'Twas brillig, and Lennart Poettering at 17/02/10 02:36 did gyre and gimble: > On Thu, 11.02.10 23:34, Colin Guthrie (gmane at colin.guthr.ie) wrote: > >> >> 'Twas brillig, and Colin Guthrie at 11/02/10 22:52 did gyre and gimble: >>> 'Twas brillig, and pl bossart at 11/02/10 21:41 did gyre and gimble: >>>> Hi Lennart, hope you had a good time in India. >>>> paplay seems to be broken in git master, git bisect tells me the culprit is: >>>> >>>> fb55798a3eb73b4c00225232408b766f74533409 is first bad commit >>>> commit fb55798a3eb73b4c00225232408b766f74533409 >>>> Author: Lennart Poettering <lennart at poettering.net> >>>> Date: Fri Jan 15 01:25:21 2010 +0100 >>>> >>>> pacat: allow configuration of latency in msec >>>> >>>> With this commit paplay is stuck, there's no audio out and no update >>>> of time/latency. >>> >>> Ahh so that's where the problem lies... I'll try and fix if Lennart >>> doesn't beat me to it. >> >> This fixes it for me: >> ff2091b pacat: Don't use any buffer attr if we don't set any >> latency/process time params >> >> May not be the nicest fix in the world but it'll do until Lennart can >> review properly :) > > Actually I had I really wanted to write what I wrote. i.e. passing > NULL as buffer_attr means "use the default latency" (which is > 250ms). Passing a buffer_attr with all values set to -1 means "i don't > care about latency" (which ideally means 2s latency). > > So, yes, I actually meant what I did in fb55798a... Now the question > is why this doesn't work for Pierre. For clarity (not sure if I mentioned it) but I did experience the same problem too, so it wasn't just Pierre that was affected. To be fair tho', I did only test it with small samples - e.g. less than 2s, so perhaps the problem is that the buffer simply never gets full enough to start playing? 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/]