**** BEGIN LOGGING AT Wed Nov 30 02:59:57 2011 Nov 30 10:57:40 Um, the latest I checked at raspberrypi.org, the final version is using Broadcom BCM2835, ARM11 compatible - isn't that one of the "newer" versions, that should be supported by Ubuntu? Nov 30 10:59:20 WaltherF1: Unless something's changed, the Pi is armv6, and we build for v7. Nov 30 11:03:55 infinity: They published a new graphic for the final version, http://www.raspberrypi.org/wp-content/uploads/2011/11/Raspi-Model-AB-Mono-1.png Nov 30 11:04:04 infinity: and it exclusively states ARM11 Nov 30 11:04:11 WaltherF1: ARM11 = ARMv6 Nov 30 11:04:31 ah, okay, sorry for the confusion Nov 30 11:04:50 ...and btw, imo that's not a neat way of arranging your versioning -.- Nov 30 11:04:57 (not your fault!) Nov 30 11:05:22 http://en.wikipedia.org/wiki/List_of_ARM_microprocessor_cores Nov 30 11:05:25 And yeah, it's confusing. Nov 30 11:05:43 send your thanks to ARM for that kind of crappy versioning :) Nov 30 11:06:05 The Cortexes continue the confusion with going from A8 to A9 to A7 and A15... Nov 30 11:06:11 yeah Nov 30 11:06:45 at least A7 will then be v7 Nov 30 11:06:48 Hah. Nov 30 11:07:49 heh Nov 30 11:09:27 duh, at least ubuntu is sensible with yy.mm Nov 30 11:09:30 janimo, oh, btw, having the ac100 kernel built for el as well as hf would be nice Nov 30 11:09:52 ogra_: Kernels will come. Nov 30 11:09:54 WaltherF1, well, but we use funny names too :) Nov 30 11:10:05 I'll bug people more insistently when the archive is actually installable. Nov 30 11:10:11 infinity, that kernel will need jani or me touching it ... its not maintained by the kernel team Nov 30 11:10:16 ogra_: el and hf? Nov 30 11:10:19 ogra_: I know. Or me. :P Nov 30 11:10:24 indeed :) Nov 30 11:10:26 lilstevie: armel and armhf. Nov 30 11:10:27 kernels shouldn't have that issue Nov 30 11:10:34 infinity: yeah I am aware Nov 30 11:10:40 lilstevie: The kernel needs to exist in the hf archive. Nov 30 11:10:45 lilstevie, but you want an armhf .deb ;) Nov 30 11:10:48 Which means, sadly, building it twice. Nov 30 11:11:02 mer guys assured me that one kernel should be universal Nov 30 11:11:03 (Or building it as arch:all, but we really don't need ARM kernels on powerpc...) Nov 30 11:11:06 its all about dpkg ... not actually the binary Nov 30 11:11:14 ah Nov 30 11:11:21 lilstevie: Identical build. Just needs to be in both _armel.deb and _armhf.deb form. :P Nov 30 11:11:32 kinda crappy system Nov 30 11:11:45 we could as well just cp them on ports.u.c :P Nov 30 11:11:46 Well, dpkg could handle it fine, to be fair. Nov 30 11:11:51 We could do multiarch kernels. Nov 30 11:11:53 And should, eventually. Nov 30 11:12:01 But we'd need some installer mangling to make that work. Nov 30 11:12:26 ah Nov 30 11:12:42 The path of least resistance for now is just two builds. It's not world-ending. Nov 30 11:12:50 Just a waste of 4 hours of a panda's time. Nov 30 11:13:25 heh Nov 30 11:13:28 and given that one of the arches will go away anyway ... Nov 30 11:13:45 yeah Nov 30 11:14:09 ogra_: I kinda want to rebootstrap armel as v4T as a community project (and maintain bi-arch/multi-arch with it)... Is that wrong? :P Nov 30 11:14:31 I'd have to do it out of work hours, but I'm quite tempted, to be honest. Nov 30 11:14:37 well, the pi people would love you forever for that :) Nov 30 11:14:44 infinity: not really, that would solve the problems with all these *plug devices wanting ubuntu Nov 30 11:14:45 i would help := Nov 30 11:14:47 :) Nov 30 11:15:23 now this sounds good, I'd be eager to help with all my knowledge as well Nov 30 11:15:24 I would donate CPU time on my trimslice as a builder if that would help Nov 30 11:15:48 is