/srv/irclogs.ubuntu.com/2011/10/19/#ubuntu-x.txt

cndbryceh, RAOF: I have a small patch for oneiric's x server: http://paste.ubuntu.com/71263500:19
cndit's an obvious fix for a code path that I must have missed testing00:19
cndbut it is hit everytime a touchpad gesture is accepted by our new gesture recognizer00:19
RAOFCausing a memory leak?00:20
cndwhat's the best way to get this into oneiric, since precise won't have these patches00:20
cndRAOF, no, dereferencing a NULL ptr00:20
cndit's not obvious from the patch00:20
RAOFFair enough.00:20
cndbut the if statement is literally "if (first)"00:21
cndthe if statement that you can't quite see at the top of the patch00:21
RAOFNothing in oneiric actually hits this, right?  At least, *my* touchpad gestures don't seem to crash the server :)00:21
cndno, you'll only hit it if you are using touch grabs on trackpads00:22
cndwhich I am doing with my new gesture recognizer00:22
RAOFBut it's still SRUable even if it doesn't apply to precise.00:22
cndyeah00:22
cndso that's my question00:22
cndshould I just leave precise as is?00:22
cndcreate a new git branch for oneiric00:22
cndand then upload following the sru procedure?00:23
cndbecause precise hasn't actually deviated from oneiric yet, has it?00:23
cndso I *could* upload it to precise too00:23
RAOFPrecise hasn't deviated just yet, no.00:24
RAOFI've got an ubuntu+1 branch already which will become ubuntu once it's ready.  Until then, feel free to push to ubuntu and from there to oneiric-proposed.00:25
cndRAOF, so I should skip precise?00:26
RAOFWe can copy the upload from oneiric-proposed into oneiric-updates & precise (if we haven't already uploded a new server to precise, in which case we don't need to anyway ?)00:27
RAOFIn short: yeah, skip precise.00:27
cndok00:27
cndRAOF, bryceh: I'm going to upload to oneiric-proposed00:37
cndheads up, just in case you were working on an sru too00:38
brycehalright00:39
cndRAOF, are you able to release a package into -updates?01:12
RAOFI am, yes.01:12
cndthere's an sru that has passed verification (though hasn't been tagged), and it's been a week01:12
cndI was wondering if you could take a look01:12
RAOFWhy hasn't it been tagged?01:12
cndnot sure01:13
RAOFUnless it's tagged it doesn't show up as green on the SRU page.01:13
cndwho is supposed to tag it?01:13
cndbug 82795801:13
ubot4Launchpad bug 827958 in libgrip (Ubuntu Precise) (and 3 other projects) "eog crashed with SIGSEGV in gdk_x11_window_get_xid() (affects: 36) (dups: 6) (heat: 160)" [Undecided,Triaged] https://launchpad.net/bugs/82795801:13
RAOFAnyone can tag it; the person who's done the verification is fine.01:13
cndoh, in the past someone always came by and tagged it for me01:14
cndthey read the comments, and then tagged it as appropriate01:14
macooh oh there are people01:14
macocan someone help me with how to respond to http://lists.alioth.debian.org/pipermail/pkg-nvidia-devel/2011-October/006721.html ? when you're done with that bug01:14
cndRAOF, I may see the issue01:15
cndI'm moving bugs to fix committed01:15
cndinstead of leaving them as in progress01:15
cndRAOF, wiki.ubuntu.com/SRU says the SRU team will set the verification tags as appropriate01:16
RAOFFix committed is the right status for "uploaded to -proposed".01:16
cndok01:16
cndnm then01:16
RAOFYeah.  We do set those tags.01:17
cndhmmm, who sets the verification-needed tag?01:17
RAOFThe person who accepts it into -proposed.01:17
cndthere seems to be a whole in the process, according to the wiki page01:17
cndoh, I'm getting ahead of myself01:18
cndthe bug has verification-needed01:18
RAOFI think one reason that bug hasn't been tagged as verified is that it's not immediately obvious that it did succeed.01:18
cndthat's one thing I worried about01:18
RAOFSo people browsing through their SRU mail won't necessarily go "ok, time to set verification-succeeded"01:19
cndahh01:19
cndRAOF, does this bug look ok to you for being verified?01:19
RAOFAnd it's perfectly reasonable to raise these to ubuntu-sru members.01:19
RAOFIt does, yes.01:19
cndok01:19
cndit's only been 8 days, so just one day past the waiting period01:20
cndso I didn't want to grab someone too fast01:20
