io_uring: clean up local tw add-wait sync

Kill a smp_mb__after_atomic() right before wake_up, it's useless, and
add a comment explaining implicit barriers from cmpxchg and
synchronsation around ->cq_wait_nr with the waiter.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/3007f3c2d53c72b61de56919ef56b53158b8276f.1705438669.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index d40c767a62163d3e71489350dc370f1ad1309efb..3ab7e6a46149b140faf68c4057ec1282cad1b16f 100644 (file)
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -1332,6 +1332,14 @@ static inline void io_req_local_work_add(struct io_kiocb *req, unsigned flags)
        } while (!try_cmpxchg(&ctx->work_llist.first, &first,
                              &req->io_task_work.node));
 
+       /*
+        * cmpxchg implies a full barrier, which pairs with the barrier
+        * in set_current_state() on the io_cqring_wait() side. It's used
+        * to ensure that either we see updated ->cq_wait_nr, or waiters
+        * going to sleep will observe the work added to the list, which
+        * is similar to the wait/wawke task state sync.
+        */
+
        if (!first) {
                if (ctx->flags & IORING_SETUP_TASKRUN_FLAG)
                        atomic_or(IORING_SQ_TASKRUN, &ctx->rings->sq_flags);
@@ -1346,8 +1354,6 @@ static inline void io_req_local_work_add(struct io_kiocb *req, unsigned flags)
        /* either not enough or the previous add has already woken it up */
        if (nr_wait > nr_tw || nr_tw_prev >= nr_wait)
                return;
-       /* pairs with set_current_state() in io_cqring_wait() */
-       smp_mb__after_atomic();
        wake_up_state(ctx->submitter_task, TASK_INTERRUPTIBLE);
 }