**** BEGIN LOGGING AT Fri Feb 12 02:59:56 2010 Feb 12 04:56:27 Anyone know how to fix this stage-manager build error? line 34, in read_cache cache[data[0]]['ts'] = int(data[1]) IndexError: list index out of range Feb 12 05:19:11 hi , i ve added xsetroot on my angstrom OS , i can hide mouse cursor on starting xsetroot command on xterm but when i try to run it on .xinitrc script , its still showing me cursor on boot.Where will i need to run xsetroot -cursor command on startup ? which file ? Feb 12 06:32:56 Hi All,Am getting the error ERROR: Please set the 'PERSISTENT_DIR' or 'CACHE' variable. Feb 12 06:33:03 while try to build Feb 12 06:33:24 how to resolve this Feb 12 06:35:48 have u set the necessary environment variables? Feb 12 06:35:55 like OEPATH Feb 12 06:36:31 @siji Feb 12 06:37:47 i think yes Feb 12 06:38:17 in your local.conf what is the value of BBFILES Feb 12 06:38:38 BBFILES := "${OETREE}/openembedded/recipes/*/*.bb" Feb 12 06:38:56 I want to build it for Beagle board +Angstrom Feb 12 06:39:49 so change your local.conf Feb 12 06:39:54 and there Feb 12 06:40:05 edit MACHINE="beagleboard" Feb 12 06:40:11 i did it Feb 12 06:41:15 and u have also set the environment variable BBPATH Feb 12 06:41:35 how to do that Feb 12 06:42:42 export BBPATH=/your_path/build:/your_path/openembedded Feb 12 06:42:47 try this and then tell Feb 12 06:42:49 ok Feb 12 06:42:52 ok Feb 12 06:42:53 the result Feb 12 06:42:58 sure Feb 12 06:48:42 Spyzer,http://debian.pastebin.com/m362c3303 Feb 12 06:48:48 Getting this error now Feb 12 06:50:17 google out to check how to configure the default shell of ubuntu or debian or the distro which ur using(i think its debian based) Feb 12 06:50:37 ubuntu in default has the dash shell enabled as the default one Feb 12 06:50:51 you can change that easily Feb 12 06:50:59 dpkg-reconfigure dash Feb 12 06:51:00 but i don't remember Feb 12 06:51:03 ok ok Feb 12 06:51:06 yups Feb 12 06:52:22 Am using ubuntu Feb 12 06:52:32 , not working Feb 12 06:52:44 Googling now ;) Feb 12 07:04:17 I think i need some help Feb 12 07:04:34 couldnt able to find any proper solution Feb 12 07:04:36 ;( Feb 12 07:26:41 hi Feb 12 07:26:58 Anybody can have a look on my local.conf Feb 12 07:27:05 http://debian.pastebin.com/m1a0b1a81 Feb 12 07:27:13 I feel like smthing i wrong there Feb 12 07:36:17 siji doesn't look too bad, depending on what env settings you have (e.g. machine) what makes you think it is wrong ? Feb 12 07:36:57 oh ya beagle board Feb 12 07:37:01 i added there Feb 12 07:37:51 actually am cross checking all the conf files Feb 12 07:39:50 siji I suggest you follow this: http://www.angstrom-distribution.org/building-angstrom quite closely Feb 12 07:40:20 only change you need to make as far as i recall is editing the local.conf as it mentions /OE not ${OETREE}/OE Feb 12 07:40:27 on a few places Feb 12 07:42:33 ok Feb 12 07:50:08 Hello, does anyone know how to build gcj for arm (at91sam9260) ? Feb 12 07:55:05 ,pls have a look on my sainty.conf Feb 12 07:55:18 BB_MIN_VERSION = "1.8.18" Feb 12 07:55:18 SANITY_ABIFILE = "${TMPDIR}/abi_version" Feb 12 07:55:18 SANITY_PRFILE = "${TMPDIR}/distro_pr" Feb 12 07:55:18 INHERIT += "sanity" Feb 12 07:56:09 too short you need two more lines: Feb 12 07:56:28 i take that back was not seeing the end of the window Feb 12 07:56:54 your sanity.conf is ok, but really the only conf file you should touch is local.conf Feb 12 07:57:20 all others should be left untouched Feb 12 07:57:28 git diff can tell you if it is changed Feb 12 07:57:40 man git-diff Feb 12 07:57:48 ok Feb 12 07:59:03 now the error is Feb 12 07:59:04 ERROR: Openembedded's config sanity checker detected a potential misconfiguration. Feb 12 07:59:05 Either fix the cause of this error or at your own risk disable the checker (see sanity.conf). Feb 12 07:59:05 Following is the list of potential problems / advisories: Feb 12 07:59:05 Using dash as /bin/sh causes various subtle build problems, please use bash instead. Feb 12 08:00:31 I did sudo dpkg-reconfigure dash Feb 12 08:00:38 still the errors are there Feb 12 08:10:58 @siji when u did dpkg reconfigure did u disbale dash Feb 12 08:11:12 *disbale disable Feb 12 08:11:20 how? Feb 12 08:12:05 You mean by selecting no from the menu? Feb 12 08:12:18 yes Feb 12 08:12:36 yes i did Feb 12 08:12:40 so did u select no in the menu Feb 12 08:12:42 ok Feb 12 08:12:46 mv: cannot stat `/bin/sh.distrib': No such file or directory Feb 12 08:12:57 I got msg like these Feb 12 08:13:09 wow, never faced error like that Feb 12 08:13:16 oh :( Feb 12 08:13:39 if u type /bin/sh what happens Feb 12 08:13:45 in console Feb 12 08:13:52 k Feb 12 08:14:22 it's executing Feb 12 08:14:46 the mv: cannot.. error when does it occur Feb 12 08:14:50 on bitbake Feb 12 08:14:54 executuin Feb 12 08:15:01 or on dpkg reconfigure Feb 12 08:15:19 on dpkg reconfigure Feb 12 08:19:13 ,that got resolved Feb 12 08:19:28 i did sudo mv /bin/sh /bin/sh.dash Feb 12 08:19:30 and Feb 12 08:19:51 sudo ln -s /bin/bash /bin/sh Feb 12 08:20:02 ver nice Feb 12 08:20:19 so now try bitbake... Feb 12 08:20:20 new errors Feb 12 08:20:26 trying Feb 12 08:20:27 like... Feb 12 08:20:51 NOTE: Psyco JIT Compiler (http://psyco.sf.net) not available. Install it to increase performance. Feb 12 08:21:02 I think need to install it am i right? Feb 12 08:21:24 visit openmebdded wiki Feb 12 08:21:28 ok Feb 12 08:21:32 there are listed many prerequisites Feb 12 08:21:42 u need to have installed Feb 12 08:21:53 before bitbaking anything, ok Feb 12 08:22:48 ok Feb 12 08:29:43 hi, i want to print a debug message in kernel_do_stage() I added print 'hi' as the first line of kernel_do_stage, but I get a warning about unknown mime type for "hi" and no such file "hi". what is your python n00b doing wrong ?? Feb 12 08:34:16 do_stage() is not a python function Feb 12 08:34:25 it's shell Feb 12 08:41:28 heh, yeah, it is unfortunate that "print" exists as this obscure binary Feb 12 08:41:51 * pb__ stabs mime-support Feb 12 08:46:53 mickey|zzZZzz: pb_ yeah found that out didn't know about the print binary Feb 12 08:47:04 * eFfeM is sometimes puzzled by what is python and what is shell Feb 12 08:47:22 eFfeM: if it says "python do_stage() { ...