**** BEGIN LOGGING AT Tue May 04 03:00:01 2010 May 04 03:59:25 so can anybody help with the error from the bitbake run? May 04 06:01:43 hi all, i built own_customized_image with qte support(libraries) int he image, but when i am running a sample program ./sample -qws such an error occurs May 04 06:02:06 QFontDatabase: Cannot find font directory /usr/lib/fonts May 04 06:02:55 I googled the error and found out that this thing happens when the prefix option while "configure" is set to the build machine environment, kindly tell me how should i resolve this?? May 04 06:31:21 03Koen Kooi  07org.openembedded.dev * r4b90bc3e1b 10openembedded.git/recipes/lxde/lxpanel_0.5.5.bb: lxpanel: add 0.5.5 May 04 06:31:22 03Koen Kooi  07org.openembedded.dev * r4e8115848c 10openembedded.git/recipes/lxde/lxsession_0.4.4.bb: lxsession: add 0.4.4 May 04 06:31:22 03Koen Kooi  07org.openembedded.dev * r975ba8dfb6 10openembedded.git/recipes/lxde/lxde-common_0.5.0.bb: lxde-common: add 0.5.0 May 04 06:31:23 03Koen Kooi  07org.openembedded.dev * ra1039d23be 10openembedded.git/recipes/lxde/lxmenu-data_0.1.1.bb: lxmenu-data: add 0.1.1 May 04 06:31:24 03Koen Kooi  07org.openembedded.dev * r87aadf260b 10openembedded.git/recipes/lxde/menu-cache_0.3.2.bb: menu-cache: add 0.3.2 May 04 06:31:25 03Koen Kooi  07org.openembedded.dev * r7d2d7e0e9f 10openembedded.git/recipes/ti/ (matrix-gui/init matrix-gui_svn.bb): matrix-gui: update to latest svn May 04 06:31:27 03Koen Kooi  07org.openembedded.dev * rb8bc13a12d 10openembedded.git/recipes/linux/ (50 files in 3 dirs): May 04 06:31:27 linux-omap-psp 26.32: sync with beagleboard validation kernel again May 04 06:31:27 This adds back support for the ds1307 RTC and adds support for the AT24 eeprom May 04 06:35:26 anybody knows how to fix the error in oe_runconf? May 04 06:58:13 03Stefan Schmidt  07org.openembedded.dev * r88eae52b1c 10openembedded.git/recipes/ti/matrix-gui_svn.bb: (log message trimmed) May 04 06:58:13 matrix-gui_svn.bb: Fix parsing of the recipe May 04 06:58:13 NOTE: Handling BitBake files: \ (6840/8344) [81 %]ERROR: May 04 06:58:13 /home/stefan/Projekte/OpenEmbedded/build/openmoko/openembedded/recipes/ti/matrix-gui_svn.bb:22: May 04 06:58:13 unparsed line: 'do_install(52) {' while parsing May 04 06:58:14 /home/stefan/Projekte/OpenEmbedded/build/openmoko/openembedded/recipes/ti/matrix-gui_svn.bb May 04 06:58:15 ERROR: May 04 07:11:18 03Koen Kooi  07org.openembedded.dev * r402a721af6 10openembedded.git/ (4 files in 3 dirs): linux-omap-psp 2.6.32: refresh touchbook defconfig, add disabled patch to enable 800MHz operation May 04 08:10:25 morning May 04 08:31:46 hey mr florian May 04 08:33:35 mornin XorA May 04 08:34:02 ant_work: get back to kexec hacking, you have no time to talk :-D May 04 08:35:36 ;) May 04 08:35:56 XorA: would you mind to explain how that udev-124 hack would be packaged? May 04 08:37:05 good moring May 04 08:37:08 +n May 04 08:38:01 ant_work: I beleive only the udev executable should be packaged May 04 08:38:12 ant_work: Im pretty sure koen tested it as he showed me a bootlog May 04 08:38:30 I will edit manually the rootfs to test the hack May 04 08:38:40 then we'll see about packaging :) May 04 08:42:15 good morning May 04 08:43:11 XorA: I just discovered a fantastic italo-scottish singer :-) May 04 08:45:02 mckoan: heh May 04 08:45:09 mckoan: sounds scary May 04 08:56:34 hi, I'm building some modules using recipes which "inherit module" and it appears that do_strip_modules doesn't works in that case May 04 08:57:15 from run.do_strip_modules.xxx, it seems there is a cd ${S} before doing do_strip_modules May 04 08:59:31 then do_strip_modules tests for ${PKGD}/lib/modules and looks for .ko and strip them May 04 09:00:17 but in then end, modules are not stripped May 04 09:00:48 do you have an idea of what could be wrong there ? May 04 09:03:02 I thought this issue was already