=== jalcine is now known as webjadmin_ === webjadmin_ is now known as jalcine === jalcine is now known as webjadmin_ === webjadmin_ is now known as jalcine [00:27] gah... [00:27] I hate it when apport makes a public bug a dupe of a private one [00:27] especially a private one I can't see [00:38] thumper: as a workaround you can ask in #ubuntu-bugs for a specific bug to be publicized [00:38] jbicha: thanks === hggdh_ is now known as hggdh [00:40] jbicha: I thought the SONAME bump was being reverted for cogl [00:44] jbicha: per the bug cogl 1.9.10 was going to revert the bump, so why upload now? [00:45] micahg: the soname for 1.9.8 was cogl9, after that upload, the clutter devs bumped it again to cogl10 & then I complained so they reverted that change [00:45] cogl5 is currently in the archives so they've been making lots of changes this cycle [00:46] jbicha: ah, ok [00:46] it was already up to 10? [00:46] wow [00:46] well it's better than gjs which doesn't bother changing their soname [00:47] they must bump the soname every time they add something to the changelog [00:49] every tarball release where the API was changed [00:50] they change the api alot, must mean they aren't satisfied with it [00:51] * micahg is glad someone else is dealing with gjs/gnome-shell at this point :) [00:57] bjsnider: welcome to intel driver pain in your desktop packages, git master or bust :) [01:01] is network-manager not meant to remember passwords now ? [01:03] lifeless: certainly does here [01:03] hmm [01:03] well, it isn't here :( [01:03] i just switched to another ap and rebooted an hour ago [01:04] and when I select edit on the network connectin in 'edt connections' it says 'Error initializing editor\nNo agents were available for this request.' [01:05] jbicha: I think you just need libgl1-mesa-dev on arm* as well [01:05] jbicha: but you should verify that with janimo [01:06] * micahg can actually try it in a chroot [01:06] something wrong with the last mesa upload? we didnt touch libgl1-mesa-dev, just egl [01:09] thumper, which bug is it? [01:10] micahg: yeah GL_PACK_INVERT_MESA is in gl [01:11] micahg: actually mesa-common-dev [01:12] Sarvatt: right, but libgl1-mesa-dev pulls that in on all the other archs [01:12] * micahg guesses technically it would just be better to add mesa-common-dev explicitly [01:12] but you dont need libGL.so, mesa-common-dev might be more correct (but could be wrong) [01:31] hi Ursinha, sorry didn't notice [01:31] I should file a bug about quassel not keeping urgency [01:31] although I do wonder if it is Qt or just Quassel [01:36] Ursinha: sorry, can't find the bug now :( [01:49] cyphermox: do you have a ppa for oneiric with backported fixes for bug #869635 yet? i think i'm getting hit at least by the modemmanager bug, so i may set one up if not [01:49] Launchpad bug 869635 in wpasupplicant "long delay at shutdown/reboot - network-manager doesn't close correctly" [Undecided,Confirmed] https://launchpad.net/bugs/869635 [01:49] my computer can't do hw virtualization so I don't know if I can run an arm chroot [01:49] jbicha: HW virt has nothing to do with emulating arm. [01:49] hw virt only helps with running x86 VMs. [01:50] so it doesn't need kvm? [01:50] jbicha: i think you just need to install qemu-user-static [02:21] broder: I installed that and ran sudo sbuild-createchroot --keyring=/etc/apt/trusted.gpg --arch=armhf precise /build/chroot/precise-armhf-sbuild http://ports.ubuntu.com/ubuntu-ports/ [02:21] but I get Failure trying to run: chroot /build/chroot/precise-armhf-sbuild mount -t proc proc /proc [02:27] chroot: failed to run command `mount': No such file or directory [02:48] jbicha: huh. i've never tried with sbuild-createchroot. i always use mk-sbuild [02:58] I was just using http://wiki.debian.org/sbuild [02:59] jbicha: i'm running it now to make sure it still works, but i think `mk-sbuild --arch=armhf precise` will work with no other options [03:00] once you have qemu-user-static installed so that you can run armhf binaries [03:01] that command is a bit shorter, trying that now [03:02] mk-sbuild also has a --eatmydata option if you're into that === jalcine is now known as JackyAlcine === JackyAlcine is now known as jalcine [03:56] does the HUD any time i tap alt quickly enough, regardless of other keypresses? it seems to if i press alt-left arrow or alt-right arrow quickly enough (for back/forward in chrome), which is really irritating [04:00] ah, yes. bug #934610. /me subscribes [04:00] Launchpad bug 934610 in unity "HUD - Typing a key combination (e.g. Alt-x) fast enough triggers the HUD" [High,Confirmed] https://launchpad.net/bugs/934610 [04:00] broder, unrelated... [04:00] can I get glxinfo from you [04:01] I need to collect enough samples to get a feel for driver support for a new unity feature... [04:01] sure: http://paste.ubuntu.com/882762/ [04:03] DBO: that's my desktop. i can get you my laptop, too, if i ever manage to unbreak apt [04:04] broder, the laptop is of higher interest to me [04:04] though I find it odd your nvidia card is not getting GL 4.0 support [04:04] still we only care about 3.0 [04:04] it's a pretty crappy card [04:05] 9500 GT is plenty fast for what I need [04:05] in fact its downright overkill [04:05] i just don't know if i should expect the card itself to have GL 4.0, but i don't really know what i'm talking about when it comes to this sort of thing [04:06] broder, its mostly a driver support issue [04:06] huh, ok [04:06] still, 32 stream units 8.8 billion texels/sec fill rate [04:06] you should be able to kill the effects [04:12] DBO: do you know if checkbox or ubuntu friendly or something is getting glxinfo data? that might be a better source of broad data than asking me :) [04:12] no idea really [04:13] even if it were I dont know if I would have access [04:13] you could talk to ara [04:13] yeah I was hoping to luck upon someone with a G45 chip tonight [04:13] that would seal me requirements quite nicely [04:14] that's...intel g45? [04:14] yeah [04:15] sometimes called the x4500 [04:15] or GMA4500 [04:15] sometimes HD is tacked on the end [04:15] all the same thing for what I care about [04:15] how does this look: 00:02.0 VGA compatible controller [0300]: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller [8086:2a42] (rev 07) [04:16] err, hmm, that laptop has a...G45 audio codec, i think? [04:16] * micahg has that same chiupset [04:17] thats perfect [04:17] give me that [04:17] glxinfo [04:18] micahg: can you get it? i don't have precise installed on that machine at present [04:19] * micahg installs mesa-utils [04:22] DBO: http://paste.ubuntu.com/882776/ [04:22] oh thats *perfect* [04:22] I can just make the cutoff opengl 3.0 support [04:22] no custom rules [04:23] I love it [04:23] thank you micahg [04:23] DBO: you're welcome [04:24] hey RAOF saw your name against the lightdm corruption after boot...how is that going? [04:24] robert_ancell: ^^ [04:26] jasoncwarner_, still getting that annoying white box thingie? [04:26] DBO..yeah [04:26] awesome... [04:27] though the above is a bit different...but yes, getting the white box thingy for sure [04:27] DBO: fwiw, I saw that today [04:27] DBO: any idea yet on that one? makes me soooooo angry ;) [04:27] no [04:27] assuming it's the same white box, thumper thought it probably was :) [04:27] jasoncwarner_, does it block