/srv/irclogs.ubuntu.com/2011/02/21/#ubuntu-x.txt

raevolthere's been a lot of work done on it since it was pulled into maverick before release, and i'd like to take advantage of some of the new code :/00:00
RAOFThere's always xorg-edgers.00:02
raevolyea i'm not that harcore :[00:02
raevolhardcore* god my typing is terrible today00:02
cndRAOF, I have good news :)00:14
cndI'm right now packaging all the stuff up to test and push to my ppa tonight00:15
RAOFcnd: Excellent.00:15
RAOFI'll try to have 1.10RC2 merged by the time you're ready, then :)00:16
RAOFcnd: You've been a-workin' on the weekend? :(00:20
cndRAOF, umm, and last weekend too...00:21
cndI haven't had a day off for two weeks now :(00:21
cndRAOF, when will you be uploading xorg-server to ubuntu (or to git)?00:22
cndwhen I'll be able to get the source, basically00:22
RAOFI'll probably be able to get it done today, so that would be in <6hrs time.00:23
cndok, so after I00:24
cndI'm asleep :)00:24
RAOFI was thinking that we'd use this opportunity to also fold in the Xi 2.1 stuff; that'll just be a couple of patches, right?00:24
cndyeah00:24
cndso if you want, you can push to git00:24
cndand then I can push my updates to git00:24
cndand then you can upload00:24
cndsound good?00:24
RAOFThat'd work.00:25
cndRAOF: when do you want to push by?00:25
RAOFIf you've got stuff ready now you could just push to git and I'd just fold that in with the rest of the merge.\00:26
cndit's not ready quite yet00:26
RAOFOr you can push after me, which would be tomorrow morning for you I guess ;)00:26
cndI'm making packages right now00:26
cndand then I need to test locally00:26
RAOFSoft!00:26
cndand I was hoping to have one day tomorrow to have people bang on it00:26
RAOFReal men test in the archive :)00:27
cndheh00:27
RAOFOk.00:27
RAOFThere's no OMGHUGE rush for the merge; a day's testing wouldn't hurt it, either.00:27
cndk00:27
cndI'm so excited :)00:29
RAOFObviously FF is our deadline :)00:29
RAOFAnd you just can't hide it?00:29
cndespecially since it actually works :)00:29
RAOF:D00:29
cndthe one thing holding me back is that I'm still glossing over a few things00:29
cndlike what do you do for pointer emulation of a new touch when there's an active grab on the pointer?00:30
cndcorner cases like that00:30
RAOFFiendish protocol-level corner cases.  Yay.00:30
cndI think it won't fall over, but I'm not 100% sure it'll behave as one might assume00:30
RAOFAnd you'll also get to take a couple of days off once the FF deadline hits, right?00:30
cndheh, I hope so :)00:31
cndoubiwann is awesome, he'll probably make me take the time off :)00:31
RAOF:)00:31
cndtesting looks good, time to push to git repos and make source packages :)01:03
cndtjaalton, it looks like you've uploaded a new libxi, but the git tree hasn't been updated01:13
cndcan you update it for me?01:13
RAOFOoh, funky.  Damage bugs!01:31
cndit takes waaaay too long to upload a qt package...01:56
cnd207 MB source tarball...01:56
RAOFXi2.1 Qt?01:58
tjaaltoncnd: it's been synced from debian, there were no ubuntu changes06:40
RAOFOh, no.  intel/nouveau switcheroo problems :(06:58
RAOFI suspect bug #718620 of being switcheroo related.06:58
ubot4Launchpad bug 718620 in xorg-server (Ubuntu) "Xorg assert failure: X: ../../dix/pixmap.c:118: AllocatePixmap: Assertion `pScreen->totalPixmapSize > 0' failed. (affects: 9) (dups: 11) (heat: 92)" [Medium,New] https://launchpad.net/bugs/71862006:58
tjaaltonRAOF: regarding bug 718331, there's a newer upstream release which could help07:45
ubot4Launchpad bug 718331 in xf86-input-wacom (Ubuntu Natty) (and 1 other project) "Xorg crashes on wacom input moule (affects: 5) (dups: 1) (heat: 30)" [High,Incomplete] https://launchpad.net/bugs/71833107:45
RAOFtjaalton: Aah, quite possibly.07:45
RAOFArgh.  Why doesn't everyone have debugging symbols installed all the time? :(07:45
tjaalton0.10.11, though .12 should be arriving too at some point07:45
tjaaltonright :)07:45
tjaaltonall the xorg-packages should imho also have a -dbg counterpart07:46
