**** BEGIN LOGGING AT Mon Aug 09 02:59:57 2010 Aug 09 06:26:30 RP: 6b6933550f60098388ed77c964f223d8f919379c stages link to workdir (${D}${libdir}/pkgconfig/xcalibrateproto.pc) and then qa_staging fails (when workdir is pruned by rm_work), can it be just relative path? "ln -sf xcalibrateproto.pc ${D}${libdir}/pkgconfig/xcalibrateext.pc" Aug 09 06:29:21 gm everyone Aug 09 06:34:33 JaMa: yes try this ln -s xcalibrateproto.pc ${D}${libdir}/pkgconfig/xcalibrateext.pc Aug 09 06:34:55 thats correct Aug 09 06:35:04 and commit it you have my ack Aug 09 06:36:51 eFfeM_work: gm Aug 09 06:37:06 khem thanks again for your fixes, I don't think I would have been able to resolve this myself Aug 09 06:37:22 eFfeM_work: no worries Aug 09 06:37:38 khem: ok, thanks for quick reply Aug 09 06:37:43 still fighting autotools for urjtag git head where it refuses to give MKINSTALLDIR the right value Aug 09 06:38:10 eFfeM_work: heh another daemon to be slain Aug 09 06:38:17 btw if there is anyone that has a suggestion how to spit out an error in src_distribute_local.bbclass, please enlighten me Aug 09 06:38:18 http://lists.linuxtogo.org/pipermail/openembedded-devel/2010-August/022613.html Aug 09 06:38:43 btw when you go into fixing do_stage you can also see if a new version is available and some old ones can be punted Aug 09 06:39:05 khem if you feel like giving my autotools issue a quick stab, see http://lists.linuxtogo.org/pipermail/openembedded-devel/2010-August/022614.html Aug 09 06:39:06 eFfeM_work: I will read it tomorrow Aug 09 06:39:10 khem agree on the versions Aug 09 06:39:17 fine, np Aug 09 06:39:27 right now I am pushing my son's swing :) Aug 09 06:39:37 ah cool Aug 09 06:39:44 where are you located ? Aug 09 06:39:49 he doesnt like if I turn on the button Aug 09 06:40:02 CA.USA Aug 09 06:40:30 thought that, then you.re at 11.40 pm, your son should be asleep Aug 09 06:40:42 * eFfeM_work has kids too so he knows it does not always work taht way Aug 09 06:40:53 khem: any news about alignment traps in eglibc-2.12? is it reported upstream or it has to be fixed in OE? Aug 09 06:41:07 he is a full pack of energy Aug 09 06:41:38 JaMa: if anyone who will fix it will be me :) Aug 09 06:41:41 and I havent done it Aug 09 06:42:00 its a complex problem to debug Aug 09 06:42:05 03Martin Jansa  07org.openembedded.dev * r620f1cef28 10openembedded.git/recipes/xorg-proto/calibrateproto_git.bb: Aug 09 06:42:05 calibrateproto: fix qa_staging Aug 09 06:42:05 * use relative path, otherwise we stage link to workdir, which is later cleaned by rm_work and qa_staging fails Aug 09 06:42:05 Acked-by: Khem Raj Aug 09 06:42:05 Signed-off-by: Martin Jansa Aug 09 06:42:18 khem: ok, thanks Aug 09 06:42:30 JaMa: but I remember it Aug 09 06:42:51 I just asked because SHR users also asked :) Aug 09 06:43:04 * JaMa can live with it (with disabled logging) Aug 09 06:43:04 hmm Aug 09 06:43:23 its only reproducable on gentoo Aug 09 06:43:29 and I dont have a gentoo box around Aug 09 06:43:42 khem: do you want access to mine? Aug 09 06:44:49 JaMa: sure I will ask you when I will work on it Aug 09 06:45:18 5ok Aug 09 06:45:23 its will take couple of days atleast I imagine Aug 09 06:45:51 can you ask SHR user to verify his host Aug 09 06:45:56 or build Aug 09 06:46:33 btw. can someone ack my openmoko distro file patch plz Aug 09 06:46:37 you mean it's reproducible only on gentoo as buildhost? Aug 09 06:46:42 yes Aug 09 06:46:53 no, we have it