discussed/fixed May 04 09:03:37 do you have a link to the discussion ? May 04 09:04:22 http://bugs.openembedded.net/show_bug.cgi?id=1182 May 04 09:04:23 iirc it was fixed only in kernel.bbclass May 04 09:04:42 so similar fix should be applied to module.bbclass maybe May 04 09:05:13 ok I'll look fro this ... in fact compat-wireless I send yesterday generates 19MB .ko ;-) May 04 09:06:49 http://cgit.openembedded.org/?url=openembedded/commit/&id=18a377dcf50c5c10ebc73a36963a6e83d1bbad5d May 04 09:18:26 XorA: http://www.paolonutini.com/home.htm May 04 09:34:55 ant_work: OK I found the problem : ${KERNEL_MAJOR_VERSION} is not set May 04 09:35:20 mckoan: Ive got my New Shoes on :-) he is quite famous round here May 04 09:35:31 it's only set in kernel.bbclass May 04 09:35:39 oh..these KERNEL_ version vars are a can of worms May 04 09:35:56 iirc there are 3 of them May 04 09:59:46 ericben: nice catch, btw May 04 10:02:59 ant_work: well now I get ~300 to 400kB packages which is better than the > 20MB I had before ;-) May 04 10:04:05 do you have g3 in your flags? May 04 10:04:22 in which flags ? May 04 10:04:36 compiler flags May 04 10:07:47 I didn't change anything to the compiler flags so it should be using the default kernel ones May 04 10:11:00 just check the size of git-native and you'll see May 04 10:11:38 git-native will use the same flags as the kernel or modules ? May 04 10:12:10 iirc it does not override, so yes May 04 10:13:35 3.5MB May 04 10:13:55 for git binary May 04 10:14:06 ok, I can't check from here May 04 10:14:41 I was talking about this: May 04 10:14:43 FULL_OPTIMIZATION = "-fexpensive-optimizations -frename-registers -fomit-frame-pointer -O2 -ggdb3" May 04 10:14:54 (angstrom-glibc.inc) May 04 10:15:08 03Eric Benard  07org.openembedded.dev * rff57f14c99 10openembedded.git/recipes/mplayer/ (files/fix-emu_qtx_api.diff mplayer_svn.bb): (log message trimmed) May 04 10:15:08 mplayer_svn : fix compilation problem May 04 10:15:08 - this patch fix the following error (met for a x86 target) : May 04 10:15:08 | loader/module.o: In function `MODULE_GetProcAddress': May 04 10:15:08 | module.c:(.text+0x125): undefined reference to `report_entry' May 04 10:15:08 | module.c:(.text+0x12f): undefined reference to `report_ret' May 04 10:15:08 | module.c:(.text+0x139): undefined reference to `wrapper_target' May 04 10:15:47 are you sure these flags will be used for kernel & modules ? May 04 10:15:58 ericben: yes, but not for native packages May 04 10:16:08 (for angstrom-glibc at least) May 04 10:16:16 see angstrom-uclibc May 04 10:16:31 ok but kernel& modules don't use glibc or uclibc May 04 10:16:44 so why would they get these flags ? May 04 10:16:58 oops, sotrry, I was talking about general binaries May 04 10:17:07 ok I better understand now ;-) May 04 10:17:14 me too I'm in the run for minimal size May 04 10:18:22 and angstrom-uclibc was not the way :/ May 04 10:18:33 oops,I didn't sleep well(moskito),And I forget the fact that the patch could not apply while rebasing May 04 10:18:54 I'll test and if there is an issue,revert and ask for rebase May 04 10:26:31 03Koen Kooi  07org.openembedded.dev * r6810c91805 10openembedded.git/recipes/mozilla/ (16 files in 2 dirs): firefox: add 3.6.3 May 04 10:26:42 03Koen Kooi  07org.openembedded.dev * r94b64e22fa 10openembedded.git/recipes/gnome-mplayer/gecko-mediaplayer_svn.bb: geck-mediaplaer svn: bump SRCREV May 04 10:49:25 03Steffen Sledz  07org.openembedded.dev * r2ac717172c 10openembedded.git/recipes/rsyslog/ (rsyslog.inc rsyslog_5.4.0.bb): May 04 10:49:25 rsyslog: recipe cleanup May 04 10:49:25 Signed-off-by: Steffen Sledz May 04 10:49:28 03Steffen Sledz  07org.openembedded.dev * rcef6bb4d55 10openembedded.git/recipes/rsyslog/rsyslog_5.5.4.bb: May 04 10:49:28 rsyslog: new version 5.5.4 May 04 10:49:28 Signed-off-by: Steffen Sledz May 04 11:02:17 http://tinderbox.openembedded.net/public/logs/task/5729419.txt