/srv/irclogs.ubuntu.com/2010/06/08/#ubuntu-x.txt

Sarvattwhoa! syncpackage -d unstable -r maverick -V 1:1.0.1.xsf1-3 --uploader="Robert Hooker <sarvatt@ubuntu.com>" -b 589528 libxprintutil   00:01
Sarvattsyncpackage -d unstable -r maverick -V 1:2.3.0-3 --uploader="Robert Hooker <sarvatt@ubuntu.com>" -b 589530 vesa00:01
RAOFI've pushed to the branch again, now as 2:1.7.6-2ubuntu7.1 and targetting lucid-proposed.00:02
Sarvattoops xserver-xorg-video-vesa for that second one00:02
Sarvattat the end00:02
Sarvattsyncpackage -d unstable -r maverick -V 1:1.4.11.dfsg-4 --uploader="Robert Hooker <sarvatt@ubuntu.com>" -b 589531 xserver-xorg-video-mga00:02
Sarvattsyncpackage -d unstable -r maverick -V 6.8.1-3 --uploader="Robert Hooker <sarvatt@ubuntu.com>" -b 589540 xserver-xorg-video-r12800:03
Sarvattthat rocks, auto uploads and closes the bugs00:03
brycehSarvatt, ok syncs updated, ack'd, and archive subbed00:04
Sarvattsyncpackage -d unstable -r maverick -V 1:12.6.9-2 --uploader="Robert Hooker <sarvatt@ubuntu.com>" -b 589925 xserver-xorg-input-vmmouse00:04
Sarvatti dont know raof's email for the evdev one00:04
brycehRAOF, should be 7.2 - remember my "don't email bryce" patch the other day ;-)00:05
RAOFAaah!00:05
brycehsorry, didn't know you were planning to maintain the lucid xserver in git else I'd have committed it there00:06
Sarvattsyncpackage -d unstable -r maverick -V 0.8.8-4 --uploader="Robert Hooker <sarvatt@ubuntu.com>" -b 589535 xf86-input-evtouch00:06
brycehactually if I'd known I'd have just slipped it in with these four, that'd be easiest00:06
RAOFbryceh: Yeah.  These fixes got a little bit lost in the ubuntu branch.00:06
Sarvattahhh synaptics, merging00:06
Sarvatterr building i should say00:07
RAOFSarvatt: It's already merged, innit?00:07
Sarvattyup00:07
brycehhttps://wiki.ubuntu.com/X/PackageNotes updated00:08
RAOFbryceh: Launchpad doesn't seem to think you've uploaded xorg-server with the “don't email bryce” patch?  It should therefore be possibly to merge the patch in with the other four?00:08
brycehRAOF, alrighty, you mind merging it in?00:09
RAOFNot at all.00:09
brycehworst case the admins will ask us to redo it and untangle00:09
brycehI got an ack but no indication it's been accepted into the queue yet00:09
Sarvattwe arent in import freeze so you can just upload directly cant ya? those lines i pasted earlier automatically grab everything and upload and close the bugs00:10
brycehoh do they?  hmm00:11
Sarvattyeah need ubuntu-dev-tools from bzr though! its an awesome script00:11
brycehbryce@blumonc:~/src/xorg-server$ syncpackage -d unstable -r maverick -V 0.8.8-4 --uploader="Robert Hooker <sarvatt@ubuntu.com>" -b 589535 xf86-input-evtouch00:11
brycehCommand not found00:11
brycehbryce@blumonc:~/src/xorg-server$ apt-cache search syncpackage00:11
brycehbryce@blumonc:~/src/xorg-server$ 00:11
Sarvattyeah ubuntu-dev-tools git, its not released yet00:12
brycehgit?00:12
Sarvatthttp://sarvatt.com/downloads/merges/xserver-xorg-input-synaptics/ (its building now to be sure itsall there)00:12
Sarvattbzr00:12
Sarvattsorry :)00:12
brycehdo I just run from the bzr tree or is there some installation magic?00:13
Sarvattdeb is here if you are on i386 http://sarvatt.com/downloads/merges/00:13
brycehok00:14
Sarvattoh its all arches00:14
brycehE: No credentials found for 'ubuntu-dev-tools', please see the manage-credentials manpage for help on how to create one for this consumer.00:15
Sarvatttoo good to be true huh00:16
Sarvattoh guess i already created credentials00:17
brycehok sorted that out00:18
brycehhmm, doesn't like me trying to upload stuff as you00:18
Sarvatttry passing -k KEYID?00:18
brycehshould I specify myself as the uploader?00:19
brycehor do I need to just import your key or something?00:19
