**** BEGIN LOGGING AT Fri Dec 18 02:59:57 2009 Dec 18 05:26:40 03Chris Larson  07org.openembedded.dev * r386409b14f 10openembedded.git/recipes/gettext/ (gettext-native_0.14.1.bb gettext-native_0.17.bb): Dec 18 05:26:40 gettext-native: rename the M4 var to better reflect its contents Dec 18 05:26:40 Signed-off-by: Chris Larson Dec 18 08:16:58 morning Dec 18 08:34:01 good morning Dec 18 08:34:17 hi marco Dec 18 09:27:43 hi florian Dec 18 09:30:19 good morning Dec 18 09:30:36 good morning Dec 18 09:55:46 morning Dec 18 09:56:15 hi hrw Dec 18 10:24:17 03Martin Jansa  07org.openembedded.dev * r9c0a5ca7a8 10openembedded.git/conf/distro/include/sane-srcrevs.inc: sane-srcrevs: update shr-launcher revision for new release Dec 18 10:37:28 is there a way to use a second toolchain in a recipe, to compile, for example, code for a dsp? (omap case) Dec 18 11:10:52 mickey|office: good morning Dec 18 11:10:55 florian: good morning Dec 18 11:12:50 hi pb_ Dec 18 11:13:16 hi pb Dec 18 11:13:29 woglinde: cacao-initial: resolve.c:356: resolve_classref_or_classinfo: Assertion `c->state & 0x0002' failed. Dec 18 11:13:42 woglinde: did you got it anytime? Dec 18 11:13:50 hrw amd64? Dec 18 11:13:54 yes Dec 18 11:14:02 better user jamvm Dec 18 11:14:02 and I know - use jamvm you will tell Dec 18 11:14:08 *g* Dec 18 11:14:13 but jamvm do not work for me at all Dec 18 11:14:24 hm uh? Dec 18 11:14:41 whats the error? Dec 18 11:14:52 will do build and report Dec 18 11:45:39 woglinde: http://hrw.pastebin.ca/1719350 Dec 18 11:46:58 woglinde: http://hrw.pastebin.ca/1719352 is jar content Dec 18 11:47:05 hrw which version is ecj? Dec 18 11:47:31 3.4.1 Dec 18 11:47:45 will build 3.4 Dec 18 11:47:48 yes Dec 18 11:58:41 woglinde: http://hrw.pastebin.ca/1719356 is strace of java command Dec 18 11:58:49 woglinde: as with 3.4 ecj it is same Dec 18 12:00:01 morning all Dec 18 12:00:06 zecke: hi Dec 18 12:01:59 hi RP Dec 18 12:03:06 hi RP Dec 18 12:03:45 woglinde: for me it looks like jamvm does not even try to open ECJ stuff Dec 18 12:03:45 hi rp Dec 18 12:03:52 hrw hm Dec 18 12:04:10 woglinde: will be back in 1-1.5h - xmas stuff :( Dec 18 12:04:21 hi rp Dec 18 12:04:31 greetings from the frozen south-east Dec 18 12:04:44 pb *g* here too Dec 18 12:05:23 pb_: The north is also rather frozen but I guess its more "normal" here :) Dec 18 12:05:25 hm seems classpath problem Dec 18 12:05:36 a frozen classpath? ;) Dec 18 12:05:41 no Dec 18 12:05:44 hrw's problem Dec 18 12:06:11 woglinde: don't take me too serious Dec 18 12:06:21 florian never Dec 18 12:06:22 *g* Dec 18 12:07:09 :-) Dec 18 12:10:22 someone fixed git sendmail hook? Dec 18 12:10:29 I cannt believe it Dec 18 12:11:02 or is it a feature when you have > 1000 commits? Dec 18 12:11:13 do to be synced Dec 18 12:22:55 yes lots of snow in south east :) Dec 18 12:23:11 i have between 6 inch to 1ft snow outside! Dec 18 13:04:38 Anybody from Italy around? Dec 18 13:08:19 Laibsch: kick mckoan Dec 18 13:09:51 Laibsch: yep! Dec 18 13:10:08 thanks Dec 18 13:10:16 mckoan: Can I PM you? Dec 18 13:18:52 yes of course Dec 18 13:51:49 morning Dec 18 13:52:15 jo kergoth Dec 18 13:52:43 hms Dec 18 13:52:56 I didnt have pushed my simplyfied gettext-native stuff jet Dec 18 13:53:45 morning Dec 18 13:55:22 re Dec 18 13:55:31 woglinde: ideas? Dec 18 13:55:41 woglinde: that test build was fresh .dev Dec 18 13:56:17 hrw whats your classpath version? Dec 18 13:56:58 woglinde: initial 0.93 native 0.97.2 Dec 18 13:57:55 try native 0.98 Dec 18 13:58:03 jo effem Dec 18 13:58:15 jepp Dec 18 13:58:31 newer jamvm has problems with old classpath Dec 18 14:00:02 hi woglinde, all Dec 18 14:01:42 ok Dec 18 14:01:51 hi eFfeM Dec 18 14:01:54 woglinde: need to rebuld jamvm after classpath or not? Dec 18 14:02:06 hm dont know Dec 18 14:02:11 I think not Dec 18 14:02:12 hi mickeyl eFfeM, rest Dec 18 14:04:32 good morning Dec 18 14:04:41 hi recalcati Dec 18 14:04:53 I'm trying to understand which are Texas irc channels Dec 18 14:05:14 I tried #arago, but it not so populated Dec 18 14:05:31 try #beagle Dec 18 14:05:44 #linux-omap as well Dec 18 14:06:04 thx, anyone specific for da vinci? Dec 18 14:06:25 hmm, davinci is a bit of a mystery to me Dec 18 14:06:34 #gst_ti Dec 18 14:06:54 woglinde: I saw it, but now I try again , thx Dec 18 14:08:49 woglinde: | java/lang/NoClassDefFoundError: java/lang/Class Dec 18 14:08:56 woglinde: | make[1]: Leaving directory `/home/hrw/devel/build/angstrom/tmp/work/x86_64-linux/classpath-native-0.98-r1/classpath-0.98/tools' Dec 18 14:08:59 ;( Dec 18 14:09:32 woglinde: http://hrw.pastebin.ca/1719462 in other words Dec 18 14:11:11 hrw hm try build from scratch Dec 18 14:11:28 and maybee hostsystem antlr is interferring Dec 18 14:11:41 hmmm Dec 18 14:12:19 no antle in system Dec 18 14:12:25 will scratch Dec 18 15:32:11 woglinde_: from scratch it built Dec 18 15:32:21 woglinde_: going to do tests with stable now Dec 18 15:59:34 hi kergoth Dec 18 16:00:08 kergoth: I've been thinking about bitbake 1.10. I think I'm going to change my mind and add a patch to bitbake master which disables the client/server stuff, then call it 1.10 Dec 18 16:08:16 morning all Dec 18 16:28:08 hi rkirti Dec 18 16:28:12 rkirti whats the status? Dec 18 16:28:14 re pb Dec 18 16:30:44 woglinde_: just began.. my localcopy now has revamped stuff for base level libs - upto libhildon. Been a little slow due to bad net access Dec 18 16:31:03 morning pb_ Dec 18 17:01:59 hi woglinde_, rkirti Dec 18 17:23:52 woglinde_: it works! Dec 18 17:24:27 rkirti: you plan to use official tarball releases or maemo.gitorius.org repository? Dec 18 17:25:30 be back in 2-2.5h from now Dec 18 17:25:43 hrw: wherever possible I used sources from the maemo repo. Dec 18 17:25:54 (official tarball) Dec 18 17:29:43 hrw jo Dec 18 18:22:46 | Collected errors: | * ERROR: Cannot satisfy the following dependencies for task-base: | * libssl0.9.8 (>= 0.9.8j) * libcrypto0.9.8 (>= 0.9.8j) * libssl0.9.8 (>= 0.9.8j) * libcrypto0.9.8 (>= 0.9.8j) * Dec 18 18:22:52 Hello, I have a question: I would like to