[00:25] Oh, man. std::tie! === duflu_ is now known as duflu [08:16] is that an outcome of std::vote? :P [09:46] * duflu resizes a surface, and kernel panics trusty [09:52] And again :P [09:55] stop resizing surfaces then ... [10:00] http://search.dilbert.com/comic/Definition%20Of%20Insanity [10:12] RAOF: thanks! Then let me refrase the question, in a more direct way [10:13] RAOF: the Nokia N900 and N9 phones come with EGL drivers (binary only, not open source); is there some chance to get Mir (and Ubuntu Touch) working on these phones? [10:13] they are real Linux phones, not Android [10:14] mardy, what is the graphics buffer management, modesetting api used by those drivers? [10:14] mardy: you can possibly do it with nitdroid drivers but you'd have to suffer from terrible tearing [10:14] plus the problem is that n900 doesn't grok thumb2 due to a HW bug [10:15] tvoss: no idea [10:16] Stskeeps: Nemo is using the original EGL binaries, right? [10:17] Stskeeps: I'm not sure what the problem is; are these binary blobs tied in some way to X11? [10:17] mardy, would be good to find out to answer your question [10:17] mardy: yeah and x11.. we've tried with wayland on there but you need to spend immense time to get proper vsync going on that rather custom stuff [10:17] Stskeeps, +1, reusing x11 interfaces is close to impossible [10:17] so i've practically given up on those devices for any purpose [10:20] Stskeeps: I see... is nitdroid reusing the original drivers, or is it using android drivers taken from some other device (with compatible HW)? [10:21] uses android drivers of dubious origin and licensing [10:21] cool :-) [10:21] Stskeeps: TBH, I don't care about licensing, I would just distribute instructions, not images [10:21] :nod: [10:22] but even then, the problem would be that tearing would be quite bad and direct rendering/hw composition is not that sane on those devices [10:23] again because of the drivers? Or is there some HW limitation (let's forget the N900 for a moment) [10:26] drivers plus custom vsync handling in display/kernel [10:26] mardy, if we do not have the hwc available, results are pretty bad. The same happens for the N7 [10:27] mmm... I guess I won't spend much time on this then :-) [12:49] greyback, ping [12:50] tvoss: pong [12:50] I saw you've added comments to the doc, I'll have a look shortly [12:51] greyback, hey, looking at the scenegraph document right now. I was thinking if we should add an introductory paragraph on different functionalities touched/covered by the scenegraph [12:51] obviously rendering, but also, input, application handling, focus handling etc === alan_g is now known as alan_g|lunch [12:52] which would then allow us to augment the proposals with details on how things would be solved for those areas, too [12:52] tvoss: makes sense [12:52] greyback, cool [12:54] tvoss: after that's done, how best you think to proceed? I've had a meeting with some of the Mir team on the rendering side, I think they were cautiously open to the idea. Input side, I need to get racarr and dandrader and myself in a meeting too. [12:56] greyback, we should decide on a rough plan and have an iteration 0 asap to flesh out remainding details and identify blockers early on [12:56] greyback, separate branch, full steam ahead. Once complete, we refactor into trunk [12:56] greyback, makes sense? [12:57] tvoss: yes. [12:58] tvoss: ok, so I'll meet with racarr and dandrader on input matters this evening. If no major blockers are evident from that, I'll send round a mail with a proposed plan, set up a meeting to discuss it, and if I get go-ahead, execute [12:59] tvoss: how that sound? [13:00] greyback, +1. I'm specifically interested in details on how you plan on updating focus, and thus app/surface/input details, atomically [13:07] tvoss: ok, I'd better get those details sorted === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader === alan_g|lunch is now known as alan_g === dandrader is now known as dandrader|lunch [15:45] Morning [15:50] Afternoon [15:54] Evening [15:55] racarr: heya, have you time soon to talk about Mir and Qt? [15:56] greyback: Ok we have our standup in 4 minutes but besides that [15:56] when is good for you? [15:57] racarr: well it's getting late here, so I'd prefer reasonably soon. dandrader|lunch should be there, he's been looking more at the input side than I have [15:58] greyback: Ok how about in 30 minutes? or tomorrow before the standup [15:59] dandrader|lunch: will you be back in 30 mins? [15:59] racarr: I'll be here anyway, so we can chat [15:59] ok [16:00] Is there no standup now or is my google hangout not