**** BEGIN LOGGING AT Tue Jun 12 02:59:56 2007 Jun 12 03:02:55 03polyonymous 07org.oe.dev * rb267dbeb... 10/ (10 files in 3 dirs): Jun 12 03:02:55 konqueror-embedded: updated to 20070212 snapshot. Jun 12 03:02:55 Credits go to Marek Vasut of bug #1995, which is being closed. Jun 12 03:03:00 03polyonymous 07org.oe.dev * r3c3d3447... 10/ (1 conf/distro/include/preferred-opie-versions-1.2.3-pre.inc): conf/distro/include/preferred-opie-versions-1.2.3-pre.inc: bump preferred konqueror version Jun 12 03:04:00 * * OE Bug 1995 has been RESOLVED (FIXED) by polyonymous@klever.net Jun 12 03:04:02 * *  Konqueror Embedded 2007 preliminary .bb file and patches Jun 12 03:04:04 * * http://bugs.openembedded.org/show_bug.cgi?id=1995 Jun 12 03:07:01 * * OE Bug 844 has been RESOLVED (WONTFIX) by polyonymous@klever.net Jun 12 03:07:02 * *  konqueror-embedded build fix. sort of Jun 12 03:07:05 * * http://bugs.openembedded.org/show_bug.cgi?id=844 Jun 12 04:18:55 03mwester 07org.oe.dev * re3ec97ab... 10/ (5 files in 3 dirs): Unslung: /usr/bin/mkfs.ext3 - strange script to workaround an mke2fs failure Jun 12 06:06:04 koen|away: http://www.atoker.com/blog/2007/06/12/webkitgtk-is-coming/ Jun 12 06:34:06 morning all Jun 12 06:38:39 03Jake 07org.oe.dev * r38e38f27... 10/ (6 files in 2 dirs): gpsd: mark /etc/default/gpsd as CONFFILE Jun 12 07:19:15 Hello, I have problems building uboot 1.1.4 for gumstix. (http://rafb.net/p/jMS9HS60.html) basically it ends because of an EABI version mismatch. also, its built in the wrong directory (i think), in armv5te-angstrom-linux-gnueabi instead of gumstix-angstrom-linux-gnueabi. What could be the cause ? Jun 12 07:25:21 morning Jun 12 07:42:00 hi everybody Jun 12 07:42:48 hi Thomas Jun 12 08:00:14 hi hrw, do you have any ideas on my problem ? Jun 12 08:02:36 let me see Jun 12 08:03:31 can you try uboot 1.2.0? Jun 12 08:03:55 03koen 07org.oe.dev * r6c2b3981... 10/ (4 files in 3 dirs): uclibc: make a bfin arch config instead of a machine config Jun 12 08:04:03 03koen 07org.oe.dev * r4bb26bc7... 10/ (1 classes/insane.bbclass classes/siteinfo.bbclass): siteinfo,insane: blackfin is uclinux, not linux Jun 12 08:04:12 03koen 07org.oe.dev * r08e772f1... 10/ (3 files in 3 dirs): angstrom: blackfin updates Jun 12 08:04:25 no, only uboot 1.14 supports gumstix Jun 12 08:05:13 uboot 120 fail because target gumstix doesn't exist. latest cvs from gumstix uses uboot 1.1.4 Jun 12 08:05:26 and I do not know how goes 1.1.4 supports eabi if at all Jun 12 08:07:49 good morning all Jun 12 08:07:53 hi koen Jun 12 08:08:47 after all, uboot is only the bootloader, do I care if it use another abi ? Jun 12 08:09:30 koen@lieve:/media/hda4/OE/build/tmp-new/angstrom/cross/bin$ ls | grep gcc-4.1.2 Jun 12 08:09:30 avr32-angstrom-linux-uclibc-gcc-4.1.2 Jun 12 08:09:31 bfin-angstrom-uclinux-uclibc-gcc-4.1.2 Jun 12 08:09:33 :D Jun 12 08:11:03 koen: we need a way to force OABI in angstrom Jun 12 08:11:43 doesnt angstrom-legacy already forces oabi ? Jun 12 08:11:44 I knew machines which run angstrom but need uboot oabi (not only neo1973) Jun 12 08:11:51 legacy force many things Jun 12 08:13:08 wouldn't it be better to fix uboot? Jun 12 08:13:31 thesing: yes, but u-boot is not a single piece of software Jun 12 08:13:32 thesing: fixing uboot need time which need money Jun 12 08:13:50 u-boot is 30 different bootloaders glued together Jun 12 08:14:14 ~lart apt-get in OE Jun 12 08:14:15 * ibot whacks apt-get in OE upside the head Jun 12 08:14:41 I have deb only build here and bitbake does not generate rootfs Jun 12 08:20:30 ~seen lrg Jun 12 08:20:33 lrg was last seen on IRC in channel #oe, 7d 15m 22s ago, saying: 'morning koen, hrw, XorA '. Jun 12 08:25:16 the gcc that seems to be used to build uboot on gumstix is named arm-linux-uclibcgnueabi-gcc, so i'd say that uboot should support eabi Jun 12 08:29:28 some uboot boards support EABI Jun 12 08:37:00 03koen 07org.oe.dev * r02e9149a... 10/ (4 files in 3 dirs): gpe-login: update to 0.90 Jun 12 08:48:13 morning all Jun 12 08:58:31 * steliosk waved bye bye to the dual opteron.... Jun 12 09:01:32 steliosk: ;( Jun 12 09:05:14 it was a 3 day "test-drive"... and as it happens with all good things they come to an end.... Jun 12 09:06:22 it would make a nice autobuild server, could have finished close to 10 full builds in a day or so Jun 12 09:09:32 hrw : BTW have you ever measured if there is any difference in time when building 32bit OE on a 64bit target ? Jun 12 09:13:28 * koen looks at the big atmel box ups dropped off Jun 12 09:16:42 big and atmel are words that hardly ever go in the same sentence. strange Jun 12 09:22:16 koen: at91sam9263ek or atr06663ek? Jun 12 09:23:04 steliosk: my 64bit box is faster then my 32bit one Jun 12 09:23:14 steliosk: and I do not have 32bit env on my 64bit one Jun 12 09:24:03 I hope that atmel devboard is smaller then the one which I have here Jun 12 09:27:45 at91 Jun 12 09:35:27 well it looks like it makes a difference to enable multiple threads with bitbake even in a single core/cpu machine Jun 12 09:35:55 steliosk: multi threads are must have Jun 12 09:36:10 216 minutes with single thread and no parallel make Jun 12 09:36:35 185 minutes with 3 thread and -j2 Jun 12 09:37:04 these are on an AMD64 3000+ Jun 12 09:38:07 lrg: Do you mind jumping on #openmoko. I have issues with the codec and the interface to the Bluetooth PCM. Jun 12 09:38:47 steliosk: I have amd64 3200+ Jun 12 09:46:14 heh Jun 12 09:46:52 the flash tool for the at91 is a .exe Jun 12 09:47:00 normal thing Jun 12 09:47:01 03ifaistos 07org.oe.dev * rcaeab840... 