**** BEGIN LOGGING AT Thu Feb 28 10:59:56 2008 Feb 28 11:01:19 03Laibsch 07org.oe.documentation * rb19cee79... 10/ (1 usermanual/chapters/common_use_cases.xml): common_use_cases.xml: correct path to sane-srcrevs.inc. Thanks Junqian for spotting the error. Feb 28 11:01:20 03Laibsch 07org.oe.documentation * r96793b73... 10/ (1 usermanual/usermanual.xml): usermanual.xml: add 2008 to the copyright years Feb 28 11:01:26 03mickeyl 07org.oe.dev * rafcd0ed7... 10/ (1 packages/python/python-2.5.2/sitecustomize.py): python 2.5.2 silence sitecustomize.py Feb 28 11:01:31 03mickeyl 07org.oe.dev * rd954d523... 10/ (11 files in 3 dirs): dbus 1.1.4 update to 1.1.20 (new default version) Feb 28 11:01:37 03mickeyl 07org.oe.dev * r3280ed49... 10/ (4 files in 3 dirs): Feb 28 11:01:37 busybox 1.9.1 factor more common things into busybox.inc Feb 28 11:01:37 disable utilities again until we're a proper replacement for tinylogin Feb 28 11:01:42 03mickeyl 07org.oe.dev * rcdf0fce8... 10/ (1 conf/distro/include/preferred-om-2008-versions.inc): preferred-om-2008-versions.inc: update entries for dbus and dbus-glibc Feb 28 11:01:48 03mickeyl 07org.oe.dev * ra7772b58... 10/ (1 packages/freesmartphone/pyneod_svn.bb): pyneod svn fix init script name Feb 28 11:02:15 o.. dbus-glibc Feb 28 11:02:20 * hrw waits for dbus-uclibc ;D Feb 28 11:07:29 oops Feb 28 11:08:00 mickeyl: I just removed recipe which you had to look at Feb 28 11:08:03 few years ago Feb 28 11:09:21 heh Feb 28 11:09:22 which one? Feb 28 11:09:39 kgrok Feb 28 11:12:01 hmm, don't remember this one Feb 28 11:12:03 excellent Feb 28 11:12:05 less recipes Feb 28 11:12:07 less maintenance :) Feb 28 11:13:26 hi all, does someone have trouble with kernel 2.6.24 and jffs2 images created with mtd-utils < 1.1.0 ? Feb 28 11:13:37 mickeyl: look at nonworking/python/ please Feb 28 11:15:17 i have to repack the jffs2 images created with oe and mtd-utils 1.0.0 in order to have a working image on simpad Feb 28 11:15:23 with kernel 2.6.24 Feb 28 11:15:51 hrw: any thing you want to bring back in particular? Feb 28 11:17:56 mickeyl: no, just want to know does some of it is not obsoleted already Feb 28 11:18:51 gm Feb 28 11:18:54 hrw: right. I'll take a look and throw away (perhaps) Feb 28 11:19:54 mickeyl:hrw: have you compiled busybox-static lately? Feb 28 11:20:29 mickeyl: One of your long requested features is now reality (recursive stamps) Feb 28 11:20:35 I found a broken file... /include/asm/posix_types.h is uncomplete Feb 28 11:20:51 only under uclibc Feb 28 11:23:10 RP: woot! that's really good news Feb 28 11:23:15 ant|work: no, not for ages Feb 28 11:23:17 ant|work: I did not build any image from OE during last month Feb 28 11:23:28 ant|work: i'll write that down and try when i have a chance Feb 28 11:26:59 k Feb 28 11:27:30 mickeyl: BB_STAMP_POLICY="full". BB_SCHEDULER="completion" is good for rm_work too... Feb 28 11:28:06 hmm yeah! Feb 28 11:28:17 i guess we need to have good documentation on all that new stuff ;) Feb 28 11:28:43 mickeyl: Its in the Changelog, the bitbake manual needs some tlc Feb 28 11:28:50 *nod* Feb 28 11:29:09 bitbake pretty much does all i need nowadays Feb 28 11:29:16 the only thing really missing is Feb 28 11:29:20 regexp in FILES Feb 28 11:29:26 that would make a lot of recipes much simpler Feb 28 11:30:03 Agreed, we just need a way to transition to that... Feb 28 11:30:15 right Feb 28 11:30:24 Its an OE problem, bitbake doesn't care about FILES... Feb 28 11:30:34 oh of course, right Feb 28 11:31:18 until then i find myself hacking more and more with PACKAGES Feb 28 11:31:25 e.g. the order sucks a bit Feb 28 11:31:33 we should have said ${PN}-dev ${PN} Feb 28 11:31:38 instead of ${PN} ${PN}-dev Feb 28 11:31:54 because i have a hard time getting binconfig and pkgconfig stuff into ${PN}-dev Feb 28 11:32:58 Maybe a recipe can elect to use "RFILES" instead of FILES Feb 28 11:33:21 Then we can transition on a per recipe basis Feb 28 11:33:22 yes, with any RFILES overriding FILES -- that would seamlessly Feb 28 11:33:27 +be Feb 28 11:33:27 mickeyl: Error: "be" is not a valid command. Feb 28 11:33:33 ~lart dtahtbot Feb 28 11:33:33 * ibot holds dtahtbot to the floor and spanks him with a cat-o-nine-tails Feb 28 11:34:00 +rootme Feb 28 11:34:02 XorA: Error: "rootme" is not a valid command. Feb 28 11:34:27 XorA: ubuntu style: To execute rootme you have to "apt-get install root-kit" Feb 28 11:34:30 mickeyl: We'd have to stipulate FILES and RFILES are not to be mixed... Feb 28 11:35:07 RP: i agree. how can we check that? Feb 28 11:35:22 or rather Feb 28 11:35:33 should we check that and issue a warning or try to rely on the users? Feb 28 11:35:56 mickeyl: Set RFILES_ACTIVE = "1" in converted recipes? Feb 28 11:36:42 hmm, with at some point of time, that becoming the default? Feb 28 11:37:31 something more automatic would be nice though Feb 28 11:37:31 mickeyl: We could add "FILES_ACTIVE" to every .bb and then make RFILES the default :) Feb 28 11:37:52 RP: that sounds better ;) Feb 28 11:38:03 or: if RFILES != "" then ignore FILES Feb 28 11:38:10 and default RFILES="" Feb 28 11:38:15 yeah, that was my first idea Feb 28 11:38:21 but its ugly Feb 28 11:38:26 RFILES just overriding FILES Feb 28 11:38:29 but we want a sane set of RFILES defaults Feb 28 11:38:35 *nod* Feb 28 11:38:48 Ideally recipes won't need to mess with them as much as they currently do Feb 28 11:38:57 No more .debug madness Feb 28 11:39:02 yeah! Feb 28 11:39:06 that's really annyoing Feb 28 11:39:11 03hrw 07org.oe.dev * r5d2e5eb2... 