**** BEGIN LOGGING AT Tue Aug 10 02:59:58 2010 Aug 10 03:00:20 any idea where I should untar the version I just made on my 32-bit machine and how to continue with my build? Aug 10 03:12:13 jkridner: copy xdctools_X.Y.Z into workdir (tmp/work/armv7a*/xdctools-*/xdctools-X.Y.Z) Aug 10 03:12:39 armv7a* or beagleboard*? Aug 10 03:12:41 edit xdctools.inc and remove "require ti-eula-unpack.inc" Aug 10 03:13:30 right, beagleboard* then Aug 10 03:14:02 k. I wasn't sure about the -X.Y.Z either, so thanks for that! Aug 10 03:14:12 you should have xdctools* directory in workdir already, just copy the extracted xdctools-X.Y.Z into that directory Aug 10 03:16:13 also, I think the new names are ti-xdc*, but I'm not sure of that. Aug 10 03:16:56 oops, I did until I did -c clean. Aug 10 03:17:16 the name from the recipe is ti-xdctools, but the one which gets extracted is just xdctools :) Aug 10 03:19:28 I did -c unpack to get back to the place where the unpack was meant to have occured. Aug 10 03:20:26 jkridner: tmp/work/beagleboard-*-linux-gnueabi/ti-xdctools-3.16.01.27-r2/xdctools_3_16_01_27 Aug 10 03:21:04 uh, do you have your _ and . swapped? Aug 10 03:29:00 installing 'ia32-libs' enables the installer to work! Aug 10 03:29:07 so, that is a much better work-around. Aug 10 03:29:28 great find! Aug 10 03:29:49 article: http://www.debian-administration.org/articles/531 Aug 10 04:19:17 hi Aug 10 04:21:13 i have problem with oe build Aug 10 04:21:47 build failed at do_rootfs for x11-image Aug 10 04:21:56 showing an error "opkg_install_cmd: Cannot install package locale-base-en-us." Aug 10 04:32:23 hiy Aug 10 04:32:49 I'm have an error due to local include when compiling gcc Aug 10 04:32:57 did anyone have an idea? Aug 10 04:57:02 i have problem with oe build Aug 10 04:57:15 | fixproto: populating `include-fixed' Aug 10 04:57:15 | CROSS COMPILE Badness: /usr/include in INCLUDEPATH: /usr/include Aug 10 04:57:15 | fix-header: Internal error: abort in add_path, at c-incpath.c:425 Aug 10 04:57:26 none see it? Aug 10 04:57:31 build failed at do_rootfs for x11-image showing an error "opkg_install_cmd: Cannot install package locale-base-en-us." Aug 10 04:57:36 any idea ? Aug 10 05:33:30 Tartarus: ref yesterdays discussion. my gcc rpath actually points to /data/src/lead-dev/openembedded/build/angstrom-dev/sysroots/i686-linux/usr/lib, which is in the tree where I compiled the sdk Aug 10 05:42:31 which bitbake version is recommended atm? Aug 10 06:03:25 When I have compiled an SDK, and "readelf -d gcc" shows an RPATH inside the tree I compiled in, isn't that odd? Doesn't seem very portable.. Aug 10 06:41:51 hmm... Aug 10 06:42:00 task-boot.bb sucks in PREFERRED_PROVIDER_virtual/update-alternatives Aug 10 06:42:11 which is probably opkg, but not necessarily Aug 10 06:42:51 so its nigh on impossible to get a rootfs image without opkg binaries Aug 10 06:43:04 even if ONLINE_PACKAGE_MANAGEMENT="none" Aug 10 06:43:43 grg: but distro can set something else to DISTRO_UPDATE_ALTERNATIVES Aug 10 06:44:24 JaMa|Wrk, yes, that is the way to get an image without opkg Aug 10 06:44:55 so why it sucks? you have to install at least one u-a provider Aug 10 06:45:28 and preferably not both (u-a-cworth and opkg) as it was before Aug 10 06:45:38 you have to really work hard to get an image without opkg Aug 10 06:46:16 are postinst called even with ONLINE_PACKAGE_MANAGEMENT="none"? Aug 10 06:46:50 forgive me, i don't know what that means :) Aug 10 06:47:33 i've been trying to build minimal/minimal-image with ONLINE_PACKAGE_MANAGEMENT="none" and it certainly results in opkg binaries in the image Aug 10 06:48:12 the compiled SDK apparently works only on my system.. Aug 10 06:49:11 grg: I mean postinst scripts from packages, because always after reflashing image and first opkg call I see lot of those executed (and often setting alternatives) Aug 10 06:49:29 grg: no idea how it works/doesn't work if you have ONLINE_PACKAGE_MANAGEMENT="none" Aug 10 06:50:02 JaMa|Wrk, there appear to be no calls to opkg in /etc/init.d with my minimal-image Aug 10 06:50:21 so presumably this would break all