**** BEGIN LOGGING AT Tue Dec 13 02:59:57 2011 Dec 13 07:56:29 When I Run the $bitbake console image It gives warning to run with sudo but When I try with Sudo it gives bitbake command not found... Please Help me! Dec 13 08:00:12 Basith: guess, you have a typo, try bitbake console-image Dec 13 08:05:29 Sorry! I don't have typo. any other Idea! Dec 13 08:11:30 CFP announced for ELC Dec 13 08:11:36 https://events.linuxfoundation.org/events/embedded-linux-conference/cfp Dec 13 08:50:51 morning all Dec 13 08:51:42 when I build $bitbake console image I get the error: 'ftp://ftp.ayamura.org/pub/gnu/autoconf/autoconf-2.68.tar.bz2/autoconf-2.68.tar.bz2' failed with signal 4, output: Dec 13 08:52:26 morning bluelightning Dec 13 08:54:42 Hi all Dec 13 09:01:29 good morning Dec 13 09:21:57 Is there a way to find out which package pulls in a certain dependency? Dec 13 09:23:29 I noticed the bitbake -g option but using dotty on a file with hundreds if not thousands of lines is a challenge : ) Dec 13 09:30:31 kenws: try bitbake -g -u depexp Dec 13 09:31:08 bluelightning: thanks! I'll have a look Dec 13 09:34:17 When I Build Bitbake ftp file downloading failed. When I test wget ftp://url working fine only in the root . I think that's why Not working when I run bitbake with normal user. so what can i do? Dec 13 09:36:03 Basith: What's the error when using wget as a regular user? Dec 13 09:36:32 mornin Dec 13 09:36:39 Maybe there are different proxy settings (just a wild guess) Dec 13 09:37:00 mornin Jin^eLD Dec 13 09:38:39 Yes I have configured proxy settings. Not downloading files via ftp:// server. I try to download the file via root user it is downloaded. Dec 13 09:40:19 Basith: is it just ftp or other protocols as well? (mabe its pasv related) Dec 13 09:42:08 download file http working fine. ftp also worked when i try to download the file in root user. but normal users it say connected then wait for long time anonymous user then return failed Dec 13 09:43:51 $ bitbake console-image it gives the warn message "Pseudo is not present but is required" Dec 13 09:46:56 Basith: wrt to the wget, I'd check for different environment variables (proxy settings) and compare the .wgetrc for root and regular users Dec 13 09:49:04 When I Buid gives "Pseudo is not present but is required" Dec 13 09:50:15 Basith_: that's normal, it then proceeds to build pseudo Dec 13 09:52:39 but When i try to Run $sudo bitbake console image it gives the error " bitbake command not found" Dec 13 10:06:54 pls help me! Dec 13 10:07:03 um, why are you trying to run bitbake under sudo? Dec 13 10:07:19 sudo is evil anyway Dec 13 10:08:08 initially I tried without sudo It failed to download the files from ftp server Dec 13 10:31:49 NOTE: package autoconf-native-2.68-r4: task do_fetch: Started .. . after long time it gives error ftp failed. but when i try to download that file using wget ftp://file It worked only in root . normal after long time it gives the error Dec 13 10:44:57 morning Dec 13 10:45:08 hi otavio Dec 13 11:00:45 hi ericben Dec 13 11:32:35 * kenws wonders why depexp shows eglibc-locale as a reverse dependency of eglibc-locale Dec 13 11:38:11 http://imagebin.org/188369 - that's strange, isn't it? Dec 13 11:48:38 kenws: it's an odd artifact I agree Dec 13 11:48:46 I don't think it does any harm Dec 13 11:49:03 Any idea on how to find out what pulls in eglibc-locale? Dec 13 