[ Upstream commit 52fde2c07d ]
The nvme specification only requires qword alignment for segment
descriptors, and the driver already guarantees that. The spec has always
allowed user data to be dword aligned, which is what the queue's
attribute is for, so relax the alignment requirement to that value.
While we could allow byte alignment for some controllers when using
SGLs, we still need to support PRP, and that only allows dword.
Fixes: 3b2a1ebceb ("nvme: set dma alignment to qword")
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 78288665b5 ]
In current implementation we set the non-mdts limits by calling
nvme_init_non_mdts_limits() from nvme_init_ctrl_finish().
This also tries to set the limits for the discovery controller which
has no I/O queues resulting in the warning message reported by the
nvme_log_error() when running blktest nvme/002: -
[ 2005.155946] run blktests nvme/002 at 2022-04-09 16:57:47
[ 2005.192223] loop: module loaded
[ 2005.196429] nvmet: adding nsid 1 to subsystem blktests-subsystem-0
[ 2005.200334] nvmet: adding nsid 1 to subsystem blktests-subsystem-1
<------------------------------SNIP---------------------------------->
[ 2008.958108] nvmet: adding nsid 1 to subsystem blktests-subsystem-997
[ 2008.962082] nvmet: adding nsid 1 to subsystem blktests-subsystem-998
[ 2008.966102] nvmet: adding nsid 1 to subsystem blktests-subsystem-999
[ 2008.973132] nvmet: creating discovery controller 1 for subsystem nqn.2014-08.org.nvmexpress.discovery for NQN testhostnqn.
*[ 2008.973196] nvme1: Identify(0x6), Invalid Field in Command (sct 0x0 / sc 0x2) MORE DNR*
[ 2008.974595] nvme nvme1: new ctrl: "nqn.2014-08.org.nvmexpress.discovery"
[ 2009.103248] nvme nvme1: Removing ctrl: NQN "nqn.2014-08.org.nvmexpress.discovery"
Move the call of nvme_init_non_mdts_limits() to nvme_scan_work() after
we verify that I/O queues are created since that is a converging point
for each transport where these limits are actually used.
1. FC :
nvme_fc_create_association()
...
nvme_fc_create_io_queues(ctrl);
...
nvme_start_ctrl()
nvme_scan_queue()
nvme_scan_work()
2. PCIe:-
nvme_reset_work()
...
nvme_setup_io_queues()
nvme_create_io_queues()
nvme_alloc_queue()
...
nvme_start_ctrl()
nvme_scan_queue()
nvme_scan_work()
3. RDMA :-
nvme_rdma_setup_ctrl
...
nvme_rdma_configure_io_queues
...
nvme_start_ctrl()
nvme_scan_queue()
nvme_scan_work()
4. TCP :-
nvme_tcp_setup_ctrl
...
nvme_tcp_configure_io_queues
...
nvme_start_ctrl()
nvme_scan_queue()
nvme_scan_work()
* nvme_scan_work()
...
nvme_validate_or_alloc_ns()
nvme_alloc_ns()
nvme_update_ns_info()
nvme_update_disk_info()
nvme_config_discard() <---
blk_queue_max_write_zeroes_sectors() <---
Signed-off-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit da42761181 ]
In nvme_alloc_admin_tags, the admin_q can be set to an error (typically
-ENOMEM) if the blk_mq_init_queue call fails to set up the queue, which
is checked immediately after the call. However, when we return the error
message up the stack, to nvme_reset_work the error takes us to
nvme_remove_dead_ctrl()
nvme_dev_disable()
nvme_suspend_queue(&dev->queues[0]).
Here, we only check that the admin_q is non-NULL, rather than not
an error or NULL, and begin quiescing a queue that never existed, leading
to bad / NULL pointer dereference.
Signed-off-by: Kyle Smith <kyles@hpe.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit a4a6f3c8f6 ]
nvme_mpath_init_identify() invoked from nvme_init_identify() fetches a
fresh ANA log from the ctrl. This is essential to have an up to date
path states for both existing namespaces and for those scan_work may
discover once the ctrl is up.
This happens in the following cases:
1) A new ctrl is being connected.
2) An existing ctrl is successfully reconnected.
