On Sat, 2012-07-28 at 16:32 +0300, Tanu Kaskinen wrote: > on my free time. I'd like to work on the bugs too, but it's lower in > priority. So, if you want to get bugs fixed, the most reliable way to > achieve that is to submit patches. You seem to have spent quite a lot of > time studying the PulseAudio code already - would you be willing to > prepare proper patches for the bugs that you've found? Not that much time - I investigated bugs 2 and 4 one evening without prior familiarity with PulseAudio code, then implemented the client-side workaround for 2 and noticed and analyzed bug 3 another day. Fixing things like bug 4 should be quicker for people who are already familiar with the program and have a vision of how things architecturally _should_ work (not just producing correct output in current case). Anyway, I now wrote patches for bugs 3 and 4 plus one bug I had not analyzed before (attached). > Answer to "do people test the low-latency case only": speaking for > myself, my audio needs are mostly covered by Totem and Iceweasel (flash > and html5), and I haven't had audio problems. I'm not sure what > latencies those request, so let's check... Totem wants 200 ms latency > and Iceweasel wants 500 ms (both flash and html5). I think 500 ms should be enough to very clearly notice the rewind bug (number 4). Maybe it's a simple implementation that doesn't use cork for accurate pause? BTW I think that's the same bug an mplayer2 user had reported as PulseAudio bug #48943; the talk about Nautilus in that bug report probably boils down to "do something else long enough for the sink to be suspended". -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-core-fix-underrun_for-calculation-when-resampling.patch Type: text/x-patch Size: 1979 bytes Desc: not available URL: <http://lists.freedesktop.org/archives/pulseaudio-discuss/attachments/20120729/fa469432/attachment.bin> -------------- next part -------------- A non-text attachment was scrubbed... Name: 0002-core-fix-broken-rewinds-after-sink-suspend.patch Type: text/x-patch Size: 4160 bytes Desc: not available URL: <http://lists.freedesktop.org/archives/pulseaudio-discuss/attachments/20120729/fa469432/attachment-0001.bin> -------------- next part -------------- A non-text attachment was scrubbed... Name: 0003-core-adjust-playing_for-and-underrun_for-at-rewind.patch Type: text/x-patch Size: 3709 bytes Desc: not available URL: <http://lists.freedesktop.org/archives/pulseaudio-discuss/attachments/20120729/fa469432/attachment-0002.bin>
