At Sun, 9 Mar 2008 18:20:47 -0400 (EDT), Mike Gorse wrote: > > There was a change in alsa-lib 1.0.16 which looks like it was designed to > make dmix skip samples in the case of underruns, but it causes the first > sample to be skipped since dmix->slave_hw_ptr == dmix->slave_appl_ptr. > The following patch fixes this and fixes a small typo in the comment: A good catch! Applied to HG tree now. Thanks. Takashi > > --- src/pcm/pcm_dmix.c.orig 2008-01-18 20:00:10.000000000 -0500 > +++ src/pcm/pcm_dmix.c 2008-03-09 18:01:13.000000000 -0400 > @@ -311,9 +311,9 @@ > if (size >= pcm->boundary / 2) > size = pcm->boundary - size; > > - /* the slave_app_ptr can be far behing the slave_hw_ptr */ > + /* the slave_app_ptr can be far behind the slave_hw_ptr */ > /* reduce mixing and errors here - just skip not catched writes */ > - if (dmix->slave_hw_ptr < dmix->slave_appl_ptr) > + if (dmix->slave_hw_ptr <= dmix->slave_appl_ptr) > slave_size = dmix->slave_appl_ptr - dmix->slave_hw_ptr; > else > slave_size = dmix->slave_appl_ptr + (dmix->slave_boundary - dmix->slave_hw_ptr); > > Thanks, > -- Mike Gorse / AIM:linvortex / http://mgorse.freeshell.org -- > _______________________________________________ > Alsa-devel mailing list > Alsa-devel@xxxxxxxxxxxxxxxx > http://mailman.alsa-project.org/mailman/listinfo/alsa-devel > _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxx http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