10/ (7 files in 4 dirs): kgrok: remove from repository - I added it few years ago and no one fixed it Feb 28 11:39:15 03hrw 07org.oe.dev * r950d3330... 10/ (5 files in 4 dirs): lftp: removed 3.2.1 from nonworking - we have 3.5.1 Feb 28 11:39:21 03hrw 07org.oe.dev * rb14797ca... 10/ (15 files in 7 dirs): apache: drop 2.0.x from nonworking - we have 2.2.x working Feb 28 11:39:26 03hrw 07org.oe.dev * r22acec5f... 10/ (3 files in 3 dirs): lzma: dropped nonworking 0.01 version Feb 28 11:39:32 03hrw 07org.oe.dev * r17c77827... 10/ (10 files in 6 dirs): dropped minisip and its components from nonworking - we have them in packages/ too Feb 28 11:41:07 mickeyl: I'm also concious we're going to need to convert our staging routines at some point too to get speed from packages staging. I'm hoping we can reuse do_install as do_stage in future. I'm thinking a flag might be needed for that too though :/ Feb 28 11:41:35 First I'll make it work though, then we can make it fast... Feb 28 11:41:41 RP: i see where you want to go, it will not be simple though Feb 28 11:41:45 we have some nasty stage tricks Feb 28 11:41:53 in more than just a few recipes Feb 28 11:42:04 mickeyl: What kind of nasty tricks? Feb 28 11:42:31 well, casese where we need to install additional or much less or completely different things into staging Feb 28 11:42:58 those won't be easy to unify Feb 28 11:43:04 but in the end it's probably worth it Feb 28 11:43:06 mickeyl: For speed the main issue is we need to start using DESTDIR for staging (i.e files aren't directly written there) Feb 28 11:43:35 mickeyl: Converting anything using autotools shouldn't be too bad... Feb 28 11:44:45 mickeyl: The need for completely different things into staging should hopefully be reducing a lot too Feb 28 11:45:46 RP: agreed Feb 28 11:45:53 03mickeyl 07org.oe.dev * r90ce09fb... 10/ (1 classes/distutils-base.bbclass): distutils-base.bbclass: package ${PN}-dev first, so that pkgconfig gets a chance to land there Feb 28 11:45:57 03mickeyl 07org.oe.dev * r71782975... 10/ (1 packages/python/python-dbus_0.82.4.bb): python-dbus 0.82.4 fix packaging, so that development stuff actually ends up in -dev packages Feb 28 11:46:01 03mickeyl 07org.oe.dev * rfc5266af... 10/ (1 packages/images/moko-underground-image.bb): moko-underground-image: add pyneod, pyneog, gsm0710muxd Feb 28 11:48:24 Down to 15 bitbake bugs :) Feb 28 11:49:03 RP: I'm actually forced to mix glibc and uclibc for that initramfs...this leads to staging issues, like preferred providers for libc (it finds an "externa-toolchain") Feb 28 11:49:41 RP: which workaround would you propose? (http://lists.linuxtogo.org/pipermail/openembedded-devel/2008-February/004470.html) Feb 28 11:51:25 ant|work: I did have some ideas but I was getting rather annoyed with the way Paul was doing things and took a step back from that. I've now forgotten the idea I had. I'll have to think about it again at some point... Feb 28 11:52:01 RP: I'm at the point I got an initramfs-kexec-image built under uclibc/klibc Feb 28 11:52:11 what notation is that? bash? if [ "${TARGET_OS}" = "linux-gnueabi" -o "${TARGET_OS}" = "linux-uclibcgnueabi" ]; then Feb 28 11:52:25 Jin^eLD: yes, shell script Feb 28 11:52:44 ant|work: Wipe external-toolchain out locally ;-) Feb 28 11:52:52 Koen's changes seem to be blackfin oriented, not uclinux in general Feb 28 11:53:06 Jin^eLD: yes, that would make sense Feb 28 11:53:35 RP: I thought I should build the initramfs-bootmenu-image under uclibc just to see if it fits in /mtd Feb 28 11:53:49 RP: well, not quite :) the uclinux related stuff is probably similar to a number of different machines Feb 28 11:54:16 but I guess he was only focusing on bfin Feb 28 12:05:26 03rpurdie 07org.oe.dev * r81d8dc81... 10/ (1 classes/sip.bbclass classes/sip3.bbclass): sip(3).bbclass: Fix do_generate task dependencies (#3765) Feb 28 12:05:40 03rpurdie 07org.oe.dev * r5210c083... 10/ (1 classes/mono.bbclass packages/mono/README): Feb 28 12:05:40 mono.bclass: Fix the location used for mono clilibs information to be host Feb 28 12:05:40 specific otherwise this will break when multiple arch builds are made. shlibs Feb 28 12:05:41 was fixed in a similar way. Note that mono apps will need to be rebuilt after Feb 28 12:05:43 this change Feb 28 12:05:47 03mickeyl 07org.oe.dev * re657326e... 10/ (1 packages/python/python-gst_0.10.10.bb): python-gst 0.10.10 fix packaging, increase package granularity Feb 28 12:10:57 im using dosfstools ; i am not able to format a 4G disk using it. mkfs.vfat doesnot give me an error but when i mount it. i get a fs of 1G size only Feb 28 12:15:01 please see : http://pastebin.ca/921857 Feb 28 12:24:26 03rpurdie 07org.oe.dev * r69da36fd... 10/ (1 packages/radlib/x): radlib: Drop strange file Feb 28 12:24:38 03rpurdie 07org.oe.dev * r36b72e8f... 10/ (6 files in 6 dirs): nonworking: Attempt to fix up various random broken staging references whilst in the area Feb 28 12:24:45 03rpurdie 07org.oe.dev * r30b674cf... 10/ (5 files in 3 dirs): uqm: Fix 0.5.0 build assumptions with a patch, fix staging references Feb 28 12:24:51 03rpurdie 07org.oe.dev * r837d64a3... 