**** BEGIN LOGGING AT Mon Jun 25 02:59:58 2012 Jun 25 07:22:15 hello all Jun 25 07:22:53 while compiling bitbake midori for ARM ... I m getting errors.... /bin/grep: /mnt/sda3/OPENEMBEDDED/build/tmp/sysroots/armv4t-angstrom-linux-gnueabi/usr/lib/libiconv.la: No such file or directory Jun 25 07:22:54 | /bin/sed: can't read /mnt/sda3/OPENEMBEDDED/build/tmp/sysroots/armv4t-angstrom-linux-gnueabi/usr/lib/libiconv.la: No such file or directory Jun 25 07:23:01 can anyone help me... Jun 25 08:07:59 good morning Jun 25 08:30:20 hi mckoan Jun 25 09:21:10 hi Jun 25 09:21:34 how to convince OE that package-r2 should be rebuilt nevermind of state of tmpdir? Jun 25 09:30:26 hrw: -c cleansstate will delete it and its sstate Jun 25 09:31:16 thanks Jun 25 09:31:25 I just removed stamps for now Jun 25 09:33:55 morning all Jun 25 09:34:15 ok, time to migrate my hacks into bbappends Jun 25 09:35:03 will mylayer/recipes/package/files/file-from-src-uri be used instead of oecore one? Jun 25 09:35:49 hrw: if you extend FILESEXTRAPATHS within the bbappend, yes Jun 25 09:36:46 thanks Jun 25 09:39:27 cool, works Jun 25 09:42:23 next question: best way to handle PR in own layers? PR_append = ".hrw1"? Jun 25 09:42:58 cause setting PR directly will not work with changes in parent layer, and PR += ".hrw1" adds space first Jun 25 09:49:06 ~curse you lighttpd in meta-oe! Jun 25 09:49:06 May you be reincarnated as a Windows XP administrator, you lighttpd in meta-oe! ! Jun 25 09:50:24 heh, this curse is stronger and stronger Jun 25 09:51:06 how layering handle situation when >1 layer has bbappend for same recipe? Jun 25 09:52:06 hrw: the bbappends are just appended in layer priority order Jun 25 09:52:08 stacks or take highest priority one? Jun 25 09:52:22 we lack uninherit then ;( Jun 25 09:52:27 yo dawg, we herd yo like layers, so we put more layers in yo layers Jun 25 09:53:12 hrw: that would be impossible with the way the parser works; the question is what is it that you want to uninherit specifically? Jun 25 09:53:28 I would like to kill 12 recipes in meta-oe due to their use of 'inherit systemd' Jun 25 09:53:31 bluelightning: systemd Jun 25 09:53:46 hmm, why am I not surprised Jun 25 09:53:53 it's so cool technology! Jun 25 09:54:31 as I said last week there are patches on the list to move it out to a separate layer Jun 25 09:54:54 until then it's probably easiest to just rip out the stuff you don't want and have your own branch :/ Jun 25 09:55:18 sure Jun 25 09:55:24 but thanks for reminder Jun 25 09:55:32 rather unsatisfactory situation, I know :( Jun 25 09:57:39 ok, now I see more or less how to handle that Jun 25 09:57:41 thanks Jun 25 09:57:48 * hrw -> out Jun 25 12:03:03 hi all Jun 25 12:19:53 someone can help me with a "g_mutex_lock" error? Jun 25 12:20:22 http://paste.ubuntu.com/1058991/ Jun 25 12:30:07 ok, wordpress runs fine on hrw-lamp-image ;) Jun 25 12:30:47 meta-wordpress? :) Jun 25 12:30:55 silvio__: you built gobject-introspection-1.29.17 with different options then system glib Jun 25 12:31:28 silvio__: but no ideas currently for fix Jun 25 12:31:40 Hello ! I have a probleme in bitbake compilation : http://pastebin.com/CNau5CTU Jun 25 12:31:47 dm8tbr: meta-hrw rather Jun 25 12:32:25 asilhol: that's openembedded classic. I do not know who supports it Jun 25 12:32:41 thanks hrw Jun 25 12:33:00 could be usefull cleanand rebuild? Jun 25 12:33:07 could be usefull clean and rebuild? Jun 25 12:34:12 hrw : do you talk to me ?