11:50:23 kenws: opkg whatdepends eglibc-locale Dec 13 11:50:33 probably easiest if you have a running system Dec 13 11:51:06 I'm using an external toolchain and currently oe is using the libc of the sysroot (might not be ideal but is a start). It works fine for the minimal core image but anything bugger (sato or qt4) pulls in eglibc-locale Dec 13 11:53:02 hi i wonder anyone could assist me Dec 13 11:53:17 i have a xxx.cpio i manage to embed into my kernel Dec 13 11:53:23 i boot it up and got it working Dec 13 11:53:48 any idea how do i do this in another way but seperating the two files? Dec 13 11:53:53 by* Dec 13 11:54:06 load kernel follow by loading .cpio Dec 13 11:54:14 instead of combining them together Dec 13 11:59:27 Jin^eLD: thanks for the hint but as soon as eglibc-locale gets pulled in the build fails (since the libc is coming from the external toolchain) Dec 13 11:59:48 kenws: ah, you have this problem during the build Dec 13 11:59:55 yep Dec 13 12:00:22 kensw: try -g maybe, for the dependency graph? the output .dot is human readable Dec 13 12:00:44 Jin^eLD: yeah, it's quite large : ) Dec 13 12:01:03 well, you then just search for the eglibc-locale string Dec 13 12:01:15 and see what depends on it Dec 13 12:08:54 so, the task-depends.dot shows the dependencies of the various tasks involved Dec 13 12:09:14 I think it spits out 3 dots Dec 13 12:09:18 while the package-depends.dot show deps for the packages Dec 13 12:09:22 yeah Dec 13 12:09:32 what's the purpose of pn-depends.dot? Dec 13 12:09:48 I am not sure, I usually found what I was looking for in the package-depends Dec 13 12:13:43 ok and I guess an edge means the package on the left depends on whatever is stated on the right Dec 13 12:22:25 hm, it states that the "eglibc-locale-dbg" depends on "eglibc-locale" but nothing depends on "eglibc-locale-dbg" Dec 13 12:24:53 http://paste.ubuntu.com/768880/ Dec 13 12:25:29 weird, I don't see how the eglibc-locale gets pulled in Dec 13 12:26:26 kenws: have you searched the file for all occurances? Dec 13 12:26:40 and nothing depends on the eglibc-binaries either Dec 13 12:26:55 Jin^eLD: yes, it's a `grep glibc package-depends.dot` Dec 13 12:27:41 btw what did you bitbake with -g? Dec 13 12:27:43 which recipe? Dec 13 12:27:58 bitbake -g core-image-sato Dec 13 12:28:11 (but I'm using an external toolchain recipe) Dec 13 12:29:55 my recipe basically does: http://paste.ubuntu.com/768888/ Dec 13 12:31:34 Do I need to "cleansstate world" prior generating the graph with -g? Dec 13 12:31:47 can't say much, I never tried external toolchains Dec 13 12:31:57 kenws: no Dec 13 12:32:36 bluelightning: ok Dec 13 12:32:52 kenws: btw what target are you specifying with -g? Dec 13 12:33:12 bluelightning: bitbake -g core-image-sato Dec 13 12:33:39 ok Dec 13 12:34:10 the core-image-minimal works fine, but I'd like to build something bigger Dec 13 12:35:44 heya, I have a problem with bitbake eating all my ram, when trying to build core-image-minimal Dec 13 12:37:50 VQuickSilver: What's the error? Is the oom killer killing your processes? How much RAM do you have. Do you specify BB_NUMBER_THREADS and PARALLEL_MAKE? Dec 13 12:38:45 kenws, I have to kill it manually before it takes all my ram, and computer stars trashing, I don't have swap, I have 4GB of ram Dec 13 12:39:23 kenws, BB_NUMBER_THREADS=2 but I see the same behaviour with 1 Dec 13 12:39:25 hi wonder