postinst scripts, yes Aug 10 06:51:13 khem: I forgot to mention yesterday, that those alignment traps happen only on armv4t (armv5 with same gcc and eglibc doesn't have those) I'm rebuilding armv5 image with same distribution (SHR instead of Angstrom) to confirm that it's not because my armv5 image is built with slightly different CFLAGS Aug 10 06:51:46 grg: ah than maybe you don't need u-a provider so much as nobody will call it automatically :) Aug 10 06:52:53 I'm have an error due to local include when compiling gcc gcc-4.3.3/build.x86_64-linux.sh4-angstrom-linux Aug 10 06:53:02 JaMa|Wrk, so i could make DISTRO_UPDATE_ALTERNATIVES dependent upon ONLINE_PACKAGE_MANAGEMENT!="none" ? Aug 10 06:53:08 | fixproto: populating `include-fixed' Aug 10 06:53:08 | CROSS COMPILE Badness: /usr/include in INCLUDEPATH: /usr/include Aug 10 06:53:08 | fix-header: Internal error: abort in add_path, at c-incpath.c:425 Aug 10 06:53:14 noone see this? Aug 10 06:54:28 j24, i haven't used sh4, nor have i noticed anyone else using it... Aug 10 06:54:36 j24, perhaps try the mailing list? Aug 10 06:54:39 grg: no strong opinion about this one.. but maybe you would like some u-a provider even with ONLINE_PACKAGE_MANAGEMENT == none (to maintain alternatives already installed for you) Aug 10 06:55:31 grg: but I agree that u-a-cworth will be better u-a provider in case ONLINE_PACKAGE_MANAGEMENT == none Aug 10 06:56:03 hmm... which edge cases do we support... :) Aug 10 06:58:47 grg: how gcc will include /usr/include Aug 10 06:59:02 j24, it shouldn't Aug 10 06:59:42 j24, paste the relevant parts of your build log along with your local.conf into an email and send it to the mailing list Aug 10 06:59:53 grg: add? Aug 10 07:00:04 grg: what is the adress? Aug 10 07:00:44 j24, http://wiki.openembedded.net/index.php/Mailing_lists Aug 10 07:01:03 j24, you want the openembedded-devel list Aug 10 07:01:20 grg: tks Aug 10 07:05:49 I need a bit of help on this. I compile an SDK in /data/src/lead-dev/openembedded, and the resulting gcc compiler for arm has an RPATH in /data/src/lead-dev/openembedded/build/angstrom-dev/sysroots/i686-linux/usr/lib, and will only work on my machine as long as that directory exists. Aug 10 07:07:29 the previous SDK I built also had this odd RPATH, but still worked on other systems, so something must have happened "under the hood" Aug 10 07:11:34 tasslehoff, ask on the mailing list. OE's toolchain guy is probably asleep by now Aug 10 07:11:54 grg: will do Aug 10 07:23:14 gm everyone Aug 10 07:23:44 morning Aug 10 07:25:25 morning hrw, mickey|office Aug 10 07:26:11 good morning eFfeM_work Aug 10 07:26:17 mickey|office: are you still interested in/working on/whatever in oe/openmoko ? Aug 10 07:27:58 eFfeM_work: i don't care at all about openmoko distro and the gtk+ based openmoko stuff; i still care about the om-gta* machines though. Aug 10 07:28:24 for me, the openmoko software side is completely over Aug 10 07:28:32 and no one is using any of the distros anymore Aug 10 07:28:46 people with OM devices have migrated to SHR or to one of the non-OE distros Aug 10 07:29:31 well the issue is that conf/distro/include/preferred-om-2008-versions.inc pins a lot of old recipes, some of which use legacy staging etc, blocking removal of them Aug 10 07:29:45 right Aug 10 07:29:47 yank it Aug 10 07:29:52 btw it also list recipes that are gone already Aug 10 07:29:57 yank the inc file ? Aug 10 07:30:02 no one is building OM 2007.x nor OM 2008.x nor OM 2009.x out of OE anymore Aug 10 07:30:10 the inc and the distro Aug 10 07:30:24 we can add an oe.error( "please use SHR" ), if someone choses openmoko Aug 10 07:31:11 what would be the best way procedurewise? Guess if I nuke it right away people won't be happy wrt the procedure Aug 10 07:31:17 guess it is best if you rm it Aug 10 07:31:42 I can do so as well, but guess in that case I have to RFC it and ask for at least two acks Aug 10 07:32:56 eFfeM_work: Acked-by: Marcin Juszkiewicz Aug 10 07:33:11 hehe Aug 10 07:33:11 eFfeM_work: Acked-by: Michael 'Mickey' Lauer Aug 10 07:33:11 eFfeM_work: mickey|office will give you second