**** BEGIN LOGGING AT Tue Oct 24 02:59:57 2006 Oct 24 07:30:15 morning all Oct 24 07:33:20 ade|desk: Hi. Oct 24 07:36:46 good morning all Oct 24 07:38:00 morning koen, den-ros Oct 24 07:38:54 hey ade|desk Oct 24 07:55:00 good morning Oct 24 07:55:29 good morning. Oct 24 07:58:30 What do I do with source code that only has configure.in and Makefile.am stuff? I know this has to do with autotools, but I already inherited that. How do I produce the configure script, etc.? Oct 24 08:00:59 Laibsch: does it include an autogen.sh? Oct 24 08:01:59 florian_kc: no Oct 24 08:02:53 Laibsch: maybe search for one... you usually just need to run autoconf, automake and maybe some more in the correct order Oct 24 08:12:23 just autoconf;automake gave me the configure script. but still no Makefile (Makefile.{in|am} are there). OE works fine with the released source code for granule but for the CVS source it bails out (" make: *** No rule to make target `install'. Stop.") which is why I am asking. Oct 24 08:13:51 Laibsch: the configure script will create the Makefile Oct 24 08:14:14 Aha Oct 24 08:15:40 re Oct 24 08:15:44 morning Oct 24 08:16:40 hi hrw|work Oct 24 08:19:44 'autoreconf -i' Oct 24 08:23:46 9408 revs in .dev, neat Oct 24 08:43:12 * koen inherits insane.bbclass Oct 24 08:48:27 all hail QA master zecke Oct 24 08:48:57 koen: do you know if postgres can do UPDATE TABLE foo ADD COLUMN moo varchar(255) NULL; Oct 24 08:50:38 it should be able to Oct 24 08:51:12 zecke: some versions ago it was not able Oct 24 08:51:53 s/UPDATE/ALTER/ ? Oct 24 08:52:04 koen: right! Oct 24 08:52:13 koen: okay I will try now Oct 24 09:02:44 koen: http://www.openembedded.org/bonsai/query/description/bug/?items=60 Oct 24 09:05:09 hmm Oct 24 09:05:16 I think I killed it :) Oct 24 09:05:21 http://www.openembedded.org/bonsai/query/description/bug/bugs.openembedded.org/show_bug.cgi?id=1483 ;) Oct 24 09:05:31 zecke: I think I see the change :) Oct 24 09:05:39 someone forgot about http:// Oct 24 09:06:40 hrw|work: right Oct 24 09:07:55 I need to adjust the regexp for the bug filter though Oct 24 09:13:09 hmm I need to make this smarter Oct 24 09:13:16 GNOME bug #xxx Oct 24 09:13:26 slugos ticket # Oct 24 09:31:17 koen: could you reload apache please? Oct 24 09:31:40 or wait Oct 24 09:36:59 koen: reload apache please Oct 24 09:37:48 morning Oct 24 09:39:48 I hate my job Oct 24 09:40:25 hehe Oct 24 09:40:42 quick heads up, what files must be shipped with any GPL code ? Oct 24 09:46:31 zecke: done Oct 24 09:46:39 ade|desk: source files and buildscripts iirc Oct 24 09:46:46 hey XorA|gprs Oct 24 09:47:39 koen i was wondering in the case of license files Oct 24 09:48:43 ade|desk: without the license files it isn't gpl :) Oct 24 09:49:20 did gprs survive that? Oct 24 09:50:06 XorA|gprs: I think so :) Oct 24 09:50:07 single file source, with standard preamble , does that need a seperate files in addition? Oct 24 09:50:08 hah it seems to work now Oct 24 09:50:55 hey koen Oct 24 09:50:55 ade|desk: your in uk? Oct 24 09:51:01 yup Oct 24 09:52:56 ade|desk: heading for linuxworld? Oct 24 09:53:26 XorA|gprs: sadly no Oct 24 09:59:52 * ade|desk feels better after ranting at some uni student over GPL violation, egged on by bad coding firstly. Oct 24 10:01:35 :) Oct 24 10:03:26 zecke: I use 6to4 for IPv6 (you asked about this a day or so back). Anything in particular you wanted to know? Oct 24 10:04:26 v8jlene: on OSX? Oct 24 10:04:41 v8jlene: Which relais do you use, should ping6 and traceroute6 work? Oct 24 10:04:51 or is it likely a firewall filters them away Oct 24 10:05:04 zecke: Nope... only linux, cisco and XP/2K... Oct 24 10:05:49 yay, my memory arrived! Oct 24 10:06:39 Firewalls need to allow ipv4 packets of type "ipv6" (ie, ipv6-tunneled-over-ipv4) through. Oct 24 10:07:06 heh heh Oct 24 10:20:21 im off for a bit before me battery dies Oct 24 10:45:00 03hrw 07org.oe.dev * r6d96c62d... 10/ (1 conf/distro/celinux-test.conf): celinux-test: set fixed glibc version Oct 24 10:47:27 03hrw 07org.oe.dev * rf6cde9cc... 10/ (1 conf/distro/celinux-test.conf): celinux-test: set fixed toolchain stuff to not get sharprom crap info Oct 24 10:49:23 03mickeyl 07org.oe.dev * r08caa01f... 10/ (1 packages/qemu/qemu-native.inc): qemu-native * make do_stage more reliable Oct 24 10:50:34 03hrw 07org.oe.dev * rcaf46dcf... 10/ (1 conf/distro/celinux-test.conf): celinux-test: bump omap5912osk kernel to 2.6.18-omap1 Oct 24 10:52:00 03hrw 07org.oe.dev * rfb2a1b3e... 10/ (15 files in 7 dirs): Oct 24 10:52:00 openembedded-essentials: package for Debian compatible distributions to get all Required Software installed Oct 24 10:52:00 It probably need some work but it is good start. Oct 24 10:52:39 03lenehan 07org.oe.dev * r0178f355... 10/ (1 packages/patch/patch.inc packages/patch/patch_2.5.9.bb): Oct 24 10:52:39 patch 2.5.9: Use alternatives for patch since busybox also provides a patch Oct 24 10:52:39 command. As per #1472. Oct 24 10:52:48 03lenehan 07org.oe.dev * ra173ba23... 10/ (1 packages/findutils/findutils_4.1.20.bb): Oct 24 10:52:48 findutils 4.1.20: Use alternatives for find and xargs since these commands Oct 24 10:52:48 are also provided by busybox. Closes #1472. Oct 24 10:52:53 03lenehan 07org.oe.dev * rf5c0d86d... 10/ (1 packages/havp/havp_0.82.bb): havp 0.82: Remove the .debug directory from the main package. Oct 24 10:53:23 zecke: http://rafb.net/paste/results/YyNHJ887.html Oct 24 10:54:06 03lenehan 07org.oe.dev * r2ec4d476... 