**** BEGIN LOGGING AT Sat Jan 06 02:59:59 2007 Jan 06 04:12:57 koen|away: I solved the diet-x11 compile failures by installing the x-dev package on Debian. Seems that compiling diet-x11 in OE requires X libraries on the host (i.e. it doesn't use the X headers built by OE in staging). Jan 06 05:00:30 is STAGING_BINDIR_CROSS meant to be created by default, or does each package have to make sure it doess an install -d on it before trying to copy something into it? Jan 06 05:12:38 * rwhitby fixes tcl to do so, which fixes the tk build failure. Jan 06 08:14:26 03hrw 07org.oe.dev * r44f1e05f... 10/ (3 files in 2 dirs): gnutls: update to 1.6.1 Jan 06 08:31:02 03hrw * r734 10bitbake/lib/bb/fetch/cvs.py: CVS Fetcher: Added support for non-default port (needed or rxvt-unicode in OE). Jan 06 08:31:58 hrw|gone, hi :) Jan 06 08:36:46 hi Jan 06 08:36:48 hi Marex Jan 06 08:37:35 03hrw 07org.oe.dev * r271d07ae... 10/ (10 files in 5 dirs): progear: added modules for AC detection and Battery status Jan 06 08:37:37 hi hrw, you're working on that marvell 8385 driver, right? :) Jan 06 08:37:40 03hrw 07org.oe.dev * r1a562cd9... 10/ (1 conf/machine/progear.conf): progear: ship AC & Battery modules Jan 06 08:37:46 03hrw 07org.oe.dev * r42afce36... 10/ (1 packages/rxvt-unicode/rxvt-unicode_cvs.bb): rxvt-unicode: fix SRC_URI - need BitBake trunk r734 or later Jan 06 08:37:53 Marex: I started it Jan 06 08:38:01 hrw, well ... I have 8305 in lifedrive and I'd like to test it Jan 06 08:38:20 could you possibly pack the whole source tree and mail it to me, please? Jan 06 08:38:21 8305 != 8385 rather Jan 06 08:38:27 mmnt Jan 06 08:38:39 88W8305 - TKIP (firmware) AES (hardware) WPA (firmware) POWER Save Jan 06 08:38:39 88W8381 - 8305 + TKIP (hardware) 802.11e (QoS) 802.11i (Security) BT coex Jan 06 08:38:39 88W8385 - 8381 + a/g Jan 06 08:38:39 88W8388 - 8385 + TCP/IP termination NAND Flash I/F Audio Codec I/F Jan 06 08:38:46 ah Jan 06 08:38:49 :) Jan 06 08:38:56 please please :) Jan 06 08:39:07 we have a project on linuxtogo.org with SVN repo Jan 06 08:39:09 hrw, I havent managed to patch that driver myself :( Jan 06 08:39:13 yeah ... I know Jan 06 08:39:16 feel free to look and join Jan 06 08:39:42 I'm not building it in armstrom/oe/whatever though Jan 06 08:39:51 and I dunno what that quilt does Jan 06 08:39:55 im makefile Jan 06 08:40:08 I'd just like compiling it and testing it Jan 06 08:40:40 currently driver is not ready for 2.6 Jan 06 08:40:43 http://web.mit.edu/jbarnold/8381/ WOW 8-o Jan 06 08:40:50 hrw, ah :( Jan 06 08:42:19 hrw, so there's no way testing it on 2.6 ? Jan 06 08:42:36 hrw, as far as I understood what you wrote on your homepage, you can use that with 2.6 Jan 06 08:42:41 You don't have permission to access /jbarnold/8381/CF-8381-LINUX-X86-5.11.7.p2-210.P4-1.tar on this server. ;( Jan 06 08:43:06 I have to get it under 2.6 Jan 06 08:44:51 03hrw 07bitbake-1.6 * r735 10/lib/bb/fetch/cvs.py: CVS Fetcher: Added support for non-default port (needed for rxvt-unicode in OE). Jan 06 08:47:10 hrw, ok, I'll look forward Jan 06 08:47:15 and I'll try asking that guy Jan 06 08:50:52 cu Jan 06 08:51:37 03rwhitby 07org.oe.dev * r87414339... 10/ (1 packages/tcltk/tcl_8.4.11.bb): tcl: make sure ${STAGING_BINDIR_CROSS} exists before trying to install tclConfig.sh into it Jan 06 08:51:46 03rwhitby 07org.oe.dev * r13aacca4... 10/ (1 packages/meta/slugos-native.bb): slugos-native: Changed libc6-dev to glibc Jan 06 08:51:51 03rwhitby 07org.oe.dev * rac0ea819... 10/ (1 packages/meta/slugos-packages.bb): slugos-packages: demoted asterisk Jan 06 08:55:49 hrw|gone, ok ... I dont have to go ;) Jan 06 08:56:05 ho, bye hrw|gone Jan 06 08:56:08 *ok Jan 06 08:56:10 and thanx Jan 06 09:11:36 03pfalcon 07org.oe.dev * r6b8185c8... 10/ (1 packages/meta/package-index.bb): Jan 06 09:11:36 package-index: Fix DEPENDS. Jan 06 09:11:36 * Actually depends on ipkg-utils-native, not ipkg-native. Jan 06 09:20:38 03rwhitby 07org.oe.dev * r66b41b89... 10/ (1 packages/meta/slugos-packages.bb): slugos-packages: demoted cron Jan 06 09:21:15 hrw|gone, j/w how did you managed to use that card with 2.6 then? Jan 06 09:39:42 Good morning Jan 06 09:40:00 hey sirfred Jan 06 09:57:42 Hi! Jan 06 09:57:58 koen, Patch for http://bugs.openembedded.org/show_bug.cgi?id=1569 attached, please have a look Jan 06 10:04:31 Anyone have an idea if busybox supports nohup? Jan 06 10:06:50 I see. I hacked defconfig for 1.2.2 which has def_pref=-1 Jan 06 10:28:43 * koen wonders how buggy busybox 1.3.x is Jan 06 10:28:51 psokolovsky_: does that patch work? Jan 06 10:29:03 koen, yep ;-) Jan 06 10:38:35 koen: why in machine reuirements in h3800.conf there is a kernel26 not 24? Jan 06 10:59:30 koen, so how about adding nohup to busybox? it's pretty useful thing to have it by default. Jan 06 11:09:28 psokolovsky_: what's the size increase? Jan 06 11:10:23 koen, I'm building it now (this time, should finally be in, I hope ;-)), will check and tell Jan 06 11:12:54 koen, though looking at the source, should be 200 bytes at most ;-) Jan 06 11:13:27 that sounds small enough Jan 06 11:24:06 koen, ok, exe size increase is zero, everything went into alignment space ;-) Jan 06 11:39:40 koen, so, ok to commit? Jan 06 11:41:50 philippe: and .. did you build h3800 image? Jan 06 11:49:20 crap. I can't do "bitbake linux-handhelds-2.6-2.6.18-hh0" Jan 06 11:52:25 psokolovsky_: sounds ok to me Jan 06 11:52:38 ok Jan 06 12:02:07 03rwhitby 07org.oe.dev * rb87c66f1... 10/ (1 packages/meta/slugos-packages.bb): slugos-packages: demoted ftpd-topfield Jan 06 12:04:41 03rwhitby 07org.oe.dev * r7e5569d0... 10/ (1 packages/meta/slugos-packages.bb): slugos-packages: demoted libgphoto2 Jan 06 12:08:43 hi Jan 06 12:08:56 koen: did not yet tried busybox 1.3.x? Jan 06 12:09:45 hrw, Hi! do you have and idea how can I build specific version of pkg like linux-handhelds-2.6 ? ;-) Jan 06 12:10:38 linux-handhelds-2.6-PV Jan 06 12:10:57 hrw, doesn't work ;-) Jan 06 12:11:21 then blame hh.org kernel maintainer Jan 06 12:12:02 hrw, or bitbake one? ;-) just thought you might know some trick... Jan 06 12:12:41 psokolovsky_: PN-PV is trick Jan 06 12:13:46 hrw, how stupid thing knows on which dash is to split? apparently, currently it doesn't try its best. anyway, should be separate option to pass in a version. Jan 06 12:14:06 psokolovsky_: patches welcome Jan 06 12:15:45 03rwhitby 07org.oe.dev * rcead2098... 10/ (1 packages/meta/slugos-packages.bb): slugos-packages: demoted gphoto2 Jan 06 12:26:30 03pfalcon 07org.oe.dev * r7d43d11a... 10/ (6 files in 6 dirs): (log message trimmed) Jan 06 12:26:30 busybox 1.2.1, 1.2.2: Enabled nohup applet. Jan 06 12:26:30 * Being able to login via telnet/ssh/on-device shell and run application such Jan 06 12:26:30 that it continues to run after logout, should be pretty useful feature for Jan 06 12:26:30 embedded system. Jan 06 12:26:30 * With common config (also used by angstrom), doesn't lead to increase of Jan 06 12:26:32 exe size (due to alignment padding), and the applet code itself is small Jan 06 12:39:58 hey hrw Jan 06 12:40:45 hrw: I didn't test busybox 1.3.0 yet Jan 06 12:40:58 03rwhitby 07org.oe.dev * re0d0eae2... 10/ (1 packages/meta/slugos-packages.bb): slugos-packages: demoted netpbm Jan 06 12:41:04 hrw: I use task-proper-tools a lot :) Jan 06 12:44:22 ;) Jan 06 12:44:36 I'm updating .oz354x now to some newer stuff Jan 06 12:44:45 and wonder about going to busybox 1.2/1.3 Jan 06 12:44:58 e2fs tools in busybox.. less etc Jan 06 12:45:09 bbiab Jan 06 12:48:31 ~pastebin Jan 06 12:48:35 it has been said that pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.ca, or http://channels.debian.net/paste Jan 06 12:48:57 hrw|afk: preparing for a new oz release? Jan 06 12:50:30 hrw|afk: since most oz users think 'I can only update with a reflash' you might as well release 3.5.4.4 Jan 06 12:54:41 gremlin[it]: hi! :) Jan 06 12:55:37 hi univac ! Jan 06 13:04:36 03rwhitby 07org.oe.dev * r8ce2ae26... 10/ (1 packages/meta/slugos-packages.bb): slugos-packages: demoted puppy Jan 06 13:05:55 koen: OZ 3.5.5 Jan 06 13:07:31 another stupid question: and how do I tell ipkg to install specific pkg version? Jan 06 13:07:48 sirfred: congrats on that xserver, Im just trying to get a working image to test Jan 06 13:18:13 Good morning! Jan 06 13:18:41 A package I am trying to build fails in configure with "configure: error: "POSIX termios operations are required"" What package provides this termios? Jan 06 13:19:22 Or is this some incorrect include? Jan 06 13:28:13 ~lart nslu-someone who broke madwifi Jan 06 13:28:13 * ibot shoots nslu-someone who broke madwifi in his sleep Jan 06 13:28:21 patching file hal/public/xscale-be-elf.inc Jan 06 13:28:21 Hunk #1 FAILED at 77. Jan 06 13:28:26 NOTE: package madwifi-ng-r1871-20061229-r0: task do_patch: started Jan 06 13:28:36 mtn: multiple update candidates: Jan 06 13:28:36 mtn: 6390f37ba1af225733a09408253deb2c380f9603 xora@openembedded.org 2005-06-14T23:52:17 Jan 06 13:28:36 mtn: 8ce2ae26934b58d193c011f402be69699132abb0 rwhitby@nslu2-linux.org 2007-01-06T13:16:02 Jan 06 13:28:45 2005-06-14T23:52:17???? Jan 06 13:28:47 hrw|afk: see oe mailinglist about madwifi Jan 06 13:29:17 heh - my date wins Jan 06 13:29:55 mwester: see note above re madwifi Jan 06 13:30:11 koen, so once more confirmed that #1569 works ok w/kernel upgrade/downgrade. ok to commit? Jan 06 13:30:11 rwhitby: are all nslu people with commit access subscribed to the oe mailinglist? Jan 06 13:30:17 psokolovsky_: yes Jan 06 13:30:27 rwhitby: if not, could you urge them do to so? Jan 06 13:31:04 koen: will do. Jan 06 13:31:20 (I believe they are, but will confirm with them) Jan 06 13:37:35 03koen 07org.oe.dev * r59377891... 