**** BEGIN LOGGING AT Fri Oct 13 02:59:57 2006 Oct 13 03:53:38 * v8jlene is back Oct 13 06:53:53 03lenehan 07org.oe.dev * ra0475d48... 10/ (1 site/i486-linux): site i486-linux: Add required entry for p3scan. Oct 13 07:21:16 <_law_> any ideas why latest oz-dev images doesnt mount /home auto. on startup? if i do an mount -a after booting it works. Oct 13 07:21:45 _law_: anything in dmesg? Oct 13 07:25:54 morning Oct 13 07:26:10 _law_: mount -at doesnt work in busybox Oct 13 07:26:22 _law_: you might need util-linux-mount/umount installed Oct 13 07:32:09 <_law_> XorA, and mount -at is called on startup? Oct 13 07:32:12 <_law_> mount -a works Oct 13 07:33:49 _law_: mount -at nfs barfs on startup and stops the other mounts Oct 13 07:34:18 _law_: Angstrom speficially calls in util-linux-mount, util-linux-umount to get round this Oct 13 07:34:37 _law_: if you look at busy an OE patch to mount was not forward ported, this I suspect is the culprit Oct 13 07:35:36 03lenehan 07org.oe.dev * r24a7d797... 10/ (1 packages/clamav/clamav_0.88.4.bb): Oct 13 07:35:36 clamav 0.84: Don't stage the libraries libtool archive file, libclamav.la, Oct 13 07:35:36 as things fail to link when they try and use it. They link fine if it Oct 13 07:35:36 doesn't exist so remove it. Oct 13 07:36:37 03lenehan 07org.oe.dev * r05ae7db2... 10/ (10 files in 4 dirs): Oct 13 07:36:37 p3scan 2.9.05d: Add p3scan, a transparent proxy for POP3, POP3S and SMTP Oct 13 07:36:37 with virus scanning support. Default configuration uses clamav directly via Oct 13 07:36:37 libclamav. Oct 13 07:36:38 * Varoudis kalimera kosme Oct 13 07:37:10 <_law_> XorA, so whats the best solution? :-) Oct 13 07:37:27 <_law_> add patch / or install util-linux-mount/umoun t Oct 13 07:42:22 _law_: Well investigating that patch would be good, but in the interim install util-linux-(u)mount Oct 13 07:44:05 * XorA wonders why clamav.la fails Oct 13 07:53:42 <_law_> XorA, i?ll try to apply the patch... Oct 13 07:55:54 morning Oct 13 07:58:31 morning all Oct 13 08:07:19 <_law_> hmm busybox has got a total now mount.c :-( Oct 13 08:08:08 <_law_> s/now/new Oct 13 08:13:24 _law_: that'll be why patch was dropped then, at least we now know Oct 13 08:13:54 _law_: have to wait for 1.2.2 then, koen said there were a lot of mount fixes in svn Oct 13 08:15:43 <_law_> XorA, ok nice, i?ll install linux-utils meanwhile Oct 13 08:18:08 uptime: error while loading shared libraries: libproc-3.2.6.so: cannot open shared object file: Input/output error Oct 13 08:18:16 hmm, I think that machine might be dead :-) Oct 13 08:30:37 good morning Oct 13 08:31:01 hi florian_kc koen Oct 13 08:32:16 <_law_> hi florian_kc Oct 13 08:33:46 koen: started packaged-staging .dev build Oct 13 08:40:44 good morning all Oct 13 08:40:44 hrw|work: cool Oct 13 08:40:44 I should do the same sometime :) Oct 13 08:46:15 koen: stage-manager is only in packaged-staging? Oct 13 08:48:47 iirc, yes Oct 13 08:48:54 and the modified ipkg-build as well Oct 13 08:49:52 ;( Oct 13 08:51:00 hm.. checkout that branch or not.. Oct 13 08:52:30 fetched with wget Oct 13 08:55:39 koen: using p-s is more pain then usage Oct 13 09:01:34 good morning all. myfoobar-image-1.0-r0/temp/run.do_rootfs.11530: line 192: makedevs: command not found Oct 13 09:02:06 likewise: you wouldn't be running ubuntu, right? Oct 13 09:03:07 koen: Angstrom on 32bit works better Oct 13 09:05:17 koen: yes, is this a new issue? Oct 13 09:08:07 ~hail MACHINE="native" Oct 13 09:08:24 * ibot bows down to MACHINE="native" and chants, "I'M NOT WORTHY!!" Oct 13 09:08:46 likewise: ubuntu replaced 'bash' with 'dash' Oct 13 09:08:46 and that breaks Oct 13 09:09:03 btw, WTF is going on with the topic? Oct 13 09:10:31 * XorA hasnt seen problems on ubuntu yet Oct 13 09:12:48 building makedevs-native fixes it for me now... I wonder if I already have the dash replacement then... Oct 13 09:13:50 /bin/sh -> bash, bash --help: GNU bash, version 3.1.17(1)-release-(x86_64-pc-linux-gnu) Oct 13 09:13:58 koen: glibc 2.3.5 fails: | checking installed Linux kernel header files... TOO OLD! Oct 13 09:14:22 hrw|work: bump native to glib 2.5? Oct 13 09:14:23 likewise: ok, must be something else Oct 13 09:14:24 koen: linux-libc-headers were build Oct 13 09:14:34 hrw|work: no idea on that Oct 13 09:14:41 likewise: this bug is from other build Oct 13 09:14:53 likewise: machine=native use host libc Oct 13 09:15:11 do we already have linux headers exctracted from the kernel source tree? Oct 13 09:15:36 hrw|work: yes I saw the native machine, didn't know it uses the host libc though... Oct 13 09:16:55 will try to build 2.6.18 l-l-h Oct 13 09:20:01 03koen 07org.oe.dev * r5dbaa074... 10/ (1 conf/machine/htcuniversal.conf): htc-universal: convert to task-base Oct 13 09:22:41 03mickeyl 07org.oe.dev * rbd364f8b... 10/ (4 files in 2 dirs): appweb: refactor get_appweb_host function out into common .inc Oct 13 09:22:50 03mickeyl 07org.oe.dev * rbbeec593... 10/ (6 files in 3 dirs): matrixssl: remove 1.0-beta, refactor get_os function out into common .inc Oct 13 09:22:54 03mickeyl 07org.oe.dev * r39cd1690... 10/ (1 classes/sanity.bbclass): sanity.bbclass: loop over required utilities Oct 13 09:22:59 03mickeyl 07org.oe.dev * rf152f765... 10/ (5 files in 3 dirs): kbdd: remove obsolete 20040904 version, catch up with rename of linux_modules bbclass Oct 13 09:23:05 03mickeyl 07org.oe.dev * ra0629352... 10/ (4 files in 2 dirs): rename linux_modules to linux-kernel-base, add common kernel_get*version functions to this class Oct 13 09:23:11 03mickeyl 07org.oe.dev * r98967f30... 10/ (1 classes/base.bbclass): base.bbclass: make showing display revision work with the mtn 0.30 workspace format Oct 13 09:23:25 that should be it Oct 13 09:23:27 morning, btw. Oct 13 09:23:56 hey mickeyl Oct 13 09:24:03 hi mickeyl Oct 13 09:24:19 mickeyl: openembedded-private ML? Oct 13 09:24:47 ? Oct 13 09:25:03 (just been back from devdays without internet access) Oct 13 09:25:14 03koen 07org.oe.dev * r6f62cb71... 10/ (1 conf/machine/h5xxx.conf): h5xxx: convert to task-base Oct 13 09:25:20 03koen 07org.oe.dev * r3eca6e20... 10/ (1 conf/machine/htcuniversal.conf): htcuniversal: correct module names Oct 13 09:25:33 mickeyl: we need openembedded-private list Oct 13 09:25:37 ah Oct 13 09:25:45 and mailman hates me Oct 13 09:25:45 i think koen did something like that Oct 13 09:25:48 heh Oct 13 09:25:54 moderated subscription, archives for members only Oct 13 09:25:56 mailman is actually pretty decent Oct 13 09:25:59 mickeyl: yeah, mailman deleted that list *twice* Oct 13 09:26:02 oops Oct 13 09:26:05 let me handle that Oct 13 09:26:11 but not now Oct 13 09:26:13 on weekend Oct 13 09:26:14 ok Oct 13 09:26:18 *terribly busy* Oct 13 09:26:27 it is already weekend :) Oct 13 09:26:28 mickeyl: we have too much addresses in oedem thread Oct 13 09:26:38 hrw|work: please yank the vanille.de Oct 13 09:26:39 koen: but not first weekend in month Oct 13 09:27:07 mickeyl: your sig say 'm******@vanille.de' Oct 13 09:27:23 mickeyl: nice 302 to vanille-media.de Oct 13 09:27:40 mickeyl: where to yank it from? Oct 13 09:30:12 koen: from the too many addresses in oedem thread? Oct 13 09:30:27 pl Oct 13 09:30:28 ok Oct 13 09:30:28 03koen 07org.oe.dev * r71b1cbe6... 10/ (1 packages/tasks/task-base.bb): task-base: don't use obsolete wpa-supplicant-nossl Oct 13 09:31:07 koen: wpa-supplicant is heavy ;( Oct 13 09:31:26 we need to find a way to not add it for small flash devices Oct 13 09:32:08 actually i'm not sure how to treat vanille.de and vanille-media.de in the future. originally i planned to keep the first one for private stuff and the 2nd one for professional, but it looks like this split would be arbitrary and not so clever. i will perhaps just leave things as they are, vanille.de and vanille-media.de being just aliases Oct 13 09:32:25 * mickeyl brews a coffee, brb Oct 13 09:32:44 mickeyl: same problem here. hrw.one.pl and haerwu.one.pl (to be done). Oct 13 09:32:58 or will do blog.hrw.one.pl and hrw.one.pl Oct 13 09:33:41 koen: l-l-h 2.6.18 gave me glibc configured Oct 13 09:35:09 hrw|work: can't wpa-supplicant use gnutls or something? Oct 13 09:35:33 only 0.5.x ones which are not called stable Oct 13 09:35:41 morning all Oct 13 09:35:45 hi RP Oct 13 09:39:54 RP: Hi Oct 13 09:39:58 mickeyl: Hi Oct 13 09:43:17 Hi Oct 13 09:44:43 how can I use a TARGET_SYS to replace "powerpc-linux" in PREFERRED_PROVIDER_virtual/powerpc-linux-libc-for-gcc="glibc-intermediate"? Oct 13 09:47:24 O_Neil: it's the other way around. If you change TARGET_SYS, then this line will be ignored Oct 13 09:48:51 because some stuff that was pulled in because of the old TARGET_SYS did a depend on virtual/powerpc-linux-libc-for-gcc, and it would now depend on, say, virtual/arm-linux-libc-for-gcc when you change TARGET_SYS Oct 13 09:49:06 yes Oct 13 09:49:26 I do not want to redefine it for each platform Oct 13 09:49:35 Supported TLS/crypto libraries Oct 13 09:49:36 * OpenSSL (default) * GnuTLS Oct 13 09:49:48 koen: 0.4.x? Oct 13 09:49:53 hrw|work: so wpa-supplicant can be a lot smaller Oct 13 09:49:59 can it be set automatically ? Oct 13 09:51:07 hrw|work: ifeq ($(CONFIG_TLS), gnutls), in 0.4.8 Oct 13 09:52:46 good Oct 13 09:54:54 PREFERRED_PROVIDERS += " virtual/${TARGET_PREFIX}libc-for-gcc:glibc-intermediate" Oct 13 09:55:07 can this solve that problem? Oct 13 09:56:44 hey RP Oct 13 09:56:53 yo den-ros Oct 13 10:00:21 03pfalcon 07org.oe.dev * r5369901f... 10/ (6 files in 4 dirs): Oct 13 10:00:21 pocketsphinx: New package, the CMU PocketSphinx speech recognition engine for handhelds. Oct 13 10:00:21 * Also, a dependent sphinxbase package. Oct 13 10:00:21 * Package directory is called sphinx, to fit future ports of full-fledged Oct 13 10:00:21 Sphinx engines. Oct 13 10:00:22 * Package is rough for ipkg conventions, just autotools' make install. Whoever Oct 13 10:00:24 actually does speech recognition with it, please add sane packaging ;-). Oct 13 10:10:44 03mickeyl 07org.oe.dev * rea1e13af... 10/ (4 files in 2 dirs): task-bootstrap*: refactor common function into task-bootstrap.inc Oct 13 10:10:50 03mickeyl 07org.oe.dev * r08b850dd... 10/ (1 packages/p4/p4.inc): p4: add p4.inc which defines the common p4_arch function Oct 13 10:11:36 mickeyl: die, task-bootstrap, die Oct 13 10:11:40 i know Oct 13 10:11:48 no reason to brake it right now though Oct 13 10:18:06 mickeyl: http://dominion.kabel.utwente.nl/koen/cms/madness :) Oct 13 10:18:33 coool! :D Oct 13 10:20:04 OE is going for the desktop distro market now? :) Oct 13 10:20:46 03koen 07org.oe.dev * rfbafee33... 