**** BEGIN LOGGING AT Mon Apr 14 02:59:57 2008 Apr 14 07:00:55 Anybody here with edit access to www.openembedded.org/wiki/MainPage? Apr 14 07:01:11 It's been porn spammed and my corporate firewall won't let me connect Apr 14 07:03:19 I can connect with lynx, but I can't see any way to register as a user. Obviously the spammer succeeded... Apr 14 07:03:36 And you're all asleep :( Apr 14 07:10:27 Araneidae : looks ok to me. where is the spam ? Apr 14 07:11:18 The entry "Mon, 2007-02-12 11:26 -- Koen Kooi" is porn spam Apr 14 07:13:12 I've only got lynx access, but it begins: "free hardcore porn movies mature movie sex ..." and goes on from there Apr 14 07:15:42 steliosk, do you see it? Apr 14 07:16:06 Araneidae : no on firefox i don't see anything Apr 14 07:16:57 That's interesting. Look at the page source then: `curl www.openembedded.org/wiki/MainPage | less` and scroll down to Koen Kooi Apr 14 07:17:17 Apr 14 07:18:06 yeah it shows in the source code of the page Apr 14 07:18:37 Interesting. Can you kill it, please? I'd like to get at that page from work! Apr 14 07:19:44 * steliosk tries to remember his wiki password Apr 14 07:20:10 There doesn't seem to be a way to register as a new user, but at least you can ask it to mail you your password Apr 14 07:21:11 doing that Apr 14 07:21:30 I hate my corporate firewall Apr 14 07:22:23 Btw, do you know where OE puts its log files? I've got a failed poky build and I haven't found the logs yet! Apr 14 07:25:14 morning Apr 14 07:25:56 Araneidae : should be ok now Apr 14 07:26:23 I guess there's some caching to get past. Apr 14 07:26:46 Yes, that's better. Thanks Apr 14 07:27:24 Araneidae : there are no actual build logs on OE. (not sure about Poky though) Apr 14 07:27:51 if it fails you see it on the console Apr 14 07:28:18 I have a console log entry, but it's less than detailed Apr 14 07:28:46 NOTE: package glibc-2.5-r8: task do_package: started Apr 14 07:28:50 FATAL: kernel too old Apr 14 07:29:09 No build logs? That's not nice Apr 14 07:29:48 There was the sepuku bbclass that was logging builds to a web server but i am not sure if it works now on .dev Apr 14 07:29:48 hello - supposing i'd start a new distro based on OE some day, how are the chances to be incorporated into the official repo? Apr 14 07:30:47 i'm working with a local overlay now, and it will take some time until it's somewhat stable Apr 14 07:30:59 mrt2k7 : FYI first make sure that you understand the difference between a "distro" and an "image" in OE Apr 14 07:31:26 then post to the bugtracker the patch Apr 14 07:31:48 steliosk: ok...i'll try to Apr 14 07:32:31 i'm not just playing around with including or excluding packages, so i think it's more than a custom image? Apr 14 07:32:32 mrt2k7 : A distro defines the toolchain version and versions of packages that get used Apr 14 07:32:59 mrt2k7 : an image selects what gets inside it Apr 14 07:33:04 So a distro can have multiple images? That makes sense Apr 14 07:33:14 yes Apr 14 07:33:17 yeah sounds like what i' going to do Apr 14 07:34:27 the only problem i may see: i'm not building for "real" embedded devices, i found out that OE is the most powerful framework for "just enough operating systems" (JeOS, kind of buzzword these days) Apr 14 07:35:07 are you targeting virtual machines ? Apr 14 07:35:10 there is ubuntu-JeOS, some gentoo approaches etc., but nothing really compares to OE Apr 14 07:35:10 yes Apr 14 07:35:36 qemu/kvm, to be specific Apr 14 07:35:50 works fine although a vmware image would be nice ;) Apr 14 07:36:17 i had troubles to boot the vmware image i once created Apr 14 07:36:38 qemu is so nice because you just pass the kernel and the rootfs to the CLI Apr 14 07:36:48 never tried to build one myself Apr 14 07:37:01 it builds, but i don't know how to make it bootable