**** BEGIN LOGGING AT Sat Nov 28 02:59:56 2009 Nov 28 03:18:17 ~lart sdl developers Nov 28 03:18:17 * ibot resizes sdl developers's terminal to 40x24 Nov 28 03:18:33 if test "x$prefix" != xNONE; then Nov 28 03:18:33 PATH="$prefix/bin:$prefix/usr/bin:$PATH" Nov 28 03:18:34 fi Nov 28 03:18:37 in sdl.m4 Nov 28 03:22:09 args Nov 28 03:22:17 seems modification from gnome-games Nov 28 09:22:09 03Thomas Zimmermann  07org.openembedded.dev * r74ecf8c0f8 10openembedded.git/recipes/openttd/openttd_0.7.4-RC1.bb: Nov 28 09:22:09 openttd: correct path for config file Nov 28 09:22:09 Signed-off-by: Klaus Kurzmann Nov 28 09:22:10 03Thomas Zimmermann  07org.openembedded.dev * r5c0217046a 10openembedded.git/recipes/scummvm/ (3 files in 2 dirs): Nov 28 09:22:13 scummvm: copy config from openmoko to shr Nov 28 09:22:15 Signed-off-by: Klaus Kurzmann Nov 28 09:22:17 03Thomas Zimmermann  07org.openembedded.dev * r232a4e92c6 10openembedded.git/ (2 files in 2 dirs): Nov 28 09:22:20 numberx: new recipe Nov 28 09:22:22 Signed-off-by: Klaus Kurzmann Nov 28 09:22:24 03Thomas Zimmermann  07org.openembedded.dev * r38c5fb678f 10openembedded.git/ (3 files in 3 dirs): Nov 28 09:22:27 scummvm: update version to 1.0.0 Nov 28 09:22:29 Signed-off-by: Klaus Kurzmann Nov 28 09:22:31 03Thomas Zimmermann  07org.openembedded.dev * r4f6a1c0e29 10openembedded.git/recipes/tasks/task-shr-feed.bb: Nov 28 09:22:34 task-shr-feed: add gtkmm, solves SHR bug #546 Nov 28 09:22:36 Signed-off-by: Klaus Kurzmann Nov 28 09:22:38 03Thomas Zimmermann  07org.openembedded.dev * r6cc4cbdabe 10openembedded.git/ (2 files in 2 dirs): Nov 28 09:22:45 neomis: new recipe Nov 28 09:22:47 Signed-off-by: Klaus Kurzmann Nov 28 09:22:49 03Thomas Zimmermann  07org.openembedded.dev * r111ebb91f3 10openembedded.git/ (3 files in 3 dirs): Nov 28 09:22:52 di: new recipe Nov 28 09:22:54 * add di (disk info) Nov 28 09:22:56 * add di to task-shr-feed Nov 28 09:22:58 * this solves SHR bug #501 Nov 28 09:23:00 Signed-off-by: Klaus Kurzmann Nov 28 09:28:46 03Klaus Kurzmann  07org.openembedded.dev * r29fb31b1a5 10openembedded.git/recipes/tasks/task-shr-feed.bb: Nov 28 09:28:46 task-shr-feed: add make Nov 28 09:28:46 Signed-off-by: Klaus Kurzmann Nov 28 11:31:25 hi there Nov 28 11:31:54 has anyone built a minimal image with additional qt4-embedded support? Nov 28 11:32:58 i built an image with a recipe with omap3-console-image from gumstix.net and i'd like to add qt4-embedded to my image Nov 28 11:33:49 is it enough to add DEPENDS = "qt-embedded" and inherit qtopia4core to this recipe? Nov 28 11:34:37 tobi_, I think DEPENDS is for building....I think you need to add qt-embedded to RDEPENDS Nov 28 11:35:26 or to the task Nov 28 11:36:02 or or to IMAGE_INSTALL Nov 28 11:42:27 hm Nov 28 11:44:00 I Nov 28 11:44:08 'll look if it's in the manual Nov 28 11:44:13 this is the recipe i've used Nov 28 11:44:15 hello Nov 28 11:44:18 http://pastebin.com/m4c815278 Nov 28 11:44:25 plz could i ask a bsd related question Nov 28 11:44:40 so you think i can just add one RDEPEND line? Nov 28 11:44:45 right after DEPEND? Nov 28 11:44:50 ppl any recommandable cheap dev board + with lcd for embedded dev that support freebsd/openbsd Nov 28 11:44:51 http://docs.openembedded.org/usermanual/usermanual.html#commonuse_new_image Nov 28 11:45:12 tobi_, so it's PACKAGE_INSTALL or maybe it was changed to IMAGE_INSTALL Nov 28 11:46:06 thx Nov 28 11:46:44 I think to be on the safe side, a RDEPEND is needed as well Nov 28 11:47:01 tobi_, maybe do the following thing: Nov 28 11:47:06 and qt4-embedded is a Nov 28 11:47:09 sorry ;) Nov 28 11:47:16 require the_image_you_have.bb Nov 28 11:47:21 and add that: Nov 28 11:47:43 IMAGE_INSTALL += " something " Nov 28 11:47:50 something beeing the packages you want to add Nov 28 11:48:14 and don't forget to add the DEPENDS in order to make it build the deps Nov 28 11:48:25 and forget about the RDEPEND Nov 28 11:49:17 thx, have you had a look at the recipe i posted? Nov 28 11:49:28 no idea about bsd board Nov 28 11:53:40 gm Nov 28 12:10:36 tobi_, yes it was an immage recipe Nov 28 12:11:03 tobi_, I bet it was an already existing recipe Nov 28 12:11:13 yes it is ;) Nov 28 12:11:22 as i told you... the one i get from gumstix.net Nov 28 12:11:43 however, i think i'll add these couple of lines to this image recipe Nov 28 12:22:29 hm, do you know how to configure a kernel in oe? is it possible to do it if menuconfig? Nov 28 12:23:10 tobi_, yes but I don't remember the exact task to do that Nov 28 12:23:23 tobi_, it was discussed recently in the mailing list Nov 28 12:24:13 tobi_, try that : bitbake virtual/kernel -c menuconfig Nov 28 12:31:00 i've found something similar to this in a mailing list for the beagle board Nov 28 12:31:28 i run this command, however, bitbake wanted to download something but couldnt :S Nov 28 12:32:24 and told me something like: NOTE: Staging package found but invalid for /.../linux-omap3_2.6.31.bb Nov 28 13:00:44 hi, I'm trying to build an opie-image. though it fails to build an rootfs. At the end I get many dependency errors, like this one:"Cannot satisfy the following dependencies for task-opie-base: * opie-freetype *". I got no idea what I am doing wrong. has any one an idea what the problem could be? Nov 28 13:12:36 mmm Nov 28 13:12:37 Boost_REGEX_LIBRARY_RELEASE:FILEPATH=/usr/lib/libboost_regex-mt.so Nov 28 13:12:49 wesnoth = boost+cmake Nov 28 13:12:57 I'll hardcode theses things Nov 28 13:13:16 but everytime I try it takes the /usr/lib and /usr/include things Nov 28 13:36:40 I'll try CMAKE_FIND_ROOT_PATH Nov 28 13:36:49 I hope it's not already defined Nov 28 13:36:57 (like in a site file for instance) Nov 28 13:53:16 CMAKE_FIND_ROOT_PATH:UNINITIALIZED=/home/embedded/oetmp_openmoko/staging/armv4t-oe-linux-gnueabi Nov 28 13:53:26 is it normal to have UNINITIALIZED in this Nov 28 13:53:27 ? Nov 28 14:01:00 RP: ping Nov 28 14:01:03 mornign folks Nov 28 14:01:19 hi mickeyl Nov 28 14:03:20 strange...-DCMAKE_FIND_ROOT_PATH=${STAGING_DIR_HOST} Nov 28 14:03:24 in the cmake class Nov 28 14:03:52 gnutoo hm Nov 28 14:04:05 you could look at my boos_1.4x recipe Nov 28 14:04:14 maybee its helping Nov 28 14:04:17 I use boost 1.4 for wesnoth Nov 28 14:04:27 but boost 1.41.0 is not found Nov 28 14:04:33 hm Nov 28 14:04:35 if it's your recipe I've comments: Nov 28 14:04:39 I wanted updated it anyway Nov 28 14:04:47 I had to upgrade to 1.41 because of mpi Nov 28 14:04:48 the version in oe is rather old Nov 28 14:04:56 it failled on checking for an mpi lib Nov 28 14:05:09 1.41 does not have this problem Nov 28 14:05:20 also I modified locally the inc file Nov 28 14:05:26 hm 1.41 is out? Nov 28 14:05:32 check it in please Nov 28 14:05:33 I think so Nov 28 14:05:36 ok Nov 28 14:05:43 hm you have to use the cmake version Nov 28 14:05:47 not the official Nov 28 14:05:50 do you want to see the changes Nov 28 14:05:55 yes that's what I use Nov 28 14:05:58 for 1.40.0 it was it at least Nov 28 14:06:06 does perl-5.8.8 in OE works for cross compile? Nov 28 14:06:12 because in offical version cmake was brokenm Nov 28 14:06:12 but you seemed to have made the recipe too quickely Nov 28 14:06:16 bits of bjam were left Nov 28 14:06:24 in the python function Nov 28 14:06:40 but in another hand I based my work on 1.40.0 Nov 28 14:06:44 so it was helpfull Nov 28 14:07:14 Openfree, I don't know but I already cross-compiled perl long time ago Nov 28 14:07:19 in oe Nov 28 14:08:42 woglinde, http://pastebin.com/m108f26ff Nov 28 14:08:46 hm release notes says Nov 28 14:08:55 1.41.0 isnt cmake ready Nov 28 14:08:59 and here's the recipe if you want: Nov 28 14:09:02 the cmake version bill come later Nov 28 14:09:36 http://pastebin.com/m7f07eb59 Nov 28 14:09:58 I've got a cmake version that at least configure compiles and install Nov 28 14:10:08 my problem is rather in wesnoth's cmake Nov 28 14:10:25 did you use a newer version? Nov 28 14:10:52 wesnoth's cmake find /usr/lib/something and /usr/include instead of the staging ones Nov 28 14:10:53 yes Nov 28 14:10:55 1.6.5 Nov 28 14:11:12 here's the preview recipe: Nov 28 14:11:31 hm Nov 28 14:11:41 I will not yet find time to look into it Nov 28 14:12:06 http://pastebin.com/m68db8e26 Nov 28 14:12:07 ok Nov 28 14:12:14 I'll try something Nov 28 14:12:31 -DBoost_ADDITIONAL_VERSIONS=1.41 Nov 28 14:12:32 and not Nov 28 14:12:37 -DBoost_ADDITIONAL_VERSIONS=1.41.0 Nov 28 14:12:42 I'll check the dirs Nov 28 14:13:05 indeed doesn't sound right Nov 28 14:13:51 what I want to know is can I modify the boost 1.4x Nov 28 14:13:54 inc file Nov 28 14:14:11 and commit my recipe of 1.41.0 Nov 28 14:14:59 I made too much kernel debugging on htcdream for alsa...I miss oe Nov 28 14:15:14 so at a point I coundn't resist more.... Nov 28 14:15:16 lol Nov 28 14:16:12 and also some people want wesnoth such as koen,shr people and there is even a port of wesnoth to the iphone so we must have wesnoth Nov 28 14:18:26 another question: what should I do if it fails on armv4? Nov 28 14:18:30 at runtime Nov 28 14:18:52 and that I can't test it in other devices without huge modifications(screen size issues) Nov 28 14:24:04 mmm still the same issue Nov 28 14:24:06 but Nov 28 14:24:07 X11_X11_INCLUDE_PATH:PATH=/home/embedded/oetmp_openmoko/staging/armv4t-oe-linux-gnueabi/usr/include Nov 28 14:24:15 so it finds the rest Nov 28 14:36:25 mmm is rm CMakeCache.txt and sh run.do_compile.21907 the good thing to do? Nov 28 14:36:55 because it sees that configure variable aren't taken into account this way Nov 28 14:41:24 wau Nov 28 14:41:31 now I have beagle demo image Nov 28 14:41:42 03Klaus Kurzmann  07org.openembedded.dev * r78c6b57a70 10openembedded.git/conf/distro/include/sane-srcrevs-fso.inc: Nov 28 14:41:42 sane-srcrevs-fso: bump fsodeviced to get the alsa player ringtone fix Nov 28 14:41:42 Signed-off-by: Klaus