| Sarvatt | knittl: probaby fallout from nvidia-96,173 and fglrx having been forcibly removed from jockey for now because they dont work | 00:03 |
|---|---|---|
| RAOF | Good morning. | 00:07 |
| RAOF | asac: Good gallium for 965, for Maverick? No. | 00:08 |
| RAOF | Even though you're not here. | 00:08 |
| Sarvatt | probably not maverick+1 either | 00:08 |
| RAOF | Not unless someone wants to devote some resources to it. | 00:10 |
| RAOF | Gah. Stupid UTC+10 | 00:10 |
| RAOF | You wake up just as the people looking into the bug you're after go to sleep | 00:10 |
| ajmitch | RAOF: yay for mondays? :) | 00:10 |
| RAOF | Eh. | 00:15 |
| * RAOF is disappointed that Adobe's DNG converter doesn't process the lens distortion data into it's nice, documented format. | 00:16 | |
| knittl | morning. how do i find out which graphics driver is currently in use? | 08:50 |
| RAOF | /var/log/Xorg.0.log will tell you what drivers have loaded, and which one(s) are in use | 08:51 |
| knittl | thanks RAOF | 09:03 |
| knittl | in maverick i get an error/warning message on boot | 09:06 |
| knittl | conflicting fb hw, disabling generic driver | 09:06 |
| knittl | something like that. is that ok? | 09:06 |
| RAOF | That's not a warning, and it's expected. | 09:06 |
| knittl | Aug 16 09:52:18 kbook kernel: [ 16.914621] fb: conflicting fb hw usage nouveaufb vs VESA VGA - removing generic driver | 09:07 |
| knittl | ok :) | 09:07 |
| RAOF | C'mon. Fix the server crash, fix the server crash! | 09:12 |
| knittl | does standby work for you? | 09:13 |
| RAOF | Yes. | 09:13 |
| RAOF | Even on my nouveau laptop. | 09:13 |
| RAOF | Urgh. It won't fix the server crash until it compiles! | 09:14 |
| === yofel_ is now known as yofel | ||
| czajkowski | RAOF: are you about ? | 11:23 |
| czajkowski | RAOF: just wondering if you could possibly look into/comment on https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/605041 | 11:25 |
| ubot4` | Launchpad bug 605041 in xorg-server (Ubuntu) "High usage of the cpu - up to 100% (affects: 1) (heat: 8)" [Undecided,Confirmed] | 11:25 |
| gord | these nvidia 2d performance regressions on anyones radar? was only happening to me a few weeks ago but noticed a few others coming forward too. i know its kinda hard with binary drivers =\ | 11:45 |
| tseliot | gord: ?? | 12:10 |
| tseliot | gord: if you file a bug report about it, I'll tell Nvidia about it | 12:11 |
| gord | tseliot, ever since around the end of last month, nvidia binary drivers have been suffering some huge 2d performance issues. 3d works fine, but just drawing a gtk window or switching tabs or anything is painful. makes xorg cpu usage go crazy. its hard to define though and doesn't seem to affect everyone | 12:11 |
| gord | tseliot, will do | 12:11 |
| tseliot | ok, thanks | 12:12 |
| RAOF | gord: I thought that was fixed with the new cairo upload? | 13:37 |
| RAOF | It certainly fixed it for the nouveau drivers :) | 13:37 |
| * RAOF goes back to sleep | 13:37 | |
| czajkowski | :( | 13:40 |
| Sarvatt | http://www.nvnews.net/vbulletin/forumdisplay.php?f=14 is the support forum for nvidia where the people who can actually fix the blob post in case it helps | 15:30 |
| Sarvatt | just looking through the first few pages it seems like 2D has regressed a ton on the GPU's that can't do trapezoid acceleration on 256.xx | 15:32 |
| Sarvatt | aka i dont see anyone on a gt200 or newer (or ion) complaining about 2D performance. the new cairo disabled server side gradients unconditionally and that was something the blob accelerated, I wouldn't be surprised if that caused even more of a performance hit and that was uploaded aug. 4th | 15:39 |
| tseliot | aah, so it's the issue with cairo. Now it makes sense | 15:44 |
| gord | Sarvatt, my desktop at home has a gtx260 that has 2d performance issues | 15:47 |
| Milos_SD | Hi | 15:48 |
| Milos_SD | Sarvatt, | 15:48 |
| gord | its just that it has a quad core i7, so you don't notice it. its only when looking at top and seeing xorg go crazy you notice. i would assume a lot of other people on gt200's have good cpu's like that taking the brunt of the performance problems | 15:48 |
| Sarvatt | gord: does nvidia-settings -a PixmapCacheRoundSizeKB=16000 help any? | 15:49 |
| seb128 | ideally the cairo changes would be on some drivers only | 15:49 |
| seb128 | if somebody want to work on that ;-) | 15:49 |
| Milos_SD | I think there is a but with new libpixman package that was sent today | 15:50 |
| seb128 | if filter on known buggy drivers | 15:50 |
