linux-kernel - [PATCH] io_uring: Fix WARNING in io_cqring_event

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20240702-bug9-v1-1-475cb52d3ee6@gmail.com>
Date: Tue, 02 Jul 2024 17:48:27 -0700
From: Pei Li <peili.dev@...il.com>
To: Jens Axboe <axboe@...nel.dk>, Pavel Begunkov <asml.silence@...il.com>
Cc: io-uring@...r.kernel.org, linux-kernel@...r.kernel.org, 
 skhan@...uxfoundation.org, syzkaller-bugs@...glegroups.com, 
 linux-kernel-mentees@...ts.linuxfoundation.org, 
 syzbot+f7f9c893345c5c615d34@...kaller.appspotmail.com, 
 Pei Li <peili.dev@...il.com>
Subject: [PATCH] io_uring: Fix WARNING in io_cqring_event_overflow

Acquire ctx->completion_lock in io_add_aux_cqe().

syzbot reports a warning message in io_cqring_event_overflow(). We were
supposed to hold ctx->completion_lock before entering this function, but
instead we did not.

This patch acquires and releases ctx->completion_lock when entering and
exiting io_add_aux_cqe().

Fixes: f33096a3c99c ("io_uring: add io_add_aux_cqe() helper")
Reported-by: syzbot+f7f9c893345c5c615d34@...kaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=f7f9c893345c5c615d34
Signed-off-by: Pei Li <peili.dev@...il.com>
---
syzbot reports a warning message in io_cqring_event_overflow(). We were
supposed to hold ctx->completion_lock before entering this function, but
instead we did not.

The call stack is as follows:

Call Trace:
 <TASK>
 __io_post_aux_cqe io_uring/io_uring.c:816 [inline]
 io_add_aux_cqe+0x27c/0x320 io_uring/io_uring.c:837
 io_msg_tw_complete+0x9d/0x4d0 io_uring/msg_ring.c:78
 io_fallback_req_func+0xce/0x1c0 io_uring/io_uring.c:256
 process_one_work kernel/workqueue.c:3224 [inline]
 process_scheduled_works+0xa2c/0x1830 kernel/workqueue.c:3305
 worker_thread+0x86d/0xd40 kernel/workqueue.c:3383
 kthread+0x2f0/0x390 kernel/kthread.c:389
 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:144
 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244
 </TASK>

In io_add_aux_cqe(), we should acquire this lock beforehead.

This patch acquires and releases ctx->completion_lock when entering and
exiting io_add_aux_cqe().
---
 io_uring/io_uring.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 4e2836c9b7bf..0f62332e95ff 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -834,8 +834,10 @@ bool io_post_aux_cqe(struct io_ring_ctx *ctx, u64 user_data, s32 res, u32 cflags
  */
 void io_add_aux_cqe(struct io_ring_ctx *ctx, u64 user_data, s32 res, u32 cflags)
 {
+	io_cq_lock(ctx);
 	__io_post_aux_cqe(ctx, user_data, res, cflags);
 	ctx->submit_state.cq_flush = true;
+	io_cq_unlock_post(ctx);
 }
 
 /*

---
base-commit: 74564adfd3521d9e322cfc345fdc132df80f3c79
change-id: 20240702-bug9-ad8a36681e2d

Best regards,
-- 
Pei Li <peili.dev@...il.com>