**** BEGIN LOGGING AT Mon May 14 02:59:57 2007 May 14 03:08:42 summatusmentis: no, it doesn't May 14 03:08:51 Zero_Chaos: alright, thanks May 14 05:47:07 goodmorning all May 14 06:23:39 koen|away: http://bugs.openembedded.org/show_bug.cgi?id=2311 , I have attached all the bitbake recipies for openalchemy, please have a look at it when you can and let me know if anything needs to be fixed. May 14 06:30:14 morning May 14 06:31:40 XorA hi ! May 14 06:36:36 good morning May 14 07:04:55 has anyone been able to build asterisk_1.2.18 with uclibc ? May 14 07:21:41 good morning all May 14 07:23:40 morning koen May 14 07:27:28 hey rwhitby May 14 07:27:33 how was your trip? May 14 07:29:41 short but long May 14 07:38:25 rwhitby: the task-base provides thingy was raised on the mailinglist by likewise, with different solutions May 14 07:39:01 rwhitby: http://lists.linuxtogo.org/pipermail/openembedded-devel/2007-May/002085.html May 14 07:40:23 ok, I don't care about #1, but do care about #2. I've just put back my PROVIDES="${PACKAGES}" work-around in my local builds. May 14 07:49:16 heh, maemo has "Compilable sources" as an item on their roadmap May 14 08:00:05 hi. pls help me again May 14 08:03:33 while building uclibc for oe it fails. database is fresh. May 14 08:04:24 | In file included from ../../ldso/include/ldso.h:26, May 14 08:04:24 | from ldso.c:33: May 14 08:04:24 | ../../ldso/include/dl-syscall.h: In function '_dl_exit': May 14 08:04:24 | ../../ldso/include/dl-syscall.h:63: error: '__NR_exit' undeclared (first use in this function) May 14 08:04:53 and too much undeclared __NR_xxx... May 14 08:06:10 distro, machine? May 14 08:07:12 /xorg-lib/libxi failed om my machine . it expects sed to be located in /usr/bin/ May 14 08:07:19 Hi May 14 08:08:22 here is the log http://paste-it.net/2178/raw/ May 14 08:09:41 keesj: where is you sed located? May 14 08:11:00 generic-uclibc for arm (ep9312) May 14 08:11:49 sandworm: use DISTRO=angstrom-2007.1 in uclibc mode May 14 08:14:01 in /bin/sed May 14 08:14:58 koen: how to set uclibc mode ? May 14 08:15:15 ANGSTROM_MODE = "uclibc" in local.conf May 14 08:18:21 koen: trying.... May 14 08:19:26 I'd to an 'mtn pull && mtn update' first :) May 14 08:20:03 03koen 07org.oe.dev * r55b586c6... 10/ (1 packages/uclibc/uclibc_0.9.29.bb): uclibc 0.9.29: fix FILESPATH, add ep93xx config May 14 08:20:07 03koen 07org.oe.dev * r36e9da9d... 10/ (8 files in 5 dirs): uclibc 0.9.29: add fic-gta01, poodle and (now really) ep93xx configs May 14 08:20:29 koen: sure, about 30min ago May 14 08:25:03 koen: are you going to release both glibc and uclibc images for Angstrom? May 14 08:25:17 koen: what's the release plan? May 14 08:25:24 xjqian: if people are interested May 14 08:25:48 xjqian: my main plan is to have a small image a la debian-installer for people to use May 14 08:27:56 koen: my concern is wether we are going to have enough developers to support both May 14 08:28:11 same here May 14 08:28:24 that's why I said "if people are interested" :) May 14 08:28:55 koen: if we're going to switch to uclibc, switch now since it's still early and ditch glibc May 14 08:29:22 I don't want uclibc on PDA type devices May 14 08:29:32 I want it for routers, NASs and stuff May 14 08:30:58 do you mean the benefit of uclibc is only space saving rather than speed/efficiency? May 14 08:31:26 xjqian: the unbenefit is the ton of support for a lot of apps May 14 08:31:59 and it isn't ABI compatible with itself, so upgrading via ipkg/deb/rpm/whatever is nigh impossible May 14 08:33:06 XorA: quote from uclibc website: " uClibc (aka µClibc/pronounced yew-see-lib-see) is a C library for developing embedded Linux systems. It is much smaller than the GNU C Library, but nearly all applications supported by glibc also work perfectly with uClibc" May 14 08:33:18 keesj: it seems libtool assumes sed is in /usr/bin May 14 08:33:25 XorA: do you mean they are bragging? May 14 08:33:50 hmm , so perhaps I need to check what the lsb is saying about that May 14 08:33:57 xjqian: arch != i386 == breakage gcc != 2.95.X == breakage, Ill extrapolate libc != glibc == breakage May 14 08:34:43 uclibc 0.9.29 doesn't do 'index()' anymore, so I had to patch a few apps yesterday May 14 08:34:53 it isn't as easy as the uclibc website makes it out to be :( May 14 08:35:40 hehe, I see, everybody is putting their positive side up front. thanks, I understand better now about uclibc May 14 08:36:24 but for 8M flash is no way to use glibc... May 14 08:36:39 xjqian: and they forget to tell you that if you want proper i18n you need libiconv, which is huge May 14 08:37:00 koen: its simpler to convert the whole world to english :-D May 14 08:37:09 kernel eats 1.5M May 14 08:37:16 thanks, I see the niche for uclibc now May 14 08:39:01 koen: 0.9.29 supposedly is abi-compatible with 0.9.28? May 14 08:40:08 i am using uclibc (0.9.28 though) with OE for sometine now, and except a few problems with some programs that need patching to work with uclibc works fine May 14 08:41:59 uclibc fails in odd and cryptic ways though ;) May 14 08:42:17 considering that most flash deprived ;) devices in the market use uclibc/busybox it is a viable option May 14 08:42:43 NAiL: During build or when the system is working ? May 14 08:42:45 Hi! May 14 08:42:59 psokolovsky : hi ! May 14 08:43:03 koen: so, maybe it's easier patch uclibc with index() than every app? May 14 08:45:11 Ifaistos: during build mostly May 14 08:46:20 NAiL: Most of the problems i found where configuration related. Either uclibc had something not configured in, or site files was missing something May 14 08:46:52 NAiL: The other problem was that with gcc 4.x you need to use a rather "recent" version of uclibc May 14 08:47:27 NAiL: so cvs/svn was the only option until the 0.9.29 came :) May 14 08:47:56 Ifaistos: yeah, I know May 14 08:57:43 good morning May 14 09:00:56 morning May 14 09:01:33 good morning May 14 09:06:30 CROSS COMPILE Badness: /usr/include in INCLUDEPATH: /usr/include/ May 14 09:06:31 cc1: internal compiler error: in add_path, at c-incpath.c:362 May 14 09:06:34 ??? May 14 09:07:05 in uclibc-0.9.29-r0 for angstrom May 14 09:17:28 03rwhitby 07org.oe.dev * r1e54d761... 