**** BEGIN LOGGING AT Wed Feb 13 02:59:58 2013 Feb 13 08:01:55 good morning Feb 13 08:19:35 morning Feb 13 09:21:46 morning all Feb 13 09:28:46 hi bluelightning, all Feb 13 09:29:18 hi all Feb 13 09:32:39 morning mckoan, hrw Feb 13 09:34:27 | built in linker script:11 cannot move location counter backwards (from 000000000040104e to 0000000000401000) Feb 13 09:34:41 I hate when binutils does that Feb 13 11:25:21 File "/usr/lib/python2.7/multiprocessing/synchronize.py", line 147, in __init__ Feb 13 11:25:21 SemLock.__init__(self, SEMAPHORE, 1, 1) Feb 13 11:25:21 OSError: [Errno 13] Permission denied Feb 13 11:25:28 can someone remind me solution? Feb 13 11:27:41 eh. /dev/shm ;D Feb 13 11:47:27 I hate build failures which are 12.04 only... Feb 13 12:37:29 is it possible to add md5 summing for generated images ? Feb 13 14:26:50 * mwester checks his email a second time, and notes that the ML must be dead again... :) Feb 13 14:28:45 ya, my mail shows the same thing Feb 13 14:28:59 oe-core, it's been nearly 12 hours.. oe-dev about 16 Feb 13 14:32:16 rob_w: have you solved your doubt about -native and -target binaries? Feb 13 14:33:00 florian: hey, anything amiss with the mailing list server? Feb 13 14:35:37 bluelightning: maybe... reatarted mailman, let's see if this fixes it Feb 13 14:35:40 oh, that's why I'm getting so much work done today! ;-) Feb 13 15:01:45 ant_work, no Feb 13 15:02:00 ant_work, can you enlight me ? did you understand what i meant ? Feb 13 15:03:11 rob_w: partially Feb 13 15:03:33 is it a -cross recipe? Feb 13 15:03:48 check this https://code.google.com/p/ada-util/source/browse/trunk/support/utilgen.c?r=688 Feb 13 15:03:57 we are trying to add ada support Feb 13 15:04:12 with ada usingt this above utilgen tool to determine arch flags Feb 13 15:04:26 rob_w: are you aware of https://github.com/Lucretia/meta-ada ? Feb 13 15:04:41 bluelightning, iam working with Luke on this Feb 13 15:04:48 he is my partner here Feb 13 15:04:50 rob_w: ah I see Feb 13 15:04:53 :-) Feb 13 15:05:04 sorry for the interruption, carry on :) Feb 13 15:05:11 so right now , wrongly we crosscompile utilgen.c Feb 13 15:05:26 and excute it which obviously fails Feb 13 15:05:39 so i am wondering what the best practice would be Feb 13 15:06:00 would we need crosscompile it and execute inside a qemu ? Feb 13 15:06:22 if it can't be built to work in a "cross" manner then yes Feb 13 15:06:29 depends on what it actually does Feb 13 15:06:46 it generates a ada header where the flags are determined Feb 13 15:07:47 is there any other exiting recipe/package which does such thing .. crosscompile a helper code and execute it inside a target qemu env ? Feb 13 15:08:13 the recently landed fontconfig class does that I believe Feb 13 15:08:22 we used to do it for libc locales Feb 13 15:08:39 could you link me such thing pls .. Feb 13 15:08:52 as we are not ony latest but poky denzil Feb 13 15:09:23 rob_w: http://cgit.openembedded.org/openembedded-core/tree/meta/classes/fontcache.bbclass Feb 13 15:09:32 sorry, I meant fontcache not fontconfig Feb 13 15:10:15 ${@qemu_run_binary() Feb 13 15:10:16 nice Feb 13 15:10:18 http://cgit.openembedded.org/openembedded-core/tree/meta/classes/qemu.bbclass Feb 13 15:10:21 yep Feb 13 15:11:09 problem with qemu_run_binary though, it only runs on targets qemu suppors Feb 13 15:11:37 which is arm for me .. so thats fine i guess Feb 13 15:12:15 qemu doesn't support all of the arm variants, like the coretex a9 Feb 13 15:13:30 another approach would be to alter the header generation so that it can be run in a cross-build situation Feb 13 15:13:57 i.e. instead of probing the current host, pickup the target parameters from elsewhere Feb 13 15:14:06 (assuming that's what it's doing) Feb 13 15:14:07 ya.. I know with the configure site stuff, we used the compiler preprocessor to generate certain code and then pulled out the values from that.. Feb 13 15:14:16 way nicer for cross compiling then trying to run a binary.. Feb 13 15:14:31 (I don't know if you can do this with the ada compiler, but I suspect it's possible) Feb 13 17:45:24 is there anyway to tell a bitbake (from a recipe) that I always want to build the recipe from scratch? Feb 13 17:45:31 ignore the states cache