**** BEGIN LOGGING AT Fri Sep 07 02:59:56 2007 Sep 07 03:18:05 has something completely stuffed up do_package_write_ipk? my autobuilders are falling over on packages that have been stable for years ... Sep 07 03:31:56 * * OE Bug 2946 has been created by autobuild(AT)openembedded.org Sep 07 03:31:59 * * linux-rp-2.6.19+git-r5-do_fetch Sep 07 03:32:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2946 Sep 07 04:30:57 * * OE Bug 2947 has been created by autobuild(AT)openembedded.org Sep 07 04:30:59 * * angstrom-x11-image-1.0-r0-do_rootfs Sep 07 04:31:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2947 Sep 07 04:54:35 03mickeyl 07org.oe.dev * r19384222... 10/ (3 files in 2 dirs): task-openmoko-[debug|feed]: sort alphabetically, add some new packages Sep 07 05:44:35 hmm. i'm trying to build the freetype libraries and header files and make them available to other bb recipes. I've run "bitbake -c stage freetype" but it complains about do_stage not being a valid task (KeyError: 'do_stage'). Sep 07 05:44:44 however the .bb file for it does have a do_stage. Sep 07 05:45:18 if i bitbake -b org.openembedded.dev/packages/freetype/freetype_2.3.4.bb -c stage, i get a "InvalidTask event exception" Sep 07 05:50:04 -c populate_staging Sep 07 05:50:21 or in doubts -c listtasks Sep 07 05:52:28 ah. ta. Sep 07 06:14:35 hi koen Sep 07 06:14:54 hey m3l3m01t Sep 07 06:14:59 good morning all Sep 07 06:15:12 I have implemented "packaged staging" feature Sep 07 06:15:41 it create a ipk for each package Sep 07 06:16:01 named ${PN}-stage_???.ipk Sep 07 06:16:14 attach the patch to the bugtracker :) Sep 07 06:17:21 it can be automatically reused and sync with a rsync repository Sep 07 06:17:32 but it is too large Sep 07 06:18:16 and I updated bitbake either Sep 07 06:19:48 a patch is nevel too large to attach to bugzilla Sep 07 06:19:53 never* Sep 07 06:20:24 yes, I means its not easy to generate a patch Sep 07 06:20:36 'mtn diff' 'svn diff' Sep 07 06:21:04 because my work is not based on current OE repository Sep 07 06:21:20 'mtn pull ; mtn up ; mtn diff' Sep 07 06:21:24 it still use bitbake 1.6.2 Sep 07 06:22:33 I want to send you some code, from that you can see how it is designed Sep 07 06:22:51 like I said, attach it to the bugtracker Sep 07 06:24:16 OK, but it wont includes all the changes Sep 07 06:30:25 that's what 'mtn diff' and 'svn diff' are for Sep 07 06:32:52 koen, hello Sep 07 06:33:06 hey step2back Sep 07 06:33:24 does anybody do qtopia package for OE? Sep 07 06:33:38 zecke added it this week Sep 07 06:33:51 do - I do n't know correct verb. Sep 07 06:34:00 COOL! Sep 07 06:34:25 qtopia core and platform? Sep 07 06:37:59 core Sep 07 06:38:17 what about platform? Sep 07 06:38:37 no idea Sep 07 06:39:11 http://www.openembedded.org:1081/revision/browse/19384222035eabe4e0ef0f06f73f2c03d05d00df/packages/qte Sep 07 06:39:44 zecke, does plan it? Sep 07 06:42:29 oh... it was on my hdd already, :(. sorry for worry Sep 07 06:42:41 .. about qtopia core Sep 07 06:44:55 koen, thanks. Sep 07 06:49:39 i have a qtopia phone bb Sep 07 06:49:42 FYI Sep 07 07:02:53 koen, please see Bug 2948 Sep 07 07:04:33 thanks Sep 07 07:07:48 koen, please let me know if you find it is useful and interest in it. Sep 07 07:08:01 will do Sep 07 07:19:21 good morning all Sep 07 07:20:10 morning steliosk! Sep 07 07:21:17 cyberdeck : hi ! Sep 07 07:24:34 cyberdeck : btw what gcc 4 versions did you used to build the xilinx kernel that failed to boot ? 4.1.1 or 4.2 ? Sep 07 07:58:56 good morning all. Sep 07 07:59:42 Does anybody happen to know wether its possible to create an initrd with oe ? Sep 07 07:59:52 RP: I think i solved the "no locks" issue. You need to have statd running on the client as well Sep 07 08:01:59 03koen 07org.oe.dev * racbe1a36... 10/ (1 packages/linux/linux.inc): linux.inc: add configure magic for replacing bootlogo Sep 07 08:02:35 morning Sep 07 08:02:42 hey hrw Sep 07 08:07:18 steliosk: good to know Sep 07 08:11:34 morning hrw Sep 07 08:20:28 hi XorA Sep 07 08:20:31 morning Sep 07 08:20:50 hey XorA Sep 07 08:22:28 hrw : hi ! Sep 07 08:22:52 hrw : well i installed nfs-common in ubuntu and it then worked Sep 07 08:23:56 hrw : but doing some googling it showed that the lock issue is related to statd on the client Sep 07 08:24:18 server need statd too Sep 07 08:24:50 yes Sep 07 08:26:01 03ospite 07org.oe.dev * r415bbaa4... 10/ (4 files in 3 dirs): ezxd: add svn version, replaces opentapi Sep 07 08:26:06 03koen 07org.oe.dev * r54a190ec... 10/ (1 conf/machine/include/motorola-ezx-base.inc): motorola-ezx-base.inc: replace opentapi with ezxd Sep 07 08:26:13 03koen 07org.oe.dev * rbbcc5fe5... 10/ (1 conf/machine/include/motorola-ezx-base.inc): motorola-ezx-base.inc: include ezxd in task-boot instead of task-base Sep 07 08:26:19 03koen 07org.oe.dev * rab566828... 10/ (1 packages/tasks/task-boot.bb): task-boot: bump PR Sep 07 08:26:25 03koen 07org.oe.dev * r73cd7668... 10/ (44 files in 9 dirs): linux-ezx: use linux.inc to avoid duplication, update patches, add asoc patch Sep 07 08:39:52 busy week: http://lists.linuxtogo.org/pipermail/openembedded-devel/2007-September/002956.html Sep 07 08:42:58 koen: you wrote bluez-utils recipes? Sep 07 08:52:03 hrw: no, I only updated it to dbus-ified bluez according to marcel's instructions Sep 07 08:52:12 ok Sep 07 08:52:30 they have many duplicated stuff in .bb/.inc Sep 07 08:58:51 gm Sep 07 09:04:24 cbrake_away: ping ? I've got some questions about cm-x270 and openembedded usage with it Sep 07 09:06:34 hrw: indeed, fixing that is waaaaay down on my TODO Sep 07 09:08:49 Deleting temp doesn't fix anything Sep 07 09:08:51 YOU LIED Sep 07 09:08:54 :D Sep 07 09:09:10 koen: I am updating bluez in Poky so 3.17 will get a bit cleaned but no cups Sep 07 09:10:08 3.18 is out Sep 07 09:15:59 nice.. Sep 