**** BEGIN LOGGING AT Fri Apr 07 09:59:57 2006 Apr 07 10:01:11 but they don't, they use font-sizes like the rest of matchbox does. And without the bug, the font is perfectly readable. With out the bug a font-size of "7" is pretty big Apr 07 10:01:56 CoreDump|home: does XXpx not mean pixels, not font size? Apr 07 10:02:05 CoreDump|home: all the themes use XXpx Apr 07 10:02:23 hmmm Apr 07 10:02:27 dunno actually Apr 07 10:03:02 for example in blondie Apr 07 10:03:04 I can only say that once every blue moon, the bug disappears, and then the title text is the same as the desktop font Apr 07 10:03:23 remove the px from that line and blondie looks fine at 200dpi Apr 07 10:03:59 we would need to use _MACHINE overrides then so that the other machines are not affected Apr 07 10:04:23 CoreDump|home: what we need to to remove those and see if they work on 100DPI machine Apr 07 10:04:28 * XorA has no access to one Apr 07 10:04:44 I have one Apr 07 10:04:55 can do so when I get back from work today Apr 07 10:05:18 cool, that was the point I got to with "fixing" inability to test at 100dpi Apr 07 10:06:16 XorA: in a perfect world, the bug in matchbox would get fixed ;) Now we need to put up with these ugly hacks :\ Apr 07 10:08:38 http://handhelds.org/scap/port.24807.png <- this is X at 100 DPI with fixed panel Apr 07 10:09:07 hmmm Apr 07 10:09:13 matchbox panel is broken Apr 07 10:09:18 it ships with the wrong icons Apr 07 10:09:37 i was wondering about the panel icon, yeah Apr 07 10:09:44 thought the use a new one now Apr 07 10:10:46 hmmm Apr 07 10:10:54 mb-panel defaults to mig icons in OE Apr 07 10:11:14 oh dear Apr 07 10:11:24 I hope matchbox did change defaults..... Apr 07 10:11:50 CoreDump|home: see http://pastebin.com/645784 Apr 07 10:12:18 heh Apr 07 10:12:39 * CoreDump|home didn't see any icon OE patches when he added .9.3 Apr 07 10:12:55 so it's probably coming from upstream Apr 07 10:13:01 damnit Apr 07 10:17:47 03koen 07org.oe.dev * r3a5a5c33... 10/packages/matchbox-panel/matchbox-panel_0.9.3.bb: matchbox-panel 0.9.3: default preference = -1 due to segfault and iconsize issues Apr 07 10:19:30 problem is that this version fixes the panel-menu problems with regards to font-size Apr 07 10:21:05 oops Apr 07 10:21:29 ~lart winfixer Apr 07 10:21:29 * ibot stabs winfixer Apr 07 10:22:28 koen: I've just flashed a clean image with .9.3. The icon is back to normal. The old image was screwed for whatever reason Apr 07 10:25:02 koen: do you know if glibc-2.4 also has some support for non-EABI softfloat ? Apr 07 10:27:30 koen: it misses bits/fenv.h, details here http://pastebin.ca/48509 Apr 07 10:27:38 schurig: it should have, but pb_ would be the authorative source Apr 07 10:27:50 schurig: ah, the glibc .bb has a remark about fenv.h stuff Apr 07 10:28:00 koen: I've only seen you and ph5 in the last few monotone logs :-) Apr 07 10:29:26 koen: my idea is to make a non-EABI image with softfloat, but with glibc 2.4 and gcc 4.1. Somehow I perceive EABI as too experimental right now. Maybe I'm wrong with that opinion ... Apr 07 10:29:45 schurig: http://bugs.openembedded.org/show_bug.cgi?id=741 Apr 07 10:30:01 schurig: you can't even get real EABI yet Apr 07 10:30:12 schurig: since real EABI requires thumb-interwork to be enabled Apr 07 10:33:17 hrw|work: OZ is moving to berlios? Apr 07 10:34:06 koen: only wiki will be there Apr 07 10:34:35 I see Apr 07 10:34:41 hrw|work: oh, not sure if that is an improvement. Berlios is sometimes sloooooow Apr 07 10:35:05 koen: we have people who want to write docs, which is good Apr 07 10:35:05 schurig: I'm not docs manager - lardman|work is Apr 07 10:35:29 hrw|work: at least it was 4 months ago, maybe they have improved. Usually the wiki was slow at then evenings (middle european time) Apr 07 10:35:48 schurig: we'll have to see how things work out Apr 07 10:36:22 lardman|work: I used to work in the wiki of the destar project at berlios Apr 07 10:36:36 schurig: ah, okay Apr 07 10:36:40 does sysvinit have any sort of debug flag? Apr 07 10:37:14 lardman|work: just make some test page below http://openfacts.berlios.de/index-en.phtml?title=DeStar and access it at 19:00 MEST Apr 07 10:38:02 lardman|work: but maybe it was just openfacts that was slow. And maybe they fixed that. I had page load times with 10 seconds then Apr 07 10:38:21 (right now it's fast) Apr 07 10:38:29 TheMasterMind1: iirc it does Apr 07 10:38:35 speaking of init... Apr 07 10:38:43 schurig: Yes, it's fast now. We'll have to see how things go. As you say, they may have fixed things Apr 07 10:38:47 who wants to add bootchart support to OE? Apr 07 10:38:56 schurig: hopefully anyway ;) Apr 07 10:39:04 koen: who wants to add InitNG to OE? Apr 07 10:39:59 bootchart is fud Apr 07 10:40:07 s/fud/fluff Apr 07 10:40:08 koen: I'm using initng on Debian SID here and at home, the software is quite young (and you see it), but it's fantastic fast, e.g. 8 seconds from "Freeing init memory" to start of kdm Apr 07 10:40:25 schurig: I do, but then I realised that we need to adapt each package that ships an initscript Apr 07 10:40:29 koen: on some mere 1 GHz Athlon machine Apr 07 10:40:37 schurig: 15s on my home machine and it can forget abot many things ;( Apr 07 10:40:47 schurig: I can get faster to kdm (if that is the goal) Apr 07 10:40:51 schurig: at work if I use initng then cups (started) does not work Apr 07 10:41:03 zecke: init=/usr/bin/kdm? Apr 07 10:41:12 hrw|work: hmm, I have cups running here and at home, no problem Apr 07 10:41:18 hrw|work: no, I patch the kernel (1tic faster) Apr 07 10:41:27 hrw|work: I think k* belongs into the kernel Apr 07 10:41:37 hrw|work: that would be the ultimate. Thrash away those old style daemons, and write them in some proper way Apr 07 10:42:20 ~lart linux anyway Apr 07 10:42:20 * ibot teaches linux anyway the basics, including how to RTM Apr 07 10:42:43 * TheMasterMind1 grrs at sysvinit Apr 07 10:43:29 * Bernardo recalls all the problems he has had until now with simpleinit Apr 07 10:54:23 koen: your comment http://bugs.openembedded.org/show_bug.cgi?id=741#c8 doesn't work for me, because there is no mp_clz_tab_* in ports/sysdeps/arm Apr 07 11:00:05 koen: oh, it see, it's not there at do_munge time, but get's placed there at do_patch time Apr 07 11:01:34 koen: arm-longlong.patch adds it, but it expands to "/* nothing */" if __ARM_ARCH_5TE__ is defined, which should be for Intel XScale PXA25x Apr 07 12:03:21 03coredump 07org.oe.oz354x * r730b0eb0... 10/packages/keylaunch/ (3 files in 3 dirs): keylaunch: Enable Softkeys for SL-Cxxxx Apr 07 12:03:25 03coredump 07org.oe.oz354x * r7f808bc9... 10/packages/keymaps/ (3 files in 3 dirs): keymaps: Map softkey-keycodes to functionkeys for SL-Cxxxx Apr 07 13:30:00 all: anyone tried making a .bb for librsvg from CVS? It uses libcairo instead of libart AFAIK. Apr 07 13:32:59 likewise: not yet Apr 07 13:33:08 zecke h3630,3635,3641,3650,3660,3670,3730,3760,3765,etc thats 9 that HP still has support docs for Apr 07 13:33:19 likewise: on x86 libcairo should be faster, but I'm not sure on arm Apr 07 13:34:00 koen: ok, maybe I'll take the job. I think librsvg moves towards libcairo for future development. (and then we need to make cairo faster on ARM maybe). Apr 07 13:34:36 likewise: I'm still hoping for crunch support in libgcc :) Apr 07 13:34:42 cairo isnt exactly fast, hence it doenst have full support in e17 Apr 07 13:35:09 emte: depends on your backend Apr 07 13:35:22 emte: on my machine the glitz backend is *super* fast Apr 07 13:35:24 this is raster we are talking about Apr 07 13:35:34 the super fanatic about rendering speed Apr 07 13:36:37 koen: you got a EP93xx around as well then? Apr 07 13:37:24 emte: e17? Apr 07 13:37:31 enlightenment Apr 07 13:37:44 likewise: yes Apr 07 13:37:58 emte: ok, I am an embedded guy, not using X much Apr 07 13:38:02 emte: okay... Apr 07 13:38:25 emte: Opie doesn't need glibc locales available, so for 50% of familiar it is not a Show stopper Apr 07 13:38:26 zecke, you questioned me on the `10+ statement Apr 07 13:38:35 koen: still looking for an embeddable GPU that supports OpenGL :-) Apr 07 13:38:45 emte: well, these are two hardware revisions you cite Apr 07 13:39:03 likewise: OpenGL ES or how this embedded spec