**** BEGIN LOGGING AT Fri Apr 27 03:00:00 2012 Apr 27 07:02:04 Hi. I'm looking at making a hardfp build, and I want it to be compatible with concensus from https://wiki.linaro.org/OfficeofCTO/HardFloat/LinkerPathCallApr2012. According to the mailinglist, there is some proposition to change to this after yocto 1.2 is released. Any ideas when this release should be, and thereby when we can expect to have this fixed? Apr 27 08:04:31 Is there a way to list all licenses (GPL,BSD,...) used in an OpenEmbedded distribution? Apr 27 08:10:17 I think there is some rudimentary functionality. Yocto is working on this SPDX thing. Apr 27 08:25:05 good morning Apr 27 08:31:42 good morning Apr 27 08:42:31 Hi. I have tried to apply the following two patches to get renamed ld-linux.so.3 to ld-linux-armhf.so.3, so the powervr driver can link to it. Problem is, with these two patches I get a kernel panic saying: Kernel panic - not suncing: No init found. Try passing init= option to kernel. See Linux Documentation/init.txt for guidance. Apr 27 08:42:46 the patches are in: http://pastebin.com/VNNAzksg, and are for both gcc and eglibc Apr 27 08:43:28 there is also another ont to eglibc, which I forgot: http://pastebin.com/dzVbxNWx Apr 27 08:43:42 any ideas why this panics? Apr 27 08:45:52 matiert: try remove the init= option from your kernel commandline Apr 27 09:00:58 ericbutters: ok, one more question while I'm at it. Should it still be init when I'm using systemd? Apr 27 09:05:23 matiert: sorry can not say that.. Apr 27 09:06:06 ok Apr 27 09:06:08 thanks Apr 27 09:17:36 ericbutters: there are no init= option on my kernel commandline Apr 27 09:20:54 ok.. maybe you can try with init=linuxrc .. but i am not an expert.. ;) Apr 27 09:22:07 If I use OpenEmbedded to make a commercial product, what should I do wrt licensing? Apr 27 09:23:23 ericbutters: weird thing is, if I don't have those three patches, everything works fine, except for one binary blob which links again ld-linux-armhf.so.3 Apr 27 09:23:58 Should I add license notices for all contained packages or is that not necessary? Apr 27 09:24:24 kobe: that depends on the license conditions of the particular packages you are shipping. OE itself doesn't impose any requirements. Apr 27 09:31:59 ok, thanks. Apr 27 09:47:21 when I apply the following patches in openembedded for eglibc and gcc, everything seems to compile fine, but when booting the kernel, I get kernel panic saying: Kernel panic - not suncing: No init found. Try passing init= option to kernel. See Linux Documentation/init.txt for guidance. Apr 27 09:48:02 sorry, should post it somewhere else Apr 27 10:06:56 martiert: you mention PowerVR and ARMhf so I assume your looking at an OMAP board? If you are, #Beagle may be a good spot to ask the question and link to the patches that are giving you grief. Apr 27 10:49:18 ok, thanks Apr 27 10:59:25 <_tasslehoff_> I've installed xserver-nodm-init and some more x-stuff, and it seems that X starts fine. But... how can I test it? My goal is to run the OGRE demos, but that's a bit heavy to start with :) Apr 27 11:07:59 khem: Hi, I've got the Qt4.8 based qt4e-demo-image running with GCC 4.7. I had to apply two tiny patches to Qt: http://lists.linuxtogo.org/pipermail/openembedded-core/2012-April/021550.html Apr 27 11:08:38 * kenws runs to the meeting he's too late for : ) Apr 27 11:15:29 * _tasslehoff_ has working xeyes Apr 27 11:41:16 morning all Apr 27 12:00:53 hi pb_, all Apr 27 13:42:38 how do i configure x11 to autologin? i use x11-image and i do not want the loginscreen.. Apr 27 13:43:01 so could i say in recipes/images/x11-image.bb ? Apr 27 14:50:47 hi. Are there tasks/recipes for a self-hosted micro image somewhere? Apr 27 14:54:09 blindvt: you could try OE-Core directly with TCLIBC=uclibc Apr 27 14:54:14 qemumips says: Missing or unbuildable dependency chain was: ['self-hosted-image', 'syslinux'] (?!) Apr 27 14:54:19 and there are self hosted images there Apr 27 14:54:44 oh that must be some dependency hosed'ness Apr 27 14:54:53 qemux86-64 says: ['self-hosted-image', 'task-self-hosted', 'eglibc-utils'] (well TCLIBC=uClibc here; What do i need rpcgen for in this case?) Apr 27 14:55:16 kenws: good job with QT 4.8.1 yes I saw your patches lately Apr 27 14:55:18 khem, that's what iĆ am trying, yes. Doesn't work out too well Apr 27 14:55:35 blindvt: what are you trying to build imagewise Apr 27 14:56:11 I can buils core-image-sato successfully Apr 27 14:56:24 with latest uclibc/master and 0.9.33 Apr 27 14:56:26 khem, anyway. with my current incarnation of future, x86, x86-64, ppc, mips boot and run the micro image and arm somehow fails Apr 27 14:56:30 and it boots too Apr 27 14:56:48 console images work I know Apr 27 14:56:57 try to boot core-image-sato Apr 27 14:57:06 khem, does core-image-sato exercise threads a bit? Apr 27 14:57:06 thats where bug or bugs are Apr 27 14:57:13 a lot Apr 27 14:57:23 oki Apr 27 14:57:25 it used pango Apr 27 14:57:37 thats why I am not too keen on future branch Apr 27 14:57:52 1, the patches are not well coagulated Apr 27 14:57:59 some patches can be sqaushed Apr 27 14:58:16 and some changes are unexplained Apr 27 14:58:35 so I would think take the obvious patch by patch which dont introduce any regression Apr 27 14:58:52 I dont want to debug uclibc regressions Apr 27 14:58:57 it takes freaking too long Apr 27 14:59:07 and is wasteful effort to clean others shit Apr 27 14:59:16 khem, i have coalesced some of the patches to make bisect easier Apr 27 15:00:50 * blindvt__ nods khem Apr 27 15:01:35 let me do me a core-image-sato-dev Apr 27 15:02:01 yes Apr 27 15:02:26 boot for arm mips ppc x86 amd64 should work Apr 27 15:02:46 soon we will also have mips64 and sh4 Apr 27 15:03:15 my blind guess is that if there's something broken in my current set of patches then it might be the hidden locks. We don't have a current compiler (as in current trunk so i can fix stuff) in oe, do we? Apr 27 15:04:33 no. We don't even have 4.7, let alone 4.8 it seems. Oh well :/ Apr 27 15:07:38 khem, mips64 and sh4{,eb} would be nice indeed. Anyone planning m68k, sparc* and xtensa? Apr 27 15:08:47 coldfire should be fun, given the expected state of that toolchain ;) Apr 27 18:45:39 khem, who had the bright idea to add posix_spawn to librt?! Apr 27 22:15:03 khem ping? Apr 27 22:56:47 JaMa: hi, see this patch could solve outstanding UBI issue with subpages and nand write verify Apr 27 22:56:50 http://lists.infradead.org/pipermail/linux-mtd/2012-April/040737.html **** ENDING LOGGING AT Sat Apr 28 03:00:00 2012