All of lore.kernel.org
 help / color / mirror / Atom feed
From: Pavel Begunkov <asml.silence@gmail.com>
To: Jens Axboe <axboe@kernel.dk>, io-uring@vger.kernel.org
Subject: [PATCH 1/6] io_uring: don't change sqpoll creds if not needed
Date: Thu, 24 Jun 2021 15:09:55 +0100	[thread overview]
Message-ID: <c54368da2357ac539e0a333f7cfff70d5fb045b2.1624543113.git.asml.silence@gmail.com> (raw)
In-Reply-To: <cover.1624543113.git.asml.silence@gmail.com>

SQPOLL doesn't need to change creds if it's not submitting requests.
Move creds overriding into __io_sq_thread() after checking if there are
SQEs pending.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 fs/io_uring.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 14a90e4e4149..cf72cc3fd8f4 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -6999,6 +6999,10 @@ static int __io_sq_thread(struct io_ring_ctx *ctx, bool cap_entries)
 
 	if (!list_empty(&ctx->iopoll_list) || to_submit) {
 		unsigned nr_events = 0;
+		const struct cred *creds = NULL;
+
+		if (ctx->sq_creds != current_cred())
+			creds = override_creds(ctx->sq_creds);
 
 		mutex_lock(&ctx->uring_lock);
 		if (!list_empty(&ctx->iopoll_list))
@@ -7015,6 +7019,8 @@ static int __io_sq_thread(struct io_ring_ctx *ctx, bool cap_entries)
 
 		if (to_submit && wq_has_sleeper(&ctx->sqo_sq_wait))
 			wake_up(&ctx->sqo_sq_wait);
+		if (creds)
+			revert_creds(creds);
 	}
 
 	return ret;
@@ -7066,7 +7072,6 @@ static int io_sq_thread(void *data)
 
 	mutex_lock(&sqd->lock);
 	while (1) {
-		int ret;
 		bool cap_entries, sqt_spin, needs_sched;
 
 		if (io_sqd_events_pending(sqd) || signal_pending(current)) {
@@ -7079,13 +7084,8 @@ static int io_sq_thread(void *data)
 		sqt_spin = false;
 		cap_entries = !list_is_singular(&sqd->ctx_list);
 		list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) {
-			const struct cred *creds = NULL;
+			int ret = __io_sq_thread(ctx, cap_entries);
 
-			if (ctx->sq_creds != current_cred())
-				creds = override_creds(ctx->sq_creds);
-			ret = __io_sq_thread(ctx, cap_entries);
-			if (creds)
-				revert_creds(creds);
 			if (!sqt_spin && (ret > 0 || !list_empty(&ctx->iopoll_list)))
 				sqt_spin = true;
 		}
-- 
2.32.0


  reply	other threads:[~2021-06-24 14:10 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-24 14:09 [PATCH for-next 0/6] straightforward for-next cleanups Pavel Begunkov
2021-06-24 14:09 ` Pavel Begunkov [this message]
2021-06-24 14:09 ` [PATCH 2/6] io_uring: refactor io_sq_thread() Pavel Begunkov
2021-06-24 14:09 ` [PATCH 3/6] io_uring: fix code style problems Pavel Begunkov
2021-06-24 14:09 ` [PATCH 4/6] io_uring: update sqe layout build checks Pavel Begunkov
2021-06-24 14:09 ` [PATCH 5/6] io_uring: simplify struct io_uring_sqe layout Pavel Begunkov
2021-06-24 14:10 ` [PATCH 6/6] io_uring: refactor io_openat2() Pavel Begunkov
2021-06-24 15:35 ` [PATCH for-next 0/6] straightforward for-next cleanups 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=c54368da2357ac539e0a333f7cfff70d5fb045b2.1624543113.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.