High CPU utilization on "native_queued_spin_lock_slowpath" due to lock
contention is possible for mq-deadline and bfq IO schedulers
when nr_hw_queues is more than one.
It is because kblockd work queue can submit IO from all online CPUs
(through blk_mq_run_hw_queues()) even though only one hctx has pending
commands.
The elevator callback .has_work for mq-deadline and bfq scheduler considers
pending work if there are any IOs on request queue but it does not account
hctx context.
Add a per-hctx 'elevator_queued' count to the hctx to avoid triggering
the elevator even though there are no requests queued.
[jpg: Relocated atomic_dec() in dd_dispatch_request(), update commit message per Kashyap]
Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: John Garry <john.garry@huawei.com>
Tested-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For when using a shared sbitmap, no longer should the number of active
request queues per hctx be relied on for when judging how to share the tag
bitmap.
Instead maintain the number of active request queues per tag_set, and make
the judgement based on that.
Originally-from: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Tested-by: Don Brace<don.brace@microsemi.com> #SCSI resv cmds patches used
Tested-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The per-hctx nr_active value can no longer be used to fairly assign a share
of tag depth per request queue for when using a shared sbitmap, as it does
not consider that the tags are shared tags over all hctx's.
For this case, record the nr_active_requests per request_queue, and make
the judgement based on that value.
Co-developed-with: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Tested-by: Don Brace<don.brace@microsemi.com> #SCSI resv cmds patches used
Tested-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Some SCSI HBAs (such as HPSA, megaraid, mpt3sas, hisi_sas_v3 ..) support
multiple reply queues with single hostwide tags.
In addition, these drivers want to use interrupt assignment in
pci_alloc_irq_vectors(PCI_IRQ_AFFINITY). However, as discussed in [0],
CPU hotplug may cause in-flight IO completion to not be serviced when an
interrupt is shutdown. That problem is solved in commit bf0beec060
("blk-mq: drain I/O when all CPUs in a hctx are offline").
However, to take advantage of that blk-mq feature, the HBA HW queuess are
required to be mapped to that of the blk-mq hctx's; to do that, the HBA HW
queues need to be exposed to the upper layer.
In making that transition, the per-SCSI command request tags are no
longer unique per Scsi host - they are just unique per hctx. As such, the
HBA LLDD would have to generate this tag internally, which has a certain
performance overhead.
However another problem is that blk-mq assumes the host may accept
(Scsi_host.can_queue * #hw queue) commands. In commit 6eb045e092 ("scsi:
core: avoid host-wide host_busy counter for scsi_mq"), the Scsi host busy
counter was removed, which would stop the LLDD being sent more than
.can_queue commands; however, it should still be ensured that the block
layer does not issue more than .can_queue commands to the Scsi host.
To solve this problem, introduce a shared sbitmap per blk_mq_tag_set,
which may be requested at init time.
New flag BLK_MQ_F_TAG_HCTX_SHARED should be set when requesting the
tagset to indicate whether the shared sbitmap should be used.
Even when BLK_MQ_F_TAG_HCTX_SHARED is set, a full set of tags and requests
are still allocated per hctx; the reason for this is that if tags and
requests were only allocated for a single hctx - like hctx0 - it may break
block drivers which expect a request be associated with a specific hctx,
i.e. not always hctx0. This will introduce extra memory usage.
This change is based on work originally from Ming Lei in [1] and from
Bart's suggestion in [2].
[0] https://lore.kernel.org/linux-block/alpine.DEB.2.21.1904051331270.1802@nanos.tec.linutronix.de/
[1] https://lore.kernel.org/linux-block/20190531022801.10003-1-ming.lei@redhat.com/
[2] https://lore.kernel.org/linux-block/ff77beff-5fd9-9f05-12b6-826922bace1f@huawei.com/T/#m3db0a602f095cbcbff27e9c884d6b4ae826144be
Signed-off-by: John Garry <john.garry@huawei.com>
Tested-by: Don Brace<don.brace@microsemi.com> #SCSI resv cmds patches used
Tested-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Introduce pointers for the blk_mq_tags regular and reserved bitmap tags,
with the goal of later being able to use a common shared tag bitmap across
all HW contexts in a set.
