=== Amaranth__ is now known as Amaranth [06:58] RAOF, bryceh, Sarvatt: so uploading 1.10 et al will be postponed to next week? [07:03] tjaalton, probably safest bet [07:05] bryceh: ok [07:06] although looks like raof fixed up -geode and -ati, so maybe it's good to go now [07:11] i could upload those when tseliot is around to upload nvidia [07:12] though I think most of the input drivers are rebuilt for no good reason, only evdev and synaptics support XI2.1 [07:15] tjaalton: tseliot is on vacation until monday, he said whats in the PPA is good to go outside of a version change in the changelog [07:15] hi Sarvatt [07:16] on the one hand it seems like it'd be better to wait and do it like Monday morning tjaalton's time, which would mean we'd have one of us around at all times in case of problems [07:17] on the other hand, it seems unlikely that there's going to be problems, and doing it now gives a good bit of additional time for testing before beta [07:19] on the other hand if RAOF has it all under his trigger to rebuild&dput the drivers, I'll gladly let him to do the uploading :) [07:20] he doesn't have coredev yet [07:20] oh right [07:21] tjaalton, hmm, I think the weight of the balance favors having more testing time [07:21] well, to publish them somewhere for dget [07:21] plus you me and sarvatt are going to be around for the next day, we can handle problems [07:21] RAOF: PING! [07:21] :) [07:21] tjaalton, I'm going to boot on -ati now just to doublecheck but otherwise I say go for it :-) [07:22] bryceh: okay [07:22] shouldn't be that late down under.. [07:26] hmm, -ati box got dropped to failsafe mode [07:28] oo, libx11 1.4.2 [07:32] hmm, something's busted with -ati [07:32] it works when I run 'startx' as root or a user, but when gdm starts it, boom\ [07:33] no obvious errors tho [07:33] strange [07:33] the previous version works? [07:34] yep [07:34] well, the version from before updating the X stack worked [07:34] there's 6.14.1 btw [07:34] not packaged yet [07:34] obviously downgrading would not go on this xserver [07:34] right [07:40] sheesh, why no errors anywhere? It doesn't even generate a new Xorg.0.log, all that gets created is the failsafe-x one [07:44] tjaalton, do you have a ati system handy you could test putting the PPA on? [07:45] bryceh: not set up, I do have an old firegl and an agp-board, but they're not in a case or anything [07:46] let me try booting earlier kernel [07:47] no go [07:47] however [07:47] since startx works... I think the problem may not be X [07:47] right [07:48] nothing in the gdm log either? [07:48] I notice the session with startx loads with a gnome-terminal open but no decorations, no panel, no background [07:48] no gdm log is produced [07:49] hmm [07:50] .xsession-errors shows "Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0. [07:50] maybe startx isn't setting up auth [07:53] oh [07:53] heh [07:54] when I upgraded to the ppa, since -ati wasn't there, apparently apt decided to uninstall all of gnome [07:54] hah [07:55] * bryceh apt-get install ubuntu-desktop ;-) [07:56] tjaalton: PONG! [07:57] RAOF: :) [07:57] RAOF: so we were thinking about pushing the final abi to the archive [07:58] ok, desktop happy now :-) [07:58] could I just pull all the packages from the ppa and push them as-is [07:58] without the rebuild hassle [07:59] RAOF: oh, and isn't only evdev and synaptics needing rebuilds for the new XI? [07:59] the ppa has all the input drivers [07:59] rebuilt [08:05] what package provides window decorations? [08:05] bryceh: unity? [08:06] no, got that installed [08:07] compiz-gnome? [08:07] aha, point to RAOF [08:07] tjaalton: Actually, none of the input drivers need rebuilding; that was an artefact of my over-eager script. [08:07] RAOF: heh, ok that's good [08:08] alrighty, that looks proper [08:08] Let me run through http://cooperteam.net/Packages and check that all's good to go. [08:08] RAOF: sure, it's easier to wget from there anyway, I guess [08:08] RAOF, do you want to send a note to the lists or would you prefer me to? [08:09] (or shall we just surprise them?) [08:10] Bah, stupid compiz crash. [08:11] surprise attack is the best kind [08:11] I think it'd be easier if you send the note; I'm pretty much done for the day. [08:11] alrighty [08:11] Hm, don't need the already-uploaded mesa in there… [08:24] Looks like everything's there. [08:24] 1.10-0ubuntu1 builds? [08:25] And installs and runs. [08:25] It's already built in the PPA. [08:25] oh cool, so the build-failure was in the packaging? [08:25] actually, the ppa seems to have that one as pending [08:26] What build-failure? [08:27] Oh, ah. [08:27] the diff between rc3 and final [08:27] No. There was a cherry-pick to take from post final to make sdksyms work. [08:27] Suddenly, context happens :) [08:27] ah [08:28] Because the wonderful udeb-builds-don't-work problem that has plagued me locally became a global problem. [08:28] ok then, I'll pull those and upload, and will sort out any kinks during the day, and hand over to bryceh/Sarvatt when they start their day :) [08:29] Sweet. [08:29] You've got an nvidia driver to go? [08:30] I can pull the one from the ppa [08:32] RAOF: push the server changes and others, if you haven't yet [08:32] to git [08:34] Done for