**** BEGIN LOGGING AT Wed May 30 02:59:57 2007 May 30 03:02:50 Zero_Chaos: bitbake version is 1.8.x, against .dev May 30 03:17:27 can someone point me in the right direction? http://pastebin.ca/520964 May 30 05:41:52 goodmorning May 30 05:51:32 steliosk: morning May 30 06:40:12 Anyone here have angstrom-minimal-image building with OABI? May 30 07:09:48 03lenehan 07org.oe.dev * r4aa7a6a4... 10/ (3 files in 2 dirs): May 30 07:09:48 perl/perl-native 5.8.8: Change the installation directories used for May 30 07:09:48 perl-native to remove the various directories and to remove the host May 30 07:09:48 specific path components. This is to make it easier to refer to the staged May 30 07:09:48 perl location from other places, such as when trying to tell things where May 30 07:09:48 the perl include files are located. May 30 07:09:52 03lenehan 07org.oe.dev * r3fde4b2f... 10/ (4 files in 4 dirs): (log message trimmed) May 30 07:09:54 perl 5.8.8: Patch MakeMaker in perl-native to check for PERL_INC in the May 30 07:09:56 environment and use that instead of the configured location of PERL_INC. May 30 07:09:58 Without this PERL_INC for non-native recipes ends up pointing at where the May 30 07:10:00 headers will be located on the target. In theory we could override this in May 30 07:10:02 cpan class when calling build.PL but for some packages, such as May 30 07:10:04 libxml-parser-perl, that only fixes the top-level makefile and is not May 30 07:10:08 03lenehan 07org.oe.dev * rd6cea9d9... 10/ (3 files in 3 dirs): May 30 07:10:10 perl-native 5.8.8: Fix an issues with the perl makedepend script not May 30 07:10:12 properly terminating a quote. This is in theory the fix for the /bin/sh -> May 30 07:10:14 /bin/dash problems with perl, however since I don't have any issues here with May 30 07:10:16 dash (ubuntu 6.10) I can't confirm this. The patch is an obvious fix though May 30 07:10:18 so I'm adding it. May 30 07:10:20 03lenehan 07org.oe.dev * r8a4d576b... 10/ (44 files in 2 dirs): May 30 07:10:24 perl 5.8.8: Bump PR on modules to force the -native versions to be rebuilt May 30 07:10:26 to take into account the new staging location of perl. This will also force May 30 07:10:28 a rebuild of libxml-parser-perl which should now work. May 30 07:34:22 good morning all May 30 07:47:53 hi i solved kernel do_stage problem it turned out my fault, linux-2.6.21/include/asm(this should be symlink) May 30 07:48:17 and had another problem May 30 07:48:39 Collected errors:| Cannot find package task-base. May 30 08:05:21 When packaging my filetree into a ipkg with the default bb class, empty directories are removed. Is there an option to set in my .bb file to circumvent this? May 30 08:06:01 * * OE Bug 2116 has been REOPENED by happyarch@gmail.com May 30 08:06:03 * *  db1200 machine description broken [...] May 30 08:06:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2116 May 30 08:22:57 I compiled a static hello world app using both sbox and gcc 4.1.2 , they work fine on the 770 and N800 but fail on my s3c2440 based system May 30 08:23:57 morning May 30 08:24:40 when I strace the program (I don't have many different tools here yet) I get execve("/mnt/sdcard/tmp/init.std", ["/mnt/sdcard/tmp/init.std"], [/* 49 vars */]) = 0 uname({sys="", node="", ...}) = -1098057662 --- SIGSEGV (Segmentation fault), binaries generated by sbox have uname({sys="Linux", node="(none)", ...}) = 0 May 30 08:26:43 is that sys compiled into the file or is it given by the env? May 30 08:33:57 hey XorA & keesj May 30 08:50:25 * CoreDump|afk reads "Comment #5 from Rolf Leggewie" to bug 1573: Guess which DISTRO the guy's been using? =D May 30 08:55:38 Damn these cl0n3rz May 30 08:55:41 morning May 30 08:55:42 * CoreDump|afk chuckles May 30 08:55:49 morning hrw May 30 09:01:04 has anyone looked at this http://www.piro.hopto.org/~piro/zaurus/bootloader/patch-2.6.16.16-z.diff May 30 09:01:42 koen: only one additional patch is needed, but it will get landed soon May 30 09:02:02 * * OE Bug 1879 has been RESOLVED (FIXED) by lenehan@twibble.org May 30 09:02:04 * *  libxml-parser-perl build failing [...] May 30 09:02:06 * * http://bugs.openembedded.org/show_bug.cgi?id=1879 May 30 09:02:14 koen: Laibsch: the WebKit Bugzilla has a concept of review. On attachments people can set review:? and then people can do review:+ or review:- May 30 09:02:26 koen: on webkit.org/pending-patches you can see the patches needing review May 30 09:02:33 XorA: what it gives? May 30 09:03:39 zecke: 404 May 30 09:03:55 /pending-review May 30 09:06:09 zecke: I don't see a patch for the pcre problem May 30 09:06:52 koen: no, there is no problem besides using #include "" May 30 09:07:25 koen: the issue is that we have -I/staging/include/ before the local -IJavaScriptCore/kjs/pcre directories May 30 09:07:29 morning all May 30 09:07:34 hey RP May 30 09:07:47 RP hi May 30 09:07:48 moring RP May 30 09:08:09 zecke: dare I ask why kjs has its own pcre? May 30 09:08:22 or is that covered by the 'k' in kjs? May 30 09:08:45 koen: to work with unicode strings May 30 09:09:01 * * OE Bug 413 has been RESOLVED (FIXED) by lenehan@twibble.org May 30 09:09:03 * *  sdl-perl installs to incorrect location [...] May 30 09:09:05 this is why you see the compile error May 30 09:09:05 * * http://bugs.openembedded.org/show_bug.cgi?id=413 May 30 09:09:39 hrw: alternative to altboot in kernel space it looks like May 30 09:13:02 * * OE Bug 1066 has been RESOLVED (FIXED) by lenehan@twibble.org May 30 09:13:04 * *  Cross Compile Badness in compiling sdl-perl [...] May 30 09:13:06 * * http://bugs.openembedded.org/show_bug.cgi?id=1066 May 30 09:20:20 XorA: I think it would be better to fix kexec-tools to support commandlines for arm. May 30 09:22:37 * CoreDump|afk could pretty much live w/ both May 30 09:23:16 CoreDump|home, morning May 30 09:23:23 any progress regarding gspca? May 30 09:24:05 thesing: probably May 30 09:24:20 thesing: the modules just looked interesting and I wondered if anyone had tried it May 30 09:24:59 dcordes: driver loades and detects camera. I couldn't verify if it actually works yet as mplayer in the feed is lacking the ability to do so May 30 09:25:31 CoreDump|home: patches welcome May 30 09:25:44 XorA: on first look it seems to be the way the openbsd people boot their kernel on zaurus. May 30 09:25:48 CoreDump|home, can I help? :) May 30 09:26:00 XorA: For now I wrote a bug report May 30 09:26:18 CoreDump|home: I saw, maybe if my new RAM works I can look tonight May 30 09:26:34 * XorA wishes mplayer was modular May 30 09:26:43 dcordes: sure, see bug 2400 May 30 09:26:55 Laibsch: how does cjkv IM situation looks for gtk/x11 environment? May 30 09:26:55 XorA: that would be most welcome =) May 30 09:27:05 XorA: not only you May 30 09:27:28 hrw: I thought you OH guys were all gstreamered up these days May 30 09:27:30 mickeyl: morning May 30 09:27:42 XorA: I use mplayer on desktop May 30 09:27:43 03koen 07org.oe.dev * r7b962d35... 10/ (4 files in 3 dirs): empathy: sync with poky May 30 09:27:49 XorA: the .bb file disables a ton of switches for ./configure, I don't believe this is a sane thing to do for a generic package. May 30 09:27:50 or kaffeine (for dvds) May 30 09:28:09 CoreDump|home: it is sane, I went through and did every one of those on purpose May 30 09:28:50 XorA: even embedded devices can make use of some of the more uncommon functions =) May 30 09:28:51 03lenehan 07org.oe.dev * r6eac237f... 10/ (1 packages/sdlperl/sdl-perl_1.20.3.bb): (log message trimmed) May 30 09:28:51 sdl-perl 1.20.3: Fix the build of sdl-perl May 30 09:28:51 * Use the cpan class to do all the work instead of manually trying to do May 30 09:28:51 what cpan.bbclass does. May 30 09:28:51 * Add some sed to stop it searching the system paths for SDL headers and to May 30 09:28:52 work around it expecting smpeg.h to be in an smpeg subdirectory. May 30 09:28:56 * Update the homepage URL. May 30 09:29:20 CoreDump|home, is it mplayer's fault not detecting v4l or rather the present v4l system being incomplete? May 30 09:29:33 CoreDump|home: you either have to a) disable feature, or b) enable feature and add depends, I wont have different people getting different mplayer packages by luck May 30 09:29:46 By the looks of it, it is a broken configure script May 30 09:30:55 CoreDump|home: so I am happy if you make sure it always works and enable the feature May 30 09:31:37 that would be a difficult thing to do for many of these May 30 09:31:47 XorA: is http://ewi546.ewi.utwente.nl/tmp/mplayer-sync-svn-with-release.diff ok to commit? May 30 09:37:33 RP, hrw: libxml-parser-perl should build and work now.. it used to build but didn't work (.so's were for the host), hasn't built for a while - didn't think anyone would be using it much since it has never actually worked and I don't see anything in OE that depends on it. May 30 09:37:49 v8jlene: I am building it now in poky, oe/glibc oe/uclibc May 30 09:37:56 NOTE: package libxml-parser-perl-2.34-r9: task do_build: completed May 30 09:37:58 poky May 30 09:38:22 Expat.so: ELF 32-bit LSB shared object, ARM, version 1 (SYSV), stripped May 30 09:38:32 v8jlene: THX a lot May 30 09:38:46 intltool on device will finally work May 30 09:38:49 v8jlene: I'd noticed the problem when I tried a uclibc build but hadn't worked out how to fix it. Having a working version is much appreciated! :) May 30 09:39:14 koen: looks fine May 30 09:39:21 It's been on my list for months to fix... some talk about it in the IRC logs finally got me around to looking into it ;) May 30 09:39:32 v8jlene: I will add it to sdk image in poky now it stands a chance of working :) May 30 09:40:01 -I/home/hrw/devel/OH/poky/build/tmp/staging/i686-linux/lib/perl/5.8.8/CORE - should it use host includes during build? May 30 09:40:56 v8jlene: it failed for angstrom/uclibc ;( May 30 09:41:00 | /home/hrw/devel/build/uclibc-angstrom/tmp/staging/i686-linux/lib/perl/5.8.8/CORE/reentr.h:612: error: field '_crypt_struct' has incomplete type May 30 09:41:05 hrw: Includes are indentical for -native and host... so I use the one lot of headers May 30 09:41:11 v8jlene: With uclibc builds, I did notice that it had failures since my host has glibc which has a recursive crypt function whereas uclibc doesn't May 30 09:41:23 v8jlene: ok May 30 09:42:04 v8jlene: That error is because it thinks it has a recursive crypt which uclibc hasn't May 30 09:42:10 hrw: Ah damn it... I'd fixed that thing in a few places for uclibc. My uclibc build hasn't caught up yet. I'll fix that it for uclibc tommorow... glibc was the aim for today. May 30 09:42:26 ok May 30 09:43:15 03koen 07org.oe.dev * r166a4964... 