inputs? [04:28] DBO: block input? no, I can do my normal computing stuff, just have this stupid white box over everything including teh dash and launcher etc [04:28] oh oh oh oh I know what [04:28] open uhm [04:29] the on screen keyboard [04:29] that should render above the dash [04:29] does that also render above the white box? [04:30] i don't have the whtie box at the moment, but as soon as I do, I'll test that [04:30] (shoudln't be long now..rebooted about an hour ago ;) ) [04:32] wish I thought of that a while ago [04:32] that will immediately point to the problem area [04:33] jasoncwarner_: I've been firefighting other things. I'll kick into that now. [04:33] RAOF, isn't that as simple as clearing the framebuffer? [04:34] DBO: No; not if you want a seamless transition. [04:35] RAOF, oh... ok :P [04:35] DBO: There isn't a modeset between lightdm and the regular session; it's not X not having anything to draw after a modeset, it's what unity-greeter has drawn to the root window going mad after unity-greeter quits. [04:36] Oh, huh. I thought we couldn't advertise GL 3.0 without S3TC compression support. Apparently we do! [04:37] RAOF, it's GL [04:37] lying is never out of the question [04:38] Heh. [04:38] * RAOF is not at all surprised that the nvidia 9600 doesn't support GL 4.0; it's not a DX11 part. [04:40] DBO: By the way, the Intel sandybridge chips *also* support GL 3.0 on Precise. Just on the off-chance you're hoping 3.0 support implies a certain level of performance :) [04:41] RAOF, I am hoping that 3.0 implies newer nvidia/amd or sandybridge [04:41] but not any older intel [04:41] I believe that is correct [04:42] I think that's correct, too. [04:42] What effect are you trying to do? [04:47] hopefully nothing that requires GLX_ARB_create_context [04:48] Sarvatt, actually we are simply using the 3.0 as an evil performance cutoff [04:49] since apparently running a small benchmark on a users machine is evil... [04:49] DBO: every time you start a unity session or something? that would be evil :P [04:50] pfft it would take like 20ms [04:50] render a couple frames into an FBO [04:50] time it [04:50] done [04:50] That seems like a reasonable solution. [04:51] yeah considered evil... [04:51] Although perhaps not *during* boot. [04:51] nah was going to do it on the first alt-tab [04:51] * RAOF was thinking storing it as configuration, and doing it on first boot. [04:51] that would not account for hardware upgrades [04:51] driver improvements [04:51] Match up to GL renderer string, and you can cache it. [04:52] or us just improving the performance of the effect [04:52] lord knows we are doing an expensive effect that maybe some smart mathy person can make cheaper [04:52] if gl version string isnt' the same, retest? [04:52] Would require a package upgrade, which can trivially trigger a re-benchmark. [04:54] Can we argue with whoever said benchmarking to determine what level of effects should be attempted? :) [04:54] yeah... [04:54] still... [04:54] it was implied to be evil! [04:54] brb [04:54] Doing it in a hot-path isn't going to be a particularly awesome move, but you don't *have* to do it in a hot path :) [04:55] actually [04:55] bed now [04:55] cheers [04:56] G'night! [04:56] Dear universe: why isn't there an IDE that uses autofoo as its project description. [04:58] would be funny if querying the cached gl version string took more time than doing the benchmark though, could totally see that :) [04:59] RAOF: lol good question. [05:01] Sarvatt: Eh, they're already querying the GL vendor string as a part of startup; that info is free. [05:47] Good Morning. [05:48] pretty funny, In fedora 18 the scrollbar reminds me more of ubuntu than anything else [06:04] Good morning [06:08] morning [06:26] good morning [06:36] didrocks: bonjour [06:36] didrocks: ca va? [06:36] guten morgen pitti [06:37] I'm fine, thanks! And you? [06:38] quite well indeed! [06:38] didrocks: I don't dare asking you my question.. [06:40] pitti: hum, what question? :) [06:40] didrocks: what do I need to reset in unity to get your alt fix? [06:41] pitti: euh, nothing normally, just logging out/in after upgrading should be fine [06:41] not working for you? [06:41] (working well here and confirmed on the french forum) [06:41] it still neither works in terminal nor firefox [06:41] hum… [06:41] I got unity 5.6.0-0ubuntu3 yesterday evening, and have booted this morning [06:41] I wonder if you don't have a keybinding blocking [06:42] (a compiz one) [06:42] unity --reset, but that will reset all your compiz keybindings [06:42] probably ok, I can restore them [06:46] keep me posted [07:19] doctor appointment, bbl [07:20] pitti: FYI, on alt + keys, seems that daniel and you still have it not working, so we are looking into it [07:26] didrocks: this may be related to pitti's issue, but alt-f10 doesn't work for me. expected or part of the same bug? [07:27] (nothing happens, and some percentage of the time compiz seems to lock up entirely) [07:47] broder: you mean, it doesn't show the appmenu for you? [07:51] broder: I had that once yesterday, but a second try didn't reproduce === tkamppeter_ is now known as tkamppeter [08:52] hey [08:54] bonjour seb128 [08:56] hey pitti, wie gehts? [08:56] gut, danke! und dir? (I'm in my 1-on-1) [08:57] pitti, gut, danke! (heh, finish, my 1:1 is supposed to be in 3 minutes :p) [09:06] seb128: he's all your's; sorry for running over :) [09:06] pitti, no worry, I've time it's my start of day ;-) [09:06] it's for jasoncwarner_ that it's starting being late :p [09:06] nice, dpkg -P python-gobject now is only held back by four packages [09:07] I filed a debian bug for python-zeitgeist (didn't want to introduce a delta for this) [09:07] update-manager and system-config-printer-gnome need porting to GI [09:07] (or s-c-p dropped, rather) [09:07] and python-gst0.10 is only due to sessioninstaller and rhythmbox [09:07] at some point we want to make these work with GI gst [09:08] Greetings. [09:09] oh, the rb dependency is obsolete [09:09] hey didrocks [09:09] pitti: is it just python-gobject -> python-gi? [09:09] hey hey rickspencer3 [09:10] so, when I rebooted, the super key didn;t work [09:10] then opened an app? [09:10] I opened the dash manually, then it worked [09:10] and super key works? [09:10] didrocks, do you want me to look for a rope for you? [09:10] didrocks, sort of [09:10] mvo: no, update-manager-core actually has a few "gobject" imports [09:10] mvo: once that's ported, we can switch [09:10] didrocks, when xchat was busy, it wouldn't work [09:10] seb128: sometimes, I wish people would directly annoy dx, yeah :) [09:11] rickspencer3: interesting, one sec [09:11] seb128, in my defense, I asked didrocks already if he wanted me to check it out ;) [09:11] so, it wasn't a randome ping [09:11] rickspencer3: can you please add and +1 on bug #953089? [09:11] Launchpad bug 953089 in unity-distro-priority "Unity 5.6: Super key doesn't start the dash" [Medium,Fix committed] https://launchpad.net/bugs/953089 [09:11] didrocks, yeah [09:11] rickspencer3: then, I'll ping duflu about it [09:11] didrocks: I think you need an IRC autoresponder /(unity|alt|super).