3) An existing ctrl is being reset.
While in (1) ctrl->namespaces is empty, (2 & 3) may have namespaces, and
nvme_read_ana_log() may call nvme_update_ns_ana_state().
This result in a hang when the ANA state of an existing namespace changes
and makes the disk live: nvme_mpath_set_live() issues IO to the namespace
through the ctrl, which does NOT have IO queues yet.
See sample hang below.
Solution:
- nvme_update_ns_ana_state() to call set_live only if ctrl is live
- nvme_read_ana_log() call from nvme_mpath_init_identify()
therefore only fetches and parses the ANA log;
any erros in this process will fail the ctrl setup as appropriate;
- a separate function nvme_mpath_update()
is called in nvme_start_ctrl();
this parses the ANA log without fetching it.
At this point the ctrl is live,
therefore, disks can be set live normally.
Sample failure:
nvme nvme0: starting error recovery
nvme nvme0: Reconnecting in 10 seconds...
block nvme0n6: no usable path - requeuing I/O
INFO: task kworker/u8:3:312 blocked for more than 122 seconds.
Tainted: G E 5.14.5-1.el7.elrepo.x86_64 #1
Workqueue: nvme-wq nvme_tcp_reconnect_ctrl_work [nvme_tcp]
Call Trace:
__schedule+0x2a2/0x7e0
schedule+0x4e/0xb0
io_schedule+0x16/0x40
wait_on_page_bit_common+0x15c/0x3e0
do_read_cache_page+0x1e0/0x410
read_cache_page+0x12/0x20
read_part_sector+0x46/0x100
read_lba+0x121/0x240
efi_partition+0x1d2/0x6a0
bdev_disk_changed.part.0+0x1df/0x430
bdev_disk_changed+0x18/0x20
blkdev_get_whole+0x77/0xe0
blkdev_get_by_dev+0xd2/0x3a0
__device_add_disk+0x1ed/0x310
device_add_disk+0x13/0x20
nvme_mpath_set_live+0x138/0x1b0 [nvme_core]
nvme_update_ns_ana_state+0x2b/0x30 [nvme_core]
nvme_update_ana_state+0xca/0xe0 [nvme_core]
nvme_parse_ana_log+0xac/0x170 [nvme_core]
nvme_read_ana_log+0x7d/0xe0 [nvme_core]
nvme_mpath_init_identify+0x105/0x150 [nvme_core]
nvme_init_identify+0x2df/0x4d0 [nvme_core]
nvme_init_ctrl_finish+0x8d/0x3b0 [nvme_core]
nvme_tcp_setup_ctrl+0x337/0x390 [nvme_tcp]
nvme_tcp_reconnect_ctrl_work+0x24/0x40 [nvme_tcp]
process_one_work+0x1bd/0x360
worker_thread+0x50/0x3d0
Signed-off-by: Anton Eidelman <anton@lightbitslabs.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8832cf9221 ]
Any attempt to flush kernel-global WQs has possibility of deadlock
so we should simply stop using them, instead introduce nvmet_wq
which is the generic nvmet workqueue for work elements that
don't explicitly require a dedicated workqueue (by the mere fact
that they are using the system_wq).
Changes were done using the following replaces:
- s/schedule_work(/queue_work(nvmet_wq, /g
- s/schedule_delayed_work(/queue_delayed_work(nvmet_wq, /g
- s/flush_scheduled_work()/flush_workqueue(nvmet_wq)/g
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit bc360b0b16 ]
Add quirks to not fail the initialization and to have quick resume
latency after cold/warm reboot.
Signed-off-by: Monish Kumar R <monish.kumar.r@intel.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 66dd346b84 ]
Qemu unconditionally reports a UUID, which depending on the qemu version
is either all-null (which is incorrect but harmless) or contains a single
bit set for all controllers. In addition it can also optionally report
a eui64 which needs to be manually set. Disable namespace identifiers
for Qemu controlles entirely even if in some cases they could be set
correctly through manual intervention.
Reported-by: Luis Chamberlain <mcgrof@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 00ff400e6d ]
Add a quirk to disable using and exporting namespace identifiers for
controllers where they are broken beyond repair.