May 04 11:02:20 weird one May 04 11:04:33 updating bitbake ... May 04 11:05:46 03Koen Kooi  07org.openembedded.dev * r80ed94e7db 10openembedded.git/recipes/xorg-xserver/xorg-xserver-common.inc: xorg-xserver-common: make xvfb depend on xkeyboard-config May 04 11:05:47 03Koen Kooi  07org.openembedded.dev * rd09096ccb5 10openembedded.git/recipes/xkeyboard-config/xkeyboard-config_1.4.bb: xkeyboard-config: convert to new style staging May 04 11:22:26 http://tinderbox.openembedded.net/builds/70529/ May 04 11:22:30 Any thoughts? May 04 11:34:09 Crofton|work: For me this reads like an unitialized value... May 04 11:34:23 yeah May 04 11:34:27 I wonder where it comes from May 04 11:34:41 or does not :) May 04 11:37:50 and why does it only seem to happen to me? May 04 11:40:18 you got too scared and the variable ran away May 04 11:41:02 Crofton|work: I have seen that but I forget what causes it May 04 11:45:37 remember, or I will fly over and waterboard you May 04 11:45:42 volano permitting May 04 11:46:08 Crofton|work: I shall do your legs as well :-D May 04 11:47:45 heh, took me ages to plan that volcano to keep the yanks out of EU :-) May 04 11:49:18 hey all, i have bitbaked a customized at embedded image. After it is built i try to run a sample at embedded program like this " ./sample -qws" but then i get the following error May 04 11:49:20 QFontDatabase: Cannot find font directory /usr/lib/fonts May 04 11:49:43 kindly help May 04 11:50:16 most people got this error when they "configure"d the qt embedded with an improper prefix May 04 11:50:31 but for me bitbake did the configuration and all so how to resolve this May 04 11:50:34 ?? May 04 11:52:48 Spyzer: install some fonts? May 04 11:55:15 but i added RDEPENDS += qte as a line in my image file May 04 11:55:39 and many qt libraries were installed also in the built image, then why not the QFontDir May 04 11:55:41 ?? May 04 12:02:36 Spyzer: I think you need qt4-embedded-fonts May 04 12:05:31 so do i need to add REDPENDS += qt4-embedded-fonts in my inage.,bb file ??? May 04 12:07:12 Spyzer: add it to IMAGE_INSTALL. I'm just repeating what ericben wrote on the mailing list today :) May 04 12:11:32 Hi! - How do I strip my so-files my binaries so the debug-symbols end up in .debug/ ... my so-files are place in /usr/lib/vtk-5.4 May 04 12:41:27 hi there. i had a quick question about an OE build issue I had May 04 12:43:28 I'm building for a beagleboard, on an Ubuntu (10.04 LTS) host May 04 12:43:51 and I found that I was getting an error with some locale building, relating to qemu-arm May 04 12:44:37 some digging on google showed that this is related to a debian kernel setting restricting the lowest mmap address that qemu-arm can use, and thus it was failing May 04 12:45:07 I don't see anything about this in the OE help and thought it would be useful to other newbies like myself. The wiki page is locked and so I wondered who to contact ? May 04 12:46:23 when you run bitbake does it make a suggestion for setting some things in sysctl? May 04 12:46:33 03Koen Kooi  07org.openembedded.dev * r092a9d27ab 10openembedded.git/recipes/lighttpd/lighttpd.inc: lighttpd: add missing dep on zlip May 04 12:46:37 yes the problem is solved May 04 12:46:42 urg the wiki page is locked? May 04 12:46:47 Seems so ? May 04 12:47:00 you made an account? May 04 12:48:15 which page are you trying to edit? May 04 12:48:35 doh - no I hadn't logged into an account May 04 12:48:48 ah, so you can edit with an account? May 04 12:49:06 I'm also a little reluctant to dive in and start modifying your wiki pages to be honest. is that acceptable behaviour from a newbie ? May 04 12:49:24 well, it depends if your answer is correct :) May 04 12:49:34 which page? May 04 12:49:53 in this case you will add a note specific to one distro May 04 12:50:06 and not rewrite an entire process May 