10/ (1 packages/uboot/u-boot.inc): Jun 12 09:47:01 packages/uboot/u-boot.inc : Disable parallel make for u-boot as it fails for no apparent Jun 12 09:47:01 reason when its enabled Jun 12 09:47:26 koen: board which I have here is flashable from win32 too (but only it uboot fails) Jun 12 09:51:43 holtmann: now in #openmoko Jun 12 09:53:15 * koen launches parallels desktop Jun 12 10:07:35 good morning Jun 12 10:26:27 03rpurdie * r833 10/ (6 files in 4 dirs): Jun 12 10:26:27 runqueue.py: Correctly redirect stdin when forking. Jun 12 10:26:27 cooker.py: If parsing errors are found, exit, too many users miss the errors Jun 12 10:28:15 steliosk: Could you try bitbake 1.8 latest svn and see if that uclibc problem is any better please? Jun 12 10:29:40 03pH5 07org.oe.dev * r00ba67c0... 10/ (8 files in 3 dirs): Jun 12 10:29:40 libgsmd: update patches Jun 12 10:29:40 - drop plugin (accepted upstream) and qualcomm-extreply patches Jun 12 10:29:40 - start the modem-alive thread only after the interpreter is ready Jun 12 10:29:40 - add vendor plugin configurable extreply character support Jun 12 10:29:41 - add a command line option to set the initial interpreter_ready state Jun 12 10:29:43 - fix csq readings in the tihtc vendor module Jun 12 10:31:00 03rpurdie 07bitbake-1.8 * r834 10/lib/bb/runqueue.py: runqueue.py: Message fix from hrw Jun 12 10:31:06 03rpurdie * r835 10bitbake/lib/bb/runqueue.py: runqueue.py: Message fix from hrw Jun 12 11:08:56 hello Jun 12 11:09:15 good morening. Jun 12 11:09:18 and morning too Jun 12 11:13:02 * * OE Bug 2323 has been REOPENED by skliarieh@gmail.com Jun 12 11:13:04 * *  angstrom collie: suspend support Jun 12 11:13:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2323 Jun 12 11:25:34 koen, ping Jun 12 11:28:01 Crofton: pong Jun 12 11:28:28 I'm wondering how "safe" it is to put email addresses on drupal pages Jun 12 11:28:36 like, will they get harvested .... Jun 12 11:29:06 as safe as putting it on regular webpages Jun 12 11:29:09 or do they do clever things to make that difficult .... Jun 12 11:29:27 * Crofton is sick of filthy spammers Jun 12 11:30:54 RP; Will do that and let you know Jun 12 11:31:12 Crofton : How was bed time reading ? Jun 12 11:31:37 I haven't had time to loo kat the stuff you sent yet Jun 12 11:31:55 This is a long term project :) Jun 12 11:32:09 hopefully, I can figure things out, before I have to panic Jun 12 11:32:19 * Crofton is planning ahead Jun 12 11:37:39 How can I make my OE enviornment make use of a kernel patch? Jun 12 11:38:05 I changed the defconf for my device. Jun 12 11:39:24 examing the far to numerous examples in packages/linux Jun 12 11:40:01 * * OE Bug 2323 has been RESOLVED (FIXED) by bugs.openembedded.org@rolf.leggewie.biz Jun 12 11:40:01 Crofton, in the oe manual? Jun 12 11:40:03 * *  angstrom collie: suspend support Jun 12 11:40:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2323 Jun 12 11:40:08 no Jun 12 11:40:23 org.oe.dev/packages/linux Jun 12 11:40:33 * Crofton does not read the manual enough Jun 12 11:42:28 Crofton, what examples do you mean there? Jun 12 11:42:45 there are many kernel bbfiles with patches Jun 12 11:45:02 re Jun 12 11:46:14 Crofton, I applied that patch manually to the defconf file Jun 12 11:46:24 it is already modified Jun 12 11:46:34 atmel does have nice boxes: http://www.flickr.com/photos/koenkooi/542193498/ Jun 12 11:46:52 I am confused, what is the problem? Jun 12 11:47:14 there is some avr32 chatter on the u-bbot list Jun 12 11:47:49 Crofton : This is for you then :) http://www.xilinx.com/ise/optional_prod/planahead.htm Jun 12 11:48:18 heh Jun 12 11:48:27 koen: how big (in cm) is at91sam board? Jun 12 11:49:30 13x20 Jun 12 11:49:41 approx a5 sized Jun 12 11:49:54 metrics Jun 12 11:50:28 39 squibles by 96 pinkies for the imperial people Jun 12 11:50:44 guys, has anyone tried building apt-native? I had a problem with oe_libinstall, it scrambled the library name incorrectly; I patched the run.do_stage_base script to test and it worked; but I wonder if that's a general problem that went unnoticed, or if it is something about my system Jun 12 11:51:44 Jin^eLD: i did Poky build with apt-native today Jun 12 11:51:52 Jin^eLD: so it works atleast in Poky Jun 12 11:51:59 doh... Jun 12 11:53:04 Crofton, my problem is that I can't compile the gspcav1 kernel module because the kernel was lacking some V4l configuration entries which I added to packages/linux/linux-rp-2.6.20/defconf-akita with CoreDump|afk's help. Now I don't know how to proceed in order to build the module. Jun 12 11:54:00 so you updated the defconfig and rebuilt the kernel Jun 12 11:54:15 and the module is built from another bb file? Jun 12 11:56:04 