and you can remove in next 5 minutes ;D Aug 10 07:33:27 wanna 3rd? Aug 10 07:33:29 the one thing to remember Aug 10 07:33:41 openmoko is an example in lots of wikis Aug 10 07:33:46 so it would be better not to remove it altogether Aug 10 07:33:49 but add the error i talked about Aug 10 07:33:58 yep Aug 10 07:34:01 we did the same in a couple of other cases Aug 10 07:34:04 JaMa|Wrk: you also happy with it? Aug 10 07:34:09 eFfeM_work: sure Aug 10 07:34:28 like in conf/machine/corgi.conf Aug 10 07:34:39 right Aug 10 07:34:44 or the infamous husky Aug 10 07:34:49 iirc Aug 10 07:34:57 or boxer etc Aug 10 07:35:01 morning Aug 10 07:35:03 ah boxer it was Aug 10 07:35:06 morning soltys Aug 10 07:35:15 hej soltys Aug 10 07:35:18 I'll peek into it, not sure if I have the time to do so during the day, but otherwise will look at it tonight @ home Aug 10 07:37:53 while we're at it, are there other distro's/machines that should be retired ? Aug 10 07:38:33 most likely yes, but we can't do that until the actual maintainers pop up Aug 10 07:38:59 we could ask on the mailing list for something like "is this distro still being used?" and then wait 1 month for answers Aug 10 07:39:15 morning florian_kc Aug 10 07:39:51 good morning Aug 10 07:42:08 mickey|office: another good reason to have explicit maintainers (if not for the recipes then at least for distro and machine) Aug 10 07:43:02 but as long as a machine or distro does not pin things or block others, I have no problems to keep it around Aug 10 07:43:58 (maintainers) true Aug 10 07:49:19 did a quick check, several distro's do not have a maintainer listed Aug 10 07:49:54 will write an RFC to the ML to add maintainers to distro files (and probably machine and recipe as well) Aug 10 07:55:50 Digging a bit here. My SDKs gcc for arm says "error while loading shared libraries: libmpfr.so.4: cannot open shared object file" since the library can't be found in the RPATH. I tried specifying LD_LIBRARY_PATH without luck. Is it looking for an x86 or arm library? Aug 10 07:59:33 tasslehoff: if it gives you this message when you just run gcc its an x86 library missing Aug 10 08:03:19 florian: ok. my sdk only contains libmpfr.so.4 for arm, so I probably need to find out which ubuntu-package will give me that library then Aug 10 08:03:48 hm, I only have libmpfr.so.1 Aug 10 08:04:07 btw mickey|office as python wiz is there anything you can say about http://lists.linuxtogo.org/pipermail/openembedded-devel/2010-August/022659.html Aug 10 08:04:19 guess something similar applies to perl too Aug 10 08:09:40 eFfeM_work: i'll take a look Aug 10 08:09:53 tnx Aug 10 08:24:04 It seems difficult to get libmpfr.so.4 on Ubuntu 10.04. Can I somehow embed it in my SDK instead, and set LD_LIBRARY_PATH to look there as well? Aug 10 08:29:05 tasslehoff: you can grab libmpfr4 from ubuntu 10.10 Aug 10 08:29:15 shr people: found this in machine/shr.conf: Aug 10 08:29:16 # set default locale for images - we have to do that before Aug 10 08:29:16 # including angstrom-2007-for-openmoko.inc because otherwise Aug 10 08:29:16 # it will be already set to en-gb Aug 10 08:29:29 but angstrom-2007-for-openmoko.inc is not included Aug 10 08:30:41 guess the comment should be updated Aug 10 08:31:10 hrw: is it also a solution to RDEPEND on mpfr in my task-toolchain-host.bb? Aug 10 08:31:29 eFfeM_work: I'll remove it Aug 10 08:31:37 cool Aug 10 08:32:47 actually found it while checking if the file was still in use Aug 10 08:33:09 no-one is, but koen is listed as maintainer, so better consult him first Aug 10 08:34:33 no Aug 10 08:34:56 JaMa|Wrk: while we're at it: you asked "wanna 3rd" for the openmoko removal, guess it does not harm so yes,please Aug 10 08:35:15 hrw guess that was to me, was the no on removal or consulting koen? Aug 10 08:35:15 we used it too in SHR.. but the stuff pinned there was too old and mostly overwritten later by us.. so I removed it later Aug 10 08:35:43 eFfeM_work: I'll ack it as soon as I see the patch Aug 10 08:36:00 ok Aug 10 08:36:20 eFfeM_work: no it was to tasslehoff Aug 10 08:36:27 tasslehoff: no, it is not a solution