/srv/irclogs.ubuntu.com/2012/09/26/#ubuntu-kernel.txt

RedoubtDoes anyone know at what level a tap-to-click is handled?00:49
RedoubtWhat decides?00:49
RAOFRedoubt: The X driver.00:49
RedoubtRAOF: Thank you-- is it purely time-based?00:50
RAOFI believe so; you'd want to check the finite state automaton in xf86-input-synaptics00:50
RedoubtRAOF: Okay, I appreciate the help!00:54
jgriffithsmb`: ping01:05
RedoubtAre funkeys (volume up, volume down, etc) handled in the userspace nowadays?01:21
RAOFYes01:23
RedoubtRAOF: I'm having some trouble figuring out where, exactly. Any pointers?01:25
RAOFgnome-settings-daemon does the handling in GNOME/Unity; KDE will have an equivalent.01:25
RedoubtRAOF: So if I have a funkey that adjusts my keyboard backlight brightness, does gnome-settings-daemon turn that into... an ACPI signal back to my keyboard?01:27
RAOFThat depends :)01:27
RedoubtHaha01:27
RedoubtHow do I learn that stuff?01:27
RAOFSo, posibilities include: your firmware doesn't generate an event at all, and just twiddles backlight brightness itself. Or, generates an event that gets sent to gnome-settings-daemon and then g-s-d doesn't know what to do with it.01:28
RedoubtAh, I didn't know that first one was an option. Is there a way to confirm one way or another?01:30
RedoubtThe backlight _does_ change, just not in an ideal manner. I was hoping I could take a look at the logic01:32
=== doko_ is now known as doko
Davieysmb: Are we blocked on bug 1023755?07:19
ubot2Launchpad bug 1023755 in linux "Precise kernel locks up while dd to /dev/mapper files > 1Gb (was: Unable to delete volume)" [Undecided,Confirmed] https://launchpad.net/bugs/102375507:19
smbDaviey, In the sense that I cannot see in a low-level experiment that the dd I use hangs. I get an IO error and the snapshot is invalidated as I would expect. I would at least need someone affected getting back in the report and saying whether the reproduction happens on them07:43
smbDaviey, There is the other part that maybe a slightly bigger snapshot size may circumvent the issue, too.07:44
smbUnfortunately there is no pre-existing simple way to convert the snapshot storage/meta-data area into a linear volume and erase that later.07:46
smb(at least I found none)07:46
=== henrix_ is now known as henrix
Davieysmb: yeah, i think there is a genuine bug there07:55
ppisatibrb08:04
* apw yawns08:24
cooloneymoring smb, ppisati and apw 08:25
apwmorning cooloney 08:25
smbMorning all08:25
* smb is lazy08:25
interciao a tutti10:37
inter!list10:37
ubot2$who: No warez here! This is not a file sharing channel (or network); read the channel topic. If you're looking for information about me, type « /msg ubottu !bot ». If you're looking for a channel, see « /msg ubottu !alis ».10:37
henrixapw: are you around?11:04
henrixguess not...11:07
apwhenrix, yep11:18
henrixapw: i've a question11:23
smbI think the answer is it was not really expected but is possible due to them being compatible... :-P11:27
* henrix -> lunch11:35
=== Amoz_ is now known as Amoz
Pirx-Danfordcheers, found out I need an older kernel after upgrading to 12.04, luckily the old kernel I used still boots, but is there a maintained 2.6 line for 12.04?11:59
Pirx-Danfordokay so I found http://kernel.ubuntu.com/~kernel-ppa/mainline/ - is there any other way?12:22
* henrix -> brb12:41
ppisatiogra_: i left my board compiling since yesterday and with vm.min_free=32K i can confirm that12:48
ppisatilp 69037012:48
ubot2Launchpad bug 690370 in linux-ti-omap4 "Strange out of memory on pandaboard" [Medium,Triaged] https://launchpad.net/bugs/69037012:48
ppisatiogra_: is fixed12:48
ogra_cool12:48
ppisatiogra_: now, casper had that fix but12:48
