**** BEGIN LOGGING AT Mon Aug 15 02:59:57 2011 Aug 15 08:41:15 morning all Aug 15 08:43:55 03Joshua Lock  07master * rbca739cdf5 10bitbake.git/lib/bb/ui/crumbs/tasklistmodel.py: Aug 15 08:43:55 bb/ui/crumbs/tasklistmodel: simplify conditional in include_item Aug 15 08:43:55 No need to check if the name ends with -native or -cross as path will be Aug 15 08:43:55 None in this instance. Aug 15 08:43:55 Signed-off-by: Joshua Lock Aug 15 08:43:57 03Denys Dmytriyenko  07master * r0175fb834e 10bitbake.git/lib/bb/providers.py: (log message trimmed) Aug 15 08:43:57 lib/bb/providers: make "checking PREFERRED_PROVIDER_%s" a debug message Aug 15 08:43:57 In verbose mode there are hundreds of these "checking PREFERRED_PROVIDER_%s" Aug 15 08:43:58 messages, cluttering the output and obscuring the more important resulting Aug 15 08:43:58 "selecting %s to satisfy runtime %s due to %s" messages. Individual "checking" Aug 15 08:43:59 lines are more suited for debug mode, similar to "sorted providers for %s Aug 15 08:44:00 are: %s", hence convert logger.verbose() to logger.debug(). Aug 15 08:44:00 03Richard Purdie  07master * r6f624e399e 10bitbake.git/ (7 files in 3 dirs): (log message trimmed) Aug 15 08:44:00 bitbake/logging: Overhaul internal logging process Aug 15 08:44:01 At the moment it bugs me a lot that we only have one effective logging Aug 15 08:44:01 level for bitbake, despite the logging module having provision to do Aug 15 08:44:14 makes the msg code overly and needlessly complex. Aug 15 08:44:14 It also removes the bb.msg.debug/note/warn/plain/error functions since Aug 15 08:44:14 (47 lines omitted) Aug 15 09:20:06 hi all Aug 15 09:28:53 hi pb_ Aug 15 11:39:29 has something changed in bitbake's handling of native dependencies? Aug 15 11:40:13 my bitbake completely ignores DEPENDS_virtclass-native Aug 15 11:41:41 which seems to be where every recipe puts its native dependencies but which bitbake just ignores Aug 15 11:42:14 03Richard Purdie  07master * ra1980fc89e 10bitbake.git/lib/bb/build.py: Aug 15 11:42:14 build.py: Fix cut and paste error Aug 15 11:42:14 Signed-off-by: Richard Purdie Aug 15 12:47:39 I have a patch to add support for my machine Aug 15 12:48:12 the patch includes a new patch file for a kernel recipe Aug 15 12:48:40 using "git am" gives whitespace errors because of the formatting of the inner patch file Aug 15 12:49:36 (not because of whitespace problems in the inner patch, just the formatting of it) Aug 15 12:50:35 last time I submitted an oe-related patch, the whitespace errors were a problem Aug 15 12:50:47 should I submit the patch? Aug 15 13:25:54 if I ask a question in the #oe IRC channel, will it get answered? Aug 15 13:46:30 rah: if someone's around that knows the answer, sure... Aug 15 13:47:13 rah: re whitespace "errors", frankly I can't imagine why anyone would really care enough to reject a patch based solely upon that Aug 15 13:47:53 ah Aug 15 13:47:56 bluelightning: ok Aug 15 13:48:18 bluelightning: and thank you for responding :-) Aug 15 13:48:23 no worries :) Aug 15 13:48:39 re your native issue, not sure... if it's behaving as you suggest it sounds like something's broken Aug 15 13:49:37 I wonder, what does bitbake -e recipename | grep DEPENDS show? Aug 15 13:52:24 # DEPENDS= Aug 15 13:52:42 # DEPENDS_virtclass-native=None Aug 15 13:54:27 http://codepad.org/opNFcX0c Aug 15 13:54:32 that's the complete output Aug 15 14:00:32 I think something is broken Aug 15 14:01:16 hmm :/ Aug 15 14:19:45 rah: seems to be working here with both angstrom and poky... Aug 15 14:20:06 DEPENDS="pkgconfig-native autoconf-native automake-native help2man-native libtool-native gnu-config-native util-macros-native util-macros-native xtrans-native libxdmcp-native libxau-native bigreqsproto-native xproto-native xextproto-native xcmiscproto-native xf86bigfontproto-native kbproto-native inputproto-native libxcb-native xproto-native" Aug 15 14:20:13 hmm Aug 15 14:20:22 I'm using SHR Aug 15 14:20:42 I've also created my own machine configuration Aug 15 14:21:02 would the machine configuration make a difference? Aug 15 14:21:29 shouldn't... if it was playing with OVERRIDES or similar then