=== henrix_ is now known as henrix
=== henrix_ is now known as henrix
=== RAOF_ is now known as RAOF
[05:30] <ox8085_> unable to patch my wireless card bcm4312 for packet injection
=== smb` is now known as smb
=== fmasi_afk is now known as fmasi
=== jussi01 is now known as jussi
[08:00] <jpds> When will raring pieces be available in http://archive.ubuntu.com/ubuntu/dists/precise-updates/main/installer-amd64/current/images/ ?
[08:37] <apw> RAOF, that mir staging PPA is all out of sync and eating my machine today
[08:44] <apw> RAOF, there are no bees
[08:48] <xnox> apw: Debian is removing a few obscure filesystems from their kernel config http://lists.debian.org/debian-boot/2013/06/msg00107.html and to this effect various parts of d-i are also removed to no longer offer unsupported filesystems.
[08:49] <xnox> apw: are our kernels going to follow suite, or shall some of the d-i bits continue to be maintained to continue supporting those/some fs mentioned above?
[08:50] <RAOF> apw: That's unfortunate! I've been pouring bees upstream!@
[08:51] <apw> xnox, i think this we are aware of this, i guess we'll discuss with foundations and see that they plan for d-i in light of that, though as cjw is heavily involved i am sure he is aware
[08:52] <apw> xnox, looking at the list, there is no standard install which would use those filesystem types for install, i don't see us removing them unless they get broken, wich if they are not maintained i suspect they would
[08:53] <xnox> apw: ok. in debian they went on a removal spree of (partman-$foo) and we haven't merged d-i stack in a while. both colin and i busy with other ubuntu-touch things =)
[08:53] <dupondje> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1203211 => Somebody can check this out ?
[08:53] <ubot2`> Ubuntu bug 1203211 in linux (Ubuntu) "Modprobe doesn't recognize any parameters on 3.10.0-4" [Undecided,Confirmed]
[08:53] <apw> xnox, yeah, i think i'll see you sometime next week so we can discuss it in person :)
[08:55] <xnox> apw: well, unless you are planning on kidnapping me, you will only see colin/steve in person sometime next week. =)
[08:55]  * xnox guesses i somehow don't know something =)
[09:12] <apw> dupondje, thanks ... am look at that indeed
[09:23] <dupondje> thx apw, as it seems quite broken :)
[09:23] <apw> dupondje, indeed quite
[10:09] <psivaa> apw: I have reported bug 1203694 not realising the existence of the above bug. Assume both are related. Pls feel free to mark it dupe if you think its appt.
[10:09] <ubot2`> Launchpad bug 1203694 in linux (Ubuntu) "virtual-host saucy server smoke tests fail because kvm_intel and kvm_amd are not present in lsmod output" [Undecided,Confirmed] https://launchpad.net/bugs/1203694
[10:09] <apw> psivaa, yeah prolly the same, thanks
[10:09] <psivaa> yw
[10:12] <apw> dupondje, smb, could you test the kernels here: http://people.canonical.com/~apw/lp1203211-saucy/
[10:12] <smb> apw, yup
[10:13] <apw> and let me know if that works for one
[10:20] <smb> apw, seems to wfm. kvm_intel with nested=1 is loaded
[10:21] <apw> smb, thanks ... sigh
[10:26] <infinity> apw: What did you change?  Do I want to know?
[10:32] <smb> infinity, These are not the changes you are looking for...
[10:32] <apw> infinity, reverted a patch we carry which is an optimisation for init, whcih i suspect (if these all become working) has been miss forward ported to this release
[10:33] <apw> we don't need it per-se any how so i would propose to revert it and upload without and think about it at my leisure
[10:33] <infinity> apw: Curious.  But also, WTF did -3 work? :P
[10:33] <infinity> apw: Just dumb luck, I guess?
