**** BEGIN LOGGING AT Sun Apr 11 02:59:56 2010 Apr 11 06:44:47 Hmm, I shouldn't be able to connect to here... Apr 11 06:46:24 oh? Apr 11 06:46:28 * RP__ notes for future reference, looks like IPv6 works :) Apr 11 06:46:51 * RP__ -> plane Apr 11 06:59:00 IPv6 ftw! :D Apr 11 07:02:39 So, I'm trying to write a recipe for Parrot, but during the configure Parrot tries to check your compiler by making a small executable, and running it. This is failing since the compiled-for-system, and the compiled-on-system aren't compatible (unless I'm completely missing something here). Anyone have any suggestions on how to deal with this, or is Parrot just being especially paranoid in its configure Apr 11 07:02:41 script? Apr 11 07:06:12 usually there is a way to tell configure that this is for crosscompiling Apr 11 07:06:53 dm8tbr: It's not autotools, it's just a big Perl5 script. Apr 11 07:07:17 oh, someone reinvented configure? ain't that grand... Apr 11 07:07:59 Yeah, I had a feeling this was going to involve patching Parrot's build system. :-\ Apr 11 07:08:06 well check if this is in any other embedded distro. they must have solved that. no? Apr 11 07:08:33 That's what I was asking here for. Wondering if anyone had any pointers on where to start looking. Apr 11 07:08:53 how about openwrt? Apr 11 07:09:52 Just because I'm feeling a bit dense right now, you're talking about checking openwrt itself, not Parrot on openwrt, right? Apr 11 07:10:22 I meant to check if there is a working parrot for openwrt Apr 11 07:10:49 There isn't. Apr 11 07:11:08 though I've got a hunch you'll just end up removing that run check Apr 11 07:12:16 Yeah, there aren't any recipes for Parrot at all yet. One of the core committers and I are working on fixing that. ;-) Apr 11 07:12:45 Ok. Off to fix Parrot's build system then. Apr 11 07:12:55 Thanks for the help. Apr 11 07:13:07 then better add some crosscompile switch to parrot :) Apr 11 07:13:30 would either of you know why when I build an image manually it's smaller than the one that is supplied precompiled for download? Apr 11 07:15:03 porcoesphino: I can come up with some wild-ass-guesses, but nothing I'd put any weight on. I'm very new to OE, and all this. Apr 11 07:16:11 I'll take a wild-ass guess Apr 11 07:16:17 it's better than what I have :P Apr 11 07:16:56 Oh, I'm trying to do a gumstix overo build if that changes any of your guessing :P Apr 11 07:17:56 Well, exactly how big is the size difference in the images? Apr 11 07:33:50 does someone have tkunze zaurus patches? (TKSRC from oe.dev returns 403 and 2 patches are missing checksums :/) Apr 11 07:52:39 03Martin Jansa  07org.openembedded.dev * rea62b370f4 10openembedded.git/recipes/intone/intone_svn.bb: intone: bump SRCREV for segmentation fault fix after latest EFL bump Apr 11 08:53:26 jhelwig: the prebuild gumstix images (omap3-console-image) are 35M mine is 27M Apr 11 08:53:50 *prebuilt Apr 11 09:12:15 Hmm... I'm using the below to test quickly compare file sizes and names in the image build: Apr 11 09:12:16 tar tvfj Angstrom-omap3-console-image-glibc-ipk-2010.4-test-20100411-overo.rootfs.tar.bz2 | awk '{printf "%10s %s\n",$3,$6}' | sort -k 2 > 11.txt Apr 11 09:12:53 my first build didn't even have ifup Apr 11 09:13:04 I'm closer now, 31M but still not right Apr 11 09:13:10 time to test the image again... Apr 11 09:14:01 