07 09:26:11 good monring Sep 07 09:26:44 hey florian Sep 07 09:31:10 gm all Sep 07 09:31:23 hi koen, hi likewise Sep 07 09:33:24 koen: bluez 3.18 in poky Sep 07 09:33:53 hi lrg Sep 07 09:34:11 florian: hey Sep 07 09:37:51 steliosk: That makes sense, I thought you said that was running on the client already? Sep 07 09:43:39 hey lrg Sep 07 09:43:54 hey koen Sep 07 09:44:51 RP: did you see http://bugs.openembedded.org/show_bug.cgi?id=2948 ? Sep 07 09:48:30 !oebug 2948 Sep 07 09:48:32 * * Bug 2948, Status: NEW, Created: 2007-09-06 23:56 Sep 07 09:48:33 * * m3l3m01t(AT)gmail.com: Packaged Staging feature Sep 07 09:48:34 * * http://bugs.openembedded.org/show_bug.cgi?id=2948 Sep 07 09:52:50 hrw: booked Restaurant Hotel Sep 07 09:54:18 steliosk: both gcc 4.1.x and 4.2.x failed to boot my kernel ;( Sep 07 09:54:47 bye Sep 07 09:55:46 03xora 07org.oe.dev * r93d42c42... 10/ (1 packages/xournal/xournal_0.4.0.1.bb): xournal/xournal_0.4.0.1.bb : version bump, bugfix release Sep 07 09:56:29 hmmm, im having problems compiling openembedded, pysqlite complains about BB_URI_HEADREVS table Sep 07 09:57:19 XorA: do you have an url for that hotel? Sep 07 09:57:45 koen: HONIGMOND on zeckes oedem2007 page Sep 07 09:57:51 ah Sep 07 09:58:03 koen: got price list if you want them Sep 07 09:58:47 €95 for a single room, right? Sep 07 09:58:57 59 Sep 07 09:59:21 http://www.honigmond-berlin.de/assets/images/autogen/a_preise1.jpg Sep 07 09:59:23 hmm Sep 07 09:59:39 koen: I asked for a quote Sep 07 10:00:04 probably quite season Sep 07 10:00:10 quiet Sep 07 10:00:15 damn I can never spell that word Sep 07 10:01:37 * koen tries to locate email address on the honingmond webpage Sep 07 10:04:26 saladino: you need pysqlite2 which supports sqlite3 Sep 07 10:04:57 RP: nop. checked the server only Sep 07 10:05:51 cyberdeck : its weird. I get both the xilinx and dht-walnut, magicbox machines which use ppc405 to build and boot fine Sep 07 10:06:55 cyberdeck : but we are using our own distro, OPLinux, although i dough this is the problem Sep 07 10:07:11 cyberdeck : what's your build machine i686/amd64 ? Sep 07 10:13:35 XorA: 59 sounds nice Sep 07 10:13:51 XorA: did they have englis page or only german? Sep 07 10:15:39 XorA: and which one of them you chosen? honigmond or garden? Sep 07 10:20:24 hrw: Restaurant Hotel Sep 07 10:20:34 hrw: only german, but they email in English Sep 07 10:21:27 XorA: Tieckstrasse 12? Sep 07 10:25:01 XorA: did they have any group disconts? Sep 07 10:25:25 hrw: I asked, but they didnt answer that question Sep 07 10:25:36 ok Sep 07 10:25:45 hrw: 59 EUR is bargain basement in UK so I wont grumble too much Sep 07 10:26:51 hrw: and about all I can say in German these days is Do you speak English Sep 07 10:27:33 'Ich sprechen Deutsch niht' is my favorite (wrong written) Sep 07 10:27:42 angstrom 2007.1 recently switched to binutils-cross-2.18. It fails on me. Sep 07 10:28:01 I think I saw a similar remark here yesterday (Crofton?) Sep 07 10:28:02 likewise: 64bit host? Sep 07 10:28:16 hrw: yes Sep 07 10:28:20 * XorA has to also practice breavhearting Scotland Sep 07 10:28:21 likewise: can you check bugtracker? we ave bug for it Sep 07 10:28:34 !oebug 2928 Sep 07 10:28:36 * * Bug 2928, Status: NEW, Created: 2007-09-05 08:32 Sep 07 10:28:37 * * philip(AT)balister.org: binutils-cross 2.18 fails to build on x86_64 Sep 07 10:28:38 * * http://bugs.openembedded.org/show_bug.cgi?id=2928 Sep 07 10:29:08 ~lart openmoko uboot Sep 07 10:29:08 * ibot takes out openmoko uboot with the trash Sep 07 10:30:42 hi, im hanging with a libz.la file not found error, i cant find that file via google nor am i able to generate that file from the zlib source. can any1 provide me with that file? Sep 07 10:30:47 need it for compiling gtk Sep 07 10:35:54 thanks. Sep 07 10:35:58 Hi, I'm having a problem building. I'm getting "configure: error: compiler support for __thread is required" in glibc-intermediate_2.5.bb and I'm not sure how to progress. There's something in the FAQ about NPTL, but adding "PREFERRED_PROVIDER_virtual/${TARGET_PREFIX}libc-for-gcc = glibc-intermediate" complains about a missing glibc version (iirc). Sep 07 10:36:12 eglibc builds look sane now, however needs multi-target testing. Sep 07 10:37:04 likewise: almost sane Sep 07 10:37:20 koen: you still hit issues (locale gen?) Sep 07 10:37:23 likewise: I don't like that it does rm STAGING/libc6.so Sep 07 10:59:54 hrw, im using pysqlite2 -> pysqlite2.dbapi2.OperationalError:no such table: BB_URI_HEADREVS while evaluating Sep 07 11:00:24 im able to open the bb_persist_data.sqlite3 db with sqlite3 commandline utility Sep 07 11:00:29 and the table exists Sep 07 11:16:04 ~lart u-boot git Sep 07 11:16:04 * ibot decapitates u-boot git conan the destroyer style Sep 07 11:16:16 um Conan Sep 07 11:16:35 ~praise the Governator Sep 07 11:16:36 All hail the Governator! Sep 07 11:17:32 what's all the uboot bashing about. I still need to switch to uboot :-) Sep 07 11:17:56 well, the git repo is giving the fetcher errors, so the parsing fails Sep 07 11:17:58 likewise: their git server died Sep 07 11:18:31 oh yeah... Sep 07 11:18:37 the reason I still can't build the davinci image ;) Sep 07 11:31:57 * * OE Bug 2925 has been RESOLVED (FIXED) by Sep 07 11:31:59 * *  Use linux decompressor instead of u-boot's for uImage kernels Sep 07 11:32:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2925 Sep 07 11:32:39 saladino: Is your sqlite >= 3.3 ? Sep 07 11:34:27 http://pastebin.ca/685440 is patch from one of angstrom users Sep 07 11:41:00 03koen 07org.oe.dev * refc0ab04... 