**** BEGIN LOGGING AT Wed Jun 16 02:59:56 2010 Jun 16 06:20:03 morning Jun 16 06:35:49 hi, someone seeing dependency loops with current oe.dev? Jun 16 07:42:08 morning Jun 16 07:55:25 good morning Jun 16 08:08:49 03Rui Miguel Silva Seabra  07org.openembedded.dev * r7764d0b251 10openembedded.git/recipes/json-c/json-c_0.9.bb: Jun 16 08:08:49 json-c: add recipe for json C library Jun 16 08:08:49 Signed-off-by: Martin Jansa Jun 16 08:15:34 03Robert Schuster  07org.openembedded.dev * r588548bdc6 10openembedded.git/recipes/llvm/ (llvm2.7/link.patch llvm2.7_2.7.bb): llvm2.7 2.7: Removed the link patch - its not needed for the CMake-build Jun 16 08:25:10 JaMa: someone has reported something about loop but about 2-3 days ago Jun 16 08:26:09 hi Jun 16 08:28:25 woglinde: morning Jun 16 08:30:23 Jay7: it was gnu-config, reported today Jun 16 08:31:40 03Martin Jansa  07org.openembedded.dev * rfe6e787870 10openembedded.git/recipes/ (libjson/libjson_0.6.bb json-c/json-c_0.9.bb): Jun 16 08:31:40 json-c: move and rename to libjson as older version was Jun 16 08:31:40 * also convert old version to new staging Jun 16 08:31:40 Signed-off-by: Martin Jansa Jun 16 08:52:40 he zecke Jun 16 09:11:18 "ld: cannot find -lgcc_eh" - someone remembers that? Jun 16 09:11:37 03Graeme Gregory  07org.openembedded.dev * r102ef5fb54 10openembedded.git/recipes/linux/linux.inc: Jun 16 09:11:37 linux.inc : remove gcc 4.2.1 "fix" that breaks all modern kernel compiles Jun 16 09:11:37 modern avr32 kernels really need that -mno-pic Jun 16 09:11:49 03Graeme Gregory  07org.openembedded.dev * rd93a88d73c 10openembedded.git/recipes/binutils/ (2 files): binutils_2.18.atmel.1.0.1.bb : add latest binutils for AVR32 Jun 16 09:14:08 hrw hm did you rebuild gcc? Jun 16 09:14:46 damn gcc-4.3.1 has no easy atomic support Jun 16 09:14:59 woglinde: I am working on linaro cross toolchain and got it in 1st stage eglibc Jun 16 09:15:04 khem: hi, how things? Jun 16 09:15:18 khem crashed Jun 16 09:15:24 woglinde: so my gcc do not have libgcc_eh yet Jun 16 09:16:15 but I think I know where to look now Jun 16 11:23:31 hi Jun 16 11:23:36 some python guru here Jun 16 11:23:49 I want to edit a %s in place and cutr something Jun 16 11:27:58 khem: hm I don't know. I was just wondering in how far linuxfoundation associates with meego. maybe it's just image / marketing related Jun 16 11:28:14 woglinde: ??? Jun 16 11:28:21 pong! Jun 16 11:28:46 I don't understand :) Jun 16 11:28:54 "edit a %s in place and cutr something" Jun 16 11:34:29 akheron I found what I need Jun 16 11:34:33 but thanks for help Jun 16 11:34:41 :D Jun 16 11:35:27 every time I coworker asks me to look at something, he figures it out before I manage to walk the 5 meters between us Jun 16 11:35:37 I didn't know this works over IRC, too Jun 16 11:35:44 hehe Jun 16 11:45:41 anyone experience with quilt snapshot ? if I do that I get rm: cannot remove `.pc/.snap/': Is a directory Jun 16 11:45:41 even if I removed the directory just before it and no snapshot data is stored; after making changes quilt diff --snapshot does not give a diff Jun 16 12:34:48 03Martin Jansa  07org.openembedded.dev * r770b4b5d0f 10openembedded.git/recipes/navit/navit.inc: Jun 16 12:34:48 navit: add STAGING_DIR_NATIVE prefix to moc_location returned by pkg-config Jun 16 12:34:48 Signed-off-by: Martin Jansa Jun 16 12:39:09 03Dmitry Eremin-Solenikov  07org.openembedded.dev * ref1b61cdaa 10openembedded.git/recipes/opkg/ (opkg-nogpg-nocurl_svn.bb opkg-nogpg_svn.bb opkg_svn.bb): Jun 16 12:39:09 Revert "opkg, opkg-nogpg, opkg-nogpg-nocurl: make ${PN}-dev provide virtual/opkg-dev" Jun 16 12:39:09 This reverts