**** BEGIN LOGGING AT Mon Jul 06 02:59:59 2015 Jul 06 06:18:19 howdy! is there a way to "depend" on a busybox configuration? e.g. a recipe that need some switch in a busybox call to be available Jul 06 06:25:35 not really Jul 06 06:25:40 the config isn't exported Jul 06 06:25:55 gah. Jul 06 06:26:01 the best you can do is try to read teh PACKAGE_CONFIG for busybox, but that will run into scoping problems Jul 06 06:26:18 option b is to have test for it in e.g. the postinst Jul 06 06:26:26 or depend on the full util Jul 06 06:26:50 .. which is both hard if the the util is there anyways, but a command line switch isnt Jul 06 06:28:03 i remember doing something similar awhile back, ended up using DISTRO_FEATURES with "busybox-" to enable said feature, which can also be used to detect if the feature is there Jul 06 06:28:47 nrossi: ok, but still this is invasive to the busybox recipe, right? Jul 06 06:29:48 LetoThe2nd: yep still required a bbappend for the busybox recipe Jul 06 06:30:04 hmkay Jul 06 06:30:57 sounds a bit clumsy if you don't mind the expression, but might work Jul 06 06:39:44 LetoThe2nd: The other way you could do it is with a PROVIDES in busybox which is an empty/virtual package Jul 06 06:47:13 nrossi: ah, thanks! Jul 06 07:33:58 good morning Jul 06 07:54:41 hi mckoan Jul 06 07:59:26 hi woglinde Jul 06 08:26:03 morning all Jul 06 08:32:24 hi bl Jul 06 08:35:07 hi woglinde Jul 06 08:43:36 hi bluelightning Jul 06 08:43:42 hi mckoan Jul 06 10:37:25 gm all Jul 06 10:38:38 I notice that when I change MACHINE after a completed (image) build, and I build the image for another MACHINE, many (all) *_setscene tasks are called. Why is this? Jul 06 11:16:07 like2wise, I will guess they are taking packages that are already in shared state and building the sysroot Jul 06 11:16:12 but I do not really know Jul 06 11:40:31 Crofton, that might be it: sysroot is machine dependent, not arch dependent Jul 06 11:43:47 wasnt there a bitbake target which would summarize all license files used in a image build ? Jul 06 11:46:46 rob_w: look in tmp/deploy/images/$MACHINE and look for the file named after the image and timestamp corresponding to the desired build Jul 06 11:48:24 LetoThe2nd: u mean the .manifest ? Jul 06 11:48:57 rob_w: don't know the exact filenames, just have a look into them Jul 06 11:56:41 like2wise, you see the same thing if you delete tmp, it repopulates tmp from sstate Jul 06 12:13:22 ok i got under tmp/deploy/licenses/my-image/ Jul 06 12:13:25 thats good . Jul 06 12:13:47 is there a bitabke target for a certain image , which would tar up all source codes used ? Jul 06 12:17:13 rob_w: see http://www.yoctoproject.org/docs/1.8/dev-manual/dev-manual.html#maintaining-open-source-license-compliance-during-your-products-lifecycle Jul 06 12:17:20 (basically, archiver.bbclass) Jul 06 12:18:07 is python allowed in conf files, a distro conf for example Jul 06 12:19:34 I'm trying to do something like def func () { return d.getVar('VAR', True) } VAR := ${@func} Jul 06 12:29:06 ok, I got it working with a 1 liner Jul 06 12:31:23 jackmitchell: obligatory to post it :) Jul 06 12:31:55 (nothing more frustrating than reading IRC logs that someone fixed your problem) Jul 06 12:34:24 like2wise: it's on the mailing list, but for completeness Jul 06 12:34:25 SDKPATH := "${@os.path.abspath(d.getVar('TOPDIR', True)+"/../sdk/"+d.getVar('SDK_VERSION', True)+"/"+d.getVar('SDK_ARCH', True))}" Jul 06 12:38:29 jackmitchell: thanks Jul 06 13:30:56 Does anyone know, what would cause my rootfs to boot in Read Only? Jul 06 13:39:42 realBigfoot: is the rootfs on a sd card? Jul 06 13:53:28 nrossi, yes when I make the image and boot it on a minnow Jul 06 13:54:43 realBigfoot: did you check to see if the kernel is giving a message regarding setting the block device to read-only mode? also those pesky lock switches on the edge of the card can be annoying :) Jul 06 13:56:43 nrossi, it builds successfully without any msg. I double checked the swtich and it is in the right position :( Jul 06 13:57:22 realBigfoot: got a spare card to make sure its not just the card playing dirty? Jul 06 13:57:43 haha nrossi, good call! I will check it out Jul 06 13:57:46 also check your /etc/fstab to make sure it has not got "ro" in the options for the rootfs mount Jul 06 13:59:13 nrossi, my fstab is-> "/dev/root / auto defaults 1 1" Jul 06 13:59:39 only my / is read only, I have tested here /tmp and it works Jul 06 14:00:20 realBigfoot: /tmp should be a tmpfs mount, (you can check $ cat /proc/mounts // $ mount) Jul 06 14:01:59 nrossi, alright my /dev/root is ro and my /tmp is a tmpfs but is has rw attributed Jul 06 14:02:13 nrossi, but why would it be generated like this ? Jul 06 14:02:23 realBigfoot, also can you see what args are passed to the kernel Jul 06 14:02:40 Crofton, how do I do this ? Jul 06 14:02:56 on mounts ? Jul 06 14:03:07 look at the boot messages Jul 06 14:03:17 pastebin the boot log Jul 06 14:03:31 Crofton, alright I will try it Jul 06 17:05:53 http://stackoverflow.com/questions/31226962/workaround-for-tar-1-27s-bug-in-ubuntu-14-04 Jul 06 17:08:15 Being from France, I wrote LICENCE_FLAGS_WHITELIST instead of LICENSE_FLAGS_WHITELIST Jul 06 18:26:57 jfk Jul 06 18:27:16 I'm about to submit this line as wic.bbclass Jul 06 18:27:27 DEPENDS_prepend = "parted-native mtools-native dosfstools-native " Jul 06 18:32:21 Crofton|work: I think we could go further than that - at OEDAM Martin and I were talking about a multiimage.bbclass that actually lets you construct the images through the build system, and I believe that Ed may be working on something like that Jul 06 18:32:52 sure, I can see adding something to run wic etc Jul 06 18:33:11 everytime I clean tmp, I get burnined running my script because I lose those bits Jul 06 18:33:56 Ed also needs to make sure he has a way to pass a thread count to xz for compressing, makes a big time difference Jul 06 18:34:35 ok... if there's not already a bug filed, it would be worth filing something so the issue is tracked Jul 06 18:34:39 I gotta go catch a train, bbl Jul 06 18:34:49 I sent it as review for some of his pactehs **** ENDING LOGGING AT Tue Jul 07 02:59:58 2015