**** BEGIN LOGGING AT Mon Jun 22 02:59:57 2009 Jun 22 03:06:21 03Angus Ainslie  07fso/milestone5.5 * rdee0f6ddc7 10openembedded.git/conf/distro/include/sane-srcrevs-fso.inc: sane-srcrevs-fso : bump frameworkd Jun 22 03:06:21 03Angus Ainslie  07fso/milestone5.5 * ra37c69a8c9 10openembedded.git/: Merge branch 'fso/milestone5.5' of git@git.openembedded.org:openembedded into fso/milestone5.5 Jun 22 03:33:12 hi, is there someone there...how do I make a tar rootfs if I already built the rootfs with another filesystem? Jun 22 03:35:14 because obviously the rootfs dir exported via NFS doesn't work Jun 22 03:40:02 hi guys, anyone hit a problem building anything console-base-image where it fails compiling m4 or other packages with a weird "stddef.h" error two or more data types in declaration specifiers? Jun 22 03:40:37 is this some weird native gcc problem with 4.4? Jun 22 03:46:44 gnutoo_, add tar toI IMAGE_FSTYPES Jun 22 03:46:54 Tartarus, already done Jun 22 03:47:16 Tartarus, but it doesn't want to create the tarball...maybe because I need to clean and rebuild something? Jun 22 03:48:20 Tartarus, I've done: bitbake -c clean micro-base-image.bb && bitbake micro-base-image.bb and the tarball is not created Jun 22 04:18:27 sigh Jun 22 04:18:38 python's broken again >.< Jun 22 04:18:40 stupid fedora Jun 22 04:43:00 hey guys Jun 22 04:43:14 how can I ditch the autotools.bbclass definition of EXTRA_AUTOMAKE > Jun 22 04:43:39 it's passing a weird-ass value (--exclude=autopoint) which seems not to be valid with every version of autoreconf I can find Jun 22 07:07:39 is there any timeline for the next stable release? Jun 22 07:27:34 anyone here familiar with ntpd ? Jun 22 07:27:56 i keep getting error Jun 22 07:27:58 Stopping ntpd: start-stop-daemon: warning: failed to kill 1551: No such process Jun 22 07:27:59 done Jun 22 07:27:59 Starting ntpd: done Jun 22 07:28:26 but it doesn't change the time nor it will be available in processes Jun 22 07:28:50 morning Jun 22 07:30:30 ibot: botmail to kergoth I built console-image for angstrom-2008.1/at91sam9263ek from RPM packages out-of-box Jun 22 07:34:18 morning Jun 22 07:53:14 how do I rebuild a package? Jun 22 07:53:18 e.g. the kernel Jun 22 07:53:36 bitbake virtual/kernel :d Jun 22 07:54:35 bitbake -cclean virtual/kernel i mean Jun 22 07:54:49 and then build it again *-) Jun 22 07:55:02 ahh, the -cclean was missing :-) Jun 22 07:55:06 thanks Jun 22 08:28:19 morning Jun 22 08:34:37 morning. Jun 22 08:45:07 good morning Jun 22 08:48:11 good morning Jun 22 08:54:29 someone know how to force OO.org to really use DPI from system? Jun 22 08:54:49 I want to print A4 paper but 100% zoom on screen > A4 sheet Jun 22 08:56:11 hm Jun 22 09:01:54 hi florian Jun 22 09:22:05 * * OE Bug 5246 has been created by simaskonfa(AT)gmail.com Jun 22 09:22:07 * * opie for qemuarm freezes (udev start sequence problem) Jun 22 09:22:09 * * http://bugs.openembedded.org/show_bug.cgi?id=5246 Jun 22 09:40:44 I havent added a do_stage function to my recipe, yet the .pc file (which is the only thing to be staged) is there in the staging area correctly. Does bitbake automatically stage everything in /usr/lib ? Jun 22 09:44:10 are there any application available to parse gps nmea string in command line Jun 22 09:52:05 nope Jun 22 09:52:29 oespirit are you have inherit pkgconfig in your recipe? Jun 22 09:52:42 woglinde: yes Jun 22 09:52:56 that is doing staging for you Jun 22 09:53:21 but only for pkgconf stuff Jun 22 09:53:35 woglinde: oh ok Jun 22 09:53:44 woglinde: thanks Jun 22 09:54:10 webos seems a nice thing, but they don't use oe Jun 22 09:54:48 recalcati_oe: they use Jun 22 09:55:25 recalcati_oe: we got this information confirmed from Palm developers with permission to note that on website Jun 22 09:55:32 recalcati_oe they use oe Jun 22 09:55:41 half of the patches are from oe Jun 22 09:55:48 my zlib autotools