Aug 10 08:36:32 ok Aug 10 08:38:17 hrw: ok. grabbed it from 10.10, and then libgmp.so.10 is the next one missing. the search goes on :) Aug 10 08:39:35 03Martin Jansa  07org.openembedded.dev * r652a4ea9ab 10openembedded.git/conf/distro/shr.conf: shr: cleanup distro config a bit Aug 10 08:41:33 and that one does not exist even in 10.10. hmm. Aug 10 08:42:56 tasslehoff: libgmp3c2: /usr/lib/libgmp.so.3.5.2 Aug 10 08:45:03 hrw: that one I have. should/could I just create a symlink? Aug 10 08:45:55 hrw: that worked, though I feel a bit dirty now :) Aug 10 08:46:16 tasslehoff: you cleanly do it wrong Aug 10 08:50:18 hrw: probably, but I don't know where/what? the gcc from my sdk wants a version of libmpfr that I can't get on Ubuntu 10.04, and I'm not sure why. Aug 10 08:51:11 I think that it should be from OE libmpfr-native Aug 10 08:52:57 hrw: meaning that when developing against the sdk I should use bitbake recipes that depend on libmpfr-native? sorry if I'm being dense here. Aug 10 08:54:46 you built sdk with OE - right? Aug 10 08:54:57 check did it built libmpfr-native during build Aug 10 09:02:59 openmoko removal patch sent to ML Aug 10 09:04:38 with all Acks? Aug 10 09:04:45 yes Aug 10 09:04:57 please have a look if the msg text is ok Aug 10 09:04:58 thx Aug 10 09:05:12 if JaMa|Wrk also acks it I'll push it tonight Aug 10 09:05:25 still not in my maildir Aug 10 09:06:20 give servers a bit of time Aug 10 09:06:21 it'll come. it is on the web already Aug 10 09:06:22 http://lists.linuxtogo.org/pipermail/openembedded-devel/2010-August/022680.html Aug 10 09:07:21 eFfeM_work: s/aug/Aug Aug 10 09:07:48 will fix Aug 10 09:08:12 "It is recommented to use DISTRO='shr' instead."? Aug 10 09:08:23 "It is recommended to use DISTRO='shr' instead."? Aug 10 09:08:45 actually wanted to add the git checkout cmd but my python was not good enough to get he } of the git command in the text Aug 10 09:08:50 hrw: fixed Aug 10 09:09:03 git has lot of manuals Aug 10 09:09:35 yup, that's why i did not spent too much time on this Aug 10 09:09:43 s/this/it/ Aug 10 09:12:43 good riddance to OM distro :-) Aug 10 09:13:02 ;D Aug 10 09:13:16 need to remember new phrase Aug 10 09:14:40 JaMa|Wrk: thanks for the ack, v2 with reworked text has been posted Aug 10 09:14:53 XorA|gone: do I recall correctly that Koen is on holiday? Aug 10 09:14:57 * XorA|gone digs a 6ft x 4ft hole in the ground and starts piling in old configs Aug 10 09:15:05 eFfeM_work: no idea Aug 10 09:15:08 eFfeM_work: yep Aug 10 09:15:23 eFfeM_work: as last OM maintainer you can add my ack to killing it :-) Aug 10 09:15:37 XorA|gone: ok Aug 10 09:16:02 it was only ever a rebadge of Angstrom anyway Aug 10 09:16:05 eFfeM_work: Koen on holidays == less problems with pushing bigger changes? :D Aug 10 09:17:10 if he is on hols we can burglarise and get his stash of omap boards Aug 10 09:17:27 XorA|gone: I would also like to kill angstrom-2007-for-openmoko.inc, no one is using it any more but it lists koen as maintainer so would like to have some blessing for removing it from the angstrom team (or should I better wait until he's back; it does not really harm, butnow I am on it and otherwise it might be forgotten Aug 10 09:17:51 hrw, do not seed exciting ideas :P Aug 10 09:17:53 eFfeM_work: you can nuke that safely Aug 10 09:17:54 naughty boy Aug 10 09:18:10 eFfeM_work: he created that file for me Aug 10 09:18:19 XorA|gone: I'll nuke it Aug 10 09:18:51 hrw: I can't find any libmpfr-native folders in my tmpdir, but I have an x86 version in "tmpdir/sysroots/i686-linux/usr/lib" where I compiled the sdk Aug 10 09:18:55 eFfeM_work: save this log to attach on Koen's scream in ML :) Aug 10 09:19:19 and wrt the boards: I know his physical address (unless he moved recently). ofc there must be a bounty before I reveal it Aug 10 09:19:35 Jay7: rikers bot will log all of this Aug 10 09:19:59 eFfeM_work: I think that many people know where Koen lives Aug 10 09:20:13 I am just too lazy to dig in emails ;D Aug 10 09:20:23 :-) Aug 10 09:20:53 his guinea pigs will be an issue, they are vicious :-D Aug 10 09:21:12 btw anyone who can