**** BEGIN LOGGING AT Mon May 10 02:59:57 2010 May 10 04:07:00 ALL, who is the maintainer of wxWidgets in OE? is he around? how can he be contacted? May 10 04:53:01 IgorK: some packages dont have dedicated maintainers May 10 04:53:13 did CXXFLAGS work for you May 10 05:09:25 khem: yes, wxWidgets and my program compiled fine May 10 05:16:00 IgorK: ok May 10 05:16:21 then you are set May 10 05:16:26 have fun May 10 05:16:57 wxwidget needs to be fixed May 10 05:17:12 I will see if i have time May 10 05:17:34 cxxflags is a workaround May 10 05:17:47 you can use it for now May 10 05:19:06 khem: According to this: http://patchwork.openembedded.org/patch/1401/ wxWidgets-2.9.0 currently in the OE tree, but it still does not have this line... May 10 05:20:21 thanx 4 the help, now I will c if the program starts... ;-) May 10 05:27:42 as I said May 10 05:27:50 this is not a proper fix May 10 05:27:57 only a workaround May 10 06:39:34 good morning May 10 08:05:31 good Morning May 10 08:06:04 do you know, that the webserver is still sleeping? May 10 08:35:05 good morning May 10 08:41:56 git server also looks to be down May 10 08:43:28 gm florian and all May 10 08:43:59 hi mckoan May 10 08:46:06 hey mckoan May 10 08:51:43 hi, http://cgit.openembedded & dgit://git.openembedded.org seems down May 10 08:52:54 trolls ate the cables again :-) May 10 08:53:53 luckilly there are mirrors of git May 10 08:54:14 and the trolls also ate patchwork's patches since 05/05 ;-) May 10 08:55:48 thats what we get for hosting in troll country May 10 08:55:51 XorA: what are the git mirrors's URL ? May 10 08:55:59 ericben: see channel topic May 10 08:56:10 XorA: good idea thanks ;-) May 10 09:16:16 morning May 10 09:20:48 hey mickey|away May 10 09:20:52 hey mickey|office May 10 09:20:58 Hi XorA May 10 09:21:04 hey Brokie May 10 09:21:19 mickey|office: your not by any chance going to UDS? May 10 09:21:45 My employer disappeared today May 10 09:21:48 ... wierd May 10 09:21:53 dissapeared?? May 10 09:21:56 ya May 10 09:21:58 as in bankrupt? May 10 09:22:00 poof May 10 09:22:02 no May 10 09:22:08 as in nowhere to be found May 10 09:22:17 * XorA hopes you got the money first May 10 09:22:21 lol May 10 09:22:32 i got the money thursday May 10 09:22:42 then he disappeared May 10 09:22:57 mafia finally found him :-) May 10 09:23:01 lol May 10 09:23:03 nah May 10 09:23:35 is there a bitbake target for the Tegra250 yet? May 10 09:23:47 or is that something I would have to do? May 10 09:23:54 I have 3 here May 10 09:23:58 wanna get started May 10 09:24:27 some people muttered about tegras but I dont think Ive seen anything go in yet May 10 09:24:47 broonie might know, he has probably come across them for audio work May 10 09:25:13 ya, they are good for audio processing, given they have in and out built in DAC and ADC May 10 09:26:06 and at the office there are codecs for 1080p H.264 AVC May 10 09:26:21 XorA: no, but hrw|gone is going. May 10 09:26:42 mickey|office: yeah already arrange to beer with him May 10 09:26:48 cool May 10 09:26:51 where is UDS? May 10 09:26:52 greet him from me May 10 09:26:57 Brokie: Brussels May 10 09:27:00 oi May 10 09:27:09 House arrest will prevent my attendance... May 10 09:27:23 you havent hacked your tag yet :-D May 10 09:27:34 yes, but she comes now and then... May 10 09:27:59 would be hard to explain the dog wearing my bracelet May 10 09:28:55 broonie: The nVidia BSPs are totally non-mainline. May 10 09:29:18 for Windows Mobile 7? May 10 09:29:20 Brokie: ^^^ tab completion is great. May 10 09:29:22 yes May 10 09:29:48 Do you mean for Windows Mobile 7? May 10 09:29:57 No, their Linux code. May 10 09:30:16 I dont comprehend your statement May 10 09:31:15 treat me like I'm 7, with terms like "mainline" and "BSP" and "Apple sucks" May 10 09:31:49 The code which nVidia provides for their processors does not look