**** BEGIN LOGGING AT Tue Mar 01 02:59:57 2011 Mar 01 06:55:46 gm Mar 01 06:58:54 khem would it be ok if I push a patch for nios2 gcc? this is for 4.1.2 and is behind a SRC_URI_append_nios2 so no one else should be affected. or should this still go through a review? The patch is an update to the nios2 version you pointed me at last week, issue for review is that a lot of new files are added, making the patch about 500 K Mar 01 07:15:45 hello anyone around ? Mar 01 07:21:06 my postinst scripts don't do anything Mar 01 07:26:37 NOONE ? Mar 01 07:43:35 good morning Mar 01 07:46:34 03Andrea Adami  07org.openembedded.dev * r276d6c3df7 10openembedded.git/recipes/linux/linux-kexecboot.inc: Mar 01 07:46:34 linux-kexecboot.inc: fix CMDLINE_DEBUG tag Mar 01 07:46:34 * bump INC_PR Mar 01 07:46:34 Signed-off-by: Andrea Adami Mar 01 07:50:17 03Koen Kooi  07org.openembedded.dev * r497a5b3641 10openembedded.git/recipes/alsa/alsa-lib_1.0.24.1.bb: Mar 01 07:50:17 alsa-lib: add 1.0.24.1, gives us UCM Mar 01 07:50:17 Signed-off-by: Koen Kooi Mar 01 07:50:24 03Koen Kooi  07org.openembedded.dev * r679b4bb094 10openembedded.git/recipes/alsa/ (2 files in 2 dirs): Mar 01 07:50:24 alsa-utils: add 1.0.24.2 Mar 01 07:50:24 Signed-off-by: Koen Kooi Mar 01 08:56:09 eFfeM_work: hmmm 500k may be if there are some recipe changes that needs attention otherwise its fine Mar 01 08:56:26 mckoan, can you tell me why the glibc-gconv all iso files are not in final rootfs Mar 01 09:03:49 khem: I know it's probably better to resolve in oe-core, but what do you think about adding -g to CFLAGS in sane-toolchain-*? Mar 01 09:04:46 good thing Mar 01 09:04:58 send patch Mar 01 09:05:01 ok Mar 01 09:05:21 khem: with -ggdb2 or without any -ggdb? Mar 01 09:05:38 only -g Mar 01 09:05:48 morning Mar 01 09:12:46 khem: sent to ml Mar 01 09:38:41 Hi, I am trying to write my first recipe. my do_configure() simple calls ./configure I am getting a configuration error "If you meant to cross compile, use `--host'." Any pointers? http://pastie.org/1620040 Mar 01 09:39:08 inherit autotools Mar 01 09:39:17 no need for run configure manually Mar 01 09:40:23 ok.. i tried that but was not winning... let me try again.... then i do not need to define do_configure() do_compile() or do_install() correct? Mar 01 09:41:39 03Martin Jansa  07master * re493739240 10openembedded.git/recipes/ (shr/libphone-ui_git.bb tasks/task-shr-minimal.bb): Mar 01 09:41:39 libphone-ui: add libphone-ui-config to libphone-ui RDEPENDS instead of task-shr-minimal-apps Mar 01 09:41:39 Signed-off-by: Martin Jansa Mar 01 09:41:49 also, how do I add dependencies? what is the variable name? DEPENDS += ? Mar 01 09:42:17 DEPENDS = Mar 01 09:42:22 thx Mar 01 09:42:34 DEPENDS += you only need then using requie blafile Mar 01 09:43:18 ok so if i have 2 dependencies do I comma separate or two DEPENDS = statements? Mar 01 09:43:39 guyvdb_: space separated Mar 01 09:43:44 k Mar 01 09:43:50 guyvdb_: read existing recipes for examples Mar 01 09:44:10 I have been... but it is a little confusing... think i am getting it Mar 01 09:44:27 ? Mar 01 09:44:43 are your software based on autotools or not? Mar 01 09:44:49 yes Mar 01 09:45:02 inherit autotools Mar 01 09:45:27 yea got that from your last msg.. trying now Mar 01 09:45:34 if it gives error pastebin them Mar 01 09:45:41 ok thx Mar 01 09:59:09 morning all Mar 01 10:00:44 hi bluelightning Mar 01 10:03:22 I see there is a variable SOURCEFORGE_MIRROR is there an APACHE_MIRROR? Mar 01 10:03:39 or is that defined in the recipie Mar 01 10:03:41 grep is your friend Mar 01 10:03:47 true :) Mar 01 10:03:47 guyvdb_: check conf/bitbake.conf for list of MIRRORS Mar 01 10:04:45 got it thx Mar 01 10:07:41 Please be patient with me. I am new to this. If I have DEPENDS = boost and I want a particular boost do I do PREFERS_boost = "x.x.x" Mar 01 10:08:04 isnt possible Mar 01 10:08:18 defint it in your local.conf for now Mar 01 10:08:23 ok Mar 01 10:08:41 PREFFERED_VERSION_boost = "344.3434" Mar 01 10:08:49 gm gnutoo Mar 01 10:10:15 guyvdb_: and PREFFERED_VERSION_boost is not for your recipe.. but needs to be set for whole distribution if you really need different boost version than default Mar 01 10:10:29 ok... Mar 01 10:12:01 03Koen Kooi  07org.openembedded.dev * rd70aabfb6a 10openembedded.git/recipes/opkg/opkg.inc: Mar 01 10:12:01 opkg: move to r609 Mar 01 10:12:01 This fixes a few segfaults Mar 01 10:12:01 Signed-off-by: Koen Kooi Mar 01 10:12:23 grg: ping Mar 01 10:51:19 03Stefan Schmidt  07master * r0b3127a0d3 10openembedded.git/recipes/llvm/llvm.inc: Mar 01 10:51:19 llvm.inc: Empty TOOLCHAIN_OPTIONS to fix the build. Mar 01 10:51:19 This fixes a problem introduced by ebd905685a339814c36bf658a8e58bf8f457f14e Mar 01 10:51:19 Signed-off-by: Stefan Schmidt Mar 01 10:51:19 Acked-by: Khem Raj Mar 01 11:11:50 hi pb Mar 01 11:11:53 hi woglinde Mar 01 11:12:27 I think i found a bug in the boost 1.4.5 recipie. I could not get it to compile until I added a DEPENDS bzip2 to it Mar 01 11:12:49 hm yes Mar 01 11:12:52 might bee Mar 01 11:13:20 what should I do? Leave it? Create a patch ? Mar 01 11:13:36 create patch send it to ml Mar 01 11:13:40 or Mar 01 11:13:45 let me do add the depence Mar 01 11:13:56 ok so small please add it Mar 01 11:30:39 03Mike Westerhof  07org.openembedded.dev * rf2e8329c76 10openembedded.git/recipes/opkg/opkg_svn.bb: Mar 01 11:30:39 SlugOS: opkg_svn.bb -- remove SlugOS-specific patch that no longer applies. Mar 01 11:30:39 This patch has been pushed upstream. No PR bump required. Mar 01 11:34:36 ugh Mar 01 11:34:44 boost suffers from bad rpath Mar 01 11:35:11 how is that fixed? Mar 01 11:35:45 can I look into it with some guidance Mar 01 11:36:46 hm I will fix it Mar 01 11:36:55 stupid linker options Mar 01 11:37:22 k thx... will you push it when done? Mar 01 11:39:36 03Martin Jansa  07master * rb528b222cb 10openembedded.git/recipes/shr/initscripts-shr_0.0.1.bb: Mar 01 11:39:36 initscripts-shr: fix missing quote Mar 01 11:39:36 Signed-off-by: Martin Jansa Mar 01 11:39:50 guyvdb_ sure Mar 01 11:54:25 hi obi Mar 01 11:54:56 hi! Mar 01 12:02:59 woglinde: I am trying to build netscape portable runtime can you look at this pastie for me http://pastiebin.com/?page=p&id=4d6c8d8a198fd Mar 01 12:04:19 I think the error is : |gnu-configize: cannot create ${srcdir}/build/autoconf: No such file or directory Mar 01 12:04:28 hm dont we have a recipe for it? Mar 01 12:04:56 I did not see one... Mar 01 12:05:06 this nspr? Mar 01 12:05:28 guyvdb_: it's in mozilla/ Mar 01 12:05:35 ahh Mar 01 12:05:50 ok thx Mar 01 12:05:52 (confusingly) Mar 01 12:06:09 it is a dependency i need... thought I would have to build it :) Mar 01 12:06:31 woglinede disregard pastie Mar 01 12:06:39 sure Mar 01 12:06:49 but the mozilla stuff is a bit outdated Mar 01 12:06:59 I think there are newer versions Mar 01 12:07:30 4.8.7 is latest version i will look at what we have and see if i can get it to latest Mar 01 12:13:21 args Mar 01 12:13:27 include/boost is 46 mb Mar 01 12:13:29 wtf Mar 01 12:14:41 woglinde: hi Mar 01 12:14:49 woglinde: what days will you be at cebit? Mar 01 12:15:34 thursday to satuarday Mar 01 12:15:44 btw its on the oe cebit page Mar 01 12:16:31 pb ping Mar 01 12:17:11 woglinde: the cebit page is to ugly to read. :) Mar 01 12:17:35 * stefan_schmidt could as well be lazy ;) Mar 01 12:17:50 what? Mar 01 12:17:51 woglinde: So we will see us on thrusday. Good. Mar 01 12:17:54 they is small Mar 01 12:18:05 I mean ugly not big :) Mar 01 12:18:15 make it better Mar 01 12:18:20 it has all infos Mar 01 12:18:23 Well, I didn't even bother to have a look :P Mar 01 12:18:27 03Martin Jansa  07master * r8322ddb9ee 10openembedded.git/recipes/shr/ (3 files in 3 dirs): Mar 01 12:18:27 libphone-ui: move config files back to libphone-ui repo Mar 01 12:18:27 * install right one in do_install_append if machine specific is available Mar 01 12:18:27 * this helps keeping OE config and libphone-ui sources consistent Mar 01 12:18:27 Signed-off-by: Martin Jansa Mar 01 12:18:35 hm btw. not the one that mickeyl is ill Mar 01 12:18:38 woglinde: ah, sorry, I meant the official cebit site Mar 01 12:18:42 not the one from OE Mar 01 12:18:46 oh okay Mar 01 12:18:47 lol Mar 01 12:18:54 hall 2 Mar 01 12:19:06 yeah, been there yesterday Mar 01 12:19:12 hm maybee