**** BEGIN LOGGING AT Tue Nov 20 02:59:57 2007 Nov 20 05:41:56 mickey|ICE1092: do you still have the OE poodle? Nov 20 05:42:37 good morning Nov 20 05:42:57 hvontres|home: yes, it should sit somewhere in my office, packed Nov 20 05:43:04 you want it? Nov 20 05:44:01 mickey|ICE1092: a second machine for testing would be good. Do you rememeber if it is a pxa250 or a pxa255? Nov 20 05:44:34 i don't remember, but i _think_ it's a 250 Nov 20 05:44:37 hrw|gone: ? Nov 20 05:44:55 hvontres|home: send me your snail mail address and i'm going to ship it to ya Nov 20 05:45:48 msg mickey|ICE1092 Henry von Tresckow,1646 Shire Ave, Oceanside,CA 92057 Nov 20 05:45:58 mickey|ICE1092: oops... Nov 20 05:46:07 * Zero_Chaos gives hvontres|home a / Nov 20 05:46:56 * hvontres|home also accepts cold hard cash at that address :) Nov 20 05:47:12 * Zero_Chaos sends hvontres|home a letter bomb Nov 20 05:47:16 heh Nov 20 05:47:18 usa? Nov 20 05:47:18 canada Nov 20 05:47:21 there goes this month's salary for shipping :D Nov 20 05:47:23 heh Nov 20 05:47:23 (that looks like a poodle) Nov 20 05:48:04 mickey|ICE1092: well, I was talking to XorA yesterday, and he was going to be in forida in Febuary Nov 20 05:53:49 mickey|ICE1092: I just wanted to see if yours was a 255 or not.. in case there are any differeneces Nov 20 05:54:26 CoreDump: ping Nov 20 05:56:01 good Nov 20 05:58:31 hvontres|home: hrw should know for sure, he had it before me Nov 20 05:59:52 mickey|ICE1092: Ok. have a good trip Nov 20 06:00:22 goodmorning all Nov 20 06:00:32 thanks Nov 20 06:00:37 4 hours to Berlin... Nov 20 06:00:49 morning steliosk. pdf will come as soon as I'm back in Frankfurt (tomorrow, given there's no train strike) Nov 20 06:01:34 mickey|ICE1092 : Thanks. Nov 20 06:05:48 * steliosk looks at his todo list and wonders how much long it can get Nov 20 06:08:39 unlimited :D Nov 20 06:16:11 mickey|ICE1092 : look like it... Nov 20 06:32:51 03ifaistos 07org.oe.dev * r9e8440e0... 10/ (5 files in 3 dirs): Nov 20 06:32:51 packages/glibc/glibc_2.5.bb : Add a new patch for the powerpc targets with softfloat Nov 20 06:32:51 support that handles a case of strong alias for two (soft) float point functions. Nov 20 06:32:51 Original patch provided by Hiroki Kaminaga kaminaga at sm dot sony dot co dot jp Nov 20 07:08:08 mickey|ICE1092: tunnel? Nov 20 07:09:44 yep Nov 20 07:09:44 gprs reception totally sucks on this track Nov 20 07:17:29 on a build server my OE build breaks with "ERROR: cannot flush diversion to temporary file: No space left on device" written in some log_do.configure. Is this temp file written to /tmp ? Nov 20 07:18:21 the error comes from M4 .. Nov 20 08:09:01 hi mickey, ICE1092 as in a train or in a gcc ICE? (just made me wonder...) Nov 20 08:09:57 hvontres|home: all OE poodles are 250 Nov 20 08:10:32 mickey|ICE1092: GPRS/UMTS sucks on every track in .de above 200km/h Nov 20 08:11:29 nik0n: good thing that the danish IC3 trains don't go above 180km/h. Nov 20 08:11:57 esben: ICE in germany goes to 300 - 320 Nov 20 08:12:14 like munich - nuremberg Nov 20 08:12:35 Today the pilot project called "Internet in Train" for the danish railroad (DSB) will be launched. It is UMTS based, and the router is of-course running OE :-) Nov 20 08:12:45 I should now ;-) Nov 20 08:13:18 question: i build up an image with a 2.4 series kernel and a jffs2 root filesystem. when i boot the image i get tons of "jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00080014: 0xf824 instead" erros, but the erros dont appear when booting up a 2.6 series kernel Nov 20 08:13:30 talking from 2.4.21 and 2.6.20.4 on a glomation ep9315 Nov 20 08:14:22 esben: train Nov 20 08:14:24 hehe Nov 20 08:14:56 btw, i set up the eraseblock size as mentioned ion my flash chips specification Nov 20 08:15:03 s/ion/in/ Nov 20 08:15:20 I'm afraid my first thought was that you must have stumpled upon a gcc bug :-) Nov 20 08:15:36 hrw|gone: thanks Nov 20 08:16:30 esben: i used gcc-cross-kernel-3.3.4 for linux 2.4.21 compilation Nov 20 08:18:20 nik0n: sorry, I was refering to ICE1092. Nov 20 08:49:52 03jeremy_laine 07org.oe.dev * r665b4cb1... 