[02:00] tjaalton: btw I tried manually copying the libdrm-dev headers back into /usr/include/drm, but it made no difference; -ati still hangs on X start. [02:01] surprisingly, vesa actually looks pretty good, although it's a lower res and no dual-head [08:02] bryce: the headers are used only when building the drivers [08:02] or do you mean you rebuilt -ati? [08:05] ah. no didn't rebuild it [08:08] shouldn't make a difference [08:08] so it just hangs? [08:08] yep, I'd had it working fine and dandy with intrepid, then did the upgrade [08:09] X fails to start, and switching to vts just displays corrupted screens [08:09] so my next hypothesis is an EXA/XAA problem [08:09] ah, so you can change the vt when it happens? [08:09] is it a r5xx? [08:10] RV630 [08:10] hmm, dunno which generation that is [08:10] if it's DRI capable, try disabling it [08:11] RV630 [Radeon HD 2600XT] [08:11] ok, it's not that [08:11] maybe it's trying to use the wrong output? [08:12] bryceharrington.org/files/Xorg.0.log.old [08:14] I don't think it's a wrong output issue - the card has 2 outputs and I have monitors connected to both. [08:14] heh, ok [08:14] there's a new upstream version in experimental.. try that [08:16] yeah [08:16] hmm, this looks suspicious: [08:16] Output 69 disable success [08:16] Output 51 disable success [08:16] Blank CRTC 0 success [08:16] Disable CRTC 0 success [08:16] Blank CRTC 1 success [08:16] Disable CRTC 1 success [08:17] right [08:17] that's why I suspect an initialization issue [08:31] bug 312924 looks similar [08:31] Launchpad bug 312924 in xorg "HD3450 configuration fails on reboot in jaunty" [Undecided,Incomplete] https://launchpad.net/bugs/312924 [08:44] yeah [08:45] and bug 311748 [08:45] Launchpad bug 311748 in xorg "HD3470 AND HD3200 X56TR laptop x doesn't start / black screen" [Undecided,Incomplete] https://launchpad.net/bugs/311748 [08:50] I've duped them [08:50] em, the former with the latter [08:51] great [08:51] I'm back to work tomorrow, and I think I'll focus the day on sorting out this particular bug (or upstreaming it at least) [08:52] the new upstream version didn't help [08:52] ? [08:52] right, I'll try that tomorrow first [08:53] ah, ok [08:53] (today I've been doing shop project stuff, and tonight am installing the new bookcase I made) [08:56] my wife's excited because I've got a pile of old books out in the main room that will be moved in now, and it'll eliminate clutter she's had to look at [08:56] hehe [08:57] it's -12C outside, so I'll stay in and weed the bugs [08:58] ouch :-) [09:01] btw, I've changed my contract to 20h/week so I can concentrate on my studies the next two months (at least), so might be that I need to lower my commitment to X/ubuntu for a while :/ [09:07] ah, good to know [09:07] how'd your tests go btw? [09:07] good practice :) [09:07] next ones the week after next week [09:08] I was just so jetlagged the whole week, and I actually had to skip the first one because I couldn't stay awake for more than 5min [09:09] ..at a time [09:09] :-/ I was worried something like that'd happen. Hope your professors were sympathetic [09:11] well they want to get rid of us (slackers), so they have no choice :) [09:11] I'll merge -ati, it can't be worse than the current version :) [09:15] howdy! any chance to get the fix to bug #255991 moved from hardy-proposed to hardy-updates? [09:15] Launchpad bug 255991 in xserver-xorg-video-geode "xf86-video-geode: DDC probing broken on GX2/CS5535 since 2.9.0 (patch)" [Undecided,Fix committed] https://launchpad.net/bugs/255991 [09:18] Q-FUNK: isn't that something for an archive-admin? [09:20] yeah, need to talk to pitti or slangasek [09:20] tjaalton: I'm not sure. last time I asked, I was told to test and ensure that it doesn't introduce any regression. [09:20] ok [09:25] tjaalton: what are you studying? [09:27] tseliot: major in semiconductors & materials in electronics (blech), minor in software engineering & business [09:28] so the courses I have left are real gems, like physics II, inorganic chemistry etc :) [09:29] hehe, a lot of fun [09:29] a lot, yes [09:30] will it give you the chance to be paid more? How does it work in Finland? [09:30] sure [09:30] if you have a degree it gives you more options [09:30] ok, then it's definitely worth the effort [09:31] although there are plenty of experts who have dropped out and made a good living [09:32] you already have a job, right? [09:32] sure, had one for ten years now.. one reason why it's been taking so long :) [09:33] aah, I thought you started your degree course recently [09:34] more recently [09:34] heh [09:34] good luck then [09:35] thanks [18:52] bryce: the new ati version didn't help your setup? [18:53] I've installed it; I'll reboot after I finish fiddling with a few bug reports [18:53] looks like evertjan didn't get the new version yet [18:53] II) Module radeon: vendor="X.Org Foundation" compiled for 1.5.99.3, module version = 6.9.0 [19:45] Hi guys, i was just reading http://blog.qa.ubuntu.com/node/9 and