brycehgpg: skipped "KEYID": secret key not available00:20
Sarvattnah just do it as yourself, more important we get them uploaded anyway00:20
jcristaureplace it with your actual key id :)00:20
brycehaha00:21
brycehdpkg-buildpackage: binary and diff upload (original source NOT included)00:21
brycehxf86-input-evtouch_0.8.8-4fakesync1_source.changes00:21
jcristauRAOF: so i'm looking at putting stuff in pristine-tar for xorg-server, and every tarball i import seems to add 1M to the repo.00:21
brycehwhat is "-b 589535"?00:21
brycehbytes?00:22
jcristaubug?00:22
Sarvattbug number it closes00:22
brycehahh00:22
brycehSuccessfully signed dsc and changes files00:23
brycehxserver-xorg-input-vmmouse_12.6.9-2_source.changes00:23
Sarvattnice, it adds Launchpad-Bugs-Fixed: to the .changes00:24
brycehbryce@blumonc:~/src/ubuntu-dev-tools$ syncpackage -k E0E67611 -d unstable -r maverick -V 6.8.1-3 --uploader="Robert Hooker <sarvatt@ubuntu.com>" -b 589540 xserver-xorg-video-r12800:24
brycehWARNING! Overwriting modified Ubuntu version 6.8.1-2ubuntu1, setting current version to 6.8.1-200:24
Sarvatthuh00:25
brycehhmm same with mga00:25
bryceh$ syncpackage -k E0E67611 -d unstable -r maverick -V 1:1.4.11.dfsg-4 --uploader="Robert Hooker <sarvatt@ubuntu.com>" -b 589531 xserver-xorg-video-mga 00:25
brycehWARNING! Overwriting modified Ubuntu version 1:1.4.11.dfsg-2ubuntu1, setting current version to 1:1.4.11.dfsg-200:25
Sarvatthmm now ubuntu-dev-tools has acksync in it too! wth00:26
brycehhrm, got it for -vmmouse too, but didn't notice it before doing the upload00:26
bryceh$ syncpackage -k E0E67611 -d unstable -r maverick -V 1:12.6.9-2 --uploader="Robert Hooker <sarvatt@ubuntu.com>" -b 589925 xserver-xorg-input-vmmouse00:26
brycehWARNING! Overwriting modified Ubuntu version 1:12.6.5-4ubuntu2, setting current version to 1:12.6.5-400:26
jcristaui guess the warning is to make sure you really want to get rid of the delta?00:27
brycehjcristau, yep00:27
brycehalso the -evtouch sync failed... needs to be fakesync'd00:27
brycehso add that one to the merges pile00:27
jcristauRAOF: so i'm looking at putting stuff in pristine-tar for xorg-server, and every tarball i import seems to add 1M to the repo :/00:28
RAOFjcristau: Urgh, really?00:29
Sarvattnice, ack-sync even has an option to build the packages too00:30
Sarvattits just a wrapper around syncrequest00:30
RAOFjcristau: I don't have much more insight than “man, that sucks”, though :)00:30
RAOFSorry, call with Rick.00:31
jcristauRAOF: i'm thinking it's because it has a lot of autotools-generated files, which aren't in git00:31
RAOFThat may well be it.00:31
brycehok looks like for -mga the only change is a patch that's already upstream, according to https://edge.launchpad.net/ubuntu/maverick/+source/xserver-xorg-video-mga/1:1.4.11.dfsg-2ubuntu100:32
brycehso I've put -mga through00:32
RAOFjcristau: Incidentally, I noticed there were a lot of files in git but not in the tarball.  I've removed them locally, so lintian now doesn't complain about changing files outside of debian/.  I'll merge that in to debian-experimental later, if you like.00:33
Sarvattyup, made sure each of those had every patch we had upstream before requesting it (or bugged jcristau to update things to sync) :D00:33
brycehSarvatt, ok are you 100% sure all of these can just sync and ignore the ubuntu changes?00:34
Sarvattsyncpackage -d unstable -r maverick -V 1:2.3.2-6 --uploader="insert roafs info here" -b 586659 xserver-xorg-input-evdev00:34
brycehotherwise we'll need to go through and check each individually00:34
Sarvattyes 100% positive, i didn't request anything that wasn't in debian or upstream already00:35
Sarvattand if it was in upstream and not debian i updated debian :D00:35
