linux-stable/net
John Fastabend 4f57c087de net: implement mechanism for HW based QOS
This patch provides a mechanism for lower layer devices to
steer traffic using skb->priority to tx queues. This allows
for hardware based QOS schemes to use the default qdisc without
incurring the penalties related to global state and the qdisc
lock. While reliably receiving skbs on the correct tx ring
to avoid head of line blocking resulting from shuffling in
the LLD. Finally, all the goodness from txq caching and xps/rps
can still be leveraged.

Many drivers and hardware exist with the ability to implement
QOS schemes in the hardware but currently these drivers tend
to rely on firmware to reroute specific traffic, a driver
specific select_queue or the queue_mapping action in the
qdisc.

By using select_queue for this drivers need to be updated for
each and every traffic type and we lose the goodness of much
of the upstream work. Firmware solutions are inherently
inflexible. And finally if admins are expected to build a
qdisc and filter rules to steer traffic this requires knowledge
of how the hardware is currently configured. The number of tx
queues and the queue offsets may change depending on resources.
Also this approach incurs all the overhead of a qdisc with filters.

With the mechanism in this patch users can set skb priority using
expected methods ie setsockopt() or the stack can set the priority
directly. Then the skb will be steered to the correct tx queues
aligned with hardware QOS traffic classes. In the normal case with
single traffic class and all queues in this class everything
works as is until the LLD enables multiple tcs.

To steer the skb we mask out the lower 4 bits of the priority
and allow the hardware to configure upto 15 distinct classes
of traffic. This is expected to be sufficient for most applications
at any rate it is more then the 8021Q spec designates and is
equal to the number of prio bands currently implemented in
the default qdisc.

This in conjunction with a userspace application such as
lldpad can be used to implement 8021Q transmission selection
algorithms one of these algorithms being the extended transmission
selection algorithm currently being used for DCB.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-01-19 23:31:10 -08:00
..
9p net: cleanup unused macros in net directory 2011-01-19 23:20:04 -08:00
802 net/802: add __rcu annotations 2010-10-25 13:09:44 -07:00
8021q 8021q: vlan device is lockless do not transfer real_num_{tx|rx}_queues 2010-11-28 10:47:19 -08:00
appletalk
atm Merge branch 'for-2.6.38' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq 2011-01-07 16:58:04 -08:00
ax25 net: ax25: fix information leak to userland harder 2011-01-12 00:34:49 -08:00
batman-adv batman-adv: Use "__attribute__" shortcut macros 2011-01-16 03:25:19 +01:00
bluetooth Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/padovan/bluetooth-next-2.6 2011-01-04 14:25:28 -05:00
bridge net: bridge: check the length of skb after nf_bridge_maybe_copy_header() 2011-01-06 11:33:05 -08:00
caif net: cleanup unused macros in net directory 2011-01-19 23:20:04 -08:00
can can: test size of struct sockaddr in sendmsg 2011-01-15 20:56:42 -08:00
ceph Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client 2011-01-13 10:25:24 -08:00
core net: implement mechanism for HW based QOS 2011-01-19 23:31:10 -08:00
dcb dcb: use after free in dcb_flushapp() 2011-01-06 11:16:54 -08:00
dccp Merge branch 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial 2011-01-13 10:05:56 -08:00
decnet net: cleanup unused macros in net directory 2011-01-19 23:20:04 -08:00
dns_resolver Net: dns_resolver: Makefile: Remove deprecated kbuild goal definitions 2010-11-22 08:16:10 -08:00
dsa net/dsa: don't use flush_scheduled_work() 2010-12-24 15:59:06 +01:00
econet Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 2010-12-17 12:27:22 -08:00
ethernet eth: fix new kernel-doc warning 2011-01-12 19:00:40 -08:00
ieee802154 net: RCU conversion of dev_getbyhwaddr() and arp_ioctl() 2010-12-08 10:07:24 -08:00
ipv4 Merge branch 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial 2011-01-13 10:05:56 -08:00
ipv6 Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 2011-01-18 16:28:31 -08:00
ipx BKL: introduce CONFIG_BKL. 2010-10-21 15:44:13 +02:00
irda Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 2010-12-26 22:37:05 -08:00
iucv [S390] irq: have detailed statistics for interrupt types 2011-01-05 12:47:25 +01:00
key net: return operator cleanup 2010-09-23 14:33:39 -07:00
l2tp Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 2010-12-08 13:47:38 -08:00
lapb Net: lapb: Makefile: Remove deprecated kbuild goal definitions 2010-11-22 08:16:14 -08:00
llc net: RCU conversion of dev_getbyhwaddr() and arp_ioctl() 2010-12-08 10:07:24 -08:00
mac80211 mac80211: use maximum number of AMPDU frames as default in BA RX 2011-01-13 15:46:45 -05:00
netfilter netfilter: ctnetlink: fix loop in ctnetlink_get_conntrack() 2011-01-13 17:03:39 +01:00
netlabel net: kill unused macros 2010-12-19 21:59:35 -08:00
netlink netlink: test for all flags of the NLM_F_DUMP composite 2011-01-09 16:25:03 -08:00
netrom
packet net: cleanup unused macros in net directory 2011-01-19 23:20:04 -08:00
phonet phonet: some signedness bugs 2011-01-10 13:33:17 -08:00
rds net: cleanup unused macros in net directory 2011-01-19 23:20:04 -08:00
rfkill rfkill: remove dead code 2010-11-15 13:24:06 -05:00
rose Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 2010-09-27 01:03:03 -07:00
rxrpc rxrpc: rxrpc_workqueue isn't used during memory reclaim 2011-01-14 09:25:11 -08:00
sched Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6 2011-01-14 13:25:30 -08:00
sctp Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 2010-12-17 12:27:22 -08:00
sunrpc Merge branch 'for-2.6.38' of git://linux-nfs.org/~bfields/linux 2011-01-14 13:17:26 -08:00
tipc tipc: update log.h re-include protection to reflect new name 2011-01-01 14:56:18 -08:00
unix af_unix: implement socket filter 2011-01-18 21:33:05 -08:00
wanrouter net: cleanup unused macros in net directory 2011-01-19 23:20:04 -08:00
wimax
wireless cfg80211: fix transposition of words in printk 2011-01-04 14:43:01 -05:00
x25 Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6 2010-12-08 13:47:38 -08:00
xfrm xfrm: check trunc_len in XFRMA_ALG_AUTH_TRUNC 2011-01-11 14:03:09 -08:00
compat.c net: Limit socket I/O iovec total length to INT_MAX. 2010-10-28 11:47:52 -07:00
Kconfig Merge branch 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial 2011-01-13 10:05:56 -08:00
Makefile net: Add batman-adv meshing protocol 2010-12-16 13:44:24 -08:00
nonet.c llseek: automatically add .llseek fop 2010-10-15 15:53:27 +02:00
socket.c pass default dentry_operations to mount_pseudo() 2011-01-12 20:03:43 -05:00
sysctl_net.c
TUNABLE