is named? Apr 07 13:39:04 yes Apr 07 13:39:47 likewise: i2700g from intel? Apr 07 13:39:58 likewise: dell uses it in their x5* series Apr 07 13:40:38 likewise: iirc it's inside the pxa272 SoC Apr 07 13:41:00 zecke: I do not know if Glitz supports OpenGL ES yet... but yes it is interesting. Apr 07 13:41:08 koen: will look into i2700g Apr 07 13:41:34 OpenGL ES is mostly an (s/w) API spec, AFAIK (but I have not looked into it deeply) Apr 07 13:42:10 koen, what other distros are you running? Apr 07 13:42:11 likewise: and TI claims that omap2 and omap3 also support 3d (OGL/ES) Apr 07 13:42:37 emte: OZ, angstrom and some custom in-house ones Apr 07 13:43:01 when has OZ ran on a h36? Apr 07 13:43:23 when you say DISTRO="openzaurus=3.5.4" in local.conf Apr 07 13:43:42 that's the point of OE: distros and machines are orthogonal Apr 07 13:43:51 hmm, the last time i tried that it didnt like me Apr 07 13:44:47 hi all Apr 07 13:45:40 hey cedric Apr 07 13:45:45 just wondering if the librsvg 2.11.1 is fixed in .dev Apr 07 13:46:28 what is broken? Apr 07 13:47:15 can't build it, can't resolve external entities to xsl files Apr 07 13:47:44 cedric: is there a bug report already? Apr 07 13:47:55 nope Apr 07 13:47:59 :-) Apr 07 13:48:02 consider creating one Apr 07 13:48:22 bugtracker.... we definitelly need bugsquashing day(s) Apr 07 13:48:30 cedric: some one just talked about librsvg... Apr 07 13:48:32 hrw|work: yes! Apr 07 13:49:12 zecke: this will also show does my machine is stable for builds... Apr 07 13:49:31 I do not remember exactly when I build something not on ewi. Apr 07 13:50:41 is the libtool patch already in? Apr 07 13:50:45 yes Apr 07 13:50:47 * koen has been out of the loop Apr 07 13:50:49 cool Apr 07 13:51:01 gtk now sucks a bit less (but still) Apr 07 13:51:09 can't wait to test my first gpe image with kernel 2.6 for poodle... Apr 07 13:51:45 hrw|work: rm_work doesn't break anymore? Apr 07 13:52:07 koen: it sometimes break by its own Apr 07 13:52:37 koen: and iirc pango and/or gtk libtool still need editing by hand (but its better then it was) Apr 07 13:53:14 zecke: where about that "someone"? Apr 07 13:53:31 cedric: likewise wanted to build librsvg cvs - not more Apr 07 13:54:21 ~ewi Apr 07 13:54:22 hmm... ewi is ewi546.ewi.utwente.nl - main backup server for OE monotone. Hosts database snapshot: http://ewi546.ewi.utwente.nl/OE/OE.db.bz2 Apr 07 13:55:11 koen: can you setup rsyncd on ewi? Apr 07 13:55:38 hrw|work: isn't ssh+rsync enough? Apr 07 13:55:59 hrw|work: I can't Apr 07 13:56:29 ok Apr 07 14:02:44 someone tried to build minimo from .dev? Apr 07 14:14:34 hostap-modules RPROVIDES = "virtual/kernel-hostap" - they do not need to PROVIDE it or need to? Apr 07 14:15:10 someone? Apr 07 14:15:44 this depends on whether something else depends on it or not Apr 07 14:16:04 hi mickeyl Apr 07 14:16:09 hi hrw Apr 07 14:17:41 machines in .oz354x depend/rdepend on hostap-modules(-cs) or kernel-hostap(-cs) Apr 07 14:18:47 hostap-daemon only RDEPEND on "virtual/kernel-hostap" Apr 07 14:19:04 hey mickeyl Apr 07 14:19:24 but I will do PROVIDES Apr 07 14:20:43 mickeyl: thanks for adding gperiodic Apr 07 14:22:39 np, it was a four liner :) Apr 07 14:23:18 mickeyl: the .po file I made for that was a bit longer :) Apr 07 14:23:24 03hrw 07org.oe.dev * rdf9cea22... 10/packages/gpe-wlancfg/ (gpe-wlancfg-makefile.patch gpe-wlancfg_0.2.6.bb): drop gpe-wlancfg - gpe-conf is used to configure network Apr 07 14:23:28 03hrw 07org.oe.dev * raea2bd16... 10/packages/hostap/ (6 files): Apr 07 14:23:28 hostap-modules: hostap-modules-cs now RDEPEND on hostap-conf Apr 07 14:23:28 - its easier to update only hostap-conf when new card is added Apr 07 14:23:40 Thu Jul 29 12:18:05 EDT 1999 - Version 1.2.5 is avilable in the usual place. This version adds Dutch support provided by: Koen Kooi. Apr 07 14:23:46 my first contribution to open source :) Apr 07 14:23:51 ;) Apr 07 14:24:10 cedric: yes, wanted to build librsvg cvs (which now depends on libcairo instead of libart afaik) Apr 07 14:24:21 koen: Congratulations :-) Apr 07 14:26:02 hrw|work: I suspect there is a DEPENDS issue for hostap in the branch which won't exist in .dev due to the RDEPENDS revamp in .dev Apr 07 14:26:29 RP: in subject of last revision? Apr 07 14:27:08 RP : where shall i searsh to desactive crosshair test ? Apr 07 14:27:09 hrw|work: No, the PROVIDES issue you mentioned Apr 07 14:27:16 matchbox or kdrive config ? Apr 07 14:27:22 RP: linux-openzaurus.inc also PROVIDES it Apr 07 14:27:29 Genesis: xserver-common-kdrive IIRC Apr 07 14:27:37 thx i look Apr 07 14:27:40 Genesis: If you're using poky that is Apr 07 14:28:03 yeap i use it for the moment Apr 07 14:28:13 i like it. Apr 07 14:28:35 hrw|work: In .dev we should only need it as a runtime entity? Apr 07 14:29:29 hrw|work: Unless hostap-utils has to build against it? Apr 07 14:30:15 RP: hostap-utils probably does not need hostap modules - but is useless without them Apr 07 14:30:46 hrw|work: It should just have a RRECOMMEND then and we shouldn't have any DEPENDS anywhere? Apr 07 14:31:30 RP: looks like it Apr 07 14:33:05 I want to copy some stuff from .oz354x to .dev Apr 07 14:35:41 likewise: did u succeed? otherwise there a version form the 01/04/2006 --> 2.14.3 Apr 07 14:35:56 s/there/there is Apr 07 14:39:26 03hrw 07org.oe.oz354x * r45a167d4... 10/packages/zbedic/ (libbedic_1.0.bb zbedic_1.0.bb): zbedic: upstream released source tarballs so we switch to using it Apr 07 14:39:31 03hrw 07org.oe.dev * r629ed29d... 10/packages/zbedic/ (9 files in 3 dirs): zbedic: updated to 1.0 Apr 07 14:45:25 03hrw 07org.oe.dev * rbb216866... 10/packages/zaurus-updater/ (tosa/updater.sh zaurus-updater.bb): zaurus-updater: added zImage.bin to tosa updater.sh Apr 07 14:50:15 ~lart http://www.ripe.net/fcgi-bin/webupdates.pl Apr 07 14:50:16 * ibot steals http://www.ripe.net/fcgi-bin/webupdates.pl's mojo Apr 07 14:54:50 03hrw 07org.oe.dev * r63e51933... 10/packages/groff/ (groff_1.19.1.bb groff_1.19.2.bb): Apr 07 14:54:50 groff: fixed paths so manpages are packaged Apr 07 14:54:50 - disabled PARALLEL_MAKE Apr 07 14:56:21 03hrw 07org.oe.dev * r2667e0b5... 10/packages/otpkeygen/otpkeygen_1.3.0.bb: otpkeygen: install correct binary Apr 07 15:00:48 debian sid has changed its use of wpa_supplicant Apr 07 15:01:15 the "new way" is to start wpa_supplicant daemon on demand, before interface goes up Apr 07 15:02:11 cedric: sorry I am not buildling now, but I will see about librsvg 2.4.13 Apr 07 15:02:25 cedric: thanks for the pointer Apr 07 15:03:09 cedric: did you mean 2.4.13 is in OE? I updated yesterday, but I have 2.11.1 Apr 07 15:03:25 03hrw 07org.oe.dev * r8a44e2ba... 10/packages/meta/ (6 files): Apr 07 15:03:25 bootstrap/gpe/opie/sectest-gpe/e/pivotboot images: dropped openzaurus/opensimpad feed urls Apr 07 15:03:25 - they pointed to openzaurus.org which does not have feeds Apr 07 15:03:25 - .dev is not for release stuff Apr 07 15:05:37 like|gaming: sorry 2.14.3, need to add a new bb file, I'll try tonite/this weekend Apr 07 15:06:12 ~lart 'find' on ewi Apr 07 15:06:12 * ibot slams 'find' on ewi against a large cement Tux Apr 07 15:06:18 hrw@bitbake:~/devel/oe/org.openembedded/packages$ find initscripts/initscripts-1.0/ -name keym* Apr 07 15:06:22 find: paths must precede expression Apr 07 15:06:25 what is wrong? Apr 07 15:06:38 hrw|work: -name 'keym*' Apr 07 15:06:50 argh.. in sid it work without '' Apr 07 15:06:58 thx Apr 07 15:08:12 it'll work without '' if you don't have any files matching "keym*" in the current directory Apr 07 15:08:39 I don't think the version of find actually has anything to do with it. Apr 07 15:09:03 ah.. ok Apr 07 15:14:45 hrw|work: in the image I built last night (oz354x) the cardmgr applet seems to have stopped working with the transition to pcmciautils Apr 07 15:15:08 Bernardo: yep - we need to disable it Apr 07 15:15:33 :( Apr 07 15:16:18 we don't have a alternative yet, right? Apr 07 15:24:54 03hrw 07org.oe.dev * r94b56490... 