Apr 14 07:37:18 and kvm is very fast and extremely small Apr 14 07:37:23 * steliosk adds another thing to the todo list Apr 14 07:37:55 * mrt2k7 get his phenom B3 today and can't wait anymore :) Apr 14 07:38:14 well i have been playing around automagically creating a grub multipartion image from inside OE Apr 14 07:38:31 sounds reasonable Apr 14 07:38:46 did you document it somewhere? Apr 14 07:38:56 but need to find a way to mount a loop filesystem without been root Apr 14 07:39:20 not yet, just figuring out what it needs to get done Apr 14 07:39:54 do you have blog or st. else where you would put it? (and I could see it in my rss? ;)) Apr 14 07:40:05 it is/was possible to generate syslinux images from within OE, but nothing with grub Apr 14 07:40:06 (when you are finished some day) Apr 14 07:40:18 but lilo would work? Apr 14 07:42:11 its not a matter of bootloader unfortunately, but that it takes root privilages to mount a loop filesystem, and create the partitions Apr 14 07:54:03 morning Apr 14 07:54:15 Araneidae: look into tmp/work/*/*/temp/ for build logs Apr 14 07:55:41 hrw : morning ! Apr 14 08:19:54 morning! Apr 14 09:55:55 Odd: to complete my build I had to add the line ENABLE_BINARY_LOCALE_GENERATION = "0" to glibc_2.5.bb Apr 14 09:56:30 Otherwise the quem-arm call around the locale builder fails with a kernel version error. Apr 14 09:56:39 What's the "right" way to solve this? Apr 14 09:56:49 Araneidae: what distro you use on host? Apr 14 09:59:07 zecke, florian: good morning Apr 14 09:59:37 hi Apr 14 10:00:08 RP: let me draw a conclusion on insane.bbclass and warning about cross packages Apr 14 10:00:16 Red Hat Enterprise Linux 4 Apr 14 10:00:43 Araneidae: without replacing your kernel that is the right way Apr 14 10:01:18 Well, I'm hoping there's a slightly more global place to put that hack? Apr 14 10:01:42 Or can I persuade it to do a fully emulated qemu, qemu-system-arm perhaps? Apr 14 10:02:54 Araneidae: put it in local.conf Apr 14 10:03:20 Much better idea! Apr 14 10:03:36 Araneidae: problem is Redhat in RHEL and Fedora Core include a kernel patch that breaks qemu Apr 14 10:03:50 Araneidae: hrw somewhere has a link that details the full situation Apr 14 10:06:01 hrw, if you have that link I'd be grateful Apr 14 10:10:59 Araneidae: poky website -> support -> handbook -> faq Apr 14 10:11:41 http://pokylinux.org/doc/poky-handbook.html#id3326890 Apr 14 10:16:56 Very clear, thank you Apr 14 10:18:26 it happen under all redhat based distros Apr 14 10:18:32 rhel, fedora, centos Apr 14 10:19:47 Trying to see what execshield is all about. Apr 14 10:20:45 Hmm. Reminiscent of a recent problem with Wine: the latest Ubuntu prevents mapping the bottom of memory, which can be troublesome. Apr 14 10:33:23 re Apr 14 10:50:11 XorA: ping Apr 14 10:50:16 ScaredyCat: pong Apr 14 10:50:45 whee do we talk about om's oe .. is here ok ? Apr 14 10:51:15 now that they're separated... as such Apr 14 10:51:37 there seems to be an issue with building linux-openmoko_2.6.24+git.bb Apr 14 10:52:16 seems to be a git based problem Apr 14 10:52:17 fatal: failed to unpack tree object f54af99850ae25da80e491b1864bb90e14c80b1f Apr 14 10:52:21 ScaredyCat: yeah :-( Apr 14 10:52:27 bt there's not much debug info Apr 14 10:52:29 ScaredyCat: I know not why that is happening Apr 14 10:53:13 ScaredyCat: it seems if you fetched during time OM git was fucked, then your tarball ends up fucked Apr 14 10:53:22 :/ Apr 14 10:53:28 ScaredyCat: and the tarball on downloads.om.org suffers as well Apr 14 10:53:35 I don't seem to have the tarball though Apr 14 10:53:40 ScaredyCat: rm -rf ../sources/git/git.openmoko.kernel* Apr 14 10:53:59 ScaredyCat: rm -rf ../sources/*git.openmoko.kernel.