cndI just happened to notice that you are an sru member :)01:20
RAOFMarked as verification-done; once I've got through the pending-sru queue I'll run through the srus that are verified.01:23
cndRAOF, ta!01:25
cndmaco, if you need info on the nvidia binary driver, tseliot would be the one to ask01:26
macocnd: as far as i can tell, he's NEVER on irc01:26
cndmaco, he's in italy01:27
cndmaybe your timezones aren't overlapping01:27
cndyou could also email him at alberto.milone@ubuntu.com01:27
macoi'm in the eastern us. we should overlap at least for the morning01:27
cndyeah, I know he's on in the morning01:27
macoi'll look again tomorrow then01:28
cndI can try to ping him tomorrow morning as well01:28
cndbut I'm on the west coast, so I may not catch him either :)01:28
cndmaco, he'll be in orlando next week too01:29
cndso you should be able to catch him on irc during the day then as well01:29
cndoops, nm, looked at the wrong alberto01:29
cndhe'll be at uds in a week and a half01:30
RAOFmaco: nvidia-settings is open-source, though.  I've not checked the source, but the fallback-to-compile-time-variable sounds like it would work.01:30
macoyeah i missed that bit of the code when i was going "see if it makes a call to pkg-config...oh hey look it does, ok i guess that bug reporter is right"01:31
* RAOF is surprised by "do_butmap_changed"06:14
arahey!08:41
arais there any main bug where you guys are tracking the issue with jockey and hybrid gpu systems? 08:41
ara(where jockey will show drivers, but will be unable to load them)08:42
tjaaltonara: maybe filed against jockey, dunno08:47
aratseliot1, ^ any ideas?08:47
tseliot1ara: no, sorry, I've never seen that bug report08:51
araOK, thanks, then I will suppose that the master bug is https://bugs.launchpad.net/ubuntu/+source/jockey/+bug/66044308:53
ubot4Launchpad bug 660443 in jockey (Ubuntu) "Recommended drivers don't work with Nvidia Optimus (affects: 3) (dups: 1) (heat: 20)" [Undecided,Confirmed]08:53
araalthough it is strange to not see more activity on it08:53
tseliot1ara: yes, that one should work as the master bug09:11
eltociaHi, my keyboard and mouse are forzen at log in (11.04 64), I edited xorg.conf adding AutoAddDevices false, and only the mouse came back. I reinstalled xserver-xorg and fglxr with no success. Any suggestion? thx10:44
tjaalton"frozen" means you have no mouse/kbd10:50
tjaaltonif you add that option you're on your own10:50
eltociayes, they are dead at log in. but I can switch to terminal with alt F1, after pressing alt-sysrq-r. I tried to restore a backup'd copy of xorg.conf, too. But it didn't work10:52
tjaaltonfile a bug without the AutoAddDevices-nonsense10:54
eltociaok, I'll do it. With that line in xorg.conf I'm able to use the mouse, though. With the original conf, or with a new one generated with X -configure, mouse doesn't work, unless I edit the config10:57
tjaaltonwhat if you start it without xorg.conf?10:59
eltociathey're still dead11:01
tjaaltoninstall pastebinit, and run 'pastebinit /var/log/Xorg.0.log'11:02
tjaaltonpost the link here11:02
eltociaok, I'll need to reboot to do that, because I'm using an other OS right now. I'll come back as soon as I have done it. 11:05
tociaHi, here's my log as requested before. http://paste.ubuntu.com/71309711:35
tjaaltontocia: nope, you have the xorg.conf in place11:36
tociaI suspect it could be a layout issue, since HAL is working, xorg is working and it's not the hardware. But I still can't understand why the mouse is dead until I add the option AutoAddevices "False". Well, I'll need to look for something else, or just do a clean install and see if it works11:42
tjaaltonHAL isn't used11:42
tjaaltonudev is11:42
tjaaltonwhy can't you just move the xorg.conf aside and redo what I asked11:42
tociaI did boot without xorg.conf, mouse and keyboard are still dead, but in this case when I switch to a terminal with alt F1 my screen freezes11:44
tjaaltonfor starters, purge fglrx11:44
tociadone, nothing changed, I purged xserver-xorg, too11:45
tjaaltonuh11:45
tjaaltonno need to do that11:45
tociait was a suggestion, I tried lot of things this two days11:46
