From 52dd86406dfa322c8d42b3a4328858abdc6f1d85 Mon Sep 17 00:00:00 2001
From: Dylan Yudaken <dylany@fb.com>
Date: Fri, 25 Mar 2022 02:37:55 -0700
Subject: [PATCH] io_uring: enable EPOLLEXCLUSIVE for accept poll
Git-commit: 52dd86406dfa322c8d42b3a4328858abdc6f1d85
Patch-mainline: v5.18-rc1
References: bsc#1205205
When polling sockets for accept, use EPOLLEXCLUSIVE. This is helpful
when multiple accept SQEs are submitted.
For O_NONBLOCK sockets multiple queued SQEs would previously have all
completed at once, but most with -EAGAIN as the result. Now only one
wakes up and completes.
For sockets without O_NONBLOCK there is no user facing change, but
internally the extra requests would previously be queued onto a worker
thread as they would wake up with no connection waiting, and be
punted. Now they do not wake up unnecessarily.
Co-developed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Dylan Yudaken <dylany@fb.com>
Link: https://lore.kernel.org/r/20220325093755.4123343-1-dylany@fb.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Gabriel Krisman Bertazi <krisman@suse.de>
---
fs/io_uring.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -959,6 +959,7 @@ struct io_op_def {
/* set if opcode supports polled "wait" */
unsigned pollin : 1;
unsigned pollout : 1;
+ unsigned poll_exclusive : 1;
/* op supports buffer selection */
unsigned buffer_select : 1;
/* do prep async if is going to be punted */
@@ -1040,6 +1041,7 @@ static const struct io_op_def io_op_defs
.needs_file = 1,
.unbound_nonreg_file = 1,
.pollin = 1,
+ .poll_exclusive = 1,
},
[IORING_OP_ASYNC_CANCEL] = {},
[IORING_OP_LINK_TIMEOUT] = {
@@ -6132,7 +6134,8 @@ static int io_arm_poll_handler(struct io
} else {
mask |= POLLOUT | POLLWRNORM;
}
-
+ if (def->poll_exclusive)
+ mask |= EPOLLEXCLUSIVE;
if (!(issue_flags & IO_URING_F_UNLOCKED) &&
!list_empty(&ctx->apoll_cache)) {
apoll = list_first_entry(&ctx->apoll_cache, struct async_poll,