mirrors
/
linux-stable
mirror of https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
1
0
Fork 0
Code Issues Releases Wiki Activity
linux-stable/drivers/net/ethernet/intel/igb/igb_main.c

9643 lines
257 KiB
C
Raw Normal View History

intel: add SPDX identifiers to all the Intel drivers Add the SPDX identifiers to all the Intel wired LAN driver files, as outlined in Documentation/process/license-rules.rst. Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-22 17:08:48 +00:00
// SPDX-License-Identifier: GPL-2.0
net: intel: Cleanup the copyright/license headers After many years of having a ~30 line copyright and license header to our source files, we are finally able to reduce that to one line with the advent of the SPDX identifier. Also caught a few files missing the SPDX license identifier, so fixed them up. Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Acked-by: Shannon Nelson <shannon.nelson@oracle.com> Acked-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-26 15:08:09 +00:00
/* Copyright(c) 2007 - 2018 Intel Corporation. */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#include <linux/module.h>
#include <linux/types.h>
#include <linux/init.h>
igb: do vlan cleanup - unify vlan and nonvlan rx path - kill adapter->vlgrp and igb_vlan_rx_register - allow to turn on/off rx/tx vlan accel via ethtool (set_features) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-21 03:27:27 +00:00
#include <linux/bitops.h>
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#include <linux/vmalloc.h>
#include <linux/pagemap.h>
#include <linux/netdevice.h>
#include <linux/ipv6.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#include <net/checksum.h>
#include <net/ip6_checksum.h>
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
#include <net/pkt_sched.h>
igb: Add the skeletons for tc-flower offloading This adds basic functions needed to implement offloading for filters created by tc-flower. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:58 +00:00
#include <net/pkt_cls.h>
igb: stub support for SIOCSHWTSTAMP Signed-off-by: John Ronciak <john.ronciak@intel.com> Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-12 05:03:42 +00:00
#include <linux/net_tstamp.h>
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#include <linux/mii.h>
#include <linux/ethtool.h>
net: introduce IFF_UNICAST_FLT private flag Use IFF_UNICAST_FTL to find out if driver handles unicast address filtering. In case it does not, promisc mode is entered. Patch also fixes following drivers: stmmac, niu: support uc filtering and yet it propagated ndo_set_multicast_list bna, benet, pxa168_eth, ks8851, ks8851_mll, ksz884x : has set ndo_set_rx_mode but do not support uc filtering Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-08-16 06:29:00 +00:00
#include <linux/if.h>
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#include <linux/if_vlan.h>
#include <linux/pci.h>
#include <linux/delay.h>
#include <linux/interrupt.h>
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
#include <linux/ip.h>
#include <linux/tcp.h>
#include <linux/sctp.h>
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#include <linux/if_ether.h>
igb: Add support for pci-e Advanced Error Reporting Add the calls necessary to enable advanced error reporting for igb on systems with AER enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-27 08:24:37 +00:00
#include <linux/aer.h>
Add appropriate <linux/prefetch.h> include for prefetch users After discovering that wide use of prefetch on modern CPUs could be a net loss instead of a win, net drivers which were relying on the implicit inclusion of prefetch.h via the list headers showed up in the resulting cleanup fallout. Give them an explicit include via the following $0.02 script. ========================================= #!/bin/bash MANUAL="" for i in `git grep -l 'prefetch(.*)' .` ; do grep -q '<linux/prefetch.h>' $i if [ $? = 0 ] ; then continue fi ( echo '?^#include <linux/?a' echo '#include <linux/prefetch.h>' echo . echo w echo q ) | ed -s $i > /dev/null 2>&1 if [ $? != 0 ]; then echo $i needs manual fixup MANUAL="$i $MANUAL" fi done echo ------------------- 8\<---------------------- echo vi $MANUAL ========================================= Signed-off-by: Paul <paul.gortmaker@windriver.com> [ Fixed up some incorrect #include placements, and added some non-network drivers and the fib_trie.c case - Linus ] Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-05-22 20:47:17 +00:00
#include <linux/prefetch.h>
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
#include <linux/pm_runtime.h>
igb: allow setting MAC address on i211 using a device tree blob The Intel i211 LOM PCIe Ethernet controllers' iNVM operates as an OTP and has no external EEPROM interface [1]. The following allows the driver to pickup the MAC address from a device tree blob when CONFIG_OF has been enabled. [1] http://www.intel.com/content/www/us/en/embedded/products/networking/i211-ethernet-controller-datasheet.html Signed-off-by: John Holland <jotihojr@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-02-18 11:10:52 +00:00
#include <linux/etherdevice.h>
igb: add IGB_DCA instead of selecting INTEL_IOATDMA Add a bool IGB_DCA defined to y if IGB and DCA are enabled, but IGB isn't y while DCA=m. And thus remove the need to select INTEL_IOATDMA when IGB is enabled, so that non-x86 architectures can build the igb driver. Based on work/patch from Brice Goglin <brice@myri.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-10-17 18:08:31 +00:00
#ifdef CONFIG_IGB_DCA
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
#include <linux/dca.h>
#endif
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
#include <linux/i2c.h>
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#include "igb.h"
igb: Bump version of driver Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-17 16:44:53 +00:00
#define MAJ 5
igb: bump version to igb-5.4.0 Bump igb version to match other igb drivers. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-08-24 15:40:23 +00:00
#define MIN 4
igb: bump version to igb-5.3.0 Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-05-20 22:40:20 +00:00
#define BUILD 0
igb: Bump version to 3.0.6 This patch updates igb version to 3.0.6. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-03-12 04:58:19 +00:00
#define DRV_VERSION __stringify(MAJ) "." __stringify(MIN) "." \
igb: Change version to remove number after -k in kernel versions. This patch changes the way versioning is done for igb in the kernel by removing the number after the "k." It has been determined that just the "k" is sufficient to identify a kernel version and the following number was used in an inconsistent manner. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-05-26 03:02:26 +00:00
__stringify(BUILD) "-k"
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
enum queue_mode {
QUEUE_MODE_STRICT_PRIORITY,
QUEUE_MODE_STREAM_RESERVATION,
};
enum tx_queue_prio {
TX_QUEUE_PRIO_HIGH,
TX_QUEUE_PRIO_LOW,
};
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
char igb_driver_name[] = "igb";
char igb_driver_version[] = DRV_VERSION;
static const char igb_driver_string[] =
"Intel(R) Gigabit Ethernet Network Driver";
igb: Copyright string update to year 2013 This patch updates Copyright year to 2013 v2: Changed Copyright year on Makefile Signed-off-by: Akeem G. Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-08 18:31:12 +00:00
static const char igb_copyright[] =
igb: Update license text to remove FSF address and update copyright. This patch updates the license text to remove address of Free Software Foundation and refer users to www.gnu.org instead. This patch also updates the copyright dates in appropriate igb driver files. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-26 01:58:57 +00:00
"Copyright (c) 2007-2014 Intel Corporation.";
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static const struct e1000_info *igb_info_tbl[] = {
[board_82575] = &e1000_82575_info,
};
igb: Cleanups to replace deprecated DEFINE_PCI_DEVICE_TABLE This patch changes implementation to remove use of DEFINE_PCI_DEVICE_TABLE. This patch fixes WARNING:DEFINE_PCI_DEVICE_TABLE Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:47:08 +00:00
static const struct pci_device_id igb_pci_tbl[] = {
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I354_BACKPLANE_1GBPS) },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I354_SGMII) },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I354_BACKPLANE_2_5GBPS) },
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I211_COPPER), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I210_COPPER), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I210_FIBER), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I210_SERDES), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I210_SGMII), board_82575 },
igb: Add device support for flashless SKU of i210 device This patch adds the specific device id support for versions of i210 that do not have flash installed. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-16 19:18:36 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I210_COPPER_FLASHLESS), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I210_SERDES_FLASHLESS), board_82575 },
igb: add support for Intel I350 Gigabit Network Connection This patch adds support for the the I350 Gigabit network connection which is the follow-on part to the 82580. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> CC: James Hearn <james.r.hearn@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-22 14:08:06 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I350_COPPER), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I350_FIBER), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I350_SERDES), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I350_SGMII), board_82575 },
igb: Add full support for 82580 devices This patch makes use of the 82580 PHY and MAC support added and adds a set of supported device IDs for said hardware. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-19 12:42:21 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82580_COPPER), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82580_FIBER), board_82575 },
igb: Add support for i340 Quad Port Fiber Adapter This patch enables support for Intel i340 Quad Port Fiber Adapter. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-01-14 05:33:46 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82580_QUAD_FIBER), board_82575 },
igb: Add full support for 82580 devices This patch makes use of the 82580 PHY and MAC support added and adds a set of supported device IDs for said hardware. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-19 12:42:21 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82580_SERDES), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82580_SGMII), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82580_COPPER_DUAL), board_82575 },
igb: Add support for DH89xxCC This patch adds support for the Intel(r) DH89xxCC series. The new device will be using Intel(r) i347-AT4 and Marvell(r) M88E1322 and M88E1112 PHYs. Support for these devices has also been added here. Signed-off-by: Joseph Gasparakis <joseph.gasparakis@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-09-22 17:56:44 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_DH89XXCC_SGMII), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_DH89XXCC_SERDES), board_82575 },
igb: Some fine tuning This patch does the following: 1. Changes the existing supported device id's so now DH89xxCC is not supported when EEPROM is not read. 2. Adds two more device ids for DH89xxCC in backplane mode and SFP. 3. Driver now initializes previously possibly uninitialized value in igb_reset_mdicnfg_82580(). Signed-off-by: Joseph Gasparakis <joseph.gasparakis@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-12-09 01:41:01 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_DH89XXCC_BACKPLANE), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_DH89XXCC_SFP), board_82575 },
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82576), board_82575 },
igb: add support for another dual port 82576 non-security nic Adding device id to support 82576NS dual port copper NIC. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-13 20:42:15 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82576_NS), board_82575 },
igb: add support for 82576NS SerDes adapter This patch adds the device ID necessary to support the 82576NS SerDes adapter. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:27 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82576_NS_SERDES), board_82575 },
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82576_FIBER), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82576_SERDES), board_82575 },
igb: add support for 82576 mezzanine card This patch adds support for a new 82576 mezzanine adapter. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:09:48 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82576_SERDES_QUAD), board_82575 },
igb: Add support for 82576 ET2 Quad Port Server Adapter Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-19 06:07:48 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82576_QUAD_COPPER_ET2), board_82575 },
igb: add support for 82576 quad copper adapter Add support for 82576 copper adapter and necessary code to restrict wol for quad port adapter to first port. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-13 20:42:35 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82576_QUAD_COPPER), board_82575 },
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82575EB_COPPER), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82575EB_FIBER_SERDES), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82575GB_QUAD_COPPER), board_82575 },
/* required last entry */
{0, }
};
MODULE_DEVICE_TABLE(pci, igb_pci_tbl);
static int igb_setup_all_tx_resources(struct igb_adapter *);
static int igb_setup_all_rx_resources(struct igb_adapter *);
static void igb_free_all_tx_resources(struct igb_adapter *);
static void igb_free_all_rx_resources(struct igb_adapter *);
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
static void igb_setup_mrqc(struct igb_adapter *);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static int igb_probe(struct pci_dev *, const struct pci_device_id *);
net/intel: remove __dev* attributes CONFIG_HOTPLUG is going away as an option. As result the __dev* markings will be going away. Remove use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit. Signed-off-by: Bill Pemberton <wfp5p@virginia.edu> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Jesse Brandeburg <jesse.brandeburg@intel.com> Cc: Bruce Allan <bruce.w.allan@intel.com> Cc: Carolyn Wyborny <carolyn.wyborny@intel.com> Cc: Don Skidmore <donald.c.skidmore@intel.com> Cc: Greg Rose <gregory.v.rose@intel.com> Cc: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Cc: Alex Duyck <alexander.h.duyck@intel.com> Cc: John Ronciak <john.ronciak@intel.com> Cc: Tushar Dave <tushar.n.dave@intel.com> Cc: e1000-devel@lists.sourceforge.net Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-03 14:24:25 +00:00
static void igb_remove(struct pci_dev *pdev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static int igb_sw_init(struct igb_adapter *);
igb: call ndo_stop() instead of dev_close() when running offline selftest Calling dev_close() causes IFF_UP to be cleared which will remove the interfaces routes and some addresses. That's probably not what the user intended when running the offline selftest. Besides this does not happen if the interface is brought down before the test, so the current behaviour is inconsistent. Instead call the net_device_ops ndo_stop function directly and avoid touching IFF_UP at all. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-02-03 08:20:50 +00:00
int igb_open(struct net_device *);
int igb_close(struct net_device *);
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
static void igb_configure(struct igb_adapter *);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static void igb_configure_tx(struct igb_adapter *);
static void igb_configure_rx(struct igb_adapter *);
static void igb_clean_all_tx_rings(struct igb_adapter *);
static void igb_clean_all_rx_rings(struct igb_adapter *);
igb: Remove adapter struct from these function call parameters Simplify a bunch of calls to these functions by moving the adapter struct out of the function arguments since that is not needed. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 17:59:48 +00:00
static void igb_clean_tx_ring(struct igb_ring *);
static void igb_clean_rx_ring(struct igb_ring *);
igb: add support for set_rx_mode netdevice operation This patch adds support for the set_rx_mode netdevice operation so that igb can better support multiple unicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:48:56 +00:00
static void igb_set_rx_mode(struct net_device *);
ethernet/intel: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Switches test of .data field to .function, since .data will be going away. Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: intel-wired-lan@lists.osuosl.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-17 00:29:35 +00:00
static void igb_update_phy_info(struct timer_list *);
static void igb_watchdog(struct timer_list *);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static void igb_watchdog_task(struct work_struct *);
igb: drop the "adv" off function names relating to descriptors Many of the function names in the hot path are carrying an extra "_adv" suffix on the end of them to represent the fact that they are using advanced descriptors instead of legacy descriptors. However since all igb uses are advanced descriptors adding the extra suffix doesn't really add any additional data. Since this is the case it is easiest to just drop the suffix and save us from having to store the extra characters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:59 +00:00
static netdev_tx_t igb_xmit_frame(struct sk_buff *skb, struct net_device *);
net: make ndo_get_stats64 a void function The network device operation for reading statistics is only called in one place, and it ignores the return value. Having a structure return value is potentially confusing because some future driver could incorrectly assume that the return value was used. Fix all drivers with ndo_get_stats64 to have a void function. Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-07 03:12:52 +00:00
static void igb_get_stats64(struct net_device *dev,
struct rtnl_link_stats64 *stats);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static int igb_change_mtu(struct net_device *, int);
static int igb_set_mac(struct net_device *, void *);
igb: Add workaround for VLAN tag stripping on 82576 There was a workaround partially implemented for the 82576 that is needed in order for VLAN tag stripping to function correctly. The original code had side effects that would make it so the workaround was active on all MACs. I have updated the code so that the workaround is enabled, but limited to the 82576, or activated if we exceed the available unicast addresses. The workaround has a side effect of mirroring all of the traffic outgoing from the VFs back to the PF. As such it is not recommended to use the 82576 in promiscuous mode as it will take a performance hit, though this is now consistent with the performance as seen on the out-of-tree igb driver. I also limited the scope of the UTA bits all being set to only when the VMOLR register is enabled. This should limit the effects of the UTA register so that we don't pick up any excess traffic unless promiscuous mode has been enabled on the PF, whereas before the PF would have ended up in something equivalent to unicast promiscuous mode with VLAN filtering otherwise. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:43 +00:00
static void igb_set_uta(struct igb_adapter *adapter, bool set);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static irqreturn_t igb_intr(int irq, void *);
static irqreturn_t igb_intr_msi(int irq, void *);
static irqreturn_t igb_msix_other(int irq, void *);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
static irqreturn_t igb_msix_ring(int irq, void *);
igb: add IGB_DCA instead of selecting INTEL_IOATDMA Add a bool IGB_DCA defined to y if IGB and DCA are enabled, but IGB isn't y while DCA=m. And thus remove the need to select INTEL_IOATDMA when IGB is enabled, so that non-x86 architectures can build the igb driver. Based on work/patch from Brice Goglin <brice@myri.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-10-17 18:08:31 +00:00
#ifdef CONFIG_IGB_DCA
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
static void igb_update_dca(struct igb_q_vector *);
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
static void igb_setup_dca(struct igb_adapter *);
igb: add IGB_DCA instead of selecting INTEL_IOATDMA Add a bool IGB_DCA defined to y if IGB and DCA are enabled, but IGB isn't y while DCA=m. And thus remove the need to select INTEL_IOATDMA when IGB is enabled, so that non-x86 architectures can build the igb driver. Based on work/patch from Brice Goglin <brice@myri.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-10-17 18:08:31 +00:00
#endif /* CONFIG_IGB_DCA */
igb: Introduce multiple TX queues with infrastructure This code adds multiple Tx queue infrastructure much like we previously did in ixgbe. The MSI-X vector mapping is the bulk of the change. IAM can now be safely enabled and we've verified that it does work correctly. We can also eliminate the tx ring lock. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:06:51 +00:00
static int igb_poll(struct napi_struct *, int);
igb: Add support for bulk Tx cleanup & cleanup boolean logic This patch enables bulk free in Tx cleanup for igb and cleans up the boolean logic in the polling routines for igb in the hopes of avoiding any mix-ups similar to what occurred with i40e and i40evf. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-03-07 17:30:21 +00:00
static bool igb_clean_tx_irq(struct igb_q_vector *, int);
drivers/net/intel: use napi_complete_done() As per Eric Dumazet's previous patches: (see commit (24d2e4a50737) - tg3: use napi_complete_done()) Quoting verbatim: Using napi_complete_done() instead of napi_complete() allows us to use /sys/class/net/ethX/gro_flush_timeout GRO layer can aggregate more packets if the flush is delayed a bit, without having to set too big coalescing parameters that impact latencies. </end quote> Tested configuration: low latency via ethtool -C ethx adaptive-rx off rx-usecs 10 adaptive-tx off tx-usecs 15 workload: streaming rx using netperf TCP_MAERTS igb: MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.48 10^6bits/s over 1.000 seconds ending at 1440193171.589 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1176930056 1475.36 797726 16384.00 71905 MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.49 10^6bits/s over 0.997 seconds ending at 1440193142.763 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1175182320 50476.00 23282 16384.00 71816 i40e: Hard to test because the traffic is incoming so fast (24Gb/s) that GRO always receives 87kB, even at the highest interrupt rate. Other drivers were only compile tested. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-09-24 23:35:47 +00:00
static int igb_clean_rx_irq(struct igb_q_vector *, int);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static int igb_ioctl(struct net_device *, struct ifreq *, int cmd);
static void igb_tx_timeout(struct net_device *);
static void igb_reset_task(struct work_struct *);
igb: Cleanups to fix line length warnings This patch fixes WARNING:LONG_LINE found with checkpatch check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:46:33 +00:00
static void igb_vlan_mode(struct net_device *netdev,
netdev_features_t features);
net: vlan: prepare for 802.1ad VLAN filtering offload Change the rx_{add,kill}_vid callbacks to take a protocol argument in preparation of 802.1ad support. The protocol argument used so far is always htons(ETH_P_8021Q). Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-19 02:04:28 +00:00
static int igb_vlan_rx_add_vid(struct net_device *, __be16, u16);
static int igb_vlan_rx_kill_vid(struct net_device *, __be16, u16);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static void igb_restore_vlan(struct igb_adapter *);
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
static void igb_rar_set_index(struct igb_adapter *, u32);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
static void igb_ping_all_vfs(struct igb_adapter *);
static void igb_msg_task(struct igb_adapter *);
static void igb_vmm_control(struct igb_adapter *);
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
static int igb_set_vf_mac(struct igb_adapter *, int, unsigned char *);
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
static void igb_flush_mac_table(struct igb_adapter *);
static int igb_available_rars(struct igb_adapter *, u8);
static void igb_set_default_mac_filter(struct igb_adapter *);
static int igb_uc_sync(struct net_device *, const unsigned char *);
static int igb_uc_unsync(struct net_device *, const unsigned char *);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
static void igb_restore_vf_multicasts(struct igb_adapter *adapter);
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
static int igb_ndo_set_vf_mac(struct net_device *netdev, int vf, u8 *mac);
static int igb_ndo_set_vf_vlan(struct net_device *netdev,
net: Update API for VF vlan protocol 802.1ad support Introduce new rtnl UAPI that exposes a list of vlans per VF, giving the ability for user-space application to specify it for the VF, as an option to support 802.1ad. We adjusted IP Link tool to support this option. For future use cases, the new UAPI supports multiple vlans. For now we limit the list size to a single vlan in kernel. Add IFLA_VF_VLAN_LIST in addition to IFLA_VF_VLAN to keep backward compatibility with older versions of IP Link tool. Add a vlan protocol parameter to the ndo_set_vf_vlan callback. We kept 802.1Q as the drivers' default vlan protocol. Suitable ip link tool command examples: Set vf vlan protocol 802.1ad: ip link set eth0 vf 1 vlan 100 proto 802.1ad Set vf to VST (802.1Q) mode: ip link set eth0 vf 1 vlan 100 proto 802.1Q Or by omitting the new parameter ip link set eth0 vf 1 vlan 100 Signed-off-by: Moshe Shemesh <moshe@mellanox.com> Signed-off-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-22 09:11:15 +00:00
int vf, u16 vlan, u8 qos, __be16 vlan_proto);
net-next:v4: Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool. o min_tx_rate puts lower limit on the VF bandwidth. VF is guaranteed to have a bandwidth of at least this value. max_tx_rate puts cap on the VF bandwidth. VF can have a bandwidth of up to this value. o A new handler set_vf_rate for attr IFLA_VF_RATE has been introduced which takes 4 arguments: netdev, VF number, min_tx_rate, max_tx_rate o ndo_set_vf_rate replaces ndo_set_vf_tx_rate handler. o Drivers that currently implement ndo_set_vf_tx_rate should now call ndo_set_vf_rate instead and reject attempt to set a minimum bandwidth greater than 0 for IFLA_VF_TX_RATE when IFLA_VF_RATE is not yet implemented by driver. o If user enters only one of either min_tx_rate or max_tx_rate, then, userland should read back the other value from driver and set both for IFLA_VF_RATE. Drivers that have not yet implemented IFLA_VF_RATE should always return min_tx_rate as 0 when read from ip tool. o If both IFLA_VF_TX_RATE and IFLA_VF_RATE options are specified, then IFLA_VF_RATE should override. o Idea is to have consistent display of rate values to user. o Usage example: - ./ip link set p4p1 vf 0 rate 900 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 900 (Mbps), max_tx_rate 900Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 300 min_tx_rate 200 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 300 (Mbps), max_tx_rate 300Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 600 rate 300 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5, tx rate 600 (Mbps), max_tx_rate 600Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 13:59:05 +00:00
static int igb_ndo_set_vf_bw(struct net_device *, int, int, int);
igb: add support for spoofchk config Add support for spoofchk configuration per VF via iproute2 tool. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-03 20:27:48 +00:00
static int igb_ndo_set_vf_spoofchk(struct net_device *netdev, int vf,
bool setting);
igb: add VF trust infrastructure * Add a per-VF value to know if a VF is trusted, by default don't trust VFs. * Implement netdev op to trust VFs (igb_ndo_set_vf_trust) and add trust status to ndo_get_vf_config output. * Allow a trusted VF to change MAC and MAC filters even if MAC has been administratively set. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-01-17 10:53:39 +00:00
static int igb_ndo_set_vf_trust(struct net_device *netdev, int vf,
bool setting);
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
static int igb_ndo_get_vf_config(struct net_device *netdev, int vf,
struct ifla_vf_info *ivi);
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
static void igb_check_vf_rate_limit(struct igb_adapter *);
igb: add support of RX network flow classification This patch is meant to allow for RX network flow classification to insert and remove Rx filter by ethtool. Ethtool interface has it's own rules manager Show all filters: $ ethtool -n eth0 4 RX rings available Total 2 rules Signed-off-by: Ruhao Gao <ruhao.gao@ni.com> Signed-off-by: Gangfeng Huang <gangfeng.huang@ni.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-07-06 05:22:54 +00:00
static void igb_nfc_filter_exit(struct igb_adapter *adapter);
static void igb_nfc_filter_restore(struct igb_adapter *adapter);
igb: fix a compile warning control these three function declarations and definitions with same macro CONFIG_PCI_IOV drivers/net/ethernet/intel/igb/igb_main.c:165: warning: ‘igb_vf_configure’ declared ‘static’ but never defined drivers/net/ethernet/intel/igb/igb_main.c:166: warning: ‘igb_find_enabled_vfs’ declared ‘static’ but never defined drivers/net/ethernet/intel/igb/igb_main.c:167: warning: ‘igb_check_vf_assignment’ declared ‘static’ but never defined Signed-off-by: RongQing Li <roy.qing.li@gmail.com> Acked-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-18 22:52:35 +00:00
#ifdef CONFIG_PCI_IOV
igb: Check if subordinate VFs are assigned to virtual machines Kvm and the Xen pci-back driver will set a flag in the virtual function pci device dev_flags when the VF is assigned to a guest VM. Before destroying subordinate VFs check to see if the flag is set and if so skip the call to pci_disable_sriov() to avoid system crashes. Copy the maintainer for the Xen pci-back driver. Also CC'ing maintainers of all drivers found to call pci_disable_sriov(). V2 - Fix uninitialized variable warning Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Christian Benvenuti <benve@cisco.com> Cc: Sathya Perla <sathya.perla@emulex.com> Cc: Dimitris Michailidis <dm@chelsio.com> Cc: Jon Mason <jdmason@kudzu.us> Cc: James Smart <james.smart@emulex.com> Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-14 02:57:14 +00:00
static int igb_vf_configure(struct igb_adapter *adapter, int vf);
igb: fix driver reload with VF assigned to guest commit fa44f2f185f7f9da19d331929bb1b56c1ccd1d93 broke reloading of igb, when VFs are assigned to a guest, in several ways. 1. on module load adapter->vf_data does not get properly allocated, resulting in a null pointer exception when accessing adapter->vf_data in igb_reset() on module reload. modprobe -r igb ; modprobe igb max_vfs=7 [ 215.215837] igb 0000:01:00.1: removed PHC on eth1 [ 216.932072] igb 0000:01:00.1: IOV Disabled [ 216.937038] igb 0000:01:00.0: removed PHC on eth0 [ 217.127032] igb 0000:01:00.0: Cannot deallocate SR-IOV virtual functions while they are assigned - VFs will not be deallocated [ 217.146178] igb: Intel(R) Gigabit Ethernet Network Driver - version 5.0.5-k [ 217.154050] igb: Copyright (c) 2007-2013 Intel Corporation. [ 217.160688] igb 0000:01:00.0: Enabling SR-IOV VFs using the module parameter is deprecated - please use the pci sysfs interface. [ 217.173703] igb 0000:01:00.0: irq 103 for MSI/MSI-X [ 217.179227] igb 0000:01:00.0: irq 104 for MSI/MSI-X [ 217.184735] igb 0000:01:00.0: irq 105 for MSI/MSI-X [ 217.220082] BUG: unable to handle kernel NULL pointer dereference at 0000000000000048 [ 217.228846] IP: [<ffffffffa007c5e5>] igb_reset+0xc5/0x4b0 [igb] [ 217.235472] PGD 3607ec067 PUD 36170b067 PMD 0 [ 217.240461] Oops: 0002 [#1] SMP [ 217.244085] Modules linked in: igb(+) igbvf mptsas mptscsih mptbase scsi_transport_sas [last unloaded: igb] [ 217.255040] CPU: 4 PID: 4833 Comm: modprobe Not tainted 3.11.0+ #46 [...] [ 217.390007] [<ffffffffa007fab2>] igb_probe+0x892/0xfd0 [igb] [ 217.396422] [<ffffffff81470b3e>] local_pci_probe+0x1e/0x40 [ 217.402641] [<ffffffff81472029>] pci_device_probe+0xf9/0x110 [...] 2. A follow up issue, pci_enable_sriov() should only be called if no VFs were still allocated on module unload. Otherwise pci_enable_sriov() gets called multiple times in a row rendering the NIC unusable until reset. 3. simply calling igb_enable_sriov() in igb_probe_vfs() is not enough as the interrupts need to be re-setup. Switching that to igb_pci_enable_sriov(). Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Sibai Li <Sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-09-24 05:18:39 +00:00
static int igb_pci_enable_sriov(struct pci_dev *dev, int num_vfs);
igb: make sure SR-IOV init uses the right number of queues Recent changes to igb_probe_vfs() could lead to the PF holding onto all of the queues. Reorder igb_probe_vfs() to be before gb_init_queue_configuration() and add some more error checking. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-08-08 00:27:39 +00:00
static int igb_disable_sriov(struct pci_dev *dev);
static int igb_pci_disable_sriov(struct pci_dev *dev);
igb: fix a compile warning control these three function declarations and definitions with same macro CONFIG_PCI_IOV drivers/net/ethernet/intel/igb/igb_main.c:165: warning: ‘igb_vf_configure’ declared ‘static’ but never defined drivers/net/ethernet/intel/igb/igb_main.c:166: warning: ‘igb_find_enabled_vfs’ declared ‘static’ but never defined drivers/net/ethernet/intel/igb/igb_main.c:167: warning: ‘igb_check_vf_assignment’ declared ‘static’ but never defined Signed-off-by: RongQing Li <roy.qing.li@gmail.com> Acked-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-18 22:52:35 +00:00
#endif
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
static int igb_suspend(struct device *);
static int igb_resume(struct device *);
static int igb_runtime_suspend(struct device *dev);
static int igb_runtime_resume(struct device *dev);
static int igb_runtime_idle(struct device *dev);
static const struct dev_pm_ops igb_pm_ops = {
SET_SYSTEM_SLEEP_PM_OPS(igb_suspend, igb_resume)
SET_RUNTIME_PM_OPS(igb_runtime_suspend, igb_runtime_resume,
igb_runtime_idle)
};
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static void igb_shutdown(struct pci_dev *);
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
static int igb_pci_sriov_configure(struct pci_dev *dev, int num_vfs);
igb: add IGB_DCA instead of selecting INTEL_IOATDMA Add a bool IGB_DCA defined to y if IGB and DCA are enabled, but IGB isn't y while DCA=m. And thus remove the need to select INTEL_IOATDMA when IGB is enabled, so that non-x86 architectures can build the igb driver. Based on work/patch from Brice Goglin <brice@myri.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-10-17 18:08:31 +00:00
#ifdef CONFIG_IGB_DCA
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
static int igb_notify_dca(struct notifier_block *, unsigned long, void *);
static struct notifier_block dca_notifier = {
.notifier_call = igb_notify_dca,
.next = NULL,
.priority = 0
};
#endif
igb: this patch addes the sr-iov enablement option via num_vfs parameter This code adds a module parameter called num_vfs which defines if the driver should attempt to use sr-iov and if so how many VFs should be enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:30 +00:00
#ifdef CONFIG_PCI_IOV
igb: Cleanups to fix static initialization This patch fixes ERROR:INITIALISED_STATIC from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:46:48 +00:00
static unsigned int max_vfs;
igb: remove sysfs entry that was used to set the number of vfs This patch removes the sysfs entry num_vfs which was added to support enabling pci virtual functions for 82576. To prevent VFs from loading automatically a module parameter "max_vfs" was added so that the number of VFs per PF can be limited. This is especially useful when 4 or more 82576 ports are on the system because otherwise to load all VFs would result in 8 interface per physical port. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-07 14:37:52 +00:00
module_param(max_vfs, uint, 0);
igb: Cleanups for messaging This patch fixes WARNING:PREFER_PR_LEVEL and WARNING:SPLIT_STRING from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:17 +00:00
MODULE_PARM_DESC(max_vfs, "Maximum number of virtual functions to allocate per physical function");
igb: remove sysfs entry that was used to set the number of vfs This patch removes the sysfs entry num_vfs which was added to support enabling pci virtual functions for 82576. To prevent VFs from loading automatically a module parameter "max_vfs" was added so that the number of VFs per PF can be limited. This is especially useful when 4 or more 82576 ports are on the system because otherwise to load all VFs would result in 8 interface per physical port. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-07 14:37:52 +00:00
#endif /* CONFIG_PCI_IOV */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static pci_ers_result_t igb_io_error_detected(struct pci_dev *,
pci_channel_state_t);
static pci_ers_result_t igb_io_slot_reset(struct pci_dev *);
static void igb_io_resume(struct pci_dev *);
netdev: make pci_error_handlers const Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2012-09-07 16:33:15 +00:00
static const struct pci_error_handlers igb_err_handler = {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
.error_detected = igb_io_error_detected,
.slot_reset = igb_io_slot_reset,
.resume = igb_io_resume,
};
igb: Move DMA Coalescing init code to separate function. This patch moves the DMA Coalescing feature initialization code from igb_reset to a new function and replaces it with a call to the new function. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-13 17:29:59 +00:00
static void igb_init_dmac(struct igb_adapter *adapter, u32 pba);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static struct pci_driver igb_driver = {
.name = igb_driver_name,
.id_table = igb_pci_tbl,
.probe = igb_probe,
net/intel: remove __dev* attributes CONFIG_HOTPLUG is going away as an option. As result the __dev* markings will be going away. Remove use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit. Signed-off-by: Bill Pemberton <wfp5p@virginia.edu> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Jesse Brandeburg <jesse.brandeburg@intel.com> Cc: Bruce Allan <bruce.w.allan@intel.com> Cc: Carolyn Wyborny <carolyn.wyborny@intel.com> Cc: Don Skidmore <donald.c.skidmore@intel.com> Cc: Greg Rose <gregory.v.rose@intel.com> Cc: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Cc: Alex Duyck <alexander.h.duyck@intel.com> Cc: John Ronciak <john.ronciak@intel.com> Cc: Tushar Dave <tushar.n.dave@intel.com> Cc: e1000-devel@lists.sourceforge.net Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-03 14:24:25 +00:00
.remove = igb_remove,
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#ifdef CONFIG_PM
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
.driver.pm = &igb_pm_ops,
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#endif
.shutdown = igb_shutdown,
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
.sriov_configure = igb_pci_sriov_configure,
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
.err_handler = &igb_err_handler
};
MODULE_AUTHOR("Intel Corporation, <e1000-devel@lists.sourceforge.net>");
MODULE_DESCRIPTION("Intel(R) Gigabit Ethernet Network Driver");
intel-ethernet: use correct module license We recently updated all our SPDX identifiers to correctly indicate our net/ethernet/intel/* drivers were always released and intended to be released under GPL v2, but the MODULE_LICENSE declaration was never updated. Fix the MODULE_LICENSE to be GPL v2, for all our drivers. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-09-15 00:37:57 +00:00
MODULE_LICENSE("GPL v2");
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
MODULE_VERSION(DRV_VERSION);
intel: make wired ethernet driver message level consistent (rev2) Dan Carpenter noticed that ixgbevf initial default was different than the rest. But the problem is broader than that, only one Intel driver (ixgb) was doing it almost right. The convention for default debug level should be consistent among Intel drivers and follow established convention. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-03-13 06:04:20 +00:00
#define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV|NETIF_MSG_PROBE|NETIF_MSG_LINK)
static int debug = -1;
module_param(debug, int, 0);
MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
struct igb_reg_info {
u32 ofs;
char *name;
};
static const struct igb_reg_info igb_reg_info_tbl[] = {
/* General Registers */
{E1000_CTRL, "CTRL"},
{E1000_STATUS, "STATUS"},
{E1000_CTRL_EXT, "CTRL_EXT"},
/* Interrupt Registers */
{E1000_ICR, "ICR"},
/* RX Registers */
{E1000_RCTL, "RCTL"},
{E1000_RDLEN(0), "RDLEN"},
{E1000_RDH(0), "RDH"},
{E1000_RDT(0), "RDT"},
{E1000_RXDCTL(0), "RXDCTL"},
{E1000_RDBAL(0), "RDBAL"},
{E1000_RDBAH(0), "RDBAH"},
/* TX Registers */
{E1000_TCTL, "TCTL"},
{E1000_TDBAL(0), "TDBAL"},
{E1000_TDBAH(0), "TDBAH"},
{E1000_TDLEN(0), "TDLEN"},
{E1000_TDH(0), "TDH"},
{E1000_TDT(0), "TDT"},
{E1000_TXDCTL(0), "TXDCTL"},
{E1000_TDFH, "TDFH"},
{E1000_TDFT, "TDFT"},
{E1000_TDFHS, "TDFHS"},
{E1000_TDFPC, "TDFPC"},
/* List Terminator */
{}
};
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* igb_regdump - register printout routine */
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
static void igb_regdump(struct e1000_hw *hw, struct igb_reg_info *reginfo)
{
int n = 0;
char rname[16];
u32 regs[8];
switch (reginfo->ofs) {
case E1000_RDLEN(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_RDLEN(n));
break;
case E1000_RDH(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_RDH(n));
break;
case E1000_RDT(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_RDT(n));
break;
case E1000_RXDCTL(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_RXDCTL(n));
break;
case E1000_RDBAL(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_RDBAL(n));
break;
case E1000_RDBAH(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_RDBAH(n));
break;
case E1000_TDBAL(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_RDBAL(n));
break;
case E1000_TDBAH(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_TDBAH(n));
break;
case E1000_TDLEN(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_TDLEN(n));
break;
case E1000_TDH(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_TDH(n));
break;
case E1000_TDT(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_TDT(n));
break;
case E1000_TXDCTL(0):
for (n = 0; n < 4; n++)
regs[n] = rd32(E1000_TXDCTL(n));
break;
default:
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
pr_info("%-15s %08x\n", reginfo->name, rd32(reginfo->ofs));
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
return;
}
snprintf(rname, 16, "%s%s", reginfo->name, "[0-3]");
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
pr_info("%-15s %08x %08x %08x %08x\n", rname, regs[0], regs[1],
regs[2], regs[3]);
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
}
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* igb_dump - Print registers, Tx-rings and Rx-rings */
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
static void igb_dump(struct igb_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
struct e1000_hw *hw = &adapter->hw;
struct igb_reg_info *reginfo;
struct igb_ring *tx_ring;
union e1000_adv_tx_desc *tx_desc;
struct my_u0 { u64 a; u64 b; } *u0;
struct igb_ring *rx_ring;
union e1000_adv_rx_desc *rx_desc;
u32 staterr;
igb: avoid unnecessary conversions from u16 to int There are a number of places where we have values that are stored as u16 but are being converted to int unnecessarily. In order to avoid that we should convert all variables that deal with the next_to_clean, next_to_use, and count to u16 values. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:26 +00:00
u16 i, n;
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
if (!netif_msg_hw(adapter))
return;
/* Print netdevice Info */
if (netdev) {
dev_info(&adapter->pdev->dev, "Net device Info\n");
net: Remove usage of net_device last_rx member The network stack no longer uses the last_rx member of struct net_device since the bonding driver switched to use its own private last_rx in commit 9f242738376d ("bonding: use last_arp_rx in slave_last_rx()"). However, some drivers still (ab)use the field for their own purposes and some driver just update it without actually using it. Previously, there was an accompanying comment for the last_rx member added in commit 4dc89133f49b ("net: add a comment on netdev->last_rx") which asked drivers not to update is, unless really needed. However, this commend was removed in commit f8ff080dacec ("bonding: remove useless updating of slave->dev->last_rx"), so some drivers added later on still did update last_rx. Remove all usage of last_rx and switch three drivers (sky2, atp and smc91c92_cs) which actually read and write it to use their own private copy in netdev_priv. Compile-tested with allyesconfig and allmodconfig on x86 and arm. Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Jay Vosburgh <j.vosburgh@gmail.com> Cc: Veaceslav Falico <vfalico@gmail.com> Cc: Andy Gospodarek <andy@greyhouse.net> Cc: Mirko Lindner <mlindner@marvell.com> Cc: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Tobias Klauser <tklauser@distanz.ch> Acked-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Jay Vosburgh <jay.vosburgh@canonical.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-18 16:45:01 +00:00
pr_info("Device Name state trans_start\n");
pr_info("%-15s %016lX %016lX\n", netdev->name,
netdev->state, dev_trans_start(netdev));
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
}
/* Print Registers */
dev_info(&adapter->pdev->dev, "Register Dump\n");
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
pr_info(" Register Name Value\n");
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
for (reginfo = (struct igb_reg_info *)igb_reg_info_tbl;
reginfo->name; reginfo++) {
igb_regdump(hw, reginfo);
}
/* Print TX Ring Summary */
if (!netdev || !netif_running(netdev))
goto exit;
dev_info(&adapter->pdev->dev, "TX Rings Summary\n");
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
pr_info("Queue [NTU] [NTC] [bi(ntc)->dma ] leng ntw timestamp\n");
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
for (n = 0; n < adapter->num_tx_queues; n++) {
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
struct igb_tx_buffer *buffer_info;
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
tx_ring = adapter->tx_ring[n];
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
buffer_info = &tx_ring->tx_buffer_info[tx_ring->next_to_clean];
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
pr_info(" %5d %5X %5X %016llX %04X %p %016llX\n",
n, tx_ring->next_to_use, tx_ring->next_to_clean,
igb: Use dma_unmap_addr and dma_unmap_len defines This change is meant to improve performance on systems that do not require the DMA unmap calls. On those systems we do not need to make use of the unmap address for Tx or the unmap length so we can drop both thereby reducing the size of the Tx buffer info structure. In addition I have changed the logic to check for unmap length instead of unmap address when checking to see if a buffer needs to be unmapped from DMA use. The reasons for this change is that on some platforms it is possible to receive a valid DMA address of 0 from an IOMMU. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-18 01:56:27 +00:00
(u64)dma_unmap_addr(buffer_info, dma),
dma_unmap_len(buffer_info, len),
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
buffer_info->next_to_watch,
(u64)buffer_info->time_stamp);
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
}
/* Print TX Rings */
if (!netif_msg_tx_done(adapter))
goto rx_ring_summary;
dev_info(&adapter->pdev->dev, "TX Rings Dump\n");
/* Transmit Descriptor Formats
*
* Advanced Transmit Descriptor
* +--------------------------------------------------------------+
* 0 | Buffer Address [63:0] |
* +--------------------------------------------------------------+
* 8 | PAYLEN | PORTS |CC|IDX | STA | DCMD |DTYP|MAC|RSV| DTALEN |
* +--------------------------------------------------------------+
* 63 46 45 40 39 38 36 35 32 31 24 15 0
*/
for (n = 0; n < adapter->num_tx_queues; n++) {
tx_ring = adapter->tx_ring[n];
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
pr_info("------------------------------------\n");
pr_info("TX QUEUE INDEX = %d\n", tx_ring->queue_index);
pr_info("------------------------------------\n");
igb: Cleanups for messaging This patch fixes WARNING:PREFER_PR_LEVEL and WARNING:SPLIT_STRING from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:17 +00:00
pr_info("T [desc] [address 63:0 ] [PlPOCIStDDM Ln] [bi->dma ] leng ntw timestamp bi->skb\n");
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
for (i = 0; tx_ring->desc && (i < tx_ring->count); i++) {
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
const char *next_desc;
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
struct igb_tx_buffer *buffer_info;
igb: Replace E1000_XX_DESC_ADV with IGB_XX_DESC Since igb only uses advanced descriptors we might as well just use an IGB specific define and drop the _ADV suffix for the descriptor declarations. In addition this can be further reduced by assuming that it will be working on pointers since that is normally how the Tx descriptors are handled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:05 +00:00
tx_desc = IGB_TX_DESC(tx_ring, i);
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
buffer_info = &tx_ring->tx_buffer_info[i];
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
u0 = (struct my_u0 *)tx_desc;
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
if (i == tx_ring->next_to_use &&
i == tx_ring->next_to_clean)
next_desc = " NTC/U";
else if (i == tx_ring->next_to_use)
next_desc = " NTU";
else if (i == tx_ring->next_to_clean)
next_desc = " NTC";
else
next_desc = "";
igb: Cleanups for messaging This patch fixes WARNING:PREFER_PR_LEVEL and WARNING:SPLIT_STRING from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:17 +00:00
pr_info("T [0x%03X] %016llX %016llX %016llX %04X %p %016llX %p%s\n",
i, le64_to_cpu(u0->a),
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
le64_to_cpu(u0->b),
igb: Use dma_unmap_addr and dma_unmap_len defines This change is meant to improve performance on systems that do not require the DMA unmap calls. On those systems we do not need to make use of the unmap address for Tx or the unmap length so we can drop both thereby reducing the size of the Tx buffer info structure. In addition I have changed the logic to check for unmap length instead of unmap address when checking to see if a buffer needs to be unmapped from DMA use. The reasons for this change is that on some platforms it is possible to receive a valid DMA address of 0 from an IOMMU. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-18 01:56:27 +00:00
(u64)dma_unmap_addr(buffer_info, dma),
dma_unmap_len(buffer_info, len),
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
buffer_info->next_to_watch,
(u64)buffer_info->time_stamp,
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
buffer_info->skb, next_desc);
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
igb: fix panic while dumping packets on Tx hang with IOMMU This patch resolves a "BUG: unable to handle kernel paging request at ..." oops while dumping packet data. The issue occurs with IOMMU enabled due to the address provided by phys_to_virt(). This patch avoids phys_to_virt() by making using skb->data and the address of the pages allocated for Rx. Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
2012-07-28 05:07:48 +00:00
if (netif_msg_pktdata(adapter) && buffer_info->skb)
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
print_hex_dump(KERN_INFO, "",
DUMP_PREFIX_ADDRESS,
igb: fix panic while dumping packets on Tx hang with IOMMU This patch resolves a "BUG: unable to handle kernel paging request at ..." oops while dumping packet data. The issue occurs with IOMMU enabled due to the address provided by phys_to_virt(). This patch avoids phys_to_virt() by making using skb->data and the address of the pages allocated for Rx. Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
2012-07-28 05:07:48 +00:00
16, 1, buffer_info->skb->data,
igb: Use dma_unmap_addr and dma_unmap_len defines This change is meant to improve performance on systems that do not require the DMA unmap calls. On those systems we do not need to make use of the unmap address for Tx or the unmap length so we can drop both thereby reducing the size of the Tx buffer info structure. In addition I have changed the logic to check for unmap length instead of unmap address when checking to see if a buffer needs to be unmapped from DMA use. The reasons for this change is that on some platforms it is possible to receive a valid DMA address of 0 from an IOMMU. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-18 01:56:27 +00:00
dma_unmap_len(buffer_info, len),
true);
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
}
}
/* Print RX Rings Summary */
rx_ring_summary:
dev_info(&adapter->pdev->dev, "RX Rings Summary\n");
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
pr_info("Queue [NTU] [NTC]\n");
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
for (n = 0; n < adapter->num_rx_queues; n++) {
rx_ring = adapter->rx_ring[n];
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
pr_info(" %5d %5X %5X\n",
n, rx_ring->next_to_use, rx_ring->next_to_clean);
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
}
/* Print RX Rings */
if (!netif_msg_rx_status(adapter))
goto exit;
dev_info(&adapter->pdev->dev, "RX Rings Dump\n");
/* Advanced Receive Descriptor (Read) Format
* 63 1 0
* +-----------------------------------------------------+
* 0 | Packet Buffer Address [63:1] |A0/NSE|
* +----------------------------------------------+------+
* 8 | Header Buffer Address [63:1] | DD |
* +-----------------------------------------------------+
*
*
* Advanced Receive Descriptor (Write-Back) Format
*
* 63 48 47 32 31 30 21 20 17 16 4 3 0
* +------------------------------------------------------+
* 0 | Packet IP |SPH| HDR_LEN | RSV|Packet| RSS |
* | Checksum Ident | | | | Type | Type |
* +------------------------------------------------------+
* 8 | VLAN Tag | Length | Extended Error | Extended Status |
* +------------------------------------------------------+
* 63 48 47 32 31 20 19 0
*/
for (n = 0; n < adapter->num_rx_queues; n++) {
rx_ring = adapter->rx_ring[n];
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
pr_info("------------------------------------\n");
pr_info("RX QUEUE INDEX = %d\n", rx_ring->queue_index);
pr_info("------------------------------------\n");
igb: Cleanups for messaging This patch fixes WARNING:PREFER_PR_LEVEL and WARNING:SPLIT_STRING from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:17 +00:00
pr_info("R [desc] [ PktBuf A0] [ HeadBuf DD] [bi->dma ] [bi->skb] <-- Adv Rx Read format\n");
pr_info("RWB[desc] [PcsmIpSHl PtRs] [vl er S cks ln] ---------------- [bi->skb] <-- Adv Rx Write-Back format\n");
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
for (i = 0; i < rx_ring->count; i++) {
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
const char *next_desc;
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
struct igb_rx_buffer *buffer_info;
buffer_info = &rx_ring->rx_buffer_info[i];
igb: Replace E1000_XX_DESC_ADV with IGB_XX_DESC Since igb only uses advanced descriptors we might as well just use an IGB specific define and drop the _ADV suffix for the descriptor declarations. In addition this can be further reduced by assuming that it will be working on pointers since that is normally how the Tx descriptors are handled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:05 +00:00
rx_desc = IGB_RX_DESC(rx_ring, i);
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
u0 = (struct my_u0 *)rx_desc;
staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
if (i == rx_ring->next_to_use)
next_desc = " NTU";
else if (i == rx_ring->next_to_clean)
next_desc = " NTC";
else
next_desc = "";
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
if (staterr & E1000_RXD_STAT_DD) {
/* Descriptor Done */
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
pr_info("%s[0x%03X] %016llX %016llX ---------------- %s\n",
"RWB", i,
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
le64_to_cpu(u0->a),
le64_to_cpu(u0->b),
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
next_desc);
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
} else {
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
pr_info("%s[0x%03X] %016llX %016llX %016llX %s\n",
"R ", i,
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
le64_to_cpu(u0->a),
le64_to_cpu(u0->b),
(u64)buffer_info->dma,
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
next_desc);
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
igb: fix panic while dumping packets on Tx hang with IOMMU This patch resolves a "BUG: unable to handle kernel paging request at ..." oops while dumping packet data. The issue occurs with IOMMU enabled due to the address provided by phys_to_virt(). This patch avoids phys_to_virt() by making using skb->data and the address of the pages allocated for Rx. Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
2012-07-28 05:07:48 +00:00
if (netif_msg_pktdata(adapter) &&
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
buffer_info->dma && buffer_info->page) {
igb: drop support for single buffer mode This change removes support for single buffer mode from igb and makes the driver function in packet split always. The advantage to doing this is that we can reduce total memory allocation overhead significantly as we will only need to allocate one 1K slab per packet and then make use of a reusable half page instead of allocating a 2K slab per packet. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:38 +00:00
print_hex_dump(KERN_INFO, "",
DUMP_PREFIX_ADDRESS,
16, 1,
igb: fix panic while dumping packets on Tx hang with IOMMU This patch resolves a "BUG: unable to handle kernel paging request at ..." oops while dumping packet data. The issue occurs with IOMMU enabled due to the address provided by phys_to_virt(). This patch avoids phys_to_virt() by making using skb->data and the address of the pages allocated for Rx. Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
2012-07-28 05:07:48 +00:00
page_address(buffer_info->page) +
buffer_info->page_offset,
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
igb_rx_bufsz(rx_ring), true);
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
}
}
}
}
exit:
return;
}
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/**
* igb_get_i2c_data - Reads the I2C SDA data bit
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
* @hw: pointer to hardware structure
* @i2cctl: Current value of I2CCTL register
*
* Returns the I2C data bit value
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
**/
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
static int igb_get_i2c_data(void *data)
{
struct igb_adapter *adapter = (struct igb_adapter *)data;
struct e1000_hw *hw = &adapter->hw;
s32 i2cctl = rd32(E1000_I2CPARAMS);
igb: Cleanups to remove return parentheses This patch fixes ERROR:RETURN_PARENTHESES from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 02:11:17 +00:00
return !!(i2cctl & E1000_I2C_DATA_IN);
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
}
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/**
* igb_set_i2c_data - Sets the I2C data bit
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
* @data: pointer to hardware structure
* @state: I2C data value (0 or 1) to set
*
* Sets the I2C data bit
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
**/
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
static void igb_set_i2c_data(void *data, int state)
{
struct igb_adapter *adapter = (struct igb_adapter *)data;
struct e1000_hw *hw = &adapter->hw;
s32 i2cctl = rd32(E1000_I2CPARAMS);
if (state)
i2cctl |= E1000_I2C_DATA_OUT;
else
i2cctl &= ~E1000_I2C_DATA_OUT;
i2cctl &= ~E1000_I2C_DATA_OE_N;
i2cctl |= E1000_I2C_CLK_OE_N;
wr32(E1000_I2CPARAMS, i2cctl);
wrfl();
}
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/**
* igb_set_i2c_clk - Sets the I2C SCL clock
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
* @data: pointer to hardware structure
* @state: state to set clock
*
* Sets the I2C clock line to state
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
**/
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
static void igb_set_i2c_clk(void *data, int state)
{
struct igb_adapter *adapter = (struct igb_adapter *)data;
struct e1000_hw *hw = &adapter->hw;
s32 i2cctl = rd32(E1000_I2CPARAMS);
if (state) {
i2cctl |= E1000_I2C_CLK_OUT;
i2cctl &= ~E1000_I2C_CLK_OE_N;
} else {
i2cctl &= ~E1000_I2C_CLK_OUT;
i2cctl &= ~E1000_I2C_CLK_OE_N;
}
wr32(E1000_I2CPARAMS, i2cctl);
wrfl();
}
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/**
* igb_get_i2c_clk - Gets the I2C SCL clock state
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
* @data: pointer to hardware structure
*
* Gets the I2C clock state
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
**/
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
static int igb_get_i2c_clk(void *data)
{
struct igb_adapter *adapter = (struct igb_adapter *)data;
struct e1000_hw *hw = &adapter->hw;
s32 i2cctl = rd32(E1000_I2CPARAMS);
igb: Cleanups to remove return parentheses This patch fixes ERROR:RETURN_PARENTHESES from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 02:11:17 +00:00
return !!(i2cctl & E1000_I2C_CLK_IN);
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
}
static const struct i2c_algo_bit_data igb_i2c_algo = {
.setsda = igb_set_i2c_data,
.setscl = igb_set_i2c_clk,
.getsda = igb_get_i2c_data,
.getscl = igb_get_i2c_clk,
.udelay = 5,
.timeout = 20,
};
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_get_hw_dev - return device
* @hw: pointer to hardware structure
*
* used by hardware layer to print debugging information
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: update hw_debug macro to make use of netdev_dbg call This change updates the igb driver to make use of the netdev_dbg function macros now provided in netdevice.h This is meant to be provided as an alternative to the patch provided by Joe Perches. It also removes igb_get_time_str since I found that it is unused code that is no longer used even in debug. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-25 13:10:08 +00:00
struct net_device *igb_get_hw_dev(struct e1000_hw *hw)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct igb_adapter *adapter = hw->back;
igb: update hw_debug macro to make use of netdev_dbg call This change updates the igb driver to make use of the netdev_dbg function macros now provided in netdevice.h This is meant to be provided as an alternative to the patch provided by Joe Perches. It also removes igb_get_time_str since I found that it is unused code that is no longer used even in debug. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-25 13:10:08 +00:00
return adapter->netdev;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: access to NIC time Adds the register definitions and code to read the time register. Signed-off-by: John Ronciak <john.ronciak@intel.com> Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-12 05:03:41 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_init_module - Driver Registration Routine
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_init_module is the first routine called when the driver is
* loaded. All it does is register with the PCI subsystem.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static int __init igb_init_module(void)
{
int ret;
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
pr_info("%s - version %s\n",
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_driver_string, igb_driver_version);
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
pr_info("%s\n", igb_copyright);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add IGB_DCA instead of selecting INTEL_IOATDMA Add a bool IGB_DCA defined to y if IGB and DCA are enabled, but IGB isn't y while DCA=m. And thus remove the need to select INTEL_IOATDMA when IGB is enabled, so that non-x86 architectures can build the igb driver. Based on work/patch from Brice Goglin <brice@myri.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-10-17 18:08:31 +00:00
#ifdef CONFIG_IGB_DCA
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
dca_register_notify(&dca_notifier);
#endif
igb: Fix DCA errors and do not use context index for 82576 82576 was being incorrectly flagged as needing a context index. It does not as each ring has it's own table of 2 contexts. Driver was registering after registering the driver instead of the other way around. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-31 08:52:30 +00:00
ret = pci_register_driver(&igb_driver);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return ret;
}
module_init(igb_init_module);
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_exit_module - Driver Exit Cleanup Routine
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_exit_module is called just before the driver is removed
* from memory.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void __exit igb_exit_module(void)
{
igb: add IGB_DCA instead of selecting INTEL_IOATDMA Add a bool IGB_DCA defined to y if IGB and DCA are enabled, but IGB isn't y while DCA=m. And thus remove the need to select INTEL_IOATDMA when IGB is enabled, so that non-x86 architectures can build the igb driver. Based on work/patch from Brice Goglin <brice@myri.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-10-17 18:08:31 +00:00
#ifdef CONFIG_IGB_DCA
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
dca_unregister_notify(&dca_notifier);
#endif
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
pci_unregister_driver(&igb_driver);
}
module_exit(igb_exit_module);
igb: re-order queues to support cleaner use of ivar on 82576 The 82576 adapter orders the queues in pairs when virtualization is in use. The queue ordering previously conflicted with the ordering when sr-iov was enabled. This new ordering allows a PF to allocate 2 queues without using any VF resources. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 09:34:11 +00:00
#define Q_IDX_82576(i) (((i & 0x1) << 3) + (i >> 1))
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_cache_ring_register - Descriptor ring to register mapping
* @adapter: board private structure to initialize
igb: re-order queues to support cleaner use of ivar on 82576 The 82576 adapter orders the queues in pairs when virtualization is in use. The queue ordering previously conflicted with the ordering when sr-iov was enabled. This new ordering allows a PF to allocate 2 queues without using any VF resources. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 09:34:11 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Once we know the feature-set enabled for the device, we'll cache
* the register offset the descriptor ring is assigned to.
igb: re-order queues to support cleaner use of ivar on 82576 The 82576 adapter orders the queues in pairs when virtualization is in use. The queue ordering previously conflicted with the ordering when sr-iov was enabled. This new ordering allows a PF to allocate 2 queues without using any VF resources. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 09:34:11 +00:00
**/
static void igb_cache_ring_register(struct igb_adapter *adapter)
{
igb: change queue ordering for 82576 based adapters This patch changes the queue ordering for 82576 adapters so that if VFs are enabled the queues will first be allocated out of the PF pool. Any remaining queues will be allocated out of other VMDq pools. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:40 +00:00
int i = 0, j = 0;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
u32 rbase_offset = adapter->vfs_allocated_count;
igb: re-order queues to support cleaner use of ivar on 82576 The 82576 adapter orders the queues in pairs when virtualization is in use. The queue ordering previously conflicted with the ordering when sr-iov was enabled. This new ordering allows a PF to allocate 2 queues without using any VF resources. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 09:34:11 +00:00
switch (adapter->hw.mac.type) {
case e1000_82576:
/* The queues are allocated for virtualization such that VF 0
* is allocated queues 0 and 8, VF 1 queues 1 and 9, etc.
* In order to avoid collision we start at the first free queue
* and continue consuming queues in the same sequence
*/
igb: change queue ordering for 82576 based adapters This patch changes the queue ordering for 82576 adapters so that if VFs are enabled the queues will first be allocated out of the PF pool. Any remaining queues will be allocated out of other VMDq pools. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:40 +00:00
if (adapter->vfs_allocated_count) {
igb: when number of CPUs > 4 combine tx/rx queues to allow more queues This patch makes it so that nics such as 82576 and newer can support more hardware queues when there are more than 4 cpus by combining a tx/rx queue pair onto one interrupt so that 8 queue pairs can be supported and thus allow for more queues. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:19 +00:00
for (; i < adapter->rss_queues; i++)
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
adapter->rx_ring[i]->reg_idx = rbase_offset +
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
Q_IDX_82576(i);
igb: change queue ordering for 82576 based adapters This patch changes the queue ordering for 82576 adapters so that if VFs are enabled the queues will first be allocated out of the PF pool. Any remaining queues will be allocated out of other VMDq pools. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:40 +00:00
}
igb: Cleanups to fix missing break in switch statements This patch fixes WARNING:MISSING_BREAK found with checkpatch check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-17 04:10:13 +00:00
/* Fall through */
igb: re-order queues to support cleaner use of ivar on 82576 The 82576 adapter orders the queues in pairs when virtualization is in use. The queue ordering previously conflicted with the ordering when sr-iov was enabled. This new ordering allows a PF to allocate 2 queues without using any VF resources. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 09:34:11 +00:00
case e1000_82575:
igb: Add full support for 82580 devices This patch makes use of the 82580 PHY and MAC support added and adds a set of supported device IDs for said hardware. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-19 12:42:21 +00:00
case e1000_82580:
igb: add support for Intel I350 Gigabit Network Connection This patch adds support for the the I350 Gigabit network connection which is the follow-on part to the 82580. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> CC: James Hearn <james.r.hearn@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-22 14:08:06 +00:00
case e1000_i350:
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
case e1000_i354:
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
case e1000_i210:
case e1000_i211:
igb: Cleanups to fix missing break in switch statements This patch fixes WARNING:MISSING_BREAK found with checkpatch check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-17 04:10:13 +00:00
/* Fall through */
igb: re-order queues to support cleaner use of ivar on 82576 The 82576 adapter orders the queues in pairs when virtualization is in use. The queue ordering previously conflicted with the ordering when sr-iov was enabled. This new ordering allows a PF to allocate 2 queues without using any VF resources. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 09:34:11 +00:00
default:
igb: change queue ordering for 82576 based adapters This patch changes the queue ordering for 82576 adapters so that if VFs are enabled the queues will first be allocated out of the PF pool. Any remaining queues will be allocated out of other VMDq pools. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:40 +00:00
for (; i < adapter->num_rx_queues; i++)
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
adapter->rx_ring[i]->reg_idx = rbase_offset + i;
igb: change queue ordering for 82576 based adapters This patch changes the queue ordering for 82576 adapters so that if VFs are enabled the queues will first be allocated out of the PF pool. Any remaining queues will be allocated out of other VMDq pools. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:40 +00:00
for (; j < adapter->num_tx_queues; j++)
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
adapter->tx_ring[j]->reg_idx = rbase_offset + j;
igb: re-order queues to support cleaner use of ivar on 82576 The 82576 adapter orders the queues in pairs when virtualization is in use. The queue ordering previously conflicted with the ordering when sr-iov was enabled. This new ordering allows a PF to allocate 2 queues without using any VF resources. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 09:34:11 +00:00
break;
}
}
igb: add register rd/wr for surprise removal Add initial register rd/wr for surprise removal (LER). Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-13 04:29:01 +00:00
u32 igb_rd32(struct e1000_hw *hw, u32 reg)
{
struct igb_adapter *igb = container_of(hw, struct igb_adapter, hw);
locking/atomics: COCCINELLE/treewide: Convert trivial ACCESS_ONCE() patterns to READ_ONCE()/WRITE_ONCE() Please do not apply this to mainline directly, instead please re-run the coccinelle script shown below and apply its output. For several reasons, it is desirable to use {READ,WRITE}_ONCE() in preference to ACCESS_ONCE(), and new code is expected to use one of the former. So far, there's been no reason to change most existing uses of ACCESS_ONCE(), as these aren't harmful, and changing them results in churn. However, for some features, the read/write distinction is critical to correct operation. To distinguish these cases, separate read/write accessors must be used. This patch migrates (most) remaining ACCESS_ONCE() instances to {READ,WRITE}_ONCE(), using the following coccinelle script: ---- // Convert trivial ACCESS_ONCE() uses to equivalent READ_ONCE() and // WRITE_ONCE() // $ make coccicheck COCCI=/home/mark/once.cocci SPFLAGS="--include-headers" MODE=patch virtual patch @ depends on patch @ expression E1, E2; @@ - ACCESS_ONCE(E1) = E2 + WRITE_ONCE(E1, E2) @ depends on patch @ expression E; @@ - ACCESS_ONCE(E) + READ_ONCE(E) ---- Signed-off-by: Mark Rutland <mark.rutland@arm.com> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: davem@davemloft.net Cc: linux-arch@vger.kernel.org Cc: mpe@ellerman.id.au Cc: shuah@kernel.org Cc: snitzer@redhat.com Cc: thor.thayer@linux.intel.com Cc: tj@kernel.org Cc: viro@zeniv.linux.org.uk Cc: will.deacon@arm.com Link: http://lkml.kernel.org/r/1508792849-3115-19-git-send-email-paulmck@linux.vnet.ibm.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-10-23 21:07:29 +00:00
u8 __iomem *hw_addr = READ_ONCE(hw->hw_addr);
igb: add register rd/wr for surprise removal Add initial register rd/wr for surprise removal (LER). Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-13 04:29:01 +00:00
u32 value = 0;
if (E1000_REMOVED(hw_addr))
return ~value;
value = readl(&hw_addr[reg]);
/* reads should not return all F's */
if (!(~value) && (!reg || !(~readl(hw_addr)))) {
struct net_device *netdev = igb->netdev;
hw->hw_addr = NULL;
igb: Do not call netif_device_detach() when PCIe link goes missing When the driver notices that PCIe link is gone by reading 0xffffffff from a register it clears hw->hw_addr and then calls netif_device_detach(). This happens when the PCIe device is physically unplugged for example the user disconnected the Thunderbolt cable. However, netif_device_detach() prevents netif_unregister() from bringing the device down properly including tearing down MSI-X vectors. This triggers following crash during the driver removal: igb 0000:0b:00.0 enp11s0f0: PCIe link lost, device now detached ------------[ cut here ]------------ kernel BUG at drivers/pci/msi.c:352! invalid opcode: 0000 [#1] PREEMPT SMP PTI ... Call Trace: pci_disable_msix+0xc9/0xf0 igb_reset_interrupt_capability+0x58/0x60 [igb] igb_remove+0x90/0x100 [igb] pci_device_remove+0x31/0xa0 device_release_driver_internal+0x152/0x210 pci_stop_bus_device+0x78/0xa0 pci_stop_bus_device+0x38/0xa0 pci_stop_bus_device+0x38/0xa0 pci_stop_bus_device+0x26/0xa0 pci_stop_bus_device+0x38/0xa0 pci_stop_and_remove_bus_device+0x9/0x20 trim_stale_devices+0xee/0x130 ? _raw_spin_unlock_irqrestore+0xf/0x30 trim_stale_devices+0x8f/0x130 ? _raw_spin_unlock_irqrestore+0xf/0x30 trim_stale_devices+0xa1/0x130 ? get_slot_status+0x8b/0xc0 acpiphp_check_bridge.part.7+0xf9/0x140 acpiphp_hotplug_notify+0x170/0x1f0 ... To prevent the crash do not call netif_device_detach() in igb_rd32(). This should be fine because hw->hw_addr is set to NULL preventing future hardware access of the now missing device. Link: https://bugzilla.kernel.org/show_bug.cgi?id=198181 Reported-by: Ferenc Boldog <ferenc.boldog@gmail.com> Reported-by: Nikolay Bogoychev <nheart@gmail.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-01-23 10:28:41 +00:00
netdev_err(netdev, "PCIe link lost\n");
igb: add register rd/wr for surprise removal Add initial register rd/wr for surprise removal (LER). Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-13 04:29:01 +00:00
}
return value;
}
igb: cleanup IVAR configuration This change is meant to cleanup some of the IVAR register configuration. igb_assign_vector had become pretty large with multiple copies of the same general code for setting the IVAR. This change consolidates most of that code by adding the igb_write_ivar function which allows us just to compute the index and offset and then use that information to setup the IVAR. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:52 +00:00
/**
* igb_write_ivar - configure ivar for given MSI-X vector
* @hw: pointer to the HW structure
* @msix_vector: vector number we are allocating to a given ring
* @index: row index of IVAR register to write within IVAR table
* @offset: column offset of in IVAR, should be multiple of 8
*
* This function is intended to handle the writing of the IVAR register
* for adapters 82576 and newer. The IVAR table consists of 2 columns,
* each containing an cause allocation for an Rx and Tx ring, and a
* variable number of rows depending on the number of queues supported.
**/
static void igb_write_ivar(struct e1000_hw *hw, int msix_vector,
int index, int offset)
{
u32 ivar = array_rd32(E1000_IVAR0, index);
/* clear any bits that are currently set */
ivar &= ~((u32)0xFF << offset);
/* write vector and valid bit */
ivar |= (msix_vector | E1000_IVAR_VALID) << offset;
array_wr32(E1000_IVAR0, index, ivar);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#define IGB_N0_QUEUE -1
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
static void igb_assign_vector(struct igb_q_vector *q_vector, int msix_vector)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
struct igb_adapter *adapter = q_vector->adapter;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
int rx_queue = IGB_N0_QUEUE;
int tx_queue = IGB_N0_QUEUE;
igb: cleanup IVAR configuration This change is meant to cleanup some of the IVAR register configuration. igb_assign_vector had become pretty large with multiple copies of the same general code for setting the IVAR. This change consolidates most of that code by adding the igb_write_ivar function which allows us just to compute the index and offset and then use that information to setup the IVAR. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:52 +00:00
u32 msixbm = 0;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
if (q_vector->rx.ring)
rx_queue = q_vector->rx.ring->reg_idx;
if (q_vector->tx.ring)
tx_queue = q_vector->tx.ring->reg_idx;
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
switch (hw->mac.type) {
case e1000_82575:
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* The 82575 assigns vectors using a bitmask, which matches the
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* bitmask for the EICR/EIMS/EIMC registers. To assign one
* or more queues to a vector, we write the appropriate bits
* into the MSIXBM register for that vector.
*/
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (rx_queue > IGB_N0_QUEUE)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
msixbm = E1000_EICR_RX_QUEUE0 << rx_queue;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (tx_queue > IGB_N0_QUEUE)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
msixbm |= E1000_EICR_TX_QUEUE0 << tx_queue;
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
if (!(adapter->flags & IGB_FLAG_HAS_MSIX) && msix_vector == 0)
igb: make certain to reassign legacy interrupt vectors after reset This change corrects an issue that will cause false hangs when using either 82575 or 82580 in legacy interrupt mode. The issue is caused when there is a slow traffic flow and an "ethtool -r" is executed while using legacy or MSI interrupts. MSI-X is not affected by this issue due to the fact that we were already reconfiguring the vectors after reset. If possible it would be best to push this for net-2.6 since it is resolving a bug but if that is not possible then net-next-2.6 will be fine. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-03 21:59:51 +00:00
msixbm |= E1000_EIMS_OTHER;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
array_wr32(E1000_MSIXBM(0), msix_vector, msixbm);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
q_vector->eims_value = msixbm;
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
break;
case e1000_82576:
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* 82576 uses a table that essentially consists of 2 columns
igb: cleanup IVAR configuration This change is meant to cleanup some of the IVAR register configuration. igb_assign_vector had become pretty large with multiple copies of the same general code for setting the IVAR. This change consolidates most of that code by adding the igb_write_ivar function which allows us just to compute the index and offset and then use that information to setup the IVAR. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:52 +00:00
* with 8 rows. The ordering is column-major so we use the
* lower 3 bits as the row index, and the 4th bit as the
* column offset.
*/
if (rx_queue > IGB_N0_QUEUE)
igb_write_ivar(hw, msix_vector,
rx_queue & 0x7,
(rx_queue & 0x8) << 1);
if (tx_queue > IGB_N0_QUEUE)
igb_write_ivar(hw, msix_vector,
tx_queue & 0x7,
((tx_queue & 0x8) << 1) + 8);
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
q_vector->eims_value = BIT(msix_vector);
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
break;
igb: Add full support for 82580 devices This patch makes use of the 82580 PHY and MAC support added and adds a set of supported device IDs for said hardware. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-19 12:42:21 +00:00
case e1000_82580:
igb: add support for Intel I350 Gigabit Network Connection This patch adds support for the the I350 Gigabit network connection which is the follow-on part to the 82580. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> CC: James Hearn <james.r.hearn@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-22 14:08:06 +00:00
case e1000_i350:
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
case e1000_i354:
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
case e1000_i210:
case e1000_i211:
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* On 82580 and newer adapters the scheme is similar to 82576
igb: cleanup IVAR configuration This change is meant to cleanup some of the IVAR register configuration. igb_assign_vector had become pretty large with multiple copies of the same general code for setting the IVAR. This change consolidates most of that code by adding the igb_write_ivar function which allows us just to compute the index and offset and then use that information to setup the IVAR. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:52 +00:00
* however instead of ordering column-major we have things
* ordered row-major. So we traverse the table by using
* bit 0 as the column offset, and the remaining bits as the
* row index.
*/
if (rx_queue > IGB_N0_QUEUE)
igb_write_ivar(hw, msix_vector,
rx_queue >> 1,
(rx_queue & 0x1) << 4);
if (tx_queue > IGB_N0_QUEUE)
igb_write_ivar(hw, msix_vector,
tx_queue >> 1,
((tx_queue & 0x1) << 4) + 8);
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
q_vector->eims_value = BIT(msix_vector);
igb: Add full support for 82580 devices This patch makes use of the 82580 PHY and MAC support added and adds a set of supported device IDs for said hardware. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-19 12:42:21 +00:00
break;
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
default:
BUG();
break;
}
igb: remove unecessary q_vector declarations and remove itr_shift This change removes the use of itr_shift since a mac type call can be used just as easily to identify the only HW that needs to have the itr shifted. In addition it removes two unecessary declarations of a q_vector pointer from the initialization path. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:00:41 +00:00
/* add q_vector eims value to global eims_enable_mask */
adapter->eims_enable_mask |= q_vector->eims_value;
/* configure q_vector to set itr on first interrupt */
q_vector->set_itr = 1;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_configure_msix - Configure MSI-X hardware
* @adapter: board private structure to initialize
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_configure_msix sets up the hardware to properly
* generate MSI-X interrupts.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_configure_msix(struct igb_adapter *adapter)
{
u32 tmp;
int i, vector = 0;
struct e1000_hw *hw = &adapter->hw;
adapter->eims_enable_mask = 0;
/* set vector for other causes, i.e. link changes */
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
switch (hw->mac.type) {
case e1000_82575:
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
tmp = rd32(E1000_CTRL_EXT);
/* enable MSI-X PBA support*/
tmp |= E1000_CTRL_EXT_PBA_CLR;
/* Auto-Mask interrupts upon ICR read. */
tmp |= E1000_CTRL_EXT_EIAME;
tmp |= E1000_CTRL_EXT_IRCA;
wr32(E1000_CTRL_EXT, tmp);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
/* enable msix_other interrupt */
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
array_wr32(E1000_MSIXBM(0), vector++, E1000_EIMS_OTHER);
igb: add NAPI Rx queue support Update the NAPI implementation to use the new napi_struct infrstructure. This removes the need of multiple net_device structs to implement a multiqueue NAPI. Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:39 +00:00
adapter->eims_other = E1000_EIMS_OTHER;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
break;
case e1000_82576:
igb: Add full support for 82580 devices This patch makes use of the 82580 PHY and MAC support added and adds a set of supported device IDs for said hardware. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-19 12:42:21 +00:00
case e1000_82580:
igb: add support for Intel I350 Gigabit Network Connection This patch adds support for the the I350 Gigabit network connection which is the follow-on part to the 82580. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> CC: James Hearn <james.r.hearn@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-22 14:08:06 +00:00
case e1000_i350:
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
case e1000_i354:
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
case e1000_i210:
case e1000_i211:
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
/* Turn on MSI-X capability first, or our settings
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* won't stick. And it will take days to debug.
*/
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
wr32(E1000_GPIE, E1000_GPIE_MSIX_MODE |
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
E1000_GPIE_PBA | E1000_GPIE_EIAME |
E1000_GPIE_NSICR);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
/* enable msix_other interrupt */
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
adapter->eims_other = BIT(vector);
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
tmp = (vector++ | E1000_IVAR_VALID) << 8;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
wr32(E1000_IVAR_MISC, tmp);
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
break;
default:
/* do nothing, since nothing else supports MSI-X */
break;
} /* switch (hw->mac.type) */
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
adapter->eims_enable_mask |= adapter->eims_other;
igb: remove unecessary q_vector declarations and remove itr_shift This change removes the use of itr_shift since a mac type call can be used just as easily to identify the only HW that needs to have the itr shifted. In addition it removes two unecessary declarations of a q_vector pointer from the initialization path. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:00:41 +00:00
for (i = 0; i < adapter->num_q_vectors; i++)
igb_assign_vector(adapter->q_vector[i], vector++);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
wrfl();
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_request_msix - Initialize MSI-X interrupts
* @adapter: board private structure to initialize
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_request_msix allocates MSI-X vectors and requests interrupts from the
* kernel.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static int igb_request_msix(struct igb_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
igb: release already assigned MSI-X interrupts if setup fails During MSI-X setup the system might run out of vectors. If this happens the already assigned vectors for this NIC should be freed before trying the disable MSI-X. Failing to do so results in the following oops. kernel BUG at drivers/pci/msi.c:341! [...] Call Trace: [<ffffffff8128f39d>] pci_disable_msix+0x3d/0x60 [<ffffffffa037d1ce>] igb_reset_interrupt_capability+0x27/0x5c [igb] [<ffffffffa037d229>] igb_clear_interrupt_scheme+0x26/0x2d [igb] [<ffffffffa0384268>] igb_request_irq+0x73/0x297 [igb] [<ffffffffa0384554>] __igb_open+0xc8/0x223 [igb] [<ffffffffa0384815>] igb_open+0x13/0x15 [igb] [<ffffffff8144592f>] __dev_open+0xbf/0x120 [<ffffffff81443e51>] __dev_change_flags+0xa1/0x180 [<ffffffff81445828>] dev_change_flags+0x28/0x70 [<ffffffff814af537>] devinet_ioctl+0x5b7/0x620 [<ffffffff814b01c8>] inet_ioctl+0x88/0xa0 [<ffffffff8142e8a0>] sock_do_ioctl+0x30/0x70 [<ffffffff8142ecf2>] sock_ioctl+0x72/0x270 [<ffffffff8118062c>] do_vfs_ioctl+0x8c/0x340 [<ffffffff81180981>] sys_ioctl+0xa1/0xb0 [<ffffffff815161a9>] system_call_fastpath+0x16/0x1b Code: 48 89 df e8 1f 40 ed ff 4d 39 e6 49 8b 45 10 75 b6 48 83 c4 18 5b 41 5c 41 5d 41 5e 41 5f c9 c3 48 8b 7b 20 e8 3e 91 db ff eb ae <0f> 0b eb fe 0f 1f 84 00 00 00 00 00 55 48 89 e5 0f 1f 44 00 00 RIP [<ffffffff8128e144>] free_msi_irqs+0x124/0x130 RSP <ffff880037503bd8> Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:17 +00:00
int i, err = 0, vector = 0, free_vector = 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
err = request_irq(adapter->msix_entries[vector].vector,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
igb_msix_other, 0, netdev->name, adapter);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (err)
igb: release already assigned MSI-X interrupts if setup fails During MSI-X setup the system might run out of vectors. If this happens the already assigned vectors for this NIC should be freed before trying the disable MSI-X. Failing to do so results in the following oops. kernel BUG at drivers/pci/msi.c:341! [...] Call Trace: [<ffffffff8128f39d>] pci_disable_msix+0x3d/0x60 [<ffffffffa037d1ce>] igb_reset_interrupt_capability+0x27/0x5c [igb] [<ffffffffa037d229>] igb_clear_interrupt_scheme+0x26/0x2d [igb] [<ffffffffa0384268>] igb_request_irq+0x73/0x297 [igb] [<ffffffffa0384554>] __igb_open+0xc8/0x223 [igb] [<ffffffffa0384815>] igb_open+0x13/0x15 [igb] [<ffffffff8144592f>] __dev_open+0xbf/0x120 [<ffffffff81443e51>] __dev_change_flags+0xa1/0x180 [<ffffffff81445828>] dev_change_flags+0x28/0x70 [<ffffffff814af537>] devinet_ioctl+0x5b7/0x620 [<ffffffff814b01c8>] inet_ioctl+0x88/0xa0 [<ffffffff8142e8a0>] sock_do_ioctl+0x30/0x70 [<ffffffff8142ecf2>] sock_ioctl+0x72/0x270 [<ffffffff8118062c>] do_vfs_ioctl+0x8c/0x340 [<ffffffff81180981>] sys_ioctl+0xa1/0xb0 [<ffffffff815161a9>] system_call_fastpath+0x16/0x1b Code: 48 89 df e8 1f 40 ed ff 4d 39 e6 49 8b 45 10 75 b6 48 83 c4 18 5b 41 5c 41 5d 41 5e 41 5f c9 c3 48 8b 7b 20 e8 3e 91 db ff eb ae <0f> 0b eb fe 0f 1f 84 00 00 00 00 00 55 48 89 e5 0f 1f 44 00 00 RIP [<ffffffff8128e144>] free_msi_irqs+0x124/0x130 RSP <ffff880037503bd8> Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:17 +00:00
goto err_out;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
for (i = 0; i < adapter->num_q_vectors; i++) {
struct igb_q_vector *q_vector = adapter->q_vector[i];
igb: release already assigned MSI-X interrupts if setup fails During MSI-X setup the system might run out of vectors. If this happens the already assigned vectors for this NIC should be freed before trying the disable MSI-X. Failing to do so results in the following oops. kernel BUG at drivers/pci/msi.c:341! [...] Call Trace: [<ffffffff8128f39d>] pci_disable_msix+0x3d/0x60 [<ffffffffa037d1ce>] igb_reset_interrupt_capability+0x27/0x5c [igb] [<ffffffffa037d229>] igb_clear_interrupt_scheme+0x26/0x2d [igb] [<ffffffffa0384268>] igb_request_irq+0x73/0x297 [igb] [<ffffffffa0384554>] __igb_open+0xc8/0x223 [igb] [<ffffffffa0384815>] igb_open+0x13/0x15 [igb] [<ffffffff8144592f>] __dev_open+0xbf/0x120 [<ffffffff81443e51>] __dev_change_flags+0xa1/0x180 [<ffffffff81445828>] dev_change_flags+0x28/0x70 [<ffffffff814af537>] devinet_ioctl+0x5b7/0x620 [<ffffffff814b01c8>] inet_ioctl+0x88/0xa0 [<ffffffff8142e8a0>] sock_do_ioctl+0x30/0x70 [<ffffffff8142ecf2>] sock_ioctl+0x72/0x270 [<ffffffff8118062c>] do_vfs_ioctl+0x8c/0x340 [<ffffffff81180981>] sys_ioctl+0xa1/0xb0 [<ffffffff815161a9>] system_call_fastpath+0x16/0x1b Code: 48 89 df e8 1f 40 ed ff 4d 39 e6 49 8b 45 10 75 b6 48 83 c4 18 5b 41 5c 41 5d 41 5e 41 5f c9 c3 48 8b 7b 20 e8 3e 91 db ff eb ae <0f> 0b eb fe 0f 1f 84 00 00 00 00 00 55 48 89 e5 0f 1f 44 00 00 RIP [<ffffffff8128e144>] free_msi_irqs+0x124/0x130 RSP <ffff880037503bd8> Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:17 +00:00
vector++;
igb: improve handling of disconnected adapters Clean up array_rd32 so that it uses igb_rd32 the same as rd32, per the suggestion of Alexander Duyck, and use io_addr in more places, so that we don't have the need to call E1000_REMOVED (which simply looks for a null hw_addr) nearly as much. Signed-off-by: Jarod Wilson <jarod@redhat.com> Acked-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-10-19 15:52:04 +00:00
q_vector->itr_register = adapter->io_addr + E1000_EITR(vector);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
if (q_vector->rx.ring && q_vector->tx.ring)
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
sprintf(q_vector->name, "%s-TxRx-%u", netdev->name,
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
q_vector->rx.ring->queue_index);
else if (q_vector->tx.ring)
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
sprintf(q_vector->name, "%s-tx-%u", netdev->name,
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
q_vector->tx.ring->queue_index);
else if (q_vector->rx.ring)
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
sprintf(q_vector->name, "%s-rx-%u", netdev->name,
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
q_vector->rx.ring->queue_index);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
else
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
sprintf(q_vector->name, "%s-unused", netdev->name);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
err = request_irq(adapter->msix_entries[vector].vector,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
igb_msix_ring, 0, q_vector->name,
q_vector);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (err)
igb: release already assigned MSI-X interrupts if setup fails During MSI-X setup the system might run out of vectors. If this happens the already assigned vectors for this NIC should be freed before trying the disable MSI-X. Failing to do so results in the following oops. kernel BUG at drivers/pci/msi.c:341! [...] Call Trace: [<ffffffff8128f39d>] pci_disable_msix+0x3d/0x60 [<ffffffffa037d1ce>] igb_reset_interrupt_capability+0x27/0x5c [igb] [<ffffffffa037d229>] igb_clear_interrupt_scheme+0x26/0x2d [igb] [<ffffffffa0384268>] igb_request_irq+0x73/0x297 [igb] [<ffffffffa0384554>] __igb_open+0xc8/0x223 [igb] [<ffffffffa0384815>] igb_open+0x13/0x15 [igb] [<ffffffff8144592f>] __dev_open+0xbf/0x120 [<ffffffff81443e51>] __dev_change_flags+0xa1/0x180 [<ffffffff81445828>] dev_change_flags+0x28/0x70 [<ffffffff814af537>] devinet_ioctl+0x5b7/0x620 [<ffffffff814b01c8>] inet_ioctl+0x88/0xa0 [<ffffffff8142e8a0>] sock_do_ioctl+0x30/0x70 [<ffffffff8142ecf2>] sock_ioctl+0x72/0x270 [<ffffffff8118062c>] do_vfs_ioctl+0x8c/0x340 [<ffffffff81180981>] sys_ioctl+0xa1/0xb0 [<ffffffff815161a9>] system_call_fastpath+0x16/0x1b Code: 48 89 df e8 1f 40 ed ff 4d 39 e6 49 8b 45 10 75 b6 48 83 c4 18 5b 41 5c 41 5d 41 5e 41 5f c9 c3 48 8b 7b 20 e8 3e 91 db ff eb ae <0f> 0b eb fe 0f 1f 84 00 00 00 00 00 55 48 89 e5 0f 1f 44 00 00 RIP [<ffffffff8128e144>] free_msi_irqs+0x124/0x130 RSP <ffff880037503bd8> Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:17 +00:00
goto err_free;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb_configure_msix(adapter);
return 0;
igb: release already assigned MSI-X interrupts if setup fails During MSI-X setup the system might run out of vectors. If this happens the already assigned vectors for this NIC should be freed before trying the disable MSI-X. Failing to do so results in the following oops. kernel BUG at drivers/pci/msi.c:341! [...] Call Trace: [<ffffffff8128f39d>] pci_disable_msix+0x3d/0x60 [<ffffffffa037d1ce>] igb_reset_interrupt_capability+0x27/0x5c [igb] [<ffffffffa037d229>] igb_clear_interrupt_scheme+0x26/0x2d [igb] [<ffffffffa0384268>] igb_request_irq+0x73/0x297 [igb] [<ffffffffa0384554>] __igb_open+0xc8/0x223 [igb] [<ffffffffa0384815>] igb_open+0x13/0x15 [igb] [<ffffffff8144592f>] __dev_open+0xbf/0x120 [<ffffffff81443e51>] __dev_change_flags+0xa1/0x180 [<ffffffff81445828>] dev_change_flags+0x28/0x70 [<ffffffff814af537>] devinet_ioctl+0x5b7/0x620 [<ffffffff814b01c8>] inet_ioctl+0x88/0xa0 [<ffffffff8142e8a0>] sock_do_ioctl+0x30/0x70 [<ffffffff8142ecf2>] sock_ioctl+0x72/0x270 [<ffffffff8118062c>] do_vfs_ioctl+0x8c/0x340 [<ffffffff81180981>] sys_ioctl+0xa1/0xb0 [<ffffffff815161a9>] system_call_fastpath+0x16/0x1b Code: 48 89 df e8 1f 40 ed ff 4d 39 e6 49 8b 45 10 75 b6 48 83 c4 18 5b 41 5c 41 5d 41 5e 41 5f c9 c3 48 8b 7b 20 e8 3e 91 db ff eb ae <0f> 0b eb fe 0f 1f 84 00 00 00 00 00 55 48 89 e5 0f 1f 44 00 00 RIP [<ffffffff8128e144>] free_msi_irqs+0x124/0x130 RSP <ffff880037503bd8> Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:17 +00:00
err_free:
/* free already assigned IRQs */
free_irq(adapter->msix_entries[free_vector++].vector, adapter);
vector--;
for (i = 0; i < vector; i++) {
free_irq(adapter->msix_entries[free_vector++].vector,
adapter->q_vector[i]);
}
err_out:
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return err;
}
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_free_q_vector - Free memory allocated for specific interrupt vector
* @adapter: board private structure to initialize
* @v_idx: Index of vector to be freed
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
*
igb: Fix queue allocation method to accommodate changing during runtime When changing number of queues using ethtool's set_channels during runtime, a queue allocation could fail, which can leave the device in a down state. In order to preserve the usability of the device in this scenario, this patch changes the driver to allocate the number of queues only if they have not been allocated already. The first allocation is then done for the max number of queues, which is the default queues for this driver. With this change, queue quantity changes are not subject to queue allocation failures. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:29 +00:00
* This function frees the memory allocated to the q_vector.
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
**/
static void igb_free_q_vector(struct igb_adapter *adapter, int v_idx)
{
struct igb_q_vector *q_vector = adapter->q_vector[v_idx];
igb: Fix queue allocation method to accommodate changing during runtime When changing number of queues using ethtool's set_channels during runtime, a queue allocation could fail, which can leave the device in a down state. In order to preserve the usability of the device in this scenario, this patch changes the driver to allocate the number of queues only if they have not been allocated already. The first allocation is then done for the max number of queues, which is the default queues for this driver. With this change, queue quantity changes are not subject to queue allocation failures. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:29 +00:00
adapter->q_vector[v_idx] = NULL;
/* igb_get_stats64() might access the rings on this vector,
* we must wait a grace period before freeing it.
*/
igb: Fixes needed for surprise removal support This patch adds some checks in order to prevent panic's on surprise removal of devices during S0, S3, S4. Without this patch, Thunderbolt type device removal will panic the system. Signed-off-by: Yanir Lubetkin <yanirx.lubetkin@intel.com> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-22 07:52:54 +00:00
if (q_vector)
kfree_rcu(q_vector, rcu);
igb: Fix queue allocation method to accommodate changing during runtime When changing number of queues using ethtool's set_channels during runtime, a queue allocation could fail, which can leave the device in a down state. In order to preserve the usability of the device in this scenario, this patch changes the driver to allocate the number of queues only if they have not been allocated already. The first allocation is then done for the max number of queues, which is the default queues for this driver. With this change, queue quantity changes are not subject to queue allocation failures. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:29 +00:00
}
/**
* igb_reset_q_vector - Reset config for interrupt vector
* @adapter: board private structure to initialize
* @v_idx: Index of vector to be reset
*
* If NAPI is enabled it will delete any references to the
* NAPI struct. This is preparation for igb_free_q_vector.
**/
static void igb_reset_q_vector(struct igb_adapter *adapter, int v_idx)
{
struct igb_q_vector *q_vector = adapter->q_vector[v_idx];
igb: Fix Null-pointer dereference in igb_reset_q_vector When igb_set_interrupt_capability() calls igb_reset_interrupt_capability() (e.g., because CONFIG_PCI_MSI is unset), num_q_vectors has been set but no vector has yet been allocated. igb_reset_interrupt_capability() will then call igb_reset_q_vector, which assumes that the vector is allocated. As this is not the case, we are accessing a NULL-pointer. This patch fixes it by checking that q_vector is indeed different from NULL. Fixes: 02ef6e1d0b0023 (igb: Fix queue allocation method to accommodate changing during runtime) Cc: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Christoph Paasch <christoph.paasch@uclouvain.be> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-21 10:48:19 +00:00
/* Coming from igb_set_interrupt_capability, the vectors are not yet
* allocated. So, q_vector is NULL so we should stop here.
*/
if (!q_vector)
return;
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
if (q_vector->tx.ring)
adapter->tx_ring[q_vector->tx.ring->queue_index] = NULL;
if (q_vector->rx.ring)
igb: Fix NULL assignment to incorrect variable in igb_reset_q_vector adapter->tx_ring is set to NULL where rx_ring should be. Fixes: 5536d2102a2d ("igb: Combine q_vector and ring allocation into a single function") Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-04-13 09:15:11 +00:00
adapter->rx_ring[q_vector->rx.ring->queue_index] = NULL;
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
netif_napi_del(&q_vector->napi);
igb: Fix queue allocation method to accommodate changing during runtime When changing number of queues using ethtool's set_channels during runtime, a queue allocation could fail, which can leave the device in a down state. In order to preserve the usability of the device in this scenario, this patch changes the driver to allocate the number of queues only if they have not been allocated already. The first allocation is then done for the max number of queues, which is the default queues for this driver. With this change, queue quantity changes are not subject to queue allocation failures. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:29 +00:00
}
static void igb_reset_interrupt_capability(struct igb_adapter *adapter)
{
int v_idx = adapter->num_q_vectors;
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
if (adapter->flags & IGB_FLAG_HAS_MSIX)
igb: Fix queue allocation method to accommodate changing during runtime When changing number of queues using ethtool's set_channels during runtime, a queue allocation could fail, which can leave the device in a down state. In order to preserve the usability of the device in this scenario, this patch changes the driver to allocate the number of queues only if they have not been allocated already. The first allocation is then done for the max number of queues, which is the default queues for this driver. With this change, queue quantity changes are not subject to queue allocation failures. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:29 +00:00
pci_disable_msix(adapter->pdev);
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
else if (adapter->flags & IGB_FLAG_HAS_MSI)
igb: Fix queue allocation method to accommodate changing during runtime When changing number of queues using ethtool's set_channels during runtime, a queue allocation could fail, which can leave the device in a down state. In order to preserve the usability of the device in this scenario, this patch changes the driver to allocate the number of queues only if they have not been allocated already. The first allocation is then done for the max number of queues, which is the default queues for this driver. With this change, queue quantity changes are not subject to queue allocation failures. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:29 +00:00
pci_disable_msi(adapter->pdev);
while (v_idx--)
igb_reset_q_vector(adapter, v_idx);
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
}
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_free_q_vectors - Free memory allocated for interrupt vectors
* @adapter: board private structure to initialize
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* This function frees the memory allocated to the q_vectors. In addition if
* NAPI is enabled it will delete any references to the NAPI struct prior
* to freeing the q_vector.
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
**/
static void igb_free_q_vectors(struct igb_adapter *adapter)
{
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
int v_idx = adapter->num_q_vectors;
adapter->num_tx_queues = 0;
adapter->num_rx_queues = 0;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
adapter->num_q_vectors = 0;
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
igb: Fix queue allocation method to accommodate changing during runtime When changing number of queues using ethtool's set_channels during runtime, a queue allocation could fail, which can leave the device in a down state. In order to preserve the usability of the device in this scenario, this patch changes the driver to allocate the number of queues only if they have not been allocated already. The first allocation is then done for the max number of queues, which is the default queues for this driver. With this change, queue quantity changes are not subject to queue allocation failures. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:29 +00:00
while (v_idx--) {
igb_reset_q_vector(adapter, v_idx);
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
igb_free_q_vector(adapter, v_idx);
igb: Fix queue allocation method to accommodate changing during runtime When changing number of queues using ethtool's set_channels during runtime, a queue allocation could fail, which can leave the device in a down state. In order to preserve the usability of the device in this scenario, this patch changes the driver to allocate the number of queues only if they have not been allocated already. The first allocation is then done for the max number of queues, which is the default queues for this driver. With this change, queue quantity changes are not subject to queue allocation failures. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:29 +00:00
}
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_clear_interrupt_scheme - reset the device to a state of no interrupts
* @adapter: board private structure to initialize
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* This function resets the device so that it has 0 Rx queues, Tx queues, and
* MSI-X interrupts allocated.
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
*/
static void igb_clear_interrupt_scheme(struct igb_adapter *adapter)
{
igb_free_q_vectors(adapter);
igb_reset_interrupt_capability(adapter);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_set_interrupt_capability - set MSI or MSI-X if supported
* @adapter: board private structure to initialize
* @msix: boolean value of MSIX capability
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Attempt to configure interrupts using the best available
* capabilities of the hardware and kernel.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
static void igb_set_interrupt_capability(struct igb_adapter *adapter, bool msix)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
int err;
int numvecs, i;
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
if (!msix)
goto msi_only;
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
adapter->flags |= IGB_FLAG_HAS_MSIX;
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
igb: move initialization of number of queues into set_interrupt_capability This patch moves the initialization of the number of queues into set_interrupt_capability. This allows the number of queues to increase in the unlikely event that the system initially fails to allocate enough msi-x interrupts, does a suspend/resume, and then can allocate enough interrupts on resume. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:15:45 +00:00
/* Number of supported queues. */
igb: when number of CPUs > 4 combine tx/rx queues to allow more queues This patch makes it so that nics such as 82576 and newer can support more hardware queues when there are more than 4 cpus by combining a tx/rx queue pair onto one interrupt so that 8 queue pairs can be supported and thus allow for more queues. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:19 +00:00
adapter->num_rx_queues = adapter->rss_queues;
igb: Use only a single Tx queue in SR-IOV mode The 82576 expects the second rx queue in any pool to receive L2 switch loop back packets sent from the second tx queue in another pool. The 82576 VF driver does not enable the second rx queue so if the PF driver sends packets destined to a VF from its second tx queue then the VF driver will never see them. In SR-IOV mode limit the number of tx queues used by the PF driver to one. This patch fixes a bug reported in which the PF cannot communciate with the VF and should be considered for 2.6.34 stable. CC: stable@kernel.org Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-07-01 13:38:16 +00:00
if (adapter->vfs_allocated_count)
adapter->num_tx_queues = 1;
else
adapter->num_tx_queues = adapter->rss_queues;
igb: move initialization of number of queues into set_interrupt_capability This patch moves the initialization of the number of queues into set_interrupt_capability. This allows the number of queues to increase in the unlikely event that the system initially fails to allocate enough msi-x interrupts, does a suspend/resume, and then can allocate enough interrupts on resume. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:15:45 +00:00
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* start with one vector for every Rx queue */
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
numvecs = adapter->num_rx_queues;
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* if Tx handler is separate add 1 for every Tx queue */
igb: when number of CPUs > 4 combine tx/rx queues to allow more queues This patch makes it so that nics such as 82576 and newer can support more hardware queues when there are more than 4 cpus by combining a tx/rx queue pair onto one interrupt so that 8 queue pairs can be supported and thus allow for more queues. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:19 +00:00
if (!(adapter->flags & IGB_FLAG_QUEUE_PAIRS))
numvecs += adapter->num_tx_queues;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
/* store the number of vectors reserved for queues */
adapter->num_q_vectors = numvecs;
/* add 1 vector for link status interrupts */
numvecs++;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
for (i = 0; i < numvecs; i++)
adapter->msix_entries[i].entry = i;
igb: Use pci_enable_msix_range() instead of pci_enable_msix() As result of deprecation of MSI-X/MSI enablement functions pci_enable_msix() and pci_enable_msi_block() all drivers using these two interfaces need to be updated to use the new pci_enable_msi_range() and pci_enable_msix_range() interfaces. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Jesse Brandeburg <jesse.brandeburg@intel.com> Cc: e1000-devel@lists.sourceforge.net Cc: netdev@vger.kernel.org Cc: linux-pci@vger.kernel.org Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-18 10:11:43 +00:00
err = pci_enable_msix_range(adapter->pdev,
adapter->msix_entries,
numvecs,
numvecs);
if (err > 0)
igb: Move the calls to set the Tx and Rx queues into igb_open This change helps to address locking issues seen with netif_set_real_num_tx_queues and netif_set_real_num_rx_queues when used in the igb_set_interrupt_capability function. To resolve these locking issues I have moved the two function calls into __igb_open so that they can be called while the RTNL lock is held. An added advantage to this is that the number of queues is not updated until the last possible moment so if there are any issues in allocating MSI-X interrupts or resources for the rings we have time to change the values prior to updating the netdev. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:22 +00:00
return;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_reset_interrupt_capability(adapter);
/* If we can't do MSI-X, try MSI */
msi_only:
igb: Unset IGB_FLAG_HAS_MSIX-flag when falling back to msi-only Prior to cd14ef54d25 (igb: Change to use statically allocated array for MSIx entries), having msix_entries different from NULL was an indicator that MSIX is enabled. In igb_set_interrupt_capabiliy we may fall back to MSI-only. Prior to the above patch msix_entries was set to NULL by igb_reset_interrupt_capability. However, now we are checking the flag for IGB_FLAG_HAS_MSIX and so the stack gets completly confused: [ 42.659791] ------------[ cut here ]------------ [ 42.715032] WARNING: CPU: 7 PID: 0 at net/sched/sch_generic.c:264 dev_watchdog+0x15c/0x1fb() [ 42.848263] NETDEV WATCHDOG: eth0 (igb): transmit queue 0 timed out [ 42.923253] Modules linked in: [ 42.959875] CPU: 7 PID: 0 Comm: swapper/7 Not tainted 3.14.0-rc2-mptcp #437 [ 43.043184] Hardware name: HP ProLiant DL165 G7, BIOS O37 01/26/2011 [ 43.119215] 0000000000000108 ffff88023fdc3da8 ffffffff81487847 0000000000000108 [ 43.208165] ffff88023fdc3df8 ffff88023fdc3de8 ffffffff81034e7d ffff88023fdc3dd8 [ 43.297120] ffffffff813fff10 ffff880236018000 ffff880236b178c0 0000000000000008 [ 43.386071] Call Trace: [ 43.415303] <IRQ> [<ffffffff81487847>] dump_stack+0x49/0x62 [ 43.484174] [<ffffffff81034e7d>] warn_slowpath_common+0x77/0x91 [ 43.556049] [<ffffffff813fff10>] ? dev_watchdog+0x15c/0x1fb [ 43.623759] [<ffffffff81034f2b>] warn_slowpath_fmt+0x41/0x43 [ 43.692511] [<ffffffff813fff10>] dev_watchdog+0x15c/0x1fb [ 43.758141] [<ffffffff813ffdb4>] ? __netdev_watchdog_up+0x64/0x64 [ 43.832091] [<ffffffff8103cd04>] call_timer_fn+0x17/0x6f [ 43.896682] [<ffffffff8103cebe>] run_timer_softirq+0x162/0x1a2 [ 43.967511] [<ffffffff81038520>] __do_softirq+0xcd/0x1cc [ 44.032104] [<ffffffff81038689>] irq_exit+0x3a/0x48 [ 44.091492] [<ffffffff81026d43>] smp_apic_timer_interrupt+0x43/0x50 [ 44.167525] [<ffffffff8148c24a>] apic_timer_interrupt+0x6a/0x70 [ 44.239392] <EOI> [<ffffffff8100992c>] ? default_idle+0x6/0x8 [ 44.310343] [<ffffffff81009b31>] arch_cpu_idle+0x13/0x18 [ 44.374934] [<ffffffff81066126>] cpu_startup_entry+0xa7/0x101 [ 44.444724] [<ffffffff81025660>] start_secondary+0x1b2/0x1b7 [ 44.513472] ---[ end trace a5a075fd4e7f854f ]--- [ 44.568753] igb 0000:04:00.0 eth0: Reset adapter [ 46.206945] random: nonblocking pool is initialized [ 46.465670] irq 44: nobody cared (try booting with the "irqpoll" option) [ 46.545862] CPU: 7 PID: 0 Comm: swapper/7 Tainted: G W 3.14.0-rc2-mptcp #437 [ 46.640610] Hardware name: HP ProLiant DL165 G7, BIOS O37 01/26/2011 [ 46.716641] ffff8802363f8c84 ffff88023fdc3e38 ffffffff81487847 00000000a03cdb6d [ 46.805598] ffff8802363f8c00 ffff88023fdc3e68 ffffffff81068489 0000007f81825400 [ 46.894539] ffff8802363f8c00 0000000000000000 0000000000000000 ffff88023fdc3ea8 [ 46.983484] Call Trace: [ 47.012714] <IRQ> [<ffffffff81487847>] dump_stack+0x49/0x62 [ 47.081585] [<ffffffff81068489>] __report_bad_irq+0x35/0xc1 [ 47.149295] [<ffffffff81068683>] note_interrupt+0x16e/0x1ea [ 47.217006] [<ffffffff8106679e>] handle_irq_event_percpu+0x116/0x12e [ 47.294075] [<ffffffff810667e9>] handle_irq_event+0x33/0x4f [ 47.361787] [<ffffffff81068c95>] handle_fasteoi_irq+0x83/0xd1 [ 47.431577] [<ffffffff81003d5b>] handle_irq+0x1f/0x28 [ 47.493047] [<ffffffff81003567>] do_IRQ+0x4e/0xd4 [ 47.550358] [<ffffffff8148b06a>] common_interrupt+0x6a/0x6a [ 47.618066] <EOI> [<ffffffff8100992c>] ? default_idle+0x6/0x8 [ 47.689016] [<ffffffff81009b31>] arch_cpu_idle+0x13/0x18 [ 47.753605] [<ffffffff81066126>] cpu_startup_entry+0xa7/0x101 [ 47.823397] [<ffffffff81025660>] start_secondary+0x1b2/0x1b7 [ 47.892146] handlers: [ 47.919301] [<ffffffff812fbd7d>] igb_intr So, this patch unsets the flag to indicate that we are not using MSIX. This patch does exactly this: Unsetting the flag when falling back to MSI. Fixes: cd14ef54d25b (igb: Change to use statically allocated array for MSIx entries) Cc: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Christoph Paasch <christoph.paasch@uclouvain.be> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-21 11:02:09 +00:00
adapter->flags &= ~IGB_FLAG_HAS_MSIX;
igb: remove sysfs entry that was used to set the number of vfs This patch removes the sysfs entry num_vfs which was added to support enabling pci virtual functions for 82576. To prevent VFs from loading automatically a module parameter "max_vfs" was added so that the number of VFs per PF can be limited. This is especially useful when 4 or more 82576 ports are on the system because otherwise to load all VFs would result in 8 interface per physical port. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-07 14:37:52 +00:00
#ifdef CONFIG_PCI_IOV
/* disable SR-IOV for non MSI-X configurations */
if (adapter->vf_data) {
struct e1000_hw *hw = &adapter->hw;
/* disable iov and allow time for transactions to clear */
pci_disable_sriov(adapter->pdev);
msleep(500);
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
kfree(adapter->vf_mac_list);
adapter->vf_mac_list = NULL;
igb: remove sysfs entry that was used to set the number of vfs This patch removes the sysfs entry num_vfs which was added to support enabling pci virtual functions for 82576. To prevent VFs from loading automatically a module parameter "max_vfs" was added so that the number of VFs per PF can be limited. This is especially useful when 4 or more 82576 ports are on the system because otherwise to load all VFs would result in 8 interface per physical port. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-07 14:37:52 +00:00
kfree(adapter->vf_data);
adapter->vf_data = NULL;
wr32(E1000_IOVCTL, E1000_IOVCTL_REUSE_VFQ);
intel drivers: repair missing flush operations after review of all intel drivers, found several instances where drivers had the incorrect pattern of: memory mapped write(); delay(); which should always be: memory mapped write(); write flush(); /* aka memory mapped read */ delay(); explanation: The reason for including the flush is that writes can be held (posted) in PCI/PCIe bridges, but the read always has to complete synchronously and therefore has to flush all pending writes to a device. If a write is held and followed by a delay, the delay means nothing because the write may not have reached hardware (maybe even not until the next read) Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-07-20 00:56:21 +00:00
wrfl();
igb: remove sysfs entry that was used to set the number of vfs This patch removes the sysfs entry num_vfs which was added to support enabling pci virtual functions for 82576. To prevent VFs from loading automatically a module parameter "max_vfs" was added so that the number of VFs per PF can be limited. This is especially useful when 4 or more 82576 ports are on the system because otherwise to load all VFs would result in 8 interface per physical port. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-07 14:37:52 +00:00
msleep(100);
dev_info(&adapter->pdev->dev, "IOV Disabled\n");
}
#endif
igb: add support for seperate tx-usecs setting in ethtool This patch adds support for a seperate tx-usecs interrupt moderation setting in ethtool which is supported when tx and rx interrupt vectors are sperated. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:45:42 +00:00
adapter->vfs_allocated_count = 0;
igb: when number of CPUs > 4 combine tx/rx queues to allow more queues This patch makes it so that nics such as 82576 and newer can support more hardware queues when there are more than 4 cpus by combining a tx/rx queue pair onto one interrupt so that 8 queue pairs can be supported and thus allow for more queues. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:19 +00:00
adapter->rss_queues = 1;
igb: add support for seperate tx-usecs setting in ethtool This patch adds support for a seperate tx-usecs interrupt moderation setting in ethtool which is supported when tx and rx interrupt vectors are sperated. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:45:42 +00:00
adapter->flags |= IGB_FLAG_QUEUE_PAIRS;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
adapter->num_rx_queues = 1;
igb: Introduce multiple TX queues with infrastructure This code adds multiple Tx queue infrastructure much like we previously did in ixgbe. The MSI-X vector mapping is the bulk of the change. IAM can now be safely enabled and we've verified that it does work correctly. We can also eliminate the tx ring lock. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:06:51 +00:00
adapter->num_tx_queues = 1;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
adapter->num_q_vectors = 1;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (!pci_enable_msi(adapter->pdev))
igb: Add support for quad port WOL and feature flags Change igb from using a series of boolean operators to using a single flags value that contains a number of different bit flags for all the different features of the adapter. This patch also adds WOL support for quad port adapters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:46 +00:00
adapter->flags |= IGB_FLAG_HAS_MSI;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
static void igb_add_ring(struct igb_ring *ring,
struct igb_ring_container *head)
{
head->ring = ring;
head->count++;
}
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_alloc_q_vector - Allocate memory for a single interrupt vector
* @adapter: board private structure to initialize
* @v_count: q_vectors allocated on adapter, used for ring interleaving
* @v_idx: index of vector in adapter struct
* @txr_count: total number of Tx rings to allocate
* @txr_idx: index of first Tx ring to allocate
* @rxr_count: total number of Rx rings to allocate
* @rxr_idx: index of first Rx ring to allocate
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* We allocate one q_vector. If allocation fails we return -ENOMEM.