Kurzmann Nov 28 14:41:50 is beagle board fast? Nov 28 14:41:57 could it replace a computer? Nov 28 14:42:08 s/computer/x86 computer/ Nov 28 14:42:56 depends Nov 28 14:43:02 because for an arm it's fast but I don't know the comparison with an x86 Nov 28 14:43:10 you can watch videos via dsp Nov 28 14:43:15 like a sempron 2500+ Nov 28 14:43:17 yes Nov 28 14:43:18 I know Nov 28 14:43:19 and openoffice should work Nov 28 14:43:21 ogg support too Nov 28 14:43:22 wow Nov 28 14:43:24 when I finished Nov 28 14:43:29 I didn't knew for openoffice Nov 28 14:43:33 the recipe Nov 28 14:43:33 is it in oe? Nov 28 14:43:36 wow!!!!!!! Nov 28 14:43:36 not yet Nov 28 14:43:38 lol Nov 28 14:43:47 how many compilation hours? Nov 28 14:43:53 hm Nov 28 14:44:03 can say yet Nov 28 14:44:07 problem is idl stuff Nov 28 14:44:13 on a peintium M 2GHz + 1.5 GB of ram it's about 5h Nov 28 14:44:35 what's idl? Nov 28 14:44:44 hm some abstraction stuff Nov 28 14:44:48 ok Nov 28 14:44:51 praeprocessor Nov 28 14:45:14 I always wanted openoffice but...the compilation hours and retrying from scratch scared me Nov 28 14:45:31 I even tried it on debian on the openmoko Nov 28 14:45:44 but it was compiled for a higher arm architecture Nov 28 14:45:55 but on the touchbook it would be so great Nov 28 14:50:18 do you have a touchbook? Nov 28 14:50:21 not yet Nov 28 14:50:35 I'm evaluating the fact of buying omap3 based stuff Nov 28 14:50:43 like a beagle board for my desktop Nov 28 14:50:49 and a touchbook for my laptop Nov 28 14:50:55 hm better by the igep 2.0 Nov 28 14:51:00 buy Nov 28 14:51:03 yes but that's harder to find Nov 28 14:51:10 but it has more ram Nov 28 14:51:28 because ram is an issue Nov 28 14:51:46 specially if I want gnome Nov 28 14:52:06 www.myigep.com Nov 28 14:52:15 harder to finf? Nov 28 14:52:23 they ship right now from spain Nov 28 14:52:28 ah ok Nov 28 14:52:30 with faster cpu since october Nov 28 14:52:44 can I trust them? Nov 28 14:52:52 does it works well etc... Nov 28 14:53:18 145 euros Nov 28 14:53:23 + 20 euros shiipig Nov 28 14:53:29 yes does Nov 28 14:53:35 I think you can trust them Nov 28 14:53:38 sounds good Nov 28 14:53:46 and you get wlan/ether/bt on the board Nov 28 14:53:52 beagle has usbnet only Nov 28 14:55:17 if that's the board I saw I knew Nov 28 14:55:38 I'm considering IGEP also Nov 28 14:55:53 since BB isn't available until January 2010... Nov 28 14:57:56 it looks like it's BB+wifi+ethernet+bt Nov 28 14:58:27 +more RAM Nov 28 14:58:29 mmm on the pdf the cpu looks the same bug I know it's 700Mhz not 600....where can I find up to date specs? Nov 28 15:03:03 ah I found it in the shop Nov 28 15:04:34 ah wow no customs taxes.... Nov 28 15:04:47 so it would be a lot cheaper Nov 28 15:06:42 you're from EU? Nov 28 15:08:38 ppl any dev board (arm based) compatible with openbsd Nov 28 15:08:44 cheap one Nov 28 15:08:44 :D Nov 28 15:11:07 yes Nov 28 15:11:34 mmm I think my problems come from UNINITIALIZED Nov 28 15:15:24 I'll try to initialize but I bet I found a problem in cmake.bbclass.... Nov 28 15:15:26 I'll check Nov 28 15:16:21 ynezz, for instance if I pay shipping for the bug device I've to pay customs....and that's not cheap....last time it was ~50E Nov 28 15:17:35 hey, has anyone run qt apps on a gumstix with a 4.3" touchscreen? Nov 28 15:17:51 i managed to create an image with qt4-embedded ;) Nov 28 15:18:46 args nfs over usbnet suckz Nov 28 15:19:05 however, although i calibrated the touchscreen with ts_calibrate, the calibration for the qt apps seems not to work Nov 28 15:19:22 tobi hm Nov 28 15:19:35 have you the tslib modul in? Nov 28 15:19:37 in qt Nov 28 15:19:43 installed on the gumstick? Nov 28 15:21:49 hm, i just added qt4-embedded in my recipe Nov 28 15:21:55 how can i check this? Nov 28 15:31:54 tobi_, there was a thread about qt + tslib on the mailing list Nov 28 15:32:02 maybe it has the variable Nov 28 15:32:16 like QWS_PROTO_MOUSE or something like this Nov 28 15:32:36 i'm having a look at: http://www.qtforum.org/article/28281/touchscreen-calibration-problem.html Nov 28 15:34:31 hm, now the mouse isn't moving at all :s Nov 28 15:35:31 ts_print or ts_print_raw works? Nov 28 15:40:41 hm, what should happen if i run ts_print? Nov 28 15:40:48 i can calibrate with ts_calibrate Nov 28 15:41:01 and the mouse moves when i run a qt app Nov 28 15:41:12 however, the limits are wrong and it behaves stupid :D Nov 28 15:41:31 03Michael 'Mickey' Lauer  07org.openembedded.dev * r052e3287b4 10openembedded.git/ (2 files in 2 dirs): fsodeviced: explicitly enable the rfkill and canberra player plugins Nov 28 15:41:55 if i set this: export QWS_MOUSE_PROTO=Tslib:/dev/input/touchscreen0, the mouse doesn't work at all Nov 28 15:42:12 it's typo? Nov 28 15:42:14 Tslib: Nov 28 15:42:14 any idea why after manual bitbake -b you can't compile anything anymore? Nov 28 15:42:28 shouldn't it be tslib: ? Nov 28 15:42:35 i tried