**** BEGIN LOGGING AT Wed Dec 26 02:59:56 2007 Dec 26 03:36:04 * * OE Bug 3552 has been created by  Dec 26 03:36:06 * * gcc-4.1.1-r17-do_install Dec 26 03:36:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3552 Dec 26 04:59:04 03pfalcon 07org.oe.dev * rbc06ee21... 10/ (3 files in 3 dirs): Dec 26 04:59:04 initrdscripts: Do machine detection function refactor per RFC. Dec 26 04:59:04 * Need to fix this, need to use good common codebase, because debugging Dec 26 04:59:04 why some adhoc code doesn't work here or there is big waste of time. Dec 26 04:59:14 03pfalcon 07org.oe.dev * rcd5d3c7e... 10/ (3 files in 3 dirs): Dec 26 04:59:14 psplash svn: Use standard OE init.d library for machine detection. Dec 26 04:59:14 * Fixes #3471. Dec 26 04:59:19 03pfalcon 07org.oe.dev * rc96094d9... 10/ (3 files in 3 dirs): Dec 26 04:59:19 ipaq-sleep 0.9: More fixes. Dec 26 04:59:19 * Don't try to interpret APM status when it's not available. Dec 26 04:59:19 * Don't assume X mode is always active. Dec 26 04:59:24 03pfalcon 07org.oe.dev * rc2046d55... 10/ (3 files in 3 dirs): Dec 26 04:59:24 ipaq-sleep 0.9: Fix issue with wrong autosuspend timer calc. Dec 26 04:59:25 * This code is sick. Dec 26 04:59:28 03pfalcon 07org.oe.dev * rc05cfe12... 10/ (1 packages/gsm/files/default packages/gsm/gsmd.inc): gsmd svn: Update for machine_id() func refactor. Dec 26 05:11:39 summatusmentis: what's up? Dec 26 05:34:04 * * OE Bug 3553 has been created by  Dec 26 05:34:06 * * base-image-1.0-r0-do_rootfs Dec 26 05:34:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3553 Dec 26 05:45:05 * * OE Bug 3554 has been created by  Dec 26 05:45:07 * * nas-server-image-1.0-r0-do_rootfs Dec 26 05:45:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3554 Dec 26 06:58:21 03xjqian 07org.oe.dev * rc0e108d7... 10/ (3 files in 3 dirs): zaurusd: add tosa default sound state patch Dec 26 07:27:03 * * OE Bug 3295 has been REOPENED by xjqian(AT)gmail.com Dec 26 07:27:05 * *  usb ports management Dec 26 07:27:07 * * http://bugs.openembedded.org/show_bug.cgi?id=3295 Dec 26 07:40:03 * * OE Bug 2468 has been RESOLVED (FIXED) by Dec 26 07:40:06 * *  python-pyqt4 fails do_compile Dec 26 07:40:07 * * http://bugs.openembedded.org/show_bug.cgi?id=2468 Dec 26 09:25:05 * * OE Bug 3551 has been RESOLVED (FIXED) by rod(AT)whitby.id.au Dec 26 09:25:07 * *  git-native fails to build - error in git-gui Makefile install command Dec 26 09:25:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3551 Dec 26 09:25:16 * * OE Bug 3555 has been created by c.romain(AT)laposte.net Dec 26 09:25:18 * * matchbox-desktop do_package QA issue Dec 26 09:25:20 * * http://bugs.openembedded.org/show_bug.cgi?id=3555 Dec 26 09:25:28 * * OE Bug 3556 has been created by c.romain(AT)laposte.net Dec 26 09:25:30 * * matchbox-applet-inputmanager do_package QA issue Dec 26 09:25:32 * * http://bugs.openembedded.org/show_bug.cgi?id=3556 Dec 26 09:25:40 * * OE Bug 3557 has been created by c.romain(AT)laposte.net Dec 26 09:25:42 * * matchbox-panel do_package QA issue Dec 26 09:25:44 * * http://bugs.openembedded.org/show_bug.cgi?id=3557 Dec 26 09:25:53 * * OE Bug 3558 has been created by c.romain(AT)laposte.net Dec 26 09:25:54 * * xkbd do_package QA issue Dec 26 09:25:57 * * http://bugs.openembedded.org/show_bug.cgi?id=3558 Dec 26 09:26:05 * * OE Bug 3559 has been created by c.romain(AT)laposte.net Dec 26 09:26:06 * * matchbox-wm do_package QA issue Dec 26 09:26:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3559 Dec 26 09:27:53 03rwhitby 07org.oe.dev * ra42936f8... 10/ (1 packages/git/git.inc): git: Add patch to put space before -m argument. Fixes #3551. No PR bump, cause no change to compilation or packaged content - if it built for you previously, then you don't need the patch, but it will do no harm. Dec 26 09:41:05 * * OE Bug 3553 has been RESOLVED (INVALID) by Dec 26 09:41:07 * *  base-image-1.0-r0-do_rootfs Dec 26 09:41:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3553 Dec 26 09:41:17 * * OE Bug 3554 has been RESOLVED (INVALID) by Dec 26 09:41:19 * *  nas-server-image-1.0-r0-do_rootfs Dec 26 09:41:21 * * http://bugs.openembedded.org/show_bug.cgi?id=3554 Dec 26 09:57:05 * * OE Bug 3560 has been created by milan.plzik(AT)gmail.com Dec 26 09:57:07 * * h5000 fingerprint scanner driver compiled Dec 26 09:57:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3560 Dec 26 11:25:57 03koen 07org.oe.dev * r8f4f4a96... 