10/ (1 packages/curl/curl_7.16.0.bb): curl 7.16: package ca-certs (88kB), needed for ssl (e.g https) and RRECOMMEND it for libcurl Jan 06 13:38:30 koen: hm.. another thing to pluck.. Jan 06 13:39:27 hrw|afk: copying curl 7.16.0.bb would be faster Jan 06 13:39:48 koen: I already plucked addon of it Jan 06 13:40:17 hrw: https works with gpe-mini-browser, but only for sites that don't popup a dialog in other browsers Jan 06 13:40:33 hrw: since webcore (or gpe-m-b) lacks a ssl manager Jan 06 13:41:05 ok Jan 06 13:41:47 first I need to finish kernel for my progear and try vesa-tng on it Jan 06 13:41:58 standard vesa fb works from time to time Jan 06 13:42:06 03rwhitby 07org.oe.dev * r198a1c85... 10/ (1 packages/meta/slugos-packages.bb): slugos-packages: demoted nfs-utils Jan 06 13:42:12 03pfalcon 07org.oe.dev * rc048c58d... 10/ (1 classes/kernel.bbclass): Jan 06 13:42:12 kernel.bbclass: Create relative symlinks from "zImage" to actual kernel image in use. Jan 06 13:42:12 * Make sure update-alternatives created relative symlinks in /boot (or whatever) Jan 06 13:42:12 kernel image directory is used. This is required for tools like LAB (Linux As Jan 06 13:42:12 Bootloader) which may mount rootfs over some other temporary fs tree to get access Jan 06 13:42:14 to kernel to boot. Jan 06 13:42:16 * Closes #1569. Jan 06 13:42:36 ok. madwifi-ng built. but I will not push changeset (ugly) Jan 06 13:46:48 hrw: can you pastebin your fix ? Jan 06 13:49:10 rwhitby: remove SRC_URI, use SRC_URI_slugos as SRC_URI, drop last entry (I build for x86) Jan 06 13:49:51 * koen mumbles something about madwifi-ng on the mailinglist Jan 06 13:52:06 morning all Jan 06 13:52:17 03koen 07org.oe.dev * r4a88d44f... 10/ (1 conf/machine/efika.conf): efika: remove pci from machine features to avoid troubles with building task-base Jan 06 13:53:40 likewise: http://gitweb.freedesktop.org/?p=cairo;a=commit;h=c96a71e709e537f690da6d4a184aa4c64fe11028 Jan 06 13:54:43 koen: thanks, but I think most of the graphics I do are not lines that need to be clipped. Jan 06 13:55:28 likewise: maybe cairo-clocks uses it :) Jan 06 13:55:37 koen: did you try out efika sound yet? I was about to test it when I run into nfs-server problems of my fresh ubuntu feisty installation. Jan 06 13:56:03 koen: no, as far as I could see, it's vectors lie within the drawn rectangle, but who knows. Jan 06 13:57:44 likewise: haven't tried yet, need to fix dhcp first Jan 06 13:58:05 koen: ok, I will spend some time on that today Jan 06 14:03:39 likewise: Can I bother you a bit with icecc? Are you still using it? Jan 06 14:05:00 Laibsch: yes, but not at home (where I am now). I still have to set it up at home. Jan 06 14:06:01 Laibsch: but yes, you may bother me at home :-) Jan 06 14:06:18 Cool Jan 06 14:06:48 I am trying to get this working. And at least I do not get any obvious errors or crashes anymore thanks to your and Ifaistos' work Jan 06 14:07:03 I have two machines right now, both running ubuntu edgy. Jan 06 14:07:15 And they seem to talk to each other. Jan 06 14:07:21 * likewise should have sticked with edgy as well Jan 06 14:07:22 But no jobs ever get distributed :-/ Jan 06 14:07:34 likewise: What are you on now? feisty? Jan 06 14:07:36 Laibsch: you enabled the extensive logging of your scheduler Jan 06 14:07:42 yes Jan 06 14:07:46 Laibsch: yes feisty Jan 06 14:07:49 I can show you in a minute Jan 06 14:08:05 Laibsch: yup, show me some scheduler log Jan 06 14:08:16 As I said and as you will see, the client contacts the scheduler. Jan 06 14:08:35 I tried distributing jobs both with OE and from a general make. Jan 06 14:08:48 * Laibsch prepares pasting log Jan 06 14:10:00 Laibsch: pastebin.ca Jan 06 14:10:31 http://oz.leggewie.org/wip/icecc_scheduler.log Jan 06 14:11:06 .10 is the scheduler. .12 the client Jan 06 14:11:06 * likewise remembers that Ifaistos said the scheduler from icecc CVS had unbalanced scheduling cost calculation Jan 06 14:11:54 Laibsch: I only see a login from one machine (Rie) Jan 06 14:12:29 Do we have virtual/glib ? Jan 06 14:13:05 Laibsch: to start with, I would expect >1 machines to login to the scheduler Jan 06 14:13:05 psokolovsky_: what for? Jan 06 14:13:39 hrw, to select either 1.2 or 2.0 per distro/package like Jan 06 14:13:52 psokolovsky_: we have glib-2.0 and glib-1.2 iirc Jan 06 14:14:05 and glib-1.2 apps should die Jan 06 14:14:22 morning Jan 06 14:14:26 hi chouimat Jan 06 14:14:30 morning chouimat Jan 06 14:15:02 http://news.bbc.co.uk/2/hi/americas/6236585.stm Jan 06 14:16:32 hrw, there're few perversions, as you know. mc depends on glib, and it used to depend on small 1.2 only. I'm trying to build small initrd, and so far stuff doesn't fit in 16Mb ;-/ Jan 06 14:17:17 psokolovsky_: which stuf in your rootfs depend on glib? Jan 06 14:17:22 NOTE: package angstrom-bootstrap-image-1.0-r3: task do_build: completed Jan 06 14:17:30 new stuff for my progear Jan 06 14:20:49 hrw, I told - mc (midnight commander) Jan 06 14:21:08 mc for small rootfs?? Jan 06 14:23:12 hrw, yes. mc is normal file manager which includes viewer and editor. pretty neat thing. Jan 06 14:24:15 * Laibsch curses DSL Jan 06 14:24:41 psokolovsky_: I know what mc is. I use it since I use unix systems Jan 06 14:25:03 psokolovsky_: I cant just understand why mc in small rootfs Jan 06 14:25:13 hrw, cool, so what do you suggest regarding glib monster? Jan 06 14:25:21 drop mc Jan 06 14:25:25 lol Jan 06 14:25:41 mc is requirement Jan 06 14:25:52 then live with glib Jan 06 14:27:00 likewise: Any idea about the problem? Any idea what to try out? Jan 06 14:27:02 hrw, I want to be able to use glib 1.2 for such cases, but big one for full-fledged system. this calls for virtual/glib Jan 06 14:27:16 psokolovsky_: no - this call for mc-glib12.bb Jan 06 14:27:43 whatever, though I'd think virtual is more clean Jan 06 14:27:46 Laibsch: I would first check the log of the two icecc daemons to see whether they successfully logged in to the scheduler. Jan 06 14:27:59 psokolovsky_: how many apps will use it? mc only? Jan 06 14:28:59 likewise: two ?? icecc daemons? AFAIK, the client .12 does not produce a log. How do I activate that? Jan 06 14:29:13 I have one scheduler at .10 and one client at .12 Jan 06 14:29:40 Laibsch: ah ok. but which machine starts the build? Jan 06 14:29:51 the scheduler at .10 Jan 06 14:30:36 hrw, I want to create devel/test-support image, ~4mb, and put all stuff needed to investigate and test system there Jan 06 14:31:20 psokolovsky_: I understand what you want. just do not want to populate repo with another virtual/ which will be used by 1-2 packages Jan 06 14:31:28 Laibsch: re-reading your log with this info... Jan 06 14:31:38 hrw: another q: can I depend on order of PACKAGES/FILES somehow to tell it: package files1 to X, files to Y, and the rest to Z? Jan 06 14:32:32 psokolovsky_: currently you can. Jan 06 14:32:44 hrw, sure, that's why I ask for suggestions how to handle that. I understand that creating a virtual for one package probably not good (though glib is pretty big lib to let that ;-) ) Jan 06 14:32:56 Laibsch: can't fetch your log anymore, it seems Jan 06 14:33:06 hrw, but not recommended? Jan 06 14:33:15 psokolovsky_: virtual/foo implies you can exchange them at will, which you can't Jan 06 14:33:15 Laibsch: vm, got it Jan 06 14:33:18 likewise: I had a DSL reconnect Jan 06 14:33:28 My IP changed. Jan 06 14:33:33 Laibsch: yup got it now. Jan 06 14:33:36 OK Jan 06 14:33:48 koen, what you mean? distro config defines them (bitbake.conf can provide default) Jan 06 14:34:28 I can't build gtk 2.x against glib 1.2 Jan 06 14:34:40 psokolovsky_: very small amount of apps can switch glib 2.0/1.2 without tweaking Jan 06 14:34:52 --with-glib-1.2 contra --with-glib-2.0 Jan 06 14:34:53 etc Jan 06 14:35:00 practically none Jan 06 14:35:17 psokolovsky_: create mc-glib12.bb Jan 06 14:35:25 will be simplest Jan 06 14:35:33 NOTE: package wpa-supplicant-0.5.7: completed Jan 06 14:35:33 koen, so, package which can be build only against 2.0, will define DEPENDS=glib-2.0, and ones which can be build against either - DEPENDS=virt/glib Jan 06 14:35:47 hrw, ok, thanks Jan 06 14:35:47 psokolovsky_: no Jan 06 14:35:50 Laibsch: I never see .12 login to the scheduler successfully, although the sched does receive broadcasts from it. Jan 06 14:36:01 psokolovsky_: but how many of them can switch glib without extra switches? Jan 06 14:36:09 dunno ;-) Jan 06 14:36:30 psokolovsky_: then check first Jan 06 14:36:37 psokolovsky_: virtual/foo may only be used if it's needed (it's not) and if you