**** BEGIN LOGGING AT Thu Jul 29 02:59:57 2010 Jul 29 03:06:44 03Chris Larson  07org.openembedded.dev * r73ae235a69 10openembedded.git/classes/sanity.bbclass: Jul 29 03:06:44 sanity.bbclass: kill usage of 'print' Jul 29 03:06:44 Signed-off-by: Chris Larson Jul 29 03:16:04 hi Jul 29 03:16:25 I am getting error in bitbake fso-illume-image Jul 29 03:16:35 the present error is Jul 29 03:16:49 unknown package gst-plugin-modplug Jul 29 03:16:52 opkg_install_cmd: Cannot install package gst-plugin-modplug Jul 29 03:17:09 khem: hi Jul 29 03:17:19 any one awake here..? Jul 29 03:27:58 can anyone here tell me how to bitbake and make a developement image. ie. one with the include dir filled with headers for native compilation on the target **** BEGIN LOGGING AT Thu Jul 29 05:10:14 2010 **** BEGIN LOGGING AT Thu Jul 29 05:24:17 2010 **** ENDING LOGGING AT Thu Jul 29 05:37:22 2010 **** BEGIN LOGGING AT Thu Jul 29 05:54:14 2010 Jul 29 06:29:06 woglinde: (mpfr) this is more a gcc problem I think... even the old one (2.3.2) does not build anymore Jul 29 06:33:26 woglinde: http://pastebin.ca/1910737 (error in do_compile) Jul 29 06:54:44 hi mmoku Jul 29 06:58:41 hmm... Jul 29 06:58:51 nothing in mysql.err... Jul 29 06:59:09 is there something I should pass to mysqld for extra logging? Jul 29 07:00:31 porcoesphino whats up? Jul 29 07:00:38 haha Jul 29 07:00:40 oops Jul 29 07:00:45 wrong channel Jul 29 07:00:50 still, you might have some ideas Jul 29 07:01:09 I've altered mysql5_blah.inc to include ssl support Jul 29 07:01:16 works fine locally but not remotely Jul 29 07:01:25 sorry I will touch mysql only again when I get pid well for the pain Jul 29 07:01:29 I can connect without ssl remotely Jul 29 07:01:37 paid Jul 29 07:01:47 haha Jul 29 07:01:49 yeah Jul 29 07:01:50 fair enough Jul 29 07:01:56 enable remote access Jul 29 07:02:02 but any ideas for debugging why remote wont work? Jul 29 07:02:05 ? Jul 29 07:02:08 check lsof -i Jul 29 07:02:10 woglinde: hi, there was some way to turn off thumb mode for one recipe only? Jul 29 07:02:19 I can connect to the database without ssl remotely... Jul 29 07:02:38 to see which ports are open Jul 29 07:02:38 mrmoku grep for INSTRUCTION in the recipes Jul 29 07:02:43 sorry... "check lsof -i"? file or command? Jul 29 07:03:04 ha Jul 29 07:03:06 woglinde: k, thanks Jul 29 07:03:09 good point.. Jul 29 07:03:09 lsof -i Jul 29 07:03:09 who made you unix admin? Jul 29 07:03:09 or use good old netstat Jul 29 07:03:11 03Roman I Khimov  07org.openembedded.dev * r86cd1c3033 10openembedded.git/recipes/at/ (8 files in 2 dirs): Jul 29 07:03:11 at: add version 3.1.12 Jul 29 07:03:11 Thanks to Buildroot guys for patches. Jul 29 07:03:11 Signed-off-by: Roman I Khimov Jul 29 07:04:14 haha Jul 29 07:04:28 I'm a coder, I know shit about networking :( Jul 29 07:04:52 o.O Jul 29 07:05:02 even as coder you should knowabout networking Jul 29 07:05:09 otherwise you are coding shit Jul 29 07:05:16 haha Jul 29 07:05:19 with no understandings Jul 29 07:05:22 depends what you're coding Jul 29 07:05:46 and I'm learning, just still learning the tools Jul 29 07:05:55 03Roman I Khimov  07org.openembedded.dev * ra8f765e41e 10openembedded.git/recipes/iproute2/iproute2_2.6.34.bb: Jul 29 07:05:55 iproute2 2.6.34: install tc Jul 29 07:05:55 Wasn't installed for some reason. Jul 29 07:05:55 Signed-off-by: Roman I Khimov Jul 29 07:06:28 he learn postgres and sqlite Jul 29 07:06:35 woglinde: ok, with ARM_INSTRUCTION_SET = "arm" Jul 29 07:06:45 mpfr builds again for armv4t Jul 29 07:07:09 good evening Jul 29 07:07:10 03Roman I Khimov  07org.openembedded.dev * r6adaccf006 10openembedded.git/recipes/dropbear/ (dropbear.inc dropbear/init): Jul 29 07:07:11 dropbear: use pidfile for daemon start/stop/restart Jul 29 07:07:11 Old init script killed all dropbear processes when doing stop/restart Jul 29 07:07:11 including open SSH sessions which is very annoying. Jul 29 07:07:11 Signed-off-by: Roman I Khimov Jul 29 07:07:24 mrmoku yes thats what I thought yesterday evening Jul 29 07:08:00 where should I put "IMAGE_ROOTFS_SIZE" variable? putting it in the local.conf does not make any differences.... Jul 29 07:08:48 I use it this way "IMAGE_ROOTFS_SIZE_ext2 = 1024000", but this value is not used anywhere Jul 29 07:09:13 03Roman I Khimov  07org.openembedded.dev * r9475bf23d1 10openembedded.git/recipes/perl/ (perl-rdepends_5.10.1.inc perl_5.10.1.bb): Jul 29 07:09:13 perl 5.10.1: add perl-module-utf8-heavy dependency Jul 29 07:09:13 Signed-off-by: Roman I Khimov Jul 29 07:09:26 igork in the machinefile Jul 29 07:09:38 woglinde: want to point me to some docs on ports, ssl and mysql or am I on the wrong track there? Jul 29 07:10:00 yes google for it Jul 29 07:10:16 hehe Jul 29 07:10:16 there should be plenty stuff how to enable it Jul 29 07:10:22 fair point :P Jul 29 07:10:29 in the openembedded/conf/machines/, right? Jul 29 07:10:50 lsof -i shows you on which port and ips mysql runs Jul 29 07:11:00 if the ssl port is only on localhost Jul 29 07:11:21 igork exactly Jul 29 07:12:06 okay I have to leace the train Jul 29 07:12:07 yeah, lsof doesn't appear to be installed on the beagle. is there an equivilant flag for netstat? Jul 29 07:12:10 till later maybe Jul 29 07:12:26 opkg install lsof Jul 29 07:12:33 netstat -na Jul 29 07:12:36 thx Jul 29 07:12:49 dammit >_< Jul 29 07:19:49 03Klaus Kurzmann  07org.openembedded.dev * r6ff7435ae8 10openembedded.git/recipes/mpfr/mpfr_3.0.0.bb: Jul 29 07:19:49 mpfr_3.0.0.bb: set ARM_INSTRUCTION_SET to arm for armv4t Jul 29 07:19:49 Signed-off-by: Klaus Kurzmann Jul 29 07:34:46 mrnng Jul 29 07:34:56 hi hrw, all Jul 29 07:35:22 ls Jul 29 08:14:52 morgen Jul 29 08:15:06 is it legal to use bash variable mangling in a recipe? e.g. ${docdir#${prefix}} ? Jul 29 08:22:01 zub: no Jul 29 08:22:16 zub: recipes should run on dash and other posix compiant shells as well Jul 29 08:22:22 hm, ok Jul 29 08:22:36 then I need to do that in a block of python code Jul 29 08:22:57 you can use the $@{} to insert chunks of python Jul 29 08:23:14 good tip, thx Jul 29 08:23:29 I may have got the exact syntax