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
**/
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
static int igb_alloc_q_vector(struct igb_adapter *adapter,
int v_count, int v_idx,
int txr_count, int txr_idx,
int rxr_count, int rxr_idx)
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
{
struct igb_q_vector *q_vector;
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
struct igb_ring *ring;
int ring_count, size;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
/* igb only supports 1 Tx and/or 1 Rx queue per vector */
if (txr_count > 1 || rxr_count > 1)
return -ENOMEM;
ring_count = txr_count + rxr_count;
size = sizeof(struct igb_q_vector) +
(sizeof(struct igb_ring) * ring_count);
/* allocate q_vector and rings */
igb: Fix queue allocation method to accommodate changing during runtime When changing number of queues using ethtool's set_channels during runtime, a queue allocation could fail, which can leave the device in a down state. In order to preserve the usability of the device in this scenario, this patch changes the driver to allocate the number of queues only if they have not been allocated already. The first allocation is then done for the max number of queues, which is the default queues for this driver. With this change, queue quantity changes are not subject to queue allocation failures. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:29 +00:00
q_vector = adapter->q_vector[v_idx];
igb: Fix oops caused by missing queue pairing When initializing igb driver (e.g. 82576, I350), IGB_FLAG_QUEUE_PAIRS is set if adapter->rss_queues exceeds half of max_rss_queues in igb_init_queue_configuration(). On the other hand, IGB_FLAG_QUEUE_PAIRS is not set even if the number of queues exceeds half of max_combined in igb_set_channels() when changing the number of queues by "ethtool -L". In this case, if numvecs is larger than MAX_MSIX_ENTRIES (10), the size of adapter->msix_entries[], an overflow can occur in igb_set_interrupt_capability(), which in turn leads to an oops. Fix this problem as follows: - When changing the number of queues by "ethtool -L", set IGB_FLAG_QUEUE_PAIRS in the same way as initializing igb driver. - When increasing the size of q_vector, reallocate it appropriately. (With IGB_FLAG_QUEUE_PAIRS set, the size of q_vector gets larger.) Another possible way to fix this problem is to cap the queues at its initial number, which is the number of the initial online cpus. But this is not the optimal way because we cannot increase queues when another cpu becomes online. Note that before commit cd14ef54d25b ("igb: Change to use statically allocated array for MSIx entries"), this problem did not cause oops but just made the number of queues become 1 because of entering msi_only mode in igb_set_interrupt_capability(). Fixes: 907b7835799f ("igb: Add ethtool support to configure number of channels") CC: stable <stable@vger.kernel.org> Signed-off-by: Shota Suzuki <suzuki_shota_t3@lab.ntt.co.jp> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-07-01 00:25:52 +00:00
if (!q_vector) {
igb: Fix queue allocation method to accommodate changing during runtime When changing number of queues using ethtool's set_channels during runtime, a queue allocation could fail, which can leave the device in a down state. In order to preserve the usability of the device in this scenario, this patch changes the driver to allocate the number of queues only if they have not been allocated already. The first allocation is then done for the max number of queues, which is the default queues for this driver. With this change, queue quantity changes are not subject to queue allocation failures. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:29 +00:00
q_vector = kzalloc(size, GFP_KERNEL);
igb: Fix oops caused by missing queue pairing When initializing igb driver (e.g. 82576, I350), IGB_FLAG_QUEUE_PAIRS is set if adapter->rss_queues exceeds half of max_rss_queues in igb_init_queue_configuration(). On the other hand, IGB_FLAG_QUEUE_PAIRS is not set even if the number of queues exceeds half of max_combined in igb_set_channels() when changing the number of queues by "ethtool -L". In this case, if numvecs is larger than MAX_MSIX_ENTRIES (10), the size of adapter->msix_entries[], an overflow can occur in igb_set_interrupt_capability(), which in turn leads to an oops. Fix this problem as follows: - When changing the number of queues by "ethtool -L", set IGB_FLAG_QUEUE_PAIRS in the same way as initializing igb driver. - When increasing the size of q_vector, reallocate it appropriately. (With IGB_FLAG_QUEUE_PAIRS set, the size of q_vector gets larger.) Another possible way to fix this problem is to cap the queues at its initial number, which is the number of the initial online cpus. But this is not the optimal way because we cannot increase queues when another cpu becomes online. Note that before commit cd14ef54d25b ("igb: Change to use statically allocated array for MSIx entries"), this problem did not cause oops but just made the number of queues become 1 because of entering msi_only mode in igb_set_interrupt_capability(). Fixes: 907b7835799f ("igb: Add ethtool support to configure number of channels") CC: stable <stable@vger.kernel.org> Signed-off-by: Shota Suzuki <suzuki_shota_t3@lab.ntt.co.jp> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-07-01 00:25:52 +00:00
} else if (size > ksize(q_vector)) {
kfree_rcu(q_vector, rcu);
q_vector = kzalloc(size, GFP_KERNEL);
} else {
igb: Fix oops on changing number of rings When changing the number of rings by ethtool -L, q_vectors are reused, which causes oops because of uninitialized pointers. - When an rx is reused as a tx, q_vector->rx.ring is not set to NULL, which misleads igb_poll() to determine that it has an rx ring although it actually points to the tx ring. - When a tx is reused as an rx, q_vector->rx.ring->skb (q_vector->ring[0].skb) has a value that was used as tx_stats before. Fix these problems by zeroing it out on reuseing it. Fixes: 02ef6e1d0b00 ("igb: Fix queue allocation method to accommodate changing during runtime") Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-04-13 09:15:10 +00:00
memset(q_vector, 0, size);
igb: Fix oops caused by missing queue pairing When initializing igb driver (e.g. 82576, I350), IGB_FLAG_QUEUE_PAIRS is set if adapter->rss_queues exceeds half of max_rss_queues in igb_init_queue_configuration(). On the other hand, IGB_FLAG_QUEUE_PAIRS is not set even if the number of queues exceeds half of max_combined in igb_set_channels() when changing the number of queues by "ethtool -L". In this case, if numvecs is larger than MAX_MSIX_ENTRIES (10), the size of adapter->msix_entries[], an overflow can occur in igb_set_interrupt_capability(), which in turn leads to an oops. Fix this problem as follows: - When changing the number of queues by "ethtool -L", set IGB_FLAG_QUEUE_PAIRS in the same way as initializing igb driver. - When increasing the size of q_vector, reallocate it appropriately. (With IGB_FLAG_QUEUE_PAIRS set, the size of q_vector gets larger.) Another possible way to fix this problem is to cap the queues at its initial number, which is the number of the initial online cpus. But this is not the optimal way because we cannot increase queues when another cpu becomes online. Note that before commit cd14ef54d25b ("igb: Change to use statically allocated array for MSIx entries"), this problem did not cause oops but just made the number of queues become 1 because of entering msi_only mode in igb_set_interrupt_capability(). Fixes: 907b7835799f ("igb: Add ethtool support to configure number of channels") CC: stable <stable@vger.kernel.org> Signed-off-by: Shota Suzuki <suzuki_shota_t3@lab.ntt.co.jp> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-07-01 00:25:52 +00:00
}
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
if (!q_vector)
return -ENOMEM;
/* initialize NAPI */
netif_napi_add(adapter->netdev, &q_vector->napi,
igb_poll, 64);
/* tie q_vector and adapter together */
adapter->q_vector[v_idx] = q_vector;
q_vector->adapter = adapter;
/* initialize work limits */
q_vector->tx.work_limit = adapter->tx_work_limit;
/* initialize ITR configuration */
igb: improve handling of disconnected adapters Clean up array_rd32 so that it uses igb_rd32 the same as rd32, per the suggestion of Alexander Duyck, and use io_addr in more places, so that we don't have the need to call E1000_REMOVED (which simply looks for a null hw_addr) nearly as much. Signed-off-by: Jarod Wilson <jarod@redhat.com> Acked-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-10-19 15:52:04 +00:00
q_vector->itr_register = adapter->io_addr + E1000_EITR(0);
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
q_vector->itr_val = IGB_START_ITR;
/* initialize pointer to rings */
ring = q_vector->ring;
igb: Use rx/tx_itr_setting when setting up initial value of itr It turns out that the InterruptThrottleRate module parameter was only having the effect of locking the ITR at the starting ITR value. This was because the values stored in rx_itr_setting and tx_itr_setting were being ignored when configuring the initial itr_val of the q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-12 02:31:01 +00:00
/* intialize ITR */
if (rxr_count) {
/* rx or rx/tx vector */
if (!adapter->rx_itr_setting || adapter->rx_itr_setting > 3)
q_vector->itr_val = adapter->rx_itr_setting;
} else {
/* tx only vector */
if (!adapter->tx_itr_setting || adapter->tx_itr_setting > 3)
q_vector->itr_val = adapter->tx_itr_setting;
}
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
if (txr_count) {
/* assign generic ring traits */
ring->dev = &adapter->pdev->dev;
ring->netdev = adapter->netdev;
/* configure backlink on ring */
ring->q_vector = q_vector;
/* update q_vector Tx values */
igb_add_ring(ring, &q_vector->tx);
/* For 82575, context index must be unique per ring. */
if (adapter->hw.mac.type == e1000_82575)
set_bit(IGB_RING_FLAG_TX_CTX_IDX, &ring->flags);
/* apply Tx specific ring traits */
ring->count = adapter->tx_ring_count;
ring->queue_index = txr_idx;
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
ring->cbs_enable = false;
ring->idleslope = 0;
ring->sendslope = 0;
ring->hicredit = 0;
ring->locredit = 0;
net: Explicitly initialize u64_stats_sync structures for lockdep In order to enable lockdep on seqcount/seqlock structures, we must explicitly initialize any locks. The u64_stats_sync structure, uses a seqcount, and thus we need to introduce a u64_stats_init() function and use it to initialize the structure. This unfortunately adds a lot of fairly trivial initialization code to a number of drivers. But the benefit of ensuring correctness makes this worth while. Because these changes are required for lockdep to be enabled, and the changes are quite trivial, I've not yet split this patch out into 30-some separate patches, as I figured it would be better to get the various maintainers thoughts on how to best merge this change along with the seqcount lockdep enablement. Feedback would be appreciated! Signed-off-by: John Stultz <john.stultz@linaro.org> Acked-by: Julian Anastasov <ja@ssi.bg> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru> Cc: "David S. Miller" <davem@davemloft.net> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org> Cc: James Morris <jmorris@namei.org> Cc: Jesse Gross <jesse@nicira.com> Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Mirko Lindner <mlindner@marvell.com> Cc: Patrick McHardy <kaber@trash.net> Cc: Roger Luethi <rl@hellgate.ch> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Simon Horman <horms@verge.net.au> Cc: Stephen Hemminger <stephen@networkplumber.org> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Cc: Wensong Zhang <wensong@linux-vs.org> Cc: netdev@vger.kernel.org Link: http://lkml.kernel.org/r/1381186321-4906-2-git-send-email-john.stultz@linaro.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2013-10-07 22:51:58 +00:00
u64_stats_init(&ring->tx_syncp);
u64_stats_init(&ring->tx_syncp2);
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
/* assign ring to adapter */
adapter->tx_ring[txr_idx] = ring;
/* push pointer to next ring */
ring++;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
}
igb: Use node specific allocations for the q_vectors and rings This change is meant to update the ring and vector allocations so that they are per node instead of allocating everything on the node that ifconfig/modprobe is called on. By doing this we can cut down significantly on cross node traffic. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:20 +00:00
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
if (rxr_count) {
/* assign generic ring traits */
ring->dev = &adapter->pdev->dev;
ring->netdev = adapter->netdev;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
/* configure backlink on ring */
ring->q_vector = q_vector;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
/* update q_vector Rx values */
igb_add_ring(ring, &q_vector->rx);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
/* set flag indicating ring supports SCTP checksum offload */
if (adapter->hw.mac.type >= e1000_82576)
set_bit(IGB_RING_FLAG_RX_SCTP_CSUM, &ring->flags);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Cleanups to change comment style on license headers This patch fixes WARNING:NETWORKING_BLOCK_COMMENT_STYLE from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:46:06 +00:00
/* On i350, i354, i210, and i211, loopback VLAN packets
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
* have the tag byte-swapped.
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
*/
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
if (adapter->hw.mac.type >= e1000_i350)
set_bit(IGB_RING_FLAG_RX_LB_VLAN_BSWAP, &ring->flags);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
/* apply Rx specific ring traits */
ring->count = adapter->rx_ring_count;
ring->queue_index = rxr_idx;
net: Explicitly initialize u64_stats_sync structures for lockdep In order to enable lockdep on seqcount/seqlock structures, we must explicitly initialize any locks. The u64_stats_sync structure, uses a seqcount, and thus we need to introduce a u64_stats_init() function and use it to initialize the structure. This unfortunately adds a lot of fairly trivial initialization code to a number of drivers. But the benefit of ensuring correctness makes this worth while. Because these changes are required for lockdep to be enabled, and the changes are quite trivial, I've not yet split this patch out into 30-some separate patches, as I figured it would be better to get the various maintainers thoughts on how to best merge this change along with the seqcount lockdep enablement. Feedback would be appreciated! Signed-off-by: John Stultz <john.stultz@linaro.org> Acked-by: Julian Anastasov <ja@ssi.bg> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru> Cc: "David S. Miller" <davem@davemloft.net> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org> Cc: James Morris <jmorris@namei.org> Cc: Jesse Gross <jesse@nicira.com> Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Mirko Lindner <mlindner@marvell.com> Cc: Patrick McHardy <kaber@trash.net> Cc: Roger Luethi <rl@hellgate.ch> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Simon Horman <horms@verge.net.au> Cc: Stephen Hemminger <stephen@networkplumber.org> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Cc: Wensong Zhang <wensong@linux-vs.org> Cc: netdev@vger.kernel.org Link: http://lkml.kernel.org/r/1381186321-4906-2-git-send-email-john.stultz@linaro.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2013-10-07 22:51:58 +00:00
u64_stats_init(&ring->rx_syncp);
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
/* assign ring to adapter */
adapter->rx_ring[rxr_idx] = ring;
}
return 0;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
}
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_alloc_q_vectors - Allocate memory for interrupt vectors
* @adapter: board private structure to initialize
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* We allocate one q_vector per queue interrupt. If allocation fails we
* return -ENOMEM.
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
**/
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
static int igb_alloc_q_vectors(struct igb_adapter *adapter)
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
{
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
int q_vectors = adapter->num_q_vectors;
int rxr_remaining = adapter->num_rx_queues;
int txr_remaining = adapter->num_tx_queues;
int rxr_idx = 0, txr_idx = 0, v_idx = 0;
int err;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
if (q_vectors >= (rxr_remaining + txr_remaining)) {
for (; rxr_remaining; v_idx++) {
err = igb_alloc_q_vector(adapter, q_vectors, v_idx,
0, 0, 1, rxr_idx);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
if (err)
goto err_out;
/* update counts and index */
rxr_remaining--;
rxr_idx++;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
}
}
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
for (; v_idx < q_vectors; v_idx++) {
int rqpv = DIV_ROUND_UP(rxr_remaining, q_vectors - v_idx);
int tqpv = DIV_ROUND_UP(txr_remaining, q_vectors - v_idx);
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
err = igb_alloc_q_vector(adapter, q_vectors, v_idx,
tqpv, txr_idx, rqpv, rxr_idx);
if (err)
goto err_out;
/* update counts and index */
rxr_remaining -= rqpv;
txr_remaining -= tqpv;
rxr_idx++;
txr_idx++;
}
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
return 0;
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
err_out:
adapter->num_tx_queues = 0;
adapter->num_rx_queues = 0;
adapter->num_q_vectors = 0;
while (v_idx--)
igb_free_q_vector(adapter, v_idx);
return -ENOMEM;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_init_interrupt_scheme - initialize interrupts, allocate queues/vectors
* @adapter: board private structure to initialize
* @msix: boolean value of MSIX capability
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* This function initializes the interrupts and allocates all of the queues.
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
**/
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
static int igb_init_interrupt_scheme(struct igb_adapter *adapter, bool msix)
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
{
struct pci_dev *pdev = adapter->pdev;
int err;
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
igb_set_interrupt_capability(adapter, msix);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
err = igb_alloc_q_vectors(adapter);
if (err) {
dev_err(&pdev->dev, "Unable to allocate memory for vectors\n");
goto err_alloc_q_vectors;
}
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
igb_cache_ring_register(adapter);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
return 0;
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
err_alloc_q_vectors:
igb_reset_interrupt_capability(adapter);
return err;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_request_irq - initialize interrupts
* @adapter: board private structure to initialize
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Attempts to configure interrupts using the best available
* capabilities of the hardware and kernel.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static int igb_request_irq(struct igb_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
struct pci_dev *pdev = adapter->pdev;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
int err = 0;
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
if (adapter->flags & IGB_FLAG_HAS_MSIX) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
err = igb_request_msix(adapter);
igb: add NAPI Rx queue support Update the NAPI implementation to use the new napi_struct infrstructure. This removes the need of multiple net_device structs to implement a multiqueue NAPI. Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:39 +00:00
if (!err)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
goto request_done;
/* fall back to MSI */
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
igb_free_all_tx_resources(adapter);
igb_free_all_rx_resources(adapter);
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb_clear_interrupt_scheme(adapter);
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
err = igb_init_interrupt_scheme(adapter, false);
if (err)
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
goto request_done;
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb_setup_all_tx_resources(adapter);
igb_setup_all_rx_resources(adapter);
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
igb_configure(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: add NAPI Rx queue support Update the NAPI implementation to use the new napi_struct infrstructure. This removes the need of multiple net_device structs to implement a multiqueue NAPI. Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:39 +00:00
igb: Make certain one vector is always assigned in igb_request_irq This change makes certain that one interrupt is always initialized in igb_request_irq. In addition we drop the use of adapter->pdev and instead just call pdev since we made a local copy of the pointer earlier in the function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:45 +00:00
igb_assign_vector(adapter->q_vector[0], 0);
igb: Add support for quad port WOL and feature flags Change igb from using a series of boolean operators to using a single flags value that contains a number of different bit flags for all the different features of the adapter. This patch also adds WOL support for quad port adapters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:46 +00:00
if (adapter->flags & IGB_FLAG_HAS_MSI) {
igb: Make certain one vector is always assigned in igb_request_irq This change makes certain that one interrupt is always initialized in igb_request_irq. In addition we drop the use of adapter->pdev and instead just call pdev since we made a local copy of the pointer earlier in the function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:45 +00:00
err = request_irq(pdev->irq, igb_intr_msi, 0,
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
netdev->name, adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (!err)
goto request_done;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* fall back to legacy interrupts */
igb_reset_interrupt_capability(adapter);
igb: Add support for quad port WOL and feature flags Change igb from using a series of boolean operators to using a single flags value that contains a number of different bit flags for all the different features of the adapter. This patch also adds WOL support for quad port adapters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:46 +00:00
adapter->flags &= ~IGB_FLAG_HAS_MSI;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Make certain one vector is always assigned in igb_request_irq This change makes certain that one interrupt is always initialized in igb_request_irq. In addition we drop the use of adapter->pdev and instead just call pdev since we made a local copy of the pointer earlier in the function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:45 +00:00
err = request_irq(pdev->irq, igb_intr, IRQF_SHARED,
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
netdev->name, adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: fix legacy mode irq issue I booted an igb kernel with the option pci=nomsi and instantly noticed that interrupts no longer worked on my igb device. I took a look at the interrupt initialization and quickly discovered a comment stating: "DO NOT USE EIAME or IAME in legacy mode" It seemed a bit odd that bits to enable IAM were being set in legacy interrupt mode, so I dropped out the following parts and interrupts began working fine again. [Updated code flow and a nitpick spelling error --Auke] Signed-off-by: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2008-02-15 22:05:25 +00:00
if (err)
igb: Make certain one vector is always assigned in igb_request_irq This change makes certain that one interrupt is always initialized in igb_request_irq. In addition we drop the use of adapter->pdev and instead just call pdev since we made a local copy of the pointer earlier in the function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:45 +00:00
dev_err(&pdev->dev, "Error %d getting interrupt\n",
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
err);
request_done:
return err;
}
static void igb_free_irq(struct igb_adapter *adapter)
{
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
if (adapter->flags & IGB_FLAG_HAS_MSIX) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
int vector = 0, i;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
free_irq(adapter->msix_entries[vector++].vector, adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: avoid unnecessarily creating a local copy of the q_vector This is mostly a drop of unnecessary pointer defines for q_vector when we don't have issues with line width and don't have multiple references to the pointer. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:34 +00:00
for (i = 0; i < adapter->num_q_vectors; i++)
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
free_irq(adapter->msix_entries[vector++].vector,
igb: avoid unnecessarily creating a local copy of the q_vector This is mostly a drop of unnecessary pointer defines for q_vector when we don't have issues with line width and don't have multiple references to the pointer. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:34 +00:00
adapter->q_vector[i]);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
} else {
free_irq(adapter->pdev->irq, adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_irq_disable - Mask off interrupt generation on the NIC
* @adapter: board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_irq_disable(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* we need to be careful when disabling interrupts. The VFs are also
igb: cleanup interrupt enablement in regards to msix_other This patch changes a few things so that instead of firing a link status interrupt directly the get_link_status bit is set and the watchdog is scheduled. In addition the mailbox bit is now only enabled if VFs are enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:59 +00:00
* mapped into these registers and so clearing the bits can cause
* issues on the VF drivers so we only need to clear what we set
*/
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
if (adapter->flags & IGB_FLAG_HAS_MSIX) {
igb: only disable/enable interrupt bits for igb physical function The igb_irq_disable/enable calls were causing virtual functions associated with the igb physical function to have their interrupts disabled. In order to prevent this from occuring we should only clear/set the bits related to the physical function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:49:15 +00:00
u32 regval = rd32(E1000_EIAM);
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: only disable/enable interrupt bits for igb physical function The igb_irq_disable/enable calls were causing virtual functions associated with the igb physical function to have their interrupts disabled. In order to prevent this from occuring we should only clear/set the bits related to the physical function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:49:15 +00:00
wr32(E1000_EIAM, regval & ~adapter->eims_enable_mask);
wr32(E1000_EIMC, adapter->eims_enable_mask);
regval = rd32(E1000_EIAC);
wr32(E1000_EIAC, regval & ~adapter->eims_enable_mask);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: add NAPI Rx queue support Update the NAPI implementation to use the new napi_struct infrstructure. This removes the need of multiple net_device structs to implement a multiqueue NAPI. Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:39 +00:00
wr32(E1000_IAM, 0);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
wr32(E1000_IMC, ~0);
wrfl();
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
if (adapter->flags & IGB_FLAG_HAS_MSIX) {
igb: Use irq_synchronize per vector when using MSI-X Synchronize all IRQs when using MSI-X. Similar to ixgbe. Issue was reported on e1000e, but the patch is also valid for igb. CC: Jean Delvare <jdelvare@suse.de> Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-02 14:40:52 +00:00
int i;
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: Use irq_synchronize per vector when using MSI-X Synchronize all IRQs when using MSI-X. Similar to ixgbe. Issue was reported on e1000e, but the patch is also valid for igb. CC: Jean Delvare <jdelvare@suse.de> Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-02 14:40:52 +00:00
for (i = 0; i < adapter->num_q_vectors; i++)
synchronize_irq(adapter->msix_entries[i].vector);
} else {
synchronize_irq(adapter->pdev->irq);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_irq_enable - Enable default interrupt generation settings
* @adapter: board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_irq_enable(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
if (adapter->flags & IGB_FLAG_HAS_MSIX) {
igb: Fix features that are currently 82580 only and should also be i350 This change allows support for per packet timesync and global device reset on the i350 adapter. These features were supported on both 82580 and i350 however it looks like several checks where not updated and as such the i350 support was not enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:55 +00:00
u32 ims = E1000_IMS_LSC | E1000_IMS_DOUTSYNC | E1000_IMS_DRSTA;
igb: only disable/enable interrupt bits for igb physical function The igb_irq_disable/enable calls were causing virtual functions associated with the igb physical function to have their interrupts disabled. In order to prevent this from occuring we should only clear/set the bits related to the physical function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:49:15 +00:00
u32 regval = rd32(E1000_EIAC);
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: only disable/enable interrupt bits for igb physical function The igb_irq_disable/enable calls were causing virtual functions associated with the igb physical function to have their interrupts disabled. In order to prevent this from occuring we should only clear/set the bits related to the physical function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:49:15 +00:00
wr32(E1000_EIAC, regval | adapter->eims_enable_mask);
regval = rd32(E1000_EIAM);
wr32(E1000_EIAM, regval | adapter->eims_enable_mask);
igb: add NAPI Rx queue support Update the NAPI implementation to use the new napi_struct infrstructure. This removes the need of multiple net_device structs to implement a multiqueue NAPI. Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:39 +00:00
wr32(E1000_EIMS, adapter->eims_enable_mask);
igb: cleanup interrupt enablement in regards to msix_other This patch changes a few things so that instead of firing a link status interrupt directly the get_link_status bit is set and the watchdog is scheduled. In addition the mailbox bit is now only enabled if VFs are enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:59 +00:00
if (adapter->vfs_allocated_count) {
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
wr32(E1000_MBVFIMR, 0xFF);
igb: cleanup interrupt enablement in regards to msix_other This patch changes a few things so that instead of firing a link status interrupt directly the get_link_status bit is set and the watchdog is scheduled. In addition the mailbox bit is now only enabled if VFs are enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:59 +00:00
ims |= E1000_IMS_VMMB;
}
wr32(E1000_IMS, ims);
igb: add NAPI Rx queue support Update the NAPI implementation to use the new napi_struct infrstructure. This removes the need of multiple net_device structs to implement a multiqueue NAPI. Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:39 +00:00
} else {
igb: Add full support for 82580 devices This patch makes use of the 82580 PHY and MAC support added and adds a set of supported device IDs for said hardware. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-19 12:42:21 +00:00
wr32(E1000_IMS, IMS_ENABLE_MASK |
E1000_IMS_DRSTA);
wr32(E1000_IAM, IMS_ENABLE_MASK |
E1000_IMS_DRSTA);
igb: add NAPI Rx queue support Update the NAPI implementation to use the new napi_struct infrstructure. This removes the need of multiple net_device structs to implement a multiqueue NAPI. Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:39 +00:00
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
static void igb_update_mng_vlan(struct igb_adapter *adapter)
{
igb: rework handling of the vfta and vlvf registers in relation to mng_vlan This patch corrects some errors in how vlans are being handled when vfs start interacting with the management vlans. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:35 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: Merge VLVF configuration into igb_vfta_set This change makes it so that we can merge the configuration of the VLVF registers into the setting of the VFTA register. By doing this we simplify the logic and make use of similar functionality that we have already added for ixgbe making it easier to maintain both drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:04 +00:00
u16 pf_id = adapter->vfs_allocated_count;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
u16 vid = adapter->hw.mng_cookie.vlan_id;
u16 old_vid = adapter->mng_vlan_id;
igb: rework handling of the vfta and vlvf registers in relation to mng_vlan This patch corrects some errors in how vlans are being handled when vfs start interacting with the management vlans. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:35 +00:00
if (hw->mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN) {
/* add VID to filter table */
igb: Merge VLVF configuration into igb_vfta_set This change makes it so that we can merge the configuration of the VLVF registers into the setting of the VFTA register. By doing this we simplify the logic and make use of similar functionality that we have already added for ixgbe making it easier to maintain both drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:04 +00:00
igb_vfta_set(hw, vid, pf_id, true, true);
igb: rework handling of the vfta and vlvf registers in relation to mng_vlan This patch corrects some errors in how vlans are being handled when vfs start interacting with the management vlans. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:35 +00:00
adapter->mng_vlan_id = vid;
} else {
adapter->mng_vlan_id = IGB_MNG_VLAN_NONE;
}
if ((old_vid != (u16)IGB_MNG_VLAN_NONE) &&
(vid != old_vid) &&
igb: do vlan cleanup - unify vlan and nonvlan rx path - kill adapter->vlgrp and igb_vlan_rx_register - allow to turn on/off rx/tx vlan accel via ethtool (set_features) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-21 03:27:27 +00:00
!test_bit(old_vid, adapter->active_vlans)) {
igb: rework handling of the vfta and vlvf registers in relation to mng_vlan This patch corrects some errors in how vlans are being handled when vfs start interacting with the management vlans. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:35 +00:00
/* remove VID from filter table */
igb: Merge VLVF configuration into igb_vfta_set This change makes it so that we can merge the configuration of the VLVF registers into the setting of the VFTA register. By doing this we simplify the logic and make use of similar functionality that we have already added for ixgbe making it easier to maintain both drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:04 +00:00
igb_vfta_set(hw, vid, pf_id, false, true);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_release_hw_control - release control of the h/w to f/w
* @adapter: address of board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_release_hw_control resets CTRL_EXT:DRV_LOAD bit.
* For ASF and Pass Through versions of f/w this means that the
* driver is no longer loaded.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_release_hw_control(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
u32 ctrl_ext;
/* Let firmware take over control of h/w */
ctrl_ext = rd32(E1000_CTRL_EXT);
wr32(E1000_CTRL_EXT,
ctrl_ext & ~E1000_CTRL_EXT_DRV_LOAD);
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_get_hw_control - get control of the h/w from f/w
* @adapter: address of board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_get_hw_control sets CTRL_EXT:DRV_LOAD bit.
* For ASF and Pass Through versions of f/w this means that
* the driver is loaded.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_get_hw_control(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
u32 ctrl_ext;
/* Let firmware know the driver has taken over */
ctrl_ext = rd32(E1000_CTRL_EXT);
wr32(E1000_CTRL_EXT,
ctrl_ext | E1000_CTRL_EXT_DRV_LOAD);
}
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
static void enable_fqtss(struct igb_adapter *adapter, bool enable)
{
struct net_device *netdev = adapter->netdev;
struct e1000_hw *hw = &adapter->hw;
WARN_ON(hw->mac.type != e1000_i210);
if (enable)
adapter->flags |= IGB_FLAG_FQTSS;
else
adapter->flags &= ~IGB_FLAG_FQTSS;
if (netif_running(netdev))
schedule_work(&adapter->reset_task);
}
static bool is_fqtss_enabled(struct igb_adapter *adapter)
{
return (adapter->flags & IGB_FLAG_FQTSS) ? true : false;
}
static void set_tx_desc_fetch_prio(struct e1000_hw *hw, int queue,
enum tx_queue_prio prio)
{
u32 val;
WARN_ON(hw->mac.type != e1000_i210);
WARN_ON(queue < 0 || queue > 4);
val = rd32(E1000_I210_TXDCTL(queue));
if (prio == TX_QUEUE_PRIO_HIGH)
val |= E1000_TXDCTL_PRIORITY;
else
val &= ~E1000_TXDCTL_PRIORITY;
wr32(E1000_I210_TXDCTL(queue), val);
}
static void set_queue_mode(struct e1000_hw *hw, int queue, enum queue_mode mode)
{
u32 val;
WARN_ON(hw->mac.type != e1000_i210);
WARN_ON(queue < 0 || queue > 1);
val = rd32(E1000_I210_TQAVCC(queue));
if (mode == QUEUE_MODE_STREAM_RESERVATION)
val |= E1000_TQAVCC_QUEUEMODE;
else
val &= ~E1000_TQAVCC_QUEUEMODE;
wr32(E1000_I210_TQAVCC(queue), val);
}
igb: Only change Tx arbitration when CBS is on Currently the data transmission arbitration algorithm - DataTranARB field on TQAVCTRL reg - is always set to CBS when the Tx mode is changed from legacy to 'Qav' mode. Make that configuration a bit more granular in preparation for the upcoming Launchtime enabling patches, since CBS and Launchtime can be enabled separately. That is achieved by moving the DataTranARB setup to igb_config_tx_modes() instead. Similarly, when disabling CBS we must check if it has been disabled for all queues, and clear the DataTranARB accordingly. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:56 +00:00
static bool is_any_cbs_enabled(struct igb_adapter *adapter)
{
int i;
for (i = 0; i < adapter->num_tx_queues; i++) {
if (adapter->tx_ring[i]->cbs_enable)
return true;
}
return false;
}
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
static bool is_any_txtime_enabled(struct igb_adapter *adapter)
{
int i;
for (i = 0; i < adapter->num_tx_queues; i++) {
if (adapter->tx_ring[i]->launchtime_enable)
return true;
}
return false;
}
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
/**
igb: Refactor igb_configure_cbs() Make this function retrieve what it needs from the Tx ring being addressed since it already relies on what had been saved on it before. Also, since this function will be used by the upcoming Launchtime patches rename it to better reflect its intention. Note that Launchtime is not part of what 802.1Qav specifies, but the i210 datasheet refers to this set of functionality as "Qav Transmission Mode". Here we also perform a tiny refactor at is_any_cbs_enabled(), and add further documentation to igb_setup_tx_mode(). Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:55 +00:00
* igb_config_tx_modes - Configure "Qav Tx mode" features on igb
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
* @adapter: pointer to adapter struct
* @queue: queue number
*
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
* Configure CBS and Launchtime for a given hardware queue.
* Parameters are retrieved from the correct Tx ring, so
* igb_save_cbs_params() and igb_save_txtime_params() should be used
igb: Refactor igb_configure_cbs() Make this function retrieve what it needs from the Tx ring being addressed since it already relies on what had been saved on it before. Also, since this function will be used by the upcoming Launchtime patches rename it to better reflect its intention. Note that Launchtime is not part of what 802.1Qav specifies, but the i210 datasheet refers to this set of functionality as "Qav Transmission Mode". Here we also perform a tiny refactor at is_any_cbs_enabled(), and add further documentation to igb_setup_tx_mode(). Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:55 +00:00
* for setting those correctly prior to this function being called.
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
**/
igb: Refactor igb_configure_cbs() Make this function retrieve what it needs from the Tx ring being addressed since it already relies on what had been saved on it before. Also, since this function will be used by the upcoming Launchtime patches rename it to better reflect its intention. Note that Launchtime is not part of what 802.1Qav specifies, but the i210 datasheet refers to this set of functionality as "Qav Transmission Mode". Here we also perform a tiny refactor at is_any_cbs_enabled(), and add further documentation to igb_setup_tx_mode(). Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:55 +00:00
static void igb_config_tx_modes(struct igb_adapter *adapter, int queue)
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
{
igb: Refactor igb_configure_cbs() Make this function retrieve what it needs from the Tx ring being addressed since it already relies on what had been saved on it before. Also, since this function will be used by the upcoming Launchtime patches rename it to better reflect its intention. Note that Launchtime is not part of what 802.1Qav specifies, but the i210 datasheet refers to this set of functionality as "Qav Transmission Mode". Here we also perform a tiny refactor at is_any_cbs_enabled(), and add further documentation to igb_setup_tx_mode(). Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:55 +00:00
struct igb_ring *ring = adapter->tx_ring[queue];
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
struct net_device *netdev = adapter->netdev;
struct e1000_hw *hw = &adapter->hw;
igb: Only change Tx arbitration when CBS is on Currently the data transmission arbitration algorithm - DataTranARB field on TQAVCTRL reg - is always set to CBS when the Tx mode is changed from legacy to 'Qav' mode. Make that configuration a bit more granular in preparation for the upcoming Launchtime enabling patches, since CBS and Launchtime can be enabled separately. That is achieved by moving the DataTranARB setup to igb_config_tx_modes() instead. Similarly, when disabling CBS we must check if it has been disabled for all queues, and clear the DataTranARB accordingly. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:56 +00:00
u32 tqavcc, tqavctrl;
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
u16 value;
WARN_ON(hw->mac.type != e1000_i210);
WARN_ON(queue < 0 || queue > 1);
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
/* If any of the Qav features is enabled, configure queues as SR and
* with HIGH PRIO. If none is, then configure them with LOW PRIO and
* as SP.
*/
if (ring->cbs_enable || ring->launchtime_enable) {
set_tx_desc_fetch_prio(hw, queue, TX_QUEUE_PRIO_HIGH);
set_queue_mode(hw, queue, QUEUE_MODE_STREAM_RESERVATION);
} else {
set_tx_desc_fetch_prio(hw, queue, TX_QUEUE_PRIO_LOW);
set_queue_mode(hw, queue, QUEUE_MODE_STRICT_PRIORITY);
}
/* If CBS is enabled, set DataTranARB and config its parameters. */
igb: Refactor igb_configure_cbs() Make this function retrieve what it needs from the Tx ring being addressed since it already relies on what had been saved on it before. Also, since this function will be used by the upcoming Launchtime patches rename it to better reflect its intention. Note that Launchtime is not part of what 802.1Qav specifies, but the i210 datasheet refers to this set of functionality as "Qav Transmission Mode". Here we also perform a tiny refactor at is_any_cbs_enabled(), and add further documentation to igb_setup_tx_mode(). Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:55 +00:00
if (ring->cbs_enable || queue == 0) {
igb: Fix the transmission mode of queue 0 for Qav mode When Qav mode is enabled, queue 0 should be kept on Stream Reservation mode. From the i210 datasheet, section 8.12.19: "Note: Queue0 QueueMode must be set to 1b when TransmitMode is set to Qav." ("QueueMode 1b" represents the Stream Reservation mode) The solution is to give queue 0 the all the credits it might need, so it has priority over queue 1. A situation where this can happen is when cbs is "installed" only on queue 1, leaving queue 0 alone. For example: $ tc qdisc replace dev enp2s0 handle 100: parent root mqprio num_tc 3 \ map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@2 hw 0 $ tc qdisc replace dev enp2s0 parent 100:2 cbs locredit -1470 \ hicredit 30 sendslope -980000 idleslope 20000 offload 1 Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-03-31 00:06:52 +00:00
/* i210 does not allow the queue 0 to be in the Strict
* Priority mode while the Qav mode is enabled, so,
* instead of disabling strict priority mode, we give
* queue 0 the maximum of credits possible.
*
* See section 8.12.19 of the i210 datasheet, "Note:
* Queue0 QueueMode must be set to 1b when
* TransmitMode is set to Qav."
*/
igb: Refactor igb_configure_cbs() Make this function retrieve what it needs from the Tx ring being addressed since it already relies on what had been saved on it before. Also, since this function will be used by the upcoming Launchtime patches rename it to better reflect its intention. Note that Launchtime is not part of what 802.1Qav specifies, but the i210 datasheet refers to this set of functionality as "Qav Transmission Mode". Here we also perform a tiny refactor at is_any_cbs_enabled(), and add further documentation to igb_setup_tx_mode(). Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:55 +00:00
if (queue == 0 && !ring->cbs_enable) {
igb: Fix the transmission mode of queue 0 for Qav mode When Qav mode is enabled, queue 0 should be kept on Stream Reservation mode. From the i210 datasheet, section 8.12.19: "Note: Queue0 QueueMode must be set to 1b when TransmitMode is set to Qav." ("QueueMode 1b" represents the Stream Reservation mode) The solution is to give queue 0 the all the credits it might need, so it has priority over queue 1. A situation where this can happen is when cbs is "installed" only on queue 1, leaving queue 0 alone. For example: $ tc qdisc replace dev enp2s0 handle 100: parent root mqprio num_tc 3 \ map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@2 hw 0 $ tc qdisc replace dev enp2s0 parent 100:2 cbs locredit -1470 \ hicredit 30 sendslope -980000 idleslope 20000 offload 1 Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-03-31 00:06:52 +00:00
/* max "linkspeed" idleslope in kbps */
igb: Refactor igb_configure_cbs() Make this function retrieve what it needs from the Tx ring being addressed since it already relies on what had been saved on it before. Also, since this function will be used by the upcoming Launchtime patches rename it to better reflect its intention. Note that Launchtime is not part of what 802.1Qav specifies, but the i210 datasheet refers to this set of functionality as "Qav Transmission Mode". Here we also perform a tiny refactor at is_any_cbs_enabled(), and add further documentation to igb_setup_tx_mode(). Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:55 +00:00
ring->idleslope = 1000000;
ring->hicredit = ETH_FRAME_LEN;
igb: Fix the transmission mode of queue 0 for Qav mode When Qav mode is enabled, queue 0 should be kept on Stream Reservation mode. From the i210 datasheet, section 8.12.19: "Note: Queue0 QueueMode must be set to 1b when TransmitMode is set to Qav." ("QueueMode 1b" represents the Stream Reservation mode) The solution is to give queue 0 the all the credits it might need, so it has priority over queue 1. A situation where this can happen is when cbs is "installed" only on queue 1, leaving queue 0 alone. For example: $ tc qdisc replace dev enp2s0 handle 100: parent root mqprio num_tc 3 \ map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@2 hw 0 $ tc qdisc replace dev enp2s0 parent 100:2 cbs locredit -1470 \ hicredit 30 sendslope -980000 idleslope 20000 offload 1 Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-03-31 00:06:52 +00:00
}
igb: Only change Tx arbitration when CBS is on Currently the data transmission arbitration algorithm - DataTranARB field on TQAVCTRL reg - is always set to CBS when the Tx mode is changed from legacy to 'Qav' mode. Make that configuration a bit more granular in preparation for the upcoming Launchtime enabling patches, since CBS and Launchtime can be enabled separately. That is achieved by moving the DataTranARB setup to igb_config_tx_modes() instead. Similarly, when disabling CBS we must check if it has been disabled for all queues, and clear the DataTranARB accordingly. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:56 +00:00
/* Always set data transfer arbitration to credit-based
* shaper algorithm on TQAVCTRL if CBS is enabled for any of
* the queues.
*/
tqavctrl = rd32(E1000_I210_TQAVCTRL);
tqavctrl |= E1000_TQAVCTRL_DATATRANARB;
wr32(E1000_I210_TQAVCTRL, tqavctrl);
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
/* According to i210 datasheet section 7.2.7.7, we should set
* the 'idleSlope' field from TQAVCC register following the
* equation:
*
* For 100 Mbps link speed:
*
* value = BW * 0x7735 * 0.2 (E1)
*
* For 1000Mbps link speed:
*
* value = BW * 0x7735 * 2 (E2)
*
* E1 and E2 can be merged into one equation as shown below.
* Note that 'link-speed' is in Mbps.
*
* value = BW * 0x7735 * 2 * link-speed
* -------------- (E3)
* 1000
*
* 'BW' is the percentage bandwidth out of full link speed
* which can be found with the following equation. Note that
* idleSlope here is the parameter from this function which
* is in kbps.
*
* BW = idleSlope
* ----------------- (E4)
* link-speed * 1000
*
* That said, we can come up with a generic equation to
* calculate the value we should set it TQAVCC register by
* replacing 'BW' in E3 by E4. The resulting equation is:
*
* value = idleSlope * 0x7735 * 2 * link-speed
* ----------------- -------------- (E5)
* link-speed * 1000 1000
*
* 'link-speed' is present in both sides of the fraction so
* it is canceled out. The final equation is the following:
*
* value = idleSlope * 61034
* ----------------- (E6)
* 1000000
igb: Clarify idleslope config constraints By design, the idleslope increments are restricted to 16.384kbps steps. Add a comment to igb_main.c making that explicit and add one example that illustrates the impact of that. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-11-10 22:21:50 +00:00
*
* NOTE: For i210, given the above, we can see that idleslope
* is represented in 16.38431 kbps units by the value at
* the TQAVCC register (1Gbps / 61034), which reduces
* the granularity for idleslope increments.
* For instance, if you want to configure a 2576kbps
* idleslope, the value to be written on the register
* would have to be 157.23. If rounded down, you end
* up with less bandwidth available than originally
* required (~2572 kbps). If rounded up, you end up
* with a higher bandwidth (~2589 kbps). Below the
* approach we take is to always round up the
* calculated value, so the resulting bandwidth might
* be slightly higher for some configurations.
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
*/
igb: Refactor igb_configure_cbs() Make this function retrieve what it needs from the Tx ring being addressed since it already relies on what had been saved on it before. Also, since this function will be used by the upcoming Launchtime patches rename it to better reflect its intention. Note that Launchtime is not part of what 802.1Qav specifies, but the i210 datasheet refers to this set of functionality as "Qav Transmission Mode". Here we also perform a tiny refactor at is_any_cbs_enabled(), and add further documentation to igb_setup_tx_mode(). Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:55 +00:00
value = DIV_ROUND_UP_ULL(ring->idleslope * 61034ULL, 1000000);
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
tqavcc = rd32(E1000_I210_TQAVCC(queue));
tqavcc &= ~E1000_TQAVCC_IDLESLOPE_MASK;
tqavcc |= value;
wr32(E1000_I210_TQAVCC(queue), tqavcc);
igb: Refactor igb_configure_cbs() Make this function retrieve what it needs from the Tx ring being addressed since it already relies on what had been saved on it before. Also, since this function will be used by the upcoming Launchtime patches rename it to better reflect its intention. Note that Launchtime is not part of what 802.1Qav specifies, but the i210 datasheet refers to this set of functionality as "Qav Transmission Mode". Here we also perform a tiny refactor at is_any_cbs_enabled(), and add further documentation to igb_setup_tx_mode(). Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:55 +00:00
wr32(E1000_I210_TQAVHC(queue),
0x80000000 + ring->hicredit * 0x7735);
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
} else {
/* Set idleSlope to zero. */
tqavcc = rd32(E1000_I210_TQAVCC(queue));
tqavcc &= ~E1000_TQAVCC_IDLESLOPE_MASK;
wr32(E1000_I210_TQAVCC(queue), tqavcc);
/* Set hiCredit to zero. */
wr32(E1000_I210_TQAVHC(queue), 0);
igb: Only change Tx arbitration when CBS is on Currently the data transmission arbitration algorithm - DataTranARB field on TQAVCTRL reg - is always set to CBS when the Tx mode is changed from legacy to 'Qav' mode. Make that configuration a bit more granular in preparation for the upcoming Launchtime enabling patches, since CBS and Launchtime can be enabled separately. That is achieved by moving the DataTranARB setup to igb_config_tx_modes() instead. Similarly, when disabling CBS we must check if it has been disabled for all queues, and clear the DataTranARB accordingly. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:56 +00:00
/* If CBS is not enabled for any queues anymore, then return to
* the default state of Data Transmission Arbitration on
* TQAVCTRL.
*/
if (!is_any_cbs_enabled(adapter)) {
tqavctrl = rd32(E1000_I210_TQAVCTRL);
tqavctrl &= ~E1000_TQAVCTRL_DATATRANARB;
wr32(E1000_I210_TQAVCTRL, tqavctrl);
}
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
}
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
/* If LaunchTime is enabled, set DataTranTIM. */
if (ring->launchtime_enable) {
/* Always set DataTranTIM on TQAVCTRL if LaunchTime is enabled
* for any of the SR queues, and configure fetchtime delta.
* XXX NOTE:
* - LaunchTime will be enabled for all SR queues.
* - A fixed offset can be added relative to the launch
* time of all packets if configured at reg LAUNCH_OS0.
* We are keeping it as 0 for now (default value).
*/
tqavctrl = rd32(E1000_I210_TQAVCTRL);
tqavctrl |= E1000_TQAVCTRL_DATATRANTIM |
E1000_TQAVCTRL_FETCHTIME_DELTA;
wr32(E1000_I210_TQAVCTRL, tqavctrl);
} else {
/* If Launchtime is not enabled for any SR queues anymore,
* then clear DataTranTIM on TQAVCTRL and clear fetchtime delta,
* effectively disabling Launchtime.
*/
if (!is_any_txtime_enabled(adapter)) {
tqavctrl = rd32(E1000_I210_TQAVCTRL);
tqavctrl &= ~E1000_TQAVCTRL_DATATRANTIM;
tqavctrl &= ~E1000_TQAVCTRL_FETCHTIME_DELTA;
wr32(E1000_I210_TQAVCTRL, tqavctrl);
}
}
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
/* XXX: In i210 controller the sendSlope and loCredit parameters from
* CBS are not configurable by software so we don't do any 'controller
* configuration' in respect to these parameters.
*/
igb: Fix format with line continuation whitespace The line continuation unintentionally adds whitespace so instead use a coalesced format to remove the whitespace. Miscellanea: o Use a more typical style for ternaries and arguments for this logging message Signed-off-by: Joe Perches <joe@perches.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-11-01 07:03:31 +00:00
netdev_dbg(netdev, "Qav Tx mode: cbs %s, launchtime %s, queue %d idleslope %d sendslope %d hiCredit %d locredit %d\n",
ring->cbs_enable ? "enabled" : "disabled",
ring->launchtime_enable ? "enabled" : "disabled",
queue,
ring->idleslope, ring->sendslope,
ring->hicredit, ring->locredit);
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
}
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
static int igb_save_txtime_params(struct igb_adapter *adapter, int queue,
bool enable)
{
struct igb_ring *ring;
if (queue < 0 || queue > adapter->num_tx_queues)
return -EINVAL;
ring = adapter->tx_ring[queue];
ring->launchtime_enable = enable;
return 0;
}
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
static int igb_save_cbs_params(struct igb_adapter *adapter, int queue,
bool enable, int idleslope, int sendslope,
int hicredit, int locredit)
{
struct igb_ring *ring;
if (queue < 0 || queue > adapter->num_tx_queues)
return -EINVAL;
ring = adapter->tx_ring[queue];
ring->cbs_enable = enable;
ring->idleslope = idleslope;
ring->sendslope = sendslope;
ring->hicredit = hicredit;
ring->locredit = locredit;
return 0;
}
igb: Refactor igb_configure_cbs() Make this function retrieve what it needs from the Tx ring being addressed since it already relies on what had been saved on it before. Also, since this function will be used by the upcoming Launchtime patches rename it to better reflect its intention. Note that Launchtime is not part of what 802.1Qav specifies, but the i210 datasheet refers to this set of functionality as "Qav Transmission Mode". Here we also perform a tiny refactor at is_any_cbs_enabled(), and add further documentation to igb_setup_tx_mode(). Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:55 +00:00
/**
* igb_setup_tx_mode - Switch to/from Qav Tx mode when applicable
* @adapter: pointer to adapter struct
*
* Configure TQAVCTRL register switching the controller's Tx mode
* if FQTSS mode is enabled or disabled. Additionally, will issue
* a call to igb_config_tx_modes() per queue so any previously saved
* Tx parameters are applied.
**/
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
static void igb_setup_tx_mode(struct igb_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
struct e1000_hw *hw = &adapter->hw;
u32 val;
/* Only i210 controller supports changing the transmission mode. */
if (hw->mac.type != e1000_i210)
return;
if (is_fqtss_enabled(adapter)) {
int i, max_queue;
/* Configure TQAVCTRL register: set transmit mode to 'Qav',
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
* set data fetch arbitration to 'round robin', set SP_WAIT_SR
* so SP queues wait for SR ones.
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
*/
val = rd32(E1000_I210_TQAVCTRL);
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
val |= E1000_TQAVCTRL_XMIT_MODE | E1000_TQAVCTRL_SP_WAIT_SR;
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
val &= ~E1000_TQAVCTRL_DATAFETCHARB;
wr32(E1000_I210_TQAVCTRL, val);
/* Configure Tx and Rx packet buffers sizes as described in
* i210 datasheet section 7.2.7.7.
*/
val = rd32(E1000_TXPBS);
val &= ~I210_TXPBSIZE_MASK;
val |= I210_TXPBSIZE_PB0_8KB | I210_TXPBSIZE_PB1_8KB |
I210_TXPBSIZE_PB2_4KB | I210_TXPBSIZE_PB3_4KB;
wr32(E1000_TXPBS, val);
val = rd32(E1000_RXPBS);
val &= ~I210_RXPBSIZE_MASK;
igb: Change RXPBSIZE size when setting Qav mode Section 4.5.9 of the datasheet says that the total size of all packet buffers combined (TxPB 0 + 1 + 2 + 3 + RxPB + BMC2OS + OS2BMC) must not exceed 60KB. Today we are configuring a total of 62KB, so reduce the RxPB from 32KB to 30KB in order to respect that. The choice of changing RxPBSIZE here is mainly because it seems more correct to give more priority to the transmit packet buffers over the receiver ones when running in Qav mode. Also, the BMC2OS and OS2BMC sizes are already too short. Signed-off-by: Jesus Sanchez-Palencia <jesus.s.palencia@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-11-17 00:19:23 +00:00
val |= I210_RXPBSIZE_PB_30KB;
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
wr32(E1000_RXPBS, val);
/* Section 8.12.9 states that MAX_TPKT_SIZE from DTXMXPKTSZ
* register should not exceed the buffer size programmed in
* TXPBS. The smallest buffer size programmed in TXPBS is 4kB
* so according to the datasheet we should set MAX_TPKT_SIZE to
* 4kB / 64.
*
* However, when we do so, no frame from queue 2 and 3 are
* transmitted. It seems the MAX_TPKT_SIZE should not be great
* or _equal_ to the buffer size programmed in TXPBS. For this
* reason, we set set MAX_ TPKT_SIZE to (4kB - 1) / 64.
*/
val = (4096 - 1) / 64;
wr32(E1000_I210_DTXMXPKTSZ, val);
/* Since FQTSS mode is enabled, apply any CBS configuration
* previously set. If no previous CBS configuration has been
* done, then the initial configuration is applied, which means
* CBS is disabled.
*/
max_queue = (adapter->num_tx_queues < I210_SR_QUEUES_NUM) ?
adapter->num_tx_queues : I210_SR_QUEUES_NUM;
for (i = 0; i < max_queue; i++) {
igb: Refactor igb_configure_cbs() Make this function retrieve what it needs from the Tx ring being addressed since it already relies on what had been saved on it before. Also, since this function will be used by the upcoming Launchtime patches rename it to better reflect its intention. Note that Launchtime is not part of what 802.1Qav specifies, but the i210 datasheet refers to this set of functionality as "Qav Transmission Mode". Here we also perform a tiny refactor at is_any_cbs_enabled(), and add further documentation to igb_setup_tx_mode(). Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:55 +00:00
igb_config_tx_modes(adapter, i);
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
}
} else {
wr32(E1000_RXPBS, I210_RXPBSIZE_DEFAULT);
wr32(E1000_TXPBS, I210_TXPBSIZE_DEFAULT);
wr32(E1000_I210_DTXMXPKTSZ, I210_DTXMXPKTSZ_DEFAULT);
val = rd32(E1000_I210_TQAVCTRL);
/* According to Section 8.12.21, the other flags we've set when
* enabling FQTSS are not relevant when disabling FQTSS so we
* don't set they here.
*/
val &= ~E1000_TQAVCTRL_XMIT_MODE;
wr32(E1000_I210_TQAVCTRL, val);
}
netdev_dbg(netdev, "FQTSS %s\n", (is_fqtss_enabled(adapter)) ?
"enabled" : "disabled");
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_configure - configure the hardware for RX and TX
* @adapter: private board structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_configure(struct igb_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
int i;
igb_get_hw_control(adapter);
igb: add support for set_rx_mode netdevice operation This patch adds support for the set_rx_mode netdevice operation so that igb can better support multiple unicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:48:56 +00:00
igb_set_rx_mode(netdev);
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
igb_setup_tx_mode(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_restore_vlan(adapter);
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
igb_setup_tctl(adapter);
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
igb_setup_mrqc(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_setup_rctl(adapter);
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
igb: add support of RX network flow classification This patch is meant to allow for RX network flow classification to insert and remove Rx filter by ethtool. Ethtool interface has it's own rules manager Show all filters: $ ethtool -n eth0 4 RX rings available Total 2 rules Signed-off-by: Ruhao Gao <ruhao.gao@ni.com> Signed-off-by: Gangfeng Huang <gangfeng.huang@ni.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-07-06 05:22:54 +00:00
igb_nfc_filter_restore(adapter);
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
igb_configure_tx(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_configure_rx(adapter);
igb: fix init on 82575 with MNG enabled This patch resolves an issue seen on 82575 adapters with managability pass-thru enabled, which could cause the system to panic. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:29 +00:00
igb_rx_fifo_flush_82575(&adapter->hw);
igb: remove IGB_DESC_UNUSED since it is better handled by a function call This patch removes IGB_DESC_UNUSED and replaces it with a function call instead in order to cleanup some of the ugliness introduced by the macro. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-20 00:16:50 +00:00
/* call igb_desc_unused which always leaves
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
* at least 1 descriptor unused to make sure
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* next_to_use != next_to_clean
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
for (i = 0; i < adapter->num_rx_queues; i++) {
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
struct igb_ring *ring = adapter->rx_ring[i];
igb: drop the "adv" off function names relating to descriptors Many of the function names in the hot path are carrying an extra "_adv" suffix on the end of them to represent the fact that they are using advanced descriptors instead of legacy descriptors. However since all igb uses are advanced descriptors adding the extra suffix doesn't really add any additional data. Since this is the case it is easiest to just drop the suffix and save us from having to store the extra characters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:59 +00:00
igb_alloc_rx_buffers(ring, igb_desc_unused(ring));
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
}
igb: Power down link when interface is down This changes the behavior of the driver to power down the link when the associated interface is down, unless management is enabled. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:01:59 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_power_up_link - Power up the phy/serdes link
* @adapter: address of board private structure
igb: Power down link when interface is down This changes the behavior of the driver to power down the link when the associated interface is down, unless management is enabled. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:01:59 +00:00
**/
void igb_power_up_link(struct igb_adapter *adapter)
{
igb: reset PHY in the link_up process to recover PHY setting after power down. There was a previous patch to resolve issue with 82576 losing PHY setting after PHY power down. However that previous implementation triggered speed mismatch and occasional link lost. Now, this patch resolves both initial PHY setting and speed mismatch issues. Signed-off-by: Akeem G. Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-07-17 04:51:18 +00:00
igb_reset_phy(&adapter->hw);
igb: Power down link when interface is down This changes the behavior of the driver to power down the link when the associated interface is down, unless management is enabled. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:01:59 +00:00
if (adapter->hw.phy.media_type == e1000_media_type_copper)
igb_power_up_phy_copper(&adapter->hw);
else
igb_power_up_serdes_link_82575(&adapter->hw);
igb: bring link up when PHY is powered up Call igb_setup_link() when the PHY is powered up. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Reported-by: Jeff Westfahl <jeff.westfahl@ni.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-06-17 06:58:11 +00:00
igb_setup_link(&adapter->hw);
igb: Power down link when interface is down This changes the behavior of the driver to power down the link when the associated interface is down, unless management is enabled. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:01:59 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_power_down_link - Power down the phy/serdes link
* @adapter: address of board private structure
igb: Power down link when interface is down This changes the behavior of the driver to power down the link when the associated interface is down, unless management is enabled. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:01:59 +00:00
*/
static void igb_power_down_link(struct igb_adapter *adapter)
{
if (adapter->hw.phy.media_type == e1000_media_type_copper)
igb_power_down_phy_copper_82575(&adapter->hw);
else
igb_shutdown_serdes_link_82575(&adapter->hw);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
/**
* Detect and switch function for Media Auto Sense
* @adapter: address of the board private structure
**/
static void igb_check_swap_media(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
u32 ctrl_ext, connsw;
bool swap_now = false;
ctrl_ext = rd32(E1000_CTRL_EXT);
connsw = rd32(E1000_CONNSW);
/* need to live swap if current media is copper and we have fiber/serdes
* to go to.
*/
if ((hw->phy.media_type == e1000_media_type_copper) &&
(!(connsw & E1000_CONNSW_AUTOSENSE_EN))) {
swap_now = true;
} else if (!(connsw & E1000_CONNSW_SERDESD)) {
/* copper signal takes time to appear */
if (adapter->copper_tries < 4) {
adapter->copper_tries++;
connsw |= E1000_CONNSW_AUTOSENSE_CONF;
wr32(E1000_CONNSW, connsw);
return;
} else {
adapter->copper_tries = 0;
if ((connsw & E1000_CONNSW_PHYSD) &&
(!(connsw & E1000_CONNSW_PHY_PDN))) {
swap_now = true;
connsw &= ~E1000_CONNSW_AUTOSENSE_CONF;
wr32(E1000_CONNSW, connsw);
}
}
}
if (!swap_now)
return;
switch (hw->phy.media_type) {
case e1000_media_type_copper:
netdev_info(adapter->netdev,
"MAS: changing media to fiber/serdes\n");
ctrl_ext |=
E1000_CTRL_EXT_LINK_MODE_PCIE_SERDES;
adapter->flags |= IGB_FLAG_MEDIA_RESET;
adapter->copper_tries = 0;
break;
case e1000_media_type_internal_serdes:
case e1000_media_type_fiber:
netdev_info(adapter->netdev,
"MAS: changing media to copper\n");
ctrl_ext &=
~E1000_CTRL_EXT_LINK_MODE_PCIE_SERDES;
adapter->flags |= IGB_FLAG_MEDIA_RESET;
break;
default:
/* shouldn't get here during regular operation */
netdev_err(adapter->netdev,
"AMS: Invalid media type found, returning\n");
break;
}
wr32(E1000_CTRL_EXT, ctrl_ext);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_up - Open the interface and prepare it to handle traffic
* @adapter: board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
int igb_up(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
int i;
/* hardware has been reset, we need to reload some things */
igb_configure(adapter);
clear_bit(__IGB_DOWN, &adapter->state);
igb: avoid unnecessarily creating a local copy of the q_vector This is mostly a drop of unnecessary pointer defines for q_vector when we don't have issues with line width and don't have multiple references to the pointer. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:34 +00:00
for (i = 0; i < adapter->num_q_vectors; i++)
napi_enable(&(adapter->q_vector[i]->napi));
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
if (adapter->flags & IGB_FLAG_HAS_MSIX)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_configure_msix(adapter);
igb: make certain to reassign legacy interrupt vectors after reset This change corrects an issue that will cause false hangs when using either 82575 or 82580 in legacy interrupt mode. The issue is caused when there is a slow traffic flow and an "ethtool -r" is executed while using legacy or MSI interrupts. MSI-X is not affected by this issue due to the fact that we were already reconfiguring the vectors after reset. If possible it would be best to push this for net-2.6 since it is resolving a bug but if that is not possible then net-next-2.6 will be fine. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-03 21:59:51 +00:00
else
igb_assign_vector(adapter->q_vector[0], 0);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Clear any pending interrupts. */
igb: Clear TSICR interrupts together with ICR Issuing "ip link set up/down" can block TSICR interrupts, what results in missing PTP Tx timestamp and no PPS pulse generation. Problem happens when the link is set up with the TSICR interrupts pending. ICR is cleared before enabling interrupts, while TSICR is not. When all TSICR interrupts are pending at this moment, time_sync interrupt will never be generated. TSICR should be cleared as well. In order to reproduce the issue: 1. Setup linux with IEEE 1588 grandmaster and PPS output enabled 2. Continue setting link up/down with random intervals between commands 3. Wait until PPS is not generated ( only one pulse is generated and PPS dies), and ptp4l complains constantly about Tx timeout. Signed-off-by: Joanna Yurdal <jyu@trackman.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-05-16 11:14:11 +00:00
rd32(E1000_TSICR);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
rd32(E1000_ICR);
igb_irq_enable(adapter);
igb: delay VF reset notification until after interrupts are enabed This update delays the VF reset notification until after interrupts are enabled. Otherwise there is a chance of having the VF try to reset itself too soon and being ignored by the PF as a result. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:45 +00:00
/* notify VFs that reset has been completed */
if (adapter->vfs_allocated_count) {
u32 reg_data = rd32(E1000_CTRL_EXT);
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: delay VF reset notification until after interrupts are enabed This update delays the VF reset notification until after interrupts are enabled. Otherwise there is a chance of having the VF try to reset itself too soon and being ignored by the PF as a result. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:45 +00:00
reg_data |= E1000_CTRL_EXT_PFRSTD;
wr32(E1000_CTRL_EXT, reg_data);
}
e1000/e1000e/igb/ixgb: do not use netif_wake_queue un-necessarily It was pointed out that the Intel wired ethernet drivers do not need to wake the tx queue since netif_carrier_on/off will take care of the qdisc management in order to guarantee the correct handling of the transmit routine enable state. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-21 18:42:05 +00:00
netif_tx_start_all_queues(adapter->netdev);
igb: cleanup interrupt enablement in regards to msix_other This patch changes a few things so that instead of firing a link status interrupt directly the get_link_status bit is set and the watchdog is scheduled. In addition the mailbox bit is now only enabled if VFs are enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:59 +00:00
/* start the watchdog. */
hw->mac.get_link_status = 1;
schedule_work(&adapter->watchdog_task);
igb: Fix for devices using ethtool for EEE settings This patch fixes a problem where using ethtool for EEE setting was not working correctly. This patch also fixes a problem where the function that checks for EEE status on i354 devices was not being called and was causing warnings with static analysis tools. Reported-by: Rashika Kheria <rashika.kheria@gmail.com> Reported-by: Josh Triplett <josh@joshtriplett.org> Reported-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-12 03:58:22 +00:00
if ((adapter->flags & IGB_FLAG_EEE) &&
(!hw->dev_spec._82575.eee_disable))
adapter->eee_advert = MDIO_EEE_100TX | MDIO_EEE_1000T;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return 0;
}
void igb_down(struct igb_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
igb: replace unecessary &adapter->hw with just hw where applicable This patch just cleans up some unecessary references to the adapter->hw member when it has already been placed in a local variable named hw. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:35 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
u32 tctl, rctl;
int i;
/* signal that we're down so the interrupt handler does not
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* reschedule our watchdog timer
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
set_bit(__IGB_DOWN, &adapter->state);
/* disable receives in the hardware */
rctl = rd32(E1000_RCTL);
wr32(E1000_RCTL, rctl & ~E1000_RCTL_EN);
/* flush and sleep below */
igb: Fix error of RX network flow classification After add an ethertype filter, if user change the adapter speed several times, the error "ethtool -N: etype filters are all used" is reported by igb driver. In older patch, function igb_nfc_filter_exit() and igb_nfc_filter_restore() is not paried. igb_nfc_filter_restore() exist in igb_up(), but function igb_nfc_filter_exit() is exist in __igb_close(). In the process of speed changing, only igb_nfc_filter_restore() is called, it will take a position of ethertype bitmap. Reproduce steps: Step 1: Add a etype filter by ethtool $ethtool -N eth0 flow-type ether proto 0x88F8 action 1 Step 2: Change the adapter speed to 100M/full duplex $ethtool -s eth0 speed 100 duplex full Step 3: Change the adapter speed to 1000M/full duplex ethtool -s eth0 speed 1000 duplex full Repeat step2 and step3, then dmesg the system log, you can find the error message, add new ethtype filter is also failed. This fixing is move igb_nfc_filter_exit() from __igb_close() to igb_down() to make igb_nfc_filter_restore()/igb_nfc_filter_exit() is paired. Signed-off-by: Gangfeng Huang <gangfeng.huang@ni.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-27 01:17:53 +00:00
igb_nfc_filter_exit(adapter);
igb: use netif_carrier_off earlier when bringing if down Use netif_carrier_off() first, since that will prevent the stack from queuing more packets to this IF. This operation is fast, and should behave much nicer when trying to bring down an interface under load. Reported-by: Eliezer Tamir <eliezer.tamir@linux.intel.com> Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-03-21 00:41:53 +00:00
netif_carrier_off(netdev);
net: Use queue aware tests throughout. This effectively "flips the switch" by making the core networking and multiqueue-aware drivers use the new TX multiqueue structures. Non-multiqueue drivers need no changes. The interfaces they use such as netif_stop_queue() degenerate into an operation on TX queue zero. So everything "just works" for them. Code that really wants to do "X" to all TX queues now invokes a routine that does so, such as netif_tx_wake_all_queues(), netif_tx_stop_all_queues(), etc. pktgen and netpoll required a little bit more surgery than the others. In particular the pktgen changes, whilst functional, could be largely improved. The initial check in pktgen_xmit() will sometimes check the wrong queue, which is mostly harmless. The thing to do is probably to invoke fill_packet() earlier. The bulk of the netpoll changes is to make the code operate solely on the TX queue indicated by by the SKB queue mapping. Setting of the SKB queue mapping is entirely confined inside of net/core/dev.c:dev_pick_tx(). If we end up needing any kind of special semantics (drops, for example) it will be implemented here. Finally, we now have a "real_num_tx_queues" which is where the driver indicates how many TX queues are actually active. With IGB changes from Jeff Kirsher. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-17 08:56:23 +00:00
netif_tx_stop_all_queues(netdev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* disable transmits in the hardware */
tctl = rd32(E1000_TCTL);
tctl &= ~E1000_TCTL_EN;
wr32(E1000_TCTL, tctl);
/* flush both disables and wait for them to finish */
wrfl();
igb: Cleanups to fix msleep warnings This patch fixes WARNING:MSLEEP found by checkpatch check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:46:40 +00:00
usleep_range(10000, 11000);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Fix possible panic caused by Rx traffic arrival while interface is down This patch reorders disabling napi and irqs during igb_down. This is done to avoid possible panic's found in other Intel drivers when Rx traffic arrives while interface is going down. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-30 00:21:32 +00:00
igb_irq_disable(adapter);
igb: Implementation of 1-sec delay for i210 devices This patch adds 1 sec delay mechanism to i210 device family, in order to avoid erroneous link issue with the link partner. Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-08-28 02:22:43 +00:00
adapter->flags &= ~IGB_FLAG_NEED_LINK_UPDATE;
igb: Fix possible panic caused by Rx traffic arrival while interface is down This patch reorders disabling napi and irqs during igb_down. This is done to avoid possible panic's found in other Intel drivers when Rx traffic arrives while interface is going down. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-30 00:21:32 +00:00
for (i = 0; i < adapter->num_q_vectors; i++) {
igb: Fixes needed for surprise removal support This patch adds some checks in order to prevent panic's on surprise removal of devices during S0, S3, S4. Without this patch, Thunderbolt type device removal will panic the system. Signed-off-by: Yanir Lubetkin <yanirx.lubetkin@intel.com> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-22 07:52:54 +00:00
if (adapter->q_vector[i]) {
napi_synchronize(&adapter->q_vector[i]->napi);
napi_disable(&adapter->q_vector[i]->napi);
}
igb: Fix possible panic caused by Rx traffic arrival while interface is down This patch reorders disabling napi and irqs during igb_down. This is done to avoid possible panic's found in other Intel drivers when Rx traffic arrives while interface is going down. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-30 00:21:32 +00:00
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
del_timer_sync(&adapter->watchdog_timer);
del_timer_sync(&adapter->phy_info_timer);
igb: update stats before doing reset in igb_down It was seen with repeated interface up/down testing that there was a large stray between the stats reported by the queues and the stats reported by the HW. It was found to be an issue in that hw stats were being reset without first being recorded. This change records the stats before wiping them from the system via the reset. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:22:32 +00:00
/* record the stats before reset*/
igb: reduce CPU0 latency when updating statistics This change is based off of the work and suggestion of Jan Jablonsky <jan.jablonsky@thalesgroup.com>. The Watchdog workqueue in igb driver is scheduled every 2s for each network interface. That includes updating a statistics protected by spinlock. Function igb_update_stats in this case will be protected against preemption. According to number of a statistics registers (cca 60), processing this function might cause additional cpu load on CPU0. In case of statistics spinlock may be replaced with mutex, which reduce latency on CPU0. CC: Bernhard Kaindl <bernhard.kaindl@thalesgroup.com> CC: Jan Jablonsky <jan.jablonsky@thalesgroup.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-08-02 17:13:10 +00:00
mutex_lock(&adapter->stats64_lock);
igb: Remove useless argument Given that all callers of igb_update_stats() pass the same two arguments: (adapter, &adapter->stats64), the second argument can be removed. Signed-off-by: Benjamin Poirier <bpoirier@suse.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-16 22:55:16 +00:00
igb_update_stats(adapter);
igb: reduce CPU0 latency when updating statistics This change is based off of the work and suggestion of Jan Jablonsky <jan.jablonsky@thalesgroup.com>. The Watchdog workqueue in igb driver is scheduled every 2s for each network interface. That includes updating a statistics protected by spinlock. Function igb_update_stats in this case will be protected against preemption. According to number of a statistics registers (cca 60), processing this function might cause additional cpu load on CPU0. In case of statistics spinlock may be replaced with mutex, which reduce latency on CPU0. CC: Bernhard Kaindl <bernhard.kaindl@thalesgroup.com> CC: Jan Jablonsky <jan.jablonsky@thalesgroup.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-08-02 17:13:10 +00:00
mutex_unlock(&adapter->stats64_lock);
igb: update stats before doing reset in igb_down It was seen with repeated interface up/down testing that there was a large stray between the stats reported by the queues and the stats reported by the HW. It was found to be an issue in that hw stats were being reset without first being recorded. This change records the stats before wiping them from the system via the reset. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:22:32 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
adapter->link_speed = 0;
adapter->link_duplex = 0;
igb: fix EEH recovery during reset on PPC EEH is not recovering in a reasonable amount of time on PPC during igb_down(). Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-25 00:01:15 +00:00
if (!pci_channel_offline(adapter->pdev))
igb_reset(adapter);
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
/* clear VLAN promisc flag so VFTA will be updated if necessary */
adapter->flags &= ~IGB_FLAG_VLAN_PROMISC;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_clean_all_tx_rings(adapter);
igb_clean_all_rx_rings(adapter);
igb: do not re-register DCA requester on device reset The current driver is re-registering the DCA requester after every reset. Instead of doing this we should only be updating the on board DCA registers and not unregistering/re-registering our requester. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-21 13:06:56 +00:00
#ifdef CONFIG_IGB_DCA
/* since we reset the hardware DCA settings were cleared */
igb_setup_dca(adapter);
#endif
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
void igb_reinit_locked(struct igb_adapter *adapter)
{
WARN_ON(in_interrupt());
while (test_and_set_bit(__IGB_RESETTING, &adapter->state))
igb: Cleanups to fix msleep warnings This patch fixes WARNING:MSLEEP found by checkpatch check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:46:40 +00:00
usleep_range(1000, 2000);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_down(adapter);
igb_up(adapter);
clear_bit(__IGB_RESETTING, &adapter->state);
}
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
/** igb_enable_mas - Media Autosense re-enable after swap
*
* @adapter: adapter struct
**/
igb: simplify and clean up igb_enable_mas() igb_enable_mas() should only be called for the 82575 and has no clear return so changing it to void. Also simplify the odd conditional expression. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-05-02 07:39:03 +00:00
static void igb_enable_mas(struct igb_adapter *adapter)
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
{
struct e1000_hw *hw = &adapter->hw;
igb: simplify and clean up igb_enable_mas() igb_enable_mas() should only be called for the 82575 and has no clear return so changing it to void. Also simplify the odd conditional expression. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-05-02 07:39:03 +00:00
u32 connsw = rd32(E1000_CONNSW);
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
/* configure for SerDes media detect */
igb: simplify and clean up igb_enable_mas() igb_enable_mas() should only be called for the 82575 and has no clear return so changing it to void. Also simplify the odd conditional expression. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-05-02 07:39:03 +00:00
if ((hw->phy.media_type == e1000_media_type_copper) &&
(!(connsw & E1000_CONNSW_SERDESD))) {
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
connsw |= E1000_CONNSW_ENRGSRC;
connsw |= E1000_CONNSW_AUTOSENSE_EN;
wr32(E1000_CONNSW, connsw);
wrfl();
}
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
void igb_reset(struct igb_adapter *adapter)
{
igb: add pci_dev in few spots to clean up use of dev_err/info/warn This patch relpaces several references to adapter->pdev->dev with just pdev->dev. This allows for cleanup of several multiline dev_err/info calls. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:55 +00:00
struct pci_dev *pdev = adapter->pdev;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
struct e1000_mac_info *mac = &hw->mac;
struct e1000_fc_info *fc = &hw->fc;
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
u32 pba, hwm;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Repartition Pba for greater than 9k mtu
* To take effect CTRL.RST is required.
*/
igb: change pba size determination from if to switch statement As additional hardware is added to the igb driver it is easier to support the expansion via switch statements instead of using nested ifs. For this reason I am changing this to a switch statement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:21:31 +00:00
switch (mac->type) {
igb: add support for Intel I350 Gigabit Network Connection This patch adds support for the the I350 Gigabit network connection which is the follow-on part to the 82580. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> CC: James Hearn <james.r.hearn@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-22 14:08:06 +00:00
case e1000_i350:
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
case e1000_i354:
igb: Add full support for 82580 devices This patch makes use of the 82580 PHY and MAC support added and adds a set of supported device IDs for said hardware. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-19 12:42:21 +00:00
case e1000_82580:
pba = rd32(E1000_RXPBS);
pba = igb_rxpbs_adjust_82580(pba);
break;
igb: change pba size determination from if to switch statement As additional hardware is added to the igb driver it is easier to support the expansion via switch statements instead of using nested ifs. For this reason I am changing this to a switch statement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:21:31 +00:00
case e1000_82576:
igb: use packet buffer sizes from RXPBS register This patch changes the configuration for 82576 so that it uses the actual value of the 82576 rx packet buffer size instead of just assuming the value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:38 +00:00
pba = rd32(E1000_RXPBS);
pba &= E1000_RXPBS_SIZE_MASK_82576;
igb: change pba size determination from if to switch statement As additional hardware is added to the igb driver it is easier to support the expansion via switch statements instead of using nested ifs. For this reason I am changing this to a switch statement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:21:31 +00:00
break;
case e1000_82575:
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
case e1000_i210:
case e1000_i211:
igb: change pba size determination from if to switch statement As additional hardware is added to the igb driver it is easier to support the expansion via switch statements instead of using nested ifs. For this reason I am changing this to a switch statement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:21:31 +00:00
default:
pba = E1000_PBA_34K;
break;
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
if (mac->type == e1000_82575) {
u32 min_rx_space, min_tx_space, needed_tx_space;
/* write Rx PBA so that hardware can report correct Tx PBA */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
wr32(E1000_PBA, pba);
/* To maintain wire speed transmits, the Tx FIFO should be
* large enough to accommodate two full transmit packets,
* rounded up to the next 1KB and expressed in KB. Likewise,
* the Rx FIFO should be large enough to accommodate at least
* one full receive packet and is similarly rounded up and
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* expressed in KB.
*/
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
min_rx_space = DIV_ROUND_UP(MAX_JUMBO_FRAME_SIZE, 1024);
/* The Tx FIFO also stores 16 bytes of information about the Tx
* but don't include Ethernet FCS because hardware appends it.
* We only need to round down to the nearest 512 byte block
* count since the value we care about is 2 frames, not 1.
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
*/
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
min_tx_space = adapter->max_frame_size;
min_tx_space += sizeof(union e1000_adv_tx_desc) - ETH_FCS_LEN;
min_tx_space = DIV_ROUND_UP(min_tx_space, 512);
/* upper 16 bits has Tx packet buffer allocation size in KB */
needed_tx_space = min_tx_space - (rd32(E1000_PBA) >> 16);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* If current Tx allocation is less than the min Tx FIFO size,
* and the min Tx FIFO size is less than the current Rx FIFO
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
* allocation, take space away from current Rx allocation.
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
*/
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
if (needed_tx_space < pba) {
pba -= needed_tx_space;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* if short on Rx space, Rx wins and must trump Tx
* adjustment
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (pba < min_rx_space)
pba = min_rx_space;
}
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
/* adjust PBA for jumbo frames */
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
wr32(E1000_PBA, pba);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
/* flow control settings
* The high water mark must be low enough to fit one full frame
* after transmitting the pause frame. As such we must have enough
* space to allow for us to complete our current transmit and then
* receive the frame that is in progress from the link partner.
* Set it to:
* - the full Rx FIFO size minus one full Tx plus one full Rx frame
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
*/
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
hwm = (pba << 10) - (adapter->max_frame_size + MAX_JUMBO_FRAME_SIZE);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Use a 32-bit mask when calculating the flow control watermarks For some devices, the result of the flow control high watermark gets truncated when programming it into the registers because of the mask used. Switch the mask to 32-bit to prevent this from happening. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-11-08 04:03:58 +00:00
fc->high_water = hwm & 0xFFFFFFF0; /* 16-byte granularity */
igb: correctly offset 82575 flow control watermarks by 16 bytes The watermark values for the 82575 were not being set correctly. As a result the high and low watermark values were set to the same value which can lead to excess xon/xoff packets being generated. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-23 13:21:27 +00:00
fc->low_water = fc->high_water - 16;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
fc->pause_time = 0xFFFF;
fc->send_xon = 1;
igb: cleanup flow control configuration to make requested/current more clear This patch cleans up the flow control configuration for igb to make it a bit more readable in regards to what the requested and current modes are. This should help with the maintainability of the current igb driver in regards to flow control. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:10:24 +00:00
fc->current_mode = fc->requested_mode;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* disable receive for all VFs and wait one second */
if (adapter->vfs_allocated_count) {
int i;
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
for (i = 0 ; i < adapter->vfs_allocated_count; i++)
igb: Fix overwrite of the VF's flags The vf flags were being overwritten causing the flag that indicates that the PF has set the VF MAC address to get cleared. This would allow the VF to override the MAC address assigned by the Host VMM using the ip link set command. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 05:43:21 +00:00
adapter->vf_data[i].flags &= IGB_VF_FLAG_PF_SET_MAC;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* ping all the active vfs to let them know we are going down */
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
igb_ping_all_vfs(adapter);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* disable transmits and receives */
wr32(E1000_VFRE, 0);
wr32(E1000_VFTE, 0);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Allow time for pending master requests to run */
igb: replace unecessary &adapter->hw with just hw where applicable This patch just cleans up some unecessary references to the adapter->hw member when it has already been placed in a local variable named hw. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:35 +00:00
hw->mac.ops.reset_hw(hw);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
wr32(E1000_WUC, 0);
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
if (adapter->flags & IGB_FLAG_MEDIA_RESET) {
/* need to resetup here after media swap */
adapter->ei.get_invariants(hw);
adapter->flags &= ~IGB_FLAG_MEDIA_RESET;
}
igb: simplify and clean up igb_enable_mas() igb_enable_mas() should only be called for the 82575 and has no clear return so changing it to void. Also simplify the odd conditional expression. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-05-02 07:39:03 +00:00
if ((mac->type == e1000_82575) &&
(adapter->flags & IGB_FLAG_MAS_ENABLE)) {
igb_enable_mas(adapter);
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
}
igb: replace unecessary &adapter->hw with just hw where applicable This patch just cleans up some unecessary references to the adapter->hw member when it has already been placed in a local variable named hw. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:35 +00:00
if (hw->mac.ops.init_hw(hw))
igb: add pci_dev in few spots to clean up use of dev_err/info/warn This patch relpaces several references to adapter->pdev->dev with just pdev->dev. This allows for cleanup of several multiline dev_err/info calls. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:55 +00:00
dev_err(&pdev->dev, "Hardware Error\n");
igb: Add DMA Coalescing feature to driver This patch add DMA Coalescing which is a power-saving feature that coalesces DMA writes in order to stay in a low-power state as much as possible. Feature is disabled by default. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-03-12 04:43:54 +00:00
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
/* RAR registers were cleared during init_hw, clear mac table */
igb_flush_mac_table(adapter);
__dev_uc_unsync(adapter->netdev, NULL);
/* Recover default RAR entry */
igb_set_default_mac_filter(adapter);
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Flow control settings reset on hardware reset, so guarantee flow
igb: Force flow control off during reset when forcing speed. During igb_reset(), we initiate a hardware reset which will clear our flow control settings. For auto-negotiation, we re-negotiate them when linking up again, but we need to force them off properly for the forced speed case. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-18 02:57:44 +00:00
* control is off when forcing speed.
*/
if (!hw->mac.autoneg)
igb_force_mac_fc(hw);
igb: Move DMA Coalescing init code to separate function. This patch moves the DMA Coalescing feature initialization code from igb_reset to a new function and replaces it with a call to the new function. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-13 17:29:59 +00:00
igb_init_dmac(adapter, pba);
igb: Enable hwmon data output for thermal sensors via I2C. Some of our adapters have internal sensors that report thermal data. This patch enables reporting of that data via sysfs. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:01:42 +00:00
#ifdef CONFIG_IGB_HWMON
/* Re-initialize the thermal sensor on i350 devices. */
if (!test_bit(__IGB_DOWN, &adapter->state)) {
if (mac->type == e1000_i350 && hw->bus.func == 0) {
/* If present, re-initialize the external thermal sensor
* interface.
*/
if (adapter->ets)
mac->ops.init_thermal_sensor_thresh(hw);
}
}
#endif
igb: Fix code comment Recently added code comment was missing a space that is needed. Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-13 23:07:14 +00:00
/* Re-establish EEE setting */
igb: Fix for devices using ethtool for EEE settings This patch fixes a problem where using ethtool for EEE setting was not working correctly. This patch also fixes a problem where the function that checks for EEE status on i354 devices was not being called and was causing warnings with static analysis tools. Reported-by: Rashika Kheria <rashika.kheria@gmail.com> Reported-by: Josh Triplett <josh@joshtriplett.org> Reported-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-12 03:58:22 +00:00
if (hw->phy.media_type == e1000_media_type_copper) {
switch (mac->type) {
case e1000_i350:
case e1000_i210:
case e1000_i211:
igb: add flags to set eee advertisement mode Change e1000_set_eee and e1000_set_eee_i35(0|4) to allow changes in the advertised EEE speeds from ethtool. Adds two boolean flags to e1000_set_eee_i35(0|4) to pass in advertised speed data. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-08-29 06:43:13 +00:00
igb_set_eee_i350(hw, true, true);
igb: Fix for devices using ethtool for EEE settings This patch fixes a problem where using ethtool for EEE setting was not working correctly. This patch also fixes a problem where the function that checks for EEE status on i354 devices was not being called and was causing warnings with static analysis tools. Reported-by: Rashika Kheria <rashika.kheria@gmail.com> Reported-by: Josh Triplett <josh@joshtriplett.org> Reported-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-12 03:58:22 +00:00
break;
case e1000_i354:
igb: add flags to set eee advertisement mode Change e1000_set_eee and e1000_set_eee_i35(0|4) to allow changes in the advertised EEE speeds from ethtool. Adds two boolean flags to e1000_set_eee_i35(0|4) to pass in advertised speed data. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-08-29 06:43:13 +00:00
igb_set_eee_i354(hw, true, true);
igb: Fix for devices using ethtool for EEE settings This patch fixes a problem where using ethtool for EEE setting was not working correctly. This patch also fixes a problem where the function that checks for EEE status on i354 devices was not being called and was causing warnings with static analysis tools. Reported-by: Rashika Kheria <rashika.kheria@gmail.com> Reported-by: Josh Triplett <josh@joshtriplett.org> Reported-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-12 03:58:22 +00:00
break;
default:
break;
}
}
igb: Power down link when interface is down This changes the behavior of the driver to power down the link when the associated interface is down, unless management is enabled. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:01:59 +00:00
if (!netif_running(adapter->netdev))
igb_power_down_link(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_update_mng_vlan(adapter);
/* Enable h/w to recognize an 802.1Q VLAN Ethernet packet */
wr32(E1000_VET, ETHERNET_IEEE_VLAN_TYPE);
igb: Prevent dropped Tx timestamps via work items and interrupts. In rare circumstances, it's possible a descriptor writeback will occur before a timestamped Tx packet will go out on the wire, leading to the driver believing the hardware failed to timestamp the packet. Schedule a work item for 82576 and use the available time sync interrupt registers on 82580 and above to account for this. Cc: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-18 07:26:33 +00:00
/* Re-enable PTP, where applicable. */
igb: re-use igb_ptp_reset in igb_ptp_init Modify igb_ptp_init to take advantage of igb_ptp_reset, and remove duplicated work that was occurring in both igb_ptp_reset and igb_ptp_init. In total, resetting the TSAUXC register, and resetting the system time both happen in igb_ptp_reset already. igb_ptp_reset now also takes care of starting the delayed work item for overflow checks, as well. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-05-24 20:56:29 +00:00
if (adapter->ptp_flags & IGB_PTP_ENABLED)
igb_ptp_reset(adapter);
igb: Prevent dropped Tx timestamps via work items and interrupts. In rare circumstances, it's possible a descriptor writeback will occur before a timestamped Tx packet will go out on the wire, leading to the driver believing the hardware failed to timestamp the packet. Schedule a work item for 82576 and use the available time sync interrupt registers on 82580 and above to account for this. Cc: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-18 07:26:33 +00:00
igb: replace unecessary &adapter->hw with just hw where applicable This patch just cleans up some unecessary references to the adapter->hw member when it has already been placed in a local variable named hw. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:35 +00:00
igb_get_phy_info(hw);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
net: introduce and use netdev_features_t for device features sets v2: add couple missing conversions in drivers split unexporting netdev_fix_features() implemented %pNF convert sock::sk_route_(no?)caps Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-15 15:29:55 +00:00
static netdev_features_t igb_fix_features(struct net_device *netdev,
netdev_features_t features)
igb: do vlan cleanup - unify vlan and nonvlan rx path - kill adapter->vlgrp and igb_vlan_rx_register - allow to turn on/off rx/tx vlan accel via ethtool (set_features) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-21 03:27:27 +00:00
{
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Since there is no support for separate Rx/Tx vlan accel
* enable/disable make sure Tx flag is always in same state as Rx.
igb: do vlan cleanup - unify vlan and nonvlan rx path - kill adapter->vlgrp and igb_vlan_rx_register - allow to turn on/off rx/tx vlan accel via ethtool (set_features) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-21 03:27:27 +00:00
*/
net: vlan: rename NETIF_F_HW_VLAN_* feature flags to NETIF_F_HW_VLAN_CTAG_* Rename the hardware VLAN acceleration features to include "CTAG" to indicate that they only support CTAGs. Follow up patches will introduce 802.1ad server provider tagging (STAGs) and require the distinction for hardware not supporting acclerating both. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-19 02:04:27 +00:00
if (features & NETIF_F_HW_VLAN_CTAG_RX)
features |= NETIF_F_HW_VLAN_CTAG_TX;
igb: do vlan cleanup - unify vlan and nonvlan rx path - kill adapter->vlgrp and igb_vlan_rx_register - allow to turn on/off rx/tx vlan accel via ethtool (set_features) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-21 03:27:27 +00:00
else
net: vlan: rename NETIF_F_HW_VLAN_* feature flags to NETIF_F_HW_VLAN_CTAG_* Rename the hardware VLAN acceleration features to include "CTAG" to indicate that they only support CTAGs. Follow up patches will introduce 802.1ad server provider tagging (STAGs) and require the distinction for hardware not supporting acclerating both. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-19 02:04:27 +00:00
features &= ~NETIF_F_HW_VLAN_CTAG_TX;
igb: do vlan cleanup - unify vlan and nonvlan rx path - kill adapter->vlgrp and igb_vlan_rx_register - allow to turn on/off rx/tx vlan accel via ethtool (set_features) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-21 03:27:27 +00:00
return features;
}
net: introduce and use netdev_features_t for device features sets v2: add couple missing conversions in drivers split unexporting netdev_fix_features() implemented %pNF convert sock::sk_route_(no?)caps Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-15 15:29:55 +00:00
static int igb_set_features(struct net_device *netdev,
netdev_features_t features)
igb: convert to ndo_fix_features Private rx_csum flags are now duplicate of netdev->features & NETIF_F_RXCSUM. Removing this needs deeper surgery. Things noticed: - HW VLAN acceleration probably can be toggled, but it's left as is - the resets on RX csum offload change can probably be avoided - there is A LOT of copy-and-pasted code here Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-06-08 08:38:01 +00:00
{
net: introduce and use netdev_features_t for device features sets v2: add couple missing conversions in drivers split unexporting netdev_fix_features() implemented %pNF convert sock::sk_route_(no?)caps Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-15 15:29:55 +00:00
netdev_features_t changed = netdev->features ^ features;
igb: Support RX-ALL feature flag. This allows the NIC to receive all frames available, including those with bad FCS, un-matched vlans, ethernet control frames, and more. Tested by sending frames with bad FCS. Signed-off-by: Ben Greear <greearb@candelatech.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-03-06 09:41:58 +00:00
struct igb_adapter *adapter = netdev_priv(netdev);
igb: convert to ndo_fix_features Private rx_csum flags are now duplicate of netdev->features & NETIF_F_RXCSUM. Removing this needs deeper surgery. Things noticed: - HW VLAN acceleration probably can be toggled, but it's left as is - the resets on RX csum offload change can probably be avoided - there is A LOT of copy-and-pasted code here Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-06-08 08:38:01 +00:00
net: vlan: rename NETIF_F_HW_VLAN_* feature flags to NETIF_F_HW_VLAN_CTAG_* Rename the hardware VLAN acceleration features to include "CTAG" to indicate that they only support CTAGs. Follow up patches will introduce 802.1ad server provider tagging (STAGs) and require the distinction for hardware not supporting acclerating both. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-19 02:04:27 +00:00
if (changed & NETIF_F_HW_VLAN_CTAG_RX)
igb: do vlan cleanup - unify vlan and nonvlan rx path - kill adapter->vlgrp and igb_vlan_rx_register - allow to turn on/off rx/tx vlan accel via ethtool (set_features) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-21 03:27:27 +00:00
igb_vlan_mode(netdev, features);
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
if (!(changed & (NETIF_F_RXALL | NETIF_F_NTUPLE)))
igb: Support RX-ALL feature flag. This allows the NIC to receive all frames available, including those with bad FCS, un-matched vlans, ethernet control frames, and more. Tested by sending frames with bad FCS. Signed-off-by: Ben Greear <greearb@candelatech.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-03-06 09:41:58 +00:00
return 0;
igb: add support of RX network flow classification This patch is meant to allow for RX network flow classification to insert and remove Rx filter by ethtool. Ethtool interface has it's own rules manager Show all filters: $ ethtool -n eth0 4 RX rings available Total 2 rules Signed-off-by: Ruhao Gao <ruhao.gao@ni.com> Signed-off-by: Gangfeng Huang <gangfeng.huang@ni.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-07-06 05:22:54 +00:00
if (!(features & NETIF_F_NTUPLE)) {
struct hlist_node *node2;
struct igb_nfc_filter *rule;
spin_lock(&adapter->nfc_lock);
hlist_for_each_entry_safe(rule, node2,
&adapter->nfc_filter_list, nfc_node) {
igb_erase_filter(adapter, rule);
hlist_del(&rule->nfc_node);
kfree(rule);
}
spin_unlock(&adapter->nfc_lock);
adapter->nfc_filter_count = 0;
}
igb: Support RX-ALL feature flag. This allows the NIC to receive all frames available, including those with bad FCS, un-matched vlans, ethernet control frames, and more. Tested by sending frames with bad FCS. Signed-off-by: Ben Greear <greearb@candelatech.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-03-06 09:41:58 +00:00
netdev->features = features;
if (netif_running(netdev))
igb_reinit_locked(adapter);
else
igb_reset(adapter);
igb: convert to ndo_fix_features Private rx_csum flags are now duplicate of netdev->features & NETIF_F_RXCSUM. Removing this needs deeper surgery. Things noticed: - HW VLAN acceleration probably can be toggled, but it's left as is - the resets on RX csum offload change can probably be avoided - there is A LOT of copy-and-pasted code here Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-06-08 08:38:01 +00:00
return 0;
}
igb: Enable use of "bridge fdb add" to set unicast table entries This change makes it so that we can use the bridge utility to add a FDB entry for the PF to an igb port. By doing this we can enable the VFs to talk to virtual ports residing on top of the PF. In addition this should also address issues with MACVLANs trying to reside on top of the PF as well as they would have had similar issues when added to the PF with SR-IOV enabled. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:34 +00:00
static int igb_ndo_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
struct net_device *dev,
const unsigned char *addr, u16 vid,
u16 flags)
{
/* guarantee we can provide a unique filter for the unicast address */
if (is_unicast_ether_addr(addr) || is_link_local_ether_addr(addr)) {
struct igb_adapter *adapter = netdev_priv(dev);
int vfn = adapter->vfs_allocated_count;
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
if (netdev_uc_count(dev) >= igb_available_rars(adapter, vfn))
igb: Enable use of "bridge fdb add" to set unicast table entries This change makes it so that we can use the bridge utility to add a FDB entry for the PF to an igb port. By doing this we can enable the VFs to talk to virtual ports residing on top of the PF. In addition this should also address issues with MACVLANs trying to reside on top of the PF as well as they would have had similar issues when added to the PF with SR-IOV enabled. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:34 +00:00
return -ENOMEM;
}
return ndo_dflt_fdb_add(ndm, tb, dev, addr, vid, flags);
}
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
#define IGB_MAX_MAC_HDR_LEN 127
#define IGB_MAX_NETWORK_HDR_LEN 511
static netdev_features_t
igb_features_check(struct sk_buff *skb, struct net_device *dev,
netdev_features_t features)
{
unsigned int network_hdr_len, mac_hdr_len;
/* Make certain the headers can be described by a context descriptor */
mac_hdr_len = skb_network_header(skb) - skb->data;
if (unlikely(mac_hdr_len > IGB_MAX_MAC_HDR_LEN))
return features & ~(NETIF_F_HW_CSUM |
NETIF_F_SCTP_CRC |
NETIF_F_HW_VLAN_CTAG_TX |
NETIF_F_TSO |
NETIF_F_TSO6);
network_hdr_len = skb_checksum_start(skb) - skb_network_header(skb);
if (unlikely(network_hdr_len > IGB_MAX_NETWORK_HDR_LEN))
return features & ~(NETIF_F_HW_CSUM |
NETIF_F_SCTP_CRC |
NETIF_F_TSO |
NETIF_F_TSO6);
/* We can only support IPV4 TSO in tunnels if we can mangle the
* inner IP ID field, so strip TSO if MANGLEID is not supported.