do some deep debugging on my image with gdb. I have DEBUG_BUILD="1" and INHIBIT_PACKAGE_STRIP="1". Is there a way that my image includes all -dbg packages in the rootfs? Dec 18 18:22:57 how can i find out where tehse dependancies are defined? Dec 18 18:24:03 this happended after i set preffered version of openssl to 0.9.8g because 0.9.8j refuses to build Dec 18 19:33:43 kfazz: bitbake -g should generate the dependency files for you Dec 18 19:34:23 jo khem Dec 18 19:34:29 woglinde: hey Dec 18 19:34:34 so the dish is waiting Dec 18 19:34:36 till later Dec 18 19:34:46 woglinde: guten appetit Dec 18 19:36:29 kfazz: you are trying to use a lesser version and some deps seems to be already using the 0.9.8j you should rebuild these recipes too Dec 18 19:39:14 re Dec 18 19:39:17 wb Dec 18 19:42:42 re Dec 18 19:52:43 rkirti: hi! Dec 18 19:52:47 hrw: wb Dec 18 20:15:08 khem since when we have version deps? Dec 18 20:18:50 woglinde: since forever Dec 18 20:18:57 woglinde: bitbake just ignores that Dec 18 20:19:27 woglinde: but fun starts when you do versions in RDEPENDS/RRECOMMENDS - opkg/ipkg handles them Dec 18 20:19:28 hrw hm ah but opkg not Dec 18 20:19:33 okay Dec 18 20:19:42 I used it in buglabs trunk Dec 18 20:19:51 hrw java is running fine now? Dec 18 20:20:03 on my machine yes. on BL buildbot not yet Dec 18 20:20:15 I will talk with robert on monday so we set some defaults Dec 18 20:20:45 classpath-native 0.98 libecj-bootstrap 3.4 is what I use Dec 18 20:20:51 robert mentioned there was some problems with jamvm and classpath 0.98 Dec 18 20:20:57 but seems there are solved now Dec 18 20:21:13 on 64bit Dec 18 20:23:09 I do not build on 32bit Dec 18 20:23:58 jepp Dec 18 20:24:20 robert said the problems were on 64bit Dec 18 20:34:31 thanks Dec 18 20:38:47 woglinde: http://hrw.pastebin.ca/1719754 - ideas? Dec 18 20:39:09 woglinde: this happens on one machine, works on mine Dec 18 20:41:46 hrw hm Dec 18 20:42:23 seems antlr.Tool is not in proper place or so Dec 18 20:50:07 it is inside of zip/jar or as a file? Dec 18 20:50:18 bugadmin@dauber:/opt/sandbox/hrw/trunk/build$ find tmp -name antlr* Dec 18 20:50:18 tmp/work/armv6-angstrom-linux-gnueabi/vim-7.0-r1/vim70/runtime/syntax/antlr.vim Dec 18 20:53:21 hrw hm dont know Dec 18 20:53:37 ok Dec 18 21:11:44 woglinde: fond Dec 18 21:11:52 hm? Dec 18 21:11:57 what was it? Dec 18 21:11:58 woglinde: --disable-gjdoc == do not use antlr == builds Dec 18 21:12:03 jupp Dec 18 21:12:14 that was my second tought Dec 18 23:08:41 whats wrong with SRCREV of u-boot for beagle ? Dec 18 23:09:05 git seems not to find it Dec 18 23:18:19 anyone have an hint about the reason opie-image now fails with: Dec 18 23:18:22 ERROR: Task /oe/org.openembedded.dev/recipes/libqpe/libqpe-opie_1.2.4.bb (do_package) has circular dependency on /oe/org.openembedded.dev/recipes/libopie/libopie2_1.2.4.bb (do_package) Dec 18 23:27:28 ant__: this here 6855e395d3277f7040a45148b0d8f25ff74eeea2 Dec 18 23:27:40 ant__: Can you comment out Dec 18 23:28:09 inherit opie in