for that wrong, please check Jul 29 08:23:42 I can never remeber if its $@{ or ${@ Jul 29 08:23:46 ok :) Jul 29 08:24:03 ./opie-reader/uqtreader_cvs.bb: soso="${so#${S}/opiedir/plugins/reader/}" Jul 29 08:24:04 :-P Jul 29 08:24:15 (I was grepping meanwhile) Jul 29 08:25:29 altogether 3 recipes use bash variable mangling Jul 29 08:26:00 BUILD_OS := "${@os.uname()[0].lower()}" Jul 29 08:26:29 bah, thats annoying Jul 29 08:41:38 03Koen Kooi  07org.openembedded.dev * r9043bdf8e8 10openembedded.git/contrib/angstrom/omap3-mkcard.sh: angstrom omap3-mkcard.sh: fix rootfs name Jul 29 08:41:49 03Koen Kooi  07org.openembedded.dev * r2d8b413001 10openembedded.git/recipes/xorg-driver/xf86-video-v4l2_git.bb: xf86-video-v4l2: add git version Jul 29 08:46:18 hehe thats gonna drive people crazy :-) Jul 29 08:46:27 why is my ubuntu called Angstrom Jul 29 08:49:22 ;D Jul 29 08:53:08 (10:22:04 AM) XorA: zub: recipes should run on dash and other posix compiant shells as well Jul 29 08:53:09 classes/sanity.bbclass: messages = messages + "Using dash as /bin/sh causes various subtle build problems, please use bash instead.\n" Jul 29 08:53:29 seems there is still some work to do then Jul 29 08:54:04 eFfeM_work: dont be obtuse, thats a whole different issue Jul 29 08:54:37 OE should run on dash, we cant say the same for software we build, for example gnome which doesnt Jul 29 08:55:14 although ubuntu may have kicked their ass over that issue now Jul 29 08:55:30 hmm, I always itnerpreted the message in sanity.bbclass as: do not build OE with /bin/sh set to dash Jul 29 08:55:57 OE itself should have zero bashisms, any that exist are bugs Jul 29 09:00:16 there was a day in history where one brave soul fixed all the bashisms to the point x11-image built perfectly on dash Jul 29 09:02:21 then new ver of gtk+ arrived and ruined it Jul 29 09:02:38 ;-) Jul 29 09:04:10 actually perhaps OE should build it's own shell and depend on that (bash-native or so), we also use our own make instead of trying to support all makes that are around, so why not do the same for sh ? Jul 29 09:04:47 becasue /bin/sh is built into the scripts Jul 29 09:05:03 running OE as root would go horribly wrong Jul 29 09:06:40 what about a chroot-ed env ? (and running OE as root is a bad thing to do anyway) Jul 29 09:06:55 * eFfeM_work is off to a meeting back later Jul 29 09:29:40 eFfeM_work: to call chroot() you should be root Jul 29 09:29:42 so.. Jul 29 09:31:10 good morning Jul 29 09:34:37 Jay7: fun with the new devices? Jul 29 09:35:00 ant_work: a bit :) Jul 29 09:35:21 and a lot of work Jul 29 09:35:29 and too hot again :( Jul 29 09:37:04 Jay7: does that mean box reached you? Jul 29 09:37:37 XorA: your box is still in NL.. Jul 29 09:37:52 but I have received Ben NanoNote and Efika MX Jul 29 09:39:28 XorA: btw, can I ask sometimes your advices related to ELC-E travel? Jul 29 09:40:03 Jay7: you can, pb_ lives actually in the town with the venue though Jul 29 09:40:24 ah, thanks :) Jul 29 09:40:34 unfortuneately someone booked UDS the same week so I might have to go there instead Jul 29 09:40:49 yeah.. I've seen this Jul 29 09:41:42 Jay7: check OEDEM/2009 wiki page ;D Jul 29 09:42:13 hrw: did they announce venue for UDS yet? Jul 29 09:42:18 nope Jul 29 09:42:40 which makes me angry as I have to pass though whole us visa crap to get to uds Jul 29 09:42:45 somewhere nearer civilisation Jul 29 09:42:51 I hope Jul 29 09:43:04 .pl needs visas for USA? Jul 29 09:43:39 heh... I would like to get 1€ each time when someone asks that question Jul 29 09:43:52 yes, we do and that suxx Jul 29 09:44:09 * Jay7 have agreed with hrw about suxx Jul 29 09:44:41 yeah that does suck, even the visa waiver sucks ass Jul 29 09:44:55 have to fill in my details in 3 different systems to get visa waiver Jul 29 09:46:33 * Jay7 need visa almost to all countries :\ Jul 29 09:47:09 Jay7: Polish visa too? Jul 29 09:47:28 so far Ive avoided going anywhere needing visa Jul 29 09:47:42 hrw: seems yes Jul 29 09:47:49 visa sux ;) Jul 29 09:48:38 XorA: I visited Turkey but it was just a matter of 15$ or 10€ at airport Jul 29 09:48:40 that is another reason to relocate into EU :) Jul 29 09:49:39 Jay7: and change nationality? Jul 29 09:50:01 hrw: yep Jul 29 09:50:25 I see no good future here Jul 29 09:50:46 but obviously you have to choose your EU country carefully :-) Jul 29 09:52:07 yep Jul 29 09:52:19 XorA: true :) Jul 29 10:26:32 03Koen Kooi  07org.openembedded.dev * r0391d04b77 10openembedded.git/recipes/xmms/xmms_1.2.10.bb: xmms: bump PR Jul 29 10:26:33 03Koen Kooi  07org.openembedded.dev * rd981b11b1c 10openembedded.git/recipes/scummvm/scummvm_1.1.1.bb: scummvm: bump PR Jul 29 10:26:33 03Koen Kooi  07org.openembedded.dev * r67d66df55b 10openembedded.git/recipes/sox/sox_13.0.0.bb: sox: bump PR Jul 29 10:26:35 03Koen Kooi  07org.openembedded.dev * r4e00989ccf 10openembedded.git/recipes/gnome/gnome-media_2.28.0.bb: gnome-media: bump PR Jul 29 10:26:40 03Koen Kooi  07org.openembedded.dev * rbf282e09e0 10openembedded.git/recipes/musicpd/mpd_0.15.bb: mpd: bump PR Jul 29 10:27:33 it seems gitserver is up and running Jul 29 11:13:17 anyone ever saw the issue that flash_lock would give 'Could not lock MTD device' (and it's unlock counterpart). I can erase the partition without problem and I can also write to it; but I would want to lock it afterwards to avoid accidental destruction Jul 29 11:31:48 are you sure your flash actually supports locking? Jul 29 11:32:39 pb_ i think so, u-boot supports it (but will recheck) guess it has to do with driver and kernel config Jul 29 11:32:52 that error message basically just means that ioctl(MTDLOCK) failed, which might be for any number of reasons but, most likely, means that either your flash isn't lockable or the driver doesn't know how to do it. Jul 29 11:33:23 i guess the latter (was peeking into the driver src) Jul 29 11:43:25 hm, interesting observation, protect in u-boot works (no error) after that erase of the protected area also gives no error, but read back of the area still gives the original data (which is what I expected), if