both :) Nov 28 15:42:40 everything fails with configure: error: C preprocessor "arm-oe-linux-gnueabi-gcc -E" fails sanity check Nov 28 15:42:41 after that Nov 28 15:44:37 it's "Tslib" according to http://doc.trolltech.com/4.1/qtopiacore-envvars.html Nov 28 15:44:51 i'm afraid ts is not set up propperly :( Nov 28 15:45:23 hm, ts_test works :S Nov 28 15:46:31 03Michael 'Mickey' Lauer  07org.openembedded.dev * rb160b53076 10openembedded.git/ (4 files in 2 dirs): Nov 28 15:46:31 fsodatad: new recipe; FSO data storage service Nov 28 15:46:31 fsomusicd: new recipe; FSO music player service Nov 28 15:47:40 g.m. Nov 28 15:49:43 according to http://pastebin.com/m60d92d2e, tslib should be built in qt4 :s Nov 28 15:50:35 do i miss to run an important service? Nov 28 15:53:07 mmm how do I find out why cmake didn't find boost? Nov 28 15:53:14 is there a trace or debug mode? Nov 28 15:54:44 tobi_: remember that qt4.6.0rc1 is supporting linuxinput Nov 28 15:55:11 yeah, but is there a bitbake recipe for qt4-embedded 4.6? Nov 28 15:56:19 hm, i'm running one of the demos in the /usr/bin/qtopia, can it be, that the demos where compiled without tslib support? :s Nov 28 15:57:33 is qte dlopening or linking against tslib? Nov 28 15:57:52 if linking, you might check the binary Nov 28 15:58:30 how can i check this? Nov 28 15:58:43 does this give any information? http://pastebin.com/m60d92d2e Nov 28 15:59:14 ldd Nov 28 15:59:44 -plugin-mouse-tslib indicates that it builds a tslib plugin Nov 28 15:59:46 unfortunately ldd is not on the gumstix Nov 28 16:00:06 perhaps you can build it seperately? Nov 28 16:00:13 or install Nov 28 16:00:58 you can also check w/ strace whether the process tries to open one of the input device nodes Nov 28 16:03:13 hm, i think i'll try to replace "-plugin-mouse-tslib -qt-mouse-pc -qt-mouse-qvfb" with "-qt-mouse-tslib" Nov 28 16:04:11 worth a try Nov 28 16:04:18 what does ./configure --help say? Nov 28 16:04:31 if i change org.openembedded.dev/recipes/qt4/qt4-embedded.inc how can i tell bitbake to rebuild qt4-embedded? will it recognise itself? Nov 28 16:04:54 try bbnew Nov 28 16:05:02 like: Nov 28 16:05:24 bbnew () Nov 28 16:05:24 { Nov 28 16:05:24 nice -n 20 bitbake -c clean -b $1 && nice -n 20 bitbake $BBD -b $1 Nov 28 16:05:24 } Nov 28 16:05:28 mutatis mutandis :) Nov 28 16:05:52 iow: clean and build again Nov 28 16:06:21 bitbake would only attempt to rebuild it if the PR was increased Nov 28 16:06:32 hm sorry, i can't follow :) Nov 28 16:07:08 ok, nm, clean and rebuild Nov 28 16:09:57 mickeyl: good morning Nov 28 16:10:19 nm clean and rebuild? :) Nov 28 16:10:29 sorry... i'm completely new to oe :D :P Nov 28 16:10:36 morning pb___ Nov 28 16:10:40 do you mean a bitbake -c clean qt4-embedded? Nov 28 16:10:45 yes Nov 28 16:11:08 okay... and a bitbake -c nm is neeeded as well? :s Nov 28 16:11:32 no Nov 28 16:11:37 that was short for "nevermind" Nov 28 16:13:39 eV members, vote early, vote often, vote NOW. Nov 28 16:14:05 * Crofton|work apologizes for the spam, but XorA|gone is away for most of today and the vote ends "soon" Nov 28 16:14:28 hehe ;) Nov 28 16:15:27 thanks! :D Nov 28 16:16:46 Crofton|work: trying to build the gnome image on 64 bit Ubuntu has proven to be a very frustrating experience :-( Nov 28 16:16:58 hmm Nov 28 16:17:11 I haven't had any particular probs on my 64 bit machine Nov 28 16:17:16 python related issues Nov 28 16:17:36 * mwester goes to look for the voting instructions email Nov 28 16:20:16 I have a recipe for qt4.6.0 rc1 . tomorrow I can post it. Nov 28 16:20:47 compilation works fine, only install and ipk is not ok Nov 28 16:21:28 Crofton|work: root issue seems to be: "LONG_BIT definition appears wrong for platform (bad gcc/glibc config?)" Nov 28 16:21:54 but it diguises itself on some packages as a messing header error Nov 28 16:21:59 missing Nov 28 16:23:36 looking in the archives I see that it has been discussed before and a few packages were fixed Nov 28 16:23:44 but sadly not all Nov 28 16:33:08 monday I can send a patch . byebye Nov 28 16:49:21 May you help me, please? I have a doubt about qt libraries, I need to install the qt4.5 libraries in my OE distro, but I don't want to install x11 libraries or xorg. I've just need the qt4.5 libraries to use with my apps. How is the easiest way to install qt4.5 libraries? Nov 28 16:51:36 mickeyl,hi you told that you stopped hoping about the htc series,but htcdream got nearly everything but pm and alsa,for alsa I'm stuch at kexec: I need a dump(I've no serial cable)....what do you think....should I abandon and write an userspace pulseaudio plugin instead that use the g1's custom audio? Nov 28 16:52:01 s/stuch/stuck Nov 28 16:52:31 in my opinion pulseaudio sucks big time on embedded systems except those who put lots of time into it (like Palm) Nov 28 16:52:41 but any audio is better than no audio Nov 28 16:52:46 so why not trying to do