10/ (1 packages/linux/linux.inc): linux.inc: use kernel decompressor (fast!) instead of uboot one (sloooooooow), closes #2925 Sep 07 11:41:15 koen, thanks for fixing that Sep 07 11:42:17 ok, I need to hit the road jack Sep 07 11:42:38 next stop Toronto, in between 10 hours in a car with my mother .... Sep 07 11:52:39 | make[1]: *** [nptl/others] Segmentation fault Sep 07 11:52:42 wtf? :)) Sep 07 11:52:58 thats when trying to build /armv5te-angstrom-linux-gnueabi/glibc-2.5-r7/glibc-2.5 Sep 07 11:59:56 03koen 07org.oe.dev * r820b4577... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom 2008: switch to xorg 7.3 Sep 07 12:03:22 koen: I have seen you post cairo perf diffs once. Could you get cairo "perf/" to build in OE? I added EXTRA_OEMAKE += "perf", but I think it needs a libtool-related patch. Sep 07 12:05:58 steliosk: my build machine is x86_x64 with 8GB RAM ;) Sep 07 12:11:21 cyberdeck: you may post the login accounts here :-) Sep 07 12:14:56 I get an exception when building: http://pastebin.ca/685476 Sep 07 12:15:16 seems to be related to a missing sysvinit-2.86-r35/install/sysvinit-pidof.lock file or something Sep 07 12:15:18 any ideas? Sep 07 12:15:27 trying to build angstrom for the TI DaVinci board Sep 07 12:16:11 pb in your tmp... Sep 07 12:16:42 just bitbake -c clean sysvinit and start over, I think. Sep 07 12:17:17 ok let me try that.. Sep 07 12:18:22 thanks! it seemed to have done the trick Sep 07 12:18:44 np. Sep 07 12:19:04 likewise: native build.... Sep 07 12:19:34 koen: ok Sep 07 12:19:37 koen: tnx Sep 07 12:20:16 ok, need to rm -rf tmp again... Sep 07 12:20:35 what happens if I keep the tmp/cross dir ? will it use it or not ? Sep 07 12:21:12 Or maybe I need to specify in the .conf that I have already a toolchain ? Sep 07 12:30:51 cbrake_away: still not there ? :-) (or any other cmx270 user) Sep 07 12:36:49 nud: cbrake is from US Sep 07 12:37:38 nud: hello Sep 07 12:37:48 cbrake: hello :-) Sep 07 12:38:56 cbrake: I'm trying to setup openembedded on a cm-x270w board these days Sep 07 12:38:57 nud: so you are using the cm-x270? Sep 07 12:39:36 I read quite a big part of your BEC blog entries but I guess I'm not smart enough to get it working anyway Sep 07 12:39:43 nud: you might be wondering how to program the jffs2 image ... Sep 07 12:39:50 well, I'm not even there Sep 07 12:39:53 :( Sep 07 12:40:24 I compiled angstrom-minimal-image-with-mtd-utils using linux-uclibc as a target Sep 07 12:40:37 nud: ok, what issues are you running into. The cm-x270 support in OE is mostly there, but there are still a few loose ends to tie up. Sep 07 12:40:58 nud: does it boot? Sep 07 12:41:04 I flashed the kernel using zImage-....bin, and then the ramdisk using minimalist-image-mtdutils.cpio.gz Sep 07 12:41:14 nud: nod Sep 07 12:41:20 then bootos, but I get "done, booting the kernel" and that's it Sep 07 12:41:46 nud: did you run "ramdisk on" and then "save" in the BL console? Sep 07 12:41:54 mmh Sep 07 12:41:56 oops Sep 07 12:41:56 :-) Sep 07 12:42:02 I knew I was missing something Sep 07 12:42:43 * nud tries Sep 07 12:43:03 nud: the last time I tried angstrom-minimal-image-with-mtd-utils initramfs, it did not boot all the way up, so let me know what happens. I have a custom initramfs image that is working ... Sep 07 12:43:24 yes I saw your image on your website Sep 07 12:44:08 nud: I would eventually like to getting standard stuff in OE working though ... Sep 07 12:44:22 no, it still hangs on "done, booting the kernel" Sep 07 12:44:28 or at least, it doesn't show anything Sep 07 12:44:40 I tried doing bootos "console=/dev/ttySA0" Sep 07 12:44:41 nud: can you pastebin the entire kernel boot log? Sep 07 12:44:58 nud: ohh, you are not getting any kernel messages :-( Sep 07 12:45:04 yes, that's the point Sep 07 12:45:23 nud: I think the console for the W module is console=ttyS0 Sep 07 12:45:34 nud: I have L modules Sep 07 12:46:17 what I should do is bitbake sysvinit -c rebuild and then the same for angstrom-image... right ? Sep 07 12:47:02 nud: also, leave off the "/dev/" -- I think that may mess things up, so it should be console=ttyS0,38400 Sep 07 12:48:15 nud: yes, if you change the SERIAL_CONSOLE in machine.conf. You can override this in local.conf if you want as well. Sep 07 12:48:26 yes that's what I did Sep 07 12:48:38 I don't like touching upstream files Sep 07 12:48:53 I'm rebuilding sysvinit Sep 07 12:48:58 nud: we may eventually want to create different machine files for L & W modules Sep 07 12:49:00 I'll tell you as soon as I've updated the image Sep 07 12:49:38 nud: you don't have to rebuild anything to test the kernel, just run the following at the BL prompt: Sep 07 12:49:46 03likewise 07org.oe.dev * r1e52f7b1... 10/ (1 packages/memdev2 packages/memdev2/memdev2_1.0.0.bb): devmem2: Simple program to read/write from/to any location in memory. Sep 07 12:51:04 oh yeah it works with the explicit "console=ttyS0,38400" argument \o/ Sep 07 12:51:20 bootos "console=ttyS0,38400 monitor=8 bpp=16 mem=64M mtdparts=physmap-flash.0:256k(boot)ro,0x180000(kernel),-(root);cm-x270-nand:64m(app),-(data) rdinit=/sbin/init" Sep 07 12:51:23 missing the root though Sep 07 12:51:36 likewise: how memdev2 differ from devmem2 whic is in repo? Sep 07 12:51:44 nud: you don't need a root with an initramfs, but the rdinit=/sbin/init is needed Sep 07 12:52:15 what do you mean ? Sep 07 12:52:50 cbrake: I will be able to paste you the kernel log now ;-) Sep 07 12:53:24 nud: you don't need a root= option on kernel cmdline if you are booting from an initramfs Sep 07 12:53:33 hrw: oh no :-/ Sep 07 12:53:38 ah ok Sep 07 12:53:52 I don't have anything on RAM yet but the minimal-image should give a prompt, shouldn't it ? Sep 07 12:53:53 likewise: you can always disapprove own commit ;) Sep 07 12:54:37 likewise: merge both recipes even Sep 07 12:54:52 cbrake: this is what I get: http://rafb.net/p/ASFSHF55.html Sep 07 12:55:26 everything seems fine but the mount errors and the fact I don't have a login after that Sep 07 12:55:38 what's a reason for not