I boot into linux after having the memory protected in u-boot, flash_erase also does not report an error, but the original data remains there Jul 29 11:46:13 i had expected some verification that the erase is successful Jul 29 11:47:13 that sounds like another driver bug, most likely Jul 29 11:49:11 peeking into the unlock thing first apparently the low level chip driver is not registering lock and unlock functions Jul 29 11:49:21 but haven't foudn that low leve driver for my chip yet Jul 29 11:52:38 you should be able to tell from the dmesg output at bootup what driver it is using. Jul 29 11:52:41 what sort of chip is it? Jul 29 11:53:34 Spansion nor flash, type S29GL01GP Jul 29 11:53:44 on powerpc Jul 29 11:55:35 I think those are cfi of some sort. Jul 29 11:55:53 yes Jul 29 11:56:27 not exactly sure any more where it lives, but I'll find it Jul 29 12:02:13 :sh Jul 29 12:02:27 oopsie Jul 29 12:02:49 pb_: drivers/mtd/maps/physmap_of.c (thanks to my good old friend grep) Jul 29 12:04:59 eFfeM_work: that's just the mapping, not the actual driver. you should be using something in chips/, probably one of the cfi_cmdset_xx.c files Jul 29 12:05:52 tnx, will check Jul 29 12:07:05 pb_ yeah I have there: built-in.o cfi_cmdset_0002.o cfi_probe.o cfi_util.o chipreg.o gen_probe.o Jul 29 12:07:05 will look into the cmdset one, many thanks for getting me on the rails with thsi Jul 29 12:10:42 03Klaus Kurzmann  07org.openembedded.dev * r77f69004cf 10openembedded.git/recipes/mpfr/ (mpfr.inc mpfr_3.0.0.bb): Jul 29 12:10:42 mpfr: move ARM_INSTRUCTION_SET to mpfr.inc Jul 29 12:10:42 Signed-off-by: Klaus Kurzmann Jul 29 12:14:35 hi Jul 29 12:15:05 i tried flashing the kernel linux2.6.32 into my openmoko Jul 29 12:15:10 pb_: this mpfr thing would need your help ;) Jul 29 12:15:13 but it returns error Jul 29 12:15:58 the error is vfs: cannot open root device " or unknown-block(0,0) Jul 29 12:16:23 Please append a correct "root=" boot option Jul 29 12:16:39 ranjith: give a root= argument on your boot cmd line Jul 29 12:16:54 (as u-boot bootargs) Jul 29 12:17:14 in the error message its also shown that Jul 29 12:17:30 the available partitions are mtdblock0 and mtdblock1 Jul 29 12:17:41 but my rootfs is in the 6th block Jul 29 12:17:49 ranjith: and perhaps when asking questions better hang around in the channel (saw a few questions from you from 3 am GMT or so, but then the channel is not too crowded) Jul 29 12:18:04 ok Jul 29 12:18:13 ranjith: use mtdparts to specify your partitions on the cmd line or compile them in Jul 29 12:18:46 i checked the mtdparts and it also shows that the rootfs is in the 6th partition Jul 29 12:19:08 yes but apparently the kernel does not know (or your root= is not good) Jul 29 12:19:10 you need Jul 29 12:19:14 and by default the root is /dev/mtdblock6 Jul 29 12:19:15 root=/ev/mtdblock6 Jul 29 12:19:37 is there a /dev/mtdblock6 ? Jul 29 12:19:39 its already there as root=/dev/mtdblock6 Jul 29 12:19:48 yes there are 6 partitions Jul 29 12:20:04 and does the kernel in the log show that 6 partitions in the table Jul 29 12:20:20 ah, if there are 6 partitions then it probably is /dev/mtdblock5, numbering starts at 0 Jul 29 12:20:24 i dint get that.. Jul 29 12:20:38 the numbering starts at 0 Jul 29 12:20:51 i have also tried giving it as /dev/mtdblock5 Jul 29 12:21:00 but that too din't work Jul 29 12:21:03 first partition is /dev/mtdblock0, 2nd is mtdblock1 etc so partition 6 is mtdblock5 Jul 29 12:21:34 guess then the root= is not good or not parsed Jul 29 12:21:37 i had also tried that .. ie giving root=/dev/mtdblock5 .. buat that too dint work Jul 29 12:21:56 so is there any way to solve it Jul 29 12:22:23 the NULL or (0,0) seem to indicate it is not parsed or not known otherwise it would say something like (92,5) or somethign like that Jul 29 12:22:28 I'm making up the 92,5 number Jul 29 12:22:53 ofc it can be solved, but I'm not too sure how Jul 29 12:23:33 is it safe to re-flash the u-boot or use Qi instead? Jul 29 12:23:56 * eFfeM_work does not know what Qi is Jul 29 12:24:06 and reflashing u-boot always has some risk in it Jul 29 12:24:31 ls Jul 29 12:24:34 :q Jul 29 12:24:34 can i do it safely Jul 29 12:24:39 tried root=/dev/mtdbloc6 rootfstype=jffs2 Jul 29 12:24:42 ? Jul 29 12:24:58 ant_work: yes i ahve tried it Jul 29 12:25:08 which board is it? Jul 29 12:25:19 neo freerunner Jul 29 12:26:01 eFfeM_work: ie is there any document or guideline through which will help for a safe flashing of the u-boot or Qi Jul 29 12:26:05 ranjith: please pastebin your boot log Jul 29 12:26:11 ~pastebin Jul 29 12:26:12 [~pastebin] A "pastebin" is a web-based service where you should paste anything over 3 lines so you don't flood the channel. Here are links to a few : http://www.pastebin.com , http://pastebin.ca , http://channels.debian.net/paste , http://paste.lisp.org , http://bin.cakephp.org/ , http://asterisk.pastey.net/ , or install pastebinit with yum or aptitude. Jul 29 12:26:22 do you have jffs2 compiled in kernel? Jul 29 12:26:24 ranjith: no idea, but agree with mckoan Jul 29 12:26:41 ant_work: if the fs type is wrong you'll get a different msg Jul 29 12:26:55 let's examine the log first Jul 29 12:27:51 eFfeM_work: "Qi" upstream http://code.google.com/p/0xlab-bootloader/ Jul 29 12:27:57 ranjith: you could quickly compile /recipes/kexecboot. This scans /proc/partitions and so we'll see where the problem is Jul 29 12:28:06 could run it as binary, no flashing Jul 29 12:28:39 we grab the 'mtdparts' passed by the kernel, though Jul 29 12:28:45 this must be correct Jul 29 12:28:58 xranby: thanks for the info, i'm not really into the handheld and openmoko stuff etc, I normally use u-boot Jul 29 12:32:40 eFfeM_work: sorry for the google link it seems dead, this project page on gittorious are more alive http://gitorious.org/0xlab-bootloader Jul 29 12:33:13 I shall get the boot log in 5 minutes Jul 29 12:33:33 i have flashed another kernel now to see the bootargs for that.. Jul 29 12:33:40 eFfeM_work: i have used Qi on a freerunner and its a quite magical experience.. the bootloaded automatically