**** BEGIN LOGGING AT Wed Mar 17 02:59:57 2010 Mar 17 04:24:12 Laibsch: hi Mar 17 04:33:24 zecke: howdy Mar 17 04:33:59 zecke: sorry havent done much on security front been sucked up in some personal work filing taxes etc. Mar 17 05:02:23 which flag set allow me do "insecure download" Mar 17 06:03:01 03Holger Hans Peter Freyther  07org.openembedded.dev * rb6faa2ebb0 10openembedded.git/ (3 files in 2 dirs): Mar 17 06:03:01 nsd: Bump to 3.2.4 Mar 17 06:03:01 This addresses CVE-2009-1755 which is in any nsd below 3.2.2. Mar 17 06:03:02 03Holger Hans Peter Freyther  07org.openembedded.dev * rdaba9a8ffb 10openembedded.git/recipes/cdrtools/cdrtools-native_2.01.bb: cdrtools-native: Upgrade to the latest version Mar 17 06:03:03 03Holger Hans Peter Freyther  07org.openembedded.dev * rd457414508 10openembedded.git/ (3 files in 3 dirs): Mar 17 06:03:04 maradns: Upgrade to version 1.4.03 Mar 17 06:03:04 * Patch the Makefiles so LDFLAGS is passed to the linker Mar 17 06:03:06 03Holger Hans Peter Freyther  07org.openembedded.dev * r8d574e65a9 10openembedded.git/ (4 files in 2 dirs): Mar 17 06:03:06 lighttpd: Upgrade to 1.4.26 Mar 17 06:03:06 This should address CVE-2010-0295. Mar 17 06:03:07 03Holger Hans Peter Freyther  07org.openembedded.dev * r9912eb3d0b 10openembedded.git/ (6 files in 3 dirs): Mar 17 06:03:07 socat: Upgrade to version 1.7.1.2 Mar 17 06:03:07 Upgrade socat to the latest available version Mar 17 06:24:30 Good morning. I need to disable sound in my Beagleboard-kernel, but how? Can I use menuconfig in some way to alter the defconfig? Mar 17 06:25:50 tasslehoff: bitbake -cmenuconfig virtual/kernel (or maybe virtual/linux) Mar 17 06:26:06 tasslehoff: and then copy the .config back to recipes/linux/linux-your-version/beagleboard/defconfig Mar 17 06:27:04 zecke: thanks a bunch. Mar 17 06:29:48 zecke: I got this output http://pastebin.ca/1843293 Mar 17 06:30:58 tasslehoff: you are on a remote system? Mar 17 06:31:21 zecke: nope. Mar 17 06:32:06 tasslehoff: did you ever play with TERMCMD in your local conf or such? Mar 17 06:34:48 zecke: no Mar 17 06:37:02 zecke: mine is based on contrib/angstrom/local.conf Mar 17 06:39:27 tasslehoff: weird, can you do a ps xua | grep make? is something like menuconfig running? Mar 17 06:39:41 tasslehoff: when I run it on my remote server I get a failed task because it fails to open the display Mar 17 06:40:09 tasslehoff: but you get a success, which means a gnome-terminal or such must have been opened somewhere, or in different configs a screen Mar 17 06:41:01 zecke: ps shows nothing. hmm. that's odd. Mar 17 06:41:37 tasslehoff: it is, you can run bitbake with -DDD and -vvv to see what is happening. :} Mar 17 06:43:12 zecke: I tried in xterm, and I think I saw a quick flash of a window appearing and disappearing again Mar 17 06:47:19 03Holger Hans Peter Freyther  07org.openembedded.dev * re54dcda0f3 10openembedded.git/ (3 files in 3 dirs): Mar 17 06:47:19 wv: Upgrade to 1.2.4 to address multiple flaws Mar 17 06:47:19 Addressed CVE-2006-4513. Mar 17 06:47:29 03Holger Hans Peter Freyther  07org.openembedded.dev * r3391bc4c4d 10openembedded.git/ (3 files in 3 dirs): Mar 17 06:47:30 bogofilter: Upgrade to 0.96.6 Mar 17 06:47:30 * 0.96.6 is the latest release of the ancient 0.96 release Mar 17 06:47:30 * It is addressing CVE-2005-4592 Mar 17 06:50:22 zecke: http://pastebin.ca/1843299 <- ran with -DDD Mar 17 06:51:16 tasslehoff: well, your gnome-terminal can not be launched. but that is in the domain of your config. Mar 17 06:57:09 zecke: yep. can I tell it to use xterm instead of gnome-terminal? was that the TERMCMD you mentioned? Mar 17 07:05:07 zecke: seems it is a bad git-revision. tried an up-to-date org.openembedded.dev, and it works. Mar 17 07:08:25 03Holger Hans Peter Freyther  07org.openembedded.dev * rd3a2941010 10openembedded.git/recipes/pulseaudio/ (files/CVE-2009-1299.patch pulseaudio_0.9.15.bb): Mar 17 07:08:25 pulseaudio-0.9.15: Fix a flaw in the tmp directory handling Mar 17 07:08:25 * Address CVE-2009-1299, compare with http://www.debian.org/security/2010/dsa-2017 Mar 17 07:35:21 thebohemian: hey Mar 17 07:35:59 zecke: yo :) Mar 17 07:36:42 thebohemian: how is life? Mar 17 07:37:06 zecke: very nice :) Mar 17 07:37:18 zecke: apprendo español ahora. Mar 17 07:38:22 zecke: work is a bit busy though but it would be worse if nothing is to do ;) Mar 17 07:40:04 nice Mar 17 07:53:58 good morning Mar 17 07:57:26 zecke: I managed to configure, but I'm not sure where it stored the result. "git diff" shows nothing :) Mar 17 08:01:36 morning Mar 17 08:04:39 did anyone try runit instead of sysvinit with oe? or something else that would monitor and restart services if they crash? Mar 17 08:15:15 tasslehoff: it is inside the tmp/work... /linux../.config Mar 17 08:15:30 zecke: thanks Mar 17 08:34:16 03Koen Kooi  07org.openembedded.dev * r01c5424aa6 10openembedded.git/recipes/xorg-lib/ (2 files in 2 dirs): pixman git: fix typo in patch Mar 17 08:34:16 03Koen Kooi  07org.openembedded.dev * r0785f3f8ca 10openembedded.git/contrib/angstrom/sort.sh: angstrom feed sorter: add smdk6410 Mar 17 08:34:24 03Koen Kooi  07org.openembedded.dev * r7779f2044a 10openembedded.git/conf/distro/include/angstrom-2008-preferred-versions.inc: angstrom: pin lighttpd to 1.4.26 Mar 17 09:50:06 03Koen Kooi  07org.openembedded.dev * ree29421cc8 10openembedded.git/recipes/gstreamer/gst-ffmpeg_0.10.10.bb: gst-ffmpeg: add 0.10.10, needed for newer ffmpeg_svn.bb Mar 17 09:50:07 03Koen Kooi  07org.openembedded.dev * r155d69ac42 10openembedded.git/recipes/gimp/gegl_svn.bb: gegl: convert to new-style staging Mar 17 09:50:08 03Koen Kooi  07org.openembedded.dev * rbe85c2cf4c 10openembedded.git/recipes/gimp/gegl_git.bb: gegl: add git version Mar 17 