10/ (1 packages/p3scan/p3scan_2.9.05d.bb): p3scan 2.9.05d: Remove the .debug directory from the main package. Oct 24 10:54:21 03koen 07org.oe.dev * r09f7b1fc... 10/ (1 conf/distro/include/angstrom-glibc.inc): angstrom-glibc: set preferred provider for virtual/libc Oct 24 10:54:28 03koen 07org.oe.dev * r4228f36c... 10/ (1 conf/distro/include/angstrom-glibc.inc): angstrom-glibc: remove stray space Oct 24 10:55:34 koen: okay Oct 24 10:55:50 replace str.contains(x) with x in str: Oct 24 10:56:14 koen: alternatively I can do that Oct 24 10:56:22 zecke: go ahead :) Oct 24 10:56:34 food first Oct 24 10:57:39 zecke|food: and can the message be changed to "non dev package contains .so: glibc-foo-bla.ipk" ? Oct 24 10:57:57 besides .ipk Oct 24 11:01:29 koen, I have an image with glbc-2.5 patched to print out some info about the struct offsets in the check_pf call Oct 24 11:07:01 03mickeyl 07org.oe.dev * ra7356e33... 10/ (2 files in 2 dirs): sharprom-toolchain 1.0 PROVIDE your stuff only for DISTRO=sharprom-compatible Oct 24 11:08:15 mickey|zzZZzz: thx Oct 24 11:14:30 argh.. what a shitty day Oct 24 11:14:39 even monotone hates me Oct 24 11:14:43 and it's raining as well Oct 24 11:15:07 mtn: misuse: path 'packages/gpe-calendar/gpe-calendar_svn.bb' already exists, cannot create Oct 24 11:15:11 ARGH Oct 24 11:15:15 fscking monotone Oct 24 11:15:26 hrw@bitbake:~/devel/oe/org.openembedded$ rm -rf _MTN/tmp/;mtn revert --missing;mtn update Oct 24 11:15:41 sick but only one method which work Oct 24 11:16:37 fuck it. OZ will not be updated today - I do not want to fight with monotone again, again, again, again and then again, again Oct 24 11:16:58 mtn update suxx Oct 24 11:17:23 I just added support for the compulab pxa270 module. The kernel build includes a binary module for the NAND flash driver. Should this module be checked into OE, or what is the best way to handle proprietary modules? Oct 24 11:17:50 cbrake: download the module from the net Oct 24 11:18:05 cbrake: add recipe, fetch module from net Oct 24 11:18:09 cbrake: I don't think we want binary blobs in OE Oct 24 11:18:12 cbrake: do not forget about license Oct 24 11:19:05 koen: OK, it is currently buried in a zip file, but it would not be a big deal to extract it. Oct 24 11:19:53 cbrake: DEPENDS += "zip-native" and it will work Oct 24 11:20:09 hrw@bitbake:~/devel/oe$ rm -rf org.openembedded/ Oct 24 11:20:12 fscking monotone Oct 24 11:20:39 hrw|work: no idea what the license is on this module -- I'll check with compulab. Oct 24 11:21:14 zecke|food: http://rafb.net/paste/results/ROvJkD44.html Oct 24 11:21:15 hrw|work, I will flash the OSK with thte new kernel later today Oct 24 11:21:22 hopefully that gives you some good news Oct 24 11:31:26 koen: looks good Oct 24 11:31:26 Crofton: ok Oct 24 11:37:02 If I want to copy over file files to the rootfs before OE generates the rootfs image do I just add a function do_populate_staging() to my meta file and copy all the files I want to the rootfs inside of it? Oct 24 11:37:57 Gerrath: generate proper recipes for them Oct 24 11:39:01 koen, it is mainly config files such as network settings so I didn't want to modify the OE default files, I figured it would be cleaner to just copy them over in my personal meta file. Oct 24 11:39:21 it isn't cleaner :) Oct 24 11:40:05 koen, ok, so I should modify the existing OE files and then just generate a personal patch file to apply to newer versions of OE when I update. Oct 24 11:40:20 or use BBCOLLECTIONS Oct 24 11:40:53 Gerrath: just have a scond directory in your bbpath, and ensure you pn/pr iis highe than oe one Oct 24 11:41:05 http://www.openembedded.org/user-manual&dpage=ch04s03 Oct 24 11:41:27 03mickeyl 07org.oe.dev * ra7356e33... 10/ (2 files in 2 dirs): sharprom-toolchain 1.0 PROVIDE your stuff only for DISTRO=sharprom-compatible Oct 24 11:41:29 damn phone keyboard Oct 24 11:41:34 03hrw 07org.oe.dev * r6d96c62d... 10/ (1 conf/distro/celinux-test.conf): celinux-test: set fixed glibc version Oct 24 11:41:38 03hrw 07org.oe.dev * rf6cde9cc... 10/ (1 conf/distro/celinux-test.conf): celinux-test: set fixed toolchain stuff to not get sharprom crap info Oct 24 11:41:44 03mickeyl 07org.oe.dev * r08caa01f... 10/ (1 packages/qemu/qemu-native.inc): qemu-native * make do_stage more reliable Oct 24 11:41:49 03hrw 07org.oe.dev * rcaf46dcf... 10/ (1 conf/distro/celinux-test.conf): celinux-test: bump omap5912osk kernel to 2.6.18-omap1 Oct 24 11:41:50 CIA-4: fuck Oct 24 11:41:55 03hrw 07org.oe.dev * rfb2a1b3e... 10/ (15 files in 7 dirs): Oct 24 11:41:55 openembedded-essentials: package for Debian compatible distributions to get all Required Software installed Oct 24 11:41:55 It probably need some work but it is good start. Oct 24 11:42:00 03lenehan 07org.oe.dev * r0178f355... 10/ (1 packages/patch/patch.inc packages/patch/patch_2.5.9.bb): Oct 24 11:42:00 patch 2.5.9: Use alternatives for patch since busybox also provides a patch Oct 24 11:42:00 command. As per #1472. Oct 24 11:42:05 03lenehan 07org.oe.dev * ra173ba23... 10/ (1 packages/findutils/findutils_4.1.20.bb): Oct 24 11:42:05 findutils 4.1.20: Use alternatives for find and xargs since these commands Oct 24 11:42:05 are also provided by busybox. Closes #1472. Oct 24 11:42:10 03lenehan 07org.oe.dev * rf5c0d86d... 10/ (1 packages/havp/havp_0.82.bb): havp 0.82: Remove the .debug directory from the main package. Oct 24 11:42:13 koen, XoA|gprs. thanks. Oct 24 11:42:16 03lenehan 07org.oe.dev * r2ec4d476... 