I should add it Mar 01 12:19:17 yesterday? Mar 01 12:19:19 Was hanging out with raster and some other E guys Mar 01 12:19:53 just to get known to the booth and the linux media guys Mar 01 12:19:54 building up booths? Mar 01 12:19:58 no real cebit yesterday Mar 01 12:20:11 so you will help out robert Mar 01 12:20:14 everything was up already Mar 01 12:20:44 woglinde: I may be able to help out some hours on thursday, but not the full day Mar 01 12:21:04 every hour helps robert Mar 01 12:21:19 That is one of the problem I see if we plan to go somewhere without enough people acknowledge that they will be around... Mar 01 12:22:10 I really hate standing at a booth and normally I just don't do it. But I should be able to free Robert some hours to get food and relax a bit Mar 01 12:23:10 fine Mar 01 12:23:34 * stefan_schmidt is away for lunch, bbl Mar 01 12:23:50 yeah Mar 01 12:23:52 till later Mar 01 12:27:26 boost pushed Mar 01 12:27:41 03Henning Heinold  07org.openembedded.dev * ref37c1c2fe 10openembedded.git/recipes/boost/ (3 files in 2 dirs): Mar 01 12:27:41 boost: fix rpath isusse with version 1.45 and add bzip do depends Mar 01 12:27:41 * add a patch to rip out -Wl,R stuff from gcc jam defs Mar 01 12:27:41 * add bzip to depends as on the recipes before Mar 01 12:27:41 * remove the doubled icu depend Mar 01 12:27:41 * bump INC_PR und PR Mar 01 12:54:41 stefan_schmidt, it is more fun than it sounds, but breaks are still appreciated :) Mar 01 12:56:47 Crofton|work: I did it once for Openmoko at Linuxtag and I really don't like it. Anyway, Robert will need a break. Mar 01 12:57:33 well, go in with an open mind :) Mar 01 13:00:46 Crofton|work: no, I never do this :) Mar 01 13:03:18 crofton cebit is diffrent Mar 01 13:03:22 to much normal people Mar 01 13:03:28 so I have two goals now Mar 01 13:03:36 get RP to speak to large groups Mar 01 13:03:39 maybee the dont come to the opensource booths's Mar 01 13:03:40 hey Mar 01 13:03:44 and stefan_schmidt to work stands :) Mar 01 13:03:53 crofton and a lot of theft is on cebit Mar 01 13:03:59 what wrong? ERROR: EOL while scanning string literal (FILESPATH, line 1) while parsing on each bb file Mar 01 13:04:05 theft? Mar 01 13:04:13 crofton yes Mar 01 13:04:17 weird Mar 01 13:04:23 the steal even your underware Mar 01 13:04:33 there are many things in the owrld I do not understand Mar 01 13:04:46 its germany Mar 01 13:04:51 and big fair Mar 01 13:04:57 woglinde: what I hear from the E both of today there is a steady stream of people. Perhaps a bit different as they show fancy pixel stuff :) Mar 01 13:05:25 They should try to steal my underwear ;) Mar 01 13:05:46 :) Mar 01 13:06:08 woglinde: having a second thought about openjdk it makes sense to just go with the newer version you already fixed. I just realized that it also contains security updates Mar 01 13:06:31 yeah Mar 01 13:06:35 but not the latest Mar 01 13:06:40 I am waiting for buglabs Mar 01 13:06:57 yes and also please merge still needed bits from angstrom-java to sane-toolchain-java if those still exists Mar 01 13:07:16 woglinde: I send ken my prio list regarding the work. Thomas should have it soon. Mar 01 13:07:20 but I was able to build new openjdk without angstrom-java included Mar 01 13:07:32 jama will do Mar 01 13:07:37 JaMa|Off: its angstrom-jalimo? Mar 01 13:07:37 I am waiting for buglabs Mar 01 13:07:42 *g* Mar 01 13:07:45 woglinde: heh Mar 01 13:07:59 stefan_schmidt: ah, yes Mar 01 13:08:30 Hopefully Ken and Thomas have the details out this week. I want you guys to start soon. Mar 01 13:08:42 * stefan_schmidt is looking forward to reliable openjdk builds Mar 01 13:08:50 * JaMa|Off too Mar 01 13:08:54 you will get it Mar 01 13:08:59 *promised* Mar 01 13:09:00 That will be a relief Mar 01 13:09:17 woglinde: good :) Mar 01 13:09:41 maybee with llvm2.8 Mar 01 13:10:06 maybe, but other items have a higher prio for me. Mar 01 13:10:13 sure Mar 01 13:17:48 woglinde I see you pushed boost... thx Mar 01 13:17:56 03Koen Kooi  07org.openembedded.dev * rcce5e64822 10openembedded.git/recipes/pulseaudio/pulseaudio_git.bb: Mar 01 13:17:56 pulseaudio: add git version Mar 01 13:17:56 armv4t/armv4t patches were not forward ported, so don't use this on those archs Mar 01 13:17:56 Signed-off-by: Koen Kooi Mar 01 13:18:24 guyvdb_ sure Mar 01 13:57:28 hi kergoth Mar 01 14:00:25 morning Mar 01 14:30:22 Hello Mar 01 14:30:49 Can somebody help me bitbaking console-image for ea3250? Mar 01 14:31:00 It keeps erroring when patching the kernel Mar 01 14:31:14 hm whats ea3250 Mar 01 14:31:16 lets see Mar 01 14:31:23 Embedded Artist 3250 Mar 01 14:31:58 hm yeah Mar 01 14:32:13 which kernelversion it is trying to build at your side? Mar 01 14:32:14 1 out of 1 hunk FAILED -- rejects in file arch/arm/mach-lpc32xx/clocks-lpc32xx.c Mar 01 14:32:35 linux-ea3250_2.6.27.8.bb Mar 01 14:32:40 okay Mar 01 14:33:33 re jama Mar 01 14:35:10 Any idea? =) Mar 01 14:35:31 hm I will look at the patch Mar 01 14:35:56 Thanks Mar 01 14:36:23 It is the first patch Mar 01 14:36:23 Applying patch kernel-arm-2.6.27.8-lpc32xx.patch Mar 01 14:36:27 Wich fails Mar 01 14:38:23 can you go to the source directory Mar 01 14:38:33 and do quilt push -f Mar 01 14:38:39 and look at the .rej files? Mar 01 14:38:52 Sure one sec Mar 01 14:39:32 quilt is not installed, will install it Mar 01 14:40:37 prjt61@prjt61-1:~/projects/sources$ quilt push -f Mar 01 14:40:38 No patches in series Mar 01 14:41:40 that's not what he meant by source directory :) Mar 01 14:41:49 bitbake -e virtual/kernel|grep S= Mar 01 14:44:00 Ugh what is that output? Mar 01 14:45:04 I don't understand Mar 01 14:45:13 What source directory do I need to go to? Mar 01 14:47:45 the source directory where should go Mar 01 14:47:57 bitbake builds things. each recipe has a source tree Mar 01 14:48:17 Yea that is my ~/projects/sources folder Mar 01 14:48:25 huh? Mar 01 14:48:39 Or do you mean the recipes folder? Mar 01 14:48:39 no. the tree where the recipe's source archives are *extracted* Mar 01 14:48:42 not your *download* directory Mar 01 14:48:43 no Mar 01 14:48:45 not the recipes folder Mar 01 14:48:52 the tree where bitbake extracted the files for the recipe Mar 01 14:49:08 The build folder? Mar 01 14:49:24 * kergoth gives up Mar 01 14:49:27 Lol Mar 01 14:49:30 $BBPATH ? Mar 01 14:49:35 no, no, no. Mar 01 14:49:41 paste output of bitbake -e virtual/kernel|grep S= Mar 01 14:49:44 we're talking about bitbake's *OUTPUT* Mar 01 14:49:48 which goes in tmp in the build dir Mar 01 14:49:50 not *input* Mar 01 14:49:53 they're two completely different things Mar 01 14:49:57 woglinde, that is ALOT of text wich doesn't makes sense Mar 01 14:50:08 its a list of variables, key=value Mar 01 14:50:12 fine the one that's just S= Mar 01 14:50:25 that's the definition of the S variable for the virtual/kernel recipe, wh ich is the path to the extracted source tree Mar 01 14:51:43 I am in the tmp dir wich has these folders: abi_version cache deploy distro_pr pkgdata pstage saved_tmpdir stamps sysroots tmp usr work Mar 01 14:53:08 try grep "WORKDIR=" Mar 01 14:53:13 cd there Mar 01 14:53:22 and then to next linux dir Mar 01 14:56:44 I think I found it Mar 01 14:57:01 Applied patch kernel-arm-2.6.27.8-lpc32xx.patch (forced; needs refresh) Mar 01 14:57:23 now find -name "*.rej" Mar 01 14:57:47 ./arch/arm/Kconfig.rej Mar 01 14:58:36 okay Mar 01 14:58:41 ~pastebin Mar 01 14:58:42 [~pastebin] A "pastebin" is a web-based service where you should paste anything over 3 lines so you don't flood the channel. Here are links to a few : http://www.pastebin.com , http://pastebin.ca , http://channels.debian.net/paste , http://paste.lisp.org , http://bin.cakephp.org/ , http://asterisk.pastey.net/ , or install pastebinit with yum or aptitude. Mar 01 14:58:50 pastebin this file Mar 01 14:59:58 http://pastebin.com/rNQjHxTj Mar 01 15:00:37 prjt61 are you familiar with it? Mar 01 15:00:46 With pastebin? Mar 01 15:01:12 no with fixing it up yourself Mar 01 15:01:21 Not really... sorry Mar 01 15:01:27 This is the first time I work with OE Mar 01 15:01:29 okay mom Mar 01 15:01:31 It is for a school project Mar 01 15:03:15 I will fix it for you Mar 01 15:03:26 but will last a while Mar 01 15:03:41 Hmm I can help Mar 01 15:04:23 If I understand what to do offcourse Mar 01 15:04:57 