**** BEGIN LOGGING AT Tue Feb 21 02:59:58 2012
Feb 21 04:29:08 <JonnyRo> What machine name should be used for the imx.53 platform?
Feb 21 04:43:27 <JonnyRo> Ah, i see its not supported yet
Feb 21 07:51:11 <pb_> morning all
Feb 21 08:27:09 <woglinde> snow snow snow again
Feb 21 08:28:11 <JaMa> ski ski ski again
Feb 21 08:29:22 <woglinde> no mountains here
Feb 21 08:29:59 * JaMa has to go 150km, but it's worth it :)
Feb 21 08:30:24 <woglinde> "riesengebirge" ?
Feb 21 08:30:28 <JaMa> well 150 to nearest and quite small skicentre
Feb 21 08:30:51 <JaMa> yes or Ore Mountains
Feb 21 08:32:01 <Noor> hi all
Feb 21 08:32:04 <JaMa> then you should buy kite to power your skies :)
Feb 21 08:32:16 <woglinde> hi noor
Feb 21 08:32:26 <Noor> is there a way of ignoring LIC_FILES_CHKSUM in oe-core
Feb 21 08:32:35 <JaMa> afaik no
Feb 21 08:32:36 <Noor> hi woglinde 
Feb 21 08:32:53 <woglinde> think why you cannt ignore it
Feb 21 08:33:12 <woglinde> and it is not hard to add it
Feb 21 08:33:36 <Noor> I have a recipe which only have prebuild binaries ... it doenot have any licensing file .... how can I make it work in oe-core
Feb 21 08:33:51 <Noor> doesnot
Feb 21 08:34:26 <JaMa> add LICENSE file to SRC_URI
Feb 21 08:34:41 <JaMa> even if it's just local file in metadata
Feb 21 08:34:42 <woglinde> are you talking about oracle arm java jvm?
Feb 21 08:34:52 <woglinde> *just a wild guess*
Feb 21 08:35:05 <JaMa> and use ${WORKDIR}/LICENSE in LIC_FILES_CHKSUM
Feb 21 08:35:16 <Noor> woglinde: u asking from me ..... no I got that recipe from somewhere else
Feb 21 08:35:32 <Noor> JaMa: I need to create my own then
Feb 21 08:35:39 <woglinde> JaMa hm isnt there one for binary in core?
Feb 21 08:35:43 <Noor> and add it in SRC_URI
Feb 21 08:36:07 <Noor> or it can be an empty one?
Feb 21 08:36:13 * Noor checks
Feb 21 08:36:14 <woglinde> but yes its maybee better to make your own
Feb 21 08:36:30 <woglinde> I would not ship an empty license file
Feb 21 08:36:30 <JaMa> woglinde: there are all common licenses, but I would rather copy whatever license agreement you had to agree on web or something where you found what to put in LICENSE field
Feb 21 08:36:40 <woglinde> jama yes
Feb 21 08:37:31 <Noor> thanks guys got the idea that what I need to do
Feb 21 08:38:31 <JaMa> there is also CLOSED license if it does apply
Feb 21 08:39:13 <kenws> Noor: Not sure if it helps but the recipe for the external binary toolchain is using file://${COREBASE}/LICENSE and file://${COREBASE}/meta/COPYING.MIT
Feb 21 08:39:20 <kenws> http://git.openembedded.org/openembedded-core/tree/meta/recipes-core/meta/external-csl-toolchain.bb
Feb 21 08:39:43 <JaMa> then it doesn't check LIC_FILES_CHKSUM, but having only binaries in SRC_URI does not always imply CLOSED license
Feb 21 08:39:48 <woglinde> kenws no licese doesnt mean ots MIT
Feb 21 08:39:50 <woglinde> its
Feb 21 08:40:12 <JaMa> heh I wish it would :)
Feb 21 08:41:27 <kenws> well, this one explicitly specifies LICENSE = "MIT"
Feb 21 08:41:56 <JaMa> yes because it's just for metadata
Feb 21 08:41:59 <kenws> it just came to my mind as it also package binary files
Feb 21 08:42:07 <JaMa> like task* recipes
Feb 21 08:42:31 <woglinde> kenws o.O
Feb 21 08:42:44 <woglinde> that doesnt mean it has the same license
Feb 21 08:43:01 <kenws> JaMa: hm, "just for metadata" - could you elaborate? I'm afraid I don't get it.