including ipkg by default in angstrom-minimal-image ? Sep 07 12:55:59 maybe it's due to the wrong tty in local.conf Sep 07 12:57:12 nud: looks good -- you may want to untar the *.tar version of the image, and see what is in inittab Sep 07 12:57:29 nud: and then change things until you get a getty starting on ttyS0 Sep 07 12:59:23 likewise: what's the difference between memdev2 and devmem2? Sep 07 12:59:59 koen: 07 14:52 < hrw> likewise: how memdev2 differ from devmem2 whic is in repo? Sep 07 13:00:07 little-endian, big-endian ;) Sep 07 13:01:00 cbrake: how do I generate a cpio.gz file from the untarred version ? Sep 07 13:01:21 nud: I would look at the tar version only to see what is in it Sep 07 13:01:32 nud: and change the OE build until you get what you want in the tar version Sep 07 13:01:56 nud: the tar version allows you to see whats in the image if it is not booting Sep 07 13:02:02 hrw, koen: Disapproving. Apply coffee. Sep 07 13:02:16 * nud is still in very early stage of oe-padawan Sep 07 13:04:00 03likewise 07org.oe.dev * rf545cd81... 10/ (1 packages/memdev2 packages/memdev2/memdev2_1.0.0.bb): Sep 07 13:04:00 disapproval of revision '1e52f7b1fc531c1622587e798a3e97e77aa163c8' Sep 07 13:04:00 memdev2: revert add, devmem2 is already there. Apply coffee. Sep 07 13:04:25 that was really stupid. Sep 07 13:05:16 Hi! Sep 07 13:05:20 !oebug 2931 Sep 07 13:05:22 * * Bug 2931, Status: NEW, Created: 2007-09-05 19:01 Sep 07 13:05:23 * * : libgcc-related SO link errors after recent changes to gcc.bb Sep 07 13:05:24 * * http://bugs.openembedded.org/show_bug.cgi?id=2931 Sep 07 13:05:33 koen: can you please have a look ^^^ Sep 07 13:06:58 cbrake: thanks, it's much more fun when there is an actual improvement :-) Sep 07 13:07:02 nud: add "cpio" into IMAGE_FSTYPES Sep 07 13:07:29 hrw: there is already cpio.gz Sep 07 13:07:47 I thought he was talking about editing the tar contents and then using the modified version Sep 07 13:11:33 cbrake: I've got the login shell and it seems to work Sep 07 13:12:10 nud: excellent :-) Sep 07 13:12:31 now I've got to make a NAND image Sep 07 13:12:41 psokolovsky: was libxml added to gstreamer recently? Sep 07 13:12:52 something I didn't understand: is the nand image a full system or just an "extension" to what is on the NOR disk ? Sep 07 13:13:40 psokolovsky: lots of google hits on illegal use of _div0 which is only meant for libgcc itself, not applications built with gcc. libxml is (was) one of the callers. Sep 07 13:14:38 I suspect gcc-cross doesn't recreate libgcc.a if it's already in cross/lib Sep 07 13:14:39 nud: NAND image is a full system Sep 07 13:15:00 nud: if you use the CMDLINE that is in the defconfig, and disable the ramdisk in the BL, the kernel will boot into the NAND fs Sep 07 13:15:21 nud: I recommend building the NAND filesys with regular glibc, and angstrom-console-image for starters Sep 07 13:15:24 I build an armv4 toolchain after an armv5 and I'm having troubles with clz (armv5) not being defined Sep 07 13:15:49 cbrake: ok, so I can compile a glibc angstrom-console-image and put it in the NAND memory Sep 07 13:16:31 nud: right. Eventually you may want to boot through the nor rootfs every time and do a switch-root to nand, but that is not necessary for getting started and only complicates things Sep 07 13:16:50 cbrake: cmx270 hhas 128 or 512M nand? Sep 07 13:17:00 hrw: depends which one you buy Sep 07 13:17:18 cbrake: well I didn't know how to do it and you explained how to setup a NOR one :p Sep 07 13:17:36 the update thingie is interesting though Sep 07 13:17:57 hrw: I think the devkits come with 512M, but production modules, you can order about anything you want. Sep 07 13:18:01 I'll look at it later, now I'd like to have a functionnal image on the NAND Sep 07 13:18:44 nud: yes, I'm implemented update functionality for 3 projects, and it has worked out very well. mtd-utils sure makes it easy ... Sep 07 13:19:16 btw cbrake: doesn't angstrom use the proprietary NAND flash driver from compulab ? Sep 07 13:19:58 nud: only in the 2.6.16 kernel. The 2.6.22 kernel includes a standard MTD driver that you can run jffs2 on top of. Sep 07 13:20:25 ah Sep 07 13:20:27 great Sep 07 13:20:42 I was wondering why it built despite the proprietary module being compiled using FPA Sep 07 13:21:00 nud: yes, works much better than the proprietary driver -- at least for smaller root file systems Sep 07 13:21:24 on debian I had the driver taking 50% CPU from time to time Sep 07 13:21:36 without me touching any file Sep 07 13:22:30 nud: yeah, we had a lot of trouble with it. Especially before we enabled kernel pre-emption. It would lock the system for 4-5 seconds at a time :-\ Sep 07 13:23:01 wow Sep 07 13:23:07 nothing great for responsive systems Sep 07 13:23:43 nud: jffs2 works well for filesystems < 128MB. Anything larger, the mount times can get a little long. So I usually split up the NAND into partitions. The main rootfs partition is farily small 64MB, and then I mount the other partitions later in the boot process. Sep 07 13:24:44 ok Sep 07 13:24:53 I don't know how to do that but I'll try :p Sep 07 13:25:35 nud: real time response of the system now is great -- at least for sending out rs485 packets every 30ms. Timing is very accurate. Sep 07 13:25:42 likewise: I saw google hits. what's needed is fix for OE however. it's clear that there's breakage in gcc build. Well, to start with, it would be nice if someone confirmed the issue. But I get it on 2 boxes. Sep 07 13:26:11 psokolovsky: but are we sure this is a gcc issue? Sep 07 13:26:20 cbrake: I'm doing something with sound and video, unfortunately some libs use floating points calculation Sep 07 13:26:36 I let you imagine what it gives with debian's FPA-enabled build Sep 07 13:26:52 psokolovsky: never mind, you