[00:05] deathcrawler: you need to enable it [00:06] deathcrawler: but it's on the build already [00:06] thanks === chihchun_afk is now known as chihchun [01:51] robert_ancell: I also had this xorg crash that needs triage https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/1231507 [01:51] Error: ubuntu bug 1231507 not found [04:17] what causes '(WW) "xmir" is not to be loaded by default. Skipping.' ? === chihchun is now known as chihchun_afk [04:43] jrr: Missing package xserver-xorg-xmir ? === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [08:34] * duflu reboots for some live image testing [08:47] alan_g: Did you resolve bug 1216237? [08:48] bug 1216237 in Mir "Mir silently overwrites and reuses /tmp/mir_socket, rendering the previous server useless" [Medium,Triaged] https://launchpad.net/bugs/1216237 [08:48] Or someone, somewhere [08:48] duflu: not intentionally [08:48] alan_g: How about https://code.launchpad.net/~alan-griffiths/mir/socket-connection/+merge/187326 ? [08:49] duflu: that shouldn't have change the default behaviour [08:49] Weird [08:49] are you seeing a change? [08:49] Yes, I get a nice error message now... std::exception::what: bind: Address already in use [08:50] Hence bug resolved apparently [08:50] \o/ [08:50] But resolved, really, by what? [08:51] I guess I'm the only one to touch that code - will check [08:56] duflu: you're right - I did fix it. (By removing some misguided code) [08:57] alan_g: OK, please link branch and update etc [08:58] duflu: ack [09:00] alan_g: So how are we making it clear which branch a bug fix is in? Just say fixed in project Mir (meaning development branch) but not in "mir (Ubuntu)"? [09:01] duflu: I was assuming we'd just reflect the state of lp:mir - it shouldn't be more than a day behind development-branch [09:02] but it hasn't been discussed [09:02] WTF?! [09:02] * duflu looks [09:03] alan_g: You're saying all dev commits will automatically enter lp:mir ?! [09:03] I thought it was only fixes from lp:mir which entered Ubuntu... [09:12] no, development-branch is just for staging changes so that we can control ABI changes more effectively while saucy is put together [09:15] * duflu rolls eyes at lp:mir r1083 [09:15] It's fine so long as you never want to track history or bisect things :( [09:21] duflu: we just work here. [09:30] alan_g: Is it just good luck that I have not "Broken pipe" any more? [09:30] -t [09:31] duflu: probably, I still see it occasionally [09:31] Oh. [09:31] * duflu bisects on dev [09:32] Oh, is this the case you had an MP for? [09:32] that got fixed [09:32] (sort of) [09:32] But racarr is still working on a better version [09:33] alan_g: That's kind of what I'm seeing [09:34] There are other cases where it happens [09:45] alan_g: What's the commit? I can't see it and have resorted to bisecting dev [09:49] duflu: https://code.launchpad.net/~alan-griffiths/mir/socket-connection/+merge/187326 lines 1030-1032 [09:52] alan_g: Annoyingly I can't reproduce it even with the revision before that landed :/ === seb128_ is now known as seb128 [10:09] alf_: oh btw did you get what I said? that it it's probably a bug in the compositor now afaict? :P [10:19] mlankhorst: alf_ is on vacation today [10:20] ah k === mzanetti is now known as mzanetti|lunch === chihchun is now known as chihchun_afk === hikiko is now known as hikiko|lunch === alan_g is now known as alan_g|lunch === mzanetti|lunch is now known as mzanetti === alan_g|lunch is now known as alan_g [12:48] hi greyback [12:48] olli: hey [12:48] did the foregrounding fix make it in :) [12:48] olli: not yet, being tested now [12:48] ok [12:49] how does it look? === hikiko|lunch is now known as hikiko [12:50] olli: there's a bug in teds code, but it's almost working. === pete-woods is now known as pete-woods-foodl [13:38] alan_g, ping [13:38] hi hikiko [13:39] hi [13:39] are you busy? can I ask you something on a test? [13:39] ask away [13:40] well there are some DefaultDisplayServerTextFixture* tests that fail from time to time [13:41] and the problem is that they cant detect the server [13:41] ack [13:42] I wonder if that's related to your task or it's a different problem, it shouldnt be related since we use a mock server isn't it? [13:44] DefaultDisplayServerTextFixture forks a real server with a test config [13:45] so it's the real server that appears and