Files
linux/io_uring
Hagar Hemdan b571a36750 io_uring: fix possible deadlock in io_register_iowq_max_workers()
[ Upstream commit 73254a297c ]

The io_register_iowq_max_workers() function calls io_put_sq_data(),
which acquires the sqd->lock without releasing the uring_lock.
Similar to the commit 009ad9f0c6 ("io_uring: drop ctx->uring_lock
before acquiring sqd->lock"), this can lead to a potential deadlock
situation.

To resolve this issue, the uring_lock is released before calling
io_put_sq_data(), and then it is re-acquired after the function call.

This change ensures that the locks are acquired in the correct
order, preventing the possibility of a deadlock.

Suggested-by: Maximilian Heyne <mheyne@amazon.de>
Signed-off-by: Hagar Hemdan <hagarhem@amazon.com>
Link: https://lore.kernel.org/r/20240604130527.3597-1-hagarhem@amazon.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
2024-07-25 09:53:23 +02:00
..
2023-12-29 11:58:41 -08:00
2024-02-09 11:54:19 -07:00
2022-12-07 06:47:13 -07:00
2024-06-16 13:51:04 +02:00
2024-02-09 09:04:39 -07:00
2024-02-12 19:02:11 -07:00
2023-08-11 10:42:57 -06:00
2024-02-09 09:04:39 -07:00
2024-02-09 09:04:39 -07:00
2024-02-12 19:02:11 -07:00
2024-02-06 19:55:14 -07:00