**** BEGIN LOGGING AT Tue Aug 30 02:59:57 2011 Aug 30 04:00:47 following directions found at https://github.com/openembedded/bitbake and building on a x86 box, and it seems to be building x86 binaries, and not arm binaries Aug 30 04:01:27 i think i am missing some core tidbit on getting open embedded to cross compile Aug 30 04:04:12 in the absence of help, i chastise myself and tell myself firmly to google the answer... Aug 30 04:05:28 i think i need to be using bash, source a file, then do the make Aug 30 04:12:55 change the target MACHINE in conf/local.conf Aug 30 04:39:40 i am missing something... something that makes it buidl arm binaries Aug 30 04:56:34 i am beginning to think the docs on openembedded.org are out of data or incomplete Aug 30 05:06:44 If you have the MACHINE set properly, bitbake will build the cross-compiler and then build ARM binaries for you Aug 30 05:07:20 That means it will end up building a number of native binaries before it gets to building things for your target Aug 30 05:09:59 oh Aug 30 05:10:25 NOTE: package perl-native-5.10.1-r10: task do_compile: Failed Aug 30 05:10:57 just confirm, building OE, rocket science, or not rocket science? Aug 30 05:11:24 Not rocket science, but there's a bit of a learning curve Aug 30 05:11:56 trying following the instructions on the oe site, and on teh chumby sites Aug 30 05:12:05 neither are working. its like there is a omission Aug 30 05:12:37 are these instructions correct looking https://github.com/clearwater/chumby-oe Aug 30 05:14:05 i have heard there are issues with corrupted fetches, requiring the builder to delete the local copy to force a re-get. Aug 30 05:14:09 is this still needed? Aug 30 05:14:15 I started out using the Angstrom Setup Scripts http://www.angstrom-distribution.org/building-angstrom Aug 30 05:15:03 I haven't had problems with corrupted fetches; most recipes include MD5 checksums Aug 30 05:17:08 after this current running attempt i will try those instructions. thanks! Aug 30 05:18:24 gm Aug 30 05:28:16 eFfeM_work: gm Aug 30 05:28:35 hi Noor, doing well? Aug 30 05:28:52 eFfeM_work: yeah .... u? Aug 30 05:29:21 yeah, need to report on my project in 1.5 hr to mgmt, but feel everything is under control :-) Aug 30 05:29:27 hope they feel the same :-) Aug 30 05:30:28 eFfeM_work: best of luck Aug 30 05:30:36 eFfeM_work: on whcih target you are working? Aug 30 05:31:10 it'll work out fine, this is on an in house developed powerpc board, but the system is derived from a freescale ref design Aug 30 05:32:03 ahan ... coool Aug 30 05:32:50 what board are you working with ? Aug 30 05:38:21 eFfeM_work: PPC, ARM and MIPS :d Aug 30 05:38:51 rigth now finding it little difficult to get uboot for MALTA MIPS board Aug 30 05:39:02 it does not have support for that Aug 30 05:39:15 quite a range! I've worked with arm on earlier projects and play with it at home (sheevaplug, beagleboard), but have hardly time recently Aug 30 05:39:20 been ages since I used mips Aug 30 05:39:56 your best bet with u-boot is use something that is close to your board as a starter Aug 30 05:40:00 but it can be tough Aug 30 05:40:47 eFfeM_work: yeah ..... team just figured it out how to flash the u-boot using YAMON Aug 30 05:41:04 ah ok Aug 30 07:12:47 03Paul Menzel  07master * r69d98e8604 10openembedded.git/recipes/genext2fs/files/volume.patch: Aug 30 07:12:48 genext2fs: Remove unused `recipes/genext2fs/files/volume.patch` Aug 30 07:12:48 This is a fix up for Aug 