**** BEGIN LOGGING AT Tue Feb 05 02:59:59 2013 **** ENDING LOGGING AT Tue Feb 05 05:35:20 2013 **** BEGIN LOGGING AT Tue Feb 05 06:02:19 2013 Feb 05 08:03:59 Hello eveybody. Does anyone tried to run strace with cortex-m3 cpu ? it seems that strace doesn't understand well cpu registers.... Feb 05 08:17:37 good morning Feb 05 08:47:49 morning Feb 05 09:14:32 silly question, anybody know a tablet with ubuntu preinstalled? Feb 05 09:17:48 mckoan: no idea Feb 05 09:18:15 mckoan: found one "coming soon": http://store.davincimt.com/tablet-nibbio-fullhd/Coupon-Nibbio-FullHd Feb 05 09:22:18 morning all Feb 05 09:35:20 Zagor: thx, I'd prefer something from a well known company https://wiki.ubuntu.com/ARM/TabletList Feb 05 09:37:00 morning Feb 05 09:39:34 http://pastebin.com/vdB8JLG4 <= any idea why i get "configure.in:1: error: possibly undefined macro: dnl" ? Feb 05 09:39:59 the recipe only does an "inherit autotools" Feb 05 10:36:57 http://marcin.juszkiewicz.com.pl/2013/02/05/fosdem-2013/ Feb 05 10:37:07 dnl is comment mark Feb 05 10:40:19 hrw, nice, thank you! Feb 05 10:44:29 | ../lib/.libs/librpm.so: undefined reference to `__secure_getenv' Feb 05 10:49:30 hello Feb 05 10:53:41 Is it possible to check some variables after parsing ? Feb 05 10:57:40 ascor_: bitbake something -e Feb 05 11:06:26 Ah, I did'nt understood that all variables were in the environment... Feb 05 11:06:43 Is it normal that I also get python source code ? Feb 05 12:07:58 morning all Feb 05 13:13:01 vivijim: stop flooding! Feb 05 15:24:53 variable TOOLCHAIN_OPTIONS should be set by default in do_compile() ? Feb 05 15:25:30 grep and you'll see where it's defined and where it's used. Feb 05 16:05:23 khem, ping? Feb 05 18:31:22 does anyone knows a bitbake-mode for emacs? Feb 05 18:54:22 someone should add wget --no-check-certificate 'https://bugs.gentoo.org/attachment.cgi?id=146294' -O ../openembedded-core/meta/recipes-extended/screen/screen-4.0.3/header_guards.patch Feb 05 18:54:55 (https://bugs.gentoo.org/show_bug.cgi?id=168281 for the details) Feb 05 20:13:35 Wondering if OE is a good way to get an ARM system up with web/database server and one QT gui application. Not finding much via google yet. Feb 05 20:18:53 JohnFarnark, sounds easonable Feb 05 20:18:56 reasonable Feb 05 20:21:08 Seemed to be - at the point of fumbling in the dark to figure out what options are. Looking to build a minimalist image for a TI Sitara based control system. Feb 05 20:29:01 basically, start and ask questions L( Feb 05 20:29:05 :) Feb 05 20:30:38 I did find this link on the OE site... seems that it might refer to a old workflow (Years mentioned are about 2009, and Ubuntu 8.04) Feb 05 20:30:42 http://free-electrons.com/doc/openembedded.pdf Feb 05 20:40:59 Anyone have any thoughts on the behavior and output format of https://gist.github.com/kergoth/4710876? Crofton|work, blindvt`? Feb 05 20:41:13 * kergoth feels this isn't quite right, but can't quite figure out what he wants to see instead Feb 05 20:41:39 * kergoth ponders Feb 05 20:43:39 kergoth looks nice I would say Feb 05 20:43:42 gi effem Feb 05 20:43:46 hi even Feb 05 20:44:37 hhi woglinde, all Feb 05 20:44:46 I'm worried the amount of output could get unwieldy with certain queries Feb 05 20:44:49 oops, I'm stuttering Feb 05 20:44:55 hey eFfeM, woglinde Feb 05 21:17:56 very interesting Feb 05 21:39:16 calling it a day, nite all Feb 05 21:39:27 nite effem Feb 05 21:45:13 I am going too Feb 05 22:31:06 * kergoth ponders Feb 05 22:32:09 poll: is it more useful to have a subcommand to search recipes/provides, and another to search packages/rprovides/packages_dynamic, or a single