10/ (1 packages/mplayer/mplayer_0.0+1.0rc1.bb): mplayer 1.0rc1: sync with svn version for pxa optimizations May 30 09:44:02 Hello May 30 09:44:33 I just ran mtn update and it tells me that it has "multiple update candidates" May 30 09:44:46 Specifically: May 30 09:44:51 mtn: 6eac237f0d909ba98b365d88b337ff3e822907e1 lenehan@openembedded.org 2007-05-30T09:00:31 May 30 09:44:52 mtn: 7b962d3510d731aaa784edc4ce4ccd71427cbf5f koen@openembedded.org 2007-05-30T09:35:41 May 30 09:45:07 Should I do the koen one? May 30 09:45:33 Maybe this is a bug or something May 30 09:45:52 B_Lizzard: maybe you should read the monotone docs May 30 09:46:05 :) May 30 09:46:07 B_Lizzard: select any May 30 09:46:19 and read monotone manual May 30 09:46:20 I've never seen this before, that's why I'm asking you May 30 09:46:26 Maybe it was a bug or something May 30 09:46:35 I'll just choose the most recent one May 30 09:46:51 hrw: currently there is none. I was hoping a bit for the openmoko people. uim is in and compiles but the package apparently does nothing now and I have had no time to fix it. recently, freak_nl gave me a bb form scim that he says compiles for him (in 4 minutes on a 2GHz machine). It has been doing something for 24h on my slow machine now without finishing. I think something in it is broken. scim has a gtk bother called skim. May 30 09:46:59 hrw: But currently: nothing May 30 09:47:18 ok May 30 09:49:30 Should I run mtn merge, then? May 30 09:49:44 :/ May 30 09:49:48 B_Lizzard, just repull May 30 09:49:58 Aha, OK May 30 09:50:00 Thanks May 30 09:51:32 mickeyl: o qmake guru, how can I reorder -I statements when compiling? May 30 09:53:46 03polyonymous 07org.oe.dev * r0e01e62b... 10/ (3 files in 3 dirs): gcc: Make gcc build on amd64 by having it forget target configure cache when configuring for build machine. Closes bug #1951 May 30 09:54:01 * * OE Bug 1951 has been RESOLVED (FIXED) by polyonymous@klever.net May 30 09:54:04 * *  gcc fails to build on x86_64 [...] May 30 09:54:05 * * http://bugs.openembedded.org/show_bug.cgi?id=1951 May 30 09:54:32 mickeyl: http://pastebin.ca/521979 May 30 09:54:34 ah.. #1951... May 30 09:56:01 I hope it doesn't break anything for anyone (it shouldn't) May 30 09:57:58 polyonymous: it should not May 30 09:58:08 I tested that patch on my box May 30 09:58:36 Me too ;-) May 30 09:59:20 koen: can you bump angstrom to use xrandr 1.2? May 30 10:01:01 * * OE Bug 2401 has been created by koen@dominion.kabel.utwente.nl May 30 10:01:03 * * Webkit fails to build if pcre*.h is present in staging May 30 10:01:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2401 May 30 10:09:04 there is a line in openssl.inc which recently got changed from: May 30 10:09:06 export EX_LIBS = "-lgcc -ldl -L${STAGING_LIBDIR}" May 30 10:09:19 rather... to the above May 30 10:09:26 from: export EX_LIBS = "-lgcc -ldl ${TARGET_LDFLAGS}" May 30 10:09:47 the new version does not work when using a pre built toolchain May 30 10:10:00 changing it back fixes... anyone know why this was chnged? May 30 10:10:33 tkp: Take a look at who made the change and ask him May 30 10:10:46 Seems to be the best and fastest solution May 30 10:11:19 mtn log --diffs $whateverfile May 30 10:11:35 Laibsch: where would I find that out? All I know is that it was done in the last 3 months May 30 10:11:37 ah. k May 30 10:11:38 thnks May 30 10:13:35 mtn annotate is easier and faster May 30 10:14:18 koen: seems it was you? May 30 10:14:41 your note just says "openssl: fix FILES, spotted by Junqian Gordon Xu" May 30 10:15:25 i'll file a bug May 30 10:16:31 fix FILES doesn't sound exactly like changing EX_LIBS. May 30 10:16:52 no. but it got done in that commit May 30 10:18:23 the EX_LIBS statement hasn't been changed in years May 30 10:19:10 see http://ewi546.ewi.utwente.nl/tmp/openssl-annotated.txt May 30 10:19:32 and "a9e89996.. by koen 2005-06-30" is when I imported our bk tree in monotone May 30 10:20:56 interesting... ok then I know what happened May 30 10:21:02 * * OE Bug 2402 has been created by stv@kirkconsulting.co.uk May 30 10:21:04 * * Recent change to openssl.inc breaks compilation when May 30 10:21:06 * * http://bugs.openembedded.org/show_bug.cgi?id=2402 May 30 10:21:11 I think I must have made this change in my local checkout some time ago May 30 10:21:23 but blitzed it with a recent OE update May 30 10:21:53 koen: i can't figure out this error message: Cannot find package task-based   May 30 10:21:55 either way... changing to my version fixes when using external toolchain May 30 10:22:24 machine feature seems fine.. May 30 10:33:50 STK1000 is 532USD May 30 10:34:20 hi hrw May 30 10:34:58 AT91SAM9263-EK is even more priced May 30 10:35:09 hrw: I know :) May 30 10:35:16 995USD May 30 10:35:31 9263EK is good base for slug replacement ;D May 30 10:36:51 koen: can you bump angstrom to use xrandr 1.2? May 30 10:36:53 Hey, did you guys see that Yoggie Pico thing on linuxdevices? May 30 10:37:10 hrw: didn't I already do that? May 30 10:37:31 They crammed a 520MHz Xscale PXA270 CPU, 128MB RAM and 128MB ROM in a USB stick May 30 10:37:33 :/ May 30 10:38:11 * XorA watches to see if koen explodes May 30 10:38:13 koen: I got 1.0.2 built for uclibc build May 30 10:38:14 It boots Linux 2.6, and acts as a firewall May 30 10:38:40 hrw: I suspect CIA is a bit lagging behind :) May 30 10:39:07 yes, it is. May 30 10:40:02 03hrw 07org.oe.dev * r917b0d82... 10/ (4 files in 3 dirs): xrandr: fix uclibc build (from Poky) May 30 10:40:11 03polyonymous 07org.oe.dev * r134a9599... 10/ (4 files in 3 dirs): opie-reader: fix copying of unicode text blocks, since upstream isn't likely to incorporate this one-liner patch anytime soon. May 30 10:40:19 03koen 07org.oe.dev * r004380a4... 10/ (1 conf/distro/angstrom-2007.1.conf): angstrom: bump xrandr to 1.2.x May 30 10:40:26 03polyonymous 07org.oe.dev * rd355cc8d... 10/ (3 files in 3 dirs): May 30 10:40:26 opie-multikey: Let it load ru.keymap even with utf8 locale set by converting comments into utf-safe encoding. May 30 10:40:26 Again, can't wait for upstream for a couple of years more. May 30 10:40:30 ah May 30 10:40:34 03koen 07org.oe.dev * r202111bb... 10/ (1 packages/webkit/webkit_svn.bb): webkit: unbreak qmake usage a bit May 30 10:40:49 koen: unbreak icu for webkit :-) May 30 10:41:24 :) May 30 10:41:40 * XorA cant install icu on debian May 30 10:42:17 http://www.mvista.com/vision/welcome.html - do we want to go there and lart them? May 30 10:43:01 hrw, but it's on the wrong side of the globe :) May 30 10:43:27 we should give a talk :) May 30 10:43:34 santa clara is an awfull place May 30 10:43:46 thank $deity for SF being close May 30 10:43:52 heh May 30 10:44:00 sourceforge? :) May 30 10:44:00 California is a dreaful place May 30 10:44:06 run by an actor May 30 10:44:20 Crofton, for what I know California is a great place to live in if you happen to be an orange. May 30 10:44:33 the people are all odd May 30 10:44:38 polyonymous: not if you hate getting squeezed May 30 10:44:43 Crofton: you are all odd to us May 30 10:44:53 my rental car had a warning that people likes to steal this car May 30 10:44:57 XorA, dunno if I do, I don't happen to be an orange. May 30 10:45:10 Crofton: you can buy "The Governator" shirts at the airports.... May 30 10:45:16 Crofton, well, they have to have hobbies there. May 30 10:45:22 there was a warning that people do not stop going through traffic lights until after they go read May 30 10:45:25 er red May 30 10:45:26 Crofton: and your country was run by an actor for 8 years :-) May 30 10:45:38 don't remind me May 30 10:45:58 XorA: that one atleast was from US :D May 30 10:46:03 Crofton: and we only have incompetent swizzle sticks, feel yourself lucky May 30 10:46:54 any country where a wrestling superstar can become governer rocks by me May 30 10:52:04 yay! May 30 10:52:08 webkit compiled May 30 10:52:25 * koen moves on to packaging and staging May 30 10:54:43 RP: how's 2.6.21 doing on your machines? May 30 10:56:44 koen: with recent fixes it works May 30 10:59:10 what's the best fs to have oe build environment on? May 30 10:59:36 polyonymous: ext3, xfs May 30 10:59:39 a big one ;-) May 30 10:59:42 even reiserfs3 May 30 10:59:43 anything that doesn't contain 'reiser' of 'fat' May 30 10:59:44 NTFS? May 30 10:59:49 :) May 30 11:00:12 damn, that many answers I could give myself :) Just thought you have a kind of consensus on it :) May 30 11:00:30 I have it on ext4 now. May 30 11:00:37 no-one said a Mac filesystem, so that's consensus. May 30 11:00:55 it works on HFS+ as well May 30 11:01:39 so, no one noticed that any fs performs better than others. Well, that's okay with me. I'm only asking because I'm doing rm my old tmp now and it takes time :) May 30 11:02:00 can i skip the "task-base"? May 30 11:02:40 or just ipkg install *.ipk -o / is fine? May 30 11:04:06 polyonymous: jfs is good for delete quickly May 30 11:04:33 mv tmp tmp.del; rm -rf tmp.del& ? May 30 11:04:59 Crofton, I've moved a while ago, now I'm just deleting it :) Yes, I can ^Z bg May 30 11:05:00 :-? May 30 11:05:36 nauj27, Might be a good choice, indeed. I've had troubles with it as my root filesystem when I had lockups data was corrupted, but for OE built filesystem that should do. May 30 11:05:49 s/built/build/ May 30 11:07:36 I'm installing oe environment on NTFS with cygwin just now ^_^ May 30 11:07:47 * CoreDump|home wonders why changes /etc/gpe/* are ignored by GPE... May 30 11:08:00 nauj27, seen psychiatrist lately? :) May 30 11:08:43 LOL ^_^U May 30 11:10:17 polyonymous: it's the work computer with winxp only... and I think should be a lot faster than vmware (if it work..) May 30 11:10:28 nauj27: colinux is not known for you? May 30 11:10:52 crink: why you want to skip task-base? May 30 11:10:59 koen: It seems to work well apart from on qemuarm May 30 11:11:02 * nauj27 is now googling... May 30 11:11:08 hrw: nup May 30 11:11:15 nauj27, do you consider running xp in vmware as well? :) May 30 11:11:52 grrr May 30 11:12:02 hrw: http://projects.linuxtogo.org/pipermail/openembedded-issues/2007-April/002212.html May 30 11:12:49 polyonymous: I think on it, but it's original xp adapted for this toshiba model... May 30 11:13:17 great features of this operating system (ironic) May 30 11:14:21 crink: it will take some time but I will check it May 30 11:15:14 hrw: thx May 30 11:15:57 arm platform so easier than mipsel platform. May 30 11:16:05 in my experience May 30 11:16:56 crink: more common so better tested May 30 11:21:31 6.4M armv5te/libwebkitgdk1_0.0+svn20070530-r0_armv5te.ipk May 30 11:23:42 03koen 07org.oe.dev * rd837d15b... 10/ (1 conf/machine/db1200.conf): db1200: remove PACKAGE_ARCHS May 30 11:29:48 03koen 07org.oe.dev * r4181e4ee... 