APPLIED/CMT: [focal:linux-bluefield][PATCH 0/1] UBUNTU: SAUCE: Revert "block: don't ignore REQ_NOWAIT for direct IO"
kelsey.skunberg at canonical.com
Wed Oct 6 22:37:07 UTC 2021
Applied to Focal/bluefield master-next. Brad Figg provided the secondary ACK
on Mattermost. Thank you!
On 2021-10-06 19:06:42 , Wen-chien Jesse Sung wrote:
> BugLink: https://launchpad.net/bugs/1946219
> == Impact ==
> After configuration multipath with io uring virtio_blk module is stuck
> == Fix ==
> The issue was introduced by one of the stable updates merged in the base
> kernel Ubuntu-5.4.0-74.83. After reverting the commit the issue can not be
> reproduced any more.
> commit d6757b3e34831b7615edd3268b878536761f8b2e
> Author: Pavel Begunkov <asml.silence at gmail.com>
> Date: Fri Nov 20 17:10:28 2020 +0000
> block: don't ignore REQ_NOWAIT for direct IO
> BugLink: https://bugs.launchpad.net/bugs/1926490
> [ Upstream commit f8b78caf21d5bc3fcfc40c18898f9d52ed1451a5 ]
> If IOCB_NOWAIT is set on submission, then that needs to get propagated to
> REQ_NOWAIT on the block side. Otherwise we completely lose this
> information, and any issuer of IOCB_NOWAIT IO will potentially end up
> blocking on eg request allocation on the storage side.
> Signed-off-by: Pavel Begunkov <asml.silence at gmail.com>
> Signed-off-by: Jens Axboe <axboe at kernel.dk>
> Signed-off-by: Sasha Levin <sashal at kernel.org>
> Signed-off-by: Kamal Mostafa <kamal at canonical.com>
> Signed-off-by: Stefan Bader <stefan.bader at canonical.com>
> == Risk of Regression ==
> Medium. The commit has a good reason to be part of the stable updates. However
> the kernel passes Cert and Nvidia tests without the commit so reverting the
> commit should be an acceptable approach for now.
> Wen-chien Jesse Sung (1):
> UBUNTU: SAUCE: Revert "block: don't ignore REQ_NOWAIT for direct IO"
> fs/block_dev.c | 4 ----
> 1 file changed, 4 deletions(-)
> kernel-team mailing list
> kernel-team at lists.ubuntu.com
More information about the kernel-team