commit aa45596db92a18851de6f2f241e578047f488307. Jun 16 12:39:09 Per request from Koen Kooi. Jun 16 12:39:10 Signed-off-by: Dmitry Eremin-Solenikov Jun 16 12:39:10 Acked-by: Graeme Gregory Jun 16 12:39:11 Acked-by: Philip Balister Jun 16 12:39:11 03Dmitry Eremin-Solenikov  07org.openembedded.dev * rc8f11e5a55 10openembedded.git/recipes/tasks/task-sdk-base.bb: Jun 16 12:39:12 Revert "task-sdk-base: replace ipkg-dev with virtual/opkg-dev" Jun 16 12:39:12 This reverts commit eeffe0a6eee97604679cf9a3cf40ed4604ae6493. Jun 16 12:39:13 Per request from Koen Kooi, as this breaks task-sdk-native, etc. Jun 16 12:39:13 Signed-off-by: Dmitry Eremin-Solenikov Jun 16 12:39:14 Acked-by: Graeme Gregory Jun 16 12:39:14 Acked-by: Philip Balister Jun 16 12:39:15 03Dmitry Eremin-Solenikov  07org.openembedded.dev * re8e5a9a02a 10openembedded.git/: Merge branch 'org.openembedded.dev' of git://git.openembedded.org/openembedded into org.openembedded.dev Jun 16 12:42:16 03Dmitry Eremin-Solenikov  07org.openembedded.dev * rb6f550fa7d 10openembedded.git/recipes/tasks/task-sdk-base.bb: Jun 16 12:42:16 task-sdk-base: drop ipkg-dev from list of packages Jun 16 12:42:16 As per discussion on oe-devel, drop ipkg-dev from list of task-sdk-base deps. Jun 16 12:42:16 Signed-off-by: Dmitry Eremin-Solenikov Jun 16 13:16:43 mickey|wm, ping Jun 16 13:16:44 hi Jun 16 13:16:55 oops wrong channel Jun 16 13:30:03 03Koen Kooi  07org.openembedded.dev * r54ca80948f 10openembedded.git/recipes/gnu-config/gnu-config_git.bb: gnu-config git: don't use git since git-native needs gnu-config-native. Switch to angstrom sourcemirror for http fetching Jun 16 14:24:43 hi, anyone a pointer or suggestion how to set a kconfig var depending on a choice? tried select, that works in menuconfig if I select it, but my configure then runs make oldconfig and that resets my var again Jun 16 14:25:35 select isn't a dependency, it's a way to force setting one option from another when its set Jun 16 14:25:50 i'm certain there are many examples of dependency in the kernel sources Jun 16 14:26:10 basically what I want is set another value if a choice is made Jun 16 14:26:32 doc is somewhat terse, didn't find a good example yet Jun 16 14:26:45 i thought thats what select was, *shrug* Jun 16 14:27:29 yes and select works for me if I seet the value in menuconfig, but when my do_configure then runs make oldconfig it resetst the damned var, even though my defconfig has the right value Jun 16 14:28:06 weird. Jun 16 14:29:24 http://www.xkcd.com/754/ Jun 16 14:29:50 (the mouse-over is spot-on, too. :) ) Jun 16 14:30:59 kergoth: think I got it just config XYX followed by def_bool y that makes it a bool with default value y and as it has no text it is not shown nor selected Jun 16 14:31:02 exactly what I need Jun 16 14:31:04 morning Jun 16 14:31:09 yo Jun 16 14:31:09 ah Jun 16 14:32:02 still learning something new every day Jun 16 14:32:17 unfortunately at my age you are also forgetting something old every day :-) Jun 16 14:33:24 Anyone else want to ack the linux-libc-headers-native thing? Jun 16 14:33:33 esp anyone that had comments the first time Jun 16 15:06:15 hi woglinde Jun 16 15:10:25 meh Jun 16 15:16:42 hi kergoth Jun 16 15:17:16 hi woglinde thebohemian Jun 16 15:17:17 hi all Jun 16 15:17:22 hi stefan_schmidt Jun 16 15:17:24 hi stefan Jun 16 15:17:30 hi stefan_schmidt Jun 16 15:17:31 hi woglinde Jun 16 15:17:45 thebohemian: for the gcc patch just check it into the overlay Jun 16 15:17:56 woglinde, I asked something yesterday, I'll re-ask : it was about navit Jun 16 15:18:13 hi GNUtoo|laptop Jun 16 15:18:19 is libnavit_la_LDFLAGS defined