[PATCH v3 RESEND] io_uring: do the sqpoll napi busy poll outside the submission block

[PATCH v3 RESEND] io_uring: do the sqpoll napi busy poll outside the submission block

[Olivier Langlois]

there are many small reasons justifying this change.

1. busy poll must be performed even on rings that have no iopoll and no
   new sqe. It is quite possible that a ring configured for inbound
   traffic with multishot be several hours without receiving new request
   submissions
2. NAPI busy poll does not perform any credential validation
3. If the thread is awaken by task work, processing the task work is
   prioritary over NAPI busy loop. This is why a second loop has been
   created after the io_sq_tw() call instead of doing the busy loop in
   __io_sq_thread() outside its credential acquisition block.

Signed-off-by: Olivier Langlois <olivier@trillion01.com>
---
 io_uring/sqpoll.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/io_uring/sqpoll.c b/io_uring/sqpoll.c
index 3b50dc9586d1..f9964eb03803 100644
--- a/io_uring/sqpoll.c
+++ b/io_uring/sqpoll.c
@@ -195,9 +195,6 @@ static int __io_sq_thread(struct io_ring_ctx *ctx, bool cap_entries)
 			ret = io_submit_sqes(ctx, to_submit);
 		mutex_unlock(&ctx->uring_lock);
 
-		if (io_napi(ctx))
-			ret += io_napi_sqpoll_busy_poll(ctx);
-
 		if (to_submit && wq_has_sleeper(&ctx->sqo_sq_wait))
 			wake_up(&ctx->sqo_sq_wait);
 		if (creds)
@@ -322,6 +319,10 @@ static int io_sq_thread(void *data)
 		if (io_sq_tw(&retry_list, IORING_TW_CAP_ENTRIES_VALUE))
 			sqt_spin = true;
 
+		list_for_each_entry(ctx, &sqd->ctx_list, sqd_list)
+			if (io_napi(ctx))
+				io_napi_sqpoll_busy_poll(ctx);
+
 		if (sqt_spin || !time_after(jiffies, timeout)) {
 			if (sqt_spin) {
 				io_sq_update_worktime(sqd, &start);
-- 
2.46.0

Re: [PATCH v3 RESEND] io_uring: do the sqpoll napi busy poll outside the submission block

[Jens Axboe]

On Mon, 16 Sep 2024 15:17:56 -0400, Olivier Langlois wrote:
> there are many small reasons justifying this change.
> 
> 1. busy poll must be performed even on rings that have no iopoll and no
>    new sqe. It is quite possible that a ring configured for inbound
>    traffic with multishot be several hours without receiving new request
>    submissions
> 2. NAPI busy poll does not perform any credential validation
> 3. If the thread is awaken by task work, processing the task work is
>    prioritary over NAPI busy loop. This is why a second loop has been
>    created after the io_sq_tw() call instead of doing the busy loop in
>    __io_sq_thread() outside its credential acquisition block.
> 
> [...]

Applied, thanks!

[1/1] io_uring: do the sqpoll napi busy poll outside the submission block
      commit: 53d69bdd5b19bb17602cb224e01aeed730ff3289

Best regards,
-- 
Jens Axboe