from different distro too, mmt Aug 09 06:46:54 I cant see it on my debian or ubuntu Aug 09 06:47:12 atleast last time Aug 09 06:47:17 debian 5.0.5, the same Aug 09 06:47:24 ok Aug 09 06:47:35 and IIRC it happens only when eglibc-2.12 is built with gcc-4.5 Aug 09 06:47:43 good, let me see if I will have some time to look into it Aug 09 06:47:48 its mostly a gcc bug Aug 09 06:47:55 nothing to do with eglibc Aug 09 06:48:31 and it exists in 4.4 too just that it does not trigger there thats my guess Aug 09 06:48:51 alright time to sleep now Aug 09 06:48:53 ttyl Aug 09 06:49:05 * khem -> sleep () Aug 09 06:49:50 gnite Aug 09 06:53:08 morning Aug 09 07:07:30 good morning Aug 09 07:12:30 khem: thx for the first step about klibc. Unfortunately something still needs to be fixed. The recipe stages much more things in sysroots using the old do_stage. Well, I'm unsure all is needed but the 'legacy' staging-klibc- package has klcc both in /sysroots/armv5te and in sysroots/i686/bin/armv5te iirc. Now the couple of klibc-static packages fail to build. Try bitbake kexecboot for e.g. qemuarm Aug 09 07:13:17 klibc libs seem not staged... Aug 09 07:15:02 JaMa: I lost time with xcalibrateproto.pc so I could not rebuild from scratch but I fear you'll see couple of failures (kexec-tools kexecboot) Aug 09 07:16:09 ant_work: I had rebuild from scratch ok.. but not really sure if it was before your revert or after or partialy before.. Aug 09 07:17:23 heh, I had a commit almost like khem but this wont stage klcc. Then I removed $D in my local copy and it works(I know it's evil). Unfortunately I did commit the broken one :/ Aug 09 07:18:10 khem committed almost right version (imho the 1.5.18 inc edit is superfluous, all includes klibc.inc) Aug 09 07:18:25 still there is some magic in do_stage wrt libs Aug 09 07:18:34 which we have to do by hand I suppose Aug 09 07:19:22 but I'm a bit lost after variables renaming/redefinition.... Aug 09 07:19:55 I think installing klcc on host is wrong. This should not be packaged, even Aug 09 07:20:11 s/host/target host/ Aug 09 09:45:30 good morning Aug 09 09:50:49 hi florian Aug 09 09:55:12 ich wuerde das js nichts davon wissen lassen, sondern nur eine moeglichkeit in js zu sagen setPropety(path, prop, bool persitent) Aug 09 09:55:15 oops wrong window :) Aug 09 10:10:42 :-) Aug 09 10:11:42 happens all the time ;) I use bitchx in split screen mode in a terminal and usually forget to switch between channels ;) Aug 09 10:21:31 ;) Aug 09 10:21:55 split screen and/or split terminal are useful things Aug 09 10:22:37 yes :) I still need to find some time to add UTF-8 support to BitchX :> I just can't switch to another client Aug 09 10:24:08 I used ircII, epic3, epic4/2000 and then migrated to irssi Aug 09 10:29:38 When i run facedetect --cascade=/usr/share/opencv/haarcascades/haarcascade_frontalface_default.xml /usr/share/opencv/samples/lena.jpg Aug 09 10:29:45 i get an error (result:1327): Gtk-WARNING **: cannot open display: Aug 09 10:29:54 Did any fix for this error ? Aug 09 10:31:26 i installed the opencv packages using opkg ( opkg install opencv-apps,opkg install opencv-samples) Aug 09 10:36:59 Ananth_: just a wild guess, is your DISPLAY variable set? Aug 09 10:37:17 No Aug 09 10:38:03 I got some from some forum that to follow like this echo $DISPLAY Aug 09 10:38:07 sudo -i Aug 09 10:38:24 xauth merge ~/.Xauthority Aug 09 10:38:31 export DISPLAY=:0.0 Aug 09 10:38:46 But after this i