can exchage them at will (you can't) Jan 06 14:37:13 koen, ok, that was just question for you guys ;-) Jan 06 14:37:42 likewise: So what do I do about it? Jan 06 14:38:18 Laibsch: check versions of sched and daemon. Should be from the same icecc release. See if you can enable as much debugging/logging as possible. Jan 06 14:38:44 Laibsch: I really need to get into this again, I cannot help much in detail right now... Jan 06 14:40:07 likewise: sched and client have the same exact package of icecc installed. The client is not creating a log for some reason. I will try to increase verbosity there. Thanks for your help. Jan 06 14:40:57 Laibsch: Hmm, I remember I had to log-less daemon too. I am not sure if I could fix it with that version. icecc is not really a stable package to work with. Jan 06 14:41:44 Is there an IRC chat or an ml where these people hang out? I had the impression icecc was kind of stale. Jan 06 14:43:23 Laibsch: I think it was spawned by the OpenSUSE KDE team for their build setup. Jan 06 14:43:37 yes Jan 06 14:44:03 Maybe I can get lucky in #opensuse Jan 06 14:44:21 Laibsch: already asking Jan 06 14:46:19 Laibsch: I'll ask at the coolo at kde.org email address mentioned on the icecream page. Jan 06 14:47:07 likewise: OK, thanks. Jan 06 14:47:33 I just joined #opensuse-project and #opensuse-buildservice Jan 06 14:47:44 Sounds like possible candidates. Jan 06 14:48:59 koen, is there any trick to include just files in a dir, but not any subdirs, FILES = "foo/*" doesn't do the trick, it picks up "foo/bar/" too. Jan 06 14:49:17 Hi gerwinin Jan 06 14:49:31 psokolovsky_: it's a python match, so check the python docs Jan 06 14:49:38 ok Jan 06 14:49:46 I suspect it has to do with indenting the matches Jan 06 14:50:07 rwhitby: ERROR: /home/koen/OE/monotone/org.openembedded.dev/packages/meta/slugos-packages.bb:207: unparsed line: ' nfs-utils UCLIBC_UNSUPPORTABLE_PACKAGES = " libpam rng-tools postfix yp-tools ypbind ypserv "' while parsing /home/koen/OE/monotone/org.openembedded.dev/packages/meta/slugos-packages.bb Jan 06 14:51:02 where is the correct place to make suggestions or place wishes for bitbake? Jan 06 14:51:22 mr_nice: bugtracker or bitbake-dev mailinglist Jan 06 14:51:36 koen: thx Jan 06 14:52:58 hrw|afk: Packaged contents of foo into /data/build/koen/OE/build/tmp/angstrom/deploy/glibc/ipk/foo_test-20070106-r0_arm-oabi.ipk Jan 06 14:55:42 likewise: coolo @ irc.kde.org seems to be our guy. Jan 06 14:56:03 check: http://people.kde.org/stephan.html Jan 06 14:56:18 03koen 07org.oe.dev * r991984e4... 10/ (1 conf/distro/angstrom-2007.1-oabi.conf): angstrom 2007.1-oabi: tag packages as arm-oabi to minimize confusion Jan 06 15:01:18 What's the common way to setup a password for root, for dropbear/ssh access to a board? Jan 06 15:03:01 likewise: He does not seem to be online. Have you mailed him? I don't want to bother him if you did. Jan 06 15:03:01 03hrw 07org.oe.dev * rfab85738... 10/ (7 files in 2 dirs): wpa-supplicant: added 0.5.7, unified 0.5.x, added script to autostart wpa-supplicant on ifup Jan 06 15:03:14 Laibsch: yes I did. Jan 06 15:04:04 Ok Jan 06 15:06:17 likewise, there's an image postprocessing func, which makes passwd empty, it can be made to stuff another value. and debug builds allow to ssh with empty password. Jan 06 15:07:03 psokolovsky_: thanks, I'll search for that. Jan 06 15:08:19 likewise: the base-passwd recipe also has a way to set a password Jan 06 15:09:23 XorA|gon1: np. Please keep me informed about your experience with that server. Jan 06 15:11:05 koen: the "do_install_append_openmn()" task? Jan 06 15:12:13 likewise: yes Jan 06 15:28:05 # NOTE: rng-tools is only here until argp-standalone can be built! nfs-utils \ Jan 06 15:28:05 UCLIBC_UNSUPPORTABLE_PACKAGES = "\ Jan 06 15:28:13 ~lart nslu-someone Jan 06 15:28:14 * ibot pours hot grits down the front of nslu-someone's pants Jan 06 15:38:15 hrw: http://lists.linuxtogo.org/pipermail/angstrom-distro-devel/2007-January/000101.html Jan 06 15:39:33 koen: I noticed it already Jan 06 15:39:59 koen: I'll run task-base run on celf machine Jan 06 15:42:06 arm this time only Jan 06 15:42:56 a1200 a780 amsdelta collie compulab-pxa270 devkitidp-pxa255 ep93xx gumstix h1910 h1940 h2200 h3600 h3900 h4000 h5000 h6300 htcblueangel htcuniversal hx2000 hx4700 jornada56x jornada7xx ks8695 logicpd-pxa270 magician mainstone mtx-3 mx21ads mx31ads navman-icn330 netbook-pro nokia770 omap1510inn omap1610h2 omap1710h3 omap2420h4 omap5912osk palmld palmtx qemuarm rokr-e2 rx3000 simpad smdk2440 triton Jan 06 15:43:02 nice set of arm we have Jan 06 15:43:12 now time to remove armv4 from list.. Jan 06 15:44:35 grrrrrrrr Jan 06 15:45:20 gremlin_it, btw, did you hear that anstrom now offers "unofficial" OABI support? don't you want to try it? Jan 06 15:46:04 psokolovsky_ yes yes i notice ... now i'm spending my time on usb ... after this i'll try the distro ... Jan 06 15:46:20 gremlin_it, nice! Jan 06 15:46:32 [hrw@misc1 angstrom]$ for machine in a1200 a780 amsdelta compulab-pxa270 devkitidp-pxa255 ep93xx gumstix h1910 h1940 h2200 h3900 h4000 h5000 h6300 htcblueangel htcuniversal hx2000 hx4700 ks8695 logicpd-pxa270 magician mainstone mtx-3 mx21ads mx31ads navman-icn330 netbook-pro nokia770 omap1510inn omap1610h2 omap1710h3 omap2420h4 omap5912osk palmld palmtx rokr-e2 rx3000 smdk2440 triton ; do MACHINE=$machine && echo $MACHINE && echo "MACHINE='$MACHINE'" >conf/au Jan 06 15:46:43 koen (or someelfe) make a good job to offer oabi version of angstrom Jan 06 15:47:40 koen: looks like all (nearly?) !armv4 machines use correct tune settings Jan 06 15:49:22 03pfalcon 07org.oe.dev * r63418984... 10/ (4 files in 2 dirs): mc: Unify recipes. Jan 06 15:49:27 03pfalcon 07org.oe.dev * r128ea1f5... 10/ (4 files in 2 dirs): Jan 06 15:49:27 mc: More fine-grained packaging. Jan 06 15:49:27 * Split extfs scripts (mc-extfs) and syntax highlighting (mc-hilite), each Jan 06 15:49:27 100KB+, out of main package. Jan 06 15:50:19 psokolovsky_: does mc atleast RRECOMMENDS them? Jan 06 15:50:59 hrw, sure ;-) Jan 06 15:51:41 NOTE: package linux-2.6.19-r0: task do_compile: started Jan 06 15:55:37 Can anyone help me with building mysql-native? See bug 1743. Jan 06 15:56:26 I keep getting configure: error: "MySQL needs a off_t type." Jan 06 15:57:16 hi koen, can you benchmark the final version of my optimized unquantizer for ffmpeg/mplayer on XScale cpu? Jan 06 15:58:56 http://www.oesf.org/forums/index.php?showtopic=22280&view=findpost&p=149865 Jan 06 15:59:29 it should be about 10% faster now, but i wonder how it perform on xscale Jan 06 16:02:01 ssvb: we fount this code sniplet from the manual, but to be honest we could not do much with it.. not enough asm experience, especially none on arm :) but that anyway looks like something required for loking, does arm has any atomic increment instruction? Jan 06 16:02:11 03pfalcon 07org.oe.dev * r7187c31b... 10/ (1 conf/machine/asus730.conf): asus730: New machine, Asus MyPal A730 - PXA27x, VGA, other goodies. Jan 06 16:02:28 on x86 we only need an extra lock thing on SMP, but on UP there is a code sniplet that just does the math and thats it Jan 06 16:03:17 koen: does gpe-scap replace gpe-screenshot? Jan 06 16:03:22 s/fount/found/g Jan 06 16:03:35 hrw: yes Jan 06 16:03:49 koen: will change then this in .oz354x Jan 06 16:04:10 Jin^eLD: there is no special instruction for atomic increment, I can try to make some working code from that example Jan 06 16:05:18 ssvb: that would be great, you'd help us a lot with that! Jan 06 16:05:41 it would be for mediatomb.org, GPL UPnP server Jan 06 16:06:51 basically we are looking for atomic inc/dec that is faster, than a construct using pthread mutex locks Jan 06 16:07:02 ssvb, so, source/patches are in maemo svn? Jan 06 16:08:11 psokolovsky_: yes, I also submitted this patch upstream, so there is a chance that it will get into the next version of mplayer or anything using ffmpeg Jan 06 16:09:10 ok, nice. too bad I don't have time to test it now, but thanks for your work on optimizing video stuff! Jan 06 16:15:03 Can anyone help me with building mysql-native? See bug 1743. Jan 06 16:15:32 I keep getting configure: error: "MySQL needs a off_t type." Jan 06 16:15:38 03hrw 07org.oe.dev * r469ff24b... 10/ (1 conf/machine/progear.conf): progear: added modules for sound, pcmcia, i2c Jan 06 16:15:44 03hrw 07org.oe.dev * re7b51cdf... 10/ (1 conf/machine/progear.conf): progear: remove APM from features Jan 06 16:16:09 dothedog: it builds for me here Jan 06 16:16:40 hrw: Yeah, I just keep getting that off_t type error on configure. Not sure what I am doing wrong. Jan 06 16:19:53 libgcc1_4.1.1-r9_arm-oabi.ipk Jan 06 16:19:54 yay! Jan 06 16:19:58 03pfalcon 07org.oe.dev * ra44afc79... 