somewhere? Jun 16 15:18:44 gnutoo ask in #navit Jun 16 15:18:52 ok thanks Jun 16 15:34:42 hmm, RP around lately? haven't seen him on gtalk/jabber since getting back from vacation Jun 16 15:37:25 DJWillis: ping Jun 16 15:37:55 Jay7: pong Jun 16 15:38:13 DJWillis: did you know who made boards for pandora? Jun 16 15:38:42 who is manufacturer i mean Jun 16 15:39:20 Jay7: made? Design is in-house, fab is a mix of a company in China and one in Texas. I have the names but not sure if it is public or not (i'll ask). Jun 16 15:39:25 Good morning all Jun 16 15:40:01 DJWillis: ok, np :) Jun 16 15:40:16 what do I need to build to get libXxf86vm library? Jun 16 15:53:26 lol Jun 16 15:53:29 spain lost Jun 16 16:05:24 hi DJWillis , I commited a new scummvm some time ago Jun 16 16:20:33 cbrake, you here? Jun 16 16:20:33 03Stefan Schmidt  07org.openembedded.dev * r90b7b9a709 10openembedded.git/recipes/u-boot/ (files/bug20/bug-uboot.patch u-boot_git.bb): Jun 16 16:20:33 u-boot_git: Cleanup and refresh bug20 patch. Jun 16 16:20:33 Thanks to Jens Seidel for spotting. Jun 16 16:20:34 03Stefan Schmidt  07org.openembedded.dev * r02c2f941d1 10openembedded.git/MAINTAINERS: MAINTAINERS: Mention bug20 in my machines Jun 16 16:28:14 03Marcin Juszkiewicz  07org.openembedded.dev * r18f4823adb 10openembedded.git/MAINTAINERS: MAINTAINERS: removed my machines/distros Jun 16 16:44:25 woglinde: yeah the old bayern coarch still delivers it for swiss Jun 16 16:45:25 he khem my friend Jun 16 16:45:42 we fixed all of our problems Jun 16 16:45:54 we need two patches for gcc-4.3.x Jun 16 16:46:02 robert posted the first Jun 16 16:46:14 the second fixes linking with g++ and lgcc Jun 16 16:48:02 damnit Jun 16 16:48:47 kergoth? Jun 16 16:49:04 woglinde: what are those Jun 16 16:49:07 woglinde: care to post Jun 16 16:49:19 * khem checks the ml Jun 16 16:49:26 khem gcc-arm-intrinsic Jun 16 16:49:46 ups gcc-arm-atomic-intinsic Jun 16 16:54:55 khem second patch is gcc-4.4-20090519-arm-eabi-atomic-builtins-unbreak-v2.patch Jun 16 16:55:55 woglinde: ok that patch seems fine Jun 16 16:56:22 hm I will send it to the ml Jun 16 16:57:10 woglinde: ok you mean this backport http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40133 Jun 16 16:57:13 hms I deleted all refrences to robers patch Jun 16 16:57:36 khem luckily it works out of the box Jun 16 16:57:41 woglinde: i got systemd compile native. but it seemd it needs at least kernel 2.6.32 Jun 16 16:57:42 .. devtmpfs.. former devfs .... Jun 16 16:57:46 with gcc-4.3.1 and gcc-4.3.3 Jun 16 16:57:59 pwgen accept4 is the problem Jun 16 16:58:05 I changed it already Jun 16 16:58:22 will commit later Jun 16 17:00:41 hmm, any git guru can confirm that git push origin origin/stable/2009 will push into the stable branch from my local current branch? Jun 16 17:01:03 * stefan_schmidt wonders why git push makes him more trouble then the other commands Jun 16 17:02:58 stefan_schmidt: git push origin :stable/2009 Jun 16 17:03:16 stefan_schmidt: but try with --dry-run first Jun 16 17:03:21 khem: ah, thanks Jun 16 17:03:34 khem: yeah, dry-run would better Jun 16 17:03:53 and also add -v for more out put Jun 16 17:04:10 khem: yup Jun 16 17:04:39 khem: these refspecs are still somehow magic for me :) I need to take an hour to really understand them at some point :) Jun 16 17:04:44 03Robert Schuster  07org.openembedded.dev * r85e466de20 10openembedded.git/recipes/cacao/ (cacao.inc cacao_0.99.3.bb cacao_0.99.4.bb): cacao *: Install libjvm.so into separate folder in ${libdir}. Jun 16 17:04:46 03Robert Schuster  07org.openembedded.dev * rcaa1cc5652 10openembedded.git/recipes/jamvm/ (5 files): jamvm *: Install libjvm.so into separate folder in ${libdir}. Jun 16 17:05:49 ah, more java magic Jun 16 17:05:54 That is also strange stuff Jun 16 17:06:06 * stefan_schmidt did forgot almost everything from his uni java courses :) Jun 16 17:08:54 khem: worked like a charm, thanks a lot Jun 16 17:09:01 * stefan_schmidt needs a break for dinner Jun 16 17:25:14 woglinde: and the mount stuff. they want to mount devtmpfs /dev .. but this feature is new in 2.6.32 not in my 2.6.28 .. Jun 16 17:25:39 pwgen ughs okay Jun 16 17:25:44 than we are out of luck Jun 16 17:25:55 okay Jun 16 17:26:13 maybe a small workaround . by mounting tmpfs and do a couple of device nodes Jun 16 17:26:16 I think you are not in the mood the backport it to 2.6.28 Jun 16 17:26:23 hm yes Jun 16 17:27:33 i fear if devtmpfs is implemented like devfs , the backport will be a "interesting" task Jun 16 17:28:00 *g* Jun 16 17:30:15 whew Jun 16 17:30:17 finally Jun 16 17:30:20 hmm Jun 16 17:31:51 cool Jun 16 17:31:54 whatever it is Jun 16 17:32:04 we finsihed shark support this day too Jun 16 17:33:54 conversion of bitbake to use the python logging module is complete, was working on it late last night, fixed the last bug with it Jun 16 17:34:27 cool Jun 16 17:34:32 brb later Jun 16 17:34:39 bye pwgen Jun 16 17:34:45 03Robert Schuster  07org.openembedded.dev * rfffb474ad6 10openembedded.git/recipes/llvm/ (llvm-gcc4-cross_svn.bb llvm-gcc4_svn.bb llvm-native_2.5.bb): Jun 16 17:34:45 llvm-native 2.5: Removed unneeded recipe (superseeded by llvm2.5-native). Jun 16 17:34:45 llvm-gcc4 svn: Removed unused recipe. Jun 16 17:34:45 llvm-gcc4-cross svn: Removed unused recipe. Jun 16 17:34:48 the nice thing is it means you don't have to duplicate the domain in every message you output Jun 16 17:35:04 one logging.getLogger() at the top, then logger.info() or whatever from then on Jun 16 17:35:31 bb.{note,info,*} will still exist, of course, for convenience Jun 16 17:35:49 * kergoth_ isn't yet fully sure he prefers this, did it to see how the code looked afterwards Jun 16 18:30:37 03Tom Rini  07org.openembedded.dev * r0d40d05185 10openembedded.git/conf/local.conf.sample: Jun 16 18:30:37 local.conf.sample: Document ASSUME_PROVIDED for linux-libc-headers-native Jun 16 18:30:37 Signed-off-by: Tom Rini Jun 16 18:30:43 03Tom Rini  07org.openembedded.dev * r293fa519d1 10openembedded.git/ (4 files in 2 dirs): (log message trimmed) Jun 16 18:30:43 linux-libc-headers-native: Add, make it a dependency for most native recipes Jun 16 18:30:43 On some host distributions the provided linux kernel headers are too old to Jun 16 18:30:43 compile utilities we need. Given that we need these utilities to run things on Jun 16 18:30:44 the target the best solution is to provide linux-libc-headers-native. Rather Jun 16 18:30:44 than get things into an inconsistent state, we make linux-libc-headers-native Jun 16 18:30:45 be a default dependency. Jun 16 18:31:28 hmm Jun 16 18:32:59 03Mike Westerhof  07org.openembedded.dev * r1f05adf3f6 10openembedded.git/conf/distro/ (3 files in 2 dirs): Jun 16 18:32:59 SlugOS: Initial commit for SlugOS 5.5 - update toolchain, use eglibc Jun 16 18:32:59 Switch to gcc 4.4.4, update toolchains to match, switch to eglibc Jun 16 18:32:59 Also add some PREFERRED_PROVIDERS. Jun 16 18:36:45 anyone know if v4l2-ctl from ivtvdriver.org is available in an OE recipe? On distros like ubuntu its in the ivtv-utils package so I would think thats where it would be in OE if it existed (which isn't there) Jun 16 18:37:20 or is there is any other cmdline app to get/set v4l2 device controls thats in OE that anyone knows of? Jun 16 18:37:54 