[vlc-devel] vlc: svn commit r23063 (funman)

Remi Denis-Courmont rdenis at simphalempin.com
Wed Nov 14 09:53:40 CET 2007


Written properly there is no need to check BEFORE. if you do, it implies you are leaking conditions signal (being signaled while no thread is waiting on)

Remi Denis

-- message original --
Sujet:	Re: [vlc-devel] vlc: svn commit r23063 (funman)
De:	Rafaël Carré <funman at videolan.org>
Date:		14.11.2007 07:37

Le Wed, 14 Nov 2007 08:09:29 +0200,
"Remi Denis-Courmont" <rdenis at simphalempin.com> a écrit :

> I disagree. There may be cases where you need to wait on yourself
> while dying. This breaks it. Wrong.

What are those cases ?

The other solution is to add that check before each call
vlc_object_timedwait() that needs it.

>  Remi Denis
> 
> -- message original --
> Sujet:	[vlc-devel] vlc: svn commit r23063 (funman)
> De:	Subversion daemon <svn at videolan.org>
> Date:		14.11.2007 00:31
> 
> r23063 | funman | 2007-11-14 01:30:45 +0100 (Wed, 14 Nov 2007) | 2
> lines Changed paths:
>    M /trunk/src/misc/objects.c
> 
> make vlc_object_timedwait() return immediately if the object or
> libvlc is dying
> 
>  > http://trac.videolan.org/vlc/changeset/23063
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel
> 
> _______________________________________________
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
> http://mailman.videolan.org/listinfo/vlc-devel


-- 
Rafaël Carré

_______________________________________________
vlc-devel mailing list
To unsubscribe or modify your subscription options:
http://mailman.videolan.org/listinfo/vlc-devel

<<signature.asc>>




More information about the vlc-devel mailing list