it might, but I doubt it would be Aug 15 14:22:27 one way to check would be to switch the machine to something else and then try bitbake -e again Aug 15 15:06:17 hi all Aug 15 15:18:25 hi lumag Aug 15 15:19:07 bluelightning, hi. I've seen some mail regarding meta-opie. Nice to hear! Aug 15 15:19:45 lumag: yes, would like to get this sysroot issue nailed, causes some breakage with quicklaunched apps Aug 15 15:19:59 other than that it works pretty well :) Aug 15 15:21:00 :) Aug 15 15:38:07 | Variable 'libdir' not defined in '/home/otavio/hacking/el/tmp-eglibc-eglibc/sysroots/ossystems-x86//usr/lib/pkgconfig/xcb-proto.pc' Aug 15 15:38:10 | Aug 15 15:38:11 humm Aug 15 15:38:13 | Consider adjusting the PKG_CONFIG_PATH environment variable if you Aug 15 15:38:34 | ERROR: Function 'do_configure' failed (see /home/otavio/hacking/el/tmp-eglibc-eglibc/work/i586-oe-linux/libxcb-1.7-r1/temp/log.do_configure.32511 for further information) Aug 15 15:46:13 wow, webkit sure does take a long time to clone Aug 15 15:56:58 http://tech.slashdot.org/submission/1756570/Google-to-acquire-Motorola-Mobility-for-125-bill Aug 15 15:57:03 interesting Aug 15 15:59:53 agreed; it's downright fascinating that slashdot is still around ;) Aug 15 16:27:03 * pb_ stabs pkg-config Aug 15 16:27:05 and cmake Aug 15 16:27:08 morning kergoth Aug 15 16:27:29 hey Aug 15 16:33:33 fontconfig.pc:Libs.private: -lexpat -L{libdir} -lfreetype -liconv Aug 15 16:33:35 hah Aug 15 16:34:09 03Richard Purdie  07master * rd5abdacaf9 10bitbake.git/lib/bb/ (msg.py runqueue.py taskdata.py ui/ncurses.py): Aug 15 16:34:09 Fixup remaining bb.msg.domain users Aug 15 16:34:09 Signed-off-by: Richard Purdie Aug 15 16:41:34 does anybody know where fontconfig's "fix-pkgconfig.patch" came from? Aug 15 16:41:47 it seems to be somewhat inaccurately named since, as far as I can tell, it serves only to break the pkgconfig data. Aug 15 17:20:06 Do someone know about libxcb being broken? Aug 15 17:21:25 otavio, sent a patch while ago. Aug 15 17:21:50 lumag: is it somewhere I can cherry-pick it? Aug 15 17:22:00 oe-core ML Aug 15 17:22:06 lumag: or can you upload it somewhere I can download it and apply? Aug 15 17:22:10 * otavio checks Aug 15 17:22:37 It's easy so you can cherry-pick it manually. Aug 15 17:23:07 lumag: which subject? Aug 15 17:24:08 search for libxcb please Aug 15 17:27:19 lumag: I did but it is for native and nativesdk Aug 15 17:27:29 lumag: I don't think this is my issue Aug 15 17:28:33 otavio, strange, what's your issue? Aug 15 17:29:19 | Variable 'libdir' not defined in '/home/otavio/hacking/el/tmp-eglibc-eglibc/sysroots/ossystems-x86//usr/lib/pkgconfig/xcb-proto.pc' Aug 15 17:29:29 ERROR: Task 6 (/home/otavio/hacking/el/meta-openembedded/meta-oe/recipes-graphics/xcb/libxcb_1.7.bb, do_configure) failed with exit code '1' Aug 15 17:31:32 Could you show that xcb-proto.pc? Aug 15 17:31:56 otavio, because mine contains definition of libdir Aug 15 17:31:58 lumag: it really lack libdir Aug 15 17:32:34 otavio, you can try rebuilding xcb-proto (bitbake -c clean xcb-proto, bitbake everything_as_usual) Aug 15 17:32:39 ahhhhhhh Aug 15 17:32:44 lumag: meta-oe has it redefined Aug 15 17:34:50 otavio, strangely enough, Koen had removed xcb-proto & libxcb from meta-oe. But I see no real diffs between xcb-proto in meta-oe and oe-core Aug 15 17:35:39 lumag: this is what was missing Aug 15 17:35:43 lumag: just updated meta-oe Aug 15 17:38:18 lumag: now added your patch and seems to be building fine :-) Aug 15 17:38:30 otavio :) Aug 15 18:02:41 well now the meta-oe recipe should just be bbappend Aug 15 18:15:52 khem: yes; indeed Aug 15 18:52:32 hi, I can't build gst-plugins-ugly recipe in openembedded-core Aug 15 18:52:35 I get this -> ERROR: gst-plugins-ugly was skipped: because it requires commercial license to ship product Aug 15 18:52:44 what do I need to do to be able to build it? Aug 15 18:55:41 pespin: set COMMERCIAL_LICENSE = "" in your local.conf Aug 15 18:59:37 Evening! Aug 15 19:00:53 lumag: hi :) Aug 15 19:01:13 After recent openssl version bump I had to rebuild python-native, git-native and rpm-native (missing OPENSSL_0.9.8 versions) Aug 15 19:02:36 Hi