10/ (1 packages/p3scan/p3scan_2.9.05d.bb): p3scan 2.9.05d: Remove the .debug directory from the main package. Oct 24 11:42:28 03koen 07org.oe.dev * r09f7b1fc... 10/ (1 conf/distro/include/angstrom-glibc.inc): angstrom-glibc: set preferred provider for virtual/libc Oct 24 11:42:39 03koen 07org.oe.dev * r4228f36c... 10/ (1 conf/distro/include/angstrom-glibc.inc): angstrom-glibc: remove stray space Oct 24 11:42:42 hrw|work: I fixed a bug in the cia-script, but forgot to update the leaves file Oct 24 11:42:44 24 14:02 Ignoring ALL from CIA* Oct 24 11:45:15 * koen waits for glibc to finish generating locales Oct 24 11:45:52 hi fooks Oct 24 11:46:09 hey likewise Oct 24 11:46:21 whoohaa, it looks like bug squashing day? Oct 24 11:46:37 or are you guys introducing new bugs to solve later? Oct 24 11:46:38 likewise: no, it looks like CIA flood days Oct 24 11:50:04 koen: gnumeric 1.7.2 need pkgconfig 0.18+ Oct 24 11:53:23 hrw|work: feel free to add 0.21 :) Oct 24 11:54:25 yeah sure Oct 24 12:04:24 koen, I'm assuming the higher the BBFILE_PRIORITY the higher its priority, in other words I would put my personal bb directory at a higer priority than the OE so that my bb files are the ones that get built. Oct 24 12:04:41 Gerrath: yes Oct 24 12:05:43 "and yes I am also a Trolltech employee, so there is no lack of communication between the two groups/community" Oct 24 12:05:46 ouch Oct 24 12:08:43 zecke|food: he means the 'lorn community' Oct 24 12:09:06 koen: This sounds like Erich Honecker Oct 24 12:09:44 hrw|work: http://blogs.qtdeveloper.net/archives/2006/10/23/the-death-of-opie-is-greatly-exaggerated/ Oct 24 12:09:53 hrw|work: check Daniel :) Oct 24 12:11:12 heh Oct 24 12:11:42 funny how users can't grok "incompatible due to lazy vendors" Oct 24 12:13:07 zecke|food: will reply after food Oct 24 12:13:51 funny how you suspect someone is clueless the moment he writes "I read from oesf forums.." Oct 24 12:14:10 suihkulokki: right :) Oct 24 12:16:51 hrw|work: I just replied Oct 24 12:30:55 zecke|food: what do you think of my patch to insane.bbclass? Oct 24 12:32:17 13:49 < zecke|food> koen: looks good Oct 24 12:32:43 zecke|food: I'm waiting on glibc to complete the locales before committing Oct 24 12:43:43 mickey|zzZZzz: get your creditcard ready :) Oct 24 12:58:46 03koen 07org.oe.dev * r4147a2c5... 10/ (1 classes/insane.bbclass): insane.bbclass: fix some python stuff and make error message a bit more helpfull Oct 24 12:59:43 zecke|food: can insane.bbclass do logging as well? Oct 24 13:00:02 zecke|food: my screen scrollback buffer is too small for glibc errors :( Oct 24 13:04:14 koen: sys.exit(-1) you mean? Oct 24 13:04:48 zecke|food: appending the errors to a global log file would do Oct 24 13:05:03 zecke|food: and send the erros to tinderbox/bonsai/lxr :) Oct 24 13:05:34 koen: lxr another mozilla beast :} Oct 24 13:05:41 :) Oct 24 13:06:08 koen: we need to put the css of django somewhere Oct 24 13:06:24 zecke|food: lets ask hrw|work :) Oct 24 13:07:28 koen: not for this one, it is more like mkdir /var/www/media Oct 24 13:08:13 und jetzt? Oct 24 13:11:33 koen: and then we will install some CSS files there Oct 24 13:11:48 koen: and chage settingy.py to point to http://www.openembedded.org/media Oct 24 13:12:05 zecke|food: that qtblog is moderated? Oct 24 13:12:28 it wasn't Oct 24 13:12:58 then my long post dissapeared Oct 24 13:14:16 zecke|food: what needs to go into ~/website/media? Oct 24 13:15:33 koen: good question Oct 24 13:16:11 http://www.trolltech.com/developer/task-tracker/index_html?method=entry&id=125546 Oct 24 13:16:22 nice, Qtopia tasktracker is open/oublic Oct 24 13:19:17 interesting Oct 24 13:20:33 mickeyl: http://www.apple.com/de/macbookpro/ Oct 24 13:21:17 meron? Oct 24 13:21:34 mickeyl: I think so Oct 24 13:21:47 finall :D Oct 24 13:21:49 finally, even Oct 24 13:21:51 that's good news Oct 24 13:22:05 yeah, right. core 2 duo Oct 24 13:22:25 oh time to get a new macbook :) Oct 24 13:22:54 zecke|food: and we still get the educational discount :) Oct 24 13:23:04 koen: i have rescheduled this to be my personal present after a successful public defense Oct 24 13:23:41 mickeyl: hehe Oct 24 13:23:59 hehe Oct 24 13:24:16 mickeyl: any news on the ncurses demo? Oct 24 13:24:51 unfortunately not, but i think i will do it today Oct 24 13:24:57 you know who is to blame? Oct 24 13:24:59 gtk+ Oct 24 13:25:01 and it's theming Oct 24 13:25:09 this is so horribly undocumented and cumbersume it hurts Oct 24 13:25:13 mickeyl: you mean it's lack of theming Oct 24 13:25:15 mickeyl: just set a different StyleSheet? Oct 24 13:25:41 i have spent 3 * 13 hours now to programatically recreate a layout and it's still not done Oct 24 13:25:49 mickeyl: it took me some hours to get the gpe-theme-clearlooks package into a workable state Oct 24 13:26:03 zecke|food: heh Oct 24 13:27:11 it sounds terrible. Does GTK+ have any design tools? Oct 24 13:27:26 scx: design tools, yes, theme tools, no Oct 24 13:45:53 <_law_> is there any why do output only directories with ls? Oct 24 13:47:36 not with ls Oct 24 13:47:47 i'd use find Oct 24 13:47:48 find . -type d -maxdepth 1 Oct 24 13:48:57 alias ls='find . -type d -maxdepth 1' :) Oct 24 13:49:19 <_law_> ok tnaks Oct 24 13:49:22 <_law_> thanks Oct 24 13:49:37 anyone know how can I get glib to build with linuxthreads instead of nptl? Oct 24 13:50:28 there is the GLIBC_ADDONS var, which I have used to add nptl before Oct 24 13:50:43 tkp: use glibc < 2.4 Oct 24 13:50:49 NOTE: DO PACKAGE QANOTE: Package: db-dbgERROR: QA Issue: non dev package contains .so: db-dbgNOTE: Package: dbERROR: QA Issue: non dev package contains .so: dbERROR: QA Issue: non dev package contains .so: db Oct 24 13:50:53 * koen hugs zecke|food Oct 24 13:50:54 zecke|food: really? Oct 24 13:51:03 tkp: yes. Oct 24 13:51:17 isn't linuxthreads available for 2.4 and up? Oct 24 13:51:21 koen: I love your OSX terminal :) Oct 24 13:51:31 tkp: read the release notes of 2.4 Oct 24 13:53:19 tkp: http://sources.redhat.com/cgi-bin/cvsweb.cgi/libc/NEWS?rev=1.172&content-type=text/x-cvsweb-markup&cvsroot=glibc Oct 24 13:53:26 