working? [16:21] kdub: ping [16:21] kdub: lemme know when you are around! [16:22] greyback: can i just lurk/listen in ? [16:22] kgunn: sure [16:30] greyback: ok hangout? [16:31] racarr: yeah, let's do it. alf_ wanna join? === dandrader|lunch is now known as dandrader [16:31] dandrader: ah, good timing :) [16:32] https://plus.google.com/hangouts/_/7ecpju9thmntft9e183v78t8bc?authuser=1&hl=en [16:32] greyback, I'm back [16:32] dandrader: short notice, but racarr & I are having a meeting now [16:34] greyback: I am in another meeting, will join as soon as I can, don't wait [16:37] kgunn: we're in meeting, can you join? [16:39] yep [16:40] greyback: where you at? [16:40] kgunn: https://plus.google.com/hangouts/_/7ecpju9thmntft9e183v78t8bc?authuser=1&hl=en [16:41] alf_: can you join now? [16:41] ricmm, pong [16:41] greyback: yes, joining [16:43] alan_g: ^^ hangout for the scenegraph:input side of things [16:46] alf_: ack [17:17] hi, how do I enable mir on trusty? [17:18] is there a wiki explaining this? (for desktop) [17:21] gema, this should help: https://wiki.ubuntu.com/Mir/Installing [17:22] tvoss: thanks :D [17:32] kgunn, tvoss: so it looks like unity and java based apps hate xmir on intel [17:32] davmor2, on trusty? [17:33] tvoss: no I haven't done the migration to trust that won't happen to beta time when things settle a bit [17:33] tvoss: this was on Saucy for all the commercial apps [17:34] davmor2, so unity and java apps together fail? [17:35] tvoss: unity3d sorry the games engine and java based apps. Java says things like no GLX and unity3d Flashes like a strobe light [17:36] davmor2, now that is interesting. And it works flawlessly without xmir? [17:37] tvoss: indeed [17:37] davmor2, okay, did you file bugs yet? [17:38] tvoss: https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0Ai33BkOcORLLdGI2dE5hWHN1cXRaOXVRZHdrY2E5eWc&usp=drive_web#gid=0 no bugs filed yet completing the app migration was the primary task which got completed Friday evening [17:39] davmor2, ack and thx, will take a look after dinner === tvoss is now known as tvoss|dinner [17:40] tvoss: if you look for the orange apps from the key those are all the ones that just wouldn't work due to xmir but were released as they worked flawlessly without xmir (which was then nolonger the default) the unity3d apps were let through as they were fine other than the strobe light effect [17:49] tvoss|dinner: http://ubuntuone.com/4EniBLWsk5yxjJYyDjfTGy for the strobe effect [17:50] davmor2: ack will look at after i go for a run [17:51] kgunn: no rush just had time to go back and see if it was xmir or not :) [17:54] it that the DPMS flicker? === alan_g is now known as alan_g|EOD [18:11] brainwash: no idea only happens on the unity3d games I can dig about and find you some that don't cost anything if you want to confirm it and figure it out for yourself :) [18:14] bug 1232343 [18:14] bug 1232343 in XMir "XMir Screen flickering when running steam or SDL games" [High,New] https://launchpad.net/bugs/1232343 [18:18] davmor2: brainwash: this could also be related. https://bugs.launchpad.net/xmir/+bug/1231632 [18:18] Ubuntu bug 1231632 in XMir "Setting DPMS mode for output to 0 causes screen to blink/flicker once" [Medium,Triaged] [18:20] yeah, annoying stuff === dandrader is now known as dandrader|bbl [18:43] kgunn, ping [18:48] brainwash: I think a lot of the apps that have the problem are window/mac conversions so I'm not sure if it is just the way it calls glx that is causes xmir to reveal a bug that is normally hidden in X or an issue in xmir itself which is also really hard to tell [18:59] ricmm, where is the hwcomposer library that we're trying to get working? [18:59] i think it was removed from the nex7 build [19:09] has anyone ran the mir server natively on a 32 bit desktop in while? As im getting a failure to create a GBM buffer (for the cursor) similar to this: [19:09] https://bugs.launchpad.net/mir/+bug/1203070 [19:09] Ubuntu bug 1203070 in Unity System Compositor "unity-system-compositor doesn't start on some ati card (with opensource driver)" [Critical,Fix released] [19:11] on an intel chip [19:18] bschaefer, can you please file a bug? [19:19] tvoss|dinner, yes i can [19:22] bschaefer, thx [19:23] np! [19:24] https://bugs.launchpad.net/mir/+bug/1247943 [19:24] Ubuntu bug 1247943 in Mir "Mir server failing to start when compiled for 32bit" [Undecided,New] [19:24] hopefully RAOF or duflu could take a look [19:29] kdub: I think it uses the default hwcomposer lib [19:49] tvoss|dinner: pong [19:52] need to reboot...got some desktop weirdness happening === dandrader|bbl is now known as dandrader