=== DarkPlayer_ is now known as DarkPlayer === kentb is now known as kentb-afk [06:24] moin === smb` is now known as smb [07:15] ppisati, morning [07:15] smb: ciao [07:21] * ppisati goes for the monday morning dist-upgrade... [07:23] * smb waits to hear ppisati 's level of cursing [07:23] can't be, just 267pkgs to upgrade, what could go wrong? :) [07:28] Not so much what *could* but what *will* :) [07:51] * apw wonders if ppisati survived [07:51] welcome back smb [07:51] apw, Morning [07:52] morning [08:03] yep, i'm alive [08:04] from the album "Imaginations from the Other Side", Blind Guardian [08:08] ppisati, :) === ikonia_ is now known as ikonia [09:03] brb === Guest8464 is now known as fmasi === fmasi is now known as fmasi_afk === fmasi_afk is now known as fmasi === davmor2_ is now known as davmor2 === ghostcube_ is now known as ghostcube [11:16] https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/1217196 [11:17] this bug has been assigned to gnome-shell, but the original submitter claims the problem appeared with a kernel update [11:18] i just happen to be having the same problem, ie.. i seem to have a setup which can reproduce the above issue [11:19] what i want to try to do is downgrade to an older saucy kernel, to see if the problem goes away [11:20] snadge, sounds worth testing indeed [11:20] theres no real reason that contrl-alt-t shouldn't open a terminal in metacity.. it does in unity.. i assumed it was a change in something relating to that [11:21] but i've noticed that sometimes my keypresses arn't being registered.. particularly space [11:21] i just assumed my keyboard was becoming faulty / less responsive [11:21] it seems highly unlikely that input handling is completely borked and noone has noticed, as kernel side that is completely orthoganal to the upper windowing system [11:22] ie it work be broke for everyone using any windowing at all, and the bug implies that downgrading from gnome-shell has fixed it for the reported no ? [11:23] im using gnome in fallback mode with metacity.. im assuming the bug is related [11:23] it started happening after an update a few weeks to a month ago [11:23] snadge, hard to say, but the test is easy to confirm if it is the kernel, boot an older kernel [11:23] and if it goes away there is a relationship, it could still not be a bug in the kernel of course, but indicative [11:27] where can i find the older saucy kernels? [11:29] snadge, well you should first check if you have them in your /boot already as you can have more than one kernel installed at the same time anyhow [11:29] yeah i've cleaned most of them out [11:29] snadge, otherwise they are in the launchpad librarian which has everything ever published: https://launchpad.net/ubuntu/+source/linux/+publishinghistory [11:30] that link has all the versions for the linunx package [11:30] look for lines for 'Saucy' 'release' [11:32] the version i want has been deleted [11:33] it being 'deleted' just means it is not in the archive any more, because it was no longer referenced by a live archive pocket [11:33] the link on the rigth still has the launchpad librarian copies [11:33] yeah im looking at that now.. udeb files and lots of module files.. what a headache [11:34] this is new to me [11:34] i have to download all of them ? [11:34] you can ignore _all_ .udebs [11:34] you need just the common headers (an _all package) and the _.debs you need for your architecture and flavour [11:40] ok im going to try 3.9.0-7 [11:54] ok that one just freezes and wont input anything [11:54] have to hard reset it === fmasi is now known as fmasi_afk [12:49] ok so if i dont install linux-image-extra.. my pc crashes ;) [13:52] snadge, heh that is half of the kernel after all [13:56] metacity shortcut problem doesnt appear to be kernel related anyway === kentb-afk is now known as kentb === fmasi_afk is now known as fmasi === kentb is now known as kentb-afk === fmasi is now known as fmasi_on_the_pho [15:03] snadge, good to know [15:03] kamal, i am assuming you are going to be applying your own patch for kteam-toosl === kentb-afk is now known as kentb === fmasi_on_the_pho is now known as fmasi [16:08] so, trying to compile saucy kernel for mako... complains that it can't find prebuilts/gcc/linux-x86/arm/arm-eabi-4.6/bin/arm-eabi-gcc-4.7 [16:08] so i probably have to install the toolchain there? do most people just symlink to a toolchain or something? [16:12] rtg, ^^^ can you help kdub a bit ? they are trying to get kernel patches in for Mir [16:12] or apw [16:13] kdub, I think I was cross-compiling that kernel in Raring. lemme check [16:13] yes, i'm at the point where i'm waiting on a fence from the sync driver forever, want to dig a bit into what is going on [16:13] thangs rtg [16:14] yeah i thought you needed the appropriate cross compiler in your chroot [16:15] yep, requires gcc-4.7-arm-linux-gnueabi in Raring [16:15] i think we installed these two: gcc-4.7-arm-linux-gnueabihf g++-4.7-arm-linux-gnueabihf [16:15] kdub, ^ [16:15] or something didn't we ? [16:16] apw, right, gcc-4.7-arm-linux-gnueabihf [16:16] I think that gets everything you need [16:16] okay, giving it a try [16:17] i think the one without hf gives you the el compilers, i _think_ [16:17] kdub, echo "dpkg-buildpackage -d -B -aarmhf -us -uc"|schroot -c raring-amd64 [16:17] but install the whole damn lot :) [16:17] apw, yes I'll apply the kteam-tools patch -- if I can get it past the whitespace police! ;-) thanks for the ack, man. [16:17] * apw looks at kamal sideways ... [16:18] but if you look at me sideways, then you'll see that my tabs and spaces don't line up! [16:18] heh indeed [16:18] apw, impertinent pup today, isn't he ? [16:18] always, rtg, always [16:19] kdub, or if you have a specific patch you want built we might be able to build it for you [16:19] depending on how desperate timewise you are [16:19] * ogra_ guesses more desparate than we all can imagine [16:20] eh, i don't have a specific patch, just trying to get to the root of the problem still [16:21] there's a moderately big delta between what qcom's current drivers put out and cm-10.1 kernel [16:21] so i'll probably have to come up with some patch (once I understand a bit more) that fixes things [16:21] good job we have so long before the release [16:22] apw, touch works a lot different QA wise [16:22] we dont have freezes and milestones [16:23] every daily image we release is handled like a milstone instead [16:31] cking, have you got a bit to have a look at CONFIG_NO_HZ_FULL_ALL as well ? [16:32] rtg, yep, I can spin a kernel and test that tomorrow morning [16:32] cking, IIRC that was the config option that was giving us the most problems [16:33] rtg, yep, I think I recall that too [16:33] let me back up a bit... the way i'm used to building kernels for android is in the android source tree, the ubuntu touch kernels aren't built like that though, right? [16:34] kdub, nope, they are debian packaged [16:35] ah :) i'll pursue that route then [16:36] rtg, when you considering this nwe upload, for after thursday ? [16:37] rtg, i want to switch out some hyperv bits for the next one [16:38] apw, dunno, anytime its convenient for you [16:38] rtg, well we are fozded now through 'final beta' anyhow, so i'll get those patches sorted today [16:38] frozded [16:39] apw, frozded -> frozen ? [16:39] yeah, past-made-up-participle of freeze [16:39] as on erm 9pm my time today, except it seems to have happened already [16:39] as of [16:40] apw, k-freeze isn't until next week [16:40] rtg, indeed, but we are in archive freeze now so anything you upload won't be getting out of -proposed till friday [16:41] apw, grumble. so much for development velocity. [16:41] yeah we can upload, but we won't get any images, but i'll get my patches switched round next [16:41] so i won't be holding u up [16:44] rtg, ok those patches slammed in [17:08] it seems there's a new embedded linux distro for set top box&c: [17:08] http://store.steampowered.com/livingroom/SteamOS/ [17:11] ppisati, keep us informed as you find out more, i assume it will be linux under the hood after all the owrk they have done with us [17:11] apw: yes, it's linux [17:12] apw: "SteamOS combines the rock-solid architecture of Linux with a gaming experience built for the big screen." [17:12] * ppisati wonders on top of which distro they built this thing [17:16] ppisati, yeah i wonder indeed [17:27] * ogra_ bets its another ubuntu-core breed [17:27] like the OS in the tesla and in the google cars [17:43] rtg, yeah I am not in favor of disabling it wholesale [17:44] rtg, but we need to turn off CONFIG_HVC_DCC coz that causes hangs as well ... [17:44] ppisati, ^ is cced on that discussion [17:44] manjo, lets run all those changes through ppisati [17:44] yep ack [17:44] he is cced on all those conversations [17:47] manjo: ??? [17:47] manjo: email subj? [17:47] ppisati, email galore for you :) [17:47] ppisati, "Disable broken CPU offloading armhf" [17:47] ppisati, sub gotomeeting [17:48] and sub what rtg just mentioned [17:48] so 2 issues [17:49] ppisati, would you like to listen in on those calls ? I know you have better things to do ... so your call on that [17:49] ppisati, todays call starts in 15mts [17:49] ppisati, tomorrow onwards calls at 10:30cst [17:50] manjo, prolly ought to take this to #hwe [17:50] yeah [17:50] rtg: i never received that email, let me grep my inbox [17:51] ppetraki, switching channels [17:51] ppisati, ^ oops [17:51] too many nics with pp [17:51] ppisati, join hwe [17:52] manjo: i'm already there [17:52] manjo: ah no === fmasi is now known as fmasi_afk [17:52] manjo: i'm on hyperscale [18:56] rtg, for building that arm kernel, what variant should debootstrap be? buildd isn't working for some reason === fmasi_afk is now known as fmasi [19:02] kdub, you need to install the kernel build-depends in there obviously, but i assume you have done that [19:04] rtg, #hwe === fmasi is now known as fmasi_afk [19:47] apw, got the build started... had to do 'fakeroot debian/rules clean' first [21:12] jsalisbury: ping [21:49] * rtg -> EOD === kentb is now known as kentb-out