The most directly visible problem with non-unique namespace identifiers
is that they break the /dev/disk/by-id/ links, with the link for a
supposedly unique identifier now pointing to one of multiple possible
namespaces that share the same ID, and a somewhat random selection of
which one actually shows up.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 726be2c72e upstream.
commit 2f4c9ba23b ("nvme: export zoned namespaces without Zone Append
support read-only") marks zoned namespaces without append support
read-only. It does iso by setting NVME_NS_FORCE_RO in ns->flags in
nvme_update_zone_info and checking for that flag later in
nvme_update_disk_info to mark the disk as read-only.
But commit 73d90386b5 ("nvme: cleanup zone information initialization")
rearranged nvme_update_disk_info to be called before
nvme_update_zone_info and thus not marking the disk as read-only.
The call order cannot be just reverted because nvme_update_zone_info sets
certain queue parameters such as zone_write_granularity that depend on the
prior call to nvme_update_disk_info.
Remove the call to set_disk_ro in nvme_update_disk_info. and call
set_disk_ro after nvme_update_zone_info and nvme_update_disk_info to set
the permission for ZNS drives correctly. The same applies to the
multipath disk path.
Fixes: 73d90386b5 ("nvme: cleanup zone information initialization")
Signed-off-by: Pankaj Raghav <p.raghav@samsung.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 5974ea7ce0 upstream.
A NVMe subsystem with multiple controller can have private namespaces
that use the same NSID under some conditions:
"If Namespace Management, ANA Reporting, or NVM Sets are supported, the
NSIDs shall be unique within the NVM subsystem. If the Namespace
Management, ANA Reporting, and NVM Sets are not supported, then NSIDs:
a) for shared namespace shall be unique; and
b) for private namespace are not required to be unique."
Reference: Section 6.1.6 NSID and Namespace Usage; NVM Express 1.4c spec.
Make sure this specific setup is supported in Linux.
Fixes: 9ad1927a3b ("nvme: always search for namespace head")
Signed-off-by: Sungup Moon <sungup.moon@samsung.com>
[hch: refactored and fixed the controller vs subsystem based naming
conflict]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 841aee4d75 ]
Put NVMe/TCP sockets in their own class to avoid some lockdep warnings.
Sockets created by nvme-tcp are not exposed to user-space, and will not
trigger certain code paths that the general socket API exposes.
Lockdep complains about a circular dependency between the socket and
filesystem locks, because setsockopt can trigger a page fault with a
socket lock held, but nvme-tcp sends requests on the socket while file
system locks are held.
======================================================
WARNING: possible circular locking dependency detected
5.15.0-rc3 #1 Not tainted
------------------------------------------------------
fio/1496 is trying to acquire lock:
(sk_lock-AF_INET){+.+.}-{0:0}, at: tcp_sendpage+0x23/0x80
but task is already holding lock:
(&xfs_dir_ilock_class/5){+.+.}-{3:3}, at: xfs_ilock+0xcf/0x290 [xfs]
which lock already depends on the new lock.
other info that might help us debug this:
chain exists of:
sk_lock-AF_INET --> sb_internal --> &xfs_dir_ilock_class/5
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(&xfs_dir_ilock_class/5);
lock(sb_internal);
lock(&xfs_dir_ilock_class/5);
lock(sk_lock-AF_INET);
*** DEADLOCK ***
6 locks held by fio/1496:
#0: (sb_writers#13){.+.+}-{0:0}, at: path_openat+0x9fc/0xa20
#1: (&inode->i_sb->s_type->i_mutex_dir_key){++++}-{3:3}, at: path_openat+0x296/0xa20
#2: (sb_internal){.+.+}-{0:0}, at: xfs_trans_alloc_icreate+0x41/0xd0 [xfs]
#3: (&xfs_dir_ilock_class/5){+.+.}-{3:3}, at: xfs_ilock+0xcf/0x290 [xfs]
#4: (hctx->srcu){....}-{0:0}, at: hctx_lock+0x51/0xd0
#5: (&queue->send_mutex){+.+.}-{3:3}, at: nvme_tcp_queue_rq+0x33e/0x380 [nvme_tcp]
This annotation lets lockdep analyze nvme-tcp controlled sockets
independently of what the user-space sockets API does.
