[prev in list] [next in list] [prev in thread] [next in thread] 

List:       mms
Subject:    Re: Alsaplayer bug resolved (was Re: mmsv2 1.0.4 released)
From:       Anders Rune Jensen <anders () gnulinux ! dk>
Date:       2005-09-25 17:56:45
Message-ID: 1127671005.5512.13.camel () localhost
[Download RAW message or body]


On Sun, 2005-09-25 at 14:50 +0200, Esprit wrote:
> Esprit a écrit :
> 
> > Anders Rune Jensen a écrit :
> >
> >> On Sat, 2005-09-24 at 14:39 +0200, Esprit wrote:
> >>  
> >>
> >>> Anders Rune Jensen a écrit :
> >>>
> >>>   
> >>>
> >>>> Hi all
> >>>>
> >>>> [...]
> >>>>
> >>>>     
> >>>
> >>> Great Work !
> >>>
> >>> But I have a strange feature now... :-D
> >>> When using alsaplayer as audio backend, if I run mms,
> >>> and want to play a video file, mms will crash after having
> >>> launch cxfe, which crashed because the audio device is
> >>> used by alsaplayer (if I 'ps' after mms crash, I can see the
> >>> alsaplayer process still alive).
> >>>   
> >>
> >>
> >> Hmm doesn't sound that good. Will try and see if I can reproduce.
> >>
> >>  
> >>
> > I will try to have a look at this this afternoon (it's raining cats 
> > and dogs,
> > so I have time to spend tracking bugs ;-)  ).
> >
> Found it !
> 
> In common.cpp, line 378, adding :
> ----------------
>         // kill the player since it's holding the audio device hostage
>         audio_state->p->stop_player();
> -----------------
> seems OK, the player is stopped before playing a movie, and we can
> still access the audio player after the video is finished (I first added
> "delete audio_state;", like in the 1.0.3 version, but after watching a
> video, if I was going back to audio, mms was crashing...).
> 
> Hope this is the right solution...

Your solution is almost correct. audio_state->p->stop() stops the player
but that was only done if something was playing. It clearly also needs
to be done when something wasn't because the audio player can still hog
the drive. So moving the audio_state->p->stop() out of the if clause
will cause it to work in both situations (playing and not-playing).
Thanks again for the report and the suggested fix.

I have commited the fix to mmsv2--mainline--1.0.5--patch-1.

> Yann

-- 
Anders Rune Jensen
http://www.cs.auc.dk/~arj/

PGP/GnuPG key: 1024D/62C2D7F0 @ pgp.mit.edu
Fingerprint: 6A03 907E 92E1 47EB 4EAB  76B6 068A ACD1 62C2 D7F0


["signature.asc" (application/pgp-signature)]

[prev in list] [next in list] [prev in thread] [next in thread]