**** BEGIN LOGGING AT Fri Oct 23 02:59:57 2009 Oct 23 03:44:52 grg: hi Oct 23 03:45:13 hello Oct 23 03:45:16 I've cross compiled the module I was trying yesterday remember Oct 23 03:45:18 ;) Oct 23 03:45:23 ok Oct 23 03:45:30 happens that...still didn't work Oct 23 03:45:30 working for you? Oct 23 03:45:30 so Oct 23 03:45:34 another gunny night Oct 23 03:45:57 did you try the dm9000 in kernel driver? Oct 23 03:46:17 strange thing you know...my omap does not have dm9000 Oct 23 03:46:42 I needed to build an image using narcissus since I didn't have access to the internet to get gcc Oct 23 03:50:16 guess I'll give up and build a new image using OE Oct 23 03:50:34 I mean, trying to add the correct module Oct 23 03:52:33 grg, question will sound kinda silly, but, I'really new to OE....where should I modify the recipe of the imagem in order to add a module which I have the source code and makefile? Oct 23 03:53:09 you will need to change the kernel config Oct 23 03:53:19 check the OE documentation Oct 23 03:53:25 ok Oct 23 03:53:30 i havent really done this the "right way" myself Oct 23 03:54:01 rsrs Oct 23 03:54:12 you were cross compiling the modules as wellw Oct 23 03:54:13 ? Oct 23 03:59:29 i build a kernel+modules out of the OE tree Oct 23 04:00:11 mainly because i have to do development their Oct 23 04:00:16 there Oct 23 04:03:52 hmm Oct 23 04:04:01 got it Oct 23 04:04:19 I only built the default kernels till now Oct 23 04:04:41 and still hoping I'll never have to write a module or a driver rsrs Oct 23 04:12:33 all you need to do is find the driver in the menus and enable it as a module. then rebuild the kernel recipe Oct 23 04:16:15 i want to write a module to toggle a gpio pin on my gumstix overo earth, what is the basic procedure? Oct 23 04:17:08 MonkeySilent, read LDD3 Oct 23 04:17:12 i have the physical address to the gpio pin I want, i am guessing I have to configure the pin to mode 4 etc Oct 23 04:17:15 and the OE manual Oct 23 04:17:25 and the omap docs Oct 23 04:17:59 i have read them breifly but i cant find what i need Oct 23 04:18:17 the omap docs are massive, yes Oct 23 04:18:18 i mean the omap3503 manual is 3k pages Oct 23 04:18:32 sorry, i cant help with that. Oct 23 04:19:02 you could read the board init code for an omap based board, that would probably have some readable gpio toggling code Oct 23 05:08:26 hmm Oct 23 05:09:24 grg, I'm building a new image...hope it works if you cuild take a look at the recipe http://www.pastebin.ca/1639408 Oct 23 05:10:14 no idea Oct 23 05:10:34 i just copied/pasted and didn't really think about my image Oct 23 05:14:36 rsrs Oct 23 05:14:48 what's rsrs? Oct 23 05:15:10 if it works I'll let people know...maybe pot it ina blog rsrs = haha in brazil Oct 23 05:15:26 by the way, isn't it late where you are now? Oct 23 05:15:32 *post .. Oct 23 05:15:35 not really. im in australia Oct 23 05:15:45 its 3:45pm Oct 23 05:16:17 man...here is the same...but in the morning: 3:16 am Oct 23 05:16:51 gonna sleep and pray for OE not to crash :) thanks Oct 23 05:17:43 i hear sacrificing a virgin goat helps with that Oct 23 05:19:40 I'd prefer sacrificing a sadistic professor haha Oct 23 05:19:51 night everybody..or day for some Oct 23 05:28:14 use minimal on .dev works well for me Oct 23 05:39:25 03Leon Woestenberg  07shr/merge * r23402486d5 10openembedded.git/recipes/gnash/gnash-fb_0.8.5.bb: gnash-fb: Fix DEPENDS from cairo to agg. Oct 23 05:39:26 03Koen Kooi  07shr/merge * r1df7956a05 10openembedded.git/ (conf/checksums.ini recipes/cairo/cairo_1.8.8.bb): cairo: add 1.8.8 Oct 23 05:39:27 03Koen Kooi  07shr/merge * ra1a6def64f 10openembedded.git/recipes/poppler/poppler_0.12.1.bb: poppler: add 0.12.1, latest stable Oct 23 05:39:27 03Koen Kooi  07shr/merge * rc938428039 10openembedded.git/conf/distro/include/angstrom-2008-preferred-versions.inc: angstrom: prefer cairo 1.8.8, less bugs and faster Oct 23 05:39:30 03Holger Hans Peter Freyther  07shr/merge * rb294033116 10openembedded.git/recipes/binutils/binutils-cross-sdk.inc: Oct 23 05:39:33 binutils-cross-sdk: Avoid QA error on building Oct 23 05:39:35 On a x86-64 build host trying to build meta-toolchain Oct 23 05:39:37 for mips ended up with a .debug dir in the cross directory. Oct 23 05:39:39 Fix the issue by placing this directory in the -dbg package Oct 23 05:39:41 03Koen Kooi  07shr/merge * r5f88338d6a 10openembedded.git/recipes/gnome/gdm_2.28.0.bb: gdm: add dep on libcanberra and bump PR Oct 23 05:39:44 03Marcin Juszkiewicz  07shr/merge * r82ffbc9a75 10openembedded.git/recipes/xorg-xserver/ (2 files in 2 dirs): xserver-xorg-conf: added config fot at91sam9263ek Oct 23 05:39:53 03Koen Kooi  07shr/merge * r3d4264f134 10openembedded.git/recipes/tasks/angstrom-task-gnome.bb: angstrom task gnome: remove non-building uim bit Oct 23 05:39:56 03Koen Kooi  07shr/merge * r7d22ae8ecd 10openembedded.git/recipes/u-boot/u-boot-git/new-pinmux.patch: u-boot git: refresh beagle pinmux patch Oct 23 05:39:59 03Koen Kooi  07shr/merge * r341b8924f5 10openembedded.git/recipes/tasks/angstrom-task-gnome.bb: angstrom-task-gnome: various tasks or a GNOME desktop based image Oct 23 05:40:02 03Mark Brown  07shr/merge * rbca3b59f77 10openembedded.git/ (2 files in 2 dirs): Oct 23 05:40:04 smdk6410: Add initial recipe for SMDK6410 Oct 23 05:40:06 This is essentially just the defconfig for mainline; pretty much just Oct 23 05:40:08 build tested at present. Oct 23 05:40:10 03Michael Smith  07shr/merge * rda48a59464 10openembedded.git/ (3 files in 3 dirs): Oct 23 05:40:18 tzdata: add 2009o as 2009n is no longer available Oct 23 05:40:20 tzcode-native 2009k: use tzdata 2009o Oct 23 05:40:22 Signed-off-by: Michael Smith Oct 23 05:40:24 03Jeremy Laine  07shr/merge * r3e9bb4f23c 10openembedded.git/recipes/klibc/ (klibc_1.5.15.inc klibc_1.5.inc): Oct 23 05:40:27 klibc_*.inc: fix QA about packaging (klibc.so only for -dev) Oct 23 05:40:29 Signed-off-by: Andrea Adami Oct 23 05:40:33 03Khem Raj  07shr/merge * ra1dc338502 10openembedded.git/recipes/serial-utils/serial-forward.bb: Oct 23 05:40:36 serial-forward.bb: Fix missing GNU_HASH QA Error. Oct 23 05:40:38 Signed-off-by: Khem Raj Oct 23 05:40:40 03Khem Raj  07shr/merge * r077f07332c 10openembedded.git/ (161 files in 7 dirs): Oct 23 05:40:42 gcc-4.4.2: Add new recipes for gcc 4.2.2 Oct 23 05:40:44 * Switch default gcc to 4.4.2 for sane toolchain Oct 23 05:40:50 * Switch binutils to 2.20 Oct 23 05:40:52 (17 lines omitted) Oct 23 07:13:18 good morning Oct 23 07:26:02 03Koen Kooi  07org.openembedded.dev * rdf4e832331 10openembedded.git/ (3 files in 3 dirs): Oct 23 07:26:02 lyx: add 1.6.4.1 Oct 23 07:26:02 Deleting libtool.m4 allows us to use autoreconf :) Oct 23 07:40:56 03Marco Cavallini  07org.openembedded.dev * r8257ab0867 10openembedded.git/recipes/bluez/bluez4.inc: bluez4.inc : temporary solution for KaeilOS just to replace at least bluez-libs Oct 23 07:40:59 03Marco Cavallini  07org.openembedded.dev * re9b1030238 10openembedded.git/recipes/bluez/bluez4_4.35.bb: bluez4_4.35.bb : temporary solution for KaeilOS just to replace at least bluez-libs Oct 23 07:41:00 03Marco