* Apr 14 10:54:12 ScaredyCat: then comment CVS_TARBALL_STASH in openmoko.conf Apr 14 10:54:59 ta.. Apr 14 10:55:06 when did it bugger up ? Apr 14 10:55:59 ScaredyCat: friday Apr 14 10:56:07 Hmmm.... Apr 14 10:56:24 ScaredyCat: sysadmin set some memory limits and git-server overran them Apr 14 10:56:24 I didn't even have my laptop on on Friday... Apr 14 10:56:51 in fact sunday night was the first time... Apr 14 10:57:01 oh well.. Apr 14 11:06:14 mmmm Apr 14 11:06:31 not so good when a mirror gives you a 403 Apr 14 11:06:39 whoes mirror Apr 14 11:08:30 ewi546.ewi.utwente.nl Apr 14 11:08:55 and removing the git*om stuff doesn;t seem to cure the problem either Apr 14 11:09:09 fatal: failed to unpack tree object 3484443ede44120e7011c2ff18a4553cae33d7d3 Apr 14 11:09:25 ScaredyCat: shit Apr 14 11:09:40 ScaredyCat: it seemed to work here, maybe the error just hasnt been thrown yet Apr 14 11:16:44 03koen 07org.oe.dev * re3a757b2... 10/ (1 conf/distro/include/angstrom-glibc.inc): angstrom-glibc: reintroduce -frename-registers for non armv7a builds Apr 14 11:21:21 ScaredyCat: NOTE: package linux-openmoko-1_2.6.24+git20080414-r0: task do_fetch: completed Apr 14 11:22:04 did it get a tarball or build it ? Apr 14 11:22:29 ScaredyCat: it did a fresh checkout from scratch Apr 14 11:22:49 f54af99850ae25da80e491b1864bb90e14c80b1f ? Apr 14 11:22:59 /home/dp/zaurus/sources/git_git.openmoko.org.git.kernel.git_f54af99850ae25da80e491b1864bb90e14c80b1f.tar.gz Apr 14 11:25:31 http://pastebin.ca/984547 Apr 14 11:47:50 ScaredyCat: looks like you didnt nuke traces in sources directoru Apr 14 11:48:10 ? Apr 14 11:49:19 ScaredyCat: all the old opemmoko kernel tarballs and the git/*openmoko.kernel* directories Apr 14 11:56:10 missed the first one Apr 14 11:58:34 still doesn;t seem to make any difference :( Apr 14 12:02:08 :-( that seems to work here Apr 14 12:02:25 ScaredyCat: gimme a while, ill upload me tarball Apr 14 12:02:35 thanks Apr 14 12:10:19 NOTE: package linux-openmoko-1_2.6.24+git20080414-r0: task do_compile: started Apr 14 12:10:34 I just removed sources/git* :) Apr 14 12:19:45 everything? Apr 14 12:21:39 Yes Apr 14 12:22:11 mmm.... bit drastic .. no? Apr 14 12:22:38 It's just sources Apr 14 12:22:48 I didn't remove my whole temp Apr 14 12:25:13 yea, but they're sources from a particular point in time.. Apr 14 12:25:20 * ScaredyCat likes to hoard stuff Apr 14 12:57:09 Hello all. I have a small ipkg question, is this an appropriate place to ask? Apr 14 12:57:46 jussio1: possibly Apr 14 12:58:29 jussio1: if its about a distro use of ipkg then find the distro support, if its about ipkg/opkg itself then propably here is fine Apr 14 13:00:03 if you'd said #openmoko he'd have got recursion Apr 14 13:01:55 thanks guys. getting some help from thos in #openmoko :) Apr 14 13:11:02 ok, Apr 14 13:11:07 got past it now... Apr 14 13:11:09 but Apr 14 13:11:24 defconfig-gta01 is missing from the repo Apr 14 13:11:51 last time I just copied it in... but... well... it should be there, tight? Apr 14 13:11:58 s/tight/right/ Apr 14 13:16:36 I got all the way to cacao-initial_0.98.bb, as usually Apr 14 13:18:53 but you're trying to build everything - right, cacao-initial isn't used Apr 14 13:18:57 ? Apr 14 13:19:51 No, it's after the om image and qt image is built Apr 14 13:21:12 The new ecore and edje stuff was built before at least Apr 14 13:31:29 03bluelightning 07org.oe.dev * r1bf5b0ba... 10/ (4 files in 3 dirs): opie-cardapplet: add recipe. courtesy of bluelightning. hopefully closes bug 944. Apr 14 13:39:22 03hrw 07org.oe.dev * rbfd97eee... 10/ (3 files in 2 dirs): opkg: really disable GPG in native and sdk (from Poky) Apr 14 13:58:55 hrw : Any idea how to handle additional arch's for i86 targets when optimizing for p4, p4 celleron etc Apr 14 13:59:09 we now have x86 i585 and i686 packages Apr 14 13:59:21 and we have "geode" Apr 14 13:59:23 a p4 what arch should use Apr 14 13:59:25 ? Apr 14 13:59:39 thats what no one know Apr 14 13:59:57 i386 generic "generic" Apr 14 13:59:59 as p4 can have sse, sse2 but maybe not sse3 Apr 14 14:00:10 i am afraid that we will end up with same arch but different optimization Apr 14 14:00:33 not each i686 is i686 indeed Apr 14 14:01:02 http://blog.haerwu.biz/2007/08/26/x86-architecture-mess/ Apr 14 14:01:23 Which optimisations can be used on "i686´´? It has MMX for sure but does it have SSE? SSEv2? SSEv3? 