**** BEGIN LOGGING AT Fri Nov 25 02:59:57 2011
Nov 25 06:47:53 <reisei> good morning!
Nov 25 08:48:07 <RP__> morning all
Nov 25 09:01:52 <bluelightning> morning all
Nov 25 09:04:20 <JaMa|Off> morning
Nov 25 09:10:53 <eFfeM_work> hi all
Nov 25 09:15:33 <JaMa|Off> bluelightning: can I use something like if [[ "${CMDLINE_DEBUG}" =~ "debug" ]]; in linux-kexecboot.inc? I don't see any sanity check for bash version or don't know if it's available in dash
Nov 25 09:16:08 <bluelightning> JaMa|Off: I was thinking about that yesterday
Nov 25 09:16:13 <bluelightning> that's a bashism to be sure...
Nov 25 09:16:19 <JaMa|Off> other option is to grep for it :)
Nov 25 09:16:35 <bluelightning> or, you could use sed
Nov 25 09:17:23 <JaMa|Off> if you prefer if echo "${CMDLINE_DEBUG}" | grep "debug" >/dev/null ; then
Nov 25 09:17:54 <bluelightning> actually you could still use base_contains within the shell script
Nov 25 09:18:08 <bluelightning> that's probably a bit less nasty
Nov 25 09:19:37 <JaMa|Off> ah ok
Nov 25 09:22:16 <mgundes> Hi everyone
Nov 25 09:22:26 <mgundes> I have question again :)
Nov 25 09:22:27 <bluelightning> hello mgundes
Nov 25 09:22:42 <bluelightning> sure, fire away
Nov 25 09:23:12 <mgundes> When a rootfs image created OE install all ipkgs to rootfs as far as I know
Nov 25 09:23:36 <mgundes> Is there a way to tell OE install a specific ipkg before another one
Nov 25 09:24:02 <mgundes> I want to overwrite a file from one package with file of another package
Nov 25 09:24:22 <pwgen> hi all
Nov 25 09:24:31 <mgundes> so that I wonder if there is a way to tell OE about ipk installation sequence for some packages
Nov 25 09:24:44 <pwgen> netwalker goes oe-core ...
Nov 25 09:25:02 <bluelightning> mgundes: well, in the purest sense the way to ensure some ordering is to set up a dependency relationship
Nov 25 09:25:19 <bluelightning> mgundes: but I don't think that will help in your case; opkg will complain about the overwriting
Nov 25 09:25:20 <JaMa|Off> bluelightning: ok, I'm testing this https://gitorious.org/shr/meta-handheld/commit/b3b1be4c80285d9441614e3537dcdafaa9204bc2?format=patch
Nov 25 09:26:12 <mgundes> in fact I want to overwrite update-alternatives script which comes from opkg-nopg
Nov 25 09:26:27 <mgundes> I want to add sync end of function in script
Nov 25 09:26:29 <bluelightning> mgundes: the best way to solve it is not to have that file in two packages; either by excluding it from the first or changing the file in the first and not having it in the second
Nov 25 09:27:08 <bluelightning> mgundes: if the modification is that trivial, just patch the script where it is now...
Nov 25 09:27:30 <bluelightning> you can do that via a bbappend in your own layer on top if you want to keep things tidy
Nov 25 09:27:45 <mgundes> it is in opkg-nopg package and it is on OE side
Nov 25 09:28:21 <bluelightning> that's fine, the recipe can still be bbappended
Nov 25 09:28:41 <mgundes> I did not use bbappend
Nov 25 09:28:44 <bluelightning> and in your bbappend, add to SRC_URI
Nov 25 09:28:50 <bluelightning> well, you could
Nov 25 09:28:56 <mgundes> then I will do some search about it
Nov 25 09:28:58 <mgundes> ok 
Nov 25 09:29:08 <mgundes> let me check and try, thanks
Nov 25 09:29:25 <ant_work> pwgen: which kernel?
Nov 25 09:29:41 <pwgen> stock 2.6.28. 
Nov 25 09:30:26 <ant_work> no chances with 3.x ?
Nov 25 09:31:27 <pwgen> sorry but 2.6.31 and 35 are have anoying screen flickering 
Nov 25 09:32:02 <pwgen> does 3.0 have arm support for cortex a8
Nov 25 09:33:15 <mgundes> bluelightning, bbappend can be used in oe-classic also, right?