checked the display macro Aug 09 10:39:00 root@overo:~# echo $DISPLAY Aug 09 10:39:01 :0.0 Aug 09 10:39:30 And error got from the application as (result:1473): Gtk-WARNING **: cannot open display: :0.0 Aug 09 10:39:34 Now any guess ? Aug 09 10:40:02 Note: i did xauth merge ~root/.Xauthority Aug 09 10:40:15 but i got error on that step that Aug 09 10:40:21 unable to open file /home/root/.Xauthority Aug 09 11:48:55 RP: ping? Aug 09 12:07:26 mickey|office: do you happen to know any dedicated server hosting providers in germany that you could recommend? Aug 09 12:08:24 pb_: check one where linuxtogo is hosted. iirc strato.de Aug 09 12:16:11 hrw, pb_: it used to be, its at Hetzner now. The experience with them is good so far. Aug 09 12:16:42 ah,ok Aug 09 12:32:22 Jay7: hm I still get the same freeze after initialized empty git repo when trying to do_fetch linux-qsd8 Aug 09 12:32:53 Jay7: I think it is really not downloading anything. In tcpdump I can only see the ssh traffic between me and the build host Aug 09 12:36:21 I suspected it is due to bad bitbake revision but I downloaded different one (I'm on master HEAD) now and the problem persists Aug 09 12:36:29 I really don't want to rm -rf Aug 09 13:00:10 nobody came across this problem ? Aug 09 13:00:36 or is it too distro or machine specific to ask here Aug 09 13:04:39 dcordes: every git? Aug 09 13:04:51 tried bitbake -cclean git-native; bitbake git-native Aug 09 13:04:58 ? Aug 09 13:05:14 wut? Aug 09 13:07:15 eFfeM_work: no others work Aug 09 13:07:24 dcordes: I've seen it. I gave up, cloned by hand and then just put it into OE's download dir Aug 09 13:07:39 (happend for some other linux recipe for me) Aug 09 13:07:51 i never had problems using 1.10 bb head from two weeks ago or so Aug 09 13:08:14 but recently i noticed donwloads from angstrom-distribution are slooooooow Aug 09 13:08:45 zub: hm interesting you also think it might be git-native problem ? Aug 09 13:09:32 dcordes: don't know; I tried to figure out, but gave up :-( Aug 09 13:09:44 dcordes: but git clone (usin my distro's git) worked ok Aug 09 13:09:49 ok I will start over Aug 09 13:10:07 zub: it seemed to me cloning the same git manually on same host worked Aug 09 13:10:11 too Aug 09 13:10:28 if I will get the same error I will just coyp it manually Aug 09 13:10:46 zub: do you have to put the do_fetch stamp ? Aug 09 13:10:51 dcordes: what if you try running OE git-native by hand? maybe it tells you more of what's happening Aug 09 13:10:52 dcordes: it just timeouts.. happends a lot with huge git repos (like kernel) from github/gitorious etc.. Aug 09 13:11:19 JaMa: hm maybe bitbake should be patched to be more patient then ? Aug 09 13:11:58 dcordes: nope, I just removed all except for .git and moved it into downloads/git/correct-dir-name/ Aug 09 13:12:09 dcordes: just look into downloads/git Aug 09 13:12:29 (might have been suboptimal, but worked) Aug 09 13:12:42 dcordes: I think that remote end is not enough patient (and I don't see git option for that on client side) Aug 09 13:13:03 dcordes: the message is usually "Remote end hung up", right? Aug 09 13:13:25 JaMa: no I see no message but Aug 09 13:13:32 initialized empty git repo in ... Aug 09 13:13:35 dcordes: fetching with -D? Aug 09 13:13:58 no I just bitbake linux-qsd8 Aug 09 13:19:16 03Martin Jansa  07org.openembedded.dev * rbf3cf7b95c 10openembedded.git/recipes/tasks/task-x11.bb: task-x11: fix PACKAGE_ARCH override for