**** BEGIN LOGGING AT Thu Jul 17 02:59:57 2008 Jul 17 03:07:07 * * OE Bug 2542 has been RESOLVED (WONTFIX) by raj.khem(AT)gmail.com Jul 17 03:07:09 * * Allow glibc to build with i686 Jul 17 03:07:11 * * http://bugs.openembedded.net/show_bug.cgi?id=2542 Jul 17 06:34:39 good morning! Jul 17 06:34:42 ping khem about bug 2542 Jul 17 07:19:41 Laibsch: hey Jul 17 07:19:53 hey Jul 17 07:19:59 why wontfix? Jul 17 07:20:16 Isn't this "don't intend to fix in the proposed way"? Jul 17 07:20:30 but certainly WILLFIX Jul 17 07:23:27 Laibsch: hmmm I think yes but that would be a separate bug and IIRC we already fixed it Jul 17 07:24:20 Then the bug is fixed Jul 17 07:24:26 certainly not wontfix Jul 17 07:25:04 I don't think it would be a separate bug in any case Jul 17 07:25:19 I'll adjust the status Jul 17 07:25:28 thanks for marking this Jul 17 07:27:34 Laibsch: ok Jul 17 07:28:06 * * OE Bug 2542 has been REOPENED by Jul 17 07:28:08 * * Allow glibc to build with i686 Jul 17 07:28:10 * * http://bugs.openembedded.net/show_bug.cgi?id=2542 Jul 17 07:40:53 morning Jul 17 07:51:42 03dirk 07org.oe.dev * r4401f16b... 10/ (3 files in 2 dirs): stress: upgrade to 1.0.0 Jul 17 08:10:59 03utx 07org.oe.dev * r78b94abf... 10/ (3 files in 3 dirs): Jul 17 08:10:59 linux-rp-2.6.26: Added zaurus-i2c-init.patch: Jul 17 08:10:59 * initialize I2C on Poodle and Spitz Jul 17 08:10:59 * sound works again Jul 17 08:10:59 * suspend does not hang if sound modules are loaded Jul 17 08:17:34 03dirk 07org.oe.dev * r4d0f3853... 10/ (6 files in 3 dirs): Jul 17 08:17:34 busybox: add new version 1.11.1 Jul 17 08:17:34 - "Starting syslogd/klogd: start-stop-daemon: lseek: Invalid argument" fixed in this version Jul 17 08:26:39 morning Jul 17 08:50:28 good morning (UTC) Jul 17 08:52:27 good morning Jul 17 08:52:42 it ugt anyway ;) Jul 17 08:53:23 morgen florian Jul 17 08:54:10 xjqian: hi Gordon, wb Jul 17 08:54:31 Hi, ant Jul 17 08:57:31 florian: good morning Jul 17 09:04:43 good morning Jul 17 09:12:22 hi do13 Jul 17 09:12:48 hi florian Jul 17 09:18:54 how to set the jffs2 image size? i tried ROOT_FLASH_SIZE and IMAGE_ROOTFS_SIZE_jffs2 but they don't seem to work. it's still making a jffs2 just enough for what's in it. Jul 17 09:20:09 pigeon_: you need padding? Jul 17 09:20:21 pigeon_: look at conf/machine/collie.conf then Jul 17 09:20:53 hmm, basicially i want to have free space in the output jffs2 file system, is that what's the padding for? Jul 17 09:23:08 pigeon_: jffs2 use whole flash partition for own use Jul 17 09:23:26 so if you have 32M partition and rootfs.jffs2 is 17M then you will get 15M free space Jul 17 09:23:57 hmm, when i mount it though, it said there's 0 free space Jul 17 09:24:02 padding is required only for devices with stupid bootloaders which require full clean partition Jul 17 09:24:32 pigeon_: what is size of target flash partition? what is size of generated rootfs.jffs2? Jul 17 09:25:21 the nand flash is 256MB, the generated jffs2 is 4M, i flashed onto it, booted linux and using that as rootfs, but df shows 4M, 0 freespace. Jul 17 09:25:50 show me /proc/mtd of that device? Jul 17 09:26:40 ahha! Jul 17 09:26:53 that explains a lot :) Jul 17 09:27:04 mtd0: 00400000 00020000 "Bootstrap" Jul 17 09:35:39 thanks btw Jul 17 10:48:48 hrw: have you 1 min for me before lunch? Jul 17 10:52:07 go Jul 17 10:56:55 hrw: hi Jul 17 10:57:18 I overlooked you mentioned udev_124 Jul 17 10:57:37 Poky use it since few days Jul 17 10:57:50 would