*/
if (skb->encapsulation && !(features & NETIF_F_TSO_MANGLEID))
features &= ~NETIF_F_TSO;
return features;
}
igb: Refactor igb_offload_cbs() Split code into a separate function (igb_offload_apply()) that will be used by ETF offload implementation. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:57 +00:00
static void igb_offload_apply(struct igb_adapter *adapter, s32 queue)
{
if (!is_fqtss_enabled(adapter)) {
enable_fqtss(adapter, true);
return;
}
igb_config_tx_modes(adapter, queue);
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
if (!is_any_cbs_enabled(adapter) && !is_any_txtime_enabled(adapter))
igb: Refactor igb_offload_cbs() Split code into a separate function (igb_offload_apply()) that will be used by ETF offload implementation. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:57 +00:00
enable_fqtss(adapter, false);
}
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
static int igb_offload_cbs(struct igb_adapter *adapter,
struct tc_cbs_qopt_offload *qopt)
{
struct e1000_hw *hw = &adapter->hw;
int err;
/* CBS offloading is only supported by i210 controller. */
if (hw->mac.type != e1000_i210)
return -EOPNOTSUPP;
/* CBS offloading is only supported by queue 0 and queue 1. */
if (qopt->queue < 0 || qopt->queue > 1)
return -EINVAL;
err = igb_save_cbs_params(adapter, qopt->queue, qopt->enable,
qopt->idleslope, qopt->sendslope,
qopt->hicredit, qopt->locredit);
if (err)
return err;
igb: Refactor igb_offload_cbs() Split code into a separate function (igb_offload_apply()) that will be used by ETF offload implementation. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:57 +00:00
igb_offload_apply(adapter, qopt->queue);
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
return 0;
}
igb: Add support for adding offloaded clsflower filters This allows filters added by tc-flower and specifying MAC addresses, Ethernet types, and the VLAN priority field, to be offloaded to the controller. This reuses most of the infrastructure used by ethtool, but clsflower filters are kept in a separated list, so they are invisible to ethtool. To setup clsflower offloading: $ tc qdisc replace dev eth0 handle 100: parent root mqprio \ num_tc 3 map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 \ queues 1@0 1@1 2@2 hw 0 (clsflower offloading depends on the netword driver to be configured with multiple traffic classes, we use mqprio's 'num_tc' parameter to set it to 3) $ tc qdisc add dev eth0 ingress Examples of filters: $ tc filter add dev eth0 parent ffff: flower \ dst_mac aa:aa:aa:aa:aa:aa \ hw_tc 2 skip_sw (just a simple filter filtering for the destination MAC address and steering that traffic to queue 2) $ tc filter add dev enp2s0 parent ffff: proto 0x22f0 flower \ src_mac cc:cc:cc:cc:cc:cc \ hw_tc 1 skip_sw (as the i210 doesn't support steering traffic based on the source address alone, we need to use another steering traffic, in this case we are using the ethernet type (0x22f0) to steer traffic to queue 1) Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:59 +00:00
#define ETHER_TYPE_FULL_MASK ((__force __be16)~0)
#define VLAN_PRIO_FULL_MASK (0x07)
static int igb_parse_cls_flower(struct igb_adapter *adapter,
struct tc_cls_flower_offload *f,
int traffic_class,
struct igb_nfc_filter *input)
{
struct netlink_ext_ack *extack = f->common.extack;
if (f->dissector->used_keys &
~(BIT(FLOW_DISSECTOR_KEY_BASIC) |
BIT(FLOW_DISSECTOR_KEY_CONTROL) |
BIT(FLOW_DISSECTOR_KEY_ETH_ADDRS) |
BIT(FLOW_DISSECTOR_KEY_VLAN))) {
NL_SET_ERR_MSG_MOD(extack,
"Unsupported key used, only BASIC, CONTROL, ETH_ADDRS and VLAN are supported");
return -EOPNOTSUPP;
}
if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_ETH_ADDRS)) {
struct flow_dissector_key_eth_addrs *key, *mask;
key = skb_flow_dissector_target(f->dissector,
FLOW_DISSECTOR_KEY_ETH_ADDRS,
f->key);
mask = skb_flow_dissector_target(f->dissector,
FLOW_DISSECTOR_KEY_ETH_ADDRS,
f->mask);
if (!is_zero_ether_addr(mask->dst)) {
if (!is_broadcast_ether_addr(mask->dst)) {
NL_SET_ERR_MSG_MOD(extack, "Only full masks are supported for destination MAC address");
return -EINVAL;
}
input->filter.match_flags |=
IGB_FILTER_FLAG_DST_MAC_ADDR;
ether_addr_copy(input->filter.dst_addr, key->dst);
}
if (!is_zero_ether_addr(mask->src)) {
if (!is_broadcast_ether_addr(mask->src)) {
NL_SET_ERR_MSG_MOD(extack, "Only full masks are supported for source MAC address");
return -EINVAL;
}
input->filter.match_flags |=
IGB_FILTER_FLAG_SRC_MAC_ADDR;
ether_addr_copy(input->filter.src_addr, key->src);
}
}
if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_BASIC)) {
struct flow_dissector_key_basic *key, *mask;
key = skb_flow_dissector_target(f->dissector,
FLOW_DISSECTOR_KEY_BASIC,
f->key);
mask = skb_flow_dissector_target(f->dissector,
FLOW_DISSECTOR_KEY_BASIC,
f->mask);
if (mask->n_proto) {
if (mask->n_proto != ETHER_TYPE_FULL_MASK) {
NL_SET_ERR_MSG_MOD(extack, "Only full mask is supported for EtherType filter");
return -EINVAL;
}
input->filter.match_flags |= IGB_FILTER_FLAG_ETHER_TYPE;
input->filter.etype = key->n_proto;
}
}
if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_VLAN)) {
struct flow_dissector_key_vlan *key, *mask;
key = skb_flow_dissector_target(f->dissector,
FLOW_DISSECTOR_KEY_VLAN,
f->key);
mask = skb_flow_dissector_target(f->dissector,
FLOW_DISSECTOR_KEY_VLAN,
f->mask);
if (mask->vlan_priority) {
if (mask->vlan_priority != VLAN_PRIO_FULL_MASK) {
NL_SET_ERR_MSG_MOD(extack, "Only full mask is supported for VLAN priority");
return -EINVAL;
}
input->filter.match_flags |= IGB_FILTER_FLAG_VLAN_TCI;
input->filter.vlan_tci = key->vlan_priority;
}
}
input->action = traffic_class;
input->cookie = f->cookie;
return 0;
}
igb: Add the skeletons for tc-flower offloading This adds basic functions needed to implement offloading for filters created by tc-flower. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:58 +00:00
static int igb_configure_clsflower(struct igb_adapter *adapter,
struct tc_cls_flower_offload *cls_flower)
{
igb: Add support for adding offloaded clsflower filters This allows filters added by tc-flower and specifying MAC addresses, Ethernet types, and the VLAN priority field, to be offloaded to the controller. This reuses most of the infrastructure used by ethtool, but clsflower filters are kept in a separated list, so they are invisible to ethtool. To setup clsflower offloading: $ tc qdisc replace dev eth0 handle 100: parent root mqprio \ num_tc 3 map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 \ queues 1@0 1@1 2@2 hw 0 (clsflower offloading depends on the netword driver to be configured with multiple traffic classes, we use mqprio's 'num_tc' parameter to set it to 3) $ tc qdisc add dev eth0 ingress Examples of filters: $ tc filter add dev eth0 parent ffff: flower \ dst_mac aa:aa:aa:aa:aa:aa \ hw_tc 2 skip_sw (just a simple filter filtering for the destination MAC address and steering that traffic to queue 2) $ tc filter add dev enp2s0 parent ffff: proto 0x22f0 flower \ src_mac cc:cc:cc:cc:cc:cc \ hw_tc 1 skip_sw (as the i210 doesn't support steering traffic based on the source address alone, we need to use another steering traffic, in this case we are using the ethernet type (0x22f0) to steer traffic to queue 1) Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:59 +00:00
struct netlink_ext_ack *extack = cls_flower->common.extack;
struct igb_nfc_filter *filter, *f;
int err, tc;
tc = tc_classid_to_hwtc(adapter->netdev, cls_flower->classid);
if (tc < 0) {
NL_SET_ERR_MSG_MOD(extack, "Invalid traffic class");
return -EINVAL;
}
filter = kzalloc(sizeof(*filter), GFP_KERNEL);
if (!filter)
return -ENOMEM;
err = igb_parse_cls_flower(adapter, cls_flower, tc, filter);
if (err < 0)
goto err_parse;
spin_lock(&adapter->nfc_lock);
hlist_for_each_entry(f, &adapter->nfc_filter_list, nfc_node) {
if (!memcmp(&f->filter, &filter->filter, sizeof(f->filter))) {
err = -EEXIST;
NL_SET_ERR_MSG_MOD(extack,
"This filter is already set in ethtool");
goto err_locked;
}
}
hlist_for_each_entry(f, &adapter->cls_flower_list, nfc_node) {
if (!memcmp(&f->filter, &filter->filter, sizeof(f->filter))) {
err = -EEXIST;
NL_SET_ERR_MSG_MOD(extack,
"This filter is already set in cls_flower");
goto err_locked;
}
}
err = igb_add_filter(adapter, filter);
if (err < 0) {
NL_SET_ERR_MSG_MOD(extack, "Could not add filter to the adapter");
goto err_locked;
}
hlist_add_head(&filter->nfc_node, &adapter->cls_flower_list);
spin_unlock(&adapter->nfc_lock);
return 0;
err_locked:
spin_unlock(&adapter->nfc_lock);
err_parse:
kfree(filter);
return err;
igb: Add the skeletons for tc-flower offloading This adds basic functions needed to implement offloading for filters created by tc-flower. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:58 +00:00
}
static int igb_delete_clsflower(struct igb_adapter *adapter,
struct tc_cls_flower_offload *cls_flower)
{
igb: Add support for adding offloaded clsflower filters This allows filters added by tc-flower and specifying MAC addresses, Ethernet types, and the VLAN priority field, to be offloaded to the controller. This reuses most of the infrastructure used by ethtool, but clsflower filters are kept in a separated list, so they are invisible to ethtool. To setup clsflower offloading: $ tc qdisc replace dev eth0 handle 100: parent root mqprio \ num_tc 3 map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 \ queues 1@0 1@1 2@2 hw 0 (clsflower offloading depends on the netword driver to be configured with multiple traffic classes, we use mqprio's 'num_tc' parameter to set it to 3) $ tc qdisc add dev eth0 ingress Examples of filters: $ tc filter add dev eth0 parent ffff: flower \ dst_mac aa:aa:aa:aa:aa:aa \ hw_tc 2 skip_sw (just a simple filter filtering for the destination MAC address and steering that traffic to queue 2) $ tc filter add dev enp2s0 parent ffff: proto 0x22f0 flower \ src_mac cc:cc:cc:cc:cc:cc \ hw_tc 1 skip_sw (as the i210 doesn't support steering traffic based on the source address alone, we need to use another steering traffic, in this case we are using the ethernet type (0x22f0) to steer traffic to queue 1) Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:59 +00:00
struct igb_nfc_filter *filter;
int err;
spin_lock(&adapter->nfc_lock);
hlist_for_each_entry(filter, &adapter->cls_flower_list, nfc_node)
if (filter->cookie == cls_flower->cookie)
break;
if (!filter) {
err = -ENOENT;
goto out;
}
err = igb_erase_filter(adapter, filter);
if (err < 0)
goto out;
hlist_del(&filter->nfc_node);
kfree(filter);
out:
spin_unlock(&adapter->nfc_lock);
return err;
igb: Add the skeletons for tc-flower offloading This adds basic functions needed to implement offloading for filters created by tc-flower. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:58 +00:00
}
static int igb_setup_tc_cls_flower(struct igb_adapter *adapter,
struct tc_cls_flower_offload *cls_flower)
{
switch (cls_flower->command) {
case TC_CLSFLOWER_REPLACE:
return igb_configure_clsflower(adapter, cls_flower);
case TC_CLSFLOWER_DESTROY:
return igb_delete_clsflower(adapter, cls_flower);
case TC_CLSFLOWER_STATS:
return -EOPNOTSUPP;
default:
cls_flower: fix error values for commands not supported by drivers -EOPNOTSUPP is the error value that should be reported if a flower command is not supported by a driver. Fix it in couple of Intel drivers. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-06-24 08:38:39 +00:00
return -EOPNOTSUPP;
igb: Add the skeletons for tc-flower offloading This adds basic functions needed to implement offloading for filters created by tc-flower. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:58 +00:00
}
}
static int igb_setup_tc_block_cb(enum tc_setup_type type, void *type_data,
void *cb_priv)
{
struct igb_adapter *adapter = cb_priv;
if (!tc_cls_can_offload_and_chain0(adapter->netdev, type_data))
return -EOPNOTSUPP;
switch (type) {
case TC_SETUP_CLSFLOWER:
return igb_setup_tc_cls_flower(adapter, type_data);
default:
return -EOPNOTSUPP;
}
}
static int igb_setup_tc_block(struct igb_adapter *adapter,
struct tc_block_offload *f)
{
if (f->binder_type != TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS)
return -EOPNOTSUPP;
switch (f->command) {
case TC_BLOCK_BIND:
return tcf_block_cb_register(f->block, igb_setup_tc_block_cb,
net: sched: pass extack pointer to block binds and cb registration Pass the extact struct from a tc qdisc add to the block bind function and, in turn, to the setup_tc ndo of binding device via the tc_block_offload struct. Pass this back to any block callback registrations to allow netlink logging of fails in the bind process. Signed-off-by: John Hurley <john.hurley@netronome.com> Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-06-25 21:30:04 +00:00
adapter, adapter, f->extack);
igb: Add the skeletons for tc-flower offloading This adds basic functions needed to implement offloading for filters created by tc-flower. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:58 +00:00
case TC_BLOCK_UNBIND:
tcf_block_cb_unregister(f->block, igb_setup_tc_block_cb,
adapter);
return 0;
default:
return -EOPNOTSUPP;
}
}
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
static int igb_offload_txtime(struct igb_adapter *adapter,
struct tc_etf_qopt_offload *qopt)
{
struct e1000_hw *hw = &adapter->hw;
int err;
/* Launchtime offloading is only supported by i210 controller. */
if (hw->mac.type != e1000_i210)
return -EOPNOTSUPP;
/* Launchtime offloading is only supported by queues 0 and 1. */
if (qopt->queue < 0 || qopt->queue > 1)
return -EINVAL;
err = igb_save_txtime_params(adapter, qopt->queue, qopt->enable);
if (err)
return err;
igb_offload_apply(adapter, qopt->queue);
return 0;
}
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
static int igb_setup_tc(struct net_device *dev, enum tc_setup_type type,
void *type_data)
{
struct igb_adapter *adapter = netdev_priv(dev);
switch (type) {
net_sch: cbs: Change TC_SETUP_CBS to TC_SETUP_QDISC_CBS Change TC_SETUP_CBS to TC_SETUP_QDISC_CBS to match the new convention.. Signed-off-by: Nogah Frankel <nogahf@mellanox.com> Signed-off-by: Jiri Pirko <jiri@mellanox.com> Reviewed-by: Simon Horman <simon.horman@netronome.com> Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-06 06:23:43 +00:00
case TC_SETUP_QDISC_CBS:
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
return igb_offload_cbs(adapter, type_data);
igb: Add the skeletons for tc-flower offloading This adds basic functions needed to implement offloading for filters created by tc-flower. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:58 +00:00
case TC_SETUP_BLOCK:
return igb_setup_tc_block(adapter, type_data);
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
case TC_SETUP_QDISC_ETF:
return igb_offload_txtime(adapter, type_data);
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
default:
return -EOPNOTSUPP;
}
}
igb: convert to net_device_ops Convert to new network device ops interface. Compile tested only. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 06:20:44 +00:00
static const struct net_device_ops igb_netdev_ops = {
igb: cleanup whitespace issues in igb_main.c This patch goes through and cleans up whitespace issues in igb_main.c to help improve readability. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:52:50 +00:00
.ndo_open = igb_open,
igb: convert to net_device_ops Convert to new network device ops interface. Compile tested only. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 06:20:44 +00:00
.ndo_stop = igb_close,
igb: drop the "adv" off function names relating to descriptors Many of the function names in the hot path are carrying an extra "_adv" suffix on the end of them to represent the fact that they are using advanced descriptors instead of legacy descriptors. However since all igb uses are advanced descriptors adding the extra suffix doesn't really add any additional data. Since this is the case it is easiest to just drop the suffix and save us from having to store the extra characters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:59 +00:00
.ndo_start_xmit = igb_xmit_frame,
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
.ndo_get_stats64 = igb_get_stats64,
igb: add support for set_rx_mode netdevice operation This patch adds support for the set_rx_mode netdevice operation so that igb can better support multiple unicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:48:56 +00:00
.ndo_set_rx_mode = igb_set_rx_mode,
igb: convert to net_device_ops Convert to new network device ops interface. Compile tested only. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 06:20:44 +00:00
.ndo_set_mac_address = igb_set_mac,
.ndo_change_mtu = igb_change_mtu,
.ndo_do_ioctl = igb_ioctl,
.ndo_tx_timeout = igb_tx_timeout,
.ndo_validate_addr = eth_validate_addr,
.ndo_vlan_rx_add_vid = igb_vlan_rx_add_vid,
.ndo_vlan_rx_kill_vid = igb_vlan_rx_kill_vid,
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
.ndo_set_vf_mac = igb_ndo_set_vf_mac,
.ndo_set_vf_vlan = igb_ndo_set_vf_vlan,
net-next:v4: Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool. o min_tx_rate puts lower limit on the VF bandwidth. VF is guaranteed to have a bandwidth of at least this value. max_tx_rate puts cap on the VF bandwidth. VF can have a bandwidth of up to this value. o A new handler set_vf_rate for attr IFLA_VF_RATE has been introduced which takes 4 arguments: netdev, VF number, min_tx_rate, max_tx_rate o ndo_set_vf_rate replaces ndo_set_vf_tx_rate handler. o Drivers that currently implement ndo_set_vf_tx_rate should now call ndo_set_vf_rate instead and reject attempt to set a minimum bandwidth greater than 0 for IFLA_VF_TX_RATE when IFLA_VF_RATE is not yet implemented by driver. o If user enters only one of either min_tx_rate or max_tx_rate, then, userland should read back the other value from driver and set both for IFLA_VF_RATE. Drivers that have not yet implemented IFLA_VF_RATE should always return min_tx_rate as 0 when read from ip tool. o If both IFLA_VF_TX_RATE and IFLA_VF_RATE options are specified, then IFLA_VF_RATE should override. o Idea is to have consistent display of rate values to user. o Usage example: - ./ip link set p4p1 vf 0 rate 900 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 900 (Mbps), max_tx_rate 900Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 300 min_tx_rate 200 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 300 (Mbps), max_tx_rate 300Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 600 rate 300 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5, tx rate 600 (Mbps), max_tx_rate 600Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 13:59:05 +00:00
.ndo_set_vf_rate = igb_ndo_set_vf_bw,
igb: add support for spoofchk config Add support for spoofchk configuration per VF via iproute2 tool. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-03 20:27:48 +00:00
.ndo_set_vf_spoofchk = igb_ndo_set_vf_spoofchk,
igb: add VF trust infrastructure * Add a per-VF value to know if a VF is trusted, by default don't trust VFs. * Implement netdev op to trust VFs (igb_ndo_set_vf_trust) and add trust status to ndo_get_vf_config output. * Allow a trusted VF to change MAC and MAC filters even if MAC has been administratively set. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-01-17 10:53:39 +00:00
.ndo_set_vf_trust = igb_ndo_set_vf_trust,
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
.ndo_get_vf_config = igb_ndo_get_vf_config,
igb: do vlan cleanup - unify vlan and nonvlan rx path - kill adapter->vlgrp and igb_vlan_rx_register - allow to turn on/off rx/tx vlan accel via ethtool (set_features) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-21 03:27:27 +00:00
.ndo_fix_features = igb_fix_features,
.ndo_set_features = igb_set_features,
igb: Enable use of "bridge fdb add" to set unicast table entries This change makes it so that we can use the bridge utility to add a FDB entry for the PF to an igb port. By doing this we can enable the VFs to talk to virtual ports residing on top of the PF. In addition this should also address issues with MACVLANs trying to reside on top of the PF as well as they would have had similar issues when added to the PF with SR-IOV enabled. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:34 +00:00
.ndo_fdb_add = igb_ndo_fdb_add,
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
.ndo_features_check = igb_features_check,
igb: Add support for CBS offload This patch adds support for Credit-Based Shaper (CBS) qdisc offload from Traffic Control system. This support enable us to leverage the Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav standard which was merged into 802.1Q in 2014. It enables traffic prioritization and bandwidth reservation via the Credit-Based Shaper which is implemented in hardware by i210 controller. The patch introduces the igb_setup_tc() function which implements the support for CBS qdisc hardware offload in the IGB driver. CBS offload is the only traffic control offload supported by the driver at the moment. FQTSS transmission mode from i210 controller is automatically enabled by the IGB driver when the CBS is enabled for the first hardware queue. Likewise, FQTSS mode is automatically disabled when CBS is disabled for the last hardware queue. Changing FQTSS mode requires NIC reset. FQTSS feature is supported by i210 controller only. Signed-off-by: Andre Guedes <andre.guedes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Henrik Austad <henrik@austad.us> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-17 01:01:28 +00:00
.ndo_setup_tc = igb_setup_tc,
igb: convert to net_device_ops Convert to new network device ops interface. Compile tested only. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 06:20:44 +00:00
};
igb: Update firmware info output Our NVM image creation tools have evolved over the years and there are multiple versions contained in them, depending on the tool used to create them. This patch outputs the NVM versions available in ethtool -i output. rc2: (not sure why others show in log but not in the message) Added additional call to igb_set_fw_version per Community feedback. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-06-14 16:04:19 +00:00
/**
* igb_set_fw_version - Configure version string for ethtool
* @adapter: adapter struct
**/
void igb_set_fw_version(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
igb: Update firmware version info for ethtool output. There are multiple places in our device nvm where the version is stored. This update fixes some output errors with some types of images and refactors the way the version data is gathered and stored for ethtool output. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-10-18 07:16:19 +00:00
struct e1000_fw_version fw;
igb_get_fw_version(hw, &fw);
switch (hw->mac.type) {
igb: Fix get_fw_version function for all parts This patch fixes issues found with older parts and older NVM tools in the display of the version in ethtool. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-16 19:25:33 +00:00
case e1000_i210:
igb: Update firmware version info for ethtool output. There are multiple places in our device nvm where the version is stored. This update fixes some output errors with some types of images and refactors the way the version data is gathered and stored for ethtool output. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-10-18 07:16:19 +00:00
case e1000_i211:
igb: Fix get_fw_version function for all parts This patch fixes issues found with older parts and older NVM tools in the display of the version in ethtool. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-16 19:25:33 +00:00
if (!(igb_get_flash_presence_i210(hw))) {
snprintf(adapter->fw_version,
sizeof(adapter->fw_version),
"%2d.%2d-%d",
fw.invm_major, fw.invm_minor,
fw.invm_img_type);
break;
}
/* fall through */
igb: Update firmware version info for ethtool output. There are multiple places in our device nvm where the version is stored. This update fixes some output errors with some types of images and refactors the way the version data is gathered and stored for ethtool output. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-10-18 07:16:19 +00:00
default:
/* if option is rom valid, display its version too */
if (fw.or_valid) {
snprintf(adapter->fw_version,
sizeof(adapter->fw_version),
"%d.%d, 0x%08x, %d.%d.%d",
fw.eep_major, fw.eep_minor, fw.etrack_id,
fw.or_major, fw.or_build, fw.or_patch);
/* no option rom */
igb: Fix get_fw_version function for all parts This patch fixes issues found with older parts and older NVM tools in the display of the version in ethtool. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-16 19:25:33 +00:00
} else if (fw.etrack_id != 0X0000) {
igb: Update firmware version info for ethtool output. There are multiple places in our device nvm where the version is stored. This update fixes some output errors with some types of images and refactors the way the version data is gathered and stored for ethtool output. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-10-18 07:16:19 +00:00
snprintf(adapter->fw_version,
igb: Fix get_fw_version function for all parts This patch fixes issues found with older parts and older NVM tools in the display of the version in ethtool. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-16 19:25:33 +00:00
sizeof(adapter->fw_version),
"%d.%d, 0x%08x",
fw.eep_major, fw.eep_minor, fw.etrack_id);
} else {
snprintf(adapter->fw_version,
sizeof(adapter->fw_version),
"%d.%d.%d",
fw.eep_major, fw.eep_minor, fw.eep_build);
igb: Update firmware version info for ethtool output. There are multiple places in our device nvm where the version is stored. This update fixes some output errors with some types of images and refactors the way the version data is gathered and stored for ethtool output. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-10-18 07:16:19 +00:00
}
break;
igb: Update firmware info output Our NVM image creation tools have evolved over the years and there are multiple versions contained in them, depending on the tool used to create them. This patch outputs the NVM versions available in ethtool -i output. rc2: (not sure why others show in log but not in the message) Added additional call to igb_set_fw_version per Community feedback. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-06-14 16:04:19 +00:00
}
}
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
/**
* igb_init_mas - init Media Autosense feature if enabled in the NVM
*
* @adapter: adapter struct
**/
static void igb_init_mas(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
u16 eeprom_data;
hw->nvm.ops.read(hw, NVM_COMPAT, 1, &eeprom_data);
switch (hw->bus.func) {
case E1000_FUNC_0:
if (eeprom_data & IGB_MAS_ENABLE_0) {
adapter->flags |= IGB_FLAG_MAS_ENABLE;
netdev_info(adapter->netdev,
"MAS: Enabling Media Autosense for port %d\n",
hw->bus.func);
}
break;
case E1000_FUNC_1:
if (eeprom_data & IGB_MAS_ENABLE_1) {
adapter->flags |= IGB_FLAG_MAS_ENABLE;
netdev_info(adapter->netdev,
"MAS: Enabling Media Autosense for port %d\n",
hw->bus.func);
}
break;
case E1000_FUNC_2:
if (eeprom_data & IGB_MAS_ENABLE_2) {
adapter->flags |= IGB_FLAG_MAS_ENABLE;
netdev_info(adapter->netdev,
"MAS: Enabling Media Autosense for port %d\n",
hw->bus.func);
}
break;
case E1000_FUNC_3:
if (eeprom_data & IGB_MAS_ENABLE_3) {
adapter->flags |= IGB_FLAG_MAS_ENABLE;
netdev_info(adapter->netdev,
"MAS: Enabling Media Autosense for port %d\n",
hw->bus.func);
}
break;
default:
/* Shouldn't get here */
netdev_err(adapter->netdev,
"MAS: Invalid port configuration, returning\n");
break;
}
}
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/**
* igb_init_i2c - Init I2C interface
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
* @adapter: pointer to adapter structure
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
**/
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
static s32 igb_init_i2c(struct igb_adapter *adapter)
{
igb: unhide invariant returns Return a 0 directly rather than a constant. Reported-by: Peter Senna Tschudin <peter.senna@gmail.com> Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-06-04 07:12:15 +00:00
s32 status = 0;
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
/* I2C interface supported on i350 devices */
if (adapter->hw.mac.type != e1000_i350)
igb: unhide invariant returns Return a 0 directly rather than a constant. Reported-by: Peter Senna Tschudin <peter.senna@gmail.com> Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-06-04 07:12:15 +00:00
return 0;
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
/* Initialize the i2c bus which is controlled by the registers.
* This bus will use the i2c_algo_bit structue that implements
* the protocol through toggling of the 4 bits in the register.
*/
adapter->i2c_adap.owner = THIS_MODULE;
adapter->i2c_algo = igb_i2c_algo;
adapter->i2c_algo.data = adapter;
adapter->i2c_adap.algo_data = &adapter->i2c_algo;
adapter->i2c_adap.dev.parent = &adapter->pdev->dev;
strlcpy(adapter->i2c_adap.name, "igb BB",
sizeof(adapter->i2c_adap.name));
status = i2c_bit_add_bus(&adapter->i2c_adap);
return status;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_probe - Device Initialization Routine
* @pdev: PCI device information struct
* @ent: entry in igb_pci_tbl
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Returns 0 on success, negative on failure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_probe initializes an adapter identified by a pci_dev structure.
* The OS initialization, configuring of the adapter private structure,
* and a hardware reset occur.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
drivers/net: fix up function prototypes after __dev* removals The __dev* removal patches for the network drivers ended up messing up the function prototypes for a bunch of drivers. This patch fixes all of them back up to be properly aligned. Bonus is that this almost removes 100 lines of code, always a nice surprise. Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-12-06 14:30:56 +00:00
static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct net_device *netdev;
struct igb_adapter *adapter;
struct e1000_hw *hw;
igb: move global_quad_port_a from global into local static define This change moves global_quad_port_a into igb_probe as a static define since it doesn't actually need to be global. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:48:31 +00:00
u16 eeprom_data = 0;
igb: Add new function to read part number from EEPROM in string format New adapters will have part numbers stored in string format rather than simple hex format. This function will read part number formats in either hex or string. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-22 17:17:21 +00:00
s32 ret_val;
igb: move global_quad_port_a from global into local static define This change moves global_quad_port_a into igb_probe as a static define since it doesn't actually need to be global. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:48:31 +00:00
static int global_quad_port_a; /* global quad port a indication */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
const struct e1000_info *ei = igb_info_tbl[ent->driver_data];
Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 Conflicts: drivers/net/igb/igb_main.c drivers/net/qlge/qlge_main.c drivers/net/wireless/ath9k/ath9k.h drivers/net/wireless/ath9k/core.h drivers/net/wireless/ath9k/hw.c
2009-03-17 22:01:30 +00:00
int err, pci_using_dac;
igb: Add new function to read part number from EEPROM in string format New adapters will have part numbers stored in string format rather than simple hex format. This function will read part number formats in either hex or string. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-22 17:17:21 +00:00
u8 part_str[E1000_PBANUM_LENGTH];
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
ixgbe/igb: catch invalid VF settings Some ixgbe cards put an invalid VF device ID in the PCIe SR-IOV capability. The ixgbe driver is only valid for PFs or non SR-IOV hardware. It seems that the same problem could occur on igb hardware as well, so if we discover we are trying to initialize a VF in ixbge_probe or igb_probe, print an error and exit. Based on a patch for ixgbe from Chris Wright <chrisw@sous-sol.org>. Signed-off-by: Andy Gospodarek <andy@greyhouse.net> Cc: Chris Wright <chrisw@sous-sol.org> Acked-by: Chris Wright <chrisw@sous-sol.org> Acked-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-07-21 06:40:31 +00:00
/* Catch broken hardware that put the wrong VF device ID in
* the PCIe SR-IOV capability.
*/
if (pdev->is_virtfn) {
WARN(1, KERN_ERR "%s (%hx:%hx) should not be a VF!\n",
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
pci_name(pdev), pdev->vendor, pdev->device);
ixgbe/igb: catch invalid VF settings Some ixgbe cards put an invalid VF device ID in the PCIe SR-IOV capability. The ixgbe driver is only valid for PFs or non SR-IOV hardware. It seems that the same problem could occur on igb hardware as well, so if we discover we are trying to initialize a VF in ixbge_probe or igb_probe, print an error and exit. Based on a patch for ixgbe from Chris Wright <chrisw@sous-sol.org>. Signed-off-by: Andy Gospodarek <andy@greyhouse.net> Cc: Chris Wright <chrisw@sous-sol.org> Acked-by: Chris Wright <chrisw@sous-sol.org> Acked-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-07-21 06:40:31 +00:00
return -EINVAL;
}
igb: remove check for needing an io port Since igb supports only pci-e nics and there is no plan to support any legacy pci parts in the driver there isn't really much need for checking to see if an io port is needed. In the unlikely event that we do begin supporting legacy pci parts then we can see about adding this code back to the driver. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:16:04 +00:00
err = pci_enable_device_mem(pdev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (err)
return err;
pci_using_dac = 0;
DMA-API: net: intel/igb: fix 32-bit DMA mask handling The fallback to 32-bit DMA mask is rather odd: err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(64)); if (!err) { err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(64)); if (!err) pci_using_dac = 1; } else { err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { dev_err(&pdev->dev, "No usable DMA configuration, aborting\n"); goto err_dma; } } } This means we only set the coherent DMA mask in the fallback path if the DMA mask set failed, which is silly. This fixes it to set the coherent DMA mask only if dma_set_mask() succeeded, and to error out if either fails. Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2013-06-10 11:24:50 +00:00
err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (!err) {
DMA-API: net: intel/igb: fix 32-bit DMA mask handling The fallback to 32-bit DMA mask is rather odd: err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(64)); if (!err) { err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(64)); if (!err) pci_using_dac = 1; } else { err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { dev_err(&pdev->dev, "No usable DMA configuration, aborting\n"); goto err_dma; } } } This means we only set the coherent DMA mask in the fallback path if the DMA mask set failed, which is silly. This fixes it to set the coherent DMA mask only if dma_set_mask() succeeded, and to error out if either fails. Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2013-06-10 11:24:50 +00:00
pci_using_dac = 1;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
} else {
DMA-API: net: intel/igb: fix 32-bit DMA mask handling The fallback to 32-bit DMA mask is rather odd: err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(64)); if (!err) { err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(64)); if (!err) pci_using_dac = 1; } else { err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { dev_err(&pdev->dev, "No usable DMA configuration, aborting\n"); goto err_dma; } } } This means we only set the coherent DMA mask in the fallback path if the DMA mask set failed, which is silly. This fixes it to set the coherent DMA mask only if dma_set_mask() succeeded, and to error out if either fails. Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2013-06-10 11:24:50 +00:00
err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (err) {
DMA-API: net: intel/igb: fix 32-bit DMA mask handling The fallback to 32-bit DMA mask is rather odd: err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(64)); if (!err) { err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(64)); if (!err) pci_using_dac = 1; } else { err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { dev_err(&pdev->dev, "No usable DMA configuration, aborting\n"); goto err_dma; } } } This means we only set the coherent DMA mask in the fallback path if the DMA mask set failed, which is silly. This fixes it to set the coherent DMA mask only if dma_set_mask() succeeded, and to error out if either fails. Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2013-06-10 11:24:50 +00:00
dev_err(&pdev->dev,
"No usable DMA configuration, aborting\n");
goto err_dma;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
}
ethernet/intel: Use pci_(request|release)_mem_regions Now that we do have pci_request_mem_regions() and pci_release_mem_regions() at hand, use it in the Intel ethernet drivers. Suggested-by: Christoph Hellwig <hch@infradead.org> Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> CC: David S. Miller <davem@davemloft.net>
2016-06-07 07:44:05 +00:00
err = pci_request_mem_regions(pdev, igb_driver_name);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (err)
goto err_pci_reg;
e1000e/igb/ixgbe: Don't report an error if devices don't support AER The only error returned by pci_{en,dis}able_pcie_error_reporting() is -EIO which simply means that Advanced Error Reporting is not supported. There is no need to report that, so remove the error check from e1000e, igb and ixgbe. Signed-off-by: Frans Pop <elendil@planet.nl> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-02 17:04:12 +00:00
pci_enable_pcie_error_reporting(pdev);
igb: Add support for pci-e Advanced Error Reporting Add the calls necessary to enable advanced error reporting for igb on systems with AER enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-27 08:24:37 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
pci_set_master(pdev);
igb: save and restore pcie/msi state to support EEH recovery To enable EEH support for pci-express network adapters, pcie/msi state needs to be saved and restored for that adapter. [after similar patches for ixgbe and e1000e from Wendy Xiong] Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Cc: Wendy Xiong <wendyx@us.ibm.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-04-23 18:09:34 +00:00
pci_save_state(pdev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
err = -ENOMEM;
igb: add vfs_allocated_count as placeholder for number of vfs This is the first step in supporting sr-iov. The vf_allocated_count value will be 0 until we actually have vfs present. In the meantime it represents an offset value for the start of the queues. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:23 +00:00
netdev = alloc_etherdev_mq(sizeof(struct igb_adapter),
igb: Remove multi_tx_table and simplify igb_xmit_frame Instead of using the multi_tx_table to map possible Tx queues to Tx rings we can just do simple subtraction for the unlikely event that the Tx queue provided exceeds the number of Tx rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:10 +00:00
IGB_MAX_TX_QUEUES);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (!netdev)
goto err_alloc_etherdev;
SET_NETDEV_DEV(netdev, &pdev->dev);
pci_set_drvdata(pdev, netdev);
adapter = netdev_priv(netdev);
adapter->netdev = netdev;
adapter->pdev = pdev;
hw = &adapter->hw;
hw->back = adapter;
intel: make wired ethernet driver message level consistent (rev2) Dan Carpenter noticed that ixgbevf initial default was different than the rest. But the problem is broader than that, only one Intel driver (ixgb) was doing it almost right. The convention for default debug level should be consistent among Intel drivers and follow established convention. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-03-13 06:04:20 +00:00
adapter->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
err = -EIO;
igb: don't unmap NULL hw_addr I've got a startech thunderbolt dock someone loaned me, which among other things, has the following device in it: 08:00.0 Ethernet controller: Intel Corporation I210 Gigabit Network Connection (rev 03) This hotplugs just fine (kernel 4.2.0 plus a patch or two here): [ 863.020315] igb: Intel(R) Gigabit Ethernet Network Driver - version 5.2.18-k [ 863.020316] igb: Copyright (c) 2007-2014 Intel Corporation. [ 863.028657] igb 0000:08:00.0: enabling device (0000 -> 0002) [ 863.062089] igb 0000:08:00.0: added PHC on eth0 [ 863.062090] igb 0000:08:00.0: Intel(R) Gigabit Ethernet Network Connection [ 863.062091] igb 0000:08:00.0: eth0: (PCIe:2.5Gb/s:Width x1) e8:ea:6a:00:1b:2a [ 863.062194] igb 0000:08:00.0: eth0: PBA No: 000200-000 [ 863.062196] igb 0000:08:00.0: Using MSI-X interrupts. 4 rx queue(s), 4 tx queue(s) [ 863.064889] igb 0000:08:00.0 enp8s0: renamed from eth0 But disconnecting it is another story: [ 1002.807932] igb 0000:08:00.0: removed PHC on enp8s0 [ 1002.807944] igb 0000:08:00.0 enp8s0: PCIe link lost, device now detached [ 1003.341141] ------------[ cut here ]------------ [ 1003.341148] WARNING: CPU: 0 PID: 199 at lib/iomap.c:43 bad_io_access+0x38/0x40() [ 1003.341149] Bad IO access at port 0x0 () [ 1003.342767] Modules linked in: snd_usb_audio snd_usbmidi_lib snd_rawmidi igb dca firewire_ohci firewire_core crc_itu_t rfcomm ctr ccm arc4 iwlmvm mac80211 fuse xt_CHECKSUM ipt_MASQUERADE nf_nat_masquerade_ipv4 tun ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_security ip6table_raw ip6table_filter ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw iptable_filter bnep dm_mirror dm_region_hash dm_log dm_mod coretemp x86_pkg_temp_thermal intel_powerclamp kvm_intel snd_hda_codec_hdmi kvm crct10dif_pclmul crc32_pclmul ghash_clmulni_intel drbg [ 1003.342793] ansi_cprng aesni_intel hp_wmi aes_x86_64 iTCO_wdt lrw iTCO_vendor_support ppdev gf128mul sparse_keymap glue_helper ablk_helper cryptd snd_hda_codec_realtek snd_hda_codec_generic microcode snd_hda_intel uvcvideo iwlwifi snd_hda_codec videobuf2_vmalloc videobuf2_memops snd_hda_core videobuf2_core snd_hwdep btusb v4l2_common btrtl snd_seq btbcm btintel videodev cfg80211 snd_seq_device rtsx_pci_ms bluetooth pcspkr input_leds i2c_i801 media parport_pc memstick rfkill sg lpc_ich snd_pcm 8250_fintek parport joydev snd_timer snd soundcore hp_accel ie31200_edac mei_me lis3lv02d edac_core input_polldev mei hp_wireless shpchp tpm_infineon sch_fq_codel nfsd auth_rpcgss nfs_acl lockd grace sunrpc ip_tables autofs4 xfs libcrc32c sd_mod sr_mod cdrom rtsx_pci_sdmmc mmc_core crc32c_intel serio_raw rtsx_pci [ 1003.342822] nouveau ahci libahci mxm_wmi e1000e xhci_pci hwmon ptp drm_kms_helper pps_core xhci_hcd ttm wmi video ipv6 [ 1003.342839] CPU: 0 PID: 199 Comm: kworker/0:2 Not tainted 4.2.0-2.el7_UNSUPPORTED.x86_64 #1 [ 1003.342840] Hardware name: Hewlett-Packard HP ZBook 15 G2/2253, BIOS M70 Ver. 01.07 02/26/2015 [ 1003.342843] Workqueue: pciehp-3 pciehp_power_thread [ 1003.342844] ffffffff81a90655 ffff8804866d3b48 ffffffff8164763a 0000000000000000 [ 1003.342846] ffff8804866d3b98 ffff8804866d3b88 ffffffff8107134a ffff8804866d3b88 [ 1003.342847] ffff880486f46000 ffff88046c8a8000 ffff880486f46840 ffff88046c8a8098 [ 1003.342848] Call Trace: [ 1003.342852] [<ffffffff8164763a>] dump_stack+0x45/0x57 [ 1003.342855] [<ffffffff8107134a>] warn_slowpath_common+0x8a/0xc0 [ 1003.342857] [<ffffffff810713c6>] warn_slowpath_fmt+0x46/0x50 [ 1003.342859] [<ffffffff8133719e>] ? pci_disable_msix+0x3e/0x50 [ 1003.342860] [<ffffffff812f6328>] bad_io_access+0x38/0x40 [ 1003.342861] [<ffffffff812f6567>] pci_iounmap+0x27/0x40 [ 1003.342865] [<ffffffffa0b728d7>] igb_remove+0xc7/0x160 [igb] [ 1003.342867] [<ffffffff8132189f>] pci_device_remove+0x3f/0xc0 [ 1003.342869] [<ffffffff81433426>] __device_release_driver+0x96/0x130 [ 1003.342870] [<ffffffff814334e3>] device_release_driver+0x23/0x30 [ 1003.342871] [<ffffffff8131b404>] pci_stop_bus_device+0x94/0xa0 [ 1003.342872] [<ffffffff8131b3ad>] pci_stop_bus_device+0x3d/0xa0 [ 1003.342873] [<ffffffff8131b3ad>] pci_stop_bus_device+0x3d/0xa0 [ 1003.342874] [<ffffffff8131b516>] pci_stop_and_remove_bus_device+0x16/0x30 [ 1003.342876] [<ffffffff81333f5b>] pciehp_unconfigure_device+0x9b/0x180 [ 1003.342877] [<ffffffff81333a73>] pciehp_disable_slot+0x43/0xb0 [ 1003.342878] [<ffffffff81333b6d>] pciehp_power_thread+0x8d/0xb0 [ 1003.342885] [<ffffffff810881b2>] process_one_work+0x152/0x3d0 [ 1003.342886] [<ffffffff8108854a>] worker_thread+0x11a/0x460 [ 1003.342887] [<ffffffff81088430>] ? process_one_work+0x3d0/0x3d0 [ 1003.342890] [<ffffffff8108ddd9>] kthread+0xc9/0xe0 [ 1003.342891] [<ffffffff8108dd10>] ? kthread_create_on_node+0x180/0x180 [ 1003.342893] [<ffffffff8164e29f>] ret_from_fork+0x3f/0x70 [ 1003.342894] [<ffffffff8108dd10>] ? kthread_create_on_node+0x180/0x180 [ 1003.342895] ---[ end trace 65a77e06d5aa9358 ]--- Upon looking at the igb driver, I see that igb_rd32() attempted to read from hw_addr and failed, so it set hw->hw_addr to NULL and spit out the message in the log output above, "PCIe link lost, device now detached". Well, now that hw_addr is NULL, the attempt to call pci_iounmap is obviously not going to go well. As suggested by Mark Rustad, do something similar to what ixgbe does, and save a copy of hw_addr as adapter->io_addr, so we can still call pci_iounmap on it on teardown. Additionally, for consistency, make the pci_iomap call assignment directly to io_addr, so map and unmap match. Signed-off-by: Jarod Wilson <jarod@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-09-10 19:37:50 +00:00
adapter->io_addr = pci_iomap(pdev, 0, 0);
if (!adapter->io_addr)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
goto err_ioremap;
igb: don't unmap NULL hw_addr I've got a startech thunderbolt dock someone loaned me, which among other things, has the following device in it: 08:00.0 Ethernet controller: Intel Corporation I210 Gigabit Network Connection (rev 03) This hotplugs just fine (kernel 4.2.0 plus a patch or two here): [ 863.020315] igb: Intel(R) Gigabit Ethernet Network Driver - version 5.2.18-k [ 863.020316] igb: Copyright (c) 2007-2014 Intel Corporation. [ 863.028657] igb 0000:08:00.0: enabling device (0000 -> 0002) [ 863.062089] igb 0000:08:00.0: added PHC on eth0 [ 863.062090] igb 0000:08:00.0: Intel(R) Gigabit Ethernet Network Connection [ 863.062091] igb 0000:08:00.0: eth0: (PCIe:2.5Gb/s:Width x1) e8:ea:6a:00:1b:2a [ 863.062194] igb 0000:08:00.0: eth0: PBA No: 000200-000 [ 863.062196] igb 0000:08:00.0: Using MSI-X interrupts. 4 rx queue(s), 4 tx queue(s) [ 863.064889] igb 0000:08:00.0 enp8s0: renamed from eth0 But disconnecting it is another story: [ 1002.807932] igb 0000:08:00.0: removed PHC on enp8s0 [ 1002.807944] igb 0000:08:00.0 enp8s0: PCIe link lost, device now detached [ 1003.341141] ------------[ cut here ]------------ [ 1003.341148] WARNING: CPU: 0 PID: 199 at lib/iomap.c:43 bad_io_access+0x38/0x40() [ 1003.341149] Bad IO access at port 0x0 () [ 1003.342767] Modules linked in: snd_usb_audio snd_usbmidi_lib snd_rawmidi igb dca firewire_ohci firewire_core crc_itu_t rfcomm ctr ccm arc4 iwlmvm mac80211 fuse xt_CHECKSUM ipt_MASQUERADE nf_nat_masquerade_ipv4 tun ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_security ip6table_raw ip6table_filter ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw iptable_filter bnep dm_mirror dm_region_hash dm_log dm_mod coretemp x86_pkg_temp_thermal intel_powerclamp kvm_intel snd_hda_codec_hdmi kvm crct10dif_pclmul crc32_pclmul ghash_clmulni_intel drbg [ 1003.342793] ansi_cprng aesni_intel hp_wmi aes_x86_64 iTCO_wdt lrw iTCO_vendor_support ppdev gf128mul sparse_keymap glue_helper ablk_helper cryptd snd_hda_codec_realtek snd_hda_codec_generic microcode snd_hda_intel uvcvideo iwlwifi snd_hda_codec videobuf2_vmalloc videobuf2_memops snd_hda_core videobuf2_core snd_hwdep btusb v4l2_common btrtl snd_seq btbcm btintel videodev cfg80211 snd_seq_device rtsx_pci_ms bluetooth pcspkr input_leds i2c_i801 media parport_pc memstick rfkill sg lpc_ich snd_pcm 8250_fintek parport joydev snd_timer snd soundcore hp_accel ie31200_edac mei_me lis3lv02d edac_core input_polldev mei hp_wireless shpchp tpm_infineon sch_fq_codel nfsd auth_rpcgss nfs_acl lockd grace sunrpc ip_tables autofs4 xfs libcrc32c sd_mod sr_mod cdrom rtsx_pci_sdmmc mmc_core crc32c_intel serio_raw rtsx_pci [ 1003.342822] nouveau ahci libahci mxm_wmi e1000e xhci_pci hwmon ptp drm_kms_helper pps_core xhci_hcd ttm wmi video ipv6 [ 1003.342839] CPU: 0 PID: 199 Comm: kworker/0:2 Not tainted 4.2.0-2.el7_UNSUPPORTED.x86_64 #1 [ 1003.342840] Hardware name: Hewlett-Packard HP ZBook 15 G2/2253, BIOS M70 Ver. 01.07 02/26/2015 [ 1003.342843] Workqueue: pciehp-3 pciehp_power_thread [ 1003.342844] ffffffff81a90655 ffff8804866d3b48 ffffffff8164763a 0000000000000000 [ 1003.342846] ffff8804866d3b98 ffff8804866d3b88 ffffffff8107134a ffff8804866d3b88 [ 1003.342847] ffff880486f46000 ffff88046c8a8000 ffff880486f46840 ffff88046c8a8098 [ 1003.342848] Call Trace: [ 1003.342852] [<ffffffff8164763a>] dump_stack+0x45/0x57 [ 1003.342855] [<ffffffff8107134a>] warn_slowpath_common+0x8a/0xc0 [ 1003.342857] [<ffffffff810713c6>] warn_slowpath_fmt+0x46/0x50 [ 1003.342859] [<ffffffff8133719e>] ? pci_disable_msix+0x3e/0x50 [ 1003.342860] [<ffffffff812f6328>] bad_io_access+0x38/0x40 [ 1003.342861] [<ffffffff812f6567>] pci_iounmap+0x27/0x40 [ 1003.342865] [<ffffffffa0b728d7>] igb_remove+0xc7/0x160 [igb] [ 1003.342867] [<ffffffff8132189f>] pci_device_remove+0x3f/0xc0 [ 1003.342869] [<ffffffff81433426>] __device_release_driver+0x96/0x130 [ 1003.342870] [<ffffffff814334e3>] device_release_driver+0x23/0x30 [ 1003.342871] [<ffffffff8131b404>] pci_stop_bus_device+0x94/0xa0 [ 1003.342872] [<ffffffff8131b3ad>] pci_stop_bus_device+0x3d/0xa0 [ 1003.342873] [<ffffffff8131b3ad>] pci_stop_bus_device+0x3d/0xa0 [ 1003.342874] [<ffffffff8131b516>] pci_stop_and_remove_bus_device+0x16/0x30 [ 1003.342876] [<ffffffff81333f5b>] pciehp_unconfigure_device+0x9b/0x180 [ 1003.342877] [<ffffffff81333a73>] pciehp_disable_slot+0x43/0xb0 [ 1003.342878] [<ffffffff81333b6d>] pciehp_power_thread+0x8d/0xb0 [ 1003.342885] [<ffffffff810881b2>] process_one_work+0x152/0x3d0 [ 1003.342886] [<ffffffff8108854a>] worker_thread+0x11a/0x460 [ 1003.342887] [<ffffffff81088430>] ? process_one_work+0x3d0/0x3d0 [ 1003.342890] [<ffffffff8108ddd9>] kthread+0xc9/0xe0 [ 1003.342891] [<ffffffff8108dd10>] ? kthread_create_on_node+0x180/0x180 [ 1003.342893] [<ffffffff8164e29f>] ret_from_fork+0x3f/0x70 [ 1003.342894] [<ffffffff8108dd10>] ? kthread_create_on_node+0x180/0x180 [ 1003.342895] ---[ end trace 65a77e06d5aa9358 ]--- Upon looking at the igb driver, I see that igb_rd32() attempted to read from hw_addr and failed, so it set hw->hw_addr to NULL and spit out the message in the log output above, "PCIe link lost, device now detached". Well, now that hw_addr is NULL, the attempt to call pci_iounmap is obviously not going to go well. As suggested by Mark Rustad, do something similar to what ixgbe does, and save a copy of hw_addr as adapter->io_addr, so we can still call pci_iounmap on it on teardown. Additionally, for consistency, make the pci_iomap call assignment directly to io_addr, so map and unmap match. Signed-off-by: Jarod Wilson <jarod@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-09-10 19:37:50 +00:00
/* hw->hw_addr can be altered, we'll use adapter->io_addr for unmap */
hw->hw_addr = adapter->io_addr;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: convert to net_device_ops Convert to new network device ops interface. Compile tested only. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 06:20:44 +00:00
netdev->netdev_ops = &igb_netdev_ops;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_set_ethtool_ops(netdev);
netdev->watchdog_timeo = 5 * HZ;
strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1);
igb: Support ports mapped in 64-bit PCI space This patch resolves an issue with 64-bit PCI addresses being truncated because the return values of pci_resource_start() and pci_resource_end() were being cast to unsigned long. Signed-off-by: Aaron Sierra <asierra@xes-inc.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-31 00:32:34 +00:00
netdev->mem_start = pci_resource_start(pdev, 0);
netdev->mem_end = pci_resource_end(pdev, 0);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* PCI config space info */
hw->vendor_id = pdev->vendor;
hw->device_id = pdev->device;
hw->revision_id = pdev->revision;
hw->subsystem_vendor_id = pdev->subsystem_vendor;
hw->subsystem_device_id = pdev->subsystem_device;
/* Copy the default MAC, PHY and NVM function pointers */
memcpy(&hw->mac.ops, ei->mac_ops, sizeof(hw->mac.ops));
memcpy(&hw->phy.ops, ei->phy_ops, sizeof(hw->phy.ops));
memcpy(&hw->nvm.ops, ei->nvm_ops, sizeof(hw->nvm.ops));
/* Initialize skew-specific constants */
err = ei->get_invariants(hw);
if (err)
igb: remove redundant count set and err_hw_init Remove the setting of ring->count variables from igb_probe as they are duplicating the same configuration that is done igb_alloc_queues. Remove the err_hw_init tag as it can be replaced by err_sw_init. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:22:11 +00:00
goto err_sw_init;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: remove redundant count set and err_hw_init Remove the setting of ring->count variables from igb_probe as they are duplicating the same configuration that is done igb_alloc_queues. Remove the err_hw_init tag as it can be replaced by err_sw_init. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:22:11 +00:00
/* setup the private structure */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
err = igb_sw_init(adapter);
if (err)
goto err_sw_init;
igb_get_bus_info_pcie(hw);
hw->phy.autoneg_wait_to_complete = false;
/* Copper options */
if (hw->phy.media_type == e1000_media_type_copper) {
hw->phy.mdix = AUTO_ALL_MODES;
hw->phy.disable_polarity_correction = false;
hw->phy.ms_type = e1000_ms_hw_default;
}
if (igb_check_reset_block(hw))
dev_info(&pdev->dev,
"PHY reset is blocked due to SOL/IDER session.\n");
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* features is initialized to 0 in allocation, it might have bits
igb: add support for NETIF_F_RXHASH This patch adds support for Rx hashing. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:29 +00:00
* set by igb_sw_init so we should use an or instead of an
* assignment.
*/
netdev->features |= NETIF_F_SG |
NETIF_F_TSO |
NETIF_F_TSO6 |
NETIF_F_RXHASH |
NETIF_F_RXCSUM |
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
NETIF_F_HW_CSUM;
igb: add support for NETIF_F_RXHASH This patch adds support for Rx hashing. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:29 +00:00
igb: Add support for generic Tx checksums This patch adds support for generic Tx checksums to the igb driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-13 15:31:23 +00:00
if (hw->mac.type >= e1000_82576)
netdev->features |= NETIF_F_SCTP_CRC;
igb: Enable the hardware traffic class feature bit for igb models This will allow functionality depending on the hardware being traffic class aware to work. In particular the tc-flower offloading checks verifies that this bit is set. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:52 +00:00
if (hw->mac.type >= e1000_i350)
netdev->features |= NETIF_F_HW_TC;
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
#define IGB_GSO_PARTIAL_FEATURES (NETIF_F_GSO_GRE | \
NETIF_F_GSO_GRE_CSUM | \
net: define gso types for IPx over IPv4 and IPv6 This patch defines two new GSO definitions SKB_GSO_IPXIP4 and SKB_GSO_IPXIP6 along with corresponding NETIF_F_GSO_IPXIP4 and NETIF_F_GSO_IPXIP6. These are used to described IP in IP tunnel and what the outer protocol is. The inner protocol can be deduced from other GSO types (e.g. SKB_GSO_TCPV4 and SKB_GSO_TCPV6). The GSO types of SKB_GSO_IPIP and SKB_GSO_SIT are removed (these are both instances of SKB_GSO_IPXIP4). SKB_GSO_IPXIP6 will be used when support for GSO with IP encapsulation over IPv6 is added. Signed-off-by: Tom Herbert <tom@herbertland.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-05-18 16:06:10 +00:00
NETIF_F_GSO_IPXIP4 | \
intel: Add support for IPv6 IP-in-IP offload This patch adds support for offloading IPXIP6 type packets that represent either IPv4 or IPv6 encapsulated inside of an IPv6 outer IP header. In addition with this change we should also be able to support FOU encapsulated traffic with outer IPv6 headers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-05-18 17:44:53 +00:00
NETIF_F_GSO_IPXIP6 | \
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
NETIF_F_GSO_UDP_TUNNEL | \
NETIF_F_GSO_UDP_TUNNEL_CSUM)
netdev->gso_partial_features = IGB_GSO_PARTIAL_FEATURES;
netdev->features |= NETIF_F_GSO_PARTIAL | IGB_GSO_PARTIAL_FEATURES;
igb: add support for NETIF_F_RXHASH This patch adds support for Rx hashing. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:29 +00:00
/* copy netdev features into list of user selectable features */
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
netdev->hw_features |= netdev->features |
NETIF_F_HW_VLAN_CTAG_RX |
NETIF_F_HW_VLAN_CTAG_TX |
NETIF_F_RXALL;
igb: add support for NETIF_F_RXHASH This patch adds support for Rx hashing. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:29 +00:00
igb: Add support for generic Tx checksums This patch adds support for generic Tx checksums to the igb driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-13 15:31:23 +00:00
if (hw->mac.type >= e1000_i350)
netdev->hw_features |= NETIF_F_NTUPLE;
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
if (pci_using_dac)
netdev->features |= NETIF_F_HIGHDMA;
igb: Add support for generic Tx checksums This patch adds support for generic Tx checksums to the igb driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-13 15:31:23 +00:00
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
netdev->vlan_features |= netdev->features | NETIF_F_TSO_MANGLEID;
igb: Add support for generic Tx checksums This patch adds support for generic Tx checksums to the igb driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-13 15:31:23 +00:00
netdev->mpls_features |= NETIF_F_HW_CSUM;
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
netdev->hw_enc_features |= netdev->vlan_features;
igb: allow vlan devices to use TSO and TCP CSUM offload Using the new interface for propagating device feature flags into VLAN deivces, turn on TSO and CSUM offload on VLAN devices. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-05 11:06:27 +00:00
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
/* set this bit last since it cannot be part of vlan_features */
netdev->features |= NETIF_F_HW_VLAN_CTAG_FILTER |
NETIF_F_HW_VLAN_CTAG_RX |
NETIF_F_HW_VLAN_CTAG_TX;
igb: Support sending custom Ethernet FCS. Including bad FCS, used generate frames with bad FCS to test other system's handling of RX of bad packets. Signed-off-by: Ben Greear <greearb@candelatech.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-03-06 09:41:53 +00:00
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
netdev->priv_flags |= IFF_SUPP_NOFCS;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
net: introduce IFF_UNICAST_FLT private flag Use IFF_UNICAST_FTL to find out if driver handles unicast address filtering. In case it does not, promisc mode is entered. Patch also fixes following drivers: stmmac, niu: support uc filtering and yet it propagated ndo_set_multicast_list bna, benet, pxa168_eth, ks8851, ks8851_mll, ksz884x : has set ndo_set_rx_mode but do not support uc filtering Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-08-16 06:29:00 +00:00
netdev->priv_flags |= IFF_UNICAST_FLT;
ethernet/intel: use core min/max MTU checking e100: min_mtu 68, max_mtu 1500 - remove e100_change_mtu entirely, is identical to old eth_change_mtu, and no longer serves a purpose. No need to set min_mtu or max_mtu explicitly, as ether_setup() will already set them to 68 and 1500. e1000: min_mtu 46, max_mtu 16110 e1000e: min_mtu 68, max_mtu varies based on adapter fm10k: min_mtu 68, max_mtu 15342 - remove fm10k_change_mtu entirely, does nothing now i40e: min_mtu 68, max_mtu 9706 i40evf: min_mtu 68, max_mtu 9706 igb: min_mtu 68, max_mtu 9216 - There are two different "max" frame sizes claimed and both checked in the driver, the larger value wasn't relevant though, so I've set max_mtu to the smaller of the two values here to retain identical behavior. igbvf: min_mtu 68, max_mtu 9216 - Same issue as igb duplicated ixgb: min_mtu 68, max_mtu 16114 - Also remove pointless old == new check, as that's done in dev_set_mtu ixgbe: min_mtu 68, max_mtu 9710 ixgbevf: min_mtu 68, max_mtu dependent on hardware/firmware - Some hw can only handle up to max_mtu 1504 on a vf, others 9710 CC: netdev@vger.kernel.org CC: intel-wired-lan@lists.osuosl.org CC: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-17 19:54:05 +00:00
/* MTU range: 68 - 9216 */
netdev->min_mtu = ETH_MIN_MTU;
netdev->max_mtu = MAX_STD_JUMBO_FRAME_SIZE;
igb: replace unecessary &adapter->hw with just hw where applicable This patch just cleans up some unecessary references to the adapter->hw member when it has already been placed in a local variable named hw. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:35 +00:00
adapter->en_mng_pt = igb_enable_mng_pass_thru(hw);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* before reading the NVM, reset the controller to put the device in a
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* known good starting state
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
hw->mac.ops.reset_hw(hw);
igb: Refactor NVM read functions to accommodate devices with no flash This patch refactors NVM read functions in order to accommodate i210 devices that do not have a flash. Previously, this was not supported on i210 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-17 19:02:53 +00:00
/* make sure the NVM is good , i211/i210 parts can have special NVM
* that doesn't contain a checksum
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
*/
igb: Refactor NVM read functions to accommodate devices with no flash This patch refactors NVM read functions in order to accommodate i210 devices that do not have a flash. Previously, this was not supported on i210 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-17 19:02:53 +00:00
switch (hw->mac.type) {
case e1000_i210:
case e1000_i211:
if (igb_get_flash_presence_i210(hw)) {
if (hw->nvm.ops.validate(hw) < 0) {
dev_err(&pdev->dev,
"The NVM Checksum Is Not Valid\n");
err = -EIO;
goto err_eeprom;
}
}
break;
default:
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
if (hw->nvm.ops.validate(hw) < 0) {
dev_err(&pdev->dev, "The NVM Checksum Is Not Valid\n");
err = -EIO;
goto err_eeprom;
}
igb: Refactor NVM read functions to accommodate devices with no flash This patch refactors NVM read functions in order to accommodate i210 devices that do not have a flash. Previously, this was not supported on i210 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-17 19:02:53 +00:00
break;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: allow setting MAC address on i211 using a device tree blob The Intel i211 LOM PCIe Ethernet controllers' iNVM operates as an OTP and has no external EEPROM interface [1]. The following allows the driver to pickup the MAC address from a device tree blob when CONFIG_OF has been enabled. [1] http://www.intel.com/content/www/us/en/embedded/products/networking/i211-ethernet-controller-datasheet.html Signed-off-by: John Holland <jotihojr@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-02-18 11:10:52 +00:00
if (eth_platform_get_mac_address(&pdev->dev, hw->mac.addr)) {
/* copy the MAC address out of the NVM */
if (hw->mac.ops.read_mac_addr(hw))
dev_err(&pdev->dev, "NVM Read Error\n");
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
memcpy(netdev->dev_addr, hw->mac.addr, netdev->addr_len);
remove init of dev->perm_addr in drivers perm_addr is initialized correctly in register_netdevice() so to init it in drivers is no longer needed. Signed-off-by: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-01-08 01:38:26 +00:00
if (!is_valid_ether_addr(netdev->dev_addr)) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
dev_err(&pdev->dev, "Invalid MAC Address\n");
err = -EIO;
goto err_eeprom;
}
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
igb_set_default_mac_filter(adapter);
igb: Update firmware info output Our NVM image creation tools have evolved over the years and there are multiple versions contained in them, depending on the tool used to create them. This patch outputs the NVM versions available in ethtool -i output. rc2: (not sure why others show in log but not in the message) Added additional call to igb_set_fw_version per Community feedback. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-06-14 16:04:19 +00:00
/* get firmware version for ethtool -i */
igb_set_fw_version(adapter);
igb: add defaults for i210 TX/RX PBSIZE Set the defaults on probe for the packet buffer size registers for the i210. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-05-29 05:47:26 +00:00
/* configure RXPBSIZE and TXPBSIZE */
if (hw->mac.type == e1000_i210) {
wr32(E1000_RXPBS, I210_RXPBSIZE_DEFAULT);
wr32(E1000_TXPBS, I210_TXPBSIZE_DEFAULT);
}
ethernet/intel: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Switches test of .data field to .function, since .data will be going away. Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: intel-wired-lan@lists.osuosl.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-17 00:29:35 +00:00
timer_setup(&adapter->watchdog_timer, igb_watchdog, 0);
timer_setup(&adapter->phy_info_timer, igb_update_phy_info, 0);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
INIT_WORK(&adapter->reset_task, igb_reset_task);
INIT_WORK(&adapter->watchdog_task, igb_watchdog_task);
igb: remove redundant count set and err_hw_init Remove the setting of ring->count variables from igb_probe as they are duplicating the same configuration that is done igb_alloc_queues. Remove the err_hw_init tag as it can be replaced by err_sw_init. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:22:11 +00:00
/* Initialize link properties that are user-changeable */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
adapter->fc_autoneg = true;
hw->mac.autoneg = true;
hw->phy.autoneg_advertised = 0x2f;
igb: cleanup flow control configuration to make requested/current more clear This patch cleans up the flow control configuration for igb to make it a bit more readable in regards to what the requested and current modes are. This should help with the maintainability of the current igb driver in regards to flow control. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:10:24 +00:00
hw->fc.requested_mode = e1000_fc_default;
hw->fc.current_mode = e1000_fc_default;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_validate_mdi_setting(hw);
igb: No longer rely on APME to determine WoL settings Historically, we've been using the APME bit to determine whether a device supports wake on a given port or not. However, this bit specifies the default wake setting, rather than the wake support. Change the behavior so that we use a flag to keep the capabilities separate from the enablement while meeting customer requirements. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-11-09 05:49:54 +00:00
/* By default, support wake on port A */
igb: support wol on second port We need to support wol on the second port for situations such as when the lan ports are on the motherboard itself. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-13 20:41:17 +00:00
if (hw->bus.func == 0)
igb: No longer rely on APME to determine WoL settings Historically, we've been using the APME bit to determine whether a device supports wake on a given port or not. However, this bit specifies the default wake setting, rather than the wake support. Change the behavior so that we use a flag to keep the capabilities separate from the enablement while meeting customer requirements. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-11-09 05:49:54 +00:00
adapter->flags |= IGB_FLAG_WOL_SUPPORTED;
/* Check the NVM for wake support on non-port A ports */
if (hw->mac.type >= e1000_82580)
igb: Add full support for 82580 devices This patch makes use of the 82580 PHY and MAC support added and adds a set of supported device IDs for said hardware. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-19 12:42:21 +00:00
hw->nvm.ops.read(hw, NVM_INIT_CONTROL3_PORT_A +
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
NVM_82580_LAN_FUNC_OFFSET(hw->bus.func), 1,
&eeprom_data);
igb: support wol on second port We need to support wol on the second port for situations such as when the lan ports are on the motherboard itself. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-13 20:41:17 +00:00
else if (hw->bus.func == 1)
hw->nvm.ops.read(hw, NVM_INIT_CONTROL3_PORT_B, 1, &eeprom_data);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: No longer rely on APME to determine WoL settings Historically, we've been using the APME bit to determine whether a device supports wake on a given port or not. However, this bit specifies the default wake setting, rather than the wake support. Change the behavior so that we use a flag to keep the capabilities separate from the enablement while meeting customer requirements. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-11-09 05:49:54 +00:00
if (eeprom_data & IGB_EEPROM_APME)
adapter->flags |= IGB_FLAG_WOL_SUPPORTED;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* now that we have the eeprom settings, apply the special cases where
* the eeprom may be wrong or the board simply won't support wake on
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* lan on a particular port
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
switch (pdev->device) {
case E1000_DEV_ID_82575GB_QUAD_COPPER:
igb: No longer rely on APME to determine WoL settings Historically, we've been using the APME bit to determine whether a device supports wake on a given port or not. However, this bit specifies the default wake setting, rather than the wake support. Change the behavior so that we use a flag to keep the capabilities separate from the enablement while meeting customer requirements. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-11-09 05:49:54 +00:00
adapter->flags &= ~IGB_FLAG_WOL_SUPPORTED;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
break;
case E1000_DEV_ID_82575EB_FIBER_SERDES:
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
case E1000_DEV_ID_82576_FIBER:
case E1000_DEV_ID_82576_SERDES:
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Wake events only supported on port A for dual fiber
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* regardless of eeprom setting
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (rd32(E1000_STATUS) & E1000_STATUS_FUNC_1)
igb: No longer rely on APME to determine WoL settings Historically, we've been using the APME bit to determine whether a device supports wake on a given port or not. However, this bit specifies the default wake setting, rather than the wake support. Change the behavior so that we use a flag to keep the capabilities separate from the enablement while meeting customer requirements. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-11-09 05:49:54 +00:00
adapter->flags &= ~IGB_FLAG_WOL_SUPPORTED;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
break;
igb: add support for 82576 quad copper adapter Add support for 82576 copper adapter and necessary code to restrict wol for quad port adapter to first port. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-13 20:42:35 +00:00
case E1000_DEV_ID_82576_QUAD_COPPER:
igb: restrict WoL for 82576 ET2 Quad Port Server Adapter Restrict Wake-on-LAN to first port on 82576 ET2 quad port NICs, as it is only supported there. Signed-off-by: Stefan Assmann <sassmann@redhat.com> Acked-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-09 09:51:34 +00:00
case E1000_DEV_ID_82576_QUAD_COPPER_ET2:
igb: add support for 82576 quad copper adapter Add support for 82576 copper adapter and necessary code to restrict wol for quad port adapter to first port. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-13 20:42:35 +00:00
/* if quad port adapter, disable WoL on all but port A */
if (global_quad_port_a != 0)
igb: No longer rely on APME to determine WoL settings Historically, we've been using the APME bit to determine whether a device supports wake on a given port or not. However, this bit specifies the default wake setting, rather than the wake support. Change the behavior so that we use a flag to keep the capabilities separate from the enablement while meeting customer requirements. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-11-09 05:49:54 +00:00
adapter->flags &= ~IGB_FLAG_WOL_SUPPORTED;
igb: add support for 82576 quad copper adapter Add support for 82576 copper adapter and necessary code to restrict wol for quad port adapter to first port. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-13 20:42:35 +00:00
else
adapter->flags |= IGB_FLAG_QUAD_PORT_A;
/* Reset for multiple quad port adapters */
if (++global_quad_port_a == 4)
global_quad_port_a = 0;
break;
igb: No longer rely on APME to determine WoL settings Historically, we've been using the APME bit to determine whether a device supports wake on a given port or not. However, this bit specifies the default wake setting, rather than the wake support. Change the behavior so that we use a flag to keep the capabilities separate from the enablement while meeting customer requirements. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-11-09 05:49:54 +00:00
default:
/* If the device can't wake, don't set software support */
if (!device_can_wakeup(&adapter->pdev->dev))
adapter->flags &= ~IGB_FLAG_WOL_SUPPORTED;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/* initialize the wol settings based on the eeprom settings */
igb: No longer rely on APME to determine WoL settings Historically, we've been using the APME bit to determine whether a device supports wake on a given port or not. However, this bit specifies the default wake setting, rather than the wake support. Change the behavior so that we use a flag to keep the capabilities separate from the enablement while meeting customer requirements. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-11-09 05:49:54 +00:00
if (adapter->flags & IGB_FLAG_WOL_SUPPORTED)
adapter->wol |= E1000_WUFC_MAG;
/* Some vendors want WoL disabled by default, but still supported */
if ((hw->mac.type == e1000_i350) &&
(pdev->subsystem_vendor == PCI_VENDOR_ID_HP)) {
adapter->flags |= IGB_FLAG_WOL_SUPPORTED;
adapter->wol = 0;
}
igb: enable WoL for OEM devices regardless of EEPROM setting Override EEPROM settings for specific OEM devices. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-05 18:08:28 +00:00
/* Some vendors want the ability to Use the EEPROM setting as
* enable/disable only, and not for capability
*/
if (((hw->mac.type == e1000_i350) ||
(hw->mac.type == e1000_i354)) &&
(pdev->subsystem_vendor == PCI_VENDOR_ID_DELL)) {
adapter->flags |= IGB_FLAG_WOL_SUPPORTED;
adapter->wol = 0;
}
if (hw->mac.type == e1000_i350) {
if (((pdev->subsystem_device == 0x5001) ||
(pdev->subsystem_device == 0x5002)) &&
(hw->bus.func == 0)) {
adapter->flags |= IGB_FLAG_WOL_SUPPORTED;
adapter->wol = 0;
}
if (pdev->subsystem_device == 0x1F52)
adapter->flags |= IGB_FLAG_WOL_SUPPORTED;
}
igb: No longer rely on APME to determine WoL settings Historically, we've been using the APME bit to determine whether a device supports wake on a given port or not. However, this bit specifies the default wake setting, rather than the wake support. Change the behavior so that we use a flag to keep the capabilities separate from the enablement while meeting customer requirements. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-11-09 05:49:54 +00:00
device_set_wakeup_enable(&adapter->pdev->dev,
adapter->flags & IGB_FLAG_WOL_SUPPORTED);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* reset the hardware with the new settings */
igb_reset(adapter);
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
/* Init the I2C interface */
err = igb_init_i2c(adapter);
if (err) {
dev_err(&pdev->dev, "failed to init i2c interface\n");
goto err_eeprom;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* let the f/w know that the h/w is now under the control of the
igb: Cleanups to change comment style on license headers This patch fixes WARNING:NETWORKING_BLOCK_COMMENT_STYLE from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:46:06 +00:00
* driver.
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_get_hw_control(adapter);
strcpy(netdev->name, "eth%d");
err = register_netdev(netdev);
if (err)
goto err_register;
igb: fix link down inital state As reported by Andrew Lutomirski <amluto@gmail.com> All the intel wired ethernet drivers were calling netif_carrier_off and netif_stop_queue (or variants) before calling register_netdevice This is incorrect behavior as was pointed out by davem, and causes ifconfig and friends to report a strange state before first link after the driver was loaded, since without a netif_carrier_off, the stack assumes carrier_on, but before register_netdev, netlink messages are not sent out telling link state. This apparently confused *some* versions of networkmanager. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Reported-by: Andrew Lutomirski <amluto@gmail.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-17 20:44:32 +00:00
/* carrier off reporting is important to ethtool even BEFORE open */
netif_carrier_off(netdev);
igb: add IGB_DCA instead of selecting INTEL_IOATDMA Add a bool IGB_DCA defined to y if IGB and DCA are enabled, but IGB isn't y while DCA=m. And thus remove the need to select INTEL_IOATDMA when IGB is enabled, so that non-x86 architectures can build the igb driver. Based on work/patch from Brice Goglin <brice@myri.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-10-17 18:08:31 +00:00
#ifdef CONFIG_IGB_DCA
igb: Fix DCA errors and do not use context index for 82576 82576 was being incorrectly flagged as needing a context index. It does not as each ring has it's own table of 2 contexts. Driver was registering after registering the driver instead of the other way around. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-31 08:52:30 +00:00
if (dca_add_requester(&pdev->dev) == 0) {
igb: Add support for quad port WOL and feature flags Change igb from using a series of boolean operators to using a single flags value that contains a number of different bit flags for all the different features of the adapter. This patch also adds WOL support for quad port adapters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:46 +00:00
adapter->flags |= IGB_FLAG_DCA_ENABLED;
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
dev_info(&pdev->dev, "DCA enabled\n");
igb_setup_dca(adapter);
}
igb: access to NIC time Adds the register definitions and code to read the time register. Signed-off-by: John Ronciak <john.ronciak@intel.com> Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-12 05:03:41 +00:00
#endif
igb: Enable hwmon data output for thermal sensors via I2C. Some of our adapters have internal sensors that report thermal data. This patch enables reporting of that data via sysfs. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:01:42 +00:00
#ifdef CONFIG_IGB_HWMON
/* Initialize the thermal sensor on i350 devices. */
if (hw->mac.type == e1000_i350 && hw->bus.func == 0) {
u16 ets_word;
igb: Tidy up wrapping for CONFIG_IGB_PTP. For users without CONFIG_IGB_PTP=y, we should not be compiling any PTP code into the driver. Tidy up the wrapping in igb to support this. Cc: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Matthew Vick <matthew.vick@intel.com> Acked-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-10 05:40:43 +00:00
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Read the NVM to determine if this i350 device supports an
igb: Enable hwmon data output for thermal sensors via I2C. Some of our adapters have internal sensors that report thermal data. This patch enables reporting of that data via sysfs. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:01:42 +00:00
* external thermal sensor.
*/
hw->nvm.ops.read(hw, NVM_ETS_CFG, 1, &ets_word);
if (ets_word != 0x0000 && ets_word != 0xFFFF)
adapter->ets = true;
else
adapter->ets = false;
if (igb_sysfs_init(adapter))
dev_err(&pdev->dev,
"failed to allocate sysfs resources\n");
} else {
adapter->ets = false;
}
#endif
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
/* Check if Media Autosense is enabled */
adapter->ei = *ei;
if (hw->dev_spec._82575.mas_capable)
igb_init_mas(adapter);
igb: fix hw timestamping Hardware timestamping for Intel 82580 didn't work in either 2.6.36 or 2.6.37. Comparing it to Intel's igb-2.4.12 I found that the timecounter_init clock/counter initialization was done too early. Signed-off-by: Anders Berggren <andfers@halon.se> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-04 07:32:32 +00:00
/* do hw tstamp init after resetting */
igb: offer a PTP Hardware Clock instead of the timecompare method This commit removes the legacy timecompare code from the igb driver and offers a tunable PHC instead. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-03-16 10:55:37 +00:00
igb_ptp_init(adapter);
igb: fix hw timestamping Hardware timestamping for Intel 82580 didn't work in either 2.6.36 or 2.6.37. Comparing it to Intel's igb-2.4.12 I found that the timecounter_init clock/counter initialization was done too early. Signed-off-by: Anders Berggren <andfers@halon.se> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-04 07:32:32 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
dev_info(&pdev->dev, "Intel(R) Gigabit Ethernet Network Connection\n");
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
/* print bus type/speed/width info, not applicable to i354 */
if (hw->mac.type != e1000_i354) {
dev_info(&pdev->dev, "%s: (PCIe:%s:%s) %pM\n",
netdev->name,
((hw->bus.speed == e1000_bus_speed_2500) ? "2.5Gb/s" :
(hw->bus.speed == e1000_bus_speed_5000) ? "5.0Gb/s" :
"unknown"),
((hw->bus.width == e1000_bus_width_pcie_x4) ?
"Width x4" :
(hw->bus.width == e1000_bus_width_pcie_x2) ?
"Width x2" :
(hw->bus.width == e1000_bus_width_pcie_x1) ?
"Width x1" : "unknown"), netdev->dev_addr);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Don't look for a PBA in the iNVM when flashless When a part is flashless, do not look for a PBA in the iNVM. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-08-23 07:49:00 +00:00
if ((hw->mac.type >= e1000_i210 ||
igb_get_flash_presence_i210(hw))) {
ret_val = igb_read_part_string(hw, part_str,
E1000_PBANUM_LENGTH);
} else {
ret_val = -E1000_ERR_INVM_VALUE_NOT_FOUND;
}
igb: Add new function to read part number from EEPROM in string format New adapters will have part numbers stored in string format rather than simple hex format. This function will read part number formats in either hex or string. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-22 17:17:21 +00:00
if (ret_val)
strcpy(part_str, "Unknown");
dev_info(&pdev->dev, "%s: PBA No: %s\n", netdev->name, part_str);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
dev_info(&pdev->dev,
"Using %s interrupts. %d rx queue(s), %d tx queue(s)\n",
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
(adapter->flags & IGB_FLAG_HAS_MSIX) ? "MSI-X" :
igb: Add support for quad port WOL and feature flags Change igb from using a series of boolean operators to using a single flags value that contains a number of different bit flags for all the different features of the adapter. This patch also adds WOL support for quad port adapters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:46 +00:00
(adapter->flags & IGB_FLAG_HAS_MSI) ? "MSI" : "legacy",
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
adapter->num_rx_queues, adapter->num_tx_queues);
igb: Fix for devices using ethtool for EEE settings This patch fixes a problem where using ethtool for EEE setting was not working correctly. This patch also fixes a problem where the function that checks for EEE status on i354 devices was not being called and was causing warnings with static analysis tools. Reported-by: Rashika Kheria <rashika.kheria@gmail.com> Reported-by: Josh Triplett <josh@joshtriplett.org> Reported-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-12 03:58:22 +00:00
if (hw->phy.media_type == e1000_media_type_copper) {
switch (hw->mac.type) {
case e1000_i350:
case e1000_i210:
case e1000_i211:
/* Enable EEE for internal copper PHY devices */
igb: add flags to set eee advertisement mode Change e1000_set_eee and e1000_set_eee_i35(0|4) to allow changes in the advertised EEE speeds from ethtool. Adds two boolean flags to e1000_set_eee_i35(0|4) to pass in advertised speed data. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-08-29 06:43:13 +00:00
err = igb_set_eee_i350(hw, true, true);
igb: Fix for devices using ethtool for EEE settings This patch fixes a problem where using ethtool for EEE setting was not working correctly. This patch also fixes a problem where the function that checks for EEE status on i354 devices was not being called and was causing warnings with static analysis tools. Reported-by: Rashika Kheria <rashika.kheria@gmail.com> Reported-by: Josh Triplett <josh@joshtriplett.org> Reported-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-12 03:58:22 +00:00
if ((!err) &&
(!hw->dev_spec._82575.eee_disable)) {
adapter->eee_advert =
MDIO_EEE_100TX | MDIO_EEE_1000T;
adapter->flags |= IGB_FLAG_EEE;
}
break;
case e1000_i354:
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
if ((rd32(E1000_CTRL_EXT) &
igb: Fix for devices using ethtool for EEE settings This patch fixes a problem where using ethtool for EEE setting was not working correctly. This patch also fixes a problem where the function that checks for EEE status on i354 devices was not being called and was causing warnings with static analysis tools. Reported-by: Rashika Kheria <rashika.kheria@gmail.com> Reported-by: Josh Triplett <josh@joshtriplett.org> Reported-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-12 03:58:22 +00:00
E1000_CTRL_EXT_LINK_MODE_SGMII)) {
igb: add flags to set eee advertisement mode Change e1000_set_eee and e1000_set_eee_i35(0|4) to allow changes in the advertised EEE speeds from ethtool. Adds two boolean flags to e1000_set_eee_i35(0|4) to pass in advertised speed data. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-08-29 06:43:13 +00:00
err = igb_set_eee_i354(hw, true, true);
igb: Fix for devices using ethtool for EEE settings This patch fixes a problem where using ethtool for EEE setting was not working correctly. This patch also fixes a problem where the function that checks for EEE status on i354 devices was not being called and was causing warnings with static analysis tools. Reported-by: Rashika Kheria <rashika.kheria@gmail.com> Reported-by: Josh Triplett <josh@joshtriplett.org> Reported-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-12 03:58:22 +00:00
if ((!err) &&
(!hw->dev_spec._82575.eee_disable)) {
adapter->eee_advert =
MDIO_EEE_100TX | MDIO_EEE_1000T;
adapter->flags |= IGB_FLAG_EEE;
}
}
break;
default:
break;
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
}
igb: Add Energy Efficient Ethernet (EEE) for i350 devices. This patch adds the EEE feature for i350 devices, enabled by default. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-03-12 04:42:13 +00:00
}
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
pm_runtime_put_noidle(&pdev->dev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return 0;
err_register:
igb_release_hw_control(adapter);
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
memset(&adapter->i2c_adap, 0, sizeof(adapter->i2c_adap));
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
err_eeprom:
if (!igb_check_reset_block(hw))
igb: do not use phy ops in ethtool test cleanup for non-copper parts Currently the igb driver is experiencing a panic due to a null function pointer being used during the cleanup of the ethtool looback test on fiber/serdes parts. This patch prevents that and adds a check prior to calling any phy function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-22 05:30:24 +00:00
igb_reset_phy(hw);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (hw->flash_address)
iounmap(hw->flash_address);
err_sw_init:
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
kfree(adapter->mac_table);
igb: Fix a memory leak in igb_probe In error handling code of igb_probe, the memory adapter->shadow_vfta allocated by kcalloc in igb_sw_init is not freed. So when register_netdev or igb_init_i2c is failed, a memory leak will occur. This patch adds kfree to fix it. Signed-off-by: Jia-Ju Bai <baijiaju1990@163.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-08-05 14:05:16 +00:00
kfree(adapter->shadow_vfta);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb_clear_interrupt_scheme(adapter);
igb: make sure SR-IOV init uses the right number of queues Recent changes to igb_probe_vfs() could lead to the PF holding onto all of the queues. Reorder igb_probe_vfs() to be before gb_init_queue_configuration() and add some more error checking. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-08-08 00:27:39 +00:00
#ifdef CONFIG_PCI_IOV
igb_disable_sriov(pdev);
#endif
igb: don't unmap NULL hw_addr I've got a startech thunderbolt dock someone loaned me, which among other things, has the following device in it: 08:00.0 Ethernet controller: Intel Corporation I210 Gigabit Network Connection (rev 03) This hotplugs just fine (kernel 4.2.0 plus a patch or two here): [ 863.020315] igb: Intel(R) Gigabit Ethernet Network Driver - version 5.2.18-k [ 863.020316] igb: Copyright (c) 2007-2014 Intel Corporation. [ 863.028657] igb 0000:08:00.0: enabling device (0000 -> 0002) [ 863.062089] igb 0000:08:00.0: added PHC on eth0 [ 863.062090] igb 0000:08:00.0: Intel(R) Gigabit Ethernet Network Connection [ 863.062091] igb 0000:08:00.0: eth0: (PCIe:2.5Gb/s:Width x1) e8:ea:6a:00:1b:2a [ 863.062194] igb 0000:08:00.0: eth0: PBA No: 000200-000 [ 863.062196] igb 0000:08:00.0: Using MSI-X interrupts. 4 rx queue(s), 4 tx queue(s) [ 863.064889] igb 0000:08:00.0 enp8s0: renamed from eth0 But disconnecting it is another story: [ 1002.807932] igb 0000:08:00.0: removed PHC on enp8s0 [ 1002.807944] igb 0000:08:00.0 enp8s0: PCIe link lost, device now detached [ 1003.341141] ------------[ cut here ]------------ [ 1003.341148] WARNING: CPU: 0 PID: 199 at lib/iomap.c:43 bad_io_access+0x38/0x40() [ 1003.341149] Bad IO access at port 0x0 () [ 1003.342767] Modules linked in: snd_usb_audio snd_usbmidi_lib snd_rawmidi igb dca firewire_ohci firewire_core crc_itu_t rfcomm ctr ccm arc4 iwlmvm mac80211 fuse xt_CHECKSUM ipt_MASQUERADE nf_nat_masquerade_ipv4 tun ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_security ip6table_raw ip6table_filter ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw iptable_filter bnep dm_mirror dm_region_hash dm_log dm_mod coretemp x86_pkg_temp_thermal intel_powerclamp kvm_intel snd_hda_codec_hdmi kvm crct10dif_pclmul crc32_pclmul ghash_clmulni_intel drbg [ 1003.342793] ansi_cprng aesni_intel hp_wmi aes_x86_64 iTCO_wdt lrw iTCO_vendor_support ppdev gf128mul sparse_keymap glue_helper ablk_helper cryptd snd_hda_codec_realtek snd_hda_codec_generic microcode snd_hda_intel uvcvideo iwlwifi snd_hda_codec videobuf2_vmalloc videobuf2_memops snd_hda_core videobuf2_core snd_hwdep btusb v4l2_common btrtl snd_seq btbcm btintel videodev cfg80211 snd_seq_device rtsx_pci_ms bluetooth pcspkr input_leds i2c_i801 media parport_pc memstick rfkill sg lpc_ich snd_pcm 8250_fintek parport joydev snd_timer snd soundcore hp_accel ie31200_edac mei_me lis3lv02d edac_core input_polldev mei hp_wireless shpchp tpm_infineon sch_fq_codel nfsd auth_rpcgss nfs_acl lockd grace sunrpc ip_tables autofs4 xfs libcrc32c sd_mod sr_mod cdrom rtsx_pci_sdmmc mmc_core crc32c_intel serio_raw rtsx_pci [ 1003.342822] nouveau ahci libahci mxm_wmi e1000e xhci_pci hwmon ptp drm_kms_helper pps_core xhci_hcd ttm wmi video ipv6 [ 1003.342839] CPU: 0 PID: 199 Comm: kworker/0:2 Not tainted 4.2.0-2.el7_UNSUPPORTED.x86_64 #1 [ 1003.342840] Hardware name: Hewlett-Packard HP ZBook 15 G2/2253, BIOS M70 Ver. 01.07 02/26/2015 [ 1003.342843] Workqueue: pciehp-3 pciehp_power_thread [ 1003.342844] ffffffff81a90655 ffff8804866d3b48 ffffffff8164763a 0000000000000000 [ 1003.342846] ffff8804866d3b98 ffff8804866d3b88 ffffffff8107134a ffff8804866d3b88 [ 1003.342847] ffff880486f46000 ffff88046c8a8000 ffff880486f46840 ffff88046c8a8098 [ 1003.342848] Call Trace: [ 1003.342852] [<ffffffff8164763a>] dump_stack+0x45/0x57 [ 1003.342855] [<ffffffff8107134a>] warn_slowpath_common+0x8a/0xc0 [ 1003.342857] [<ffffffff810713c6>] warn_slowpath_fmt+0x46/0x50 [ 1003.342859] [<ffffffff8133719e>] ? pci_disable_msix+0x3e/0x50 [ 1003.342860] [<ffffffff812f6328>] bad_io_access+0x38/0x40 [ 1003.342861] [<ffffffff812f6567>] pci_iounmap+0x27/0x40 [ 1003.342865] [<ffffffffa0b728d7>] igb_remove+0xc7/0x160 [igb] [ 1003.342867] [<ffffffff8132189f>] pci_device_remove+0x3f/0xc0 [ 1003.342869] [<ffffffff81433426>] __device_release_driver+0x96/0x130 [ 1003.342870] [<ffffffff814334e3>] device_release_driver+0x23/0x30 [ 1003.342871] [<ffffffff8131b404>] pci_stop_bus_device+0x94/0xa0 [ 1003.342872] [<ffffffff8131b3ad>] pci_stop_bus_device+0x3d/0xa0 [ 1003.342873] [<ffffffff8131b3ad>] pci_stop_bus_device+0x3d/0xa0 [ 1003.342874] [<ffffffff8131b516>] pci_stop_and_remove_bus_device+0x16/0x30 [ 1003.342876] [<ffffffff81333f5b>] pciehp_unconfigure_device+0x9b/0x180 [ 1003.342877] [<ffffffff81333a73>] pciehp_disable_slot+0x43/0xb0 [ 1003.342878] [<ffffffff81333b6d>] pciehp_power_thread+0x8d/0xb0 [ 1003.342885] [<ffffffff810881b2>] process_one_work+0x152/0x3d0 [ 1003.342886] [<ffffffff8108854a>] worker_thread+0x11a/0x460 [ 1003.342887] [<ffffffff81088430>] ? process_one_work+0x3d0/0x3d0 [ 1003.342890] [<ffffffff8108ddd9>] kthread+0xc9/0xe0 [ 1003.342891] [<ffffffff8108dd10>] ? kthread_create_on_node+0x180/0x180 [ 1003.342893] [<ffffffff8164e29f>] ret_from_fork+0x3f/0x70 [ 1003.342894] [<ffffffff8108dd10>] ? kthread_create_on_node+0x180/0x180 [ 1003.342895] ---[ end trace 65a77e06d5aa9358 ]--- Upon looking at the igb driver, I see that igb_rd32() attempted to read from hw_addr and failed, so it set hw->hw_addr to NULL and spit out the message in the log output above, "PCIe link lost, device now detached". Well, now that hw_addr is NULL, the attempt to call pci_iounmap is obviously not going to go well. As suggested by Mark Rustad, do something similar to what ixgbe does, and save a copy of hw_addr as adapter->io_addr, so we can still call pci_iounmap on it on teardown. Additionally, for consistency, make the pci_iomap call assignment directly to io_addr, so map and unmap match. Signed-off-by: Jarod Wilson <jarod@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-09-10 19:37:50 +00:00
pci_iounmap(pdev, adapter->io_addr);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
err_ioremap:
free_netdev(netdev);
err_alloc_etherdev:
ethernet/intel: Use pci_(request|release)_mem_regions Now that we do have pci_request_mem_regions() and pci_release_mem_regions() at hand, use it in the Intel ethernet drivers. Suggested-by: Christoph Hellwig <hch@infradead.org> Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> CC: David S. Miller <davem@davemloft.net>
2016-06-07 07:44:05 +00:00
pci_release_mem_regions(pdev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
err_pci_reg:
err_dma:
pci_disable_device(pdev);
return err;
}
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
#ifdef CONFIG_PCI_IOV
igb: fix driver reload with VF assigned to guest commit fa44f2f185f7f9da19d331929bb1b56c1ccd1d93 broke reloading of igb, when VFs are assigned to a guest, in several ways. 1. on module load adapter->vf_data does not get properly allocated, resulting in a null pointer exception when accessing adapter->vf_data in igb_reset() on module reload. modprobe -r igb ; modprobe igb max_vfs=7 [ 215.215837] igb 0000:01:00.1: removed PHC on eth1 [ 216.932072] igb 0000:01:00.1: IOV Disabled [ 216.937038] igb 0000:01:00.0: removed PHC on eth0 [ 217.127032] igb 0000:01:00.0: Cannot deallocate SR-IOV virtual functions while they are assigned - VFs will not be deallocated [ 217.146178] igb: Intel(R) Gigabit Ethernet Network Driver - version 5.0.5-k [ 217.154050] igb: Copyright (c) 2007-2013 Intel Corporation. [ 217.160688] igb 0000:01:00.0: Enabling SR-IOV VFs using the module parameter is deprecated - please use the pci sysfs interface. [ 217.173703] igb 0000:01:00.0: irq 103 for MSI/MSI-X [ 217.179227] igb 0000:01:00.0: irq 104 for MSI/MSI-X [ 217.184735] igb 0000:01:00.0: irq 105 for MSI/MSI-X [ 217.220082] BUG: unable to handle kernel NULL pointer dereference at 0000000000000048 [ 217.228846] IP: [<ffffffffa007c5e5>] igb_reset+0xc5/0x4b0 [igb] [ 217.235472] PGD 3607ec067 PUD 36170b067 PMD 0 [ 217.240461] Oops: 0002 [#1] SMP [ 217.244085] Modules linked in: igb(+) igbvf mptsas mptscsih mptbase scsi_transport_sas [last unloaded: igb] [ 217.255040] CPU: 4 PID: 4833 Comm: modprobe Not tainted 3.11.0+ #46 [...] [ 217.390007] [<ffffffffa007fab2>] igb_probe+0x892/0xfd0 [igb] [ 217.396422] [<ffffffff81470b3e>] local_pci_probe+0x1e/0x40 [ 217.402641] [<ffffffff81472029>] pci_device_probe+0xf9/0x110 [...] 2. A follow up issue, pci_enable_sriov() should only be called if no VFs were still allocated on module unload. Otherwise pci_enable_sriov() gets called multiple times in a row rendering the NIC unusable until reset. 3. simply calling igb_enable_sriov() in igb_probe_vfs() is not enough as the interrupts need to be re-setup. Switching that to igb_pci_enable_sriov(). Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Sibai Li <Sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-09-24 05:18:39 +00:00
static int igb_disable_sriov(struct pci_dev *pdev)
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
{
struct net_device *netdev = pci_get_drvdata(pdev);
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
/* reclaim resources allocated to VFs */
if (adapter->vf_data) {
/* disable iov and allow time for transactions to clear */
igb: Use pci_vfs_assigned instead of igb_vfs_are_assigned This change makes it so that the igb driver uses the generic helper pci_vfs_assigned instead of the igb specific function igb_vfs_are_assigned. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 00:03:26 +00:00
if (pci_vfs_assigned(pdev)) {
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
dev_warn(&pdev->dev,
"Cannot deallocate SR-IOV virtual functions while they are assigned - VFs will not be deallocated\n");
return -EPERM;
} else {
pci_disable_sriov(pdev);
msleep(500);
}
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
kfree(adapter->vf_mac_list);
adapter->vf_mac_list = NULL;
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
kfree(adapter->vf_data);
adapter->vf_data = NULL;
adapter->vfs_allocated_count = 0;
wr32(E1000_IOVCTL, E1000_IOVCTL_REUSE_VFQ);
wrfl();
msleep(100);
dev_info(&pdev->dev, "IOV Disabled\n");
/* Re-enable DMA Coalescing flag since IOV is turned off */
adapter->flags |= IGB_FLAG_DMAC;
}
return 0;
}
static int igb_enable_sriov(struct pci_dev *pdev, int num_vfs)
{
struct net_device *netdev = pci_get_drvdata(pdev);
struct igb_adapter *adapter = netdev_priv(netdev);
int old_vfs = pci_num_vf(pdev);
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
struct vf_mac_filter *mac_list;
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
int err = 0;
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
int num_vf_mac_filters, i;
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
if (!(adapter->flags & IGB_FLAG_HAS_MSIX) || num_vfs > 7) {
igb: don't allow SR-IOV without MSI-X MSI-X interrupts are required for SR-IOV operation. Check to make sure they're enabled before allowing the user to turn on VFs. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-06-20 06:03:36 +00:00
err = -EPERM;
goto out;
}
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
if (!num_vfs)
goto out;
igb: fix driver reload with VF assigned to guest commit fa44f2f185f7f9da19d331929bb1b56c1ccd1d93 broke reloading of igb, when VFs are assigned to a guest, in several ways. 1. on module load adapter->vf_data does not get properly allocated, resulting in a null pointer exception when accessing adapter->vf_data in igb_reset() on module reload. modprobe -r igb ; modprobe igb max_vfs=7 [ 215.215837] igb 0000:01:00.1: removed PHC on eth1 [ 216.932072] igb 0000:01:00.1: IOV Disabled [ 216.937038] igb 0000:01:00.0: removed PHC on eth0 [ 217.127032] igb 0000:01:00.0: Cannot deallocate SR-IOV virtual functions while they are assigned - VFs will not be deallocated [ 217.146178] igb: Intel(R) Gigabit Ethernet Network Driver - version 5.0.5-k [ 217.154050] igb: Copyright (c) 2007-2013 Intel Corporation. [ 217.160688] igb 0000:01:00.0: Enabling SR-IOV VFs using the module parameter is deprecated - please use the pci sysfs interface. [ 217.173703] igb 0000:01:00.0: irq 103 for MSI/MSI-X [ 217.179227] igb 0000:01:00.0: irq 104 for MSI/MSI-X [ 217.184735] igb 0000:01:00.0: irq 105 for MSI/MSI-X [ 217.220082] BUG: unable to handle kernel NULL pointer dereference at 0000000000000048 [ 217.228846] IP: [<ffffffffa007c5e5>] igb_reset+0xc5/0x4b0 [igb] [ 217.235472] PGD 3607ec067 PUD 36170b067 PMD 0 [ 217.240461] Oops: 0002 [#1] SMP [ 217.244085] Modules linked in: igb(+) igbvf mptsas mptscsih mptbase scsi_transport_sas [last unloaded: igb] [ 217.255040] CPU: 4 PID: 4833 Comm: modprobe Not tainted 3.11.0+ #46 [...] [ 217.390007] [<ffffffffa007fab2>] igb_probe+0x892/0xfd0 [igb] [ 217.396422] [<ffffffff81470b3e>] local_pci_probe+0x1e/0x40 [ 217.402641] [<ffffffff81472029>] pci_device_probe+0xf9/0x110 [...] 2. A follow up issue, pci_enable_sriov() should only be called if no VFs were still allocated on module unload. Otherwise pci_enable_sriov() gets called multiple times in a row rendering the NIC unusable until reset. 3. simply calling igb_enable_sriov() in igb_probe_vfs() is not enough as the interrupts need to be re-setup. Switching that to igb_pci_enable_sriov(). Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Sibai Li <Sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-09-24 05:18:39 +00:00
if (old_vfs) {
dev_info(&pdev->dev, "%d pre-allocated VFs found - override max_vfs setting of %d\n",
old_vfs, max_vfs);
adapter->vfs_allocated_count = old_vfs;
} else
adapter->vfs_allocated_count = num_vfs;
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
adapter->vf_data = kcalloc(adapter->vfs_allocated_count,
sizeof(struct vf_data_storage), GFP_KERNEL);
/* if allocation failed then we do not support SR-IOV */
if (!adapter->vf_data) {
adapter->vfs_allocated_count = 0;
err = -ENOMEM;
goto out;
}
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
/* Due to the limited number of RAR entries calculate potential
* number of MAC filters available for the VFs. Reserve entries
* for PF default MAC, PF MAC filters and at least one RAR entry
* for each VF for VF MAC.
*/
num_vf_mac_filters = adapter->hw.mac.rar_entry_count -
(1 + IGB_PF_MAC_FILTERS_RESERVED +
adapter->vfs_allocated_count);
adapter->vf_mac_list = kcalloc(num_vf_mac_filters,
sizeof(struct vf_mac_filter),
GFP_KERNEL);
mac_list = adapter->vf_mac_list;
INIT_LIST_HEAD(&adapter->vf_macs.l);
if (adapter->vf_mac_list) {
/* Initialize list of VF MAC filters */
for (i = 0; i < num_vf_mac_filters; i++) {
mac_list->vf = -1;
mac_list->free = true;
list_add(&mac_list->l, &adapter->vf_macs.l);
mac_list++;
}
} else {
/* If we could not allocate memory for the VF MAC filters
* we can continue without this feature but warn user.
*/
dev_err(&pdev->dev,
"Unable to allocate memory for VF MAC filter list\n");
}
igb: fix driver reload with VF assigned to guest commit fa44f2f185f7f9da19d331929bb1b56c1ccd1d93 broke reloading of igb, when VFs are assigned to a guest, in several ways. 1. on module load adapter->vf_data does not get properly allocated, resulting in a null pointer exception when accessing adapter->vf_data in igb_reset() on module reload. modprobe -r igb ; modprobe igb max_vfs=7 [ 215.215837] igb 0000:01:00.1: removed PHC on eth1 [ 216.932072] igb 0000:01:00.1: IOV Disabled [ 216.937038] igb 0000:01:00.0: removed PHC on eth0 [ 217.127032] igb 0000:01:00.0: Cannot deallocate SR-IOV virtual functions while they are assigned - VFs will not be deallocated [ 217.146178] igb: Intel(R) Gigabit Ethernet Network Driver - version 5.0.5-k [ 217.154050] igb: Copyright (c) 2007-2013 Intel Corporation. [ 217.160688] igb 0000:01:00.0: Enabling SR-IOV VFs using the module parameter is deprecated - please use the pci sysfs interface. [ 217.173703] igb 0000:01:00.0: irq 103 for MSI/MSI-X [ 217.179227] igb 0000:01:00.0: irq 104 for MSI/MSI-X [ 217.184735] igb 0000:01:00.0: irq 105 for MSI/MSI-X [ 217.220082] BUG: unable to handle kernel NULL pointer dereference at 0000000000000048 [ 217.228846] IP: [<ffffffffa007c5e5>] igb_reset+0xc5/0x4b0 [igb] [ 217.235472] PGD 3607ec067 PUD 36170b067 PMD 0 [ 217.240461] Oops: 0002 [#1] SMP [ 217.244085] Modules linked in: igb(+) igbvf mptsas mptscsih mptbase scsi_transport_sas [last unloaded: igb] [ 217.255040] CPU: 4 PID: 4833 Comm: modprobe Not tainted 3.11.0+ #46 [...] [ 217.390007] [<ffffffffa007fab2>] igb_probe+0x892/0xfd0 [igb] [ 217.396422] [<ffffffff81470b3e>] local_pci_probe+0x1e/0x40 [ 217.402641] [<ffffffff81472029>] pci_device_probe+0xf9/0x110 [...] 2. A follow up issue, pci_enable_sriov() should only be called if no VFs were still allocated on module unload. Otherwise pci_enable_sriov() gets called multiple times in a row rendering the NIC unusable until reset. 3. simply calling igb_enable_sriov() in igb_probe_vfs() is not enough as the interrupts need to be re-setup. Switching that to igb_pci_enable_sriov(). Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Sibai Li <Sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-09-24 05:18:39 +00:00
/* only call pci_enable_sriov() if no VFs are allocated already */
if (!old_vfs) {
err = pci_enable_sriov(pdev, adapter->vfs_allocated_count);
if (err)
goto err_out;
}
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
dev_info(&pdev->dev, "%d VFs allocated\n",
adapter->vfs_allocated_count);
for (i = 0; i < adapter->vfs_allocated_count; i++)
igb_vf_configure(adapter, i);
/* DMA Coalescing is not supported in IOV mode. */
adapter->flags &= ~IGB_FLAG_DMAC;
goto out;
err_out:
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
kfree(adapter->vf_mac_list);
adapter->vf_mac_list = NULL;
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
kfree(adapter->vf_data);
adapter->vf_data = NULL;
adapter->vfs_allocated_count = 0;
out:
return err;
}
#endif
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/**
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
* igb_remove_i2c - Cleanup I2C interface
* @adapter: pointer to adapter structure
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
**/
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
static void igb_remove_i2c(struct igb_adapter *adapter)
{
/* free the adapter bus structure */
i2c_del_adapter(&adapter->i2c_adap);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_remove - Device Removal Routine
* @pdev: PCI device information struct
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_remove is called by the PCI subsystem to alert the driver
* that it should release a PCI device. The could be caused by a
* Hot-Plug event, or because the driver is going to be removed from
* memory.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
net/intel: remove __dev* attributes CONFIG_HOTPLUG is going away as an option. As result the __dev* markings will be going away. Remove use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit. Signed-off-by: Bill Pemberton <wfp5p@virginia.edu> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Jesse Brandeburg <jesse.brandeburg@intel.com> Cc: Bruce Allan <bruce.w.allan@intel.com> Cc: Carolyn Wyborny <carolyn.wyborny@intel.com> Cc: Don Skidmore <donald.c.skidmore@intel.com> Cc: Greg Rose <gregory.v.rose@intel.com> Cc: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Cc: Alex Duyck <alexander.h.duyck@intel.com> Cc: John Ronciak <john.ronciak@intel.com> Cc: Tushar Dave <tushar.n.dave@intel.com> Cc: e1000-devel@lists.sourceforge.net Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-03 14:24:25 +00:00
static void igb_remove(struct pci_dev *pdev)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct net_device *netdev = pci_get_drvdata(pdev);
struct igb_adapter *adapter = netdev_priv(netdev);
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
pm_runtime_get_noresume(&pdev->dev);
igb: Enable hwmon data output for thermal sensors via I2C. Some of our adapters have internal sensors that report thermal data. This patch enables reporting of that data via sysfs. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:01:42 +00:00
#ifdef CONFIG_IGB_HWMON
igb_sysfs_exit(adapter);
#endif
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
igb_remove_i2c(adapter);
igb: Update PTP function names/variables and locations. Where possible, move PTP-related functions into igb_ptp.c and update the names of functions and variables to match the established coding style in the files and specify that they are PTP-specific. Cc: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-10 05:40:44 +00:00
igb_ptp_stop(adapter);
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* The watchdog timer may be rescheduled, so explicitly
igb[v],ixgbe: don't use flush_scheduled_work() All three drivers use flush_scheduled_work() similarly during driver detach. Replace it with explicit cancels. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: e1000-devel@lists.sourceforge.net Cc: netdev@vger.kernel.org
2010-12-12 15:45:14 +00:00
* disable watchdog from being rescheduled.