}" then it's python Feb 12 08:47:27 if it doesn't say python there, it's shell Feb 12 08:47:38 actually for me kernel_do_stage is not working without anything in the log so instrumenting it Feb 12 08:48:03 pb__, ah ok, the bbclasses still have quite some mysteries for me, thanks for the info Feb 12 08:48:48 actually since it is shell I can just do a set + v :-) Feb 12 08:49:45 set -v Feb 12 09:02:40 for Machine="qemuarm" I am doing a bitbake opie-image and am getting this sort of error Feb 12 09:02:41 http://pastebin.com/m514eb7a5 Feb 12 09:02:53 pls help me with this one Feb 12 09:05:57 anyone Feb 12 09:08:26 hm, for some reason the kernel build will not generate an include/asm symlink for me, anyone know if something has changed there (moving from 2.6.31 to 2.6.33rc3 Feb 12 09:10:34 http://pastebin.com/m514eb7a5 Feb 12 09:12:00 Spyzer: no idea on those pacakges but maybe your MC&DriverOnlyInstallers.zip is truncated, I suggest removing it from the download dir and retry Feb 12 09:13:06 i can remove it from the download directory but then what if bitbake downloads it again, how will i tell bitbake not to ignore MC&DriverOnlyInstaller Feb 12 09:14:30 Spyzer: don't understand about ignoring, it will download and reapply, you might want to bitbake -c clean the package that uses it though Feb 12 09:14:35 hi RP__ Feb 12 09:14:41 morning all Feb 12 09:14:51 morning Feb 12 09:15:07 plus where is the location from where i have to delete, by download directory u mean build/sources or something else Feb 12 09:15:09 ? Feb 12 09:15:51 bitbake -c.. should i do that in build directory Feb 12 09:16:30 you do have a downloads dir somewhere, depends a little bit on config, use find to locate the zip file Feb 12 09:17:01 and bitbake -c can be done whereever you want to if your env settings are ok, but you need to provide the package Feb 12 09:26:01 @eFfem: So i should do bitbake -c opie-image bcoz the build was made for bitbake opie-image Feb 12 09:26:03 ?? Feb 12 09:27:34 Spyzer: bitbake -cclean for the failing package name Feb 12 09:27:59 ok Feb 12 09:28:26 failing package is something like spectrum-fw Feb 12 09:28:41 never done that actually, but will try if error shows, as of now i have deleted .zip file and am bitbaking again, till now no errors Feb 12 09:29:57 good morning Feb 12 09:31:08 morning Feb 12 09:33:36 hi Spyzer Feb 12 09:34:07 hi florian Feb 12 09:34:19 am doing bitbake for MACHINE=qemuarm" Feb 12 09:34:28 may be then i can emulate opie Feb 12 09:34:32 easily on qemu Feb 12 09:34:57 yes this should be possible... but iirc there is a quite good qemu emulation for the mini2440 too Feb 12 09:35:31 you know i tried everything for mini2440 supported at google code qemu version Feb 12 09:35:34 but to no avail Feb 12 09:35:46 i posted my problems here and other places Feb 12 09:36:00 but the only soln. i got was to build the "qemuarm" image Feb 12 09:36:11 but oh well.. anything is better than nothing Feb 12 09:43:55 anyone a hint on what is wrong here ? Feb 12 09:44:18 I am building a kernel and kernel_do_stage() does not work Feb 12 09:44:56 it says ASMDIR=`readlink include/asm` but there is no include/asm dir in my tree Feb 12 09:45:15 (and the kernel build properly and if I retrieve it from the image subdir it boots nicely) Feb 12 09:55:53 nevermind found it, had to pull a more recent kernel.bbclass Feb 12 10:21:09 @eFfem i did a bitbake -c clean spectrum-fw Feb 12 10:21:23 and now bitbake opie-image is kinda sttoped at this Feb 12 10:21:25 NOTE: package spectrum-fw-1.0-r1: task do_configure: Started Feb 12 10:21:32 the cursor is just blinking Feb 12 10:21:37 should i wait for it Feb 12 10:21:43 it has been 30 min alredy Feb 12 10:22:58 would you ple tellme Feb 12 10:23:09 what to do now Feb 12 10:23:35 Spyzer: i don't know this package at all, guess it requires some input, but you are building an image that i never build before Feb 12 10:24:08 you might want to look into the temp dir of the package to see what is in log.do_configure Feb 12 10:24:26 * eFfeM does not even know what spectrum-fw is Feb 12 10:24:57 maybe an emulator for zx spectrum? Feb 12 10:25:00 ok Feb 12 10:26:01 well GNUtoo i changed the machine to qemuarm and doing bitbake opie-image, first there were regarding spectrum-fw do_configure stopped Feb 12 10:26:06 failed i mean Feb 12 10:26:29 and so i cleared the files like bitbake -c clean spectrum-fw Feb 12 10:26:37 and then i did bitbake opie-image Feb 12 10:26:47 and? Feb 12 10:26:57 NOTE: Running task 5024 of 5046 (ID: 696, /home/abhishek/embedded/toolchain/openembedded/recipes/orinoco/spectrum-fw.bb, do_configure) Feb 12 10:26:59 NOTE: package spectrum-fw-1.0-r1: task do_configure: Started Feb 12 10:27:12 and this is there on the console for last 30 min or so Feb 12 10:27:29 can it take so much time.. Feb 12 10:27:38 shold i wait or should i ctrl+c Feb 12 10:27:44 Spyzer, if you think it takes too much time there is a solution Feb 12 10:27:47 and bitbake again Feb 12 10:27:48 no don't ctrl+c Feb 12 10:27:55 go into: Feb 12 10:28:19 tmpdir/work/armv-something/spectrum-fw_something/temp Feb 12 10:28:35 and look at the log.do_configure Feb 12 10:29:08 should i paste bin that, its 2000 lines Feb 12 10:29:54 but each line has the same sort of structure Feb 12 10:30:46 just have a brief look at it, all i can think is that my download speed is too slow, that is the reason Feb 12 10:30:48 http://pastebin.com/m5a0e4264 Feb 12 10:31:24 strange...it seem it's downloading something Feb 12 10:31:37 yeah i think that's the problem, before configuring, it has to actually download something to configure Feb 12 10:31:37 maybe related to what raster said yesterday Feb 12 10:31:52 maybe it dowloads macros Feb 12 10:32:26 oh, so i should wait, it means? Feb 12 10:33:40 I think so but...I think you/we should fix that dowload... Feb 12 10:33:43 previously it was giving different errors regarding this, http://pastebin.com/m514eb7a5 Feb 12 10:34:27 ok , but how do I do that? Feb 12 10:34:55 let's wait the dowload is nearly finished Feb 12 10:35:33 my connection is as good as a dial up i think we will have to wait much longer Feb 12 10:36:02 I'll meet you after some time, if you would be there Feb 12 10:36:04 ok? Feb 12 10:37:02 GNUtoo: haven't really found out why xinit is not in xfce46-image Feb 12 10:37:48 ok Feb 12 10:38:02 maybe I should have a look...but after commiting stuff Feb 12 10:38:15 xinit is a binary right Feb 12 10:38:23 it's not a lib... Feb 