you check the bb I posted and eventualy add your changes? Jul 17 10:58:12 where it is posted? Jul 17 10:58:26 ML -> oebug 4118 Jul 17 10:58:58 !oebug 4118 Jul 17 10:58:59 * * Bug 4118, Status: NEW, Created: 2008-03-25 08:47 Jul 17 10:59:00 * * xjqian(AT)gmail.com: udev-118 failed to create device nodes on Tosa Jul 17 10:59:01 * * http://bugs.openembedded.net/show_bug.cgi?id=4118 Jul 17 10:59:08 approaching the solution ...giving udevsettle a timeout of 5 is not enough on c7x0 Jul 17 10:59:20 mmc and mtd are skipped Jul 17 10:59:28 I'll post new logs Jul 17 10:59:52 to get it work I have to comment out the udevsettle line Jul 17 11:00:11 this happens apparently only on c7x0 Jul 17 11:00:21 and Tosa Jul 17 11:00:55 hmm. will have to unpack my c7x0 then Jul 17 11:01:05 oops sorry...;-) Jul 17 11:01:35 do you have test image? Jul 17 11:01:51 I can provide it later evening Jul 17 11:02:27 ok Jul 17 11:03:42 hrw: already moved to linux Jul 17 11:03:56 linux-rp_2.6.26 ? Jul 17 11:04:22 I'v seen a fresh commit today Jul 17 11:04:25 I have 2.6.26-rc4 on device Jul 17 11:05:01 ok, I had some suspend issues as I told you. Need to retest Jul 17 11:06:50 ahemm, in OE we have r0 Jul 17 11:26:44 hrw: nice to see Poky has a brand new udev_124 init! Jul 17 11:27:02 ant->canapka Jul 17 11:27:04 bbl Jul 17 11:31:07 03utx 07org.oe.dev * ra8473c5c... 10/ (1 packages/linux/linux-rp-2.6.26/zaurus-i2c-init.patch): Typo in poodle chunk fixed. Jul 17 12:02:05 re Jul 17 12:06:52 ant|work: wb Jul 17 12:07:07 sandwich was indeed ugly :-( Jul 17 12:23:02 hello there! Jul 17 12:23:27 i got nice kernel patches for squashfs3.3+lzma for 2.6.24 kernel here Jul 17 12:23:34 http://rabenfrost.net/celtune/misc/linux-openmoko-2.6.24-squashfs-lzma-patches.tar.bz2 Jul 17 12:24:03 they were tested and the first working ipkg packages for openmoko gta01 and freerunner are out there Jul 17 12:24:21 does someone want to include the patch into oe? Jul 17 12:25:44 RP or hrw maybe? Jul 17 12:26:43 josch|nsn: OE handle lot of devices Jul 17 12:26:54 which one you are interested in? Jul 17 12:28:11 openmoko neo1973 and freerunner Jul 17 12:29:31 so talk with their maintainers Jul 17 12:31:07 hrw: mickeyl is still asleep :( Jul 17 12:32:48 * XorA is awake though Jul 17 12:32:52 \o/ Jul 17 12:33:16 josch: please file an OE bug, and Ill see what I can do Jul 17 12:33:24 XorA: okay thanks Jul 17 12:33:40 at least if bug is filed, mickey|zzZZzz can see it later Jul 17 12:33:46 ;) Jul 17 12:35:36 is this device-specific? Jul 17 12:35:44 i doubt it Jul 17 12:37:07 XorA, hrw: if this is not device-specific, maybe one of you can indeed be bothered to take a look at the overall quality of the patch? I'd certainly like to see it in OE. Jul 17 12:37:39 Laibsch: currently I am writing few other things and this is kernel patch. Jul 17 12:37:46 I never was kernel hacker Jul 17 12:38:54 Laibsch: well I assume each kernel needs patched individually Jul 17 12:39:01 XorA: did I mention we have sd working on kaiser? :P Jul 17 12:39:08 dcordes: you did not Jul 17 12:39:17 dcordes: so booting problems are solved :-) Jul 17 12:39:18 you can even boot from it Jul 17 12:39:24 yyep Jul 17 12:39:45 * XorA dances around the channel Jul 17 12:40:08 I shall be without internet access for a couple of weeks from friday, but Ill certainly look at kaiser again when I return :-D Jul 17 12:40:09 XorA: would you like a build until it is in the git? Jul 17 12:40:28 * XorA is busy sorting