=== releaselogger is now known as apachelogger [04:51] Wow, have to give the ati gallium devs credit, they're quickly approaching performance parity with the catalyst drivers. Now all we need is some form Video Acceleration for r300-r500 and I'll be a very happy camper. [04:54] ripps: I'd try a little reverse-engineering if I were you; the gallium VA architecture is starting to materialise, so there'd be something to hook in to :) [05:01] ripps, yes but the tests were conducted with extremely old hardware. [05:01] bjsnider: like mine. I have an rv350 radeon 9600 pro [05:02] This is very good news for me [05:02] it would be helpful if the code also worked for the newer stuff too [05:03] at least r600g has xvmc, I'm still wallowing in xv only. I'd really like to offload just a small bit of video decoding to the gpu [05:04] there's no video decoding hardware on an old thing like that [05:05] shouldn't matter, from what I can tell, the upcoming gallium va architecture will shaders, not custom the uvd stuff in the catalyst anyway. [05:06] it would be most important to have help decoding h.264, and i seriously doubt an old gpu would help much [05:06] an old system like you undoubtedly have would not be much good for h.264 video [05:06] any help, no matter how small would be help greatly [05:07] you know what would be a better solution though... [05:19] new equipment === jbs is now known as Bernardo|Away === Bernardo|Away is now known as jbs [05:21] Actually, I've been looking at an upgrade soon. And I was planning to get an nvidia geforce gt 240. Mostly because I wanted access to a decent gpu that allowed for hd playback, and vdpau looked appealing. [05:22] However, seeing how much progress ati gallium is making, I'm kinda of two minds about it. [05:23] that has vdpau feature set c, which means it decodes just about everything [05:23] but hte gt 210 can come without a fan, has the same vdpau features, and makes no noise [05:24] actually, which is better? gt 240 or fermi 430? Both are bout the same in price, but the gt240 has gddr5, so shouldn't it be faster? [05:24] yeah, although the fermi has a more advanced gpu, with opengl 4 [05:25] they both have feature set c [05:25] a more powerful card will be louder, which is a personal pet peeve, and i don't see how you take advantage of the speed on linux [05:26] it seems fermi is more target toward the htpc crowd, while gt is more focused toward gamers. Now I hardly ever play PC games, but it be nice to know I could If I ever wanted to. [05:26] you can't play them on linux though [05:26] unless you dual-boot it's irrelevant === jbs is now known as Bernardo|Away === Bernardo|Away is now known as jbs [05:30] I'm not opposed to dual-booting [05:31] But, the last time I had a windows partition, I hadn't boot into it in over a year. Then I erased the drive it was on a few months ago, and I haven't found the need to reinstall windows on it yet. [05:32] has that computer got a pci-express slot? it sounds like you've got an agp card now [05:32] bjsnider: my current computer doesn't, but I got an old mobo and cpu from dad recently, I just need a decent pci-e video card and a cpu fan. [05:34] my current computer is very old, so I'm looking forward to the new one. Unfortunately, the new cpu isn't that much faster than my current one, it's just 64-bit [05:43] bjsnider: so, you think I'd like the fermi 430 better because it would produce less heat, and therefore less noise? === jbs is now known as Bernardo|Away === Bernardo|Away is now known as jbs === jbs is now known as Bernardo|Away === Bernardo|Away is now known as jbs [06:29] lastlog jbs [06:31] hmm, is jbs a bot or ..? looks like he isnt active he except for the nickchange spam [06:32] active here* === jbs is now known as Bernardo|Away === Bernardo|Away is now known as jbs === jbs is now known as Bernardo|Away === Bernardo|Away is now known as jbs === jbs is now known as Bernardo|Away === Bernardo|Away is now known as jbs [13:24] hello ubuntu-x [13:24] how are we today? [13:28] ok, not a lot happening in here I can see. i'll just let my question linger then: i can see from old IRC logs that the llvmpipe driver was once included in the xorg-edgers PPA, is there any chance it will be that again in the future? i would really like to try it out, but I must admit that compiling mesa (and all the other things necessary) are too much for me right now, so it'd be perfect if xorg-edgers packaged this. any chance [13:28] of this happening in the future? running the Gallium drivers for my RV670 now BTW, runs *perfectly* (and quicker than mesa classic!). thanks for this awesome PPA! [14:34] \o [14:34] http://imagebin.org/133512 is some weird stuff happening with a friends install [14:35] he has ubuntu 10.10 ... [14:36] Also bug 702341 << some weird stuff with his synaptics touchpad ( Cross posted from #ubuntu-kernel since i dont know which channel this should go in ) [14:36] Launchpad bug 702341 in xserver-xorg-input-synaptics (Ubuntu) "synaptics touchpad does not work (affects: 1) (heat: 416)" [Undecided,New] https://launchpad.net/bugs/702341 [14:36] shadeslayer: did you try kicking the friend? (not the install) [14:36] * evilvish hides.. [14:37] heh [14:37] evilvish: didnt work :P [14:37] he's new to this ;) [14:39] * evilvish smiles at the scan! and wonders what