if anyone know how to load .cpio.gz and kernel? Dec 13 12:40:05 kenws, I'm using shr Makefile Dec 13 12:40:08 VQuickSilver: Do you know what is eating all your memory? Dec 13 12:40:10 I've 4GB of ram too and I'm fine with the following settings: Dec 13 12:40:37 PARALLEL_MAKE = "-j 4" Dec 13 12:40:38 BB_NUMBER_THREADS = "4" Dec 13 12:41:09 kenws, I now it's bitbake cause if I do ctrl+c to it ram it's freed Dec 13 12:41:19 kenws, I see the ram used with conky Dec 13 12:41:31 VQuickSilver: where does it begin to fail? Dec 13 12:41:49 VQuickSilver: I meant what exactly allocates the memory (is it the gcc during the compile for example) Dec 13 12:42:02 htop can gives % on memory Dec 13 12:42:10 kenws, I don't know, I don't see any messages on screen Dec 13 12:42:27 VQuickSilver, try htop Dec 13 12:42:42 set the sorting to memory instead of cpu Dec 13 12:42:45 and look Dec 13 12:43:08 GNUtoo, ok Dec 13 12:48:19 I can see a lot of bitbake threads Dec 13 12:49:14 if there are no indicators that something strange is going on you may just create a swap file and mount it Dec 13 12:49:35 (to check if adding some memory really helps) Dec 13 12:49:54 kenws, ok I'm gong to add a 2GB swap file Dec 13 12:50:29 you may limit BB_NUMBER_THREADS to one Dec 13 12:50:34 kenws, there is no way to make bitbake print what is doing? seems like a old school fork bomb Dec 13 12:51:35 hm, is it really bitbake or some other processes that bitbake starts? Dec 13 12:51:39 PARALLEL_MAKE = "-j 2" Dec 13 12:51:39 BB_NUMBER_THREADS = "2" Dec 13 12:51:58 you may check using ps -fux|less Dec 13 12:52:08 kenws, wait a second I will copy the htop line Dec 13 12:53:25 /bin/sh /OE/shr-core/openembedded-core/scripts/bitbake core-image-minimal Dec 13 12:53:56 lots of them with diferent PIDS Dec 13 12:54:05 and 0.1 MEM Dec 13 12:57:18 anyone know what this means? Dec 13 12:57:19 ## Loading Ramdisk Image at 23000000 ... Bad Magic Number Dec 13 12:58:54 kaylessa, maybe that at 0x23000000 there is something else than a ramdisk image Dec 13 12:59:12 ok thanks i try load at other address Dec 13 12:59:38 that's from your bootloader right? Dec 13 12:59:39 and also what is the correct way of cleaning of all bitbake cache etc...? in the case I want to build another image? Dec 13 12:59:46 yup from bootloader Dec 13 12:59:48 uboot Dec 13 12:59:49 VQuickSilver: hm, smells fishy - I see only one "/bin/sh /path/to/bitbake" (that starts python /path/to/bitbake) Dec 13 13:00:07 VQuickSilver: you could do a: bitbake -ccleansstate world Dec 13 13:00:17 (but maybe there are better ways) Dec 13 13:00:22 kenws, ok Dec 13 13:00:29 thanks Dec 13 13:00:46 or rm/move your tmpdir Dec 13 13:00:58 since you have only one machine you're building for Dec 13 13:01:03 erm is this command valid in uboot? Dec 13 13:01:21 bootm XXXXXXXX(kernel) XXXXXXXX(cpio.gz) Dec 13 13:01:51 kaylessa, can you include the cpio.gz inside the kenrel itself? Dec 13 13:01:57 or does it makes it too big? Dec 13 13:02:12 i manage to embed cpio into my kernel and got it working Dec 13 13:02:16 I guess uboot has an help Dec 13 13:02:18 but my prof wan to seperate it Dec 13 13:02:21 ok Dec 13 13:02:37 i am now trying to load them seperately Dec 13 13:02:43 wondering if anyone could help Dec 13 13:02:44 ok Dec 13 13:03:30 I usually don't use initramfs on embedded stuff Dec 13 13:03:35 so I don't know it well Dec 13 13:03:39 :( Dec 13 