[00:16] <thomi> is anyone able to help me diagnose why inotify seems to not work on a live CD boot, as described here? https://bugs.launchpad.net/ubuntu/+source/linux-meta/+bug/1015348
[00:16] <ubot2> Ubuntu bug 1015348 in linux-meta "openssh-server package does not start sshd on a live CD boot" [Undecided,New]
[00:29] <thomi> Oh, I see it's already been reported: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/882147
[00:29] <ubot2> Ubuntu bug 882147 in linux "overlayfs does not implement inotify interfaces correctly" [High,Triaged]
[05:21] <scientes> can i get on that backport track with kernel for precise NOW?
[05:21] <scientes> There are various reasons I want a 3.4-based kernel
[06:44] <tjaalton> scientes: https://launchpad.net/~ubuntu-x-swat/+archive/q-lts-backport
[06:44] <tjaalton> scientes: what reasons?
=== smb` is now known as smb
[07:32] <smb> janimo, You mail look into git for you answer: eb66750db88b59b28ac91c627dd0984b5508f5a6 (at least for maint-startnewrelease). And it was not me putting it there. ;)
[07:33] <smb>  s/mail/may/
[07:33] <smb> Morning, btw
[08:01] <ppisati> moin
=== yofel_ is now known as yofel
[08:59] <smb> tjaalton, Hey, I just noted on your submissions to stable that they only went there (and yourself). Usually it is better to cc at least the maintainer(s) and the author (depending on Greg's mood (if it is for one of his stable trees) the sub-system mailing list too.
[09:00] <smb> Reason is that maintainers not always (if at all) look at stable but are normally required to ack or nak
[09:01] <tjaalton> smb: oh
[09:02] <tjaalton> it's not mentioned on Documentation/stable_kernel_rules.txt though
[09:03] <smb> tjaalton, I try to put everyone who signed-off and the author (if not equal to one s-o-b) on cc. Maybe not, but I could give you examples of me being told so. ;) And it just gives you a much better chance at being seen/noticed. :)
[09:05] <tjaalton> smb: heh, ok
[10:18] <apw> bjf, henrix, looks like this might be a regression-update: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1015480
[10:19] <ubot2> Ubuntu bug 1015480 in linux "webcam not found (no device found)" [Undecided,New]
[10:19] <henrix> apw: checking
[10:19] <apw> cking, hey you looked at bug #904261, did the fix for that make it into the main kernel ?
[10:19] <ubot2> Launchpad bug 904261 in linux "Lenovo U300s does not suspend" [Medium,In progress] https://launchpad.net/bugs/904261
[10:20] <henrix> apw: yeah, it looks like. thanks
[10:20] <cking> apw, not yet, I'm waiting for a specific patch to land
[10:20] <apw> cking, so i think you supplied a custom kernel, and that has now been superceeded
[10:21] <dileks> hi
[10:21] <cking> apw, yep, that's true
[10:21] <apw> cking, if you have the patch around it might be nice to spin a replacement
[10:21] <dileks> is a wubi-installation a somehow elegant way to "bypass" secure-boot problem :-)?
[10:22] <cking> apw, I could do that, but I could be doing this until 3.6 at this rate
[10:22] <apw> cking, if this one fixes it then lets just SAUCE it in
[10:22] <cking> apw, and there is a user-space hackaround for it
[10:22] <cking> apw, the patch is really ugly and won't get upstreamed, it was just a sanity check to see why things were locking up
[10:23] <apw> the userspace work around is reported (by a normal) to have stopped working
[10:23] <apw> http://marc.info/?l=linux-pm&m=133573212512122&w=2 <-- this one ?