ppisati1) does it carry it to the installed system?12:48
ppisati2) what someone upgrades from [N/O] to P?12:49
ppisati*what if12:49
ppisatiis the fix carried over?12:49
ppisatiin Q is fixed at kernel level, so no prob there12:49
ppisatiogra_: ^12:49
ogra_it is a file that jasper creates in /etc/sysctl.d ... 12:50
ogra_that file isnt owned by any package, so upgrades should still have it12:50
ogra_(i.e. there is nothing removin that file)12:50
ppisatiwhat if someone comes from M?12:50
ppisatiand was N the first were this fix showed up?12:51
ppisati*first release12:51
ogra_coming from M you have an armel install and *wnat* to reinstall anyway 12:51
ogra_*want12:51
ppisatiwell, you can reach P12:51
ogra_theoretically, yeah12:51
ogra_err12:52
ogra_no, you cant 12:52
ppisatibecasue in that lp, mdz said he had a 11.04 system12:52
ogra_there was some kernel guy who decided to only build panda kernels for armhf ;)12:52
ppisatiand he still experience that bug12:52
ppisatithat wasn't me :)12:52
ogra_heh12:52
ogra_it was a good decision ... butu upgraders will be stuck on their old kernel12:53
ppisatiuhm ok12:53
ogra_why isnt that a duplicate for the oter bug ? 12:53
ogra_ppisati, i think we should just duplicate it ... natty will be dead in a few weeks 12:58
ogra_and i really wouldnt suggest to anyone to stay on armel, while it is still there in the archive, its totally untested since two reeases12:59
ppisatiogra_: added a comment for mdz in that lp bug12:59
ogra_(and in quantal it even switched its default arch)13:00
pocoporco1hello@all13:00
pocoporco1can somebody help me? i'm searching for an implementation of the wlan management frames in the kernel :(13:01
pocoporco1or... could somebody tell me, if i'm searching right here in ubuntu-KERNEL? i'm not sure where i have to search13:02
rtgherton, re: https://lists.ubuntu.com/archives/kernel-team/2012-September/022121.html - did you notice that I did cherrypick the upstream patch (for Quantal) and not the one in the email ?13:06
hertonrtg, no, didn't notice you cherry-picked it already the right patch13:07
rtgherton, depending on who submits the request, I almost always verify first.13:08
hertonrtg, sent the ack for the cherry-pick on precise13:10
rtgapw, https://lists.ubuntu.com/archives/kernel-team/2012-September/022108.html13:10
apwrtg seen it, will get to thinking on shortly13:11
=== yofel_ is now known as yofel
* rtg thinks ogasawara must be having desktop issues13:34
ogasawaraI'm here, although I'm on my laptop as I am having issues with my desktop (ie I'm pretty sure my hard drive is dieing).13:35
rtgogasawara, oh, not the refurb Vostro ?13:36
rtgor its the desktop the auguring in ?13:37
ogasawarait's my desktop that is cratering, I'm on my adamo13:37
rtgogasawara, fortunately you live in the silicon forest where hard drives are readily available13:38
ogasawaraindeed13:42
amitkrtg: what, ogasawara moved to Fry's?!13:46
rtgamitk, she lives just down the street from Fry's IIRC13:47
rtgamitk, no, thats a different place.13:47
tjaaltonrtg: hey, have you tried building the ubuntu-r kernel yet? I'm thinking of using it as a base for debugging intel drm crack..13:47
rtgtjaalton, it buildds for armhf, amd64, and i38613:48
rtgbuilds*13:48
tjaaltonrtg: cool, enough for me13:48
rtgusing the quantal toolchain13:48
tjaaltonyeah13:48
rtgtjaalton, its doesn't work worth a dang on my AMD Fusion(?) GPU13:49
tjaaltonoh, heh13:49
apwhenrix, ok a buildd admin did indeed flip a machine to lpia and back14:01
henrixapw: ah, ok. cool! thanks for the update14:01
infinitySorry if I confused people. :P14:03
henrixheh, no problem. i was a little bit confused indeed :)14:04
apwinfinity, its good for us14:07