Nov 25 09:33:30 <bluelightning> mgundes: if you use bitbake 1.12.0 or newer yes
Nov 25 09:33:42 <ant_work> pwgen: sure
Nov 25 09:34:15 <mgundes> I use 1.8.18, I will upgrade it
Nov 25 09:35:58 <florian> good morning
Nov 25 09:38:23 <pwgen> ant_work: i tried these kernels from darion76. but he didnt finished them. they needed these patches from lineo arm
Nov 25 09:38:30 <Jin^eLD> mornin
Nov 25 09:38:51 <ant_work> pwgen: I see, i.MX515
Nov 25 09:39:06 <ant_work> iirc there was some work for efika
Nov 25 09:41:20 <bluelightning> JaMa|Off: looks good to me, if it works I'll merge it
Nov 25 09:42:51 <stomach_contents> Hey guys. I'm trying in vain to build a kernel (or anything) for overo using Angstrom with MACHINE=overo ./oebb.sh bitbake virtual/kernel but fails with a sanity warning about MACHINE. Seems to not know overo?  Any ideas? Works perfectly for beagleboard.
Nov 25 09:44:51 <Jin^eLD> if I want to generate a toolchain that could be used without OE, is it still "meta-toolchain" in OE-core? or what are these -sdk recipes?
Nov 25 09:46:15 <bluelightning> stomach_contents: do you have some kind of additional layer providing the overo machine? I'm guessing not...
Nov 25 09:46:55 <pwgen> <ant_work> but there are patch for 2.6.38 out 
Nov 25 09:51:24 <pwgen> with 2.6.28 i can not run systemd.  
Nov 25 09:52:27 <pwgen> i build the 2.6.28 out of the efikamx recipe, but its generates allways a u-boot image, how can i for it to build an std zImage ? 
Nov 25 09:57:38 <Jin^eLD> pwgen: try setting KERNEL_IMAGETYPE = "zImage" I think that should work?
Nov 25 09:57:42 <ant_work> yes
Nov 25 09:58:33 <Noor> hi guys
Nov 25 09:58:50 <pwgen> Arrghh i see ..3 line ... KERNEL_IMAGETYPE = "uImage" .... THX
Nov 25 09:58:57 <Noor> I need some help regarding understanding sdk recipes
Nov 25 09:59:31 <Noor> I have libltdl-dev in RDEPENDS in one of my tasks
Nov 25 09:59:54 <Noor> when I build that task I get NOTE: preferred version 2.4 of libtool-sdk not available (for item libltdl-dev)
Nov 25 10:00:29 <Noor> although libltdl-dev is present in PACKAGES in libtool_2.4.bb recipes
Nov 25 10:00:58 <Noor> why it saying that it needs libtool-sdk version 2.4
Nov 25 10:01:10 <Noor> and where this dependency is set
Nov 25 10:03:22 <Noor> the older version of libtool-sdk are requiring libtool_PV.bb file
Nov 25 10:03:58 <Noor> so when I add that in libtool-sdk_2.4.bb the warning is removed but libtool-sdk is not building
Nov 25 10:04:16 <Noor> so what is going here ... can anybody help
Nov 25 10:08:11 <pwgen> <ant_work> kernel upgrade/patching without a serial console is quite hard
Nov 25 10:09:50 <ant_work> pwgen: netconsole as last resort ?
Nov 25 10:11:55 <pwgen> .. this thingy has only usb wlan sound plugs . no network, i have to unscrew it and find the 3.3V serial port.
Nov 25 10:12:42 <pwgen> and with a bit f luck there are jtag ports, and with mor luck i get the openmoko debug board connected
Nov 25 10:13:08 <ant_work> ouch
Nov 25 10:14:23 <pwgen> unless i get the power regulator and backlight runninf i will see nothing .. and one of the worsted thing is , kexec is not working (:-((
Nov 25 10:15:36 <pwgen> but the good news are, i could build all angtrom images, and core-images 
Nov 25 10:16:13 <ant_work> you'll have issues with udev if not using systemd
Nov 25 10:16:22 <pwgen> whats about oe-stats. is this not supported in oe-core ?
Nov 25 10:17:01 <Jin^eLD> ant_work: why that? curious because I am not using systemd, what issues?