Cavallini  07org.openembedded.dev * re93d6c8559 10openembedded.git/recipes/bluez/bluez4_4.37.bb: bluez4_4.37.bb : temporary solution for KaeilOS just to replace at least bluez-libs Oct 23 07:49:33 03Koen Kooi  07org.openembedded.dev * rd15a13236e 10openembedded.git/recipes/redland/redland_1.0.8.bb: redland: add more DEPENDS Oct 23 07:49:34 03Steve Sakoman  07org.openembedded.dev * r6b73e92d64 10openembedded.git/recipes/ckermit/ckermit_211.bb: ckermit: fix GNU_HASH QA error Oct 23 07:49:35 03Koen Kooi  07org.openembedded.dev * r98e736cd3d 10openembedded.git/classes/image.bbclass: image bbclass: tweak lingue script Oct 23 07:49:36 03Steve Sakoman  07org.openembedded.dev * r4933bba055 10openembedded.git/recipes/file-roller/file-roller_2.28.0.bb: file-roller: fix packaging Oct 23 07:49:37 03Steve Sakoman  07org.openembedded.dev * r75f1abb702 10openembedded.git/recipes/evince/evince_2.28.0.bb: evince: stage evince headers to make nautilus happy Oct 23 07:49:40 03Koen Kooi  07org.openembedded.dev * r0e2ebb72bf 10openembedded.git/conf/distro/include/sane-srcrevs.inc: opkg-*: bump SRCREV Oct 23 07:49:43 03Steve Sakoman  07org.openembedded.dev * r88b3efc813 10openembedded.git/recipes/gnome/gnome-themes_2.28.0.bb: gnome-themes: clean up packaging Oct 23 07:49:46 03Steve Sakoman  07org.openembedded.dev * rf1bcec3227 10openembedded.git/recipes/dhcp/dhcp_3.0.2.bb: dhcp: create /var/lib/dhcp3 rather than /var/lib/dhcp Oct 23 07:49:49 03Koen Kooi  07org.openembedded.dev * r11581aaa9e 10openembedded.git/recipes/ekiga/ekiga_git.bb: ekiga git: bump SRCREV, change PV Oct 23 07:49:54 03Steve Sakoman  07org.openembedded.dev * rbe42c722e3 10openembedded.git/recipes/gnome/system-tools-backends_2.6.1.bb: system-tools-backends: fix packaging Oct 23 07:49:57 03Steve Sakoman  07org.openembedded.dev * r450852afb7 10openembedded.git/recipes/usb-gadget-mode/ (files/usb-gadget usb-gadget-mode.bb): usb-gadget: add serial gadget option Oct 23 07:50:00 03Steve Sakoman  07org.openembedded.dev * r791937c975 10openembedded.git/recipes/gnome/system-tools-backends_2.8.1.bb: system-tools-backends: disable 2.8.1 for now Oct 23 07:50:05 03Koen Kooi  07org.openembedded.dev * r9d98659a0e 10openembedded.git/recipes/bluez/bluez4_4.53.bb: bluez4: update to 4.53 Oct 23 07:50:08 03Koen Kooi  07org.openembedded.dev * rcc300143a1 10openembedded.git/recipes/powervr-drivers/ (libgles-omap3-3.00.00.09/rc.pvr libgles-omap3_3.00.00.09.bb): libgles-omap3: merge in changes from overo repo Oct 23 08:10:10 03Michael Smith  07org.openembedded.dev * rb57a5b04cd 10openembedded.git/recipes/cairo/ (cairo_1.8.8.bb files/dolt-fix.patch): (log message trimmed) Oct 23 08:10:10 cairo 1.8.8: fix libtool name in doltlibtool Oct 23 08:10:10 In OE libtool is renamed to ${host_alias}-libtool. Change doltlibtool Oct 23 08:10:10 script accordingly. Oct 23 08:10:10 Ripped from patch for xserver-xorg 1.6.1 Oct 23 08:10:14 (c10e9906a2ecea7e9c04bf967a8882d6b4cd1077) Oct 23 08:10:16 Signed-off-by: Michael Smith Oct 23 08:16:22 msmith__: you might want to put the patch to webkit-gtk too? Oct 23 08:21:00 zecke: good evening Oct 23 08:21:15 pb__: hey! it is not that late Oct 23 08:22:46 oh right, yeah, afternoon I guess Oct 23 08:25:44 I am not usually at work so early :-} Oct 23 09:05:13 morning Oct 23 09:07:57 03ghost  07org.openembedded.dreambox * r22a1904d33 10openembedded.git/packages/dreambox/dreambox-dvb-modules.bb: (log message trimmed) Oct 23 09:07:57 dreambox-dvb-modules.bb: another dm800/dm8000 driver update Oct 23 09:07:57 dm800 changes Oct 23 09:07:57 ------------- Oct 23 09:07:57 * tuner detection fixes Oct 23 09:07:59 dm800 and dm8000 changes Oct 23 09:08:01 ------------------------ Oct 23 09:08:07 03ghost  07org.openembedded.dreambox * rbdf5b9b0e1 10openembedded.git/packages/dreambox/dreambox-dvb-modules.bb: (log message trimmed) Oct 23 09:08:10 dreambox-dvb-modules.bb: update dm800 and dm8000 drivers Oct 23 09:08:12 changes dm800 Oct 23 09:08:14 ------------- Oct 23 09:08:16 * fix completely broken 1080p24/25/30 output (broken since previous drivers) Oct 23 09:08:18 * fix strange kernel ooopses on flash access (i.e. packages upgrades or install new ipk packages) Oct 23 09:08:22 changes dm800 and dm8000 Oct 23 09:08:24 03ghost  07org.openembedded.dreambox * r04686ab35b 10openembedded.git/packages/enigma2/enigma2-skins.bb: enigma2-skins.bb: add meta support Oct 23 09:08:29 03ghost  07org.openembedded.dreambox * r9db3fec3c7 10openembedded.git/packages/dreambox/dreambox-secondstage.bb: dreambox-secondstage.bb: update dm8000 secondstage loader because of changed oled initialisation sequence Oct 23 09:08:35 03ghost  07org.openembedded.dreambox * re6cebc5c9b 10openembedded.git/: Merge branch 'org.openembedded.dreambox' of git://git.openembedded.net/openembedded into org.openembedded.dreambox Oct 23 09:08:38 03ghost  07org.openembedded.dreambox * rb8a294c073 10openembedded.git/packages/enigma2/enigma2.bb: enigma2.bb: add gst-plugin-autodetect... its needed since gstreamer update to get the alsasink working again (i.e. for flac, ogg playback) Oct 23 09:08:46 03ghost  07org.openembedded.dreambox * r8e536df7ca 10openembedded.git/packages/gstreamer/gst-plugin-dvbmediasink.bb: update mediasink Oct 23 09:08:49 03ghost  07org.openembedded.dreambox * r4c8b61858e 10openembedded.git/packages/dreambox/dreambox-dvb-modules.bb: Oct 23 09:08:52 dreambox-dvb-modules.bb: update dm8000 drivers Oct 23 09:08:54 * fix no more working non linear scaling Oct 23 09:08:58 * fix inaccurate oled brightness controls since new 2nd stage loader with fix for new oled displays Oct 23 09:09:00 03ghost  07org.openembedded.dreambox * rb42f30663c 10openembedded.git/packages/gstreamer/ (3 files in 2 dirs): Oct 23 09:09:03 packages/gstreamer: update gst-plugins-ugly and gst-plugins-bad Oct 23 09:09:05 small fix for ac3 detection in mpegtsdemux Oct 23 09:09:07 03ghost  07org.openembedded.dreambox.release_25 * rf39f525831 10openembedded.git/packages/dreambox/dreambox-secondstage.bb: dreambox-dvb-secondstage.bb: use latest secondstage loader for dm800 and dm8000 Oct 23 09:09:13 03ghost  07org.openembedded.dreambox.release_25 * r52924526e5 10openembedded.git/packages/gstreamer/gst-plugin-dvbmediasink.bb: update mediasink Oct 23 09:09:16 03ghost  07org.openembedded.dreambox.release_25 * r905034087e 10openembedded.git/packages/dreambox/dreambox-dvb-modules.bb: dreambox-dvb-modules.bb: fix typo Oct 23 09:09:23 03ghost  07org.openembedded.dreambox.release_25 * r16b13cd547 10openembedded.git/packages/enigma2/enigma2.bb: enigma2.bb: push srcdate for release Oct 23 09:09:28 03ghost  07org.openembedded.dreambox.release_25 * rf91d0602d4 10openembedded.git/packages/dreambox/dreambox-dvb-modules.bb: dreambox-dvb-modules.bb: backport some changes/bugfixes from current dreambox-dvb-modules for dm800/dm8000 .. but dont backport new ETIMEDOUT handling for faster service searching Oct 23 09:28:24 03Henning Heinold  07org.openembedded.dev * r402bfde509 10openembedded.git/recipes/linux-libc-headers/linux-libc-headers_2.6.20.bb: Oct 23 09:28:24 linux-libc-headersr-2.6.20: add unifdef.patch Oct 23 