seems something changed in this file upstream Mar 01 15:05:02 and you have applie it manually Mar 01 15:05:07 but I will do it for you Mar 01 15:14:10 Hmm ok Mar 01 15:14:33 Are you writing a new patch file based on the .rej or something? Mar 01 15:15:52 yes Mar 01 15:17:05 hm args Mar 01 15:17:13 stupid dos endings Mar 01 15:17:23 Lol Mar 01 15:17:35 hm not only Mar 01 15:20:31 hm its really the last patch Mar 01 15:20:38 :) Mar 01 15:20:52 Aww Mar 01 15:21:02 ERROR: EOL while scanning string literal (FILESPATH, line 1) while parsing /stuff/openembedded/recipes/freesmartphone/fso-gpsd_git.bb that happens on each bb file - latest git checkout - any hints? Mar 01 15:21:06 hes been at it all day :) Mar 01 15:21:24 prjt61 have 5 minutes Mar 01 15:21:32 You had to go through all patch files manually? Mar 01 15:21:34 Ok sure Mar 01 15:21:39 no Mar 01 15:21:41 only the last Mar 01 15:21:41 Take your time Mar 01 15:21:53 And please explain after what you did exactly Mar 01 15:22:03 So I can do it my self next time Mar 01 15:22:04 =) Mar 01 15:22:04 hm look at the commits Mar 01 15:22:17 I applied manually Mar 01 15:22:24 the patch which didnt worked Mar 01 15:25:00 Where can I check the commit? Mar 01 15:25:36 mom Mar 01 15:26:15 first test it again Mar 01 15:26:27 Ok, performing bitbake console-image Mar 01 15:27:04 what is a .lock file in my downloads directory? Mar 01 15:27:44 prjt61? Mar 01 15:27:50 Yes? Mar 01 15:27:52 I didnt checked yet in something Mar 01 15:28:01 Ahh ok Mar 01 15:28:04 jkridner|work1 file from old downloader Mar 01 15:28:09 jkridner|work1 you can delete it Mar 01 15:32:26 prjt61 pushed Mar 01 15:32:28 03Henning Heinold  07org.openembedded.dev * r524f98da7d 10openembedded.git/recipes/linux/linux-ea3250-2.6.27.8/lpc32xx-spwm.patch: linux-ea3250-2.6.27.8: refresh lpc32xx-spwm.patch, so it can apply cleanly Mar 01 15:32:39 do git pull --rebase Mar 01 15:32:43 Nice Mar 01 15:33:02 I cannot figure out how to build nspr. It is trying to run nsinstall but cannot find the binary... any ideas? http://pastiebin.com/?page=p&id=4d6cbeb55a44c Mar 01 15:33:07 bitbake -c clean virtual/kernel Mar 01 15:33:26 run bitbake virtual/kernel Mar 01 15:34:04 what the hell is nsinstall Mar 01 15:34:19 hmmm.... bitbake seems to be deleting the copy of the file I downloaded. :( Mar 01 15:34:40 I dropped a copy into the downloads folder and each time I run the build it deletes it. Mar 01 15:34:45 I guess it is something in the nspr-tools-native package... but cannot figure it out Mar 01 15:35:00 guyvdb_ hm yes Mar 01 15:35:03 that makes sense Mar 01 15:35:16 the file cannt exectuted natively Mar 01 15:35:21 if it is arm Mar 01 15:35:24 yea... Mar 01 15:35:26 maybee with qemu Mar 01 15:35:36 but not for installing something Mar 01 15:35:39 thats stupid Mar 01 15:35:46 yup Mar 01 15:36:34 okay Mar 01 15:36:35 yeah, just build an nsinstall-native if there isn't one already Mar 01 15:36:45 have to catch my son Mar 01 15:36:49 till later Mar 01 15:36:52 I'm not quite sure why mozilla need their own install, presumably due to reasons. Mar 01 15:36:57 woglinde, it's cleaning Mar 01 15:37:05 bye woglinde thx for all the help today Mar 01 15:37:12 pb_ i will try that Mar 01 15:37:16 yeah no prob Mar 01 15:37:26 pb but the dep should have build it Mar 01 15:38:13 hmm, OT, but anyone know of any good resources for improving one's writing skill, particularly as a software engineer? Mar 01 15:38:45 good question. no, but I could use some. my writing sucks. Mar 01 15:38:50 woglinde, it errors when cleaning Mar 01 15:46:58 creating a .md5 file seemed to fix the problem. Mar 01 15:47:44 Nice I just saw the failing package succeeding =) Mar 01 15:50:23 ok i am stuck. nspr creates executable in working nspr-${v}/mozilla/nsprpub/config/nsinstall and nspr-tools-native creates a native version of this in the same place but nspr fails because when nsinstall runs the arm version is at that location. Not the native version... any ideas Mar 01 15:56:09 guyvdb: it should be running the version from your native sysroot Mar 01 16:08:45 pb_ do_compile() invokes ../../../config/./nsinstall which is the arm binary Mar 01 16:09:08 that is from within the directory tmp/work/armv6-angstrom-linux-gnueabi/nspr-4.7.1-r1/nspr-4.7.1/mozilla/nsprpub/pr/include/md' Mar 01 16:09:28 * Tartarus drafts the changes since section on 2011.03 Mar 01 16:10:06 http://wiki.openembedded.org/index.php/Release-2011.03#Changes_since_the_2010.12_Release Mar 01 16:10:17 but I'm only up to 161c773156ac98e8648ca779926514ba4f5dccfd in the log Mar 01 16:10:19 guyvdb_: then I guess do_compile() needs to be patched Mar 01 16:21:14 anyone have any idea why system-tools-backends decides to re-run configure during do_compile Mar 01 16:21:20 on a Fedora machine? Mar 01 16:21:55 Hmm Mar 01 16:22:05 The libtool bit at the end? Mar 01 16:22:31 Try changing it from cp ...libtool ${S} to EXTRA_OEMAKE='LIBTOOL=${...}libtool' Mar 01 16:22:36 it feels like it found another automake on the build machine Mar 01 16:22:37 or whatever the correct incantation that others do Mar 01 16:23:10 Also, maybe we should switch it to just a do_configure_prepend() ? Mar 01 16:23:50 heh Mar 01 16:23:56 not my area of expertise ... Mar 01 16:24:00 heh Mar 01 16:24:11 I'll try the EXTRA_OEMAKE approach real quick Mar 01 16:24:11 Well, just the sed's Mar 01 16:24:17 And then the EXTRAQ_OEMAKE bi Mar 01 16:24:18 t Mar 01 16:25:06 dman lunch time already Mar 01 16:25:13 EXTRA_OEMAKE = "LIBTOOL=${STAGING_BINDIR_CROSS}/${TARGET_PREFIX}libtool" Mar 01 16:25:15 ? Mar 01 16:26:03 looks right Mar 01 16:26:20 ok Mar 01 16:26:22 bbaib Mar 01 16:26:30 running now Mar 01 16:26:32 re Mar 01 16:26:46 woglinde ? Mar 01 16:26:47 bye Mar 01 16:27:13 morning Mar 01 16:27:15 prjt61 re == rehi Mar 01 16:27:36 Sorry? Mar 01 16:27:54 yesterday when I got back I heard of a "problem" what was it woglinde? Mar 01 16:28:28 ka6sox makeing accounts in the wiki Mar 01 16:28:36 bluelightning had the problems Mar 01 16:28:58 prjt61 its simple means I am back Mar 01 16:29:15 Ahh ok Mar 01 16:29:16 bluelightning, you about? Mar 01 16:29:36 ka6sox-away: yep, was unable to create a wiki account yesterday; also noticed Administrators group seems to be empty Mar 01 16:29:37 It returns a new error Mar 01 16:30:12 ka6sox-away: or at least, that's what it looked like, perhaps the error page is broken Mar 01 16:30:16 prjt61 okay whats next? Mar 01 16:30:30 woglinde, http://pastebin.com/2u4nuRme Mar 01 16:31:16 http://wiki.openembedded.org/index.php/Release-2011.03#Changes_since_the_2010.12_Release any comments? Not done yet in reading the log 'tho Mar 01 16:32:24 prjt61 I really wonder that someone checked if this compiles Mar 01 16:33:28 Well, it did gave me an error when cleaning Mar 01 16:33:36 Could that be a/the problem? Mar 01 16:33:59 args damn Mar 01 16:34:00 no Mar 01 16:34:05 stupid me Mar 01 16:34:09 =) Mar 01 16:39:20 I messed up the patch Mar 01 16:39:30 will compile it now my self Mar 01 16:39:35 soI sure its working Mar 01 16:39:39 sorry Mar 01 16:39:48 It's ok Mar 01 16:39:55 I don't blame you Mar 01 16:39:57 =) Mar 01 16:41:38 Alot of students over here will be pleased if the kernel compiles Mar 01 16:49:56 woglinde, not to rush you or anything, but how long will it take to test the source? Mar 01 16:54:52 prjt61 you can yuick fix it yourself Mar 01 16:54:58 run the clean Mar 01 16:55:25 Cleaning Mar 01 16:55:29 edit recipes/linux/linux-ea3250-2.6.27.8/lpc32xx-spwm.patch Mar 01 16:55:36 go the end Mar 01 16:56:27 remove the double plus Mar 01 16:56:31 so its looks like Mar 01 16:56:32 +#if defined(CONFIG_SPWM_LPC32XX) Mar 01 16:56:33 +extern void lpc32xx_spwm_init(void); Mar 01 16:56:33 +#endif Mar 01 16:56:44 03Tom Rini  07master * r9e1d7a09fb 10openembedded.git/recipes/avahi/ (9 files in 2 dirs): Mar 01 16:56:44 avahi: Drop older versions Mar 01 16:56:44 We drop a number of un-pinned older versions and clean up slightly Mar 01 16:56:44 the files directory. Mar 01 16:56:44 Signed-off-by: Tom Rini Mar 01 16:56:46 when I have checked it in Mar 01 16:56:49 again Mar 01 16:57:04 do git checkout recipes/linux/linux-ea3250-2.6.27.8/lpc32xx-spwm.patch and git pull --rebase Mar 01 16:57:14 otherwise no updates will work for you anymore Mar 01 16:57:19 Ok Mar 01 16:57:23 I made the