13:03:56 but I wonder if the kenrel accepet as argument the address of the cpio.gz Dec 13 13:03:56 know anyone that could help? Dec 13 13:04:09 look at the kenrel-parameters.txt Dec 13 13:04:17 yup i setenv bootargs initrd=XXXXXXXX(cpio.gz) Dec 13 13:04:20 but it doesnt work Dec 13 13:04:26 ah ok Dec 13 13:04:40 but at XXXXXX... is there already an initrd.gz? Dec 13 13:04:45 *cpio.gz Dec 13 13:04:56 i doubt so Dec 13 13:05:01 i only load a kernel Dec 13 13:05:15 i have to load between 20000000-2FFFFFFF Dec 13 13:05:21 basically you have to load it in memory Dec 13 13:05:21 i load kernel at 20004000 Dec 13 13:05:29 and then pass the address to the kernel Dec 13 13:05:29 and cpio.gz at other place Dec 13 13:05:30 try that Dec 13 13:05:32 it doesnt work Dec 13 13:05:44 yup i pass the address over Dec 13 13:05:47 it doesnt read Dec 13 13:05:50 lol Dec 13 13:06:05 when your kenrel boot it prints the cmdline at the very beginning Dec 13 13:06:22 (and once it's booted you have it at /proc/cmdline) Dec 13 13:06:40 let me try load kernel only Dec 13 13:07:47 GNUtoo, something is wrong with my install cause even bitbake -ccleansstate world takes all my ram Dec 13 13:08:00 hmmm Dec 13 13:08:10 does bitbake use a wrapper? Dec 13 13:08:32 I don't remember well Dec 13 13:08:41 but it seem that there should be a bitbake wrapper Dec 13 13:08:47 for pseudo or something like that Dec 13 13:08:57 I'm going to make an chroot-update etc... to see if it solves the problem Dec 13 13:08:58 that may explain the sh Dec 13 13:09:25 maybe I erased some file that I shouldn't Dec 13 13:09:27 btw for htop: Dec 13 13:09:29 F6 Dec 13 13:09:33 sort by Dec 13 13:09:40 you did that? Dec 13 13:10:10 GNUtoo, yeah, like a said lot of processes with 0.1% RAM usage Dec 13 13:10:27 ok Dec 13 13:10:29 When I Build OpenEmbedded I get the error: task do_unpack: Failed with gcc-cross-initial file. Please help me! Dec 13 13:11:00 Basith, maybe clean the toolchain and retry Dec 13 13:11:00 ah busy day Dec 13 13:11:04 many people need help Dec 13 13:11:08 indeed Dec 13 13:11:21 i setenv bootargs initrd=21000000 Dec 13 13:11:32 Kernel command line: initrd=21000000 Dec 13 13:11:40 but still cannot find my cpio.gz :( Dec 13 13:11:48 yes but what does it says Dec 13 13:11:53 Please append a correct "root=" boot option Dec 13 13:11:59 Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0) Dec 13 13:12:16 ah but does the initrd run? Dec 13 13:12:24 meaning? Dec 13 13:12:33 meaning does it run before panicking Dec 13 13:12:38 yup it run Dec 13 13:12:51 then it doesn't find the thing to switch_root too Dec 13 13:12:56 beware with switch_root btw Dec 13 13:12:59 ?? Dec 13 13:13:02 whats that? Dec 13 13:13:03 it's the most dangerous command ever Dec 13 13:13:24 just read the man Dec 13 13:14:10 u mean manual? Dec 13 13:14:17 man switch_root Dec 13 13:14:28 it implements an efficient rm -rf * Dec 13 13:15:02 o.O? Dec 13 13:15:13 see ./filesystems/ramfs-rootfs-initramfs.txt in the kernel documentation Dec 13 13:15:15 i am still kinda newb to embedded linux Dec 13 13:15:18 ok Dec 13 13:15:45 btw please don't use that knowledge of switch_root for evil purposes....it would be devastating Dec 13 13:16:04 more confused lol Dec 13 13:16:45 same error after I have cleaned and retry. Dec 13 13:17:01 Doesn't uboot require an uInitrd? Dec 13 13:17:28 and uImage Dec 13 13:17:33 kaylessa, does it work with 0x