10/ (1 packages/python/python-pyopenssl_0.6.bb): python-pyopenssl_0.6.bb: add simple Python wrapper around OpenSSL Nov 20 08:54:22 morning! Nov 20 08:56:56 n8 all Nov 20 08:57:03 morning all Nov 20 08:57:58 RP: any thoughts on #3333 ? Nov 20 08:58:06 !oebug 3333 Nov 20 08:58:08 * * Bug 3333, Status: NEW, Created: 2007-11-18 23:41 Nov 20 08:58:08 * * hvontres(AT)gmail.com: linux-rp-2.6.23: sd drivers do not load before looking for root fs Nov 20 08:58:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3333 Nov 20 09:09:21 mtn: help required for 3-way merge <= what so I do with this nasty error? Nov 20 09:11:28 can I somehow revert my changes I made with monotone? Nov 20 09:16:04 josch: Tried to take a look in the manual.., found "monotone revert ." That's in the root of openembedded I guess, or maybe one above Nov 20 09:16:57 ah Nov 20 09:17:03 yepp it was mtn revert . Nov 20 09:17:09 :) Nov 20 09:17:27 morning Nov 20 09:19:59 good morning all Nov 20 09:24:07 is anybody using openvz or something similar to run OE inside a minimal environment? Nov 20 09:25:56 It might be possible to reduce the number of hard-to-reproduce build problems by tracking down and eliminating dependencies of host libs and tools (not specifically mentioned in ASSUME_PROVIDED) Nov 20 09:26:13 s/of/on/ Nov 20 09:48:42 hrw: Were SD drivers made modular? Nov 20 09:50:07 esben: People have tried it now and again. I just replied to your email Nov 20 09:51:05 * * OE Bug 3334 has been RESOLVED (WONTFIX) by hvontres(AT)gmail.com Nov 20 09:51:07 * *  -rootdelay locks up linux-rp-2.6.23 Nov 20 09:51:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3334 Nov 20 09:53:49 RP: not in OE Nov 20 09:59:07 RP: only for collie/2.6 Nov 20 10:00:05 ah, glibc got bumped Nov 20 10:01:10 * koen hopes that the "Need different builddirs" myth is now gone Nov 20 10:01:28 http://lists.linuxtogo.org/pipermail/openembedded-devel/2007-November/003494.html Nov 20 10:02:13 ooooh Nov 20 10:02:16 "phobia of using the real tools." Nov 20 10:02:19 that's nice Nov 20 10:02:23 * koen writes down Nov 20 10:09:50 i compiled a linux 2.4.21 system witrh uclibc 0.9.29 and nfsroot, when invoking ifconfig i get: Nov 20 10:09:54 root@192:/proc$ ifconfig Nov 20 10:09:56 any1 ever saw this? Nov 20 10:09:58 ifconfig: error: no inet socket available: Function not implemented Nov 20 10:10:11 kernel too stripped? Nov 20 10:10:50 no, my kernel with the veondor ramdisk wirks on ifconfig Nov 20 10:10:55 i guess its a uclibc issue Nov 20 10:19:15 koen: :) Nov 20 10:19:44 good morning Nov 20 10:24:05 * * OE Bug 3331 has been marked as DUPLICATE of bug 629 by Nov 20 10:24:06 * *  ping: permission denied. (are you root?) Nov 20 10:24:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3331 Nov 20 10:26:24 RP: does bitbake support 'aspect orientated' programming with tasks? Nov 20 10:26:36 e.g. "Prepend this bit to all tasks" Nov 20 10:29:55 is there an "OE official" standing on (L)GPL 2 vs. 3? Nov 20 10:30:03 RP: or a way to say "if then skip to do_stage" Nov 20 10:30:20 esben: yes: They are both FSF licenses Nov 20 10:30:23 the newest version of gmp has some improved powerpc support, but license have been changed to LGPL 3. Nov 20 10:30:44 esben: its upto distro so sort out what they distribute Nov 20 10:30:48 should we keep a recipe for the latest v2 version in case someone cares? Nov 20 10:31:20 koen: Not in the slightest Nov 20 10:32:06 but if a distro decides to stick with v2 licensed packages, then OE metadata might better have v2 licensed versions of as many packages as possible. Nov 20 10:32:12 esben: Yes, that would be a good idea Nov 20 10:32:25 esben: We need to make sure the LICENCE tags are accurate too Nov 20 10:33:02 * RP has already had discussions about this with poky. Its a mess :( Nov 20 10:33:16 should we use LGPL3 then? Nov 20 10:33:16 RP: I was thinking of tarring up ${D} into deploy/staging-pk/pn-pv-pr and untarring those for packaged-staging Nov 20 10:33:25 with LGPL being alias for LGPL2? Nov 20 10:34:10 esben: I think we need to better define LGPL and that is the safest meaning, yes Nov 20 10:34:28 "LGPL3+" might be a tag we need to start using Nov 20 10:34:44 or "LGPL2+" etc. Nov 20 10:34:44 LGPL3 or LGPL3+ ? Nov 20 10:34:52 03jeremy_laine 07org.oe.dev * r9f9b86bb... 10/ (1 packages/python/python-cheetah_2.0.1.bb): python-cheetah_2.0.1.bb: package version 2.0.1 of the cheetah template engine Nov 20 10:34:59 3+ means it has the "or later version" clause Nov 20 10:35:40 koen: Thats broadly along the lines I was thinking too Nov 20 10:36:05 RP: I know :) Nov 