09:28:24 * bump PR Oct 23 09:28:25 03Henning Heinold  07org.openembedded.dev * rb6db22fb40 10openembedded.git/conf/checksums.ini: checksum.ini: add sums for mkbc.c from jalimo overlay Oct 23 09:28:40 woglinde: you track Poky? Oct 23 09:29:21 03Richard Purdie  07master * r05faed8848 10bitbake.git/lib/bb/fetch/hg.py: Oct 23 09:29:21 Add hg fetcher auto revisioning, patch from Carl Simonson Oct 23 09:29:21 Signed-off-by: Richard Purdie Oct 23 09:29:28 hrw no Oct 23 09:29:31 03Richard Purdie  07master * r8462cc8e1d 10bitbake.git/lib/bb/build.py: Oct 23 09:29:31 build.py: Make sure the logfile exists before checking its size. Previously this broke upon recursive calls. Oct 23 09:29:31 Signed-off-by: Richard Purdie Oct 23 09:30:10 hrw I am trying got get chinook-compat useable again Oct 23 09:30:17 to build openjdk Oct 23 09:31:21 woglinde: I just pushed 2.6.20 headers fix into Poky Oct 23 09:32:22 *g* Oct 23 09:32:36 strangly it compiled yesterday Oct 23 09:32:38 the headers Oct 23 09:32:44 and today I needed the fix Oct 23 09:34:11 you did host upgrade? Oct 23 09:34:19 lol Oct 23 09:34:25 the headers are old Oct 23 09:34:32 and 2.6.23 needed the fix Oct 23 09:43:34 03Richard Purdie  07master * r14748abaf4 10bitbake.git/lib/bb/fetch/svk.py: Oct 23 09:43:34 fetch/svk: Fix up breakage. Evidently nobody uses svk Oct 23 09:43:34 Signed-off-by: Richard Purdie Oct 23 09:44:16 jo mickeyl Oct 23 09:44:25 morning all Oct 23 09:44:34 jo bitbake-master Oct 23 09:45:51 morsche Oct 23 09:45:56 hi mickeyl Oct 23 09:46:05 yo RP Oct 23 09:46:54 03Henning Heinold  07org.openembedded.dev * r1ec5d58130 10openembedded.git/recipes/gcc/ (4 files in 2 dirs): Oct 23 09:46:54 gcc-cross_csl-arm-2005q3: put into useable state again Oct 23 09:46:54 * add some patches needed for libstdc++ compilation Oct 23 09:46:54 * disable fortran and objc build for now Oct 23 09:46:54 * bump PR Oct 23 09:47:04 03Henning Heinold  07org.openembedded.dev * r1258abe38a 10openembedded.git/conf/distro/chinook-compat.conf: chinook-compat: add new toolchain options and stick to util-linux Oct 23 09:52:09 jo morning florian Oct 23 09:55:16 pb__: could you check if my last email makes sense? I said that when having a libfoo.a and .debug/libfoo.a linking libfoo.a will not automatically take in debug symbols Oct 23 09:55:29 hi florian RP mickeyl Oct 23 09:56:09 hey hrw Oct 23 09:56:36 zecke: yes, I think that is correct Oct 23 09:56:37 mickey|office: you iphone modplayer... where from you are fetching mods? Oct 23 09:57:01 hrw: mirrors of the modland collection Oct 23 09:58:12 morning xora Oct 23 09:58:25 hi XorA Oct 23 09:58:35 hey hrw mickey|office Oct 23 09:59:22 good morning Oct 23 10:01:02 * XorA thinks ubuntu will have a difficult time competing with Win7 Oct 23 10:01:34 XorA: it reminds me of code classic Oct 23 10:02:23 I installed Karmic RC1 and Win7 on the same machine last night, one works perfect, one is totally fscked, Ill let you guess which is which Oct 23 10:03:15 03Thomas Zimmermann  07shr/merge * re75593bd5c 10openembedded.git/ (conf/checksums.ini recipes/cellhunter/cellhunter_0.5.0.bb): Oct 23 10:03:15 Cellhunter: update to 0.5.0 Oct 23 10:03:15 Signed-off-by: Klaus Kurzmann Oct 23 10:03:26 03Thomas Zimmermann  07shr/merge * re41832586c 10openembedded.git/recipes/netbase/ (5 files in 4 dirs): Oct 23 10:03:26 shr/merge: netbase: update config files for shr and om-* Oct 23 10:03:26 Signed-off-by: Klaus Kurzmann Oct 23 10:08:27 the makefile of omapfbplay says "omapfbplay: omapfbplay.o yuv.o", and that's about it. what is it that makes OE link the .o's into an executable in the end? Oct 23 10:35:13 if the rule says "omapfbplay: omapfbplay.o yuv.o", OE compiles and links. If I change the name of the target, it will only compile the two object files Oct 23 10:37:46 tassleho1f: thats nothing to do with OE, I would suggest reading the very good GNU Make manual Oct 23 10:38:51 XorA: I'll take the suggestion. Oct 23 10:44:25 maybee reading autools manual is better Oct 23 10:44:45 * tassleho1f curses implicit rules Oct 23 10:45:05 hello every one Oct 23 10:46:55 somone refer me to this irc chat. to solve my issue with beagleboard. Oct 23 10:46:55 kindly read http://groups.google.com/group/beagleboard/browse_thread/thread/c850aaf45cba63e9# my post Oct 23 10:48:04 i really need help on it. its urgent Oct 23 10:49:40 fadi: first of all - those 3 messages on start are harmless Oct 23 10:49:55 fadi: so if you got "beagleboard login:" on serial port you have it working Oct 23 10:51:05 fadi: you might try #angstrom or #beagle for those issues. Oct 23 10:53:07 hrw: thanks Oct 23 10:53:39 what about first issue? is this normal that screen is freeze Oct 23 10:54:16 03Martin Jansa  07shr/merge * re601ac0b73 10openembedded.git/conf/distro/ (include/preferred-shr-versions.inc shr.conf): Oct 23 10:54:16 shr/merge: Replace angstrom-2007 include with sane-toolchain and remove preferred versions for toolchain from shr-versions Oct 23 10:54:16 Signed-off-by: Klaus Kurzmann Oct 23 10:54:17 03Martin Jansa  07shr/merge * rc55c21ec80 10openembedded.git/conf/distro/include/preferred-shr-versions.inc: Oct 23 10:54:17 fadi: an upto date Angstrom also requires an uptodate u-boot Oct 23 10:54:19 shr/merge: Remove preferred versions for xorg packages, defined in separate include Oct 23 10:54:21 Signed-off-by: Klaus Kurzmann Oct 23 10:54:23 03Martin Jansa  07shr/merge * r876903a4dc 10openembedded.git/conf/distro/shr.conf: Oct 23 10:54:26 Set DISTRO_REVISION (now sync with Angstrom) Oct 23 10:54:28 Signed-off-by: Klaus Kurzmann Oct 23 10:54:34 fadi: the fact you get a u-boot logo indicates its not upto date Oct 23 10:56:28 XorA: ohoo. is there any better way to update and refresh my NAND? Oct 23 10:56:52 fadi: http://www.angstrom-distribution.org/demo/beagleboard/README.txt Oct 23 10:57:48 fadi: but this really is the wrong channel, goto #beagle Oct 23 10:58:32 XorA: sorry my mistake. i will switch the channel Oct 23 11:18:33 03Klaus Kurzmann  07shr/merge * re72932858a 10openembedded.git/conf/distro/include/shr-autorev.inc: Oct 23 11:18:33 shr-autorev.inc: sync with shr-autorev-unstable.inc before removing that one Oct 23 11:18:33 Signed-off-by: Klaus Kurzmann Oct 23 11:18:35 03Klaus Kurzmann  07shr/merge * rcb55283b24 10openembedded.git/conf/distro/include/shr-autorev-unstable.inc: Oct 23 11:18:35 shr-autorev-unstable.inc: remove it - we don't want (and need) two autorev includes Oct 23 11:18:37 Signed-off-by: Klaus Kurzmann Oct 23 11:18:39 03Klaus Kurzmann  07shr/merge * r3b75820ce8 10openembedded.git/conf/distro/shr.conf: Oct 23 11:18:41 shr.conf: require shr-autorev.inc Oct 23 11:18:43 Signed-off-by: Klaus Kurzmann Oct 23 11:23:36 03Klaus Kurzmann  07shr/merge * r0a9e591dc2 10openembedded.git/conf/distro/include/shr-autorev.inc: Oct 23 11:23:36 shr-autorev.inc: remove doubled revs for mesa-dri and libdrm Oct 23 11:23:36 Signed-off-by: Klaus Kurzmann Oct 23 12:04:36 03Leon Woestenberg  07org.openembedded.dev * r449fca9901 10openembedded.git/ (3 files in 3 dirs): Oct 23 12:04:36 u-boot: Patched and tested a GIT version for MACHINE=calamari. Oct 