out RL at the moment Jul 17 12:40:55 http://www.headfuzz.co.uk/files/android/zImage uuh rl Jul 17 12:42:38 XorA: couple of weeks w/o internet? That sounds scary. Jul 17 12:42:55 Laibsch: new flat, need to transwer Jul 17 12:42:59 transfer Jul 17 12:43:20 so Ill be GPRS only unless someone is kind enough to leave an open accesspoint Jul 17 12:44:01 you can use mine but I fear it lacks txpower Jul 17 12:45:00 :-D Jul 17 12:45:12 its a large block of flats, Im sure someone failed to set password Jul 17 12:46:29 XorA: Not even 3G coverage? Jul 17 12:46:46 broonie: depends what side of the building I am on Jul 17 12:46:54 broonie: not sure where Im going to put my office yet Jul 17 12:47:37 Ah, my flat's the same - I hardly get 2G coverage at the street side but the back of the building has 3G. Jul 17 12:48:15 broonie: the flats form a U and reception inside the U isnt too hot, but the Dock Place side should be ok Jul 17 12:48:31 broonie: I can spy on your old office from my bedroom window Jul 17 12:48:34 Heh, just back next to my old work. Jul 17 12:48:58 my wifi should reach the bar, the bar has good beer, I see some benefits :-) Jul 17 12:50:13 there was a tip on the old openembedded index page to generate sources directory of libre software Jul 17 12:50:52 Which bar? Jul 17 12:50:56 does someone remember ? Jul 17 12:51:00 the one on the end of dock place Jul 17 12:51:08 Sirius or the Cameo? Jul 17 12:51:11 used to do a good line of guest ales Jul 17 12:51:19 other end Jul 17 12:51:23 Oh, right - the Waterfront. Jul 17 12:51:28 next to rennies lock Jul 17 12:51:37 Yeah, that's quite nice if tiny. The restraunt's really great too. Jul 17 12:52:13 restaurant has changed Jul 17 12:52:20 looks snazzier now Jul 17 12:52:31 Yeah, they redecorate every few years. Jul 17 12:53:22 Old Dock Bar, thats the one I was thinking o Jul 17 12:53:23 f Jul 17 12:55:00 broonie: http://www.citikey.com/business/10302327/ thats the one :-D Jul 17 12:55:48 Ah, yes. It's OK but always felt a bit lacking in atmosphere. Also, they were done for trading without a license while Aculab was there :/ Jul 17 12:56:37 hmm, used to be a good joint, went in there a lot while I was living there Jul 17 12:57:03 but I guess that might be later than you would if you were working Jul 17 12:57:38 Yes, we always worked from the bar! Jul 17 12:57:47 Genesis: src_distribute* classes Jul 17 12:58:41 broonie: heh, no I meant, I normally went in about 10pm, not 6pm like I guess you would if going after work Jul 17 12:59:31 Ah, we were known to stay fairly late. Jul 17 13:00:28 * XorA sometimes wonders how any software ever gets written Jul 17 13:00:37 thx Marcin :) Jul 17 13:00:41 the amount of beer consumed by the average software engineer Jul 17 13:03:29 XorA: ballmer peak Jul 17 13:04:42 maybe vista programmers didnt drink enough beer Jul 17 13:05:47 yeap they drink cola Jul 17 13:05:55 cool aid Jul 17 13:05:57 too sugar , not enough alcohol Jul 17 13:05:58 :) Jul 17 13:06:09 too much sugar. Jul 17 13:06:24 they best thing to do with sugar is ferment it :-D Jul 17 13:07:22 yeap ! Jul 17 13:07:27 i do my kombucha Jul 17 13:07:33 it's fun Jul 17 13:14:06 * * OE Bug 4437 has been created by j.schauer(AT)email.de Jul 17 13:14:08 * * squashfs3.3+lzma patch for 2.6.24 kernel Jul 17 13:14:10 * * http://bugs.openembedded.net/show_bug.cgi?id=4437 Jul 17 13:18:25 hrw: any idea what's happened to outo in org.openembedded.dev? Jul 17 13:19:37 hrw: oh, its in