[10:33] <apw> some of that i suspect yes
[10:33] <apw> now i have found a blatent unititialised pointer in lightdm which
[10:33] <apw> seeming should stop it working for anyone ever
[10:34] <apw> and yet has only started hitting my machine today randomly
[10:34] <apw> now to try and find the frigging source branches for it
[10:34] <apw> oh god ... it has a packaging only branch ... i hate those even more
[10:35] <infinity> Heh.  Just pull-lp-source
[10:35] <apw> yeah i am going to, it is clear seb128 has done that in the past as well :)
[10:37] <apw> now i could just edit this shit, and debcommit it and it will make me a quilt patch right ?
[10:38] <apw> arse, not that, its erm, why are names so hard
[10:41] <infinity> apw: dpkg-source --commit
[10:41] <apw> now, why doesn't debcommit do that
[10:41] <infinity> apw: debcommit is for magically committing to packaging branches (it even sorts out if you're in git, bzr, svn, etc)
[11:01] <apw> infinity, is it reasonable to just upload this lightdm change
[11:05] <dupondje> apw: can only test this evening if still needed :)
[11:05] <apw> dupondje, ok
[11:05] <apw> i guess smb's and my confirmation will have to do
[11:15]  * ppisati -> out for lunch
=== ara is now known as Guest46239
[12:13] <infinity> apw: Just uploading it is probably fine, but if the packaging branch is up to date, you should commit there too.
[12:28]  * henrix -> lunch
[12:39] <zequence> infinity: If I were to update linux-lowlatency for the development release now and again, how would the uploading work? Do you have something similar to SRU for the development release as well? I think I read something about you changing routines for saucy, or maybe I was just imagining?
[12:42] <infinity> zequence: kernels to straight to the archive for devel releases, no interim PPA business.
[12:43] <infinity> zequence: For the master kernel, they're using tracking bugs, but no need for derivatives.
[12:50] <apw> zequence, infinity, i suspect once i've done a couple to amke sure its working ok, you could do them, and just upload them to your normal ppa
[12:51] <apw> then we can copy them out in the normal way
[12:52] <apw> infinity, ok this revert is looking like it is doing the business on the module param issue
[12:52] <apw> so i'll get that preped and uploaded shortly
[12:52] <apw> rtg, yo
[12:52] <rtg> apw, dude
[12:52] <apw> rtg, i am just prepping a 3.10.0-4 upload to quickly fix a module load iss
[12:53] <apw> rtg, but i see there is a .2 in stable and now wonder if we should shove that in
[12:53] <rtg> apw, wanna pick up a stable ?
[12:53] <apw> thoughts?
[12:53] <apw> i'll take that as a yes :)
[12:54] <zequence> apw: infinity: ack
[12:54] <rtg> apw, yeah, I'll get the -rc2 in shape in the meantime. gonna start prodding the dkms folks to be ready for a switch by te end of the week.
[12:54] <rtg> apw, don't forget the tracking bug :)
[12:54] <apw> rtg, ok ... revert the 'all parameters' patch en-toto for now, we don't require it in fact and it is causing these issues right now; i'll fix it up
[12:54] <apw> and resubmit once i have figured out the issue
[13:00] <rtg> apw, 'UBUNTU: SAUCE: (no-up) add option to hand off all kernel parameters to init' is the patch causing problems ?
[13:00] <apw> yes .... revert that, upstart will cope (and be marginally slower)
[13:01] <apw> but something in the forward porting (or lack thereof) has broken it all of a sudden in -4
[13:01] <rtg> apw, ack, I'll just drop it from unstable for now
[13:01] <apw> though i suspect -3 working is complete luck there
=== medberry is now known as med_
[13:27] <infinity> apw: Wait, is that the patch responsible for my upstart getting passed things like --debug and --no-sessions from the cmdline?
[13:27] <infinity> apw: Losing that will be a bit irritating.