zecke|food: ta Oct 24 13:55:57 drepper should add "The following bugs were closed as invalid during the release cycle" Oct 24 13:58:07 zecke: could you add a bonsai page that lists referenced bugs and the status of those bugs? Oct 24 13:58:29 koen: all referenced bugs? Oct 24 13:58:48 zecke: all bugs named in commit messages Oct 24 13:59:23 koen: I will need xmlrpc.cgi on the Bugzilla Oct 24 14:00:48 zecke: have you mail ggilbert(_) about that? Oct 24 14:00:55 mailed* Oct 24 14:01:01 koen: no, mail? Oct 24 14:01:16 greag -at- treke.net ? Oct 24 14:01:52 greg* Oct 24 14:03:04 ERROR: QA Issue: non dev package contains .so: libxcursor Oct 24 14:03:57 bad :) Oct 24 14:04:32 looks like we will get lot of bugraports soon Oct 24 14:07:42 hmm Oct 24 14:07:43 compared to a couple of months ago, i have the feeling that OE has stabilized a lot Oct 24 14:08:19 i keep getting less broken builds Oct 24 14:08:19 sounds like we should think about making our first snapshot any time soon Oct 24 14:08:19 mickeyl: we need the coverage matrix first Oct 24 14:08:22 yep Oct 24 14:09:34 I'm thinking about letting angstrom inherit insane.bbclass by default Oct 24 14:09:51 need to get rid of the py2.4'isms first, tough Oct 24 14:09:58 though* Oct 24 14:10:05 morning Oct 24 14:10:11 hey chouimat Oct 24 14:10:20 koen: right, let us move to py2.5 Oct 24 14:11:32 i thought about adding 2.5, but that would break PyQt/E and as long as we don't have Opie in obsolete I don't want to do that Oct 24 14:12:01 mickeyl: only 7 days left :) Oct 24 14:12:24 mickeyl: why would that break? Oct 24 14:12:55 i can't upgrade sip, python-sip, nor pyQt. the last versions with -e support are in OE, newer versions won't work. Oct 24 14:12:59 and I don't want to start Oct 24 14:13:00 python24 Oct 24 14:13:02 python25 Oct 24 14:13:03 etc. Oct 24 14:13:06 that's just annoying Oct 24 14:13:15 i will maintain _one_ Python version, not multiple Oct 24 14:14:50 mickeyl/zecke : could one of you make a list of ${PN}/providers/providees that we can feed into a database? Oct 24 14:16:42 should be easy, i don't know the current data structure, but the shell does something similar Oct 24 14:17:00 that's why I asked you :) Oct 24 14:18:32 2.5!!!! 2.5! 2.5! 2.5! 2.5! 2.5! YAY!!!! GO! Go! go! Oct 24 14:18:44 * chouimat sucks at being a cheerleader :) Oct 24 14:18:50 yeha! Oct 24 14:20:14 zecke: I'll start a build from scratch and log the output to a file Oct 24 14:25:30 mickeyl: does your move of opie to packages/obsolete include stuff like libsdl-qpe? Oct 24 14:25:58 yes. this move will affect everything qt/embedded. including zillions of qpe additional software Oct 24 14:26:06 s/will/would/ Oct 24 14:26:09 .. Oct 24 14:26:32 BBFILES = "PKGDIR/*/*.bb PKGDIR/obsolete/*/*.bb" Oct 24 14:26:44 i mean, it wouldn't even a loss for qtopia4, because all qpe stuff would have to be ported anyway Oct 24 14:27:24 right, so your move stimulates people to move to qtopia-core 4 :) Oct 24 14:27:28 ~qtopia core Oct 24 14:27:30 rumour has it, qtopia core is qt/embedded. please don't support stupid marketing tricks. Oct 24 14:27:42 ~boxer Oct 24 14:27:44 rumour has it, boxer is Sharp SL-C860, or a dog. in reality, this is an SL-C760 with a different case color. please don't support stupid marketing tricks. Oct 24 14:28:07 koen: or as the founder of Trolltech said. Qtopia Core is actually the normal Qt Oct 24 14:30:04 mickeyl: obsolete is the wrong word Oct 24 14:30:19 mickeyl: there is no replacement/easy upgrade :) Oct 24 14:30:23 hi Oct 24 14:30:24 mickeyl: I'd welcome more qt4 apps in OE Oct 24 14:30:33 mickeyl: especially if they can run on x11 Oct 24 14:30:47 koen: there are not too many yet though Oct 24 14:31:01 when i added qt 4.1.2 (or so) i add all qt4 apps i've found Oct 24 14:33:55 I plan to write Opie^WHrw^W Notes with qt4/x11 Oct 24 14:35:48 cool Oct 24 14:35:56 could you do it like omnioutliner? Oct 24 14:36:03 i'd like to have that tool on pda Oct 24 14:36:15 mickeyl: thats the tool which you used on oedem? Oct 24 14:36:25 zecke: 'http://www.openemnedded.org/bonsai/do_query' ?!? Oct 24 14:36:26 unfortunately not, omnioutliner is MAC only Oct 24 14:36:33 hi Oct 24 14:36:35 tkoutliner comes not even close Oct 24 14:36:36 yo woglinde Oct 24 14:36:40 hey woglinde Oct 24 14:36:55 mickeyl: hard to write app similiar to something which I never saw Oct 24 14:37:47 http://www.omnigroup.com/applications/omnioutliner/ Oct 24 14:38:06 mickeyl: figmet! Oct 24 14:38:12 he zecke Oct 24 14:38:29 woglinde: how man girls where in this class? 100? Oct 24 14:38:37 zecke hm Oct 24 14:38:38 80 Oct 24 14:38:43 woglinde: I have the wrong subject :} Oct 24 14:38:57 and 4 erasmus pupils from france Oct 24 14:39:02 women too Oct 24 14:40:54 jesus Oct 24 14:40:55 http://groovy.codehaus.org/ Oct 24 14:42:19 no thanks, i won't trade my : for {} Oct 24 14:42:20 :D Oct 24 14:42:27 mickeyl: have you seen http://www.delicious-monster.com/ yet? Oct 24 14:43:02 no, that looks good Oct 24 14:43:15 it's great Oct 24 14:43:29 I even bought it when it was released a year ago :) Oct 24 14:43:48 lol Oct 24 14:44:00 koen: do you know a public RTP video stream? Oct 24 14:44:05 mickeyl: it's made by ex-omni employees Oct 24 14:44:21 zecke: didn't the fluendo people have one? Oct 24 14:45:22 zecke: http://stream.fluendo.com/demo/ ? Oct 24 14:49:28 koen: that is through http :( Oct 24 14:51:41 ah, right Oct 24 14:53:04 mickeyl: I'd love to have a simple applet that connects to a rtp proxy and streams video + audio from your phone over gprs/edge Oct 24 14:53:58 koen: sounds good... the thing is... i'm not sure if the hardware makes streaming video in a convincing quality possible. Oct 24 