Signed-off-by: John Garry <john.garry@huawei.com>
Tested-by: Don Brace<don.brace@microsemi.com> #SCSI resv cmds patches used
Tested-by: Douglas Gilbert <dgilbert@interlog.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pass hctx/tagset flags argument down to blk_mq_init_tags() and
blk_mq_free_tags() for selective init/free.
For now, make it include the alloc policy flag, which can be evaluated
when needed (in blk_mq_init_tags()).
Signed-off-by: John Garry <john.garry@huawei.com>
Tested-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The function does not set the depth, but rather transitions from
shared to non-shared queues and vice versa.
So rename it to blk_mq_update_tag_set_shared() to better reflect
its purpose.
[jpg: take out some unrelated changes in blk_mq_init_bitmap_tags()]
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: John Garry <john.garry@huawei.com>
Tested-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
BLK_MQ_F_TAG_SHARED actually means that tags is shared among request
queues, all of which should belong to LUNs attached to same HBA.
So rename it to make the point explicitly.
[jpg: rebase a few times, add rnbd-clt.c change]
Suggested-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Tested-by: Douglas Gilbert <dgilbert@interlog.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
c616cbee97 ("blk-mq: punt failed direct issue to dispatch list") supposed
to add request which has been through ->queue_rq() to the hw queue dispatch
list, however it adds request running out of budget or driver tag to hw queue
too. This way basically bypasses request merge, and causes too many request
dispatched to LLD, and system% is unnecessary increased.
Fixes this issue by adding request not through ->queue_rq into sw/scheduler
queue, and this way is safe because no ->queue_rq is called on this request
yet.
High %system can be observed on Azure storvsc device, and even soft lock
is observed. This patch reduces %system during heavy sequential IO,
meantime decreases soft lockup risk.
Fixes: c616cbee97 ("blk-mq: punt failed direct issue to dispatch list")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
SCHED_RESTART code path is relied to re-run queue for dispatch requests
in hctx->dispatch. Meantime the SCHED_RSTART flag is checked when adding
requests to hctx->dispatch.
memory barriers have to be used for ordering the following two pair of OPs:
1) adding requests to hctx->dispatch and checking SCHED_RESTART in
blk_mq_dispatch_rq_list()
2) clearing SCHED_RESTART and checking if there is request in hctx->dispatch
in blk_mq_sched_restart().
Without the added memory barrier, either:
1) blk_mq_sched_restart() may miss requests added to hctx->dispatch meantime
blk_mq_dispatch_rq_list() observes SCHED_RESTART, and not run queue in
dispatch side
or
2) blk_mq_dispatch_rq_list still sees SCHED_RESTART, and not run queue
in dispatch side, meantime checking if there is request in
hctx->dispatch from blk_mq_sched_restart() is missed.
IO hang in ltp/fs_fill test is reported by kernel test robot:
https://lkml.org/lkml/2020/7/26/77
Turns out it is caused by the above out-of-order OPs. And the IO hang
can't be observed any more after applying this patch.
Fixes: bd166ef183 ("blk-mq-sched: add framework for MQ capable IO schedulers")
Reported-by: kernel test robot <rong.a.chen@intel.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: David Jeffery <djeffery@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Fix a kernel-doc warning in block/blk-mq.c:
../block/blk-mq.c:1844: warning: Function parameter or member 'at_head' not described in 'blk_mq_request_bypass_insert'
Fixes: 01e99aeca3 ("blk-mq: insert passthrough request into hctx->dispatch directly")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: André Almeida <andrealmeid@collabora.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: linux-block@vger.kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
-----BEGIN PGP SIGNATURE-----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=abJG
-----END PGP SIGNATURE-----
Merge tag 'for-5.9/block-20200802' of git://git.kernel.dk/linux-block
Pull core block updates from Jens Axboe:
"Good amount of cleanups and tech debt removals in here, and as a
result, the diffstat shows a nice net reduction in code.