command which outputs both, distinguished in its output? I want to avoid subcommand proliferation, but we also want to keep usability, which means we don't want the output to be too complex or confusing Feb 05 22:45:07 hi guys. potentially newbie question here .. Feb 05 22:45:45 i need to replace something in RDEPENDS_${PN} .. but I can't quite seem to nail down what syntax exactly is required to do something like RDEPENDS_${PN} := ${RDEPENDS_${PN}}.replace(a,b) .. obv that doesn't work Feb 05 22:46:47 RDEPENDS_${PN} := "${@RDEPENDS.replace('a', 'b')}" Feb 05 22:46:53 use inline python syntax (${@}) Feb 05 22:47:19 so i don't need to have the _${PN} there ? Feb 05 22:47:28 that was a typo Feb 05 22:47:38 * kergoth sighs Feb 05 22:47:52 i swear i've tried that, but will do again Feb 05 22:47:57 i may have tried something else Feb 05 22:48:02 RDEPENDS_${PN} = "${@d.getVar('RDEPENDS_${PN}', True).replace('a', 'b')}" Feb 05 22:48:11 there are a ton of example uses of ${@} in the metadata, including this sort of thing Feb 05 22:48:20 erm, forgot the := that time, but you get the idea, i hope Feb 05 22:48:53 if you in tend to just remove a word from a space separated list of words, so if you want to remove an rdepend, rather than really doing a replacement within an rdepend, this isn't the best approach Feb 05 22:49:06 in this case, i am just removing Feb 05 22:49:09 in that case, use oe_filter_out Feb 05 22:49:11 grep for uses of it Feb 05 22:50:46 .. i guess this comes up with a new question, then. is "d" a special ? Feb 05 22:50:54 d is the metadata Feb 05 22:51:06 a DataSmart object, the class is defined in bitbake/lib/bb/data_smart.py Feb 05 22:51:13 d is used everywhere python is used inthe metadata Feb 05 22:51:36 anonymous python functions, event handlers (indirectly, the name is actually e.data there, as its a member of the event object), inline python Feb 05 22:51:46 it's made available by bitbake itself Feb 05 22:52:12 that would be handy knowledge Feb 05 22:53:57 doesn't seem to work either way :-S Feb 05 22:54:59 d.getVar() returns a NoneType Feb 05 22:55:00 "doesn't seem to work" tells me nothing. i use oe_filter_out in tons of bbappends across 5+ layers Feb 05 22:55:13 ah, right, hold on a second Feb 05 22:55:19 my apologies, i'm getting to the errors :) Feb 05 22:55:22 this is a special case, due to the variable reference in the variable name Feb 05 22:55:31 bitbake is expanding the ${PN} ont he right side before it calls getVar Feb 05 22:55:48 there's no way to escape it directly (that is, \${PN} won't cause bitbake ot not expand it) Feb 05 22:56:29 RDEPENDS_${PN} := "${@d.getVar('RDEPENDS_$' + '{PN}', True).replace('foo', 'bar')}" Feb 05 22:56:34 for example Feb 05 22:56:47 not pretty, i know. would be nice if bitbake just had -= Feb 05 22:56:58 sadly its less trivial than you'd think Feb 05 22:58:17 That gives me SyntaxError: EOL while scanning string literal (RDEPENDS_${PN}[:=], line 1) Feb 05 22:58:51 you're absolutely right it would be nice to have, and yes it is less trivial than one would think. and once you define string subtraction, some people want it to work differently Feb 05 22:59:56 thats the thing, if you try to do an immediate -=, then you have to answer questions like does it expand the current value first? Feb 05 23:01:19 EricBlade: could try d.getVar('RDEPENDS_foo', True).. pre-expanding the variable name Feb 05 23:27:50 kergoth: this is exactly the same error i'm getting, whether i use a literal or try to use the ${PN} .. http://lists.linuxtogo.org/pipermail/bitbake-devel/2011-January/000800.html **** ENDING LOGGING AT Wed Feb 06 02:59:58 2013