3DNow! technology? It depends on cpu... for example PentiumPro (the first "i686´´) has MMX but lack any other addons. Athlon64 will have most of them (or even all in newest cores) but under 32bit Linux it still be "i686´´. Apr 14 14:01:57 x86 is mess Apr 14 14:02:44 i do not disagree but we need to find a way to handle the mess in OE Apr 14 14:03:16 i.e i am adding support for the intel d201gly2 board which is p4 celeron Apr 14 14:03:26 steliosk: a config file with all the supported option for each cpu ... I think gentoo has something like this somewhere Apr 14 14:03:39 what arch should the packages be generated ? Apr 14 14:03:57 brb Apr 14 14:04:06 steliosk: i686? Apr 14 14:04:31 i686-p4 Apr 14 14:04:33 and if you have multimachine builds and have for example i686 packages from i686-generic that are build with specific optimization how do they mix with the ones build for the new target ? Apr 14 14:04:44 or rather i686p4 Apr 14 14:05:45 steliosk: you should have no problem by mixing the packages Apr 14 14:06:59 they mix ugly Apr 14 14:07:03 i am not so sure. i686 arch is -march=pentiumpro -mtune=pentiumpro d201gly2 for example is -march=pentium4 -mtune=celeron Apr 14 14:07:23 steliosk: use i686p4 as PACKAGE_ARCH Apr 14 14:07:55 hrw : Do we need to rfc that ? Apr 14 14:08:03 otherwise your packages will be installable on progear which has transmeta crusoe but will not be able to run SSE Apr 14 14:08:07 steliosk: would be nice Apr 14 14:11:43 hrw : nocona has x86-32-nocona Apr 14 14:12:00 it looks like everyone is using a different naming Apr 14 14:12:31 03hrw 07org.oe.dev * race8bd50... 10/ (1 packages/opkg/opkg-nogpg_svn.bb): opkg: added version without GPG support (from Poky) Apr 14 14:13:42 yep Apr 14 14:13:49 I used 'geode' for amd geode lx Apr 14 14:14:07 and it use amd k6-2 optimisations Apr 14 14:15:19 i think better to send an email to the devs list describing the problem and then discuss about an rfc Apr 14 14:17:39 yo Laibsch Apr 14 14:18:00 howdy! Apr 14 14:18:19 Laibsch: finally re-installed brain and got an envelope ready to send you tomorrow Apr 14 14:18:35 Hehe Apr 14 14:18:39 Don't worry Apr 14 14:18:49 I was away for a few days anyway Apr 14 14:18:55 I'll be just in time Apr 14 14:19:00 a modern concept Apr 14 14:19:09 Thanks again! Apr 14 14:19:38 no probs Apr 14 14:20:03 ~lart Monday's Apr 14 14:20:03 * ibot smacks Monday's up side the head with a clue-by-4 Apr 14 14:27:01 Greetings Apr 14 14:27:08 http://www.youtube.com/watch?v=AgqEIp2YmtE Apr 14 14:27:14 :> Apr 14 14:27:52 NineX: a bit late ;D Apr 14 14:28:03 hrw: ? Apr 14 14:28:31 Engineer's Guide to Cats http://www.youtube.com/watch?v=mHXBL6bzAR4 Apr 14 14:28:36 hrw: ^^^^ Apr 14 14:28:40 hrw: there are still users in poland that use dialin connection Apr 14 14:28:43 NineX: that link goes over net for quite long time ;) Apr 14 14:28:46 today Apr 14 14:29:07 hrw: aa ok Apr 14 14:29:18 * hrw neved owned modem Apr 14 14:29:45 my second machine had modem connected for ~20 minutes during one home computer party Apr 14 14:30:16 and once I did webservice at work remote hacking via modem (1 minute work) Apr 14 14:30:36 * Crofton owned a 1200 baud modem Apr 14 14:30:47 * Crofton used a 300 baud modem Apr 14 14:30:52 * chouimat|work too Apr 14 14:31:24 * mwester owned a 110 baud TeleType