diff options
| author | Pavel Begunkov <asml.silence@gmail.com> | 2024-03-15 15:29:51 +0000 |
|---|---|---|
| committer | Sasha Levin <sashal@kernel.org> | 2024-03-26 18:20:09 -0400 |
| commit | 4cbc5e93bf6eb238cfaa9b839083cd66d2914ce4 (patch) | |
| tree | 60b14eb445e8c0cd3a8ca69ddd16aad2fab0b5be /io_uring | |
| parent | 943c14ece95eb1cf98d477462aebcbfdfd714633 (diff) | |
| download | linux-stable-4cbc5e93bf6eb238cfaa9b839083cd66d2914ce4.tar.gz linux-stable-4cbc5e93bf6eb238cfaa9b839083cd66d2914ce4.tar.bz2 linux-stable-4cbc5e93bf6eb238cfaa9b839083cd66d2914ce4.zip | |
io_uring: fix poll_remove stalled req completion
[ Upstream commit 5e3afe580a9f5ca173a6bd55ffe10948796ef7e5 ]
Taking the ctx lock is not enough to use the deferred request completion
infrastructure, it'll get queued into the list but no one would expect
it there, so it will sit there until next io_submit_flush_completions().
It's hard to care about the cancellation path, so complete it via tw.
Fixes: ef7dfac51d8ed ("io_uring/poll: serialize poll linked timer start with poll removal")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/c446740bc16858f8a2a8dcdce899812f21d15f23.1710514702.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Diffstat (limited to 'io_uring')
| -rw-r--r-- | io_uring/poll.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/io_uring/poll.c b/io_uring/poll.c index 48ca0810a54a..5cf4fffe8b6c 100644 --- a/io_uring/poll.c +++ b/io_uring/poll.c @@ -981,7 +981,6 @@ int io_poll_remove(struct io_kiocb *req, unsigned int issue_flags) struct io_hash_bucket *bucket; struct io_kiocb *preq; int ret2, ret = 0; - struct io_tw_state ts = { .locked = true }; io_ring_submit_lock(ctx, issue_flags); preq = io_poll_find(ctx, true, &cd, &ctx->cancel_table, &bucket); @@ -1030,7 +1029,8 @@ found: req_set_fail(preq); io_req_set_res(preq, -ECANCELED, 0); - io_req_task_complete(preq, &ts); + preq->io_task_work.func = io_req_task_complete; + io_req_task_work_add(preq); out: io_ring_submit_unlock(ctx, issue_flags); if (ret < 0) { |