Nov 25 10:17:19 <ant_work> i.e. cgroups are required
Nov 25 10:17:35 <Jin^eLD> what is cgroups? well let me google...
Nov 25 10:18:09 <ant_work> udev 169 and up require kernel 2.6.36 for ARM
Nov 25 10:18:16 <Jin^eLD> isn't that something that only became available in the recent kernel? and how are those cgroups related to systemd?
Nov 25 10:18:26 <ant_work> http://git.kernel.org/?p=linux/hotplug/udev.git;a=commit;h=67a77c8bf299f6264f001677becd056316ebce2f
Nov 25 10:19:12 <Jin^eLD> but how is that related to systemd? I somehow don't get tha part
Nov 25 10:19:25 <ant_work> the issue is with udev
Nov 25 10:20:04 <ant_work> but iirc koen has som epatches for older kernels
Nov 25 10:34:32 <Jin^eLD> 1. fixed some more live update problems, updated nspr
Nov 25 10:34:35 <Jin^eLD> oops wrong window
Nov 25 11:00:56 <pwgen> nice, core-image-sato image on netwalker 
Nov 25 11:05:38 <ant_work> pwgen: are you including meta-oe layer?
Nov 25 11:26:29 <pwgen> yes
Nov 25 11:26:45 <pwgen> its a long list
Nov 25 11:27:47 <CIA-15> 03Richard Purdie <richard.purdie@linuxfoundation.org> 07master * r5ace320ccc 10bitbake.git/lib/bb/siggen.py: 
Nov 25 11:27:48 <CIA-15> bitbake/siggen.py: Don't backtrace if the taskhash data isn't present
Nov 25 11:27:48 <CIA-15> This allows the code to safely fall back to dumping the basehash data
Nov 25 11:27:48 <CIA-15> if the taskhash data isn't present for some reason. We could effecitvely
Nov 25 11:27:48 <CIA-15> obsolete the runtime option and use this approach instead exclusively.
Nov 25 11:27:48 <CIA-15> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Nov 25 11:30:28 <pwgen> ant_work: http://pastebin.com/Dm4WTd2t 
Nov 25 11:31:06 <JaMa|Off> bluelightning: tested and updated in jansa/pull branch
Nov 25 11:31:49 <sledges> hello
Nov 25 11:32:54 <sledges> How could I make ltp recipe work? (from http://comments.gmane.org/gmane.comp.handhelds.openembedded/47909 :: Is there a way to specify the file name the downloaded file should be saved as?) as I can't wait for upstream to changing their naming convention..
Nov 25 11:35:21 <pwgen> ant_work: but it need still a lot of additional work ( ubi-fs .. get it run out of the box )
Nov 25 12:00:44 <ant_work> pwgen: did you test core-image-sato runtime?
Nov 25 12:01:26 <pb_> hi all
Nov 25 12:01:58 <sledges> hi pb_
Nov 25 12:01:59 <ant_work> hi pb_ 
Nov 25 12:02:39 <sledges> I am writing custom do_unpack. How to get the directory where the tarball was fetched into (presumably the sources one)
Nov 25 12:02:40 <pwgen> all images were not running out of the box ( sometimes it needs tuning /etc/inittab and missing additional packages )
Nov 25 12:04:23 <Jin^eLD> hmm, when I get this: http://pastebin.mozilla.org/1388221 does that mean it did not like that configure refenreces its own location by an absolute path, or does that mean that the configure that triggered the error is in that path?
Nov 25 12:04:37 <Jin^eLD> (both is true, want to understand what to fix here)
Nov 25 12:09:00 <bluelightning> sledges: you may find classes/base.bbclass useful as a reference
Nov 25 12:09:45 <pb_> Jin^eLD: the latter
Nov 25 12:09:52 <pb_> you need to inspect config.log to find the actual error
Nov 25 12:10:04 <Jin^eLD> ok I see
Nov 25 12:10:26 <Jin^eLD> trying to compile new 1.8.5 spidermonkey that switched to autotools, but intsead of doing it the proper way they are doing some really twisted stuff
Nov 25 12:11:45 <bluelightning> sledges: or not... probably ${DL_DIR}/filename
Nov 25 12:33:24 <bluelightning> JaMa|Off: merged, thanks!
Nov 25 12:38:01 <ant_work> bluelightning: now we have only h1940 in 2.6.39 ...