libosso now. ok so just deps need to be fixed up :) Jul 17 13:29:42 outo? Jul 17 13:51:44 josch Jul 17 13:51:49 er Jul 17 13:51:56 hi Jul 17 13:51:58 josch: nice xkcd reference awhile back :) Jul 17 13:53:41 woglinde: Morning Jul 17 13:54:46 hi prioko Jul 17 13:55:31 !seen zecke Jul 17 13:55:32 zecke (n=ich@118-166-68-228.dynamic.hinet.net) was last seen quitting from #oe 22 hours, 8 minutes ago stating (Read error: 104 (Connection reset by peer)). Jul 17 13:55:57 prioko *g* what you want from zecke? Jul 17 13:57:37 hi kgilmer Jul 17 13:57:50 morning woglinde :) Jul 17 13:58:25 we are moving office and don't have reliable networking yet...so I'm not here very often :( Jul 17 13:58:35 oh Jul 17 13:58:41 bigger nicer office? Jul 17 13:58:46 or smaller cheaper? Jul 17 13:58:48 *g* Jul 17 13:58:59 yep :) bigger office in a hot part of town. soho nyc. Jul 17 13:59:08 hehe Jul 17 14:18:53 http://blog.haerwu.biz/2008/07/17/using-poky-sdk-to-build-software/ Jul 17 14:25:16 woglinde: He is looking into the qtopia-not-building issue I'm having Jul 17 14:25:43 hrw: neat article. Does packaged staging make that work, or would this have worked before? Jul 17 14:26:24 josch|nsn: you forgot to add a defconfig to those patches, do I need to set the Additional Options for Constrained Systems? Jul 17 14:26:34 cbrake: before Jul 17 14:26:51 cbrake: I just had to take time and write it Jul 17 14:27:00 s/it/post about it Jul 17 15:04:28 bye Jul 17 15:22:34 re Jul 17 15:35:06 hi Jul 17 15:35:37 Marcel_M, XorA asked wether the module needs Additional Options for Constrained Systems activated Jul 17 15:36:08 pfu got no idea if its needed Jul 17 15:36:47 Marcel_M, i dont even know what it really does - did you have it activated? Jul 17 15:36:54 i didnt compile that option in Jul 17 15:36:55 no Jul 17 15:37:12 XorA, there you have it :) Jul 17 15:40:28 coo;l Jul 17 15:55:43 anyone building a powerpc arch machine for KERNEL_IMAGETYPE=uImage? I find that kernel.bbclass makes in invalid test for compressed kernel for this architecture when running mkimage Jul 17 15:56:29 hm Jul 17 15:56:38 In troubleshooting this issue I find that my arm based machine builds a uImage without the do_deploy task defined in kernel.bbclass - that I don't quite understand Jul 17 15:57:00 as far I know mkimage sets only a smalle header in front of your bzImage/zImage Jul 17 15:57:53 yes it does... but the issue I see is that in do_deploy for kernel.bbclass there is a test for arch/powerpc/boot/compressed/vmlinux which I don't believe is correct... it ends up telling mkimage that the image is compressed when it shouldn't Jul 17 15:58:24 hm Jul 17 15:58:28 let me see Jul 17 15:58:37 the compression field in the 64 byte uboot header that mkimage applies tells uboot if it should uncompress prior to jumping to the kernel vs the kernel having its own decomp Jul 17 15:58:47 only foer powerpc? Jul 17 15:59:04 in my case, I'm trying to build a compressed kernel, in which case I don't want uboot to decomp it Jul 17 15:59:36 if test -e arch/${ARCH}/boot/compressed/vmlinux ; then Jul 17 15:59:36 ${OBJCOPY} -O binary -R .note -R .comment -S arch/${ARCH} Jul 17 15:59:37 /boot/compressed/vmlinux linux.bin Jul 17 15:59:40 this one? Jul 17 15:59:52 well... the stuff in kernel.bbclass::do_deploy doesn't get called for my arm machine, so I can't understand for that machine where mkimage gets called... I think the kernel build does it Jul 17 16:01:07 are you sure? Jul 17 