**** BEGIN LOGGING AT Thu Jan 19 02:59:58 2006 Jan 19 03:00:11 <[lala]> RP: i can imagine that. but i can help testing if u send me a patch Jan 19 03:00:43 [lala]: Its certianly possible without the hardware - just more difficult :) Jan 19 03:01:08 [lala]: What were the steps you took to trigger the oops? Jan 19 03:02:16 <[lala]> RP: essentially you have the hciattach running, then you unplug the card and try to kill the still running hciattach Jan 19 03:03:08 <[lala]> RP: killing hciattach only works properly as long as the card is still inserted Jan 19 03:03:34 [lala]: So it looks like hciattach holds something open and tries to access the card after removal of the device... Jan 19 03:03:53 <[lala]> yeah, i suppose so Jan 19 03:03:57 That'll be why the uart function causes the oops Jan 19 03:04:13 its trying to access hardware that doesn't exist... Jan 19 03:04:26 <[lala]> it also works if you pull the card and leave the hciattach alone, insert the card and then kill hciattach - no oops Jan 19 03:05:23 <[lala]> of course, cardmgr wants to start a new hciattach after reinsertion, but that one fails naturally since still a hciattach is running Jan 19 03:06:00 [lala]: You could also report this bug to the mainline linux developers as you're using a very recent kernel Jan 19 03:06:26 <[lala]> but the old running hciattach is no good either, a hcitool scan doesnt work with that process Jan 19 03:06:54 hciattach should really exit once the card is removed Jan 19 03:07:12 <[lala]> that was what i expected as well :) Jan 19 03:07:42 It just sounds like the driver is badly written... Jan 19 03:24:38 Hm.. is anyone having problems with libtool-native from .dev? Jan 19 03:27:55 Some packages fetch sources from cvs... Do they actually check whether the CVS has changed before deciding to rebuild? Jan 19 03:29:17 only if $PV or $PR or $CVSDATE have changed Jan 19 03:30:34 How is CVSDATE determined? Jan 19 03:32:52 DataBeaver: if SRCDATE is not set then current date in GMT is used Jan 19 03:34:59 And it checks that the cvs actually has some changes before rebuilding the package? Jan 19 03:36:55 well, yes. It's using cvs update IIRC Jan 19 03:37:27 k Jan 19 03:47:03 03coredump 07org.oe.dev * re89614e3... 10/packages/initscripts/ (2 files in 2 dirs): initscrpts: Removed -de extension. Jan 19 03:47:07 03coredump 07org.oe.dev * r74fd84d0... 10/packages/keymap-locales/ (3 files in 3 dirs): Add keymap-locales package. Jan 19 04:00:08 Some kind of progress meter would be nice... Jan 19 04:03:52 XorA_: wb Jan 19 04:05:36 ADSL seems to be totally dead now Jan 19 04:09:49 <[lala]> CoreDump|home: that keymap pkg is a good idea i think. i tried hooking up an usb kbd some times ago but wasn't much useful without a proper keymap Jan 19 04:11:16 i c. Problem is that such a keymap would probably change both, on-board keyboard _and_ USB keyboard. Might mess things up quite a bit Jan 19 04:30:35 * aoe ended up patching the usb keyboard driver builtin map Jan 19 04:31:08 of course if the zaurus keyboard driver delivered reasonable keypresses, this wouldn't be necessary Jan 19 04:31:22 right Jan 19 04:35:13 koen|food: ping Jan 19 04:46:16 <|anja|> hello I have compiled gcc from openembedded few days ago, and I have problem my c can not create binaries on my pda, because my libgcc uses hardware fpu but my gcc uses software. can anybody hep please? Jan 19 04:47:00 hi zecke Jan 19 04:47:25 hey Jan 19 04:47:45 |anja|: read the local.conf.sample Jan 19 04:47:53 |anja|: specially on TARGET_FPU ;) Jan 19 04:50:32 qt/e is at 2.3.12 in snapshots.. Jan 19 04:52:40 <|anja|> |zecke|: I had set TARGET_FPU = "soft" so where is the problem???? Jan 19 04:53:41 |anja|: then please rephrase your question. You run gcc on your PDA? Jan 19 04:53:53 |anja|: or you have a libgcc on your PDA that uses hard float? Jan 19 04:54:01 |anja|: and you compiled using soft float? Jan 19 04:54:40 <|anja|> |zecke|: yes i run it on pda, and I hope that i compiled it using float Jan 19 04:55:10 |anja|: you run gcc on your PDA or an application compiled with gcc? Jan 19 04:55:25 <|anja|> run gcc on pda Jan 19 04:56:02 |anja|: where is your libgcc from? Jan 19 04:57:13 <|anja|> compiled averything is compiled with bitbake exept kernel using acer N30 Jan 19 04:57:58 hmm there is one thing I do not get... Jan 19 04:58:17 if you compiled an image using OE and you have TGARGET_FPU = "soft" why is libgcc hard float compiled? Jan 19 04:58:51 <|anja|> yes and that is the problem Jan 19 04:59:09 |anja|: I have not seen that problem so far Jan 19 04:59:31 |anja|: I would propose you look into the config.log's of gcc Jan 19 05:03:42 <|anja|> |zecke|: Another question. I am new to bitbake, how do I recompile packages, please? Jan 19 05:05:19 bitbake -c clean -b $path_to_.bb ; bitbake -b $path_to_.bb Jan 19 05:06:04 <|anja|> thans Jan 19 05:07:26 np Jan 19 05:15:34 <|anja|> does any have no idea why it is not working touchscreen in my new gpe-image??? how gpe finds touchscreen where is the config or it is all done by detect-stylus by setting variables Jan 19 05:16:00 |anja|: normally detect-stylus finds a mouse input device Jan 19 05:17:41 <|anja|> |zecke|: so problem is that detect-stylus is not in my rc scripts? image was created by bitbake gpe-iamge Jan 19 05:18:02 03tmbinc 07org.oe.dreambox * r172de177... 10/packages/dreambox/dreambox-dvb-modules.bb: dreambox-dvb-modules: update to version 20060116 Jan 19 05:18:07 03tmbinc 07org.oe.dreambox * rad640a85... 