| *** edcragg has quit IRC | 00:51 | |
| *** straycat_ has joined #baserock | 02:42 | |
| *** straycat has quit IRC | 02:46 | |
| *** zoli__ has joined #baserock | 04:47 | |
| *** zoli__ has quit IRC | 05:27 | |
| *** zoli__ has joined #baserock | 05:28 | |
| *** paulw has joined #baserock | 06:18 | |
| *** zoli__ has quit IRC | 06:25 | |
| *** zoli__ has joined #baserock | 06:44 | |
| *** zoli__ has quit IRC | 06:52 | |
| *** zoli__ has joined #baserock | 07:34 | |
| *** bashrc has joined #baserock | 08:07 | |
| *** jonathanmaw has joined #baserock | 08:13 | |
| *** ssam2 has joined #baserock | 08:47 | |
| *** ChanServ sets mode: +v ssam2 | 08:47 | |
| ssam2 | I want to merge https://gerrit.baserock.org/1003/ (migrate to definitions version 6 in definitions) today, unless anyone objects | 08:53 |
|---|---|---|
| *** edcragg has joined #baserock | 09:00 | |
| *** flatmush has quit IRC | 09:14 | |
| *** flatmush has joined #baserock | 09:14 | |
| *** lmackenzie_2015 has joined #baserock | 09:22 | |
| *** franred has joined #baserock | 09:32 | |
| *** ssam2 has quit IRC | 09:53 | |
| *** mdunford has quit IRC | 10:05 | |
| *** ssam2 has joined #baserock | 10:08 | |
| *** ChanServ sets mode: +v ssam2 | 10:08 | |
| *** paulw has quit IRC | 10:27 | |
| *** lachlanmackenzie has quit IRC | 10:27 | |
| *** persia has quit IRC | 10:27 | |
| *** _longines has quit IRC | 10:27 | |
| *** paulw has joined #baserock | 10:29 | |
| *** lachlanmackenzie has joined #baserock | 10:29 | |
| *** persia has joined #baserock | 10:29 | |
| *** _longines has joined #baserock | 10:29 | |
| *** paulw has quit IRC | 10:30 | |
| *** lachlanmackenzie has quit IRC | 10:30 | |
| *** persia has quit IRC | 10:30 | |
| *** _longines has quit IRC | 10:30 | |
| *** mdunford has joined #baserock | 10:31 | |
| *** paulw has joined #baserock | 10:32 | |
| *** lachlanmackenzie has joined #baserock | 10:32 | |
| *** persia has joined #baserock | 10:32 | |
| *** _longines has joined #baserock | 10:32 | |
| *** ssam2 has quit IRC | 10:49 | |
| *** ssam2 has joined #baserock | 11:06 | |
| *** ChanServ sets mode: +v ssam2 | 11:06 | |
| *** paulw has quit IRC | 11:29 | |
| *** paulw has joined #baserock | 11:34 | |
| *** paulw has quit IRC | 12:14 | |
| *** paulw has joined #baserock | 12:26 | |
| *** paulw has quit IRC | 12:27 | |
| *** paulw has joined #baserock | 12:27 | |
| *** wschaller has joined #baserock | 13:22 | |
| *** zoli__ has quit IRC | 13:29 | |
| *** jonathanmaw has quit IRC | 13:35 | |
| rjek | ssam2: The old adage is that C++ compilers work by sending your source into the future to be compiled by Skynet. The computations required for the transmission require lots of RAM. | 13:44 |
| robtaylor | rjek: indeed, its a little know fact that compiling C++ is AI-Complete | 13:48 |
| rjek | :) | 13:49 |
| *** paulw has quit IRC | 13:55 | |
| *** mdunford has quit IRC | 13:58 | |
| *** paulw has joined #baserock | 14:00 | |
| *** paulw has quit IRC | 14:02 | |
| *** mdunford has joined #baserock | 14:13 | |
| paulsherwood | i've asked this before, but i can't remember the answer... can i build armv7 natively on aarch64 (eg moonshot?) | 14:18 |
| rjek | It's complex. | 14:18 |
| radiofree | i have a mustang on my desk if you'd like to try it | 14:18 |
| robtaylor | rjek: can you boot an armv7 rootfs on one? | 14:18 |
| rjek | You can ask GCC to generate 32 bit ARM code, and it will link. There is no support for interworking between 64 and 32 bit instruction sets like there is between 32 and 16. | 14:18 |
| radiofree | booting armv7 rootfs would probably be the easiest thing | 14:19 |
| rjek | ARMv8 does not /require/ a CPU to contain support for the 32 bit instruction set. | 14:19 |
| robtaylor | rjek: but they all do.. | 14:19 |