10/ (1 packages/linux/linux-handhelds-2.6.inc): linux-handhelds-2.6: Add asus730 to COMPATIBLE_MACHINE. Jan 06 16:20:21 hrw: is there anything I can try to figure this out? Jan 06 16:20:24 koen, cool! Jan 06 16:20:45 dothedog: you use x86-64? Jan 06 16:21:20 Not on this build. I am also trying it on a different machine that is, but this build is on a x86 smp machine. Jan 06 16:21:39 Jin^eLD: I googled for "atomic increment arm" and found a few interesting links (some of them contain code for gcc): http://lists.helixcommunity.org/pipermail/common-dev/2003-November/000534.html Jan 06 16:22:05 hrw: uname -a Jan 06 16:22:05 Linux i10178 2.6.18.2-34-default #1 SMP Mon Nov 27 11:46:27 UTC 2006 i686 i686 i386 GNU/Linux Jan 06 16:22:49 ssvb: I did a lot of googling yesterday, maybe I missed something, let me see Jan 06 16:22:59 ssvb: I'll run the unq tests in a few minutes Jan 06 16:25:36 ssvb: I saw that post, but we were not sure, so we went with the kernel sources which did not work; but indeed, we can try this first Jan 06 16:26:26 ssvb: http://rafb.net/p/ro3rKq56.html Jan 06 16:28:27 koen: thanks, everything looks ok, but i guess cpu clock freq detection is still wrong :) Jan 06 16:29:25 hrw: I am running gcc version 4.1.2 20061115 (prerelease) (SUSE Linux). Should I try gcc 3.3? Jan 06 16:30:00 17:58 hrw@home:~$ gcc --version Jan 06 16:30:00 gcc (GCC) 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) Jan 06 16:31:33 hrw: anything I can try? what sets off_t? Jan 06 16:38:04 dothedog: edit OE/site/i686-linux and add "ac_cv_sizeof_off_t=${ac_cv_sizeof_off_t=4}" line Jan 06 16:47:17 hrw: shouldn't kernel-module-yenta-socket be in task-base pcmcia? Jan 06 16:47:47 koen: and you will add there ALL pcmcia controlers? Jan 06 16:48:02 koen: same for alsa sound module? Jan 06 16:48:36 hmmm Jan 06 16:49:33 psokolovsky_, gremlin_it: which kernel version is the best for h3600? Jan 06 16:50:34 hrw: that got me past the configure problem, now it dies on compile. See: http://pastebin.ca/307990 Jan 06 16:51:00 koen for a user ? Jan 06 16:51:09 : error: #error "Error: InnoDB's ulint must be of the same size as void*" Jan 06 16:51:21 gremlin_it: yes, which 2.6.x kernel? Jan 06 16:51:32 dothedog: can you try to build mysql without OE? Jan 06 16:51:55 dothedog: unpack mysql 4.1.18 archive and build it for your host environment Jan 06 16:52:01 mhhh for a user problaly 2.4 ... but 2.6 is ok for latest 2.6.19-hh7 last i test Jan 06 16:52:03 dothedog: I wonder will it fail Jan 06 16:52:14 I'll give it a try. Jan 06 16:52:26 uf.. gpe-image or c7x0/oz354x builds Jan 06 16:52:30 dothedog: thx Jan 06 16:53:01 hrw: I'm trying on gpe-image Jan 06 16:53:49 dothedog: I'm working on OZ 3.5.5 now Jan 06 16:53:57 cu guys Jan 06 16:54:03 hrw: 3.5.5 = 2.6 except collie? Jan 06 16:54:25 hrw: Thanks. Jan 06 17:04:17 hrw|gone, RP: http://bugs.openembedded.org/show_bug.cgi?id=1753 Jan 06 17:07:48 Jin|away: the mediatomb .bb needs to be modified a bit for the new location of taglib-config Jan 06 17:32:26 wsuetholz: Are you there? Jan 06 17:35:10 koen: ... in another words, the latest. 2.6.19-hh7 gremlin_it mentioned is the last tagged. Jan 06 17:36:52 hrw|gone: I just configured/compiled mysql 4.1.18 AND 4.1.22 outside of OE and they both worked. Jan 06 17:54:36 Is there a way to trace back the source package of the files in the resulting rootfs? "/ipk/Packages.filelist" almost provides this info, but only for the last build?? Jan 06 17:56:09 ipkg -whatprovides ? Jan 06 17:56:21 there is some switch for that Jan 06 17:56:25 never used it, though Jan 06 17:57:12 03koen 07org.oe.dev * r99de646e... 10/ (1 conf/distro/angstrom-2007.1-oabi.conf): angstrom oabi: add kernel version for h3600 Jan 06 18:03:33 Likewise : did you get my message ? Jan 06 18:04:05 gerwinin: I did not see any mesg from you? IRC or email? Jan 06 18:08:00 hi everybody..! can someone please help me - i already have a working kernel, so basically i want to use OE without rebuilding it. is there something i need to do for that? does OE always build whole of the rootfs? Jan 06 18:08:30 shirour: OE builds what you tell it to build Jan 06 18:11:22 koen: but if i tell it to build "vim", wouldn't it look for all of its dependencies? Jan 06 18:11:37 and once it sees i haven't done anything yet, build everything from scratch (i.e. kernel etc')? Jan 06 18:12:05 for example - let's say i want to build something for opie, how would it know which libs i already have ? Jan 06 18:12:10 shirour, kernel is not dependency of vim. but glibc is, for example Jan 06 18:12:35 koen, how is efika support? Jan 06 18:14:01 psokolovsky: hi, thanks..! so how do i know how far back would it go? Jan 06 18:14:23 let's say i want to compile something for opie, will it build opie from scratch? Jan 06 18:14:35 shirour, it would go deep enought, but not up to kernel ;-) use "-v -n" for bitbake Jan 06 18:14:44 shirour, yep. Jan 06 18:14:59 is there anyway by which i can copy my exist rootfs so that oe will know about the stuff i currently have? Jan 06 18:15:27 (*any way, sorry0 Jan 06 18:15:30 shirour, yes, RTFM ;-) Jan 06 18:15:37 ok, thans Jan 06 18:15:39 thanks Jan 06 18:16:06 but can you guide me a bit on where to look? Jan 06 18:16:13 will it be enough if i just copy the rootfs somewhere? Jan 06 18:16:41 shirour, start with looking at local.conf - it's well of knowledge ;-) Jan 06 18:17:13 shirour, rootfs is not really enough, you need properly installed dev packages Jan 06 18:21:33 psokolovsky: and how long + how much space would it take to compile opie from scratch? Jan 06 18:21:49 shirour, a day/5Gb Jan 06 18:22:14 posokolvsky: mm... but that is not necessary, right..?? Jan 06 18:23:02 shirour, actually, that will save you lotsa time and effort - to just leave it overnight ;-) Jan 06 18:23:38 mmm... Jan 06 18:23:50 but i need 5 gb for that Jan 06 18:23:51 03likewise 07org.oe.dev * rb06b77d4... 10/ (1 packages/alsa/alsa-utils_1.0.13.bb): alsa-utils: Fixed typo's in recipe: speakertest vs. speaker-test. Jan 06 18:29:18 posokolvsky: thanks for everything..! bye Jan 06 18:31:39 philippe: ping Jan 06 18:43:31 koen|away: only in trunk (bitbake bug)? Jan 06 18:45:21 koen: Have you seen this: http://handhelds.org/hypermail/htc-pxa/current/0061.html ? Do you agreed with mickeyl? Jan 06 18:45:30 03pfalcon 07org.oe.dev * rfe96c9a0... 10/ (1 packages/qte/qte-common_2.3.10.inc): qte: Add asus730 overrides. Jan 06 18:45:37 03pfalcon 07org.oe.dev * re6925e9d... 10/ (1 packages/opie-pics/opie-pics.inc): opie-pics: Stop hardcoding machines, use machine features (GUI_MACHINE_CLASS instead). Jan 06 18:47:20 zecke, Hi! Want to use you invitation to ask stupid opie q's ;-): Where can I read about OPIE debug logging? Jan 06 18:48:26 posokolvsky: like oDebug? the way it is implemented or how you can configure it? Jan 06 18:49:01 http://handhelds.org/cgi-bin/cvsweb.cgi/opie/libopie2/opiecore/odebug.cpp?rev=1.9&content-type=text/x-cvsweb-markup Jan 06 18:49:02 zecke, how to configure, or more specifically what actions I should perform to get the log to file for example? ;-) Jan 06 18:49:04 impl Jan 06 18:50:10 posokolvsky: come to #opie Jan 06 18:54:00 re Jan 06 18:54:08 koen: ok, whats the new location now? Jan 06 19:03:32 Jin^eLD: STAGING_BINDIR_CROSS iirc Jan 06 19:03:45 zecke: and in 1.6 I think Jan 06 19:03:50 ok, I'll change that Jan 06 19:03:59 zecke: it was introduced with hrw's latest csets Jan 06 19:04:30 so just like that? --with-taglib-cfg=${STAGING_BINDIR_CROSS}/taglib-config" Jan 06 19:05:53 posokolvsky: first: mails about about go to the angstrom mailinglist Jan 06 19:06:06 koen: right, looks weird Jan 06 19:06:10 hrw|gone: ping? Jan 06 19:06:30 cvs co server:2401/ I doubt this works.. Jan 06 19:06:42 posokolvsky: write a mail to the angstrom devel list and we might consider your requests Jan 06 19:06:49 koen, well, as Angstrom doesn't support and doesn't have plans to support opie, how it can go there? Jan 06 19:06:56 koen, ok Jan 06 19:06:58 posokolvsky: so stop pissing me off by using some random shit mailinglist Jan 06 19:07:07 lol Jan 06 19:07:16 koen: hehe, don't get pissed off so easily Jan 06 19:07:31 zecke: don't piss me off so easiliy ;) Jan 06 19:07:35 create some vodoo dolls, this helped me Jan 06 19:09:31 hrw|gone: ping me once you are back Jan 06 19:10:13 interesting, I have an efika zImage , exts fs and tarball Jan 06 19:10:39 Crofton: efika should build from an unpatched .dev now Jan 06 19:11:27 yep Jan 06 19:11:37 I just finshed the build I started a few days ago Jan 06 19:12:06 posokolvsky: mickeyl and I still stand with the conditions we have set for opie inclusion into angstrom Jan 06 19:12:38 posokolvsky: so find at least to people committed to maintaining opie-image for angstrom and we can start to talk Jan 06 19:12:55 koen, We already discussed this and I got defibitive response fromm you - OPIE is not currently supported. Jan 06 19:13:05 _currently_ Jan 06 19:13:47 koen, with the kernel and the ext2 image, what are my chances of getting the efika bootable? Jan 06 19:13:59 (why do I always have to repeat myself?) Jan 06 19:14:06 Crofton: about 90% Jan 06 19:14:21 koen, In this regard, I'd like to still make opie easily buildable for folks, and let developers gather around that, not each sit in its own corner. So, it calls for separate distro (sigh). And the real q here is how to call that distro. Jan 06 19:14:38 posokolvsky: whatever Jan 06 19:14:46 koen, guess I need to read up on copying images to the hd via the bot