**** BEGIN LOGGING AT Thu Apr 16 02:59:57 2009 Apr 16 04:34:31 03Mike Westerhof  07org.openembedded.dev * re9a0747532 10openembedded.git/ (5 files in 3 dirs): Apr 16 04:34:31 Revert "Openmoko: linux-openmoko-2.6.29 - commit new recipe tied to upstream kernel" Apr 16 04:34:31 This reverts commit fb7eba513b2d59c33163bd380019a4cabc8b6782. Apr 16 04:34:31 Change rejected. Apr 16 05:39:00 03Angus Ainslie  07fso/milestone5.5 * r9e4ede86c5 10openembedded.git/recipes/openmoko-projects/paroli_git.bb: paroli: better separation of shipped files Apr 16 05:39:11 03Mike Westerhof  07fso/milestone5.5 * r01742d3aa2 10openembedded.git/ (2 files in 2 dirs): Apr 16 05:39:11 Openmoko: linux-openmoko-2.6.29 - commit new recipe tied to upstream kernel Apr 16 05:39:11 git repo, which is currently at 2.6.29-rc3 + Openmoko patches. Note that Apr 16 05:39:11 the preferred kernel remains unchanged; select this kernel via local.conf. Apr 16 06:00:53 morning guys Apr 16 06:01:03 i have a problem with the latest gstreamer prerelease tarballs Apr 16 06:01:15 they now include an own shave-libtool -_- Apr 16 06:12:12 okay looks like i got it to compile by removing the shave m4 and the respective configure.ac line Apr 16 06:41:10 * MostAwesomeDude is back with more stupid questions Apr 16 06:41:48 task-x11 pulls in xserver-nodm, which prevents gpe-dm from auto-loading. Is there a simple alternative or do I need to roll my own task-x11? Apr 16 06:46:55 good morning Apr 16 06:47:20 oi Apr 16 06:50:39 MostAwesomeDude: are you bitbaking gpe-image and seeing this Apr 16 06:51:27 khem: oswald-image, but same thing. I'm currently depending on task-x11 and angstrom-task-gpe-base. Apr 16 06:55:58 MostAwesomeDude: task-x11 could be conflicting with task-gpe-base Apr 16 06:56:32 Hi Apr 16 06:56:55 Hello Apr 16 06:57:00 khem: Well, what happens is that xserver-nodm starts, then gpe-dm starts. Apr 16 06:57:10 And gpe-dm fails due to xserver already running. Apr 16 06:57:28 So the best course of action, we thought, was just to keep xserver-nodm-init from being installed. Apr 16 06:57:42 good morning Apr 16 06:57:45 why do you need task-x11 Apr 16 06:58:16 khem: Because we need an xserver and task-gpe-base doesn't depend on it. Apr 16 06:59:03 ? Apr 16 07:00:30 My name has been changed as Guest54006, anyone can tell me how to change it? Apr 16 07:00:49 Guest54006: /nick Apr 16 07:01:20 Hi Apr 16 07:01:45 No use Apr 16 07:02:20 Hi Apr 16 07:02:41 MostAwesomeDude: you could just DEPEND on virtual/xserver Apr 16 07:02:58 I inputted /nick Jack8312, but it looks no use. Apr 16 07:03:07 khem: Hm. I suppose that would work. Apr 16 07:03:11 Lemme see if that works. Apr 16 07:08:07 03Carsten Haitzler (Rasterman  07org.openembedded.dev * r48488aba06 10openembedded.git/recipes/pointercal/ (files/palmt650/pointercal pointercal_0.0.bb): pointercal: update palmt650 pre-seeded pointercal data (i just re-ran it) Apr 16 07:22:09 at91bootstrap_2.10.bb has as SRC_URI of file://defconfig. But this defconfig file is not on FILESPATH to be found, can someone shed some light on how it is supposed to be found? Apr 16 07:30:45 i try to compile libomxil-bellagio-0.9.0-r6 and i get a make[2]: *** [doxygen-build.stamp] Segmentation fault when it goes into the doc folder of the said source Apr 16 07:31:00 any clue why this occurs. am i using a old doxygen? Apr 16 07:48:35 Does Angstrom use a different FPE (floating point emulater) as the NWFPE in kernel normally? Apr 16 07:53:51 tsjsieb: that depends on the processor Apr 16 07:54:09 and toolchain Apr 16 07:56:02 o.k. I'm currently testing to create a custom kernel with OE and I have it enabled in the defconfig I give, but it seems that something overrides this option, so it is disabled in the .config file Apr 16 07:57:51 * mpoullet|work is back from holidays Apr 16 07:57:55 good morning Apr 16 08:03:42 tsjsieb: I think angstrom normally uses soft-float, i.e. no FPE at all, but you should probably check with #angstrom. Apr 16 08:04:41 o.k. thnx I will search for that :) Apr 16 08:25:19 * khem sleep Apr 16 08:28:10 on a zaurus sl-5500 how do i issue a ctrl-c Apr 16 08:30:17 good morning Apr 16 08:30:34 BusError: The invitation isn't sticky Apr 16 08:38:40 up already florian Apr 16 08:38:52 what was that like 4hrs? Apr 16 08:39:48 ntfn: oops no, somethign like 10 Apr 16 08:40:45 damn i need to sleep Apr 16 08:41:01 florian: good morning Apr 16 08:41:09 i just cant stop watching my build Apr 16 08:57:53 morning Apr 16 08:59:35 morning hrw! Apr 16 09:03:33 hrw: morning Apr 16 09:08:58 03Carsten Haitzler (Rasterman  07stable/2009 * rb9e483eac7 10openembedded.git/recipes/sysstat/sysstat_8.1.6.bb: Apr 16 09:08:58 sysstat: fix build on omap3 to actually work (gnu_hash problem like ltrace) Apr 16 09:08:58 Signed-off-by: Koen Kooi Apr 16 09:08:58 Acked-by: Denys Dmytriyenko Apr 16 09:10:45 hi thebohemian Apr 16 09:10:54 hi marcin Apr 16 09:11:17 i use totem on my board, when i play it allows me to select a movie file, but it does not play it. Apr 16 09:11:28 has anyone used totem here? Apr 16 09:11:59 thebohemian: java.bbclass change: + ln -sf $destname $dir/$1 Apr 16 09:12:13 thebohemian: instead of "ln -s $destname $dir/$1" Apr 16 09:13:26 thebohemian: thebohemian but why I had to do that I do not remember ;( Apr 16 09:35:06 hrw, no way to build rpm-native in stable :( ... Apr 16 10:02:43 anybody try buld fennec? i build failed. log is configure: error: Compiler cannot compile macros used in autoconf tests Apr 16 10:03:34 yan: thats new Apr 16 10:05:47 yan: discuss on