adjustment Mar 01 16:57:38 okay start build again Mar 01 16:58:20 Ok it started building Mar 01 16:58:48 Now the patch fails again Mar 01 16:58:53 hms Mar 01 16:59:27 can you pastebin the recipes/linux/linux-ea3250-2.6.27.8/lpc32xx-spwm.patch file? Mar 01 17:01:16 or only the last 10 lines Mar 01 17:01:33 http://pastebin.com/ZLeNpVmL Mar 01 17:01:37 Oh, it's the complete file =) Mar 01 17:02:34 no problem Mar 01 17:02:57 that looks sane Mar 01 17:03:04 args Mar 01 17:03:08 hm Mar 01 17:03:19 are you on an utf-8 linux? Mar 01 17:03:31 Ubuntu 10.10 Mar 01 17:03:33 maybee your editor changed the line endings Mar 01 17:03:36 hm okay Mar 01 17:03:38 strange Mar 01 17:03:40 I used gedit Mar 01 17:04:15 hm sure made the -c clean? Mar 01 17:04:35 Yes, I can do it again Mar 01 17:05:41 hm only last 10 min here Mar 01 17:06:00 than the build should be finished Mar 01 17:06:40 2 out of 2 hunks FAILED -- rejects in file arch/arm/mach-lpc32xx/sys-lpc32xx.h Mar 01 17:06:40 Patch lpc32xx-spwm.patch does not apply (enforce with -f) Mar 01 17:10:21 hm okay Mar 01 17:10:55 hm why the hell the kernelbuild needs pam Mar 01 17:23:10 hi risca Mar 01 17:23:17 Hi Mar 01 17:24:00 all fine now? Mar 01 17:24:34 hi woglinde Mar 01 17:24:42 did you have the time to fix curl? Mar 01 17:25:59 Well... the fbv package seems fine, but I got some compilation errors related with GCC>=4.4.2 Mar 01 17:26:24 http://debbugs.gnu.org/cgi-bin/bugreport.cgi?bug=5047 Mar 01 17:26:56 ? Mar 01 17:27:13 GNUtoo hm um what was wrong with curl? Mar 01 17:27:20 hm the host includes Mar 01 17:27:25 because of ldap Mar 01 17:27:27 yes Mar 01 17:27:29 okay I will check it Mar 01 17:27:32 I did that locally to fix: Mar 01 17:27:37 + deps.append("libgcrypt" + native) Mar 01 17:27:43 but that's the wrong fix Mar 01 17:27:45 as you told me Mar 01 17:28:03 yeah fix is to leave ldap complete out Mar 01 17:28:14 indeed we don't need ldap Mar 01 17:28:28 okay let me see Mar 01 17:29:03 tartarus ping Mar 01 17:33:45 pong Mar 01 17:34:51 tartarus when will you make release? Mar 01 17:35:47 3h30m or so Mar 01 17:36:17 hm ah okay Mar 01 17:36:18 need a delay? Mar 01 17:36:26 so no chances for openjdk Mar 01 17:36:56 Well, can you have it ready your today? Mar 01 17:36:59 but the curl-native fix will go in Mar 01 17:37:07 no openjdk needs more time Mar 01 17:37:37 gnutoo ping Mar 01 17:37:58 gnutoo whats the reason BBCLASSEXTEND = "native nativesdk sdk" Mar 01 17:38:10 is in the versions file Mar 01 17:38:16 instead of the common? Mar 01 17:38:25 for curl? Mar 01 17:39:42 03Henning Heinold  07org.openembedded.dev * r95e8fd9661 10openembedded.git/recipes/linux/linux-ea3250-2.6.27.8/lpc32xx-spwm.patch: linux-ea3250-2.6.27.8: fix my stupid cut and paste in lpc32xx-spwm.patch Mar 01 17:40:36 hmmm Mar 01 17:40:47 woglinde, hi Mar 01 17:41:07 no idea Mar 01 17:41:10 I just hit the bug Mar 01 17:41:16 I don't maintain curl Mar 01 17:41:23 but I can look Mar 01 17:43:46 do you want me to look? Mar 01 17:43:58 na I will fix it Mar 01 17:44:04 ok thanks Mar 01 17:44:12 what curl problem? Mar 01 17:44:28 Tartarus when you have installed ldap-dev on your host system Mar 01 17:44:41 curls tries to build ldaps support Mar 01 17:44:48 which drags in host libs Mar 01 17:44:52 in target or host or sdk? Mar 01 17:45:14 hm Mar 01 17:45:15 args Mar 01 17:45:21 damn Mar 01 17:45:32 gnutoo do you have the error at hand? Mar 01 17:48:41 I guess ldap needs to be another cookie in curl-common.inc Mar 01 17:48:46 And either disabled or enabled outright Mar 01 17:48:55 (and openldap added as a dep for target) Mar 01 17:49:22 yes Mar 01 17:49:29 s/gnutoo/ GNUtoo|laptop Mar 01 17:49:32 I disabled it Mar 01 17:49:38 because else I don't get highlighten Mar 01 17:49:52 dont what we need ldap or ldaps support for Mar 01 17:50:35 yeah, I think ldap support in curl is a fairly fringe interest Mar 01 17:50:42 I've never heard of any apps that actually use it Mar 01 17:50:50 hi all - anybody working with angstrom on omap (overo)? since recently upgrading from a late-last-year snapshot enlightenment/X isn't working right - have some artifacts on menu items and frequent crashes. Mar 01 17:51:34 wait a sec Mar 01 17:51:44 I think I don't have the error in tmpdir anymore Mar 01 17:51:47 let me look better Mar 01 17:52:07 ~pastebin Mar 01 17:52:07 [~pastebin] A "pastebin" is a web-based service where you should paste anything over 3 lines so you don't flood the channel. Here are links to a few : http://www.pastebin.com , http://pastebin.ca , http://channels.debian.net/paste , http://paste.lisp.org , http://bin.cakephp.org/ , http://asterisk.pastey.net/ , or install pastebinit with yum or aptitude. Mar 01 17:52:17 yes I'll look at the pastebin Mar 01 17:52:26 nee one with file Mar 01 17:53:16 and .ca suckz Mar 01 17:53:47 you're on a mobile device? Mar 01 17:53:53 you act like if you were on one Mar 01 17:53:59 that doesn't have Tab Mar 01 17:54:06 no Mar 01 17:55:05 hello Mar 01 17:55:32 I am trying to cross compile an app, and have problems with pkg-config Mar 01 17:55:58 http://www.pastie.org/pastes/1589070/text Mar 01 17:56:04 the .pc files in sysroots/armv7a-angstrom-linux-gnueabi/usr/lib/pkgconfig/ have paths like /usr/lib/ Mar 01 17:56:22 but actually it should be something like sysroots/armv7a-angstrom-linux-gnueabi/usr/lib/ Mar 01 17:56:43 how does OE handle this? does it prepend a path to the pkg-config ones? Mar 01 17:56:49 dv_, try running pkg-config Mar 01 17:57:03 like pkg-config --libs --cflags in the devshell Mar 01 17:57:09 for having the devshell do Mar 01 17:57:13 bitbake devshell Mar 01 17:57:16 and look in: Mar 01 17:57:17 hm Mar 01 17:57:25 dv thats okay otherwise no recipe would build Mar 01 17:57:39 does OE use chroot? Mar 01 17:57:44 tmpdir deploy/addons Mar 01 17:57:46 dv no Mar 01 17:57:49 sysroot Mar 01 17:57:54 no but there is a devshell Mar 01 17:57:57 learn about it Mar 01 17:58:07 it's a nice thing when developing recipes Mar 01 17:58:10 hm seems do compile Mar 01 17:58:22 and how does the devshell deal with this? how does it convert /usr/lib to the right path? Mar 01 17:58:25 the log is config.log Mar 01 17:58:30 I will have a look at it Mar 01 17:58:43 damn where is a site with file paste again Mar 01 17:59:23 woglinde, http://www.pastie.org/pastes/1589070/text Mar 01 17:59:41 [OE::arm-oe-linux-gnueabi-shr-nokia900]:~$ pkg-config --libs --cflags glib-2.0 => -I/home/gnutoo/embedded/oe/oetmps/nokia900/sysroots/armv7a-oe-linux-gnueabi/usr/include/glib-2.0 -I/home/gnutoo/embedded/oe/oetmps/nokia900/sysroots/armv7a-oe-linux-gnueabi/usr/lib/glib-2.0/include -lglib-2.0 Mar 01 17:59:49 Tartarus, still no build Mar 01 18:00:06 http://pastebin.com/kCXKK9yU Mar 01 18:00:19 see around line 26 Mar 01 18:02:05 damn privatepaste.com dont works too Mar 01 18:02:13 I am not sure where the automake 1.7.9 is coming frmo Mar 01 18:02:22 lol Mar 01 18:03:05 hm whats system-backend-tools needed for? Mar 01 18:04:39 gnome systems tools Mar 01 18:04:53 yes i know Mar 01 18:05:02 but what are the really usefull for? Mar 01 18:05:08 dunno Mar 01 18:05:17 it is broken for some build machines Mar 01 18:05:20 including mine Mar 01 18:05:29 hm yeah Mar 01 18:05:46 I'd rather fix the recipe than remove it from the image :) Mar 01 18:06:02 annoyingly, it works for some people Mar 01 18:06:44 the recipe looks shitty Mar 01 18:06:54 woglinde, I'll get back to you on those compilation errors I mentioned (with solution I hope). Now it is time for the cinemas :D Mar 01 18:07:56 risca sure Mar 01 18:22:26 gnutoo ping Mar 01 18:23:45 damn Mar 01 18:37:23 03Stefan Schmidt  07master * rd6f710f0ba 10openembedded.git/conf/distro/include/angstrom-jalimo.conf: Mar 01 18:37:23 angstrom-jalimo.conf: Bump openjdk version to include security and build fixes Mar 01 18:37:23 Signed-off-by: Stefan Schmidt Mar 01 18:37:23 Acked-by: Koen Kooi Mar 01 18:38:33 hm I will push the curl stuff Mar 01 18:45:45 hi all Mar 01 18:45:49 he khem Mar 01 18:45:56 woglinde: howdy Mar 01 18:46:17 hm I will push the curl-native explicit disable ldap stuff Mar 01 18:47:46 hi khem. Mar 01 18:48:34 HopsNBarley: hey Mar 01 18:48:43 hm strange system-tools-backends worked here without problems Mar 01 18:49:06 I am contemplating a libtool'less world Mar 01 18:49:17 \o/ Mar 01 18:49:19 khem? Mar 01 18:49:21 