04 12:50:12 you will be ok :) May 04 12:50:15 I was thinking of either the general one May 04 12:50:16 http://wiki.openembedded.net/index.php/Getting_started#Start_building May 04 12:50:25 or more likely the Ubuntu specific section May 04 12:50:31 I see the link above is locked May 04 12:50:38 yeah, that would be the place May 04 12:50:42 I'll just check the distro-specfic page May 04 12:50:52 it is likely locked due to over enthusiastic people :) May 04 12:51:54 OK so this is maybe the right place - May 04 12:51:55 http://wiki.openembedded.net/index.php/OEandYourDistro#Ubuntu May 04 12:52:01 that's locked too (I'm logged in now) May 04 12:52:34 you know, just somewhere where it's really obvious this needs to be done, for beginners like me :) May 04 12:52:37 weird, I can edit it ... May 04 12:53:25 ah sorry - I'm misreading - It's locked unless you are registered. OK I'll add a line in there for Ubuntu if that's ok May 04 12:53:32 thanks for the help May 04 12:53:33 that's fine May 04 12:53:40 thanks! May 04 12:53:45 :) bfn May 04 12:54:28 Is it possible to make OE strip libraries that are not located in /usr/lib or /usr7local/lib ? May 04 13:31:01 re May 04 13:54:40 I can't compile python-pyqt with the latest from dev : http://pastebin.org/201108 May 04 14:42:21 tasslehoff: eww C++, I dont do that much :-) May 04 14:56:11 RP, around? May 04 14:56:55 hi, I'd like to fix sato image: it lacks : May 04 14:56:57 *dbus-x11 May 04 14:57:02 *matchbox-common May 04 14:57:30 but I wonder if theses dependencies should not rather go in the other recipes rather than in the image bb May 04 15:16:15 hrw|gone, thanks for the beaver :) May 04 16:23:35 Hi, guys May 04 16:27:06 CRofton: thank u 4 the help yesterday May 04 16:27:26 np May 04 16:27:30 we like happy users May 04 16:27:45 unfortunately the recipe didn't build, it gave an error during configure process May 04 16:27:58 so i am not yet "happy" May 04 16:28:11 do u want me to repost the error? May 04 16:28:16 hi IgorK May 04 16:28:31 hi florian, how's life May 04 16:28:46 still not bored :-) May 04 16:29:03 i decided to write a bb recipe 4 my program and failed ;-) May 04 16:29:26 u r still @ work? May 04 16:30:00 IgorK: yes May 04 16:31:10 having fun? May 04 16:31:50 CRofton: I had a power surge yesterday, so couldn't continue. when I got back u left the channel... May 04 16:35:13 Here is the error output: http://pastebin.com/i8wfEmrv May 04 16:41:57 IgorK: sure :-) May 04 16:42:32 IgorK: looks like there is a missing " or \ May 04 17:23:16 florian: thanx, this is fixed. but now I have this: http://pastebin.com/YLVKPCiN May 04 17:26:29 IgorK: get rid of your own do_configure and inherit autotools like other recipes do May 04 17:35:09 florian: I copied my recipe from the wxwidgets one. it has both "inherit autotools pkgconfig" and "do_configure()" function May 04 17:35:46 should I put "inherit autotools" and comment out "do_configure()"? May 04 17:37:06 IgorK: yes May 04 18:11:49 hi May 04 18:15:18 Anyone have any thoughts on how I should manage flags with this metadata hashing stuff? May 04 18:15:49 just changing task order around likely doesn't actually affect the output, it affects how things are done, not what things are done.. but adding a new task that deploys something new or something, that should likely be captured May 04 18:18:35 that is, its important for a full recipe signature. per task signatures obviously wouldn't be an issue, could exclude the flags entirely May 04 18:18:40 hey all, when i was bitbaking an image an error occurred with a package named linux-rp which is as such May 04 18:18:48 tar: . file changed as we read it May 04 18:18:59 what does this error signify?? May 04 18:19:12 i mean how did this occur May 04 18:19:25 sounds like it signifies exactly what it says ;) May 04 18:19:32 what task did it fail in? May 04 18:20:04 linux_rp 2.6.25 May 