channel, its a new bug, Ive not seen before Apr 16 10:05:57 yan: but Im at work no build machine handy Apr 16 10:06:57 XorA: i check configure.in ,error is about assertion_macros,i dont know what it mean Apr 16 10:07:16 mozilla build system is complicated Apr 16 10:08:13 it's not easy to understand configure.in and mozconfig file Apr 16 10:08:38 yan: I know, Im normally responsible for new versions of firefox :-D Apr 16 10:08:48 still search for answer,thanks for attention Apr 16 10:09:04 firefox is too big to run on arm platform Apr 16 10:09:30 fennec fancy ui is impressive Apr 16 10:13:35 fscking git Apr 16 10:13:51 yan: erm, depends on the arm Apr 16 10:15:13 it seems firefox build is similar to fennec build.just mozconfig file difference Apr 16 10:27:28 pretty much Apr 16 10:30:01 -> lunch Apr 16 10:34:48 03Jeremy Lainé  07org.openembedded.dev * r42e6449391 10openembedded.git/classes/siteinfo.bbclass: siteinfo.bbclass: add entry for arm-darwin9 Apr 16 10:37:47 * hrw works on long list of 'want to have in stable' things Apr 16 10:49:06 Quick question. My overlay needs a checksum.ini for a package in it, but that breaks all the other packages because apparently bitbake only recognizes one conf/checksums.ini. Is there some workaround? Apr 16 10:49:26 there was a patch on OE ML Apr 16 10:49:28 I sent a patch yesterday Apr 16 10:49:33 test it Apr 16 10:50:25 http://article.gmane.org/gmane.comp.handhelds.openembedded/22714 Apr 16 10:51:10 morning all Apr 16 10:52:43 hi Richard Apr 16 10:54:29 03Jeremy Lainé  07org.openembedded.dev * re0d9f95533 10openembedded.git/recipes/odcctools/ (odcctools-cross-sdk_svn.bb odcctools-cross_svn.bb): odcctools: add cctools for Darwin (from Poky) Apr 16 10:55:06 heh, someone is playing with darwin9? :) Apr 16 10:56:00 RP: I have to tell one thing. Apr 16 10:56:25 RP: switching from Poky'pinky' to OE stable/2009 is not as painful as I thought Apr 16 10:57:01 hrw: cool :) Apr 16 10:57:25 hrw: We did push a lot of Poky's bleeding edge stuff into OE ;-) Apr 16 10:57:33 yep Apr 16 10:57:51 * hrw hails git for making poky<>oe easier Apr 16 10:59:00 hrw: I'm still very sad to see the switch :( Apr 16 10:59:24 RP: so do I but how many projects uses Poky now? Apr 16 11:00:48 hrw: There is actually some growth atm ;-) Apr 16 11:01:26 Things will become public when they become public :/ Apr 16 11:01:34 lD Apr 16 11:03:32 * XorA sees RP is taking politician lessons Apr 16 11:05:05 03Robert Schuster  07org.openembedded.dev * r8fa95dc3f6 10openembedded.git/classes/java.bbclass: java.bbclass: Use force (-f) for ln (suggested by Marcin). Apr 16 11:05:12 hrw: pushed ... :) Apr 16 11:21:13 XorA: Gah ;-) Apr 16 11:22:01 * RP dislikes politics Apr 16 11:28:37 RP: when you next biking around the edin area? Apr 16 11:32:16 XorA: I rarely make it up to edin :( Apr 16 11:32:55 That's no good! Apr 16 11:32:56 XorA: I've taken up offroad trail riding more recently which is a lot more local too. You've probably seen the photos on te blog... Apr 16 11:33:13 yeah looks fun Apr 16 11:33:35 * RP is currently healing a fractured rib :/ Apr 16 11:33:42 eeeks Apr 16 11:35:00 Apparently it was an impressive crash to watch Apr 16 11:36:33 Hey guys. Apr 16 11:37:07 Quick question about policies: I have a bunch of package I need to commit on a specific branch, all related to the same distro, including some specific config/PROVIDE-ing packages related to e-wm or framework Apr 16 11:37:15 Should I put that in the same subdirectory of recipes/? Apr 16 11:37:36 Or should I put that in the subdir related to the actual package? (e.g. my e-wm config goes in e17/) Apr 16 11:49:15 huhu Apr 16 11:50:26 RP: hi, while you're here, question for you: can/should we remove the o-hand/poky bits from linux-rp.* files? Apr 16 11:58:56 ant_work: Why? Apr 16 11:59:13 03Koen Kooi  07stable/2009 * rb816a3b15b 10openembedded.git/recipes/python/ (2 files in 2 dirs): Apr 16 11:59:13 python 2.6.1: fix build for architectures lacking libffi support Apr 16 11:59:13 Signed-off-by: Koen Kooi Apr 16 11:59:13 Acked-by: Marcin Juszkiewicz Apr 16 11:59:15 03Koen Kooi  07stable/2009 * ra64c682acf 10openembedded.git/conf/checksums.ini: Apr 16 11:59:17 checksums: sync with .dev Apr 16 11:59:19 Acked-by: Marcin Juszkiewicz Apr 16 12:00:53 RP: well, there is no distro 'poky' io OE and the kernel bootlogo was defaulting to o-hand Apr 16 12:01:03 so I started commenting out... Apr 16 12:01:33 ant_work: That code was designed to work in both Poky and OE so no, it should not be removed or commented out Apr 16 12:01:48 The OH logo was only enabled for the Poky distro iirc Apr 16 12:02:27 * RP needs to do something about the OH logo, probably replacement with a beaver Apr 16 12:02:38 penguin is now free Apr 16 12:02:54 ant_work: free? Apr 16 12:04:03 now yes Apr 16 12:04:22 linux mascotte changed with 2.6.29 Apr 16 12:04:22 ant_work: What do you mean by free? Apr 16 12:04:29 joking Apr 16 12:05:03 ant_work: ah, right :) Apr 16 12:06:27 RP: any hope you'll make your hands dirty against power management forlinux-rp- 2.6.29 ? Apr 16 12:06:57 ant_work: There is always hope but I have my hands pretty full atm Apr 16 12:07:09 ant_work: Whats up exactly? Apr 16 12:07:23 LKML thread about spitz vs. 2.6.29 Apr 16 12:07:49 I think I did read that Apr 16 12:08:06 btw: I tried to catch Pavel too :-) Apr 16 12:08:15 If I look at it, it won't be a for a couple of weeks Apr 16 12:09:05 Basically whenever I next try and write a linux-rp_2.6.29 Apr 16 12:09:24 RP: 2.6.26 needs some polishing wrt recharge/suspend