task-x11-server, _pn- didn't work it seems Aug 09 13:19:25 03Martin Jansa  07org.openembedded.dev * r96b2d0344d 10openembedded.git/recipes/tasks/task-shr-feed.bb: task-shr-feed: add xfsprogs Aug 09 14:10:26 florian: thanks, Hetzner looks like a good option. Aug 09 14:11:46 pb_: yw... so far we really can't complain. the support was very responsive and you get quite powerful machines quite cheap Aug 09 14:12:55 yeah, that's definitely cheaper than the hosting people we use in the UK. not quite sure how it compares to our site in the Netherlands but I suspect Hetzner is cheaper than them too. Aug 09 14:13:18 and the support we get from the Netherlands folk is fairly sucky, so I am not very keen on going back to them again. :-} Aug 09 14:13:49 pb_: the only computing centre i really can recommend is Hetzner Aug 09 14:13:56 Hetzner i never had trouble with Aug 09 14:14:10 mickey|office: very good, thanks Aug 09 14:14:35 florian: d'oh -- aren't you on vacation? Aug 09 14:15:01 mickey|office: not anymore Aug 09 14:15:15 florian: that was short. just one week? Aug 09 14:15:46 mickey|office: yes indeed... too much of work here for longer vacations. Aug 09 14:16:03 florian: i see. beware of burn out though Aug 09 14:16:32 usually summer time is quite relaxed here but this year we have a pile of strange projects Aug 09 14:16:39 mickey|office: true Aug 09 14:18:06 things like: we discussed a project almost a year ago... now the customer replies to some old email telling us to do this project right now Aug 09 14:18:24 hey guys Aug 09 14:18:29 g'day kergoth_ Aug 09 14:25:04 florian: sounds familiar to me. sucky Aug 09 14:25:08 hi kergoth_ Aug 09 14:32:58 hi all Aug 09 14:33:53 hi stefan_schmidt Aug 09 14:57:19 hi HopsNBarley ! haven't seen you in a while Aug 09 14:57:45 hey Jin^eLD - well, i took an ill-advised detour away from embedded, what can I say? Aug 09 14:57:52 anybody up on bug 5398, the antlr-native problem? Aug 09 14:58:35 oh.. I hope you're well again Aug 09 14:58:58 I still did not have time to bring the storcenter to live Aug 09 14:59:00 Jin^eLD, no - i wasn't sick! thanks anyway. How've you been? Aug 09 14:59:30 ok I guess, shifted hobbies from coding to working on my car :) Aug 09 15:00:39 nice. quite similar. i used to be a lawn mower mechanic :) Aug 09 15:00:39 have you been doing anything on openprotium? Aug 09 15:00:48 sadly, no, but i need to. Aug 09 15:01:08 I started from zero on mechanics when I bought that car last year :) Aug 09 15:02:57 but you already knew code! Aug 09 15:03:15 well that did not help me at all with the car though ;) Aug 09 15:31:17 aja stimmt ja Aug 09 15:31:20 ops wrong window Aug 09 15:31:28 03Stefan Schmidt  07org.openembedded.dev * r9d0e260c60 10openembedded.git/recipes/pointercal/ (files/bug20/pointercal pointercal_0.0.bb): pointercal: Add bug20 support and bump PR Aug 09 15:31:51 03Stefan Schmidt  07org.openembedded.dev * r9b1a9d8fd1 10openembedded.git/recipes/firmwares/marvell-sdio-tf-fw_9.0.7-p2.bb: firmware: Add recipe to work with the marvell-sdio-tf-fw Aug 09 16:15:57 anyone know why the default hostap-daemon picked up for angstrom-2008 would be 0.5.10 and not 0.6.9? grepping through preffered versions didn't yield anything obvious (stable) Aug 09 17:46:49 khem: about klibc: http://pastebin.ca/1913941 vs. http://pastebin.ca/1913940 Aug 09 17:49:26 old do_stage vs new do_install Aug 09 17:55:07 khem, now e.g. kexec-tools-klibc-static fails like