[PATCH] io_uring/napi: cap busy_poll_to 10 msec

[PATCH] io_uring/napi: cap busy_poll_to 10 msec

[Jens Axboe]

Currently there's no cap on the maximum amount of time that napi is
allowed to poll if no events are found, which can lead to kernel
complaints on a task being stuck as there's no conditional rescheduling
done within that loop.

Just cap it to 10 msec in total, that's already way above any kind of
sane value that will reap any benefits, yet low enough that it's
nowhere near being able to trigger preemption complaints.

Fixes: 8d0c12a80cde ("io-uring: add napi busy poll support")
Signed-off-by: Jens Axboe <axboe@kernel.dk>

---

diff --git a/io_uring/napi.c b/io_uring/napi.c
index 4a10de03e426..8d68366a4b90 100644
--- a/io_uring/napi.c
+++ b/io_uring/napi.c
@@ -276,6 +276,8 @@ static int io_napi_register_napi(struct io_ring_ctx *ctx,
 	/* clean the napi list for new settings */
 	io_napi_free(ctx);
 	WRITE_ONCE(ctx->napi_track_mode, napi->op_param);
+	/* cap NAPI at 10 msec of spin time */
+	napi->busy_poll_to = min(10000, napi->busy_poll_to);
 	WRITE_ONCE(ctx->napi_busy_poll_dt, napi->busy_poll_to * NSEC_PER_USEC);
 	WRITE_ONCE(ctx->napi_prefer_busy_poll, !!napi->prefer_busy_poll);
 	return 0;

-- 
Jens Axboe