**** BEGIN LOGGING AT Thu Sep 06 02:59:56 2007 Sep 06 03:40:33 OE build problems again... bb.fetch.FetchErrors now. Sep 06 03:41:22 "Could not resolve hostname 'svn.openmoko.org': Temporary failure in name resolution (htp://svn.openmoko.org)" Sep 06 03:41:56 Of course, manually resolving said name works just fine... Sep 06 03:46:10 BTW, given this sort of misbehavior, is it not true that it is now impossible to build with OE offline? Sep 06 04:05:39 mwester-road: which negates one of the advantages of using monotone ... ;-) Sep 06 04:12:57 * * OE Bug 2932 has been created by autobuild(AT)openembedded.org Sep 06 04:12:59 * * libnotify-0.4.4-r0-do_compile Sep 06 04:13:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2932 Sep 06 04:20:57 * * OE Bug 2933 has been created by autobuild(AT)openembedded.org Sep 06 04:20:59 * * gpe-bluetooth-0.55-r0-do_compile Sep 06 04:21:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2933 Sep 06 05:12:36 ggod morning all Sep 06 05:16:44 mwester-road: http://bugs.openembedded.org/show_bug.cgi?id=2897 Sep 06 05:44:24 good morning all Sep 06 05:46:01 Angstrom-x11-pimlico-image-uclibc-ipk-2007.9-test-20070906-ep93xx.rootfs.tar.gz Sep 06 06:16:57 03ifaistos 07org.oe.dev * r28d859e7... 10/ (4 files in 3 dirs): conf/machine/i586-generic.conf : Rename the included tune file from .conf to .inc Sep 06 06:17:02 03ifaistos 07org.oe.dev * r67a0d6fb... 10/ (4 files in 3 dirs): conf/machine/i686-generic.conf : Rename the included tune file from .conf to .inc Sep 06 06:17:09 03ifaistos 07org.oe.dev * re0ffb9b2... 10/ (1 site/ix86-common): site/ix86-common : Add a def for libpcap Sep 06 06:17:19 03koen 07org.oe.dev * r22435246... 10/ (1 conf/distro/angstrom-2008.1.conf): angstrom-2008.1: make binutils overridable Sep 06 06:17:24 03koen 07org.oe.dev * r42c7569b... 10/ (1 conf/distro/include/sane-srcrevs.inc): sane-srcrevs.inc: start collecting SRCREVs to compat parse-errors due to e.g. berlios being down Sep 06 06:18:28 koen: heheh Sep 06 06:20:54 hvontres|home: the openmoko people will throw a fit is angstrom starts to use it by default :( Sep 06 06:24:25 koen: why? they don't have to.... Sep 06 06:24:49 hvontres|home: the include angstrom-2007.1 Sep 06 06:25:00 koen: ahhh..... Sep 06 06:26:10 koen: thanks for pointing me at the bug btw... I was stating to think I had busted something on my setup. I recently got a nice new HP dv2000 notebook so I can build stuff on the way home from work.. Sep 06 06:26:36 koen: but that doesn't work too well if you have to be on the NET just to parse stuff :( Sep 06 06:27:10 * hvontres|home likes his new dual-core turion.... so much faster than the old PII-333's Sep 06 06:28:19 koen: hello Sep 06 06:28:22 n8 all Sep 06 06:28:32 hey khem Sep 06 06:28:36 hvontres|home: 'night Sep 06 06:28:38 koen: I have tried to build angstrom-console-image Sep 06 06:28:50 but I think libtool is biting me Sep 06 06:29:05 build is failing in binutils Sep 06 06:29:30 because it is tryin to use 'flex -lm' instead of 'flex' Sep 06 06:29:34 !oebug 2928 Sep 06 06:29:35 * * Bug 2928, Status: NEW, Created: 2007-09-05 08:32 Sep 06 06:29:36 * * philip(AT)balister.org: binutils-cross fails to build on x86_64 Sep 06 06:29:37 * * http://bugs.openembedded.org/show_bug.cgi?id=2928 Sep 06 06:30:30 koen: when was libtool-1.5 made compulsory on OE because OE was building for me last week Sep 06 06:31:53 it should use 1.5.10 Sep 06 06:31:56 has some one tried angstrom-console-image lately on Fedora 7 Sep 06 06:33:06 koen: when I see the generated ld/libtool file it says Generated automatically by ltconfig (GNU libtool 1.4a-GCC3.0 (1.641.2.256 2001/05/28 20:09:07 with GCC-local changes)) Sep 06 06:34:24 koen: is it possible to chose older libtool just to verify Sep 06 06:39:57 * * OE Bug 2888 has been RESOLVED (FIXED) by Sep 06 06:39:59 * *  xserver-kdrive-imageon-1.3.0.0-r0-do_patch Sep 06 06:40:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2888 Sep 06 06:40:09 * * OE Bug 2880 has been RESOLVED (FIXED) by Sep 06 06:40:11 * *  linux-handhelds-2.6-2.6.21-hh12-r2-do_compile Sep 06 06:40:13 * * http://bugs.openembedded.org/show_bug.cgi?id=2880 Sep 06 06:40:21 koen: 2.18 builds ok Sep 06 06:40:54 koen: 2930 can you have a look at the patch I posted Sep 06 06:43:57 * * OE Bug 2934 has been created by autobuild(AT)openembedded.org Sep 06 06:43:59 * * gtk+-2.10.14-r7-do_configure Sep 06 06:44:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2934 Sep 06 06:53:53 khem: didn't I fix that 20 minutes ago? Sep 06 06:54:29 koen: I did not update yet Sep 06 06:55:58 revision 22435246c6164e8d9181239f72a4b3ddae29e35b Sep 06 06:56:30 you mean fix for 2930 right ? Sep 06 06:56:45 yes Sep 06 06:57:26 ok Sep 06 06:57:37 then you can close it Sep 06 07:09:57 * * OE Bug 2930 has been RESOLVED (FIXED) by Sep 06 07:09:59 * *  angstrom-2008.1 overwrites the PREFERRED_VERSION for binutils Sep 06 07:10:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2930 Sep 06 07:24:24 03koen 07org.oe.dev * rce91420f... 10/ (1 packages/pango/pango.inc): pango.inc: also depend on virtual/libiconv, should fix #2934 Sep 06 07:28:58 morning Sep 06 07:48:01 hey hrw Sep 06 07:54:09 morning koen, please can you merge that ts72xx support when you have some time? thanks Sep 06 07:54:27 what's the bugnumber for that again? Sep 06 07:55:05 [oe-issues] [Bug 2918] New: Support for Technologic Systems TS-72xx SBC Sep 06 07:56:28 just started to wonder if ts72xx is right machine name, since there's ts7400 which is almost same as ts72xx Sep 06 07:57:13 don't know if it have different machine ID in kernel also, since I don't own that board Sep 06 07:58:04 maybe we can add ts74xx machine which will just include ts72xx once it's confirmed working by some ts7400 owner Sep 06 07:59:58 hm i might lookup that machine ID in TS's 2.4 kernel Sep 06 08:05:57 bluez-gnome fails Sep 06 08:08:56 morning Sep 06 08:10:10 hey XorA Sep 06 08:21:18 XorA: did you find oedem hotel finally? Sep 06 08:22:06 hrw: the Restaurant Hotel is much cheaper than the Garden Hotel Sep 06 08:22:26 how muc for single? Sep 06 08:22:41 59E for single Sep 06 08:22:57 * * OE Bug 2918 has been RESOLVED (FIXED) by Sep 06 08:22:59 * *  Support for Technologic Systems TS-72xx SBC Sep 06 08:23:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2918 Sep 06 08:23:27 59€ yeah I found the key Sep 06 08:23:49 lucky you - Polish keymap lack euro sign Sep 06 08:24:03 hrw: my keyboard lacks Euro symbol Sep 06 08:24:17 XorA: you look at keys when you type? Sep 06 08:24:34 hrw: yes I never learned touch typing Sep 06 08:25:08 ouc.. Sep 06 08:25:21 hrw: double in Restaurant Hotel is 110€ Sep 06 08:25:44 hrw: proably too much typing on zx81 keyboard ruined for life my typing Sep 06 08:26:33 XorA: especially in basic? Sep 06 08:26:54 10 PRINT "I am god" Sep 06 08:26:57 20 GOTO 10 Sep 06 08:27:27 I prefer it as one linr Sep 06 08:27:35 hrw: cant on zx81 Sep 06 08:27:45 argh Sep 06 08:27:51 hrw: it didnt support multi statement lines, that only came in in Spectrum Sep 06 08:28:42 * XorA suddenly has horrific memories of COMAL Sep 06 08:29:39 XorA: you forgot the space after god. Sep 06 08:29:56 and a semicolon! Sep 06 08:30:01 or was that BBC basic Sep 06 08:30:25 Spacemonkey: I think you think BBC, but the ; was only needed for multiple statments per line Sep 06 08:31:13 * koen looks at http://tinyurl.com/ynmwfh and cries Sep 06 08:31:45 XorA: the multiple statements per line was the most satisfying, that way you used to get a scrolling effect! Sep 06 08:32:07 koen, thanks Sep 06 08:33:52 03ynezz 07org.oe.dev * ra0c2c46e... 10/ (1 conf/machine/ts72xx.conf): ts72xx: add machine description for ts72xx based devices Sep 06 08:33:58 Spacemonkey: great thing was procedured basic Sep 06 08:33:59 03ynezz 07org.oe.dev * r92277d43... 10/ (35 files in 8 dirs): linux: add ts72xx support to 2.6.22.6 and 2.6.22+2.6.23-rc5 (see version policy for the renaming reason), closes #2918 Sep 06 08:34:13 on $var exec prc1,proc2,proc3,proc4 Sep 06 08:34:26 similiar to "on $var goto 100, 1000,2342,324; Sep 06 08:34:32 s/;/: Sep 06 08:34:51 hrw: thats as cryptic as C :) Sep 06 08:35:47 Was anyone here at LinuxConf? Sep 06 08:37:21 10 pr. "hello world":g. 10 Sep 06 08:39:41 hrw: dont get me started on hello world! I am sure the example in Kernighan and Ritchie is incorrect! Sep 06 08:59:57 * * OE Bug 2935 has been created by autobuild(AT)openembedded.org Sep 06 08:59:59 * * libmokojournal2-0.1.0+svnr2780-r2-do_compile Sep 06 09:00:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2935 Sep 06 09:11:57 * * OE Bug 2936 has been created by autobuild(AT)openembedded.org Sep 06 09:11:59 * * libiconv-1.11-r4-do_configure Sep 06 09:12:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2936 Sep 06 09:13:57 * * OE Bug 2936 has been RESOLVED (INVALID) by Sep 06 09:13:59 * *  libiconv-1.11-r4-do_configure Sep 06 09:14:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2936 Sep 06 09:44:08 / work around for annoying things. Sep 06 09:44:14 what a useful comment :-( Sep 06 09:44:37 where? Sep 06 09:44:42 mozilla source Sep 06 09:44:47 heh Sep 06 09:44:54 webkit! Sep 06 09:46:02 ah mozilla actually ignores Alt TAB Sep 06 09:46:10 WTF are they dealing with raw keys :-( Sep 06 09:58:09 XorA: firefox is still busted on eabi? Sep 06 10:00:59 koen: yes Sep 06 10:01:20 koen: and I dont have a powerful enough machine to debug, maybe when GTA02 is released Sep 06 10:02:43 ahah, found the mozilla keyboard problem Sep 06 10:03:56 not sure there is a clean fix as Mozilla is technically right and handhelds stuff is wrong Sep 06 10:04:58 file a bug against the handhelds stuff :) Sep 06 10:05:07 bug(tm) Sep 06 10:05:12 stuff(tm) Sep 06 10:05:16 koen: I mean handhelds as in general handheld software, not .org Sep 06 10:05:24 Mod4 is META Sep 06 10:05:35 and this is true for most desktop software Sep 06 10:05:40 we use Mod4 to mean Fn Sep 06 10:05:58 hence why it works when we change to mod5 Sep 06 10:06:24 my next commit should make you feel a bit better :) Sep 06 10:06:30 moin Sep 06 10:06:36 hey zecke Sep 06 10:06:36 although keylaunch is the only app that needs changed AFAIK Sep 06 10:06:48 keylaunch needs dying Sep 06 10:06:53 03koen 07org.oe.dev * r89b1d1f0... 