10/ (1 packages/uclibc/uclibc.inc): uclibc: Clean up so only one set of headers are staged and better staging variables are used Feb 28 12:24:57 03rpurdie 07org.oe.dev * r4d7696be... 10/ (1 packages/tremor/tremor_svn.bb): tremor: Use autotools_stage_all Feb 28 12:25:09 shri: first of all: "mkfs.vfat -F 32" Feb 28 12:25:15 03rpurdie 07org.oe.dev * rea8187e2... 10/ (1 packages/startup-notification/startup-notification_0.5.bb): startup-notification: Use autotools_stage_all Feb 28 12:25:21 03rpurdie 07org.oe.dev * rdc145b86... 10/ (7 files in 4 dirs): qt, qte, qt4: Use better staging variables Feb 28 12:25:27 03rpurdie 07org.oe.dev * ra350c95a... 10/ (4 files in 2 dirs): libpixman: Use autotools_stage_all Feb 28 12:25:34 03rpurdie 07org.oe.dev * rabefd5c1... 10/ (5 files in 2 dirs): classes: Improve various staging variable references Feb 28 12:25:39 03rpurdie 07org.oe.dev * raea71c0a... 10/ (3 files in 2 dirs): eglibc: Remove hardcoded layout, improve staging path references Feb 28 12:25:44 03rpurdie 07org.oe.dev * r2c34202b... 10/ (1 packages/meta/meta-toolchain.bb): Feb 28 12:25:44 meta-toolchain: Sync with Poky. Add site-config file with distilled site Feb 28 12:25:44 configuration from OE, add environment setup script to tarball, add version file Feb 28 12:25:44 to tarball. Generate extras package with tarball separately since this Feb 28 12:25:45 information is only needed when using the toolchain with OE as an external Feb 28 12:25:47 toolchain. Change the way the target packages are installed so fix bugs with the Feb 28 12:25:49 previous approach Feb 28 12:29:45 03mickeyl 07org.oe.dev * re81217cc... 10/ (1 conf/distro/include/sane-srcrevs.inc): sane-srcrevs: bump pyneod, pyneog, gsm0710muxd, catch up with renaming gpio to s3c24xx-gpio Feb 28 12:31:35 hi, all! Feb 28 12:31:55 Do anybody remmeber which touch screen is on mainstone? Feb 28 12:53:04 morning all. Feb 28 12:53:50 afternoon Feb 28 12:54:01 hi thesing Feb 28 12:54:07 thesing: how goes your pc? Feb 28 12:54:39 hrw: I got a new board a week ago. I told you, did I? Feb 28 12:55:05 but I did not asked does it work ;) Feb 28 12:55:35 I works. Feb 28 12:55:43 great Feb 28 12:55:47 slapin: The coments in arch/arm/mach-pxa/mainstone.c say IIRC. Feb 28 12:56:54 I can't use usb-storage as a normal user, but I think this is a debian-sid thing. I think it worked in September. Feb 28 12:57:36 * Crofton now understands the lichenstein references from the OE ev meeting Feb 28 12:58:58 broonie, which tree? Feb 28 12:59:04 mainline Feb 28 12:59:21 broonie, actually I look for patch to support. Feb 28 12:59:46 broonie, I have checked out tree and there's no any mention of touchscreen in mainline tree Feb 28 13:00:18 slapin: Oh, the touchscreen itself. It'll be using whatever the default thing on the AC97 bus is which is some Phillips thing IIRC. Feb 28 13:00:26 I thought you meant the screen. Feb 28 13:00:43 ALSA should be able to tell you. Feb 28 13:01:56 broonie, as per 2.6.9 kernel from them it is UCB1X00, and I see it under drivers/mfd, but is is only for StrongArm, while my mainstone is pxa270 Feb 28 13:02:41 * broonie has no idea about drivers for it - I've only used the touchscreen on Mainstone with Wolfson codecs doing the input side. Feb 28 13:08:04 does not it depends on which screen/touchscreen is connected? Feb 28 13:09:54 03thesing 07org.oe.dev * r452ba8f4... 10/ (4 files in 2 dirs): Feb 28 13:09:54 klibc.inc: reformat SRC_URI, add fstype-sane-vfat-and-jffs2 patch Feb 28 13:09:54 Change of SRC_URI was necessary because the eabi patch wasn't always appllied Feb 28 13:09:54 by bitbake. Feb 28 13:10:01 03dyoung 07org.oe.dev * r9366ce82... 10/ (1 packages/gcc/gcc_4.2.2.bb): bump PR, forgot to do it 11 days ago when re-adding 801 series patch. Feb 28 13:10:06 03rpurdie 07org.oe.dev * r22484b6d... 10/ (4 files in 2 dirs): mpfr: Convert to autotools_stage_all Feb 28 13:10:11 03rpurdie 07org.oe.dev * r3e7486d8... 10/ (1 packages/sox/sox-native_13.0.0.bb): sox-native: Fix STAGING_DIR reference Feb 28 13:10:16 03rpurdie 07org.oe.dev * ra4ca2e50... 10/ (1 packages/skippy/skippy-xd_0.5.0.bb): skippy-xd: Fix hardcoded staging paths Feb 28 13:10:22 03rpurdie 07org.oe.dev * rdf91e6d4... 10/ (3 files in 2 dirs): radlib: Fix STAGING_DIR reference Feb 28 13:10:28 03rpurdie 07org.oe.dev * ra6d0bc76... 10/ (3 files in 2 dirs): xsp, xpext: Convert to autotools_stage_all Feb 28 13:10:32 03rpurdie 07org.oe.dev * rf6f72dca... 10/ (1 packages/roadster/roadster_cvs.bb): roadster: Improve staging variables Feb 28 13:10:33 hrw: that'll usually need configuring, yes. Feb 28 13:10:37 03rpurdie 07org.oe.dev * r753c8290... 10/ (1 packages/pwc/pwc_10.0.12-rc1.bb): pwc: Improve staging variables Feb 28 13:10:41 03rpurdie 07org.oe.dev * rc8935fe4... 10/ (1 packages/php/php_5.2.0.bb): php: Improve staging variables Feb 28 13:10:47 03rpurdie 07org.oe.dev * r92f6802c... 10/ (1 packages/pax-utils/pax-utils-native_0.1.13.bb): pax-utils-native: Improve staging variables Feb 28 13:10:51 03rpurdie 07org.oe.dev * r444bba9d... 10/ (1 packages/netatalk/netatalk_2.0.3.bb): netatalk: Fix hardcoded staging paths Feb 28 13:10:56 03rpurdie 07org.oe.dev * ref74485c... 10/ (1 packages/libaio/libaio_0.3.103.bb): libaio: Fix hardcoded staging paths Feb 28 13:11:01 03rpurdie 07org.oe.dev * r8f84fa28... 