=== tkamppeter__ is now known as tkamppeter [02:50] hyperair: #compiz-dev could probably just take it from you [02:51] Amaranth: a bit too late, the patch's already committed =) [02:51] but yeah i took it to #compiz-dev [03:02] hyperair: Did you see the PPA I started for compiz++? [03:02] Amaranth: YOU DID?! [03:02] I got compiz and libcompizconfig done, iirc [03:02] >=O [03:02] after all the efforts i went through compiling and stuffing into /opt! [03:03] But then seb128 said to merge with the debian packaging and I couldn't get any answer from the debian packager on my changes [03:03] ah i see. [03:03] So that's where I stopped working on it [03:03] I rewrote the compiz package to use dh 7 [03:03] awesome! [03:03] will there be a daily PPA for compiz++? [03:04] And I was going to include the plugin packs as submodules of the compiz package so they would always be in sync and we could easily creatively split them up based on usage [03:04] ah i see [03:04] that's an interesting idea. [03:04] Maybe, I might just go implement all my ideas anyway and get seb128 to recognize the awesomeness [03:04] but that'd require you to generate your own tarballs, right? [03:05] Oh, crap, I deleted my PPA because some idiots were trying to use it and it didn't have all the packages [03:05] lol [03:05] I hope I still have the source packages somewhere... [03:05] you should always maintain these things in git =p [03:05] hyperair: I believe dh 7 supports a source package with multiple tarballs [03:05] oh yeah it does. [03:05] Or maybe I would have to roll my own 'compiz' tarball [03:05] Oh, I still have my bzr branches [03:06] you might, if you're going for the submodule method [03:06] https://code.launchpad.net/~amaranth/compiz/0.9 [03:06] git-buildpackage hates submodules in any case. [03:07] Nice simple rules file [03:07] why are you still depending on cdbs, though? [03:08] the translation stripping stuff [03:09] afaik you don't need cdbs for that. [03:09] my dh7 packages don't. [03:09] e.g. nautilus-share [03:09] Do you strip the translations though? You actually have to put something in rules to do it [03:10] From the gconf schemas, I mean [03:10] eh? [03:10] gconf? [03:10] oh nautilus-share doesn't have gconf [03:11] wait a sec, i could have sworn i made nautilus-share use dh7 [03:11] why is it still using cdbs D= [03:12] anyway, i have to be going. ttyl === nigelbabu is now known as nigelb === almaisan-away is now known as al-maisan === al-maisan is now known as almaisan-away [06:21] Who feels like sponsoring some bugfixes! [06:22] RAOF: do you have any idea if the subpar performance of unity is related to nvidia drivers? [06:23] It's sub-par on nvidia? [06:23] well, not even a subpar performance - just that using unity (with some pretty standard apps) makes my laptop boil :/ [06:23] I should probably install the blob on one of my nvidia machines, but Alberto does such a good job of maintaining it! [06:24] temperature is ~7-8 degrees higher then what I get on standard desktop and it can raise up to ~90 [06:24] I guess it's possible that unity hits codepaths poorly-optimised by the nvidia drivers. I'd be a bit surprised, though, since AFAIK unity tends to be developed on a combination of intel & nvidia boxes. [06:25] I can see mutter constantly use some cpu even if I just type in the terminal.. [06:25] I'd give #ayatana a try, or... #ubuntu-dx? Where do our DX guys hang out? :) [06:26] on #ayatana afair [06:26] do you know who I should ask directly? [06:27] Not really; both DBO and MacSlow would probably be knowledgable in the appropriate areas, though. [06:27] kklimonda, whats up? [06:27] thanks, I'll ask them [06:27] DBO: unity is boiling my laptop :/ [06:27] well, not exactly but it's noticeably hotter [06:27] kklimonda, a little more explanation would be nice :) [06:28] DBO: there isn't much else - just by running unity (firefox, evolution, terminal, few