16:01:08 hmm... perhaps the issue is in my kernel build Jul 17 16:01:25 uncompressed kernel should be in kernel root called vmlinux correct? Jul 17 16:01:29 how looks the do_deploy() function? Jul 17 16:01:38 your look at the temp dir Jul 17 16:01:45 and a compressed one would be in boot/$ARCH/compressed/vmlinux Jul 17 16:01:51 yes Jul 17 16:02:00 thats what do_deploy looks for Jul 17 16:02:30 ah... then I bet do_deploy is doing the right thing and the issue is that my kernel build is not building a compressed vmlinux Jul 17 16:02:56 *g* Jul 17 16:03:54 03xora 07org.oe.dev * r912f3884... 10/ (7 files in 3 dirs): linux-openmoko_2.6.24+git.bb : add squashfs to the build #4437 Jul 17 16:06:06 so I have KERNEL_IMAGETYPE="uImage" in my machine conf... not sure why that isn't causing the kernel to build a compressed image Jul 17 16:06:18 hm Jul 17 16:06:28 you could look at the kernel Makefile Jul 17 16:06:39 for arm, KERNEL_IMAGETYPE="uImage" is causing the kernel recipe to do a 'make uImage' in kernel, but it would seem that for powerpc its doing a 'make vmlinux' Jul 17 16:07:51 where would I find the actual cmd that builds the kernel? in build/tmp/work/.../linux-.../temp/run.do_compile.*? Jul 17 16:09:06 * * OE Bug 4437 has been RESOLVED (FIXED) by dp(AT)xora.org.uk Jul 17 16:09:08 * * squashfs3.3+lzma patch for 2.6.24 kernel Jul 17 16:09:10 * * http://bugs.openembedded.net/show_bug.cgi?id=4437 Jul 17 16:09:32 tharvey yes Jul 17 16:09:38 and run_do_deploy Jul 17 16:10:33 thanks XorA:) Jul 17 16:10:34 josch|nsn: done Jul 17 16:10:51 ~hail xora Jul 17 16:10:53 * ibot bows down to xora and chants, "I'M NOT WORTHY!!" Jul 17 16:11:07 woglinde: nothing to do with me, I just applied the patches given Jul 17 16:11:12 *g* Jul 17 16:11:18 I know Jul 17 16:11:58 XorA|gone, oh nice! thank you very much! Jul 17 16:13:45 woglinde, do_compile is a bit hard to follow - and I don't see any of the std output that building a kernel manually would produce - are you sure I'm in the right file? Jul 17 16:14:30 oic.... log.do_compile Jul 17 16:21:20 tharvey you have run.do and log.do Jul 17 16:21:34 run.do are the commands which will be executed Jul 17 16:21:40 and log.do the output Jul 17 16:30:14 tharvey: there is a better way of autoloading modules. Jul 17 16:30:29 did you see the poky ml entry on this a few days back? Jul 17 16:31:23 kgilmer, no... I didn't - I guess I need to watch that list! Jul 17 16:31:51 kgilmer, is there a good archive of it? Jul 17 16:34:17 tharvey, from RP ~ Jul 17 16:34:20 You can autoload a module by adding a line like: Jul 17 16:34:20 module_autoload_ohci-hcd = "ohci-hcd" Jul 17 16:34:20 to the kernel recipe where "ohci-hcd" is the module name. Jul 17 16:35:48 i am now trying to add my own UDEV rules to my target. Jul 17 16:35:55 anyone now the best way of doing that? Jul 17 16:37:47 kgilmer, nice... thanks for the info! Jul 17 16:38:30 sure thing tharvey Jul 17 16:38:39 i see this variable in the udev recipe, UDEV_EXTRAS Jul 17 16:38:52 seems like something i might be able to use but can't find anything that uses it... Jul 17 16:39:06 why do you say thats a 'better way'? seems like thats a per-kernel way where touching /etc/modules is a 'per distro' way? - I'm actually not sure what populates /etc/modules so perhaps its not as easy as I would expect to mod that in the image Jul 17 16:40:16 well for us our kernel is machine specific. we can define what modules to autoload. Jul 17 16:40:34 it's