10/ (3 files in 3 dirs): linux-x86: enable some more filesystems for the guinness machine Oct 13 10:24:46 koen: Сongratulation! Oct 13 10:25:48 mickeyl: #1463 - can you look at it and add comment? Oct 13 10:25:56 03likewise 07org.oe.dev * r43309b85... 10/ (1 packages/lzma/lzma_4.17.bb): lzma: do not stage target binary, DEPEND on zlib - close #1476 Oct 13 10:28:29 hrw|work: yep Oct 13 10:29:10 koen: nice! Oct 13 10:29:17 koen: which machine/distro did you use? Oct 13 10:31:45 RP: absolutely :D Oct 13 10:32:12 can I specify package version in DEPENDS? Oct 13 10:32:40 Someone should write a parser to pull all the packages and patches from debian/ubunutu :) Oct 13 10:34:36 schurig: angstrom Oct 13 10:34:49 RP: and maemo :-) Oct 13 10:34:50 koen: and normal linux-x86 Oct 13 10:34:55 wow, ipkg is *fast* on a dual amd 2800+ :) Oct 13 10:35:14 schurig: MACHINE=guinness, for my mainboard Oct 13 10:35:27 RP: I started on an overlay driver, having trouble scibbling on the framebufer and it showing up though Oct 13 10:35:38 weird, people give beer names to mainboards Oct 13 10:35:52 schurig: not as wierd as B52 songs Oct 13 10:35:53 XorA: You might have to tweak the kernel driver too... Oct 13 10:36:14 RP: but fb2overlay shows up where and the size I expect it :-) Oct 13 10:36:26 XorA: That's a good start :) Oct 13 10:36:45 RP: but this is going to be slow work, as I can only grab 5 mins here and there Oct 13 10:36:58 ipkg is *fast* on a dual amd 2800+ :) Oct 13 10:37:36 Guinness must donate now Angstrom :) Oct 13 10:37:48 * mickeyl tries to see whether his new shiny Google SOC T-Shirt fits Oct 13 10:38:16 if I will get M instead of XL then Ania will get it Oct 13 10:40:02 * den-ros think that OE logo T-Shirt more cool than Google, but will prefer T-Shirt with big A :) Oct 13 10:40:20 hehe Oct 13 10:40:22 black Oct 13 10:40:22 den-ros: I have OE T-Shirt already Oct 13 10:40:22 :) Oct 13 10:40:23 with golden A Oct 13 10:40:32 hrw|work: :) Oct 13 10:40:52 its good to have 1.2G of mem Oct 13 10:41:48 mickeyl: Mickeyl did You see a little joke image where A compared with pyramid with eye on back side of US$ ? ;) Oct 13 10:42:19 heh, no Oct 13 10:42:54 mickeyl: I show it Koen but remove unfortunly already. Oct 13 10:44:27 thx for url koen . Oct 13 11:15:24 Hi! Oct 13 11:15:41 hey psokolovsky_ Oct 13 11:15:51 psokolovsky_: did you forget --author with your last commit? Oct 13 11:16:09 where bugs regarding linuxtogo.org infrastructure should be submitted? ;-) Oct 13 11:16:33 psokolovsky_: to linuxtogo@linuxtogo.org? Oct 13 11:17:51 psokolovsky_: http://www.linuxtogo.org/contact Oct 13 11:18:16 koen, Hi! Nope, my piece. There was a guy on familiar list crying that he can't build it for month+. So, I did it for exercise ;-) Oct 13 11:18:37 koen, thanks. i.e. there's no bugtracker for that, right? Oct 13 11:18:43 psokolovsky_: because VoodooZ has been talking about it for days :) Oct 13 11:19:11 really? I wonder, if he's just that guy ;-) Oct 13 11:19:30 psokolovsky_: there is http://bugs.linuxtogo.org, but that isn't used for the site (yet) Oct 13 11:19:44 ok Oct 13 11:20:40 there just appears rendering bug with projects' viewcvs (due to CSS I guess). will post details to that email later Oct 13 11:20:58 hahahaha. Oct 13 11:21:10 What are we talking about here. I miss that part! :) Oct 13 11:22:02 the glibc thing or the sphinx thing? the glibc one was another slugos guy Oct 13 11:22:03 VoodooZ, there's pocketsphinx in mtn now, if you're interested Oct 13 11:22:08 nice! Oct 13 11:22:32 So it should propagate to slugos' head eventually or? Oct 13 11:22:53 VoodooZ, not tested on device, so please do that ;-) Oct 13 11:23:06 I want to see how they did it as I almost had it working but found out the Makefile for it says it's not crosstool compatible! :( Oct 13 11:23:22 I will do that on my slug with my USB audio dongle. Oct 13 11:23:40 What's the best for audio lib nowadays? OSS or alsa? Oct 13 11:23:47 alsa Oct 13 11:24:03 ok. that's what I tried last time. Oct 13 11:24:16 I got flite working ok too. Oct 13 11:24:30 VoodooZ, issues like that I pretty common when crosscompiling. .bb made to be pretty clean though ;-) Oct 13 11:24:42 so Now my robot should be able to talk back to me (for debugging) and perhaps even listen to. Oct 13 11:25:01 I see. Oct 13 11:25:13 I've learned a great deal in the process anyways. Oct 13 11:25:14 VoodooZ, cool! On how many MHz it runs? Oct 13 11:25:21 Not enough unforunately Oct 13 11:25:40 It's running a modded slug (NSLU2) so 266Mhz. Oct 13 11:26:05 I'm keeping an eye open for anything faster for my future project as the vision stuff is really demanding. Oct 13 11:26:43 I'll probably end up with a PC based solution but power consumption and size is very important for a mobile robot of this size. Oct 13 11:27:13 VoodooZ: or use more slugs :) Oct 13 11:27:14 well, I probably won't be able to switch routes by voice command on my ~200MHz asus wl500g then ;-) Oct 13 11:27:25 The slug is great for that as it has USB2 for the camera. Most other board don't. Oct 13 11:27:44 the gumstix was my first choice but it doesn't have usb host mode. Oct 13 11:27:57 armadeus.com Oct 13 11:28:00 :) Oct 13 11:28:03 koen, yeah, I thought about that. dual slug. Oct 13 11:28:36 how would you do the interprocessor comm though? via ethernet using one of the MPC lib? Oct 13 11:29:47 Genesis, looks good but for vision it's too slow. Oct 13 11:30:04 Having hardware I2C on the board would definitely save me time too. Oct 13 11:30:44 psokolovsky_, I've read a paper about pocketsphinx and it was not far from real-time on a 200Mhz ARM so I wouldn't discount it. Oct 13 11:30:55 More of a novelty but still fun. Oct 13 11:31:39 I'm sure it can be further optimized for smaller vocabularies (for stuff such as: "Computer: Turn on Lights" Oct 13 11:31:53 I'm not expecting to use it for dictation. Oct 13 11:32:16 Those ARM boards are popping up everywhere lately though. Oct 13 11:33:14 * VoodooZ is updating slugos' HEAD Oct 13 11:34:11 Huh, since when do I "exist" here? CIA-4> likewise org.oe.dev * r43309b85... / (1 packages/lzma/lzma_4.17.bb) Oct 13 11:34:14 hmmmm. It appears the monotone.nslu2-linux.com server is down. Oct 13 11:34:21 likewise: mtn commit --author Oct 13 11:35:09 when do we use --author? Oct 13 11:35:48 Crofton: when we commit someone's patches Oct 13 11:35:53 when you commit a recipe someone else wrote Oct 13 11:35:59 ok Oct 13 11:36:08 I thought so, but wanted to make certain Oct 13 11:36:08 i thought author was the author of the software Oct 13 11:36:18 i should correct my recipes Oct 13 11:36:20 when you want to blame someone else :) Oct 13 11:36:30 but why MAINTAINER so Oct 13 11:36:37 it author is a bit same Oct 13 11:36:54 Genesis: AUTHOR is upstream Oct 13 11:37:06 AUTHOR = "creator of software" Oct 13 11:37:18 mtn commit --author "creator of recipe" Oct 13 11:37:33 s/recipe/recipe or patch Oct 13 11:40:36 I would like to populate my image using ipkg on my build host, but do not need ipkg support on the target. Currently, my package DEPENDS on "ipkg-collateral ipkg ipkg-link" which are also set in IPKG_INSTALL. Can I just remove those packages from IPKG_INSTALL to achieve this? Oct 13 11:42:19 likewise: they are added in rootfs_ipk.bbclass Oct 13 11:43:06 koen, How would you go around doing the vision stuff on multiple boards using only ethernet? Oct 13 11:43:31 I guess sending the whole image to both would be slow. Would be great for stereoscopic stuff though. Oct 13 11:44:47 VoodooZ: do vision stuff on one slug and audio on the other? Oct 13 11:45:55 yeah but the vision stuff is the real killer and the most important. voice can be ran in a low priority thread as it's not that important. Detecting obstacles is though. Oct 13 11:48:00 oh well, my next project will be on a much bigger base so I'll have more space for power so I'll just drop a portable in. Not as cool