checks any available sd card for a kernel to boot and if it finds one then it loads it and starts up the system.. Jul 29 12:33:42 i shall be back soon Jul 29 12:34:13 eFfeM_work: if it dont find any sd card then Qi starts the system from the internal flash Jul 29 12:34:14 xranby: not sure if it fits my hw, I am on an embedded ppc board Jul 29 12:39:28 eFfeM_work: if you want to be able to simply switch kernel on your ppc board without having to fiddle with the bootloader then Qi are a quite good solution. i dont think Qi are limited to only boot from sd cards/fralsh.. Jul 29 12:49:17 hi all i got it working Jul 29 12:49:30 ant_work: its working now Jul 29 12:49:41 great Jul 29 12:49:46 i made some changes in the uboot command line from this page http://wiki.openmoko.org/wiki/Kernel Jul 29 12:50:08 what was wrong that i was confused with the instructions provided in the wikipage. Jul 29 12:50:31 i have made some changes in the page as per the instructions i used to make the kernel working Jul 29 12:51:34 ranjith: great you got it working and thanks for updating the doc Jul 29 13:04:10 does oe keep a cache of my .bb recipe? im changing my recipe but its still trying to apply a path Jul 29 13:04:12 patch* Jul 29 13:12:36 marcompile: no cache for that, but did you bump PR or do a bitbake -cclean of the recipe? Jul 29 13:17:53 I now deleted the .patch file and it worked Jul 29 13:19:41 thanks Jul 29 13:20:20 deleting the patch file should not be needed if the file is removed from teh recipe Jul 29 13:27:03 eFfeM_work, I am making an asterisk 1.6 non beta recipe Jul 29 13:28:16 what's the difference between python () { and python __anonymous () { ? Jul 29 13:38:53 I still don't get the versioning of packages in oe Jul 29 13:39:26 I made a julius_4.1.5.bb but when I "bitbake julius" it builds the 4.1.2 (the old one) Jul 29 13:40:42 I still can build with bitbake -b ./julius_4.1.5.bb Jul 29 13:49:06 marcompile: http://old.nabble.com/How-does-OE-choose-%22preferred%22-version--td27992038.html maybe this helps Jul 29 13:50:04 zub thank you very much Jul 29 13:50:30 yw Jul 29 13:50:51 can I ask for (p)review of http://linux.fjfi.cvut.cz/~zub/OE/cmake-2.8.2/cmake_2.8.2.bb ? Jul 29 13:51:06 I yet have to test that on the target, but seems to compile ok Jul 29 13:51:26 I mainly want to know if the variable stripping thing is done ok, or if there are better ways Jul 29 14:08:03 03Koen Kooi  07org.openembedded.dev * r962d52ae83 10openembedded.git/recipes/x-load/signgp-native.bb: signgp: convert to BBCLASSEXTEND Jul 29 14:11:35 ant_work: which mpfr thing is that? Jul 29 14:11:38 kergoth: good morning Jul 29 14:12:05 hey Jul 29 14:17:33 hello Jul 29 14:17:38 pb_: was mpfr: move ARM_INSTRUCTION_SET to mpfr.inc Jul 29 14:18:06 I didn't follow but must be the usual thumb breakage Jul 29 14:18:39 for armv4t Jul 29 14:19:22 http://pastebin.com/dRX5PHv2 im having issues resolving that error; i can execute ipkg-make-index manually with those arguements and it works fine... Jul 29 14:20:53 im guessing an ipk doesnt have a control.tar.gz; though even adding prints to the py i wasnt able to identify which ipk is causing the issue Jul 29 14:32:31 ant_work: in Ubuntu I think mpfr 2.4.2 is built as thumb Jul 29 14:33:16 hopefully is not a side-efffect of CROSS works Jul 29 14:38:19 ant_work: if you have a pointer to the original issue then I can certainly have a look at it Jul 29 14:40:15 [20100728 20:40:59] selected processor does not support `umull r6,r3,r2,r1' Jul 29 14:41:12 building shr-unstable image it seems Jul 29 14:41:40 log http://pastebin.com/hF2PaqKu Jul 29 14:47:06 hrw: mpfr-3.0.0 Jul 29 14:48:45 ant_work: ah. ubuntu is at 2.4.2 now Jul 29 14:49:05 will stumble in it ;) Jul 29 14:50:42 pb_: fwiw I yet have toexperience the benefits of thumbs on old limited devices like Zaurus Jul 29 14:51:11 but the wind blows that direction... Jul 29 14:53:16 hrw: strange the hack was moved in the inc, thus for older versions too Jul 29 15:15:03 pb_: I added that arm to mpfr, because thumb does not build for armv4t Jul 29 15:16:38 ant_work: I moved it then to the inc, because older version of mpfr did not build either with thumb mode Jul 29 15:17:29 which images do require mpfr? Jul 29 15:17:34 rebuilding zlib-native and python-native solved the problem; tarfile.ReadError: file could not be opened successfully (http://pastebin.com/dRX5PHv2) Jul 29 15:18:27 I think it is not built for console-iamge nor for x11-image Jul 29 15:18:48 shr Jul 29 15:18:50 mrmoku: can you not fix mpfr to build in thumb mode? that's usually fairly trivial to arrange. Jul 29 15:18:52 builds it Jul 29 15:19:22 pb_: I would need more instructions on how to do that :) Jul 29 15:19:33 * ant_work did his job ;) Jul 29 15:19:40 okay I have to leave Jul 29 15:19:43 ant? Jul 29 15:19:47 ant_work, woglinde: mpfr is a dep of gcc, no? Jul 29 15:19:55 gcc-4.5 Jul 29 15:19:57 hm yes Jul 29 15:20:08 okay Jul 29 15:20:08 mrmoku: those umull instructions are probably coming from some inline asm statement. usually there is a C fallback anyway, so disabling the eleet asm for thumb is often an adequate solution. Jul 29 15:20:12 till lter Jul 29 15:20:34 that's often as easy as finding the "#ifdef __arm__" that guards the macro in question, and changing it to "#if defined(__arm__) && !defined(__thumb__)" Jul 29 15:20:49 pb_: ok, will take a look at the code Jul 29 15:21:17 alternatively, if it's just one source file that has the problem, you can arrange for (just) that one to be built in arm mode Jul 29 15:21:25 see glib, for example, which does that for some of the gatomic bits Jul 29 15:21:43 ok Jul 29 15:48:24 hm.. linux.conf.au Jul 29 15:55:32 pb_: heh Jul 29 15:55:33 #if 1 || defined (__arm_m__) /* `M' series has widening multiply support */ Jul 29 15:55:33 #define umul_ppmm(xh, xl, a, b) \ __asm__ ("umull %0,%1,%2,%3" : "=&r" (xl), "=&r" (xh) : "r" (a), "r" (b)) Jul 29 15:58:18 ok, is there a way in git of finding files no matter what branch they are in? Jul 29 15:59:02 or from cgit finding out what branches a file exists one Jul 29 15:59:03 