[gstreamer-bugs] [Bug 545807] [baseaudiosink] audible crack when starting the pipeline

[GStreamer (bugzilla.gnome.org)]

If you have any questions why you received this email, please see the text at
the end of this email. Replies to this email are NOT read, please see the text
at the end of this email. You can add comments to this bug at:
  http://bugzilla.gnome.org/show_bug.cgi?id=545807

  GStreamer | gst-plugins-base | Ver: git




------- Comment #18 from Eero Nurkkala  2009-05-01 18:23 UTC -------
(In reply to comment #17)
> 
> A better solution would be to keep the hardware read pointer and our write
> pointer within a reasonable distance regardless of hw write() behaves and
> blocks. The sunaudio driver has something like that. The thing is that all
> alternatives to write() have historically been very flaky on many alsa
> implementations. What we really want to use is plain mmap (like pulseaudio) and
> fix all the alsa driver out there that would fail.
> 

(Just to make sure we're talking about the same thing; something like:
gst-launch filesrc ! mp3dec ! alsa or pulsesink)
Just a sidenote, this stuff also occur with pulsesink.

I'm not exatly sure what you're saying - but I interpret is as:
audioringbuffer_thread_func should not get too far (from the data provided by
mp3dec or wavdec etc)? That is something like the patch attached to this bug
does; but probably not in the most optimal way.

Like it's now, I see no single problem with alsasink or pulsesink, they're
fine. It's the very best that Gstreamer tries (and does so) to keep these
buffers all filled up.

So by fixing all alsa drivers, you mean the kernel drivers? Or gstreamer
alsa/pulseaudio drivers?


I've tried modifying the audioringbuffer_thread_func (thread) to be more
scheduler friendly - but with no success so far. Possibly that could be another
way around the problem?


-- 
See http://bugzilla.gnome.org/page.cgi?id=email.html for more info about why you received
this email, why you can't respond via email, how to stop receiving
emails (or reduce the number you receive), and how to contact someone
if you are having problems with the system.

You can add comments to this bug at http://bugzilla.gnome.org/show_bug.cgi?id=545807.