before the address in the kernel paramteter Dec 13 13:17:34 ? Dec 13 13:17:46 let me try :) Dec 13 13:18:07 ah sorry Dec 13 13:18:15 It already works and doesn't find the rootfs right? Dec 13 13:18:27 paste the full serial log Dec 13 13:18:31 so we see more clearly Dec 13 13:19:16 ok Dec 13 13:19:17 (I'm doing 2 things at once and it doesn't help....) Dec 13 13:22:45 https://docs.google.com/document/d/1Up4nJXmJD6uEy4Esn-0h_wG5eVr28B4yPYgetLAb3oI/edit?hl=en_US Dec 13 13:23:16 kaylessa, use a pastebin next time Dec 13 13:23:23 ok Dec 13 13:24:02 kaylessa, first I don't see any initrd loaded Dec 13 13:24:08 second the kernel is very old.... Dec 13 13:24:11 2.6.21 Dec 13 13:24:23 I wonder if 2.6.21 has initrfamfs Dec 13 13:24:28 maybe it has only initrd Dec 13 13:24:34 yea it should have Dec 13 13:24:40 ok Dec 13 13:24:47 i have another one that has embeded rfs Dec 13 13:24:49 and it works Dec 13 13:24:59 just seperating doesnt work Dec 13 13:25:00 ok Dec 13 13:25:22 what are all the command you type for loading the initrd? Dec 13 13:25:26 btw pastebin: http://www.pastie.org/pastes/new Dec 13 13:26:43 first i have my .cpio.gz file and my kernel Dec 13 13:26:54 i load .cpio.gz at 23000000 Dec 13 13:26:58 kernel at 20004000 Dec 13 13:27:11 then i setenv bootargs initrd=0x23000000 Dec 13 13:27:23 then bootm 20004000 23000000 Dec 13 13:27:51 kaylessa: and the .cpio.gz contains the stuff for your initrd? Dec 13 13:28:13 inside is a .cpio file Dec 13 13:28:20 i presume it should? Dec 13 13:28:33 I don't know if things are different with the oe-uboot but I needed to create one like this: Dec 13 13:29:32 mkimage -A arm -O linux -T ramdisk -C none -a 0x0 -e 0x0 -n "My Initrd" -d /path/to/your.cpio.gz uInitrd Dec 13 13:30:18 to boot it I used something like this: fatload mmc 0:1 0x80200000 uImage; fatload mmc 0:1 0x81600000 uInitrd; bootm 0x80200000 0x81600000 Dec 13 13:30:47 (that was for my pandaboard, so I probably won't work on your system) Dec 13 13:30:57 mine is a very old board Dec 13 13:31:08 AT91RM9200 Dec 13 13:31:11 I think bootm foo 23000000 is redondant with initrd=0x23000000 Dec 13 13:31:34 When bulid OE, I get the error gcc-4_5-branch_gcc.gnu.org_.svn.gcc.branches_181733_.tar.gz unpack failed. Dec 13 13:31:39 yea if i didnt setenv initrd Dec 13 13:31:42 but I still think you need to create a uInitrd Dec 13 13:31:46 i will get bad magic number Dec 13 13:34:01 Basith: maybe svn checkout of the gcc failed - you could check: downloads/gcc-4_5-branch_gcc.gnu.org_.svn.gcc.branches_181733_.tar.gz and downloads/svn/gcc.gnu.org/svn/ Dec 13 13:35:32 ahhh i am going crazzyyy Dec 13 13:51:35 kaylessa, try to find some documentation on it instead of blindly trying everything Dec 13 13:51:40 I'll go bye Dec 13 14:02:41 I am having trouble using fetch mirrors Dec 13 14:02:58 I've add an entry on KERNELORG_MIRROR at mirrors.bbclass but seems to not get used Dec 13 14:04:45 How do I enforce the use of the mirror? Dec 13 14:10:45 otavio: afair, I just added KERNELORG_MIRROR = "http://mirror.gbxs.net/pub" to my local.conf Dec 13 14:12:14 but I'd also expect that altering the meta/classes/mirrors.bbclass works too Dec 13 14:12:28 kenws: right but mirrors.bbclass seems unused Dec 13 14:12:44 mirrors.bbclass is inherited by base Dec 13 14:12:50 I don't see how it could possibly be unused Dec 13 14:12:53 grep is your friend Dec 13 14:13:21 kergoth: but seems to