09:50:11 03Koen Kooi  07org.openembedded.dev * r10ca2773f9 10openembedded.git/recipes/ffmpeg/ffmpeg.inc: ffmpeg: install more avutil headers, partially fixes the breakages introduced by the SRCREV bump Mar 17 09:50:12 03Koen Kooi  07org.openembedded.dev * r6cb0ae5c19 10openembedded.git/recipes/gimp/babl_git.bb: babl: add git version Mar 17 09:50:13 03Koen Kooi  07org.openembedded.dev * r5a70033fdb 10openembedded.git/classes/sourcepkg.bbclass: sourcepkg: save lots of space by only making the diff arch specific Mar 17 09:50:23 03Koen Kooi  07org.openembedded.dev * r4935517e32 10openembedded.git/recipes/orc/orc.inc: orc: add license stub, packaging still needs to change to comply with it Mar 17 10:46:09 03Martin Jansa  07org.openembedded.dev * rc588c7eceb 10openembedded.git/recipes/findutils/ (6 files): Mar 17 10:46:09 findutils: convert to BBCLASSEXTEND, add 4.5.5 instead of 4.4, use more specific LICENCE GPLv2|3 Mar 17 10:46:09 Signed-off-by: Martin Jansa Mar 17 10:46:10 03Martin Jansa  07org.openembedded.dev * r806b1f04ac 10openembedded.git/recipes/openmoko-3rdparty/ventura_svn.bb: Mar 17 10:46:10 ventura: bump SRCREV for https handling Mar 17 10:46:10 Signed-off-by: Martin Jansa Mar 17 10:46:11 03Martin Jansa  07org.openembedded.dev * r67b3baea2f 10openembedded.git/ (81 files in 25 dirs): Mar 17 10:46:12 sane-srcrevs: move SRCREVs to recipes for openmoko related svn recipes Mar 17 10:46:12 * As suggested few times on ML Mar 17 10:46:12 * I would move git revs too, but after merging SRCPV (don't want to Mar 17 10:46:12 resolve so much conflicts now) Mar 17 10:46:13 Signed-off-by: Martin Jansa Mar 17 11:01:55 03Graeme Gregory  07org.openembedded.dev * r60314be80a 10openembedded.git/recipes/linux/ (2 files in 2 dirs): Mar 17 11:01:55 linux-omap-zoomsync_2.6.32.bb : update to the latest zoomsync kernel Mar 17 11:01:55 This is the version ti integration tree is using, updated the defconfig Mar 17 11:01:55 to include the wifi driver and UVC devices. Mar 17 11:08:38 Hi guys. Could you please verify that `angstrom-dev/staging/i686-linux/usr/include/lzo*` are not empty for you. Mar 17 11:09:14 The staging area for the host system. Mar 17 11:09:57 PaulePanter: why? Mar 17 11:10:42 zecke: They are empty on my system and therefore compilation fails for `mtd-utils-native` for example. Mar 17 11:11:29 I am doing `bitbake console-image` with GCC 4.4.3 over my old temporary directory. Mar 17 11:11:49 It might be that this is confusing bitbake.(?) Mar 17 11:17:04 PaulePanter: well, it is more a question of depedencies. mtd-utils-native DEPENDS on zlo-native Mar 17 11:17:09 PaulePanter: was it build? Mar 17 11:17:44 zecke: How can I verify it. The date stamp suggest that the files were built. But they are empty. Mar 17 11:18:05 good morning mickeyl Mar 17 11:18:09 morning pb_ Mar 17 11:18:20 zecke: Anyways doing `bitbake -c clean lzo-native` gave me files with content. Mar 17 11:18:22 and a very fine morning it is here. is it still snowing where you are? Mar 17 11:18:50 pb_: hi Mar 17 11:19:00 zecke: hail zecke Mar 17 11:22:06 pb_: no, luckily the beginnings of spring have appeared Mar 17 11:22:31 oh the irony though... we're driving to austria on saturday to go skiing... *cough* Mar 17 11:22:42 very good. yeah, same here. we have flowers, lambs gambolling in the fields, the whole works. Mar 17 11:22:50 ah, heh. oh well :-) Mar 17 11:23:10 good morning Mar 17 11:23:13 florian: good morning Mar 17 11:24:10 morning florian Mar 17 11:27:03 morning Mar 17 11:27:20 hi methril Mar 17 11:27:26 morning all Mar 17 11:27:32 hi mckoan Mar 17 11:42:23 mickeyl, where in Austria? We were in St Anton a week ago :) Mar 17 11:43:01 Crofton|work: Gaschurn. St.Anton is cool; was it fun? Mar 17 11:43:07 yeah Mar 17 11:43:11 really good Mar 17 11:43:18 they had 10" of snow the day we arrived Mar 17 11:43:59 ncie :) Mar 17 11:44:00 nice Mar 17 11:47:16 http://www.flickr.com/photos/32615155@N00/sets/72157623466216879/ Mar 17 11:47:37 aah, very cool Mar 17 11:54:37 actually, I'm not skiing anymore Mar 17 11:54:45 03Michael 'Mickey' Lauer  07org.openembedded.dev * rb32478bf38 10openembedded.git/recipes/images/fso-image.inc: fso-image.inc: remove unconditional autostart of bluetooth, avahi, connman. FSO uses dbus system activation for that Mar 17 11:54:45 03Michael 'Mickey' Lauer  07org.openembedded.dev * r3c3e5c5d52 10openembedded.git/recipes/tasks/task-fso2-compliance.bb: task-fso2-compliance: remove unnecessary dependencies Mar 17 11:54:46 03Michael 'Mickey' Lauer  07org.openembedded.dev * red446723c8 10openembedded.git/recipes/tasks/task-cli-tools.bb: task-cli-tools: add gdb to task-cli-tools-debug Mar 17 11:54:47 03Michael 'Mickey' Lauer  07org.openembedded.dev * r00e29e85c1 10openembedded.git/recipes/freesmartphone/fsodatad_git.bb: fsodatad: DEPEND on mobile-broadband-provider info Mar 17 11:54:50 but i'm coming with the group to enjoy the evenings Mar 17 11:55:02 will work while they're having fun Mar 17 11:55:04 *shrug* Mar 17 12:07:24 morning Mar 17 12:08:01 morning, hrw. Mar 17 12:17:50 03Joshua Lock  07org.openembedded.dev * rf94496c1c2 10openembedded.git/classes/packaged-staging.bbclass: (log message trimmed) Mar 17 12:17:50 packaged-staging: Fix packagaging of cross packages Mar 17 12:17:50 packagedstaging_fastpath() was only copying the contents of CROSS_DIR to Mar 17 12:17:50 PSTAGE_TMPDIR resulting in the folders contents being packaged and then Mar 17 12:17:50 installed incorrectly at the top level of CROSS_DIR rather than in HOST_ARCH Mar 17 12:17:50 specific sub directories. Mar 17 12:17:51 This patch fixes that issue by copying the directory and its contents rather Mar 17 12:19:23 Does anyone want to pull cherry pick f1cab4b4124b7ed94b58445f59e1c89b7a13e76e