Nov 25 12:38:11 <bluelightning> ah right yes
Nov 25 12:38:20 <bluelightning> h1940 should be able to move up
Nov 25 12:38:25 <bluelightning> but I should really test it...
Nov 25 12:38:43 <ant_work> happily was not needing any patch (apparently)
Nov 25 12:39:58 <Jin^eLD> is meta-toolchain doing the same in OE-core as in OE-classic? and what are the -sdk recipes?
Nov 25 12:41:12 <anarsoul> bluelightning: 3.0 should work with h1940
Nov 25 12:42:08 <anarsoul> not sure about 3.1
Nov 25 12:42:38 <anarsoul> there were some changes in adc driver, so touchscreen and battery can be broken with 3.1
Nov 25 12:44:11 <anarsoul> and bluetooth issue is not resolved yet
Nov 25 12:44:17 <ant_work> then we can start removing 2.6.39 righ tnow and 3.0.1 very soon
Nov 25 12:45:16 <ant_work> anarsoul: so you mean we should add h1940 to 3.0.1 isn't ?
Nov 25 12:46:15 <anarsoul> ant_work: well, it worked for me with 3.0, not sure if I applied any extra patches
Nov 25 12:49:02 <anarsoul> ant_work: my h1940 is briked, planning to unbrick it this weekend. Then I can test if 3.1 works
Nov 25 12:49:35 <ant_work> great, thx
Nov 25 12:51:01 <ant_work> less kernel recipes is always better
Nov 25 12:51:42 <bluelightning> indeed
Nov 25 13:16:11 <ant_work> JaMa|Off: we need to refine the cmdline mangling, things have changed for arm with 3.x
Nov 25 13:21:42 <sledges> bluelightning: many thanks, worked like charm and you've saved my day :) not really that I could get more info from the net on that DL_DIR!
Nov 25 13:21:52 <sledges> hence I'd have never found it..
Nov 25 13:22:07 <bluelightning> sledges: no problem :)
Nov 25 13:40:37 <JaMa|Off> ant_work: agreed
Nov 25 14:05:34 <Crofton|work> bluelightning, have you had a chance to look at the bb file I sent you?
Nov 25 14:07:04 <bluelightning> Crofton|work: I haven't really I'm afraid no
Nov 25 14:08:20 <bluelightning> that reminds me, I need to set up my home machine for builds, the work one is packed up at the moment
Nov 25 14:08:34 <bluelightning> qt takes so long to build I usually prefer not to do it on my laptop :)
Nov 25 14:11:19 <Crofton|work> heh
Nov 25 15:09:36 <mgundes> Are we allowed to create a branch of OE on remote server :)
Nov 25 15:11:01 <mgundes> sometimes it is hard to solve problems with our own recipes and if we are allowed to create a personal branch I can do some changes on OE recipes up to my personal purpose.
Nov 25 15:11:07 <mgundes> is it possible?
Nov 25 15:12:00 <HeinervdmWork> just make one on github or so
Nov 25 15:14:00 <An-iSociaL> afternoon
Nov 25 15:14:01 <mgundes> good idea
Nov 25 15:19:22 <stefan_schmidt> mgundes: if you have access you should be able to add remote branches. Deleting is not allowed though. Needs to be done manually
Nov 25 15:19:36 <stefan_schmidt> mgundes: Or use another git hoster. As you prefer
Nov 25 15:21:20 <mgundes> I am trying to fork and test submit issue on github
Nov 25 17:36:43 <Jin^eLD> hmm has this been fixed in OE core? http://patches.openembedded.org/patch/14191/ I'm seeing the problem when building in a VM
Nov 25 17:40:31 <Jin^eLD> can i simply disable buildstats somehow?
Nov 25 17:50:07 <Jin^eLD> hmm, could not find who is inheriting it... put an empty buildstats.bbclass in my layer :P
Nov 25 18:00:52 <kergoth> Jin^eLD: a quick grep/ack shows that base.bbclass inherits it.
