'Twas brillig, and Ozan ?a?layan at 02/06/09 11:38 did gyre and gimble: > Then I've done a bisect between those tags and the problem seems to be > introduced by the following commit: > > commit d5f46e824e3f8a042e6f67dd4c3fc385545edd74 > Author: Lennart Poettering <lennart at poettering.net> > Date: Tue Jan 27 04:39:07 2009 +0100 > > move flat volume logic into the core. while doing so add > n_volume_steps field to sinks/sources > > > Hope this helps, thanks. Very long shot.....: I've not tried this or checked to see if it is correct or not, but looking just at your commit above the only change to alsa-sink.c that I can spot where it may affect things is in the hunk on line 932 in this page in the function sink_get_volume_cb(): http://colin.guthr.ie/git/pulseaudio/diff/src/modules/alsa/alsa-sink.c?id=d5f46e824e3f8a042e6f67dd4c3fc385545edd74 There was a call to: - pa_sink_set_soft_volume(s, &r); removed. I'm not sure if this is meant after the changes or not. I do see that in the sink_get_volume_cb() method it does still call this function, so perhaps this was removed accidentally in sink_get_volume_cb()? You could always try adding in a pa_sink_set_soft_volume(s, &s->soft_volume); As the new form of this call would pass in both s and a member variable contained inside s, I suspect the API change makes this unnecessary now, so this isn't the magical bullet, but I just figured I'd mention it here and you can debug - I have no time at present. The same change is also present in alsa-source.c in that commit (hunk at line 879) Hopefully Lennart will look in more depth and find where the real regression snuck in!! Col -- Colin Guthrie gmane(at)colin.guthr.ie http://colin.guthr.ie/ Day Job: Tribalogic Limited [http://www.tribalogic.net/] Open Source: Mandriva Linux Contributor [http://www.mandriva.com/] PulseAudio Hacker [http://www.pulseaudio.org/] Trac Hacker [http://trac.edgewall.org/]