*bug/ -> "please ask in #unity" [09:12] anything else I can do to help, let me know [09:12] pitti: well, if only upstream was answering on their channel :/ [09:12] mvo: and gtk as well, of course [09:13] pitti: oh, let me check [09:14] pitti: that is the old DistUpgrade/DistUpgradeViewGtk.py we actually need it like this to support lucid->precise upgrades [09:14] mvo: I moved all packages which use GI to python-gi yesterday [09:14] pitti: or am I missing something? [09:14] good morning everyone [09:14] mvo: well, almost all, there were three which I sent a debian bug for [09:15] mvo: ah, I see [09:15] mvo: well, it's fine; we don't _need_ to drop it in 12.04 yet [09:15] whoever is on precise and running firefox could probably switch to the firefox-next PPA now, to start testing the first post-release update for 12.04 ;) [09:15] mvo: I just did that dependency cleanup yesterday so that it's easier to see what's left to do [09:15] mvo: I figure we can drop the static bits in 12.10? [09:15] pitti: indeed, we can't as its still needed for the upgrade, once 12.04 is out we can drop it [09:15] mvo: perfect [09:15] pitti: great, thanks for the double-check! [09:16] so s-c-p won't be ported, but I figure we'll use the GNOME upstream printing UI in 12.10 [09:17] mvo: I think I already tried moving the gst bits in sessioninstaller to GI, but the gst GIR is not working very well [09:17] anyway, no need to spend vast amount of effort on this now [09:18] mvo: btw, any chance we can look at bug 950676 this week? [09:18] Launchpad bug 950676 in apt "lucid->precise upgrade failure due to gir1.0->gir1.2 conflicts" [High,New] https://launchpad.net/bugs/950676 [09:20] pitti: yes! [09:20] killall -9 unity ; unity & should be sufficient after an upgrade, right? [09:21] ok, unity --replace is needed [09:24] didrocks: i expect alt-f10 to maximize the window, but it doesn't [09:26] aha! apparently i'm supposed to retrain myself to use super-up. didn't notice that before - must not have done a unity --reset yet [09:27] broder: alt-f10 doesn't maximize, it's supposed now to open the appmenu [09:29] didrocks: yep, got it. i was having problems immediately after upgrading, so i checked ccsm and alt-f10 was still listed as being the maximize. since then i've done a unity --reset, and it's now showing super-up [09:29] so it seems like i was left without a functioning maximize shortcut immediately after upgrading, which seems suboptimal [09:29] broder: yeah, compiz isn't great for migrating settings unfortunatly [09:29] chop the last 4 words from that sentence ;) [09:29] * chrisccoulson trolling [09:30] :-) [09:30] seems like it might be worth just coming up with some condition where it's reasonable to reset the settings on upgrade. maybe first time you login to the ubuntu session or something? [09:31] hey chrisccoulson, how are you? [09:31] hi seb128. i'm good thanks, but a bit tired. how are you? [09:32] chrisccoulson, lucky you, you can troll without being trolled back, I can't remember having hit any firefox bug this cycle :p [09:32] heh [09:32] chrisccoulson, same here, good but tired, I'm working on it through coffee though ;-) [09:32] yeah, it's fairly solid atm [09:32] and i even run nightly builds :) [09:33] chrisccoulson, time for you to tackle your g-s-d login measurement workitem then! ;-) [09:33] yeah, i'll do that later. just getting firefox 12 beta stuff ready :) [09:33] (already) [09:33] chrisccoulson, is 12 the one which will have some ui changes? [09:34] seb128, what sort of changes? it looks fairly similar to the current version here [09:34] chrisccoulson, dunno, I read some article online which stated they were looking at changing the ui, how tabs are rendered and a few other things [09:34] oh, the theme refresh [09:35] no, that hasn't happened yet [09:35] if that happens, then it won't be before firefox 14, seeing as nothing has landed on trunk yet [09:35] ok [09:35] chrisccoulson, didn't they want to change the startpage as well? [09:35] didrocks: so, unity --reset destroyed all my settings, but alt+left/up are still broken in ffox and weechat :/ [09:36] i'm not sure what's going on with the startpage. i thought they were going to have a home tab, which would be the access point to the app store [09:36] anyway, will deal with this later on, need to run to the supermarked quickly [09:36] but i don't really know what is happening with that just yet [09:36] pitti: yeah, duflu found another place where it can causes this [09:36] chrisccoulson, ok, nothing fancy in 12 yet then ;-) [09:36] pitti: so maybe another today [09:36] what happens to the startpage is probably going to be driven by the requirements of https://wiki.mozilla.org/Web_Apps_integration [09:37] chrisccoulson, is mozilla trying to turn my browser into an OS? ;-) [09:37] lol [09:38] and I can't even say "I will stay on the lts", you will bring the new crack there for us :p [09:38] i really want to provide our own branded local startpage, so that we benefit from some of the features of the upstream startpage [09:38] as it stands, we will miss out on features that are accessed via the upstream startpage [09:39] right [09:39] chrisccoulson, btw did you get anywhere about the search entry not respecting your local for searches? [09:39] or rather google not respecting your locale [09:39] not yet. that reminds me, i need to email someone at google about that [10:06] wow, unlucky day for me :/ [10:07] rickspencer3, what happened? [10:07] I did a dist-upgrade yesterday, eveyrthing was fine [10:07] booted this morning, got a disk check ... then unity didn't quite finish loading [10:07] this is on my workstation [10:07] I did a hard reboot [10:07] I suppose I should have ssh'ed in and gotten logs and stuff [10:07] dang [10:08] worked perfectly this time :) [10:08] brb === Omega is now known as mightbereptar [10:33] pitti, hi [10:33] hello tkamppeter [10:34] pitti, I have committed cups-filters 1.0.5 now. Canb you upload it to Debian and Ubuntu? It fixes all PostScript printer problems. [10:34] tkamppeter: nice! so did it help to reduce the PS level? [10:36] tkamppeter: I'll have a look at debian bug 663564, and then uplaod [10:36] Debian bug 663564 in cups-filters "cups-filters: 'make clean' leaves some object files over" [Normal,Open] http://bugs.debian.org/663564 [10:36] tkamppeter: that's also something which sohuld go upstream [10:38] * desrt yawns [10:42] pitti, reducing of the PS level did not help. The solution was adding certain workaround code to deactivate some PS features which the interpreters do not like. [10:42] desrt, hey, wth, waking up at the middle of the night now? ;-) [10:42] can't sleep :) [10:42] pitti, The "make clean" issue and some improvement on texttopdf will come in 1.0.6. [10:42] hey desrt [10:43] tkamppeter: ah, so should I leave this bug alone for now? [10:43] pitti, please upload 1.0.5 as it is now, as the bugs it fixes are of high importance. [10:43] ack [10:43] pitti: good morning [10:43] seb128: how's glib? [10:44] i notice you did a subtle increase of the priority on my XUL bug :) [10:44] desrt, glib is just building, I cleared night backlog first [10:45] sounds good. looks like no new crisis for me overnight [10:45] desrt, XUL> no, I just added an Ubuntu component to the bug and make it match the upstream settings [10:46] * desrt notices some nice new gnome upgrades and wonders if it's jbicha or ricotz [10:47] remind me again: what is the dpkg equivalent for cat /var/lib/dpkg/info/${pkgname}.list? [10:49] desrt, dpkg -L binary? [10:50] thanks [10:50] looks like the new gnome-shell upload has problems :( [10:50] desrt, what sort of problems? [10:51] JS ERROR: !!! message = '"missing ; before statement"' [10:51] JS ERROR: !!! Exception was: SyntaxError: missing ; before statement [10:51] JS ERROR: !!! lineNumber = '12' [10:51] JS ERROR: !!! fileName = '"/usr/share/gnome-shell/js/misc/config.js"' [10:51] JS ERROR: !!! stack = '"@/usr/share/gnome-shell/js/ui/main.js:15 [10:51] these sort [10:51] urg [10:51] const HAVE_NETWORKMANAGER = @HAVE_NETWORKMANAGER@; [10:51] oof. [10:51] hum [10:52] i'm not a javascript genius or anything... but that looks more like automake than javascript :) [10:52] desrt, so interesting, new cogl got uploaded and binaries got rejected [10:52] I guess that's because it failed to build on armel and somebody figured it should build everywhere or nowhere ;-) [10:52] what does it mean for binaries to be rejected? [10:52] desrt, that they will never reach you [10:53] curious [10:53] how is that related? [10:53] desrt, related to what? [10:53] the gnome-shell failure [10:53] oh, I'm not speaking about your bug [10:53] ah. okay [10:53] i though you had an explanation :) [10:53] desrt, rather commenting on the "new gnome upgrades" [10:54] desrt, we were > < that close to get the new clutter in :p [10:54] desrt, which meant the new gnome-shell in [10:54] oh [10:54] it's just that armel ruined your party [10:54] so now i have a new ultra-high priority bug :) [10:54] got a log? [10:54] desrt, https://launchpadlibrarian.net/96702927/buildlog_ubuntu-precise-armel.cogl_1.9.8-0ubuntu1_FAILEDTOBUILD.txt.gz [10:55] did you report that already? [10:55] desrt, jbicha used https://bugs.launchpad.net/ubuntu/+source/clutter-1.0/+bug/941617 [10:55] Launchpad bug 941617 in clutter-1.0 "FFe: Update clutter/cogl to 1.9" [Wishlist,Confirmed] [10:56] desrt, oh, upstream, dunno if he did [10:56] ahh [10:56] no arm builds for ppa, i'm guessing [10:56] desrt, no [10:56] good morning [10:57] jbicha: good morning [10:57] hey jbicha, how are you? [10:57] jbicha: bad gnome-shell upload? :( [10:58] desrt, hum, glib with your cancellable patches failed to build [10:58] ah. it's a good morning for everybody [10:58] seb128: test failures? [10:59] desrt, http://pastebin.ubuntu.com/883085/ [10:59] seb128: you applied one patch and forgot the other [10:59] desrt, bah, some days I hate quilt [11:00] desrt, sorry about that [11:00] :) [11:00] no worries. only increased my heart rate for 1 minute :) [11:00] I had both but I ls >> serie (forgot the 's') [11:00] ahh [11:01] hum [11:01] most curious! [11:01] ./cogl/cogl.c:#ifndef GL_PACK_INVERT_MESA [11:01] ./cogl/cogl.c:#define GL_PACK_INVERT_MESA 0x8758 [11:02] desrt: do you have an ARM chroot? I tried to make one yesterday but I was getting memory errors [11:02] jbicha: i have an arm system, in fact [11:03] jbicha: the fix here is quite obvious, though [11:03] desrt, mhr3: btw the libgee warning issue is fixed after yesterday's "rebuild with current vala" update [11:03] jbicha: the code that uses GL_PACK_INVERT_MESA was originally added to cogl.c with this #ifndef business above [11:03] http://irclogs.ubuntu.com/2012/03/14/%23ubuntu-desktop.html#t01:05 [11:03] the code was later moved to cogl-framebuffer.c [11:03] seb128, awesome, thx [11:04] but the #ifndef didn't follow it [11:04] the #ifndef just needs to be moved over [11:05] pitti, \o/ syncdaemon bug [11:05] just pushed the fix upstream [11:05] seb128: with that I'm on par with you again! [11:05] pitti, ;-) [11:05] actually no, I uploaded a fix in ubuntu-defaults-builder this morning [11:05] so, +1 [11:06] pitti, I've 3 fixes lined up for upload [11:06] * pitti sighs [11:06] pitti, you need to get an extra 2 this afternoon :p [11:06] jbicha: checking what's the deal with the mesa headers on ARM [11:06] clearly there is some expectation that this symbol may not exist [11:07] jbicha: have you reported the bug upstream yet? i can do so, with a patch, if you like [11:08] desrt: no, I had assumed I was missing some build-depends, thanks [11:08] jbicha: seems quite unlikely [11:08] chrisccoulson, don't feed the trolls! (reading your -discuss reply) [11:09] heh :) [11:09] i love feeding the trolls! [11:09] ;-) [11:10] jbicha: recent bug. happened on feb 25 [11:15] jbicha: https://bugzilla.gnome.org/show_bug.cgi?id=672038 [11:15] Gnome bug 672038 in general "Move the fallback define for GL_PACK_INVERT_MESA" [Normal,Unconfirmed] [11:15] this actually looks vaguely like a mesa bug.... [11:16] but i don't know enough to say [11:16] because in the mesa headers you see this: [11:16] #ifndef GL_MESA_pack_invert #define GL_PACK_INVERT_MESA 0x8758 #endif [11:16] and later on, this: [11:16] #ifndef GL_MESA_pack_invert #define GL_MESA_pack_invert 1 #endif [11:16] if those were in the other order, this would to working fine [11:17] desrt: have you test built on ARM yet? [11:17] nope [11:17] let me do that. [11:17] i think it may take a while [11:18] my arm system isn't the zippiest thing :) [11:20] desrt, ok, so glib seems to work fine for me, you have another chance to stop me before I upload it :p [11:20] seb128: i didn't find any problems either [11:20] desrt, ok, let's upload [11:21] hello mvo, i committed some fixes to the aptdaemon branch [11:22] glatzor, hey [11:23] jbicha: test-building now [11:23] desrt: Timo attached almost the same patch to the Ubuntu bug :) [11:23] hmmm, i make the same assumption in firefox as dbusmenu is making for bug 953562 [11:23] Launchpad bug 953562 in glib2.0 "dbusmenu assumes GDBus cancellation is reliable (and it's not)" [High,Fix committed] https://launchpad.net/bugs/953562 [11:23] jbicha: the problem is pretty obvious [11:24] chrisccoulson, so you should be happy with the new upload ;-) [11:24] devscripts has the worst recommends list of anything in the entire archive :p [11:25] at least it doesn't install postfix anymore :) [11:25] hey seb128 ! how are you? [11:26] glatzor, I'm thanks, how are you? [11:26] *** parse error [11:26] seb128, fine. except that I have to leave for work now :/ :) [11:26] glatzor, ok, ttyl! [11:26] desrt, ? [11:27] 07:26 < seb128> glatzor, I'm thanks, how are you? [11:27] ups, "good" ;-) [11:27] so i looked at the background thing last night, btw [11:27] it's not as easy as all that [11:28] mostly because lightdm does its image loading in a thread (as not to block the main UI thread when switching users) [11:28] and gnomebg is not threadsafe [11:28] seb128: hey. hold the press on your glib upload. [11:28] (unless it's too late) [11:28] desrt, not too late [11:29] i have another few patches for you :p [11:29] desrt, the log ones? [11:29] ya [11:29] desrt, yeah, I noticed in git, both are needed? [11:29] there are 3, actually [11:29] although the one that's not in the bug is a windows fix [11:29] and very embarassing :) [11:30] desrt, I don't care about win bugs for Ubuntu uploads :p [11:30] desrt, the buffer size one? [11:30] ya [11:30] pretty hilarious [11:30] yeah, "don't care" ;-) [11:30] "allocate an 80 bytes buffer" [11:30] "now read a 100 bytes string into it" [11:30] ...2 lines apart [11:31] desrt, ok, I'm uploading that dbusmenu fix for now, I will do another upload later today or tomorrow, I've some packaging stuff to fix, I will include the log stuff then [11:32] cool [11:40] pitti: so, the additional alt fixes