Nov 25 18:01:51 <Jin^eLD> kergoth: yes I found that too but I do not know how to "uninherit" it
Nov 25 18:04:11 <Jin^eLD> so I put an empty buildstats.bbclass in my layer to "wipe" over the default one
Nov 25 18:04:31 <woglinde> jo zecke
Nov 25 18:05:11 <zecke> woglinde: yo
Nov 25 18:05:36 <zecke> woglinde: did you hear about Taku yet? some small buildroot for Rasperry Pi and Qt5
Nov 25 18:06:42 <woglinde> no
Nov 25 18:06:57 <woglinde> hi kergoth
Nov 25 18:07:11 <kergoth_> hey
Nov 25 18:08:45 <kergoth_> i should really set up an irc proxy again
Nov 25 18:08:47 <kergoth_> too many me's
Nov 25 18:12:28 <kergoth_> Totally OT, but does anyone own a herman miller embody or steelcase leap? I'm trying to make up my mind between them and having a hell of a time
Nov 25 18:13:29 * Jin^eLD does not even know what kergoth is talking about
Nov 25 18:13:35 <Crofton|work> chairs?
Nov 25 18:14:27 <khem> kergoth_: too expensive for my taste 
Nov 25 18:14:52 <kergoth_> that's what i thought too, until I started having back problems
Nov 25 18:15:00 <kergoth_> the $100 chair from staples isn't cutting it anymore
Nov 25 18:15:10 <kergoth_> heh
Nov 25 18:15:22 <khem> hmmm
Nov 25 18:16:07 <khem> I have a good one at work I would recommend that to you if I knew the make and model
Nov 25 18:16:26 <khem> I will take a look on it monday
Nov 25 18:16:30 <khem> its really nice chair
Nov 25 18:16:31 <kergoth_> cool, thanks
Nov 25 18:16:50 <Crofton|work> I'm with kergoth, the cheaper chairs just suck
Nov 25 18:17:04 * mwester has discovered that fact, as well.
Nov 25 18:17:15 * kergoth_ wishes he had used a better one 10 years ago and maybe his back wouldn't be as bad as it is now
Nov 25 18:17:20 <khem> Crofton|work: I hardly sit on the chairs other than office one
Nov 25 18:17:32 <khem> at home I dont sit at one place for long 
Nov 25 18:17:47 <khem> kergoth_: excercise :)
Nov 25 18:18:23 <kergoth_> yeah, I'm working on that too
Nov 25 18:18:30 <kergoth_> strengthening the core and working on flexibility should help
Nov 25 18:18:38 <khem> http://hermanmiller.com/Products/Embody-Chairs
Nov 25 18:18:41 <kergoth_> also standing a portion of the day helps a lot
Nov 25 18:18:54 <khem> actually now I think my office chair is similar to the orange one in shape
Nov 25 18:19:47 <khem> kergoth_: there are effective yoga postures for back
Nov 25 18:19:58 <khem> I have myself tried that
Nov 25 18:20:03 <kergoth_> hmm, sounds interesting
Nov 25 18:20:13 <khem> and they also helped a friend with chronic back problem
Nov 25 18:20:29 <kergoth_> i figured i'll start doing some basic stretching and warmup type exercises every morning
Nov 25 18:20:52 <recalcati> good morning, as I can compile many recipes in oe-core, gstreamer-ti there is no way. problem on fecth. I'm doing something wring?
Nov 25 18:21:21 <kergoth_> fetch failures are usually caused by upstream locations removing the files
Nov 25 18:21:27 <woglinde> jupp
Nov 25 18:21:33 <recalcati> yes, but are ti related
Nov 25 18:21:35 <khem> http://commons.wikimedia.org/wiki/File:02_Purna_Sarpasana.jpg
Nov 25 18:21:43 <woglinde> recalcati hm?
Nov 25 18:21:55 <woglinde> recalcati without errors we cannt tell
Nov 25 18:22:04 <woglinde> and not all ti stuff is open available
Nov 25 18:22:09 <woglinde> for some you have to register
Nov 25 18:22:15 <woglinde> and download manually
Nov 25 18:22:21 <kergoth_> haha. I decided to try importing the kernel tree into mercurial as a test of hg-git. it'll be done in 9 hours.
Nov 25 18:22:44 <woglinde> lol
Nov 25 18:23:16 * kergoth_ is trying to get as comfortable with mercurial as he is with git, since they're so similar
Nov 25 18:23:49 <bluelightning> kergoth_: I've had good results with rebuildyourback.com FWIW
Nov 25 18:23:52 <recalcati> see here  http://pastebin.com/ZKyx51WA
Nov 25 18:24:21 <recalcati> woglinde: so I wait for ti, right?
