[00:30] checking for arm-linux-gnueabi-pkg-config... (cached) /usr/bin/pkg-config [00:30] thbbt [01:06] * utlemming is away: Gone away for now [02:22] * utlemming is back. [03:00] * utlemming is away: Gone away for now === Jack87 is now known as Jack87|Away [05:20] * utlemming is back. === Jack87|Away is now known as Jack87 [05:51] * utlemming is away: Gone away for now === Jack87 is now known as Jack87|Away === Jack87|Away is now known as Jack87 === Quintasan_ is now known as Quintasan === Jack87 is now known as Jack87|Away [12:40] * ogra_ sees ubuntu-core armhf and applauds infinity ... [12:41] :) [12:41] yay [12:57] ogra_: Short-lived, mind you. The buildd is acting up. [12:57] oh, still ? [12:58] ogra_: It'll be doing dailies "later"... When lamont and I can diagnose the issue. [12:58] ogra_: "still"? It's a new buildd. [12:58] * ogra_ just tried to read the webkit ftbfs log .. [12:58] Don't worry about it. [12:58] smehow it exhausts my 512MB ram [12:58] Unless you're an ELF wizard. [12:58] * ogra_ doesnt think he is :P [12:59] The webkit FTBFS (and a few others) is due to libicudata.so.48 being a braindead non-library with a hand-crafted (and incorrect) ELF header. [12:59] I'll get to it soon. [12:59] yeah, just saw that in brtty [12:59] * infinity nods. [12:59] i just dont get why the log is to big for my ram ... [13:00] Cause web browsers suck. [13:00] heh, yeah [13:00] 5MB log = 500MB to render. [13:00] plus caching ... [13:00] There's a reason I have a laptop with 4G of RAM. [13:00] Which is now not enough. :P [13:00] Seemed good 4 years ago. [13:01] well, usually my 2G on the ac100 are enough, just slow [13:01] (due to swapping) [13:02] Hrm. [13:02] Sleep or fix icu, sleep or fix icu... [13:02] sleep ! [13:02] if you can [13:02] I probably can. [13:03] I only slept 3 hours yesterday. From 8am to 11am. [13:03] * ogra_ tries a hf chroot from -core [13:03] So, I'm at 3 hours of sleep in the last ~48... [13:03] yeah, go to bed ... [13:03] ogra_: Should work. Let me know if it's broken. I didn't actually check the manifest or anything. [13:04] tar acts up :( [13:04] ... [13:04] Let me check.. [13:04] * ogra_ tries again [13:04] but my system might be at fault after two times running out of ram for the webkit log [13:05] i should probably reboot to clear all caches [13:05] tar was happy here. [13:06] hmm, untarring with sudo works [13:06] weird [13:06] ogra@horus:~/chroot-hf$ sudo chroot . [13:06] root@horus:/# dpkg --print-architecture [13:06] armhf [13:06] root@horus:/# [13:06] \o/ [13:07] great, i have something to work in [13:07] apt-get update and upgrade work, I call it a success. [13:07] build-essential won't work for you. No linux-libc-dev. [13:07] Let me fix that for you. :P [13:08] no ! [13:08] go sleep ! [13:08] It's 2 seconds to give you a link. :P [13:08] its saturday [13:08] k === infinity changed the topic of #ubuntu-arm to: Ubuntu ARMv7 Discussion & Development | https://wiki.ubuntu.com/ARM | Submit a Bug? https://bugs.launchpad.net/ubuntu/+filebug | Get oneiric while it's hot ! http://cdimage.ubuntu.com/releases/11.10/release/ | Logs at http://irclogs.ubuntu.com/ | armhf builds are running | armhf headers at http://people.canonical.com/~ubuntu-archive/armhf/linux-libc-dev_3.2.0-2.4_armhf.deb [13:08] * infinity points up. [13:08] Enjoy. [13:08] * ogra_ wishes we could quiten the perl locale warnings in perl itself one day [13:09] Or you could not have mismatched locales. [13:09] LANG=C chroot chroot/ su - [13:09] sure i just alwas forget to call chroot with LANG=C [13:10] i mean the message could be a one liner instead of spamming half a terminal every time [13:11] And I just verified that with the above linux-libc-dev, build-essential installs in core. [13:11] So, yay. [13:11] Enjoy. [13:11] yeah, build-essential seems to install fine [13:12] Actually. [13:12] If you want to be super effin' helpful, grab the linux source package and make it stop failing on armhf. :P [13:12] Upload with extreme prejudice, and then push a patch to apw. [13:12] Sick of linux-libc-dev not being in the archive. [13:13] * apw can hear you [13:13] apw: Can you hear the grump from across the ocean? ;) [13:14] why the heck is everyone working on a sat. [13:14] ogra_: It's still Friday for me. [13:14] And will be until I sleep.. [13:14] and