On Thu, 14 Oct 2021 16:17:04 +0800, Ming Lei wrote:
> request queue quiescing has been applied in lots of block drivers and
> block core from different/un-related code paths. So far, both quiesce
> and unquiesce changes the queue state unconditionally. This way has
> caused trouble, such as, driver is quiescing queue for its own purpose,
> however block core's queue quiesce may come because of elevator switch,
> updating nr_requests or other queue attributes, then un-expected
> unquiesce may come too early.
>
> [...]
Applied, thanks!
[1/6] nvme: add APIs for stopping/starting admin queue
commit: a277654bafb51fb8b4cf23550f15926bb02536f4
[2/6] nvme: apply nvme API to quiesce/unquiesce admin queue
commit: 6ca1d9027e0d9ce5604a3e28de89456a76138034
[3/6] nvme: prepare for pairing quiescing and unquiescing
commit: ebc9b95260151d966728cf0063b3b4e465f934d9
[4/6] nvme: paring quiesce/unquiesce
commit: 9e6a6b1212100148c109675e003369e3e219dbd9
[5/6] nvme: loop: clear NVME_CTRL_ADMIN_Q_STOPPED after admin queue is reallocated
commit: 1d35d519d8bf224ccdb43f9a235b8bda2d6d453c
[6/6] blk-mq: support concurrent queue quiesce/unquiesce
commit: e70feb8b3e6886c525c88943b5f1508d02f5a683
Best regards,
--
Jens Axboe