20 10:36:22 RP: but afaik we lack a way to say: "screw tasks, just untar" Nov 20 10:36:59 hehe, and mpfr is LGPL2.1+ then :-) Nov 20 10:38:06 koen: wouldn't it be possible to keep all packages staging files in seperate directories and then do a massive symlinking when populating staging, which should be much faster than untarring? Nov 20 10:38:43 I did something similar many years ago as sysadmin for a campus software collection. Nov 20 10:39:20 we installed each software package in it's own specific directory, even with multiple versions of each application and library. Nov 20 10:39:40 you're assuming filesystems support symlinks :) Nov 20 10:40:05 we then generated so-called software collections by using the fewest number of symlinks needed. Nov 20 10:40:45 koen: stop using your MSDOS Nov 20 10:41:44 koen: but which build environment do we support that doesn't have symlinks? Nov 20 10:41:46 cygwin supports symlinks Nov 20 10:41:53 (badly) Nov 20 10:42:25 I tried symlinks in for packaged-staging Nov 20 10:42:31 "didn't work as expected" Nov 20 10:42:50 koen: there are enough usage of "ln -sf" in many of our recipes not to mentioned makefiles of the software packages we use. Nov 20 10:43:07 my opinion on licenses: as long as they are accepted as opensource I accept them Nov 20 10:43:11 hmmm, what about using something like fuse? Nov 20 10:43:22 overlay the build dirs into deployed? Nov 20 10:43:26 koen: what more precisely were not as expected? Nov 20 10:43:43 Stephmw: that would rule out all non-Linux hosts! Nov 20 10:44:07 esben: :D Nov 20 10:44:09 esben: "too many level of symlinks", symlink loops, etc Nov 20 10:44:37 koen: I think working on symlinks for packaged staging would be a nice thing when you have it fully working with tar files Nov 20 10:44:59 koen: but making it work with the simpler tar approach first is a good idea :-) Nov 20 10:46:08 koen: I might be able to dig out the old code we used for generating symlinks based software collections in the old days. It is in Python :-) Nov 20 10:52:02 scary Nov 20 10:52:09 my bank has an MSN bot Nov 20 10:53:43 drat Nov 20 10:53:49 movial is now MoVial Nov 20 10:54:03 time to change all the stationary Nov 20 10:56:11 http://linuxdevices.com/news/NS6212564523.html looks nice to tinker with] Nov 20 10:56:34 * koen needs some more time to power up and check all the devboard gathering dust Nov 20 10:56:51 koen: you did not know about Hammer before? Nov 20 10:57:07 hrw: I did well before it went public :) Nov 20 10:57:08 koen: its prpplague toy Nov 20 10:57:22 I know prpplague personally :) Nov 20 10:58:28 who do we need to comfort? Nov 20 10:58:30 * Stephmw hides Nov 20 11:00:05 * * OE Bug 3085 has been REOPENED by eha(AT)doredevelopment.dk Nov 20 11:00:06 * *  gcc-cross-4.2.1 fails to build Nov 20 11:00:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3085 Nov 20 11:01:15 03ifaistos 07org.oe.dev * reef79e90... 10/ (4 files in 3 dirs): packages/liboil/liboil_0.3.11.bb : Correctly detect if the powerpc target has fpu Nov 20 11:02:33 koen : Any idea why ipaq-sleep gets in all machines using angstrom ? Does it actually work on anything else other than an ipaq ? Nov 20 11:06:59 steliosk: it sould be renamed to gpe-sleep iirc Nov 20 11:25:38 * XorA sees the news about libFLAC and goes ouch Nov 20 11:26:52 or thin-frontend-for-apm-or-sysfs-sleep-functionality Nov 20 11:27:34 I guess 'losless' doesn't apply to your CC# being stolen :) Nov 20 11:27:58 well you were obviously a dirty pirate anyway :-)_ Nov 20 11:28:31 yarrrrrrrr Nov 20 11:33:47 Which news? Nov 20 11:40:05 * * OE Bug 3251 has been RESOLVED (FIXED) by skoroneos(AT)digital-opsis.com Nov 20 11:40:07 * *  popt-1.7-r4-do_compile Nov 20 11:40:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3251 Nov 20 11:42:19 koen : got x11-image to build for dht :) Nov 20 11:43:06 koen: autobuilder does not provided spitz images... Nov 20 11:43:43 broonie: Vulnerabilities in libFLAC, see slashdot :) Nov 20 11:45:05 morning Nov 20 11:46:15 hi Rob Nov 20 11:47:11 s/b/d/ Nov 20 11:48:13 sorry Nov 20 11:48:27 np Nov 20 11:48:37 it's a common error :-) Nov 20 12:02:38 Argh! cygwin builds breaks in glibc because of case insensitive filesystem :-( Nov 20 12:03:20 * florian shooths monotone Nov 20 12:03:27 "mtn: misuse: path 'packages/ettercap/files' already exists, cannot create" Nov 