tjaaltonsuggestion by whom?11:46
tjaaltonanyway, unless you've rebooted and removed the xorg.conf, you still have the fgrlx module loaded etc11:47
tociaIt was a suggestion in the ubuntu chat, I don't remember who said that to me. I did reboot when I rm the xorg.conf11:48
tjaaltonso did you purge the 'fglrx' package before reboot?11:49
tociayes11:51
tjaaltonwhat does 'apt-cache policy xserver-xorg-input-evdev' tell11:52
tociaI need to reboot to do anything on ubuntu, with no keyboard on the GUI, I have to browse the internet on win7.11:55
tjaaltonyou don't need to browse internet..11:56
tociaand to chat here, too11:56
tjaaltoni suspect you don't have -evdev installed11:56
tjaaltonreinstall xserver-xorg and see if it installs the missing drivers11:57
tociaI already did a reinstall of that package after I purged it. But I can do it again. along with fglrx11:59
tociaAnyway, Is that right to purge fglrx? sudo apt-get remove --purge xorg-driver-fglrx fglrx*12:01
tociaI purged fglrx, reboot, reinstalled xserver-org, reboot, same as before. apt-cache policy xserver-xorg-input-evdev says: Installed 1:2.6.0-1ubuntu12/ Candidate 1:2.6.0-1ubuntu12/ Version Table:/ *** 1:2.6.0-1ubuntu12 0/500 http://ch.archive.ubuntu.com/ubuntu/ natty/main amd64 Packages/ 100 /var/lib/dpkg/status12:34
tjaaltonok, so pastebinit Xorg.0.log12:35
tjaaltonit worked with 11.04 at some point?12:36
tociaIt worked since last weeks, I upgraded to 11.10, I had some problems and I reinstalled 11.04, it worked for a couple of days until I had this problem. When I was installing some package with Synaptic suddenly mouse and keyboard frozen, I rebooted and they were still dead12:40
tjaaltonok12:42
tjaaltoni'd rather fix the issues with 11.10 though12:42
tociaI had black screen and fglrx problems after the second reboot of the installation, I wasn't able to resolve them, so I reinstalled 11.04. I'll come back with the log in 512:47
tociahttp://paste.ubuntu.com/71316012:57
tjaaltontocia: and again, you have the xorg.conf there..13:01
tjaaltontocia: pastebinit /etc/X11/xorg.conf13:01
tociammh, I lookd for the xorg.conf after the reinstall of xserver-xorg and I didn't find it, now I see it's been used, too13:03
tociathat's the old one I used before the issues http://paste.ubuntu.com/713172/13:05
tocianow I reboot and look if there's a new one13:05
tjaaltondoesn't the default ati driver work, or why do you use fglrx?13:06
tociayes, they work, but I'm using the fglrx since very long without any problem13:08
tociaI'll try to switch back to ati drivers and see if something changes13:09
tjaaltonbut had issues in 11.10.13:10
tjaaltoni'd say install that and use the default driver..13:11
jjohansenSo I have been suffering from an odd corruption for a while13:13
jjohansenhttp://people.canonical.com/~jj/screeshot.png13:13
tociaOk, I'll give a shot at that.13:13
tjaaltonjjohansen: xterm?13:13
tjaaltonoh13:14
jjohansentjaalton: no not just xterm, all windows13:14
tjaaltonnot just the term13:14
jjohansengoes away when the window get refreshed13:14
tjaaltonwhich hw/driver?13:14
jjohansentjaalton: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (primary) (rev 0c) (prog-if 00 [VGA controller])13:18
jjohansenKernel modules: intelfb, i91513:18
macotseliot1: hello?13:18
jjohansensorry, was doing lspci in a terminal on the wrong machine at firts :)13:18
tseliot1maco: hi13:19
macotseliot1: i need help replying to debian nvidia people13:19
macotseliot1: can you look at http://lists.alioth.debian.org/pipermail/pkg-nvidia-devel/2011-October/006721.html ?13:19
tjaaltonjjohansen: this with oneiric?13:20
jjohansentjaalton: yes and natty13:20
jjohansentjaalton: actually I can't say most recent natty, I upgraded the machine early in the cycle13:20
tjaaltonmy x61 seems to work fine13:21
jjohansenhrmmm, I was afraid of that.  Possibly a hardware issue on my end13:22
tjaaltonis this with dualscreen or some other special case?13:22
jjohansennope, regular laptop, shows up after a day or two uptime, no suspend needed13:23
jjohanseneven13:23
tjaaltonah, ok13:23
tjaaltontbh i don't use the machine much anymore13:23
macotseliot1: this was related to bug 72145313:24