hrw, could you take a quick look please: http://pastebin.ca/561259 Jun 12 11:59:48 $libname is 'libapt-inst-libc6.5-6'? Jun 12 12:03:37 yes Jun 12 12:04:26 but after sedding the -6 gets stripped away Jun 12 12:05:06 and libname gets changed, but is used later for installation Jun 12 12:05:18 so then the installation fails, because the name is wrong Jun 12 12:06:22 hard to tell for me as I never touched that part of code Jun 12 12:07:46 whom should I bother? Jun 12 12:08:45 I guess I'll file a bugreport.. don't think that I am doing something wrong there Jun 12 12:17:23 Crofton, yes I made the zImage.bin but I tried to make the module from the same bb file as before. Is that the mistake? Jun 12 12:17:43 I'm not sure Jun 12 12:17:47 sorry Jun 12 12:22:01 * * OE Bug 2473 has been created by slapinid@gmail.com Jun 12 12:22:03 * * libgpevtype: wrong depends, fix inside Jun 12 12:22:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2473 Jun 12 12:24:18 he, the bug already exists Jun 12 12:24:26 2017 Jun 12 12:36:44 !oebug 2017 Jun 12 12:36:45 * * Bug 2017, Status: NEW, Created: 2007-03-18 03:12 Jun 12 12:36:46 * * c.romain(AT)laposte.net: apt-0.6.46.2 does not build Jun 12 12:36:47 * * http://bugs.openembedded.org/show_bug.cgi?id=2017 Jun 12 12:37:42 if a makefile has not install: target ,, there will be no package created , correc t? Jun 12 12:38:35 rob_w|mis: that depends on your recipe?! Jun 12 12:39:33 zecke, i am creatiung a new recipe for app which has no make install target .. i added do_install() into my bb and now i need it to create a ipkg package ? Jun 12 12:41:14 rob_w|mis: do_install() { install by hand? } Jun 12 12:42:30 yes i did that and it puts that data into its WORKDIR/install/appname/ Jun 12 12:43:32 oh no i made it not install i made it copy to $(D)/mypath and it ends up in WORKDIR/image/mypath Jun 12 12:43:43 rob_w|mis: just check the manual... Jun 12 12:43:50 ok Jun 12 12:44:42 what use is a manual if one gets ZERO feedback on it... Jun 12 12:45:01 * * OE Bug 2474 has been created by u.luckas@road.de Jun 12 12:45:03 * * linux-hotplug: mmc.agent shoud be block.agent Jun 12 12:45:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2474 Jun 12 12:45:13 zecke, ideology Jun 12 12:45:37 RP: #2474 is quite interesting. it might even help with the boot time? Jun 12 12:46:07 rob_w|mis: well, we could have a lot of pages "This page is left blank" Jun 12 12:46:19 heh Jun 12 12:53:59 btw, I get ERROR: pkg-native_1.13.22.bb depends on itself (eventually), does anyone have the same problem? Jun 12 13:05:56 uhm.. did anyone succeed in building dpkg? for me configure complains that it can not run program when cross copiling, when doing the va_copy check Jun 12 13:11:51 03Laibsch 07org.oe.dev * r2820300e... 10/ (1 packages/qt/qt-x11-free-common.inc): qt-x11-free: move some debug files into proper package Jun 12 13:12:00 03Laibsch 07org.oe.dev * r2639e563... 10/ (1 packages/gsoap/gsoap_2.7.7.bb): gsoap: remove erroneous PACKAGES-line Jun 12 13:12:24 zecke: Yes, it could Jun 12 13:16:04 but we do not have linux-hotplug in our 2.6 images Jun 12 13:23:01 hrw: we don't use the mmc.agent anymore? Jun 12 13:24:50 morning Jun 12 13:26:20 mmc.agent is part of linux-hotplug iirc Jun 12 13:27:16 hmm openssl doenst show a linux-gnueabi-armeb switch Jun 12 13:27:40 it is Jun 12 13:27:40 rob_w|mis: which version? the one build by angstrom is likely to have it patched Jun 12 13:27:54 hrw is right, we don't use hotplug at all (I removed it from poky) Jun 12 13:28:03 itbake -b ../org.openembedded.dev/packages/openssl/openssl_0.9.7g.bb Jun 12 13:28:14 RP: how are mmc cards mounted? Jun 12 13:28:24 rob_w|mis: is that the version used by angstrom? Jun 12 13:28:31 i check Jun 12 13:28:36 03hrw 07org.oe.dev * r97b9d760... 10/ (1 packages/udev/udev_092.bb): udev: make only udev RPROVIDE hotplug not all packages (from Poky) Jun 12 13:29:09 zecke: mount.sh script in udev (nasty hack really) Jun 12 13:29:26 RP: without any sleep? Jun 12 13:29:37 hm. close as INVALID? WONTFIX? Jun 12 13:29:44 zecke: Why would it need to sleep? Jun 12 13:30:29 hrw: well, it is valid as long as we have linux-hotplug in the tree Jun 12 13:30:42 then left as NEW Jun 12 13:31:07 You might want linux-hotplug with old kernels... Jun 12 13:31:12 zecke yep was the version Jun 12 13:32:41 rob_w|mis: so what is your question? ever considered to grep eabi in the openssl dir? Jun 12 13:33:33 XorA|holiday: http://www.cnn.com/interactive/entertainment/0705/gallery.lego.art/frameset.exclude.html Jun 12 13:33:47 it only lists linux-gnueabi-arm but not linux-gnueabi-armeb Jun 12 13:34:17 rob_w|mis: ah :) Jun 12 13:34:41 there is patch by oe to make gnueabi-arm -- so will adapt this Jun 12 13:36:03 http://news.yahoo.com/s/nm/20070612/tc_nm/htc_phone_usa_dc_2;_ylt=Av0IBLihRoJ5iJwvfFUtK6EE1vAI Jun 12 13:38:48 zecke, ah the adaption works .. i will file a bug for this change Jun 12 13:42:34 hi Jun 12 13:42:35 uhh waht is that ? ERROR: Cannot satisfy the following dependencies for libcrypto0.9.7: Jun 12 13:42:35 libc6 (>= 2.5) Jun 12 14:05:01 * * OE Bug 2475 has been created by slapinid@gmail.com Jun 12 