10/ (1 packages/ltrace/ltrace_0.4.bb): ltrace: Improve staging variables Feb 28 13:11:06 03rpurdie 07org.oe.dev * re3295b5f... 10/ (4 files in 3 dirs): linphone: Improve staging variables Feb 28 13:11:11 03rpurdie 07org.oe.dev * r824b8111... 10/ (1 packages/libxslt/libxslt_1.1.17.bb): libxslt: Improve staging variables Feb 28 13:11:18 03rpurdie 07org.oe.dev * r1ffdb359... 10/ (3 files in 2 dirs): konqueror: Improve staging variables Feb 28 13:11:23 03rpurdie 07org.oe.dev * r6bf5262f... 10/ (3 files in 2 dirs): gxine: Improve staging variables Feb 28 13:11:28 03rpurdie 07org.oe.dev * ra27d281e... 10/ (1 packages/gpgme/gpgme.inc): gpgme: Improve staging variables Feb 28 13:11:32 03rpurdie 07org.oe.dev * r8a395148... 10/ (1 packages/dtnrg/dtn_2.5.0.bb): dtnrg: Improve staging variables Feb 28 13:11:36 03rpurdie 07org.oe.dev * r00dee336... 10/ (1 packages/callweaver/callweaver_1.2.0-rc5.bb): callweaver: Improve staging variables Feb 28 13:11:41 03rpurdie 07org.oe.dev * r2eef75a4... 10/ (1 packages/ark3116/ark3116_0.4.1.bb): ark1136: Improve staging variables Feb 28 13:11:46 03rpurdie 07org.oe.dev * rc6bd8d86... 10/ (1 packages/emacs/emacs_cvs.bb): emacs: Improve staging variables Feb 28 13:11:50 03rpurdie 07org.oe.dev * rb6622369... 10/ (1 packages/dejagnu/dejagnu-qemu_1.0.bb): dejagnu: Improve staging variables Feb 28 13:11:55 03rpurdie 07org.oe.dev * ra1412674... 10/ (1 packages/erlang/erlang_R11B.1.bb): erlang: Improve staging variables Feb 28 13:16:43 yay Feb 28 13:17:08 ho :P Feb 28 13:32:00 mickeyl: good morning Feb 28 13:32:56 zecke: good evening Feb 28 13:33:04 RP: you make good use of quad? :D Feb 28 13:33:09 pb_: master! Feb 28 13:33:55 hrw: it certainly helps ;-) Feb 28 13:42:23 RP: you joined the quad fun? Feb 28 13:43:30 XorA: yes, my desktop became a little too underpowered Feb 28 13:44:05 heh heh Feb 28 13:44:26 of course you now add all these new features to bb and you need an 8 core :-) Feb 28 13:45:04 XorA: None of the new features should require cpu ;-) Feb 28 13:45:26 XorA: If anything its all the extra recipes causing problems... Feb 28 13:45:58 I make bitbake parsing 25% faster, we add 25% more recipes... Feb 28 13:46:07 :-D Feb 28 13:46:15 time to make cache building multi threaded Feb 28 13:46:33 XorA: I was thinking about that ;-) Feb 28 13:47:55 A full reparse is ~80seconds on this machine Feb 28 13:48:29 I will wait for quad core price drop Feb 28 13:48:41 today quad core platform is 1/4 of my car Feb 28 13:49:07 hrw: now you have two people draining your wallet, it shall never happen :-)] Feb 28 13:49:28 ;DD Feb 28 13:49:46 XorA: and both are women kind Feb 28 13:49:56 ~hail qemuarm Feb 28 13:49:57 * ibot bows down to qemuarm and chants, "I'M NOT WORTHY!!" Feb 28 13:50:31 hrw: you will need bigger house for the shoes Feb 28 13:50:32 you can get a quad system with 1TB disk space and 4GB ram for about 1.4K EUR which is OK I think? Feb 28 13:50:58 Jin^eLD: let me check prices here Feb 28 13:51:40 hrw: I put together and ordered a system for the company I work at and it was about that price, imho quite acceptable Feb 28 13:51:54 it probably got cheaper in the meantime Feb 28 13:57:36 600EUR without gfx card (cpu+mainboard+2xhdd+4gb ram+case) Feb 28 13:57:53 wow, dropped to half of the price Feb 28 14:02:52 cheapest ram, cheapest cpu (q6600) Feb 28 14:06:33 03tmbinc 07org.oe.dreambox * r97437d26... 10/ (3 files in 3 dirs): dreambox-buildimage-native: support for different flash sizes Feb 28 14:06:38 03mickeyl 07org.oe.dev * r0e004b0b... 10/ (6 files in 6 dirs): ode 0.8 update to 0.9 and bring back from nonworking Feb 28 14:06:42 03mickeyl 07org.oe.dev * r425c38aa... 10/ (3 files in 3 dirs): python-pyode 1.2.0 bring back from nonworking Feb 28 14:06:47 03rpurdie 07org.oe.dev * r72256473... 10/ (1 classes/distutils.bbclass): distutils.bbclass: Tweak staging variables Feb 28 14:06:50 l8r Feb 28 14:06:53 03rpurdie 07org.oe.dev * r2a0669f0... 10/ (5 files in 2 dirs): mono: Use STAGING_DATADIR Feb 28 14:06:57 03rpurdie 07org.oe.dev * r20be77ba... 10/ (4 files in 2 dirs): libidl: Use autotools_stage_all Feb 28 14:07:02 03rpurdie 07org.oe.dev * r36937c20... 10/ (4 files in 2 dirs): fltk: Use autotools_stage_all Feb 28 14:07:06 03rpurdie 07org.oe.dev * rc6edddd9... 10/ (4 files in 2 dirs): libsvg-cairo/xsvg: Use autotools_stage_all Feb 28 14:07:12 03rpurdie 07org.oe.dev * r33c9650c... 10/ (7 files in 6 dirs): enca, motion, iperf, gdk-pixbuf, gtk+-directfb, python-sword: Use staging prefix/exec_prefix Feb 28 14:07:18 03rpurdie 07org.oe.dev * rdd10c4b3... 10/ (1 packages/fs/fs_cvs.bb): fs: Use autotools_stage_all Feb 28 14:07:22 03rpurdie 07org.oe.dev * rc7964e38... 10/ (1 packages/avr-libc/avr-libc_1.0.3.bb): avr-libc: Fix STAGING_DIR references Feb 28 14:07:28 03rpurdie 07org.oe.dev * r7389f2e9... 10/ (6 files in 4 dirs): rootfs_deb.bbclass: Fix apt.conf handling (from poky) Feb 28 14:07:33 03rpurdie 07org.oe.dev * r4e11ae88... 10/ (1 classes/package_deb.bbclass): package_deb.bbclass: Update apt.conf location (from poky) Feb 28 14:07:40 03mickeyl 07org.oe.dev * r6e7a8376... 