04 18:20:14 did it download incorrectly May 04 18:20:16 that isn't a task. May 04 18:20:27 fetch, unpack, patch, configure, compile, install, ... May 04 18:20:32 look at the messages May 04 18:20:34 ideally, pastebin the log May 04 18:20:47 how do i actually resolve this?? task was staging_packager May 04 18:21:01 do you think I have a crystal ball? May 04 18:21:09 i can't fix a problem if you haven't given us enough information to even diagnose it May 04 18:21:12 i am sorry :( May 04 18:21:21 i will do it right away May 04 18:21:36 don't apologize to me, you're the one affected by our inability to help you :) May 04 18:21:46 i am actually on a non GUI environment and hence facing some problems May 04 18:22:02 regarding proper communication May 04 18:22:31 again, you ran bitbake, you can see its output. copy/paste that into a pastebin web site and put the link here May 04 18:23:35 just let me say one more thing(i ask for ur patience, i do understand what u r saying) that in the log file there is only one line written and that is May 04 18:23:37 tar: . file changed as we read it May 04 18:23:51 as far as the shell errors are concerned May 04 18:24:00 bitbake outputs things May 04 18:24:01 NOTE: messages May 04 18:24:04 i am unable to recreate them May 04 18:24:18 hmm May 04 18:24:36 well, there's not enough information to diagnose. does it say which tar file it was extracting? May 04 18:25:21 still its something like NOTE: linux-rp_2.6.25+2.6.something do_package_stage failed May 04 18:25:37 okay, that's what i originally asked for, thank you May 04 18:25:39 but it didn't say anything about any tar package May 04 18:26:03 sorry to say I don't know what caused that :9 May 04 18:26:05 :( May 04 18:26:13 maybe someone else has seen it before May 04 18:26:36 should i try to find and delete the corresponding tar package May 04 18:26:39 ?? May 04 18:26:45 and let bitbake redownload it May 04 18:27:09 it wasn't the source tarball May 04 18:27:22 it was most likelyl the tar file inside of an ipk file for the packaged-staging stuff May 04 18:27:38 i'd suggest doing a bitbake -c clean linux-rp and rebuilding, hopefully it was just a fluke :) May 04 18:28:44 thnx i shall try it right away :) May 04 18:51:38 * Tartarus tries perl-native non-legacy staging on 5.8.8 the will force 5.10 to be picked May 04 18:54:44 * kergoth_ thinks about how to handle variable references of things which aren't literal strings, from python code.. maybe we just need a flag to indicate which variables are utilized May 04 18:59:39 ugh, need to rewrite the blacklist transformation May 04 18:59:46 hrm May 04 19:56:10 mmm boost 1.36 is still the default in angstrom... May 04 19:59:55 grr May 04 20:00:05 * kergoth_ fights with python May 04 20:05:52 be carefull it's venomous May 04 20:08:04 heh May 04 20:08:14 perl-native 5.8.8 is good non-legacy, trying 5.10.1 now May 04 20:11:19 Tartarus: nice May 04 20:24:58 GNUtoo, I have a patch to remove all the DEF_PREF = "-1" in the boost recipes May 04 20:25:15 once I confirm the stuff I care about is ok with it, I am going to submit it May 04 20:25:20 ok thanks May 04 20:25:39 it is not Angstrom, just the ways the recipes were added May 04 20:25:47 boost tends to make rendom api changes May 04 20:25:57 I know May 04 20:26:05 that the api changes between versions May 04 20:26:07 but I think everything we care about is ok with later boosts now May 04 20:26:08 which is an issue May 04 20:26:09 yeah May 04 20:26:12 ok May 04 20:26:13 nice May 04 20:26:57 the only issue is the most recent recipe used the cmake versio of boost, which does not seem well maintained May 04 20:28:09 I know May 04 20:28:20 I think the last commit was made long time ago May 04 20:28:25 yeah May 04 20:28:32 what are our options with that May 04 20:28:33 ? May 04 20:28:34 there is a later version of boost May 04 20:28:43 ah? May 04 20:28:48 