brycehalrighty00:35
brycehxserver-xorg-video-r128_6.8.1-3_source.changes done00:36
RAOFsyncpackage -d unstable -r maverick -V 1:2.3.2-6 --uploader="Christopher James Halse Rogers <raof@ubuntu.com>" -b 586659 xserver-xorg-input-evdev00:36
RAOF^^ That's what you were after, Sarvatt  ;)00:36
jcristauRAOF: i tend to ignore this lintian bit.  the files are in upstream git, so deleting them means merge pain if they get modified upstream, and unless they're huge removing them doesn't buy us much. i don't care much though.00:36
brycehRAOF, thanks00:36
brycehxserver-xorg-input-evdev_2.3.2-6_source.changes uploaded00:37
jcristauRAOF: (if they're in the debian branch but not upstream, then that's a different story, but i think we mostly got rid of that a while back)00:37
RAOFjcristau: Fair enough.00:38
brycehlibxprintutil_1.0.1.xsf1-3_source.changes uploaded00:38
brycehSarvatt, got more for me?00:38
Sarvattgoing back over and seeing what ya uploaded00:39
bryceheverything in the Waiting list at https://wiki.ubuntu.com/X/PackageNotes was done00:39
brycehexcept one00:39
bryceh-evtouch has to be fakesync'd00:39
Sarvattsweet, edit-patch automatically adds patches if you have any local ones now too00:45
bryceh15 min remain... anything else I can help y'all with today?00:45
RAOFThe xorg-server SRU would be good.  I've merged your patch, and double checked that it builds.  I'll just push up to git.00:49
brycehok00:50
RAOFPushed in git.00:52
brycehuploaded to lucid-proposed00:57
RAOFThanks!00:57
RAOFbryceh: Have a good dinner tonight!01:01
brycehno prob, good luck with the rest of the uploads01:02
brycehhopefully you'll be able to find other sponsors but if you get in a pinch lemme know01:02
RAOFI think I'll be able to tap some of #ubuntu-desktop :)01:02
Sarvattargh, had to run out there, take care and thanks bryce01:25
Sarvattnot to ppa-purge all these machines and upgrade as things come :)01:27
Sarvattnow to do that rather01:28
Sarvattand \o/ you dropped the ati patch series, that ended up not doing anything for those bug reports anyway01:28
Sarvattoh.. you probably just forgot to git add huh? :( we had some other real post 6.13 fixes in there01:29
RAOFSarvatt: What are you talking about?01:36
Sarvattati?01:38
Sarvattin git?01:38
RAOFHm.  Maybe I did forget to git add.01:38
RAOFWell, it hasn't been uploaded yet.01:39
Sarvatti think i hit a new point of lazyness when i rename a tarball and include all the new commits as patches in the diff :D02:15
Sarvattmv libdrm_2.4.20+git20100527.607e228c.orig.tar.gz libdrm_2.4.20+git20100607.66375fd6.orig.tar.gz it is!02:16
Sarvattsuch a pain syncing libdrm and people think they are using the old one because of the name02:17
Sarvattbjsnider: your vaapi acceleration stuff is all in there now02:17
bjsniderin the libdrm?02:18
RAOFAh, the multi ringbuffer patch has finally landed?02:21
Sarvattyeah02:21
RAOFThat was a bit of a drama.02:22
bjsniderwish i had an intel card to test it on02:22
bjsniderwait, no i don't02:22
RAOFHah.  You totally do.  Intel hardware rocks.02:29
bjsnideryes it does02:31
jcristauRAOF: so running git import-orig on all tarballs to get the upstream/* tags on http://git.debian.org/?p=users/jcristau/xorg-server.git + the pristine-tar branch there adds up to ~15M afaict, which is more reasonable.02:44
RAOFThat's not terrible.  It's mostly a one-off cost, too, and bandwith is (generally) cheaper than my attention :)02:45
jcristaui downloaded all tarballs from snapshot.debian.org so that should be pretty much all xorg-server tarballs ever uploaded to debian02:46
RAOFExcellent!02:48
RAOFSarvatt:  You've added 102-disable-page-flipping-v2.patch to xf86-video-intel - on what hardware is that meant to be a problem?  Is there a bug link available?03:18