infinityhenrix: For the record, I'll be doing the usual copy-from-ppa and mangle overrides dances today for all the things you guys have been uploading this week.14:17
henrixinfinity: ack, thanks.14:17
ppisatiis the a way to specify to dpkg-buildpackage to build just one flavour? e.g. omap14:35
ppisati*there14:35
rtgppisati, --target=14:35
ppisatirtg: thanks14:36
=== ogasawara_ is now known as ogasawara
* rtg knocks the number of linux-firmware bugs down to 2.14:47
ogra_does that ean we need to file new ones ?14:48
ogra_*mean14:48
ogra_so you dont run out of work :)14:48
rtgogra_, oh please, I _live_ for firmware bugs :)14:48
ogra_hahaha14:49
rtgskaet, I think you should drop the  rls-mgr-q-tracking tag on bug #1006427 as ppc is a community supported arch and should not be allowed to hold up the release.15:00
ubot2Launchpad bug 1006427 in linux "(PowerBookG4) Live image won't boot, stuck on "stdin: Not a typewriter"" [High,Incomplete] https://launchpad.net/bugs/100642715:00
skaetrtg,  that's a stale tag,  not being used anymore.    Need to do a cruft cleanup.15:01
rtgskaet, what are you using to compile http://reports.qa.ubuntu.com/reports/rls-mgr/rls-q-tracking-bug-tasks.html ?15:02
skaetvalid tags are:  rls-q-incoming,   rls-q-notfixing.   Otherwise status is tracked by whether its targetted to a series or not.15:02
skaetrtg,  bdmurray looks for those bugs targetted to a series,   as decided at last UDS,   there was an email from slangasek about earlier in cycle,  I'll go get the blueprint...15:03
rtgskaet, don't bother, its not that important to me (cause I won't remember for more  then a day at the most)15:03
rtgskaet, mostly I just wanted bug 1006427 off my radar15:04
ubot2Launchpad bug 1006427 in linux "(PowerBookG4) Live image won't boot, stuck on "stdin: Not a typewriter"" [High,Incomplete] https://launchpad.net/bugs/100642715:04
rtgppisati, tag, you're it: bug #104585515:05
ubot2Launchpad bug 1045855 in linux-ti-omap4 "usb keyboard doesn't work during installation of ubuntu-server on panda" [High,Triaged] https://launchpad.net/bugs/104585515:05
skaetrtg,  then add the tag rls-q-notfixing to it,  and that should bump it from the list15:06
rtgjsalisbury, recent responses on https://bugs.launchpad.net/bugs/104502715:07
ubot2Launchpad bug 1045027 in linux "[regression] iPXE kills kvm with KVM: entry failed, hardware error 0x80000021" [Critical,In progress]15:07
rtgskaet, done15:07
ppisatirtg: weird15:07
rtgppisati, might be bogus, it just needs some love15:08
ppisatirtg: ack15:08
ppisatiogra_: how is the server installation handled these days? i remember the installer back then via serial console (and thus you needed serial access), how about now?15:11
rtgppisati, bug #746137 is also in your bailiwick.15:11
ubot2Launchpad bug 746137 in linux-ti-omap4 "Page allocation failure on Pandaboard and Beagle XM" [High,Confirmed] https://launchpad.net/bugs/74613715:11
ogra_ppisati, theoretically you should be able to isntall via framebuffer, practically bug 1028664 might get in your way15:12
ubot2Launchpad bug 1028664 in debian-installer "keyboard does not work on quantal-server-armhf+omap4.img" [Critical,Fix released] https://launchpad.net/bugs/102866415:12
ogra_ppisati, there is a preEnv.txt-serial you can copy over preEnv.txt on the first partition of the SD before installing, that will get you the serial setup15:14
ppisatiogra_: no, i want to test lp 102866415:17
ubot2Launchpad bug 1028664 in debian-installer "keyboard does not work on quantal-server-armhf+omap4.img" [Critical,Fix released] https://launchpad.net/bugs/102866415:17
ppisatiogra_: but i remembered that it wasn't possible to do a server installation without serial console15:17