I have it working *almost* at one time Mar 01 18:50:13 so shit on me if I will break the release Mar 01 18:50:14 *g* Mar 01 18:50:51 woglinde: you better test it out Mar 01 18:51:01 woglinde, yeah, apparently this is specific to some fedora's Mar 01 18:51:05 release is at final stages Mar 01 18:51:12 Crofton|work: hello Mar 01 18:51:16 hey Mar 01 18:51:25 khem curls break when ldap-dev is installed anyway Mar 01 18:51:29 fedora is shitty Mar 01 18:51:36 woglinde: ok Mar 01 18:51:48 and I tested Mar 01 18:51:59 it dont breaks when ldap-dev isnt istalled Mar 01 18:52:44 hm whats the release cycle now? Mar 01 18:52:47 3 months? Mar 01 18:52:54 03Marc Reilly  07master * r9bbb63ad7f 10openembedded.git/recipes/qt4/ (2 files in 2 dirs): Mar 01 18:52:55 qt-4.7.1: patch gstreamer phonon plugin to use host rgb endianess Mar 01 18:52:55 This patch was obtained from Nokia bug tracker. Mar 01 18:52:55 Without it, the R and B channels are wrong when playing on ARM system. Mar 01 18:52:55 Signed-off-by: Marc Reilly Mar 01 18:52:55 Signed-off-by: Tom Rini Mar 01 18:53:02 hm no we need more time for yocto Mar 01 18:53:14 03Denis Carikli  07master * re124d03f53 10openembedded.git/recipes/ (netbase/netbase_4.21.bb opkg/files/configure opkg/opkg.inc): Mar 01 18:53:14 netbase and opkg: reorder init scripts to make networking come up at first boot Mar 01 18:53:14 Without that fix, usb0 doesn't come up at first boot Mar 01 18:53:14 (tested on the bug machine, with Angstrom on a console-image). Mar 01 18:53:14 Signed-off-by: Denis 'GNUtoo' Carikli Mar 01 18:53:14 Signed-off-by: Tom Rini Mar 01 18:53:24 woglinde: yes quarterly Mar 01 18:53:32 tom is pushing to fast Mar 01 18:53:32 03Otavio Salvador  07master * rcffa6d72a8 10openembedded.git/recipes/freerdp/freerdp.inc: Mar 01 18:53:32 freerdp: rdepends on xprop Mar 01 18:53:32 Signed-off-by: Otavio Salvador Mar 01 18:53:32 Acked-by: Phil Blundell Mar 01 18:53:32 Signed-off-by: Tom Rini Mar 01 18:53:38 morning khem Mar 01 18:53:50 03Jan Kobler  07master * raffc152bd4 10openembedded.git/ (8 files in 8 dirs): (log message trimmed) Mar 01 18:53:50 New machine pcm043 Phytec phyCORE-i.MX35 Mar 01 18:53:50 The new machine pcm043 is the board Phytec phyCORE-i.MX35. Mar 01 18:53:50 The machine name pcm043 is chosen, because it is already used in the linux kernel Mar 01 18:53:50 for this board. Mar 01 18:53:50 BSP is derived from BSP of machine mx31ads. Mar 01 18:53:51 Linux kernel is based on Phytec Linux BSP for phyCORE-i.MX35 PD 10.1.1 Mar 01 18:53:53 03Jan Kobler  07master * r1840fd720b 10openembedded.git/recipes/linux/ (87 files in 2 dirs): (log message trimmed) Mar 01 18:53:53 linux-2.6.31 config and patch files for pcm043 Mar 01 18:53:53 These config and patch files are from the Phytec Linux BSP for phyCORE-i.MX35 PD10.1.1. Mar 01 18:53:53 They were taken from the archive Mar 01 18:53:53 ftp://ftp.phytec.de/pub/Products/phyCORE-iMX350/Linux/PD10.1.1/OSELAS.BSP-Phytec-phyCORE-i.MX35-PD10.1.1.tar.gz Mar 01 18:53:54 The patch files are from the folder in this archive Mar 01 18:53:54 OSELAS.BSP-Phytec-phyCORE-i.MX35-PD10.1.1/configs/phyCORE-i.MX35-2010.08.0/patches/linux-2.6.31.6/generic Mar 01 18:54:40 I hope gnutoo can if it a test Mar 01 18:56:00 * Tartarus promises to slow down after the release is out :) Mar 01 18:56:17 ka6sox-away: hi Mar 01 18:56:28 Tartarus: one thing that worries me is the runtime testing Mar 01 18:56:56 Tartarus: I have seen that sometimes it takes a lot of time to get it to runnable state Mar 01 18:57:09 and more commits means more searching if so happens Mar 01 18:57:38 Yes, I look forward to being able to leverage the run time stuff poky has for qemu Mar 01 18:59:37 Tartarus, you didn't have anymore more idea on my system tools headache Mar 01 18:59:49 I guess you are a bad guy to ask sinc you do not have fedora :( Mar 01 19:00:22 Crofton|work: whats the issue Mar 01 19:03:28 http://pastebin.com/kCXKK9yU Mar 01 19:03:32 see around line 26 Mar 01 19:04:24 hmm I see Mar 01 19:08:06 Crofton|work: try this patch http://pastebin.com/d9iG7FLa Mar 01 19:14:28 hi blindvt Mar 01 19:14:43 re gnutoo Mar 01 19:14:55 gnutoo can you test the curl stuff? Mar 01 19:15:14 woglinde, hi Mar 01 19:15:34 khem http://pastebin.com/HH3dWsin Mar 01 19:16:57 solved that with a suggestion from Tartarus earlier Mar 01 19:17:34 no Mar 01 19:17:42 same issue Mar 01 19:17:52 crofton? Mar 01 19:17:55 curl? Mar 01 19:18:21 ? Mar 01 19:18:45 khem btw. uclibc-get needs some newer libc-kernelheaders then .32 Mar 01 19:19:37 03Henning Heinold  07org.openembedded.dev * r0f7996bb4d 10openembedded.git/recipes/curl/ (curl-common.inc curl_7.21.1.bb curl_7.21.2.bb): (log message trimmed) Mar 01 19:19:37 curl: fix native build when ldap-headers are installed on the host Mar 01 19:19:37 * there were serval build error reported for curl-native, Mar 01 19:19:37 when ldap-dev stuff was installed on the build-host, Mar 01 19:19:37 so we explicit disable the build of ldap and ldaps protocls Mar 01 19:19:37 for native, nativesdk and sdk Mar 01 19:19:38 * put the BBCLASSEXTEND line into common.inc rather than into Mar 01 19:25:21 woglinde_, khem, speaking of kernel-headers. Is there a story of not providing libc-kernel-headers by means of the kernel recipes? Mar 01 19:25:38 s/ of /behind/ Mar 01 19:25:52 hm there was some discussion on the ml Mar 01 19:25:53 yes Mar 01 19:26:03 blindvt: I always get tempted to do so Mar 01 19:26:10 but that wouldnt solve it for older kernels Mar 01 19:26:15 but in large scheme of things we wont be able to share Mar 01 19:26:25 where the define in socket.h still is Mar 01 19:26:26 so say you have two armv7 machines Mar 01 19:26:37 then you have to build complete rfs separately Mar 01 19:26:45 iow multimachine wont work Mar 01 19:27:43 for older (let's say < 2.6.19) the sanitized headers are of course a must, but for everything vaguely recent, the separate libc-kernel-headers sound a bit unclean Mar 01 19:28:27 blindvt: yes I know but there is a tradeoff here Mar 01 19:28:36 by using same kernel Mar 01 19:28:38 khem, i don't get what you mean. Can you explain? Mar 01 19:28:55 right now if you build for many machines in same tmpdir Mar 01 19:29:07 then say you have two machines which are similar Mar 01 19:29:12 say based on armv7 Mar 01 19:29:25 then it shared most of arch specific packages Mar 01 19:29:39 but with such a setup this will be wrong Mar 01 19:30:00 and to get it working you have to setup separate tmpdir per machine Mar 01 19:31:47 | config.status: error: po/Makefile is not ready. Mar 01 19:31:57 anyone have any thoughts on where that comes from? Mar 01 19:32:02 Crofton|work: I am building it locally here Mar 01 19:32:07 lets see where I end up Mar 01 19:32:23 fedora? Mar 01 19:33:00 khem, i'm talking about tar -xf 2.6.37 && for a in foo bar baz etc;do make -C there install_headers ARCH=$a;done I mean, you'll most likely need the kernel for building later on anyway, so why not just use that to spit out headers for libc? The ultimate outcome of the current way (separate packages) is pretty much the same (except you have to do the work twice for every kernel version, once for the kernel itself and once for the libc-he Mar 01 19:33:01 aders) as with a "single" unpacked kernel sourcetree that you get headers out of and later may build a binary Mar 01 19:36:23 blindvt: because kernel headers are different version most of times Mar 01 19:36:37 blindvt: real kernel uses vendor trees etc. Mar 01 19:36:47 for kernel headers we always use upstream release Mar 01 19:38:08 anyone knowwhat po/Makefile is? Mar 01 19:38:44 makefile in the po directory? :p Mar 01 19:38:57 Crofton|work: thats for documentation Mar 01 19:39:00 something translation related? Mar 01 19:39:06 yep Mar 01 19:39:47 http://www.gnu.org/software/hello/manual/gettext/PO-Files.html Mar 01 19:40:43 weird Mar 01 19:40:57 I am not wise in the ways of translations :) Mar 01 19:41:09 ouch not my day, first I send a mail without topic, and then I send a wrong mail Mar 01 19:41:16 Crofton|work: as a workaround it can be disabled Mar 01 19:41:41 Crofton|work: once it builds on my slooow machine I will know more Mar 01 19:41:52 GNUtoo|laptop: heh happens Mar 01 19:42:03 I sometimes type passwords into IRC channels Mar 01 19:42:08 ouch Mar 01 19:42:54 ok Mar 