patches are theire Jun 22 09:56:32 nice, but by now they don't publicize it. I see the webpage with the list of source and oe is not mentioned Jun 22 09:57:03 oe is very great: Texas is using it inside bsp support for omap3 evm Jun 22 09:57:05 recalcati_oe: they do not have to Jun 22 09:57:41 hrw: ok Jun 22 09:58:13 basically they do not even have to share their change to OE (due to MIT license) Jun 22 09:58:49 hrw did you add plam to the flyers? Jun 22 09:58:52 *g* Jun 22 10:01:29 but I wonder how it comply with GPL which require 'any associated interface definition files, plus the scripts used to control compilation and installation of the executable" Jun 22 10:01:49 woglinde: Mickeyl did flyers and we did not add Palm or Montavista to it Jun 22 10:01:53 hrw right Jun 22 10:02:02 for GPL Jun 22 10:02:19 woglinde: palm pre contains gpl kernel and few other things Jun 22 10:04:13 hrw yes I know Jun 22 10:04:22 and I know the build stuff Jun 22 10:05:50 but should I first get palm pre or is fetching webosdoctorp100ewwsprint.jar enough to ask? Jun 22 10:58:38 hi, how do I get a tarball from the rootfs? I've added tar to IMAGE_FSTYPES *after* having bitbaken the image...when I clean + re-bitbake the image it doesn't create the tarball... Jun 22 10:59:28 hello Jun 22 10:59:56 i am experiencing an issue with a fresh OE/bitbake install Jun 22 11:00:53 flameman, pastebin the error Jun 22 11:00:57 http://www.elinux.org/Flameman/bitbake Jun 22 11:00:58 what machine/distro? Jun 22 11:01:08 i documented everything there Jun 22 11:02:43 flyers will be sent today to me Jun 22 11:03:08 When do we need to setup the booth Jun 22 11:04:21 what do you thing about this issue ? Jun 22 11:06:21 do you have symlinks in the path to the oe dir? Jun 22 11:06:31 yes Jun 22 11:06:55 /src/oe ---> /mnt/diske/src/oe Jun 22 11:07:08 flameman: so you have an answer Jun 22 11:07:16 I am pretty sure that symlinks in that path lead to problems Jun 22 11:07:57 :) Jun 22 11:09:44 anyone? Jun 22 11:13:47 should I recreate my tmpfs? Jun 22 11:13:53 s/tmpfs/tmpdir Jun 22 11:22:59 I don't like the idea of re-creating the tmpdir only for that... Jun 22 11:23:15 GNUtoo: bitbake image again Jun 22 11:23:43 or check machine config - maybe it hardcode image formats Jun 22 11:25:44 hrw, ah ok because I already cleaned and bitbaked the image ...thanks a lot Jun 22 11:27:20 thanks a lot Jun 22 11:27:41 np Jun 22 11:30:09 morning Jun 22 12:05:07 does somebody know what can cause the following when trying to use sdma? "DMA transaction error with device 0" Jun 22 12:05:19 The fundamental problem showing of OE: Jun 22 12:05:46 Running an overo attached to kb/mouse/monitor running a flashy gui demo image is cool Jun 22 12:06:11 but does not show off our ability to generate firmware for truly embedded devices Jun 22 12:06:13 but Jun 22 12:06:25 embedded devices are "dull" :) Jun 22 12:12:38 Crofton|work: yep Jun 22 12:16:53 lol Jun 22 12:19:48 We seriously need to work out an environment for people to do sw development for OE derived distro's and promote it heavily Jun 22 12:19:56 and yes, I still need to pack Jun 22 12:21:16 Is is possible to place bb-files in build/recipes ? Jun 22 12:22:20 ???? Jun 22 12:22:49 you wan't a sort of 'overlay'? Jun 22 12:24:03 sgh, yes, it is possible to use an "overlay" like that, you have to add it to the local.conf file Jun 22 12:24:24 I used to do it before git, but now I find it easier to just use git branches Jun 22 12:24:29 valhalla: aaaahh .... thank you Jun 22 12:30:40 has anyone used the u-boot EABI hacks uboot-eabi-fix-HACK.patch or uboot-eabi-fix-HACK2.patch under .dev/recipes/u-boot? Jun 22 12:31:20 i keep getting an EABI version mismatch error when building u-boot and i'm unsure are those hacks still usable Jun 22 12:32:33 Viltapi: they should be used only by Zaurus...I did some builds couple of months ago Jun 22 12:37:08 