ppisatiogra_: let's see this framebuffer thing15:18
ogra_you should just see the d-i screen right after booting15:18
ogra_it comes up really fast 15:18
ppisatibesides, someone should really tell what's the rationale in requiring a monitor for a server installation... but that's another story...15:18
ppisatiogra_: on the screen, right?15:19
ogra_yes15:20
ogra_ppisati, we used to default to serial and i agree with you that servers should be serial ...15:20
ogra_but with us forcing unity on everyone people asked for a way to install xubuntu or lubuntu desktops ... 15:21
ogra_for which server or netinst is your best chouce15:21
ogra_and tehse people want to use framebuffer15:21
ppisatii disagree here15:21
ppisatiif i install a server (a dell blade&c) i'll never attach an lcd to it perhaps15:22
ogra_oh. sure15:22
ppisatiso if you install an ubuntu server, just use the serial console15:22
rtgarges, has jjohansen's AA bug fix been exposed publicly yet ?15:22
ogra_but server serves the purpose of the now dropped alternate image now as well15:22
* ppisati shrugs15:23
argesrtg, what do you mean? the LP #?15:23
rtgarges, the patch15:23
ppisatiogra_: anyway, i'm downloading the daily server to check it out15:23
argesrtg, https://bugs.launchpad.net/bugs/105607815:23
ubot2Launchpad bug 1056078 in linux "freeing apparmor profiles cause irq stack overflow" [Critical,Confirmed]15:23
argesrtg, yea its there15:23
ogra_ppisati, but we should probably consider defaulting to serial and make framebuffer the optional choiuce (i.e. flip the defaults)15:23
rtgarges, cool, I'll pluck that into Q15:24
ppisatiogra_: wait, let me just check this usb thing15:24
ppisatiogra_: so i don't have to flip the new defaults15:24
ogra_default is framebuffer15:24
ogra_you should be able to but as is15:25
ogra_*boot15:25
ppisatidd-ing right now...15:26
rtgarges, is bug #979498 related to bug #1056078 ?15:30
ubot2Launchpad bug 979498 in linux "Critical: Natty kernel hangs in xlog_grant_log_space" [High,In progress] https://launchpad.net/bugs/97949815:30
ubot2Launchpad bug 1056078 in linux "freeing apparmor profiles cause irq stack overflow" [Critical,Confirmed] https://launchpad.net/bugs/105607815:30
argesrtg, nope15:30
argesrtg, that one will be a bit more tricky... its a series of 13 or so patches that are in upstream nfs now (i think)15:31
argeserr15:31
argesxfs15:31
rtgok15:31
rtgarges, I'm gonna change the patch description from jjohansen to ':115:31
rtgUBUNTU SAUCE: apparmor: fix stack overflow15:31
ppisatiogra_: screen is black and installer started on serial console15:32
ppisatiogra_: http://cdimage.ubuntu.com/ubuntu-server/daily/current/quantal-server-armhf+omap4.img15:32
ogra_it doesnt for me here 15:32
ogra_you didnt copy anything around on the image/SD ?15:32
ppisatiogra_: nope15:33
ppisatiogra_: wget and then dd15:33
ogra_well, thats tecnically not possible 15:33
* ppisati is technically impossible!15:33
ogra_d-i wont start on serial if there isnt a console= line15:33
ogra_and there isnt one15:33
ogra_check preEnv.txt please15:33
ppisatiok15:33
ppisatiogra_: bootargs=vram=40M mem=456M@0x80000000 mem=512M@0xA0000000   cdrom-detect/try-usb=true15:34
ppisatipreEnv.txt15:34
ogra_yeah, nothing in there that could even remotely cause d-i to start on serial15:35
ogra_(and in fact it doesnt here)15:35
ppisatiAH!15:35
ppisatithis time it started on the framebuffer!15:36
ppisatii'm not kidding15:36
* ogra_ scratches head15:36
ppisatiand indeed usb is dead15:36
ogra_i suspect we're missing HID stuff in the udebs15:37
rtgbjf, another regression test for your suite: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1056078/comments/215:37
