[00:21] is anyone here using an acer iconia W500 tablet with precise? [00:22] touch events seem to get all messed up after a couple of minutes [00:35] mdeslaur, don't think anyone here has that hw. certainly haven't heard of that problem so far, although I haven't been perusing the precise input bugs [00:36] mdeslaur, not aware of any recent sru's for X's input bits, possibly kernel got some update? Tried booting an earlier kernel? [00:36] hrm, using mtview...when the problem occurs, mtview is drawing events two inches away from my finger [00:36] them I switch windows and back, then it's back to being fine for a bit [00:37] bryceh: I tried with precise release, precise with updates, and precise with proposed [00:37] mdeslaur, I do have a fujitsu touch screen, which tjaalton tinkered with last uds, which had a problem like that, except it was like that all the time [00:38] and iirc it was only the x coord that was off [00:38] hrm, i bought this one because everyone was saying oneiric worked great on it...so I guess I'll try oneiric [00:38] mdeslaur, ok so all three of those versions have the same problem? [00:39] bryceh: yeah, precise is a no-go [00:39] precise has the franken-xserver as far as the input stack goes. that's been at the heart of a variety of input bugs in precise [00:42] mdeslaur, alright so booting different kernels probably won't do anything. xorg-edgers *might* be worth testing. But I'd check with tjaalton since he might know something relevant. [00:42] bryceh: cool, thanks [00:47] * mdeslaur tries xorg-edgers crack [00:49] this makes mdeslaur a crackhead [00:50] would you like treatment? [00:50] bjsnider: this free sample is great! [00:50] :) [00:55] lol [00:55] xorg-edgers broke unity [00:55] \o/ [01:04] "broke' is such a strong word. [01:04] it is functionally challenged [01:05] hehe [01:05] unity looks all nice and clean without any icons being drawn in it [01:12] that's a glass-half-full approach [03:52] mdeslaur: that's thd mesa blocker bug for us.. anyway, you could've tried the backport stack ppa [04:11] mdeslaur: if you need to revert and are on amd64, dpkg -l | egrep 'sarvatt|ricotz|edgers' will give you a full package list, you just remove the ppa sources, apt-get update then add every package returned there to a sudo apt-get install package1/precise package2/precise package3/precise line, just ignore libdrm-nouveau2 [04:27] uploaded mesa_8.0.3+8.0.2-0ubuntu3.2 [04:27] to precise-proposed [06:57] * mlankhorst slaps Sarvatt [06:58] what did he do this time?-) [06:59] he knows what he's forgotten :D [08:24] i'll package -omap, obsoletes -omapfb [08:25] oh it's packaged already [08:39] and git messed up [08:44] uploaded anyway [08:44] didn't push the branch, it's maintained in collab-maint but wanted to ask the maintainer first [08:51] meh, -msm [08:51] no upstream changes since 2010 [08:51] why do we still carry it? [08:51] ogra_: do you know where xf86-video-msm is used? [08:52] oh htc dream / nexus one [08:52] i say purge it [08:54] vbox next, and after that the stack can move [08:54] debfx: i'm looking into patching vbox to build against 1.13.. === seb128_ is now known as seb128 [09:57] tjaalton: if you want I can do that, it needs 2 things, 1. Disable XAA, 2. compat header [09:58] mlankhorst: ok, go ahead :) [09:59] i started but still trying to figure out the way it's built [10:01] who cares, debuild it.. :p [10:06] i'll just look at a build log :) [10:12] debuild -nc >:) [10:27] but I don't see vboxmouse being built so I guess it uses another way :) [10:34] oh great it's too dumb to even have xaa, makes things easier though :) [10:34] yeah [10:37] compat-api.h is identical on every driver? [10:37] no [10:38] but the idea is the same [10:38] ok [10:42] the only difference is in those #ifdef XF86_SCRN_INTERFACE, I think things are changed there as needed by the specific driver :) === yofel_ is now known as yofel [11:33] tjaalton: hi, so xorg-edgers didn't solve my touch issue anyway...cnd said it's probably a bug in the gesture stack they're aware of [11:35] mdeslaur: ok, good to know [11:58] sigh [11:58] mlankhor1t: wanna fix -msm too?