Nov 25 18:24:55 <kergoth_> bluelightning: thanks, I'll check that out
Nov 25 18:25:20 <kergoth_> its getting to the point that the more time I spend in my computer chair, the more my lower back hurts the next day. like clockwork, every time
Nov 25 18:25:25 <bluelightning> back trouble is an occupational hazard of our profession I fear :/
Nov 25 18:25:46 <khem> true
Nov 25 18:25:52 <kergoth_> indeed
Nov 25 18:26:09 <woglinde> recalcati write to the meta-ti list
Nov 25 18:26:19 <kergoth_> polar opposite of folks in construction, etc, who also have that srot of problem, but more due to stress on the body than inactivity
Nov 25 18:26:19 <woglinde> or ask koen in #beagle
Nov 25 18:26:43 <khem> kergoth_: yeah irony of life
Nov 25 18:26:59 <khem> kergoth_: swimming also helps actually if you like swimming 
Nov 25 18:27:27 <kergoth_> hmm, that's an interesting idea, I don't do that much, maybe I should
Nov 25 18:28:03 <recalcati> woglinde: is ok, I didn0t know there was that list. thx
Nov 25 18:28:08 <khem> I am not a good swimmer so I struggle but I try weeklly once :)
Nov 25 18:28:15 <kergoth_> more and more I'm taking to exercises that work more of the body, rather than the focused stuff like weight lifting machines do. better for flexibility, and all the muscles that those don't tend to use
Nov 25 18:28:26 <kergoth_> swimming seems like it'd be a good chocie for that
Nov 25 18:28:33 <kergoth_> heh
Nov 25 18:28:40 <kergoth_> anyway
Nov 25 18:29:08 <khem> I dont do weights since I dont want to look cool hunk :)
Nov 25 18:29:50 <kergoth_> don't necessarily bulk up, depending on what you do. I've heard that weight lifting can increase how many calories you burn when at rest. dunno if thats true or not, but its certainly an interesting possibility
Nov 25 18:30:00 * kergoth_ needs to finish reading the 4 hour body
Nov 25 18:30:34 <khem> yes muscles are natural calorie burners thats true
Nov 25 18:30:52 <khem> but then they need to be excercised as well
Nov 25 18:31:46 <kergoth_> I'd really like to do more activities where the exercise / body improvement is a side effect. Easier to get motivated to, say, go hiking than to go to the gym
Nov 25 18:31:47 <kergoth_> heh
Nov 25 18:32:15 <khem> exactly
Nov 25 18:32:30 <khem> and it has to be sustainable change
Nov 25 18:32:50 <khem> sometimes one goes to gym for a while then stops for months
Nov 25 18:33:17 <kergoth_> Yeah, I've never been able to sustain it in the long term :\
Nov 25 18:33:48 <JaMa|Off> works for me :)
Nov 25 18:35:46 <kergoth_> I'm *terrible* at building habits. I don't even stick to a proper morning/evening routine every day
Nov 25 18:35:46 <kergoth_> heh
Nov 25 18:36:23 <JaMa|Off> kergoth_: and it's true that you're burning more at rest but only when your body is rebuilding muscles
Nov 25 18:36:39 <JaMa|Off> kergoth_: so it takes training with higher intensity to get this effect
Nov 25 18:36:56 <khem> anaerobic excercises
Nov 25 18:37:06 <JaMa|Off> or at least interval training with some smaller part in anaerobic spectrum
Nov 25 18:38:11 <kergoth_> ah, I see
Nov 25 18:39:27 <JaMa|Off> but wrt total amount of fat burned during excersises you shouldn't spend whole time in anaerobic, because the percentage of energy from fat is better with lower hr
Nov 25 18:39:42 <JaMa|Off> and the anaerobic part is only to kick metabolism for few next days
Nov 25 18:40:18 <kergoth_> I have a different problem from many, I need to bulk up, I accidentally lost like 20 pounds over the past year due to improving our diet
Nov 25 18:40:18 <kergoth_> heh
Nov 25 18:40:34 <woglinde> hm
Nov 25 18:41:03 * kergoth_ gets back to code
Nov 25 18:41:13 * JaMa|Off gets back to beer
Nov 25 18:41:19 <woglinde> jama lol
Nov 25 18:41:26 * Crofton|work curses cmake
Nov 25 18:41:34 <Crofton|work> well cmake+Qt
Nov 25 18:41:38 * JaMa|Off curses staropramen
Nov 25 18:41:39 <woglinde> crofton *g*
Nov 25 18:41:54 <woglinde> jama why you drink it
Nov 25 18:41:59 <woglinde> even we german avoid it
Nov 25 18:42:13 <woglinde> urpils krusovic
Nov 25 18:42:29 <JaMa|Off> woglinde: I'm visiting my father this weekend and he gave it to me..