use the mainline supported configure stuff May 04 20:28:57 which I think is still based on bjam :( May 04 20:28:58 ah yes indeed May 04 20:29:08 I just looked at the wikipedia and it says 1.4.2 May 04 20:29:21 ouch bjam... May 04 20:29:33 morning :) May 04 20:29:52 * kergoth_ grumbles May 04 20:30:51 http://tree.celinuxforum.org/pipermail/celinux-dev/2010-May/002026.html May 04 20:31:22 wow May 04 20:32:55 damnit May 04 20:32:57 i hate when i lose code May 04 20:33:16 i know i have this somewhere .. don't even know what machine it was on, much less what file and what dir May 04 20:33:19 hrmph May 04 20:35:58 * Jay7 trying to remember who have done bootchart logging in oe-devel ML.. May 04 20:38:10 grrr May 04 20:38:20 we really need better tools for debugging problems in the python code in recipes May 04 20:40:50 kergoth_: RFC it :) May 04 20:42:50 well, the question isn't whether it should be done, the question is *how* May 04 20:47:08 * kergoth_ says screw it and does a pickle.dump() of the entire datastore so he can test generating signatures against it outside of OE May 04 20:48:14 may be implement some tracing code in bitbake? May 04 20:48:42 yeah something like that would be nice May 04 20:48:58 or even the ability to jump in with an interpreter in a task May 04 21:03:01 oh man, that was easy.. pickling the data and testing outside of OE, took all of 20 seconds to fix this problem that's been troubling me on and off all day May 04 21:03:04 * kergoth_ rolls eyes May 04 21:07:54 hmm.. i wonder if there's a way to eval/exec a python ast.. seems silly to go from the string again when you've already compiled to the ast May 04 21:08:19 03Tom Rini  07org.openembedded.dev * r12c5d82663 10openembedded.git/recipes/perl/ (perl-native_5.10.1.bb perl-native_5.8.8.bb): May 04 21:08:19 perl-native: Switch away from legacy staging, bump PR. May 04 21:08:19 Signed-off-by: Tom Rini May 04 21:09:54 * Tartarus waits on patchwork May 04 21:12:23 03Eric BENARD  07org.openembedded.dev * r8c7f50cd36 10openembedded.git/recipes/crda/crda_1.1.1.bb: May 04 21:12:23 crda: add version 1.1.1 May 04 21:12:23 - http://wireless.kernel.org/en/developers/Regulatory/CRDA May 04 21:12:23 - CRDA acts as the udev helper for communication between the kernel and May 04 21:12:23 userspace for regulatory compliance. May 04 21:12:23 Signed-off-by: Eric Benard May 04 21:12:23 Signed-off-by: Tom Rini May 04 21:12:34 03Eric Benard  07org.openembedded.dev * r0e35beaa5a 10openembedded.git/recipes/python/python-m2crypto_0.18.2.bb: May 04 21:12:34 python-m2crypto: add native May 04 21:12:34 - this package is needed to compile crda May 04 21:12:34 Signed-off-by: Eric Benard May 04 21:12:35 Signed-off-by: Tom Rini May 04 21:17:16 i'm trying to boot from jffs2 and get No filesystem could mount root, tried: ext2 vfat ntfs fuseblk May 04 21:17:37 so the kernel says it has no jffs2 support? May 04 21:18:07 try to pass rootfstype=jffs2 to kernel in bootloader May 04 21:18:12 did that May 04 21:19:02 then seems yes, kernel have no jffs2 support May 04 21:19:25 when i boot that kernel with root on nfs /proc/filesystems includes jffs2 May 04 21:20:22 hm.. May 04 21:21:31 is there any extra option to allow boot from jffs2 ? May 04 21:22:11 I can't remember any May 04 21:22:12 W1zzard: apart from jffs2 support in the kernel you need to pass "rootfstype=jffs2" to it. May 04 21:22:22 btw May 04 21:22:24 florian: hey May 04 21:22:37 its not one of the default filesystems which are tried per default May 04 21:22:40 florian: about OE stand on LinuxTag May 04 21:22:41 hi Jay7 May 04 21:22:53 florian: can anyone show kexecboot there? ;) May 04 21:23:44 Jay7: I guess we shiuld have some quite good OE-built examples here :-) May 04 21:24:25 yeah :) May 04 21:24:28 damn, forgot about the methodpool May 04 21:24:31 grr May 04 21:25:17 I hope there will be support for TouchBook