10/ (5 files in 4 dirs): linux-dht-walnut: first stab at converting it to a more generic recipe Dec 26 11:27:04 * * OE Bug 3295 has been RESOLVED (WORKSFORME) by dbaryshkov(AT)gmail.com Dec 26 11:27:06 * *  usb ports management Dec 26 11:27:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3295 Dec 26 11:28:23 03rwhitby 07org.oe.dev * r44926b8d... 10/ (1 packages/atftp/atftp_0.7.bb): atftp: Updated to use latest patches from Debian, fixes CLK_TCK compilation failure. Dec 26 11:49:44 morning all Dec 26 11:49:56 morning Dec 26 11:57:05 * * OE Bug 3561 has been created by c.romain(AT)laposte.net Dec 26 11:57:07 * * bluez-cups-backend-3.23 do_patch failure Dec 26 11:57:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3561 Dec 26 11:58:05 * * OE Bug 3562 has been created by  Dec 26 11:58:07 * * linux-dht-walnut-2.6.20-r4-do_deploy Dec 26 11:58:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3562 Dec 26 11:59:42 03freyther 07org.oe.dev * r4277f2f4... 10/ (7 files in 5 dirs): QtopiaCore: Do the minor/bugfix only upgrade from 4.3.2 to 4.3.3 Dec 26 11:59:46 03freyther 07org.oe.dev * r0c915035... 10/ (1 packages/qmake/qmake2-native_2.10a.bb): packages/qmake/qmake2-native_2.10a: No tabs, 4 spaces for indention Dec 26 11:59:52 03freyther 07org.oe.dev * re574d2ad... 10/ (1 packages/uicmoc/uicmoc4-native.inc): packages/uicmoc/uicmoc4-native.inc: No tabs, 4-spaces for indention Dec 26 12:00:01 03freyther 07org.oe.dev * r0e05da82... 10/ (6 files in 3 dirs): QtX11: Upgrade from 4.3.2 to 4.3.3 as well. Dec 26 12:00:07 03freyther 07org.oe.dev * r85e0c839... 10/ (19 files in 9 dirs): Dec 26 12:00:07 Remove some GNUism and replace cp -a with cp -pPR (once again) Dec 26 12:00:07 I didn't touch the to be installed files, as we have GNU coreutils or Dec 26 12:00:07 at least a compatible busybox version on our targets so far. Dec 26 12:00:23 03freyther 07org.oe.dev * rea1df980... 10/ (1 conf/distro/include/angstrom-2007-preferred-versions.inc): conf/distro/include/angstrom-2007-preferred-versions.inc: Propose to update Qt4/X11 to 4.3.3 Dec 26 12:00:46 03freyther 07org.oe.dev * r541cfc2f... 10/ (5 files in 2 dirs): Dec 26 12:00:46 packages/qmake/qmake2: Make it more easy to do updates Dec 26 12:00:46 Create a qmake2-native.inc, get the files from qmake2-native. Recreate Dec 26 12:00:46 qmake2-native_2.10a.bb and require the above file and say which version Dec 26 12:00:47 of QtopiaCore to get. Dec 26 12:00:54 03freyther 07org.oe.dev * rf9cf9e90... 10/ (19 files in 5 dirs): Dec 26 12:00:54 Move Qt/X11 and Qt/Embedded (known as QtopiaCore) to a qt4 directory. We will see a great unification in this directory. Dec 26 12:00:56 Besides the name Qt/X11 and Qt/Emebedded (QtopiaCore) are the same package Dec 26 12:00:58 with a different configure script. The aim is to share configuring, compiling, Dec 26 12:01:00 staging and packaging between these modules and making it more easy to try a bleeding Dec 26 12:01:02 qt snapshot. Dec 26 12:01:07 03freyther 07org.oe.dev * r3b045b5a... 10/ (1 packages/qt4/qt4_arch.inc): packages/qt4/qt4_arch.inc: No tabs, 4 spaces are wonderful Dec 26 12:01:17 03freyther 07org.oe.dev * r259d2314... 10/ (4 files in 2 dirs): packages/qt4: Move QtopiaCore over to the qt4_arch.inc Dec 26 12:01:28 03freyther 07org.oe.dev * r9261783a... 10/ (4 files in 2 dirs): Dec 26 12:01:28 package/qt4/qt4-x11-free,qtopia-core: Start sharing packaging Dec 26 12:01:28 The font and translations are a bit hairy and need further work. The idea Dec 26 12:01:28 is that at one point in time we can install Qt/X11 and Qt/Embedded side Dec 26 12:01:29 by side again. Headers and tools still need some time. Dec 26 12:01:40 03freyther 07org.oe.dev * rca108ea1... 10/ (5 files in 2 dirs): packages/qt4/qt4-x11-free, qtopia-core: Move boilerplate lines to a .inc file Dec 26 12:01:51 03freyther 07org.oe.dev * rdb92c83d... 10/ (4 files in 2 dirs): packages/qt4/qtopia-core,qt4-x11-free: Make it share the DEPENDS and inherits Dec 26 12:02:00 03freyther 07org.oe.dev * r63730890... 