10/ (4 files in 3 dirs): sysconf: Remove /linuxrc and spurious debug messages May 14 09:17:41 03rwhitby 07org.oe.dev * rf7790705... 10/ (5 files in 5 dirs): ixp4xx-kernel: Dropped support for 2.6.21, we're sticking to 2.6.20 stable and 2.6.22 development May 14 09:17:54 03rwhitby 07org.oe.dev * r769dfe2e... 10/ (6 files in 3 dirs): arm-kernel-shim: Updated to version 1.4 May 14 09:18:06 03rwhitby 07org.oe.dev * rf40c7790... 10/ (3 files in 2 dirs): ftpd-topfield: Updated to 0.7.3 May 14 09:19:04 xjqian: Thank you "taking over the tosa" May 14 09:19:24 xjqian: You have an environment to compile, right? May 14 09:19:55 Why don't you go ahead and do a bitbake "opie-image" to see if the opie issues are dealt with? May 14 09:20:03 Thanks again for your work May 14 09:34:46 "linux violates 107 patents, of which 42 are for the the kernel" May 14 09:35:03 I always thought linux *was* the kernel May 14 09:35:29 koen: Only if you're not in marketing May 14 09:38:25 koen: I think you and mickey have "the master key" to bugzilla, right? As you know, I am thinking about workflow for our bugs. I think there are features in bugzilla we do not use and that could help us. I was wondering if it was possible to get master access to bugzilla May 14 09:38:45 Things I am looking for are flags, "watch this", etc. May 14 09:39:00 I do not intend to make any changes without prior consultation. May 14 09:39:16 But to get the imagination going it would be nice to look at the real thing. May 14 09:43:23 Laibsch: sure, ask treke/ggilbert May 14 09:43:37 Laibsch: but I'd wait a bit, since he is trying to upgrade bugzilla to 3.0 May 14 09:49:31 Looking won't hurt anyone I hope ;-) May 14 09:50:20 koen: why ?: NOTE: package uclibc-0.9.29-r0: task do_compile: started May 14 09:50:21 ERROR: function do_compile failed May 14 09:50:21 ERROR: log data follows (./openembedded/build/tmp/work/ep93xx-angstrom-linux-uclibcgnueabi/uclibc-0.9.29-r0/temp/log.do_compile.4109) May 14 09:50:21 | NOTE: make OPTIMIZATION= CPU_CFLAGS=-isystem/build/tmp/staging/arm-angstrom-linux-uclibcgnueabi/include -fexpensive-optimizations -fomit-frame-pointer -frename-registers -Os STRIPTOOL=true LD=arm-angstrom-linux-uclibcgnueabi-ld LOCALE_DATA_FILENAME=uClibc-locale-030818.arm.tgz CC=ccache arm-angstrom-linux-uclibcgnueabi-gcc -march=armv4t -mtune=arm920t May 14 09:50:24 | make[1]: `conf' is up to date. May 14 09:50:26 | CROSS COMPILE Badness: /usr/include in INCLUDEPATH: /usr/include/ May 14 09:50:28 | cc1: internal compiler error: in add_path, at c-incpath.c:362 May 14 09:50:52 Laibsch: no problem. May 14 09:51:36 sandworm: you did update, right? May 14 09:51:44 yes! May 14 09:51:44 Laibsch: I'm running out of space on my laptop now. I will bake the opie-image once I got more space May 14 09:52:07 2007-05-14 11:29 OE.mtn May 14 09:52:32 let me start a build May 14 09:52:56 xjqian: Maybe I can bake one for you May 14 09:53:03 Oh, koen already offered May 14 09:53:07 ask me what? May 14 09:53:09 That will be MUCH faster May 14 09:53:18 Laibsch: a build for ep93xx/uclibc May 14 09:53:19 also uclibc .config have a line "TARGET_alpha=y", not TARGET_arm May 14 09:53:30 I don't even have opie in my BBFILES :) May 14 09:53:45 ggilbert: Would it be possible to give me superpowers on bugzilla? May 14 09:53:51 oh, I imagine so May 14 09:53:54 koen: OK, I was surprised May 14 09:53:56 let me look May 14 09:54:06 xjqian: Be patient. I will get you an image May 14 09:54:26 also, I can hold off on bugzilla 3 if it'll break stuff for you guys May 14 09:54:38 I was just going to try doing a test upgrade on another machine May 14 09:54:38 ggilbert: I don't think so May 14 09:54:49 thanks for the offer May 14 09:54:52 I don't know how this will affect stuff like the build boxes May 14 09:54:53 An upgrade would be very much appreciated May 14 09:54:53 I have to run May 14 09:55:16 sandworm: what does 'mtn automate heads' say for you? May 14 09:55:32 have to run, though. see you guys. and looking forward to the opie-image May 14 09:55:40 ggilbert: I think we'll welcome bugzilla 3 :) May 14 09:55:54 alright, then I'll get it done sometime this week May 14 09:56:18 Laibsch: what's your bugzilla username? May 14 09:56:34 sandworm: it should be 11c89795a895612815318e54f362f6cbe08afbe0 May 14 09:56:42 mickey|uni: guten morgen! May 14 09:56:48 koen, I'm assuming that since you told him to look me up, that you don't have any problems with him having access? May 14 09:56:49 Guten Morgen! May 14 09:56:56 ggilbert: correct :) May 14 09:56:59 hehe May 14 09:57:09 and I suspect mickey|uni will agree as well May 14 09:57:16 who... what? May 14 09:57:18 * mickey|uni out of the loop May 14 09:57:33 mickey|uni: laibsch, more bugzilla powers May 14 09:57:35 mickey|uni: About you owing me 2,000 USD May 14 09:57:40 or what koen said May 14 09:57:47 hah May 14 09:57:50 you are buying beer for everyone May 14 09:57:55 i'm broke... moving agency took all my money May 14 09:58:08 hehe May 14 09:58:20 take it back May 14 09:58:21 yeah, more bugzilla power for laibsch is ok May 14 09:58:23 moving far? May 14 09:58:44 not at all, but apparantly it took them very long to (de)assemble the sleeping room May 14 09:58:53 ah May 14 09:59:02 ggilbert: bugs.openembedded.org AT rolf.leggewie.biz May 14 09:59:08 anyway, since friday we're sleeping in the new flat May 14 09:59:32 mickey|uni: congrats May 14 09:59:59 Laibsch: you should be good to go now May 14 10:00:01 koen: nothing said :( May 14 10:00:10 ggilbert: Cool. Thanks a bunch May 14 10:00:52 Wow, all these nice new shiny Knöpfsche ;-) May 14 10:01:08 -> buttons May 14 10:01:17 * Laibsch hugs multimachine May 14 10:01:21 sandworm: it looks like you're missing http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/uclibc/uclibc-0.9.29/ep93xx May 14 10:01:30 xjqian: Your tosa build is on the way May 14 10:01:56 ggilbert: thanks again for maintaining our bugzilla May 14 10:02:05 koen, glad to help. May 14 10:03:09 ~lart PC pool workstations May 14 10:03:09 * ibot pulls out a ClueBat (tm) and thwaps PC pool workstations May 14 10:09:25 koen: compiled, thanx. but why i not see