https://bugzilla.novell.com/show_bug.cgi?id=405354 User wolfgang@rosenauer.org added comment https://bugzilla.novell.com/show_bug.cgi?id=405354#c2 Wolfgang Rosenauer changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|wolfgang@rosenauer.org | --- Comment #2 from Wolfgang Rosenauer 2008-07-01 07:59:40 MDT --- (In reply to comment #0 from Wolfgang Rosenauer)

int rc = snd_pcm_open(&mHandle, settings.value("audio/device", "default").toByteArray(), SND_PCM_STREAM_PLAYBACK, SND_PCM_ASYNC);

(In reply to comment #1 from Takashi Iwai)

The async handler doesn't work with pulse plugin. If your code uses it, no chance that it'll work with pulse or any other external plugins. (That's why flash also doesn't work with alsa-pulse plugin.)

So I guess the above flag defines to use the async handler? I see no other definitions. Output is done with snd_pcm_writei(). Thanks for the hint. (BTW: why does Flash work then at all? Is there some workaround?) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=405354 Takashi Iwai changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |wolfgang@rosenauer.org -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=405354 User wolfgang@rosenauer.org added comment https://bugzilla.novell.com/show_bug.cgi?id=405354#c5 Wolfgang Rosenauer changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|wolfgang@rosenauer.org | --- Comment #5 from Wolfgang Rosenauer 2008-07-01 08:34:23 MDT --- clear NEEDINFO flag -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=405354 User wolfgang@rosenauer.org added comment https://bugzilla.novell.com/show_bug.cgi?id=405354#c7 Wolfgang Rosenauer changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|wolfgang@rosenauer.org | --- Comment #7 from Wolfgang Rosenauer 2008-07-02 01:31:46 MDT --- (In reply to comment #6 from Takashi Iwai)

Hm, this sounds like rather a problem in the app code. Is the start_threshold set properly?

That could really be an app issue. It's just that it always worked before pulseaudio was introduced. We used snd_pcm_start() to activate the stream but I've changed that to use start_threshold=64 (which is the number of frames per period AFAICS) with no change in behaviour.

snd_pcm_writei() doesn't care whether the samples are played or not. It cares only where the samples are put on the buffer or not.

So what could hold alsa from actually playing its buffer? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=405354 User wolfgang@rosenauer.org added comment https://bugzilla.novell.com/show_bug.cgi?id=405354#c9 --- Comment #9 from Wolfgang Rosenauer 2008-07-04 12:30:51 MDT --- I fixed some other issues in our code meanwhile but pulseaudio still blocks the sound output. Is there any debugging I can do to actually see what fails? I'm checking all the return values of the ALSA calls and as soon as PA is not running everything works as it should. 'pasuspender' doesn't make it work though. All I see is that Alsa's buffer is filled and the device/stream is in state 'PREPARED' before it is written to it's buffer. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.