- Softirq completion cleanups (Christoph)
- Stop using ->queuedata (Christoph)
- Cleanup bd claiming (Christoph)
- Use check_events, moving away from the legacy media change
(Christoph)
- Use inode i_blkbits consistently (Christoph)
- Remove old unused writeback congestion bits (Christoph)
- Cleanup/unify submission path (Christoph)
- Use bio_uninit consistently, instead of bio_disassociate_blkg
(Christoph)
- sbitmap cleared bits handling (John)
- Request merging blktrace event addition (Jan)
- sysfs add/remove race fixes (Luis)
- blk-mq tag fixes/optimizations (Ming)
- Duplicate words in comments (Randy)
- Flush deferral cleanup (Yufen)
- IO context locking/retry fixes (John)
- struct_size() usage (Gustavo)
- blk-iocost fixes (Chengming)
- blk-cgroup IO stats fixes (Boris)
- Various little fixes"
* tag 'for-5.9/block-20200802' of git://git.kernel.dk/linux-block: (135 commits)
block: blk-timeout: delete duplicated word
block: blk-mq-sched: delete duplicated word
block: blk-mq: delete duplicated word
block: genhd: delete duplicated words
block: elevator: delete duplicated word and fix typos
block: bio: delete duplicated words
block: bfq-iosched: fix duplicated word
iocost_monitor: start from the oldest usage index
iocost: Fix check condition of iocg abs_vdebt
block: Remove callback typedefs for blk_mq_ops
block: Use non _rcu version of list functions for tag_set_list
blk-cgroup: show global disk stats in root cgroup io.stat
blk-cgroup: make iostat functions visible to stat printing
block: improve discard bio alignment in __blkdev_issue_discard()
block: change REQ_OP_ZONE_RESET and REQ_OP_ZONE_RESET_ALL to be odd numbers
block: defer flush request no matter whether we have elevator
block: make blk_timeout_init() static
block: remove retry loop in ioc_release_fn()
block: remove unnecessary ioc nested locking
block: integrate bd_start_claiming into __blkdev_get
...
Drop the repeated word "the".
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
tag_set_list is only accessed under the tag_set_lock lock. There is
no need for using the _rcu list functions.
The _rcu list function were introduced to allow read access to the
tag_set_list protected under RCU, see 705cda97ee ("blk-mq: Make it
safe to use RCU to iterate over blk_mq_tag_set.tag_list") and
05b7941394 ("Revert "blk-mq: don't handle TAG_SHARED in restart"").
Those changes got reverted later but the cleanup commit missed a
couple of places to undo the changes.
Fixes: 97889f9ac2 ("blk-mq: remove synchronize_rcu() from blk_mq_del_queue_tag_set()"
Signed-off-by: Daniel Wagner <dwagner@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We've already validated the 'q->elevator' before calling
->ops.completed_request() in blk_mq_sched_completed_request(), thus no
need to validate rq->internal_tag again. Rmove it.
Signed-off-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move .nr_active update and request assignment into blk_mq_get_driver_tag(),
all are good to do during getting driver tag.
Meantime blk-flush related code is simplified and flush request needn't
to update the request table manually any more.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Current handling of q->mq_ops->queue_rq result is a bit ugly:
- two branches which needs to 'continue' have to check if the
dispatch local list is empty, otherwise one bad request may
be retrieved via 'rq = list_first_entry(list, struct request, queuelist);'
- the branch of 'if (unlikely(ret != BLK_STS_OK))' isn't easy
to follow, since it is actually one error branch.
