[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <76eef510a89966e04184893a3c53c6cc67b81579.1629655338.git.olivier@trillion01.com>
Date: Sun, 22 Aug 2021 17:05:59 -0400
From: Olivier Langlois <olivier@...llion01.com>
To: Jens Axboe <axboe@...nel.dk>,
Pavel Begunkov <asml.silence@...il.com>,
Oleg Nesterov <oleg@...hat.com>,
Steven Rostedt <rostedt@...dmis.org>,
Ingo Molnar <mingo@...hat.com>,
"Eric W. Biederman" <ebiederm@...ssion.com>,
io-uring@...r.kernel.org, linux-fsdevel@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: [PATCH 2/3] io_uring: Clear TIF_NOTIFY_SIGNAL when cancelling requests
It is a reasonable expectation from a function to leave the task struct
in a clean state. During io_uring_try_cancel_requests TIF_NOTIFY_SIGNAL
can be set. Make sure that it is cleared by replacing calls to
io_run_task_work with tracehook_notify_signal
Signed-off-by: Olivier Langlois <olivier@...llion01.com>
---
fs/io_uring.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/io_uring.c b/fs/io_uring.c
index a2e20a6fbfed..a9c83a5fc9f1 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -9032,7 +9032,7 @@ static void io_uring_try_cancel_requests(struct io_ring_ctx *ctx,
ret |= io_poll_remove_all(ctx, task, cancel_all);
ret |= io_kill_timeouts(ctx, task, cancel_all);
if (task)
- ret |= io_run_task_work();
+ ret |= tracehook_notify_signal();
if (!ret)
break;
cond_resched();
--
2.32.0
Powered by blists - more mailing lists