help me or give me some advise on an autotools issue: http://lists.linuxtogo.org/pipermail/openembedded-devel/2010-August/022614.html ? Aug 10 09:21:47 * eFfeM_work is not scared of pigs, I was raised on a farm with plenty of pigs around, including vicious ones Aug 10 09:22:03 eFfeM_work: Guinea Pigs != Pigs Aug 10 09:22:08 just need a good ale fork or whatever the english name is Aug 10 09:22:33 eFfeM_work: http://www.youtube.com/watch?v=jI-ObHzkw78 Aug 10 09:22:37 real pigs are much bigger and the adult males can be very agressive too Aug 10 11:20:40 03Martin Jansa  07org.openembedded.dev * r5e69441e65 10openembedded.git/recipes/pidgin/pidgin_2.7.2.bb: Aug 10 11:20:40 pidgin: add 2.7.2 version Aug 10 11:20:40 Signed-off-by: Martin Jansa Aug 10 11:20:48 03Martin Jansa  07org.openembedded.dev * r27870493b2 10openembedded.git/recipes/xfsprogs/xfsprogs_3.1.2.bb: Aug 10 11:20:48 xfsprogs: add 3.1.2 version Aug 10 11:20:48 Signed-off-by: Martin Jansa Aug 10 12:10:34 * mwester reads the backlog and discovers the plot. Aug 10 12:15:40 03Martin Jansa  07org.openembedded.dev * rdb4786b7be 10openembedded.git/recipes/nano/ (nano-2.2.5/ncursesw.includedir.patch nano_2.2.5.bb): Aug 10 12:15:40 nano: add version 2.2.5 and fix for using system ncursesw Aug 10 12:15:40 Signed-off-by: Martin Jansa Aug 10 12:21:06 rwhitby: as we're about to terminate the openmoko distro, is there still a use case for mokoslug? or should that be terminated as well ? Aug 10 12:21:29 mwester: the plot thickens ;-) Aug 10 12:24:39 eFfeM_work: celinux-test gmustix mamona opennm openprotium wrt54oe etc will be next? Aug 10 12:25:15 eFfeM_work: celinux-test was done for CELF few years ago and probably never updated. Aug 10 12:25:54 hrw, if they are not used, why not, but actually I think some of them are still used (guess gumstix is, perhaps wrt54oe too) Aug 10 12:26:57 hrw: if they are used and don't block others I have no problems keeping them; reason for explictly asking about mokoslug was its description Aug 10 12:27:00 "The Linux Server companion for Openmoko phones" Aug 10 12:27:07 sure Aug 10 12:27:33 so w/o openmoko this could be pretty useless too Aug 10 12:28:31 xxs1500 machine... I had to remove it long time ago D: Aug 10 12:28:40 it is unbuildable since eons Aug 10 12:29:21 ;) Aug 10 12:29:40 hrw, then remove Aug 10 12:29:55 guess there are lots of other unused machines around, but they do not really harm Aug 10 12:30:35 will leave it. maybe one day someone will use it agan Aug 10 12:31:21 eFfeM_work, mokoslug can be removed as well. Aug 10 12:31:31 mwester: ok Aug 10 12:31:44 will take care of that tonight Aug 10 12:37:00 does anyone know why EXTRA_AUTORECONF = "--exclude=autopoint" is in autotools.bbclass? Aug 10 12:37:27 I'm wondering how the correct version of po/Makefile.in.in gets generated without it Aug 10 12:39:33 eFfeM_work: once I wondered about a matrix of working distro/machine Aug 10 12:39:55 at least of 'used' pairs Aug 10 12:40:09 it seems some recipes include: Aug 10 12:40:10 do_configure_prepend () { autopoint --force Aug 10 12:40:10 } Aug 10 12:40:18 guess that is what I need to do ... Aug 10 12:40:47 ant_work: use a big system and something like for i in machinelist ; do ; MACHINE=$i bitbake x11-image;done or whatever image you want to use as ref Aug 10 12:41:10 I mean we could collect the autobuilder outputs and populate that matrix Aug 10 12:41:27 but we would have' image-level' outputs Aug 10 12:41:35 you mena the angstrom autobuilder or the tinderbox info Aug 10 12:41:42 yes Aug 10 12:41:52 i seem to recall tinderbox had some issues last time I wanted to search for something Aug 10 12:42:03 got an sql error on my pacakge search Aug 10 12:42:17 eg minimal-simpad OK, minimal-beagleboard KO, ... Aug 10 12:42:41 at least referring to a known image Aug 10 12:43:13 could be useful but is a lot of work Aug 10 12:43:14 smthg like 'minimal-console-image-simpad' Aug 10 12:44:24 eFfeM_work: terminate it too Aug 10 12:44:33 not oe related, but hey this is the oe lounge so lots of geek talk fits here: anyone aware of an SSD > 80 GB with SLC chips in it (and