ant_work: oh, well I keep getting this error "arm-angstrom-linux-gnueabi-ld: ERROR: Source object .../build/tmp/cross/arm/lib/gcc/arm-angstrom-linux-gnueabi/4.3.3/libgcc.a(_udivdi3.o) has EABI version 4, but target u-boot has EABI version 0" all the time, any idea how to get rid of it? Jun 22 12:37:57 Viltapi: if you use old uboot then you need those ones rather Jun 22 12:38:25 hrw: i'm using 1.3.4 Jun 22 12:41:03 Viltapi: we are still using U-Boot 1.1.4 (May 21 2006 - 17:28:48) Jun 22 12:41:48 we did not find any volunteer to update it Jun 22 12:42:02 ant_work: yea but i have to use a patch that is for 1.3.4 :( Jun 22 12:43:20 i wonder what those two different hacks are for Jun 22 13:00:43 mmm...is that correct: CONFIG_CMDLINE="root=/dev/nfs rw ip=dhcp nfsroot=168.0.0.2:/embedded/NFS console=ttyS0,115200 debug " because I have: bootserver=168.0.0.2, rootserver=168.0.0.2, rootpath= Jun 22 13:02:25 Viltapi: have you found it? Jun 22 13:03:02 http://markmail.org/message/467fyisljqr5fg6y#query:uboot%20eabi%20hack+page:1+mid:rv3rrhnmge62bpkn+state:results Jun 22 13:04:18 ant_work: i tried appending those files in my recipe but i get an error that the files it tries to patch cannot be found Jun 22 13:05:17 Whats the right way to go about if I want to run a bash script to modify files in ${S} before configure step ? Is do_configure_prepend(){ sh mysrcipt.sh} correct ? Jun 22 13:06:38 hey guys Jun 22 13:06:56 linking gcc's libgcc into uboot is a bit dirty, isn't it Jun 22 13:07:24 if there are some symbols missing they should be implemented in u-boot Jun 22 13:07:40 just like the kernel (or uboot-v2) does Jun 22 13:10:02 ah it mounted the root filesystem but didn't found init...sorry I'll try busybox Jun 22 13:13:02 How can I within a recipe change to a different directory than the working dir prior to executing a command ? Jun 22 13:17:14 oespirit_: either that or a separate task. if your changes aren't idempotent then a separate task might be better. Jun 22 13:18:39 BET_frogger: no, linking libgcc into uboot is perfectly respectable. it would be a bit silly for uboot to reimplement all of those operations for itself. Jun 22 13:18:52 03Koen Kooi  07org.openembedded.dev * rf5a0d86c80 10openembedded.git/recipes/zip/ (zip.inc zip_2.32.bb): zip: fix QA errors Jun 22 13:18:54 hola Jun 22 13:18:59 the kernel does, but the kernel people are notorious for their crazed behaviour Jun 22 13:19:02 g'day zecke Jun 22 13:20:46 pb_ : my script is a simple one-liner to grep Werror from all files and remove those flags with sed. Jun 22 13:21:09 zecke: I might need someone to do a bit of paid-for bitbake hacking. is that something you might be interested in? Jun 22 13:21:33 zecke, will you be at linuxtag? Jun 22 13:21:57 pb_: argh, in theory yes... in practive I'm on "customer" sites Jun 22 13:22:08 mmm...I think I've a problem with the root filesystem: Kernel panic - not syncing: Attempted to kill init! when I want to execute init=/bin/ash Jun 22 13:22:11 03Philip Balister  07org.openembedded.dev * r039c7a332f 10openembedded.git/recipes/gnuradio/gnuradio.inc: gnuradio.inc : GNU Radio is GPLv3. Update LICENSE. Jun 22 13:25:41 jo pb Jun 22 13:26:10 yo woglinde Jun 22 13:26:23 doh, why is IMAGE_EXTRA_SPACE set to 10240 by default?! Jun 22 13:26:26 * pb_ stabs bitbake.conf Jun 22 13:26:44 pb_: would it be possible for you to do it in-house with free assistance? Jun 22 13:27:15 zecke: probably :-) Jun 22 13:27:28 zecke: I'll send you an email with what we are trying to do and you can let me know your thoughts. Jun 22 13:27:42 what email address is best, your selfish one? Jun 22 13:28:10 oespirit_: look in recipes/u-boot.git.bb for sed exemples Jun 22 13:28:20 pb_: yes, or holger@freyther.de (getting old...) Jun 22 13:28:23 pb_, likely for silly, but historical reasons :) Jun 22 13:29:09 Crofton|work: yah, I guess so. it wouldn't be so