10/packages/dreambox/dreambox-dvb-tools-v3.bb: dreambox-dvb-tools-v3: update to version 1.1 Jan 19 05:18:11 03tmbinc 07org.oe.dreambox * raf7cfe1a... 10/packages/dreambox/dreambox-secondstage.bb: dreambox-secondstage/dm7025: update to version 41 Jan 19 05:18:15 03tmbinc 07org.oe.dreambox * r042d2e31... 10/packages/enigma2/enigma2.bb: enigma2: update to version 20060117, add libfribidi DEPEND Jan 19 05:18:17 hi zecke Jan 19 05:18:19 03tmbinc 07org.oe.dreambox * r4449e4d3... 10/packages/twisted/twisted_2.1.0.bb: twisted-2.1.0: correct typo Jan 19 05:18:22 |anja|: no, /etc/profiles/ should have a TSLIB_DEVICE veriable Jan 19 05:18:23 03tmbinc 07org.oe.dreambox * r843261ec... 10/packages/twisted/twisted-native_2.1.0.bb: twisted-native-2.1.0: add native version of twisted Jan 19 05:18:27 03tmbinc 07org.oe.dreambox * r49b9129b... 10/packages/nevow/nevow_0.5.0.bb: nevow-0.5.0: add package Jan 19 05:18:32 03tmbinc 07org.oe.dreambox * r3b7f95d5... 10/packages/twisted/twisted-web2_0.1.bb: twisted-web2: add package Jan 19 05:18:34 |anja|: and the content is set by detect-stylus Jan 19 05:18:36 03tmbinc 07org.oe.dreambox * re7cd5322... 10/packages/twisted/twisted-web_0.5.0.bb: twisted-web: add package Jan 19 05:18:40 03tmbinc 07org.oe.dreambox * r6fb302f7... 10/packages/twisted/twisted-web-native_0.5.0.bb: twisted-web-native: add native package Jan 19 05:21:08 zecke: Did you have a chance to look at the two bugfixes for bitbake? Jan 19 05:21:30 RP: when I left there was only one?! Jan 19 05:22:11 zecke: There's circular dependencies and the fact PREFERRED_PROVIDER wasn't working properly Jan 19 05:22:36 RP: point me to the URL's please, will apply them right away Jan 19 05:23:37 zecke: http://sam.rpsys.net/bitbake-circular-r3.patch http://www.rpsys.net/openzaurus/temp/bitbake-prefprov-r2.patch Jan 19 05:24:49 zecke: The prefprov one is almost certainly more correct. With the circular one, I wish I could find a nicer way but I don't think anything nicer is going to happen easily... Jan 19 05:25:14 circular patch looks right Jan 19 05:26:09 RP: I will need to test 'preferred' with BB Collections first Jan 19 05:26:57 RP: could 2.6.14-rc1 (default in oz354fam083 branch) get some 'make it quiet' patches? Jan 19 05:26:59 zecke: Ok, but keep in mind it will only have any affect when using buildRProvider Jan 19 05:27:27 hrw|work|busy: Add "quiet" to the kernel commandline? Jan 19 05:27:50 RP: could you make diff emit the 'method' it patches as well? :} Jan 19 05:28:17 RP: what I wonder can discriminated be removed completely? Jan 19 05:28:35 RP: added it - it show some messages on suspend/resume anyway - will look at them in a minute Jan 19 05:28:59 RP: those wm8731_write messages Jan 19 05:29:00 zecke: I made the patches using svn diff Jan 19 05:29:26 hrw|work|busy: The wm8731_write messages should be commented out - they're debugging only Jan 19 05:29:29 RP: and 'CPU0: I cache' etc Jan 19 05:30:09 hrw|work|busy: I think the cpu message was removed in later kernels... Jan 19 05:30:15 (and in mainline) Jan 19 05:31:14 RP: sure - but we will use 2.6.14-rc1 for release Jan 19 05:31:37 RP: I have one question (comitted circular patch anyway) Jan 19 05:31:47 if build_depends and build_depends.count(fn) > 1: Jan 19 05:31:48 and Jan 19 05:31:59 + if self.tryBuild(fn, item, buildAllDeps, build_depends + [fn]): Jan 19 05:32:05 look a bit unbalanced to me Jan 19 05:32:39 Imagine I have changed a .bb file by hand and now I want to run an update and monotone resuse to update the file because it's different than the last one it wrote. Jan 19 05:32:41 build_depends + [fn] would either lead to a circular dep or build_depends is empty?! Jan 19 05:32:50 How can i force monotone to updatee the file? Jan 19 05:32:57 Cobelius: commit your changes Jan 19 05:34:16 RP: the PREFERRED PROVIDER needs some testing by myself before comitting Jan 19 05:34:37 RP: I had broken that part of bitbake way too often already Jan 19 05:43:54 zecke: I understand :) Jan 19 05:44:16 RP: hehe Jan 19 05:45:05 zecke: re: the build_depends, the code looks correct - if the same function is in build_depends twice, it gives an error Jan 19 05:47:09 s/function/filename/ :) Jan 19 05:48:24 zecke: The only consolation re the PREFERRED_PROVIDER is that as it stands the, the code is totally useless so anything will be better than that ;-) Jan 19 05:48:53 (in the rdepends case) Jan 19 05:50:57 RP: at least it passed the bb collection testcase Jan 19 05:51:20 zecke: But would that trigger the Runtime provider code? Jan 19 05:51:43 RP: no, but I didn't know which method you patched ;) Jan 19 05:51:50 RP: I thought it was findBestProvider Jan 19 05:52:48 zecke: no, buildRProvider - the PREFFERRED_PROVIDER code is different beetwen build and runtime cases... Jan 19 05:57:27 hrw|work|busy: the TTF stuff looks good.No problems w/ opie-console Jan 19 05:57:43 morning gentlemen Jan 19 05:57:51 hi alan|home Jan 19 05:57:52 great Jan 19 06:02:09 anyone working at Philips here? Jan 19 06:03:25 zecke wants his TV to be repared for free... ^^ Jan 19 06:06:14 alan|home: No I want a job in a Philips research facility in Eindhoven ;) Jan 19 06:06:24 * zecke wants to leave Germany as soon as possible Jan 19 06:07:03 ok, so may i ask something too ? Jan 19 06:07:17 no you musn't Jan 19 06:07:34 alan|home: what is your question? Jan 19 06:07:46 RP: when will you arrive in Brussel? Jan 19 06:07:51 anyone working anywhere needs a definitely non developper to work for him anywhere in the world ? Jan 19 06:08:07 wanna be a lab monkey? Jan 19 06:08:23 hehe... I work in a school, isn't that the same ? Jan 19 06:09:10 03coredump 07org.oe.dev * r6b230bb5... 