12 10:38:33 so it's not automatically added in the deps Feb 12 10:39:29 GNUtoo: it is a binary, i would expect it to be part of the image Feb 12 10:39:54 yes but what if someone forgetten to add it in RDEPENDS Feb 12 10:40:12 somewhere Feb 12 10:40:14 03Martin Jansa  07org.openembedded.dev * rab3babc960 10openembedded.git/recipes/xorg-xserver/ (files/ac_check_file.cross.patch xserver-xorg_git.bb): Feb 12 10:40:14 xorg-xserver_git: remove AC_CHECK_FILE check Feb 12 10:40:14 * configure says that AC_CHECK_FILE is not supported for cross building Feb 12 10:40:14 * that DEFAULT_FONT_PATH is later set by --with-default-font-path from Feb 12 10:40:14 recipe Feb 12 10:40:15 Signed-off-by: Martin Jansa Feb 12 10:40:16 03Martin Jansa  07org.openembedded.dev * rf0184ebb8e 10openembedded.git/ (8 files in 5 dirs): Feb 12 10:40:16 xorg: add latest xorg packages for 2010-02-10 and xserver-xorg git recipe Feb 12 10:40:17 Signed-off-by: Martin Jansa Feb 12 10:40:27 btw I downloaded from your feed, it said it required X , i had no X only Xvesa, made a symlink and got some error on display forgot about what, but didn't manage to proceed Feb 12 10:40:56 it might be interesting to see what you have as output of opkg list-installed and compare with what I have Feb 12 10:41:10 (but can't do that right now as I am not near the eee) Feb 12 10:42:59 eFfeM, Xvesa seem so kdrive Feb 12 10:44:47 Anyone good at problem solving? Trying to fix: mtd-utils_1.2.0+git.bb do_populate_staging failed .... stage-manager, line 34, in read_cache cache[data[0]]['ts'] = int(data[1]) IndexError: list index out of range Feb 12 10:46:17 eFfeM, http://pastebin.com/d17244a8d Feb 12 10:48:32 GNUtoo will check the packages later, don't understand your remark on xvesa, is it ok? do you have a symlink to x ? Feb 12 10:48:55 no it's not ok Feb 12 10:49:05 eFfeM, do you know what kdrive is? Feb 12 10:49:40 I'll check something Feb 12 10:49:42 http://www.pastey.net/132889 for my bitbake build problem Feb 12 10:53:01 eFfeM, http://en.wikipedia.org/wiki/Kdrive Feb 12 10:53:27 also kdrive is unmaintained Feb 12 10:56:28 yeah, the thing compiled without erros, whew!! Feb 12 11:01:07 so GNUtoo now that i have qemu ext2 image so when i am typing the same qemu-system-arm command as from yesterday, but same errors re occuring ,help Feb 12 11:01:22 ../qemu-system-arm -M mini2440 -hda minimal-opie-image-eglibc-ipk-.dev-snapshot-20100212-qemuarm.rootfs.ext2 Feb 12 11:01:43 Spyzer, the -M mini2440 is wrong Feb 12 11:01:50 look at angstrom howto for that Feb 12 11:02:00 I'll try to find it Feb 12 11:02:19 http://www.linuxtogo.org/gowiki/Angstrom/qemu-system-arm Feb 12 11:02:23 but Feb 12 11:02:35 change: -append "root=/dev/sda" for -append "root=/dev/sda rw" Feb 12 11:07:12 eFfeM, maybe you know Xfbdev ? Xvesa is the same but for vesa Feb 12 11:08:41 WOW, everything is working , thanks a lot everyone and specially you GNUtoo, Wohooooo!!!! Feb 12 11:09:03 nice Feb 12 11:11:16 I've changed the wiki to rw Feb 12 11:11:31 ("root=/dev/sda"->"root=/dev/sda rw") Feb 12 11:15:25 GNUtoo, kdrive page was very slow, i think i've heard aobut it before Feb 12 11:15:36 ok Feb 12 11:15:46 xvesa gives some x-like display but came with a dialog with two buttons with unreadable text Feb 12 11:15:47 it's an alternative X server Feb 12 11:15:57 so better using Xorg Feb 12 11:15:58 tried all combinatiions but didn't get a good display Feb 12 11:16:06 Xorg should work Feb 12 11:16:18 I don't know why xvesa got into your image Feb 12 11:16:45 xinit wanted X, didn't have that so either i need different settings/parameters/arugments or something else Feb 12 11:16:46 according to image file and machine config they should have been ....wait a sec.... Feb 12 11:17:02 famous last words ? :-) :-) Feb 12 11:17:06 ahhh,,,, Feb 12 11:17:18 look in i686-generic.conf Feb 12 11:17:28 maybe it was modified or I didn't saw that: Feb 12 11:17:50 PREFERRED_PROVIDER_virtual/xserver = "xserver-kdrive" and XSERVER="xserver-kdrive-vesa" Feb 12 11:17:58 so basically it does that: Feb 12 11:18:03 it's set in i686-generic.conf Feb 12 11:18:10 that is there Feb 12 11:18:28 and then PREFERRED_PROVIDER_virtual/xserver ?= "xserver-xorg" Feb 12 11:18:31 see the ?= Feb 12 11:18:35 so it's not set again Feb 12 11:18:40 and so on until the image Feb 12 11:18:51 where it's not set again because of ?= here too Feb 12 11:19:05 so how do we fix that Feb 12 11:19:27 what's the parse order? Feb 12 11:20:11 it is possible to override in local.conf Feb 12 11:20:16 bacause if (eee701.conf is parsed after i686-generic.conf) and before other stuff we're doomed Feb 12 11:20:26 where is this last preferred_provider line from Feb 12 11:20:31 yes but I don't know if it works Feb 12 11:20:43 I don't know the correct order Feb 12 11:20:47 of the parsing Feb 12 11:20:51 sorry Feb 12 11:20:54 I forgetten Feb 12 11:21:04 eee701.conf includes i686-generic Feb 12 11:21:26 but if you want xorg then ?= should be = Feb 12 11:21:46 not shure Feb 12 11:22:05 because if you want to override in image after you may not be able to do it Feb 12 11:22:33 because image has ?= Feb 12 11:22:42 mmm Feb 12 11:23:10 for thinking about this issue we need the correct parsing order Feb 12 11:23:35 s./because image has ?=/ Feb 12 11:23:39 let's make a scenario Feb 12 11:23:48 what if: Feb 12 11:24:11 the image came before machine conf Feb 12 11:24:19 if you use ?= the i686-generic always wins Feb 12 11:24:25 I know Feb 12 11:24:32 you can override in local.conf Feb 12 11:24:37 I think so Feb 12 11:24:45 but that's not a good option either Feb 12 11:24:55 we need help Feb 12 11:24:58 Hello, does anyone know how to build gcj for arm ? Feb 12 11:25:10 if you want to fix it quickely simply add - in eee701.conf Feb 12 11:25:16 but I'm not shure it's the correct fix Feb 12 11:25:34 if image came before than machine conf: Feb 12 11:25:44 image set it,then machine.conf set it Feb 12 11:25:53 but image should ovveride machine Feb 12 11:26:06 and if machine is not ?= it wouldn't be able to do it Feb 12 11:26:14 for instance if you have a : Feb 12 11:26:21 micro-x11-image Feb 12 11:26:27 with kdrive instead of xorg Feb 12 11:26:34 and want kdrive in that image... Feb 12 11:26:50 and another image with xorg... Feb 12 11:27:01 FCO__, maybe there are docs in the wiki Feb 12 11:27:11 FCO__, also use org.openembedded.dev Feb 12 11:27:43 http://wiki.openembedded.net/index.php/Java Feb 12 11:30:23 GNUtoo did you do anything special to get your system running (apart from installing grub and the X thing) Feb 12 11:31:26 GNUtoo did a quick grep on XSERVER in conf/machine, eeepc is the only one that uses ?