but works faster. Oct 13 11:51:25 koen, Thanks. sphinx just showed up on slugos' head. I'll give it a try this weekend and let you know. Oct 13 11:56:15 hrw|work: OK, I removed the ipkg from my image.bb, however, my rootfs ends up with "rootfs/usr/lib/ipkg/" but I don't know wherefrom. Oct 13 11:56:43 morning Oct 13 11:56:49 morning Oct 13 11:57:33 What is a good program in oe to search for memory leaks? I noticed valgrind is not in oe.. Oct 13 11:57:53 likewise: it will contain it because it is part of image created from ipkg packages. you can remove it after unpacking packages Oct 13 11:58:33 likewise: ipkg try to configure some packages on first start anyway.. we still does not resolved r/o filesystems Oct 13 11:58:53 Gerrath: valgrind is good but won't run on arm Oct 13 11:59:11 Gerrath: Since most OE devs use arm, that probably why its not in OE Oct 13 11:59:44 RP, what do most oe devs use then for memory leaks? Oct 13 11:59:58 RP other than good programming :-) Oct 13 12:00:20 Gerrath: I was just about to say :). I guess valgrind things on x86 to check them... Oct 13 12:00:23 Gerrath: Compile it on an x86 host and use valgrind? :) Oct 13 12:00:58 v8jlene, did you hear the news? Somebody got pocketsphinx working! Was it you? there's no maintainer line. Oct 13 12:01:10 hrw|work: OK, so I should not deploy the image in a R/O environment as this might break some packages... Oct 13 12:01:59 RP, NAbyss thanks. Oct 13 12:02:01 And how's the status of uClibc based slug? Oct 13 12:03:11 v8jlene, and it looks much cleaner than mine too! Nice work whoever did. Oct 13 12:04:01 likewise: run in in RW first Oct 13 12:08:44 koen: question about p-s. Oct 13 12:09:03 koen: can it clean staging and recreate it from packages? Oct 13 12:09:42 clean no, recreate, yes Oct 13 12:10:22 koen: so if I will remove tmp/staging/ it will be able to recreate it? Oct 13 12:10:38 it should Oct 13 12:10:43 btw - I started to get spam on @openembedded.org address Oct 13 12:10:45 it calls 'ipkg install foo' Oct 13 12:11:02 testing.. Oct 13 12:12:15 it won't recreate the host part now I think of it Oct 13 12:12:29 it does not Oct 13 12:13:01 and it does not use tmp/deploy/pstage/ feed Oct 13 12:16:43 <[lala]> RP: around? Oct 13 12:17:55 [lala]: yes Oct 13 12:18:36 <[lala]> RP: is there something special about the git fetcher? it doesnt pick up prefetched tarballs ... Oct 13 12:19:00 [lala]: It does here... Oct 13 12:19:00 <[lala]> or do i have to set a special variable for that? Oct 13 12:19:31 [lala]: Was a specific revisions specified? Oct 13 12:19:56 <[lala]> i have the problem with xcalibrateext-0.0+git20060814-r0 Oct 13 12:20:49 <[lala]> git_anongit.freedesktop.org.xorg.lib.libXCalibrate_master.tar.gz and co are in my dld dir Oct 13 12:21:12 [lala]: master means its a floating revision - ie always the latest Oct 13 12:21:28 <[lala]> i have master and no master there as tarball Oct 13 12:21:46 [lala]: Also, fixed dates are meaningless to git Oct 13 12:21:52 RP: base_contains_two would be nice... Oct 13 12:22:05 <[lala]> can i set some debug level to see why it doesnt take the prefetched one? Oct 13 12:22:12 RP: ${@base_contains_two("DISTRO_FEATURES", "pcmcia", "pci", "hostap-utils", "",d)} Oct 13 12:22:47 where is linux_modules.bbclass? Oct 13 12:23:04 it got renamed Oct 13 12:23:17 ah, many *.bb still inherit it Oct 13 12:23:20 what's the new name? Oct 13 12:23:31 do they? i grepped through the tree Oct 13 12:23:32 hrw|work: That would open the door to _three, _four, _five etc :) Oct 13 12:23:34 bitbake makes a WEIRD error message because of this Oct 13 12:23:40 schurig: linux-kernel-base Oct 13 12:23:51 RP: any solution would be nice Oct 13 12:23:54 We should make it take a list :) Oct 13 12:24:05 RP: best Oct 13 12:24:13 mickeyl: bitbake tells me: "ERROR: Unable to open conf/bitbake.conf" Oct 13 12:24:45 schurig: odd, but bitbake error reporting is on our list of things to improve very high Oct 13 12:24:49 but the problem was really in lib/bb/parse/parse_py/BBHandler.py, line 108 Oct 13 12:25:00 schurig, yep, that sux. I have patch for that. Oct 13 12:25:10 schurig: What was the real problem? Oct 13 12:25:12 there it raised an IOERROR with a message what file was not found, but bitbake did eat this good error message and spit out the nonsense message Oct 13 12:25:13 ah darn Oct 13 12:25:21 i didn't thought of cINHERIT += Oct 13 12:25:24 * mickeyl fixes Oct 13 12:26:02 mickeyl: I'm on svn://svn.berlios.de/bitbake/branches/bitbake-1. Oct 13 12:26:10 psokolovsky_, pocketsphinx in mtn compiles most of it but fails bitching: Armeb-linux-libtool: link: cannot find the library `/usr/lib/libsphinxutil.la' or unhandled argument `/usr/lib/libsphinxutil.la' Oct 13 12:26:10 darn, n branches/bitbake-1.6 Oct 13 12:26:24 schurig: bitbake is nowadays the property of RP and Zecke Oct 13 12:26:34 i didn't have a chance to look at it for ages Oct 13 12:26:50 schurig: and as mickeyl said, we know the error reporting is going badly wrong atm :-/ Oct 13 12:27:16 schurig: If people report the specific cases, I will try to look at them. I wasn't aware of this one Oct 13 12:27:27 VoodooZ, weird. builds finf for h4000/angstrom. look for more places where it harcodes stuff Oct 13 12:27:55 03hrw 07org.oe.dev * r27992ad9... 10/ (1 packages/tasks/task-base.bb): task-base: install hostap-utils only if there is pci or pcmcia featured Oct 13 12:28:30 RP: the real problem was that conf/machine/include/handheld-common.conf did INHERIT += "linux_modules" and bitbake 1.6 detects this in BBHandler.py correctly, but bitbake/bin/bitbake near line 814 didn't emit that linux_modules.bb was nowhere, but said conf/bitbake.conf was not readable Oct 13 12:28:44 RP: which was not true, bitbake -D showed that it was indeed readable Oct 13 12:29:05 psokolovsky_, Yeah, I'm looking at it now... Oct 13 12:30:38 03mickeyl 07org.oe.dev * r45a56df6... 10/ (8 files in 4 dirs): Oct 13 12:30:38 many conf files: catch up with renaming linux_modules to linux-kernel-base Oct 13 12:30:38 sorry, I didn't think of INHERIT += Oct 13 12:32:32 RP: Btw, I have an RFC of branching entrire bitbake trunk/ for next versions, as well as include that in d/l tarball. will help people to get most of BB. Oct 13 12:33:05 03mickeyl 07org.oe.dev * r11fba4b6... 10/ (1 classes/sanity.bbclass): sanity.bbclass: improve error reporting for missing utilities. thanks nicholas Oct 13 12:33:36 psokolovsky_: There is a RFC coming of how were thinking of providing downloadable tarballs. Including a checked out copy of bitbake within that is probably a good idea Oct 13 12:35:16 RP, nice. I'm reaaly waiting for info of what was decided regarding BB on OEDEM (no hurry of course) Oct 13 12:35:17 schurig: Thanks, I've made a note to look at it. I can't promise when, but I will try :) Oct 13 12:35:38 psokolovsky_: I was waiting for the summary to come out. Expect something this weekend :) Oct 13 12:35:40 03rpurdie 07org.oe.dev * r53955fef... 10/ (1 MAINTAINERS): MAINTAINERS: Add me, clarify file structure slightly Oct 13 12:35:50 ok, cool Oct 13 12:39:05 I need to override a specific ac_sizeof when configure runs Oct 13 12:39:13 how does this work? Oct 13 12:39:39 basically, when I build the native omniorb, I need it to build with the sizeof long double for the target ... Oct 13 12:44:27 03koen 07org.oe.dev * r2af10c8f... 10/ (7 files in 3 dirs): procps: update to 3.2.7 Oct 13 12:51:41 argh.. Oct 13 12:51:57 I have to resubscribe to lakml because I too often send mail from wrong account Oct 13 12:53:00 03mickeyl 07org.oe.dev * r0168c996... 10/ (1 MAINTAINERS): MAINTAINERS: add me and resort (guys, it's _first_ name, not last name) Oct 13 12:53:45 thx mickeyl - I had it in todolist Oct 13 12:53:51 :) Oct 13 12:54:20 first name? Oct 13 12:54:35 mickeyl: I ordered a usb-vga thingie yesterday Oct 13 12:54:36 that's what it states in the document Oct 13 12:54:42 dunno whether it's a good idea or not Oct 13 12:54:46 koen: oh, cool. where? Oct 13 12:55:10 alpha by first name? Oct 13 12:55:23 mickeyl: cheapest online store in .nl Oct 13 12:55:47 mickeyl: firstname is from linux maintainers file Oct 13 12:55:52 ah Oct 13 12:55:54 mickeyl: so combined with my ratoc card I can do presentation with almost every toy I have :) Oct 13 12:55:58 mickeyl: I just added that to clarify the position as people were sorting by both :-/ Oct 13 12:56:06 RP: hehe, ok Oct 13 12:56:13 it should be correct now Oct 13 12:56:18 (as far as I can spell...) Oct 13 12:56:21 I thought I'd resorted it but one was out of place :) Oct 13 12:57:28 well, I think we should sort by last name so I can be at the top! Oct 13 12:57:36 heh Oct 13 12:57:39 forget it Oct 13 12:57:40 oe-stylize-maint.py coming soon from mickeyl :-) Oct 13 12:57:50 :D Oct 13 12:58:03 * XorA changes name to Aaron A Aardvark Oct 13 12:58:16 how do you sorrt the guys with funny characters Oct 13 12:58:38 take the ascii that has the minimum hamming distance Oct 13 12:59:30 A script to look for packages no one maintains would be interesting Oct 13 12:59:44 s/interesting/depressing/ ? Oct 13 13:00:05 :) Oct 13 13:00:21 at least people could look for packages to maintain that way Oct 13 13:01:33 03hrw 07org.oe.dev * r0c3f381b... 10/ (1 MAINTAINERS): MAINTAINERS: edited my entry, added WiFi, changed fonts info Oct 13 13:02:20 With the clearout, we've lost a lost of maintainers, although at least the ones we have are accurate now... Oct 13 13:08:03 if so, then the recipe lines: should exactly match the recipe file names.... Oct 13 13:08:03 Crofton: if so, then the recipe lines: should exactly match the recipe file names.... Oct 13 13:08:12 Crofton: this seems not the case now Oct 13 13:08:34 recipe names or regexp? Oct 13 13:08:53 likewise: It should be possible to take what's there and use some kind of fuzzy matching Oct 13 13:11:08 lint for the MAINTAINER file Oct 13 13:12:28 Today is my first time in this irc. So hello to everybody Oct 13 13:12:39 hi pf Oct 13 13:12:40 gm Oct 13 13:13:10 so I have a question to distcc and bitbake. Did anybody use it? Oct 13 13:14:05 I use it since one houre. it works fine. Speed up bitbake a lot. BUT...... Oct 13 13:15:25 pf: we use icecream for that Oct 13 13:16:16 icecream? Oct 13 13:17:45 pf: http://gentoo-wiki.com/HOWTO_Setup_An_ICECREAM_Compile_Cluster Oct 13 13:18:56 koen: is the conf.local for your madness (http://dominion.kabel.utwente.nl/koen/cms/madness) boot in .dev? Today I am without any targets other than this machine... Oct 13 13:25:35 Crofton: how goes work on 2.6.18-omap-booting? Oct 13 13:26:08 bitbake will override my PATH by use $PATH_prepend+PATH . So I can't prepend my own PATH to the distcc. Is there any other posibility? (if you understand what i meen) Oct 13 13:26:25 likewise: I'm running it now :) Oct 13 13:26:31 likewise: MACHINE=guinness Oct 13 13:27:03 morning Oct 13 13:27:15 hrw|work, building bootstrap-image should produce a working uboot/kernel/fs iamge Oct 13 13:27:16 atm Oct 13 13:27:36 hi hrw Oct 13 13:27:41 I am trying to sort out omniorb now Oct 13 13:28:35 hrw|work, I am watching as changes go into git Oct 13 13:29:05 koen: how many host do you use? Oct 13 13:29:09 Crofton: so current git head is working? Oct 13 13:29:14 hi chouimat siriusnova Oct 13 13:29:19 yes Oct 13 13:29:24 if you enable dspgw :) Oct 13 13:29:46 and don't enable LL DEBUG Oct 13 13:30:53 Crofton: can I then add kind of linux-omap1_2.6.18-working.bb recipe? Oct 13 13:31:05 yes Oct 13 13:31:24 using the current git tag? Oct 13 13:31:30 koen: guinness, in a can? :-) Oct 13 13:31:36 Crofton: yes Oct 13 13:31:38 brb Oct 13 13:31:40 ok Oct 13 13:37:23 Crofton: [hrw@host1 ps]$ bbrebuild ~/devel/oe/org.openembedded.dev/packages/linux/linux-omap1_2.6.18+git.bb Oct 13 13:37:54 ok Oct 13 13:37:58 Crofton: http://source.mvista.com/git/gitweb.cgi?p=linux-omap-2.6.git;a=commit;h=c6051183c597b6a0fa73cdb59aac852c6148c5b6 used Oct 13 13:38:01 bbl Oct 13 13:40:50 likewise: that's the codename of the mainboard :) Oct 13 13:40:55 mickeyl: http://dominion.kabel.utwente.nl/koen/images/angstrom-x86.jpg Oct 13 13:41:29 psokolovsky_, Did you say you knew the who create the pocketsphinx package? Oct 13 13:41:46 koen: that's not too bad! Oct 13 13:41:59 koen: how's ipkg performing on that machine? Oct 13 13:42:01 still slow? Oct 13 13:42:02 mickeyl: and it's wicked fast as well Oct 13 13:42:15 mickeyl: ipkg is faster as apt Oct 13 13:42:18 heh Oct 13 13:43:33 koen: did you anything else for speed up bitbake? It takes very long Oct 13 13:44:05 I use bitbake 1.7 and that very fast Oct 13 13:44:16 can I safely ignore this: NOTE: Multiple libraries (libsphinxutil.so.0, libsphinxad.so.0, libsphinxfe.so.0, libsphinxfeat.so.0) found and LEAD_SONAME not defined Oct 13 13:45:00 is bitbake 1.7 much faster then 1.6.1? Oct 13 13:45:36 VoodooZ: you can, although you should think about a) packing them into different packages or b) setting LEAD_SONAME for the "main" library. Oct 13 13:46:09 mickeyl, i couldn't find any info on the doc site about LEAD_SONAME. Oct 13 13:46:53 VoodooZ: whenever you can't find info about a certain VARIABLE, just grep for it in packages/*/*.bb and see how other recipies are using it Oct 13 13:48:54 mickeyl: btw, Ø comes second last in the alphabet, between Æ and Ã…. Not anywhere near O ;-) Oct 13 13:49:12 NAiL: like I said, hamming distance :D Oct 13 13:49:23 feel free to correct it Oct 13 13:49:41 (visual encoding just doesn't gets it) Oct 13 13:49:51 but that's why I would sort Ö to be near O Oct 13 13:50:06 On most cellphones, it's also sorted on the same key as O Oct 13 13:50:13 which is annoying Oct 13 13:50:45 and if I switch my phone language to english (which I prefer), it screws up the sorting too :-( Oct 13 13:51:00 VoodooZ, of course I know, because it is me ;-) Oct 13 13:51:01 so that's something you would need on a cool phone? Oct 13 13:51:12 sorting to respect locales? Oct 13 13:51:20 * mickeyl notes Oct 13 13:51:30 yes, correct sorting is a good thing Oct 13 13:51:51 when someones name starts with the last char, it shouldn't be sorted first ;) Oct 13 13:53:04 btw, on the subject of special chars... Oct 13 13:53:22 Wouldn't UTF-8 be a good choice? Oct 13 13:53:52 why is libtool finding libs in the native staging area? Oct 13 13:54:39 because libtool should've been shot on conception? Oct 13 13:54:49 besides that Oct 13 13:54:53 oh... Oct 13 13:54:54 dunno Oct 13 13:56:34 psokolovsky_, hehehe Sorry. It sounded like it was somebody else. Imissed the beginning of the conversation... Oct 13 13:56:37 03nail 07org.oe.dev * rfd7e0b1c... 10/ (1 MAINTAINERS): MAINTAINERS: Re-sort me to the bottom. Scandinavian characters and all. Oct 13 13:57:09 psokolovsky_, I tried doing the same you did for INCDIR but for LIBDIR but to no avail.. Oct 13 13:58:18 03koen 07org.oe.dev * r348938c3... 10/ (1 conf/machine/htcuniversal.conf): htcuniversal: ship acx module as well Oct 13 14:02:28 If I am building a library that uses libtool, do I need to put anything "special" in the bb file? Oct 13 14:06:55 usually not Oct 13 14:07:33 dependds on libtool-cross, just in case? Oct 13 14:08:06 libtool is doing something stupid and finding native libs in staging for me Oct 13 14:10:01 something is on crack Oct 13 14:10:09 mickeyl: did I mention that glibc 2.5 rocks? Oct 13 14:10:24 koen: no you didn't. that's actually a bunch of good news Oct 13 14:10:43 mickeyl: it solves the shm bugs in x Oct 13 14:10:55 oh great! Oct 13 14:10:55 mickeyl: although loadkeys still segfaults the first run Oct 13 14:10:58 argh Oct 13 14:10:59 :) Oct 13 14:11:04 For your amusement ..... Oct 13 14:14:39 Crofton: i don't understand any of this autofoo/libtool stuff, that's why i use qmake everywhere... better use the mailing list for that. Oct 13 14:14:47 03koen 07org.oe.dev * ra78e7ef3... 10/ (1 MAINTAINERS): MAINTAINERS: UTF-8 rocks, so let's use it Oct 13 14:14:57 mickeyl: angstrom looks ready to ship, stability wise, just need to fix busybox ps and mount Oct 13 14:15:07 NAiL: there you go Oct 13 14:15:16 koen: great, that sounds like we can meet 2007.01 Oct 13 14:15:55 psokolovsky_, It seems to work now. I'm not sure if it's because of the old crap from my poor attempt at making a package or not but I'll see. Oct 13 14:16:08 ok Oct 13 14:17:24 koen, Is Angstrom boot logo in place now with psplash? Oct 13 14:17:36 psokolovsky_: no, a crappy OE logo Oct 13 14:17:56 psokolovsky_: I need to poke mallum about the rgb value of the psplash background Oct 13 14:17:59 koen, then, it's not ready to ship, users won't get it ;-) Oct 13 14:18:04 ok Oct 13 14:26:37 koen: the Aongstom on mickeyl looks right the RP one doesnt Oct 13 14:27:02 if I mtn commit in a directory, the changes in that directory and below are commited, but nothing above? Oct 13 14:27:18 koen: Angstrom fails to work on my c7x0 Oct 13 14:28:15 XorA: I'll have to dig out my c700 to test Oct 13 14:28:40 XorA: so far only people with a zaurus reported a RO / Oct 13 14:28:46 03koen 07org.oe.dev * r172e330b... 10/ (1 MAINTAINERS): MAINTAINERS: also fix RP's entries Oct 13 14:30:30 Crofton: i think so, but to be sure, just mtn ci . Oct 13 14:30:54 It looks likes like it does :) Oct 13 14:32:47 I am going to delete the older omniorb packages since they likely do not work well atm Oct 13 14:32:54 ok, cool Oct 13 14:33:29 and I think the 4.0.7 will be "reliable" Oct 13 14:35:43 koen|away: Xfbdev hard locks the machine Oct 13 14:36:03 koen|away: loadkeys fails, so it doesnt actually look any different than glibc2.4 to me Oct 13 14:36:57 XorA: doesn't sound good Oct 13 14:37:55 03crofton 07org.oe.dev * rb7fc885d... 10/ (5 files in 3 dirs): Oct 13 14:37:55 omniorb_4.0.7.bb : Fix file to correctly build without LongDouble support. Oct 13 14:37:55 * Add patches for ARM support Oct 13 14:37:55 * Convert to auto staging. Oct 13 14:38:00 03crofton 07org.oe.dev * ra81ee1e7... 