10/ (9 files in 2 dirs): Dec 26 12:02:00 packages/qt4/qt4-x11-free,qtopia-core: Unify configure and compile of Qt4 Dec 26 12:02:00 Adjust packaging to look into the directories we configured. Dec 26 12:02:06 03freyther 07org.oe.dev * ra3aa0bc4... 10/ (1 packages/qt4/qtopia-core.inc): Dec 26 12:02:06 packages/qt4/qtopia-core.inc: Use the same patches as Qt/X11 Dec 26 12:02:06 Do not build the tools, fix the mksepcs Dec 26 12:02:17 03freyther 07org.oe.dev * rf6576dba... 10/ (1 packages/qt4/qt_configuration.inc): packages/qt4/qt_configuration.inc: Put the header files somewhere else Dec 26 12:02:23 03freyther 07org.oe.dev * rc48f617b... 10/ (1 packages/qt4/qt4-x11-free.inc packages/qt4/qtopia-core.inc): packages/qt4/qt4-x11-free,qtopia-core: Change the order, move the config flags to the .inc files Dec 26 12:02:32 03freyther 07org.oe.dev * rf03475e5... 10/ (7 files in 2 dirs): Dec 26 12:02:32 packages/qt4: Move the staging bits from qtopia-core to qt_staging.inc, use it in Qt4/X11 Dec 26 12:02:32 The qt4-x11-free_4.3.3 and qtopia-core_4.3.3 bbfiles are rather empty now. The next Dec 26 12:02:32 stop is to make it compile. Dec 26 12:02:39 03freyther 07org.oe.dev * r7419f3f4... 10/ (1 classes/qmake2.bbclass classes/qtopia4core.bbclass): classes/qmake2,qtopia4core: Change the Qt include and library dir to match the new location/names Dec 26 12:02:45 03freyther 07org.oe.dev * rc9d0f49f... 10/ (1 packages/qt4/qtopia-core_4.2.0.bb): packages/qt4/qtopia-core_4.2.0: Use the common ${PN}.inc as well Dec 26 12:02:52 03freyther 07org.oe.dev * r9a909cd8... 10/ (1 conf/distro/include/angstrom-2007-preferred-versions.inc): conf/distro/include/angstrom-2007-preferred-versions.inc: Propose to update uicmoc4-native to 4.3.3 as well Dec 26 12:02:57 03freyther 07org.oe.dev * re548ddd5... 10/ (1 packages/qt4/qt4-x11-free.inc packages/qt4/qtopia-core.inc): Dec 26 12:02:57 packages/qt4/qt4-x11-free.inc,qtopia-core.inc: Require the qt_configuration.inc as well Dec 26 12:02:57 (add fontconfig to the depends of the X11 build as well, I wish I would have an Dec 26 12:02:57 interactive mode for mtn...) Dec 26 12:03:02 03freyther 07org.oe.dev * r298bc289... 10/ (1 classes/qmake2.bbclass): Dec 26 12:03:03 classes/qmake2.bbclass: Remove the usage of the QTDIR from this line as well Dec 26 12:03:05 (sadly no equivalent to git-commit --amend is available) Dec 26 12:03:09 03freyther 07org.oe.dev * r269e8f21... 10/ (3 files in 3 dirs): Dec 26 12:03:09 packages/qmake/qmake2-native.inc: Install the specs only once, add a patch to change the prefixes at runtime Dec 26 12:03:11 qmake has built-in paths. When building qmake2-native we have a pretty boring path Dec 26 12:03:15 which is certainly wrong for qt/x11 and qt/embedded. So we need a way to change this Dec 26 12:03:17 path at runtime. We just need to point qmake to a qt.conf and everything will be fine! Dec 26 12:03:19 03freyther 07org.oe.dev * rb03e1377... 10/ (1 packages/qt4/qt_configuration.inc): Dec 26 12:03:21 packages/qt4/qt_configuration.inc: Make use of the new QT_CONF_PATH, correct the configure line Dec 26 12:03:23 Generate a qt.conf in do_configure with the paths we pass to the ./configure Dec 26 12:03:27 script. Dec 26 12:03:31 03freyther 07org.oe.dev * r2d62f1f2... 10/ (1 packages/qt4/qt_packaging.inc): packages/qt4/qt_packaging.inc: Only install Qtopia translations if they exist. Dec 26 12:03:35 03freyther 07org.oe.dev * r44cf96b6... 10/ (1 packages/qt4/qt_configuration.inc): packages/qt4/qt_configuration.inc: Keep PARALLEL_MAKE untouched the buildsystem should handle deps Dec 26 12:03:43 03freyther 07org.oe.dev * r8b7d9bf8... 