more apps but nothing cpu intensive) my laptop runs 7-8 degrees hotter then when I use standard desktop. [06:29] kklimonda, okay a small battery of questions: CPU/GPU? CPU usage levels when running unity? Standard desktop == compiz? [06:29] DBO: compiz, core2duo 2.4Ghz with nvidia quadro 140m [06:30] kklimonda, using the nvidia proprietary drivers? [06:30] yes [06:30] hows CPU? [06:31] DBO: hmm.. I get constant 20% cpu usage when I move terminal window slowly [06:32] slower I move it the higher cpu usage is :) [06:32] thats actually expected, how is it just sitting there? [06:32] it's sitting at 1-2% [06:32] 4 windows: 2 terminals, transmission and rhythmbox in background [06:33] when I type in terminal cpu usage raises to ~7% [06:33] running maverick? === almaisan-away is now known as al-maisan [06:34] DBO: yeah, up to date [06:34] the CPU usage seems fine, almost idle [06:35] it is likely the high temperature is related to GPU being powered up into a higher power state [06:35] this is one of the few times a nvidia chip is working against you here... [06:35] can I actually force it to stay in powersave? [06:36] and my nvidia is a nasty beast - it's actively working on making my laptop useless by burning itself ;) [06:37] btw, I have crashed mutter/unity twice already [06:37] and I've been running it for an hour or so :) [06:37] * DBO sighs [06:37] DBO: You should have quit about a month ago. It was much more stable ;) [06:38] it wasn't us who broke it [06:38] all these new updates came into Maverick [06:38] Oh, yeah. There was a clutter update, wasn't there. [06:38] how much can I bitch about clutter in public? [06:38] but the stacktrace actually looks useful: __strlen_sse2 () at ../sysdeps/x86_64/multiarch/../strlen.S:31 [06:38] g_strconcat (string1=0x7fc006c0c50b "app-") at /build/buildd/glib2.0-2.25.15/glib/gstrfuncs.c:310 [06:39] ? more? [06:39] DBO: I can subscribe you if you can give me your launchpad id :) [06:39] jassmith [06:39] link me the bug? [06:39] it's not retraced yet but first 3 frames are available. [06:39] bug 648548 [06:39] kklimonda: Bug 648548 on http://launchpad.net/bugs/648548 is private [06:40] subscribing you [06:40] done [06:41] kklimonda, why is a Unity bug private such that not even the unity team can look at it? [06:41] DBO: until it's retraced it's not available to anyone because CoreDump is still attached [06:41] ahhh [06:41] well, looks like we got vala fucked again, I'll look at it [06:42] !ohmy | DBO [06:42] DBO: Please remember that all Ubuntu IRC channels share the same attitude of providing friendly and polite interaction with all users of all ages and cultures. Basically, this means no foul language and no abuse towards others. [06:42] DBO: it's not that hard to make vala generate stupid things :/ [06:42] indeed [06:44] RAOF, you think I should try for Ubuntu membership [06:44] yeah, you were right - It's probably gpu reving up to make all the bling show on the screen every time I do something. [06:44] DBO: Sure, if you want to. [06:44] I mean, I know you :) [06:45] hehe I am curious how that would go [06:45] DBO: after I spent last 20 minutes doing nothing but chatting here the temperature has dropped down. [06:45] "So what have you done for Ubuntu?" "Oh you know... stuff" [06:45] kklimonda, I suspect what is going on is as you do stuff mutter in all its wisdome is getting a big ol stream of damage events [06:46] unlike compiz its not all that smart about occlusion and just goes ahead and tfp's the living heck (welcome micahg) out of your system [06:46] that's unfortunate [06:46] this is causing your X driver to enter a spin loop (TFP is awesome) and spins the nvidia driver up to a higher power state [06:47] DBO: It should be using the GL occlusion query that locked the GPU on i965 :) [06:47] in so many ways I hate linux drivers [06:47] DBO: any chance you can fix that before natty release? ;) [06:47] because