| Milos_SD | notify-osd bubble is not showed as it should | 15:50 |
| gord | Sarvatt, not on this machine (9600gtm), can't say about my desktop until i am home | 15:50 |
| Milos_SD | here is the screenshot: http://www.glowfoto.com/viewimage.php?img=16-060157L&rand=5836&t=jpg&m=08&y=2010&srv=img5 | 15:50 |
| Sarvatt | its really a problem with the blob hitting software fallbacks way too often not really cairo | 15:50 |
| Milos_SD | is there a way to downgrade that package to previuse one? | 15:52 |
| Sarvatt | Milos_SD: do you have it in /var/cache/apt/archives/ still? | 15:53 |
| Milos_SD | no I don't :( | 15:53 |
| Sarvatt | Milos_SD: amd64 or i386? | 15:55 |
| Milos_SD | amd64 | 15:55 |
| Sarvatt | https://edge.launchpad.net/ubuntu/+source/pixman/0.18.2-1/+build/1741447 | 15:56 |
| Sarvatt | https://edge.launchpad.net/ubuntu/+source/pixman/0.18.2-1/+build/1741447/+files/libpixman-1-0_0.18.2-1_amd64.deb | 15:56 |
| Sarvatt | Milos_SD: let me know if it works correctly again when you downgrade please | 15:57 |
| Milos_SD | ok, but it was this version: libpixman-1-0 (0.19.1+git20100722.bf125fbb-0ubuntu0sarvatt~lucid) to 0.19.1+git20100815.8a5d1be1-0ubuntu0sarvatt~lucid | 15:58 |
| Milos_SD | :D | 15:58 |
| Sarvatt | oh you mean in xorg-edgers! | 15:58 |
| Sarvatt | i just updated it in maverick too and thought you meant that :D | 15:59 |
| Milos_SD | in xorg-edgers for lucid | 15:59 |
| Milos_SD | :) | 15:59 |
| Sarvatt | Milos_SD: http://ppa.launchpad.net/xorg-edgers/ppa/ubuntu/pool/main/p/pixman/ | 16:00 |
| Sarvatt | grab it quick because it only stores the latest 2 versions and i'm about to update it again so the 0722 will go away :) | 16:00 |
| Milos_SD | Sarvatt, thanks | 16:02 |
| Milos_SD | I'll try if it works | 16:02 |
| Milos_SD | maybe the new version will work after you update it :) | 16:02 |
| Milos_SD | Sarvatt, it works now | 16:03 |
| Milos_SD | thanks :D | 16:03 |
| Sarvatt | possibly, they stopped supporting alpha maps with alpha maps completely and then added some handling to it if it happens just after i updated it yesterday | 16:04 |
| Sarvatt | its uploading now | 16:05 |
| Milos_SD | yes, it happened after that update | 16:05 |
| Milos_SD | :) | 16:05 |
| Sarvatt | Milos_SD: what GPU do you have and what driver are you using btw? I can't reproduce it on intel | 16:07 |
| Milos_SD | Sarvatt, Nvidia 7600 GT PCIe and 256.44 drivers | 16:08 |
| Milos_SD | in that screenshot compiz is on | 16:08 |
| Milos_SD | but if I use only metacity, that part of notify-osd is darker then other part of it | 16:08 |
| Sarvatt | yay so more 2D brokenness with the blob? :) | 16:09 |
| bjsnider | the 7k cards have significantly older hardware than 8k and newer | 16:09 |
| Milos_SD | Sarvatt, but it worked ok before that update today/yesterday :) | 16:10 |
| bjsnider | the blob is basically optimized for 8k and newer | 16:10 |
| Sarvatt | yeah but it's not broken on nouveau or intel for me so it might have broken something they didn't expect that the blob uses I'm thinking | 16:10 |
| === seb128_ is now known as seb128 | ||
| czajkowski | RAOF: are you up ? | 20:34 |
| jono | hi all | 22:07 |
| jono | my X is seg-faulting - http://pastebin.com/gUmYb3Xv - I just wanted to check if this was a known issue | 22:07 |
| jono | (in Maverick) | 22:07 |
| jcristau | that's out of date maverick | 22:08 |
| jcristau | is it reproducible with the latest version? | 22:08 |
| jcristau | could be fixed by the patch in <yunk4nrtdoi.fsf@aiko.keithp.com> if so | 22:10 |
| czajkowski | Just wondering if anyone could comment on bug https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/605041 someone was talking about it yesterday on identi.ca and looking for help | 22:11 |
| ubot4` | Launchpad bug 605041 in xorg-server (Ubuntu) "High usage of the cpu - up to 100% (affects: 1) (heat: 8)" [Undecided,Confirmed] | 22:11 |
| jono | jcristau, I just did dist-upgrade | 22:11 |
| jcristau | your log says xorg-server 2:1.8.1.902-0ubuntu2 | 22:12 |
| jcristau | but, xorg-server | 2:1.8.99.905-1ubuntu2 | maverick | source | 22:12 |
| jono | jcristau, hmmm | 22:12 |
| jono | how do I upgrade it? | 22:12 |
| jono | jcristau, when was that published? | 22:15 |
| jcristau | i have no idea | 22:16 |
| jono | it is weird that I have an old package | 22:17 |
| jono | jcristau, ok I managed to install it | 22:20 |
| jono | it was in my apt cache but not installed | 22:21 |
| jono | so I installed it and it still segfaults | 22:21 |
| jcristau | time to try that patch then | 22:21 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!