seems to work without any regression I spotted [11:40] pitti: will take my chance and upload it [11:40] \o/ [11:40] * pitti hugs didrocks [11:41] * didrocks hugs pitti back [11:43] didrocks: my god... [11:43] didrocks: this caching code you added to gnomebg [11:43] INSANE [11:43] desrt: hum, really? vuntz acked it for my defense :p [11:44] the num_monitor thing is really really difficult to understand [11:44] and possibly broken too [11:44] since it assumes that a particular monitor number will always exist at the same geometry position [11:45] jbicha: cogl test build is looking good... it's past the compiling and onto the docs [11:45] desrt: hum, the unicity is done on num monitor/geometry transformation/resolution [11:46] didrocks: so consider that i have a 'span' configuration with monitor 0 (1920x1200) on the left and monitor 1 (same) on the right [11:46] then i change the order [11:47] the filenames should probably encode the relative position of each monitor -- not just its number [11:48] and for completeness, also the total resolution of the desktop (since that could theoretically have an impact in the span case as well) [11:48] unless you have a reason to believe i am wrong :) [11:48] * desrt has never seen it actually do the wrong thing [11:50] desrt: I don't see how the span configuration changes things? As there is a support for only one wallpaper per monitor from what I know of? [11:50] the wallpapers can't span across monitors, isn't it? [11:50] yes. they can. [11:50] that's the 'spanned' option [11:51] desrt: oh really? Never noticed the option. So in this case, you're right, encoding the relative position and total resolution would be needed [11:52] didrocks: i'm sort of wondering why you bother with the monitor number at all [11:52] didrocks: it seems in the 'spanned' case i end up with a file like so: [11:52] 0_6_3840_1200_3243108fa99adb7361797c9c64f0021d [11:52] ie: the total resolution of both monitors [11:52] desrt: the monitor was added at the end of the review [11:52] from a french man leaving in Grenoble :) [11:52] which means the monitor number thing is ... not really used, i guess [11:52] ahh [11:52] vuntz added that part? [11:52] the idea was "maybe one day, we will support multiple wallpapers" [11:53] *headdesk* [11:53] don't remember if I added them or he did TBH [11:53] quite 2 years old :) [11:53] but remember that was added at the end [11:53] triggered an offset issue [11:53] and then fixed :) [11:53] * desrt was trying to make sense of GnomeBG yesterday for adding to lightdm (so we can have slideshows supported) [11:54] * didrocks tries to grab the bug reference [11:54] desrt: but yeah, not having the monitor number prevent this case (which isn't used today) [11:55] https://bugzilla.gnome.org/show_bug.cgi?id=608419 [11:55] Gnome bug 608419 in libgnome-desktop "Caching wallpaper resize to avoid some CPU cycle at startup" [Enhancement,Resolved: fixed] [11:55] in any case, i don't think it matters too much [11:55] i've stopped trying for now :) [11:55] getting GnomeBG into lightdm would be ... much less trivial than i expected [11:56] and there are more important things to do :) [11:56] jbicha: the cogl build is working [11:56] agreed :) [11:57] well, changing the monitor side, if they have exactly the same resolution and you use "spanned" wallpaper [11:57] the probability is quite low :) [11:57] not really [11:58] when you hook up two new monitors for the first time there's a 50/50 chance that they end up the right way around [11:58] hum, really? I remember digging into the glib monitor doc for that purpose [11:58] and I thought it was quite clear from there that the number were preserved [11:59] yes. that's exactly the problem :) [11:59] the number is preserved -- so it used to be in the order (0 1) and now it's in the order (1 0) [11:59] I meant, the number per monitor [12:00] like, the monitor tagged 0 will always be 0 :) [12:00] or you mean, when you switch them? [12:00] yes. exactly [12:00] maybe i plug in the monitor physically on the right side and it becomes 0 and the desktop decides it is the one on the left [12:00] ah yeah, in this case, even if the resolution is different on both, you will get the bug [12:00] hum, tricky :) [12:01] well, i'm not sure [12:01] because there is another bug that seems to prevent that, as mentioned :) [12:01] in the span case, it seems that only monitor number 0 is ever seen [12:01] with the total resolution of the entire desktop appearing in the cache filename [12:01] which is.. unexpected :) === MacSlow is now known as MacSlow|lunch [12:01] indeed :) [12:02] caching is also never done for slideshows [12:02] and it could be... [12:02] yeah, that's in a FIXME in the code, or TODO :) [12:02] ie: cache the scaled source images [12:02] before blending [12:02] one after another [12:02] we need to think about the cleanup though [12:02] ya.... [12:03] I'm still very surprised by the speed we gained with that [12:03] quite unexpected [12:03] scaling is not cheap [12:03] well, actually [12:03] the clever thing to do would be to scale on the GPU [12:03] indeed :) [12:04] but i guess that might not work if you can't upload the original image as a texture because it's too large [12:04] or if you're in a software-only situation [12:04] it's not like the GPU is totally unfamiliar with this kind of operations :) [12:04] large FBO support is better, so maybe in few years [12:04] yo/sup [12:04] (apart from cheap netbooks) [12:05] heh [12:05] i totally pictured you writing that patch on your cheap netbook :) [12:05] desrt: no, it was on my cheap laptop! :) [12:05] desrt: even if it's totally true that most of the time, I do my GNOME patches on my netbook (while travelling) [12:05] but not this one though ;) [12:06] oh wait [12:06] yeah I refactored it for gnome3 on my netbook [12:06] while doing Annecy->Lyon ;) so yeah, half true :) [12:07] ugh [12:07] annoying unity issue: go into a terminal and hold down alt+a [12:07] Could somebody give https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/931397 a look? [12:07] Launchpad bug 931397 in xorg-server "Xorg crashes with AutoAddDevices "false"" [High,Confirmed] [12:07] watch the appmenu flicker.... [12:08] desrt: yeah, this one is tracked [12:08] desrt: (TBH, we have more annoying issues, but shhhhh ;)) [12:09] didrocks: i found a really really cool one yesterday :) [12:09] if it's cool, I'm all hears :) [12:09] http://imgur.com/uGhcd [12:10] that's what happens if you write a program that repeatedly opens/closes a window quickly [12:10] seems the code that dynamically adds/removes the icon for the window to the dash has some race in it [12:11] huho, scary [12:11] please ping DBO about it [12:11] (what's this train icon?) [12:11] it's a custom gnome-terminal profile i have called 'spadina' [12:11] that sshs me to my linode (also called 'spadina') [12:12] makes sense :) [12:12] spadina is a street in downtown toronto. it has streetcars on it. [12:12] they kinda look like that :) [12:12] * desrt needed an icon *shrug* [12:12] ah ok, got it! [12:12] * didrocks still waiting for thunderbird to restart [12:12] (approx. 