Link: https://lore.kernel.org/linux-nvme/CAHj4cs9MDYLJ+q+2_GXUK9HxFizv2pxUryUR0toX974M040z7g@mail.gmail.com/
Signed-off-by: Chris Leech <cleech@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e2724cb9f0 ]
nvme_subsys_check_duplicate_ids should needs to return an error if any of
the identifiers matches, not just if all of them match. But it does not
need to and should not look at the CSI value for this sanity check.
Rewrite the logic to be separate from nvme_ns_ids_equal and optimize it
by reducing duplicate checks for non-present identifiers.
Fixes: ed754e5dee ("nvme: track shared namespaces")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit fd8099e791 ]
Pass the actual nvme_ns_ids used for the comparison instead of the
ns_head that isn't needed and use a more descriptive function name.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Revert commit 626851e922 ("nvmet: make discovery NQN configurable");
the interface was deemed incorrect and will be replaced with a different
one.
Fixes: 626851e922 ("nvmet: make discovery NQN configurable")
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
As per NVMe/TCP specification (revision 1.0a, section 3.6.2.3)
Maximum Host to Controller Data length (MAXH2CDATA): Specifies the
maximum number of PDU-Data bytes per H2CData PDU in bytes. This value
is a multiple of dwords and should be no less than 4,096.
Current code sets H2CData PDU data_length to r2t_length,
it does not check MAXH2CDATA value. Fix this by setting H2CData PDU
data_length to min(req->h2cdata_left, queue->maxh2cdata).
Also validate MAXH2CDATA value returned by target in ICResp PDU,
if it is not a multiple of dword or if it is less than 4096 return
-EINVAL from nvme_tcp_init_connection().
Signed-off-by: Varun Prakash <varun@chelsio.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Commit e7d65803e2 ("nvme-multipath: revalidate paths during rescan")
introduced the NVME_NS_READY flag, which nvme_path_is_disabled() uses
to check if a path can be used or not. We also need to set this flag
for devices that fail the ZNS feature validation and which are available
through passthrough devices only to that they can be used in multipathing
setups.
Fixes: e7d65803e2 ("nvme-multipath: revalidate paths during rescan")
Reported-by: Kanchan Joshi <joshi.k@samsung.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Tested-by: Kanchan Joshi <joshi.k@samsung.com>
When a fabrics controller claims to support an invalidate metadata
configuration we already warn and disable metadata support. No need to
also return an error during revalidation.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Tested-by: Kanchan Joshi <joshi.k@samsung.com>
Various block drivers call blk_set_queue_dying to mark a disk as dead due
to surprise removal events, but since commit 8e141f9eb8 that doesn't
work given that the GD_DEAD flag needs to be set to stop I/O.
Replace the driver calls to blk_set_queue_dying with a new (and properly
documented) blk_mark_disk_dead API, and fold blk_set_queue_dying into the
only remaining caller.
Fixes: 8e141f9eb8 ("block: drain file system I/O on del_gendisk")
Reported-by: Markus Blöchl <markus.bloechl@ipetronik.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Link: https://lore.kernel.org/r/20220217075231.1140-1-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
AER is not backed by a real request, hence we should not incorrectly
assume that when failing to send a nvme command, it is a normal request
but rather check if this is an aer and if so complete the aer (similar
to the normal completion path).
Cc: stable@vger.kernel.org
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Add NVMe request completion trace in nvme_complete_batch_req() because
nvme:nvme_complete_req tracepoint is missing in case of request batched
completion.
Signed-off-by: Bean Huo <beanhuo@micron.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Controller deletion/reset, immediately followed by or concurrent with
a reconnect, is hard failing the connect attempt resulting in a
complete loss of connectivity to the controller.
In the connect request, fabrics looks for an existing controller with
the same address components and aborts the connect if a controller
already exists and the duplicate connect option isn't set. The match
routine filters out controllers that are dead or dying, so they don't
interfere with the new connect request.
When NVME_CTRL_DELETING_NOIO was added, it missed updating the state
filters in the nvmf_ctlr_matches_baseopts() routine. Thus, when in this
new state, it's seen as a live controller and fails the connect request.
Correct by adding the DELETING_NIO state to the match checks.