* Jun 25 12:34:23 asilhol: yes Jun 25 12:34:42 asilhol: consider switch to oe-core? Jun 25 12:35:01 silvio__: you may end with same problem Jun 25 12:35:13 ok Jun 25 12:36:07 but, does'it work in AT91SAM ? Jun 25 12:39:30 hrw: could u suggest me some wayu to walk ? (yes i know google :P ) Jun 25 12:41:11 asilhol: as target? with some work it will Jun 25 12:41:31 hrw: I understand there are still projects out there using oe-classic or based on it. After one year I'd expect everyone has moved. Imho oe-classic should be ro for there is no support, just some rare patch. Jun 25 12:41:32 silvio__: do build in clean chroot to get rid of system libraries Jun 25 12:42:14 hrw : ok, great, i will try it Jun 25 12:43:24 asilhol: you may need to create meta-at91 (if such not exists) and copy there current kernel/bootloaders and adapt recipes to new world. Jun 25 12:47:01 hrw: is the same if I delete the output directory? Jun 25 12:47:09 hrw : is there tutorial to help me ? Jun 25 12:47:18 silvio__: probably would be Jun 25 12:48:02 asilhol: will have to add few variables for licensing stuff - bitbake will fail with some infos Jun 25 12:48:15 asilhol: for me oe-core is new world too ;) Jun 25 12:48:32 nevermind that I got r/w rights over 8 years ago Jun 25 12:49:17 hrw: i didt it already and i finish in the same error, now i do not now how to unlock this , i try another time Jun 25 12:50:04 oe-core is better, but my hardware is supported until now only with classic :( Jun 25 12:52:44 all we really need is for someone to step up to maintain a layer for the machine Jun 25 17:03:48 hm multi machine stuff still dont works Jun 25 17:04:03 now all the packages get recompiled with the same arch Jun 25 17:04:05 *sigh* Jun 25 17:04:32 hm or its uclibc Jun 25 19:05:01 khem hm Jun 25 19:05:03 WARNING: For recipe libgcc, the following files/directories were installed but not shipped in any package: Jun 25 19:05:06 WARNING: /usr/lib/arm-angstrom-linux-uclibceabi/4.7.2/include Jun 25 19:05:08 WARNING: /usr/lib/arm-angstrom-linux-uclibceabi/4.7.2/include/unwind.h Jun 25 19:05:25 woglinde: I havent looked into it in detail Jun 25 19:05:37 but it seems it should be packaged into libgcc-dev Jun 25 19:06:01 or probably be deleted if gcc-dev is already providing it Jun 25 19:08:36 maybee a problem with the multi-machine build Jun 25 19:09:10 changing machine here with the same arvm still lets recompile all packages Jun 25 19:09:19 args armV Jun 25 19:12:54 not related to multimachines Jun 25 19:14:43 okay Jun 25 19:27:10 RP_: would it be worth to submit a patch for bitbake so that ".*://.*/.* file://${DL_DIR}/../local/" like mirrors are working again (after mirror rewrite, protocol part can not be a regexp anymore)? or are such mirror specifications unsupported? Jun 25 20:55:56 has anyone successfully built scipy through bitbake? Jun 26 01:37:09 is anybody familiar with the gcc-4.5 recipe? Jun 26 01:38:58 i get lots of errors which all look like "error: .libs/f2c_specifics.o uses VFP register arguments, .libs/libgfortran.so.3.0.0 does not" Jun 26 01:40:12 FWIW, I've successfully compiled Angstrom, and everything else seems fine (so far) Jun 26 02:04:00 nstiurca: it means you have calling convention incompatible here Jun 26 02:04:21 nstiurca: can you see -mfloat-abi is set to ? Jun 26 02:04:32 on cmdline for the package you are compiling **** ENDING LOGGING AT Tue Jun 26 02:59:58 2012