on Jul 29 15:59:28 I think you'd have to write a short shell script for it Jul 29 15:59:34 ssh & grep :) Jul 29 15:59:39 of find Jul 29 15:59:48 that's not going to help Jul 29 15:59:55 unless you have a checkout of every branch Jul 29 16:00:03 i expect what you really want is git for-each-ref Jul 29 16:00:09 I mean ssh to git server :) Jul 29 16:00:16 oh wait, I see it, its in the URL Jul 29 16:00:16 i know Jul 29 16:00:21 and grep will still be useless there Jul 29 16:01:21 http://stackoverflow.com/questions/372506/how-can-i-search-git-branches-for-a-file-or-directory Jul 29 16:01:56 mrmoku: awesome dude, exactly what I needed :-) Jul 29 16:02:34 yw :) Jul 29 16:02:45 cute, I forgot log can show commits from all branches. I think i like for-each-ref + ls-tree better, if you don't need the history, though, as per the comments on the ls-tree answer Jul 29 16:02:57 I just googled for 'git find file in all branches' :P Jul 29 16:05:24 mrmoku: heh, right, that is a bit sad. try sticking an "!defined(__thumb__) && ..." at the start of that condition and see if the situation improves. Jul 29 16:06:47 mrmoku: fwiw, that asm is fairly useless nowadays in any case, since gcc is at least tolerably good at figuring out for itself when umull would be a win. you might find that even just changing it to "if 0" is fine :-} Jul 29 16:08:04 has anyone seen relocatable.bbclass throwing exception on Popen() while executing chrpath with "File not found"? the file is indeed there and running chrpath manually works... any ideas? Jul 29 16:08:29 (and gcc also knows that, in armv6 and higher, you can have overlapping registers in a umull which will also give you better code sometimes.) Jul 29 16:09:39 pb_: I removed the one... and get a whole bunch more asm errors in the same file :/ Jul 29 16:11:39 pb_: probably the 1 is there, because the alternative code path is broken anyway Jul 29 16:12:51 could be. what are the other errors? Jul 29 16:13:40 moment... reproducing :) Jul 29 16:15:00 pb: http://pastie.org/1065666 Jul 29 16:17:02 mrmoku: oh, right. you need to wrap the rest of that mess in #if !defined(__thumb__) as well. Jul 29 16:17:29 probably best to find the outer "#if defined(__arm__)" and add the thumb guard there. Jul 29 16:17:46 basically, all the inline asm bits will be inappropriate in thumb mode Jul 29 16:18:09 ok, trying Jul 29 16:18:36 03Roman I Khimov  07org.openembedded.dev * reb6293b2f6 10openembedded.git/recipes/squid/ (5 files in 2 dirs): Jul 29 16:18:36 squid: remove -native recipes Jul 29 16:18:36 They were only used for cf_gen compilation which is unnecessary with a little patch Jul 29 16:18:36 for squid build system. Jul 29 16:18:36 Signed-off-by: Roman I Khimov Jul 29 16:20:21 pb_: yeah, that works Jul 29 16:21:43 03Roman I Khimov  07org.openembedded.dev * rfdde749316 10openembedded.git/recipes/squid/ (squid.inc squid_2.6.STABLE14.bb squid_2.7.STABLE8.bb): Jul 29 16:21:44 good-o Jul 29 16:21:44 squid: remove SQUID_MAJOR, determine SRC_URI path from PV Jul 29 16:21:44 Signed-off-by: Roman I Khimov Jul 29 16:25:31 kudos to both Jul 29 16:26:42 ant_work: nah, kudos to pb_... I just did what he told me to do :) Jul 29 16:28:06 hi, have built console image for tx25 and flashed but its hanging in boot immediatly after * Starting Avahi mDNS/DNS-SD Daemon: avahi-daemon Jul 29 16:28:06 [ ok ] Jul 29 16:28:06 , and then saying: INIT: Id "S" respawning too fast: disabled for 5 minutes Jul 29 16:31:51 maybe there is a loop in your rcS somewhere? Jul 29 16:31:57 pb_: what about older versions of mpfr? Jul 29 16:34:40 mrmoku: I imagine the same patch will apply to them verbatim. mpfr-longlong.h doesn't change often. Jul 29 16:35:20 any that don't patch, I would be happy enough for you to either delete or ignore. personally I don't care which but I guess others will have an opinion :-} Jul 29 16:36:29 pb_: ok, thanks Jul 29 16:37:16 will try some older ones after dinner then Jul 29 16:43:15 pb_: heh, just discovered there already is a patch for it... long-long-thumb.patch. Just not used and is doing exactly the same :P Jul 29 16:43:57 anyway... dinner now Jul 29 16:45:39 anybody got any good laptop recommendations? I think I need a new one. Jul 29 16:46:17 my old x61s was looking a bit worn-out, and I think my daughter might have finally killed it this morning by pouring an entire mug of coffee into the keyboard. Jul 29 16:47:21 pb_: I got mself an X200s but I'm not so impressed by it. Jul 29 16:47:40 ah, funnily enough I was just looking at the x201 spec on the web Jul 29 16:47:42 pb_: Given my T20 and T40p before the hardware feels more like Lenovo then like IBM ;) Jul 29 16:47:53 doh Jul 29 16:47:58 pb_: shoragan got one recently and he seems to be happy Jul 29 16:48:43 yeah, I've heard several people say that they noticed the lenovo ones sucking compared to how ibm used to make them. Jul 29 16:48:53 pb_: on mine and the one froma friend we already had to change the display, the display cable, thinkpad stickers has been falling off and now BT is dead on mine Jul 29 16:49:09 but, my x61s was post-lenovo and I never had any real trouble with it. Jul 29 16:49:09 could be "monday" machines but I felt better with my machines before Jul 29 16:49:41 pb_: well, the x6x series was still designed by IBM and perhaps also QA'ed :) Jul 29 16:50:07 lenovo was doing the production for years already, but IBM had the hand on design, AQ etc Jul 29 16:50:10 yeah, true. :-} Jul 29 16:50:12 That makes a difference Jul 29 16:50:22 anybody have any suggestions for non-thinkpads that are worth looking at? Jul 29 16:50:45 But I don't have a real alternative though. Thinkpads still are one of the best ootb experience with Linux Jul 29 16:51:03 I'm happy with a Dell 4300. nice metal construction, good keyboard layout, all the special buttons work in Linux without hacking Jul 29 16:51:27 stefan_schmidt: yeah, exactly. it's a shame that the newer thinkpads seem not as good as the old ones, but as far as I can tell they are still the best option out there. Jul 29 16:52:04 hollisb: thanks, I'll take a look Jul 29 16:52:18 