that pulseaudio plugin Nov 28 16:52:57 ok Nov 28 16:53:01 thanks a lot Nov 28 16:53:03 np Nov 28 16:53:24 btw I bet no one know why my phone rebooted instead of kexecing Nov 28 16:53:36 that is to say it rebooted into android Nov 28 16:53:36 sounds like a bootloader issue Nov 28 16:53:39 ah ok Nov 28 16:53:44 proprietary bootloader Nov 28 16:53:51 of course :) Nov 28 16:53:53 I can flash a new but I'm afraid Nov 28 16:54:05 many variables Nov 28 16:54:05 don't you have the engineering bootloader on it? Nov 28 16:54:08 no Nov 28 16:54:10 yeah, many variables Nov 28 16:54:44 because italian phone,enginering bootloader may not have been tested on this phone Nov 28 16:54:55 italian phone means carrier phone Nov 28 16:55:19 btw another thing: does someone knows if fbdev for xorg can do xrandr -o 1 Nov 28 17:07:07 wesnoth building....it's hackish as the dir of boost include is specified.... Nov 28 17:07:11 it'll fail at linking Nov 28 17:07:38 I'll specify better the lib directory Nov 28 17:10:30 03Michael 'Mickey' Lauer  07org.openembedded.dev * r13920da39c 10openembedded.git/ (3 files in 2 dirs): fsolocationd: renamed recipe; source was fsogpsd Nov 28 17:11:19 what would be the correct way to add a distro specific sshd_config? Nov 28 17:11:30 as override in the openssh recipe? Nov 28 17:11:35 or in rootfs? Nov 28 17:19:42 morning Nov 28 17:25:10 mrmoku|away, mmm distro means it have to be valid for all images no? Nov 28 17:39:09 gnutoo: what do you mean? Nov 28 17:41:01 mrmoku, if you choose the distro way (in .../org.openembedded.dev/recipes/openssh/files/shr ) it will be valid for all images if you select that distro Nov 28 17:41:29 else if you choose the image way it will be valid for all distro but only for that image and the images that require that image Nov 28 17:42:06 gnutoo: yeah, and will even work on updates Nov 28 17:42:29 who selects dropbear? I bet it's the image no? Nov 28 17:42:43 so I will just add a shr folder to files :) Nov 28 17:42:45 gnutoo: thanks Nov 28 17:43:53 I bet the correct way is select openssh in the image + the config in .../org.openembedded.dev/recipes/openssh/files/shr Nov 28 17:44:19 but I'm not shure where dropbear is defined Nov 28 17:45:48 gnutoo: IIRC that is specified via a provider Nov 28 17:45:51 let me check Nov 28 17:46:08 ok Nov 28 17:46:30 btw how the updates will be handled? Nov 28 17:47:05 gnutoo: you mean from dropbear to openssh? Nov 28 17:47:10 yes Nov 28 17:47:20 if you opkg update && opkg upgrade Nov 28 17:47:23 I'm inclined to say we don't care :P Nov 28 17:47:26 do you get openssh automagically Nov 28 17:47:28 (for shr-unstable) Nov 28 17:47:29 ok Nov 28 17:47:31 hehe Nov 28 17:47:33 so reinstall Nov 28 17:47:41 or manually remove + install Nov 28 17:47:46 will post something on our ml to tell people how to do that Nov 28 17:47:54 ok fine Nov 28 17:48:28 but not everybody read the ml: I know an user who opkg update+opkg upgrade and that had to reflash because nothing worked Nov 28 17:48:39 gnutoo: DISTRO_SSH_DAEMON is the one Nov 28 17:48:41 he opkg update && opkg upgrade from an old shr to an oe one Nov 28 17:48:43 ok Nov 28 17:49:17 alternative would be to do some magic postinst scripting Nov 28 17:49:24 ah great idea Nov 28 17:49:43 update-alternatives? Nov 28 17:51:06 don't think it works via alternative Nov 28 17:51:07 s Nov 28 17:51:28 openssh has an expicit RCONFLICTS against dropbear Nov 28 17:51:34 so you can't install both Nov 28 17:54:29 switching from dropbear to openssh or vice-versa is nearly impossible with a post-inst script, because the process may be run from an SSH session. :( Nov 28 17:56:26 mwester: good point :) Nov 28 17:56:40 and a strong reason to do it while we're still unstable Nov 28 17:56:58 (without magic and manual instructions) Nov 28 17:57:52 When we switched SlugOS from dropbear to openssh (which was a very wise change!), we required a reflash. :) Nov 28 17:59:06 guess we will do the same... well maybe offer some help how to manually fix it Nov 28 18:01:53 03Martin Jansa  07org.openembedded.dev * re1d692c126 10openembedded.git/recipes/openmoko-3rdparty/babiloo-efl_bzr.bb: babilo-efl: switch SRCREV -> SRCPV, formal change as bzr returns sortable revision number Nov 28 18:02:50 mrmoku, ok Nov 28 18:38:08 03Michael 'Mickey' Lauer  07org.openembedded.dev * r37303bcdab 10openembedded.git/recipes/freesmartphone/fsomusicd_git.bb: fsomusicd: add SRCREV Nov 28 18:38:22 03Michael 'Mickey' Lauer  07org.openembedded.dev * r808a7c7048 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Nov 28 18:45:08 I don't think there's any particular reason (apart from opkg bugs and/or broken maintainer scripts) that replacing openssh with dropbear on a live system oughtn't to work just fine. Nov 28 18:45:58 Going the other way, from dropbear to openssh, might be a little bit more fraught: you might end up needing to reboot in order to get sshd restarted, but