Fixes: ecca390e80 ("nvme: fix deadlock in disconnect during scan_work and/or ana_work")
Cc: <stable@vger.kernel.org> # v5.7+
Signed-off-by: Uday Shankar <ushankar@purestorage.com>
Reviewed-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
While nvme_rdma_submit_async_event_work is checking the ctrl and queue
state before preparing the AER command and scheduling io_work, in order
to fully prevent a race where this check is not reliable the error
recovery work must flush async_event_work before continuing to destroy
the admin queue after setting the ctrl state to RESETTING such that
there is no race .submit_async_event and the error recovery handler
itself changing the ctrl state.
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
While nvme_tcp_submit_async_event_work is checking the ctrl and queue
state before preparing the AER command and scheduling io_work, in order
to fully prevent a race where this check is not reliable the error
recovery work must flush async_event_work before continuing to destroy
the admin queue after setting the ctrl state to RESETTING such that
there is no race .submit_async_event and the error recovery handler
itself changing the ctrl state.
Tested-by: Chris Leech <cleech@redhat.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Unlike .queue_rq, in .submit_async_event drivers may not check the ctrl
readiness for AER submission. This may lead to a use-after-free
condition that was observed with nvme-tcp.
The race condition may happen in the following scenario:
1. driver executes its reset_ctrl_work
2. -> nvme_stop_ctrl - flushes ctrl async_event_work
3. ctrl sends AEN which is received by the host, which in turn
schedules AEN handling
4. teardown admin queue (which releases the queue socket)
5. AEN processed, submits another AER, calling the driver to submit
6. driver attempts to send the cmd
==> use-after-free
In order to fix that, add ctrl state check to validate the ctrl
is actually able to accept the AER submission.
This addresses the above race in controller resets because the driver
during teardown should:
1. change ctrl state to RESETTING
2. flush async_event_work (as well as other async work elements)
So after 1,2, any other AER command will find the
ctrl state to be RESETTING and bail out without submitting the AER.
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
The Intel P4500/P4600 SSDs do not report a subsystem NQN despite claiming
compliance to a standards version where reporting one is required.
Add the IGNORE_DEV_SUBNQN quirk to not fail the initialization of a
second such SSDs in a system.
Signed-off-by: Zheng Wu <wu.zheng@intel.com>
Signed-off-by: Ye Jinhe <jinhe.ye@intel.com>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
-----BEGIN PGP SIGNATURE-----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=k0IZ
-----END PGP SIGNATURE-----
Merge tag 'for-5.17/drivers-2022-01-11' of git://git.kernel.dk/linux-block
Pull block driver updates from Jens Axboe:
- mtip32xx pci cleanups (Bjorn)
- mtip32xx conversion to generic power management (Vaibhav)
- rsxx pci powermanagement cleanups (Bjorn)
- Remove the rsxx driver. This hardware never saw much adoption, and
it's been end of lifed for a while. (Christoph)
- MD pull request from Song:
- REQ_NOWAIT support (Vishal Verma)
- raid6 benchmark optimization (Dirk Müller)
- Fix for acct bioset (Xiao Ni)
- Clean up max_queued_requests (Mariusz Tkaczyk)
- PREEMPT_RT optimization (Davidlohr Bueso)
- Use default_groups in kobj_type (Greg Kroah-Hartman)
- Use attribute groups in pktcdvd and rnbd (Greg)
- NVMe pull request from Christoph:
- increment request genctr on completion (Keith Busch, Geliang
Tang)
- add a 'iopolicy' module parameter (Hannes Reinecke)
- print out valid arguments when reading from /dev/nvme-fabrics
(Hannes Reinecke)
- Use struct_group() in drbd (Kees)
- null_blk fixes (Ming)
- Get rid of congestion logic in pktcdvd (Neil)
- Floppy ejection hang fix (Tasos)
- Floppy max user request size fix (Xiongwei)
- Loop locking fix (Tetsuo)
* tag 'for-5.17/drivers-2022-01-11' of git://git.kernel.dk/linux-block: (32 commits)
md: use default_groups in kobj_type
md: Move alloc/free acct bioset in to personality
lib/raid6: Use strict priority ranking for pq gen() benchmarking
lib/raid6: skip benchmark of non-chosen xor_syndrome functions
md: fix spelling of "its"
md: raid456 add nowait support
md: raid10 add nowait support
md: raid1 add nowait support
md: add support for REQ_NOWAIT
md: drop queue limitation for RAID1 and RAID10
md/raid5: play nice with PREEMPT_RT
block/rnbd-clt-sysfs: use default_groups in kobj_type
pktcdvd: convert to use attribute groups
block: null_blk: only set set->nr_maps as 3 if active poll_queues is > 0
nvme: add 'iopolicy' module parameter
nvme: drop unused variable ctrl in nvme_setup_cmd
nvme: increment request genctr on completion
nvme-fabrics: print out valid arguments when reading from /dev/nvme-fabrics
block: remove the rsxx driver
rsxx: Drop PCI legacy power management
...