no indilinx controller) Aug 10 12:44:36 rwhitby: thanks, will do Aug 10 12:55:53 mickeyl: ping Aug 10 13:57:56 it appears that packages are no longer allowed to overwrite files that other packages install at the rootfs stage Aug 10 14:00:02 my gdb-cross-sdk problem was the destination sdk already installed - remote it, then it builds fine. Aug 10 14:00:05 cbrake: yeah, I think opkg was patched to correctly diagnose that a few weeks ago Aug 10 14:00:20 s/remote/remove - yawn. Aug 10 14:02:30 pb_: that closes a convenient way to modify config files in existing packages without modifing a bunch of packages and tracking all those modifications Aug 10 14:02:36 * cbrake thinks on this some more ... Aug 10 14:03:58 cbrake: if you declare your package as Replaces: then it will still be allowed to do that Aug 10 14:09:41 pb_: that sounds great -- where do you specifiy "Replaces" ? Aug 10 14:10:07 cbrake: RREPLACES_${PN} in your .bb file Aug 10 14:10:17 pb_: ahh, excellent -- thanks Aug 10 14:48:49 hi mickey|office Aug 10 14:49:09 morning pb_ Aug 10 16:05:35 03Frans Meulenbroeks  07org.openembedded.dev * r07b7b39348 10openembedded.git/conf/distro/ (include/preferred-om-2008-versions.inc openmoko.conf): (log message trimmed) Aug 10 16:05:35 openmoko: removed distro Aug 10 16:05:35 As discussed on #oe on Aug 10, 2010. Aug 10 16:05:35 Removed because this is not used or maintained any more. Aug 10 16:05:35 A message Aug 10 16:05:36 * Aug 10 16:05:36 * Aug 10 16:06:53 03Frans Meulenbroeks  07org.openembedded.dev * r24ed3287fe 10openembedded.git/conf/distro/mokoslug.conf: Aug 10 16:06:53 mokoslug.conf: removed distro Aug 10 16:06:53 approved by mwester and rwhitby on irc Aug 10 16:06:53 Signed-off-by: Frans Meulenbroeks Aug 10 16:07:17 RIP mokoslug Aug 10 16:10:15 03Frans Meulenbroeks  07org.openembedded.dev * r335234b673 10openembedded.git/conf/distro/include/angstrom-2007-for-openmoko.inc: Aug 10 16:10:15 angstrom-2007-for-openmoko.inc: removed Aug 10 16:10:15 was not used any more. Aug 10 16:10:15 XorA said on #oe that this was created for him, and that it could go Aug 10 16:10:15 Signed-off-by: Frans Meulenbroeks Aug 10 16:10:59 RP, actually my slug is almost RIP too, I think it is more than half a year ago since it was powered Aug 10 16:26:17 I need to find time to power m sheevaplug back again Aug 10 16:27:56 * kergoth_ too Aug 10 16:31:01 each time when I launch 'git gui' I wonder why Tk is so ugly Aug 10 16:31:12 agreed Aug 10 16:33:48 well, it's 20 years old. I think the state of the art in gui toolkits has changed a bit in the past two decades. :-} Aug 10 16:34:29 eFfeM, slugs never die. Their PSU's merely fail. :) Aug 10 16:34:59 mwester: true, but the xscales in it are declared EOL Aug 10 16:35:17 i have no plans to abandon my slug Aug 10 16:36:27 "apt-get install tk8.5;update-alternatives --config wish" + wish8.5 == a bit better Aug 10 16:51:51 bye Aug 10 16:53:12 ls Aug 10 17:32:32 03Frans Meulenbroeks  07org.openembedded.dev * r3cd6729fe7 10openembedded.git/MAINTAINERS: Aug 10 17:32:33 MAINTAINERS: removed openmoko distro maintainer as openmoko has been terminated Aug 10 17:32:33 Signed-off-by: Frans Meulenbroeks Aug 10 17:45:05 03Frans Meulenbroeks  07org.openembedded.dev * r5db1701988 10openembedded.git/recipes/alsa/ (7 files): Aug 10 17:45:05 alsa: removed old versions Aug 10 17:45:05 Signed-off-by: Frans Meulenbroeks Aug 10 17:46:38 03Frans Meulenbroeks  07org.openembedded.dev * r7fcd85506e 10openembedded.git/recipes/abiword/ (3 files): Aug 10 17:46:38 abiword: removed 2.5.2; 3 years old and default pref -1 Aug 10 17:46:38 Signed-off-by: Frans Meulenbroeks Aug 10 17:52:16 03Frans Meulenbroeks  07org.openembedded.dev * rfc3233b3a2 10openembedded.git/recipes/autoconf/ (15 files in 3 dirs): Aug 10 17:52:16 autoconf: remove a few old unused versions Aug 10 17:52:16 Signed-off-by: Frans Meulenbroeks Aug 10 17:54:16 03Frans Meulenbroeks  07org.openembedded.dev * rb48cd835b4 10openembedded.git/recipes/zziplib/ (zziplib_0.10.82.bb zziplib_0.12.83.bb): Aug 10 17:54:17 zziplib: remove old versions Aug 10 17:54:17 Signed-off-by: Frans Meulenbroeks