not work Dec 13 14:13:25 does for me Dec 13 14:13:32 * kergoth shrugs Dec 13 14:13:51 what's the behavior? Dec 13 14:14:06 kergoth: i did http://paste.debian.net/149101/ Dec 13 14:14:16 kergoth: and tried to fetch util-linux Dec 13 14:14:18 kergoth: fails Dec 13 14:14:48 checked the do_fetch detailed log? should show each mirror fetch attempt Dec 13 14:17:18 hmmm Dec 13 14:18:43 kergoth: have you been able to reproduce it? Dec 13 14:18:59 * kergoth checks Dec 13 14:33:33 kergoth: ? Dec 13 14:43:01 any one need help http://pastebin.com/h6WM997L Dec 13 14:44:51 kashyap1: what does `file /usr/share/sources/git_www.denx.de.git.u-boot.git._9912121f7ed804ea58fd62f3f230b5dcfc357d88.tar.gz` give? Dec 13 14:47:35 hold on il check Dec 13 14:49:19 what do you mean by give? i tried to open that file but it gives an read error Dec 13 14:49:47 it looks like it fails when attempting to extract it Dec 13 14:50:07 actually when the build was on, the downloading of a package froze because of internet disconnection Dec 13 14:50:12 when i tried it now Dec 13 14:50:18 it doesn work Dec 13 14:50:24 I meant to run "file" on that file : ) Dec 13 14:50:33 can you give me some solution Dec 13 14:50:54 kashyap1: bitbake -c clean u-boot Dec 13 14:50:58 ok Dec 13 14:51:02 let me try Dec 13 14:51:05 then continue with the build you were running Dec 13 14:51:09 ok Dec 13 14:51:59 blue: ERROR: Nothing PROVIDES 'uboot' (but '[]' DEPENDS on or otherwise requires it) Dec 13 14:52:31 kashyap1: maybe it's u-boot vs. uboot Dec 13 14:52:49 sorry Dec 13 14:52:51 it worked Dec 13 14:55:26 kenws: bluelighting: clean boot done still doesnt work http://pastebin.com/wPTDiCQM Dec 13 15:02:58 kashyap1: well, it still looks like tar/gunzip isn't able to unpack the archive. you may check the file (and also the git tree) Dec 13 15:04:54 can i replace the file with the working one? Dec 13 15:08:43 kashyap1: check the SRC_URI of your /home/m04cde-cw-group_2/gumstix/gumstix-oe/org.openembedded.snapshot/packages/uboot/u-boot-mkimage-openmoko-native_oe.bb Dec 13 15:11:08 Kenws: i just downloaded the tar file which was not able to unpack and replaced it and it worked Dec 13 15:11:57 Kenws: what does this mean : Errno 2] No such file or directory: '/home/m04cde-cw-group_2/gumstix/gumstix-oe/user.collection/packages/*/*.bb' Dec 13 15:12:59 kashyap1: I'm not really familiar with oe, I just started Dec 13 15:13:26 maybe you configured a layer (conf/layer.conf) that sets the BBFILES variable or something like this Dec 13 15:14:14 kashyap1: Did you read the http://docs.openembedded.org/usermanual/usermanual.html ? Dec 13 15:14:20 no Dec 13 15:14:30 that might be a good thing to start with Dec 13 15:14:31 : ) Dec 13 15:14:39 thank you so much Dec 13 15:17:27 kashyap1: bitbake -c clean u-boot-mkimage-openmoko-native Dec 13 15:19:33 bluelidghting i have finished that part Dec 13 15:19:48 i mean i just replaced the tar file with the working one Dec 13 15:20:04 and it worked but i dun no if i get some problems in future Dec 13 15:20:36 im jus worried about Errno 2] No such file or directory: '/home/m04cde-cw-group_2/gumstix/gumstix-oe/user.collection/packages/*/*.bb<<<<<<<>>>>>it should not be a problem? Dec 13 15:21:56 like kenws said above, either your configuration or one of the layers you have enabled is setting BBFILES to point to files that don't exist Dec 13 15:23:09 will it create a prob later? Dec 