Apr 16 12:09:46 ant_work: I'll probably skip to 2.6.29 now Apr 16 12:10:01 I'm crossing fingers Apr 16 12:10:21 for c7x0 too Apr 16 12:11:17 btw during your absence we almost won the battle against Zaurus partition-size (kexecboot) Apr 16 12:12:05 next step being to be able to read mtdparts from Shapr's bootloader (old-params) Apr 16 12:12:12 or straight from nand dump Apr 16 12:12:39 (with 2.6.2x kernel ) Apr 16 12:14:24 RP: http://lists.linuxtogo.org/pipermail/openembedded-devel/2009-January/007617.html Apr 16 12:15:09 ant_work: cool Apr 16 12:15:46 btw CONFIG_SHARPSL_BOOTLDR_PARAMS remains a mystery..sources? Apr 16 12:16:31 ant_work: Don't go there ;-) Apr 16 12:16:39 ant_work: It was part of the sharp 2.4 junk Apr 16 12:16:46 ok, nand read then Apr 16 12:17:06 we almost know the addresses Apr 16 12:17:41 ant_work: what if u-boot + kexecboot? Apr 16 12:17:59 works but sounds insane... Apr 16 12:21:44 hrw: however the extra space could allow us to enable networking in kernel again (boot from USB / net) Apr 16 12:25:46 nfsroot over wifi/wpa... Apr 16 12:25:53 * hrw -> food Apr 16 12:34:43 ping hrw Apr 16 12:36:53 i got a problem quilding rpm packages, on rpm-native i got a lot off error such this on Q&A phase "non dbg contains .debug, rpm-native, /work/i686-linux/rpm-native-4.4.2.3-r15/install/rpm-native/staging/i686-linux/usr/lib/rpm/.debug/rpmq" Apr 16 12:43:26 03Koen Kooi  07org.openembedded.dev * ra2680a0a36 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Apr 16 12:43:36 03Koen Kooi  07org.openembedded.dev * r77623602a2 10openembedded.git/recipes/angstrom/ (3 files in 2 dirs): exquisite-theme-angstrom: add angstrom theme as a quick hack Apr 16 12:43:37 03Koen Kooi  07org.openembedded.dev * r7eec526f3c 10openembedded.git/recipes/angstrom/ (2 files in 2 dirs): exquisite-theme-angstrom: make it glow harder Apr 16 12:58:33 03Koen Kooi  07stable/2009 * r37baf47b4e 10openembedded.git/recipes/ (5 files in 3 dirs): Apr 16 12:58:33 bluez + obex: sync with .dev Apr 16 12:58:33 Acked-by: Philip Balister Apr 16 12:58:38 03Koen Kooi  07stable/2009 * rec1f47c78f 10openembedded.git/ (21 files in 6 dirs): Apr 16 12:58:38 gnome: update various bits to 2.26, sync with .dev Apr 16 12:58:38 Acked-by: Philip Balister Apr 16 12:58:40 03Koen Kooi  07stable/2009 * r27c14c917e 10openembedded.git/conf/distro/angstrom-2008.1.conf: Apr 16 12:58:43 angstrom 2009.X config file: sync with .dev Apr 16 12:58:45 Acked-by: Marcin Juszkiewicz Apr 16 12:58:47 Acked-by: Philip Balister Apr 16 12:58:51 03Koen Kooi  07stable/2009 * rfbea2ea1cc 10openembedded.git/recipes/obexftp/obexftp_0.23.bb: Apr 16 12:58:54 obexftp: update to 0.23 Apr 16 12:58:56 Signed-off-by: Koen Kooi Apr 16 12:58:58 Acked-by: Philip Balister Apr 16 12:59:00 Acked-by: Marcin Juszkiewicz Apr 16 12:59:04 03Fraxinas  07org.openembedded.dreambox * r3644614ee9 10openembedded.git/packages/gstreamer/ (files/gst-configure_skip_shave.patch gstreamer_0.10.22.2.bb): gstreamer: add recipe for gstreamer 0.10.22.2 prerelease Apr 16 12:59:08 03Fraxinas  07org.openembedded.dreambox * r3d12a3d083 10openembedded.git/packages/gstreamer/ (3 files in 2 dirs): gstreamer: add recipe for gst-plugins-base 0.10.22.2 prerelease and plugin-common patches and updates Apr 16 12:59:12 03Fraxinas  07org.openembedded.dreambox * re40fc100a5 10openembedded.git/packages/gstreamer/gst-plugins-ugly_0.10.11.bb: gstreamer: add recipe for gst-plugins-ugly 0.10.11 release Apr 16 12:59:17 03Fraxinas  07org.openembedded.dreambox * re248af93a8 10openembedded.git/packages/gstreamer/gst-plugins-bad_0.10.11.bb: gstreamer: add recipe for gst-plugins-bad 0.10.11 release Apr 16 13:01:53 03Stanislav Brabec  07org.openembedded.dev * ra873d2ce98 10openembedded.git/recipes/mc/ (mc.inc mc_4.6.0.bb mc_4.6.1.bb mc_4.6.2.bb): mc: Fixed invalid references to staging bindir. Apr 16 13:42:47 Hey there, would anyone know what would be the proper way to handle a multi-arch package which is getting its sources from git? Having different subdirs in git named after the MACHINE? Then how to handle default? Apr 16 13:45:21 Ainulindale, take a look at linux-openmoko-2.6.28 Apr 16 13:45:35 the defconfig is machine-dependant Apr 16 13:45:51 Sir yes sir :-) Apr 16 13:46:40 Hmmmm Apr 16 13:46:56 Ok then if and so on Apr 16 13:47:44 Would it be a proper way then in configure_prepend to do S = something depending on the arch? Apr 16 13:50:06 shoragan: then another problem, I'd like to check out from two different git repositories Apr 16 13:50:11 How will that behave? Apr 16 13:50:22 (especially for gitr) Apr 16 13:51:26 Ainulindale, no idea :) Apr 16 13:51:41 well I'll try Apr 16 13:51:57 shoragan: as what we do is get some files from fso, then overwrite some other with ours Apr 16 13:52:17 Maybe we should patch rather than overwrite but it'd be a bit difficult to follow changes Apr 16 13:53:25 why don't you send us patches? Apr 16 13:53:42 or maintain a git branch Apr 16 13:54:33 because it's just config Apr 16 13:54:39 and I'd like that config to be "themable" Apr 16 13:54:48 so we have a frameworkd-config-shr Apr 16 13:54:55 Why not a frameworkd-config-shoragan Apr 16 13:55:00 with its ringtone, rules, and such Apr 16 13:55:06 That's why I built it that way Apr 16 13:55:40 then you should probably maintain a git branch, that way you can easily pull new changes Apr 16 13:56:43 perhaps using an oe distro here would make things easier, but you'll have to ask mickey Apr 16 14:04:42 Would anyone know how I would be able to deliberately put two different git repos