mention the gcc build itself fails. Sep 07 13:26:58 likewise: if the problem is in libgcc, whose the issue then? no, I give only 80% probability. and I rebuild gcc/glibc/binutils from scratch of course Sep 07 13:27:16 likewise: gcc build doesn't fail Sep 07 13:27:53 hi mickeyl Sep 07 13:28:03 hi mickeyl Sep 07 13:28:27 psokolovsky: From what I gather, _div0() must not be used outside of gcc/libgcc. So I would first see if that happens or not. Sep 07 13:28:44 hello mickeyl Sep 07 13:29:05 Dammit a lot of websites are down today Sep 07 13:29:15 jpeg website is down Sep 07 13:29:21 likewise: well, I don't feel like debugging such stuff. just would like someone to confirm, and hope that gcc master, koen, can beat it out Sep 07 13:29:25 hey mickeyl Sep 07 13:29:34 morning guys Sep 07 13:29:41 i have a slight problem with the toolchain Sep 07 13:29:43 anyone ever seen: Sep 07 13:30:01 psokolovsky: I built armv4 r6, then armv5te r6 and it is still working Sep 07 13:30:16 hrw: can you run command in that ticket? Sep 07 13:30:31 "Source object liblto.a(rsa.o) has EABI version 0, but target Sep 07 13:30:31 gllin has EABI version 4" Sep 07 13:30:44 hrw: bitbake gst-plugins-base -c rebuild Sep 07 13:30:52 mickeyl: seen that Sep 07 13:31:09 mickeyl: it means that the lib has been compiled with the wrong flags Sep 07 13:31:22 psokolovsky: started witout -crebuild Sep 07 13:31:32 koen: right. which flags should i tell them to use? Sep 07 13:31:53 mickeyl: at this point I'd really say "use OE" Sep 07 13:32:04 psokolovsky: it was built here - now it packaging Sep 07 13:32:11 "EABI version 0" sounds a lot like oabi Sep 07 13:32:14 psokolovsky: rebuilding now Sep 07 13:32:27 hrw: can you rebuild? you could have it build before recent gcc change Sep 07 13:32:34 koen: well, they did use OE... they used the toolchain i provided Sep 07 13:32:43 psokolovsky: no - that toolchain is r6 Sep 07 13:33:11 psokolovsky: I built it on wed-fri Sep 07 13:34:45 mickeyl: then tell them to not use -mabi=apcs and friends Sep 07 13:34:57 mickeyl: uboot has(had?) the same problem Sep 07 13:34:58 koen: ok, worth a try Sep 07 13:35:26 hrw: so, did rebuild succeed? Sep 07 13:36:08 I suppose at worse gllin could be -static :-) Sep 07 13:36:18 psokolovsky: in progress Sep 07 13:36:42 koen: so EABI version 0 ist just a code name for OABI ? Sep 07 13:36:47 koen: is that an accepted fact? ;) Sep 07 13:38:02 psokolovsky: NOTE: package gst-plugins-base-0.10.14-r2: task do_compile: completed Sep 07 13:38:20 the EABI version 0 message gets triggers by -mabi=apcs, which is OABI Sep 07 13:41:00 hrw: weird. ok, thanks for confirmation. time to kill tmp/, I guess... Sep 07 13:42:44 maybe Sep 07 13:43:11 koen: please make autobuild uploading logs Sep 07 13:43:24 hrw: fix seppuku Sep 07 13:43:32 koen: #2946 is useless without it Sep 07 13:43:47 ERROR: This autoconf log indicates errors, it looked at host includes. Rerun configure task after fixing this. Path was '/home/leon/sandbox/x86core2/openembedded/build/tmp/work/i686-angstrom-linux/gst-plugins-base-0.10.14-r2/gst-plugins-base-0.10.14' Sep 07 13:43:48 hrw: that's a seppuku problem Sep 07 13:44:08 hrw: see http://bugs.openembedded.org/show_bug.cgi?id=2945 Sep 07 13:44:12 same autobuilder Sep 07 13:44:42 hrw: I suspect do_fetch either doesn't leave logs, or fails inside bitbake, which seppuku can't catch Sep 07 13:48:15 bye all Sep 07 13:48:52 cu hrw Sep 07 13:50:37 well, for some reason my angstrom-minimal-image-with-mtd-tools doesn't contain mtdtools Sep 07 13:50:51 could it be because of the lack of IMAGE_INSTALL definition in it , Sep 07 13:50:52 ? Sep 07 14:03:26 nud: right, the format of the image recipies has changed Sep 07 14:03:41 nud: so it should be IMAGE_INSTALL += ... Sep 07 14:04:34 nud: I'll push a fix as soon as I get my tree updated ... Sep 07 14:06:10 cbrake: ah, ok, thanks Sep 07 14:12:39 XorA: so you booked the Garden Hotel yet? Sep 07 14:13:13 mickeyl: Restaurant Hotel, its 1/3 the price Sep 07 14:13:45 XorA: but still Honigmond? Sep 07 14:13:58 mickeyl: yeah Sep 07 14:14:01 ok Sep 07 14:14:05 * mickeyl tries to book that now Sep 07 14:14:18 single room they quoted me 59 EUR Sep 07 14:14:28 for which nights? Sep 07 14:14:36 6 thru 9? Sep 07 14:14:49 arrive 5th, leave 11th Sep 07 14:15:11 arriving 8th, leaving 11th Sep 07 14:16:36 its a bit of a hike for me to berlin so I prefer to be able to use 6th and 11th to do OE stuff Sep 07 14:16:48 10th Sep 07 14:17:16 BTW how far is airport to OEDEM location? Sep 07 14:17:23 since when does bitbake -v no longer prints the paths? Sep 07 14:17:27 i.e foo -> bar -> foobar Sep 07 14:17:57 * * OE Bug 2931 has been RESOLVED (INVALID) by Sep 07 14:17:59 * *  libgcc-related SO link errors after recent changes to gcc.bb Sep 07 14:18:00 didn't rp add another swithc for that? Sep 07 14:18:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2931 Sep 07 14:18:57 * koen mails honingmond to get a quote, but still reserves a spot at casa holger Sep 07 14:19:52 I know a nice thai place to eat in kreuzberg :) Sep 07 14:19:59 i can't find anything except -v Sep 07 14:20:02 yay, I like Thai Sep 07 14:20:19 it's a schnell-imbiss, so pretty cheap as well Sep 07 14:20:37 hi Sep 07 14:21:43 i'm having problems on openmoko-image build with package binutils-cross-2.18 which fails on do_populate_staging cause of missing file: tmp/cross/lib/libiberty.a Sep 07 14:21:53 anyone familiar with this problem? Sep 07 14:24:27 it seems OEDEM is pretty close to monsieur vuong as well Sep 07 14:24:50 nud: image is fixed -- "mtn pull && mtn up" to get the fix Sep 07 14:25:11 cbrake: you're my hero ! Sep 07 14:25:26 koen: ;) Sep 07 14:25:34 koen: I will not propose to go there this year Sep 07 14:26:19 zecke: I can take xora there on the 5th :) Sep 07 14:26:35 take me with you Sep 