13 15:23:40 it's hard to say Dec 13 15:23:59 if there aren't meant to be any files in user.collection then no Dec 13 15:24:36 referring to http://gumstix.org/software-development/open-embedded/161-openembedded-directory-layout.html Dec 13 15:24:52 looks like it's somewhere you can put your own recipes, so no it won't be an issue Dec 13 15:26:09 thanks Dec 13 16:20:42 has anyone created a recipe for adding posix_ipc to python? Dec 13 16:49:33 hi Dec 13 16:49:37 is there anyone who could help me in writing a recipe? Dec 13 16:54:13 taedium: please just ask your question(s) :) Dec 13 16:56:10 I'm trying to write a recipe for jzmq (java bindings for zeromq); configure fails because it doesn't find java include files Dec 13 16:57:07 EXTRA_OECONF seems unuseful since configure recognize only a few options not related to java Dec 13 16:58:13 maybe I have to change configure itself? anyway, it is generated automatically when I run bitbake, so I don't know how Dec 13 16:58:34 you could overwrite the do_configure task Dec 13 16:58:50 (I guess) Dec 13 16:59:54 I tried, with a "sed" followed by the "./configure", but it seems to execute the standard configure, as if no sed has been executed Dec 13 17:01:07 is oe-core prefered over oe-classic these days? Dec 13 17:01:33 awozniak: yes Dec 13 17:01:38 for new projects Dec 13 17:02:00 I'm not familiar at all with oe-core. Is there a guide to switching over? Dec 13 17:03:17 The root of my problem ,really is ERROR: 'org.openembedded.dev/recipes/iproute2/iproute2_2.6.38.bb' failed, and it looks like there might be a fix in oe-core Dec 13 17:04:43 awozniak: I started with these two wiki pages: http://www.openembedded.org/wiki/OpenEmbedded-Core and https://wiki.yoctoproject.org/wiki/OpenEmbedded-Core Dec 13 17:04:59 kenws: thx Dec 13 17:06:17 If you're just looking for some recipes provided by oe-core you may use the git web interface Dec 13 17:06:21 http://git.openembedded.org/openembedded-core/tree/meta/recipes-connectivity/iproute2 Dec 13 17:07:32 Can I safely move those recipes into my oe-classic tree to get around this problem? Or are there deeper changes? Dec 13 17:08:13 I think that depends on the recipe. In general I would not do that : ) Dec 13 17:08:20 hehehe, ok thanks Dec 13 18:21:30 ericben, hi Dec 13 18:28:35 GNUtoo: did your compiler problems get solved Dec 13 18:29:05 khem, yes it did by passing that flag: Dec 13 18:29:21 -mno-unaligned-access Dec 13 18:29:27 like you advised me to do Dec 13 18:29:45 altough I was just pointed to that: Dec 13 18:29:48 http://comments.gmane.org/gmane.linux.ports.arm.kernel/117863 Dec 13 18:31:02 khem, thanks a lot Dec 13 18:48:36 Hm, now the do_rootfs task of core-image-sato fails with: http://paste.ubuntu.com/769256/ Dec 13 18:48:51 (I'm using an external toolchain) Dec 13 18:48:52 Did anyone see such an error before? Dec 13 18:49:28 task-core-apps-x11-pimlico doesn't seem to depend on external-csl-toolchain-gconv: http://paste.ubuntu.com/769257 Dec 13 18:51:43 Oh, I looked at the -dev package but the base package also doesn't depend on external-csl-toolchain-gconv: http://paste.ubuntu.com/769264/ Dec 13 19:01:49 strange Dec 13 19:36:45 *G Dec 13 19:36:50 oops Dec 13 19:39:28 mario-goulart, hi Dec 13 19:39:32 the sources directory looks really different in oe-core from oe-classic. Is there a migration strategy? I don't want to have to wait forever for