**** BEGIN LOGGING AT Tue Jun 05 02:59:57 2007 Jun 05 03:07:01 * * OE Bug 2414 has been created by summatusmentis@gmail.com Jun 05 03:07:03 * * mb-applet-wireless segfaults Jun 05 03:07:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2414 Jun 05 03:33:31 03coredump2 07org.oe.dev * rdd7855d9... 10/ (3 files in 3 dirs): netbase: Add verbose examples for WEP and WPA encrytion to /etc/network/interfaces Jun 05 03:33:35 03coredump2 07org.oe.dev * rad72560a... 10/ (4 files in 4 dirs): keylaunch: Swap F9 and F10 for SL-Cxx00....again *sigh* Jun 05 03:33:40 03coredump2 07org.oe.dev * rff99d101... 10/ (3 files in 3 dirs): gpe-login: Allow brightness adjustment via Fn+3 / Fb+4 for spitz, akita and c7x0, closing bug 2230 Jun 05 03:33:44 03coredump2 07org.oe.dev * r2056e2a0... 10/ (4 files in 3 dirs): xserver-common: Add the enhanced and very nicely commented xmodmap by Jeroen Hoek for SL-Cxx00 devices, closes bug 2404, deprecates at-fix-slcxxxx.patch and softkeys-slcxxxx-xmodmap.patch. Thanks Jeroen! Jun 05 03:33:48 03coredump2 07org.oe.dev * rdfea4f33... 10/ (3 files in 3 dirs): initscripts: mountall.sh: Create roots home directory on boot if it is missing, closing bug 2109 Jun 05 03:33:53 03coredump2 07org.oe.dev * r93c76dab... 10/ (1 packages/keylaunch/keylaunch_2.0.10.bb): keylaunch: Import forgotten patch from oz354x branch, closing bug #2231 Jun 05 03:33:58 03coredump2 07org.oe.dev * rc891266b... 10/ (3 files in 3 dirs): Jun 05 03:33:58 zaurusd: Fix a typo (?) in zaurus-mixer-callback, causing the speaker on spitz Jun 05 03:33:58 to always be turned off. This commit fixes output via speaker and headphones. Jun 05 03:33:58 Please review change on models != spitz as the problem could just as well have Jun 05 03:33:59 been caused by a driver bug. Jun 05 07:04:20 Hello, I have strange problems which halt the compilation of the linux kernel Jun 05 07:05:10 make oldconfig hangs. I have no idea why Jun 05 07:06:17 it waits for user input *probably* because your configs differ a lot Jun 05 07:07:09 what version of bitbake are you using ? Jun 05 07:08:12 version 827 Jun 05 07:08:36 1.6 or 1.8 ? Jun 05 07:09:53 1.8.3 Jun 05 07:10:25 yes I waits for some user interaction... but I didn't modified (yed) the config Jun 05 07:12:05 so maybe if I make makeoldconfig, make modifications then copy .config to defconfig, it should work ? Jun 05 07:12:19 yes that should work Jun 05 07:31:25 good morning all Jun 05 07:33:37 koen hi ! Jun 05 07:34:31 RP: I think i found a way to avoid bitbake "hanging" when doing oldconfig and the files differ alot Jun 05 07:34:59 RP: we need to change from oemake oldconfig to yes "" | oemake oldconfig Jun 05 07:35:41 RP: in this way the defaults will be accepted and won't hang waiting for user input Jun 05 07:36:44 morning Jun 05 07:36:54 hey hrw & steliosk Jun 05 07:38:02 hrw hi ! Jun 05 07:38:54 hey lrg Jun 05 07:39:19 hi Liam Jun 05 07:57:18 hi Jun 05 07:58:12 how can i build some package (i.e. busybox) with config placed outside of oe tree ? Jun 05 07:58:36 morning Jun 05 07:58:58 sandworm: use BB collections? Jun 05 07:59:40 sandworm: or just overwrite OE config with your, or use overrides power by creating dir for machine with its config etc Jun 05 08:00:37 alsa 0.14 released Jun 05 08:00:46 I know Jun 05 08:00:52 just installed it on my c760 Jun 05 08:01:27 hmm. now it has a mix of armv5te and armv4t packages ;) Jun 05 08:05:11 morning koen, hrw, XorA Jun 05 08:05:26 emacs 22 released :-D Jun 05 08:05:34 been ages for that OS to release Jun 05 08:06:07 ~prod lrg Jun 05 08:06:10 even Debian releases are more often Jun 05 08:06:20 * ibot zaps lrg with a high voltage cattle prod Jun 05 08:07:20 hrw: are two last recipes allowing me to place config file _outside_ of org.openemdedded.dev ? Jun 05 08:08:05 sandworm: you are making own distro? or need that config for own machine? Jun 05 08:08:06 hrw: collections: good. is it require to place all package files or only modified ? Jun 05 08:08:13 modified Jun 05 08:08:37 hrw: own distro at last Jun 05 08:11:00 sandworm: cd OE/packages/busybox/files/; mkdir sandworm-distro; cp mybusyboxconfig sandworm-distro/defconfig Jun 05 08:17:44 Hi! Jun 05 08:18:14 hrw: is this the one sa1100_udc you know about - http://balloonboard.org/pipermail/balloon/2006-June/000499.html ? Jun 05 08:20:05 psokolovsky: same Jun 05 08:20:47 hrw: i.e., you know about anything "better" than that? it doesn't work, indeed. Jun 05 08:21:25 no Jun 05 08:21:48 hrw: but older (essentially 2.4-based) adhoc driver for sa1100 (as currently in hh.org cvs) does work for me on h3600, but rather flaky Jun 05 08:26:45 I updated the angstrom arm-oabi feeds today Jun 05 08:26:56 (speaking of strongarm) Jun 05 08:29:45 does anyone know where oemake function is defined ? Jun 05 08:32:32 classes/*.bbclass Jun 05 08:33:26 base.bbclass from the looks of it Jun 05 08:35:16 steliosk: thats the default one Jun 05 08:45:37 openmoko bitbake patch is awesome Jun 05 08:45:49 parsing for 45 minutes and still not end Jun 05 08:46:12 like zecke said "it tells people SRCDATE=now is bad" Jun 05 08:46:30 03koen 07org.oe.dev * rf600bb50... 10/ (1 conf/distro/angstrom-2007.1.conf): angstrom: add gstreamer feed Jun 05 08:49:37 hrw : any reason it is so slow ? Jun 05 08:50:36 steliosk: for every svn recipe it contacts the svn server Jun 05 08:50:58 so if you have 300 svn recipes -> 300 svn connects Jun 05 08:52:30 people have reported that parsing can take 10 hours Jun 05 08:57:23 03koen 07org.oe.dev * r7abc4507... 10/ (1 packages/gaim/pidgin.inc packages/gaim/pidgin_2.0.1.bb): pidgin: fix RRECOMMENDS Jun 05 09:05:08 Hi all Jun 05 09:05:35 RP: ping Jun 05 09:22:22 hello...adding new packages to the final build is as simple as adding DISTRO_EXTRA_RDEPENDS=""..right? Jun 05 09:22:31 to the local.conf Jun 05 09:24:22 thesing: pong Jun 05 09:25:18 !oebug 2415 Jun 05 09:25:19 * * Bug 2415, Status: NEW, Created: 2007-06-05 02:25 Jun 05 09:25:20 * * marek.vasut(AT)gmail.com: Vectoroids doesnt compie (fix included) Jun 05 09:25:21 * * http://bugs.openembedded.org/show_bug.cgi?id=2415 Jun 05 09:26:00 * * OE Bug 2415 has been created by marek.vasut@gmail.com Jun 05 09:26:02 * * Vectoroids doesnt compie (fix included) Jun 05 09:26:04 * * http://bugs.openembedded.org/show_bug.cgi?id=2415 Jun 05 09:26:06 RP: Is it possible that current kernels ignore the atag list from r2? Jun 05 09:26:36 thesing: You mean Zaurus ones? Jun 05 09:26:48 RP: no all arm kernels. Jun 05 09:27:04 thesing: No, normal arm kernels should read it Jun 05 09:27:12 thesing: Zaurus ones won't though Jun 05 09:27:44 RP: do you know the relevant files? Jun 05 09:28:32 hi RP Jun 05 09:28:39 RP, why is that so? :x Jun 05 09:29:22 thesing: The machine definition files usually say where to look for the atag params Jun 05 09:30:01 Marex: Because the Zaurus has a brain dead bootloader, we can't change the options it passes and they 2.4 commandline is not useful on a 2.6 kernel Jun 05 09:30:02 RP: thats what I found out. Jun 05 09:30:24 RP, ah :D Jun 05 09:30:36 RP, they why dont you just switch to uboot? Jun 05 09:30:43 *then ... Jun 05 09:30:55 RP: lookup machine_type should copy r2 to mach_desc->boot_params, if r2 != 0 right? Jun 05 09:31:23 