in SRC_URI and use the gitr I want? Apr 16 14:08:29 are they for different machines? How would you choose between them? Apr 16 14:10:14 http://cgit.openembedded.net/cgit.cgi?url=openembedded/tree/recipes/linux/linux-msm7xxxx_git.bb Apr 16 14:10:32 Has examples of overriding the git rev for different machines. Apr 16 14:19:15 03Koen Kooi  07org.openembedded.dev * rf9a6f7e2fb 10openembedded.git/recipes/angstrom/ (10 files in 3 dirs): exquisite-theme-angstrom: import needed files after rasterman fixed the images Apr 16 14:31:45 Ainulindale: I guess that depends on how "the one that you want" is determined. Can you be more specific about what you're trying to do? Apr 16 14:33:27 well I have two gits Apr 16 14:33:31 I'm interested in the two srcrevs Apr 16 14:33:38 I was using SRCREV_FORMAT to do that properly Apr 16 14:33:41 with name= in my URIs Apr 16 14:33:46 it was working well Apr 16 14:33:53 I then tried to set specific rev for my names in the bb recipe Apr 16 14:34:02 i.e. SRCREV_pn-pn_NAME Apr 16 14:34:08 And that isn't working unless it's autorev Apr 16 14:35:11 Let me test some more and I'll pastebin what is necessary to understand my mess :-) Apr 16 14:35:33 I think mostly my problem comes from a non updated cache Apr 16 14:36:07 ah, you need SRCREV_pn-recipe = "REVONE-REVTWO" to match you SRCREV_FORMAT Apr 16 14:36:23 well having a look at bb it seemed this wasn't done like that Apr 16 14:36:49 rev = data.getVar("SRCREV_pn-" + pn + "_" + ud.parm['name'], d, 1) Apr 16 14:36:52 Especially that Apr 16 14:36:52 but git has the added problem is it always uses a directory called git/ and it all goes wrong from there Apr 16 14:37:03 Well it was fetching correctly here Apr 16 14:37:10 And I think we can override that Apr 16 14:37:41 I'll tell you if my thing works anyway :-) Apr 16 14:46:49 Seems I'm getting 1 everytime. Hmpf. Apr 16 14:47:52 re Apr 16 14:51:14 XorA: solution doesn't work, unfortunately Apr 16 14:51:32 Mine does but bitbake will generate a package with -1 instead of revA-revB Apr 16 14:52:20 What would generate the "package" name? Apr 16 14:56:09 oh. I belive that qmake2-native ought to die. qt4-tools-native should be used in place of it Apr 16 14:56:23 anyone has any objection to it? Apr 16 14:57:48 morning Apr 16 14:57:55 morning Apr 16 14:59:30 Well, nobody knows then :-/ Apr 16 14:59:38 mickey|zzZZzz? (worth a try :-p) Apr 16 15:02:18 Ainulindale: can you pastebin the actual bb file you're using? Apr 16 15:03:47 hi kergoth Apr 16 15:04:09 http://pastebin.com/m445c8a8f <= pb_ Apr 16 15:05:15 pb_: I think I got why Apr 16 15:05:32 And it seems I was right Apr 16 15:05:39 ah, very good :-) Apr 16 15:05:48 So, the problem is Apr 16 15:05:59 (let my bitbake end properly) Apr 16 15:06:11 (it's packaging) Apr 16 15:06:19 frameworkd-config-shr-0.8.5.1+gitr1277+542c7e13e021b609ec0aa1c20615d940adfcf403-5+fddd6e828e791f4ea69f89f15a8cdd7b1d5b540e-r3 <= nice name :-/ Apr 16 15:06:29 Will be great to dictate over the phone Apr 16 15:07:01 SRCREV_pn-frameworkd-config-shr = "${AUTOREV}" Apr 16 15:07:01 SRCREV_pn-frameworkd-config-shr_FSO_REV = "${AUTOREV}" Apr 16 15:07:01 SRCREV_pn-frameworkd-config-shr_SHR_REV = "fddd6e828e791f4ea69f89f15a8cdd7b1d5b540e" Apr 16 15:07:04 That's what I needed Apr 16 15:07:14 Without the first one, bitbake won't compute the proper name Apr 16 15:07:15 heh, right. you should really embrace the idea of making releases rather than using git checkouts :-} Apr 16 15:07:19 (which sounds a bit dumb) Apr 16 15:07:27 yeah I know :-) Apr 16 15:09:15 it does seem a little bit odd that that first line is needed, yes Apr 16 15:09:16 oh well Apr 16 15:09:26 Well as long as it's working properly, I don't care for now Apr 16 15:30:17 03Koen Kooi  07org.openembedded.dev * re9c95296cb 10openembedded.git/ (7 files in 3 dirs): linux-omap 2.6.29: sync with dss2 tree for YUV rotation Apr 16 15:41:48 can someone build 'alsa-oss' package? Apr 16 16:03:58 hmm. merging my changes is more complicated Apr 16 16:12:32 mmm: http://rafb.net/p/nUSido29.html Apr 16 16:12:49 I don't understand why I can bzcat the file but oe can't Apr 16 16:13:21 I'll check in what dir it is Apr 16 16:16:35 I'll try a quick fix Apr 16 16:30:40 mmm....now it fails to compile: http://rafb.net/p/Rwzb2L81.html Apr 16 16:57:21 mmm...I'll try a more recent version Apr 16 16:58:12 hey Apr 16 16:58:17 is "pkg-config --variable=defsdir pygtk-2.0" supposed to return /usr/share/pygtk/2.0/defs/ inside bitbake environment? Apr 16 16:58:27 looks like i don't understand everything :( Apr 16 17:04:09 Tartarus: ping Apr 16 17:05:07 ok then, maybe later... Apr 16 17:05:09 more recent version has long double prolbem...maybe looking at boost's patch Apr 16 17:05:24 * denix|away is away: I'm not here Apr 16 17:06:59 hi denix|away Apr 16 17:07:04 bah Apr 16 17:07:10 mmm...I'll remove that...UCLIBCXX_HAS_LONG_DOUBLE=y Apr 16 17:31:57 mmm...LDFLAGS...again... Apr 16 17:32:50 strange errors with bitbake here Apr 16 17:33:22 it says "could not include required file *****.inc while parsing *****.bb Apr 16 17:33:22 poi prosegue" Apr 16 17:33:33 is it ok if I only add --hash-style=gnu to the LDFLAGS? Apr 16 17:34:05 flameman, ??? come prosegue??? Apr 16 17:34:19 flameman, check if it's required or included Apr 16 17:34:25 in the recipe Apr 16 17:35:54 is it normal when bitbake sends --prefix=/usr to configure? Apr 16 17:35:59 instead of staging dir? Apr 16 17:39:54 dos1, i think so...but I'm not shure...I think it's the run prefix Apr 16 17:41:06 Gnutoo: $(datadir) etc. are in /usr too Apr 16 17:41:29 dos1, I think so...look at the existings recipes that work Apr 16 17:48:34 this is the error i get Apr 16 17:48:35 http://www.webalice.it/speeder/bitbake-error.txt Apr 16 18:02:37 strange... Apr 16 18:02:55 I've still problems with hash Apr 16 18:03:43 I'll try modifying the makefile...just to be shure Apr 16 18:11:25 flameman: do you have gpe-contacts.inc in metadata Apr 16 18:11:49 i don't know Apr 16 18:12:09 now i am git cloning the stable branch of bitbake Apr 16 18:32:30 how do I generate package for development (including headers and .so) Apr 16 18:32:52 keesj: you should have -dev package for the recipe Apr 16 18:35:05 I am using angstrom I guess I need to serach in the local.conf to enable -dev package generation for common packages? Apr 16 18:35:59 hmm I dont think so, which package are you having issue with Apr 16 18:36:16 keesj, they are already there Apr 16 18:37:15 in most cases -dev and -dbg packages are generated for recipes which include the stuff for such (-dev and -dbg) packages Apr 16 18:38:45 could some one help me to the bitbake already working ? Apr 16 18:38:47 I do see a lot of those packages indeed Apr 16 18:39:26 * keesj is obviousely trying to be able to compile on the target... Apr 16 18:39:31 i still get http://www.webalice.it/speeder/bitbake-error.txt Apr 16 18:40:39 flameman: either set your BBFILES or stop overriding BBMASK. the default BBMASK masks out nonworking from the parsing Apr 16 18:40:40 stuff/openembedded/recipes/nonworking/e2fsprogs/e2fsprogs_1.41.2 Apr 16 18:40:46 might be the clue Apr 16 18:42:26 kergoth this is the way i use bitbake http://www.webalice.it/speeder/bitbake-do-it.sh Apr 16 18:42:36 should i post the local.conf ? Apr 16 18:42:50 i just told you what you need to fix. you expect me to fix it for you, too? Apr 16 18:43:32 no, the point is .. i don't understand what you have suggest to do Apr 16 18:43:47 mmm....is is a good idea to link /usr/lib/libuClibc++.so.0 to libstdc++.so.6 ...is it the same ABI ? Apr 16 18:44:09 (could it be that it's not linked because of a bug?) Apr 16 18:44:25 flameman, grep for BBMASK. It seems that your variable includes nonworking stuff Apr 16 18:44:32 if you didn't understand what i just said, then you need to read the documentation. if you don't know what BBFILES is, which is one of the two most important variables in the use OpenEmbedded, you need to read up Apr 16 18:45:18 flameman, (BBMASK and BBFILES) Apr 16 18:45:38 i am new about that, anyway ... yesterday bitbake was working, today i svn update and the stuff stops to work Apr 16 18:45:47 this seems to me be a bit "strange" Apr 16 18:45:53 anyway i will read the doc Apr 16 18:46:08 svn update? oe uses git Apr 16 18:47:15 also i rm -r all the stuff and i git clone back again Apr 16 18:47:33 you can see what i did in the .sh i past included Apr 16 18:47:45 (that comes from the wiki) Apr 16 18:48:34 I'd better start from the beginning: Apr 16 18:50:06 I've a FAM problem : http://rafb.net/p/3zGDCx34.html Apr 16 18:50:11 what could I do Apr 16 18:50:25 do I need a specific libstdc++? Apr 16 18:50:56 I've already uclibc and I can't use recent glibc because they require nptl(I've a 2.4.19 kernel) Apr 16 18:51:11 Gnutoo: you need to install libstdc++ Apr 16 18:51:11 and I can't change that kernel Apr 16 18:51:24 khem, yes I've cross-compiled uclibc++ Apr 16 18:51:40 hmmm Apr 16 18:51:52 (I did a ugly hack but the new 0.2.2 version compiled) Apr 16 18:51:59 s/but/and/ Apr 16 18:51:59 is it staged Apr 16 18:52:07 can you use libstdc++ Apr 16 18:52:17 yes but under another name Apr 16 18:52:27 libuClibc++.so.0 Apr 16 18:52:42 they are not same Apr 16 18:52:43 should I make a simlink to it ? Apr 16 18:52:47 ok Apr 16 18:52:55 try to symlink Apr 16 18:52:59 so they are not compatible? Apr 16 18:53:00 ok Apr 16 18:53:03 thanks a lot Apr 16 18:53:31 I dont know if it will work out for you Apr 16 18:53:54 FAMopen might just be made to care for libstdc++ only Apr 16 18:54:11 I have not heard many people using glibc+uclibc++ Apr 16 18:57:59 i'm using uclibc Apr 16 18:58:54 oh that might be better than Apr 16 19:02:41 doesn't work Apr 16 19:05:03 is linker finding the c++ libraries ? Apr 16 19:05:24 if not then force it to link to uclibc++ provided libs Apr 16 19:07:58 ok Apr 16 19:08:02 thanks Apr 16 19:22:18 I'll try to do something like CXX=g++-uc Apr 16 19:43:40 mmm I think I will have to stage g++-uc Apr 16 19:43:45 but in the cross dir Apr 16 19:44:04 that is at /home/embedded/oetmp_pma/cross/arm-oabi/bin/ Apr 16 19:45:04 I'll look in the core recipes like GCC to see if there is a variable for that Apr 16 19:53:18 hmm, the dot files generated by bitbake -g are ... interesting Apr 16 19:56:36 gremlin[it]: hi Apr 16 19:56:49 hi mckoan ... Apr 16 19:58:05 * gremlin[it] spend evening with a broker for house loan ... grrrrrrrrrr Apr 16 19:59:01 gremlin[it]: happens :-) Apr 16 19:59:34 * gremlin[it] hope happen just a couple of time in a life ... Apr 16 20:00:01 gremlin[it]: indeed! :-D good luck and best wishes Apr 16 20:00:43 yes, i hope to be close to the end ... on august i want be in my house ! Apr 16 20:01:05 gremlin[it], congratulations! Apr 16 20:01:38 thanks Crofton|work Apr 16 20:01:52 and now let's bitbake ... Apr 16 20:05:05 I'm assuming this is mainly a bug in bitbake -g, but I notice that the dot files have a lot of lines like Apr 16 20:05:08 "task-boot" -> "b" [style=dashed] Apr 16 20:05:17 in other words the dependency names are being split characterwise :) Apr 16 20:07:15 