theere actually much work to do apart from changing compiler flags and do a complete rebuild ? Nov 30 11:16:18 (and getting approval for the resource costs indeed) Nov 30 11:16:26 ogra_: And back out any braindead patches we have that assume thumb/vfp/etc. Nov 30 11:16:43 well, just re-roll debian then Nov 30 11:16:46 ogra_: Basically, revert to pure Debian compat for armel. Which shouldn't be too hard. Nov 30 11:16:56 yeah Nov 30 11:17:20 infinity, but why bother duplicating Debian, it's already there? Nov 30 11:17:20 The upshot is that we can do it on v7 hardware and just cheat. Nov 30 11:17:38 davidm: Because we get a TON of people who are grumpy when we say "use Debian for you < v7 device". Nov 30 11:17:44 davidm, there is no unity :) Nov 30 11:18:07 davidm: And when our official v7 solution becomes armhf, it might be a nice community service to offer a baselive v4T again. Nov 30 11:18:52 Gentlemen (and ladies?), I appreciate this. Thank you. Nov 30 11:18:59 indeed totally unsupported etc Nov 30 11:19:02 WaltherF1: Nothing to appreciate yet. :P Nov 30 11:19:12 WaltherF1: I need to get a lot of people to agree to my insanity before it could ever happen. Nov 30 11:19:18 infinity: heh Nov 30 11:19:44 WaltherF1: (And the official answer for now on the Pi is, indeed, "use Debian") Nov 30 11:19:46 and I suspect with the current lack of HW it's not going to happen Nov 30 11:19:48 infinity: But hey, I got you convinced to my insanity already, isn't that a start :P Nov 30 11:20:06 WaltherF1, even if it would happen that would be in a year from now or so Nov 30 11:20:21 davidm: Get me one 4U Calxeda box and we can build 4 arm* ports and keep up. :P Nov 30 11:20:30 yeah yeah, i'm not going to spoil this anywhere... But I can guarantee, (if)/when you'll announce the support officially, internet will shit bricks Nov 30 11:20:53 WaltherF1: It would never be "official". Nov 30 11:21:07 WaltherF1: It would be as unofficial as our powerpc port (which people keep trying to drop...) Nov 30 11:21:10 So... Nov 30 11:21:18 I wouldn't hold your breath, is the take-home message. ;) Nov 30 11:21:21 Still, I'd like to do it. Nov 30 11:21:49 we should also add nslu2 images back :P Nov 30 11:21:58 once we have such a prot Nov 30 11:22:00 *port Nov 30 11:23:59 sneaky.. so when fedora moves to armhf we grab the legacy armv5 armel turf? Nov 30 11:25:03 xranby, its just a weird idea by a sleep lacking developer :P Nov 30 11:25:53 xranby: Fedo--who? Doesn't evern v5 user run Debian? :) Nov 30 11:26:04 s/evern/every/ Nov 30 11:27:58 at least some old fedora verson was preinstalled on the openrd boards Nov 30 11:30:31 also some netgear NAS like the Stora are based on a locked down fedora Nov 30 11:36:46 infinity: but.. yeah.. i agree I do run debian on my openrd machines and on my Stora :) Nov 30 16:16:38 NCommander: ping Nov 30 16:18:18 infinity: ping...any idea when we can expect to have an armhf image for folks to test out? Nov 30 16:19:18 robbiew: Desktop won't be for quite a while (hard to estimate while build-deps are being unsnagged), server might be doable (with some bits missing) in a week or so? Nov 30 16:19:53 robbiew: Very hand-wavy at this point, hard to commit until I can get a good overview of what I still need to manually bootstrap. Nov 30 16:19:59 ack Nov 30 16:20:07 robbiew: Plus, I'll need to force some server bits to the front of the queue to make that estimate. ;) Nov 30 16:20:07 infinity: cool, thx Nov 30 16:20:12 heh Nov 30 16:20:26 infinity: looks like the ARM Supercomputer folks are willing to testdrive our bits Nov 30 16:20:35 so the sooner the better ;) Nov 30 16:20:42 robbiew: I'd like to testdrive theirs too! Nov 30 16:20:48 (Man, that sounds bad) Nov 30 16:20:50 I'm working on that Nov 30 16:21:29 robbiew: I'll see what I can do about reordering and jamming some more stage-2 bootstrap bits in for server dep loops. Nov 30 16:22:06 robbiew: Of course, we won't have installable images for those folks (as I assume they use kernels we don't ship), but I also assume they're smart enough to figure out how to install minimal chroots and go from there. Nov 30 16:22:32 infinity: yeah...given they've been using Ubuntu already...I think so :P Nov 30 16:23:25 robbiew: Feel free to keep giving me gentle prods over the next week or two. I'll see what I can do about delivering most of server-ship. Nov 30 16:23:39 (Hey, you already have MySQL, we're done, right?) Nov 30 16:23:47 infinity, robbiew: a debootstrappable buildd chroot would be appreciated as the first step ;-P Nov 30 16:23:47 (Right?) Nov 30 16:23:56 doko: Yeah, yeah. Working on it. Nov 30 16:24:16 couldn't resist ... Nov 30 16:24:31 infinity:cool...who should I talk to about producing an ARM HPC kernel for them...is that you all or #ubuntu-kernel folks Nov 30 16:24:44 doko: You can cheat and debootstrap --minbase from http://people.canonical.com/~ubuntu-archive/armhf/ for now, add ports.u.c to the sources.list, and install build-essential. Nov 30 16:24:48 They need kernel scheduler and parameters tuned for single-threaded execution, as there is no user interaction or time-sharing in an HPC platform Nov 30 16:25:52 robbiew: Well, I assume it would be targetted specifically at their hardware as well. You could talk to davidm about it. I'm fairly sure it won't be his team that ends up doing the work, but he'll know who should. :P Nov 30 16:26:19 doko: --variant=minbase, that is. But you know what I mean. Nov 30 16:26:47 doko: But I'll get --variant=buildd working from ports fairly soon. Need a kernel and a few other bits. Nov 30 16:27:08 infinity: ack, thx Nov 30 16:27:22 infinity, minbose is supposed to work? Nov 30 16:27:23 infinity: yeah...could be something done as part of a Canonical/BSC relationship Nov 30 16:27:34 doko: minbase works from people. Nov 30 16:27:41 ahh Nov 30 16:27:44 doko: Probably not from ports yet. Nov 30 16:28:31 doko: buildd would theoretically work from people too, except my stage-2 archive doesn't take overrides into account, so sections/priorities are wrong, and debootstrap has a bit of a hissy fit about the whole deal. Nov 30 16:29:42 doko: But yeah, if you want to play, minbase from people, add ports at the top of sources.list (and keep people), install build-essential, have fun. Nov 30 16:32:20 well, shouldnt -core be buildable already ? Nov 30 16:32:36 we could probably make A1 with it ;) Nov 30 16:32:59 (for hf that is) Nov 30 16:34:04 ogra_: Not yet. Nov 30 16:35:29 * infinity tests this statement. Nov 30 16:35:59 if minbase and apt work i would have guessed it should be buildable Nov 30 16:36:15 infinity, did you ever get a reply from apw about linux-libc-dev built from separate source? Nov 30 16:36:15 Note that I said to debootstrap minbase from my stage-2. Nov 30 16:36:20 Not from the archive. ;) Nov 30 16:36:20 (which i would expect to be reqs for the buildd setup) Nov 30 16:36:27 ah Nov 30 16:36:34 that one i missed indeed :) Nov 30 16:36:39 doko: linux-libc-dev is buildable out-of-band with DEB_STAGE=stage1 Nov 30 16:36:49 doko: Works well enough for bootstrapping. Nov 30 16:37:24 ogra_: Yeah, minbase is still missing a few bits in the archive. Will sort it. Nov 30 16:38:15 Oh, or not just yet. Nov 30 16:38:21 Need ruby1.9 to build libselinux. Nov 30 16:38:24 La la la. Nov 30 16:38:26 Sorting slowly. ;) Nov 30 16:38:31 no hurry, i wasnt actually serious