01 19:43:17 I'm going to look at some bus timingas an interrupt response times then Mar 01 19:54:25 hi Mar 01 21:01:07 03Simon Busch  07org.openembedded.dev * re2121e9641 10openembedded.git/recipes/freesmartphone/ (cornucopia.inc frameworkd_git.bb libgisi_git.bb msmcomm.inc): Mar 01 21:01:07 freesmartphone: bump SRCREV of several recipes Mar 01 21:01:07 Signed-off-by: Simon Busch Mar 01 21:01:08 03Simon Busch  07org.openembedded.dev * r2bd8ccc062 10openembedded.git/recipes/palmpre/tsmd_git.bb: Mar 01 21:01:08 tsmd: install script to /usr/bin and not to /bin and bmp PR Mar 01 21:01:08 Signed-off-by: Simon Busch Mar 01 21:01:09 03Simon Busch  07org.openembedded.dev * rb25fa087d8 10openembedded.git/recipes/e17/elfe_svn.bb: Mar 01 21:01:10 elfe: bump SRCREV Mar 01 21:01:10 Signed-off-by: Simon Busch Mar 01 21:01:22 *ding* Mar 01 21:01:32 dong Mar 01 21:01:45 bah, didn't annotate the tag Mar 01 21:02:43 OK, ding for real :) Mar 01 21:02:55 gratz Mar 01 21:35:44 I'm experiencing some building errors that has to do with gcc 4.4.2 producing invalid makefiles Mar 01 21:36:02 hm why you need gcc-4.4.2? Mar 01 21:36:05 use the 4.5 Mar 01 21:36:08 its stable Mar 01 21:36:44 Hmm... I wonder why my oe build went with 4.4.2? Mar 01 21:36:52 I'm using DISTRO=minimal Mar 01 21:37:23 hm seems nobody updated minimal Mar 01 21:37:41 minimal isnt that well maintained Mar 01 21:37:44 angstroem is Mar 01 21:37:46 shr is Mar 01 21:37:48 and micro Mar 01 21:38:14 minimal is one of first distros who migrated to 4.5 Mar 01 21:38:24 hm Mar 01 21:38:29 That's odd Mar 01 21:38:29 khem maintaining it Mar 01 21:38:33 lets see Mar 01 21:39:15 PREFERRED_GCC_VERSION ?= "4.5" Mar 01 21:39:22 yes minimal is using 4.5 from sane-toolchain.inc Mar 01 21:39:24 risca whats your local.conf looks like? Mar 01 21:39:25 wrong local.conf? Mar 01 21:39:50 Coming up Mar 01 21:39:52 the special settings for machine are all not 4.4 Mar 01 21:40:11 ANGSTROM_GCC_VERSION_armv4 ?= "4.4.2" Mar 01 21:40:15 4.3.2 4.2.1 4.1.2 4.1.1 Mar 01 21:40:20 ie, but that would be angstrom and armv4 Mar 01 21:40:38 http://pastebin.com/CVqvkPMH Mar 01 21:40:38 and what gcc is producing makefiles ? :) Mar 01 21:41:17 risca: whats the arch Mar 01 21:41:34 read Mar 01 21:41:36 TARGET_OS = "linux" Mar 01 21:41:40 you dont need to set Mar 01 21:41:46 armv4 works well with minimal Mar 01 21:41:54 risca ah you are using the ea thingie too Mar 01 21:42:03 Yeah, that's right :D Mar 01 21:42:06 MACHINE = "ea3250" Mar 01 21:42:10 That's the machine Mar 01 21:42:14 I fixed the kernel today Mar 01 21:42:23 Of ea3250? Mar 01 21:42:31 yes Mar 01 21:42:43 the one patch didnt apply correctly Mar 01 21:42:54 Yeah, that's right! Mar 01 21:43:10 I forgot about that. I came across that error too Mar 01 21:43:14 hehe Mar 01 21:43:19 now its fixed Mar 01 21:43:25 Officially Mar 01 21:43:26 maybee you are in the same group Mar 01 21:43:32 as the other guy Mar 01 21:43:36 reporting it to my Mar 01 21:43:38 to me Mar 01 21:43:55 What other guy? Mar 01 21:44:16 I don't think so Mar 01 21:44:17 prjt61 Mar 01 21:44:44 I'm basically alone working on the ea3250 in my group. It's kinda like a school project Mar 01 21:45:26 So, I don't need to set TARGET_OS=linux? Mar 01 21:45:39 read the comment above the line Mar 01 21:45:47 rsica he said the same Mar 01 21:45:54 school project Mar 01 21:45:54 Ah, missed that line... Mar 01 21:45:56 *g* Mar 01 21:46:11 He isn't from Sweden, is he? Mar 01 21:46:43 prjt61 [~prjt61@145.93.21.43] Mar 01 21:47:46 Nope, Netherlands Mar 01 21:48:09 I guess there is more than 1 student working on a ea3250 ^^ Mar 01 21:48:21 hehe seems so Mar 01 21:48:25 hi likewise Mar 01 21:49:16 Let's see. What version of gcc is angstrom-2008 providing? Mar 01 21:50:06 ? Mar 01 21:50:15 you said you used minimal Mar 01 21:50:24 dont mixup distro with image Mar 01 21:51:13 I'm not. I'm just lazy... Mar 01 21:51:34 I entered DISTRO = angstrom-2008.1 first, then changed to minimal Mar 01 21:51:57 As I changed to minimal I saw that bitbake recompiled gcc though Mar 01 21:52:18 hm Mar 01 21:52:25 better start from scratch Mar 01 21:52:29 with diffrent distro Mar 01 21:52:38 I was afraid you would say that Mar 01 21:53:23 I guess I would have seen these error coming my way ^^ Mar 01 21:53:46 Changing DISTRO in the middle of a build can never be good Mar 01 21:54:49 So, what distro _should_ I use? Mar 01 21:55:28 ? Mar 01 21:55:33 what you want Mar 01 21:56:16 I'll have another go with minimal then Mar 01 21:56:36 Are there any distros that are more stable than others? Mar 01 21:56:47 That you know of? Mar 01 22:01:49 hi woglinde, all Mar 01 22:02:48 angstroem 2010 seems stable Mar 01 22:04:36 I think I'll go with Ångström 2010. I think that was what I used before my old computer crashed and burned Mar 01 22:04:49 Crofton|work: system-tools-backends builds fine for arm for me with the patch I posted :( Mar 01 22:05:41 Crofton|work: it seems a fedora issue to me Mar 01 22:05:42 on fedora? Mar 01 22:05:47 yeah, but how Mar 01 22:05:50 no I have ubuntu Mar 01 22:06:14 Crofton|work: I guess there is some host package that you dont have installed which I tend to have Mar 01 22:06:37 so you ended in error: po/Makefile is not ready. Mar 01 22:06:43 yeah Mar 01 22:06:55 I wonder if there is some race Mar 01 22:06:56 very mysterious Mar 01 22:07:10 Does it happen second time ? Mar 01 22:07:22 khem I didnt patch it and it worked fine Mar 01 22:07:45 woglinde: yeah I remember I did that patch some months ago Mar 01 22:07:51 but seems its not needed Mar 01 22:08:19 still it should not harm to have it as it is its emulating what autotools would do otherwise Mar 01 22:08:20 khem, are you using bbthreads? Mar 01 22:08:28 Crofton|work: yes 2 Mar 01 22:08:37 crofton I have 4 Mar 01 22:08:41 and it worked Mar 01 22:08:45 the patch helps in the sense that I get further along .... Mar 01 22:08:50 but I didnt build from scratch Mar 01 22:08:59 I built from scratch Mar 01 22:09:09 stupid fedoara Mar 01 22:09:19 * khem has slackware Mar 01 22:09:32 Crofton|work: what distro/machine/image is affected? Mar 01 22:09:33 stupid OE for having dependencies on the build machine Mar 01 22:09:41 angstrom gnome image Mar 01 22:09:48 oh.. Mar 01 22:09:53 it's large :) Mar 01 22:10:01 Crofton|work: can you see if it happens again Mar 01 22:10:12 want to make sure its not some sort of race Mar 01 22:10:34 k Mar 01 22:10:50 Crofton|work: Do you have gettext installed ? Mar 01 22:10:58 on this F fox Mar 01 22:11:24 * khem clarifies F = Fedora no tangents please Mar 01 22:11:26 yes Mar 01 22:11:51 ok Mar 01 22:11:54 let me see a bit Mar 01 22:13:40 turned off || make and bbthreads Mar 01 22:14:09 same Mar 01 22:18:12 03Martin Jansa  07master * r8c6a3711e3 10openembedded.git/recipes/webm/libvpx.inc: Mar 01 22:18:12 libvpx: force arm mode to be able to link with ffmpeg where arm mode already is forced Mar 01 22:18:12 * v2: updated INC_PR after 0478950960e5c600efc77c25559fc7eee311f686 Mar 01 22:18:12 Signed-off-by: Tom Rini Mar 01 22:18:14 03Tom Rini  07master * r5205a53c99 10openembedded.git/recipes/webm/libvpx.inc: Mar 01 22:18:14 Revert "libvpx: force arm mode to be able to link with ffmpeg where arm mode already is forced" Mar 01 22:18:14 We back this out after the release in order to not mask the problem. Mar 01 22:18:14 This reverts commit 8c6a3711e329522bf1fd2e6c020509c122c235ab. Mar 01 22:18:35 *g* Mar 01 22:18:55 Crofton|work: try adding --disable-nls to EXTRA_OECONF Mar 01 22:20:51 ok Mar 01 22:22:22 Crofton|work: next try this patch http://paste.ubuntu.com/574188/ Mar 01 22:24:02 disbale did not help Mar 01 22:26:29 and that patch gets me back to compile rerunning configure Mar 01 22:27:36 Crofton|work: and where does it end ? Mar 01 22:28:00 heh Mar 01 22:28:26 it never does Mar 01 22:29:23 Crofton|work: you can post the config.log somewhere for me ? Mar 01 22:29:35 for which run? Mar 01 22:29:59 there should a file called config.log ib builddir Mar 01 22:30:03 Okay, running task 80 of 5250 ^^ Time to turn down the lid and watch some big bang theory and hope it's gotten past task 3000 in the morning =) Mar 01 22:30:06 and also configure step Mar 01 22:33:21 for me it never tries to create po/Makefile Mar 01 22:34:26 do yuo have gettext installed? Mar 01 22:34:38 yes Mar 01 