I refuse to use gnome shell [06:47] the crash? yes [06:47] mutter being stupid :) [06:47] Speaking of which… sponsorship of mesa & xserver-xorg-video-intel from http://cooperteam.net/Packages/ would be appreciated :) [06:48] the performance? my god man I rewrite the entire logic of how clutter picks actors (no small task, they chiding me for even suggesting i could do it) [06:48] rewrote* [06:48] I wonder if I can at least force nvidia to stay in powersave mode.. [06:48] It's not like I use it for anything [06:49] the worst purchase ever - for various reasons [06:49] oh wait, the ibook that broke 4 times was probably even worse.. [06:51] no you cant [06:51] nvidia has openly stated that [06:51] basically they need to be able to power up in dual monitor situations [06:51] so rather than giving you that control [06:51] well you get the idea... [06:52] right, users are obviously too stupid for their own good so It's important to think for them.. [07:02] DBO: btw, do you know if it's reported that, when an application is maximised, I can't double click on the top panel (where appmenu-gtk lives) to unmaximise it? [07:02] I haven't seen it reported but maybe I'm not looking at the right packae [07:02] kklimonda, thats more of a feature issue than a bug [07:02] I know that code is not written [07:02] DBO: oh, so it's on the radar? [07:03] mmm no comment [07:03] :) === al-maisan is now known as almaisan-away === almaisan-away is now known as al-maisan [07:09] kklimonda: yeah you can, with the registrydwords module option [07:09] Sarvatt: any drawbacks? [07:09] i can't remember the exact stanza for minimum all the time though.. [07:09] other then not being able to rev it up without restarting X :) [07:10] PowerMizerEnable=0x1 PerfLevelSrc=0x2233 PowerMizerDefault=0x3 -- thats always min power on battery, adaptive on ac [07:11] google, here I come [07:11] you can probably tweak it with nvidia-settings on the command line, i know xorg.conf stopped working around the 180 series [07:11] lemme boot up my nvidia and see how I did it [07:12] thanks [07:20] ok level 1 is max performance 3 is min power for each of the settings, PerfLevelSrc=0x2222 makes it fixed frequency on ac and battery, 2233 is fixed on battery adaptive on ac [07:20] need to add options nvidia NVreg_RegistryDwords="foo" in a /etc/modprobe.d/whatever.conf [07:21] might be options nvidia-current actually [07:21] looks like i'm not using any of the settings at the moment on that machine to reference, I always get thrown off when I go to disable it [07:22] options nvidia NVreg_RegistryDwords="PowerMizerEnable=0x1; PerfLevelSrc=0x2222; PowerMizerDefault=0x3" [07:22] that should work [07:22] vdpau needs more than the min frequency so you might not want to use it :) [07:23] I have never really had much luck with vdpau anyway [07:23] option nvidia or nvidia-current? [07:23] options* [07:24] ok, lets see what happens [07:25] oh! I was using xorg.conf, I guess it does work again! [07:26] i have Option "RegistryDwords" "PowerMizerEnable=0x1; PerfLevelSrc=0x2233; PowerMizerDefault=0x3; PowerMizerDefaultAC=0x2" in the device section of my xorg.conf [07:27] it only worked via a kernel module option from 180-195 from what I remember, but its working via xorg.conf in this 260 driver [07:27] hmm, it doesn't seem to work - at least nvidia-settings says that gpu revs up as always.. [07:27] is modprobe -r nvidia-current && modprobe nvidia-current enough? [07:27] what GPU do you have? [07:27] quadro 140m [07:27] if its before 8xxx it proably wont work [07:27] no, it's 8xxx [07:27] oh thats the same as I have and its working [07:28] hmm.. then I'm probably doing it wrong [07:28] its sitting at 169/100 in nvidia-settings no matter what I do with the xorg.conf [07:29] (on battery) [07:30] http://pastebin.com/kpb2Quzg [07:31] I should remove the last one? [07:31] yeah no need for the module option, xorg.conf