handled at 'design time' vs after the image hsa been generated, like with /etc/modules Jul 17 16:41:04 I've found that you pretty much have to have a machine-specific kernel - if you want to have a custom defconfig that is and if you want to not touch the oe base files but instead add your own local tree Jul 17 16:41:04 i guess it depends on usage. for us which modules to autoload is machine specific, not distro specific. Jul 17 16:43:19 he zattoo Jul 17 16:43:21 looks like /etc/modules is populated from packages/modutils which makes sense - you can put a machine specific file in packages/modutils/files it looks like Jul 17 16:46:17 kgilmer, do you know how packages deal with auto-loading modules? I wonder if their config/install scripts edit /etc/modules? Would seem perfect to have a /etc/modules.d that module packages could manage Jul 17 16:46:19 is the error message "preferred version of glibc 2.6.1 not available" something that could break a bitbake build? Jul 17 16:46:46 kdubois, no... but I would sure like to understand how to get rid of it Jul 17 16:47:24 hrm, i'm at a loss then why fso-image compile fails Jul 17 16:48:44 pastbin the bitbake output and lets see... Jul 17 16:50:00 the 'preferred version' msg is supposed to be saying that you have set PREFERRED_VERSION_glibc=2.6.1 but that version isn't available - yet angstrom sets this and glibc-2.6.1 'is' available and 'is' whats being built for me, so I don't understand the msg Jul 17 16:51:33 tharvey: http://pastebin.ca/1074481 Jul 17 16:53:07 the error from the log is FATAL: oe_libinstall: unable to locate shared library Jul 17 16:55:15 kdubois, also pastebin the log it refers you to: /home/kevin/source/fso/build/tmp/work/armv4t-angstrom-linux-gnueabi/libsdl-x11-1.2.11-r0/temp/log.do_stage.27617 Jul 17 16:57:38 woglinde, do you have experience with powerpc systems? Jul 17 16:57:45 http://pastebin.ca/1074485 Jul 17 16:58:59 kdubois, I'm not familiar with the image your trying to build - are you sure its supposed to be buildable? I've found many machines that are not currently buildable in the oe tree... perhaps the same is true for images Jul 17 16:59:24 is that log saying that your missing a library needed to continue? Jul 17 16:59:29 package-builds for libsdl-x11 http://tinderbox.openembedded.net/packages/libsdl-x11/ Jul 17 16:59:39 i am just going off of what the openmoko wiki suggested i use... Jul 17 17:00:05 failures only on do_populate_staging...suspicious Jul 17 17:01:44 if anyone can point me towards a fso image that builds, that would be great Jul 17 17:02:07 looking at the log.do_compile for my powerpc kernel build I see that mkimage was run during the kernel build but shows a Load ADdress and Entry Point of 0x00000000 - that seems suspicious as O Jul 17 17:02:22 as I've specified UBOOT_ENTRYPOINT=800000 Jul 17 17:03:00 tharvey a little bit Jul 17 17:03:07 but supper now Jul 17 17:04:15 still not clear why the powerpc kernel build seems to do a mkimage yet kernel.bbclass::do_compile 'also' does one - thats probably why my resulting uImage doesn't boot Jul 17 17:31:52 hrm, one problem may be that when trying to download libsdl, it cant connect Jul 17 17:35:21 yeah, libsdl.org is down, so i think thats my problem, ill wait for someone to bring it back up... Jul 17 17:37:30 kdubois, so you have a fetch error then (must have preceded the other errors in your output) - make sure you always focus on the first error shown Jul 17 17:37:52 you