22:34:55 gettext (GNU gettext-runtime) 0.18.1 Mar 01 22:34:55 khem, what timezone are you on? And do you ever sleep? Mar 01 22:35:07 grg PST8PDT Mar 01 22:35:31 grg: I do sleep (open eyed) Mar 01 22:35:50 I sleep more than ka6sox-away Mar 01 22:36:43 khem is AI behind OE Mar 01 22:36:46 may be OE-based :) Mar 01 22:36:55 Jay7, too true Mar 01 22:38:16 i don't see a conf/machine/khem.conf :) Mar 01 22:38:55 it's private branch I suspect :) Mar 01 22:38:59 * khem fetches coffee Mar 01 22:39:07 hehe Mar 01 22:41:58 JaMa, belated pong Mar 01 22:44:52 hm.. Mar 01 22:44:59 do we have release? Mar 01 22:45:16 * Jay7 missed something Mar 01 22:48:41 * ant__ doesn't see fresh commits...CIA-47 is sleeping? Mar 01 22:50:51 been laggy all day Mar 01 22:54:26 Tartarus: btw I'm having fun with a mipsel needing uImage... I added your hack about dynamic entrypoint Mar 01 22:54:58 now I'm dubious about a patch adding uImage target to mips kernels... Mar 01 22:55:40 done that, I'd use the code in kernel.bbclass Mar 01 22:57:23 i think the mips-linux tree recently accepted a patch to the arch Makefile for a uImage target Mar 01 22:57:34 is that the same patch? Mar 01 22:57:44 I didn't find it in linux-next today Mar 01 22:57:51 moment Mar 01 22:58:05 we have problem with kexec on mipsel Mar 01 22:58:24 it doesn't working in short words Mar 01 22:58:59 it is possible to load uncompressed vmlinux with kexec -l, but kexec -e freezes machine Mar 01 22:59:11 Here's the patch. It may not have been applied. http://www.spinics.net/lists/mips/msg39827.html Mar 01 23:00:04 http://patchwork.linux-mips.org/patch/1998/ Mar 01 23:00:08 ^^ state new Mar 01 23:00:12 ? Mar 01 23:00:29 same Mar 01 23:00:39 yeah, looks like it Mar 01 23:01:35 Jay7: remember that mips can handle compressed kernel since 2.6.33 (when launched by kexec) Mar 01 23:01:45 u-boot decompress in the other cases Mar 01 23:10:20 * grg is rewriting opkg Mar 01 23:10:35 ... and is accepting design ideas and feature requests Mar 01 23:11:15 I hope to have code for public consumption by the end of March. Mar 01 23:11:22 not sure fi this is lacking in opkg -- or just the poky implementation.. Mar 01 23:11:27 directory attributes and ownership.. Mar 01 23:11:49 fray, can you clarify that? Mar 01 23:12:03 grg: great, what are the considerations Mar 01 23:12:15 without the packager having specific directory ownership and attributes it's not possible have a consistent known owner/group/mode per directory.. which results in whoever created it is what ends up existing.. Mar 01 23:12:32 this is a problem for some directories who don't have the permissive 0755 -- or have an owner/group such as spool/mail Mar 01 23:12:43 (and especially directories that are setup with inheriting modes) Mar 01 23:13:00 RPM has this functionality -- but we do NOT use it today in Poky... Mar 01 23:13:02 Current design goals: http://pastebin.com/C54iRuNi Mar 01 23:13:04 (future enhancement Mar 01 23:14:07 grg: looks nice. Will you give it a yet different name or still call it opkd Mar 01 23:14:12 fray, are you talking about chown/chmod when the package is untarred? Mar 01 23:14:14 package summary/description.. I know dpkg has this, (and rpm).. opkg MIGHT, but I'm not completely sure.. Mar 01 23:14:20 khem, i'm calling it ypkg Mar 01 23:14:23 no, actually "who owns what" on directories.. Mar 01 23:14:52 grg: opkg is better imo since o can be interpreted as openembedded Mar 01 23:15:07 now that openmoko is history Mar 01 23:15:07 for instance.. /var/spool/mail would be owned by root:mail with mode 0775 Mar 01 23:15:19 khem, other projects are using opkg Mar 01 23:15:26 if something else comes in and tries to "own" /var/spool/mail with different permissions -- an error/conflict should be generated Mar 01 23:15:43 this is different from the implicit creation of unowned directories.. Mar 01 23:15:46 fray: yes thats a nice feature I would say Mar 01 23:15:48 this also affects install ordering.. Mar 01 23:16:03 since software that goes into /var/spool/mail needs to be reordered after the directory that owns it.. Mar 01 23:16:14 fray, ok. opkg does chown/chmod dirs, but doesn't check the ownership when it extracts new packages over that dir. Mar 01 23:16:14 reordered ? Mar 01 23:16:17 if nothing claims to own a directory then the default mask (0755) and owner/group (root:root) should be applied Mar 01 23:16:49 it's the conflict finding and install ordering that is important with this.. (just like file extraction and conflict control) Mar 01 23:17:04 yapkg Mar 01 23:17:09 yet another :) Mar 01 23:17:42 a name change is but a sed away... Mar 01 23:17:51 package install ordering matters vs simply changing owner/group/mode when you hit that package due to windows where the wrong modes might be set.. (i.e. something is set too permissive) Mar 01 23:18:16 these are all considerations for specific types of systems.. many systems frankly have one user (root) and don't care about permissions.. Mar 01 23:18:59 the other feature that I'd like to see.. multilib reconciliation.. in otherwords.. normally installing two files of the same name -- but different contents should lead to a conflict... Mar 01 23:19:08 does CIA still work? Mar 01 23:19:10 I just made some commits Mar 01 23:19:22 but if the type of one is ELF32 and the other ELF64.. a system/package policy may be able to say "hey I want the ELF64 version".. no conflict/error is generated.. Mar 01 23:19:30 and all non-conflicting files are installed from both.. Mar 01 23:19:52 this is REALLY useful on larger systems that may have a primary ABI installed, and one or more secondary ABIs for compatibility Mar 01 23:20:28 (that of course requires type knowledge and policy control during conflict resolution) Mar 01 23:20:30 GNUtoo|laptop: yes CIA works but sometimes it holds the information for further scrutiny :) Mar 01 23:21:09 holding the information is not nice Mar 01 23:21:26 * fray has a /ignore CIA-* Mar 01 23:21:49 lol Mar 01 23:22:08 fray, I'll make conflict resolution be easily extendable for a start and we can play with those ideas. Mar 01 23:22:23 grg, works for me.. :) Mar 01 23:22:46 the other thing I suggest you make extendable -- package validation -- and subsequent file validation Mar 01 23:22:52 yes Mar 01 23:22:55 absolutely Mar 01 23:23:03 there are lots of different groups that want different kinds of validation at both levels.. Mar 01 23:23:18 post-file processing is another thign (this is different from install scripts.. it's really attribute processing for things like selinux) Mar 01 23:24:04 I think someone can just consult on writing policies for selinux Mar 01 23:24:06 * grg probably needs to read up on selinux and ACLs, etc. Mar 01 23:24:45 grg, I strongly suggest you avoid anything to do with embedding selinux knoweldge.. otherwise you might go insane Mar 01 23:25:07 external control is the best way to do it.. (of course someone has to figure out how to feed policy data and such.. but there is no clear answer to all of that) Mar 01 23:25:15 i just want to make sure i don't unintentionally make it hard to do selinux stuff Mar 01 23:25:16 dont overcomplicate opkg I would say Mar 01 23:25:25 keep is simple but no simpler Mar 01 23:25:30 khem, ya.. Mar 01 23:25:49 if someone want selinux kind of things he better use rpm Mar 01 23:25:54 install ordering.. install actions.. external helpers for conflict resolution.. Mar 01 23:26:04 khem, rpm is the wrong place to do selinux behavior.. Mar 01 23:26:05 it shouldn't be that much code. Opkg is probably twice as much code as it needs to be. Mar 01 23:26:17 IMHO Fedora folks don't know what they're doing any better then anyone else when it comes to packaging and selinux Mar 01 23:26:35 it's the ordering and conflict finding that becomes difficult IMHO.. Mar 01 23:26:40 grg: IMO one aim for the rewrite should be small footprint Mar 01 23:26:54 khem, yes Mar 01 23:26:59 ah selinux? Mar 01 23:27:09 maybe better not running everything as root first? Mar 01 23:27:16 else selinux is nice Mar 01 23:27:17 i've actually started writing code already... using valgrind lots to check memory usage Mar 01 23:27:36 the other thing that people overlook.. ;) there needs to be a simple way to query whats installed.. :P Mar 01 23:27:47 grg: there still are enough people