like Linux code. May 10 09:32:01 ... May 10 09:32:11 what does it look like? May 10 09:32:15 BSD? May 10 09:32:19 Their own creation. May 10 09:32:31 is it based on Linux? May 10 09:32:39 is the source avail? May 10 09:32:51 I have the Avalon SDK with WinMo 7 May 10 09:32:54 Yes, their Linux kernel code is available. May 10 09:33:11 At least some has been published, anyway. May 10 09:33:23 but it is, as I understand, not on par with the tarvalds kernel sources? May 10 09:34:18 has the chip been mapped out yet? May 10 09:34:42 so conversion of mainline sources can be poerted half with some ease? May 10 09:34:56 s/poerted/ported May 10 09:36:54 i see it May 10 09:36:58 wow May 10 09:37:20 whatsup with that bootloader? May 10 09:38:42 hi Brokie, don't know if that helsp but tegra2 git is available here for android : http://android.git.kernel.org/?p=kernel/tegra.git;a=summary May 10 09:41:38 i found all nv's gits, I got a 10 page book of them with my SDK May 10 09:41:42 but thanks May 10 09:51:32 Seems funny this thing runs the goldfish kernel with only 2 modificqations May 10 10:21:29 Tartarus: I've rebuild for calamari from scratch without problem (with TARGET_OS = "linux-gnuspe" May 10 10:21:29 ) May 10 11:52:43 morning May 10 12:08:04 bitbake nano fails for me with cp: cannot stat `/home/soltys/oe/ngw100/tmp/sysroots/avr32-angstrom-linux-uclibc/usr/share/gettext/config.rpath': No such file or directory May 10 12:08:20 is there any known solution for that ? May 10 12:10:05 (bitbake gettext fails with the same error on one of dependencies..) May 10 12:10:06 try 'bitbake gettext' first May 10 12:10:08 ok May 10 12:15:06 anyway if it depends on gettext it should pull it as dependency.. May 10 12:52:11 is git down? May 10 12:52:33 Crofton|work: yes May 10 12:52:35 Crofton|work: it would appear so from peoples comments all day May 10 12:52:43 * XorA isnt doing OE today May 10 12:57:34 i built a my-own-customized-image.bb, see it here http://pastebin.com/pTRGNbvv May 10 12:57:48 It has a size of 64 Mb, can I reduce it somehow?? May 10 12:58:03 by changing some params?? May 10 13:11:23 removing al the 'IMAGE_INSTALL +=' will help a lot ... :p May 10 13:12:01 but that's probably not wat you want to hear from me May 10 13:25:56 anyone using mono? I have updated mono in OE to version 2.6.3 and fixed the build issues with it and would like someone else to try it before I commit it. May 10 13:27:33 this sucks.. nano whants gettext but gettext wants libiconv wchith fails becouse there is no gettext oO May 10 13:46:21 soltys, hi,you've got commit access? May 10 14:13:31 GNUtoo: hi, nope May 10 14:14:02 GNUtoo: I'm new to this May 10 14:14:21 GNUtoo: and finaly I've got linux buildhost not bsd one ;) May 10 14:14:25 soltys, ah ok so what did you meant by " before I commit it.",did you meant before I send it to mailing list? May 10 14:14:37 ah ok nice May 10 14:15:13 GNUtoo: I didn't wrote anything about commiting but svolpe did ;) May 10 14:15:50 ahhh sorry May 10 14:16:13 again I read too fast + autocompletion of names May 10 14:16:27 soltys, do you use uclibc? May 10 14:17:10 GNUtoo: yes May 10 14:17:19 ah ok May 10 14:17:43 mmm I don't know....sorry May 10 14:17:56 but you could check if there is inherit gettext May 10 14:18:01 in nano May 10 14:19:32 no only autotools May 10 14:19:54 try inherit gettext then, it can't hurt May 10 14:20:22 ok May 10 14:22:17 it stil fails May 10 14:22:52 libiconv pulled by gettext fails becouse it can't find gettext.. May 10 14:23:43 is your gettext error similar to this: configure.in:18: error: possibly undefined macro: AM_GLIB_GNU_GETTEXT ? May 10 14:25:42 rhc: no, nano and libiconv (pulled by gettext) fail with `/home/soltys/oe/ngw100/tmp/sysroots/avr32-angstrom-linux-uclibc/usr/share/gettext/config.rpath': No such file or directory May 10 14:43:49 cbrake, you about? May 10 