10/ (1 packages/webkit/webkit_svn.bb): webkit: add install and packaging, still needs do_stage() May 30 11:36:24 03hrw 07org.oe.dev * r3b2a4c2d... 10/ (1 packages/perl/libxml-parser-perl-native_2.34.bb): libxml-parser-perl-native: fix RDEPENDS (from Poky) May 30 11:37:09 * koen does a little dance: http://www.angstrom-distribution.org/repo/?action=details&pnm=libwebkitgdk-gdklauncher May 30 11:38:19 * koen wonders where all those spurious -dev dependencies come from May 30 11:49:26 hrw: What does an RDEPENDS on a -native package do? Isn't DEPENDS enough to get whatever it needs built? May 30 11:49:49 v8jlene: enough to build - sure. May 30 11:50:07 v8jlene: but it include libxml-parser-perl so gets RDEPENDS="perl" May 30 11:51:49 hrw: It's a native package though... I thought RDEPENDS were only related to packing and installing packages.. (which you don't do with -native). May 30 11:52:57 Guess I need to re-think that thinking... May 30 11:53:26 v8jlene: it is native indeed May 30 11:53:55 v8jlene: that RDEPENDS was added to keep bitbake quiet so it will not build perl for libxml-parser-perl-native May 30 11:56:35 v8jlene: Ah, ok, because bitnake builds all RDEPENDS... yep that explains it. I'll look at fixing that in cpan.bbclass so it's global fixed for all perl modules. Thanks. May 30 11:56:46 thx May 30 12:24:10 RP: do you know the kexec-arm internals? May 30 12:24:47 Hello; any idea when/whether will starling ipk's get fixed? May 30 12:25:00 ERROR: QA issue: WebKitGdk.pc failed sanity test (reference to workdir) May 30 12:30:01 hrw: re perl - "the headers are the same"... yeah sure... I'm an idiot. Staging and using perl's headers fixes libxml-parser-perl for uclibc... bit more testing and I'll push the fix out now. May 30 12:32:22 ok May 30 12:32:41 mmp: what is wrong with it now? May 30 12:33:30 hrw: unsatisfiable dependencies... May 30 12:33:47 mmp: which ones? May 30 12:34:26 hrw: http://www.angstrom-distribution.org/repo/?action=details&pnm=starling May 30 12:34:27 mmp: starling 0.1 or svn? one of them was fixed some time ago May 30 12:34:30 hi Cliff May 30 12:34:35 ah.. package built. May 30 12:35:04 just lots of package on which starling depends don't exist... May 30 12:35:54 s/lots/4/ May 30 12:36:42 I filed also a bug:) http://bugs.openembedded.org/show_bug.cgi?id=2129 May 30 12:37:25 !oebug 2129 May 30 12:37:27 * * Bug 2129, Status: NEW, Created: 2007-04-21 10:37 May 30 12:37:28 * * milan.plzik(AT)gmail.com: starling depends on non-existing packages May 30 12:37:29 * * http://bugs.openembedded.org/show_bug.cgi?id=2129 May 30 12:37:39 hrw: are you there May 30 12:37:56 yes May 30 12:37:57 mickey|lunch: May I ask you a brief question about your alice experience in private chat? May 30 12:38:38 i have seen different error now May 30 12:38:48 ERROR: Cannot satisfy the following dependencies for task-base: May 30 12:38:48 | kernel update-rc.d update-rc.d update-rc.d update-alternatives update-rc.d update-rc.d task-distro-base update-rc.d task-base-kernel26 prism3-firmware prism3-support spectrum-fw hostap-conf task-base-usbhost task-base-nfs task-base-ppp May 30 12:39:40 /stuff/tmp/deploy/glibc/ipk May 30 12:39:42 crink: show me "ls tmp/deploy/*/ipk" May 30 12:39:52 Packages Packages.filelist Packages.gz Packages.stamps all db1200 i686 mipsel May 30 12:40:42 i cp all/* db1200/ and cp all/* mipsel/ May 30 12:40:51 DO NOT DO THAT May 30 12:40:59 :) May 30 12:41:08 or we will punish you May 30 12:41:24 hrw: to solve problem May 30 12:41:32 i have no choice May 30 12:41:34 thats not solution May 30 12:42:17 hrw: ah one thing, why oe choose kernel 2.4 at first in db1200? May 30 12:42:19 * XorA gueses machine.conf doesnt list the package arches May 30 12:42:40 crink: does db1200 has kernel in OE? May 30 12:42:46 angstrom, and db1200 May 30 12:43:15 angstrom-2007.1.conf May 30 12:43:46 /stuff/org.openembedded.dev/packages/linux May 30 12:44:04 linux-h6300-omap1-2.6.12-rc5 linux-h6300-omap1_2.6.12-rc5.bb linux-mtx-1-2.4.24 linux-mtx-1u_2.4.27.bb linux-mtx-3-2.6.15 linux-mtx-3_2.6.15.bb May 30 12:44:05 linux-h6300-omap1-2.6.14.3 linux-h6300-omap1_2.6.14.3.bb linux-mtx-1-2.4.27 linux-mtx-2-2.4.27 linux-mtx-3-2.6.15.4 linux-tornado-omap2 May 30 12:44:05 linux-h6300-omap1-2.6.16.16 linux-h6300-omap1_2.6.16.16.bb linux-mtx-1_2.4.27.bb linux-mtx-2_2.4.27.bb linux-mtx-3_2.6.15.4.bb linux-tornado-omap2_2.6.16.16.bb May 30 12:44:20 crink: ARGH! May 30 12:44:23 ~pastebin May 30 12:44:35 ok May 30 12:44:35 and we know that packages/linux dir exists May 30 12:44:39 extra, extra, read all about it, pastebin is 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 May 30 12:44:46 next time May 30 12:45:11 I don't blog much, but how do I get http://blog.leggewie.org/?cat=6 accepted? May 30 12:45:14 angstrom choose these kernel, so i remove it from unused dir May 30 12:45:24 /s/from/to/ May 30 12:45:41 Laibsch: wordpress? May 30 12:46:39 hrw: problem is although it should have to choose 2.6, but it choose wrong(2.4. and other) kernel.. May 30 12:46:41 Laibsch: accepted on planet.linuxtogo.org? May 30 12:46:57 crink: submit working kernel recipe/patchset for db1200 May 30 12:47:47 !oebug 1791 May 30 12:47:48 hrw: my mipsel kernel is for customized au1200, not the reference board, so it will be the mass. May 30 12:47:49 * * Bug 1791, Status: NEW, Created: 2007-01-15 07:18 May 30 12:47:50 * * wb_viva(AT)yahoo.com.cn: db1200 lack kernel information May 30 12:47:51 * * http://bugs.openembedded.org/show_bug.cgi?id=1791 May 30 12:48:17 crink: then create machine config for your board and give us kernel for it May 30 12:48:24 hrw: Yes, wordpress. How do I make that category of my blog appear on planet.ltg.org? May 30 12:48:44 hrw: what is the machine config? you mean kernel .config? May 30 12:48:48 Laibsch: join #linuxtogo and request it? May 30 12:49:21 crink: oe/conf/machine/crink-strange-machine-based-on-au1200-but-different-enough-to-need-own-config.conf May 30 12:50:37 hrw: OK May 30 12:51:17 hrw: not so much different, problem is default oe mipsel kernel is compiled well, to correct the compile error, it takes time.. May 30 12:51:28 s/is/isn't/ May 30 12:52:14 FYI, minimo mipsel is not working atm, May 30 12:52:27 though it build well. May 30 12:53:20 i built my own, i have to test it with new rootfs, so i need the rootfs. May 30 12:54:55 03lenehan 07org.oe.dev * r077d4c67... 10/ (4 files in 3 dirs): (log message trimmed) May 30 12:54:55 perl 5.8.8: Fixes for uclibc and libxml-parser-perl. May 30 12:54:55 * Stage the headers from perl May 30 12:54:55 * Modify cpan.bbclass to use the appropriate headers. May 30 12:54:55 * Remove prototypes as well as disabling functions that uclibc doesn't May 30 12:54:56 support. May 30 12:54:58 * Bump PR on libxml-parser-perl to force it to be rebuilt. May 30 12:55:49 * crink built gtk webcore 1 months ago May 30 13:03:51 v8jlene: I will test it May 30 13:09:23 nauj27: You're welcome ;-) May 30 13:09:45 crink: so does it use mainline 2.6.21? May 30 13:10:25 hrw: Ok. Does this look sane: http://pastebin.ca/522510 - so for native modules I only add the DEPENDS on perl-native, non-native get DEPENDS on perl and perl-native and RDEPENDS on perl. May 30 13:10:37 hrw: Appears to work as expected. May 30 13:11:02 * * OE Bug 2305 has been RESOLVED (FIXED) by polyonymous@klever.net May 30 13:11:04 * *  pidgin dependencies may need some love. [...] May 30 13:11:05 03polyonymous 07org.oe.dev * r4be62db5... 10/ (1 packages/gaim/pidgin.inc): pidgin: repackaged for sane interdependencies handling. Closes #2305 May 30 13:11:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2305 May 30 13:11:06 hrw: yes May 30 13:13:10 crink: then give me defconfig for it May 30 13:14:27 Laibsch: :) May 30 13:14:52 v8jlene: do not remove expat-native from libxml-parser-perl-native please May 30 13:15:22 v8jlene: and non-native perl stuff need to DEPEND only on perl (which depend on perl-native) May 30 13:16:06 crink: if machine does not have defined preferred kernel then OE choose any kernel for it May 30 13:17:53 hrw: i can't send defconfig to you May 30 13:18:05 maybe port is blocked.. May 30 13:18:21 crink: do it via OE bugtracker May 30 13:18:27 I do not use dcc May 30 13:18:46 hrw: libxml-parser-perl adds a DEPENDS on expat-native, so that dependency is already handled May 30 13:18:51 hrw: OE bugtracker and where? May 30 13:20:40 hrw: can i make bugreport? May 30 13:21:00 crink: You never have to ask for permission to write a bug report! May 30 13:21:38 * XorA denies crink permission May 30 13:21:50 there thats bettern we have no bugs :-) May 30 13:21:58 Laibsch: i don't have to :-) May 30 13:22:04 scareilly I have had to work with project managers that did that May 30 13:23:34 * crink think defconfig doesn't make any solution May 30 13:25:28 v8jlene: ok May 30 13:26:02 * * OE Bug 2403 has been created by hma@syd.odn.ne.jp May 30 13:26:04 * * minimo build process needs libidl-dev on host side. May 30 13:26:06 * * http://bugs.openembedded.org/show_bug.cgi?id=2403 May 30 13:26:47 ah, more bugs from Japan! May 30 13:26:49 Nice! May 30 13:26:56 hrw: i don't care about working zImage, only i need rootfs, so the install process without problem is enough to me. May 30 13:27:15 crink: building rootfs need also kernel in OE May 30 13:27:33 Laibsch: more occassions for japanese chars? May 30 13:27:36 hrw: i am using nfs and minicom now May 30 13:27:45 nfs root boot and kernel boot. May 30 13:28:00 no, just referring to 2403 which is from Hamazaki-San May 30 13:28:18 crink: Are you Hamazaki-San? Your IP is Korean. May 30 13:28:26 old rootfs using libc 2.3.90 this is the problem May 30 13:28:34 Laibsch: yes May 30 13:28:38 i am south korean May 30 13:29:09 * crink don't like japanese personally May 30 13:29:59 Well, I know. A lot of tension in Asia. I hope you guys can work it out. May 30 13:30:23 crink: What Korean input methods are most popular/suitable for an embedded device? May 30 13:30:40 I'd like to add them to OE May 30 13:30:44 !oebug 1094 May 30 13:30:46 * * Bug 1094, Status: ASSIGNED, Created: 2006-06-17 11:55 May 30 13:30:47 * * bugs.openembedded.org(AT)rolf.leggewie.biz: Meta-bug: CJKV for Opie/GPE May 30 13:30:48 * * http://bugs.openembedded.org/show_bug.cgi?id=1094 May 30 13:30:51 Laibsch: offtopic but say something, russian and usa is same situation i know May 30 13:31:00 Laibsch: scim scim-hangul May 30 13:31:17 crink: I am using tftp to upload kernel to board and boot from MMC or download kernel/rootfs and store in nand May 30 13:31:23 Laibsch: i wrote a patch for matchbox-keyboard for hangul May 30 13:31:46 crink: does that patch was sent to OpenedHand? May 30 13:31:51 crink: Not