10/ (1 packages/qt4/qt_configuration.inc): Dec 26 12:03:43 packages/qt4/qt_configuration.inc: ccache and pch conflict with each other Dec 26 12:03:43 The time spent on recompiling outweights the time saved using pch. So disable Dec 26 12:03:45 pch and have better recompile times. Dec 26 12:03:49 03freyther 07org.oe.dev * refd90f8b... 10/ (4 files in 2 dirs): Dec 26 12:03:49 packages/qt4/qt4-x11-free,qtopia-core: Work on properly packaging Qt4 Dec 26 12:03:51 -QT_LIBRARY_NAME lost the lib prefix to make it usable with .pc files Dec 26 12:03:55 -Add headers to the -dev packages Dec 26 12:03:57 -Look in the right directory for the plugins remains Dec 26 12:03:59 -Get rid off the mkspecs before packaging Dec 26 12:04:01 -Attempt to package translations Dec 26 12:04:03 03freyther 07org.oe.dev * r48e0262a... 10/ (1 packages/qt4/qt_packaging.inc): Dec 26 12:04:05 packages/qt4/qt_packaging.inc: Next round of typo fixing and trying Dec 26 12:04:07 -Don't install every library to libQtUiTools-dev :} Dec 26 12:04:09 -Try to put the .la and .prl into the -dev packages as well Dec 26 12:04:13 03freyther 07org.oe.dev * r48dfd4dc... 10/ (1 packages/qt4/qt_packaging.inc): packages/qt4/qt_packaging.inc: Sadly the previous glob didn't work. Write .so, .la, .prl in the long form Dec 26 12:04:20 03freyther 07org.oe.dev * r4ac1d57c... 10/ (1 packages/qt4/qt_packaging.inc): packages/qt4/qt_packaging.inc: All Qt modules are roughly similar, have one python function to generate the files for the dev package Dec 26 12:04:23 03freyther 07org.oe.dev * r909c536d... 10/ (1 packages/qt4/qt_packaging.inc): Dec 26 12:04:25 packages/qt4/qt_packaging.inc: Do not remove everything from the bindir, fix the glob Dec 26 12:04:26 -${PN}-doc already has the html documentation, no need for -documentation package Dec 26 12:04:31 -Package the demos and exmaples Dec 26 12:04:33 -.debug and translations are missing Dec 26 12:04:37 03freyther 07org.oe.dev * r680bcf9a... 10/ (14 files in 6 dirs): Dec 26 12:04:39 packages/qt4/: Share some more patches (no host includes, cross-compile) between Embedded and X11 Dec 26 12:04:41 -Use the CROSSARCH hack/patch on QtopiaCore as well Dec 26 12:04:43 -Do not search /usr/include for freetype and openssl on Qt as well Dec 26 12:04:45 -Do not build a qmake version (needed a diversion due different configure scripts) Dec 26 12:04:49 03freyther 07org.oe.dev * r4ff36886... 10/ (1 packages/qt4/files/0005-fix-mkspecs.patch): packages/qt4/files/0005-fix-mkspecs.patch: Remove the O2 we want OE to decide the tuning! Dec 26 12:04:54 03freyther 07org.oe.dev * r085f876e... 10/ (4 files in 4 dirs): Dec 26 12:04:56 qmake,Qt/X11,QtopiaCore: Allow OE to set the tuning options. Allow to add a default module Dec 26 12:04:58 -Do not hardcode O2, we have flags in OE to do such things and passing both Dec 26 12:05:00 didn't make sense Dec 26 12:05:02 -With QtopiaCore QtGui is using Tcp/Udp/Sockets for the QWS implementation. This means when Dec 26 12:05:04 building a gui application you will need the network module. Allow to add modules using Dec 26 12:05:06 environment variables. Dec 26 12:05:10 03freyther 07org.oe.dev * r53b9cb02... 10/ (1 packages/qt4/qt_packaging.inc): Dec 26 12:05:12 packages/qt4/qt_packing.inc: Examples/Demos are in ${QT_DIR_NAME}, disable them due QA issues Dec 26 12:05:13 Disable packaging of the examples/demos as we need to get rid off Dec 26 12:05:16 the .debug directories inside the nested hierachy of the examples Dec 26 12:05:18 and demos. Dec 26 12:05:26 03freyther 07unknown * r5d2b4cb4... 10/ (1 packages/qt4/qt4-x11-free.inc packages/qt4/qtopia-core.inc): packages/qt4/qt4-x11-free.inc,qtopia-core.inc: Bump the PR as the packaging did change Dec 26 12:05:29 03freyther 07org.oe.dev * r49e39c42... 10/ (1 packages/qt4/qt_staging.inc): packages/qt4/qt_staging.inc: Fix place of the include and library directory Dec 26 12:07:48 03rwhitby 07org.oe.dev * rfd336c87... 10/ (1 packages/tasks/task-nas-server.bb): nas-server: added support for flashing openmoko phones from a usb-enabled nas-server Dec 26 12:14:04 * * OE Bug 3563 has been created by  Dec 26 12:14:06 * * linux-dht-walnut-2.6.20-r4-do_populate_staging Dec 26 12:14:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3563 Dec 26 12:15:05 * * OE Bug 3564 has been created by c.romain(AT)laposte.net Dec 26 12:15:07 * * matchbox-applet-startup-monitor do_package QA issue Dec 26 12:15:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3564 Dec 26 12:28:04 * * OE Bug 3565 has been created by  Dec 26 12:28:06 * * linux-dht-walnut-2.6.20-r4-do_install Dec 26 12:28:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3565 Dec 26 12:28:16 * * OE Bug 