blk-mq: don't count completed flush data request as inflight in case of quiesce
authorMing Lei <ming.lei@redhat.com>
Fri, 1 Dec 2023 08:56:05 +0000 (16:56 +0800)
committerJens Axboe <axboe@kernel.dk>
Fri, 1 Dec 2023 14:34:47 +0000 (07:34 -0700)
commit0e4237ae8d159e3d28f3cd83146a46f576ffb586
tree5930c3cfef59c10ec74806e4d64d3bba08b98289
parent3649ff0a0b152b5f00e8f56a5ce0da0945aae278
blk-mq: don't count completed flush data request as inflight in case of quiesce

Request queue quiesce may interrupt flush sequence, and the original request
may have been marked as COMPLETE, but can't get finished because of
queue quiesce.

This way is fine from driver viewpoint, because flush sequence is block
layer concept, and it isn't related with driver.

However, driver(such as dm-rq) can call blk_mq_queue_inflight() to count &
drain inflight requests, then the wait & drain never gets done because
the completed & not-finished flush request is counted as inflight.

Fix this issue by not counting completed flush data request as inflight in
case of quiesce.

Cc: Mike Snitzer <snitzer@kernel.org>
Cc: David Jeffery <djeffery@redhat.com>
Cc: John Pittman <jpittman@redhat.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20231201085605.577730-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
block/blk-mq.c