this file in 3-hours-old database ? May 14 10:10:13 sandworm: because I added it ~2.5 hours ago :) May 14 10:12:06 koen :) May 14 10:13:36 koen: what target should i tell bitbake to make minimal router image May 14 10:27:50 03koen 07org.oe.dev * rc3843c7f... 10/ (1 packages/bluez/bluez-libs_3.10.bb): bluez-libs: update to 0.3.10 May 14 10:32:13 bbiab, dinner May 14 10:32:16 lunch, even May 14 10:33:53 koen: NOTE: package linux-2.6.21-r0: task do_unpack: started May 14 10:33:53 NOTE: Unpacking openembedded/sources/linux-2.6.21.tar.bz2 to openembedded/build/tmp/work/ep93xx-angstrom-linux-uclibcgnueabi/linux-2.6.21-r0/ May 14 10:33:53 NOTE: Unpacking openembedded/org.openembedded.dev/packages/linux/linux/defconfig to openembedded/build/tmp/work/ep93xx-angstrom-linux-uclibcgnueabi/linux-2.6.21-r0/ May 14 10:33:53 cp: невозможно выполнить stat для `openembedded/org.openembedded.dev/packages/linux/linux/defconfig': No such file or directory May 14 10:33:57 NOTE: Task failed: May 14 10:34:12 that looks like a bug in OE May 14 10:34:39 koen: now kernel build fails May 14 10:38:21 yes, because it's the wrong kernel May 14 10:38:29 I think I know why May 14 10:38:54 hmmm May 14 10:39:04 * koen scratches head May 14 10:39:20 it should build the .19+.20rc7 kernel May 14 10:43:09 koen: Done. May 14 10:43:41 I added a disabled-by-default bluez-utils 3.10 to OE, and I'm now going to try to get that in shape May 14 10:44:02 do you have a sample initscript? May 14 10:44:05 I am working on the perfect configure line for you. May 14 10:44:19 03koen 07org.oe.dev * r05b9706d... 10/ (1 packages/bluez/bluez-utils_3.10.bb): (log message trimmed) May 14 10:44:19 bluez-utils: groundwork for 0.3.10 recipe - DO NOT USE FOR PRODUCTION May 14 10:44:19 Since this was a big update from 0.3.9: May 14 10:44:19 * DEFAULT_PREFERENCE = "-1" May 14 10:44:19 * the TI initscript patch doesn't apply - needs updating (affected machine: hx4700, htcuniversal and others) May 14 10:44:20 * The "any" -> BCSP patch for serial cards was declared crap by Marcel Holtmann and thus removed May 14 10:44:24 * The initscript *needs* updating May 14 10:44:35 ah, there it is May 14 10:45:39 koen: and how can i tell to build .19xxxx kernel for angstrom ? May 14 10:46:11 koen: apologies for the erroneous removal of ixp4xx-kernel 2.6.21 - the next commit should reinstate it. May 14 10:46:48 rwhitby: no problem, I just wanted to highlight it for other kernel maintainers May 14 10:46:51 and thank you for handling the error graciously. May 14 10:47:34 I'll try to be even more gracefull in the future May 14 10:48:20 no, any more graceful and I wouldn't learn the lesson from the mistake. May 14 10:49:46 ggilbert: are you fiddling with bugzilla? May 14 10:49:52 yeah May 14 10:49:56 I see nothing but white pages May 14 10:50:07 oh Odd. May 14 10:50:14 I thought I put up a maintenance message May 14 10:50:32 OK May 14 10:50:36 oh May 14 10:50:42 I think I know why that doesn't show :) May 14 10:50:48 ;-) May 14 10:50:52 rwhitby: EABI/EB works, EABI/EB-uclibc builds, but is untested May 14 10:51:00 ... because I'm fiddling with bugzilla :) May 14 10:51:35 I'll see if google has it May 14 10:52:23 koen: I'll have to try a mokoslug eabi/be-uclibc build then ... May 14 10:52:42 ~lart OSX keyspan driver May 14 10:52:43 * ibot drops a truckload of VAXen on OSX keyspan driver May 14 10:52:59 I can only interrupt redboot on my avila using a linux computer... May 14 10:56:44 holtmann: is scripts/bluetooth.init up to data? May 14 10:57:50 koen: In the CVS, yes. For the release, I forgot it. Preparing a bluez-utils-3.10.1 with fixes right now. May 14 11:02:36 ggilbert: Please let me know when you are done May 14 11:09:52 should just be a minute or two May 14 11:13:01 ok May 14 11:13:04 you should be good to go May 14 11:13:37 Yes, back alive May 14 11:13:44 Is that already the new bugzilla? May 14 11:13:49 Or just a template? May 14 11:13:49 yup May 14 11:13:52 with the expected css breakage May 14 11:13:53 COOL May 14 11:13:58 Good job! May 14 11:14:00 and html guru's in da houz? May 14 11:14:02 there shouldn't be much css breakage May 14 11:14:35 this time the templates didn't entirely shit themselves May 14 11:14:36 ah May 14 11:14:42 * koen stabs browser cache May 14 11:14:58 oh I think I do see some May 14 11:16:19 Laibsch: The bugzilla guys actually do a good job at making updates painless May 14 11:16:34 Laibsch: they'd probably be even easier if I learned how to handle templates properly May 14 11:18:15 koen: I am doing a bluez-utils-3.10.1 today to make it easier for OpenMoko. May 14 11:26:09 holtmann: should bluetooth.init be used as-is, or is it recommended to edit it? May 14 11:26:19 (the one in cvs/3.10.1) May 14 11:27:30 Can be used as is. With the 3.10 you only need to start "hcid -s". Maybe you wanna use "hcid -s -x" to enable some experimental options. May 14 11:28:21 Experimental means that I might change the API. May 14 11:28:51 any docs on what -x does right now? May 14 11:29:07 * koen is too lazy to ask lots of questions May 14 11:29:38 It enables the D-Bus methods marked with [experimental] in dbus-api.txt May 14 11:30:01 Mainly the additional authorization framework. May 14 11:32:19 heh, familiar names in the copyright statements :) May 14 11:34:22 holtmann: any idea on putting back bluez2-compatible auth handler (process-based, not dbus-based)? May 14 11:36:15 psokolovsky: No. May 14 11:36:47 holtmann: no idea or no putting back? May 14 11:38:45 I am not putting it back, because it is not a clean approach. May 14 11:39:08 Normally the PIN requester and the PIN responder have to run in a different user context. May 14 11:39:09 koen: NOTE: DO PACKAGE QA May 14 11:39:09 NOTE: Checking Package: update-modules-dbg May 14 11:39:09 NOTE: Checking Package: update-modules May 14 11:39:09 ERROR: Can not check RPATH, scanelf (part of pax-utils-native) not found May 14 11:39:28 sandworm: please upgrade to bitbake 1.8.2 May 14 11:39:47 sigh. people just forget what unix is, with all these leenookses May 14 11:40:09 koen: also build process periodically interrupts witj this message. restart continues ok. May 14 