in principle it ought to work mostly ok in that direction too. Nov 28 18:46:33 ok thanks Nov 28 18:59:13 03Martin Jansa  07org.openembedded.dev * ra5d21cde98 10openembedded.git/conf/distro/include/ (3 files): Nov 28 18:59:13 shr: bump linux-libc-headers to 2.6.31, use EFL_SRCREV from sane-srcrevs Nov 28 18:59:13 Signed-off-by: Martin Jansa Nov 28 19:06:23 03Martin Jansa  07org.openembedded.dev * r2ba6e73864 10openembedded.git/recipes/linux/ (2 files in 2 dirs): Nov 28 19:06:23 linux-openmoko-2.6.31: enable NFS and INOTIFY_USER for gta02 Nov 28 19:06:23 Signed-off-by: Martin Jansa Nov 28 19:16:29 I hope it runs now : NOTE: Running task 1085 of 1522 (ID: 7, /home/embedded/oe/org.openembedded.dev/recipes/wesnoth/wesnoth_1.6.5.bb, do_install) Nov 28 19:30:31 I get a problem with fetch u-boot from de git repository, anyone may help me to solve it, please. I paste the log error in the follow link: http://pastebin.com/m63a3cf7c . Thanks Nov 28 19:37:46 03Klaus Kurzmann  07org.openembedded.dev * r6e1292fe4d 10openembedded.git/recipes/python/python-phoneutils_git.bb: Nov 28 19:37:46 python-phoneutils_git: fix DEPENDS Nov 28 19:37:46 Signed-off-by: Klaus Kurzmann Nov 28 19:48:43 03Klaus Kurzmann  07org.openembedded.dev * r0a4733c115 10openembedded.git/recipes/e17/exhibit_svn.bb: Nov 28 19:48:43 exhibit: fix SRC_URI - exhibit moved to OLD Nov 28 19:48:43 Signed-off-by: Klaus Kurzmann Nov 28 20:07:42 mmm Nov 28 20:07:47 I've a lot of ERROR: Error, lockfile path does not exist!: /home/embedded/oetmp_openmoko/work/armv4t-oe-linux-gnueabi/util-macros-1_1.3.0-r0/packages-split Nov 28 20:07:54 what should I do? Nov 28 20:08:04 rebuild clean them + rebuild Nov 28 20:09:24 btw how do SHR guys built boost 1.40.0 Nov 28 20:09:32 did you have MPI on your system? Nov 28 20:26:39 I got a ERROR with u-boot recipe, when bitbake was trying to fetch the file from a git repository. I look in the recipe and I haven't founded do_fetch() function. Is it correct? what shoud I do to solve it? Nov 28 20:29:24 * ynezz loves git reflog, saved him few hours of work... Nov 28 20:36:09 May you help me with git, please? I can't to build the u-boot recipe, and get a fetch error. For helper, the log error is here: http://pastebin.com/m63a3cf7c Nov 28 20:37:01 roliveira, what machine .... Nov 28 20:37:59 Crofton|work: i486 machines, thanks Nov 28 20:38:23 Does u-boot support intel machine? Nov 28 20:42:45 Crofton|work: I don't know, i used the default configuration and the minimal distro. What is better for intel machine? Nov 28 20:45:28 there are some bootloader such as: Nov 28 20:45:37 *bios+{grub,lilo etc...} Nov 28 20:45:40 *coreboot Nov 28 20:45:52 *special free embedded bios for alix boards Nov 28 20:45:54 *other Nov 28 20:47:31 gnutoo: Crofton|work: Thanks all, I looked for u-boor targed architecture and I saw that uboot doesn't support intel machine Nov 28 20:48:53 roliveira, what do you want exactly? to replace the BIOS or to have a bootloader on top of that? Nov 28 20:50:54 gnutoo: I want to have a bootloader, I removed the flag uboot in the MACHINE_FEATURES Nov 28 20:52:01 roliveira, yes but what kind of bootloader? one that sits on top of the BIOS? or one that replace it? Nov 28 20:52:08 what's your machine? Nov 28 20:53:13 the machine is AMD Geode NSV Nov 28 20:57:26 mmm I don't find it on the net Nov 28 20:57:29 but... Nov 28 20:58:31 gnutoo: but now I get a big one ERROR with ipk packet, the last one, could you help me, i paste the error in the follow link because is bigger: http://pastebin.com/m19aeb646 Nov 28 20:59:20 but if you have a BIOS and don't care about the fact that SMI/SMM has a lot of power and that the bios is proprietary(I looked for a coreboot mainboard but didn't find one in my area)....you could use the bios Nov 28 21:00:18 mmm Nov 28 21:01:32 it says that "${@ " ".join( [ ( "/etc/opkg/%s-feed.conf" % feed ) for feed in "all ${PACKAGE_EXTRA_ARCHS} rumba".split() ] ) }" is failing in /home/ricardo/OEBASE/openembedded/recipes/meta/distro-feed-configs.bb Nov 28 21:02:37 you could print all the variables? Nov 28 21:02:44 of "${@ " ".join( [ ( "/etc/opkg/%s-feed.conf" % feed ) for feed in "all ${PACKAGE_EXTRA_ARCHS} rumba".split() ] ) }" Nov 28 21:04:22 gnutoo: i'm doing that, just a minuto, thanks Nov 28 21:04:54 EOF = end of file btw Nov 28 21:05:44 re Nov 28 21:07:14 gnutoo: How do I show all variables? I tried bitbake -v -D, but showed the same thing Nov 28 21:07:31 print Nov 28 21:07:36 in the python function Nov 28 21:10:30 gnutoo: I paste here the file content DESCRIPTION = "Configuration files for online package repositories aka feeds" Nov 28 21:10:30 PR = "r1" Nov 28 21:10:30 DISTRO_FEED_PREFIX ?= "remote" Nov 28 21:10:30 DISTRO_FEED_URI ?