30 07:12:48 1b2afa96f12bc4e324e1793260aea91d203311d0 genext2fs: remove old versions Aug 30 07:12:48 which did not remove the patch. Aug 30 07:12:48 Signed-off-by: Paul Menzel Aug 30 07:16:15 how do I change the kernel in Angstrom? I tried adding "PREFERRED_PROVIDER_virtual/kernel_beagleboard = "linux-sakoman-pm" to my local.conf. That made the specified kernel compile as part of my rootfs, but its still linux-omap_2.6.39 that's packed into the rootfs. Aug 30 07:24:18 guessing "MACHINE=chumby ./oebb.sh bitbake console-image" takes a day or so.... Aug 30 08:14:30 Anyone? I've tried setting all the preferred providers I can think of, but it still chooses linux-omap_2.6.39.bb Aug 30 08:18:54 tasslehoff: You could send a message to the mailing list if nobody answers in this channel. Aug 30 08:19:53 PaulePanter: good idea Aug 30 10:53:52 03Jesse Gilles  07org.openembedded.dev * r16a8a1e917 10openembedded.git/recipes/ecj/files/ecj.in: (log message trimmed) Aug 30 10:53:52 classpath in 2011.03-maintenance Aug 30 10:53:52 On Fri, 2011-08-05 at 11:38 -0400, George C. Huntington, III wrote: Aug 30 10:53:52 > I posted a patch that makes the wrapper script return the error code Aug 30 10:53:52 > if it Aug 30 10:53:53 > fails after 5 retries. it is on patchwork. Aug 30 10:53:53 > http://patches.openembedded.org/patch/7249/ Aug 30 11:32:56 oe-core seems unbuildable like oe in the old times, because of missing downloads (:/(( Aug 30 12:34:55 if from the pull request one asks to resend one recipe ..... i assume I only need to send pull request of that recipe not the whole bunch of recipes that i send in previous pull request Aug 30 12:46:54 Noor: traditionally it's the whole pull request that gets resent Aug 30 12:47:20 if you use git rebase -i it's pretty easy to change the commit that needs fixing Aug 30 12:47:26 bluelightning: :O ok Aug 30 12:47:41 bluelightning: yeah doing that Aug 30 12:48:22 I know you've had a tough time getting these recipes in, but once you get used to things it gets easier I can assure you :) Aug 30 12:58:05 bluelightning: :D Aug 30 13:01:09 bluelightning: need minor help ... i have updated my local repo but when i give 'git push origin master' command it says Aug 30 13:01:10 To git@github.com:Noor-Ahsan/meta-oe.git ! [rejected] master -> master (non-fast-forward) error: failed to push some refs to 'git@github.com:Noor-Ahsan/meta-oe.git' To prevent you from losing history, non-fast-forward updates were rejected Merge the remote changes before pushing again. See the 'Note about fast-forwards' section of 'git push --help' for details. Aug 30 13:01:43 Noor: since it's your own repo/branch, you can use the -f option to force-push Aug 30 13:02:18 (be careful with that one in future though, you can of course cause huge damage if you push to the wrong repo) Aug 30 13:03:20 bluelightning: cool ... thanks ... yeah should be careful about the repo Aug 30 13:03:50 need to learn about fast-forward thingi as well Aug 30 13:04:31 a fast-forward merge just means the changes to be merged follow directly on from the ones currently in the repository Aug 30 13:04:34 it was saying something about + sign to make non-fast-forward to fast-forward Aug 30 13:06:00 hmm, there are some advanced things you can do with pushing, I don't use them myself though Aug 30 13:15:00 bluelightning: is it necessary to have HOMEPAGE in recipe .... i cant find the HOPEPAGE for debianutils Aug 30 13:15:03 HOMEPAGE Aug 30 13:15:19 Noor: it's