All of lore.kernel.org
 help / color / mirror / Atom feed
From: Pavel Begunkov <asml.silence@gmail.com>
To: io-uring@vger.kernel.org
Cc: Jens Axboe <axboe@kernel.dk>, asml.silence@gmail.com
Subject: [PATCH liburing 2/2] io_uring: fix SQPOLL timeout-new test
Date: Sun,  3 Oct 2021 12:11:00 +0100	[thread overview]
Message-ID: <5faf8ade7e161931d76a47b809650e68a1b361ba.1633259449.git.asml.silence@gmail.com> (raw)
In-Reply-To: <cover.1633259449.git.asml.silence@gmail.com>

Happens pretty rarely, but there were cases when CQE waitinig in
test_return_before_timeout() time outs before the SQPOLL thread kicks in
and executes submitted requests, give SQPOLL a little bit more time.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 test/timeout-new.c | 22 ++++++++++++++--------
 1 file changed, 14 insertions(+), 8 deletions(-)

diff --git a/test/timeout-new.c b/test/timeout-new.c
index 19c5ac3..6efcfb4 100644
--- a/test/timeout-new.c
+++ b/test/timeout-new.c
@@ -53,14 +53,12 @@ static int test_return_before_timeout(struct io_uring *ring)
 	struct io_uring_cqe *cqe;
 	struct io_uring_sqe *sqe;
 	int ret;
+	bool retried = false;
 	struct __kernel_timespec ts;
 
-	sqe = io_uring_get_sqe(ring);
-	if (!sqe) {
-		fprintf(stderr, "%s: get sqe failed\n", __FUNCTION__);
-		return 1;
-	}
+	msec_to_ts(&ts, TIMEOUT_MSEC);
 
+	sqe = io_uring_get_sqe(ring);
 	io_uring_prep_nop(sqe);
 
 	ret = io_uring_submit(ring);
@@ -69,13 +67,21 @@ static int test_return_before_timeout(struct io_uring *ring)
 		return 1;
 	}
 
-	msec_to_ts(&ts, TIMEOUT_MSEC);
+again:
 	ret = io_uring_wait_cqe_timeout(ring, &cqe, &ts);
-	if (ret < 0) {
+	if (ret == -ETIME && (ring->flags & IORING_SETUP_SQPOLL) && !retried) {
+		/*
+		 * there is a small chance SQPOLL hasn't been waked up yet,
+		 * give it one more try.
+		 */
+		printf("warning: funky SQPOLL timing\n");
+		sleep(1);
+		retried = true;
+		goto again;
+	} else if (ret < 0) {
 		fprintf(stderr, "%s: timeout error: %d\n", __FUNCTION__, ret);
 		return 1;
 	}
-
 	io_uring_cqe_seen(ring, cqe);
 	return 0;
 }
-- 
2.33.0


  parent reply	other threads:[~2021-10-03 11:11 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-03 11:10 [PATCH liburing 0/2] timeout tests Pavel Begunkov
2021-10-03 11:10 ` [PATCH liburing 1/2] io_uring: test IORING_TIMEOUT_ETIME_SUCCESS Pavel Begunkov
2021-10-03 11:11 ` Pavel Begunkov [this message]
2021-10-03 13:07 ` [PATCH liburing 0/2] timeout tests Jens Axboe

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5faf8ade7e161931d76a47b809650e68a1b361ba.1633259449.git.asml.silence@gmail.com \
    --to=asml.silence@gmail.com \
    --cc=axboe@kernel.dk \
    --cc=io-uring@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.