10/ (33 files in 20 dirs): (log message trimmed) Apr 07 15:24:54 keymaps handling moved from initscripts to keymaps package Apr 07 15:24:54 - loading keymap require loadkeys from console-tools (which takes 300K) Apr 07 15:24:54 - all Zaurus machines (except collie) will get this package installed Apr 07 15:24:54 - collie 2.4 keymap is the same as kernel one so it is not needed on device Apr 07 15:24:56 anyway users can install it if want to make some changes in mapping Apr 07 15:24:58 Zaurus machines conf: dropped console-tools from RDEPENDS, added keymaps instead Apr 07 15:30:51 03hrw 07org.oe.oz354x * r207ae6e0... 10/packages/meta/meta-opie.bb: Apr 07 15:30:51 meta-opie: install opie-pcmciaapplet only on 2.4 machines Apr 07 15:30:51 - 2.6 ones will get pcmciautils and opie-pcmciaapplet needs pcmcia-cs Apr 07 15:32:37 enough for today Apr 07 15:33:11 I think that most of my 3.5.4x work in now in .dev too - if something is fscked then someone will notice and complain Apr 07 15:33:32 hrw|work: Looks good to me - all nice to have in .dev :) Apr 07 15:34:10 RP: we need to think how to handle switch "pcmciautils" <> "pcmcia-cs hostap-conf orinoco-conf" Apr 07 15:34:58 bluez-utils-(no)dbus contain /etc/pcmcia/bluetooth.conf which is also not needed with pcmciautils rather... and other pcmcia.conf files Apr 07 15:35:13 ok time to go home - cu all Apr 07 15:35:18 bye hrw|work Apr 07 15:35:22 hrw|work: In bitbake.conf we declare PCMCIAUTILS as "pcmciatutils" and PCMCIACS as "pcmcia-cs hostap-conf orinoco-conf" ? Apr 07 15:35:33 cu hrw|gone Apr 07 15:35:44 RP: sounds ugly Apr 07 15:35:45 cu Apr 07 15:36:18 hrw|gone: task-pcmcia and task-pcmciacs perhaps... Apr 07 15:50:51 bonne soiree Apr 07 16:27:32 so when is our bug squashing day? Apr 07 16:28:21 zecke: propose a date on oe@ Apr 07 16:31:13 during a weekend is best :) Apr 07 16:31:24 I'm currently trying to generate reports... Apr 07 16:35:11 zecke: looking at the last few weeks, it seems that people are most active during weekdays Apr 07 16:35:59 http://bugs.kde.org/reports.cgi?product=konqueror&output=show_chart&datasets=NEW%3A&datasets=ASSIGNED%3A&datasets=REOPENED%3A&datasets=UNCONFIRMED%3A&datasets=RESOLVED%3A&banner=1 Apr 07 16:37:10 treke|home: ping Apr 07 16:51:10 heh Apr 07 16:51:12 "If it worked, we wouldn't need a bug reporting system." Apr 07 16:56:56 koen: For bugfixing, a weekend or evening would be best for me as otherwise it will be tricky with work :-/ Apr 07 16:57:41 what is our goal? Apr 07 16:57:51 Close Bugs? What kind of bugs? Apr 07 16:58:04 Generally improve infrastructure to find bugs more easy? Apr 07 16:58:18 zecke: Apply patches, fix reported bugs and try and reduce the amount of open bugs Apr 07 17:02:40 RP: fridays are good, or do you put deadlines in fridays? Apr 07 17:04:52 zecke: Fridays are bad for me as usually about now I'm in a pub ;-) Apr 07 17:05:05 RP: no Wireless in your pub? Apr 07 17:05:16 zecke: Unless you think I fix bugs better when I'm not sober :) Apr 07 17:05:17 RP: by what time are you pissed? Apr 07 17:05:29 usually... Apr 07 17:05:45 zecke: Arguably, some of my best code has been written on Friday evenings mind ;-) Apr 07 17:05:58 Or early Saturday morning :-/ Apr 07 17:06:21 zecke: No wireless I've managed to break into yet... Apr 07 17:06:50 OT: Is the o-hand crew scattered or do you have one office where you all sit together? Apr 07 17:07:00 zecke: I'm usually back by 10pm but not entirely sober :) Apr 07 17:07:05 zecke: scattered Apr 07 17:07:37 zecke: One person is 100 miles north, the next nearest 300 miles south :) Apr 07 17:07:47 "usually" means I'll be leaving shortly btw ;-) Apr 07 17:11:10 http://www.oesf.org/forums/index.php?showtopic=18700&hl= Apr 07 17:11:14 what do you think about that? Apr 07 17:12:56 mickeyl: let's start our own research institute :) Apr 07 17:13:10 koen: on my way ;) Apr 07 17:13:18 mickeyl: I wonder what the difference is betweeen their distro and 'MACHINE=n30 bitbke gpe-image' Apr 07 17:13:31 koen: the difference is the "hh.org build scripts" Apr 07 17:13:33 *cough* Apr 07 17:14:09 considering that they base on GPE they should donate some percents of their sales to hh.org Apr 07 17:14:15 or the GPE team, for that matter Apr 07 17:14:26 imo Apr 07 17:14:26 or send a mail to gpe@ Apr 07 17:14:44 then again Apr 07 17:14:46 ~emulate zautrix Apr 07 17:14:48 this is exactly what the GPL is about. grab the stuff and run away Apr 07 17:14:50 heh Apr 07 17:15:07 just point TT to their site Apr 07 17:15:30 I bet they'll get all redfaced about the 'x is faster as framebuffer' Apr 07 17:15:36 hehe, yep Apr 07 17:15:57 ah well Apr 07 17:16:07 angstrom still has the coolest installer ;) Apr 07 17:16:08 he what about the award winning Qtopia? Apr 07 17:16:17 Opie died on his demand anyway... Apr 07 17:16:34 yeah. we let it die just because of his editorial and now... he doesn't even use qtopia Apr 07 17:16:35 what a shame Apr 07 17:16:48 ~lart guylhem Apr 07 17:16:49 * ibot takes a large goose feather pillow and swings it wildly in guylhem's direction, hitting guylhem and sending guylhem flying into the closet Apr 07 17:17:02 03mickeyl 07org.oe.dev * rc0c009a8... 10/packages/konqueror/konqueror-embedded_svn.bb: konqueror-embedded-svn: fix DEPENDS line Apr 07 17:18:16 hrw|gone: ping Apr 07 17:18:25 ~lart guylhem Apr 07 17:18:25 * ibot takes guylhem to the vet for a "special" visit Apr 07 17:18:44 hrw|gone: if you have a bit of time... could you cook konq_svn + qt-mt-static and have a go at that w/ 3.5.4 (or 3.5.4.x) Apr 07 17:27:14 * RP -> pub :) Apr 07 17:27:24 RP: enjoy :) Apr 07 17:44:15 27000848 2006-04-06 21:41 opie-image-c7x0-20060406094017.rootfs.img Apr 07 17:44:18 hmm Apr 07 17:44:21 how could that get so large? Apr 07 17:44:36 this breaks c700 and c750 Apr 07 17:46:01 check the ipkg status file Apr 07 17:47:32 mickeyl: modules? Apr 07 18:04:47 possible. need to investigate Apr 07 18:18:59 http://www.letux.org/releases/Quick%20Start%20Letux%20on%20Acer%20n30.pdf Apr 07 18:19:00 lol Apr 07 18:19:09 they refer to their site as letox.org Apr 07 18:21:56 I noticed Apr 07 18:22:04 I also noticed you can't buy it yet Apr 07 18:22:51 hmm where is Qtopia4.1 support for LeTux Apr 07 18:22:55 I want VoIP :} Apr 07 18:23:07 zecke: V=Voice? Video? Apr 07 18:26:49 mickeyl: do you want me to try building konq_svn for 3.5.4.1? Apr 07 18:27:41 * Bernardo wonders which head should he choose, hrw's or coredump's one Apr 07 18:28:01 externe.net is also... Apr 07 18:28:57 koen: "Thanks to X11 support, is offers a faster display than most framebuffer based systems equipping other free software handhelds." Apr 07 18:29:08 koen: So X is faster because it does not use a FrameBuffer? Apr 07 18:29:27 zecke: maybe the n30 has a real video chip with x accell drivers Apr 07 18:30:35 zecke: It's kinda susupicous trolltech doesn't publish benchmarks and comparisons Apr 07 18:31:02 koen benchmarks can be faked Apr 07 18:31:28 yeah, TT would obviously need to ly ;) Apr 07 18:31:31 lie Apr 07 18:31:33 hrm Apr 07 18:31:34 puh Apr 07 18:32:37 koen: I think QtE is more efficient that kdrive Apr 07 18:32:54 koen: see the Stuff you do with EXA and the issues you have... buffer management Apr 07 18:33:07 koen: QtE's buffer support is way better ;) Apr 07 18:33:25 zecke: could be, show me some real numbers of repeatable benchmarks Apr 07 18:33:39 proof by hand waving does not impress me Apr 07 18:33:41 thats what you should do Apr 07 18:34:50 * koen is thinking about patching http://cvs.gnome.org/viewcvs/beagle/Filters/FilterEbuild.cs?annotate=1.3 to support bitbake recipes Apr 07 18:48:11 ljp: any news on Qtopia4.1 GPL? Apr 07 18:50:45 strange and weird, the alphagrip is Apr 07 19:13:23 hmm, an SD card is hardly "pre-installed" on a device Apr 07 19:43:45 joshua_, that would be the same thing as how most sim devices work would it not? Apr 07 19:45:47 my guess would be that the term refers to shipping conditions Apr 07 19:46:06 ex. digital cameras, cellphones etc Apr 07 19:46:24 joshua_: maybe it has a special bootloader flashed ;) Apr 07 19:46:47 joshua_: maybe you can't remove the SD card without violating DMCA or DRM systems... Apr 07 19:47:02 lol Apr 07 19:47:27 i would think its just extra storage so they can advertise it as a "media player" as well Apr 07 19:48:32 RP_: where do you host rpsys.net? Apr 07 19:51:48 emte, no, no, it evidnetly boots off of the SD, and otherwise runs windows Apr 07 19:52:06 "it runs Linux natively! it's preinstalled!" ... if you