from my tree (zecke maybe?) as it fixes the broken download URL for Socat since the recipe bump. http://git.openpandora.org/cgi-bin/gitweb.cgi?p=openembedded.git;a=summary Mar 17 12:19:55 anyone know what OE package would provide "resize"? Mar 17 12:20:25 like "eval `resize`" when using a ttyS console with an xterm Mar 17 12:20:48 thought it was ncurses but no Mar 17 12:24:20 th1_, busybox comes with a resize applet Mar 17 12:25:39 a fabulous one, btw ;) Mar 17 12:28:07 morning all Mar 17 12:37:42 hi, how can i set the kernel version to a fixed value, i want to compile Angstrom.2008 for kernel 2.26.27 and i'm using "PREFERRED_VERSION_linux = "2.6.27" but i get everything compiled for kernel 2.6.27.45 ... because i'm using a prebuild kernel, the modules doesn't want to load because of the different kernel version. Mar 17 12:43:13 hi rp Mar 17 12:44:51 marekp: all our 2.6 kernels are updated to more or less latest stable updates Mar 17 12:46:05 can i somehow manually use another kernel version? Mar 17 12:46:34 edit recipe Mar 17 12:46:42 and drop stable patch from it Mar 17 13:04:58 03Martin Jansa  07org.openembedded.dev * rc240d00f5f 10openembedded.git/conf/distro/include/ (sane-srcrevs-fso.inc sane-srcrevs.inc shr-autorev.inc): Mar 17 13:04:58 shr: bump SRCREVs for SHR projects, fsogsmd, remove frameworkd-config from autorev Mar 17 13:04:58 Signed-off-by: Martin Jansa Mar 17 13:08:15 03Martin Jansa  07org.openembedded.dev * r51b6131073 10openembedded.git/conf/distro/include/sane-srcrevs-fso.inc: fsogsmd: fix SRCREV bump Mar 17 13:08:44 RP_: hey! Mar 17 13:09:09 RP_, thanks for rememberig the TSC election :) Mar 17 13:28:49 stupid udev is making my usb nic eth3 when I plug it in Mar 17 13:29:00 any idea how I can make it pick something better? Mar 17 13:29:05 I think is is confused Mar 17 13:29:34 what do you have in /etc/udev/rules.d? Mar 17 13:29:58 you'll probably find there is some kind of persistent net state file in there which has enumerated the cards that it expects to be eth0-eth2 Mar 17 13:30:31 zecke: ping Mar 17 13:30:34 whatever the default is :) Mar 17 13:30:35 JaMa: hi Mar 17 13:31:04 zecke: Hi, socat-1.7.1.2 seems only available on SRC_URI without /Archive/ Mar 17 13:31:56 JaMa: oops, i wonder how I fetched it then. feel free to update that. Mar 17 13:32:18 zecke: ok, I'll, I was curious how you managed to fetch it for checksums :) Mar 17 13:33:10 JaMa: do the checksums match yours? Mar 17 13:33:30 yes Mar 17 13:33:35 ah, so delete the persitent file and let udev start over ... Mar 17 13:33:41 JaMa: funny Mar 17 13:33:55 JaMa: thanks for updating findutils. It just built for you? I was fighting with two versions of it Mar 17 13:35:32 zecke: 4.4.2 was failing.. then I noticed that there are much newer versions on that alpha mirror and the 4.5* compiled just fine without any change for that lfseek (or something like that) Mar 17 13:36:51 03Martin Jansa  07org.openembedded.dev * rd7c9e12cd2 10openembedded.git/recipes/socat/socat_1.7.1.2.bb: Mar 17 13:36:51 socat_1.7.1.2: change SRC_URI, it's available only without /Archive/ Mar 17 13:36:51 Signed-off-by: Martin Jansa Mar 17 13:38:30 we only have about ~10 things with outstanding CVEs Mar 17 13:38:43 and samba-essential which is so important that it is unmaintained for years Mar 17 13:42:14 zecke: maybe you should resent that list with remaining items, if you have it Mar 17 13:43:23 JaMa: yeah, I should :) Mar 17 13:43:51 postresql, curl-sdk (i'm doing that right), neon, apr, dnsmasq, vnc are outstanding Mar 17 13:44:01 we have a ipsec-tools on the list and also a even newer grip Mar 17 13:44:28 and I applied a pulseaudio fix now... looks like debian released their security report too early Mar 17 14:00:33 03Martin Jansa  07org.openembedded.dev * rbb75d7f3b6 10openembedded.git/conf/distro/include/sane-srcrevs-fso.inc: Mar 17 14:00:33 sane-srcrevs-fso: bump all (because I forgot to bump libfso-glib, fos-specs before for newer fsogsmd) Mar 17 14:00:33 Signed-off-by: Martin Jansa Mar 17 14:18:58 pb_, any opinion to the basic idea (the impl can still be improved)? http://article.gmane.org/gmane.comp.handhelds.openembedded/30917 Mar 17 14:31:05 03Graeme Gregory  07org.openembedded.dev * rd8a743b064 10openembedded.git/recipes/linux/linux-omap-zoomsync_2.6.32.bb: linux-omap-zoomsync_2.6.32.bb : reinstate bootlogo Mar 17 14:35:45 adam@e ~/oe $ bitbake console-image ERROR: Importing sqlite3 and pysqlite2 failed, please install one of them. Python 2.5 or a 'python-pysqlite2' like package is likely to be what you need. adam@e ~/oe $ python --version Python 2.6.4 adam@e ~/oe $ sqlite3 --version 3.6.23 Mar 17 14:36:00 Any clues? Mar 17 14:37:43 awozniak, Which distro? Mar 17 14:37:59 my host? gentoo Mar 17 14:38:51 Hm, sorry can't help overly specifically, but there is probably a python-sqlite binding package that you haven't installed. Mar 17 14:40:38 awozniak, I think it may be called pysqlite Mar 17 14:41:13 trying "emerge pysqlite" now Mar 17 14:48:48 still no-one with channel op privs around who can change the topic ???? Mar 17 15:02:38 03Graeme Gregory  07org.openembedded.dev * r1b8480c8b5 10openembedded.git/recipes/linux/ (2 files in 2 dirs): linux-omap-zoomsync_2.6.32.bb : update defconfig Mar 17 15:16:31 blindvt`: sorry, which basic idea are we talking about? Mar 17 15:16:48 oh, right, I see your link now. doh. :-} Mar 17 15:17:06 let me have a look Mar 17 15:18:49 blindvt`: looks pretty reasonable overall. does uclibc-arch really need to be a bbclass rather than just part of uclibc.inc? Mar 17 15:19:40 and, do you really need an -initial variant of the nptl branch? not that it really matters, but still. Mar 17 15:22:25 * kergoth yawns Mar 17 15:27:18 pb_, well, i think it has