seems to be happening for RDEPENDS but not DEPENDS Apr 16 20:12:08 hello, arm-angstrom-linux-uclibc-* that are in ${CROSS_DIR}/bin that is /home/embedded/oetmp/cross/arm-oabi/bin are defined how(how their name is defined?) Apr 16 20:13:22 Gnutoo, your question is ambiguous Apr 16 20:13:26 try harder :) Apr 16 20:13:54 ok how the name arm-angstrom-linux-uclibc get generated...what are the variables that govern that? Apr 16 20:14:35 I'll check for CROSS_DIR variable in bitbake -i Apr 16 20:15:14 iirc "linux" is ${DISTRO_OS}, "angstrom" is ${DISTRO_NAME}... Apr 16 20:15:28 gremlin[it]: what are you cooking tonight? Apr 16 20:15:29 ok thanks Apr 16 20:15:31 "angstrom" is ${DISTRO_VENDOR} Apr 16 20:15:38 grep for "DISTRO_" somewhere in openembedded/classes Apr 16 20:15:46 I think Apr 16 20:15:49 CROSS_DIR matches... Apr 16 20:15:52 thanks a lot Apr 16 20:16:04 dhd, don't know - no OE copy on the machine Apr 16 20:16:10 and arm is ${ARCH} Apr 16 20:16:30 i'll try theses variables in bitbake -i Apr 16 20:16:32 mckoan, KaeliOS on qemux86 with rpm sauce :) Apr 16 20:17:35 gremlin[it]: hmmmm taste good ! Apr 16 20:18:04 mmm...TARGET_OS is linux-uclibc Apr 16 20:18:11 ARCH is none...that is strange Apr 16 20:18:21 ahh, TARGET_VENDOR Apr 16 20:18:33 ah TARGET_ARCH Apr 16 20:28:52 Gnutoo: still fighting against uclibc ? Apr 16 20:29:03 yes and no Apr 16 20:29:08 I will try something Apr 16 20:29:36 and after I would need advise on the best way to prepare the recipe for review Apr 16 20:29:54 what I will try: Apr 16 20:30:13 export CXX=arm...-g++-uc Apr 16 20:30:29 and it was uclibc++ Apr 16 20:30:31 not uclibc Apr 16 20:30:37 uclibc is ok now Apr 16 20:31:00 in uclibc I think I changed the defconfig Apr 16 20:31:14 and made better kernel headers Apr 16 20:36:44 dhd: bitbake -g suffers a known bug... Apr 16 20:38:38 also, I have my own custom image type, which does not depend on the whole task-base, but does depend on task-base-usbhost - yet for some reason, bitbake wants to build all of task-base anyway Apr 16 20:39:37 in particular, task-base-bluetooth pulls in a lot of stuff I don't ever want to build Apr 16 20:40:00 should I just avoid touching anything from task-base and add the relevant packages explicitly? Apr 16 20:40:36 dhd: for a fix look at http://bugs.openembedded.net/show_bug.cgi?id=5002 Apr 16 20:40:42 kergoth: ^^^ Apr 16 20:41:29 03Stanislav Brabec  07org.openembedded.dev * r0ff8efdbbf 10openembedded.git/recipes/navit/ (navit.inc navit_0.0.4.bb navit_0.1.0.bb navit_svn.bb): navit: Modules .so files must go to the main package, modules .la can go away. Apr 16 20:41:30 03Stanislav Brabec  07org.openembedded.dev * r6898e13ea0 10openembedded.git/recipes/espeak/espeak_1.37.bb: espeak: Fixed GNU_HASH QA error. Apr 16 20:42:08 ant__: looks correct to me at first glance. might want to run it by RP or zecke to get a second opinion Apr 16 20:42:30 almost 3 months... Apr 16 20:42:46 since Roman posted the fix Apr 16 20:42:49 ant__: thanks! Apr 16 20:42:52 mmm...it still fail: http://rafb.net/p/S7m1G277.html Apr 16 20:43:10 where did he post the fix? Apr 16 20:43:21 attachment to the bug Apr 16 20:43:38 ...but nobody cares about bugzilla lately Apr 16 20:43:58 urg, it is in bitbake Apr 16 20:44:02 yep Apr 16 20:44:04 fewer people can commit that Apr 16 20:44:09 I bet I'll have to hack the build system Apr 16 20:44:24 today RP was there..I forgot... Apr 16 20:44:25 ant__, best thing is to ask people who you know can commit to bitbake Apr 16 20:44:33 i can commit it, but i want to test it out a bit first, i donj't want to inadvertantly break anything Apr 16 20:44:38 heh Apr 16 20:44:44 kergoth: :) Apr 16 20:45:11 there is another patch of Roman, supposedly solving some race issues Apr 16 20:45:15 there are a number of Roman's patches in patchwork that need reviewing Apr 16 20:45:26 (which I happily don't have..yet) Apr 16 20:45:31 those are easier to find Apr 16 20:45:53 Crofton|work: seems a valid contributor Apr 16 20:46:05 yeah Apr 16 20:46:06 today monopolyzed ML :-) Apr 16 20:46:14 he touiches stuff that scares me Apr 16 20:46:22 yeah.. Apr 16 20:46:28 * kergoth always avoided looking anywhere near the multilib crap Apr 16 20:46:33 Tom and Khem should evaluate Apr 16 20:46:44 Stanislav too Apr 16 20:47:15 toolchain is yet a mistery for me in some aspects ;-) Apr 16 20:47:58 kergoth: yea, multilib is not for purists Apr 16 20:49:24 I'll try and look it over Apr 16 20:50:01 toolchain is relatively complex, just due to how many steps are involved Apr 16 20:52:08 let say I thought 'I do understand' just before playing with cross-compile Apr 16 20:52:13 :/ Apr 16 20:58:48 03Stanislav Brabec  07org.openembedded.dev * r0a9c8004d0 10openembedded.git/recipes/pulseaudio/libatomics-ops_1.2.bb: libatomics-ops: Package all installed files. Apr 16 20:58:49 03Stanislav Brabec  07org.openembedded.dev * r888279d292 10openembedded.git/recipes/eds/eds-dbus_svn.bb: eds-dbus: Package missing files, remove unneeded .la files. Apr 16 20:58:49 03Stanislav Brabec  07org.openembedded.dev * r96cb8adcf1 10openembedded.git/recipes/libsndfile/libsndfile1_1.0.17.bb: libsndfile1: Package unpackaged octave files to subpackage libsndfile1-octave. Apr 16 20:59:13 oh, Stanislav is fixing navit...good Apr 16 21:02:22 hey dhd Apr 16 21:02:40 hey Apr 16 21:02:52 So, another canadian user, eh? ;) Apr 16 21:02:59 haha yes Apr 16 21:03:05 Sorry, had to Apr 16 21:03:21 amazingly enough I am actually Canadian Apr 16 21:03:30 but mainly I just don't enjoy using Cygwin :) Apr 16 21:04:04 