3566 has been created by deller(AT)gmx.de Dec 26 12:28:18 * * Poodle backlight control driver unreliable (too dark, jumping light intensity) Dec 26 12:28:20 * * http://bugs.openembedded.org/show_bug.cgi?id=3566 Dec 26 12:57:04 * * OE Bug 3567 has been created by deller(AT)gmx.de Dec 26 12:57:06 * * add sample wpa_supplicant config to /etc/network/ interfaces file Dec 26 12:57:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3567 Dec 26 12:57:16 * * OE Bug 3563 has been RESOLVED (FIXED) by Dec 26 12:57:18 * * linux-dht-walnut-2.6.20-r4-do_populate_staging Dec 26 12:57:21 * * http://bugs.openembedded.org/show_bug.cgi?id=3563 Dec 26 12:57:28 * * OE Bug 3565 has been RESOLVED (FIXED) by Dec 26 12:57:30 03koen 07org.oe.dev * rdad602c7... 10/ (3 files in 3 dirs): linux-dht-walnut: switch to uImage Dec 26 12:57:30 * *  linux-dht-walnut-2.6.20-r4-do_install Dec 26 12:57:33 * * http://bugs.openembedded.org/show_bug.cgi?id=3565 Dec 26 12:57:41 * * OE Bug 3562 has been RESOLVED (FIXED) by Dec 26 12:57:43 * *  linux-dht-walnut-2.6.20-r4-do_deploy Dec 26 12:57:45 * * http://bugs.openembedded.org/show_bug.cgi?id=3562 Dec 26 13:44:12 03koen 07org.oe.dev * r7abdd570... 10/ (6 files in 4 dirs): Dec 26 13:44:12 linux-dht-walnut: add 2.6.23 Dec 26 13:44:12 * disabled because uImage is 3.3MB larger as .20 and uboot doesn't boot it due to that size Dec 26 13:44:17 03koen 07org.oe.dev * r1416d1e4... 10/ (1 conf/distro/angstrom-2007.1.conf): angstrom 2007: RC3 Dec 26 14:18:56 morning Dec 26 14:19:26 mornin' CoreDump Dec 26 14:29:00 morning CoreDump Dec 26 14:36:06 RP: you don't know any zaurus-like device / board with existing u-boot config by chance? ;) Dec 26 14:36:38 CoreDump: Not really. You're going to look into uboot? Dec 26 14:37:22 hrw updated the pdaxrom patch to a recent u-boot Dec 26 14:38:07 all that appears to be missing for working SL-C7x0 and SL-Cxx00 support are working configuration headers (the format has changed) Dec 26 14:38:41 * RP isn't keen on encouranging people to use uboot :( Dec 26 14:39:00 RP: we would provide NAND-restore images for users Dec 26 14:39:05 * * OE Bug 3568 has been created by deller(AT)gmx.de Dec 26 14:39:07 * * Poodle - WPA wifi hangs system when unplugging power cable Dec 26 14:39:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3568 Dec 26 14:39:48 I did the pdax u-boot install yesterday, including restore to SHARP. Absolutely painless. Dec 26 14:40:24 it would solve the kernel problem perfectly Dec 26 14:41:57 I believe most users "bricking" their Z were using damaged or wrong u-boot images Dec 26 14:42:27 (the u-boot.zip didn't come with checksums heh) Dec 26 14:43:08 Recipe question. I have been browsing around the wiki and various recipies but I can't really find anything good. The program I want to compile needs to execute some stuff natively. Can this be done using qemu, and where should I look for instructions or ideas on how to do this? Dec 26 14:44:02 libc6 locale generation uses qemu for this if I am not mistaken Dec 26 14:44:14 (I have it turned off usually) Dec 26 14:44:31 * schme looksies. ty. Dec 26 14:45:06 Compiling on the Z itself was just no fun. After 7 hours it spat a internal compiler error: seg fault at me (: Dec 26 14:45:20 =) Dec 26 14:48:44 CoreDump: I think you're too optmisitic about users... Dec 26 14:48:47 this "clisp" of your seems to be a real PITA Dec 26 14:49:44 they were using the u-boot binary on SL-C3000 and SL-C3100 but it only works on SL-C3200 Dec 26 14:49:45 * RP is trying to improve his dev environment to help productivity and make me more active again... Dec 26 14:50:32 I would certainly prefer u-boot with JFFS2 / ext2 kernel-loading and native root-on-anything to a kexec "hack" Dec 26 14:51:01 I want to make kexec kernel commandline work properly... Dec 26 14:51:55 That would open up kernel as a bootloader options. Dec 26 14:52:29 CoreDump: Indeed. It has some none autotools ./configure. I hand it ./configure --host=arm-linux-angstrom-gnueabi . It does some stuff. enters src/ and runs the autotools conf there and that is handeh --host=i686-linux. GREAT. I managed to bypass that with some ugly hackery and the build just fails anyway. The