| robtaylor | (so far) | 14:19 |
| rjek | I have not conducted a survey to confirm that, but it seems likely doesn't it? :) | 14:19 |
| robtaylor | paulsherwood: so, just boot an armv7 rootfs :) | 14:20 |
| paulsherwood | rewind... to be more precise... i want to run ybd on moonshot. moonshot runs ubuntu, little endian. would the armv7 definitions build or not? | 14:20 |
| rjek | I don't think morph would enjoy trying to build a 32 bit ARMv7 world while running on a 64 bit ARMv8 would, though, so easiest solution is probably to boot the ARMv7 world and build there | 14:20 |
| * paulsherwood doens't have ability to boot his target machine into anything other than ubuntu, it seems | 14:20 | |
| rjek | paulsherwood: If you can control what parameters get passed to GCC (to set -march=armv7 and similar) it might | 14:21 |
| rjek | paulsherwood: Unpack a 32 bit world, and chroot? | 14:21 |
| paulsherwood | rjek: you assume i understand more than i do | 14:21 |
| radiofree | i can test that out if you want | 14:21 |
| paulsherwood | do you have access to a moonshot? | 14:22 |
| rjek | No, but he has a Mustang. | 14:22 |
| radiofree | i have a mustang next to me | 14:22 |
| paulsherwood | ah! :) | 14:22 |
| rjek | paulsherwood: You can use a tool called "debootstrap" to build a Debian or Ubuntu chroot system which you can then enter, or you can somehow arrange for GCC to always be invoked with -march=armv7. I suspect compiling "natively" via a chroot would probably be easiest/most reliable | 14:23 |
| rjek | Actually, I've never tried debootstrapping Ubuntu; I have no idea if that works. | 14:23 |
| radiofree | if this works i suppose i could leave it populating armv7 caches on cache.br | 14:24 |
| rjek | debootstrap --arch=armhf --include=build-essential jessie my-jessie-chroot http://ftp.uk.debian.org/debian && time-passes && chroot my-jessie-chroot | 14:24 |
| radiofree | i can imagine it's quite fast | 14:25 |
| rjek | Oh you need --foreign in there too, and after chrooting run /debootstrap/debootstrap --second-stage to complete the install | 14:25 |
| *** rdale has joined #baserock | 14:28 | |
| radiofree | does baserock publish armv7 rootfs images? | 14:31 |
| radiofree | erm.. archives | 14:31 |
| rjek | radiofree: Only x86 for chroot tarballs IIRC | 14:34 |
| rjek | But you should be able to download the ARMv7 raw disk image and use kpartx to loopback mount it | 14:35 |
| paulsherwood | this all sounds lovely, but it might as well be in a foreign language to me :) | 14:36 |
| radiofree | rjek: that's exactly what i did | 14:37 |
| radiofree | "mount img.img foo/" is good enough | 14:37 |
| rjek | Ah, no partition table? Bonus. | 14:37 |
| bwh | since I just did it today, this is *all* the steps for a foreign debootstrap: https://blog.night-shade.org.uk/2013/12/building-a-pure-debian-armhf-rootfs/ | 14:47 |
| paulsherwood | bwh: great! | 14:48 |
| bwh | vmdebootstrap is a wrapper that builds a disk image complete with kernel and boot loader, but I don't know if it does cross-building or supports boot loaders other than grub | 14:48 |
| * rjek moans a bit at bwh's CSS. | 14:49 | |
| bwh | rjek: Not me, I just read and (roughly) followed that | 14:49 |
| rjek | Oh, right, sorry, misinterpreted what you said | 14:49 |
| bwh | unless you're talking about CSS on decadent.org.uk which is, ahem, minimal | 14:50 |
| rjek | For some reason I read night-shade as decadent-place | 14:50 |
| radiofree | ok this mustang is building a jetson armv7 image in a chroot | 14:52 |
| rjek | \o/ | 14:53 |
| paulsherwood | kewl! | 14:53 |
| rjek | Do you have a Jetson to try the result on? :) | 14:54 |
| radiofree | i do | 14:54 |
| radiofree | if it works maybe we can setup an arm mason on it? | 14:54 |