bollywood has to do with capacitors ;p [14:40] lol : [14:40] evilvish: http://imagebin.org/133511 [14:40] not limited to a particular area [14:40] and is completely randomn [14:42] Mobility Radeon HD 5000 Series << Graphics Card [14:46] shadeslayer: hey, so your exams are over now? [14:46] evilvish: yep :) [14:46] why are you evil? :P [14:48] shadeslayer: you use evilslayer too, right? ;) [14:48] not right now :P [14:57] ripps, it's tough to say. i use a graphics card that has feature set c, no fan, and the speed is fine for me, but i don't do any gaming or anything. it runs the composited desktop admirably. for me, graphics card noise is a huge annoyance. for you, it might not be === yofel_ is now known as yofel [15:14] hi all, can anyone confirm that xorg edgers (and the new xorg) work with nvidia? would like to upgrade to do some testing with unity [15:16] not with the proprietary drivers [15:16] great, will hold off for a while then [15:33] is there any chance that the llvmpipe driver will be included in xorg-edgers again any time soon? i can see from IRC logs that it once was. [15:43] just testing IRC via Empathy [15:47] seems to work :) [16:15] i'm off for now. if anyone knows anything about when the llvmpipe driver might be available in xorg edgers PPA, please write an answer here. i'll just look in the IRC logs for the answer, when time permits. good day everyone! [17:23] bryceh: ugh what a mess in that bug you linked. looks like ati broke userspace and that commit was picked up in 2.6.35 stable? [17:27] http://cgit.freedesktop.org/mesa/mesa/commit/?h=7.9&id=773e8fadc8b19aaba5d13f75ac5810badb3968c4 (post 7.9.0...) [17:51] if a thin like that is in the stable kernel, that ought to make linus very happy [17:54] https://bugzilla.kernel.org/show_bug.cgi?id=24802 [17:54] thats.. nasty [17:54] bugzilla.kernel.org bug 24802 in Video(DRI - non Intel) "Graphics errors with Radeon KMS driver on RV770" [Normal,New] [17:55] mesa 7.9.1+ needs that kernel commit to work right, mesa 7.9.0 wont work right with that kernel commit [17:55] Sarvatt, yeah ugh [17:57] what the heck do we do? [17:58] pull the mesa patch, with versioned depends on the kernel? :P [17:59] or drop the kernel commit [17:59] yeah its dropped, but newer kernels dont work right on maverick without updating mesa somehow now [18:00] at least its easy to use metacity in maverick (unlike natty) to work around it :) [18:30] http://www.bryceharrington.org/X/Reports/ubuntu-x-swat/totals-natty-workqueue.svg [18:31] 4 new bugs in xorg-server today [18:31] 3 of which are apport crash hook bugs [18:32] which is interesting because I thought hook was busted... did it suddenly start working again? [18:32] why pixman all the sudden? hmm [18:33] wow you're right, haha [18:33] how the heck did that happen [18:34] hmm, I overheard someone else having trouble getting apport to capture crashes, I wonder if something got fixed? [18:38] doesn't look like its working quite right though, hmm [18:38] oh? [18:38] StacktraceTop: [18:38] XISendDeviceHierarchyEvent (flags=0xbfcccf9c) [18:38] DisableDevice (dev=0xaaa8e70, sendevent=1 '\001') [18:38] ?? () [18:38] they're all that [18:38] hrm [18:38] Sarvatt, speaking of apport, dunno if you noticed this change went in: [18:39] retracer isn't working right [18:39] revno: 1822 [18:39] committer: Martin Pitt [18:39] branch nick: trunk [18:39] timestamp: Tue 2011-01-11 16:26:12 -0600 [18:39] message: [18:39] hookutils.py, attach_dmesg(): Do not overwrite already existing dmesg. [18:39] I chatted with pitti about our gpu freeze capture troubles last week [18:40] problem was when the gpu freeze handler triggers, it *does* capture the dmesg and other stuff [18:40] but after rebooting when apport fires up to report the bug, it RE-collected those files, overwriting the ones that had been captured earlier [18:40] now it no longer does that [18:41] http://launchpadlibrarian.net/62489709/ThreadStacktrace.txt vs https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/705078/+attachment/1799814/+files/ThreadStacktrace.txt [18:41] Sarvatt: Error: Bug #705078 is private. [18:41] ahh cool [18:41] the retraces aren't working for sure, darn [18:42] hrm [18:43] Sarvatt, yeah you're right, the retracer is attaching something completely unrelated [18:43] Sarvatt, I'll email pitti about this [18:43] actually I'll file an apport bug [18:53] Sarvatt, hmm actually looking closer, all three bugs have the same stack up through #9 [18:54] thats where the real stack starts [18:54] rest just looks like cleaning up after the crash [18:54] Sarvatt: sorry to highlight you, but one of my friends has this http://imagebin.org/133511 [18:54] with a radeon 5400... [18:54] Sarvatt, ideas on what is in #4-9? [18:54] any ideas would be helpful :) [18:56] shadeslayer: I dont see anything wrong there? :) [18:57] Sarvatt: the weird colors [18:57] on the profile pic [18:57] Sarvatt: http://imagebin.org/133511 << [18:58] err [18:58] http://imagebin.org/133512 [19:00] Sarvatt, bug #705572 [19:00] Launchpad bug 705572 in apport (Ubuntu) "retracer attaching incomplete backtrace (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/705572 [19:01] Sarvatt: whereas actually : http://imagebin.org/133588 [19:14] Sarvatt, notice the xorg.0.log.old's are filled with bo map failed: Cannot allocate memory messages