2min here) [12:12] * didrocks looks at chrisccoulson meanwhile [12:12] http://people.csail.mit.edu/hasinoff/spadiners/510-streetcar.jpg [12:14] desrt: the picture looks accurate [12:17] cnd: You have any status on https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/931397 ? [12:17] Launchpad bug 931397 in xorg-server "Xorg crashes with AutoAddDevices "false"" [High,Confirmed] [12:20] DBO: hey [12:20] DBO: http://imgur.com/uGhcd happens if you run http://fpaste.org/79H2/ for a little while [12:21] even after you quit the program... [12:21] and unity tends to be using a lot of CPU at that point as well (and doesn't go back down) [12:21] restarting unity fixes both issues [12:24] dupondje, hey, I recommend you use #ubuntu-x for xorg issues [12:25] * dupondje joins :) thx [12:34] seb128: thanks for the gio-symbols catch [12:34] desrt, yw ;-) [12:39] pitti, patch for debian bug 663564 is applied to the upstream BZR of cups-filters now, so will be available in 1.0.6. [12:39] Debian bug 663564 in cups-filters "cups-filters: 'make clean' leaves some object files over" [Normal,Open] http://bugs.debian.org/663564 [12:41] tkamppeter: hey, did you get the time to look at why I can't print anymore btw? :) [12:58] seb128: can you approve cogl & gjs from the new queue? [13:00] jbicha, done [13:01] didrocks, this is a known bug, thew update to the new gutenprint does not update your PPD. [13:01] didrocks, for the time being, run [13:02] sudo cups-genppdupdate [13:02] tkamppeter: thanks! Will try that :) === marrusl_ is now known as marrusl === greyback is now known as greyback|lunch === m_conley_away is now known as m_conley [13:16] jbicha: fixed build has been fired off, i guess? [13:20] desrt, cogl? it built and got newed already [13:20] oh. that was fast. [13:21] i guess the builders are not loaded down for the day yet :) [13:21] desrt, seems like builders are more powerful that your arm device ;-) [13:21] desrt, well, archive builders are usually fine [13:21] desrt, the lag is often on ppa builders [13:21] ah. didn't realise PPA was a different set [13:21] i thought they used the same x86(64) ones [13:21] desrt, the ppa ones are virtual buildders with a limit set of archs [13:21] and just the arm ones were for archive-only [13:22] desrt, archive buildders are real hardware builders :p [13:22] :) [13:22] desrt, https://launchpad.net/builders btw [13:22] i was just looking for that [13:22] didn't it used to be called soyuz? [13:23] aw crap. unity crashed again -- in a really interesting way [13:23] no, "soyuz" is the launchpad codename for the buildds part [13:23] it redirects to launchpad.net/launchpad nowadays though [13:23] i can still type into IRC and my screen is updating (i see my music player ticking off seconds) [13:23] but i can't interact with any chrome (like moving windows) [13:23] nor can i change the focus [13:23] *fun* [13:24] good old compiz restart fixes all :) [13:24] we still have lpia builders? fascinating. [13:25] * desrt thought those were done [13:25] tkamppeter: ah, great [13:25] desrt, yeah, and seems people still use it :p [13:26] desrt, the ppa stats have "lpia 1 4 jobs (1 hour 40 minutes) " [13:26] there's only one... so i'm guessing it's only used for some special-cases? [13:26] desrt, likely for people still running builds on ubuntu version where lpia was supported [13:26] not sure, that would be a lamont's question ;-) [13:27] or maybe just "in case we would need one a day for a reason" [13:27] ah. good call. [13:28] ah nice. jbicha did a fixed gnome-shell upload [13:28] didrocks: The following packages will be REMOVED: unity [13:29] The following packages will be upgraded: unity-common [13:29] desrt: you are on amd64? [13:29] yup... [13:29] that's not supposed to happen anymore, right? [13:29] desrt: seems the unity-common in i386 is published [13:29] and the unity package built on amd64 for you not [13:29] desrt: hum, no, it's supposed to happen still :) [13:29] ah. okay. [13:29] so the usual "don't blindly dist-upgrade" :) [13:30] exactly! [13:30] sorry for the bug [13:30] no worry ;) [13:30] at least, you asked before telling yes :p [13:31] * desrt runs apt-get with -y from now on for extra fun [13:31] tip of the day: add --force-yes as well :) [13:31] is that for the "Yes, I really mean to destroy my system" prompt? :) [13:31] ;) [13:32] (btw, when compiz crash, it's normal you can still type on IRC and the screen refreshes by Xorg) [13:32] but as you told, you can't switch the focus [13:32] and of course, no chrome [13:32] well, there was chrome [13:32] i just couldn't interact with it [13:32] like dragging the title bar did nothing [13:32] chrome, like window decoration? [13:32] ya [13:32] ah interesting, that shouldn't happen indeed :) [13:33] pitti, can you upload CUPS to Debian and Ubuntu? It has important fixes on the IPP backend. [13:33] i don't know if it was a crash as much as a weird sort of lockup [13:33] well, lockup, if compiz is still running, freeze the screen most of the time, no more refresh [13:33] ya. that's why i was surprised [13:36] tkamppeter: ok, sure; still fighting with the kfreebsd FTBFS, I'll have another look at this; my first attempt didn't succeed [13:40] tkamppeter: building now === jbicha is now known as Guest97816 === Guest97816 is now known as jbicha_ [13:44] what is the key to enable/disable the display of the content of ~/Desktop on the desktop ? [13:45] jibel, [13:46] org.gnome.desktop.background show-desktop-icons [13:48] seb128, thanks, that's also the key to bug 953089 \o/ [13:48] Launchpad bug 953089 in unity-distro-priority "Unity 5.6: key bindings doesn't work on empty workspace" [Medium,Fix committed] https://launchpad.net/bugs/953089 [13:48] when show-desktop-icons is disabled, no keybinding work [13:49] hum [13:49] why do people do stuff like that?! [13:50] still a bug, but yeah, stupid options keep bitting us [13:53] it also breaks refresh of the title in the menu bar [13:54] jibel, right, we should not let people turn that off [13:54] jibel, I'm pondering forcing the option under unity === greyback|lunch is now known as greyback [13:56] seb128,is it on or off after an upgrade from oneiric ? [13:56] jibel, it was always on in any ubuntu version === MacSlow|lunch is now known as MacSlow [14:22] seb128: not in une though [14:23] (sorry, catching up, was building the list for priority on top of priority to prioritize more bugs) [14:23] huats: are you there [14:23] ? [14:23] so maybe une -> natty -> oneiric -> precise [14:23] found a bug related to you i guess [14:24] but it was the default key, so should be migrated to the new default after update/login [14:26] didrocks: seb128: pitti: are you guys running the new n-m/n-m-g from ~ubuntu-desktop ppa? I'd upload today if possible [14:27] cyphermox: ah, I'm not; seems I missed the announcement/ping for that [14:27] cyphermox: I added the ppa this morning, but didn't get a chance to reboot yet [14:27] * pitti upgrades [14:27] cyphermox, I didn't restart yet, will do in a bit [14:28] thanks guys [14:32] s9iper1: I am [14:32] but not sure I'll be able to answer you [14:33] just query me in private [14:33] I'll answer it once I'llbe able (tonight) [14:59] * desrt does the happy-cogl dance [15:00] jbicha_: looks like the gnome-shell lg crasher is fixed with the latest upload [15:02] desrt: upload? you mean the gnome3 ppa or what? [15:02] yes. i'd imagine so. [15:03] dupondje, no status to report [15:03] wow, yeah it works here too now, I did pull from the Debian packaging & rebuilt a few things.... [15:04] dupondje, why