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
set_bit(__IGB_DOWN, &adapter->state);
del_timer_sync(&adapter->watchdog_timer);
del_timer_sync(&adapter->phy_info_timer);
igb[v],ixgbe: don't use flush_scheduled_work() All three drivers use flush_scheduled_work() similarly during driver detach. Replace it with explicit cancels. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: e1000-devel@lists.sourceforge.net Cc: netdev@vger.kernel.org
2010-12-12 15:45:14 +00:00
cancel_work_sync(&adapter->reset_task);
cancel_work_sync(&adapter->watchdog_task);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add IGB_DCA instead of selecting INTEL_IOATDMA Add a bool IGB_DCA defined to y if IGB and DCA are enabled, but IGB isn't y while DCA=m. And thus remove the need to select INTEL_IOATDMA when IGB is enabled, so that non-x86 architectures can build the igb driver. Based on work/patch from Brice Goglin <brice@myri.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-10-17 18:08:31 +00:00
#ifdef CONFIG_IGB_DCA
igb: Add support for quad port WOL and feature flags Change igb from using a series of boolean operators to using a single flags value that contains a number of different bit flags for all the different features of the adapter. This patch also adds WOL support for quad port adapters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:46 +00:00
if (adapter->flags & IGB_FLAG_DCA_ENABLED) {
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
dev_info(&pdev->dev, "DCA disabled\n");
dca_remove_requester(&pdev->dev);
igb: Add support for quad port WOL and feature flags Change igb from using a series of boolean operators to using a single flags value that contains a number of different bit flags for all the different features of the adapter. This patch also adds WOL support for quad port adapters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:46 +00:00
adapter->flags &= ~IGB_FLAG_DCA_ENABLED;
igb: remove unused defines This patch removes all of the unused defines from the .h files contained in igb. For some defines there was a use and so I plugged them into the correct locations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-16 07:59:44 +00:00
wr32(E1000_DCA_CTRL, E1000_DCA_CTRL_DCA_MODE_DISABLE);
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
}
#endif
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Release control of h/w to f/w. If f/w is AMT enabled, this
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* would have already happened in close and is redundant.
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_release_hw_control(adapter);
igb: this patch addes the sr-iov enablement option via num_vfs parameter This code adds a module parameter called num_vfs which defines if the driver should attempt to use sr-iov and if so how many VFs should be enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:30 +00:00
#ifdef CONFIG_PCI_IOV
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
igb_disable_sriov(pdev);
igb: this patch addes the sr-iov enablement option via num_vfs parameter This code adds a module parameter called num_vfs which defines if the driver should attempt to use sr-iov and if so how many VFs should be enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:30 +00:00
#endif
igb: cleanup whitespace issues in igb_main.c This patch goes through and cleans up whitespace issues in igb_main.c to help improve readability. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:52:50 +00:00
igb: Teardown SR-IOV before unregister_netdev() When the .remove() callback for a PF is called, SR-IOV support for the device is disabled, which requires unbinding and removing the VFs. The VFs may be in-use either by the host kernel or userspace, such as assigned to a VM through vfio-pci. In this latter case, the VFs may be removed either by shutting down the VM or hot-unplugging the devices from the VM. Unfortunately in the case of a Windows 2012 R2 guest, hot-unplug is broken due to the ordering of the PF driver teardown. Disabling SR-IOV prior to unregister_netdev() avoids this issue. Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-07-29 20:38:15 +00:00
unregister_netdev(netdev);
igb_clear_interrupt_scheme(adapter);
igb: don't unmap NULL hw_addr I've got a startech thunderbolt dock someone loaned me, which among other things, has the following device in it: 08:00.0 Ethernet controller: Intel Corporation I210 Gigabit Network Connection (rev 03) This hotplugs just fine (kernel 4.2.0 plus a patch or two here): [ 863.020315] igb: Intel(R) Gigabit Ethernet Network Driver - version 5.2.18-k [ 863.020316] igb: Copyright (c) 2007-2014 Intel Corporation. [ 863.028657] igb 0000:08:00.0: enabling device (0000 -> 0002) [ 863.062089] igb 0000:08:00.0: added PHC on eth0 [ 863.062090] igb 0000:08:00.0: Intel(R) Gigabit Ethernet Network Connection [ 863.062091] igb 0000:08:00.0: eth0: (PCIe:2.5Gb/s:Width x1) e8:ea:6a:00:1b:2a [ 863.062194] igb 0000:08:00.0: eth0: PBA No: 000200-000 [ 863.062196] igb 0000:08:00.0: Using MSI-X interrupts. 4 rx queue(s), 4 tx queue(s) [ 863.064889] igb 0000:08:00.0 enp8s0: renamed from eth0 But disconnecting it is another story: [ 1002.807932] igb 0000:08:00.0: removed PHC on enp8s0 [ 1002.807944] igb 0000:08:00.0 enp8s0: PCIe link lost, device now detached [ 1003.341141] ------------[ cut here ]------------ [ 1003.341148] WARNING: CPU: 0 PID: 199 at lib/iomap.c:43 bad_io_access+0x38/0x40() [ 1003.341149] Bad IO access at port 0x0 () [ 1003.342767] Modules linked in: snd_usb_audio snd_usbmidi_lib snd_rawmidi igb dca firewire_ohci firewire_core crc_itu_t rfcomm ctr ccm arc4 iwlmvm mac80211 fuse xt_CHECKSUM ipt_MASQUERADE nf_nat_masquerade_ipv4 tun ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_security ip6table_raw ip6table_filter ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw iptable_filter bnep dm_mirror dm_region_hash dm_log dm_mod coretemp x86_pkg_temp_thermal intel_powerclamp kvm_intel snd_hda_codec_hdmi kvm crct10dif_pclmul crc32_pclmul ghash_clmulni_intel drbg [ 1003.342793] ansi_cprng aesni_intel hp_wmi aes_x86_64 iTCO_wdt lrw iTCO_vendor_support ppdev gf128mul sparse_keymap glue_helper ablk_helper cryptd snd_hda_codec_realtek snd_hda_codec_generic microcode snd_hda_intel uvcvideo iwlwifi snd_hda_codec videobuf2_vmalloc videobuf2_memops snd_hda_core videobuf2_core snd_hwdep btusb v4l2_common btrtl snd_seq btbcm btintel videodev cfg80211 snd_seq_device rtsx_pci_ms bluetooth pcspkr input_leds i2c_i801 media parport_pc memstick rfkill sg lpc_ich snd_pcm 8250_fintek parport joydev snd_timer snd soundcore hp_accel ie31200_edac mei_me lis3lv02d edac_core input_polldev mei hp_wireless shpchp tpm_infineon sch_fq_codel nfsd auth_rpcgss nfs_acl lockd grace sunrpc ip_tables autofs4 xfs libcrc32c sd_mod sr_mod cdrom rtsx_pci_sdmmc mmc_core crc32c_intel serio_raw rtsx_pci [ 1003.342822] nouveau ahci libahci mxm_wmi e1000e xhci_pci hwmon ptp drm_kms_helper pps_core xhci_hcd ttm wmi video ipv6 [ 1003.342839] CPU: 0 PID: 199 Comm: kworker/0:2 Not tainted 4.2.0-2.el7_UNSUPPORTED.x86_64 #1 [ 1003.342840] Hardware name: Hewlett-Packard HP ZBook 15 G2/2253, BIOS M70 Ver. 01.07 02/26/2015 [ 1003.342843] Workqueue: pciehp-3 pciehp_power_thread [ 1003.342844] ffffffff81a90655 ffff8804866d3b48 ffffffff8164763a 0000000000000000 [ 1003.342846] ffff8804866d3b98 ffff8804866d3b88 ffffffff8107134a ffff8804866d3b88 [ 1003.342847] ffff880486f46000 ffff88046c8a8000 ffff880486f46840 ffff88046c8a8098 [ 1003.342848] Call Trace: [ 1003.342852] [<ffffffff8164763a>] dump_stack+0x45/0x57 [ 1003.342855] [<ffffffff8107134a>] warn_slowpath_common+0x8a/0xc0 [ 1003.342857] [<ffffffff810713c6>] warn_slowpath_fmt+0x46/0x50 [ 1003.342859] [<ffffffff8133719e>] ? pci_disable_msix+0x3e/0x50 [ 1003.342860] [<ffffffff812f6328>] bad_io_access+0x38/0x40 [ 1003.342861] [<ffffffff812f6567>] pci_iounmap+0x27/0x40 [ 1003.342865] [<ffffffffa0b728d7>] igb_remove+0xc7/0x160 [igb] [ 1003.342867] [<ffffffff8132189f>] pci_device_remove+0x3f/0xc0 [ 1003.342869] [<ffffffff81433426>] __device_release_driver+0x96/0x130 [ 1003.342870] [<ffffffff814334e3>] device_release_driver+0x23/0x30 [ 1003.342871] [<ffffffff8131b404>] pci_stop_bus_device+0x94/0xa0 [ 1003.342872] [<ffffffff8131b3ad>] pci_stop_bus_device+0x3d/0xa0 [ 1003.342873] [<ffffffff8131b3ad>] pci_stop_bus_device+0x3d/0xa0 [ 1003.342874] [<ffffffff8131b516>] pci_stop_and_remove_bus_device+0x16/0x30 [ 1003.342876] [<ffffffff81333f5b>] pciehp_unconfigure_device+0x9b/0x180 [ 1003.342877] [<ffffffff81333a73>] pciehp_disable_slot+0x43/0xb0 [ 1003.342878] [<ffffffff81333b6d>] pciehp_power_thread+0x8d/0xb0 [ 1003.342885] [<ffffffff810881b2>] process_one_work+0x152/0x3d0 [ 1003.342886] [<ffffffff8108854a>] worker_thread+0x11a/0x460 [ 1003.342887] [<ffffffff81088430>] ? process_one_work+0x3d0/0x3d0 [ 1003.342890] [<ffffffff8108ddd9>] kthread+0xc9/0xe0 [ 1003.342891] [<ffffffff8108dd10>] ? kthread_create_on_node+0x180/0x180 [ 1003.342893] [<ffffffff8164e29f>] ret_from_fork+0x3f/0x70 [ 1003.342894] [<ffffffff8108dd10>] ? kthread_create_on_node+0x180/0x180 [ 1003.342895] ---[ end trace 65a77e06d5aa9358 ]--- Upon looking at the igb driver, I see that igb_rd32() attempted to read from hw_addr and failed, so it set hw->hw_addr to NULL and spit out the message in the log output above, "PCIe link lost, device now detached". Well, now that hw_addr is NULL, the attempt to call pci_iounmap is obviously not going to go well. As suggested by Mark Rustad, do something similar to what ixgbe does, and save a copy of hw_addr as adapter->io_addr, so we can still call pci_iounmap on it on teardown. Additionally, for consistency, make the pci_iomap call assignment directly to io_addr, so map and unmap match. Signed-off-by: Jarod Wilson <jarod@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-09-10 19:37:50 +00:00
pci_iounmap(pdev, adapter->io_addr);
igb: remove unnecessary adapter->hw calls when just hw-> will do. There were several spots in the code making calls to adapter->hw when they could have just been accessing hw-> directly. I cleaned up the spots where this was visibly apparent. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:20:31 +00:00
if (hw->flash_address)
iounmap(hw->flash_address);
ethernet/intel: Use pci_(request|release)_mem_regions Now that we do have pci_request_mem_regions() and pci_release_mem_regions() at hand, use it in the Intel ethernet drivers. Suggested-by: Christoph Hellwig <hch@infradead.org> Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> CC: David S. Miller <davem@davemloft.net>
2016-06-07 07:44:05 +00:00
pci_release_mem_regions(pdev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
kfree(adapter->mac_table);
igb: VFTA Table Fix for i350 devices Due to a hardware problem, writes to the VFTA register can theoretically fail. Although the likelihood of this is very low. This patch adds a shadow vfta in the adapter struct for reading and adds new write functions for these devices to work around the problem. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-14 00:13:49 +00:00
kfree(adapter->shadow_vfta);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
free_netdev(netdev);
e1000e/igb/ixgbe: Don't report an error if devices don't support AER The only error returned by pci_{en,dis}able_pcie_error_reporting() is -EIO which simply means that Advanced Error Reporting is not supported. There is no need to report that, so remove the error check from e1000e, igb and ixgbe. Signed-off-by: Frans Pop <elendil@planet.nl> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-02 17:04:12 +00:00
pci_disable_pcie_error_reporting(pdev);
igb: Add support for pci-e Advanced Error Reporting Add the calls necessary to enable advanced error reporting for igb on systems with AER enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-27 08:24:37 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
pci_disable_device(pdev);
}
igb: move vf init into a seperate function This patch moves VF initialization into a seperate function to help improve the readability of igb_probe. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:53 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_probe_vfs - Initialize vf data storage and add VFs to pci config space
* @adapter: board private structure to initialize
igb: move vf init into a seperate function This patch moves VF initialization into a seperate function to help improve the readability of igb_probe. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:53 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* This function initializes the vf specific data storage and then attempts to
* allocate the VFs. The reason for ordering it this way is because it is much
* mor expensive time wise to disable SR-IOV than it is to allocate and free
* the memory for the VFs.
igb: move vf init into a seperate function This patch moves VF initialization into a seperate function to help improve the readability of igb_probe. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:53 +00:00
**/
net/intel: remove __dev* attributes CONFIG_HOTPLUG is going away as an option. As result the __dev* markings will be going away. Remove use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit. Signed-off-by: Bill Pemberton <wfp5p@virginia.edu> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Jesse Brandeburg <jesse.brandeburg@intel.com> Cc: Bruce Allan <bruce.w.allan@intel.com> Cc: Carolyn Wyborny <carolyn.wyborny@intel.com> Cc: Don Skidmore <donald.c.skidmore@intel.com> Cc: Greg Rose <gregory.v.rose@intel.com> Cc: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Cc: Alex Duyck <alexander.h.duyck@intel.com> Cc: John Ronciak <john.ronciak@intel.com> Cc: Tushar Dave <tushar.n.dave@intel.com> Cc: e1000-devel@lists.sourceforge.net Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-03 14:24:25 +00:00
static void igb_probe_vfs(struct igb_adapter *adapter)
igb: move vf init into a seperate function This patch moves VF initialization into a seperate function to help improve the readability of igb_probe. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:53 +00:00
{
#ifdef CONFIG_PCI_IOV
struct pci_dev *pdev = adapter->pdev;
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: move vf init into a seperate function This patch moves VF initialization into a seperate function to help improve the readability of igb_probe. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:53 +00:00
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
/* Virtualization features not supported on i210 family. */
if ((hw->mac.type == e1000_i210) || (hw->mac.type == e1000_i211))
return;
igb: fix NULL derefs due to skipped SR-IOV enabling The combined effect of commits 6423fc3416 ("igb: do not re-init SR-IOV during probe") and ceee3450b3 ("igb: make sure SR-IOV init uses the right number of queues") causes VFs no longer getting set up, leading to NULL pointer dereferences due to the adapter's ->vf_data being NULL while ->vfs_allocated_count is non-zero. The first commit not only neglected the side effect of igb_sriov_reinit() that the second commit tried to account for, but also that of setting IGB_FLAG_HAS_MSIX, without which igb_enable_sriov() is effectively a no-op. Calling igb_{,re}set_interrupt_capability() as done here seems to address this, but I'm not sure whether this is better than sinply reverting the other two commits. Signed-off-by: Jan Beulich <jbeulich@suse.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-10-19 10:23:29 +00:00
/* Of the below we really only want the effect of getting
* IGB_FLAG_HAS_MSIX set (if available), without which
* igb_enable_sriov() has no effect.
*/
igb_set_interrupt_capability(adapter, true);
igb_reset_interrupt_capability(adapter);
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
pci_sriov_set_totalvfs(pdev, 7);
igb: do not re-init SR-IOV during probe During driver probing the following code path is triggered. igb_probe ->igb_sw_init ->igb_probe_vfs ->igb_pci_enable_sriov ->igb_sriov_reinit Doing the SR-IOV re-init is not necessary during probing since we're starting from scratch. Here we can call igb_enable_sriov() right away. Running igb_sriov_reinit() during igb_probe() also seems to cause occasional packet loss on some onboard 82576 NICs. Reproduced on Dell and HP servers with onboard 82576 NICs. Example: Intel Corporation 82576 Gigabit Network Connection [8086:10c9] (rev 01) Subsystem: Dell Device [1028:0481] Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-07-10 13:01:12 +00:00
igb_enable_sriov(pdev, max_vfs);
igb: Check if subordinate VFs are assigned to virtual machines Kvm and the Xen pci-back driver will set a flag in the virtual function pci device dev_flags when the VF is assigned to a guest VM. Before destroying subordinate VFs check to see if the flag is set and if so skip the call to pci_disable_sriov() to avoid system crashes. Copy the maintainer for the Xen pci-back driver. Also CC'ing maintainers of all drivers found to call pci_disable_sriov(). V2 - Fix uninitialized variable warning Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Christian Benvenuti <benve@cisco.com> Cc: Sathya Perla <sathya.perla@emulex.com> Cc: Dimitris Michailidis <dm@chelsio.com> Cc: Jon Mason <jdmason@kudzu.us> Cc: James Smart <james.smart@emulex.com> Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-14 02:57:14 +00:00
igb: move vf init into a seperate function This patch moves VF initialization into a seperate function to help improve the readability of igb_probe. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:53 +00:00
#endif /* CONFIG_PCI_IOV */
}
igb: add function to get maximum RSS queues This patch adds a new function igb_get_max_rss_queues() to get maximum RSS queues, this will reduce duplicate code and facilitate future maintenance. Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-09-19 13:40:54 +00:00
unsigned int igb_get_max_rss_queues(struct igb_adapter *adapter)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct e1000_hw *hw = &adapter->hw;
igb: add function to get maximum RSS queues This patch adds a new function igb_get_max_rss_queues() to get maximum RSS queues, this will reduce duplicate code and facilitate future maintenance. Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-09-19 13:40:54 +00:00
unsigned int max_rss_queues;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Streamline RSS queue and queue pairing assignment logic. Rather than spread out the complexity of the RSS queue and queue pairing assignment logic, place it all in one location for simplicity and readability. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-05-18 04:54:58 +00:00
/* Determine the maximum number of RSS queues supported. */
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
switch (hw->mac.type) {
igb: Streamline RSS queue and queue pairing assignment logic. Rather than spread out the complexity of the RSS queue and queue pairing assignment logic, place it all in one location for simplicity and readability. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-05-18 04:54:58 +00:00
case e1000_i211:
max_rss_queues = IGB_MAX_RX_QUEUES_I211;
break;
case e1000_82575:
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
case e1000_i210:
igb: Streamline RSS queue and queue pairing assignment logic. Rather than spread out the complexity of the RSS queue and queue pairing assignment logic, place it all in one location for simplicity and readability. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-05-18 04:54:58 +00:00
max_rss_queues = IGB_MAX_RX_QUEUES_82575;
break;
case e1000_i350:
/* I350 cannot do RSS and SR-IOV at the same time */
if (!!adapter->vfs_allocated_count) {
max_rss_queues = 1;
break;
}
/* fall through */
case e1000_82576:
if (!!adapter->vfs_allocated_count) {
max_rss_queues = 2;
break;
}
/* fall through */
case e1000_82580:
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
case e1000_i354:
igb: Streamline RSS queue and queue pairing assignment logic. Rather than spread out the complexity of the RSS queue and queue pairing assignment logic, place it all in one location for simplicity and readability. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-05-18 04:54:58 +00:00
default:
max_rss_queues = IGB_MAX_RX_QUEUES;
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
break;
igb: Streamline RSS queue and queue pairing assignment logic. Rather than spread out the complexity of the RSS queue and queue pairing assignment logic, place it all in one location for simplicity and readability. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-05-18 04:54:58 +00:00
}
igb: add function to get maximum RSS queues This patch adds a new function igb_get_max_rss_queues() to get maximum RSS queues, this will reduce duplicate code and facilitate future maintenance. Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-09-19 13:40:54 +00:00
return max_rss_queues;
}
static void igb_init_queue_configuration(struct igb_adapter *adapter)
{
u32 max_rss_queues;
max_rss_queues = igb_get_max_rss_queues(adapter);
igb: Streamline RSS queue and queue pairing assignment logic. Rather than spread out the complexity of the RSS queue and queue pairing assignment logic, place it all in one location for simplicity and readability. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-05-18 04:54:58 +00:00
adapter->rss_queues = min_t(u32, max_rss_queues, num_online_cpus());
igb: Fix oops caused by missing queue pairing When initializing igb driver (e.g. 82576, I350), IGB_FLAG_QUEUE_PAIRS is set if adapter->rss_queues exceeds half of max_rss_queues in igb_init_queue_configuration(). On the other hand, IGB_FLAG_QUEUE_PAIRS is not set even if the number of queues exceeds half of max_combined in igb_set_channels() when changing the number of queues by "ethtool -L". In this case, if numvecs is larger than MAX_MSIX_ENTRIES (10), the size of adapter->msix_entries[], an overflow can occur in igb_set_interrupt_capability(), which in turn leads to an oops. Fix this problem as follows: - When changing the number of queues by "ethtool -L", set IGB_FLAG_QUEUE_PAIRS in the same way as initializing igb driver. - When increasing the size of q_vector, reallocate it appropriately. (With IGB_FLAG_QUEUE_PAIRS set, the size of q_vector gets larger.) Another possible way to fix this problem is to cap the queues at its initial number, which is the number of the initial online cpus. But this is not the optimal way because we cannot increase queues when another cpu becomes online. Note that before commit cd14ef54d25b ("igb: Change to use statically allocated array for MSIx entries"), this problem did not cause oops but just made the number of queues become 1 because of entering msi_only mode in igb_set_interrupt_capability(). Fixes: 907b7835799f ("igb: Add ethtool support to configure number of channels") CC: stable <stable@vger.kernel.org> Signed-off-by: Shota Suzuki <suzuki_shota_t3@lab.ntt.co.jp> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-07-01 00:25:52 +00:00
igb_set_flag_queue_pairs(adapter, max_rss_queues);
}
void igb_set_flag_queue_pairs(struct igb_adapter *adapter,
const u32 max_rss_queues)
{
struct e1000_hw *hw = &adapter->hw;
igb: Streamline RSS queue and queue pairing assignment logic. Rather than spread out the complexity of the RSS queue and queue pairing assignment logic, place it all in one location for simplicity and readability. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-05-18 04:54:58 +00:00
/* Determine if we need to pair queues. */
switch (hw->mac.type) {
case e1000_82575:
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
case e1000_i211:
igb: Streamline RSS queue and queue pairing assignment logic. Rather than spread out the complexity of the RSS queue and queue pairing assignment logic, place it all in one location for simplicity and readability. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-05-18 04:54:58 +00:00
/* Device supports enough interrupts without queue pairing. */
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
break;
igb: Streamline RSS queue and queue pairing assignment logic. Rather than spread out the complexity of the RSS queue and queue pairing assignment logic, place it all in one location for simplicity and readability. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-05-18 04:54:58 +00:00
case e1000_82576:
case e1000_82580:
case e1000_i350:
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
case e1000_i354:
igb: Streamline RSS queue and queue pairing assignment logic. Rather than spread out the complexity of the RSS queue and queue pairing assignment logic, place it all in one location for simplicity and readability. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-05-18 04:54:58 +00:00
case e1000_i210:
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
default:
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* If rss_queues > half of max_rss_queues, pair the queues in
igb: Streamline RSS queue and queue pairing assignment logic. Rather than spread out the complexity of the RSS queue and queue pairing assignment logic, place it all in one location for simplicity and readability. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-05-18 04:54:58 +00:00
* order to conserve interrupts due to limited supply.
*/
if (adapter->rss_queues > (max_rss_queues / 2))
adapter->flags |= IGB_FLAG_QUEUE_PAIRS;
igb: Unpair the queues when changing the number of queues By the commit 72ddef0506da ("igb: Fix oops caused by missing queue pairing"), the IGB_FLAG_QUEUE_PAIRS flag can now be set when changing the number of queues by "ethtool -L", but it is never cleared unless the igb driver is reloaded. This patch clears it if queue pairing becomes unnecessary as a result of "ethtool -L". Signed-off-by: Shota Suzuki <suzuki_shota_t3@lab.ntt.co.jp> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-12-11 09:44:00 +00:00
else
adapter->flags &= ~IGB_FLAG_QUEUE_PAIRS;
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
break;
}
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_sw_init - Initialize general software structures (struct igb_adapter)
* @adapter: board private structure to initialize
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_sw_init initializes the Adapter private data structure.
* Fields are initialized based on PCI device information and
* OS network device settings (MTU size).
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
**/
static int igb_sw_init(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
struct net_device *netdev = adapter->netdev;
struct pci_dev *pdev = adapter->pdev;
pci_read_config_word(pdev, PCI_COMMAND, &hw->bus.pci_cmd_word);
/* set default ring sizes */
adapter->tx_ring_count = IGB_DEFAULT_TXD;
adapter->rx_ring_count = IGB_DEFAULT_RXD;
/* set default ITR values */
adapter->rx_itr_setting = IGB_DEFAULT_ITR;
adapter->tx_itr_setting = IGB_DEFAULT_ITR;
/* set default work limits */
adapter->tx_work_limit = IGB_DEFAULT_TX_WORK;
adapter->max_frame_size = netdev->mtu + ETH_HLEN + ETH_FCS_LEN +
VLAN_HLEN;
adapter->min_frame_size = ETH_ZLEN + ETH_FCS_LEN;
igb: add support of RX network flow classification This patch is meant to allow for RX network flow classification to insert and remove Rx filter by ethtool. Ethtool interface has it's own rules manager Show all filters: $ ethtool -n eth0 4 RX rings available Total 2 rules Signed-off-by: Ruhao Gao <ruhao.gao@ni.com> Signed-off-by: Gangfeng Huang <gangfeng.huang@ni.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-07-06 05:22:54 +00:00
spin_lock_init(&adapter->nfc_lock);
igb: reduce CPU0 latency when updating statistics This change is based off of the work and suggestion of Jan Jablonsky <jan.jablonsky@thalesgroup.com>. The Watchdog workqueue in igb driver is scheduled every 2s for each network interface. That includes updating a statistics protected by spinlock. Function igb_update_stats in this case will be protected against preemption. According to number of a statistics registers (cca 60), processing this function might cause additional cpu load on CPU0. In case of statistics spinlock may be replaced with mutex, which reduce latency on CPU0. CC: Bernhard Kaindl <bernhard.kaindl@thalesgroup.com> CC: Jan Jablonsky <jan.jablonsky@thalesgroup.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-08-02 17:13:10 +00:00
mutex_init(&adapter->stats64_lock);
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
#ifdef CONFIG_PCI_IOV
switch (hw->mac.type) {
case e1000_82576:
case e1000_i350:
if (max_vfs > 7) {
dev_warn(&pdev->dev,
"Maximum of 7 VFs per PF, using max\n");
igb: Fix null pointer dereference The max_vfs= option has always been self limiting to the number of VFs supported by the device. fa44f2f1 added SR-IOV configuration via sysfs, but in the process broke this self correction factor. The failing path is: igb_probe igb_sw_init if (max_vfs > 7) { adapter->vfs_allocated_count = 7; ... igb_probe_vfs igb_enable_sriov(, max_vfs) if (num_vfs > 7) { err = -EPERM; ... This leaves vfs_allocated_count = 7 and vf_data = NULL, so we bomb out when igb_probe finally calls igb_reset. It seems like a really bad idea, and somewhat pointless, to set vfs_allocated_count separate from vf_data, but limiting max_vfs is enough to avoid the null pointer. Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Acked-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-13 15:50:24 +00:00
max_vfs = adapter->vfs_allocated_count = 7;
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
} else
adapter->vfs_allocated_count = max_vfs;
if (adapter->vfs_allocated_count)
dev_warn(&pdev->dev,
"Enabling SR-IOV VFs using the module parameter is deprecated - please use the pci sysfs interface.\n");
break;
default:
break;
}
#endif /* CONFIG_PCI_IOV */
igb: assume MSI-X interrupts during initialization In igb_sw_init() the sequence of calls was changed from igb_init_queue_configuration() igb_init_interrupt_scheme() igb_probe_vfs() to igb_probe_vfs() igb_init_queue_configuration() igb_init_interrupt_scheme() This results in adapter->flags not having the IGB_FLAG_HAS_MSIX bit set during igb_probe_vfs()->igb_enable_sriov(). Therefore SR-IOV does not get enabled properly and we run into a NULL pointer if the max_vfs module parameter is specified (adapter->vf_data does not get allocated, crash on accessing the structure). [ 7.419348] BUG: unable to handle kernel NULL pointer dereference at 0000000000000048 [ 7.419367] IP: [<ffffffffa02161c6>] igb_reset+0xe6/0x5d0 [igb] [ 7.419370] PGD 0 [ 7.419373] Oops: 0002 [#1] SMP [ 7.419381] Modules linked in: ahci(+) libahci igb(+) i40e(+) vxlan ip6_udp_tunnel udp_tunnel megaraid_sas(+) ixgbe(+) mdio [ 7.419385] CPU: 0 PID: 4 Comm: kworker/0:0 Not tainted 4.2.0+ #153 [ 7.419387] Hardware name: Dell Inc. PowerEdge R720/0C4Y3R, BIOS 1.6.0 03/07/2013 [...] [ 7.419431] Call Trace: [ 7.419442] [<ffffffffa0217236>] igb_probe+0x8b6/0x1340 [igb] [ 7.419447] [<ffffffff814c7f15>] local_pci_probe+0x45/0xa0 Prevent this by setting the IGB_FLAG_HAS_MSIX bit before calling igb_probe_vfs(). The real interrupt capabilities will be checked during igb_init_interrupt_scheme() so this is safe to do. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-09-17 12:46:10 +00:00
/* Assume MSI-X interrupts, will be checked during IRQ allocation */
adapter->flags |= IGB_FLAG_HAS_MSIX;
treewide: kzalloc() -> kcalloc() The kzalloc() function has a 2-factor argument form, kcalloc(). This patch replaces cases of: kzalloc(a * b, gfp) with: kcalloc(a * b, gfp) as well as handling cases of: kzalloc(a * b * c, gfp) with: kzalloc(array3_size(a, b, c), gfp) as it's slightly less ugly than: kzalloc_array(array_size(a, b), c, gfp) This does, however, attempt to ignore constant size factors like: kzalloc(4 * 1024, gfp) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( kzalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | kzalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( kzalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(__u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(char) * COUNT + COUNT , ...) | kzalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_ID) + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_ID + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_CONST) + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_CONST + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_ID) + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_ID + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_CONST) + COUNT_CONST, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_CONST + COUNT_CONST, sizeof(THING) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ - kzalloc + kcalloc ( - SIZE * COUNT + COUNT, SIZE , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( kzalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( kzalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( kzalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products, // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( kzalloc(C1 * C2 * C3, ...) | kzalloc( - (E1) * E2 * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * (E3) + array3_size(E1, E2, E3) , ...) | kzalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants, // keeping sizeof() as the second factor argument. @@ expression THING, E1, E2; type TYPE; constant C1, C2, C3; @@ ( kzalloc(sizeof(THING) * C2, ...) | kzalloc(sizeof(TYPE) * C2, ...) | kzalloc(C1 * C2 * C3, ...) | kzalloc(C1 * C2, ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (E2) + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * E2 + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (E2) + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * E2 + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - (E1) * E2 + E1, E2 , ...) | - kzalloc + kcalloc ( - (E1) * (E2) + E1, E2 , ...) | - kzalloc + kcalloc ( - E1 * E2 + E1, E2 , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-12 21:03:40 +00:00
adapter->mac_table = kcalloc(hw->mac.rar_entry_count,
sizeof(struct igb_mac_addr),
igb: Replace GFP_ATOMIC with GFP_KERNEL in igb_sw_init() igb_sw_init() is never called in atomic context. It calls kzalloc() and kcalloc() with GFP_ATOMIC, which is not necessary. GFP_ATOMIC can be replaced with GFP_KERNEL. This is found by a static analysis tool named DCNS written by myself. Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-07-27 08:04:53 +00:00
GFP_KERNEL);
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
if (!adapter->mac_table)
return -ENOMEM;
igb: make sure SR-IOV init uses the right number of queues Recent changes to igb_probe_vfs() could lead to the PF holding onto all of the queues. Reorder igb_probe_vfs() to be before gb_init_queue_configuration() and add some more error checking. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-08-08 00:27:39 +00:00
igb_probe_vfs(adapter);
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
igb_init_queue_configuration(adapter);
igb: when number of CPUs > 4 combine tx/rx queues to allow more queues This patch makes it so that nics such as 82576 and newer can support more hardware queues when there are more than 4 cpus by combining a tx/rx queue pair onto one interrupt so that 8 queue pairs can be supported and thus allow for more queues. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:19 +00:00
igb: VFTA Table Fix for i350 devices Due to a hardware problem, writes to the VFTA register can theoretically fail. Although the likelihood of this is very low. This patch adds a shadow vfta in the adapter struct for reading and adds new write functions for these devices to work around the problem. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-14 00:13:49 +00:00
/* Setup and initialize a copy of the hw vlan table array */
ethernet: Remove unnecessary alloc/OOM messages, alloc cleanups alloc failures already get standardized OOM messages and a dump_stack. Convert kzalloc's with multiplies to kcalloc. Convert kmalloc's with multiplies to kmalloc_array. Fix a few whitespace defects. Convert a constant 6 to ETH_ALEN. Use parentheses around sizeof. Convert vmalloc/memset to vzalloc. Remove now unused size variables. Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-02-03 17:43:58 +00:00
adapter->shadow_vfta = kcalloc(E1000_VLAN_FILTER_TBL_SIZE, sizeof(u32),
igb: Replace GFP_ATOMIC with GFP_KERNEL in igb_sw_init() igb_sw_init() is never called in atomic context. It calls kzalloc() and kcalloc() with GFP_ATOMIC, which is not necessary. GFP_ATOMIC can be replaced with GFP_KERNEL. This is found by a static analysis tool named DCNS written by myself. Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-07-27 08:04:53 +00:00
GFP_KERNEL);
igb: check memory allocation failure Check memory allocation failures and return -ENOMEM in such cases, as already done for other memory allocations in this function. This avoids NULL pointers dereference. Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Tested-by: Aaron Brown <aaron.f.brown@intel.com Acked-by: PJ Waskiewicz <peter.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-08-27 06:39:51 +00:00
if (!adapter->shadow_vfta)
return -ENOMEM;
igb: VFTA Table Fix for i350 devices Due to a hardware problem, writes to the VFTA register can theoretically fail. Although the likelihood of this is very low. This patch adds a shadow vfta in the adapter struct for reading and adds new write functions for these devices to work around the problem. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-14 00:13:49 +00:00
igb: move vf init into a seperate function This patch moves VF initialization into a seperate function to help improve the readability of igb_probe. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:53 +00:00
/* This call may decrease the number of queues */
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
if (igb_init_interrupt_scheme(adapter, true)) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
dev_err(&pdev->dev, "Unable to allocate memory for queues\n");
return -ENOMEM;
}
/* Explicitly disable IRQ since the NIC can be in any state. */
igb_irq_disable(adapter);
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
if (hw->mac.type >= e1000_i350)
igb: Add DMA Coalescing feature to driver This patch add DMA Coalescing which is a power-saving feature that coalesces DMA writes in order to stay in a low-power state as much as possible. Feature is disabled by default. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-03-12 04:43:54 +00:00
adapter->flags &= ~IGB_FLAG_DMAC;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
set_bit(__IGB_DOWN, &adapter->state);
return 0;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_open - Called when a network interface is made active
* @netdev: network interface device structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Returns 0 on success, negative value on failure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* The open entry point is called when a network interface is made
* active by the system (IFF_UP). At this point all resources needed
* for transmit and receive operations are allocated, the interrupt
* handler is registered with the OS, the watchdog timer is started,
* and the stack is notified that the interface is ready.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
static int __igb_open(struct net_device *netdev, bool resuming)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
struct pci_dev *pdev = adapter->pdev;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
int err;
int i;
/* disallow open during test */
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
if (test_bit(__IGB_TESTING, &adapter->state)) {
WARN_ON(resuming);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return -EBUSY;
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
}
if (!resuming)
pm_runtime_get_sync(&pdev->dev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: fix link down inital state As reported by Andrew Lutomirski <amluto@gmail.com> All the intel wired ethernet drivers were calling netif_carrier_off and netif_stop_queue (or variants) before calling register_netdevice This is incorrect behavior as was pointed out by davem, and causes ifconfig and friends to report a strange state before first link after the driver was loaded, since without a netif_carrier_off, the stack assumes carrier_on, but before register_netdev, netlink messages are not sent out telling link state. This apparently confused *some* versions of networkmanager. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Reported-by: Andrew Lutomirski <amluto@gmail.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-17 20:44:32 +00:00
netif_carrier_off(netdev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* allocate transmit descriptors */
err = igb_setup_all_tx_resources(adapter);
if (err)
goto err_setup_tx;
/* allocate receive descriptors */
err = igb_setup_all_rx_resources(adapter);
if (err)
goto err_setup_rx;
igb: Power down link when interface is down This changes the behavior of the driver to power down the link when the associated interface is down, unless management is enabled. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:01:59 +00:00
igb_power_up_link(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* before we allocate an interrupt, we must be ready to handle it.
* Setting DEBUG_SHIRQ in the kernel makes it fire an interrupt
* as soon as we call pci_request_irq, so we have to setup our
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* clean_rx handler before we do so.
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_configure(adapter);
err = igb_request_irq(adapter);
if (err)
goto err_req_irq;
igb: Move the calls to set the Tx and Rx queues into igb_open This change helps to address locking issues seen with netif_set_real_num_tx_queues and netif_set_real_num_rx_queues when used in the igb_set_interrupt_capability function. To resolve these locking issues I have moved the two function calls into __igb_open so that they can be called while the RTNL lock is held. An added advantage to this is that the number of queues is not updated until the last possible moment so if there are any issues in allocating MSI-X interrupts or resources for the rings we have time to change the values prior to updating the netdev. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:22 +00:00
/* Notify the stack of the actual queue counts. */
err = netif_set_real_num_tx_queues(adapter->netdev,
adapter->num_tx_queues);
if (err)
goto err_set_queues;
err = netif_set_real_num_rx_queues(adapter->netdev,
adapter->num_rx_queues);
if (err)
goto err_set_queues;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* From here on the code is the same as igb_up() */
clear_bit(__IGB_DOWN, &adapter->state);
igb: avoid unnecessarily creating a local copy of the q_vector This is mostly a drop of unnecessary pointer defines for q_vector when we don't have issues with line width and don't have multiple references to the pointer. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:34 +00:00
for (i = 0; i < adapter->num_q_vectors; i++)
napi_enable(&(adapter->q_vector[i]->napi));
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Clear any pending interrupts. */
igb: Clear TSICR interrupts together with ICR Issuing "ip link set up/down" can block TSICR interrupts, what results in missing PTP Tx timestamp and no PPS pulse generation. Problem happens when the link is set up with the TSICR interrupts pending. ICR is cleared before enabling interrupts, while TSICR is not. When all TSICR interrupts are pending at this moment, time_sync interrupt will never be generated. TSICR should be cleared as well. In order to reproduce the issue: 1. Setup linux with IEEE 1588 grandmaster and PPS output enabled 2. Continue setting link up/down with random intervals between commands 3. Wait until PPS is not generated ( only one pulse is generated and PPS dies), and ptp4l complains constantly about Tx timeout. Signed-off-by: Joanna Yurdal <jyu@trackman.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-05-16 11:14:11 +00:00
rd32(E1000_TSICR);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
rd32(E1000_ICR);
igb: add NAPI Rx queue support Update the NAPI implementation to use the new napi_struct infrstructure. This removes the need of multiple net_device structs to implement a multiqueue NAPI. Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:39 +00:00
igb_irq_enable(adapter);
igb: delay VF reset notification until after interrupts are enabed This update delays the VF reset notification until after interrupts are enabled. Otherwise there is a chance of having the VF try to reset itself too soon and being ignored by the PF as a result. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:45 +00:00
/* notify VFs that reset has been completed */
if (adapter->vfs_allocated_count) {
u32 reg_data = rd32(E1000_CTRL_EXT);
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: delay VF reset notification until after interrupts are enabed This update delays the VF reset notification until after interrupts are enabled. Otherwise there is a chance of having the VF try to reset itself too soon and being ignored by the PF as a result. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:45 +00:00
reg_data |= E1000_CTRL_EXT_PFRSTD;
wr32(E1000_CTRL_EXT, reg_data);
}
igb/ixgbe/e1000e: resolve tx multiqueue bug With the recent changes to tx mutiqueue, igb/ixgbe/e1000e was not calling netif_tx_start_all_queues() before calling netif_tx_wake_all_queues(). This causes an issue during loading of the driver. In addition, updated e1000e to use the updated tx mutliqueue api. Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-18 11:33:03 +00:00
netif_tx_start_all_queues(netdev);
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
if (!resuming)
pm_runtime_put(&pdev->dev);
igb: cleanup interrupt enablement in regards to msix_other This patch changes a few things so that instead of firing a link status interrupt directly the get_link_status bit is set and the watchdog is scheduled. In addition the mailbox bit is now only enabled if VFs are enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:59 +00:00
/* start the watchdog. */
hw->mac.get_link_status = 1;
schedule_work(&adapter->watchdog_task);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return 0;
igb: Move the calls to set the Tx and Rx queues into igb_open This change helps to address locking issues seen with netif_set_real_num_tx_queues and netif_set_real_num_rx_queues when used in the igb_set_interrupt_capability function. To resolve these locking issues I have moved the two function calls into __igb_open so that they can be called while the RTNL lock is held. An added advantage to this is that the number of queues is not updated until the last possible moment so if there are any issues in allocating MSI-X interrupts or resources for the rings we have time to change the values prior to updating the netdev. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:22 +00:00
err_set_queues:
igb_free_irq(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
err_req_irq:
igb_release_hw_control(adapter);
igb: Power down link when interface is down This changes the behavior of the driver to power down the link when the associated interface is down, unless management is enabled. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:01:59 +00:00
igb_power_down_link(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_free_all_rx_resources(adapter);
err_setup_rx:
igb_free_all_tx_resources(adapter);
err_setup_tx:
igb_reset(adapter);
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
if (!resuming)
pm_runtime_put(&pdev->dev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return err;
}
igb: call ndo_stop() instead of dev_close() when running offline selftest Calling dev_close() causes IFF_UP to be cleared which will remove the interfaces routes and some addresses. That's probably not what the user intended when running the offline selftest. Besides this does not happen if the interface is brought down before the test, so the current behaviour is inconsistent. Instead call the net_device_ops ndo_stop function directly and avoid touching IFF_UP at all. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-02-03 08:20:50 +00:00
int igb_open(struct net_device *netdev)
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
{
return __igb_open(netdev, false);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_close - Disables a network interface
* @netdev: network interface device structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Returns 0, this is not allowed to fail
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* The close entry point is called when an interface is de-activated
* by the OS. The hardware is still under the driver's control, but
* needs to be disabled. A global MAC reset is issued to stop the
* hardware, and all transmit and receive resources are freed.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
static int __igb_close(struct net_device *netdev, bool suspending)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct igb_adapter *adapter = netdev_priv(netdev);
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
struct pci_dev *pdev = adapter->pdev;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
WARN_ON(test_bit(__IGB_RESETTING, &adapter->state));
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
if (!suspending)
pm_runtime_get_sync(&pdev->dev);
igb_down(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_free_irq(adapter);
igb_free_all_tx_resources(adapter);
igb_free_all_rx_resources(adapter);
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
if (!suspending)
pm_runtime_put_sync(&pdev->dev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return 0;
}
igb: call ndo_stop() instead of dev_close() when running offline selftest Calling dev_close() causes IFF_UP to be cleared which will remove the interfaces routes and some addresses. That's probably not what the user intended when running the offline selftest. Besides this does not happen if the interface is brought down before the test, so the current behaviour is inconsistent. Instead call the net_device_ops ndo_stop function directly and avoid touching IFF_UP at all. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-02-03 08:20:50 +00:00
int igb_close(struct net_device *netdev)
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
{
igb: Free IRQs when device is hotplugged Recently I got a Caldigit TS3 Thunderbolt 3 dock, and noticed that upon hotplugging my kernel would immediately crash due to igb: [ 680.825801] kernel BUG at drivers/pci/msi.c:352! [ 680.828388] invalid opcode: 0000 [#1] SMP [ 680.829194] Modules linked in: igb(O) thunderbolt i2c_algo_bit joydev vfat fat btusb btrtl btbcm btintel bluetooth ecdh_generic hp_wmi sparse_keymap rfkill wmi_bmof iTCO_wdt intel_rapl x86_pkg_temp_thermal coretemp crc32_pclmul snd_pcm rtsx_pci_ms mei_me snd_timer memstick snd pcspkr mei soundcore i2c_i801 tpm_tis psmouse shpchp wmi tpm_tis_core tpm video hp_wireless acpi_pad rtsx_pci_sdmmc mmc_core crc32c_intel serio_raw rtsx_pci mfd_core xhci_pci xhci_hcd i2c_hid i2c_core [last unloaded: igb] [ 680.831085] CPU: 1 PID: 78 Comm: kworker/u16:1 Tainted: G O 4.15.0-rc3Lyude-Test+ #6 [ 680.831596] Hardware name: HP HP ZBook Studio G4/826B, BIOS P71 Ver. 01.03 06/09/2017 [ 680.832168] Workqueue: kacpi_hotplug acpi_hotplug_work_fn [ 680.832687] RIP: 0010:free_msi_irqs+0x180/0x1b0 [ 680.833271] RSP: 0018:ffffc9000030fbf0 EFLAGS: 00010286 [ 680.833761] RAX: ffff8803405f9c00 RBX: ffff88033e3d2e40 RCX: 000000000000002c [ 680.834278] RDX: 0000000000000000 RSI: 00000000000000ac RDI: ffff880340be2178 [ 680.834832] RBP: 0000000000000000 R08: ffff880340be1ff0 R09: ffff8803405f9c00 [ 680.835342] R10: 0000000000000000 R11: 0000000000000040 R12: ffff88033d63a298 [ 680.835822] R13: ffff88033d63a000 R14: 0000000000000060 R15: ffff880341959000 [ 680.836332] FS: 0000000000000000(0000) GS:ffff88034f440000(0000) knlGS:0000000000000000 [ 680.836817] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 680.837360] CR2: 000055e64044afdf CR3: 0000000001c09002 CR4: 00000000003606e0 [ 680.837954] Call Trace: [ 680.838853] pci_disable_msix+0xce/0xf0 [ 680.839616] igb_reset_interrupt_capability+0x5d/0x60 [igb] [ 680.840278] igb_remove+0x9d/0x110 [igb] [ 680.840764] pci_device_remove+0x36/0xb0 [ 680.841279] device_release_driver_internal+0x157/0x220 [ 680.841739] pci_stop_bus_device+0x7d/0xa0 [ 680.842255] pci_stop_bus_device+0x2b/0xa0 [ 680.842722] pci_stop_bus_device+0x3d/0xa0 [ 680.843189] pci_stop_and_remove_bus_device+0xe/0x20 [ 680.843627] trim_stale_devices+0xf3/0x140 [ 680.844086] trim_stale_devices+0x94/0x140 [ 680.844532] trim_stale_devices+0xa6/0x140 [ 680.845031] ? get_slot_status+0x90/0xc0 [ 680.845536] acpiphp_check_bridge.part.5+0xfe/0x140 [ 680.846021] acpiphp_hotplug_notify+0x175/0x200 [ 680.846581] ? free_bridge+0x100/0x100 [ 680.847113] acpi_device_hotplug+0x8a/0x490 [ 680.847535] acpi_hotplug_work_fn+0x1a/0x30 [ 680.848076] process_one_work+0x182/0x3a0 [ 680.848543] worker_thread+0x2e/0x380 [ 680.848963] ? process_one_work+0x3a0/0x3a0 [ 680.849373] kthread+0x111/0x130 [ 680.849776] ? kthread_create_worker_on_cpu+0x50/0x50 [ 680.850188] ret_from_fork+0x1f/0x30 [ 680.850601] Code: 43 14 85 c0 0f 84 d5 fe ff ff 31 ed eb 0f 83 c5 01 39 6b 14 0f 86 c5 fe ff ff 8b 7b 10 01 ef e8 b7 e4 d2 ff 48 83 78 70 00 74 e3 <0f> 0b 49 8d b5 a0 00 00 00 e8 62 6f d3 ff e9 c7 fe ff ff 48 8b [ 680.851497] RIP: free_msi_irqs+0x180/0x1b0 RSP: ffffc9000030fbf0 As it turns out, normally the freeing of IRQs that would fix this is called inside of the scope of __igb_close(). However, since the device is already gone by the point we try to unregister the netdevice from the driver due to a hotplug we end up seeing that the netif isn't present and thus, forget to free any of the device IRQs. So: make sure that if we're in the process of dismantling the netdev, we always allow __igb_close() to be called so that IRQs may be freed normally. Additionally, only allow igb_close() to be called from __igb_close() if it hasn't already been called for the given adapter. Signed-off-by: Lyude Paul <lyude@redhat.com> Fixes: 9474933caf21 ("igb: close/suspend race in netif_device_detach") Cc: Todd Fujinaka <todd.fujinaka@intel.com> Cc: Stephen Hemminger <stephen@networkplumber.org> Cc: stable@vger.kernel.org Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-12-12 19:31:30 +00:00
if (netif_device_present(netdev) || netdev->dismantle)
igb: close/suspend race in netif_device_detach Similar to ixgbe, when an interface is part of a namespace it is possible that igb_close() may be called while __igb_shutdown() is running which ends up in a double free WARN and/or a BUG in free_msi_irqs(). Extend the rtnl_lock() to protect the call to netif_device_detach() and igb_clear_interrupt_scheme() in __igb_shutdown() and check for netif_device_present() to avoid calling igb_clear_interrupt_scheme() a second time in igb_close(). Also extend the rtnl lock in igb_resume() to netif_device_attach(). Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Acked-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-11-15 16:54:26 +00:00
return __igb_close(netdev, false);
return 0;
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_setup_tx_resources - allocate Tx resources (Descriptors)
* @tx_ring: tx descriptor ring (for a specific queue) to setup
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Return 0 on success, negative on failure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: add pci device pointer to ring structure This patch adds a pci device pointer to the ring structure. The main use of this pointer is for memory mapping/unmapping of the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:51:47 +00:00
int igb_setup_tx_resources(struct igb_ring *tx_ring)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: convert igb from using PCI DMA functions to using DMA API functions This patch makes it so that igb now uses the DMA API functions instead of the PCI API functions. To do this the pci_dev pointer that was in the rings has been replaced with a device pointer, and as a result all references to [tr]x_ring->pdev have been replaced with [tr]x_ring->dev. This patch is based of of work originally done by Nicholas Nunley. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 13:09:25 +00:00
struct device *dev = tx_ring->dev;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
int size;
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
size = sizeof(struct igb_tx_buffer) * tx_ring->count;
igb: Remove logic that was doing NUMA pseudo-aware allocations This change removes the code that was doing the NUMA allocations for the q_vectors, rings, and ring resources. The problem is the logic used assumed that the NUMA nodes were always interleved and that is not always the case. At some point I hope to add this functionality back in a more controlled manner in the future. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:27:55 +00:00
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
tx_ring->tx_buffer_info = vmalloc(size);
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
if (!tx_ring->tx_buffer_info)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
goto err;
/* round up to nearest 4K */
igb: transition driver to only using advanced descriptors Currently the driver uses advanced descriptors for its main functionality, but then uses legacy when testing. This patch changes this so that advanced descriptors are used throughout and all mentions of legacy descriptors are removed. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-16 08:00:20 +00:00
tx_ring->size = tx_ring->count * sizeof(union e1000_adv_tx_desc);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
tx_ring->size = ALIGN(tx_ring->size, 4096);
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
tx_ring->desc = dma_alloc_coherent(dev, tx_ring->size,
&tx_ring->dma, GFP_KERNEL);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (!tx_ring->desc)
goto err;
tx_ring->next_to_use = 0;
tx_ring->next_to_clean = 0;
igb: Use node specific allocations for the q_vectors and rings This change is meant to update the ring and vector allocations so that they are per node instead of allocating everything on the node that ifconfig/modprobe is called on. By doing this we can cut down significantly on cross node traffic. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:20 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return 0;
err:
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
vfree(tx_ring->tx_buffer_info);
igb: Remove logic that was doing NUMA pseudo-aware allocations This change removes the code that was doing the NUMA allocations for the q_vectors, rings, and ring resources. The problem is the logic used assumed that the NUMA nodes were always interleved and that is not always the case. At some point I hope to add this functionality back in a more controlled manner in the future. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:27:55 +00:00
tx_ring->tx_buffer_info = NULL;
dev_err(dev, "Unable to allocate memory for the Tx descriptor ring\n");
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return -ENOMEM;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_setup_all_tx_resources - wrapper to allocate Tx resources
* (Descriptors) for all queues
* @adapter: board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Return 0 on success, negative on failure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static int igb_setup_all_tx_resources(struct igb_adapter *adapter)
{
igb: cleanup code related to ring resource allocation and free This patch cleans up some of the ring alloc and free code to better handle exceptions such as attempting to free resources on an already freed ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:20 +00:00
struct pci_dev *pdev = adapter->pdev;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
int i, err = 0;
for (i = 0; i < adapter->num_tx_queues; i++) {
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
err = igb_setup_tx_resources(adapter->tx_ring[i]);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (err) {
igb: cleanup code related to ring resource allocation and free This patch cleans up some of the ring alloc and free code to better handle exceptions such as attempting to free resources on an already freed ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:20 +00:00
dev_err(&pdev->dev,
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
"Allocation for Tx Queue %u failed\n", i);
for (i--; i >= 0; i--)
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
igb_free_tx_resources(adapter->tx_ring[i]);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
break;
}
}
return err;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_setup_tctl - configure the transmit control registers
* @adapter: Board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: re-use ring configuration code in ethtool testing Since all of the ring code is now specific to the ring instead of the adapter struct it is possible to cut a large section of code out of the ethtool testing configuraiton since we can just use the existing functions to configure the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:54:23 +00:00
void igb_setup_tctl(struct igb_adapter *adapter)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct e1000_hw *hw = &adapter->hw;
u32 tctl;
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
/* disable queue 0 which is enabled by default on 82575 and 82576 */
wr32(E1000_TXDCTL(0), 0);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Program the Transmit Control Register */
tctl = rd32(E1000_TCTL);
tctl &= ~E1000_TCTL_CT;
tctl |= E1000_TCTL_PSP | E1000_TCTL_RTLC |
(E1000_COLLISION_THRESHOLD << E1000_CT_SHIFT);
igb_config_collision_dist(hw);
/* Enable transmits */
tctl |= E1000_TCTL_EN;
wr32(E1000_TCTL, tctl);
}
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_configure_tx_ring - Configure transmit ring after Reset
* @adapter: board private structure
* @ring: tx ring to configure
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Configure a transmit ring after a reset.
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
**/
igb: re-use ring configuration code in ethtool testing Since all of the ring code is now specific to the ring instead of the adapter struct it is possible to cut a large section of code out of the ethtool testing configuraiton since we can just use the existing functions to configure the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:54:23 +00:00
void igb_configure_tx_ring(struct igb_adapter *adapter,
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
struct igb_ring *ring)
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
{
struct e1000_hw *hw = &adapter->hw;
igb: Update RXDCTL/TXDCTL configurations This change cleans up the RXDCTL and TXDCTL configurations and optimizes RX performance by allowing back write-backs on all hardware other than 82576. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:27 +00:00
u32 txdctl = 0;
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
u64 tdba = ring->dma;
int reg_idx = ring->reg_idx;
wr32(E1000_TDLEN(reg_idx),
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
ring->count * sizeof(union e1000_adv_tx_desc));
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
wr32(E1000_TDBAL(reg_idx),
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
tdba & 0x00000000ffffffffULL);
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
wr32(E1000_TDBAH(reg_idx), tdba >> 32);
igb: use igb_adapter->io_addr instead of e1000_hw->hw_addr When running as guest, under certain condition, it will oops as following. writel() in igb_configure_tx_ring() results in oops, because hw->hw_addr is NULL. While other register access won't oops kernel because they use wr32/rd32 which have a defense against NULL pointer. [ 141.225449] pcieport 0000:00:1c.0: AER: Multiple Uncorrected (Fatal) error received: id=0101 [ 141.225523] igb 0000:01:00.1: PCIe Bus Error: severity=Uncorrected (Fatal), type=Unaccessible, id=0101(Unregistered Agent ID) [ 141.299442] igb 0000:01:00.1: broadcast error_detected message [ 141.300539] igb 0000:01:00.0 enp1s0f0: PCIe link lost, device now detached [ 141.351019] igb 0000:01:00.1 enp1s0f1: PCIe link lost, device now detached [ 143.465904] pcieport 0000:00:1c.0: Root Port link has been reset [ 143.465994] igb 0000:01:00.1: broadcast slot_reset message [ 143.466039] igb 0000:01:00.0: enabling device (0000 -> 0002) [ 144.389078] igb 0000:01:00.1: enabling device (0000 -> 0002) [ 145.312078] igb 0000:01:00.1: broadcast resume message [ 145.322211] BUG: unable to handle kernel paging request at 0000000000003818 [ 145.361275] IP: [<ffffffffa02fd38d>] igb_configure_tx_ring+0x14d/0x280 [igb] [ 145.400048] PGD 0 [ 145.438007] Oops: 0002 [#1] SMP A similar issue & solution could be found at: http://patchwork.ozlabs.org/patch/689592/ Signed-off-by: Cao jin <caoj.fnst@cn.fujitsu.com> Acked-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-11-08 07:06:20 +00:00
ring->tail = adapter->io_addr + E1000_TDT(reg_idx);
igb: Update RXDCTL/TXDCTL configurations This change cleans up the RXDCTL and TXDCTL configurations and optimizes RX performance by allowing back write-backs on all hardware other than 82576. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:27 +00:00
wr32(E1000_TDH(reg_idx), 0);
igb: change the head and tail offsets into pointers Since we are writting to the head/tail pointers frequently we might as well save ourselves some processing time by converting the head and tail offsets directly to pointers. This will shave a few cycles off the rx/tx path and allows us to move one step closer to the rings being a bit more independant of each other. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:51:27 +00:00
writel(0, ring->tail);
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
txdctl |= IGB_TX_PTHRESH;
txdctl |= IGB_TX_HTHRESH << 8;
txdctl |= IGB_TX_WTHRESH << 16;
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
/* reinitialize tx_buffer_info */
memset(ring->tx_buffer_info, 0,
sizeof(struct igb_tx_buffer) * ring->count);
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
txdctl |= E1000_TXDCTL_QUEUE_ENABLE;
wr32(E1000_TXDCTL(reg_idx), txdctl);
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_configure_tx - Configure transmit Unit after Reset
* @adapter: board private structure
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Configure the Tx unit of the MAC after a reset.
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
**/
static void igb_configure_tx(struct igb_adapter *adapter)
{
igb: Wait 10ms just once after TX queues reset Move 10ms sleep out of function resetting TX queue. Reset all the TX queues in one turn and wait for all of them just once. Use usleep_range() instead of mdelay() in order not to affect transmission on other interfaces. Signed-off-by: Sergey Nemov <sergey.nemov@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-05-18 09:58:40 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
int i;
igb: Wait 10ms just once after TX queues reset Move 10ms sleep out of function resetting TX queue. Reset all the TX queues in one turn and wait for all of them just once. Use usleep_range() instead of mdelay() in order not to affect transmission on other interfaces. Signed-off-by: Sergey Nemov <sergey.nemov@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-05-18 09:58:40 +00:00
/* disable the queues */
for (i = 0; i < adapter->num_tx_queues; i++)
wr32(E1000_TXDCTL(adapter->tx_ring[i]->reg_idx), 0);
wrfl();
usleep_range(10000, 20000);
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
for (i = 0; i < adapter->num_tx_queues; i++)
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
igb_configure_tx_ring(adapter, adapter->tx_ring[i]);
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_setup_rx_resources - allocate Rx resources (Descriptors)
* @rx_ring: Rx descriptor ring (for a specific queue) to setup
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Returns 0 on success, negative on failure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: add pci device pointer to ring structure This patch adds a pci device pointer to the ring structure. The main use of this pointer is for memory mapping/unmapping of the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:51:47 +00:00
int igb_setup_rx_resources(struct igb_ring *rx_ring)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: convert igb from using PCI DMA functions to using DMA API functions This patch makes it so that igb now uses the DMA API functions instead of the PCI API functions. To do this the pci_dev pointer that was in the rings has been replaced with a device pointer, and as a result all references to [tr]x_ring->pdev have been replaced with [tr]x_ring->dev. This patch is based of of work originally done by Nicholas Nunley. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 13:09:25 +00:00
struct device *dev = rx_ring->dev;
igb: Remove logic that was doing NUMA pseudo-aware allocations This change removes the code that was doing the NUMA allocations for the q_vectors, rings, and ring resources. The problem is the logic used assumed that the NUMA nodes were always interleved and that is not always the case. At some point I hope to add this functionality back in a more controlled manner in the future. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:27:55 +00:00
int size;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
size = sizeof(struct igb_rx_buffer) * rx_ring->count;
igb: Remove logic that was doing NUMA pseudo-aware allocations This change removes the code that was doing the NUMA allocations for the q_vectors, rings, and ring resources. The problem is the logic used assumed that the NUMA nodes were always interleved and that is not always the case. At some point I hope to add this functionality back in a more controlled manner in the future. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:27:55 +00:00
igb: Clear Rx buffer_info in configure instead of clean This change makes it so that instead of going through the entire ring on Rx cleanup we only go through the region that was designated to be cleaned up and stop when we reach the region where new allocations should start. In addition we can avoid having to perform a memset on the Rx buffer_info structures until we are about to start using the ring again. By deferring this we can avoid dirtying the cache any more than we have to which can help to improve the time needed to bring the interface down and then back up again in a reset or suspend/resume cycle. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:25:50 +00:00
rx_ring->rx_buffer_info = vmalloc(size);
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
if (!rx_ring->rx_buffer_info)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
goto err;
/* Round up to nearest 4K */
igb: Remove logic that was doing NUMA pseudo-aware allocations This change removes the code that was doing the NUMA allocations for the q_vectors, rings, and ring resources. The problem is the logic used assumed that the NUMA nodes were always interleved and that is not always the case. At some point I hope to add this functionality back in a more controlled manner in the future. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:27:55 +00:00
rx_ring->size = rx_ring->count * sizeof(union e1000_adv_rx_desc);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
rx_ring->size = ALIGN(rx_ring->size, 4096);
igb: Combine q_vector and ring allocation into a single function This change combines the the allocation of q_vectors and rings into a single function. The advantage of this is that we are guaranteed we will avoid overlap in the L1 cache sets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:17 +00:00
rx_ring->desc = dma_alloc_coherent(dev, rx_ring->size,
&rx_ring->dma, GFP_KERNEL);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (!rx_ring->desc)
goto err;
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
rx_ring->next_to_alloc = 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
rx_ring->next_to_clean = 0;
rx_ring->next_to_use = 0;
return 0;
err:
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
vfree(rx_ring->rx_buffer_info);
rx_ring->rx_buffer_info = NULL;
igb: Remove logic that was doing NUMA pseudo-aware allocations This change removes the code that was doing the NUMA allocations for the q_vectors, rings, and ring resources. The problem is the logic used assumed that the NUMA nodes were always interleved and that is not always the case. At some point I hope to add this functionality back in a more controlled manner in the future. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:27:55 +00:00
dev_err(dev, "Unable to allocate memory for the Rx descriptor ring\n");
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return -ENOMEM;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_setup_all_rx_resources - wrapper to allocate Rx resources
* (Descriptors) for all queues
* @adapter: board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Return 0 on success, negative on failure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static int igb_setup_all_rx_resources(struct igb_adapter *adapter)
{
igb: cleanup code related to ring resource allocation and free This patch cleans up some of the ring alloc and free code to better handle exceptions such as attempting to free resources on an already freed ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:20 +00:00
struct pci_dev *pdev = adapter->pdev;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
int i, err = 0;
for (i = 0; i < adapter->num_rx_queues; i++) {
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
err = igb_setup_rx_resources(adapter->rx_ring[i]);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (err) {
igb: cleanup code related to ring resource allocation and free This patch cleans up some of the ring alloc and free code to better handle exceptions such as attempting to free resources on an already freed ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:20 +00:00
dev_err(&pdev->dev,
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
"Allocation for Rx Queue %u failed\n", i);
for (i--; i >= 0; i--)
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
igb_free_rx_resources(adapter->rx_ring[i]);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
break;
}
}
return err;
}
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_setup_mrqc - configure the multiple receive queue control registers
* @adapter: Board private structure
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
**/
static void igb_setup_mrqc(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
u32 mrqc, rxcsum;
igb: Expose RSS indirection table for ethtool This patch adds the ethtool callbacks necessary to change the RETA indirection table from userspace. In order to achieve this, we add the indirection table field (rss_indir_tbl) in the board specific data structure (struct igb_adapter) to preserve the values across hardware resets. The indirection table must be initialized with default values in the following cases: * at module init time * when the number of RX queues changes. For this reason we add a new field (rss_indir_tbl_init) in igb_adapter that keeps track of the number of RX queues. Whenever the number of RX queues changes, the rss_indir_tbl is modified and initialized with default values. The rss_indir_tbl_init is updated accordingly. CC: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: Laura Mihaela Vasilescu <laura.vasilescu@rosedu.org> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-31 20:19:54 +00:00
u32 j, num_rx_queues;
igb: use netdev_rss_key_fill() helper Use of well known RSS key increases attack surface. Switch to a random one, using generic helper so that all ports share a common key. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-16 14:23:14 +00:00
u32 rss_key[10];
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
igb: use netdev_rss_key_fill() helper Use of well known RSS key increases attack surface. Switch to a random one, using generic helper so that all ports share a common key. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-16 14:23:14 +00:00
netdev_rss_key_fill(rss_key, sizeof(rss_key));
igb: Simplify how we populate the RSS key Instead of storing the RSS key as a character array we can simplify the configuration by making it a u32 array. This allows us to just write one value per register without any unnecessary operations to construct the value. This change will produce the same exact key, the only difference is that I translated the u8 array to a u32 array which will be correctly ordered on writes to hardware by the cpu_to_le32 operations that are built into the writel calls. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:28:16 +00:00
for (j = 0; j < 10; j++)
igb: use netdev_rss_key_fill() helper Use of well known RSS key increases attack surface. Switch to a random one, using generic helper so that all ports share a common key. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-16 14:23:14 +00:00
wr32(E1000_RSSRK(j), rss_key[j]);
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
igb: when number of CPUs > 4 combine tx/rx queues to allow more queues This patch makes it so that nics such as 82576 and newer can support more hardware queues when there are more than 4 cpus by combining a tx/rx queue pair onto one interrupt so that 8 queue pairs can be supported and thus allow for more queues. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:19 +00:00
num_rx_queues = adapter->rss_queues;
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
igb: Change how we populate the RSS indirection table This patch cleans up our RSS indirection table configuration so that we generate the same table regardless of CPU endianness. In addition it changes the table setup so that instead of doing a modulo based setup it is instead a divisor based setup. The advantage to this is that we should be able to take the Rx hash and compute the Rx queue with very little CPU overhead if needed. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:28:11 +00:00
switch (hw->mac.type) {
case e1000_82576:
/* 82576 supports 2 RSS queues for SR-IOV */
igb: Expose RSS indirection table for ethtool This patch adds the ethtool callbacks necessary to change the RETA indirection table from userspace. In order to achieve this, we add the indirection table field (rss_indir_tbl) in the board specific data structure (struct igb_adapter) to preserve the values across hardware resets. The indirection table must be initialized with default values in the following cases: * at module init time * when the number of RX queues changes. For this reason we add a new field (rss_indir_tbl_init) in igb_adapter that keeps track of the number of RX queues. Whenever the number of RX queues changes, the rss_indir_tbl is modified and initialized with default values. The rss_indir_tbl_init is updated accordingly. CC: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: Laura Mihaela Vasilescu <laura.vasilescu@rosedu.org> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-31 20:19:54 +00:00
if (adapter->vfs_allocated_count)
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
num_rx_queues = 2;
igb: Change how we populate the RSS indirection table This patch cleans up our RSS indirection table configuration so that we generate the same table regardless of CPU endianness. In addition it changes the table setup so that instead of doing a modulo based setup it is instead a divisor based setup. The advantage to this is that we should be able to take the Rx hash and compute the Rx queue with very little CPU overhead if needed. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:28:11 +00:00
break;
default:
break;
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
}
igb: Expose RSS indirection table for ethtool This patch adds the ethtool callbacks necessary to change the RETA indirection table from userspace. In order to achieve this, we add the indirection table field (rss_indir_tbl) in the board specific data structure (struct igb_adapter) to preserve the values across hardware resets. The indirection table must be initialized with default values in the following cases: * at module init time * when the number of RX queues changes. For this reason we add a new field (rss_indir_tbl_init) in igb_adapter that keeps track of the number of RX queues. Whenever the number of RX queues changes, the rss_indir_tbl is modified and initialized with default values. The rss_indir_tbl_init is updated accordingly. CC: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: Laura Mihaela Vasilescu <laura.vasilescu@rosedu.org> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-31 20:19:54 +00:00
if (adapter->rss_indir_tbl_init != num_rx_queues) {
for (j = 0; j < IGB_RETA_SIZE; j++)
igb: Cleanups to fix line length warnings This patch fixes WARNING:LONG_LINE found with checkpatch check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:46:33 +00:00
adapter->rss_indir_tbl[j] =
(j * num_rx_queues) / IGB_RETA_SIZE;
igb: Expose RSS indirection table for ethtool This patch adds the ethtool callbacks necessary to change the RETA indirection table from userspace. In order to achieve this, we add the indirection table field (rss_indir_tbl) in the board specific data structure (struct igb_adapter) to preserve the values across hardware resets. The indirection table must be initialized with default values in the following cases: * at module init time * when the number of RX queues changes. For this reason we add a new field (rss_indir_tbl_init) in igb_adapter that keeps track of the number of RX queues. Whenever the number of RX queues changes, the rss_indir_tbl is modified and initialized with default values. The rss_indir_tbl_init is updated accordingly. CC: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: Laura Mihaela Vasilescu <laura.vasilescu@rosedu.org> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-31 20:19:54 +00:00
adapter->rss_indir_tbl_init = num_rx_queues;
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
}
igb: Expose RSS indirection table for ethtool This patch adds the ethtool callbacks necessary to change the RETA indirection table from userspace. In order to achieve this, we add the indirection table field (rss_indir_tbl) in the board specific data structure (struct igb_adapter) to preserve the values across hardware resets. The indirection table must be initialized with default values in the following cases: * at module init time * when the number of RX queues changes. For this reason we add a new field (rss_indir_tbl_init) in igb_adapter that keeps track of the number of RX queues. Whenever the number of RX queues changes, the rss_indir_tbl is modified and initialized with default values. The rss_indir_tbl_init is updated accordingly. CC: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: Laura Mihaela Vasilescu <laura.vasilescu@rosedu.org> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-31 20:19:54 +00:00
igb_write_rss_indir_tbl(adapter);
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Disable raw packet checksumming so that RSS hash is placed in
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
* descriptor on writeback. No need to enable TCP/UDP/IP checksum
* offloads as they are enabled by default
*/
rxcsum = rd32(E1000_RXCSUM);
rxcsum |= E1000_RXCSUM_PCSD;
if (adapter->hw.mac.type >= e1000_82576)
/* Enable Receive Checksum Offload for SCTP */
rxcsum |= E1000_RXCSUM_CRCOFL;
/* Don't need to set TUOFL or IPOFL, they default to 1 */
wr32(E1000_RXCSUM, rxcsum);
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
igb: Support for modifying UDP RSS flow hashing This patch provides ability to enable or disable UDP RSS hashing. It gives users option of generating RSS hash based on the UDP source and destination ports numbers. Currently, UDP flow hash is always disabled in igb-driver. Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:21 +00:00
/* Generate RSS hash based on packet types, TCP/UDP
* port numbers and/or IPv4/v6 src and dst addresses
*/
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
mrqc = E1000_MRQC_RSS_FIELD_IPV4 |
E1000_MRQC_RSS_FIELD_IPV4_TCP |
E1000_MRQC_RSS_FIELD_IPV6 |
E1000_MRQC_RSS_FIELD_IPV6_TCP |
E1000_MRQC_RSS_FIELD_IPV6_TCP_EX;
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
igb: Support for modifying UDP RSS flow hashing This patch provides ability to enable or disable UDP RSS hashing. It gives users option of generating RSS hash based on the UDP source and destination ports numbers. Currently, UDP flow hash is always disabled in igb-driver. Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:21 +00:00
if (adapter->flags & IGB_FLAG_RSS_FIELD_IPV4_UDP)
mrqc |= E1000_MRQC_RSS_FIELD_IPV4_UDP;
if (adapter->flags & IGB_FLAG_RSS_FIELD_IPV6_UDP)
mrqc |= E1000_MRQC_RSS_FIELD_IPV6_UDP;
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
/* If VMDq is enabled then we set the appropriate mode for that, else
* we default to RSS so that an RSS hash is calculated per packet even
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* if we are only using one queue
*/
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
if (adapter->vfs_allocated_count) {
if (hw->mac.type > e1000_82575) {
/* Set the default pool for the PF's first queue */
u32 vtctl = rd32(E1000_VT_CTL);
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
vtctl &= ~(E1000_VT_CTL_DEFAULT_POOL_MASK |
E1000_VT_CTL_DISABLE_DEF_POOL);
vtctl |= adapter->vfs_allocated_count <<
E1000_VT_CTL_DEFAULT_POOL_SHIFT;
wr32(E1000_VT_CTL, vtctl);
}
igb: when number of CPUs > 4 combine tx/rx queues to allow more queues This patch makes it so that nics such as 82576 and newer can support more hardware queues when there are more than 4 cpus by combining a tx/rx queue pair onto one interrupt so that 8 queue pairs can be supported and thus allow for more queues. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:19 +00:00
if (adapter->rss_queues > 1)
igb: rename igb define to be more generic E1000_MRQC_ENABLE_RSS_4Q enables 4 and 8 queues depending on the part so rename to be generic. Similarly, E1000_MRQC_ENABLE_VMDQ_RSS_2Q has no numeric meaning so rename to be more generic. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-11 17:34:50 +00:00
mrqc |= E1000_MRQC_ENABLE_VMDQ_RSS_MQ;
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
else
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
mrqc |= E1000_MRQC_ENABLE_VMDQ;
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
} else {
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
if (hw->mac.type != e1000_i211)
igb: rename igb define to be more generic E1000_MRQC_ENABLE_RSS_4Q enables 4 and 8 queues depending on the part so rename to be generic. Similarly, E1000_MRQC_ENABLE_VMDQ_RSS_2Q has no numeric meaning so rename to be more generic. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-11 17:34:50 +00:00
mrqc |= E1000_MRQC_ENABLE_RSS_MQ;
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
}
igb_vmm_control(adapter);
wr32(E1000_MRQC, mrqc);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_setup_rctl - configure the receive control registers
* @adapter: Board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: re-use ring configuration code in ethtool testing Since all of the ring code is now specific to the ring instead of the adapter struct it is possible to cut a large section of code out of the ethtool testing configuraiton since we can just use the existing functions to configure the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:54:23 +00:00
void igb_setup_rctl(struct igb_adapter *adapter)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct e1000_hw *hw = &adapter->hw;
u32 rctl;
rctl = rd32(E1000_RCTL);
rctl &= ~(3 << E1000_RCTL_MO_SHIFT);
igb: loopback bits not correctly cleared from RCTL register This change forces the bits to 0 by using an &= operation with an inverted mask of all options instead of using an |= with a value of 0. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-25 09:04:03 +00:00
rctl &= ~(E1000_RCTL_LBM_TCVR | E1000_RCTL_LBM_MAC);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: loopback bits not correctly cleared from RCTL register This change forces the bits to 0 by using an &= operation with an inverted mask of all options instead of using an |= with a value of 0. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-25 09:04:03 +00:00
rctl |= E1000_RCTL_EN | E1000_RCTL_BAM | E1000_RCTL_RDMTS_HALF |
igb: remove unnecessary adapter->hw calls when just hw-> will do. There were several spots in the code making calls to adapter->hw when they could have just been accessing hw-> directly. I cleaned up the spots where this was visibly apparent. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:20:31 +00:00
(hw->mac.mc_filter_type << E1000_RCTL_MO_SHIFT);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* enable stripping of CRC. It's unlikely this will break BMC
igb: reenable CRC stripping in hardware We can remove a clunky workaround for not having the hardware strip the CRC. 82575 silicon as well as the older PCI Express e1000e hardware all work OK in this respect. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:08:29 +00:00
* redirection as it did with e1000. Newer features require
* that the HW strips the CRC.
igb: misc whitespace/formatting cleanups This patch is intended to hold several whitespace, formatting, and comment cleanups that have been found while cleaning up the igb driver. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-12 18:16:59 +00:00
*/
igb: reenable CRC stripping in hardware We can remove a clunky workaround for not having the hardware strip the CRC. 82575 silicon as well as the older PCI Express e1000e hardware all work OK in this respect. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:08:29 +00:00
rctl |= E1000_RCTL_SECRC;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: cleanup whitespace issues in igb_main.c This patch goes through and cleans up whitespace issues in igb_main.c to help improve readability. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:52:50 +00:00
/* disable store bad packets and clear size bits. */
igb: prevent skb_over panic w/ mtu smaller than 1K A panic has been observed with frame sizes smaller than 1K. This has been root caused to the hardware spanning larger frames across multiple buffers and then reporting the original frame size in the first descriptor. To prevent this we can enable set the LPE bit which in turn will restrict packet sizes to those set in the RLPML register. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-31 08:52:57 +00:00
rctl &= ~(E1000_RCTL_SBP | E1000_RCTL_SZ_256);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
/* enable LPE to allow for reception of jumbo frames */
igb: remove rx_ps_hdr_len This patch removes the rx_ps_hdr_len which isn't really needed since we can now use rx_buffer_len less than 1K to indicate that we are in a packet split mode. We also don't need it since we always use a half page for the data buffers when receiving so we always know the size to map/unmap. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:48 +00:00
rctl |= E1000_RCTL_LPE;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: move SRRCTL register configuration into ring specific config The SRRCTL register exists per ring. Instead of configuring all of them in the RCTL configuration which is meant to be global it makes more sense to move this out into the ring specific configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:51:07 +00:00
/* disable queue 0 to prevent tail write w/o re-config */
wr32(E1000_RXDCTL(0), 0);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add pf side of VMDq support Add the pf portion of vmdq support. This provides enough support so that VMDq is enabled, and the pf is functional without enabling vfs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:44 +00:00
/* Attention!!! For SR-IOV PF driver operations you must enable
* queue drop for all VF and PF queues to prevent head of line blocking
* if an un-trusted VF does not provide descriptors to hardware.
*/
if (adapter->vfs_allocated_count) {
/* set all queue drop enable bits */
wr32(E1000_QDE, ALL_QUEUES);
}
igb: Support RX-ALL feature flag. This allows the NIC to receive all frames available, including those with bad FCS, un-matched vlans, ethernet control frames, and more. Tested by sending frames with bad FCS. Signed-off-by: Ben Greear <greearb@candelatech.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-03-06 09:41:58 +00:00
/* This is useful for sniffing bad packets. */
if (adapter->netdev->features & NETIF_F_RXALL) {
/* UPE and MPE will be handled by normal PROMISC logic
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* in e1000e_set_rx_mode
*/
igb: Support RX-ALL feature flag. This allows the NIC to receive all frames available, including those with bad FCS, un-matched vlans, ethernet control frames, and more. Tested by sending frames with bad FCS. Signed-off-by: Ben Greear <greearb@candelatech.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-03-06 09:41:58 +00:00
rctl |= (E1000_RCTL_SBP | /* Receive bad packets */
E1000_RCTL_BAM | /* RX All Bcast Pkts */
E1000_RCTL_PMCF); /* RX All MAC Ctrl Pkts */
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
rctl &= ~(E1000_RCTL_DPF | /* Allow filtered pause */
igb: Support RX-ALL feature flag. This allows the NIC to receive all frames available, including those with bad FCS, un-matched vlans, ethernet control frames, and more. Tested by sending frames with bad FCS. Signed-off-by: Ben Greear <greearb@candelatech.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-03-06 09:41:58 +00:00
E1000_RCTL_CFIEN); /* Dis VLAN CFIEN Filter */
/* Do not mess with E1000_CTRL_VME, it affects transmit as well,
* and that breaks VLANs.
*/
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
wr32(E1000_RCTL, rctl);
}
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
static inline int igb_set_vf_rlpml(struct igb_adapter *adapter, int size,
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
int vfn)
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
{
struct e1000_hw *hw = &adapter->hw;
u32 vmolr;
igb: Do not factor VLANs into RLPML calculation The RLPML registers already take the size of VLAN headers into account when determining the maximum packet length. This is called out in EAS documents for several parts including the 82576 and the i350. As such we can drop the addition of size to the value programmed into the RLPML registers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:47 +00:00
if (size > MAX_JUMBO_FRAME_SIZE)
size = MAX_JUMBO_FRAME_SIZE;
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
vmolr = rd32(E1000_VMOLR(vfn));
vmolr &= ~E1000_VMOLR_RLPML_MASK;
vmolr |= size | E1000_VMOLR_LPE;
wr32(E1000_VMOLR(vfn), vmolr);
return 0;
}
igb: Fix VLAN tag stripping on Intel i350 Problem: When switching off VLAN offloading on an i350, the VLAN interface gets unusable. For testing, set up a VLAN on an i350 and some remote machine, e.g.: $ ip link add link eth0 name eth0.42 type vlan id 42 $ ip addr add 192.168.42.1/24 dev eth0.42 $ ip link set dev eth0.42 up Offloading is switched on by default: $ ethtool -k eth0 | grep vlan-offload rx-vlan-offload: on tx-vlan-offload: on $ ping -c 3 -I eth0.42 192.168.42.2 [...works as usual...] Now switch off VLAN offloading and try again: $ ethtool -K eth0 rxvlan off Actual changes: rx-vlan-offload: off tx-vlan-offload: off [requested on] $ ping -c 3 -I eth0.42 192.168.42.2 PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da ta. --- 192.168.42.2 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms I can only reproduce it on an i350, the above works fine on a 82580. While inspecting the igb source, I came across the code in igb_set_vmolr which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and for all, and in all of the igb code there's no other place where the STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb unconditionally, independently of the offloading setting. I compared that to the latest Intel igb-5.3.3.5 driver from http://sourceforge.net/projects/e1000/ which in fact sets and clears the STRVLAN flag independently from igb_set_vmolr in its own function igb_set_vf_vlan_strip, depending on the vlan settings. So I included the STRVLAN handling from the igb-5.3.3.5 driver into our current igb driver and tested the above scenario again. This time ping still works after switching off VLAN offloading. Tested on i350, with and without addtional VFs, as well as on 82580 successfully. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-28 12:53:23 +00:00
static inline void igb_set_vf_vlan_strip(struct igb_adapter *adapter,
int vfn, bool enable)
igb: add pf side of VMDq support Add the pf portion of vmdq support. This provides enough support so that VMDq is enabled, and the pf is functional without enabling vfs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:44 +00:00
{
struct e1000_hw *hw = &adapter->hw;
igb: Fix VLAN tag stripping on Intel i350 Problem: When switching off VLAN offloading on an i350, the VLAN interface gets unusable. For testing, set up a VLAN on an i350 and some remote machine, e.g.: $ ip link add link eth0 name eth0.42 type vlan id 42 $ ip addr add 192.168.42.1/24 dev eth0.42 $ ip link set dev eth0.42 up Offloading is switched on by default: $ ethtool -k eth0 | grep vlan-offload rx-vlan-offload: on tx-vlan-offload: on $ ping -c 3 -I eth0.42 192.168.42.2 [...works as usual...] Now switch off VLAN offloading and try again: $ ethtool -K eth0 rxvlan off Actual changes: rx-vlan-offload: off tx-vlan-offload: off [requested on] $ ping -c 3 -I eth0.42 192.168.42.2 PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da ta. --- 192.168.42.2 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms I can only reproduce it on an i350, the above works fine on a 82580. While inspecting the igb source, I came across the code in igb_set_vmolr which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and for all, and in all of the igb code there's no other place where the STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb unconditionally, independently of the offloading setting. I compared that to the latest Intel igb-5.3.3.5 driver from http://sourceforge.net/projects/e1000/ which in fact sets and clears the STRVLAN flag independently from igb_set_vmolr in its own function igb_set_vf_vlan_strip, depending on the vlan settings. So I included the STRVLAN handling from the igb-5.3.3.5 driver into our current igb driver and tested the above scenario again. This time ping still works after switching off VLAN offloading. Tested on i350, with and without addtional VFs, as well as on 82580 successfully. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-28 12:53:23 +00:00
u32 val, reg;
igb: add pf side of VMDq support Add the pf portion of vmdq support. This provides enough support so that VMDq is enabled, and the pf is functional without enabling vfs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:44 +00:00
igb: Fix VLAN tag stripping on Intel i350 Problem: When switching off VLAN offloading on an i350, the VLAN interface gets unusable. For testing, set up a VLAN on an i350 and some remote machine, e.g.: $ ip link add link eth0 name eth0.42 type vlan id 42 $ ip addr add 192.168.42.1/24 dev eth0.42 $ ip link set dev eth0.42 up Offloading is switched on by default: $ ethtool -k eth0 | grep vlan-offload rx-vlan-offload: on tx-vlan-offload: on $ ping -c 3 -I eth0.42 192.168.42.2 [...works as usual...] Now switch off VLAN offloading and try again: $ ethtool -K eth0 rxvlan off Actual changes: rx-vlan-offload: off tx-vlan-offload: off [requested on] $ ping -c 3 -I eth0.42 192.168.42.2 PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da ta. --- 192.168.42.2 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms I can only reproduce it on an i350, the above works fine on a 82580. While inspecting the igb source, I came across the code in igb_set_vmolr which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and for all, and in all of the igb code there's no other place where the STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb unconditionally, independently of the offloading setting. I compared that to the latest Intel igb-5.3.3.5 driver from http://sourceforge.net/projects/e1000/ which in fact sets and clears the STRVLAN flag independently from igb_set_vmolr in its own function igb_set_vf_vlan_strip, depending on the vlan settings. So I included the STRVLAN handling from the igb-5.3.3.5 driver into our current igb driver and tested the above scenario again. This time ping still works after switching off VLAN offloading. Tested on i350, with and without addtional VFs, as well as on 82580 successfully. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-28 12:53:23 +00:00
if (hw->mac.type < e1000_82576)
return;
igb: add pf side of VMDq support Add the pf portion of vmdq support. This provides enough support so that VMDq is enabled, and the pf is functional without enabling vfs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:44 +00:00
igb: Fix VLAN tag stripping on Intel i350 Problem: When switching off VLAN offloading on an i350, the VLAN interface gets unusable. For testing, set up a VLAN on an i350 and some remote machine, e.g.: $ ip link add link eth0 name eth0.42 type vlan id 42 $ ip addr add 192.168.42.1/24 dev eth0.42 $ ip link set dev eth0.42 up Offloading is switched on by default: $ ethtool -k eth0 | grep vlan-offload rx-vlan-offload: on tx-vlan-offload: on $ ping -c 3 -I eth0.42 192.168.42.2 [...works as usual...] Now switch off VLAN offloading and try again: $ ethtool -K eth0 rxvlan off Actual changes: rx-vlan-offload: off tx-vlan-offload: off [requested on] $ ping -c 3 -I eth0.42 192.168.42.2 PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da ta. --- 192.168.42.2 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms I can only reproduce it on an i350, the above works fine on a 82580. While inspecting the igb source, I came across the code in igb_set_vmolr which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and for all, and in all of the igb code there's no other place where the STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb unconditionally, independently of the offloading setting. I compared that to the latest Intel igb-5.3.3.5 driver from http://sourceforge.net/projects/e1000/ which in fact sets and clears the STRVLAN flag independently from igb_set_vmolr in its own function igb_set_vf_vlan_strip, depending on the vlan settings. So I included the STRVLAN handling from the igb-5.3.3.5 driver into our current igb driver and tested the above scenario again. This time ping still works after switching off VLAN offloading. Tested on i350, with and without addtional VFs, as well as on 82580 successfully. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-28 12:53:23 +00:00
if (hw->mac.type == e1000_i350)
reg = E1000_DVMOLR(vfn);
else
reg = E1000_VMOLR(vfn);
val = rd32(reg);
if (enable)
val |= E1000_VMOLR_STRVLAN;
else
val &= ~(E1000_VMOLR_STRVLAN);
wr32(reg, val);
igb: add pf side of VMDq support Add the pf portion of vmdq support. This provides enough support so that VMDq is enabled, and the pf is functional without enabling vfs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:44 +00:00
}
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
static inline void igb_set_vmolr(struct igb_adapter *adapter,
int vfn, bool aupe)
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
{
struct e1000_hw *hw = &adapter->hw;
u32 vmolr;
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* This register exists only on 82576 and newer so if we are older then
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
* we should exit and do nothing
*/
if (hw->mac.type < e1000_82576)
return;
vmolr = rd32(E1000_VMOLR(vfn));
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
if (aupe)
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
vmolr |= E1000_VMOLR_AUPE; /* Accept untagged packets */
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
else
vmolr &= ~(E1000_VMOLR_AUPE); /* Tagged packets ONLY */
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
/* clear all bits that might not be set */
vmolr &= ~(E1000_VMOLR_BAM | E1000_VMOLR_RSSE);
igb: when number of CPUs > 4 combine tx/rx queues to allow more queues This patch makes it so that nics such as 82576 and newer can support more hardware queues when there are more than 4 cpus by combining a tx/rx queue pair onto one interrupt so that 8 queue pairs can be supported and thus allow for more queues. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:19 +00:00
if (adapter->rss_queues > 1 && vfn == adapter->vfs_allocated_count)
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
vmolr |= E1000_VMOLR_RSSE; /* enable RSS */
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* for VMDq only allow the VFs and pool 0 to accept broadcast and
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
* multicast packets
*/
if (vfn <= adapter->vfs_allocated_count)
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
vmolr |= E1000_VMOLR_BAM; /* Accept broadcast */
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
wr32(E1000_VMOLR(vfn), vmolr);
}
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_configure_rx_ring - Configure a receive ring after Reset
* @adapter: board private structure
* @ring: receive ring to be configured
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Configure the Rx unit of the MAC after a reset.
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
**/
igb: re-use ring configuration code in ethtool testing Since all of the ring code is now specific to the ring instead of the adapter struct it is possible to cut a large section of code out of the ethtool testing configuraiton since we can just use the existing functions to configure the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:54:23 +00:00
void igb_configure_rx_ring(struct igb_adapter *adapter,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
struct igb_ring *ring)
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
{
struct e1000_hw *hw = &adapter->hw;
igb: Use length to determine if descriptor is done This change makes it so that we use the length of the packet instead of the DD status bit to determine if a new descriptor is ready to be processed. The obvious advantage is that it cuts down on reads as we don't really even need the DD bit if going from a 0 to a non-zero value on size is enough to inform us that the packet has been completed. In addition I have updated the code so that we only reset the Rx descriptor length for descriptor zero when resetting a ring instead of having to do a memset with 0 over the entire ring. By doing this we can save some time on initialization. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:25:41 +00:00
union e1000_adv_rx_desc *rx_desc;
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
u64 rdba = ring->dma;
int reg_idx = ring->reg_idx;
igb: Update RXDCTL/TXDCTL configurations This change cleans up the RXDCTL and TXDCTL configurations and optimizes RX performance by allowing back write-backs on all hardware other than 82576. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:27 +00:00
u32 srrctl = 0, rxdctl = 0;
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
/* disable the queue */
igb: Update RXDCTL/TXDCTL configurations This change cleans up the RXDCTL and TXDCTL configurations and optimizes RX performance by allowing back write-backs on all hardware other than 82576. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:27 +00:00
wr32(E1000_RXDCTL(reg_idx), 0);
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
/* Set DMA base address registers */
wr32(E1000_RDBAL(reg_idx),
rdba & 0x00000000ffffffffULL);
wr32(E1000_RDBAH(reg_idx), rdba >> 32);
wr32(E1000_RDLEN(reg_idx),
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
ring->count * sizeof(union e1000_adv_rx_desc));
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
/* initialize head and tail */
igb: use igb_adapter->io_addr instead of e1000_hw->hw_addr When running as guest, under certain condition, it will oops as following. writel() in igb_configure_tx_ring() results in oops, because hw->hw_addr is NULL. While other register access won't oops kernel because they use wr32/rd32 which have a defense against NULL pointer. [ 141.225449] pcieport 0000:00:1c.0: AER: Multiple Uncorrected (Fatal) error received: id=0101 [ 141.225523] igb 0000:01:00.1: PCIe Bus Error: severity=Uncorrected (Fatal), type=Unaccessible, id=0101(Unregistered Agent ID) [ 141.299442] igb 0000:01:00.1: broadcast error_detected message [ 141.300539] igb 0000:01:00.0 enp1s0f0: PCIe link lost, device now detached [ 141.351019] igb 0000:01:00.1 enp1s0f1: PCIe link lost, device now detached [ 143.465904] pcieport 0000:00:1c.0: Root Port link has been reset [ 143.465994] igb 0000:01:00.1: broadcast slot_reset message [ 143.466039] igb 0000:01:00.0: enabling device (0000 -> 0002) [ 144.389078] igb 0000:01:00.1: enabling device (0000 -> 0002) [ 145.312078] igb 0000:01:00.1: broadcast resume message [ 145.322211] BUG: unable to handle kernel paging request at 0000000000003818 [ 145.361275] IP: [<ffffffffa02fd38d>] igb_configure_tx_ring+0x14d/0x280 [igb] [ 145.400048] PGD 0 [ 145.438007] Oops: 0002 [#1] SMP A similar issue & solution could be found at: http://patchwork.ozlabs.org/patch/689592/ Signed-off-by: Cao jin <caoj.fnst@cn.fujitsu.com> Acked-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-11-08 07:06:20 +00:00
ring->tail = adapter->io_addr + E1000_RDT(reg_idx);
igb: Update RXDCTL/TXDCTL configurations This change cleans up the RXDCTL and TXDCTL configurations and optimizes RX performance by allowing back write-backs on all hardware other than 82576. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:27 +00:00
wr32(E1000_RDH(reg_idx), 0);
igb: change the head and tail offsets into pointers Since we are writting to the head/tail pointers frequently we might as well save ourselves some processing time by converting the head and tail offsets directly to pointers. This will shave a few cycles off the rx/tx path and allows us to move one step closer to the rings being a bit more independant of each other. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:51:27 +00:00
writel(0, ring->tail);
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
igb: move SRRCTL register configuration into ring specific config The SRRCTL register exists per ring. Instead of configuring all of them in the RCTL configuration which is meant to be global it makes more sense to move this out into the ring specific configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:51:07 +00:00
/* set descriptor configuration */
igb: drop support for single buffer mode This change removes support for single buffer mode from igb and makes the driver function in packet split always. The advantage to doing this is that we can reduce total memory allocation overhead significantly as we will only need to allocate one 1K slab per packet and then make use of a reusable half page instead of allocating a 2K slab per packet. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:38 +00:00
srrctl = IGB_RX_HDR_LEN << E1000_SRRCTL_BSIZEHDRSIZE_SHIFT;
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
if (ring_uses_large_buffer(ring))
srrctl |= IGB_RXBUFFER_3072 >> E1000_SRRCTL_BSIZEPKT_SHIFT;
else
srrctl |= IGB_RXBUFFER_2048 >> E1000_SRRCTL_BSIZEPKT_SHIFT;
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
srrctl |= E1000_SRRCTL_DESCTYPE_ADV_ONEBUF;
igb: Fix features that are currently 82580 only and should also be i350 This change allows support for per packet timesync and global device reset on the i350 adapter. These features were supported on both 82580 and i350 however it looks like several checks where not updated and as such the i350 support was not enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:55 +00:00
if (hw->mac.type >= e1000_82580)
igb: add per-packet timestamping This patch adds support for per-packet timestamping for the 82580 adapter. The rx timestamp code is also pulled out of the inlined rx hotpath and instead moved to a seperate function. This version adds a comment explaining the per-packet timestamping code added to igb_hwtstamp_ioctl(). Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-26 11:36:47 +00:00
srrctl |= E1000_SRRCTL_TIMESTAMP;
igb: only support SRRCTL_DROP_EN when using multiple queues The SRRCTL.DROP_EN bit should only be set when we are supporting multiple queues. This bit is meant to prevent head of line blocking and is unnecessary in the single queue case. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:03:38 +00:00
/* Only set Drop Enable if we are supporting multiple queues */
if (adapter->vfs_allocated_count || adapter->num_rx_queues > 1)
srrctl |= E1000_SRRCTL_DROP_EN;
igb: move SRRCTL register configuration into ring specific config The SRRCTL register exists per ring. Instead of configuring all of them in the RCTL configuration which is meant to be global it makes more sense to move this out into the ring specific configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:51:07 +00:00
wr32(E1000_SRRCTL(reg_idx), srrctl);
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
/* set filtering for VMDQ pools */
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
igb_set_vmolr(adapter, reg_idx & 0x7, true);
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
rxdctl |= IGB_RX_PTHRESH;
rxdctl |= IGB_RX_HTHRESH << 8;
rxdctl |= IGB_RX_WTHRESH << 16;
igb: Update RXDCTL/TXDCTL configurations This change cleans up the RXDCTL and TXDCTL configurations and optimizes RX performance by allowing back write-backs on all hardware other than 82576. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:27 +00:00
igb: Clear Rx buffer_info in configure instead of clean This change makes it so that instead of going through the entire ring on Rx cleanup we only go through the region that was designated to be cleaned up and stop when we reach the region where new allocations should start. In addition we can avoid having to perform a memset on the Rx buffer_info structures until we are about to start using the ring again. By deferring this we can avoid dirtying the cache any more than we have to which can help to improve the time needed to bring the interface down and then back up again in a reset or suspend/resume cycle. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:25:50 +00:00
/* initialize rx_buffer_info */
memset(ring->rx_buffer_info, 0,
sizeof(struct igb_rx_buffer) * ring->count);
igb: Use length to determine if descriptor is done This change makes it so that we use the length of the packet instead of the DD status bit to determine if a new descriptor is ready to be processed. The obvious advantage is that it cuts down on reads as we don't really even need the DD bit if going from a 0 to a non-zero value on size is enough to inform us that the packet has been completed. In addition I have updated the code so that we only reset the Rx descriptor length for descriptor zero when resetting a ring instead of having to do a memset with 0 over the entire ring. By doing this we can save some time on initialization. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:25:41 +00:00
/* initialize Rx descriptor 0 */
rx_desc = IGB_RX_DESC(ring, 0);
rx_desc->wb.upper.length = 0;
igb: Update RXDCTL/TXDCTL configurations This change cleans up the RXDCTL and TXDCTL configurations and optimizes RX performance by allowing back write-backs on all hardware other than 82576. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:27 +00:00
/* enable receive descriptor fetching */
rxdctl |= E1000_RXDCTL_QUEUE_ENABLE;
igb: move the tx and rx ring specific config into seperate functions This change makes the tx and rx config a bit cleaner by breaking out the ring specific configuration from the generic rx and tx configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:50:29 +00:00
wr32(E1000_RXDCTL(reg_idx), rxdctl);
}
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
static void igb_set_rx_buffer_len(struct igb_adapter *adapter,
struct igb_ring *rx_ring)
{
/* set build_skb and buffer size flags */
igb: Add support for padding packet With the size of the frame limited we can now write to an offset within the buffer instead of having to write at the very start of the buffer. The advantage to this is that it allows us to leave padding room for things like supporting XDP in the future. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:14 +00:00
clear_ring_build_skb_enabled(rx_ring);
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
clear_ring_uses_large_buffer(rx_ring);
if (adapter->flags & IGB_FLAG_RX_LEGACY)
return;
igb: Add support for padding packet With the size of the frame limited we can now write to an offset within the buffer instead of having to write at the very start of the buffer. The advantage to this is that it allows us to leave padding room for things like supporting XDP in the future. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:14 +00:00
set_ring_build_skb_enabled(rx_ring);
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
#if (PAGE_SIZE < 8192)
if (adapter->max_frame_size <= IGB_MAX_FRAME_BUILD_SKB)
return;
set_ring_uses_large_buffer(rx_ring);
#endif
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_configure_rx - Configure receive Unit after Reset
* @adapter: board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Configure the Rx unit of the MAC after a reset.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_configure_rx(struct igb_adapter *adapter)
{
drivers/net/igb: fix sparse warning: symbol shadows an earlier one Impact: Move variable declaration as far inner as possible. Fix this sparse warning: drivers/net/igb/igb_main.c:1895:21: warning: symbol 'j' shadows an earlier one drivers/net/igb/igb_main.c:1855:16: originally declared here Signed-off-by: Hannes Eder <hannes@hanneseder.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-19 03:36:04 +00:00
int i;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add combined function for setting rar and pool bits This patch adds igb_rar_qsel which sets the mac address and pool bits for a given mac address in the receive address register table. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:32:49 +00:00
/* set the correct pool for the PF default MAC address in entry 0 */
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
igb_set_default_mac_filter(adapter);
igb: add combined function for setting rar and pool bits This patch adds igb_rar_qsel which sets the mac address and pool bits for a given mac address in the receive address register table. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:32:49 +00:00
igb: move the multiple receive queue configuration into seperate function This patch moves the multiple receive queue configuration into a seperate function from igb_configure_rx. We can essentially do the configuration for the multiple receive queues just prior to enabling the RX and this will allow us to seperate the queue enablement from the receive queue layout configuration. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:25 +00:00
/* Setup the HW Rx Head and Tail Descriptor Pointers and
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* the Base and Length of the Rx Descriptor Ring
*/
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
for (i = 0; i < adapter->num_rx_queues; i++) {
struct igb_ring *rx_ring = adapter->rx_ring[i];
igb_set_rx_buffer_len(adapter, rx_ring);
igb_configure_rx_ring(adapter, rx_ring);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_free_tx_resources - Free Tx Resources per Queue
* @tx_ring: Tx descriptor ring for a specific queue
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Free all transmit software resources
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: Fix tx/rx_ring_count parameters for igb on suspend/resume/ring resize When suspending the device the ring structure is freed which causes it to loose track of the count. To resolve this we need to move the ring count outside of the ring structure and store it in the adapter struct. In addition to resolving the suspend/resume issue this patch also addresses issues seen in the event of memory allocation errors causing uneven ring sizes on multiple queues. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 08:48:10 +00:00
void igb_free_tx_resources(struct igb_ring *tx_ring)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: Remove adapter struct from these function call parameters Simplify a bunch of calls to these functions by moving the adapter struct out of the function arguments since that is not needed. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 17:59:48 +00:00
igb_clean_tx_ring(tx_ring);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
vfree(tx_ring->tx_buffer_info);
tx_ring->tx_buffer_info = NULL;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: cleanup code related to ring resource allocation and free This patch cleans up some of the ring alloc and free code to better handle exceptions such as attempting to free resources on an already freed ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:20 +00:00
/* if not set, then don't free */
if (!tx_ring->desc)
return;
igb: convert igb from using PCI DMA functions to using DMA API functions This patch makes it so that igb now uses the DMA API functions instead of the PCI API functions. To do this the pci_dev pointer that was in the rings has been replaced with a device pointer, and as a result all references to [tr]x_ring->pdev have been replaced with [tr]x_ring->dev. This patch is based of of work originally done by Nicholas Nunley. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 13:09:25 +00:00
dma_free_coherent(tx_ring->dev, tx_ring->size,
tx_ring->desc, tx_ring->dma);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
tx_ring->desc = NULL;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_free_all_tx_resources - Free Tx Resources for All Queues
* @adapter: board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Free all transmit software resources
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_free_all_tx_resources(struct igb_adapter *adapter)
{
int i;
for (i = 0; i < adapter->num_tx_queues; i++)
igb: Fixes needed for surprise removal support This patch adds some checks in order to prevent panic's on surprise removal of devices during S0, S3, S4. Without this patch, Thunderbolt type device removal will panic the system. Signed-off-by: Yanir Lubetkin <yanirx.lubetkin@intel.com> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-22 07:52:54 +00:00
if (adapter->tx_ring[i])
igb_free_tx_resources(adapter->tx_ring[i]);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_clean_tx_ring - Free Tx Buffers
* @tx_ring: ring to be cleaned
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: Remove adapter struct from these function call parameters Simplify a bunch of calls to these functions by moving the adapter struct out of the function arguments since that is not needed. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 17:59:48 +00:00
static void igb_clean_tx_ring(struct igb_ring *tx_ring)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
u16 i = tx_ring->next_to_clean;
struct igb_tx_buffer *tx_buffer = &tx_ring->tx_buffer_info[i];
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
while (i != tx_ring->next_to_use) {
union e1000_adv_tx_desc *eop_desc, *tx_desc;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
/* Free all the Tx ring sk_buffs */
dev_kfree_skb_any(tx_buffer->skb);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
/* unmap skb header data */
dma_unmap_single(tx_ring->dev,
dma_unmap_addr(tx_buffer, dma),
dma_unmap_len(tx_buffer, len),
DMA_TO_DEVICE);
igb, ixgbe: netdev_tx_reset_queue incorrectly called from tx init path igb and ixgbe incorrectly call netdev_tx_reset_queue() from i{gb|xgbe}_clean_tx_ring() this sort of works in most cases except when the number of real tx queues changes. When the number of real tx queues changes netdev_tx_reset_queue() only gets called on the new number of queues so when we reduce the number of queues we risk triggering the watchdog timer and repeated device resets. So this is not only a cosmetic issue but causes real bugs. For example enabling/disabling DCB or FCoE in ixgbe will trigger this. CC: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: John Fastabend <john.r.fastabend@intel.com> Tested-by: John Bishop <johnx.bishop@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-23 12:22:39 +00:00
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
/* check for eop_desc to determine the end of the packet */
eop_desc = tx_buffer->next_to_watch;
tx_desc = IGB_TX_DESC(tx_ring, i);
/* unmap remaining buffers */
while (tx_desc != eop_desc) {
tx_buffer++;
tx_desc++;
i++;
if (unlikely(i == tx_ring->count)) {
i = 0;
tx_buffer = tx_ring->tx_buffer_info;
tx_desc = IGB_TX_DESC(tx_ring, 0);
}
/* unmap any remaining paged data */
if (dma_unmap_len(tx_buffer, len))
dma_unmap_page(tx_ring->dev,
dma_unmap_addr(tx_buffer, dma),
dma_unmap_len(tx_buffer, len),
DMA_TO_DEVICE);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
/* move us one more past the eop_desc for start of next pkt */
tx_buffer++;
i++;
if (unlikely(i == tx_ring->count)) {
i = 0;
tx_buffer = tx_ring->tx_buffer_info;
}
}
/* reset BQL for queue */
netdev_tx_reset_queue(txring_txq(tx_ring));
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
/* reset next_to_use and next_to_clean */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
tx_ring->next_to_use = 0;
tx_ring->next_to_clean = 0;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_clean_all_tx_rings - Free Tx Buffers for all queues
* @adapter: board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_clean_all_tx_rings(struct igb_adapter *adapter)
{
int i;
for (i = 0; i < adapter->num_tx_queues; i++)
igb: Fixes needed for surprise removal support This patch adds some checks in order to prevent panic's on surprise removal of devices during S0, S3, S4. Without this patch, Thunderbolt type device removal will panic the system. Signed-off-by: Yanir Lubetkin <yanirx.lubetkin@intel.com> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-22 07:52:54 +00:00
if (adapter->tx_ring[i])
igb_clean_tx_ring(adapter->tx_ring[i]);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_free_rx_resources - Free Rx Resources
* @rx_ring: ring to clean the resources from
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Free all receive software resources
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: Fix tx/rx_ring_count parameters for igb on suspend/resume/ring resize When suspending the device the ring structure is freed which causes it to loose track of the count. To resolve this we need to move the ring count outside of the ring structure and store it in the adapter struct. In addition to resolving the suspend/resume issue this patch also addresses issues seen in the event of memory allocation errors causing uneven ring sizes on multiple queues. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-20 08:48:10 +00:00
void igb_free_rx_resources(struct igb_ring *rx_ring)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: Remove adapter struct from these function call parameters Simplify a bunch of calls to these functions by moving the adapter struct out of the function arguments since that is not needed. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 17:59:48 +00:00
igb_clean_rx_ring(rx_ring);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
vfree(rx_ring->rx_buffer_info);
rx_ring->rx_buffer_info = NULL;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: cleanup code related to ring resource allocation and free This patch cleans up some of the ring alloc and free code to better handle exceptions such as attempting to free resources on an already freed ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:49:20 +00:00
/* if not set, then don't free */
if (!rx_ring->desc)
return;
igb: convert igb from using PCI DMA functions to using DMA API functions This patch makes it so that igb now uses the DMA API functions instead of the PCI API functions. To do this the pci_dev pointer that was in the rings has been replaced with a device pointer, and as a result all references to [tr]x_ring->pdev have been replaced with [tr]x_ring->dev. This patch is based of of work originally done by Nicholas Nunley. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 13:09:25 +00:00
dma_free_coherent(rx_ring->dev, rx_ring->size,
rx_ring->desc, rx_ring->dma);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
rx_ring->desc = NULL;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_free_all_rx_resources - Free Rx Resources for All Queues
* @adapter: board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Free all receive software resources
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_free_all_rx_resources(struct igb_adapter *adapter)
{
int i;
for (i = 0; i < adapter->num_rx_queues; i++)
igb: Fixes needed for surprise removal support This patch adds some checks in order to prevent panic's on surprise removal of devices during S0, S3, S4. Without this patch, Thunderbolt type device removal will panic the system. Signed-off-by: Yanir Lubetkin <yanirx.lubetkin@intel.com> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-22 07:52:54 +00:00
if (adapter->rx_ring[i])
igb_free_rx_resources(adapter->rx_ring[i]);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_clean_rx_ring - Free Rx Buffers per Queue
* @rx_ring: ring to free buffers from
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: Remove adapter struct from these function call parameters Simplify a bunch of calls to these functions by moving the adapter struct out of the function arguments since that is not needed. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 17:59:48 +00:00
static void igb_clean_rx_ring(struct igb_ring *rx_ring)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: Clear Rx buffer_info in configure instead of clean This change makes it so that instead of going through the entire ring on Rx cleanup we only go through the region that was designated to be cleaned up and stop when we reach the region where new allocations should start. In addition we can avoid having to perform a memset on the Rx buffer_info structures until we are about to start using the ring again. By deferring this we can avoid dirtying the cache any more than we have to which can help to improve the time needed to bring the interface down and then back up again in a reset or suspend/resume cycle. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:25:50 +00:00
u16 i = rx_ring->next_to_clean;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
if (rx_ring->skb)
dev_kfree_skb(rx_ring->skb);
rx_ring->skb = NULL;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Free all the Rx ring sk_buffs */
igb: Clear Rx buffer_info in configure instead of clean This change makes it so that instead of going through the entire ring on Rx cleanup we only go through the region that was designated to be cleaned up and stop when we reach the region where new allocations should start. In addition we can avoid having to perform a memset on the Rx buffer_info structures until we are about to start using the ring again. By deferring this we can avoid dirtying the cache any more than we have to which can help to improve the time needed to bring the interface down and then back up again in a reset or suspend/resume cycle. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:25:50 +00:00
while (i != rx_ring->next_to_alloc) {
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
struct igb_rx_buffer *buffer_info = &rx_ring->rx_buffer_info[i];
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: update driver to make use of DMA_ATTR_SKIP_CPU_SYNC The ARM architecture provides a mechanism for deferring cache line invalidation in the case of map/unmap. This patch makes use of this mechanism to avoid unnecessary synchronization. A secondary effect of this change is that the portion of the page that has been synchronized for use by the CPU should be writable and could be passed up the stack (at least on ARM). The last bit that occurred to me is that on architectures where the sync_for_cpu call invalidates cache lines we were prefetching and then invalidating the first 128 bytes of the packet. To avoid that I have moved the sync up to before we perform the prefetch and allocate the skbuff so that we can actually make use of it. Link: http://lkml.kernel.org/r/20161110113611.76501.98897.stgit@ahduyck-blue-test.jf.intel.com Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Chris Metcalf <cmetcalf@mellanox.com> Cc: David Howells <dhowells@redhat.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Cc: Helge Deller <deller@gmx.de> Cc: James Hogan <james.hogan@imgtec.com> Cc: Jonas Bonn <jonas@southpole.se> Cc: Keguang Zhang <keguang.zhang@gmail.com> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Rich Felker <dalias@libc.org> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Steven Miao <realmz6@gmail.com> Cc: Tobias Klauser <tklauser@distanz.ch> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-14 23:05:30 +00:00
/* Invalidate cache lines that may have been written to by
* device so that we avoid corrupting memory.