xserver, ati, intel. Checking nouveau. [08:34] cool [08:35] * RAOF should update the rebuild-the-world script to do the git-y thing, too. [08:37] Hm, whoops. Looks like nouveau's git is already out of date. [08:49] Aaand nouveau's in git too. [08:50] niice [08:52] uploading [08:54] 1Mbit uplink ftw [08:54] not [09:04] damnit [09:04] nvidia-current doesn't build-depend on a specific server version [09:07] We should update nvidia-current (and fglrx) to just list the abis it supports in the packaging, and then have an alternate depends on all of them. [09:07] but now it's getting the wrong abi in depends [09:08] Yeah. [09:09] Re-rebuild it is :) [09:09] well, I'll add the build-dep [09:13] there [10:35] nice, nvidia failed to build on amd64 [10:35] cdbs : Depends: python-scour but it is not going to be installed [10:36] ?!? [10:37] tjaalton: Do you have a link to the build log? [10:38] http://launchpadlibrarian.net/66646771/buildlog_ubuntu-natty-amd64.nvidia-graphics-drivers_270.30-0ubuntu2_FAILEDTOBUILD.txt.gz [10:38] 0ubuntu1 built fine [10:38] so probably some chroot problem [10:40] That would be my guess, too. [10:44] Glide built just fine on the same box just now. [11:07] I'll retry it [12:22] built fine === njpatel is now known as njpatel_ === seb128_ is now known as seb128 [14:52] hey, I have a small question, I just bought a second video card so now I have an ATI and a nVidia in the one machine. The hardware tool picked up on the nv one [14:52] but not the ATI [14:53] So I installed the driver manually, will that break anything or was that just a bug with the detection process? [14:53] its in 10.10 I havent tested it in 11.04 since its my work machine [14:54] (and the fact I just came from 11.04 and the closed source ATI driver was broken) [14:56] the closed driver won't coexist peacefully [14:56] drivers, even [14:58] jcristau: hmmm well I could just use the ATI card with the open source driver since it works fine without the closed one but its a shame that they dont work. [14:59] On windows the only things that clash are the control centers [14:59] but I suppose they must have put a lot of effort into making them work on windows obviously [15:01] Ok im upgrading to 11.04 then since ill be using the nv driver as my main [15:01] ill report bugs since thats all the rage :) [15:37] was there a change somewhere that hides cursors from touch devices somehow? [15:37] I can't complain :) [15:37] but I just did a dist-upgrade [15:37] not just touch devices [15:37] it's been awhile [15:37] tjaalton, can you describe when the cursor disappears? [15:38] it only seems to disappear when I use my touchscreen [15:38] hmm not sure [15:38] doesn't seem to happen here [15:38] maybe I was wrong [15:41] tjaalton, it seems that the cursor is hidden any time the contents of the x window under it changes [15:41] if you run "xinput test-xi2" [15:41] then move your cursor over the terminal window [15:41] it will flicker [15:41] as raw motion events are printed [15:42] same thing if you hold a button down and drag on the root window in a gnome desktop [15:44] apw: did you update mesa between tests? [15:45] apw: https://lists.ubuntu.com/archives/natty-changes/2011-March/009496.html [15:46] pretty big change for i386 mesa yesterday [15:57] Sarvatt, nope flipping just the kernel is enough to bust me, so i assume its enabling the optimisation which triggered it [15:57] Sarvatt, and indeed, we have a panic so ... i think mesa is off the hook [16:01] ah, phew [17:32] Sarvatt: oh, btw, mesa 7.11 snapshots from xorg-edgers seem to have fixed my fullscreen flash issues [17:33] and the odd screen flashing I'd get while loading an HTML5 video [17:36] bryceh, I've pushed a fix to xorg-server packaging tree [17:36] just fyi [17:37] it doesn't have to be uploaded asap, as I think only xinput test-xi2 exploits the bug [17:39] I've also pushed a fix for synaptics that isn't needed asap === evilvish is now known as vish === njpatel is now known as njpatel|away [18:53] hey [18:54] I just installed a fresh Kubuntu Natty from alpah3, and upgrade [18:54] after the upgrade (before everything was fine) [18:54] xrandr is not showing my crtc [18:55] http://paste.kde.org/7606/ xrandr output [18:56] I'm going to reboot and try to start the kernel without the HDMI cable plugged [19:00] same result :( [19:00] * afiestas is scared [19:21] I think I got the problem, intel driver wasn't isntaleld :/ [19:21] rebooting [19:29] Mueeheh dunno why but intel driver was not installed, really impressive the fbdev performance [19:30] afiestas, look at your /var/log/dpkg.log [19:32] bryceh: http://paste.ubuntu.com/582223/ [19:51] afiestas: your mirror was lagging behind, and you forced the upgrade [20:09] tjaalton: okz, thanks [20:10] afiestas: so the xserver upgrade probably removed it, along with xserver-xorg-video-all [20:11] yes, because I didn't had vesa either [20:11] the odd thing is that KWin was perfectly working with fbdev :s [20:20] but slow? [20:54] Sarvatt, i just tested the preinstall script in nvidia-common and it doesn't actually suppress the nvidia-installer. it simply adds a yes/no question. the user can hit enter and continue. there is not even a detailed error message like "this installer is not compatible with ubuntu...".