[10:23] <cking> yep
[10:24] <apw> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/904261/comments/32
[10:24] <ubot2> Ubuntu bug 904261 in linux "Lenovo U300s does not suspend" [Medium,In progress]
[10:24] <cking> it's more complex that that though
[10:24] <apw> i cannot tell easily as the machine was installed by someone else, but i am suspicious he had your kernel
[10:24] <apw> and its gone now, and he is not using the work around, hmmm
[10:25] <cking> apw, fundamentally it is partly to do with a BIOS issue
[10:25] <apw> cking, so the kernel you did was jsut a bodge and not something we can put in master
[10:26] <cking> apw, it was a bodge and I'm waiting for the real fix to land - which is going to be somewhat different
[10:27] <cking> ..as this fix may cause other issues on other machines if I read it correctly
[10:28] <apw> cking, i wonder if we need a genric mechanism for these
[10:28] <apw> like ubuntu-bodge-lpNNNN
[10:28] <apw> which explicitly has to be turned on
[10:28] <apw> ie deliberatly not letting it be DMI enabled etc, to cover the space before an upstream fix lands
[10:29] <cking> makes sense - or I write in the bug report "THIS IS A TEST KERNEL AND IS NOT SUPPORTED!!"
[10:29] <apw> you can write that, and what happens is someone installs it for the VP of something, and then when it gets replace, i get involved via rick
[10:30] <apw> its a general problem though, that we have no way to put 'crap needed for broken machines' in the tree safely right no
[10:30] <apw> now
[10:44] <ppisati> cooloney: still around?
[11:28] <janimo> smb, ok. I noticed it because I have my @ubuntu.com address the default for gpg and debemail :). I am not sure why only Canonical people should run this script, it makes perfect sense even for local tweaks and rebuilds to keep the packaging cleaner
[11:30] <smb> janimo, I guess it is valid to ask whether it may be changed to be configurable there. Just as an answer to "do you know why" it is "yes, because" ;)
[11:34] <janimo> smb, :)
[11:34] <janimo> probably a config option would make sense
[11:35] <janimo> if we are to be nitpicky canonical@foo.com also matches currently :)
[11:35] <smb> janimo, Or probably an environment variable to override the requirement (to avoid having an additional config file)
[11:35] <janimo> ok, I thought there was a conffile already but if not envvar is better
[11:48] <smb> janimo, Ok, I did a quick stab at it. Now you can at least override the check (I admit the envvar name is long)
[11:49] <janimo> smb, nice thanks ! btw can you add me to (janimo/jani.monoses@canonical.com) to the mainstscript.cfg list? The docs said I should add myself but could not push
[11:54] <smb> janimo, I could but I think this list is more intended as a default with a bit more limitation on the scope
[11:55] <janimo> smb, ok great than. I thought it was used by some scripts to match commit authors to irc names for some reason
[11:55] <janimo> maybe to ping  by bot or something
[11:57] <smb> janimo, Maybe you are even talking about a different config file... Which one was that exactly?
[11:57] <smb> There is a exmaple-kteam.rc which would go to ~/.kteam.rc but that is solely used to make maint-modify-patch wrork
[11:58] <janimo> smb, yes, that one
[11:59] <janimo> ah ok then. no worries :)
[11:59] <janimo> I had it copied to ~/.maintscripts.cfg based on what I read somewhere (wiki most likely)
[11:59] <smb> Ah, ok. That is really only to avoid typing email addresses for adding signed-off-by and acked and such. You can easily create your own fitting your needs
[12:26] <Defusal> the latest cpuset package says "ImportError: No module named cpuset.main" when running "cset", does anyone know how i can fix this?
[13:30] <cooloney> ppisati: yeah, i'm here, just felt a little bit head ache and took a reset
[13:33] <Defusal> does anyone know how to make a default group for all system-wide processes (starting with init) that are not part of another cgroup in cgconfig.conf?