14:05:03 * * gpe-announce build failure (due to libsoundgen staging) Jun 12 14:05:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2475 Jun 12 14:50:37 hrw: http://www.flickr.com/photos/koenkooi/542494461/ Jun 12 14:54:46 thx Jun 12 14:54:53 koen: vga or qvga screen? Jun 12 14:55:30 I suspect qgva Jun 12 14:55:36 Hello, in which step do you apply patchs ? (deploy, install, configure ...) Jun 12 14:55:43 I'm missing the datacard to flash the board... Jun 12 14:57:23 koen: manual say 3.5" 1/4 VGA - so qvga Jun 12 14:58:05 The datacard should arrive tomorrow if all goes well Jun 12 14:58:40 what datacard? Jun 12 14:58:41 then I can look at adding the kernel to OE and integrate the recipes atmel made Jun 12 14:58:59 hrw: a flash chip that goes in the sd slot Jun 12 14:59:05 it's needed to flash the board Jun 12 14:59:20 * koen would have loaded uboot into ram and used that Jun 12 14:59:45 ah Jun 12 15:00:15 by default no sign of life at all? Jun 12 15:00:44 at91rm9200 has minimal loader in cpu chip iirc - but you have at91sam Jun 12 15:01:21 the minimal loader talks over usb and there is an apps to use that Jun 12 15:01:24 * wick7xx is away: sl33p & Jun 12 15:01:40 but all their files and scripts need that flashcard.... Jun 12 15:02:32 ;( Jun 12 15:02:51 ~seen ph5 Jun 12 15:02:56 ph5 was last seen on IRC in channel #htc-linux, 2h 3m 43s ago, saying: 'bbl'. Jun 12 15:03:23 theBrave: do_patch ;-) Jun 12 15:05:02 hi koen hrw Jun 12 15:05:11 hey chouimat Jun 12 15:05:32 Laibsch, but there is no log.do_patch.xxxx in my tmp directory :( does that means that it doesn't patch ? Jun 12 15:05:44 patch do not make logs Jun 12 15:05:59 hrw: when I asked "is it possible to flash from ram" they responded with "we will send you a flashcard" :) Jun 12 15:06:03 theBrave: cd ${S} && quilt applied == log.do_patch Jun 12 15:06:18 koen: not tried with sd/mmc card instead? Jun 12 15:09:46 I tried, but dataflash seems to be raw NAND instead of a block device Jun 12 15:10:00 or raw NOR Jun 12 15:10:24 rather nor Jun 12 15:10:49 thanks hrw, now oe build but linux doesn't boot Jun 12 15:11:07 or more exactly make lot of errors booting Jun 12 15:11:08 theBrave: so you need to fix kernel Jun 12 15:11:15 yes :( Jun 12 15:11:18 theBrave: decide - boot or lot of errors Jun 12 15:11:22 and what errors Jun 12 15:11:33 i just hope a patch wasn't applied Jun 12 15:12:53 let me see Jun 12 15:15:32 re mickey_fic Jun 12 15:15:52 hi hrw Jun 12 15:16:04 * mickey_fic building openmoko-devel-image, DISTRO=angstrom-2007.1 Jun 12 15:16:09 hi mickey_fic Jun 12 15:16:29 hi chouimat Jun 12 15:16:50 ... Jun 12 15:16:55 re Jun 12 15:17:52 hey mickey_fic Jun 12 15:17:55 how do i install ipks from qemu, will adding the ipks to the rootf tarball and then using the application installer to install it, work Jun 12 15:18:03 ? Jun 12 15:18:12 oops...wrong channel Jun 12 15:18:13 :D Jun 12 15:19:01 koen: Dataflash is dataflash - SPI accessed Jun 12 15:19:05 * mickey_fic just learned that emotion has a gstreamer backend Jun 12 15:19:11 florian: hey Jun 12 15:19:24 yo zecke , hey florian Jun 12 15:19:54 mickey_fic: xine is evil and must die Jun 12 15:20:07 * chouimat kicks koen Jun 12 15:20:24 koen: really? why do you think so? Jun 12 15:20:32 hi zecke Jun 12 15:20:54 mickey_fic: bad design, low quality code, etc Jun 12 15:21:10 koen but it works :) Jun 12 15:21:19 chouimat: actually, it doesn't Jun 12 15:21:30 mickey_fic: how are tests coming along? Jun 12 15:21:32 koen it works enough for me ... Jun 12 15:21:40 xine plays sound at double speed on almost every ARM device I have Jun 12 15:22:02 koen I don't use it on ARM Jun 12 15:22:17 chouimat: I don't use it anymore Jun 12 15:22:57 koen the arm devices I have have no sound device so ... Jun 12 15:22:58 florian: no tests yet... OpenMoko is busy with kicking FIC's ass :) we'll test on weekend Jun 12 15:23:26 actually FIC is a pretty open minded company. It's exciting Jun 12 15:23:28 florian: hey, could you do me a favor and mail joey and ask how active ffis.de is? Jun 12 15:24:05 florian: stillin bejing? Jun 12 15:24:06 mickey_fic: hehe okay, sounds liek some progress at least ;) Jun 12 15:24:14 absolutely Jun 12 15:24:20 lots of changes Jun 12 15:24:21 for the good Jun 12 15:24:23 zecke: right, that's a good idea Jun 12 15:24:55 koen: they have all broken drivers :) Jun 12 15:24:59 * chouimat bangs his head on the desk Jun 12 15:25:19 koen: check the Opie xine.conf to enable soft delay calculation Jun 12 15:25:36 mickey_fic: nope, returned saturday and started to enjoy some time at heome :-) Jun 12 15:25:38 zecke: soft delay didn't help either Jun 12 15:25:39 eh home Jun 12 15:25:53 in contrast to nils ;) Jun 12 15:30:07 * chouimat should stop reading openwrt forums Jun 12 15:30:53 mickey_fic: when will the new products be announced? Jun 12 15:34:18 koen: no idea. that's a bit out of scope until our team restructuring has been done Jun 12 15:35:42 03florian 07org.oe.dev * r7d055c0a... 10/ (4 files in 3 dirs): gtk+: Add patch to fix file selector segfault contributed by HEXaBYTE. Jun 12 15:41:55 Is there ever going to be a need for empty directories in OE metadata? I dropped a bunch of them locally and would like to know if this is safe to commit (as I assume) Jun 12 15:44:13 brb Jun 12 15:52:03 zecke: mail sent Jun 12 15:52:30 koen: That patch should bring you closer to a stable release :-) Jun 12 15:54:13 nice Jun 12 15:56:37 Laibsch: feel free to clean up that way Jun 12 15:56:38 thanks Jun 12 15:57:01 mickey_fic: Done. Thanks for the feedback Jun 12 15:57:12 Laibsch: they are leftovers from the time mtn didn't track directories Jun 12 15:57:45 koen: Well, they are no more ;-) Jun 12 16:02:47 cute, the sam9263 even has some 2d accell Jun 12 16:13:17 bye Jun 12 16:16:37 hi Jun 12 16:16:44 03Laibsch 07org.oe.dev * r87fab946... 