annoying if IMAGE_EXTRA_SPACE didn't override IMAGE_ROOTFS_SIZE. Jun 22 13:29:10 yeah would be good if zecke get paid fo hacking on oe Jun 22 13:30:12 as it stands, the behaviour is that you get the larger of IMAGE_ROOTFS_SIZE or (size of rootfs contents)+IMAGE_EXTRA_SPACE. Jun 22 13:32:10 GNUtoo: "init=/bin/busybox ash" Jun 22 13:32:28 I'll try but I suspect a problem with my rootfs Jun 22 13:32:36 such as library problems or similar Jun 22 13:32:42 or worse abi problems Jun 22 13:32:49 thanks a lot Jun 22 13:33:09 if it were just a wrong commandline then you would probably expect to see a busybox usage message. Jun 22 13:33:28 but, I think the misfeature that required you to say "busybox ash" was fixed some time ago. iirc, just "init=/bin/ash" should be okay nowadays. Jun 22 13:33:52 so, I think you are probably right that you have bigger problems than that. can't hurt to try though. Jun 22 13:34:16 I didn't try it yet because I had to recompile the kernel...I'll do it now Jun 22 13:34:32 ok thanks Jun 22 13:35:59 03Klaus Kurzmann  07shr/import * r81cafd49dd 10openembedded.git/conf/distro/include/shr-autorev.inc: Jun 22 13:35:59 shr-autorev.inc: set rev for connman to the equivalent of 0.19 Jun 22 13:35:59 Signed-off-by: Klaus Kurzmann Jun 22 13:36:00 03Klaus Kurzmann  07shr/import * ra2dcc35078 10openembedded.git/conf/distro/include/shr-autorev-unstable.inc: Jun 22 13:36:02 shr-autorev-unstable.inc: set connman to AUTOREV Jun 22 13:36:04 Signed-off-by: Klaus Kurzmann Jun 22 13:36:06 03Klaus Kurzmann  07shr/import * ra2543819ec 10openembedded.git/recipes/connman/connman_git.bb: Jun 22 13:36:09 connman_git.bb: build connman from git Jun 22 13:36:11 Signed-off-by: Klaus Kurzmann Jun 22 13:36:35 Can I do "cd src" within a bb-file ? Jun 22 13:37:14 yes Jun 22 13:37:52 how .... I can't figure it out ... :( Jun 22 13:38:19 grep inside recipes Jun 22 13:38:35 by the way where can I find or how can I bitbake a cross-ldd? Jun 22 13:38:38 in which stage do you need it= Jun 22 13:38:53 GNUtoo: cross-ldd? Jun 22 13:38:57 hrw, ok thanks Jun 22 13:39:06 GNUtoo: cross-objdump -x binary|grep NEEDED Jun 22 13:39:13 hrw, ok thanks a lot Jun 22 13:39:17 i get this issue http://www.webalice.it/mr.ddc/bitbake.err Jun 22 13:40:09 hm why its parsing obsolete? Jun 22 13:40:20 too old bitbake? Jun 22 13:41:04 it's fresh form git pull Jun 22 13:41:10 bitbake is fresh from the svn Jun 22 13:41:13 hm Jun 22 13:41:20 strange Jun 22 13:41:22 woglinde: ok .. Jun 22 13:41:32 sgh which stage? Jun 22 13:41:44 compile unpack staging? Jun 22 13:43:08 hi prpplague Jun 22 13:43:32 hrw: greetings Jun 22 13:43:46 hrw: what's cookin today? Jun 22 13:44:21 webos on beagleboard Jun 22 13:44:53 hrw *g* Jun 22 13:45:49 hrw stealing stuff from the image? Jun 22 13:46:17 woglinde: why stealing? Jun 22 13:46:38 zecke the nonfree stuff Jun 22 13:46:43 woglinde: I think Palm gave a two week period of legal license violation to everyone Jun 22 13:47:18 zeck I want the upstart.bb Jun 22 13:47:40 woglinde: I can provide you upstart.bb but without any tweaks done Jun 22 13:47:59 hrw hm? Jun 22 13:48:44 woglinde: rootfs contains /usr/lib/ipkg/ so when upstart is built (this part lacks cflags etc) I can package results in similar packages. no warranty that it will even work Jun 22 13:50:18 hm what are libmatthew-java and palmipc-java Jun 22 13:51:10 woglinde: http://www.matthew.ath.cx/projects/java/libmatthew-java-0.7.tar.gz Jun 22 13:51:27 hrw: webos eh? Jun 22 13:51:38 palmipc-java is Palm only Jun 22 13:51:43 prpplague: Palm Pre system Jun 22 13:51:44 hrw: i'm still doing wince stuff :( Jun 22 13:52:21 prpplague uh Jun 22 13:53:05 woglinde: not by choice... Jun 22 13:54:01 hrw hm plamipc-java contains libmathew-java.jars and java-dbus Jun 22 13:56:03 hm dosfstools too, I hope they learned from tomtom Jun 22 14:35:52 mmm...hello world + while (1){sleep(1)} works...statically Jun 22 14:37:57 mmm....only static things work... Jun 22 14:45:31 how can I get more verbose things...when I execute the dynamic hello world it doesn't print anything...and exits Jun 22 14:51:08 GNUtoo: try a statically linked strace + your hello world Jun 22 14:51:16 ok thanks Jun 22 15:16:40 mmm...segmentation fault...http://www.pastebin.ca/1469948 do I need gdb-static over serial?