Re: Race in vt_event_wait() during suspend/resume

[Alan Cox <alan@lxorguk.ukuu.org.uk>]

On Fri, 18 May 2012 16:34:53 +0530
Rabin Vincent <rabin@rab.in> wrote:

> pm_restore_console() is called from the suspend/resume path,
> this calls vt_move_to_console(), which calls vt_event_wait().
> 
> There's a race in this path which causes the process
> which requests the suspend to sleep indefinitely waiting
> for an event which already happened:
> 
> P1					P2
>  vt_move_to_console()
>    set_console()
>      schedule_console_callback()
>    vt_waitactive()
>      check n == fg_console +1
> 					console_callback()
> 					  switch_screen()
> 					  vt_event_post() // no waiters
> 
>      vt_event_wait() // forever
> 
> I think the following should fix it.  Comments?
> 

Looks right to me.

Alan