Feb 21 08:43:21 <woglinde> meta info in the resulting package
Feb 21 08:43:28 <woglinde> like rpm or deb or ipk
Feb 21 08:43:30 <kenws> ah
Feb 21 08:43:34 <kenws> got it, thx
Feb 21 08:44:02 <woglinde> kenws btw. the recipes inside oe are all MIT
Feb 21 08:44:02 <JaMa> kenws: I mean that this license does apply to external-csl-toolchain.bb recipe etc. not to the binaries which are used thanks to this recipe
Feb 21 08:44:59 <kenws> JaMa: yes, it's for the recipe (not the code sourcery toolchain itself)
Feb 21 08:45:05 <kenws> woglinde: ok
Feb 21 10:09:39 <mertsas> I wan't the shared glib-2.0 libraries to be added to my image. How can I do that?
Feb 21 10:09:53 <mckoan> good morning
Feb 21 10:10:47 <woglinde_> hi mckoan
Feb 21 10:11:01 <mckoan> woglinde_: hi
Feb 21 10:11:04 <bluelightning> hi all
Feb 21 10:11:09 <woglinde_> hi bluelightning
Feb 21 10:11:10 <mertsas> morning mckoan
Feb 21 10:23:26 <woglinde> hi mickeyl
Feb 21 10:45:46 <mertsas> how do I get x11 on my image? I tried adding DEPENDS += "x11-common" RDEPENDS += "x11-common", but I don't see any startx or anything in the resulting image
Feb 21 10:50:11 <JaMa> look at x11-common recipe what it provides first
Feb 21 10:53:54 <Noor> I have noticed that QA test function has been changed
Feb 21 10:54:30 <Noor> previously INSAVE-SKIP_$[PN] = "true" ignores all the tests
Feb 21 10:54:37 <Noor> but this is not the came now
Feb 21 10:55:03 <Noor> you need to give explicit test which you want to ignore .... how can I ignore all the tests?
Feb 21 10:55:21 <pb_> just list them all?  there aren't very many, after all.
Feb 21 10:55:22 <woglinde> damn I dont understand the oe git setup
Feb 21 10:55:25 <pb_> why do you want to ignore them all though?
Feb 21 10:55:34 <woglinde> how do I push a new personal branch now?
Feb 21 10:55:47 <pb_> same as always, isn't it?
Feb 21 10:56:06 <woglinde> push origin foo/bar?
Feb 21 10:56:18 <pb_> right
Feb 21 10:56:24 <Noor> pb_ just comparing how can i do what I was doing previously
Feb 21 10:56:58 <woglinde> dont works here hm
Feb 21 10:57:00 <Noor> So setting it to true will not help
Feb 21 10:57:58 <icanicant> How is auto-loading of kernel modules supposed to be implemented in Core? I can't find any recipes/classes that put anything in /etc/modules-load.d, so it seems that autoload modules should go in /etc/modutils and modutils-initscripts / modutils.sh is still required in the systemd world.
Feb 21 10:58:54 <woglinde> pb -> http://pastebin.com/Z2XnytBd
Feb 21 11:00:06 <woglinde> is my setup broken?
Feb 21 11:02:23 <pb_> well, that sounds rather like the branch you're trying to push just doesn't exist in your local repo.
Feb 21 11:02:45 <pb_> what does "git branch -a" say?
Feb 21 11:04:17 <woglinde> okay but I thought it create it when pushing
Feb 21 11:04:44 <pb_> it creates it on the remote, but obviously there needs to be something to push first.
Feb 21 11:04:53 <pb_> are you trying to just create an empty branch?
Feb 21 11:05:12 <woglinde> no
Feb 21 11:05:28 <woglinde> I have a local branch k3
Feb 21 11:05:33 <woglinde> and want to push it
Feb 21 11:05:41 <woglinde> to woglinde/foo
Feb 21 11:05:50 <woglinde> where woglinde/foo do not yet exist
Feb 21 11:06:17 <pb_> ah.  so you need to tell it to push k3 then.
Feb 21 11:06:38 <pb_> git push -n origin k3:woglinde/kaeilos-2011
Feb 21 11:06:44 <woglinde> args
Feb 21 11:06:48 <woglinde> the : again
Feb 21 11:06:55 <woglinde> okay thanks
Feb 21 11:06:55 <pb_> at the moment you're just asking it to push "woglinde/kaeilos-2011", which is clearly no good if you don't have a branch by that name
Feb 21 11:07:04 <woglinde> yeah understood
Feb 21 11:10:30 <woglinde> pb is deleting of the user branches now allowed?