07 14:29:15 * RP would prefer to try somewhere different to monsieur vuong this year :) Sep 07 14:29:20 :D Sep 07 14:29:38 * RP should really think about hotels and flights and things... Sep 07 14:29:41 there should be enough italian restaurants in berlin Sep 07 14:30:01 It doesn't have to be Italian :) Sep 07 14:30:37 once I learned that german restaurants serve eisbein, I'll go for italian or oriental ;) Sep 07 14:30:47 hahaha Sep 07 14:31:10 i rarely find any germans who love those "german" meals Sep 07 14:32:31 mickeyl: prop. you should go asking some bavarians, they might ... ;) Sep 07 14:32:46 mickeyl: Thats interesting, I'd guess most people here don't mind tradiaitonal English food... Sep 07 14:33:03 RP: indian food? Sep 07 14:33:10 indian food rocks Sep 07 14:33:14 Lan Sep 07 14:33:18 Lamm madras Sep 07 14:33:19 mmh Sep 07 14:33:59 Every time I get back from work I say to myself you have to try that restaurant... there is indian restaurant in my house... Sep 07 14:34:08 zecke: As long as they don't do curry :/ Sep 07 14:35:10 there's a lot more to indian food than curry Sep 07 14:35:39 RP: is there still a way to see the paths (foo->bar) w/ current BitBake? Sep 07 14:35:52 or should we all use -g ? Sep 07 14:36:06 mickeyl: I know, I just can't stand even the smell of it which makes trying anything else indian tricky... Sep 07 14:36:12 mickeyl: -g I think Sep 07 14:36:13 koen: eisbein? Sep 07 14:36:39 * XorA is kinda ruined on the british take on Indian food, but I love Tandori Sep 07 14:36:45 XorA: pigleg with extra servings of blubbery fat Sep 07 14:36:53 mickeyl: The problem is we do all the dependency computation in advance of all possible combinations so the old way of printing things wasn't really effecitve anymore Sep 07 14:36:59 koen: they should sell that in Scotland, deep fried Sep 07 14:37:05 :) Sep 07 14:37:27 XorA: Are you sure they don't already? Sep 07 14:37:34 RP: :) Sep 07 14:37:58 RP: I can hook you up with a special diet. Stefan and Daniel started to like it... Sep 07 14:38:19 zecke: What kind of special diet? Sep 07 14:38:46 http://www.fresenius-kabi.com/internet/kabi/gb/fkintpub.nsf/Content/Product%20Features%20Fresubin%20original%20drink Sep 07 14:40:09 zecke: :) Sep 07 14:41:50 zecke: that sounds hideous and I should avoid Sep 07 14:42:15 XorA: that was almost my only food for a year Sep 07 14:42:58 anyone up for answering a bitbake question? Sep 07 14:43:34 morning Sep 07 14:43:34 dotmef: try asking Sep 07 14:43:35 zecke: how did that happen? ;) Sep 07 14:43:39 oh - and hello all :) Sep 07 14:43:44 hi chouimat Sep 07 14:43:57 florian: darm kaputt :) Sep 07 14:44:25 zecke: oh Sep 07 14:44:26 03cbrake 07org.oe.dev * r600ac42c... 10/ (1 packages/images/angstrom-minimal-image-with-mtd-utils.bb): angstrom-minimal-image-with-mtd-utils: change RDEPENDS to IMAGE_INSTALL Sep 07 14:44:48 RP: sure thing. would a comment in the middle of a SRC_URI definition break a BB recipe? Sep 07 14:45:07 florian: http://de.wikipedia.org/wiki/Morbus_Crohn Sep 07 14:45:33 dotmef: In theory yes, in practise bitbakes parser is a bit lax and people have taken advantage of that Sep 07 14:46:08 zecke: Oh nasty... I know that quite well, a good friend of mine suffers from that too. Sep 07 14:47:09 RP: ah. thanks. is it worth raising a bug on? webkit-gtk has such a comment. Sep 07 14:47:53 dotmef: Its never seemed to cause a problem so far... Sep 07 14:49:19 RP: I suppose it could be BB version dependent then Sep 07 14:50:14 RP: it's been killing my Moko builds, but I've changed the recipe to remove the comment Sep 07 14:50:54 dotmef: Killing them how? Sep 07 14:51:25 RP: it doesn't seem to build at all, and skips straight to packaging, which fails Sep 07 14:52:22 RP: er, sorry - I meant webkit-gtk doesn't seem to build Sep 07 14:57:04 dotmef: Looking at the file, I doubt that comment is the problem Sep 07 14:57:11 gah Sep 07 14:57:15 online reservations broken Sep 07 14:57:45 RP: yep - I think you maybe right. although, removing the comment fixed that package. Sep 07 14:58:27 RP: the libmokogsmd2 package failed in a similar way, and doesn't have such a comment in its .bb Sep 07 14:58:52 * koen guesses it's the pkgdata changes Sep 07 15:00:35 sounds a bit like it Sep 07 15:00:45 koen - sorry, I'm a bit n00bish. is pkgdata a meta-data format within BB? Sep 07 15:04:57 dotmef: We made a chance recently and the commit message advised that people would need to rebuild a certain amount of their tree Sep 07 15:04:59 it's something we moved around last tuesdayy Sep 07 15:07:01 RP, koen: right - yep. I had to chuck the accumulated tmp build files around then Sep 07 15:07:27 there, train tickets for OEdem booked Sep 07 15:08:03 dam you guys with no water in the way Sep 07 15:08:09 +n Sep 07 15:08:44 XorA: swim at this time of year is quite nice , so i hear Sep 07 15:09:05 ade|desk: I better book swimming lessons then Sep 07 15:09:26 i'll get my wife to teach you Sep 07 15:09:47 * ade|desk parcels up a debbie package Sep 07 15:10:17 grr postage is crazy for live creatures Sep 07 15:10:54 koen, RP: based on what you've said, I tried a simple "touch" of the .bb. this seems to fix the issue Sep 07 15:10:57 * * OE Bug 2949 has been created by nicolas.ferre(AT)rfo.atmel.com Sep 07 15:10:59 * * vlc-gpe do_compile error modules/gui/pda/pda.c Sep 07 15:11:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2949 Sep 07 15:11:14 him all! Sep 07 15:11:19 *hi all! Sep 07 15:11:22 is it ok to use device such as adm232a on 3.3v lines? Sep 07 15:11:37 (5v rs232 line driver) Sep 07 15:11:56 * dotmef slaps himself on the back of the head Sep 07 15:13:16 XorA: Learning to swim was mandatory when I to school... Sep 07 15:13:39 * RP has wondered whether thats just a coastal thing Sep 07 15:14:48 RP: Stranraer has a coast, it