If command prep fails, current handling will orphan subsequent requests
in the list. Consider a simple example:
rqlist = [ 1 -> 2 ]
When prep for request '1' fails, it will be appended to the
'requeue_list', leaving request '2' disconnected from the original
rqlist and no longer tracked. Meanwhile, rqlist is still pointing to the
failed request '1' and will attempt to submit the unprepped command.
Fix this by updating the rqlist accordingly using the request list
helper functions.
Fixes: d62cbcf62f ("nvme: add support for mq_ops->queue_rqs()")
Signed-off-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220105170518.3181469-5-kbusch@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
While the 'iopolicy' sysfs attribute can be set at runtime, most
storage arrays prefer to use the 'round-robin' iopolicy per default.
We can use udev rules to set this, but is getting rather unwieldy
for rebranded arrays as we would have to update the udev rules
anytime a new array shows up, leading to the same mess we currently
have in multipathd for configuring the RDAC arrays.
Hence this patch adds a module parameter 'iopolicy' to allow the
admin to switch the default, and to do away with the need for a
udev rule here.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
The variable 'ctrl' became useless since the code using it was dropped
from nvme_setup_cmd() in the commit 292ddf67bbd5 ("nvme: increment
request genctr on completion"). Fix it to get rid of this compilation
warning in the nvme-5.17 branch:
drivers/nvme/host/core.c: In function ‘nvme_setup_cmd’:
drivers/nvme/host/core.c:993:20: warning: unused variable ‘ctrl’ [-Wunused-variable]
struct nvme_ctrl *ctrl = nvme_req(req)->ctrl;
^~~~
Fixes: 292ddf67bbd5 ("nvme: increment request genctr on completion")
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
The nvme request generation counter is intended to catch duplicate
completions. Incrementing the counter on submission means duplicates can
only be caught if the request tag is reallocated and dispatched prior to
the driver observing the corrupted CQE. Incrementing on completion
removes this window, making it possible to detect duplicate completions
in consecutive entries.
Signed-off-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Currently applications have a hard time figuring out which
nvme-over-fabrics arguments are supported for any given kernel;
the ioctl will return an error code on failure, and the application
has to guess whether this was due to an invalid argument or due
to a connection or controller error.
With this patch applications can read a list of supported
arguments by simply reading from /dev/nvme-fabrics, allowing
them to validate the connection string.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
This enables the block layer to send us a full plug list of requests
that need submitting. The block layer guarantees that they all belong
to the same queue, but we do have to check the hardware queue mapping
for each request.
If errors are encountered, leave them in the passed in list. Then the
block layer will handle them individually.
This is good for about a 4% improvement in peak performance, taking us
from 9.6M to 10M IOPS/core.
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a nvme_prep_rq() helper to setup a command, and nvme_queue_rq() is
adapted to use this helper.
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We'll need it for batched submit as well. Since we now have a copy
helper, get rid of the nvme_submit_cmd() wrapper.
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
nvmet_tcp_handle_req_failure needs to understand weather to prepare
for incoming data or the next pdu. However if we misidentify this, we
will wait for 0-length data, and queue the response although nvmet_req_init
already did that.
The particular command was namespace management command with no data,
which was incorrectly categorized as a command with incapsule data.
Also, add a code comment of what we are trying to do here.