boot from SD Apr 07 19:53:03 joshua_: if we are talking about letux.org, you can remove ppc and install linux as well Apr 07 19:53:09 joshua_: it was somewhere on the site as well Apr 07 19:54:41 hmm Apr 07 19:57:10 ahh Apr 07 19:57:28 os on a chip systems Apr 07 20:00:07 hi Apr 07 20:00:44 mickeyl: I can check konq/e 3.5 but probably not during weekend Apr 07 20:01:22 mickeyl: which srcdate is good to test? 20060102 is a bit old probably.. Apr 07 20:01:48 hrw|husky: hi. 20060404 Apr 07 20:02:04 if that works, we will pin it down Apr 07 20:02:09 ok Apr 07 20:02:53 dcopidl-native is in oe or has to be assumed? dont know does it is on ewi or not Apr 07 20:03:22 * hrw|husky does not have 354x builddir on own machines Apr 07 20:03:50 ~hail anonymous donor for pretec wifi card Apr 07 20:03:58 * ibot bows down to anonymous donor for pretec wifi card and chants, "I'M NOT WORTHY!!" Apr 07 20:04:25 hrw|husky: oh, I fear it is not on ewi Apr 07 20:04:49 ok. can do build at work Apr 07 20:06:00 first some rm.. Apr 07 20:06:44 whichi debian package provides dcip-idl? Apr 07 20:06:46 which* Apr 07 20:06:51 someone know how much does portrait vga linuxpda costs now? Apr 07 20:07:01 hrw|husky: a lot Apr 07 20:07:20 hrw|husky: unless you buy a dell which will break down after a week Apr 07 20:07:48 koen|tv: kdelibs4-dev Apr 07 20:07:55 really a dell is so fragile ? Apr 07 20:08:14 ok. so ill stay with collie/shit and c760 Apr 07 20:08:20 gremlin[it]: dell is crap - regardless of what they produce Apr 07 20:08:35 but dell is good for p0rn Apr 07 20:08:48 and porn is good, so dell is good Apr 07 20:08:50 Get:19 http://ftp.nl.debian.org unstable/main kdelibs4-dev 4:3.5.2-2 [1388kB]Fetched 13.7MB in 1s (11.6MB/s) Apr 07 20:09:04 koen|tv: a bit slow... :) Apr 07 20:09:11 :) Apr 07 20:09:23 they still won't allow me to use the Gbit interface Apr 07 20:09:46 mhhh i have a dell laptop and never had problem ... also desktop machine ad university was good .... Apr 07 20:10:12 zecke: nice monologue on #opie Apr 07 20:10:51 hrw|husky: hehe, is he on your ignore list Apr 07 20:10:59 koen|tv: they do not want you to saturate connection Apr 07 20:11:12 zecke: l1nux or nesse? :) Apr 07 20:11:15 * koen|tv wonders how to saturate 20Gbit/s Apr 07 20:11:32 hrw|husky: l1nux Apr 07 20:11:34 | /home/ich/hacking/embedded/oe/builds/c7x0/tmp/work/i686-linux/quilt-native-0.42-r0/temp/run.do_compile.9566: line 629: make: command not found Apr 07 20:11:36 koen|tv: 2 networds cards.. Apr 07 20:11:43 heh Apr 07 20:12:35 RP_: may I extend your sanity bbclass? Apr 07 20:14:14 RP_: speaking about pcmcia-cs/pcmciautils.. i would rather use base_conditionals in machine confs and move pcmcia_manager into distro conf Apr 07 20:15:00 mickeyl: pcmciautils does not need /etc/pcmcia/*.conf files - right? Apr 07 20:16:27 right Apr 07 20:16:58 hrw|husky: do you have any idea why recent opie-image in .dev is so chunky? Apr 07 20:17:03 26MB Apr 07 20:17:14 doesn't fit a c700 or c750 Apr 07 20:17:33 RP, mickeyl: but if we use pcmciautils then hostap-conf/orinoco-conf will not be needed due to mapping in driver then: 1. some users will complain that their cards are not working (lack of map) 2. how to deal with hostap modules rdepending on hosatp-conf? Apr 07 20:17:41 mickeyl: kernel-modules Apr 07 20:17:52 hrw|husky: darn. what can i do to remove those? Apr 07 20:17:54 mickeyl: rebuild kernel if not latest Apr 07 20:18:11 i did that. why do we have DEFAULT_PREFERENCE=-1 in the 2.6.16 one? Apr 07 20:18:14 iirc RP fixed that today Apr 07 20:18:35 ah Apr 07 20:18:35 mickeyl: update :) Apr 07 20:18:35 thanks Apr 07 20:18:40 * mickeyl mt pull Apr 07 20:19:46 l8er Apr 07 20:20:06 jezz.. its pulling here (271 revs) sine 16... Apr 07 20:20:14 hrw|husky: we need to do some thorough tests re. the new pcmcia mapping Apr 07 20:21:14 mickey|tv: thats why 3.5.4.1-aplha2 is with pcmciautils - i already removed opie-pcmciaapllet from opie-image/2.6 Apr 07 20:22:35 ~lart ipkg for being slooooow with oz feeds Apr 07 20:22:36 * ibot DoSes ipkg for being slooooow with oz feeds Apr 07 20:24:00 hm.. we have "maintainer" not "Maintainer" in packages.. Apr 07 20:25:01 hrw|husky: oh Apr 07 20:25:32 * koen|tv is tempted to use apt + .debs in images Apr 07 20:26:03 koen|tv: Isn't apt written in c++? Apr 07 20:26:12 koen|tv:will yoou patch it to not use /var? Apr 07 20:26:35 zecke: he has lot of unused flash space.. Apr 07 20:27:49 * koen|tv has root on sd Apr 07 20:29:02 * hrw|husky will have when find time to rebuild kernel etc.. Apr 07 20:29:52 zecke: libopie recipes need to be changed so -dev will get built Apr 07 20:33:20 cu - battery ending Apr 07 20:57:11 hrw|gone: yes, I have the beginning of a patch for that already Apr 07 21:08:34 03koen 07org.oe.dev * rba0fa166... 10/conf/distro/angstrom-2006.9.conf: angstrom-2006.9: insist a bit harder on using the right toolchain bits for EABI to reduce PEBKAC problems Apr 07 21:09:29 ERROR: dependency glibc-charmap-utf-8 (for e-image) not satisfied Apr 07 21:09:30 NOTE: no buildable providers for e-image Apr 07 21:09:32 hrm Apr 07 21:09:45 why does e-image seems to be perpetually broken? Apr 07 21:12:45 RP_: hey! Apr 07 21:12:54 zecke: hi :) Apr 07 21:13:02 zecke: Why the question about rpsys.net? Apr 07 21:14:51 RP: I need to find a place where to 'host' some of my services Apr 07 21:15:42 RP: I have done some changes to sanity.*. Should I send you a patch, or can I commit it? Apr 07 21:15:54 RP: I'm checking for ${USER} (no just kidding)... Apr 07 21:16:04 zecke: Feel free to upgrade it Apr 07 21:16:14 RP: I'm running the test exactly once, I'm checking for executables as well Apr 07 21:16:19 RP: make, patch, diffutils Apr 07 21:16:32 zecke: Sounds good to me - something I wanted to do Apr 07 21:18:31 zecke: rpsys.net isn't really hosted anywhere but is a couple of ADSL lines to different locations with an understanding service provider (or used to be anyway)... Apr 07 21:19:19 RP: ah okay, you only have a DNS available Apr 07 21:20:16 03freyther 07org.oe.dev * rb41b048c... 10/ (classes/sanity.bbclass conf/sanity.conf): Apr 07 21:20:16 conf/sanity.conf: Apr 07 21:20:16 Set BB_MIN_VERSION here Apr 07 21:20:16 classes/sanity.bbclass: Apr 07 21:20:16 -Run the sanitycheck exactly once (on BuildStart) Apr 07 21:20:17 -Check for common utilities/executables as well Apr 07 21:20:58 RP: I will spend the next movie in hacking the cache :) Apr 07 21:21:18 s/in/with/ Apr 07 21:21:53 zecke: I run the primary myself and have the secondary hosted independently (along with a secondary MX) Apr 07 21:22:35 RP: I want to do a small survey but my Uni disabled php/mysql/cgi due breakins Apr 07 21:23:03 zecke: This was the kind of reason I run my own servers... Apr 07 21:23:18 It has mixed benefits... Apr 07 21:24:04 disabling mysql is the only wise thing you can do with mysql Apr 07 21:24:28 zecke: On my wishlist for sanity.bbclass is checking for symlinks in the paths Apr 07 21:25:18 RP: I'm 'green' I turn off as my electronic as possible ;) Apr 07 21:25:33 RP: so running a server myself (dyndns.org) is not an option Apr 07 21:28:25 zecke: You'd probably go green at the number of computers doing things here :-/ Apr 07 21:29:13 :-))) Apr 07 21:29:43 zecke: Admittedly, rpsys.net does do several different roles though Apr 07 21:31:06 damn atmel controllers Apr 07 21:32:22 hi Apr 07 21:32:52 hi CoreDump|home Apr 07 21:37:52 Somebody experience with gps receivers ? Apr 07 21:39:20 hey CoreDump|home & gerwinin Apr 07 21:40:27 what kind of gps? Apr 07 21:40:58 if it's supported by gpsd it shouldn't be a problem Apr 07 21:47:10 kergoth: ping Apr 07 21:48:34 zecke: Does BuildStart get run even when you use bitbake -b ? Apr 07 21:49:02 it should, otherwise we would have to use PkgStarted Apr 07 21:50:59 zecke: I seem to remember having this problem. I'll test at some point Apr 07 21:55:44 03coredump 07org.oe.dev * rbdc6f8c0... 