Nov 25 18:42:35 <JaMa|Off> but normally it's not so bad
Nov 25 18:42:42 <woglinde> uhm
Nov 25 18:44:08 <woglinde> hm so I hope this was the last error in meta-java
Nov 25 18:44:24 <JaMa|Off> woglinde: where lives meta-java?
Nov 25 18:44:30 <woglinde> *g*
Nov 25 18:44:32 <woglinde> here
Nov 25 18:44:39 <woglinde> until the build finishes
Nov 25 18:44:49 <woglinde> than I will push it to my github account
Nov 25 18:44:59 <JaMa|Off> *g*
Nov 25 18:45:34 <woglinde> needs still some polish for PR stuff and some licenses mapping
Nov 25 18:46:17 <woglinde> and of course updates to the newer icedtea releases
Nov 25 18:57:35 <woglinde> hms
Nov 25 18:57:38 <woglinde> typo in llvm
Nov 25 18:58:48 <woglinde> hm how is the syntax to apply a patch only when uclibc and arm is used
Nov 25 19:00:02 <khem> woglinde: you need a python func I guess
Nov 25 19:00:20 <woglinde> hrms
Nov 25 19:00:25 <woglinde> or you fix fenv quickly
Nov 25 19:00:31 <Jin^eLD> does SRC_URI_append_uclibc no longer work?
Nov 25 19:00:40 <Jin^eLD> I remember stuff like that worked in classic at least
Nov 25 19:00:42 <khem> yes 
Nov 25 19:00:52 <Jin^eLD> ah, uclibc AND arm...
Nov 25 19:00:52 <khem> but he is asking for 2 overrides
Nov 25 19:01:05 <Jin^eLD> yes, I was too slow on that :P
Nov 25 19:01:45 <woglinde> crofton what was the problem with cmake and qt again?
Nov 25 19:01:51 <woglinde> maybee its the problem I posted
Nov 25 19:01:54 <Crofton|work> FindQt4 macro hates cross
Nov 25 19:02:08 <Crofton|work> I am having to overrisde a bunch of stuff it should detect
Nov 25 19:02:11 <woglinde> wrong gcc-cross default c++ include path
Nov 25 19:02:16 <Crofton|work> basically the macro uses qmake
Nov 25 19:02:34 <Crofton|work> but won't pass qmake a spec file, so itreirns the build system paths into the sysroot
Nov 25 19:03:00 <woglinde> hm okay
Nov 25 19:04:08 <Crofton|work> basically, working out how to tell the macro where stuff is
Nov 25 19:04:52 <woglinde> khem hm cannt I define a variable with _uclibc_append bla and use that in SRC_URI_arm_append ?
Nov 25 19:05:04 <Crofton|work> testing changes takes forever also 
Nov 25 19:05:17 <Crofton|work> if I turn on || make, usually the error message is ahrd to find
Nov 25 20:17:12 <Crofton|work> bluelightning, still awake?
Nov 25 20:34:32 <Jin^eLD> in OE core, is meta-toolchain doing the same as in classic? and what exactly are the -sdk recipes?
Nov 25 20:34:56 <woglinde> sdk is for sdk
Nov 25 20:35:13 <Jin^eLD> what is "sdk" in this sence?
Nov 25 20:35:47 <woglinde> sdk
Nov 25 20:35:56 <woglinde> software development kit
Nov 25 20:35:58 <Jin^eLD> "software development kit" - what is that?
Nov 25 20:36:12 <Jin^eLD> yes I know the word ) but what is that in OE sence, what is the difference between -sdk and normal packages?
Nov 25 20:36:46 <woglinde> sdk has some special cases where native dont fit
Nov 25 20:37:23 <Jin^eLD> I still do not really undstand the difference :P
Nov 25 20:37:44 <Jin^eLD> but it is NOT meta-toolchain? so that if I want to build a standalone toolchain for use outside of OE, then meta-toolchain is still the recipe I want?
