**** BEGIN LOGGING AT Thu Jul 10 02:59:58 2014 Jul 10 08:01:08 how does the recipe deal with ASSUME_PROVIDED? does it search in PATH or somewhere else? Jul 10 09:30:33 ouch.. "bitbake -c cleanall" removes the downloaded git repository for the package aswell ? Jul 10 09:32:10 I guess it does, all according to the doc :-D Jul 10 11:36:31 hi there. i hadn't made a build for some time... today i am facing this git config issue Jul 10 11:36:31 ERROR: Fetcher failure: Fetch command failed with exit code 128, output: Jul 10 11:36:31 error: Malformed value for push.default: simple Jul 10 11:36:31 error: Must be one of nothing, matching, tracking or current. Jul 10 11:36:52 is there anything that can be done for that (beyond just removing the line in my .gitconfig)? Jul 10 11:37:12 looks like a version mismatch Jul 10 11:37:47 yeah, that's a new option added in git 2.0 about the push for tracking branches Jul 10 11:38:13 with git 2.0 if you don't set it in ~/.gitconfig, you get a warning message everytime.. Jul 10 11:38:33 the problem is that you get a failure apparently with this option and older git.. Jul 10 11:38:35 I bet git-native in OE isn't 2.9 Jul 10 11:38:38 2.0* Jul 10 11:39:14 I had git-native in assume-provided and set push.default to simple Jul 10 11:39:27 right. should we upgrade git? Jul 10 11:39:28 that broke bitbake when extracting git URIs ;( Jul 10 11:39:39 that was with 1.4 or 1.5 Jul 10 11:39:48 I think it has been fixed Jul 10 11:45:07 koen: damned... git-native is in ASSUME_PROVIDED by *default*... the problem is that I was building in a old chroot... Jul 10 11:48:44 Building recipe: shr-theme-neo (1501/1885) Jul 10 12:26:13 hi morphis, i am trying to use meta-android layer, and noticed that it 'depends' on meta-oe since it's using FREESMARTPHONE_GIT var.. not sure how you want to handle or fix that.. but the README doesn't mention the dependency. Jul 10 12:39:04 morphis: also commit "android-tools: split out machine specific configuration" breaks the build of ext4.fastboot images, since android-tools-native-conf does not exist Jul 10 13:09:47 ndec: thanks for the pointer Jul 10 13:09:57 will try to fix it Jul 10 13:10:52 morphis: ok! in fact I was wondering if we could/should move the android image types stuff into oe-core... it's not uncommon to have a board that comes with fastboot. Jul 10 13:11:15 ndec: good idea Jul 10 13:11:30 instead of pulling complete meta-android in Jul 10 13:12:43 yes. indeed. Jul 10 13:49:21 morning Jul 10 14:06:50 Hey guys; is there a known issue building mozjs right now? Jul 10 14:13:34 morphis: so , i was able to create a ext4.fasboot image and flash it. i will try to move the necessary pieces into oe-core. then we can discuss on the list if that can be merged.. Jul 10 14:14:09 i think i will move only the rootfs image stuff, not the abootimg since this one is more difficult to manage and more 'custom' that ext4.fastboot Jul 10 14:59:06 ndec: I've added dependency in README Jul 10 14:59:20 ok Jul 10 15:49:39 Anyone have any comments? Can someone else try to bitbake mozjs using the lastest git? Jul 10 16:45:00 hi koen , do you remember why you changed git SRC_URI from kernel.org to http://git-core.googlecode.com? (that was in 2011 ;-) Jul 10 16:45:16 i was trying to update to 2.0, but it seems that they no longer update the release there Jul 10 16:45:28 so, i am tempted to switch back to k.org.. Jul 10 16:48:18 there was a period of time when kernel.org was down.. it might have been related Jul 10 16:51:12 ah. right. so given that the official git release announcement claim that tarball are on k.org (http://git.661346.n2.nabble.com/ANNOUNCE-Git-v2-0-1-td7614048.html), i guess it's fine to switch back? Jul 10 18:59:08 I recently updated our build system to Daisy, and upgraded bitbake to 1.22 Jul 10 19:00:05 however, I am getting errors running bitbake, seems like the sanity check is failing on check_app_exists since it can't import "which" Jul 10 19:00:14 from bb import data, which Jul 10 19:00:36 not sure why it's failing, anyone has seen something similar? Jul 10 19:01:07 the error: "ImportError: cannot import name which" Jul 10 19:03:06 you're trying to use a mismatched bitbake version, i'd say Jul 10 19:03:09 which was moved to bb.utils Jul 10 19:03:22 are you running the bitbake you think you are running? Jul 10 19:06:33 ah, that make sense, so looks like I am using a bitbake that might be expecting which in a different location Jul 10 19:06:55 let me make sure my paths are fine Jul 10 19:07:00 you're using metadata that's expecting an old bitbake :) Jul 10 19:56:37 two questions: Jul 10 19:56:50 1. any problem adding chrono and atomic to boost's recipe? Jul 10 19:57:00 2. what about adding log4cpp support? Jul 10 19:57:21 (I know there's log4cxx and log4cplus in there already, but since everyone needs their own special logger, apparently...) Jul 10 20:49:20 trondeau: Just FYI, I setup a bbappends locally to add atomic to BOOST_LIBS Jul 10 20:49:55 dgeiger: yeah, that's what I'm doing, too. It's easy enough. **** ENDING LOGGING AT Fri Jul 11 02:59:58 2014