ubot4Launchpad bug 721453 in nvidia-settings (Ubuntu Oneiric) (and 1 other project) "nvidia-settings should depend on pkg-config and python-gtk (affects: 3) (dups: 2) (heat: 26)" [Undecided,Fix committed] https://launchpad.net/bugs/72145313:24
tseliot1maco: yes, sorry, I'm multitasking (poorly). I'll have a look at it13:25
macothanks13:25
jjohansentjaalton: well it is getting a little long in the tooth, but I haven't found a machine to replace it with yet.  It has a nice 12.1" 1400x1050 ips display and they don't see to do anything close anymore13:25
tseliot1maco: oh, so the fix is already in13:26
macotseliot1: in ubuntu, yes, i was suggesting it to debian as well and then the usefulness of adding pkg-config came up and i dont really know13:26
macoi know its on my system and nvidia-settings works for me!13:27
tseliot1maco: do they use my code to make use of policykit (through pygtk2) in Debian?13:27
macoi dont know. is that whats breaking it here? ubuntu-specific patchiness?13:28
tseliot1maco: I'm wondering why the package would need pkg-config at runtime13:31
tseliot1maco: screen-resolution-extra is what the nvidia-settings panel uses (thanks to an Ubuntu patch) to allow users to write xorg.conf through policykit (i.e. without having to run the whole app as root)13:33
macotseliot1: http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/precise/nvidia-settings/precise/view/head:/src/XF86Config-parser/Generate.c i was looking at line 217 when i went "hey yep that sure looks like it uses pkg-config"13:34
macowhat i missed was line 248 13:35
tseliot1maco: now, that's new to me. Forget what I said about policykit, it their code that's using pkg-config at runtime.13:36
=== kinoucho` is now known as kinouchou
tseliot1maco: do you need anything else from me to help you reply that email?14:00
macotseliot1: is it that netiher is required in debian?14:19
tseliot1maco: I believe that pkg-config is needed in debian (since the Nvidia code calls it) but pygtk2 is not (as we use it only in Ubuntu). Good catch BTW14:21
macotseliot1: debian people said that with the fallback they dont think pkg-config is necessary either14:22
tseliot1maco: if the problem doesn't affect debian users, then I guess they're right. This would mean that, somehow, the fallback is broken in ubuntu. I don't think our move to multi-arch could have caused this. Did you see what find_libdir() returns when using the fallback?14:26
macono. im leaving a message on the bug asking the original reporter how he determined pkg-config was needed14:29
Dice-Manhey guys14:31
Dice-Mani got a problem with my touchpad14:31
tseliot1maco: it sounds like a good idea ;)14:34
=== soreau_ is now known as soreau
brycehheh, ironically just had the most annoying tech support call ever, with my mother, setting up dual head21:32
mdeslaurbryceh: hehe :)21:34
brycehshe wanted the displays mirrored but at their own native resolutions21:38
brycehsecondary problem - seems to be no way in gui to set which monitor is primary, so with it !cloned everything was on the wrong monitor21:38
mdeslaurbryceh: remind me to buy your mother any hardware I need fixed :)21:40
brycehheh21:40
brycehit's funny to realize the dorky minuscule things that roadblock non-technical users21:41
brycehlike, dragging the panel toolbars from screen to screen is hard because there's no visual clue that things are "dragging"21:42
brycehor in the monitor gui that you have to click on the laptop box to set the laptop settings.  _quite_ perplexing to mom21:42
mdeslauryeah, doing tech support for non-technical users is a real eye opener for details like that21:43
mdeslaurmy mother in law used to click column titles in evolution to sort her mail by date or subject, and if she moved the mouse by a couple of pixels while clicking, evolution thought you were trying to drag away the column to delete it.21:44
brycehyeah my dad has shaky hands and so certain seemingly simple tasks can be kind of challenging for him21:46
brycehand touchpads are right out21:46
mdeslaurbryceh: now fix my bugs, or I'll send you mom a poulsbo21:49
mdeslaur:)21:49
brycehheh21:50

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