pb_: As long as you stay with the T or X series you should be quite ok imho Jul 29 16:52:29 pb_: but avoid the consumer series :) Jul 29 16:52:32 Yeah, the Dells work really well out of the box - I've got a 4300 and an M1330 which have both been very straightforward. Jul 29 16:52:59 03Denys Dmytriyenko  07org.openembedded.dev * rb1a73fa466 10openembedded.git/recipes/liboil/ (5 files): Jul 29 16:52:59 liboil: change the license to liboil Jul 29 16:52:59 Legally, it is more correct to call the mix of licenses (like used by liboil) Jul 29 16:52:59 with the application-specific name, rather than a generic "various". Jul 29 16:52:59 Signed-off-by: Denys Dmytriyenko Jul 29 16:54:00 stefan_schmidt: yeah, I have always had X-series myself. we have a bunch of T-series lying around the office and they always feel like they have been built from depleted uranium or something. Jul 29 16:54:24 pb_: heh Jul 29 16:54:59 pb_: I'm still pondering with myself about a Macbookpro after the x200. I like the hardware. And it would be linux only anyway :) Jul 29 16:55:14 Still not sure if I like to feed my money into apple Jul 29 16:55:30 but my machine should hold another year hopefully Jul 29 16:57:19 03Denys Dmytriyenko  07org.openembedded.dev * rd8a6e4a9f6 10openembedded.git/recipes/meta/external-toolchain-csl.bb: Jul 29 16:57:19 external-toolchain-csl: fix the license for -dev packages Jul 29 16:57:19 Signed-off-by: Denys Dmytriyenko Jul 29 16:57:59 stefan_schmidt: if you want lnx on it they why Macbook pro Jul 29 16:58:12 its just expensive piece of hardware Jul 29 16:58:40 I have used on but felt its not worth the money Jul 29 16:59:01 yeah, I have been tempted by macbooks on and off. my last apple machine was an ibook back in 2003 or thereabouts, haven't really used one seriously since then. Jul 29 17:00:10 pb_: I still use it on and off but not for serious stuff Jul 29 17:00:20 multimedia pictures Jul 29 17:00:23 thats it Jul 29 17:02:15 khem: well, for my x200s I also paid 1600€ Jul 29 17:02:27 khem: MBP would be about the same for the 13" I wanted Jul 29 17:02:42 khem: But I would really miss my touchstick :( Jul 29 17:03:32 stefan_schmidt: In my experience I could buy a lot better hardware with same amount of money I would spend on a MBP Jul 29 17:04:09 i want to build binaries for arm/android what is the bset way to do this? Jul 29 17:04:11 if you dont mind the coolness factor then its expensive Jul 29 17:04:12 khem: Maybe I just don't have a good overview on the actual notebook market, but most of the big players are not what I want Jul 29 17:04:19 is there a distro for android in oe? Jul 29 17:05:57 cmkinne: No, but someone posted patches a while back Jul 29 17:06:09 do you nkow where i acn find those? Jul 29 17:06:14 Didn't look and see if it got all of the hard cases tho Jul 29 17:06:23 i want to get ssh and busybox on my android emulator Jul 29 17:06:24 cmkinne, other than googling it, no Jul 29 17:06:32 k Jul 29 17:06:34 khem: I would be lying if it would not be a small percentage of coolness factor as well. ;) Jul 29 17:06:44 emulator? no need for ssh, adb should be good enough for that :) Jul 29 17:06:48 busybox is another question Jul 29 17:06:49 khem: I have to evaluate the market shortly before buying a new anyway Jul 29 17:07:08 stefan_schmidt: yes do that Jul 29 17:07:11 I would to get a dual core cortex A9 notebook :) Jul 29 17:07:21 heh not a bad idea Jul 29 17:07:29 not sure if this is realistic in ~1 year Jul 29 17:22:43 pb_: me too switched to a Dell after being disappointed by the last thinkpad Jul 29 17:23:33 stefan_schmidt: that's the good thing with dell notebooks... they too have a trackpoint :) Jul 29 17:24:57 mrmoku|dinner: yeah, I know. But I'm not really happy with what they offer. Need to check before buying Jul 29 17:25:06 But its pb_ who is looking not me :) Jul 29 17:25:29 03Roman I Khimov  07org.openembedded.dev * r0b89ca3d10 10openembedded.git/recipes/squid/ (squid.inc squid/squid.init): Jul 29 17:25:29 squid: fix missing /etc/squid.conf message from init Jul 29 17:25:29 Like this: Jul 29 17:25:29 /etc/init.d/squid: line 18: /etc/squid.conf: No such file or directory Jul 29 17:25:29 We actually have squid config in /etc/squid/squid.conf Jul 29 17:25:29 Signed-off-by: Roman I Khimov Jul 29 17:25:30 03Roman I Khimov  07org.openembedded.dev * rdc009cff04 10openembedded.git/recipes/squid/ (7 files in 2 dirs): Jul 29 17:25:31 squid: add version 3.1.4 Jul 29 17:25:31 Signed-off-by: Roman I Khimov Jul 29 17:25:31 03Roman I Khimov  07org.openembedded.dev * r7e76202b53 10openembedded.git/recipes/squid/ (squid.inc squid/squid.init): Jul 29 17:25:32 squid: fix obsolete "-D" cli option warning for squid version 3 Jul 29 17:25:32 Signed-off-by: Roman I Khimov Jul 29 17:25:50 stefan_schmidt: yeah, but you commented about missing it :) Jul 29 17:26:13 mrmoku: heh, correct Jul 29 17:29:50 pb_: just to be sure, because the other patch was added to all archs... it is needed for armv4t only, right? Jul 29 17:32:46 03Roman I Khimov  07org.openembedded.dev * r6425b4db03 10openembedded.git/recipes/ (barnyard/barnyard_0.2.0.bb snort/snort_2.8.5.3.bb): (log message trimmed) Jul 29 17:32:46 barnyard/snort: switch to Gentoo mirrors for SRC_URI Jul 29 17:32:46 snort.org changed source URLs to Jul 29 17:32:46 http://www.snort.org/downloads/116 Jul 29 17:32:46 alike that redirect somewhere to Amazon S3 with parameters added to the link that Jul 29 17:32:46 makes wget save files like "snort-version.tar.gz?lots-of-gibberish" which breaks Jul 29 17:32:47 build. Workaround that. Jul 29 17:53:57 03Klaus Kurzmann  07org.openembedded.dev * r53126bb60b 10openembedded.git/recipes/mpfr/ (mpfr.inc mpfr_3.0.0.bb): Jul 29 17:53:57 mpfr-3.0.0.bb: use existing long-long-thumb.patch instead of arm mode for armv4t Jul 29 17:53:57 Signed-off-by: Klaus Kurzmann Jul 29 17:55:31 03Koen Kooi  07org.openembedded.dev * ra3b093595a 10openembedded.git/recipes/xmms/xmms-tremor_1.0.bb: xmms-tremor: bump PR Jul 29 17:58:11 what is the best way to add xserver/gnome to a distro like angstrom? I thought it would