**** BEGIN LOGGING AT Fri Sep 23 02:59:56 2011
Sep 23 06:36:40 <tasslehoff> getting started with bblayers and need some help. my first case is that I need to add my own /etc/fstab and /etc/profile. I use angstrom, which also has a .bbappend for base-files. Will OE then use things from all 3 recipes, based on priority?
Sep 23 06:37:57 <GNUtoo> tasslehoff, maybe try it to see....
Sep 23 06:38:37 <tasslehoff> GNUtoo: yep. just want to understand it as well, but I'll try first :)
Sep 23 06:38:52 <GNUtoo> ok
Sep 23 06:49:47 <tasslehoff> and it seems it did
Sep 23 06:50:18 * tasslehoff likes his new oe-core much better than his oe-classic
Sep 23 06:52:03 <JaMa> tasslehoff: yes it will append all .bbappend files it can find
Sep 23 06:52:30 <tasslehoff> how should one use PRINC? I see that angstrom has it set to 8 for base-files, while I used 1 after looking at an example in khem's online braindump.
Sep 23 06:53:02 <JaMa> tasslehoff: probably increased 8 times.. that's why there is 8
Sep 23 06:53:19 <GNUtoo> hi JaMa 
Sep 23 06:53:37 <JaMa> hi
Sep 23 06:54:07 <tasslehoff> JaMa: ah. of course. I use that one to trigger a rebuild...
Sep 23 07:26:32 <tasslehoff> what's the proper way to make sure rtc_twl.ko is loaded on startup? in my old patch I did an insmod in hwclock.sh, but it seems a bit hackish.
Sep 23 07:29:33 <JaMa> tasslehoff: grep for module-autoload
Sep 23 07:29:44 <JaMa> tasslehoff: usually in machine confs
Sep 23 07:43:57 <tasslehoff> JaMa: thx
Sep 23 07:44:10 <ant_work> JaMa: heh, I see you updated libx11, thx
Sep 23 07:46:20 <ant_work> ah, no, is the old patch
Sep 23 07:47:03 <ant_work> somehow with the -contrib patches libx11 failed do_configure
Sep 23 07:47:07 <Noor> hi all 
Sep 23 07:47:36 <ant_work> 'morning
Sep 23 07:49:11 <Noor> I am working on sstate-cache ..... got a scenario where I have to ignore one varaible which is being used in base-files'd do_install funtion .... I know we can exclude the varaibles using vardepsexclude but cant find the method with which I can exclude dependency of particular task for a partular recipe
Sep 23 07:49:36 <JaMa> ant_work: ?
Sep 23 07:50:04 <JaMa> ant_work: and no need to revert anything.. those branches are rebased quite often so this change was already included there..
Sep 23 07:50:18 <ant_work> ?
Sep 23 07:50:30 <JaMa> 00:24:00 < ant_home> JaMa: last 3 oe-core patches about xorg need to be reverted to apply your patchset clean
Sep 23 07:50:54 <ant_work> yes, they did conflict with oe-core
Sep 23 07:51:03 <JaMa> no they didn't!
Sep 23 07:51:12 <ant_work> then it was too late ;)
Sep 23 07:51:28 <JaMa> because jansa/x11 was already updated when otavio patches were pushed to oe-core..
Sep 23 07:52:12 <ant_work> anyway I could fix that
Sep 23 07:52:41 <ant_work> then launched the build and zZZZ, this morning build failed
Sep 23 07:53:05 <JaMa> and what's do_configure error?
Sep 23 07:53:11 <ant_work> no tme to inspect the log, just seen it was libx11
Sep 23 07:53:42 <ant_work> checking for xxyy  .. not found akind
Sep 23 07:54:37 <tasslehoff> Is it possible to append to a machine-conf from my meta-layer?
Sep 23 07:54:38 <JaMa> hmm xxyy and aking are just "foo" and "bar" or is it real message?
