[02:36] hello [02:36] ive built a vanilla 3.2.10 kernel with perf enabled, since i wnated to use the performance counters for a personal project [02:37] with make-kpkg i get the different deb packages, however the perf tool consistently fails the symbol test [02:37] if i boot the same kernel with qemu and into a debian/ubuntu qemu image, the perf tool works fine [03:43] can the linux-tools packages get built through make-kpkg as well? === kengyu_ is now known as kengyu [06:27] '~$ nmon' > r prints: " Linux: Linux version 2.6.32-38-generic (buildd@zirconium)." Is »buildd« the username of a computer named »zirconium«? What kind of user is that? [08:15] * apw yawns ... another 80MB of updates [08:16] moin [08:17] ppisati, morning [08:18] morning... lots of fun to look forward to [08:22] smb, a shiney new unity ... i wonder what fun that brings ... any guess where the keys are today [08:22] * smb needs to install the shiny thing to know... [08:24] * apw watches it install ... excitemente [08:24] apw, At least nothing gets removed... I think I saw in the upload mails that this will be famous 5.6 we have all been waiting for... [08:25] yeah what did 5.6 contains that we needed ? was that the alt not being annoying one ? [08:26] * ppisati loves the smell of apw's positivty wrt to unity in the morning :) [08:27] apw, yes I think so (or generally the "even without multitouch you can press two keys at once, now stop giving me the help screen on super+something" [08:28] new flash it seems in the bargain [08:28] apw: hehe. Over on this side of the fence, work has ground to halt as everyone upgrades the galaxy s2 phones to Android 4.x and deals with the consequences :) [08:28] heheh ... that sounds fun [08:29] * ppisati -> reboot [08:29] - Dash - update Dash keyboard shortcuts so the 'CTRL + TAB' switches [08:29] between Lenses and 'TAB' by itself moves the focus between categories [08:29] new keybindings ... [08:30] - Alt+Tab default delay of 150ms is too long (LP: #888636) [08:31] - Numbers on Launcher icons sticking (LP: #942359) [08:31] so far, so good [08:31] * apw reboots to get the new shiney [08:32] * smb wonders whether you need fps experience to have the reflexes required to operate it... [08:37] unity2d: works here, still no overlay shortcut text and (thanks God) still no launcher on the second screen (with that bloody sticky edge) [08:38] * smb ready to do the same (reboot) [08:42] well i was able to reboot and login ... [08:43] glad to see they havent fixed the help [08:44] at least it does not come up when you press super+alt... [08:45] btw, did you see that the unity/UX/usability/$something_team is running a poll about multi screen setup + multi launcher setup? [08:45] ppisati, nope, where is that as i want to be chosing [08:45] you can even vote AGAINST sticky edge on the second screen and multi panel setup [08:45] hold on [08:46] nope [08:46] ppisati, though asking me to chose when i can only try one of them is a bit daft [08:46] actually the poll is "omg ubuntu" made for the unity team [08:46] big cockup... :P [08:47] oh so that won't have any input then [08:47] yep :( [08:48] * ppisati curls in sadness in a corner, you can't stop the sticky screeny edge, they will assimilate us... [08:48] heheh [08:48] Hm, still got the issue that clicking on an existing app in launcher moves the app and not the viewport... :/ [08:48] at least for some of them [08:48] smb, i still get launcher and help on pressing super-alt together if super is the first key that closes ... however short a time [08:56] apw, Yes, true. same here [08:57] * apw also notes the resize space round a window has stopped working so you have to hit the 1px again [08:58] alt+right drag? [08:58] apw, hm there seems to be some space to left and bottom at least [08:58] jk-, oh i know how to get round it, but i doubt its expected to have gone [08:59] smb, not for me [08:59] to the right as well if only that stupid hidden scrollbar would not be [09:00] top is only 1px... [09:00] not any side for me, i have 1px and 1px only everywhere [09:00] apw, the special apw settings ...? [09:00] perhaps so .. sigh [09:01] apw, Or maybe... is that on a external screen? [09:02] nope all of 'em [09:04] Just was wondering whether the issue rises because there is more than one screen, which I do not have [09:07] * apw doesn't really care ... i have it ... and so a bug i will file [09:08] Given the speed of updates one feels like the white rabbit... I am too late for updates... I am too late for updates... [09:47] /msg smb ping [09:47] ops [09:48] apw, For fun I did bug 953858 [09:48] Launchpad bug 953858 in unity "Selecting running app in launcher moves app not viewport" [Undecided,New] https://launchpad.net/bugs/953858 [09:49] oh thats oss as it does not do