Aug 10 17:56:33 * mwester imagines effeM's home; a spartan place with all non-functional or not-currently-used items absent completely... :p :D Aug 10 17:58:32 that looks like good image.. Aug 10 17:59:17 * JaMa doesn't like stuff which looks functional and after few hours waiting for miracles to happen you discover that it's long dead :) Aug 10 17:59:51 Ah, but there is a place in life for non-functional things, too. I appreciate my dog, for example. Aug 10 18:00:30 and he has his place as long as you appreciate him Aug 10 18:00:57 mwester: don't even dare to suggest that to my wife, she'll make me clean up for at least a week Aug 10 18:01:02 too much hw around the compu Aug 10 18:01:13 but after you as his owner lost interest and stop feeding him .. who knows how long it can be still considered a dog.. Aug 10 18:01:32 and at some point he becomes smelly Aug 10 18:01:38 away for coffee break Aug 10 18:36:57 03Stanislav Brabec  07org.openembedded.dev * rcdf13c967f 10openembedded.git/recipes/initscripts/ (files/functions initscripts_1.0.bb): Aug 10 18:36:57 initscripts: Fixed empty machine_id returned on some devices. Aug 10 18:36:57 Signed-off-by: Stanislav Brabec Aug 10 18:38:27 03Christian Rüb  07org.openembedded.dev * ref6107ce7e 10openembedded.git/recipes/openmoko-3rdparty/advancedcaching_git.bb: Aug 10 18:38:27 advancedcaching: new version 0.6.1.5 Aug 10 18:38:27 Signed-off-by: Christian Rüb Aug 10 18:38:27 Signed-off-by: Martin Jansa Aug 10 18:38:27 03Martin Jansa  07org.openembedded.dev * r75cbe0974c 10openembedded.git/recipes/shr/elementary-theme-gry_git.bb: Aug 10 18:38:28 elementary-theme-gry: bump SRCREV Aug 10 18:38:29 Signed-off-by: Martin Jansa Aug 10 18:45:43 huhu Aug 10 18:49:43 re Aug 10 18:49:59 03Frans Meulenbroeks  07org.openembedded.dev * rb1fbc3fd45 10openembedded.git/recipes/zeroconf/ (3 files in 2 dirs): Aug 10 18:49:59 zeroconf 0.6.1: removed (5 years old) Aug 10 18:49:59 Signed-off-by: Frans Meulenbroeks Aug 10 18:50:09 03Frans Meulenbroeks  07org.openembedded.dev * r6c51e308c7 10openembedded.git/recipes/zeroconf/files/ (4 files): Aug 10 18:50:09 zeroconf: renamed files to zeroconf-0.9 Aug 10 18:50:09 Signed-off-by: Frans Meulenbroeks Aug 10 18:51:38 mickeyl: JaMa, hrw|gone, XorA|gone, was wonderig if after nuking openmoko distro there is still room for the openmoko* recipes Aug 10 18:53:17 yes, there is Aug 10 18:53:30 most of them are not bound to openmoko hardware Aug 10 18:53:30 ok, was just wondering Aug 10 18:53:47 I wrote recipes for most of om2007.2 ones Aug 10 18:53:48 true (but the name is somewhat misleading in that case) Aug 10 18:53:58 and many from them are also used by SHR Aug 10 18:54:25 eFfeM: no, they were wrote for openmoko company and have such names Aug 10 18:54:33 Openmoko-contacts for example Aug 10 18:54:39 ah ok Aug 10 18:54:49 * eFfeM does not know too much about the openmoko history Aug 10 18:55:01 * JaMa is usually putting new stuff (originally writen with openmoko phones in mind) to openmoko-3rdparty Aug 10 19:02:41 03Jason Kridner  07org.openembedded.dev * r7ac33686b5 10openembedded.git/recipes/beagleboard/beagleboard-test-scripts_git.bb: Aug 10 19:02:41 beagleboard-test-scripts: added new scripts Aug 10 19:02:41 testneon: runs burn-neon for 10 seconds Aug 10 19:02:41 testdsp: runs loopgpp dsplink example Aug 10 19:02:41 testmem: runs memtester Aug 10 19:02:42 Signed-off-by: Jason Kridner Aug 10 19:02:42 Signed-off-by: Denys Dmytriyenko Aug 10 19:02:44 03Jason Kridner  07org.openembedded.dev * r3ebb92b7f1 10openembedded.git/recipes/beagleboard/beagleboard-test-scripts_git.bb: Aug 10 19:02:44 beagleboard-test-scripts: updated to latest Aug 10 19:02:44 Signed-off-by: Jason Kridner Aug 10 19:02:45 Acked-by: Paul Menzel Aug 10 19:02:45 Signed-off-by: Denys Dmytriyenko Aug 10 19:20:15 03Frans Meulenbroeks  07org.openembedded.dev * r7f1a5e873d 10openembedded.git/recipes/cdparanoia/cdparanoia_9.8alpha.bb: Aug 10 19:20:15 cdparanoia_9.8alpha.bb: removed old version Aug 10 19:20:15 Signed-off-by: Frans Meulenbroeks Aug 10 19:20:33 03Simon Busch  07org.openembedded.dev * re9fdbfcbba 