machine... Apr 14 14:31:33 mwester, wins Apr 14 14:31:47 my first internet access was 9600 serial terminal at university Apr 14 14:32:03 hmm Apr 14 14:32:07 beloved HP2623A from 70's (probably) Apr 14 14:32:17 why does linux-davinci kernel recipe hang during fetch Apr 14 14:32:27 after a 403 from ewi546 Apr 14 14:32:38 my first internet connection was in 1995 Apr 14 14:32:44 hrw: 9600 is high speed :-) Apr 14 14:32:51 XorA: in 1996? Apr 14 14:32:57 XorA: history Apr 14 14:33:13 hrw: I was using a 2400 modem until 1998 Apr 14 14:33:24 XorA: you win Apr 14 14:33:46 I started with sunOS account which had 1/1.5MB quota Apr 14 14:33:51 hrw: used to be lots of "free" X25 gateways you could dial into in the UK Apr 14 14:33:52 i need to search my ond 9600 exentral modem Apr 14 14:34:05 but does not stopped users from having 50M of data on account ;D Apr 14 14:34:08 * XorA has a zx spectrum 1200/75 modem which Ive never used Apr 14 14:34:23 XorA: you win! Apr 14 14:35:07 i have somwhere my Amstrad shneider cpc 464 Apr 14 14:35:23 with inregrated tape recorder Apr 14 14:35:25 I had a 6128 which I sold a couple of years ago Apr 14 14:35:37 I always sold old computers when buy new Apr 14 14:35:46 NineX: there was no others cpc464 ;D Apr 14 14:35:51 hrw: i have prety collection Apr 14 14:36:12 I decided ZX Spectrum was the only ones I really liked and sold the rest Apr 14 14:37:05 o i found LDW 2000 tygrys turbo floppy for my atari (tygrys turbo was a pretty polish technology to make floppydisc destroyer) Apr 14 14:37:30 I owned few machines: Atari 65XE + tape + disk drive + 12" green display. then amiga600 + hdd + same display. then amiga1200 + hdd + cd + same display (replaced by mono vga), then switched to pc Apr 14 14:37:37 NineX: 240KB per page one? Apr 14 14:37:46 yeap Apr 14 14:38:34 I had sn-360 - ds/dd 360KB per disk. Polish, slow, ugly but better then tape Apr 14 14:39:01 hmm, I think I am pulling a load of stuff from linux-davinci git Apr 14 14:41:55 hrw: do you remeber tyrpo cartriges to atari? Apr 14 14:42:13 hrw: big fun you wait to game load ~10 minutes tahn 1 hour Apr 14 14:42:25 s/tyrpo/turbo/ Apr 14 14:43:02 first cartrige for my atari was assembler editor/compiler :> Apr 14 14:43:50 NineX: which turbo? I had 2000F and there was looot of them Apr 14 14:44:25 I once bought original Atari Assembler. costed 2 pln before denomination ;D Apr 14 14:45:16 hrw: i don't remember i need to find it Apr 14 14:45:49 ok - but this is #oe not #old-computers Apr 14 14:45:59 hehehe Apr 14 14:46:08 port fo oe to comodore ;) Apr 14 14:46:12 we make a 8bit compile farm for OE :-) Apr 14 14:46:33 XorA: 64x zx81 with microdrives Apr 14 14:46:38 XorA: cluster from 40 000 c64 Apr 14 14:46:44 nice cluster would it make Apr 14 14:46:54 we could put smallC in OE as a cross compiler :-) Apr 14 14:46:55 * mwester can contribute a KIM-1 (6502-based) to the build farm Apr 14 14:47:20 mwester: first C= computer one? Apr 14 14:47:45 nope - it ws original MOS product Apr 14 14:48:57 then C= bought mos and re-released it Apr 14 15:05:23 bye Apr 14 17:28:49 is there any easy way to setup OpenMoko simultaneously? Apr 14 17:32:12 s/simultaneously/simultaneously with another OE distro/ Apr 14 17:49:08 probably not a good idea . OE is multimachine not multi-distro Apr 14 17:51:04 * * OE Bug 4178 has been created by gwossum(AT)acm.org Apr 14 17:51:06 * * binutils 2.17.atmel.1.2.6 and gcc 4.2.2.atmel.1.0. 