[03:28] ayan: Hi again :) [03:29] Also, the contents of /var/log/Xorg.0.log and the output of dmesg are likely to be useful in debugging this. [03:29] ROAF: hi! [03:29] (In a pastebin, if that's not obvious ☺) [03:49] ayan: Still here? [03:50] Hm, interesting. Compiz seems to have kernel-panicked nouveau, but ssh soldiers on. [05:58] Balls. There's an X crasher in -radeon's DRI2 swapbuffers handling again. [09:42] what, have I pushed something to our xorg-server branch.. [09:43] oh, the merge. just the diff post on debian-x@ looks funny [09:45] tjaalton: Would you like to test some mesa 7.10.2? :) [09:45] RAOF: sure [09:45] I'll push it to git. [09:45] ok, I'll build from there [09:46] Oh, yeah. [09:46] I'll throw my binaries into http://cooperteam.net/Packages for you! [09:47] cool, clean it up a bit and I can wget the whole dir :) [09:47] i have intel to test it on [09:47] 965 === lilstevie|lagche is now known as lilstevie [09:50] tjaalton: Now available for your pleasure. i386 & amd64! [09:54] RAOF: thanks, downloading.. [10:14] I suppose Radeon HD 6950's are supposed to work in natty? [10:15] Are, or aren't? [10:15] I *think* they are, actually. [10:21] are [10:21] somebody complained his new HD 6950 didn't work in 10.10, I told him to try the natty beta live cd [10:23] what happened? [10:23] It should be supported by fglrx if nothing else. [10:24] he's downloading... [10:24] heh, ok [10:25] I'll let you know ;) [10:26] download is going to take 3 hours... :-( [10:27] at 64k/s? [10:27] slow.. [10:30] RAOF: mesa seems to not burst into flames here [10:31] It also hasn't burst into flames on my radeon card, where the bulk of the changes were. [10:31] However, I don't have an r300 to test, and that would be good. [10:31] i do, actually [10:31] at least I think it is one [10:32] Anything r300-r500 would do; that's 9600 (I think) - HD 2xxx (again, I think) :) [10:41] yesh, R300 NG [FireGL X1] [10:47] seems to work [10:47] RAOF: ^ [10:51] so, we have libdrm, xorg-server, mesa, libx11 bugfix releases pending.. next thursday is beta2 [10:51] when do we start uploading these?-) [10:56] Now, I think. [10:58] Although we might not want libdrm 2.4.24 [10:58] ok [10:58] because of then nouveau changes? [11:00] great, the 10.10 installer doesn't recognize my ssd drive [11:01] so I can't test the joystick patch on it [11:01] don't have a spare hd [11:08] tseliot: looks like the dvi-dsub -dongle works after all, just had to make sure it's connected to the right dvi port [11:09] and the "no ssd" problem is a bios failure [11:09] sometimes it doesn't seem to probe it [11:15] tjaalton: what is it that makes a dvi port the right port? [11:16] does it have to be the 1st dvi connector? [11:19] probably [11:19] dvi cable works on both === yofel_ is now known as yofel [13:43] cnd: hey, there's bug 745065 which says that wacom multitouch is broken in natty when it worked in lucid & maverick [13:43] Launchpad bug 745065 in xf86-input-wacom (Ubuntu) "Touchscreen not detected as multi-input (affects: 1) (heat: 6)" [Low,Confirmed] https://launchpad.net/bugs/745065 [13:43] cnd: ideas about that on? [13:43] *one [13:45] tjaalton, I'll take a look [13:46] cnd: thanks [16:17] RAOF: forgot to push mesa? [16:17] probably too late to ask :) === soreau_ is now known as soreau [16:43] so when I resize a window, it should actually resize, right? [16:43] (natty beta) [16:44] though to be fair, that's from a day or 3 ago - dist-upgrading now [16:44] works here [16:44] tjaalton: uxterm and resize it [16:44] pulling down the bottom of the window [16:44] and then resize it again [16:45] lamont: yep [16:45] looks fine to me [16:45] yep thats busted here too, the xterm doesn't resize [16:45] I'm using "ubuntu classic" atm [16:45] me too :) [16:45] ah wait [16:45] it works on an updated system [16:46] Sarvatt, bryceh_: the archive will beta-freeze on monday :/ [16:46] let me finish the dist-upgrade and confirm that I'm still broken (hoping I can just go "all better, thanks") [16:46] 3 day old natty + nvidia blob doesnt resize, fully up to date intel resizes properly [16:46] and that says 8 minutes [16:46] intel chipset here [16:46] nvidia here, upgraded yesterday [16:47] 1 day 25 min uptime here, but that's 12-18 