23 12:04:36 Tested on the Freescale MPC8536DS board. Oct 23 12:04:36 Includes a workaround for a known buggy compiler. Oct 23 12:04:36 Signed-off-by: Leon Woestenberg Oct 23 12:06:33 hi , all thanks for all people , this irc channel is really helpful and helping people Oct 23 12:11:33 03Koen Kooi  07org.openembedded.dev * r4249269bf8 10openembedded.git/ (conf/checksums.ini recipes/bluez/bluez4_4.56.bb): bluez4: update to 4.46, 4.53 has build from scratch problems Oct 23 12:24:05 zecke_: now that i see how webkit-gtk works around the dolt ${TARGET_PREFIX}libtool stuff, I think i actually like it better :) Oct 23 12:24:22 hack hack hack in a recipe instead of hack hack hack in some autoconf cruft Oct 23 12:26:28 03Marco Cavallini  07org.openembedded.dev * rc8dc29972b 10openembedded.git/recipes/glibc/glibc-package.bbclass: glibc-package.bbclass : added KaeilOS hack supporting glibc 2.9 Oct 23 12:26:29 03Marco Cavallini  07org.openembedded.dev * r0e586e8140 10openembedded.git/conf/distro/include/kaeilos.inc: kaeilos.inc : temporarily aligned to latest Angstrom settings Oct 23 12:26:30 03Marco Cavallini  07org.openembedded.dev * r15bef3123e 10openembedded.git/conf/distro/kaeilos.conf: Oct 23 12:26:30 distro/kaeilos.conf : switched to glibc 2.9 Oct 23 12:26:32 * see also glibc-package.bbclass : added KaeilOS hack supporting glibc 2.9 Oct 23 12:26:34 * see also bluez4* : temporary solution just to replace at least bluez-libs Oct 23 12:26:36 * bumped PR to be sure glibc will be aligned on resulting image Oct 23 12:32:20 03Marco Cavallini  07koen/static-libs-rework * r0e586e8140 10openembedded.git/conf/distro/include/kaeilos.inc: kaeilos.inc : temporarily aligned to latest Angstrom settings Oct 23 12:32:21 03Koen Kooi  07koen/static-libs-rework * rc7cc360774 10openembedded.git/classes/package.bbclass: package bbclass: don't strip static libs so aggressively Oct 23 12:32:21 03Koen Kooi  07koen/static-libs-rework * r6591ebc188 10openembedded.git/conf/bitbake.conf: bitbake conf: put static libs in ${PN}-static Oct 23 12:32:22 03Marco Cavallini  07koen/static-libs-rework * r15bef3123e 10openembedded.git/conf/distro/kaeilos.conf: Oct 23 12:32:24 distro/kaeilos.conf : switched to glibc 2.9 Oct 23 12:32:26 * see also glibc-package.bbclass : added KaeilOS hack supporting glibc 2.9 Oct 23 12:32:28 * see also bluez4* : temporary solution just to replace at least bluez-libs Oct 23 12:32:30 * bumped PR to be sure glibc will be aligned on resulting image Oct 23 12:32:32 03Marco Cavallini  07koen/static-libs-rework * rc8dc29972b 10openembedded.git/recipes/glibc/glibc-package.bbclass: glibc-package.bbclass : added KaeilOS hack supporting glibc 2.9 Oct 23 12:32:35 03Koen Kooi  07koen/static-libs-rework * r82cffa0b57 10openembedded.git/recipes/glibc/glibc-package.bbclass: (log message trimmed) Oct 23 12:32:38 glibc package bbclass: tweak library packaging Oct 23 12:32:40 Static libs move around a bit: Oct 23 12:32:42 glibc-dev: Oct 23 12:32:49 /usr/lib/libpthread_nonshared.a Oct 23 12:32:50 /usr/lib/libc_nonshared.a Oct 23 12:32:52 glibc-static: Oct 23 12:34:21 what do you use to checksum sorting? contrib/source-checker/oe-checksums-sorter.py sorts just alphabetically and in checksums.ini is acpid-1.0.10 AFTER acpid-1.0.2 so maybe someone has sorter respecting versions... Oct 23 12:39:12 i think most people just use the sorter when they have more than one new package Oct 23 12:39:31 it's not that important anyways, it's rather for limiting the amount of merge conflicts Oct 23 12:42:31 mickey|office: good afternoon Oct 23 12:42:42 good afternoon to you Oct 23 12:42:42 he pb Oct 23 12:42:51 ho woglinde Oct 23 12:43:31 ~seen vivi Oct 23 12:43:33 mickey|office: i haven't seen 'vivi' Oct 23 12:43:37 JaMa: the checksums sorter should be used always Oct 23 12:50:53 XorA: this one? Oct 23 12:51:30 XorA: I'm using it always.. but I don't want to move acpid and dbus (which are now sorted version wise) just because oe-checksums-sorter.py sorts it alphabetically Oct 23 12:52:47 I wouldn't worry about it. If you don't move them now, someone else will next time they sort the file. Oct 23 12:53:44 I never used the sorter Oct 23 12:53:59 well, use the sorter :) Oct 23 12:54:07 The exact order is pretty much irrelevant: you could sort it according to the number of characters in the filename, or alphabetically by sha256sum, or anything else that you wanted. The only purpose of the sorter is to have a (roughly) deterministic order so that conflicts will be less bad if two people add the same file. Oct 23 12:54:09 OK, if you all use that sorter, then its fine.. Oct 23 12:54:59 pb_, exactly :) Oct 23 12:55:03 Or of course you could join my campaign for the abolition of checksums.ini and not worry about it. :-) Oct 23 12:55:12 pb_, not likely Oct 23 12:55:25 hehe Oct 23 12:55:25 it actually is helpful to know when upstream dorks with stuff Oct 23 12:55:38 03Koen Kooi  07koen/static-libs-rework * rbfa6dd452d 10openembedded.git/recipes/glibc/glibc-package.bbclass: (log message trimmed) Oct 23 12:55:38 glibc package bbclass: tweak library packaging Oct 23 12:55:38 libcidn and libmemusage have there own package now Oct 23 12:55:38 Static libs move around a bit: Oct 23 12:55:38 glibc-dev: Oct 23 12:55:39 /usr/lib/libpthread_nonshared.a Oct 23 12:55:43 /usr/lib/libc_nonshared.a Oct 23 12:56:27 Crofton: its a bad database design though :-) Oct 23 12:56:30 Crofton: I agree with that aim, but I am not convinced that the current checksums.ini is a good way of achieving it. Oct 23 12:57:05 or, even, that it does achieve it in any meaningful sense, since the usual response to a change of upstream checksum seems just to be to grumble a bit and then update the file to match the new one. Oct 23 12:57:46 I changed the asterisk checksum after working out that upstream had to remove some sound files for copyright reasons Oct 23 13:05:16 right, but looking at the history of checksums.ini I also see things like "update entry for libxml-twig-perl; The Twig people are idiots because they keep replacing the tarball", "Upstream package changed (!). New checksums", "It seems that the upstream tarball was changed. Update checksum" and suchlike. Oct 23 13:05:49 in the twig case it looks like the new tarball did certainly have different contents to the old one; it is hard to tell from a superficial inspection whether this was true of the other two packages or not. Oct 23 13:06:03 pb__: ack... noone really cares about _what_ has changed. So how it is used currently its mostly overhead... Oct 23 13:06:16 right, exactly. Oct 23 13:06:55 and, worse, I think it fools people into thinking that they are protected from unexpected upstream changes when in fact they are not. Oct 23 13:07:55 indeed Oct 23 13:08:41 i.e. if I was a naive user, I might think that there was no chance that the twig tarball could change without me noticing, because the checksum is right there in checksums.ini. but, of course, if someone has updated checksums.ini without me noticing that either, I actually have no protection at all. Oct 23 13:09:30 very true Oct 23 13:14:59 