Sep 23 07:55:12 <ant_work> rephrased: it was th etypical kind of configure/not found error
Sep 23 07:55:49 <ant_work> just can't remember whic lib was not found :p
Sep 23 07:56:21 <ant_work> np, I'll retry this evening
Sep 23 08:00:19 <JaMa> ok
Sep 23 08:00:31 * JaMa I guess libxcb or something like that
Sep 23 08:00:55 <JaMa> but maybe you were building libx11-trim-native or something like that
Sep 23 08:01:01 <JaMa> I prefer libx11-native
Sep 23 08:01:43 <JaMa> you would need patches from another x11 branch like this http://git.shr-project.org/git/?p=meta-smartphone.git;a=shortlog;h=refs/heads/x11
Sep 23 08:01:45 <ant_work> hmm, probably the -trim iirc
Sep 23 08:02:03 <JaMa> but applied to meta-angstrom or whatever you're using
Sep 23 08:02:56 <ant_work> ok, I'm building with no-distro atm
Sep 23 08:27:07 <alx_> hi all
Sep 23 08:27:52 <alx_> I'm working with oe-core (yocto) and I'm trying to have a first build working, but it hangs in compiling qemu-native asking for libGL.so and libGLU.so
Sep 23 08:27:59 <alx_> how to solve this issue ?
Sep 23 08:32:21 <JaMa> alx_: are you using external-toolchain?
Sep 23 08:32:47 <alx_> JaMa, no. I just follow http://www.yoctoproject.org/docs/yocto-quick-start/yocto-project-qs.html
Sep 23 08:33:00 <alx_> but it seems that something is broken
Sep 23 08:33:05 <JaMa> alx_: otherwise you usually don't need qemu-native.. except ie external-toolchain which was pulled to build because of bug yesterday
Sep 23 08:33:39 <alx_> is it normal that meta isn't under git ?
Sep 23 08:33:45 <JaMa> http://lists.linuxtogo.org/pipermail/openembedded-core/2011-September/009986.html
Sep 23 08:34:19 <alx_> JaMa, I just try my first image : bitbake poky-image-minimal
Sep 23 08:34:22 <alx_> and this don't work.
Sep 23 08:34:44 <florian> good morning
Sep 23 08:35:07 <alx_> morning florian 
Sep 23 09:02:19 <pb_> hi all
Sep 23 09:11:40 <bluelightning> hi pb_, all
Sep 23 09:13:16 <florian> hi alx_, hi bluelightning 
Sep 23 09:16:37 <alx_> bluelightning, florian how to change from Xvesa to xserver-xorg in a build ?
Sep 23 09:16:54 <alx_> "image build" like qt4-x11-demo-image
Sep 23 09:17:55 <florian> alx_: It should be a setting in the device definition which xserver to use.
Sep 23 09:18:31 <tasslehoff> is it possible to make SRC_URI_myboard be "SRC_URI_beagleboard + mypatches". And is it a bad thing to do?
Sep 23 09:45:18 <alx_> florian, mmm where to for i686-generic ?
Sep 23 09:55:20 <florian> alx_: i686-generic.conf
Sep 23 09:55:44 <tasslehoff> is there a bitbake invocation that will show me the PV that would be built?
Sep 23 09:57:11 <alx_> florian PREFERRED_PROVIDER_virtual/xserver   ----    XSERVER ?
Sep 23 09:57:16 <alx_> how to change ?
Sep 23 09:59:11 <mckoan> good morning
Sep 23 10:02:18 <alx_> how to add a bsp to oe-core ?
Sep 23 10:03:48 <sanket> hello
Sep 23 10:04:10 <sanket> can anyone help me how to compile DOSEMU in OE...
Sep 23 10:06:22 <florian> alx_: There are two settings - one is the actual package to install and the other is the dependency to get it compiled. Check the other files, there should be some example
Sep 23 10:15:35 <ensc|w> how can I create a literal '${libdir}' string from within a .bb? E.g. for "echo 'Libs: -L${libdir}' >> x.pc" where ${libdir} is the 'libdir' Variable from .pc but not the expanded OE variable
Sep 23 10:17:40 <pb_> yeah, that is a bit tiresome.  the only way I found was to do something like:
Sep 23 10:17:47 <pb_> libstring="-L$"
Sep 23 10:17:57 <pb_> libstring="$libstring{libdir}"
Sep 23 10:18:05 <pb_> echo "Libs: $libstring" ...