also didnt have a public pool Sep 07 15:15:12 RP: I also have injuries that make swimming difficult Sep 07 15:15:23 XorA: ah :( Sep 07 15:15:41 wierdly its an injury that doesnt bother me for any other exercise :-) Sep 07 15:16:39 XorA: Whilst I once could swim, I discovered last year that my change in size in the intervening 15+ years means I really need to learn again... Sep 07 15:17:01 RP: heh heh Sep 07 15:17:27 RP: we invented boatds for a reason, because water is cold :-) Sep 07 15:17:57 XorA: yes, at least in this country :) Sep 07 15:18:22 hehe Sep 07 15:18:45 * XorA has been in the med, relative to the 40C air temp, the water still felt freezing :-D Sep 07 15:19:32 the water is never cold, its purely refreshing , cough Sep 07 15:19:37 florian: yesterday vlc-gpe compilation problem solved : Sep 07 15:20:10 florian: http://bugs.openembedded.org/show_bug.cgi?id=2949 Sep 07 15:21:50 XorA: Its heat I can't stand, I can manage cold water much better :) Sep 07 15:22:34 lobster ? Sep 07 15:23:12 noglitch: ah great... even if I wonder why this is necessary at all - if it helps i'll apply it. Sep 07 15:23:37 Anyone here who does not agree? Sep 07 15:24:50 florian, noglitch: PATH should be setup to use that anyway. Is the configure script doing something nasty to PATH? Sep 07 15:27:07 cbrake: last time I annoy you today: do you have a sample image bb file to setup the partitions and such ? Sep 07 15:29:29 has anyone mentioned libts problems on install lately, i just reflashed my c860 and libts-0.0 is trying to overwrite files from libts-1.0 Sep 07 15:30:05 nud: parititions are already set up in the kernel: cat /proc/mtd Sep 07 15:30:06 03koen 07org.oe.dev * r3b618e17... 10/ (3 files in 3 dirs): dbus: remove --reverse (yes, that is bad) to fix shutting down of dbus services on halt/reboot Sep 07 15:30:09 03koen 07org.oe.dev * r3b618e17... 10/ (3 files in 3 dirs): dbus: remove --reverse (yes, that is bad) to fix shutting down of dbus services on halt/reboot Sep 07 15:30:30 03koen 07org.oe.dev * r3b618e17... 10/ (3 files in 3 dirs): dbus: remove --reverse (yes, that is bad) to fix shutting down of dbus services on halt/reboot Sep 07 15:30:33 nud: so to program a partition, just use the correct mtdx -- same with mounting Sep 07 15:30:57 cbrake: but you were speaking about dividing the NAND into several partitions ? Sep 07 15:31:11 nud: right, this is done in the kernel Sep 07 15:31:15 ah, ok Sep 07 15:31:51 koen, RP - thanks for your help earlier. Sep 07 15:31:58 cheerio, all Sep 07 15:32:07 nud: the 2.6.22 has this all done for you assuming you use the default defconfig. The partitioning is done on the kernel command line -- for both NOR and NAND flash. MTD is great Sep 07 15:32:59 any guesses if logfs will be ready any time soon? Sep 07 15:33:47 cbrake: ok Sep 07 15:33:55 thanks Sep 07 15:35:33 which head in the mtn repository should i be pulling from? Sep 07 15:35:51 koen: BTW where is epiphany .bb's :-D Sep 07 15:36:18 and ephy-webkit's ? ;-) Sep 07 15:41:34 koen: still no modules.dep for me :-( Sep 07 15:42:13 koen: did a clean build, only diff is to force binutils for my amd64 Sep 07 15:42:31 koen: http://rafb.net/p/Oga9MC37.html Sep 07 15:43:09 any ideas? Sep 07 15:47:46 hm.. can anybody give me a tip why non-subscribers are allowed to post to openembedded-commits@ ? Sep 07 15:47:56 * * OE Bug 2950 has been created by wirelessdreamer(AT)gmail.com Sep 07 15:47:58 * * libts version missmatches Sep 07 15:48:00 * * http://bugs.openembedded.org/show_bug.cgi?id=2950 Sep 07 15:48:07 there is spam coming in via that hole Sep 07 15:52:18 noglitch: Anything intersting in that vlc config script? Sep 07 15:55:34 florian: well, I am no so used to go that deep into configure scripts... I am trying Sep 07 15:56:22 noglitch: maybe just grep for PATH in the toplevel source directory Sep 07 16:23:52 night all Sep 07 16:26:43 florian: I made a diff between both log.do_configure and the only difference is this : http://pastebin.ca/685831 at the head of the file Sep 07 16:28:30 florian: sorry but I must leave... maybe I we can go further next week Sep 07 16:30:44 noglitch: same here ok... Sep 07 16:31:12 florian: the difference appears in the log.do_configure when it compiles ok (to make myself clear) Sep 07 16:32:00 noglitch: ok, that needs some more investigating... it looks like it really does somethign nasty Sep 07 16:32:10 bbiab+ Sep 07 16:32:27 good WE all... bye ! Sep 07 16:35:42 XorA|gone: I'll add bleeding gtk and glib this weekend for epi **** ENDING LOGGING AT Fri Sep 07 17:09:29 2007 **** BEGIN LOGGING AT Fri Sep 07 17:09:50 2007 Sep 07 17:30:06 drat, the restaurant hotel seems fully booked Sep 07 17:33:52 or rather, the cheap Xora rooms are Sep 07 17:36:51 koen: there is a hostel in the novalis-street. It looks crappy aeh cheap ;) Sep 07 17:55:24 hi folks.. has anyone built pygame recently? I have having trouble getting a dependency, smpeg, to fetch. using a the dev branch Sep 07 18:03:42 hi, I'm considering to get some XScale (iwmmxt) powered device, does anyone know exact Motorola A1200 hardware specs? Or would it be better to try getting Sharp Zaurus instead? A phone would be probably more useful for everyday use though :) Sep 07 18:10:48 ssvb: the default firmware on the moto is crap, and the replacement (openezx) isn't functional yet Sep 07 18:17:23 koen: do you know offhand which recipe should be creating modules.dep? Sep 07 18:17:48 I have some time this afternoon to debug and would like a good place to start :-) Sep 07 18:26:54 re Sep 07 18:27:20 koen: thanks for the information, I'm checking openezx wiki now Sep 07 18:53:29 sakoman: afaik it's done in the kernel-module postinsts Sep 07 18:58:57 OK, thanks! Sep 07 19:00:10 fontconfig-native 2.3.91 (and, I believe, fontconfig 2.3.91) doesn't seem