tjaaltonbut wacom is maintained by ron07:46
RAOFNah; -dbgsym is easy.07:46
tjaaltonit's not apt-get'able?07:46
RAOF*Debian* should have -dbgsym, and we should drop all the trivial -dbg packages.07:46
RAOFtjaalton: It totally is.07:46
tjaaltonoh07:46
RAOFSee https://wiki.ubuntu.com/DebuggingProgramCrash :)07:47
tjaaltoni'm missing the source then07:47
tjaaltonfixed07:48
RAOFI wonder whether we shouldn't have that in the default sources.07:48
RAOFAt least during development.07:49
tjaaltonright07:49
tjaaltonhuh, uses a different signing key07:50
tjaaltonoh wacom 0.10.11 was released last week, so it's quite fresh07:54
RAOFcnd: xserver is in git, for your delectation.08:02
RAOFtjaalton: I don't suppose you have a switchable graphics system to confirm my suspicion that 718620 is switcheroo-based (and get a better backtrace at the same time ☺)?08:02
tjaaltonRAOF: nope..08:03
tjaaltondoes switcheroo work on a desktop with embedded and discrete graphics, or is it laptop only? (not that I have such a desktop either)08:03
RAOFIt requires ACPI support; it's likely that only laptop manufacturers bother.08:04
tjaaltonok08:04
tjaaltonRAOF: i've a patch to fix mumble, will push soon08:57
tjaaltonto xorg-server08:57
RAOFtjaalton: Sweet.  I'd also like to convince myself that 718620 isn't an xserver problem.09:25
tjaaltonRAOF: yeah, nasty..09:27
tjaaltonlunch->09:28
RAOFIt's probably something in the plymouth integration patch that doesn't work when copy_from_fb fails.09:31
tjaaltoneh, https://launchpad.net/xserver11:51
cndtjaalton, true, but I still can't find it in git12:54
cndI can push the changes to git though12:54
tjaaltoncnd: the ubuntu-branch there is stale, last updated 21 months ago12:58
tjaaltoncnd: so you can pretty much do whatever you like with it :)12:58
cndyeah13:02
apwcnd, playing with mumble to try and work out why the PPT buttons don't work13:25
apwits looking like the xserver hasn't sent us the events we asked for ... is there any way to ask the Select extenstion whats goign on?13:26
apwor any small examples of asking for key events i can use for testing13:26
jcristauthere's a fix for the mumble thing in git, aiui13:26
apwoh ... pthth ... 13:26
apwjcristau, got a pointer to the git so i can look ?13:27
jcristauhttp://git.debian.org/?p=pkg-xorg/xserver/xorg-server.git;a=commitdiff;h=74acb8b958a0e19beae993f0c1f4627ab2296ee013:27
apwjcristau, oh so its an xserver issue ...13:28
jcristau(also at http://patchwork.freedesktop.org/patch/4200/)13:28
apwdo we already have that fix coming?  and i am just behind or ?13:29
jcristauit's queued for whenever the next upload is, i think13:29
apwjcristau, heh then thank you i can go back to ignoring it for a bit :)13:29
tjaaltonapw: you're subscribed to the bug, should've seen that it's "fix committed" :)13:33
cndapw: the next upload should be tomorrow13:35
cndwith my multitouch stuff :)13:35
apwtjaalton, heh oops14:36
apwcnd, sounds good ... except i expect it'll be all broke :)14:36
cndapw, of course it will!14:37
cndI'm going to try my hardest to somehow cause a kernel panic from multitouch :)14:37
ricotzSarvatt, hi, perhaps it is useful to propose the pixman update before FF while the notify-osd update seems to take longer15:11
lilsteviis there a way to get an onscreen keyboard to automatically launch with netbook launcher15:17
tjaaltonricotz: national holiday in US today ;)15:33
ricotztjaalton, oh, ok ;)15:34
=== lilstevi is now known as lilstevie
Sarvattricotz: yeah been 2 weeks now, sounds good to me :)16:03
ricotzSarvatt, are you able to sponsor it?16:16
Sarvattricotz: nope I don't have permissions for it. tjaalton, would ya be willing to sponsor a pixman 0.20.2 update for ricotz? his update looked good to me when I reviewed it16:17
ricotztjaalton, that would be great http://people.ubuntu.com/~ricotz/pixman/16:18
Sarvatt0.21.x isn't appropriate for natty, 0.20.2 is the latest stable release, there is just an issue with notify-osd popups looking screwed up that has been waiting to have the fix uploaded since december but feature freeze and all, dont think we can wait anymore16:21