so, it sounds like we need to talk to people making casual updates then Oct 23 13:15:08 what exactly do you propose? Oct 23 13:15:36 I think no one, really means the people that want to get rid of checksums Oct 23 13:16:38 As a first step I would propose putting the checksums in the .bb recipes themselves, rather than centrally in checksums.ini, and requiring at a minimum a PR bump when the checksum changes. Oct 23 13:17:27 That would go some way towards eliminating "submarine" changes that can go completely undetected since everybody with the same PR will have built from the same source. Oct 23 13:17:46 pb__: sounds like you got another topic for OEDEM Oct 23 13:18:09 pb__: but in general your reasoning is sound, I had a lot of issue at OM with merging checksums Oct 23 13:18:39 XorA: good idea, I will add it to the agenda Oct 23 13:19:31 pb__: I thank that was the thrust of the last oe-devel debate about checksums ;-) - Works for me but would beed work on the PR stuff to tie it in with the checksum but it would help with the things that only come up on an rm -rf tmp ;-) Oct 23 13:20:37 pb_, and avoid the merge headache .... Oct 23 13:21:27 -r :-) Oct 23 13:23:50 are those two errors already known in .dev http://pastebin.com/m3b424f4a ? Oct 23 13:24:51 hihi checksum.ini Oct 23 13:25:16 i hate that file Oct 23 13:25:26 ynezz, why? Oct 23 13:25:42 it's pain to work with it Oct 23 13:26:01 too many commands for me and if you don't use OE daily it's pain^2 Oct 23 13:26:14 ynezz: do not panic Oct 23 13:26:27 one variable in local.conf and your builds ignore checksums.ini Oct 23 13:27:23 yes, I know Oct 23 13:27:40 * JaMa has 2 scripts to work with checksums.. Oct 23 13:28:22 1 adds checksum for url I already know.. 2nd calls -c unpack for list of recipes and then download all urls missing in checksums.ini and adds it.. Oct 23 13:28:46 JaMa: add them to contrib/ Oct 23 13:29:21 hrw: they are ugly, but I can make them a bit more usable I guess Oct 23 13:29:27 would be nice Oct 23 13:29:40 hrw: but still calling sorter after every checksum is a bit slow :) Oct 23 13:30:00 JaMa: was it you that did the improved sorter? Oct 23 13:31:16 XorA: I proposed something about checksums.ini but its quite a long time ago.. need to check my outbox :) Oct 23 13:31:34 JaMa: I meant the --inplace, that was cool :-D Oct 23 13:31:55 http://shr.pastebin.ca/1640075 - those 2 ugly scripts, suggestions before making them nicer? Oct 23 13:32:08 XorA: I don't remember --inplace.. so probably not me.. Oct 23 13:34:39 XorA: I proposed using only filenames in checksums.ini if possible (no conflict) and changed base.class and checksums.ini for that (about a year ago) Oct 23 13:36:35 JaMa: if possible was problem Oct 23 13:37:33 yes but now the same file from different mirror with different checksum is maybe problem too Oct 23 13:38:11 I like pb__ idea to move checksums to recipes, or at least to recipe directory Oct 23 13:38:21 the locate new sources based on checksum was cool :-) Oct 23 13:39:00 now if only we could insert all the sources into freenet and just pull them out when needed :-) Oct 23 13:39:27 :) Oct 23 13:40:08 JaMa: format with [filename]\nurl0=somewher\nurl1=otherplace\nmd5sum= etc would be handy Oct 23 13:47:30 Who knows about util-linux? Oct 23 13:47:51 SlugOS builds are moaning about a PREFERRED_PROVIDER for some time now, and I've been ignoring that. Oct 23 13:48:04 Is there any reason I don't want util-linux-ng? Oct 23 14:05:19 03Henning Heinold  07org.openembedded.dev * rde7bfd4382 10openembedded.git/recipes/util-linux-ng/ (3 files in 2 dirs): Oct 23 14:05:19 util-linux-ng: for chinook-compat let util-linux-ng compile with older libtool Oct 23 14:05:19 *bump INC_PR Oct 23 14:07:33 03Henning Heinold  07koen/static-libs-rework * rde7bfd4382 10openembedded.git/recipes/util-linux-ng/ (3 files in 2 dirs): Oct 23 14:07:33 util-linux-ng: for chinook-compat let util-linux-ng compile with older libtool Oct 23 14:07:33 *bump INC_PR Oct 23 14:07:35 03Koen Kooi  07koen/static-libs-rework * r7c8ae637e9 10openembedded.git/conf/bitbake.conf: bitbake conf: put static libs in ${PN}-static Oct 23 14:07:36 03Koen Kooi  07koen/static-libs-rework * re1e484e471 10openembedded.git/classes/package.bbclass: package bbclass: don't strip static libs so aggressively Oct 23 14:07:39 03Koen Kooi  07koen/static-libs-rework * r08af581548 10openembedded.git/recipes/glibc/glibc-package.bbclass: (log message trimmed) Oct 23 14:07:42 glibc package bbclass: tweak library packaging Oct 23 14:07:44 libcidn and libmemusage have there own package now Oct 23 14:07:46 Static libs move around a bit: Oct 23 14:07:48 glibc-dev: Oct 23 14:07:50 /usr/lib/libpthread_nonshared.a Oct 23 14:07:52 /usr/lib/libc_nonshared.a Oct 23 14:18:40 quilt is so broken. Depending on the obsolete RFC822 date output is just insane given that 2822 replaced it some 8 years ago Oct 23 14:19:20 * blindvt adds dummy 822 support to return 2822 ourput to busybox and --without-date to quilt Oct 23 14:20:04 nice testbed you have ;) Oct 23 15:09:10 03Martin Jansa  07shr/merge * rad75ad90dd 10openembedded.git/ (6 files in 5 dirs): Oct 23 15:09:10 shr/merge: Latest xorg stuff 2009-10-23 only xserver stay with 1.7.1 instead of 1.7.99.1 Oct 23 15:09:10 Signed-off-by: Klaus Kurzmann Oct 23 15:16:16 03Marco Cavallini  07org.openembedded.dev * r15bef3123e 10openembedded.git/conf/distro/kaeilos.conf: Oct 23 15:16:16 distro/kaeilos.conf : switched to glibc 2.9 Oct 23 15:16:16 * see also glibc-package.bbclass : added KaeilOS hack supporting glibc 2.9 Oct 23 15:16:16 * see also bluez4* : temporary solution just to replace at least bluez-libs Oct 23 15:16:16 * bumped PR to be sure glibc will be aligned on resulting image Oct 23 15:16:22 03Marco Cavallini  07org.openembedded.dev * r0e586e8140 10openembedded.git/conf/distro/include/kaeilos.inc: kaeilos.inc : temporarily aligned to latest Angstrom settings Oct 23 15:16:25 03Marco Cavallini  07org.openembedded.dev * rc8dc29972b 10openembedded.git/recipes/glibc/glibc-package.bbclass: glibc-package.bbclass : added KaeilOS hack supporting glibc 2.9 Oct 23 15:16:28 03Marco Cavallini  07org.openembedded.dev * r0bcccaf028 10openembedded.git/recipes/xorg-xserver/xserver-xorg-conf/ronetix-pm9263/xorg.conf: xserver-xorg-conf: added config for ronetix-pm9263 Oct 23 15:16:33 03Marco Cavallini  07org.openembedded.dev * re528052544 10openembedded.git/: Merge branch 'org.openembedded.dev' of git://git.openembedded.org/openembedded into org.openembedded.dev Oct 23 15:21:25 hmmm... strange, I already sent tree of these patches, I only added one file now :-S Oct 23 15:23:41 OH NO!!!!! Oct 23 15:26:10 hm Oct 23 15:26:14 ? Oct 23 15:27:03 mckoan hm try look with qgit Oct 23 15:27:19 my checkins and koen was nearly at the same time Oct 23 15:27:50 I hope that didnt break something Oct 23 15:30:17 cgit is really handy Oct 23 15:30:51 I suspect its an artifact of the merge as the child refs changed when woglinde and koen checked in Oct 23 15:30:52 woglinde: I can't fisure out what happened Oct 23 15:32:02 gitk will show you Oct 23 15:32:25 