8 for AVR32 Apr 14 17:51:08 * * http://bugs.openembedded.net/show_bug.cgi?id=4178 Apr 14 17:51:40 steliosk: so you think I should set up OE twice? Apr 14 17:55:33 Is it the correct place to ask questions about bitbake and its bb files? Apr 14 17:55:35 summatusmentis: yes, that's how I do it. However, you can share the downloads/sources directory using sym links -- that can save a lot of duplicate downloads. Apr 14 17:55:48 summatusmentis: I have been using the same setup for OE work and openmoko. I just set up different conf files and separate tmp directories. Apr 14 17:56:48 apaulsen: how do you have different conf files? something like OE/build/{conf_oe,conf_om} ? Apr 14 17:57:22 that's what I do, and just link local.conf to the one I want to use Apr 14 17:58:11 hmm, that's intriguing Apr 14 17:58:15 and define the corresponding tmp directory in the conf file as well. so conf_om would specify tmp_om Apr 14 17:59:47 Only problem with that is that om uses a different branch, so you can't really share the OE and OM openembedded stuff; you'll also need to separate that. Apr 14 18:00:17 Can bitbake handle Subversion authentication, if possible interactively? I'm trying to write a package that accesses the source tree at work. And this source tree only accepts authenticated users. Apr 14 18:00:59 summatusmentis: I haven't used om in a few months so I can't say if it still works :) Apr 14 18:01:35 Nahor: I usually work around this problem by checking out the source somewhere once, because .svn will remember the auth token and store it in your home directory Apr 14 18:01:39 mwester: so you just maintain two different dirs? Apr 14 18:03:01 uh oh... I can't find my stylus Apr 14 18:03:35 Jin: Yes, that was my fallback. It would have been cleaner if bitbake could handle it though. Guess I'll do that then. Thanks Apr 14 18:04:03 summatusmentis: if they do use a different branch you would have to use a separate dir, but could share the download directory, as mwester said Apr 14 18:04:05 Nahor: I was also looking for some auth via bitbake but did not find anything so far, not sure if I missed it or if it is not there Apr 14 18:05:12 Jin: I would not be surprised if it's not there. I don't think it usually required Apr 14 18:05:49 Jin: I've seen a "name" option which I assume would be the username to use for anonymous login Apr 14 18:06:00 Jin: but so far, nothing about a password Apr 14 18:06:03 * * OE Bug 4179 has been created by jin(AT)mediatomb.cc Apr 14 18:06:05 * * external toolchain generated by meta-toolchain depends on the original Apr 14 18:06:08 * * http://bugs.openembedded.net/show_bug.cgi?id=4179 Apr 14 18:07:18 I also did not find anything regarding pw Apr 14 18:07:37 yay, my bug :> guys - a really weirdo problem with meta-toolchain in oe.stable Apr 14 18:07:52 so if anyone wants to have a look Apr 14 18:07:55 :) Apr 14 18:08:12 * Jin^eLD went into the "annoy devs after submitting a bug" mode :> Apr 14 18:13:08 Jin: looking at the source code, it seems to support a "svn://user:pwd@svn.server.com/...." Apr 14 18:16:09 hmm, I could not get it to work... but I did not look at the source Apr 14 18:20:36 Jin: it works Apr 14 18:20:58 Jin: the only problem is that the password is then in clear in the bb file :( Apr 14 18:21:32 oh... Apr 14 18:23:56 flo_lap: ping :> Apr 14 18:24:02 do you have a second? Apr 14 18:25:08 Jin^eLD: yes Apr 14 18:25:09 re Apr 14 18:25:39 flo_lap: some real weirdness with meta-toolchain again :) it can be easily worked around, but still good to be aware of the problem Apr 14 18:25:49 so if you have time to read the PR: http://bugs.openembedded.net/show_bug.cgi?id=4179 Apr 14 18:26:23 took quite some time to figure out what it was, ran into this twice Apr 14 18:32:04 * * OE Bug 4180 has been created by grigorios(AT)prasinos.eu Apr 14 18:32:06 * * WPA not working on C750 with Angstrom 2007.12-r13 Apr 14 18:32:08 * * http://bugs.openembedded.net/show_bug.cgi?id=4180 Apr 14 18:34:54 hrm, when I run "ipkg update" , it downloads the list files to /var/lib/ipkg/packagefile but when i go to install something, it cant find anything. could someone offerme some pointers on how to fix