diff mbox series

[09/11] io_uring: fix skipping disabling sqo on exec

Message ID 4bb1c422df133f0e883fefe221ffc866bfce7aa9.1611659564.git.asml.silence@gmail.com
State New
Headers show
Series io_uring for-stable | expand

Commit Message

Pavel Begunkov Jan. 26, 2021, 11:17 a.m. UTC 
[ Upstream commit 0b5cd6c32b14413bf87e10ee62be3162588dcbe6 ]

If there are no requests at the time __io_uring_task_cancel() is called,
tctx_inflight() returns zero and and it terminates not getting a chance
to go through __io_uring_files_cancel() and do
io_disable_sqo_submit(). And we absolutely want them disabled by the
time cancellation ends.

Cc: stable@vger.kernel.org # 5.5+
Reported-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Fixes: d9d05217cb69 ("io_uring: stop SQPOLL submit on creator's death")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/io_uring.c | 4 ++++
 1 file changed, 4 insertions(+)
diff mbox series

Patch

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 12fa5e09cefa..5ead8b6aeda2 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -8886,6 +8886,10 @@  void __io_uring_task_cancel(void)
 	/* make sure overflow events are dropped */
 	atomic_inc(&tctx->in_idle);
 
+	/* trigger io_disable_sqo_submit() */
+	if (tctx->sqpoll)
+		__io_uring_files_cancel(NULL);
+
 	do {
 		/* read completions before cancelations */
 		inflight = tctx_inflight(tctx);