10/ (46 files in 24 dirs): remove a bunch of empty directories Jun 12 16:26:40 ~seen thesing Jun 12 16:26:50 thesing was last seen on IRC in channel #oe, 8h 13m 42s ago, saying: 'wouldn't it be better to fix uboot?'. Jun 12 16:46:00 i cant understand why 2 object files inside one package will have difffern eabi versions .. Jun 12 16:46:14 can i disable ldd checking for those and go without with some luck ? Jun 12 16:48:35 who is the eabi guru here ? Jun 12 17:10:19 hey guys, I have some dud parts of the flash in my z, does it have some way to manage this and not use these areas or will this cause issues with even a newly flashed image? Jun 12 17:22:22 uhoh madwifi doesnt know -march=armv5te Jun 12 17:34:26 * rob_w hails the gods of madwifi , grant me a hal for armv5te pleease Jun 12 17:34:51 rob_w: file a bug, and be polite Jun 12 17:35:04 rob_w: and like I said, ask in #oe and #angstrom, they might be able to help Jun 12 17:35:06 Zero_Chaos, i do what ever you want Jun 12 17:35:51 rob_w: #angstrom uses eabi so you may get good luck there. Are you using OE and failing to build madwifi? Jun 12 17:36:55 yeah i see my chance now in compiling angstrom with oabi and see if xcale-hal blinks Jun 12 17:37:11 rob_w: that's icky Jun 12 17:37:22 what will i loose with eabi ? quite some right Jun 12 17:37:28 so I hear Jun 12 17:37:36 I'm not the eabi guru you seek ;-) Jun 12 17:37:55 well enough to open my eyes in some way .. Jun 12 17:38:28 now i really understand the error i get fully .. eabi version 0 is oabi ,, right Jun 12 17:38:47 * Zero_Chaos doesn't know much about this stuff Jun 12 17:40:39 rob_w: poke the madwifi maintainer in OE about their crazy whackelf patch Jun 12 17:41:16 i.e. ask on the oe-devel mailinglist Jun 12 17:42:27 iam a very bad emailer , but feels this time i ve no chance Jun 12 17:42:47 koen, going oabi , is this some try worth ? Jun 12 17:42:55 no Jun 12 17:43:11 03koen 07org.oe.dev * re65ca569... 10/ (3 files in 3 dirs): linux nokia800: add defconfig for 770 Jun 12 17:43:16 03koen 07org.oe.dev * r76e34db5... 10/ (4 files in 4 dirs): linux 2.6.21: elaborate support for at91 and avr32 Jun 12 17:43:20 03nicolas.ferre 07org.oe.dev * rad05460d... 10/ (1 conf/machine/at91sam9263ek.conf): at91sam9263ek: add support for the at91sam9263-ek evaluation board Jun 12 17:44:27 is there a shot wiki on what the differences / effects are ? Jun 12 17:47:01 * * OE Bug 2476 has been created by bugs.openembedded.org@rolf.leggewie.biz Jun 12 17:47:03 * * spitz default opie image contains unneeded bluetooth support Jun 12 17:47:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2476 Jun 12 17:50:00 * koen wonders what is wrong with people Jun 12 17:50:13 eabi is http://wiki.debian.org/ArmEabiPort Jun 12 17:50:19 ~eabi Jun 12 17:50:21 it has been said that eabi is the new "Embedded" ABI by ARM ltd which allow mixing soft and hardfloat code. Not yet supported for armv4l cpu's like StrongARM. Embedded distro with EABI: Angstrom. More info: http://wiki.debian.org/ArmEabiPort and http://www.angstrom-distribution.org Jun 12 17:50:41 first they demand wifi support for zaurusses, even if the device doesn't have it by default, and later they complain about bluetooth support being present... Jun 12 17:50:47 koen, you have only just figured out most people are wrong Jun 12 17:52:01 koen: Need an update in logic? Jun 12 17:52:13 people is not a single entity Jun 12 17:52:19 That is just for starters Jun 12 17:52:36 you can't have it both ways Jun 12 17:52:49 either wifi + bluetooth support or no wifi+bluetooth support Jun 12 17:53:00 one depends on the other? Jun 12 17:53:21 I don't think so Jun 12 17:53:33 with 128 Megs of flash, I'd rather have both Jun 12 17:53:46 Zero_Chaos: You can install whatever you want Jun 12 17:53:55 And angstrom is not installed to flash anyways Jun 12 17:53:56 Laibsch: I know I can, I'm just saying Jun 12 17:54:02 Laibsch: it is on my system Jun 12 17:54:44 koen: wifi depends on bt? Jun 12 17:55:52 bt support is exactly the same issue as wifi support Jun 12 17:56:22 if you want to exclude one you'll have to exclude the other as well if you want to be consistent Jun 12 17:56:22 OK, I agree with you in that both are optional except such strange devices like the tosa when considering the Z world Jun 12 17:56:34 (I don't have experience with other devices) Jun 12 17:57:27 well, where it differs I think is that it would be nice to have some kind of network