10/ (8 files in 3 dirs): midori: update to 0.6, add .desktop entry, remove obsolete versions Sep 06 10:07:07 YAY Sep 06 10:08:00 * koen wonders if epiphany webkit builds yet Sep 06 10:12:15 http://blogs.gnome.org/epiphany/2007/07/30/epiphany-2196-released-webkit-back-end/ Sep 06 10:13:31 ~slay keylaunch Sep 06 10:13:32 * ibot brandishes Excalibur! "With this sword, I vanquish thee, keylaunch!" apt lops off keylaunch's head. Sep 06 10:13:42 heh. Sep 06 10:14:02 hey everyone. Sep 06 10:14:57 so I do not need to find a clean way to modify keylaunch so it comply with this (Fn = mod5) Sep 06 10:15:05 if keylaunch is about to die. Sep 06 10:24:22 * koen can't find the epiphany recipes XorA gave him Sep 06 10:24:42 thats coz I didnt give you any Sep 06 10:24:54 * koen bites tongue to not respond to avr32 mail Sep 06 10:26:15 what is avr32 anyway ? Sep 06 10:26:41 oh, I see. Sep 06 10:29:46 XorA, the brightness bug is not fixed in 2.6.22 for me. Sep 06 10:30:11 (I did not re-flash the whole system, only the kernel and ipkg upgrade) Sep 06 10:31:08 koen: can I ask you a quick question about the kb9202_defconfig you and hrw defined a while back? Sep 06 10:31:19 nodens: odd, I saw it until I reflashed, but there shouldnt be any diference Sep 06 10:31:23 russf: sure Sep 06 10:31:46 XorA, I'll try a full reflash? Sep 06 10:31:58 -? Sep 06 10:31:58 nodens: not unless you really want Sep 06 10:32:04 koen: thanks. I need to add MTD support, similar to the at91*_defconfigs. How should I go about that? Sep 06 10:32:21 XorA, are you set to auto-dim in your brightness settings ? Sep 06 10:32:21 * XorA knows its a pain to re-create all those settings Sep 06 10:32:45 I can always backup /etc and /home :) Sep 06 10:32:50 add it, copy your defconfig over the one in OE, 'mtn diff /to/defconfig > diff.txt', attach diff.txt to bugzilla Sep 06 10:33:10 s/add/edit/ Sep 06 10:33:15 koen: cool. thanks! Sep 06 10:33:56 nodens: yes Sep 06 10:34:14 nodens: just tried it on mains and battery with autodim on Sep 06 10:35:43 mwester: Your fakeroot patch in OE does not seem to work for me... "LT_INIT" does not seem to be defined here. Sep 06 10:38:14 XorA, I'll try to reflash then... or maybe with altboot first Sep 06 10:47:48 03koen 07org.oe.dev * rbd4ed15a... 10/ (1 packages/openmoko2/openmoko-dates2_svn.bb): openmoko-dates2: merge from poky Sep 06 10:47:55 03koen 07org.oe.dev * r35021294... 10/ (16 files in 5 dirs): pimlico: sync with poky Sep 06 11:00:44 mwester: I take that back, I need the patch for both fakeroot and fakeroot-native. I'll push that change... Sep 06 11:01:09 hi, i get a strange FATAL: oe_runmake failed Sep 06 11:01:18 error when compiling glib for MACHINE=native Sep 06 11:01:25 any1 goit an idea? Sep 06 11:01:30 which glib? Sep 06 11:01:52 glib-2.0-native-2.12.4-r1 Sep 06 11:02:22 pastebin log? give distro/host? Sep 06 11:06:48 hrw: http://www.pastebin.org/1841 Sep 06 11:09:25 that libtool is missing the 3figures patch Sep 06 11:11:01 koen: midori just needs the download pr0n button Sep 06 11:14:20 ok seems like that someone removed the 3figures patch from OE 0.35 Sep 06 11:14:26 XorA: haha Sep 06 11:15:08 XorA: my first browser mockup had a gender division. male: just one button to download porn, in the 2nd iteration this button was removed and directly started downloading... Sep 06 11:15:17 but I have given up on the gender separation Sep 06 11:15:57 nik0n: what is 'OE 0.35'? Sep 06 11:16:45 XorA: do you have the epophany recipes still around? Sep 06 11:17:00 open embedded 0.35 Sep 06 11:17:13 there is no OE 0.35 Sep 06 11:18:39 sorry, i mean OE-this-is-for-mtn-0.35.mtn.bz206-Sep-2007 Sep 06 11:19:19 i used another version of libtool now with a 3figures patch. works now, thx Sep 06 11:20:49 nik0n: 0.35 is monotone version Sep 06 11:21:54 03florian 07org.oe.dev * r471f6e7d... 10/ (1 packages/gpephone/gpe-applauncher_0.10.bb): gpe-applauncher: Add 0.10 Sep 06 11:21:59 03florian 07org.oe.dev * r0ffedb7c... 10/ (1 packages/fakeroot/fakeroot_1.7.1.bb): fakeroot: Apply patch for fakeroot-native to fakeroot too. Sep 06 11:23:57 * * OE Bug 2937 has been created by autobuild(AT)openembedded.org Sep 06 11:23:59 * * gnome-desktop-2.18.3-r2-do_configure Sep 06 11:24:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2937 Sep 06 11:28:55 koen: Thanks for the pointer to the bug. I didn't think I had strong feelings on the entire svn/srcdate stuff, but I suddenly find that I'm developing an opinion... Sep 06 11:28:57 * * OE Bug 2937 has been RESOLVED (FIXED) by Sep 06 11:28:59 * *  gnome-desktop-2.18.3-r2-do_configure Sep 06 11:29:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2937 Sep 06 11:29:26 It seems a sad thing to me that something such as parsing bb files is now dependent upon the vaguaries of the internet... Sep 06 11:29:56 mwester-road: I'd like a way that makes those errors show up in OE (e.g. as a do_fetch error) instead of a parse error Sep 06 11:30:57 * * OE Bug 2938 has been created by autobuild(AT)openembedded.org Sep 06 11:30:59 * * libwnck-2.13.5-r0-do_package_write_deb Sep 06 11:31:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2938 Sep 06 11:32:19 is it just me, or is the OW site very slow today? Sep 06 11:32:30 s/OW/OE Sep 06 11:32:43 * hrw now has broadcom based router, bcm4306 in laptop, bcm4318 in alix Sep 06 11:33:11 russf: Its pretty fast for me... Sep 06 11:33:29 florian: ok. thx Sep 06 11:33:40 * russf looks elsewhere for the problem Sep 06 11:34:34 russf: I just checked the server... its more or less idle. Sep 06 11:35:44 florian: traceroute is showing a problem after or in q9.net Sep 06 11:37:23 koen: no, they never worked at all so are long gone Sep 06 11:38:57 * * OE Bug 2939 has been created by autobuild(AT)openembedded.org Sep 06 11:38:59 * * libgpewidget-0.115-r0-do_compile Sep 06 11:39:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2939 Sep 06 11:56:31 hi there... im searching for florian .. about commit 0ffedb7c7f00db7f23e28aec3cbb44abdd01876f Sep 06 11:58:19 03koen 07org.oe.dev * r4f51dba4... 10/ (1 packages/gnome/gnome-desktop_2.18.3.bb): gnome-desktop: update to fix issues with newer gnome-common, closes #2937 Sep 06 11:58:57 it seems work-with-older-libtool-patch gets applied (or at least tried to do so) in fakeroot-native now.. so i had to drop the patch from the reciepe there. Sep 06 12:03:59 03koen 07org.oe.dev * r4f51dba4... 10/ (1 packages/gnome/gnome-desktop_2.18.3.bb): gnome-desktop: update to fix issues with newer gnome-common, closes #2937 Sep 06 12:04:59 03koen 07org.oe.dev * r4f51dba4... 10/ (1 packages/gnome/gnome-desktop_2.18.3.bb): gnome-desktop: update to fix issues with newer gnome-common, closes #2937 Sep 06 12:20:05 IOError: [Errno 2] No such file or directory: '/home/gg/zaurus/build-neo/tmp/work/armv4t-angstrom-linux-gnueabi/xf86dgaproto-1_2.0.2-r1/install/xf86dgaproto-dbg.lock' Sep 06 12:20:10 Im confused Sep 06 12:20:49 i got such errors also Sep 06 12:21:07 rebuilding the package manualy fixed the problem Sep 06 12:21:20 http://rafb.net/p/I9lyjw25.html Sep 06 12:21:51 XorA i got those all the time. basically you need to rebuild everything due to some ipkg-poackaging change introduced in 209ccafa5a29dc4d6ec16bbb89bfd465904ce377 Sep 06 12:22:11 ynezz thats what i did last night.. don't .. will make you crazy Sep 06 12:22:15 oh damnation, I forgot this build tree is older than that Sep 06 12:22:28 ~lart me for general stupidity Sep 06 12:22:28 * ibot hauls xora up by the scruff of the neck and spanks him until he waddles for general stupidity Sep 06 12:22:32 so i have to rebuild from scratch right? Sep 06 12:22:47 ynezz: you can move the files to the right location Sep 06 12:23:25 it won't happen again right? Sep 06 12:24:02 i've seen similar questions from others here also, so it might be FAQ :) Sep 06 12:24:34 and on #openmoko also of course Sep 06 12:27:47 ynezz: org.oe.dev is development tree Sep 06 12:28:04 so many things happen - most of them are announced on ML Sep 06 12:28:57 * * OE Bug 2940 has been created by autobuild(AT)openembedded.org Sep 06 12:28:59 * * fakeroot-native-1.7.1-r2-do_patch Sep 06 12:29:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2940 Sep 06 12:29:46 03xora 07org.oe.dev * rb7169a80... 10/ (1 packages/fakeroot/fakeroot-native_1.7.1.bb): fakeroot/fakeroot-native_1.7.1.bb : dont apply same patch twice Sep 06 12:29:53 florian: http://bugs.openembedded.org/show_bug.cgi?id=2940 Sep 06 12:30:08 hrw: ok Sep 06 12:31:57 * * OE Bug 2940 has been RESOLVED (FIXED) by Sep 06 12:31:59 * *  fakeroot-native-1.7.1-r2-do_patch Sep 06 12:32:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2940 Sep 06 12:38:19 koen: ah sorry, I didn't see that the fakeroot-native bb does it in this way.# Sep 06 12:38:42 btw. anyone here who tried OE for a mips bigendian target? Sep 06 12:44:01 argh... another business card of someone who wants to buy my car Sep 06 12:44:06 hrw: what's recommended then? there's no org.oe.stable, so I have just to stop pulling if I'm ok with that version, rename that dir stable and make a new one as testing ? Sep 06 12:46:33 ynezz: follow .dev I would say or use one ver which works for you Sep 06 12:46:43 morning Sep 06 12:47:31 florian: is it more rsut than car? Sep 06 12:48:31 XorA: ah no, its still 100% car :-) Sep 06 12:52:17 how would you handle this scenario: i'm using package umbaumba_1.0-r0, built image using it, distributed image, pull, pull(some fixes added -r1) ... weeks passed ... pull(some fixes added -r2), now I've received some bugreports but my working image is using umbaumba-r2 and I've to built umbaumba-r0 to find and kill the bugs Sep 06 12:53:43 but other packages umbaumba depends on could've changed also, so I might not be able to reproduce some of the bugs anymore Sep 06 12:53:56 WORKSFORME :-) Sep 06 12:54:13 the only solution is to have some frozen repo right? Sep 06 12:54:43 if -r2 can't reproduce the bugs, tell your users to use -r2 Sep 06 12:54:56 yes sure Sep 06 12:56:57 * * OE Bug 2938 has been RESOLVED (FIXED) by Sep 06 12:56:59 * *  libwnck-2.13.5-r0-do_package_write_deb Sep 06 12:57:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2938 Sep 06 12:58:02 hmm, next toolchain problem: Sep 06 12:58:04 | /space/fic/openmoko-oe.dev/fic-gta01/tmp/work/i686-armv4t-sdk-angstrom-linux-gnueabi/meta-toolchain-1.0-r0/temp/run.do_populate_sdk.32458: line 559: cd: /space/fic/openmoko-oe.dev/fic-gta01/tmp/work/i686-armv4t-sdk-angstrom-linux-gnueabi/meta-toolchain-1.0-r0/sdk/image//usr/local/openmoko/arm/arm-angstrom-linux-gnueabi/lib/pkgconfig: No such file or directory Sep 06 12:59:00 guess we should check whether this directory exists before attempting to fix the pkgconfig files Sep 06 12:59:06 * mickey|zzZZzz adds chec Sep 06 12:59:28 mickey|zzZZzz: moring Sep 06 13:00:01 mickey|zzZZzz: /lib/pkgconfig looks awfully hardcoded to me Sep 06 13:00:02 03koen 07org.oe.dev * r7a785952... 10/ (1 packages/gnome/libwnck_2.13.5.bb): libwnck: remove empty homepage field, fixes #2938 Sep 06 13:00:18 (just like all the /lib and /bin entries in rootfs*.bbclass) Sep 06 13:00:23 koen: ya, that as well Sep 06 13:01:53 mickeyl: lots of python stuff in http://tinyurl.com/ynmwfh Sep 06 13:02:43 koen: ooh, nasty :/ Sep 06 13:03:02 dunno how i have time to look at it though Sep 06 13:03:10 hopefully on weekend Sep 06 13:03:31 NOTE: package gnome-panel-2.18.1-r1: task do_build: completed Sep 06 13:06:06 * chouimat is sad pavarotti is dead Sep 06 13:07:29 mickeyl: openmoko-dates2 is in OE now Sep 06 13:07:52 koen: nice. tried it yet? Sep 06 13:08:03 nope Sep 06 13:08:15 both my OE machines run an autobuilder nowadays Sep 06 13:08:49 sakoman, ping Sep 06 13:09:05 03koen 07org.oe.dev * r5cca19f6... 