10/packages/keymaps/ (4 files in 4 dirs): keymaps: Include softkey mapping from .oz354x for SL-Cxxxx Apr 07 21:55:49 03coredump 07org.oe.dev * refd0638b... 10/packages/keylaunch/ (4 files in 4 dirs): keylaunch: Add default mapping for the LCD softkeys on SL-Cxxxx Apr 07 21:56:25 kergoth: could you add survey.openembedded.org to point to 213.239.211.73? Apr 07 21:58:55 RP: do you have the 'build' with the broken bitbake around? Apr 07 22:01:05 zecke: I can probably recreate it tomorrow - I just reverted to svn r410 or something... Apr 07 22:01:36 zecke: atm, its busy in a glibc build and I'd prefer to leave it alone :) Apr 07 22:01:50 RP: it would be interesting to see if a stamp was created Apr 07 22:01:57 NOTE: package glibc-2.3.5+cvs20050627-r3: task do_package: started Apr 07 22:02:41 zecke: what kind of survey do you want to do? Apr 07 22:03:17 koen: Should "xyz" (where xyz could be koen) be kicked out of the OE 'Team' Apr 07 22:03:21 zecke: It can't have been a stamp as wiping the cache changed the behaviour Apr 07 22:03:29 zecke: cool Apr 07 22:03:30 koen: I would like to get an idea for who we are working for Apr 07 22:04:16 zecke: letux.org ;) Apr 07 22:04:32 koen: they use cross tools ;) Apr 07 22:04:45 Koen: do you have any experience with ttl on a gps module ? Apr 07 22:04:45 koen: where you able to download the SD card? Apr 07 22:04:53 zecke: I wasn't Apr 07 22:05:10 koen: florian should revoke the GPE license for letux.org :} Apr 07 22:05:36 gerwinin: after fighting with cmos and ttl I gave up on line level adjustments Apr 07 22:06:35 zecke: You'll have to be careful to define "who". I work for several different masters :) Apr 07 22:06:45 koen: I am planning to do so as well , I got now a buetooth heartmonitor working and since I am working with this gps module I only have boiling plastics Apr 07 22:07:10 Koen : I think I have to live with a hearthmonitor without gps than :) Apr 07 22:07:17 bluetooth heartmonitor.... Apr 07 22:07:19 RP: well, I will learn about surveys and abuse OE ;) Apr 07 22:07:23 sounds like the project my uni did Apr 07 22:07:26 (is doing) Apr 07 22:07:41 RP: I would like to know what the biggest use of OE is Apr 07 22:07:44 zecke: you have SPSS installed? Apr 07 22:08:08 koen: no Apr 07 22:08:11 Koen: I read about it , pretty fun how they did it , I am making it for myself since I have a high cholestorol Apr 07 22:08:51 zecke: At a guess, NSLU2, Familiar and OZ Apr 07 22:09:00 but it would be interesting to know Apr 07 22:09:35 Zecke: a lot of taiwanese folks are running it on there smartphones Apr 07 22:09:44 Zecke: check linuxdevices.com Apr 07 22:10:03 gerwinin: pointed it to us :) Apr 07 22:10:10 gerwinin: I know that GSM module very well Apr 07 22:10:54 Zecke : I am going to Taiwan soon I will ask the people there for some smartphones for us to work on Apr 07 22:11:21 Zecke: I am busy at my work now with the imcosys thingie to get opie on Apr 07 22:11:37 gerwinin: good luck, get back to us Apr 07 22:11:45 gerwinin: I think Opie Qt4 porting will start soon Apr 07 22:11:56 gerwinin: I hope we can leverage from Qtopia4.1.1 and share the platform Apr 07 22:12:29 zecke: I will , I am trying to contribute my epia distri and asked france for a page but it seems he is a bit asleep lately Apr 07 22:13:29 zecke: it is a pitty because I am trying to port via's webpad now to linux as well and I could use some help Apr 07 22:14:20 zecke: I hope so as well Apr 07 22:14:51 ljp: lorn, any news about Qtopia4.1.1? Apr 07 22:17:00 RP: I'm counting +12 variables we utilise Apr 07 22:18:43 RP: What I want to find out: Who is using OpenEmbedded for what Apr 07 22:19:17 RP: What set of packages do they use. GUI Opie/GPE or custom, what attracts them, what is more important network, GUI, base... Apr 07 22:19:18 that would indeed be interesting to know Apr 07 22:19:38 zecke: That sounds about right. You have the Rx and Rx_y versions of the variables. The number sounds about right Apr 07 22:19:39 RP: what do they create, distris or only cross compiling help Apr 07 22:19:53 Zecke : a lot of people are using it to build routers, firewalls, and asterisk based boxes Apr 07 22:20:40 and what itches they have Apr 07 22:20:46 zecke: Do you plan to cover what their plans for feedback into the community are? Apr 07 22:21:00 Zecke: the asians are using it to build webpads , smartphones (for their home market) Apr 07 22:21:16 * koen wants a 'call home' feature in OE Apr 07 22:21:20 heh Apr 07 22:21:24 [x] "I pay MVista but use OE" Apr 07 22:21:43 koen: I think www.oesources.org access.log is that feature Apr 07 22:21:49 zecke: yeah Apr 07 22:22:04 zecke: I was about to say "my access.log has interesint entries" Apr 07 22:22:12 treke|home: ping :) Apr 07 22:22:15 interesting* Apr 07 22:23:01 I just get hits from Lineo :-/ Apr 07 22:23:11 =) Apr 07 22:23:13 mvista is a pain in the ase my boss paid for it as well but we are never using it Apr 07 22:23:22 I have heard Lineo will acquire o-hand.com soon ;) Apr 07 22:23:24 You probably see wierd hits from AMD because we check oesources first Apr 07 22:23:28 even for our internal kernels Apr 07 22:24:22 gerwinin: Convince him to put mone into OE ;-) Apr 07 22:24:41 zecke: I wonder how much they'd pay for us :} Apr 07 22:25:40 RP: we want to but everytime we try to set something up a server or a project it seems that it takes ages and ages before somebody wakes up Apr 07 22:25:41 I imagine how much they charge for a port 'Zaurus kernel from 2.0 to 2.6' Apr 07 22:25:54 gerwinin: :( Apr 07 22:26:14 gerwinin: e.g. hh.org is in the process of being restored Apr 07 22:26:26 gerwinin: and france is the only labor-force... Apr 07 22:26:38 zecke: I know Apr 07 22:26:50 I think OE needs to keep separate from hh.org if it can as much as possible Apr 07 22:27:09 gerwinin: what would be really benefitical for us... would be a meating and Leffe for mickeyl Apr 07 22:27:14 hh.org is taking care of that themselves Apr 07 22:27:27 koen: psst ;) Apr 07 22:27:35 I wonder which OE devs will have an 'accident' with his account next Apr 07 22:27:42 zecke: I tried to setup once an oe meeting but only 1 philips guy showed up Apr 07 22:27:59 okay, he didn't have a long journey Apr 07 22:28:06 zecke: no that is true Apr 07 22:28:22 zecke: but a leffe will do especially the triple onces Apr 07 22:28:27 gerwinin: It needs to be more carefully planed Apr 07 22:28:43 gerwinin: e.g. a bug hunting session... maybe next to a bigger event Apr 07 22:28:59 like the next Chaos Congress or something similiar Apr 07 22:29:02 next to easter! Apr 07 22:29:13 find some easter eggs first and move on to bugs Apr 07 22:29:20 koen: hehe my term is about to start then... Apr 07 22:29:31 OH is keen on OE but can't find a way to generate much income from it which makes it difficult to allocate time to work on it :-/ Apr 07 22:29:59 zecke: I just spent money on a new osciloscoop so I am not sure I will be recovered financially from that :) Apr 07 22:30:16 RP: what about a market place site on oe.org? Apr 07 22:30:30 RP: These companies are willing to save your ass: OH,... Apr 07 22:30:50 zecke: I'm sure OH would happily be listed :) Apr 07 22:30:56 RP: a 'colleague' of mine had a hard time finding his way into OE... Apr 07 22:31:10 RP: You can generate some money from it but the problems are the startup costs for hardware development :) Apr 07 22:31:11 RP: we need to really improve our presentation to the outside Apr 07 22:31:27 zecke: There are a lot of people in the position and we do need to work on presentation Apr 07 22:31:31 zecke: I offered to be pr person but no reaction :( Apr 07 22:31:44 gerwinin: I have missed that :( Apr 07 22:31:51 gerwinin: I know :-/ Apr 07 22:31:51 gerwinin: that is due the missing pr person ;) Apr 07 22:32:03 gerwinin: Somehow I missed this as well :-/ Apr 07 22:32:20 zecke: I am still available for a couple of hours per week so .... Apr 07 22:32:47 gerwinin: mail oe@handhelds.org, I'll reply then! Apr 07 22:32:54 zecke: okay Apr 07 22:33:00 gerwinin: I think an agenda and plan for OE PR would be needed first Apr 07 22:33:12 gerwinin: how do we want to present OE? How to achieve it? Apr 07 22:33:16 gerwinin: where to present it? Apr 07 22:33:23 the poster design was a good start to get the PR stuff rolling Apr 07 22:33:27 zecke: I think that is a good starting point Apr 07 22:33:42 zecke: http://presentationzen.blogs.com/ Apr 07 22:33:53 zecke: the first thing I want to improve is the communication with the manufacturers Apr 07 22:34:08 koen: hehe, I would like to give a Lessig-style