Hello does anyone know about gtk+-directfb recipe, being in the nonworking folder? I am trying to get GTK to work ontop of Directfb Jun 16 18:40:24 so Tartarus decided to push it anyway Jun 16 18:41:17 khem, eh? Jun 16 18:42:11 khem, did you miss my request for how do we move from here the other day? Jun 16 18:43:33 The suggestion was to document ASSUME_PROVIDED Jun 16 18:44:02 khem, you can always RFC making that be default or something Jun 16 18:45:56 Tartarus: I wonder when the next request to have gcc-native and glibc-native is going to come Jun 16 18:46:19 because someone wants to build something on a really arcane system Jun 16 18:46:38 we could end of building a build system before we really build OE Jun 16 18:46:46 there is no end to it Jun 16 18:48:07 khem: You really won't like what RP has then Jun 16 18:48:11 Since yes, he's got that Jun 16 18:48:27 And that's how the crosssdk / nativesdk stuff that's been in transition for some time does Jun 16 18:48:47 RHEL4, btw, is not IMHO arcane Jun 16 18:48:58 It's a commercial distro with another 18 months of support to it Jun 16 18:49:23 its not arcane yet because you are not needing gcc-native for it Jun 16 18:50:23 khem: there's really a better case for glibc/gcc/etc, for SDK exports Jun 16 18:50:32 The "what host will this work on" dance sucks, hard Jun 16 18:50:53 So if the SDK contains glibc, et al, that's that Jun 16 18:50:55 Tartarus: they should be generated host specific Jun 16 18:50:59 And that's RPs arguement, not mine Jun 16 18:51:06 or otherwise just make them static Jun 16 18:51:13 and be done with it Jun 16 18:51:22 there is no need to over complicate things Jun 16 18:51:23 khem: static isn't a magic bullet, there's still a few things that can get tripped up. Jun 16 18:51:28 LSB is more magic bulletish Jun 16 18:52:25 once you have a static toolchain what is left thats host specific for SDK Jun 16 18:52:49 its least errorprone path Jun 16 18:53:09 khem: RP is on vacation atm ;) Jun 16 18:53:14 That's his arguement, not mine Jun 16 18:53:35 Mine is that OE should support RHEL4 without a massive headache or external stuff being pulled in Jun 16 18:53:45 Yikes Jun 16 18:53:48 Since needing N overlays isn't the community OE model Jun 16 18:54:06 * Tartarus hasn't even proposed svn-native, also needed for RHEL4 Jun 16 18:54:21 Tartarus: and cvs-native ? Jun 16 18:54:25 (I also wanna poke that problem a bit more) Jun 16 18:54:44 khem: I won't argue against cvs-native, it's already in the default ASSUME_PROVIDED list Jun 16 18:55:40 Tartarus: how about creating host specific rpms instead Jun 16 18:56:03 or whatever the build box uses for package management Jun 16 18:56:08 khem: having a meta target to make a "sane host environment" to install is something I'd like to look at someday Jun 16 18:56:29 Since then yes, it'd be easier to even assume git-mnative Jun 16 18:56:40 *git-native Jun 16 18:57:27 right now a build from scratch half of the time we are building host stuff Jun 16 18:57:40 Yes, there's a lot of native stuff built Jun 16 18:57:40 thats a sin on modern distros Jun 16 18:57:49 Some of which isn't strictly needed Jun 16 18:57:55 Others are crappy tools :( Jun 16 18:57:57 experienced users wont feel these pinch Jun 16 18:58:00 bison-native for example Jun 16 18:58:07 but think from a new user pov Jun 16 18:58:20 From a new user pov linux-libc-headers-native is a no-op Jun 16 18:58:32 Or just about Jun 16 18:58:43 It's a new recipe, same deps as linux-libc-headers for the target Jun 16 18:58:44 why ? just because you assume that he is going to use the local.conf.sample ? Jun 16 18:59:16 No, because