(I should find a second serial port) Jun 22 15:18:42 bye Jun 22 15:20:01 GNUtoo: do you have another rootfs that does work? Jun 22 15:20:15 if yes, it might be easiest to boot with that, then copy your failing binaries into it and debug them that way. Jun 22 15:20:18 pb_, yes but: Jun 22 15:20:33 *booth working and non-working use uclibc Jun 22 15:21:02 *the other rootfs is not generated with oe Jun 22 15:21:31 it doesn't really matter what the other rootfs was generated with, so long as you can boot into it and run gdb or some such. Jun 22 15:21:45 boot the working gdb, copy the broken binaries into a ramdisk, then run them under the debugger Jun 22 15:22:40 ok thanks Jun 22 15:24:26 *one has 2.4 and the other has 2.6...maybe I should try and flash a working 2.6 into the device Jun 22 15:25:46 the arch is mips Jun 22 15:25:50 or mipsel Jun 22 15:25:53 I should check Jun 22 15:26:36 mipsel Jun 22 15:27:12 is there anything like oabi and eabi in mips? Jun 22 15:47:50 GNUtoo: not sure but libffi mentions N32, N64 and perhaps something else (and there is the little+big endianess support, of course) Jun 22 15:48:03 ok thakns Jun 22 15:48:23 thebohemian: hello Jun 22 15:48:47 * thebohemian untarring libffi source Jun 22 15:48:53 thebohemian: my memory is short-time..did I already ask you if you're building *on* mipsel? Jun 22 15:48:55 thebohemian, by the way I've serious issues with wesnoth 1.6.x it segafaults and it's non trivial and I don't know well C++ etc... Jun 22 15:49:26 GNUtoo: there is something called o32 and n32. I have the impression that this are two ABIs Jun 22 15:49:33 thebohemian, ok Jun 22 15:50:01 can it be seen with readelf? Jun 22 15:50:04 ant_work: no I don't have such a device or are targetting one Jun 22 15:50:09 (if someone knows) Jun 22 15:50:20 GNUtoo: It would be odd if it can't Jun 22 15:50:24 ok Jun 22 15:50:26 thanks Jun 22 15:52:14 GNUtoo: the binutils just recently got support for a totally new CPU architecture called moxie: http://moxielogic.org/blog/ Jun 22 15:52:37 GNUtoo: so it would be strange if well-known stuff is not supported ;) Jun 22 15:52:47 ok Jun 22 15:52:48 thanks Jun 22 15:53:49 ant_work: do you have trouble with something mips related that I should look at? Jun 22 15:54:01 ant_work: something with the Java recipes maybe? Jun 22 15:54:08 no, thx Jun 22 15:54:24 is just flameman having had horrible problems trying a build on mipsel Jun 22 15:54:35 ah ok Jun 22 15:54:36 now he found an old pentium3 :) Jun 22 15:55:46 iirc there was some buzz about mipsel lately...probably as target, though Jun 22 15:55:55 thx anyway Jun 22 15:56:52 it seems fairly unlikely anybody would bother using mips as a host nowadays. might have made sense 15-20 years ago, but probably not now. Jun 22 15:58:31 yes i am using a native mipsel machine class 4 to build for an embedded mipsel class 3 Jun 22 15:58:58 trying to fix up the stuff cause i'd like to cross compile from my x86 Jun 22 16:02:42 03Koen Kooi  07org.openembedded.dev * r62720413fe 10openembedded.git/ (conf/checksums.ini recipes/bluez/bluez4_4.42.bb): bluez4: add 4.42 Jun 22 16:08:45 pb_: hey I was just about to revive my Octane for OE builds ;) Jun 22 16:09:38 florian: heh, clearly I was mistaken :-) Jun 22 16:13:47 bye Jun 22 16:16:46 any machines using Atom exist? curious if there is any specific tuning that should be used - I don't see anything jumping out at me Jun 22 16:17:25 florian, when do we need to setup the OE booth? Jun 22 16:17:59 i need