*/
dma_sync_single_range_for_cpu(rx_ring->dev,
buffer_info->dma,
buffer_info->page_offset,
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
igb_rx_bufsz(rx_ring),
igb: update driver to make use of DMA_ATTR_SKIP_CPU_SYNC The ARM architecture provides a mechanism for deferring cache line invalidation in the case of map/unmap. This patch makes use of this mechanism to avoid unnecessary synchronization. A secondary effect of this change is that the portion of the page that has been synchronized for use by the CPU should be writable and could be passed up the stack (at least on ARM). The last bit that occurred to me is that on architectures where the sync_for_cpu call invalidates cache lines we were prefetching and then invalidating the first 128 bytes of the packet. To avoid that I have moved the sync up to before we perform the prefetch and allocate the skbuff so that we can actually make use of it. Link: http://lkml.kernel.org/r/20161110113611.76501.98897.stgit@ahduyck-blue-test.jf.intel.com Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Chris Metcalf <cmetcalf@mellanox.com> Cc: David Howells <dhowells@redhat.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Cc: Helge Deller <deller@gmx.de> Cc: James Hogan <james.hogan@imgtec.com> Cc: Jonas Bonn <jonas@southpole.se> Cc: Keguang Zhang <keguang.zhang@gmail.com> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Rich Felker <dalias@libc.org> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Steven Miao <realmz6@gmail.com> Cc: Tobias Klauser <tklauser@distanz.ch> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-14 23:05:30 +00:00
DMA_FROM_DEVICE);
/* free resources associated with mapping */
dma_unmap_page_attrs(rx_ring->dev,
buffer_info->dma,
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
igb_rx_pg_size(rx_ring),
igb: update driver to make use of DMA_ATTR_SKIP_CPU_SYNC The ARM architecture provides a mechanism for deferring cache line invalidation in the case of map/unmap. This patch makes use of this mechanism to avoid unnecessary synchronization. A secondary effect of this change is that the portion of the page that has been synchronized for use by the CPU should be writable and could be passed up the stack (at least on ARM). The last bit that occurred to me is that on architectures where the sync_for_cpu call invalidates cache lines we were prefetching and then invalidating the first 128 bytes of the packet. To avoid that I have moved the sync up to before we perform the prefetch and allocate the skbuff so that we can actually make use of it. Link: http://lkml.kernel.org/r/20161110113611.76501.98897.stgit@ahduyck-blue-test.jf.intel.com Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Chris Metcalf <cmetcalf@mellanox.com> Cc: David Howells <dhowells@redhat.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Cc: Helge Deller <deller@gmx.de> Cc: James Hogan <james.hogan@imgtec.com> Cc: Jonas Bonn <jonas@southpole.se> Cc: Keguang Zhang <keguang.zhang@gmail.com> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Rich Felker <dalias@libc.org> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Steven Miao <realmz6@gmail.com> Cc: Tobias Klauser <tklauser@distanz.ch> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-14 23:05:30 +00:00
DMA_FROM_DEVICE,
igb: Add support for DMA_ATTR_WEAK_ORDERING Since we are already using DMA attributes in igb for Rx there is no reason why we can't also apply DMA_ATTR_WEAK_ORDERING which is needed on some platforms to improve performance. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:25:26 +00:00
IGB_RX_DMA_ATTR);
mm: rename __page_frag functions to __page_frag_cache, drop order from drain This patch does two things. First it goes through and renames the __page_frag prefixed functions to __page_frag_cache so that we can be clear that we are draining or refilling the cache, not the frags themselves. Second we drop the order parameter from __page_frag_cache_drain since we don't actually need to pass it since all fragments are either order 0 or must be a compound page. Link: http://lkml.kernel.org/r/20170104023954.13451.5678.stgit@localhost.localdomain Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-01-11 00:58:09 +00:00
__page_frag_cache_drain(buffer_info->page,
buffer_info->pagecnt_bias);
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
igb: Clear Rx buffer_info in configure instead of clean This change makes it so that instead of going through the entire ring on Rx cleanup we only go through the region that was designated to be cleaned up and stop when we reach the region where new allocations should start. In addition we can avoid having to perform a memset on the Rx buffer_info structures until we are about to start using the ring again. By deferring this we can avoid dirtying the cache any more than we have to which can help to improve the time needed to bring the interface down and then back up again in a reset or suspend/resume cycle. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:25:50 +00:00
i++;
if (i == rx_ring->count)
i = 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
rx_ring->next_to_alloc = 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
rx_ring->next_to_clean = 0;
rx_ring->next_to_use = 0;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_clean_all_rx_rings - Free Rx Buffers for all queues
* @adapter: board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_clean_all_rx_rings(struct igb_adapter *adapter)
{
int i;
for (i = 0; i < adapter->num_rx_queues; i++)
igb: Fixes needed for surprise removal support This patch adds some checks in order to prevent panic's on surprise removal of devices during S0, S3, S4. Without this patch, Thunderbolt type device removal will panic the system. Signed-off-by: Yanir Lubetkin <yanirx.lubetkin@intel.com> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-22 07:52:54 +00:00
if (adapter->rx_ring[i])
igb_clean_rx_ring(adapter->rx_ring[i]);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_set_mac - Change the Ethernet Address of the NIC
* @netdev: network interface device structure
* @p: pointer to an address structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Returns 0 on success, negative on failure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static int igb_set_mac(struct net_device *netdev, void *p)
{
struct igb_adapter *adapter = netdev_priv(netdev);
igb: remove unnecessary adapter->hw calls when just hw-> will do. There were several spots in the code making calls to adapter->hw when they could have just been accessing hw-> directly. I cleaned up the spots where this was visibly apparent. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:20:31 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
struct sockaddr *addr = p;
if (!is_valid_ether_addr(addr->sa_data))
return -EADDRNOTAVAIL;
memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
igb: remove unnecessary adapter->hw calls when just hw-> will do. There were several spots in the code making calls to adapter->hw when they could have just been accessing hw-> directly. I cleaned up the spots where this was visibly apparent. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:20:31 +00:00
memcpy(hw->mac.addr, addr->sa_data, netdev->addr_len);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add combined function for setting rar and pool bits This patch adds igb_rar_qsel which sets the mac address and pool bits for a given mac address in the receive address register table. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:32:49 +00:00
/* set the correct pool for the new PF MAC address in entry 0 */
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
igb_set_default_mac_filter(adapter);
igb: add pf side of VMDq support Add the pf portion of vmdq support. This provides enough support so that VMDq is enabled, and the pf is functional without enabling vfs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:44 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return 0;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_write_mc_addr_list - write multicast addresses to MTA
* @netdev: network interface device structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Writes multicast address list to the MTA hash table.
* Returns: -ENOMEM on failure
* 0 on no addresses written
* X on writing X addresses to MTA
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
static int igb_write_mc_addr_list(struct net_device *netdev)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
net: convert multicast list to list_head Converts the list and the core manipulating with it to be the same as uc_list. +uses two functions for adding/removing mc address (normal and "global" variant) instead of a function parameter. +removes dev_mcast.c completely. +exposes netdev_hw_addr_list_* macros along with __hw_addr_* functions for manipulation with lists on a sandbox (used in bonding and 80211 drivers) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-01 21:22:57 +00:00
struct netdev_hw_addr *ha;
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
u8 *mta_list;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
int i;
net: use netdev_mc_count and netdev_mc_empty when appropriate This patch replaces dev->mc_count in all drivers (hopefully I didn't miss anything). Used spatch and did small tweaks and conding style changes when it was suitable. Jirka Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-08 04:30:35 +00:00
if (netdev_mc_empty(netdev)) {
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
/* nothing to program, so clear mc list */
igb_update_mc_addr_list(hw, NULL, 0);
igb_restore_vf_multicasts(adapter);
return 0;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
treewide: kzalloc() -> kcalloc() The kzalloc() function has a 2-factor argument form, kcalloc(). This patch replaces cases of: kzalloc(a * b, gfp) with: kcalloc(a * b, gfp) as well as handling cases of: kzalloc(a * b * c, gfp) with: kzalloc(array3_size(a, b, c), gfp) as it's slightly less ugly than: kzalloc_array(array_size(a, b), c, gfp) This does, however, attempt to ignore constant size factors like: kzalloc(4 * 1024, gfp) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( kzalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | kzalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( kzalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(__u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(char) * COUNT + COUNT , ...) | kzalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_ID) + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_ID + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_CONST) + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_CONST + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_ID) + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_ID + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_CONST) + COUNT_CONST, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_CONST + COUNT_CONST, sizeof(THING) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ - kzalloc + kcalloc ( - SIZE * COUNT + COUNT, SIZE , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( kzalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( kzalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( kzalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products, // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( kzalloc(C1 * C2 * C3, ...) | kzalloc( - (E1) * E2 * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * (E3) + array3_size(E1, E2, E3) , ...) | kzalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants, // keeping sizeof() as the second factor argument. @@ expression THING, E1, E2; type TYPE; constant C1, C2, C3; @@ ( kzalloc(sizeof(THING) * C2, ...) | kzalloc(sizeof(TYPE) * C2, ...) | kzalloc(C1 * C2 * C3, ...) | kzalloc(C1 * C2, ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (E2) + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * E2 + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (E2) + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * E2 + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - (E1) * E2 + E1, E2 , ...) | - kzalloc + kcalloc ( - (E1) * (E2) + E1, E2 , ...) | - kzalloc + kcalloc ( - E1 * E2 + E1, E2 , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-12 21:03:40 +00:00
mta_list = kcalloc(netdev_mc_count(netdev), 6, GFP_ATOMIC);
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
if (!mta_list)
return -ENOMEM;
igb: add support for set_rx_mode netdevice operation This patch adds support for the set_rx_mode netdevice operation so that igb can better support multiple unicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:48:56 +00:00
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
/* The shared function expects a packed array of only addresses. */
net: convert multiple drivers to use netdev_for_each_mc_addr, part4 Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-22 09:22:26 +00:00
i = 0;
net: convert multicast list to list_head Converts the list and the core manipulating with it to be the same as uc_list. +uses two functions for adding/removing mc address (normal and "global" variant) instead of a function parameter. +removes dev_mcast.c completely. +exposes netdev_hw_addr_list_* macros along with __hw_addr_* functions for manipulation with lists on a sandbox (used in bonding and 80211 drivers) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-01 21:22:57 +00:00
netdev_for_each_mc_addr(ha, netdev)
memcpy(mta_list + (i++ * ETH_ALEN), ha->addr, ETH_ALEN);
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
igb_update_mc_addr_list(hw, mta_list, i);
kfree(mta_list);
net: use netdev_mc_count and netdev_mc_empty when appropriate This patch replaces dev->mc_count in all drivers (hopefully I didn't miss anything). Used spatch and did small tweaks and conding style changes when it was suitable. Jirka Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-08 04:30:35 +00:00
return netdev_mc_count(netdev);
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
}
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
static int igb_vlan_promisc_enable(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
u32 i, pf_id;
switch (hw->mac.type) {
case e1000_i210:
case e1000_i211:
case e1000_i350:
/* VLAN filtering needed for VLAN prio filter */
if (adapter->netdev->features & NETIF_F_NTUPLE)
break;
/* fall through */
case e1000_82576:
case e1000_82580:
case e1000_i354:
/* VLAN filtering needed for pool filtering */
if (adapter->vfs_allocated_count)
break;
/* fall through */
default:
return 1;
}
/* We are already in VLAN promisc, nothing to do */
if (adapter->flags & IGB_FLAG_VLAN_PROMISC)
return 0;
if (!adapter->vfs_allocated_count)
goto set_vfta;
/* Add PF to all active pools */
pf_id = adapter->vfs_allocated_count + E1000_VLVF_POOLSEL_SHIFT;
for (i = E1000_VLVF_ARRAY_SIZE; --i;) {
u32 vlvf = rd32(E1000_VLVF(i));
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
vlvf |= BIT(pf_id);
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
wr32(E1000_VLVF(i), vlvf);
}
set_vfta:
/* Set all bits in the VLAN filter table array */
for (i = E1000_VLAN_FILTER_TBL_SIZE; i--;)
hw->mac.ops.write_vfta(hw, i, ~0U);
/* Set flag so we don't redo unnecessary work */
adapter->flags |= IGB_FLAG_VLAN_PROMISC;
return 0;
}
#define VFTA_BLOCK_SIZE 8
static void igb_scrub_vfta(struct igb_adapter *adapter, u32 vfta_offset)
{
struct e1000_hw *hw = &adapter->hw;
u32 vfta[VFTA_BLOCK_SIZE] = { 0 };
u32 vid_start = vfta_offset * 32;
u32 vid_end = vid_start + (VFTA_BLOCK_SIZE * 32);
u32 i, vid, word, bits, pf_id;
/* guarantee that we don't scrub out management VLAN */
vid = adapter->mng_vlan_id;
if (vid >= vid_start && vid < vid_end)
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
vfta[(vid - vid_start) / 32] |= BIT(vid % 32);
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
if (!adapter->vfs_allocated_count)
goto set_vfta;
pf_id = adapter->vfs_allocated_count + E1000_VLVF_POOLSEL_SHIFT;
for (i = E1000_VLVF_ARRAY_SIZE; --i;) {
u32 vlvf = rd32(E1000_VLVF(i));
/* pull VLAN ID from VLVF */
vid = vlvf & VLAN_VID_MASK;
/* only concern ourselves with a certain range */
if (vid < vid_start || vid >= vid_end)
continue;
if (vlvf & E1000_VLVF_VLANID_ENABLE) {
/* record VLAN ID in VFTA */
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
vfta[(vid - vid_start) / 32] |= BIT(vid % 32);
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
/* if PF is part of this then continue */
if (test_bit(vid, adapter->active_vlans))
continue;
}
/* remove PF from the pool */
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
bits = ~BIT(pf_id);
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
bits &= rd32(E1000_VLVF(i));
wr32(E1000_VLVF(i), bits);
}
set_vfta:
/* extract values from active_vlans and write back to VFTA */
for (i = VFTA_BLOCK_SIZE; i--;) {
vid = (vfta_offset + i) * 32;
word = vid / BITS_PER_LONG;
bits = vid % BITS_PER_LONG;
vfta[i] |= adapter->active_vlans[word] >> bits;
hw->mac.ops.write_vfta(hw, vfta_offset + i, vfta[i]);
}
}
static void igb_vlan_promisc_disable(struct igb_adapter *adapter)
{
u32 i;
/* We are not in VLAN promisc, nothing to do */
if (!(adapter->flags & IGB_FLAG_VLAN_PROMISC))
return;
/* Set flag so we don't redo unnecessary work */
adapter->flags &= ~IGB_FLAG_VLAN_PROMISC;
for (i = 0; i < E1000_VLAN_FILTER_TBL_SIZE; i += VFTA_BLOCK_SIZE)
igb_scrub_vfta(adapter, i);
}
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_set_rx_mode - Secondary Unicast, Multicast and Promiscuous mode set
* @netdev: network interface device structure
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* The set_rx_mode entry point is called whenever the unicast or multicast
* address lists or the network interface flags are updated. This routine is
* responsible for configuring the hardware for proper unicast, multicast,
* promiscuous mode, and all-multi behavior.
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
**/
static void igb_set_rx_mode(struct net_device *netdev)
{
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
unsigned int vfn = adapter->vfs_allocated_count;
igb: Limit maximum frame Rx based on MTU In order to support the use of build_skb going forward it will be necessary to place a maximum limit on the amount of data we can receive when jumbo frames is not enabled. In order to do this I am adding a new upper limit for receive based on the size of a 2K buffer minus padding. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:15 +00:00
u32 rctl = 0, vmolr = 0, rlpml = MAX_JUMBO_FRAME_SIZE;
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
int count;
/* Check for Promiscuous and All Multicast modes */
if (netdev->flags & IFF_PROMISC) {
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
rctl |= E1000_RCTL_UPE | E1000_RCTL_MPE;
igb: Add workaround for VLAN tag stripping on 82576 There was a workaround partially implemented for the 82576 that is needed in order for VLAN tag stripping to function correctly. The original code had side effects that would make it so the workaround was active on all MACs. I have updated the code so that the workaround is enabled, but limited to the 82576, or activated if we exceed the available unicast addresses. The workaround has a side effect of mirroring all of the traffic outgoing from the VFs back to the PF. As such it is not recommended to use the 82576 in promiscuous mode as it will take a performance hit, though this is now consistent with the performance as seen on the out-of-tree igb driver. I also limited the scope of the UTA bits all being set to only when the VMOLR register is enabled. This should limit the effects of the UTA register so that we don't pick up any excess traffic unless promiscuous mode has been enabled on the PF, whereas before the PF would have ended up in something equivalent to unicast promiscuous mode with VLAN filtering otherwise. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:43 +00:00
vmolr |= E1000_VMOLR_MPME;
/* enable use of UTA filter to force packets to default pool */
if (hw->mac.type == e1000_82576)
vmolr |= E1000_VMOLR_ROPE;
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
} else {
if (netdev->flags & IFF_ALLMULTI) {
rctl |= E1000_RCTL_MPE;
vmolr |= E1000_VMOLR_MPME;
} else {
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Write addresses to the MTA, if the attempt fails
Fix common misspellings Fixes generated by 'codespell' and manually reviewed. Signed-off-by: Lucas De Marchi <lucas.demarchi@profusion.mobi>
2011-03-31 01:57:33 +00:00
* then we should just turn on promiscuous mode so
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
* that we can at least receive multicast traffic
*/
count = igb_write_mc_addr_list(netdev);
if (count < 0) {
rctl |= E1000_RCTL_MPE;
vmolr |= E1000_VMOLR_MPME;
} else if (count) {
vmolr |= E1000_VMOLR_ROMPE;
}
}
igb: move all multicast addresses into multicast table array This patch moves all of the multicast addresses out of the free Receive address registers and instead programs them all into the multicast table array. As a result the multicast filtering may not be as precise, but it also greatly reduces the overhead for multicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:08:54 +00:00
}
igb: Enable use of "bridge fdb add" to set unicast table entries This change makes it so that we can use the bridge utility to add a FDB entry for the PF to an igb port. By doing this we can enable the VFs to talk to virtual ports residing on top of the PF. In addition this should also address issues with MACVLANs trying to reside on top of the PF as well as they would have had similar issues when added to the PF with SR-IOV enabled. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:34 +00:00
/* Write addresses to available RAR registers, if there is not
* sufficient space to store all the addresses then enable
* unicast promiscuous mode
*/
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
if (__dev_uc_sync(netdev, igb_uc_sync, igb_uc_unsync)) {
igb: Enable use of "bridge fdb add" to set unicast table entries This change makes it so that we can use the bridge utility to add a FDB entry for the PF to an igb port. By doing this we can enable the VFs to talk to virtual ports residing on top of the PF. In addition this should also address issues with MACVLANs trying to reside on top of the PF as well as they would have had similar issues when added to the PF with SR-IOV enabled. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:34 +00:00
rctl |= E1000_RCTL_UPE;
vmolr |= E1000_VMOLR_ROPE;
igb: move all multicast addresses into multicast table array This patch moves all of the multicast addresses out of the free Receive address registers and instead programs them all into the multicast table array. As a result the multicast filtering may not be as precise, but it also greatly reduces the overhead for multicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:08:54 +00:00
}
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
/* enable VLAN filtering by default */
rctl |= E1000_RCTL_VFE;
/* disable VLAN filtering for modes that require it */
if ((netdev->flags & IFF_PROMISC) ||
(netdev->features & NETIF_F_RXALL)) {
/* if we fail to set all rules then just clear VFE */
if (igb_vlan_promisc_enable(adapter))
rctl &= ~E1000_RCTL_VFE;
} else {
igb_vlan_promisc_disable(adapter);
}
/* update state of unicast, multicast, and VLAN filtering modes */
rctl |= rd32(E1000_RCTL) & ~(E1000_RCTL_UPE | E1000_RCTL_MPE |
E1000_RCTL_VFE);
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
wr32(E1000_RCTL, rctl);
igb: move all multicast addresses into multicast table array This patch moves all of the multicast addresses out of the free Receive address registers and instead programs them all into the multicast table array. As a result the multicast filtering may not be as precise, but it also greatly reduces the overhead for multicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:08:54 +00:00
igb: Limit maximum frame Rx based on MTU In order to support the use of build_skb going forward it will be necessary to place a maximum limit on the amount of data we can receive when jumbo frames is not enabled. In order to do this I am adding a new upper limit for receive based on the size of a 2K buffer minus padding. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:15 +00:00
#if (PAGE_SIZE < 8192)
if (!adapter->vfs_allocated_count) {
if (adapter->max_frame_size <= IGB_MAX_FRAME_BUILD_SKB)
rlpml = IGB_MAX_FRAME_BUILD_SKB;
}
#endif
wr32(E1000_RLPML, rlpml);
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* In order to support SR-IOV and eventually VMDq it is necessary to set
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
* the VMOLR to enable the appropriate modes. Without this workaround
* we will have issues with VLAN tag stripping not being done for frames
* that are only arriving because we are the default pool
*/
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
if ((hw->mac.type < e1000_82576) || (hw->mac.type > e1000_i350))
igb: move all multicast addresses into multicast table array This patch moves all of the multicast addresses out of the free Receive address registers and instead programs them all into the multicast table array. As a result the multicast filtering may not be as precise, but it also greatly reduces the overhead for multicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:08:54 +00:00
return;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Add workaround for VLAN tag stripping on 82576 There was a workaround partially implemented for the 82576 that is needed in order for VLAN tag stripping to function correctly. The original code had side effects that would make it so the workaround was active on all MACs. I have updated the code so that the workaround is enabled, but limited to the 82576, or activated if we exceed the available unicast addresses. The workaround has a side effect of mirroring all of the traffic outgoing from the VFs back to the PF. As such it is not recommended to use the 82576 in promiscuous mode as it will take a performance hit, though this is now consistent with the performance as seen on the out-of-tree igb driver. I also limited the scope of the UTA bits all being set to only when the VMOLR register is enabled. This should limit the effects of the UTA register so that we don't pick up any excess traffic unless promiscuous mode has been enabled on the PF, whereas before the PF would have ended up in something equivalent to unicast promiscuous mode with VLAN filtering otherwise. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:43 +00:00
/* set UTA to appropriate mode */
igb_set_uta(adapter, !!(vmolr & E1000_VMOLR_ROPE));
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
vmolr |= rd32(E1000_VMOLR(vfn)) &
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
~(E1000_VMOLR_ROPE | E1000_VMOLR_MPME | E1000_VMOLR_ROMPE);
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
igb: Limit maximum frame Rx based on MTU In order to support the use of build_skb going forward it will be necessary to place a maximum limit on the amount of data we can receive when jumbo frames is not enabled. In order to do this I am adding a new upper limit for receive based on the size of a 2K buffer minus padding. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:15 +00:00
/* enable Rx jumbo frames, restrict as needed to support build_skb */
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
vmolr &= ~E1000_VMOLR_RLPML_MASK;
igb: Limit maximum frame Rx based on MTU In order to support the use of build_skb going forward it will be necessary to place a maximum limit on the amount of data we can receive when jumbo frames is not enabled. In order to do this I am adding a new upper limit for receive based on the size of a 2K buffer minus padding. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:15 +00:00
#if (PAGE_SIZE < 8192)
if (adapter->max_frame_size <= IGB_MAX_FRAME_BUILD_SKB)
vmolr |= IGB_MAX_FRAME_BUILD_SKB;
else
#endif
vmolr |= MAX_JUMBO_FRAME_SIZE;
vmolr |= E1000_VMOLR_LPE;
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
wr32(E1000_VMOLR(vfn), vmolr);
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
igb: move all multicast addresses into multicast table array This patch moves all of the multicast addresses out of the free Receive address registers and instead programs them all into the multicast table array. As a result the multicast filtering may not be as precise, but it also greatly reduces the overhead for multicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:08:54 +00:00
igb_restore_vf_multicasts(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Add Anti-spoofing feature support Add support for the anti-spoofing feature in the HW. Packets from VF devices with spoofed MAC addresses or VLAN tags will be blocked and an event generated. When the watchdog task runs it will call a function to check if any spoof events occurred. If an event was detected then a warning message is dumped to the system log. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 02:08:26 +00:00
static void igb_check_wvbr(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
u32 wvbr = 0;
switch (hw->mac.type) {
case e1000_82576:
case e1000_i350:
igb: Cleanups to fix assignment in if error This patch fixes ERROR:ASSIGN_IN_IF found with checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:46:13 +00:00
wvbr = rd32(E1000_WVBR);
if (!wvbr)
igb: Add Anti-spoofing feature support Add support for the anti-spoofing feature in the HW. Packets from VF devices with spoofed MAC addresses or VLAN tags will be blocked and an event generated. When the watchdog task runs it will call a function to check if any spoof events occurred. If an event was detected then a warning message is dumped to the system log. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 02:08:26 +00:00
return;
break;
default:
break;
}
adapter->wvbr |= wvbr;
}
#define IGB_STAGGERED_QUEUE_OFFSET 8
static void igb_spoof_check(struct igb_adapter *adapter)
{
int j;
if (!adapter->wvbr)
return;
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
for (j = 0; j < adapter->vfs_allocated_count; j++) {
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
if (adapter->wvbr & BIT(j) ||
adapter->wvbr & BIT(j + IGB_STAGGERED_QUEUE_OFFSET)) {
igb: Add Anti-spoofing feature support Add support for the anti-spoofing feature in the HW. Packets from VF devices with spoofed MAC addresses or VLAN tags will be blocked and an event generated. When the watchdog task runs it will call a function to check if any spoof events occurred. If an event was detected then a warning message is dumped to the system log. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 02:08:26 +00:00
dev_warn(&adapter->pdev->dev,
"Spoof event(s) detected on VF %d\n", j);
adapter->wvbr &=
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
~(BIT(j) |
BIT(j + IGB_STAGGERED_QUEUE_OFFSET));
igb: Add Anti-spoofing feature support Add support for the anti-spoofing feature in the HW. Packets from VF devices with spoofed MAC addresses or VLAN tags will be blocked and an event generated. When the watchdog task runs it will call a function to check if any spoof events occurred. If an event was detected then a warning message is dumped to the system log. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 02:08:26 +00:00
}
}
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Need to wait a few seconds after link up to get diagnostic information from
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* the phy
*/
ethernet/intel: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Switches test of .data field to .function, since .data will be going away. Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: intel-wired-lan@lists.osuosl.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-17 00:29:35 +00:00
static void igb_update_phy_info(struct timer_list *t)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
ethernet/intel: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Switches test of .data field to .function, since .data will be going away. Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: intel-wired-lan@lists.osuosl.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-17 00:29:35 +00:00
struct igb_adapter *adapter = from_timer(adapter, t, phy_info_timer);
igb: do not use phy ops in ethtool test cleanup for non-copper parts Currently the igb driver is experiencing a panic due to a null function pointer being used during the cleanup of the ethtool looback test on fiber/serdes parts. This patch prevents that and adds a check prior to calling any phy function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-22 05:30:24 +00:00
igb_get_phy_info(&adapter->hw);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: add link check function Add a link check function to contain all activities related to verifying that the link is present. The current approach is a bit cludgy and needs to be cleaned up. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:16:24 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_has_link - check shared code for link and determine up/down
* @adapter: pointer to driver private info
igb: add link check function Add a link check function to contain all activities related to verifying that the link is present. The current approach is a bit cludgy and needs to be cleaned up. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:16:24 +00:00
**/
igb: Report link status in ethtool when interface is down With this change ethtool will correctly report link status when the interface is down. Currently ethtool reports the link as not detected when the interface is down. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:01:21 +00:00
bool igb_has_link(struct igb_adapter *adapter)
igb: add link check function Add a link check function to contain all activities related to verifying that the link is present. The current approach is a bit cludgy and needs to be cleaned up. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:16:24 +00:00
{
struct e1000_hw *hw = &adapter->hw;
bool link_active = false;
/* get_link_status is set on LSC (link status) interrupt or
* rx sequence error interrupt. get_link_status will stay
* false until the e1000_check_for_link establishes link
* for copper adapters ONLY
*/
switch (hw->phy.media_type) {
case e1000_media_type_copper:
igb: Read register for latch_on without return value This patch changes register read to "just-read" without returning a value for hardware to accurately latch the register value. Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-06-06 01:31:09 +00:00
if (!hw->mac.get_link_status)
return true;
igb_main: Mark expected switch fall-throughs In preparation to enabling -Wimplicit-fallthrough, mark switch cases where we are expecting to fall through. Addresses-Coverity-ID: 200521 ("Missing break in switch") Addresses-Coverity-ID: 114797 ("Missing break in switch") Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-08-07 23:22:22 +00:00
/* fall through */
igb: add link check function Add a link check function to contain all activities related to verifying that the link is present. The current approach is a bit cludgy and needs to be cleaned up. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:16:24 +00:00
case e1000_media_type_internal_serdes:
igb: Read register for latch_on without return value This patch changes register read to "just-read" without returning a value for hardware to accurately latch the register value. Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-06-06 01:31:09 +00:00
hw->mac.ops.check_for_link(hw);
link_active = !hw->mac.get_link_status;
igb: add link check function Add a link check function to contain all activities related to verifying that the link is present. The current approach is a bit cludgy and needs to be cleaned up. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:16:24 +00:00
break;
default:
case e1000_media_type_unknown:
break;
}
igb: Implementation of 1-sec delay for i210 devices This patch adds 1 sec delay mechanism to i210 device family, in order to avoid erroneous link issue with the link partner. Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-08-28 02:22:43 +00:00
if (((hw->mac.type == e1000_i210) ||
(hw->mac.type == e1000_i211)) &&
(hw->phy.id == I210_I_PHY_ID)) {
if (!netif_carrier_ok(adapter->netdev)) {
adapter->flags &= ~IGB_FLAG_NEED_LINK_UPDATE;
} else if (!(adapter->flags & IGB_FLAG_NEED_LINK_UPDATE)) {
adapter->flags |= IGB_FLAG_NEED_LINK_UPDATE;
adapter->link_check_timeout = jiffies;
}
}
igb: add link check function Add a link check function to contain all activities related to verifying that the link is present. The current approach is a bit cludgy and needs to be cleaned up. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:16:24 +00:00
return link_active;
}
igb: introduce igb_thermal_sensor_event for sensor checking The code for thermal sensor checking should be wrapped into a function. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-04-05 04:27:15 +00:00
static bool igb_thermal_sensor_event(struct e1000_hw *hw, u32 event)
{
bool ret = false;
u32 ctrl_ext, thstat;
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
/* check for thermal sensor event on i350 copper only */
igb: introduce igb_thermal_sensor_event for sensor checking The code for thermal sensor checking should be wrapped into a function. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-04-05 04:27:15 +00:00
if (hw->mac.type == e1000_i350) {
thstat = rd32(E1000_THSTAT);
ctrl_ext = rd32(E1000_CTRL_EXT);
if ((hw->phy.media_type == e1000_media_type_copper) &&
igb: random code and comments fix This patch fixes code and comments as identified in the driver. Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-29 10:15:31 +00:00
!(ctrl_ext & E1000_CTRL_EXT_LINK_MODE_SGMII))
igb: introduce igb_thermal_sensor_event for sensor checking The code for thermal sensor checking should be wrapped into a function. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-04-05 04:27:15 +00:00
ret = !!(thstat & event);
}
return ret;
}
igb: Add message when malformed packets detected by hw This patch adds a check and prints the error cause register value when the hardware detects a malformed packet. This is a very unlikely scenario but has been seen occasionally, so printing the message to assist the user. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-07-09 04:55:45 +00:00
/**
* igb_check_lvmmc - check for malformed packets received
* and indicated in LVMMC register
* @adapter: pointer to adapter
**/
static void igb_check_lvmmc(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
u32 lvmmc;
lvmmc = rd32(E1000_LVMMC);
if (lvmmc) {
if (unlikely(net_ratelimit())) {
netdev_warn(adapter->netdev,
"malformed Tx packet detected and dropped, LVMMC:0x%08x\n",
lvmmc);
}
}
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_watchdog - Timer Call-back
* @data: pointer to adapter cast into an unsigned long
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
ethernet/intel: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Switches test of .data field to .function, since .data will be going away. Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: intel-wired-lan@lists.osuosl.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-17 00:29:35 +00:00
static void igb_watchdog(struct timer_list *t)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
ethernet/intel: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Switches test of .data field to .function, since .data will be going away. Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: intel-wired-lan@lists.osuosl.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-17 00:29:35 +00:00
struct igb_adapter *adapter = from_timer(adapter, t, watchdog_timer);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Do the rest outside of interrupt context */
schedule_work(&adapter->watchdog_task);
}
static void igb_watchdog_task(struct work_struct *work)
{
struct igb_adapter *adapter = container_of(work,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
struct igb_adapter,
watchdog_task);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: display a warning message when SmartSpeed works Current igb driver doesn't tell nothing when Link Speed is downgraded due to SmartSpeed. As a result, users suspect that there is something wrong with NIC. If the cause of it is SmartSpeed, there is no means to replace NIC. This patch make igb notify users that SmartSpeed worked. Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-16 11:05:53 +00:00
struct e1000_phy_info *phy = &hw->phy;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
struct net_device *netdev = adapter->netdev;
igb: introduce igb_thermal_sensor_event for sensor checking The code for thermal sensor checking should be wrapped into a function. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-04-05 04:27:15 +00:00
u32 link;
igb: force all queues to interrupt once every 2 seconds Set the EICS bit for each of the RX queues at least once every 2 seconds to prevent the rx queues from stalling. Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-08-26 11:25:03 +00:00
int i;
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
u32 connsw;
igb: When GbE link up, wait for Remote receiver status condition I210 device IPv6 autoconf test sometimes fails, because DAD NS for link-local is not transmitted. This packet is silently dropped. This problem is seen only GbE environment. igb_watchdog_task link up detection continues to the following process. The following cases are observed: 1.PHY 1000BASE-T Status Register Remote receiver status bit is NG. (NG status becomes OK after about 200 - 700ms) 2.In this case, the transfer packet is silently dropped. 1000BASE-T Status register [Expected]: 0x3800 or 0x7800 [problem occurred]: 0x2800 or 0x6800 Frequency of occurrence: approx 1/10 - 1/40 observed In order to avoid this problem, wait until 1000BASE-T Status register "Remote receiver status OK" After applying this patch, at least 400 runs succeed with no problems. Signed-off-by: Takuma Ueba <t.ueba11@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-12-31 05:58:14 +00:00
u16 phy_data, retry_count = 20;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add link check function Add a link check function to contain all activities related to verifying that the link is present. The current approach is a bit cludgy and needs to be cleaned up. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:16:24 +00:00
link = igb_has_link(adapter);
igb: Implementation of 1-sec delay for i210 devices This patch adds 1 sec delay mechanism to i210 device family, in order to avoid erroneous link issue with the link partner. Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-08-28 02:22:43 +00:00
if (adapter->flags & IGB_FLAG_NEED_LINK_UPDATE) {
if (time_after(jiffies, (adapter->link_check_timeout + HZ)))
adapter->flags &= ~IGB_FLAG_NEED_LINK_UPDATE;
else
link = false;
}
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
/* Force link down if we have fiber to swap to */
if (adapter->flags & IGB_FLAG_MAS_ENABLE) {
if (hw->phy.media_type == e1000_media_type_copper) {
connsw = rd32(E1000_CONNSW);
if (!(connsw & E1000_CONNSW_AUTOSENSE_EN))
link = 0;
}
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (link) {
igb: Add media switching feature for i354 PHY's This patch adds a new feature which is supported in some PHY's on some i354 devices. This feature is Auto Media Detect and allows which ever media is detected first by the PHY to be the media used and configured by the device. This is a media swapping feature that is wholly contained in the Marvell PHY. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:23:01 +00:00
/* Perform a reset if the media type changed. */
if (hw->dev_spec._82575.media_changed) {
hw->dev_spec._82575.media_changed = false;
adapter->flags |= IGB_FLAG_MEDIA_RESET;
igb_reset(adapter);
}
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
/* Cancel scheduled suspend requests. */
pm_runtime_resume(netdev->dev.parent);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (!netif_carrier_ok(netdev)) {
u32 ctrl;
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: replace unecessary &adapter->hw with just hw where applicable This patch just cleans up some unecessary references to the adapter->hw member when it has already been placed in a local variable named hw. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:35 +00:00
hw->mac.ops.get_speed_and_duplex(hw,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
&adapter->link_speed,
&adapter->link_duplex);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
ctrl = rd32(E1000_CTRL);
igb: link up/down messages must follow a specific format The system log messages created on a link status change need to follow a specific format to work with tools some customers use. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-27 08:21:39 +00:00
/* Links status message must follow this format */
igb: Cleanups for messaging This patch fixes WARNING:PREFER_PR_LEVEL and WARNING:SPLIT_STRING from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:17 +00:00
netdev_info(netdev,
"igb: %s NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n",
igb: cleanup whitespace issues in igb_main.c This patch goes through and cleans up whitespace issues in igb_main.c to help improve readability. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:52:50 +00:00
netdev->name,
adapter->link_speed,
adapter->link_duplex == FULL_DUPLEX ?
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
"Full" : "Half",
(ctrl & E1000_CTRL_TFCE) &&
(ctrl & E1000_CTRL_RFCE) ? "RX/TX" :
(ctrl & E1000_CTRL_RFCE) ? "RX" :
(ctrl & E1000_CTRL_TFCE) ? "TX" : "None");
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Fix for devices using ethtool for EEE settings This patch fixes a problem where using ethtool for EEE setting was not working correctly. This patch also fixes a problem where the function that checks for EEE status on i354 devices was not being called and was causing warnings with static analysis tools. Reported-by: Rashika Kheria <rashika.kheria@gmail.com> Reported-by: Josh Triplett <josh@joshtriplett.org> Reported-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-12 03:58:22 +00:00
/* disable EEE if enabled */
if ((adapter->flags & IGB_FLAG_EEE) &&
(adapter->link_duplex == HALF_DUPLEX)) {
dev_info(&adapter->pdev->dev,
"EEE Disabled: unsupported at half duplex. Re-enable using ethtool when at full duplex.\n");
adapter->hw.dev_spec._82575.eee_disable = true;
adapter->flags &= ~IGB_FLAG_EEE;
}
igb: display a warning message when SmartSpeed works Current igb driver doesn't tell nothing when Link Speed is downgraded due to SmartSpeed. As a result, users suspect that there is something wrong with NIC. If the cause of it is SmartSpeed, there is no means to replace NIC. This patch make igb notify users that SmartSpeed worked. Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-16 11:05:53 +00:00
/* check if SmartSpeed worked */
igb_check_downshift(hw);
if (phy->speed_downgraded)
netdev_warn(netdev, "Link Speed was downgraded by SmartSpeed\n");
igb: introduce igb_thermal_sensor_event for sensor checking The code for thermal sensor checking should be wrapped into a function. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-04-05 04:27:15 +00:00
/* check for thermal sensor event */
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
if (igb_thermal_sensor_event(hw,
igb: Cleanups to fix braces location warnings This patch fixes WARNING:BRACES and ERROR:OPEN_BRACE from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:23 +00:00
E1000_THSTAT_LINK_THROTTLE))
igb: Cleanups for messaging This patch fixes WARNING:PREFER_PR_LEVEL and WARNING:SPLIT_STRING from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:17 +00:00
netdev_info(netdev, "The network adapter link speed was downshifted because it overheated\n");
igb: introduce igb_thermal_sensor_event for sensor checking The code for thermal sensor checking should be wrapped into a function. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-04-05 04:27:15 +00:00
igb: do not modify tx_queue_len on link speed change Previously the driver tweaked txqueuelen to avoid false Tx hang reports seen at half duplex. This had the effect of overriding user set values on link change/reset. Testing shows that adjusting only the timeout factor is sufficient to prevent Tx hang reports at half duplex. Based on e1000e patch by Franco Fichtner <franco@lastsummer.de> Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-23 18:34:57 +00:00
/* adjust timeout factor according to speed/duplex */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
adapter->tx_timeout_factor = 1;
switch (adapter->link_speed) {
case SPEED_10:
adapter->tx_timeout_factor = 14;
break;
case SPEED_100:
/* maybe add some timeout factor ? */
break;
}
igb: When GbE link up, wait for Remote receiver status condition I210 device IPv6 autoconf test sometimes fails, because DAD NS for link-local is not transmitted. This packet is silently dropped. This problem is seen only GbE environment. igb_watchdog_task link up detection continues to the following process. The following cases are observed: 1.PHY 1000BASE-T Status Register Remote receiver status bit is NG. (NG status becomes OK after about 200 - 700ms) 2.In this case, the transfer packet is silently dropped. 1000BASE-T Status register [Expected]: 0x3800 or 0x7800 [problem occurred]: 0x2800 or 0x6800 Frequency of occurrence: approx 1/10 - 1/40 observed In order to avoid this problem, wait until 1000BASE-T Status register "Remote receiver status OK" After applying this patch, at least 400 runs succeed with no problems. Signed-off-by: Takuma Ueba <t.ueba11@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-12-31 05:58:14 +00:00
if (adapter->link_speed != SPEED_1000)
goto no_wait;
/* wait for Remote receiver status OK */
retry_read_status:
if (!igb_read_phy_reg(hw, PHY_1000T_STATUS,
&phy_data)) {
if (!(phy_data & SR_1000T_REMOTE_RX_STATUS) &&
retry_count) {
msleep(100);
retry_count--;
goto retry_read_status;
} else if (!retry_count) {
dev_err(&adapter->pdev->dev, "exceed max 2 second\n");
}
} else {
dev_err(&adapter->pdev->dev, "read 1000Base-T Status Reg\n");
}
no_wait:
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
netif_carrier_on(netdev);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb_ping_all_vfs(adapter);
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
igb_check_vf_rate_limit(adapter);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: remove redundant timer updates and cleanup watchdog_task The igb watchdog task is modifying the watchdog timer twice duing a single run. It only needs to be called once to reschedule itself for 2 seconds from the last time it ran. In addition I removed the allocation of the mac_info structure since it is only called twice and is easier to access via the e1000_hw struct. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:19:50 +00:00
/* link state has changed, schedule phy info update */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (!test_bit(__IGB_DOWN, &adapter->state))
mod_timer(&adapter->phy_info_timer,
round_jiffies(jiffies + 2 * HZ));
}
} else {
if (netif_carrier_ok(netdev)) {
adapter->link_speed = 0;
adapter->link_duplex = 0;
igb: introduce igb_thermal_sensor_event for sensor checking The code for thermal sensor checking should be wrapped into a function. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-04-05 04:27:15 +00:00
/* check for thermal sensor event */
igb: Convert printks to pr_<level> Based on original patch from Joe Perches <joe@perches.com> Use the current logging styles. pr_<level> conversions are now prefixed with "igb: " Correct a defect where the trailing NTU may have been printed on a separate line because of an interleaving hex_dump. Remove unnecessary uses of KERN_CONT and use single pr_info()s to avoid any possible output interleaving from other modules. Coalesce formats as appropriate. -v2 fix-up to make checkpatch.pl compliant and remove change to the copyright line CC: Joe Perches <joe@perches.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-21 20:01:34 +00:00
if (igb_thermal_sensor_event(hw,
E1000_THSTAT_PWR_DOWN)) {
igb: Cleanups for messaging This patch fixes WARNING:PREFER_PR_LEVEL and WARNING:SPLIT_STRING from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:17 +00:00
netdev_err(netdev, "The network adapter was stopped because it overheated\n");
igb: Add messaging for thermal sensor events on i350 devices This feature adds messaging to the link status change to notify the user if the device returned from a downshift or power off event due to the Thermal Sensor feature in i350 parts. Feature is only available on internal copper ports. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-03-12 08:59:47 +00:00
}
igb: introduce igb_thermal_sensor_event for sensor checking The code for thermal sensor checking should be wrapped into a function. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-04-05 04:27:15 +00:00
igb: link up/down messages must follow a specific format The system log messages created on a link status change need to follow a specific format to work with tools some customers use. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-27 08:21:39 +00:00
/* Links status message must follow this format */
igb: Cleanups for messaging This patch fixes WARNING:PREFER_PR_LEVEL and WARNING:SPLIT_STRING from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:17 +00:00
netdev_info(netdev, "igb: %s NIC Link is Down\n",
igb: link up/down messages must follow a specific format The system log messages created on a link status change need to follow a specific format to work with tools some customers use. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-27 08:21:39 +00:00
netdev->name);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
netif_carrier_off(netdev);
igb: remove redundant timer updates and cleanup watchdog_task The igb watchdog task is modifying the watchdog timer twice duing a single run. It only needs to be called once to reschedule itself for 2 seconds from the last time it ran. In addition I removed the allocation of the mac_info structure since it is only called twice and is easier to access via the e1000_hw struct. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:19:50 +00:00
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb_ping_all_vfs(adapter);
igb: remove redundant timer updates and cleanup watchdog_task The igb watchdog task is modifying the watchdog timer twice duing a single run. It only needs to be called once to reschedule itself for 2 seconds from the last time it ran. In addition I removed the allocation of the mac_info structure since it is only called twice and is easier to access via the e1000_hw struct. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:19:50 +00:00
/* link state has changed, schedule phy info update */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (!test_bit(__IGB_DOWN, &adapter->state))
mod_timer(&adapter->phy_info_timer,
round_jiffies(jiffies + 2 * HZ));
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
/* link is down, time to check for alternate media */
if (adapter->flags & IGB_FLAG_MAS_ENABLE) {
igb_check_swap_media(adapter);
if (adapter->flags & IGB_FLAG_MEDIA_RESET) {
schedule_work(&adapter->reset_task);
/* return immediately */
return;
}
}
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
pm_schedule_suspend(netdev->dev.parent,
MSEC_PER_SEC * 5);
igb: Add new feature Media Auto Sense for 82580 devices only This patch adds support for the hardware feature Media Auto Sense. This feature requires a custom EEPROM image provided by our customer support team. The feature allows hardware designed with dual PHY's, fiber and copper to be used with either media without additional EEPROM changes. Fiber is preferred and driver will swap and configure for fiber media if sensed by the device at any time. Device will swap back to copper if it is the only media detected. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-17 05:36:26 +00:00
/* also check for alternate media here */
} else if (!netif_carrier_ok(netdev) &&
(adapter->flags & IGB_FLAG_MAS_ENABLE)) {
igb_check_swap_media(adapter);
if (adapter->flags & IGB_FLAG_MEDIA_RESET) {
schedule_work(&adapter->reset_task);
/* return immediately */
return;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
}
igb: reduce CPU0 latency when updating statistics This change is based off of the work and suggestion of Jan Jablonsky <jan.jablonsky@thalesgroup.com>. The Watchdog workqueue in igb driver is scheduled every 2s for each network interface. That includes updating a statistics protected by spinlock. Function igb_update_stats in this case will be protected against preemption. According to number of a statistics registers (cca 60), processing this function might cause additional cpu load on CPU0. In case of statistics spinlock may be replaced with mutex, which reduce latency on CPU0. CC: Bernhard Kaindl <bernhard.kaindl@thalesgroup.com> CC: Jan Jablonsky <jan.jablonsky@thalesgroup.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-08-02 17:13:10 +00:00
mutex_lock(&adapter->stats64_lock);
igb: Remove useless argument Given that all callers of igb_update_stats() pass the same two arguments: (adapter, &adapter->stats64), the second argument can be removed. Signed-off-by: Benjamin Poirier <bpoirier@suse.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-16 22:55:16 +00:00
igb_update_stats(adapter);
igb: reduce CPU0 latency when updating statistics This change is based off of the work and suggestion of Jan Jablonsky <jan.jablonsky@thalesgroup.com>. The Watchdog workqueue in igb driver is scheduled every 2s for each network interface. That includes updating a statistics protected by spinlock. Function igb_update_stats in this case will be protected against preemption. According to number of a statistics registers (cca 60), processing this function might cause additional cpu load on CPU0. In case of statistics spinlock may be replaced with mutex, which reduce latency on CPU0. CC: Bernhard Kaindl <bernhard.kaindl@thalesgroup.com> CC: Jan Jablonsky <jan.jablonsky@thalesgroup.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-08-02 17:13:10 +00:00
mutex_unlock(&adapter->stats64_lock);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: check for packets on all tx rings when link is down We were previously only checking the first tx ring to see if it had any packets in it when the link when down. However we should be checking all of the rings so this patch makes it so that all of the rings are now being checked. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:38:16 +00:00
for (i = 0; i < adapter->num_tx_queues; i++) {
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
struct igb_ring *tx_ring = adapter->tx_ring[i];
igb: check for packets on all tx rings when link is down We were previously only checking the first tx ring to see if it had any packets in it when the link when down. However we should be checking all of the rings so this patch makes it so that all of the rings are now being checked. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:38:16 +00:00
if (!netif_carrier_ok(netdev)) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* We've lost link, so the controller stops DMA,
* but we've got queued Tx work that's never going
* to get done, so reset controller to flush Tx.
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* (Do the reset outside of interrupt context).
*/
igb: check for packets on all tx rings when link is down We were previously only checking the first tx ring to see if it had any packets in it when the link when down. However we should be checking all of the rings so this patch makes it so that all of the rings are now being checked. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:38:16 +00:00
if (igb_desc_unused(tx_ring) + 1 < tx_ring->count) {
adapter->tx_timeout_count++;
schedule_work(&adapter->reset_task);
/* return immediately since reset is imminent */
return;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: check for packets on all tx rings when link is down We were previously only checking the first tx ring to see if it had any packets in it when the link when down. However we should be checking all of the rings so this patch makes it so that all of the rings are now being checked. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:38:16 +00:00
/* Force detection of hung controller every watchdog period */
igb: move TX hang check flag into ring->flags This change moves the Tx hang check into the ring flags. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:19 +00:00
set_bit(IGB_RING_FLAG_TX_DETECT_HANG, &tx_ring->flags);
igb: check for packets on all tx rings when link is down We were previously only checking the first tx ring to see if it had any packets in it when the link when down. However we should be checking all of the rings so this patch makes it so that all of the rings are now being checked. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:38:16 +00:00
}
igb: make tx hang check multiqueue, check eop descriptor This change makes the tx hang check run over all tx queues instead of just queue 0. Also have hang display info on EOP descriptor instead of the descriptor at the start of the chain. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:48:51 +00:00
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Cause software interrupt to ensure Rx ring is cleaned */
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
if (adapter->flags & IGB_FLAG_HAS_MSIX) {
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
u32 eics = 0;
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: avoid unnecessarily creating a local copy of the q_vector This is mostly a drop of unnecessary pointer defines for q_vector when we don't have issues with line width and don't have multiple references to the pointer. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:34 +00:00
for (i = 0; i < adapter->num_q_vectors; i++)
eics |= adapter->q_vector[i]->eims_value;
igb: force all queues to interrupt once every 2 seconds Set the EICS bit for each of the RX queues at least once every 2 seconds to prevent the rx queues from stalling. Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-08-26 11:25:03 +00:00
wr32(E1000_EICS, eics);
} else {
wr32(E1000_ICS, E1000_ICS_RXDMT0);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Add Anti-spoofing feature support Add support for the anti-spoofing feature in the HW. Packets from VF devices with spoofed MAC addresses or VLAN tags will be blocked and an event generated. When the watchdog task runs it will call a function to check if any spoof events occurred. If an event was detected then a warning message is dumped to the system log. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 02:08:26 +00:00
igb_spoof_check(adapter);
igb: Add mechanism for detecting latched hardware Rx timestamp Add a check against possible Rx timestamp freezing in the hardware via watchdog mechanism. This situation can occur when an Rx timestamp has been latched, but the packet has been dropped because the Rx ring is full. Whenever a packet comes in that should be timestamped, the Rx timestamp gets latched into the hardware registers and we will store the jiffy value in the rx_ring. The watchdog will keep track of his own jiffy timer whenever there is no valid timestamp in the registers. If the watchdog detects a valid timestamp in the registers, meaning that no Rx packet has consumed it yet, it will check which time is most recent: the last time in the watchdog or any time in the rx_rings. If the most recent "event" was more than 5 seconds ago, it will flush the Rx timestamp and print a warning message to the syslog. Cc: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Matthew Vick <matthew.vick@intel.com> Acked-by: Richard Cochran <richardcochran@gmail.com> Acked-by: Jacob Keller <Jacob.e.keller@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-13 07:20:35 +00:00
igb_ptp_rx_hang(adapter);
igb: check for Tx timestamp timeouts during watchdog The igb driver has logic to handle only one Tx timestamp at a time, using a state bit lock to avoid multiple requests at once. It may be possible, if incredibly unlikely, that a Tx timestamp event is requested but never completes. Since we use an interrupt scheme to determine when the Tx timestamp occurred we would never clear the state bit in this case. Add an igb_ptp_tx_hang() function similar to the already existing igb_ptp_rx_hang() function. This function runs in the watchdog routine and makes sure we eventually recover from this case instead of permanently disabling Tx timestamps. Note: there is no currently known way to cause this without hacking the driver code to force it. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-03 17:29:03 +00:00
igb_ptp_tx_hang(adapter);
igb: Add Anti-spoofing feature support Add support for the anti-spoofing feature in the HW. Packets from VF devices with spoofed MAC addresses or VLAN tags will be blocked and an event generated. When the watchdog task runs it will call a function to check if any spoof events occurred. If an event was detected then a warning message is dumped to the system log. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 02:08:26 +00:00
igb: Add message when malformed packets detected by hw This patch adds a check and prints the error cause register value when the hardware detects a malformed packet. This is a very unlikely scenario but has been seen occasionally, so printing the message to assist the user. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-07-09 04:55:45 +00:00
/* Check LVMMC register on i350/i354 only */
if ((adapter->hw.mac.type == e1000_i350) ||
(adapter->hw.mac.type == e1000_i354))
igb_check_lvmmc(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Reset the timer */
igb: Implementation of 1-sec delay for i210 devices This patch adds 1 sec delay mechanism to i210 device family, in order to avoid erroneous link issue with the link partner. Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-08-28 02:22:43 +00:00
if (!test_bit(__IGB_DOWN, &adapter->state)) {
if (adapter->flags & IGB_FLAG_NEED_LINK_UPDATE)
mod_timer(&adapter->watchdog_timer,
round_jiffies(jiffies + HZ));
else
mod_timer(&adapter->watchdog_timer,
round_jiffies(jiffies + 2 * HZ));
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
enum latency_range {
lowest_latency = 0,
low_latency = 1,
bulk_latency = 2,
latency_invalid = 255
};
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_update_ring_itr - update the dynamic ITR value based on packet size
* @q_vector: pointer to q_vector
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Stores a new ITR value based on strictly on packet size. This
* algorithm is less sophisticated than that used in igb_update_itr,
* due to the difficulty of synchronizing statistics across multiple
* receive rings. The divisors and thresholds used by this function
* were determined based on theoretical maximum wire speed and testing
* data, in order to minimize response time while increasing bulk
* throughput.
igb: remove references to long gone command line parameters Command line parameters QueuePairs, Node, EEE, DMAC and InterruptThrottleRate do not exist these days. Remove all references to them in the Documentation folder and update code comments. Signed-off-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-18 07:26:48 +00:00
* This functionality is controlled by ethtool's coalescing settings.
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* NOTE: This function is called only when operating in a multiqueue
* receive environment.
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
**/
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
static void igb_update_ring_itr(struct igb_q_vector *q_vector)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
int new_val = q_vector->itr_val;
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
int avg_wire_size = 0;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
struct igb_adapter *adapter = q_vector->adapter;
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
unsigned int packets;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
/* For non-gigabit speeds, just fix the interrupt rate at 4000
* ints/sec - ITR timer value of 120 ticks.
*/
if (adapter->link_speed != SPEED_1000) {
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
new_val = IGB_4K_ITR;
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
goto set_itr_val;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
packets = q_vector->rx.total_packets;
if (packets)
avg_wire_size = q_vector->rx.total_bytes / packets;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
packets = q_vector->tx.total_packets;
if (packets)
avg_wire_size = max_t(u32, avg_wire_size,
q_vector->tx.total_bytes / packets);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
/* if avg_wire_size isn't set no work was done */
if (!avg_wire_size)
goto clear_counts;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
/* Add 24 bytes to size to account for CRC, preamble, and gap */
avg_wire_size += 24;
/* Don't starve jumbo frames */
avg_wire_size = min(avg_wire_size, 3000);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
/* Give a little boost to mid-size frames */
if ((avg_wire_size > 300) && (avg_wire_size < 1200))
new_val = avg_wire_size / 3;
else
new_val = avg_wire_size / 2;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
/* conservative mode (itr 3) eliminates the lowest_latency setting */
if (new_val < IGB_20K_ITR &&
((q_vector->rx.ring && adapter->rx_itr_setting == 3) ||
(!q_vector->rx.ring && adapter->tx_itr_setting == 3)))
new_val = IGB_20K_ITR;
igb: cap interrupts at 20K per queue when in itr mode 3 In order to maintain similar performance between MSI-X and legacy/MSI interrupts, this patch reduces the number of interrupts when receiving small packets to 20K when in interrupt throttle rate mode 3. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:03:19 +00:00
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
set_itr_val:
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (new_val != q_vector->itr_val) {
q_vector->itr_val = new_val;
q_vector->set_itr = 1;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
clear_counts:
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
q_vector->rx.total_bytes = 0;
q_vector->rx.total_packets = 0;
q_vector->tx.total_bytes = 0;
q_vector->tx.total_packets = 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_update_itr - update the dynamic ITR value based on statistics
* @q_vector: pointer to q_vector
* @ring_container: ring info to update the itr for
*
* Stores a new ITR value based on packets and byte
* counts during the last interrupt. The advantage of per interrupt
* computation is faster updates and more accurate ITR for the current
* traffic pattern. Constants in this function were computed
* based on theoretical maximum wire speed and thresholds were set based
* on testing data as well as attempting to minimize response time
* while increasing bulk throughput.
igb: remove references to long gone command line parameters Command line parameters QueuePairs, Node, EEE, DMAC and InterruptThrottleRate do not exist these days. Remove all references to them in the Documentation folder and update code comments. Signed-off-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-18 07:26:48 +00:00
* This functionality is controlled by ethtool's coalescing settings.
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* NOTE: These calculations are only valid when operating in a single-
* queue environment.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
static void igb_update_itr(struct igb_q_vector *q_vector,
struct igb_ring_container *ring_container)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
unsigned int packets = ring_container->total_packets;
unsigned int bytes = ring_container->total_bytes;
u8 itrval = ring_container->itr;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
/* no packets, exit with status unchanged */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (packets == 0)
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
return;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
switch (itrval) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
case lowest_latency:
/* handle TSO and jumbo frames */
if (bytes/packets > 8000)
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
itrval = bulk_latency;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
else if ((packets < 5) && (bytes > 512))
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
itrval = low_latency;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
break;
case low_latency: /* 50 usec aka 20000 ints/s */
if (bytes > 10000) {
/* this if handles the TSO accounting */
igb: Cleanups to fix braces location warnings This patch fixes WARNING:BRACES and ERROR:OPEN_BRACE from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:23 +00:00
if (bytes/packets > 8000)
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
itrval = bulk_latency;
igb: Cleanups to fix braces location warnings This patch fixes WARNING:BRACES and ERROR:OPEN_BRACE from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:23 +00:00
else if ((packets < 10) || ((bytes/packets) > 1200))
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
itrval = bulk_latency;
igb: Cleanups to fix braces location warnings This patch fixes WARNING:BRACES and ERROR:OPEN_BRACE from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:23 +00:00
else if ((packets > 35))
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
itrval = lowest_latency;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
} else if (bytes/packets > 2000) {
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
itrval = bulk_latency;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
} else if (packets <= 2 && bytes < 512) {
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
itrval = lowest_latency;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
break;
case bulk_latency: /* 250 usec aka 4000 ints/s */
if (bytes > 25000) {
if (packets > 35)
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
itrval = low_latency;
igb: reduce size required to trigger low latency Update the Adaptive Interrupt Moderation algorithm so that the low latency state is triggered less easily to prevent high interrupt loads. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-12 18:17:21 +00:00
} else if (bytes < 1500) {
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
itrval = low_latency;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
break;
}
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
/* clear work counters since we have the values we need */
ring_container->total_bytes = 0;
ring_container->total_packets = 0;
/* write updated itr to ring container */
ring_container->itr = itrval;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
static void igb_set_itr(struct igb_q_vector *q_vector)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
struct igb_adapter *adapter = q_vector->adapter;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
u32 new_itr = q_vector->itr_val;
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
u8 current_itr = 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* for non-gigabit speeds, just fix the interrupt rate at 4000 */
if (adapter->link_speed != SPEED_1000) {
current_itr = 0;
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
new_itr = IGB_4K_ITR;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
goto set_itr_now;
}
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
igb_update_itr(q_vector, &q_vector->tx);
igb_update_itr(q_vector, &q_vector->rx);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
current_itr = max(q_vector->rx.itr, q_vector->tx.itr);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
/* conservative mode (itr 3) eliminates the lowest_latency setting */
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
if (current_itr == lowest_latency &&
((q_vector->rx.ring && adapter->rx_itr_setting == 3) ||
(!q_vector->rx.ring && adapter->tx_itr_setting == 3)))
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
current_itr = low_latency;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
switch (current_itr) {
/* counts and packets in update_itr are dependent on these numbers */
case lowest_latency:
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
new_itr = IGB_70K_ITR; /* 70,000 ints/sec */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
break;
case low_latency:
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
new_itr = IGB_20K_ITR; /* 20,000 ints/sec */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
break;
case bulk_latency:
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
new_itr = IGB_4K_ITR; /* 4,000 ints/sec */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
break;
default:
break;
}
set_itr_now:
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (new_itr != q_vector->itr_val) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* this attempts to bias the interrupt rate towards Bulk
* by adding intermediate steps when interrupt rate is
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* increasing
*/
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
new_itr = new_itr > q_vector->itr_val ?
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
max((new_itr * q_vector->itr_val) /
(new_itr + (q_vector->itr_val >> 2)),
new_itr) : new_itr;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Don't write the value here; it resets the adapter's
* internal timer, and causes us to delay far longer than
* we should between interrupts. Instead, we write the ITR
* value at the beginning of the next interrupt so the timing
* ends up being correct.
*/
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
q_vector->itr_val = new_itr;
q_vector->set_itr = 1;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
}
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
static void igb_tx_ctxtdesc(struct igb_ring *tx_ring,
struct igb_tx_buffer *first,
u32 vlan_macip_lens, u32 type_tucmd,
u32 mss_l4len_idx)
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
{
struct e1000_adv_tx_context_desc *context_desc;
u16 i = tx_ring->next_to_use;
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
struct timespec64 ts;
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
context_desc = IGB_TX_CTXTDESC(tx_ring, i);
i++;
tx_ring->next_to_use = (i < tx_ring->count) ? i : 0;
/* set bits to identify this as an advanced context descriptor */
type_tucmd |= E1000_TXD_CMD_DEXT | E1000_ADVTXD_DTYP_CTXT;
/* For 82575, context index must be unique per ring. */
igb: Consolidate all of the ring feature flags into a single value This change moves all of the ring flags into a single value. The advantage to this is that there is one central area for all of these flags and they can all make use of the set/test bit operations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:36 +00:00
if (test_bit(IGB_RING_FLAG_TX_CTX_IDX, &tx_ring->flags))
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
mss_l4len_idx |= tx_ring->reg_idx << 4;
context_desc->vlan_macip_lens = cpu_to_le32(vlan_macip_lens);
context_desc->type_tucmd_mlhl = cpu_to_le32(type_tucmd);
context_desc->mss_l4len_idx = cpu_to_le32(mss_l4len_idx);
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
/* We assume there is always a valid tx time available. Invalid times
* should have been handled by the upper layers.
*/
if (tx_ring->launchtime_enable) {
ts = ns_to_timespec64(first->skb->tstamp);
context_desc->seqnum_seed = cpu_to_le32(ts.tv_nsec / 32);
} else {
context_desc->seqnum_seed = 0;
}
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
}
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
static int igb_tso(struct igb_ring *tx_ring,
struct igb_tx_buffer *first,
u8 *hdr_len)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
u32 vlan_macip_lens, type_tucmd, mss_l4len_idx;
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
struct sk_buff *skb = first->skb;
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
union {
struct iphdr *v4;
struct ipv6hdr *v6;
unsigned char *hdr;
} ip;
union {
struct tcphdr *tcp;
unsigned char *hdr;
} l4;
u32 paylen, l4_offset;
igb: remove open-coded skb_cow_head Signed-off-by: Francois Romieu <romieu@fr.zoreil.com> Cc: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-30 03:14:11 +00:00
int err;
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
igb: Make TSO check for CHECKSUM_PARTIAL to avoid skb_is_gso check This change is meant to reduce the overhead for workloads that are not using either TSO or checksum offloads. Most of the time the compiler should jump ahead after failing this check to the VLAN check since in the igb_tx_csum call we start with that check as well. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:22 +00:00
if (skb->ip_summed != CHECKSUM_PARTIAL)
return 0;
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
if (!skb_is_gso(skb))
return 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: remove open-coded skb_cow_head Signed-off-by: Francois Romieu <romieu@fr.zoreil.com> Cc: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-30 03:14:11 +00:00
err = skb_cow_head(skb, 0);
if (err < 0)
return err;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
ip.hdr = skb_network_header(skb);
l4.hdr = skb_checksum_start(skb);
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
/* ADV DTYP TUCMD MKRLOC/ISCSIHEDLEN */
type_tucmd = E1000_ADVTXD_TUCMD_L4T_TCP;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
/* initialize outer IP header fields */
if (ip.v4->version == 4) {
igb/igbvf: Don't use lco_csum to compute IPv4 checksum In the case of IPIP and SIT tunnel frames the outer transport header offset is actually set to the same offset as the inner transport header. This results in the lco_csum call not doing any checksum computation over the inner IPv4/v6 header data. In order to account for that I am updating the code so that we determine the location to start the checksum ourselves based on the location of the IPv4 header and the length. Fixes: e10715d3e961 ("igb/igbvf: Add support for GSO partial") Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-11-28 15:42:23 +00:00
unsigned char *csum_start = skb_checksum_start(skb);
unsigned char *trans_start = ip.hdr + (ip.v4->ihl * 4);
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
/* IP header will have to cancel out any data that
* is not a part of the outer IP header
*/
igb/igbvf: Don't use lco_csum to compute IPv4 checksum In the case of IPIP and SIT tunnel frames the outer transport header offset is actually set to the same offset as the inner transport header. This results in the lco_csum call not doing any checksum computation over the inner IPv4/v6 header data. In order to account for that I am updating the code so that we determine the location to start the checksum ourselves based on the location of the IPv4 header and the length. Fixes: e10715d3e961 ("igb/igbvf: Add support for GSO partial") Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-11-28 15:42:23 +00:00
ip.v4->check = csum_fold(csum_partial(trans_start,
csum_start - trans_start,
0));
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
type_tucmd |= E1000_ADVTXD_TUCMD_IPV4;
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
ip.v4->tot_len = 0;
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
first->tx_flags |= IGB_TX_FLAGS_TSO |
IGB_TX_FLAGS_CSUM |
IGB_TX_FLAGS_IPV4;
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
} else {
ip.v6->payload_len = 0;
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
first->tx_flags |= IGB_TX_FLAGS_TSO |
IGB_TX_FLAGS_CSUM;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
/* determine offset of inner transport header */
l4_offset = l4.hdr - skb->data;
/* compute length of segmentation header */
*hdr_len = (l4.tcp->doff * 4) + l4_offset;
/* remove payload length from inner checksum */
paylen = skb->len - l4_offset;
csum_replace_by_diff(&l4.tcp->check, htonl(paylen));
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
/* update gso size and bytecount with header size */
first->gso_segs = skb_shinfo(skb)->gso_segs;
first->bytecount += (first->gso_segs - 1) * *hdr_len;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* MSS L4LEN IDX */
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
mss_l4len_idx = (*hdr_len - l4_offset) << E1000_ADVTXD_L4LEN_SHIFT;
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
mss_l4len_idx |= skb_shinfo(skb)->gso_size << E1000_ADVTXD_MSS_SHIFT;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
/* VLAN MACLEN IPLEN */
igb/igbvf: Add support for GSO partial This patch adds support for partial GSO segmentation in the case of tunnels. Specifically with this change the driver an perform segmentation as long as the frame either has IPv6 inner headers, or we are allowed to mangle the IP IDs on the inner header. This is needed because we will not be modifying any fields from the start of the start of the outer transport header to the start of the inner transport header as we are treating them like they are just a block of IP options. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-14 21:19:38 +00:00
vlan_macip_lens = l4.hdr - ip.hdr;
vlan_macip_lens |= (ip.hdr - skb->data) << E1000_ADVTXD_MACLEN_SHIFT;
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
vlan_macip_lens |= first->tx_flags & IGB_TX_FLAGS_VLAN_MASK;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
igb_tx_ctxtdesc(tx_ring, first, vlan_macip_lens,
type_tucmd, mss_l4len_idx);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
return 1;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Add support for generic Tx checksums This patch adds support for generic Tx checksums to the igb driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-13 15:31:23 +00:00
static inline bool igb_ipv6_csum_is_sctp(struct sk_buff *skb)
{
unsigned int offset = 0;
ipv6_find_hdr(skb, &offset, IPPROTO_SCTP, NULL, NULL);
return offset == skb_checksum_start_offset(skb);
}
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
static void igb_tx_csum(struct igb_ring *tx_ring, struct igb_tx_buffer *first)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
struct sk_buff *skb = first->skb;
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
u32 vlan_macip_lens = 0;
u32 type_tucmd = 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
if (skb->ip_summed != CHECKSUM_PARTIAL) {
igb: Add support for generic Tx checksums This patch adds support for generic Tx checksums to the igb driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-13 15:31:23 +00:00
csum_failed:
igb: Use an advanced ctx descriptor for launchtime On i210, Launchtime (TxTime) requires the usage of an "Advanced Transmit Context Descriptor" for retrieving the timestamp of a packet. The igb driver correctly builds such descriptor on the segmentation flow (i.e. igb_tso()) or on the checksum one (i.e. igb_tx_csum()), but the feature is broken for AF_PACKET if the IGB_TX_FLAGS_VLAN is not set, which happens due to an early return on igb_tx_csum(). This flag is only set by the kernel when a VLAN interface is used, thus we can't just rely on it. Here we are fixing this issue by checking if launchtime is enabled for the current tx_ring before performing the early return. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-07-26 17:20:38 +00:00
if (!(first->tx_flags & IGB_TX_FLAGS_VLAN) &&
!tx_ring->launchtime_enable)
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
return;
igb: Add support for generic Tx checksums This patch adds support for generic Tx checksums to the igb driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-13 15:31:23 +00:00
goto no_csum;
}
igb: allow tx of pre-formatted vlan tagged packets When the 82575 is fed 802.1q packets, it chokes with an error of the form: igb 0000:08:00.1 partial checksum but proto=81! As the logic there was not smart enough to look into the vlan header to pick out the encapsulated protocol. There are times when we'd like to send these packets out without having to configure a vlan on the interface. Here we check for the vlan tag and allow the packet to go out with the correct hardware checksum. Thanks to Kand Ly <kand@riverbed.com> for discovering the issue and the coming up with a solution. This patch is based upon his work. Signed-off-by: Arthur Jones <ajones@riverbed.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-21 23:55:07 +00:00
igb: Add support for generic Tx checksums This patch adds support for generic Tx checksums to the igb driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-13 15:31:23 +00:00
switch (skb->csum_offset) {
case offsetof(struct tcphdr, check):
type_tucmd = E1000_ADVTXD_TUCMD_L4T_TCP;
/* fall through */
case offsetof(struct udphdr, check):
break;
case offsetof(struct sctphdr, checksum):
/* validate that this is actually an SCTP request */
if (((first->protocol == htons(ETH_P_IP)) &&
(ip_hdr(skb)->protocol == IPPROTO_SCTP)) ||
((first->protocol == htons(ETH_P_IPV6)) &&
igb_ipv6_csum_is_sctp(skb))) {
type_tucmd = E1000_ADVTXD_TUCMD_L4T_SCTP;
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
break;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb_main: Mark expected switch fall-throughs In preparation to enabling -Wimplicit-fallthrough, mark switch cases where we are expecting to fall through. Addresses-Coverity-ID: 200521 ("Missing break in switch") Addresses-Coverity-ID: 114797 ("Missing break in switch") Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-08-07 23:22:22 +00:00
/* fall through */
igb: Add support for generic Tx checksums This patch adds support for generic Tx checksums to the igb driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-13 15:31:23 +00:00
default:
skb_checksum_help(skb);
goto csum_failed;
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Add support for generic Tx checksums This patch adds support for generic Tx checksums to the igb driver. It turns out this is actually pretty easy after going over the datasheet as we were doing a number of steps we didn't need to. In order to perform a Tx checksum for an L4 header we need to fill in the following fields in the Tx descriptor: MACLEN (maximum of 127), retrieved from: skb_network_offset() IPLEN (maximum of 511), retrieved from: skb_checksum_start_offset() - skb_network_offset() TUCMD.L4T indicates offset and if checksum or crc32c, based on: skb->csum_offset The added advantage to doing this is that we can support inner checksum offloads for tunnels and MPLS while still being able to transparently insert VLAN tags. I also took the opportunity to clean-up many of the feature flag configuration bits to make them a bit more consistent between drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-13 15:31:23 +00:00
/* update TX checksum flag */
first->tx_flags |= IGB_TX_FLAGS_CSUM;
vlan_macip_lens = skb_checksum_start_offset(skb) -
skb_network_offset(skb);
no_csum:
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
vlan_macip_lens |= skb_network_offset(skb) << E1000_ADVTXD_MACLEN_SHIFT;
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
vlan_macip_lens |= first->tx_flags & IGB_TX_FLAGS_VLAN_MASK;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Add support for ETF offload Implement HW offload support for SO_TXTIME through igb's Launchtime feature. This is done by extending igb_setup_tc() so it supports TC_SETUP_QDISC_ETF and configuring i210 so time based transmit arbitration is enabled. The FQTSS transmission mode added before is extended so strict priority (SP) queues wait for stream reservation (SR) ones. igb_config_tx_modes() is extended so it can support enabling/disabling Launchtime following the previous approach used for the credit-based shaper (CBS). As the previous flow, FQTSS transmission mode is enabled automatically by the driver once Launchtime (or CBS, as before) is enabled. Similarly, it's automatically disabled when the feature is disabled for the last queue that had it setup on. The driver just consumes the transmit times from the skbuffs directly, so no special handling is done in case an 'invalid' time is provided. We assume this has been handled by the ETF qdisc already. Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-03 22:42:59 +00:00
igb_tx_ctxtdesc(tx_ring, first, vlan_macip_lens, type_tucmd, 0);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
#define IGB_SET_FLAG(_input, _flag, _result) \
((_flag <= _result) ? \
((u32)(_input & _flag) * (_result / _flag)) : \
((u32)(_input & _flag) / (_flag / _result)))
static u32 igb_tx_cmd_type(struct sk_buff *skb, u32 tx_flags)
igb: Create separate functions for generating cmd_type and olinfo This change is meant to improve the readability of the driver by separating out the cmd_type configuration and the olinfo configuration into their own functions. By doing this it is much easier to determine which ingredients go into setting up these to portions of the descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:48 +00:00
{
/* set type for advanced descriptor with frame checksum insertion */
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
u32 cmd_type = E1000_ADVTXD_DTYP_DATA |
E1000_ADVTXD_DCMD_DEXT |
E1000_ADVTXD_DCMD_IFCS;
igb: Create separate functions for generating cmd_type and olinfo This change is meant to improve the readability of the driver by separating out the cmd_type configuration and the olinfo configuration into their own functions. By doing this it is much easier to determine which ingredients go into setting up these to portions of the descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:48 +00:00
/* set HW vlan bit if vlan is present */
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
cmd_type |= IGB_SET_FLAG(tx_flags, IGB_TX_FLAGS_VLAN,
(E1000_ADVTXD_DCMD_VLE));
/* set segmentation bits for TSO */
cmd_type |= IGB_SET_FLAG(tx_flags, IGB_TX_FLAGS_TSO,
(E1000_ADVTXD_DCMD_TSE));
igb: Create separate functions for generating cmd_type and olinfo This change is meant to improve the readability of the driver by separating out the cmd_type configuration and the olinfo configuration into their own functions. By doing this it is much easier to determine which ingredients go into setting up these to portions of the descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:48 +00:00
/* set timestamp bit if present */
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
cmd_type |= IGB_SET_FLAG(tx_flags, IGB_TX_FLAGS_TSTAMP,
(E1000_ADVTXD_MAC_TSTAMP));
igb: Create separate functions for generating cmd_type and olinfo This change is meant to improve the readability of the driver by separating out the cmd_type configuration and the olinfo configuration into their own functions. By doing this it is much easier to determine which ingredients go into setting up these to portions of the descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:48 +00:00
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
/* insert frame checksum */
cmd_type ^= IGB_SET_FLAG(skb->no_fcs, 1, E1000_ADVTXD_DCMD_IFCS);
igb: Create separate functions for generating cmd_type and olinfo This change is meant to improve the readability of the driver by separating out the cmd_type configuration and the olinfo configuration into their own functions. By doing this it is much easier to determine which ingredients go into setting up these to portions of the descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:48 +00:00
return cmd_type;
}
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
static void igb_tx_olinfo_status(struct igb_ring *tx_ring,
union e1000_adv_tx_desc *tx_desc,
u32 tx_flags, unsigned int paylen)
igb: Create separate functions for generating cmd_type and olinfo This change is meant to improve the readability of the driver by separating out the cmd_type configuration and the olinfo configuration into their own functions. By doing this it is much easier to determine which ingredients go into setting up these to portions of the descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:48 +00:00
{
u32 olinfo_status = paylen << E1000_ADVTXD_PAYLEN_SHIFT;
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
/* 82575 requires a unique index per ring */
if (test_bit(IGB_RING_FLAG_TX_CTX_IDX, &tx_ring->flags))
igb: Create separate functions for generating cmd_type and olinfo This change is meant to improve the readability of the driver by separating out the cmd_type configuration and the olinfo configuration into their own functions. By doing this it is much easier to determine which ingredients go into setting up these to portions of the descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:48 +00:00
olinfo_status |= tx_ring->reg_idx << 4;
/* insert L4 checksum */
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
olinfo_status |= IGB_SET_FLAG(tx_flags,
IGB_TX_FLAGS_CSUM,
(E1000_TXD_POPTS_TXSM << 8));
igb: Create separate functions for generating cmd_type and olinfo This change is meant to improve the readability of the driver by separating out the cmd_type configuration and the olinfo configuration into their own functions. By doing this it is much easier to determine which ingredients go into setting up these to portions of the descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:48 +00:00
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
/* insert IPv4 checksum */
olinfo_status |= IGB_SET_FLAG(tx_flags,
IGB_TX_FLAGS_IPV4,
(E1000_TXD_POPTS_IXSM << 8));
igb: Create separate functions for generating cmd_type and olinfo This change is meant to improve the readability of the driver by separating out the cmd_type configuration and the olinfo configuration into their own functions. By doing this it is much easier to determine which ingredients go into setting up these to portions of the descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:48 +00:00
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
tx_desc->read.olinfo_status = cpu_to_le32(olinfo_status);
igb: Create separate functions for generating cmd_type and olinfo This change is meant to improve the readability of the driver by separating out the cmd_type configuration and the olinfo configuration into their own functions. By doing this it is much easier to determine which ingredients go into setting up these to portions of the descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:48 +00:00
}
igb: flush when in xmit_more mode and under descriptor pressure Mirror the changes made to ixgbe in commit 2367a17390138f68b3aa28f2f220b8d7ff8d91f4 ("ixgbe: flush when in xmit_more mode and under descriptor pressure") Signed-off-by: David S. Miller <davem@davemloft.net>
2014-08-28 08:39:31 +00:00
static int __igb_maybe_stop_tx(struct igb_ring *tx_ring, const u16 size)
{
struct net_device *netdev = tx_ring->netdev;
netif_stop_subqueue(netdev, tx_ring->queue_index);
/* Herbert's original patch had:
* smp_mb__after_netif_stop_queue();
* but since that doesn't exist yet, just open code it.
*/
smp_mb();
/* We need to check again in a case another CPU has just
* made room available.
*/
if (igb_desc_unused(tx_ring) < size)
return -EBUSY;
/* A reprieve! */
netif_wake_subqueue(netdev, tx_ring->queue_index);
u64_stats_update_begin(&tx_ring->tx_syncp2);
tx_ring->tx_stats.restart_queue2++;
u64_stats_update_end(&tx_ring->tx_syncp2);
return 0;
}
static inline int igb_maybe_stop_tx(struct igb_ring *tx_ring, const u16 size)
{
if (igb_desc_unused(tx_ring) >= size)
return 0;
return __igb_maybe_stop_tx(tx_ring, size);
}
igb: avoid permanent lock of *_PTP_TX_IN_PROGRESS The igb driver uses a state bit lock to avoid handling more than one Tx timestamp request at once. This is required because hardware is limited to a single set of registers for Tx timestamps. The state bit lock is not properly cleaned up during igb_xmit_frame_ring() if the transmit fails such as due to DMA or TSO failure. In some hardware this results in blocking timestamps until the service task times out. In other hardware this results in a permanent lock of the timestamp bit because we never receive an interrupt indicating the timestamp occurred, since indeed the packet was never transmitted. Fix this by checking for DMA and TSO errors in igb_xmit_frame_ring() and properly cleaning up after ourselves when these occur. Reported-by: Reported-by: David Mirabito <davidm@metamako.com> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-03 17:28:55 +00:00
static int igb_tx_map(struct igb_ring *tx_ring,
struct igb_tx_buffer *first,
const u8 hdr_len)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
struct sk_buff *skb = first->skb;
igb: Use dma_unmap_addr and dma_unmap_len defines This change is meant to improve performance on systems that do not require the DMA unmap calls. On those systems we do not need to make use of the unmap address for Tx or the unmap length so we can drop both thereby reducing the size of the Tx buffer info structure. In addition I have changed the logic to check for unmap length instead of unmap address when checking to see if a buffer needs to be unmapped from DMA use. The reasons for this change is that on some platforms it is possible to receive a valid DMA address of 0 from an IOMMU. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-18 01:56:27 +00:00
struct igb_tx_buffer *tx_buffer;
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
union e1000_adv_tx_desc *tx_desc;
igb: Improve performance and reduce size of igb_tx_map This change is meant to both improve the performance and reduce the size of igb_tx_map. To do this I have expanded the work done in the main loop by pushing first into tx_buffer. This allows us to pull in the dma_mapping_error check, the tx_buffer value assignment, and the initial DMA value assignment to the Tx descriptor. The net result is that the function reduces in size by a little over a 100 bytes and is about 1% or 2% faster. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:24 +00:00
struct skb_frag_struct *frag;
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
dma_addr_t dma;
igb: Improve performance and reduce size of igb_tx_map This change is meant to both improve the performance and reduce the size of igb_tx_map. To do this I have expanded the work done in the main loop by pushing first into tx_buffer. This allows us to pull in the dma_mapping_error check, the tx_buffer value assignment, and the initial DMA value assignment to the Tx descriptor. The net result is that the function reduces in size by a little over a 100 bytes and is about 1% or 2% faster. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:24 +00:00
unsigned int data_len, size;
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
u32 tx_flags = first->tx_flags;
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
u32 cmd_type = igb_tx_cmd_type(skb, tx_flags);
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
u16 i = tx_ring->next_to_use;
tx_desc = IGB_TX_DESC(tx_ring, i);
igb: Improve performance and reduce size of igb_tx_map This change is meant to both improve the performance and reduce the size of igb_tx_map. To do this I have expanded the work done in the main loop by pushing first into tx_buffer. This allows us to pull in the dma_mapping_error check, the tx_buffer value assignment, and the initial DMA value assignment to the Tx descriptor. The net result is that the function reduces in size by a little over a 100 bytes and is about 1% or 2% faster. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:24 +00:00
igb_tx_olinfo_status(tx_ring, tx_desc, tx_flags, skb->len - hdr_len);
size = skb_headlen(skb);
data_len = skb->data_len;
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
dma = dma_map_single(tx_ring->dev, skb->data, size, DMA_TO_DEVICE);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Improve performance and reduce size of igb_tx_map This change is meant to both improve the performance and reduce the size of igb_tx_map. To do this I have expanded the work done in the main loop by pushing first into tx_buffer. This allows us to pull in the dma_mapping_error check, the tx_buffer value assignment, and the initial DMA value assignment to the Tx descriptor. The net result is that the function reduces in size by a little over a 100 bytes and is about 1% or 2% faster. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:24 +00:00
tx_buffer = first;
for (frag = &skb_shinfo(skb)->frags[0];; frag++) {
if (dma_mapping_error(tx_ring->dev, dma))
goto dma_error;
/* record length, and DMA address */
dma_unmap_len_set(tx_buffer, len, size);
dma_unmap_addr_set(tx_buffer, dma, dma);
tx_desc->read.buffer_addr = cpu_to_le64(dma);
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
while (unlikely(size > IGB_MAX_DATA_PER_TXD)) {
tx_desc->read.cmd_type_len =
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
cpu_to_le32(cmd_type ^ IGB_MAX_DATA_PER_TXD);
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
i++;
tx_desc++;
if (i == tx_ring->count) {
tx_desc = IGB_TX_DESC(tx_ring, 0);
i = 0;
}
igb: Improve performance and reduce size of igb_tx_map This change is meant to both improve the performance and reduce the size of igb_tx_map. To do this I have expanded the work done in the main loop by pushing first into tx_buffer. This allows us to pull in the dma_mapping_error check, the tx_buffer value assignment, and the initial DMA value assignment to the Tx descriptor. The net result is that the function reduces in size by a little over a 100 bytes and is about 1% or 2% faster. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:24 +00:00
tx_desc->read.olinfo_status = 0;
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
dma += IGB_MAX_DATA_PER_TXD;
size -= IGB_MAX_DATA_PER_TXD;
tx_desc->read.buffer_addr = cpu_to_le64(dma);
}
if (likely(!data_len))
break;
igb: Combine all flag info fields into a single tx_flags structure This change is meant to combine all of the TX flags fields into one u32 flags field so that it can be stored into the tx_buffer_info structure. This includes the time stamp flag as well as mapped_as_page flag info. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:59 +00:00
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
tx_desc->read.cmd_type_len = cpu_to_le32(cmd_type ^ size);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: cleanup tx dma so map & unmap use matching calls The igb driver was using map_single to map the skbs and then unmap_page to unmap them. This update changes that so instead uses skb_dma_map and skb_dma_unmap. In addition the next_to_watch member of the buffer_info struct was being set uneccesarily. I removed the spots where it was set without being needed. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-20 00:17:43 +00:00
i++;
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
tx_desc++;
if (i == tx_ring->count) {
tx_desc = IGB_TX_DESC(tx_ring, 0);
igb: cleanup tx dma so map & unmap use matching calls The igb driver was using map_single to map the skbs and then unmap_page to unmap them. This update changes that so instead uses skb_dma_map and skb_dma_unmap. In addition the next_to_watch member of the buffer_info struct was being set uneccesarily. I removed the spots where it was set without being needed. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-20 00:17:43 +00:00
i = 0;
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
}
igb: Improve performance and reduce size of igb_tx_map This change is meant to both improve the performance and reduce the size of igb_tx_map. To do this I have expanded the work done in the main loop by pushing first into tx_buffer. This allows us to pull in the dma_mapping_error check, the tx_buffer value assignment, and the initial DMA value assignment to the Tx descriptor. The net result is that the function reduces in size by a little over a 100 bytes and is about 1% or 2% faster. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:24 +00:00
tx_desc->read.olinfo_status = 0;
igb: cleanup tx dma so map & unmap use matching calls The igb driver was using map_single to map the skbs and then unmap_page to unmap them. This update changes that so instead uses skb_dma_map and skb_dma_unmap. In addition the next_to_watch member of the buffer_info struct was being set uneccesarily. I removed the spots where it was set without being needed. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-20 00:17:43 +00:00
net: add skb frag size accessors To ease skb->truesize sanitization, its better to be able to localize all references to skb frags size. Define accessors : skb_frag_size() to fetch frag size, and skb_frag_size_{set|add|sub}() to manipulate it. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-10-18 21:00:24 +00:00
size = skb_frag_size(frag);
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
data_len -= size;
dma = skb_frag_dma_map(tx_ring->dev, frag, 0,
igb: Improve performance and reduce size of igb_tx_map This change is meant to both improve the performance and reduce the size of igb_tx_map. To do this I have expanded the work done in the main loop by pushing first into tx_buffer. This allows us to pull in the dma_mapping_error check, the tx_buffer value assignment, and the initial DMA value assignment to the Tx descriptor. The net result is that the function reduces in size by a little over a 100 bytes and is about 1% or 2% faster. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:24 +00:00
size, DMA_TO_DEVICE);
igb: remove use of skb_dma_map from driver This change removes skb_dma_map/unmap calls from the igb driver due to the fact that the call is incompatible with iommu enabled kernels. In order to prevent warnings about using the wrong unmap call I have added a mapped_as_page value to the buffer_info structure to track if the mapped region is a page or a buffer. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-02 16:47:18 +00:00
igb: Use dma_unmap_addr and dma_unmap_len defines This change is meant to improve performance on systems that do not require the DMA unmap calls. On those systems we do not need to make use of the unmap address for Tx or the unmap length so we can drop both thereby reducing the size of the Tx buffer info structure. In addition I have changed the logic to check for unmap length instead of unmap address when checking to see if a buffer needs to be unmapped from DMA use. The reasons for this change is that on some platforms it is possible to receive a valid DMA address of 0 from an IOMMU. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-18 01:56:27 +00:00
tx_buffer = &tx_ring->tx_buffer_info[i];
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
/* write last descriptor with RS and EOP bits */
igb: Update igb Tx flags to improve code efficiency This change is meant to improve the efficiency of the Tx flags in igb by aligning them with the values that will later be written into either the cmd_type or olinfo. By doing this we are able to reduce most of these functions to either just a simple shift followed by an or in the case of cmd_type, or an and followed by an or in the case of olinfo. In order to avoid type conversion errors I also adjusted the locations where we were switching between CPU and little endian. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:23 +00:00
cmd_type |= size | IGB_TXD_DCMD;
tx_desc->read.cmd_type_len = cpu_to_le32(cmd_type);
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
igb: Improve performance and reduce size of igb_tx_map This change is meant to both improve the performance and reduce the size of igb_tx_map. To do this I have expanded the work done in the main loop by pushing first into tx_buffer. This allows us to pull in the dma_mapping_error check, the tx_buffer value assignment, and the initial DMA value assignment to the Tx descriptor. The net result is that the function reduces in size by a little over a 100 bytes and is about 1% or 2% faster. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-11-13 04:03:24 +00:00
netdev_tx_sent_queue(txring_txq(tx_ring), first->bytecount);
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
/* set the timestamp */
first->time_stamp = jiffies;
intel-ethernet: software timestamp skbs as late as possible Many of the Intel Ethernet drivers call skb_tx_timestamp() earlier than necessary. Move the calls to this function to the latest point possible, just prior to notifying hardware of the new Tx packet when we bump the tail register. This affects i40e, iavf, igb, igc, and ixgbe. The e100, e1000, e1000e, fm10k, and ice drivers already call the skb_tx_timestamp() function just prior to indicating the Tx packet to hardware, so they do not need to be changed. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-10-05 16:33:56 +00:00
skb_tx_timestamp(skb);
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Force memory writes to complete before letting h/w know there
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
* are new descriptors to fetch. (Only applicable for weak-ordered
* memory model archs, such as IA-64).
*
* We also need this memory barrier to make certain all of the
* status bits have been updated before next_to_watch is written.