Feb 21 11:10:47 <pb_> don't think so
Feb 21 11:18:47 <icanicant> the "module_autoload_name" directive puts modules in "/etc/modutils" but I don't see systemd doing anything with this. is module_autoload just not working properly yet with systemd? it looks like modutils-initscripts/modutils.sh should not be used with systemd, but kernel modules don't load without it.
Feb 21 11:19:03 <woglinde> wow another git feature mastered
Feb 21 11:19:09 <woglinde> rebase --onto
Feb 21 11:19:26 <pb_> eleet
Feb 21 11:20:54 <woglinde> for cherry-picking  more patches at once
Feb 21 11:25:12 <mckoan> a great day to master git commands today :-D
Feb 21 11:25:56 <JaMa> mertsas: you probably want something like task-core-x11.bb
Feb 21 11:26:42 <woglinde> icanicant what modules you need to load which udev not can?
Feb 21 11:29:46 <icanicant> woglinde: in this case it's iscsi_tcp
Feb 21 11:33:42 <mckoan> woglinde: please let's use mckoan/kaeilos-2012
Feb 21 11:35:09 * mckoan -> lunch (bb in 1,5 h)
Feb 21 11:36:02 <woglinde> icanicant okay
Feb 21 11:36:22 <woglinde> icanicant than you have to look into the systemd documentation
Feb 21 11:37:06 <bluelightning> ideally we would have some way of having these autoloaded without caring which init system was in use...
Feb 21 11:41:55 <icanicant> it's not really a question about how to do it, it's a question about how to do it the "correct" way in OE Core.
Feb 21 11:42:22 <icanicant> i can make it work but what is the recommended way of doing it?
Feb 21 11:42:59 <icanicant> there is already the "autoload_module_name" directive which looks useful but it turns out that this doesn't work as expected, probably since systemd
Feb 21 11:50:26 <icanicant> bluelightning: yes i think the module_autoload_ code just needs updating it for the systemd world. i'll see if i can knock up a patch.
Feb 21 11:51:43 <JaMa> icanicant: when you're working on it, please look if there is possibility to force loading in specified order
Feb 21 11:52:31 <JaMa> icanicant: or at least load more modules from one module_autoload_ directive (ie usb controller + g_ether) in one module_autoload_
Feb 21 11:59:43 <icanicant> JaMa: sounds useful
Feb 21 12:00:50 <JaMa> yup I have usecase for it already :)
Feb 21 12:01:01 <JaMa> SHR root@gjama ~ $ cat /etc/modules-load.d/g_ether.conf
Feb 21 12:01:01 <JaMa> s3c2410_udc
Feb 21 12:01:01 <JaMa> g_ether
Feb 21 12:36:07 <jkd> hi, Im trying to build using oe-core, I've ran the oe-init-build-env script, and then I run bitbake, and do not get past parsing
Feb 21 12:36:26 <jkd> I only get "ERROR: Command execution failed:" 
Feb 21 12:36:34 <jkd> -v or -DDD do not give more info
Feb 21 12:36:42 <jkd> the cooker.log only includes the above line
Feb 21 12:36:50 <jkd> any ideas?
Feb 21 12:39:42 <Noor> om which ML we are sending patches for meta-oe?
Feb 21 12:40:18 <JaMa> openembedded-devel
Feb 21 12:41:27 <Noor> ooo yeah ...... why I missed that
Feb 21 12:41:52 <Noor> there is only one patch should I sent pull request or just one patch will also work
Feb 21 12:42:25 <Noor> looks like only patch will do the work
Feb 21 12:42:42 <Noor> cant see recent pull requests 
Feb 21 12:53:29 <JaMa> then look better
Feb 21 12:53:50 <JaMa> there are at least 2 from today
Feb 21 12:54:00 <JaMa> but single patch with right prefix will work too
Feb 21 13:07:50 <mckoan> re
Feb 21 14:00:17 <jkd> Is there anyway to get more info on why bitbake fails? first run after init oe-core it fails with only ERROR: Command execution failed: Exited with 1, -v and/or -DDD gives nothing more and cooker.log only gives same line
Feb 21 14:02:27 <woglinde> does python work?
