**** BEGIN LOGGING AT Mon Mar 04 02:59:58 2013 Mar 04 03:10:52 mewyn: why? you can define an image using all discrete tools rather than busybox but I expect that BB will still be built but not used. Mar 04 08:55:58 good morning Mar 04 09:30:43 JaMa: wrt the -2 klibc recipes failing in world I'll retest for qemu with 3.8 headers Mar 04 09:31:23 ant_work: they fail every now and then (even before 3.8 headers) Mar 04 09:31:37 yes, that's really suspicious Mar 04 09:32:35 as for kexec-tools-klibc, first it fails do_configure for qemux86 then do_compile for qemux86-64 ?? Mar 04 09:32:57 some race? headers not staged? Mar 04 09:35:11 it has the very same .inc of oe-core. DEPENDS = "zlib xz" Mar 04 09:38:04 I'll try to update to 2.0.3 as well Mar 04 11:25:33 Hi oe Mar 04 11:27:16 As a complete newbie to oe, can someone advise where to start? Should I go with Yocto or stay with OpenEmbedded Core? Or are they the same? I'll admit I'm struggling a bit with the documentation Mar 04 11:28:53 My goal is to compile an image for an Arndale Board: ArndaleBoard.org Mar 04 11:29:14 I have a toolchain already Mar 04 11:38:09 bdholt1: oe-core is the core of Yocto, you can use both but probably at the moment Yocto has better documentation Mar 04 11:38:15 bdholt1: http://www.yoctoproject.org/docs/current/yocto-project-qs/yocto-project-qs.html Mar 04 11:38:34 bdholt1: Yocto has its IRC channel too Mar 04 11:42:02 thanks very much Mar 04 11:43:49 bdholt1: yw Mar 04 13:35:45 I want to modify ntpd.service, do I use a bbappend and if I append it SRC_URI does it replace the one that is defined in the ntpd recipe? Mar 04 13:36:16 OWayne: see FILESEXTRAPATH usage in other bbappends Mar 04 13:36:47 and that's enough, no need to append it to SRC_URI again Mar 04 13:37:34 ah right, so it will search my layer first when processing the recipe Mar 04 14:12:51 is it possible to mix native and non-native stuff in one recipe? Mar 04 14:13:21 I have a package here that generates native helpers in addition to its libraries. Mar 04 16:18:12 hmm, ellcc looks interesting Mar 04 17:49:51 hello folks Mar 04 17:50:20 I've requested a build machine from my university and they are now expecting me to handle the specs Mar 04 17:51:46 moin Mar 04 17:51:57 of course, more horsepower is always good but I need to say what I need. What's the typical configuration for relatively fast build system? Mar 04 17:53:00 I'm thinking of requesting 8-cores, 8GB ram, and 80GB disk. Mar 04 17:53:43 eren: depends on what builds do you want to do there Mar 04 17:54:00 eren: for my big targets and multiple build environments 80G seems really low Mar 04 17:54:11 but if it's SSD then I can understand Mar 04 17:54:25 JaMa: not a world build, I will just target arm and intel Mar 04 17:54:31 for now Mar 04 17:55:07 JaMa: I don't know the configuration of the host machine, but the server was bought 2 weeks ago. I guess the price was relatively high Mar 04 17:55:44 I would go for bigger disk if it's not ssd Mar 04 17:56:08 it's not, I believe. It was bought to meet the general needs of the uni Mar 04 17:56:17 JaMa: so, 8cores would be enough? Mar 04 17:56:41 I will say that it will be my minumum. Mar 04 17:56:43 sure if you don't need to sit before it and wait for image build Mar 04 17:56:56 I don't, definitely :) Mar 04 17:56:59 I'm running official builder with 2 virtual cores fine Mar 04 17:57:19 only when rebuilding from scratch it's really pain and takes like 2 weeks Mar 04 17:57:32 wow Mar 04 17:57:32 but incremental builds without OEBasicHash are still fine on that old machine Mar 04 17:57:49 I will probably build from scratch frequently Mar 04 17:58:44 JaMa: I hope they will allocate my optimal configuration, but I doubt that. I would end up with 4-cores with 4gb ram :) Mar 04 17:58:56 but my builds are for 6 MACHINEs and with pretty big feed Mar 04 17:59:07 ah Mar 04 17:59:34 building core-image is fine on any hardware Mar 04 17:59:52 except my laptop Mar 04 18:00:23 you wouldn't want to wait 4 hours just for core-image-minimal :) Mar 04 18:00:41 you can still run it before going to bed Mar 04 18:02:17 of course, but debug-fix-compile cycle takes lots of time Mar 04 20:00:52 hi all Mar 04 20:09:31 darknighte, ping **** ENDING LOGGING AT Tue Mar 05 02:59:58 2013