come with the x11 build but doesn't seem to be there Jul 29 18:03:22 nm feel like an idiot booted wrong kernel... Jul 29 18:05:08 03Koen Kooi  07org.openembedded.dev * rad2db9ff3d 10openembedded.git/recipes/musicpd/mpd_0.15.12.bb: mpd: add 0.15.12 Jul 29 18:08:39 03Koen Kooi  07org.openembedded.dev * r005408184b 10openembedded.git/recipes/linux/multi-kernel.inc: multi-kernel: fix race condition Graeme reported by depending on mkimage Jul 29 18:10:58 03Denys Dmytriyenko  07org.openembedded.dev * r0f7a7ef4c3 10openembedded.git/recipes/x-load/signgp.bb: Jul 29 18:10:59 signgp: set license to BSD Jul 29 18:10:59 Signed-off-by: Denys Dmytriyenko Jul 29 18:13:59 Hello Jul 29 18:14:05 I Have errors here Jul 29 18:14:12 can anyone help ? Jul 29 18:16:31 Angelox_123, no chance if you don't paste all required info to some paste thing like paste.debian.net, no Jul 29 18:16:52 ok Jul 29 18:17:10 wait..its of QT Compile...this channel can help ???? Jul 29 18:19:26 {standard input}: Assembler messages: Jul 29 18:19:27 {standard input}:165: Error: immediate expression expected -- `wstrw wcgr0,[r5,r0,asl#2]' Jul 29 18:19:27 gmake: *** [.obj/release-shared-emb-arm/qscriptengine.o] Error 1 Jul 29 18:19:45 ON COMMAND Jul 29 18:19:46 arm-linux-g++ -c -pipe -DQWS -fno-exceptions -fno-rtti -Wall -W -O2 -fPIC -DQT_SHARED -DQT_NO_DEBUG -DQT_NO_CUPS -DQT_NO_NIS -D_LARGEFILE_SOURCE -D_LARGE_FILES -D_FILE_OFFSET_BITS=64 -DQT_NO_IMAGEIO_MNG -DQT_NO_IMAGEIO_JPEG -DQT_NO_STYLE_MAC -DQT_NO_STYLE_AQUA -DQT_NO_STYLE_INTERLACE -DQT_NO_STYLE_WINDOWSXP -DQT_NO_STYLE_COMPACT -DQT_NO_STYLE_POCKETPC -DQT_NO_QWS_VFB -DQT_NO_QWS_VNC -DQT_NO_QWS_VGA16 -DQT_NO_QWS_TRANSFORMED -DQT_NO_QWS Jul 29 18:19:50 _SNAP -DQT_NO_QWS_MACH64 -DQT_NO_QWS_VOODOO3 -DQT_NO_QWS_MATROX -DQT_NO_QWS_SHADOWFB -DQT_NO_QWS_REPEATER -DQT_NO_QWS_KBD_SL5000 -DQT_NO_QWS_KBD_USB -DQT_NO_QWS_KBD_YOPY -DQT_NO_QWS_KBD_VR41 -DQT_NO_QWS_MOUSE_BUS -DQT_NO_QWS_MOUSE_LINUXTP -DQT_NO_QWS_MOUSE_VR41 -DQT_NO_QWS_MOUSE_YOPY -I/root/qt-estress/qt-embedded-free-3.3.5/mkspecs/qws/linux-arm-g++ -I. -I3rdparty/freetype/src -I3rdparty/freetype/include -I3rdparty/freetype/builds/uni Jul 29 18:19:55 x -I3rdparty/freetype2/include -I3rdparty/libpng -I3rdparty/zlib -I../include -I.moc/release-shared-emb-arm/ -o .obj/release-shared-emb-arm/qscriptengine.o kernel/qscriptengine.cpp Jul 29 18:30:42 how do I create the Packages, Packages.gz and the files necessary to make mu /stuff/tmp/deploy/glibc/ipk a repository exported through http ? Jul 29 18:31:27 marcompile, dunno, but i'd search for feed Jul 29 18:33:18 Angelox_123: forget iwmmxt; it's broken both in kernel and in recent gccs Jul 29 18:34:25 BlindMan, kk, thanks Jul 29 18:34:45 BlindMan, mt, sorry Jul 29 18:43:02 marcompile: you mean blindvt ;) Jul 29 18:44:47 yeah Jul 29 18:49:34 BlindMan, hi (and gimme five ;) Jul 29 18:54:23 hehe Jul 29 18:54:24 *clap* Jul 29 18:54:35 hi there *wave+ Jul 29 19:00:20 any workaround to the error: "cannot find -ltinfo" when compiling alsamixer on F13 ? Jul 29 19:26:44 what does a kernel command line root arg with hex value mean? i.e. root=1f01. I have a karo tx25 and this is the default root= arg for booting there supplied kernel. Jul 29 19:38:46 03Chase Maupin <[chasemaupin03@gmail.com]> 07org.openembedded.dev * rd90020bab7 10openembedded.git/recipes/x-load/ (files/signGP.c signgp.bb): Jul 29 19:38:46 signGP: add BSD header and copyright Jul 29 19:38:46 * Add the BSD header to the signGP.c file with copyright year 2010. Jul 29 19:38:46 * Bump PR Jul 29 19:38:46 Signed-off-by: Chase Maupin Jul 29 19:38:46 Signed-off-by: Denys Dmytriyenko Jul 29 19:40:57 dan2003, it just specifies the filesystem, so 001 is e.g. /dev/ram0 if memory serves me right Jul 29 19:41:21 yeah sorry, found out it specifies major, minor dev node Jul 29 19:43:03 my rootfs is missing a dev node ttymxc0 (207,16) which i think is what os preventing it from booting - just trying to make a new jffs2 with it added in to see if it is the cause. Jul 29 19:55:25 03Roman Khimov  07org.openembedded.dev * r3fe37352fc 10openembedded.git/recipes/base-files/ (7 files in 7 dirs): Jul 29 19:55:25 base-files: profile: don't set TZ if /etc/TZ is present Jul 29 19:55:25 uClibc-based systems can have system-wide TZ in /etc/TZ, thus we shouldn't Jul 29 19:55:25 set TZ variable if it is present. Jul 29 19:55:25 Signed-off-by: Roman I Khimov Jul 29 19:55:25 Acked-by: Khem Raj Jul 29 20:17:38 does avahi-daemon worg with normal dhcp servers or not? on my zaurus (runign angstrom) i had to make a hack script that runs udhcpc -i wlan0 to get it on the lan automatically at boot. Now this tx25 running openembedded says its started avahi-daemon but it never gets an ip from my dns server Jul 29 20:20:51 dan2003 i just starting trying to identify where the dhcp is called during the networking init script; currently it tries to launch dhclient but with the wrong arguements Jul 29 20:23:11 warflyr, The serial console on my board never comes up. the last text i see is the bit about starting avahi-daemon followd by a msg form init every 5 mins to do with the serial console not working - so i have no way to get in as the lan is not working either :( Jul 29 20:25:31 im wondering if this is related.. Jul 29 20:25:32 Starting udev/etc/rcS.d/S03udev: line 50: can't create /tmp/uname: Read-only file system Jul 29 20:25:32 touch: /tmp/uname: Read-only file system Jul 29 20:25:32 Remounting root file system.. Jul 29 20:25:50 dan2003 setting a static ip in /etc/network/interfaces prior to booting may help you out... did you add your serial console to the init tab? Jul 29 20:26:28 echo "S:2345:respawn:/sbin/getty 115200 ttyS2" >> /etc/inittab Jul 29 20:26:30 something like that Jul 29 20:26:35 yup - bu the device node was missing - i added it but still sint working (has a funny name ttymxc0) Jul 29 20:27:52 only way i have to make the changes is to edit the rootfs on the pc, make a new .jffs2, and flash it , which is somewhat tedious Jul 29 20:27:52 what machine? Jul 29 20:28:19 karo tx25, (which is basically just a freescale mx257) Jul 29 20:30:08 hmm Jul 29 