Sarvatti dont have a bug link handy because wenever shipped it, it's been a problem all throughout the late 2.10+ cycle and still a problem today in git and fedora is shipping the same thing03:21
Sarvattthey haven't figured out where the problem is yet03:21
RAOFSarvatt: I ask this because I'm very happily running without that patch on my invincible laptop.03:21
Sarvatt915-X3100 seems to be the range bad off03:21
RAOFFair enough.03:22
Sarvattand some of the devs are having it on GM45 too03:22
Sarvatt(what you have)03:22
* RAOF can recognise his own chipset :)03:22
Sarvatthere is fedoras http://cvs.fedoraproject.org/viewvc/F-13/xorg-x11-drv-intel/intel-2.11-no-pageflipping.patch?view=markup03:23
Sarvattsome people like hyperair can't go 10 minutes with it enabled, figured it was better to patch it preemptively since there isn't any kind of freeze03:24
Sarvatthttps://bugzilla.redhat.com/show_bug.cgi?id=58842103:24
ubot4bugzilla.redhat.com bug 588421 in xorg-x11-drv-intel "Pageflipping disabled in F13" [Medium,Assigned]03:24
Sarvatthttp://www.pubbs.net/201005/fedora/1529-page-flipping-on-intel-211-driver.html03:25
RAOFMan, that's a useless bug report :)03:26
Sarvatt(the patch was from jbarnes, I just changed the info message in it)03:26
rippsThat's it, I'm taking the plunge and upgrading to maverick03:31
RAOFGet in quick, before xserver 1.8 has been published :)03:33
rippsI'm already using 1.8 from xorg-edgers, is it very different?03:35
Sarvattppa-purge first!!03:35
Sarvattno its the same but ati has had a ton of changes since 6.1303:36
Sarvattword of warning though, i'm putting edgers on 1.9 any day now :D03:36
Sarvattjust in maverick03:36
rippsgood, I'll probably try it from maverick03:36
rippsxorg-edgers has maverick packages, right?03:37
Sarvattyeah, basically the same as lucid now, i upload the same stuff to both03:37
Sarvattmaverick's i686 instead of i486 though03:37
Sarvattif you're on x8603:37
rippsI use a athlonxp 2500+, I'm pretty sure that's i68603:38
stentenWait, would being on xorg-edgers interfere with the 1.8 switch in Maverick?03:38
rippsI'm purging xorg-edgers now03:39
stentenMost of the people on the Maverick Forum are running edgers.03:39
rippsI have alot of ppa's, any other's I should purge, like nautilus-elementary?03:39
RAOFstenten: Yeah, they're not really helping test X in Ubuntu.03:39
rippsI should probably purge the Unity and gnome-shell stuff too, while I'm at it03:40
Sarvattwell they probably wont be on edgers soon with all the breakage incoming :)03:41
RAOFSarvatt: So, it looks like there's at least one pageflipping bug that's been fixed upstream.  I think I'll pull those patches and drop the pageflipping disablement patch.03:42
Sarvattokie, the bugs will be hard hangs with the system completely dead and no way to get info out of it when you come across them03:43
RAOFHm.  I might check that we've actually got the kernel side of the fix - it doesn't seem to be in yet.03:44
Sarvatti've been dropping the patch from edgers every kernel upgrade to see if its fixed and readding it when its not03:46
Sarvattsome of the hangs might be fixed in rc2 though, i saw a few commits03:47
RAOFIt doesn't look like it's in 2.6.35-rc2, so that puts the kybosh on enabling page flipping it seems.03:47
Sarvatti think it was merged in the rc2 cycle but the commits were old enough to be back before rc103:49
Sarvattdrm/i915: Kill dangerous pending-flip debugging is one03:52
RAOFThere should also be a gen 3 patch; that doesn't seem to be in any git repository.03:52
Sarvattlooks like thats it upstream :(03:53
RAOFI'm talking about http://lists.freedesktop.org/archives/intel-gfx/2010-April/006463.html03:53