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
A crash happens when trying to disconnect a reconnecting ctrl:
1) The network was cut off when the connection was just established,
scan work hang there waiting for some IOs complete. Those I/Os were
retried because we return BLK_STS_RESOURCE to blk in reconnecting.
2) After a while, I tried to disconnect this connection. This
procedure also hangs because it tried to obtain ctrl->scan_lock.
It should be noted that now we have switched the controller state
to NVME_CTRL_DELETING.
3) In nvme_check_ready(), we always return true when ctrl->state is
NVME_CTRL_DELETING, so those retrying I/Os were issued to the bottom
device which was already freed.
To fix this, when ctrl->state is NVME_CTRL_DELETING, issue cmd to bottom
device only when queue state is live. If not, return host path error to
the block layer
Signed-off-by: Ruozhu Li <liruozhu@huawei.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Set ana_log_size to 0 when ana_log_buf is freed to make sure
nvme_mpath_init_identify will do the right thing when retrying
after an earlier failure.
Signed-off-by: Hou Tao <houtao1@huawei.com>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
The write pointer in NVMe ZNS is invalid for a zone in zone state full.
The same also holds true for ZAC/ZBC.
The current behavior for NVMe is to simply propagate the wp reported by
the drive, even for full zones. Since the wp is invalid for a full zone,
the wp reported by the drive may be any value.
The way that the sd_zbc driver handles a full zone is to always report
the wp as zone start + zone len, regardless of what the drive reported.
null_blk also follows this convention.
Do the same for NVMe, so that a BLKREPORTZONE ioctl reports the write
pointer for a full zone in a consistent way, regardless of the interface
of the underlying zoned block device.
blkzone report before patch:
start: 0x000040000, len 0x040000, cap 0x03e000, wptr 0xfffffffffffbfff8
reset:0 non-seq:0, zcond:14(fu) [type: 2(SEQ_WRITE_REQUIRED)]
blkzone report after patch:
start: 0x000040000, len 0x040000, cap 0x03e000, wptr 0x040000 reset:0
non-seq:0, zcond:14(fu) [type: 2(SEQ_WRITE_REQUIRED)]
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
The only fabrics target that supports metadata handling through the
separate integrity buffer is RDMA. It is currently usable only if the
size is 8B per block and formatted for protection information. If an
rdma target were to export a namespace with a different format (ex:
4k+64B), the driver will not be able to submit valid read/write commands
for that namespace.
Suppress setting the metadata feature in the namespace so that the
gendisk capacity will be set to 0. This will prevent read/write access
through the block stack, but will continue to allow ioctl passthrough
commands.
Cc: Max Gurtovoy <mgurtovoy@nvidia.com>
Cc: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
The driver assigned nvme handle isn't persistent across reboots, so is
not enough information to match up where the collisions are occuring.
Add the subsys nqn string to the output so that it can more easily be
identified later.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=215099
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Remove the gendisk aregument to blk_execute_rq and blk_execute_rq_nowait
given that it is unused now. Also convert the boolean at_head parameter
to actually use the bool type while touching the prototype.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Just use the disk attached to the request_queue instead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20211126121802.2090656-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Submit I/O requests with the IOCB_NOWAIT flag set only if
the underlying filesystem supports it.
Fixes: 50a909db36 ("nvmet: use IOCB_NOWAIT for file-ns buffered I/O")
Signed-off-by: Maurizio Lombardi <mlombard@redhat.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Write Zeroes sets PRACT when block integrity is enabled (as it should),
but neglects to also set the reftag which is expected by reads. This
causes protection errors on reads.
Fix this by setting the reftag for type 1 and 2 (for type 3, reads will
not check the reftag).
Signed-off-by: Klaus Jensen <k.jensen@samsung.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
This particular Kioxia device times out and aborts I/O during any load,
but it's more easily observable with discards (fstrim).
The device gets to a state that is also not possible to use
"nvme set-feature" to disable APST.
Booting with nvme_core.default_ps_max_latency=0 solves the issue.
We had a dozen or so of these devices behaving this same way in
customer environments.
Signed-off-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Release the page frag cache when tearing down the io queues
Signed-off-by: Maurizio Lombardi <mlombard@redhat.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: John Meneghini <jmeneghi@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>