14:54:02 (that is, the CPU) Oct 24 14:54:34 but you should certainly give that a try once we ship it Oct 24 14:55:22 * koen reads http://www.amidasimputer.com/flip-flop/ Oct 24 14:56:13 "motion sensor" is really a bad name for that Oct 24 14:56:23 yeah Oct 24 14:56:36 since it can't sense constant motion Oct 24 14:56:53 accelerometers Oct 24 14:58:19 some people need to learn about tslib plugins: http://www.oesf.org/forums/index.php?showtopic=21619&pid=144687&st=0&#entry144687 Oct 24 14:58:48 hey pH5 Oct 24 14:59:15 hey koen Oct 24 15:00:01 koen: http://bugs.openembedded.org/show_bug.cgi?id=543 looks like candidate for insane.bbclass checks Oct 24 15:00:47 yes Oct 24 15:04:10 zecke: yes to #543? Oct 24 15:04:29 yes Oct 24 15:16:11 a+ Oct 24 15:27:42 trying to compile glibc-2.3.5+cvs20050627 for a c3 based machine... Oct 24 15:28:13 but it chokes saying that tune=c3 is not a valid option Oct 24 15:28:48 presumably that option was introduced to glibc after 2.4.5 Oct 24 15:29:07 how can I remove that option for that specific package only? Oct 24 15:29:42 tkp: compiler tuning is a property of the compiler and not glibc Oct 24 15:30:00 zecke: actually... after looking at config.log I see: Oct 24 15:30:00 ccache i586-oe-linux-gcc -march=c3 -mtune=c3 -V &5 Oct 24 15:30:03 tkp: see which gcc was built and consult a GCC manual, the FSF sells printed manuals as well Oct 24 15:30:16 `-V' must come at the start of the command line Oct 24 15:30:38 tkp: make sure you look at the right error Oct 24 15:31:12 well the lasty one is configure:2712: error: cannot compute suffix of object files: cannot compile Oct 24 15:31:17 last* Oct 24 15:31:29 tkp: well that doesn't matter what it says last Oct 24 15:31:35 gcc is 3.3.4 Oct 24 15:31:53 tkp: this is autoconf, it tests stuff... and a test can be positive or negative Oct 24 15:31:55 ... Oct 24 15:31:58 *sigh* Oct 24 15:32:06 tkp: look at the error message Oct 24 15:32:13 tkp: then grep for it in the config.log Oct 24 15:35:03 sorry dude Oct 24 15:35:20 tkp: there is one thing I don't get with you guys :) Oct 24 15:35:40 Why you toss so many things around, instead of looking carefully into the problem Oct 24 15:36:11 that's called being "overwhelmed"! I know the feeling. Oct 24 15:36:18 * CosmicPenguin discovers that the git fetcher is probably a little bit more complicated then it should be Oct 24 15:36:23 tkp: okay back to your issue. So you have no clue what happens? So what is the error message you get? Oct 24 15:36:26 I'm just not entirely sure which error I should be looking into Oct 24 15:36:27 http://rafb.net/paste/results/VcFfPI30.html Oct 24 15:36:28 You don't know which direction to throw yourself in. Oct 24 15:36:37 tkp: no, I don't want to read it Oct 24 15:36:40 and it said: Oct 24 15:36:41 | checking for suffix of object files... configure: error: cannot compute suffix of object files: cannot compile Oct 24 15:36:41 | See `config.log' for more details. Oct 24 15:36:52 see config.log Oct 24 15:37:22 tkp: good, okay so i586-oe-linux-gcc is a C compile from the GNU project Oct 24 15:37:31 tkp: and it says it doesn't do mtune=c3? Oct 24 15:37:36 and in there I see: error: invalid option `tune=c3' Oct 24 15:37:37 exactly Oct 24 15:37:40 tkp: so let us go gcc.gnu.org Oct 24 15:37:46 thats what I asked about first time round Oct 24 15:38:13 tkp: and then you began tossing stuff around Oct 24 15:38:40 * zecke wonders why you such an old compiler Oct 24 15:39:00 because the software I'm trying to build requires it Oct 24 15:39:06 tkp: there is no gcc 3.3.4 release :} Oct 24 15:39:14 at least to http://gcc.gnu.org/news.html Oct 24 15:39:56 now. that's my mistaker Oct 24 15:39:58 tkp: okay, so go to gcc.gnu.org, are you there yet? Oct 24 15:40:02 that should be 3.4.4 darn it Oct 24 15:40:09 I'm there yeah Oct 24 15:40:17 go to the 3.4.6 manual Oct 24 15:41:09 http://gcc.gnu.org/onlinedocs/gcc-3.4.4/gcc/i386-and-x86_002d64-Options.html#i386-and-x86_002d64-Options Oct 24 15:41:27 or http://gcc.gnu.org/onlinedocs/gcc-3.4.6/gcc/i386-and-x86_002d64-Options.html Oct 24 15:41:49 right... thats what I was looking for. Oct 24 15:41:55 tkp: hehe, this looks like a funny problem :) Oct 24 15:43:49 tkp: any idea why mtune=c3 doesn't work even if it should be there? Oct 24 15:44:20 zecke: nope. not yet Oct 24 15:44:55 well... for a start, that page is from the 3.3.3 manual... I'm using 3.3.4 Oct 24 15:45:01 tkp: my guess is, it doesn't have mmx enabled and cs3 would require it :) Oct 24 15:45:07 I mean... page is from 3.4.4 Oct 24 15:45:12 "-march=cpu-type Oct 24 15:45:12 Generate instructions for the machine type cpu-type. The choices for cpu-type are the same as for -mtune. Moreover, specifying -march=cpu-type implies -mtune=cpu-type. " Oct 24 15:45:18 I need to run/catch the bus Oct 24 15:45:32 tkp: you want to change CFLAGS which is set somewhere in conf/* Oct 24 15:45:34 later :) Oct 24 15:45:38 later zecke Oct 24 15:45:46 so 'ccache i586-oe-linux-gcc -march=c3 -mtune=c3 -V &5' is a bit redundant Oct 24 15:45:59 (according to the gcc page) Oct 24 15:47:23 well, first off, I should actually try and build with 3.4.4 and not 3.3.4! Oct 24 15:47:31 damn typos Oct 24 15:57:55 I get this after "bitbake -n virtual/kernel": http://en.pastebin.ca/219273. Any advice? Oct 24 15:59:13 pH5: I'd suggest moving off of the bitbake development branch and getting the latest checkout from head perhaps? Oct 24 16:00:29 cu Oct 24 16:08:10 pH5: looks like a deadly typo in some conf file Oct 24 16:08:41 pH5: try stracing it Oct 24 16:22:23 yuck, stracing python? Oct 24 16:23:03 * pH5 fears Oct 24 16:24:34 it should end with the last file it tries to open Oct 24 16:24:40 that