10/ (4 files in 3 dirs): omniorb : Delete old and very likely non-working version of omniorb bb files. Oct 13 14:49:16 what is the good way to modify a recipe/file in OE but don't lost the original file , when update for example , is there a sort of overlay ? Oct 13 14:49:47 Genesis: Modify as necessary, then revert with monotone when you want to go back? Oct 13 14:49:49 my specific file in overlay , near the normal tree Oct 13 14:50:02 a bit like on gentoo Oct 13 14:51:10 i can also prefix the recipe with a id , and precise this in the distro file Oct 13 14:51:28 but it' s not dynamic Oct 13 14:54:05 Genesis: Look into BitBake Collections: BBFILE_COLLECTIONS = "upstream local" Oct 13 14:54:08 Genesis, problem's not clear. just put your file besides the rest - nothing will overwrite it. or make your own subtree for your recipes. or create private branch. Oct 13 14:54:23 Genesis: this way, local is an overlay over upstream Oct 13 14:54:35 Genesis: much like Gentoo Oct 13 14:56:01 i'll try thx guy Oct 13 15:13:20 Does the PXA250 have an integrated 2D engine? Oct 13 15:14:05 CosmicPenguin: iirc not Oct 13 15:14:12 CosmicPenguin: only a fb Oct 13 15:14:17 again, iir Oct 13 15:14:18 c Oct 13 15:14:42 Somwhere along the line - I thought that a Xscale appeared with 2D on it Oct 13 15:14:42 hmm Oct 13 15:20:10 can someone please tell me which package i need to get for STL support, thanks Oct 13 15:22:23 hi, all! Oct 13 15:22:48 CosmicPenguin: pxa270 and beyond have overlay support Oct 13 15:22:54 could anybody help with u-boot pxafb.c? Oct 13 15:23:15 cosmicpenguin: the xscale chipset doesn't have integrated video, Oct 13 15:23:54 CosmicPenguin: Developments beyond the 270 might have more features but the 270 basically just does overlays Oct 13 15:24:00 but the zaurus C7x0/8x0 have an ATI video chip on the board Oct 13 15:24:17 its got some accelerated FB function, but no 2d accel Oct 13 15:24:51 althought RP could give you a more accurate description ;) Oct 13 15:25:13 raduga: take a look at Xw100 that has 2d accelleration on w100 Oct 13 15:25:14 s/xscale/pxa250/ Oct 13 15:25:15 sorry Oct 13 15:25:43 koen: thx ;) Oct 13 15:26:02 the ati w100 that some zaurus have, actually does 2d accel? (sorry, I didn't know) Oct 13 15:26:18 or is this a custom Xscale, with ati built in? Oct 13 15:26:37 * raduga crawls back into a silent space Oct 13 15:26:38 raduga: separate chip which does support some 2D operations Oct 13 15:27:04 rp: thanks. Oct 13 15:27:09 enough operations that Xw100 makes gnome/firefox usable Oct 13 15:28:34 speaking of X, RP any news on Xiwmmx? Oct 13 15:28:58 Heh - ATI? :) Oct 13 15:29:13 Doesn't give as much performance as we'd like as the data has to be 8byte aligned Oct 13 15:29:26 i like ati :) Oct 13 15:29:56 raduga: You like closed specs? Oct 13 15:30:44 rp: i like the extent to which ati has sometimes historically allowed sufficent openness for people to make reasonable drivers for their cards Oct 13 15:30:57 rp: maybe not in the last decade Oct 13 15:31:32 rp: but at least, once apon a time, ati was much more helpful than the average video chipset maker Oct 13 15:31:58 there is no ati Oct 13 15:32:03 there is only atimd Oct 13 15:32:09 rp: since my statement "i like ati" doesn't really correspond well with modern reality, please feel free to ignore Oct 13 15:37:23 raduga: even the ATI linux drivers dont work, at least nvidia with closed specs works :-) Oct 13 15:41:01 Can OE package icecream cross-compiler chroot environments for us? from the icecream web site: "How to package such a cross compiler is pretty straightforward if you look what's inside the tarballs generated by icecc" Oct 13 15:41:42 likewise: you;d have to ask icecream master zecke Oct 13 15:42:41 koen: I think I will be secretly deploying Ubuntu VMWare images across my colleague's machines to build an icecream park :-) Oct 13 15:42:56 :) Oct 13 15:48:46 * koen shows http://ewi546.ewi.utwente.nl/tmp/die-pre-xorg-die.diff to the world Oct 13 15:49:40 koen: looks good to me Oct 13 15:54:10 likewise: if you want to know how to build icecc tarball look at suse cross-*-icecream.src.rpm :) Oct 13 15:54:17 I'll hold off with commiting it for a bit longer :) Oct 13 15:55:07 chouimat|away: hmm, I'll just follow the docs instead and learn a bit. Oct 13 15:55:33 likewise: afaik icecream.bbclass tar up the toolchains for you Oct 13 15:55:42 cool Oct 13 15:55:43 * ade|desk waves good bye to the diet-x11_6.2.1 he love so much Oct 13 15:56:17 mickeyl: do you know when zeckeis scheduled to arrive in berlin? Oct 13 15:56:29 koen: go on commit it ;) Oct 13 15:57:22 ooh i like the latex saving in oo.org 2.0.4 Oct 13 16:00:47 koen: thanks, icecc.bbclass is there. Oct 13 16:05:56 hmm, I cannot deduce where I should inherit icebb.class. Oct 13 16:06:11 likewise: local.conf? Oct 13 16:06:31 koen: no, that is toooo obvious Oct 13 16:15:21 ade|desk: finished saying goodbye to diet-x11 6 ? Oct 13 16:22:39 re Oct 13 16:22:51 hrw: wb Oct 13 16:24:17 ~hail ssh keys for their usefullness on ssh chains Oct 13 16:24:18 * ibot bows down to ssh keys for their usefullness on ssh chains and chants, "I'M NOT WORTHY!!" Oct 13 16:24:22 NOTE: package linux-omap1-2.6.18+git-r0: task do_build: completed Oct 13 16:26:09 Crofton will be happy Oct 13 16:32:29 I get "NOTE: exceptions.TypeError:coercing to Unicode: need string or buffer, NoneType found while evaluating:" and then some 8 lines of output. Does the error relate to the first line of the output, or the last line? Oct 13 16:32:46 (that's with INHERIT += "icecc" btw) Oct 13 16:33:43 it's amazing how abiword *instantly* starts up on angstrom/x86 Oct 13 16:34:32 a+ Oct 13 16:38:33 its amazing how fast kernels can be built when no one use dual core machine Oct 13 16:38:50 Crofton: I cleaned linux-omap1 kernels Oct 13 16:39:11 Crofton: rebuilding now and will push if it will work Oct 13 16:42:27 http://pastebin.com/805946 Oct 13 16:42:45 koen: that is angstrom on guinness? Oct 13 16:54:13 Who edited MAINTAINERS on a Mac? ;-) Oct 13 16:54:35 lol Oct 13 16:54:37 koen Oct 13 16:54:37 psokolovsky_: that would be me Oct 13 16:54:44 hey zecke Oct 13 16:54:50 koen: my bonsai works with mtn now :) Oct 13 16:54:54 zecke: cool Oct 13 16:55:00 there is one issue with utf-8 somewhere which I have not found Oct 13 16:55:09 koen: and you can subscribe to a rss feed Oct 13 16:55:10 koen, have a look at it ;-) in hex or on linux ;-) Oct 13 16:55:19 koen: feed://127.0.0.1:8000/bonsai/feeds/author/njs/ Oct 13 16:55:19 zecke: I've been waiting for you to return so we can set it up on oe.org Oct 13 16:55:50 can we do that today or tomorrow? Oct 13 16:55:53 NOTE: package dpkg-1.13.22-r0: task do_compile: failed Oct 13 16:56:18 zecke: lets to it now! Oct 13 16:56:19 we need to say 'we do not want selinux' Oct 13 16:56:42 koen: I fear, I eat atm :) Oct 13 16:56:42 03koen 07org.oe.dev * r989e5fa0... 10/ (92 files in 7 dirs): remove pre xorg stuff, as indicated in http://lists.openembedded.org/pipermail/openembedded-devel/2006-October/000538.html Oct 13 16:56:51 03koen 07org.oe.dev * rb2c9193c... 10/ (1 site/i686-linux): site/i686-linux: add ettercap sock_len Oct 13 16:57:06 koen: do you have added a postgres database? Oct 13 16:57:36 zecke: no Oct 13 16:57:49 zecke: can we start with sqlite? Oct 13 16:58:02 koen: sure Oct 13 16:58:54 we need to define rules for openembedded-announce mailing list Oct 13 16:59:15 to sent there info about pre-X.org stuff removal there or not etc Oct 13 16:59:22 koen: Damn, why did you mail gpe@hh.org? Oct 13 17:03:09 zecke: put your stuff in /home/oe/website/bonsai Oct 13 17:04:19 bbl Oct 13 17:05:42 zecke: Is this ok: http://rafb.net/paste/results/EQkFpd22.html ? Oct 13 17:07:42 koen: no Oct 13 17:08:02 * zecke moves into his 'nasty' room Oct 13 17:08:38 zecke: ;D Oct 13 17:08:50 03pfalcon 07org.oe.dev * r91bdb4f0... 10/ (1 MAINTAINERS): MAINTAINERS: Fix line-endings: CR => LF. Oct 13 17:10:18 koen: SetHandler is python Oct 13 17:10:31 koen: and we need one SetEnv PYTHONPATH ... Oct 13 17:10:46 BTW: is Hans Reiser in jail for real? Oct 13 17:11:00 zecke: arrest Oct 13 17:11:11 Crofton: check my changesets Oct 13 17:11:22 ~hail MACHINE="native" Oct 13 17:11:25 * ibot bows down to MACHINE="native" and chants, "I'M NOT WORTHY!!" Oct 13 17:11:26 ~hail MACHINE="native" again Oct 13 17:11:27 * ibot bows down to MACHINE="native" again and chants, "I'M NOT WORTHY!!" Oct 13 17:11:33 zecke: http://rafb.net/paste/results/zwH0bn29.html Oct 13 17:11:55 koen: you missed the '...' in SetEnv PYTHONPATH Oct 13 17:13:34 hrw, I will look over them in a little while Oct 13 17:14:06 did you change the kernel version on the omap5912osk machine file? Oct 13 17:14:29 Crofton: not touched it Oct 13 17:14:35 zecke: seriously, what needs to be put into pythonpath? Oct 13 17:14:47 koen: http://rafb.net/paste/results/Vnk9Wf33.html Oct 13 17:15:45 ok Oct 13 17:16:02 Crofton: kernel version should be set in distro anyway Oct 13 17:16:09 koen: obvioulsy change /space/webapps with something else Oct 13 17:16:10 right :) Oct 13 17:16:18 I keep forgetting Oct 13 17:16:34 hi all Oct 13 17:16:51 I am trying to figure out a libtool problem right now, but I will look over changesets Oct 13 17:17:05 and attempt to get in a position to refalsh my OSK Oct 13 17:18:05 I can set kernel version in distro and overried in local.conf? Oct 13 17:18:32 bbiab Oct 13 17:21:27 <[lala]> how can i add more mirrors to the MIRRORS() array in base.bbclass w/o modifing the class itself? Oct 13 17:24:52 Crofton|home: another changes for you Oct 13 17:27:39 have to go - cu Oct 13 17:40:40 hello from dresden Oct 13 17:41:10 mmmm.. dresden Oct 13 17:41:28 I am jealous Oct 13 17:41:40 yes, CosmicPenguin, 2km from the amd fabs :) Oct 13 17:44:11 greentux: do you like beer? Oct 13 17:44:23 CosmicPenguin: yes why not! Oct 13 17:44:34 