Nov 25 20:38:36 <woglinde> checkout if there is a .bb
Nov 25 20:38:38 <woglinde> for it
Nov 25 20:38:43 <Jin^eLD> I believe there is
Nov 25 20:39:26 <woglinde> and its maybee using sdk class
Nov 25 20:39:44 <Jin^eLD> I guess I'll try it out and see how it gos
Nov 25 20:55:15 <khem> Jin^eLD: yes meta-toolchain generated SDK
Nov 25 20:55:39 <Jin^eLD> so the -sdk packages are actually *for* the meta toolchain?
Nov 25 20:55:56 <khem> Jin^eLD: in oe-core there should be nativesdk 
Nov 25 20:57:16 <Jin^eLD> and what is it? is it the same as meta-toolchain? I still seem not to understand the difference
Nov 25 20:57:53 <khem> ok. meta-toolchain is a meta target for sdk 
Nov 25 20:58:25 <khem> it will consist of packages for target as well as the host system on which sdk will run
Nov 25 20:58:37 <Jin^eLD> ok now I thik I get it
Nov 25 21:00:17 <khem> so any image you have ending with -sdk will actually make sure that dev packages and toolchain compoents are packaged for that image
Nov 25 21:01:32 <Jin^eLD> aah, I see
Nov 25 21:01:52 <Jin^eLD> so like an additional meta-toolchain target, but in a form of a rootfs
Nov 25 21:03:15 <khem> right
Nov 25 21:03:34 <khem> basically it will make the image debuggable
Nov 25 21:05:51 <Jin^eLD> thanks for the pointers, I'll try that
Nov 25 21:06:48 <khem> nativesdk packages are targetted for SDK_HOST
Nov 25 21:15:30 <khem> woglinde: I am not sure about if those _appends will work that way
Nov 25 21:15:58 <woglinde> hm yes
Nov 25 21:16:33 <woglinde> hm libiconv 1.14 is out
Nov 25 21:49:24 <Jin^eLD> these udev -> glib / consolekit / systemd dependencies are so annoying :(
Nov 25 21:57:20 * kergoth_ hacks on an hg equivalent to his git-cache bits
Nov 25 22:13:52 <bluelightning> Crofton|work: still awake, was afk
Nov 25 22:17:36 <Crofton|work> I made some progress with qt
Nov 25 22:17:47 <Crofton|work> basically, set lots of vars to tell findqt4 where stuff is
Nov 25 22:17:52 <Crofton|work> it is gross
Nov 25 22:37:40 <bluelightning> Crofton|work: hmm :/
Nov 25 22:38:21 <bluelightning> I wonder if it would be more practical to patch FindQt4 so that you can specify a common path
Nov 25 22:40:07 <clockwize> hi guys, i'm trying to do the getting started guide, i've gone through everything and now i'm on bitbake nano step and I get this: https://gist.github.com/1394583
Nov 25 22:40:11 <clockwize> can anyone advise?
Nov 25 22:55:16 <Crofton|work> or adjust toolchain file
Nov 25 22:55:43 <Crofton|work> I'll be out tonight, I will send you updated bb file so you can see what I had to do
Nov 25 23:07:18 <bluelightning> Crofton|work: ok thanks
Nov 25 23:10:18 <clockwize> anyone..
Nov 25 23:10:31 <clockwize> uclibc_0.9.30.3.bb syntax error?
Nov 25 23:17:07 <rob_w> or a in classes/conf/roofs_....
Nov 25 23:17:39 <rob_w> did ur set work before ?
Nov 26 01:51:33 <rlrosa> hi
Nov 26 01:52:09 <rlrosa> i checked out oe (the setup-script repo), configured  for a beagleboard and tried:
Nov 26 01:52:09 <rlrosa> bitbake console-image
Nov 26 01:52:28 <rlrosa> i get errors in gpio code: | arch/arm/plat-omap/gpio.c:1677: error: 'gpio_pads_config' undeclared (first use in this function)
Nov 26 01:52:47 <rlrosa> i checked out remote.origin.url=git://git.angstrom-distribution.org/setup-scripts
Nov 26 01:53:20 <rlrosa> any tips?
Nov 26 02:23:06 <rlrosa> anybody around?
**** ENDING LOGGING AT Sat Nov 26 02:59:57 2011