=== duflu_ is now known as duflu [01:15] Anyone know if VT switching has been descoped/deprioritized for USC? [01:16] I mean... https://bugs.launchpad.net/mir/+bug/1192429 [01:16] Launchpad bug 1192429 in Unity System Compositor "unity-system-compositor is on multiple simultaneous (and random) VTs" [High,Triaged] [01:50] duflu: Man, is your house built on quicksand or something? [01:51] RAOF: Old suburbs... [01:52] …are nice, but obviously don't have the maintenance thing worked out. [01:54] RAOF: Hey it's better than it used to be. For the first few years here I had a blackout every month or two [01:54] Although *my* old suburb doesn't seem to suffer from those problems (so far‽). They *are* digging up the pavement, but that's to lay pipes for sweet, sweet fibre. [01:54] RAOF: You win :/ [02:47] * RAOF heads out to lunch === tvoss_ is now known as tvos === tvos is now known as tvoss [06:12] hi [06:13] Hi! [06:13] g'day mate [06:13] :) [06:14] RAOF: but what good is fibre if you are limited by laggy intercontinental links. :P [06:15] mlankhorst: All the content that I can get locally is fast?\ [06:15] Also, my uplink will be faster than the current anaemic 200-odd-KB/sec [06:16] And *those* links aren't saturated :) [06:17] Oooh! *That* might be why lseek isn't working! [06:19] Dear launchpad: 250kB/s. Really? [06:19] Error: Launchpad bug 250 could not be found [06:19] _Really?_ [06:22] lol [06:30] RAOF, is there any script (eg ident) that I could run to fix all the formatting errors before I submit a branch? [06:30] indent* [06:31] Not that I'm aware of. [06:32] If you find one or make one (maybe it's possible to make an appropriate indent incantation?) please let me know :) [06:33] sure :) [06:34] Grr. We're going to need to throw away logind. [06:52] good morning [07:00] Good morning dholbach! [07:11] hey RAOF [07:39] ARGH! === tvoss is now known as tvoss_test === tvoss_test is now known as tvoss [08:05] alf__, ping [08:05] tvoss: pong [08:13] RAOF, ping [08:24] ok that should fix the random lockups I was getting on radeon, had too much debug stuff on and the one I wrote broke :( [08:24] robert_ancell: kgunn: I just accepted Mir in universe \o/ [08:25] didrocks, yay! [08:25] didrocks: a serious thanks for all the help [08:26] kgunn: no worry ;) [08:26] olli_: ^ [08:26] boom [08:26] kgunn, missing context [08:26] but I assume something hit universe? [08:27] robert_ancell, ^? [08:27] robert_ancell: kgunn: I just accepted Mir in universe \o/ [08:27] olli_: yeah, "something" :-) [08:27] congrats guys ;-) [08:27] tvoss, ^ [08:28] (it just hit it, didn't crash ;)) [08:28] beautiful landing [08:29] didrocks, \o/ [08:30] didrocks, beautifully work as well if I want to install it? ;-) [08:33] seb128: needs now xmir being in distro and the patches for the drivers [08:33] then u-s-c will land in distro [08:33] ok, a bit of extra waiting, I guess [08:33] I'm sure there will be an email announcing the details once things are in place, going to wait for that ;-) [08:35] be warned, seb128 is ready! [08:36] ;-) [08:57] kgunn, are you held up? [08:59] robert_ancell: Pong [09:00] RAOF, hey, is there still a regression in the -staging PPA that stops us copying to system-compositor-testing? [09:01] Hm, yes. [09:01] The intel driver will regress if we do that. [09:01] robert_ancell: I found the new lightdm in staging broken my (raring) system on the weekend. Had to revert the new lightdm from staging [09:01] RAOF, what needs to be done to fix that? [09:02] robert_ancell: Fix the patch, basically. I'll upload a new intel tomorrow to the distro, and also to staging. [09:02] RAOF, can we revert for now so we can fix asap? [09:02] Actually, if Mir has landed in-distro, maybe I'll just upload straight to distro? [09:02] Is this related to https://bugs.launchpad.net/xmir/+bug/1203776 ? [09:02] Launchpad bug 1203776 in XMir "X crashes w/ xserver-xorg-video-intel 2:2.21.9+xmir5870-1~saucy1 from System Compositor Testing PPA" [Critical,Triaged] [09:02] RAOF, can't until the MIR is complete [09:02] duflu: Yes. [09:02] robert_ancell: Oh, that's right. Not yet in main. [09:03] RAOF, but once that is done, yes [09:05] RAOF, can we revert? [09:05] Looking. [09:06] Oh, that's