RP: thats at least what arch/arm/setup.c uses Jun 05 09:31:30 Marex: Risky, a lot of work and not supportable for not enough gain Jun 05 09:31:49 thesing: I have no idea on the specifics without looking at the code Jun 05 09:31:52 RP: but that function doesn't do that (arch/arm/kernel/head_common.S) Jun 05 09:33:23 RP, chmm ... well ... ok :] Jun 05 09:34:21 RP: OK I will do some try-and-error than. Jun 05 09:53:40 any idea why bitbake 1.8 tries to build perl-native when using minimal-image ? Jun 05 09:55:38 tries and fails -> http://rafb.net/p/rIkkFG87.html Jun 05 09:58:41 it builds all from task-base Jun 05 09:59:57 hmmm Jun 05 10:02:02 steliosk: you can extract task-boot from task-base and send us a patch. Jun 05 10:03:05 koen, why rrecommend libjabber? In theory pidgin should rrecommend protocol jabber which in turn should rdepend on libjabber, no? Jun 05 10:07:26 psokolovsky,hrw I was now able to use the cde sa1100 driver for simpad :) Jun 05 10:07:57 thething does the cde driver work without changes on collie? Jun 05 10:08:12 s/thething/thesing Jun 05 10:09:12 mr_nice: what is cde driver? Jun 05 10:09:26 sorry cdc Jun 05 10:09:30 usb ethernet Jun 05 10:09:40 you mentioned todeay Jun 05 10:10:01 I fixed it diry for my simpad Jun 05 10:10:04 and it works Jun 05 10:10:42 you need to comment out the ifdev DEV_CONFIG_CDC Jun 05 10:10:58 mr_nice: cool! where patches can be found? Jun 05 10:11:33 psokolovsky: I can make one Jun 05 10:11:52 psokolovsky: but we will need to write some code to make it less dirty Jun 05 10:13:11 mr_nice: I'd appreciate if you can make patches, I'd be glad to test it on h3600. Jun 05 10:13:33 can take a half hour because of slow machine. I will poste it as soon as I have created the patch Jun 05 10:15:56 mr_nice: thanks! Jun 05 10:17:16 mr_nice: please send me tha patch as email. I will test it this evening. Jun 05 10:17:43 thesing: ok Jun 05 10:18:01 thanks. Jun 05 10:21:29 * thesing goes to lunch Jun 05 10:21:35 smacznego thesing Jun 05 10:21:50 bbs Jun 05 10:24:20 hi Jun 05 10:31:42 polyonymous: there is no protocol for jabber..... Jun 05 10:32:00 koen, hmm... ahh... then it must be protocol xmpp Jun 05 10:32:44 D'oh Jun 05 10:32:46 right Jun 05 10:45:40 hi Jun 05 10:45:45 hi zecke Jun 05 10:46:45 hey zecke Jun 05 10:49:45 man what a week Jun 05 10:51:43 zecke: how was linuxtag? Jun 05 10:55:25 okay Jun 05 10:55:36 mickeyl didn't catch a cold ;) Jun 05 10:55:49 03koen 07org.oe.dev * rab9e6b01... 10/ (1 packages/gaim/pidgin.inc): pidgin: recommend xmpp instead of jabber Jun 05 10:56:01 zecke: I wasn't there, so mickey|dinner was safe ;) Jun 05 10:56:18 ah you take the responsibility for mickey Jun 05 10:56:25 's sicknes at FOSDEM? Jun 05 10:56:31 yeah Jun 05 10:56:48 and the resulting delay for OpenMoko? Jun 05 10:56:52 mickey|dinner: did you hear it?! Jun 05 11:01:07 http://www.angstrom-distribution.org/unstable/images/collie/20070604/ Jun 05 11:02:41 anybody tried and had success with the 8GB SDHC cards in zaurus? seen sucess reports with the 4GB ones, just off to hong kong shortly and may pick a couple up if they work Jun 05 11:03:49 dan2003: 4GB sd or 4GB sdhc? Jun 05 11:04:07 thesing,psokolovsky sorry I was to fast :( *shame. It worked once but now it seems it was just luck not the code change. I will try to figgure it out :/ Jun 05 11:04:34 thesing: http://www.angstrom-distribution.org/unstable/images/collie/20070604/ Jun 05 11:08:11 hrw: i read somewhere some1 sayin gthey had a 4GB SDHC workin, but then im pretty sure the same place said anything above 2GB is SDHC, is that not the case? Jun 05 11:08:32 there are 4GB sd cards too Jun 05 11:08:43 ooooo Jun 05 11:08:50 2.6.21 has sdhc support Jun 05 11:09:01 mr_nice: anyway, if you have patches, I'd like to look at them and see what the kind of code it is. is that what was in initial 2.6 releases in mach-sa11xx/usb/ ? Jun 05 11:09:53 dan2003: 4gb can be either sdhc or normal sd, that's the trick Jun 05 11:10:33 thesing: Do you see any reason not to try out http://tree.celinuxforum.org/CelfPubWiki/SuspendToDiskForARM Jun 05 11:10:35 AFAIK, we don't yet have a hibernate functionality in OE. I'd like to see this to be able to change batteries and come back to the same session: http://www.oesf.org/forums/index.php?showtopic=21752 Jun 05 11:10:46 hey psokolovsky, long time no see. Jun 05 11:11:06 hi. summer time ;-) Jun 05 11:12:49 You mean Jun 05 11:12:51 You mean Jun 05 11:13:03 You are a geek and still have a real life? Jun 05 11:13:11 Come one. You crank me up ;-) Jun 05 11:14:07 koen: I guess the X11 collie image is only for boot from SD? Jun 05 11:14:18 Is the minimalist image set up to boot from SD? Jun 05 11:14:21 yes, it's >14MB Jun 05 11:14:40 Laibsch: the kernel decides what to boot, not the image Jun 05 11:15:14 OK, the kernel inside the image is not set up to boot from SD, is it? Jun 05 11:16:17 psokolovsky: http://shortlink.co.uk/m1o .the cdc code is taken from thesing and this was taken from balloon. Jun 05 11:17:47 psokolovsky: sorry wrong version.. mom Jun 05 11:17:58 ok Jun 05 11:18:39 psokolovsky: http://shortlink.co.uk/m1p Jun 05 11:19:39 thanks, will try it Jun 05 11:19:43 RP: I think I have not gotten your postal address yet. Jun 05 11:19:57 XorA: The cable is probably gonna be posted today Jun 05 11:20:12 Laibsch: One could try it. Could you at it to buzilla? Jun 05 11:20:33 Sure Jun 05 11:20:59 Laibsch: cool Jun 05 11:21:02 Laibsch: normal post? Jun 05 11:21:17 I think so. What is unnormal post? Jun 05 11:21:45 Laibsch: cool, as it probably arrive when I on holiday I didnt want it signed for Jun 05 11:21:55 as thats a touch hard to do from another country :-) Jun 05 11:24:53 I could send it your hotel in AMS if you prefer Jun 05 11:25:01 or delay the shipment a couple of days Jun 05 11:25:15 But I think since there is nothing to sign it should be OK as is Jun 05 11:25:28 Laibsch: it will be fine Jun 05 11:25:40 Laibsch: and be nice to get back and it sits on doormat :-) Jun 05 11:26:19 !oebug 2417 Jun 05 11:26:21 * * Bug 2417, Status: NEW, Created: 2007-06-05 04:25 Jun 05 11:26:22 * * marek.vasut(AT)gmail.com: ScummVM doesnt compie (fix included) Jun 05 11:26:23 * * http://bugs.openembedded.org/show_bug.cgi?id=2417 Jun 05 11:28:01 * * OE Bug 2417 has been created by marek.vasut@gmail.com Jun 05 11:28:03 * * ScummVM doesnt compie (fix included) Jun 05 11:28:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2417 Jun 05 11:28:13 * * OE Bug 2416 has been created by bugs.openembedded.org@rolf.leggewie.biz Jun 05 11:28:15 * * RFE: Add hibernation support for ARM Jun 05 11:28:17 * * http://bugs.openembedded.org/show_bug.cgi?id=2416 Jun 05 11:28:39 Marex looks like you are doing lot of opie fixes Jun 05 11:29:36 yeah we have a new website Jun 05 11:29:47 :) Jun 05 11:30:00 zecke: new opie website? Jun 05 11:30:29 schurig: companies website, with actual content :) Jun 05 11:32:03 hrw, slapin_fs told me some time ago, that every time I put bug to OE bugtracker, I should call oebug here Jun 05 11:32:24 hrw, and ... no ... Im just compiling OE and fixing all I meet through the way ;) Jun 05 11:33:03 Marex: welll, cdbot2 is tracking the bug tracker itself Jun 05 11:33:34 bye Jun 05 11:33:56 