[Mplayer-dev-eng] release 0.18...

[Arpi]

Hi,

> >> >> (if it's truth that mplayer runs several processes on separate cpus)
> >> >not true. it runs in only one process/thread now.
> >> >
> >> Then I dont know! If mplayer consist from one process without threads which can be run
> >> on other cpus - there must be no difference between uni- and multi- processor systems.
> >> In this case problem is not in mplayer but outside.
> >> But I think you are wrong.
> >> Please look at line # 61 of codecctrl.c:
> >>       if((child_pid=fork())==0){
> >It's the old code. Not used in the current version.
> >
> Then may be better to move out it to SHIT subdir?
It can be removed, but not yet.

> >> IMHO fork creates process which can be run on any cpu by kernel opinion.
> >> and it placed in mpeg_codec_controller :(
> >> But seems that this function never called. Sorry if I'm wrong.
> >Somewhere in mplayer.c there is some #ifdef-s to enable old fork() code,
> >but as latest changes wasn't applied to that code, it may be broken.
> >
> I've found only one file with one fork...
in codecctrl.c it is the old code, and it's included only if it's
enabled.

but i think it won't compile, because wasn't updated after sh_* changes.


A'rpi / Astral & ESP-team

--
mailto:arpi at thot.banki.hu
http://esp-team.scene.hu

_______________________________________________
Mplayer-dev-eng mailing list
Mplayer-dev-eng at lists.sourceforge.net
http://lists.sourceforge.net/lists/listinfo/mplayer-dev-eng