10/ (3 files in 3 dirs): python-pycap 0.1.6 bring back from nonworking Feb 28 14:08:47 thx mickey|lunch Feb 28 14:32:09 XorA: I saw you played with upstart in with busybox Feb 28 14:32:25 how did that experiment end? Feb 28 14:32:32 keesj: niope Feb 28 14:34:01 this is what I saw http://lists.openmoko.org/pipermail/distro-devel/2007-October/000067.html Feb 28 14:35:07 morning Feb 28 14:36:20 keesj: thats as far as I ever got Feb 28 14:37:36 Thanks Feb 28 14:38:11 I am looking it now , so I just wondered Feb 28 14:56:24 RP: can LED triggers be changed at run-time using sysfs? Feb 28 14:56:42 cbrake: sure Feb 28 14:56:53 RP: very neat :-) Feb 28 14:57:56 n810 keyboard backlight is handled as LED Feb 28 14:58:08 looks nice with heartbeat trigger ;D Feb 28 15:02:19 03mickeyl 07org.oe.dev * r73105a15... 10/ (4 files in 2 dirs): dbus [all] fix packaging, make the libdbus RRECOMMEND the actual message bus system Feb 28 15:03:37 mickey|lunch: isnt r73105a15... just a hack to avoid setting DISTRO policy correctly? Feb 28 15:08:28 bye Feb 28 15:30:08 anyone know what happened to madwifi-ng and gtk+directfb, neither seem to be in the build environment and so the slugos build of packages doesn't complete Feb 28 16:02:48 03mickeyl 07org.oe.dev * r52629bff... 10/ (7 files in 5 dirs): python-m2crypto 0.18 update to 0.18.2 and bring back from nonworking Feb 28 16:48:55 anyone interested in the btopush library and bluespam? Some guy has made bb files for these in #gumstix Feb 28 16:49:12 * Crofton is not certain we should make it easy for people to do bluetooth spam Feb 28 16:51:24 XorA, ping Feb 28 16:52:14 Crofton: yo Feb 28 16:52:38 the gumstix guys are trying to rewrite the tsc2003 touchscreen driver Feb 28 16:52:57 do you know much about the chip, apparently it is driving them insane Feb 28 16:55:28 Crofton: I got the driver off a mailing list, the chip datasheet was fairly clear Feb 28 16:55:42 Crofton: datasheet is public Feb 28 16:56:28 apparently the behaviors are strange :) Feb 28 16:57:07 phrases like "interupts go wacky" are used in #gumstix :) Feb 28 16:57:18 the penirq signal is the culprit Feb 28 16:57:34 Wolfson board didnt have IRQ wired Feb 28 16:57:39 doing x/y readings cause it to toggle Feb 28 16:57:52 ah Feb 28 16:58:01 we're just about to give up on using the irq Feb 28 16:58:21 thanks XorA Feb 28 16:58:29 works fine in polling mode, and I cut out the irq handling in the driver I gave you Feb 28 16:58:41 that explains the difference in temperment between you and #gusmtix :) Feb 28 16:58:42 just reading the 2 paragraphs describing its bizarrre behaviour are enough to give you a headache Feb 28 16:59:12 turn off penirq in handler? Feb 28 16:59:17 poorly written docs about a poorly design chunk or hw :-) Feb 28 16:59:38 buy Wolfson :-D Feb 28 16:59:39 XorA: yes, that's the next experiment Feb 28 16:59:54 sakoman_: what do you use to base your driver on? Feb 28 16:59:56 They aren't paying me though :-) Feb 28 17:00:16 The code you gave Crofton Feb 28 17:00:41 sakoman_: ah, that'll be the problem, that was a totally hacked up of the original Feb 28 17:00:54 I went back and grabbed the original Feb 28 17:00:54 sakoman_: you would be better basing on the original driver probably Feb 28 17:01:02 same issues :-) Feb 28 17:01:32 The authors agree that the penirq stuff is ugly Feb 28 17:01:33 heh heh, so the code I ripped out didnt work anyway :-D Feb 28 17:02:59 03mickeyl 07org.oe.dev * r1d8d7681... 10/ (3 files in 3 dirs): python-egenix-md-base 2.0.6 update to 3.0.0 and bring back from nonworking Feb 28 17:03:05 03mickeyl 07org.oe.dev * ref6de30c... 10/ (78 files in 16 dirs): remove obsolete stuff from nonworking (e17, efl, xfree86) Feb 28 17:03:23 * CosmicPenguin giggles at that Feb 28 17:12:02 ~ban irq Feb 28 17:12:46 FYI Angstrom running on atngw100 (avr32)-> http://www.linuxdevelopment.org/projects/embedded/avr32/shots/dscn4201.jpg Feb 28 17:13:05 http://www.linuxdevelopment.org/projects/embedded/avr32/shots/dscn4202.jpg Feb 28 18:23:57 03rpurdie 07org.oe.dev * r03537602... 10/ (38 files in 13 dirs): obsolete: Drop old premodular xlibs and xproto Feb 28 18:32:13 http://www.zinkwazi.com/wp/2008/02/28/greg-gerg-gilbert/ :( Feb 28 18:35:49 Wow - that is most uncool Feb 28 18:36:29 ouch Feb 28 18:37:05 Yeah.. just saw that on the opie ml Feb 28 18:38:08 :( Feb 28 18:38:53 forwarded to oe ml Feb 28 18:57:05 * * OE Bug 2406 has been RESOLVED (FIXED) by rodrigo.vivi(AT)gmail.com Feb 28 18:57:06 * * Build breaks when the user has a ~/.quiltrc file Feb 28 18:57:08 * * http://bugs.openembedded.org/show_bug.cgi?id=2406 Feb 28 19:22:55 thesing: NOTE: Task failed: Error: /oe/org.openembedded.dev/packages/klibc/files/./fstype-sane-vfat-and-jffs2-for-1.5.patch not found. Too many '-'? Feb 28 19:25:30 For a do_func[depends] += line, is virtual/kernel:do_deploy correct syntax? Feb 28 19:25:38 or is a virtual done slightly differently? Feb 28 20:13:04 mhhh poky have their own full set of recipes or share a part with OE ? Feb 28 20:16:45 gremlin[it]: Most things are kept in sync Feb 28 20:17:20 * RP just read about Greg :( Feb 28 20:18:41 but one can build with poky metadata without any OE recipe, right ? Feb 28 20:20:22 gremlin[it], right Feb 28 20:20:22 gremlin[it]: yes Feb 28 20:20:32 poky is allegedly meta that all works :) Feb 28 20:20:43 er recipes that all work :) Feb 28 20:21:00 Crofton: The autobuilder says there is nothing alleged about it ;-) Feb 28 20:21:21 do you test :) Feb 28 20:21:50 sad news about Greg Feb 28 20:22:05 hi all,i instal on HTCMagician Angstrom with OpenMoko - how i can chenge locale on ru ? Feb 28 20:27:44 with the new recipes i saw on stable branch 'meta-toolchain*' and 'task-sdk*' will