*/
igb: Use dma_wmb() instead of wmb() before doorbell writes igb writes to doorbells to post transmit and receive descriptors; after writing descriptors to memory but before writing to doorbells, use dma_wmb() rather than wmb(). wmb() is more heavyweight than necessary before doorbell writes. On x86, this avoids SFENCEs before doorbell writes in both the tx and rx refill paths. Signed-off-by: Venkatesh Srinivas <venkateshs@google.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-05-25 04:13:21 +00:00
dma_wmb();
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
/* set next_to_watch value indicating a packet is present */
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
first->next_to_watch = tx_desc;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
i++;
if (i == tx_ring->count)
i = 0;
igb: remove use of skb_dma_map from driver This change removes skb_dma_map/unmap calls from the igb driver due to the fact that the call is incompatible with iommu enabled kernels. In order to prevent warnings about using the wrong unmap call I have added a mapped_as_page value to the buffer_info structure to track if the mapped region is a page or a buffer. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-02 16:47:18 +00:00
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
tx_ring->next_to_use = i;
igb: remove use of skb_dma_map from driver This change removes skb_dma_map/unmap calls from the igb driver due to the fact that the call is incompatible with iommu enabled kernels. In order to prevent warnings about using the wrong unmap call I have added a mapped_as_page value to the buffer_info structure to track if the mapped region is a page or a buffer. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-02 16:47:18 +00:00
igb: flush when in xmit_more mode and under descriptor pressure Mirror the changes made to ixgbe in commit 2367a17390138f68b3aa28f2f220b8d7ff8d91f4 ("ixgbe: flush when in xmit_more mode and under descriptor pressure") Signed-off-by: David S. Miller <davem@davemloft.net>
2014-08-28 08:39:31 +00:00
/* Make sure there is space in the ring for the next send. */
igb_maybe_stop_tx(tx_ring, DESC_NEEDED);
if (netif_xmit_stopped(txring_txq(tx_ring)) || !skb->xmit_more) {
net: Remove ndo_xmit_flush netdev operation, use signalling instead. As reported by Jesper Dangaard Brouer, for high packet rates the overhead of having another indirect call in the TX path is non-trivial. There is the indirect call itself, and then there is all of the reloading of the state to refetch the tail pointer value and then write the device register. Move to a more passive scheme, which requires very light modifications to the device drivers. The signal is a new skb->xmit_more value, if it is non-zero it means that more SKBs are pending to be transmitted on the same queue as the current SKB. And therefore, the driver may elide the tail pointer update. Right now skb->xmit_more is always zero. Signed-off-by: David S. Miller <davem@davemloft.net>
2014-08-25 22:51:53 +00:00
writel(i, tx_ring->tail);
/* we need this if more than one processor can write to our tail
* at a time, it synchronizes IO on IA64/Altix systems
*/
mmiowb();
}
igb: avoid permanent lock of *_PTP_TX_IN_PROGRESS The igb driver uses a state bit lock to avoid handling more than one Tx timestamp request at once. This is required because hardware is limited to a single set of registers for Tx timestamps. The state bit lock is not properly cleaned up during igb_xmit_frame_ring() if the transmit fails such as due to DMA or TSO failure. In some hardware this results in blocking timestamps until the service task times out. In other hardware this results in a permanent lock of the timestamp bit because we never receive an interrupt indicating the timestamp occurred, since indeed the packet was never transmitted. Fix this by checking for DMA and TSO errors in igb_xmit_frame_ring() and properly cleaning up after ourselves when these occur. Reported-by: Reported-by: David Mirabito <davidm@metamako.com> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-03 17:28:55 +00:00
return 0;
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
dma_error:
dev_err(tx_ring->dev, "TX DMA map failed\n");
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
tx_buffer = &tx_ring->tx_buffer_info[i];
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
/* clear dma mappings for failed tx_buffer_info map */
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
while (tx_buffer != first) {
if (dma_unmap_len(tx_buffer, len))
dma_unmap_page(tx_ring->dev,
dma_unmap_addr(tx_buffer, dma),
dma_unmap_len(tx_buffer, len),
DMA_TO_DEVICE);
dma_unmap_len_set(tx_buffer, len, 0);
igb: Fix TX map failure path When the driver cannot map a TX buffer, instead of rolling back gracefully and retrying later, we currently get a panic: [ 159.885994] igb 0000:00:00.0: TX DMA map failed [ 159.886588] Unable to handle kernel paging request at virtual address ffff00000a08c7a8 ... [ 159.897031] PC is at igb_xmit_frame_ring+0x9c8/0xcb8 Fix the erroneous test that leads to this situation. Signed-off-by: Jean-Philippe Brucker <jean-philippe.brucker@arm.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-10-19 19:07:36 +00:00
if (i-- == 0)
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
i += tx_ring->count;
igb: Use dma_unmap_addr and dma_unmap_len defines This change is meant to improve performance on systems that do not require the DMA unmap calls. On those systems we do not need to make use of the unmap address for Tx or the unmap length so we can drop both thereby reducing the size of the Tx buffer info structure. In addition I have changed the logic to check for unmap length instead of unmap address when checking to see if a buffer needs to be unmapped from DMA use. The reasons for this change is that on some platforms it is possible to receive a valid DMA address of 0 from an IOMMU. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-18 01:56:27 +00:00
tx_buffer = &tx_ring->tx_buffer_info[i];
igb: remove use of skb_dma_map from driver This change removes skb_dma_map/unmap calls from the igb driver due to the fact that the call is incompatible with iommu enabled kernels. In order to prevent warnings about using the wrong unmap call I have added a mapped_as_page value to the buffer_info structure to track if the mapped region is a page or a buffer. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-02 16:47:18 +00:00
}
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
if (dma_unmap_len(tx_buffer, len))
dma_unmap_single(tx_ring->dev,
dma_unmap_addr(tx_buffer, dma),
dma_unmap_len(tx_buffer, len),
DMA_TO_DEVICE);
dma_unmap_len_set(tx_buffer, len, 0);
dev_kfree_skb_any(tx_buffer->skb);
tx_buffer->skb = NULL;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
tx_ring->next_to_use = i;
igb: avoid permanent lock of *_PTP_TX_IN_PROGRESS The igb driver uses a state bit lock to avoid handling more than one Tx timestamp request at once. This is required because hardware is limited to a single set of registers for Tx timestamps. The state bit lock is not properly cleaned up during igb_xmit_frame_ring() if the transmit fails such as due to DMA or TSO failure. In some hardware this results in blocking timestamps until the service task times out. In other hardware this results in a permanent lock of the timestamp bit because we never receive an interrupt indicating the timestamp occurred, since indeed the packet was never transmitted. Fix this by checking for DMA and TSO errors in igb_xmit_frame_ring() and properly cleaning up after ourselves when these occur. Reported-by: Reported-by: David Mirabito <davidm@metamako.com> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-03 17:28:55 +00:00
return -1;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: drop the "adv" off function names relating to descriptors Many of the function names in the hot path are carrying an extra "_adv" suffix on the end of them to represent the fact that they are using advanced descriptors instead of legacy descriptors. However since all igb uses are advanced descriptors adding the extra suffix doesn't really add any additional data. Since this is the case it is easiest to just drop the suffix and save us from having to store the extra characters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:59 +00:00
netdev_tx_t igb_xmit_frame_ring(struct sk_buff *skb,
struct igb_ring *tx_ring)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
struct igb_tx_buffer *first;
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
int tso;
igb: minor type cleanups This change cleans up some instances where unsigned int and u32 were being used interchangeably, and cleans up hdr_len which was being set to 0 twice. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:04:56 +00:00
u32 tx_flags = 0;
igb: Don't use NETDEV_FRAG_PAGE_MAX_SIZE in descriptor calculation This change updates igb so that it will correctly perform the descriptor count calculation. Previously it was taking NETDEV_FRAG_PAGE_MAX_SIZE into account with isn't really correct since a different value is used to determine the size of the pages used for TCP. That is actually determined by SKB_FRAG_PAGE_ORDER. Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-07 04:11:45 +00:00
unsigned short f;
igb: Update igb to use a path similar to ixgbe to determine when to stop Tx After reviewing the igb and ixgbe code I realized there are a few issues in how the code is structured. Specifically we are not checking the size of the buffers being used in transmits and we are not using the same value to determine when to stop or start a Tx queue. As such the code is prone to be buggy. This patch makes it so that we have one value DESC_NEEDED that we will use for starting and stopping the queue. In addition we will check the size of buffers being used when setting up a transmit so as to avoid a possible buffer overrun if we were to receive a frame with a block of data larger than 32K in skb->data. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-09 04:27:48 +00:00
u16 count = TXD_USE_COUNT(skb_headlen(skb));
igb: Cleanup protocol handling in transmit path This change is meant to cleanup the protocol handling in the transmit path so that it correctly offloads software VLAN tagged frames. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:53 +00:00
__be16 protocol = vlan_get_protocol(skb);
igb: minor type cleanups This change cleans up some instances where unsigned int and u32 were being used interchangeably, and cleans up hdr_len which was being set to 0 twice. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:04:56 +00:00
u8 hdr_len = 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Update igb to use a path similar to ixgbe to determine when to stop Tx After reviewing the igb and ixgbe code I realized there are a few issues in how the code is structured. Specifically we are not checking the size of the buffers being used in transmits and we are not using the same value to determine when to stop or start a Tx queue. As such the code is prone to be buggy. This patch makes it so that we have one value DESC_NEEDED that we will use for starting and stopping the queue. In addition we will check the size of buffers being used when setting up a transmit so as to avoid a possible buffer overrun if we were to receive a frame with a block of data larger than 32K in skb->data. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-09 04:27:48 +00:00
/* need: 1 descriptor per page * PAGE_SIZE/IGB_MAX_DATA_PER_TXD,
* + 1 desc for skb_headlen/IGB_MAX_DATA_PER_TXD,
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
* + 2 desc gap to keep tail from touching head,
* + 1 desc for context descriptor,
igb: Update igb to use a path similar to ixgbe to determine when to stop Tx After reviewing the igb and ixgbe code I realized there are a few issues in how the code is structured. Specifically we are not checking the size of the buffers being used in transmits and we are not using the same value to determine when to stop or start a Tx queue. As such the code is prone to be buggy. This patch makes it so that we have one value DESC_NEEDED that we will use for starting and stopping the queue. In addition we will check the size of buffers being used when setting up a transmit so as to avoid a possible buffer overrun if we were to receive a frame with a block of data larger than 32K in skb->data. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-09 04:27:48 +00:00
* otherwise try next time
*/
igb: Don't use NETDEV_FRAG_PAGE_MAX_SIZE in descriptor calculation This change updates igb so that it will correctly perform the descriptor count calculation. Previously it was taking NETDEV_FRAG_PAGE_MAX_SIZE into account with isn't really correct since a different value is used to determine the size of the pages used for TCP. That is actually determined by SKB_FRAG_PAGE_ORDER. Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-07 04:11:45 +00:00
for (f = 0; f < skb_shinfo(skb)->nr_frags; f++)
count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size);
igb: Update igb to use a path similar to ixgbe to determine when to stop Tx After reviewing the igb and ixgbe code I realized there are a few issues in how the code is structured. Specifically we are not checking the size of the buffers being used in transmits and we are not using the same value to determine when to stop or start a Tx queue. As such the code is prone to be buggy. This patch makes it so that we have one value DESC_NEEDED that we will use for starting and stopping the queue. In addition we will check the size of buffers being used when setting up a transmit so as to avoid a possible buffer overrun if we were to receive a frame with a block of data larger than 32K in skb->data. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-09 04:27:48 +00:00
if (igb_maybe_stop_tx(tx_ring, count + 3)) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* this is a hard error */
return NETDEV_TX_BUSY;
}
igb: use timecompare to implement hardware time stamping Both TX and RX hardware time stamping are implemented. Due to hardware limitations it is not possible to verify reliably which packet was time stamped when multiple were pending for sending; this could be solved by only allowing one packet marked for hardware time stamping into the queue (not implemented yet). RX time stamping relies on the flag in the packet descriptor which marks packets that were time stamped. In "all packet" mode this flag is not set. TODO: also support that mode (even though it'll suffer from race conditions). Signed-off-by: John Ronciak <john.ronciak@intel.com> Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-12 05:03:43 +00:00
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
/* record the location of the first descriptor for this packet */
first = &tx_ring->tx_buffer_info[tx_ring->next_to_use];
first->skb = skb;
first->bytecount = skb->len;
first->gso_segs = 1;
igb: Pull adapter out of main path in igb_xmit_frame_ring We only need the adapter pointer in the case of ptp. As such we can pull the adapter out of the main path and place it inside the if statement to avoid the temptation of accessing the adapter pointer in the fast path. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-07 08:55:46 +00:00
if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) {
struct igb_adapter *adapter = netdev_priv(tx_ring->netdev);
igb: Prevent dropped Tx timestamps via work items and interrupts. In rare circumstances, it's possible a descriptor writeback will occur before a timestamped Tx packet will go out on the wire, leading to the driver believing the hardware failed to timestamp the packet. Schedule a work item for 82576 and use the available time sync interrupt registers on 82580 and above to account for this. Cc: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-18 07:26:33 +00:00
igb: Fix a test with HWTSTAMP_TX_ON 'HWTSTAMP_TX_ON' should be handled as a value, not as a bit mask. The modified code should behave the same, because HWTSTAMP_TX_ON is 1 and no other possible values of 'tx_type' would match the test. However, this is more future-proof, should other values be allowed one day. See 'struct hwtstamp_config' in 'include/uapi/linux/net_tstamp.h' This fixes a warning reported by smatch: igb_xmit_frame_ring() warn: bit shifter 'HWTSTAMP_TX_ON' used for logical '&' Fixes: 26bd4e2db06be ("igb: protect TX timestamping from API misuse") Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-02-06 19:47:59 +00:00
if (adapter->tstamp_config.tx_type == HWTSTAMP_TX_ON &&
igb: protect TX timestamping from API misuse HW timestamping can only be requested for a packet if the NIC is first setup via ioctl(SIOCSHWTSTAMP). If this step was skipped, then the igb driver still allowed TX packets to request HW timestamping. In this situation, the _IGB_PTP_TX_IN_PROGRESS flag was set and would never clear. This prevented any future HW timestamping requests to succeed. Fix this by checking that the NIC is configured for HW TX timestamping before accepting a HW TX timestamping request. Signed-off-by: Cliff Spradlin <cspradlin@google.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-06-19 20:30:43 +00:00
!test_and_set_bit_lock(__IGB_PTP_TX_IN_PROGRESS,
igb: fix race conditions on queuing skb for HW time stamp igb has a single set of TX time stamping resources per NIC. Use a simple bit lock to avoid race conditions and leaking skbs when multiple TX rings try to claim time stamping. Signed-off-by: Jakub Kicinski <kubakici@wp.pl> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-03-15 14:55:32 +00:00
&adapter->state)) {
igb: Pull adapter out of main path in igb_xmit_frame_ring We only need the adapter pointer in the case of ptp. As such we can pull the adapter out of the main path and place it inside the if statement to avoid the temptation of accessing the adapter pointer in the fast path. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-07 08:55:46 +00:00
skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
tx_flags |= IGB_TX_FLAGS_TSTAMP;
adapter->ptp_tx_skb = skb_get(skb);
adapter->ptp_tx_start = jiffies;
if (adapter->hw.mac.type == e1000_82576)
schedule_work(&adapter->ptp_tx_work);
igb: add statistic indicating number of skipped Tx timestamps The igb driver can only handle one Tx timestamp request at a time. This means it is possible for an application timestamp request to be ignored. There is no easy way for an administrator to determine if this occurred. Add a new statistic which tracks this, tx_hwtstamp_skipped. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-03 17:28:59 +00:00
} else {
adapter->tx_hwtstamp_skipped++;
igb: Pull adapter out of main path in igb_xmit_frame_ring We only need the adapter pointer in the case of ptp. As such we can pull the adapter out of the main path and place it inside the if statement to avoid the temptation of accessing the adapter pointer in the fast path. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-07 08:55:46 +00:00
}
igb: use timecompare to implement hardware time stamping Both TX and RX hardware time stamping are implemented. Due to hardware limitations it is not possible to verify reliably which packet was time stamped when multiple were pending for sending; this could be solved by only allowing one packet marked for hardware time stamping into the queue (not implemented yet). RX time stamping relies on the flag in the packet descriptor which marks packets that were time stamped. In "all packet" mode this flag is not set. TODO: also support that mode (even though it'll suffer from race conditions). Signed-off-by: John Ronciak <john.ronciak@intel.com> Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-12 05:03:43 +00:00
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
net: rename vlan_tx_* helpers since "tx" is misleading there The same macros are used for rx as well. So rename it. Signed-off-by: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-13 16:13:44 +00:00
if (skb_vlan_tag_present(skb)) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
tx_flags |= IGB_TX_FLAGS_VLAN;
net: rename vlan_tx_* helpers since "tx" is misleading there The same macros are used for rx as well. So rename it. Signed-off-by: Jiri Pirko <jiri@resnulli.us> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-13 16:13:44 +00:00
tx_flags |= (skb_vlan_tag_get(skb) << IGB_TX_FLAGS_VLAN_SHIFT);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
/* record initial flags and protocol */
first->tx_flags = tx_flags;
first->protocol = protocol;
igb: cleanup igb xmit frame path This patch cleans up the xmit frame path for igb to better handle xmit frame errors and avoid null pointer exceptions. It also cleans up some whitespace issues found in the xmit frame path. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:50:57 +00:00
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
tso = igb_tso(tx_ring, first, &hdr_len);
if (tso < 0)
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
goto out_drop;
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
else if (!tso)
igb_tx_csum(tx_ring, first);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: avoid permanent lock of *_PTP_TX_IN_PROGRESS The igb driver uses a state bit lock to avoid handling more than one Tx timestamp request at once. This is required because hardware is limited to a single set of registers for Tx timestamps. The state bit lock is not properly cleaned up during igb_xmit_frame_ring() if the transmit fails such as due to DMA or TSO failure. In some hardware this results in blocking timestamps until the service task times out. In other hardware this results in a permanent lock of the timestamp bit because we never receive an interrupt indicating the timestamp occurred, since indeed the packet was never transmitted. Fix this by checking for DMA and TSO errors in igb_xmit_frame_ring() and properly cleaning up after ourselves when these occur. Reported-by: Reported-by: David Mirabito <davidm@metamako.com> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-03 17:28:55 +00:00
if (igb_tx_map(tx_ring, first, hdr_len))
goto cleanup_tx_tstamp;
igb: add a flags value to the ring This patch adds a flags value to the ring that cleans up some of the last remaining items from the ring in order to help seperate it from the adapter struct. By implementing these flags it becomes possible for different rings to support different functions such as rx checksumming. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:52:46 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return NETDEV_TX_OK;
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
out_drop:
igb: Don't bother clearing Tx buffer_info in igb_clean_tx_ring In the case of the Tx rings we need to only clear the Tx buffer_info when we are resetting the rings. Ideally we do this when we configure the ring to bring it back up instead of when we are taking it down in order to avoid dirtying pages we don't need to. In addition we don't need to clear the Tx descriptor ring since we will fully repopulate it when we begin transmitting frames and next_to_watch can be cleared to prevent the ring from being cleaned beyond that point instead of needing to touch anything in the Tx descriptor ring. Finally with these changes we can avoid having to reset the skb member of the Tx buffer_info structure in the cleanup path since the skb will always be associated with the first buffer which has next_to_watch set. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:26:02 +00:00
dev_kfree_skb_any(first->skb);
first->skb = NULL;
igb: avoid permanent lock of *_PTP_TX_IN_PROGRESS The igb driver uses a state bit lock to avoid handling more than one Tx timestamp request at once. This is required because hardware is limited to a single set of registers for Tx timestamps. The state bit lock is not properly cleaned up during igb_xmit_frame_ring() if the transmit fails such as due to DMA or TSO failure. In some hardware this results in blocking timestamps until the service task times out. In other hardware this results in a permanent lock of the timestamp bit because we never receive an interrupt indicating the timestamp occurred, since indeed the packet was never transmitted. Fix this by checking for DMA and TSO errors in igb_xmit_frame_ring() and properly cleaning up after ourselves when these occur. Reported-by: Reported-by: David Mirabito <davidm@metamako.com> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-03 17:28:55 +00:00
cleanup_tx_tstamp:
if (unlikely(tx_flags & IGB_TX_FLAGS_TSTAMP)) {
struct igb_adapter *adapter = netdev_priv(tx_ring->netdev);
dev_kfree_skb_any(adapter->ptp_tx_skb);
adapter->ptp_tx_skb = NULL;
if (adapter->hw.mac.type == e1000_82576)
cancel_work_sync(&adapter->ptp_tx_work);
clear_bit_unlock(__IGB_PTP_TX_IN_PROGRESS, &adapter->state);
}
igb: push data into first igb_tx_buffer sooner to reduce stack usage Instead of storing most of the data for the TX hot path in the stack until we are ready to write the descriptor we can save ourselves some time and effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into the first igb_tx_buffer since that is where we will end up putting it anyway. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:15 +00:00
igb: Consolidate creation of Tx context descriptors into a single function This patch is meant to simplify the transmit path by reducing the overhead for creating a transmit context descriptor. The current implementation is split with igb_tso and igb_tx_csum doing two separate implementations on how to setup the tx_buffer_info structure and the tx_desc. By combining them it is possible to reduce code and simplify things since now only one function will create context descriptors. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:32 +00:00
return NETDEV_TX_OK;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
net: Remove ndo_xmit_flush netdev operation, use signalling instead. As reported by Jesper Dangaard Brouer, for high packet rates the overhead of having another indirect call in the TX path is non-trivial. There is the indirect call itself, and then there is all of the reloading of the state to refetch the tail pointer value and then write the device register. Move to a more passive scheme, which requires very light modifications to the device drivers. The signal is a new skb->xmit_more value, if it is non-zero it means that more SKBs are pending to be transmitted on the same queue as the current SKB. And therefore, the driver may elide the tail pointer update. Right now skb->xmit_more is always zero. Signed-off-by: David S. Miller <davem@davemloft.net>
2014-08-25 22:51:53 +00:00
static inline struct igb_ring *igb_tx_queue_mapping(struct igb_adapter *adapter,
struct sk_buff *skb)
igb: Remove multi_tx_table and simplify igb_xmit_frame Instead of using the multi_tx_table to map possible Tx queues to Tx rings we can just do simple subtraction for the unlikely event that the Tx queue provided exceeds the number of Tx rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:10 +00:00
{
net: Remove ndo_xmit_flush netdev operation, use signalling instead. As reported by Jesper Dangaard Brouer, for high packet rates the overhead of having another indirect call in the TX path is non-trivial. There is the indirect call itself, and then there is all of the reloading of the state to refetch the tail pointer value and then write the device register. Move to a more passive scheme, which requires very light modifications to the device drivers. The signal is a new skb->xmit_more value, if it is non-zero it means that more SKBs are pending to be transmitted on the same queue as the current SKB. And therefore, the driver may elide the tail pointer update. Right now skb->xmit_more is always zero. Signed-off-by: David S. Miller <davem@davemloft.net>
2014-08-25 22:51:53 +00:00
unsigned int r_idx = skb->queue_mapping;
igb: Remove multi_tx_table and simplify igb_xmit_frame Instead of using the multi_tx_table to map possible Tx queues to Tx rings we can just do simple subtraction for the unlikely event that the Tx queue provided exceeds the number of Tx rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:10 +00:00
if (r_idx >= adapter->num_tx_queues)
r_idx = r_idx % adapter->num_tx_queues;
return adapter->tx_ring[r_idx];
}
igb: drop the "adv" off function names relating to descriptors Many of the function names in the hot path are carrying an extra "_adv" suffix on the end of them to represent the fact that they are using advanced descriptors instead of legacy descriptors. However since all igb uses are advanced descriptors adding the extra suffix doesn't really add any additional data. Since this is the case it is easiest to just drop the suffix and save us from having to store the extra characters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:59 +00:00
static netdev_tx_t igb_xmit_frame(struct sk_buff *skb,
struct net_device *netdev)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct igb_adapter *adapter = netdev_priv(netdev);
igb: make tx ring map and free functionality non-static This change makes a minor change to the xmit_frame_ring_adv funcition in that it moves 2 checks from it into the xmit_frame_adv since the checks were not ring specific. In addition it exports the xmit_frame_ring_adv and the unmap_and_free_tx_resource calls so that they can be used by other code such as the ethtool loopback testing calls. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:54:43 +00:00
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* The minimum packet size with TCTL.PSP set is 17 so pad the skb
igb: Remove multi_tx_table and simplify igb_xmit_frame Instead of using the multi_tx_table to map possible Tx queues to Tx rings we can just do simple subtraction for the unlikely event that the Tx queue provided exceeds the number of Tx rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:10 +00:00
* in order to meet this minimum size requirement.
*/
ethernet/intel: Use eth_skb_pad and skb_put_padto helpers Update the Intel Ethernet drivers to use eth_skb_pad() and skb_put_padto instead of doing their own implementations of the function. Also this cleans up two other spots where skb_pad was called but the length and tail pointers were being manipulated directly instead of just having the padding length added via __skb_put. Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-03 16:17:39 +00:00
if (skb_put_padto(skb, 17))
return NETDEV_TX_OK;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Remove multi_tx_table and simplify igb_xmit_frame Instead of using the multi_tx_table to map possible Tx queues to Tx rings we can just do simple subtraction for the unlikely event that the Tx queue provided exceeds the number of Tx rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:10 +00:00
return igb_xmit_frame_ring(skb, igb_tx_queue_mapping(adapter, skb));
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_tx_timeout - Respond to a Tx Hang
* @netdev: network interface device structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static void igb_tx_timeout(struct net_device *netdev)
{
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
/* Do the reset outside of interrupt context */
adapter->tx_timeout_count++;
igb: make tx hang check multiqueue, check eop descriptor This change makes the tx hang check run over all tx queues instead of just queue 0. Also have hang display info on EOP descriptor instead of the descriptor at the start of the chain. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:48:51 +00:00
igb: Fix features that are currently 82580 only and should also be i350 This change allows support for per packet timesync and global device reset on the i350 adapter. These features were supported on both 82580 and i350 however it looks like several checks where not updated and as such the i350 support was not enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:55 +00:00
if (hw->mac.type >= e1000_82580)
igb: Add full support for 82580 devices This patch makes use of the 82580 PHY and MAC support added and adds a set of supported device IDs for said hardware. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-19 12:42:21 +00:00
hw->dev_spec._82575.global_device_reset = true;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
schedule_work(&adapter->reset_task);
igb: fix two minor items found during code review This patch addresses two minor items I found while cleaning up the igb driver for our sourceforge version. The first clears the context index if we don't flag that we need it. The second item is that eims_other should be used instead of bit defines when setting all of the EICS bits prior to reset. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:22:52 +00:00
wr32(E1000_EICS,
(adapter->eims_enable_mask & ~adapter->eims_other));
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
static void igb_reset_task(struct work_struct *work)
{
struct igb_adapter *adapter;
adapter = container_of(work, struct igb_adapter, reset_task);
igb: add registers etc. printout code just before resetting adapters This patch adds registers (,tx/rx rings' status and so on) printout code just before resetting adapters. This will be helpful for detecting the root cause of adapters reset. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 14:39:30 +00:00
igb_dump(adapter);
netdev_err(adapter->netdev, "Reset adapter\n");
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_reinit_locked(adapter);
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_get_stats64 - Get System Network Statistics
* @netdev: network interface device structure
* @stats: rtnl_link_stats64 pointer
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
net: make ndo_get_stats64 a void function The network device operation for reading statistics is only called in one place, and it ignores the return value. Having a structure return value is potentially confusing because some future driver could incorrectly assume that the return value was used. Fix all drivers with ndo_get_stats64 to have a void function. Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-01-07 03:12:52 +00:00
static void igb_get_stats64(struct net_device *netdev,
struct rtnl_link_stats64 *stats)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
struct igb_adapter *adapter = netdev_priv(netdev);
igb: reduce CPU0 latency when updating statistics This change is based off of the work and suggestion of Jan Jablonsky <jan.jablonsky@thalesgroup.com>. The Watchdog workqueue in igb driver is scheduled every 2s for each network interface. That includes updating a statistics protected by spinlock. Function igb_update_stats in this case will be protected against preemption. According to number of a statistics registers (cca 60), processing this function might cause additional cpu load on CPU0. In case of statistics spinlock may be replaced with mutex, which reduce latency on CPU0. CC: Bernhard Kaindl <bernhard.kaindl@thalesgroup.com> CC: Jan Jablonsky <jan.jablonsky@thalesgroup.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-08-02 17:13:10 +00:00
mutex_lock(&adapter->stats64_lock);
igb: Remove useless argument Given that all callers of igb_update_stats() pass the same two arguments: (adapter, &adapter->stats64), the second argument can be removed. Signed-off-by: Benjamin Poirier <bpoirier@suse.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-16 22:55:16 +00:00
igb_update_stats(adapter);
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
memcpy(stats, &adapter->stats64, sizeof(*stats));
igb: reduce CPU0 latency when updating statistics This change is based off of the work and suggestion of Jan Jablonsky <jan.jablonsky@thalesgroup.com>. The Watchdog workqueue in igb driver is scheduled every 2s for each network interface. That includes updating a statistics protected by spinlock. Function igb_update_stats in this case will be protected against preemption. According to number of a statistics registers (cca 60), processing this function might cause additional cpu load on CPU0. In case of statistics spinlock may be replaced with mutex, which reduce latency on CPU0. CC: Bernhard Kaindl <bernhard.kaindl@thalesgroup.com> CC: Jan Jablonsky <jan.jablonsky@thalesgroup.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-08-02 17:13:10 +00:00
mutex_unlock(&adapter->stats64_lock);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_change_mtu - Change the Maximum Transfer Unit
* @netdev: network interface device structure
* @new_mtu: new value for maximum frame size
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Returns 0 on success, negative on failure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static int igb_change_mtu(struct net_device *netdev, int new_mtu)
{
struct igb_adapter *adapter = netdev_priv(netdev);
igb: add pci_dev in few spots to clean up use of dev_err/info/warn This patch relpaces several references to adapter->pdev->dev with just pdev->dev. This allows for cleanup of several multiline dev_err/info calls. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:55 +00:00
struct pci_dev *pdev = adapter->pdev;
igb: Update max_frame_size to account for an optional VLAN tag if present This patch modifies the max_frame_size in order account for an optional VLAN tag. In order to support this we must also increase the MAX_STD_JUMBO_FRAME_SIZE to account for the 4 extra bytes. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:32 +00:00
int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Update MTU so that it is always at least a standard frame size This change makes it so that we limit the lower bound for max_frame_size to the size of a standard Ethernet frame. This allows for feature parity with other Intel based drivers such as ixgbe. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-07-16 00:20:34 +00:00
/* adjust max frame to be at least the size of a standard frame */
if (max_frame < (ETH_FRAME_LEN + ETH_FCS_LEN))
max_frame = ETH_FRAME_LEN + ETH_FCS_LEN;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
while (test_and_set_bit(__IGB_RESETTING, &adapter->state))
igb: Cleanups to fix msleep warnings This patch fixes WARNING:MSLEEP found by checkpatch check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:46:40 +00:00
usleep_range(1000, 2000);
igb: misc whitespace/formatting cleanups This patch is intended to hold several whitespace, formatting, and comment cleanups that have been found while cleaning up the igb driver. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-12 18:16:59 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* igb_down has a dependency on max_frame_size */
adapter->max_frame_size = max_frame;
igb: cleanup whitespace issues in igb_main.c This patch goes through and cleans up whitespace issues in igb_main.c to help improve readability. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:52:50 +00:00
igb: move rx_buffer_len into the ring structure This patch moves the rx_buffer_len value into the ring structure. This allows greater flexibility and the option of doing things such as supporting packet split only on some queues, or enabling virtualization. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:52:07 +00:00
if (netif_running(netdev))
igb_down(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add pci_dev in few spots to clean up use of dev_err/info/warn This patch relpaces several references to adapter->pdev->dev with just pdev->dev. This allows for cleanup of several multiline dev_err/info calls. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:55 +00:00
dev_info(&pdev->dev, "changing MTU from %d to %d\n",
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
netdev->mtu, new_mtu);
netdev->mtu = new_mtu;
if (netif_running(netdev))
igb_up(adapter);
else
igb_reset(adapter);
clear_bit(__IGB_RESETTING, &adapter->state);
return 0;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_update_stats - Update the board statistics counters
* @adapter: board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: Remove useless argument Given that all callers of igb_update_stats() pass the same two arguments: (adapter, &adapter->stats64), the second argument can be removed. Signed-off-by: Benjamin Poirier <bpoirier@suse.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-16 22:55:16 +00:00
void igb_update_stats(struct igb_adapter *adapter)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: Remove useless argument Given that all callers of igb_update_stats() pass the same two arguments: (adapter, &adapter->stats64), the second argument can be removed. Signed-off-by: Benjamin Poirier <bpoirier@suse.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-05-16 22:55:16 +00:00
struct rtnl_link_stats64 *net_stats = &adapter->stats64;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
struct e1000_hw *hw = &adapter->hw;
struct pci_dev *pdev = adapter->pdev;
igb: count Rx FIFO errors correctly Don't aggregate rx_no_buffer_count into rx_fifo_errors. RNBC counts packets that get queued temporarily in the adapter's FIFO. These packets are not dropped and are not errors. The correct counter is rx_missed_errors (MPC). Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-23 18:34:38 +00:00
u32 reg, mpc;
igb: only process global stats in igb_update_stats This patch moves the update of adapter->net_stats.rx/tx values out of the interrupt routine and into igb_update_stats by just adding together the tx/rx byte/packet counts for the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:48:12 +00:00
int i;
u64 bytes, packets;
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
unsigned int start;
u64 _bytes, _packets;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Prevent stats update while adapter is being reset, or if the pci
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
* connection is down.
*/
if (adapter->link_speed == 0)
return;
if (pci_channel_offline(pdev))
return;
igb: only process global stats in igb_update_stats This patch moves the update of adapter->net_stats.rx/tx values out of the interrupt routine and into igb_update_stats by just adding together the tx/rx byte/packet counts for the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:48:12 +00:00
bytes = 0;
packets = 0;
igb: Added rcu_lock to avoid race This patch adds rcu_lock to avoid possible race condition with igb_update_stats function accessing the rings in free_ q_vector. CC: Eric Dumazet <edumazet@google.com> Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-06-27 09:10:23 +00:00
rcu_read_lock();
igb: only process global stats in igb_update_stats This patch moves the update of adapter->net_stats.rx/tx values out of the interrupt routine and into igb_update_stats by just adding together the tx/rx byte/packet counts for the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:48:12 +00:00
for (i = 0; i < adapter->num_rx_queues; i++) {
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
struct igb_ring *ring = adapter->rx_ring[i];
igb: fix stats for i210 rx_fifo_errors RQDPC on i210/i211 is R/W not ReadClear. Clear after reading. Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-08 05:36:15 +00:00
u32 rqdpc = rd32(E1000_RQDPC(i));
if (hw->mac.type >= e1000_i210)
wr32(E1000_RQDPC(i), 0);
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
igb: Remove artificial restriction on RQDPC stat reading For some reason the reading of the RQDPC register was being artificially limited to 4K. Instead of limiting the value we should read the value and add the full amount. Otherwise this can lead to a misleading number of dropped packets when the actual value is in fact much higher. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-08 05:23:22 +00:00
if (rqdpc) {
ring->rx_stats.drops += rqdpc;
net_stats->rx_fifo_errors += rqdpc;
}
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
do {
net: Replace u64_stats_fetch_begin_bh to u64_stats_fetch_begin_irq Replace the bh safe variant with the hard irq safe variant. We need a hard irq safe variant to deal with netpoll transmitting packets from hard irq context, and we need it in most if not all of the places using the bh safe variant. Except on 32bit uni-processor the code is exactly the same so don't bother with a bh variant, just have a hard irq safe variant that everyone can use. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-14 04:26:42 +00:00
start = u64_stats_fetch_begin_irq(&ring->rx_syncp);
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
_bytes = ring->rx_stats.bytes;
_packets = ring->rx_stats.packets;
net: Replace u64_stats_fetch_begin_bh to u64_stats_fetch_begin_irq Replace the bh safe variant with the hard irq safe variant. We need a hard irq safe variant to deal with netpoll transmitting packets from hard irq context, and we need it in most if not all of the places using the bh safe variant. Except on 32bit uni-processor the code is exactly the same so don't bother with a bh variant, just have a hard irq safe variant that everyone can use. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-14 04:26:42 +00:00
} while (u64_stats_fetch_retry_irq(&ring->rx_syncp, start));
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
bytes += _bytes;
packets += _packets;
igb: only process global stats in igb_update_stats This patch moves the update of adapter->net_stats.rx/tx values out of the interrupt routine and into igb_update_stats by just adding together the tx/rx byte/packet counts for the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:48:12 +00:00
}
igb: Rework how netdev->stats is handled This patch does some refactoring work that I felt was needed after reviewing the changes recently submitted relating to the replacement of net_stats with netdev->stats. This patch essentially creates two different collections of stats. The first handles the adapter specific states and is stored in gstring_stats, and the second is for netdev specific stats and is stored in gstring_net_stats. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:38 +00:00
net_stats->rx_bytes = bytes;
net_stats->rx_packets = packets;
igb: only process global stats in igb_update_stats This patch moves the update of adapter->net_stats.rx/tx values out of the interrupt routine and into igb_update_stats by just adding together the tx/rx byte/packet counts for the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:48:12 +00:00
bytes = 0;
packets = 0;
for (i = 0; i < adapter->num_tx_queues; i++) {
igb: Allocate rings seperately instead of as a block This change makes it so that the rings are allocated seperately. As a result we can allocate them on seperate nodes at some point in the future if we so desire. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:39 +00:00
struct igb_ring *ring = adapter->tx_ring[i];
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
do {
net: Replace u64_stats_fetch_begin_bh to u64_stats_fetch_begin_irq Replace the bh safe variant with the hard irq safe variant. We need a hard irq safe variant to deal with netpoll transmitting packets from hard irq context, and we need it in most if not all of the places using the bh safe variant. Except on 32bit uni-processor the code is exactly the same so don't bother with a bh variant, just have a hard irq safe variant that everyone can use. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-14 04:26:42 +00:00
start = u64_stats_fetch_begin_irq(&ring->tx_syncp);
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
_bytes = ring->tx_stats.bytes;
_packets = ring->tx_stats.packets;
net: Replace u64_stats_fetch_begin_bh to u64_stats_fetch_begin_irq Replace the bh safe variant with the hard irq safe variant. We need a hard irq safe variant to deal with netpoll transmitting packets from hard irq context, and we need it in most if not all of the places using the bh safe variant. Except on 32bit uni-processor the code is exactly the same so don't bother with a bh variant, just have a hard irq safe variant that everyone can use. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-14 04:26:42 +00:00
} while (u64_stats_fetch_retry_irq(&ring->tx_syncp, start));
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
bytes += _bytes;
packets += _packets;
igb: only process global stats in igb_update_stats This patch moves the update of adapter->net_stats.rx/tx values out of the interrupt routine and into igb_update_stats by just adding together the tx/rx byte/packet counts for the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:48:12 +00:00
}
igb: Rework how netdev->stats is handled This patch does some refactoring work that I felt was needed after reviewing the changes recently submitted relating to the replacement of net_stats with netdev->stats. This patch essentially creates two different collections of stats. The first handles the adapter specific states and is stored in gstring_stats, and the second is for netdev specific stats and is stored in gstring_net_stats. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:38 +00:00
net_stats->tx_bytes = bytes;
net_stats->tx_packets = packets;
igb: Added rcu_lock to avoid race This patch adds rcu_lock to avoid possible race condition with igb_update_stats function accessing the rings in free_ q_vector. CC: Eric Dumazet <edumazet@google.com> Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-06-27 09:10:23 +00:00
rcu_read_unlock();
igb: only process global stats in igb_update_stats This patch moves the update of adapter->net_stats.rx/tx values out of the interrupt routine and into igb_update_stats by just adding together the tx/rx byte/packet counts for the rings. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:48:12 +00:00
/* read stats registers */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
adapter->stats.crcerrs += rd32(E1000_CRCERRS);
adapter->stats.gprc += rd32(E1000_GPRC);
adapter->stats.gorc += rd32(E1000_GORCL);
rd32(E1000_GORCH); /* clear GORCL */
adapter->stats.bprc += rd32(E1000_BPRC);
adapter->stats.mprc += rd32(E1000_MPRC);
adapter->stats.roc += rd32(E1000_ROC);
adapter->stats.prc64 += rd32(E1000_PRC64);
adapter->stats.prc127 += rd32(E1000_PRC127);
adapter->stats.prc255 += rd32(E1000_PRC255);
adapter->stats.prc511 += rd32(E1000_PRC511);
adapter->stats.prc1023 += rd32(E1000_PRC1023);
adapter->stats.prc1522 += rd32(E1000_PRC1522);
adapter->stats.symerrs += rd32(E1000_SYMERRS);
adapter->stats.sec += rd32(E1000_SEC);
igb: count Rx FIFO errors correctly Don't aggregate rx_no_buffer_count into rx_fifo_errors. RNBC counts packets that get queued temporarily in the adapter's FIFO. These packets are not dropped and are not errors. The correct counter is rx_missed_errors (MPC). Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-23 18:34:38 +00:00
mpc = rd32(E1000_MPC);
adapter->stats.mpc += mpc;
net_stats->rx_fifo_errors += mpc;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
adapter->stats.scc += rd32(E1000_SCC);
adapter->stats.ecol += rd32(E1000_ECOL);
adapter->stats.mcc += rd32(E1000_MCC);
adapter->stats.latecol += rd32(E1000_LATECOL);
adapter->stats.dc += rd32(E1000_DC);
adapter->stats.rlec += rd32(E1000_RLEC);
adapter->stats.xonrxc += rd32(E1000_XONRXC);
adapter->stats.xontxc += rd32(E1000_XONTXC);
adapter->stats.xoffrxc += rd32(E1000_XOFFRXC);
adapter->stats.xofftxc += rd32(E1000_XOFFTXC);
adapter->stats.fcruc += rd32(E1000_FCRUC);
adapter->stats.gptc += rd32(E1000_GPTC);
adapter->stats.gotc += rd32(E1000_GOTCL);
rd32(E1000_GOTCH); /* clear GOTCL */
igb: count Rx FIFO errors correctly Don't aggregate rx_no_buffer_count into rx_fifo_errors. RNBC counts packets that get queued temporarily in the adapter's FIFO. These packets are not dropped and are not errors. The correct counter is rx_missed_errors (MPC). Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-23 18:34:38 +00:00
adapter->stats.rnbc += rd32(E1000_RNBC);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
adapter->stats.ruc += rd32(E1000_RUC);
adapter->stats.rfc += rd32(E1000_RFC);
adapter->stats.rjc += rd32(E1000_RJC);
adapter->stats.tor += rd32(E1000_TORH);
adapter->stats.tot += rd32(E1000_TOTH);
adapter->stats.tpr += rd32(E1000_TPR);
adapter->stats.ptc64 += rd32(E1000_PTC64);
adapter->stats.ptc127 += rd32(E1000_PTC127);
adapter->stats.ptc255 += rd32(E1000_PTC255);
adapter->stats.ptc511 += rd32(E1000_PTC511);
adapter->stats.ptc1023 += rd32(E1000_PTC1023);
adapter->stats.ptc1522 += rd32(E1000_PTC1522);
adapter->stats.mptc += rd32(E1000_MPTC);
adapter->stats.bptc += rd32(E1000_BPTC);
igb: remove adaptive IFS from driver Adaptive IFS support has been included in the igb driver since its initial release, but it is not a feature on any igb NICs. This patch removes it from the driver. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:59 +00:00
adapter->stats.tpt += rd32(E1000_TPT);
adapter->stats.colc += rd32(E1000_COLC);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
adapter->stats.algnerrc += rd32(E1000_ALGNERRC);
igb: only read phy specific stats if in internal phy mode There are a couple statistics registers that are not meant to be read when in SGMII/serdes mode. This patch adds a check to verify mode before reading and updating these statistics. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:03:58 +00:00
/* read internal phy specific stats */
reg = rd32(E1000_CTRL_EXT);
if (!(reg & E1000_CTRL_EXT_LINK_MODE_MASK)) {
adapter->stats.rxerrc += rd32(E1000_RXERRC);
igb: Fix stats output on i210/i211 parts. Due to a hardware issue, on i210 and i211 parts, the TNCRS statistic provides an invalid value. This patch changes the update stats function to increment the stat only for non-i210/i211 parts. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-12 04:36:24 +00:00
/* this stat has invalid values on i210/i211 */
if ((hw->mac.type != e1000_i210) &&
(hw->mac.type != e1000_i211))
adapter->stats.tncrs += rd32(E1000_TNCRS);
igb: only read phy specific stats if in internal phy mode There are a couple statistics registers that are not meant to be read when in SGMII/serdes mode. This patch adds a check to verify mode before reading and updating these statistics. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:03:58 +00:00
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
adapter->stats.tsctc += rd32(E1000_TSCTC);
adapter->stats.tsctfc += rd32(E1000_TSCTFC);
adapter->stats.iac += rd32(E1000_IAC);
adapter->stats.icrxoc += rd32(E1000_ICRXOC);
adapter->stats.icrxptc += rd32(E1000_ICRXPTC);
adapter->stats.icrxatc += rd32(E1000_ICRXATC);
adapter->stats.ictxptc += rd32(E1000_ICTXPTC);
adapter->stats.ictxatc += rd32(E1000_ICTXATC);
adapter->stats.ictxqec += rd32(E1000_ICTXQEC);
adapter->stats.ictxqmtc += rd32(E1000_ICTXQMTC);
adapter->stats.icrxdmtc += rd32(E1000_ICRXDMTC);
/* Fill out the OS statistics structure */
igb: Rework how netdev->stats is handled This patch does some refactoring work that I felt was needed after reviewing the changes recently submitted relating to the replacement of net_stats with netdev->stats. This patch essentially creates two different collections of stats. The first handles the adapter specific states and is stored in gstring_stats, and the second is for netdev specific stats and is stored in gstring_net_stats. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:38 +00:00
net_stats->multicast = adapter->stats.mprc;
net_stats->collisions = adapter->stats.colc;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Rx Errors */
/* RLEC on some newer hardware can be incorrect so build
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* our own version based on RUC and ROC
*/
igb: Rework how netdev->stats is handled This patch does some refactoring work that I felt was needed after reviewing the changes recently submitted relating to the replacement of net_stats with netdev->stats. This patch essentially creates two different collections of stats. The first handles the adapter specific states and is stored in gstring_stats, and the second is for netdev specific stats and is stored in gstring_net_stats. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:38 +00:00
net_stats->rx_errors = adapter->stats.rxerrc +
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
adapter->stats.crcerrs + adapter->stats.algnerrc +
adapter->stats.ruc + adapter->stats.roc +
adapter->stats.cexterr;
igb: Rework how netdev->stats is handled This patch does some refactoring work that I felt was needed after reviewing the changes recently submitted relating to the replacement of net_stats with netdev->stats. This patch essentially creates two different collections of stats. The first handles the adapter specific states and is stored in gstring_stats, and the second is for netdev specific stats and is stored in gstring_net_stats. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:38 +00:00
net_stats->rx_length_errors = adapter->stats.ruc +
adapter->stats.roc;
net_stats->rx_crc_errors = adapter->stats.crcerrs;
net_stats->rx_frame_errors = adapter->stats.algnerrc;
net_stats->rx_missed_errors = adapter->stats.mpc;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Tx Errors */
igb: Rework how netdev->stats is handled This patch does some refactoring work that I felt was needed after reviewing the changes recently submitted relating to the replacement of net_stats with netdev->stats. This patch essentially creates two different collections of stats. The first handles the adapter specific states and is stored in gstring_stats, and the second is for netdev specific stats and is stored in gstring_net_stats. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-12 18:37:38 +00:00
net_stats->tx_errors = adapter->stats.ecol +
adapter->stats.latecol;
net_stats->tx_aborted_errors = adapter->stats.ecol;
net_stats->tx_window_errors = adapter->stats.latecol;
net_stats->tx_carrier_errors = adapter->stats.tncrs;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Tx Dropped needs to be maintained elsewhere */
/* Management Stats */
adapter->stats.mgptc += rd32(E1000_MGTPTC);
adapter->stats.mgprc += rd32(E1000_MGTPRC);
adapter->stats.mgpdc += rd32(E1000_MGTPDC);
igb: Add stats output for OS2BMC feature on i350 devices This patch adds statistics output for OS2BMC feature which is configured by eeprom on capable devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-26 07:42:37 +00:00
/* OS2BMC Stats */
reg = rd32(E1000_MANC);
if (reg & E1000_MANC_EN_BMC2OS) {
adapter->stats.o2bgptc += rd32(E1000_O2BGPTC);
adapter->stats.o2bspc += rd32(E1000_O2BSPC);
adapter->stats.b2ospc += rd32(E1000_B2OSPC);
adapter->stats.b2ogprc += rd32(E1000_B2OGPRC);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: refactor time sync interrupt handling The code that handles the time sync interrupt is repeated in three different places. This patch refactors the identical code blocks into a single helper function. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:10 +00:00
static void igb_tsync_interrupt(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
igb: enable internal PPS for the i210 The i210 device can produce an interrupt on the full second. This patch allows using this interrupt to generate an internal PPS event for adjusting the kernel system time. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:20 +00:00
struct ptp_clock_event event;
net: igb: avoid using timespec We want to deprecate the use of 'struct timespec' on 32-bit architectures, as it is will overflow in 2038. The igb driver uses it to read the current time, and can simply be changed to use ktime_get_real_ts64() instead. Because of hardware limitations, there is still an overflow in year 2106, which we cannot really avoid, but this documents the overflow. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: intel-wired-lan@lists.osuosl.org Reviewed-by: Richard Cochran <richardcochran@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-30 11:26:33 +00:00
struct timespec64 ts;
igb: enable auxiliary PHC functions for the i210 The i210 device offers a number of special PTP Hardware Clock features on the Software Defined Pins (SDPs). This patch adds support for two of the possible functions, namely time stamping external events, and periodic output signals. The assignment of PHC functions to the four SDP can be freely chosen by the user. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:26 +00:00
u32 ack = 0, tsauxc, sec, nsec, tsicr = rd32(E1000_TSICR);
igb: enable internal PPS for the i210 The i210 device can produce an interrupt on the full second. This patch allows using this interrupt to generate an internal PPS event for adjusting the kernel system time. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:20 +00:00
if (tsicr & TSINTR_SYS_WRAP) {
event.type = PTP_CLOCK_PPS;
if (adapter->ptp_caps.pps)
ptp_clock_event(adapter->ptp_clock, &event);
ack |= TSINTR_SYS_WRAP;
}
igb: refactor time sync interrupt handling The code that handles the time sync interrupt is repeated in three different places. This patch refactors the identical code blocks into a single helper function. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:10 +00:00
if (tsicr & E1000_TSICR_TXTS) {
/* retrieve hardware timestamp */
schedule_work(&adapter->ptp_tx_work);
igb: enable internal PPS for the i210 The i210 device can produce an interrupt on the full second. This patch allows using this interrupt to generate an internal PPS event for adjusting the kernel system time. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:20 +00:00
ack |= E1000_TSICR_TXTS;
igb: refactor time sync interrupt handling The code that handles the time sync interrupt is repeated in three different places. This patch refactors the identical code blocks into a single helper function. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:10 +00:00
}
igb: enable internal PPS for the i210 The i210 device can produce an interrupt on the full second. This patch allows using this interrupt to generate an internal PPS event for adjusting the kernel system time. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:20 +00:00
igb: enable auxiliary PHC functions for the i210 The i210 device offers a number of special PTP Hardware Clock features on the Software Defined Pins (SDPs). This patch adds support for two of the possible functions, namely time stamping external events, and periodic output signals. The assignment of PHC functions to the four SDP can be freely chosen by the user. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:26 +00:00
if (tsicr & TSINTR_TT0) {
spin_lock(&adapter->tmreg_lock);
net: igb: avoid using timespec We want to deprecate the use of 'struct timespec' on 32-bit architectures, as it is will overflow in 2038. The igb driver uses it to read the current time, and can simply be changed to use ktime_get_real_ts64() instead. Because of hardware limitations, there is still an overflow in year 2106, which we cannot really avoid, but this documents the overflow. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: intel-wired-lan@lists.osuosl.org Reviewed-by: Richard Cochran <richardcochran@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-30 11:26:33 +00:00
ts = timespec64_add(adapter->perout[0].start,
adapter->perout[0].period);
/* u32 conversion of tv_sec is safe until y2106 */
igb: enable auxiliary PHC functions for the i210 The i210 device offers a number of special PTP Hardware Clock features on the Software Defined Pins (SDPs). This patch adds support for two of the possible functions, namely time stamping external events, and periodic output signals. The assignment of PHC functions to the four SDP can be freely chosen by the user. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:26 +00:00
wr32(E1000_TRGTTIML0, ts.tv_nsec);
net: igb: avoid using timespec We want to deprecate the use of 'struct timespec' on 32-bit architectures, as it is will overflow in 2038. The igb driver uses it to read the current time, and can simply be changed to use ktime_get_real_ts64() instead. Because of hardware limitations, there is still an overflow in year 2106, which we cannot really avoid, but this documents the overflow. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: intel-wired-lan@lists.osuosl.org Reviewed-by: Richard Cochran <richardcochran@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-30 11:26:33 +00:00
wr32(E1000_TRGTTIMH0, (u32)ts.tv_sec);
igb: enable auxiliary PHC functions for the i210 The i210 device offers a number of special PTP Hardware Clock features on the Software Defined Pins (SDPs). This patch adds support for two of the possible functions, namely time stamping external events, and periodic output signals. The assignment of PHC functions to the four SDP can be freely chosen by the user. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:26 +00:00
tsauxc = rd32(E1000_TSAUXC);
tsauxc |= TSAUXC_EN_TT0;
wr32(E1000_TSAUXC, tsauxc);
adapter->perout[0].start = ts;
spin_unlock(&adapter->tmreg_lock);
ack |= TSINTR_TT0;
}
if (tsicr & TSINTR_TT1) {
spin_lock(&adapter->tmreg_lock);
net: igb: avoid using timespec We want to deprecate the use of 'struct timespec' on 32-bit architectures, as it is will overflow in 2038. The igb driver uses it to read the current time, and can simply be changed to use ktime_get_real_ts64() instead. Because of hardware limitations, there is still an overflow in year 2106, which we cannot really avoid, but this documents the overflow. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: intel-wired-lan@lists.osuosl.org Reviewed-by: Richard Cochran <richardcochran@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-30 11:26:33 +00:00
ts = timespec64_add(adapter->perout[1].start,
adapter->perout[1].period);
igb: enable auxiliary PHC functions for the i210 The i210 device offers a number of special PTP Hardware Clock features on the Software Defined Pins (SDPs). This patch adds support for two of the possible functions, namely time stamping external events, and periodic output signals. The assignment of PHC functions to the four SDP can be freely chosen by the user. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:26 +00:00
wr32(E1000_TRGTTIML1, ts.tv_nsec);
net: igb: avoid using timespec We want to deprecate the use of 'struct timespec' on 32-bit architectures, as it is will overflow in 2038. The igb driver uses it to read the current time, and can simply be changed to use ktime_get_real_ts64() instead. Because of hardware limitations, there is still an overflow in year 2106, which we cannot really avoid, but this documents the overflow. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: intel-wired-lan@lists.osuosl.org Reviewed-by: Richard Cochran <richardcochran@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-30 11:26:33 +00:00
wr32(E1000_TRGTTIMH1, (u32)ts.tv_sec);
igb: enable auxiliary PHC functions for the i210 The i210 device offers a number of special PTP Hardware Clock features on the Software Defined Pins (SDPs). This patch adds support for two of the possible functions, namely time stamping external events, and periodic output signals. The assignment of PHC functions to the four SDP can be freely chosen by the user. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:26 +00:00
tsauxc = rd32(E1000_TSAUXC);
tsauxc |= TSAUXC_EN_TT1;
wr32(E1000_TSAUXC, tsauxc);
adapter->perout[1].start = ts;
spin_unlock(&adapter->tmreg_lock);
ack |= TSINTR_TT1;
}
if (tsicr & TSINTR_AUTT0) {
nsec = rd32(E1000_AUXSTMPL0);
sec = rd32(E1000_AUXSTMPH0);
event.type = PTP_CLOCK_EXTTS;
event.index = 0;
event.timestamp = sec * 1000000000ULL + nsec;
ptp_clock_event(adapter->ptp_clock, &event);
ack |= TSINTR_AUTT0;
}
if (tsicr & TSINTR_AUTT1) {
nsec = rd32(E1000_AUXSTMPL1);
sec = rd32(E1000_AUXSTMPH1);
event.type = PTP_CLOCK_EXTTS;
event.index = 1;
event.timestamp = sec * 1000000000ULL + nsec;
ptp_clock_event(adapter->ptp_clock, &event);
ack |= TSINTR_AUTT1;
}
igb: enable internal PPS for the i210 The i210 device can produce an interrupt on the full second. This patch allows using this interrupt to generate an internal PPS event for adjusting the kernel system time. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:20 +00:00
/* acknowledge the interrupts */
wr32(E1000_TSICR, ack);
igb: refactor time sync interrupt handling The code that handles the time sync interrupt is repeated in three different places. This patch refactors the identical code blocks into a single helper function. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:10 +00:00
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static irqreturn_t igb_msix_other(int irq, void *data)
{
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
struct igb_adapter *adapter = data;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: add NAPI Rx queue support Update the NAPI implementation to use the new napi_struct infrstructure. This removes the need of multiple net_device structs to implement a multiqueue NAPI. Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:39 +00:00
u32 icr = rd32(E1000_ICR);
/* reading ICR causes bit 31 of EICR to be cleared */
igb: add counter for dma out of sync errors Add a counter for dma out of sync errors reported via interrupt. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:19:08 +00:00
igb: add support for device reset interrupt This patch adds support for the global device reset interrupt. Without this change the drivers will report tx hangs on all ports when a global device reset occurs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-07 17:41:00 +00:00
if (icr & E1000_ICR_DRSTA)
schedule_work(&adapter->reset_task);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (icr & E1000_ICR_DOUTSYNC) {
igb: add counter for dma out of sync errors Add a counter for dma out of sync errors reported via interrupt. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:19:08 +00:00
/* HW is reporting DMA is out of sync */
adapter->stats.doosync++;
igb: Add Anti-spoofing feature support Add support for the anti-spoofing feature in the HW. Packets from VF devices with spoofed MAC addresses or VLAN tags will be blocked and an event generated. When the watchdog task runs it will call a function to check if any spoof events occurred. If an event was detected then a warning message is dumped to the system log. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 02:08:26 +00:00
/* The DMA Out of Sync is also indication of a spoof event
* in IOV mode. Check the Wrong VM Behavior register to
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* see if it is really a spoof event.
*/
igb: Add Anti-spoofing feature support Add support for the anti-spoofing feature in the HW. Packets from VF devices with spoofed MAC addresses or VLAN tags will be blocked and an event generated. When the watchdog task runs it will call a function to check if any spoof events occurred. If an event was detected then a warning message is dumped to the system log. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 02:08:26 +00:00
igb_check_wvbr(adapter);
igb: add counter for dma out of sync errors Add a counter for dma out of sync errors reported via interrupt. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:19:08 +00:00
}
igb: cleanup igb_netpoll to be more friendly with napi & GRO This patch cleans up igb_netpoll so that it is more friendly with both the current napi and newly introduced GRO features. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:19:29 +00:00
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* Check for a mailbox event */
if (icr & E1000_ICR_VMMB)
igb_msg_task(adapter);
if (icr & E1000_ICR_LSC) {
hw->mac.get_link_status = 1;
/* guard against interrupt when we're going down */
if (!test_bit(__IGB_DOWN, &adapter->state))
mod_timer(&adapter->watchdog_timer, jiffies + 1);
}
igb: refactor time sync interrupt handling The code that handles the time sync interrupt is repeated in three different places. This patch refactors the identical code blocks into a single helper function. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:10 +00:00
if (icr & E1000_ICR_TS)
igb_tsync_interrupt(adapter);
igb: Prevent dropped Tx timestamps via work items and interrupts. In rare circumstances, it's possible a descriptor writeback will occur before a timestamped Tx packet will go out on the wire, leading to the driver believing the hardware failed to timestamp the packet. Schedule a work item for 82576 and use the available time sync interrupt registers on 82580 and above to account for this. Cc: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-18 07:26:33 +00:00
igb: add NAPI Rx queue support Update the NAPI implementation to use the new napi_struct infrstructure. This removes the need of multiple net_device structs to implement a multiqueue NAPI. Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:39 +00:00
wr32(E1000_EIMS, adapter->eims_other);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return IRQ_HANDLED;
}
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
static void igb_write_itr(struct igb_q_vector *q_vector)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: remove unecessary q_vector declarations and remove itr_shift This change removes the use of itr_shift since a mac type call can be used just as easily to identify the only HW that needs to have the itr shifted. In addition it removes two unecessary declarations of a q_vector pointer from the initialization path. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:00:41 +00:00
struct igb_adapter *adapter = q_vector->adapter;
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
u32 itr_val = q_vector->itr_val & 0x7FFC;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (!q_vector->set_itr)
return;
igb: misc whitespace/formatting cleanups This patch is intended to hold several whitespace, formatting, and comment cleanups that have been found while cleaning up the igb driver. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-12 18:16:59 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (!itr_val)
itr_val = 0x4;
igb: Introduce multiple TX queues with infrastructure This code adds multiple Tx queue infrastructure much like we previously did in ixgbe. The MSI-X vector mapping is the bulk of the change. IAM can now be safely enabled and we've verified that it does work correctly. We can also eliminate the tx ring lock. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:06:51 +00:00
igb: remove unecessary q_vector declarations and remove itr_shift This change removes the use of itr_shift since a mac type call can be used just as easily to identify the only HW that needs to have the itr shifted. In addition it removes two unecessary declarations of a q_vector pointer from the initialization path. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:00:41 +00:00
if (adapter->hw.mac.type == e1000_82575)
itr_val |= itr_val << 16;
igb: Introduce multiple TX queues with infrastructure This code adds multiple Tx queue infrastructure much like we previously did in ixgbe. The MSI-X vector mapping is the bulk of the change. IAM can now be safely enabled and we've verified that it does work correctly. We can also eliminate the tx ring lock. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:06:51 +00:00
else
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
itr_val |= E1000_EITR_CNT_IGNR;
igb: Introduce multiple TX queues with infrastructure This code adds multiple Tx queue infrastructure much like we previously did in ixgbe. The MSI-X vector mapping is the bulk of the change. IAM can now be safely enabled and we've verified that it does work correctly. We can also eliminate the tx ring lock. Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:06:51 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
writel(itr_val, q_vector->itr_register);
q_vector->set_itr = 0;
igb: Improve multiqueue AIM support Improve multiqueue performance Change itr_val to reflect ITR timer value instead of ints/sec Cleaned up AIM algorithms in general Based on work by Mitch Williams Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:14:44 +00:00
}
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
static irqreturn_t igb_msix_ring(int irq, void *data)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
struct igb_q_vector *q_vector = data;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
/* Write the ITR value calculated from the previous interrupt. */
igb_write_itr(q_vector);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
napi_schedule(&q_vector->napi);
igb: add NAPI Rx queue support Update the NAPI implementation to use the new napi_struct infrstructure. This removes the need of multiple net_device structs to implement a multiqueue NAPI. Signed-off-by: PJ Waskiewicz <peter.p.waskiewicz.jr.@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-27 18:00:39 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
return IRQ_HANDLED;
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
}
igb: add IGB_DCA instead of selecting INTEL_IOATDMA Add a bool IGB_DCA defined to y if IGB and DCA are enabled, but IGB isn't y while DCA=m. And thus remove the need to select INTEL_IOATDMA when IGB is enabled, so that non-x86 architectures can build the igb driver. Based on work/patch from Brice Goglin <brice@myri.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-10-17 18:08:31 +00:00
#ifdef CONFIG_IGB_DCA
igb: Split igb_update_dca into separate Tx and Rx functions This change makes it so that igb_update_dca is broken into two halves, one for Rx and one for Tx. The advantage to this is primarily readability. In addition I am enabling relaxed ordering for reads from hardware since this is supported on all of the igb parts. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:27 +00:00
static void igb_update_tx_dca(struct igb_adapter *adapter,
struct igb_ring *tx_ring,
int cpu)
{
struct e1000_hw *hw = &adapter->hw;
u32 txctrl = dca3_get_tag(tx_ring->dev, cpu);
if (hw->mac.type != e1000_82575)
txctrl <<= E1000_DCA_TXCTRL_CPUID_SHIFT;
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* We can enable relaxed ordering for reads, but not writes when
igb: Split igb_update_dca into separate Tx and Rx functions This change makes it so that igb_update_dca is broken into two halves, one for Rx and one for Tx. The advantage to this is primarily readability. In addition I am enabling relaxed ordering for reads from hardware since this is supported on all of the igb parts. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:27 +00:00
* DCA is enabled. This is due to a known issue in some chipsets
* which will cause the DCA tag to be cleared.
*/
txctrl |= E1000_DCA_TXCTRL_DESC_RRO_EN |
E1000_DCA_TXCTRL_DATA_RRO_EN |
E1000_DCA_TXCTRL_DESC_DCA_EN;
wr32(E1000_DCA_TXCTRL(tx_ring->reg_idx), txctrl);
}
static void igb_update_rx_dca(struct igb_adapter *adapter,
struct igb_ring *rx_ring,
int cpu)
{
struct e1000_hw *hw = &adapter->hw;
u32 rxctrl = dca3_get_tag(&adapter->pdev->dev, cpu);
if (hw->mac.type != e1000_82575)
rxctrl <<= E1000_DCA_RXCTRL_CPUID_SHIFT;
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* We can enable relaxed ordering for reads, but not writes when
igb: Split igb_update_dca into separate Tx and Rx functions This change makes it so that igb_update_dca is broken into two halves, one for Rx and one for Tx. The advantage to this is primarily readability. In addition I am enabling relaxed ordering for reads from hardware since this is supported on all of the igb parts. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:27 +00:00
* DCA is enabled. This is due to a known issue in some chipsets
* which will cause the DCA tag to be cleared.
*/
rxctrl |= E1000_DCA_RXCTRL_DESC_RRO_EN |
E1000_DCA_RXCTRL_DESC_DCA_EN;
wr32(E1000_DCA_RXCTRL(rx_ring->reg_idx), rxctrl);
}
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
static void igb_update_dca(struct igb_q_vector *q_vector)
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
{
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
struct igb_adapter *adapter = q_vector->adapter;
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
int cpu = get_cpu();
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (q_vector->cpu == cpu)
goto out_no_update;
igb: Split igb_update_dca into separate Tx and Rx functions This change makes it so that igb_update_dca is broken into two halves, one for Rx and one for Tx. The advantage to this is primarily readability. In addition I am enabling relaxed ordering for reads from hardware since this is supported on all of the igb parts. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:27 +00:00
if (q_vector->tx.ring)
igb_update_tx_dca(adapter, q_vector->tx.ring, cpu);
if (q_vector->rx.ring)
igb_update_rx_dca(adapter, q_vector->rx.ring, cpu);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
q_vector->cpu = cpu;
out_no_update:
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
put_cpu();
}
static void igb_setup_dca(struct igb_adapter *adapter)
{
igb: do not re-register DCA requester on device reset The current driver is re-registering the DCA requester after every reset. Instead of doing this we should only be updating the on board DCA registers and not unregistering/re-registering our requester. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-21 13:06:56 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
int i;
igb: Add support for quad port WOL and feature flags Change igb from using a series of boolean operators to using a single flags value that contains a number of different bit flags for all the different features of the adapter. This patch also adds WOL support for quad port adapters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:46 +00:00
if (!(adapter->flags & IGB_FLAG_DCA_ENABLED))
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
return;
igb: do not re-register DCA requester on device reset The current driver is re-registering the DCA requester after every reset. Instead of doing this we should only be updating the on board DCA registers and not unregistering/re-registering our requester. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-21 13:06:56 +00:00
/* Always use CB2 mode, difference is masked in the CB driver. */
wr32(E1000_DCA_CTRL, E1000_DCA_CTRL_DCA_MODE_CB2);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
for (i = 0; i < adapter->num_q_vectors; i++) {
igb: remove unecessary q_vector declarations and remove itr_shift This change removes the use of itr_shift since a mac type call can be used just as easily to identify the only HW that needs to have the itr shifted. In addition it removes two unecessary declarations of a q_vector pointer from the initialization path. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:00:41 +00:00
adapter->q_vector[i]->cpu = -1;
igb_update_dca(adapter->q_vector[i]);
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
}
}
static int __igb_notify_dca(struct device *dev, void *data)
{
struct net_device *netdev = dev_get_drvdata(dev);
struct igb_adapter *adapter = netdev_priv(netdev);
igb: add pci_dev in few spots to clean up use of dev_err/info/warn This patch relpaces several references to adapter->pdev->dev with just pdev->dev. This allows for cleanup of several multiline dev_err/info calls. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:55 +00:00
struct pci_dev *pdev = adapter->pdev;
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
struct e1000_hw *hw = &adapter->hw;
unsigned long event = *(unsigned long *)data;
switch (event) {
case DCA_PROVIDER_ADD:
/* if already enabled, don't do it again */
igb: Add support for quad port WOL and feature flags Change igb from using a series of boolean operators to using a single flags value that contains a number of different bit flags for all the different features of the adapter. This patch also adds WOL support for quad port adapters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:46 +00:00
if (adapter->flags & IGB_FLAG_DCA_ENABLED)
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
break;
if (dca_add_requester(dev) == 0) {
igb: Fix DCA errors and do not use context index for 82576 82576 was being incorrectly flagged as needing a context index. It does not as each ring has it's own table of 2 contexts. Driver was registering after registering the driver instead of the other way around. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-31 08:52:30 +00:00
adapter->flags |= IGB_FLAG_DCA_ENABLED;
igb: add pci_dev in few spots to clean up use of dev_err/info/warn This patch relpaces several references to adapter->pdev->dev with just pdev->dev. This allows for cleanup of several multiline dev_err/info calls. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:55 +00:00
dev_info(&pdev->dev, "DCA enabled\n");
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
igb_setup_dca(adapter);
break;
}
/* Fall Through since DCA is disabled. */
case DCA_PROVIDER_REMOVE:
igb: Add support for quad port WOL and feature flags Change igb from using a series of boolean operators to using a single flags value that contains a number of different bit flags for all the different features of the adapter. This patch also adds WOL support for quad port adapters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:46 +00:00
if (adapter->flags & IGB_FLAG_DCA_ENABLED) {
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
/* without this a class_device is left
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* hanging around in the sysfs model
*/
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
dca_remove_requester(dev);
igb: add pci_dev in few spots to clean up use of dev_err/info/warn This patch relpaces several references to adapter->pdev->dev with just pdev->dev. This allows for cleanup of several multiline dev_err/info calls. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:55 +00:00
dev_info(&pdev->dev, "DCA disabled\n");
igb: Add support for quad port WOL and feature flags Change igb from using a series of boolean operators to using a single flags value that contains a number of different bit flags for all the different features of the adapter. This patch also adds WOL support for quad port adapters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:46 +00:00
adapter->flags &= ~IGB_FLAG_DCA_ENABLED;
igb: remove unused defines This patch removes all of the unused defines from the .h files contained in igb. For some defines there was a use and so I plugged them into the correct locations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-16 07:59:44 +00:00
wr32(E1000_DCA_CTRL, E1000_DCA_CTRL_DCA_MODE_DISABLE);
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
}
break;
}
igb: Fix DCA errors and do not use context index for 82576 82576 was being incorrectly flagged as needing a context index. It does not as each ring has it's own table of 2 contexts. Driver was registering after registering the driver instead of the other way around. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-31 08:52:30 +00:00
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
return 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
static int igb_notify_dca(struct notifier_block *nb, unsigned long event,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
void *p)
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
{
int ret_val;
ret_val = driver_for_each_device(&igb_driver.driver, NULL, &event,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
__igb_notify_dca);
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
return ret_val ? NOTIFY_BAD : NOTIFY_DONE;
}
igb: add IGB_DCA instead of selecting INTEL_IOATDMA Add a bool IGB_DCA defined to y if IGB and DCA are enabled, but IGB isn't y while DCA=m. And thus remove the need to select INTEL_IOATDMA when IGB is enabled, so that non-x86 architectures can build the igb driver. Based on work/patch from Brice Goglin <brice@myri.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-10-17 18:08:31 +00:00
#endif /* CONFIG_IGB_DCA */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Check if subordinate VFs are assigned to virtual machines Kvm and the Xen pci-back driver will set a flag in the virtual function pci device dev_flags when the VF is assigned to a guest VM. Before destroying subordinate VFs check to see if the flag is set and if so skip the call to pci_disable_sriov() to avoid system crashes. Copy the maintainer for the Xen pci-back driver. Also CC'ing maintainers of all drivers found to call pci_disable_sriov(). V2 - Fix uninitialized variable warning Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Christian Benvenuti <benve@cisco.com> Cc: Sathya Perla <sathya.perla@emulex.com> Cc: Dimitris Michailidis <dm@chelsio.com> Cc: Jon Mason <jdmason@kudzu.us> Cc: James Smart <james.smart@emulex.com> Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-14 02:57:14 +00:00
#ifdef CONFIG_PCI_IOV
static int igb_vf_configure(struct igb_adapter *adapter, int vf)
{
unsigned char mac_addr[ETH_ALEN];
igb: Don't give VFs random MAC addresses If the user has not assigned a MAC address to a VM, then don't give it a random one. Instead, just give it zeros and let it figure out what to do with them. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> CC: Andy Gospodarek <andy@greyhouse.net> CC: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Stefan Assmann <sassmann@redhat.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-18 08:57:20 +00:00
eth_zero_addr(mac_addr);
igb: Check if subordinate VFs are assigned to virtual machines Kvm and the Xen pci-back driver will set a flag in the virtual function pci device dev_flags when the VF is assigned to a guest VM. Before destroying subordinate VFs check to see if the flag is set and if so skip the call to pci_disable_sriov() to avoid system crashes. Copy the maintainer for the Xen pci-back driver. Also CC'ing maintainers of all drivers found to call pci_disable_sriov(). V2 - Fix uninitialized variable warning Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Christian Benvenuti <benve@cisco.com> Cc: Sathya Perla <sathya.perla@emulex.com> Cc: Dimitris Michailidis <dm@chelsio.com> Cc: Jon Mason <jdmason@kudzu.us> Cc: James Smart <james.smart@emulex.com> Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-14 02:57:14 +00:00
igb_set_vf_mac(adapter, vf, mac_addr);
igb: add support for spoofchk config Add support for spoofchk configuration per VF via iproute2 tool. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-03 20:27:48 +00:00
/* By default spoof check is enabled for all VFs */
adapter->vf_data[vf].spoofchk_enabled = true;
igb: add VF trust infrastructure * Add a per-VF value to know if a VF is trusted, by default don't trust VFs. * Implement netdev op to trust VFs (igb_ndo_set_vf_trust) and add trust status to ndo_get_vf_config output. * Allow a trusted VF to change MAC and MAC filters even if MAC has been administratively set. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-01-17 10:53:39 +00:00
/* By default VFs are not trusted */
adapter->vf_data[vf].trusted = false;
igb: Change how we check for pre-existing and assigned VFs Adapt the pre-existing and assigned VFs code to the ixgbe way introduced in commit 9297127b9cdd8d30c829ef5fd28b7cc0323a7bcd. Instead of searching the enabled VFs we use pci_num_vf to determine enabled VFs. By comparing to which PF an assigned VF is owned it's possible to decide whether to leave it enabled or not. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Acked-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Robert Garrett <robertx.e.garrett@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-18 04:06:11 +00:00
return 0;
igb: Check if subordinate VFs are assigned to virtual machines Kvm and the Xen pci-back driver will set a flag in the virtual function pci device dev_flags when the VF is assigned to a guest VM. Before destroying subordinate VFs check to see if the flag is set and if so skip the call to pci_disable_sriov() to avoid system crashes. Copy the maintainer for the Xen pci-back driver. Also CC'ing maintainers of all drivers found to call pci_disable_sriov(). V2 - Fix uninitialized variable warning Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Christian Benvenuti <benve@cisco.com> Cc: Sathya Perla <sathya.perla@emulex.com> Cc: Dimitris Michailidis <dm@chelsio.com> Cc: Jon Mason <jdmason@kudzu.us> Cc: James Smart <james.smart@emulex.com> Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-14 02:57:14 +00:00
}
#endif
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
static void igb_ping_all_vfs(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
u32 ping;
int i;
for (i = 0 ; i < adapter->vfs_allocated_count; i++) {
ping = E1000_PF_CONTROL_MSG;
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
if (adapter->vf_data[i].flags & IGB_VF_FLAG_CTS)
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
ping |= E1000_VT_MSGTYPE_CTS;
igb_write_mbx(hw, &ping, 1, i);
}
}
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
static int igb_set_vf_promisc(struct igb_adapter *adapter, u32 *msgbuf, u32 vf)
{
struct e1000_hw *hw = &adapter->hw;
u32 vmolr = rd32(E1000_VMOLR(vf));
struct vf_data_storage *vf_data = &adapter->vf_data[vf];
igb: clear VF_PROMISC bits instead of setting all other bits This change corrects an issue in which we were setting all flag bits except for promisc instead of clearing the promisc bits due to the incorrect use of an |= instead of an &=. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-09-22 17:56:20 +00:00
vf_data->flags &= ~(IGB_VF_FLAG_UNI_PROMISC |
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
IGB_VF_FLAG_MULTI_PROMISC);
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
vmolr &= ~(E1000_VMOLR_ROPE | E1000_VMOLR_ROMPE | E1000_VMOLR_MPME);
if (*msgbuf & E1000_VF_SET_PROMISC_MULTICAST) {
vmolr |= E1000_VMOLR_MPME;
igb: clear VF_PROMISC bits instead of setting all other bits This change corrects an issue in which we were setting all flag bits except for promisc instead of clearing the promisc bits due to the incorrect use of an |= instead of an &=. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-09-22 17:56:20 +00:00
vf_data->flags |= IGB_VF_FLAG_MULTI_PROMISC;
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
*msgbuf &= ~E1000_VF_SET_PROMISC_MULTICAST;
} else {
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* if we have hashes and we are clearing a multicast promisc
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
* flag we need to write the hashes to the MTA as this step
* was previously skipped
*/
if (vf_data->num_vf_mc_hashes > 30) {
vmolr |= E1000_VMOLR_MPME;
} else if (vf_data->num_vf_mc_hashes) {
int j;
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
vmolr |= E1000_VMOLR_ROMPE;
for (j = 0; j < vf_data->num_vf_mc_hashes; j++)
igb_mta_set(hw, vf_data->vf_mc_hashes[j]);
}
}
wr32(E1000_VMOLR(vf), vmolr);
/* there are flags left unprocessed, likely not supported */
if (*msgbuf & E1000_VT_MSGINFO_MASK)
return -EINVAL;
return 0;
}
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
static int igb_set_vf_multicasts(struct igb_adapter *adapter,
u32 *msgbuf, u32 vf)
{
int n = (msgbuf[0] & E1000_VT_MSGINFO_MASK) >> E1000_VT_MSGINFO_SHIFT;
u16 *hash_list = (u16 *)&msgbuf[1];
struct vf_data_storage *vf_data = &adapter->vf_data[vf];
int i;
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
/* salt away the number of multicast addresses assigned
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
* to this VF for later use to restore when the PF multi cast
* list changes
*/
vf_data->num_vf_mc_hashes = n;
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
/* only up to 30 hash values supported */
if (n > 30)
n = 30;
/* store the hashes for later use */
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
for (i = 0; i < n; i++)
trivial: remove unnecessary semicolons Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2009-08-18 18:18:35 +00:00
vf_data->vf_mc_hashes[i] = hash_list[i];
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* Flush and reset the mta with the new values */
igb: add support for set_rx_mode netdevice operation This patch adds support for the set_rx_mode netdevice operation so that igb can better support multiple unicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:48:56 +00:00
igb_set_rx_mode(adapter->netdev);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
return 0;
}
static void igb_restore_vf_multicasts(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
struct vf_data_storage *vf_data;
int i, j;
for (i = 0; i < adapter->vfs_allocated_count; i++) {
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
u32 vmolr = rd32(E1000_VMOLR(i));
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
vmolr &= ~(E1000_VMOLR_ROMPE | E1000_VMOLR_MPME);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
vf_data = &adapter->vf_data[i];
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
if ((vf_data->num_vf_mc_hashes > 30) ||
(vf_data->flags & IGB_VF_FLAG_MULTI_PROMISC)) {
vmolr |= E1000_VMOLR_MPME;
} else if (vf_data->num_vf_mc_hashes) {
vmolr |= E1000_VMOLR_ROMPE;
for (j = 0; j < vf_data->num_vf_mc_hashes; j++)
igb_mta_set(hw, vf_data->vf_mc_hashes[j]);
}
wr32(E1000_VMOLR(i), vmolr);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
}
static void igb_clear_vf_vfta(struct igb_adapter *adapter, u32 vf)
{
struct e1000_hw *hw = &adapter->hw;
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
u32 pool_mask, vlvf_mask, i;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
/* create mask for VF and other pools */
pool_mask = E1000_VLVF_POOLSEL_MASK;
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
vlvf_mask = BIT(E1000_VLVF_POOLSEL_SHIFT + vf);
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
/* drop PF from pool bits */
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
pool_mask &= ~BIT(E1000_VLVF_POOLSEL_SHIFT +
adapter->vfs_allocated_count);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* Find the vlan filter for this id */
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
for (i = E1000_VLVF_ARRAY_SIZE; i--;) {
u32 vlvf = rd32(E1000_VLVF(i));
u32 vfta_mask, vid, vfta;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* remove the vf from the pool */
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
if (!(vlvf & vlvf_mask))
continue;
/* clear out bit from VLVF */
vlvf ^= vlvf_mask;
/* if other pools are present, just remove ourselves */
if (vlvf & pool_mask)
goto update_vlvfb;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
/* if PF is present, leave VFTA */
if (vlvf & E1000_VLVF_POOLSEL_MASK)
goto update_vlvf;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
vid = vlvf & E1000_VLVF_VLANID_MASK;
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
vfta_mask = BIT(vid % 32);
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
/* clear bit from VFTA */
vfta = adapter->shadow_vfta[vid / 32];
if (vfta & vfta_mask)
hw->mac.ops.write_vfta(hw, vid / 32, vfta ^ vfta_mask);
update_vlvf:
/* clear pool selection enable */
if (adapter->flags & IGB_FLAG_VLAN_PROMISC)
vlvf &= E1000_VLVF_POOLSEL_MASK;
else
vlvf = 0;
update_vlvfb:
/* clear pool bits */
wr32(E1000_VLVF(i), vlvf);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
}
igb: set vf rlpml wasn't taking vlan tag into account This patch updates things so that vlan tags are taken into account when setting the receive large packet maximum length. This allows the VF driver to correctly receive full sized frames when vlans are enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:49:33 +00:00
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
static int igb_find_vlvf_entry(struct e1000_hw *hw, u32 vlan)
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
{
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
u32 vlvf;
int idx;
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
/* short cut the special case */
if (vlan == 0)
return 0;
/* Search for the VLAN id in the VLVF entries */
for (idx = E1000_VLVF_ARRAY_SIZE; --idx;) {
vlvf = rd32(E1000_VLVF(idx));
if ((vlvf & VLAN_VID_MASK) == vlan)
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
break;
}
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
return idx;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
igb: make igb_update_pf_vlvf static Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:29 +00:00
static void igb_update_pf_vlvf(struct igb_adapter *adapter, u32 vid)
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
{
struct e1000_hw *hw = &adapter->hw;
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
u32 bits, pf_id;
int idx;
igb: rework handling of the vfta and vlvf registers in relation to mng_vlan This patch corrects some errors in how vlans are being handled when vfs start interacting with the management vlans. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:35 +00:00
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
idx = igb_find_vlvf_entry(hw, vid);
if (!idx)
return;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
/* See if any other pools are set for this VLAN filter
* entry other than the PF.
*/
pf_id = adapter->vfs_allocated_count + E1000_VLVF_POOLSEL_SHIFT;
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
bits = ~BIT(pf_id) & E1000_VLVF_POOLSEL_MASK;
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
bits &= rd32(E1000_VLVF(idx));
/* Disable the filter so this falls into the default pool. */
if (!bits) {
if (adapter->flags & IGB_FLAG_VLAN_PROMISC)
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
wr32(E1000_VLVF(idx), BIT(pf_id));
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
else
wr32(E1000_VLVF(idx), 0);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
}
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
static s32 igb_set_vf_vlan(struct igb_adapter *adapter, u32 vid,
bool add, u32 vf)
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
{
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
int pf_id = adapter->vfs_allocated_count;
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
int err;
igb: set vf rlpml wasn't taking vlan tag into account This patch updates things so that vlan tags are taken into account when setting the receive large packet maximum length. This allows the VF driver to correctly receive full sized frames when vlans are enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:49:33 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
/* If VLAN overlaps with one the PF is currently monitoring make
* sure that we are able to allocate a VLVF entry. This may be
* redundant but it guarantees PF will maintain visibility to
* the VLAN.
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
*/
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
if (add && test_bit(vid, adapter->active_vlans)) {
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
err = igb_vfta_set(hw, vid, pf_id, true, false);
if (err)
return err;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
err = igb_vfta_set(hw, vid, vf, add, false);
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
if (add && !err)
return err;
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
/* If we failed to add the VF VLAN or we are removing the VF VLAN
* we may need to drop the PF pool bit in order to allow us to free
* up the VLVF resources.
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
*/
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
if (test_bit(vid, adapter->active_vlans) ||
(adapter->flags & IGB_FLAG_VLAN_PROMISC))
igb_update_pf_vlvf(adapter, vid);
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
return err;
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
}
static void igb_set_vmvir(struct igb_adapter *adapter, u32 vid, u32 vf)
{
struct e1000_hw *hw = &adapter->hw;
if (vid)
wr32(E1000_VMVIR(vf), (vid | E1000_VMVIR_VLANA_DEFAULT));
else
wr32(E1000_VMVIR(vf), 0);
}
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
static int igb_enable_port_vlan(struct igb_adapter *adapter, int vf,
u16 vlan, u8 qos)
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
{
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
int err;
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
err = igb_set_vf_vlan(adapter, vlan, true, vf);
if (err)
return err;
igb_set_vmvir(adapter, vlan | (qos << VLAN_PRIO_SHIFT), vf);
igb_set_vmolr(adapter, vf, !vlan);
/* revoke access to previous VLAN */
if (vlan != adapter->vf_data[vf].pf_vlan)
igb_set_vf_vlan(adapter, adapter->vf_data[vf].pf_vlan,
false, vf);
adapter->vf_data[vf].pf_vlan = vlan;
adapter->vf_data[vf].pf_qos = qos;
igb: Fix VLAN tag stripping on Intel i350 Problem: When switching off VLAN offloading on an i350, the VLAN interface gets unusable. For testing, set up a VLAN on an i350 and some remote machine, e.g.: $ ip link add link eth0 name eth0.42 type vlan id 42 $ ip addr add 192.168.42.1/24 dev eth0.42 $ ip link set dev eth0.42 up Offloading is switched on by default: $ ethtool -k eth0 | grep vlan-offload rx-vlan-offload: on tx-vlan-offload: on $ ping -c 3 -I eth0.42 192.168.42.2 [...works as usual...] Now switch off VLAN offloading and try again: $ ethtool -K eth0 rxvlan off Actual changes: rx-vlan-offload: off tx-vlan-offload: off [requested on] $ ping -c 3 -I eth0.42 192.168.42.2 PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da ta. --- 192.168.42.2 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms I can only reproduce it on an i350, the above works fine on a 82580. While inspecting the igb source, I came across the code in igb_set_vmolr which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and for all, and in all of the igb code there's no other place where the STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb unconditionally, independently of the offloading setting. I compared that to the latest Intel igb-5.3.3.5 driver from http://sourceforge.net/projects/e1000/ which in fact sets and clears the STRVLAN flag independently from igb_set_vmolr in its own function igb_set_vf_vlan_strip, depending on the vlan settings. So I included the STRVLAN handling from the igb-5.3.3.5 driver into our current igb driver and tested the above scenario again. This time ping still works after switching off VLAN offloading. Tested on i350, with and without addtional VFs, as well as on 82580 successfully. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-28 12:53:23 +00:00
igb_set_vf_vlan_strip(adapter, vf, true);
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
dev_info(&adapter->pdev->dev,
"Setting VLAN %d, QOS 0x%x on VF %d\n", vlan, qos, vf);
if (test_bit(__IGB_DOWN, &adapter->state)) {
dev_warn(&adapter->pdev->dev,
"The VF VLAN has been set, but the PF device is not up.\n");
dev_warn(&adapter->pdev->dev,
"Bring the PF device up before attempting to use the VF device.\n");
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
}
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
return err;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
static int igb_disable_port_vlan(struct igb_adapter *adapter, int vf)
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
{
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
/* Restore tagless access via VLAN 0 */
igb_set_vf_vlan(adapter, 0, true, vf);
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
igb_set_vmvir(adapter, 0, vf);
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
igb_set_vmolr(adapter, vf, true);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
/* Remove any PF assigned VLAN */
if (adapter->vf_data[vf].pf_vlan)
igb_set_vf_vlan(adapter, adapter->vf_data[vf].pf_vlan,
false, vf);
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
adapter->vf_data[vf].pf_vlan = 0;
adapter->vf_data[vf].pf_qos = 0;
igb: Fix VLAN tag stripping on Intel i350 Problem: When switching off VLAN offloading on an i350, the VLAN interface gets unusable. For testing, set up a VLAN on an i350 and some remote machine, e.g.: $ ip link add link eth0 name eth0.42 type vlan id 42 $ ip addr add 192.168.42.1/24 dev eth0.42 $ ip link set dev eth0.42 up Offloading is switched on by default: $ ethtool -k eth0 | grep vlan-offload rx-vlan-offload: on tx-vlan-offload: on $ ping -c 3 -I eth0.42 192.168.42.2 [...works as usual...] Now switch off VLAN offloading and try again: $ ethtool -K eth0 rxvlan off Actual changes: rx-vlan-offload: off tx-vlan-offload: off [requested on] $ ping -c 3 -I eth0.42 192.168.42.2 PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da ta. --- 192.168.42.2 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms I can only reproduce it on an i350, the above works fine on a 82580. While inspecting the igb source, I came across the code in igb_set_vmolr which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and for all, and in all of the igb code there's no other place where the STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb unconditionally, independently of the offloading setting. I compared that to the latest Intel igb-5.3.3.5 driver from http://sourceforge.net/projects/e1000/ which in fact sets and clears the STRVLAN flag independently from igb_set_vmolr in its own function igb_set_vf_vlan_strip, depending on the vlan settings. So I included the STRVLAN handling from the igb-5.3.3.5 driver into our current igb driver and tested the above scenario again. This time ping still works after switching off VLAN offloading. Tested on i350, with and without addtional VFs, as well as on 82580 successfully. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-28 12:53:23 +00:00
igb_set_vf_vlan_strip(adapter, vf, false);
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
return 0;
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
}
net: Update API for VF vlan protocol 802.1ad support Introduce new rtnl UAPI that exposes a list of vlans per VF, giving the ability for user-space application to specify it for the VF, as an option to support 802.1ad. We adjusted IP Link tool to support this option. For future use cases, the new UAPI supports multiple vlans. For now we limit the list size to a single vlan in kernel. Add IFLA_VF_VLAN_LIST in addition to IFLA_VF_VLAN to keep backward compatibility with older versions of IP Link tool. Add a vlan protocol parameter to the ndo_set_vf_vlan callback. We kept 802.1Q as the drivers' default vlan protocol. Suitable ip link tool command examples: Set vf vlan protocol 802.1ad: ip link set eth0 vf 1 vlan 100 proto 802.1ad Set vf to VST (802.1Q) mode: ip link set eth0 vf 1 vlan 100 proto 802.1Q Or by omitting the new parameter ip link set eth0 vf 1 vlan 100 Signed-off-by: Moshe Shemesh <moshe@mellanox.com> Signed-off-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-22 09:11:15 +00:00
static int igb_ndo_set_vf_vlan(struct net_device *netdev, int vf,
u16 vlan, u8 qos, __be16 vlan_proto)
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
{
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
struct igb_adapter *adapter = netdev_priv(netdev);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
if ((vf >= adapter->vfs_allocated_count) || (vlan > 4095) || (qos > 7))
return -EINVAL;
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
net: Update API for VF vlan protocol 802.1ad support Introduce new rtnl UAPI that exposes a list of vlans per VF, giving the ability for user-space application to specify it for the VF, as an option to support 802.1ad. We adjusted IP Link tool to support this option. For future use cases, the new UAPI supports multiple vlans. For now we limit the list size to a single vlan in kernel. Add IFLA_VF_VLAN_LIST in addition to IFLA_VF_VLAN to keep backward compatibility with older versions of IP Link tool. Add a vlan protocol parameter to the ndo_set_vf_vlan callback. We kept 802.1Q as the drivers' default vlan protocol. Suitable ip link tool command examples: Set vf vlan protocol 802.1ad: ip link set eth0 vf 1 vlan 100 proto 802.1ad Set vf to VST (802.1Q) mode: ip link set eth0 vf 1 vlan 100 proto 802.1Q Or by omitting the new parameter ip link set eth0 vf 1 vlan 100 Signed-off-by: Moshe Shemesh <moshe@mellanox.com> Signed-off-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-22 09:11:15 +00:00
if (vlan_proto != htons(ETH_P_8021Q))
return -EPROTONOSUPPORT;
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
return (vlan || qos) ? igb_enable_port_vlan(adapter, vf, vlan, qos) :
igb_disable_port_vlan(adapter, vf);
}
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
static int igb_set_vf_vlan_msg(struct igb_adapter *adapter, u32 *msgbuf, u32 vf)
{
int add = (msgbuf[0] & E1000_VT_MSGINFO_MASK) >> E1000_VT_MSGINFO_SHIFT;
int vid = (msgbuf[1] & E1000_VLVF_VLANID_MASK);
igb: Fix VLAN tag stripping on Intel i350 Problem: When switching off VLAN offloading on an i350, the VLAN interface gets unusable. For testing, set up a VLAN on an i350 and some remote machine, e.g.: $ ip link add link eth0 name eth0.42 type vlan id 42 $ ip addr add 192.168.42.1/24 dev eth0.42 $ ip link set dev eth0.42 up Offloading is switched on by default: $ ethtool -k eth0 | grep vlan-offload rx-vlan-offload: on tx-vlan-offload: on $ ping -c 3 -I eth0.42 192.168.42.2 [...works as usual...] Now switch off VLAN offloading and try again: $ ethtool -K eth0 rxvlan off Actual changes: rx-vlan-offload: off tx-vlan-offload: off [requested on] $ ping -c 3 -I eth0.42 192.168.42.2 PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da ta. --- 192.168.42.2 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms I can only reproduce it on an i350, the above works fine on a 82580. While inspecting the igb source, I came across the code in igb_set_vmolr which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and for all, and in all of the igb code there's no other place where the STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb unconditionally, independently of the offloading setting. I compared that to the latest Intel igb-5.3.3.5 driver from http://sourceforge.net/projects/e1000/ which in fact sets and clears the STRVLAN flag independently from igb_set_vmolr in its own function igb_set_vf_vlan_strip, depending on the vlan settings. So I included the STRVLAN handling from the igb-5.3.3.5 driver into our current igb driver and tested the above scenario again. This time ping still works after switching off VLAN offloading. Tested on i350, with and without addtional VFs, as well as on 82580 successfully. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-28 12:53:23 +00:00
int ret;
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
if (adapter->vf_data[vf].pf_vlan)
return -1;
igb: Retain HW VLAN filtering while in promiscuous + VT mode When using the new bridge FDB interface to allow SR-IOV virtual function network devices to communicate with SW bridged network devices the physical function is placed into promiscuous mode and hardware VLAN filtering is disabled. This defeats the ability to use VLAN tagging to isolate user networks. When the device is in promiscuous mode and VT mode simultaneously ensure that VLAN hardware filtering remains enabled. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-26 06:19:41 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
/* VLAN 0 is a special case, don't allow it to be removed */
if (!vid && !add)
return 0;
igb: Fix VLAN tag stripping on Intel i350 Problem: When switching off VLAN offloading on an i350, the VLAN interface gets unusable. For testing, set up a VLAN on an i350 and some remote machine, e.g.: $ ip link add link eth0 name eth0.42 type vlan id 42 $ ip addr add 192.168.42.1/24 dev eth0.42 $ ip link set dev eth0.42 up Offloading is switched on by default: $ ethtool -k eth0 | grep vlan-offload rx-vlan-offload: on tx-vlan-offload: on $ ping -c 3 -I eth0.42 192.168.42.2 [...works as usual...] Now switch off VLAN offloading and try again: $ ethtool -K eth0 rxvlan off Actual changes: rx-vlan-offload: off tx-vlan-offload: off [requested on] $ ping -c 3 -I eth0.42 192.168.42.2 PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da ta. --- 192.168.42.2 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms I can only reproduce it on an i350, the above works fine on a 82580. While inspecting the igb source, I came across the code in igb_set_vmolr which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and for all, and in all of the igb code there's no other place where the STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb unconditionally, independently of the offloading setting. I compared that to the latest Intel igb-5.3.3.5 driver from http://sourceforge.net/projects/e1000/ which in fact sets and clears the STRVLAN flag independently from igb_set_vmolr in its own function igb_set_vf_vlan_strip, depending on the vlan settings. So I included the STRVLAN handling from the igb-5.3.3.5 driver into our current igb driver and tested the above scenario again. This time ping still works after switching off VLAN offloading. Tested on i350, with and without addtional VFs, as well as on 82580 successfully. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-28 12:53:23 +00:00
ret = igb_set_vf_vlan(adapter, vid, !!add, vf);
if (!ret)
igb_set_vf_vlan_strip(adapter, vf, !!vid);
return ret;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
static inline void igb_vf_reset(struct igb_adapter *adapter, u32 vf)
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
{
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
struct vf_data_storage *vf_data = &adapter->vf_data[vf];
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
/* clear flags - except flag that indicates PF has set the MAC */
vf_data->flags &= IGB_VF_FLAG_PF_SET_MAC;
vf_data->last_nack = jiffies;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* reset vlans for device */
igb_clear_vf_vfta(adapter, vf);
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
igb_set_vf_vlan(adapter, vf_data->pf_vlan, true, vf);
igb_set_vmvir(adapter, vf_data->pf_vlan |
(vf_data->pf_qos << VLAN_PRIO_SHIFT), vf);
igb_set_vmolr(adapter, vf, !vf_data->pf_vlan);
igb: Fix VLAN tag stripping on Intel i350 Problem: When switching off VLAN offloading on an i350, the VLAN interface gets unusable. For testing, set up a VLAN on an i350 and some remote machine, e.g.: $ ip link add link eth0 name eth0.42 type vlan id 42 $ ip addr add 192.168.42.1/24 dev eth0.42 $ ip link set dev eth0.42 up Offloading is switched on by default: $ ethtool -k eth0 | grep vlan-offload rx-vlan-offload: on tx-vlan-offload: on $ ping -c 3 -I eth0.42 192.168.42.2 [...works as usual...] Now switch off VLAN offloading and try again: $ ethtool -K eth0 rxvlan off Actual changes: rx-vlan-offload: off tx-vlan-offload: off [requested on] $ ping -c 3 -I eth0.42 192.168.42.2 PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da ta. --- 192.168.42.2 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms I can only reproduce it on an i350, the above works fine on a 82580. While inspecting the igb source, I came across the code in igb_set_vmolr which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and for all, and in all of the igb code there's no other place where the STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb unconditionally, independently of the offloading setting. I compared that to the latest Intel igb-5.3.3.5 driver from http://sourceforge.net/projects/e1000/ which in fact sets and clears the STRVLAN flag independently from igb_set_vmolr in its own function igb_set_vf_vlan_strip, depending on the vlan settings. So I included the STRVLAN handling from the igb-5.3.3.5 driver into our current igb driver and tested the above scenario again. This time ping still works after switching off VLAN offloading. Tested on i350, with and without addtional VFs, as well as on 82580 successfully. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-28 12:53:23 +00:00
igb_set_vf_vlan_strip(adapter, vf, !!(vf_data->pf_vlan));
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* reset multicast table array for vf */
adapter->vf_data[vf].num_vf_mc_hashes = 0;
/* Flush and reset the mta with the new values */
igb: add support for set_rx_mode netdevice operation This patch adds support for the set_rx_mode netdevice operation so that igb can better support multiple unicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:48:56 +00:00
igb_set_rx_mode(adapter->netdev);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
static void igb_vf_reset_event(struct igb_adapter *adapter, u32 vf)
{
unsigned char *vf_mac = adapter->vf_data[vf].vf_mac_addresses;
igb: Don't give VFs random MAC addresses If the user has not assigned a MAC address to a VM, then don't give it a random one. Instead, just give it zeros and let it figure out what to do with them. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> CC: Andy Gospodarek <andy@greyhouse.net> CC: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Stefan Assmann <sassmann@redhat.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-18 08:57:20 +00:00
/* clear mac address as we were hotplug removed/added */
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
if (!(adapter->vf_data[vf].flags & IGB_VF_FLAG_PF_SET_MAC))
igb: Don't give VFs random MAC addresses If the user has not assigned a MAC address to a VM, then don't give it a random one. Instead, just give it zeros and let it figure out what to do with them. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> CC: Andy Gospodarek <andy@greyhouse.net> CC: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Tested-by: Stefan Assmann <sassmann@redhat.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-18 08:57:20 +00:00
eth_zero_addr(vf_mac);
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
/* process remaining reset events */
igb_vf_reset(adapter, vf);
}
static void igb_vf_reset_msg(struct igb_adapter *adapter, u32 vf)
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
{
struct e1000_hw *hw = &adapter->hw;
unsigned char *vf_mac = adapter->vf_data[vf].vf_mac_addresses;
u32 reg, msgbuf[3];
u8 *addr = (u8 *)(&msgbuf[1]);
/* process all the same items cleared in a function level reset */
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
igb_vf_reset(adapter, vf);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* set vf mac address */
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
igb_set_vf_mac(adapter, vf, vf_mac);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* enable transmit and receive for vf */
reg = rd32(E1000_VFTE);
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
wr32(E1000_VFTE, reg | BIT(vf));
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
reg = rd32(E1000_VFRE);
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
wr32(E1000_VFRE, reg | BIT(vf));
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: Fix overwrite of the VF's flags The vf flags were being overwritten causing the flag that indicates that the PF has set the VF MAC address to get cleared. This would allow the VF to override the MAC address assigned by the Host VMM using the ip link set command. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 05:43:21 +00:00
adapter->vf_data[vf].flags |= IGB_VF_FLAG_CTS;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* reply to reset with ack and vf mac address */
igb: Indicate failure on vf reset for empty mac address Commit 5ac6f91d changed the igb driver to expose a zero (empty) mac address to the VF on reset rather than a random one. However, that behavioral change also requires igbvf driver changes which can be hard especially when we want to talk to proprietary guest OSs. Looking at the code previous to the commit in Linux that made igbvf work with empty mac addresses (8d56b6d), we can see that on reset failure the driver will try to generate a new mac address with both the old and the new code. Furthermore, ixgbe does send reset failure when it detects an empty mac address (35055928c). So I think it's safe to make igb behave the same. With this patch I can successfully run a Windows 8.1 guest with an empty mac address and an assigned igbvf device that has no mac address set by the host. If anyone is aware of a guest driver that chokes on NACK returns of VF RESET commands, please speak up. Signed-off-by: Alexander Graf <agraf@suse.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-10-09 05:33:55 +00:00
if (!is_zero_ether_addr(vf_mac)) {
msgbuf[0] = E1000_VF_RESET | E1000_VT_MSGTYPE_ACK;
memcpy(addr, vf_mac, ETH_ALEN);
} else {
msgbuf[0] = E1000_VF_RESET | E1000_VT_MSGTYPE_NACK;
}
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb_write_mbx(hw, msgbuf, 3, vf);
}
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
static void igb_flush_mac_table(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
int i;
for (i = 0; i < hw->mac.rar_entry_count; i++) {
adapter->mac_table[i].state &= ~IGB_MAC_STATE_IN_USE;
memset(adapter->mac_table[i].addr, 0, ETH_ALEN);
adapter->mac_table[i].queue = 0;
igb_rar_set_index(adapter, i);
}
}
static int igb_available_rars(struct igb_adapter *adapter, u8 queue)
{
struct e1000_hw *hw = &adapter->hw;
/* do not count rar entries reserved for VFs MAC addresses */
int rar_entries = hw->mac.rar_entry_count -
adapter->vfs_allocated_count;
int i, count = 0;
for (i = 0; i < rar_entries; i++) {
/* do not count default entries */
if (adapter->mac_table[i].state & IGB_MAC_STATE_DEFAULT)
continue;
/* do not count "in use" entries for different queues */
if ((adapter->mac_table[i].state & IGB_MAC_STATE_IN_USE) &&
(adapter->mac_table[i].queue != queue))
continue;
count++;
}
return count;
}
/* Set default MAC address for the PF in the first RAR entry */
static void igb_set_default_mac_filter(struct igb_adapter *adapter)
{
struct igb_mac_addr *mac_table = &adapter->mac_table[0];
ether_addr_copy(mac_table->addr, adapter->hw.mac.addr);
mac_table->queue = adapter->vfs_allocated_count;
mac_table->state = IGB_MAC_STATE_DEFAULT | IGB_MAC_STATE_IN_USE;
igb_rar_set_index(adapter, 0);
}
igb: Allow filters to be added for the local MAC address Users expect that when adding a steering filter for the local MAC address, that all the traffic directed to that address will go to some queue. Currently, it's not possible to configure entries in the "in use" state, which is the normal state of the local MAC address entry (it is the default), this patch allows to override the steering configuration of "in use" entries, if the filter to be added match the address and address type (source or destination) of an existing entry. There is a bit of a special handling for entries referring to the local MAC address, when they are removed, only the steering configuration is reset. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:55 +00:00
/* If the filter to be added and an already existing filter express
* the same address and address type, it should be possible to only
* override the other configurations, for example the queue to steer
* traffic.
*/
static bool igb_mac_entry_can_be_used(const struct igb_mac_addr *entry,
const u8 *addr, const u8 flags)
{
if (!(entry->state & IGB_MAC_STATE_IN_USE))
return true;
if ((entry->state & IGB_MAC_STATE_SRC_ADDR) !=
(flags & IGB_MAC_STATE_SRC_ADDR))
return false;
if (!ether_addr_equal(addr, entry->addr))
return false;
return true;
}
igb: Add support for MAC address filters specifying source addresses Makes it possible to direct packets to queues based on their source address. Documents the expected usage of the 'flags' parameter. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:53 +00:00
/* Add a MAC filter for 'addr' directing matching traffic to 'queue',
* 'flags' is used to indicate what kind of match is made, match is by
* default for the destination address, if matching by source address
* is desired the flag IGB_MAC_STATE_SRC_ADDR can be used.