20 12:12:04 * * OE Bug 2348 has been RESOLVED (FIXED) by heinold(AT)inf.fu-berlin.de Nov 20 12:12:06 * *  alsa-utils build fails against uclibc Nov 20 12:12:08 * * http://bugs.openembedded.org/show_bug.cgi?id=2348 Nov 20 12:14:08 esben: that's why zecke and I reformatted our macs Nov 20 12:17:17 crosstool has a fix for glibc-2.3.6... let's see if it works when upported to 2.6.1 Nov 20 12:22:31 uh, we even have a similar fix for glibc-2.2.5 Nov 20 12:22:34 glibc-2.2.5-cygwin.patch Nov 20 12:23:05 so I guess glibc-2.2.5 is the last glibc that MIGHT have worked on cygwin for OE :-) Nov 20 12:34:22 rwhitby, nice website. Have you built a rootfs in OE and used it on your N800 yet or do you mainly use maemo for the N800? I'm interesting in hearing how well the OE build works on the N800. Nov 20 12:34:51 svolpe_gerrath: I run vendor firmware on the n800 - I primarily use it for business voip. Nov 20 12:35:06 (gizmo and skype) Nov 20 12:35:45 svolpe_gerrath: you have n800? Nov 20 12:38:22 hrw, yes. Nov 20 12:40:45 hrw, I wanted to work on getting OE running on it, I've been using OE a lot and work and wanted to start doing it on something that I'm not tied up with deadlines so I could get more into it :-) Nov 20 12:47:53 re Nov 20 12:49:01 svolpe_gerrath: http://pokylinux.org/autobuild/poky/3193/ has poky image for n800 Nov 20 12:49:26 angstrom probably also has images for n800 Nov 20 12:49:27 anyone has a hint on that? http://pastebin.ca/790495 Nov 20 12:56:33 hrw, great that will be a nice staring point. How does it compare to the stock OS on the N800? Nov 20 12:57:40 its open but you loose Nov 20 12:57:59 Jin^eLD: I would rebuild automake-native Nov 20 13:00:42 03Cwiiis 07org.oe.dev * rec970839... 10/ (1 packages/openmoko2/libjana_svn.bb): libjana_svn.bb : build and package examples Nov 20 13:00:53 hrw, lol that was a very short but concise answer :-) Nov 20 13:01:49 svolpe_gerrath: you loose wifi (no rights to distribute), browser (can install other) and few other things Nov 20 13:02:32 hrw, I thought I remember reading that there was a wifi hack, that is a shame that the wifi is closed up. Nov 20 13:03:28 there is wifi hack indeed Nov 20 13:03:36 but we cant distribute wifi module Nov 20 13:03:51 we can create package which will hackk it on device itself anyway Nov 20 13:05:56 hey mickey|berlin Nov 20 13:06:08 hi Nov 20 13:06:56 mickey|berlin: seems you travel a lot these past days Nov 20 13:07:01 ff, siegen, berlin Nov 20 13:07:29 hrw: he got the same quilt problem again that the other guy had last week, it seems to be something they do or something they do different, no idea, both have debian Nov 20 13:07:36 but when I did a clean rebuild myself on their machines it worked Nov 20 13:07:38 doh... Nov 20 13:07:43 anyway, thanks for the automake hint Nov 20 13:08:08 koen: the final journey. when i'm back this time, I'll stay at home until FOSDEM Nov 20 13:08:43 koen: did you tried armv6 vfp + thumb? Nov 20 13:11:44 mickey|berlin: afaik you can only use vfp with thumb2 Nov 20 13:11:48 ehm Nov 20 13:11:53 s/mickey|berlin/hrw/ Nov 20 13:12:02 koen: thumb2 is armv7 iirc Nov 20 13:12:07 hrw: right Nov 20 13:12:38 see http://syslog.movial.fi/archives/46-N800-VFP-or-not-to-VFP.html for what happens with thumb1 and vfp Nov 20 13:13:31 tx Nov 20 13:14:07 that article also mentions the nonsense metric nokia uses for performance measures Nov 20 13:14:19 hrmpf! circular dependencies suck Nov 20 13:15:14 to get rid of the dependency on gfortran on host, I introduce a circular dependency on gmp and mpfr (which depends back on gcc-cross) Nov 20 13:34:14 hrw: I can't download today's tosa installkit. is it the expected behaviour? Nov 20 13:35:57 why the hell the menuconfig target tries to spawn a gnome-terminal ? Nov 20 13:36:13 nud: it spawn TERMCMD Nov 20 13:36:24 lumag: 403? Nov 20 13:36:28 hrw: and can you make it spawn nothing ? Nov 20 13:37:01 nud: tasks started by bitbake are not connected to your terminal... Nov 20 13:37:10 hrw: now it's ok Nov 20 13:37:25 lumag: server overload probably Nov 20 13:37:34 hrw: the issue is that I have no X server available Nov 20 13:37:35 ok Nov 20 13:38:22 nud: I usually enter dir in shell and call 'ARCH=arm make menuconfig' by hand Nov 20 13:39:02 that's what I do too, but I found the bitbake way quite cool Nov 20 13:39:24 which linux powered device (other then n8x0 from nokia) is armv6? Nov 20 13:39:34 device != devboard Nov 20 13:40:31 hi Cliff Nov 20 13:40:40 hrw: Hello Marcin! Nov 20 13:42:04 * * OE Bug 2928 has been RESOLVED (FIXED) by philip(AT)balister.org Nov 20 13:42:07 * *  binutils-cross 2.18 fails to build on x86_64 Nov 20 13:42:08 * * http://bugs.openembedded.org/show_bug.cgi?id=2928 Nov 20 13:43:27 hrw: zune2 :) Nov 20 13:44:02 ;D Nov 20 13:44:33 iirc there was some mx31 pmp somewhere Nov 20 13:44:37 can't remember the details Nov 20 13:44:54 * koen just sold his n800 to finance the n810 Nov 20 13:45:59 koen: for how much? Nov 20 13:46:14 €99 to my dad :) Nov 20 13:46:38 plus a spare gps I had to cover S&H for the n810 Nov 20 13:47:00 ;) Nov 20 13:47:32 lets see how much will it costs finally in europe Nov 20 13:50:24 1000usd .... Nov 20 13:51:46 what?!?! Nov 20 13:51:51 only 10 eurocent? Nov 20 13:52:04 heh Nov 20 13:52:25 * koen wonders what happens if the OPEC does dump the dollar Nov 20 13:52:38 never happen Nov 20 13:52:55 bush family has secret agreement with Saudi Royal family ... Nov 20 13:53:09 but not with the 11 other members Nov 20 13:53:26 if they do, saudi's flood the market with oil and drive price down Nov 20 13:53:47 * Crofton|home recites what he rad last night (well he is making up the secret agreement bit) Nov 20 13:54:33 only 426 days + 15 hours to go Nov 20 13:55:04 so we get cheap oil decoupled from the dollar Nov 20 13:55:46 the idea is the Saudi's drive peoples incmoe down so they stay on the dollar Nov 20 13:55:54 koen: China has over billion (10^12) usd reserve - let them start changing it to eur/chf and us notes will be more worth as toilet paper then value on them Nov 20 13:56:19 China's currency is tied to usd Nov 20 13:56:23 * XorA offers poor Crofton|home a box in the garden to live in Nov 20 13:56:28 :) Nov 20 13:57:04 * koen doesn't even have a garden Nov 20 13:57:04 although you might get kicked out if Scotland ever goes insane and declares independance Nov 20 13:57:10 hrw: tested current pimlico image. It works, except regular suspend/resume problems. Nov 20 13:57:14 koen: fit him in with the piggies Nov 20 13:57:51 lumag: ok Nov 20 14:02:53 03rwhitby 07org.oe.dev * r6d3696d4... 10/ (1 packages/net-tools/net-tools_1.60.bb): net-tools: Put quotes around flags so that expansion with spaces does not cause syntax errors. Nov 20 14:02:57 03rwhitby 07org.oe.dev * rada36533... 10/ (1 packages/meta/slugos-packages.bb): slugos-packages: Demote packages that no longer build Nov 20 14:03:05 03ifaistos 07org.oe.dev * r73862e82... 10/ (1 conf/distro/angstrom-2007.1.conf): conf/distro/angstrom-2007.1.conf : Set gcc to 4.1.1 for the xilinx ml403 board Nov 20 14:04:45 steliosk, why gcc 4.1.1? Nov 20 14:05:21 powerpc patches Nov 20 14:06:23 millions of patches, patches for free Nov 20 14:06:31 ah Nov 20 14:06:32 million of patches, patches for me Nov 20 14:06:56 * CM hands koen a 3-stringed guittar Nov 20 14:07:08 and 2 stringed bass? Nov 20 14:07:14 Of course :D Nov 20 14:07:40 patches come from OE, they were put there by a team, in a repository online Nov 20 14:09:59 bye! Nov 20 14:10:08 hmmm Nov 20 14:10:24 "Come on OE" to the tune of come on eileen would be neat as well Nov 20 14:11:12 You really should pick up that guitar ;) Nov 20 14:12:52 sadly I'm tonedeaf Nov 20 14:13:00 or whatever the PC term for that is Nov 20 14:14:12 "Musically challanged" maybe ;) Nov 20 14:18:58 morning Nov 20 14:26:26 hi Nov 20 14:26:59 hrw: I've just tested the new updater.sh on the akita Nov 20 14:27:08 works? Nov 20 14:27:20 besides a huge (more than one minute) wait before flashing the kernel Nov 20 14:27:29 strange Nov 20 14:27:30 it gives errors on boot Nov 20 14:27:55 jffs gives hundreds of errors of not finding the magic bitmap Nov 20 14:28:05 and then a /bin/sh error Nov 20 14:28:08 There's any guidelines about how to maintain multiple "recipes" for the oe building environment? Nov 20 14:28:39 I mean to maintain different machine support setups for some system, for example Nov 20 14:29:05 * * OE Bug 3339 has been created by eha(AT)doredevelopment.dk Nov 20 14:29:07 * * Upgrade to mpfr-2.3.0 Nov 20 14:29:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3339 Nov 20 14:29:13 I just reflashed with the older updater.sh, same zimage and initrd, and everything was fine Nov 20 14:29:21 ok Nov 20 14:29:30 thx for info Nov 20 14:29:45 when you have a new one to test, let me know Nov 20 14:29:56 it's the least I can do Nov 20 14:30:46 ok Nov 20 14:31:34 Bernardo: long wait was between "kernel" info and progressbar? Nov 20 14:36:00 hrw: Yes, I think so. Nov 20 14:36:05 strange Nov 20 14:36:20 I can prepare test ver for you Nov 20 14:36:30 * Bernardo wonders what broke the cursor orientation in latest opie image... Nov 20 14:36:59 sure, I have to reinstall my packages, so a reflash won't break much Nov 20 14:37:26 03crofton 07org.oe.dev * r5c78b207... 