| rjek | Or you could chroot into the chroot from inside your chroot and try building another chroot there. | 14:54 |
| radiofree | although i had to use ybd because i'd have to hack morph not to get the arch from uname | 14:54 |
| paulsherwood | w00t! :) | 14:54 |
| rjek | radiofree: Does "linux32" do on ARMv8 what it does on x86? | 14:55 |
| rjek | (What does uname return on the Mustang?) | 14:55 |
| radiofree | ooh right, i could have done that | 14:55 |
| paulsherwood | radiofree: any chance yo ucould paste the log somewhere, if/when it inishes? | 14:55 |
| rjek | setarch seems to be the general thing | 14:55 |
| radiofree | rjek: ah no i couldn't do that | 14:56 |
| radiofree | linux32 uname -a | 14:56 |
| radiofree | Linux baserock-mustang 4.1.0-rc7 #7 SMP PREEMPT Mon Jun 8 17:32:42 BST 2015 armv8l GNU/Linux | 14:56 |
| rjek | setarch armv7l uname -a ? | 14:56 |
| radiofree | uname -a | 14:56 |
| radiofree | Linux baserock-mustang 4.1.0-rc7 #7 SMP PREEMPT Mon Jun 8 17:32:42 BST 2015 aarch64 GNU/Linux | 14:56 |
| radiofree | apparently armv7l is an "Unrecognized architecture" | 14:57 |
| radiofree | anyway, you can just pass the arch to ybd | 14:57 |
| * rjek nods | 14:57 | |
| rjek | armv7l was a guess for what it returns on a real ARMv7 box; it may be wrong. | 14:58 |
| radiofree | paulsherwood: the log is stdout right? | 14:58 |
| paulsherwood | radiofree: yup. i tend to tee it | 14:58 |
| rjek | Anyway, uname is full of lies at the best of times, perhaps morph shouldn't be using it | 14:58 |
| radiofree | paulsherwood: can you give me the command please | 14:58 |
| rjek | ybd ... | tee logfile.txt | 14:58 |
| paulsherwood | yup. for example ../ybd/ybd.py clusters/ci.morph | tee some-memorable-name.log | 14:59 |
| * radiofree adds sys.stdout.flush() to the log() function | 15:01 | |
| paulsherwood | patch please? :) | 15:02 |
| radiofree | ybd fetches everything before building? | 15:11 |
| * radiofree doesn't think he's actually built anything yet | 15:11 | |
| paulsherwood | radiofree: do you have gits already on that box? | 15:12 |
| radiofree | not in the chroot | 15:16 |
| paulsherwood | ah, ok | 15:16 |
| paulsherwood | do you have enough storage in the chroot for this activity? | 15:16 |
| radiofree | i have 28G free | 15:17 |
| radiofree | 36G free now | 15:17 |
| paulsherwood | not enough | 15:18 |
| radiofree | 40.9G? | 15:18 |
| paulsherwood | for one system maybe | 15:18 |
| radiofree | the ssd is only 120G | 15:19 |
| radiofree | i have built a devel system on the mustang before, so 44G is probably enough | 15:20 |
| radiofree | no wait, it's 64G | 15:20 |
| radiofree | (the ssd) | 15:20 |
| paulsherwood | well it will probably run out of space | 15:20 |
| radiofree | really? it didn't run out space when i built a devel system previously | 15:21 |
| paulsherwood | oh? maybe not then :) | 15:21 |
| paulsherwood | radiofree: thanks for the patch! | 15:24 |
| ssam2 | I just submitted my first ever patch to bitbake-devel@lists.openembedded.org :-) | 15:46 |
| ssam2 | only as a "proof of concept", though, it's still a really ugly hack | 15:46 |
| *** rdale has quit IRC | 15:49 | |
| paulsherwood | ssam2: cool! is there any demo or something to see? | 15:50 |
| radiofree | it failed! mount(/, MS_PRIVATE | MS_REC): Invalid argument | 15:55 |
| radiofree | i guess it's because i'm chrooting in a chroot? | 15:55 |
| radiofree | i'll use schroot? i know that works | 15:55 |
| radiofree | anyone know where to get schroot source? | 16:02 |
| radiofree | configure: error: Boost.Format (Boost C++ Libraries) is not installed, but is required by schroot | 16:04 |
| radiofree | what's the correct chroot foo for launching a baserock chroot | 16:04 |