but i wouldnt worry about it Oct 23 15:34:25 XorA: I'm using gitk, I see my patches at 09:40 and now again in a branch Oct 23 15:34:50 mckoan: well thats normal then, branches are totally seperate Oct 23 15:36:24 mckoan try --rebase before pushing always Oct 23 15:37:12 woglinde: I always do git pull before pushing Oct 23 15:37:45 git pull --rebase is probably what you really mean though Oct 23 15:37:51 yep Oct 23 15:38:11 * mwester learned that the hard way Oct 23 15:38:21 ok thanx all, I'm going to do that next time Oct 23 15:40:17 is there some trick to honor proxy settings? I find myself changing the default bitbake.conf to +FETCHCOMMAND_wget = "/usr/bin/env 'PATH=${PATH} http_proxy=${http_proxy} ftp_proxy=${ftp_proxy} no_proxy=${no_proxy}' wget -t 5 --passive-ftp -P ${DL_DIR} ${URI}" Oct 23 15:40:49 blindvt: you need to sety the EXTRA WHITE stuff Oct 23 15:40:53 what's this ? $ git pull --rebase Oct 23 15:40:54 It looks like git-am is in progress. Cannot rebase. Oct 23 15:40:56 blindvt: look in the manual Oct 23 15:41:02 git am --abort Oct 23 15:41:27 koan@quad:openembedded$ git pull --rebase Oct 23 15:41:27 Current branch org.openembedded.dev is up to date. Oct 23 15:41:29 thx Oct 23 15:48:00 anyone familiar with X? do the kdrive x servers use the xf86-input- modules, or is their input built in? Oct 23 15:49:38 XorA, ah, thanks! ( http://paste.debian.net/49767/ ) Oct 23 15:52:00 kergoth: built in Oct 23 15:52:05 k Oct 23 15:52:11 any idea what input devices it supports, offhand? Oct 23 15:52:13 just ts? Oct 23 15:52:24 kergoth: evdev I think Oct 23 15:52:30 hmm, k Oct 23 15:52:33 thanks Oct 23 15:52:40 * kergoth hasn't messed with X in some time Oct 23 15:52:50 I have vague memorirs of -mouse help being an argument Oct 23 15:53:10 but I could very well be mixing stuff up Oct 23 15:58:04 morning kergoth Oct 23 15:58:08 hey pb_ Oct 23 15:59:10 hrmph, hard to get work done when the work git server is down.. Oct 23 16:01:57 kergoth: IIRC kdrive uses only input built in Oct 23 16:04:03 kergoth: yes, looks like git server is down Oct 23 16:04:09 have a nice rest of the day Oct 23 16:11:05 03Marco Cavallini  07koen/static-libs-rework * r0e586e8140 10openembedded.git/conf/distro/include/kaeilos.inc: kaeilos.inc : temporarily aligned to latest Angstrom settings Oct 23 16:11:06 03Koen Kooi  07koen/static-libs-rework * rfa1c574301 10openembedded.git/conf/bitbake.conf: bitbake conf: put static libs in ${PN}-static Oct 23 16:11:07 03Marco Cavallini  07koen/static-libs-rework * re528052544 10openembedded.git/: Merge branch 'org.openembedded.dev' of git://git.openembedded.org/openembedded into org.openembedded.dev Oct 23 16:11:09 03Marco Cavallini  07koen/static-libs-rework * rc8dc29972b 10openembedded.git/recipes/glibc/glibc-package.bbclass: glibc-package.bbclass : added KaeilOS hack supporting glibc 2.9 Oct 23 16:11:12 03Koen Kooi  07koen/static-libs-rework * r1b521e0edd 10openembedded.git/classes/package.bbclass: package bbclass: don't strip static libs so aggressively Oct 23 16:11:15 03Marco Cavallini  07koen/static-libs-rework * r15bef3123e 10openembedded.git/conf/distro/kaeilos.conf: Oct 23 16:11:18 distro/kaeilos.conf : switched to glibc 2.9 Oct 23 16:11:20 * see also glibc-package.bbclass : added KaeilOS hack supporting glibc 2.9 Oct 23 16:11:22 * see also bluez4* : temporary solution just to replace at least bluez-libs Oct 23 16:11:24 * bumped PR to be sure glibc will be aligned on resulting image Oct 23 16:11:26 03Marco Cavallini  07koen/static-libs-rework * r0bcccaf028 10openembedded.git/recipes/xorg-xserver/xserver-xorg-conf/ronetix-pm9263/xorg.conf: xserver-xorg-conf: added config for ronetix-pm9263 Oct 23 16:11:36 03Koen Kooi  07koen/static-libs-rework * rbe0dfd0214 10openembedded.git/recipes/glibc/glibc-package.bbclass: (log message trimmed) Oct 23 16:11:39 glibc package bbclass: tweak library packaging Oct 23 16:11:41 libcidn and libmemusage have there own package now Oct 23 16:11:45 Static libs move around a bit: Oct 23 16:11:47 glibc-dev: Oct 23 16:11:49 /usr/lib/libpthread_nonshared.a Oct 23 16:11:51 hms+ Oct 23 16:11:51 /usr/lib/libc_nonshared.a Oct 23 16:11:53 again? Oct 23 16:12:06 there is something worng Oct 23 16:12:23 cia error? Oct 23 16:12:44 woglinde: last one was koen's branch Oct 23 16:13:23 previously was oe.dev Oct 23 16:13:34 * mwester wishes we could color-code or otherwise separate the CIA messages for the major branches in OE (dev, stable, koen's) Oct 23 16:13:49 ah Oct 23 16:15:42 woglinde: anyway patches appear sent 4 or even 5 times ..see openembeded-commit ML ;) Oct 23 16:15:48 bbl Oct 23 17:15:55 jo thesing Oct 23 17:16:04 hi all Oct 23 17:16:11 hi woglinde Oct 23 17:16:33 do you have a patch for armv4 eabi ready? Oct 23 18:05:59 heys Oct 23 18:08:00 how can i tell which ipk files are required to install on device? I'm working on getting fuse over, and first of all there are 2 directories: armv5te and magician... not sure which one it is, and then for the fuse package there are mutliple ipks listed: fuse-utils-dbg... fuse-utils... libfuse2_2.... libfuse-dev... Oct 23 18:08:22 those are the ones in the arm5te folder Oct 23 18:08:49 then the one in the magicial folder is kernel-module-fuse-2.6...ipk Oct 23 18:09:09 Im guessing that the magician folder correspends to kernel modules only or something? Oct 23 18:12:02 person456: the magician folder has packages specific to the magician. Oct 23 18:12:28 mainly kernel modules. Oct 23 18:12:54 thesing: thats what it looks like yeah Oct 23 18:13:20 thesing: ok so i guess i just need all ls * | grep fuse packages Oct 23 18:13:42 thesing: what do you think Oct 23 18:14:15 I don't know fuse very much. But it can't hurt to insall all those. (but not the -dbg packages) Oct 23 18:14:45 thesing: what are the dbg packages for... some sort of debug thing obviously, but what do those contain typically Oct 23 18:14:55 thesing: a binary compiled for debug mode? Oct 23 18:14:55 and you don't need the -dev packages. (if you don't want to compile stuff that needs them) Oct 23 18:15:05 yes. Oct 23 18:15:45 thesing: well ok so the dev package contains like the header files and stuff Oct 23 18:16:04 thesing: im new to ipk Oct 23 18:16:13 thesing: :) Oct 23 18:16:38 it's OE that controls what files go into what packages, not the file format. it can make deb, rpm, ipk, .. :) Oct 23 18:19:08 woglinde, khem: can you please tell me what to change to compile eabi for collie? Oct 23 18:21:01 is there a way to package a bunch of ipks into one ipk? Oct 23 18:21:05 * kergoth plays with witpa Oct 23 18:21:12 person456: why would you want to? Oct 23 18:22:41 kergoth: wouldnt it be annoying to type ipkg install .pkh for every single pkg? Oct 23 18:23:00 why would you want to install every package? Oct 23 18:23:03 kergoth: can you do ipkg install *.ipk? Oct 23 18:23:14 yes, ipkg can accept wildcards Oct 23 18:23:19 see opkg --help Oct 23 18:23:24 shows examples right at hte bottom Oct 23 18:23:27 oh ok then you are right Oct 23 18:24:09 still seems pointless to install packages you don't need, but suit yourself :) Oct 23 18:30:38 if i have 3 different ipks that depend on eachother in some order, do i have to install them in the order required or do you know if ipkg will seek out local ipk dependencies in current directory Oct 23 18:33:21 if they are dependent on an order, then I assume the best way to have a distributable folder of ipks would be to prefix the order of installation to the front and then do ipkg install *.ipk Oct 23 18:33:33 cause then it would do 01_... 