Streamline this handling, so the code becomes more readable, meantime
potential kernel oops can be avoided in case that the last request in
local dispatch list is failed.
Fixes: fc17b6534e ("blk-mq: switch ->queue_rq return value to blk_status_t")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
dm-multipath is the only user of blk_mq_queue_inflight(). When
dm-multipath calls blk_mq_queue_inflight() to check if it has
outstanding IO it can get a false negative. The reason for this is
blk_mq_rq_inflight() doesn't consider requests that are no longer
MQ_RQ_IN_FLIGHT but that are now MQ_RQ_COMPLETE (->complete isn't
called or finished yet) as "inflight".
This causes request-based dm-multipath's dm_wait_for_completion() to
return before all outstanding dm-multipath requests have actually
completed. This breaks DM multipath's suspend functionality because
blk-mq requests complete after DM's suspend has finished -- which
shouldn't happen.
Fix this by considering any request not in the MQ_RQ_IDLE state
(so either MQ_RQ_COMPLETE or MQ_RQ_IN_FLIGHT) as "inflight" in
blk_mq_rq_inflight().
Fixes: 3c94d83cb3 ("blk-mq: change blk_mq_queue_busy() to blk_mq_queue_inflight()")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The make_request_fn is a little weird in that it sits directly in
struct request_queue instead of an operation vector. Replace it with
a block_device_operations method called submit_bio (which describes much
better what it does). Also remove the request_queue argument to it, as
the queue can be derived pretty trivially from the bio.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The queue can be trivially derived from the bio, so pass one less
argument.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move .nr_active update and request assignment into blk_mq_get_driver_tag(),
all are good to do during getting driver tag.
Meantime blk-flush related code is simplified and flush request needn't
to update the request table manually any more.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
It is used by blk-mq.c only, so move it to the source file.
Suggested-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk_mq_get_driver_tag() is only used by blk-mq.c and is supposed to
stay in blk-mq.c, so move it and preparing for cleanup code of
get/put driver tag.
Meantime hctx_may_queue() is moved to header file and it is fine
since it is defined as inline always.
No functional change.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
More and more drivers want to get batching requests queued from
block layer, such as mmc, and tcp based storage drivers. Also
current in-tree users have virtio-scsi, virtio-blk and nvme.
For none, we already support batching dispatch.
But for io scheduler, every time we just take one request from scheduler
and pass the single request to blk_mq_dispatch_rq_list(). This way makes
batching dispatch not possible when io scheduler is applied. One reason
is that we don't want to hurt sequential IO performance, becasue IO
merge chance is reduced if more requests are dequeued from scheduler
queue.
Try to support batching dispatch for io scheduler by starting with the
following simple approach:
1) still make sure we can get budget before dequeueing request
2) use hctx->dispatch_busy to evaluate if queue is busy, if it is busy
we fackback to non-batching dispatch, otherwise dequeue as many as
possible requests from scheduler, and pass them to blk_mq_dispatch_rq_list().
Wrt. 2), we use similar policy for none, and turns out that SCSI SSD
performance got improved much.
In future, maybe we can develop more intelligent algorithem for batching
dispatch.
Baolin has tested this patch and found that MMC performance is improved[3].
[1] https://lore.kernel.org/linux-block/20200512075501.GF1531898@T590/#r
[2] https://lore.kernel.org/linux-block/fe6bd8b9-6ed9-b225-f80c-314746133722@grimberg.me/
[3] https://lore.kernel.org/linux-block/CADBw62o9eTQDJ9RvNgEqSpXmg6Xcq=2TxH0Hfxhp29uF2W=TXA@mail.gmail.com/
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Tested-by: Baolin Wang <baolin.wang7@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Baolin Wang <baolin.wang7@gmail.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pass obtained budget count to blk_mq_dispatch_rq_list(), and prepare
for supporting fully batching submission.