| paulsherwood | radiofree: http://wiki.baserock.org/guides/chroot/ | 16:06 |
| radiofree | i'd rather do this without baserock-chroot | 16:10 |
| radiofree | since that needs schroot which would involve me building boost | 16:10 |
| radiofree | known issue! | 16:11 |
| paulsherwood | ack | 16:11 |
| * radiofree tries the fix | 16:11 | |
| paulsherwood | radiofree: does your system have linux-user-chroot? | 16:11 |
| paulsherwood | if not, adding it might be a soln | 16:12 |
| paulsherwood | ybd defaults to luc if it finds it, falls back to chroot | 16:12 |
| radiofree | yes it has it | 16:13 |
| paulsherwood | is it using it? | 16:13 |
| paulsherwood | (ie does your log show any message about rogue build? | 16:14 |
| paulsherwood | if it mentions rogue build, it's not finding luc | 16:14 |
| radiofree | restarting usually fixes things, i'll try that | 16:14 |
| radiofree | working! | 16:17 |
| radiofree | http://wiki.baserock.org/guides/build-failures/#index7h2 | 16:17 |
| radiofree | + mount bind /dev | 16:17 |
| radiofree | and proc | 16:18 |
| paulsherwood | what was the fix? | 16:21 |
| radiofree | [17:17:41] <radiofree> http://wiki.baserock.org/guides/build-failures/#index7h2 | 16:21 |
| radiofree | [17:17:47] <radiofree> + mount bind /dev | 16:21 |
| radiofree | [17:18:45] <radiofree> and proc | 16:21 |
| paulsherwood | ah, so should ybd do that? | 16:22 |
| radiofree | no you do this outside of the chroot | 16:22 |
| radiofree | paulsherwood: basically http://fpaste.org/260741/07790351/ | 16:24 |
| paulsherwood | radiofree: aha, ok. thanks! | 16:25 |
| radiofree | it's using all 8 cores to build | 16:25 |
| radiofree | which is nice | 16:25 |
| paulsherwood | :) | 16:26 |
| radiofree | http://fpaste.org/260742/44077917/ | 16:26 |
| radiofree | for my v7 chroot i just downloaded a jetson devel image | 16:27 |
| radiofree | mount -o jetson.img /mnt | 16:27 |
| radiofree | mkdir v7chroot | 16:27 |
| paulsherwood | ack, thanks | 16:28 |
| radiofree | cp -r /mnt/systems/factory/run/* /src/v7chroot/ | 16:28 |
| radiofree | gcc fails | 16:30 |
| radiofree | stage1 | 16:30 |
| radiofree | gcc.c:(.rodata+0x1b0): undefined reference to `host_detect_local_cpu(int, char const**)' | 16:30 |
| *** zoli__ has joined #baserock | 16:35 | |
| radiofree | fixed | 16:54 |
| radiofree | i added the commands to a script and ran it with /usr/bin/linux32 -B ./launch.sh | 16:54 |
| *** bashrc has quit IRC | 17:17 | |
| *** zoli__ has quit IRC | 17:28 | |
| *** franred has quit IRC | 17:32 | |
| radiofree | this might actually work | 17:32 |
| radiofree | http://fpaste.org/260774/78316214/ | 17:32 |
| radiofree | 15-08-28 17:30:03 [14/280/282] [glibc] Elapsed time for assembly 00:24:17 ?? | 17:32 |
| *** lmackenzie_2015 has quit IRC | 17:51 | |
| *** ssam2 has quit IRC | 17:54 | |
| *** mdunford has quit IRC | 17:58 | |
| *** wschaller has quit IRC | 18:15 | |
| *** zoli__ has joined #baserock | 19:12 | |
| *** zoli__ has quit IRC | 19:23 | |
| *** zoli__ has joined #baserock | 19:25 | |
| *** edcragg has quit IRC | 19:30 | |
| *** zoli__ has quit IRC | 19:38 | |
| paulsherwood | radiofree: assembly is a bit of a movable feast... may or may not include build time of dependencies too | 21:14 |
| paulsherwood | Elapsed time for build is more interesting in general i think | 21:14 |
| *** zoli__ has joined #baserock | 21:18 | |
| *** zoli__ has joined #baserock | 21:19 | |
| *** zoli__ has quit IRC | 21:23 | |
| radiofree | paulsherwood: that time is totally inaccurate | 23:38 |
| * radiofree will check on the progress tomorrow | 23:39 | |
| radiofree | A fast arm build server for caches will be pretty neat to have | 23:39 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!