works [07:31] sorry about that [07:31] i.e. PowerMizerDefaultAC=0x2 [07:32] try Option "RegistryDwords" "PowerMizerEnable=0x1; PerfLevelSrc=0x2222; PowerMizerDefault=0x3; PowerMizerDefaultAC=0x3" [07:32] that should be always lowest power mode [07:33] dont have an ac adapter handy to check it [07:33] nouveau will have power management for natty at least [07:34] yeah, it works now - everything is so slow :D [07:34] well, not everything but the fade-in/fade-out is [07:35] might want to use adaptive that only goes up to performance level 1, but heck if I know what that combination would be :) [07:35] yeah, that would probably be perfect.. Good to know that nvidia provides a detailed documentation for this feature ;) [07:35] i know you can do that though because i've had it before [07:37] 0x2 maybe? [07:38] *that* might be the thing you need to use the kernel module options to accomplish [07:38] hmm [07:38] actually, try this [07:38] Option "RegistryDwords" "PowerMizerEnable=0x1; PerfLevelSrc=0x3333; PowerMizerDefault=0x3; PowerMizerDefaultAC=0x2" [07:39] * Sarvatt thinks default sets the max level but isn't sure [07:39] http://tutanhamon.com.ua/technovodstvo/NVIDIA-UNIX-driver/ do you think it's up to date? [07:40] yeah those two default values pick the performance level to use, with 1 being the fastest level and 3 being the slowest level [07:41] so Option "RegistryDwords" "PowerMizerEnable=0x1; PerfLevelSrc=0x3333; PowerMizerDefault=0x2; PowerMizerDefaultAC=0x2" would do it [07:41] the options have been there for years and years so I'm sure it is [07:42] it seem to be stuck at the Level 1 [07:42] or not [07:43] it just took a while [07:43] it go own to 0 or up to 2? [07:43] only down to 0 [07:43] thanks [07:43] sweet, no problem [08:12] good morning [08:15] Aloha, didrocks! [08:15] hey RAOF, how are you? [08:15] Pretty good. [08:16] Oh, whoops. Just cliked on a link. I may be slightly less than responsive while KDE tries to launch hundreds of winebrowser instances for no particularly good reason. [08:18] :) [08:19] Yay! It's done. [08:19] Good morning [08:21] Hey pitti, didrocks. [08:21] Guten Morgen pitti [08:21] hey TheMuso [08:21] hey TheMuso, had a nice weekend? [08:21] bonjour didrocks [08:22] pitti: Certainly did thanks, a little warmer than I'd like for early sprig however. [08:22] early/mid spring even [08:22] lucky you - it has been raining here since Saturday, and will most of the week [08:23] And we've got strong wind here - although it's settled down a bit during the day. [09:01] hello [09:01] hey seb128 :) [09:02] lut didrocks [09:02] welcome back! how were your holidays? [09:02] how are you? [09:02] great, I managed to not work at all [09:02] I almost didn't turn the computer on [09:02] how did those 10 days go? [09:02] busy :) [09:03] but good I think! [09:03] a lot of RC bug closed and such [09:03] and a lot of uploads :) [09:03] happy to hear you didn't turn the computer on during your holidays :) [09:03] hey seb128 - welcome back! [09:04] hey mvo, thanks! [09:04] didrocks, great [09:04] didrocks, where do we stand quality and updates wise [09:04] desktop and une? [09:05] seb128: desktop has few remaining RC bugs, I didn't get anything worrying on the list [09:05] ok, great [09:05] I finished updating gnome 2.31.92 last week, all went well [09:05] how did the meeting go? [09:05] (apart from glib and gtk I let you that part :p) [09:05] brb coffee ready [09:05] alright [09:05] new GNOME today right? [09:05] 2.32? [09:05] right, finishing my emails first and then, jump on them :) [09:06] unity got a lot, a lot, a lot of bug fixes :) [09:06] it's really good those days! [09:06] still some few bugs which should be fixed as of today [09:06] a lot of unexpected extra work like libindicator soname bump… but overall all went well [09:07] and both meetings were smooth :) nothing particular [09:11] ok great [09:11] I'm catching up on