= all others use = Feb 12 11:31:47 actually almost the only one Feb 12 11:32:03 overlooked om-gta* Feb 12 11:33:37 ah ok Feb 12 11:34:20 so maybe fix locally and ask for review Feb 12 11:34:36 or maybe I should ask for review because I know the problem better Feb 12 11:58:46 03Graeme Gregory  07org.openembedded.dev * rcc57f26d64 10openembedded.git/recipes/gnome/gconf_2.26.2.bb: gconf_2.26.2.bb : add --enable-gtk to ensure consistant packages Feb 12 11:58:56 03Graeme Gregory  07org.openembedded.dev * r4e45a92b85 10openembedded.git/: Merge branch 'org.openembedded.dev' of git+ssh://git.openembedded.org/openembedded into org.openembedded.dev Feb 12 12:09:32 XorA: do you know about "git rebase -i origin"? Feb 12 12:09:51 Laibsch: yes and do you know rebase is bad? Feb 12 12:10:15 that is not true Feb 12 12:10:23 it's OK for local commits Feb 12 12:10:32 stuff that hasn't been pushed yet Feb 12 12:11:06 Laibsch: my tree is public rebase is bad Feb 12 12:11:11 "git rebase -i origin" should never ever create a problem when pushing to origin Feb 12 12:12:05 XorA: I think you are mistaken, but it's your choice. Feb 12 12:12:20 Laibsch: why get so anal about normal operation of git? Feb 12 12:12:29 I am not getting anal Feb 12 12:12:35 I asked a simple question Feb 12 12:13:03 Laibsch: and if you look on www.xora.org.uk/git you will see I expose my tree for other Angstrom devs before it hits OE Feb 12 12:13:48 GNUtoo eee701 is new, you made it, i think you and me are the only users atm and it is similar to what is in other recipes, so my suggestion is just to change it Feb 12 12:14:55 XorA: You could still push a rebased rather than a merged commit. But again, it's your choice. Feb 12 12:15:19 and no, it's not the end of the world Feb 12 12:15:22 Laibsch: then git pull fails as a non fast forward Feb 12 12:15:28 I was just wondering if you're aware of it or not Feb 12 12:29:56 ok thanks Feb 12 12:41:55 mmm libunique fails: http://pastebin.com/dd25c75c is it because of version mismatch between libunique and glib? Feb 12 12:42:48 mmm hrw|gone is gone Feb 12 12:46:48 eFfeM, btw you made a commit replacing inherit native by BBCLASSEXTENDS="native", I think I've forgetten what BBCLASSEXTENDS does Feb 12 12:47:12 I may have read it very quickely long time ago and never used it Feb 12 12:47:55 GNUtoo it can be used in a non-native recipe and then the same recipe also can be used to produce the -native variant Feb 12 12:48:10 nice thanks a lot Feb 12 12:48:28 so x.bb and x-native.bb can be replaced by x.bb with BBCLASSEXTENDS added Feb 12 12:48:34 np, my pleasure Feb 12 12:48:34 ok Feb 12 12:53:24 03Thomas Zimmermann  07org.openembedded.dev * r3b9b55a651 10openembedded.git/recipes/tasks/task-shr-feed.bb: Feb 12 12:53:24 task-shr-feed: add mpd (solves SHR bug: #946) Feb 12 12:53:24 Signed-off-by: Martin Jansa Feb 12 12:59:18 03Martin Jansa  07org.openembedded.dev * raf52177ab7 10openembedded.git/conf/distro/shr.conf: Feb 12 12:59:19 shr: add bluetooth to DISTRO_FEATURES to get bluetooth modules, remove pci, pcmcia, nfs, smbfs, also remove kernel-modules from DISTRO_EXTRA_RRECOMMENDS which are pulled by task-base Feb 12 12:59:19 Signed-off-by: Martin Jansa Feb 12 12:59:19 03Martin Jansa  07org.openembedded.dev * r6b9ab500f7 10openembedded.git/conf/machine/om-gta02.conf: Feb 12 12:59:20 om-gta02: install only really important kernel-modules Feb 12 12:59:20 Signed-off-by: Martin Jansa Feb 12 13:11:48 GNUtoo, thank you. I will try Feb 12 13:12:15 ok Feb 12 13:21:06 eFfeM, I've fixes that depends on ?= in machines.conf so I'll send a mail Feb 12 13:22:51 GNUtoo ok Feb 12 13:23:08 eFfeM, we could also try to move the require at the end of the file Feb 12 13:24:24 GNUtoo compare with what others do Feb 12 13:24:31 other are wrong Feb 12 13:24:35 for ?= Feb 12 13:24:40 I had the issue with bug device Feb 12 13:25:12 too bad that hrw|gone is gone Feb 12 13:25:36 maybe I'll try the fix of putting on the botom of the file Feb 12 13:25:43 I saw that in other files Feb 12 13:25:52 fine with me Feb 12 13:26:36 but if it doesn't work Feb 12 13:26:49 and think of it the other way: basically it is our party (and mostly yours), so feel free to do it in a way that you feel best, you're not harming others by changing eee701.conf Feb 12 13:27:04 just give it a try (or i will do at some later point in time) Feb 12 13:27:09 other have a problem too....that's why I'm writing a mail Feb 12 13:27:16 for instance bug+e-image Feb 12 13:27:21 bug is a machine Feb 12 13:33:49 it doesn't change a thing...I write the mail Feb 12 13:34:11 hi Feb 12 13:35:30 eFfeM, sorry you were right about = Feb 12 13:35:42 hrw, hi, just to check parse order: Feb 12 13:36:09 i686-generic.conf,then eee701.conf then xfce46-image is the right order? Feb 12 13:36:23 GNUtoo: ? Feb 12 13:36:44 hrw, we have issues with xorg/kdrive variable Feb 12 13:36:53 so we want to know parsing order Feb 12 13:37:16 let me guess - i686-generic or eee701 set preference to xserver-kdrive? Feb 12 13:37:25 and both are same arch? Feb 12 13:37:45 hrw, i686-generic : kdrive,then eee701 require i686-generic but set xorg Feb 12 13:37:48 we want xorg Feb 12 13:38:06 but I'm not shure = wouldn't prevent an image for overriding after that Feb 12 13:38:23 GNUtoo just try Feb 12 13:38:29 = should set Feb 12 13:38:39 so that's why I need parsing order to see clear in this problem Feb 12 13:39:25 GNUtoo: or you need to do "bitbake --help" and check for "-e" option Feb 12 13:39:30 ok Feb 12 13:41:21 eFfeM, I've changes in my git tree that indicates that the problem is real and that I already faced it with machine=bug and image = e-image Feb 12 13:41:37 ok, will read your email Feb 12 13:41:42 not sent yet Feb 12 13:41:49 I'm trying to see clear before Feb 12 13:42:08 btw, I have a new kdrive mostly working in Poky in case anyone is interested Feb 12 13:43:47 RP__: it did not died yet? Feb 12 13:44:34 hrw: You can still build kdrive from the xorg-xserver and it still works... Feb 12 13:44:42 good Feb 12 13:45:05 There are