connectivity built in. IOW not necessary to install packages via CF or SD or something else Jun 12 17:57:45 Many people have wifi and thus IMHO it makes sense to have wifi. Jun 12 17:58:01 many people have bluetooth as well for network access Jun 12 17:58:11 Really? Jun 12 17:58:15 consumes far less power as wifi and is fast enough Jun 12 17:58:18 I don't know a single one Jun 12 17:58:36 yes you do :) Jun 12 17:59:30 that depends on the semantics you read into the sentence Jun 12 17:59:33 But I got your hint Jun 12 17:59:50 That is one. One blessed with probably any toy available to mankind Jun 12 17:59:57 That does not count too much ;-) Jun 12 18:00:07 :D Jun 12 18:00:11 Let's rephrase Jun 12 18:00:28 I don't know a single person with BT card but without wifi card Jun 12 18:00:41 I know quite a few with wifi card but without bt card Jun 12 18:00:47 How's that? Jun 12 18:01:23 (following the "first internet connection for update and install anything you like"-argument) Jun 12 18:05:10 koen: Can you tell me how I can make the decision of whether opie-image includes task-bluetooth depend on whether MACHINE_FEATURES has bluetooth or not? Jun 12 18:05:23 koen: I'd like to prepare a patch to discuss on the ML Jun 12 18:06:02 all 'standard' angstrom images depend on task-base-extended, and that is not going to change Jun 12 18:06:26 patches to exclude task-opie-bluetooth will be accepted Jun 12 18:07:09 you are welcome to create an image that uses task-base instead of task-base-extended Jun 12 18:07:23 OK, can you answer the question? Just a hint should be enough. Jun 12 18:07:33 task-base-extended includes the bt dependency? Jun 12 18:07:39 yes Jun 12 18:07:44 opie-image includes it as well Jun 12 18:08:12 task-base-extended includes wifi and bt if the device has usb|cf|pccard Jun 12 18:08:22 I'd like to learn how to do this for opie-image although it won't have an effect. Jun 12 18:08:34 how do I realize this conditional DEPENDS? Jun 12 18:09:45 have a look at task-base.bb or angstrom-x11-image.bb Jun 12 18:10:14 OK Jun 12 18:53:43 hi folks Jun 12 18:54:30 why could busybox init say: Jun 12 18:54:51 Bad inittab entry: id:3:initdefault: Jun 12 18:54:52 ? Jun 12 19:17:28 koen: to build angstrom uclib do both ANGSTROM_MODE and TARGET_OS need to be set? Jun 12 19:17:48 just Angstrom_mode I think Jun 12 19:19:08 only angstrom-mode Jun 12 19:19:20 that triggers python magic to set target os Jun 12 19:21:01 * * OE Bug 2477 has been created by bugs.openembedded.org@rolf.leggewie.biz Jun 12 19:21:03 * * RFE: Better wifi support Jun 12 19:21:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2477 Jun 12 19:24:47 angstrom-console-image doesn't seem to be populating&creating /bin and other in the rootfs Jun 12 19:25:28 (glibc build) Jun 12 19:25:55 I don't have anstrom-console-image in my tree anymore Jun 12 19:26:07 ljp: angstrom-minimal-image? Jun 12 19:27:41 hmm Jun 12 19:28:53 hi ljp Jun 12 19:31:49 no angstromg-minimal-image in my tree Jun 12 19:32:01 hi chouimat Jun 12 19:33:55 ljp : You need ANGSTROM_MODE only Jun 12 19:35:01 * * OE Bug 2478 has been created by bugs.openembedded.org@rolf.leggewie.biz Jun 12 19:35:03 * * module packages should call depmod Jun 12 19:35:04 * * http://bugs.openembedded.org/show_bug.cgi?id=2478 Jun 12 19:35:44 * polyonymous agrees with cdbot2 and Laibsch Jun 12 19:36:22 * Zero_Chaos thinks added a bug like that is completely pointless Jun 12 19:36:49 Zero_Chaos, that is, if you can fix it? Jun 12 19:36:58 it's important, but it needs to be opened for each module recipe Jun 12 19:37:09 well Jun 12 19:37:13 polyonymous: well I was just about to fix the modules I'm responsible for. Jun 12 19:37:16 anyone familiar with busybox? Jun 12 19:38:01 Zero_Chaos, well, I don't know about kernel bbs, I thought all kernel modules (except for a few external ones) are built in one bb? Jun 12 19:38:09 Zero_Chaos: You are welcome to give me the patch and I'll apply it. Jun 12 19:38:14 This is anything but pointless Jun 12 19:38:24 If nothing else it serves as a reminder for me. Jun 12 19:38:32 Since I don't have the time to look into it now. Jun 12 19:38:38 I also don't think it's pointless. Jun 12 19:38:40 Laibsch: patch Jun 12 19:38:43 Laibsch: pkg_postinst() { depmod } Jun 12 19:38:52 Laibsch: now add it to every module ;-) Jun 12 19:39:03 Zero_Chaos: See, I think about something more elaborate Jun 12 19:39:11 And as I said I don't have the time now Jun 12 19:39:22 But I do have the time for a short bug report so it does not get forgotten Jun 12 19:39:33 Zero_Chaos: But you don't have to like Jun 12 19:39:34 it Jun 12 19:39:39 I don't care ;-) Jun 12 19:39:42 I think it's anyway easier for those who are going to toy with kernel bbs, anyway. Jun 12 19:40:07 Laibsch: I am just saying seems like a better note left for you on your desktop than in the bug tracker Jun 12 19:40:09 Yes, I am sure there can be a general solution. To include that line for all module packages Jun 12 19:40:28 Zero_Chaos: You wanna see the mess on my desktop? ;-) Jun 12 19:40:35 Laibsch: join the club Jun 12 19:40:46 Zero_Chaos, wrong. If someone is going to