With the obtained budget count, it is easier to put extra budgets
in case of .queue_rq failure.
Meantime remove the old 'got_budget' parameter.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Tested-by: Baolin Wang <baolin.wang7@gmail.com>
Reviewed-by: Christoph Hellwig <hch@infradead.org>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Baolin Wang <baolin.wang7@gmail.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When BLK_STS_RESOURCE or BLK_STS_DEV_RESOURCE is returned from
.queue_rq, the 'list' variable always holds this rq which isn't
queued to LLD successfully.
So blk_mq_dispatch_rq_list() always returns false from the branch
of '!list_empty(list)'.
No functional change.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Tested-by: Baolin Wang <baolin.wang7@gmail.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Christoph Hellwig <hch@infradead.org>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Baolin Wang <baolin.wang7@gmail.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move code for getting driver tag and budget into one helper, so
blk_mq_dispatch_rq_list gets a bit simplified, and easier to read.
Meantime move updating of 'no_tag' and 'no_budget_available' into
the branch for handling partial dispatch because that is exactly
consumer of the two local variables.
Also rename the parameter of 'got_budget' as 'ask_budget'.
No functional change.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Tested-by: Baolin Wang <baolin.wang7@gmail.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Christoph Hellwig <hch@infradead.org>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Baolin Wang <baolin.wang7@gmail.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
All requests in the 'list' of blk_mq_dispatch_rq_list belong to same
hctx, so it is better to pass hctx instead of request queue, because
blk-mq's dispatch target is hctx instead of request queue.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Tested-by: Baolin Wang <baolin.wang7@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Baolin Wang <baolin.wang7@gmail.com>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk-mq budget is abstract from scsi's device queue depth, and it is
always per-request-queue instead of hctx.
It can be quite absurd to get a budget from one hctx, then dequeue a
request from scheduler queue, and this request may not belong to this
hctx, at least for bfq and deadline.
So fix the mess and always pass request queue to get/put budget
callback.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Tested-by: Baolin Wang <baolin.wang7@gmail.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Baolin Wang <baolin.wang7@gmail.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Just check for a non-NULL elevator directly to make the code more clear.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
It is natural to release driver tag when this request is completed by
LLD or device since its purpose is for LLD use.
One big benefit is that the released tag can be re-used quicker since
bio_endio() may take too long.
Meantime we don't need to release driver tag for flush request.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This is a variant of blk_mq_complete_request_remote that only completes
the request if it needs to be bounced to another CPU or a softirq. If
the request can be completed locally the function returns false and lets
the driver complete it without requring and indirect function call.
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a helper to decide if we can complete locally or need an IPI.
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We don't really care if we get migrated during the I/O completion.
In the worth case we either perform an IPI that wasn't required, or
complete the request on a CPU which we just migrated off.
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move the call to blk_should_fake_timeout out of blk_mq_complete_request
and into the drivers, skipping call sites that are obvious error
handlers, and remove the now superflous blk_mq_force_complete_rq helper.
This ensures we don't keep injecting errors into completions that just
terminate the Linux request after the hardware has been reset or the
command has been aborted.
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Both the softirq path for single queue devices and the multi-queue
completion handler share the same logic to figure out if we need an
IPI for the completion and eventually issue it. Merge the two
versions into a single unified code path.
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Let the compile optimize out the entire IPI path, given that we are
obviously not going to use it.
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Even for single queue devices there is no point in offloading a polled
completion to the softirq, given that blk_mq_force_complete_rq is called
from the polling thread in that case and thus there are no starvation
issues.
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
By open coding raise_blk_irq in the only caller, and replacing the
ifdef CONFIG_SMP with an IS_ENABLED check the flow in the caller
can be significantly simplified.