help from the nice lisp people mostly consists of "cross compiling is dumb." Dec 26 14:53:36 =D Dec 26 14:54:03 yay for x86 closemindedness Dec 26 14:54:24 CoreDump: btw, did you find a test case for the bitbake issue? Dec 26 14:54:34 I believe so, hold on Dec 26 14:55:04 * * OE Bug 3567 has been RESOLVED (FIXED) by Dec 26 14:55:06 * *  add sample wpa_supplicant config to /etc/network/interfaces file Dec 26 14:55:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3567 Dec 26 14:55:13 It's strange though. clisp brags about being the most portable clisp and lists a fuckofalot of platforms it runs on. Which is all good if the fucker compiled. (: Dec 26 14:56:04 My latest idea is to see what their silly ./configure actually does and slap it all in a do_configure () and then cd src and do the autotools bit. Dec 26 14:56:25 Pain in the arse, but a lot of fun. This is the charm of OE I suppose :D Dec 26 14:56:54 RP: meh, I hade a reproducable case where "bitbake image" would always b0rk on a certain package due to non-built deps. I had to reboot the box and the problem is gone again.... Dec 26 14:57:05 03coredump2 07org.oe.dev * re0812b86... 10/ (3 files in 3 dirs): netbase: Update /e/n/interfaces to include new WPA example configuration. Closing bug #3567. Dec 26 14:57:21 schme: indeed =) Dec 26 14:57:27 CoreDump: gah :( Dec 26 14:57:48 CoreDump: Did you note which image and deps they were along with the task graph output? Dec 26 14:58:03 um no Dec 26 14:58:29 but fear not. openmoko-devel-image hits the bug a bazillion times each build Dec 26 14:58:42 I shall have a testcase again soonish Dec 26 14:59:06 * RP wonders why he doesn't see the bug... Dec 26 14:59:19 ok, failed to build settings_daemon Dec 26 14:59:23 ok, I'm not building om images but poky should show the same problem... Dec 26 14:59:51 CoreDump: failed because which dep wasn't built? Dec 26 14:59:56 rebuild fixed it *sigh* Dec 26 14:59:59 RP: of course Dec 26 15:00:30 CoreDump: Are these froms scratch images or an already populated tree with bumped PRs etc? Dec 26 15:00:46 It could be some I/O starvation thingy.. notince that non-bitbake HDD-access is _really_ slow when BB runs w/ 16 threads Dec 26 15:01:02 RP: always started w/ a cleam tmp Dec 26 15:01:05 *clean Dec 26 15:01:20 CoreDump: How many cores has that machine? Dec 26 15:01:25 4 Dec 26 15:01:45 Sounds like the machine just has really poor IO performance Dec 26 15:02:09 hrmm, Intel Quadcore ain't the best in I/O Dec 26 15:03:02 Thinking about it, bitbake is incredibly racy from that PoV... Dec 26 15:03:09 =D Dec 26 15:03:40 what kind of logging should I run to see why package "X" failed? Dec 26 15:03:57 The log from the failed task should tell you? Dec 26 15:04:03 nahh not that =) Dec 26 15:04:11 it is always a missing dependency. Dec 26 15:04:23 I mean how to see why the deps were not build yet Dec 26 15:04:54 All I can suggest is using bitbake -DDD and seeing why it built X Dec 26 15:05:39 bitbake shouldn't be racy with IO. If it is, it suggests something kernel related broke :/ Dec 26 15:06:12 FWIW, poky's autobuilder is now a quadcore intel btw Dec 26 15:06:44 I've seen two odd build failures on that and they looked like parallel make problems Dec 26 15:08:22 usually the depends bug stops the build in some do_configure Dec 26 15:08:37 with configure right complaining about a missing build-time dep Dec 26 15:09:04 or a do_compile halting with a missing header Dec 26 15:10:29 What we need to know is why its doing that. The output from bitbake -DDD before the crash would help. Also, once it crashes it would help if you could leave that build as per when it crashed for debugging Dec 26 15:10:59 specifically so the stamps directory is intact Dec 26 15:11:09 sure Dec 26 15:11:19 a -DDD build is running ATM Dec 26 15:34:59 03pfalcon 07org.oe.dev * r847c5e13... 10/ (1 packages/linux/linux-handhelds-2.6-2.6.21/h5000/defconfig): linux-handhelds-2.6 2.6.21: h5000: Enable new keys driver. Dec 26 15:35:04 03pfalcon 07org.oe.dev * r5411de36... 10/ (1 packages/linux/linux-handhelds-2.6-2.6.21/h5000/defconfig): linux-handhelds-2.6 2.6.21: h5000: Enable ASoC sound driver. Dec 26 15:35:10 03pfalcon 07org.oe.dev * r9d717f53... 