mess with kernel build and also knows of this bug he might fix it while he's at it. Jun 12 19:41:21 this is a pointless debate that I will not continue, it doesn't really matter Jun 12 20:01:26 re Jun 12 20:03:28 polyonymous: wrt to your broken wifi card, contact gerwin and inquire if he needs the OE devices wifi card Jun 12 20:03:32 http://www.openembedded.org/wiki/ProjectDevices Jun 12 20:03:47 I think it would be fair if you used it for further dev work Jun 12 20:04:05 Maybe you can discuss the issue on the ML as well Jun 12 20:04:22 Laibsch, but do I need it? Jun 12 20:08:30 I've gone to read the page, though and I think "Project will lend the devices for a limiting time" should be a "limited time" :) Jun 12 20:19:11 * koen adjusts #2478 Jun 12 20:20:01 * * OE Bug 1006 has been RESOLVED (FIXED) by bugs.openembedded.org@rolf.leggewie.biz Jun 12 20:20:03 * *  tmp sticky bit Jun 12 20:20:05 * * http://bugs.openembedded.org/show_bug.cgi?id=1006 Jun 12 20:24:01 * * OE Bug 1369 has been RESOLVED (WONTFIX) by bugs.openembedded.org@rolf.leggewie.biz Jun 12 20:24:03 * *  SD not appearing in /dev OZ-3.5.4.2-rc2 Jun 12 20:24:05 * * http://bugs.openembedded.org/show_bug.cgi?id=1369 Jun 12 20:53:08 are there any prism2 usb drivers in the oe repository yet? Jun 12 20:53:24 I think so Jun 12 20:53:27 linux-wlan-ng Jun 12 20:53:33 builds for tosa I think Jun 12 20:55:07 thanks Jun 12 21:00:41 someone should translate http://www.trusted code.de/paper/OpenEmbedded.pdf into en Jun 12 21:16:42 what happened to hostapd? Jun 12 21:19:03 dcordes: should still be there Jun 12 21:19:38 Zero_Chaos, it seems broken for me Jun 12 21:26:37 does anyone know if get fetcher is broken ? Jun 12 21:26:43 s/get/git Jun 12 21:27:31 doesn't seem to be Jun 12 21:49:52 morning Jun 12 21:50:01 good one Jun 12 21:52:58 hmm.. reload icons consistently locks up opie... Jun 12 22:04:39 polyonymous: for me, just took ~10s of no feedback ;-) Jun 12 22:05:28 psokolovsky_, hmm.. I don't think I have that many icons to take a long while :) Anyway, this still must be my system, I need a fresh install to see if it's a bug. Jun 12 22:07:17 XorA|holiday, could you send me the link to the "usb-power-limit patched" kernel again in case you still have it online? I accidently reflashed and did not back it up :( Jun 12 22:15:34 * v8jlene is back Jun 12 22:15:36 morning Jun 12 22:15:43 hi v8jlene Jun 12 22:21:13 cu all Jun 12 22:33:45 Hmm, netbase is broken... I suspect that "deleting empty directories" commit Jun 12 22:38:19 v8jlene : hi ! Jun 12 22:38:38 steliosk: hi Jun 12 22:38:56 did you have any problems building uclibc recently ? Jun 12 22:41:13 My last uclibc build was on 2007-05-30, so haven't tried a fresh build for about two weeks now Jun 12 22:44:45 v8jlene : try if you have some spare time and let me know. The order of execution of install and staging has changed and it breaks uclibc. We found it foing some work on avr32 but i assume it must be the same on all architectures Jun 12 22:46:11 bye Jun 12 22:46:25 steliosk: Just started a rebuild now. It's funny you mention the order though - I've had to add "oe_runmake ... dev_Install" to do_install for previous uclibc versions. Jun 12 22:48:27 * v8jlene was correct, deleting those empty directories did break netbase Jun 12 22:57:16 has anyone tried a native qt build lately, ipkg says qmake should be part of qt4-x11-free-dev, and live in /us/bin, but i just force reinstalled and qmake is still missing Jun 12 23:03:22 I don't think we have qmake packaged, we only have qmake2-native Jun 12 23:03:32 and qt4-x11-free has no qmake Jun 12 23:04:42 so, basically, if by native qt build you mean building on oe device that wouldn't work, methinks. Jun 12 23:08:25 i was hoping to have functional build environments for when i was out and about, guess i'll have to live with just gtk builds on the run Jun 12 23:15:10 or contribute :) Jun 12 23:25:51 zecke: I got a reply from Joey... ffis is active and alive. Jun 12 23:27:54 florian: awesome, the question is can OE be an associated project? Jun 12 23:28:03 is that too much work for them do we have the freedom? Jun 12 23:28:28 zecke: that's what i was going to ask next :-) Jun 12 23:28:46 but before this i'll try to get a little bit of sleep. Jun 12 23:28:55 * zecke heads home as well Jun 12 23:29:10 where home is somewhere :} Jun 12 23:29:26 later Jun 12 23:29:28 sleep well Jun 12 23:29:30 uh Jun 12 23:29:35 zecke: good night Jun 12 23:45:55 polyonymous, ping Jun 12 23:46:00 pong Jun 12 23:47:11 polyonymous, hi, is STAGING_DIR correct variable for OE receipt? Jun 12 23:48:14 Marex, (a) why me? (b) generally yes, but it depends on the usage (c) I think you mean "recipe" :) Jun 12 23:48:45 oops ... right Jun 12 23:49:06 polyonymous, I guess because you're one of OE core team? Jun 12 23:49:32 I don't know what part of the team is 'core', but anyway, I'm not a .bb-king (yet). Jun 12 23:49:47 There are people around who know definitely better. Jun 12 23:50:43 but they wont help as I noticed today Jun 12 23:51:32 Well, they may or may not, you never know :) Jun 12 23:51:55 well ... right :) Jun 12 23:53:24 * Marex has still weird feeling about this qtopia-free Jun 12 23:53:55 Marex, what are you trying to do with it? :) Just wondering :) Jun 12 23:54:14 get it to compile correctly within OE Jun 12 23:54:47 it can be useful as startpoint for opie ressurection Jun 12 23:54:53 Marex: STAGING_DIR is the base directory for staged files, there's separate defines for include and library directories. Jun 12 23:54:53 ah, just compile. And what would you do once it compiles? :) Jun 12 23:55:16 polyonymous, Ill try it of course :) Jun 12 23:55:38 Marex, try with what? Jun 12 23:55:42 I think it's just a core? Jun 12 23:55:43 v8jlene, hmm ... doesnt look like STAGING_DIR works ... but STAGING_LIBDIR/.. fixed the issue :b Jun 12 23:55:48 (not that I know) Jun 12 23:55:53 polyonymous, qtopia-free isnt just a core Jun 12 23:55:55 :) Jun 12 23:56:15 ah Jun 12 23:56:25 Marex: The staging directories are listed here: http://www.openembedded.org/user-manual&dpage=directories_staging Jun 12 23:56:50 sorry, I was thinking about qtopiacore, of course. Jun 12 23:57:01 Marex: Yeah, the lack of a directory pointing to STAGING_LIBDIR/.. is rather annoying. Jun 12 23:57:08 I wish they hadn't renamed Qt/E Jun 12 23:57:10 polyonymous, I know, it's confusing Jun 12 23:57:14 a directory? I mean a variable Jun 12 23:58:47 bipolar, hrw said you're running something-kde on qt4? Jun 12 23:59:40 polyonymous: hehe.. yeah. I'm building KDE on Qtopia-Core. Jun 13 00:00:00 polyonymous: at least, I'm trying to. Not a lot of time. :P Jun 13 00:00:17 bipolar, and how is it going? I'm dreaming of having kdepim(pi) on qt4-x11 Jun 13 00:00:49 err... I'm going the other way around. I want to put kdepim on Qt/e :) Jun 13 00:01:10 Ah, no big progress? I've taken a look at the new kde build system and to be true I don't fancy the idea of toying with it unless I'm at least half-sure I will get some results :) Jun 13 00:01:27 well, I put (pi) in parentheses :) Jun 13 00:01:46 03lenehan 07org.oe.dev * r0a38e756... 10/ (9 files in 4 dirs): Jun 13 00:01:46 poptop 1.3.4: Add 1.3.4 with some changes: Jun 13 00:01:46 * Mark config files as such Jun 13 00:01:46 * Add an init script Jun 13 00:01:46 * Make bcrelay and the wtmp plugin available via seperate packages Jun 13 00:01:46 * Make the -doc package available. Jun 13 00:03:42 * Marex became pretty evil :B Jun 13 00:04:31 but looking at all those pims I'm really thinking that maybe I should take a week off of everything and hack together my own :) Jun 13 00:04:41 * Marex nearly made qtopia-free obey his orders Jun 13 00:04:51 polyonymous: I'm still not used to cmake. Jun 13 00:05:05 bipolar, that's a part of nightmare, indeed :) Jun 13 00:07:47 polyonymous: so far, it's been really simple to get kdebase to *build*. it's not stable though. Jun 13 00:07:55 at least, when the code is buildable :P Jun 13 00:08:27 what are kdepim's deps? Jun 13 00:09:49 kdelib, kdepimlib, kdebase Jun 13 00:10:01 Which I've built Jun 13 00:10:12 but kdelibs needs work to make it stable. Jun 13 00:10:25 there are too many assumptions in the code. Jun 13 00:10:35 and they all depend on boost too? Jun 13 00:10:59 er.... I dunno. maybe boost is a requirement of kdepim. Jun 13 00:11:04 I havn't needed it yet Jun 13 00:14:02 good to know you haven't :) Jun 13 00:15:27 does anyone know of a good way to get cvs stats? Like a graph showing everyone who's commited code and the # of lines, etc.... Jun 13 00:18:43 bipolar, find, cat, grep, awk, wc? :b Jun 13 00:19:04 and then something like rrdtool Jun 13 00:21:47 bipolar: Something like this: http://statcvs.sourceforge.net/ Jun 13 00:21:49 gah... ugly Jun 13 00:22:22 v8jlene: hmm... interesting Jun 13 00:22:35 v8jlene: thanks. I'll see if that works. Jun 13 00:23:34 and I personally agree with Linus on cvs :) Jun 13 00:23:58 he certionly knows what he's talking about. Jun 13 00:24:10 I didn't get a chance to watch the whole video though. Jun 13 00:24:13 I should do that now Jun 13 00:24:20 We all know what he's talking about :) Jun 13 00:24:40 Well, I wouldn't say I learned much from the video, but since it's well-know it's a nice thing to refer to :) Jun 13 01:29:11 03lenehan 07org.oe.documentation * rb56888d6... 10/ (1 usermanual/chapters/usage.xml): usermanual/usage: Add some details on what configuration files exist in OE. Jun 13 01:56:26 03lenehan 07org.oe.documentation * r10f66fad... 10/ (8 files in 3 dirs): Jun 13 01:56:26 usermanual: Split each of the original chapters out of usermanual.xml into a Jun 13 01:56:26 seperate file per chapter in the chapters subdirectory. In preparation for Jun 13 01:56:26 cleaning up some of the overlap between chapters and to make life a bit Jun 13 01:56:26 easier for editing and merging. Jun 13 01:56:28 03lenehan 07org.oe.documentation * r8f846078... 10/ (1 usermanual/chapters/usage.xml): Jun 13 01:56:31 usermanual usage: Add the missing section_id on the configuration chapter so Jun 13 01:56:33 it gets a filename that'll remain constant. Jun 13 02:25:00 kergoth: I thought you might like to see this: http://www.bipolardevelopment.com/opiestats/ **** ENDING LOGGING AT Wed Jun 13 02:59:57 2007