It didn't have su either!? Shit, I didn't notice any errors come up. That's a tad worrying. Apr 11 09:53:00 dxx@dxx-desktop:~/oe$ bitbake Apr 11 09:53:00 Traceback (most recent call last): Apr 11 09:53:00 File "/home/dxx/oe/bitbake/bin/bitbake", line 30, in Apr 11 09:53:00 import bb Apr 11 09:53:00 File "/home/dxx/oe/bitbake/lib/bb/__init__.py", line 27, in Apr 11 09:53:01 import bb.msg Apr 11 09:53:01 File "/home/dxx/oe/bitbake/lib/bb/msg.py", line 50, in Apr 11 09:53:02 "Util")) Apr 11 09:53:02 File "/home/dxx/oe/bitbake/lib/bb/msg.py", line 34, in _NamedTuple Apr 11 09:53:03 Tuple = collections.namedtuple(name, " ".join(fields)) Apr 11 09:53:03 AttributeError: 'module' object has no attribute 'namedtuple' Apr 11 09:54:25 help,I don't know what's wrong. Apr 11 10:09:41 you're using a bad version of bitbake Apr 11 10:10:07 it's nolonger compatible with the current oe (as far as I can tell, but I am new to oe and git) Apr 11 10:10:19 run: Apr 11 10:10:27 cd ~/oe/bitbake Apr 11 10:10:41 git checkout 1.8.18 Apr 11 10:11:12 vb: well, if you're building for overo, not the new COM Apr 11 10:11:55 have you got a build to work before? my images are smaller than they should be. I'm wondering if its because I don't have a com.gumstix directory, but the overo tutorial doesn't use one... Apr 11 10:24:35 vb: you solve your problem? I wrote some stuff not noticing you were gone Apr 11 10:25:09 your bitbake is too recent a version, I don't think it's currently compatible with the current oe. Try running: Apr 11 10:25:29 cd ~/oe/bitbake; git checkout 1.8.18; bitbake Apr 11 10:36:03 is there some way to force bitbake ignore COMPATIBLE_MACHINE? Apr 11 10:36:40 I need it only for -c fetch (also would be nice to prettend to be every available machine to catch SRC_URI_append_machine :/ Apr 11 10:36:53 * JaMa hates recipes/linux dir now :) Apr 11 10:49:07 porcoesphino: now is ok Apr 11 10:49:15 thanks Apr 11 10:49:18 np Apr 11 10:49:48 vb: have you got a working image before? Apr 11 10:50:45 no Apr 11 11:31:56 vb: damn, I'm still having some fun setting one up myselg Apr 11 12:08:31 anyone around that might be able to help me troubleshoot a recipe i wrote? Apr 11 12:08:35 http://pastebin.com/aBkHP5Ly Apr 11 12:09:27 compilation fails for some reason with this message: Apr 11 12:09:29 | CROSS COMPILE Badness: /usr/include in INCLUDEPATH: /usr/include/fltk-1.1 Apr 11 12:09:29 | cc1plus: internal compiler error: in add_path, at c-incpath.c:425 Apr 11 12:09:29 | Please submit a full bug report, Apr 11 12:09:30 | with preprocessed source if appropriate. Apr 11 12:09:30 | See for instructions. Apr 11 12:09:30 | scons: *** [Opcodes/ampmidid.os] Error 1 Apr 11 12:09:30 | scons: building terminated because of errors. Apr 11 12:09:31 | FATAL: scons build execution failed. Apr 11 12:21:50 for reference, i'm trying to get Csound onto my Overo COM Apr 11 12:23:48 i would greatly appreciate it if anyone could help me sort this out! Apr 11 12:50:18 03Martin Jansa  07org.openembedded.dev * r9c58805777 10openembedded.git/recipes/obsolete/ (7 files in 6 dirs): Apr 11 12:50:18 obsolete: add include files to make recipes parse-able Apr 11 12:50:18 Signed-off-by: Martin Jansa Apr 11 12:50:20 03Martin Jansa  07org.openembedded.dev * r6ea9a5e807 10openembedded.git/recipes/linux/ (8 files in 3 dirs): Apr 11 12:50:20 linux-openmoko-2.6.32: add patches from qtmoko, move defconfig to kernel tree Apr 11 12:50:20 Signed-off-by: Martin Jansa Apr 11 12:50:21 03Martin Jansa  07org.openembedded.dev * r692072ca86 10openembedded.git/recipes/openh323/openh323_1.13.5.4.bb: Apr 11 12:50:21 openh323_1.13.5.4: add missing checksum Apr 11 12:50:21 Signed-off-by: Martin Jansa Apr 11 12:50:22 03Martin Jansa  07org.openembedded.dev * r0189e3a0d7 10openembedded.git/recipes/uclibc/ (6 files): Apr 11 12:50:22 uclibc: add checksums to recipes Apr 11 12:50:23 Signed-off-by: Martin Jansa Apr 11 12:50:23 03Martin Jansa  07org.openembedded.dev * raf977daa11 10openembedded.git/recipes/libelf/libelf_0.8.6.bb: Apr 11 12:50:24 liberf_0.8.6: use same name SRC_URI param as those 2 URLs provide the same file Apr 11 12:50:24 Signed-off-by: Martin Jansa Apr 11 13:40:44 any idea why a boot process could hang right after 'Freeing init memory: ...K'? Apr 11 13:40:49 no ABI mismatch Apr 11 13:54:28 mickeyl: busybox instead of sysvinit? Apr 11 13:57:18 hmm Apr 11 13:57:20 possible Apr 11 13:57:26 let me attach a serial cable Apr 11 13:57:30 thank god i have one for this machine Apr 11 13:59:44 aah Apr 11 13:59:48 it was common cause on gta0* with broken u-a it got /sbin/init targeting busybox Apr 11 14:00:03 qi mounts the rootfs ro Apr 11 14:00:19 need to tweak the command line arguments Apr 11 14:04:55 hmm Apr 11 14:04:58 no getty on serial Apr 11 14:04:58 :/ Apr 11 14:05:42 INIT: Id "S" respawning too fast: disabled for 5 minutes Apr 11 14:05:42 INIT: no more processes left in this runlevel Apr 11 14:05:43 hmm Apr 11 14:42:33 is here tkunze? or someone with his zaurus patches? http://www2.informatik.hu-berlin.de/~tkunze/zaurus/patches/ says 403 for everything :/ Apr 11 14:58:09 any kirkwood maintainer? linux-kirkwook_2.6.31 recipe is requesting non-existent SRCREV, can I bump it to some which exists? Apr 11 14:59:37 i guess any existing one sounds better than a non-existing one Apr 11 15:01:45 but I cannot test it on device.. Apr 11 15:02:00 * JaMa just running download test for all machines/SRC_URIs :) Apr 11 15:24:16 It's been 6 months or more since I last built and tested a kirkwood kernel... Apr 11 15:25:11 You might ask the angstrom folks; I know angstrom also runs on the Sheevaplug (which is the kirkwood I know works with OE) Apr 11 15:26:53 it's strange.. I changed it.. and now it says it doesn't exist again.. I'll probably ignore that and leave it as it was Apr 11 19:28:39 mickey|bbl: "INIT: Id "S" respawning too fast:" == getty running for not existing /dev/tty*** Apr 11 19:28:43 morning Apr 11 19:29:13 JaMa: MACHINE=sheevaplug bitbake -cfetch linux-kirkwood-2.6.31 also fails? Apr 11 19:35:57 hrw: I used SOURCE_MIRROR_FETCH to ignore COMPATIBLE_* Apr 11 19:36:16 SOURCE_MIRROR_FETCH? first heard Apr 11 19:36:37 hrw: maybe the downloaded .tar.gz with git clone is somehow broken a bit Apr 11 19:37:08 hrw: I've found it in base.bbclass (after removing ^COMPATIBLE_ from whole oe.dev tree :) Apr 11 23:08:16 hmm Apr 12 00:16:07 03Michael 'Mickey' Lauer  07org.openembedded.dev * r61d6b15d54 10openembedded.git/recipes/freesmartphone/ (zhone2/zhone2 zhone2_git.bb): zhone2: install init script Apr 12 00:37:43 hmm Apr 12 00:37:57 * kergoth looks over some old prototype code he did for variable reference tracking **** ENDING LOGGING AT Mon Apr 12 02:59:56 2010