**** BEGIN LOGGING AT Tue Jun 07 02:59:57 2011 Jun 07 07:39:08 khem, RP__: please keep me informed about status of porting oe-core to BSD Jun 07 08:33:13 bluelightning: amazing flood of patches, thx Jun 07 08:35:46 ant_work: no problem :) glad to get things tidied up Jun 07 11:05:59 RP__, regarding http://bugzilla.pokylinux.org/show_bug.cgi?id=1139 , you can just tell me here if you have some things I should try for you to get a better picture of it Jun 07 11:59:27 mlip: What happens if you remove recipes-kernel/linux/files/configs/ ? Jun 07 11:59:51 (from your local layer) Jun 07 12:01:35 RP__, I had recipes-kernel/linux/linux-omap/configs/ , but it fails in both cases (with, and without this dir) Jun 07 12:02:08 RP__, also failes with files/configs/ in my local overlay Jun 07 17:26:26 zeddii, pinf Jun 07 17:26:28 ping even Jun 07 17:26:56 zeddii, any particular reason the beagleboard-non_hardware.cfg has things like CPU_IDEL and DEBUG config settings in it? Jun 07 17:27:11 zeddii, seems like those should be part of the base config Jun 07 17:27:31 no reason that I recall. that's probably just where they landed in the mad shuffle. Jun 07 17:27:41 ok Jun 07 17:28:12 zeddii, I've got the patches applied and am now looking into meta Jun 07 17:28:27 zeddii, I can do this config iterations outside of bitbake right? Jun 07 17:28:37 zeddii, I think we've had this discussion Jun 07 17:28:53 but I don't think I ever saw a "use these commands" Jun 07 17:29:04 I sent something. let me mine my sentbox Jun 07 17:30:19 I have a blurb from you discussing the general approach (whittling away at the override warnings, etc) Jun 07 17:30:27 I'm just not sure how to do that outside of the bitbake env Jun 07 17:30:33 and bitbake would really slow this down Jun 07 17:36:16 * zeddii got NMId. looking now. Jun 07 17:36:30 * dvhart reading "configme" Jun 07 17:36:40 not sure how to apply an external defconfig Jun 07 17:36:51 you don't with configme. Jun 07 17:37:49 would I just add it to the existing scc and then it will warn when the meta-ti defconfig conflicts with the existing base config? Jun 07 17:37:57 if you've already gone through one linux-yocto config cycle, just head into the source tree and run what you see in the yocto kernel class Jun 07 17:38:40 configme --reconfig --output ../ <$machine> Jun 07 17:39:06 and change the settings in the .cfg files in the meta branch, configme will re-process the ones that are referenced in the meta-series. Jun 07 17:40:03 so you would do this from a yocto build dir - rather than out of my linux-yocto-2.6.37.git tree? Jun 07 17:40:25 if you don't have a meta-series, configme won't have anything to operate on. Jun 07 17:41:01 sorry, that term is new to me and doesn't appear in the configme script Jun 07 17:41:10 what does "meta-series" mean ? Jun 07 17:42:00 it's the output of the scripts. it's a meta-series in the sense of a patch series, but it has more than just patches. Jun 07 17:42:22 patchme and configme process a meta-series generated by create/update. Jun 07 17:42:49 I see Jun 07 17:43:06 ok, working from within the yocto build dir makes sense for other reasons as well Jun 07 17:43:13 I'll try that Jun 07 17:43:24 aha. I did send the steps. I found at least a reference to it. I'll forward it your way. Jun 07 17:43:30 thanks Jun 07 18:02:30 zeddii, thanks: There were 566 kernel config options redefined during processing this BSP. Jun 07 18:02:37 time to start whittling.... Jun 07 18:30:30 RP__, I noticed you took some of the user/group addition patches, but not the bbclass file itself. Let me know if you have any issues I need to resolve with that. Jun 07 18:31:18 once the bbclass is in place, then we can update the couple of recipes that already try to use user/group adding Jun 07 20:38:34 Hey there. Really stupid question -- I've done the bitbake and it's finished building all of the packages for the poky-image-sato recipe Jun 07 20:38:42 ...now what? :-)( Jun 07 20:38:44 :-) Jun 07 20:39:05 my understanding is there should be an image somewhere that I can put on the SD card for my beagleboard but it's not immediately obvious where Jun 07 20:40:08 jefferai, look in build/tmp/deploy/images Jun 07 20:40:16 there you'll find the kernel and rootfs images Jun 07 20:41:17 ah, cool, thanks Jun 07 20:41:23 I didn't expect to find it in tmp Jun 07 20:41:31 as I figured when things are built they're not tmp anymore :-) Jun 07 20:42:02 are all of the generated ipks/rpms used in the distribution, or do I need to set up some sort of a package management server if I want to be able to dynamically add packages onto the target system? Jun 07 20:42:57 jefferai, you can set up a package repository if you'd like, or simply copy packages onto the target and install them Jun 07 20:43:16 each image type has its own list of packages that get installed. you can also modify that and rebuild the image Jun 07 20:48:13 Yep Jun 07 20:49:27 I'm mainly not sure out of the few thousand packages, some of which are rpm and some of which are ipk, which packages are in the image by default, and why some are rpm and some ipk Jun 07 20:51:37 The package type is determined by the PACKAGE_CLASSES variable as set in your local.conf. I believe by default it includes both package_rpm and package_ipk, which means both package types will get built for all target pacakges. Jun 07 20:52:13 The first one specified is the package management system on the target, if I'm not mistaken. Jun 07 20:52:31 zenlinux: right, it is set for both -- but there are about 1500 ipks and about 5000 rpms Jun 07 20:52:52 (perhaps due to splitting things into multiple RPMs but not multiple ipks?) Jun 07 20:53:41 jefferai, that's a pretty big disparity. I was going to assume that ipk packages were not generated for -native packages, but I don't think there are 3500 more -native recipes that you built. Jun 07 20:53:59 Take what I say with a small grain of salt, I'm not 100% certain Jun 07 20:55:43 sure Jun 07 20:55:59 although Jun 07 20:56:03 the ipks are all armv7a Jun 07 20:56:11 the rpms are aplit between all/armv7a/beagleboard Jun 07 20:56:15 (beagleboard is my actual target) Jun 07 21:08:54 zeddii, we seem to have several CONFIG_*=n in the base config Jun 07 21:09:12 zeddii, I presume those should all be converted to "# CONFIG_* is not set" ? **** ENDING LOGGING AT Wed Jun 08 02:59:56 2011