10/ (4 files in 3 dirs): wifistix : Sync with gumstix repo. Nov 20 14:38:26 hrw: what do you think about something like this: http://ewi546.ewi.utwente.nl/tmp/branding.diff.txt Nov 20 14:40:07 koen: such stuff is in plans Nov 20 14:41:15 hrw: I was also thinking about building a static (c)fdisk Nov 20 14:42:02 Bernardo: http://linuxtogo.org/~hrw/updater-bernardo.bin Nov 20 14:42:21 koen: there are many tweaks which we can add... Nov 20 14:45:22 * koen installs claws on his n800 Nov 20 14:46:22 koen: s/his/his father/ Nov 20 14:47:10 hrw: ok, will test it now Nov 20 14:48:43 Bernardo: this one got one tweak present only in akita updater Nov 20 14:50:42 I remember vaguely something when I looked at it two years ago Nov 20 14:51:13 Right now it cleared the screen, and is showing the "--- Zaurus CF updater ---"line Nov 20 14:51:27 Model Nov 20 14:51:29 kernel Nov 20 14:51:54 "special tweak for the akita" :) Nov 20 14:52:52 it is only when kernel flashing? Nov 20 14:53:19 and hangs after showing the "zImage.bin:1256136 bytes" line Nov 20 14:53:26 ok, flahsed now Nov 20 14:53:43 only shows when kernel flashing, yes Nov 20 14:53:56 it is now flashing the root fs Nov 20 14:54:19 took 2 minutes to flash the zImage Nov 20 14:54:26 only one for rootfs Nov 20 14:54:43 and now is at a prompt, do you want me to check anything ? Nov 20 14:55:08 reboot or exit Nov 20 14:55:24 ok, rebooting Nov 20 14:55:28 shell start was added to not reboot automatically so you see output Nov 20 14:55:46 jffs2 errors like crazy Nov 20 14:56:37 "jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at..." Nov 20 14:56:39 kernel boots normally and rootfs is wrong? right? Nov 20 14:57:20 looks like it, it then hangs just after INIT, with a /bin/sh error Nov 20 14:57:46 ok Nov 20 14:57:58 bbl Nov 20 14:58:21 so it looks like kernel tweak for akita is not needed Nov 20 14:58:54 but the kernel flashing is still very slow, with no progress indication Nov 20 15:00:17 Bernardo: with previous my version it gave progress? Nov 20 15:00:40 I'm not sure, but I don't think so Nov 20 15:00:44 have to go, work Nov 20 15:00:47 bbl Nov 20 15:01:09 bye Nov 20 15:01:15 * Bernardo is away: Ausente por agora. Nov 20 15:03:38 hrw: the "regular" akita updater also doesn't show progress when flashing kernel Nov 20 15:09:20 I know Nov 20 15:11:03 he, the fix worked. glibc-intermediate builds on cygwin now :-) Nov 20 15:11:29 and gcc-cross is in do_compile. a shame I have to leave for today now :-D Nov 20 15:16:34 ok, developing for a 2.4 series kernel is a mess Nov 20 15:18:22 indeed Nov 20 15:23:11 Crofton|home : gcc 4.1.2 does not builds for ppc/softfloat :( Nov 20 15:27:16 hrw, I just built a package in OE and installed it on my n800. However, it would not just install it since OE built packages register as armv6 and the maemo packages register as armel. I manually extracted the ipk file and installed it, everything worked fine as I expected. Is there an easy way around this? I realize I should just do a seperate rootfs but there are a couple of packages (udhcpd) that I need running right now from OE. Can Nov 20 15:27:16 I tell OE to build armel packages? Isn't armv6 and armel the same? Nov 20 15:35:21 dpkg --force-arch -i package.deb Nov 20 15:35:33 and it should not complain Nov 20 15:35:51 or give PACKAGE_ARCH="armel" in distro config Nov 20 16:03:02 bye Nov 20 16:10:15 hrw, thanks! Nov 20 17:04:52 is there any uml based image? Nov 20 17:07:24 any linux dostro which works out of the box inside qemu -arm ? Nov 20 17:30:47 no_mind: http://www.angstrom-distribution.org/unstable/autobuild/qemuarm/ Nov 20 17:36:03 koen is there any uml image? Nov 20 17:36:18 chouimat|work: no tthat I know off Nov 20 17:36:32 :( Nov 20 18:34:04 * * OE Bug 3340 has been created by dbaryshkov(AT)gmail.com