10/ (1 packages/gnome/gnome-panel_2.18.1.bb): gnome-desktop: fix packaging Sep 06 13:09:05 one with ANGSTROM_PKG_FORMAT = "deb" and one with ANGSTROM_PKG_FORMAT = "ipkg" Sep 06 13:09:15 koen scary Sep 06 13:10:41 mickeyl: I wrote that recipe but did not tested it on phone yet Sep 06 13:12:08 btw: e-mails suxx Sep 06 13:12:55 pop3 fetch all to one box... imap is slow as fetch each time... disconnected/offline imap is easy to get out-of-sync when account used on more then one machine Sep 06 13:12:57 * * OE Bug 2941 has been created by autobuild(AT)openembedded.org Sep 06 13:12:59 * * gnome-panel-2.18.1-r1-do_package Sep 06 13:13:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2941 Sep 06 13:13:34 hrw: I have no sync problems using offlineimap program Sep 06 13:13:49 XorA: offlineimap require me to have imapd Sep 06 13:14:37 hrw: but if you fetching from IMAP you have an imapd Sep 06 13:15:05 XorA: my mail client fetch from imap Sep 06 13:15:08 or is this because Evolution junk doesnt support folders well :-) Sep 06 13:15:19 crofton: pong Sep 06 13:15:23 offlineimap force me to ave imapd on local machine Sep 06 13:15:55 need to refalsh uboot, have you done this? Sep 06 13:15:56 hrw: I have no imapd on local machine, just a mail client that understand folders Sep 06 13:16:51 claws? Sep 06 13:16:57 Crofton: yes Sep 06 13:17:04 mutt :-D Sep 06 13:17:10 but claws does as well :-) Sep 06 13:17:10 I have ckermit running and talking to gumstix Sep 06 13:17:23 hi Sep 06 13:17:25 florian: Hi, while trying to compile vlc-gpe I fall into this error : cc1: internal compiler error while compiling modules/gui/pda/pda.c Sep 06 13:17:26 and I made a .kermrc with a bunch of stuff in it froma web page Sep 06 13:17:28 OK, that's a good first step! Sep 06 13:17:31 heh Sep 06 13:17:34 * XorA dismisses Evolution as being not worth the mouse click Sep 06 13:17:39 I think I am to the scary part ... Sep 06 13:17:56 Wat procedure will you be using? Sep 06 13:17:59 florian: as I saw you work on it recently, I wonder if you manage to compile it... on wich machine ? Sep 06 13:18:12 http://docwiki.gumstix.com/U-Boot#Loading_new_stuff_to_flash Sep 06 13:18:18 * rwhitby uses IMAP on Thunderbird on laptop, and Chatter on Treo650, and offline works perfectly in both cases. Sep 06 13:18:37 (oh, and web interface sometimes too. www.fastmail.com.au) Sep 06 13:18:55 never had a folder get out of sync Sep 06 13:19:10 Yes, I used the procedure starting at the line "or by serial (kermit)" Sep 06 13:19:32 It's all over relatively quickly :-) Sep 06 13:19:48 why am I having issues escaping from kernit ... Sep 06 13:20:06 ctl \ c Sep 06 13:20:13 ah Sep 06 13:20:21 * Crofton hates kermit Sep 06 13:20:34 You'll grow to love it over time :-) Sep 06 13:20:48 why kermit? Sep 06 13:20:51 noglitch: sounds nasty... Sep 06 13:21:05 woglinde, it is the only thing that do do serail downloads well Sep 06 13:21:18 noclouds: I built it for my Palm TX without too much of trouble - for which device did you try? Sep 06 13:21:21 crofton try cutecom Sep 06 13:21:46 rwhitby: I tried thunderbird, but its devs are sick. to have folder sorted by date AND threaded I need to use menu with eac folder Sep 06 13:21:47 I am hoping this os over with Sep 06 13:22:07 crofton only cavete with cutecom is you can use vi in proper way Sep 06 13:22:15 caveate Sep 06 13:22:17 hms Sep 06 13:22:40 Crofton: minicom is usable, picocom is usable but stupid Sep 06 13:22:52 hrw: agreed. But it is an existence proof that IMAP offline does work perfectly if the clients and server handle imap correctly. Sep 06 13:23:02 minicom is stupid in some ways Sep 06 13:23:22 So don' let me hear you dissin' IMAP when it's your clients or your server that's at fault. Sep 06 13:23:27 woglinde: but atleast it is fullscreen terminal Sep 06 13:23:28 you cant see all the kernel output e.g the option line Sep 06 13:23:52 hrw yes I have used it for many years Sep 06 13:24:05 screen is an other option, but has some problems too Sep 06 13:24:06 woglinde: like many of us here I think Sep 06 13:24:22 sakoman, just getting errors, nothing is transsfering Sep 06 13:24:58 sakoman, can you pastebin your .kermrc Sep 06 13:25:02 woglinde: have you turned on line wrap in minicom to see lines > 80 chars? Sep 06 13:25:05 What kind of errors? Sep 06 13:25:13 rwhitby hm oh Sep 06 13:25:21 Ctrl-A W Sep 06 13:25:23 sakoman, does not say Sep 06 13:25:25 I am wasnt aware of this option Sep 06 13:25:50 florian: it is for at91sam9263 which is a arm926 device Sep 06 13:25:52 try "robust" before send? Sep 06 13:26:29 florian: seems to be a confusion of include path while x-compiling Sep 06 13:27:04 Crofton: did you read "Using Kermit" section of http://docwiki.gumstix.com/Tutorial Sep 06 13:27:22 a little :) Sep 06 13:27:40 I use that setup and procedure, including the robust command before send Sep 06 13:27:42 noclouds: can you pastebin the full error message? Sep 06 13:28:56 I put all of the "set" commands in a text file and "take" the file at the start of a kermit session Sep 06 13:29:16 I put all the commands in .kermrc Sep 06 13:29:35 I use kermit on other stuff too, so it's an easy way to have multiple configs Sep 06 13:30:12 florian: here it is : http://rafb.net/p/gU47ZE42.html Sep 06 13:30:30 Crofton, so at what point do you start to get errors? Sep 06 13:30:51 immiediatley Sep 06 13:31:03 er, how do a take the file? Sep 06 13:31:19 Is it after the "send" command when the download status page is drawn? Sep 06 13:31:30 anybody here have a debian based distro using oe? i. e. build .deb instead of .ipk packages Sep 06 13:31:44 yeah Sep 06 13:31:49 just "take filename" Sep 06 13:32:46 What kind of machine are you using? Sep 06 13:32:59 noglitch: you are right, looks like your gtk-2.0.pc is wrong Sep 06 13:33:02 linux Sep 06 13:33:12 real serial port or USB adaptor? Sep 06 13:33:18 real serial Sep 06 13:33:18 * florian wonders how that happened Sep 06 13:33:46 but it should work! :-) Sep 06 13:34:01 yeah Sep 06 13:35:03 I'll do some more reading Sep 06 13:35:19 I've heard of some folks having issues like this on particular PC hw, switching to another machine made it go away Sep 06 13:35:43 Have another machine you could try? Sep 06 13:36:17 probably, need to find one wih a serial port :) Sep 06 13:36:27 I've noticed, that some notebook ACs causing problems to some USB/RS232 adapters Sep 06 13:36:27 HW was usually something funky like a Mac running Ubuntu Sep 06 13:36:47 if you've your AC near adapter or cable it might be a problem Sep 06 13:37:28 I have no issues with my Ubuntu/AMD64 or Ubuntu/x86 laptop, both using USB/serial adaptors Sep 06 13:37:40 Got to run, bbl Sep 06 13:38:53 anyone tried to build something with a meta-toolchain generated toolchain yet? Sep 06 13:39:57 * * OE Bug 2941 has been RESOLVED (FIXED) by Sep 06 13:39:59 * *  gnome-panel-2.18.1-r1-do_package Sep 06 13:40:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2941 Sep 06 13:40:06 florian: so, ok, is there a solution, how can I fix this ? Sep 06 13:40:11 it can't find -lgcc_s here Sep 06 13:41:22 mickeyl: this worked for me lately, but i built meta-sdk iirc Sep 06 13:41:46 ah, well, that's completely different Sep 06 13:42:00 especially since the latest mergfe Sep 06 13:42:06 mickeyl: does it have the libgcc_s.so symlink? Sep 06 13:42:11 florian: how do you feel about altering keylaunch to not use Meta as its key, see bug #2326 for the reasons Sep 06 13:42:13 koen: from where to where? Sep 06 13:42:30 ah Sep 06 13:42:46 from libgcc_s.so -> libgcc_s.so.whatever Sep 06 13:42:47 let me check Sep 06 13:43:35 noglitch: You need to find out how this path ended up there, so check if your pc file in staging is correct. If it is it must be something in the configure process of vlc. Sep 06 13:43:55 koen: ooh, good catch! that was missing. how should we fix the toolchain package then? Sep 06 13:43:56 Requested 'glib-2.0 >= 2.13.4' but version of GLib is 2.12.12Requested 'gtk+-2.0 >= 2.11.6' but version of GTK+ is 2.10.14Requested 'gtk+-unix-print-2.0 >= 2.11.6' but version of GTK+ is 2.10.14 Sep 06 13:44:02 ~lart epiphany Sep 06 13:44:02 * ibot lowers epiphany's priority Sep 06 13:44:36 mickeyl: yes, it should include the libgcc_s-dev package *or* make the symlink manually Sep 06 13:45:24 ok, which one do we prefer? Sep 06 13:45:58 I prefer the package approach, but historcially OE did it manually for sdk stuff Sep 06 13:46:08 because we lacked package Sep 06 13:46:21 even after we had the package :( Sep 06 13:46:24 XorA: hmm... sounds like a plan, i would not expect this to break any other platform Sep 06 13:46:38 koen: you need bleeding edge of gnome for epipany Sep 06 13:46:52 hrw: I noticed Sep 06 13:47:21 florian: will need keymaps rejigged in xserver-common I assume Sep 06 13:47:23 the 2.19.x PV was a big clue Sep 06 13:48:00 XorA: i guess so... Sep 06 13:48:25 * florian does not have time currently :-( Sep 06 13:48:27 florian: ok, I shall put together an RFC sometime in next few days and post it to lists Sep 06 13:48:32 florian: just in case Sep 06 13:48:48 florian: then do the work a little later Sep 06 13:48:56 XorA: ok Sep 06 13:49:14 *sigh* who broke pulseaudio startup? Sep 06 13:49:22 I'll try to follow, but its unlikely that I manage to do anything before mid of October. Sep 06 13:49:33 it seems we still do too many changes without checking on actual devices :/ Sep 06 13:49:49 phew, this one I didn't touch ;) Sep 06 13:51:14 XorA: http://hocwp.free.fr/xbindkeys/xbindkeys.html maybe instead of keylaunc? Sep 06 13:51:48 hrw: doesnt fix gpe though Sep 06 13:52:24 gpe buttons&keys app? Sep 06 13:55:46 root@fic-gta01:/tmp$ ./testing Sep 06 13:55:47 Hello World Sep 06 13:55:50 * mickeyl dances a jig Sep 06 13:56:03 * CM is mildly impressed ;) Sep 06 13:56:05 weeee! Sep 06 13:57:01 heh Sep 06 13:57:09 finally a working external toolchain Sep 06 13:57:13 this is important for me Sep 06 13:57:19 * mickeyl brabbles about gllin-eabi Sep 06 13:57:51 XorA, hrw: Fixing gpe-conf would be easy... I guess instead of xbindkeys we could use matchbox as well, but it does not cover long key presses you might want for some functions. Sep 06 13:58:13 mickeyl hehe Sep 06 13:58:19 except matchbox keybindings dont work here Sep 06 13:58:40 * florian edits siteinfo.bbclass to try MIPS (without el) Sep 06 13:58:59 florian: what device are you hacking on? Sep 06 13:59:00 florian oh which device? Sep 06 13:59:04 hehe Sep 06 13:59:09 florian: mickeyl's neod seems to handle long presses Sep 06 13:59:19 maybee the one nils told a little bit Sep 06 14:00:03 * koen makes a note to visit fotoimpex in berlin the next time Sep 06 14:00:26 koen is this a museum? Sep 06 14:00:50 woglinde: "Fotoversand & Fachgeschäft für analoge Fotomaterialien" Sep 06 14:00:59 hm googled it Sep 06 14:01:20 yeah isnt that far from zeckes work Sep 06 14:01:36 mickeyl, woglinde: A Basler eXcite camera: http://www.baslerweb.com/beitraege/beitrag_en_19756.html Sep 06 14:02:03 are we gonna take half a day for the locals to show us sites of Berlin? Sep 06 14:02:15 OE on the prowl as it were Sep 06 