thingy? Jan 06 19:14:55 posokolvsky: if you don't want to listen what mickeyl and I say, fine Jan 06 19:15:15 Crofton: yep, or setup bootp Jan 06 19:15:21 posokolvsky: why don't you wait for Lorn's Opie II? Jan 06 19:15:44 koen, I want to listen. And mikey asked to call it angstrom-opie ;-) Jan 06 19:15:53 I want to get my efika going in a free standing mode Jan 06 19:16:02 (why do I always have to repeat myself?) Jan 06 19:16:18 zecke, too long a wait ;-). and I have an itch to do "right disto management", lol ;-E Jan 06 19:16:21 koen, because you won't say what people want to hear Jan 06 19:16:23 :) Jan 06 19:16:31 koen: commited the .bb in rev 957, but I did not test it yet Jan 06 19:16:45 Jin^eLD: ok, will test later Jan 06 19:16:58 posokolvsky: well, he is full time paid developer. So it can't take too long *bites on his tongue* Jan 06 19:17:10 posokolvsky: why do 4 lines in a local.conf need a seperate distro? Jan 06 19:17:53 posokolvsky: if future opie maintainers are too stupid to understand that I (and mickeyl) will be glad to ban opie Jan 06 19:17:53 koen, because you refuse to re-add those 4 lines! I wish those 4 lines being in *ansgtrom* very much myself. Jan 06 19:18:16 and remove it from OE completely Jan 06 19:18:17 koen, well, you want 2 people, get them: hrw and me. Jan 06 19:18:38 posokolvsky: hrw doesn't want to maintain opie for angstrom Jan 06 19:19:31 03zecke123 * r736 10/ (2 files in 2 dirs): Make cvs_port a string and an empty one as well. This should fix the concatination error koen was seeing and describing in #1753 Jan 06 19:20:23 koen, then, so far noone but me wants to mainatain, either. But few folks want to use. My plan is to let them do that, make them want to develop/mainatin it, and resubmit Ansgtrom support request at later time. Why do you think it's stupid? it's just real sitaution with it. Jan 06 19:20:56 posokolvsky: you didn't listen to what I said Jan 06 19:21:25 the problem is the users will become *very* demanding Jan 06 19:21:58 specially Opie users Jan 06 19:22:04 :) Jan 06 19:24:49 koen, you said about "to" (I guess, two) people to maintain OPIE in Angstrom. If I missed something, sorry, I post that msg to the list, let's please spend another, but final round there. Thanks. Jan 06 19:25:10 posokolvsky: the next commit is for you Jan 06 19:25:19 posokolvsky: and spellcheck your nick Jan 06 19:26:30 03koen 07org.oe.dev * r44124b9f... 10/ (2 files in 2 dirs): add an .inc for those poor clueless users that demand opie support but surprisingly do not want to do anything to support it Jan 06 19:27:09 posokolvsky: do 'include conf/distro/include/opie-users-are-too-stupid-too-put-this-in-local.conf.inc' and you get opie support Jan 06 19:27:44 Crofton: I am now running EFIKA over DHCP/NFS. Do you want to run from HDD or USB? Jan 06 19:27:51 hdd :) Jan 06 19:27:54 lol Jan 06 19:28:09 * Crofton hates opie for the amount of pain it has inflicted Jan 06 19:28:24 Crofton: pain? what pain? Jan 06 19:28:40 the never ending opie questions Jan 06 19:28:40 ~lart ALSA Jan 06 19:28:40 * ibot blames ALSA for all the evil in the world Jan 06 19:28:53 zecke: bait & switch Jan 06 19:28:53 koen: thanks ;-I Jan 06 19:29:04 a guy here spent so time figuring out opie, then found out it was based on old qt Jan 06 19:29:07 koen: I wonder about the second "too" Jan 06 19:29:24 zecke: I'[m too stupid too spellcheck :) Jan 06 19:29:29 gopefully opie II appears and solves these problems Jan 06 19:29:35 zecke: it's a turingtest Jan 06 19:29:58 koen: you need to have opie-turing installed and you expose them to chicken vs. egg problem? Jan 06 19:30:07 zecke: sort off Jan 06 19:30:14 * chouimat is bored ... now snow and stupid rain outside can't go out and buy food Jan 06 19:30:20 s/now/no Jan 06 19:30:32 chouimat, sounds like you may starve Jan 06 19:30:52 Crofton I have a lot of ramen here :) Jan 06 19:30:58 * lpotter sees nothing but blue sky and warm sunshine Jan 06 19:31:08 chouimat: order a pizza then Jan 06 19:31:26 lpotter: it's January here ... supposed to be around -10C not 10C Jan 06 19:31:56 chouimat: 10C? How did you manage to get snow then? Jan 06 19:32:00 likewise: might be a good idea too ... Jan 06 19:32:03 likewise: no snow :) Jan 06 19:32:22 chouimat: ah, I read "now snow" Jan 06 19:32:51 january here too.. :) Jan 06 19:33:09 and supposed to be 30 :) Jan 06 19:34:20 likewise: can't type today ... brain too fast for my hands :) Jan 06 19:34:39 chouimat: sure :-) Jan 06 19:34:41 lpotter: you're in summer :) Jan 06 19:35:56 "The good folks at embeddedarm.com apparently ship their ARM devices with NetBSD preinstalled. This is really good news. Once you have a NetBSD version installed, staying in sync with reality is quite easy because you can build your own worlds on the device either way. You can get a 160GB usb harddisk for €100, which is a good bargain. You configure something like 8 gigabytes of swap space, and the machine can handle a true load of services, althought Jan 06 19:36:22 hehe Jan 06 19:38:11 zecke: http://bsdblog.livejournal.com/ Jan 06 19:38:44 koen: thanks for viewing my flickr pictures Jan 06 19:39:01 zecke: :) Jan 06 19:41:29 posokolvsky: 1) get something working to show 2) show the patches for step 1 3)... 4) profit Jan 06 19:41:49 ok Jan 06 19:41:59 Yay! Got EFIKA making sound (and /boot/zImage sounds better then Marco Borsato). Jan 06 19:42:17 lol Jan 06 19:42:22 posokolvsky: fwiw gpe support is about to get pulled as well Jan 06 19:42:40 OpenMoko :) Jan 06 19:43:06 koen, pulled where? Jan 06 19:43:15 posokolvsky: from angstrom Jan 06 19:43:19 hi! can someone please help me? when i'm running "bitbake opie-image" i get "ERROR: Unable to open conf/bitbake.conf"... any idea? Jan 06 19:43:41 posokolvsky: next month there will be angstrom-bootstrap-image and angstrom-base-x11-image Jan 06 19:44:28 shirour: set BBPATH? Jan 06 19:45:47 koen, that's your decisions, as a maintainer. there're always some alternatives for people to bother with, if you don't want to concentrate them around angstrom. I perssnally interested in Linux on PDAs more than in politics around it, so anything works for me ;-) Jan 06 19:47:18 posokolvsky: you need to have something working before you can interest people Jan 06 19:47:35 posokolvsky: otherwise it will stay a developer toy forever (like oz and fam are) Jan 06 19:48:10 koen, in open-source world those things are usually done in parallel ;-) Jan 06 19:48:43 koen, that's why I usually call Angstrom a last hope in local user forums. please don't screw it up ;-) Jan 06 19:48:44 zecke: i did Jan 06 19:49:27 shirour: well, it is either incorrect or the bad error handling of bitbake has seen a parser error somewhere Jan 06 19:49:35 posokolvsky: angstrom will virtually support anything that has dedicated maintainers for it Jan 06 19:49:55 well, then everything's ok Jan 06 19:49:59 shirour: check BBPATH, use bitbake -v (upto -vvv) to see what it is doing. same for -D Jan 06 19:53:45 posokolvsky: you have mail Jan 06 19:54:33 03pfalcon 07org.oe.dev * rdd99a3ac... 10/ (1 packages/opie-button-settings/opie-button-settings.inc): Jan 06 19:54:33 opie-button-settings: Use ipaq button images for bunch more ipaqs. Jan 06 19:54:33 * Well, at least it's better than completely unfamiliar Zaurus button Jan 06 19:54:33 images. Jan 06 19:54:37 03koen 07org.oe.dev * r3492d999... 10/ (3 files in 2 dirs): make include zecke compliant Jan 06 19:54:53 koen: more spelling errors? Jan 06 19:55:13 zecke: I hid a frikandel inside it Jan 06 19:56:50 thanks all! bye Jan 06 19:57:15 YEAH my Stub Creator is done!!!! Jan 06 20:03:11 Does anybody else gets lots of: Jan 06 20:03:12 NOTE: Couldn't find shared library provider for libc.so.6 Jan 06 20:03:12 NOTE: Couldn't find shared library provider for libmb.so.1 Jan 06 20:03:32 I wonder, if RP missed something with that refactor, or I misupdated something... Jan 06 20:03:34 not since objdump got fixed Jan 06 20:03:39 posokolvsky: not recently Jan 06 20:03:41 03cbrake 07org.oe.dev * r1d4c5bb3... 10/ (4 files in 3 dirs): u-boot 1.1.6: add patch so PXA255 IDP builds Jan 06 20:03:49 03cbrake 07org.oe.dev * r14721209... 10/ (6 files in 3 dirs): devkitidp-pxa255 2.6.19: move kernel to 2.6.19 and update patch Jan 06 20:04:30 03rwhitby 07org.oe.dev * ra88c6575... 10/ (1 packages/meta/slugos-packages.bb): slugos-packages: Demoted erlang, and fixed syntax errors. Jan 06 20:05:01 03cbrake 07org.oe.dev * ra7c10592... 