14:44:42 thanks for fixing git May 10 14:44:42 03Martin Jansa  07org.openembedded.dev * rbf9d1fad96 10openembedded.git/recipes/emacs/ (emacs-x11_23.1.bb emacs.inc files/emacs23.1-use-qemu.patch): (log message trimmed) May 10 14:44:42 emacs: add missing chunk in use-qemu.patch for 23.1, create libc.so -> libc.so.6 link May 10 14:44:42 * prefix-args were built with target compiler but later executed in host May 10 14:44:42 environment, older emacs versions used prefix-args from qemu, but this May 10 14:44:43 part was missing in emacs23.1-use-qemu.patch May 10 14:44:43 * Paul Fertser suggested that prefix-args should be built with host May 10 14:44:44 compiler instead and then without need to use qemu-arm May 10 14:44:45 03Martin Jansa  07org.openembedded.dev * ra8d24bab84 10openembedded.git/ (2 files in 2 dirs): May 10 14:44:45 task-shr-feed: add emacs-x11 and set P_V to 23.1 May 10 14:44:46 Signed-off-by: Martin Jansa May 10 14:45:17 ka6sox: yes May 10 14:45:23 OOM May 10 14:45:30 ka6sox: ohh May 10 14:46:24 a python process seems to have gotten stuck (bitbake?) May 10 14:47:19 03Michael 'Mickey' Lauer  07org.openembedded.dev * r651255194b 10openembedded.git/recipes/vala/ (vala.inc vala_0.8.0.128-1150.bb vala_0.8.1.1.bb): vala: 0.8.0.128 -> 0.8.1.1 May 10 14:47:19 ka6sox: does not seem like we should be running bitbake on this machine May 10 14:47:50 stucked bitbake in do_setscene while rebuilding something seems quite common now (no idea why) May 10 14:52:50 JaMa: there is an extra lockfile call May 10 14:53:40 https://www.cvg.de/people/ensc/bb-lock.patch May 10 14:54:03 ensc|w: so it's just waiting for something in multi BB_NUMBER_THREADS or is it posible deadlock? May 10 14:54:47 ah thanks May 10 14:54:54 no, it is locked already in the same method May 10 14:55:20 some python versions might close 'lf' so this deadlock was unnoticed May 10 14:56:40 sigh... writing useful commit messages takes more time than developing the patch itself :( May 10 14:57:14 and after -c clean it doesn't happen (not sure why, looking in that bbclass code) May 10 15:02:00 mickey|office: is it possible your new vala fso release changed checksum? May 10 15:02:07 ka6sox: thanks for fixing melo May 10 15:03:03 mickey|office: mine are SRC_URI[md5sum] = "5135c4ea472269de59121b3380bcf1cd" May 10 15:03:03 SRC_URI[sha256sum] = "07fcd4140ef24dc9977aef89eab023e9e2814be4baee5e210dcdce64d084743e" May 10 15:11:32 cbrake, yw...glad I could fix it easily May 10 15:14:19 stupid gettext ;< May 10 15:17:44 khem, any news about the gdb-cross problem? It still makes meta-toolchain unbuildable. May 10 15:18:38 JaMa|Off: possible, I updated it 2 minutes after initial upload May 10 15:18:43 but before i pushed to OE May 10 15:18:54 if you happen to have downloaded it exactly then it would be differing May 10 15:27:22 03Koen Kooi  07org.openembedded.dev * rad7c0216a5 10openembedded.git/recipes/ccache/ccache_2.4.bb: ccache: fix QA error May 10 15:27:35 03Koen Kooi  07org.openembedded.dev * racd000615c 10openembedded.git/recipes/qt4/ (files/qte.sh qt-4.6.2.inc qt4-embedded.inc): May 10 15:27:35 qt4-embedded: fix decorations and mouse configs May 10 15:27:35 Imported from http://arago-project.org/git/?p=arago.git;a=commitdiff;h=0fe5a365d34a07f93861741e25c5cfaaf662aa1c May 10 15:27:35 Original patch done by Denys Dmytriyenko May 10 15:27:36 03Koen Kooi  07org.openembedded.dev * r1a733aa29a 10openembedded.git/recipes/linux/ (66 files in 4 dirs): linux-omap-psp 2.6.32: rediff touchbook and beagleboard patches, tweak beagleboard defconfig May 10 15:30:33 hi hrw - new job brings travelling, eh? May 10 15:30:54 yes May 10 15:31:14 and due to one vulcano I know where to get free wifi on Köln Hbf May 10 15:31:28 hehe May 10 15:31:30 köln May 10 15:31:32 nice May 10 15:32:09 I am still sleepless May 10 15:42:07 hrw-uds: hi, still around? May 10 15:42:52 yep