why is linux not on the ftbfs list if it fails [13:15] Cause there've been a few new uploads since. [13:15] And it's in needs-build. [13:15] ah [13:15] (But those uploads will all fail too) [13:16] due to the headers ? or due to the broked buildd ? [13:16] Neither... [13:16] Due to the arm build of the kernel not being fixed yet. [13:16] All the recent linux uploads have been for other arch stuff. [13:17] s/arm/armhf/ [13:17] oh, i thought we had that bit before alpha [13:17] armel works, but who cares about that? Yesterday's news. [13:17] yeah [13:17] old arch ... [13:17] obsolete crap :P [13:27] infinity, ahh balls more arm fookage, i hate arm [13:28] apw: It hates you too, honey. [13:28] you got somewhere you can test build a fix, i am tired of waiting for for uploads to fail [13:28] Sure. [13:31] apw, scheat.canonical.com [13:31] doko: There's no armhf chroot on scheat, unless lamont just fixed that? [13:32] ahh, I assume he's waiting on linux-libc-dev ;p [13:32] Yeah, cause he wants to debootstrap it and use his script. [13:32] apw: Fresh chroot all set up here to test. [13:34] apw, pretty please build linux-libc-dev from separate source [13:35] doko, i don't understand you have the thing buildable anyhow [13:35] maintaining that for 100s of uploads a cycle is pure madness for the 1 time every 5 cycles we need it [13:35] infinity, http://people.canonical.com/~apw/armhf/0001-UBUNTU-armhf-add-d-i-configuration.patch [13:35] Indeed. [13:35] not that this isn't madenning [13:36] arm being so very differnet from our core arches is hard on the head [13:36] apw: no, you wouldn't need to upload it hundreds of times [13:36] its version locked to the version of the kernel right now, as it comes from the kernel source [13:37] doko: I'm with apw, honestly. DEB_STAGE= fixes the bootstrap issue. After that, we need things to build ANYWAY. [13:37] So, saying "having linux-libc-dev come from another source package means the kernel can FTBFS forever" is a bit pointless. [13:37] ok, I'll re-merge gcj and gnat into gcc [13:37] right, i thought that fixed you _if_ i could get the krenl to build, sigh [13:38] infinity, how long will your test take ? i will prepare a source upload, and can push it later [13:38] (based on that outcome) [13:39] apw: I dunno. About 4 hours. But I might be asleep when it finishes. ;) [13:39] well ... yell if and when you find out, and i can get it uploaded [13:39] apw: Still better than wasting buildd time. [13:40] Oh, it's already wasting buildd time. :P [13:40] Well, better than wasting more! [13:41] infinity, heh yeah [13:42] infinity, i wonder thinking about bootstrap, i wonder what would happen if we had just uploaded the main source package but with DEB_STAGE=stage1 set, just once [13:42] that would have just bumped the useless packages, and made libc, and then uploaded it again without [13:42] then you'd have an old libc-dev and nothing else, and i'd still be ok the whole time right ? [13:42] apw: That would work fine, yeah. [13:42] apw: And if this test build fails, we might want to do that. :P [13:43] yeah, i'll think about that while you test :) [13:44] Anyhow, test build is going. [13:45] ~4 hours to build, ~∞ hours to nap. [13:45] Ish. [13:58] infinity, why was the eglibc rebuild needed? [14:00] doko: nscd depended on libsfgcc1. [14:00] bah [14:00] Some 500ish binary packages did. :P [14:01] Everything's reuploaded and building. [14:01] All better "soon". [14:02] dont exaggerate, i only had ~300 mails on -changes when i got up :P [14:02] ogra_: ~300 source packages, ~500 binary. [14:02] heh, k :) [14:03] (sorry i'm in fighting mode trying to help quietening a troll on ubuntu-users :P ) [14:04] Right. Nap time. [14:04] ogra_: If doko puts armhf buildds on manual, smack his knuckles with a ruler and remind him that building something (even if it's the "wrong" thing) is better than being idle and building nothing. :P [14:05] * ogra_ looks for that 350km long ruler ... :P [14:07] infinity, well, 15min isn't worth this [17:13] * utlemming is back. [20:24] So I have a kernel booting on the nook color but it's failing to go further than halting at udev [21:34] Xase: standard kernel issue with android devices, enable kernel maintained devtmpfs [21:35] standard kernel config issue* [21:53] hey anyone active ?