zecke, ah ... I see now ... that's why it's here twice :X sorry Jun 05 11:34:02 Marex: as to #2417 not before hell freezes... Jun 05 11:34:13 Marex: don't hardcode qte-mt! Jun 05 11:34:24 Marex: don't hardcode Qt as well... Jun 05 11:35:08 zecke, ok ok ... Jun 05 11:35:45 Laibsch: Did sleeping fast help ? ;) Jun 05 11:37:45 zecke, well ... I dunno how to fix it properly ... sorry Jun 05 11:38:18 Marex: well, give me a second to take a look Jun 05 11:38:28 zecke, thanx Jun 05 11:39:02 Marex: as to qte vs. qte-mt vs. qt vs. qt-mt some of the qmake(-*) classes will tell you what to do... Jun 05 11:39:04 Marex: look into palmtop.bbclass to check which var control qte/qte-mt and use it Jun 05 11:39:36 hrw, ok Jun 05 11:40:18 Marex: this brings us back to the virtual/libsdl game. what is your libsdl linking against? Jun 05 11:40:41 Marex: and it is always sane to attach the actual error message Jun 05 11:40:55 ok ... just a second then Jun 05 11:41:09 zecke, btw that error message was 4000+ lines long ;) Jun 05 11:41:29 Marex: it certainly is not longer than 10 lines Jun 05 11:41:36 that's why zecke saud 'attach', not 'paste' Jun 05 11:42:28 thesing: #openzaurus - user has 4GB SD cards and can test. will you help him? Jun 05 11:42:52 hrw: sure Jun 05 11:43:10 zecke: http://www.angstrom-distribution.org/repo/?name=qt4&action=search Jun 05 11:43:32 hehe Jun 05 11:51:42 koen, what kernel do you use for N800? Jun 05 11:51:54 the nokia one Jun 05 11:52:12 which is based on linux-omap Jun 05 11:52:17 (tony works for nokia) Jun 05 11:52:22 there are a number of patches going into the omap git that mention the n800 Jun 05 11:52:28 Crofton: if you build own one then you will loose wifi support due to binary driver Jun 05 11:52:38 hrw: not anymore Jun 05 11:52:54 koen: so what is a way to get it? Jun 05 11:53:10 I received a patch for OE that uses umac.ko to build a working version Jun 05 11:53:23 what is the name of the wifi chip? Jun 05 11:53:32 cx3110 Jun 05 11:53:33 c3110 iirc Jun 05 11:53:35 koen: how? Jun 05 11:55:51 hrw: http://ewi546.ewi.utwente.nl/tmp/cx311x_test.txt Jun 05 11:56:42 koen: ah.. that one.. Jun 05 11:57:00 http://use.the.umac.ko.from.your.own.nokia800/umac.ko is nice Jun 05 11:58:04 that is a scary looking thing Jun 05 12:02:09 thesing: A little bit. But I feel a bit tired. And you? Jun 05 12:03:29 03koen 07org.oe.dev * r225d874e... 10/ (4 files in 3 dirs): Jun 05 12:03:29 c3110x: add patch and magic to get a working umac.ko for the nokia 800 Jun 05 12:03:29 * you need to provide your own umac.ko to avoid licensing troubles Jun 05 12:03:41 Laibsch: me too. Jun 05 12:05:33 anyone got a big SCSI disk spare? Jun 05 12:06:02 zecke: 8" one? Jun 05 12:06:54 hrw: I don't know sizes. The 'normal' size as of 2007 Jun 05 12:07:13 03xora 07org.oe.dev * r9bd0abfd... 10/ (3 files in 2 dirs): xserver-kdrive-1.3.0.0 : add missing patches, stolen from poky Jun 05 12:07:19 03xora 07org.oe.dev * r755520d5... 10/ (1 packages/kazehakase/kazehakase_0.4.7.bb): kazehakase_0.4.7.bb : new version of browser Jun 05 12:08:40 XorA: Ah, the kazehakase browser. Didn't loose interest now that minimo is supposedly compiling? Jun 05 12:08:57 Laibsch: no, I just need time to fixup the guys code a bit Jun 05 12:09:14 Laibsch: minimo compiles no supposedly Jun 05 12:09:33 minimo is still a dog though ;-( Jun 05 12:09:42 thesing: Did you turn off debug logging in default compilation of collie kernel? There is no more reaction now when I insert the 1G card of mine. Jun 05 12:10:32 thesing: Wait, I forgot the module. Jun 05 12:10:46 Would be nice if it autoloaded Jun 05 12:11:42 Laibsch: add it to linux-rp.inc Jun 05 12:14:10 thesing: The card itself is still not recognized. I tested one of my 256MB cards and it worked fine. Jun 05 12:14:24 card itself = 1G card Jun 05 12:14:29 from A-Data Jun 05 12:16:29 someone used PCI cards which adds rs232/parallel ports? Jun 05 12:17:14 Laibsch: I think this is because of the mmc-spi layer. I will iron out the final bugs, when everthing else works more or less. Jun 05 12:17:15 * koen looks in the obsolete bin Jun 05 12:17:30 nope, only ISA rs232/parallel cards Jun 05 12:17:37 koen: I am thinking about buying one as I lack serial ports Jun 05 12:17:50 hrw: usb ports? Jun 05 12:18:16 koen: pci card with 2 rs + 1 lpt is 45 pln. serial->usb cable is nearly same price Jun 05 12:18:22 ah, right Jun 05 12:18:36 and serial->usb cable does not have to work. I have one non-working here Jun 05 12:19:21 hrw : I used a multi serial one (8 ports) but 2 serial +1 printer port are seen like build in ports usually Jun 05 12:19:30 XorA: http://software.twotoasts.de/media/midori/ Jun 05 12:19:55 koen: Access forbidden! Jun 05 12:19:57 XorA: sorry, http://software.twotoasts.de/?page=midori Jun 05 12:20:15 steliosk: I can get 4 port one but more expensive Jun 05 12:20:26 patches to compile it against recent webcore should be on the gtk-webcore mb list Jun 05 12:21:08 and even 6 serial one.. Jun 05 12:21:33 koen: fix webcore staging then :-) Jun 05 12:21:49 XorA: webcore changed API and moved headers around Jun 05 12:21:50 hrw : pci serail cards are better than usb->serial especially if you want to control some of the i/o pins directly Jun 05 12:21:54 gtk-webcore, that is Jun 05 12:22:07 hrw : for usb->serial get something with ftdi chipset Jun 05 12:22:16 steliosk: or pl2303 Jun 05 12:22:17 koen, zecke, sorry, I dunno how to fix it properly ... LDFLAGS += isnt correct too I guess Jun 05 12:23:13 Marex: without an error log, there is no sign of an error Jun 05 12:23:31 hrw : Well i am not so "fond" of pl2323 and friends Jun 05 12:23:32 Marex: I'm mere mortal, I can only fix what I understand, I can only understand when I see stuff Jun 05 12:23:50 zecke, ok ... just a second Jun 05 12:24:25 koen: my uclibc console-image hangs after Freeing init memory. How likely is it that the init output goes somewhere else? Jun 05 12:24:42 zecke: depends on the machine Jun 05 12:24:54 koen: neo and the emulator :) Jun 05 12:25:24 nothing on serial as well? Jun 05 12:27:15 koen: you make me read documentation... Jun 05 12:33:10 I know, I'm a bastard Jun 05 12:33:31 koen: I'm too lazy, any idea what to pass to -serial ? Jun 05 12:33:48 I never used the emulator Jun 05 12:33:59 ask balrog-kun on #openmoko Jun 05 12:36:24 zecke, http://disk.jabbim.cz/marex@jabber.cz/log.do_compile.23896 this is the log Jun 05 12:37:09 Marex: I'm not an angstrom dev, but attach it to the bugtracker? Jun 05 12:37:14 Marex: prefilter it? Jun 05 12:37:46 zecke, I thought you'll take a look ... Jun 05 12:39:03 hey mickeyl Jun 05 12:39:52 Marex: I take a look at bug reports, attach the log to the bug tracker Jun 05 12:41:07 ok ... I dont care ... the patch fixed the problem for me, do whatever you want with it ... Jun 05 12:41:08 Marex: sorry. For issues you see. E.g. a link failure, compile error. Take the relevant part of the log and attach it to a bug report Jun 05 12:42:37 Marex: we might have a different understanding of quality, For me quality starts with the process. Document the issue (attaching an error log to the bug tracker), attaching a fix and saying why one believes to fix the issue Jun 05 12:42:52 mickeyl: http://article.gmane.org/gmane.comp.window-managers.enlightenment.devel/13140 and http://article.gmane.org/gmane.comp.window-managers.enlightenment.devel/13124 Jun 05 12:44:22 zecke, maybe next time ... Jun 05 12:47:52 Marex: well, I don't care but is it really too difficult to attach an error output to the bugtracker Jun 05 12:48:57 zecke, Im past that package so it is ... Jun 05 12:49:13 that's why I said maybe next time Jun 05 12:49:51 Marex: OE did a lot of work for your Jun 05 12:50:08 * zecke shakes head and places Marex onto his ignore list. Luckily marex will never need to rebuild scummvm... Jun 05 12:50:09 Be kind to OE and document your problems problems Jun 05 12:50:24 problems proprerly Jun 05 12:51:09 Laibsch, I will ... next time I get to a problem Jun 05 12:52:02 which will probably be sooner than you think.... Jun 05 12:52:30 maybe next build? if there is no fix :b Jun 05 12:53:23 steliosk, right ;) Jun 05 12:53:40 steliosk, right now :B Jun 05 12:54:00 Marex: And there you have it Jun 05 12:54:06 I hope NOBODY will help Jun 05 12:54:15 Since you showed to be an ignorant Jun 05 12:54:23 somebody who just takes and does not give Jun 05 12:54:24 Laibsch, Ill help myself and commit patch :B Jun 05 12:54:30 We don't appreciate such people Jun 05 12:54:34 Laibsch, I commited a patch :E Jun 05 12:54:46 to every bug I found and managed to fix Jun 05 12:55:11 zecke : how difficult it will be to get oemake to recognize if its called to make oldconfig ? Jun 05 12:55:43 steliosk: oe_runmake? it is shell and I have no clue about shell ;) Jun 05 12:56:48 wb steliosk! Jun 05 12:56:55 zecke : here is the situation : if oemake oldconfig gets a config that's significally different (i.e has new options) it halts waiting for user input Jun 05 12:57:10 heh Jun 05 12:57:27 even detecting this happens would be useful Jun 05 12:57:46 I had a very similar issue Saurday and wasted a couple of hours before I figured it out Jun 05 12:58:02 zecke : which can be solved by changing oemake oldconfig to yes "" | oemake olsconfig Jun 05 12:58:31 steliosk, it may be better to detect and quit Jun 05 12:58:43 so the user can provide manaul intervention Jun 05 12:59:03 in my case, I had a badly screwed up config, that should not have been auto fixed Jun 05 12:59:04 Crofton : it does not quit... that's the problem. And if you do auto builds it hangs there Jun 05 12:59:18 yes, I understand Jun 05 12:59:40 we need a way to handle this Jun 05 12:59:50 I agree :) Jun 05 13:00:13 Now that it has happened to me, I will detect it faster in the future Jun 05 13:00:23 IMO yes "" | oldconfig is the right solution Jun 05 13:00:24 but auto detections would be very good Jun 05 13:00:56 Crofton: isn't there a way to automatically say no? Jun 05 13:01:10 koen : we either change all references to that or handle it automagically by bitbake/bbclasses Jun 05 13:01:46 zecke: the problem is that it's now always "y" or "n" Jun 05 13:01:54 s/now/not/ Jun 05 13:02:08 what about "m: Jun 05 13:02:11 er "m: Jun 05 13:02:13 arg Jun 05 13:02:15 "m: Jun 05 13:02:24 you know what I mean Jun 05 13:02:56 choosing the default is the sane option for automagic configurations Jun 05 13:03:14 you could make a case for erroring out and yelling at the maintainer to fix the defconfig Jun 05 13:03:21 in my case, I had messed up the efika config so it was asking about x86 things Jun 05 13:03:29 koen, I agree Jun 05 13:04:28 so we need something that can be set to "auto" (i.e yes) or "fail" if this happens Jun 05 13:04:42 steliosk, you are having this issue against git kernels? Jun 05 13:05:03 nop uclibc-initial and bitbake 1.8 Jun 05 13:05:05 steliosk: how to detect it? checking for '? and [Y/n] [y/N] [y/n/M]? Jun 05 13:05:15 steliosk: alternatively we auto return? Jun 05 13:05:17 hmm Jun 05 13:05:50 zecke : no i mean set it in advance to handle it accordingly Jun 05 13:06:14 steliosk: how do you want to fix the .config in advance? Jun 05 13:06:28 steliosk: you have to see it blocks! and wants an input. How do you want to handle that? Jun 05 13:06:41 error out Jun 05 13:07:05 waiting for user input (which can not be entered) is useless Jun 05 13:07:06 error out :) Jun 05 13:07:18 okay, once again Jun 05 13:07:23 How to decide when to error out Jun 05 13:07:30 a) Always error out on invocation Jun 05 13:07:35 * Crofton gets nervous about machines building kernel configs :) Jun 05 13:07:36 b) Have some criteria to test Jun 05 13:07:52 c) patch the buildsystem to error out? Jun 05 13:08:22 zecke : Is this an exam ? :) Jun 05 13:08:22 For "standard" kernels, this shouldn't happen Jun 05 13:08:28 since source is fixed Jun 05 13:08:45 in my case, I had to do a quick hack to build ehci_hcd for Efika Jun 05 13:08:52 quick hack took abobut 3 hours Jun 05 13:09:06 For the git recipes, this can happen with time Jun 05 13:09:22 how about piping "yes n" to stdin of make oldconfig by default Jun 05 13:09:28 so the user of the git recipes shoudl be told to update defconfig Jun 05 13:09:41 zecke : ideal would be to have a "special" oemake_oldconfig that takes as a parameter what to do Jun 05 13:10:27 zecke : default would be to error out if it asks for user input Jun 05 13:10:39 XorA: do youknow RP's address? Jun 05 13:10:51 zecke : or in certain cases it should igonre it (i.e say yes to all) Jun 05 13:10:52 Laibsch: only that he is south of me Jun 05 13:11:04 steliosk: sorry, you fail to understand my question Jun 05 13:11:16 steliosk: how do you want to detect it asks for input? Jun 05 13:11:42 zecke : check for ? Y/n as you said Jun 05 13:11:49 i don't see another way Jun 05 13:12:13 steliosk: its Y/M/N normally Jun 05 13:12:42 zecke : i am not sure if oldconfig has any option to exit with an error if user input is required .... Jun 05 13:13:07 zecke : there was something called silent_oldconfig (or something like that) that could do it Jun 05 13:13:20 XorA: I guess he can be happy he is not to the left of you Jun 05 13:14:01 Laibsch: is there an easy way to get a cvs list of all bug mutation using a shellscript? Jun 05 13:14:01 Laibsch: true Jun 05 13:14:26 Laibsch: to generate something like http://lists.linuxtogo.org/pipermail/openembedded-devel/2007-June/002240.html Jun 05 13:14:29 koen: Don't understand the question Jun 05 13:14:36 OK, hold one Jun 05 13:14:57 I have shell script to do it for monotone, but I'd like one for bugzilla as well Jun 05 13:15:09 (or a volunteer to do it manually every week) Jun 05 13:16:00 koen: make a bugzilla query, save the xml and do xslt? Jun 05 13:16:08 * steliosk was told while in the army that volunteers die first Jun 05 13:16:25 zecke: that's an option too Jun 05 13:16:40 koen: you mean a CSV list? Jun 05 13:16:47 zecke: I'm trying to get the infrastructure for weekly OE releases going Jun 05 13:17:16 koen: take the code from the (uninstalled) ikada server Jun 05 13:17:22 ~time Jun 05 13:17:26 koen: The best I can come up with is a query and then parse the HTML Jun 05 13:17:43 time is probably 1 dimensional, or everlasting, an illusion, or 2007.06.05 13:17:46 GMT Jun 05 13:17:47 I know bugzilla can output csv data Jun 05 13:17:54 that's how I did it in the email :) Jun 05 13:18:16 but I used Excel instead of awk to parse it Jun 05 13:18:18 koen: But bugzilla seems to interpret 1W and 7d very differently. The 7d interpretation seems to be more to the truth Jun 05 13:18:22 * koen hangs head in shame Jun 05 13:18:41 koen: whats wrong with gnumeric Jun 05 13:18:53 XorA: no gnumeric on my mac Jun 05 13:18:54 koen: I stumbled the "cvs" word which I did not understand Jun 05 13:19:04 koen : That's why you ask for volunteer.... to do excel dirty job hehe Jun 05 13:19:04 koen: mac sucl Jun 05 13:19:06 ~lart my typos Jun 05 13:19:06 * ibot wallops my typos with a main rotation server that needs rehubbing. It won't take long Jun 05 13:19:54 koen: You want all changes, all status changes or all fixed bugs? Jun 05 13:20:16 Laibsch: all fixed/duplicate/wontfix/invalid bugs to start with Jun 05 13:20:24 later on all new bugs as well Jun 05 13:21:08 koen: http://tinyurl.com/2tgeky ? Jun 05 13:21:14 Laibsch: in the end something like this should be done by a script: http://mail.gnome.org/archives/gtk-devel-list/2007-June/msg00018.html Jun 05 13:21:33 koen: I think above URL is for you. Jun 05 13:21:59 http://tinyurl.com/ywlmdp Jun 05 13:24:18 zecke : let me know if you come up with something for handling the oldconfig Jun 05 13:24:55 koen: I am no XML guru. But the bugzilla should be easy to beat into shape. Jun 05 13:25:03 bugzilla output Jun 05 13:26:15 Wiat, bugzilla can output straight csv Jun 05 13:27:14 Laibsch: http://rafb.net/p/UtFu1s50.html Jun 05 13:27:53 the output: http://rafb.net/p/Y9akYL48.html Jun 05 13:28:21 looks good, except for the useless use of cat award you'd get ;-) Jun 05 13:28:38 I think that should do it Jun 05 13:41:49 wow.. there are also pci express cards with serial ports Jun 05 13:52:18 hrw: where??? Jun 05 13:52:30 * mmp is going to buy a pile, just to be sure:) Jun 05 13:55:00 mmp: in Polish auctioning webservice Jun 05 13:55:16 I don't want to ask stupid questions but how would I compile a bunch of *.o files with bitbake? Jun 05 13:55:45 I thought it would work the same like make Jun 05 13:57:17 you don't, you write a recipe for it Jun 05 13:59:01 Laibsch: http://ewi546.ewi.utwente.nl/tmp/ChangeLog Jun 05 14:00:06 sorry there are no *.o files at all. the source looks like this: http://pastebin.ca/540124 Jun 05 14:00:27 koen: and what can I do with that recipe then? Jun 05 14:00:32 bitbake it Jun 05 14:00:48 hrw: mm, not that far; maybe at your southern neighbors there could be something similar... :) Jun 05 14:01:01 dcordes: you know what a Makefile is? Jun 05 14:01:50 mmp: Czech or Slovak? same allegro service Jun 05 14:01:55 hrw: and apparently there is:) Jun 05 14:01:56 Laibsch: now I just need to make that a cronjob :) Jun 05 14:01:57 Slovak Jun 05 14:02:01 * * OE Bug 2418 has been created by marek.vasut@gmail.com Jun 05 14:02:03 * * Keyring wont compile, fix included Jun 05 14:02:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2418 Jun 05 14:02:12 hrw: I even googled shop which seems to sell them officially Jun 05 14:02:52 Laibsch, zecke satisfied? Jun 05 14:03:09 Marex: that patch doesn't apply to OE Jun 05 14:03:14 Without looking, I am Jun 05 14:03:20 Since you do seem to make an effort Jun 05 14:03:23 good job! Jun 05 14:03:26 zecke: files that tell make what to do? Jun 05 14:03:34 mmp: hm. now to decide.. 48 PLN for 2xRS,1xLPT or 105 PLN for 6xRS Jun 05 14:03:35 koen, that patch does apply to that package Jun 05 14:03:47 03koen 07org.oe.dev * rcc77b9d1... 10/ (1 contrib/mtn2cl/mtn2cl.sh): mtn2cl: make it select comments for the last week automagically and also append fixed bugs Jun 05 14:03:58 I wonder does 2xPCI with 2xRS (so 4xRS in total) will works Jun 05 14:04:00 koen: good work! Jun 05 14:04:12 hrw: the first one seems to have wider use - at least you can work with some parallel port devices... Jun 05 14:04:14 koen: does it still do rm? Jun 05 14:04:21 hrw: yes Jun 05 14:04:35 koen: then create temporary dir for them please Jun 05 14:04:43 hrw: you need a mtn workspace anyway, so I run the script in that dir Jun 05 14:04:47 hrw: feel free to fix it Jun 05 14:04:56 hrw: if your machine doesn't have parport (I found it quite useful at least because some dev boards have parport interface) Jun 05 14:05:07 * zecke wonders if mStartTime needs to be void* in the first place Jun 05 14:05:08 s/found/find :) Jun 05 14:05:30 dcordes: your .bb would just call oe_runmake (I assume the Makefile is sane) Jun 05 14:05:44 mmp: my machine lack parport but I never had to use it for something other then old printer Jun 05 14:05:59 Do we have libqt4 in OE? Jun 05 14:06:08 * Laibsch kind of lost overview of the QT stuff Jun 05 14:06:10 qt4-x11-free Jun 05 14:06:14 OK, nice Jun 05 14:06:17 mmp: and for sure I do not need it to do ARM JTAG - my Neo DebugBoard has 20 pin JTAG port added already Jun 05 14:07:00 koen, and now? (keyring-oe.patch) Jun 05 14:07:12 zecke: erm. ok :) Jun 05 14:08:06 Laibsch: http://www.angstrom-distribution.org/repo/?name=qt4&action=search Jun 05 14:10:37 can't somebody compile the module for me? :) Jun 05 14:10:52 hrw: well, I have one devkit which unfortunately uses parport (and it's also useful for home-made AVR programmer:) Jun 05 14:11:09 http://bitbake.berlios.de/manual/ch02.html this looks so complicated for me with those variables and so on Jun 05 14:12:05 dcordes: I think you'd have more success asking for help in setting up OE to compile for yourself Jun 05 14:12:10 And it is very easy Jun 05 14:12:24 * koen mentions http://www.angstrom-distribution.org/building-angstrom again Jun 05 14:12:26 And you've already been pointed to some good docu Jun 05 14:12:40 exactly, koen had mentioned that link already Jun 05 14:12:41 Laibsch: I already did... I have bitbake and the dependencies installed but I have no idea how to compile Jun 05 14:12:45 err write recipes? Jun 05 14:12:48 What is so difficult? Jun 05 14:13:00 which doku? Jun 05 14:13:03 dcordes: Just a minute Jun 05 14:13:11 doku: link above from koen Jun 05 14:13:26 dcordes: You know how to compile stuff in general? Jun 05 14:13:31 koen: angsrom uclibc EABI does not work in the non-upstream qemu emulator, please fix it Jun 05 14:13:50 zecke: :) Jun 05 14:13:52 Laibsch, koen zecke logs added ... Jun 05 14:14:09 good job! Jun 05 14:14:28 Laibsch: make Jun 05 14:14:42 ? Jun 05 14:15:12 We don't have python-numpy or python-matplotlib in OE yet, do we? (yes, koen I checked the feed browser ;-)) Jun 05 14:15:20 dcordes: yes and no. Jun 05 14:15:28 that is not the whole truth Jun 05 14:15:30 numpy iirc we have Jun 05 14:15:32 Laibsch: dunno, I don't use python on the device Jun 05 14:15:38 Laibsch: I know how to run a configure script and how to type in make lol Jun 05 14:15:41 ok and make install Jun 05 14:15:53 OK, let's start from that Jun 05 14:16:03 nice Jun 05 14:16:23 I'll help you a bit Jun 05 14:16:30 But I want you to read http://www.openembedded.org/user-manual Jun 05 14:16:43 It might be too much for you right now, but it is good stuff Jun 05 14:16:48 and a fairly easy read Jun 05 14:17:21 who is this "Holger Hans Peter Freyther" guy? Jun 05 14:17:30 http://www.openembedded.org/user-manual&dpage=chapter_recipes is particularly important for your current question Jun 05 14:17:39 Marex: zecke Jun 05 14:17:52 dcordes: http://www.openembedded.org/user-manual&dpage=chapter_recipes is particularly important for your current question Jun 05 14:17:57 * koen hints at http://www.openembedded.org/people Jun 05 14:18:03 ah ... no wonder Jun 05 14:18:24 Marex: I feel your manners are slipping again Jun 05 14:18:33 And that after you made such good progress Jun 05 14:18:44 You would not be where you are now without zecke Jun 05 14:18:52 So pay some due respect, please! Jun 05 14:18:57 dcordes: Read at least that part and once you've done that ping me Jun 05 14:19:49 Laibsch, that's because what happened here made me quite angry Jun 05 14:20:51 Laibsch: roger that. Jun 05 14:24:13 Laibsch: does cross compile mean that the binaries can be executed on every oe distro? Jun 05 14:24:24 no Jun 05 14:24:35 they can execute on the processor they are compiledfor Jun 05 14:24:50 which can be different than the one you are compiling on Jun 05 14:25:02 ah ok so I have that recipe and can tell the compiler to compile it for a certain architecture? Jun 05 14:25:18 and that will work on any distro using that architecture? Jun 05 14:25:24 well, the machine file well tell OE what cross compiler is needed Jun 05 14:25:45 ok Jun 05 14:28:41 koen: alsa 1.0.14 landed in Poky Jun 05 14:29:16 wait.. do I have to write the whole recipes myself? Jun 05 14:29:32 that syntax page is too much for me Jun 05 14:31:13 dcordes: yes, you write it yourself Jun 05 14:31:22 Take an example and adapt it Jun 05 14:31:28 You don't need all fields Jun 05 14:31:30 oh no Jun 05 14:31:40 a couple of them will be sufficient Jun 05 14:31:44 I will do this tomorrow I think Jun 05 14:31:52 google for "my first bb" or something like that Jun 05 14:32:02 I think there are some very easy walk-throughs Jun 05 14:32:49 morning Jun 05 14:32:53 * koen wonders how people can find crosscompiling easy and writing recipes hard Jun 05 14:32:56 hey chouimat Jun 05 14:35:16 :) Jun 05 15:50:03 http://pastebin.ca/540321 Jun 05 15:50:10 these people annoy me Jun 05 15:50:37 Between the one day window to turn in time sheets on time, me forgetting and the guy that has to sign being out of town it is no wonder Jun 05 15:50:43 I just don't care Jun 05 15:56:43 Crofton wow :) Jun 05 15:57:02 Apparently, I am a problem Jun 05 15:57:22 * Crofton takes pride in causing trouble Jun 05 15:57:38 Crofton sound like someone with no real power that want to abuse it :p Jun 05 15:57:44 yeah Jun 05 15:58:01 apparently it annoy auditors, but the system is really stupid Jun 05 15:58:11 koen http://www.groklaw.net/article.php?story=20070604183519938 Jun 05 15:58:15 Crofton ^^^^ Jun 05 16:04:12 do we have a good link explain OABI vs EABI? Jun 05 16:05:02 Crofton: the Debian Wiki? Jun 05 16:05:08 Crofton: the CodeSourcery FAQ? Jun 05 16:05:22 Crofton I found some on arm.com Jun 05 16:05:22 yeah, looking at debian wiki now Jun 05 16:14:28 guys, I found a forum post about some GPL tool that can create an NK bin with a linux kernel inside, in order to trick eboot into booting linux instead of WinCE... does anyone by chance know what tool that is? Jun 05 17:00:42 hi folks, newbie here.. i wonder if someout there can give me a hint on trying to build an "image" as in "opie-image" or "gpe-image" Jun 05 17:01:50 see you all Jun 05 17:03:27 the "GettingStarted" on the wiki indicated that I can just run "bitbake opie-image" and it should start up, but when i do it, it complains about no provider. I do have my local.conf set up with the BBFILES pointing to my packages dir Jun 05 17:03:48 i am able to successfully build other packages, just not the "image" ones... Jun 05 17:07:58 tony_: Works for me Jun 05 17:08:22 angstrom-opie-image, that is Jun 05 17:08:35 hmm, let me try it now Jun 05 17:09:18 ERROR: No providers of build target angstrom-opie-image (for []) Jun 05 17:09:32 now, i am trying to do a build for machine "mainstone" Jun 05 17:09:48 there is a lot i don't understand, so maybe i broke something Jun 05 17:10:37 right now i am trying to do a build of "opie-clock" and it's chugging away.... Jun 05 17:10:56 and all the packages are in the same path Jun 05 17:26:01 * * OE Bug 2419 has been created by marek.vasut@gmail.com Jun 05 17:26:03 * * qpegps wont compile, fix follows Jun 05 17:26:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2419 Jun 05 17:30:50 I get some errors while trying to build angstrom: http://pastebin.ca/541888 Jun 05 17:31:22 tony_, see this http://www.angstrom-distribution.org/building-angstrom this http://www.openembedded.org/wiki/GettingStarted and this http://www.linuxtogo.org/gowiki/OpieWithAngstrom Jun 05 17:31:42 tony_, if you combine those three, you'll be able to compile opie with angstrom Jun 05 17:32:05 it'll be similar in others imo ... Jun 05 17:34:31 Marex, thanks... i have been using the GettingStarted wiki as my guide. I don't think what I am doing is relevent to Angstrom, though, i am trying to do a build for the PXA310, and have been using the cross compiler build I got from marvell Jun 05 17:35:33 I am new here... but I got the impression that Angstrom is a distro specific for the iPAQ? Jun 05 17:37:12 tony_, not at all ;) Jun 05 17:37:37 but that's not important ... what exactly failed? Jun 05 17:38:22 Just a note to anyone who's interested, #opie is back up Jun 05 17:40:34 Marex, well, i follow the GettingStarted wiki, and am able to get things set up to the point of "bitbake nano", which succeeds. but when I try to do a "bitbake opie-image", it complains "ERROR: No providers of build target angstrom-opie-image (for [])" Jun 05 17:41:29 tony_, ah ... so you're trying to build angstrom, right? Jun 05 17:41:49 not yet... i had my distro set as generic Jun 05 17:42:02 oops Jun 05 17:42:24 tony_, add the following line to your local.conf Jun 05 17:42:25 require conf/distro/include/preferred-opie-versions-1.2.3-pre.inc Jun 05 17:42:50 oh? ok, that looks interesting.. let me check Jun 05 17:44:38 the thing is... i get the same error even when i try "bitbake helloworld-image" Jun 05 17:45:11 i can complete "bitbake helloworld" fine Jun 05 17:49:34 tony_, probably because there is no "helloworld-image" Jun 05 17:51:12 Marex, maybe i misunderstand... i have a file called "org.openembedded.dev/packages/images/helloworld-image.bb" Jun 05 17:51:29 tony_, pastebin it somewhere Jun 05 17:51:38 tony_: for Opie check the FAQ of angstrom? Jun 05 17:52:17 I am new to IRC, what does pastebin mean? Jun 05 17:52:24 ~pastebin Jun 05 17:52:35 somebody said pastebin was a place to paste your stuff without flooding the channel - try http://pastebin.ca, or http://channels.debian.net/paste, or http://rafb.net/paste/, or http://pastebin.com is usually painfully too slow and unresponsive to use, use one of the other pastebin sites, or dpaste.com is a very nice pastebin as well Jun 05 17:53:33 got it Jun 05 17:54:33 could somebody have a look at my paste? http://pastebin.ca/541888 Jun 05 17:55:38 Marex, here is helloworld-image.bb http://paste.debian.net/29693 Jun 05 17:56:02 but doesn't everyone have it? i got it when i downloaded everything via monotone Jun 05 17:56:41 dcordes_: it looks like shasum-native is failing. could you pasetebin /home/erdinger/OE/build/tmp/akita/work/i686-linux/shasum-native-1.0-r1/temp/log.do_compile.8901 