right. [09:06] I'm not sure that we *do* regress with what's in staging. [09:06] Or, at least, I'm not sure that the bugs in staging are the bugs in -intel :/ [09:07] Anyway, reverting. [09:19] alf__: sees you've re-approved a lot of spurious failures. Is there a pattern to worry about? [09:19] *seems [09:22] alan_g: I am not sure yet, let's see how the second reapproval goes [09:24] RAOF: is the -testing ppa intel issue simply worked around by accelOption = uxa [09:24] in xorg conf [09:24] kgunn: Yes. [10:07] RAOF, is xserver-xorg-video-intel - 2:2.21.9+xmir15871-2~saucy1 the intel driver that should work? [10:07] Yes [10:08] RAOF, ta [10:08] tvoss is smoke testing [10:08] flickering is really bad [10:08] on my machine [10:09] but....push it [10:10] kgunn: With 2:2.21.9+xmir15871-2~saucy1? [10:10] Really? [10:10] That's new as of 5 minutes ago. [10:10] RAOF: oh...need to test that [10:13] robert_ancell: Oh, how do I build lightdm from trunk? [10:13] I know I've asked you this before, but I think logind has gone and become more annoying. [10:13] RAOF, are you getting llvmpipe issues? [10:14] robert_ancell: As in - software rendering? Yes. [10:14] RAOF, can you look in /var/log/auth.log? [10:14] I had these issues last night [10:14] What would be in there? Is *that* where logind does it's logging? [10:15] RAOF, I had pam_systemd failing and reverting to the CK backup. Apparently that doesn't work for DRM anymore [10:15] RAOF, symptons are loginctl showing no sessions, ck-list-sessions showing sessions and the auth.log complaining about a /var/run directory already existing [10:15] robert_ancell: Ah, no. That's not my problem. I've got loginctl showing my session, just that it's not active. [10:16] RAOF, ok. To build lightdm it just get tarball and overlay debian/ from latest package [10:16] yes, yuck [10:16] RAOF, what do you need exactly? [10:17] robert_ancell: I need to write a blueprint, but we basically need to replace logind [10:17] robert_ancell: But http://paste.ubuntu.com/5928462/ should work as a stop-gap. [10:17] RAOF, shouldn't that be XDG_VTNR? [10:17] alf__: I'm looking into the graphicsplatform dependencies on frontend... [10:17] robert_ancell: Possibly. I'll recheck logind code. [10:18] RAOF, and that's now set in src/x-server.c [10:18] Is there any reason InternalNativeSurface::get_parameters() uses the surface, not the buffer to get the size? [10:24] alan_g: At the moment the buffers and the surface have the same properties. I am not sure if/how support for resizing will affect this. [10:26] alf__: Would you be happy if I used the buffer properties and left any resizing issues for the future? [10:28] robert_ancell: You never actually set XDG_VTNR for unity seats. [10:29] RAOF, ah, k [10:30] Because you (quite sensibly) never set the xserver's VT number for unity seats. [10:32] RAOF, this is for fallback? [10:32] No, this is for u-s-c [10:32] RAOF, ah, ok [10:33] alan_g: Sure, but I would also like to get Kevin's opinion later, since he is more familiar with the needs of the internal client. [10:33] I *think* we need to set the vt so that logind doesn't have a screaming hissy fit. [10:33] RAOF, can you file a bug / make branch - I'll just push that into main [10:34] alf__: Sure, I'm just using you as a first sanity check. ;) [10:34] * robert_ancell has to wait again before updating the -testing PPA while mir decides to build *again* [10:35] robert_ancell: Will do, once I've tested that this actually fixes my problem. [10:35] RAOF, k [10:39] alf__: Have you ever seen gcc optimize out functions, causing mock expectation failures? [10:47] thomi, can you block the CI uploading to the staging PPA? [10:47] thomi, there's new revisions landing faster than they're building [10:47] robert_ancell: shouldn't we just let it catch up? [10:48] thomi, we'll never get a finished build at this rate [10:48] it's building 894 at the moment, but 895 is already in trunk [10:49] I need to get a built version and copy it over to -testing [10:49] To the disassembler... :S [10:49] robert_ancell: why can't we copy over the copy that's older than trunk? [10:49] thomi, when I go to copy from the PPA it only shows the building version, not the old one [10:50] hmmm [10:50] didrocks, ^ don't know if we can get the previous built version? [10:51] robert_ancell: we're on it [10:52] duflu: could it be that the mocks don't override the correct signature? (gmock doesn't use "override") [10:53] alan_g: Thought of that. It's a match [10:54] duflu: or the base class forgot "virtual"? [10:55] * alan_g knows these are silly, but so is the symptom [10:55] alan_g: Nope [10:58] This is about the time I start adding #error to prove I'm looking at the right code. ;) [11:00] alan_g, alf__: OK, I forgot LD_LIBRARY_PATH and unit-tests got some symbols from the wrong libmirserver [11:00] What a fine waste of evening [11:01] * alan_g wishes he hadn't done that too [11:02] Of course, with rpath it would have been fine if I ran it on the build machine. But this is cross-compiled [11:26] RAOF: ping [11:27] kgunn: Pong [11:27] RAOF: hey...so robert & i got sunk by the very latest staging [11:27] seems x crashes...and boots us back out to the greeter (over and over) [11:27] i got logs [11:27] Yes please! [11:27] will attach them to a bug [11:31] RAOF: https://pastebin.canonical.com/95196/ [11:31] there's that for the moment...launchpad being fussy [11:31] or our wifi is flakey [11:33] Hm. That's the wrong log :) [11:33] Or, rather, it's lightdm's log, which is useless, and I should really fix it's uselessness. [11:33] kgunn: Got /var/log/Xorg.0.log? [11:34] checking [11:36] RAOF: robert just looked over my shoulder...says "its a clean run"...so the xorg.0.log didn't have any useful info [11:36] kgunn: Ah. [11:36] Xorg.0.log.old might? [11:37] right...was looking in old [11:37] Ah. [11:37] robert's checking to see if he has a log [11:38] he's gonna turn on once more....just to make certain [11:41] RAOF: now he's having issues repro'ing.... [11:41] i'm pretty reliable...i'll give another go [11:43] Cool. [11:44] RAOF, ping [11:45] tvoss: Pong. [11:45] okay, disabling the process group leader thingy helps [11:46] RAOF, Xorg.0.log says [ 6.649] (II) intel(0): SNA compiled: xserver-xorg-video-intel 2:2.21.9+xmir15871-2~saucy1 (Chris Halse Rogers ) [11:46] [ 6.657] (II) intel(0): SNA initialized with Sandybridge (gen6, gt2) backend [11:47] Sweet. And setting XDG_VTNR does indeed fix acceleration, even with the tty detaching. [11:47] RAOF, okay, where do we need to set that? [11:48] RAOF, however, I'm experiencing the slowness issue right from the beginning [11:49] tvoss: In the lightdm branch I'm just proposing. [11:49] RAOF, awesome [11:50] so no need for me to mp removal of the session group leader stuff, right? [11:50] tvoss: Correct. [11:50] cool [11:50] RAOF: hey [11:50] just noticed xorg updated in main [11:50] arg [11:50] ...so new problem :) [11:50] right in the middle of me trying to repro [11:51] :( [11:51] RAOF: how long to rebase your patches on that ?...like...are we torpedo'd for the night ? (e.g. you have to sleep at some point) [11:52] kgunn, where are you? [11:52] in murry2 [11:52] kgunn, wanna hop over to the main ball room? [11:52] tvoss: ack...robert said he'll have the first dance [11:52] kgunn, fair :) [11:52] (seriously...i think maybe a 15min break to eat a bite...but i'll be right there after that) [11:53] RAOF: i'll still try to pin the ppa on staging to avoid the xorg from main....and then try to repro again [11:53] repro the xorg crash that punts me back out to greeter that is [11:54] I'll rebase xserver now. [11:54] I will be sleeping soon :) [12:12] tvoss: New xserver in staging, lightdm branch proposed for merging. [12:12] tvoss: Now, sleep. [12:13] RAOF, ack [12:13] tvoss: Oh, I haven't put up a blueprint for the logind work yet - I'll do that in the morning. [12:14] RAOF, no worries, go to sleep dude [12:14] and thanks for all the help === alan_g is now known as alan_g|lunch === hikiko is now known as hikiko|lunch === alan_g|lunch is now known as alan_g === hikiko|lunch is now known as hikiko [13:21] RAOF: so as a test...i pinned staging, updated...then robert corrected/built https://code.launchpad.net/~raof/lightdm/more-logind-workarounds [13:21] and then built u-s-c (so staging + local lightdm & u-s-c) [13:22] still get the x crash that punts me to greeter [13:22] good news, i gotta xorg.0.log [13:22] will