14:02:24 * florian cries for more time Sep 06 14:02:56 * mickeyl joins florian Sep 06 14:03:21 xora we will see Sep 06 14:03:26 mickeyl: you are the lucky one, you moved already and cancelled OSiM ;) Sep 06 14:03:35 florian: right ;) Sep 06 14:03:43 i really regret not to be there Sep 06 14:03:48 after all it's OpenMoko 1th annivesary Sep 06 14:03:54 but i just can't do it Sep 06 14:09:08 Crofton: back now, any progress? Sep 06 14:27:27 any volunteers for looking at adding texlive to OE? Sep 06 14:28:52 s/volunteers/insane Sep 06 14:28:55 woglinde: hi Sep 06 14:32:55 NOTE: package texlive-2007-r0: task do_fetch: started Sep 06 14:34:33 16:34 hrw@home:~$ apt-cache search ^texlive|wc 81 510 4013 Sep 06 14:41:37 83% [================================================================================================================================> ] 59,163,496 120.69K/s ETA 01:58 Sep 06 14:42:49 nice, can then tweak CV on neo :p Sep 06 14:43:17 * XorA does his CV on zaurus with abiword Sep 06 14:44:09 sharp killed zaurus line right? Sep 06 14:44:29 CL-3200 year+ old is last one? Sep 06 14:44:41 yes Sep 06 14:45:00 although with their new touch/scanner LCD I wouldnt be suprised to see new one Sep 06 14:46:30 CV? I keep such stuff online now Sep 06 14:47:13 I don't have one, it was just meant as joke Sep 06 14:50:56 RP: could it be that bitbake 1.8.9 breaks build.exec_task('do_clean', data) in seppuku.bbclass? Sep 06 14:51:41 koen: Richard is off today Sep 06 14:51:58 ah Sep 06 14:58:16 sakoman, got it to load, trick is to start kermit from a cold boot Sep 06 15:02:13 ~log Sep 06 15:02:14 log is, like, as piece of wood, or a record, or the opposite of exponentiation Sep 06 15:02:19 ~logs Sep 06 15:02:20 logs is, like, apt/ibot/infobot/jbot/purl all log daily to http://ibot.rikers.org// where channelname is html encoded ie: %23debian | lines that start with a space are not shown | some channels have stats at http://ibot.rikers.org/stats/.html.gz Sep 06 15:04:27 koen: sorry! back to the kb9202, if I may. Having made those changes in the org.oe.dev tree, how do I force bitbake to update the /tmp tree, get the kernel config changes, and rebuild, without 'rm -rf ./tmp'? bitbake -c rebuild task-base did not help. Sep 06 15:04:43 re Sep 06 15:04:49 bitbake -c rebuild virtual/kernel Sep 06 15:05:20 koen: gracias! Sep 06 15:05:56 crofton: good news! now we can get to the real issues:-) Sep 06 15:07:49 well Sep 06 15:07:55 I have the file in RAM :) Sep 06 15:08:10 and have confirmed CRC matches what Craig said it was Sep 06 15:08:19 * koen suspect qtopia people developed texlive Sep 06 15:08:30 it includes copies of the kitchensync and a dog Sep 06 15:09:46 koen hmmmm? Sep 06 15:10:21 a drowned dog? Sep 06 15:10:30 XorA, when you tested the suspend/brightness issue did you suspended your apitz with on/off or with the auto-suspend ? Sep 06 15:10:39 nodens: on/off Sep 06 15:10:46 s/apitz/spitz Sep 06 15:10:49 now, we make a sacrifice and reboot Sep 06 15:10:49 Crofton|home: nervous about reflashing uboot? :-) Sep 06 15:11:09 XorA, ok, this has always worked here Sep 06 15:11:22 but auto suspend breaks Sep 06 15:11:25 I bricked the OSK a few times Sep 06 15:11:31 nodens: oh, that was doing the zero brightness thing to me before I made new image :-) Sep 06 15:11:40 nodens: must be slight variations on the problem Sep 06 15:11:41 finally figured out the usb flasher Sep 06 15:11:50 I think it may be another bug Sep 06 15:12:22 haven't bricked a gumstix yet, but I'm sure there will be a first time! Sep 06 15:12:33 ok, I have a new uboot Sep 06 15:12:33 because fn+4 does'nt work when it's auto suspend Sep 06 15:12:43 hi koen! Sep 06 15:12:46 excellent! Sep 06 15:12:47 The bad news I am off to Toronto this evening :( Sep 06 15:12:57 so I won't be abel to do very much Sep 06 15:12:58 I'm sorry to hear that Sep 06 15:13:01 yeah Sep 06 15:13:07 Funeral is Sat Sep 06 15:13:26 hey HopsNBarley Sep 06 15:13:28 She had lived with Parkinsons for 40 years Sep 06 15:13:33 Crofton|home: my condolances Sep 06 15:13:37 that is a rough disease Sep 06 15:13:47 koen, thanks Sep 06 15:14:20 Crofton|home, condoleances. both my grand fathers died of parkinson Sep 06 15:14:38 funny how many you know have it Sep 06 15:14:51 a guy at schools uncle is 50 or so with it Sep 06 15:14:56 koen, finally crawling out from under my consulting rock. i'll be around more often. Sep 06 15:15:06 Crofton|home: my consolences. Sep 06 15:15:44 fortunately, if you can bear the medication, you can live a long time with it... Sep 06 15:16:07 yeah Sep 06 15:16:14 Crofton|home, sorry for your loss. i have a friend recently diagnosed. Sep 06 15:16:53 one of my grand could't bear the medication... fast evolution, not funny at all Sep 06 15:17:17 She was just a little to old for the implant Sep 06 15:17:24 I wish she had tried it Sep 06 15:17:34 My aunt, the cyborg Sep 06 15:18:03 arf Sep 06 15:18:09 hmm, I have a light on the switch now Sep 06 15:18:26 Crofton|home: amazing the boot speed difference with new compression scheme Sep 06 15:19:11 ok, I will work on geting that pushed Sep 06 15:19:20 unless someone already pushed? Sep 06 15:20:14 Don't know, haven't done a pull/update Sep 06 15:24:52 hmm dhcp command is not getting an ip Sep 06 15:25:04 fixed ip doesn't seem to work Sep 06 15:25:56 are you running with Craig's set of patches? Sep 06 15:26:04 he made the u-boot Sep 06 15:26:17 trying to get u-boot to tftp images Sep 06 15:26:23 so I can reflash Sep 06 15:26:30 Ah, OK I thought you had booted a kernel Sep 06 15:26:44 heh Sep 06 15:26:49 one step at a time :) Sep 06 15:27:00 You can reflash via serial, just takes a bit longer Sep 06 15:27:10 yeah Sep 06 15:27:21 I am all set up for tftp already Sep 06 15:27:25 so I want to try that Sep 06 15:28:00 Still not getting modules.dep populated during the build :-( Sep 06 15:28:07 Anyone ever figure that out? Sep 06 15:28:14 not sure Sep 06 15:28:26 there is a bunch of stuff in BZ that needs processing Sep 06 15:28:39 Seems like it worked for a while & then broke again Sep 06 15:28:40 I may take the gumstix along in case I get some time Sep 06 15:28:47 yeah Sep 06 15:28:54 Easy to pack, I do it all the time Sep 06 15:28:55 hoepfully there is stuff in BZ Sep 06 15:29:00 yeah Sep 06 15:29:10 BZ? Sep 06 15:29:47 bugzilla Sep 06 15:29:57 Ah, ok :-) Sep 06 15:37:25 I think craig has some work to do on the u-boot Sep 06 15:37:39 it hangs after a while while messing with the network Sep 06 15:37:50 What error do you get for the dhcp command? Sep 06 15:38:02 It keeps retrying Sep 06 15:38:10 ping is not reporting anything alive Sep 06 15:38:42 I get: Sep 06 15:38:47 GUM> dhcp Sep 06 15:38:47 BOOTP broadcast 1 Sep 06 15:38:47 BOOTP broadcast 2 Sep 06 15:38:47 *** Unhandled DHCP Option in OFFER/ACK: 52 Sep 06 15:38:47 *** Unhandled DHCP Option in OFFER/ACK: 52 Sep 06 15:38:48 DHCP client bound to address 192.168.0.17 Sep 06 15:38:50 Using SMC91C1111-0 device Sep 06 15:38:52 TFTP from server 0.0.0.0; our IP address is 192.168.0.17; sending through gateway 192.168.0.1 Sep 06 15:38:56 Filename '�'. Sep 06 15:38:58 Load address: 0xa2000000 Sep 06 15:39:00 Loading: * Sep 06 15:39:02 TFTP error: 'No such file or directory' (0) Sep 06 15:39:04 Starting again Sep 06 15:39:27 I have a different network interface Sep 06 15:39:44 Ah, that's right! Sep 06 15:39:52 LAN9x17 Sep 06 15:40:02 Keep forgetting you are on the bleeding edge :-) Sep 06 15:40:28 yeah Sep 06 15:55:55 bye Sep 06 16:09:57 * * OE Bug 2942 has been created by nicolas.ferre(AT)rfo.atmel.com Sep 06 16:10:00 * * Adding at91 two more machines Sep 06 16:10:02 * * http://bugs.openembedded.org/show_bug.cgi?id=2942 Sep 06 16:11:57 * * OE Bug 2943 has been created by autobuild(AT)openembedded.org Sep 06 16:11:59 * * libmimedir-0.4.2-r0-do_compile Sep 06 16:12:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2943 Sep 06 16:12:09 * * OE Bug 2944 has been created by autobuild(AT)openembedded.org Sep 06 16:12:11 * * libglade-2.5.1-r1-do_compile Sep 06 16:12:13 * * http://bugs.openembedded.org/show_bug.cgi?id=2944 Sep 06 16:19:00 does anybody know why kismet is built with --disable-setuid in OE ? Sep 06 16:19:30 nodens: because at the time it was built for OZ that only had a root user Sep 06 16:19:43 nodens: I've fixed it since, but waiting on a commit (later this month) Sep 06 16:19:57 Ok. Sep 06 16:20:13 I'm trying to write a recipe for kismet-newcore Sep 06 16:20:17 seems easy enough Sep 06 16:20:52 I'll need to add conflicts/provide/replace in the ipkg, though. Sep 06 16:20:56 nodens: no need, I have it all done Sep 06 16:21:00 oh. Sep 06 16:21:15 I'm glad I asked, then ;) Sep 06 16:21:19 nodens: problem is, kismet-newcore has no "make install". I will resolve this in a few days and bug my friend to commit Sep 06 16:21:32 nodens: give it a week, and if you see nothing then bother me ;-) Sep 06 16:21:40 ok :) Sep 06 16:22:00 I'll just use the stable one until then :) Sep 06 16:22:09 nodens: I would give you my newcore reciepe now, but it is completely pointless as the "make install" is borked. Sep 06 16:22:58 sakoman, did the serial load thing, now I haved duplicated asmola__ issue from yesterday Sep 06 16:23:01 time for lunch Sep 06 16:23:07 actually, I guess it does kind of work, in a hackish way... but yeah, give me a week Sep 06 16:25:00 Zero_Chaos, ok. Sep 06 16:26:54 rwhitby: I guess you've seen this already, but in case you haven't: http://chezphil.org/slugterm/ Sep 06 16:28:00 Crofton|home: excellent! What *was* Asmola's issue from yesterday? Sep 06 16:28:17 bioot hangs access hwclock Sep 06 16:28:28 I suspect we have fixed and vroke a couple fo things Sep 06 16:32:49 My main issue is still the lack of modules.dep at build time, I think the hwclock thing is related Sep 06 16:32:59 * florian pushes some MIPS bits Sep 06 16:33:12 When I get a modules.dep at build time I don't see the hwclock issue Sep 06 16:34:01 When I don't see modules.dep at build, it gets created at first boot and all is well till reboot Sep 06 16:34:23 then I get fatal module errors and the hwclock thing Sep 06 16:34:58 why not buildin the rtc module? Sep 06 16:44:57 * * OE Bug 2942 has been RESOLVED (FIXED) by Sep 06 16:44:59 * *  Adding at91 two more machines Sep 06 16:45:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2942 Sep 06 16:46:30 build in, that is Sep 06 16:56:33 03florian 07org.oe.dev * rfca56676... 10/ (1 classes/siteinfo.bbclass): siteinfo.bbclass: Add mips (MSB) architecture Sep 06 16:56:41 03florian 07org.oe.dev * r31702ffb... 10/ (1 site/mips-linux): site: Add mips site information (untested) Sep 06 16:56:48 03nicolas.ferre 07org.oe.dev * r9c6ceb61... 