[PATCH for-6.1] io_uring: disallow self-propelled ring polling

[PATCH for-6.1] io_uring: disallow self-propelled ring polling

[Pavel Begunkov]

When we post a CQE we wake all ring pollers as it normally should be.
However, if a CQE was generated by a multishot poll request targeting
its own ring, it'll wake that request up, which will make it to post
a new CQE, which will wake the request and so on until it exhausts all
CQ entries.

Don't allow multishot polling io_uring files but downgrade them to
oneshots, which was always stated as a correct behaviour that the
userspace should check for.

Cc: stable@vger.kernel.org
Fixes: aa43477b04025 ("io_uring: poll rework")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 io_uring/poll.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/io_uring/poll.c b/io_uring/poll.c
index f500506984ec..9fe31c1ebb7b 100644
--- a/io_uring/poll.c
+++ b/io_uring/poll.c
@@ -239,6 +239,8 @@ static int io_poll_check_events(struct io_kiocb *req, bool *locked)
 			continue;
 		if (req->apoll_events & EPOLLONESHOT)
 			return IOU_POLL_DONE;
+		if (io_is_uring_fops(req->file))
+			return IOU_POLL_DONE;
 
 		/* multishot, just fill a CQE and proceed */
 		if (!(req->flags & REQ_F_APOLL_MULTISHOT)) {
-- 
2.38.1

Re: [PATCH for-6.1] io_uring: disallow self-propelled ring polling

[Jens Axboe]

On Fri, 18 Nov 2022 15:41:41 +0000, Pavel Begunkov wrote:
> When we post a CQE we wake all ring pollers as it normally should be.
> However, if a CQE was generated by a multishot poll request targeting
> its own ring, it'll wake that request up, which will make it to post
> a new CQE, which will wake the request and so on until it exhausts all
> CQ entries.
> 
> Don't allow multishot polling io_uring files but downgrade them to
> oneshots, which was always stated as a correct behaviour that the
> userspace should check for.
> 
> [...]

Applied, thanks!

[1/1] io_uring: disallow self-propelled ring polling
      commit: 7fdbc5f014c3f71bc44673a2d6c5bb2d12d45f25

Best regards,
-- 
Jens Axboe