this? Apr 14 19:34:35 zecke: ping Apr 14 19:38:56 pong Apr 14 19:42:54 zecke: Where do the header files of qt3 end up if not in the -dev package? Apr 14 19:43:29 * mickeyl suspects Nirvana Apr 14 19:43:40 flo_lap: hehe, manually copied into staging. I didn't care for qt3 when doing the 'cleanup' Apr 14 19:44:14 zecke: I just tried to build it again, but after building qt4 it seems qt3 is not buildable any more. Apr 14 19:44:31 that makes fixing it quite hard... Apr 14 19:47:04 * flo_lap is tempted to try crosscompiling it manually, but that might be a major pain Apr 14 19:49:20 flo_lap: why is it unbuildable after qt4 is built? there shouldn't be any kind of conflict Apr 14 19:49:53 zecke: I suspect an incompatible render extensions Qt4 pulls in. Apr 14 19:50:35 I get errors like this: Apr 14 19:50:37 shared-mt/qdesktopwidget_x11.o kernel/qdesktopwidget_x11.cpp Apr 14 19:50:38 | /disk/d1/oe/build_multimachine/staging/arm-angstrom-linux-gnueabi/include/X11/extensions/Xrandr.h:88: error: 'Connection' does not name a type Apr 14 19:52:50 flo_lap: yeah, somehow the headers of randrproto changed... they lack such definitions as XID... so a missing X.h include Apr 14 19:54:35 zecke: ok, so I patch that qt stuff to include X.h before, right? Apr 14 20:01:24 flo_lap: well, try that (I did that to Xglamo for the gta02). But the proper fix is somewhere else :( Apr 14 20:03:20 zecke: I just need to make it build for now :) Apr 14 20:03:43 hehe Apr 14 20:03:55 And to avoid questions: Not deliberately! Apr 14 20:10:14 hi everybody Apr 14 20:11:26 hi The_Niz Apr 14 20:11:31 hrmm Apr 14 20:11:38 hi thesing Apr 14 20:15:58 hi all Apr 14 20:16:02 zecke:flo_lap: about renderproto changes, look at how sirfred closed oebug 3635 Apr 14 20:16:18 thesing: sorry I could not test your patches yet, any progress in the bootimage field? Apr 14 20:16:38 damn webchat... Apr 14 20:16:42 thanks Apr 14 20:16:46 !oebug 3635 Apr 14 20:16:47 * * Bug 3635, Status: RESOLVED (FIXED), Created: 2008-01-06 14:38 Apr 14 20:16:48 * * : xserver-kdrive-imageon-1.2.0-r3-do_compile Apr 14 20:16:49 * * http://bugs.openembedded.org/show_bug.cgi?id=3635 Apr 14 20:18:05 ant|work: no I'm doing kernel work atm. Apr 14 20:18:32 I see Apr 14 20:49:47 is it just me, or is oe.org down? Apr 14 20:51:42 thesing: it is... strange Apr 14 20:51:48 * flo_lap is not guilty Apr 14 20:52:17 o-hand is down to. Apr 14 20:56:35 strange... I got pages from the server a few minutes ago. and it was quite fast Apr 14 20:58:47 thesing: Looks like network trouble, it took me some time to get a connection but the machine is more or less idle. Apr 14 20:58:50 hi all! Apr 14 20:59:04 hi denix Apr 14 20:59:42 can I ask you something? Apr 14 21:00:40 I'm trying to use monotone, but my firewall block me... how can I manage to use it on port 80? Apr 14 21:02:22 ps: I want to build h6300 images Apr 14 21:03:04 Last time I ran into this I used a SOCKS proxy Apr 14 21:03:38 I cannot change firewall settings Apr 14 21:05:00 thesing: its back for me Apr 14 21:06:30 flo_lap: not here yet. Apr 14 21:15:10 hi marex! Apr 14 21:16:05 hello, no idea what you want from me Apr 14 21:16:39 hehehe you was developing for h6300 right? Apr 14 21:17:26 no Apr 14 21:17:36 ops, then sorry Apr 14 21:17:49 what do you need about h6300 ? Apr 14 21:18:15 I want to build angstrom images Apr 14 21:18:49 I've my pc at work that is doing nothing during the night and I would love to compile and test angstrom images :) Apr 14 21:21:38 and what's the problem ? Apr 14 21:22:47 monotone, I cannot use it because of my firewall :( Apr 14 21:23:05 and in my laptop will take years to compile... do you have any hint? Apr 14 21:23:23 denix: just download a snapshot Apr 14 21:23:40 theres is an howto using a snapshot? Apr 14 21:24:41 denix: ths the mtn databse