Sarvattyeah i saw it on the lists03:53
Sarvattapril...03:53
RAOFSo, the result is: page flipping is probably still broken, leave it disabled.03:54
RAOFAnyway, lunch time!03:58
Sarvattwill ask jbarnes whats up03:58
Sarvattthat one patch of his fixing hangs on dpms events still isn't upstream and it was from january i think..03:59
RAOFHah.  We get 1.8 uploaded and 1.9 rc1 gets branched.04:11
Sarvattyeah been testing it the past few days, intel is horribly broken :)04:16
Sarvatti'm blaming the gtk stuff since other people arent seeing it04:17
Sarvattwas hoping bgnr support could make the cut :(04:20
Sarvattholy crap, i got almost10k karma from uploading all the x driver/libs/protos04:22
Sarvattever figure out whats forcing 96 dpi everywhere?04:29
Sarvattwe had a patch doing it before -     - 138_fedora_xserver-1.3.0-default-dpi.patch04:32
Sarvatt      Changes default dpi to 96.04:32
Sarvatthmm might be gdm's xsettings manager plugin?04:34
rippsokay, system has been cleaned of a lot of ppas05:15
RAOFWe interrupt your regularly scheduled IRC proxy to bring you 5.8 GiB swap usage.06:11
RAOFSarvatt: http://bugs.freedesktop.org/show_bug.cgi?id=23705 is why X is set to 96 DPI06:11
ubot4Freedesktop bug 23705 in Server/general "xserver 1.7.0rc0 uses wrong dimensions" [Normal,Reopened]06:11
RAOFI'd like to revert that and let GNOME and KDE handle it if they feel they must.06:12
=== AlanChicken is now known as AlanBell
lucazadehi Sarvatt10:29
lucazadedid you read my message yesterday about poulsbo patch?10:30
asacRAOF: !!10:45
asacRAOF: did you get my message about stuff failing on armel?10:46
RAOFasac: Hi!11:19
RAOFasac: Ah, yes I did, but for some reason I didn't read it as something that needed responding to - I got it quite a lot after you sent it.11:20
RAOFasac: Does any arm hardware actually have intel graphics hardware?  If not, it shouldn't be hard to just drop the dependency and drop the intel build (on arm, so the same package builds appropriately in the archive)11:21
jcristaubuilding libdrm-intel on arm is nonsense afaik11:22
asacRAOF: it fails on armel with intel .so not there11:24
asacyes11:24
RAOFAha.  Ok.  I'll update the packaging to handle that, then.11:28
RAOF(Tomorrow)11:28
RAOFHm.  Now that there's an openvg library to build against, I wonder how cairo's openvg backend does… :)11:29
* RAOF is off to the movies. See you tomorrow!11:29
asacRAOF: whats the idea?11:29
asaci can do that too then?11:29
asacRAOF: we could --disable-intel on !x86/amd64 i guess?11:32
RAOFasac: The idea would be, in debian/rules, to not build anything that requires libdrm_intel on arm.  Right something like that.11:33
asacRAOF: yeah. so just not on armel or everywhere? how does the package actually build for all the archs in maverick?11:33
RAOFI didn't think intel built by default, but perhaps it does.11:33
asacit seems to do ... but i see that the build succeeded everywhere in maverick ... which feels like this might not be the problem11:34
asacanyway, enjoy your evening. i will try :)11:34
RAOFThe current archive package would build everywhere because it only enables i915 and i965 on amd64 and i386.  But it also --disable-gallium's11:34
RAOFTa.  Good luck!11:34
apwSarvatt, seems i am about to expire out of xorg-edgers are you able to wave a wand over that?12:52
Sarvattasac: you need libdrm-intel1 to build plymouth on arm :)16:45
asacSarvatt: why is that?16:48
asacthen i will just enable all the intel packages everywhere16:48
asacthey seem to build ... not sure why just the intel1 package was disabled on arm in the first plac16:48
tseliotasac: because plymouth has a drm plugin with different backends for radeon, intel and nouveau. It's what it uses to map the framebuffer, etc.16:51
asackk16:52