[13:36] <ppisati> cooloney: no prob, i was actually looking for ming, so i wanted to know if you have seen him around on irc
[13:39] <cooloney> ppisati: np, he is in #hwe channel
[13:39] <ppisati> cooloney: ack
[13:40] <cooloney> ppisati: but don't know whether he is online
[13:40] <ppisati> cooloney: i'll ping him tomorrow then
[13:40] <ppisati> cooloney: we hit the same mmc card regression on oamp3
[13:40] <ppisati> cooloney: and i was wondering if he saw a panic during about voltage_scaling
[13:40] <ppisati> *during boot
[13:41] <cooloney> ppisati: oh, i saw you mentioned that yesterday. is that related to filesystem or just voltage thing
[13:42] <ppisati> cooloney: one is about mmc card (sever file system corruption) both of us hit it, we bisected and came down to the same problem
[13:42] <ppisati> cooloney: but in the mean time, they introduced another regression
[13:43] <ppisati> cooloney: and it _seems_ (i'm rebuilding openocd to jtag it) to be voltage scaling required
[13:43] <cooloney> ppisati: oh, what a mess. is that in mainline kernel? unforunately, i don't have omap3 board
[13:43] <ppisati> cooloney: actually that's something to do with ubuntu config (omap2plus_defconfig doesn't exhibit it)
[13:43] <ppisati> no prob
[13:43] <cooloney> ppisati: hmmm. got it.
[13:44] <ppisati> but i wanted to know if he saw the secont one too since we are working on mailine both for omap3
[14:08] <tlei> ppisati, ping
[14:08] <ppisati> tlei: hey
[14:10] <tlei> ppisati, I am ming, and bryan said you are looking for me
[14:10] <ppisati> tlei: yep, besides the mmc regression, didn't you hit a kernel hang during boot?
[14:10] <ppisati> tlei: last kernel message complains about voltage_scaling
[14:11] <tlei> ppisati, what is log?
[14:11] <ppisati> tlei: something like this - http://paste.ubuntu.com/1049450/
[14:11] <ppisati> tlei: to rreproduce it, grab mainline (3.5rcx), use the ubuntu config file, compile and try to boot it
[14:12] <ppisati> tlei: usually around the second reboot, it hangs (something not de/initialized properly seems)
[14:12] <tlei> ppetraki, looks no such problem on beagle-xm revB with 3.5-rc2
[14:13] <ppisati> tlei: are you using ubuntu config file? doesn't show up with omap2plus_defconfig
[14:14] <tlei> ppisati, oh, I seldom tried 2nd reboot since there is a crash triggered by reboot, and I am a bit lazy to apply one patch to fix it
[14:14] <ppisati> tlei: i usually reset in that case :)
[14:15] <tlei> ppetraki, no, I seldom tried ubuntu kernel on beagle-xm, and it is very slow with my low speed mmc card
[14:15] <ppisati> tlei: ack, no prob
[14:16] <tlei> tlei, ok, I will try it, but I remembered one reboot crash problem not fixed, so can you reboot successfully without any out-of-tree patch? 
[14:16] <tlei> tlei, just on beagle, and panda is not affected
[14:17] <tlei> ppisati,  ok, I will try it, but I remembered one reboot crash problem not fixed, so can you reboot successfully without any out-of-tree patch? 
[14:17] <ppisati> tlei: nope, when it crashes at reboot i simply reset the aobrd
[14:17] <tlei> ppisati,  just on beagle, and panda is not affected
[14:17] <ppisati> tlei: beaglexm yes
[14:18] <tlei> ppisati, if so, I am sure I have not meet the kernel hang problem with the reset button
[14:18] <ppisati> tlei: you can easily grab Q/master-next, revert the autocmd12 commit and build
[14:18] <tlei> ppisati, maybe I need to try ubuntu config, instead of the omap plus config
[14:19] <ppisati> tlei: first boot ok, from the second one onward, solid hangs during boot
[14:19] <ppisati> tlei: 100% reproducible
[14:19] <tlei> ppisati, this line may be a clue:  voltdm_scale: No voltage scale API registered for vdd_mpu_iva
[14:23] <ppisati> tlei: yep, that's why i thought it's voltage scaling related
[14:27] <tlei> ppisati, looks I didn't enable voltdm_scale
[14:27] <ppisati> tlei: in your config?