about A1 anyway Nov 30 16:39:37 looks like the bootstrap is stalled now, universe packages start building Nov 30 16:43:55 robbiew: pong Nov 30 16:44:12 doko: I'm revisiting some snags. Nov 30 16:45:00 NCommander: nevermind...all sorted ;) Nov 30 16:45:25 robbiew: k :-) Nov 30 16:47:44 NCommander: got another question for ya...any reason why hdf5 isn't tagged for armhf builds? -> https://launchpad.net/ubuntu/precise/+source/hdf5 Nov 30 16:48:00 hdf4 is, but our supercomputer friends in barcelona would like hdf5 :) Nov 30 16:48:10 robbiew: its probably in P-a-s Nov 30 16:48:22 looking Nov 30 16:48:48 it probably simply missesw armhf in debian/control Nov 30 16:48:53 *misses Nov 30 16:49:09 What ogra_ said. Nov 30 16:49:11 ogra_: any all Nov 30 16:49:19 It's not P-a-s. Nov 30 16:49:26 hrm, no its got a armhf build record Nov 30 16:49:33 robbiew: https://launchpad.net/ubuntu/+source/hdf5/1.8.4-patch1-3ubuntu1/+build/2963282 Nov 30 16:49:36 Oh, then it just needs to build. :P Nov 30 16:49:40 ah Nov 30 16:49:52 cool...nevermind :) Nov 30 16:50:15 robbiew: always use the source overview page to look at what builds are doing what where: https://launchpad.net/ubuntu/+source/hdf5 Nov 30 16:51:19 robbiew, http://qa.ubuntuwire.org/ftbfs/primary-precise-armhf.html might also be intresting for you in case you look for failed builds (not actually instresting at this stage of the bootstrap but will be soon) Nov 30 16:51:26 doko: We have a fair few packages that need -lm fixes. You could queue those up for uploads right after the freeze is over, if you're feeling helpful. ;) Nov 30 16:51:37 NCommander: thanks...I **thought** I did :/ Nov 30 16:52:03 robbiew: you were looking at the percise version page. LP's URLs for finding stuff are kinda irritating Nov 30 16:52:18 robbiew: having the release name in the URL is a drastically different page. Annoys me to no end. Nov 30 16:52:19 indeed Nov 30 16:52:27 +1 infinity Nov 30 16:52:50 (Try having the release name in a bug URL and see how that screws with your head) Nov 30 16:52:58 Like, being unable to target things, etc. Nov 30 17:02:02 infinity, hmm, new -as-needed ftbfs? Nov 30 17:04:12 doko: No, it was the "something used to needlessly link against libm and doesn't anymore" thing. Nov 30 17:04:32 doko: Remember that landed late in the oneiric cycle and then seb128 rapidly reverted it? Nov 30 17:04:42 ahh, the seb128^Bgnome mess Nov 30 17:04:48 doko: Well, it's landed for good now. Nov 30 17:05:32 doko: https://launchpad.net/ubuntu/+source/gnome-desktop/1:2.32.1-0ubuntu6/+build/2962501 <-- For example. Nov 30 17:06:08 I'm working on the "stuff that's FTBFS due to changes in if_packet.h" failures right now. Nov 30 17:18:26 ahh, like klibc? Nov 30 17:20:02 doko: klibc, iproute... I'm sure I've seen a couple more. Will comb. Nov 30 17:20:45 Of course, I also need a nap at some point. Nov 30 17:20:51 python distracted me last night. Nov 30 17:21:04 ctypes? Nov 30 17:21:07 ctypes is hilariously broken. Nov 30 17:21:09 Yeah. Nov 30 17:21:16 lool sent a patch Nov 30 17:21:22 Yeah, I'm building with it. Nov 30 17:21:32 But that's just papering over awful code. Dec 01 02:09:25 infinity, something is f*cked, python2.7 depends on libsfgcc1 Dec 01 02:31:03 doko: Err, does it? Dec 01 02:31:08 doko: Just now? Dec 01 02:32:41 doko: Grr. My build of 5ubuntu1.1 must have been in a dirty chroot. Though, the part where it picked up that dependency is worrying. Dec 01 02:34:07 doko: Yeah, confirmed that I built it in a chroot with gcc-multilib installed. But ouch that it didn't do the right thing. :/ Dec 01 02:42:37 doko: Erk, and gcj-4.6 depends on libsfgcc1 ... **** ENDING LOGGING AT Thu Dec 01 02:59:58 2011