= "http://my-distribution.example/remote-feed/" Nov 28 21:10:30 do_compile() { Nov 28 21:10:34 mkdir -p ${S}/${sysconfdir}/opkg Nov 28 21:10:36 for feed in all ${PACKAGE_EXTRA_ARCHS} ${MACHINE_ARCH}; do Nov 28 21:10:38 echo "src/gz ${DISTRO_FEED_PREFIX}-${feed} ${DISTRO_FEED_URI}/${feed}" > ${S}/${sysconfdir}/opkg/${feed}-feed.conf Nov 28 21:10:41 done Nov 28 21:10:43 } Nov 28 21:10:45 do_install () { Nov 28 21:10:47 install -d ${D}${sysconfdir}/opkg Nov 28 21:10:49 install -m 0644 ${S}/${sysconfdir}/opkg/* ${D}${sysconfdir}/opkg/ Nov 28 21:10:51 } Nov 28 21:10:53 PACKAGE_ARCH = "${MACHINE_ARCH}" Nov 28 21:10:55 #def distro_feed_configs(d): Nov 28 21:10:57 # import bb Nov 28 21:10:59 # parchs = bb.data.getVar( "PACKAGE_EXTRA_ARCHS", d, 1 ).split() Nov 28 21:11:01 # march = bb.data.getVar( "MACHINE_ARCH", d, 1 ).split() Nov 28 21:11:05 # archs = [ "all" ] + parchs + march Nov 28 21:11:07 # confs = [ ( "${sysconfdir}/opkg/%s-feed.conf" % feed ) for feed in archs ] Nov 28 21:11:09 # return " ".join( confs ) Nov 28 21:11:11 # Nov 28 21:11:13 #CONFFILES_${PN} += '${@distro_feed_configs(d)}' Nov 28 21:11:15 CONFFILES_${PN} += '${@ " ".join( [ ( "${sysconfdir}/opkg/%s-feed.conf" % feed ) for feed in "all ${PACKAGE_EXTRA_ARCHS} ${MACHINE_ARCH}".split() ] ) }' Nov 28 21:11:22 roliveira, do you know what a pastebin is? Nov 28 21:12:18 gnutoo, I don't think so, but he is about to find out Nov 28 21:12:32 sorry guys, i paste in the paste bin, but the command didn't copy the link http://pastebin.com/m2508166e Nov 28 21:12:54 gnutoo: sorry Nov 28 21:13:44 ~pastebin Nov 28 21:13:45 [~pastebin] A "pastebin" is a web-based service where you can paste anything over 3 lines without flooding 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://www.rafb.net/paste Nov 28 21:13:49 roliveira, first do you need feeds? Nov 28 21:14:05 woglinde, wesnoth work to some extend Nov 28 21:14:25 gnutoo cool Nov 28 21:14:35 basically : Nov 28 21:14:50 *assertions in 480x640 in the tutorial Nov 28 21:15:29 *multiplayer on the device nearly worked...but a button moved itself when you click on it => imporssible/difficult to choose a computer opponent Nov 28 21:15:44 but it doesn't segfault anymore Nov 28 21:16:03 in #wesnoth-dev they told me that: Nov 28 21:16:08 gnutoo hehe Nov 28 21:16:18 " gnutoo: depending on which cmake version you got you might also just want to remove cmake/FindBoost.cmake" Nov 28 21:16:29 " since in version >=cmake-2.6 the working findscript is already shipped" Nov 28 21:16:34 gnutoo: thanks, i will look for that Nov 28 21:16:41 that would prevent hardcoded things Nov 28 21:16:57 roliveira, because you commented things Nov 28 21:17:25 roliveira, and I bet the last line...needed the previous things that you commented Nov 28 21:18:05 roliveira, and I bet you don't own that address either: http://my-distribution.example/remote-feed/ Nov 28 21:19:12 gnutoo: I've just tried the minimal OE distro, in the tutorial, to learned about, I'll see and study about, thanks a lot and sorry with the paste code, it was a big mistake Nov 28 21:20:29 roliveira, basically what's your goal? do you need feeds in order to update the packages or you only need an image? Nov 28 21:21:36 gnutoo: I'll need the feed in the future Nov 28 21:21:45 ok Nov 28 21:22:00 so you'll have to look into it and make your own image and feed recipe Nov 28 21:22:37 gnutoo: this is a very good thing to update automatically, i'll look and do that Nov 28 21:22:47 gnutoo: thank Nov 28 21:29:05 hi, i'm trying to update the php recipe but for some reason I get .o files and .lo files for libtool, but no .so files, anyone an idea on the direction to look in Nov 28 21:30:17 hm you should get .la files Nov 28 21:30:24 and from .la files .so Nov 28 21:30:30 gm Nov 28 21:30:45 hi likewise Nov 28 21:30:45 woglinde: not in the dir where the .o files are Nov 28 21:30:50 likewise: hi Nov 28 21:30:59 .lib Nov 28 21:31:34 I'm trying to download http://www.oesf.org/forum/index.php?act=attach&type=post&id=2510 from oesf forum but I get an error Nov 28 21:31:47 "Sorry, but you do not have permission to use this feature. If you are not logged in, you may do so using the form below if available." Nov 28 21:31:49 woglinde: not in lib dir or so either, ran a find on the dir Nov 28 21:32:32 I cant seem to download or add posts or contact an admin on oesf Nov 28 21:33:23 pumpapa: not sure what you want to d/l there Nov 28 21:33:50 i'm using oe for quite a while and never had the desire to get tomething from it, but it looks like you need to have an account Nov 28 21:34:01 a file wifi_net.rar.txt which is supposed to fix my wifi problems Nov 28 21:34:22 I do have an account, and I am loggen in Nov 28 21:35:31 pumpapa: no idea ask the admin Nov 28 21:35:42 #oe is not really an oesf support channel Nov 28 21:36:16 when I try to contact admin I get the same error. Nov 28 21:36:39 ah special forces again Nov 28 21:36:41 *g* Nov 28 21:36:54 who is oesf? Nov 28 21:37:16 some forum Nov 28 21:37:21 http://www.oesf.org/forum Nov 28 21:37:33 with people who thinks is better to edit