diff options
| author | Hui Su <sh_def@163.com> | 2026-08-18 18:33:36 +0800 |
|---|---|---|
| committer | Jens Axboe <axboe@kernel.dk> | 2026-08-25 11:51:38 -0600 |
| commit | 14572de82e5022899e5856008bc9cac97004a88c (patch) | |
| tree | 57e94133da53531a77dec8ac3921424ae7675d0c /scripts/const_structs.checkpatch | |
| parent | ba77efee1b95b4ad7559b1cdbe7cd7fa36dca95b (diff) | |
| download | linux-14572de82e5022899e5856008bc9cac97004a88c.tar.gz linux-14572de82e5022899e5856008bc9cac97004a88c.zip | |
io_uring/waitid: honor task_work cancellation
io_waitid_cb() may run through the fallback task_work path when
task_work_add() can no longer queue work to the originating task. The
fallback runs from a kworker and io_uring marks such task work as
canceled through tw.cancel.
io_waitid_cb() currently ignores tw.cancel and calls __do_wait().
waitid is task-context dependent: __do_wait() performs child lookup
relative to current, and the retry path also uses
current->signal->wait_chldexit. If the callback runs from the fallback
kworker, current is therefore not the task that submitted the request.
Honor tw.cancel before entering __do_wait(). Complete the request with
-ECANCELED and skip the siginfo copy, since canceled task work may run
without the submitting task's userspace execution context.
Keep the existing siginfo handling for normal waitid completion and
explicit cancellation.
Fixes: f31ecf671ddc ("io_uring: add IORING_OP_WAITID support")
Cc: stable@vger.kernel.org
Signed-off-by: Hui Su <sh_def@163.com>
Link: https://patch.msgid.link/20260818103336.1922818-2-sh_def@163.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'scripts/const_structs.checkpatch')
0 files changed, 0 insertions, 0 deletions