disappears? [13:45] Does it appear? [13:47] well if I run the tests with gdb [13:48] some rare times I see that in launch_server_process [13:48] we get PID 0 and is_test_process becomes false [13:48] does this mean that the server appears? [13:48] I am not very familiar with this part of code [13:51] when there's a fork what gdb does depends on what you've told it to do. If you see fork() return 0, then you're in the child process [13:51] That is you're debugging the server, not the test process [13:52] yes [13:53] so in this case there's a server running isnt it? [13:53] What you'll probably see is that the test process times out waiting for the server to come up before you've told it to continue [13:55] hikiko: the server process may be running, but it needs to create a socket for connections before anything detects it [13:55] and it can't do that while it is stopped in the debugger [13:56] that's interesting :) [13:58] so, what do you think that is happening? the server starts and then ? [13:59] I mean, in case that I dont stop it from gdb [14:00] if I dont add a breakpoint but just print a message when the server is forked I still get some messages [14:00] I don't know what's happening. I'd turn on some logging and see if that gives any clues. [14:01] yes, that's what I'll do then, I ll leave gdb for the moment [14:01] E.g. does the server initialize properly? Does it crash? [14:01] Does it just take a long time to start? [14:03] You know how to control the logs? [14:04] how? I was about to just print debug messages in stderr but I'd like to do it the proper way if you can give me some guidance [14:05] hey guys congratulations mir on maguro is a hell of a lot more stable on the current images [14:06] hikiko: Using the MIR_SERVER_* or MIR_CLIENT_* environment variables - they are documented by ./mir_demo_server_basic --help [14:07] ok, I ll check it :) thanks a lot [14:08] davmor2: thank you!...i am glad to know that.... [14:08] hikiko: And if you do need to use gdb, then this explains the multi-process options: http://sourceware.org/gdb/onlinedocs/gdb/Forks.html [14:08] davmor2: you would be the first person to say such a thing [14:09] hikiko: ....http://unity.ubuntu.com/mir/....or even http://unity.ubuntu.com/mir/component_reports.html [14:09] hikiko: they've been there for ages ? [14:10] kgunn: It's still slower than SF, but at least it doesn't crash if you open an app and the launcher, Used for about 30 minutes no crashes, it was just too slow to use all day :) [14:10] davmor2: got it...would you mind taking a video of that (just opening apps, swiping the ui, launcher, dash etc....and rotation as well?) [14:11] kgunn, I am not looking to trace anything for the moment but thanks I ll have this in mind [14:11] +get a look [14:11] kgunn: rotation is laggy on SF there is a bug for that, mir is slightly slower but not much [14:12] davmor2: yeah...sometimes, the "slow" or "lag" is actually animation _velocities_ set by qt (especially for things that have that rubber band effect) [14:12] kgunn: I'll get on with some work for a bit first then look at doing 2 videos one on SF and one on Mir for comparisons [14:13] kgunn: yeah if you start top in the terminal and rotate qml-scene is using 132% of the cpu which is why it's slow I guess :) [14:14] kgunn: as soon as the rotation finishes qml-scene disappears off the list :) [14:14] davmor2: great....I'd love it if you could also take some measurements.... https://pastebin.canonical.com/98152/ [14:14] davmor2: that'll print render in ms to the console from the device [14:15] davmor2: i had someone measure before and it was like 24 ms or some such [14:16] davmor2: so you get ~45 fps experience....but i'm curious what the spikes are [14:16] kgunn: right I'll do that as I do the video and upload it to u1 for a share [14:16] davmor2: that's awesome...thanks for the insights [14:16] Morning [14:16] racarr: mornin' [14:20] Rats! racarr woke up before I started reviewing things. [14:31] Is Mir going to be enabled on the 13.10 phone Image ? === pete-woods-foodl is now known as pete-woods [14:43] om26er: that is the aim [14:43] kgunn, How can I help ? :) === alan_g is now known as alan_g|tea === alan_g|tea is now known as alan_g [14:56] om26er: thanks for the offer....lemme think on that....is it an open ended offer ? [14:57] kgunn, given my skill set I would