Sep 23 10:18:27 <CIA-77> 03Nicu Pavel <npavel@mini-box.com> 07master * rff3282af48 10openembedded.git/recipes/linux/ (picosam9/defconfig picosam9_git.bb): (log message trimmed)
Sep 23 10:18:27 <CIA-77> Add linux kernel recipe for picoSAM9G45 board[1]
Sep 23 10:18:27 <CIA-77> This kernel is based on stock 2.6.32 with AT91 patches[2], Android patches[3]
Sep 23 10:18:27 <CIA-77> and board specific patches. Some of the board specific patches had been submited
Sep 23 10:18:27 <CIA-77> upstream[4]. Custom Android related patches won't be submited upstream. Future
Sep 23 10:18:28 <CIA-77> updates are planned to include patches from Android[3] and AT91 Linux tree[5].
Sep 23 10:18:29 <CIA-77> 1. http://www.mini-box.com/pico-SAM9G45-X
Sep 23 10:18:37 <pb_> which sucks, of course, but there doesn't appear to be any obvious better solution.
Sep 23 10:18:38 <CIA-77> 03Nicu Pavel <npavel@mini-box.com> 07master * r5477979116 10openembedded.git/conf/machine/picosam9.conf: 
Sep 23 10:18:38 <CIA-77> Add machine configuration for picoSAM9G45 board[1].
Sep 23 10:18:38 <CIA-77> More information about this board and software support can be found at [2].
Sep 23 10:18:38 <CIA-77> 1. http://www.mini-box.com/pico-SAM9G45-X
Sep 23 10:18:38 <CIA-77> 2. http://arm.mini-box.com/index.php?title=Main_Page
Sep 23 10:18:38 <CIA-77> Signed-off-by: Nicu Pavel <npavel@mini-box.com>
Sep 23 10:18:39 <CIA-77> Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
Sep 23 11:34:09 <kobe_> A recipe that I included in my image needs sources from the linux.org server, which is currently down. I downloaded the tar.gz file from a different server and saved it in the folder to which bitbake downloads all sources. Unfortunately, bitbake does not seem to use this tar.gz file. What can I do about this?
Sep 23 11:45:55 <ant_work> kobe_:  clean that foo recipe and create a fake foo.done in /downloads
Sep 23 11:50:28 <kobe_> ok, thanks
Sep 23 12:11:17 <tasslehoff> In oe-classic I use task-proper-tools.bb, but that seems to be gone now. Is there a replacement for it, or is the convenience-task gone?
Sep 23 12:26:23 <lautriv> heyda ;-)
Sep 23 12:30:35 <lautriv> i started to dig around a motorola hce700g ( kinda smartphone with barcodescanner ) but it's somewhat hard to get detailed information ( actually building oe/aurora with a slightly changed hx4700.conf ) can anyone provide more details about this machine ?
Sep 23 12:46:19 <lautriv> :-( failed to fetch util-linux-ng-2.18, even i had all files on latest version.
Sep 23 12:55:52 <alx_> any expert of yocto or oe-core ?
Sep 23 12:56:03 <woglinde> depends
Sep 23 12:56:06 <alx_> can't understand how to build core-image-minimal on emenlow
Sep 23 12:56:32 <woglinde> hm?
Sep 23 12:56:37 <woglinde> whats the error?
Sep 23 12:57:28 <alx_> mm well
Sep 23 12:57:47 <alx_> I'll tell you in a while .. removed all and started again 
Sep 23 12:59:18 <woglinde> o.O
Sep 23 13:02:31 <alx_> woglinde, try and error is a way :)
Sep 23 13:03:39 <lumag> hi all
Sep 23 13:05:33 <lumag> JaMa, I'm trying to debug your issues with gdb/glibc-thread-db. Do you have some time to help me?
Sep 23 13:11:28 <bluelightning> msm: btw: http://bugzilla.pokylinux.org/show_bug.cgi?id=1513
Sep 23 13:19:34 <tasslehoff> If I want to do "require foo.bb", must foo.bb be in the same directory?
Sep 23 13:20:01 <Crofton|work> no
Sep 23 13:20:29 <Crofton|work> but you may need the path relative to BBFILES or whatever is point at the base of your oe checkout
Sep 23 13:20:29 <JaMa> lumag: some yes
Sep 23 13:20:31 <tasslehoff> Crofton|work: ah, it's just that I need to give it a path, I suspect
Sep 23 13:20:36 <tasslehoff> yeah :)
Sep 23 13:20:40 <Crofton|work> just a little path :)
Sep 23 13:20:48 <Crofton|work> not the full path
Sep 23 13:21:15 <tasslehoff> I'm trying to require console-base-image.bb from my layer, but that's maybe bad practice?