10/ (1 conf/machine/magician.conf): magician.conf: Reorder autoloads from machine-independent to machine-specific. Dec 26 15:35:15 03pfalcon 07org.oe.dev * re4084010... 10/ (1 conf/machine/h5000.conf): h5000.conf: Ship and autoload ASoC sound modules. Dec 26 15:35:19 03pfalcon 07org.oe.dev * rac1747cc... 10/ (1 conf/machine/fic-gta01.conf conf/machine/fic-gta02.conf): Dec 26 15:35:19 fic-gta01.conf, fic-gta02.conf: Remove non-machine-specific modules. Dec 26 15:35:19 * Machine configs should contain only machine-specific modules, Dec 26 15:35:19 task-base pulls other modules consistently across machines, based on Dec 26 15:35:19 *_FEATURES. Dec 26 15:35:25 03pfalcon 07org.oe.dev * rb935ae17... 10/ (25 files in 25 dirs): linux-handhelds-2.6 2.6.21: Fix issue with asoc-using devices having static sound subsys. Dec 26 15:35:31 03pfalcon 07org.oe.dev * r13c20f26... 10/ (1 packages/linux/linux-handhelds-2.6_2.6.21-hh20.bb): linux-handhelds-2.6 2.6.21-hh20: Bump for recent changes. Dec 26 15:35:36 03pfalcon 07org.oe.dev * r363edf7a... 10/ (1 packages/tasks/task-base.bb): task-base: Bump PR for h5000 & fic-gta* changes. Dec 26 15:42:22 RP: http://hentges.net/tmp/debug/ <- cache, stamps and output from -DDD of a failure Dec 26 15:48:46 CoreDump: thanks, I'll have a look :) Dec 26 15:48:58 RP: thanks ;) Dec 26 15:53:39 CoreDump: openmoko-messages2 seems to be missing a DEPENDS on libjana? Dec 26 15:58:04 * * OE Bug 3569 has been created by jbs(AT)bandos.homelinux.com Dec 26 15:58:06 * * Konqueror-embedded broken in autoconf Dec 26 15:58:07 "TARGET_DEVICE_IP=192.168.1.x bitbake linux-rp -c bootkern" is a very scary command :} Dec 26 15:58:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3569 Dec 26 16:02:53 RP: well, if I keep rebuilding the image eventually libjana will be build and om-messages will configure Dec 26 16:03:20 CoreDump: right, but libjana is not in om-messages DEPENDS Dec 26 16:03:26 CoreDump: That is a metadata bug Dec 26 16:04:08 but liblana would be in a DEP from some .bb which is usually compiled _before_ om-m2 Dec 26 16:04:18 CoreDump: ok, which .bb is that? Dec 26 16:04:39 I have not the slightest idea, you are the one with the python foo =D Dec 26 16:04:53 fact is, eventually, libjana will be built Dec 26 16:05:12 Yes, libjana is depended on by other packages Dec 26 16:05:23 but does om-messages depend on those other packages? Dec 26 16:05:26 also I tried adding a missing dependency to another .bb and it would _still_ fail with that DEP on the next rebuild Dec 26 16:05:43 no clue Dec 26 16:05:55 I'll add the DEP and rebuild Dec 26 16:06:21 openmoko-dialer2, om-today2, om-dates2 and om-worldclock2 have libjana as DEPENDS at a quick glance Dec 26 16:06:35 om-messages2 should not be DEPENDing on them, right? Dec 26 16:07:41 no Dec 26 16:07:54 it shouldn't Dec 26 16:09:10 so the question would be: with an empty TMPDIR, would om-messages2 build w/o the libjana dep Dec 26 16:09:27 I don't think it would Dec 26 16:09:34 neither do I Dec 26 16:09:40 It needs libjana and libjana is not listed in DEPENDS Dec 26 16:09:43 I've added the dep and am rebuilding Dec 26 16:11:15 the image compiled Dec 26 16:11:20 * CoreDump rm -rf ../../out/ ; bitbake openmoko-messages2 Dec 26 16:11:35 (w/o libjana dep) Dec 26 16:12:35 My money is on that failing and the DEPENDS being incorrect Dec 26 16:13:03 Over time we've found and fixed a ton of these dependency glitches in poky Dec 26 16:14:05 * * OE Bug 3570 has been created by jbs(AT)bandos.homelinux.com Dec 26 16:14:06 03pfalcon 07org.oe.dev * re544e652... 