and Ben NanoNote at that moment.. May 04 21:25:27 zauruses are too old already :) May 04 21:25:45 problem solved .. for some reason my bootloader corrupts the setenv string i'm giving it setenv bootargs=mem=64 rootfstype=jffs2 ended up as mem=64=rootfstype=jffs2, note the = before rootfstype t.t fixed by setenv bootargs=mem=64M foo=bar rootfstype=jffs2 May 04 21:25:56 can you still load linux on one of those old sharp zaurus'? May 04 21:26:51 rhc: yes :) May 04 21:27:35 but newer kernels have problems with offline charging May 04 21:27:56 latest really working kernel is 2.6.24 afaik May 04 21:35:30 shit.. I know I've seen this one before May 04 21:35:41 hrm May 04 21:35:49 oh.. i bet i know.. shit May 04 21:35:50 hmm May 04 21:38:15 * kergoth_ ponders May 04 21:39:34 this is a potential problem.. to do the reference tracking it has to expand the tasks... if the task uses a var that has a snippet that pulls something from staging (i.e. the kernel version functions), it'll explode, because it's parsing earlier than the code expects May 04 21:44:49 hmmm... i forgot about __radd__.. that'd be helpful for getting variable typing into bitbake May 04 22:15:44 03Andrea Adami  07org.openembedded.dev * rc504712189 10openembedded.git/recipes/opie-init/opie-init_1.2.4.bb: May 04 22:15:45 opie-init: uncomment c7x0 specific stuff May 04 22:15:45 * was commented in ver 1.2.4 but there are still bits in opie-init.inc May 04 22:15:45 * unbreak the build for the moment May 04 22:15:45 * still needs testing about w100 in opie May 04 22:16:27 ant__: hey May 04 22:16:35 hello May 04 22:17:08 remember the initrd! I had a patch somewhere... May 04 22:26:47 http://github.com/kergoth/OE-Signatures - could use code review / thoughts / opinions :) May 04 22:30:51 isnt there a kernel command line option to define jffs2 partition via cmdline ? May 04 22:32:01 define define May 04 22:32:20 tell the kernel where the partitions are May 04 22:32:29 without modifying the kernel source May 04 22:32:33 mtdparts= in cmdline May 04 22:32:45 thanks! May 04 22:32:51 yw May 04 22:54:32 argh May 04 22:57:27 this is way more work than i expected when i started working on this :) May 04 23:36:37 kergoth: Yeah usually happens in untreaded zones :) May 04 23:36:58 very true May 04 23:45:33 kergoth: once I wanted to fix an allocation bug in assembler May 04 23:45:45 and I ended up writing a whole new memory pool manager :) May 04 23:50:17 khem: hehe, yikes May 04 23:51:21 think I'm getting close to having these classes in shape.. now whether they're worth merging into bitbake or not is something else entirely.. as is trying to figure out how the fuck to do that May 05 01:25:55 hi , i m having a winsock socket problem on windows , i ve disconnect my modem on internet but my c++ code still hangs on recv method , it should return -1 or 0 when i disconnect modem but its not ? i know its openembedded channel but there is only 1 channel to ask here... :( May 05 01:27:05 huh? May 05 01:27:11 there are tons of irc channels on the network May 05 01:27:21 if you don't know about them or can't find the right one, that doesn't make it on topic here May 05 01:28:33 Also, winsock? Is this 1996? May 05 01:30:58 lol May 05 01:31:00 yea :D May 05 01:31:09 i really hated it but i must do May 05 01:31:17 my code hangs on recv May 05 01:31:30 although modem dc May 05 01:31:32 yes, we heard you the first time May 05 01:31:40 and again, go elsewhere May 05 01:31:42 try google, it helps May 05 01:31:57 i'm sure there are thousands of sockets tutorials May 05 01:33:10 i tried all on google May 05 01:33:15 none of them worked May 05 01:33:19 not our problem May 05 01:33:23 setsockopt(socket2,SOL_SOCKET,SO_KEEPALIVE,(char*)&optval,sizeof(bool)); May 05 01:33:36 SO_KEEPALIVE should solve the problem May 05 01:33:43 but its not solving... **** ENDING LOGGING AT Wed May 05 02:59:56 2010