Nov 20 18:34:06 * * fix for the fix of wlan-ng-modules 0.2.7 Nov 20 18:34:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3340 Nov 20 18:42:23 How I can force a rebuilding of a package on the tmp/work? Nov 20 18:42:38 bitbake -c rebuild foo Nov 20 18:42:43 Let's say, clean the tmp/work from the package Nov 20 18:42:46 Zero_Chaos: good, thanks Nov 20 18:43:00 -c clean if you wish to just clean it out Nov 20 18:43:06 :-) Nov 20 18:46:14 Does someone has problems building libxau? It looks to fail to build shared library Nov 20 18:49:01 !praise OE Nov 20 18:49:03 * cdbot2 bows down to OE and chants, "I'M NOT WORTHY!!" Nov 20 18:49:39 I think i found. My fault :-) Nov 20 18:49:44 ~emulate stewie Nov 20 18:49:45 It's like an orgy in my mouth! Nov 20 19:12:25 CoreDump: Do you still have the donated poodle? Nov 20 19:12:58 sort of Nov 20 19:13:02 * mwester hopes he's been feeding it well, and taking it on frequent walks. Nov 20 19:13:13 it doesn't make woof woof any longer Nov 20 19:13:23 That's a bad, bad sign. :( Nov 20 19:13:52 * mwester really show go do some work, but lacks motivation at the moment... Nov 20 19:13:52 CoreDump: what happend?? Nov 20 19:13:57 it won't charge and doesn't even come on w/ a pre-charged battery Nov 20 19:14:20 hvontres|home: no idea. I kept it in a drawer for like 3 month, wouldn't work afterwards Nov 20 19:14:26 CoreDump: hmm, that is a bad sign... Nov 20 19:14:37 it is at work waiting further inspection Nov 20 19:14:45 CoreDump: heh... Nov 20 19:15:36 I saw you'll get one from mickey_away, right? Nov 20 19:15:42 CoreDump: well looks like between mwester and uli, we finally have a shot at having kexec commandline support :) Nov 20 19:16:15 CoreDump: Actually, I think XorA|gone is going to send me his when he comes over in Febuary Nov 20 19:16:25 i see Nov 20 19:16:40 a fixed kexec would indeed be really useful Nov 20 19:17:39 CoreDump: I should fix up those patches and get them in the BTS. Right now, I am trying to get 2.6.23 to boot with SD support Nov 20 19:19:12 CoreDump: did you ever get a chance to play with my menu system? Nov 20 19:19:22 afraid not Nov 20 19:19:31 CoreDump: np Nov 20 19:19:38 I didn't play w/ anything zaurus related for some time Nov 20 19:20:26 CoreDump: well, it might even be usefull on OM...:) Nov 20 19:21:02 indeed Nov 20 19:21:26 but once the OM folks fix u-boot, there would not be much demand for altboot Nov 20 19:21:44 CoreDump: it only need three keys to work:) (up, down and pick) Nov 20 19:22:03 neo has like two Nov 20 19:22:15 (hehe) Nov 20 19:22:42 CoreDump: use the buttons for up down and mash the screen for select :) Nov 20 19:22:59 "mashing the screen" was one the two buttuns Nov 20 19:23:04 *buttons Nov 20 19:23:22 the PWR button couldn't be abused for anything useful last time I checked Nov 20 19:24:56 * hvontres|home wishes he could justify a new n810.... Nov 20 19:25:59 well, by the sounds of it, the n810 has inherited all the problems of the n770 Nov 20 19:26:27 being semi-closed, crappy closed-source flashing utility etc etc Nov 20 19:26:43 hrw|gone_: and koen would know Nov 20 19:26:45 CoreDump: heh....maybe I can pick up a cheap Cxxx Nov 20 19:27:34 CoreDump: But at least koen and hrw|gone_ have them.... Nov 20 19:27:40 maybe one of these sub-sub notebooks when their prices go down... Nov 20 19:28:39 CoreDump: well, I picked up a 14" hp dv2416us for $600 not to long ago. so $500 for a handheld seems outragous to men Nov 20 19:28:44 err me Nov 20 19:28:50 right Nov 20 19:29:35 CoreDump: but then again, a $500 iPhone also seems like ripoff :) Nov 20 19:29:40 it is Nov 20 19:31:36 * hvontres|home spent $20 for his cell, a motorola C139 with prepaid airtime... Nov 20 19:32:09 hvontres|home: 500$ is nothing to look cool :) Nov 20 19:37:44 chouimat|work: hehe Nov 20 19:38:35 chouimat|work: well, I have the "I have a house and a Family" budget restriction...plus I'd rather buy a $400 planer for that kind of money :) Nov 20 19:38:42 CoreDump: https://garage.maemo.org/projects/oxffff/ Nov 20 19:38:49 * mwester just picked up an N800 for $200 -- very nice little device. Nov 20 19:39:30 mwester: not bad, but I have kinda gotten used to the built in keyboard:) Nov 20 19:39:52 koen: that is something, at least ;) Nov 20 19:39:56 Yeah, I'd like the N810 but I too have the "I have a house and a