to focus on catching a plne Jun 22 16:44:40 g'day kergoth Jun 22 16:51:17 I am bitbacking gpe-image for hours now, and for 160min I have opkg-cl running 100% CPU Jun 22 16:51:20 is that normal? Jun 22 16:55:15 pb_: hey Jun 22 16:56:15 pb_: do you happen to know if the shlibs based rdeps get used by bitbake when it's figuring out its runqueue? Jun 22 16:59:10 Treibholz, it depends on the iron you use, but my machine use 2 days + the nights from scratch Jun 22 16:59:46 Treibholz, so i cxan say yes! Jun 22 17:00:19 dth: it's not compiling anymore! Jun 22 17:00:35 NOTE: Running task 2986 of 3524 (ID: 10, /home/kumbach/src/HTC/openembedded/recipes/images/opie-image.bb, do_rootfs) Jun 22 17:00:48 for ~3h now Jun 22 17:00:52 what is he dooing? Jun 22 17:01:15 he's running task 2986! :-) Jun 22 17:01:27 opkg-cl is at 100% Jun 22 17:01:51 2986 from? Jun 22 17:02:05 3524 Jun 22 17:02:38 so take your time Jun 22 17:07:41 I guess I should answer Crofton|work via email :) Jun 22 17:07:53 pb_: :-) Jun 22 17:09:17 Treibholz: That's suspicious... you could stop it and take a look at the last log. I suspect that's a problem dunning do_rootfs. Jun 22 17:17:59 Treibholz: is a known issue Jun 22 17:18:05 it's OPIE btw, not GPE Jun 22 17:20:07 ant_work: sorry, yes. but I had the same Problem with GPE, before. Jun 22 17:20:40 I'm now trying to build gpe again and have a look tomorrow :-) Jun 22 17:21:03 unfortunately I seem unable to reach openembedded site / ML to show you the quick hack Jun 22 17:26:41 Treibholz: quick fix http://cgit.openembedded.net/cgit.cgi?url=openembedded/commit/&id=097a6b10895c48b4e58efdd7eb60877ed2e6cf10 Jun 22 17:28:23 in my case in OPIE image opkg-cl hangs asking you if you want upgrade bluez->bluez4 Jun 22 17:28:37 some missing provider thing Jun 22 17:28:44 have to go now, bbl Jun 22 17:28:48 thanks. Jun 22 17:57:19 03Koen Kooi  07org.openembedded.dev * r2efbc18b22 10openembedded.git/ (conf/checksums.ini recipes/pulseaudio/pavucontrol_0.9.8.bb): pavucontrol: add 0.9.8 Jun 22 18:41:15 Crofton: are you in Europe already? Jun 22 18:41:44 no Jun 22 18:41:51 In the airport in Roanoke :) Jun 22 18:42:00 florian, ping Jun 22 18:42:17 Crofton: ah, have a safe flight then! :) Jun 22 18:42:30 thanks Jun 22 19:06:58 I need to make a recipe for a kernel ofan embedded device not in OE's repos Jun 22 19:07:37 which kernel would you recomend me to look as an example? Jun 22 19:07:41 *kernel recipe Jun 22 19:08:39 it doesn't really matter, the kernels are all much the same. probably one of the vanilla linux_2.6.xx ones would be simplest. Jun 22 19:09:04 linux_2.6.30.bb for example Jun 22 19:16:30 03Denys Dmytriyenko  07org.openembedded.dev * radb9174a2f 10openembedded.git/recipes/base-files/base-files/profile: Jun 22 19:16:30 /etc/profile: stricter root PATH - require absolute pathname for bins in cwd Jun 22 19:16:30 Having current directory (either '.' or empty string) in PATH is considered Jun 22 19:16:30 dangerous for root. Jun 22 19:16:30 Signed-off-by: Denys Dmytriyenko Jun 22 19:16:33 Acked-by: Phil Blundell Jun 22 19:16:35 Acked-by: Koen Kooi Jun 22 19:18:36 oh well... missed Crofton again Jun 22 19:18:53 Who is able to fix the weserver? Jun 22 19:21:55 cbrake? mickey|zzZZzz? Jun 22 19:27:32 florian: hello Jun 22 19:28:20 cbrake: looks like the wiki doesn't work any more... Jun 22 19:28:34 florian: wonder if Rolf moved it to melo .... Jun 22 19:28:40 florian: checking ... Jun 22 19:28:46 03Sebastian Krzyszkowiak  07shr/import * r26299d4f4c 10openembedded.git/ (2 files in 2 dirs): Jun 22 19:28:46 opimd-utils: add recipe for opimd test scripts Jun 22 19:28:46 Signed-off-by: Klaus Kurzmann Jun 22 19:30:40 wiki is still at amethyst Jun 22 19:32:51 looks like the mysql server died Jun 22 19:33:05 florian: yup, I can't even stop it Jun 22 19:33:32 mysqld 57MB -- hmmm Jun 22 19:33:35 I know why I dislike wikies needing