file - unack it and check it out like you would do after pulling it from the server Apr 14 21:25:36 flo_lap: so I don't need to use monotone... Apr 14 21:25:49 how many giga it will use? Apr 14 21:26:22 denix: you need to, but you don't connect to a server. Apr 14 21:26:31 denix, just download the OE.mtn.bz2 or how's it called Apr 14 21:26:46 bunzip it and use it ... Apr 14 21:27:13 well anyway, if you have some external victim, you can use httptunnel to bypass the firewall/proxy Apr 14 21:27:45 unfortunately I've no other machines where I can do it :( Apr 14 21:28:20 somebody can suggest to the oe group to put a monotone server on port 80? :D Apr 14 21:31:47 no Apr 14 21:35:26 denix: or you can download the snapshot... Apr 14 21:35:55 port 80 is actually occupied with somethign different ;) Apr 14 21:36:23 hehehe right :) Apr 14 21:36:25 how can I let bitbake assume QA issues are not errors but warnings? Apr 14 21:36:25 <|miska|> :-D Apr 14 21:36:27 8080? Apr 14 21:38:33 good morning! Apr 14 21:39:28 I'm playing around with x11-image on the TI OMAP3 EVM and encountering an issue I haven't seen before Apr 14 21:40:13 The touchscreen seems to work enough to move focus to text entry boxes on the "New user details" screen at boot Apr 14 21:40:33 And it also will activate the OK button Apr 14 21:40:52 But soft keyboard entry doesn't seem to work Apr 14 21:41:09 The keys highlight, but no text appears in the text boxes Apr 14 21:41:42 Any pointers on where to start looking? I assume this is the xkbd app? Apr 14 21:44:18 sakoman: yes it is... Apr 14 21:44:47 flo_lap: could it be a config issue? Apr 14 21:44:57 sakoman: I have seen this effect too... iirc that was on an X86 some time ago. Apr 14 21:45:55 flo_lap: any resolution to the issue? Apr 14 21:46:01 sakoman: did you build this from the unstable branch? Apr 14 21:46:17 if .dev = unstable, yes :-) Apr 14 21:46:35 sakoman: iirc i threw away that image and used one built from stable :-} Apr 14 21:47:06 sakoman: Maybe xtest extensions are missing, xkbd uses these to generate the events. Apr 14 21:47:28 thanks! I'll look into that Apr 14 21:47:37 flo_lap, OMAP3 is going to need bleeding edge gcc Apr 14 21:47:58 hmmm indeed Apr 14 21:48:12 * flo_lap has not yet seen such a device Apr 14 21:48:27 but OMAP3 sounds promising :) Apr 14 21:48:53 hmm . . . whaty package builds xtest? Apr 14 21:49:37 could it be libxtst? Apr 14 21:50:23 grep -r -i xtest pacakges/* Apr 14 21:51:36 I think it must be libxtst rather than xtest Apr 14 21:52:06 in xorg-lib Apr 14 21:56:16 ibot: Botmail for zecke: Found the issue: task-sdk-extra was polluting staging with an outdated randr.h Apr 14 21:58:00 nite Apr 14 22:09:06 nite! Apr 14 22:34:53 * flo_lap is amazed how long Qt is able to occupy his Xeons Apr 14 22:36:57 mickey|zzZZzz: good night Apr 14 22:37:09 night florian Apr 14 22:37:17 (feel the power of c++ ;) Apr 14 23:09:55 Is it normal that if I set EXTRA_OEMAKE in the bb file, make doesn't run the same compiler? Apr 14 23:10:20 If I don't set EXTRA_OEMAKE, make will run "ccache arm-linux-uclibc-g++" with a bunch of other parameters Apr 14 23:10:39 If I set it, make runs "g++" Apr 14 23:24:17 Found the solution, I needed to add a "-e" in EXTRA_OEMAKE Apr 14 23:48:16 Jin|away: very strange... Apr 14 23:56:06 03xjqian 07org.oe.dev * r4eb84840... 10/ (36 files in 10 dirs): wlan-ng: drop obsolete (<0.2.7) recipies. Apr 15 00:36:27 03tmbinc 07org.oe.dreambox * r5e51a233... 10/ (3 files in 3 dirs): linux-dm800: add linux-2.6.12-fixup-memsize.patch to use memory size from bootloader Apr 15 00:36:30 03tmbinc 07org.oe.dreambox * re1d2b261... 10/ (1 packages/dreambox/dreambox-secondstage.bb): dreambox-secondstage/dm800: update to version 53 to fix memory size for kernel **** ENDING LOGGING AT Tue Apr 15 02:59:56 2008