10/ (9 files in 6 dirs): at91sam*: add 2 more at91sam machines, close #2942 Sep 06 16:58:13 koen: that might fix the rtc problem for them, but I suspect they will then run into the other problems I am seeing with modules Sep 06 16:58:26 one step at a time :) Sep 06 16:59:00 I think all are probably related to broken modules Sep 06 16:59:33 Need to make some time to debug Sep 06 16:59:48 Too big a "to do" list Sep 06 17:02:57 koen: would be nice if there was an intermediate step between task-boot and task-base Sep 06 17:03:13 moving from boot to base adds 1200 tasks to the build! Sep 06 17:03:26 and many extra tens of minutes :-( Sep 06 17:03:39 koen@bitbake:/data/build/koen/OE/build/tmp/angstrom/deploy/glibc/ipk$ tar ztvf ../images/gumstix-connex/Angstrom-minimalist-image-glibc-ipk-2007.9-test-20070905-gumstix-connex.rootfs.tar.gz | grep modules.dep Sep 06 17:03:39 -rw-r--r-- root/root 724 2007-09-05 12:48 ./lib/modules/2.6.21/modules.dep Sep 06 17:03:56 sakoman: tasks to *build*, not to install Sep 06 17:04:19 Yes, but it is the build that hurts so much :-) Sep 06 17:04:32 koen@bitbake:/data/build/koen/OE/build/tmp/angstrom/deploy/glibc/ipk$ tar ztvf ../../uclibc/images/gumstix-connex/Angstrom-minimalist-image-uclibc-ipk-2007.9-test-20070905-gumstix-connex.rootfs.tar.gz | grep modules.dep Sep 06 17:04:33 -rw-r--r-- root/root 724 2007-09-05 13:25 ./lib/modules/2.6.21/modules.dep Sep 06 17:05:05 hmm . . . so why am I missing it in my build Sep 06 17:06:47 mtn status says I'm good Sep 06 17:07:05 time for rm -rf tmp? Sep 06 17:07:18 can't go wrong with that Sep 06 17:07:23 sigh Sep 06 17:08:20 http://rafb.net/p/DqgACu23.html has been chugging along for the past day for me Sep 06 17:09:08 now you just need a way to test all those ... Sep 06 17:09:45 koen: I have no build failures, just failure to function properly :-) Sep 06 17:10:25 sakoman, you aren't getting any build failures for gumstix connex, which angstrom image, a custom one? Sep 06 17:11:00 re Sep 06 17:11:15 zecke: wb Sep 06 17:11:33 asmola__: did you remove TMP after tuesday's bbclass update? Sep 06 17:11:52 yeah Sep 06 17:12:05 asmola: yes, simple custom image Sep 06 17:13:59 morning Sep 06 17:14:01 asmola: http://www.sakoman.net/oe/sakoman-s100-image.bb Sep 06 17:14:07 /bin/sh: ./otangle: cannot execute binary file Sep 06 17:14:11 morning! Sep 06 17:14:12 * koen stabs texlive Sep 06 17:14:25 koen, hmm, i thought i did, now i am wondering if it was tuesday or wednesday Sep 06 17:14:36 sakoman: the DEPENDS aren't needed anymore in image recipes :) Sep 06 17:14:50 excellent! they are now gone! Sep 06 17:15:03 hughescr, morning Sep 06 17:15:12 seemed a silly duplication Sep 06 17:15:24 lots of junk dna in the recipes Sep 06 17:15:29 you can just say 'inherit image ; IMAGE_INSTALL = "stuff I want"' Sep 06 17:15:57 I am here, headed home to work on gumstix for a few hours, then off to Toronto Sep 06 17:16:10 hopefulyl, I can resolve the verdex not booting issue Sep 06 17:16:59 bbiab Sep 06 17:16:59 koen: like this? http://www.sakoman.net/oe/sakoman-s100-image.bb Sep 06 17:20:27 sakoman: and IMAGE_BASENAME defaults to ${PV}, so you can remove that as well :) Sep 06 17:20:41 hm. is it known that the links to viewmtn in the commitlog mails are broken? Sep 06 17:22:19 koen: I love removing stuff, less to go wrong :-) Sep 06 17:28:02 bbl Sep 06 17:34:33 sakoman: (and others) I've set up an OE branch for gumstix rapid iteration, targetting getting to where we've got a buildroot replacement set of machine/distro/image Sep 06 17:34:44 It's pretty much OE + my bug patches right now Sep 06 17:34:52 available at host rungie.com Sep 06 17:34:58 branch is called org.gumstix.oe Sep 06 17:35:05 open for pulling from Sep 06 17:35:16 and I can easily add anyone's monotone keys for writing too Sep 06 17:35:51 and how are you planning to get that into OE proper? Sep 06 17:36:02 heh, now we can test monotone hard :) Sep 06 17:36:21 koen: asking me? Sep 06 17:36:29 yes Sep 06 17:36:59 koen: Bad fat momentum :) Sep 06 17:37:05 guys slow down .... Sep 06 17:37:50 koen: I'm just trying to allow closer/easier/faster cooperation among folks working on gumstix stuff which is still basically highly experimental Sep 06 17:38:06 and how are you planning to get that into OE proper? Sep 06 17:38:18 so I can share patches easily with crofton, sakoman, asmola, etc and we'll all be on the same page, even if those patches aren't yet suitable to merge back into OE upstream Sep 06 17:38:31 and how are you planning to get that into OE proper? Sep 06 17:38:33 koen, I am certainly working on the getting stuff in to gumstix Sep 06 17:38:40 er OE Sep 06 17:38:51 eventually things on our branch will settle down and we'll have a coherent "this is how we think it makes sense to do gumstix, which btw is working" Sep 06 17:38:55 but, for obivous reasons, this process is being delayed Sep 06 17:39:04 if it's not the right way, we can get it to be the right way based on a working system Sep 06 17:39:16 instead of trying to both solve workingness and correctness at the same time, which is much slower Sep 06 17:39:34 my feelings are is that anyone who wants to try maintaining a vendor branch with monotine is learning stuff we need to know Sep 06 17:39:42 ie the focus of the branch is workingness, not so much correctness Sep 06 17:39:50 then we can fix correctness when merging back Sep 06 17:40:45 Crofton: yes and long term this will be an ongoing vendor branch, providing a bit more stability for gumstix OEMs Sep 06 17:41:11 I know multiple people here are interested in this problem Sep 06 17:41:19 org.oe.dev frankly is a bit unstable imo for a base for product development; one really needs some less fluid base to build a large-volume embedded system on Sep 06 17:41:35 there will be a angstrom stable branch in the near future :) Sep 06 17:41:45 that might well help :) Sep 06 17:41:46 koen, good :) Sep 06 17:41:51 it will make my life better Sep 06 17:42:15 koen: here, here on the stable branch! Sep 06 17:42:38 ultimately though, I have a feeling our customers are going to look to us to provide the stability they need, on a schedule related to gumstix product cycles rather than the broader angstrom timeline Sep 06 17:42:59 so we might have the org.gumstix.oe branch based on angstrom-stable instead of .dev, but the concept will still probably be there Sep 06 17:43:28 but it'd be a lot easier to keep in sync with an angstrom-stable branch probably then a .dev Sep 06 17:44:04 ok, I am really interested in this conversation Sep 06 17:44:16 but I want to try and fix some things b4 I have to leave Sep 06 17:44:41 So anyway, anyone: feel free to mtn pull rungie.com Sep 06 17:44:52 and to add to the complexity, each gumstix customer is likely to have org.customer.oe branches Sep 06 17:45:01 and feel free to send me your keys if you want to mtn push Sep 06 17:45:39 btw which OE mailing list would be the best to discuss this kind of thing on? Sep 06 17:45:45 oe-dev Sep 06 17:45:58 traffic on the one I subscribed to, which I forget which one that was, seems low and not on these types of meta-topics Sep 06 17:46:04 koen: ok Sep 06 17:46:22 sakoman: not probably *each*, but many Sep 06 17:46:52 true, probably mainly the non-hobbiest customers Sep 06 17:48:18 anybody using Fedora 7 for host machine here Sep 06 17:48:47 yes Sep 06 17:48:48 khem: Not at the moment, I have one in vmware almost ready to test w/ OE Sep 06 17:49:03 Hi khem, Crofton|home Sep 06 17:49:09 high Sep 06 17:49:23 khem, it works great for me Sep 06 17:49:33 unless it is x84_64 Sep 06 17:49:45 and then you have to hack binutils-cross Sep 06 17:50:19 Crofton|home: Do you have all updates applied to your host F7 Sep 06 17:50:25 yes Sep 06 17:50:31 I should at least Sep 06 17:50:36 Crofton|home: so it must be same as me Sep 06 17:50:48 I hope so, what is the problem? Sep 06 17:50:50 Have you tried a clean build lately Sep 06 17:51:03 past couple of days Sep 06 17:51:04 I am trying angstrom-console-image Sep 06 17:51:10 problem? Sep 06 17:51:14 and I am getting lex problem Sep 06 17:51:28 hmm someone else had that Sep 06 17:51:32 wait maybe it was ytou Sep 06 17:51:37 first I got it with binutils-2.17.50.0.12 Sep 06 17:52:03 and then when I switch to use 2.18 this works so my build goes further Sep 06 17:52:28 but dies with same message when doing pcmciautils_014.bb Sep 06 17:52:52 problem is that it uses -lm option in addition Sep 06 17:52:57 Let me pull and run a build on that machine Sep 06 17:53:00 which is not recognised by flex Sep 06 17:53:08 I am glibc Sep 06 17:53:23 I think libc's are not problem Sep 06 17:53:38 its libtool Sep 06 17:53:54 hmmm Sep 06 17:54:07 I better make sure I use libtool-1.5.10 then :) Sep 06 17:54:28 yep, using stock libtool Sep 06 17:54:36 http://pastebin.ca/684328 Sep 06 17:54:55 khem, I am leaving for a few days this afternoon, and need to fix somethign for asmola__ Sep 06 17:55:10 sure Sep 06 17:55:14 especially since helping him has attracted even more work to me :) Sep 06 17:55:22 :) Sep 06 17:55:28 gumstix Sep 06 17:55:40 yeah Sep 06 17:55:54 they are all the rage in the SDR world for some reason Sep 06 17:56:21 hmm, I am pulling 231 revs Sep 06 17:59:18 alright, so the number one question is why doesn't the gumtix mount recognize the size option as the OSK moutn appears to see Sep 06 18:00:46 Crofton: udev-115 Sep 06 18:00:54 my patch there fixes the size problem Sep 06 18:00:54 why? Sep 06 18:01:10 in the udev init script, it tries to pass the size option when mounting tmpfs Sep 06 18:01:20 my udev-115 removes that arg Sep 06 18:01:30 hughescr, the question I have is "Why does it work on the OSK?" Sep 06 18:01:36 ah Sep 06 18:01:41 different tmpfs implementation? Sep 06 18:01:48 which uses mount-util Sep 06 18:02:03 and I think we tried to fix up minimal-image to do this Sep 06 18:02:29 basically, OE has a huge testing issue Sep 06 18:02:38 the problem is udev is trying to pass a mount option which the filesystem doesn't grok Sep 06 18:02:59 it's most likely a difference between the OSK kernel and the gumstix kernel Sep 06 18:03:09 in the tmpfs filesystem implementation Sep 06 18:03:36 why would they be different? Sep 06 18:04:01 who can help with pcmciautils Sep 06 18:04:25 tmpfs can be ramfs or something more fancy, IIRC Sep 06 18:04:39 it's a kernel option, IIRC Sep 06 18:05:01 yeah, I had to adjust the gusmtix defconfig so that it is included by default Sep 06 18:05:06 maybe I buggered that Sep 06 18:05:07 crofton: linux .config file, or kernel version, or patch set Sep 06 18:05:12 those are your 3 choices :) Sep 06 18:06:09 bother someone is hammering my dsl ... Sep 06 18:06:25 aha Sep 06 18:06:40 org.openembedded.dev/packages/linux/linux-2.6.21/ramfs-mode-support.patch Sep 06 18:06:45 that could be the problem right there Sep 06 18:06:48 khem, I will run a clean build on a f7 bx Sep 06 18:06:59 can't do anymore than that for you :( Sep 06 18:07:01 in adding mode support, it looks like dave might have stopped other params (like size) from working Sep 06 18:07:24 Crofton|home: no thats more than enough please run angstrom-console-image Sep 06 18:07:43 will do Sep 06 18:07:46 updating first Sep 06 18:08:35 hughescr, ramfs: Bad mount option size Sep 06 18:08:41 look familiar ... Sep 06 18:08:44 Crofton: yup Sep 06 18:09:10 ~lart