can search for the source from a mirror and place it in sources manually Jul 17 17:38:36 when you add a file manually to sources you need to touch the corresponding .md5 file there too Jul 17 17:40:16 right, i didnt notice that error because i left my machine while it was compiling, and it pops up in the middle of a very long log Jul 17 17:56:36 re Jul 17 18:03:16 so what I'm seeing with my powerpc linux kernel is that a 'make uImage' builds a non-compressed vmlinux, then compresses that and runs mkimage on it whereas my arm linux build 'make uImage' creates a compressed kernel and does mkimage on that (telling uboot to not uncompress) it Jul 17 18:03:31 so in the powerpc architecture is there no support for the kernel do its own decompression? Jul 17 18:04:06 not sure I understand why the big difference in what 'make uImage' does for both - the resulting powerpc uImage doesn't boot for me either which I haven't been able to figure out Jul 17 18:12:31 hi, i'm trying compile directfb on OE. i get error message: "checking for mipsel-angstrom-linux-gcc... .. C compiler default output file name". OE old version compile it Jul 17 18:13:59 i not found this error in bugtrack Jul 17 18:18:33 hi, i resolved problem with tslib package. compile tslib before directfb. Jul 17 18:18:55 fenrir I am working on getting directfb in better shape Jul 17 18:19:09 but in the moment I havent this much time Jul 17 18:19:32 and yes I fixed directfb localy to get it run with the newer tslib Jul 17 18:21:58 woglinde, ok. thanks. Jul 17 18:22:29 fenrrir hm maybee I should concentrate on one think now Jul 17 18:23:41 woglinde, ok. no problem. my problem is resolvd too. good bye Jul 17 18:24:47 woglinde: Like libftdi? :D Jul 17 18:25:25 piroko *g* Jul 17 18:29:55 woglinde: I'm not sure what that means :( Jul 17 18:31:24 prikoko concentrate on one thing Jul 17 18:31:33 instead of doing 13 things half Jul 17 18:33:26 woglinde: Ah Jul 17 19:04:10 03mickeyl 07org.oe.dev * r4c3e9561... 10/ (5 files in 5 dirs): frameworkd git add button/switches configuration Jul 17 19:06:19 sgrunt, apr and libtool are driving me crazy Jul 17 19:14:57 anybody have tried to compile subversion, on .dev branch Jul 17 19:14:58 ? Jul 17 19:17:22 diego hehe yes I am investigate this too Jul 17 19:17:51 unforantly I am not yet figured out how to disable rpath stuff Jul 17 19:17:57 for libtool2 Jul 17 19:18:07 apr now compile with warning, and apr-util failed on configure Jul 17 19:18:32 (i've just moved out libtool2* recipe ^^) Jul 17 19:19:30 yes I read your bugreport Jul 17 19:19:43 but thats not an solution Jul 17 19:20:00 but I wonder our autools stuff should add the new libtool Jul 17 19:20:21 so hehe Jul 17 19:20:33 but now I am finished the directfb stuff Jul 17 19:20:37 time now for libtool Jul 17 19:22:43 tell me if i can help you in some way Jul 17 19:24:30 diego hm dont think so Jul 17 19:25:00 diego I am consider to rewrite the configure.in Jul 17 19:25:08 for apr Jul 17 19:30:37 btw libtool error are weird: http://pastebin.com/m7aa39d3c Jul 17 19:31:32 diego hm no Jul 17 19:31:38 its clear what happen Jul 17 19:31:52 really? Jul 17 19:32:11 yes Jul 17 19:32:27 Has anyone been able to successfully compile qtopia on OE? Jul 17 19:32:29 apr ships an older libtool shell script in his sources Jul 17 19:32:46 oe with angstroen is now at libtool2 Jul 17 19:33:00 and copying the newer .m4 files into the config dir Jul 17 19:33:09 but i've compiled it