this http://pastebin.ca/1913945 Aug 09 17:55:10 bbl Aug 09 18:33:54 I just struggled with the bitbake git fetcher (http://bitbake.berlios.de/manual/ch03s07.html) and found that SRCREV must be defined to do a successful fetch, but that's not mentioned in the docs. SRCREV = "HEAD" worked fine for my needs. Aug 09 18:34:48 you're probably better off using SRCREV = "${AUTOREV}" than HEAD Aug 09 18:35:13 what's AUTOREV? Aug 09 18:36:18 kergoth_: did you using ArchLinux as buildhost? Aug 09 18:36:29 * Jay7 trying to remember Aug 09 18:42:34 I´ve created an SDK that a colleague and myself use to develop for the beagleboard. It has worked great, but after my latest update&build we have a weird issue. On his system GCC won't run, complaining that it doesn't find libmpfr.so.4, while it works on my system. We have the same OS, all the latest upgrades, and the mentioned file can be found in his SDK as well as mine. Anyone got a clue what could cause stuff like that? Aug 09 18:42:42 ant_: hmm ok I will look into it in a moment Aug 09 18:46:06 * Jay7 is selecting between ArchLinux and Debian for build host.. Aug 09 18:49:13 Shouldn't be possible that something installed on my host when I created the SDK is missing on his system, and therefore he can't launch his gcc-for-arm compiler? Aug 09 18:50:55 He just needs to install some packages, yes Aug 09 18:52:05 Well, actually, did he install it SDK_PATH ? Aug 09 18:53:52 Tartarus: we have the SDK in version controlled, so a scons-build points to the needed locations. going back to before my latest SDK-commit, everything works well for him. Aug 09 18:56:52 tasslehoff: arm-...-gcc is normally linked with an RPATH of SDK_PATH (OE variable) Aug 09 18:57:05 So if it's not installed into SDK_PATH, it may not find certain libraries Aug 09 18:57:24 * Tartarus isn't 100% sure if we $ORIGIN mangle sdk recipes right now, but I don't think so Aug 09 18:59:15 Tartarus: hm. ok. so it could actually be because I have done an OE-build in the same shell (and ran an env-setup) that it works? Aug 09 18:59:29 No Aug 09 18:59:38 Where did you install the SDK Aug 09 18:59:44 and where did your coworker install it to? Aug 09 18:59:56 hi all Aug 09 18:59:59 And, what is SDK_PATH set to (bitbake your-sdk-target -e | grep -E ^SDK_PATH Aug 09 19:02:33 we both tried checking out a tree to /data/src/ , and that would place our toolchain in /data/src/toolchains/arm/. We use OE/bitbake to compile our rootfs, but not to compile the (small) app we have problems with. We have a scons-setup that sets up the needed paths and environment variables. Aug 09 19:03:00 Tartarus: I'm at home researching a bit now. Can't test any commands until I get to my workpc tomorrow :-/ Aug 09 19:03:20 So, maybe here's some confusion Aug 09 19:03:34 Are you talking about using a tarball created by bitbake meta-toolchain-foo Aug 09 19:03:48 Or the internal to OE cross compiler that lives inside of /data/src/tmp/ ... ? Aug 09 19:04:02 Tartarus: the first one. Aug 09 19:04:21 Right Aug 09 19:04:34 So, when you did bitbake, did you set SDK_PATH=/data/src/toolchains/arm/ ? Aug 09 19:04:37 Or no Aug 09 19:05:04 Tartarus: that I did not. Aug 09 19:05:28 re Aug 09 19:06:44 is there anyone who can explain me how to emit an error in a bbclass file ? Aug 09 19:15:02 03Henri Bragge  07org.openembedded.dev * rf3114e1b53 10openembedded.git/recipes/gsnmp/ (gsnmp-0.3.0/quote-fix.patch gsnmp_0.3.0.bb): Aug 09 19:15:02 gsnmp-0.3.0: fix quoting in autoconf macros Aug 09 19:15:02 Signed-off-by: Henri Bragge Aug 09 19:21:34 is it preferred to have checksums immediately after the SRC_URI ? Aug 09 19:27:14 03Frans Meulenbroeks  07org.openembedded.dev * r8f738d5d18 10openembedded.git/recipes/cdparanoia/ (cdparanoia_10.2.bb cdparanoia_9.8alpha.bb cdparanoia_svn.bb): Aug 09 19:27:14 cdparanoia: oe-stylize Aug 09 19:27:14 Signed-off-by: Frans Meulenbroeks Aug 09 19:29:46 03Frans Meulenbroeks  07org.openembedded.dev * r8888b934f9 10openembedded.git/recipes/tgt/ (tgt_0.9.10.bb tgt_1.0.7.bb): Aug 09 19:29:47 tgt: oe-stylize Aug 09 19:29:47 Signed-off-by: Frans Meulenbroeks Aug 09 19:29:59 guess everyone is on vacation ; lucky guys ;-) Aug 09 19:32:48 Tartarus: when I launch gcc for arm and it complaints about libmpfr missing, is it the x86 or arm library it misses? Aug 09 19:33:42 It's a host library, due to how arm-...-gcc is linked not being found Aug 09 19:33:54 do a readelf -d on it and grep for RPATH Aug 09 19:34:04 Those are the places that host libraries are looked for, outside of the usual spots Aug 09 19:35:16 Tartarus: since it finds it on my system it probably is a packet missing on his host. thanks for helping. Aug 09 19:48:48 03Frans Meulenbroeks  07org.openembedded.dev * r5566943a65 10openembedded.git/recipes/mythtv/ (5 files): Aug 09 19:48:48 mythtv: removed tabs Aug 09 19:48:49 Signed-off-by: Frans Meulenbroeks Aug 09 19:48:59 03Frans Meulenbroeks  07org.openembedded.dev * rd2426a2442 10openembedded.git/recipes/cdstatus/ (cdstatus_0.96.05.bb cdstatus_0.97.01.bb): Aug 09 19:48:59 cdstatus: removed tabs Aug 09 19:48:59 Signed-off-by: Frans Meulenbroeks Aug 09 19:49:43 03Frans Meulenbroeks  07org.openembedded.dev * r108144464d 10openembedded.git/MAINTAINERS: Aug 09 19:49:43 MAINTAINERS: updated my entry Aug 09 19:49:43 Signed-off-by: Frans Meulenbroeks Aug 09 19:50:38 * Jay7 is thinking that MAINTAINTER should go into recipe.. Aug 09 19:52:39 Jay7: me too, but I did only get limited feedback on my email to do so Aug 09 19:53:11 basically I now use git blame to identify what I consider to be the owner, Aug 09 19:53:38 the person with the most lines assigned to him/her is the owner ;-) Aug 09 20:22:02 eFfeM: obviously i missed your mail... when did you send it? Aug 09 20:29:54 florian, on maintainers ? Aug 09 20:30:01 * eFfeM sends many mails Aug 09 20:30:37 eFfeM: yes Aug 09 20:31:56 florian: http://lists.linuxtogo.org/pipermail/openembedded-devel/2010-July/022161.html Aug 09 20:32:53 there are some replies, but not really a conclusion Aug 09 20:39:19 btw here is the thread on the discussion to *remove* maintainers from the recipes Aug 09 20:39:20 http://lists.linuxtogo.org/pipermail/openembedded-devel/2006-October/000521.html Aug 09 20:41:58 calling it a day, cya tomorrow & remember: have fun! Aug 09 20:54:14 khem: any test I can do? Aug 09 21:14:52 argh ..build started in /openembedded and not in openembedded/build....there was a guard once... Aug 09 21:15:44 old bitbake perhaps Aug 09 21:20:37 khem - hey Aug 09 21:22:30 HopsAtWork: hey Aug 09 21:30:27 03Jason Kridner  07org.openembedded.dev * ra5ee540c8a 10openembedded.git/recipes/images/beagleboard-test-image.bb: Aug 09 21:30:27 beagleboard-test-image: add additional packages Aug 09 21:30:27 Add sox for audio testing. Aug 09 21:30:27 Add kernel-module-mt9t112 for camera testing. Aug 09 21:30:27 Add u-boot-mkimage for editing the boot.scr and user.scr