Feb 21 14:04:48 <jkd> yes
Feb 21 14:26:33 <mckoan> ao2: ping
Feb 21 14:44:44 <bluelightning> ericben: ping
Feb 21 14:56:32 <ant_work> pb_: do you know the list of users with write access on the git.openembedded.org repo?
Feb 21 15:11:22 <ao2> mckoan, pong
Feb 21 16:13:48 <rob_w> would it be possible to upgrade a glibc based system into a eglibc one by just using opkg ?
Feb 21 16:14:24 <Jin^eLD> rob_w: yes
Feb 21 16:14:36 <Jin^eLD> we did that when we moved from OE classic to OE core
Feb 21 16:15:06 <Jin^eLD> we had to do some tuning of the packages though
Feb 21 16:15:15 <rob_w> so is eglibc some much better ? 
Feb 21 16:15:23 <Jin^eLD> which, shame on me, I still have not found time to go through and submit patches for those changes
Feb 21 16:15:52 <rob_w> ah never mind .. i iwll read myself into it 
Feb 21 16:15:53 <Jin^eLD> no idea :) but angstrom on OE coreused eglibc by default so it seemed like the way to go
Feb 21 17:38:34 <bluelightning> Crofton|work: hi, will you have a chance to test my qt patches this week?
Feb 21 17:38:50 <Crofton|work> trying to
Feb 21 17:39:10 <bluelightning> ok, that would be great, thanks
Feb 21 17:39:19 <Crofton|work> I have to get some guys setup to do local builds to support stuff in the field
Feb 21 18:15:01 <Crofton|work> anyone using fedora 16 with oe-clssic?
Feb 21 18:15:09 <Crofton|work> or oe-core?
Feb 21 18:16:01 <woglinde> I am not
Feb 21 18:16:13 <woglinde> I try to stay away from rpm
Feb 21 18:28:27 <pb_> ant_work: not offhand, but I can look it up
Feb 21 18:28:41 <pb_> going home now, but remind me later and I will investigate
Feb 21 18:29:12 <pb_> also, which particular repo are we talking about?
Feb 21 18:33:09 <woglinde> ah kergoth linked me
Feb 21 20:21:52 <florian> Which setting do I change to set the prefix of the toolchain? 
Feb 21 20:22:48 <florian> I would like to unify it a little bit...
Feb 21 20:46:56 <muriani> gaaaaah openmoko svn is down again
Feb 21 21:19:31 <reaperofsouls> florian: TARGET_PREFIX?
Feb 21 21:33:34 <JDuke128> hello , on Windows every window has ID named HWND , on Linux , how can i access other apps windows? they have something like HWND ? My real problem is , i've java app , i want to show some video on C++ side combining with java but java slows up , I want to bypass Java side and directly render some video on HWND on Windows but i could not found any solution on Linux , how can i fix problem
Feb 21 21:44:09 <florian> re
Feb 21 21:44:37 <florian> reaperofsouls: is this for the whole thing?
Feb 21 21:50:02 <reaperofsouls> florian: Where ever CC is used for and alike its ${TARGET_PREFIX}gcc, same for the various others. Normally it derived from ${TARGET_SYS}-
Feb 21 21:51:38 <florian> hmm true... I wonder what it will break if I change it :)
Feb 21 21:51:54 <reaperofsouls> Sorry thats not exactly correct by default it is HOST_PREFIX=${TARGET_PREFIX} which for target apps resolves right.
Feb 21 21:52:20 * florian wants toolchains for multiple targets all looking the same
Feb 21 21:52:38 <reaperofsouls> florian: We provide external toolchains, and it is what we set, though the stuff we are currently providing is more in line with classic atm, but it looks like it should carry in to core.
Feb 21 21:52:38 <florian> sich as arm-linux-gcc and mipsel-linux-gcc
Feb 21 21:53:11 <florian> true... it has to work for external ones
Feb 21 21:57:09 <reaperofsouls> Not sure if that would have the desired effect on the internally generated toolchains.
Feb 21 21:59:07 <reaperofsouls> A quick look at gcc-configure-common, I think the answer is probably though some one like khem would know.
Feb 21 22:15:53 <khem> reaperofsouls: no effect. It should work fine if you override it provided its a valid gcc triplet
Feb 21 22:16:07 <khem> since thats what we configure internal toolchain with
Feb 21 22:16:18 <khem> for --program-prefix
**** ENDING LOGGING AT Wed Feb 22 02:59:58 2012