who prefer to have package management software to be small and work on their contrained systems Mar 01 23:28:05 (ya I know obvious.. but it's easy to forget.. I've seen designs focus on all of the install stuff.. and then have no way to figure out what was actually installed) Mar 01 23:28:06 opkg list installed Mar 01 23:28:06 I wrote policies long time ago Mar 01 23:28:09 already exists Mar 01 23:28:11 03Andrea Adami  07org.openembedded.dev * r5c6097d75d 10openembedded.git/conf/machine/ben-nanonote.conf: Mar 01 23:28:11 ben-nanonote.conf: prepare for standard uboot-mkimage task. Mar 01 23:28:11 * enable linux-kexecboot by default. Mar 01 23:28:11 Signed-off-by: Andrea Adami Mar 01 23:28:15 03Andrea Adami  07org.openembedded.dev * r3c344717b4 10openembedded.git/recipes/linux/linux-kexecboot.inc: Mar 01 23:28:15 linux-kexecboot.inc: apply lzma/gzip mangling only to arm kernels Mar 01 23:28:15 * refine custom mips uboot-mkimage, waiting for uImge target in kernel. Mar 01 23:28:15 Signed-off-by: Andrea Adami Mar 01 23:28:20 I've even the outdated book Mar 01 23:28:22 khem, I know -- point is that it needs to continue to exist.. ;) Mar 01 23:28:37 GNUtoo|laptop: ok so CIA found no problem with your patch no secrets revealed Mar 01 23:28:48 lol Mar 01 23:28:55 fray, don't worry, stuff like that will necessarily exist because i'll want it for testing while i code other parts Mar 01 23:29:27 but yes.. my number one request is directory ownership and having that affect install order.. ;) Mar 01 23:29:43 maybe CIA knows that I have good commit messages, so CIA doesn't bother showing the commits Mar 01 23:29:57 (might be nice to have a policy to decide what to do if a directory is un-owned .. error, warning or create it implicitly) Mar 01 23:30:23 fray, can you provide an example of directory ownership affecting the install order? Mar 01 23:30:36 how that would work, beyond a DEPENDS Mar 01 23:30:44 typical system (not great example) Mar 01 23:30:55 a filesystem / base package is layed down that creates the /bin directory.. Mar 01 23:31:02 that should be isntalled before anything that goes into /bin Mar 01 23:31:52 ok. Well rpm allows for pacakges to Provide anything. Like files, dirs, etc. Mar 01 23:32:21 I don't see any reason why that wouldn't be supportable with current opkg Mar 01 23:32:29 I've had this problem in the past, and I wish I could remember exactly what it was.. but the security issue was there was a race window where software/data files were installed -- but it was before the more restrictive permissions came in.. this race allowed someone logged in to see things they shouldn't have Mar 01 23:32:38 was an issue for a carrier grade customer I had a while back.. Mar 01 23:32:56 grg, as for provide/depends.. there really are two types.. "states" and "implicit".. Mar 01 23:33:13 the files are "implicit".. while the stated are just that, verbatim what someone typed + versioning Mar 01 23:33:31 when you do resolution, you simply check stated -- fail, fall back to implicit.. Mar 01 23:34:01 (or do it the other way around -- or create a list of both to compare against..) ideas all the same Mar 01 23:34:07 ohh pkgname is also implicit dep Mar 01 23:35:52 (the second you start treating package name, file names, etc differently from a dependency resolution standpoint, the more complexity, the more bugs.. keep that simple and have some simple routines to promote the implied deps and explicitly stated deps for resolution Mar 01 23:36:44 those implied deps/provides could be put in the package at package creation time Mar 01 23:36:56 slightly more fragile, but far simpler Mar 01 23:37:06 they exist in the package already -- just depends on how you want to capture and process them.. Mar 01 23:37:11 i'm pretty sure that's what rpm does Mar 01 23:37:19 RPM has some internal data structure optimizations.. for instance filenames.. Mar 01 23:37:25 put into the package's meta data i mean Mar 01 23:37:50 instead of listing the archive each time, a copy of the files and associate meta data are stored as a list of.. all paths within the package.. and a list of all filenames.. with a pointer to the associated path.. Mar 01 23:37:56 on large packages this saves a lot of room Mar 01 23:38:10 (the meta data in RPM isn't compressed -- other then my the format fo the data itself) Mar 01 23:38:41 (the other thing RPM has that I really like is per-file dependency data.. but it's more useful for debugging that anythign practical these days) Mar 01 23:39:24 http://gate.crashing.org/~fray/rpm/rpm-xml-desc.jpg Mar 01 23:39:34 that is the description of the rpm format from a few years ago.. (like 7) ;) Mar 01 23:39:46 but you can see all of the metadata that is stored outside of the archvie.. Mar 01 23:40:17 in fact the internal cpio archive in an RPM doesn't necessarily have the same permissions, owners, groups as what will be installed onto the target.. thats handled directly for the associated data Mar 01 23:40:48 one other trick with signing.. there has to be a way to sign a package, without the signature affecting the package so you can't validate it.. ;) that can always be tricky as well Mar 01 23:43:26 03Andrea Adami  07org.openembedded.dev * r43bdeb023b 10openembedded.git/recipes/linux/linux-kexecboot.inc: Mar 01 23:43:26 linux-kexecboot.inc: explicit we don't want modules. Mar 01 23:43:26 * fix harmless typo... Mar 01 23:43:26 Signed-off-by: Andrea Adami Mar 01 23:45:28 03Denis 'GNUtoo' Carikli  07org.openembedded.dev * rfa59c1a71e 10openembedded.git/recipes/shr/ (shr-settings/htcdream/elementary.sh shr-settings_git.bb): Mar 01 23:45:29 shr-settings: add specific elementary.sh for the htcdream machine Mar 01 23:45:29 Signed-off-by: Denis 'GNUtoo' Carikli Mar 01 23:45:29 03Denis 'GNUtoo' Carikli  07org.openembedded.dev * rd50ab86bdb 10openembedded.git/recipes/xfce-base/xfce4-mixer_4.6.1.bb: (log message trimmed) Mar 01 23:45:29 xfce4-mixer 4.6.1: add gst-plugins-base dependency Mar 01 23:45:30 Without that fix we have: Mar 01 23:45:30 checking for gstreamer-plugins-base-0.10 >= 0.10.2... not found Mar 01 23:45:31 *** The required package gstreamer-plugins-base-0.10 was not found on your system. Mar 01 23:45:31 *** Please install gstreamer-plugins-base-0.10 (atleast version 0.10.2) or adjust Mar 01 23:45:47 *** the PKG_CONFIG_PATH environment variable if you Mar 01 23:57:40 ah finally Mar 02 00:00:02 03Andrea Adami  07org.openembedded.dev * r2479922601 10openembedded.git/recipes/linux/ (3 files in 2 dirs): Mar 02 00:00:02 linux-kexecboot: update 2.6.37 defconfig for ben-nanonote. Bump INC_PR. Mar 02 00:00:02 Signed-off-by: Andrea Adami Mar 02 00:24:39 03Andrea Adami  07org.openembedded.dev * r701c15ba40 10openembedded.git/recipes/linux/ (5 files in 2 dirs): Mar 02 00:24:39 linux: add initial ben-nanonote defconfig to 2.6.37. Mar 02 00:24:39 Signed-off-by: Andrea Adami Mar 02 00:24:39 03Andrea Adami  07org.openembedded.dev * r67e38549fd 10openembedded.git/conf/machine/ben-nanonote.conf: Mar 02 00:24:39 ben-nanonote.conf: default to "linux" as virtual/kernel. Mar 02 00:24:40 * jlime distro has own override, so no harm. Mar 02 00:24:40 Signed-off-by: Andrea Adami Mar 02 00:26:42 03Andrea Adami  07org.openembedded.dev * r09037741d0 10openembedded.git/recipes/linux/linux-kexecboot_2.6.37.bb: Mar 02 00:26:42 linux-kexecboot: remove redundant entry in SRC_URI for ben-nanonote. Mar 02 00:26:42 Signed-off-by: Andrea Adami Mar 02 00:32:48 ERROR: Trying to resolve runtime dependency gdk-pixbuf-loader-png resulted in conflicting PREFERRED_PROVIDER entries being found. Mar 02 00:32:49 The providers found were: ['/storage/oe/openembedded/recipes/gtk+/gtk+_2.20.1.bb', 'virtual:native:/storage/oe/openembedded/recipes/gtk+/gtk+_2.20.1.bb'] Mar 02 00:32:51 wtf? Mar 02 00:33:07 same for every pixbuf-loader-* Mar 02 00:33:32 non-fatal but looking strange Mar 02 00:34:18 minimal/ben-nanonote/native-sdk-image Mar 02 00:38:41 * Jay7 -> sleep Mar 02 00:39:01 gn Mar 02 00:53:25 nn Crofton Mar 02 00:54:23 03Andrea Adami  07org.openembedded.dev * r3dc72c4ca0 10openembedded.git/recipes/linux/linux-kexecboot.inc: Mar 02 00:54:24 linux-kexecboot.inc: fix typo for task override (mips/mipsel). Mar 02 00:54:24 Signed-off-by: Andrea Adami Mar 02 00:55:03 'nite **** ENDING LOGGING AT Wed Mar 02 02:59:57 2011