ubot2Launchpad bug 1056078 in linux "freeing apparmor profiles cause irq stack overflow" [Critical,Confirmed]15:37
ogra_but didnt get to that big yet15:37
ogra_*bug15:37
bjfrtg, reading15:37
ppisatii want to wipe the sd and try again now15:37
bjfrtg, we _do_ run the apparmor qrt tests, i'd like to see this added to those. however, i'm happy to run this separately also.15:38
rtgbjf, It doesn't matter to me where it lands15:38
rtgvi ~/tmp/jj15:39
bjfrtg, i understand15:39
ppisatiogra_: dd it again and this time it went immediately to framebuffer15:45
ppisatiogra_: *bah*15:45
ogra_phew, ok, lets call that cosmic rays thenb15:46
ppisatialways on me?15:46
ogra_looking at the d-i detection code there is no reason why you did see the serial console  15:46
ppisatii didn't take a screenshot15:46
ppisatinet time i'll do15:46
ppisatianyhow15:46
ogra_from the black screen ? 15:46
ppisatinope, from the terminal15:47
ogra_i belive you it was black when the serial console was up ;)15:47
ogra_ah, k15:47
ppisatiogra_: during the desktop install, the usb keyboard is working (i know it because i use it for input), so why it's not working here?15:58
ogra_ppisati, d-i can only use modules that are provided as a udeb 15:58
ogra_while desktop just has all modules available that are in the kernel package15:59
ppisatiok15:59
ogra_i would suspect we are missing a hid udeb here 15:59
rtgsforshee, could you have a look at bug #117414061 ? It kind of looks like an interrupt issue.16:03
jsalisburybjf, should I add 1056969 to the hotlist, or are you looking at it?16:03
rtgsforshee, make that bug #105696916:04
ubot2Launchpad bug 1056969 in linux "USB accessories stop working" [High,Confirmed] https://launchpad.net/bugs/105696916:04
bjfjsalisbury: should be on the list16:04
jsalisburybjf, ack16:04
sforsheertg, ack16:04
penguin42are there older nightly deb's anywhere from Linus's git tree - I'm looking for stuff pre 3.6.-rc116:06
rtgpenguin42, http://kernel.ubuntu.com/~kernel-ppa/mainline/16:07
penguin42rtg: That only has the main tags, rather than Linus's tree16:08
rtgpenguin42, I'm not sure what you're telling me. those kernels are built from vanilla sources out of Linus tree.16:08
penguin42rtg: But only at specific tags; I'm looking for the nightlys that occurred prior to the 3.6.-rc1 tag16:09
rtgpenguin42, oh, that we don't have16:09
penguin42rtg: But (after 3.5 release)16:09
penguin42rtg: Yeh the nightly build only has about a weeks saved16:10
rtghenrix, pushed precise16:21
henrixrtg: ack, thanks16:21
jsalisburyrtg, bug 1041883 bisected down to commit d3a67dbe37d32ba10da0de2f14151d606856e5d2 as the cause of the regression.  16:28
ubot2Launchpad bug 1041883 in linux "Recent patch to asus-wmi module makes system unbootable" [High,In progress] https://launchpad.net/bugs/104188316:28
jsalisburyrtg, it may be related to CONFIG_X86_X32=y16:28
jsalisburyrtg, I think that was your first hunch :-)16:29
rtgjsalisbury, indeed. apw ^^16:29
ckingapw, any notion why we enabled that back in 3.5.0-11.11?16:35
apwcking, yeah it was something we are interested in for next cycle maybe16:36
apwcking, so we enabled it to see if it was a problem and to allow testing16:36
apwcking, it is the 32bit pointer 64bit registers mode thing16:36
ckingso lets turn it off and explore that for Q+1 then..16:36
* cking recalls that now16:36
apwcking, if jsalisbury proves it is the issue i am all for that16:36
tjaaltonrtg: ubuntu-r kernel oopses immediately after loading.. oh well16:37
* ppisati -> gym16:42
penguin42jsalisbury: on the iPXE crash, I'm curious - does it trigger for you (it's trivial to try if you have KVM installed)16:53
jsalisburypenguin42, I'm not able to reproduce the bug, but there are two people in that bug that can.16:54