really, their situation is getter better while inner-Asia seems to get worse. And of course China-US is going to get worse, unluckily. May 30 13:31:52 hrw: yes May 30 13:32:21 crink: I am working on scim in general. Or to be more correct, freak_NL did most of the work May 30 13:32:53 Laibsch: you are wrong, problem is usa, in 200years ago they start to origianl indian, see http://www.freepeltier.org May 30 13:33:05 s/to/to kill/ May 30 13:33:22 see you all! May 30 13:33:30 s/origianl indian/original american/ May 30 13:33:35 crink: It seems compilation never finishes for me which is why I wanted to ask hrw if it compiles for him. Just a minute and I'll put up a diff you can both apply. May 30 13:34:23 crink: Have you sent the patch upstream? I think o-hand is doing the work on matchbox, right? May 30 13:34:28 white ppl never free from racism... May 30 13:34:48 crink: That is clearly a racist remark, right there ;-) May 30 13:34:52 or :-( May 30 13:35:29 Laibsch: i don't care about racism, it just big fish and small fish world, jewish mind controlling use May 30 13:35:34 /s/use/usa/ May 30 13:35:42 jungle May 30 13:36:02 crink: Let's try and keep the politics and history out of this channel. I think we are on slippery ground. May 30 13:36:27 And we don't have to agree with our historical and contextual analysis May 30 13:36:48 Laibsch: yes May 30 13:38:04 we can all agree to hate the country to our left :-) May 30 13:39:10 XorA: What country would that be for me? Maybe the one you are in? May 30 13:39:15 :-) May 30 13:39:29 Laibsch: for me I think that is iceland or canada May 30 13:39:48 Laibsch: but I cant quite remeber which way is north (and being scotland I cant see the sun May 30 13:39:58 Well, if you go that far, any country is to your left ;-) May 30 13:40:24 Laibsch: rule is go left, whatever country you hit first in a straight line May 30 13:40:41 and what is left? May 30 13:40:44 Laibsch: i sent my patch to mallum http://www.advogato.org/person/mallum May 30 13:40:47 Wouldn't that be Ireland? May 30 13:40:49 so if I spin around then its Netherlands :-) May 30 13:41:10 Laibsch: you personally, not your countries left May 30 13:41:29 crink: How long ago? Can you please also attach it to http://bugs.openembedded.org ? May 30 13:41:52 XorA: Oh, so if I turn around the country on my left is now the country on my right? May 30 13:42:06 XorA: That'll get complicate soon! :-) May 30 13:42:09 Laibsch: now you get it, that way everyone gets an equal dose of hatage May 30 13:42:14 and lovage as well May 30 13:42:29 cool idea! May 30 13:42:32 Laibsch: about 1 years ago, May 30 13:42:35 spread the hate! May 30 13:42:40 Oh just everybody hate the US; it's easier that way. May 30 13:42:45 :p May 30 13:42:48 crink: Please attach it to the BTS. May 30 13:42:55 watch it! May 30 13:43:10 We can at least apply it to OE and ask mallum for inclusion. May 30 13:43:15 We'll send Bush to ive in your county May 30 13:43:16 i have to search the patch from *my* computer, it takes time. May 30 13:43:28 I#d appreciate it May 30 13:43:32 maybe next week i can upload it. May 30 13:43:40 no rush May 30 13:43:47 Crofton: he does already live in my country ;) (Chicago) May 30 13:44:01 heh May 30 13:44:20 I tried to get the Mexican here to take TX back :) May 30 13:47:02 Crofton: I suggested that we trade with Mexico: They get California if we get Cancun. May 30 13:47:51 yay I get to hate Wales May 30 13:48:29 * crink was banned from debian-offtopic when he free speech about racism.. May 30 13:48:37 have to go May 30 13:48:39 bye May 30 13:54:40 Well, sorry to say this now that you're gone, but I think I can understand the ban O:-) May 30 13:55:18 * Laibsch is glad this did not escalate May 30 13:56:23 03hrw 07org.oe.dev * r9f8cf61e... 10/ (5 files in 2 dirs): linux-mtx*: added COMPATIBLE_MACHINE May 30 13:56:43 I would then add him to ignore list and forgot about him May 30 13:58:14 03koen 07org.oe.dev * rc7115238... 10/ (1 packages/icu/icu_3.6.bb): icu: depend on icu-native, thanks XorA May 30 13:58:58 hrw: no, I'll just refrain from raising politics. As I said, we don't have to agree on politics. In this case it seems wiser to communicate about other things so as not to give an opportunity for frustration on either side. May 30 13:59:36 I have no issue with him personally. And if we can get some Korean input for OE that would be great. I have no clue about Korean. May 30 13:59:45 re May 30 14:00:01 s/raising politics/raising politics with him/ May 30 14:05:06 Laibsch: we are developers. this is developers channel May 30 14:07:10 hrw: yes, of course. May 30 14:07:19 cbrake: hey Cliff, can you /msg me? May 30 14:07:27 I don't intend to make this my political discussion forum May 30 14:07:37 If I do, just tell me to shut up ;-) May 30 14:08:09 ~boot Laibsch May 30 14:08:12 * ibot inconspicously shuffles over to get a straight line across Laibsch to the window above the dumpster, takes aim, and punts Laibsch out to where he belongs! May 30 14:08:26 :-D May 30 14:09:04 XorA : country left when looking north on south ? :) May 30 14:09:37 steliosk: it doesnt matter, always country to left which ever way you face May 30 14:09:39 ~zap XorA from the dumpster May 30 14:09:40 * ibot takes out a cattle prod and gives XorA from the dumpster a good jolt. May 30 14:10:04 jsheldon: just sent /msg -- let me if you did not get it May 30 14:11:14 morning May 30 14:20:58 cbrake: did you get my replies? May 30 14:21:31 jsheldon: no I did not, hmm May 30 14:22:15 cbrake: ok I'll be back in a minute May 30 14:30:39 hi all, which is the difference between depends and rdepends? May 30 14:30:49 * koen bangs head on desk May 30 14:30:54 cp is not install... May 30 14:30:57 hey vivijim May 30 14:31:08 vivijim: compile dependencies and runtime dependencies May 30 14:31:18 vivijim: DEPENDS = buildtime, RDEPENDS = runtime (Depends: in the .deb control file) May 30 14:31:18 hi koen! May 30 14:32:47 uhm but in a task my rdepends and my depends are not been built... It is very strange because it was used to work some days ago... May 30 14:33:04 bitbake 1.6.x or 1.8.x? May 30 14:33:30 in either case you need RP or zecke :) May 30 14:34:45 03hrw 07org.oe.dev * r577e91f0... 10/ (1 packages/linux/linux-tornado-omap2_2.6.16.16.bb): linux-tornado-omap2: add COMPATIBLE_MACHINE May 30 14:34:50 03hrw 07org.oe.dev * re15bb626... 10/ (4 files in 2 dirs): linux-h6300-omap1: add COMPATIBLE_MACHINE May 30 14:35:29 koen: bitbake 1.6 May 30 14:35:31 hrw: building for progear? ;) May 30 14:35:46 vivijim: update to the bitbake-1.8 branch from svn May 30 14:35:51 Italians know how to run a bike race! May 30 14:35:54 that solves a lot of other problems May 30 14:36:13 Crofton: sounds a bit backwards, running in a bike rice May 30 14:36:29 :) May 30 14:36:38 I have never seen anything this steep May 30 14:36:40 race* May 30 14:36:52 koen: that depends on the amount of steroids you have in your body ;-) May 30 14:37:01 koen: I was using the 1.6 one because I had another problem with 1.8 (I cannot remember what)... But I'll try it... May 30 14:37:25 vivijim: dependencies are known to break in 1.6 IIRC May 30 14:37:36 I need to bother Rishi about getting a DV EVM again ... May 30 14:37:55 koen: no - db1200 May 30 14:39:03 I can't remember if I sent an RFC for COMPATIBLE_MACHINE May 30 14:39:23 koen http://www.cbc.ca/canada/calgary/story/2007/05/29/museum-ab.html May 30 14:40:02 *twitch* May 30 14:40:08 ~seen mrevell May 30 14:40:38 Laibsch: i haven't seen 'mrevell' May 30 14:41:10 03koen 07org.oe.dev * r5d75706a... 10/ (1 packages/webkit/webkit_svn.bb): webkit: fix packaging May 30 14:42:22 I think I just made webkit 4MB smaller.. May 30 14:58:41 grr hippy debian 2.6.21 kernel May 30 15:03:09 nice koen May 30 15:04:02 2.2M armv4t/libwebkitgdk1_0.0+svn20070530-r1_armv4t.ipk May 30 15:04:07 instead of 6.4MB May 30 15:04:16 and it works as well :) May 30 15:04:43 is 2.6.21 from poky ready for OE? May 30 15:05:58 koen scary :) May 30 15:07:46 is there already some browser in OE using webkit? :) May 30 15:07:54 yes May 30 15:08:04 which one? May 30 15:08:07 the one included in webkit :) May 30 15:08:14 :-)) May 30 15:08:36 I guess it doesn't fit 240x320 screen very well yet, does it? May 30 15:08:43 'ipkg install libwebkitgdk-gdklauncher; DISPLAY=:0 GdkLauncher' May 30 15:09:03 mmp: it only has an url bar and a 'go' button May 30 15:09:13 sounds promising... May 30 15:09:51 * mmp will try in \varepsilon time :) May 30 15:10:38 it doesnt have scroll bars so its kinda useless May 30 15:10:54 zecke has a patch for that :) May 30 15:11:56 XorA: some browser allow "scrolling" when doing text selection, isn't it applicable in this case? :) May 30 15:12:10 or joypad might also help.. May 30 15:12:46 mmp: I think you find GdkLauncher is a very basic demo, it doesn nothing at all apart from render page May 30 15:13:05 neep it integrated into gpe-mini-browser May 30 15:13:20 koen: I am flashing it on my c7x0 May 30 15:13:25 XorA: well, now I'm forced to use dillo, as that's probably the only correctly working browser in repo... May 30 15:13:27 mickeyl: May I ask you a brief question about your alice experience in private chat? May 30 15:13:37 mmp: links-x11 is more correctly working May 30 15:13:37 mmp: and minimo May 30 15:13:45 koen: minimo crashes on ssl May 30 15:14:00 XorA: I did not manage to install it due lack of space... May 30 15:14:20 koen: err, I tried minimo few months ago, did not seem to work, but I have not tried very hard... . May 30 15:14:32 XorA: ssl or sites with broken ssl? May 30 15:14:39 koen: ssl at all May 30 15:14:56 koen: it either puts up a requesters that doesnt work, or just crashes May 30 15:15:04 I wonder why people keep claiming gmail works... May 30 15:15:14 no ssl :-) May 30 15:15:44 or maybe it sites with certain fields in certificates (note this is different from broken) May 30 15:16:24 gpe-minibrowser lacks a requester, so it only works with sites that work with curl-certs May 30 15:16:24 koen: why not using install in webkit? May 30 15:16:40 XorA: no reason May 30 15:18:58 * koen wonders how to write a do_stage for webkit May 30 15:19:12 koen: use $EDITOR May 30 15:19:36 do we have any funding for a new OEDEM yet? May 30 15:19:44 can we do one in Alborg? May 30 15:20:30 koen: gpe-mini-browser works? :) I've seen it only rendering certain pages; even google did not work. May 30 15:21:19 koen: you were asking yesterday why I'm trying to port Kde4 to qtopia-core. Really, it's just to make porting apps easier, not for running the full DE on a handheld. That would be kind of pointless :) May 30 15:21:26 Poky on c7x0 works with 2.6.21 May 30 15:21:42 