database servers ;) Jun 22 19:34:07 florian: heh Jun 22 19:34:20 03Dmitry Eremin-Solenikov  07org.openembedded.dev * r2e32cd12e1 10openembedded.git/recipes/iproute2/ (3 files in 2 dirs): Jun 22 19:34:20 iproute2: add new upstream version Jun 22 19:34:20 Signed-off-by: Dmitry Eremin-Solenikov Jun 22 19:34:20 florian: a bunch of apache instances seem to be going away Jun 22 19:34:21 03Dmitry Eremin-Solenikov  07org.openembedded.dev * r0319a593f7 10openembedded.git/conf/checksums.ini: Jun 22 19:34:22 checksums: add an entry for iproute2-2.6.29 Jun 22 19:34:24 Signed-off-by: Dmitry Eremin-Solenikov Jun 22 19:34:26 03Dmitry Eremin-Solenikov  07org.openembedded.dev * r0dcaad7774 10openembedded.git/: Merge branch 'org.openembedded.dev' of git://git.openembedded.org/openembedded into org.openembedded.dev Jun 22 19:35:15 florian: sigh -- * /etc/init.d/mysql: ERROR: The partition with /var/lib/mysql is too full! Jun 22 19:35:31 florian: out of disk space :-\ Jun 22 19:40:22 cbrake: ah well, that's a real showstopper Jun 22 19:40:37 florian: yup Jun 22 19:44:17 florian: man linuxtogo site seems slow today Jun 22 19:45:53 re Jun 22 19:46:00 kergoth: read botmail? Jun 22 19:46:07 ok, sounds like I need to do some admin job as well Jun 22 19:46:11 hi hrw Jun 22 19:47:52 florian: i just did a svnadmin load for the openocd-redux repo Jun 22 19:47:55 prpplague: yes i see... quite some load currently. Jun 22 19:48:09 florian: related to my checkin? Jun 22 19:48:43 prpplague: unlikely... looks like some apache hickup or similar Jun 22 19:48:46 hrw: yep, thanks, I'm trying what you've had good luck with, and will go from there :) Jun 22 19:48:52 florian: ok thanks Jun 22 19:49:38 kergoth: I built images for arm and x86 with rpm packages and was able to install rpms on target device (using rpm and/or yum) Jun 22 19:49:50 excellent Jun 22 19:51:10 kergoth: and now, when I know that it works for me, I prefer to forget that I ever touched that stuff Jun 22 19:51:16 hehe Jun 22 19:51:18 understandable Jun 22 19:51:24 i wouldn't be messing with it if i had a choice in the matter Jun 22 19:52:43 mvl? Jun 22 19:55:44 pb__: ok, I'll try Jun 22 19:59:12 03Marcin Juszkiewicz  07org.openembedded.dev * rad42bf9065 10openembedded.git/recipes/nbench-byte/ (nbench-byte/nbench_32bits.patch nbench-byte_2.2.3.bb): Jun 22 19:59:12 nbench-byte: added BYTEmark benchmark Jun 22 19:59:12 BYTE Magazine's native benchmarks (also called BYTEmark) designed to Jun 22 19:59:12 expose the capabilities of a system's CPU, FPU and memory system. Jun 22 19:59:12 http://www.tux.org/~mayer/linux/ Jun 22 19:59:16 Recipe from old Atmel overlay. Jun 22 19:59:18 03Alan Carvalho de Assis  07org.openembedded.dev * r0526f70737 10openembedded.git/ (4 files in 4 dirs): Jun 22 19:59:21 Add basic support for i.MX27ADS board Jun 22 19:59:23 This patch adds basic support on OpenEmbedded for i.MX27ADS board. Jun 22 19:59:25 It uses the mainline kernel instead Freescale one. The mainline doesn't Jun 22 19:59:27 have support for USB, Camera, etc. Jun 22 19:59:29 Signed-off-by: Alan Carvalho de Assis Jun 22 19:59:31 Acked-by: Marcin Juszkiewicz Jun 22 20:00:05 03Marcin Juszkiewicz  07org.openembedded.dev * rf642b5b1e0 10openembedded.git/conf/checksums.ini: checksums.ini: added few entries Jun 22 20:07:13 03Florian Boor  07org.openembedded.dev * rcabde07cc8 10openembedded.git/conf/checksums.ini: checksums.ini: Add zbar Jun 22 20:59:42 03Stanislav Brabec  07org.openembedded.dev * rd9dc503e42 10openembedded.git/recipes/keymaps/keymaps_1.0.bb: keymaps: RDEPENDS fix. Jun 22 20:59:43 03Stanislav Brabec  07org.openembedded.dev * ree7dab99e1 10openembedded.git/recipes/ (3 files in 3 dirs): xserver-common, zaurusd: Use native xserver orientation for spitz, akita and borzoi, updated X11/Xinit.d/11zaurus. Jun 22 20:59:43 03Stanislav Brabec  07org.openembedded.dev * r35b5303461 10openembedded.git/recipes/zaurusd/ (4 files in 2 dirs): zaurusd: Support configurable portrait and landscape xrandr orientation. Jun 22 21:22:39 hmm, what was roman khimov's irc nick again? Jun 22 21:23:51 his name Jun 22 21:24:20 ah Jun 22 21:42:05 * * OE Bug 5251 has been created by ronald(AT)sysrealtime.com Jun 22 21:42:07 * * CROSS COMPILE Badness: Jun 22 21:42:09 * * http://bugs.openembedded.org/show_bug.cgi?id=5251 Jun 22 21:56:26 just a last question before going to sleep: is there a possibility to create a root filesystem(for NFS) that has debug symbol on all libraries and binaries? Jun 22 21:57:04 it's because I can't run the root filesystem at all(all non-static binaries segfault) Jun 22 21:57:18 but I can use gdb from another filesystem on flash Jun 22 22:04:53 GNUtoo: add the dbg packages you want to the rootfs. or don't, and use a gdbserver Jun 22 22:07:21 kergoth, ok I'll make a custom image with all the dbg packages thanks a lot Jun 22 22:07:31 kergoth, and I use gdbserver Jun 22 22:07:37 kergoth, but I usually do: Jun 22 22:07:47 set sysroot /embedded/NFS Jun 22 22:07:52 i can't think of any better way of doing it, offhand. unless you were to do something like RRECOMMENDS_${PN} += "${PN}-dbg" or something, perhaps Jun 22 22:07:54 * kergoth shrugs Jun 22 22:08:10 file /embedded/NFS/hello Jun 22 22:08:20 target remote etc... Jun 22 22:08:31 ok thanks a lot Jun 22 22:09:47 because otherwise I would have to do a custom gdb script if I remember well... Jun 22 22:10:03 kergoth: about recipe_sanity...http://fr.pastebin.ca/1470392 Jun 22 22:10:24 the recipe itself hasn't any RDEPENDS Jun 22 22:10:32 these are included... Jun 22 22:11:15 ant__: there's no way for recipe_sanity to check that for sure, classes and stuff get merged at parse time. Jun 22 22:11:27 ant__: its left to the user to figureo ut if they need to fix a .inc or whatever instead of the recipe itself Jun 22 22:12:18 ant__: maybe just silence the message for images :) images don't emit packages anyway, so its not a problem Jun 22 22:12:32 ant__: i'll make it shut up about those when PACKAGES is empty. Jun 22 22:12:36 I see...I suspect lot of 'unexplicable' things are caused by RDEPENDS vs. RDEPENDS_${PN} Jun 22 22:13:05 indeed, it probably explains a lot of "how did all this extra crap get pulled into my image" problems Jun 22 22:13:06 heh Jun 22 22:13:37 nice catch, congrats :) Jun 22 22:15:09 don't congrat me, it was pb_ 's idea, i just added it to recipe_sanity :) Jun 22 22:15:21 * kergoth 's trying to add as many useful checks to it as possible Jun 22 22:16:48 yea, pb's post in the ML is very explicative Jun 22 22:24:09 kergoth: another question... recipe_sanity: LICENSE should be defined to something other than default (unknown) Jun 22 22:24:57 any sources fetched by this recipe Jun 22 22:35:14 hm.. I added a new recipe (zaurus-installer), just a wrapper putting together zaurus-updater and a kexecboot-enabled kernel.. Jun 22 22:35:43 well..zaurus-updater has LICENSE = "zaurus-updater" Jun 22 22:35:45 sic Jun 22 22:36:21 but DEPENDS = "encdec-updater-native" (LICENSE = "GPL") Jun 22 22:37:15 Yes, it depends on what the two actually do together :) Jun 22 22:37:41 And, heh, I wonder why gdbserver.inc has depends on curses/readline, don't need that for just gdbserver Jun 22 22:37:47 * Tartarus goes to check his logs once done Jun 22 22:45:20 hm, am I alone in finding nautilus's default thumbnail for jpegs a bit disturbing? Jun 23 01:20:16 03Tom Rini  07org.openembedded.dev * r85d95bc654 10openembedded.git/ (conf/checksums.ini recipes/netperf/netperf_2.4.4.bb): netperf: Fix SRC_URI (2.4.4 is in archive now) Jun 23 02:29:11 Anyone know of any companies that sell touchscreen dev platforms? Kind of like what restraunts have to place orders to the kitchen? Jun 23 02:32:58 er, omap3evm has a ts and runs 1500USD, i think **** ENDING LOGGING AT Tue Jun 23 02:59:57 2009