10/ (3 files in 3 dirs): devkitidp-pxa255 2.6.11: drop this version Jan 06 20:05:53 A package I am trying to build for Sharp ROM fails in configure with "configure: error: "POSIX termios operations are required"" What package provides this termios? Or is this some incorrect include? Jan 06 20:06:07 Laibsch: glibc Jan 06 20:06:26 Laibsch: paste the config.log please Jan 06 20:06:39 Ic, so in the case of Sharp ROM it should be provided and thus becomes an inclusion problem, right? Jan 06 20:06:58 Laibsch: either your ancient toolchain is broken^has features, or the configure check is broken Jan 06 20:07:12 who knows ;-) Jan 06 20:07:14 Laibsch: yes, or some mental outage of the developer in question Jan 06 20:08:13 zecke: http://oz.leggewie.org/wip/log.do_configure.31592 Jan 06 20:08:50 My openzaurus-unstable is not yet back so I thought I'd take advantage of the prebuilt Sharp one ;-) Jan 06 20:08:54 Laibsch: config.log please ;) Jan 06 20:09:09 * Laibsch scratches head Jan 06 20:09:13 just a moment Jan 06 20:09:21 we had that already :) Jan 06 20:10:38 zecke: http://oz.leggewie.org/wip/config.log Jan 06 20:11:29 /usr/include/bits/pthreadtypes.h:69: warning: unnamed struct/union that defines no instances Jan 06 20:11:39 /usr/lib/libncurses.so: could not read symbols: Invalid operation Jan 06 20:11:44 note the /usr/lib Jan 06 20:11:50 and the /usr/include Jan 06 20:12:32 zecke: I think what is needed is /usr/local/arm/2.95.3/arm-linux/include/linux/termios.h or some other /usr/local/arm/2.95.3/arm-linux/include/*/termios.h Jan 06 20:12:35 Sharp GCC lacks my clever hack :) Jan 06 20:12:42 Laibsch: time to learn m4 :) Jan 06 20:12:58 ah, fuck it. Jan 06 20:13:00 m4 makes my eyes bleed Jan 06 20:13:12 I'll just wait until OE toolchain is up again. Jan 06 20:13:26 Laibsch: collect a pile of dog shit, put it in a brown paper-bag, burn this paper bag, place it in front of the "Devloper Under Test" door and ring the bell Jan 06 20:13:27 If it builds for OZ, I'll just close the bug report anyhow ;-) Jan 06 20:13:46 zecke: Sounds like a plan. Jan 06 20:14:10 Laibsch: this dev is moron and puts -I/usr/include somewhere :) Jan 06 20:14:40 zecke: Couldn't I just work with OE_EXTRA_CONF or something so this is included? Jan 06 20:14:48 configure:8168: ccache arm-linux-gcc -march=armv4 -o conftest -I/export/interim/openembedded/tmp/sharprom-compatible/staging/arm-linux/include -fexpensive-optimizations -fomit-frame-pointer -O2 -I/export/interim/openembedded/tmp/sharprom-compatible/staging/arm-linux/include -lncurses -I/usr/include -L/usr/lib -L/export/interim/openembedded/tmp/sharprom-compatible/staging/arm-linux/lib -Wl,-rpath-link,/export/interim/openembedded/tmp/sharprom-compatible Jan 06 20:14:51 to be exact Jan 06 20:15:09 Laibsch: no, fix the evil at the root and shoot the developer Jan 06 20:15:13 Laibsch: could you post the configure.ac ? Jan 06 20:16:30 03pfalcon 07org.oe.dev * r75de3114... 10/ (3 files in 2 dirs): Jan 06 20:16:30 matchbox-panel: Update from using specific machine overrides to using machine Jan 06 20:16:30 features (GUI_MACHINE_CLASS). Jan 06 20:16:34 03cbrake 07org.oe.dev * r21538320... 10/ (1 MAINTAINERS): MAINTAINERS: add devkitidp to my entry Jan 06 20:16:42 koen: http://oz.leggewie.org/wip/configure.ac Jan 06 20:16:58 zecke: the nokia n800 tablet has a armv6 :) Jan 06 20:17:49 n800? Jan 06 20:19:06 Laibsch: adding ac_cv_sys_posix_termios to the arm sitefile would probably the easiest workaround Jan 06 20:19:11 anyone here about the Marvell 88W8385 driver? :( Jan 06 20:19:55 two heads in mtn Jan 06 20:20:33 Laibsch: try --with-openssl-dir=/bla/bla in EXTRA_OECONF and see what happends Jan 06 20:21:15 zecke: /bla/bla being the path to termios.h? Jan 06 20:22:07 Laibsch: "/bla/bla" means /bla/bla Jan 06 20:23:08 wow, interesting concept ;-) Jan 06 20:23:48 Laibsch: close to "more is more" Jan 06 20:24:19 I thought "less is more".... Jan 06 20:27:15 I hired a new test engineer: http://www.sidebranch.com/openembedded/efika_senna.jpg Jan 06 20:27:44 likewise: I can read your card number and expiry date Jan 06 20:28:05 zecke: good could you order me some nice hardware then ? Jan 06 20:28:15 sure, plastic dolls? Jan 06 20:28:44 zecke: if there are any new models I don't have yet, why not! Jan 06 20:29:01 likewise: they have this new one, it is really exciting Jan 06 20:29:04 zecke: but the place is pretty crowded with pussy as you can see Jan 06 20:30:40 cool http://www.iwantoneofthose.com/usb-hamster-wheel/index.html Jan 06 20:30:51 likewise: zecke ^^^^ Jan 06 20:31:50 lol Jan 06 20:32:42 zecke: hours of fun at the office :) Jan 06 20:44:25 anyone mind if I merge dev to elimnate the two heads? Jan 06 20:46:08 go ahead Jan 06 20:46:25 in process Jan 06 20:47:32 done Jan 06 21:01:51 ~lart me Jan 06 21:01:51 * ibot hereby declares hrw a troll Jan 06 21:02:06 cbrake: big THX for fixing devkitidp target Jan 06 21:03:21 zecke: pöng Jan 06 21:03:55 hrw: do you agree with the fix to cvs? Jan 06 21:04:35 zecke: yes Jan 06 21:05:16 ~lart trolltech for "| kernel/qapplication_qws.cpp:2133: error: cast from 'void*' to 'int' loses precision Jan 06 21:05:17 * ibot pulls out his louisville slugger and uses trolltech's head to break the homerun record for "| kernel/qapplication_qws.cpp:2133: error: cast from 'void*' to 'int' loses precision Jan 06 21:09:21 zecke: next time I will triple check Jan 06 21:09:52 cbrake: NOTE: Error on parsing: /home/hrw/devel/oe/org.openembedded.dev/packages/linux/devkitidp-pxa255_2.6.19.bb:1 Jan 06 21:10:00 cbrake: I will post a fix Jan 06 21:18:09 cbrake: pushed Jan 06 21:18:29 03hrw 07org.oe.dev * r47def83c... 10/ (1 packages/linux/devkitidp-pxa255_2.6.19.bb): devkitidp-pxa255: use COMPATIBLE_MACHINE instead of python function, use KERNELORG_MIRROR Jan 06 21:19:53 lpotter: ping? Jan 06 21:40:10 zecke: It does not look like your /bla/bla suggestion really changed much Jan 06 21:40:21 And I would have been very surprised if it had. Jan 06 21:41:01 Laibsch: The error is: Someone is putting "-I/usr/include" into CPPFLAGS Jan 06 21:41:30 Laibsch: now your task is to find out who is doing that. I do not feel like looking at pwsafe as disgusting autotools usage makes me puke Jan 06 21:41:44 and I'm hacking away on stuff... Jan 06 21:41:56 sure, I did not expect you to fix this. Jan 06 21:42:05 And it is not a high prio for me either Jan 06 21:42:25 I just found this RFP on the BTS and thought I would have a go at it. Jan 06 21:42:55 hehe, well I don't like to know about errors... Mail upstream and ask him to fix his buildsystem Jan 06 21:43:00 WNPP RFP? Jan 06 21:44:45 hrw, yes Jan 06 21:45:28 !oebug 1065 Jan 06 21:45:29 * * Bug 1065, Status: NEW, Created: 2006-06-05 20:58 Jan 06 21:45:30 * * shadow(AT)serverart.org: \[request for package\] pwsafe - password keyring utility Jan 06 21:45:31 * * http://bugs.openembedded.org/show_bug.cgi?id=1065 Jan 06 21:46:38 zecke: -I/usr/include should be replaced by what? -Iusr/include? I understand this is something about relative and absolute paths, right? Jan 06 21:47:02 Laibsch: no this is about including types from the host Jan 06 21:47:33 Laibsch: do you just call ./configure, or regenerate the autofoo? Jan 06 21:48:26 hrw: pong Jan 06 21:49:09 lpotter: qt/e 3.3.5 does not build for x86-64 ;( and I suppose that it will never will? Jan 06 21:49:23 zecke: I inherit autotools in pwsafe bb. then I say "rebuild pwsafe" in bitbake shell. (Not sure if that is what you wanted to know) Jan 06 21:49:48 hrw: not real sure, havent really done much with qte3 Jan 06 21:49:52 lpotter: can TT add 'x86-64 -> arm' combo into Qtopia test queue? Jan 06 21:50:54 morning all Jan 06 21:50:58 hi RP Jan 06 21:51:07 i can find out if brisbane has any 64bit machines Jan 06 21:51:30 Laibsch: do_configure_prepend() { rm ${S}/configure } Jan 06 21:51:53 lpotter: would be nice - I do not have x86 machines here to do qtopia/opieII work on them Jan 06 21:52:06 ibot, seen prpplague* Jan 06 21:52:24 i haven't seen 'prpplague*', Marex Jan 06 21:52:29 ibot, seen prpplague Jan 06 21:52:32 prpplague was last seen on IRC in channel #edev, 1d 11m 45s ago, saying: 'wonder if thats the same cpu thats in the xgp mini'. Jan 06 21:54:14 lpotter: and "kernel/qapplication_qws.cpp:2133: error: cast from 'void*' to 'int' loses precision" does not looks like 64bit safe.. (qt/e 3.3.5) Jan 06 21:55:23 ~hail someone for 64bit-cleanup.patch Jan 06 21:55:26 blame oslo Jan 06 21:55:26 * ibot bows down to someone for 64bit-cleanup.patch and chants, "I'M NOT WORTHY!!" Jan 06 21:59:27 ~hail mtn pluck Jan 06 21:59:31 * ibot bows down to mtn pluck and chants, "I'M NOT WORTHY!!" Jan 06 21:59:35 ~blame oslo Jan 06 21:59:37 * ibot blames oslo (and Canada) for all the evil in the world Jan 06 22:16:43 http://www.reallylinux.com/docs/linuxlovingmen.shtml Jan 06 22:26:27 hrw: I config/compiled mysql_4.1.18 in suse. I actually got mysql-native compiled however, I'm not sure what I did is correct. See http://bugs.openembedded.org/show_bug.cgi?id=1743. I updated it. Jan 06 22:27:05 dothedog: I saw that - will include in branch Jan 06 22:27:25 dothedog: thx for digging Jan 06 22:27:32 hrw: No problem. Jan 06 22:27:50 Now I can't compile mysql-4.1.18 Jan 06 22:28:25 dothedog: what this time failed? Jan 06 22:28:46 I get this: Jan 06 22:28:49 http://pastebin.ca/308609 Jan 06 22:29:50 dothedog: it should be ./gen_lex_hash rather.. Jan 06 22:29:57 please submit bug Jan 06 22:30:04 hrw: will do Jan 06 22:33:10 hrw: it is here: http://bugs.openembedded.org/show_bug.cgi?id=1755 Jan 06 22:33:45 Laibsch: configure is not getting regenerated, so we should remove it in the do_configure_prepend Jan 06 22:34:12 dothedog: thx Jan 06 22:34:16 !oebug 1755 Jan 06 22:34:17 * * Bug 1755, Status: NEW, Created: 2007-01-06 15:04 Jan 06 22:34:18 * * dothedog(AT)comcast.net: mysql-4.1.18 build fails on gen_lex_hash Jan 06 22:34:19 * * http://bugs.openembedded.org/show_bug.cgi?id=1755 Jan 06 22:34:41 zecke: Now I understand. Jan 06 22:36:47 .oz354x got 16 changesets today from me.. Jan 06 22:37:06 hrw: Wow Jan 06 22:38:01 result of 7:00 wakeup.. Jan 06 22:38:15 hrw: what time is it for you? Jan 06 22:38:19 00:07 Jan 06 22:38:40 hrw: already tomorrow for you ;) Jan 06 22:42:01 !oebug 1756 Jan 06 22:42:02 * * Bug 1756, Status: NEW, Created: 2007-01-06 15:11 Jan 06 22:42:03 * * papercrane(AT)reversefold.com: glibc binary locale generation fails with "Unable to load interpreter" Jan 06 22:42:04 * * http://bugs.openembedded.org/show_bug.cgi?id=1756 Jan 06 22:42:11 fun ^_^ Jan 06 22:42:29 I've got a patch on that bug that fixes binary locale generation for me. Jan 06 22:57:13 wb koen - cant sleep? Jan 06 22:57:25 koen: http://www.reallylinux.com/docs/linuxlovingmen.shtml Jan 06 22:58:10 hrw: I'm reading a book I can't seen to put away :) Jan 06 23:01:10 03joaquinduran 07org.oe.dev * rfcedd511... 