koen: I'm just curious, because it's browser with probably the smallest footprint, but I was not able to use it effectively. May 30 15:23:41 clothing keyboard May 30 15:24:04 oh wrong window, sorry :) May 30 15:27:49 koen since rdepends is the runtime I believe that I should use the names of the dep packages right? but when I'm trying to use the libz1 I get this error message: "No providers of runtime build target libz1" What is wrong? BTW: the bitbake1.8 seens to be solved my rdepends problem... May 30 15:28:42 vivijim: OE takes care of the renaming, so you can do "RDEPENDS = zlib" and the .deb will have Depends: libz1 May 30 15:29:11 uhm ok thanks May 30 15:31:02 did someone had situation when rpcgen was unable to locale cpp which is in PATH? May 30 16:03:17 RP: ping May 30 16:08:47 koen|away, recent Angstrom still boots on the OSK May 30 16:08:56 Crofton: ;( May 30 16:09:14 Crofton: we were unable to break it.. sorry May 30 16:09:21 heh May 30 16:09:22 well try to do that later May 30 16:09:26 ;DD May 30 16:11:07 26254 hrw 15 0 244m 233m 3788 S 0 11.5 85:18.77 irssi May 30 16:11:27 fsck May 30 16:13:08 thesing: do you have a copy of the 5500 service manual? If not, I could send you one May 30 16:13:48 who still does not have copy of collie service manual? May 30 16:14:23 hvontres|poodle: No I don't have one. May 30 16:14:28 me either May 30 16:14:47 if its useful I want it ;) May 30 16:14:48 I hanvn't touched my 5500 in a while though May 30 16:14:57 thesing: I will put in online May 30 16:15:38 thanks May 30 16:15:39 ~lart syscallX May 30 16:15:40 * ibot throws a AN/M-8 smoke grenade at syscallX May 30 16:15:55 hrw: thanks May 30 16:18:33 http://blog.haerwu.biz/tmp/sl5500smpg.pdf May 30 16:22:47 wow its indeed useful. I was looking for something like this for a while. May 30 16:23:13 thesing: iirc I already asked you about that manual in pst May 30 16:23:15 thesing: pong May 30 16:23:17 s/pst/past May 30 16:23:32 * hvontres|poodle wishes he could find one for 5600.... May 30 16:23:51 RP: do you know details about kexec-arm? May 30 16:24:00 thesing: Some details May 30 16:24:18 hrw: don't happen to have that for slc-1000 do you :-) May 30 16:24:26 RP: do you know where it creates the ATAG list? May 30 16:24:40 thesing: It doesn't May 30 16:25:21 RP: so it uses the one from the old kernel? May 30 16:25:39 thesing: No, it doesn't use one May 30 16:26:00 Zero_Chaos: there was no English mannuals for cxx00 May 30 16:26:16 hrw: yeah, I kinda figured. bastards. May 30 16:26:20 hrw: thanks anyway May 30 16:27:07 RP: so how it gets information about the memory layout etc? May 30 16:27:23 thesing: it doesn't May 30 16:27:26 thesing: did you looked at kernel init asm code? May 30 16:27:39 thesing: There is no commandline handling May 30 16:27:53 hrw no sl-5000d manual? May 30 16:28:07 chouimat: its same machine just other ram chips May 30 16:28:15 k May 30 16:28:25 chouimat: sl5500 manual notify when 5500<>5000 May 30 16:28:47 RP: if I wanted to implement commandline handling I would create a ATAG list and pass it to the new kernel? May 30 16:28:56 hrw perfect then May 30 16:29:09 thesing: Then you'd need to implement it May 30 16:29:38 hrw: which one? arch/arm/kernel/head.S ? May 30 16:29:57 RP: Ok. Thanks. May 30 16:31:10 RP, hrw: do you know if the Cxx0 machines use a Locomo chip or just the sl-5xxx ones? May 30 16:31:38 koen|away, really away? May 30 16:31:38 just the sl5xxx May 30 16:31:59 thesing: boot/compressed/head-sharpsl.S May 30 16:33:23 mickeyl: thanks. May 30 16:33:40 hrw: this is for PXA machines so I didn't look at it. Should I? May 30 16:33:40 * hvontres|poodle will need to do some digging in locomo.c later... May 30 16:34:18 hvontres|poodle: locomo.c is not that bad ;) May 30 16:49:40 re May 30 16:49:48 anyone know where the arm gcc compiler flags are documented? May 30 16:50:19 thesing: actually, I need to take a look at what changed/ should have changed between 2.6.17 and 2.6.20. May 30 16:50:38 Crofton: http://gcc.gnu.org/onlinedocs/gcc/ARM-Options.html May 30 16:50:55 thesing: it looks like the lococmo is not coming back from suspend properly on poodle May 30 16:51:08 Crofton: or http://gcc.gnu.org/onlinedocs/gcc-4.1.2/gcc/ARM-Options.html#ARM-Options May 30 16:51:11 I think u-boot is building with old abi and trying to link against a new abi lib May 30 16:51:30 Crofton: -mapcs32 and stuff? May 30 16:51:38 yeah May 30 16:51:44 hvontres|poodle: the is a fixme for poodle in the file May 30 16:51:57 I have -mapscs-32,-mabi=apcs-gnu May 30 16:52:07 -PLATFORM_CPPFLAGS +=$(call cc-option,-mapcs-32,-mabi=apcs-gnu) May 30 16:52:08 +PLATFORM_CPPFLAGS +=$(call cc-option) May 30 16:52:20 thesing: I'll take a look when I get home tonight. I am at work and don't have the souce files ATM May 30 16:52:20 apcs == OABI May 30 16:52:39 I wonder if u-boot can build with EABI? May 30 16:52:58 Crofton: one way to find out.....:) May 30 16:53:20 right May 30 16:53:38 hi all May 30 16:53:43 I am also concerned it is pulling symbols from libgcc May 30 16:53:55 AAPCS or EABI => arm-linux-gnueabi May 30 16:53:56 APCS or OABI => arm-linux-gnu and arm-apple-darwin May 30 16:53:59 mrdata: hi May 30 16:54:09 thesing: hi ;-) May 30 16:54:10 Crofton: -lgcc_eh for static linking iirc May 30 16:55:23 Crofton: didn;t you build the osk5912 uboot with eabi? May 30 16:55:55 http://geektechnique.org/projectlab/781/slurpr-the-mother-of-all-wardrive-boxes May 30 16:56:15 I think the issue revolves around things getting linked from libgcc May 30 16:57:38 so OSK u-boot uses EABI tool with OABI flags, but never links against libgccc May 30 16:58:21 yeah OSK won't build with EABI May 30 17:01:07 confirmed I need to use old abi to build u-boot forOSK May 30 17:01:44 http://geektechnique.org/projectlab/781/slurpr-the-mother-of-all-wardrive-boxes May 30 17:01:58 you posted that url before :) May 30 17:02:08 sorry May 30 17:02:21 like 6 minutes ago :) May 30 17:02:22 I couldn't remember, I must be getting old May 30 17:02:24 koen any place where EABI (or AAPCS) and OABI (or APCS) are defined? May 30 17:02:43 chouimat: somewhere on arm.com I'd guess May 30 17:02:47 we need to wean all embedded guys away from debian :) May 30 17:03:16 and from timesys and from mv and from... May 30 17:05:34 Crofton: right on! May 30 17:05:40 chouimat|lunch: on http://www.heyrick.co.uk/assembler/ there is something about APCS May 30 17:06:05 I am working on some davinci guys May 30 17:06:08 cu May 30 17:06:15 cu May 30 17:06:15 I really need rishi to send me the EVM May 30 17:06:37 thesing: thanks May 30 17:07:00 ~arm-arm May 30 17:07:08 ~arm arm May 30 17:07:10 methinks arm arm is see armarm May 30 17:07:18 ~armarm May 30 17:07:19 i heard armarm is at http://www.altera.com/literature/third-party/ddi0100e_arm_arm.pdf May 30 17:07:24 ibot: arm-arm is see armarm May 30 17:07:26 hrw|gone: okay May 30 17:07:45 * hrw|gone has arm arm somewhere in drawer May 30 17:07:46 bye May 30 17:08:14 the last time I saw an arm-arm was in brazil :) May 30 17:08:36 next to a huuuuge omap board May 30 17:09:03 * chouimat waits for the next edition ... and since the cad is doing better it will be cheaper to get than 90CAD :) May 30 17:27:46 03koen 07org.oe.dev * r1d416eab... 10/ (3 files in 3 dirs): openmoko-session: sync a bit with openmoko svn May 30 17:48:11 zecke: hi, have you been on linuxtag? May 30 17:50:37 Crofton: did you see http://www.techworld.com/security/news/index.cfm?NewsID=8965 ? May 30 17:51:19 I heard about that May 30 17:52:01 mr_nice: yes May 30 17:53:27 zecke: was it well visited? May 30 17:54:27 * mr_nice hopes it May 30 17:54:32 it was the first day May 30 17:58:16 * mr_nice crosses fingers that linuxtag can be saved May 30 17:59:09 zecke: http://www.angstrom-distribution.org/repo/?action=details&pnm=libwebkitgdk-gdklauncher May 30 18:04:09 zecke: do you know which files need to get staged? May 30 18:06:35 zecke: http://www.youtube.com/view_play_list?p=AA11ACE45D6B0E4C May 30 18:08:08 Eh, pidgin has another "problem" May 30 18:09:05 Seems that libpurple plugins depend on pidgin-plugin-extplacement, which depends on gtk2 May 30 18:09:13 For no reason :) May 30 18:09:34 Um, libpurple *protocols* depend on pidgin-plugin-extplacement May 30 18:11:57 are you sure? May 30 18:12:22 does anyone knows which keyboard driver gpes x11 uses? May 30 18:12:30 angstrom@serenity:~/website/unstable/feed/armv5te$ grep pidgin-plugin-extplacement base/Packages May 30 18:12:31 Package: pidgin-plugin-extplacement May 30 18:12:32 Filename: pidgin-plugin-extplacement_2.0.0-r1_armv5te.ipk May 30 18:12:37 and that's it May 30 18:12:45 nothing in Depends: and friends May 30 18:12:55 Hmm May 30 18:12:57 Wait a sec May 30 18:13:03 I'm going to check the package May 30 18:14:30 Eh, for me it has Depends: libpurple, gtk+ (>= 2.8.9), atk (>= 1.10.3), glibc (>= 2.5), pango (>= 1.16.2), cairo (>= 1.4.4), glib-2.0 (>= 2.12.12), freetype (>= 2.1.10), fontconfig (>= 2.4.1), libpng12 (>= 1.2.12), libxrender (>= X11R7.1-0.9.1), libx11 (>= X11R7.1-1.0.1), libgcc (>= 4.1.1) May 30 18:14:43 In the control file May 30 18:15:21 I'm guessing it inherits it's dependancies from pidgin May 30 18:15:34 Because they're supposed to be part of the same package May 30 18:15:47 that's not what you said May 30 18:15:50 you said May 30 18:16:07 [20:09] B_Lizzard: Um, libpurple *protocols* depend on pidgin-plugin-extplacement May 30 18:16:12 which they don't May 30 18:16:20 Eh, they do for me May 30 18:16:35 pidgin-plugin-extplacement depends on all that stuff May 30 18:16:39 Lemme check the protocols May 30 18:16:56 Slight misunderstanding there... May 30 18:17:42 Ok, say libpurple-protocol-msn depends on... May 30 18:17:48 pidgin-plugin-extplacement, glib-2.0 (>= 2.12.12), glibc (>= 2.5), libgcc (>= 4.1.1) May 30 18:17:49 since extplacement is a *GUI* plugin, I'm not surprised it depends on gtk+ May 30 18:17:50 For me May 30 18:17:57 Yeah May 30 18:18:13 http://www.angstrom-distribution.org/repo/?action=details&pnm=libpurple-protocol-msn May 30 18:18:21 Yeah, I kinda mixed up things there, sorry May 30 18:18:25 The first time May 30 18:18:27 Depends: libglib-2.0-0 (>= 2.12.12), libc6 (>= 2.5) May 30 18:18:32 :/ May 30 18:18:37 Seems something went wrong May 30 18:18:50 I'll clean, mtn update and rebuild :/ May 30 18:22:15 Dunno how it could've mixed things up, though May 30 18:22:48 03koen 07org.oe.dev * r910bc48d... 10/ (1 packages/webkit/webkit_svn.bb): webkit: add more depends May 30 18:31:19 Eh, again May 30 18:31:21 WTF? May 30 18:32:13 I'll try and add pidgin-plugin-extplacement to BBMASK or something May 30 18:33:31 03koen 07org.oe.dev * r2ce404b9... 