vendors with large patch sets Sep 06 18:09:11 * ibot urinates on vendors with large patch sets Sep 06 18:09:12 :) Sep 06 18:09:12 I'm guessing that without the ramfs_parse_option fn being in there, there's some default fs_parse_option thing which kicks in and groks size Sep 06 18:09:23 which might well also now solve the mode thing Sep 06 18:09:23 yeah Sep 06 18:09:27 thanks for catching that Sep 06 18:10:21 I have a Q: how do people handle patchsets when developing inside OE? Sep 06 18:10:27 what is this patch trying to do? Sep 06 18:10:32 ie quilt won't parse a .bb file as a series file Sep 06 18:10:34 poorly :) Sep 06 18:10:43 webkit-gtk-0.0+svnrwebcore-rwebkit-r1 Sep 06 18:10:55 shouldn't the svn revision be there somehwere? Sep 06 18:10:56 there was long discussion on the dev list about that Sep 06 18:10:59 Crofton: the patch is adding "mode" as an option to ramfs, which didn't used to work Sep 06 18:11:01 it might work now Sep 06 18:11:10 hmm, you are .21 Sep 06 18:11:17 if you have a working OSK, can you try mounting a ramfs with a "mode=xxxx" option? Sep 06 18:11:24 see if the kernel does it right Sep 06 18:11:40 oh, what version is osk? Sep 06 18:11:45 .22 Sep 06 18:12:13 ok Sep 06 18:12:41 let's look at the .22 ramfs code and see if maybe they added parse_option there Sep 06 18:12:51 http://rafb.net/p/QqyM6s28.html Sep 06 18:12:53 I'm pretty sure that the ramfs-mode patch is one which went upstream Sep 06 18:13:04 and might have been folded in with some extra bonus stuff like size too Sep 06 18:13:21 ok, so mode's working :) Sep 06 18:13:44 but I notice udev's not mounted a /dev tmpfs Sep 06 18:13:50 oh, /dev/shm Sep 06 18:13:52 duh Sep 06 18:15:43 I don't see a parse_option in 2.6.22 ramfs Sep 06 18:16:20 Does libidl fuck up at package_write? Sep 06 18:16:41 Bitbake bails out due to this: Sep 06 18:16:42 IOError: [Errno 2] No such file or directory: '/usr/local/oe/tmp/work/sh3-linux/libidl-0.8.6-r3/install/libidl-dbg.lock' Sep 06 18:17:42 Crofton: what .bb is osk using to build linux? Sep 06 18:17:58 linux-omap1 Sep 06 18:18:02 ah found it Sep 06 18:18:08 I need to move it to linux.inc Sep 06 18:18:58 Crofton: large patchsets is way nicer than one big honking patch like http://www.muru.com/linux/omap/patches/patch-2.6.22-omap1.bz2 Sep 06 18:18:59 ;) Sep 06 18:19:07 well Sep 06 18:19:16 they are moving stuff upstream :) Sep 06 18:19:48 this is basically a patch to get from the kernel.org tarball to the git version on teh omap kernel Sep 06 18:20:43 Crofton|home: Do u add $OEDIR/build/tmp/staging/i686-linux/bin Sep 06 18:20:45 khem, angstrom-console-image Sep 06 18:20:48 to ur path Sep 06 18:20:52 no Sep 06 18:21:03 ok Sep 06 18:21:30 so bb knows to pick stuff from staging? Sep 06 18:21:35 Crofton: yeah but it's one big 3.3MB patchfile Sep 06 18:21:55 with all the changes rolled into one so it's impossible to tell what the differences are.... Sep 06 18:22:14 hughescr, if you care, I think you go play with the git repo Sep 06 18:22:21 ie hard to locate where/if they've implemented something to allow mode= in ramfs, or if it's just coming in form upstream Sep 06 18:22:35 hughescr, understood Sep 06 18:22:48 but, I suspect stuff like that is from upstream Sep 06 18:22:53 just saying, lots of ways to skin a cat; all of them suck for the cat Sep 06 18:23:16 rofl Sep 06 18:23:33 basically, the kernel.org patch + the linunx-omap1 patch give us a "stable" reproducible kernel Sep 06 18:23:48 A friend of mine skinned a cat once and tanned the hide Sep 06 18:23:49 without a ramfs patch, I don't see how ramfs can be supporting mode Sep 06 18:24:41 I get NOTE: Psyco JIT Compiler (http://psyco.sf.net) not available. Install it to increase performance. Sep 06 18:24:43 in 2.6.22 static int ramfs_fill_super(struct super_block * sb, void * data, int silent) Sep 06 18:24:44 is Sep 06 18:24:49 inode = ramfs_get_inode(sb, S_IFDIR | 0755, 0); Sep 06 18:24:52 ie hard-coded mode Sep 06 18:25:16 * Crofton|home wonders Sep 06 18:25:20 so osk I think must be using somethign else not ramfs for tmpfs Sep 06 18:25:22 but I have installed python-psyco-1.5.1-5.fc7 Sep 06 18:25:30 khem: you on an intel machine? Sep 06 18:25:46 yes Sep 06 18:26:46 don't know then :) Sep 06 18:28:15 mickeyl: looks like SRCREV_FORMAT is broken in your bitbake Sep 06 18:28:32 koen: oh. ok ;) Sep 06 18:28:51 yup, got it Crofton Sep 06 18:28:57 this is off on gumstix, on in osk Sep 06 18:28:57 Use full shmem filesystem Sep 06 18:28:57 uboot-openmoko_svn.bb has a parse error .... Sep 06 18:29:08 under "Configure standard kernel features" Sep 06 18:29:15 under "General Setup" Sep 06 18:29:18 mickeyl: and in mine :( Sep 06 18:29:29 koen: bummer Sep 06 18:29:35 ie choosing ramfs vs shmfs for tmpfs Sep 06 18:29:41 bummer is such an awesome word Sep 06 18:29:53 hehe Sep 06 18:30:01 hughescr, looking now, I wa at the TMPFS/RAMFS line Sep 06 18:30:02 mickeyl: it used to work when bitbake 1.8.8 wasn't released Sep 06 18:30:21 lets pester RP tomorrow ;) Sep 06 18:30:25 what is the name of the kconfig var Sep 06 18:30:26 zecke: or do you have an idea? Sep 06 18:30:29 khem: I tested, committed and pushed the "touch all configures before build" patch. Sep 06 18:30:43 so I think the answer is to remove the ramfs mode patch, and switch that on; or keep the ramfs patch and remove the size arg in udev's init Sep 06 18:30:52 khem: (for eglibc that is) Sep 06 18:30:54 likewise: ok Sep 06 18:30:55 libwebkitgdk1_0.0+svnrwebcore-rwebkit-r1_arm.deb Sep 06 18:30:56 I will try that Sep 06 18:31:13 we want things to work the same on as many machines as possible' Sep 06 18:31:29 SRCREV_FORMAT = "webcore-rwebkit" <- that worked in the past Sep 06 18:32:06 Crofton: yeah, but I'm also trying to make a roots which fits in 4MB :) Sep 06 18:32:22 heh Sep 06 18:32:25 Crofton: sorry bootloader + uImage + rootfs Sep 06 18:32:31 right Sep 06 18:32:38 Crofton: on verdex though, np switching to shmfs Sep 06 18:32:45 and I don't know how big the size difference is Sep 06 18:32:51 might still not be a problem Sep 06 18:32:57 * * OE Bug 2929 has been RESOLVED (FIXED) by likewise(AT)gmx.net Sep 06 18:32:59 * *  eglibc should not regenerate configure scripts Sep 06 18:33:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2929 Sep 06 18:33:01 hughescr, lets get a working config in Sep 06 18:33:03 mickeyl: ah, it looks like someone reverted parts of the webkit .bb or I didn't check in all my stuff Sep 06 18:33:09 then we can work on the edge cases Sep 06 18:33:55 the problem OE tries to tackle is hard Sep 06 18:34:10 I am not sure if we even understand what problem we are trying to sovle :) Sep 06 18:34:59 mickeyl: fixed Sep 06 18:35:25 libidl stages for me. Who asked? Sep 06 18:36:06 B_Lizzard: libidl stages for me. Sep 06 18:36:55 yeah, i cleaned it, and it built correctly, but it does the same for many packages after that Sep 06 18:37:07 I'll give you a pastebin Sep 06 18:37:28 B_Lizzard: remove temp Sep 06 18:37:44 ehm? Sep 06 18:37:47 tmp? Sep 06 18:37:48 koen: cool, thanks Sep 06 18:37:56 and start reading oe-devel Sep 06 18:38:24 ok... Sep 06 18:38:30 http://pastebin.ca/684379 Sep 06 18:38:44 in case it's something else Sep 06 18:38:47 :/ Sep 06 18:38:47 woglinde: ping Sep 06 18:39:01 two heads beat as one ... Sep 06 18:39:40 B_Lizzard: could you please investigate the permissions on these lock files using "ls -ald some_lock_file" ?> Sep 06 18:39:48 Ok.... Sep 06 18:39:51 Crofton: heh. that stupid 80s song Sep 06 18:40:16 :/ Sep 06 18:41:24 hey do13 Sep 06 18:41:26 how are things? Sep 06 18:41:39 hi mickeyl Sep 06 18:42:54 fine Sep 06 18:43:10 likewise: he needs to remove tmp Sep 06 18:43:14 It doesn't even exist Sep 06 18:43:18 Yeah, seems like that Sep 06 18:43:31 Sorry for the mixup, it's probably my setup or something :? Sep 06 18:43:33 :/ Sep 06 18:43:38 mickeyl: busy with openmoko? Sep 06 18:43:46 koen: how could you tell? hi, btw :-) Sep 06 18:44:43 The package is already built, but bitbake seems to want to rebuild it Sep 06 18:44:45 :/ Sep 06 18:45:08 likewise: because everyone needed to after tuesday due to the package*bbclass changes Sep 06 18:45:17 do13: terribly, yes. Sep 06 18:45:24 do13: although it's good in some sense :) Sep 06 18:46:22 mickeyl: good news, we have another doughter Sep 06 18:46:46 should be daughter :) Sep 06 18:46:53 yay! Sep 06 18:46:55 congrats Sep 06 18:46:57 great Sep 06 18:47:05 thanks Sep 06 18:47:28 do13: congrats! Sep 06 18:47:35 lol Sep 06 18:47:39 do13 gratz Sep 06 18:47:56 thanks Sep 06 18:49:53 do13: Congratulations! How many days already? Sep 06 18:50:00 too late, daughter calls :-) Sep 06 18:55:33 koen, is this still a parsing problem? uboot-openmoko_svn.bb Sep 06 18:56:25 Crofton|home: uboot-openmoko also pull uboot git Sep 06 18:56:30 which is &#(@*&$(@*#&$@*( broken Sep 06 19:00:42 03likewise 07org.oe.dev * r4a859de9... 10/ (4 files in 2 dirs): eglibc: touch configure files to prevent them from being regenerated. Sep 06 19:00:46 03koen 07org.oe.dev * rb90c6e33... 10/ (1 packages/webkit/webkit-gtk_svn.bb): webkit: reinstate ;name inj SRC_URI Sep 06 19:00:53 03koen 07org.oe.dev * rabf691e4... 10/ (1 conf/machine/nokia770.conf): nokia770: fix XSERVER Sep 06 19:03:09 koen, does this mean if fetching ails, parsing breaks? Sep 06 19:03:19 yes Sep 06 19:04:12 argh Sep 06 19:04:22 * Crofton|home is not a fan of this behavior .... Sep 06 19:05:59 Crofton|home: http://bugs.openembedded.org/show_bug.cgi?id=2897 Sep 06 19:08:34 can someone check if http://www.openembedded.org:1081/revision/filechanges/6a2d2b93d5aaaeac208513989dea7830e85864d2/classes/package_ipk.bbclass works for them? Sep 06 19:10:29 yikes! i just did fresh check out, and now it get all these errors from bitback regarding "get_srcrev" Sep 06 19:10:40 bitbake, i mean :-) Sep 06 19:10:44 ":'module' object has no attribute 'get_srcrev' while evaluating:" Sep 06 19:10:45 tony_: upgrade bitbake Sep 06 19:10:56 oh? what is the preferef version? Sep 06 19:10:57 and start using a distro that includes sanity.bbclass Sep 06 19:12:08 koebn: what does that class change? Sep 06 19:12:11 koen: what does that class change? Sep 06 19:12:25 koen, what bitbake version do you recommend? Sep 06 19:12:36 likewise: it moves pkgdata around to stop cross-build contamination Sep 06 19:12:39 tony_: 1.8.8 Sep 06 19:12:52 thanks, i will give it a try Sep 06 19:14:25 koen: starting a build Sep 06 19:14:35 koen: I don't see why 1.8.9 would break an exec_task call Sep 06 19:14:42 koen: how should I see if it works as you expect it to behave? Sep 06 19:15:08 RP: the task won't run if you don't inherit the class to run it Sep 06 19:15:15 RP: IOW: I'm a moron Sep 06 19:15:24 koen: ah, ok :) Sep 06 19:15:32 http://www.openembedded.org/aggregator/sources works again Sep 06 19:16:20 next step is to update ciabot-email to use correct URLs Sep 06 19:18:13 done Sep 06 19:19:06 the new and improved viewmtn now has a filehistory as well for easier web-based blaming :) Sep 06 19:19:48 crofton: what did you and Craig decide