to be a class since it decides upon configuration of alot of packages. I previously said: i feel pretty dumb to spell out the obvious, but, just to reiterate: an arch constrains (hard/soft constraint, via e.g. float, mmu ,endian, abi, stack growth direction and so forth), a toolchain features/provides (binfmt etc, and highlevel -- or midlevel if you prefer -- stuff like malloc_0, lfs, nls, network, Mar 17 15:27:18 etc); Then, from a TARGET_OS level, a package features/provides stuff, a distro requires stuff Mar 17 15:28:27 cbrake: ping? Mar 17 15:28:45 oneshel: yes Mar 17 15:29:05 pb_, so we have to take machine contraints into account as well as distro wishes (e.g.: distro foo wants ipv4; distro bar wants LFS and IPv6 and no IPv4) Mar 17 15:29:39 sorry to bother you, just wondering if you'd have some time this afternoon to hop on IM with me and talk about the kernel changes Mar 17 15:30:12 btw, this is Jim from russound, don't know if I've talked t you you on IRC with this nick yet Mar 17 15:31:54 oneshel: hey Jim Mar 17 15:32:28 cbkrake: hey Cliff! Mar 17 15:32:29 * cbrake adds oneshel to his secret decoder list ... Mar 17 15:32:47 hehe Mar 17 15:33:01 pb_, with the RFC hunks, i am able to turn off, for example, ipv6 globally; automake packages would be built with --disable-ipv6, libc and busybox will be built without IPv6 support. Same thing for --disable-largefile. And the same thing for passing the proper ac_cv_malloc_0 settings to configure, again, just an example. If i can do that without a class, then that'd be even better, but i think i cannot Mar 17 15:34:25 i haven't seen the patch you're proposing, I don't think, but is there some reason you can't obey the ipv6 DISTRO_FEATURE? Mar 17 15:35:10 * kergoth grumbles, tempted to down a bunch of caffeine despite attempts to ditch it Mar 17 15:36:09 kergoth, it does that via DISTRO_FEATURES, yes. See the class in http://thread.gmane.org/gmane.comp.handhelds.openembedded/30305/focus=30917 Mar 17 15:36:18 * kergoth looks Mar 17 15:36:42 * kergoth kicks gmane Mar 17 15:36:53 features_to_uclibc_settings() Mar 17 15:37:36 it should be in your inbox, too; Subject: [oe] [PATCH 3/3] uClibc: redo configuration Mar 17 15:37:42 k Mar 17 15:38:57 remember that it's an RFC. No distro asks for ipv4 or ipv6 at the moment, so can't go in as is. Mar 17 15:39:06 * kergoth_ nods Mar 17 15:58:23 blindvt`: so, which packages exactly would you envisage inheriting uclibc-arch.bbclass? Mar 17 15:58:29 maybe I am missing the point slightly Mar 17 15:59:21 hm. more likely that the one who misses the point is me Mar 17 16:01:13 pb_, noone, so it can go to the .inc. But how would i negotiate e.g. ac_cv_malloc_0 ? Mar 17 16:02:34 or ld hash-style Mar 17 16:04:08 ld hash-style seems to be dealt with by ${TARGET_LINK_HASH_STYLE}, for better or worse. Mar 17 16:04:19 arguably for worse, since bitbake.conf refers to that variable but doesn't define it. still... Mar 17 16:05:04 for ac_cv_malloc_0, I guess that is a question of whether you think it is a knob that DISTROs will want to play with. if you're going to take the view that malloc(0) will always be NULL for uclibc, it can just go in the uclibc site files as such. Mar 17 16:06:31 pb_, i don't think distros would care much, but whatever it is set to in the .config, it would be passed to configure invocations Mar 17 16:07:31 * Tartarus tries pw-am.sh Mar 17 16:08:54 * Tartarus waitts for patchwork to get the ack Mar 17 16:09:56 pb_, these .config settings somehow have to be passed to the rest but they only live in the .config ATM (which is fine for most of them). In my old setup, i made all of them available to the whole build machinery, so i could easily access them. Mar 17 16:11:09 pb_, but i think that i can't access them from other recipes if they are in the uClibc.bb, no? Mar 17 16:11:19 blindvt`: yeah. so, the more canonical oe way of dealing with that is to have the primary configuration done in the DISTRO (somehow) and for both the .config and the other recipes to draw their information from that. Mar 17 16:11:42 trying to export configuration bits from uclibc.bb to other recipes is a rather fraught process. it is just about possible to make it work, by installing special bits into staging, but it is not a very good idea. Mar 17 16:15:26 MWelchUK_work: ping Mar 17 16:15:48 Tartarus, lo Mar 17 16:15:54 Hey Mar 17 16:16:09 Can you redo the 8641 patch real quick? The xilinx stuff means it doesn't apply anymore Mar 17 16:16:28 Agh! :-) Mar 17 16:16:32 indeed Mar 17 16:16:43 thanks Mar 17 16:17:24 np, just need to update my tree... Mar 17 16:24:04 Tartarus, should be on it's way to the list and you. Mar 17 16:24:41 * Tartarus waits on patchwork again Mar 17 16:37:06 MWelchUK_work: done, thanks Mar 17 16:37:14 03Martyn Welch  07org.openembedded.dev * r86ab2dc660 10openembedded.git/ (6 files in 5 dirs): Mar 17 16:37:14 Add Freescale mpc8641_hpcn demo board Mar 17 16:37:14 A demo board from Freescale, based on the MPC8641D Freescale SOC. Mar 17 16:37:14 Acked-by: Tom Rini Mar 17 16:37:14 Signed-off-by: Tom Rini Mar 17 16:37:30 I'm having problems with my first ever OpenEmbedded build, maybe somebody can offer some advice... Mar 17 16:37:36 Tartarus, Cheers, much appreciated. Mar 17 16:37:44 mikeul: if you have a question, just ask it Mar 17 16:39:07 Upon 'bitbake -k console-image', I get 9 errors, so I'm looking at the first, "ERROR: '/data/oe/openembedded/recipes/shared-mime-info/shared-mime-info-native_0.51.bb' failed" Mar 17 16:39:33 above that will be more information, which task failed, and a link to the log file. Mar 17 16:39:50 I do 'bitbake -b ...._0.51.bb' and I get http://pastebin.com/G06wBVPu Mar 17 16:43:51 odd. thats a perl script, but its not running