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a helper to deduplicate the logic that raises the block softirq.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
__blk_complete_request is only called from the blk-mq code, and
duplicates a lot of code from blk-mq.c. Move it there to prepare
for better code sharing and simplifications.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There is an issue when tune the number for read and write queues,
if the total queue count was not changed. The hctx->type cannot
be updated, since __blk_mq_update_nr_hw_queues will return directly
if the total queue count has not been changed.
Reproduce:
dmesg | grep "default/read/poll"
[ 2.607459] nvme nvme0: 48/0/0 default/read/poll queues
cat /sys/kernel/debug/block/nvme0n1/hctx*/type | sort | uniq -c
48 default
tune the write queues to 24:
echo 24 > /sys/module/nvme/parameters/write_queues
echo 1 > /sys/block/nvme0n1/device/reset_controller
dmesg | grep "default/read/poll"
[ 433.547235] nvme nvme0: 24/24/0 default/read/poll queues
cat /sys/kernel/debug/block/nvme0n1/hctx*/type | sort | uniq -c
48 default
The driver's hardware queue mapping is not same as block layer.
Signed-off-by: Weiping Zhang <zhangweiping@didiglobal.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Allocation of the driver tag in the case of using a scheduler shares very
little code with the "normal" tag allocation. Split out a new helper to
streamline this path, and untangle it from the complex normal tag
allocation.
This way also avoids to fail driver tag allocation because of inactive hctx
during cpu hotplug, and fixes potential hang risk.
Fixes: bf0beec060 ("blk-mq: drain I/O when all CPUs in a hctx are offline")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: John Garry <john.garry@huawei.com>
Cc: Dongli Zhang <dongli.zhang@oracle.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Most of blk-mq drivers depend on managed IRQ's auto-affinity to setup
up queue mapping. Thomas mentioned the following point[1]:
"That was the constraint of managed interrupts from the very beginning:
The driver/subsystem has to quiesce the interrupt line and the associated
queue _before_ it gets shutdown in CPU unplug and not fiddle with it
until it's restarted by the core when the CPU is plugged in again."
However, current blk-mq implementation doesn't quiesce hw queue before
the last CPU in the hctx is shutdown. Even worse, CPUHP_BLK_MQ_DEAD is a
cpuhp state handled after the CPU is down, so there isn't any chance to
quiesce the hctx before shutting down the CPU.
Add new CPUHP_AP_BLK_MQ_ONLINE state to stop allocating from blk-mq hctxs
where the last CPU goes away, and wait for completion of in-flight
requests. This guarantees that there is no inflight I/O before shutting
down the managed IRQ.
Add a BLK_MQ_F_STACKING and set it for dm-rq and loop, so we don't need
to wait for completion of in-flight requests from these drivers to avoid
a potential dead-lock. It is safe to do this for stacking drivers as those
do not use interrupts at all and their I/O completions are triggered by
underlying devices I/O completion.
[1] https://lore.kernel.org/linux-block/alpine.DEB.2.21.1904051331270.1802@nanos.tec.linutronix.de/
[hch: different retry mechanism, merged two patches, minor cleanups]
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
blk_mq_alloc_request_hctx is only used for NVMeoF connect commands, so
tailor it to the specific requirements, and don't bother the general
fast path code with its special twinkles.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.de
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Replace various magic -1 constants for tags with BLK_MQ_NO_TAG.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
To prepare for wider use of this constant give it a more applicable name.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Don't split request initialization between __blk_mq_alloc_request and
blk_mq_rq_ctx_init. Also remove the op argument as it can be derived
from the blk_mq_alloc_data structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The bio argument is entirely unused, and the request_queue can be passed
through the alloc_data, given that it needs to be filled out for the
low-level tag allocation anyway. Also rename the function to
__blk_mq_alloc_request as the switch between get and alloc in the call
chains is rather confusing.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Kashyap Desai
John Garry
Hannes Reinecke
Ming Lei
Randy Dunlap
Linus Torvalds
Daniel Wagner
Baolin Wang
Jens Axboe
Christoph Hellwig
Weiping Zhang