OE talk Apr 07 22:34:22 zecke: I would also like to make a list with professional people busy with oe like rp Apr 07 22:34:24 gerwinin: a clear message is most important Apr 07 22:34:31 zecke: agree Apr 07 22:34:32 gerwinin: Why is OE right for you! Apr 07 22:34:41 zecke: free culture? Apr 07 22:34:43 gerwinin: How to get help from OE (community, professional) Apr 07 22:34:48 koen: yes, I love that one Apr 07 22:34:48 I really wish somebody would have made a talk at CELF Apr 07 22:34:59 I feel bad talking about OE but not being able to get into the details Apr 07 22:35:12 gerwinin: How to make OE fit you! Apr 07 22:35:15 Cosmic : I did one time a talk about oe in holland Apr 07 22:35:19 gerwinin: How to get your improvements back Apr 07 22:35:23 * koen feels bad to have too little time to tour the world to spread the word Apr 07 22:35:30 gerwinin: but thats not CELF, is it? :) Apr 07 22:35:53 * zecke really wants to hold talks and give tutorials about OE Apr 07 22:35:54 zecke: what I would like to see is mainly bb files of what people made Apr 07 22:36:35 CosmicPenguin: I'd like to talk to CELF about OE... Apr 07 22:36:47 CosmicPenguin: I've noticed that OE is too technical for 99% of the audience Apr 07 22:36:55 RP: Maybe next year.. :( Apr 07 22:37:03 koen: yeah, but I think they want to hear bout it Apr 07 22:37:11 koen: This can be easily avoided Apr 07 22:37:18 koen: take a heavy autotools user Apr 07 22:37:23 koen: try cross tools, fail Apr 07 22:37:26 koen: use OE, win Apr 07 22:37:26 i also noticed that interactive sessions work really well Apr 07 22:37:28 Koen: but I learned that oe is a good starting point to start embedded development , it even makes it a lot easier Apr 07 22:37:36 CosmicPenguin: Maybe. The problem is I really really hate giving presentations... Apr 07 22:38:21 RP: I think a good OE presentation consists out of two people Apr 07 22:38:26 interacting on stage ;) Apr 07 22:38:32 the secret to a succesfull presentation: black turtlenecks Apr 07 22:39:11 zecke: I can see the advantages in that :) Apr 07 22:39:29 I like telling people what I know, but I'm not a very good speaker Apr 07 22:39:35 like the dillo presentation at fosdem 2005? Apr 07 22:39:49 koen: I have not seen that one ;) Apr 07 22:40:00 RP: just assign someone the role of a cross tool user Apr 07 22:40:09 RP: and help him with the power of OE Apr 07 22:40:11 zecke: mail sent Apr 07 22:40:47 zecke: i am not subscribed to the mailing list so replies to my adres :) Apr 07 22:41:28 zecke: take a look at the trainings at http://free-electrons.com/training Apr 07 22:41:52 koen: I know people that are making about 4000 euro a month with oe trainings Apr 07 22:42:05 Thats some good scratch Apr 07 22:42:24 gerwinin: Why don't we get feedback from these people? Apr 07 22:42:54 RP: because there is not spokesperson and it is hard to contact the people from oe Apr 07 22:43:24 RP: the problem is for example the wiki Apr 07 22:43:41 I think that's an issue we need to address - maybe have a panel of core developers or something Apr 07 22:44:11 RP: That is what I want but also fear... Apr 07 22:44:35 RP: for example I have been bombarded with mails about those asterisk boxes I created once but as long as it is hard to contribute to oe Apr 07 22:44:50 My bb files do not get through Apr 07 22:44:54 zecke: I know - double edged sword :-/ Apr 07 22:45:02 I gouess you could make a foundation or something - raise some money to beef up the infrastructure Apr 07 22:45:44 CosmicPenguin: I only know the german laws for associations... Apr 07 22:46:02 CosmicPenguin: we would have to have an founding assembly with five members Apr 07 22:46:28 Cosmicp: I think we should focus also more on cheaper hardware that for the regular developer it is easier to start with oe Apr 07 22:46:29 zecke: all german citizens? Apr 07 22:46:29 I guess then there's the question of which country it would exist in etc :-/ Apr 07 22:46:54 Cosmicp: that is also why I am trying to create a distri made with oe for epia Apr 07 22:47:12 I think the better system would be to keep funds out of it if possible. If a sponsor wanted to sponsor development, that could be taken up with the people doing the development directly Apr 07 22:47:26 under supervison from the panel etc. Apr 07 22:48:16 gerwinin: see, thats your first problem Apr 07 22:48:25 gerwinin: if you were using Geodes, I could get you hardware... :) Apr 07 22:49:06 seeding hardware instead of funds is also a nice way to contribute Apr 07 22:49:16 although I could see that some people would prefer money Apr 07 22:49:33 koen: hardware works well for certain circumstances Apr 07 22:49:49 koen: for others, money can have other benefits :-/ Apr 07 22:49:50 Koen : I agree upon donating hardware I did it once and that works great Apr 07 22:50:31 koen: In my case, money helps more than hardware as it buys time for me to work on OE in work time Apr 07 22:50:52 RP: that's exactly what I was getting at :) Apr 07 22:51:10 CosmicPenguin: not necessary, but physically in one place Apr 07 22:51:18 koen: right :) Apr 07 22:51:45 CosmicPenguin: Do you produce that Geode design as well ( the one I saw on linuxdevices.com) Apr 07 22:52:27 zlib_inflate looks to have made -mm :) Apr 07 22:53:00 RP: congrats Apr 07 22:53:23 RP: yipee :) Apr 07 22:53:43 RP: but we can launch to start with an oe partner programm Apr 07 22:54:07 zecke: thats our RDK for our new processor - we don't make them, but thats our design Apr 07 22:55:48 Cosmic would your company be interested in donating some hardware to oe ? Apr 07 22:56:06 gerwinin: possibly Apr 07 22:56:26 I think we need to develope the people behind OE page ASAP but nominate a few people to be core developers. These would show key points of contact for OE... Apr 07 22:56:27 gerwinin: doesn't hurt to ask Apr 07 22:56:44 * CosmicPenguin has to run Apr 07 22:56:51 How we nominate the core developers is tricky... Apr 07 22:57:12 RP: how about if we seperate the wiki from an actual website ? Apr 07 22:57:24 gerwinin: It already is, isn't it? Apr 07 22:57:24 RP: http://oe.linuxtogo.org/people.php + http://ewi546.ewi.utwente.nl/maps/OE.php ? Apr 07 22:57:25 by self appointment Apr 07 22:58:05 zecke: shall I coordinate this ? Apr 07 22:58:14 koen: aha :) Apr 07 22:59:07 RP: what I do a lot is asking hardware in return of help with oe :) Apr 07 23:00:11 koen: I'm not on you map :) Apr 07 23:00:44 koen: We should list email addresses on that people's page (with people's consent) - I'm happy for mine to be listed Apr 07 23:01:34 RP: that map is pretty old Apr 07 23:02:49 RP: and I think a 'contact form' mapping people to interest would be a good thing as well Apr 07 23:03:03 Koen : If you want I can set up the main webpage of oe and change the layout a bit Apr 07 23:03:45 Koen: to make hardware donations possible and to make it a bit more attractive for publicity Apr 07 23:04:11 gerwinin: sure, but I'm not sure which person(s) run the website Apr 07 23:04:31 Koen: Should we be allowing banners and stuff on the website for sponsoring ? Apr 07 23:06:54 I really don't know Apr 07 23:07:19 We need some kind of committee :-( Apr 07 23:07:39 RP: Sound like a good idea Apr 07 23:08:20 RP: but I need to watch a bit out that I keep on having enough time for coding and hardware hacking :) Apr 07 23:09:00 gerwinin: I just see it as something having enough power to match decisions Apr 07 23:09:25 RP: http://oe.handhelds.org/people.php lists the monotone keys, so we just need to enable mail forwarding from openembedded.org to solve the problem :) Apr 07 23:09:34 RP: yes Apr 07 23:10:15 koen: "just". We need a hsting provider to do that :-/ Apr 07 23:10:55 actually... Apr 07 23:11:16 I think I have a 2 year old mail from michaelo about @openembedded.org addresses Apr 07 23:11:56 anyway Apr 07 23:11:59 'night all Apr 07 23:12:16 Worst case, I guess you could point oe.org at rpsys.net and I could forward as appropriate :) Apr 07 23:12:22 'night koen|sleep Apr 07 23:12:38 (by point, I just mean the MX) Apr 07 23:13:58 good night koen Apr 07 23:15:47 OT: someone broke initscripts? Apr 07 23:17:38 * gerwinin needs to save money for a stronger laptop Apr 07 23:26:00 good nite Apr 07 23:54:04 anybody has