are you running with auto add devices as false? [15:05] jbicha_, I promoted the cogl binary and retried builds [15:05] jbicha_, clutter builds I mean [15:06] cnd: seems like its a required setting for bumblebee [15:06] dupondje, what's bumblebee? [15:07] nvidia optimus implementation. [15:07] why would it be required for it? [15:07] http://www.bumblebee-project.org/ [15:08] Not sure why its needed btw. [15:09] dupondje, where do you see anything that says it is needed? === jbicha_ is now known as jbicha [15:11] https://github.com/Bumblebee-Project/Bumblebee/blob/master/conf/xorg.conf.nouveau [15:12] the bugreporter is the main dev of Bumblebee btw === zyga is now known as zyga-afk [15:34] i'm disappointed that the troll didn't respond to my email on the gwibber thread [15:44] didrocks: rebooted with new unity, and there's still a problem [15:44] cyphermox: new nm/nm-applet from PPA, working fine for WPA2 and system-wide connection [15:44] pitti: thanks [15:45] WPA2 I'm using too, testing WPA2 ad-hoc as a replacement for the WPA ad-hoc fubar [15:45] pitti: :/ we kill latest opportunities though [15:45] works well here, but since yesterday [15:45] didrocks: now that my Alt key is fully working, what do I annoy you with now? [15:45] (weechat in a terminal) [15:46] * pitti hugs didrocks, SCNR [15:46] * didrocks hugs pitti too [15:46] working well in weechat and firefox again!! [15:46] stop doing that to my little heart! [15:46] :) [15:46] phew ;) [15:55] chrisccoulson: I was wondering why you wanted to break my Flash plugin, but it turns out I just needed to update my partner sources.list to precise [15:56] jbicha, it's all part of a larger conspiracy :) [15:57] the conspiracy that the partner repo stays empty most of the development cycle? or the one involving Gwibber? [16:07] hmm, clutter fails on armel & armhf too https://launchpadlibrarian.net/96782888/buildlog_ubuntu-precise-armhf.clutter-1.0_1.9.14-0ubuntu1_FAILEDTOBUILD.txt.gz [16:12] Sweetshark: is bug 906402 still required for precise? [16:12] Launchpad bug 906402 in graphite2 "[MIR] graphite2" [Undecided,Incomplete] https://launchpad.net/bugs/906402 [16:13] jbicha, you're my hero <3 [16:13] pitti: no, IMHO. [16:14] and everyone else working on gnome 3.4 for official repos [16:15] Sweetshark: ah, good; could you please close the bug then, saying that/why it's not needed any more? [16:16] pitti, jibel_: I still cant reproduce bug 916868 -- I just did a complete upgraded from a virgin oneiric install. [16:16] Launchpad bug 916868 in libreoffice "package libreoffice-common 1:3.5.0~beta2-2ubuntu3 failed to install/upgrade: subprocess installed post-installation script returned error exit status 134 (dup-of: 915271)" [Undecided,New] https://launchpad.net/bugs/916868 [16:16] Launchpad bug 915271 in libreoffice "package libreoffice-core 1:3.4.4-0ubuntu2 failed to install/upgrade: rmdir: failed to remove `usr/lib/libreoffice/basis3.4/program/': Directory not empty" [Critical,Fix released] https://launchpad.net/bugs/915271 [16:16] desrt: the clutter failure is upstream too, right? [16:16] jibel, jibel_: Is it possible to run jenkins with debug=70000? [16:27] Sweetshark, sure. first thing tomorrow [16:28] jibel: great! that helps me a bunch. thanks! === zyga-afk is now known as zyga [16:33] cyphermox: seems I stil have network :) [16:36] didrocks: great [16:36] I'll upload this after lunch I think [16:36] others tested this as I fit in two extra bug fixes on top [16:36] cyphermox: yeah, I tried wired and wifi and both are fine here [16:37] nice :) [16:37] anyone tried 3G? (though I did, and I know it works) [16:37] I just ordered 4 new 3G devices to help with testing, and slowly build up my box of pain [16:37] * didrocks doesn't have a 3G card support in the computer [16:38] didrocks: don't worry about it === mao42ranma is now known as maoranma [17:12] good night everyone! [18:03] * didrocks waves good evening [18:05] Is anyone else having trouble clicking on scrollbar grips this past week? [18:12] mterry: if you mean overlay-scrollbar, i've always had problems with that. [18:12] dobey, I do mean o-s, but I've only had recent problems. It goes away too quickly [18:14] ah [18:15] mterry: the fade-out time seems fine here. i have also been experiencing some of my mouse clicks being ignored all over the place, though [18:16] dobey, no, not that. It just goes away before I can get to it [18:17] mterry: it doesn't pop up where your mouse is? [18:18] dobey, well, it does, but kind of to the right, as expected. When I try to move over to it, it goes away [18:18] mterry, do you use a pad? [18:18] mterry, or any touch device? [18:18] seb128, laptop touchpad [18:18] mterry, it's https://bugzilla.gnome.org/show_bug.cgi?id=672009 [18:18] Gnome bug 672009 in gtk "gtk 3.3.18, menu items are not highlighted on mouseover with pads devices" [Normal,New] [18:19] mterry, gtk missing enter events on touch devices [18:19] huh [18:19] likely a side effect of the xinput 2.2 smooth scrolling work [18:19] there is a small testcase with a gtkbutton on the bug [18:19] it only gets the first enter event not the next ones [18:19] works fine if you use a mouse [18:19] seb128, ah, I'm opening an email on d-d-l from you now that sounds related :) [18:19] it's something xinputish [18:20] eh, not quite [18:20] mterry, yeah, the email is gtk doing a slightly incompatible change in their abi [18:20] yeah, i'm not on my laptop [18:20] which is another "fun" issue [18:20] seb128, but cool, good to know it's a known issue anyway [18:21] mterry, still hacking on unity-greeter? as much as I love smooth animation I think it's time you stop investing on that to tackle some bugs ;-) [18:23] seb128, no, done with that. I got as far as I cared to with GTK+. Expect a release tomorrow. Now I'm fixing up quickly's tutorial. And then hopefully bugs (though I'll probably give deja-dup bugs a priority unless you've got something you care about) [18:23] Oh, happy pi day btw! [18:24] :) [18:24] lol, "pi day" [18:24] mterry, we have plenty on the milestoned list ;-) [18:26] how is 14/03/2012 == pi? [18:26] silly 'mericans [18:27] dobey, right, i thought exactly the same: https://twitter.com/#!/chrisccoulson/status/179955372759584769 ;) [18:28] heh [18:28] i like Laney's suggestion [18:28] yeah [18:28] 22/7 ;) [18:28] although, 3.333... isn't really pi either [18:29] not that i oppose the consumption of pie on any day [18:30] dobey: your math is off :P [18:30] eh? [18:30] 22/7 != 3.333 [18:31] uh [18:32] ah right [18:32] anyway, it's not pi [18:32] 22/7 =~ 3.14285 which is closer than 3.14 as was pointed out on slashdot [18:34] π =~ 3.14159 [18:34] heh, "slashdot" [18:34] i know what pi is === chaoticuk_ is now known as chaoticuk [18:36] jbicha, argh gnome-shell does not start anymore [18:36] and pi number of days into the existence of earth, it was still a big ball of plasma [18:36] jbicha, since the cogl/clutter update [18:38] jbicha, I have a backtrace if you want [18:41] jbicha, https://bugs.launchpad.net/ubuntu/+source/cogl/+bug/955347 [18:41] Launchpad bug 955347 in cogl "Crash in gnome-shell" [Undecided,New] [18:57] hey, last December it was mentioned that you guys want a dedicated gnome shell distro [18:57] I made ugr last year, but had to stop due to health issues === chaoticuk_ is now known as chaoticuk [19:18] seb128: ^^ [19:21] soaringsky, hi, yes? [19:21] soaringsky, talk to jbicha and ricotz maybe, they maintain the gnome-shell stack [19:22] soaringsky, it's just lacking a few extra volunteers to work on it and make a proper derivative like xubuntu etc [19:22] jbicha: ricotz: ^^ [19:22] seb128: yeah, I have those volunteers [19:22] great, so wait for jbicha to ping back ;-) [19:23] seb128: I was almost to that point last year, but I got really sick. now I'm starting to get better and working on ugr again [19:24] soaringsky, ok, great, I never understood the ugr stuff, like why doing a separate project rather than an official variant [19:24] but if both efforts can be merged that would be great [19:25] seb128: I couldn't start it out as an official variant. I had to start somewhere [19:25] seb128: I wrote the initial preview in two days. I never could have became official in that time [19:26] well, let's not argue on that ;-) [19:26] seems there is enough people interested to do an official variant [19:26] so that's great [19:26] seb128: but my goal has always been to become Gubuntu :) [19:32] isn't ~/.config/user-dirs.dirs supposed to get sourced on log-in so those things env up in the ENV? [19:32] dobey, no [19:32] hmpf [19:32] dobey, glib has an api giving you those folder which parses the file IIRC [19:33] seb128: so one is just supposed to parse the file? [19:33] i know glib has api [19:33] so who decided to make gcalctool unresizable but also make it wrap the stuff in the top pane ? [19:33] but python doesn't [19:33] that really doesnt fly if you have a long addition to do [19:34] xclaesse: yes I know it's broken, there are several uploads, you can use the gnome3 ppa if you don't want to wait [19:35] soaringsky: well it doesn't start as official, ask Lubuntu about how long that took [19:35] jbicha, that's why you use breaks usually ;-) [19:35] dobey, the point is that you should be using glib [19:35] but an important first step is getting as much as possible in Ubuntu's official repositories [19:35] jbicha: ik, thats why I started unofficial [19:36] seb128: unfortunately that's not so much an option for this. the whole cross-platform thing and all [19:36] seb128: so i guess i have to implement parsing that file [19:36] jbicha, so gnome-shell 3.3 is going to precise soon? [19:36] dobey, right, if you don't want to use standards libs I guess you have to do things yourself [19:36] seb128: could you approve mutter for 386 & amd64 from the new queue? [19:36] jbicha: yeah, I have a few minor things like cantarell and shell tweaks. then its just the metapackage [19:37] seb128: what i *want* to do is not relevant :) [19:37] soaringsky: cantarell is in Ubuntu, I think shell tweaks should be a packaged extension if you need them [19:37] dobey, glib works fine under other platforms you support, you could use it [19:38] dobey, and if some Xsession script was exporting the xdg dir in the environment that wouldn't help you on other platforms anyway [19:38] seb128: feel free to convince people who make that decision. and good luck with that. [19:38] seb128: no, but it would make my life easier on linux [19:38] well you have to implement the parsing anyway [19:38] because os.environ.get() is a lot less code [19:38] so you can as well use it accross platforms [19:39] jbicha: oh, cool. I don't have more than a handful of my own packages. everything else is already in repos [19:39] jbicha, done [19:40] jbicha, bah, clutter ftbfs on armel [19:40] desrt, ^ [19:40] https://launchpadlibrarian.net/96782888/buildlog_ubuntu-precise-armhf.clutter-1.0_1.9.14-0ubuntu1_FAILEDTOBUILD.txt.gz [19:40] "/usr/bin/install: will not overwrite just-created `/build/buildd/clutter-1.0-1.9.14/debian/tmp//usr/lib/arm-linux-gnueabihf/pkgconfig/clutter-glx-1.0.pc' with `clutter-glx-1.0.pc'" [19:40] wth? [19:41] desrt, that was just a fyi seems you watch the new gnome-shell landing [19:42] seb128: yeah, I did mention it failed but I don't really have the ability to fix arm failures :( [19:43] jbicha, seb128, hi [19:43] hey ricotz [19:43] jbicha: back in the day when gnome-shell was young, we had a lot of version in compatability issues with shell extensions. is that still the case? [19:43] jbicha, I see gnome3 ppa also has totem 3.3, that won't be merged into main I guess? [19:44] i dont have time to look into it, but this is caused by my arm patch to unify the pkg-config file name [19:44] soaringsky: my personal opinion is that we shouldn't bother packaging extensions, but point people to extensions.gnome.org [19:44] jbicha: ok [19:44] ricotz: ah, what happens if we don't use that patch? [19:44] seb128, jbicha, the x11 backend installs a clutter-glx.pc too which should be work around somehow [19:45] jbicha, just drop the patch, if there is an issue and some user try to run gnome-shell on arm they will complain [19:45] it's not like your armel gnome-shell userbase was big [19:45] jbicha, dropping it might be fine if all rdepens have caught up to use clutter.pc instead of a backend specific one [19:46] it's maybe desrt only :p [19:46] ricotz, jbicha: if that's not the case you will know soon with rebuilds [19:46] ricotz: ok, I'll drop it and if it's needed we can add it again later [19:46] seb128, some rdepends arent really handling gl vs gles2 [19:47] jbicha, ok [19:47] jbicha, i guess doko will complain if it breaks [19:48] jbicha: extensions.gnome.org looks great. I'll probably code a quick app that has that site embedded [19:48] jbicha, ok FYI adding the gnome3 ppa and gnome-shell 3.3, the crash is fixed :) [19:48] soaringsky: you should talk to GNOME about that [19:48] /usr/bin/install: will not overwrite just-created `/build/buildd/clutter-1.0-1.9.14/debian/tmp//usr/lib/arm-linux-gnueabihf/pkgconfig/clutter-glx-1.0.pc' with `clutter-glx-1.0.pc' [19:48] xclaesse: yes, it's a temporary problem in the normal archives [19:49] seb128: highly curious [19:49] jbicha, next time please make cogl breaks gnome-shell [19:49] seb128: looks like a packaging problem more than anything else? [19:49] desrt, read scrollback, it's ricotz's fault [19:49] jbicha: k, thanks. do you know of any way to have extensions installed by default? [19:49] desrt, or don't read scrollback [19:49] desrt, unping, nothing for you, it's being addressed, just keep the "gnome-shell on its way" info ;-) [19:52] * desrt is very very tired [19:54] soaringsky: Debian enables the alternative status menu (aka "don't hide the shut down button") by default [19:54] http://packages.qa.debian.org/g/gnome-shell-extensions.html [19:57] jbicha: ty [19:58] THOSE BASTARDS [19:58] how dare they modify our perfect UI [19:59] who do they think they are?! [20:26] desrt, perhaps you need to hit them with a trademark stick like some other projects do? That will stop people from modifying your perfect UI ;) [20:27] actually, i shouldn't put ideas in to peoples heads [20:32] chrisccoulson: but then they'll have to ship it as goneme instead of gnome [20:50] hi people [20:50] I have a really annoying problem that effectively stops me being able to use a vt [20:51] I get a bucketload of "iwlwifi 0000:05:00.0: GF was set with SGI:SISO" sent to standard out on the vt [20:51] anywhere from 10 to 20 / second [20:51] can't use gdb with that going on === chaoticuk_ is now known as chaoticuk === JackyAlcine is now known as webjadmin_