Sep 23 13:21:29 <tasslehoff> my layer requiring something from the angstrom layer
Sep 23 13:21:37 <Crofton|work> i use include files from oe.dev in my layer
Sep 23 13:21:48 <Crofton|work> just need somelike recipes/foo/foo.inc
Sep 23 13:21:58 <woglinde> hm good my cortexa9 and mpfu vfp-d16 seems to work
Sep 23 13:22:34 <tasslehoff> cool
Sep 23 13:22:57 <tasslehoff> Crofton|work: ^, but probably woglinde ^ as well :)
Sep 23 13:23:22 * woglinde wonders why nobody yet made cortexa9
Sep 23 13:23:34 <woglinde> till later
Sep 23 13:24:13 <lumag> JaMa, It seems I've found at least one of the problems. Do you know if anybody uses eglibc-thread-db-nativesdk/glibc-thread-db-nativesdk
Sep 23 13:32:59 <JaMa> lumag: don't know.. I'm not using sdk at all
Sep 23 13:33:30 <lumag> JaMa, ok. I've found how to fix the problem, but I don't know why it does so.
Sep 23 13:35:04 <lumag> JaMa, Basically if I remove ${PKGSUFFIX} from offending glibc-thread-db RPROVIDES it works ok.
Sep 23 13:35:33 <lumag> btw: where does PREFERRED_PROVIDER_glibc-thread-db get from?
Sep 23 13:35:37 <JaMa> lumag: but then you will get multiple providers, don't you?
Sep 23 13:35:46 <lumag> no.
Sep 23 13:36:07 <JaMa> lumag: I've added it to tcmode-eglibc locally
Sep 23 13:36:14 <lumag> I mean just RPROVIDES_glibc-thread-db = "eglibc-thread-db"
Sep 23 13:36:16 <lumag> JaMa, ok
Sep 23 13:36:38 <JaMa> lumag: and same RPROVIDES from eglibc-nativesdk
Sep 23 13:36:51 <JaMa> lumag: that's why you will get multiple runtime providers IMHO
Sep 23 13:37:15 <lumag> that doesn't matter if it has the same value
Sep 23 13:38:13 <JaMa> do you have patch for me to test?
Sep 23 13:38:51 <lumag> Revert RP's revert patch and remove ${PKGSUFFIX} from glibc-thread-db lines (from the left and from the right).
Sep 23 13:42:34 <lautriv> is there actually a problem with oe and some mirrors ? i try to build a minimal image (console) but get errors on certain sources (unable to fetch)
Sep 23 13:49:31 <woglinde> lautriv you are aware that kernel.org is down since two weeks?
Sep 23 13:51:40 <alx_> two weeks ?
Sep 23 13:52:12 <pb_> seems like more than that
Sep 23 13:52:22 <lautriv> woglinde, kernel is not the problem but util-linux-ng and binutils
Sep 23 13:52:23 <lautriv> alx_, i guess more
Sep 23 13:53:41 <pb_> yeah, looks like a bit more than 3 weeks now
Sep 23 13:53:56 <lautriv> but feels like a half year
Sep 23 13:55:13 <pb_> oh well, those crazy kernel dudes
Sep 23 14:03:56 <ant_work> pb_: klibc is on the same server ;)
Sep 23 14:04:29 <pb_> heh
Sep 23 14:04:33 <pb_> yes, I guess it would be
Sep 23 14:07:12 <mckoan> in my final installed-package-sizes.txt I noticed a huge package
Sep 23 14:07:19 <mckoan> 58924   KiB /armv5te/libc6-dev_2.9-r21.11+svnr10153.8_armv5te.ipk
Sep 23 14:07:47 <mckoan> I think it is an error because libc6-dev isn't really required in a target fs
Sep 23 14:08:03 <JaMa> check depends.dot
Sep 23 14:10:06 <mckoan> JaMa: I find libc6 in depends.dot but not libc6-dev
Sep 23 14:11:13 <ant_work> pb_: for the impatient reviewer, I linked a mirror in my RFC
Sep 23 14:11:32 <mckoan> I wonder which recipe could be responsible for its inclusion
Sep 23 14:12:33 <mckoan> maybe recipes/eglibc/eglibc*
Sep 23 14:12:49 <mckoan> any hint please?