-) [12:03] aw was just done with virtualbox :p [12:04] trying to update sbuild so it supports cross-building [12:05] seems virtualbox builds now [12:05] cool [12:05] do you have ubuntu in a vm to test with? [12:05] not vbox [12:06] hm, could install it on the quantal machine [12:06] or maybe not [12:06] I'll see if I can make it do the netboot thing [12:06] if the dependencies get right, just ship it [12:07] because netboot [12:07] nah I want to send it in if it's not too much trouble [12:09] let debfx handle that part :) [12:09] heh sure [12:10] in the meantime we can upload it to proposed [12:11] I'm just rechecking if it builds on old one first [12:11] if so I'll just make it a dumb quilt patch [12:12] * mlankhor1t assumes building == working === mlankhor1t is now known as mlankhorst [12:13] tjaalton: where's msm then? [12:16] ok seems to work until it complains I overrode my opengl, ship it.. http://people.canonical.com/~mlankhorst/vbox-x1.13.patch [12:16] dget https://launchpad.net/ubuntu/+archive/primary/+files/xf86-video-msm_1.0.1%2Bgit20100122.5f7df591-1ubuntu5.dsc [12:16] thanks [12:17] tjaalton: yeah but it is really a 2010 snapshot? :o [12:17] yup [12:17] no new upstream versions [12:17] have fun with it kthxbye :) [12:17] can we drop it and never speak of it again? [12:17] :o [12:18] well, you can probably wait until we've got confirmation that it's still used [12:21] yay [12:22] so ship it to quantal? >:D [12:22] mlankhorst: can't VBOXDRIScreenInit use SCREEN_INIT_ARGS_DECL as well? [12:23] probably but this works so why bother [12:23] heh [12:23] yeah [12:23] that function converts it anyhow [12:24] would also mean defining SCREEN_INIT_ARGS to pass it along [12:24] it's mostly meant to make xorg<->driver api easier, and this is driver internal :) [12:25] i mean so that it builds against earlier abi too, ie. upstream could take it as-is [12:25] but oh well, they can sort it out [12:25] it will build against earlier api now.. [12:25] I've built against both for testing [12:26] ah, ok [12:28] VBOXDRIScreenInit is no longer SCREEN_INIT_ARGS_DECL with either X api, it doesn't pass the pScrn or index on newer :) [12:29] gotcha [12:29] but geode was a bigger mess, at least this one followed the standard api making it easy :) [12:33] test building before upload [12:40] yep, ship it [12:42] mmh, -ivtvdev is hopefully the last one that remains [12:42] skip it [12:43] it's part of universe right? [12:43] multiverse [12:43] even worse :) [12:44] not much happened there either since 2010 [12:44] if anything [12:44] so yeah skip it [12:45] added it to the purge list [12:45] 14 drivers there now [12:46] :-) [12:46] afk for a bit -> [12:47] I'm going to need that list at one point! :D [13:07] mlankhorst: bug 1034793 :) [13:07] Launchpad bug 1034793 in xserver-xorg-video-voodoo (Ubuntu) "Please remove obsolete video drivers from the archive" [Wishlist,Confirmed] https://launchpad.net/bugs/1034793 [13:12] yeah omapfb has been replaced and doesn't even work right in precise [13:24] brb [13:52] mlankhorst, tjaalton, hi :), be aware that 304.37 still seems to be broken with 1.13rc4 [14:30] sigh ._. [14:49] tough [14:50] ricotz: what's broken then? [14:52] mlankhorst, xorg crashes while just starting firefox or running g-s, same issue since 304.x introduced 1.13 support [14:53] http://paste.debian.net/plain/183527 [14:53] havent retraced it though since it has the same symptom [14:54] (i mentioned it to aaronp) [14:58] * mlankhorst tries i915 with valgrind [15:03] hey a conditional jump depending on uninitialised bytes in vprintf.. [15:05] oh and still no synaptics debugging symbols? what went wrong there [15:09] oh, no merge from debian-experimental which had it [15:12] tjaalton: hey can you upload synaptics? [16:04] mlankhorst: yup, uploaded [16:04] thanks :) [16:05] hmm, rejected [16:06] o.O [16:06] did I not push the previous changed, meh [16:07] yup.. oh well, I'll merge with origin [16:11] there [16:12] ah k :) [20:11] holy crap, unity work with today's mesa checkout [20:11] \o/\o/ [20:13] Sarvatt: ship it! [20:14] * mlankhorst is having some fun with wine [20:14] someone actually respond to the pull request? [20:15] sending one with a commit starting with [HACK] maybe wasnt a good idea :P [20:19] Sarvatt: do you think they honestly care about what I send any more? [20:26] I was hoping to stir up some discussion but that seemed to have failed [20:30] where was this? [20:37] wine ml [20:42] sent a pull request to their patch mailing list in the hopes of stirring up some discussion, surprised Sarvatt knows :p [20:43] ah [20:50] * mlankhorst pokes Sarvatt [21:21] mlankhorst: #nouveau scrollback :) [21:22] ah ofc [21:25] really frustrating though [21:30] was surprised to see fedora was shipping it even, any idea if YokoZar is going to enable it in ubuntu? [21:32] no idea, I've been using it for ages lol [21:32] holy crap, wine in wheezy got upgraded to 1.4.1 [21:32] the ubuntu-wine ppa has it though [21:32] quite stable [21:42] Sarvatt: fedora ships wine modifications? not that I know of at least.. [21:42] http://pkgs.fedoraproject.org/cgit/wine.git/tree/wine.spec [21:42] check out may 28th [21:42] oh :p [21:42] http://pkgs.fedoraproject.org/cgit/wine.git/tree/wine-pulse-1.5.4.patch they use that, no attribution on it so no clue if its yours or the other one [21:43] well it is, just an old version === JanC_ is now known as JanC