not mandatory no Aug 30 13:15:32 bluelightning: thanks Aug 30 17:38:55 florian: Crofton denix sorry, i'm late. did you start the meeting already? Aug 30 17:45:15 I do no tthink so Aug 30 17:45:31 thinks are a little hectic at my end, we received new hw last night Aug 30 17:45:33 no not that I'm aware of Aug 30 17:45:43 oh well Aug 30 17:46:15 I think I confused things with my email also Aug 30 17:46:21 heh, ok Aug 30 17:46:24 since we had talked about next week Aug 30 17:46:36 oops Aug 30 17:46:45 I am leaving CA tomorrow night, but will be on vacation until a week from Tuesday Aug 30 17:46:54 need to sort this out :) Aug 30 17:47:19 I'll keep an eye here though Aug 30 18:57:04 Hi everyone Aug 30 18:57:45 I am wondering is there any difference between arm-none-linux-gnueabi-* and arm-oe-linux-gnueabi-* cross compilers. Aug 30 18:58:04 Can I use one for instead of another one? Aug 30 21:01:36 03Joshua Lock  07master * r12c6f98ba7 10bitbake.git/lib/bb/ui/crumbs/tasklistmodel.py: Aug 30 21:01:37 ui/crumbs/tasklistmodel: remove unnecessary check Aug 30 21:01:37 Cheaper to set COL_INC to True regardless of whether it's already set. Aug 30 21:01:37 Signed-off-by: Joshua Lock Aug 30 21:01:47 03Joshua Lock  07master * r82845569c3 10bitbake.git/lib/bb/ui/crumbs/tasklistmodel.py: Aug 30 21:01:47 ui/crumbs/tasklistmodel: loop optimisation in include_item() Aug 30 21:01:47 Rather than trying to iterate the model to find whether the item is already Aug 30 21:01:47 included and then iterate the model again to find the items path attempt to Aug 30 21:01:47 find the path first and if the path is found test whether the COL_INC of Aug 30 21:01:47 the row is set. Aug 30 21:01:48 Signed-off-by: Joshua Lock Aug 30 21:01:49 03Joshua Lock  07master * rd245053626 10bitbake.git/lib/bb/ui/crumbs/tasklistmodel.py: Aug 30 21:01:49 ui/crumbs/tasklistmodel: optimise find_path_for_item() Aug 30 21:01:49 Rather than calling get_path() for each iterated value use the get_value() Aug 30 21:01:50 method to lookup the COL_NAME value and only call get_path() for a match. Aug 30 21:01:50 This should save some time by potentially removing N-1 calls to get_path() Aug 30 21:01:51 from the loop. Aug 30 21:01:51 Signed-off-by: Joshua Lock Aug 30 21:01:52 03Joshua Lock  07master * r85c5b8b3b9 10bitbake.git/lib/bb/ui/crumbs/tasklistmodel.py: Aug 30 21:01:53 ui/crumbs/tasklistmodel: don't add empty entries to COL_BINB Aug 30 21:02:49 loop which modifies the binb column. Aug 30 21:02:49 Fixes [YOCTO #1420] Aug 30 21:02:49 (1 lines omitted) Aug 30 22:05:05 I am wondering is there any difference between arm-none-linux-gnueabi-* and arm-oe-linux-gnueabi-* cross compilers? Aug 30 22:07:56 i think that there is no difference Aug 30 22:12:17 is there a simple way to remove a problem package from a build? Aug 30 22:12:36 core-image-lsb is including screen… but it's not cross compiling properly atm Aug 30 22:12:44 and i want to just move on for a bit and ignore it for now Aug 30 22:12:55 but i dont want to change core-image-lsb besides that one bit Aug 30 22:31:19 Not that I know of Aug 30 22:31:45 You can use bitbake -k to continue after the failing build, but the final image build step will still fail Aug 30 22:33:02 housel: eh i already started hacking configure.in Aug 30 22:33:11 not how i wanted to spend my time Aug 30 23:23:43 thank you mike111, I have just recognized your answer Aug 30 23:24:43 I thought so **** ENDING LOGGING AT Wed Aug 31 02:59:57 2011