seb128it's easy enough to grab the notify-osd commit and upload to natty16:23
seb128don't bother about small breakages in unstable series16:24
ricotzseb128, yeah, macslow said he wanted to do it soon16:24
seb128usually having things broken motivate to get the fix in when that has been not really moving for a while16:24
seb128he was sick this weekend and has work to finish before feature freeze16:24
seb128so it's likely to not land before feature freeze16:25
ricotzright, but we wanted to prevent a bug flood ;)16:25
ricotzok16:25
Sarvattseb128: well I was getting a crazy amount of emails about the problem just from edgers and was worried about when it actually went in the distro16:25
seb128can't you just upload a notify-osd with the patch in natty?16:26
Sarvattnot sure the guy signed the contributor agreement16:26
seb128Sarvatt, no need to have a c-a for a distro patch16:27
seb128that's only an issue to merge in the upstream trunk16:28
tjaaltonSarvatt: sure, i'll upload it later when i'm back home16:44
ari-tczewshrugs, when we will get new nvidia driver?16:53
ricotztjaalton, thanks16:55
ricotzari-tczew, depends on when nvidia decides to release a working one16:57
ari-tczewricotz: changing resolution and fan's speed every boot making me crazy16:58
ricotzari-tczew, nouveau should work pretty fine detecting the right resolution, of course it doesnt take care of your fan speed17:00
ari-tczewricotz: 1024x768 is not native resolution for my lcd17:01
ari-tczewso it's not prettty17:01
ricotzok, if you are nouveau and nvidia blob is removed, it looks like a bug which might be reported17:02
ricotzprobably an edid problem17:02
=== yofel_ is now known as yofel
alex_mayorgaon bug 553789 should I provide further peeks?17:42
ubot4Launchpad bug 553789 in xserver-xorg-video-nouveau (Ubuntu) (and 2 other projects) "X freeze/crash with nouveau driver (affects: 22) (dups: 5) (heat: 108)" [High,Confirmed] https://launchpad.net/bugs/55378917:43
alex_mayorgaX keeps freezing every 2-4 hours or so17:43
alex_mayorgawhat else should I gather?17:44
alex_mayorgaI have an ssh window to the frozen laptop17:44
alex_mayorgabryceh: ping17:47
tjaaltonalex_mayorga: no17:50
alex_mayorgatjaalton: so for now is just patience and frequent reboot?17:52
alex_mayorgaor should I try edgers or something?17:52
alex_mayorgaor is there a way to restar X from an ssh window?17:52
hyperairrestart gdm17:53
hyperairor rather, sudo restart gdm17:53
alex_mayorgahyperair: thanks! Let me try that17:53
alex_mayorgadidn't do unfortunately17:55
tjaaltonalex_mayorga: there is no known fix, nor is there one on the horizon17:58
alex_mayorgatjaalton: I see, this would bug me on every distro right?17:58
tjaaltona workaround is to disable acceleration, the upstream bug, at least, has th details17:59
alex_mayorgaalso I hear the nvidia blob is not much better anyway17:59
tjaaltonyes17:59
alex_mayorgalet me review, but I think accel is already off here17:59
alex_mayorgais there other driver that I can use other than nouveau18:08
alex_mayorgavesa perhaps?18:08
tjaaltonnot unless you disable kms18:14
yofelreally? I did try vesa as a replacement for nouveau once, did work, but with rather low resolution18:15
yofelcould be that I had nvidia installed and just nouvea blacklisted maybe18:16
tjaaltonprobably18:17
yofelI'll try nouveau.noaccel=1 later, I don't use it anyway18:19
yofeltjaalton: would my bug be a dup of that one too? I think so since I've an GT218 here18:21
tjaaltonyofel: dunno, i'd have to check18:23
yofel bug 71190818:24
ubot4Launchpad bug 711908 in xserver-xorg-video-nouveau (Ubuntu) "[natty] frequent nouveau freeze on GT218 [NVS 3100M] (rev a2) (affects: 2) (heat: 12)" [Undecided,New] https://launchpad.net/bugs/71190818:24
tjaaltonyofel: yeah looks like it18:27
tjaaltoni can reproduce it when i _disable_ dri18:28
tjaaltonthough i have nv5x18:28
tjaaltonand had to switch firefox to chromium18:28
tjaaltonotherwise it would crash, though the bt looks different18:29