some fixes it would be good to upstream Feb 12 13:47:26 I'm trying test case Feb 12 13:53:43 it seem to work....I'll put some =,sorry for the noise Feb 12 13:54:33 strange issue Feb 12 13:54:40 it didn't work before but now it does Feb 12 14:01:40 maybe the thing whas that the image hardcoded kdrive not depending on xserver Feb 12 14:01:53 maybe Feb 12 14:02:00 because e-image does it Feb 12 14:02:08 which is a bug Feb 12 14:04:21 wait a sec.... Feb 12 14:04:24 yes I know Feb 12 14:04:51 it parses local.conf before machine.conf Feb 12 14:04:55 maybe that was the issue Feb 12 14:05:25 you define XSERVER in local.conf and it's not taken into account Feb 12 14:11:59 has anyone seen race conditions with packaged staging? http://thread.gmane.org/gmane.comp.handhelds.openembedded/30040 Feb 12 14:30:36 denix0 i've seen some other issues ; eg packages that upon restore are marked as invalid or if the kernel is restored it still wants to run kernel_do_deploy, didn't really find a solution yet Feb 12 14:31:54 eFfeM: my problem seems to be with creating pstage, not using it... Feb 12 14:32:14 making a clean build - no TMPDIR/DEPLOY_DIR Feb 12 14:33:39 denix yeah but i think the root cause of the kernel do_deploy is that the timestamp for packaged staging is older than the one from do_deploy Feb 12 14:33:56 whici is somewhat similar to your tar problem Feb 12 14:35:00 ah, I think Tartarus has fixed it recently? Feb 12 14:36:27 eFfeM: any comments on IPKG_BUILD.XXX directory clash? Feb 12 14:38:06 denix0 don't know/haven't heard of that problem Feb 12 14:52:00 read OEML in few minutes - sent patchset for review Feb 12 14:53:50 eFfeM: please try to be polite when you post on the mailing list. there is no need for abrupt comments like the one you sent this morning. Feb 12 14:56:08 damn I am tired today Feb 12 14:56:47 pb_ the mail to marcos ? Feb 12 14:56:51 yeah Feb 12 14:57:16 it seemed a bit of a harsh response to a reasonable suggestion Feb 12 14:57:27 hm, ok will let it pass by next time Feb 12 14:57:56 woglinde: can I ask about classpath? Feb 12 14:58:01 actually i consider it somewhat rude to provide a patch to a very basic utility and in the cover mail indicate that you didn't even test it yourself Feb 12 14:58:20 er, I think you misunderstood his mail Feb 12 14:58:30 his patch was for a completely different issue than the suggestion that you responded to. Feb 12 14:58:39 did you actually look at the patch? Feb 12 14:58:42 eFfeM: so I am rude too - check my latest patchset on oeml Feb 12 15:01:04 pb_ obviously from my reply i did not really look at the patch, just did, and i understand what you mean; then again if you only see his msg text and the files he is changing (which is roughly what is visible in my window), it is not more than understandable that one thinks the patch is on the SRC_URI move Feb 12 15:01:48 it helps if people state somewhere in the mail where the patch is about and do not address completely different issues Feb 12 15:46:35 hi folks Feb 12 15:46:55 is someone aware of openssl which can be used fro crosscompiling for mips? Feb 12 15:47:12 version 0.9.8g preferably :) Feb 12 15:47:28 Blinder: what is the error you see when using bitbake openssl? Feb 12 15:48:01 zecke: ah, sorry, was rater OT, generally speaking, not oe specific, my bad, sorry Feb 12 16:01:12 hi, Feb 12 16:01:22 I've the wesnoth_1.6.5 recipe Feb 12 16:01:45 I'd like to have a default EXTRA_OECMAKE with default settings Feb 12 16:01:59 and to override that for machines with big screen or eee701 Feb 12 16:02:04 should I do that with python Feb 12 16:02:05 ? Feb 12 16:02:11 like this for instance: Feb 12 16:03:15 ${@base_conditional('', '', '', ', d)}" Feb 12 16:03:37 the variable name would be bigscreen or eee701 Feb 12 16:03:48 but I don't even know what how big is bigscreen Feb 12 16:04:06 it would start at 800x480 in the case of wesnoth Feb 12 16:04:30 before I had something more hackish Feb 12 16:04:48 playing with variables and overriding Feb 12 16:05:01 but that required to create a temporary variable Feb 12 16:05:39 I wonder what to do with -DENABLE_LOW_MEM=ON Feb 12 16:05:56 if someone has an eee701 with 2GB of ram I bet that's not low mem Feb 12 16:06:17 but on 64MB it is low mem Feb 12 16:06:23 yes Feb 12 16:06:51 is base conditional ok for bigscreen? Feb 12 16:07:25 because I could keep lowmem but big screen is more than an issue Feb 12 16:07:40 it shrink graphics,you have issues with GUI etc... Feb 12 16:07:52 so better having it only when realy necessary Feb 12 16:08:38 I'm not even shure 64M would run wesnoth Feb 12 16:08:47 110 is ok Feb 12 16:09:45 * XorA wishes there was a git link command Feb 12 16:10:15 What would that do? Feb 12 16:10:22 ln -s ? Feb 12 16:11:38 broonie: a safe ln -s Feb 12 16:13:01 bye all Feb 12 16:16:24 XorA: you mean it'd work on filesystems that don't support links, by doing a secondary checkout of the file or something? Feb 12 16:16:27 that could be interesting Feb 12 16:22:34 kergoth: yes, rather than just recording a ln -s it would actually associate the two files Feb 12 16:24:43 anyone knows what are smallscreen or bigscreen? Feb 12 16:24:50 or how I can say: Feb 12 16:27:17 ${@base_conditional(''MACHINE_DISPLAY_something, 'smaller than 800x480', "-DGUI=tiny", "",d )}" Feb 12 16:28:49 MACHINE_GUI_CLASS, you mean? Feb 12 16:29:03 that's pretty much a historical artifact, I am not sure it is such a great idea to use it in new code. Feb 12 16:29:31 originally, I think "smallscreen" meant 320x240, and "bigscreen" was VGA or anything larger. Feb 12 16:29:39 pb_, ok so what should I do....I want to use tiny gui only with resolution smaller than 800x480 Feb 12 16:29:40 ok Feb 12 16:29:50 so not suitable for my goal Feb 12 16:30:12 there's MACHINE_DISPLAY_HEIGHT_PIXELS etc Feb 12 16:30:16 (I suspected it looking at gumstix commits) Feb 12 16:30:18 ok Feb 12 16:30:21 I know that Feb 12 16:30:25 which I think might be an opieism but is probably your best option Feb 12 16:30:31 hi pb Feb 12 16:30:35 hi blindman Feb 12 16:30:42 so I should make an anonymous python function? Feb 12 16:30:47 or is there something easier? Feb 12 16:31:12 yes, you probably want to write a python expression Feb 12 16:31:17 I could also make a temporary variable and override it per machine but it would be ugly Feb 12 16:31:20 ok Feb 12 16:31:45 so there is no python like @base_conditional and I should write a function...ok Feb 12 16:32:13 *like @base_conditional that does that Feb 12 16:32:20 it doesn't need to be an actual function, just a ${@...