11:40:12 sandworm: I'm still trying to findout why it bulds the wrong kernel May 14 11:41:41 koen: have you fix a bug with shasum hangs on 1.8.2 ? May 14 11:41:53 yes May 14 11:46:38 should i update db again now ? so i still see do_fetch hangup on 1.8.2 May 14 11:47:36 sandworm: do me a favour and try bitbake nano >/dev/null May 14 11:47:50 das that fetch and compile for you? May 14 11:48:00 >/some/file wouls work as well May 14 11:49:18 * CoreDump|afk shudders at his shoddy English writing skills May 14 11:49:41 Hello, I'm making a image by installing ipkgs into the rootfs, but now I'm confronted with the fact that I can't use ipkg -force-overwrite. I tried to use IPKG_ARGS=="-force-overwrite", but that doesn't do it. Any suggestions? May 14 11:49:58 += ofcourse May 14 11:50:09 CoreDump|afk: bb1.8.2 hangs in do_fetch as usually. May 14 11:50:19 even with redirection? May 14 11:51:07 Pat75: well, what about fixing the packages that try to overwrite stuff? May 14 11:52:16 CoreDump|afk: Exactly, I just want a config file in /etc to be different for my image, but don't want to do this with sed during runtime on the sustem May 14 11:53:24 right, well, there are multiple ways to go about it. You could, for example, run sed at image-creation time. May 14 11:53:48 CoreDump|afk: oops! with redirection works May 14 11:54:14 sandworm: that's the same bug I see here. The BB uber-hackers are informed but have no solution yet May 14 11:54:46 Pat75: also, you could create a custom version of the package that contains said config file with your very own version of it May 14 11:55:03 anyway, gotta go to work, later all May 14 11:55:32 03rwhitby 07org.oe.dev * rc226efda... 10/ (5 files in 5 dirs): ixp4xx-kernel: reinstated 2.6.21 - apologies to all affected by my previous mistake May 14 11:55:34 What I did was a package that (r)depends on the original package, and create a local package that overrides the conf of the original one May 14 11:58:11 CoreDump| afk: Seems a bit silly to me to create a whole new package with only a diff in the config files. May 14 12:04:14 psokolovsky, RP, v8jlene: I'm going to apply the var-volatiles patch now May 14 12:06:43 koen: please do, I wondered myself where v8jlene is... May 14 12:06:54 in .au :) May 14 12:07:19 ;-) May 14 12:16:16 hi there May 14 12:16:45 does anyone of you have experience with oe on a nokia n800 and sound integration May 14 12:17:36 03lenehan 07org.oe.dev * rd8f8ffd5... 10/ (38 files in 38 dirs): basefiles, initscripts: make /var non-volatile, as discussed on the mailinglist May 14 12:19:05 psokolovsky: there you go May 14 12:19:29 thanks! May 14 12:21:28 holtmann: the /var issues has been fixed now May 14 12:22:35 What issues? May 14 12:24:54 holtmann: that /var was tmpfs May 14 12:25:45 hi May 14 12:28:41 koen: Ah. That one. Good. May 14 12:56:50 hi all May 14 12:59:06 i got a lot of error "ERROR: Could not inherit file classes/rootfs_tar.bbclass while parsing /media/spare/oe/org.openembedded.dev/packages/images/twin-image.bb" and don't know why ... it happen on a fresh installation of OE with bitbake 1.8.3 revision 817 ... May 14 13:06:24 hi pH5 May 14 13:06:42 hi gremlin[it] May 14 13:06:50 hi, somebody with e17 on Collie? May 14 13:07:30 i got a probably stupid problem with OE ... i got a lot of ""ERROR: Could not inherit file classes/rootfs_tar.bbclass while parsing ...zzxxdd.bb" ... May 14 13:15:05 koen, success loading code into my SFFSDR board May 14 13:15:25 koen, now I need to find out how to setup the DDR interface May 14 13:20:04 ho May 14 13:24:08 Crofton: cool May 14 13:24:13 zecke: hey May 14 13:27:01 CoreDump|afk: Something is wrong with your svn server. Can you please fix it? Thank you May 14 13:27:16 * zecke has headache May 14 13:27:29 koen: did you cross compile icu yet? May 14 13:27:40 Laibsch: http://lists.linuxtogo.org/pipermail/openembedded-commits/2007-May/005485.html May 14 13:27:49 zecke: I didn't see a commit :) May 14 13:28:30 koen: fixing icu will be a torture, I will give you 5$ if you do it May 14 13:29:02 koen: it has a bunch of native apps... and you can imagine the rest May 14 13:29:13 that's like €0.10! May 14 13:29:29 * Crofton kicks koen May 14 13:29:35 :) May 14 13:29:46 koen: that much? May 14 13:29:56 3.7 euro May 14 13:30:18 Crofton: at ELC the Robs responded to every dollar amount with "That's like 10p" :) May 14 13:30:28 heh May 14 13:30:58 =) May 14 13:31:15 * Crofton remembers when the pound was $4 May 14 13:31:16 * koen plays a bit more with the wiimote May 14 13:31:45 koen: cool, i havn't got to play with mine much yet. :( May 14 13:31:48 koen: if you add it I will integrate the rss reader with webkit and probably look into setting up gprs as well May 14 13:32:14 zecke: I don't even know what icu is May 14 13:32:24 koen: icu-project.org, unicode lib May 14 13:32:42 crap May 14 13:32:48 isn't that like 16MB? May 14 13:33:08 sourcecode? I don't think so May 14 13:33:32 Hmm, they have wild donkeys on Sardinia May 14 13:35:09 zecke: I'll have to wait for prints to flatten anyway May 14 13:36:47 zecke: looks like MIT license, correct? May 14 13:37:22 koen: looks like, but they name it IUC License 1.5 May 14 13:37:42 it even has autofoo May 14 13:37:59 koen: yes, and they cross compile and run the executable :) May 14 13:41:20 koen: slimming webkit will be an issue :} May 14 13:41:32 does someone have build andimage for quemux86 ??? May 14 13:42:25 zecke: I've been thinking about qemu.bbclass and having a qemu_do_compile() May 14 13:42:54 koen: we could contribute to scratchbox2 May 14 13:43:03 heh, no May 14 13:43:22 koen: or we could teach gcc to have the native and arm backend compiled in May 14 13:43:29 koen: and by pure magic select the right one May 14 13:43:42 koen: or we use LLVM and only on packaging we compile into a binary May 14 13:43:53 this class would use the gcc instead of gcc-cross May 14 13:44:02 inside a chroot that's built on demand May 14 13:44:18 NOTE: package icu-3.6-r0: task do_compile: started May 14 13:44:32 koen: I was as far as that ;) May 14 13:44:39 /bin/sh: ../bin/pkgdata: cannot execute binary file May 14 13:45:00 hmm May 14 13:45:12 * koen