it with the perl interpreter Mar 17 16:44:01 looks like the #! may be screwed up? Mar 17 16:44:05 exactly. Mar 17 16:44:28 Tartarus, please could you mark the patch "[oe,v2] commit 099dc400f4a7fb056b59a6faaaac6964823eda89" as dead in Patchwork - it's the mpc8641_hpcn patch, but I screwed up the subject. Mar 17 16:45:17 Indeed, the first line of the perl script is just "#!" with no path to an interpreter. But how the heck did that happen? Mar 17 16:45:58 good question :) most often thats a @PERL@ in the .in, substituted by the autoconf configure script.. i have no idea how the intltool-native configure could possibly have gotten that far.. if it didn't find perl, it should have died.. hmm Mar 17 16:46:43 03David Lanzendörfer  07org.openembedded.dev * r741d040e05 10openembedded.git/recipes/xserver-kdrive-common/ (xserver-kdrive-common/Xserver xserver-kdrive-common_0.1.bb): Mar 17 16:46:43 xserver-kdrive-common: add trout board to Xserver start script (kdrive) Mar 17 16:46:43 This patch should allow kdrive to use the touchscreen if you are booting SHR on trout bases devices like htcdream Mar 17 16:46:43 Signed-off-by: Martin Jansa Mar 17 16:46:44 03David Lanzendörfer  07org.openembedded.dev * r801d00f929 10openembedded.git/recipes/shr/alsa-scenarii-shr_git.bb: Mar 17 16:46:44 alsa-scenarii-shr: make alsa-scenarii-shr available for htcdream Mar 17 16:46:45 Signed-off-by: Martin Jansa Mar 17 16:46:45 03Martin Jansa  07org.openembedded.dev * rba4ccaa713 10openembedded.git/recipes/xorg-util/ (16 files): Mar 17 16:46:46 util-macros: convert to BBCLASSEXTEND, add version 1.6.1 Mar 17 16:46:46 Signed-off-by: Martin Jansa Mar 17 16:46:47 03David Lanzendörfer  07org.openembedded.dev * rd3780146af 10openembedded.git/recipes/firmwares/wl1251-fw_1.0-p1.bb: Mar 17 16:46:47 Added wl1251 firmware Mar 17 16:46:48 Hi Mar 17 16:46:48 I'd like to introduce a the bitbake file for the wl1251 firmware Mar 17 16:46:49 this firmware is needed to get wl1251 like its in the htcdream to initialize Mar 17 16:47:04 I have perl installed natively in /usr/bin/perl Mar 17 16:47:07 wl1251 is in htcdream? Mar 17 16:47:22 Signed-off-by: Martin Jansa Mar 17 16:47:46 where do I find the intltool-native configure script? Mar 17 16:51:39 kergoth, it was helpful just to know that the same thing jumped out at you as I had noticed. I just have to read up on autoconf and trace the problem back a bit. Mar 17 16:54:00 intltool-native is a recipe, you could INHERIT += "devshell", then bitbake -c devshell intltool-native to spawn a shell in its source tree Mar 17 16:55:37 you mean add that INHERIT line in the intltool-native .bb file? Mar 17 16:57:59 03Michael 'Mickey' Lauer  07org.openembedded.dev * r76dc5ff734 10openembedded.git/recipes/vala/ (vala-native_0.7.10.4.bb vala.inc vala_0.7.10.4.bb): vala: 0.7.10.4 -> 0.7.10.5 Mar 17 16:58:42 mikeul: no, to local.conf. devshell is a mechanism to spawn a shell inside the patched source tree of any recipe Mar 17 17:05:51 Does packaged-staging need 1 ack or 2? Mar 17 17:08:34 2, I guess Mar 17 17:08:56 Someone else wanna look at the os.unlink patch? :) Mar 17 17:09:17 * Tartarus is going to pw-am that and then change the note to a debug, about rm'ing files at all Mar 17 17:11:35 Hi, does somebody know what i have to do to enable debugging for the head.S by using OE ? Mar 17 17:12:15 ./arch/arm/boot/compressed/head.S Mar 17 17:12:37 hrw: yep, on connected up on SDIO Mar 17 17:14:49 hrw: however packaging the NVS calibration data like that recipe does is bad as that calibration data can very from batch to batch and module to module (for example, we don't even use that calibration data on the WL1251 on the OpenPandora, we read it from an EEPROM on the wifi module). Mar 17 17:15:23 DJWillis: no binary for calibration even? Mar 17 17:16:13 hrw: nope, the WL1251 is a bit of a POS and the mainline driver from Nokia, while it helps, smooths over the fact Ti would rather not support it and don't like providing help ;-) Mar 17 17:16:46 suxx Mar 17 17:17:43 root@bug20:~# busybox Mar 17 17:17:44 -sh: busybox: command not found Mar 17 17:19:29 hrw: That damm WiFi module has caused me more sleepless nights in the last year then I would like to admit to ;-). That firmware package is really HTC specific, ideally the NVS should be 'per device' at least. I'll tweak later if I get a chance, merge in the firmware recipe from the OpenPandora overlay into that new one. Mar 17 17:40:45 03Koen Kooi  07org.openembedded.dev * r6c5f9d4325 10openembedded.git/recipes/xorg-lib/ (18 files in 2 dirs): pixman git: remove patches that were applied upstream Mar 17 17:45:17 I've got an iPaq h5550 lying around that I have loaded pre-built OE images on in the past. I'd like to build an image based on GPE - can anyone tell me which would be best to build. Mar 17 17:53:56 MWelchUK: x11-image for start Mar 17 17:54:45 hrw, ah - cheers. Mar 17 18:01:51 any opinions on the OS for a OE build machine at OSUOSL? Mar 17 18:02:10 kergoth: Ah, keep learning more python as I go ;) Mar 17 18:02:22 So I guess os.walk and then os.rmdir on the dirs we get too Mar 17 18:08:53 cbrake: debian testing or ubuntu 9.10 Mar 17 18:11:00 cbrake, Fedora Mar 17 18:11:18 cbrake, real answer, which ever you are most comfortable with Mar 17 18:11:23 debian stable or fedora Mar 17 18:11:44 heh, nothing like variety :-) Mar 17 18:11:48 :) Mar 17 18:12:07 debian stable is the sanest choice Mar 17 18:12:30 fedora could get a bit annoying to upgrade on headless box Mar 17 18:12:47 * florian votes for Debian stable Mar 17 18:13:18 cbrake, what is this machines job? Mar 17 18:17:33 whatever we want Mar 17 18:17:42 just a scratch machine to do autobuilds, etc Mar 17 18:18:13 OE builds happilly on debian stable and