} expression would be enough Feb 12 16:33:08 ah wow Feb 12 16:33:31 so I put my function in @ and it would be ok...thanks Feb 12 16:34:26 I'll try base_confditional Feb 12 16:36:42 I'll craft something Feb 12 16:36:49 base_conditional won't work Feb 12 16:41:11 nice Feb 12 16:41:12 base_ifelse Feb 12 16:42:04 yeah, you can use base_ifelse if you want, though it doesn't particularly buy much. Feb 12 17:07:26 btw thanks a lot Feb 12 17:18:50 Tartarus: ping me or denix0 when you are back Feb 12 17:21:54 ah ok lol...I was doing ${@base_ifelse( "( bb.data.getVar('MACHINE_DISPLAY_WIDTH_PIXELS', d, 1) < 800 ) or ( bb.data.getVar('MACHINE_DISPLAY_HEIGHT_PIXELS',d,1) < 480 )","-DGUI=tiny","")} \ Feb 12 17:22:01 but condition is not a string Feb 12 17:25:26 pb_, thanks a lot Feb 12 17:29:05 hi hrw|gone do we put kernel-modules in EXTRA_RDEPENDS or in EXTRA_RECOMMANDS for machine bug ? Feb 12 17:43:08 what is the best way to create a dir after unpack but before the build starts? Feb 12 17:43:29 tried an append to do_unpack with midir in it but that just gives a syntax error .... Feb 12 17:47:24 eFfeM, what's the syntax error? maybe add a task because you may not be able to mix python and bash Feb 12 17:48:35 btw fix comming Feb 12 17:48:39 for eeepc Feb 12 17:54:38 GNUtoo guess it is indeed python moved to do_configure_prepend, but that didn't help too much either Feb 12 17:54:46 need a better plan :-) Feb 12 17:55:38 couldn't you add a task? Feb 12 17:55:50 or do a python mkdir Feb 12 17:56:34 os.mkdir: mkdir(path [, mode=0777]) Feb 12 18:03:18 g.m. Feb 12 18:05:09 hi recalcati Feb 12 18:05:41 GNUtoo actually there is more work needed, it is a recipe that desires to be build in a separate dir, but if i do the mkdir the paches do not apply Feb 12 18:06:05 need a better solution, will look into it Feb 12 18:07:37 I'm copying akita for my pxa270 rework. Starting from stable/2009 is good (gcc4.2.4) or dev branch can be better (gcc4.3.3) . the base image compilation works in bot cases . Feb 12 18:09:21 ok Feb 12 18:09:50 mmm that doesn't work as expected: ${@base_ifelse( ( ( bb.data.getVar('MACHINE_DISPLAY_WIDTH_PIXELS', d, 1) < 800 ) or ( bb.data.getVar('MACHINE_DISPLAY_HEIGHT_PIXELS',d,1) < 480 ) ),"-DGUI=tiny","")} Feb 12 18:10:57 it always do "" Feb 12 18:11:05 s/do/output Feb 12 18:13:50 think i already solved my problem, there was a variable for it :-) Feb 12 18:14:24 ok Feb 12 18:14:29 what was the variable? Feb 12 18:18:11 what is d in bb.data.getVar(,d,1) Feb 12 18:18:24 because in the python console It's not defined Feb 12 18:21:38 of course it isn't defined Feb 12 18:21:44 it's the bitbake/oe metadata. Feb 12 18:22:06 ok Feb 12 18:22:20 how can I do that in python shell: Feb 12 18:22:23 ${@base_ifelse( ( ( bb.data.getVar('MACHINE_DISPLAY_WIDTH_PIXELS', d, 1) < 800 ) or ( bb.data.getVar('MACHINE_DISPLAY_HEIGHT_PIXELS',d,1) < 480 ) ),"-DGUI=tiny","")} Feb 12 18:22:39 because something is wrong somewhere and it doesn't work Feb 12 18:22:45 as expected Feb 12 18:22:52 basically it always returns "" Feb 12 18:24:42 03Graeme Gregory  07org.openembedded.dev * rbe1c605149 10openembedded.git/conf/machine/omapzoom.conf: Feb 12 18:24:42 omapzoom.conf : this is an omap3 device using same kernel as most others Feb 12 18:24:42 Change it to use the omap3.inc so it gets the same features beagle/zoom2/hawk Feb 12 18:24:42 get. Feb 12 18:25:02 GNUtoo are these vars defined Feb 12 18:25:05 ok got it Feb 12 18:25:47 >>> d = init() Feb 12 18:25:51 so I must find init Feb 12 18:35:29 got it I think Feb 12 18:35:34 but it doesn't return anything Feb 12 18:35:40 d = bb.data.init() Feb 12 18:40:27 03Frans Meulenbroeks  07org.openembedded.dev * ra6cbed9c4b 10openembedded.git/recipes/wt/wt_2.2.3.bb: Feb 12 18:40:27 wt: removed do_stage Feb 12 18:40:27 Signed-off-by: Frans Meulenbroeks Feb 12 18:40:29 03Frans Meulenbroeks  07org.openembedded.dev * r3e5ac9a9cb 10openembedded.git/ (3 files in 2 dirs): Feb 12 18:40:29 wt3: build outside the source tree Feb 12 18:40:29 this builds wt3 outside the source tree. Feb 12 18:40:29 not doing so is not the recommended solution and actually breaks in 3.1.0 Feb 12 18:40:29 fixed it to use a build dir Feb 12 18:40:30 while at it also inlined the checksums and removed them from checksums.ini Feb 12 18:40:30 Signed-off-by: Frans Meulenbroeks Feb 12 18:40:32 03Frans Meulenbroeks  07org.openembedded.dev * r6adbcf3963 10openembedded.git/recipes/wt/ (3 files in 2 dirs): Feb 12 18:40:32 wt3: added recipe for 3.1.0a Feb 12 18:40:32 Signed-off-by: Frans Meulenbroeks Feb 12 18:46:39 sigh Feb 12 18:46:52 how do I make it parse? Feb 12 18:49:47 getVar gets a variable from the metadata, as parsed from a recipe. do you honestly think a getVar on an empty datastore created with bb.data.init() is going to return anything? Feb 12 18:49:51 ahhh Feb 12 18:49:52 ok Feb 12 18:50:01 bb.data.getVar('MACHINE_DISPLAY_WIDTH_PIXELS', d, 1) < 800 is false Feb 12 18:50:06 because it should be Feb 12 18:50:09 bb.data.getVar('MACHINE_DISPLAY_WIDTH_PIXELS', d, 1) < "800" Feb 12 18:50:25 int(bb.data.getVar("MACHINE_DISPLAY_WIDTH_PIXELS", d, 1) < 800 Feb 12 18:50:26 is better Feb 12 18:50:35 kergoth, indeed that was my problem I wondered how to make it parse Feb 12 18:50:36 integer comparison rather than string comparison Feb 12 18:50:42 ok Feb 12 18:50:47 indeed Feb 12 18:50:54 erm, missed a closing paren, but you get the point :) Feb 12 18:51:49 ok thanks a lot Feb 12 18:52:48 btw for the record is it possible to populate the datastore whitin python because I thought it was already populated when I did parse in bitbbake -i Feb 12 19:01:25 GNUtoo: rrecommends Feb 12 19:01:38 ok thanks Feb 12 19:07:15 eFfeM-away: can you switch to use utf-8 on default on kirkwoods? thx Feb 12 19:30:40 mickeyl: good morning Feb 12 19:31:57 morning pb__ Feb 12 19:45:36 JaMa, I think my patch is not needed anymore with your imlib2 bump Feb 12 19:45:41 I won't push it then Feb 12 19:46:12 GNUtoo: ok, good to know, thanks Feb 12 19:46:44 GNUtoo: I'll bump it a bit more after reboot as epsilon patch is already aplied upstream, so to remove it from oe.dev Feb 12 