10/packages/altboot/files/ (altboot-menu/Advanced/40-bootNFS init.altboot): altboot: Fixed a typo preventing any NFS boots Jan 19 06:10:13 flashing ELPP kernel.. Jan 19 06:10:39 <[lala]> CoreDump|home: does that altboot also work for borzoi? Jan 19 06:11:15 that would require some porting. Currently not, sry Jan 19 06:13:06 well, does borzoi have its root on HDD or flash? Jan 19 06:13:19 <[lala]> with oz-devel: hdd Jan 19 06:13:46 with 3.5.4: flash Jan 19 06:13:55 ah Jan 19 06:14:16 so altboot might work with akita settings in 3.5.4 out-of-the-box Jan 19 06:15:51 is the "monotone update -rjan2006prebreakage" still required ? Jan 19 06:16:14 alan|home: yes if you do not have latest bitbake+patches Jan 19 06:16:44 zecke: the cvs bitbake is already patched ? Jan 19 06:17:26 alan|home: no Jan 19 06:17:33 ok. thanks zecke Jan 19 06:17:44 ~lart mickeyl for ELPP="yes" instead of ELPP=1 Jan 19 06:17:44 * ibot plops mickeyl into a giant vat of herring for ELPP="yes" instead of ELPP=1 Jan 19 06:19:03 huh? I'm using the head branch with svn bitbake and it works quite fine... Jan 19 06:19:32 DataBeaver: there is one pending patch Jan 19 06:21:08 And I'm supposed to need that? Jan 19 06:23:51 Does anyone know how "Ctrl-Alt-Fn" is mapped? Can't find it in my dumpkey on the desktop but it still works heh Jan 19 06:25:05 zecke: The plane lands at 13:00 on the Friday and leaves at 15:xx on the Monday if I remember correctly Jan 19 06:25:23 ~spitz Jan 19 06:25:29 zecke: FOSDEM is just on the Sat/Sun but I'm there slightly longer than that Jan 19 06:25:32 somebody said spitz was the Sharp SL-C3000, or a dog Jan 19 06:34:02 c7x0 + ELPP + gpe-bootsplash == funny Jan 19 06:37:41 03coredump 07org.oe.dev * r282da9e1... 10/packages/initscripts/ (4 files in 4 dirs): Jan 19 06:37:41 initscripts: Jan 19 06:37:41 - Allow VT change by pressing Fn-LeftArrow and Fn-RightArrow Jan 19 06:37:41 - Sync Borzoi and Spitz keymaps with Akita Jan 19 06:44:37 anyone try to build qt-embedded-2.3.10? Jan 19 06:45:50 AvengerMoJo: yesterday I built it for 6 machines Jan 19 06:46:40 lunch time.cu Jan 19 06:46:54 hrw|work|busy: .... :( I couldn't get it to build at all Jan 19 06:47:09 * AvengerMoJo need to hit himself with a brick Jan 19 06:47:48 hrw|work|busy: anything special I need to know? Jan 19 06:52:48 AvengerMoJo: no, what is your error? Jan 19 06:59:32 zecke: I'm not even cross compile Jan 19 06:59:48 zecke: I'm just trying to build it and see it work on my x86 enviornment Jan 19 07:00:15 zecke: it fail right away on .. Jan 19 07:00:40 zecke: something like QRecList not declair ... let me check Jan 19 07:01:21 AvengerMoJo: well TT does only support gcc 1.4 and older for Qt and Qtopia Jan 19 07:02:07 TT? Jan 19 07:02:33 Trolltech Jan 19 07:02:45 oh ... you mean gcc 4.1? Jan 19 07:02:49 Opie has some Qt patches including build fixes for gcc3.4 and gcc4 Jan 19 07:03:13 AvengerMoJo: no they work and test only with gcc 2.95.2 or such (which I label as gcc1.4) Jan 19 07:03:35 oh Jan 19 07:03:37 hum ... Jan 19 07:03:58 I see what you mean ... you mean my gcc is too new I need 2.95 .. Jan 19 07:04:13 yeah .. I'm using 4 Jan 19 07:04:34 AvengerMoJo: no apply the patched we have created?! Jan 19 07:05:07 well not yet .. I want to see it build from their source or snapshot just for testing :) Jan 19 07:05:27 I guess I need to get the source from you oe instead hehe :P Jan 19 07:05:30 have fun Jan 19 07:05:41 zecke: :P Jan 19 07:06:09 zecke: I'm just trying to build a testing enviornment Jan 19 07:06:16 using qvfb Jan 19 07:06:36 zecke: any idea? Jan 19 07:06:44 AvengerMoJo: what did I say Jan 19 07:07:03 :P Jan 19 07:07:12 already :P do it right away Jan 19 07:07:16 thanks Jan 19 07:07:31 AvengerMoJo: wait a second ;) Jan 19 07:07:53 zecke: ? Jan 19 07:07:54 AvengerMoJo: MACHINE="native" bitbake qte Jan 19 07:08:11 AvengerMoJo: I use qt/e from OE to develop OPIE stuff Jan 19 07:08:33 morning Jan 19 07:08:43 hrw|work|busy: oh :) didn't know Jan 19 07:08:44 hi chouimat Jan 19 07:09:14 AvengerMoJo: I use MACHINE='native' and then normal build of opie stuff with bitbake + testing in qvfb Jan 19 07:09:33 cya later Jan 19 07:10:10 hrw|work|busy: ....I see ... I know want to build the binary just for test ... .can I some how copy it out? Jan 19 07:10:37 hrw|work|busy: can I use my qvfb on my currect distribution install qt3? Jan 19 07:10:47 I never did it this way Jan 19 07:10:49 AvengerMoJo: yes Jan 19 07:11:41 but I will still need the qpe library right Jan 19 07:11:52 so I can build that from oe Jan 19 07:11:53 AvengerMoJo: which you can build using bitbake Jan 19 07:12:26 and I don't have to worry about the path or anything? Jan 19 07:12:43 just cp those binary library? :) Jan 19 07:12:54 AvengerMoJo: and use proper LD_LIBRARY_PATH Jan 19 07:14:55 thanks Jan 19 07:15:12 I just wonder why it is so difficult to compile qpe ... :( Jan 19 07:16:04 AvengerMoJo: difficult? 