hours post-upgrade [16:47] 8 min download, nfc how long after that for processing [16:47] xterm hasn't changed [16:47] the server has [16:47] compiz has [16:48] yeah - was just going to say "or has the server..." compiz has [16:48] gnmoe has [16:48] gnome even [16:48] lamont: [16:48] compiz (1:0.9.4+bzr20110406-0ubuntu1) natty; urgency=low [16:48] * new upstream bzr tarball: [16:48] - display/size problems with xterm (LP: #748137) [16:49] sure is nice when bugs are already fixed :P [16:49] hehe [16:58] heh [16:58] ta [17:00] ok got 7.10.2 on all my machines, now to break things! [17:01] it's indestructible [17:02] i'll upload libx11 1.4.2-1u1 [17:03] 1.4.3 would need newer xutils-dev [17:03] bryceh_: btw, xutils-dev doesn't show on versions-current.html? [17:06] [ 5.435] [17:06] X.Org X Server 1.10.0 [17:06] so friggin fast! [17:06] heh [17:07] ssd luv [17:09] it does, indeed, appear to be fixed. [17:16] libx11 uploaded [17:19] guinness-time \o\ |o| /o/ :) [17:20] * Sarvatt is jealous [17:20] speaking of which, need to put more in the fridge :) [17:25] darn, new libva in sid but its too late for natty [17:26] into x-updates it goes [17:32] hrm 270.40 nvidia driver but its not a normal release, just on the CUDA page [17:37] tjaalton: looks like the buildds are screwed up? libx11 failed [17:38] oh nevermind klingon failed [17:42] Sarvatt: great, i get to drop it [17:42] oh [17:43] how the hell did it build here then [17:43] series file was all fail [17:47] tjaalton: 100_latin_locale_alias.diff101_klingon_locale_alias.diff yea :) [17:48] 102_double_arrows_Compose.diff is failing after fixing that [17:48] yeah i dropped the double [17:48] err, the other entry i mean [17:49] klingon is fine [17:49] tjaalton, ah, +packagebugs changed from "in ubuntu" to "in Ubuntu", which broke my regex. So it has been using an old list of our X packages. Page is being refreshed now. [17:49] bryceh_: great, thanks [17:54] tjaalton: http://sarvatt.com/downloads/patches/102_double_arrows_Compose.diff [17:56] oh bah wait [17:56] 102_double_arrows_Compose.diff was in the series twice.. [17:56] lol [17:59] http://www.bryceharrington.org/Arsenal/Reports/ubuntu-x-swat/versions-current.html updated [18:15] https://bugs.launchpad.net/ubuntu/natty/+source/nvidia-graphics-drivers/+bug/752634 -- isn't that the intended behavior from the nvidia blob? I've never used multi monitor on it but fairly sure you have to set up additional monitors through the nvidia-settings gui and not xrandr [18:15] Launchpad bug 752634 in nvidia-graphics-drivers (Ubuntu Natty) (and 1 other project) "[Lenovo T410] Displayport cannot output to a monitor (affects: 1) (heat: 6)" [Medium,New] [18:30] Sarvatt, yes that's correct [19:11] phew, only 2 bugs from the cert runs, I was worried :) [19:14] ram usage with the blob in natty is outrageous [19:14] compiz ends up at >500 mb after a couple of days [19:14] bjsnider: update [19:14] new cairo should be published now [19:14] well, i ran all updates as of this morning [19:14] it was uploaded like an hour ago I think [19:15] what is it, libcairo? [19:16] it's not in there so far [19:19] https://lists.ubuntu.com/archives/natty-changes/2011-April/011061.html [19:19] hmm should be published, might not be on your mirror yet [19:21] Sarvatt: what are the issues? [19:24] cairo linking to a 24MB nvidia libGL making the world use outrageous amounts of memory :) [19:24] https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers/+bug/725434 [19:24] Launchpad bug 725434 in nvidia-graphics-drivers (Ubuntu Natty) (and 3 other projects) "Nvidia drivers lead to extra memory usage for each process using libGL (affects: 32) (dups: 1) (heat: 188)" [Medium,Won't fix] [19:26] Sarvatt: eh? clearly the Linux VM should handle that well or is allocating a lot of data as well? [21:55] RAOF, bryceh_ : what are your thoughts on bug 738687? should that patch be added to our mesa? [21:55] Launchpad bug 738687 in mesa (Ubuntu) "R600 driver crashes when adding effects in kwin (affects: 2) (heat: 18)" [Medium,Triaged] https://launchpad.net/bugs/738687 [22:04] kees: thats actually queued up to upload in mesa 7.10.2 whenever RAOF comes on next [22:04] kees, yeah if it's in the stable tree should be a no-brainer