asacwill remember that when touching it next time16:52
asac(reenable intel everywhere)16:52
tseliotgood16:58
jcristautseliot, Sarvatt: plymouth needs to be able to disable some of the backends.17:24
tseliotjcristau, Sarvatt: I *think* it's possible to disable the drm renderer (thus disabling radeon, intel and nouveau)17:26
jcristaunot good enough17:27
tseliotI'm not even sure about that. I haven't checked17:28
nperryHello guys, wondered if somone could have a look at xserver-xorg-video-nouveau for maverick, it seems to of failed its last build becuase its still depending on  Xserver 1.718:57
seb128nperry, it didn't fail it's depwaiting19:01
seb128there is a known soyuz issue that it's taking a while to see that the depwait is cleared19:02
seb128the new xorg has only be newed some hours ago so you just have to wait19:02
nperryAh ok, no problems!19:05
brycehthere was an email to ubuntu-devel@ sent about this the other day20:06
RAOFasac: Any joy with the mesa build?23:47
asacRAOF: let me check ;)23:47
asactook more time then i had before leaving so felt good23:48
asacthumbs up: https://edge.launchpad.net/~asac/+archive/armel1/+build/177812623:48
asacyay23:48
RAOFYay!23:48
asacRAOF: anyway, i was told that we need to resurrect intel for something23:48
asacwill just enable it everywhere23:48
asacit seems to build everywhere at least ;)23:48
asacright23:48
asachere the reason:23:49
asac17:51 < tseliot> asac: because plymouth has a drm plugin with different backends for radeon, intel and nouveau. It's what it uses to map the framebuffer, etc.23:49
asacRAOF: ^^23:49
asacfor me it feels not particular !armel specific to build a driver for an intel chip23:49
asacas long as it builds it should be fine imo23:49
asacso anything against just setting the intel1 thing to any?23:49
RAOFApart from the fact that there's no hardware it could possibly drive on !{i386,amd64}, not really.23:50
RAOFIt'll just needlessly inflate the build times & package sizes on those archs.23:50
asacright23:50
asacbut i think things like radeon and ati could in theory exist23:50
asacwhy not intel ;)23:50
asacif its a usb/pci card, then it should just work23:51
* asac considers to find a card for his intel chipset ;)23:51
RAOFIntel don't make usb/pci cards :)23:51
* cwillu feels a sudden urge to check for video chipsets on digikey :p23:51
asacthey dont? hmm ... /me sees a market gap ;)23:51
asacopportunity its called23:51
asacbtw, http://identi.ca/notice/35424509 ... just so you know ;)23:51
jcristauasac: plymouth just needs fixing.  it's a 5 minutes job..23:53
RAOFThat would be the other option, yes.23:54
asacok i let you guys figure ouw what you want ... imo usb/pci based drivers don't need to be disabled just because its arm and its called intel ;)23:54
jcristauit's not usb/pci.  it's integrated graphics on intel kit.23:55
jcristauit comes on the chipset with an intel cpu.  or on the same chip as the cpu, lately.23:56
asacjcristau: but the driver is registered as pci ;)23:56
asacisnt it?23:56
* asac types lspci23:57
cwilluasac, when a device becomes available, the driver can be trivially enabled23:57
cwilluuntil then, why waste the resources?23:57
asaccwillu: right. why waste resources and disable it ;) ... we can just keep it enabled and dont need to touch packging23:57
lifelessRAOF: so, I have a gt200 (gtx260) reva23:58
lifelessRAOF: and I play games on it23:58
RAOFA fine use of a dedicated GPU.  GO on.23:58
lifelessRAOF: also, remind me about cedega full screen being offset by a couple of inches23:58
lifelessRAOF: what driver should I run on the gt200 - is nouveau there yet for (say) wow, civIV, steam:source games ?23:59
RAOFNo, I don't think so.23:59

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