downloads I've already got. Dec 13 19:40:08 Hi GNUtoo :-) Dec 13 19:41:07 awozniak: what happens if you point DL_DIR to the old downloads Dec 13 19:41:27 awozniak: what do you mean? the downloaded tarballs, etc are reused just fine Dec 13 19:41:43 the .md5 stamp was changed to .done, etc, but that won't cause it to redownload Dec 13 19:43:27 kergoth: I'm used to symlinking to an existing sources directory, with tarballs and md5sum files, but the environment I'm looking at now has a sources directory with a whole bunch of nested subdirectories. I'm following instructions at http://www.angstrom-distribution.org/building-angstrom Dec 13 19:43:46 the only nested subdirs I know of are git/svn/etc for the checkouts. Dec 13 19:44:35 ok, looks like my old "sources" is now "sources/downloads" Dec 13 19:44:48 that'd be an angstrom specific layout thing Dec 13 19:45:14 there is an angstrom channel for their specifics, if you have further questions about that Dec 13 19:45:25 kergoth: thx Dec 13 19:46:09 8np Dec 13 19:48:40 kergoth: were you able to reproduce the mirror issue? Dec 13 19:49:18 do someone has recipes for firefox and/or chromium for oe-core/meta-oe? Dec 13 20:14:31 re Dec 13 22:13:24 is there documentation on amend.bbclass somewhere? Dec 13 22:46:01 why is telnetd in busybox disabled ? Dec 13 22:46:24 just not needed in default config Dec 13 22:46:40 you can have your own .config where you enable it Dec 13 22:48:11 i have already done this, but now i have to branch the modified defconfig Dec 13 22:49:14 i am working on an ubifs based liveramdisk image and dropbear allways creates new keys. so i want use telnet Dec 13 23:05:38 pwgen: what you should do ideally is do that via a bbappend in your own customisation layer Dec 13 23:08:26 that will avoid branching Dec 13 23:08:52 it has been suggested we might want to move to a config fragment system for busybox like the kernel configuration in linux-yocto Dec 13 23:09:44 : its my own layer, with an epia-m MACHINE target , so i can add a recipe with and alternate busybox Dec 13 23:11:05 that and eglibc... Dec 13 23:11:14 (latest eglibc has kconfig support now as well) Dec 13 23:30:40 pwgen: bbappends are better for maintenance long-term than just duplicating the recipe Dec 13 23:31:10 Please Helphttp://pastebin.com/uH36LevP Dec 13 23:32:52 kashyap1: looks like a broken file again, just do bitbake -c cleanall gconf-dbus Dec 13 23:38:53 BlueLightning: ERROR: Task do_cleanall does not exist for target gconf-dbus,,,,,,,,,,,,,,,,,,i tried only clean which worked but the error still exist Dec 13 23:42:19 kashyap1: -c cleansstate (not -c do_cleanall) Dec 13 23:42:59 ah or even -c cleanall (but still without do_) Dec 13 23:45:08 jama ERROR: Task do_cleansstate does not exist for target gconf-dbus Dec 13 23:46:08 hmm show the exact command line you have executed Dec 13 23:47:33 bitbake -c cleansstate gconf-dbus Dec 13 23:51:31 ah sorry I didn't notice how ancient version you have.. Dec 13 23:52:10 what ancient version jama Dec 13 23:52:29 JaMa: cleansstate will not work, that does not delete downloaded source Dec 13 23:52:37 which is what we want to do in this case Dec 13 23:53:19 * bluelightning -> bed Dec 13 23:56:19 Friends, the file which was being downloaded was corrupt Dec 13 23:56:33 i have downloaded the correct tar ball Dec 13 23:56:37 and it worked **** ENDING LOGGING AT Wed Dec 14 02:59:57 2011