to include [22:05] kees, how'd you run into this issue? [22:05] cool; thanks guys. how should I mark that bug so it disappears from the sponsorship queue? [22:05] bryceh_: I'm patch pilot, and it was marked for sponsorship [22:05] oh patch piloting, of course [22:06] kees, hmm [22:07] kees, good question... [22:08] kees, raof probably won't be on until monday and since the freeze comes monday there could be a chance it gets missed [22:08] kees, so maybe go ahead and sponsor the patch in? [22:08] sarvatt, what do you think? [22:09] I'd rather not disrupt RAOF's schedule... [22:10] bryceh_, the freeze will not be before the european day so there is time for an upload for .au guys [22:10] but RAOF is no core-dev [22:11] okay, I'll just mention in the bug it's scheduled... [22:11] if he could just push the git branch we could upload during the weekend [22:11] kees, yeah do that [22:11] ok well fwiw, I'm on vacation monday7 [22:11] it's likely we will clean pending uploads on monday [22:11] I've tested 7.10.2 on r300 and intel [22:12] and RAOF took care of r600 i think [22:12] i've been running it on 945 and sandybridge [22:12] kees, sounds like assign to RAOF I guess [22:13] bryceh_: yeah, was just doing that :) [22:13] 965 intel here [22:14] if the archive is frozen at 9:00 UTC on monday, I'll have plenty of time uploading it :) [22:33] I've got a Sandybridge laptop that is having what seems to be KDE specific screen corruption issues. Is there a way I can turn specific driver features on/off at runtime to see if I can narrow down where the issue is? [22:33] It's not tied to compositing as it happens with that totally disabled. [22:36] ScottK, modinfo on the drm driver gives the kernel tunables; man intel gives some X driver options; driconf has the mesa tunables [22:38] bryceh_: The bug is https://bugs.launchpad.net/ubuntu/+source/xorg/+bug/750964 - Any suggestions on where best to start? [22:38] Launchpad bug 750964 in xorg (Ubuntu) "Screen corruption in menus and titlebars (affects: 2) (heat: 12)" [Undecided,New] [22:39] ScottK, maybe rerun apport-collect xorg and see if it'll pick up the missing log files [22:39] I'll try it. [22:40] ScottK, if you suspect it's particular to kwin, I'd suggest starting with driconf since that controls 3D and is pretty easy to use [22:41] also, an important factoid is when the issue first started happening [22:41] It happened since I got the laptop on Monday. [22:41] since if it's a recent regression, then downgrading packages one by one can be a straightforward way to pinpoint things [22:42] So no idea how long it's been a problem. [22:42] ah, so not a regression then [22:42] Nope. [22:42] Just really, really annoying. [22:44] I assume I need to restart X after making changes in driconf? [22:44] you can also flip on kernel drm debug (drm.debug=0x03) although it'll spew a lot of output [22:44] yep [22:45] Let me play with driconf first. It's got a lot of knobs to turn. [22:45] yep [22:47] i really don't understand bug 754565 [22:47] Launchpad bug 754565 in xorg (Ubuntu) (and 1 other project) "Regression: shift+click on a launcher icon to open a new application instance gone (affects: 2) (heat: 10)" [High,Triaged] https://launchpad.net/bugs/754565 [22:49] tjaalton, hum yeah me neither [22:49] tjaalton, isn't really much of a bug report [22:49] bryceh_: nope [22:49] tjaalton, close it and tell them we're not mind readers? ;-) [22:50] hehe [22:50] I'll think of something [22:55] there [23:10] No luck with driconf [23:50] ScottK: the config in freedesktop bug 35808 comment 11 may be worth to test [23:50] Freedesktop bug 35808 in Driver/intel "[SNB] KDE graphic glitches" [Normal,New] http://bugzilla.freedesktop.org/show_bug.cgi?id=35808 [23:50] albert23: Thanks. [23:52] ScottK: also if it doesn't and you want to see if it does with updated drivers - https://launchpad.net/~sarvatt/+archive/gold [23:52] Thanks. [23:53] Mine is a 6320 (vice 6420) so it's very close. [23:53] I don't have the kms issue he has. [23:53] i've got 2 6420's here, it must really be KDE specific [23:56] Ah. That did it. [23:57] Option "DebugFlushBatches" "True" and it's all better. [23:59] Sarvatt: I do believe it's KDE specific. Even non-KDE apps in a KDE session are OK.