=== Aki-Thinkpad is now known as akiva-thinkpad
=== akiva-thinkpad is now known as Akiva-Thinkpad
=== Akiva-Thinkpad is now known as akiva-mobile
=== akiva-mobile is now known as Akiva-Thinkpad
=== chihchun is now known as chihchun_afk
=== chihchun_afk is now known as chihchun
[08:08] <duflu> alf___: ping
[08:08] <alf___> duflu: hey
=== alf___ is now known as alf_
[08:08] <alf_> my tail grew while I was away
[08:09] <duflu> alf_: Hey can you refresh my memory about where to find the Qt compositor code that determines how long compositor buffers are held?
[08:10] <duflu> Or if you can't remember, I shall dig it up later
[08:11] <alf_> duflu: I am not sure I get your question, but related code is in lp:qtmir (i.e. the glue between qt an mir used by QtCompositor)
[08:12] <duflu> alf_: Thanks. It's been quite a while since I jumped between projects
[08:13] <alf_> duflu: by "Qt compositor" you mean the new version, right (THE "QtCompositor")?
[08:15] <duflu> alf_: I was away for the changeover. Only vaguely familiar with the fact that things are changing there
[08:17] <alan_g> RAOF: provides was in the incantation mterry gave me. And my reading of the documentation didn't make it clear that it was wrong.
[08:17] <alan_g> Thanks for the fix
[08:18] <duflu> Oh render_surfaces isn't green any more?
[08:18] <duflu> Horrible green isn't fashionable?
[08:19] <alan_g> ...a shame it didn't work
[08:22] <alf_> duflu: lp:qtmir , src/modules/Unity/Application/mirsurfaceitem.cpp should be a good start for what you are looking for. Note, again, that this is the code for the upcoming "QtCompositor" Gerry is working on, *not* what's in the images right now.
[08:26] <alf_> duflu: I think the relevant code for the current version is in lp:unity-mir
[08:30] <duflu> alf_: Thanks, good to know the difference
[09:41] <alf_> duflu: I have managed to reproduce 1335481 locally. Are you investigating or should I look into it?
[09:41] <alf_> duflu: bug 1335481
[09:41] <ubot5> bug 1335481 in mir (Ubuntu) "unity-system-compositor crashed with SIGABRT - assertion failed at buffer_queue.cpp:136 - "!pending_client_notifications.empty()"" [Critical,Triaged] https://launchpad.net/bugs/1335481
[09:41] <duflu> alf_: Go nuts, I'm not working on it
[09:41] <alf_> duflu: ack
[10:44] <mlankhorst> is mir broken on nouveau? getting thread 1 stuck on pthread_mutex_lock in #7  mir::scene::Observers::add_observer (this=this@entry=0xc54ee8, observer=std::shared_ptr (count 1, weak 0) 0xc55978) at /build/buildd/mir-0.4.1+14.10.20140714/src/server/scene/surface_stack.cpp:285
[10:53] <duflu> mlankhorst: I was considering retesting more hardwares this week. Will do so by the weekend
[10:53] <duflu> rsalveti: Got some fun for you
[10:53] <alan_g> mlankhorst: that looks like it could be a problem we fixed in -r 1764 (can't find the bug reference just now).
[10:54] <alan_g> I don't think the fix has quite made it to archive.
[10:54] <duflu> rsalveti: https://bugs.launchpad.net/ubuntu/+source/libhybris/+bug/1343198
[10:54] <ubot5> Ubuntu bug 1343198 in Mir "[android] Mir spends 12%+ CPU time in get_hooked_symbol (libhybris-common)" [Medium,In progress]
[10:54] <mlankhorst> alan_g: is there a ppa for testing?
[10:55] <alan_g> camako: ^
[11:03] <camako> mlankhorst, https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-018
[11:07] <mlankhorst> I thought the train was crashed by a aiplane?
=== desrt is now known as desrt|ses
=== chihchun is now known as chihchun_afk
=== dandrader_ is now known as dandrader
=== dandrader is now known as dandrader|afk
[13:59] <tedg> dednick, So the silo18 has everything in it now?
=== dandrader|afk is now known as dandrader
[14:00]  * tedg notices a new branch added there
[14:24] <dednick> tedg: new branch? didnt realise
[14:24] <dednick> tedg: it's still missing a bit from unity8
[14:24] <tedg> dednick, I think so, but I'm loosing track of which ones I need :-)
[14:26] <tedg> dednick, Oh, is that in a silo somewhere?
[14:27] <dednick> tedg: i'm not really sure. things are a bit fluid at the moment. Saviq?
[14:27] <Saviq> dednick, it could be added now
[14:27] <tedg> dednick, So I got silo 18 installed but I'm not seeing a socket in the runtime dir.
[14:27] <Saviq> dednick, previous landing *just* happened
[14:27] <tedg> Is that expected?
[14:28] <tedg> Saviq, While you're here, when does dash get it's own PID?
[14:28] <dednick> tedg: that's the bit that's missing.
[14:28] <Saviq> tedg, soon after qtcomp
[14:28] <Saviq> tedg, meaning a week, max two away
[14:29] <dednick> tedg: if you add a MIR_SERVER_PROMPT_FILE=1 to your unity8.conf upstart file it will work.
[14:29] <tedg> Oh, my. I was hoping that was going to be an hour, max two away...
[14:29] <tedg> dednick, So can I ask the trusted prompt session to overlay on the Unity8 PID?
[14:30] <tedg> Or is that crossing the streams?
[14:30] <dednick> tedg: no, it wont work on the unity8 pid. only an application
[14:31] <tedg> dednick, Uhg, okay.
[14:31] <tedg> Saviq, Can we split the dash out before the qtcompositor landing?
=== alan_g is now known as alan_g|tea
[14:31] <tedg> I realize it won't be as beautiful…
[14:33] <Saviq> tedg, no we can't
[14:33] <tedg> Saviq, If I ask really nicely. PLEASE! :-)
[14:33]  * tedg gets some kitten pictures off the Internet
[14:34] <Saviq> tedg, really, no, it's already done based on qtcomp, and qtcomp is landing within days
[14:34] <Saviq> and is more important
[14:35] <tedg> Hmm, okay.
=== greyback is now known as greyback|lunch
=== ara is now known as Guest35403
=== greyback|lunch is now known as greyback
=== alan_g is now known as alan_g|EOD
=== dandrader is now known as dandrader|lunch
=== dandrader|lunch is now known as dandrader