'bitbake libqpe' is enough iirc Jan 19 07:24:39 thanks Jan 19 07:33:09 well gpe built ... Jan 19 07:33:28 i am not quite sure how to detect the presence of SYS IPC tho ... Jan 19 08:09:56 IMAGE_ROOTFS_SIZE_ext2 = "2097152" <- And I was wondering why I'm getting "Not enough memory" on a 2GB box heh Jan 19 08:19:24 CoreDump|home: all recent altboot changes should go into branch? Jan 19 08:19:38 yes, please Jan 19 08:19:45 ok Jan 19 08:22:31 CoreDump|home: pushed Jan 19 08:22:38 ty Jan 19 08:23:02 CoreDump|home: which exactly fixes about libsdl should go into branch? Jan 19 08:23:14 sec Jan 19 08:24:20 probably the push that introduced file://fixlibs.patch;patch=1" Jan 19 08:24:22 ok Jan 19 08:24:47 could you isolate that patch and add it to #350? I'm quite busy at work Jan 19 08:25:02 ok Jan 19 08:25:10 thx Jan 19 08:25:20 also other things. Jan 19 08:26:07 I'm thinking about adding openzaurus related bluetooth config which will have HID disabled - most of users does not use due to lack of BT and it give error on start now Jan 19 08:27:19 i agree Jan 19 08:33:18 hmm I can't identify the revision that brought this patch Jan 19 08:36:10 Revision: e5ed779bbcac2d7c6e324cc7fcabb5259dbed245 Jan 19 08:36:32 heh Jan 19 08:36:43 just cp .dev Jan 19 08:36:53 works perfectly Jan 19 08:41:41 something more (from SDL stuff) need fixing? Jan 19 08:42:40 ~curse php for weak unicode support Jan 19 08:42:42 May the fleas of a thousand camels infest your most sensitive regions, php for weak unicode support ! Jan 19 08:47:55 not that I knew of Jan 19 08:49:07 without it we still get libsdl-qpe and libsdl-x11 which are not replaceable? Jan 19 08:52:13 without the patch libsdl-qpe comiles fine, but all appsneeding SDL will not Jan 19 08:53:01 aha Jan 19 08:53:47 don't ask me why =) SDL from .dev works as expected as far as I can tell Jan 19 08:54:04 simple a -Dmain=SDLMain was left behind Jan 19 08:55:05 i wish Akitas graphics card was a weensy little bit faster Jan 19 09:02:56 zecke: pong Jan 19 09:03:23 koen: hi Jan 19 09:05:10 hey hrw|work|busy Jan 19 09:05:53 koen: I cleaned a bit my builddirs Jan 19 09:06:33 hm.. branch changeset tree looks interesting... too much merges I did lately Jan 19 09:09:15 I'll build another set of familar rc images this weekend Jan 19 09:11:31 last push Jan 19 09:12:56 * law_ flashes his collie Jan 19 09:15:57 law_: after flash you could update altboot package Jan 19 09:16:55 hrw|work|busy: ok Jan 19 09:21:22 hrw|work|busy: opie-image (collie) is an tar.bz2 not .bin Jan 19 09:21:53 law_: opie-image-initrd.bin: bzip2 compressed data, block size = 900k Jan 19 09:22:09 yep. will fix it Jan 19 09:22:48 opie-image-initrd.bin: Linux jffs2 filesystem data little endian Jan 19 09:22:50 i?ll try gpe firs Jan 19 09:22:51 fixed Jan 19 09:23:11 law_: thx for noticing Jan 19 09:23:19 hrw|work|busy: np Jan 19 09:25:31 openzaurus.org is funny: "Since this is a Debian based distro can we import someprogram.arm.rpm?" Jan 19 09:25:40 ~curse rpm Jan 19 09:25:45 May you be reincarnated as a Windows XP administrator, rpm ! Jan 19 09:27:05 where the hell did someone find netwinder rpms these days :-) Jan 19 09:27:37 koen|away: where it is? Jan 19 09:32:32 hr http://openzaurus.org/wordpress/about/ Jan 19 09:33:01 * koen|away really away now Jan 19 09:38:54 I will moderate that comment Jan 19 09:39:31 XorA: netwinder.org, I guess Jan 19 09:46:05 found first problem: after every restart clock looks like this: http://handhelds.org/scap/port.27152.png Jan 19 09:46:26 after one minute this changes :-) Jan 19 09:47:33 changes to.. what? Jan 19 09:48:03 raduga: full clock Jan 19 09:48:26 possibly its still loading? Jan 19 09:49:01 raduga: no if the times changes i scale right i think Jan 19 09:50:05 strg +c also doesnt work with collie (in gpe) Jan 19 09:50:52 fn +c | fn + shift + c | calendar +c doesnt work for strg + c:-( Jan 19 10:00:51 hrw|work|busy: suspend/resume scripts for nokia bluetooth cards are not available Jan 19 10:10:26 law_: ah.. that one.. will add it soon - have one change to bluez-utils in queue Jan 19 10:18:05 mithro: hi Jan 19 10:18:10 where are you? Jan 19 10:19:45 Istanbul! Jan 19 10:22:41 i'm still pondering about Fosdem Jan 19 10:26:38 mithro: its a good chance to see a few OE people in one place! Jan 19 10:27:07 I was about to book my flight yesterday... Jan 19 10:29:40 zecke: but? Jan 19 10:30:07 mithro: didn't knew when the others would be there :} Jan 19 10:30:22 or if I should sleep in Eindhoven (more than one hour train travel each day9 Jan 19 10:31:26 what/where/when are other people going to be there? Jan 19 10:31:42 I would need to get a flight from Rome up to where Fosdem is Jan 19 10:32:04 mithro: koen, mickeyl, O-hand-Crew, pb_(?) Jan 19 10:32:06 FOSDEM is on the Sat/Sun. I'll get there midday Fri and leave midday Mon Jan 19 10:32:48 lrg, greentux and do13 as well I believe Jan 19 10:36:11 cu Jan 19 10:37:20 compiled gpe-image and udev seems like not working I am missing /dev/input/event* Jan 19 10:38:51 * mithro is just thinking about cheap accom.... maybe better if share with someone? Jan 19 10:40:13 how can I say that udhcp will not be installed in my Image? Jan 19 10:41:49 yipee! svn conversion is done :} Jan 19 10:42:14 zecke: eekk svn! Jan 19 10:42:26 Head for the hills! Jan 19 10:42:29 mithro: better than that security hole called darcs :} Jan 19 10:43:15 mithro: I find it unbelievable that a month old security issue is not fixed Jan 19 10:44:51 what was the security hole? Jan 19 10:45:13 i remeber reading it and thinking that it was non-consequential/didn't effect 99% of people Jan 19 10:45:15 mithro: when applying a patch you can escape with '../../' Jan 19 10:45:36 and with a distributed scm you pull from people you do not trust too much Jan 19 10:46:34 mithro: but we have a current OE darcs as well Jan 19 10:48:05 zecke: i thought you said that it died after 631 patches? Jan 19 10:48:20 how can I say that udhcp will not be installed in my Image? Jan 19 10:48:43 mithro: well the darcs pull dies on os x with more than 300 patches Jan 19 10:49:07 mithro: but a get works Jan 19 10:49:30 ahh okay Jan 19 10:49:33 o well... Jan 19 10:49:49 as long as you don't go to svn I'll be happy ;) Jan 19 10:51:52 I have an irrational grudge against SVN :/ Jan 19 10:51:53 hmm, what is the latest on source control? Jan 19 10:52:48 Crofton: monotone 0.26 is in testing with