May 10 15:43:05 someone has XorA's jabber id? May 10 15:48:52 hrw-uds: He's most likely on a plane at the minute. May 10 15:49:24 broonie: Koen told that he was online ~20 minutes ago May 10 15:49:38 anyway will catch him later May 10 15:50:16 I need to fix kde battery levels May 10 15:50:36 warning about low level at 15% is a bit too early when 1h of work time left May 10 15:50:38 hrw-uds: Ah, well. Probably texting him will get his interest. May 10 15:51:01 yep May 10 15:57:15 RP: you here? May 10 16:37:38 ao2: I havent yet looked into gdb-cross-sdk issue May 10 17:51:06 I don't have luck to oe or sth is wrong... May 10 17:51:51 I can't build anithing that pulls gettext because it wont build... (it pulls libiconv with fails due to missing gettext...) May 10 17:55:54 is anyone familar with qt embedded? May 10 17:56:07 and yeah, hi all :D btw :D May 10 17:58:21 soltys: stuff that needs gettext inherits the gettext bbclass which pulls in gettext-native, which builds just fine. May 10 17:59:24 kergoth: that is right but cp: cannot stat `/home/soltys/oe/ngw100/tmp/sysroots/avr32-angstrom-linux-uclibc/usr/share/gettext/config.rpath': No such file or directory May 10 18:00:05 then you're buliding a recipe that needs gettext and doesnt inherit gettext. add an inherit of gettext to the recipe and move on May 10 18:00:07 it should look in tmp/systoots/i686-linux/ .. May 10 18:15:48 khem: That optspace patch was to fix the gnutls thing? Or the angstrom thing? May 10 18:18:23 blah there is inherit gettext ;/ May 10 18:44:32 Tartarus: optspace patch is to fix the undefined symbol problem you reported few weeks back May 10 18:49:02 gnutls fails to build on minimal and calamari with that applied, fresh tmpdir May 10 18:50:13 Tartarus: whats the error ? May 10 18:50:33 Tartarus: but you get thru the eglibc buildright ? May 10 18:50:45 Always have on e500v2 May 10 18:50:51 just not e300c3 / ppc603e May 10 18:52:58 yeah why you went thru e500 I explained in the email May 10 18:53:05 the new patch extends over to all ppcs May 10 18:53:16 so now e300c3 / ppc603e should also build May 10 18:53:27 coming to gnutls whats the error message May 10 18:53:40 It's pastebin'd in the email :) May 10 18:53:52 ah email May 10 18:53:54 And that patch doesn't help angstrom either May 10 18:53:56 fyi May 10 18:53:58 i'll try e300 now May 10 18:54:10 I have not intended to fix angstrom May 10 18:54:28 there is magin in sane-toolchains to select right addons for eglibc May 10 18:54:32 and glibc May 10 18:54:47 thats why distro's using sane-toolchain will work with SPE May 10 18:54:52 I am not sure about angstrom May 10 18:55:04 s/magin/magic May 10 19:17:38 Tartarus: gnutls problem is different but similar May 10 19:17:46 Tartarus: I have seen it in past May 10 19:18:52 Tartarus: do you know which DSO is it ? May 10 19:19:07 it seems it is missing -lgcc when linking this DSO May 10 19:21:31 * khem wonders how useful -Os is in ppc context May 10 19:24:51 khem: No idea, sorry May 10 19:25:32 khem: I'll kick another build once I confirm the e300c3 stuff May 10 19:27:23 Tartarus: ok May 10 19:33:34 Hi, ALL May 10 19:34:11 Is there anyway to find out why the program starts as a console application? May 10 19:34:28 on the angstrom kernel May 10 19:35:30 I'm using arm-angstrom-linux-gnueabi-g++ compiler. Is there a switch to compile it this way? May 10 19:41:22 IgorK: do you want to compile kernel with g++ ? May 10 19:45:14 03Michael 'Mickey' Lauer  07org.openembedded.dev * rb8e5860405 10openembedded.git/recipes/vala/vala_0.8.1.1.bb: vala: fix tarball checksums; tarball was corrupt when i generated mine May 10 19:47:55 khem: e300c3 works better now, trying a full image May 10 19:49:56 khem: it is compiled from an OE tree, so I take it is compiled with the cross g++ May 10 19:50:59 I'm talking about my program. I'm using X11+GPE+wxWidgets