to do re: size option on tmpfs? Sep 06 19:20:12 remove ramfs patch and Craig's udev patch? Sep 06 19:23:47 I am trying that Sep 06 19:24:01 I am about to test the new kernl Sep 06 19:24:31 good luck! Sep 06 19:25:40 sadly, I think that that patch was in response to a tmpfs mode bug I reported to gumstix a few months back :-) Sep 06 19:25:48 rofl Sep 06 19:26:15 :) Sep 06 19:26:32 guess the fix wasn't quite right -- surprising cause the guy who submitted it s usually really good about not breaking things Sep 06 19:27:14 well, the OSK is running .22, so maybe what we need isn't in .21 Sep 06 19:28:26 deleting temp fixes nothing Sep 06 19:29:54 bye Sep 06 19:31:47 hangs on loading kernel ... Sep 06 19:31:55 ooooooh Sep 06 19:31:59 trying to flash again Sep 06 19:35:57 Oh ffs Sep 06 19:36:04 This keeps going on and on and on Sep 06 19:47:29 mtn: updating along branch 'org.openembedded.dev' Sep 06 19:47:33 mtn: selected update target 5cca19f60a761cd1b4ec84d916a7eb7f46c8b563 Sep 06 19:47:45 mtn: misuse: path _MTN/tmp/1 already exists Sep 06 19:47:55 eh, what's that? Sep 06 19:48:28 update monotone Sep 06 19:48:35 to the latest version Sep 06 19:49:00 ynezz: rm -rf _MTN/tmp Sep 06 19:49:07 That won't do much Sep 06 19:49:13 i've been using my version for more then week without problems Sep 06 19:49:42 monotone 0.31 on ubuntu Sep 06 19:49:51 ynezz: rm -rf _MTN/tmp Sep 06 19:50:01 Well, there are two choices, either start over, or update your monotone Sep 06 19:50:09 I had the same exact problem Sep 06 19:50:22 I had to update to monotone 0.35 Sep 06 19:50:55 Or start a new org.openembedded.dev Sep 06 19:51:09 Crofton|home: the build worked for me now Sep 06 19:51:10 B_Lizzard: or just remove that directory and mtn update again... Sep 06 19:51:22 It'll bring up the same problem, zecke_ Sep 06 19:51:54 khem, I am in parsing hades Sep 06 19:51:55 Well, try deleting that directory Sep 06 19:52:10 It might work, in case it's not the same exact problem Sep 06 19:52:13 koen: How can I enable network access with angstrom-console-image is it included or do I need more packages Sep 06 19:52:25 But updating monotone resolved it permanently for me Sep 06 19:52:33 I mean internet access Sep 06 19:52:40 zecke_: now it ends like this, http://pastebin.ca/684466 Sep 06 19:53:19 Hmm Sep 06 19:53:30 mtn revert --missing Sep 06 19:53:36 And mtn update again Sep 06 19:53:57 mtn: misuse: path _MTN/tmp/1 already exists Sep 06 19:54:02 same problem Sep 06 19:54:04 :) Sep 06 19:54:07 jesus :) Sep 06 19:54:09 Update monotone Sep 06 19:54:20 Maybe getdeb has something Sep 06 19:54:34 ynezz: what happens if you do a checkout out of the db directly (in contrast to update) Sep 06 19:54:55 I will try Sep 06 20:02:33 Crofton: did your reflash work? I'm doing a build with the same changes Sep 06 20:03:26 zecke_: mtn: misuse: 12107 missing files; use 'mtn ls missing' to view Sep 06 20:03:52 no, does not boot Sep 06 20:03:55 ynezz: ouch, what was your command? Sep 06 20:04:11 mtn checkout openembedded; mtn update Sep 06 20:04:24 ynezz: from within your working copy? Sep 06 20:04:34 ynezz: do it from outside (as I said...) Sep 06 20:05:41 sakoman, does this mean anything to you? Sep 06 20:05:43 http://rafb.net/p/yncoiy79.html Sep 06 20:06:04 ynezz: but it is definately strange Sep 06 20:06:04 zecke_: i don't get it, hm what do you mean? download DB again and then do checkout? Sep 06 20:06:14 cd .. Sep 06 20:06:24 that's what i did of course Sep 06 20:06:26 mtn --db whatever/your/db co --branch=org.openembedded.dev Sep 06 20:06:35 ah Sep 06 20:07:20 bother, I had a patched linux.inc Sep 06 20:09:29 Crofton|home, i just booted over here cause something looked fishy, it doesn't seem to know the kernel image is compressed Sep 06 20:10:07 I think it is using the kernel decompressor now Sep 06 20:10:24 zecke_: this has fixed it, thanks :) Sep 06 20:10:28 Crofton: that's all you get? Sep 06 20:10:49 If so I think your kernel build is broken somehow Sep 06 20:11:02 yeah, I had somethign based of Craigs patch in Sep 06 20:11:10 I have reverted to match mtn and will see what happens Sep 06 20:13:06 Crofton, the toher thing i am notcing is that your kernel copy is copying from 0x00f00000 and mine is going from 0x01f00000 Sep 06 20:13:44 http://pastebin.com/m71b6b5a Sep 06 20:14:01 maybe you have more flash than I? Sep 06 20:14:16 yep Sep 06 20:14:52 i thought it ended up in the same place regardless of which version but maybe not, oh well Sep 06 20:15:24 asmola: it loads at an offset from *top* of flash Sep 06 20:15:29 ahhhhhhh Sep 06 20:15:39 good to know sakoman, thanks Sep 06 20:16:05 GUM> katload Sep 06 20:16:05 1 Usage: Sep 06 20:16:05 katload - Load kernel to RAM from offset at top of flash Sep 06 20:16:10 :-) Sep 06 20:16:30 you actually read the manual? :) Sep 06 20:16:47 no, I just forgot to type 100000 one time :-) Sep 06 20:18:16 :) Sep 06 20:28:56 * * OE Bug 2945 has been created by autobuild(AT)openembedded.org Sep 06 20:28:58 * * matchbox-wm-1.2-r1-do_compile Sep 06 20:29:00 * * http://bugs.openembedded.org/show_bug.cgi?id=2945 Sep 06 20:36:47 bugger all Sep 06 20:37:13 my efforts at booting the gusmtix are not going well Sep 06 20:37:28 re Sep 06 20:41:28 Crofton: can't help, have no gumstix :-) Sep 06 20:42:09 Crofton: btw. I closed my screen session and restarted it now angstrom-console-image builds file strange Sep 06 20:43:09 KhemWork: if a restart succeeds but the first build did not, it could be the sanity checker that made the first build fail on sanity (but the task will succeed). Sep 06 20:43:31 I have reverted to match mtn and will see what happens Sep 06 20:43:37 KhemWork: I hope that sentence made sense Sep 06 20:43:43 likewise: hehe Sep 06 20:44:21 Crofton: are you actually able to build your davinci image? I have the board right here, but I can not build your image because package u-boot-git-r1: task do_fetch: failed Sep 06 20:44:24 likewise: actually let me check if pcmciautils built or not Sep 06 20:44:30 KhemWork: The sanity checker can fail a build because it saw something strange in the build output. The task does succeed however. During a restart the task in not run again. Sep 06 20:44:38 s/in/is Sep 06 20:44:52 KhemWork: ... so the sanity failure does not trigger the second time. Sep 06 20:45:15 I'd really like to test it though.. but stuck because of this build issue Sep 06 20:46:36 likewise: actually it rebuilt the package Sep 06 20:46:42 so I think it was not the sanity checked Sep 06 20:46:49 s/checked/checker/ Sep 06 20:47:20 KhemWork: ok, that makes it stranger, because I only know of the sanity issue regarding restarts (zecke knows more on this, he wrote most of it) Sep 06 20:47:29 zecke is not here I think Sep 06 20:48:23 likewise: it could be but actually it was calling flex -lm before and dying in compile this time when it restarted the compile task went on to complete successfully Sep 06 20:51:25 Jin^eLD, uboot git suxors ..... Sep 06 20:51:40 Crofton: can I skip it somehow, I have the bootloader flashed Sep 06 20:51:45 so I do not really need to build it Sep 06 20:52:04 we need mkimage from it also ... Sep 06 20:52:27 Crofton: but mkimage can be build for uboot-utils, no? Sep 06 20:52:48 at least I remember using something like that for my other project where I also did not build uboot Sep 06 20:53:33 Crofton: just a sec; I'm near having a nice pretty much working thing you can mtn pull from my DB to get a working gumstix bootable image Sep 06 20:53:51 hughescr, I am about to head out the door :( Sep 06 20:54:04 anyone know how to get /dev/dsp to show up ? is it just a matter of including alsa-lib or what makes initscripts add it ? Sep 06 20:54:09 I will be in Toronto for a few days Sep 06 20:54:15 but I'll give it a try :-) Sep 06 20:54:18 I have an audio device that is looking for /dev/dsp Sep 06 20:54:21 and it isn't there Sep 06 20:54:22 let me know when it is ready Sep 06 20:54:26 Crofton: so.. any hints for me? it's kind of well, suxxors in the sense that I could and I want to test your image but can't build it :P Sep 06 20:54:29 can you email notes on pulling etc Sep 06 20:54:37 Crofton: np it'll be there when you get back :) Sep 06 20:54:56 I'll take it with me in case I have time to kill Sep 06 20:55:00 christopher: /dev/dsp requires not just alsa, but snd-pcm-oss Sep 06 20:55:19 ok. so, when I include mplayer, why wouldn't it require those too? Sep 06 20:55:23 snd-pcm-oss provides the bridge from OSS (which does /dev/dsp) to ALSA Sep 06 20:55:26 but that's good to know Sep 06 20:55:41 mplayer probably does things the alsa way if you have alsa configured Sep 06 20:55:53 instead of using the oss compatibility from snd-pcm-oss Sep 06 20:56:06 Crofton: yup will email Sep 06 20:56:07 ok. I don't know if I have alsa configured or not. Sep 06 20:56:15 the kernel is setup to use the AT91 SSC Sep 06 20:56:17 do you have /dev/sound Sep 06 20:56:20 or /proc/asouns Sep 06 20:56:31 no /dev/sound Sep 06 20:56:37 I mean /proc/asound Sep 06 20:56:41 but have /proc/asound Sep 06 20:56:47 anything in /proc/asound? Sep 06 20:57:07 cards, devices, oss, pcm, seq, timers, version Sep 06 20:58:34 anything in cards? Sep 06 20:58:49 nothing in cards Sep 06 20:58:57 I was just noticing that. not sure how to get it into cards Sep 06 20:59:03 ok, so sounds like you have the basic alsa modules loaded, but no actual card driver Sep 06 20:59:20 ok. hmmm. Sep 06 20:59:36 if it is just a matter of having a ssc connected through i2s to an audio dac...what is the card driver? Sep 06 20:59:49 that is, would the kernel declare that as a 'card' ? Sep 06 21:00:06 it'll depend on what the audio dac is Sep 06 21:00:13 hughescr, did I mention the network gets the link up, but I can not load stuff via it Sep 06 21:00:16 yes, the kernel considers the dac to be a card Sep 06 21:00:18 ping does not work Sep 06 21:00:25 Crofton: hmm Sep 06 21:00:29 you mean in u-boot or linux? Sep 06 21:00:44 u-boot Sep 06 21:00:48 ok, I'll check Sep 06 21:01:05 all I tested yesterday was that it detected the controller and stuff -- didn't actually plug it into a network :) Sep 06 21:01:10 :) Sep 06 21:01:11 no ethernet cable handy Sep 06 21:01:19 I had to go buy one Sep 06 21:01:53 I would just like to say the USAir reservation confirmation email suxs Sep 06 21:02:44 anyone know what's going on here? http://pastebin.com/m14643d9 Sep 06 21:04:12 two head Sep 06 21:04:29 in your branch Sep 06 21:04:36 hmm interesting, I wonder how that happened, given that I'm the only person developing on the branch Sep 06 21:04:37 you're right Sep 06 21:04:39 use mtn merge to merge them Sep 06 21:05:03 somewhere there is a viewmtn program that is helpful Sep 06 21:05:27 hey, looks like someone just pulled from me, seeing as how it seemed to think my sqlite DB was locked :) Sep 06 21:05:58 Is there any way to either replace sqlite with a more fine-grained locking DB, or to have mtn wait on the lock acquire instead of just failing Sep 06 21:05:58 ? Sep 06 21:06:45 I guess I can just replicate the online DB and then have 2 and sync one to the other Sep 06 21:08:48 yep I definitley screwed up the kconfig Sep 06 21:09:24 ok, no more work, need to make sure I have everything .... Sep 06 21:12:32 thanks for being patient toady, you guys have been a big help Sep 06 21:14:36 Crofton: hope you have a non-eventful flight Sep 06 21:14:47 seems the best you can hope for these dayas :-) Sep 06 21:21:09 koen: finished that epiphany .bb yet :-) Sep 06 21:41:07 03pH5 07org.oe.dev * rd4e02155... 