Sep 23 14:16:16 <ant_work> mckoan: one of those .txt from testlab should have dep graph
Sep 23 14:18:54 <mckoan> ant_work: any graph use depends.dot to create it
Sep 23 14:19:19 <mckoan> ant_work: in depends.dot there is any occurence of libc6-dev
Sep 23 14:19:39 <mckoan> so I argue it is not required and included in fs erroneously
Sep 23 14:19:59 <mckoan> but who's the guilty? :-D
Sep 23 14:20:20 <JaMa> you of course :)
Sep 23 14:21:00 <ant_work> I guess a blob of 58MiB will not pass unnoticed 
Sep 23 14:21:20 <JaMa> yes it's image or configuration issue IMHO
Sep 23 14:21:34 <mckoan> JaMa: hmmmm
Sep 23 14:22:20 <mckoan> maybe inherited by a classpath-dev inclusion in my image
Sep 23 14:22:36 * mckoan is checking
Sep 23 14:28:05 <woglinde> hehe
Sep 23 14:28:14 <woglinde> so no oe-core
Sep 23 14:29:16 <woglinde> Angstrom-console-image-uclibc-ipk-v2011.09-core-ac100.rootfs.tar.bz2
Sep 23 14:31:25 <mckoan> that's the problem: classpath-dev includes libc6-dev
Sep 23 14:31:35 <mckoan> thank you guys ;-)
Sep 23 14:32:39 <woglinde> ?
Sep 23 14:32:45 <woglinde> you found it out yourself
Sep 23 14:35:13 <JaMa> but it wasn't error, but feature :)
Sep 23 14:36:26 <mckoan> often talking with somebody helps ;-)
Sep 23 14:52:35 <ant_work> bbl
Sep 23 15:47:38 <CMoH> hmm... can i make a minimal build with bitbake? using a file provider
Sep 23 15:47:59 <CMoH> using bitbake -f or such
Sep 23 15:51:11 <CMoH> -f, --force           force run of specified cmd, regardless of stamp status
Sep 23 15:51:15 <CMoH> appears to not work :)
Sep 23 15:51:27 <msm> bluelightning: thanks for the note about the bitbake -D fix =)
Sep 23 15:51:49 <bluelightning> msm: no worries, should get merged soon I would think
Sep 23 15:55:07 <bluelightning> argh just found an issue :/
Sep 23 16:00:37 <bluelightning> CMoH: it works here... you need to combine it with -c
Sep 23 16:02:12 <CMoH> bluelightning, how are you using it?
Sep 23 16:03:48 <CMoH> bitbake -f -c build?
Sep 23 16:04:21 <bluelightning> CMoH: I don't think that will work, it doesn't force all dependent tasks to run, just the task you name
Sep 23 16:04:26 <bluelightning> build doesn't do anything by itself
Sep 23 16:04:56 <CMoH> well but if i run all tasks with bitbake -f -c fetch -f -c unpack .... -f -c compile   
Sep 23 16:04:57 <bluelightning> my test was install but any of the individual tasks should work
Sep 23 16:05:11 <CMoH> compile gets to do a full rebuild 
Sep 23 16:05:38 <bluelightning> I think it's not really designed for what you're trying to do there
Sep 23 16:06:36 <CMoH> i'd like to get a minimal rebuild on the sources i'm changing in the development tree (which is a pain if any change calls for a full rebuild)
Sep 23 16:06:55 <CMoH> ty bluelightning, i've had this answer before :)
Sep 23 16:07:27 <CMoH> but thinking a bit out of the box, what would in theory be needed to achieve my goal?
Sep 23 16:07:59 <CMoH> can i symlink tmp/work/.../my-package to dev packages and start with the compile task onwards?
Sep 23 16:08:12 <khem> CMoH: you could use srctree
Sep 23 16:08:21 <CMoH> s|dev packages/~/src/my-package|
Sep 23 16:08:30 <CMoH> srctree?
Sep 23 16:09:59 <CMoH> aha!
Sep 23 16:10:03 <CMoH> i found the bbclass
Sep 23 16:13:12 <Keizer> Does OE have a kernel builder like OpenWRT does?