So I'm gonna give both of those patches a shot w/ DISTRO=minimal and MACHINE=ep93xx Apr 16 21:04:19 The drive letter thing is odd Apr 16 21:04:30 as I did a buncha tests locally, but not quite w/ DISTRO=minimal Apr 16 21:04:42 yeah - I can send you my distro/machine if you like Apr 16 21:04:55 but it is basically ep93xx minus some machine features Apr 16 21:05:30 the board is a Qwerk, which is essentially an edb9302a plus some robot peripherals hooked up to GPIO lines Apr 16 21:05:37 Well, so long as minimal/ep93xx build currently, thats good enough :) Apr 16 21:06:14 hmm Apr 16 21:06:21 you didn't c&p into the email did you? Apr 16 21:06:32 Hunk #1 FAILED at 1. Apr 16 21:06:32 Hunk #5 succeeded at 139 (offset 6 lines). Apr 16 21:06:32 1 out of 5 hunks FAILED -- saving rejects to file recipes/meta/canadian-sdk.bb.rej Apr 16 21:06:42 oh the patches are against stable/2009 Apr 16 21:06:47 Ah Apr 16 21:06:52 I should have mentioned that... Apr 16 21:07:21 Er, I didn't think I did driveletter in stable.. Apr 16 21:09:30 ahh canadian-sdk.bbclass is a bit different in .dev, I see Apr 16 21:09:41 sorry, I should be using .dev but I have enough problems already... Apr 16 21:09:52 heh, I can understand that Apr 16 21:09:53 actually quite different it seems Apr 16 21:10:12 Did you try moving stuff from .dev to stable w/o git cherry-pick? Apr 16 21:11:08 no, not yet... Apr 16 21:11:22 how does SDK_REALPATH_MINGW get picked up? is it in the canadian recipes? Apr 16 21:12:02 oh, I see Apr 16 21:12:09 see ae48ce485442fa2c1a7d0552bf6b4ea10fe692cf Apr 16 21:12:11 so I could cherry-pick the gcc recipes and the canadian classes Apr 16 21:12:15 dhd: .dev is not so bad today...just checkout a local branch and don't pull / rebase on it Apr 16 21:12:38 make your stable today... Apr 16 21:12:56 thanks, that sounds reasonable Apr 16 21:13:08 or better tomorrow...lot of hanging patches :-) Apr 16 21:13:37 yeah, unfortunately I am going to a conference for a week on Saturday Apr 16 21:14:07 well, you can cherry-pick those Apr 16 21:14:40 but really was unfortunate moment when stable was branched Apr 16 21:14:42 yeah, that seems like a wise idea - I think that I want to keep tracking stable/2009 but incorporate stuff that this project needs Apr 16 21:15:06 Well, getting canadian stuff into stable is on my todo list Apr 16 21:15:20 really it should just be git cherry-pick, submit and then push Apr 16 21:15:53 cool, thanks! Apr 16 21:16:15 it's very nice to be able to build toolchains for people running Eclipse on Windows Apr 16 21:16:41 yeap Apr 16 21:17:05 the drive letter thing is needed so they can install on a diff drive than where they work Apr 16 21:17:10 ahh Apr 16 21:17:11 had someone report that as a problem :) Apr 16 21:17:25 there is one other problem with the MinGW canadian-sdk Apr 16 21:17:33 which is that some Linux header files have names that differ in case only Apr 16 21:17:38 Yes :( Apr 16 21:17:41 we heve on-line log of this channel ? Apr 16 21:17:47 Locally, I've got an "rm -rf netfilter" Apr 16 21:17:49 right now I just have to tell people to click "Yes to All" Apr 16 21:17:58 gremlin[it]: ~logs Apr 16 21:18:06 ah, I'll do that - I didn't have time to look for them all Apr 16 21:18:11 There's no real good solution to the problem Apr 16 21:18:13 ~logs Apr 16 21:18:14 All conversations are logged to http://ibot.rikers.org/channel, where "channel" is replaced by the URL-encoded channel name, such as %23freenode for #freenode. Lines starting with spaces are not logged. Apr 16 21:18:23 It will break those, of course Apr 16 21:18:31 But it's a limitation on Windows Apr 16 21:18:42 !logs Apr 16 21:18:42 Channel logs for #oe are archived at: Apr 16 21:18:43 http://hentges.net/tmp/logs/irc/%23oe Apr 16 21:18:44 Live-logs are available at Apr 16 21:18:46 http://hentges.net/tmp/logs/irc/livelogs/%23oe.livelog Apr 16 21:18:48 See ?? help-logs for usage instructions Apr 16 21:18:53 there are livelog^^ Apr 16 21:19:02 yeah, but really, those header files are probably not useful to the average person building embedded apps from Windows anyway Apr 16 21:19:06 if they cared, they'd use Linux :) Apr 16 21:19:16 Basically Apr 16 21:19:20 it's netfilter fun Apr 16 21:19:43 thanks Jay7 Apr 16 21:19:48 Case insensitivity is bad. Apr 16 21:19:48 rm -rf ${SDK_OUTPUT}/${prefix}/${TARGET_SYS}/usr/include/linux/netfilter Apr 16 21:20:00 netfilter* even, stupid 'screen' :) Apr 16 21:20:27 Also means no kernel building on wIndows Apr 16 21:20:58 I think I've a big problem with uclibc++: http://rafb.net/p/ePGWYB73.html Apr 16 21:21:04 Tartarus: and on OSX... Apr 16 21:21:14 ant__, OSX by default :) Apr 16 21:21:24 he Apr 16 21:21:35 Windows it's much harder to get a case sensitive fs, iirc Apr 16 21:21:52 osx at least has that as an option to hfs+, but it breaks something or another Apr 16 21:21:55 or used to Apr 16 21:24:35 ok probably fixed the rpm-native issue ... Apr 16 21:25:45 Tartarus: why loose time searching for new problems with rare buildhosts?There are enough natives...! Apr 16 21:34:14 hello all Apr 16 21:36:44 dead channel is dead :I Apr 16 21:39:38 not really Apr 16 21:40:17 gn Apr 16 21:40:39 ah Apr 16 21:41:04 i guess im just used to more lively channels Apr 16 21:41:25 heh Apr 16 21:41:30 this channel can be very lively Apr 16 21:41:43 sounds promising then :3 Apr 16 21:41:52 the .eu guys are in bed and the .us guys are working Apr 16 21:42:14 makes sense Apr 16 21:42:37 tbh though, im hunting for answers Apr 16 21:42:42 Only us Atlantic Tuna here. Apr 16 