10openembedded.git/recipes/tzdata/tzdata_2010j.bb: Aug 10 19:20:33 tzdata: bump version to 2010k Aug 10 19:20:33 Signed-off-by: Simon Busch Aug 10 19:20:33 Signed-off-by: Roman I Khimov Aug 10 19:20:34 03Simon Busch  07org.openembedded.dev * rac61355f70 10openembedded.git/recipes/tzcode/tzcode-native_2010j.bb: Aug 10 19:20:34 tzcode-native: bump version to 2010k Aug 10 19:20:35 Signed-off-by: Simon Busch Aug 10 19:20:35 Signed-off-by: Roman I Khimov Aug 10 20:22:27 03Frans Meulenbroeks  07org.openembedded.dev * re69f211928 10openembedded.git/recipes/dpkg/ (7 files in 2 dirs): Aug 10 20:22:27 dpkg: remove old unpinned versions Aug 10 20:22:27 Signed-off-by: Frans Meulenbroeks Aug 10 20:47:55 03Koen Kooi  07org.openembedded.dev * rcafabdb4f7 10openembedded.git/conf/distro/include/angstrom-2008-preferred-versions.inc: Aug 10 20:47:55 Revert "angstrom: jpeg 8b doesn't build, prefer 6b for the time being" Aug 10 20:47:55 it seems to build again and Martin has sorted out the upgrade path issues on other commits Aug 10 20:47:55 This reverts commit 6b099fc3110b76c260c3adafe6f8960e27e42d8b. Aug 10 20:55:51 khem: would it be of any help if I'd prune the klibc recipes and just keep last one? Aug 10 20:57:41 ant_: yes unused ones can be deleted Aug 10 20:57:49 ant_: I owe you the fix Aug 10 20:57:52 ok, atm absolutely nobody using klibc apart our kexecboot things Aug 10 20:58:06 ant_: ok Aug 10 20:58:10 heh, sorry but seems beyond my poor skills ;) Aug 10 20:58:44 I'll clean the old recipes but the staging hell is up to you :p Aug 10 21:00:18 fwiw dietlibc segfaults on do_compile.. heh, this *libc are sons of lesser gods Aug 10 21:02:16 ant_: http://pastebin.com/QnFeH94x Aug 10 21:02:19 try this one Aug 10 21:05:03 ok, but I already tried it w/out success Aug 10 21:05:38 klibc headers are lost Aug 10 21:06:32 I even tried oe_runmake 'DESTDIR=${D}' install Aug 10 21:06:36 same Aug 10 21:12:24 khem: no klcc, no headers, misplaced klibc.so Aug 10 21:14:24 ant_: thats strange Aug 10 21:14:31 let my tree build Aug 10 21:14:37 and then I will experiment Aug 10 21:15:06 try kexec-tools-klibc-static as test against klibc, pls Aug 10 21:16:14 it is like the sysroot phases are skipped Aug 10 21:16:34 I mean, comparing the logs of the old/new Aug 10 21:17:10 khem, there is an isystem patch..perhaps this? Aug 10 21:18:21 seems not the case Aug 10 22:23:25 Is there a way to build libgles (for omap3530) without X11? Older libgles .bb built for raw/non-x11 Aug 10 22:32:32 beagleboard Buffer I/O error?? Aug 11 01:19:00 hmmm Aug 11 01:19:12 * kergoth wonders if anyone had any thoughts on his bitbake 'logging' branch Aug 11 01:35:35 khem: ping Aug 11 01:35:51 khem: I wanna make a bugfix to gcc-configure-cross.inc for relocation, bump prs or no? Aug 11 01:47:26 Tartarus: what sort of fix it is Aug 11 01:50:04 Tartarus: and I think its modifying the binaries so PR bump is needed I think Aug 11 01:54:29 So all of the PRs? ug Aug 11 01:54:38 :( Aug 11 01:54:48 And dropping --with-local-prefix and --with-gxx-include-dir Aug 11 01:54:52 what is the change btw Aug 11 01:55:00 from EXTRA_OECONF_PATHS Aug 11 01:55:18 seems ok Aug 11 01:55:21 yeah Aug 11 01:55:25 its well tested here Aug 11 01:55:29 pushed it to pb's branch iirc Aug 11 01:55:41 ok Aug 11 01:55:52 ran into it again doing relocation tests on upstream, i'll do the PR dance tonight or tomorrow then, thanks Aug 11 01:56:03 sure Aug 11 01:56:47 Tartarus: I guess SDK needs love too wrt to relocatability Aug 11 02:04:24 Yes Aug 11 02:04:35 It's just a bit more difficult since there's some other hard-coded things around Aug 11 02:05:05 yeah SDK I plan to look into it seriously Aug 11 02:05:09 in coming days Aug 11 02:09:06 03Tom Rini  07org.openembedded.dev * r15ce294f13 10openembedded.git/recipes/gcc/ (18 files): Aug 11 02:09:06 gcc-configure-cross.inc: Drop --with-local-prefix and --with-gxx-include-dir Aug 11 02:09:06 We don't need these paths to be set and they further break things if Aug 11 02:09:06 we use --sysroot rather than -isystem Aug 11 02:09:06 Signed-off-by: Tom Rini **** ENDING LOGGING AT Wed Aug 11 02:59:57 2010