[gst-devel] Patches for 0.6

[Benjamin Otte]

Hi 0.6 team,

Here's my list of patches that have gone into 0.7 since Feb. 1 and my
thoughts if they should be applied to 0.6.

1) Simple fix (misnamer) in the docs
File:	gstreamer/docs/gst/tmpl/gstcaps.sgml
Patch:
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gstreamer/docs/gst/tmpl/gstcaps.sgml.diff?r1=1.26&r2=1.27&diff_format=u

2) Fix for xvideosink (see bug 105256)
File:	gst-plugins/sys/xvideo/xvideosink.c
Patch:	http://bugzilla.gnome.org/showattachment.cgi?attach_id=14560
Comment:
I believe it's sane to apply all changes to the videosink since we branched.
(There's two other patches in there.) So you would just replace xvideosink.c
with HEAD xvideosink.c

3) Fix for mu-law / a-law
File:	Makefile.am
	alaw.c
	alaw-decode.c
	alaw-encode.c
        mulaw.c
	mulaw-decode.c
	mulaw-encode.c
	alaw-conversion.c (removed)
	alaw-conversion.h (removed)
Patch:
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gst-plugins/gst/law/Makefile.am.diff?r1=1.7&r2=1.8&diff_format=u
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gst-plugins/gst/law/alaw.c.diff?r1=1.6&r2=1.7&diff_format=u
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gst-plugins/gst/law/alaw-decode.c.diff?r1=1.4&r2=1.5&diff_format=u
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gst-plugins/gst/law/alaw-encode.c.diff?r1=1.4&r2=1.5&diff_format=u
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gst-plugins/gst/law/mulaw.c.diff?r1=1.6&r2=1.7&diff_format=u
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gst-plugins/gst/law/mulaw-decode.c.diff?r1=1.7&r2=1.8&diff_format=u
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gst-plugins/gst/law/mulaw-encode.c.diff?r1=1.4&r2=1.6&diff_format=u
Comment:
I believe mulaw and alaw converters are broken in 0.6. ("gst-launch filesrc
location=some.mp3 ! mad ! alawenc ! alawdec ! osssink") The above patches
fix this.
So IMO they should go in.

4) ALSA
Comment:
Since Alsa is disabled by default anyway, we should IMO not mess with it and
keep it that way. The current stuff is more stable, but it's still not
finished. So let's keep it.

5) Small fix for opt scheduler
File:	gstreamer/gst/schedulers/gstoptimalscheduler.c
Patch:
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gstreamer/gst/schedulers/gstoptimalscheduler.c.diff?r1=1.40&r2=1.41&diff_format=u
Comment:
Bugfix - initialize the thread system when using an optimal scheduler with
threads.

6) GThread Scheduler
File:	gstreamer/gst/schedulers/Makefile.am
	gstreamer/gst/schedulers/cothreads_compat.h
	gstreamer/gst/schedulers/gthread-cothreads.h (added)
Patch:
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gstreamer/gst/schedulers/Makefile.am.diff?r1=1.25&r2=1.26&diff_format=u
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gstreamer/gst/schedulers/cothreads_compat.h.diff?r1=1.10&r2=1.11&diff_format=u
http://cvs.sf.net/cgi-bin/viewcvs.cgi/gstreamer/gstreamer/gst/schedulers/gthread-cothreads.h?rev=1.4
Comment:
This scheduler is supposed to be portable. If there are still problems on
some archs that don't work because of buggy cothreads and we do not want to
use opt, we should go for this. Since I'm not hearing about non-working
schedulers anymore and we probably don't want to change our cothreads
implementation anyway, this should stay out.
But I don't know that much about GStreamer on Solaris/FreeBSD/whatever, so
it's up to people running those systems to test that.


Cheers,
Benjamin