Feb 13 17:56:06 hi hrw Feb 13 18:24:22 GNUtoo-m4a785t-m: what's up? I will be afk most of evening Feb 13 18:26:00 hrw, hi, I wanted to ask you some questions about the ARM chromebook Feb 13 18:26:10 like how did the bootchain work Feb 13 18:45:10 scoutcamper|away: thanks dude Feb 13 20:45:13 re Feb 13 20:45:28 GNUtoo-m4a785t-m: samsung preboot, (uboot, vboot), kernel Feb 13 20:46:30 GNUtoo-m4a785t-m: preboot we skip, vboot is glued to u-boot and checks are kernels signed, devmode enabled etc Feb 13 20:47:56 GNUtoo-m4a785t-m: possible changes are preboot, uboot/vboot, uboot, kernel - so only second uboot needs to be signed and kernel can be in rootfs Feb 13 20:48:18 GNUtoo-m4a785t-m: other change is to replace uboot+vboot with normal uboot but can brick Feb 13 20:48:41 ok Feb 13 20:49:06 thanks a lot Feb 13 20:49:10 you welcome Feb 13 20:49:12 GNUtoo-m4a785t-m: got one? Feb 13 20:51:25 no, I was just looking for infos on it Feb 13 20:51:35 I got even more infos Feb 13 20:51:55 I don't know what vboot is but I think that you can replace what's in the flash Feb 13 20:52:24 preboot == BL1 is signed Feb 13 20:52:29 yep Feb 13 20:52:42 but after that you have uboot and you must disassemble the laptop to reflash it Feb 13 20:52:46 you can replace uboot/vboot with normal uboot Feb 13 20:52:47 no Feb 13 20:52:57 just need to take 5th screw from case Feb 13 20:53:02 ah ok Feb 13 20:53:16 it is flash write protect screw Feb 13 20:53:16 I thought that there was 2 ways: Feb 13 20:53:21 ah ok Feb 13 20:53:24 it depends on model Feb 13 20:53:37 ok, I was told that (to remove #WP) : Feb 13 20:53:50 The WP pin makes a connection with the D-panel (the bottom of the case), so just removing the bottom panel will allow the flash to be modified. Feb 13 20:58:52 GNUtoo-m4a785t-m: in snow it is just one screw Feb 13 20:59:09 GNUtoo-m4a785t-m: it was publically available but page got rewritten few times since then Feb 13 20:59:41 ok Feb 13 21:01:46 thanks a lot Feb 13 21:03:31 you welcome Feb 13 21:39:55 hello all Feb 13 22:05:42 What is the license for recipes that have no license like task-base.bb ? Feb 13 22:05:55 This is in an old OE-dev tree. Feb 13 22:06:08 I have to make a list for mgmt. Feb 13 22:06:12 ? Feb 13 22:06:23 recipes have MIT normaly Feb 13 22:06:34 software themselv have others Feb 13 22:06:44 that's what I mean Feb 13 22:07:01 task-base.bb does not resolv in any package Feb 13 22:07:02 I guess I'll say that any sort of procedural OE recipe is MIT Feb 13 22:07:06 just build an image Feb 13 22:07:19 and look in deploy dir for the license dir Feb 13 22:07:37 In an old OE-dev tree, there is recipes/tasks/task-base.bb Feb 13 22:07:51 yes and? Feb 13 22:08:08 So, can I assume that anything in the OE tree unmarked is MIT? Feb 13 22:08:15 ???? Feb 13 22:08:29 again .bb files are licenseds under MIT Feb 13 22:08:38 ok, thanks Feb 13 22:09:02 the license in the .bb files are the license whicht the resulting packages become Feb 13 22:09:15 not the license of the bb file Feb 13 22:10:22 I was just trying to process an okpg info directory of packages in my image and dump out a list of packages and their licenses Feb 13 22:10:57 some of the procedural packages in OE have no license defined in the .bb file. I didn't know that they were MIT. Feb 13 22:11:08 thanks for the info Feb 13 22:11:08 sr105: OE classic may lack license information for some recipes. Feb 13 22:11:19 s/for/in/ Feb 13 22:11:26 I've been tracking down the "unknown"s Feb 13 22:11:59 We're migrating to the newer OE, but we're shipping classic, so I have to make this list. Feb 13 22:12:15 sure Feb 13 22:12:43 no Feb 13 22:12:45 no no Feb 13 22:13:01 again for editing .bb files you have to do it under MIT license Feb 13 22:13:19 when there is no license inside the .bb the license is unknow or was not put in Feb 13 22:13:33 you have to examine the sources for the right license Feb 13 22:14:21 hm wonder if ross will talk about mali and tegra too Feb 13 22:20:54 For the unknowns, I looked up the license in the source. I only asked here, because I didn't know that the OE-specific task and image recipes were MIT Feb 13 22:21:04 Thanks for your help. Feb 13 22:24:42 no prob **** ENDING LOGGING AT Thu Feb 14 02:59:59 2013