build relocable toolchains ? Feb 28 20:28:23 can suck task / meta build .rpm packages ? Feb 28 20:28:29 can such .... Feb 28 20:33:25 Crofton: We do test ;-) Feb 28 20:34:05 gremlin[it]: We have a package_rpm.bbclass but I don't know what state its in Feb 28 20:34:24 gremlin[it]: meta-toolchain builds a toolchain to run in a predefined location, it is not relocatable Feb 28 20:34:40 you can specify the location though Feb 28 20:37:40 mhh ok ... cause on FOSDEM we talk about make our toolachians (gcc mainly) relocatable as for example neuros one is and another i don't remember (both are freely available with sources) Feb 28 20:52:09 gremlin[it]: It would be interesting to look at the patches for that Feb 28 20:53:01 RP, no patch just an already patched archive .... :s Feb 28 20:53:17 at least for the neuros Feb 28 20:53:48 mhh the other can be foresight linux or something like ... my memory is really bad for names Feb 28 20:59:35 er, RP, wasn't the sysroot stuff you did to gcc stuff to make it relocatble? Feb 28 20:59:47 Or are there still a few missing bits that people like CodeSourcery have? Feb 28 21:00:05 Tartarus: No, sysroot is something different Feb 28 21:00:20 Tartarus: I don't think CodeSourcery is reclocatable either Feb 28 21:00:25 RP, well, part of it, at least from how I've played w/ the CS stuff Feb 28 21:00:31 RP, current CS stuff is Feb 28 21:00:42 so long as part of the base dir structure is in tact Feb 28 21:01:11 Tartarus: Is think something they're recently added? Feb 28 21:01:16 s/think/this/ Feb 28 21:01:28 Sourcery G++ 4.2 has it Feb 28 21:01:37 Haven't used anything older Feb 28 21:01:53 The CSL stuff usually filters into gcc Feb 28 21:02:03 Yeah Feb 28 21:02:12 I just hope it's not some sort of value-add thing they won't push back Feb 28 21:02:15 'cuz it's quite nice Feb 28 21:02:16 Although they tend to 'forget' important patches :/ Feb 28 21:02:45 03pfalcon 07org.oe.dev * r8ea8e071... 10/ (1 packages/linux/linux-handhelds-2.6.inc): Feb 28 21:02:45 linux-handhelds-2.6: Add support for building kernel with internal initramfs Feb 28 21:02:45 * KERNEL_INITRAMFS_PATH should be set to teh full path of cpio.gz image of Feb 28 21:02:45 initramfs. It is assumed to be built with OE as previous step, so usually Feb 28 21:02:45 points to deploy dir. Feb 28 21:02:49 03pfalcon 07org.oe.dev * r75180fb8... 10/ (1 conf/distro/angstrom-2008.1.conf): Feb 28 21:02:50 angstrom-2008.1.conf: Set initramfs-bootmenu-image as kernel builtin initramfs Feb 28 21:02:52 * Applies to linux-handhelds-2.6 so far. Feb 28 21:04:04 Although is it really gcc that's not relocatible? Feb 28 21:04:15 I don't think we had to do anything back at montavista for that to work Feb 28 21:04:23 * Tartarus looks for khem Feb 28 21:05:28 Tartarus: I know this didn't work back in the 3.4/3.4 days, I'm not sure I've ever tried since ;-) Feb 28 21:05:52 really? Feb 28 21:06:01 I'd have sworn MV did it back with that version Feb 28 21:06:03 and older Feb 28 21:06:22 sysroot Feb 28 21:06:48 yes it works on MV compiler 3.4 Feb 28 21:06:57 the toolchain of neuros is gcc 3.4.6 ;) Feb 28 21:06:59 because sysroot was patched in? Feb 28 21:08:28 heh Feb 28 21:08:29 you can use --sysroot for gcc to look for header and libs Feb 28 21:08:42 sysroot and my understanding of a relocatable compiler are two different things Feb 28 21:08:51 RP: yes Feb 28 21:09:40 sysroot mostly works in gcc 3.4 onwards. I'm not sure about reclocation though Feb 28 21:11:08 RP: if you put gcc under a/b or a/b/c it works does that suffice for relocation ? Feb 28 21:11:54 RP: sysroot helps gcc to become relocatable Feb 28 21:12:12 KhemHome: I mean if you compile it to live in /usr/local/a and then move it to /opt/b, does it still work? Feb 28 21:12:20 yes it does Feb 28 21:12:34 I know it never used to ;-) Feb 28 21:16:45 i remember too ... fixed position and no symlink !!! Feb 28 21:17:08 KhemHome: I just tried it "as: unrecognized option `-mfloat-abi=soft'" Feb 28 21:17:08 RP: So now I have a compiler compiled to be using /opt/montavista and have it installed under /opt/khem/montavista Feb 28 21:17:33 RP: hmmm Feb 28 21:17:58 RP: actually do you have as in your gcc relocated tree for you target ? Feb 28 21:18:21 KhemHome: Its a compile which worked and I just moved the directory Feb 28 21:18:39 RP: OK so I believe it has Feb 28 21:19:21 KhemHome: I've been burnt by this stuff before :/ Feb 28 21:19:49 RP: I remember this problem and there was a way I fixed it Feb 28 21:19:54 let me look into my notes Feb 28 21:21:38 RP: where is as located in your relocated tree ? Feb 28 21:21:44 KhemHome: I understood a lot of things here : http://dev.gentoo.org/~vapier/eh/index.xml?part=1&chap=4 Feb 28 21:21:48 is it a binary or symlink Feb 28 21:22:02 yeah Feb 28 21:22:06 there's a dir you need to move too Feb 28 21:22:26 ie arm-linux-gnueabi/bin Feb 28 21:22:31 KhemHome: "We will be cross-compiling using the sysroot method. But wtf is sysroot ?" Feb 28 21:22:36 needs to be in the same dir as bin/arm-linux-gnueabi-gcc Feb 28 21:22:37 actually creating as as relative symlink should help Feb 28 21:22:46 KhemHome: A ton of broken symlinks ;-) Feb 28 21:22:53 RP: there you go Feb 28 21:22:55 ant__, magic to say that the / you want is really /my/other/target/ Feb 28 21:23:02 RP: it should be relative Feb 28 21:23:02 KhemHome: Fixing the as one let it work... Feb 28 21:23:10 RP: yep Feb 28 21:23:25 now your compiler should work from anywhere Feb 28 21:23:41 RP, so you'll have, say: Feb 28 21:23:47 $ ls bin/arm-linux-gcc Feb 28 21:23:47 bin/arm-linux-gcc Feb 28 21:23:53 $ ls arm-linux/bin Feb 28 21:23:54 ar as ld libusb-config nm objcopy objdump ranlib strip tput tset Feb 28 21:23:56 ? Feb 28 21:24:28 RP: we should put this into OE toolchain Feb 28 21:25:39 Tartarus: ls arm-poky-linux-gnueabi/bin/ -la Feb 28 21:25:39 lrwxrwxrwx 1 richard richard 60 2008-02-28 21:14 ar -> /usr/local/poky/eabi-glibc/arm/bin/arm-poky-linux-gnueabi-ar Feb 28 21:25:39 lrwxrwxrwx 1 richard richard 61 2008-02-28 21:22 as -> /usr/local/poky/eabi-glibc/arm3/bin/arm-poky-linux-gnueabi-as Feb 28 21:25:49 ant__: sysroot is logical rootfilesystem where you point your gcc and binutils to look for headers and libraries Feb 28 21:25:53 You can see the one I 'fixed' Feb 28 21:26:17 KhemHome: Yes, its something we should fix. It never used to work at all, even allowing for the symlinks Feb 28 21:26:20 Tartarus: these are absolute links Feb 28 21:26:20 RP, right Feb 28 21:26:42 RP: right, are you going to take a stab at it Feb 28 21:26:54 RP: have you seen Paul committed the initramfs stuff for linux-handelds ? Feb 28 21:26:55 RP, I'm saying the whole of /usr/local/poky/eabi-glibc could become /opt/eabi-glibc Feb 28 21:27:30 KhemHome: I probably will at some point, yes. I've already altered this code a lot recently to improve the situation... Feb 28 21:27:33 and it _should_ look for /opt/eabi-glibc/bin/ar Feb 28 21:27:35 er Feb 28 21:27:47 /opt/eabi-glbic/arm/bin/ar, as, etc Feb 28 21:28:00 Tartarus: I agree it shows signs of working. I'd also swear it never used to ;-) Feb 28 21:28:19 woo :) Feb 28 21:28:53 RP: It works pretty reliable this way I have done it long ago here and seen no problems Feb 28 21:31:04 Tartarus: how is samba stuff Feb 28 21:31:22 KhemHome, ant was gonna verify it Feb 28 21:31:27 ok Feb 28 21:31:28 the big problem, at least Feb 28 21:31:40 KhemHome: At some point I'd like to rewrite most of the gcc recipes, they're horrible ;-) Feb 28 21:31:59 Tartarus: I am having trouble booting my board and I am working from home so could not get it working with latest .dev Feb 28 21:32:20 RP: I have not given a serious glare at those too :) Feb 28 21:32:39 tar jcf Feb 28 21:32:40 ga Feb 28 21:32:58 * KhemHome wishes if he had more time to work on OE Feb 28 21:35:49 ant__: sysroot is like pretending to gcc as if the / was located at /a/b/c/d/ that's where it will look for /usr/lib and /usr/include etc. the stuff it will need for linking and compiling and linker is smart to remove the sysroot part from final binary so in the end it will have paths relative to / in the binary Feb 28 21:36:04 * * OE Bug 3892 has been created by lee295012(AT)googlemail.com Feb 28 21:36:06 * * Angstrom and rx1950 no console/x11 Feb 28 21:36:08 * * http://bugs.openembedded.net/show_bug.cgi?id=3892 Feb 28 21:37:36 KhemHome: "... But wtf is sysroot ?" was an excerpt from that page, not my doubt ;-) Feb 28 21:38:01 Tartarus: I'm recompiling all the stuff again, hope I'll try samba in a couple of hours Feb 28 21:38:05 ant__: LOL Feb 28 21:38:54 KhemHome: but don't ask me too difficult questions, we have RP for this =) Feb 28 21:39:35 bbl Feb 28 22:18:09 03rpurdie * r1025 10/ (2 files in 2 dirs): build.py: Fix task variable expansion issue Feb 28 22:35:45 hey...i've got a short question again indeed ;) Feb 28 22:37:15 can i use ipkg on a running, bare built oe-image to upgrade its functionality? i'm coming from debian, so i think of something like dpkg -i Feb 28 22:38:10 i'm think of building an image just with everyting neccessary for a working ipkg system and then upgrading it by internet Feb 28 23:29:05 re Feb 28 23:33:04 * * OE Bug 3893 has been created by mark.just.mark(AT)gmail.com Feb 28 23:33:06 * * H4350 hangs after ipkg update, upgrade Feb 28 23:33:08 * * http://bugs.openembedded.net/show_bug.cgi?id=3893 Feb 28 23:44:45 ~seen ggilbert Feb 28 23:44:48 ggilbert was last seen on IRC in channel #oe, 8d 12h 1m 3s ago, saying: 'morning'. Feb 28 23:52:52 wow Feb 29 00:23:04 * * OE Bug 3886 has been RESOLVED (FIXED) by Feb 29 00:23:06 * * udev-118 initscript is very inefficent Feb 29 00:23:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3886 Feb 29 00:39:48 03mickeyl 07org.oe.dev * rac57dc4d... 10/ (1 packages/udev/udev-118/init packages/udev/udev_118.bb): udev 118 simplify init script, closes #3886 Feb 29 00:39:53 03mickeyl 07org.oe.dev * r46a9855b... 10/ (1 packages/linux/linux-handhelds-2.6.inc): linux-handhelds-2.6.inc: COMPATIBLE_MACHINE += eseries, closes #3175 Feb 29 00:43:06 thesing: Patch fstype-sane... is insane! klibc does not build in uclibc "mode" **** BEGIN LOGGING AT Fri Feb 29 03:59:33 2008 Feb 29 06:43:56 03likewise 07org.oe.dev * rfc26b505... 10/ (5 files in 4 dirs): linux-rt-2.6.24: Update to 2.6.24.3-rt3. Add missing include patch. Add MPC8313E-RDB LED driver. Fix LED numbering order. Feb 29 06:44:01 03likewise 07org.oe.dev * r59cc00aa... 10/ (2 files in 2 dirs): linux-2.6.24: Fix MPC8313E-RDB LED numbering order. Feb 29 06:44:06 03dyoung 07org.oe.dev * rd11d5824... 