seen the xstroke-0.6.tar.gz downloading problem? Apr 07 23:54:19 xstroke.org is changed to something totally unrelated. Apr 08 00:23:31 Hehe Apr 08 00:23:36 Right when I get in :p Apr 08 00:24:38 oh, accesslog Apr 08 00:24:55 hmm. I'll have to fiddle with that. Right now it's all in one Apr 08 00:30:07 419 unique ips Apr 08 00:30:28 that's quite a few people building oe Apr 08 00:33:26 Yep, just a few :) Apr 08 00:34:33 tomorrow I get to figure out how to do builds :p Apr 08 00:36:12 the bulk of the ips that have reverse dns configured seem to be non us ips. Apr 08 00:37:22 more ips that resolve to aol then I would have guessed Apr 08 00:37:37 Scary.. Apr 08 00:38:36 CosmicPenguin: You don't happen to be around in any form, do you? Apr 08 00:56:45 g'day treke|home Apr 08 01:12:45 Hey pb Apr 08 06:41:40 how do you override bootstrap rrecommends Apr 08 07:27:05 good morning Apr 08 08:35:18 03coredump 07org.oe.oz354x * r23b4360b... 10/packages/xserver-common/xserver-common_1.8.bb: xserver-common: re-enable 100DPI patch for 200DPI machines Apr 08 08:35:23 03coredump 07org.oe.dev * r78a4544b... 10/packages/gpe-session-scripts/gpe-session-scripts_0.66.bb: gpe-sessionscripts: Set font-size for 200DPI machines back to 12. Anything smallr is barely readable. Apr 08 08:35:26 03coredump 07org.oe.dev * r6e205d2f... 10/packages/gpe-session-scripts/gpe-session-scripts_0.66.bb: disapproval of revision '78a4544b0da0db7bef0ee08f054a51321fc0d271' Apr 08 08:35:30 03coredump 07org.oe.oz354x * rdb18e20e... 10/packages/matchbox-wm/matchbox-wm_1.0.bb: matchbox-wm: Upgrade to version _1.0 from .dev Apr 08 08:35:35 03coredump 07org.oe.oz354x * r6ddcd71c... 10/conf/distro/preferred-gpe-versions-2.7.inc: preferred-gpe-versions-2.7.inc: Upgrade to matchbox-wm 1.0 Apr 08 08:37:15 does anyone know good resources, wikis, hints, tips to get smaller binaries out of a c++ codebase? Apr 08 08:38:06 I have an oe platform, and a c++ app I want/need to have run on it, but it compiles huge (~1mb'ish) Apr 08 08:38:22 good morning all Apr 08 08:39:42 muy: see http://free-electrons.com/doc/embedded_linux_optimizations/img47.html Apr 08 08:40:10 I know -Os and -s :) Apr 08 08:40:35 nice link though, thx Apr 08 08:41:47 morning all Apr 08 08:41:57 hey RP Apr 08 08:42:08 muy u can try if u don't use to exclude support for rtti or exceptions ... with -fno-rtti and (not sure) -fno-exceptions Apr 08 08:42:13 muy: you could ask zecke and kergoth about visibility Apr 08 08:43:05 gremlin[it]: I definitely use it for exceptions, it's the main adv c++ is providing me over c in this case -- proper deinitialization of device drivers, etc Apr 08 08:43:22 muy .. ok and rtti ? Apr 08 08:43:27 worth a shot :) Apr 08 08:44:12 muy: with some horrid set of options, it's possible to get rid of libstdc++ Apr 08 08:44:20 ah Apr 08 08:44:22 (you can't use the STL, then, of course, but perhaps you don't need to) Apr 08 08:45:04 is there a way to build libstdc++ smaller / links in less code? Apr 08 08:48:39 err.. probably I should remove that dependency on boost, then -fno-rtti would pass Apr 08 08:49:59 oh my f*cking god Apr 08 08:50:18 hey CoreDump|afk Apr 08 08:50:20 * CoreDump|afk just resolved the idiotic 200DPI bug Apr 08 08:50:25 hey koen Apr 08 08:51:09 look what I found in /etc/X11/Xdefaults Apr 08 08:51:12 Xft.dpi: 100 Apr 08 08:51:30 remove that and the machine runs with 200 DPI perfectly fine... Apr 08 08:51:55 CoreDump|afk: Nice find :) Apr 08 08:52:14 CoreDump|afk: Does this mean we can build alpha2? :) Apr 08 08:52:32 after a few more tests, yeah Apr 08 08:53:41 Hmm. My zlib changes broke ppc/powerpc :-( Apr 08 09:03:10 ehm Apr 08 09:03:18 onotone: bytes in | bytes out | revs in | revs out | revs written Apr 08 09:03:19 monotone: 308.0 k | 1.4 M | 0 | 450 | 0 Apr 08 09:03:26 =D Apr 08 09:03:31 * koen wonders what happened to nslu2-linux.org Apr 08 09:03:40 nice one, been busy huh? ;) Apr 08 09:04:01 450 csets added in 15 minutes? Apr 08 09:05:00 now that's going to be a bitch to pull Apr 08 09:05:25 I suspect something went wrong on nslu2-linux.org Apr 08 09:05:35 like someone put an old oe.db in place Apr 08 09:05:45 hmm Apr 08 09:07:05 eeew Apr 08 09:07:05 pd3fa4f.wpdabw01.ap.so-net.ne.jp - - [08/Apr/2006:11:16:37 +0200] "GET /icons/back.gif HTTP/1.0" 200 216 "http://ewi546.ewi.utwente.nl/mirror/www.openzaurus.org/official/" "Mozilla/4.08 (PDA; SL-C3100/1.0,Qtopia/1.5.5) NetFront/3.1" Apr 08 09:07:15 03koen 07org.oe.dev * re42c8683... 10/packages/gpe-windowlist/gpe-windowlist_cvs.bb: gpe-windowlist: add cvs version of gpe-windowlist, a panel applet that lists all active windows Apr 08 09:07:54 wth? Apr 08 09:07:56 koen: yay! I am _so_ going to try this applet Apr 08 09:07:57 monotone: common ancestor 776f76fe495350009610233da25b36aa5c732d3f jbowler@nslu2-linux.org 2005-09-26T23:18:06 found Apr 08 09:08:00 2005! Apr 08 09:10:54 meh my tree is so out of date, i need to push all these changes Apr 08 09:10:56 CoreDump|afk: don't forget to move your SRCDATE to tomorrow, the applet landed in cvs last night Apr 08 09:11:30 will do, thanks Apr 08 09:13:32 how do you undo a commit in monotone? Apr 08 09:13:46 i want to undo a commit i did, but still keep the changes i made Apr 08 09:14:48 mehh Apr 08 09:15:02 wlan is totally screwed in .354x Apr 08 09:15:09 anyway, gotta go to work Apr 08 09:15:11 later Apr 08 09:15:27 thx for the tips all Apr 08 09:17:20 TheMasterMind1: monotone kill_rev_locally Apr 08 09:19:13 * koen fires up monotone-viz Apr 08 09:20:02 monotone: unknown command 'kill_rev_locally' Apr 08 09:20:08 'db kill_rev_locally' Apr 08 09:20:24 the name is ugly because the functionality is ugly :-) Apr 08 09:21:05 (both in the sense of usually not being a good idea, and because it's easy for it to have basically no effect and just not work, i.e., if you have already pushed the rev in question.) Apr 08 09:21:22 gotcha Apr 08 09:22:11 hrm Apr 08 09:22:45 crap Apr 08 09:23:37 njs: what's the best way to merge two heads where one has ancestors in a different branch? Apr 08 09:24:00 koen: can you give more details? Apr 08 09:25:24 ok sorry, stupid questions, how do i figure out the id of the commit i made Apr 08 09:26:07 is it the revision hash? Apr 08 09:26:24 njs: http://ewi546.ewi.utwente.nl/tmp/viewmtn/headofbranch.psp?branch=org.openembedded.dev and http://dominion.kabel.utwente.nl/koen/pda/files/screwage.tiff Apr 08 09:26:39 TheMasterMind1: yes Apr 08 09:26:43 TheMasterMind1: that is what 'id' means Apr 08 09:26:57 TheMasterMind1: what messages confused you? Apr 08 09:27:48 koen: err. do they share _any_ history? Apr 08 09:27:56 well after the kill monotone log shows nothing at all, whereas before it showed all the revs before mine Apr 08 09:28:21 and mtsh says terminate called after throwing an instance of 'boost::bad_lexical_cast' Apr 08 09:28:22 TheMasterMind1: oh, another reason 'db kill_rev_locally' has a nasty name, it sort of screws up your working copy Apr 08 09:28:28 njs: ehm right till the branch point eons ago, --lca gives monotone: common ancestor 776f76fe495350009610233da25b36aa5c732d3f jbowler@nslu2-linux.org 2005-09-26T23:18:06 found Apr 08 09:29:00 TheMasterMind1: (in a very simple way, the MT/revision file now contains the hash of a revision that doesn't exist anymore. You can fix it up by hand by doing 'monotone automate heads > MT/revision', probably. Make sure that there is only one head when you do this.) Apr 08 09:29:21 koen: slick :-( Apr 08 09:29:39 koen: have you tried normal 'merge --lca'? Apr 08 09:29:39 define 'only one head' Apr 08 09:29:52 TheMasterMind1: "monotone heads" prints 1 line, not more than that :-) Apr 08 09:30:26 njs: that's with merge --lca, I aborted it before it makes me merge a zillion 'conflicts' Apr 08 09:30:35 koen: ah. grr. Apr 08 09:30:48 stupid 0.25 Apr 08 09:31:15 koen: what happened was that you have two branches that are supposed to diverge permanently, but accidentally one rev was put into the other branch? Apr 08 09:31:22 koen: or... what? Apr 08 09:31:22 Hi! Apr 08 09:31:23 rwhitby, NAiL: monotone.nslu2-linux.org needs to serve all branches now, otherwise no-one can pull anymore from your server Apr 08 09:31:34 njs: that's what happened Apr 08 09:31:37 hey danboid Apr 08 09:31:52 