Re: OSS emulation delays (?)

[Takashi Iwai <tiwai@suse.de>]

[-- Attachment #1: Type: text/plain, Size: 1077 bytes --]

At Fri, 24 Oct 2003 14:49:43 +0100,
Steffen Sauder wrote:
> 
> > With the xmms ALSA driver, this is not a
> >problem.  Unfortunately mplayer and ALSA don't seem to get along with
> >my card (ice1712... maudio delta-44), but with audio disabled seek
> >times are fine.
> >  
> >
> Same problem here again, mplayer's alsa-output has been broken for me 
> since I have that card (1.5 years). Audio seems to be running fine, but 
> video plays at about one fps, and after few seconds the 
> "buggy-audio-driver" message pops up. Because mplayer is the only app 
> having problem with the alsa-driver, I always thought it was mplayer's 
> bug, but I didn't dare to post on their mailinglists :o).

ok, finally i can reproduce this bug, too.
this happens when the plug layer is used, e.g. sample-rate or format
conversion is needed.  that's why this doesn't appear on most
machines.

the attached patch should fix the problem.

(but of course, it has nothing to do with the oss-emulation
 problem...  please let me know if you find the exact alsa version
 getting broken.) 

Takashi

[-- Attachment #2: pcm-plugin-avail-fix.dif --]
[-- Type: application/octet-stream, Size: 769 bytes --]

Index: alsa-lib/src/pcm/pcm_plugin.c
===================================================================
RCS file: /suse/tiwai/cvs/alsa/alsa-lib/src/pcm/pcm_plugin.c,v
retrieving revision 1.44
diff -u -r1.44 pcm_plugin.c
--- alsa-lib/src/pcm/pcm_plugin.c	25 Jul 2003 16:56:37 -0000	1.44
+++ alsa-lib/src/pcm/pcm_plugin.c	28 Oct 2003 17:56:39 -0000
@@ -622,8 +622,10 @@
 	}
 	status->appl_ptr = *pcm->appl.ptr;
 	status->hw_ptr = *pcm->hw.ptr;
-	status->avail = pcm->buffer_size;
-	snd_pcm_plugin_delay(pcm, &status->delay);
+	if (plugin->client_frames) {
+		status->delay = plugin->client_frames(pcm, status->delay);
+		status->avail = plugin->client_frames(pcm, status->avail);
+	}
 	if (!snd_atomic_read_ok(&ratom)) {
 		snd_atomic_read_wait(&ratom);
 		goto _again;