**** BEGIN LOGGING AT Wed Oct 24 03:00:02 2012 Oct 24 13:53:57 bluelightning: hey Oct 24 13:54:04 hi ant_work Oct 24 13:54:48 oneday you'll have to ask Intel guys to learn about COMPATIBLE_MACHINE :) Oct 24 13:55:19 I started a discussion about linux-yocto skeleton Oct 24 13:55:44 I see now again things like Oct 24 13:55:46 +COMPATIBLE_MACHINE_fri2-noemgd = "fri2-noemgd" Oct 24 13:56:50 so if one/we include meta-intel theit kernel recipes are available to our machine with consequent virtual/provider clash Oct 24 13:57:28 I tried to explain that it is good habit to 'mask' the recipe for the other devices Oct 24 13:58:51 (this is not my case but iirc the setup-scripts of some distro do stack many layers, BSP or not) Oct 24 14:08:35 hmm that fri2 statement is not going to do much is it... Oct 24 14:27:43 see #12 Oct 24 14:27:44 http://cgit.openembedded.org/openembedded-core/tree/meta-skeleton/recipes-kernel/linux/linux-yocto-custom.bb Oct 24 14:28:20 and, wow, 64 Oct 24 14:30:25 (note that there is lot of black-magic behind SRC_URI interaction with kernel-tools, the instructions don't go in depth) Oct 24 14:31:29 I was helped in 2 monster email-threads with Bruce Oct 24 14:32:21 (31 and 25 emails respectively :) Oct 24 14:33:28 ..now we have 'state of the art' linux-yocto-out-of-tree support :) Oct 24 14:34:23 in fact, one could just borow meta-hh to deploy for a new board without own git tree Oct 24 14:34:33 *borrow... Oct 24 14:35:17 nice :) Oct 24 14:35:19 it's a super-skeleton in some way Oct 24 14:35:47 btw I'll have to sen you a patch for it :) Oct 24 14:36:13 Darren pushed today 3.4 -tiny so we can update bot kernels Oct 24 14:36:26 ok... I still need to apply your 3.2 patch as well right? Oct 24 14:36:26 argh..I can't type today Oct 24 14:36:41 yes, this is independent but necessary Oct 24 14:37:24 ok I'll apply that this evening Oct 24 14:38:15 if you think it is ok, you could apply with D_P = "-1" Oct 24 14:38:16 https://github.com/andrea-adami/meta-handheld/commit/18ad805bf7134e3c64810bbaaca179d2cd3c07f8 Oct 24 14:38:36 or maybe wait, I hope to clean it more thi sweekend Oct 24 14:39:13 you know, we have now to decide how to shrink Oct 24 14:39:28 one way is to redefine own KERNEL_FEATURES Oct 24 14:39:44 then we can compile as module most drivers Oct 24 14:39:57 and we can get rid of many drivers... Oct 24 14:40:27 linux-yocto-standard-base is based on a real PC Oct 24 14:40:29 :/ Oct 24 14:40:36 in mind Oct 24 14:41:32 anyway, is worth for you to look at it. ipaq should be very easy port **** ENDING LOGGING AT Thu Oct 25 03:00:00 2012