that smb` [09:49] (for me) === Kiall_ is now known as Kiall [09:50] heh .. that looks odd indeed [09:50] mention the bug on #ubuntu-unity [09:50] apw, Since you said you do not use multiple desktops that much, you would less likely. And it does not seem to happen with all applications [09:51] oh dear thats most odd, oh yeah and i would have to start using them to hit it [10:24] ppisati, in your precise/ti-omap4 rebase i see two commits i am not expecting ... right at the bottom of your delta ... [10:24] ppisati, UBUNTU: SAUCE: ata_piix: defer to the Hyper-V drivers by default [10:24] ppisati, UBUNTU: ubuntu: nx-emy - i387: NX emulation [10:25] ppisati, i am not expecting either of those to be in your delta [10:25] ppisati, if you concur i can just rebase them out of existance [10:38] apw: let me check [10:41] sudo aa-disable sbin.dhclient [10:46] apw: go ahead with those two commits, seems i rebased on master [10:46] apw: and these were there [10:46] ppisati, so drop them yes [10:53] flag [11:18] * apw takes that as a yes [11:19] apw: yes [12:14] ppisati, ok uploaded ... [12:16] ppisati, and i've prepared linux-meta ready for when its built [12:25] apw: cool [12:25] * ppisati fails to understand poll() and decides it's time for lunch() [12:51] cking, about ? your ecryptfs patches presumably are for arm too [12:52] apw, yep and yep [12:57] cking, applied and applied [12:57] ta [13:55] * apw lunches [14:15] apw: just fyi, I've rebased master-next to v3.2.10 which happened to also include the fix for CVE-2012-1146 you pushed [14:15] ogasawara: ** RESERVED ** This candidate has been reserved by an organization or individual that will use it when announcing a new security problem. When the candidate has been publicized, the details for this candidate will be provided. (http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2012-1146) [14:27] if i have a 2GB DIMM installed on a system and /proc/meminfo is only reporting 1653076 kB for MemTotal, what accounts for the lost megabytes? [14:43] * ogasawara back in 20 [14:59] ** [14:59] ** Ubuntu Kernel Team Meeting - Today @ 17:00 UTC - #ubuntu-meeting [14:59] ** [15:23] jsalisbury, thats in 2 hours from your announcement right [15:24] apw, correct. [15:25] apw, it's an hour later in the US now due to the recent daylight time change. [15:25] weeks of confusion have begun :) [15:25] jsalisbury, thought so ... just checking you were on the same page as me [15:33] bjf, uploading :( [15:34] smb, heh [15:38] jsalisbury, did I lose audio? [15:38] arges, I don't think so [15:42] apw: still under analysis, but i would say bugs 922906 and 924400 are the same [15:42] Launchpad bug 922906 in linux "BUG: unable to handle kernel NULL pointer dereference at 0000009c" [Medium,In progress] https://launchpad.net/bugs/922906 [15:42] Launchpad bug 924400 in linux "kernel NULL pointer dereference at 00000000000001f0" [High,Incomplete] https://launchpad.net/bugs/924400 [15:42] apw: (can't speak on mumble!!!) [15:43] apw: i've been looking at these bugs, and they seem to be a common issue w/ unmount and inotify [15:44] smb, whats your trick for fixing speaking on mumble [15:44] either restart [15:44] or change output from pulse to alsa and back [15:44] heh i'll try that one. actually, it looks like my push-to-speak keys are not working... [15:44] umount and inotify ... hmm ... that rings a bell, i wonder if there was someting committed to linus recently [15:45] henrix, ahh did you add a keyboard since mumble started, that can break PPT keys [15:45] henrix, That sometimes seems to be because mumble tries to play something but cannot [15:47] well, it's working now! [15:47] thanks [15:48] its basically a frigile POS ... [16:09] bjf, i see a small formatting error in 'eCryptfs: Copy up lower inode attrs after setting lower xattr' [16:09] bjf, and i'd like to rebuild tip to fix it (buglink is missing prefix) [16:10] apw, wfm [16:10] apw, master or master-next ? [16:10] this is master-next [16:10] apw, series? [16:10] master is 'tough its done' [16:10] apw, you can do anything to master-next :-) [16:10] bjf, i need to check lucid natty and oneiric [16:10] just you are pushing too :) [16:12] apw, we should be done now, i fixed a commit for colin this a.m. and am done [16:13] bjf, ahh i see now, its actually an upstream link so its as intended... === cmagina_ is now known as cmagina [16:16] cking, ok your patches for bug #926292 didn [16:16] Launchpad bug 926292 in linux "automake distdir.test fails because of an EPERM error" [Medium,Fix committed] https://launchpad.net/bugs/926292 [16:16] didn't have a proper buglink, most are already past the point of fixing, so you'll have to move them Fix Released manually ... [16:17] ack [16:22] * apw gets blinded by the sun ... gah [16:22] * ogasawara