http://www.watzke.de/ball/ball.php Oct 13 17:44:47 CosmicPenguin: I know tha location Oct 13 17:44:49 Highly recommend - if you're in Dresden on Tuesday, thats Masse beer night Oct 13 17:45:01 Your at Dresden this week? Oct 13 17:45:07 No, not now Oct 13 17:45:24 But I've come in contact with much beer in Dresden in the past Oct 13 17:45:43 http://www.waldschloesschen.de/ Oct 13 17:45:45 I like that place too Oct 13 17:46:10 Who is the libtool guru? Oct 13 17:46:18 fine, is there any chance in the future? Oct 13 17:46:50 not right now - but I'm actively seeking out any chance to get there Oct 13 17:47:22 CosmicPenguin: trade you a bottle of beer for w100 specs :) Oct 13 17:47:29 w100? Oct 13 17:47:33 I wonder, how comes that OE generats >= version constraint for each RDEPENDS, and that constraint is set to the *current* version of dependency? Oct 13 17:47:36 ati imageon w100 Oct 13 17:47:39 heh Oct 13 17:47:48 Yeah - well, thats not likely going to happen any time soon Oct 13 17:47:57 CosmicPenguin: your loss ;) Oct 13 17:48:03 CosmicPenguin: hey :) Oct 13 17:48:06 I am only permitted to give you as many Geode specs as you can handle Oct 13 17:48:21 schurig: hey! Oct 13 17:48:25 schurig: how are you? Oct 13 17:48:46 zecke: fine! Oct 13 17:48:55 What's the kernel module name for those cheap usb sound card dongle you can find all over ebay? Oct 13 17:48:59 zecke: and you? How's health and study? Oct 13 17:49:17 schurig: health okay, study... well as well :) Oct 13 17:49:31 schurig: I attended the Qt Dev Days, just returned home safely Oct 13 17:50:44 zecke: mickeyl was there as well Oct 13 17:50:51 zecke: anything interesting for you? Oct 13 17:51:03 schurig: I even slept in the same room as mickeyl ;) Oct 13 17:51:23 schurig: QGraphicsView and CascadingStyle Sheets is awesome Oct 13 17:51:24 zecke: I should have been there, too. Could have slept at my sisters place ... Oct 13 17:51:44 a moose bit my sister Oct 13 17:51:58 Crofton: did the moose survive? Oct 13 17:52:03 or by my brother-in-law, they both are in Munich Oct 13 17:52:27 what is the command to show files in the repo that are not in mtn? Oct 13 17:52:40 Crofton: isn't it mtn status? Oct 13 17:52:48 Crofton: 'mtn' and read the output :) Oct 13 17:52:52 koen: is there any acitivy for supporting olympus r1000? Oct 13 17:53:04 greentux: afaik not Oct 13 17:53:13 greentux: last I heard they stopped the port Oct 13 17:53:14 Crofton: ls unknown ;) Oct 13 17:53:23 http://www.wilwheaton.net/2006/02/a_moose_bit_my_sister_once.php Oct 13 17:53:43 koen: who is "they"? Oct 13 17:53:54 greentux: sdg Oct 13 17:54:12 koen: hey all pda i find in the net have todo with sdg... gggrrml Oct 13 17:54:32 get a greenphone Oct 13 17:54:37 koen: sdg? Oct 13 17:54:40 zecke: closed kernel Oct 13 17:54:50 greentux: == secure kernel Oct 13 17:55:17 greentux: and no, you can get the source as well, but it ancient Oct 13 17:55:28 zecke: mmmh Oct 13 17:55:56 zecke: from trolltech? Oct 13 17:56:28 zecke: and i would prefer 2.6 :) as i said Oct 13 17:57:38 greentux: 2.4 ancient, full source if you pay for it Oct 13 17:57:45 greentux: but this device lacks wlan, but is a phone Oct 13 17:58:18 zecke: a nice pda would be enough as first step... Oct 13 17:59:02 VoodooZ: doesn't kernel-module-snd-usb-audio-2.6 do the job? Oct 13 17:59:07 greentux: oh still at the 1st step? Oct 13 17:59:38 kerwood, thanks I'll try that and alsa... Oct 13 17:59:42 zecke: yes the way to road is hard way... :) Oct 13 18:00:16 zecke: first step was zaurus, works perfekt, but is not customer compatible Oct 13 18:01:51 kerwood, I'm just not sure if I need all the OSS crap and the various mixers. I'm hoping pocketsphinx and flite use alsa and not the older OSS. Oct 13 18:03:59 oh, man ... i'm freaking out. i'm trying to build another dev computer with OE and i can't expand the OE.mtn database Oct 13 18:04:36 it's says "wanted schema 'ae1968...' but got '9d2b5...'" Oct 13 18:04:42 any idea what i'm supposed to do? Oct 13 18:04:54 bigd0g: mtn --version Oct 13 18:05:05 bigd0g: happen to have updated to a new version of mtn? Oct 13 18:05:27 zecke: it's a new build computer, so i just pulled it from a repo Oct 13 18:05:31 it should be mtn 0.30 Oct 13 18:05:33 let me check Oct 13 18:05:44 yeah. mtn 0.3 Oct 13 18:06:08 bigd0g: and where did you get the mtn db from? Oct 13 18:06:21 bigd0g: http://venge.net/monotone/UPGRADE Oct 13 18:06:30 bigd0g: you need to either get the 0.30 mtn snapshot or run the upgrade sequence on your db Oct 13 18:06:42 0.30 OE db snapshot I mean Oct 13 18:06:43 bigd0g: but OE provides two db's one for mtn >= 0.26 <= 0.29 and one for >= 0.30 Oct 13 18:06:44 i tried the 'migrate' sequence... that seems to fail Oct 13 18:06:52 upgrade is worth it ... Oct 13 18:06:54 bigd0g: perhaps you did it in the wrong order? Oct 13 18:06:58 i had gotten the snapshot from openembedded.org/snapshots/OE.mtn.bz2 Oct 13 18:07:46 JustinP: i tried 'mtn --db=OE.mtn migrate' Oct 13 18:07:46 JustinP: is that incorrect? Oct 13 18:07:46 yes Oct 13 18:07:46 look at the link zecke gave you Oct 13 18:07:59 or look at http://www.openembedded.org/snapshots/ Oct 13 18:07:59 bigd0g: try http://www.openembedded.org/snapshots/OE-this-is-for-mtn-0.30.mtn.bz2 Oct 13 18:08:14 koen: please install pysqlite2 Oct 13 18:08:16 zecke: thanks for the link. /me looks Oct 13 18:09:07 magical. mtn syntax ftw Oct 13 18:09:09 bigd0g, hrw has reaarnged the kernel bb files Oct 13 18:09:19 they shoudl work, but I haven't had a chance to test Oct 13 18:09:19 bigd0g: with the file from my link, you don't need any migration or update, works out-of-the-box with mtm 0.30 (at least the one that is in Debian) Oct 13 18:09:43 Crofton: ok, thanks. i'll have to take a look Oct 13 18:09:53 Crofton: flying back home tomorrow. i'll have another board there, tho Oct 13 18:09:55 bigd0g: this snapshot is just short of 477 revisions from current head Oct 13 18:09:56 cool Oct 13 18:09:58 <[lala]> can anyone tell me how to add a custom mirror in local.conf? Oct 13 18:10:05 schurig: cheers Oct 13 18:10:14 <[lala]> MIRRORS () {...} doesnt work Oct 13 18:10:22 hopefully by the time you have your act together I will have tested Oct 13 18:11:26 haha ;) Oct 13 18:11:38 i've gotta convince the RA that wants my assistance that OE is the way to go Oct 13 18:11:48 he's using the crippled MontaVista at kernel 2.4 Oct 13 18:11:49 ick Oct 13 18:12:00 ug Oct 13 18:12:07 i'm sure it won't be too hasrd Oct 13 18:12:13 we should be very solid for producing base images now Oct 13 18:12:21 he should've seen the postings to the OMAP list by Dirk telling the other n00bs to stop using kernel 2.4 Oct 13 18:12:38 yeah, the 2.6.18-omap build works great Oct 13 18:12:46 even my custom spinoff with framebuffer support Oct 13 18:13:07 BTW: I got brainwashed by TT Oct 13 18:13:25 there's some other issue with framebuffer support that Dirk's config fixes... i haven't had a chance to isolate it yet, tho Oct 13 18:13:48 Texas Instruments? Oct 13 18:13:53 that the defconfig + fb support enabled doesn't select. that causes the kernel to not compile fully. some other dependency, i imagine Oct 13 18:13:54 <[lala]> koen, zecke: any hint for my mirror question? Oct 13 18:14:09 [lala]: no Oct 13 18:14:26 I think we will try and move the defconfig to support fb as modules Oct 13 18:14:38 [lala]: I would mkdir classes, copy base.bbclass to this directory Oct 13 18:14:44 [lala]: and change it Oct 13 18:14:51 may be beneficial. most would be using it with an lcd at some point, i would imagine Oct 13 18:15:05 especially with the qvga out Oct 13 18:15:13 [lala]: but something like MIRRORS_append in local.conf could work already Oct 13 18:15:16 <[lala]> zecke: i thought there would be an easier solution, because that way i have to manually update the class Oct 13 18:16:00 I need to get my OSK with the QVGA board to boot again Oct 13 18:16:58 <[lala]> zecke: MIRRORS_append () { ... } leads to a parse error Oct 13 18:17:27 [lala]: hmm, Oct 13 18:17:36 <[lala]> zecke: same for MIRROR or MIRROR_prepend Oct 13 18:17:51 [lala]: let me setup bonsai first. I think we will be done in about an hour Oct 13 18:18:02 <[lala]> zecke: thanks Oct 13 18:18:10 [lala]: there is an easy way, I don't remember it though :) Oct 13 18:18:17 <[lala]> :-) Oct 13 18:34:01 mickeyl: yes! Oct 13 18:34:12 hey zecke! Oct 13 18:34:12 back home? Oct 13 18:35:08 yes Oct 13 18:35:29 what did you do the remaining time? Oct 13 18:38:01 need to spend some time with Sabine now Oct 13 18:38:06 cu later! Oct 13 18:42:49 JustinP: hey :) Oct 13 18:45:52 <_law_|iBook> bbl cu Oct 13 18:49:16 ERROR: name 'base_contains' is not defined while parsing /usr/src/sdkmnci/org.openembedded.dev/packages/tasks/task-base.bb Oct 13 18:49:32 how comes this? Oct 13 18:49:49 dunoo Oct 13 18:51:58 schurig: perhaps you don't have the correct stuff set up to use task-base for your distro/machine? Oct 13 18:52:02 zecke: hey Oct 13 18:52:35 JustinP: Do you think it is time to change the semantic of include/inherit? Oct 13 18:52:49 JustinP: I try DISTRO = "angstrom-2007.1" Oct 13 18:52:50 JustinP: actually the semantic of include must stay Oct 13 18:53:03 and MACHINE = "gumstix" Oct 13 18:53:10 schurig: your base.bbclass is too old Oct 13 18:53:35 