10/ (1 packages/webkit/webkit_svn.bb): webkit: fix typo May 30 18:53:20 Strange problem though May 30 18:53:58 Maybe if someone (re)builds pidgin, it'll happen to him too May 30 18:54:08 And I won't feel like a lone arsehole May 30 18:58:37 03ifaistos 07org.oe.dev * r5bbcb1c9... 10/ (3 files in 3 dirs): May 30 18:58:37 conf/machine/sequoia.conf: Sequoia board is using AMCC's ppc440EPx cpu with an FP unit and May 30 18:58:37 so it needs to build with -mcpu=440fp for hard fp support May 30 19:13:38 Again! May 30 19:15:06 * cbrake wonders about the state of Java support in OE ... May 30 19:16:45 wassup with pidgin? May 30 19:18:09 For some reason, libpurple-protocol-* depend on pidgin-plugin-extplacement May 30 19:18:20 Which depends on gtk2 and so on and so forth May 30 19:18:35 hmm... May 30 19:18:41 I've cleaned it, updated OE, BBMASKed it... May 30 19:19:13 koen, Crofton : here is a paper on the security stuff: http://www.arxiv.org/ftp/physics/papers/0612/0612153.pdf May 30 19:19:19 I was checking the bb file, nothing seems to match May 30 19:19:27 And koen doesn't have the same issues May 30 19:19:37 As it seems May 30 19:19:53 so, B_Lizzard do you suspect that gtk2 gets linked in for no reason or what? May 30 19:20:15 Well, it's logical that pidgin-plugin-extplacement depends on all that stuff May 30 19:20:19 It's a GUI plugin May 30 19:20:35 But the protocols getting linked with it is not May 30 19:20:36 Yes, so illogical part is dependence on extplacement? May 30 19:20:41 Yes May 30 19:21:29 all protocols depend on it? May 30 19:21:39 Yeah May 30 19:22:42 hmm... May 30 19:22:49 Depends: pidgin-plugin-relnot, libpurple-libjabber (>= 2.0.0), libc6 (>= 2.5) May 30 19:22:52 (xmpp) May 30 19:23:26 no extlacement... May 30 19:23:30 Say, libpurple-protocol-msn May 30 19:23:31 extplacement even. May 30 19:23:34 For me, it depends on May 30 19:23:41 Depends: pidgin-plugin-extplacement, glib-2.0 (>= 2.12.12), glibc (>= 2.5), libgcc (>= 4.1.1) May 30 19:23:41 let's see. I don't have this one installed. May 30 19:23:51 I have noooo idea what's going on May 30 19:24:57 And it having no version after the package name is also a bit fishy May 30 19:25:03 Depends: pidgin-plugin-relnot, libglib-2.0-0 (>= 2.12.12), libc6 (>= 2.5) May 30 19:25:07 The dependancies May 30 19:25:23 It depends on another plugin for you? :/ May 30 19:25:27 Yes, because plugins have no versions. May 30 19:25:33 they all depend on relnot May 30 19:25:58 Mine seems to have gone wayward May 30 19:26:20 I'll check my distro file and my local.conf May 30 19:26:35 Although I'm pretty sure that's not the reason. :| May 30 19:26:45 probably stray shlibs in staging May 30 19:27:08 I don't even understand how would stray shlibs make it depend on extplacement. May 30 19:27:24 But on the other hand I don't understand why does it depend on relnot. May 30 19:27:26 * mwester imagines a caretaker script, carefully sheparding stray shlibs out of staging... May 30 19:28:18 mwester: packaged-staging May 30 19:28:25 mwester, I think that's what these 'packaged staging' talks were about, but I have no idea what's that :) May 30 19:28:40 besides what the name implies, of course. May 30 19:28:52 check packaged-staging.bbclass :) May 30 19:29:13 Sure there's a number of ways to find this out ;-) Thanks for the pointer, though. May 30 19:29:19 Hmm, I found a couple of references to that plugin in shlibs May 30 19:29:28 I'll check those out, thanks May 30 19:30:41 I was under impression that these deps are generated automatically, roughly speaking, from ldd output... May 30 19:30:49 Yeah, me too May 30 19:30:56 And I do not see how this plugin depends on relnot. May 30 19:30:57 Or if they're forced by the bb file May 30 19:31:12 I know they aren't. May 30 19:31:13 normal reason for these sorts of problems is two library in two packages with the same name May 30 19:31:16 I mean, in this case. May 30 19:31:32 hmm.. May 30 19:31:43 check ${STAGING_DIR}/shlibs/*.list May 30 19:31:52 bitbake effectively picks a package at random to depend on May 30 19:31:55 minimo, thunderbird, firefox all do it to each other May 30 19:32:28 polyonymous: OE stores the shlib list in ${PN}.list May 30 19:32:42 koen, yup, I see. May 30 19:32:54 so if idd say "libfobar1" it looks in to *.list and adds the match to DEPENDS May 30 19:33:17 but there's only one relnot in staging... May 30 19:33:29 which is an incentive to packages all shlibs seperatly ;) May 30 19:33:48 and relnot is the only so in its package. May 30 19:34:11 Same with extplacement May 30 19:34:37 luckily relnot isn't causing any problems :) May 30 19:35:17 :) May 30 19:35:42 Still, this random behaviour isn't exactly what's intended to be May 30 19:36:02 right, someone should find a cause. May 30 19:36:18 And I'm afraid I won't be this someone atm ;-) May 30 19:36:56 Well, as I need finch without having to install gtk2, I'm looking into it May 30 19:37:09 I'll see if I can come up with something May 30 19:37:25 (I very much doubt it :/ ) May 30 19:37:26 Good :) May 30 19:37:30 Baad :) May 30 19:37:49 :) May 30 19:42:50 koen, can you point me to the code that handles these dependencies? May 30 19:43:08 polyonymous: I suspect package.bbclass May 30 19:43:13 ok. Thanks. May 30 19:43:20 grep for shlib :) May 30 19:43:29 also good pointer, thanks :) May 30 19:43:56 I think I'm going to do a rewrite of packaged-staging this weekend May 30 19:45:43 I have to admit, digging python code is a pain in certain part for me, since I don't know python. But I'm getting to know it, unfortunately :) May 30 19:46:06 So it basically thinks that it depends on extplacement.so, goes into shlibs, checks the package under which extplacement.so is in, and creates a dependancy for pidgin-plugin-extplacement May 30 19:46:45 or it depends on somethingelse.so and that's in extplacement.list May 30 19:47:08 only extplacement.so is in extplacement.list May 30 19:49:10 We could either see why it links to that (or why it thinks it needs to link to that) [Normal Version], or just make no package correspond to extplacement.so (or whatever it is) [Extremely Hackish and Localized Version] May 30 19:49:47 B_Lizzard, rm ${STAGING_DIR}/shlibs/pidgin-plugin-extplacement.*? :) May 30 19:50:06 Yeah, but those get remade if you rebuild pidgin May 30 19:50:20 ah, indeed. May 30 19:50:22 I'll suspend the proccess once they are made, delete them, and continue May 30 19:50:36 Quite obviously not what all users must be made to do May 30 19:50:56 pidgin doesnt have recursive configure does it, where it can enable extra features if it detects an extplacement inststalled? May 30 19:51:08 re May 30 19:51:21 hrw: wb May 30 19:51:23 XorA|gone, extplacement is a part of it. May 30 19:51:27 extplacement is part of pidgin May 30 19:52:04 And it doesn't seem to be a problem with extplacement only May 30 19:52:27 It could be another package May 30 19:52:58 yes, in my case it's relnot, which protocols depend on for no apparent reason... May 30 19:53:08 Yeah, exactly May 30 20:07:17 someone remember how to get as much informations about pccmia/cf card as possible even if there is no driver for it? May 30 20:07:55 I remember a tool to dump all the status regs May 30 20:08:17 but that will probably not work with a recent kernel (if I could locate it) May 30 20:08:49 hrw: does pcmcia have /sys entries nowadays? May 30 20:09:03 Yeah May 30 20:09:15 pccardctl May 30 20:09:18 koen: its collie user so 2.4 kernel May 30 20:09:24 yes, it does. May 30 20:09:30 dunno about 2.4 May 30 20:09:37 dump_cis probably will help May 30 20:12:10 did someone had situation when rpcgen was unable to locale cpp which is in PATH? May 30 20:15:12 locale? May 30 20:15:56 locate, I think May 30 20:16:13 * polyonymous pretends he's mod_speling.so May 30 20:16:24 you'll never know with glibc :) May 30 20:16:36 glibc == drepperware May 30 20:17:03 :))) May 30 20:17:08 locate cpp |head -n1 == /etc/alternatives/cpp == /usr/bin/cpp May 30 20:21:29 well, shlib code doesn't relnot to deps list... May 30 20:22:01 The shlib entries are added *after* pidgin has been installed and packaged May 30 20:22:13 and? May 30 20:22:17 So they would do nothing in this case May 30 20:22:25 hmm.. May 30 20:22:37 Well, they do. May 30 20:22:42 Really? May 30 20:22:43 I mean, not shlib files May 30 20:23:00 I don't have a faintest clue whether it uses files or not. May 30 20:23:15 if n in shlib_provider.keys May 30 20:23:27 Did you manage to take relnot out of the packages? May 30 20:23:31 anyway, this is not where relnot gets added May 30 20:23:38 Aha May 30 20:23:39 haven't even tried to May 30 20:24:35 03xora 07org.oe.dev * r5f43096a... 10/ (3 files in 3 dirs): May 30 20:24:35 evince_0.9.0.bb : on my kde system trying to run scrollkeeper just hangs May 30 20:24:35 forever so removing it the brutal way as the --disable-scrollkeeper May 30 20:24:35 seems to be a no-op May 30 20:25:01 * * OE Bug 2342 has been RESOLVED (FIXED) by dp@xora.org.uk May 30 20:25:03 * *  evince needs scrollkeeper-native [...] May 30 20:25:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2342 May 30 20:25:16 How does bitbake determine which packages depend upon which other than ldd and DEPENDS in the bb file? May 30 20:25:36 RDEPENDS + RRECOMMENDS May 30 20:25:46 DEPENDS plays no role May 30 20:25:49 It's not the shlib entries, because they simply do not exist during the packaging phase May 30 20:25:51 I think it has to do with some other automated machinery... May 30 20:25:52 Oh, yeah May 30 20:25:57 (and it's OE, not bitbake) May 30 20:26:11 OE then :) May 30 20:26:47 This is a bit convoluted... May 30 20:27:03 isn't do_package writing out the shlibs? May 30 20:27:15 and do_package_write will add them to the control file? May 30 20:27:18 Yeah, but only after the install and packagin phase May 30 20:28:08 Then probably it has lists generated somewhere in memory. May 30 20:34:02 * * OE Bug 2404 has been created by mail@jeroenhoek.nl May 30 20:34:04 * * Zaurus clamshells need a custom modmap May 30 20:34:06 * * http://bugs.openembedded.org/show_bug.cgi?id=2404 May 30 20:35:51 As expected, msn doesn't link to any of the plugins May 30 20:36:01 So it's not ldd May 30 20:37:13 I know it isn't. May 30 20:38:06 Since this didn't happen before, I'm guessing it's the pidgin.inc file May 30 20:38:31 could be an off by one error somewhere May 30 20:38:45 I keep havin libglib-2.0-dev being added to libmimedir Depends: May 30 20:38:55 I'll check a diff May 30 20:42:38 The diff seems to be nothing... May 30 20:42:43 http://www.openembedded.org/viewmtn/diff.psp?id1=077d4c675ec4c8e5510e74f9033fd6c1079b78c6&id2=4be62db5e8ff5d5cad0e43cd8487d753ab05b5db May 30 20:42:59 Nothing that would break things