tjaaltonricotz: no _source.changes for pixman?19:05
ricotztjaalton, sorry, should be there now19:09
tjaaltonricotz: thanks, uploaded19:15
cndbryceh, or anyone else who's a core dev: can someone review the utouch-grail upload and push to ubuntu?20:08
cndhttps://bugs.launchpad.net/ubuntu/+source/utouch-grail/+bug/72278020:08
ubot4Launchpad bug 722780 in utouch-grail (Ubuntu) "Upload utouch-grail 1.0.19 to Ubuntu (affects: 1) (heat: 8)" [Undecided,New]20:08
cndit's a dependency of the new xi 2.1 work20:08
tjaaltoncnd: i can check it out a bit later20:09
cndtjaalton, that would be great!20:09
cndlet me know if you need anything20:10
tjaaltonsure20:10
tjaaltoncnd: ok, what exactly should I do to upload it?22:12
cndusually didrocks helps out, but he seems unavailable22:12
cndwe give him a link to our packaging branch22:13
cndhe builds the package and signs it22:13
cndand uploads it22:13
cndI can build it for you instead22:13
cndif you'd prefer that22:13
tjaaltonthe source, yeah22:13
tjaaltonor wait22:13
tjaaltonI'll try something first22:13
cndsure22:13
cndthe packaging branch is linked in the bug, if you've got that handy22:14
tjaaltonyep22:14
tjaaltonhmm no, I'd need the tarball anyway etc. guess it's best if you prepare the source and I'll debsign&dput it to the archive22:15
cndok22:16
tjaaltonI haven't used bzr for packaging, so :)22:16
cndahh22:16
tjaaltondon't know what the workflow is22:16
cndit's actually quite nice, especially if you are doing packaging of new upstream sources22:17
cndthere's a command that takes the upstream tarball and the current packaging branch, and does crazy commits and merges for you22:18
cndand then gives you a new commit in your packaging branch that's all ready for you to upload22:18
tjaaltonalrighty22:18
cndtjaalton, it's up at people.canonical.com/~cndougla/utouch/22:21
tjaaltonoh it was a native package22:21
tjaaltonbut thanks, will upload22:22
cndwait22:23
cndno, it's should be native22:23
cndtjaalton, why do you think it's native?22:23
tjaaltoncnd: normally the upstream tarball is a separate file22:23
cndoh wait22:23
cndI forgot to put it there22:23
cndlet me upload it too22:23
tjaaltonthat explains it then :)22:24
tjaaltonhmm EPERM too22:24
tjaalton+NO22:24
tjaalton:)22:24
cnd?22:25
cndI just uploaded the tarball22:25
tjaaltoncan't access them22:25
tjaaltonforbidden22:25
tjaaltonchmod 755?22:25
cndI'll try22:25
tjaaltonor 644 at least22:25
cndtjaalton, I did chmod a+x *22:26
cndsorry a+r22:26
cndso hopefully you can get them now22:26
tjaaltonyep22:26
tjaaltonforgot to finalize the release?-)22:27
cndoh yeah...22:27
cndgrrr22:27
cndsorry22:27
tjaaltonheh22:28
cndusually didrocks does that for us :)22:28
tjaaltonindeed, since he'd have to sign it himself22:28
tjaaltonum22:28
cndwell, if you mean putting your name in the changelog22:28
tjaaltoni mean create the source package22:28
cndyou can sign without doing that22:28
tjaaltonyes22:28
cndanyways, I'll create a new package as released22:29
cndand upload again22:29
tjaaltoncool22:29
cndsorry for the trouble22:29
tjaaltonnp22:29
cndtjaalton, ok, the new package is ready for you22:36
tjaaltoncnd: and uploaded22:37
cndtjaalton, thanks!22:37
cndwe need to make a packaging group for utouch so I can upload them...22:39
cndI just never got around to requesting it22:39
tjaaltonheh, right22:39
brycehcnd, yep22:42
brycehcnd, or just get core-dev ;-)22:46
cndbryceh, how high is the bar for that?22:46
cndI've so far only touched linux-firmware, utouch*, and x stuff22:47
brycehcnd, generally you'd get motu first, which that's probably enough.22:48
brycehit's more about "do we trust him" than "does he have enough experience"22:48
cndI don't touch much in universe :)22:49
brycehagain, it's not so much a point of having experience in the area as much as being trusted to upload stuff in that area22:49
cndand most of the stuff I do need privs for are core packages22:49
cndok22:49
cndmaybe I should at least attempt motu now22:50