should be the one with the error in it Oct 24 16:24:40 115MB log, that's not so bad :) Oct 24 16:29:26 pH5: I crashed bitbake with 'foo ? = bar' today Oct 24 16:29:32 (note the extra space) Oct 24 16:29:39 heh, thanks a lot koen Oct 24 16:29:58 it wasn't a typo, but a missing directory Oct 24 16:30:16 that was still in BBPATH somehow Oct 24 16:34:49 right Oct 24 16:35:07 bitbake should error out a little more obvious with that Oct 24 16:39:10 Could anyone check if they got a file called fix-header inside their tmp/ tree? Oct 24 16:39:34 Im getting errors while building gcc 4.1.1 with this file. Checking inside gcc.log (buildlog) I cannot see anything about this file until it fails Oct 24 16:46:55 koen@bitbake:/data/build/koen/OE/build/tmp/angstrom$ find . -name fix-header* Oct 24 16:46:55 ./work/armv5te-angstrom-linux-gnueabi/gcc-cross-initial-4.1.1-r7/gcc-4.1.1/gcc/fix-header.c Oct 24 16:46:55 ./work/armv5te-angstrom-linux-gnueabi/gcc-cross-4.1.1-r7/gcc-4.1.1/gcc/fix-header.c Oct 24 16:46:55 koen@bitbake:/data/build/koen/OE/build/tmp/angstrom$ Oct 24 16:54:19 I know this is still listed as a bug in the bugtracker... but does any one know anything more about this one: http://bugs.openembedded.org/show_bug.cgi?id=291? Oct 24 16:56:13 koen, thats same place as I get. (only under work/ and nothing under staging/) but why the hell does it complain about it then. I assume it compiles nicely for you Oct 24 16:57:10 it does Oct 24 16:57:35 thats really weird. Oct 24 16:58:03 that's gcc for you Oct 24 16:58:14 do you hear flute music when it errors out? Oct 24 16:58:22 if so, get some stallman repellent ;) Oct 24 16:58:27 haha Oct 24 17:00:10 koen, we got any gcc experts in the house? Oct 24 17:00:39 Kristoffer, not really i assume :) Oct 24 17:00:51 TheCan, :P Oct 24 17:01:58 koen@bitbake:~/OE/monotone/org.openembedded.dev/classes$ grep "QA Issue" ~/OE/build/insane-QA.log | wc -l Oct 24 17:02:00 524 Oct 24 17:02:03 anyone ever tried using LD_ASSUME_KERNEL on an oe built distro with any success? Oct 24 17:02:21 If I attemp to do /lib/libdl.so.2, I get a segfault Oct 24 17:08:14 * koen waits for the localegeneration to complete for the 3rd time today Oct 24 17:10:32 * chouimat is waiting for emerge -e world to finish Oct 24 17:17:41 heh Oct 24 17:17:54 * koen spots someone with a LeChuck tattoo on tv Oct 24 17:22:15 * Crofton_ googles "lechuck tattoo" Oct 24 17:22:56 ~praise wikipedia Oct 24 17:22:58 All hail wikipedia! Oct 24 17:23:52 Crofton_: never played one of monkey island games? Oct 24 17:24:11 nope :) Oct 24 17:24:29 my game playing is severely limited Oct 24 17:24:31 install scummvm and get addicted :) Oct 24 17:24:34 lest I become a junky Oct 24 17:25:10 * Crofton_ played Space Invaders when it first appeared Oct 24 17:26:37 hrw|gone, looks like the 2.6.18-omap1 kernel boots Oct 24 17:28:48 http://pastebin.ca/219405 Oct 24 17:29:07 results of printf in glibc-2.5 and trying to run dropbear Oct 24 17:33:21 * Crofton_ kicks myself for not patching out the assert Oct 24 17:36:29 it would appear that the struct that the author of this code expected to be be 1 byte, is alloted 4 bytes Oct 24 17:36:36 this the padding thing beices stupid Oct 24 17:38:12 koen, you around? Oct 24 17:50:26 Crofton_: great to hear Oct 24 17:57:18 Linux omap5912osk 2.6.18-omap1 #1 Mon Oct 23 22:26:44 EDT 2006 armv5tejl unknown Oct 24 17:57:18 root@omap5912osk:~$ Oct 24 17:57:30 this glibc thing is annying Oct 24 17:58:39 Crofton_, glibc 2.5 working out nicely? :) Oct 24 17:58:45 hhe Oct 24 17:58:47 well Oct 24 17:58:54 there is an assert that fails Oct 24 18:01:51 hmmm look slike assert crept in from a Fedora branch Oct 24 18:01:57 Crofton_, :( Oct 24 18:02:47 maybe not viewcvs confuses me Oct 24 18:04:59 anyway, I need to chat with someone who understands struct packing Oct 24 18:08:10 I know you lot are trying to annoy me into posting this problem to the libc-pports list .... Oct 24 18:12:57 heh, sounds like crofton's having fun Oct 24 18:15:12 right Oct 24 18:15:35 Something is fowl in how the structure gets laid out in memory Oct 24 18:15:53 and for some reason they decided to check the layout around May Oct 24 18:17:11 ok email sent Oct 24 18:17:19 we'll see if anything comes of it Oct 24 18:25:54 each time when I build kernel for my desktop I'm thinking about making -hrwX patchset Oct 24 18:27:56 mickey|sports: can you add working qmake (non-native) into OE? Oct 24 18:28:51 hi rp Oct 24 18:28:59 morning all Oct 24 18:29:16 hey rp Oct 24 18:30:14 hi RP Oct 24 18:30:38 ~lart gmail for defaulting to rtf Oct 24 18:30:38 * ibot installs a bad bootloader on gmail and turns gmail into a brick for defaulting to rtf Oct 24 18:30:51 OK I really think I sent an email to the libc-ports list Oct 24 18:34:59 RP|hel: having parted during updater.sh phase would be nice Oct 24 18:35:24 hrw, That is true... Oct 24 18:35:43 hrw, having updater fixed to work on all devices would be a good start... Oct 24 18:35:55 (one script for all devices) Oct 24 18:36:17 one script to rule them all Oct 24 18:36:19 ;D Oct 24 18:36:43 and in the darkness bind them ;-) Oct 24 18:37:19 that would only be true if we forced 2.4 on the users... Oct 24 18:41:10 maybe in next days I will finally get spitz into usable state Oct 24 18:41:20 1G rootfs should be enough for some time Oct 24 18:41:39 hrw: spitz not usable? Oct 24 18:42:14 I've never actually repartitioned mine. I really should get on and do it... :) Oct 24 18:42:29 JustinP: I nearly filled / Oct 24 18:42:52 JustinP: and parted does not like FS which I have on / Oct 24 18:42:53 hrw: :-P mine has been repartitioned since....well, nearly forever Oct 24 18:43:37 From libc-ports: Oct 24 18:43:41 Looks somewhat like a bug. I wonder if we've only run the testsuite on Oct 24 18:43:41 EABI systems without netlink. Oct 24 18:43:59 from a Daniel Jacobowitz Oct 24 18:44:22 How do I know if I am really doing EABI? Oct 24 18:45:32 RP|hel: what is mversion.bin file found in updater.sh? Oct 24 18:50:21 hrw: I think its some part of the first 7MB of flash. We don't usually touch it Oct 24 18:50:49 I have a stupid question Oct 24 18:51:06 How do I know if I am EABI or old ABI? Oct 24 18:51:53 hm if soft-fp and non-soft-fp works Oct 24 18:52:01 Crofton_: file somebinary Oct 24 18:52:06 RP|hel: I will remove it Oct 24 18:52:14 hrw he thats easy one Oct 24 18:54:26 tmp/work/ossie-nodebooter-svn-r0/nodebooter/nodeBooter: ELF 32-bit LSB executable, ARM, version 1 (ARM), for GNU/Linux 2.4.0, dynamically linked (uses shared libs), for GNU/Linux 2.4.0, not stripped Oct 24 18:54:42 what does it mean GNU/linux 2.4.0? Oct 24 18:55:11 kernel? Oct 24 18:55:20 no .... Oct 24 18:55:28 glibc that oe compiles is 2.5 Oct 24 19:01:56 http://wiki.debian.org/ArmEabiPort Oct 24 19:02:19 ~lart packaged-staging screwing up include paths Oct 24 19:02:20 * ibot blasts packaged-staging screwing up include paths with a huge firehose then strangles packaged-staging screwing up include paths with it Oct 24 19:02:38 justinp how that? Oct 24 19:02:45 yup I was right this week looked to boring and so far it is Oct 24 19:03:12 woglinde: if I use packaged-staging, angstrom fails on gcc-cross and oz fails on glibc, both due to missing includes which are quite obviously in staging Oct 24 19:03:34 woglinde: /sda1/oe/packaged-staging.angstrom.build/tmp/cross/arm-angstrom-linux-gnueabi/include/bits/sigcontext.h:28:29: error: asm/sigcontext.h: No such file or directory Oct 24 19:11:27 woglinde: :-( I see you don't know either Oct 24 19:11:47 justin_p hm I would look in the temp file Oct 24 19:11:55 temp/log_file Oct 24 19:13:07 woglinde: are you saying you'll look? I've looked but it just says that there are missing includes Oct 24 19:13:08 RP, JustinP, koen, mickeyl: http://ewi546.ewi.utwente.nl/tmp/hrw/t/updater.sh.txt to rule them all Oct 24 19:14:00 woglinde: the staging include doesn't seem to be included in the -I for the gcc cmdline.....but I have no idea why using packaged-staging causes this Oct 24 19:15:06 I meant run.do_stage Oct 24 19:15:54 hrw, do you undertstand how to build EABI tools versus old-ABI? Oct 24 19:16:16 Crofton_: I would use angstrom for it Oct 24 19:16:46 I would also need to turn on EABI support in he kernel config? Oct 24 19:16:52 yes Oct 24 19:17:05 would this annoy you? Oct 24 19:17:08 Crofton_: eabi kernel can boot oabi if has oabi support enabled Oct 24 19:17:19 Crofton_: check how linux-openzaurus handle enabling EABI Oct 24 19:18:29 the kernel bb file? Oct 24 19:18:37 linux-openzaurus.inc Oct 24 19:20:14 morning Oct 24 19:20:19 hi xora Oct 24 19:20:23 woglinde: I really have no idea.... Oct 24 19:20:54 I really should switch to angstrom .... Oct 24 19:20:56 hi XorA Oct 24 19:20:58 RP|hel: any news of any OHies at linuxworld? Oct 24 19:21:04 XorA: http://ewi546.ewi.utwente.nl/tmp/hrw/t/updater.sh.txt to rule them all Oct 24 19:21:11 I was looking at using the flash support in directfb but it seems it requires libflash3 which does not seem to be ported to OE, has anyone played flash files with directfb before? Oct 24 19:22:56 s/libflash3/libflash Oct 24 19:23:17 hrw, Sounds good although I can't test atm... Oct 24 19:23:53 XorA, I'm not aware of anyone there. Most people aren't online now but I can ask tomorrow... Oct 24 19:24:15 RP|hel: I will look will it work on tosa Oct 24 19:24:30 RP|hel: cool, would be a laugh to meet some OHies Oct 24 19:25:51 XorA: be at next OEDEM or FOSDEM Oct 24 19:27:16 FWIW, OH will have a presence at FOSDEM and I should be there Oct 24 19:31:19 hrw: FOSDEM, when is next OEDEM planned? Oct 24 19:33:14 XorA: may 2007 Oct 24 19:35:52 booting into rescue rootfs on tosa... Oct 24 19:37:16 hrw: in North America??? Oct 24 19:38:37 chouimat: no - Poznan, Poland, Europe Oct 24 19:38:47 hrw: :( Oct 24 19:45:06 looks like the glibc-2.5 problem only exists on the old-ABI Oct 24 19:51:14 looks like this updater.sh works on tosa Oct 24 19:55:10 kernel panic... Oct 24 19:58:49 mtn: connecting to monotone.openembedded.org Oct 24 19:58:49 is the only thing i see and no eth traffic or such ? Oct 24 20:02:16 hrw: should be ok for me Oct 24 20:02:35 XorA: I changed it Oct 24 20:02:46 hrw: I mean OEDEM date Oct 24 20:02:58 hrw: I was on phone so took a while to answerr Oct 24 20:03:03 XorA: ah.. Oct 24 20:03:19 XorA: I was thinking that you say about updater.sh Oct 24 20:07:32 not specifically oe-related, but I am having trouble building something and don't know where the conf-file to tweak would be: Oct 24 20:07:39 task do_configure: failed Oct 24 20:08:21 /usr/bin/autoreconf2.50: unrecognized option `--exclude=autopoint' Oct 24 20:08:21 any ideas? Oct 24 20:08:30 mccarthy: oe builds its own autoconf that supports that option. for whatever reason, you dont have autoconf-native built. Oct 24 20:08:59 so do I "bitbake autoconf-native"? Oct 24 20:09:25 looks like maybe I do :) Oct 24 20:09:35 thanks kergoth Oct 24 20:16:09 so should i run mtn pull every day ? and then how about my private changes i made to my oe ,, will they get lost ? Oct 24 20:16:25 because pulling after some days takes ages ... Oct 24 20:16:49 (especially when I cannot even get to oe.org...like right now) Oct 24 20:41:14 http://www.linuxdevices.com/news/NS8104854546.html Oct 24 20:41:39 gd evening Oct 24 20:42:25 hi greentux Oct 24 20:42:32 hi hrw Oct 24 20:44:17 * hrw want someone who