On 2014-09-21 12:28, Alexander E. Patrakov wrote: > 21.09.2014 16:19, Tanu Kaskinen wrote: >> Patch review status updated: >> http://www.freedesktop.org/wiki/Software/PulseAudio/PatchStatus/ > > "Reduce hardware pointer update syscalls" is not exactly "waiting for > review". I have tested it and found it to fail an assertion when the > sink is autosuspended. Also I did some profiling using perf, but testing > on one card does not give enough justification to reject the patch on > the "no performance improvement" basis. It does look like a valid > cleanup and simplification attempt, though. > > Of course this cannot count as a proper review, but, due to the failed > assertion, the need for a new version of the patch is quite obvious at > this point. I'm currently waiting for Peter to see if he notices any substantial perfomance improvements on his card(s). If he does, it makes sense to submit a new, and more robust, version. If not, maybe it's easier to just drop the idea (no use optimising things that don't take much time anyway). -- David Henningsson, Canonical Ltd. https://launchpad.net/~diwic