attach to bug [13:23] until then https://pastebin.canonical.com/95204/ [13:23] RAOF: ^ [13:31] mlankhorst: would you possibly be able to take a peek at https://bugs.launchpad.net/mir/+bug/1206508 [13:31] Launchpad bug 1206508 in XMir "Xorg crash on xmir punts user back out to greeter" [Critical,Triaged] [13:31] or rather the xorg log in there [13:32] kgunn: to the valgrind machine! [13:33] tvoss_, https://code.launchpad.net/~robert-ancell/lightdm/unity-session-vtnr/+merge/177591 [13:34] and make sure to install xserver-xorg-core-dbg and xserver-xorg-video-intel-dbg, else valgrind won't be useful === dholbach_ is now known as dholbach [13:40] hikiko: alf__ is right about not needing create_nested_platform(), but doing that requires a definition of NestedPlatform (and a constructor that takes the same parameters we gave the function). Can you do that with a stubbed implementation as an update? [13:44] sure alan_g right after I fix the other issues [14:32] alan_g, alf__ the only problem is that to call the constructor the NestedPlatform must inherit from mg::Platform so I have to add a stub implementation of all Platform methods (which isn't a problem since we ll need it anyway at some point) [14:32] hikiko: ack === alan_g is now known as alan_g|tea [14:33] It seems like there is an opengl packaging conflict between main and system-compositor-testing today [14:35] hi guys [14:36] anyone that has some knowledge about mir I can talk about this patch http://paste.ubuntu.com/5929237/ i made to fix the focus problems i was having with the qml-demo-shell gerry was working on based on unity-mir ? [14:36] racarr: you here already or still to early for you? === alan_g|tea is now known as alan_g [14:42] alf__: ping ↑↑↑ ? [14:43] tsdgeos: hi [14:43] tsdgeos: Looks sensible [14:45] Although at some point we may want hidden surfaces to "contain" points [14:45] Hey guys, I am running Ubuntu Touch on my Galaxy Note. How can I take a screenshot? even from the terminal?. gnome-screenshot doesn't work because it can't open the display (I am assuming this is because Mir is the display server?) [14:45] alf__: alan_g: basically i realize that msh::SingleVisibilityFocusMechanism::set_focus_to "raises" the window just because it hiddes the others, but doesn't "really raise" at the ms::SurfaceStack layer so i need that in contains otherwise the touch is directed to the "previous" window that is hidden but "on top" regarding ms::SurfaceStack [14:45] not sure if i made sense :D [14:46] tsdgeos: You make as much sense as the mechanism you describe [14:46] i'll create a Merge Request and explain the stuff there too [14:46] racarr is in the process of reworking that stuff [14:47] But a quick MP will beat him to trunk [14:47] :D [14:52] alan_g: https://code.launchpad.net/~aacid/mir/hidden_surface_no_contains/+merge/177621 [15:13] bye! [15:15] kgunn, https://launchpad.net/ubuntu/+source/lightdm/1.7.8-0ubuntu1/+build/4837955 [16:22] robert_ancell, https://code.launchpad.net/~thomas-voss/mir/revert-process-group-leader-patch [16:39] tvoss: why? o.o [16:40] looks more like mir needs to handle this case better instead.. [16:46] tvoss: does mir handle SIGINT by any chance? if not it wouldn't surprise me if X was killed because mir was.. [16:48] and if not.. Xmir really shouldn't touch the console in any way [16:57] you need to use KDSKBMUTE ioctl and tcsetattr, fwiw.. === alan_g is now known as alan_g|EOD [17:52] An interesting idea for [17:53] model synchronization... [17:54] rather than taking controller objects in constructor, certain methods could require the controller object to be passed in, i.e. create_surface, destroy_surface, move_surface, etc. [17:55] Then a central object, can lease out controllers and use that to enforce locking. so code could be written sort of like [17:57] ms::UniqueController controller(surfaces) auto surface = session->default_surface() // do stuff // surface.raise(controller) // Surface can't have been destroyed because you have a unique controller [17:57] I dunno ;) it's hazy === owner_ is now known as Guest84665 [18:59] https://orders.letsorderfood.co.uk/order.asp [19:00] http://pizzakingiom.co.uk/ [19:01] zip code IM1 2LY [19:01] robert_ancell, kgunn, tvoss, Saviq ^ [19:16] lol [19:16] @ pizza king iom [19:52] I notived the system compositor testing ppa has some new packages, is it ready for testing? [19:52] bschaefer: ^ [20:16] robotfuel, there are some, but my u-s-c was removed :( [20:17] you can try some out though :) [20:17] bschaefer: it already fixed a bug with opengl and the r600 radeon driver :) [20:18] bschaefer: I just started tests on a bunch of other systems [20:18] hopefully the getting punted to the greeter on intel is fixed. :P [20:22] robotfuel, o thats not good, and a fix would be nice! [20:23] robotfuel, awesome, should I just check jenkins for the test updates? [20:25] bschaefer: the punted to greeter issue was a bad patch from mainstream xorg. [20:25] robotfuel, o man, i've had that happen to me a couple times, then I needed to switch back to UXA and normal X and I've not gotten it yet [20:26] bschaefer: anything finished within the last ~hour minutes should have the latests results from the u-s-c ppa [20:26] I think there is only one [20:26] robotfuel, cool, let me check [20:26] * bschaefer hopes to see an ati one [20:27] bschaefer: yes the hd7450 [20:27] sweet [20:27] * bschaefer goes to look for it [20:28] bschaefer: http://10.97.2.10:8080/job/openarena-benchmark-ps-radeon-hd7450-le-xmir/48/ [20:28] just got there! [20:28] but thanks :) [20:28] robotfuel, which xorg patch is that? [20:29] robotfuel, awesome, it looks like its running xmir, and no errors in any logs i see :) [20:29] bschaefer: 2:1.14.2-0ubuntu3+xmir1 [20:29] ii xserver-common 2:1.14.2-0ubuntu3+xmir1 [20:29] ii xserver-xorg 1:7.7+1ubuntu5 [20:30] * bschaefer reboots his ati machine [20:30] robotfuel, is that the version of xmir on that ati machine? [20:31] * bschaefer has an abi break with u-s-c [20:31] err [20:31] a depends issue rather [20:31] bschaefer: yes http://10.97.2.10:8080/job/openarena-benchmark-ps-radeon-hd7450-le-xmir/48/artifact/results/dpkg-list.log has all the package versions [20:31] robotfuel, sweet, let me double check mine [20:32] robotfuel, but so far its looking like ati is working fine, at lease this run through [20:33] robotfuel, is that using the daily build ppa? [20:33] yes it is nm¡ [20:33] ! [20:33] bschaefer: no that is the unity-system-compositor ppa [20:33] o really? [20:33] 1:0.9.9~daily13.04.18.1~13.04-0ubuntu1 [20:33] made me think it was the daily next one... [20:34] robotfuel, hmm do you have a machine that you can test that out with? [20:34] cause i get this in the daily next build: [20:34] unity-system-compositor : Depends: libmirserver0 (= 0.0.8+13.10.20130729ubuntu.unity.next-0ubuntu1) but 0.0.8+13.10.20130730bzr896saucy0 is to be installed [20:34] * bschaefer hopes thats rebuilt soon [20:39] bschaefer: they all seem to be busy. [20:39] robotfuel, alright, ill do some manual testing then! Thanks [20:49] tvoss, soo daily build next ppa seems to be working for me with xmir [20:50] with an ATI card [20:50] bschaefer, \o/ [20:50] bschaefer, what was the fix? [20:50] bschaefer, we have multiple changes coming to mir/lightdm to fix permission issues and such [20:50] tvoss, im not sure, I just updated and and to downgrade my libmirserver, but things seem to be working... [20:51] tvoss, hmm now im not sure .. this strange I have TTYs [20:51] but: [20:51] root 985 0.5 0.3 178820 15360 tty7 Ssl+ 13:48 0:00 /usr/sbin/unity-system-compositor --enable-input=false --from-dm-fd 10 --to-dm-fd 13 --vt 7 [20:51] is running [20:51] along with: [20:51] root 1017 0.6 0.5 110348 22416 ? Sl 13:48 0:01 /usr/bin/X -core :0 -auth /var/run/lightdm/root/:0 -mir 0 -mirSocket /tmp/mir_socket -nolisten tcp [20:54] tvoss, strange, but I just wanted to quickly update you, im assuming its later over there :) [20:57] bschaefer, it is, thanks for the update :) [20:57] np! [21:05] haha, and ctrl+c kills everything...that seems like a mir thing :) [21:05] * bschaefer assumes mir got tty support [21:21] bschaefer, fix for that is propagating [21:21] tvoss, yup saw you fixed that ~30 min ago :) [21:21] but confirming with bregma and it looks like daily next build is working with xmir [21:21] it works beautifully [21:21] well confirming that ttys are working on xmir [21:22] which made me doubt if it