a complete new internal architecture Jan 19 10:53:05 cool Jan 19 10:53:14 Crofton: I have git, svn and darcs uptodate Jan 19 10:53:31 which is going to win? Jan 19 10:53:36 zecke: is it much effort to keep them in sync? Jan 19 10:53:45 zecke: btw whats your postal address? Jan 19 10:54:06 I have some stuff in a local monotone branch from a few months back Jan 19 10:54:07 mithro: one way is easy... but manual work Jan 19 10:54:23 I'm going to need to re-integrate at some point Jan 19 10:54:32 but I am putting it off until things calm down some Jan 19 10:54:58 Crofton: they are all calm ;) Jan 19 10:55:03 heh Jan 19 10:55:19 well, for the OE situation to calm down also Jan 19 10:55:26 hehe Jan 19 10:55:33 and I have lots to do otherwise :( Jan 19 10:56:20 I actually like svn + trac, but I am not sure how well it will work for OE Jan 19 10:56:36 since people have conditioned themselves to work in a distributed environemnt Jan 19 10:57:30 Crofton: svn+svk ;) Jan 19 10:57:41 I'm not familiar with that Jan 19 10:57:51 Crofton: it add distribution on svn Jan 19 10:58:01 Crofton: it adds even... Jan 19 10:58:11 yeah, I have that impression Jan 19 10:59:17 Crofton: once monotone 0.26 is released we will let the SCM's battle if people can host servers... Jan 19 10:59:51 time frame? Jan 19 11:00:06 SCM's at twenty paces Jan 19 11:00:42 Crofton: none ;) Jan 19 11:03:22 Crofton: I offer tarballs of the scm's and people could use them on their installation Jan 19 11:03:41 Crofton: and people can test push, pull, merge... Jan 19 11:04:05 I wish I had some time to mess with them Jan 19 11:04:18 but I am using this time to catch up on other work Jan 19 11:04:35 although I am going to get pressure on the oe area soon Jan 19 11:27:50 okay darcs, git and svn tree are current Jan 19 11:28:35 re Jan 19 11:29:13 zecke: rock Jan 19 11:29:32 pb_: now I need to make the svn accessible some how :} Jan 19 11:29:39 yeah Jan 19 11:30:50 pb_: may I ask you one thing aout bluez in OE? Jan 19 11:30:58 zecke: maybe you could host it on handhelds.org or something Jan 19 11:31:10 hrw: I guess so. don't ask to ask, and all that. Jan 19 11:31:14 pb_: yep Jan 19 11:31:29 pb_: france said I might use h7... but I need to check back with him Jan 19 11:31:41 zecke: right Jan 19 11:32:18 I could use my home dir... Jan 19 11:32:22 pb_: I want to make one change to bluez-utils: disable hidd on default in OZ as most of users does not use/have BT in zaurus and it give 'error/warning' message during start Jan 19 11:33:10 hrw: that would be fine. might as well not even install hidd by default. Jan 19 11:33:24 (on openzaurus, that is: we want to keep it in familiar) Jan 19 11:33:52 zecke: you should be able to use proj_oe's home dir, I suppose Jan 19 11:34:00 pb_: not installing maybe for next release ;) now I have 'polishing' phase Jan 19 11:34:07 heh, righto Jan 19 11:34:43 pb_: I will also add script from bug #417 to make dtl1_cs cards working after resume (drive does not support suspend/resume at all - need eject/insert) Jan 19 11:34:48 pb_: hehe, I think I need to get added to that group Jan 19 11:34:59 hrw: that sounds good Jan 19 11:35:26 zecke: ah. you should ask ueberhacker mickeyl to send a note to admin@handhelds.org and ask for you to be added. Jan 19 11:35:52 pb_: yeah he needs to return and recover from the sunburns... Jan 19 11:35:59 first Jan 19 11:36:21 yeah Jan 19 11:36:30 when's he due back? I guess it must be at almost any minute. Jan 19 11:37:22 * pb_ go home now Jan 19 11:37:25 bbiab Jan 19 11:37:28 pb_: I think he is overdue, he must feel lost with out internet by now Jan 19 11:37:30 cya Jan 19 11:37:48 zecke: heh, yeah, indeed Jan 19 11:44:36 woglinde: hey! Jan 19 11:53:38 hmm.. 6 versions of bluez-utils to bump Jan 19 11:55:38 * RP contemplates inherited PR numbering :) Jan 19 11:57:16 hello RP. Do you know if pxa cpufreq is ever going to go into mainline? Jan 19 11:57:48 mreimer: No idea. I guess someone would need to cleanup the patch and then ask rmk about it Jan 19 11:57:52 and now another changeset for bluez-utils Jan 19 11:58:09 RP: ok, I think I'll take that up when I get a chance Jan 19 11:58:32 mreimer: It'd be nice to get it into mainline... Jan 19 11:58:43 RP: is your patch the latest greatest? Jan 19 11:58:56 mreimer: There is a patch in the zaurus tree but I don't know how it compares to the other versions out there Jan 19 12:00:01 mreimer: We need to check all the peripherals that are sensitive to clock speed are actually notified of speed changes and act accrodingly Jan 19 12:00:11 We can't be that far out as it works ok on the c7x0 though Jan 19 12:00:24 RP: yes Jan 19 12:01:49 I need to write the code for the cxx00 at some point :) Jan 19 12:02:53 RP: yes, please! those are pxa27x, right? Jan 19 12:03:03 pH5: correct Jan 19 12:03:15 pH5: There are some intel patches somewhere that have some sample code Jan 19 12:08:22 zecke: at some point it might be an idea to "git repack; git prune-packed" on the git tree to compact it a bit and make it easier to download Jan 19 12:11:26 mt is complaining about oz354fam083 having multiple heads, and suggesting a merge Jan 19 12:11:44 so I try to merge and it complains monotone: misuse: unable to resolve file conflict 'packages/qpf-fonts/qpf-dejavusans_2.0.bb' -> 'packages/qpf-fonts/qpf-dejavusans_2.2.bb' vs. 