spots atmel.com email address on the list May 14 13:45:12 koen: well I had a do_configure() { oe_runconf } May 14 13:45:19 morning May 14 13:45:29 do_configure() { libtoolize --force oe_runconf } May 14 13:45:32 hey hrw May 14 14:00:26 zecke: NOTE: package icu-native-3.6-r0: task do_build: completed May 14 14:00:42 zecke: can we use those tools, or does it really needs arch specific stuff? May 14 14:02:22 dunno May 14 14:03:42 !oebug 2296 May 14 14:03:52 anyone get a chance to try out my java updates yet? May 14 14:03:59 * * Bug 2296, Status: NEW, Created: 2007-05-10 06:17 May 14 14:04:00 * * : Java updates May 14 14:04:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2296 May 14 14:04:34 oh-oh, is bugzilla already updated to 3.0? it looks funny May 14 14:05:01 apaulsen: IIRC, one patch was already committed, others in q yet... May 14 14:06:14 psokolovsky: yeah, I saw that the trivial ecj-native was committed :) May 14 14:06:58 psokolovsky: looks like lack of css May 14 14:07:34 psokolovsky: speaking of defconfigman - how much work is needed to get it support zaurus machines? May 14 14:10:32 hrw: well, as you see, I don't haste with it. I remember your request to add it to OE, but I want to do that gardually, so people were able to handle (and to be sure it does its job well). So, I'm going to handle HH.org ports with it, and that will be real good test of how well it does its job. As for Z, I don't expect too much trouble with adding draft support, though making sure it fits well (==optimal) both z and pocketpcs may take more re May 14 14:10:32 factoring. May 14 14:13:14 psokolovsky: I will check it when will find time for it May 14 14:13:21 hrw: ok! May 14 14:19:10 koen: did you checked does powertop works on arm? May 14 14:23:07 do13: did you checked wlan-ng 0.2.8? May 14 14:28:18 cu May 14 14:28:26 too much lag May 14 14:29:31 hrw|gone: not yet, since I don't now if the kernels have the right options on May 14 14:29:44 koen: rather not May 14 14:29:47 bye May 14 14:32:37 doesn't powertop on acpi c-states support? May 14 14:32:50 I've just compiled a kernel with TIMER_STATS enabled, but I fear that won't be enough. May 14 14:33:05 s/on/depend on/ May 14 14:45:49 03koen 07org.oe.dev * r370fefec... 10/ (3 files in 2 dirs): May 14 14:45:49 bluez-utils 3.10.1: configure and package according to upstream guidelines May 14 14:45:49 * still DEFAULT_PREFERENCE = -1 May 14 14:50:30 does one know how I could request the ohci status / last error? May 14 14:55:02 re May 14 14:55:53 keesj: at a guess, via /sys or /proc May 14 14:57:14 hmm, if I want to depend on a certain version of a package, how would I do that? I tried setting PREFERRED_VERSION_directfb = "0.9.24" in my task.bb file, but it did not help May 14 14:58:14 koen, http://www.balister.org/~balister/sff_sdr/I May 14 15:02:16 Crofton: thanks May 14 15:04:43 koen: lmsnesor's is compiling for you? May 14 15:04:52 zecke: no May 14 15:05:51 let me commit the 'fix' May 14 15:12:26 03koen 07org.oe.dev * rddf317b6... 10/ (1 packages/tasks/task-openmoko.bb): task-openmoko: remove unused lmsensors May 14 15:12:44 koen: hehe May 14 15:13:03 koen: why don't apply the lmsensor iconv.patch only when building for uclibc? May 14 15:13:38 dunno May 14 15:26:04 Petachi! May 14 15:34:03 zecke: that sounds like a good idea May 14 15:57:06 zecke: ping May 14 15:57:36 pong May 14 15:57:46 sorry, I found my answer May 14 15:57:51 TARGET_OS is in overrides May 14 15:58:24 What is in OVERRIDES? (zecke is da jeopardy king) May 14 16:25:07 03koen 07org.oe.dev * r14ec4d45... 10/ (1 packages/gdb/gdb_6.6.bb): gdb: only RRECOMMEND glibc stuff when building for glibc, the check isn't optimal May 14 16:25:11 03koen 07org.oe.dev * r7aa8e108... 10/ (1 packages/bluez/bluez-utils_3.10.1.bb): bluez-utils 3.10.1: move hciattach to main package till hcid can handle that functionality May 14 16:27:24 koen: is bootstrap-image a supported angstrom image? May 14 16:27:39 no May 14 16:28:08 right now the only images coming close to support are angstrom--image May 14 16:28:56 03koen 07org.oe.dev * rafd93002... 10/ (1 packages/bluez/bluez-utils_3.10.1.bb): bluez 3.10.1: ~lart linebreaks May 14 16:29:10 koen: luckily ansgtrom-console-image fails with the same error. avahi-autopid runtime build target not found May 14 16:29:12 RP: is the poky libtool change safe to merge? May 14 16:29:18 koen: building angstrom in uclibc mode :) May 14 16:30:04 koen: Yes, as long as you change base.bbclass too May 14 16:30:32 koen: I've been planning to ask about merging that one as its quite subtle and nasty... May 14 16:31:00 koen: and has existed since libtool was added to OE as I understand it, it goes way back in bitkeeper May 14 16:31:26 RP: I rebuild libtool-cross when changing machine, so I know of the troubles May 14 16:31:58 koen: I don't believe in having to rebuild anything ;-) May 14 16:32:05 koen: You mean libtool-native, right? May 14 16:32:10 * koen neither May 14 16:32:16 RP: no, libtool-cross May 14 16:32:33 otherwise you get bizarre libtool failures during builds May 14 16:32:35 koen: Hmm. This fixes a problem with libtool-native installing into cross dirs May 14 16:32:59 Proving my view that most uses of := are bugs waiting to happen May 14 16:33:16 right May 14 16:36:16 koen: did you recently build angstrom for uclibc? May 14 16:36:24 zecke: let me commit some more local diff :) May 14 16:36:27 zecke: yes May 14 16:36:40 koen: Some of the libtool-cross failures should have been fixed by the additional libtool-cross patches I added May 14 16:36:43 zecke: http://ewi546.ewi.utwente.nl/tmp/Angstrom-minimalist-image-test-20070513-hx4700.rootfs.tar.gz May 14 16:37:14 koen: awesome May 14 16:37:35 1MB :) May 14 16:37:39 and it works as well May 14 16:37:55 koen: I want to run the angstrom-console-image in my neo emulator (okay now you will shoot me) May 14 16:38:04 03koen 07org.oe.dev * rdf134424... 