Jun 05 17:57:53 tony_, I see ... well it looks like it's dublicated Jun 05 17:57:59 *duplicated ... Jun 05 17:58:09 tony_, try deleting half of that file ;) Jun 05 17:58:18 ok, marex, i will Jun 05 17:58:24 i just noticed the dupe too Jun 05 17:58:37 just curious, what happend when you try to build it? Jun 05 17:58:59 let's see Jun 05 17:59:59 tony_, it compiled fine :) Jun 05 18:00:57 hvontres|poodle: http://pastebin.ca/541970 looks like I set some variable wrong, right? Jun 05 18:01:06 marex, ok, so my problem is more fundamental Jun 05 18:01:40 ok, thanks for the help.. i have to go to a meeting... i am in over my head with all this!! Jun 05 18:05:04 bipolar: Thanks for letting us know. Let's hope it picks up speed again, soon Jun 05 18:16:17 wow I missed that completely, QtopiaCore for DirectFB Jun 05 18:17:21 zecke: nice Jun 05 18:50:31 wow ... emacs gets drag and drop Jun 05 18:51:49 Hmm... why not? It does everything a good operating system does, why not add all the GUI functions too? ;-) Jun 05 18:52:13 * mwester is a lifelong Emacs devotee... Jun 05 18:52:53 i was merely pointing that out... some 20 years behind even windows :) Jun 05 18:55:47 too bad its got support got gtk.. seems like the wrong toolkit to use :) Jun 05 19:11:00 ljp: you could always start a QtEmacs project :) Jun 05 19:11:14 i was thinking of it Jun 05 19:11:38 dcordes_: yes... but I am not sure which one got borked to do that. Jun 05 19:11:57 then we coujld add it to opie Jun 05 19:12:07 heh Jun 05 19:12:09 jk Jun 05 19:14:15 ljp: well, since emacs is 'not just an editor, its a way of life', we could make it it's own environment :) Jun 05 19:15:15 hmm. good point! Jun 05 19:15:23 ljp: opps, I guess I need to go file with the uspto now..:) Jun 05 19:15:34 doh! Jun 05 19:16:29 ljp: we could call it E-njaeiawol :) Jun 05 19:17:03 hmm.. dont know what that means... but ok! Jun 05 19:18:02 ljp: Emacs-not just an editor, its a way of life Jun 05 19:18:03 geee.. the commandline version of new emacs looks just like the old one Jun 05 19:18:08 ljp: did you look at the (x)emacs code? it's ugly and I will be a lot bored having to do a qt of it :) Jun 05 19:18:35 not really Jun 05 19:18:37 * hvontres|poodle would prefer QtVi anyways Jun 05 19:18:46 * ljp shudders Jun 05 19:18:52 hvontres|poodle: look at yzis Jun 05 19:18:57 kvim Jun 05 19:19:11 ljp: heh. Jun 05 19:19:18 http://www.yzis.org Jun 05 19:19:42 i just dont understand why I should have to type in commands before I get to start editing... Jun 05 19:20:05 too many key presses Jun 05 19:21:34 ljp hehe Jun 05 19:29:25 wrong Jun 05 19:30:12 * mwester marvels that after almost 20 years, the conflict between emacs users and vi users continues... :-D Jun 05 19:30:35 *star wars theme* Jun 05 19:31:29 * mwester is so old that he knew Luke Skywalker personally. Luke used emacs. Jun 05 19:31:37 NO Jun 05 19:33:24 Crofton : Have you tried to get efika's 2nd serial port to work? Jun 05 19:35:53 so did Darth Jun 05 19:40:02 * * OE Bug 2420 has been created by marek.vasut@gmail.com Jun 05 19:40:04 * * TxDrug wont compile, fix follows Jun 05 19:40:06 * * http://bugs.openembedded.org/show_bug.cgi?id=2420 Jun 05 20:11:27 no Jun 05 20:11:59 * Crofton saw the first showing of Star Wars in Charlottesville, VA, USA Jun 05 20:12:28 hiall. I'll just idle here for some time Jun 05 20:12:28 yeah Jun 05 20:12:29 quiet Jun 05 20:12:29 it's been quiet here all day. Jun 05 20:12:30 hiall. I'll just idle here for some time Jun 05 20:12:30 yeah Jun 05 20:12:31 quiet Jun 05 20:12:32 it's been quiet here all day. Jun 05 20:19:44 lol Jun 05 20:19:59 moo: what happened? Jun 05 20:21:06 Sorry about that...pretty embarassing there. I think that my caffinated fingers found a new feature in my irc-client... Sorry again Jun 05 20:21:13 haha Jun 05 20:21:15 * chouimat|bored hates cold rainy day ... it's BORING Jun 05 20:26:48 I went through my logs but can't find it: What was it again that I needed to change to boot from SD? Jun 05 20:27:17 Laibsch: you need to change the kernel cmdline Jun 05 20:27:45 console=ttySA0,115200n8 console=tty1 noinitrd root=/dev/mmcblk0p1 rootfstype=ext2 rootdelay =5 mem=64M fbcon=rotate:1 dyntick=enable debug Jun 05 20:27:50 is mine Jun 05 20:28:04 (without the tab) Jun 05 20:28:56 hi pvanhoof Jun 05 20:29:10 hey Jun 05 20:29:59 thesing: But where do I change it? What file in bb? Jun 05 20:30:09 linux-rp.inc? Jun 05 20:30:21 Yes Jun 05 20:31:01 But you also have to compile the mmc_drivers into the kernel. Jun 05 20:33:48 thesing: CMDLINE_ROOT ? Jun 05 20:34:06 What is yours? Can you paste a diff for your linux-rp.inc? Jun 05 20:35:21 I don't build my kernels with OE here is my working kernel: http://www2.informatik.hu-berlin.de/~tkunze/zaurus/images/zImage Jun 05 20:36:20 hm, if possible, I'd stick to OE Jun 05 20:44:28 Laibsch: you can try setting CMDLINE_ROOT_OVERRIDE="root=/dev/mmcblk0p1 rootfstype=ext2 rootdelay=5" in your local.conf Jun 05 20:50:04 after solving the my problem with the gcc path not being found with a stupid symlink, I get the following errors compiling the minimal image of angstrom: http://www.pastebin.ca/542425 How can I fix this? Jun 05 20:53:36 Jun 05 20:57:44 koen: what about Düsseldorf :) ? Jun 05 20:58:18 90 minutes ago I was there Jun 05 20:58:19 I take that as he has returned from DDF Jun 05 21:00:01 koen: it is like 20 minutes from my place Jun 05 21:00:34 ratingen here Jun 05 21:01:19 dcordes_: no way.... I was born there :) ( a while back...) Jun 05 21:01:38 hvontres|poodle: in ratingen? Jun 05 21:01:40 not really Jun 05 21:01:45 must be kidding Jun 05 21:02:27 dcordes_: yes... then I moved up to Rendsburg and now I live in beautifull Souther California :) Jun 05 21:02:45 Laibsch: I picked up my parent from the airport Jun 05 21:02:50 parents* Jun 05 21:03:00 hvontres|poodle: no wai ^^ Jun 05 21:03:46 ~dict turd Jun 05 21:03:51 Dictionary 'turd' (1 of 2): obscene terms for feces . Jun 05 21:03:53 koen: Well, that is because you are a nice guy! Jun 05 21:05:02 hvontres|poodle: man how could you leave the hood?? :) Jun 05 21:05:35 the homies is here Jun 05 21:05:41 * dcordes_ quots gta Jun 05 21:07:12 homies? hood? Jun 05 21:07:12 hvontres|poodle: how old were you when you left ratingen? Jun 05 21:07:15 WTF? Jun 05 21:07:35 dcordes_: one or two :) Jun 05 21:07:47 oh ok then Jun 05 21:10:42 hvontres|poodle: do you have any relatives here? Jun 05 21:12:56 dcordes_: yeah, most of my dad's family is still in the old country :) Jun 05 21:21:55 great feeling to live in the city where one of the almighty oe devs is from ^^ Jun 05 21:23:05 1245184 8% 18.12kB/s 0:12:21 Jun 05 21:23:15 normally that's 2.5MB/s Jun 05 21:25:37 dcordes_: hahahaha... I am just an OE newb who can build a console image for one machine :) Jun 05 21:25:51 dcordes_: but thanks anyways :) Jun 05 21:30:23 thesing: It is really nice to have suspend in the kernel now Jun 05 21:30:37 Before that I had to force that battery out so often Jun 05 21:30:43 Was a real bitch Jun 05 21:31:17 Laibsch: did the CMDLINE_ROOT_OVERRIDE thing work for you? Jun 05 21:31:48 But I think it still loses to much power in suspend. But I havent realy tested