to compile my program, but it starts as console application May 10 19:52:05 Tartarus: ok nice May 10 20:15:38 qt4 4.6.2 lacks translations. May 10 20:15:40 :-( May 10 20:15:51 this was dropped without even a comment in the commits May 10 20:16:26 and as can be seen in b4e0df20e3724251ae39945a0aa91e6d00078b29 it had users ;-) May 10 20:17:13 morning May 10 20:17:34 XorA|UDS: afternoon ;-) May 10 20:21:36 khem: gnutls also fails on e300c3 :) sec May 10 20:22:32 khem: failed trying to link ex-client-srp May 10 20:24:15 Tartarus: is it same message about linking with hidden symbols from libgcc May 10 20:39:07 otavio: I don't see translations being dropped... May 10 20:39:34 * khem mutters May 10 20:39:42 gcc 4.5 ICEs with ppc May 10 20:45:03 khem, yes, same message I do believe May 10 20:45:03 sec May 10 20:45:46 khem: http://pastebin.com/sa5sYzfG May 10 20:49:25 Tartarus: ok, now if we can find which DSO it is May 10 20:49:28 then we can fix it May 10 20:52:07 I'm getting "CROSS COMPILE Badness: /usr/include in INCLUDEPATH: /usr/include/directfb" trying to build libsdl-directfb May 10 20:53:03 looking at the directfb recipes, it looks like there's a patch that's already supposed to fix that May 10 21:10:24 Zygo: then make sure that patch is being applied when building May 10 21:26:01 khem: I gather you're still lacking in CPU cycles to debug this kind of thing locally? May 10 21:26:22 yes May 10 21:26:46 blarg May 10 21:26:53 Want the whole log.do_compile pastebin'd? May 10 21:27:40 Tartarus: hmm sure May 10 21:28:11 http://pastebin.com/rd2DWBYc May 10 21:28:15 Tartarus: if you can add -Wl,-Map,xx.map to linker commandline May 10 21:28:26 then the mapfile should show which library is asking for this symbol May 10 21:30:38 Tartarus: hmm one thing I see could be the problem May 10 21:30:50 Tartarus: -o .libs/libgnutlsxx.so.26.14.12 May 10 21:31:11 when this link step happens it adds -nostdlib to the link options May 10 21:31:18 and then links with -lgcc_s May 10 21:31:33 if you change that to -lgcc it should work May 10 21:31:45 but before please dump the symbol table of this DSO May 10 21:31:50 and grep for UND May 10 21:32:21 readelf -s .libs/libgnutlsxx.so.26.14.12 | grep _savegpr_31 May 10 21:32:34 That's all crap libtool decided to do May 10 21:32:35 right? May 10 21:32:44 dont know May 10 21:32:56 but somewhere nostdlib is added by build system May 10 21:33:23 115: 00000000 0 NOTYPE GLOBAL DEFAULT UND _savegpr_31 May 10 21:33:23 libtool is less of culprit I guess when we say -nostdlib we essential cut gcc drivers hands off when linking May 10 21:33:29 there you go May 10 21:33:41 similar lines for _30/25/29/28/27 May 10 21:33:48 so the solution is to link with -lgcc instead of -lgcc_s May 10 21:34:00 yeah all of them will then work May 10 21:34:05 Right, I'm saying libtool is what decided what libs to use or not use May 10 21:34:24 or wholesome solution is to disable -Os for gnutls completely for ppc platform May 10 21:34:41 even bigger is to disable -Os for all powerpc's May 10 21:34:51 but that will be like healthcare bill May 10 21:35:13 That feels like a bandaid, yes :) May 10 21:35:25 I guess I don't see why other arches don't blow up with -Os May 10 21:35:27 But ppc does May 10 21:36:15 ok I can explain that May 10 21:36:19 its a gcc thing May 10 21:38:59 gcc generates out of line prologue/epilogue functions when compiling with -Os on ppc May 10 21:39:10 these functions are supposed to be in libgcc.a May 10 21:39:29 because there calling convention is not standard May 10 21:40:55 it worked wrongly thus far because these functions had global visibility which was wrong and caused many different problems which were fixed when these symbols were made hidden May 10 21:41:12 but when these symbols are made hidden then the problems like we see appear May 10 21:42:28 this is ppc ABI thing May 10 