libqpe-opie.inc and see if it works ok again Dec 18 23:28:25 khem: ok, thx Dec 18 23:31:13 ant__: does that help ? Dec 18 23:31:38 he..you'll have to wait..rebuilding from scratch with new headres... ;) Dec 18 23:31:58 but yes, seems to be th eculprit Dec 18 23:33:20 ah, interestingly I could not find the commit in oe git..I googled and found one hit on gitorius.org/gumstix ... Dec 18 23:33:33 to find it in oe I have to search by name :/ Dec 18 23:33:38 ? Dec 18 23:34:00 git.openembedded.org should show you if you provide commit id Dec 18 23:34:27 there are three keyx...committer, authoir, log-msg...no ID ! Dec 18 23:34:43 cgit Dec 18 23:35:37 someone should add that field to the combo Dec 18 23:37:20 what I do is this Dec 18 23:37:28 http://git.openembedded.org/cgit.cgi/openembedded/commit/?id= Dec 18 23:37:38 and after = type the SHA of commit-id Dec 18 23:37:59 yea, but it's abug in the interface I'd say... Dec 18 23:38:51 see here: http://git.linuxtogo.org/?p=groups/kexecboot/kexecboot.git;a=summary Dec 18 23:39:16 yes there should prolly be a grep and commit id in search fields Dec 18 23:46:35 khem: I was testing the arm patch for memcpy Dec 18 23:47:10 I see the 2.6.3x kernels have an separate memcpy optimization Dec 18 23:48:07 Imean this: Dec 18 23:48:07 http://lists.infradead.org/pipermail/linux-arm-kernel/2009-November/003860.html Dec 19 00:13:18 ant__: yes fairly new Dec 19 00:13:27 is it merged yet ? Dec 19 00:13:47 I don't think so Dec 19 00:13:57 I'm testing on 2.6.26 (backporting) Dec 19 00:14:10 because 2.6.32 seems a bit unstable on armv5te Dec 19 00:14:12 it should be ok though Dec 19 00:14:17 (corgi & spitz) Dec 19 00:14:40 just dubious about the memcpy in kernel thing... Dec 19 00:16:03 khem: well, kernel boots from nand. I have to take some figures with/without Dec 19 00:16:45 first impression: faster Dec 19 00:17:11 did the opie thing work for you Dec 19 00:17:19 still rebuilding Dec 19 00:17:24 ugh Dec 19 00:17:36 can you leave a message here for me what happens Dec 19 00:17:49 yea, I'll do Dec 19 00:17:55 I will fix the orginal issue in different way to accomodate this Dec 19 00:17:57 I just commented it out Dec 19 00:18:08 k Dec 19 00:18:09 it's already in libqte Dec 19 00:18:14 inc Dec 19 00:22:25 khem: given the RMK's answer I think the patch has poor chanches;) Dec 19 00:22:27 No one has been bothered for the last 15 years to optimise it. Dec 19 00:23:11 but it is still improvement Dec 19 00:26:22 I don't understand the last comment, though Dec 19 00:26:36 Why use this, there's better {read,write}{b,w,l}s calls available which Dec 19 00:26:36 do the job in lovely optimised assembler. Dec 19 00:28:44 better than the 'real' memcpy? Dec 19 00:39:29 these are utilities in kernel Dec 19 00:41:37 and iirc gcc uses inline asm starting with -O1 Dec 19 00:42:04 I mean, in userspace Dec 19 00:43:29 khem: I'll just do the opie-benchmarks (if it builds). have to go now. I'll keep you updated. Dec 19 00:43:31 bye **** ENDING LOGGING AT Sat Dec 19 02:59:56 2009