10/ (16 files in 5 dirs): Dec 26 16:14:06 linux-rp: Consistently enable ext3 module. Dec 26 16:14:06 * Fixes #2515. Dec 26 16:14:07 * * qpegps various problems Dec 26 16:14:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3570 Dec 26 16:15:17 RP: I'd be happy if that turns out to be the problem Dec 26 16:15:35 re Dec 26 16:15:49 wb jilles Dec 26 16:15:53 wb Jin^eLD Dec 26 16:16:00 CoreDump: This is one thing packaged staging could help check out - whether our dependencies are correct without using a ton of processing power Dec 26 16:16:29 hey guys Dec 26 16:16:50 the akita config for u-boot is terrible Dec 26 16:17:06 no structure at all Dec 26 16:18:01 sash pasted and copied from the same configs I'm looking at ehe Dec 26 16:44:27 morning Dec 26 16:44:44 CoreDump: config is not everything Dec 26 16:45:02 CoreDump: u-boot newer then 1.1.4 require more board code Dec 26 16:45:13 I noticed Dec 26 16:45:24 after 1.5 hours of "playing" with the config... Dec 26 16:45:50 undefined reference to `board_nand_init' Dec 26 16:47:54 * CoreDump gave up on that... Dec 26 16:50:25 I am thinking about checking 1.1.4 and then each newer until it breaks Dec 26 16:51:57 FWIW, the pdax patch does not apply to _any_ version available in OE. Dec 26 16:52:13 thats not a problem Dec 26 16:52:14 Also I was to lazy to investigate which snapshot he used ;) Dec 26 16:52:36 ;D Dec 26 16:53:12 he he done his work correctly the first time around and got his stuff included upstream, we wouldn't have this discussion now... Dec 26 16:53:18 *had he Dec 26 16:54:04 upstream? Dec 26 16:54:10 u-boot Dec 26 16:54:15 for god sake - it came from pdaX! Dec 26 16:54:17 ;D Dec 26 16:54:21 =) Dec 26 16:56:45 I guess the remaining pdaX users would be happy as well about a fully working uboot w/ JFFS2 + ext2 kernel loading Dec 26 16:57:10 so they can use the Angstrom u-boot, and Angstrom Kernel to run a Debian system ;) Dec 26 16:57:56 03hrw 07org.oe.dev * r4b397824... 10/ (1 packages/openmoko2/openmoko-messages2_svn.bb): openmoko-messages2: added libjana to dependencies Dec 26 17:05:07 hrw: do you know if the u-boot prompt can be routed to USB via ACM device like on neo? Or is that a hardware thing? Dec 26 17:05:50 software needed Dec 26 17:05:59 I do not know does u-boot has pxa g_ether Dec 26 17:14:31 CoreDump: LABEL_2006_04_18_1106 u-boot is what I am building for corgi Dec 26 17:14:40 config changed already to more sane one Dec 26 17:15:02 that is quite ancient =) Dec 26 17:15:50 thats what pdaX used Dec 26 17:16:00 /home/hrw/devel/build/angstrom/tmp/cross/lib/gcc/arm-angstrom-linux-gnueabi/4.1.2/libgcc.a(_dvmd_lnx.o): In function `__div0': Dec 26 17:16:03 /home/hrw/devel/build/angstrom/tmp/work/armv5te-angstrom-linux-gnueabi/gcc-cross-4.1.2-r10/gcc-4.1.2/gcc/config/arm/lib1funcs.asm:1001: undefined reference to `raise' Dec 26 17:16:06 is what I got Dec 26 17:16:22 ok, I am going to wake up my wife and spend some time Dec 26 17:16:57 have fun hrw Dec 26 17:17:28 hrw|gone: cya Dec 26 17:17:41 http://www.arcamax.com/garfield Dec 26 17:17:59 CoreDump: http://pastebin.ca/832806 is corgi.h config used Dec 26 17:18:58 thanks, but I doubt that I could solve the ASM problem Dec 26 17:22:38 its oabi<>eabi problem Dec 26 17:22:43 u-boot should be oabi ;) Dec 26 17:22:45 bye for now Dec 26 17:22:55 hrw|gone: cya Dec 26 17:26:15 om has some eabi patches Dec 26 17:45:24 the "raise" thing appears to be a gcc bug...of all things Dec 26 17:54:40 hrw|gone: NOTE: package u-boot-git-r2: task do_build: completed =D Dec 26 17:55:58 now I'm going to try hacking the ENV to boot the angstrom kernel Dec 26 18:30:11 hrw|gone: ok, "our" u-boot with original pdax config works with pdax kernel & emergency setup Dec 26 18:31:15 just replacing the SHARP stuff and jumping to our kernel does not work Dec 26 18:41:46 hrw|gone: u-boot eabi patch: http://rafb.net/p/EQ2wb723.html Dec 26 19:05:54 Is there a way to make OE do remote execution ala scratchbox? Dec 26 19:17:13 icecc or something like that Dec 26 19:17:49 * schme looksies Dec 26 19:18:29 lookat classes/icecc.class Dec 26 19:18:35 that might give you some hints Dec 26 19:19:31 Ya. Checking it now. It seems to be what I'm looking for. Dec 26 19:19:50 Thanks :) Dec 26 22:16:46 console image boots ok on omap5912osk, but not base image ... Dec 26 22:26:04 * * OE Bug 3571 has been created by philip(AT)balister.org Dec 26 22:26:06 * * Base image hangs on boot for machine omap5912osk Dec 26 22:26:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3571 Dec 27 00:26:05 * * OE Bug 3328 has been RESOLVED (FIXED) by joshua(AT)willowisp.net Dec 27 00:26:07 * *  sound broken for h2200 in RC1.3 Dec 27 00:26:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3328 **** ENDING LOGGING AT Thu Dec 27 02:59:57 2007