On 17 November 2015 at 12:50, Alexander E. Patrakov <patrakov at gmail.com> wrote: > 17.11.2015 12:18, Alexander E. Patrakov пиÑ?еÑ?: >> >> 17.11.2015 12:13, arun at accosted.net wrote: >>> >>> From: Arun Raghavan <git at arunraghavan.net> >>> >>> The drain reporting improvements that were added to alsa-sink were only >>> being applied to directly connected sink inputs. This patch makes the >>> same logic also recurse down the filter hierarchy, so drains are >>> acknowledged more accurately (and not late) even if there is a filter >>> sink in between. >>> >>> Also does some minor reorganisation of the code and sprinkles in some >>> comments as documentation. >> >> >> Thanks for the patch. Do I understand correctly that it supersedes this >> one? >> >> http://thread.gmane.org/gmane.comp.audio.pulseaudio.general/21582 >> > > If GMANE is timing out for you: see > http://lists.freedesktop.org/archives/pulseaudio-discuss/2014-October/021898.html Thanks for pointing this out, I had missed it. David points out two problems: 1. The filter hands out silence if it can't get any data, and his fix disables that -- my fix forces the sink to see pending underruns across the filter hierarchy, so we don't need to propagate the underrun explicitly (but perhaps there is value in doing this anyway, I might be missing some other cases) 2. The process_underrun() callback needs to be chained -- my patch allows this to happen without modifying every filter sink implementation So imo my fix should supersede the patch you refer to, and maybe we should reason out whether to propagate underruns up the sink input hierarchy as a separate patch. Regards, Arun