10/ (1 packages/fltk/fltk_1.1.7.bb): Jan 06 23:01:10 fltk 1.1.7: Better configure options and revamped staging. Jan 06 23:01:10 * Add support for anti aliased fonts. Jan 06 23:01:10 * Enable double buffer to avoid flicking. Jan 06 23:01:10 * Enable OpenGL. Jan 06 23:01:10 * Fix dependency problems with X11 libraries. Jan 06 23:01:12 * Closes #1555 Jan 06 23:01:33 joaquinduran? Jan 06 23:01:52 who is it? Jan 06 23:02:07 or let me gues.. --author used.... Jan 06 23:06:34 03hrw 07org.oe.dev * re40db82f... 10/ (1 site/x86_64-linux): site/x86-64-linux: add ac_cv_sizeof_off_t to get mysql-native build - part of #1743 Jan 06 23:06:39 03hrw 07org.oe.dev * rd857a7c2... 10/ (1 site/ix86-common): site/ix86-linux: add ac_cv_sizeof_off_t to get mysql-native build - part of #1743 Jan 06 23:10:15 hrw: please add the test to mysql.m4 Jan 06 23:12:46 ko Jan 06 23:17:25 likewise: I have pussy on table as well :} Jan 06 23:17:46 zecke: sleeping as well? :-) Jan 06 23:18:01 likewise: currently playing with my usb harddrive Jan 06 23:18:39 zecke: you should buy anti-vibration stands for the drive then. Jan 06 23:18:57 zecke: in case she goes wild on it Jan 06 23:19:05 03koen 07org.oe.dev * rafcc206a... 10/ (1 packages/linux/linux-handhelds-2.6.inc): linux-handhelds-2.6: add support for writing reflash.ctl files containing kernel information as needed for hx4700 Jan 06 23:20:20 OK, first off, this is not a support request. I am interested in this from bug perspective. Jan 06 23:20:39 I installed xmms-embedded from the OZ feeds on my Z and it killed opie. Jan 06 23:20:53 killed i.e. I cannot restart it, the screen remains blank. Jan 06 23:21:23 I verified xmms-embedded to be the culprit be deinstalling and reinstalling all packages and dependencies that led to this. Jan 06 23:21:32 zecke: like http://pastebin.ca/308736 one? Jan 06 23:22:06 I guess there is some undeclared conflict between one of the packages that I have installed and xmms-embedded. Jan 06 23:22:29 I wonder how? Any such incompatibility known? Jan 06 23:23:07 hrw: eek Jan 06 23:23:10 I deinstalled xmms-embedded and opie runs fine again, now. So again, this has nothing to do with support. Just trying to make OE better. Jan 06 23:23:28 zecke: m4 is like perl to me Jan 06 23:23:39 hrw: it makes your eyes bleed? Jan 06 23:23:47 hrw: MYSQL_CHECK_SIZEOF Jan 06 23:23:53 03koen 07org.oe.dev * r3d9d209d... 10/ (4 files in 2 dirs): h3600 (and h3800 by inclusion), h5000, hx4700: add support from writing reflash.ctl files containing instruction to flash the rootfs using bootldr Jan 06 23:24:02 koen: especially without my RSA decrypter... Jan 06 23:24:05 Laibsch: faulty .desktop file? Jan 06 23:24:25 zecke: I'll remove that part or add definition Jan 06 23:24:49 hrw: add it please, I fill fix up the rest Jan 06 23:25:01 http://thoughtfix.blogspot.com/2007/01/critical-question-answered-yes-n800-is.html Jan 06 23:25:51 vfp and 128MB ram Jan 06 23:26:14 zecke: what about adding 2000 lines of mysql/acinclude.m4? Jan 06 23:26:50 Some Random V6 Processor rev 2 (v6l);) Jan 06 23:27:11 hrw: ain't that cute? Jan 06 23:27:21 the arm people complained about that a while ago on LAK Jan 06 23:27:26 hrw: no, I'm not in the mood of removing 1996 of them Jan 06 23:27:29 I remember Jan 06 23:27:31 zecke: sure Jan 06 23:28:00 koen: I need to admit I don't remember what they complained about :( Jan 06 23:28:10 koen: I extracted http://ewi546.ewi.utwente.nl/mirror/www.openzaurus.org/official/unstable/3.5.4/feed/opie/xmms-embedded_20040327-r3_arm.ipk and looked at the desktop file: http://rafb.net/p/NQTRrV41.html Looks OK. Jan 06 23:28:36 rmk complained about the new idiotic bitfield and catalin complained about the unprofessional message Jan 06 23:28:50 libxslt is horribly broken... it tries to install to /usr/lib Jan 06 23:29:07 seems it gets passed --prefix=/usr on configure Jan 06 23:29:49 changing it in EXTRA_OECONF does not help as it keeps being overridden. Anybody an idea? Jan 06 23:30:48 mysql.m4 is 101 lines Jan 06 23:31:20 philippe: --prefix=/usr is right?! Jan 06 23:31:25 philippe: --prefix=/usr is correct Jan 06 23:31:34 hrw: okay, okay, send me the acinclude and mysql.m4 :) Jan 06 23:31:36 philippe: you do want to install it in /usr eventually Jan 06 23:31:52 zecke: pushing mysql.m4 Jan 06 23:31:59 okay :) Jan 06 23:32:04 right, but the problem is in do_install it goes to /usr/lib on the host Jan 06 23:32:16 for the plugins? Jan 06 23:32:19 philippe: it should use DESTDIR in 'make install' Jan 06 23:32:23 koen: yep Jan 06 23:32:59 -EXTRA_OECONF = "--without-python --without-debug --without-mem-debug --without-crypto" Jan 06 23:32:59 +EXTRA_OECONF = "--without-python --without-plugins --without-debug --without-mem-debug --without-crypto" Jan 06 23:33:08 lol Jan 06 23:33:23 my 'fix' Jan 06 23:33:25 +EXTRA_OECONF = "--without-everything" Jan 06 23:33:44 :-) okay. That might work Jan 06 23:34:57 maybe I'll fix the gcc .la files tomorrow using XorA|gon1's sed magic Jan 06 23:36:13 hrw: what's the targetdate for 3.5.5? Jan 06 23:36:48 March? Jan 06 23:37:08 zecke: pushed Jan 06 23:37:19 so "fosdem + week" :) Jan 06 23:37:46 * philippe has to likes the use of FOSDME as reference :-) Jan 06 23:38:09 * koen hints philippe to bring a n800 Jan 06 23:38:48 zecke: pushed Jan 06 23:39:11 thanks Jan 06 23:39:25 philippe: my h3600 build halted at enchant, will try to push a fix for that tomorrow Jan 06 23:39:42 bugdays? :D Jan 06 23:39:56 right :) Jan 06 23:40:10 hrw: thx for the devkitidp fixes Jan 06 23:40:23 cbrake: thx for you too Jan 06 23:40:29 koen: I am doing a h3600 build atm.. I will see what happens Jan 06 23:40:38 cbrake: can you drop its u-boot recipe too? Jan 06 23:40:47 hrw: sure, will do now Jan 06 23:40:57 philippe: if you update and restart OE will even write a reflash.ctl for you :) Jan 06 23:43:22 koen: what is reflash.ctl? Jan 06 23:44:35 philippe: the file you need for bootldr Jan 06 23:45:23 CIA-4: ping Jan 06 23:46:06 koen: ??? reflashing a h3600 or 38xx is serial... Jan 06 23:46:15 philippe: nope :) Jan 06 23:46:35 philippe: put the files on cf + reflash.ctl and you even get nice menu :) Jan 06 23:46:44 03hrw 07org.oe.dev * r1fa25c4e... 10/ (1 contrib/site-conf/m4/mysql.m4): site-conf/mysql.m4: filled with some tests from mysql 4.1.18 Jan 06 23:46:48 03cbrake 07org.oe.dev * rb7d0f941... 10/ (1 packages/uboot/uboot-devkitidp-pxa255_cvs.bb): uboot-devkitidp: drop obsolete uboot recipe -- generic 1.1.6 is now used Jan 06 23:46:52 * philippe wishes he had CF now Jan 06 23:47:06 grrrr mtn: multiple update candidates: Jan 06 23:47:06 mtn: 3d9d209dc4aaf9913ae7f9891a5736abb9e2d4a7 koen@openembedded.org 2007-01-06T23:51:17 Jan 06 23:47:09 mtn: e93b0677a2514562872c3ebc886e3ac9642a1b5d hrw@openembedded.org 2007-01-07T00:01:55 Jan 06 23:47:25 philippe: mtn sync ; mtn merge ; mtn sync Jan 06 23:47:36 new nokia tablet: http://www.linuxdevices.com/news/NS9981902594.html Jan 06 23:47:57 cbrake: scroll up 23 minutes :) Jan 06 23:48:20 03hrw 07org.oe.dev * r57cd3499... 