[13:27] <apw> yes it is
[13:28] <apw> oh, no it isn't
[13:28] <apw> those are unknown which are passed on, its the known ones like quiet that are the issue
[13:28] <apw> which upstart looks for in /proc/cmdline anyhow, its all a bit stupid
[13:29] <apw> i want to rework it anyhow, cause i think when we are doing it, the init needs to know, so we need to say like "ALL" at the front or something to say they are indeed all there
[13:29] <apw> so i'll revisit it, and fix it right this time
=== fmasi is now known as fmasi_afk
=== kamal__ is now known as kamal
[13:52] <dupondje> apw: just home, will test
[13:58] <dupondje> apw: ok, it works :)
[14:22] <ppisati> brb
=== rtg is now known as rtg-afk
[14:43] <apw> dupondje, nice thanks
[14:46] <ppisati> on S/generic (but only highbank related so far) i need to run a script/function at postinst time
[14:46] <ppisati> is there a generic way to hook into the installation process, or shall i just modify debian/control-scripts/postinst and be done?
[14:47] <ppisati> what i need to do is to change some constants in /boot/boot.script, since the new kernel in S needs more space (and thus we need to move initrd further below in memory)
[14:55] <caribou_> ivoks: I have one Maas bug that I haven't had time to work on
=== caribou_ is now known as Caribou
[14:55] <Caribou> ivoks: would you have some cycles to look at it ?
[14:55] <Caribou> ivoks: pretty simple issue
[14:56] <Caribou> crap ,sorry for  the noise guys, wrong room
=== fmasi_afk is now known as fmasi
[15:25] <apw> Caribou, np
[15:54]  * apw wanders to another (hopefully cooler) location ... bah for heatwaves
=== ogasawara__ is now known as ogasawara
[16:10]  * apw discovers that this "other location" has cold cider, who would have guessed
[16:11] <apw> rtg-afk, ok kernel uploaded
[16:11] <apw> ogasawara, i uploaded a -5 kernel today to fix a module parameter issue which was affecting testing pretty badly, fyi
[16:12] <ogasawara> apw: ack
=== rtg-afk is now known as rtg
[16:35] <rtg> apw, how will reverting that parameters patch affect upstart in Lucid (with the LTS kernel) ?
[16:36] <rtg> not Lucid, but precise
[16:36] <apw> rtg, shouldn't affect it in theory, though i intend on fixing it tommorrow
[16:36] <apw> it needed a cleanup anyhow
[16:36] <rtg> apw, ack
[16:37] <apw> rtg, basically upstart has no way to know the kernel can do the feature, so it has to check, so it should find out what it needs by other means
[16:37] <lool> Hi, Xorg doesn't start anymore with 3.10.0-4 but it does with 3.10.0-3; is this a known bug?
[16:37] <apw> and we need a way to allow it not to do that, so it needs ot know it can not do that
[16:37] <lool> this is on intel graphics on a thinkpad X301
[16:37] <apw> lool, i have been some issues with lightdm and unity-greeter with later kernels
[16:37] <apw> lool, so far the bugs have not been the kernel though
[16:38] <apw> what is the last thing you see
[16:38] <lool> apw: If I try to launch X, it says no device detected
[16:38] <rtg> apw, I also sent an email to you and the k-team list about a couple of other patches that I dropped. 
[16:38] <lool> [    16.760] (EE) No devices detected.
[16:38] <lool> is what it says in the Xorg log
[16:38] <lool> it's possibly related to some Xorg side changes that are needed for the new kernel
[16:39] <apw> lool do yo uhave any i915 kernle parameters
[16:39] <dupondje> lool: prolly some modules not loaded ? :)
[16:39] <apw> (on your command line)
[16:39] <lool> /proc/cmdline: BOOT_IMAGE=/@/boot/vmlinuz-3.10.0-3-generic root=UUID=b13b1eeb-f9f4-404f-bb88-3c87466c9c14 ro rootflags=subvol=@ quiet splash vt.handoff=7
[16:39] <apw> not that then, that one may be new, though i would strongly suggest trying the .5 i just uploaded to confirm
[16:40] <lool> apw: in -proposed then?