Sep 23 16:15:41 <CMoH> let's see how things work now with the srctree thingie
Sep 23 16:17:19 <khem> hmmm srctree is not in oe-core
Sep 23 16:17:21 <khem> hmm
Sep 23 16:17:35 <khem> Keizer: what do u mean
Sep 23 16:17:49 <khem> Keizer: for OE kernel is just like any other package
Sep 23 16:18:01 <khem> just a bit complex one
Sep 23 16:18:37 <khem> you do bitbake virtual/kernel assuming you have MACHINE set correctly it will figure out which recipe to use and built it for you
Sep 23 16:18:53 <khem> in theory it even doesnt have to be linux
Sep 23 16:21:46 <CMoH> i'm not on oe-core, but on 2011.03-maintenance
Sep 23 16:22:46 <khem> ok then you can explore it
Sep 23 16:25:59 <Keizer> Ok. The OpenWRT kernel setup let's you run make menuconfig and shows arm options in the CPU area
Sep 23 16:28:07 <khem> Keizer: bitbake -clisttasks virtual/kernel
Sep 23 16:28:22 <khem> will show you all the individual tasks you can do on kernel
Sep 23 16:28:29 <khem> one of them will be menuconfig
Sep 23 16:28:45 <Keizer> pwn
Sep 23 16:31:53 <khem> Keizer: one of these days I have plans to do a meta-openwrt which will built wrt usin OE insfra
Sep 23 16:32:31 <khem> and let openwrt users peek into OE technology if they want to 
Sep 23 16:32:44 <khem> all I need is that 25th hour
Sep 23 16:37:22 <CosmicPenguin> I'm not going to lie - if I get a 25th hour, I'm going to pre-reserve it for sleeping
Sep 23 16:39:03 <khem> oh sleeping so easy I can do it with closed eyes
Sep 23 17:14:31 <woglinde_> jo rp
Sep 23 17:18:48 <RP__> hi woglinde_
Sep 23 17:59:13 <lumag> I wander, why for gcc-cross-canadian-arm build I'm getting the /usr/local/oecore-x86_64-arm/sysroots/x86_64-oesdk-linux/usr/lib/lib/libiberty.a file (notice two lib)
Sep 23 18:00:15 <CMoH> can i have multiple commands executed by a single bitbake run? 
Sep 23 18:00:28 <CMoH> like bitbake -f -c compile -c package -c ... package
Sep 23 18:00:41 <CMoH> seems like only one of them gets the -f applied
Sep 23 18:00:55 <CMoH> btw, khem, thanks srctree is working
Sep 23 18:02:04 <woglinde_> -c package is nonsens
Sep 23 18:05:02 <CMoH> woglinde_, please explain :)
Sep 23 18:05:21 <CMoH> what tasks i need to run so that i get the same effect as running bitbake -c clean package && bitbake package
Sep 23 18:05:46 <CMoH> i've just looked at the log of tasks and added all
Sep 23 18:06:02 <woglinde_> you can only run -c command package
Sep 23 18:06:21 <CMoH> -c package my-package might have been a better description
Sep 23 18:06:34 <woglinde_> yyes
Sep 23 18:06:46 <CMoH> anyway, i'm trying to force all of them to execute
Sep 23 18:06:54 <CMoH> but seems only the first one gets the -f
Sep 23 18:07:09 <woglinde_> try multiple -f
Sep 23 18:07:33 <woglinde_> -f -c bla packagename1 -f -c bla packagename2
Sep 23 18:07:40 <woglinde_> or use subshell
Sep 23 18:07:47 <woglinde_> with bitbake ... && bitbake ..
Sep 23 18:08:00 <CMoH> i have these: bitbake -f -c compile -c package -c install -c blah my-lovely-package
Sep 23 18:08:15 <woglinde_> *sigh*
Sep 23 18:08:16 <CMoH> bitbake -f -c compile -f -c package -f -c install -f -c blah my-lovely-package
Sep 23 18:08:44 <woglinde_> and?
Sep 23 18:08:46 <CMoH> bitbake -f -c compile my-package && bitbake -f -c package my-package &&  ... (this is slow)
Sep 23 18:08:56 <CMoH> well, the first one only does the do_compile task
Sep 23 18:08:59 <woglinde_> not this slow
Sep 23 18:09:01 <CMoH> the second does the same
Sep 23 18:09:24 <CMoH> the last one does them all, but each after reparsing, re-resoving missing task queue deps, re-preparing queue etc
Sep 23 18:09:44 <CMoH> which makes it slow
Sep 23 18:09:57 <CMoH> does the subshell work again? :D:D
Sep 23 18:10:02 <CMoH> that would really be awesome
Sep 23 18:10:46 <woglinde_> ?