10/ (1 packages/gcc/gcc-cross_4.2.2.bb): gcc-cross_4.2.2 -> resynchronize PR to match recently modified gcc_4.2.2 Feb 29 07:01:01 do you remember which version of bitbake was used with oz354? Feb 29 07:08:18 zecke: i think i could have have a oz tree with me at my laptop .. with the used bitbake .. i can tell you in some hours , if its enough for you Feb 29 07:08:36 rob_w: I have found the information in the OZ wiki Feb 29 07:08:40 ok Feb 29 07:09:22 i am using oz354 snapshot still for a mipsel based project for now over 2 years .. works still nice for my purpose ;-) Feb 29 07:15:48 I am looking for a way to not build all the locales with glibc? Feb 29 07:18:26 zecke: digging in the past? Feb 29 07:19:04 hvontres|home: have to... Feb 29 07:19:25 hvontres|home: I need to build an old toolchain Feb 29 07:19:56 zecke: ahh... Feb 29 07:20:07 zecke: btw, did you hear about ggilbert? Feb 29 07:20:12 :( Feb 29 07:21:01 can anyone point me to how I can get rid of unused locales in glibc? Feb 29 07:23:10 berbsd_: each one should be an ipk Feb 29 07:23:43 berbsd_: at which point get rid off them? not creating them? not installing them? Feb 29 07:24:22 I would like not to create them to make sure they don't get installed. I am trying to get rid of any localization Feb 29 07:24:55 I only need en_US Feb 29 07:26:22 ENABLE_BINARY_LOCALE_GENERATION="0" I have this in my local.conf Feb 29 07:26:36 it will not create any locales! (but glibc should be en_US by default...) Feb 29 07:28:12 I suppose I need to rebuild glibc after that :-) Thanks, I am going to give it a try. Feb 29 07:47:23 zecke: it does not seem to work as expected. The locale packages are still created. Feb 29 08:03:02 berbsd_: I see NOTE: generation of binary locales disabled. this may break i18n! Feb 29 08:03:26 berbsd_: packages are created but it is almost no overhead creating the package though Feb 29 08:04:00 berbsd_: with IMAGE_LINGUAS you might or might not control the locales getting installed into the image, grep the tree... Feb 29 08:09:59 zecke: does IMAGE_LINGUAS indicates what get installed, independently of waht locales get built? Feb 29 08:10:37 berbsd_: right, not every package you built will automatically end in an image ;) Feb 29 08:11:57 I'll won't set ENABLE... for now and experiment with IMAGE_LINGUAS thaaaaaaaaaaaaa Feb 29 08:12:30 (cat jumped on the keyboard). Thanks for the tip. Feb 29 08:12:47 good cat! Feb 29 08:21:54 night all Feb 29 09:01:50 zecke: What kind of toolchain do you need? Feb 29 09:02:20 The "old" ones in Poky should still work... Feb 29 09:10:39 hey RP Feb 29 09:10:57 I saw openedhand somewhere today - you still with them? Feb 29 09:11:09 mithro: yes Feb 29 09:11:18 RP: cool Feb 29 09:12:05 mithro: How's things? Feb 29 09:13:12 pretty good Feb 29 09:13:18 I work at the same company as rwhitby Feb 29 09:13:47 mithro: ah, cool :) Feb 29 09:15:30 RP: sadly evil closed source stuff :/ Feb 29 09:15:53 mithro: If it pays the bills though... Feb 29 09:16:18 yeah, and the people are cool Feb 29 09:16:34 bblr Feb 29 09:17:06 mithro: Makes all the difference :). cu! Feb 29 09:43:00 hmmm nice....a vendor of embedded linux is using microsoft exchange for mailserver Feb 29 09:46:08 steliosk: they need a calendar that works :-) Feb 29 10:02:05 * * OE Bug 3894 has been created by  Feb 29 10:02:07 * * gcc-cross-sdk-4.2.2-r5-do_compile Feb 29 10:02:09 * * http://bugs.openembedded.net/show_bug.cgi?id=3894 Feb 29 10:06:04 hi, there? Feb 29 10:20:43 morning Feb 29 10:20:45 hi hrw Feb 29 10:23:14 hi, i've got a strange problem with my ext2-images....even when i do a fresh build, /lost+found uses 0,5-1mb of space, but there are no files in it. Feb 29 10:23:46 tom__: From doing a du -k /lost+found? Feb 29 10:23:49 Or just a df? Feb 29 10:24:01 du -sh /* Feb 29 10:25:00 why -k ? Feb 29 10:25:41 That'll show you the number of kilobytes, rather than the human-readable format.. no real difference if it's showing 1 meg Feb 29 10:25:52 k, assumed so Feb 29 10:25:54 hmm Feb 29 10:26:07 dmesg says that root is mounted readonly Feb 29 10:26:16 but in fstab its "defaults" Feb 29 10:26:20 and i can write to / Feb 29 10:26:34 i just installed lame by ipkg, and it works Feb 29 10:26:39 tom__: ls -lRa /lost+found? Feb 29 10:26:59 its empty Feb 29 10:27:01 good morning Feb 29 10:27:08 that is what's so weird to me Feb 29 10:27:13 never seen that Feb 29 10:27:14 RP: for our benq device, I digged up the old OE data :) Feb 29 10:27:16 hi florian Feb 29 10:28:39 * florian shooths compiz Feb 29 10:28:45 lol Feb 29 10:28:54 compiz does global warming Feb 29 10:29:16 heh Feb 29 10:29:55 it seems to break the session storage - all my termainals show up on the same desk Feb 29 10:30:22 compiz is just broken Feb 29 10:31:34 NAbyss: even when i do a e2fsck on the .ext2 file its not gone Feb 29 10:31:57 i think i'll try ext3 for compare Feb 29 10:32:01 tom__: That's quite odd.. it's a recurring issue with mkfs? Feb 29 10:32:19 yes, everytime i do bitakte *-image Feb 29 10:32:23 bitbake Feb 29 10:44:41 NAbyss: ls -a /lost+found/.* seems to do it Feb 29 10:44:52 there's a root dir structure Feb 29 10:45:19 oops Feb 29 10:45:28 no...i'm stupid Feb 29 10:45:34 Yeah, that'd be /lost+found/.. Feb 29 10:45:39 exactly Feb 29 10:45:41 doh Feb 29 10:47:31 ahh Feb 29 10:47:39 could it be that it's just reserved space? Feb 29 10:47:45 like 5% normally? Feb 29 10:48:01 because it's 412k when i have a 8mb fs **** ENDING LOGGING AT Fri Feb 29 10:59:56 2008