[16:40] <apw> lool building, i'll find you a binary
[16:41] <lool> I discovered a bug in the failsafe handling while I was there
[16:41] <lool> apw: thanks
[16:43] <dupondje> https://launchpad.net/ubuntu/+source/linux/3.10.0-5.14/+build/4816287/+files/linux-image-3.10.0-5-generic_3.10.0-5.14_amd64.deb
[16:43] <dupondje> :P
[16:43] <dupondje> https://launchpad.net/ubuntu/+source/linux/3.10.0-5.14/+build/4816287/+files/linux-image-extra-3.10.0-5-generic_3.10.0-5.14_amd64.deb
[16:51] <apw> lool, if those don't do it: http://people.canonical.com/~apw/master-next-saucy/
[16:52] <rtg> my Sandy Bridge server sure doesn't get along with 3.11-rc2
[16:53] <apw> rtg, my efi obox doens't either
[16:53] <apw> it is not a happy release
[16:54] <rtg> apw, so far its not
[16:54] <apw> fingers crossed for -rc3
[16:55] <rtg> apw, yeah. in the meantime I guess I'll go work some Nexus kernel issues
[16:59] <lool> apw: good news!
[17:00] <lool> apw: it's indeed fixed with 3.10.0-5-generic
[17:00] <lool> apw: thanks for putting it together
[17:00] <apw> lool, great, that is already uploaded, so should be in the images tomorrow if we are lucky
=== fmasi is now known as fmasi_afk
=== emma is now known as em
=== ogra_` is now known as ogra_
[17:35]  * rtg -> lunch
[17:42] <cooloney> ogasawara: your video about Ubuntu for phone is great!!!
=== ayan_ is now known as ayan
=== Nafallo_ is now known as Nafallo
[18:37] <hallyn> jdstrand: we should probably discuss the kvm.ko problem here
[18:38] <rtg> infinity, were you considering making use of saucy-udebs-generic in Precise ? That patch is causing a build error in the LTS backport (though easily fixable)
[18:44] <jdstrand> hallyn: ok. still, I want to reboot into 3.10.0-5 :)
[18:44]  * jdstrand is almost done downloading
[18:45] <rsalveti> rtg: thanks for rebase & upload
[18:45] <rtg> rsalveti, I figured that was the most thorough approach
[18:46] <rsalveti> yeah, that's fine as well
[18:48] <kamal> bjf, jsalisbury, mjg59: just fyi, I have released 3.8.13.5-stable which includes the fixes "Modify UEFI anti-bricking code" and "x86/efi: Fix dummy variable buffer allocation" (plus some supporting framework)
[18:48] <mjg59> kamal: Thanks!
[18:48] <mjg59> Appreciate it
[18:49] <kamal> mjg59, likewise!
[18:54] <jdstrand> hallyn: ok, 3.10.0-5 fixes the kvm_intel nested=1 issue
[18:54] <jdstrand> let's see if it also fixes the hard lockup
[18:57] <jdstrand> that would be a 'no'
[18:58] <jdstrand> erf, ubuntu-bug won't let me report it ye
[18:58] <jdstrand> yet
[18:59] <apw> jdstrand, hard lockup doing what
[19:00] <jdstrand> apw: the 3.10 kernel, kvm and my x201s (i7) do not get along
[19:00] <apw> so kvm, ok, i have seen a lock there too
[19:00] <jdstrand> apw: qemu-kvm crashes hard unless I downgrade to a raring kernel
[19:00] <apw> when you have filed a bug, would you ping me the number so i can look at it
[19:00] <jdstrand> sure thing
[19:01]  * jdstrand wonders if there is a way to waive the official Ubuntu package check
[19:02] <jdstrand> ah, maybe I just have to add -proposed back to sources.list
[19:07] <jdstrand> ah, no it is in NEW and arm is still building
[19:07]  * jdstrand reboots in 3.8 for now
=== JanC_ is now known as JanC
[20:01]  * rtg -> EOD
=== RAOF_ is now known as RAOF