diff options
| author |   Jens Axboe <axboe@kernel.dk> | 2021-07-27 10:50:31 -0600 |
|---|---|---|
| committer | Jens Axboe <axboe@kernel.dk> | 2021-07-28 07:24:38 -0600 |
| commit | ef04688871f3386b6d40ade8f5c664290420f819 (patch) | |
| tree | 1ae872efba193225d4d350cee41c8a2524001c75 /net/unix/af_unix.c | |
| parent | io_uring: always reissue from task_work context (diff) | |
| download | wireguard-linux-ef04688871f3386b6d40ade8f5c664290420f819.tar.xz wireguard-linux-ef04688871f3386b6d40ade8f5c664290420f819.zip | |
io_uring: don't block level reissue off completion path
Some setups, like SCSI, can throw spurious -EAGAIN off the softirq
completion path. Normally we expect this to happen inline as part
of submission, but apparently SCSI has a weird corner case where it
can happen as part of normal completions.
This should be solved by having the -EAGAIN bubble back up the stack
as part of submission, but previous attempts at this failed and we're
not just quite there yet. Instead we currently use REQ_F_REISSUE to
handle this case.
For now, catch it in io_rw_should_reissue() and prevent a reissue
from a bogus path.
Cc: stable@vger.kernel.org
Reported-by: Fabian Ebner <f.ebner@proxmox.com>
Tested-by: Fabian Ebner <f.ebner@proxmox.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'net/unix/af_unix.c')
0 files changed, 0 insertions, 0 deletions