**** BEGIN LOGGING AT Sun Jul 28 02:59:58 2013 Jul 28 13:35:14 hello , i want very simple chip which has very good memory and ram and low cost , i ve seen some chip like this ; http://www.adafruit.com/products/1336 Jul 28 13:35:33 but this one is so less memory , i need 2x more memory to run freertos Jul 28 13:35:47 is there any chip like this that is able to run freertos ? Jul 28 13:42:27 is there any chip like this that is able to run freertos ? Jul 28 13:42:32 i dc sry Jul 28 16:23:26 hi, I'm trying to compile OE for the RPi, but I'm getting this error: http://ix.io/6UD . apparently, it's caused by a version mismatch and a simple rename suffices in most cases, but how do I find out what version I need to rename to? Jul 28 16:33:50 davor: check for the version of that recipe is master of meta-oe and meta Jul 28 16:36:00 davor: which meta-raspberrypi fork/branch are you building Jul 28 16:37:05 thanks zecke Jul 28 16:37:22 oo hi nerdboy! I just cloned master from sarnold's repo Jul 28 16:37:25 those look like my .bbappends Jul 28 16:37:33 that's me Jul 28 16:37:49 ahh that explains a lot :p Jul 28 16:37:54 master should build against master of the other two Jul 28 16:38:46 hmm... I was following this to get oe-core http://www.openembedded.org/wiki/OE-Core_Standalone_Setup and I got meta-oe from over here http://layers.openembedded.org/layerindex/layer/meta-oe/ Jul 28 16:39:38 i just followed the djwillis readme and extended the pi layer Jul 28 16:41:20 git://git.yoctoproject.org/poky.git <= master, top level directory Jul 28 16:41:56 git://git.openembedded.org/meta-openembedded <= master, inside poky dir Jul 28 16:42:53 git@github.com:sarnold/meta-raspberrypi.git <= master, same (inside poky dir) Jul 28 16:44:29 ah, I'll use poky then. thanks Jul 28 16:44:45 needs several layers from meta-openembedded Jul 28 16:45:19 yeah these versions seem newer (oe-core), dbus 1.6.10, mesa 9.1.5... Jul 28 16:45:47 http://paste2.org/Uv5sFgnJ Jul 28 16:46:58 oh, thanks, thought it just needed meta-oe Jul 28 16:48:06 it should work with oe-core if you copy those older ones to the newer versions... Jul 28 16:48:39 i haven't pulled in a couple weeks, they may be updated already Jul 28 16:54:39 yeah, seems it's just dbus and mesa that are outdated Jul 28 16:55:26 okay, i just pushed my latest updates, so now i can pull and see what breaks Jul 28 16:56:47 heh, plenty of commits in meta-openembedded Jul 28 16:57:45 okay, this should be a fun ride... Jul 28 16:58:45 davor: which image were you building? Jul 28 16:59:44 rpi-hwup-image Jul 28 16:59:56 seems nothing provides that in oe-core Jul 28 17:00:23 just found out, it complained about that after it checked the recipes Jul 28 17:00:27 the basic pi images should all build fine, but efl needs systemd re-enabled in distro features Jul 28 17:00:45 rpi-hwup-image is only in the meta-pi layer Jul 28 17:01:24 looks like i have three recipe errors Jul 28 17:01:55 yup, same as yours plus openssh Jul 28 17:06:36 ah, thanks Jul 28 17:06:46 where could I see available images? Jul 28 17:08:02 recipes-core/images Jul 28 17:08:13 thanks! Jul 28 17:08:36 in meta-pi... Jul 28 17:08:45 yep, found it Jul 28 17:09:04 alright, it's building with the updated recipes Jul 28 17:09:54 that was a lot of updates in both those master branches so there are many tasks to rebuild... Jul 28 17:11:38 i'll go ahead and push the meta-pi updates now, but there could be a few updates later, depending Jul 28 17:12:39 awesome, thanks! Jul 28 17:13:32 if you pull on all three master branches now, it should at least parse correctly Jul 28 17:13:53 and start building even Jul 28 17:15:07 for the rpi-* images, don't worry about swap on the sdcard Jul 28 17:16:05 er, don't worry about setting anything up Jul 28 17:19:00 nice, thanks man Jul 28 17:19:06 looks like the same updates you would need for oe-core so it should work either way Jul 28 17:19:56 you can change the size of the zram swap device in /etc/defaults/zram Jul 28 17:21:18 it'll probably take a couple to finish the first image build Jul 28 17:54:33 heh, might be more than just a few minutes nerdboy, it's only completed 190 tasks out of 3140 and it's been 30 minutes! Jul 28 17:55:16 (1131 of 6985) here Jul 28 17:55:45 but you're building hwup right? Jul 28 17:55:56 that's a *very* minimal image Jul 28 17:56:13 