Sep 23 18:10:51 <woglinde_> foo && foo
Sep 23 18:10:54 <woglinde_> is using subshell
Sep 23 18:11:05 <CMoH> after bitbake 1.8 the subshell was dropped a bit
Sep 23 18:11:30 <woglinde_> I am not talking about bitbake subshell using
Sep 23 18:11:35 <CMoH> oh
Sep 23 18:11:41 <woglinde_> I am talking about normal shell uses subshell
Sep 23 18:11:44 <woglinde_> when doing &&
Sep 23 18:11:52 <CMoH> ah, yeah
Sep 23 18:11:59 <CMoH> that was the slow option
Sep 23 18:12:10 <woglinde_> but working one
Sep 23 18:13:50 <CMoH> so i take it the answer is "no, it doesn't work for multiple commands"
Sep 23 18:14:31 <CMoH> i'll take a look into bitbake ... unfortunately one considers optimizing the build when running out of time :D:D
Sep 23 18:15:05 <rob_w> hmmm 
Sep 23 18:15:24 <rob_w> still sounds like a simple bash problem ?
Sep 23 18:17:01 <CMoH> thanks for the interest and time, woglinde_ 
Sep 23 18:21:36 <woglinde_> ?
Sep 23 18:21:53 <woglinde_> your bitbake commands even with subshell using should be finsihed now
Sep 23 18:22:05 <rob_w> :-)
Sep 23 18:32:20 <tasslehoff> I have created a SRC_URI_append_testboard in x-load_git.bbappend that contains just one patch. When I try to compile x-load_git.bb it ends up with URL: 'git://gitorious.org/x-loader/x-loader.git;branch=master;protocol=gitfile://0001-my.patch.patch'
Sep 23 18:32:38 <CMoH> woglinde_, rob_w, i had the script ready since i started asking here
Sep 23 18:32:45 <tasslehoff> sorry for the lack of paste, but the error is one another computer that I can't copy/paste from :)
Sep 23 18:32:56 <CMoH> problem is it's just as slow as recompiling after running -c clean
Sep 23 18:33:29 <rob_w> what i dont understand is why want to call those commands one by one when without -c it just does them all if there missing ?
Sep 23 18:33:43 <CMoH> but they're not missing
Sep 23 18:33:56 <CMoH> ok, a bit more info
Sep 23 18:34:04 <CMoH> i'm trying to do a minimal compile for arm using my dev tree
Sep 23 18:34:06 <rob_w> please.. just for my reference
Sep 23 18:34:13 <CMoH> someone suggested to use inherit srctree for that
Sep 23 18:34:20 <CMoH> which works just fine for the first build
Sep 23 18:34:39 <CMoH> i want to run all commands after compile after that first build, after modifying sources in the dev tree
Sep 23 18:34:53 <CMoH> running bitbake -f -c compile my-package-with-srctree
Sep 23 18:35:00 <CMoH> does the minimal compile as expected
Sep 23 18:35:12 <CMoH> but from this point on i want to package it to install it on the running system
Sep 23 18:35:18 <rob_w> i ve to check the srctree thing some more.. that seems to be my missing link here
Sep 23 18:37:16 <CMoH> so i'd like to force (-f) the whole command chain from compile to build 
Sep 23 18:37:30 <CMoH> problem is -f only seems to apply properly to the first -c 
Sep 23 19:02:44 <likewise> gm
Sep 23 19:02:51 <florian> hi likewise 
Sep 23 19:03:10 <likewise> hi florian
Sep 23 19:38:03 <lautriv> ok, another thingie of more interrest............got anyone information when kernel.org may be back ?
Sep 23 19:39:27 <lautriv> sorry, wrong channel :-)
Sep 23 19:56:33 <lumag> bye!
Sep 23 22:16:38 <woglinde_> he ant
Sep 23 22:16:41 <woglinde_> and good nite
**** ENDING LOGGING AT Sat Sep 24 02:59:57 2011