**** BEGIN LOGGING AT Mon Apr 16 02:59:59 2012 Apr 16 07:23:11 good morning Apr 16 08:41:49 Hi, I am trying to get going on oe-core (switch from classic). I am building core-image-minimal and getting a download error on linux-yocto-3.2.11 for a couple of days now. Url is git://git.yoctoproject.org/linux-yocto-3.2 .... error log is http://pastie.org/3796957. Seems like the source is no longer available. Any help? Apr 16 08:54:25 guyvdb_: there is a proper channel #yocto ;-) Apr 16 09:05:48 mckoan i am doing an oe-core build. But I am aware the oe-core and yocto share common pieces. I will chat to the in yocto. Apr 16 09:05:59 the=them Apr 16 12:25:56 hi, why I am not seeing the ralink platform in make menuconfig ? Apr 16 12:31:34 ideas? Apr 16 12:36:17 <_tasslehoff_> What's wrong with this linux_3.0.bbappend http://pastebin.com/JAKdDUCQ? It says it can't find linux-3.0/mymachine/defconfig (which exists). Apr 16 13:18:24 <_tasslehoff_> How do I append/prepend to the FILESPATH =. set in linux_3.0.bb Apr 16 13:40:45 (bump) why I am not seeing the ralink platform in make menuconfig? Apr 16 13:52:03 chiui: wrong ARCH in environment? Apr 16 14:11:40 florian, mmh but why I am seeing other arches? I'm on amd64 yet I see at least x86, arm and avr32 targets Apr 16 14:26:07 <_tasslehoff_> Can I (in core and/or classic) compile an sdk on a 64 bit machine that people on a 32 bit machine can use? Apr 16 14:40:03 _tasslehoff_: that's called 'canadian cross-compiler' and AFAIK is present but broken in OE Apr 16 14:45:16 <_tasslehoff_> mckoan: ok. looks like it's pae-kernel on the new machine as well then Apr 16 17:56:02 Hi, I am trying to get started on oe-core. I am getting a download error when I try to build core-image-minimal. It is the kernel package (linux-yocto-3.2) that fails to download for a few days now from the yocto git repos. Is the meta recipe wrong in the oe-core repos? Ideas? Apr 16 18:05:46 guyvdb_: hmmm have you tried cloning it otherwise ? Apr 16 18:06:31 unless you have some weird firewall settings which blocks git port Apr 16 18:06:51 make sure if you can clone any git repos at all Apr 16 18:11:54 khem it is not a proxy issue I do not think. I can clone from the shell. I have been chatting in yocto and they think it is a Fetcher error. See log: http://pastie.org/3799496 Apr 16 18:13:32 fwiw I use bare oe-core too Apr 16 18:13:50 and I just did bitbake -ccleanall virtual/kernel;bitbake -cfetch virtual/kernel Apr 16 18:13:55 and it fetched it alright Apr 16 18:14:37 khem let me try again. Apr 16 18:24:41 khem it seems to be fetching correctly now. I think the bitbake -c cleanall sorted it. Thanks for the help Apr 16 19:52:20 I have some proposed update-alternatives.bbclass changes: Apr 16 19:52:22 http://git.yoctoproject.org/cgit/cgit.cgi/poky-contrib/log/?h=mhatle/upd-alt Apr 16 19:52:36 if anyone wants to review it before I send it to the list, I'd like that.. Apr 16 19:53:03 hmm.. upon thinking about it.. I think RECIPE_BLACKLIST is probably better then just BLACKLIST Apr 16 19:53:17 we already have a LICENSE_BLACKLIST (don't we?) Apr 16 19:53:26 oops wrong channel.. Apr 16 19:54:01 (the blacklist comments, I am looking for feedback on the update-alteratives changes) Apr 16 19:57:21 looks pretty sane to me (upd-alt) Apr 16 19:58:08 thanks.. it's a bit messier then I'd like.. but for the upcoming oe-core freeze it's as far as I want to push it.. Apr 16 19:58:23 after freeze, I want to fix update-alternatives to be more like useradd and allow per-split settings Apr 16 19:59:16 some days it really bugs me all the ways in which we work around the limitations of our format. some of the uses of flags to implement a ghetto dictionary make me sad Apr 16 19:59:24 heh Apr 16 20:01:24 to me the trick is figure out all of the permutations for adduser, update-alternatives and update-rc.d.. once you do that.. then the implementation becomes MUCH easier.. Apr 16 20:01:38 it's the manual package update-alternatives that are still annoying and painful.. (and likely uncessary) Apr 16 20:02:02 for something like RPM, we could even change the implementation to -not- run a shell script but do it using an alterantive approach that might be more correct for RPM (vs say deb or ipk) Apr 16 20:02:08 that is where I'd like to get to at least Apr 16 20:04:29 sounds like a good goal Apr 16 20:42:26 Is this a bug or not? guile-native does not build without guile installed.... Apr 16 20:42:41 that would be a bug, yes Apr 16 21:31:24 thats a lousy guile for cross compilation Apr 16 22:33:05 Umm I just got a weird message from git send-email, did anyone get my update-alternatives patches send to the oe-core list? Apr 16 22:37:38 * fray apologies in advance if I send two copies of this.. but it does look like the first one failed. Apr 16 22:41:54 seems to be working this time Apr 16 22:59:30 apparently our IT department has decided to punk me.. so if yous ee multiple copies of emails from me.. again.. sorry Apr 16 23:04:43 sadly standard IT email handling and the typical open source patch/mailing list workflow often don't mix :( Apr 16 23:07:07 arg.. its stuck.. again... sending 8/17... Apr 16 23:07:22 (and apparently the retry caused 4/17 to go out twice) :P Apr 16 23:07:29 * fray wants to go home, but needs to get this out Apr 16 23:07:38 stoooopid exchange.. Apr 16 23:15:22 well it decided to unclog itself.. later! **** ENDING LOGGING AT Tue Apr 17 02:59:59 2012