May 30 20:43:10 (Not that I'm an expert or anything) May 30 20:43:45 This my commit, it didn't introduce new deps, I believe. May 30 20:43:54 Yeah May 30 20:43:57 I mean it's been there already. May 30 20:44:09 It took the pidgin dep out of finch, but nothing else... May 30 20:44:41 no, it also packaged a few missed out libs ;-) May 30 20:44:53 But it has nothing to do with this issue. May 30 20:45:42 I checked in the install directory of pidgin May 30 20:45:53 libpurple-protocol-msn.shlibdeps says: May 30 20:45:59 hi May 30 20:46:00 glib-2.0 (>= 2.12.12) May 30 20:46:02 glibc (>= 2.5) May 30 20:46:02 libgcc (>= 4.1.1) May 30 20:46:02 :/ May 30 20:46:05 says 'hi' :) May 30 20:46:06 Hello May 30 20:46:20 I told you it's not shlib machinery. May 30 20:46:32 you're looking in the wrong place. May 30 20:47:39 Yeah, I see that now May 30 20:47:57 This is a definate bug May 30 20:48:10 With bitbake or OE May 30 20:49:02 Eh, JLime is in LinuxFormat May 30 20:49:04 :) May 30 20:50:13 feel free to file it then :) May 30 20:50:27 Yeah, but I still don't know where the bug is May 30 20:50:38 I can't file an obscure bug report May 30 20:50:52 I mean, everything seems right May 30 20:51:27 Well, neither I do. But it's a bug due to inconsistency of expected results and real ones. May 30 20:51:45 Yeah May 30 20:52:50 I'll try and pinpoint it further May 30 20:53:04 So will I. You got me started :) May 30 20:53:09 I'll revert to an older release of bitbake, maybe it's that May 30 20:53:22 Funny part is that now that I'm back from my trip I'm not going to need finch anytime soon :) May 30 20:53:27 :D May 30 20:53:40 The fun of hacking, then! May 30 20:53:41 :) May 30 20:55:22 the fun of python hacking? :) Not my cup of tea :) May 30 20:56:12 Try hacking SH3 ASM :) May 30 20:56:22 A hellish task May 30 20:58:04 So, let's see if it's bitbake 1.8 May 30 20:58:12 It's rebuilding the cache May 30 20:58:34 Otherwise it's probably the classes May 30 21:13:50 Eh, it's not bitbake May 30 21:14:18 It's the classes May 30 21:14:28 (probably) May 30 21:14:43 uf. May 30 21:14:51 few mails to -private sent May 30 21:16:57 About this, or unrelated stuff? May 30 21:16:59 :/ May 30 21:17:21 unrelated May 30 21:17:47 :) May 30 21:25:04 * Bernardo looks at the opie@handhelds.org mailing list and dispairs May 30 21:26:55 what caused the despair? May 30 21:44:43 cu May 30 21:51:18 It's in do_split_packages May 30 21:54:22 polyonymous: people fighting over trademarks instead of colaborating writing code May 30 21:54:47 GPE? May 30 21:55:01 Bernardo, ah.. well, yes, that happens all the way. Usually, due to inability to write code :) May 30 21:55:09 polyonymous: what's in do_split_packages? May 30 21:55:18 B_Lizzard, *IT* :) May 30 21:55:22 :D May 30 21:55:35 A revelation! May 30 21:55:46 heh. I think I'll track it down. May 30 21:56:02 B_Lizzard: GPE, OPIE, etc. Everything that was hosted in hh.org now has a trademark application in George France's name May 30 21:56:15 :/ May 30 21:56:29 Did George France write everything? May 30 21:56:46 * Bernardo goes back to finding out how to get more info out of enlightnment to find why it crashes May 30 21:56:52 B_Lizzard: no idea May 30 21:56:56 And I'd think that the GPL doesn't make it easy to trademark such things May 30 21:57:10 I do recall some oe vs familiar wars, though May 30 21:57:54 aha May 30 21:58:09 Well, it does have a comment "# This is *really* broken" in there May 30 21:58:52 lol May 30 22:00:14 polyonymous: can you check what happened to billiardz gcc4 patch? The bb file referes it, but it is nowhere to be found May 30 22:00:15 Can anyone remember who the Zaurus users with suspend/resume problems were/are? May 30 22:00:41 hi RP May 30 22:00:50 hi Bernardo May 30 22:01:03 what gcc4 patch? May 30 22:01:07 polyonymous: ahahaha May 30 22:01:10 my akita suspends/resumes fine from opie menu, only gives the error I showed you May 30 22:01:27 And this is RP's comment, btw :) May 30 22:01:32 Anyway, I think I got it. May 30 22:01:35 :) May 30 22:01:58 Bernardo: Yes, I haven't forgotten about that although its low priority :/ May 30 22:02:03 I know May 30 22:02:46 I'm trying to recall who was complaining of susp errors the other day May 30 22:03:03 Okay, I'm rebuilding pidgin now and if everything goes well commit it. May 30 22:03:41 Ok, waiting with anticipation :) May 30 22:03:42 welcome home, CoreDump|home ;-) May 30 22:04:00 polyonymous: hi May 30 22:04:29 * v8jlene is back May 30 22:04:35 morning May 30 22:04:47 evening :) May 30 22:05:37 re May 30 22:07:58 burn in hell my wicked ISP May 30 22:08:14 Bernardo, GCC4 patch for billiardz is somewhere in OE bugtracker May 30 22:08:24 I recreated it since it got somehow lost May 30 22:08:49 what's billiardz, anyway? May 30 22:08:54 CoreDump|home: Can you remember who the zaurus user with suspend issues was? May 30 22:09:04 If anyone does, let me know as I have some code for them to test ;-) May 30 22:09:12 hmmm no sorry May 30 22:09:21 RP: which suspend issue? May 30 22:09:37 hvontres|poodle: The one where cxx00 doesn't come back from a suspend May 30 22:09:42 RP: it should be in the BTS May 30 22:09:47 was it not xjqsomething? May 30 22:09:56 RP: ohh, ok...:) May 30 22:10:15 cxx00? Hmm... mine does. May 30 22:10:26 But, RP, after suspend/resume my g-ether doesn't work. May 30 22:10:35 polyonymous, billiardz is quite nice game for Opie using sdl iirc May 30 22:10:48 Well, it works but it never sees the link May 30 22:10:58 Marex, ah, should try it then :) May 30 22:11:08 after pidgin issue is fixed. May 30 22:11:25 polyonymous: Most devices work but some don't May 30 22:11:54 RP, was it comment regarding usbnet or just general suspend/resume trouble? May 30 22:12:11 polyonymous: No, Its an outright resume failure I'm interested in atm May 30 22:12:20 #2306 looks like it May 30 22:12:43 RP, ah ok. If you get time to do something about g-ether and if there's anything I can help you with, let me know. May 30 22:12:54 03lenehan 07org.oe.dev * r399705df... 10/ (1 classes/cpan-base.bbclass): (log message trimmed) May 30 22:12:54 cpan-base.bbclass: Update the perl depenedencies: May 30 22:12:54 * For native modules DEPEND on perl-native and for others DEPEND on perl May 30 22:12:54 (which in turn depends on perl-native) rather then always depending on May 30 22:12:54 both perl and perl-native. May 30 22:12:55 * Only RDEPEND for non-native modules rather then both native and May 30 22:12:57 non-native. May 30 22:12:59 It's not awfully bad, but having to reboot feels somewhat windowish :) May 30 22:13:06 polyonymous: Does removing and modprobing the module fix it? May 30 22:13:12 RP, no May 30 22:13:23 polyonymous: Hmm. Thats bad then May 30 22:13:37 but it LOOKS normal, as if there's no cable. May 30 22:14:38 fwiw, I'm seeing the same bug on spitz May 30 22:14:48 g-ether? May 30 22:15:34 yes May 30 22:16:04 good, I don't feel so lonely then. May 30 22:16:08 =) May 30 22:16:46 B_Lizzard, pull May 30 22:16:55 03polyonymous 07org.oe.dev * rb8b984ca... 10/ (1 packages/gaim/pidgin.inc packages/gaim/pidgin_2.0.0.bb): pidgin: added zero-length extra_depends to do_split_packages call to avoid adding semirandom dependency by the totally broken code. May 30 22:16:56 :D May 30 22:17:08 Thanks a lot! May 30 22:17:19 I'd write in caps, but that would be geeky. May 30 22:17:27 THANKS A LOT POLYONYMOUS! May 30 22:17:40 I was about to write "Thanks for not writing it in caps" :) May 30 22:17:57 I cannot express my gratitude over standart TCP/IP enough May 30 22:18:58 I'm not sure if tcp/ip was designed for gratitude transmission. But I'm sure you can come up with GTP RFC. May 30 22:19:20 :) May 30 22:19:21 polyonymous: have you been hanging around Laibsch again :) May 30 22:19:38 I'll ssh into your PC and fill it with free porn! May 30 22:19:42 hvontres|poodle, I'm trying to learn from him :) May 30 22:19:47 That's how happy I am! May 30 22:20:00 Or is that not a good way to go? :/ May 30 22:20:08 :) May 30 22:20:12 B_Lizzard, thanks God you don't seem to be able to do that :) May 30 22:20:16 :) May 30 22:20:32 polyonymous, I hate my isp :E btw I used to play billiardz on jornada ;) May 30 22:20:33 task fetch for billiardz failed May 30 22:20:49 polyonymous, it has just one thing that is a problem ... there is no way to exit the game :D May 30 22:21:13 Marex, not technically, I hope? :) May 30 22:22:16 okay, guys, now that billiardz isn't available for download anymore, what do we do? May 30 22:23:47 find a new game to play? May 30 22:23:55 abandonware... Too many opie/qpe packages seem to be disapearing May 30 22:24:12 I can find the tarball in http://www.openzaurus.org/download/3.5.4/sources/, but apart from that... May 30 22:24:59 Marex: thanks for finding the patch, I found it was missing trying to build meta-opie-all May 30 22:25:11 Try reminiscense May 30 22:25:39 Which is the engine re-implementation of Flashback: Quest for Identity May 30 22:25:41 :) May 30 22:26:43 Also, C-Dogs May 30 22:28:09 added OZ sources to my list of mirrors. May 30 22:28:27 you hate your isp May 30 22:28:35 :) May 30 22:28:45 (how did I guess?) May 30 22:28:53 polyonymous: I had success using g_ether after resume on 3100 by unloading and reloading all usb related modules. May 30 22:29:05 thesing, tried that, didn't help. May 30 22:29:06 B_Lizzard: is there a opie port? :) May 30 22:29:18 Well, they all use SDL May 30 22:29:26 Just plain SDL May 30 22:29:36 that should build... May 30 22:29:49 So I guess that, if you link them to libsdl-qte, they should work May 30 22:29:54 well, have to go now... May 30 22:29:58 bye May 30 22:30:11 * Bernardo wishes for Ron Gilbert's games on his akita... someday May 30 22:30:16 'night May 30 22:30:20 * Bernardo is away: Ausente por agora. May 30 22:30:20 night May 30 22:31:09 Is anyone with a c7x0 running angstrom online? May 30 22:33:21 Someone will pay me for this :E May 30 22:33:27 how much? May 30 22:34:23 how much does life of some techsupp monkey cost? May 30 22:34:28 :B May 30 22:35:06 ~seen psokolovsky May 30 22:35:14 Im going to scold them tomorow pretty thoroughly May 30 22:35:47 psokolovsky was last seen on IRC in channel #handhelds, 2d 21h 52m 11s ago, saying: 'dnix: see http://defconfigman.sf.net . Check out from SVN. Do like other machines do. Voila.'. May 30 22:36:06 I cant even work on that pcmcia driver...darn May 30 22:38:45 03polyonymous 07org.oe.dev * r26f36d57... 