brycehonce you've gotten motu, spend a bit of time doing some random package sponsorship and stuff for a month or two, to gain some sponsor statements (which should be easy for you), then put in for core dev.22:50
cndok22:51
cndbryceh, RAOF: got some questions for you about versioning23:10
cndwe're pushing xi 2.1 stuff23:10
brycehok23:10
RAOFcnd: Shoot.23:10
cndbut there's no official upstream input proto packages yet23:10
cndnothing like inputproto 2.0.99.90123:10
brycehcnd, so you need a git snapshot?23:11
cndso I've been just adding them into ubuntu patches on top of what's already there23:11
cndwell, it's not even upstream git snapshot yet23:11
cndnot in the official repo23:11
cndso I've just been incremening the ubuntu version suffix23:12
cndand adding patches23:12
RAOFThat seems reasonable to me.23:13
cndone issue is that for libxi lintian complains because we're adding symbols with an ubuntu version23:13
brycehthat can probably be ignored in this case23:13
cndit's actually a lintian error23:13
RAOFLintian is wrong.  Feel free to silence that with an override if you want.23:13
cndwill that cause an issue?23:13
brycehdon't think so23:13
cndI didn't think so either, since I remember xorg-server having lintian errors too :)23:14
RAOFIt'd cause an issue in Debian because dak's started to reject packages with lintian errors; I don't believe soyuz does that :)23:14
cndahh23:14
brycehcnd, yeah doing incrementing ubuntu versions is probably fine.  I think I'd probably do it that way myself23:14
cndok23:15
cndI'm going to prepare packages right now23:15
RAOFThere's probably no way we can stick an ubuntu string into the input ABI version, is there?23:20
cndRAOF, that sounds really ugly :)23:22
cndbut it may be helpful23:23
cndRAOF, the alternative is to leave the input abi at 1223:31
cndand to assume the input abi includes the new mt stuff in evdev and synaptics23:31
cndI'm almost tempted to say that's a better route23:31
RAOFWe could also bump the minor version of the input abi?23:31
cndbecause in reality, input abi 13 should be backwards compatible with 12, but whot likes to bump the major version for some reason23:32
cndRAOF, even that could potentially be an issue if they bump the minor version upstream23:32
cndthough that's highly unlikely at this point23:32
cndrc2 is a little late for that :)23:32
cndRAOF, ok, I'm thinking that may be a good idea23:32
cndbump the minor to 12.323:33
RAOFI think aaronp has declared the ABI frozen.23:33
cndand I can convince whot to bump the major to 13 when mt is added23:33
cndleaving the major at 13 would also mean we don't have to rebuild all the input module packages23:33
RAOFMajor at 12, you mean?23:34
cndI'm glad someone has finally declared it frozen :)23:34
cndyeah, leaving it at 1223:34
cnd:)23:34
brycehseriously23:34
cndbryceh, was that in response to my comment about abi frozen declarations?23:35
brycehcnd, yep23:35
cndheh, ok23:35
cndit puzzled me for a moment, I thought you had better ideas about the input abi handling :)23:35
brycehnot seriously23:35
cndheh23:36
cndeveryone should play with qt's fingerpaint demo23:36
cndit makes me happy :)23:36
cndand they have a pinch to zoom demo where there's a giant world of cheese with mice running on it23:37
cndand you can pinch to zoom in and out23:37
RAOFOh, bah.23:38
cndRAOF, btw, thanks for fixing that doxygen issue :)23:39
cndit was really getting on my nerves :)23:39
RAOFDebian's X server Provides: xorg-video-abi-9, we Provides: xorg-video-abi-9.023:39
RAOFcnd: KiBi also fixed it :)23:39
cndahh23:39
RAOFbryceh: Do we want to go with Debian's packaging and incur another rebuild of the world, or diverge for now to include the minor version in those ABI virtual packages too?23:41
cndtjaalton, I think your xi 1.x fix is causing warnings:23:42
cndhttp://pastebin.ubuntu.com/570315/23:43
cndjust a guess since I think your patch touched that file23:43
cndRAOF, so you think you'll be ready for uploads today?23:44
RAOFcnd: I think so, yes.23:44
cndcool23:44

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!