[PATCH 5.12] io_uring: fix io_sq_offload_create error handling

[PATCH 5.12] io_uring: fix io_sq_offload_create error handling

[Pavel Begunkov]

Don't set IO_SQ_THREAD_SHOULD_STOP when io_sq_offload_create() has
failed on io_uring_alloc_task_context() but leave everything to
io_sq_thread_finish(), because currently io_sq_thread_finish()
hangs on trying to park it. That's great it stalls there, because
otherwise the following io_sq_thread_stop() would be skipped on
IO_SQ_THREAD_SHOULD_STOP check and the sqo would race for sqd with
freeing ctx.

A simple error injection gives something like this.

[  245.463955] INFO: task sqpoll-test-hang:523 blocked for more than 122 seconds.
[  245.463983] Call Trace:
[  245.463990]  __schedule+0x36b/0x950
[  245.464005]  schedule+0x68/0xe0
[  245.464013]  schedule_timeout+0x209/0x2a0
[  245.464032]  wait_for_completion+0x8b/0xf0
[  245.464043]  io_sq_thread_finish+0x44/0x1a0
[  245.464049]  io_uring_setup+0x9ea/0xc80
[  245.464058]  __x64_sys_io_uring_setup+0x16/0x20
[  245.464064]  do_syscall_64+0x38/0x50
[  245.464073]  entry_SYSCALL_64_after_hwframe+0x44/0xae

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 fs/io_uring.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 5505e19f1391..5ca3c70e6640 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -7860,10 +7860,9 @@ static int io_sq_offload_create(struct io_ring_ctx *ctx,
 			ret = PTR_ERR(tsk);
 			goto err;
 		}
-		ret = io_uring_alloc_task_context(tsk, ctx);
-		if (ret)
-			set_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state);
+
 		sqd->thread = tsk;
+		ret = io_uring_alloc_task_context(tsk, ctx);
 		wake_up_new_task(tsk);
 		if (ret)
 			goto err;
-- 
2.24.0

Re: [PATCH 5.12] io_uring: fix io_sq_offload_create error handling

[Jens Axboe]

On 3/8/21 10:30 AM, Pavel Begunkov wrote:
> Don't set IO_SQ_THREAD_SHOULD_STOP when io_sq_offload_create() has
> failed on io_uring_alloc_task_context() but leave everything to
> io_sq_thread_finish(), because currently io_sq_thread_finish()
> hangs on trying to park it. That's great it stalls there, because
> otherwise the following io_sq_thread_stop() would be skipped on
> IO_SQ_THREAD_SHOULD_STOP check and the sqo would race for sqd with
> freeing ctx.
> 
> A simple error injection gives something like this.
> 
> [  245.463955] INFO: task sqpoll-test-hang:523 blocked for more than 122 seconds.
> [  245.463983] Call Trace:
> [  245.463990]  __schedule+0x36b/0x950
> [  245.464005]  schedule+0x68/0xe0
> [  245.464013]  schedule_timeout+0x209/0x2a0
> [  245.464032]  wait_for_completion+0x8b/0xf0
> [  245.464043]  io_sq_thread_finish+0x44/0x1a0
> [  245.464049]  io_uring_setup+0x9ea/0xc80
> [  245.464058]  __x64_sys_io_uring_setup+0x16/0x20
> [  245.464064]  do_syscall_64+0x38/0x50
> [  245.464073]  entry_SYSCALL_64_after_hwframe+0x44/0xae

Applied, thanks.

-- 
Jens Axboe