[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20181022170244.GB27929@dtor-ws>
Date: Mon, 22 Oct 2018 10:02:44 -0700
From: Dmitry Torokhov <dmitry.torokhov@...il.com>
To: Sasha Levin <sashal@...nel.org>
Cc: stable@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH AUTOSEL 4.18 08/27] Input: evdev - add a schedule point
in evdev_write()
Hi Sasha,
On Mon, Oct 22, 2018 at 06:19:05AM -0400, Sasha Levin wrote:
> From: Dmitry Torokhov <dmitry.torokhov@...il.com>
>
> [ Upstream commit 36d2582ff235b4e01ad64a734c877a52dc762d9c ]
>
> Large writes to evdev interface may cause rcu stalls. Let's add
> cond_resched() to the loop to avoid this.
This is not needed in stable:
- the interfaces are not accessible for unprivileged users so there is
no DOS concerns
- the issue can only happen on non-preemptible kernels
- the issue only showed up with synthetic test and not real life use
cases
- this is not a regression
The same goes to similar patches to mousedev, joydev and uinput that you
autoselected.
Thanks.
>
> Reviewed-by: Paul E. McKenney <paulmck@...ux.ibm.com>
> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@...il.com>
> Signed-off-by: Sasha Levin <sashal@...nel.org>
> ---
> drivers/input/evdev.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c
> index c81c79d01d93..718d5622267d 100644
> --- a/drivers/input/evdev.c
> +++ b/drivers/input/evdev.c
> @@ -564,6 +564,7 @@ static ssize_t evdev_write(struct file *file, const char __user *buffer,
>
> input_inject_event(&evdev->handle,
> event.type, event.code, event.value);
> + cond_resched();
> }
>
> out:
> --
> 2.17.1
>
--
Dmitry
Powered by blists - more mailing lists