02... etc. Oct 23 18:36:24 easier to just make a package feed and install from there, like yum/apt-get, obeying dependencies Oct 23 18:36:27 you can have a local feed.. Oct 23 18:37:28 kergoth: with ipkg or opkg? I only got ipkg Oct 23 18:40:19 either Oct 23 18:40:30 google for info on opkg/ipkg package feeds Oct 23 18:41:50 03Thomas Zimmermann  07shr/merge * r73b0141a9f 10openembedded.git/conf/distro/include/preferred-shr-versions.inc: Oct 23 18:41:50 shr/merge: nfs-utils: update to 1.1.2 because 1.0.6 doesn't provide nfs-utils-client Oct 23 18:41:50 Signed-off-by: Klaus Kurzmann Oct 23 18:42:06 03Thomas Zimmermann  07shr/merge * r832ffd725a 10openembedded.git/conf/distro/shr.conf: Oct 23 18:42:06 shr/merge: correct order of required configs in shr.conf Oct 23 18:42:06 Signed-off-by: Klaus Kurzmann Oct 23 18:42:06 03Thomas Zimmermann  07shr/merge * r7f66314113 10openembedded.git/recipes/tasks/task-shr-minimal.bb: Oct 23 18:42:08 shr/merge: use util-linux-ng-fdisk instead of util-linux-fdisk Oct 23 18:42:10 Signed-off-by: Klaus Kurzmann Oct 23 18:49:40 re Oct 23 19:08:16 re Oct 23 19:12:15 mckoan: wb Oct 23 19:14:14 florian: wb Oct 23 19:15:24 wb all :-) Oct 23 19:19:44 oh well... how can a company that makes that beautiful hardware write that ugly drivers Oct 23 19:21:09 florian: with ease ;-) Oct 23 19:25:20 It looks a little bit like an interpretation of spect to make it as cheap as possible. Oct 23 19:25:32 (apart from the bad code quality) Oct 23 19:27:14 gm all Oct 23 19:28:37 Sound support... well it makes noise yes. But the alsa device does not have a single setting! Oct 23 19:28:45 hi khem Oct 23 19:35:36 florian: is mail forwarding on openembedded.org fixed Oct 23 19:36:14 khem: was it broken? Oct 23 19:39:28 florian: doesnt work for me Oct 23 19:42:37 * florian checks Oct 23 19:44:10 hmm... works for me Oct 23 19:44:29 but i don't count Oct 23 19:45:25 we had this topic some weeks ago... Oct 23 19:45:59 * florian tries to remember how mail handling for oe works Oct 23 19:47:10 florian: can you check if it is set up for me :) Oct 23 19:57:27 thesing still there? Oct 23 20:03:19 woglinde: yes Oct 23 20:04:25 khem: how exactly is it broken for you? Oct 23 20:05:21 thesing still needing the infos? Oct 23 20:06:07 yes. Oct 23 20:06:19 a patch would be best. Oct 23 20:06:39 pb_: It does not forward emails to my gmail account Oct 23 20:06:46 khem: do they bounce? Oct 23 20:06:52 or do they just go into a black hole? Oct 23 20:06:54 no idea Oct 23 20:06:56 FYI: truecrypt is really slow on embedded devices :) Oct 23 20:07:12 I tried to send email to khem@openembedded.org or kraj@openemebedd.org Oct 23 20:07:27 and was expecting to recieve on gmail account Oct 23 20:07:34 blackhole Oct 23 20:07:38 is my guess Oct 23 20:08:24 thesing here is what I done for simpad http://pastebin.ca/1640642 Oct 23 20:09:09 neither kraj nor khem seem to exist as accounts on discovery, and there doesn't seem to be any obvious forwarding mechanism in place. Oct 23 20:09:31 did forwarding use to work for you before? Oct 23 20:09:53 no I dont think it ever worke Oct 23 20:09:55 for me Oct 23 20:10:01 woglinde: and gcc 4.4 is used automatically? Oct 23 20:10:13 thesing you have to use micro distro Oct 23 20:10:23 and I only tested with uclibc Oct 23 20:10:32 dontknow if eglibc needs something special Oct 23 20:11:11 * thesing wants to use angstrom and glibc.. Oct 23 20:11:39 thesing, for what machine? Oct 23 20:11:41 then you have to bump gcc Oct 23 20:12:43 thesing: I use micro and minimal with uclibc/eglibc often without issues Oct 23 20:13:00 Crofton: collie Oct 23 20:13:15 I only need to bump gcc an binutils? Oct 23 20:13:20 I'm not sure gcc 4.4 will by you anything Oct 23 20:13:24 Crofton he wants to try out the new armveabi stuff Oct 23 20:13:27 args armv4 Oct 23 20:13:37 and that is in 4.4 not 4.3? Oct 23 20:13:38 it will get eabi working for armv4 Oct 23 20:13:42 ah Oct 23 20:13:56 see the angstrom conf file Oct 23 20:14:03 khem: is it supposed to? I'm not sure that there has ever been a global policy to do mail forwarding for all oe developers. Oct 23 20:14:05 you can set per mahicne overide Oct 23 20:14:17 there is an example in there already for some ppc machines Oct 23 20:14:31 not that I am opposed to setting one up, but that would be a certain amount of administration and is a bit more complex than just fixing a broken forward. Oct 23 20:14:38 pb_: May be Oct 23 20:14:49 I thought tit worked Oct 23 20:15:10 uclibc project is set up that way Oct 23 20:15:15 but its not a must Oct 23 20:17:21 Hmm... same here - but adding a forward on request we can do always. I hope... Oct 23 20:19:57 so binutils to 2.19.51.03 and gcc to 4.4.2 should be enough, right? Oct 23 20:20:24 thesing yes Oct 23 20:20:57 ok. thanks. Oct 23 20:21:10 good luck Oct 23 20:23:11 I hope that I don't need it ;) Oct 23 20:27:36 thesing: generally I would say use binutils 2.20 Oct 23 20:27:51 because thumb interworking is fixed in there Oct 23 20:28:06 it was in flux till 2.20 was released Oct 23 20:29:16 khem: thanks. I didn't see that there was a .20 version... Oct 23 20:29:32 kehm how thumb interworking is working with armv4? Oct 23 20:29:53 you dont need it on armv4 Oct 23 20:30:03 khem, in your patch to build.py for brevity: s/os.path.isfile(logfile) is True/os.path.isfile(logfile)/ Oct 23 20:30:47 yes I thought of it but then I thought its more readable that way :) Oct 23 20:31:30 but prolly what you suggest is equally readable Oct 23 20:31:46 khem, that would be the only "if .*is True" in the whole thing, but of course i don't really care Oct 23 20:32:52 one can argue about programming preferences all day long.. Oct 23 20:33:57 blindvt: suggest it on ml and RP will make correction Oct 23 20:34:01 if he wishes to Oct 23 20:35:51 * blindvt doesn't care enough ATM Oct 23 20:35:54 tr ant Oct 23 20:36:55 khem, let me first run one complete fetchall for a start ;) Oct 23 20:38:29 03Frederik Sdun  07shr/merge * r2387ab0fab 10openembedded.git/conf/checksums.ini: Oct 23 20:38:29 Update checksums for devmem2.c mirror Oct 23 20:38:29 Signed-off-by: Frederik 'playya' Sdun Oct 23 20:38:29 Signed-off-by: Klaus Kurzmann Oct 23 20:38:40 03Frederik Sdun  07shr/merge * r490be921dd 10openembedded.git/recipes/freesmartphone/libfsoresource_git.bb: Oct 23 20:38:40 Add new recipe libfsoresource Oct 23 20:38:40 Signed-off-by: Frederik 'playya' Sdun Oct 23 20:38:40 Signed-off-by: Klaus Kurzmann Oct 23 20:41:41 khem: the "is True" thing is known to be unpythonic. i've read multiple python best practices documents on it, and i happen to agree. it's completely unnecessary. python