family" budget concern. Nov 20 19:40:04 are there still binary-only drivers? Nov 20 19:40:15 CoreDump: OE can relink the wifi module Nov 20 19:40:20 I think the wifi, for one thing is closed. Nov 20 19:40:48 * mwester wonders if OE can build packages for the N800... Nov 20 19:41:18 mwester: in an unslung kind of way Nov 20 19:41:20 It's just an ARM, right? No reason with the right libs that OE can't build me a feed. Nov 20 19:41:32 I'm familiar with Unslung ;) I can live with that. Nov 20 19:41:41 mwester: nokia uses the codesourcery toolchain Nov 20 19:41:58 with a sort-of-EABI-but-not-quite glibc Nov 20 19:42:28 :\ Nov 20 19:43:09 time is better spent adding more maemo4 stuff to oe Nov 20 19:44:07 one hour 37 minutes left Nov 20 19:51:12 * Bernardo is back. Nov 20 20:03:04 * * OE Bug 3341 has been created by autobuild(AT)openembedded.org Nov 20 20:03:06 * * webkit-gtk-0.1+svnr27222-r3-do_configure Nov 20 20:03:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3341 Nov 20 20:33:42 Where I should send a patch? Nov 20 20:34:25 Should I report a bug with it only? Nov 20 20:34:27 otavio: either attach it to an existing bug in the bugtracker or open a new one and attach it there Nov 20 20:34:36 Jin^eLD: right Nov 20 20:35:51 re Nov 20 20:36:14 hey flo_lap Nov 20 20:41:53 hi florian Nov 20 20:45:29 hi pvanhoof Nov 20 20:45:53 hey flo_lap Nov 20 20:57:18 !lart tosa_pm.c for being total mess Nov 20 20:57:18 * cdbot2 pours hot grits down the front of tosa_pm.c for being total mess's pants Nov 20 21:07:29 But at least the kernel is trimmed to 480216 :) Nov 20 21:10:01 koen: no mainstone images? i'm shocked ;) Nov 20 21:10:11 * flo_lap just unpacked his mainstone Nov 20 21:10:47 send a mail to angstrom-devel requesting mainstone images :) Nov 20 21:35:30 koen, I'm going to try an autobuilder image an an OSK with a display Nov 20 21:35:39 which is most likely to work? Nov 20 22:02:26 03dbaryshkov 07org.oe.dev * r479b064d... 10/ (3 files in 3 dirs): wlan-ng-modules: fix for the fix of wlan-ng-modules 0.2.7 - close #3340 Nov 20 22:12:38 bye all Nov 20 22:14:21 * cbrake prefers to leave his mainstone packed up :-) Nov 20 22:33:34 re Nov 20 22:36:16 Let me ask something. I've found a build error on binutils-2.17.50.5 and fixed it. Since 2.18 is already included on packages, bugs on previous version are interesting to get solved? Nov 20 23:04:05 * * OE Bug 3340 has been RESOLVED (FIXED) by Nov 20 23:04:07 * *  fix for the fix of wlan-ng-modules 0.2.7 Nov 20 23:04:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3340 Nov 20 23:41:20 otavio: yes Nov 20 23:41:29 otavio: someone can still use them Nov 20 23:41:38 * hrw|gone -> bed Nov 20 23:42:01 hrw|gone: http://bugs.openembedded.org/show_bug.cgi?id=3342 Nov 20 23:42:04 hrw|gone: reported Nov 20 23:49:05 * * OE Bug 3342 has been created by otavio(AT)ossystems.com.br Nov 20 23:49:07 * * binutils 2.17.50.0. 5 fails to build due makeinfo version problem Nov 20 23:49:09 * * http://bugs.openembedded.org/show_bug.cgi?id=3342 Nov 21 00:32:04 * * OE Bug 3343 has been created by antti.andreimann(AT)mail.ee Nov 21 00:32:06 * * ibungif download location Nov 21 00:32:08 * * http://bugs.openembedded.org/show_bug.cgi?id=3343 Nov 21 01:01:39 03mwester 07org.oe.dev * r0eebea61... 10/ (4 files in 3 dirs): Nov 21 01:01:39 initscripts: SlugOS - add bootclean.sh to clean out /tmp, /var/lock, /var/run Nov 21 01:01:39 (required when turned-up to disk). Create SlugOS-specific volatiles config Nov 21 01:01:39 because SlugOS needs /etc/resolv.conf to be static and not volatile. Nov 21 01:01:46 03mwester 07org.oe.dev * rb2a2a4f2... 10/ (2 files in 2 dirs): Nov 21 01:01:46 wireless-tools.if-pre-up - move setting of the ESSID to the end of the script, Nov 21 01:01:46 because certain chipsets (ZD1211) use that as the "commit" operation. Nov 21 01:01:51 03mwester 07org.oe.dev * r51aad88c... 10/ (1 packages/wireless-tools/wireless-tools_28+29pre21.bb): Bump PR Nov 21 01:02:43 sakoman, ping Nov 21 01:12:58 * otavio is having problem with libtool putting the path to .la files, on dependencies, in /usr/lib instead of pointing it to stagging area Nov 21 01:13:06 Does some has any hint about it? Nov 21 01:34:02 configure.in:89: error: possibly undefined macro: AM_PATH_LIBGCRYPT <- any idea what that could be? Nov 21 01:41:05 nm **** ENDING LOGGING AT Wed Nov 21 02:59:56 2007