10/ (3 files in 3 dirs): billiardz: restored the missing gcc4 patch, thanks to Marek Vasut from bug #2051. May 30 22:39:02 * * OE Bug 2051 has been RESOLVED (FIXED) by polyonymous@klever.net May 30 22:39:03 * *  Billiardz wont build cuz of missing gcc4 [...] May 30 22:39:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2051 May 30 22:39:06 polyonymous, no need to comment ... May 30 22:40:15 Marex_Yggdrasil, if it offended you, I apologize, if you're just being modest, then I think giving credit where credit is due is the right thing to do. May 30 22:40:27 polyonymous, it was about my isp again ;) May 30 22:40:34 Ah :) May 30 22:40:47 Why would I repeat? :) May 30 22:41:03 ;) May 30 22:41:14 good bye May 30 22:45:37 it seems that one could write a driver for the sharp remote controls for collie May 30 22:45:57 koen, do you have do_split_packages wherever you have this spurious dep problem? May 30 22:46:11 I just tested the remocon entry in the diag menu and the adc gets different values. May 30 22:48:20 thesing: hehe..a hacker with a manual...an dangerours combo :) May 30 22:51:19 Hi, does somebody has a good .config for h5000 kernel? (good = most devices working) May 30 22:51:56 polyonymous: not that I know off May 30 22:52:08 libmimedir doesn't use do_split_packages May 30 22:52:22 koen, ah ok. It's just where it was being added to pidgin May 30 22:56:27 polyonymous: I ehm... *cough* had your change already locally May 30 22:56:57 koen, you could've committed it then :) May 30 22:57:40 I sure didn't enjoy digging this code ;-) May 30 23:02:13 Is QT4 finally out? I don't understand this. The commercial Qt is making progress while FOSS QT is stuck at version 2 or what? May 30 23:02:28 qt 4 has been out for a while May 30 23:02:39 qtopia 4 also has been out for a while May 30 23:02:54 qtopia4 _gpl_ is only a few months old May 30 23:02:59 so everything is peachy? May 30 23:03:02 yes May 30 23:03:15 So are we using QT4? May 30 23:03:24 What is that discussion with hh.org about? May 30 23:03:36 iirc there's a qt4-x11 recipe and a qtopia4 recipe May 30 23:03:37 Why so little activity at opieII? May 30 23:04:01 because of trademarks? May 30 23:04:05 * koen hides May 30 23:04:07 ~corgie May 30 23:04:07 So, the release of QT4 did not bring the devs back in droves? May 30 23:04:16 ~c7x0 May 30 23:04:21 no, trademarks are just the means to the battle May 30 23:04:47 ~corgi May 30 23:04:49 [corgi] sharp sl-c700, or a dog May 30 23:04:54 Why is psokolovsky apparently still "working" at hh.org? Why on old Qt2 base? May 30 23:05:20 ~husky May 30 23:05:22 [husky] Sharp Zaurus SL-C760 or a dog May 30 23:14:02 * * OE Bug 2110 has been RESOLVED (FIXED) by oe@hentges.net May 30 23:14:04 * *  Angstrom 20070402-spitz gpe [...] May 30 23:14:06 * * http://bugs.openembedded.org/show_bug.cgi?id=2110 May 30 23:14:11 http://www.linuxdevices.com/news/NS2494029776.html <- I want one May 30 23:14:21 actually, I want two :) May 30 23:14:37 really? I wasn't impressed. May 30 23:15:11 just had an interview at montavista.. support engineer type position though, bugfixing, but itd get me in the door. *crosses fingers* May 30 23:15:18 I still 'need' a small laptop like thingy May 30 23:15:28 kergoth: heretic May 30 23:15:50 I would want one if it had swivel display like Z May 30 23:16:08 and a touchscreen May 30 23:16:30 * CoreDump|home wants an affordable _hackable_ tablet May 30 23:16:42 * polyonymous joins CoreDump|home May 30 23:17:02 maybe if we will want it together it will actually happen? May 30 23:17:08 CoreDump|home: like http://ewayco.com/62-LCD-Panel-PC-Webpad-Tablet-PC-system-low-cost/01-lcd-panel-pc-webpad-tablet-pc-system-low-cost.html ? May 30 23:18:04 koen: close, but $800 is a little bit steep :( May 30 23:18:41 $800 for Q100 May 30 23:18:58 the site sucks. May 30 23:19:02 TX Tablet PC $800 May 30 23:19:04 Guess so does the tablet :) May 30 23:19:06 polyonymous: it sure does May 30 23:19:17 it's asian May 30 23:19:27 I'm not racist :) May 30 23:19:28 so of course the look of the site sucks May 30 23:19:37 ever seen adds from japan? May 30 23:19:42 they make my head explode May 30 23:19:44 it feels like mid-90s :) May 30 23:20:20 Guess they're using some mid-90 html wysiwig editor with templates :) May 30 23:20:38 anyway, they say " Tablet PC, May 30 23:20:39 Touch, $680" May 30 23:20:52 they are all using a zaurus and only links-x11 works ;) May 30 23:21:30 talking about mid-90 html, ever seen myspace? May 30 23:21:52 believe it or not - no :) May 30 23:22:15 keep it that way :) May 30 23:22:24 as long as I can :) May 30 23:23:07 03coredump2 07org.oe.dev * r01b8b653... 10/ (3 files in 3 dirs): keylaunch: Swap Calendar and Address buttons to launch the correct application whe pressed. May 30 23:23:13 03coredump2 07org.oe.dev * r038120fa... 10/ (8 files in 3 dirs): xserver-common: Import a set of patches from the .oz354x branch, fixing keymapping on akita, spitz and c7x0. May 30 23:23:19 03coredump2 07org.oe.dev * r5406c265... 10/ (1 packages/keylaunch/keylaunch_2.0.10.bb): keylaunch: Fix RDEPENDS to display-brightness for selected machines, closing bug #2110 May 30 23:25:49 time to `mount /proc/self /dev/bed -t sleepfs -o opiereader` ;-) May 30 23:28:14 polyonymous: I wish I could compiles opiereader for x =) May 30 23:28:56 http://www.rpsys.net/openzaurus/temp/power_test-r0.patch is my dose of insanity for today... May 30 23:28:58 CoreDump|home, that may be possible, I've seen the port somewhere, but haven't got time to play with it. May 30 23:28:58 there simply is no decent X11 ebook reader that comes eve close to opie-reader =( May 30 23:29:11 RP: that's a lot of asm May 30 23:29:40 koen: Its a hack together of some nasty Sharp code. If it solves problems, I will implement it properly May 30 23:30:03 03coredump2 07org.oe.dev * r143d8aa2... 10/ (4 files in 4 dirs): keylaunch: the h/w hotkeys and LCD softkeys are now in the F20+ range (there was a damn good reason for changing them back in the days, but I forgot why heh ) May 30 23:30:17 In theory it might a) reduce suspend current and increase battery life and b) solve the mysterious cxx00 doesn't resume problem May 30 23:30:30 but both theories need testing May 30 23:30:41 that's what theories are for. May 30 23:31:05 RP: I suspect zecke's rpath-hack will make packaged-staging easier May 30 23:31:20 koen: Possibly, yes May 30 23:31:32 koen: I still don't know if we shouldn't just use separate packages May 30 23:31:42 koen: That way we'd handle the pkgdata too May 30 23:32:04 I'm still thinking about a chroot May 30 23:32:30 * polyonymous is gone to opieread himself to sleep. May 30 23:32:41 RP, hey, hows it going? May 30 23:32:59 kergoth: hi, ok thanks! you? May 30 23:33:15 * thesing -> sleep May 30 23:33:17 cu May 30 23:33:26 not bad. at a horrid position working on a greenhills embedded os instead of linux, but just had an interview at mvista, so hopefully... May 30 23:33:29 how are things at ohand? May 30 23:36:16 kergoth: Much the same really. Poky is ever improving and getting nicely stress tested with autobuilds now :) May 30 23:36:23 nice May 30 23:37:01 * * OE Bug 2290 has been RESOLVED (WORKSFORME) by oe@hentges.net May 30 23:37:04 * *  angstrom-x11-image custom build breaks NAND [...] May 30 23:37:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2290 May 30 23:37:19 kergoth, working for mv would be an improvement? May 30 23:37:20 :) May 30 23:37:46 anything linux and not GD is an improvement at this point :P May 30 23:38:43 kergoth: must be bad :/ May 30 23:44:22 RP and kergoth talking at the same time? May 30 23:44:25 * CosmicPenguin falls over May 30 23:45:56 CosmicPenguin: I should ideally be sleeping now (and will be shortly) :) May 30 23:46:46 heh May 30 23:47:39 Before I go home, I want to figure out how to package the cairo testsuite May 30 23:48:04 * CoreDump|home reads the flamefest at opie-devel May 30 23:48:54 I wonder why france didn't trademark matchbox or OE May 30 23:52:28 what a load of crap May 30 23:53:49 "community" my ass May 30 23:54:30 your ass would be a better community than france May 31 00:08:33 kergoth: mv, hey. good luck. May 31 00:08:54 Just an FYI about the flamefest. The Opie trademark thing it going to be worked out. I can't discuss details right now though... next week. May 31 00:09:18 kergoth: Long time no chat! I see your name in WoW addons now! :D May 31 00:09:51 people still play wow? May 31 00:10:00 * CoreDump|home beat that addiction ages ago May 31 00:10:01 bipolar: "worked out" as in you get a few $$$$ and hh.org the trademark? May 31 00:10:01 haha May 31 00:10:01 kergoth: if you need another reference, I will give one. mv is a partner of TT May 31 00:10:15 koen: no. thats not it at all. May 31 00:10:27 koen: next week :) May 31 00:10:39 opieII owes bipolar $$$$ for useage :) May 31 00:10:46 hmmm... maybe..... May 31 00:10:47 =D May 31 00:10:49 I hope it ends with george getting sued into bankrupcy May 31 00:10:55 kergoth, good luck with MV May 31 00:11:02 koen: one can only hope... May 31 00:11:34 the GD guys I knonw are real nice, but large corporations are no fun May 31 00:12:04 * bipolar must keep his mouth zipped though the weekend May 31 00:13:21 bipolar: better do so, this channels is logged for everyone to see May 31 00:13:44 even the folks on the ban list ;) May 31 00:14:29 haha May 31 00:15:59 CoreDump|home: Us old folks waited for all the punks to get bored with it so we could enjoy playing with being invited to a duel or to join a stupid guild every 15sec... :D May 31 00:16:29 =) May 31 00:16:59 Has anyone tested angstrom-2007.1-oabi distro lately? I can't seem to build an image. May 31 00:22:36 NB, what version of bitbake should I be running? I seem to have problems with the latest one from 1.8.x branch, 1.8.3? May 31 00:22:45 03lenehan 07org.oe.dev * r4c8cdb95... 10/ (17 files in 5 dirs): May 31 00:22:45 clamav: Add 0.90.3, remove 0.88.7. May 31 00:22:45 From the release announcement: "This release fixes some security bugs May 31 00:22:45 in libclamav and improves stability under Solaris. Please see May 31 00:22:45 ChangeLog for complete list of changes." May 31 01:02:02 * * OE Bug 1155 has been RESOLVED (FIXED) by lenehan@twibble.org May 31 01:02:04 * *  dosfstools-native fails to build on 64bit [...] May 31 01:02:06 * * http://bugs.openembedded.org/show_bug.cgi?id=1155 May 31 01:04:01 * * OE Bug 1484 has been RESOLVED (FIXED) by lenehan@twibble.org May 31 01:04:03 * *  busybox "sh" is inadequate to run "ldd" [...] May 31 01:04:06 * * http://bugs.openembedded.org/show_bug.cgi?id=1484 May 31 01:06:21 CoreDump|afk, i find coding wow ui modifications more enjoyable than playing, but still play occasionally May 31 01:08:17 kergoth, what would you be doing for MV? May 31 01:08:42 fixing their davinci code to compile with gcc-4.1 ? May 31 01:08:44 :) May 31 02:51:56 hrw: are you there? **** ENDING LOGGING AT Thu May 31 02:59:56 2007