debian stable is the sanest to admin Mar 17 18:18:35 ubuntu avoid like plague, even so called "security" updates can leave your machines unbootable Mar 17 18:18:46 I guess they did make my machine more secure Mar 17 18:20:41 heh Mar 17 18:20:44 I haven't had that happen Mar 17 18:20:54 But I've had all sorts of annoying build issues on redhat stuff Mar 17 18:21:01 That said Mar 17 18:21:24 We should probably have the actual OS be something sane / stable, and do builds in chroots of redhat/debian/whatever Mar 17 18:21:40 anyway Im off for food Mar 17 18:30:11 bye Mar 17 19:52:58 mickey|tv, do you know somebody else than florian who could help with the linuxtogo server problems? Mar 17 19:55:46 speak of the devil Mar 17 19:55:59 ~ping florian Mar 17 19:56:00 pong florian Mar 17 19:56:21 oops Mar 17 19:56:37 ~lart QSC Mar 17 19:56:37 * ibot DoSes QSC Mar 17 19:56:52 florian, who's QSC? Mar 17 19:57:30 dcordes: The provider for this internet connection. Mar 17 20:23:16 * florian runs home, bbiab Mar 17 20:27:02 03David Lanzendörfer  07org.openembedded.dev * rd695f337bd 10openembedded.git/recipes/xorg-driver/ (4 files in 2 dirs): xf86-video-msm: fix build errors Mar 17 20:34:19 03Koen Kooi  07org.openembedded.dev * re04b826e9b 10openembedded.git/recipes/gstreamer/ (9 files): Mar 17 20:34:19 gstreamer: updates Mar 17 20:34:19 * switch to new-style staging Mar 17 20:34:19 * add support for new-style checksums Mar 17 20:34:19 * move plugins to newer versions, so plugins moved (e.g. ivorbis) Mar 17 20:34:19 * fixup gstconfig header parsing Mar 17 20:42:42 I thought .ipk files were just tar.gz files containing control and data archives, yet if I try to tar -xvzf one I get a 'not in gzip format'? Mar 17 20:43:22 the old, old, really really really fucking old ipk format was tar.gz within tar.gz Mar 17 20:43:40 the current format is tar.gz files within an ar archive Mar 17 20:43:46 same as .deb Mar 17 20:43:47 ah... dated google search hits :) Mar 17 20:44:11 the ar format was relatively recent back when i started OZ Mar 17 20:45:56 kergoth, cool -t hx Mar 17 20:46:01 np Mar 17 20:46:14 03Roman I Khimov  07org.openembedded.dev * r2530a0e368 10openembedded.git/recipes/ipsec-tools/ (ipsec-tools.inc ipsec-tools_0.6.6.bb ipsec-tools_0.6.7.bb): Mar 17 20:46:14 ipsec-tools: convert to INC_PR Mar 17 20:46:14 Signed-off-by: Roman I Khimov Mar 17 20:46:15 03Roman I Khimov  07org.openembedded.dev * r6b423dc2fc 10openembedded.git/recipes/ipsec-tools/ (4 files): Mar 17 20:46:15 ipsec-tools: add version 0.7.2 Mar 17 20:46:15 Fixes CVE-2008-3651 and CVE-2008-3652. Mar 17 20:46:16 Signed-off-by: Roman I Khimov Mar 17 21:20:30 anyone done any in-linux entire root filesystem update? wondering if the right technique is to try to pivot-root to something? Mar 17 21:24:46 kergoth: what are major differences between ipk and deb ? Mar 17 21:25:01 as far as i know the file format is identical Mar 17 21:25:23 but i haven't kept up on current debian distro activity Mar 17 21:25:32 jo Mar 17 21:25:33 hmm so we could use dpkg then Mar 17 21:25:42 plus apt Mar 17 21:25:44 Not exactly? Mar 17 21:25:55 ipk trimmed a bunch of stuff out in the desc, i thought Mar 17 21:26:11 that could be, control file field differences Mar 17 21:26:13 If you want deb + apt, use package_deb not package_ipk Mar 17 21:26:17 i don't really see what you're getting at, khem Mar 17 21:26:22 there's package_deb already, exactly Mar 17 21:26:46 kergoth_: I am trying to learn adavantage of ipk Mar 17 21:26:48 Probably bitrotted a bit again and pointing at versions that don't exist anymore, of course :) Mar 17 21:27:05 apt isn't designed for embedded systems Mar 17 21:27:07 khem: Memory footprint on device, due to stripping out various bits of info Mar 17 21:27:16 busybox has dpkg, but there's no small apt Mar 17 21:27:17 Also, yes, apt is C++ is hog/leaky Mar 17 21:27:56 03Mario Domenech Goulart  07org.openembedded.dev * r5eb0c55119 10openembedded.git/recipes/chicken/chicken_4.4.0.bb: Mar 17 21:27:56 chicken: added recipe for version 4.4.0 Mar 17 21:27:56 Signed-off-by: Mario Domenech Goulart Mar 17 21:27:56 Signed-off-by: Otavio Salvador Mar 17 21:27:57 03Mario Domenech Goulart  07org.openembedded.dev * r98764601e5 10openembedded.git/recipes/xinput-calibrator/ (2 files in 2 dirs): Mar 17 21:27:57 xinput-calibrator: patch to fix miny and maxx printing order for UDEV and HAL Mar 17 21:27:58 Signed-off-by: Mario Domenech Goulart Mar 17 21:27:58 Signed-off-by: Otavio Salvador Mar 17 21:27:58 and of course, apt downloads everything before installing anything, is better about rollback (afaik), etc.. none of that is done by ipkg for good reason Mar 17 21:27:59 03Mario Domenech Goulart  07org.openembedded.dev * r3e87f78f40 10openembedded.git/recipes/chicken/chicken-cross_4.4.0.bb: Mar 17 21:27:59 chicken-cross: added recipe for chicken-cross version 4.4.0 Mar 17 21:28:00 Signed-off-by: Mario Domenech Goulart Mar 17 21:28:00 Signed-off-by: Otavio Salvador Mar 17 21:28:06 03Mario Domenech Goulart  07org.openembedded.dev * r5cca66257c 10openembedded.git/recipes/xinput-calibrator/xinput-calibrator_0.6.0.bb: Mar 17 21:28:06 xinput-calibrator: install xinput_calibrator_get_hal_calibration.sh Mar 17 21:28:06 Signed-off-by: Mario Domenech Goulart Mar 17 21:28:06 Signed-off-by: Otavio Salvador Mar 17 21:28:37 kergoth_: ipkg is now opkg ? Mar 17 21:28:45 yes Mar 17 21:29:00 hm consider both prjects dead Mar 17 21:29:19 well, someone seems to be working on opkg, at least there are responses to bugs opened on the google code page Mar 17 21:29:27 yeah, opkg is still being worked on Mar 17 21:29:28 They're even in here Mar 17 21:29:34 but i expect it may be a losing battle, that codebase is .. not good Mar 17 21:29:38 heh Mar 