was xmir :) [21:22] bschaefer, we need to reenable the patch I just reverted, but a little more sophisticated [21:23] touchscreen still doesn't work right, but I doublt that has anything to do with Mir [21:23] bregma, yup, still the good ol' x input stack :) [21:23] cool :), but Ill continue to run it to see if I can get anything to go wrong (though things are looking good!) [21:23] bschaefer, awesome, thanks [21:23] np! [21:24] actually, I think my touchscreen problem is nTrig is crap [21:24] bschaefer, would be great if you could hammer it with phoronix-test-suite run pts/gaming-free and pts/gui-toolkits [21:24] tvoss, great idea, let me go run those! [21:35] bschaefer: cracking up at the idea of you hitting the ctl+c :) [21:36] imagine us here....forgetting that bugs exists & hitting ctl+c about every 5 minutes [21:36] haha [21:36] kgunn, yeah i use ctrl+c tooo often as well [21:39] kgunn: no, why did you approve [21:39] it was so easy to fix by just doing a few more CORRECT ioctl's instead of reverting.. [21:40] mlankhorst: because its really late & we need to "do something" - we did test & verified this worked....i tried not to...but unfortunately we proved the ctl+c was an issue [21:41] did you even read what I said in the review? [21:41] mlankhorst: yes... [21:43] mlankhorst: altho reading & knowing exactly what needs to be done 2 diff things [21:43] crtl-c, crtl-v from x.org, apply correct copyright header from xorg, done :P [21:52] mlankhorst, +1, it's on my plate for tomorrow morning first thing [22:29] bschaefer, pingf [22:29] olli, hello! [22:29] bschaefer, the usc-testing ppa is up [22:30] mind giving an ATI system a spin [22:30] olli, awesome, im the daily-build-next ppa and its working on ATI [22:30] kgunn reminds you to install usc at the end [22:30] let me purge, remove and install the u-s-c testing ppa and give that a test [22:30] yup [22:30] bschaefer, cool! [22:30] and to pin the ppa as well :) [22:31] bschaefer, brb [22:31] * bschaefer always forgets that part [22:31] alright [22:31] dammit Ctrl+C! [22:31] adding the mute thing would have been a lot faster :P [22:32] :) [22:32] But much less fun! [22:36] olli, also, a couple hours ago robotfuel had some ATI tests on the u-s-c ppa and things were working as well [22:37] xmir was version: 2:1.14.2-0ubuntu3+xmir1 [22:37] robotfuel, can you give the ppa a spin as well pls? [22:38] olli: ok I'll kick off new tests runs [22:38] awesome, I was just about to ask that :) [22:45] * RAOF is minding Zoë for an hour or so [22:48] dang...well its working on u-s-c testing ppa as well, and ctrl+c no longer kills the server! [22:50] olli, working on u-s-c testing ppa, and ctrl+c no longer kills the server [23:35] bschaefer: it looks like there is new packages worked on the systems [23:36] robotfuel, how new? As I just tested mine an hour or so again, and ctrl+c was no longer killing the server [23:36] bschaefer: from u-s-c [23:36] bschaefer: I didn't try control c [23:36] bschaefer: I ran the openarena benchmark on it. [23:36] cool, im still downloading games to run those... [23:41] bschaefer: I get ret = 0 for buffer 26 [23:42] ret = 0 for buffer 18 [23:42] repeated over 100 times on nvidia, it doesn't seem like openarena is running [23:42] bschaefer: in Xorg.0.log [23:43] :( [23:44] robotfuel, i was about to try running this on daily build next ppa, but there are so many games that it tests... [23:44] * bschaefer goes to look at logs [23:46] bschaefer, ping [23:46] tvoss, hello! [23:46] bschaefer, did you give the system-compositor-testing ppa a spin? [23:47] tvoss, yup! ctrl+c is no longer killing the mir server :), but robotfuel is running into openarena not being drawn? [23:47] or running [23:47] * bschaefer is still downloading games [23:48] for the test suite [23:48] bschaefer, okay, checking locally [23:49] tvoss, alright, let me do that [23:49] tvoss: it ran ok on intel and radeon, on nvidia I get: ret = 0 for buffer 26 [23:49] ret = 0 for buffer 18 repeated 100 times in Xorg.0.log and openarena hasn't finished it's first test in over an hour [23:51] I have a compiz crash Jul 30 22:50:26 ps-nvidia-gt640-he gnome-session[1300]: WARNING: App 'compiz.desktop' respawning too quickly [23:51]