=== chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === fginther` is now known as fginther [04:50] I know everyone's either Sundaying or enroute to London, but have I accidentally broken cross-compile-chroot.sh? [09:50] anpok, hey alan_g just told me that you plan to work on item 5 of https://docs.google.com/a/canonical.com/document/d/12aVPbX5qzr2GqF8yBcc17ZWsD9Z1kQfoSaYi2yMm1rE/edit#heading=h.17wq5abcr8cb [09:50] anpok, is that so? [09:52] W/Adreno-ES20( 2074): : GL_INVALID_OPERATION [09:52] W/Adreno-EGLSUB( 2074): : PMEM_INV_CACHES undefined [09:52] i see a lot of this lately in the logcat output on a mako (N4) ^^^ [10:06] dandrader: yes [10:06] there is just some recuring distraction around the split greeter work [10:43] camako: hello [10:43] camako: welcome [13:34] Are there any known problems with mir/devel? I've tried it on my mako, but the unity8 session never appears on screen (seems to be running fine). What's the easiest way to debug this? Get a display report, I'm guessing. Any other reports that might be useful? [13:39] mterry: to use mir/devel you need some changes in platform-api, unity-mir and unity-system-compositor [13:41] alf__, yup, I built those branches [13:42] mterry: oh, the latest compatibility branches need mir/devel + lp:~alan-griffiths/mir/scene-grabs-Session-and-SessionListener (yes it's a mess) [13:43] alf__, ah... I don't have that one [13:43] * mterry builds more branches [13:46] greyback: How can I get the latest Qt (the one that exhibits the blocking problem) to try on the phone? [13:47] alf__: it's already on your phone [13:47] Qt5.2 has the issue [13:48] greyback: ok, how can I trigger it? [13:49] mterry: I couldn't reproduce the problem you reported as bug 1302689 (but I wrote a test case) - maybe you can provide more details. [13:49] bug 1302689 in Mir "Session raising problems" [Undecided,In progress] https://launchpad.net/bugs/1302689 [13:50] alan_g, hmm OK. Well that's good in a way. Maybe I'm just misusing API or some such in USC. But I have a new problem when I tried to use a newer mir revision where unity8 sessions don't want to appear at all. Working on that (see scrollback) [13:52] alf__: there are compatibility branches without (as well as with) scene-grabs-Session-and-SessionListener. The ones without "proposed" in them match devel [13:52] alf__: handiest is to grab the QML code in https://bugreports.qt-project.org/browse/QTBUG-37677, save it to a file on the device as test.qml. Then as phablet user run "qmlscene test.qml --desktop_file_hint=/usr/share/application/.desktop" [13:52] alf__: that file prints the time every second to your console [13:53] (yes it's a mess) [13:53] alf__: actually let me check something... [13:53] alf__: yeah that's fine [13:54] greyback: thanks [13:54] alf__: ok, now with that program running, hit the power key, and watch the console - the time stops printing after a few seconds [13:54] alf__: we want it so that the time does not stop printing to the console [13:58] greyback: thanks, reproduced [13:58] alf__: np, lemme know if you need a volunteer to test [13:59] alan_g, ah OK. Then maybe I was using right compatibility branches anyway... But no matter. I'll try with all the proposed branches [14:03] mterry: the related changes are largely moves & renames - so if it compiles you've got a consistent set. [14:04] alan_g, that's what I figured / have [14:04] alan_g, so mir/devel is working OK for you in tests? Have you tried with a full stack with unity8 and such? [14:06] mterry: not recently enough [14:07] alan_g, well. It wasn't working great for me. If you have time, I'd appreciate a sanity check on those results. (specifically, I'm seeing no frames on screen, but no errors in logs) [14:08] alan_g, and this is without my split branch nonsense [14:08] mterry: greyback has been using it [14:09] mterry: things were working on Thursday [14:10] greyback, would you mind retesting with tip? I updated on Friday and things got worse for me. But I'm not sure exactly what revision I had before [14:12] mterry: okay [14:51] greyback: could you please test the latest version of lp:~afrantzis/mir/non-blocking-swap-buffers-spike [14:51] greyback: (merged with the version of mir/devel + Alan's branches you are using, of course) === shuduo is now known as shuduo_afk [15:11] greyback: ^^^ also don't forget that you need the change in USC too [15:14] alf__: working on that now [15:14] greyback: thanks === olli_ is now known as olli === chihchun is now known as chihchun_afk === alex_abreu is now known as alex-abrey === alex-abrey is now known as alex-abreu [16:28] alf__: happy with the new names: https://code.launchpad.net/~alan-griffiths/mir/scene-grabs-Session-and-SessionListener/+merge/214275 === chihchun_afk is now known as chihchun [16:41] alan_g: could you take https://code.launchpad.net/~alan-griffiths/unity-mir/compatibility-with-mir-changes/+merge/213079 , branch it at rev 209, and push as 0.1.8 compatible branch [16:41] it compiles for me against 0.1.8 anyway :) [16:42] greyback: your wish will be granted (as soon as I get some CPU cycles on my laptop) [16:43] * duflu warms hands on greyback's laptop vent [16:56] alf__: slow going, but finally tested: bad news though, it's still blocking [16:57] alf__: to confirm, I built your spike branch, platform-api, unity-mir and USC [16:58] after about 3-4 seconds, the small qml test app stops printing to screen (I disabled the process suspend/resume just in case, no difference) [16:58] alf__: it's interesting, as when I unlock the screen, the greeter is visible, but as soon as I tap the screen, the test app appears and starts rendering once again [17:00] s/tap the screen/tap the right edge of the screen/ [19:23] Saviq, so I lost my latest silo. Is there a silo crunch right now or could I get a new one? I've got a plan for avoiding the Mir nonsense I've dealt with recently === zoktar_ is now known as zoktar === ChickenCutlass_ is now known as ChickenCutlass [21:31] is it just me, or is khronos.org down today? === ubot5` is now known as ubot5 [21:43] down for me === fginther is now known as fginther|away