i hear you need testers [19:46] I have a quad screen setup (2 nVidia 8800GT cards, both dual head) and i currently run 2 X screens to manage it [19:46] i'd love to help you guys test, so just give me a call :) [19:46] love the sound of xrandr :) [19:56] CShadowRun: thanks, welcome to the ubuntu-x team :-) [19:57] CShadowRun: I assume you're using the -nvidia driver? [19:57] yup :) [19:58] what are you interested in testing particularly? [19:58] well i'd like to see the xrandr stuff for multiple heads [19:58] because xinerama is a bit crap, and so are multiple X screens lol [20:01] it's a shame to see multiple X screens fall into a state of disrepair, though [20:01] heh, well you're sort of stuck, since you're using the -nvidia proprietary binary driver [20:01] ah [20:01] well i can always put another installation of ubuntu on one of my drives to test with [20:01] with -nvidia you first need to wait for them to provide xrandr 1.2 support [20:02] yea [20:03] i well, i shall just twiddle my thumbs till it comes along [20:03] unless theres any better ideas :) [20:05] CShadowRun: are you using nvidia-graphics-drivers-177? Do you have jaunty installed? [20:06] if so, one area you could help is by looking at bugs in https://bugs.edge.launchpad.net/ubuntu/+source/nvidia-graphics-drivers-177 [20:08] some documentation for how to check bug reports is at https://wiki.ubuntu.com/X/Triaging [20:12] bbl - testing new -ati [20:54] I can probably load jaunty on one of my spare drives and check out some of those bugs :) [22:29] tjaalton: yep reverting the EXA patch and going back to XAA solved the issue for me [22:30] bryce: ok, so it was the same bug then [22:30] strange that the patch _looks_ ok [22:31] I'll need to look it over a bit more [22:31] there's actually 2 bugs - one is why EXA isn't working (which may or may not be the same issue as graber), and the other is why XAA can't be specified [22:31] it's the same [22:32] I bet EXA works for you if you force it in the conf [22:32] ok lemme try [22:33] you're right [22:33] bryce@chideok:~$ grep XAA /var/log/Xorg.0.log [22:33] bryce@chideok:~$ grep EXA /var/log/Xorg.0.log [22:33] (**) RADEON(0): Option "AccelMethod" "EXA" [22:34] btw, I subbed ubuntu-x to radeontool [22:34] there's no bugs against it, but I think we should probably get the package up to date with what's in airlied's git tree [22:35] ok [22:48] aha, found one issue... setting AccelMode to XAA does not actually turn off useEXA, it ust leaves the default [22:50] ahaa!! [22:50] tjaalton: ok I understand what's going wrong [22:51] great :) [22:51] it *is* an EXA problem [22:51] graber actually isn't getting EXA (neither am I) [22:51] (==) RADEON(0): No acceleration support available on R600 yet. [22:51] heh [22:52] there's logic that prevents considering EXA if CHIP_FAMILY_R600 [22:52] so it forces XAA for that [22:52] however my patch ignored that and set EXA as the default before that condition was checked [22:52] so I was forcing EXA on R600 for graber and I, thus exposing the EXA problem [22:53] maybe the other bug should be closed as well then (311748), or marked as dupe of the first? [22:53] secondarily, there's another problem I mentioned above where specifying XAA doesn't switch you to XAA, only to the default [22:54] tjaalton: no, I think that's a different problem actually [22:55] ok [22:55] they are reporting seeing that issue on intrepid, which doesn't have my patch so could not be seeing this problem [22:55] ah [22:55] right [22:56] well, now there's a change it might work [22:56] chance [22:57] yep [22:58] hmm [22:58] interesting; in reading the logic it sounds like even XAA is unsupported on this chipset. bummer [22:59] well, r6xx only gained acceleration support a couple of days ago :) [23:00] but not something for jaunty.. [23:00] well, basic performance seems fine [23:07] ok, made a new patch... time to build and test [23:10] hrm, the j key on my keyboard is not producing j's reliably... not a good keyboard for development of uanty [23:10] heh [23:11] ok here we go [23:12] seems okay [23:15] I can't test setting XAA though [23:15] at least, not easily right now [23:15] right [23:15] ship it [23:16] heh, done [23:16] hmm, with sufficient firefox windows open, performance issues become evident [23:20] tjaalton: where'd you see r6xx gaining accel support? The mainline tree seems to still have it disabled [23:21] bryce: not the -ati driver, but the fuss about the DRM code etc. radeonhd has exa support with the experimental code, probably will land in -ati too after a while [23:21] ah right [23:29] yeah, scrolling pages with the ati driver on r6xx with 1680x1050 as resolution is really a pain :) [23:29] for now I'm using my laptop as a server and using my Asus EEE (intel graphic :)) with an external screen ssh-ing my laptop [23:30] stgraber: I've uploaded a new -ati which should fix the problem you were seeing [23:30] stgraber: sorry it was my lousy coding that broke it [23:32] bryce: no problem, thanks for fixing it :)