it's a no compile recipe Jun 16 18:59:19 oh well I better eat some thing :) Jun 16 18:59:23 heh Jun 16 19:01:33 * Tartarus slowly catches up on WC matches, 12min (except stoppage) on Spain Jun 16 19:03:19 Tartarus: do you like soccer :) Jun 16 19:03:33 wow some Americans also like it Jun 16 19:04:03 other day on ESPN they were voting if Argentina's goal was better than a stike from some golfer I forgot Jun 16 19:04:13 and they voted the golf shot :) Jun 16 19:04:15 heh Jun 16 19:04:27 I don't watch ESPN, except for soccer, yes Jun 16 19:04:39 And i don't record the post game shows Jun 16 19:05:00 heh they are funny Jun 16 19:05:12 (record the 4am coverage, 7am coverage and then 11am one since i'm not done w/ the first two by then) Jun 16 19:05:25 cool Jun 16 19:05:27 just need to avoid spoilers during the morning Jun 16 19:05:36 (failed today, dang coworkers :)) Jun 16 19:05:38 I would say argentina has a chance I am rooting for them Jun 16 19:06:00 Yeah, I don't know global stuff well enough really Jun 16 19:06:15 I know the US probably won't get beyond the 2nd round Jun 16 19:06:21 But we're getting pretty good :) Jun 16 19:06:31 I have a brazilian restaurant close by my house Jun 16 19:06:39 I go there for brazil game :) Jun 16 19:06:44 fun Jun 16 19:07:00 US is pretty good Jun 16 19:07:08 There's a few places here we'd like to go for games, but finding a babysitter is hard :( Jun 16 19:07:09 they have a chance in last 8 Jun 16 19:07:16 if they play like they played last time Jun 16 19:07:31 goalkeeper is good Jun 16 19:07:33 And don't have a lazy starting 15min :) Jun 16 19:07:40 yeah, howard is great :) Jun 16 19:08:09 but it will get tougher in second round onwards he will have to stop more :) Jun 16 19:44:32 03Khem Raj  07org.openembedded.dev * r8fcab9f2a9 10openembedded.git/recipes/gnu-config/gnu-config_git.bb: Jun 16 19:44:32 gnu-config_git.bb: Add checksums Jun 16 19:44:32 Signed-off-by: Khem Raj Jun 16 20:31:59 03Koen Kooi  07org.openembedded.dev * r197acb9de5 10openembedded.git/recipes/gnu-config/gnu-config_git.bb: gnu-config git: remove duplicate, wrong checksums Jun 16 20:47:24 re Jun 16 20:51:09 03Leon Woestenberg  07org.openembedded.dev * r68651a6c43 10openembedded.git/recipes/udev/ (acl_2.2.49.bb attr_2.4.44.bb): (log message trimmed) Jun 16 20:51:09 udev/{acl,attr}: Fix install. Wanted to install to host /usr/bin/. Jun 16 20:51:09 The subdirectory Makefile includes a file, that overrides the Jun 16 20:51:09 variables inherited from the parent Makefile. Those where provided Jun 16 20:51:09 to the parent make through '-e'. Jun 16 20:51:10 Make sure the subdir makes also use '-e'. Jun 16 20:51:11 (An alternative fix is to sed the include file.) Jun 16 20:51:19 gm Jun 16 20:55:59 likewise: hey howdy Jun 16 20:57:52 re khem Jun 16 20:57:55 hi likewise Jun 16 20:59:31 hey hi all Jun 16 20:59:47 khem: tnx for the ack. the MAKEFLAGS=-e didn't work, btw. Jun 16 21:13:03 likewise: ok Jun 16 21:26:31 khem hm you made the patches-symlink stuff Jun 16 21:26:50 why the hell the patches are copied anyway to ${WORKDIR} Jun 16 21:29:14 woglinde: thats not my part :) Jun 16 21:29:35 woglinde: I have made symlinks so reduced the copies to half Jun 16 21:29:53 hm that must be late regression Jun 16 21:30:00 kergoth would get rid of the other half which is fall out from the change to simplify file:// Jun 16 21:30:11 yes its known Jun 16 21:30:15 I can remember patches werent copied to ${WORKDIR} Jun 16 21:30:18 ah okay Jun 16 21:30:28 hm woot Jun 16 21:30:42 systemd compiles and packaged Jun 16 21:34:21 bitbake seems not to differentiate between files and patches