throws a snowball at apw [16:23] * apw gets snow blinded as well [16:26] * smb doubts it would hit apw at all (being melted before ;-)) === jussi01 is now known as jussi [16:29] * apw lobs an iceburg in smb's direction [16:31] * smb wonders how that castle looks like [16:31] (burg being castle in de) [16:32] heavy and inbound [16:33] ogasawara, do they even build test upstream stable ... [16:33] apw: apparently not [16:33] apw, sometimes it feels like "he" claims he does [16:33] so much for the claims of testing it [16:34] * jussi hides from the snow fight... [16:34] * apw gets out big digger and uses it to heave a 25 tonne snow drift onto jussi [16:34] apw, Question is what config is used (something like make alldontcareconfig) [16:34] allnoconfig perhaps [16:35] heh [16:35] if it wasn't so hard to consume they could let us autobuild the -rcs for stable [16:35] apw: be nice! :( [16:35] oh no, that might be seen as contributing and he cannot be seen letting us to do that [16:36] apw, The trigger may be hard. Though the queue is applicable when the review starts [16:37] git quiltimport --patches= [16:38] if it was a git repo like it is once it is released then the current autobuilder could be pointed at it [16:38] Right, that would be too simple [16:39] /home/apw/COD/linux/drivers/mfd/wm8994-core.c:260: error: 'WM1811_JACKDET_MODE_MASK' undeclared (first use in this function) [16:39] that said the mainline builds built the tag and barfed instantly as above [16:39] broooooooonie [16:40] smb if they just pushed the -rc1 things he does, we could send him the reports ... sigh [16:41] apw, Problem seems to be that he seems to be unable to handle branches. Or he is just a je... One or the other === yofel_ is now known as yofel [17:17] cking: about bug #926292, i've tagged it verification-needed-oneiric. although the commit is queued for stable, i guess you'll want to test it, right? [17:17] Launchpad bug 926292 in linux "automake distdir.test fails because of an EPERM error" [Medium,Fix committed] https://launchpad.net/bugs/926292 [17:18] henrix, yep, I'd like to hand verify it [17:18] cking: ok, cool. just to confirm that [17:19] I'm just preparing a machine to test it as we speak, so good timing ;-) [17:19] heh [17:19] no hurry! :) === jsalisbury changed the topic of #ubuntu-kernel to: Home: https://wiki.ubuntu.com/Kernel/ || Ubuntu Kernel Team Meeting - Tues Mar 20th, 2012 - 17:00 UTC || If you have a question just ask, and do wait around for an answer! [17:57] hi. does anyone know what the status of ext4+'discard' (ssd TRIM) support is? [17:58] is this stable enough to be considered a best standard practice for use with SSDs at this point? [18:00] henrix, verification complete :-) [18:01] cking: cool, thanks! [18:15] ogasawara, you will be pleased to know there is now a 3.2.11 which builds [18:15] apw: nice [18:16] apw: I'll get 'er rebased [18:16] ogasawara, the mainline builder just announced it, and it seems built ok on all arches [18:37] Hmm... it appears brcmsmac stopped detecting my wifi card some time after 3.2.0-12 [18:37] 3.2.0-12 works, 3.2.0-18 doesn't [18:58] * jsalisbury is updating bios. Back in a bit. [18:59] '~$ nmon' > r prints: " Linux: Linux version 2.6.32-38-generic (buildd@zirconium)." Is »buildd« the username of a computer named »zirconium«? What kind of user is that? [19:02] bullgard4, buildd is the user under which builds in ubuntu are [19:02] built on the buildds [19:18] apw: Thank you very much for your help. [19:53] hi, i'm playing with kernel development on ubuntu and i'm wondering what's the best way to do incremental compiles [19:54] at the moment i'm compiling my kernel according to http://blog.avirtualhome.com/2011/10/28/how-to-compile-a-new-ubuntu-11-10-oneiric-kernel/ and https://wiki.ubuntu.com/Kernel/BuildYourOwnKernel [19:55] my problem with that is that after i change something "skipabi=true skipmodule=true fakeroot debian/rules binary-mbp" doesn't include my changes if i don't do "fakeroot debian/rules clean" [19:55] before [19:55] but cleaning and doing a full build on every change takes a lot of time [19:55] distcc is your friend [19:57] so it's not possible to do incremental builds? [19:57] i fear even with distcc/ccache it'll take ages [20:04] i prefer to do full builds.. i think of it as safer [20:05] i think if you do the plain old make [20:05] you can do incremetal builds [20:05] if you're working on a module you can use the regular linux build stuff to test it [20:05] ^^ [20:06] i'm doing that for a part of it but that doesn't work for everything [20:40] * cking hammers ecryptfs with a kernel build and calls it a day, /me --> EOD === tsimpson_ is now known as tsimpson [23:44] ah-, you can remove the build stamp in debian stamps to get an incremental build