From: "Stefan Bühler" <source@stbuehler.de>
To: Jens Axboe <axboe@kernel.dk>,
linux-block@vger.kernel.org, linux-fsdevel@vger.kernel.org
Subject: [PATCH barrier cleanup v1 3/7] io_uring: remove unnecessary barrier before reading cq head
Date: Wed, 24 Apr 2019 23:54:18 +0200 [thread overview]
Message-ID: <20190424215422.7404-3-source@stbuehler.de> (raw)
In-Reply-To: <20190424215422.7404-1-source@stbuehler.de>
The memory operations before reading cq head are unrelated and we
don't care about their order.
Document that the control dependency in combination with READ_ONCE and
WRITE_ONCE forms a barrier we need.
Signed-off-by: Stefan Bühler <source@stbuehler.de>
---
fs/io_uring.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 1f4419f38ef1..2c101230df71 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -431,8 +431,11 @@ static struct io_uring_cqe *io_get_cqring(struct io_ring_ctx *ctx)
unsigned tail;
tail = ctx->cached_cq_tail;
- /* See comment at the top of the file */
- smp_rmb();
+ /*
+ * writes to the cq entry need to come after reading head; the
+ * control dependency is enough as we're using WRITE_ONCE to
+ * fill the cq entry
+ */
if (tail - READ_ONCE(ring->r.head) == ring->ring_entries)
return NULL;
--
2.20.1
next prev parent reply other threads:[~2019-04-24 21:54 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-14 16:44 io_uring memory ordering (and broken queue-is-full checks) Stefan Bühler
2019-04-17 15:02 ` Jens Axboe
2019-04-19 9:29 ` Stefan Bühler
2019-04-19 9:57 ` [PATCH v1 1/3] io_uring: fix race condition reading SQ entries Stefan Bühler
2019-04-19 9:57 ` [PATCH v1 2/3] io_uring: fix race condition when sq threads goes sleeping Stefan Bühler
2019-04-19 9:57 ` [PATCH v1 3/3] io_uring: fix poll full SQ detection Stefan Bühler
2019-04-22 17:04 ` io_uring memory ordering (and broken queue-is-full checks) Jens Axboe
2019-04-24 21:54 ` [PATCH barrier cleanup v1 1/7] io_uring: fix notes on barriers Stefan Bühler
2019-04-24 21:54 ` [PATCH barrier cleanup v1 2/7] io_uring: remove unnecessary barrier before wq_has_sleeper Stefan Bühler
2019-04-24 21:54 ` Stefan Bühler [this message]
2019-04-24 21:54 ` [PATCH barrier cleanup v1 4/7] io_uring: remove unnecessary barrier after updating SQ head Stefan Bühler
2019-04-24 21:54 ` [PATCH barrier cleanup v1 5/7] io_uring: remove unnecessary barrier before reading SQ tail Stefan Bühler
2019-04-24 21:54 ` [PATCH barrier cleanup v1 6/7] io_uring: remove unnecessary barrier after incrementing dropped counter Stefan Bühler
2019-04-24 21:54 ` [PATCH barrier cleanup v1 7/7] io_uring: remove unnecessary barrier after unsetting IORING_SQ_NEED_WAKEUP Stefan Bühler
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=20190424215422.7404-3-source@stbuehler.de \
--to=source@stbuehler.de \
--cc=axboe@kernel.dk \
--cc=linux-block@vger.kernel.org \
--cc=linux-fsdevel@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).