Jay7, how is it going? Aug 15 19:05:16 lumag: slowly :) Aug 15 19:05:21 lumag: bitbake did not notice it ? Aug 15 19:05:51 * Jay7 is building virtualization solutions for some soho-customers Aug 15 19:12:22 khem: no Aug 15 19:41:10 khem, thanks it worked :) Aug 15 20:37:54 gdk-pixbuf-native failed, couldn't find libselinux Aug 15 20:40:18 Ashburn +1ed it Aug 15 20:40:20 ant__, here? Aug 15 20:40:25 Oops - thanks tmux Aug 15 20:41:39 CosmicPenguin: have you seen refresh issues with tmux? in particular, a few times now i've seen irssi and vim get display glitches like they aren't refreshing fully. not sure if its a recent bug from trunk or what Aug 15 20:41:50 Haven't really Aug 15 20:41:57 Mouse doesn't always work great, but other then that Aug 15 20:42:30 k Aug 15 21:03:31 lumag: hi Aug 15 21:08:54 ant__, Jay7, tested kexecboot on tosa. got wrong logo size on boot, blinking screen (probably due to COMADJ mismatch, have to test) and non-working keys :) Aug 15 21:09:42 logo size is fixed Aug 15 21:09:55 at least this... Aug 15 21:11:00 lumag: same things I've reported to you ;) Aug 15 21:11:12 :) Aug 15 21:11:17 lumag: I'm testing bare oe-core + meta-zaurus today Aug 15 21:11:33 apart missing /boot/zImage symlink Aug 15 21:11:34 screen have low brightness but sometimes backlight is switching to max :) Aug 15 21:11:39 Haven't tried with serial cable yet. Aug 15 21:11:46 as I understand at least :) Aug 15 21:11:53 Jay7, for me it's blinking ultra fast Aug 15 21:12:05 core-image-sato is booting on poodle and ctx0 Aug 15 21:12:15 but touchscreen is broken Aug 15 21:12:16 and I've seen orange vertical lines Aug 15 21:12:20 (kdrive) Aug 15 21:12:28 can't remember which version it was.. Aug 15 21:12:40 lumag: built with gcc4.6 Aug 15 21:12:49 kernel still boots Aug 15 21:12:53 last time it was because of 8-bit ram to video ram transfers Aug 15 21:13:14 but now we are using 32-bit transfers (if compiler does not break this) Aug 15 21:13:29 Jay7, good! Aug 15 21:13:39 and yes, no touchscreen Aug 15 21:13:50 I've not investigated yet Aug 15 21:14:05 ant__, I have built zImage-kexecboot-2.6.39-r41.2-tosa.bin. Is it the latest one? Aug 15 21:14:49 I still have .3 Aug 15 21:16:37 lumag: any news about locomo kernel patch? Just that one? Aug 15 21:17:50 ant__, no news unfortunately. Overwhelmed with my work, breakage of my buildserver and tending kids :/ Aug 15 21:18:14 oh, crazy days ! Aug 15 21:18:54 I tested just the old collie_keymap patch and the 'missing' keys are now found, even by kexecboot Aug 15 21:19:18 this seems indeed necessary for both machines Aug 15 21:26:09 ant__, I have .2 in linux-kexecboot_2.6.39.bb Do I miss something? Aug 15 21:27:18 http://git.shr-project.org/git/?p=meta-smartphone.git;a=commit;h=a001d4de6c0188a77356fa4ec839e28f6ef5977c Aug 15 21:27:34 ant__: so with old keymap patch poodle and collie have usable keyboards? Aug 15 21:28:11 ant__, that's an r4 in main kernel, not kexecboot one Aug 15 21:28:40 ah, ok, I meant this one Aug 15 21:28:41 Ah, I see the fix for logo size. Aug 15 21:28:46 btw add file://locomo_kbd_tweak-r2.patch Aug 15 21:30:00 lumag: there is a pending patch: http://paste.debian.net/126338/ Aug 15 21:30:07 we are overtaken channel with our zaurus + kexecboot talk ;) Aug 15 21:30:14 not only Aug 15 21:30:35 as far as I can say, Angstrom-scripts lead to very broken rootfs's Aug 15 21:30:52 now, testing oe-core + our meta-zaurus layer is only a bit better :/ Aug 15 21:30:55 ant__, Jay7, should we move to any #zaurus or #kexecboot channel? Aug 15 21:31:04 let's go on the latter Aug 15 21:31:12 we are reusing #kexecboot frequently Aug 15 21:31:20 for zaurus-related talks :) Aug 15 21:31:26 ok :) Aug 15 21:31:51 for broken rootfs, please show some successful logs ;) Aug 15 22:07:16 hi, I updated lightmediascanner and enjoy recipes to build with oe-core, but I get this autofoo error :S http://paste.pocoo.org/show/458965/ Aug 15 22:07:26 it builds fine outside of OE though Aug 16 02:57:13 hello? Aug 16 02:57:59 i know this isnt for distro help, but #cacko is empty, and the links i found for c700 cacko rom is dead **** ENDING LOGGING AT Tue Aug 16 02:59:57 2011