10/ (1 conf/distro/include/preferred-xorg-versions-X11R7.3.inc): preferred-xorg-versions-X11R7.3: add preferred versions for X11R7.3 release Sep 06 21:41:13 03pH5 07org.oe.dev * r47566414... 10/ (1 packages/xorg-proto/inputproto_1.4.2.1.bb): inputproto: add 1.4.2.1 for X11R7.3 release Sep 06 21:41:21 03pH5 07org.oe.dev * r10ac4711... 10/ (1 packages/xorg-proto/renderproto_0.9.3.bb): renderproto: add 0.9.3 for X11R7.3 release Sep 06 21:41:34 03pH5 07org.oe.dev * r392abfc8... 10/ (1 packages/xorg-lib/libxcursor_1.1.9.bb): libxcursor: add 1.1.9 for X11R7.3 release Sep 06 21:41:52 03pH5 07org.oe.dev * r3e19de3d... 10/ (1 packages/xorg-lib/libxfont_1.3.1.bb): libxfont: add 1.3.1 for X11R7.3 release Sep 06 21:42:00 03pH5 07org.oe.dev * rb5c92d2c... 10/ (1 packages/xorg-lib/libxrender_0.9.4.bb): libxrender: add 0.9.4 for X11R7.3 release Sep 06 21:42:07 03pH5 07org.oe.dev * rc50efdec... 10/ (1 packages/xorg-lib/libxrandr_1.2.2.bb): libxrandr: add 1.2.2 for X11R7.3 release Sep 06 21:42:15 03pH5 07org.oe.dev * r0af69cdb... 10/ (1 packages/xorg-driver/xf86-input-keyboard_1.2.2.bb): xf86-input-keyboard: add 1.2.2 for X11R7.3 release Sep 06 21:42:23 03pH5 07org.oe.dev * r12e5af5c... 10/ (1 packages/xorg-lib/libxi_1.1.3.bb): libxi: add 1.1.3 for X11R7.3 release Sep 06 21:42:31 03pH5 07org.oe.dev * r44a6e8a5... 10/ (1 packages/xorg-proto/xf86dgaproto_2.0.3.bb): xf86dgaproto: add 2.0.3 for X11R7.3 release Sep 06 21:42:43 03pH5 07org.oe.dev * r4fcf7c8f... 10/ (1 packages/xorg-lib/libxxf86dga_1.0.2.bb): libxxf86dga: add 1.0.2 for X11R7.3 release Sep 06 21:42:58 03pH5 07org.oe.dev * rbfbb18da... 10/ (1 conf/distro/include/preferred-xorg-versions-X11R7.3.inc): preferred-xorg-versions-X11R7.3.inc: add version 2.0.3 for xf86dgaproto Sep 06 21:43:08 03pH5 07org.oe.dev * rad31767a... 10/ (1 packages/xorg-app/xset_1.0.3.bb): xset: add 1.0.3 for X11R7.3 release Sep 06 21:43:17 03pH5 07org.oe.dev * ra9c6803e... 10/ (1 packages/xorg-app/xmodmap_1.0.3.bb): xmodmap: add 1.0.3 for X11R7.3 release Sep 06 21:43:26 03pH5 07org.oe.dev * ra28fa232... 10/ (1 packages/xorg-app/xprop_1.0.3.bb): xprop: add 1.0.3 for X11R7.3 release Sep 06 21:43:36 03pH5 07org.oe.dev * r933a004c... 10/ (1 packages/xorg-app/xinit_1.0.5.bb): xinit: add 1.0.5 for X11R7.3 release Sep 06 21:43:44 03pH5 07org.oe.dev * r432646df... 10/ (1 packages/xorg-app/xdm_1.1.6.bb): xdm: add 1.1.6 for X11R7.3 release Sep 06 21:43:49 03pH5 07org.oe.dev * raa23c804... 10/ (1 packages/xorg-app/bdftopcf_1.0.1.bb): bdftopcf: add 1.0.1 for X11R7.3 release Sep 06 21:43:55 03pH5 07org.oe.dev * r4ec2c8cb... 10/ (1 packages/xorg-app/iceauth_1.0.2.bb): iceauth: add 1.0.2 for X11R7.3 release Sep 06 21:44:00 03pH5 07org.oe.dev * r1e4fbe1f... 10/ (1 packages/xorg-app/iceauth_1.0.2.bb): iceauth-1.0.2: fix DEPENDS Sep 06 21:44:01 pH5: :) Sep 06 21:44:11 03pH5 07org.oe.dev * rf4001b78... 10/ (1 packages/xorg-app/bdftopcf_1.0.1.bb): bdftopcf-1.0.1: fix DEPENDS Sep 06 21:44:26 03pH5 07org.oe.dev * r4cd29f48... 10/ (1 packages/xorg-app/ico_1.0.2.bb): ico: add 1.0.2 for X11R7.3 release Sep 06 21:44:35 03pH5 07org.oe.dev * r12c98ff4... 10/ (1 packages/xorg-app/xbacklight_1.1.bb): xbacklight: add 1.1 for X11R7.3 release Sep 06 21:45:13 pH5: don't flood the channel :-) Sep 06 21:45:34 flo_lap: it's easier to fix bugs if you make them yourself just before Sep 06 21:45:43 * flo_lap likes these kinds floods Sep 06 21:45:49 likewise: NOTE: package xserver-kdrive-1.4: failed Sep 06 21:45:58 heh Sep 06 21:46:00 no more flooding until that is fixed :) Sep 06 21:46:09 pH5: lol Sep 06 21:46:40 uh, old pixman version. next try.. Sep 06 21:48:17 ph5: there is a one-day old fix for pixman on the cairo mailing list for RGB24 stuff: http://mail.google.com/mail/?attid=0.0.0.1&disp=attd&view=att&th=114d3165b0ee0e2c Sep 06 21:48:39 ph5: uh, you can't access that link I guess Sep 06 21:48:47 lichtgestalt: probably not :) Sep 06 21:48:52 :-) Sep 06 21:49:44 ph5: it's on the cairo mailing list, in case you see strange effects around anti-aliasing and transparency Sep 06 21:52:59 likewise: ok, found it. I'll throw it in for good measure. Sep 06 21:53:50 ph5: if it applies the fix is probably needed, I guess you are working against a recent pixman Sep 06 21:55:02 yo Sep 06 21:55:37 likewise: yes, 0.9.5 is 11 days old Sep 06 22:16:14 I'm having a weird bitbake dependency thing going on.... Sep 06 22:16:15 http://pastebin.com/m249de336 Sep 06 22:16:46 for some reason, it's trying to pull in both linux-2.6.21 and linux-2.6.15 to satisfy building a module Sep 06 22:17:13 as well as a whole bunch of other kernels too Sep 06 22:17:52 Is there a reason we decieded to slow down parsing exponentially by trying to connect to every fetch url when building the cache? Sep 06 22:18:15 it is pretty horrifing to use with no net connection Sep 06 22:20:09 and now that it is done, I can not work because I have "errors" Sep 06 22:20:11 zerochaos: I noticed the same thing Sep 06 22:20:18 pretty ugly Sep 06 22:20:20 anyone help me with this? Sep 06 22:21:09 yeah, it appears to have been "a bad idea" Sep 06 22:22:21 hughescr: do you use bitbake collections? Sep 06 22:23:08 it appears to run svn info on every svn in the repo... Pretty nasty Sep 06 22:24:32 sorry, lost cell signal, I presume no response? Sep 06 22:24:51 nickserv ghost zerochaos password Sep 06 22:24:58 nice Sep 06 22:25:48 likewise: if I don't know if I'm using them, does that mean I'm using them or not using them :) Sep 06 22:26:02 hughescr: not. :-) Sep 06 22:26:49 this might have been a caching bug actually, I've -c rebuild'ed some stuff, and it might be working now... Sep 06 22:27:20 ah crap, I think my tmpdir is all hosed now Sep 06 22:27:24 time to rebuild from 0.... Sep 06 22:27:25 sigh Sep 06 22:30:53 just the fricking rm -rf takes a long time..... Sep 06 22:31:08 I deleted my tmp, and I still have the problem. I do not think it is me. Sep 06 22:31:29 hugh: and it does not fix the problem anyway Sep 06 22:32:01 zerochaos: I have two problems -- the one I'm trying to fix is related to bitbake trying to include like 8 kernels in my image... Sep 06 22:32:08 which I think was a caching problem Sep 06 22:32:40 jeez, I need faster disks Sep 06 22:33:13 ext3 is not known for deletion speed Sep 06 22:33:22 Zero_Chaos: the correct place to discuss the bitbake fetching behaviour is in #oe Sep 06 22:33:27 no-one here can change it Sep 06 22:33:53 mv tmp tmp_& then run a new bitbake in the foreground Sep 06 22:34:07 with an rm -rf tmp_ inbetween Sep 06 22:35:52 rwhitby: thought I was in #oe Sep 06 22:36:05 well, I'd say you both are ;) Sep 06 22:36:18 Zero_Chaos: indeed you are :-) my apologies :-) Sep 06 22:36:46 * rwhitby needs to separate his #oe and #openmoko windows more Sep 06 22:36:55 * nodens is thinking he needs faster disks as well Sep 06 22:37:00 * likewise too Sep 06 22:37:24 zerochaos: this is reiser3 Sep 06 22:37:32 rwhitby: happens to the best of us Sep 06 22:37:42 so, about fixing this... Sep 06 22:38:09 .zZ Sep 06 22:38:18 Zero_Chaos: koen posted a recipe for doing sane srcrevs - dunno whether that negates the need to do svn infos Sep 06 22:38:19 a dual core cpu and 2Gb ram would be nice as well... mhmm... I think I'll ask for that as a christmas present :) Sep 06 22:38:37 I actually sped up my buildroot rebuild-from-scratch process by having the whole buildroot directory be in a loop-mounted large file formatted with ext3 (no journalling) Sep 06 22:38:44 meanwhile, I'll continue reading userfriendly while building. Sep 06 22:39:34 I have xfs here... ext3 could be faster given the kind of data we are dealing with Sep 06 22:39:35 I've got dual opterons and 2GB of RAM, but the disk is just killing me Sep 06 22:41:22 a good sata raid controller... Sep 06 22:41:36 raid0 is nice for speed Sep 06 22:41:58 with a lot of cache Sep 06 22:44:17 I have a raid0 controller Sep 06 22:44:22 with really old slow disks Sep 06 22:45:05 mhmm... I know the I in raid is for inexpensive but it is sad to have a dual opteron and old disks :) Sep 06 22:47:05 nodens: they're old opterons ;) Sep 06 22:48:20 03pH5 07org.oe.dev * r08187172... 10/ (1 packages/xorg-lib/pixman_0.9.5.bb): pixman: add 0.9.5 for X11R7.3 release Sep 06 22:48:25 03pH5 07org.oe.dev * rf6e9631c... 10/ (4 files in 3 dirs): pixman-0.9.5: add upstream alpha copy fix Sep 06 22:48:30 03pH5 07org.oe.dev * rb9e53112... 10/ (5 files in 2 dirs): xserver-kdrive: move some patches out of -common inc, preparation for 1.4 Sep 06 22:52:36 hey good thing I'm complaining about my slow disks -- looks like SMART says I've got some failed sectors on one of them Sep 06 22:53:06 arf Sep 06 22:55:17 still no errors here, about 30 000 hours running... Sep 06 22:56:26 25k hours on each disk here Sep 06 22:56:28 bonnie++ says it's not so bad. But I always feel like the disks are slow. Sep 06 22:56:37 not as much cooling as I'd like on them Sep 06 23:02:06 thanks for the fun Sep 06 23:02:08 good nite Sep 06 23:10:37 when I said raid0 above, of course I meant raid1 Sep 06 23:10:52 raid1 can be slow Sep 06 23:10:59 raid0 without raid1 is for people who hate their data :) Sep 06 23:11:09 raid10, then Sep 06 23:11:38 but the tmp dir does not need to be on a raid1 array... Sep 06 23:12:03 all tmp dirs should be on a separate raid0 array (but that means many disks) Sep 06 23:12:17 (and a controller with two channels) Sep 06 23:23:21 hughescr: sorry if I messed you up by trying to pull! Sep 06 23:23:49 sakoman: no problem Sep 06 23:24:05 not 100% sure my branch is building cleanly right now. building to check Sep 06 23:24:11 let me know when it is safe :-) Sep 06 23:24:35 did you ever figure out the secret to consistently getting modules.dep generated in the build? Sep 06 23:25:09 Seems to be inexplicably random for me Sep 06 23:25:21 But mostly not working now Sep 06 23:26:05 sakoman: nope haven't really looked at it Sep 06 23:26:29 just buy some 16GB sd disk just for building Sep 06 23:26:32 it gets generated at first boot & all is well Sep 06 23:26:47 but the next reboot dies a horrid death :-( Sep 06 23:27:22 I'd just been working under the assumption that I'd just force a depmod -a on boot Sep 06 23:27:22 before anything else happens Sep 06 23:27:40 have you tried a reboot? Sep 06 23:54:05 I just wonder if there should be 'ipkg' available in angstrom-minimal-image Sep 07 00:04:44 i was thinking that recently :) Sep 07 00:45:00 sakoman, ping Sep 07 02:45:31 aa Sep 07 02:48:58 Crofton: belated pong! **** ENDING LOGGING AT Fri Sep 07 02:59:56 2007