hi koen! Apr 08 09:32:02 is RP here? Apr 08 09:32:06 koen: so, err, you realize merging these will, in fact, merge the two branches? Apr 08 09:32:28 koen: i.e., merge in all the changes from .oz354x? Apr 08 09:32:40 hrm Apr 08 09:32:42 :-( Apr 08 09:32:54 now would probably be a good time to take down the server :-( Apr 08 09:33:00 the trolls will shout 'database corruption' again Apr 08 09:33:00 njs: thanks Apr 08 09:33:03 (you guys are so good at finding these horrible cases! :-)) Apr 08 09:33:08 yes, well Apr 08 09:33:23 if they would like to invent rollback algorithms, they are welcome Apr 08 09:33:24 njs: the bad news: the cset is in all 5 servers.... Apr 08 09:33:39 (and all my local dbs) Apr 08 09:33:48 however, since no-one in the world has actually succeeded in doing this... :-( Apr 08 09:33:49 hmm Apr 08 09:34:11 danboid: yes Apr 08 09:34:23 if monotone had a cherry picker the .oz354x branch would be a clean subset of .dev (in theory) Apr 08 09:34:44 hmm, actually, it's not even rollback, really, just revival. but still, there are no working implementations of this, it's a tricky thing that's still outside the boundaries of theory. Apr 08 09:34:52 Hi RP! Whats the status of the video accel patches for the Cxx00 under 2.6- anything been done recently? Apr 08 09:35:15 ("rollback" and "revival" are technical terms in merging theory, never mind my babbling) Apr 08 09:35:56 perhaps this calls for writing the notional "merge" command that takes two trees and declares one of them to totally win everything Apr 08 09:36:52 danboid: They've not been touched Apr 08 09:36:56 koen: are there any files deleted on the oz354x branch, that the .dev branch wants to keep? Apr 08 09:37:40 koen: that's the really obnoxious case, though in practice even it is live-with-able, just you'd have to be a little careful with merges for a bit afterwards Apr 08 09:38:03 Aww! This is the one thing that prevents me ditching cacko. Is there a good chance this will be sorted for the final release of 3.5.4.1? Apr 08 09:38:55 njs: yes .oz354x ditched some files Apr 08 09:38:55 koen: on a tangent, but not really as tangential as all that, when would you guys be planning to switch to 0.26 after it is released? Apr 08 09:39:05 danboid: I'm the only person working on c7x0 and cxx00 kernels and I have a ton of conflicting priorities so I have no idea :-( Apr 08 09:39:21 njs: my plan is to wait for 0.27 Apr 08 09:39:50 koen: ah, hmm. I don't think I can write a clobber-merger on the 0.25 codebase... :-( Apr 08 09:39:54 * njs ponders whether that is really true Apr 08 09:40:01 koen: Why 0.27? Apr 08 09:40:31 RP: to find bugs in OE + 0.26 without burdening our users Apr 08 09:40:52 * koen sends a mail to the mailinglist about the multiple heads Apr 08 09:40:57 Can we not manually remove an offending revision to "fix" things? Apr 08 09:41:10 RP: we can Apr 08 09:41:16 but *everyone* has to do that Apr 08 09:41:33 (see the 'killrev' script solution I used earlier) Apr 08 09:42:03 koen: I remember it. Can that be updated to block that revision entering the core servers? Apr 08 09:43:00 ehm Apr 08 09:43:05 hmm Apr 08 09:43:10 all core and secondary server have it already... Apr 08 09:43:14 is there some reason we didn't use the epoch stuff before to get rid of revs like that? Apr 08 09:43:14 RP: Whoah! Thats some responsibility for one guy- take care RP! Don't you work for intel doing something like this- I'm sure you could Apr 08 09:43:31 rwhitby, NAiL: could you shutdown monotone.nslu2-linux? Apr 08 09:43:50 if you bump the .dev epoch, then everyone has to take some manual action on their db before they can sync with it Apr 08 09:43:51 and if I had some money I'd be trying to bribe you into that video accel patch right now ;) Apr 08 09:44:07 (if you bump it on the server, I mean, they have to take some manual action before they can sync with that server) Apr 08 09:44:12 njs: 'everyone' and 'manual action' are things I like to avoid Apr 08 09:44:15 danboid: I don't work for Intel, no Apr 08 09:44:18 koen: I agree! Apr 08 09:44:46 njs: I have access to 4 of the 5 servers, so a tweak to that would be trivial Apr 08 09:44:58 If we can kill that revision on the core servers, and stop it coming back from anyone who already has it, we might be able to prevent most people having problems... Apr 08 09:44:59 koen: _if_, however, one is going the "kill it manually in every db" route, then epochs should make it easier Apr 08 09:45:38 We do really need to shut down the servers now to stop it spreading though... Apr 08 09:45:42 the other option is the clobber merge Apr 08 09:45:54 RP: they are down, except for nslu2-linux Apr 08 09:46:15 koen: ah, ok, good Apr 08 09:46:23 njs: the clobber merger just ignores one head? Apr 08 09:46:38 koen: basically it does a "merge" where one side wins on each and every point Apr 08 09:46:54 actually, looking at this, maybe it is trivial to add to the 0.25 line Apr 08 09:47:06 (especially since it only has to work, err, once, ever) Apr 08 09:47:42 (since the hard part of this sort of thing is always thinking through the 10 million weird rare edge cases, and in this case we could just try the simple thing and then look at it to see if it worked right before pushing it) Apr 08 09:47:44 would all clients need to support the clobber merger? Apr 08 09:48:01 (I suspect not, but I'm just checking) Apr 08 09:48:05 no, it would create a normal rev Apr 08 09:48:23 that _could_ be the result of a normal merge command, just what it would say is that the merge of A and B is... A again. Apr 08 09:49:02 This does sound like the best option... Apr 08 09:49:07 oh, hmm Apr 08 09:49:26 do you ever merge .dev -> .oz354x? Apr 08 09:49:32 njs: http://ewi546.ewi.utwente.nl/OE/snapshot.db is a db with all the screwage Apr 08 09:49:39 njs: no, we never merge that Apr 08 09:49:50 njs: we cherry pick from .dev -> .oz354x Apr 08 09:49:58 .oz354x is a stable release branch Apr 08 09:50:01 and never will? because doing this would mean that if you ever did, it would wipe out .oz354x's changes Apr 08 09:50:04 nod Apr 08 09:50:08 okay :-) Apr 08 09:52:19 njs: can something like "your revisions has ancestors in a different branch, really commit? [No]: " be implemented? Apr 08 09:52:38 since this is the second time it happened with the OE repo Apr 08 09:52:38 do you know how this happened? Apr 08 09:52:57 oh, is _this_ what happened the first time you redid things? Apr 08 09:53:01 I suspect someone did 'org.oe.oz354x:/foo/bar/ Apr 08 09:53:12 I suspect someone did 'org.oe.oz354x:/foo/bar/ $ monotone commit baz -b org.oe.dev' Apr 08 09:53:24 * njs . o O (_Why_?) Apr 08 09:53:25 :-) Apr 08 09:53:41 commit probably should not take -b Apr 08 09:53:43 njs: this is what the trolls called 'db corruption' the first time Apr 08 09:53:48 * RP suddenly realises how this was possible... Apr 08 09:53:50 ahhh Apr 08 09:54:02 I never did find out what that was about :-) Apr 08 09:54:32 koen: Arguably, it does corrupt the database :-/ Apr 08 09:54:36 the first time however it involved two branches which propagated to eachother Apr 08 09:54:50 anyway, the UI I will probably switch things to in the relatively near future (maybe by 0.27, dunno), is that the way you change your working copy branch is by running "monotone branch ". and that's the only way. Apr 08 09:55:02 RP: it creates a problem, which can be solved without losing data Apr 08 09:56:04 it this case monotone is too powerfull and flexible Apr 08 09:56:31 koen: where solving it isn't a nice experience which is the real problem :-/ Apr 08 09:56:47 thank god we aren't using git or something equally obfuscated ;) Apr 08 09:56:49 Do we know how pushed this revision? Apr 08 09:56:57 yes Apr 08 09:57:24 I assume you will have a word? Apr 08 09:57:37 the .oz354x ancesters are a big hint Apr 08 09:57:59 the key that was used to sign the erroneous branch cert is also a hint :-) Apr 08 09:58:05 heh Apr 08 09:58:31 it's even in the screenshot I posted Apr 08 09:59:14 :) Apr 08 09:59:39 I've not pulled since I read about the breakage... Apr 08 09:59:41 RP: well there's nothing really wrong with what that person did, but the circumstances make it a problem **** ENDING LOGGING AT Sat Apr 08 09:59:56 2006