understand sharp zaurus updating process Oct 24 20:52:19 where may I find gpe-scap-1.0.tar.bz2? oe is looking for it at http://handhelds.org/pub/projects/gpe/source but it's not there Oct 24 20:52:37 gpe.linuxtogo.org? Oct 24 20:53:55 thanks hrw, I'll look there Oct 24 20:54:18 hrw, I am trying to build an EABI image and kernel Oct 24 20:54:57 I added the code from the oz kernel inc to the omap1 kernel inc Oct 24 20:55:05 I'm trying to understand how to simplify zaurus updater script Oct 24 21:04:13 ibot: 0x1c00000 + 0 Oct 24 21:04:14 ping koen Oct 24 21:05:10 bc Oct 24 21:15:39 re Oct 24 21:21:26 sharp 2.4 suxx totally on tosa Oct 24 21:21:32 03xora 07org.oe.dev * r40d2d66d... 10/ (1 packages/glib-1.2/glib-1.2-native_1.2.10.bb): Oct 24 21:21:32 glib-1.2-native : apply patch from lenehan@twibble.org for staging Oct 24 21:21:32 problem. Thanks for that one. Oct 24 21:29:48 * v8jlene is back Oct 24 22:00:26 hmm.. new updater.sh works good on c7x0 but fails on tosa Oct 24 22:28:36 cu Oct 24 22:28:43 anyone have any advice on how I can copy my home directory of my latop Oct 24 22:28:46 wo using nfs Oct 24 22:35:54 Crofton: Samba? Oct 24 22:36:14 nfs? Oct 24 22:36:34 I don't use samba much Oct 24 22:36:50 Crofton: Where are you trying to copy stuff to? Oct 24 22:36:55 what does rsync do? Oct 24 22:37:05 another linux box Oct 24 22:37:15 I need to archive my home dir for a reinstall Oct 24 22:37:38 Crofton: tar it up, scp it over.. Oct 24 22:37:48 no space on laptop :) Oct 24 22:38:09 Crofton: ahh, rsync might work, I have never used it. Oct 24 22:38:33 hm scp has a -r option though Oct 24 22:39:07 wait, I think tar can work over network as well... Oct 24 22:39:46 At home I use samba to mount a windows share, since my wife's windows box has the biggest hdd's...:) Oct 24 22:40:10 my linux machines have bugger drives Oct 24 22:41:41 rsync might be a good bet tho...looking at the man-page, it also can take a user@host:dest format Oct 24 22:42:39 but I haven't installed rsync on the latop Oct 24 22:47:10 Crofton: tar over ssh Oct 24 22:47:43 Crofton: or scp -r, but it's not as good as a tarball (i.e. copying permissions) Oct 24 22:47:59 yeah and scp has no option to avoid nfs mounts Oct 24 22:48:15 Crofton: http://www.cyberciti.biz/faq/howto-use-tar-command-through-network-over-ssh-session/ ^_^ Oct 24 22:49:49 perfectly simple ;-) Oct 24 22:50:28 JustinP: Wow...never thought of using cat that way....:) Oct 24 22:51:30 heh Oct 24 22:51:53 I used that to test multiple pipes in my mini-shell I wrote in college Oct 24 22:52:02 echo "foo" | cat | cat | cat | cat | cat ... Oct 24 23:22:09 Crofton: I usually use something like this to copy stuff over ssh: tar -cppf - | ssh othermachine "cd && tar -xvppf -" Oct 24 23:22:44 ... which is basically on that page that JustinP pointed out ;) Oct 24 23:28:23 I seriously need to reread man pages Oct 24 23:28:39 I've been doing linux stuff since like kernel version 0.95a Oct 24 23:28:58 so I haven't kept up with improvements in many areas :) Oct 24 23:30:54 bleh Oct 24 23:30:58 I'm getting old Oct 24 23:31:23 * koen remembers times where guinness wouldn't make the world spin Oct 24 23:31:40 heh Oct 24 23:32:00 I remember when I could stay up all night carrying on Oct 24 23:32:26 now I could fufill wild fantasies with young women Oct 24 23:32:31 and not go to prison Oct 24 23:32:48 koen, what is the secret to building an EABI ARM system? Oct 24 23:32:52 hmm... gpe-image built on .dev, but there's no kernel or modules in the rootfs. Any ideas? Oct 24 23:33:31 Crofton: 0.95a.. you old timer! I didn't start using linux till 0.95c ;) Oct 24 23:34:30 ah the good old days Oct 24 23:34:38 Crofton: check TARGET* in the angstrom configs Oct 24 23:34:40 bined'ing the kernel to change root Oct 24 23:34:57 in distro? Oct 24 23:35:38 koen@bitbake:~/OE/monotone/org.openembedded.dev$ grep TARGET conf/distro/include/angstrom-glibc.inc Oct 24 23:35:38 TARGET_OS = "linux${@['','-gnueabi'][bb.data.getVar('TARGET_ARCH',d,1)=='arm']}" Oct 24 23:35:45 Crofton: that's the magic Oct 24 23:36:23 mreimer: h2200.conf says MACHINE_EXTRA_RDEPENDS = "mtd-utils h2200-bootloader kernel" Oct 24 23:36:35 mreimer: so the kernel should be there (when using task-base) Oct 24 23:36:44 hey koen Oct 24 23:37:07 koen: bitbake built handhelds-pxa-2.6, but for some reason do_rootfs didn't install it Oct 24 23:37:36 Cab I just set TARGET_OS = linux-gnueabi in local.conf? Oct 24 23:37:52 koen: ah, 'build task-base' is building more stuff. Thanks for the clue. Oct 24 23:38:33 Crofton: if you're only building for ARM, yes Oct 24 23:38:40 ok Oct 24 23:39:02 I am trying to build an eabi system to see if that gets around the glibc poblem I had Oct 24 23:41:50 Crofton: if you're short on time you could pluck an image from http://angstrom-distribution.org/unstable/images/20061012/ and install your EABI kernel + kernel modules Oct 24 23:43:56 have you checked to see if dropbear starts on that Oct 24 23:44:02 I have time anyway :) Oct 24 23:52:01 Crofton: I have not used dropbear on a pure glibc 2.5 system, but dropbear built against 2.4 with 'ipkg update ; ipkg upgrade' to glibc 2.5 works Oct 24 23:52:15 Crofton: note that eabi itself mandates stricter alignment Oct 24 23:52:48 it looks like the issue is contained in glibc to me Oct 24 23:53:06 I'll know more in the morning Oct 24 23:55:41 Crofton: 'msh' in #monotone on irc.oftc.net should be the main dropbear dude Oct 25 00:04:54 man Oct 25 00:05:01 we have a lot of QA issues: http://ewi546.ewi.utwente.nl/tmp/QA-issues.txt Oct 25 00:30:03 * koen zzzzzzz Oct 25 00:30:09 night koen Oct 25 00:35:19 night koen **** ENDING LOGGING AT Wed Oct 25 02:59:57 2006