From f46b9cdb22f7a167c36b6bcddaef7e8aee2598fa Mon Sep 17 00:00:00 2001 From: David Wei Date: Wed, 20 Nov 2024 14:14:52 -0800 Subject: io_uring: limit local tw done Instead of eagerly running all available local tw, limit the amount of local tw done to the max of IO_LOCAL_TW_DEFAULT_MAX (20) or wait_nr. The value of 20 is chosen as a reasonable heuristic to allow enough work batching but also keep latency down. Add a retry_llist that maintains a list of local tw that couldn't be done in time. No synchronisation is needed since it is only modified within the task context. Signed-off-by: David Wei Link: https://lore.kernel.org/r/20241120221452.3762588-3-dw@davidwei.uk Signed-off-by: Jens Axboe --- include/linux/io_uring_types.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h index aa5f5ea98076..3e934feb3187 100644 --- a/include/linux/io_uring_types.h +++ b/include/linux/io_uring_types.h @@ -335,6 +335,7 @@ struct io_ring_ctx { */ struct { struct llist_head work_llist; + struct llist_head retry_llist; unsigned long check_cq; atomic_t cq_wait_nr; atomic_t cq_timeouts; -- cgit v1.2.3