I'll leave it overnight if needed Jul 28 17:56:37 i'd say don't bother, but since it's your first one... Jul 28 17:57:35 nah, building basic Jul 28 17:57:48 still pretty minimal... Jul 28 17:58:04 well, I got a pretty slow computer though haha Jul 28 17:58:10 dual core athlon Jul 28 17:58:38 2.6 GHz Jul 28 17:58:55 the box i'm using is one of those triple-core amds Jul 28 17:59:19 on a cheap motherboard that unlocks the fourth core Jul 28 17:59:30 not too bad actually... Jul 28 17:59:36 nice Jul 28 17:59:44 far better than this though haha Jul 28 17:59:54 well, multiple cores do speed up the process significantly Jul 28 18:00:43 you should have parallel make set to -j5 and bitbake threads to at least 2 Jul 28 18:03:14 hmm, I set -j3 Jul 28 18:03:19 CPUs+1 Jul 28 18:03:30 well, cores+1 to be exact Jul 28 18:03:57 i usually do 2n+1 Jul 28 18:04:28 works fine except when the multiple threads hit a bunch of qt libs... Jul 28 18:05:08 so i actually sis crank it back a little to -j5 with 4 BB threads Jul 28 18:05:12 *did Jul 28 18:06:20 too much qt made it so hot the bios shutdown kicked in Jul 28 18:06:39 woah haha Jul 28 18:08:20 it's so damn bloaty and takes so long all threads hit "do_compile" at the same time Jul 28 18:11:44 hahah Jul 28 18:14:48 what did you set BB threads to? Jul 28 18:16:27 3 as well Jul 28 18:24:07 as long as the temperature doesn't go too high... Jul 28 18:25:13 it goes over 70 C too much i'd drop that to 2 and check the temp again Jul 28 18:25:33 *if Jul 28 18:26:37 hm, I have two temperatures for every core, they're around 70 Jul 28 18:27:08 the ACPI interface says it's 69°C so that's good Jul 28 18:28:15 around 70 is probably okay Jul 28 18:28:48 75 would be a little edgy... Jul 28 18:29:06 indeed Jul 28 18:30:52 will the process resume where it left off if I stop it to adjust the make parallel thread and BB thread parameters? Jul 28 18:31:09 (2247 of 6985) Jul 28 18:31:20 yes Jul 28 18:31:28 (284 of 3140) Jul 28 18:32:02 how about what it is compiling at this moment? will it need to restart compiling too? it's been on binutils for a while Jul 28 18:33:04 it would restart the task Jul 28 18:33:31 so anything big i would let it finish that particular compile task Jul 28 18:34:16 like cousin eddie says, "once he gets started, best to just let him finish..." Jul 28 18:34:24 haha Jul 28 18:34:44 I'll let it complete binutils and eglibc Jul 28 18:38:38 woo, it finished compiling binutils Jul 28 19:18:54 ah, nice, if I Ctrl+C, it waits for the tasks to finish Jul 28 19:39:08 boy, google really did make a mongo portage fork for their chromeos overlay... Jul 28 23:00:17 this is weird, bcm2835-bootfiles seems to be hanging on "do_fetch", it's been fetching it for about an hour already Jul 28 23:20:41 it's just github/raspberrypi/firmware.git Jul 28 23:21:07 it shouldn't take *that* long to clone... Jul 28 23:24:00 you should pull again before you build the rpi-xorg-image Jul 28 23:48:51 ah, it shot up an error Jul 28 23:48:57 unable to get from any source Jul 28 23:55:47 trying to git clone manually, seems to be working Jul 29 00:41:10 hm, I don't know how to archive the repo so bitbake actually recognizes it Jul 29 00:42:41 look in downloads/.git Jul 29 00:43:07 you could try to emulate the filename of the tarball it makes Jul 29 00:43:23 it should clone it fine if it's reachable Jul 29 00:43:40 did you try it again with bitbake? Jul 29 00:44:36 ah there it is! Jul 29 00:45:01 one moment, I'll put it in place, touch a .done file and try again Jul 29 00:45:33 to see more detail, you could try "bitbake -b path/to/recipe.bb -c fetch -D" Jul 29 00:46:15 the -b part cuts out dep resolution, among other things... Jul 29 00:49:57 here we go Jul 29 00:51:02 I just moved the "firmware" dir to downloads/git2/github.com.raspberrypi.firmware.git after removing the old dir, renamed github.com.raspberrypi.firmware.git.lock to github.com.raspberrypi.firmware.git.done and restarted, it immediately started unpacking! Jul 29 00:56:48 shouldn't be more than a few hours left, but I'm beat. off to bed, it's 3 AM already! good night, thanks for all the help nerdboy :D Jul 29 01:03:13 np Jul 29 01:04:57 don't forget to pull if you have problems with anything in meta-pi Jul 29 01:05:19 mine still hasn't finished either **** ENDING LOGGING AT Mon Jul 29 02:59:58 2013