10/ (1 conf/distro/include/angstrom-uclibc.inc): angstrom-uclibc: turn on NLS for glibc May 14 16:38:32 zecke: that should work, only the system emu needs the nptl patches May 14 16:40:38 koen: I forgot it, but what is in the 1mb image? May 14 16:40:49 taskboot + dropbear May 14 16:40:49 koen: networking scripts at least? X11? May 14 16:41:28 sysvinit, tinylogin, busybox, dropbear and some kernel modules May 14 16:42:54 koen: it starts to build May 14 16:43:01 * zecke puts a coin into koen's head May 14 16:45:14 time to make some food May 14 16:46:56 koen: you made libtool-native fail (but that is likely my issue May 14 17:04:39 koen, you did catch the link to pictures of the sff sdr board? May 14 17:20:22 later May 14 17:22:07 Packaged contents of glib-2.0-dev into /data/build/koen/OE/build/tmp/angstrom/deploy/uclibc/ipk/armv4t/libglib-2.0-dev_2.12.12-r0_armv4t.ipk May 14 17:22:15 hey koen May 14 17:22:26 Crofton|laptop: yes, I saw them May 14 17:22:30 hey mreimer_ May 14 17:28:22 03koen 07org.oe.dev * rec070b43... 10/ (3 files in 3 dirs): gettext: fix building against uclibc May 14 17:49:10 has anyone heard/used "code sourcery"? hear about it for the first time, is it any good? just curious May 14 17:49:39 looks like some commercial stuff - at least thats my impression from the webpage May 14 17:49:53 CSL toolchains suck May 14 17:50:04 took us a few months to track that down May 14 17:50:12 Jin^eLD: I think it the defacto standard for exitoc gcc's May 14 17:50:23 exitoc? May 14 17:50:43 exotic May 14 17:51:10 aah, I see.. hmm, well, I read ARM there, doesn't sound that exotic :) May 14 17:51:23 :P May 14 17:51:26 well I guess they have more.. May 14 17:51:31 for me it is :p May 14 17:51:45 :> May 14 17:52:40 but again I am quite new to the embeded world i have done moostly java&linux&opensource the last 8 years May 14 17:53:17 the CSL toolchains have a lot of black magic patching May 14 17:53:40 but what is worse, CSL employees are the release managers for gcc and glibc nowadays May 14 17:53:44 keesj: about a year ago I was new too, but then it turned out - it's not sooo much different, your software just runs a little slower :) the rest did not look way to different to me May 14 17:53:57 really? May 14 17:54:00 doh.. May 14 17:54:20 so GNU guys gave that away? or how did it turn out to be like that? May 14 17:54:24 mark mitchel, paul brook, etc are all CSL employees May 14 17:54:31 they seem to be doing a good job May 14 17:55:04 I see, well, I guess then this sourcery stuff shouldn't be too bad May 14 17:55:27 we had the csl toolchain in OE and stuff would just randomly segfault May 14 17:55:59 anyway, I was just curious, not planning to use it; just heard the name and wondered what it was May 14 17:56:09 maemo is using it May 14 17:58:13 Jin^eLD: for me it is really a differnt world, here people talk about little problems(byte order,16bpp,deps,drivers) May 14 17:59:13 and the commmunity is also very different I think May 14 17:59:30 and I love it! May 14 18:00:04 keesj: I am far away from coding stuff that is close to hardware or doing kernel hacking, I'm usually on application level :) but the move from x86 to ARM or mipsel was not really hard May 14 18:00:19 I guess because OE did the whole work :) May 14 18:00:45 it really saves you a lot of trouble May 14 18:00:54 close to hardware and at application level? May 14 18:01:21 keesj: stuff that I code compiles on x86 or ARM without any changes, or maybe just little changes May 14 18:02:27 hmmm May 14 18:02:44 why does a glibc compile take 3 weeks and a uclibc compilation 3 seconds? May 14 18:02:55 keesj: and thats very different to say, writing kernel modules or something :) so I guess it comes down to what you do May 14 18:03:45 koen: maybe glibc is using some of that patented M$ stuff :) May 14 18:04:18 koen: getting nearer to switch x11 image to uclibc then :-D May 14 18:25:16 what is the best way to install gdb-cross on a host system? A gdb-cross ipk is generated, but nothing is placed in staging. May 14 18:26:27 channel May 14 19:10:36 Hello :D May 14 20:00:55 03koen 07org.oe.dev * rba7dc233... 10/ (3 files in 3 dirs): uclibc 0.9.29: turn on SUSV3 stuff on angsgrom to build perl May 14 20:00:55 03koen 07org.oe.dev * rba7dc233... 10/ (3 files in 3 dirs): uclibc 0.9.29: turn on SUSV3 stuff on angsgrom to build perl May 14 20:00:55 03koen 07org.oe.dev * rba7dc233... 10/ (3 files in 3 dirs): uclibc 0.9.29: turn on SUSV3 stuff on angsgrom to build perl May 14 20:00:59 03koen 07org.oe.dev * rba7dc233... 10/ (3 files in 3 dirs): uclibc 0.9.29: turn on SUSV3 stuff on angsgrom to build perl May 14 20:01:39 CIA-4: stop stuttering May 14 20:02:33 koen: libiconv failed to build :) May 14 20:02:46 03koen 07org.oe.dev * rba7dc233... 10/ (3 files in 3 dirs): uclibc 0.9.29: turn on SUSV3 stuff on angsgrom to build perl May 14 20:03:01 it built over here May 14 20:03:02 03koen 07org.oe.dev * rf05d7ffb... 10/ (1 packages/install/install-native_0.1.bb): install-native: inherit native May 14 20:03:08 perl doesn't build however May 14 20:03:09 03koen 07org.oe.dev * r990392a7... 10/ (1 packages/gimp/gimp_2.3.10.bb): gimp 2.3.10: fix packaging, closes #2313 May 14 20:15:25 Sorry for being off-topic. /dev/shm has run full at 315MB. I admit to not knowing what this is for and how to reduce the usage. HELP! So far no obvious issues in running the box. But this cannot be healthy May 14 20:20:52 Laibsch: do you use tmpfs for log files? May 14 20:22:04 koen: gettext is failing now May 14 20:22:25 zecke: up to date OE repo? May 14 20:22:43 koen: well, which OE repo? May 14 20:22:56 the one and only .dev May 14 20:23:36 zecke: yes, this is tmpfs May 14 20:23:51 And there was more logging recently May 14 20:24:04 45M of logs May 14 20:24:17 But I will see if cleaning out /var helps May 14 20:24:25 Laibsch: so where is the other 250MB from? May 14 20:24:32 zecke: you need revision ec070b4345b9870ee77f712ad166b28c7ddf7f73 May 14 20:24:34 which device is that? is that your via? May 14 20:25:11 zecke: I don't know /dev/shm is empty May 14 20:25:22 And the logs were not really on tmpfs May 14 20:25:27 zecke: http://lists.linuxtogo.org/pipermail/openembedded-commits/2007-May/005502.html May 14 20:26:17 mickey|moving: so many people are popping up for WebKit/Gdk hacking, I could have done by tinderbox stuff... May 14 20:26:55 that's good news May 14 20:27:11 koen: well, let us wait and see :) May 14 20:27:18 ~lart Nokia just because... May 14 20:27:50 zecke: http://rafb.net/p/PcLw8P45.html May 14 20:27:54 * ibot brandishes Excalibur! "With this sword, I vanquish thee, Nokia just because...!" and lops off Nokia just because...'s head May 14 20:28:04 ggilbert: Are we eventually going to see a better layout in bugzilla? Is this a CSS problem? May 14 20:28:43 ~lart me May 14 20:28:43 * ibot duct-tapes laibsch to the floor and drools on him May 14 20:28:49 zecke: Nevermind May 14 20:29:23 zecke: Wrong column ;-) May 14 20:30:03 but, var was tmpfs? May 14 20:30:46 not var itself May 14 20:31:04 www.heise.de May 14 20:31:05 but /var/run and /var/lock May 14 20:31:08 argh May 14 20:34:57 zecke: which one? the one about dvb-h? May 14 20:36:37 hvontres|poodle: context? May 14 20:37:08 I guess the nokia lart May 14 20:37:38 zecke: www.heise.de May 14 20:38:05 hvontres|poodle: focus issue with X11 on OSX ;) May 14 21:00:21 psokolovsky_: did you patch ipkg? I'm reading logs, and it's looking like it May 14 21:01:22 ~lart new version of ipkg for angstrom May 14 21:01:23 * ibot takes a big bite out of new version of ipkg's jugular vein for angstrom May 14 21:05:38 summatusmentis: what's wrong with ipkg for angstrom? May 14 21:06:02 I think koen put it something like "ipkg has gone from slow and buggy to plain buggy" May 14 21:06:43 NAiL: might be my machine, I'm getting lots of errors, but I'm troubleshooting it w/ hvontres|poodle in another channel May 14 21:08:14 errors about a fork failing before execution May 14 21:08:35 hmm, ok May 14 21:11:17 NAiL: it was memory, for some reason, my machine was running out of ram May 14 21:15:07 hi all May 14 21:17:41 hi May 14 22:08:44 RP: ping May 14 22:08:51 CoreDump|home: pong May 14 22:09:00 got your mail May 14 22:09:15 speed 38400 baud; line = 32; May 14 22:09:15 tostop May 14 22:09:41 CoreDump|home: Try turning tostop off May 14 22:11:03 tostop isn't listed in the man page I looked at hence I couldn't remember what it was when I'd seen this before May 14 22:12:00 RP: sure, give me a second while I svn up bb 1.8 and reparse May 14 22:15:02 RP: wow, that did the trick May 14 22:15:33 CoreDump|home: :) May 14 22:16:00 CoreDump|home: Can you post that to the mailing list so we remember what its name is :) May 14 22:16:03 I guess every debian user needs to add stty -tostop to his shell env May 14 22:16:19 right away =) May 14 22:16:52 speed 9600 baud; line = 0; May 14 22:16:52 -brkint ixany May 14 22:16:53 -echok May 14 22:16:57 on my debian box May 14 22:17:27 speed 38400 baud; line = 0; May 14 22:17:29 eol = M-^?; eol2 = M-^?; swtch = M-^?; May 14 22:17:29 ixany iutf8 May 14 22:17:35 on mine May 14 22:17:48 heh May 14 22:18:09 I've only had that problem on one of my slackware boxes. The other slackware box was fine... May 14 22:18:30 I didn't remember what I'd changes though, typically... May 14 22:19:00 but I recongised it instantly when I saw CoreDump's output :) May 14 22:19:50 0D May 14 22:19:52 =D May 14 22:21:39 awesome :) May 14 22:21:44 so can we bump the req? May 14 22:22:32 no ;) May 14 22:22:56 zecke: It looks like we have an interesting problem with the inherits_cache? May 14 22:23:14 RP: did you see polynomous's patch? May 14 22:23:37 CoreDump|home: Note that if the terminal has the TOSTOP flag set (i.e. "stty tostop"), May 14 22:23:40 a background process which tries to write to the terminal May 14 22:23:42 will be suspended, which is undesirable. May 14 22:24:02 CoreDump|home: why in the world would you WANT that flag set ????? May 14 22:24:09 I didn't set it May 14 22:24:20 I never heard of it before =) May 14 22:24:37 My jail was set up this way by debootstrap May 14 22:24:53 CoreDump|home: ahhh, for a jail that might make sense....:) May 14 22:24:56 zecke: Whats his real name? :) May 14 22:25:16 michael krelin May 14 22:25:19 CoreDump|home: BTW: http://www.intevation.de/rt/webrt?serial_num=3709&display=History May 14 22:25:26 Right, that's the patch I was looking at May 14 22:25:39 RP: koen beat me May 14 22:25:57 zecke: the bruises will fade ;) May 14 22:26:23 zecke: I've now seen it, I've not thought about what it does. Looks like there is a bug though and it would explain a few things May 14 22:26:43 RP: I think the patch is sane. I still would like to see the circle it went May 14 22:27:01 RP: because I know that this code can detect certain circles May 14 22:27:20 zecke: Yes, I think we need some debug output before and after to compare May 14 22:27:31 zecke: 1.8 has the problem too? May 14 22:27:43 RP: yes, the same code with a small hunk May 14 22:28:20 "I want a log" is a lazy way to say to have not enough time to test myself May 14 22:28:50 zecke: I can't test now but now I know about it, I can add it to the todo list :) May 14 22:31:09 RP: which change to base.bbclass is needed for the libtool merge? May 14 22:33:05 libtool merge? May 14 22:33:13 will this fix rpath crap? May 14 22:33:17 dunno May 14 22:33:25 it will fix other libtool crap :) May 14 22:33:35 heh May 14 22:34:34 ngnng rpath May 14 22:34:38 at the next OE developer gathering I want to discuss our toolchain maintenance May 14 22:35:23 libtool does bad things when building cross, but we can not de libtool packages ... May 14 22:35:34 so we must fix libtool May 14 22:36:04 okay, I will go to bed and leave this code unfinished May 14 22:40:33 'night zecke May 14 22:42:02 koen: hey i heard opemoko is now building without alphs blending Are they using/gonna use sapwood now? May 14 22:42:25 dunno May 14 22:42:30 :/ May 14 22:42:37 I haven't seen any csets concerning alphablending May 14 22:43:06 non-alphablended theme May 14 22:43:09 robtaylor: if you are thinking me, its because Im not using openmoko May 14 22:44:12 XorA: have i got my channels confused? =) May 14 22:45:03 robtaylor: I am running Angstrom on Neo, and on #openmoko I announced it was a lot faster May 14 22:45:09 ahh May 14 22:46:28 so is openmoko still in svn, or is it seperate distro in oe? May 14 22:46:38 * robtaylor really hasn't een following things May 14 22:47:15 robtaylor: OE has a version of openmoko, they have their own