RAOF | Prf_Jakob: It's not the size of the library that's the problem, it's that the nvidia libgl *dirties* ~5MiB of memory (on x86-64, apparently *more* on i386) on load. | 01:01 |
---|---|---|
RAOF | albert23: Oh, really? Yeah, that'd be right. Gak! | 01:05 |
* RAOF wonders if multiarching mesa is acceptable at this point. Probably not! | 01:06 | |
RAOF | albert23: Got a bug filed about that? | 01:06 |
albert23 | RAOF: sure, bug 755720 | 01:07 |
ubot4 | Launchpad bug 755720 in ia32-libs (Ubuntu) "32 bit dri drivers cannot find libdricore and libglsl (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/755720 | 01:07 |
RAOF | Ta. I'll fix that with (hopefully!) today's mesa upload. | 01:07 |
albert23 | thanks | 01:10 |
bjsnider | RAOF, why wasn't that ever a problem before? | 01:12 |
RAOF | bjsnider: Possibly becuause no-one's tried it before? | 01:13 |
bjsnider | tried what? | 01:13 |
RAOF | Using the 32bit mesa on a 64bit system? | 01:14 |
RAOF | You'd mostly notice if you were using wine on x86-64, and the mesa libGL doesn't support wine very well. | 01:15 |
bjsnider | no, i meant the tmb of memory dirtied by the blob | 01:15 |
bjsnider | 5mb | 01:15 |
albert23 | RAOF: btw, I tested your mesa 7.10.2 packages on ironlake and nouveau. Both doing fine. Even unity works on nouveau :-) | 01:18 |
RAOF | albert23: Thanks :) | 01:18 |
RAOF | bjsnider: Apparently that's a well-known, longstanding property of nvidia's libGL. I just didn't know it :) | 01:18 |
bjsnider | RAOF, well, i don't know why you didn't, since it's open and well-documented, with an extremely robust public bug tracking system | 01:22 |
bjsnider | oh, and none of the code is shared between platforms. | 01:22 |
RAOF | Wait. It's 10:30. WHY ISN'T MY COFFEE MACHINE ON?! | 01:35 |
RAOF | Fun fact: chrpath on x86-64 doesn't understand IA32 objects. | 01:42 |
RAOF | Oh, this'll require an ia32-libs update, too, won't it. Hm. | 02:39 |
tjaalton | RAOF: so, we need to upload what we want for natty within the next three hours :) | 06:07 |
tjaalton | or so | 06:07 |
tjaalton | which includes your mesa and xserver 1.10.1rc2? | 06:08 |
tjaalton | breakfast-> | 06:11 |
RAOF | Someone had highlighted me before my IRC bouncer caught flames. Was it tjaalton by any chance, offering to upload mesa? | 06:45 |
tjaalton | 08:07 < tjaalton> RAOF: so, we need to upload what we want for natty within the next three hours :) | 06:50 |
tjaalton | 08:07 < tjaalton> or so | 06:50 |
tjaalton | 08:08 < tjaalton> which includes your mesa and xserver 1.10.1rc2? | 06:50 |
RAOF | Yes indeed. | 06:51 |
tjaalton | we would've uploaded mesa on friday, but you didn't push it to git :) | 06:51 |
RAOF | Yah, I saw. | 06:51 |
RAOF | It's done now, though, and available on http://cooperteam.net/Packages | 06:52 |
RAOF | With a bonus ia32-libs fix. | 06:52 |
tjaalton | i'll grab it | 06:52 |
RAOF | I've also got a nouveau upload staged, let me grab it. | 06:57 |
tjaalton | mesa uploaded | 07:06 |
* RAOF is somewhat disappointed how much more performant nouveau on his *netbook* is compared to intel on a reasonably beefy laptop. | 07:10 | |
tjaalton | hehe | 07:14 |
tjaalton | bryceh_: about? was gonna ask about xserver patch 208 ("switch on release") | 07:18 |
RAOF | I'm always a fan of breaking the X protocol :) | 07:19 |
tjaalton | heh | 07:19 |
tjaalton | well, this might create issues for a11y | 07:19 |
RAOF | Nouveau's also up on cooperteam | 07:21 |
tjaalton | uploaded | 07:22 |
RAOF | Now let's see if I can't make DBO happy and fix X/libX11/libxcb | 07:23 |
tjaalton | what bug is it? | 07:29 |
RAOF | Bug #740126 - apparently it's fairly widespread and not WaitMSC specific. | 07:32 |
ubot4 | Launchpad bug 740126 in compiz (Ubuntu) "compiz hangs randomly several times per day (affects: 3) (heat: 20)" [High,Incomplete] https://launchpad.net/bugs/740126 | 07:32 |
tjaalton | hmh | 07:33 |
tjaalton | ooh, the libx11 commit you mentioned on #u-d would fix my vdr box issues | 07:35 |
RAOF | tjaalton: But that's in the libx11 1.4.2 upload that you've already done! | 07:37 |