without the libtool2 Jul 17 19:33:14 but somehow dont overwrite the old libtool Jul 17 19:33:17 hm oh Jul 17 19:33:18 okay Jul 17 19:33:21 let me see Jul 17 19:33:33 i have found this: 4148 Jul 17 19:35:07 diego why the heck it is using the host libtool? Jul 17 19:35:13 instead of the oe one Jul 17 19:35:25 good question Jul 17 19:41:46 woglinde, not a problem with host libtool Jul 17 19:42:06 looks like apr install wrong files... Jul 17 19:43:40 http://pastebin.com/m4b4233bc Jul 17 19:44:04 this after removing host libtool, and doing a bitbake -c rebuild apr Jul 17 19:44:24 diego your install-path is totaly broken Jul 17 19:45:13 and how i'm supposed to fix it? ^^ Jul 17 19:45:32 hm Jul 17 19:45:36 dont know yet Jul 17 19:45:44 have you read the bug 4148 Jul 17 19:45:46 first I will finish libftdi now Jul 17 19:46:03 then I will work further on apr Jul 17 19:46:16 ok, thanks anyway Jul 17 19:46:26 03woglinde2 07org.oe.dev * re07ea850... 10/ (10 files in 5 dirs): Jul 17 19:46:26 directfb: fixing stuff and update to version 1.1.1 Jul 17 19:46:26 * add Version 1.1.1, but for now let it not be the default version Jul 17 19:46:26 * add ts_lib_autools.patch so directfb builds with newer libts Jul 17 19:46:26 * fix serval .pc files so they has useful entries Jul 17 19:47:47 woglinde: :D Jul 17 19:48:10 I have 256MB of NAND flash, yet the jffs2 image that gets generated by OE only has 64MB available Jul 17 19:48:17 Is there an option to change that? Jul 17 19:50:14 yes Jul 17 19:50:16 mom Jul 17 19:50:40 ? Jul 17 19:51:19 he's asking to his mom ^^ Jul 17 19:51:25 lol Jul 17 19:51:30 (mom == moment, i suppose :D) Jul 17 19:53:02 piroko hm you have to write your own machine conf Jul 17 19:53:09 Makes sense :) Jul 17 19:53:35 or can try to set the options ROOT_FLASH_SIZE = and VOLATILE_STORAGE_SIZE = in your local.conf Jul 17 19:53:48 woglinde: Already have my own machine conf :) Jul 17 19:53:53 sorry dont know if they will be overwrite Jul 17 19:54:00 hm okay Jul 17 19:54:03 I have to set both? Jul 17 19:54:06 set to 60 and 64? Jul 17 19:54:18 hm yes Jul 17 19:55:04 Alright, thank you :) Jul 17 19:56:15 hm you set it at higher values Jul 17 19:56:16 I guess Jul 17 19:56:17 piroko: btw you find a guide here: http://bec-systems.com/oe/html/index.html Jul 17 19:56:44 diego_: yeah, just hard to find what I'm looking for sometimes Jul 17 19:57:10 hm http://bec-systems.com/oe/html/commonuse_new_machine.html isnt usefull Jul 17 20:00:17 yes it's a bit terse ^^ Jul 17 20:28:44 piroko now I found the error Jul 17 20:29:27 somehow LD_RUN_PATH is set Jul 17 20:29:39 at linking Jul 17 20:31:51 args no Jul 17 21:05:14 hrw|gone: the c7xo images are here (udev_124/ 118_ init script): http://filehost.justfreespace.com/735initrd.bin http://FileHost.JustFreeSpace.Com/594zImage Jul 17 21:07:40 piroko still there? Jul 17 21:21:12 03thebohemian 07org.oe.dev * r83329363... 10/ (1 conf/distro/chinook-compat.conf): chinook-compat: Corrected dbus-glib version number. Jul 17 21:56:41 hey CoreDump, alles i.O.? Jul 17 21:56:51 sure Jul 17 21:59:45 re flo Jul 17 22:12:39 /usr/sbin/update-modules: line 194: depmod: not found Jul 17 22:12:48 why isn't depmod getting into my console image? Jul 17 22:14:02 hehe Jul 17 22:14:07 busybox Jul 17 22:14:27 ? Jul 17 22:14:56 could be that depmod ist depmod.static Jul 17 22:14:59 there was something Jul 18 01:08:23 anyone know of a fso compilation tutorial? **** ENDING LOGGING AT Fri Jul 18 02:59:56 2008