'packages/qpf-fonts/qpf-dejavusans_2.1.bb' Jan 19 12:11:49 RP: how will that work when someone pushes objects? Jan 19 12:11:54 good evening !!! Jan 19 12:12:09 but only qpf-dejavusans_2.1.bb exists Jan 19 12:12:13 what to do? Jan 19 12:12:39 RP: actually I want a worst case ;) Jan 19 12:12:45 zecke: The new ojbects just get added to the tree as normal Jan 19 12:12:57 mreimer: I'll merge branch then Jan 19 12:13:04 thanks hrw Jan 19 12:13:14 zecke: and when neccessary, someone can pack or repack it again and create more pack files Jan 19 12:13:21 mreimer: I'm pushing few things for branch and dev Jan 19 12:13:33 thanks Jan 19 12:13:47 RP: how will he push his packed tree? Jan 19 12:13:52 zecke@bitbake:~/scm_convert/gitside$ git repackPacking 28994 objects Jan 19 12:14:21 zecke: I'd presume git can handle unpacking the needed objects... Jan 19 12:14:47 zecke: I repacked my copy of that tree and it worked ok :) Jan 19 12:15:11 RP: well I ask myself what happens if you push, who will pack it on the server? Jan 19 12:15:23 Pack pack-cdbc63e435173f50ef0cac8ee5f0a587b6e86e75 created. Jan 19 12:15:27 zecke: nightly cron job? Jan 19 12:15:38 zecke: It doesn't have to be packed. Its just more efficient Jan 19 12:15:47 Instead of 30,000 files in this case ;-) Jan 19 12:16:54 RP: I like 30,000 files ;) Jan 19 12:17:05 RP: it is packed now Jan 19 12:17:20 RP: going to apply your bitbake patch soon Jan 19 12:17:46 need to find a good tv program first Jan 19 12:17:58 thanks :) Jan 19 12:18:27 mreimer: finished pushing Jan 19 12:18:38 thanks hrw Jan 19 12:20:39 gosh TV is full of crap Jan 19 12:21:52 in italy this evening start the anual edition of 'big brother' ... ;( Jan 19 12:22:04 s/anual/annual/ Jan 19 12:22:30 heh, oh dear Jan 19 12:22:38 big brother has been going on for a while over here Jan 19 12:23:08 i think is the 5th or 6th edition in italy ... ane every year more crap ;( ... Jan 19 12:24:09 yeah, I lost count of how many years it has been running for in the uk. probably 5 or 6 here as well. Jan 19 12:24:59 we have Big Brother The Village ;) Jan 19 12:25:04 the last big brother ever Jan 19 12:25:25 crumbs Jan 19 12:25:49 hihihi ... Jan 19 12:26:07 I want ... the big pengiun !!! :) *LOL* Jan 19 12:26:08 pb__: it runs for 8 month already, four houses... and not scheduled to end any time soon Jan 19 12:26:26 hrw: same problem Jan 19 12:26:55 hrw: I pulled from ewi Jan 19 12:26:55 Do we have a working gconf-dbus cvs date? or svn date? Jan 19 12:27:50 mreimer: both branches are merged here... Jan 19 12:28:05 mreimer: will try on other machine Jan 19 12:28:11 mreimer: Do you have local changes you need to revert? Jan 19 12:28:24 zecke: that does sound quite extreme Jan 19 12:28:52 RP: I do, but not in the files mt complains about when I mt merge Jan 19 12:41:47 RP: do you know if signing of commits is planned for git/cogito Jan 19 12:47:59 zecke: I don't think the kernel demands that so I doubt its high on the todo list Jan 19 12:49:37 hi i am trying to compile Tcl on my Cross development system for Openslug and i always getting this error when i am configure: Jan 19 12:49:41 RP: I have a gconf-dbus_svn20060118-r0_armv5te.ipk in my deploy/ipk. What is the problem? Jan 19 12:49:57 /home/kami/slug/openslug/tmp/work/tcl-8.4.11-r2/tcl8.4.11/unix/configure: line 7625: syntax error near unexpected token `)' Jan 19 12:49:58 /home/kami/slug/openslug/tmp/work/tcl-8.4.11-r2/tcl8.4.11/unix/configure: line 7625: ` SunOS-4*)' Jan 19 12:49:58 FATAL: oe_runconf failed Jan 19 12:50:03 pH5: PARALLEL_MAKE = "" is needed :) (I'll push) Jan 19 12:50:08 Hmm. libgpewidget 0.107 needs pkgconfig > 0.16... Jan 19 12:50:15 >= Jan 19 12:52:21 sorry this is the real error message : Jan 19 12:53:21 ~pastebin Jan 19 12:53:23 [pastebin] a place to paste your stuff without flooding the channel - try http://pastebin.com/, or http://pastebin.ca Jan 19 12:53:50 hi Jan 19 12:54:17 http://pastebin.com/513486 Jan 19 12:55:15 kami22: do you think that log is useful? Jan 19 12:57:55 RP: okay with packing things go really faster Jan 19 12:58:11 pb__: would git be acceptable to you? Jan 19 12:58:34 zecke: what do you mean? this is the error Jan 19 12:59:01 kami22: what should I say besides look at line 7624? Jan 19 12:59:03 zecke: worth a try, at least Jan 19 13:00:00 pb__: from the conversion git peforms reasonable well Jan 19 13:00:07 right Jan 19 13:00:10 pb__: now with packing the initial clone will be fast Jan 19 13:00:34 pb__: 40mb network IO, small object header Jan 19 13:00:41 pb__: ci seems fast Jan 19 13:00:50 yah. it still suffers from being a bit new and untested, but that isn't necessarily a fatal problem Jan 19 13:01:13 zecke: i am sorry this is the lines in configure script : Jan 19 13:01:17 http://pastebin.com/513503 Jan 19 13:01:32 pb__: hehe if the Linus Kernel would be as mature as BSD, the usage of git would mean something ;) Jan 19 13:01:44 kami22: well does the line look suspicious to you? Jan 19 13:01:53 kami22: was autoreconf run? Jan 19 13:01:56 i dont know :) Jan 19 13:02:05 kami22: did you solve your php problem? did you send a patch? Jan 19 13:02:18 zecke: hehe, yeah, indeed Jan 19 13:02:30 zecke: I thought it might speed it up a bit :) Jan 19 13:02:43 pb__: on the other hand, the svk cloning didn't take too long as well (longer than git though) Jan 19 13:02:46 i solved it i compile the version 4.3.10 from the openembedded packages and this worked Jan 19 13:02:50 zecke: I'll try and get a server working.... Jan 19 13:02:52 zecke: also, git's name still makes me wince from time to time, but I guess I will get over that. Jan 19 13:03:05 but i think i will try in the next days to compile a php5 version Jan 19 13:03:10 pb__: Use cognito :) Jan 19 13:03:43 pb__: on reasoanble filesystems do ln -s /usr/bin/git /usr/bin/bk Jan 19 13:03:58 this is the .bb file Jan 19 13:03:59 zecke: heh, good idea Jan 19 13:04:02 http://pastebin.com/513510 Jan 19 13:04:24 i dont know if autoreconf was started or better where it must be started Jan 19 13:04:40 kami22: did you create that file? Jan 19 13:04:50 RP: you mean cogito, right? Jan 19 13:04:59 no its from the cross development Jan 19 13:05:02 pb__: thanks, I did not want to ask that Jan 19 13:05:03 pb: names can be important though. Jan 19 13:05:27 I considered buying a PocketPC, but the name 'WinCE' made me wince then, and I never got over that Jan 19 13:06:48 RP: could you point me to your preferred provider patch? Jan 19 13:07:28 RP: r2 is the latest patch? Jan 19 13:08:16 RP: can one kill discriminated? Jan 19 13:09:43 ah it is killed Jan 19 13:13:55 zecke: can you help me? Jan 19 13:14:16 luke-jr_: you remember our previous conversation about a japanese input method? 