19:47:11 ok Feb 12 19:47:19 epsilon is a machine? Feb 12 19:48:42 no old deprecated efl package Feb 12 19:56:58 ok Feb 12 19:57:21 I check my bug changes bitbaking a modified illume-image and I push Feb 12 19:57:45 JaMa, do you think I could remove all remaining phone bits from illume-image? Feb 12 19:57:55 because that is commented: Feb 12 19:58:08 # ${ZHONE_INSTALL} \ Feb 12 19:58:27 which installed zhone+fso Feb 12 19:58:39 because I'd like to use the image for my bug device Feb 12 20:10:27 are there any up-to-date discussions about checksums.ini? Feb 12 20:10:32 GNUtoo: I agree, with name like this it should stay as general as posible Feb 12 20:10:40 ok Feb 12 20:12:33 03Klaus Kurzmann  07org.openembedded.dev * r550d4902f4 10openembedded.git/recipes/openssh/openssh-5.2p1/shr/sshd_config: Feb 12 20:12:33 openssh: allow empty passwords in sshd_config for SHR Feb 12 20:12:33 because users insist on it :( Feb 12 20:12:33 Signed-off-by: Klaus Kurzmann Feb 12 20:15:27 03Michael 'Mickey' Lauer  07org.openembedded.dev * rc9fbceb7c3 10openembedded.git/ (5 files in 2 dirs): freesmartphone: rename mickeydbus(2) to mdbus(2), fso-term is now called mterm2 Feb 12 20:15:38 03Michael 'Mickey' Lauer  07org.openembedded.dev * rfac2ce0bc2 10openembedded.git/recipes/images/fso2-console-image.bb: fso2-console-image: catch up with upstream renaming Feb 12 20:15:40 03Koen Kooi  07org.openembedded.dev * r5ab23a5665 10openembedded.git/ (conf/checksums.ini recipes/perl/perl_5.8.8.bb): perl 5.8.8: add missing checksums Feb 12 20:38:48 mickeyl: for your interest, we agreed a sale of our house yesterday. so far so good. Feb 12 20:40:18 pb__: ah very good. congrats! Feb 12 21:05:04 JaMa, do you use waker? Feb 12 21:05:16 http://pastebin.com/d55e35567 Feb 12 21:06:48 GNUtoo: no never.. but IIRC Ecore_list is now Eina_list, but I'm not sure.. Feb 12 21:07:00 ok Feb 12 21:09:21 do someone use /etc/ssl/certs on target? there is curl-certs package as part of curl but in latest version is empty => not packaged Feb 12 21:10:39 mmm no... Feb 12 21:10:47 for ssl certs Feb 12 21:13:02 03Martin Jansa  07org.openembedded.dev * rdc115b9b64 10openembedded.git/ (5 files in 3 dirs): Feb 12 21:13:02 shr-image: move some stuff from task-shr-minimal to task-shr, drop unneeded and make shr-lite-images smaller Feb 12 21:13:02 Signed-off-by: Martin Jansa Feb 12 21:13:03 03Martin Jansa  07org.openembedded.dev * r79b6c7fe7b 10openembedded.git/ (3 files in 3 dirs): Feb 12 21:13:03 EFL: bump SRCREV a bit, remove applied patch Feb 12 21:13:03 Signed-off-by: Martin Jansa Feb 12 21:13:03 03Martin Jansa  07org.openembedded.dev * rdcfdafaa68 10openembedded.git/recipes/tasks/ (task-shr-feed.bb task-shr.bb): Feb 12 21:13:03 task-shr: catch up with upstream renaming Feb 12 21:13:04 Signed-off-by: Martin Jansa Feb 12 21:16:05 mmm Feb 12 21:16:06 http://pastebin.com/d1f2481a2 Feb 12 21:29:34 GNUtoo|oeee, Ecore_List is obsolete, the deveolper should uese Eina_List instead Feb 12 21:30:04 03Sebastian Krzyszkowiak  07org.openembedded.dev * r41f1d9de92 10openembedded.git/recipes/webkit/webkit-efl_git.bb: Feb 12 21:30:04 webkit-efl: don't include error page in package Feb 12 21:30:04 Signed-off-by: Martin Jansa Feb 12 21:31:13 ok Feb 12 21:41:54 playya_, I fixed the first failure but the second... Feb 12 21:42:06 I changed ecore->eina and did -leina Feb 12 21:42:08 no success Feb 12 21:42:19 let me check Feb 12 21:42:29 its in the e svn? Feb 12 21:42:33 ok thanks a lot Feb 12 21:42:49 I looked at last checkout and it's the same Feb 12 21:42:49 yes Feb 12 21:42:51 at: Feb 12 21:43:09 http://svn.enlightenment.org/svn/e/trunk/TMP/st/waker/ Feb 12 21:43:22 without the last / Feb 12 21:44:27 03Ruud Althuizen  07org.openembedded.dev * r150b89129e 10openembedded.git/conf/machine/ea3250.conf: Feb 12 21:44:28 ea3250.conf: New machine config Feb 12 21:44:28 Machine config for the LPC32XX based Embedded Artists development board Feb 12 21:44:28 Signed-off-by: Frans Meulenbroeks Feb 12 21:44:37 03Ruud Althuizen  07org.openembedded.dev * rfb1e00bdcd 10openembedded.git/ (5 files in 3 dirs): (log message trimmed) Feb 12 21:44:38 linux-ea3250: New recipe Feb 12 21:44:38 This is a kernel recipe for the LPC32XX (LPC3250) based Embedded Artists Feb 12 21:44:38 development board. It uses public patches provided by NXP and Embedded Feb 12 21:44:38 Artists. The other two patches provide drivers for the PWM capabilities Feb 12 21:44:38 and fix one NXP patch. Feb 12 21:44:39 It can be generalized for use with the PHYTEC LPC3250 development board. Feb 12 21:44:43 i think it needs some more love to get this running again GNUtoo Feb 12 21:44:48 ok Feb 12 21:45:19 if I remove the alarms from the illume-image will somebody get mad at me? and will it remove the waker dep? Feb 12 21:45:40 I'll try Feb 12 21:45:49 maybe there's another dependency to waker Feb 12 21:46:23 ok Feb 12 21:47:11 elementary-alarm \ => # elementary-alarm \ Feb 12 21:48:00 hrw|gone: will do Feb 12 21:49:00 remind me if it has not happened in a day or two (or make the change yourself, you have my blessing :-) ), planning to do it after the current build is finished but might forget :-[ Feb 12 21:58:16 GNUtoo: Try to fix the missing dependency Feb 12 21:58:28 ok Feb 12 21:58:29 commenting out will only hide the problem Feb 12 21:59:48 you need enlightenment/ecore libs Feb 12 22:00:01 probably thats not being pulled in Feb 12 22:00:12 I'll add eina dep Feb 12 22:00:22 ecore is pulled Feb 12 22:00:28 but not eina Feb 12 22:00:34 but I already had eina in workdir Feb 12 22:00:57 khem, i should segfault if you use Ecore_List Feb 12 22:01:40 ? Feb 12 22:02:27 it uses ecore_file_ls which returns an eina list Feb 12 22:03:56 may be there is some ecore configure option touse aina Feb 12 22:05:41 nope Feb 12 22:05:43 basically my problem is that I made something like 10 commits and I'd like to push....but before pushing I like to test the last commit Feb 12 22:05:51 which depend on an image Feb 12 22:05:51 eina should replace some parts of ecore Feb 12 22:06:01 and I chose illume-image Feb 12 22:06:14 that I'll fix later Feb 12 22:06:19 for removing phone Feb 12 22:06:28 functionality,and also for that bug Feb 12 22:06:36 this is my patch: http://pastebin.com/f5fdd0cf5 Feb 12 22:06:47 but it's not really tested ;) Feb 12 22:06:48 thanks a lot I'll try Feb 12 