koen: can't be, I just did an mtn pull Oct 13 18:53:56 koen: I'm at 224a910a8fb730675edc6464917f5dad2261aeab Oct 13 18:54:21 and my bitbake is bitbake-1.6 Oct 13 18:55:21 schurig: classes/base.bbclass:def base_contains(variable, checkvalue, truevalue, falsevalue, d): Oct 13 18:55:29 schurig: so something is wrong at your end Oct 13 18:55:57 koen: yeah, the function *IS* in classes/base.bbclass. line 44 Oct 13 18:57:41 koen: when I put an "import sys; sys.exit(1)" at the top of classes/base.bbclass, then I can see clearly that this file is NOT loaded, my bitbake doesn't stop because of this Oct 13 18:59:08 schurig: so you BBPATH and BBFILES are wrong Oct 13 19:02:19 schurig: or you have a stale cache which isn't being updated Oct 13 19:02:44 I think it was my BBPATH Oct 13 19:02:59 zecke: I'm just saying what I think. It doesn't make sense to me to allow a recipe to continue when an include fails Oct 13 19:03:09 zecke: my bitbake/OE works fine with the patch I put in there ;-) Oct 13 19:03:12 zecke: at least AFAIK Oct 13 19:03:29 zecke: why would we want to allow missing include? Oct 13 19:03:36 JustinP: conf/bitbake.conf Oct 13 19:03:50 JustinP: it is okay to not have an auto.conf and site.conf or even valid ${DISTRO}.conf Oct 13 19:04:06 debatable, but sure Oct 13 19:04:12 some are I mean ;-) Oct 13 19:04:29 ~lart apache Oct 13 19:04:30 * ibot slaps apache upside and over the head with one freakishly huge killer whale named hugh Oct 13 19:04:32 ok, well, it seems like the recipes should be using "require blah.inc" then as they definately are required Oct 13 19:04:51 is there a require statement in bb? Oct 13 19:05:05 JustinP: sure Oct 13 19:05:16 JustinP: this was added as we couldn't change 'include' Oct 13 19:05:33 JustinP: back then I was too scared to change the behaviour of inherit Oct 13 19:07:45 JustinP: check the recipes before making a statement :) Oct 13 19:08:49 koen: :-/ apologies Oct 13 19:08:57 which bug was that? Oct 13 19:09:02 I'll revise/delete my comment Oct 13 19:09:20 phew... back-up dev computer built... time to head back to linux for archiving... bbl Oct 13 19:24:14 JustinP: if you want to make sure you get the file you want to "include", then simply "require" it Oct 13 19:27:12 woglinde: I can verify the fix proposed in bug 1421. glibc builds again. Oct 13 19:27:30 Anybody going to push this into .dev? Oct 13 19:27:39 I heard this might not happen. Oct 13 19:28:04 laibsch hm I can ask hrw if he allows me to push it Oct 13 19:34:38 That would be nice. I think this should be fixed and updated now that the reason for the failure is clear. Oct 13 19:35:16 Is it advisable to build openzaurus-uclibc? What is the reason that openzaurus does not use uclibc as default? Oct 13 19:35:52 why should it use uclibc? Oct 13 19:36:02 Laibsch: probably because people want too much stuff that doesn't work with uclibc. But take my comment with a grain of salt...I know nothing specific about uclibc Oct 13 19:36:33 uclibc is great if all your users speak english Oct 13 19:37:18 Internationalization is definately a problem Oct 13 19:37:35 hm Oct 13 19:37:38 slightly Oct 13 19:37:41 * Crofton is still embaressed to have thought mickey's thesis would be in Elnglish Oct 13 19:38:12 I see. Internationalizations buys me into standard right away ;-) I guess you all know I use the Z as a Japanese dictionary. Oct 13 19:38:37 koen: I thought uclibc is smaller and small is beautiful (especially for embedded) Oct 13 19:39:02 Its never that simple Oct 13 19:39:31 uclibc + iconv + gettext is roughly the size of glibc with iconv builtin Oct 13 19:39:47 Can I build packages with openzaurus uclibc and use them with a standard image? I guess not and that it will create big problems, right? Oct 13 19:39:50 and with 6gb microdrives, who's gonna care Oct 13 19:40:07 laibsch right Oct 13 19:40:29 There is never a good excuse for unnecessary bloat. (In this case I see the necessity, though, as said earlier) Oct 13 19:41:45 any 'apache' guru around? Oct 13 19:43:34 zecke: I've used apache a lot...I don't know about guru, though... Oct 13 19:43:57 JustinP: do you know 'Location'? Oct 13 19:44:40 zecke: a redirect header you mean? Oct 13 19:44:58 (Location header....) Oct 13 19:45:02 http://www.openembedded.org/bonsai/query/author/RP/ Oct 13 19:45:04 yeah Oct 13 19:45:18 what about it? Oct 13 19:45:42 JustinP: http://rafb.net/paste/results/qfuiBa55.html Oct 13 19:45:55 JustinP: this Location worked for bonsai/ but not for bonsai/query/author/RP Oct 13 19:46:03 RewriteEngine off Oct 13 19:46:55 koen: for the update process, I would like to have a mtn db Oct 13 19:47:32 some one feels like donating proper and good looking templates for the bonsai? Oct 13 19:53:15 zecke: sorry, was sidetracked. Did koen's suggestion fix it? Oct 13 19:53:21 I'm off to bed. Godd night. Oct 13 19:53:56 yes Oct 13 19:54:16 ^_^ Oct 13 19:56:53 koen: feed://www.openembedded.org/bonsai/feeds/author/rpurdie/ Oct 13 19:57:07 koen: next time richard breaks bitbake, you will know it :) Oct 13 19:57:35 :) Oct 13 19:58:36 koen: could you reload apache again? Oct 13 19:59:42 done Oct 13 20:00:08 hmm Oct 13 20:02:52 eeek Oct 13 20:04:49 hm? Oct 13 20:04:54 reboot Oct 13 20:04:57 it will help Oct 13 20:05:44 woglinde: did you use Windows recently? (Where a reboot is a cure to everything :-) Oct 13 20:06:20 zecke, did you say you were talking to some texas Instruments people? Oct 13 20:06:29 Crofton: no Oct 13 20:06:41 ah different TI then Oct 13 20:06:59 schurig he at work there is most of it windows Oct 13 20:07:07 Crofton: it was "brainwashed by TT", not by "TI" (TT is Trolltech) Oct 13 20:07:10 ah Oct 13 20:07:29 I think he mispelled it :) Oct 13 20:07:54 QT rocks! Oct 13 20:09:47 koen: we should consider using postgres ASAP Oct 13 20:10:19 zecke: right Oct 13 20:10:54 njs: http://www.openembedded.org/bonsai/feeds/author/njs/ Oct 13 20:11:45 zecke: http://www.openembedded.org/bonsai/view/scm/1/0/ Oct 13 20:12:34 koen: No lets put a best viewed with mosaic 2.3 button on the page Oct 13 20:12:37 koen: and announce it Oct 13 20:13:37 zecke: can I replace mtn.mtn with an OE.mtn? Oct 13 20:14:18 is there a valid fb.modes for qvga? Oct 13 20:14:32 koen: no, we can add OE as mtn system as well Oct 13 20:14:44 er. does anyone kow of the existence of one? Oct 13 20:14:54 oe@serenity:~/tinderbox-server/tinderbox$ ./bonsai_add.py -d "Monotone main" -t mtn -v "" -s "venge.net" -p "netsync" -m "/home/oe/tinderbox-server/mtn.mtn" -b "net.venge.monotone" Oct 13 20:15:53 koen: feel free to update to OE Oct 13 20:16:26 zecke: http://www.openembedded.org/bonsai/view/scm/3/0/ Oct 13 20:17:34 koen: ./bonsai_update.py "Openembedded dev" Oct 13 20:17:51 cool Oct 13 20:18:16 without any arguments everything will be updated Oct 13 20:19:23 koen: how do you like RSS feed? Oct 13 20:19:27 koen: do you prefer atoms? Oct 13 20:19:50 koen: sadly, during the mtn update, the bonsai is blocked :( Oct 13 20:20:52 so I will flash my simpad Oct 13 20:20:55 zecke: iirc atom feeds have a saner date and update handling Oct 13 20:21:30 koen: http://www.djangoproject.com/documentation/syndication/ Oct 13 20:21:50 koen: it is quite easy to add new outputs Oct 13 20:21:54 zecke: send me a mail with instructions on how to setup postgres for bonsai and I will have a look at it tomorrow Oct 13 20:22:17 woglinde: http://www.chicagocrime.org/ this should be of interest for you Oct 13 20:22:28 koen: mail? Oct 13 20:23:00 zecke: yes, a mail Oct 13 20:23:00 woglinde: how does one administrate postgres? Oct 13 20:23:00 hm how was the bl-command for cleaning partition Oct 13 20:23:07 zecke: so I won't forget Oct 13 20:23:12 zecke? Oct 13 20:23:14 woglinde: e.g. creating a new database Oct 13 20:23:31 create a user or user the postgres account Oct 13 20:23:39 on debian/ubuntu Oct 13 20:23:43 su - postgres Oct 13 20:23:46 create_db Oct 13 20:23:59 nothing easier than this Oct 13 20:24:19 without underscore ;) Oct 13 20:24:36 ok. found a working fb.modes :) Oct 13 20:25:08 cant remeber all exactly Oct 13 20:26:08 * koen looks at http://www.openembedded.org/bonsai/query/author/fre/ Oct 13 20:26:46 oh no push this month Oct 13 20:27:11 zecke: can we add 'maps' ? Oct 13 20:27:15 hm again what is the command in hh-bootldr do clean up a partition Oct 13 20:27:17 like freyther <> zecke123 Oct 13 20:27:18 not delete it Oct 13 20:27:38 woglinde: dunno Oct 13 20:27:44 woglinde: help, or check the manual Oct 13 20:28:18 zecke guess what I do Oct 13 20:28:30 partition delete -- deletes a partition Oct 13 20:28:41 is it reset? Oct 13 20:28:42 partition reset -- resets all partitions to the default Oct 13 20:28:42 woglinde: pflash? Oct 13 20:28:58 woglinde: it's no a 'partition' command Oct 13 20:29:00 pflash 0|1 -- (1 -> protect, 0 -> unprotect all) Oct 13 20:29:17 pflash unprotects it Oct 13 20:29:32 after that you should be able to run 'eflash' or 'erase' Oct 13 20:29:37 * koen tries to rememeber Oct 13 20:29:55 there is no erease mentioned in the help Oct 13 20:30:24 woglinde: how do I pass '/' in a URL? Oct 13 20:30:31 ??? Oct 13 20:30:51 query/file/path/to/file Oct 13 20:31:01 path/to/file should be one path :) Oct 13 20:31:16 query? Oct 13 20:31:25 what the hell you are talking