Factour out a helper to initialize a simple single hw queue tag_set from
blk_mq_init_sq_queue. This will allow to phase out blk_mq_init_sq_queue
in favor of a more symmetric and general API.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Chaitanya Kulkarni <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
Commit 545fbd0775 ("rq-qos: fix missed wake-ups in rq_qos_throttle")
tried to fix a problem that a process could be sleeping in rq_qos_wait()
without anyone to wake it up. However the fix is not complete and the
following can still happen:
CPU1 (waiter1) CPU2 (waiter2) CPU3 (waker)
rq_qos_wait() rq_qos_wait()
acquire_inflight_cb() -> fails
acquire_inflight_cb() -> fails
completes IOs, inflight
decreased
prepare_to_wait_exclusive()
prepare_to_wait_exclusive()
has_sleeper = !wq_has_single_sleeper() -> true as there are two sleepers
has_sleeper = !wq_has_single_sleeper() -> true
io_schedule() io_schedule()
Deadlock as now there's nobody to wakeup the two waiters. The logic
automatically blocking when there are already sleepers is really subtle
and the only way to make it work reliably is that we check whether there
are some waiters in the queue when adding ourselves there. That way, we
are guaranteed that at least the first process to enter the wait queue
will recheck the waiting condition before going to sleep and thus
guarantee forward progress.
Fixes: 545fbd0775 ("rq-qos: fix missed wake-ups in rq_qos_throttle")
CC: [email protected]
Signed-off-by: Jan Kara <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
After commit 07173c3ec2 ("block: enable multipage bvecs"), a bvec can
have multiple pages. But bio_will_gap() still assumes one page bvec while
checking for merging. If the pages in the bvec go across the
seg_boundary_mask, this check for merging can potentially succeed if only
the 1st page is tested, and can fail if all the pages are tested.
Later, when SCSI builds the SG list the same check for merging is done in
__blk_segment_map_sg_merge() with all the pages in the bvec tested. This
time the check may fail if the pages in bvec go across the
seg_boundary_mask (but tested okay in bio_will_gap() earlier, so those
BIOs were merged). If this check fails, we end up with a broken SG list
for drivers assuming the SG list not having offsets in intermediate pages.
This results in incorrect pages written to the disk.
Fix this by returning the multi-page bvec when testing gaps for merging.
Cc: Jens Axboe <[email protected]>
Cc: Johannes Thumshirn <[email protected]>
Cc: Pavel Begunkov <[email protected]>
Cc: Ming Lei <[email protected]>
Cc: Tejun Heo <[email protected]>
Cc: "Matthew Wilcox (Oracle)" <[email protected]>
Cc: Jeffle Xu <[email protected]>
Cc: [email protected]
Cc: [email protected]
Fixes: 07173c3ec2 ("block: enable multipage bvecs")
Signed-off-by: Long Li <[email protected]>
Reviewed-by: Ming Lei <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
Although the original intent was to use blk_update_request() in stacking
block drivers only, it is used much more widely today. Reflect this in the
documentation block above this function. See also:
* commit 32fab448e5 ("block: add request update interface").
* commit 2e60e02297 ("block: clean up request completion API").
* commit ed6565e734 ("block: handle partial completions for special
payload requests").
Cc: Christoph Hellwig <[email protected]>
Cc: Ming Lei <[email protected]>
Cc: Hannes Reinecke <[email protected]>
Signed-off-by: Bart Van Assche <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
Provided the device driver does not implement dispatch budget accounting
(which only SCSI does) the loop in __blk_mq_do_dispatch_sched() pulls
requests from the IO scheduler as long as it is willing to give out any.
That defeats scheduling heuristics inside the scheduler by creating
false impression that the device can take more IO when it in fact
cannot.
For example with BFQ IO scheduler on top of virtio-blk device setting
blkio cgroup weight has barely any impact on observed throughput of
async IO because __blk_mq_do_dispatch_sched() always sucks out all the
IO queued in BFQ. BFQ first submits IO from higher weight cgroups but
when that is all dispatched, it will give out IO of lower weight cgroups
as well. And then we have to wait for all this IO to be dispatched to
the disk (which means lot of it actually has to complete) before the
IO scheduler is queried again for dispatching more requests. This
completely destroys any service differentiation.
So grab request tag for a request pulled out of the IO scheduler already
in __blk_mq_do_dispatch_sched() and do not pull any more requests if we
cannot get it because we are unlikely to be able to dispatch it. That
way only single request is going to wait in the dispatch list for some
tag to free.
Reviewed-by: Ming Lei <[email protected]>
Signed-off-by: Jan Kara <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
The error handling on a nullb->disk allocation currently jumps to
out_cleanup_disk that calls blk_cleanup_disk with a null pointer causing
a null pointer dereference issue. Fix this by jumping to out_cleanup_tags
instead.
Addresses-Coverity: ("Dereference after null check")
Fixes: 132226b301 ("null_blk: convert to blk_alloc_disk/blk_cleanup_disk")
Signed-off-by: Colin Ian King <[email protected]>
Reviewed-by: Chaitanya Kulkarni <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
Convert the null_blk driver to use the blk_alloc_disk and blk_cleanup_disk
helpers to simplify gendisk and request_queue allocation. Note that the
blk-mq mode is left with its own allocations scheme, to be handled later.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Ulf Hansson <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
Convert the brd driver to use the blk_alloc_disk and blk_cleanup_disk
helpers to simplify gendisk and request_queue allocation. This also
allows to remove the request_queue pointer in struct request_queue,
and to simplify the initialization as blk_cleanup_disk can be called
on any disk returned from blk_alloc_disk.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Reviewed-by: Ulf Hansson <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
Add a flag to indicate that __device_add_disk did grab a queue reference
so that disk_release only drops it if we actually had it. This sort
out one of the major pitfals with partially initialized gendisk that
a lot of drivers did get wrong or still do.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Reviewed-by: Luis Chamberlain <[email protected]>
Reviewed-by: Ulf Hansson <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
Christoph Hellwig