22:07:36 ok I'll do the testing Feb 12 22:12:09 at least it compiles and does not segfault Feb 12 22:12:16 ok Feb 12 22:12:27 thanks a lot Feb 12 22:42:06 03Martin Jansa  07org.openembedded.dev * r1fac183589 10openembedded.git/recipes/tasks/task-shr-feed.bb: Feb 12 22:42:06 task-shr-feed: add older mickeyterm to feed, newer mterm2 should be installed and build with shr*image Feb 12 22:42:06 Signed-off-by: Martin Jansa Feb 12 22:42:09 03Martin Jansa  07org.openembedded.dev * r5d719eb699 10openembedded.git/recipes/ (3 files in 2 dirs): Feb 12 22:42:09 task-cli-tools-python, illume-image, fso-image-nox: catch up with upstream renaming Feb 12 22:42:09 Signed-off-by: Martin Jansa Feb 12 22:56:19 03Graeme Gregory  07org.openembedded.dev * r6687b01b72 10openembedded.git/contrib/angstrom/omap3-mkcard.sh: omap3-mkcard.sh : make fdisk parsing a little more robust Feb 12 22:56:20 03Graeme Gregory  07org.openembedded.dev * r787d023875 10openembedded.git/recipes/angstrom/ (omap3-mkcard.bb omap3-mkcard/omap3-mkcard.sh): omap3-mkcard.bb : add a recipe for mkcard.sh so we can install on targets Feb 12 23:03:03 MACHINE_EXTRA_RRECOMMENDS = "marvell-gspi-fw marvell-sdio-fw kernel-modules" doesn't ship all the modules Feb 12 23:03:27 despite of having this in task-base: RRECOMMENDS_task-machine-base = "${MACHINE_EXTRA_RRECOMMENDS}" Feb 12 23:03:38 but it ships the marvell firmwares.... Feb 12 23:03:43 strange Feb 12 23:11:19 has task-base-gumstix been changed to something else? Feb 12 23:11:29 don't know Feb 12 23:11:33 I have recompiled my kernel, and I want to inject it into my image, but can't find the right command Feb 12 23:11:44 s/it/the modules/ Feb 12 23:11:54 i want the modules in my build image Feb 12 23:11:59 ah ok Feb 12 23:12:18 rphillips, does your MACHINE_EXTRA_RRECOMMENDS have kernel-modules, if so does it work? Feb 12 23:12:24 because for me it doesn't work Feb 12 23:12:28 but it should Feb 12 23:14:19 hmm, it's commented out. Feb 12 23:14:39 what's your machine config? Feb 12 23:14:48 gumstix-custom-verdex.conf Feb 12 23:14:51 ok Feb 12 23:15:46 the comment in the file mentions task-base-gumstix Feb 12 23:16:12 mmm this file is strange Feb 12 23:16:22 Use this file as a starting point for your custom gumstix configuration Feb 12 23:17:05 so just temporarly add that: Feb 12 23:17:15 MACHINE_ESSENTIAL_EXTRA_RRECOMMENDS += "kernel-modules" Feb 12 23:17:22 and look if it works Feb 12 23:18:27 GNUtoo: which package should I tell bitbake to rebuild? Feb 12 23:18:35 the image Feb 12 23:18:38 ok Feb 12 23:19:00 else there is another way: create a feed and opkg install kernel-modules Feb 12 23:19:05 or somthing like that Feb 12 23:19:25 but re-bitbaking the image seem easier for you Feb 12 23:19:30 agreed Feb 12 23:19:50 thanks for the help Feb 12 23:19:53 btw I think I'll go to bed soon, Feb 12 23:20:18 could you do me a favor: report to the mailing list if MACHINE_ESSENTIAL_EXTRA_RRECOMMENDS += "kernel-modules" doesn't work Feb 12 23:20:36 sure. np Feb 12 23:37:29 I'll try something Feb 13 00:02:37 rphillips, are you still there Feb 13 00:02:39 ? Feb 13 00:04:30 rphillips, because I think I've the solution Feb 13 00:04:41 you need to bump PR of that recipe too: Feb 13 00:05:11 task-base.bb Feb 13 00:05:15 at least it seems Feb 13 00:05:21 I'll test the result Feb 13 01:07:38 03Denis 'Gnutoo' Carikli  07org.openembedded.dev * r4facd26dbd 10openembedded.git/conf/machine/eee701.conf: (log message trimmed) Feb 13 01:07:38 eee701 machine config: fix to use xorg instead of kdrive Feb 13 01:07:38 eFfeM pointed me out that he had a Xvesa binary in his xf46-image, Feb 13 01:07:38 for eeepc701 and no Xorg. Feb 13 01:07:38 So I investigated and found that: Feb 13 01:07:38 In eee701.conf there were ?= for XSERVER and Feb 13 01:07:38 PREFERRED_PROVIDER_virtual/xserver Feb 13 01:07:39 03Denis 'Gnutoo' Carikli  07org.openembedded.dev * r0911f0ad43 10openembedded.git/conf/machine/bug.conf: Feb 13 01:07:40 bug.conf : depends on kernel-modules Feb 13 01:07:40 Most users would want all modules,moreover some modules are needed Feb 13 01:07:41 for framebuffer Feb 13 01:07:41 03Denis 'Gnutoo' Carikli  07org.openembedded.dev * r8d9d0507c0 10openembedded.git/recipes/e17/ (entrance/rebased_config-db.patch entrance_0.9.0.010.bb): Feb 13 01:07:42 entrance: rebased config.db Feb 13 01:07:42 The config.db seem to be some openembedded settings/configuration, Feb 13 01:07:43 for launching entrance/e17 better Feb 13 01:07:43 03Denis 'Gnutoo' Carikli  07org.openembedded.dev * r60af72f1af 10openembedded.git/recipes/images/e-image-core.bb: (log message trimmed) Feb 13 01:07:55 03Denis 'Gnutoo' Carikli  07org.openembedded.dev * rcb40fbe4b6 10openembedded.git/conf/machine/bug.conf: Feb 13 01:07:55 bug.conf: added screen resolution Feb 13 01:07:56 some recipes depend on MACHINE_DISPLAY_WIDTH_PIXELS and MACHINE_DISPLAY_HEIGHT_PIXELS Feb 13 01:07:56 like wesnoth for instance Feb 13 01:07:57 03Denis 'Gnutoo' Carikli  07org.openembedded.dev * r0af03b94cd 10openembedded.git/recipes/tasks/task-base.bb: Feb 13 01:07:57 task-base : bump PR because bug.conf changed(details below) Feb 13 01:07:58 kernel-modules was added to MACHINE_EXTRA_RRECOMMENDS in bug.conf, Feb 13 01:07:58 But without bumping the PR of task-base it wouldn't include Feb 13 01:07:59 the kernel modules into the image Feb 13 01:08:00 03Denis 'Gnutoo' Carikli  07org.openembedded.dev * r8a145683d6 10openembedded.git/recipes/e17/entrance_0.9.0.010.bb: Feb 13 01:08:00 entrance: run autopoint in order to cope with lacking ./config.rpath Feb 13 01:08:01 At first I had a bad hack for it,thanks JaMa for showing me how to Feb 13 01:08:01 do it(pointing at one of your commits that did it) Feb 13 01:08:02 03Denis 'Gnutoo' Carikli  07org.openembedded.dev * r5635777f38 10openembedded.git/recipes/images/illume-image.bb: (log message trimmed) Feb 13 01:08:02 illume-image: comment additional phone functionality(zhone already commented) Feb 13 01:08:03 (13 lines omitted) Feb 13 01:28:54 g.m. Feb 13 01:29:49 apt get PACKAGE MANAGER seems to be used only in task-mamona-base.bb . So I guess is good to use opkg on my system, right? Feb 13 02:23:00 How can I receive on my mail openembedded-devel mailing list messages? I'm already subscribed, but I don't know how to enable this feature **** ENDING LOGGING AT Sat Feb 13 02:59:58 2010