about Oct 13 20:31:28 zecke: you rawurlencode it Oct 13 20:31:34 * JustinP speaks in PHP terms Oct 13 20:31:53 %2F Oct 13 20:35:11 hello. I've tinkered with my kernel a bit and I want bitbake to recompile it but it refuses without an error. I'm calling "bitbake handhelds-pxa-2.6" Oct 13 20:36:57 OE/bitbake is in a working state right now. Just not clear on how I recompile the kernel (handhelds-pxa-2.6-2.6.13-hh1+cvs20051127-r0/) Oct 13 20:38:56 bitbake -b bbfile -c clean Oct 13 20:38:58 bitbake -b bbfile Oct 13 20:39:10 clean will remove the directory though Oct 13 20:39:54 iirc, there's no handhelds-pxa-2.6-2.6.13 in repo any longer Oct 13 20:41:59 yikes. Yes I need to stick with this kernel for historical reasons. I' guess the only option is to build the kernel manually? Oct 13 20:43:15 jtodd, yep, that would be easiest if you need just kernel Oct 13 20:44:52 alright. thanks :) Oct 13 20:52:22 hms Oct 13 20:52:37 it seems glibc-2.5 istnt working right Oct 13 20:53:08 woglinde: why? Oct 13 20:53:14 on simpad Oct 13 20:53:33 no output after init Oct 13 20:54:56 CONFIG_NWFPE=y ? Oct 13 20:55:05 args right Oct 13 20:55:15 damn crappy default .config Oct 13 21:07:24 libtool gurus? Oct 13 21:07:46 nope Oct 13 21:08:37 * Crofton begs Oct 13 21:10:21 crofton maybee Oct 13 21:10:34 tell the problem Oct 13 21:11:05 one of the -L entries points to the host staging lib Oct 13 21:11:28 in .la file? Oct 13 21:11:28 and it contains a host copy of the lib, I need from the target staging area ... Oct 13 21:11:48 http://pastebin.ca/200971 Oct 13 21:11:57 .so Oct 13 21:12:53 which package? Oct 13 21:13:15 one I am working on Oct 13 21:13:24 * koen recommends 'sed' Oct 13 21:13:26 the bb file isn't commited yet Oct 13 21:13:53 hmm Oct 13 21:14:00 seeing a lot of references to x86_64 Oct 13 21:14:21 I can't even figure out were the -L that causes the probllem is injected .... Oct 13 21:14:24 hm thats not an libtool bug Oct 13 21:14:34 seems autotools Oct 13 21:14:44 Crofton: config.log is a good file Oct 13 21:14:46 I did the autofoo for the package ... Oct 13 21:14:51 hmmm Oct 13 21:14:53 good point Oct 13 21:14:57 or crappy Makefile.am and configure.ac Oct 13 21:15:19 woglinde, very very likely Oct 13 21:15:19 crofton its in the .la file Oct 13 21:15:20 I guess Oct 13 21:15:28 I wrote them Oct 13 21:15:35 hehe Oct 13 21:18:26 what is build and host refer too? Oct 13 21:18:56 build is the machine you build on, host is final hardware? Oct 13 21:19:11 hm Oct 13 21:19:14 no Oct 13 21:19:20 host is like parent Oct 13 21:19:26 target is the final Oct 13 21:19:28 the machine you build on Oct 13 21:19:43 host_cpu=arm? Oct 13 21:19:45 build is the process Oct 13 21:19:58 if you happen to own an epia borad that might work Oct 13 21:20:16 most people dont have arm boxes they can use to build on Oct 13 21:20:26 I should go drink Oct 13 21:20:50 emte: building on an arm box is slooooooooooooooow (at least on my slug) Oct 13 21:21:05 chouimat: get an ix2800 Oct 13 21:21:22 koen first I need $$$$$ ;) Oct 13 21:21:22 chouimat, yeah, i'd expect so ... especially on ones that dont have a FPU Oct 13 21:22:05 my configure lines from config.log report build=x86_64 --host=arm-linux --target=arm-linux Oct 13 21:24:00 i cant comment on that, i dont have anything near a recent version of bitbake or the OE meta Oct 13 21:24:25 it seems correct based on reading configure --help Oct 13 21:24:37 but it looks like your host is identifying wrong OR qemu is doing something Oct 13 21:24:51 host is target Oct 13 21:24:54 are my guesses Oct 13 21:25:18 that doesnt make sense Oct 13 21:25:45 if it were then you would have: build=x86_64 --host=x86_64 --target=x86_64 Oct 13 21:26:03 I am building on x86_64 for an ARM Oct 13 21:26:24 right so it should be: build=x86_64 --host=x86_64 --target=arm_linux Oct 13 21:26:37 hms Oct 13 21:26:47 do a configure --help Oct 13 21:26:54 unless, like i said earlier, qemu is involved there and emulating an arm system Oct 13 21:27:01 even without fp-emu no init with glibc-2.5 on the simpad Oct 13 21:27:05 no qemu Oct 13 21:27:35 Crofton, configure --help on what? Oct 13 21:27:45 the package your having proiblems with? Oct 13 21:28:23 anything, it explains build versus host Oct 13 21:33:00 * chouimat kicks openoffice for being a piece of shit Oct 13 21:33:43 chouimat not here Oct 13 21:34:02 right, the german ooo is scheisse Oct 13 21:34:15 noway Oct 13 21:35:20 I swear I am going to convert to cmake Oct 13 21:37:31 crap I am going to push the bb file and try it at home Oct 13 21:37:44 of course there will be some other forms of bustage there :( Oct 13 21:44:47 woglinde: here it simply die ... Oct 13 21:45:13 Crofton, in trems of autotools (as opposed to bitbake) build is the target(as in build for) and host is still the build machine ... unless you had a very wacky autotools author Oct 13 21:46:39 wacky and autotools go well together Oct 13 21:48:21 bother, now it would appear monotone.oe.org is sick Oct 13 21:48:28 I am going for beer Oct 13 21:48:38 gn all Oct 13 22:11:57 gosh... Oct 13 22:13:57 zecke: gosh? Oct 13 22:14:17 zecke: or was there a mis-placed " "? (go sh) ;-) Oct 13 22:14:33 a missing " fi" Oct 13 22:14:34 no, I'm losing against postgres Oct 13 22:14:50 ah.....yeah.... Oct 13 22:14:55 zecke: have some beers with bruce, he's a nice guy :) Oct 13 22:14:57 I set up my first postgres a little bit ago... Oct 13 22:15:07 somewhat harder than mysql ;-) Oct 13 22:15:23 luckily, Gentoo has a howto that helped me set it up Oct 13 22:15:37 of course, I'm not *using* it for anything except to test a package against it.... Oct 13 22:16:22 zecke: lets install glom :) Oct 13 22:16:23 koen: .decode('latin1').encode('utf-8') does the trick Oct 13 22:16:35 koen: the GNUstep glom? Oct 13 22:17:11 no, glom Oct 13 22:18:28 the gnustep on is called gorm Oct 13 22:19:53 koen: please delete the database again :) Oct 13 22:19:57 koen: and add the SCMs again Oct 13 22:21:09 done Oct 13 22:21:28 do you call bonsai_update, or can I call it? Oct 13 22:21:34 you Oct 13 22:22:21 * koen charges n770 and gps Oct 13 22:23:37 psokolovsky_: with Opie it is something completely different :) Oct 13 22:23:41 end Oct 13 22:23:43 psokolovsky_: I can invite you for a beer :) Oct 13 22:23:46 woglinde: rejected! Oct 13 22:23:51 no Oct 13 22:23:53 sleep now Oct 13 22:24:31 policy violation by woglinde Oct 13 22:25:23 koen: SVN updated Oct 13 22:25:46 cool Oct 13 22:26:02 let us see if mtn does the same :( Oct 13 22:26:19 or if my on the fly latin1 -> utf8 encoding is failing somewhere else Oct 13 22:27:35 nope Oct 13 22:27:38 damn Oct 13 22:30:31 * koen caches some maps with maemo-mapper Oct 13 22:30:48 koen: where are you heading to? Oct 13 22:31:00 koen: and why did you wrote this mail? Oct 13 22:31:18 to a 'veengebied' tomorrow Oct 13 22:31:22 zecke: which mail? Oct 13 22:31:36 koen: I will give you my phone number, so you can always call the 'Emergy hotline' Oct 13 22:32:48 .....3176 ? Oct 13 22:33:05 probably, Oct 13 22:33:45 BTW: could you install mc? Oct 13 22:34:23 mc installed Oct 13 22:43:25 everything indexed Oct 13 22:44:11 cool Oct 13 22:47:46 03hrw 07org.oe.dev * r2458e825... 10/ (4 files in 2 dirs): linux-omap1: split common stuff into linux-omap1.inc and fixed do_deploy() Oct 13 22:47:50 03hrw 07org.oe.dev * r8743eceb... 10/ (4 files in 3 dirs): linux-omap1: added 2.6.18 version from git which should work Oct 13 22:47:57 03pfalcon 07org.oe.dev * r948a3826... 10/ (1 MAINTAINERS): MAINTAINERS: Add myself. Oct 13 22:48:08 03hrw 07org.oe.dev * ra900b9d8... 10/ (1 conf/machine/omap5912osk.conf): omap5912osk: move PREF_VER of kernel to distro config Oct 13 22:48:17 03hrw 07org.oe.dev * r0e558e33... 10/ (1 conf/distro/openomap.conf): openomap: moved PREF_VER of ompa5912 kernel to distro config - set to 2.6.18+git Oct 13 22:48:22 03erik 07org.oe.dev * r224a910a... 10/ (1 MAINTAINERS): Oct 13 22:48:22 MAINTAINERS file Oct 13 22:48:22 * Add myself and required additional info. Oct 13 22:48:26 koen: now let us test if updating works?! Oct 13 22:48:31 03crofton 07org.oe.dev * ref615413... 10/ (1 packages/ossiecf packages/ossiecf/ossiecf_svn.bb): Oct 13 22:48:31 ossiecf.bb : Add bb file to build the open source SCA implementation from Va Tech. Oct 13 22:48:31 * bb file does not actually work at this time Oct 13 22:48:38 zecke: sure Oct 13 22:49:00 Sounds about right for something form Va Tech Oct 13 22:49:00 * CosmicPenguin runns Oct 13 22:51:11 zecke: is it possible to include a 'diff' function as well? Oct 13 22:51:38 koen: yes, it is present in the model Oct 13 22:51:40 eeks Oct 13 22:51:42 1:11 Oct 13 22:51:53 * koen goes to sleep Oct 13 22:51:59 koen: I will implement bonsai fights first Oct 13 22:52:04 bonsai fight koen/rp Oct 13 22:54:23 nite guys Oct 13 22:54:30 nite zecke Oct 13 22:58:43 koen: what is the password for the db? Oct 13 23:16:09 03justinp 07org.oe.dev * r61d62d88... 10/ (1 MAINTAINERS): MAINTAINERS: Add myself Oct 13 23:19:06 03justinp 07org.oe.dev * r2bb0057f... 10/ (1 MAINTAINERS): MAINTAINERS: fix alphabetical ordering Oct 14 01:38:46 03lenehan 07org.oe.dev * rcf3211db... 10/ (1 MAINTAINERS): MAINTAINERS: Add myself to the maintainers file. **** ENDING LOGGING AT Sat Oct 14 02:59:57 2006