most likely be helpful with testing. But maybe there are other things that I might be able to help with. [15:01] kgunn: Standup link? [15:01] Calendar never works for me :( [15:01] kgunn: Nvm it worked [15:02] racarr: this is probably the wrong channel to publish the link [15:20] greyback: Hey! Have time for a quick sync up soonish? [15:20] bye [15:20] Just want to chat with you on my plans for API rework [15:20] racarr: I'm at your disposal [15:20] greyback: Ok lets hangout. ill invite you [15:21] https://plus.google.com/hangouts/_/e458079840664e3f891b11aa2d809d703d9882ca?authuser=1&hl=en [15:21] greyback: ^ [15:35] BTW, is there to remove Mir from a cmake build? [15:35] I mean, to uninstall it from /usr/local [15:45] make uninstall [15:47] racarr, no target "uninstall" [15:54] smartboyhw, you could do: xargs rm < install_manifest.txt [15:55] bschaefer, thanks [15:55] ...I thought we had an uninstall target :( [15:57] racarr: alan_g kdub .... any volunteer to review todays dev branch merge to trunk ? [15:57] why would you ever want to uninstall? ;-) [15:57] https://code.launchpad.net/~mir-team/mir/development-branch/+merge/188086 [16:01] kgunn: LGTM [16:01] sil2100: coming? [16:01] kgunn, me too [16:09] didrocks: I didn't get any invite, so I don't even know where to come ;) [16:09] sil2100: https://plus.google.com/hangouts/_/b23d56bfd9b5678ec007684d3d181c1af9f34002 [16:27] om26er: run it all day log any crashes === jhodapp|afk is now known as jhodapp|lunch === jhodapp|lunch is now known as jhodapp === alan_g is now known as alan_g|EOD [17:31] davmor2, right. I thought it was crashing due to a single reason all those times ;) [17:32] "the many apps opened crash" [17:32] is that fixed ? [17:43] om26er: there was also the open an app and launcher crash that I reported :) === jhodapp is now known as jhodapp|afk [19:40] kdub: didn't you fix this one (thot it was most recent mp) [19:40] https://bugs.launchpad.net/mir/+bug/1231594 [19:40] Ubuntu bug 1231594 in Mir "android test failure in HWCCommon/[0,1].test_hwc_throws_on_blank_or_unblank_error" [High,New] [19:40] oh yeah, should be fixed [19:41] sometimes launchpad changes for me, sometimes not [19:41] havent figured out why [19:41] no worries...i'll change it (launchpad janitor kgunn :) [20:47] racarr: ping === jhodapp|afk is now known as jhodapp [21:48] kgunn_: pong === jhodapp is now known as jhodapp|afk [21:57] racarr, i hate my verizon router...missed this...sorry [21:58] racarr, so mterry is gonna put up an mp for ignoring arg's it doesn't uderstand as a quick fix....was just wanting to see if you could review/approve before weekend [21:58] either you or kevin [21:58] its one of our mir blockers for being default in touch (and some mutual management is kinda hot on seeing it all resolved by Monday) [21:58] :) [21:59] I reviewed the first one then just saw now there is [21:59] a second one that says something about the 13.10 development version but is targeted towards lp:mir [21:59] and contains lots of extra diff [21:59] kgunn_: > I reviewed the first one then just saw now there is [21:59] 22:02 -!- kgunn [~kgunn@pool-96-226-49-103.dllstx.fios.verizon.net] has joined #ubuntu-mir [21:59] 22:02 < racarr> a second one that says something about the 13.10 development version but is targeted towards lp:mir [22:00] 22:02 < racarr> and contains lots of extra diff [22:00] presumably the goal is to land it [22:00] to lp:mir [22:00] racarr, i asked mterry to retarget to dev branch [22:00] he said bbiab...a while ago [22:01] kgunn_: Ok I retargeted it https://code.launchpad.net/~mterry/mir/dev-unregistered-options/+merge/188177 [22:01] racarr, ...huh...you can do that ? [22:01] just waiting to see that the diff is correct [22:01] apparently! [22:01] wow [22:01] * kgunn_ thinks racarr may be a wizard [22:02] boom [22:02] all done [22:02] no but really there is just a button [22:02] resubmit proposal [22:02] I didnt know you could do it until robert ancell did it to my stuff [22:02] racarr, hey thanks!...that's cool...i had no idea, thot the owner had to [22:03] :D [22:15] racarr, kdub, mterry ok...i outta here...have a good weekend [22:15] you too! [22:19] kgunn_: Cheers! You too [22:34] racarr, that's where I put it! [22:34] I manually proposed a new branch... I thought [22:50] racarr, ah I see. I pointed it at lp:mir. Thanks for the fix, doh!