覚えていますか? Jan 19 13:14:40 zecke: while you're at it, could you consider the stamp patch at http://article.gmane.org/gmane.comp.handhelds.openembedded/8529? Jan 19 13:18:02 okay Jan 19 13:18:19 kami22: I can only encourage you to use your mind ;) Jan 19 13:18:42 okay my mind says no :) Jan 19 13:19:22 kami22: that file looks strange ;) Jan 19 13:19:31 kami22: it should not have a custom do_configure method Jan 19 13:20:22 so you mean i have to remove the do_configure command? Jan 19 13:20:34 kami22: you could try it Jan 19 13:20:41 zecke: thanks Jan 19 13:20:41 Laibsch: did you get im-ja to work? Jan 19 13:23:11 pH5: Unfortunately, no. I also have only very limited time. But I guess the real limiting factor is my knowledge about compilation systems. But it's good to see others noticed the efforts. Do you maybe want to help? Jan 19 13:23:44 Most of the information is publicly available and I'll gladly share the sources of this information with you. Jan 19 13:23:55 zecke this produces a much more erros Jan 19 13:24:05 'i am also not able to push strg + c with my collie in gpe (rxvt)' - someone has idea what reporter wanted to say? that 'strg' is unknown for me Jan 19 13:24:09 kami22: fix them one by one ;) Jan 19 13:25:06 i think i will compile it natively on the slug Jan 19 13:25:14 pH5: To quote myself from this channel from Jan 12th: Jan 19 13:25:14 but thanks a lot cu :) Jan 19 13:25:17 (21:19:13) Laibsch: Luke-Jr: Maybe you want to have a look at http://pfaffe3.dyndns.org/openembedded/org.openembedded.dev/packages/nunome/nunome_1.0.bb and see if we can get it to compile? Jan 19 13:25:17 (21:19:13) Laibsch: http://www.google.de/search?hs=Q3m&hl=en&client=firefox&rls=org.mozilla%3Aen-US%3Aunofficial&q=+site%3Apurl.rikers.org+nunome+laibsch+mickeyl&btnG=Search lists some of the previous discussions about problems I had. Jan 19 13:25:54 whois pH5 Jan 19 13:26:16 pH5: Aha, ein Deutscher ;-) Jan 19 13:31:50 kami22: this happens when people try to use autotools Jan 19 13:32:20 hrw|tv: strg = ctrl Jan 19 13:32:21 Laibsch: I'm definitely interested in this, but I'm not really up to date. Does nunome depend on qtopia? Jan 19 13:32:23 nick koen Jan 19 13:32:26 nick koen Jan 19 13:33:11 pH5: Cool. Jan 19 13:33:28 I don't think nunome depends on Qtopia. Jan 19 13:33:33 pb__: I did mean cogito, yes :) Jan 19 13:33:34 any idea why we do not supply numeric pin for bluez? its hard/impossible to enter 'BlueZ' via nokia keypad... Jan 19 13:33:51 zecke: -r2 is the latest and the discriminated variable should be removed Jan 19 13:33:53 How about you download the .bb files and try to compile them? Jan 19 13:34:09 koen: thx Jan 19 13:34:42 RP: patches are applied already, ph5's one is pending Jan 19 13:35:10 hrw|tv: doesn't opie have an interactive pin requesting thing? Jan 19 13:35:12 pH5: http://pfaffe3.dyndns.org/openembedded/org.openembedded.dev/packages/nunome/nunome_1.0.bb Jan 19 13:35:12 pH5: http://pfaffe3.dyndns.org/openembedded/org.openembedded.dev/packages/anthy/anthy_6700b.bb Jan 19 13:35:46 pb__: I use gpe now on both Z Jan 19 13:35:53 oh, right Jan 19 13:35:56 oh dear, anthy is a bitch to crosscompile. Jan 19 13:36:05 gpe definitely does have an interactive pin requester. is that not working for you? Jan 19 13:36:08 pH5: That is where I currently stand. There are still things that need to get fixed, for sure. If you do any work, let me know. Jan 19 13:36:16 pH5: ACK for anthy. Jan 19 13:36:28 pb__: my phone also ask for pin... I had to enter pin on Z and on phone Jan 19 13:36:28 the "BlueZ" string shouldn't be getting used for anything, afaik Jan 19 13:36:39 hrw|tv: I need to use a german winxp laptop for work :( Jan 19 13:37:00 pH5: If you want to we can try to get nunome to work first. I'll be out skiing from tomorrow. But I'll get back to it after that. Jan 19 13:37:02 koen: what's worse, german or winxp? ;) Jan 19 13:37:03 hrw|tv: right, it ought to prompt you to enter a pin at both ends. Jan 19 13:37:10 pH5: xp :) Jan 19 13:37:26 koen: Uff :-) Jan 19 13:37:27 ~lart qwertz for being not qwerty Jan 19 13:37:28 * ibot executes killall -KILL qwertz for being not qwerty Jan 19 13:43:07 zecke: thanks :) Jan 19 13:43:26 zecke: I can push for some more cleanup in OE now :) Jan 19 13:51:04 hrw|tv: thanks for the bluetooth update, could you put it to the update feed? Jan 19 13:53:07 Laibsch: I can't use nunome. It's qtopia based, which doesn't help much on GPE. I do have an old .bb recipe of anthy-6801, though. Jan 19 13:53:39 law_: ok Jan 19 13:54:16 law_: done Jan 19 13:54:32 pH5: you hae experience with cross-compiling anthy? Jan 19 13:54:42 hrw|tv: thanks Jan 19 13:54:49 Laibsch: ugly memories, trust me. Jan 19 13:54:50 hi marcan Jan 19 13:54:51 pH5: are you in Uni tomorrow? Jan 19 13:55:32 pH5: Well, I have