[4/6] io_uring: mark the need to lock/unlock the ring as unlikely

Commit Message

Jens Axboe Feb. 7, 2024, 5:17 p.m. UTC

Any of the fast paths will already have this locked, this helper only
exists to deal with io-wq invoking request issue where we do not have
the ctx->uring_lock held already. This means that any common or fast
path will already have this locked, mark it as such.

Signed-off-by: Jens Axboe <[e-mail]>
---
 io_uring/io_uring.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Patch

diff --git a/io_uring/io_uring.h b/io_uring/io_uring.h
index 2952551fe345..46795ee462df 100644
--- a/io_uring/io_uring.h
+++ b/io_uring/io_uring.h
@@ -208,7 +208,7 @@  static inline void io_ring_submit_unlock(struct io_ring_ctx *ctx,
 					 unsigned issue_flags)
 {
 	lockdep_assert_held(&ctx->uring_lock);
-	if (issue_flags & IO_URING_F_UNLOCKED)
+	if (unlikely(issue_flags & IO_URING_F_UNLOCKED))
 		mutex_unlock(&ctx->uring_lock);
 }
 
@@ -221,7 +221,7 @@  static inline void io_ring_submit_lock(struct io_ring_ctx *ctx,
 	 * The only exception is when we've detached the request and issue it
 	 * from an async worker thread, grab the lock for that case.
 	 */
-	if (issue_flags & IO_URING_F_UNLOCKED)
+	if (unlikely(issue_flags & IO_URING_F_UNLOCKED))
 		mutex_lock(&ctx->uring_lock);
 	lockdep_assert_held(&ctx->uring_lock);
 }