version in svn, people tend to argue May 14 22:48:27 gngngng May 14 22:48:38 OE fixes bugs in minutes, openmoko in weeks May 14 22:49:04 they openmoko people accuse OE people of sabautage May 14 22:49:17 why? May 14 22:49:31 openmoko follows the "apply-revert" method, OE the "test-apply-smile" method May 14 22:49:42 lack of intelligence seems to be the main problem May 14 22:50:03 somehome OE HEAD can break a build based on a fixed revision in the past May 14 22:50:15 somehome? May 14 22:50:19 * XorA shrugs, but I think that guy was just a moron May 14 22:50:31 with his own agenda May 14 22:50:38 s/moron/anti-OE troll/ May 14 22:50:48 there should really just be a branch in monotone May 14 22:51:17 its just a shame monotone's so slow at branch switching =) May 14 22:51:18 robtaylor: thats effectively what they have, they pick a fixed revision and branch from that in svn May 14 22:51:27 robtaylor: so nothing we do can effect that May 14 22:51:35 XorA: well,that's not really the same, is it May 14 22:51:40 and since they don't accept help.... May 14 22:51:50 yeah, this is the biggest issue May 14 22:52:04 robtaylor: the same effect, we cant remove that revision from the database or change it May 14 22:52:07 and mickeyl's been silent for ages now May 14 22:52:26 and the "we developed this patch against bitbake in secreat, and only our developer may work on it before we allow you to see it" thing May 14 22:52:27 koen: The bit where it adds libtool-cross to the dependencies May 14 22:52:36 XorA: why would you change an existing revisio? May 14 22:52:41 koen: You need the libtool-native and libtool-cross changes too May 14 22:52:48 robtaylor: mickeyl is moving and without connection May 14 22:52:58 robtaylor: see above about them claiming we sabatague their project May 14 22:53:14 XorA: i still dont understand at all May 14 22:53:23 robtaylor: in order to do that it would require rewinding the whole OE, unstiching that revision, cracking SHA-1 May 14 22:53:56 XorA: i'mreally confused what you're actually talking about, tbh May 14 22:54:12 RP, have you looked at how libtool inserts bad rpaths? May 14 22:54:20 XorA: oh, you were accused of chaging an existing revision in monotone? May 14 22:54:21 morns May 14 22:54:43 robtaylor: yes, that is the claim that was made last night by as koen says an anti OE troll May 14 22:54:52 RP: s/base.bbclass/autotools.bbclass/ May 14 22:54:57 that would be difficult May 14 22:54:59 RP: http://svn.o-hand.com/view/poky?rev=1617&view=rev :) May 14 22:55:03 Crofton: No, I've so far managed to avoid looking at libtool in any detail at all ;-) May 14 22:55:06 anyway it is time for me to sleep May 14 22:55:29 cya May 14 22:55:30 RP, one of my background hobbies is to see if I can find a fix May 14 22:55:31 XorA:night :) May 14 22:55:41 koen: Ah, right. I meant autocoolt.bbclass ;-) May 14 22:55:51 I keep wishing someone else would fix it :) May 14 22:56:04 koen: Look for a later revision though as it took three goes before I managed the right magic May 14 22:56:47 Crofton: I've patched libtool for odd issues and koen is talking about some .bb sanitising I did but I don't dare look too deeply at it ;-) May 14 22:57:30 * koen skips http://svn.o-hand.com/view/poky/trunk/meta/classes/package.bbclass?rev=1619&view=diff&r1=1619&r2=1618&p1=trunk/meta/classes/package.bbclass&p2=/trunk/meta/classes/package.bbclass for the time being May 14 22:58:26 koen: I can't remember if you'll need that or not, I was just about to warn about that May 14 22:58:51 I think you might get away without but I won't swear to it ;-) May 14 22:59:38 koen: 1620 looks like the right version May 14 23:00:00 yeah, I have that May 14 23:00:09 koen: If its not right, bitbake will complain about circular dependencies May 14 23:00:38 http://rafb.net/p/CB1JJO35.html May 14 23:01:37 koen: keep the space after libtool-native May 14 23:02:02 ah, right :) May 14 23:02:46 http://ewi546.ewi.utwente.nl/tmp/libtool.diff May 14 23:03:09 koen: looks good May 14 23:03:55 bedtime for me... May 14 23:03:59 'night all May 14 23:04:06 gn May 14 23:04:10 CoreDump|home: Something is wrong with your svn server May 14 23:06:01 Laibsch: no, the svn fetcher of OE is bugged May 14 23:06:09 I've changed a path... May 14 23:06:24 n8 RP May 14 23:06:27 Laibsch: did you see the link to the commit I pasted today? May 14 23:06:54 RP, Crofton: http://lists.linuxtogo.org/pipermail/openembedded-devel/2007-May/002164.html May 14 23:19:16 has anyone got a copy of the Intel PXA27x Processor Family Developer's Manual? its gone from downloads.intel.com, and doesnt seem to be accessible from marvell yet :/ May 14 23:20:22 * wookey_ has supplied May 14 23:20:46 wookey_: nope, fraid that was the thermal and electrical :/ May 14 23:20:57 ah, yes - just noticed timestamp May 14 23:21:15 I'll mail bisset 0 he must have it May 14 23:21:26 wookey_: rock :) May 14 23:21:41 hey people! May 14 23:21:51 * philippe needs to fix the gtk-webcore bb files... May 14 23:21:53 hey philippe :) May 14 23:22:02 long time no see :) May 14 23:22:12 hi there robtaylor! & wookey_ May 14 23:23:23 hi May 14 23:23:31 I am in your home town May 14 23:23:48 philippe: whats going on in your world atm? :) i'm knee deep in scarily large changes to hal :/ May 14 23:24:23 * philippe is just dealing with hal :-) May 14 23:24:36 wookey_: you are in Tampere? May 14 23:25:06 no brussels May 14 23:25:28 wookey_: drink a trappiste for me :) May 14 23:25:40 IF you insist :- May 14 23:25:43 =) May 14 23:25:49 * philippe insists :-) May 15 00:00:13 koen: Which one of the thousand commits you made today? May 15 00:00:14 I usually don't follow your commits closely May 15 00:04:29 koen: In case you are talking about rm -Rf sources/altboot, I don't have such a dir. May 15 00:05:07 I only have two tarballs May 15 00:52:15 fwiw monotone.openembedded.org is down May 15 01:08:40 anyone else fail kismet compile building with angstrom and bitbake 1.8.2? May 15 01:08:52 1.8.3 I mean May 15 01:10:56 Zero_Chaos: I'll try. Will take a few minutes tho May 15 01:11:33 co thanks. sanity check on config.log fails for me. **** ENDING LOGGING AT Tue May 15 02:59:58 2007