diff options
| author | Pavel Begunkov <asml.silence@gmail.com> | 2023-01-09 14:46:09 +0000 |
|---|---|---|
| committer | Jens Axboe <axboe@kernel.dk> | 2023-01-29 15:17:40 -0700 |
| commit | bca39f39058567643487cd654970717705784ba3 (patch) | |
| tree | 1f0282a4e401c5c06e453d3c0605588e89045f63 /io_uring/io_uring.h | |
| parent | 7b235dd82ad32c1626e51303d94ec5ef4d7bc994 (diff) | |
| download | lwn-bca39f39058567643487cd654970717705784ba3.tar.gz lwn-bca39f39058567643487cd654970717705784ba3.zip | |
io_uring: add lazy poll_wq activation
Even though io_poll_wq_wake()'s waitqueue_active reuses a barrier we do
for another waitqueue, it's not going to be the case in the future and
so we want to have a fast path for it when the ring has never been
polled.
Move poll_wq wake ups into __io_commit_cqring_flush() using a new flag
called ->poll_activated. The idea behind the flag is to set it when the
ring was polled for the first time. This requires additional sync to not
miss events, which is done here by using task_work for ->task_complete
rings, and by default enabling the flag for all other types of rings.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/060785e8e9137a920b232c0c7f575b131af19cac.1673274244.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'io_uring/io_uring.h')
| -rw-r--r-- | io_uring/io_uring.h | 7 |
1 files changed, 3 insertions, 4 deletions
diff --git a/io_uring/io_uring.h b/io_uring/io_uring.h index c75bbb94703c..5113e0ddb01d 100644 --- a/io_uring/io_uring.h +++ b/io_uring/io_uring.h @@ -222,7 +222,7 @@ static inline void io_commit_cqring(struct io_ring_ctx *ctx) static inline void io_poll_wq_wake(struct io_ring_ctx *ctx) { - if (waitqueue_active(&ctx->poll_wq)) + if (wq_has_sleeper(&ctx->poll_wq)) __wake_up(&ctx->poll_wq, TASK_NORMAL, 0, poll_to_key(EPOLL_URING_WAKE | EPOLLIN)); } @@ -230,8 +230,6 @@ static inline void io_poll_wq_wake(struct io_ring_ctx *ctx) /* requires smb_mb() prior, see wq_has_sleeper() */ static inline void __io_cqring_wake(struct io_ring_ctx *ctx) { - io_poll_wq_wake(ctx); - /* * Trigger waitqueue handler on all waiters on our waitqueue. This * won't necessarily wake up all the tasks, io_should_wake() will make @@ -316,7 +314,8 @@ static inline void io_req_complete_defer(struct io_kiocb *req) static inline void io_commit_cqring_flush(struct io_ring_ctx *ctx) { - if (unlikely(ctx->off_timeout_used || ctx->drain_active || ctx->has_evfd)) + if (unlikely(ctx->off_timeout_used || ctx->drain_active || + ctx->has_evfd || ctx->poll_activated)) __io_commit_cqring_flush(ctx); } |