tjaalton | RAOF: oh, good :P | 07:37 |
tjaalton | then the next yavdr release should get it | 07:37 |
RAOF | :) | 07:37 |
tjaalton | i'll upload xserver as-is | 07:51 |
seb128 | bug #748487 seems to get some duplicates | 10:53 |
ubot4 | Launchpad bug 748487 in xserver-xorg-video-nouveau (Ubuntu) "compiz assert failure: compiz: nouveau_context.c:260: nouveau_update_renderbuffers: Assertion `!ret' failed. (affects: 3) (dups: 6) (heat: 475)" [Medium,New] https://launchpad.net/bugs/748487 | 10:53 |
seb128 | not sure if it should be tracked in some way | 10:53 |
tjaalton | hmm, there was a mesa update pushed earlier today, I'll check if it has something nouveau'y in it | 10:54 |
seb128 | tjaalton, thanks ;-) | 10:56 |
tjaalton | seb128: so far no hits. maybe it needs to be upstreamed then | 11:16 |
tjaalton | which I can take care of | 11:16 |
seb128 | that would be nice | 11:17 |
seb128 | well I don't have the issue since I'm on intel but since it got a few duplicates it would be nice to get upstreamed at least | 11:17 |
tjaalton | yep | 11:20 |
=== yofel_ is now known as yfoel | ||
=== yfoel is now known as yofel | ||
bdmurray | RAOF: is the ia32-libs task in bug 755720 still valid? | 18:57 |
ubot4 | Launchpad bug 755720 in mesa (Ubuntu) (and 1 other project) "32 bit dri drivers cannot find libdricore and libglsl (affects: 1) (heat: 6)" [High,Fix released] https://launchpad.net/bugs/755720 | 18:57 |
Sarvatt__ | bdmurray: yeah ia32-libs needs to be updated | 19:05 |
Sarvatt__ | ScottK: did you have a launchpad bug for the corruption? I'm not seeing it | 19:05 |
ScottK | I did. Let me find it. | 19:05 |
Sarvatt__ | got a test package for you but need to put the bug # in the changelog | 19:05 |
ScottK | Sarvatt__: Bug #750964 | 19:06 |
ubot4 | Launchpad bug 750964 in xorg (Ubuntu) (and 1 other project) "Screen corruption in menus and titlebars (affects: 2) (heat: 12)" [High,Confirmed] https://launchpad.net/bugs/750964 | 19:06 |
Sarvatt__ | thanks a bunch, updated it with a PPA that has an xserver-xorg-video-intel with what should be the fix in it, would be much appreciated if you could test it. its a very small change at least | 19:10 |
ScottK | Will do. | 19:41 |
Sarvatt__ | nvidia 270.41.03 uploaded to x-updates if anyone was looking for it | 21:29 |
Sarvatt__ | screwed up the maverick one, fixing that up now | 21:43 |
bjsnider | what does the changelog say? | 21:53 |
Sarvatt__ | Fixed a bug causing the X server to hang every 49.7 days on 32-bit platforms. and a ton of new gpu support | 21:54 |
bjsnider | that's it? | 21:54 |
Sarvatt__ | yep | 21:55 |
bjsnider | sounds like they're not even really working on it | 21:55 |
Sarvatt__ | all the new gpu support took them a whole month to add, we have a ton of systems that were waiting on this updated driver to ship | 21:59 |
seb128 | bug #757906 seems a frequent unity crash for intel users | 22:53 |
ubot4 | Launchpad bug 757906 in unity (Ubuntu) "compiz assert failure: compiz: intel_tex_image.c:726: intelSetTexBuffer2: La declaraciĆ³n `!texImage->Data' no se cumple. (affects: 5) (dups: 4) (heat: 42)" [Medium,New] https://launchpad.net/bugs/757906 | 22:53 |
seb128 | if someone wants to have a look to it | 22:53 |
seb128 | https://bugzilla.redhat.com/show_bug.cgi?id=681161 is similar | 22:56 |
ubot4 | bugzilla.redhat.com bug 681161 in xorg-x11-drv-intel "[abrt] mutter-2.91.90-2.fc15: intelSetTexBuffer2: Process /usr/bin/mutter was killed by signal 6 (SIGABRT)" [Unspecified,New] | 22:56 |
seb128 | https://bugs.freedesktop.org/show_bug.cgi?id=35234 | 22:58 |
ubot4 | Freedesktop bug 35234 in Drivers/DRI/i965 "mutter: intel_tex_image.c:726: intelSetTexBuffer2: Assertion `!texImage->Data' failed" [Major,New] | 22:58 |
Sarvatt__ | ScottK: looks like it works, I've got packages prepared but need to figure out when its ok to upload with the freeze | 23:39 |
Sarvatt__ | http://sarvatt.com/downloads/merges/intel-natty/ | 23:45 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!