will do it for you, if necessary calling __nonzero__ on the object Oct 23 20:41:42 has anyone used truecrypt on OE device? I got it installed but its soooooo slow Oct 23 20:41:56 is there a way to speed up truecrypt somehow for embedded devices? Oct 23 20:42:04 * kergoth grumbles, need motivation Oct 23 20:42:52 kergoth is a kick in the ace enough? Oct 23 20:42:54 args Oct 23 20:42:55 ass Oct 23 20:43:01 if only.. :) Oct 23 20:44:37 woglinde: how was ubifs on simpad? Oct 23 20:44:47 thesing couldnt try Oct 23 20:45:03 because couldnt make a parameters partition Oct 23 20:45:24 I'm not sure if it's worth looking at for such small flash sizes. Oct 23 20:45:45 hm? Oct 23 20:45:59 its handling writing stuff better than jffs2 Oct 23 20:46:03 and its faster Oct 23 20:46:17 isn't it basically lvm for flash? Oct 23 20:46:59 n8 Oct 23 20:54:27 do we have sdk packages for download somewhere? Oct 23 20:55:48 hm angstroem maybee Oct 23 21:07:32 woglinde: after all that talk about stripping and static libs I fear I could have done a bad commit :/ Oct 23 21:07:57 hopefully not ... ^_^ Oct 23 21:08:06 would you give a quick look? http://cgit.openembedded.org/cgit.cgi/openembedded/commit/?id=3e9bb4f23c205dde9ab1b042b4bfc2c1b5fe8243 Oct 23 21:09:54 was: 'non dev contains .so, klibc, /work/c7x0-angstrom-linux-gnueabi/klibc-1.5.15-r2/install/klibc/lib/klibc.soì Oct 23 21:10:33 ant hms Oct 23 21:10:50 -dev and -dbg packages are there by default Oct 23 21:11:49 try PACKAGES += "${PN}-utils-sh ${PN}-utils-kinit" Oct 23 21:12:52 does anyone know a way to speed up a process running on angstrom? I compiled something for angstrom but it's way too slow. Oct 23 21:12:59 hm hm Oct 23 21:13:07 Like is there a process priority thing somewhere? Oct 23 21:13:13 MajorMonkey its nor angstroem Oct 23 21:13:15 its bitbake Oct 23 21:13:26 there are tow vars which can speed up Oct 23 21:13:30 two Oct 23 21:13:40 oh Oct 23 21:13:42 what are they Oct 23 21:13:56 http://wiki.openembedded.net/index.php/Advanced_configuration Oct 23 21:13:57 optimization flags Oct 23 21:14:24 PARALLEL_MAKE is not mentioned Oct 23 21:14:30 but is in local.conf.sample Oct 23 21:14:58 are we talking about buildtime or runtime? Oct 23 21:15:18 ant hm let the package as it is now Oct 23 21:15:26 I think libc is different Oct 23 21:15:34 woglinde: the QA has gone Oct 23 21:15:55 but iirc as you point out, pb__ said that's ok for a *libc Oct 23 21:16:38 and I could still buyild some (statical) bins against klibc after that patch Oct 23 21:17:08 actually I've rebuilt all from scratch last nite.... Oct 23 21:19:28 kergoth: yes I usually put it for my own sanity because python can return anything Oct 23 21:19:42 but then I should get rid of them from final code Oct 23 21:29:45 khem: what do you mean? all you're doing is adding unnecessary limitations. there's no reason you should have to care whether it returns true or not, only whether what it returns evaluates to true in that context Oct 23 21:32:10 yes as I said for my own sanity Oct 23 21:32:21 it need not be there in the end Oct 23 21:32:48 but its more comprehendable if I read it when so many things are going on Oct 23 21:33:06 it's similar to people using isinstance() or type() instead of using duck typing. what the thing is doesn't matter, how it behaves is what matters Oct 23 21:33:26 in the end yes Oct 23 21:33:37 bye all Oct 23 21:34:33 thesin didnt wait long enough to see me wave back Oct 23 21:43:35 khem: by your next *libc roundup, would you pls check the sanity of klibc recipes? Oct 23 21:52:02 ant__: I will try Oct 23 21:52:14 tia Oct 23 22:40:14 blindvt, khem: The version I merged takes account of that anyway ;-0 Oct 23 22:40:16 ;-) Oct 23 22:41:55 ehm..I don't remember offhand...what is "NOTE: Section = Base or Libs. Target won't be installed automatically." Oct 23 22:42:06 I'll have to grep... Oct 23 22:42:08 hi! Oct 23 22:42:20 good night to everyone! Oct 23 22:49:48 I'm trying to get a decent environment for my media player (mips), where i can run my linux programs. Althoug, I've figured out, that the best way to acomplish this, is to have an chroot setup Oct 23 22:50:51 I have the chroot binary (the player comes with busybox), do I need any other dependencies? Oct 23 23:00:55 kanniball: Do you have target hardware Oct 23 23:01:11 or are you going to use simulator like qemu or some such Oct 23 23:04:10 khem: I have the hardware Oct 23 23:15:03 ok then you could use openembedded to create a rfs to boot it Oct 23 23:15:36 that would be ideal Oct 23 23:15:52 since the root filesystem is read-only, I would not like to mess with it... Oct 23 23:15:53 do you know if the hardware you have it open enough Oct 23 23:16:35 Well one way would be that you use SDK Oct 23 23:16:37 from oe Oct 23 23:16:53 to build your applications Oct 23 23:17:01 and then run them on this pre build rfs Oct 23 23:17:11 it's not that open, because the media player components are closed source... Oct 23 23:17:23 hmm Oct 23 23:17:37 Can you scp say helloworld to it and run it Oct 23 23:18:14 khem: I've already compiled unrar and runned it ;) (I've used an 3rd party toolchain to build it) Oct 23 23:18:32 kanniball: ok you could do same with OE Oct 23 23:18:47 I would say generate SDK and use it Oct 23 23:20:26 generate the SDK is the same as building a target like console-image ? Oct 23 23:20:34 yes Oct 23 23:20:50 good... Oct 23 23:23:44 bitbake meta-toolchain Oct 23 23:23:48 another question, I've created a new machine it don't have anything special, is a copy of other mips machine... but with the right kernel version the system uses and ext3 Oct 23 23:23:54 http://gist.github.com/217264 Oct 23 23:24:34 I've read that we "users" should not create machines... but the kernel version don't make diference? Oct 23 23:54:28 well if your machine is new then you need to create Oct 23 23:54:45 but as it seems you are just needing a sdk for your already populated box Oct 23 23:55:00 so I would just make sure that you have right arch (mips or mipsel) Oct 23 23:55:03 and go Oct 23 23:55:12 with any machine that uses this arch Oct 23 23:55:48 ah ok, the cpu is MIPS 24K V7.8 (i think its a mipsel) Oct 23 23:55:58 ok Oct 23 23:57:31 then you could use db1200 or lsmipsel e.g. Oct 23 23:57:38 for MACHINE Oct 23 23:58:30 if you wanted a fully blown distribution or machine specific things like kernel tweaks which are specific to your machine then you should have machine.conf Oct 23 23:59:13 ok, so I don't need a new machine... I just want to run some programs in my player... Oct 24 00:00:07 yes Oct 24 00:00:11 chose db1200 Oct 24 00:00:19 should be good enough Oct 24 00:00:40 make sure that you chose same C library thats on your target box Oct 24 00:00:46 if it is uclibc or glibc Oct 24 00:01:13 it's uclibc Oct 24 00:01:15 say if its uclibc 'minimal-uclibc' for DISTRO otherwise 'minimal' Oct 24 00:07:53 ok Oct 24 00:09:00 thanks a lot for your attention!!! Oct 24 00:13:01 np have fun Oct 24 01:43:11 i have found gpio.c in temp i want to compile as a module how do i do it/ **** ENDING LOGGING AT Sat Oct 24 02:59:56 2009