10/ (3 files in 2 dirs): file: drop 4.16 Jan 06 23:48:29 koen: 23? Jan 06 23:48:43 cbrake: omap 2420 + 128MB ram Jan 06 23:49:25 hmmm. Management is not going to like this.... Jan 06 23:50:19 philippe: there are even pictures on flickr Jan 06 23:50:28 ~hail compusa Jan 06 23:50:30 * ibot bows down to compusa and chants, "I'M NOT WORTHY!!" Jan 06 23:50:41 koen: cool! Jan 06 23:50:43 115 recipes failed still Jan 06 23:50:55 zecke: I know :-) Jan 06 23:51:36 zecke: btw now that I have you online. I saw you played with gpe-mini-browser a while ago and had a small bug. Mind to tell me more about that? Jan 06 23:52:22 philippe: google.pl does not load in it Jan 06 23:52:43 philippe: There is not much to tell. I think http://www.heise.de as as argv[1] makes it crash, or www.heise.de as argv[1] Jan 06 23:54:00 philippe: gpe-mini-browser lack ssl-manager to view sites which use untrusted ssl certs.. Jan 06 23:54:21 philippe: I have used version 0.19 of it and it crashed on argv[1]. Otherwise it would have worked good enough Jan 06 23:54:45 philippe: at the end I used ruby and GtkMozEmbed as it is available on Ubuntu Dapper Jan 06 23:54:58 hrw: well it loads fine on my desktop. Thw images must be lacking some localisation stuff... Jan 06 23:55:19 zecke: I am sure gtk-webcore crashed on heise as you used the broken tarball release.... Jan 06 23:55:26 philippe: which one? gconv support? Jan 06 23:55:54 hrw: for the ssl koen has found a solution I believe. However I have to find out what he meant with curl -k Jan 06 23:55:54 philippe: well it crashed during parsing the url, not the site itself Jan 06 23:57:52 zecke: weird. It doesn't now and I did not change the code Jan 06 23:58:23 when I read your blog entry I checked it and could not find an error Jan 06 23:58:35 philippe: -k = --insecure == always accept certs Jan 06 23:58:50 philippe: mini-browser doesn't propagate the curl dialog Jan 06 23:58:57 so you end up with a black page Jan 06 23:59:08 philippe: I will give it a try on monday (the code is in the office) Jan 06 23:59:45 koen: I'll have to check out what happens in the gtk-webcore code... Jan 07 00:00:28 * philippe has not found one reported bug that was not gtk-webcore... Jan 07 00:00:54 philippe: did you talk with Mike Emmel? It looks like he started creating a Gtk+ backend for WebKit upstream Jan 07 00:02:23 zecke: not yet. I saw some webkit gdk code but could not figure out how it would be useful Jan 07 00:02:46 philippe: a Makefile or such is missing, but there is some code... Jan 07 00:03:13 03hrw 07org.oe.dev * r8d360ff7... 10/ (4 files in 2 dirs): ruby: dropped older then 1.8.5 Jan 07 00:03:18 03hrw 07org.oe.dev * rfe41ffdd... 10/ (8 files in 3 dirs): zlib: dropped older then 1.2.3 Jan 07 00:05:27 hrw: this is how it looks to me in pl. http://handhelds.org/~philippe/gpe-mini-browser-pl.png Jan 07 00:06:12 * koen suspects a missing local-base-pl Jan 07 00:06:31 philippe: try it on a handheld Jan 07 00:07:03 * philippe is working on that :-) Jan 07 00:07:18 bit first make sure I get gtk-webcore from svn now by default Jan 07 00:07:42 koen: and then locale-base-ch_TW when I will want to see Taiwan website? etc? Jan 07 00:08:30 03hrw 07org.oe.dev * rf416512d... 10/ (1 packages/frodo/frodo_4.1b.bb): frodo: fix SRC_URI Jan 07 00:08:56 hrw: right, I know it sucks Jan 07 00:09:51 useless Jan 07 00:10:11 so we need glibc-locale-all package to get usable GPE browser? Jan 07 00:10:39 hrw: well basically you would need the locales you need... Jan 07 00:11:04 * philippe wants one world computer language character set Jan 07 00:11:13 philippe: utf8 Jan 07 00:11:16 UTF8? Jan 07 00:11:28 koen: just thought about that... Jan 07 00:11:31 hrw: it's a theory about the locales, I solved tinymail crashes with it Jan 07 00:11:35 * hrw want someone who will fix 106 recipes... Jan 07 00:12:07 * JustinP scripts Jan 07 00:12:42 hi, do any linux handheld devices use gcc older than 3.1? Jan 07 00:12:44 3http://pastebin.ca/308834 Jan 07 00:13:00 ssvb: 3.3 is minimal Jan 07 00:13:18 ssvb: depends on what you run on the machines, not on the machines itself :) Jan 07 00:13:26 ssvb: and we ignore fact that some zaurus people use antique 2.95 Jan 07 00:14:32 ssvb: iirc kernel req 3.3 or 3.2 as minimum Jan 07 00:14:40 hrw: http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/2007-January/050547.html Jan 07 00:14:59 so this ifdef can be safely dropped? Jan 07 00:15:47 I would left it Jan 07 00:16:15 03dp 07org.oe.dev * rad31e71f... 10/ (4 files in 3 dirs): Jan 07 00:16:15 enca 1.9: Fix build. Jan 07 00:16:15 * These patches should fix build on both 32 and 64 bit systems. Jan 07 00:16:15 * Closes #1129, #1461. Jan 07 00:16:24 sharprom lusers use 2.95 and build mplayer from time to time Jan 07 00:16:46 ok, thanks Jan 07 00:23:01 03hrw 07org.oe.dev * r7dbeb5e3... 10/ (1 packages/ruby/ruby_1.8.5.bb): ruby: build more extensions (zlib, curses, openssl) - close #1725, #1699 Jan 07 00:24:12 philippe: are you? Jan 07 00:25:01 philippe: i built anstrom for h3800 now i sending it over com, maybe it'll boot. Jan 07 00:26:20 !oebug 1730 Jan 07 00:26:21 * * Bug 1730, Status: NEW, Created: 2006-12-30 16:55 Jan 07 00:26:21 koen: ? Jan 07 00:26:21 * * likewise(AT)gmx.net: add AC97 support for EFIKA - experimental Jan 07 00:26:22 * * http://bugs.openembedded.org/show_bug.cgi?id=1730 Jan 07 00:26:49 ah, I thought likewise had already pushed that Jan 07 00:26:51 03GoxboxLive 07org.oe.dev * rdf828882... 10/ (4 files in 4 dirs): keymaps: added HTC Universal keymap - close #1731 Jan 07 00:27:03 hrw: that's good yo go Jan 07 00:28:25 hrw, koen: yes, pushed that already. double-checking as I am testing it right now on the target. seems to work fine. Jan 07 00:28:45 hrw, koen: I wanted to close the bug when the test was ok. Jan 07 00:29:00 ok Jan 07 00:29:38 btw - does ipkg .163 better then .159? worth upgrade? Jan 07 00:30:20 hi univac Jan 07 00:30:26 Hi Jan 07 00:30:42 philippe: maybe you know why oe building 2.6 kernel for h3800? :) Jan 07 00:30:47 hrw: less segfaults, more bugs Jan 07 00:30:50 univac: I hope it boots :-) Jan 07 00:30:58 koen: common problem.. Jan 07 00:31:04 philippe: noo, i tried Jan 07 00:31:38 i built angstrom but unpacked this, changed kernel to 2.4.19. remove udev and change to hotplug Jan 07 00:31:51 univac: it builds for 3600, 3800 needs some reconfiguration and is slightly broken. My Ipaq is charging so I can reflash it safely now Jan 07 00:31:57 hrw: I choose .163 for angstrom to get rid of segfaults Jan 07 00:32:07 univac: I am talking about 2.6 here Jan 07 00:32:10 philippe: now i sending my hacked jffs2 :) Jan 07 00:32:12 thx who ever goxboxlive Jan 07 00:32:15 is Jan 07 00:32:24 philippe: it boots on h3800? on h3600 yes, i know Jan 07 00:32:54 philippe: could you paste defconfig of h3800? Jan 07 00:34:21 univac: you also need to change some kernel files. Once I know it boots I will put the patches in OE. Jan 07 00:34:54 Ok, i'm waiting, if you will put it, blame me Jan 07 00:35:12 you talking about 2.6.19-rc6? Jan 07 00:38:01 univac: univac Jan 07 00:38:10 yep? Jan 07 00:38:25 03pfalcon 07org.oe.dev * r6fde327b... 10/ (1 packages/fbreader/fbreader_0.7.3d.bb): fbreader 0.7.3d: Add bzip2 to DEPENDS. Jan 07 00:38:56 fbreader 0.7.3d is UNFETCHABLE Jan 07 00:39:12 univac: 2.6.16-hh8 Jan 07 00:39:29 ah Jan 07 00:41:25 hrw, fetchable for me Jan 07 00:41:43 posokolvsky: then I will have to check it again Jan 07 00:42:07 ah.. right.. from obsolete/ dir... Jan 07 00:47:45 03hrw 07org.oe.dev * r81518d02... 10/ (1 packages/fbreader/fbreader_0.7.3b.bb): fbreader: drop 0.7.3b Jan 07 00:55:37 03philippe 07org.oe.dev * r19957664... 10/ (1 packages/gtk-webcore/osb-nrcit_svn.bb): Jan 07 00:55:37 packages/gtk-webcore/osb-nrcit_svn.bb Jan 07 00:55:37 * added nrcit svn build recipe Jan 07 00:58:53 Jan 7 02:27:02 localhost dhclient: bound to 192.168.1.100 -- renewal in 44 seconds. Jan 07 00:58:55 argh Jan 07 01:00:17 bound to 192.168.1.100 -- renewal in 20415 seconds. Jan 07 01:00:20 much better Jan 07 01:03:01 ok. time to go sleep Jan 07 01:03:05 02:31 here Jan 07 01:03:55 celf machine is building angstrom task-base for all eabi arm machines, my machine builds opie-image/oz354x and gpe+xorg server for progear Jan 07 01:04:21 philippe: nah, modules don't work Jan 07 01:04:30 zecke: #kdevelop they are talking about your findbug thingy Jan 07 01:04:34 night hrw Jan 07 01:04:36 FATAL: Error inserting ipaq_hal (/lib/modules/2.4.19-rmk6-pxa1-hh37/kernel/arch/arm/mach-ipaq/ipaq_hal.o): Unknown symbol in mo) Jan 07 01:04:45 univac: you're on 2.4? Jan 07 01:04:49 chouimat|Zzzz: all varpoware Jan 07 01:05:06 zecke: :) Jan 07 01:05:16 philippe: i built angstrom with 2.6 and change kernel to old, but modules don't work grrrr Jan 07 01:05:20 dunno why Jan 07 01:06:07 insmod: error inserting 'ipaq_hal.o': -1 Unknown symbol in module Jan 07 01:06:08 hm Jan 07 01:08:47 nah Jan 07 01:08:50 depmod is too new Jan 07 01:10:17 03likewise 07org.oe.dev * r7f70b7fa... 10/ (4 files in 3 dirs): mp3blaster: Added version 3.2.3. Now hosted on sourceforge. Jan 07 01:10:55 univac : 2.6 modules with 2.4 wont work... Jan 07 01:32:44 it's 2.4.19 modules of course Jan 07 01:32:49 depmod is too new i think Jan 07 01:35:06 univac: ok. could be too.... Jan 07 01:35:28 * philippe has 2.6 booting, but something seems to be wrong with the framebuffer Jan 07 01:35:32 i waiting for 2.6 :) Jan 07 01:35:37 touschreen works?! Jan 07 01:42:12 03pfalcon 07org.oe.dev * r906e7dac... 10/ (1 conf/machine/htchimalaya.conf): Jan 07 01:42:12 htchimalaya: New machine, an HTC phone device. Jan 07 01:42:12 * Initial version was in #540, but this one is based on the current template. Jan 07 02:09:30 depmod: QM_MODULES: Function not implemented Jan 07 02:09:36 yyy Jan 07 02:17:19 night everyone Jan 07 02:19:53 grrr something removes include/X11/Xlib.h Jan 07 02:34:03 03justinp 07org.oe.oz354x * rd51c9900... 10/ (1 packages/glibc/glibc_2.3.5+cvs20050627.bb): glibc_2.3.5: remove broken te_IN locale (bug #1760) Jan 07 02:53:40 03justinp 07org.oe.oz354x * r87c2d837... 10/ (10 files in 8 dirs): glibc-package.bbclass: fix qemu -L parameter (bug #1756) **** ENDING LOGGING AT Sun Jan 07 02:59:58 2007