Jun 05 21:32:16 thesing: are you leaving the serial cable plugged in? Jun 05 21:33:27 Yes. Jun 05 21:33:55 thesing: IIRC, the 3.3V line stays up during suspend, so the cable is still sucking power Jun 05 21:34:17 thesing: asuming you are using an active cable Jun 05 21:34:52 I don't know if its active or not. Jun 05 21:34:52 hvontres|poodle: I will compile an image tonight Jun 05 21:35:39 hvontres|poodle: let me take a guess.. the poodle? Jun 05 21:36:05 Laibsch: BTW, I am making progress on my altboot mods. It seems every time I think it is ready, I find another little bug Jun 05 21:36:16 dcordes_: how'd you guess :) Jun 05 21:36:21 dunno,, Jun 05 21:37:36 hvontres|poodle: I might have an hack available to change the cmdline of a kexec kernel via sysfs soon. Jun 05 21:37:52 thesing: whooooooo :) Jun 05 21:38:14 hvontres|poodle: Is your work available for public consumption? Jun 05 21:38:26 hvontres|poodle: You know, release early and release often Jun 05 21:39:55 Laibsch: check out svn://hentges.net/server/svn/public/altboot/branches/1.0.x Jun 05 21:40:14 Laibsch: I also have a matching .bb file, but it's on my machine at home Jun 05 21:40:36 Laibsch: so far it's only been tested on poodle :) Jun 05 21:46:21 does somebody have a high resolution picture of the latest neo? Jun 05 21:59:57 later all Jun 05 22:14:17 hi Jun 05 22:14:21 heya Jun 06 00:28:06 ~lart gpe-conf for cloning "bl" but not doing it properly Jun 06 00:28:06 * ibot takes out gpe-conf with the trash for cloning "bl" but not doing it properly Jun 06 00:39:29 * crink had tried minimo on his board yesterday.. Jun 06 00:56:49 doh.. guys.. regarding ipkg feeds, I do run ipkg-make-index in my package directory, there are some newer files there; I do update/upgrade on my machine but it says nothing to do Jun 06 00:56:55 does anyone haev an idea? Jun 06 01:00:15 Jin^eLD: i think nothing you can do, wait and see. Jun 06 01:00:23 what do you mean? Jun 06 01:00:26 wait for what? :) Jun 06 01:00:36 oe update? Jun 06 01:00:42 no, its my own feed Jun 06 01:00:55 Jin^eLD: check your "Packages" files if the never packages are indeed listed in it Jun 06 01:02:11 Jin^eLD: it doesn't matter. Jun 06 01:02:28 seems like it Jun 06 01:02:36 crink: oh, you mean there is a bug in OE? Jun 06 01:03:00 CoreDump|home: yep, listed Jun 06 01:03:16 Jin^eLD: all i can say, still oe un-flexible to me. Jun 06 01:03:20 then make sure ipkg update indeed pulls in the correct Packages file Jun 06 01:04:03 CoreDump|home: well, the ipkg.conf points to the repo, and if I do ipkg install somepackage that I do not yet have it will fetch it from the repo Jun 06 01:04:20 ipkg update anyway Jun 06 01:05:02 did that of course Jun 06 01:05:03 it will tell you where it stores the downloaded "Packages" file. Make sure it contains the newer packages Jun 06 01:05:26 upgrade should install the updated stuff if that is the case Jun 06 01:05:45 the package is there Jun 06 01:05:49 * crink build rootfs again, due to failure old rootfs with new minimo. Jun 06 01:05:56 can it be because of the versioning? allthough I did follow the OE guidelines Jun 06 01:06:00 s/failure/failure of/ Jun 06 01:06:07 what are the two versions? Jun 06 01:06:26 1.0.1+svn20070605150759-r0 old one, 1.0.1+svn20070605160818-r0 Jun 06 01:06:46 that should be ok, no? Jun 06 01:06:51 yeah Jun 06 01:06:58 the right one is newer Jun 06 01:07:25 yes, so why the hell is it not being upgraded to the right one? Jun 06 01:07:39 hard to tell from here Jun 06 01:07:55 CoreDump|home: why not? Jun 06 01:08:06 CoreDump|home: well, what else could I check? Jun 06 01:08:19 crink: because I left my crystal ball at work Jun 06 01:08:31 Jin^eLD: good question Jun 06 01:08:50 CoreDump|home: i don't know what you mean, anyway thx. Jun 06 01:08:52 both files have identical PACKAGE_ARCH? Jun 06 01:09:12 yes Jun 06 01:09:27 the .bb file that builds them simply has the ${DATETIME} thing in PV Jun 06 01:09:33 right Jun 06 01:09:35 so its the same .bb file, just checks out newest code from SVN and builds Jun 06 01:09:57 what does "ipkg install $package_name" say? Jun 06 01:11:01 what?? it says package with newer version installed in root is up to date Jun 06 01:11:07 but when I do ipkg list_insatlled | grep package name Jun 06 01:11:09 I get the old version Jun 06 01:11:11 oe need real "open(flexibility) i think. the gap, scratchbox. Jun 06 01:11:44 Jin^eLD: for whatever reason ipkg believes the new version is already installed =) Jun 06 01:12:00 CoreDump|home: makes no sense dammit :) what reason would that be ? Jun 06 01:12:15 maybe an aborted earlier installation attempt Jun 06 01:12:33 * crink had succeed to compile minimo with scratchbox(mipsel-foreign toolchain, without installing it). Jun 06 01:14:01 doh... I am fighting with the upgrade stuff for 2 days now Jun 06 01:14:07 I dont thinl I aborted anything Jun 06 01:14:20 Jin^eLD: cheer up Jun 06 01:14:24 how would I "start over"? Jun 06 01:14:34 is there a way to rebuild the ipkg database on the system Jun 06 01:14:47 Jin^eLD: not really Jun 06 01:15:00 Jin^eLD: ipkg remove $package_name --force-depends Jun 06 01:15:09 ipkg install $package_name Jun 06 01:15:20 tried that before Jun 06 01:15:27 03coredump2 07org.oe.dev * rd2a5a60f... 10/ (4 files in 3 dirs): zaurusd: Add the bl-on/off-via-hinge scripts to zaurusd as the bl packages has been assimilated by gpe-conf Jun 06 01:15:27 thats how the old version got on the system :) Jun 06 01:15:37 after that ran another build to get newer stuff Jun 06 01:15:45 and wondered why it did not get upgraded Jun 06 01:34:51 that makes no sense at all Jun 06 01:35:04 on server - generated the index, packages list has the new files Jun 06 01:35:16 on client checked - old files, did ipkg update - got the new file list from server Jun 06 01:35:21 ran ipkg upgrade - nothing to be done Jun 06 01:40:31 when it gets the old package from the package hash, it somehow gets the old version Jun 06 02:32:59 03coredump2 07org.oe.dev * rda6e5b87... 10/ (10 files in 4 dirs): usb-gadget-mode: Add usb-gadget mode: An initscript to load the configured USB gadget module on boot Jun 06 02:33:03 03coredump2 07org.oe.dev * r216711af... 10/ (1 packages/tasks/task-base.bb): task-base.bb: Add usb-gadget-mode to devices with usb-gadget support Jun 06 02:33:07 03coredump2 07org.oe.dev * r114006d8... 10/ (9 files in 6 dirs): usb-gadget-mode: Add default configuration (g_ether) for akita, spitz, tosa, c7x0 and collie Jun 06 02:37:37 03lenehan 07org.oe.dev * r60acb481... 10/ (11 files in 4 dirs): Jun 06 02:37:37 ipsec-tools: Add 0.6.7, removed all 0.4 versions and fixed for Jun 06 02:37:37 building with gcc 4.2. Jun 06 02:42:02 * * OE Bug 2390 has been RESOLVED (FIXED) by lenehan@twibble.org Jun 06 02:42:04 * *  "Full" Wget breaks ipkg Jun 06 02:42:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2390 Jun 06 02:42:13 * * OE Bug 2421 has been created by oe@hentges.net Jun 06 02:42:16 * * gpe-conf usb is utterly broken Jun 06 02:42:17 * * http://bugs.openembedded.org/show_bug.cgi?id=2421 Jun 06 02:43:31 night Jun 06 02:44:36 what's the next step, after finding a segfault, and getting the output from both strace and gdb? **** ENDING LOGGING AT Wed Jun 06 02:59:58 2007