files. Aug 09 21:30:28 Signed-off-by: Jason Kridner Aug 09 21:30:28 Signed-off-by: Denys Dmytriyenko Aug 09 21:30:31 03Jason Kridner  07org.openembedded.dev * r9eab7a12e4 10openembedded.git/recipes/angstrom/ (3 files in 2 dirs): Aug 09 21:30:31 angstrom-uboot-scripts: debugged beagleboard validation boot/user scripts Aug 09 21:30:31 Signed-off-by: Jason Kridner Aug 09 21:30:31 Signed-off-by: Denys Dmytriyenko Aug 09 21:30:31 03Jason Kridner  07org.openembedded.dev * r3914198ad9 10openembedded.git/recipes/netbase/ (netbase/beagleboard/interfaces netbase_4.21.bb): Aug 09 21:30:32 netbase: beagleboard: auto configure xM Ethernet Aug 09 21:30:32 Signed-off-by: Jason Kridner Aug 09 21:30:33 Signed-off-by: Denys Dmytriyenko Aug 09 21:30:40 03Jason Kridner  07org.openembedded.dev * r74a1a1017a 10openembedded.git/recipes/images/beagleboard-test-image.bb: Aug 09 21:30:40 beagleboard-test-image: add devmem2 Aug 09 21:30:40 Signed-off-by: Jason Kridner Aug 09 21:30:40 Signed-off-by: Denys Dmytriyenko Aug 09 21:39:21 So... Aug 09 21:39:30 binutils.inc has it's own copy of export CC Aug 09 21:39:43 Put there back in 2005 Aug 09 21:40:31 03Khem Raj  07org.openembedded.dev * r8b31920550 10openembedded.git/conf/distro/include/preferred-om-2008-versions.inc: Aug 09 21:40:31 preferred-om-2008-versions.inc: Update PREFERRED_VERSION for libftdi confuse and ftdi_eeprom Aug 09 21:40:31 Signed-off-by: Khem Raj Aug 09 21:40:31 Acked-by: Martin Jansa Aug 09 22:07:28 thanks denix Aug 09 22:09:24 pb_: it seems we need another lesson about overrides (Ref: _pn-${PN}) Aug 09 22:09:33 heh Aug 09 22:10:17 denix: do you know if the xdctools are supposed to be installed to build dsplink? it seems they aren't requested to be downloaded during the build, but the 'prepsources' task fails. Aug 09 22:17:34 k, maybe all I need to do is 'bitbake ti-xdctools'. Looks like that dependency might just be missing. Aug 09 22:32:54 jkridner2: not sure, let me check Aug 09 22:34:08 denix0: my error is /home/ubuntu/angstrom-setup-scripts/build/tmp-angstrom_2008_1/work/beagleboard-angstrom-linux-gnueabi/ti-dsplink-1_1_64-r88h/temp/run.do_prepsources.29982: line 778: /home/ubuntu/angstrom-setup-scripts/build/tmp-angstrom_2008_1/sysroots/beagleboard-angstrom-linux-gnueabi/usr/share/ti/ti-xdctools-tree/xdc: No such file or directory Aug 09 22:34:50 bitbake ti-xdctools ran to completion without an issue, but didn't fix this. Aug 09 22:35:50 the folder exists, but is empty. Aug 09 22:36:04 could this be one of those staging issues? Aug 09 22:38:35 denix0: or perhaps something tied to rm_work? Aug 09 22:38:52 interesting Aug 09 22:39:12 not sure rm_work would be to blame - it's not supposed to clean in sysroots... Aug 09 22:39:42 what version of xdctools was built? Aug 09 22:40:22 ti-xdctools-3_16_01_27-r2: task do_build: Succeeded Aug 09 22:40:37 hmmm.... bitbake ti-codec-engine failed. Aug 09 22:40:45 same issue. Aug 10 02:39:56 anybody seeing a problem in gdb-cross-sdk? only on 32... my run.do_populate_sysroot file is zero length Aug 10 02:46:18 denix0: I believe the xdctools binary has a problem with x86_64, but I'm not sure still. the executable does nothing. Aug 10 02:48:16 ah, quite possible - I'm not using 64bit machines... Aug 10 02:56:52 i put the bin on a 32-bit machine and it expanded "properly"... Aug 10 02:56:59 omg, this installer is a work of junk! Aug 10 02:58:30 tell me about it... :) **** ENDING LOGGING AT Tue Aug 10 02:59:58 2010