[14:27] <tlei> ppisati, I will try to test it tomorrow
[14:27] <ppisati> tlei: no prob, it's night over there
[14:27] <tlei> ppisati, yes, I guess
[14:27] <ppisati> tlei: enjoy the... night? :)
=== skaet_ is now known as skaet
[15:07]  * ogasawara back in 15
[16:35]  * ppisati -> gym
[16:56]  * cking gives up with RTC_IRQP_SET and calls it EOD
[17:05] <scientes> tjaalton, this bug mainly https://bugs.launchpad.net/ubuntu/+source/linux/+bug/955665 However I am also trying to use the new DRM_UDL driver, but havn't had much success yet
[17:05] <ubot2> Ubuntu bug 955665 in linux ""udlfb: device_create_bin_file failed -22" boot hang with displaylink graphics" [Medium,Triaged]
[17:05] <scientes> and i noticed that the mainline builds that are marked "quantal" don't work with precise (networking doesn't come up for example)
[17:09] <apw> ogasawara, i had forgotten about that rebase bug finder ... heh
[17:09] <ogasawara> apw: I love it
[17:16] <apw> ogasawara, it is most surrpising just how many there are
[17:17] <ogasawara> apw: it did seem we had more than usual
[17:28]  * apw wanders out to do some errands, have fun
[17:28] <scientes> any previews of the signed uefi secure-boot kernels?
[17:28] <scientes> and any writeup on what features they don't have?
[17:28] <scientes> i didn't think you guys had any way to sign modules
=== rtg is now known as rtg-lunch
=== bjf is now known as bjf[afk]
=== rtg-lunch is now known as rtg
[18:42] <adam_g> other than changelog and the specific bug reports they resolve, where do ubuntu specific patches / sauce get tracked?
[19:09] <ogasawara> adam_g: If you want a current view, you'd want to look at the git tree.  All Ubuntu specific patches are prefixed with "UBUNTU:" and the sauce patches are prefixed with "UBUNTU: SAUCE:".  We also have a slightly dated view in our delta review spec.  eg.  https://wiki.ubuntu.com/KernelTeam/Specs/QKernelDeltaReview.
[19:24] <adam_g> ogasawara: thanks
[20:39] <scientes> somwhere i can track a 3.4 mainline?
[20:40] <infinity> ogasawara: Oh, your kernels should be getting published in quantal around nowish, BTW.  Sorry for the delay.  Headless chickening.
[20:40] <scientes> (for precise) the mainline builds seem like they are only against quantal
[20:40] <scientes> and i tried one and it broke my precise system
[20:41] <infinity> scientes: quantal kernels probably shouldn't break precise systems... Define "broke"?
[20:41] <scientes> infinity, no networking, no usb
[20:41] <scientes> graphics works fine
[20:42] <scientes> with the 3.4.3 quantal, the 3.4.0-precise works fine
[20:42] <infinity> Curious.
[20:42] <infinity> I really can't think of why that would be.
[20:42] <infinity> Except potentially 3.4.3 just hating you.
[20:42] <scientes> ant the lts 3.5 backport works with networking and usb
[20:42] <scientes> although it doesn't work with my touchscreen, but that is likely upstream thing
[20:45] <jwi> scientes: you installed the -extra package?
[20:45] <scientes> no
[20:45] <infinity> Or used the metapackage...
[20:45] <scientes> just the raw linux-image-fooo-amd.deb
[20:45] <infinity> Ahh.
[20:45] <infinity> That would be why.
[20:46] <infinity> A bunch of modules were broken out.  The metapackage handles that for you.
[20:46] <scientes> ahhhhhhhhhh
[20:46] <scientes> ahh i need the -extra i see
[20:47] <scientes> should have checked that
[20:48] <scientes> i couldn't find the ppa to do these automatically: http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.4.3-quantal/
[21:23] <ogasawara> scientes: did you install both the quantal linux-image and linux-image-extra's packages?  You need both.
[21:23] <scientes> yes i did
[21:23] <scientes> (now)
[21:26] <ogasawara> infinity: thanks!  /me leaves for vacation