17 21:30:13 may be oe should embrase opkg and host it along with oe and bitbake Mar 17 21:30:23 what's the point? Mar 17 21:30:48 if its dead and needs place to live Mar 17 21:30:49 oe can emit a number of different packaging formats, i don't think it'd be good to bias like that, for one, and for two, it wouldn't gain you anything Mar 17 21:30:58 as i just said like 20 seconds ago, its on google code Mar 17 21:31:05 and has active comments in the bts there Mar 17 21:31:22 kergoth_: so you suggest to use deb or rpm? Mar 17 21:31:26 hmm ok thats fine then I didnt parse that Mar 17 21:31:30 http://code.google.com/p/opkg/ Mar 17 21:31:35 otavio: not sure what you mean. Mar 17 21:32:01 otavio: I think OE loved all of them like a mother of 3 Mar 17 21:32:02 I didn't advocate any of those at any point, so I'm not sure where you got that Mar 17 21:32:11 OE can emit all of them, just deb/rpm don't get as mjuch attention Mar 17 21:32:14 kergoth_: I meant that it being a "bad codebase" so which other would be the good one to be the "default" Mar 17 21:32:15 since the main distros aren't using them Mar 17 21:32:33 otavio: as we wre just talking about in here 5 minutes ago, there's no embedded apt Mar 17 21:32:43 deb isn't suitable for all the systems opkg is Mar 17 21:32:50 * kergoth_ points up in the scrollback Mar 17 21:32:53 kergoth_: that was my concern too that OE distro use ipk heavily and that means opkg must be maintained healthy Mar 17 21:32:55 kergoth_: indeed; and apt is a very bad guy regarding memory and like Mar 17 21:33:21 khem: The problem being most people don't want to work on opkg Mar 17 21:33:24 kergoth_: there're some efforts to work on a c based apt replacement but didn't go far yet Mar 17 21:33:25 There's a few, and they're active Mar 17 21:33:39 Tartarus: hmm ok Mar 17 21:34:06 otavio: would have to consider behavioral changes too not just code, to download then install package by package, even though its less safe, due to less area to store the cached packages Mar 17 21:34:18 C version sounds interesting, though Mar 17 21:34:20 hmm Mar 17 21:34:42 * kergoth_ goes for a walk, head is pounding Mar 17 21:34:44 well; time to go Mar 17 21:34:51 cya all Mar 17 21:34:51 nite otavio Mar 17 21:35:01 woglinde: g'night Mar 17 21:35:32 apt is better than yum atleast Mar 17 21:35:46 What is the newer way of putting checksums into the package recipe? Mar 17 21:35:47 heh - say that in #meego ;-) Mar 17 21:35:48 khem: everything is better then yum ;-) Mar 17 21:35:51 cheers ;-) Mar 17 21:36:11 zen: Roughly Mar 17 21:36:27 SRC_URI = "uri://foo.tar.gz;name=archive" Mar 17 21:36:38 SRC_URI[archive.md5sum] = "..." Mar 17 21:36:45 SRC_URI[archive.sha256sum] = "..." Mar 17 21:36:47 ah Mar 17 21:36:51 thanks Mar 17 21:36:52 s/archive/whatever/g Mar 17 21:36:54 right Mar 17 21:36:57 But I think Mar 17 21:37:10 If you use archive, you can then do just SRC_URI[md5sum] = ... Mar 17 21:37:14 If there's just one to add a sum for Mar 17 21:37:20 haven't used it myself Mar 17 21:42:38 Tartarus, worked like a charm. I do like the idea of keeping the checksums with the individual recipes. Makes a lot more sense. Mar 17 21:56:13 Hi, I am working on OMAP Zoom platform. I built an Angstrom X11 image from OE, org.openembedded.dev branch and have issues in loading the file system; I get INIT: Id "S" respawning too fast: disabled for 5 minutes" error Mar 17 21:57:12 I see /etc/init.d/rc: line 1: syntax error ... Pls let me know if there are any clues; Mar 17 21:58:57 gt3: sounds like a bad build, I don't have a zoom but I have images from the near tip of oe.dev running on my OMAP3EVM, Beagle and Pandora. No issues. Are you flashing to the NAND or running from the SD? Mar 17 21:59:26 from SD Mar 17 22:04:30 DJWillis: I am trying to add support to the new OMAP Zoom36x platform and added the machine config and modified the kernel recipes Mar 17 22:05:12 the new kernel loads up fine, but file system is hanging at INIT Mar 17 22:05:54 gt3: that should pose no major issues :-o. Mar 17 22:08:15 DJWillis: it is repeating at "INIT: Id "1" respawnign too fast" message Mar 17 22:09:39 gt3: yep, so your init is all fecked up, that's quite easy to tell, are you using a bog standard image, Arago overlay? Anything special like that? Mar 17 22:10:29 DJWillis: just a standard image.. I just did "bitbake X11-image" Mar 17 22:12:06 gt3: tried console-image with nothing special in just to be sure the basics for the image are solid, do you have anything exotic in your machine file? Mar 17 22:12:50 DJWillis: I picked the "x11-image-omapzoom36x.tar.bz2" file to use; is that the correct one? I also see a Angstrom....rootfs.tar Mar 17 22:13:27 DJWIllis: no big change on the machine; very close to omapzoom2 that is already there. Mar 17 22:14:31 hi djwillis Mar 17 22:14:35 gt3: that looks like the image you should use at a guess (omapzoom36x is you machine file). Mar 17 22:14:38 hi woglinde Mar 17 22:15:09 DJWIllis: yes Mar 17 22:17:20 gt3: so the simple test would be to try a really basic console image and see how you get on, it looks like the image build is borked so building another simple image would be a good test. Mar 17 22:18:07 DJWillis: Ok. I'll try that. Thanks! Mar 17 22:18:44 gt3: dont have a 36** board to mess about with at the moment. Mar 17 22:21:17 DJWillis: I actually took the other Angsrtrom....rootfs.tar from the same location and it loads into the console Mar 17 22:22:41 DJWillis: hope I am taking the rootfs from the correct location - "build/tmp/deploy/glibc/images/omapzoom36x" Mar 17 22:24:12 gt3: interesting, one should just be linked to the other, did you get a bad extract onto your SD I wonder. Yep tmp/deploy/ is the right place. Mar 17 22:30:19 DJWillis: Do you