21:42:48 regarding SPI Apr 16 21:44:58 even more specifically about the slave-select function in a micro's implementation of spi Apr 16 21:47:04 I'm back...my Internet provider should have had a problem Apr 16 21:47:14 I think I've a big problem with uclibc++: http://rafb.net/p/ePGWYB73.html Apr 16 21:53:35 what should I do? Apr 16 22:05:16 03Angus Ainslie  07fso/milestone5.5 * r6e4ad95def 10openembedded.git/recipes/freesmartphone/frameworkd_git.bb: frameworkd : patch otimed to correct timezones Apr 16 22:05:19 03Angus Ainslie  07fso/milestone5.5 * r5e74e5d110 10openembedded.git/conf/distro/include/sane-srcrevs.inc: sane-srcrevs : bump fso frameworkd to fix idlenotifier Apr 16 22:11:18 BusError: Do you want a fix for the lowlevel_foo thingy for u-boot? Apr 16 22:12:31 I managed to kill my uboot earier today but luckily the procedure to recover is documented and the jtag dongle shipped with the device works pretty well with openocd. Apr 16 22:13:58 mmm...is it possible to mix uclibc and libstdc++ 6 and does libstdc++ depend on nptl or things not available for 2.4 kernel? Apr 16 22:24:33 florian, still can't reach mini ? Apr 16 22:25:03 BusError: no, works pretty good now. Apr 16 22:25:04 florian, you can always restart with the vivi bootloader on the NOR.. I hardly had to use JTAG to develop the u-boot port Apr 16 22:25:23 Gnutoo: C++ runtime libs do not depends on thread model Apr 16 22:25:34 wow Apr 16 22:25:44 BusError: hrm... i supposed that it would work like this but for some reason it failed for me Apr 16 22:25:50 florian, so what the lowlevel problem exactly ? Apr 16 22:25:54 thanks a lot...so how do I get it then? bitbake gcc-cross didn't stage it Apr 16 22:26:17 florian, load u-boot in ram at 33f80000 using vivi, run it, and fix the nand :> Apr 16 22:27:22 BusError: some typo inherited from some other device, it was never built and the CFLAGS to build it were wrong. Apr 16 22:27:37 unless you use an oabi toolchain maybe... Apr 16 22:27:46 I used JTAG only to figure where I was hanging :> Apr 16 22:28:02 :-) Apr 16 22:28:09 florian, I use oabi... Apr 16 22:28:20 florian, hmm I tought I had fixed that, somehow :> yes do forward it on Apr 16 22:28:36 what are the options lefts...I need glib-2.0 Apr 16 22:28:54 BusError: the dongle that is shipped with the device? using openocd or urjtag? Apr 16 22:29:09 I used openocd, and a JTAGKey Apr 16 22:29:18 urjtag doesn't grock the s3c Apr 16 22:29:31 (it doesn't find FAM...and I want to try fam because otherwise it fails because I've no inotify) Apr 16 22:30:20 florian, I don;t have a parallel port on my workstation :> Apr 16 22:30:55 :-) Apr 16 22:31:32 That's why I have a _real_ workstation and an old laptop with old-style ports :) Apr 16 22:32:41 and thats why I hsve a jtagkey :> Apr 16 22:34:44 I have something like five jtags but all of them suck for some reason. e.g. any IXP43x device fails with J-Link and ARm-USB-OCD Apr 16 22:37:57 I did a tad of kernel hacking. gave up on the OEM backlight, I think it's not hooked up at all. Instead hooked the LCD on/off to the VESA blanking, do we have proper deep screen sleep Apr 16 22:38:12 I'll push a new kernel patch later Apr 16 22:40:10 ehm..that's recent... Apr 16 22:40:10 ERROR: Multiple .bb files are due to be built which each provide bluez-libs (/oe/org.openembedded.dev/recipes/bluez/bluez4_4.35.bb /oe/org.openembedded.dev/recipes/bluez/bluez-libs_3.36.bb). Apr 16 22:40:25 BusError: do you have one with a small display? Apr 16 22:40:41 florian, yup, I have a 3.5 and a 7" Apr 16 22:41:11 mmm...found the lib...I'll try to link to it Apr 16 22:41:15 it was in cross dir Apr 16 22:41:27 BusError: I noticed that the display timing for the 3.5 is bad. Apr 16 22:41:42 Hey, I made a recipe for xf86-input-evtouch. Want a patch? Apr 16 22:42:22 florian, ? works fine here... how does it look to you? Apr 16 22:43:48 BusError: maybe the display for the micro is different (i don't hope so). i miss some colums at the right border Apr 16 22:44:34 and there is a timimg problem. its not really flickering but i can see that rows move a little bit Apr 16 22:44:38 the chassis of your screen looks different. from the only pic I saw of the micro.. Apr 16 22:45:00 i'll takea look at the datasheet some day... Apr 16 22:45:47 in the kernek siurce, try to decrease framerate. it's specced fairly high because it works fine on the mini Apr 16 22:46:41 mmm....I'll set the rpath... Apr 16 22:49:32 BTW on a general topic, how do I change the timezone on OE/Angstrom ? didn't find anything in /etc... Apr 16 22:52:04 hrm /etc/timezone I assume Apr 16 22:52:35 I assumed too. it's not there. I wonder if it's just a missing link Apr 16 22:52:54 Did you install tzdata? Apr 16 22:53:29 check /etc/bootmisc.sh Apr 16 22:53:44 now using /etc/timestamp Apr 16 22:53:49 hmm no it's not on the default angstrom it seems Apr 16 22:55:43 sorry etc/init.d/bootmisc.sh Apr 16 22:56:39 and somewhere TZ is exported Apr 16 22:58:06 yes, /etc/init.d/hwclock.sh Apr 16 22:58:47 thanks a lot for the help!!! Apr 16 22:59:42 * florian needs some sleep Apr 16 22:59:45 good night Apr 16 23:00:34 'night flo Apr 16 23:01:12 'night Apr 16 23:08:27 03Angus Ainslie  07fso/milestone5.5 * re4f800531b 10openembedded.git/recipes/freesmartphone/frameworkd/ (om-gta01/otimed.patch om-gta02/otimed.patch otimed.patch): frameworkd : otimed patch Apr 17 00:56:26 Has anybody else had problems with task-java recently? Apr 17 01:44:13 w00t i think my build finaly finished **** ENDING LOGGING AT Fri Apr 17 02:59:57 2009