*/
static int igb_add_mac_filter_flags(struct igb_adapter *adapter,
const u8 *addr, const u8 queue,
const u8 flags)
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
{
struct e1000_hw *hw = &adapter->hw;
int rar_entries = hw->mac.rar_entry_count -
adapter->vfs_allocated_count;
int i;
if (is_zero_ether_addr(addr))
return -EINVAL;
/* Search for the first empty entry in the MAC table.
* Do not touch entries at the end of the table reserved for the VF MAC
* addresses.
*/
for (i = 0; i < rar_entries; i++) {
igb: Allow filters to be added for the local MAC address Users expect that when adding a steering filter for the local MAC address, that all the traffic directed to that address will go to some queue. Currently, it's not possible to configure entries in the "in use" state, which is the normal state of the local MAC address entry (it is the default), this patch allows to override the steering configuration of "in use" entries, if the filter to be added match the address and address type (source or destination) of an existing entry. There is a bit of a special handling for entries referring to the local MAC address, when they are removed, only the steering configuration is reset. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:55 +00:00
if (!igb_mac_entry_can_be_used(&adapter->mac_table[i],
addr, flags))
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
continue;
ether_addr_copy(adapter->mac_table[i].addr, addr);
adapter->mac_table[i].queue = queue;
igb: Add support for MAC address filters specifying source addresses Makes it possible to direct packets to queues based on their source address. Documents the expected usage of the 'flags' parameter. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:53 +00:00
adapter->mac_table[i].state |= IGB_MAC_STATE_IN_USE | flags;
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
igb_rar_set_index(adapter, i);
return i;
}
return -ENOSPC;
}
igb: Add support for MAC address filters specifying source addresses Makes it possible to direct packets to queues based on their source address. Documents the expected usage of the 'flags' parameter. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:53 +00:00
static int igb_add_mac_filter(struct igb_adapter *adapter, const u8 *addr,
igb: make a few local functions static Clean up a few sparse warnings, these following functions can be made static: drivers/net/ethernet/intel/igb/igb_main.c: warning: symbol 'igb_add_mac_filter' was not declared. Should it be static? drivers/net/ethernet/intel/igb/igb_main.c: warning: symbol 'igb_del_mac_filter' was not declared. Should it be static? drivers/net/ethernet/intel/igb/igb_main.c: warning: symbol 'igb_set_vf_mac_filter' was not declared. Should it be static? Signed-off-by: Colin Ian King <colin.king@canonical.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-04-27 17:59:11 +00:00
const u8 queue)
igb: Add support for MAC address filters specifying source addresses Makes it possible to direct packets to queues based on their source address. Documents the expected usage of the 'flags' parameter. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:53 +00:00
{
return igb_add_mac_filter_flags(adapter, addr, queue, 0);
}
/* Remove a MAC filter for 'addr' directing matching traffic to
* 'queue', 'flags' is used to indicate what kind of match need to be
* removed, match is by default for the destination address, if
* matching by source address is to be removed the flag
* IGB_MAC_STATE_SRC_ADDR can be used.
*/
static int igb_del_mac_filter_flags(struct igb_adapter *adapter,
const u8 *addr, const u8 queue,
const u8 flags)
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
{
struct e1000_hw *hw = &adapter->hw;
int rar_entries = hw->mac.rar_entry_count -
adapter->vfs_allocated_count;
int i;
if (is_zero_ether_addr(addr))
return -EINVAL;
/* Search for matching entry in the MAC table based on given address
* and queue. Do not touch entries at the end of the table reserved
* for the VF MAC addresses.
*/
for (i = 0; i < rar_entries; i++) {
if (!(adapter->mac_table[i].state & IGB_MAC_STATE_IN_USE))
continue;
igb: Add support for MAC address filters specifying source addresses Makes it possible to direct packets to queues based on their source address. Documents the expected usage of the 'flags' parameter. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:53 +00:00
if ((adapter->mac_table[i].state & flags) != flags)
continue;
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
if (adapter->mac_table[i].queue != queue)
continue;
if (!ether_addr_equal(adapter->mac_table[i].addr, addr))
continue;
igb: Allow filters to be added for the local MAC address Users expect that when adding a steering filter for the local MAC address, that all the traffic directed to that address will go to some queue. Currently, it's not possible to configure entries in the "in use" state, which is the normal state of the local MAC address entry (it is the default), this patch allows to override the steering configuration of "in use" entries, if the filter to be added match the address and address type (source or destination) of an existing entry. There is a bit of a special handling for entries referring to the local MAC address, when they are removed, only the steering configuration is reset. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:55 +00:00
/* When a filter for the default address is "deleted",
* we return it to its initial configuration
*/
if (adapter->mac_table[i].state & IGB_MAC_STATE_DEFAULT) {
adapter->mac_table[i].state =
IGB_MAC_STATE_DEFAULT | IGB_MAC_STATE_IN_USE;
adapter->mac_table[i].queue =
adapter->vfs_allocated_count;
} else {
adapter->mac_table[i].state = 0;
adapter->mac_table[i].queue = 0;
memset(adapter->mac_table[i].addr, 0, ETH_ALEN);
}
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
igb_rar_set_index(adapter, i);
return 0;
}
return -ENOENT;
}
igb: Add support for MAC address filters specifying source addresses Makes it possible to direct packets to queues based on their source address. Documents the expected usage of the 'flags' parameter. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:53 +00:00
static int igb_del_mac_filter(struct igb_adapter *adapter, const u8 *addr,
const u8 queue)
{
return igb_del_mac_filter_flags(adapter, addr, queue, 0);
}
igb: Add support for enabling queue steering in filters On some igb models (82575 and i210) the MAC address filters can control to which queue the packet will be assigned. This extends the 'state' with one more state to signify that queue selection should be enabled for that filter. As 82575 parts are no longer easily obtained (and this was developed against i210), only support for the i210 model is enabled. These functions are exported and will be used in the next patch. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:54 +00:00
int igb_add_mac_steering_filter(struct igb_adapter *adapter,
const u8 *addr, u8 queue, u8 flags)
{
struct e1000_hw *hw = &adapter->hw;
/* In theory, this should be supported on 82575 as well, but
* that part wasn't easily accessible during development.
*/
if (hw->mac.type != e1000_i210)
return -EOPNOTSUPP;
return igb_add_mac_filter_flags(adapter, addr, queue,
IGB_MAC_STATE_QUEUE_STEERING | flags);
}
int igb_del_mac_steering_filter(struct igb_adapter *adapter,
const u8 *addr, u8 queue, u8 flags)
{
return igb_del_mac_filter_flags(adapter, addr, queue,
IGB_MAC_STATE_QUEUE_STEERING | flags);
}
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
static int igb_uc_sync(struct net_device *netdev, const unsigned char *addr)
{
struct igb_adapter *adapter = netdev_priv(netdev);
int ret;
ret = igb_add_mac_filter(adapter, addr, adapter->vfs_allocated_count);
return min_t(int, ret, 0);
}
static int igb_uc_unsync(struct net_device *netdev, const unsigned char *addr)
{
struct igb_adapter *adapter = netdev_priv(netdev);
igb_del_mac_filter(adapter, addr, adapter->vfs_allocated_count);
return 0;
}
igb: make a few local functions static Clean up a few sparse warnings, these following functions can be made static: drivers/net/ethernet/intel/igb/igb_main.c: warning: symbol 'igb_add_mac_filter' was not declared. Should it be static? drivers/net/ethernet/intel/igb/igb_main.c: warning: symbol 'igb_del_mac_filter' was not declared. Should it be static? drivers/net/ethernet/intel/igb/igb_main.c: warning: symbol 'igb_set_vf_mac_filter' was not declared. Should it be static? Signed-off-by: Colin Ian King <colin.king@canonical.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-04-27 17:59:11 +00:00
static int igb_set_vf_mac_filter(struct igb_adapter *adapter, const int vf,
const u32 info, const u8 *addr)
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
{
struct pci_dev *pdev = adapter->pdev;
struct vf_data_storage *vf_data = &adapter->vf_data[vf];
struct list_head *pos;
struct vf_mac_filter *entry = NULL;
int ret = 0;
switch (info) {
case E1000_VF_MAC_FILTER_CLR:
/* remove all unicast MAC filters related to the current VF */
list_for_each(pos, &adapter->vf_macs.l) {
entry = list_entry(pos, struct vf_mac_filter, l);
if (entry->vf == vf) {
entry->vf = -1;
entry->free = true;
igb_del_mac_filter(adapter, entry->vf_mac, vf);
}
}
break;
case E1000_VF_MAC_FILTER_ADD:
igb: add VF trust infrastructure * Add a per-VF value to know if a VF is trusted, by default don't trust VFs. * Implement netdev op to trust VFs (igb_ndo_set_vf_trust) and add trust status to ndo_get_vf_config output. * Allow a trusted VF to change MAC and MAC filters even if MAC has been administratively set. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-01-17 10:53:39 +00:00
if ((vf_data->flags & IGB_VF_FLAG_PF_SET_MAC) &&
!vf_data->trusted) {
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
dev_warn(&pdev->dev,
"VF %d requested MAC filter but is administratively denied\n",
vf);
return -EINVAL;
}
if (!is_valid_ether_addr(addr)) {
dev_warn(&pdev->dev,
"VF %d attempted to set invalid MAC filter\n",
vf);
return -EINVAL;
}
/* try to find empty slot in the list */
list_for_each(pos, &adapter->vf_macs.l) {
entry = list_entry(pos, struct vf_mac_filter, l);
if (entry->free)
break;
}
if (entry && entry->free) {
entry->free = false;
entry->vf = vf;
ether_addr_copy(entry->vf_mac, addr);
ret = igb_add_mac_filter(adapter, addr, vf);
ret = min_t(int, ret, 0);
} else {
ret = -ENOSPC;
}
if (ret == -ENOSPC)
dev_warn(&pdev->dev,
"VF %d has requested MAC filter but there is no space for it\n",
vf);
break;
default:
ret = -EINVAL;
break;
}
return ret;
}
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
static int igb_set_vf_mac_addr(struct igb_adapter *adapter, u32 *msg, int vf)
{
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
struct pci_dev *pdev = adapter->pdev;
struct vf_data_storage *vf_data = &adapter->vf_data[vf];
u32 info = msg[0] & E1000_VT_MSGINFO_MASK;
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* The VF MAC Address is stored in a packed array of bytes
igb: Add comment Add explanatory comment to avoid confusion when a pointer is set to the second word of an array instead of the customary cast of a pointer to the beginning of the array. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-07-01 13:39:23 +00:00
* starting at the second 32 bit word of the msg array
*/
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
unsigned char *addr = (unsigned char *)&msg[1];
int ret = 0;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
if (!info) {
igb: add VF trust infrastructure * Add a per-VF value to know if a VF is trusted, by default don't trust VFs. * Implement netdev op to trust VFs (igb_ndo_set_vf_trust) and add trust status to ndo_get_vf_config output. * Allow a trusted VF to change MAC and MAC filters even if MAC has been administratively set. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-01-17 10:53:39 +00:00
if ((vf_data->flags & IGB_VF_FLAG_PF_SET_MAC) &&
!vf_data->trusted) {
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
dev_warn(&pdev->dev,
"VF %d attempted to override administratively set MAC address\nReload the VF driver to resume operations\n",
vf);
return -EINVAL;
}
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
if (!is_valid_ether_addr(addr)) {
dev_warn(&pdev->dev,
"VF %d attempted to set invalid MAC\n",
vf);
return -EINVAL;
}
ret = igb_set_vf_mac(adapter, vf, addr);
} else {
ret = igb_set_vf_mac_filter(adapter, vf, info, addr);
}
return ret;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
static void igb_rcv_ack_from_vf(struct igb_adapter *adapter, u32 vf)
{
struct e1000_hw *hw = &adapter->hw;
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
struct vf_data_storage *vf_data = &adapter->vf_data[vf];
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
u32 msg = E1000_VT_MSGTYPE_NACK;
/* if device isn't clear to send it shouldn't be reading either */
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
if (!(vf_data->flags & IGB_VF_FLAG_CTS) &&
time_after(jiffies, vf_data->last_nack + (2 * HZ))) {
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb_write_mbx(hw, &msg, 1, vf);
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
vf_data->last_nack = jiffies;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
}
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
static void igb_rcv_msg_from_vf(struct igb_adapter *adapter, u32 vf)
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
{
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
struct pci_dev *pdev = adapter->pdev;
u32 msgbuf[E1000_VFMAILBOX_SIZE];
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
struct vf_data_storage *vf_data = &adapter->vf_data[vf];
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
s32 retval;
igb: do not drop PF mailbox lock after read of VF message When the PF receives a mailbox message from the VF, it grabs the mailbox lock, reads the VF message from the mailbox, ACKs the message and drops the lock. While the PF is performing the action for the VF message, nothing prevents another VF message from being posted to the mailbox. The current code handles this condition by just dropping any new VF messages without processing them. This results in a mailbox timeout in the VM for posted messages waiting for an ACK, and the VF is reset by the igbvf_watchdog_task in the VM. Given the right sequence of VF messages and mailbox timeouts, this condition can go on ad infinitum. Modify the PF mailbox read method to take an 'unlock' argument that optionally leaves the mailbox locked by the PF after reading the VF message. This ensures another VF message is not posted to the mailbox until after the PF has completed processing the VF message and written its reply. Signed-off-by: Greg Edwards <gedwards@ddn.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-06-28 15:22:26 +00:00
retval = igb_read_mbx(hw, msgbuf, E1000_VFMAILBOX_SIZE, vf, false);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: fix handling of mailbox collisions between PF/VF This patch changes the handling of collisions between the use of the PF/VF sides of the mailbox. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-12 06:57:34 +00:00
if (retval) {
/* if receive failed revoke VF CTS stats and restart init */
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
dev_err(&pdev->dev, "Error receiving message from VF\n");
igb: fix handling of mailbox collisions between PF/VF This patch changes the handling of collisions between the use of the PF/VF sides of the mailbox. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-12 06:57:34 +00:00
vf_data->flags &= ~IGB_VF_FLAG_CTS;
if (!time_after(jiffies, vf_data->last_nack + (2 * HZ)))
igb: do not drop PF mailbox lock after read of VF message When the PF receives a mailbox message from the VF, it grabs the mailbox lock, reads the VF message from the mailbox, ACKs the message and drops the lock. While the PF is performing the action for the VF message, nothing prevents another VF message from being posted to the mailbox. The current code handles this condition by just dropping any new VF messages without processing them. This results in a mailbox timeout in the VM for posted messages waiting for an ACK, and the VF is reset by the igbvf_watchdog_task in the VM. Given the right sequence of VF messages and mailbox timeouts, this condition can go on ad infinitum. Modify the PF mailbox read method to take an 'unlock' argument that optionally leaves the mailbox locked by the PF after reading the VF message. This ensures another VF message is not posted to the mailbox until after the PF has completed processing the VF message and written its reply. Signed-off-by: Greg Edwards <gedwards@ddn.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-06-28 15:22:26 +00:00
goto unlock;
igb: fix handling of mailbox collisions between PF/VF This patch changes the handling of collisions between the use of the PF/VF sides of the mailbox. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-12 06:57:34 +00:00
goto out;
}
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* this is a message we already processed, do nothing */
if (msgbuf[0] & (E1000_VT_MSGTYPE_ACK | E1000_VT_MSGTYPE_NACK))
igb: do not drop PF mailbox lock after read of VF message When the PF receives a mailbox message from the VF, it grabs the mailbox lock, reads the VF message from the mailbox, ACKs the message and drops the lock. While the PF is performing the action for the VF message, nothing prevents another VF message from being posted to the mailbox. The current code handles this condition by just dropping any new VF messages without processing them. This results in a mailbox timeout in the VM for posted messages waiting for an ACK, and the VF is reset by the igbvf_watchdog_task in the VM. Given the right sequence of VF messages and mailbox timeouts, this condition can go on ad infinitum. Modify the PF mailbox read method to take an 'unlock' argument that optionally leaves the mailbox locked by the PF after reading the VF message. This ensures another VF message is not posted to the mailbox until after the PF has completed processing the VF message and written its reply. Signed-off-by: Greg Edwards <gedwards@ddn.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-06-28 15:22:26 +00:00
goto unlock;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* until the vf completes a reset it should not be
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
* allowed to start any configuration.
*/
if (msgbuf[0] == E1000_VF_RESET) {
igb: do not drop PF mailbox lock after read of VF message When the PF receives a mailbox message from the VF, it grabs the mailbox lock, reads the VF message from the mailbox, ACKs the message and drops the lock. While the PF is performing the action for the VF message, nothing prevents another VF message from being posted to the mailbox. The current code handles this condition by just dropping any new VF messages without processing them. This results in a mailbox timeout in the VM for posted messages waiting for an ACK, and the VF is reset by the igbvf_watchdog_task in the VM. Given the right sequence of VF messages and mailbox timeouts, this condition can go on ad infinitum. Modify the PF mailbox read method to take an 'unlock' argument that optionally leaves the mailbox locked by the PF after reading the VF message. This ensures another VF message is not posted to the mailbox until after the PF has completed processing the VF message and written its reply. Signed-off-by: Greg Edwards <gedwards@ddn.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-06-28 15:22:26 +00:00
/* unlocks mailbox */
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb_vf_reset_msg(adapter, vf);
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
return;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
if (!(vf_data->flags & IGB_VF_FLAG_CTS)) {
igb: fix handling of mailbox collisions between PF/VF This patch changes the handling of collisions between the use of the PF/VF sides of the mailbox. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-12 06:57:34 +00:00
if (!time_after(jiffies, vf_data->last_nack + (2 * HZ)))
igb: do not drop PF mailbox lock after read of VF message When the PF receives a mailbox message from the VF, it grabs the mailbox lock, reads the VF message from the mailbox, ACKs the message and drops the lock. While the PF is performing the action for the VF message, nothing prevents another VF message from being posted to the mailbox. The current code handles this condition by just dropping any new VF messages without processing them. This results in a mailbox timeout in the VM for posted messages waiting for an ACK, and the VF is reset by the igbvf_watchdog_task in the VM. Given the right sequence of VF messages and mailbox timeouts, this condition can go on ad infinitum. Modify the PF mailbox read method to take an 'unlock' argument that optionally leaves the mailbox locked by the PF after reading the VF message. This ensures another VF message is not posted to the mailbox until after the PF has completed processing the VF message and written its reply. Signed-off-by: Greg Edwards <gedwards@ddn.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-06-28 15:22:26 +00:00
goto unlock;
igb: fix handling of mailbox collisions between PF/VF This patch changes the handling of collisions between the use of the PF/VF sides of the mailbox. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-12 06:57:34 +00:00
retval = -1;
goto out;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
switch ((msgbuf[0] & 0xFFFF)) {
case E1000_VF_SET_MAC_ADDR:
igb/igbvf: Add VF MAC filter request capabilities Add functionality for the VF to request up to 3 additional MAC filters. This is done using existing E1000_VF_SET_MAC_ADDR message, but with additional message info - E1000_VF_MAC_FILTER_CLR to clear all unicast MAC filters previously set for this VF and E1000_VF_MAC_FILTER_ADD to add MAC filter. Additional filters can be added only in case if administrator did not set VF MAC explicitly and allowed to change default MAC to the VF. Due to the limited number of RAR entries reserve at least 3 MAC filters for the PF. If SRIOV is supported by the NIC after this change RAR entries starting from 1 to (RAR MAX ENTRIES - NUM SRIOV VFS) will be used for PF and VF MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:26 +00:00
retval = igb_set_vf_mac_addr(adapter, msgbuf, vf);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
break;
igb: rework use of VMOLR in regards to PF and VFs This patch updates the use of the VMOLR to include enabling multicast promiscous for the VFs should they attempt to send over 30 multicast addresses or if they use the new message type to enable multicast promiscuous. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:16 +00:00
case E1000_VF_SET_PROMISC:
retval = igb_set_vf_promisc(adapter, msgbuf, vf);
break;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
case E1000_VF_SET_MULTICAST:
retval = igb_set_vf_multicasts(adapter, msgbuf, vf);
break;
case E1000_VF_SET_LPE:
retval = igb_set_vf_rlpml(adapter, msgbuf[1], vf);
break;
case E1000_VF_SET_VLAN:
igb: Warn on attempt to override administratively set MAC/VLAN Print a warning message to the system log when the VF attempts to override administratively set MAC/VLAN configuration. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 05:42:59 +00:00
retval = -1;
if (vf_data->pf_vlan)
dev_warn(&pdev->dev,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
"VF %d attempted to override administratively set VLAN tag\nReload the VF driver to resume operations\n",
vf);
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
else
igb: Clean-up configuration of VF port VLANs This patch is meant to clean-up the configuration of the VF port based VLAN configuration. The original logic was a bit muddled and had some undesirable side effects such as VLANs being either completely stripped from the port or VLANs being left when they shouldn't be. The idea behind this code is to avoid any events such as spurious spoof notifications when we are removing one VLAN tag and replacing it with another. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:11 +00:00
retval = igb_set_vf_vlan_msg(adapter, msgbuf, vf);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
break;
default:
igb: add pci_dev in few spots to clean up use of dev_err/info/warn This patch relpaces several references to adapter->pdev->dev with just pdev->dev. This allows for cleanup of several multiline dev_err/info calls. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:55 +00:00
dev_err(&pdev->dev, "Unhandled Msg %08x\n", msgbuf[0]);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
retval = -1;
break;
}
igb: fix handling of mailbox collisions between PF/VF This patch changes the handling of collisions between the use of the PF/VF sides of the mailbox. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-12 06:57:34 +00:00
msgbuf[0] |= E1000_VT_MSGTYPE_CTS;
out:
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
/* notify the VF of the results of what it sent us */
if (retval)
msgbuf[0] |= E1000_VT_MSGTYPE_NACK;
else
msgbuf[0] |= E1000_VT_MSGTYPE_ACK;
igb: do not drop PF mailbox lock after read of VF message When the PF receives a mailbox message from the VF, it grabs the mailbox lock, reads the VF message from the mailbox, ACKs the message and drops the lock. While the PF is performing the action for the VF message, nothing prevents another VF message from being posted to the mailbox. The current code handles this condition by just dropping any new VF messages without processing them. This results in a mailbox timeout in the VM for posted messages waiting for an ACK, and the VF is reset by the igbvf_watchdog_task in the VM. Given the right sequence of VF messages and mailbox timeouts, this condition can go on ad infinitum. Modify the PF mailbox read method to take an 'unlock' argument that optionally leaves the mailbox locked by the PF after reading the VF message. This ensures another VF message is not posted to the mailbox until after the PF has completed processing the VF message and written its reply. Signed-off-by: Greg Edwards <gedwards@ddn.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-06-28 15:22:26 +00:00
/* unlocks mailbox */
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb_write_mbx(hw, msgbuf, 1, vf);
igb: do not drop PF mailbox lock after read of VF message When the PF receives a mailbox message from the VF, it grabs the mailbox lock, reads the VF message from the mailbox, ACKs the message and drops the lock. While the PF is performing the action for the VF message, nothing prevents another VF message from being posted to the mailbox. The current code handles this condition by just dropping any new VF messages without processing them. This results in a mailbox timeout in the VM for posted messages waiting for an ACK, and the VF is reset by the igbvf_watchdog_task in the VM. Given the right sequence of VF messages and mailbox timeouts, this condition can go on ad infinitum. Modify the PF mailbox read method to take an 'unlock' argument that optionally leaves the mailbox locked by the PF after reading the VF message. This ensures another VF message is not posted to the mailbox until after the PF has completed processing the VF message and written its reply. Signed-off-by: Greg Edwards <gedwards@ddn.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-06-28 15:22:26 +00:00
return;
unlock:
igb_unlock_mbx(hw, vf);
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
}
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: replace the VF clear_to_send with a flags value In order to support future features it is easiest to replace the clear_to_send boolean with a flag value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:46:57 +00:00
static void igb_msg_task(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
u32 vf;
for (vf = 0; vf < adapter->vfs_allocated_count; vf++) {
/* process any reset requests */
if (!igb_check_for_rst(hw, vf))
igb_vf_reset_event(adapter, vf);
/* process any messages pending */
if (!igb_check_for_msg(hw, vf))
igb_rcv_msg_from_vf(adapter, vf);
/* process any acks */
if (!igb_check_for_ack(hw, vf))
igb_rcv_ack_from_vf(adapter, vf);
}
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
/**
* igb_set_uta - Set unicast filter table address
* @adapter: board private structure
igb: Add workaround for VLAN tag stripping on 82576 There was a workaround partially implemented for the 82576 that is needed in order for VLAN tag stripping to function correctly. The original code had side effects that would make it so the workaround was active on all MACs. I have updated the code so that the workaround is enabled, but limited to the 82576, or activated if we exceed the available unicast addresses. The workaround has a side effect of mirroring all of the traffic outgoing from the VFs back to the PF. As such it is not recommended to use the 82576 in promiscuous mode as it will take a performance hit, though this is now consistent with the performance as seen on the out-of-tree igb driver. I also limited the scope of the UTA bits all being set to only when the VMOLR register is enabled. This should limit the effects of the UTA register so that we don't pick up any excess traffic unless promiscuous mode has been enabled on the PF, whereas before the PF would have ended up in something equivalent to unicast promiscuous mode with VLAN filtering otherwise. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:43 +00:00
* @set: boolean indicating if we are setting or clearing bits
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
*
* The unicast table address is a register array of 32-bit registers.
* The table is meant to be used in a way similar to how the MTA is used
* however due to certain limitations in the hardware it is necessary to
Fix common misspellings Fixes generated by 'codespell' and manually reviewed. Signed-off-by: Lucas De Marchi <lucas.demarchi@profusion.mobi>
2011-03-31 01:57:33 +00:00
* set all the hash bits to 1 and use the VMOLR ROPE bit as a promiscuous
* enable bit to allow vlan tag stripping when promiscuous mode is enabled
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
**/
igb: Add workaround for VLAN tag stripping on 82576 There was a workaround partially implemented for the 82576 that is needed in order for VLAN tag stripping to function correctly. The original code had side effects that would make it so the workaround was active on all MACs. I have updated the code so that the workaround is enabled, but limited to the 82576, or activated if we exceed the available unicast addresses. The workaround has a side effect of mirroring all of the traffic outgoing from the VFs back to the PF. As such it is not recommended to use the 82576 in promiscuous mode as it will take a performance hit, though this is now consistent with the performance as seen on the out-of-tree igb driver. I also limited the scope of the UTA bits all being set to only when the VMOLR register is enabled. This should limit the effects of the UTA register so that we don't pick up any excess traffic unless promiscuous mode has been enabled on the PF, whereas before the PF would have ended up in something equivalent to unicast promiscuous mode with VLAN filtering otherwise. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:43 +00:00
static void igb_set_uta(struct igb_adapter *adapter, bool set)
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
{
struct e1000_hw *hw = &adapter->hw;
igb: Add workaround for VLAN tag stripping on 82576 There was a workaround partially implemented for the 82576 that is needed in order for VLAN tag stripping to function correctly. The original code had side effects that would make it so the workaround was active on all MACs. I have updated the code so that the workaround is enabled, but limited to the 82576, or activated if we exceed the available unicast addresses. The workaround has a side effect of mirroring all of the traffic outgoing from the VFs back to the PF. As such it is not recommended to use the 82576 in promiscuous mode as it will take a performance hit, though this is now consistent with the performance as seen on the out-of-tree igb driver. I also limited the scope of the UTA bits all being set to only when the VMOLR register is enabled. This should limit the effects of the UTA register so that we don't pick up any excess traffic unless promiscuous mode has been enabled on the PF, whereas before the PF would have ended up in something equivalent to unicast promiscuous mode with VLAN filtering otherwise. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:43 +00:00
u32 uta = set ? ~0 : 0;
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
int i;
/* we only need to do this if VMDq is enabled */
if (!adapter->vfs_allocated_count)
return;
igb: Add workaround for VLAN tag stripping on 82576 There was a workaround partially implemented for the 82576 that is needed in order for VLAN tag stripping to function correctly. The original code had side effects that would make it so the workaround was active on all MACs. I have updated the code so that the workaround is enabled, but limited to the 82576, or activated if we exceed the available unicast addresses. The workaround has a side effect of mirroring all of the traffic outgoing from the VFs back to the PF. As such it is not recommended to use the 82576 in promiscuous mode as it will take a performance hit, though this is now consistent with the performance as seen on the out-of-tree igb driver. I also limited the scope of the UTA bits all being set to only when the VMOLR register is enabled. This should limit the effects of the UTA register so that we don't pick up any excess traffic unless promiscuous mode has been enabled on the PF, whereas before the PF would have ended up in something equivalent to unicast promiscuous mode with VLAN filtering otherwise. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:43 +00:00
for (i = hw->mac.uta_reg_count; i--;)
array_wr32(E1000_UTA, i, uta);
igb: make use of the uta to allow for promiscous mode filter In order to support functions such as vlan tag stripping when SR-IOV is enabled any given packet must match at least one filter. However in the case of promiscous mode being enabled on the PF the traffic routed to it may not match any filters and is just sent to the PF by default. In order to make certain that this traffic is processed we can set all bits in the UTA registers to create a pseudo promiscous mode filter that accepts all packets. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:33:08 +00:00
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_intr_msi - Interrupt Handler
* @irq: interrupt number
* @data: pointer to a network interface device structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static irqreturn_t igb_intr_msi(int irq, void *data)
{
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
struct igb_adapter *adapter = data;
struct igb_q_vector *q_vector = adapter->q_vector[0];
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
struct e1000_hw *hw = &adapter->hw;
/* read ICR disables interrupts using IAM */
u32 icr = rd32(E1000_ICR);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb_write_itr(q_vector);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add support for device reset interrupt This patch adds support for the global device reset interrupt. Without this change the drivers will report tx hangs on all ports when a global device reset occurs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-07 17:41:00 +00:00
if (icr & E1000_ICR_DRSTA)
schedule_work(&adapter->reset_task);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (icr & E1000_ICR_DOUTSYNC) {
igb: add counter for dma out of sync errors Add a counter for dma out of sync errors reported via interrupt. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:19:08 +00:00
/* HW is reporting DMA is out of sync */
adapter->stats.doosync++;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (icr & (E1000_ICR_RXSEQ | E1000_ICR_LSC)) {
hw->mac.get_link_status = 1;
if (!test_bit(__IGB_DOWN, &adapter->state))
mod_timer(&adapter->watchdog_timer, jiffies + 1);
}
igb: refactor time sync interrupt handling The code that handles the time sync interrupt is repeated in three different places. This patch refactors the identical code blocks into a single helper function. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:10 +00:00
if (icr & E1000_ICR_TS)
igb_tsync_interrupt(adapter);
igb: Prevent dropped Tx timestamps via work items and interrupts. In rare circumstances, it's possible a descriptor writeback will occur before a timestamped Tx packet will go out on the wire, leading to the driver believing the hardware failed to timestamp the packet. Schedule a work item for 82576 and use the available time sync interrupt registers on 82580 and above to account for this. Cc: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-18 07:26:33 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
napi_schedule(&q_vector->napi);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return IRQ_HANDLED;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_intr - Legacy Interrupt Handler
* @irq: interrupt number
* @data: pointer to a network interface device structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
static irqreturn_t igb_intr(int irq, void *data)
{
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
struct igb_adapter *adapter = data;
struct igb_q_vector *q_vector = adapter->q_vector[0];
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
struct e1000_hw *hw = &adapter->hw;
/* Interrupt Auto-Mask...upon reading ICR, interrupts are masked. No
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* need for the IMC write
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
u32 icr = rd32(E1000_ICR);
/* IMS will not auto-mask if INT_ASSERTED is not set, and if it is
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* not set, then the adapter didn't send an interrupt
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (!(icr & E1000_ICR_INT_ASSERTED))
return IRQ_NONE;
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
igb_write_itr(q_vector);
igb: add support for device reset interrupt This patch adds support for the global device reset interrupt. Without this change the drivers will report tx hangs on all ports when a global device reset occurs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-01-07 17:41:00 +00:00
if (icr & E1000_ICR_DRSTA)
schedule_work(&adapter->reset_task);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (icr & E1000_ICR_DOUTSYNC) {
igb: add counter for dma out of sync errors Add a counter for dma out of sync errors reported via interrupt. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:19:08 +00:00
/* HW is reporting DMA is out of sync */
adapter->stats.doosync++;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (icr & (E1000_ICR_RXSEQ | E1000_ICR_LSC)) {
hw->mac.get_link_status = 1;
/* guard against interrupt when we're going down */
if (!test_bit(__IGB_DOWN, &adapter->state))
mod_timer(&adapter->watchdog_timer, jiffies + 1);
}
igb: refactor time sync interrupt handling The code that handles the time sync interrupt is repeated in three different places. This patch refactors the identical code blocks into a single helper function. Signed-off-by: Richard Cochran <richardcochran@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-21 20:51:10 +00:00
if (icr & E1000_ICR_TS)
igb_tsync_interrupt(adapter);
igb: Prevent dropped Tx timestamps via work items and interrupts. In rare circumstances, it's possible a descriptor writeback will occur before a timestamped Tx packet will go out on the wire, leading to the driver believing the hardware failed to timestamp the packet. Schedule a work item for 82576 and use the available time sync interrupt registers on 82580 and above to account for this. Cc: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-18 07:26:33 +00:00
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
napi_schedule(&q_vector->napi);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return IRQ_HANDLED;
}
igb: make local functions static Sparse caught two functions that were only being used in one file. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-18 22:13:26 +00:00
static void igb_ring_irq_enable(struct igb_q_vector *q_vector)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
struct igb_adapter *adapter = q_vector->adapter;
igb: update napi polling to consolidate function and return correct values igb is currently not returning the correct values for napi. In addition it is doing more work than necessary since it will not exit polling until work_done is equal to zero. This patch makes the following changes: 1. Consolidates msi-x and non-msi polling routines. 2. Corrects return values for polling routines. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:04 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
if ((q_vector->rx.ring && (adapter->rx_itr_setting & 3)) ||
(!q_vector->rx.ring && (adapter->tx_itr_setting & 3))) {
if ((adapter->num_q_vectors == 1) && !adapter->vf_data)
igb_set_itr(q_vector);
igb: update napi polling to consolidate function and return correct values igb is currently not returning the correct values for napi. In addition it is doing more work than necessary since it will not exit polling until work_done is equal to zero. This patch makes the following changes: 1. Consolidates msi-x and non-msi polling routines. 2. Corrects return values for polling routines. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:04 +00:00
else
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb_update_ring_itr(q_vector);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: update napi polling to consolidate function and return correct values igb is currently not returning the correct values for napi. In addition it is doing more work than necessary since it will not exit polling until work_done is equal to zero. This patch makes the following changes: 1. Consolidates msi-x and non-msi polling routines. 2. Corrects return values for polling routines. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:04 +00:00
if (!test_bit(__IGB_DOWN, &adapter->state)) {
igb: Change to use statically allocated array for MSIx entries This patch changes how the driver initializes MSIx and checks for MSIx configuration. This change makes it easier to reconfigure the device when queue changes happen at runtime using ethtool's set_channels feature. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:34 +00:00
if (adapter->flags & IGB_FLAG_HAS_MSIX)
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
wr32(E1000_EIMS, q_vector->eims_value);
igb: update napi polling to consolidate function and return correct values igb is currently not returning the correct values for napi. In addition it is doing more work than necessary since it will not exit polling until work_done is equal to zero. This patch makes the following changes: 1. Consolidates msi-x and non-msi polling routines. 2. Corrects return values for polling routines. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:04 +00:00
else
igb_irq_enable(adapter);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: update napi polling to consolidate function and return correct values igb is currently not returning the correct values for napi. In addition it is doing more work than necessary since it will not exit polling until work_done is equal to zero. This patch makes the following changes: 1. Consolidates msi-x and non-msi polling routines. 2. Corrects return values for polling routines. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:04 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_poll - NAPI Rx polling callback
* @napi: napi polling structure
* @budget: count of how many packets we should handle
igb: update napi polling to consolidate function and return correct values igb is currently not returning the correct values for napi. In addition it is doing more work than necessary since it will not exit polling until work_done is equal to zero. This patch makes the following changes: 1. Consolidates msi-x and non-msi polling routines. 2. Corrects return values for polling routines. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:04 +00:00
**/
static int igb_poll(struct napi_struct *napi, int budget)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
struct igb_q_vector *q_vector = container_of(napi,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
struct igb_q_vector,
napi);
igb: Refactor clean_rx_irq to reduce overhead and improve performance This change is meant to be a general cleanup and performance improvement for clean_rx_irq. The previous patch should have updated the allocation so that the rings can be treated as read-only within the clean_rx_irq function. In addition I am re-ordering the operations such that several goals are accomplished including reducing the overhead for packet accounting, reducing the number of items on the stack, and improving overall performance. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:54 +00:00
bool clean_complete = true;
drivers/net/intel: use napi_complete_done() As per Eric Dumazet's previous patches: (see commit (24d2e4a50737) - tg3: use napi_complete_done()) Quoting verbatim: Using napi_complete_done() instead of napi_complete() allows us to use /sys/class/net/ethX/gro_flush_timeout GRO layer can aggregate more packets if the flush is delayed a bit, without having to set too big coalescing parameters that impact latencies. </end quote> Tested configuration: low latency via ethtool -C ethx adaptive-rx off rx-usecs 10 adaptive-tx off tx-usecs 15 workload: streaming rx using netperf TCP_MAERTS igb: MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.48 10^6bits/s over 1.000 seconds ending at 1440193171.589 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1176930056 1475.36 797726 16384.00 71905 MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.49 10^6bits/s over 0.997 seconds ending at 1440193142.763 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1175182320 50476.00 23282 16384.00 71816 i40e: Hard to test because the traffic is incoming so fast (24Gb/s) that GRO always receives 87kB, even at the highest interrupt rate. Other drivers were only compile tested. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-09-24 23:35:47 +00:00
int work_done = 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: add IGB_DCA instead of selecting INTEL_IOATDMA Add a bool IGB_DCA defined to y if IGB and DCA are enabled, but IGB isn't y while DCA=m. And thus remove the need to select INTEL_IOATDMA when IGB is enabled, so that non-x86 architectures can build the igb driver. Based on work/patch from Brice Goglin <brice@myri.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-10-17 18:08:31 +00:00
#ifdef CONFIG_IGB_DCA
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
if (q_vector->adapter->flags & IGB_FLAG_DCA_ENABLED)
igb_update_dca(q_vector);
igb: add DCA support Add DCA support in the similar method that it was added to the ixgbe driver recently. DCA allows the network device to put data in the CPU cache and notify the chipset of that event. This reduces cache misses during receives. Signed-off-by: Jeb Cramer <cramerj@intel.com> Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:07:55 +00:00
#endif
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
if (q_vector->tx.ring)
igb: Add support for bulk Tx cleanup & cleanup boolean logic This patch enables bulk free in Tx cleanup for igb and cleans up the boolean logic in the polling routines for igb in the hopes of avoiding any mix-ups similar to what occurred with i40e and i40evf. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-03-07 17:30:21 +00:00
clean_complete = igb_clean_tx_irq(q_vector, budget);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
drivers/net/intel: use napi_complete_done() As per Eric Dumazet's previous patches: (see commit (24d2e4a50737) - tg3: use napi_complete_done()) Quoting verbatim: Using napi_complete_done() instead of napi_complete() allows us to use /sys/class/net/ethX/gro_flush_timeout GRO layer can aggregate more packets if the flush is delayed a bit, without having to set too big coalescing parameters that impact latencies. </end quote> Tested configuration: low latency via ethtool -C ethx adaptive-rx off rx-usecs 10 adaptive-tx off tx-usecs 15 workload: streaming rx using netperf TCP_MAERTS igb: MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.48 10^6bits/s over 1.000 seconds ending at 1440193171.589 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1176930056 1475.36 797726 16384.00 71905 MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.49 10^6bits/s over 0.997 seconds ending at 1440193142.763 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1175182320 50476.00 23282 16384.00 71816 i40e: Hard to test because the traffic is incoming so fast (24Gb/s) that GRO always receives 87kB, even at the highest interrupt rate. Other drivers were only compile tested. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-09-24 23:35:47 +00:00
if (q_vector->rx.ring) {
int cleaned = igb_clean_rx_irq(q_vector, budget);
work_done += cleaned;
igb: Add support for bulk Tx cleanup & cleanup boolean logic This patch enables bulk free in Tx cleanup for igb and cleans up the boolean logic in the polling routines for igb in the hopes of avoiding any mix-ups similar to what occurred with i40e and i40evf. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-03-07 17:30:21 +00:00
if (cleaned >= budget)
clean_complete = false;
drivers/net/intel: use napi_complete_done() As per Eric Dumazet's previous patches: (see commit (24d2e4a50737) - tg3: use napi_complete_done()) Quoting verbatim: Using napi_complete_done() instead of napi_complete() allows us to use /sys/class/net/ethX/gro_flush_timeout GRO layer can aggregate more packets if the flush is delayed a bit, without having to set too big coalescing parameters that impact latencies. </end quote> Tested configuration: low latency via ethtool -C ethx adaptive-rx off rx-usecs 10 adaptive-tx off tx-usecs 15 workload: streaming rx using netperf TCP_MAERTS igb: MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.48 10^6bits/s over 1.000 seconds ending at 1440193171.589 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1176930056 1475.36 797726 16384.00 71905 MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.49 10^6bits/s over 0.997 seconds ending at 1440193142.763 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1175182320 50476.00 23282 16384.00 71816 i40e: Hard to test because the traffic is incoming so fast (24Gb/s) that GRO always receives 87kB, even at the highest interrupt rate. Other drivers were only compile tested. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-09-24 23:35:47 +00:00
}
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb: Refactor clean_rx_irq to reduce overhead and improve performance This change is meant to be a general cleanup and performance improvement for clean_rx_irq. The previous patch should have updated the allocation so that the rings can be treated as read-only within the clean_rx_irq function. In addition I am re-ordering the operations such that several goals are accomplished including reducing the overhead for packet accounting, reducing the number of items on the stack, and improving overall performance. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:54 +00:00
/* If all work not completed, return budget and keep polling */
if (!clean_complete)
return budget;
igb: update napi polling to consolidate function and return correct values igb is currently not returning the correct values for napi. In addition it is doing more work than necessary since it will not exit polling until work_done is equal to zero. This patch makes the following changes: 1. Consolidates msi-x and non-msi polling routines. 2. Corrects return values for polling routines. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:39:04 +00:00
ethernet/intel: consolidate NAPI and NAPI exit While reviewing code, I noticed that Eric Dumazet recommends that drivers check the return code of napi_complete_done, and use that to decide to enable interrupts or not when exiting poll. One of the Intel drivers was already fixed (ixgbe). Upon looking at the Intel drivers as a whole, we are handling our polling and NAPI exit in a few different ways based on whether we have multiqueue and whether we have Tx cleanup included. Several drivers had the bug of exiting NAPI with return 0, which appears to mess up the accounting in the stack. Consolidate all the NAPI routines to do best known way of exiting and to just mostly look like each other. 1) check return code of napi_complete_done to control interrupt enable 2) return the actual amount of work done. 3) return budget immediately if need NAPI poll again Tested the changes on e1000e with a high interrupt rate set, and it shows about an 8% reduction in the CPU utilization when busy polling because we aren't re-enabling interrupts when we're about to be polled. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-11-08 22:55:32 +00:00
/* Exit the polling mode, but don't re-enable interrupts if stack might
* poll us due to busy-polling
*/
if (likely(napi_complete_done(napi, work_done)))
igb_ring_irq_enable(q_vector);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
ethernet/intel: consolidate NAPI and NAPI exit While reviewing code, I noticed that Eric Dumazet recommends that drivers check the return code of napi_complete_done, and use that to decide to enable interrupts or not when exiting poll. One of the Intel drivers was already fixed (ixgbe). Upon looking at the Intel drivers as a whole, we are handling our polling and NAPI exit in a few different ways based on whether we have multiqueue and whether we have Tx cleanup included. Several drivers had the bug of exiting NAPI with return 0, which appears to mess up the accounting in the stack. Consolidate all the NAPI routines to do best known way of exiting and to just mostly look like each other. 1) check return code of napi_complete_done to control interrupt enable 2) return the actual amount of work done. 3) return budget immediately if need NAPI poll again Tested the changes on e1000e with a high interrupt rate set, and it shows about an 8% reduction in the CPU utilization when busy polling because we aren't re-enabling interrupts when we're about to be polled. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-11-08 22:55:32 +00:00
return min(work_done, budget - 1);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb trivial annotations Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2008-03-16 22:23:24 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_clean_tx_irq - Reclaim resources after transmit completes
* @q_vector: pointer to q_vector containing needed info
igb: Add support for bulk Tx cleanup & cleanup boolean logic This patch enables bulk free in Tx cleanup for igb and cleans up the boolean logic in the polling routines for igb in the hopes of avoiding any mix-ups similar to what occurred with i40e and i40evf. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-03-07 17:30:21 +00:00
* @napi_budget: Used to determine if we are in netpoll
drivers/net/ethernet: Fix (nearly-)kernel-doc comments for various functions Fix incorrect start markers, wrapped summary lines, missing section breaks, incorrect separators, and some name mismatches. Delete a few that are content-free. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-10 10:56:00 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* returns true if ring is completely cleaned
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: Add support for bulk Tx cleanup & cleanup boolean logic This patch enables bulk free in Tx cleanup for igb and cleans up the boolean logic in the polling routines for igb in the hopes of avoiding any mix-ups similar to what occurred with i40e and i40evf. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-03-07 17:30:21 +00:00
static bool igb_clean_tx_irq(struct igb_q_vector *q_vector, int napi_budget)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
struct igb_adapter *adapter = q_vector->adapter;
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
struct igb_ring *tx_ring = q_vector->tx.ring;
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
struct igb_tx_buffer *tx_buffer;
igb: Change Tx cleanup loop to do/while instead of for This change makes it so that Tx cleanup is done in a do/while loop instead of a for loop. The main motivation behind this is the fact that we should never be invoked with a budget less than 1 so we can skip checking the budget before processing the first descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:28:01 +00:00
union e1000_adv_tx_desc *tx_desc;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
unsigned int total_bytes = 0, total_packets = 0;
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
unsigned int budget = q_vector->tx.work_limit;
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
unsigned int i = tx_ring->next_to_clean;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
if (test_bit(__IGB_DOWN, &adapter->state))
return true;
igb: defeature tx head writeback This patch removes tx head writeback as it was found to not provide a significant improvement in tx performance and on some systems has been seen to cause a performance degredation due to partial cache line writes. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 09:33:18 +00:00
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
tx_buffer = &tx_ring->tx_buffer_info[i];
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
tx_desc = IGB_TX_DESC(tx_ring, i);
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
i -= tx_ring->count;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Change Tx cleanup loop to do/while instead of for This change makes it so that Tx cleanup is done in a do/while loop instead of a for loop. The main motivation behind this is the fact that we should never be invoked with a budget less than 1 so we can skip checking the budget before processing the first descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:28:01 +00:00
do {
union e1000_adv_tx_desc *eop_desc = tx_buffer->next_to_watch;
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
/* if next_to_watch is not set then there is no work pending */
if (!eop_desc)
break;
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
igb: Change Tx cleanup loop to do/while instead of for This change makes it so that Tx cleanup is done in a do/while loop instead of a for loop. The main motivation behind this is the fact that we should never be invoked with a budget less than 1 so we can skip checking the budget before processing the first descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:28:01 +00:00
/* prevent any other reads prior to eop_desc */
igb: Use smp_rmb rather than read_barrier_depends The original issue being fixed in this patch was seen with the ixgbe driver, but the same issue exists with igb as well, as the code is very similar. read_barrier_depends is not sufficient to ensure loads following it are not speculatively loaded out of order by the CPU, which can result in stale data being loaded, causing potential system crashes. Cc: stable <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-11-17 17:05:47 +00:00
smp_rmb();
igb: Change Tx cleanup loop to do/while instead of for This change makes it so that Tx cleanup is done in a do/while loop instead of a for loop. The main motivation behind this is the fact that we should never be invoked with a budget less than 1 so we can skip checking the budget before processing the first descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:28:01 +00:00
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
/* if DD is not set pending work has not been completed */
if (!(eop_desc->wb.status & cpu_to_le32(E1000_TXD_STAT_DD)))
break;
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
/* clear next_to_watch to prevent false hangs */
tx_buffer->next_to_watch = NULL;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
/* update the statistics for this packet */
total_bytes += tx_buffer->bytecount;
total_packets += tx_buffer->gso_segs;
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
/* free the skb */
igb: Add support for bulk Tx cleanup & cleanup boolean logic This patch enables bulk free in Tx cleanup for igb and cleans up the boolean logic in the polling routines for igb in the hopes of avoiding any mix-ups similar to what occurred with i40e and i40evf. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-03-07 17:30:21 +00:00
napi_consume_skb(tx_buffer->skb, napi_budget);
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
/* unmap skb header data */
dma_unmap_single(tx_ring->dev,
igb: Use dma_unmap_addr and dma_unmap_len defines This change is meant to improve performance on systems that do not require the DMA unmap calls. On those systems we do not need to make use of the unmap address for Tx or the unmap length so we can drop both thereby reducing the size of the Tx buffer info structure. In addition I have changed the logic to check for unmap length instead of unmap address when checking to see if a buffer needs to be unmapped from DMA use. The reasons for this change is that on some platforms it is possible to receive a valid DMA address of 0 from an IOMMU. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-18 01:56:27 +00:00
dma_unmap_addr(tx_buffer, dma),
dma_unmap_len(tx_buffer, len),
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
DMA_TO_DEVICE);
igb: Use dma_unmap_addr and dma_unmap_len defines This change is meant to improve performance on systems that do not require the DMA unmap calls. On those systems we do not need to make use of the unmap address for Tx or the unmap length so we can drop both thereby reducing the size of the Tx buffer info structure. In addition I have changed the logic to check for unmap length instead of unmap address when checking to see if a buffer needs to be unmapped from DMA use. The reasons for this change is that on some platforms it is possible to receive a valid DMA address of 0 from an IOMMU. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-18 01:56:27 +00:00
/* clear tx_buffer data */
dma_unmap_len_set(tx_buffer, len, 0);
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
/* clear last DMA location and unmap remaining buffers */
while (tx_desc != eop_desc) {
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
tx_buffer++;
tx_desc++;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
i++;
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
if (unlikely(!i)) {
i -= tx_ring->count;
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
tx_buffer = tx_ring->tx_buffer_info;
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
tx_desc = IGB_TX_DESC(tx_ring, 0);
}
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
/* unmap any remaining paged data */
igb: Use dma_unmap_addr and dma_unmap_len defines This change is meant to improve performance on systems that do not require the DMA unmap calls. On those systems we do not need to make use of the unmap address for Tx or the unmap length so we can drop both thereby reducing the size of the Tx buffer info structure. In addition I have changed the logic to check for unmap length instead of unmap address when checking to see if a buffer needs to be unmapped from DMA use. The reasons for this change is that on some platforms it is possible to receive a valid DMA address of 0 from an IOMMU. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-18 01:56:27 +00:00
if (dma_unmap_len(tx_buffer, len)) {
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
dma_unmap_page(tx_ring->dev,
igb: Use dma_unmap_addr and dma_unmap_len defines This change is meant to improve performance on systems that do not require the DMA unmap calls. On those systems we do not need to make use of the unmap address for Tx or the unmap length so we can drop both thereby reducing the size of the Tx buffer info structure. In addition I have changed the logic to check for unmap length instead of unmap address when checking to see if a buffer needs to be unmapped from DMA use. The reasons for this change is that on some platforms it is possible to receive a valid DMA address of 0 from an IOMMU. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-18 01:56:27 +00:00
dma_unmap_addr(tx_buffer, dma),
dma_unmap_len(tx_buffer, len),
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
DMA_TO_DEVICE);
igb: Use dma_unmap_addr and dma_unmap_len defines This change is meant to improve performance on systems that do not require the DMA unmap calls. On those systems we do not need to make use of the unmap address for Tx or the unmap length so we can drop both thereby reducing the size of the Tx buffer info structure. In addition I have changed the logic to check for unmap length instead of unmap address when checking to see if a buffer needs to be unmapped from DMA use. The reasons for this change is that on some platforms it is possible to receive a valid DMA address of 0 from an IOMMU. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-18 01:56:27 +00:00
dma_unmap_len_set(tx_buffer, len, 0);
igb: consolidate creation of Tx buffer info and data descriptor This change will combine the writes of tx_buffer_info and the Tx data descriptors into a single function. The advantage of this is that we can avoid needless memory reads from the buffer info struct and speed things up by keeping the accesses to the local registers. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:09 +00:00
}
}
/* move us one more past the eop_desc for start of next pkt */
tx_buffer++;
tx_desc++;
i++;
if (unlikely(!i)) {
i -= tx_ring->count;
tx_buffer = tx_ring->tx_buffer_info;
tx_desc = IGB_TX_DESC(tx_ring, 0);
}
igb: Change Tx cleanup loop to do/while instead of for This change makes it so that Tx cleanup is done in a do/while loop instead of a for loop. The main motivation behind this is the fact that we should never be invoked with a budget less than 1 so we can skip checking the budget before processing the first descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:28:01 +00:00
/* issue prefetch for next Tx descriptor */
prefetch(tx_desc);
/* update budget accounting */
budget--;
} while (likely(budget));
igb: defeature tx head writeback This patch removes tx head writeback as it was found to not provide a significant improvement in tx performance and on some systems has been seen to cause a performance degredation due to partial cache line writes. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-26 09:33:18 +00:00
igb: Add support for byte queue limits. This adds support for byte queue limits (BQL) Since this driver collects bytes count in 'bytecount' field, use it also in igb_tx_map() Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> CC: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:36 +00:00
netdev_tx_completed_queue(txring_txq(tx_ring),
total_packets, total_bytes);
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
i += tx_ring->count;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
tx_ring->next_to_clean = i;
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
u64_stats_update_begin(&tx_ring->tx_syncp);
tx_ring->tx_stats.bytes += total_bytes;
tx_ring->tx_stats.packets += total_packets;
u64_stats_update_end(&tx_ring->tx_syncp);
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
q_vector->tx.total_bytes += total_bytes;
q_vector->tx.total_packets += total_packets;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: move TX hang check flag into ring->flags This change moves the Tx hang check into the ring flags. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:19 +00:00
if (test_bit(IGB_RING_FLAG_TX_DETECT_HANG, &tx_ring->flags)) {
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Detect a transmit hang in hardware, this serializes the
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* check with the clearing of time_stamp and movement of i
*/
igb: move TX hang check flag into ring->flags This change moves the Tx hang check into the ring flags. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:19 +00:00
clear_bit(IGB_RING_FLAG_TX_DETECT_HANG, &tx_ring->flags);
igb: Change Tx cleanup loop to do/while instead of for This change makes it so that Tx cleanup is done in a do/while loop instead of a for loop. The main motivation behind this is the fact that we should never be invoked with a budget less than 1 so we can skip checking the budget before processing the first descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:28:01 +00:00
if (tx_buffer->next_to_watch &&
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
time_after(jiffies, tx_buffer->time_stamp +
drivers/net: Move && and || to end of previous line Only files where David Miller is the primary git-signer. wireless, wimax, ixgbe, etc are not modified. Compile tested x86 allyesconfig only Not all files compiled (not x86 compatible) Added a few > 80 column lines, which I ignored. Existing checkpatch complaints ignored. Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-03 07:58:21 +00:00
(adapter->tx_timeout_factor * HZ)) &&
!(rd32(E1000_STATUS) & E1000_STATUS_TXOFF)) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* detected Tx unit hang */
igb: convert igb from using PCI DMA functions to using DMA API functions This patch makes it so that igb now uses the DMA API functions instead of the PCI API functions. To do this the pci_dev pointer that was in the rings has been replaced with a device pointer, and as a result all references to [tr]x_ring->pdev have been replaced with [tr]x_ring->dev. This patch is based of of work originally done by Nicholas Nunley. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-27 13:09:25 +00:00
dev_err(tx_ring->dev,
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
"Detected Tx Unit Hang\n"
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
" Tx Queue <%d>\n"
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
" TDH <%x>\n"
" TDT <%x>\n"
" next_to_use <%x>\n"
" next_to_clean <%x>\n"
"buffer_info[next_to_clean]\n"
" time_stamp <%lx>\n"
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
" next_to_watch <%p>\n"
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
" jiffies <%lx>\n"
" desc.status <%x>\n",
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
tx_ring->queue_index,
igb: update ring and adapter structure to improve performance This change is meant to improve performance by splitting the Tx and Rx rings into 3 sections. The first is primarily a read only section containing basic things like the indexes, a pointer to the dev and netdev structures, and basic information. The second section contains the stats and next_to_use and next_to_clean values. The third section is primarily unused values that can just be placed at the end of the ring and are not used in the hot path. The adapter structure has several sections that are read in the hot path. In order to improve performance there I am combining the frequent read hot path items into a single cache line. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:48 +00:00
rd32(E1000_TDH(tx_ring->reg_idx)),
igb: change the head and tail offsets into pointers Since we are writting to the head/tail pointers frequently we might as well save ourselves some processing time by converting the head and tail offsets directly to pointers. This will shave a few cycles off the rx/tx path and allows us to move one step closer to the rings being a bit more independant of each other. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:51:27 +00:00
readl(tx_ring->tail),
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
tx_ring->next_to_use,
tx_ring->next_to_clean,
igb: Make first and tx_buffer_info->next_to_watch into pointers This change converts two tx_buffer_info index values into pointers. The advantage to this is that we reduce unnecessary computations and in the case of next_to_watch we get an added bonus of the value being able to provide additional information as a NULL value indicates it is unset versus a 0 not having any meaning for the index value. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:43 +00:00
tx_buffer->time_stamp,
igb: Change Tx cleanup loop to do/while instead of for This change makes it so that Tx cleanup is done in a do/while loop instead of a for loop. The main motivation behind this is the fact that we should never be invoked with a budget less than 1 so we can skip checking the budget before processing the first descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:28:01 +00:00
tx_buffer->next_to_watch,
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
jiffies,
igb: Change Tx cleanup loop to do/while instead of for This change makes it so that Tx cleanup is done in a do/while loop instead of a for loop. The main motivation behind this is the fact that we should never be invoked with a budget less than 1 so we can skip checking the budget before processing the first descriptor. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-13 06:28:01 +00:00
tx_buffer->next_to_watch->wb.status);
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
netif_stop_subqueue(tx_ring->netdev,
tx_ring->queue_index);
/* we are about to reset, no point in enabling stuff */
return true;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
}
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
igb: Update igb to use a path similar to ixgbe to determine when to stop Tx After reviewing the igb and ixgbe code I realized there are a few issues in how the code is structured. Specifically we are not checking the size of the buffers being used in transmits and we are not using the same value to determine when to stop or start a Tx queue. As such the code is prone to be buggy. This patch makes it so that we have one value DESC_NEEDED that we will use for starting and stopping the queue. In addition we will check the size of buffers being used when setting up a transmit so as to avoid a possible buffer overrun if we were to receive a frame with a block of data larger than 32K in skb->data. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-09 04:27:48 +00:00
#define TX_WAKE_THRESHOLD (DESC_NEEDED * 2)
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
if (unlikely(total_packets &&
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
netif_carrier_ok(tx_ring->netdev) &&
igb_desc_unused(tx_ring) >= TX_WAKE_THRESHOLD)) {
igb: Make Tx budget for NAPI user adjustable This change is to make the NAPI budget limits for transmit adjustable. Currently they are only set to 128, and when the changes/improvements to NAPI occur to allow for adjustability, it would be possible to tune the value for optimal performance with applications such as routing. v2: remove tie between NAPI and interrupt moderation fix work limit define name (s/IXGBE/IGB/) Update patch description to better reflect patch Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2011-10-05 13:35:24 +00:00
/* Make sure that anybody stopping the queue after this
* sees the new next_to_clean.
*/
smp_mb();
if (__netif_subqueue_stopped(tx_ring->netdev,
tx_ring->queue_index) &&
!(test_bit(__IGB_DOWN, &adapter->state))) {
netif_wake_subqueue(tx_ring->netdev,
tx_ring->queue_index);
u64_stats_update_begin(&tx_ring->tx_syncp);
tx_ring->tx_stats.restart_queue++;
u64_stats_update_end(&tx_ring->tx_syncp);
}
}
return !!budget;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_reuse_rx_page - page flip buffer and store it back on the ring
* @rx_ring: rx descriptor ring to store buffers on
* @old_buff: donor buffer to have page reused
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Synchronizes page for reuse by the adapter
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
**/
static void igb_reuse_rx_page(struct igb_ring *rx_ring,
struct igb_rx_buffer *old_buff)
{
struct igb_rx_buffer *new_buff;
u16 nta = rx_ring->next_to_alloc;
new_buff = &rx_ring->rx_buffer_info[nta];
/* update, and store next to alloc */
nta++;
rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0;
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
/* Transfer page from old buffer to new buffer.
* Move each member individually to avoid possible store
* forwarding stalls.
*/
new_buff->dma = old_buff->dma;
new_buff->page = old_buff->page;
new_buff->page_offset = old_buff->page_offset;
new_buff->pagecnt_bias = old_buff->pagecnt_bias;
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
}
igb: Clean-up page reuse code This patch cleans up the page reuse code getting it into a state where all the workarounds needed are in place as well as cleaning up a few minor oversights such as using __free_pages instead of put_page to drop a locally allocated page. It also cleans up how we clear the descriptor status bits. Previously they were zeroed as a part of clearing the hdr_addr. However the hdr_addr is a 64 bit field and 64 bit writes can be a bit more expensive on on 32 bit systems. Since we are no longer using the header split feature the upper 32 bits of the address no longer need to be cleared. As a result we can just clear the status bits and leave the length and VLAN fields as-is which should provide more information in debugging. Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-14 00:56:19 +00:00
static inline bool igb_page_is_reserved(struct page *page)
{
mm: make page pfmemalloc check more robust Commit c48a11c7ad26 ("netvm: propagate page->pfmemalloc to skb") added checks for page->pfmemalloc to __skb_fill_page_desc(): if (page->pfmemalloc && !page->mapping) skb->pfmemalloc = true; It assumes page->mapping == NULL implies that page->pfmemalloc can be trusted. However, __delete_from_page_cache() can set set page->mapping to NULL and leave page->index value alone. Due to being in union, a non-zero page->index will be interpreted as true page->pfmemalloc. So the assumption is invalid if the networking code can see such a page. And it seems it can. We have encountered this with a NFS over loopback setup when such a page is attached to a new skbuf. There is no copying going on in this case so the page confuses __skb_fill_page_desc which interprets the index as pfmemalloc flag and the network stack drops packets that have been allocated using the reserves unless they are to be queued on sockets handling the swapping which is the case here and that leads to hangs when the nfs client waits for a response from the server which has been dropped and thus never arrive. The struct page is already heavily packed so rather than finding another hole to put it in, let's do a trick instead. We can reuse the index again but define it to an impossible value (-1UL). This is the page index so it should never see the value that large. Replace all direct users of page->pfmemalloc by page_is_pfmemalloc which will hide this nastiness from unspoiled eyes. The information will get lost if somebody wants to use page->index obviously but that was the case before and the original code expected that the information should be persisted somewhere else if that is really needed (e.g. what SLAB and SLUB do). [akpm@linux-foundation.org: fix blooper in slub] Fixes: c48a11c7ad26 ("netvm: propagate page->pfmemalloc to skb") Signed-off-by: Michal Hocko <mhocko@suse.com> Debugged-by: Vlastimil Babka <vbabka@suse.com> Debugged-by: Jiri Bohac <jbohac@suse.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: David Miller <davem@davemloft.net> Acked-by: Mel Gorman <mgorman@suse.de> Cc: <stable@vger.kernel.org> [3.6+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-08-21 21:11:51 +00:00
return (page_to_nid(page) != numa_mem_id()) || page_is_pfmemalloc(page);
igb: Clean-up page reuse code This patch cleans up the page reuse code getting it into a state where all the workarounds needed are in place as well as cleaning up a few minor oversights such as using __free_pages instead of put_page to drop a locally allocated page. It also cleans up how we clear the descriptor status bits. Previously they were zeroed as a part of clearing the hdr_addr. However the hdr_addr is a 64 bit field and 64 bit writes can be a bit more expensive on on 32 bit systems. Since we are no longer using the header split feature the upper 32 bits of the address no longer need to be cleared. As a result we can just clear the status bits and leave the length and VLAN fields as-is which should provide more information in debugging. Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-14 00:56:19 +00:00
}
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
static bool igb_can_reuse_rx_page(struct igb_rx_buffer *rx_buffer)
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
{
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
unsigned int pagecnt_bias = rx_buffer->pagecnt_bias;
struct page *page = rx_buffer->page;
igb: update code to better handle incrementing page count Update the driver code so that we do bulk updates of the page reference count instead of just incrementing it by one reference at a time. The advantage to doing this is that we cut down on atomic operations and this in turn should give us a slight improvement in cycles per packet. In addition if we eventually move this over to using build_skb the gains will be more noticeable. Link: http://lkml.kernel.org/r/20161110113616.76501.17072.stgit@ahduyck-blue-test.jf.intel.com Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Chris Metcalf <cmetcalf@mellanox.com> Cc: David Howells <dhowells@redhat.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Cc: Helge Deller <deller@gmx.de> Cc: James Hogan <james.hogan@imgtec.com> Cc: Jonas Bonn <jonas@southpole.se> Cc: Keguang Zhang <keguang.zhang@gmail.com> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Rich Felker <dalias@libc.org> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Steven Miao <realmz6@gmail.com> Cc: Tobias Klauser <tklauser@distanz.ch> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-14 23:05:34 +00:00
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
/* avoid re-using remote pages */
igb: Clean-up page reuse code This patch cleans up the page reuse code getting it into a state where all the workarounds needed are in place as well as cleaning up a few minor oversights such as using __free_pages instead of put_page to drop a locally allocated page. It also cleans up how we clear the descriptor status bits. Previously they were zeroed as a part of clearing the hdr_addr. However the hdr_addr is a 64 bit field and 64 bit writes can be a bit more expensive on on 32 bit systems. Since we are no longer using the header split feature the upper 32 bits of the address no longer need to be cleared. As a result we can just clear the status bits and leave the length and VLAN fields as-is which should provide more information in debugging. Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-14 00:56:19 +00:00
if (unlikely(igb_page_is_reserved(page)))
igb: don't reuse pages with pfmemalloc flag Incoming packet is dropped silently by sk_filter(), if the skb was allocated from pfmemalloc reserves and the corresponding socket is not marked with the SOCK_MEMALLOC flag. Igb driver allocates pages for DMA with __skb_alloc_page(), which calls alloc_pages_node() with the __GFP_MEMALLOC flag. So, in case of OOM condition, igb can get pages with pfmemalloc flag set. If an incoming packet hits the pfmemalloc page and is large enough (small packets are copying into the memory, allocated with netdev_alloc_skb_ip_align(), so they are not affected), it will be dropped. This behavior is ok under high memory pressure, but the problem is that the igb driver reuses these mapped pages. So, packets are still dropping even if all memory issues are gone and there is a plenty of free memory. In my case, some TCP sessions hang on a small percentage (< 0.1%) of machines days after OOMs. Fix this by avoiding reuse of such pages. Signed-off-by: Roman Gushchin <klamm@yandex-team.ru> Tested-by: Aaron Brown "aaron.f.brown@intel.com" Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-10-23 03:32:27 +00:00
return false;
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
#if (PAGE_SIZE < 8192)
/* if we are only owner of page we can reuse it */
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
if (unlikely((page_ref_count(page) - pagecnt_bias) > 1))
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
return false;
#else
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
#define IGB_LAST_OFFSET \
(SKB_WITH_OVERHEAD(PAGE_SIZE) - IGB_RXBUFFER_2048)
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
if (rx_buffer->page_offset > IGB_LAST_OFFSET)
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
return false;
#endif
igb: update code to better handle incrementing page count Update the driver code so that we do bulk updates of the page reference count instead of just incrementing it by one reference at a time. The advantage to doing this is that we cut down on atomic operations and this in turn should give us a slight improvement in cycles per packet. In addition if we eventually move this over to using build_skb the gains will be more noticeable. Link: http://lkml.kernel.org/r/20161110113616.76501.17072.stgit@ahduyck-blue-test.jf.intel.com Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Chris Metcalf <cmetcalf@mellanox.com> Cc: David Howells <dhowells@redhat.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Cc: Helge Deller <deller@gmx.de> Cc: James Hogan <james.hogan@imgtec.com> Cc: Jonas Bonn <jonas@southpole.se> Cc: Keguang Zhang <keguang.zhang@gmail.com> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Rich Felker <dalias@libc.org> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Steven Miao <realmz6@gmail.com> Cc: Tobias Klauser <tklauser@distanz.ch> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-14 23:05:34 +00:00
/* If we have drained the page fragment pool we need to update
* the pagecnt_bias and page count so that we fully restock the
* number of references the driver holds.
igb: Clean-up page reuse code This patch cleans up the page reuse code getting it into a state where all the workarounds needed are in place as well as cleaning up a few minor oversights such as using __free_pages instead of put_page to drop a locally allocated page. It also cleans up how we clear the descriptor status bits. Previously they were zeroed as a part of clearing the hdr_addr. However the hdr_addr is a 64 bit field and 64 bit writes can be a bit more expensive on on 32 bit systems. Since we are no longer using the header split feature the upper 32 bits of the address no longer need to be cleared. As a result we can just clear the status bits and leave the length and VLAN fields as-is which should provide more information in debugging. Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-14 00:56:19 +00:00
*/
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
if (unlikely(!pagecnt_bias)) {
igb: update code to better handle incrementing page count Update the driver code so that we do bulk updates of the page reference count instead of just incrementing it by one reference at a time. The advantage to doing this is that we cut down on atomic operations and this in turn should give us a slight improvement in cycles per packet. In addition if we eventually move this over to using build_skb the gains will be more noticeable. Link: http://lkml.kernel.org/r/20161110113616.76501.17072.stgit@ahduyck-blue-test.jf.intel.com Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Chris Metcalf <cmetcalf@mellanox.com> Cc: David Howells <dhowells@redhat.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Cc: Helge Deller <deller@gmx.de> Cc: James Hogan <james.hogan@imgtec.com> Cc: Jonas Bonn <jonas@southpole.se> Cc: Keguang Zhang <keguang.zhang@gmail.com> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Rich Felker <dalias@libc.org> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Steven Miao <realmz6@gmail.com> Cc: Tobias Klauser <tklauser@distanz.ch> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-14 23:05:34 +00:00
page_ref_add(page, USHRT_MAX);
rx_buffer->pagecnt_bias = USHRT_MAX;
}
igb: Clean-up page reuse code This patch cleans up the page reuse code getting it into a state where all the workarounds needed are in place as well as cleaning up a few minor oversights such as using __free_pages instead of put_page to drop a locally allocated page. It also cleans up how we clear the descriptor status bits. Previously they were zeroed as a part of clearing the hdr_addr. However the hdr_addr is a 64 bit field and 64 bit writes can be a bit more expensive on on 32 bit systems. Since we are no longer using the header split feature the upper 32 bits of the address no longer need to be cleared. As a result we can just clear the status bits and leave the length and VLAN fields as-is which should provide more information in debugging. Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-11-14 00:56:19 +00:00
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
return true;
}
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_add_rx_frag - Add contents of Rx buffer to sk_buff
* @rx_ring: rx descriptor ring to transact packets on
* @rx_buffer: buffer containing page to add
* @skb: sk_buff to place the data into
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
* @size: size of buffer to be added
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* This function will add the data contained in rx_buffer->page to the skb.
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
**/
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
static void igb_add_rx_frag(struct igb_ring *rx_ring,
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
struct igb_rx_buffer *rx_buffer,
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
struct sk_buff *skb,
unsigned int size)
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
{
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
#if (PAGE_SIZE < 8192)
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
unsigned int truesize = igb_rx_pg_size(rx_ring) / 2;
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
#else
igb: Add support for padding packet With the size of the frame limited we can now write to an offset within the buffer instead of having to write at the very start of the buffer. The advantage to this is that it allows us to leave padding room for things like supporting XDP in the future. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:14 +00:00
unsigned int truesize = ring_uses_build_skb(rx_ring) ?
SKB_DATA_ALIGN(IGB_SKB_PAD + size) :
SKB_DATA_ALIGN(size);
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
#endif
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buffer->page,
rx_buffer->page_offset, size, truesize);
#if (PAGE_SIZE < 8192)
rx_buffer->page_offset ^= truesize;
#else
rx_buffer->page_offset += truesize;
#endif
}
static struct sk_buff *igb_construct_skb(struct igb_ring *rx_ring,
struct igb_rx_buffer *rx_buffer,
union e1000_adv_rx_desc *rx_desc,
unsigned int size)
{
void *va = page_address(rx_buffer->page) + rx_buffer->page_offset;
#if (PAGE_SIZE < 8192)
unsigned int truesize = igb_rx_pg_size(rx_ring) / 2;
#else
unsigned int truesize = SKB_DATA_ALIGN(size);
#endif
unsigned int headlen;
struct sk_buff *skb;
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
/* prefetch first cache line of first page */
prefetch(va);
#if L1_CACHE_BYTES < 128
prefetch(va + L1_CACHE_BYTES);
#endif
/* allocate a skb to store the frags */
skb = napi_alloc_skb(&rx_ring->q_vector->napi, IGB_RX_HDR_LEN);
if (unlikely(!skb))
return NULL;
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
igb: Pull timestamp from fragment before adding it to skb This change makes it so that we pull the timestamp from the fragment before we add it to the skb. By doing this we can avoid a possible issue in which the fragment can possibly be less than IGB_RX_HDR_LEN due to the timestamp being pulled after the copybreak check. While making this change I realized we could also pull the rest of the igb_pull_tail function into igb_add_rx_frag since in the case of igb, unlike ixgbe, we are able to unmap the entire buffer before calling add_rx_frag so merging the two allows for sharing of code between the two merged functions. Reported-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-04-23 04:49:17 +00:00
if (unlikely(igb_test_staterr(rx_desc, E1000_RXDADV_STAT_TSIP))) {
igb_ptp_rx_pktstamp(rx_ring->q_vector, va, skb);
va += IGB_TS_HDR_LEN;
size -= IGB_TS_HDR_LEN;
}
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
/* Determine available headroom for copy */
headlen = size;
if (headlen > IGB_RX_HDR_LEN)
headlen = eth_get_headlen(va, IGB_RX_HDR_LEN);
igb: Pull timestamp from fragment before adding it to skb This change makes it so that we pull the timestamp from the fragment before we add it to the skb. By doing this we can avoid a possible issue in which the fragment can possibly be less than IGB_RX_HDR_LEN due to the timestamp being pulled after the copybreak check. While making this change I realized we could also pull the rest of the igb_pull_tail function into igb_add_rx_frag since in the case of igb, unlike ixgbe, we are able to unmap the entire buffer before calling add_rx_frag so merging the two allows for sharing of code between the two merged functions. Reported-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-04-23 04:49:17 +00:00
/* align pull length to size of long to optimize memcpy performance */
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
memcpy(__skb_put(skb, headlen), va, ALIGN(headlen, sizeof(long)));
igb: Pull timestamp from fragment before adding it to skb This change makes it so that we pull the timestamp from the fragment before we add it to the skb. By doing this we can avoid a possible issue in which the fragment can possibly be less than IGB_RX_HDR_LEN due to the timestamp being pulled after the copybreak check. While making this change I realized we could also pull the rest of the igb_pull_tail function into igb_add_rx_frag since in the case of igb, unlike ixgbe, we are able to unmap the entire buffer before calling add_rx_frag so merging the two allows for sharing of code between the two merged functions. Reported-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-04-23 04:49:17 +00:00
/* update all of the pointers */
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
size -= headlen;
if (size) {
skb_add_rx_frag(skb, 0, rx_buffer->page,
(va + headlen) - page_address(rx_buffer->page),
size, truesize);
#if (PAGE_SIZE < 8192)
rx_buffer->page_offset ^= truesize;
#else
rx_buffer->page_offset += truesize;
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
#endif
} else {
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
rx_buffer->pagecnt_bias++;
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
}
return skb;
}
igb: Re-add support for build_skb in igb This reverts commit f9d40f6a9921 ("igb: Revert support for build_skb in igb") and adds a few changes to update it to work with the latest version of igb. We are now able to revert the removal of this due to the fact that with the recent changes to the page count and the use of DMA_ATTR_SKIP_CPU_SYNC we can make the pages writable so we should not be invalidating the additional data added when we call build_skb. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:36 +00:00
static struct sk_buff *igb_build_skb(struct igb_ring *rx_ring,
struct igb_rx_buffer *rx_buffer,
union e1000_adv_rx_desc *rx_desc,
unsigned int size)
{
void *va = page_address(rx_buffer->page) + rx_buffer->page_offset;
#if (PAGE_SIZE < 8192)
unsigned int truesize = igb_rx_pg_size(rx_ring) / 2;
#else
unsigned int truesize = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) +
SKB_DATA_ALIGN(IGB_SKB_PAD + size);
#endif
struct sk_buff *skb;
/* prefetch first cache line of first page */
prefetch(va);
#if L1_CACHE_BYTES < 128
prefetch(va + L1_CACHE_BYTES);
#endif
igb/ixgbe: Fix typo in igb_build_skb and/or ixgbe_build_skb code comment There was a typo that I had left in the code comments for the igb and ixgbe functions that enabled build_skb support. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-15 17:15:59 +00:00
/* build an skb around the page buffer */
igb: Re-add support for build_skb in igb This reverts commit f9d40f6a9921 ("igb: Revert support for build_skb in igb") and adds a few changes to update it to work with the latest version of igb. We are now able to revert the removal of this due to the fact that with the recent changes to the page count and the use of DMA_ATTR_SKIP_CPU_SYNC we can make the pages writable so we should not be invalidating the additional data added when we call build_skb. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:36 +00:00
skb = build_skb(va - IGB_SKB_PAD, truesize);
if (unlikely(!skb))
return NULL;
/* update pointers within the skb to store the data */
skb_reserve(skb, IGB_SKB_PAD);
__skb_put(skb, size);
/* pull timestamp out of packet data */
if (igb_test_staterr(rx_desc, E1000_RXDADV_STAT_TSIP)) {
igb_ptp_rx_pktstamp(rx_ring->q_vector, skb->data, skb);
__skb_pull(skb, IGB_TS_HDR_LEN);
}
/* update buffer offset */
#if (PAGE_SIZE < 8192)
rx_buffer->page_offset ^= truesize;
#else
rx_buffer->page_offset += truesize;
#endif
return skb;
}
igb: drop the "adv" off function names relating to descriptors Many of the function names in the hot path are carrying an extra "_adv" suffix on the end of them to represent the fact that they are using advanced descriptors instead of legacy descriptors. However since all igb uses are advanced descriptors adding the extra suffix doesn't really add any additional data. Since this is the case it is easiest to just drop the suffix and save us from having to store the extra characters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:59 +00:00
static inline void igb_rx_checksum(struct igb_ring *ring,
igb: leave staterr in place and instead us a helper function to check bits Instead of doing a byte swap on the staterr bits in the Rx descriptor we can save ourselves a bit of space and some CPU time by instead just testing for the various bits out of the Rx descriptor directly. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:03 +00:00
union e1000_adv_rx_desc *rx_desc,
struct sk_buff *skb)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
drivers/net: avoid some skb->ip_summed initializations fresh skbs have ip_summed set to CHECKSUM_NONE (0) We can avoid setting again skb->ip_summed to CHECKSUM_NONE in drivers. Introduce skb_checksum_none_assert() helper so that we keep this assertion documented in driver sources. Change most occurrences of : skb->ip_summed = CHECKSUM_NONE; by : skb_checksum_none_assert(skb); Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-09-02 20:07:41 +00:00
skb_checksum_none_assert(skb);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: retire the RX_CSUM flag and use the netdev flag instead Since the netdev now has its' own checksum flag to indicate if Rx checksum is enabled we might as well use that instead of using the ring flag. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:57 +00:00
/* Ignore Checksum bit is set */
igb: leave staterr in place and instead us a helper function to check bits Instead of doing a byte swap on the staterr bits in the Rx descriptor we can save ourselves a bit of space and some CPU time by instead just testing for the various bits out of the Rx descriptor directly. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:03 +00:00
if (igb_test_staterr(rx_desc, E1000_RXD_STAT_IXSM))
igb: retire the RX_CSUM flag and use the netdev flag instead Since the netdev now has its' own checksum flag to indicate if Rx checksum is enabled we might as well use that instead of using the ring flag. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:57 +00:00
return;
/* Rx checksum disabled via ethtool */
if (!(ring->netdev->features & NETIF_F_RXCSUM))
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return;
igb: add a flags value to the ring This patch adds a flags value to the ring that cleans up some of the last remaining items from the ring in order to help seperate it from the adapter struct. By implementing these flags it becomes possible for different rings to support different functions such as rx checksumming. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:52:46 +00:00
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* TCP/UDP checksum error bit is set */
igb: leave staterr in place and instead us a helper function to check bits Instead of doing a byte swap on the staterr bits in the Rx descriptor we can save ourselves a bit of space and some CPU time by instead just testing for the various bits out of the Rx descriptor directly. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:03 +00:00
if (igb_test_staterr(rx_desc,
E1000_RXDEXT_STATERR_TCPE |
E1000_RXDEXT_STATERR_IPE)) {
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* work around errata with sctp packets where the TCPE aka
igb: Enable SCTP checksum offloading Originally from: Vlad Yasevich <vladislav.yasevich@hp.com> This patch, both the driver portion and the sctp code was modified by Jesse Brandeburg and is Copyright(c) 2009 Intel Corporation. Thanks go to Vlad for starting this work. Intel 82576 chipset supports SCTP checksum offloading. This patch enables this functionality in the driver. A new NETIF feature is introduced for SCTP checksum offload. If the driver supports CRC32c checksum, it can set this feature flag. The hardware can offload both transmit and receive. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-04-27 22:36:13 +00:00
* L4E bit is set incorrectly on 64 byte (60 byte w/o crc)
* packets, (aka let the stack check the crc32c)
*/
igb: Consolidate all of the ring feature flags into a single value This change moves all of the ring flags into a single value. The advantage to this is that there is one central area for all of these flags and they can all make use of the set/test bit operations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:36 +00:00
if (!((skb->len == 60) &&
test_bit(IGB_RING_FLAG_RX_SCTP_CSUM, &ring->flags))) {
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
u64_stats_update_begin(&ring->rx_syncp);
igb: move alloc_failed and csum_err stats into per rx-ring stat The allocation failed and checksum error stats are currently kept as a global stat. If we end up allocating the queues to multiple netdevs then the global counter doesn't make much sense. For this reason I felt it necessary to move the alloc_rx_buff_failed stat into the rx_stats portion of the rx_ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:52:27 +00:00
ring->rx_stats.csum_err++;
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
u64_stats_update_end(&ring->rx_syncp);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* let the stack verify checksum errors */
return;
}
/* It must be a TCP or UDP packet with a valid checksum */
igb: leave staterr in place and instead us a helper function to check bits Instead of doing a byte swap on the staterr bits in the Rx descriptor we can save ourselves a bit of space and some CPU time by instead just testing for the various bits out of the Rx descriptor directly. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:03 +00:00
if (igb_test_staterr(rx_desc, E1000_RXD_STAT_TCPCS |
E1000_RXD_STAT_UDPCS))
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
skb->ip_summed = CHECKSUM_UNNECESSARY;
igb: leave staterr in place and instead us a helper function to check bits Instead of doing a byte swap on the staterr bits in the Rx descriptor we can save ourselves a bit of space and some CPU time by instead just testing for the various bits out of the Rx descriptor directly. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:03 +00:00
dev_dbg(ring->dev, "cksum success: bits %08X\n",
le32_to_cpu(rx_desc->wb.upper.status_error));
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: add support for NETIF_F_RXHASH This patch adds support for Rx hashing. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:29 +00:00
static inline void igb_rx_hash(struct igb_ring *ring,
union e1000_adv_rx_desc *rx_desc,
struct sk_buff *skb)
{
if (ring->netdev->features & NETIF_F_RXHASH)
net: igb calls skb_set_hash Drivers should call skb_set_hash to set the hash and its type in an skbuff. Signed-off-by: Tom Herbert <therbert@google.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-18 16:46:58 +00:00
skb_set_hash(skb,
le32_to_cpu(rx_desc->wb.lower.hi_dword.rss),
PKT_HASH_TYPE_L3);
igb: add support for NETIF_F_RXHASH This patch adds support for Rx hashing. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:29 +00:00
}
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_is_non_eop - process handling of non-EOP buffers
* @rx_ring: Rx ring being processed
* @rx_desc: Rx descriptor for current buffer
* @skb: current socket buffer containing buffer in progress
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* This function updates next to clean. If the buffer is an EOP buffer
* this function exits returning false, otherwise it will place the
* sk_buff in the next buffer to be chained and return true indicating
* that this is in fact a non-EOP buffer.
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
**/
static bool igb_is_non_eop(struct igb_ring *rx_ring,
union e1000_adv_rx_desc *rx_desc)
{
u32 ntc = rx_ring->next_to_clean + 1;
/* fetch, update, and store next to clean */
ntc = (ntc < rx_ring->count) ? ntc : 0;
rx_ring->next_to_clean = ntc;
prefetch(IGB_RX_DESC(rx_ring, ntc));
if (likely(igb_test_staterr(rx_desc, E1000_RXD_STAT_EOP)))
return false;
return true;
}
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_cleanup_headers - Correct corrupted or empty headers
* @rx_ring: rx descriptor ring packet is being transacted on
* @rx_desc: pointer to the EOP Rx descriptor
* @skb: pointer to current skb being fixed
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Address the case where we are pulling data in on pages only
* and as such no data is present in the skb header.
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* In addition if skb is not at least 60 bytes we need to pad it so that
* it is large enough to qualify as a valid Ethernet frame.
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Returns true if an error was encountered and skb was freed.
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
**/
static bool igb_cleanup_headers(struct igb_ring *rx_ring,
union e1000_adv_rx_desc *rx_desc,
struct sk_buff *skb)
{
if (unlikely((igb_test_staterr(rx_desc,
E1000_RXDEXT_ERR_FRAME_ERR_MASK)))) {
struct net_device *netdev = rx_ring->netdev;
if (!(netdev->features & NETIF_F_RXALL)) {
dev_kfree_skb_any(skb);
return true;
}
}
ethernet/intel: Use eth_skb_pad and skb_put_padto helpers Update the Intel Ethernet drivers to use eth_skb_pad() and skb_put_padto instead of doing their own implementations of the function. Also this cleans up two other spots where skb_pad was called but the length and tail pointers were being manipulated directly instead of just having the padding length added via __skb_put. Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-03 16:17:39 +00:00
/* if eth_skb_pad returns an error the skb was freed */
if (eth_skb_pad(skb))
return true;
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
return false;
igb: use buffer_info->dma instead of shinfo->nr_frags to determine unmap This change makes it so that we use buffer_info->dma instead of shinfo->nr_frags to determine if we need to unmap a received skb. By doing this we can avoid a cache miss on small packets since the buffer_info structure should already be prefetched. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:10:06 +00:00
}
igb: Combine post-processing of skb into a single function This change is meant to just clean-up a number of function calls that were made at the end of the Rx clean-up path by combining them into a single function call. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:57 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_process_skb_fields - Populate skb header fields from Rx descriptor
* @rx_ring: rx descriptor ring packet is being transacted on
* @rx_desc: pointer to the EOP Rx descriptor
* @skb: pointer to current skb being populated
igb: Combine post-processing of skb into a single function This change is meant to just clean-up a number of function calls that were made at the end of the Rx clean-up path by combining them into a single function call. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:57 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* This function checks the ring, descriptor, and packet information in
* order to populate the hash, checksum, VLAN, timestamp, protocol, and
* other fields within the skb.
igb: Combine post-processing of skb into a single function This change is meant to just clean-up a number of function calls that were made at the end of the Rx clean-up path by combining them into a single function call. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:57 +00:00
**/
static void igb_process_skb_fields(struct igb_ring *rx_ring,
union e1000_adv_rx_desc *rx_desc,
struct sk_buff *skb)
{
struct net_device *dev = rx_ring->netdev;
igb_rx_hash(rx_ring, rx_desc, skb);
igb_rx_checksum(rx_ring, rx_desc, skb);
igb: fix last_rx_timestamp usage last_rx_timestamp should be updated only when rx time stamp is read. Also it's only used with NICs that have per-interface time stamping resources so it can be moved to adapter structure and set in igb_ptp_rx_rgtstamp(). Signed-off-by: Jakub Kicinski <kubakici@wp.pl> Acked-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-02 10:33:33 +00:00
if (igb_test_staterr(rx_desc, E1000_RXDADV_STAT_TS) &&
!igb_test_staterr(rx_desc, E1000_RXDADV_STAT_TSIP))
igb_ptp_rx_rgtstamp(rx_ring->q_vector, skb);
igb: Combine post-processing of skb into a single function This change is meant to just clean-up a number of function calls that were made at the end of the Rx clean-up path by combining them into a single function call. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:57 +00:00
net: vlan: rename NETIF_F_HW_VLAN_* feature flags to NETIF_F_HW_VLAN_CTAG_* Rename the hardware VLAN acceleration features to include "CTAG" to indicate that they only support CTAGs. Follow up patches will introduce 802.1ad server provider tagging (STAGs) and require the distinction for hardware not supporting acclerating both. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-19 02:04:27 +00:00
if ((dev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
igb: Combine post-processing of skb into a single function This change is meant to just clean-up a number of function calls that were made at the end of the Rx clean-up path by combining them into a single function call. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:57 +00:00
igb_test_staterr(rx_desc, E1000_RXD_STAT_VP)) {
u16 vid;
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: Combine post-processing of skb into a single function This change is meant to just clean-up a number of function calls that were made at the end of the Rx clean-up path by combining them into a single function call. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:57 +00:00
if (igb_test_staterr(rx_desc, E1000_RXDEXT_STATERR_LB) &&
test_bit(IGB_RING_FLAG_RX_LB_VLAN_BSWAP, &rx_ring->flags))
vid = be16_to_cpu(rx_desc->wb.upper.vlan);
else
vid = le16_to_cpu(rx_desc->wb.upper.vlan);
net: vlan: add protocol argument to packet tagging functions Add a protocol argument to the VLAN packet tagging functions. In case of HW tagging, we need that protocol available in the ndo_start_xmit functions, so it is stored in a new field in the skb. The new field fits into a hole (on 64 bit) and doesn't increase the sks's size. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-19 02:04:30 +00:00
__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vid);
igb: Combine post-processing of skb into a single function This change is meant to just clean-up a number of function calls that were made at the end of the Rx clean-up path by combining them into a single function call. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:57 +00:00
}
skb_record_rx_queue(skb, rx_ring->queue_index);
skb->protocol = eth_type_trans(skb, rx_ring->netdev);
}
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
static struct igb_rx_buffer *igb_get_rx_buffer(struct igb_ring *rx_ring,
const unsigned int size)
{
struct igb_rx_buffer *rx_buffer;
rx_buffer = &rx_ring->rx_buffer_info[rx_ring->next_to_clean];
prefetchw(rx_buffer->page);
/* we are reusing so sync this buffer for CPU use */
dma_sync_single_range_for_cpu(rx_ring->dev,
rx_buffer->dma,
rx_buffer->page_offset,
size,
DMA_FROM_DEVICE);
rx_buffer->pagecnt_bias--;
return rx_buffer;
}
static void igb_put_rx_buffer(struct igb_ring *rx_ring,
struct igb_rx_buffer *rx_buffer)
{
if (igb_can_reuse_rx_page(rx_buffer)) {
/* hand second half of page back to the ring */
igb_reuse_rx_page(rx_ring, rx_buffer);
} else {
/* We are not reusing the buffer so unmap it and free
* any references we are holding to it
*/
dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
igb_rx_pg_size(rx_ring), DMA_FROM_DEVICE,
IGB_RX_DMA_ATTR);
__page_frag_cache_drain(rx_buffer->page,
rx_buffer->pagecnt_bias);
}
/* clear contents of rx_buffer */
rx_buffer->page = NULL;
}
drivers/net/intel: use napi_complete_done() As per Eric Dumazet's previous patches: (see commit (24d2e4a50737) - tg3: use napi_complete_done()) Quoting verbatim: Using napi_complete_done() instead of napi_complete() allows us to use /sys/class/net/ethX/gro_flush_timeout GRO layer can aggregate more packets if the flush is delayed a bit, without having to set too big coalescing parameters that impact latencies. </end quote> Tested configuration: low latency via ethtool -C ethx adaptive-rx off rx-usecs 10 adaptive-tx off tx-usecs 15 workload: streaming rx using netperf TCP_MAERTS igb: MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.48 10^6bits/s over 1.000 seconds ending at 1440193171.589 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1176930056 1475.36 797726 16384.00 71905 MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.49 10^6bits/s over 0.997 seconds ending at 1440193142.763 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1175182320 50476.00 23282 16384.00 71816 i40e: Hard to test because the traffic is incoming so fast (24Gb/s) that GRO always receives 87kB, even at the highest interrupt rate. Other drivers were only compile tested. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-09-24 23:35:47 +00:00
static int igb_clean_rx_irq(struct igb_q_vector *q_vector, const int budget)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
struct igb_ring *rx_ring = q_vector->rx.ring;
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
struct sk_buff *skb = rx_ring->skb;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
unsigned int total_bytes = 0, total_packets = 0;
igb: Refactor clean_rx_irq to reduce overhead and improve performance This change is meant to be a general cleanup and performance improvement for clean_rx_irq. The previous patch should have updated the allocation so that the rings can be treated as read-only within the clean_rx_irq function. In addition I am re-ordering the operations such that several goals are accomplished including reducing the overhead for packet accounting, reducing the number of items on the stack, and improving overall performance. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:54 +00:00
u16 cleaned_count = igb_desc_unused(rx_ring);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Don't receive packets when the napi budget == 0 Processing any incoming packets with a with a napi budget of 0 is incorrect driver behavior. This matters as netpoll will shortly call drivers with a budget of 0 to avoid receive packet processing happening in hard irq context. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-15 01:00:06 +00:00
while (likely(total_packets < budget)) {
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
union e1000_adv_rx_desc *rx_desc;
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
struct igb_rx_buffer *rx_buffer;
unsigned int size;
igb: add page recycling support This patch adds support for page recycling by splitting the page into two usable portions and tracking the reference count. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:11:40 +00:00
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
/* return some buffers to hardware, one at a time is too slow */
if (cleaned_count >= IGB_RX_BUFFER_WRITE) {
igb_alloc_rx_buffers(rx_ring, cleaned_count);
cleaned_count = 0;
}
igb: add page recycling support This patch adds support for page recycling by splitting the page into two usable portions and tracking the reference count. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:11:40 +00:00
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
rx_desc = IGB_RX_DESC(rx_ring, rx_ring->next_to_clean);
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
size = le16_to_cpu(rx_desc->wb.upper.length);
if (!size)
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
break;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
/* This memory barrier is needed to keep us from reading
* any other fields out of the rx_desc until we know the
fm10k/igb/ixgbe: Use dma_rmb on Rx descriptor reads This change makes it so that dma_rmb is used when reading the Rx descriptor. The advantage of dma_rmb is that it allows for a much lower cost barrier on x86, powerpc, arm, and arm64 architectures than a traditional memory barrier when dealing with reads that only have to synchronize to coherent memory. In addition I have updated the code so that it just checks to see if any bits have been set instead of just the DD bit since the DD bit will always be set as a part of a descriptor write-back so we just need to check for a non-zero value being present at that memory location rather than just checking for any specific bit. This allows the code itself to appear much cleaner and allows the compiler more room to optimize. Cc: Matthew Vick <matthew.vick@intel.com> Cc: Don Skidmore <donald.c.skidmore@intel.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 23:02:28 +00:00
* descriptor has been written back
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
*/
fm10k/igb/ixgbe: Use dma_rmb on Rx descriptor reads This change makes it so that dma_rmb is used when reading the Rx descriptor. The advantage of dma_rmb is that it allows for a much lower cost barrier on x86, powerpc, arm, and arm64 architectures than a traditional memory barrier when dealing with reads that only have to synchronize to coherent memory. In addition I have updated the code so that it just checks to see if any bits have been set instead of just the DD bit since the DD bit will always be set as a part of a descriptor write-back so we just need to check for a non-zero value being present at that memory location rather than just checking for any specific bit. This allows the code itself to appear much cleaner and allows the compiler more room to optimize. Cc: Matthew Vick <matthew.vick@intel.com> Cc: Don Skidmore <donald.c.skidmore@intel.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 23:02:28 +00:00
dma_rmb();
igb: Support using build_skb in the case that jumbo frames are disabled This change makes it so that we can enable the use of build_skb for cases where jumbo frames are disabled. The advantage to this is that we do not have to perform a memcpy to populate the header and as a result we see a significant performance improvement. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-02 05:07:11 +00:00
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
rx_buffer = igb_get_rx_buffer(rx_ring, size);
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
/* retrieve a buffer from the ring */
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
if (skb)
igb_add_rx_frag(rx_ring, rx_buffer, skb, size);
igb: Re-add support for build_skb in igb This reverts commit f9d40f6a9921 ("igb: Revert support for build_skb in igb") and adds a few changes to update it to work with the latest version of igb. We are now able to revert the removal of this due to the fact that with the recent changes to the page count and the use of DMA_ATTR_SKIP_CPU_SYNC we can make the pages writable so we should not be invalidating the additional data added when we call build_skb. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:36 +00:00
else if (ring_uses_build_skb(rx_ring))
skb = igb_build_skb(rx_ring, rx_buffer, rx_desc, size);
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
else
skb = igb_construct_skb(rx_ring, rx_buffer,
rx_desc, size);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
/* exit if we failed to retrieve a buffer */
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
if (!skb) {
rx_ring->rx_stats.alloc_failed++;
rx_buffer->pagecnt_bias++;
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
break;
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
}
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
igb: Break out Rx buffer page management At this point we have 2 to 3 paths that can be taken depending on what Rx modes are enabled. In order to better support that and improve the maintainability I am breaking out the common bits from those paths and making them into their own functions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:26 +00:00
igb_put_rx_buffer(rx_ring, rx_buffer);
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
cleaned_count++;
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
/* fetch next buffer in frame if non-eop */
if (igb_is_non_eop(rx_ring, rx_desc))
continue;
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
/* verify the packet layout is correct */
if (igb_cleanup_headers(rx_ring, rx_desc, skb)) {
skb = NULL;
continue;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Combine post-processing of skb into a single function This change is meant to just clean-up a number of function calls that were made at the end of the Rx clean-up path by combining them into a single function call. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:57 +00:00
/* probably a little skewed due to removing CRC */
igb: leave staterr in place and instead us a helper function to check bits Instead of doing a byte swap on the staterr bits in the Rx descriptor we can save ourselves a bit of space and some CPU time by instead just testing for the various bits out of the Rx descriptor directly. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:03 +00:00
total_bytes += skb->len;
igb: Combine post-processing of skb into a single function This change is meant to just clean-up a number of function calls that were made at the end of the Rx clean-up path by combining them into a single function call. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:57 +00:00
/* populate checksum, timestamp, VLAN, and protocol */
igb_process_skb_fields(rx_ring, rx_desc, skb);
igb: leave staterr in place and instead us a helper function to check bits Instead of doing a byte swap on the staterr bits in the Rx descriptor we can save ourselves a bit of space and some CPU time by instead just testing for the various bits out of the Rx descriptor directly. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:03 +00:00
igb: do vlan cleanup - unify vlan and nonvlan rx path - kill adapter->vlgrp and igb_vlan_rx_register - allow to turn on/off rx/tx vlan accel via ethtool (set_features) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-21 03:27:27 +00:00
napi_gro_receive(&q_vector->napi, skb);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
/* reset skb pointer */
skb = NULL;
igb: Move rx_buffer related code in Rx cleanup path into separate function In order to try and isolate things a bit further I am moving the code related to retrieving data from the rx_buffer_info structure into a separate function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:07 +00:00
/* update budget accounting */
total_packets++;
igb: Don't receive packets when the napi budget == 0 Processing any incoming packets with a with a napi budget of 0 is incorrect driver behavior. This matters as netpoll will shortly call drivers with a budget of 0 to avoid receive packet processing happening in hard irq context. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-15 01:00:06 +00:00
}
igb: add page recycling support This patch adds support for page recycling by splitting the page into two usable portions and tracking the reference count. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:11:40 +00:00
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
/* place incomplete frames back on ring for completion */
rx_ring->skb = skb;
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
u64_stats_update_begin(&rx_ring->rx_syncp);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
rx_ring->rx_stats.packets += total_packets;
rx_ring->rx_stats.bytes += total_bytes;
igb: fix stats handling There are currently some problems with igb. - On 32bit arches, maintaining 64bit counters without proper synchronization between writers and readers. - Stats updated every two seconds, as reported by Jesper. (Jesper provided a patch for this) - Potential problem between worker thread and ethtool -S This patch uses u64_stats_sync, and convert everything to be 64bit safe, SMP safe, even on 32bit arches. It integrates Jesper idea of providing accurate stats at the time user reads them. Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-15 17:27:10 +00:00
u64_stats_update_end(&rx_ring->rx_syncp);
igb: Move ITR related data into work container within the q_vector This change moves information related to interrupt throttle rate configuration into a separate q_vector sub-structure called a work container. A similar change has already been made for ixgbe and this work is based off of that. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:45:47 +00:00
q_vector->rx.total_packets += total_packets;
q_vector->rx.total_bytes += total_bytes;
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
if (cleaned_count)
igb: drop the "adv" off function names relating to descriptors Many of the function names in the hot path are carrying an extra "_adv" suffix on the end of them to represent the fact that they are using advanced descriptors instead of legacy descriptors. However since all igb uses are advanced descriptors adding the extra suffix doesn't really add any additional data. Since this is the case it is easiest to just drop the suffix and save us from having to store the extra characters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:59 +00:00
igb_alloc_rx_buffers(rx_ring, cleaned_count);
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
drivers/net/intel: use napi_complete_done() As per Eric Dumazet's previous patches: (see commit (24d2e4a50737) - tg3: use napi_complete_done()) Quoting verbatim: Using napi_complete_done() instead of napi_complete() allows us to use /sys/class/net/ethX/gro_flush_timeout GRO layer can aggregate more packets if the flush is delayed a bit, without having to set too big coalescing parameters that impact latencies. </end quote> Tested configuration: low latency via ethtool -C ethx adaptive-rx off rx-usecs 10 adaptive-tx off tx-usecs 15 workload: streaming rx using netperf TCP_MAERTS igb: MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.48 10^6bits/s over 1.000 seconds ending at 1440193171.589 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1176930056 1475.36 797726 16384.00 71905 MIGRATED TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.0.0.1 () port 0 AF_INET : demo ... Interim result: 941.49 10^6bits/s over 0.997 seconds ending at 1440193142.763 Alignment Offset Bytes Bytes Recvs Bytes Sends Local Remote Local Remote Xfered Per Per Recv Send Recv Send Recv (avg) Send (avg) 8 8 0 0 1175182320 50476.00 23282 16384.00 71816 i40e: Hard to test because the traffic is incoming so fast (24Gb/s) that GRO always receives 87kB, even at the highest interrupt rate. Other drivers were only compile tested. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-09-24 23:35:47 +00:00
return total_packets;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Add support for padding packet With the size of the frame limited we can now write to an offset within the buffer instead of having to write at the very start of the buffer. The advantage to this is that it allows us to leave padding room for things like supporting XDP in the future. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:14 +00:00
static inline unsigned int igb_rx_offset(struct igb_ring *rx_ring)
{
return ring_uses_build_skb(rx_ring) ? IGB_SKB_PAD : 0;
}
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
static bool igb_alloc_mapped_page(struct igb_ring *rx_ring,
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
struct igb_rx_buffer *bi)
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
{
struct page *page = bi->page;
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
dma_addr_t dma;
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
/* since we are recycling buffers we should seldom need to alloc */
if (likely(page))
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
return true;
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
/* alloc new page for storage */
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
page = dev_alloc_pages(igb_rx_pg_order(rx_ring));
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
if (unlikely(!page)) {
rx_ring->rx_stats.alloc_failed++;
return false;
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
}
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
/* map page for use */
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
dma = dma_map_page_attrs(rx_ring->dev, page, 0,
igb_rx_pg_size(rx_ring),
DMA_FROM_DEVICE,
IGB_RX_DMA_ATTR);
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* if mapping failed free memory back to system since
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
* there isn't much point in holding memory we can't use
*/
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
if (dma_mapping_error(rx_ring->dev, dma)) {
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
__free_pages(page, igb_rx_pg_order(rx_ring));
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
rx_ring->rx_stats.alloc_failed++;
return false;
}
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
bi->dma = dma;
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
bi->page = page;
igb: Add support for padding packet With the size of the frame limited we can now write to an offset within the buffer instead of having to write at the very start of the buffer. The advantage to this is that it allows us to leave padding room for things like supporting XDP in the future. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:14 +00:00
bi->page_offset = igb_rx_offset(rx_ring);
igb: update code to better handle incrementing page count Update the driver code so that we do bulk updates of the page reference count instead of just incrementing it by one reference at a time. The advantage to doing this is that we cut down on atomic operations and this in turn should give us a slight improvement in cycles per packet. In addition if we eventually move this over to using build_skb the gains will be more noticeable. Link: http://lkml.kernel.org/r/20161110113616.76501.17072.stgit@ahduyck-blue-test.jf.intel.com Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Chris Metcalf <cmetcalf@mellanox.com> Cc: David Howells <dhowells@redhat.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Cc: Helge Deller <deller@gmx.de> Cc: James Hogan <james.hogan@imgtec.com> Cc: Jonas Bonn <jonas@southpole.se> Cc: Keguang Zhang <keguang.zhang@gmail.com> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Rich Felker <dalias@libc.org> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Steven Miao <realmz6@gmail.com> Cc: Tobias Klauser <tklauser@distanz.ch> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-14 23:05:34 +00:00
bi->pagecnt_bias = 1;
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
return true;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_alloc_rx_buffers - Replace used receive buffers; packet split
* @adapter: address of board private structure
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
**/
igb: drop the "adv" off function names relating to descriptors Many of the function names in the hot path are carrying an extra "_adv" suffix on the end of them to represent the fact that they are using advanced descriptors instead of legacy descriptors. However since all igb uses are advanced descriptors adding the extra suffix doesn't really add any additional data. Since this is the case it is easiest to just drop the suffix and save us from having to store the extra characters. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:59 +00:00
void igb_alloc_rx_buffers(struct igb_ring *rx_ring, u16 cleaned_count)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
union e1000_adv_rx_desc *rx_desc;
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
struct igb_rx_buffer *bi;
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
u16 i = rx_ring->next_to_use;
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
u16 bufsz;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
/* nothing to do */
if (!cleaned_count)
return;
igb: Replace E1000_XX_DESC_ADV with IGB_XX_DESC Since igb only uses advanced descriptors we might as well just use an IGB specific define and drop the _ADV suffix for the descriptor declarations. In addition this can be further reduced by assuming that it will be working on pointers since that is normally how the Tx descriptors are handled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:05 +00:00
rx_desc = IGB_RX_DESC(rx_ring, i);
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
bi = &rx_ring->rx_buffer_info[i];
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
i -= rx_ring->count;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
bufsz = igb_rx_bufsz(rx_ring);
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
do {
igb: Do not use header split, instead receive all frames into a single buffer This change makes it so that we no longer use header split. The idea is to reduce partial cache line writes by hardware when handling frames larger then header size. We can compensate for the extra overhead of having to memcpy the header buffer by avoiding the cache misses seen by leaving an full skb allocated and sitting on the ring. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:30:52 +00:00
if (!igb_alloc_mapped_page(rx_ring, bi))
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
break;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: update driver to make use of DMA_ATTR_SKIP_CPU_SYNC The ARM architecture provides a mechanism for deferring cache line invalidation in the case of map/unmap. This patch makes use of this mechanism to avoid unnecessary synchronization. A secondary effect of this change is that the portion of the page that has been synchronized for use by the CPU should be writable and could be passed up the stack (at least on ARM). The last bit that occurred to me is that on architectures where the sync_for_cpu call invalidates cache lines we were prefetching and then invalidating the first 128 bytes of the packet. To avoid that I have moved the sync up to before we perform the prefetch and allocate the skbuff so that we can actually make use of it. Link: http://lkml.kernel.org/r/20161110113611.76501.98897.stgit@ahduyck-blue-test.jf.intel.com Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Chris Metcalf <cmetcalf@mellanox.com> Cc: David Howells <dhowells@redhat.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Cc: Helge Deller <deller@gmx.de> Cc: James Hogan <james.hogan@imgtec.com> Cc: Jonas Bonn <jonas@southpole.se> Cc: Keguang Zhang <keguang.zhang@gmail.com> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Rich Felker <dalias@libc.org> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Steven Miao <realmz6@gmail.com> Cc: Tobias Klauser <tklauser@distanz.ch> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-14 23:05:30 +00:00
/* sync the buffer for use by the device */
dma_sync_single_range_for_device(rx_ring->dev, bi->dma,
igb: Add support for using order 1 pages to receive large frames This patch adds support for using 3K buffers in order 1 pages the same way we were using 2K buffers in 4K pages. We are reserving 1K of room for now to have space available for future headroom and tailroom when we enable build_skb support. One side effect of this patch is that we can end up using a larger buffer if jumbo frames is enabled. The impact shouldn't be too great, but it could hurt small packet performance for UDP workloads if jumbo frames is enabled as the truesize of frames will be larger. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:27:03 +00:00
bi->page_offset, bufsz,
igb: update driver to make use of DMA_ATTR_SKIP_CPU_SYNC The ARM architecture provides a mechanism for deferring cache line invalidation in the case of map/unmap. This patch makes use of this mechanism to avoid unnecessary synchronization. A secondary effect of this change is that the portion of the page that has been synchronized for use by the CPU should be writable and could be passed up the stack (at least on ARM). The last bit that occurred to me is that on architectures where the sync_for_cpu call invalidates cache lines we were prefetching and then invalidating the first 128 bytes of the packet. To avoid that I have moved the sync up to before we perform the prefetch and allocate the skbuff so that we can actually make use of it. Link: http://lkml.kernel.org/r/20161110113611.76501.98897.stgit@ahduyck-blue-test.jf.intel.com Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Chris Metcalf <cmetcalf@mellanox.com> Cc: David Howells <dhowells@redhat.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no> Cc: Helge Deller <deller@gmx.de> Cc: James Hogan <james.hogan@imgtec.com> Cc: Jonas Bonn <jonas@southpole.se> Cc: Keguang Zhang <keguang.zhang@gmail.com> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Rich Felker <dalias@libc.org> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Steven Miao <realmz6@gmail.com> Cc: Tobias Klauser <tklauser@distanz.ch> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-14 23:05:30 +00:00
DMA_FROM_DEVICE);
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Refresh the desc even if buffer_addrs didn't change
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
* because each write-back erases this info.