anymore Jun 16 21:38:53 It assumes a .patch or .diff is a patch now Jun 16 21:40:55 jo kergoth Jun 16 21:41:08 hey Jun 16 21:45:07 hrms Jun 16 21:45:23 why the image dont install the package I specified in INSTALL Jun 16 21:48:52 hm Jun 16 21:50:05 woglinde: add them to IMAGE_INSTALL Jun 16 21:50:38 e.g. IMAGE_INSTALL_append_pn-console-image = " libstdc++" Jun 16 21:50:39 :) Jun 16 21:51:04 that should add libstdc++ to console-image Jun 16 21:54:07 hm console-image has += Jun 16 21:54:45 whats append_pn? Jun 16 21:54:54 instead off _append? Jun 16 21:55:30 append is same as += here Jun 16 21:58:58 khem nope Jun 16 21:59:03 dont installed Jun 16 22:01:20 * mwester seems to recall that attempting to mix "+=" and _append for the same override caused problems in SlugOS IMAGE_INSTALL too. Jun 16 22:01:48 The solution was to use only one of the two mechanisms, which actually required some reorganization of the recipe(s) Jun 16 22:01:59 mwester I can see the package in IMAGE_INSTALL with bitbake -e Jun 16 22:02:43 woglinde: which package ? Jun 16 22:02:52 systemd Jun 16 22:03:00 woglinde: make sure you added the package name as it appears in .ipk Jun 16 22:03:03 and not the recipe name Jun 16 22:03:22 mwester: kudo's to you on upgrading toolchain for slug Jun 16 22:03:40 args Jun 16 22:03:42 args Jun 16 22:04:00 systemd_git-r0.5_armv7a.ipk Jun 16 22:04:01 khem - thanks for your help; your work and your suggestions were invaluable. :) Jun 16 22:04:12 what the hell Jun 16 22:04:20 mwester: merci :) Jun 16 22:05:21 woglinde: is it using master bitbake ? Jun 16 22:05:50 yes Jun 16 22:05:59 will do like uclibc_git Jun 16 22:06:47 hrms still systemd-git Jun 16 22:08:03 woglinde: systemd_git I think isnts it note the underscore Jun 16 22:08:33 *sigh* Jun 16 22:08:34 woglinde: it should be ok to use IMAGE_INSTALL += " systemd" then Jun 16 22:09:41 hm uclibc its the same Jun 16 22:09:53 khem yes but nothig happens Jun 16 22:13:23 why isn't openembedded supported on uclinux ? Jun 16 22:13:25 woglinde: paste the systemd recipe soemwhere Jun 16 22:13:33 http://blackfin.uclinux.org/ Jun 16 22:13:41 e.g. in case this implementation Jun 16 22:13:44 openembedded is metadata, a build tool Jun 16 22:13:44 djszapi: because the support needs updating Jun 16 22:13:55 do you mean oe based distros? Jun 16 22:14:29 can I use openembedded on this processor with this uclinux solution ? Jun 16 22:14:51 Tartarus: I can help in coding if it's needed ;) Jun 16 22:14:56 djszapi: you can but you have to do some work Jun 16 22:15:05 I thought there is a big antipattern in the policy for it ;) Jun 16 22:15:31 djszapi: no antipattern Jun 16 22:16:12 what kind of work must I do ? Jun 16 22:16:33 to get uclinux support upto mark Jun 16 22:16:47 what do you mean exactly ? Jun 16 22:17:28 i.e. letting OE know about uclinux like it knows about linux Jun 16 22:17:47 TARGET_OS="uclinux" and all should fall into place :) Jun 16 22:19:05 uhum Jun 16 22:21:36 So Jun 16 22:21:43 There's already the hooks for uclinux in OE Jun 16 22:21:50 It just hasn't been used in a few years Jun 16 22:21:53 So, it's bitrotten Jun 16 22:23:37 okay. Jun 16 22:23:53 I will check it as long as I will have some leisure time ;) Jun 16 22:25:19 khem helped me earlier to be updated with the differences between buildroot and openembedded Jun 16 22:25:35 but what're the main advantages ? Jun 16 22:25:56 I used to deal with buildroot earlier. Jun 16 22:35:46 khem its clash btw libattr1 and libattr2 Jun 16 22:36:30 arggsss Jun 16 22:41:42 damn Jun 16 22:41:52 whats wrong with libattr_ Jun 17 01:00:26 meh **** ENDING LOGGING AT Thu Jun 17 02:59:56 2010