to compile anymore... Sep 07 19:00:16 I've posted a bug report Sep 07 19:01:25 ...which is kind of retarded, sorry Sep 07 19:03:56 * * OE Bug 2951 has been created by B_Linuz(AT)yahoo.com Sep 07 19:03:58 * * fontconfig-native 2.3.91 fails to build Sep 07 19:04:00 * * http://bugs.openembedded.org/show_bug.cgi?id=2951 Sep 07 19:06:27 bye Sep 07 19:07:16 bye Sep 07 19:12:51 evening Sep 07 19:18:28 was anyone (or did anyone try), apart from Crofton who created it, to build an angstrom image for TI DaVinci? Sep 07 19:18:40 aem, nice sentence hehe Sep 07 19:18:44 was anyone successfull in that, I mean Sep 07 19:20:09 hi, shouldn't the news be sorted in reverse on the openembedded.org page Sep 07 19:20:30 first i thought, oh bitrot, when i saw the page the first time Sep 07 19:20:41 last update 2004-10-26 Sep 07 19:21:09 it's at least pretty uncommon 8-) Sep 07 19:24:20 Sorry about adding a pastebin URL instead of the log itself, I thought it would be construed as spam Sep 07 19:26:01 Jin^eLD: after crofton added it the uboot people fucked up their git tree Sep 07 19:26:58 koen: yes.. seems so... I was trying to workaround that by disabling the build of uboot and by using uboot-utils Sep 07 19:27:02 since I do not really need the uboot itself Sep 07 19:27:10 right Sep 07 19:27:14 but the davinci kernel bb is based on omap and that has some weird stuff in it Sep 07 19:27:17 so I got lost there Sep 07 19:27:52 he uses linux-omap.inc which has a weird DEPENDS line Sep 07 19:28:59 the board is ready on my desk, I'd really like to test the build... so any help would be greatly appreciated Sep 07 19:31:11 hmm lets see, put the uboot-utils into the omap.inc Sep 07 19:31:17 maybe that will be enough Sep 07 19:39:37 RP: ping Sep 07 19:54:46 poky for the 770 dream or not? Sep 07 19:58:19 morning Sep 07 19:58:22 keesj: dream Sep 07 20:01:39 morning hrw! Sep 07 20:02:20 ~curse remote scm checking Sep 07 20:02:21 May the fleas of a thousand camels infest your most sensitive regions, remote scm checking ! Sep 07 20:02:45 bitbake --skip-scm-checking would be nice Sep 07 20:03:14 koen: for 2951 solution can be using fontconfig(-native) 2.4.2 from Poky Sep 07 20:03:16 I can't use OE at work because of that Sep 07 20:03:44 not only you I think Sep 07 20:04:01 Like I said, it shouldn't be a parseerror Sep 07 20:04:14 or we revert the 'halt-on-parse-error' patch Sep 07 20:04:24 firewalls are so "last century" Sep 07 20:04:45 I don't know if fontconfig 2.4.1 fails, but 2.3.95 does Sep 07 20:06:32 ERROR: Multiple files due to be built which all provide libpcap (/home/hrw/devel/oe/org.openembedded/packages/libpcap/libpcap_0.9.6.bb /home/hrw/devel/oe/org.openembedded/packages/libpcap/libpcap_0.9.3.bb) Sep 07 20:06:49 that type of errors needs to be tracked Sep 07 20:07:25 I get that with libxdmcp and libxau Sep 07 20:07:48 tcpdump/tcpdump_3.9.3.bb:DEPENDS = "libpcap-0.9.3" Sep 07 20:08:29 hrw: that's the wrong way to specify versioned depends Sep 07 20:08:42 koen: but thats working way Sep 07 20:08:50 and unofficial and bad Sep 07 20:08:50 (you know that, I'm just saying it for the IRC logs) Sep 07 20:08:51 hrw: rm -rf tmp/ fixed my issue. again, thanks for checking Sep 07 20:08:57 psokolovsky: np Sep 07 20:11:01 koen: did you built bluez-gnome 0.14? Sep 07 20:11:18 yes Sep 07 20:11:26 and I just installed it on my neo Sep 07 20:12:20 koen: http://ewi546.ewi.utwente.nl/tinderbox/showlog.pl?machine_id=388&logfile=20070907204507.log Sep 07 20:14:21 make[3]: execvp: /home/hrw/devel/build/angstrom/tmp/work/armv4t-angstrom-linux-gnueabi/bluez-gnome-0.14-r1/image/usr/share/applications: Permission denied ? Sep 07 20:15:03 it should be a problem.. Sep 07 20:15:11 building it again now Sep 07 20:15:34 * koen returns to his lego Sep 07 20:18:00 !oebug 2952 Sep 07 20:18:02 * * Bug 2952, Status: NEW, Created: 2007-09-07 13:16 Sep 07 20:18:03 * * : Some tasks are skipped during build Sep 07 20:18:04 * * http://bugs.openembedded.org/show_bug.cgi?id=2952 Sep 07 20:18:13 who also has this problem? Sep 07 20:27:01 03hrw 07org.oe.dev * ra12d2158... 10/ (3 files in 2 dirs): tcpdump: update to 3.9.7 whic do not need libpcap 0.9.3 but builds with 0.9.6 Sep 07 20:27:10 03hrw 07org.oe.dev * r83398e13... 10/ (1 classes/base.bbclass): base.bbclass: change message about lack of checksums.ini entry for URI to something easier to understand Sep 07 20:27:11 bye Sep 07 20:33:25 anyone out there who has ported gtkpod to oe ? Sep 07 20:53:18 does anyone know what ipk startx gets put in? Sep 07 20:55:33 xinit Sep 07 20:57:31 thansk Sep 07 20:57:34 *thanks Sep 07 20:59:42 no problem :) Sep 07 21:01:43 Can I build both glibc and uclibc images in one workspace? Or do I need to have separate workspaces? Sep 07 21:02:59 ynezz: you can Sep 07 21:03:02 I've just built succesfuly glibc angstrom-minimal and want to built uclibc, but it fails on gcc-cross-4.2.1-r3 Sep 07 21:25:33 Zero_Chaos: pong Sep 07 21:26:56 RP: I have some updates for the kernel to include new drivers and some wifi patches. Can I get an email address for a mtn diff? Sep 07 21:27:43 Zero_Chaos: I'm fairly sure my email address is all over the oe lists Sep 07 21:28:04 okay, didn't know if you had a preference, I'll pull it off the dev list Sep 07 21:35:32 03stefan 07org.oe.dev * r76f17964... 10/ (7 files in 5 dirs): opentapi: Remove as superseeded by ezxd Sep 07 21:38:24 can I get an mtn merge? thanks Sep 07 21:40:33 thanks\ Sep 07 21:58:25 gcc-cross-4.2.1-r3 fails for me with uclibc, http://pastebin.ca/686293 Sep 07 22:20:29 I'll check if bug 2951 is due to freetype Sep 07 22:21:22 Um, it probably is, right? Sep 07 22:36:21 Yeah, it does Sep 07 22:36:52 Um, bug 2951 pretty much closed Sep 07 22:37:15 Or at least, fix known Sep 08 02:54:12 Hey Crofton! **** ENDING LOGGING AT Sat Sep 08 02:59:57 2007