*/
igb: Revert support for build_skb in igb This patch actually reverts: igb: Support using build_skb in the case that jumbo frames are disabled The reason for reverting this patch is that it can lead to data corruption. The following flow was pointed out by Ben Hutchings: 1. skb is forwarded to another device 2. Packet headers are modified and it's put into a queue 3. Second packet is received into the other half of this page 4. Page cannot be reused, so is DMA-unmapped 5. The DMA mapping was non-coherent, so unmap copies or invalidates cache The headers added in step 2 get trashed in step 5. Reported-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-17 20:41:04 +00:00
rx_desc->read.pkt_addr = cpu_to_le64(bi->dma + bi->page_offset);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
rx_desc++;
bi++;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
i++;
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
if (unlikely(!i)) {
igb: Replace E1000_XX_DESC_ADV with IGB_XX_DESC Since igb only uses advanced descriptors we might as well just use an IGB specific define and drop the _ADV suffix for the descriptor declarations. In addition this can be further reduced by assuming that it will be working on pointers since that is normally how the Tx descriptors are handled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:05 +00:00
rx_desc = IGB_RX_DESC(rx_ring, 0);
igb: split buffer_info into tx_buffer_info and rx_buffer_info In order to be able to improve the performance of the TX path it has been necessary to add addition info to the tx_buffer_info structure. However a side effect is that the structure has gotten larger and this in turn has also increased the size of the RX buffer info structure. In order to avoid this in the future I am splitting the single buffer_info structure into two separate ones and instead I will join them by making the buffer_info pointer in the ring a union of the two. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:44:22 +00:00
bi = rx_ring->rx_buffer_info;
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
i -= rx_ring->count;
}
igb: Use length to determine if descriptor is done This change makes it so that we use the length of the packet instead of the DD status bit to determine if a new descriptor is ready to be processed. The obvious advantage is that it cuts down on reads as we don't really even need the DD bit if going from a 0 to a non-zero value on size is enough to inform us that the packet has been completed. In addition I have updated the code so that we only reset the Rx descriptor length for descriptor zero when resetting a ring instead of having to do a memset with 0 over the entire ring. By doing this we can save some time on initialization. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-02-07 02:25:41 +00:00
/* clear the length for the next_to_use descriptor */
rx_desc->wb.upper.length = 0;
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
cleaned_count--;
} while (cleaned_count);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: streamline Rx buffer allocation and cleanup This change is meant to streamline the Rx buffer allocation and cleanup. This is accomplished by reducing the number of writes by only having the Rx descriptor ring written by software during allocation, and it will only be read during cleanup. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:43:43 +00:00
i += rx_ring->count;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (rx_ring->next_to_use != i) {
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
/* record the next descriptor to use */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
rx_ring->next_to_use = i;
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
/* update next to alloc since we have filled the ring */
rx_ring->next_to_alloc = i;
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Force memory writes to complete before letting h/w
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
* know there are new descriptors to fetch. (Only
* applicable for weak-ordered memory model archs,
igb: Map entire page and sync half instead of mapping and unmapping half pages This change makes it so that we map the entire page and just sync half of it for the device at a time. The advantage to this approach is that we can avoid the locking on map/unmap seen in many IOMMU implementations. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-09-25 00:31:02 +00:00
* such as IA-64).
*/
igb: Use dma_wmb() instead of wmb() before doorbell writes igb writes to doorbells to post transmit and receive descriptors; after writing descriptors to memory but before writing to doorbells, use dma_wmb() rather than wmb(). wmb() is more heavyweight than necessary before doorbell writes. On x86, this avoids SFENCEs before doorbell writes in both the tx and rx refill paths. Signed-off-by: Venkatesh Srinivas <venkateshs@google.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-05-25 04:13:21 +00:00
dma_wmb();
igb: change the head and tail offsets into pointers Since we are writting to the head/tail pointers frequently we might as well save ourselves some processing time by converting the head and tail offsets directly to pointers. This will shave a few cycles off the rx/tx path and allows us to move one step closer to the rings being a bit more independant of each other. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:51:27 +00:00
writel(i, rx_ring->tail);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
}
/**
* igb_mii_ioctl -
* @netdev:
* @ifreq:
* @cmd:
**/
static int igb_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
{
struct igb_adapter *adapter = netdev_priv(netdev);
struct mii_ioctl_data *data = if_mii(ifr);
if (adapter->hw.phy.media_type != e1000_media_type_copper)
return -EOPNOTSUPP;
switch (cmd) {
case SIOCGMIIPHY:
data->phy_id = adapter->hw.phy.addr;
break;
case SIOCGMIIREG:
igb: do not use phy ops in ethtool test cleanup for non-copper parts Currently the igb driver is experiencing a panic due to a null function pointer being used during the cleanup of the ethtool looback test on fiber/serdes parts. This patch prevents that and adds a check prior to calling any phy function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-22 05:30:24 +00:00
if (igb_read_phy_reg(&adapter->hw, data->reg_num & 0x1F,
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
&data->val_out))
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return -EIO;
break;
case SIOCSMIIREG:
default:
return -EOPNOTSUPP;
}
return 0;
}
/**
* igb_ioctl -
* @netdev:
* @ifreq:
* @cmd:
**/
static int igb_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
{
switch (cmd) {
case SIOCGMIIPHY:
case SIOCGMIIREG:
case SIOCSMIIREG:
return igb_mii_ioctl(netdev, ifr, cmd);
igb: implement SIOCGHWTSTAMP ioctl This patch adds support for the SIOCGHWTSTAMP ioctl which enables user processes to read the current hwtstamp_config settings non-destructively. Previously a process had to be privileged and could only set values, it couldn't return what is currently set without possibly overwriting the value. This patch adds support for this new operation into igb by keeping a shadow copy of the config in the adapter structure, which is returned upon request. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Acked-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-01-11 07:20:06 +00:00
case SIOCGHWTSTAMP:
return igb_ptp_get_ts_config(netdev, ifr);
igb: stub support for SIOCSHWTSTAMP Signed-off-by: John Ronciak <john.ronciak@intel.com> Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-12 05:03:42 +00:00
case SIOCSHWTSTAMP:
igb: implement SIOCGHWTSTAMP ioctl This patch adds support for the SIOCGHWTSTAMP ioctl which enables user processes to read the current hwtstamp_config settings non-destructively. Previously a process had to be privileged and could only set values, it couldn't return what is currently set without possibly overwriting the value. This patch adds support for this new operation into igb by keeping a shadow copy of the config in the adapter structure, which is returned upon request. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Acked-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-01-11 07:20:06 +00:00
return igb_ptp_set_ts_config(netdev, ifr);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
default:
return -EOPNOTSUPP;
}
}
igb: Workaround for i210 Errata 25: Slow System Clock On some devices, the internal PLL circuit occasionally provides the wrong clock frequency after power up. The probability of failure is less than one failure per 1000 power cycles. When the failure occurs, the internal clock frequency is around 1/20 of the correct frequency. Cc: stable <stable@vger.kernel.org> Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-10 08:47:15 +00:00
void igb_read_pci_cfg(struct e1000_hw *hw, u32 reg, u16 *value)
{
struct igb_adapter *adapter = hw->back;
pci_read_config_word(adapter->pdev, reg, value);
}
void igb_write_pci_cfg(struct e1000_hw *hw, u32 reg, u16 *value)
{
struct igb_adapter *adapter = hw->back;
pci_write_config_word(adapter->pdev, reg, *value);
}
igb: add completion timeout workaround for 82575/82576 The 82575 and 82576 hardware can both experience data corruption issues if a pci-e completion arrives after the timeout value. In order to avoid this we need to increase the timeout value while pci-e master is disabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:08:35 +00:00
s32 igb_read_pcie_cap_reg(struct e1000_hw *hw, u32 reg, u16 *value)
{
struct igb_adapter *adapter = hw->back;
igb: Use PCI Express Capability accessors Use PCI Express Capability access functions to simplify igb driver. [bhelgaas: split e1000e and igb into separate patches] Signed-off-by: Jiang Liu <jiang.liu@huawei.com> Signed-off-by: Yijing Wang <wangyijing@huawei.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-20 19:32:20 +00:00
if (pcie_capability_read_word(adapter->pdev, reg, value))
igb: add completion timeout workaround for 82575/82576 The 82575 and 82576 hardware can both experience data corruption issues if a pci-e completion arrives after the timeout value. In order to avoid this we need to increase the timeout value while pci-e master is disabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:08:35 +00:00
return -E1000_ERR_CONFIG;
return 0;
}
s32 igb_write_pcie_cap_reg(struct e1000_hw *hw, u32 reg, u16 *value)
{
struct igb_adapter *adapter = hw->back;
igb: Use PCI Express Capability accessors Use PCI Express Capability access functions to simplify igb driver. [bhelgaas: split e1000e and igb into separate patches] Signed-off-by: Jiang Liu <jiang.liu@huawei.com> Signed-off-by: Yijing Wang <wangyijing@huawei.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-08-20 19:32:20 +00:00
if (pcie_capability_write_word(adapter->pdev, reg, *value))
igb: add completion timeout workaround for 82575/82576 The 82575 and 82576 hardware can both experience data corruption issues if a pci-e completion arrives after the timeout value. In order to avoid this we need to increase the timeout value while pci-e master is disabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-23 18:08:35 +00:00
return -E1000_ERR_CONFIG;
return 0;
}
net: introduce and use netdev_features_t for device features sets v2: add couple missing conversions in drivers split unexporting netdev_fix_features() implemented %pNF convert sock::sk_route_(no?)caps Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-15 15:29:55 +00:00
static void igb_vlan_mode(struct net_device *netdev, netdev_features_t features)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
u32 ctrl, rctl;
net: vlan: rename NETIF_F_HW_VLAN_* feature flags to NETIF_F_HW_VLAN_CTAG_* Rename the hardware VLAN acceleration features to include "CTAG" to indicate that they only support CTAGs. Follow up patches will introduce 802.1ad server provider tagging (STAGs) and require the distinction for hardware not supporting acclerating both. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-19 02:04:27 +00:00
bool enable = !!(features & NETIF_F_HW_VLAN_CTAG_RX);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: fix recent VLAN changes that would leave VLANs disabled after reset This patch cleans up several issues with VLANs on igb after the recent changes that were meant to leave the VLANs enabled/disable via the netdev->features flags. Specifically the Rx VLAN settings were being dropped after reset due to the fact that they were not being restored correctly. In addition I removed the IRQ disable/enable since those were in place to protect the setting of vlgrp. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:08 +00:00
if (enable) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* enable VLAN tag insert/strip */
ctrl = rd32(E1000_CTRL);
ctrl |= E1000_CTRL_VME;
wr32(E1000_CTRL, ctrl);
igb: rework handling of the vfta and vlvf registers in relation to mng_vlan This patch corrects some errors in how vlans are being handled when vfs start interacting with the management vlans. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:35 +00:00
/* Disable CFI check */
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
rctl = rd32(E1000_RCTL);
rctl &= ~E1000_RCTL_CFIEN;
wr32(E1000_RCTL, rctl);
} else {
/* disable VLAN tag insert/strip */
ctrl = rd32(E1000_CTRL);
ctrl &= ~E1000_CTRL_VME;
wr32(E1000_CTRL, ctrl);
}
igb: Fix VLAN tag stripping on Intel i350 Problem: When switching off VLAN offloading on an i350, the VLAN interface gets unusable. For testing, set up a VLAN on an i350 and some remote machine, e.g.: $ ip link add link eth0 name eth0.42 type vlan id 42 $ ip addr add 192.168.42.1/24 dev eth0.42 $ ip link set dev eth0.42 up Offloading is switched on by default: $ ethtool -k eth0 | grep vlan-offload rx-vlan-offload: on tx-vlan-offload: on $ ping -c 3 -I eth0.42 192.168.42.2 [...works as usual...] Now switch off VLAN offloading and try again: $ ethtool -K eth0 rxvlan off Actual changes: rx-vlan-offload: off tx-vlan-offload: off [requested on] $ ping -c 3 -I eth0.42 192.168.42.2 PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da ta. --- 192.168.42.2 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 1999ms I can only reproduce it on an i350, the above works fine on a 82580. While inspecting the igb source, I came across the code in igb_set_vmolr which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and for all, and in all of the igb code there's no other place where the STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb unconditionally, independently of the offloading setting. I compared that to the latest Intel igb-5.3.3.5 driver from http://sourceforge.net/projects/e1000/ which in fact sets and clears the STRVLAN flag independently from igb_set_vmolr in its own function igb_set_vf_vlan_strip, depending on the vlan settings. So I included the STRVLAN handling from the igb-5.3.3.5 driver into our current igb driver and tested the above scenario again. This time ping still works after switching off VLAN offloading. Tested on i350, with and without addtional VFs, as well as on 82580 successfully. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-28 12:53:23 +00:00
igb_set_vf_vlan_strip(adapter, adapter->vfs_allocated_count, enable);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
net: vlan: prepare for 802.1ad VLAN filtering offload Change the rx_{add,kill}_vid callbacks to take a protocol argument in preparation of 802.1ad support. The protocol argument used so far is always htons(ETH_P_8021Q). Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-19 02:04:28 +00:00
static int igb_vlan_rx_add_vid(struct net_device *netdev,
__be16 proto, u16 vid)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
int pf_id = adapter->vfs_allocated_count;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: rework handling of the vfta and vlvf registers in relation to mng_vlan This patch corrects some errors in how vlans are being handled when vfs start interacting with the management vlans. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:47:35 +00:00
/* add the filter since PF can receive vlans w/o entry in vlvf */
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
if (!vid || !(adapter->flags & IGB_FLAG_VLAN_PROMISC))
igb_vfta_set(hw, vid, pf_id, true, !!vid);
igb: do vlan cleanup - unify vlan and nonvlan rx path - kill adapter->vlgrp and igb_vlan_rx_register - allow to turn on/off rx/tx vlan accel via ethtool (set_features) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-21 03:27:27 +00:00
set_bit(vid, adapter->active_vlans);
net: make vlan ndo_vlan_rx_[add/kill]_vid return error value Let caller know the result of adding/removing vlan id to/from vlan filter. In some drivers I make those functions to just return 0. But in those where there is able to see if hw setup went correctly, return value is set appropriately. Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-09 00:52:37 +00:00
return 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
net: vlan: prepare for 802.1ad VLAN filtering offload Change the rx_{add,kill}_vid callbacks to take a protocol argument in preparation of 802.1ad support. The protocol argument used so far is always htons(ETH_P_8021Q). Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-19 02:04:28 +00:00
static int igb_vlan_rx_kill_vid(struct net_device *netdev,
__be16 proto, u16 vid)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct igb_adapter *adapter = netdev_priv(netdev);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
int pf_id = adapter->vfs_allocated_count;
igb: Merge VLVF configuration into igb_vfta_set This change makes it so that we can merge the configuration of the VLVF registers into the setting of the VFTA register. By doing this we simplify the logic and make use of similar functionality that we have already added for ixgbe making it easier to maintain both drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:04 +00:00
struct e1000_hw *hw = &adapter->hw;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Merge VLVF configuration into igb_vfta_set This change makes it so that we can merge the configuration of the VLVF registers into the setting of the VFTA register. By doing this we simplify the logic and make use of similar functionality that we have already added for ixgbe making it easier to maintain both drivers. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:04 +00:00
/* remove VID from filter table */
igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE This change fixes things so that we can fully support SR-IOV or the recently added NTUPLE filtering while allowing support for VLAN promiscuous mode. By making this change we are able to support possible scenarios such as SR-IOV with the PF connected to a Linux bridge hosting other VMs. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:11:18 +00:00
if (vid && !(adapter->flags & IGB_FLAG_VLAN_PROMISC))
igb_vfta_set(hw, vid, pf_id, false, true);
igb: do vlan cleanup - unify vlan and nonvlan rx path - kill adapter->vlgrp and igb_vlan_rx_register - allow to turn on/off rx/tx vlan accel via ethtool (set_features) Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-21 03:27:27 +00:00
clear_bit(vid, adapter->active_vlans);
net: make vlan ndo_vlan_rx_[add/kill]_vid return error value Let caller know the result of adding/removing vlan id to/from vlan filter. In some drivers I make those functions to just return 0. But in those where there is able to see if hw setup went correctly, return value is set appropriately. Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-09 00:52:37 +00:00
return 0;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
static void igb_restore_vlan(struct igb_adapter *adapter)
{
igb: Always enable VLAN 0 even if 8021q is not loaded This patch makes it so that we always add VLAN 0. This is important as we need to guarantee the PF can receive untagged frames in the case of SR-IOV being enabled but VLAN filtering not being enabled in the kernel. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:54 +00:00
u16 vid = 1;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: fix recent VLAN changes that would leave VLANs disabled after reset This patch cleans up several issues with VLANs on igb after the recent changes that were meant to leave the VLANs enabled/disable via the netdev->features flags. Specifically the Rx VLAN settings were being dropped after reset due to the fact that they were not being restored correctly. In addition I removed the IRQ disable/enable since those were in place to protect the setting of vlgrp. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:08 +00:00
igb_vlan_mode(adapter->netdev, adapter->netdev->features);
igb: Always enable VLAN 0 even if 8021q is not loaded This patch makes it so that we always add VLAN 0. This is important as we need to guarantee the PF can receive untagged frames in the case of SR-IOV being enabled but VLAN filtering not being enabled in the kernel. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:54 +00:00
igb_vlan_rx_add_vid(adapter->netdev, htons(ETH_P_8021Q), 0);
igb: fix recent VLAN changes that would leave VLANs disabled after reset This patch cleans up several issues with VLANs on igb after the recent changes that were meant to leave the VLANs enabled/disable via the netdev->features flags. Specifically the Rx VLAN settings were being dropped after reset due to the fact that they were not being restored correctly. In addition I removed the IRQ disable/enable since those were in place to protect the setting of vlgrp. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-08-26 07:46:08 +00:00
igb: Always enable VLAN 0 even if 8021q is not loaded This patch makes it so that we always add VLAN 0. This is important as we need to guarantee the PF can receive untagged frames in the case of SR-IOV being enabled but VLAN filtering not being enabled in the kernel. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:54 +00:00
for_each_set_bit_from(vid, adapter->active_vlans, VLAN_N_VID)
net: vlan: prepare for 802.1ad VLAN filtering offload Change the rx_{add,kill}_vid callbacks to take a protocol argument in preparation of 802.1ad support. The protocol argument used so far is always htons(ETH_P_8021Q). Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-19 02:04:28 +00:00
igb_vlan_rx_add_vid(adapter->netdev, htons(ETH_P_8021Q), vid);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
net/igb/e1000/e1000e: more robust ethtool duplex/speed configuration This makes sure that one cannot request a 99Mbps full-duplex and get a 100Mbps half-duplex configuration in return due to the way the speed/duplex parameters are handled internally. Tested: e1000 works Signed-off-by: David Decotigny <decot@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-04-27 18:32:43 +00:00
int igb_set_spd_dplx(struct igb_adapter *adapter, u32 spd, u8 dplx)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: add pci_dev in few spots to clean up use of dev_err/info/warn This patch relpaces several references to adapter->pdev->dev with just pdev->dev. This allows for cleanup of several multiline dev_err/info calls. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:55 +00:00
struct pci_dev *pdev = adapter->pdev;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
struct e1000_mac_info *mac = &adapter->hw.mac;
mac->autoneg = 0;
net/igb/e1000/e1000e: more robust ethtool duplex/speed configuration This makes sure that one cannot request a 99Mbps full-duplex and get a 100Mbps half-duplex configuration in return due to the way the speed/duplex parameters are handled internally. Tested: e1000 works Signed-off-by: David Decotigny <decot@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-04-27 18:32:43 +00:00
/* Make sure dplx is at most 1 bit and lsb of speed is not set
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* for the switch() below to work
*/
net/igb/e1000/e1000e: more robust ethtool duplex/speed configuration This makes sure that one cannot request a 99Mbps full-duplex and get a 100Mbps half-duplex configuration in return due to the way the speed/duplex parameters are handled internally. Tested: e1000 works Signed-off-by: David Decotigny <decot@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-04-27 18:32:43 +00:00
if ((spd & 1) || (dplx & ~1))
goto err_inval;
igb: Support for 100base-fx SFP This patch adds support for 100base-fx SFP and report proper link speed/duplex via Ethtool. v2: fix smatch warnings CC: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-05 16:49:06 +00:00
/* Fiber NIC's only allow 1000 gbps Full duplex
* and 100Mbps Full duplex for 100baseFx sfp
*/
if (adapter->hw.phy.media_type == e1000_media_type_internal_serdes) {
switch (spd + dplx) {
case SPEED_10 + DUPLEX_HALF:
case SPEED_10 + DUPLEX_FULL:
case SPEED_100 + DUPLEX_HALF:
goto err_inval;
default:
break;
}
}
igb: add check for fiber/serdes devices to igb_set_spd_dplx; Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Emil Tantilov <emil.s.tantilov@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-10-12 22:27:02 +00:00
net/igb/e1000/e1000e: more robust ethtool duplex/speed configuration This makes sure that one cannot request a 99Mbps full-duplex and get a 100Mbps half-duplex configuration in return due to the way the speed/duplex parameters are handled internally. Tested: e1000 works Signed-off-by: David Decotigny <decot@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-04-27 18:32:43 +00:00
switch (spd + dplx) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
case SPEED_10 + DUPLEX_HALF:
mac->forced_speed_duplex = ADVERTISE_10_HALF;
break;
case SPEED_10 + DUPLEX_FULL:
mac->forced_speed_duplex = ADVERTISE_10_FULL;
break;
case SPEED_100 + DUPLEX_HALF:
mac->forced_speed_duplex = ADVERTISE_100_HALF;
break;
case SPEED_100 + DUPLEX_FULL:
mac->forced_speed_duplex = ADVERTISE_100_FULL;
break;
case SPEED_1000 + DUPLEX_FULL:
mac->autoneg = 1;
adapter->hw.phy.autoneg_advertised = ADVERTISE_1000_FULL;
break;
case SPEED_1000 + DUPLEX_HALF: /* not supported */
default:
net/igb/e1000/e1000e: more robust ethtool duplex/speed configuration This makes sure that one cannot request a 99Mbps full-duplex and get a 100Mbps half-duplex configuration in return due to the way the speed/duplex parameters are handled internally. Tested: e1000 works Signed-off-by: David Decotigny <decot@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-04-27 18:32:43 +00:00
goto err_inval;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: update to allow reading/setting MDI state This is the implementation for igb to allow forcing MDI state via ethtool, allowing users to work around some improperly behaving switches. Forcing in this driver is for now only allowed when auto-neg is enabled. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> CC: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown aaron.f.brown@intel.com Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-07-26 02:31:19 +00:00
/* clear MDI, MDI(-X) override is only allowed when autoneg enabled */
adapter->hw.phy.mdix = AUTO_ALL_MODES;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return 0;
net/igb/e1000/e1000e: more robust ethtool duplex/speed configuration This makes sure that one cannot request a 99Mbps full-duplex and get a 100Mbps half-duplex configuration in return due to the way the speed/duplex parameters are handled internally. Tested: e1000 works Signed-off-by: David Decotigny <decot@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-04-27 18:32:43 +00:00
err_inval:
dev_err(&pdev->dev, "Unsupported Speed/Duplex configuration\n");
return -EINVAL;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
static int __igb_shutdown(struct pci_dev *pdev, bool *enable_wake,
bool runtime)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
struct net_device *netdev = pci_get_drvdata(pdev);
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
igb: add 82576 MAC support Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:10:12 +00:00
u32 ctrl, rctl, status;
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
u32 wufc = runtime ? E1000_WUFC_LNKC : adapter->wol;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#ifdef CONFIG_PM
int retval = 0;
#endif
igb: close/suspend race in netif_device_detach Similar to ixgbe, when an interface is part of a namespace it is possible that igb_close() may be called while __igb_shutdown() is running which ends up in a double free WARN and/or a BUG in free_msi_irqs(). Extend the rtnl_lock() to protect the call to netif_device_detach() and igb_clear_interrupt_scheme() in __igb_shutdown() and check for netif_device_present() to avoid calling igb_clear_interrupt_scheme() a second time in igb_close(). Also extend the rtnl lock in igb_resume() to netif_device_attach(). Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Acked-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-11-15 16:54:26 +00:00
rtnl_lock();
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
netif_device_detach(netdev);
igb: update suspend resume Updates the suspend and resume to better handle the possibility of MSIX vector changes. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:13:38 +00:00
if (netif_running(netdev))
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
__igb_close(netdev, true);
igb: update suspend resume Updates the suspend and resume to better handle the possibility of MSIX vector changes. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:13:38 +00:00
igb: call igb_ptp_suspend during suspend/resume cycle Properly stop the extra workqueue items and ensure that we resume cleanly. This is better than using igb_ptp_init and igb_ptp_stop since these functions destroy the PHC device, which will cause other problems if we do so. Since igb_ptp_reset now re-schedules the work-queue item we don't need an equivalent igb_ptp_resume in the resume workflow. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-05-24 20:56:31 +00:00
igb_ptp_suspend(adapter);
igb: add new data structure for handling interrupts and NAPI Add a new igb_q_vector data structure to handle interrupts and NAPI. This helps to abstract the rings away from the adapter struct. In addition it allows for a bit of consolidation since a tx and rx ring can share a q_vector. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:49:27 +00:00
igb_clear_interrupt_scheme(adapter);
igb: close/suspend race in netif_device_detach Similar to ixgbe, when an interface is part of a namespace it is possible that igb_close() may be called while __igb_shutdown() is running which ends up in a double free WARN and/or a BUG in free_msi_irqs(). Extend the rtnl_lock() to protect the call to netif_device_detach() and igb_clear_interrupt_scheme() in __igb_shutdown() and check for netif_device_present() to avoid calling igb_clear_interrupt_scheme() a second time in igb_close(). Also extend the rtnl lock in igb_resume() to netif_device_attach(). Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Acked-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-11-15 16:54:26 +00:00
rtnl_unlock();
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#ifdef CONFIG_PM
igb: Fix an issue that PME is not enabled during runtime suspend I210 ethernet card doesn't wakeup when a cable gets plugged. It's because its PME is not set. Since commit 42eca2302146 ("PCI: Don't touch card regs after runtime suspend D3"), if the PCI state is saved, pci_pm_runtime_suspend() stops calling pci_finish_runtime_suspend(), which enables the PCI PME. To fix the issue, let's not to save PCI states when it's runtime suspend, to let the PCI subsystem enables PME. Fixes: 42eca2302146 ("PCI: Don't touch card regs after runtime suspend D3") Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-12-03 05:54:38 +00:00
if (!runtime) {
retval = pci_save_state(pdev);
if (retval)
return retval;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
#endif
status = rd32(E1000_STATUS);
if (status & E1000_STATUS_LU)
wufc &= ~E1000_WUFC_LNKC;
if (wufc) {
igb_setup_rctl(adapter);
igb: add support for set_rx_mode netdevice operation This patch adds support for the set_rx_mode netdevice operation so that igb can better support multiple unicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:48:56 +00:00
igb_set_rx_mode(netdev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* turn on all-multi mode if wake on multicast is enabled */
if (wufc & E1000_WUFC_MC) {
rctl = rd32(E1000_RCTL);
rctl |= E1000_RCTL_MPE;
wr32(E1000_RCTL, rctl);
}
ctrl = rd32(E1000_CTRL);
/* advertise wake from D3Cold */
#define E1000_CTRL_ADVD3WUC 0x00100000
/* phy power management enable */
#define E1000_CTRL_EN_PHY_PWR_MGMT 0x00200000
ctrl |= E1000_CTRL_ADVD3WUC;
wr32(E1000_CTRL, ctrl);
/* Allow time for pending master requests to run */
igb: replace unecessary &adapter->hw with just hw where applicable This patch just cleans up some unecessary references to the adapter->hw member when it has already been placed in a local variable named hw. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 23:51:35 +00:00
igb_disable_pcie_master(hw);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
wr32(E1000_WUC, E1000_WUC_PME_EN);
wr32(E1000_WUFC, wufc);
} else {
wr32(E1000_WUC, 0);
wr32(E1000_WUFC, 0);
}
net/igb: Fix kexec with igb (rev. 3) Impact: Fix Yinghai Lu found one system with 82575EB where, in the kernel that is kexeced, probe igb failed with -2, the reason being that the adapter could not be brought back from D3 by the kexec kernel, most probably due to quirky hardware (it looks like the same behavior happened on forcedeth). Prevent igb from putting the adapter into D3 during shutdown except when we going to power off the system. For this purpose, seperate igb_shutdown() from igb_suspend() and use the appropriate PCI PM callbacks in both of them. Signed-off-by: "Rafael J. Wysocki" <rjw@sisk.pl> Reported-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-31 21:23:50 +00:00
*enable_wake = wufc || adapter->en_mng_pt;
if (!*enable_wake)
igb: Power down link when interface is down This changes the behavior of the driver to power down the link when the associated interface is down, unless management is enabled. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:01:59 +00:00
igb_power_down_link(adapter);
else
igb_power_up_link(adapter);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* Release control of h/w to f/w. If f/w is AMT enabled, this
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* would have already happened in close and is redundant.
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_release_hw_control(adapter);
pci_disable_device(pdev);
return 0;
}
igb: Enable reading of wake up packet Currently, in igb_resume(), igb driver ignores the Wake Up Status (WUS) and Wake Up Packet Memory (WUPM) registers. This patch enables the igb driver to read the WUPM if the controller was woken by a wake up packet that is not more than 128 bytes long (maximum WUPM size), then pass it up the kernel network stack. Signed-off-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-27 00:44:35 +00:00
static void igb_deliver_wake_packet(struct net_device *netdev)
{
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
struct sk_buff *skb;
u32 wupl;
wupl = rd32(E1000_WUPL) & E1000_WUPL_MASK;
/* WUPM stores only the first 128 bytes of the wake packet.
* Read the packet only if we have the whole thing.
*/
if ((wupl == 0) || (wupl > E1000_WUPM_BYTES))
return;
skb = netdev_alloc_skb_ip_align(netdev, E1000_WUPM_BYTES);
if (!skb)
return;
skb_put(skb, wupl);
/* Ensure reads are 32-bit aligned */
wupl = roundup(wupl, 4);
memcpy_fromio(skb->data, hw->hw_addr + E1000_WUPM_REG(0), wupl);
skb->protocol = eth_type_trans(skb, netdev);
netif_rx(skb);
}
igb: mark PM functions as __maybe_unused The new wake function is only used by the suspend/resume handlers that are defined in inside of an #ifdef, which can cause this harmless warning: drivers/net/ethernet/intel/igb/igb_main.c:7988:13: warning: 'igb_deliver_wake_packet' defined but not used [-Wunused-function] Removing the #ifdef, instead using a __maybe_unused annotation simplifies the code and avoids the warning. Fixes: b90fa8763560 ("igb: Enable reading of wake up packet") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-04-27 19:09:52 +00:00
static int __maybe_unused igb_suspend(struct device *dev)
net/igb: Fix kexec with igb (rev. 3) Impact: Fix Yinghai Lu found one system with 82575EB where, in the kernel that is kexeced, probe igb failed with -2, the reason being that the adapter could not be brought back from D3 by the kexec kernel, most probably due to quirky hardware (it looks like the same behavior happened on forcedeth). Prevent igb from putting the adapter into D3 during shutdown except when we going to power off the system. For this purpose, seperate igb_shutdown() from igb_suspend() and use the appropriate PCI PM callbacks in both of them. Signed-off-by: "Rafael J. Wysocki" <rjw@sisk.pl> Reported-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-31 21:23:50 +00:00
{
int retval;
bool wake;
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
struct pci_dev *pdev = to_pci_dev(dev);
net/igb: Fix kexec with igb (rev. 3) Impact: Fix Yinghai Lu found one system with 82575EB where, in the kernel that is kexeced, probe igb failed with -2, the reason being that the adapter could not be brought back from D3 by the kexec kernel, most probably due to quirky hardware (it looks like the same behavior happened on forcedeth). Prevent igb from putting the adapter into D3 during shutdown except when we going to power off the system. For this purpose, seperate igb_shutdown() from igb_suspend() and use the appropriate PCI PM callbacks in both of them. Signed-off-by: "Rafael J. Wysocki" <rjw@sisk.pl> Reported-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-31 21:23:50 +00:00
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
retval = __igb_shutdown(pdev, &wake, 0);
net/igb: Fix kexec with igb (rev. 3) Impact: Fix Yinghai Lu found one system with 82575EB where, in the kernel that is kexeced, probe igb failed with -2, the reason being that the adapter could not be brought back from D3 by the kexec kernel, most probably due to quirky hardware (it looks like the same behavior happened on forcedeth). Prevent igb from putting the adapter into D3 during shutdown except when we going to power off the system. For this purpose, seperate igb_shutdown() from igb_suspend() and use the appropriate PCI PM callbacks in both of them. Signed-off-by: "Rafael J. Wysocki" <rjw@sisk.pl> Reported-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-31 21:23:50 +00:00
if (retval)
return retval;
if (wake) {
pci_prepare_to_sleep(pdev);
} else {
pci_wake_from_d3(pdev, false);
pci_set_power_state(pdev, PCI_D3hot);
}
return 0;
}
igb: mark PM functions as __maybe_unused The new wake function is only used by the suspend/resume handlers that are defined in inside of an #ifdef, which can cause this harmless warning: drivers/net/ethernet/intel/igb/igb_main.c:7988:13: warning: 'igb_deliver_wake_packet' defined but not used [-Wunused-function] Removing the #ifdef, instead using a __maybe_unused annotation simplifies the code and avoids the warning. Fixes: b90fa8763560 ("igb: Enable reading of wake up packet") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-04-27 19:09:52 +00:00
static int __maybe_unused igb_resume(struct device *dev)
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
{
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
struct pci_dev *pdev = to_pci_dev(dev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
struct net_device *netdev = pci_get_drvdata(pdev);
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
igb: Enable reading of wake up packet Currently, in igb_resume(), igb driver ignores the Wake Up Status (WUS) and Wake Up Packet Memory (WUPM) registers. This patch enables the igb driver to read the WUPM if the controller was woken by a wake up packet that is not more than 128 bytes long (maximum WUPM size), then pass it up the kernel network stack. Signed-off-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-27 00:44:35 +00:00
u32 err, val;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
pci_set_power_state(pdev, PCI_D0);
pci_restore_state(pdev);
igb: call pci_save_state after pci_restore_state This patch adds calls to pci_save_state() immediately after calls to pci_restore_state(). Due to a change in the behavior of pci_restore_state() it is necessary to call pci_save_state() to keep the state_saved flag. This patch is based on a similar patch for ixgbe. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:19 +00:00
pci_save_state(pdev);
igb: make ioport free This patch makes igb driver ioport-free. This corrects behavior in probe function so as not to request ioport resources as long as they are not really needed. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-06-20 03:10:30 +00:00
igb: Fixes needed for surprise removal support This patch adds some checks in order to prevent panic's on surprise removal of devices during S0, S3, S4. Without this patch, Thunderbolt type device removal will panic the system. Signed-off-by: Yanir Lubetkin <yanirx.lubetkin@intel.com> Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-22 07:52:54 +00:00
if (!pci_device_is_present(pdev))
return -ENODEV;
igb: remove check for needing an io port Since igb supports only pci-e nics and there is no plan to support any legacy pci parts in the driver there isn't really much need for checking to see if an io port is needed. In the unlikely event that we do begin supporting legacy pci parts then we can see about adding this code back to the driver. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:16:04 +00:00
err = pci_enable_device_mem(pdev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (err) {
dev_err(&pdev->dev,
"igb: Cannot enable PCI device from suspend\n");
return err;
}
pci_set_master(pdev);
pci_enable_wake(pdev, PCI_D3hot, 0);
pci_enable_wake(pdev, PCI_D3cold, 0);
igb: remove duplicate code for fallback interrupt initialization Given a small change to igb_init_interrupt_scheme() the function fits igb_request_irq() for MSI/legacy interrupts initialization as well, instead of duplicating most of its code there. Also adding a missing igb_configure() to igb_request_irq() for MSI fallback to work properly. Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-04 06:00:12 +00:00
if (igb_init_interrupt_scheme(adapter, true)) {
igb: update suspend resume Updates the suspend and resume to better handle the possibility of MSIX vector changes. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2008-07-08 22:13:38 +00:00
dev_err(&pdev->dev, "Unable to allocate memory for queues\n");
return -ENOMEM;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb_reset(adapter);
igb: move get_hw_control within igb_resume. Move igb_get_hw_control up so that it is called just after the reset in igb_resume. This notifies the HW sooner that the driver is reassuming control of the device. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:21:10 +00:00
/* let the f/w know that the h/w is now under the control of the
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* driver.
*/
igb: move get_hw_control within igb_resume. Move igb_get_hw_control up so that it is called just after the reset in igb_resume. This notifies the HW sooner that the driver is reassuming control of the device. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:21:10 +00:00
igb_get_hw_control(adapter);
igb: Enable reading of wake up packet Currently, in igb_resume(), igb driver ignores the Wake Up Status (WUS) and Wake Up Packet Memory (WUPM) registers. This patch enables the igb driver to read the WUPM if the controller was woken by a wake up packet that is not more than 128 bytes long (maximum WUPM size), then pass it up the kernel network stack. Signed-off-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-27 00:44:35 +00:00
val = rd32(E1000_WUS);
if (val & WAKE_PKT_WUS)
igb_deliver_wake_packet(netdev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
wr32(E1000_WUS, ~0);
igb: close/suspend race in netif_device_detach Similar to ixgbe, when an interface is part of a namespace it is possible that igb_close() may be called while __igb_shutdown() is running which ends up in a double free WARN and/or a BUG in free_msi_irqs(). Extend the rtnl_lock() to protect the call to netif_device_detach() and igb_clear_interrupt_scheme() in __igb_shutdown() and check for netif_device_present() to avoid calling igb_clear_interrupt_scheme() a second time in igb_close(). Also extend the rtnl lock in igb_resume() to netif_device_attach(). Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Acked-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-11-15 16:54:26 +00:00
rtnl_lock();
if (!err && netif_running(netdev))
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
err = __igb_open(netdev, true);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: close/suspend race in netif_device_detach Similar to ixgbe, when an interface is part of a namespace it is possible that igb_close() may be called while __igb_shutdown() is running which ends up in a double free WARN and/or a BUG in free_msi_irqs(). Extend the rtnl_lock() to protect the call to netif_device_detach() and igb_clear_interrupt_scheme() in __igb_shutdown() and check for netif_device_present() to avoid calling igb_clear_interrupt_scheme() a second time in igb_close(). Also extend the rtnl lock in igb_resume() to netif_device_attach(). Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Acked-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-11-15 16:54:26 +00:00
if (!err)
netif_device_attach(netdev);
rtnl_unlock();
return err;
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
}
igb: mark PM functions as __maybe_unused The new wake function is only used by the suspend/resume handlers that are defined in inside of an #ifdef, which can cause this harmless warning: drivers/net/ethernet/intel/igb/igb_main.c:7988:13: warning: 'igb_deliver_wake_packet' defined but not used [-Wunused-function] Removing the #ifdef, instead using a __maybe_unused annotation simplifies the code and avoids the warning. Fixes: b90fa8763560 ("igb: Enable reading of wake up packet") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-04-27 19:09:52 +00:00
static int __maybe_unused igb_runtime_idle(struct device *dev)
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
{
struct pci_dev *pdev = to_pci_dev(dev);
struct net_device *netdev = pci_get_drvdata(pdev);
struct igb_adapter *adapter = netdev_priv(netdev);
if (!igb_has_link(adapter))
pm_schedule_suspend(dev, MSEC_PER_SEC * 5);
return -EBUSY;
}
igb: mark PM functions as __maybe_unused The new wake function is only used by the suspend/resume handlers that are defined in inside of an #ifdef, which can cause this harmless warning: drivers/net/ethernet/intel/igb/igb_main.c:7988:13: warning: 'igb_deliver_wake_packet' defined but not used [-Wunused-function] Removing the #ifdef, instead using a __maybe_unused annotation simplifies the code and avoids the warning. Fixes: b90fa8763560 ("igb: Enable reading of wake up packet") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-04-27 19:09:52 +00:00
static int __maybe_unused igb_runtime_suspend(struct device *dev)
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
{
struct pci_dev *pdev = to_pci_dev(dev);
int retval;
bool wake;
retval = __igb_shutdown(pdev, &wake, 1);
if (retval)
return retval;
if (wake) {
pci_prepare_to_sleep(pdev);
} else {
pci_wake_from_d3(pdev, false);
pci_set_power_state(pdev, PCI_D3hot);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
return 0;
}
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
igb: mark PM functions as __maybe_unused The new wake function is only used by the suspend/resume handlers that are defined in inside of an #ifdef, which can cause this harmless warning: drivers/net/ethernet/intel/igb/igb_main.c:7988:13: warning: 'igb_deliver_wake_packet' defined but not used [-Wunused-function] Removing the #ifdef, instead using a __maybe_unused annotation simplifies the code and avoids the warning. Fixes: b90fa8763560 ("igb: Enable reading of wake up packet") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-04-27 19:09:52 +00:00
static int __maybe_unused igb_runtime_resume(struct device *dev)
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
{
return igb_resume(dev);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static void igb_shutdown(struct pci_dev *pdev)
{
net/igb: Fix kexec with igb (rev. 3) Impact: Fix Yinghai Lu found one system with 82575EB where, in the kernel that is kexeced, probe igb failed with -2, the reason being that the adapter could not be brought back from D3 by the kexec kernel, most probably due to quirky hardware (it looks like the same behavior happened on forcedeth). Prevent igb from putting the adapter into D3 during shutdown except when we going to power off the system. For this purpose, seperate igb_shutdown() from igb_suspend() and use the appropriate PCI PM callbacks in both of them. Signed-off-by: "Rafael J. Wysocki" <rjw@sisk.pl> Reported-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-31 21:23:50 +00:00
bool wake;
igb: add basic runtime PM support Use the runtime power management framework to add basic runtime PM support to the igb driver. Namely, make the driver suspend the device when the link is off and set it up for generating a wakeup event after the link has been detected again. This feature is disabled by default. Based on e1000e's runtime PM code. Changes since v1: Don't suspend the device when shutting down the interface. Avoid race between runtime suspending and ethtool operations. Signed-off-by: Zheng Yan <zheng.z.yan@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-04 20:23:37 +00:00
__igb_shutdown(pdev, &wake, 0);
net/igb: Fix kexec with igb (rev. 3) Impact: Fix Yinghai Lu found one system with 82575EB where, in the kernel that is kexeced, probe igb failed with -2, the reason being that the adapter could not be brought back from D3 by the kexec kernel, most probably due to quirky hardware (it looks like the same behavior happened on forcedeth). Prevent igb from putting the adapter into D3 during shutdown except when we going to power off the system. For this purpose, seperate igb_shutdown() from igb_suspend() and use the appropriate PCI PM callbacks in both of them. Signed-off-by: "Rafael J. Wysocki" <rjw@sisk.pl> Reported-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-31 21:23:50 +00:00
if (system_state == SYSTEM_POWER_OFF) {
pci_wake_from_d3(pdev, wake);
pci_set_power_state(pdev, PCI_D3hot);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
#ifdef CONFIG_PCI_IOV
static int igb_sriov_reinit(struct pci_dev *dev)
{
struct net_device *netdev = pci_get_drvdata(dev);
struct igb_adapter *adapter = netdev_priv(netdev);
struct pci_dev *pdev = adapter->pdev;
rtnl_lock();
if (netif_running(netdev))
igb_close(netdev);
igb: do a reset on SR-IOV re-init if device is down To properly re-initialize SR-IOV it is necessary to reset the device even if it is already down. Not doing this may result in Tx unit hangs. Cc: stable <stable@vger.kernel.org> Signed-off-by: Stefan Assmann <sassmann@kpanic.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-10 10:29:39 +00:00
else
igb_reset(adapter);
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
igb_clear_interrupt_scheme(adapter);
igb_init_queue_configuration(adapter);
if (igb_init_interrupt_scheme(adapter, true)) {
igb: missing rtnl_unlock in igb_sriov_reinit() Signed-off-by: Vasily Averin <vvs@virtuozzo.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-07-07 15:53:45 +00:00
rtnl_unlock();
igb: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-01-17 09:03:06 +00:00
dev_err(&pdev->dev, "Unable to allocate memory for queues\n");
return -ENOMEM;
}
if (netif_running(netdev))
igb_open(netdev);
rtnl_unlock();
return 0;
}
static int igb_pci_disable_sriov(struct pci_dev *dev)
{
int err = igb_disable_sriov(dev);
if (!err)
err = igb_sriov_reinit(dev);
return err;
}
static int igb_pci_enable_sriov(struct pci_dev *dev, int num_vfs)
{
int err = igb_enable_sriov(dev, num_vfs);
if (err)
goto out;
err = igb_sriov_reinit(dev);
if (!err)
return num_vfs;
out:
return err;
}
#endif
static int igb_pci_sriov_configure(struct pci_dev *dev, int num_vfs)
{
#ifdef CONFIG_PCI_IOV
if (num_vfs == 0)
return igb_pci_disable_sriov(dev);
else
return igb_pci_enable_sriov(dev, num_vfs);
#endif
return 0;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_io_error_detected - called when PCI error is detected
* @pdev: Pointer to PCI device
* @state: The current pci connection state
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* This function is called after a PCI bus error affecting
* this device has been detected.
**/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static pci_ers_result_t igb_io_error_detected(struct pci_dev *pdev,
pci_channel_state_t state)
{
struct net_device *netdev = pci_get_drvdata(pdev);
struct igb_adapter *adapter = netdev_priv(netdev);
netif_device_detach(netdev);
igb: return PCI_ERS_RESULT_DISCONNECT on permanent error PCI drivers that implement the io_error_detected callback should return PCI_ERS_RESULT_DISCONNECT if the state passed in is pci_channel_io_perm_failure. This patch fixes the issue for igb. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-06-30 12:46:34 +00:00
if (state == pci_channel_io_perm_failure)
return PCI_ERS_RESULT_DISCONNECT;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
if (netif_running(netdev))
igb_down(adapter);
pci_disable_device(pdev);
/* Request a slot slot reset. */
return PCI_ERS_RESULT_NEED_RESET;
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_io_slot_reset - called after the pci bus has been reset.
* @pdev: Pointer to PCI device
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* Restart the card from scratch, as if from a cold-boot. Implementation
* resembles the first-half of the igb_resume routine.
**/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
static pci_ers_result_t igb_io_slot_reset(struct pci_dev *pdev)
{
struct net_device *netdev = pci_get_drvdata(pdev);
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
igb: Add support for pci-e Advanced Error Reporting Add the calls necessary to enable advanced error reporting for igb on systems with AER enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-27 08:24:37 +00:00
pci_ers_result_t result;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: remove check for needing an io port Since igb supports only pci-e nics and there is no plan to support any legacy pci parts in the driver there isn't really much need for checking to see if an io port is needed. In the unlikely event that we do begin supporting legacy pci parts then we can see about adding this code back to the driver. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-06 23:16:04 +00:00
if (pci_enable_device_mem(pdev)) {
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
dev_err(&pdev->dev,
"Cannot re-enable PCI device after reset.\n");
igb: Add support for pci-e Advanced Error Reporting Add the calls necessary to enable advanced error reporting for igb on systems with AER enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-27 08:24:37 +00:00
result = PCI_ERS_RESULT_DISCONNECT;
} else {
pci_set_master(pdev);
pci_restore_state(pdev);
igb: call pci_save_state after pci_restore_state This patch adds calls to pci_save_state() immediately after calls to pci_restore_state(). Due to a change in the behavior of pci_restore_state() it is necessary to call pci_save_state() to keep the state_saved flag. This patch is based on a similar patch for ixgbe. Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-17 01:02:19 +00:00
pci_save_state(pdev);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Add support for pci-e Advanced Error Reporting Add the calls necessary to enable advanced error reporting for igb on systems with AER enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-27 08:24:37 +00:00
pci_enable_wake(pdev, PCI_D3hot, 0);
pci_enable_wake(pdev, PCI_D3cold, 0);
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: re-assign hw address pointer on reset after PCI error Whenever the igb driver detects the result of a read operation returns a value composed only by F's (like 0xFFFFFFFF), it will detach the net_device, clear the hw_addr pointer and warn to the user that adapter's link is lost - those steps happen on igb_rd32(). In case a PCI error happens on Power architecture, there's a recovery mechanism called EEH, that will reset the PCI slot and call driver's handlers to reset the adapter and network functionality as well. We observed that once hw_addr is NULL after the error is detected on igb_rd32(), it's never assigned back, so in the process of resetting the network functionality we got a NULL pointer dereference in both igb_configure_tx_ring() and igb_configure_rx_ring(). In order to avoid such bug, this patch re-assigns the hw_addr value in the slot_reset handler. Reported-by: Anthony H Thai <ahthai@us.ibm.com> Reported-by: Harsha Thyagaraja <hathyaga@in.ibm.com> Signed-off-by: Guilherme G Piccoli <gpiccoli@linux.vnet.ibm.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-11-10 18:46:43 +00:00
/* In case of PCI error, adapter lose its HW address
* so we should re-assign it here.
*/
hw->hw_addr = adapter->io_addr;
igb: Add support for pci-e Advanced Error Reporting Add the calls necessary to enable advanced error reporting for igb on systems with AER enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-27 08:24:37 +00:00
igb_reset(adapter);
wr32(E1000_WUS, ~0);
result = PCI_ERS_RESULT_RECOVERED;
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb: Add support for pci-e Advanced Error Reporting Add the calls necessary to enable advanced error reporting for igb on systems with AER enabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-27 08:24:37 +00:00
return result;
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
}
/**
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* igb_io_resume - called when traffic can start flowing again.
* @pdev: Pointer to PCI device
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* This callback is called when the error recovery driver tells us that
* its OK to resume normal operation. Implementation resembles the
* second-half of the igb_resume routine.
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
*/
static void igb_io_resume(struct pci_dev *pdev)
{
struct net_device *netdev = pci_get_drvdata(pdev);
struct igb_adapter *adapter = netdev_priv(netdev);
if (netif_running(netdev)) {
if (igb_up(adapter)) {
dev_err(&pdev->dev, "igb_up failed after reset\n");
return;
}
}
netif_device_attach(netdev);
/* let the f/w know that the h/w is now under the control of the
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* driver.
*/
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
igb_get_hw_control(adapter);
}
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
/**
* igb_rar_set_index - Sync RAL[index] and RAH[index] registers with MAC table
* @adapter: Pointer to adapter structure
* @index: Index of the RAR entry which need to be synced with MAC table
**/
static void igb_rar_set_index(struct igb_adapter *adapter, u32 index)
igb: add combined function for setting rar and pool bits This patch adds igb_rar_qsel which sets the mac address and pool bits for a given mac address in the receive address register table. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:32:49 +00:00
{
struct e1000_hw *hw = &adapter->hw;
igb: clean up code for setting MAC address Drop a bunch of hand written byte swapping code in favor of just doing the byte swapping ourselves. The registers are little endian registers storing a big endian value so if we read the MAC address array as little endian then we will get the CPU registers into the proper layout. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:23 +00:00
u32 rar_low, rar_high;
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
u8 *addr = adapter->mac_table[index].addr;
igb: add combined function for setting rar and pool bits This patch adds igb_rar_qsel which sets the mac address and pool bits for a given mac address in the receive address register table. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:32:49 +00:00
igb: Fix sparse warning about passing __beXX into leXX_to_cpup We were casting the addr as __beXX and then passing it into le32_to_cpu because the device expects the MAC address to be in network order even though the register set is little endian. Instead of casting it as __beXX we can just cast it as __leXX in order to maintain consistency since the region of memory is already in little endian order as far as we are concerned. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-03-18 23:06:53 +00:00
/* HW expects these to be in network order when they are plugged
* into the registers which are little endian. In order to guarantee
* that ordering we need to do an leXX_to_cpup here in order to be
* ready for the byteswap that occurs with writel
igb: add combined function for setting rar and pool bits This patch adds igb_rar_qsel which sets the mac address and pool bits for a given mac address in the receive address register table. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:32:49 +00:00
*/
igb: Fix sparse warning about passing __beXX into leXX_to_cpup We were casting the addr as __beXX and then passing it into le32_to_cpu because the device expects the MAC address to be in network order even though the register set is little endian. Instead of casting it as __beXX we can just cast it as __leXX in order to maintain consistency since the region of memory is already in little endian order as far as we are concerned. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-03-18 23:06:53 +00:00
rar_low = le32_to_cpup((__le32 *)(addr));
rar_high = le16_to_cpup((__le16 *)(addr + 4));
igb: add combined function for setting rar and pool bits This patch adds igb_rar_qsel which sets the mac address and pool bits for a given mac address in the receive address register table. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:32:49 +00:00
/* Indicate to hardware the Address is Valid. */
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
if (adapter->mac_table[index].state & IGB_MAC_STATE_IN_USE) {
igb: Allow to remove administratively set MAC on VFs Before libvirt modifies the MAC address and vlan tag for an SRIOV VF for use by a virtual machine (either using vfio device assignment or macvtap passthru mode), it saves the current MAC address and vlan tag so that it can reset them to their original value when the guest is done. Libvirt can't leave the VF MAC set to the value used by the now-defunct guest since it may be started again later using a different VF, but it certainly shouldn't just pick any random value, either. So it saves the state of everything prior to using the VF, and resets it to that. The igb driver initializes the MAC addresses of all VFs to 00:00:00:00:00:00, and reports that when asked (via an RTM_GETLINK netlink message, also visible in the list of VFs in the output of "ip link show"). But when libvirt attempts to restore the MAC address back to 00:00:00:00:00:00 (using an RTM_SETLINK netlink message) the kernel responds with "Invalid argument". Forbidding a reset back to the original value leaves the VF MAC at the value set for the now-defunct virtual machine. Especially on a system with NetworkManager enabled, this has very bad consequences, since NetworkManager forces all interfacess to be IFF_UP all the time - if the same virtual machine is restarted using a different VF (or even on a different host), there will be multiple interfaces watching for traffic with the same MAC address. To allow libvirt to revert to the original state, we need a way to remove the administrative set MAC on a VF, to allow normal host operation again, and to reset/overwrite the VF MAC via VF netdev. This patch implements the outlined scenario by allowing to set the VF MAC to 00:00:00:00:00:00 via RTM_SETLINK on the PF. igb_ndo_set_vf_mac resets the IGB_VF_FLAG_PF_SET_MAC flag to 0, so it's possible to reset the VF MAC back to the original value via the VF netdev. Note: Recent patches to libvirt allow for a workaround if the NIC isn't capable of resetting the administrative MAC back to all 0, but in theory the NIC should allow resetting the MAC in the first place. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <arron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-04-10 08:58:14 +00:00
if (is_valid_ether_addr(addr))
rar_high |= E1000_RAH_AV;
igb: add combined function for setting rar and pool bits This patch adds igb_rar_qsel which sets the mac address and pool bits for a given mac address in the receive address register table. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:32:49 +00:00
igb: Add support for MAC address filters specifying source addresses Makes it possible to direct packets to queues based on their source address. Documents the expected usage of the 'flags' parameter. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:53 +00:00
if (adapter->mac_table[index].state & IGB_MAC_STATE_SRC_ADDR)
rar_high |= E1000_RAH_ASEL_SRC_ADDR;
igb: Fix queue selection on MAC filters on i210 On the RAH registers there are semantic differences on the meaning of the "queue" parameter for traffic steering depending on the controller model: there is the 82575 meaning, which "queue" means a RX Hardware Queue, and the i350 meaning, where it is a reception pool. The previous behaviour was having no effect for i210 based controllers because the QSEL bit of the RAH register wasn't being set. This patch separates the condition in discrete cases, so the different handling is clearer. Fixes: 83c21335c876 ("igb: improve MAC filter handling") Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:51 +00:00
switch (hw->mac.type) {
case e1000_82575:
case e1000_i210:
igb: Add support for enabling queue steering in filters On some igb models (82575 and i210) the MAC address filters can control to which queue the packet will be assigned. This extends the 'state' with one more state to signify that queue selection should be enabled for that filter. As 82575 parts are no longer easily obtained (and this was developed against i210), only support for the i210 model is enabled. These functions are exported and will be used in the next patch. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:54 +00:00
if (adapter->mac_table[index].state &
IGB_MAC_STATE_QUEUE_STEERING)
rar_high |= E1000_RAH_QSEL_ENABLE;
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
rar_high |= E1000_RAH_POOL_1 *
adapter->mac_table[index].queue;
igb: Fix queue selection on MAC filters on i210 On the RAH registers there are semantic differences on the meaning of the "queue" parameter for traffic steering depending on the controller model: there is the 82575 meaning, which "queue" means a RX Hardware Queue, and the i350 meaning, where it is a reception pool. The previous behaviour was having no effect for i210 based controllers because the QSEL bit of the RAH register wasn't being set. This patch separates the condition in discrete cases, so the different handling is clearer. Fixes: 83c21335c876 ("igb: improve MAC filter handling") Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:51 +00:00
break;
default:
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
rar_high |= E1000_RAH_POOL_1 <<
adapter->mac_table[index].queue;
igb: Fix queue selection on MAC filters on i210 On the RAH registers there are semantic differences on the meaning of the "queue" parameter for traffic steering depending on the controller model: there is the 82575 meaning, which "queue" means a RX Hardware Queue, and the i350 meaning, where it is a reception pool. The previous behaviour was having no effect for i210 based controllers because the QSEL bit of the RAH register wasn't being set. This patch separates the condition in discrete cases, so the different handling is clearer. Fixes: 83c21335c876 ("igb: improve MAC filter handling") Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:51 +00:00
break;
}
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
}
igb: add combined function for setting rar and pool bits This patch adds igb_rar_qsel which sets the mac address and pool bits for a given mac address in the receive address register table. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-05 06:32:49 +00:00
wr32(E1000_RAL(index), rar_low);
wrfl();
wr32(E1000_RAH(index), rar_high);
wrfl();
}
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
static int igb_set_vf_mac(struct igb_adapter *adapter,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
int vf, unsigned char *mac_addr)
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
{
struct e1000_hw *hw = &adapter->hw;
igb: add support for set_rx_mode netdevice operation This patch adds support for the set_rx_mode netdevice operation so that igb can better support multiple unicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:48:56 +00:00
/* VF MAC addresses start at end of receive addresses and moves
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
* towards the first, as a result a collision should not be possible
*/
igb: add support for set_rx_mode netdevice operation This patch adds support for the set_rx_mode netdevice operation so that igb can better support multiple unicast addresses. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-09-03 14:48:56 +00:00
int rar_entry = hw->mac.rar_entry_count - (vf + 1);
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
unsigned char *vf_mac_addr = adapter->vf_data[vf].vf_mac_addresses;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: improve MAC filter handling Using the work which was done for ixgbe driver by Jacob Keller commit 5d7daa35b9eb ("ixgbe: improve mac filter handling") and Alexander Duyck commit 0f079d22834a ("ixgbe: Use __dev_uc_sync and __dev_uc_unsync for unicast addresses") and out-of-tree igb driver add functionality to manage (add and delete) MAC filters. Signed-off-by: Yury Kylulin <yury.kylulin@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-03-07 08:20:25 +00:00
ether_addr_copy(vf_mac_addr, mac_addr);
ether_addr_copy(adapter->mac_table[rar_entry].addr, mac_addr);
adapter->mac_table[rar_entry].queue = vf;
adapter->mac_table[rar_entry].state |= IGB_MAC_STATE_IN_USE;
igb_rar_set_index(adapter, rar_entry);
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
return 0;
}
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
static int igb_ndo_set_vf_mac(struct net_device *netdev, int vf, u8 *mac)
{
struct igb_adapter *adapter = netdev_priv(netdev);
igb: Allow to remove administratively set MAC on VFs Before libvirt modifies the MAC address and vlan tag for an SRIOV VF for use by a virtual machine (either using vfio device assignment or macvtap passthru mode), it saves the current MAC address and vlan tag so that it can reset them to their original value when the guest is done. Libvirt can't leave the VF MAC set to the value used by the now-defunct guest since it may be started again later using a different VF, but it certainly shouldn't just pick any random value, either. So it saves the state of everything prior to using the VF, and resets it to that. The igb driver initializes the MAC addresses of all VFs to 00:00:00:00:00:00, and reports that when asked (via an RTM_GETLINK netlink message, also visible in the list of VFs in the output of "ip link show"). But when libvirt attempts to restore the MAC address back to 00:00:00:00:00:00 (using an RTM_SETLINK netlink message) the kernel responds with "Invalid argument". Forbidding a reset back to the original value leaves the VF MAC at the value set for the now-defunct virtual machine. Especially on a system with NetworkManager enabled, this has very bad consequences, since NetworkManager forces all interfacess to be IFF_UP all the time - if the same virtual machine is restarted using a different VF (or even on a different host), there will be multiple interfaces watching for traffic with the same MAC address. To allow libvirt to revert to the original state, we need a way to remove the administrative set MAC on a VF, to allow normal host operation again, and to reset/overwrite the VF MAC via VF netdev. This patch implements the outlined scenario by allowing to set the VF MAC to 00:00:00:00:00:00 via RTM_SETLINK on the PF. igb_ndo_set_vf_mac resets the IGB_VF_FLAG_PF_SET_MAC flag to 0, so it's possible to reset the VF MAC back to the original value via the VF netdev. Note: Recent patches to libvirt allow for a workaround if the NIC isn't capable of resetting the administrative MAC back to all 0, but in theory the NIC should allow resetting the MAC in the first place. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <arron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2017-04-10 08:58:14 +00:00
if (vf >= adapter->vfs_allocated_count)
return -EINVAL;
/* Setting the VF MAC to 0 reverts the IGB_VF_FLAG_PF_SET_MAC
* flag and allows to overwrite the MAC via VF netdev. This
* is necessary to allow libvirt a way to restore the original
* MAC after unbinding vfio-pci and reloading igbvf after shutting
* down a VM.
*/
if (is_zero_ether_addr(mac)) {
adapter->vf_data[vf].flags &= ~IGB_VF_FLAG_PF_SET_MAC;
dev_info(&adapter->pdev->dev,
"remove administratively set MAC on VF %d\n",
vf);
} else if (is_valid_ether_addr(mac)) {
adapter->vf_data[vf].flags |= IGB_VF_FLAG_PF_SET_MAC;
dev_info(&adapter->pdev->dev, "setting MAC %pM on VF %d\n",
mac, vf);
dev_info(&adapter->pdev->dev,
"Reload the VF driver to make this change effective.");
/* Generate additional warning if PF is down */
if (test_bit(__IGB_DOWN, &adapter->state)) {
dev_warn(&adapter->pdev->dev,
"The VF MAC address has been set, but the PF device is not up.\n");
dev_warn(&adapter->pdev->dev,
"Bring the PF device up before attempting to use the VF device.\n");
}
} else {
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
return -EINVAL;
}
return igb_set_vf_mac(adapter, vf, mac);
}
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
static int igb_link_mbps(int internal_link_speed)
{
switch (internal_link_speed) {
case SPEED_100:
return 100;
case SPEED_1000:
return 1000;
default:
return 0;
}
}
static void igb_set_vf_rate_limit(struct e1000_hw *hw, int vf, int tx_rate,
int link_speed)
{
int rf_dec, rf_int;
u32 bcnrc_val;
if (tx_rate != 0) {
/* Calculate the rate factor values to set */
rf_int = link_speed / tx_rate;
rf_dec = (link_speed - (rf_int * tx_rate));
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
rf_dec = (rf_dec * BIT(E1000_RTTBCNRC_RF_INT_SHIFT)) /
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
tx_rate;
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
bcnrc_val = E1000_RTTBCNRC_RS_ENA;
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
bcnrc_val |= ((rf_int << E1000_RTTBCNRC_RF_INT_SHIFT) &
E1000_RTTBCNRC_RF_INT_MASK);
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
bcnrc_val |= (rf_dec & E1000_RTTBCNRC_RF_DEC_MASK);
} else {
bcnrc_val = 0;
}
wr32(E1000_RTTDQSEL, vf); /* vf X uses queue X */
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Set global transmit compensation time to the MMW_SIZE in RTTBCNRM
igb: A fix to VF TX rate limit There is a need to configure MMW_SIZE in register RTTBCNRM with a correct value. For 82576 device, the value should be 0x14. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-06-04 06:05:03 +00:00
* register. MMW_SIZE=0x014 if 9728-byte jumbo is supported.
*/
wr32(E1000_RTTBCNRM, 0x14);
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
wr32(E1000_RTTBCNRC, bcnrc_val);
}
static void igb_check_vf_rate_limit(struct igb_adapter *adapter)
{
int actual_link_speed, i;
bool reset_rate = false;
/* VF TX rate limit was not set or not supported */
if ((adapter->vf_rate_link_speed == 0) ||
(adapter->hw.mac.type != e1000_82576))
return;
actual_link_speed = igb_link_mbps(adapter->link_speed);
if (actual_link_speed != adapter->vf_rate_link_speed) {
reset_rate = true;
adapter->vf_rate_link_speed = 0;
dev_info(&adapter->pdev->dev,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
"Link speed has been changed. VF Transmit rate is disabled\n");
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
}
for (i = 0; i < adapter->vfs_allocated_count; i++) {
if (reset_rate)
adapter->vf_data[i].tx_rate = 0;
igb_set_vf_rate_limit(&adapter->hw, i,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
adapter->vf_data[i].tx_rate,
actual_link_speed);
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
}
}
net-next:v4: Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool. o min_tx_rate puts lower limit on the VF bandwidth. VF is guaranteed to have a bandwidth of at least this value. max_tx_rate puts cap on the VF bandwidth. VF can have a bandwidth of up to this value. o A new handler set_vf_rate for attr IFLA_VF_RATE has been introduced which takes 4 arguments: netdev, VF number, min_tx_rate, max_tx_rate o ndo_set_vf_rate replaces ndo_set_vf_tx_rate handler. o Drivers that currently implement ndo_set_vf_tx_rate should now call ndo_set_vf_rate instead and reject attempt to set a minimum bandwidth greater than 0 for IFLA_VF_TX_RATE when IFLA_VF_RATE is not yet implemented by driver. o If user enters only one of either min_tx_rate or max_tx_rate, then, userland should read back the other value from driver and set both for IFLA_VF_RATE. Drivers that have not yet implemented IFLA_VF_RATE should always return min_tx_rate as 0 when read from ip tool. o If both IFLA_VF_TX_RATE and IFLA_VF_RATE options are specified, then IFLA_VF_RATE should override. o Idea is to have consistent display of rate values to user. o Usage example: - ./ip link set p4p1 vf 0 rate 900 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 900 (Mbps), max_tx_rate 900Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 300 min_tx_rate 200 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 300 (Mbps), max_tx_rate 300Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 600 rate 300 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5, tx rate 600 (Mbps), max_tx_rate 600Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 13:59:05 +00:00
static int igb_ndo_set_vf_bw(struct net_device *netdev, int vf,
int min_tx_rate, int max_tx_rate)
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
{
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
int actual_link_speed;
if (hw->mac.type != e1000_82576)
return -EOPNOTSUPP;
net-next:v4: Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool. o min_tx_rate puts lower limit on the VF bandwidth. VF is guaranteed to have a bandwidth of at least this value. max_tx_rate puts cap on the VF bandwidth. VF can have a bandwidth of up to this value. o A new handler set_vf_rate for attr IFLA_VF_RATE has been introduced which takes 4 arguments: netdev, VF number, min_tx_rate, max_tx_rate o ndo_set_vf_rate replaces ndo_set_vf_tx_rate handler. o Drivers that currently implement ndo_set_vf_tx_rate should now call ndo_set_vf_rate instead and reject attempt to set a minimum bandwidth greater than 0 for IFLA_VF_TX_RATE when IFLA_VF_RATE is not yet implemented by driver. o If user enters only one of either min_tx_rate or max_tx_rate, then, userland should read back the other value from driver and set both for IFLA_VF_RATE. Drivers that have not yet implemented IFLA_VF_RATE should always return min_tx_rate as 0 when read from ip tool. o If both IFLA_VF_TX_RATE and IFLA_VF_RATE options are specified, then IFLA_VF_RATE should override. o Idea is to have consistent display of rate values to user. o Usage example: - ./ip link set p4p1 vf 0 rate 900 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 900 (Mbps), max_tx_rate 900Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 300 min_tx_rate 200 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 300 (Mbps), max_tx_rate 300Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 600 rate 300 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5, tx rate 600 (Mbps), max_tx_rate 600Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 13:59:05 +00:00
if (min_tx_rate)
return -EINVAL;
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
actual_link_speed = igb_link_mbps(adapter->link_speed);
if ((vf >= adapter->vfs_allocated_count) ||
(!(rd32(E1000_STATUS) & E1000_STATUS_LU)) ||
net-next:v4: Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool. o min_tx_rate puts lower limit on the VF bandwidth. VF is guaranteed to have a bandwidth of at least this value. max_tx_rate puts cap on the VF bandwidth. VF can have a bandwidth of up to this value. o A new handler set_vf_rate for attr IFLA_VF_RATE has been introduced which takes 4 arguments: netdev, VF number, min_tx_rate, max_tx_rate o ndo_set_vf_rate replaces ndo_set_vf_tx_rate handler. o Drivers that currently implement ndo_set_vf_tx_rate should now call ndo_set_vf_rate instead and reject attempt to set a minimum bandwidth greater than 0 for IFLA_VF_TX_RATE when IFLA_VF_RATE is not yet implemented by driver. o If user enters only one of either min_tx_rate or max_tx_rate, then, userland should read back the other value from driver and set both for IFLA_VF_RATE. Drivers that have not yet implemented IFLA_VF_RATE should always return min_tx_rate as 0 when read from ip tool. o If both IFLA_VF_TX_RATE and IFLA_VF_RATE options are specified, then IFLA_VF_RATE should override. o Idea is to have consistent display of rate values to user. o Usage example: - ./ip link set p4p1 vf 0 rate 900 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 900 (Mbps), max_tx_rate 900Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 300 min_tx_rate 200 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 300 (Mbps), max_tx_rate 300Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 600 rate 300 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5, tx rate 600 (Mbps), max_tx_rate 600Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 13:59:05 +00:00
(max_tx_rate < 0) ||
(max_tx_rate > actual_link_speed))
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
return -EINVAL;
adapter->vf_rate_link_speed = actual_link_speed;
net-next:v4: Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool. o min_tx_rate puts lower limit on the VF bandwidth. VF is guaranteed to have a bandwidth of at least this value. max_tx_rate puts cap on the VF bandwidth. VF can have a bandwidth of up to this value. o A new handler set_vf_rate for attr IFLA_VF_RATE has been introduced which takes 4 arguments: netdev, VF number, min_tx_rate, max_tx_rate o ndo_set_vf_rate replaces ndo_set_vf_tx_rate handler. o Drivers that currently implement ndo_set_vf_tx_rate should now call ndo_set_vf_rate instead and reject attempt to set a minimum bandwidth greater than 0 for IFLA_VF_TX_RATE when IFLA_VF_RATE is not yet implemented by driver. o If user enters only one of either min_tx_rate or max_tx_rate, then, userland should read back the other value from driver and set both for IFLA_VF_RATE. Drivers that have not yet implemented IFLA_VF_RATE should always return min_tx_rate as 0 when read from ip tool. o If both IFLA_VF_TX_RATE and IFLA_VF_RATE options are specified, then IFLA_VF_RATE should override. o Idea is to have consistent display of rate values to user. o Usage example: - ./ip link set p4p1 vf 0 rate 900 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 900 (Mbps), max_tx_rate 900Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 300 min_tx_rate 200 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 300 (Mbps), max_tx_rate 300Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 600 rate 300 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5, tx rate 600 (Mbps), max_tx_rate 600Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 13:59:05 +00:00
adapter->vf_data[vf].tx_rate = (u16)max_tx_rate;
igb_set_vf_rate_limit(hw, vf, max_tx_rate, actual_link_speed);
igb: add support for VF Transmit rate limit using iproute2 Implemented igb_ndo_set_vf_bw function which is being used by iproute2 tool. In addition, updated igb_ndo_get_vf_config function to show the actual rate limit to the user. The rate limitation can be configured only when the link is up. The rate limit value can be ranged between 0 and actual link speed measured in Mbps. A value of '0' disables the rate limit for this specific VF. iproute2 usage will be 'ip link set ethX vf Y rate Z'. After the command is made, the rate will be changed instantly. To view the current rate limit, use 'ip link show ethX'. The rates will be zeroed only upon driver reload or a link speed change. This feature is being supported only by 82576 device. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-02-08 02:28:46 +00:00
return 0;
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
}
igb: add support for spoofchk config Add support for spoofchk configuration per VF via iproute2 tool. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-03 20:27:48 +00:00
static int igb_ndo_set_vf_spoofchk(struct net_device *netdev, int vf,
bool setting)
{
struct igb_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
u32 reg_val, reg_offset;
if (!adapter->vfs_allocated_count)
return -EOPNOTSUPP;
if (vf >= adapter->vfs_allocated_count)
return -EINVAL;
reg_offset = (hw->mac.type == e1000_82576) ? E1000_DTXSWC : E1000_TXSWC;
reg_val = rd32(reg_offset);
if (setting)
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
reg_val |= (BIT(vf) |
BIT(vf + E1000_DTXSWC_VLAN_SPOOF_SHIFT));
igb: add support for spoofchk config Add support for spoofchk configuration per VF via iproute2 tool. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-03 20:27:48 +00:00
else
igb: use BIT() macro or unsigned prefix For bitshifts, we should make use of the BIT macro when possible, and ensure that other bitshifts are marked as unsigned. This helps prevent signed bitshift errors, and ensures similar style. Make use of GENMASK and the unsigned postfix where BIT() isn't appropriate. Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-04-13 23:08:28 +00:00
reg_val &= ~(BIT(vf) |
BIT(vf + E1000_DTXSWC_VLAN_SPOOF_SHIFT));
igb: add support for spoofchk config Add support for spoofchk configuration per VF via iproute2 tool. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-03 20:27:48 +00:00
wr32(reg_offset, reg_val);
adapter->vf_data[vf].spoofchk_enabled = setting;
igb: unhide invariant returns Return a 0 directly rather than a constant. Reported-by: Peter Senna Tschudin <peter.senna@gmail.com> Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-06-04 07:12:15 +00:00
return 0;
igb: add support for spoofchk config Add support for spoofchk configuration per VF via iproute2 tool. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-03 20:27:48 +00:00
}
igb: add VF trust infrastructure * Add a per-VF value to know if a VF is trusted, by default don't trust VFs. * Implement netdev op to trust VFs (igb_ndo_set_vf_trust) and add trust status to ndo_get_vf_config output. * Allow a trusted VF to change MAC and MAC filters even if MAC has been administratively set. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-01-17 10:53:39 +00:00
static int igb_ndo_set_vf_trust(struct net_device *netdev, int vf, bool setting)
{
struct igb_adapter *adapter = netdev_priv(netdev);
if (vf >= adapter->vfs_allocated_count)
return -EINVAL;
if (adapter->vf_data[vf].trusted == setting)
return 0;
adapter->vf_data[vf].trusted = setting;
dev_info(&adapter->pdev->dev, "VF %u is %strusted\n",
vf, setting ? "" : "not ");
return 0;
}
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
static int igb_ndo_get_vf_config(struct net_device *netdev,
int vf, struct ifla_vf_info *ivi)
{
struct igb_adapter *adapter = netdev_priv(netdev);
if (vf >= adapter->vfs_allocated_count)
return -EINVAL;
ivi->vf = vf;
memcpy(&ivi->mac, adapter->vf_data[vf].vf_mac_addresses, ETH_ALEN);
net-next:v4: Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool. o min_tx_rate puts lower limit on the VF bandwidth. VF is guaranteed to have a bandwidth of at least this value. max_tx_rate puts cap on the VF bandwidth. VF can have a bandwidth of up to this value. o A new handler set_vf_rate for attr IFLA_VF_RATE has been introduced which takes 4 arguments: netdev, VF number, min_tx_rate, max_tx_rate o ndo_set_vf_rate replaces ndo_set_vf_tx_rate handler. o Drivers that currently implement ndo_set_vf_tx_rate should now call ndo_set_vf_rate instead and reject attempt to set a minimum bandwidth greater than 0 for IFLA_VF_TX_RATE when IFLA_VF_RATE is not yet implemented by driver. o If user enters only one of either min_tx_rate or max_tx_rate, then, userland should read back the other value from driver and set both for IFLA_VF_RATE. Drivers that have not yet implemented IFLA_VF_RATE should always return min_tx_rate as 0 when read from ip tool. o If both IFLA_VF_TX_RATE and IFLA_VF_RATE options are specified, then IFLA_VF_RATE should override. o Idea is to have consistent display of rate values to user. o Usage example: - ./ip link set p4p1 vf 0 rate 900 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 900 (Mbps), max_tx_rate 900Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 300 min_tx_rate 200 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 300 (Mbps), max_tx_rate 300Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 600 rate 300 ./ip link show p4p1 32: p4p1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5, tx rate 600 (Mbps), max_tx_rate 600Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a Signed-off-by: Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 13:59:05 +00:00
ivi->max_tx_rate = adapter->vf_data[vf].tx_rate;
ivi->min_tx_rate = 0;
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
ivi->vlan = adapter->vf_data[vf].pf_vlan;
ivi->qos = adapter->vf_data[vf].pf_qos;
igb: add support for spoofchk config Add support for spoofchk configuration per VF via iproute2 tool. Signed-off-by: Lior Levy <lior.levy@intel.com> Tested-by: Sibai Li <sibai.li@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-03-03 20:27:48 +00:00
ivi->spoofchk = adapter->vf_data[vf].spoofchk_enabled;
igb: add VF trust infrastructure * Add a per-VF value to know if a VF is trusted, by default don't trust VFs. * Implement netdev op to trust VFs (igb_ndo_set_vf_trust) and add trust status to ndo_get_vf_config output. * Allow a trusted VF to change MAC and MAC filters even if MAC has been administratively set. Signed-off-by: Corinna Vinschen <vinschen@redhat.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-01-17 10:53:39 +00:00
ivi->trusted = adapter->vf_data[vf].trusted;
igb: support for VF configuration tools Add support to the igb driver for VF configuration mechanisms through the PF interface. Signed-off-by: Mitch Williams <mitch.a.williams@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-10 01:44:24 +00:00
return 0;
}
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
static void igb_vmm_control(struct igb_adapter *adapter)
{
struct e1000_hw *hw = &adapter->hw;
igb: setup vlan tag replication stripping in igb_vmm_control This update adds vlan tag stripping for inter-vf communications to the igb_vmm_control configuration function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:54:04 +00:00
u32 reg;
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
igb: cleanup usage of virtualization registers The igb driver was incorrectly attempting to write to registers that do not exist on 82580 hardware. This wasn't causing any issues that I can tell, but it is not recommended behavior either so I have changed this so we are determining which registers to write to based on mac type. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-22 14:07:46 +00:00
switch (hw->mac.type) {
case e1000_82575:
igb: Add Support for new i210/i211 devices. This patch adds new initialization functions and device support for i210 and i211 devices. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-06 23:25:19 +00:00
case e1000_i210:
case e1000_i211:
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
case e1000_i354:
igb: cleanup usage of virtualization registers The igb driver was incorrectly attempting to write to registers that do not exist on 82580 hardware. This wasn't causing any issues that I can tell, but it is not recommended behavior either so I have changed this so we are determining which registers to write to based on mac type. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-22 14:07:46 +00:00
default:
/* replication is not supported for 82575 */
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
return;
igb: cleanup usage of virtualization registers The igb driver was incorrectly attempting to write to registers that do not exist on 82580 hardware. This wasn't causing any issues that I can tell, but it is not recommended behavior either so I have changed this so we are determining which registers to write to based on mac type. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-22 14:07:46 +00:00
case e1000_82576:
/* notify HW that the MAC is adding vlan tags */
reg = rd32(E1000_DTXCTL);
reg |= E1000_DTXCTL_VLAN_ADDED;
wr32(E1000_DTXCTL, reg);
igb: Cleanups to fix missing break in switch statements This patch fixes WARNING:MISSING_BREAK found with checkpatch check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-17 04:10:13 +00:00
/* Fall through */
igb: cleanup usage of virtualization registers The igb driver was incorrectly attempting to write to registers that do not exist on 82580 hardware. This wasn't causing any issues that I can tell, but it is not recommended behavior either so I have changed this so we are determining which registers to write to based on mac type. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-22 14:07:46 +00:00
case e1000_82580:
/* enable replication vlan tag stripping */
reg = rd32(E1000_RPLOLR);
reg |= E1000_RPLOLR_STRVLAN;
wr32(E1000_RPLOLR, reg);
igb: Cleanups to fix missing break in switch statements This patch fixes WARNING:MISSING_BREAK found with checkpatch check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-17 04:10:13 +00:00
/* Fall through */
igb: add support for Intel I350 Gigabit Network Connection This patch adds support for the the I350 Gigabit network connection which is the follow-on part to the 82580. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> CC: James Hearn <james.r.hearn@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-22 14:08:06 +00:00
case e1000_i350:
/* none of the above registers are supported by i350 */
igb: cleanup usage of virtualization registers The igb driver was incorrectly attempting to write to registers that do not exist on 82580 hardware. This wasn't causing any issues that I can tell, but it is not recommended behavior either so I have changed this so we are determining which registers to write to based on mac type. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-03-22 14:07:46 +00:00
break;
}
igb: setup vlan tag replication stripping in igb_vmm_control This update adds vlan tag stripping for inter-vf communications to the igb_vmm_control configuration function. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:54:04 +00:00
igb: delay VF reset notification until after interrupts are enabed This update delays the VF reset notification until after interrupts are enabled. Otherwise there is a chance of having the VF try to reset itself too soon and being ignored by the PF as a result. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:45 +00:00
if (adapter->vfs_allocated_count) {
igb_vmdq_set_loopback_pf(hw, true);
igb_vmdq_set_replication_pf(hw, true);
igb: Add Anti-spoofing feature support Add support for the anti-spoofing feature in the HW. Packets from VF devices with spoofed MAC addresses or VLAN tags will be blocked and an event generated. When the watchdog task runs it will call a function to check if any spoof events occurred. If an event was detected then a warning message is dumped to the system log. Signed-off-by: Greg Rose <gregory.v.rose@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2010-11-06 02:08:26 +00:00
igb_vmdq_set_anti_spoofing_pf(hw, true,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
adapter->vfs_allocated_count);
igb: delay VF reset notification until after interrupts are enabed This update delays the VF reset notification until after interrupts are enabled. Otherwise there is a chance of having the VF try to reset itself too soon and being ignored by the PF as a result. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-27 15:53:45 +00:00
} else {
igb_vmdq_set_loopback_pf(hw, false);
igb_vmdq_set_replication_pf(hw, false);
}
igb: Add support for enabling VFs to PF driver. This patch adds the support to handle requests from the VF to perform operations such as completing resets, setting/reading mac address, adding vlans, adding multicast addresses, setting rlpml, and general communications between the PF and all VFs. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-20 04:40:07 +00:00
}
igb: Move DMA Coalescing init code to separate function. This patch moves the DMA Coalescing feature initialization code from igb_reset to a new function and replaces it with a call to the new function. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-13 17:29:59 +00:00
static void igb_init_dmac(struct igb_adapter *adapter, u32 pba)
{
struct e1000_hw *hw = &adapter->hw;
u32 dmac_thr;
u16 hwm;
if (hw->mac.type > e1000_82580) {
if (adapter->flags & IGB_FLAG_DMAC) {
u32 reg;
/* force threshold to 0. */
wr32(E1000_DMCTXTH, 0);
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* DMA Coalescing high water mark needs to be greater
igb: Update DMA Coalescing threshold calculation. This patch updates the DMA Coalescing feature parameters to account for larger MTUs. Previously, sufficient space may not have been allocated in the receive buffer, causing packet drop. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-11-17 08:33:12 +00:00
* than the Rx threshold. Set hwm to PBA - max frame
* size in 16B units, capping it at PBA - 6KB.
igb: Move DMA Coalescing init code to separate function. This patch moves the DMA Coalescing feature initialization code from igb_reset to a new function and replaces it with a call to the new function. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-13 17:29:59 +00:00
*/
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
hwm = 64 * (pba - 6);
igb: Update DMA Coalescing threshold calculation. This patch updates the DMA Coalescing feature parameters to account for larger MTUs. Previously, sufficient space may not have been allocated in the receive buffer, causing packet drop. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-11-17 08:33:12 +00:00
reg = rd32(E1000_FCRTC);
reg &= ~E1000_FCRTC_RTH_COAL_MASK;
reg |= ((hwm << E1000_FCRTC_RTH_COAL_SHIFT)
& E1000_FCRTC_RTH_COAL_MASK);
wr32(E1000_FCRTC, reg);
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* Set the DMA Coalescing Rx threshold to PBA - 2 * max
igb: Update DMA Coalescing threshold calculation. This patch updates the DMA Coalescing feature parameters to account for larger MTUs. Previously, sufficient space may not have been allocated in the receive buffer, causing packet drop. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-11-17 08:33:12 +00:00
* frame size, capping it at PBA - 10KB.
*/
igb: Allow asymmetric configuration of MTU versus Rx frame size Since the igb driver is using page based receive there is no point in limiting the Rx capabilities of the device. The driver can receive 9K jumbo frames at all times. The only changes needed due to MTU changes are updates for the FIFO sizes and flow-control watermarks. Update the maximum frame size to reflect the 9.5K limitation of the hardware, and replace all instances of max_frame_size with MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-01-07 07:10:39 +00:00
dmac_thr = pba - 10;
igb: Move DMA Coalescing init code to separate function. This patch moves the DMA Coalescing feature initialization code from igb_reset to a new function and replaces it with a call to the new function. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-13 17:29:59 +00:00
reg = rd32(E1000_DMACR);
reg &= ~E1000_DMACR_DMACTHR_MASK;
reg |= ((dmac_thr << E1000_DMACR_DMACTHR_SHIFT)
& E1000_DMACR_DMACTHR_MASK);
/* transition to L0x or L1 if available..*/
reg |= (E1000_DMACR_DMAC_EN | E1000_DMACR_DMAC_LX_MASK);
/* watchdog timer= +-1000 usec in 32usec intervals */
reg |= (1000 >> 5);
igb: Disable the BMC-to-OS Watchdog Enable bit for DMAC. Under certain scenarios, it's possible that bursty manageability traffic over the BMC-to-OS path may overrun the internal manageability receive buffer causing dropped manageability packets. Clearing this bit prevents this situation by interrupting coalescing to allow manageability traffic through. Signed-off-by: Matthew Vick <matthew.vick@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-04-14 05:20:32 +00:00
/* Disable BMC-to-OS Watchdog Enable */
igb: Add support for i354 devices This patch adds base support for new i354 devices. Loopback test is unsupported for this release. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-04-18 22:21:30 +00:00
if (hw->mac.type != e1000_i354)
reg &= ~E1000_DMACR_DC_BMC2OSW_EN;
igb: Move DMA Coalescing init code to separate function. This patch moves the DMA Coalescing feature initialization code from igb_reset to a new function and replaces it with a call to the new function. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-13 17:29:59 +00:00
wr32(E1000_DMACR, reg);
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* no lower threshold to disable
igb: Move DMA Coalescing init code to separate function. This patch moves the DMA Coalescing feature initialization code from igb_reset to a new function and replaces it with a call to the new function. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-13 17:29:59 +00:00
* coalescing(smart fifb)-UTRESH=0
*/
wr32(E1000_DMCRTRH, 0);
reg = (IGB_DMCTLX_DCFLUSH_DIS | 0x4);
wr32(E1000_DMCTLX, reg);
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* free space in tx packet buffer to wake from
igb: Move DMA Coalescing init code to separate function. This patch moves the DMA Coalescing feature initialization code from igb_reset to a new function and replaces it with a call to the new function. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-13 17:29:59 +00:00
* DMA coal
*/
wr32(E1000_DMCTXTH, (IGB_MIN_TXPBSIZE -
(IGB_TX_BUF_4096 + adapter->max_frame_size)) >> 6);
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/* make low power state decision controlled
igb: Move DMA Coalescing init code to separate function. This patch moves the DMA Coalescing feature initialization code from igb_reset to a new function and replaces it with a call to the new function. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-13 17:29:59 +00:00
* by DMA coal
*/
reg = rd32(E1000_PCIEMISC);
reg &= ~E1000_PCIEMISC_LX_DECISION;
wr32(E1000_PCIEMISC, reg);
} /* endif adapter->dmac is not disabled */
} else if (hw->mac.type == e1000_82580) {
u32 reg = rd32(E1000_PCIEMISC);
igb: Cleanups to fix incorrect indentation This patch fixes WARNING:LEADING_SPACE, WARNING:SPACING, ERROR:SPACING, WARNING:SPACE_BEFORE_TAB and ERROR_CODE_INDENT from checkpatch file check. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-04-11 01:45:34 +00:00
igb: Move DMA Coalescing init code to separate function. This patch moves the DMA Coalescing feature initialization code from igb_reset to a new function and replaces it with a call to the new function. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2011-10-13 17:29:59 +00:00
wr32(E1000_PCIEMISC, reg & ~E1000_PCIEMISC_LX_DECISION);
wr32(E1000_DMACR, 0);
}
}
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/**
* igb_read_i2c_byte - Reads 8 bit word over I2C
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
* @hw: pointer to hardware structure
* @byte_offset: byte offset to read
* @dev_addr: device address
* @data: value read
*
* Performs byte read operation over I2C interface at
* a specified device address.
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
**/
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
s32 igb_read_i2c_byte(struct e1000_hw *hw, u8 byte_offset,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
u8 dev_addr, u8 *data)
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
{
struct igb_adapter *adapter = container_of(hw, struct igb_adapter, hw);
igb: Fix for lockdep issue in igb_get_i2c_client This patch fixes a lockdep warning in igb_get_i2c_client by refactoring the initialization and usage of the i2c_client completely. There is no on the fly allocation of the single client needed today. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-20 07:40:55 +00:00
struct i2c_client *this_client = adapter->i2c_client;
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
s32 status;
u16 swfw_mask = 0;
if (!this_client)
return E1000_ERR_I2C;
swfw_mask = E1000_SWFW_PHY0_SM;
igb: unhide invariant returns Return a 0 directly rather than a constant. Reported-by: Peter Senna Tschudin <peter.senna@gmail.com> Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-06-04 07:12:15 +00:00
if (hw->mac.ops.acquire_swfw_sync(hw, swfw_mask))
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
return E1000_ERR_SWFW_SYNC;
status = i2c_smbus_read_byte_data(this_client, byte_offset);
hw->mac.ops.release_swfw_sync(hw, swfw_mask);
if (status < 0)
return E1000_ERR_I2C;
else {
*data = status;
igb: unhide invariant returns Return a 0 directly rather than a constant. Reported-by: Peter Senna Tschudin <peter.senna@gmail.com> Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-06-04 07:12:15 +00:00
return 0;
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
}
}
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
/**
* igb_write_i2c_byte - Writes 8 bit word over I2C
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
* @hw: pointer to hardware structure
* @byte_offset: byte offset to write
* @dev_addr: device address
* @data: value to write
*
* Performs byte write operation over I2C interface at
* a specified device address.
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
**/
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
s32 igb_write_i2c_byte(struct e1000_hw *hw, u8 byte_offset,
igb: Fix code comments and whitespace Aligns the multi-line code comments with the desired style for the networking tree. Also cleaned up whitespace issues found during the cleanup of code comments (i.e. remove unnecessary blank lines, use tabs where possible, properly wrap lines and keep strings on a single line) Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com>
2013-02-23 07:29:56 +00:00
u8 dev_addr, u8 data)
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
{
struct igb_adapter *adapter = container_of(hw, struct igb_adapter, hw);
igb: Fix for lockdep issue in igb_get_i2c_client This patch fixes a lockdep warning in igb_get_i2c_client by refactoring the initialization and usage of the i2c_client completely. There is no on the fly allocation of the single client needed today. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-02-20 07:40:55 +00:00
struct i2c_client *this_client = adapter->i2c_client;
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
s32 status;
u16 swfw_mask = E1000_SWFW_PHY0_SM;
if (!this_client)
return E1000_ERR_I2C;
igb: unhide invariant returns Return a 0 directly rather than a constant. Reported-by: Peter Senna Tschudin <peter.senna@gmail.com> Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-06-04 07:12:15 +00:00
if (hw->mac.ops.acquire_swfw_sync(hw, swfw_mask))
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
return E1000_ERR_SWFW_SYNC;
status = i2c_smbus_write_byte_data(this_client, byte_offset, data);
hw->mac.ops.release_swfw_sync(hw, swfw_mask);
if (status)
return E1000_ERR_I2C;
else
igb: unhide invariant returns Return a 0 directly rather than a constant. Reported-by: Peter Senna Tschudin <peter.senna@gmail.com> Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2014-06-04 07:12:15 +00:00
return 0;
igb: Add i2c interface to igb. Some of our adapters have sensors on them accessible via i2c and a private interface. This patch implements the kernel interface for i2c to those sensors. Subsequent patches will provide functions to export that data. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2012-12-07 03:00:30 +00:00
}
igb: Add ethtool support to configure number of channels This patch adds the ethtool callbacks necessary to configure the number of RSS queues. The maximum number of queues is in accordance with the datasheets. Signed-off-by: Laura Mihaela Vasilescu <laura.vasilescu@rosedu.org> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-01 11:33:56 +00:00
int igb_reinit_queues(struct igb_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
struct pci_dev *pdev = adapter->pdev;
int err = 0;
if (netif_running(netdev))
igb_close(netdev);
igb: Fix queue allocation method to accommodate changing during runtime When changing number of queues using ethtool's set_channels during runtime, a queue allocation could fail, which can leave the device in a down state. In order to preserve the usability of the device in this scenario, this patch changes the driver to allocate the number of queues only if they have not been allocated already. The first allocation is then done for the max number of queues, which is the default queues for this driver. With this change, queue quantity changes are not subject to queue allocation failures. Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-12-10 07:58:29 +00:00
igb_reset_interrupt_capability(adapter);
igb: Add ethtool support to configure number of channels This patch adds the ethtool callbacks necessary to configure the number of RSS queues. The maximum number of queues is in accordance with the datasheets. Signed-off-by: Laura Mihaela Vasilescu <laura.vasilescu@rosedu.org> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-01 11:33:56 +00:00
if (igb_init_interrupt_scheme(adapter, true)) {
dev_err(&pdev->dev, "Unable to allocate memory for queues\n");
return -ENOMEM;
}
if (netif_running(netdev))
err = igb_open(netdev);
return err;
}
igb: add support of RX network flow classification This patch is meant to allow for RX network flow classification to insert and remove Rx filter by ethtool. Ethtool interface has it's own rules manager Show all filters: $ ethtool -n eth0 4 RX rings available Total 2 rules Signed-off-by: Ruhao Gao <ruhao.gao@ni.com> Signed-off-by: Gangfeng Huang <gangfeng.huang@ni.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-07-06 05:22:54 +00:00
static void igb_nfc_filter_exit(struct igb_adapter *adapter)
{
struct igb_nfc_filter *rule;
spin_lock(&adapter->nfc_lock);
hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node)
igb_erase_filter(adapter, rule);
igb: Add support for adding offloaded clsflower filters This allows filters added by tc-flower and specifying MAC addresses, Ethernet types, and the VLAN priority field, to be offloaded to the controller. This reuses most of the infrastructure used by ethtool, but clsflower filters are kept in a separated list, so they are invisible to ethtool. To setup clsflower offloading: $ tc qdisc replace dev eth0 handle 100: parent root mqprio \ num_tc 3 map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 \ queues 1@0 1@1 2@2 hw 0 (clsflower offloading depends on the netword driver to be configured with multiple traffic classes, we use mqprio's 'num_tc' parameter to set it to 3) $ tc qdisc add dev eth0 ingress Examples of filters: $ tc filter add dev eth0 parent ffff: flower \ dst_mac aa:aa:aa:aa:aa:aa \ hw_tc 2 skip_sw (just a simple filter filtering for the destination MAC address and steering that traffic to queue 2) $ tc filter add dev enp2s0 parent ffff: proto 0x22f0 flower \ src_mac cc:cc:cc:cc:cc:cc \ hw_tc 1 skip_sw (as the i210 doesn't support steering traffic based on the source address alone, we need to use another steering traffic, in this case we are using the ethernet type (0x22f0) to steer traffic to queue 1) Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-04-10 17:49:59 +00:00
hlist_for_each_entry(rule, &adapter->cls_flower_list, nfc_node)
igb_erase_filter(adapter, rule);
igb: add support of RX network flow classification This patch is meant to allow for RX network flow classification to insert and remove Rx filter by ethtool. Ethtool interface has it's own rules manager Show all filters: $ ethtool -n eth0 4 RX rings available Total 2 rules Signed-off-by: Ruhao Gao <ruhao.gao@ni.com> Signed-off-by: Gangfeng Huang <gangfeng.huang@ni.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2016-07-06 05:22:54 +00:00
spin_unlock(&adapter->nfc_lock);
}
static void igb_nfc_filter_restore(struct igb_adapter *adapter)
{
struct igb_nfc_filter *rule;
spin_lock(&adapter->nfc_lock);
hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node)
igb_add_filter(adapter, rule);
spin_unlock(&adapter->nfc_lock);
}
igb: PCI-Express 82575 Gigabit Ethernet driver We are pleased to announce a new Gigabit Ethernet product and its driver to the linux community. This product is the Intel(R) 82575 Gigabit Ethernet adapter family. Physical adapters will be available to the public soon. These adapters come in 2- and 4-port versions (copper PHY) currently. Other variants will be available later. The 82575 chipset supports significantly different features that warrant a new driver. The descriptor format is (just like the ixgbe driver) different. The device can use multiple MSI-X vectors and multiple queues for both send and receive. This allows us to optimize some of the driver code specifically as well compared to the e1000-supported devices. This version of the igb driver no lnger uses fake netdevices and incorporates napi_struct members for each ring to do the multi- queue polling. multi-queue is enabled by default and the driver supports NAPI mode only. All the namespace collisions should be gone in this version too. The register macro's have been condensed to improve readability. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-24 10:22:38 +00:00
/* igb_main.c */