20:30:55 well, set static IP addresses in interfaces then try to boot; make sure you have ssh daemon installed; if you get lucky, you should be able to ssh in Jul 29 20:33:50 yeah - will do that - ui can see dropbear starting just prior to avahi so thats promising Jul 29 20:37:40 hasnt made any difference :( Jul 29 20:38:10 cant ping it? Jul 29 20:39:09 nope Jul 29 20:39:37 might remove avahi form the startup Jul 29 20:39:41 any activity lights on the board? Jul 29 20:40:03 yeah i was gonna suggest that, avahi is pretty heavy for embedded systems Jul 29 20:40:04 afraid not - but it works fine with karo's own image Jul 29 20:40:42 maybe try karos kernel/uboot with your rootfs? Jul 29 20:41:32 proab have problems with modules? Jul 29 20:41:41 guess could copy them into my rootfs Jul 29 20:43:12 yeah, use the corresponding modules Jul 29 20:59:51 03Klaus Kurzmann  07org.openembedded.dev * r84c25f7573 10openembedded.git/recipes/freesmartphone/fso-specs_git.bb: Jul 29 20:59:52 fso-specs_git.bb: bump SRCREV Jul 29 20:59:52 Signed-off-by: Klaus Kurzmann Jul 29 20:59:56 03Klaus Kurzmann  07org.openembedded.dev * rdcc69927a2 10openembedded.git/recipes/freesmartphone/cornucopia.inc: Jul 29 20:59:57 cornucopia.inc: bump FSO_CORNUCOPIA_SRCREV Jul 29 20:59:57 Signed-off-by: Klaus Kurzmann Jul 29 20:59:57 03Klaus Kurzmann  07org.openembedded.dev * r1fb667cb41 10openembedded.git/recipes/freesmartphone/libfso-glib_git.bb: Jul 29 20:59:57 libfso-glib_git.bb: bump SRCREV Jul 29 20:59:57 Signed-off-by: Klaus Kurzmann Jul 29 21:16:24 hi I've that: ./libpython2.6.so: undefined reference to `_PyParser_Grammar' Jul 29 21:16:38 the fix is so old and was pushed Jul 29 21:16:43 but I've still that Jul 29 21:16:50 machine htcdream Jul 29 21:16:52 distro shr Jul 29 21:16:58 bitbake shr-image Jul 29 21:43:22 03Andrea Adami  07org.openembedded.dev * r0ff60623c2 10openembedded.git/recipes/kexecboot/kexecboot.inc: Jul 29 21:43:22 kexecboot: remove dependency on losetup Jul 29 21:43:22 * Support for loopmount still to be defined. Jul 29 21:43:22 * While we could easily mount a loopimage and kexec the included Jul 29 21:43:22 * kernel and initrd, it would be duty of the init of the loopimage Jul 29 21:43:23 * to know about rootfs and rest of filesystem. Jul 29 21:43:24 * Problem is we don't know that init in advance. Jul 29 21:43:24 03Andrea Adami  07org.openembedded.dev * rd45a784c09 10openembedded.git/recipes/kexecboot/kexecboot_git.bb: Jul 29 21:43:25 kexecboot_git: bump to f0f9069a40f7dcf21da9d0b59f36e768a073d0c9 Jul 29 21:43:25 * new fbmenu layout Jul 29 22:23:57 etrunko: did you progress with testing the cmake? Jul 29 22:26:06 etrunko: I just tested on an om-gta02. Tested basic c, c++ compilation. Tested also FindPkgConfig (=cmake can find the shipped modules) Jul 29 22:26:42 etrunko: but tested only search for non-existend pkgconfig lingrary; but as the include was found + pkgconfig called that would probably also work Jul 29 22:26:54 ~ overall result seems to be: cmake works Jul 29 22:27:45 but as the include was found ... I mean: as cmake found FindPkgConfig module Jul 29 22:28:29 also ccmake works :) Jul 29 22:39:45 if im using bbclassextend=native; how can i have a different configure option on the native package? (or only patch the target build) Jul 29 22:39:46 zub: just did very basic test too Jul 29 22:39:58 but indeed it works Jul 29 22:56:27 warflyr: Yes, there's the virtclass-native override Jul 29 23:01:07 Tartarus, thanks Jul 29 23:01:51 03Andrea Adami  07org.openembedded.dev * r78cc524b1b 10openembedded.git/conf/machine/ (akita.conf include/zaurus-2.6.inc spitz.conf): Jul 29 23:01:51 zaurus-2.6.inc: Set PREFERRED_PROVIDER for udev-compat to udev-compat141 here Jul 29 23:01:51 * extend support to c7x0 and poodle Jul 29 23:02:02 03Andrea Adami  07org.openembedded.dev * r6293399a67 10openembedded.git/recipes/udev/ (udev.inc udev_151.bb udev_154.bb): Jul 29 23:02:03 udev: add support for udev-compat for c7x0 and poodle Jul 29 23:02:03 * following spitz + akita Jul 29 23:02:03 * fixed RDEPENDS of .154 Jul 29 23:04:41 hmm, bitbake still having issues finding the package... just sits at NOTE: Resolving any missing task queue dependencies Jul 29 23:09:14 is there something wrong with using the name 'unixodbc-native'... bitbake had this same issue with using both unixodbc-native_pv.bb (inherit native) and the bbclassextend=native method o.0; id at least expect it to find the package and throw an error... Jul 29 23:13:11 Tartarus: have you commited your patches for packaged-staging? Jul 29 23:15:51 seems so Jul 29 23:27:08 ant_: yes, aside from perl-native and guile-native (still need to write that up, bah), it's good enough for pstaging to be 1.0 I think Jul 29 23:27:15 it's there, it works, it could be improved Jul 30 01:46:24 hello, I'm a Chinese, and i'm now translating OE manual into Chinese. Jul 30 01:47:42 when I finished this , can OE add a link to official Document webpage? Jul 30 01:50:24 I am sure we can work something out Jul 30 01:50:39 what format is the translated document in ? Jul 30 01:53:28 now, I'm only working with gedit Jul 30 01:54:21 now I can not use docbook tool, beacause I could not find a better docbook editor Jul 30 01:54:38 what was the oe manual written with? Jul 30 01:57:06 and now i put the translate on my blog, so it is a blog page, but I think I need to use the docbook format. Jul 30 01:57:31 is there a nice docbook editor? Jul 30 01:57:52 or xml editor support docbook? Jul 30 02:06:23 Crofton: ? Jul 30 02:10:32 sorry afk Jul 30 02:10:47 I'm not sure Jul 30 02:13:45 Crofton: sorry for the stupid question Jul 30 02:15:19 Crofton: I found an editor , serna-free-4.2 , it's opensource, but it is based on QT, too big for me. Jul 30 02:17:52 you could ask on the oe devel list Jul 30 02:18:10 although, I am not sure how many people know about writing in Chinese Jul 30 02:18:14 thank you Jul 30 02:18:16 I am going to bed now Jul 30 02:18:22 thanks for looking at this Jul 30 02:18:28 good night crofton Jul 30 02:18:36 you're welcome **** ENDING LOGGING AT Fri Jul 30 02:59:57 2010