21:43:53 so these register storing and restoring functions are emitted once to reduce code size for whole linking unit May 10 21:44:08 Tartarus: thats the reason for these problems :) May 10 21:44:18 I hope it explains a bit for you May 10 21:44:46 a little, yeah, thanks May 10 21:45:09 But now it sounds like we need to make sure that -lgcc or so is passed and that sounds like libtool junk :) May 10 21:45:20 yes May 10 21:45:30 with -Os lgcc is always needed May 10 21:45:46 denix: humm it doesn't has them here May 10 21:46:32 Tartarus: so some packages might do -nostdlib, they will need some love May 10 21:46:45 like gnutls May 10 21:46:48 there might be more May 10 21:47:17 so either we fix those packages or drop optspace for ppc at distro level May 10 21:47:23 or machine level May 10 21:48:52 Well, libtool is what's doing it :) May 10 21:49:02 I see -nostdlib in powerpc-...-libtool May 10 21:49:45 ah May 10 21:49:52 libtool is package specific May 10 21:50:02 so gnutls configure might be interesting May 10 21:50:24 I checked, it's not. Here's the interesting bit tho, sec May 10 21:51:28 http://pastebin.com/3YG710Lf May 10 21:53:02 Tartarus: where is it adding -lgcc_s May 10 21:53:11 thats what we have to trap May 10 21:53:17 denix: has it built qt4-translaton-qt package for you? May 10 21:53:40 k, poking May 10 21:53:53 may be $deplibs or $compiler_flags May 10 21:54:03 postdeps has it May 10 21:54:08 seeing how it's populated May 10 21:54:38 actually libtool is not to blame here, its the way things are May 10 21:54:50 heh May 10 21:54:59 it thinks its good to link with shared ver of libgcc May 10 21:55:11 poor libtool is unaware of ppc ABI May 10 21:55:43 So, what's the fix, exactly? May 10 21:55:49 If we're -Os, we need non-shared libgcc? May 10 21:55:56 Or expose some stuff in shared gcc? May 10 21:56:11 yay...if directfb-config doesn't have /usr/include in cflags, it just arbitrarily adds ${SYSROOT}/usr/include/directfb May 10 21:59:16 ...or...is it supposed to do that? May 10 21:59:46 Tartarus: the right fix is to link libgcc.a instead of libgcc_s for ppc when compiled with -Os May 10 21:59:52 for all packages that is May 10 22:00:10 khem: OK May 10 22:00:55 Tartarus: I am more of opinion to switch to -O2 though May 10 22:01:05 Tartarus: I have started a build here May 10 22:01:07 for gnutls May 10 22:01:20 I should be there in an hour or so hopefully May 10 22:13:01 is there a handy-dandy guide to setting up pkg-config variables for OE cross builds? May 10 22:13:51 e.g. what should a .pc.in look like to get the right include paths for the cross build and for when (if) the .pc ends up on the target? May 10 22:14:22 Zygo: look in one of familiar recipes May 10 22:15:35 Tartarus: postdeps seems to be generated by executing the compiler somehow with -shared option May 10 22:15:53 Tartarus: but gcc always links with -lgcc and lgcc_s as I can see May 10 22:17:09 -lgcc --as-needed -lgcc_s --no-as-needed -lc -lgcc --as-needed -lgcc_s --no-as-needed May 10 22:17:28 I wonder how configure decides to filter out lgcc May 10 22:21:46 gotta run May 10 23:48:50 03Brijesh Singh  07org.openembedded.dev * rb0ccc12f3f 10openembedded.git/recipes/ti/ (matrix-gui/dm365-evm/init matrix-gui_svn.bb): May 10 23:48:50 matrix-gui_svn: add init script for dm365-evm. May 10 23:48:50 * use default Rotation to 0 degrees. May 10 23:48:50 * remove touch screen specific variables. May 10 23:48:50 Signed-off-by: Brijesh Singh May 10 23:48:50 Signed-off-by: Denys Dmytriyenko May 11 00:50:20 sigh...found an oe-devel mailing list post talking about my directfb compile error dated Oct 16, 2006 May 11 00:50:35 well, strictly speaking, directfb compiles fine, it's things that try to import directfb's .pc files that fail May 11 02:20:00 Anyone here know how to build distro with linux 2.4? **** ENDING LOGGING AT Tue May 11 02:59:56 2010