penguin42jsalisbury: Yes, I'm one of them - hmm ok, I'd assumed it was host cpu independent16:55
jsalisburypenguin42, ahh, ok :-)16:55
jsalisburypenguin42, It's strange that the bug does not exist on v3.6-rc1, since the same commit that causes the bug in 3.5.3 is applied there as well.  An email has been sent upstream requesting info as well.16:57
penguin42jsalisbury: Yeh, there are lots of other changes in 3.6 though, so I assume another one fixed it somewhere or it was an incomplete backport of a patch from 3.616:58
* cking ~~> food16:59
jsalisburypenguin42, yeah.  I'll have to look closer 16:59
apwcommit bacef661acdb634170a8faddbc1cf28e8f8b9eee17:02
apwAuthor: Jan Beulich <JBeulich@suse.com>17:02
apwDate:   Fri May 25 16:20:31 2012 +010017:02
apw    x86-64/efi: Use EFI to deal with platform wall clock17:02
jgriffithsmb: ping17:29
* rtg -> lunch17:34
rtgsmb, so what is your call on bug #1038055 ? do you still want CONFIG_DRM_CIRRUS_QEMU disabled ?18:27
ubot2Launchpad bug 1038055 in linux "graphics fail to initialise correctly, in kvm with cirrus graphics (after LUKS install)" [High,Confirmed] https://launchpad.net/bugs/103805518:27
=== Amoz is now known as Staffan_Star
=== Staffan_Star is now known as Amz
=== Amz is now known as Amoz
=== henrix is now known as henrix_
hertonrtg, that apparmor sru for lucid will not apply, wrong patch?19:35
rtgherton, perhaps, lemme check19:37
rtgherton, hmm, looks right to me. try git://kernel.ubuntu.com/rtg/ubuntu-lucid.git master-next19:40
hertonrtg, right, ignore me... now I noticed I was in the wrong branch after reviewing lts-backport oneiric today...19:41
rtgherton, np19:42
rtgherton, arges: I'm getting a lot of noise from the AA reproducer script in https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1056078/comments/2 . Is that what you would expect ?19:54
ubot2Launchpad bug 1056078 in linux "freeing apparmor profiles cause irq stack overflow" [Undecided,In progress]19:54
hertonrtg, on a precise vm I get lots of "apparmor_parser: cannot use or update cache..."19:59
rtgherton, thats what I got as well, but dmesg indicates that tehre are profiles being deleted. is that the same in yours ?20:00
hertonrtg, yes, some profile_replace messages20:01
rtgherton, ok. I guess thats good.20:02
* rtg -> EOD20:29
phillwHi, is kernel-3.6.0-0.rc6.git0.2.fc18 (or later) going to land for 12.10?20:45
bjfphillw, we are going with the 3.5 kernel20:58
phillwbjf: okies, I'll have to go figure out another route to fix bug 1043518 and https://bugzilla.redhat.com/show_bug.cgi?id=85730021:03
ubot2Launchpad bug 1043518 in ubiquity "live cd is unusable due to video degradation with the splash boot option enabled" [High,Confirmed] https://launchpad.net/bugs/104351821:03
ubot2phillw: Error: Could not parse XML returned by bugzilla.redhat.com: HTTP Error 404: Not Found (https://bugzilla.redhat.com/xml.cgi?id=857300)21:03
phillwhopefully I can nosplasj put in for the PPC architecture until 3.6 arrives for 13.0421:04
phillw*nosplash*21:04
=== henrix_ is now known as henrix
infinityphillw: You should be isolating the git commit and asking people to pull it into our kernel, not to upgrade wholesale to 3.6.021:50
phillwinfinity: my apologies, when it comes to this level of detail, I'm way out of my depth... sorry!21:51
phillwso, now that I know the correct question.. could the patch be applied to our current 3.5.x kernel. From my (very) limited reading of the bug report, it did not cause regressions and fixed the issue.21:52
phillws/reading/understanding22:02
=== henrix is now known as henrix_
=== zz_Kiall is now known as Kiall
=== Kiall is now known as zz_Kiall

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