emails, it's crazy [09:11] especially my bugmails box which has received lot of bug watch updates it seems [09:12] didrocks, seems there was some discussion about the signature [09:12] thanks for fixing it ;-) [09:12] did we just decide to go back to the hack I did without your changes to set it as default? === tkamppeter_ is now known as tkamppeter === asac_ is now known as asac [09:24] * mvo restarts to test mesa changes [09:34] seb128: yeah, quite hot discussion :) [09:34] good luck on email catching [09:34] seb128: yeah, it's your hack + some small fixes, but not the default addition [09:35] 91_git_fix_cache_issues.patch [09:35] ok [09:35] didrocks, ^ in gnome-menus [09:35] isn't the caching ubuntu specific? [09:35] seb128: no, this is the internal gnome-menus cache, not the file cache [09:35] ok [09:35] seb128: don't do like vuntz, he was blaming our cache :p [09:36] seb128: welcome back. Make sure to blame didrocks a lot! [09:36] and then, realize it was all his fault! [09:36] he didn't believe me :p [09:36] didrocks: of course I didn't believe you. As usual :-) [09:37] hey vuntz [09:50] on todays upgrade I get a lot warnings about a missing "/usr/lib/gdk-pixbuf-2.0/loaders.cache" - known issue? [09:50] sort of [09:51] that's nothing new in any case [09:51] it's there since gdk-pixbuf has its own source [09:51] I pinged you about that by then [09:51] hello [09:51] the issue is that the gdk-pixbuf postinst create the cache [09:51] but other postinsts can be run before this one and use the pixbufs [09:52] ie softwares installing icons [09:52] running gtk-update-icon-cache [09:52] mvo, ^ [09:52] baptistemm, lut [09:52] mvo, we would need a way to ensure that the gdx-pixbuf postinst is triggered before the others [09:53] but there is no way to do that right? [09:56] hm, ok [09:56] so its harmless? [09:56] not sure [09:56] I'm not sure it doesn't make the icon cache updates fail [09:56] hey seb128, welcome back! had a nice vacatin? [09:56] hey pitti! [09:56] excellent thanjks [09:56] * pitti hugs seb128 [09:56] * seb128 hugs pitti [09:56] how are things around? [09:57] pretty hectic last week, but settling down [09:57] I'm still catching up, anything I should know about? [09:57] oh? breakage? or just rc bug fixing sprint? [09:57] the latter [09:57] i386 retracer fixed, FYI (python module breakage again) [09:58] what is the shortcut in unity that was alt-f2 in most other systems? [09:59] if a "run app" feature is available, why wouldn't it be alt-f2? [10:00] I don't know [10:00] salut seb128 [10:03] I guess there is none [10:03] :/( [10:06] mvo: there is none [10:09] didrocks: bug #648625 fwiw, mutter crashing reliable for me on right click on any launcher item and clicking on a action in there [10:09] mvo: Bug 648625 on http://launchpad.net/bugs/648625 is private [10:10] mvo: waow, can't open the bug :) [10:10] mvo: ATI card? [10:10] didrocks: intel [10:11] didrocks: subscribed you [10:11] mvo: thanks [10:11] mvo: sounds like GL issue [10:12] possible [10:12] can you paste to the bug report your exact intel card? [10:13] there were a feature planned for a fallback if GL support in quicklist wasn't handle by the hw [10:13] unfortunately, that branch didn't land and may won't be in maverick [10:30] good morning everyone [10:31] chrisccoulson, hey [10:31] how are you? [10:31] hi seb128, you're back! [10:31] did you enjoy your vacation? [10:31] i'm good thanks [10:31] yes, I had a great time away from the computer [10:31] I'm happy to be back to it after 2 weeks though ;-)à [10:32] heh _;_ [10:32] oops [10:32] :-) [10:32] not sure what happened there ;) [10:32] i bet you have a lot of bug mail to get through this morning [10:32] emails yes, not only bugemails [10:32] I think I will just set the bug emails as read [10:33] I've read the ones assigned to me or the team [10:41] hey chrisccoulson [10:41] hi didrocks, how are you? [10:41] chrisccoulson: got a lot of rest this week-end, so fine, yes! ;) [10:42] and you? [10:42] yeah, i'm good thanks. i got quite a bit of rest too [10:43] didrocks: running netbook-session and switching back set the gconf key for "metacity-is-a-compoisition manager" [10:43] didrocks, hi, do you still plan to sponsor the tiny cairo fix? [10:44] didrocks: and it shufffled my custom button_layout (that one is minor) [10:44] didrocks: just fyi, not terrible [10:44] fta: not right now, but can maybe do later or hunt for another sponsor :) [10:44] mvo: hum? switching back to the desktop session change the default compositing manager? [10:48] didrocks: it appears to have set /apps/metacity/general/compositing_manager to true [10:48] interesting, that's maybe a side issue when mutter is started [10:48] * didrocks checks the value there [10:49] didrocks - any bugs you want me to fix today? :-) [10:49] i got the totem one fixed on friday [10:49] hum, it's false there, will try later on :) [10:49] chrisccoulson: yeah, I saw that! thanks a lot :) [10:50] chrisccoulson: hum, there is still those 2 ftbfs (the syncevo and the testsuite of xvfb). Not sure how important it is. Do you want to work on them? [10:50] the totem fix really rocks! ;) [10:50] yeah, sure [10:51] thanks chrisccoulson :) [10:51] chrisccoulson: I'll upload it with the new totem btw [10:53] mvo: can you log a bug and subscribe me? I'll have a test later on a fresh install :) [10:53] didrocks: sure [10:53] mvo: thanks :) [10:54] didrocks: what package? unity? [10:54] mvo: add a unity upstream task for tracking, but I would say ubuntu-netbook-default-settings for the package [10:56] chrisccoulson, having some free time today? [10:57] seb128 - yeah, i can look at some desktop bugs today [10:57] didrocks, how busy are you? [10:57] I'm trying to organize GNOME 2.32 updates ;-) [10:57] seb128: well, quite a bit right now :-) will be able to start updates on 30 minutes, I hope :) [10:58] ok [10:59] seb128: will do the gnome-games one [10:59] didrocks: bug #648662 [10:59] Launchpad bug 648662 in ubuntu-netbook-default-settings (Ubuntu) "after switching from netbook-session to desktop-session metacity is a composiotr (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/648662 [10:59] didrocks, ok [10:59] I will start with the gdk-pixbuf gtk glib [10:59] seb128: yeah, the one I even didn't try to touch :) [10:59] seb128: oh about gtk, one sec [10:59] yes? [11:00] pitti, do I need ack from somebody to sync telepathy-* updates from debian? [11:00] * didrocks looks at emails [11:00] we have the current rc tarballs and they rolled stable ones while I was away [11:00] seb128: bratche fixed bug #644730 and it's now commited upstream AFAIK, maybe should be interesting to pick it [11:00] Launchpad bug 644730 in gtk+2.0 (Ubuntu) "Remove mouse scrolling from GtkNotebook tabs (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/644730 [11:01] didrocks, ok thanks [11:01] bratsche* [11:01] yw :) [11:01] seb128: you were on the freeze exceptions delegates list for GNOME, but since you were on holiday and couldn't confirm, you got taken off, I think [11:01] seb128: but at this point I think it's better to have a bug and document it, and get a signoff [11:02] pitti, hum ok, one bug for each I guess? [11:02] seb128: tp-glib is depending on latest gir [11:02] seb128: or task, if the kind of changes are similar [11:02] and that blocks another tp package [11:02] didrocks, well quite some 2.32 tarballs will [11:02] I will just revert those commits [11:02] yeah, it's just a FYI stuff :) I didn't have the time to look at it and ken told me it wasn't trivial to revert [11:03] pitti, do you prefer a bug or a fakesync so it can be reviewed in the queue? [11:03] didrocks, ok [11:03] seb128: I don't mind much; fakesync makes it easier to review, though [11:03] didrocks, the gtk change seems controversial that will be for next cycle [11:03]