binaries Nov 28 21:41:51 edit binaries...lol like open an hex editor instead of recompiling? Nov 28 21:44:31 gnutoo yes Nov 28 21:45:43 lol Nov 28 21:46:50 yuk, hex editors Nov 28 21:48:43 so first what should I do for boost? Nov 28 21:48:54 gnutoo hm? Nov 28 21:48:56 some people like SHR were using old 1.40.0 boost Nov 28 21:49:03 and I changed the inc file Nov 28 21:49:05 old? Nov 28 21:49:19 I updated it last month Nov 28 21:49:27 before 1.36.0 was default Nov 28 21:49:32 I know Nov 28 21:49:52 by old I mean that it didn't work for me and that the one I made work for me Nov 28 21:50:05 I bet the issue is if you have mpi on your host or not Nov 28 21:50:24 so should we require mpi or switch to a newer boost without this issue Nov 28 21:50:26 ? Nov 28 21:51:28 whats mpi? Nov 28 21:51:52 something for cluster Nov 28 21:54:14 hms what was the command to generate Packages.gz Nov 28 21:54:24 gzip? Nov 28 21:54:36 bitbake package-index ? Nov 28 21:55:47 how strange....today I have forgetten that for a moment and had to look at the manual....but I use that command everytime....lol Nov 28 21:56:12 hm cool didnt know of it Nov 28 21:56:24 ??? Nov 28 21:56:48 maybe there is a missunderstanding Nov 28 21:57:27 because I can't believe you don't know package-index Nov 28 21:59:14 only one minute left to vote in the e.V member election! Nov 28 21:59:51 I already did Nov 28 22:00:22 gnutoo I dont use opkg often Nov 28 22:00:27 ah ok Nov 28 22:00:32 or using a fee Nov 28 22:00:33 d Nov 28 22:00:36 busybox's dpkg? Nov 28 22:00:36 or providing Nov 28 22:00:40 ok Nov 28 22:00:47 but now its better then copy the ikg over Nov 28 22:00:49 or apt or tar or ? Nov 28 22:00:56 I use feed to get deps Nov 28 22:01:11 hm packages is still empty Nov 28 22:01:14 whats wrong? Nov 28 22:01:32 ah okay wrong one Nov 28 22:03:45 woglinde: very good Nov 28 22:05:03 hi everybody! for some reason I'm having troubles with libdvb again... Nov 28 22:05:13 ERROR: QA Issue with libdvb: No GNU_HASH in the elf binary: '/usr/tmp2/oetmp/work/armv7a-angstrom-linux-gnueabi/libdvb-0.5.5.1-r3/packages-split/libdvb/usr/bin/dvbaudio' Nov 28 22:05:13 etc Nov 28 22:05:19 I read that that's usually because it's not using the specified LDFLAGS? Nov 28 22:05:30 but I don't see why it wouldn't be using them Nov 28 22:07:29 signal11 right Nov 28 22:07:38 sorry I didnt check the update yet in Nov 28 22:07:54 still have a bunch of patches to test Nov 28 22:08:20 well, I made a package earlier, but then decided to rebuild things Nov 28 22:08:22 and now I have this error, even after fixing the other thing Nov 28 22:08:39 it compiles, but has the QA issue in do_populate_staging and do_qa_staging Nov 28 22:09:28 signal11 right Nov 28 22:09:34 we need to pass the righr linker options Nov 28 22:10:19 lol I thought you had something like a sigkill but not a sigkill Nov 28 22:20:54 hms Nov 28 22:21:04 mplayer uses its own libavoced in oe Nov 28 22:21:06 thats not good Nov 28 22:22:27 I see stuff with --hash-style=gnu in the compile log though Nov 28 22:23:16 http://static.quaddro.net/tmp/log.do_compile.21515 Nov 28 22:34:30 signal hm its needed at linking Nov 28 22:34:38 maybee not all binarys get linked right Nov 28 22:42:21 so odd... seems like a horrible source package, but I need it! Nov 28 22:42:35 hms Nov 28 22:42:37 ERROR: from element /GstPipeline:pipeline0/GstOggDemux:oggdemux0: Internal data stream error. Nov 28 22:42:48 some saw this? Nov 28 23:24:09 hm can't build qpf-freemono package Nov 28 23:24:17 no source mirror available Nov 28 23:24:24 is it obsolete? Nov 28 23:29:25 woglinde, you're trying that: http://www.schleef.org/blog/2009/11/11/theora-on-ti-c64x-dsp-and-omap3/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+BeagleBoard+%28BeagleBoard.org%29 ? Nov 28 23:29:58 signal11, at what point are you...you could inspect the makefiles Nov 28 23:32:47 gnutoo not yet Nov 28 23:32:58 ok Nov 28 23:33:04 I am trying if the newer libvorbis and theora are working Nov 28 23:33:11 ok Nov 28 23:33:20 I updated and cleanded the recipes Nov 28 23:33:42 no need for tremor there...you have a floating point unit(I don't know which one but you have one) Nov 28 23:34:01 hm mplayer is working Nov 28 23:34:05 maybe I should import patch from the iphone wesnoth version Nov 28 23:34:06 and using tremor Nov 28 23:34:18 do it Nov 28 23:34:29 ok Nov 28 23:34:42 but I'm not shure....the screen is 480x640 Nov 28 23:35:00 the iphone is 480x320 Nov 28 23:35:02 it should do Nov 28 23:35:25 jepp Nov 28 23:37:20 jo kergoth Nov 28 23:37:48 hey Nov 28 23:40:02 I don't find the build system.... Nov 28 23:43:46 ???? Nov 28 23:54:17 good nite Nov 29 00:04:16 ouch...| -- The C compiler identification is unknown + | -- The CXX compiler identification is unknown Nov 29 00:43:16 mmm Backtrace stopped: previous frame identical to this frame (corrupt stack?) **** ENDING LOGGING AT Sun Nov 29 02:59:57 2009