mean "Angstrom-x11-image-glibc-ipk-2009.X-test-20100316-omapzoom36x.rootfs" is the same as "X11-image-.." Mar 17 22:56:43 gt3: make sure that udev is creating nodes properly Mar 17 23:34:27 03Paul Eggleton  07org.openembedded.dev * rc914d3308d 10openembedded.git/conf/distro/include/sane-srcrevs.inc: opkg: update to include upstream libopkg C++ header fixes Mar 17 23:34:33 03Paul Eggleton  07org.openembedded.dev * rbad37557be 10openembedded.git/recipes/opie-packagemanager/ (files/opkg_update.patch opie-packagemanager_1.2.4.bb): opie-packagemanager: fix for recent opkg SRCREV bump Mar 18 00:19:35 devshell seems to require 'gnome-terminal'. I wonder how you do it from the command line when you aren't running X on your build machine. Mar 18 00:20:50 hmmm.... seems to be discussed exhaustively on the web. Mar 18 00:24:47 I need to poke at devshell sometime Mar 18 00:25:23 I found there is an option to use screen, so that should work for me. Mar 18 00:26:20 gnome-terminal is a bad default Mar 18 00:27:20 I was expecting it to just come to a bash prompt with the environment setup. Mar 18 00:28:48 that would be a sane default Mar 18 00:59:27 the screen solution doesn't work for me: "Cannot open your terminal '/dev/pts/16' - please check." Mar 18 01:00:25 Did you login as another user then su - to your current user. Thats usually when I see that. The other time is when I chroot in to a chrooted filesystem and don't have pts mounted. Mar 18 01:01:51 I did. Mar 18 01:02:02 is that a problem? Mar 18 01:02:23 Yes cause your pts is owned by the user you sued from. Mar 18 01:02:41 oh. :( Mar 18 01:02:55 You can chmod 666 /dev/pts/16 as the other user or root and it should work. Mar 18 01:03:01 k Mar 18 01:06:58 hmmm... that worked, but the environment wasn't right at all. Mar 18 01:13:04 Is there some way to invoke bitbake directly on a new style native recipe? Mar 18 01:14:40 With old native recipes one can do a `bitbake -b path/to/native/recipe' but that doesn't work with the new BBCLASSEXTEND native recipes Mar 18 01:14:56 odd. this time, the environment is configured. Mar 18 01:15:19 now I just need to understand what is done instead of running ./configure. Mar 18 01:18:55 urg. arm-angstrom-linux-gnueabi-gcc isn't even in the path. :( Mar 18 01:21:00 The .bashrc is messing up my environment. Mar 18 01:39:05 hallow Mar 18 01:39:09 hallo* even Mar 18 01:39:33 was wondering if there is any way to make my own version of the x11-image to include python Mar 18 01:39:39 not as an ipk Mar 18 01:39:46 but make it come preinstalled Mar 18 01:41:24 you can add it to the image you are targeting for your distro. Mar 18 01:42:44 hmm k Mar 18 01:44:31 thanks Mar 18 02:09:30 aditya_1111: IMAGE_INSTALL_append = " python" in local.conf would append it to the image_install for any image you build Mar 18 02:09:54 aditya_1111: alternatively, modify the recipe itself as jkridner suggested, or IMAGE_INSTALL_append_pn-devel-image = " python" to append just for one image from local.conf Mar 18 02:17:04 the recipes seem to have stuff like IMAGE_INSTALL = " Mar 18 02:17:13 can I just add python to the list for my recipe? Mar 18 02:17:34 or do I have to do it in local.conf? Mar 18 02:20:25 yes, you can add it to IMAGE_INSTALL in the recipe Mar 18 02:20:35 like i said, either way Mar 18 02:23:03 anyway to force glibc to rebuild without destroying the build system? Mar 18 02:23:27 you mean like bitbake -c clean glibc; bitbake glibc? Mar 18 02:23:36 the same way you'd rebuild anything else? Mar 18 02:23:37 :) Mar 18 02:23:50 if I do bitbake -c glibc Mar 18 02:23:57 I cannot rebuild glibc Mar 18 02:23:59 ... Mar 18 02:24:17 untill I jank the entire tmp folder and start over Mar 18 02:24:25 what are you talking about? Mar 18 02:24:26 wasn't like this till last thusrday Mar 18 02:24:31 try it Mar 18 02:24:37 do Mar 18 02:24:44 bitbake -c clean glibc Mar 18 02:24:50 bitbake glibc Mar 18 02:24:56 errors out on config Mar 18 02:24:59 to be absolutely sure, you could clean the entire tooclhain Mar 18 02:25:10 to ensure that glibc-initial is available to use to build the next steps Mar 18 02:25:15 dunno how to clean the toolchain Mar 18 02:25:20 bitbake -c clean gcc-cross gcc-cross-initial gcc-cross-intermediate glibc-initial glibc Mar 18 02:25:22 now you do Mar 18 02:25:35 awesome... Mar 18 02:25:48 Now, where did you cp that from so I can be a smart ass too? Mar 18 02:26:55 the reason I ask, is if there is a problem getting just ONE locale, it will fail to build images in the future Mar 18 02:27:29 up untill friday, I was able to rebuild glibc by clean and build Mar 18 02:27:39 saturday, someone put a fork in it Mar 18 02:28:55 I didn't copy it from anywhere, I started the project, and remember how the toolchain is constructed from my initial toolchain recipes and gerald britton's later rework Mar 18 02:30:03 awesome Mar 18 02:30:13 years of advantage, for a "now you do" Mar 18 02:30:28 feel good about yourself now? Mar 18 02:30:34 sorry got distracted after your reply. thanks kergoth Mar 18 02:30:38 wonderful community spirit Mar 18 02:30:55 jeeze, what an asshole Mar 18 02:31:35 "now you do" means .. now you do, here's the answer, not get your panties in a twist and freak out for no reason Mar 18 02:31:38 aditya_1111: np Mar 18 02:32:19 heh Mar 18 02:32:32 its great to meet people who feel like they are entitled Mar 18 02:32:41 you lose your faith in humanity that much more Mar 18 02:58:25 damn Mar 18 02:58:29 I hate do_fetch errors Mar 18 02:58:35 agreed Mar 18 02:59:03 http://projects.linuxtogo.org/pipermail/openembedded-devel/2008-March/004613.html could be useful in trying to find a mirror for something, when hitting one, perhaps **** ENDING LOGGING AT Thu Mar 18 02:59:56 2010