From: Pavel Begunkov <[email protected]>
To: Jens Axboe <[email protected]>, [email protected]
Subject: [PATCH 03/11] io_uring: optimise out task_work checks on enter
Date: Mon, 22 Mar 2021 01:58:26 +0000 [thread overview]
Message-ID: <ff3273b5111fdb10eea0e3d4f81f620fb58c5a5b.1616378197.git.asml.silence@gmail.com> (raw)
In-Reply-To: <[email protected]>
io_run_task_work() does some extra work for safety, that isn't actually
needed in many cases, particularly when it's known that current is not
exiting and in the TASK_RUNNING state, like in the beginning of a
syscall.
Signed-off-by: Pavel Begunkov <[email protected]>
---
fs/io_uring.c | 16 +++++++++++++---
1 file changed, 13 insertions(+), 3 deletions(-)
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 080fac4d4543..45b49273df8b 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -2227,6 +2227,16 @@ static inline unsigned int io_put_rw_kbuf(struct io_kiocb *req)
return io_put_kbuf(req, kbuf);
}
+/* only safe when TASK_RUNNING and it's not PF_EXITING */
+static inline bool __io_run_task_work(void)
+{
+ if (current->task_works) {
+ task_work_run();
+ return true;
+ }
+ return false;
+}
+
static inline bool io_run_task_work(void)
{
/*
@@ -6729,7 +6739,7 @@ static int io_sq_thread(void *data)
}
if (sqt_spin || !time_after(jiffies, timeout)) {
- io_run_task_work();
+ __io_run_task_work();
cond_resched();
if (sqt_spin)
timeout = jiffies + sqd->sq_thread_idle;
@@ -6870,7 +6880,7 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events,
io_cqring_overflow_flush(ctx, false, NULL, NULL);
if (io_cqring_events(ctx) >= min_events)
return 0;
- if (!io_run_task_work())
+ if (!__io_run_task_work())
break;
} while (1);
@@ -9117,7 +9127,7 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit,
struct fd f;
long ret;
- io_run_task_work();
+ __io_run_task_work();
if (unlikely(flags & ~(IORING_ENTER_GETEVENTS | IORING_ENTER_SQ_WAKEUP |
IORING_ENTER_SQ_WAIT | IORING_ENTER_EXT_ARG)))
--
2.24.0
next prev parent reply other threads:[~2021-03-22 2:03 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-22 1:58 [PATCH 5.13 00/11] yet another series of random 5.13 patches Pavel Begunkov
2021-03-22 1:58 ` [PATCH 01/11] io_uring: don't clear REQ_F_LINK_TIMEOUT Pavel Begunkov
2021-03-22 1:58 ` [PATCH 02/11] io_uring: don't do extra EXITING cancellations Pavel Begunkov
2021-03-22 1:58 ` Pavel Begunkov [this message]
2021-03-22 13:45 ` [PATCH 03/11] io_uring: optimise out task_work checks on enter Jens Axboe
2021-03-22 14:53 ` Pavel Begunkov
2021-03-22 14:58 ` Jens Axboe
2021-03-22 1:58 ` [PATCH 04/11] io_uring: remove tctx->sqpoll Pavel Begunkov
2021-03-22 1:58 ` [PATCH 05/11] io-wq: refactor *_get_acct() Pavel Begunkov
2021-03-22 1:58 ` [PATCH 06/11] io_uring: don't init req->work fully in advance Pavel Begunkov
2021-03-22 1:58 ` [PATCH 07/11] io_uring: kill unused REQ_F_NO_FILE_TABLE Pavel Begunkov
2021-03-22 1:58 ` [PATCH 08/11] io_uring: optimise kiocb_end_write for !ISREG Pavel Begunkov
2021-03-22 1:58 ` [PATCH 09/11] io_uring: don't alter iopoll reissue fail ret code Pavel Begunkov
2021-03-22 1:58 ` [PATCH 10/11] io_uring: hide iter revert in resubmit_prep Pavel Begunkov
2021-03-22 1:58 ` [PATCH 11/11] io_uring: optimise rw complete error handling Pavel Begunkov
2021-03-22 2:00 ` [PATCH 5.13 00/11] yet another series of random 5.13 patches Pavel Begunkov
2021-03-22 13:49 ` 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=ff3273b5111fdb10eea0e3d4f81f620fb58c5a5b.1616378197.git.asml.silence@gmail.com \
[email protected] \
[email protected] \
[email protected] \
/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