those, too. Question is if you ever got it to compile. Jan 19 14:00:08 zecke: currently I'm doing my diploma work at the HMI in wannsee. tomorrow I'll be in adlershof. Jan 19 14:00:28 Laibsch: I just tried, and it seems I got it to build. Jan 19 14:01:05 ah okay Jan 19 14:04:45 hrw: i think alarms dont work on collie (gpe) Jan 19 14:05:20 Are there any rules for adding new packages to oe? I'd like to upload anthy and ttf-sazanami .bb files. Jan 19 14:05:38 law_: connect headphones Jan 19 14:05:52 hrw: already done Jan 19 14:06:03 law_: gpe use /dev/dsp which means headphones on collie Jan 19 14:06:04 mom i?ll test it again Jan 19 14:06:15 sound works on collie Jan 19 14:09:48 Why does gxine depend on gxine-locale* Jan 19 14:11:59 hrw: can find any bluetooth stuff in /etc/apm/scripts.d Jan 19 14:12:04 law_: on working collie I got alarm from contacts on headphones Jan 19 14:12:12 law_: /etc/apm/event.d/ Jan 19 14:12:24 DataBeaver: so that the translations are installed Jan 19 14:12:38 most binary packages depend on the corresponding translations. Jan 19 14:12:40 hrw: ah :-) why event? Jan 19 14:13:03 hrw: alarm with gpe? Jan 19 14:13:08 Well, what if I don't have any translations? Jan 19 14:13:09 law_: looked at opie-alarm which do the same. and apmd_proxy run event.d/ then script.d/ iirc Jan 19 14:13:20 DataBeaver: then none will be installed. Jan 19 14:13:36 law_: yes - alarm in gpe - but it was on runniing collie - now its snoozed to few minutes and collie suspended Jan 19 14:13:38 But I can't remove gxine-locale-de without removing gxine? Jan 19 14:14:17 indeed not (at least, not without --force-depends). that's kind of a bug. Jan 19 14:15:20 law_: now collie resumed and is beeping with gpe-contacts event Jan 19 14:16:09 hrw: gpe-calendar or? i tried the alarm with gpe-watch Jan 19 14:16:26 law_: yep - gpe-calendar sorry Jan 19 14:18:06 pb__: shouldn't the locale-xx packages be in rrecommends instead of rdepends, then? Jan 19 14:18:21 pH5: theoretically, yes, but ipkg doesn't support that Jan 19 14:18:43 so, the choice is between not being able to conveniently remove them, or having them not work at all Jan 19 14:18:55 pb__: http://ewi546.ewi.utwente.nl/OE/other_scms/git/org.openembedded.oz354fam083/ Jan 19 14:19:28 mercurial seems ueberfast Jan 19 14:19:41 pb__: another possibility would be to remove the dependencies altogether and to enhance the rootfs builder to pull existing $P-locale-... packages for every package P that has them. Jan 19 14:19:56 pH5: that'd lose for software that you install after the fact. Jan 19 14:20:15 and, also, for locales that you install after the fact. Jan 19 14:20:46 I think the current arrangement is pretty much optimal given the state of the art in ipkg. Jan 19 14:20:58 zecke: awesome Jan 19 14:21:28 ah, I see mickeyl has regained access to his email at least Jan 19 14:21:37 hrw: there are event in /var/spool/at but the dont launch, perhaps this is a timezone problem, i?ll have a look Jan 19 14:21:57 law_: dont know - I do not use gpe usually Jan 19 14:23:11 pb__: you're right. there's now way this can work correctly without ipkg understanding locales. Jan 19 14:23:37 pH5: it doesn't need to understand locales per se, it just needs to support greedy Recommends. Jan 19 14:24:08 then, like you say, the locale stuff could be switched to use RRECOMMENDS not RDEPENDS, and everything would be peachy. Jan 19 14:24:21 hrw: ok i think the problem is, that system time is 1 hour behind gpe-time Jan 19 14:24:35 pb__: but then still the user has no possibility to select the locales he wants during package installation, but he has to remove the unwanted ones afterwards. Jan 19 14:25:20 pH5: is there a real-world use case for that? most users just want the same set of locales for all packages. Jan 19 14:25:26 Nice... hostap-utils depends on kernel-modules, which depends on all of the kernel-module-* packages. Most of which I don't need. Jan 19 14:25:41 pH5: it is hard to imagine that tweaking the set of locales on a per-package basis is really a common requirement. Jan 19 14:25:50 pb__: hmm I think I need a mercurial server as well :} Jan 19 14:26:35 Or at least ipkg claims that dependency to exist when trying to remove kernel-modules. Jan 19 14:26:48 pb__: I meant on a per system basis. What if I only need -fr locales, but the package recommends -de -en and -ru, too? Jan 19 14:27:52 pH5: no packages should be recommending -de, -en and -ru. if they are, they are just broken. Jan 19 14:28:13 pb__: what should they be recommending, then? Jan 19 14:28:15 All the packages should do is depend on -locale*, and everything else will "just work". Jan 19 14:28:30 ouch Jan 19 14:28:34 or, in this future utopia of an improved ipkg, Recommend: -locale* Jan 19 14:29:07 I think I'm not really understanding how this really works. Jan 19 14:30:12 pH5: the use case goes like this: Jan 19 14:30:28 ipkg install gaim Jan 19 14:30:34 - user installs locale-base-LL-CC packages for all the locales he wants on his system (e.g. locale-base-fr-fr) Jan 19 14:30:50 - those locale-base packages each Provide: virtual-locale-LL-CC, just for general reasons Jan 19 14:31:25 - at some later date, the user installs gpe-calendar, which Depends: on gpe-calendar-locale* Jan 19 14:31:41 - gpe-calendar-locale-fr is the only installable candidate for gpe-calendar-locale*, so ipkg installs it. Jan 19 14:32:03 - user lives happily ever after Jan 19 14:32:19 actually, the second part of the user case goes like this: Jan 19 14:32:32 - at some even later date, the user learns German and installs locale-base-de-de. Jan 19 14:32:47 - now, gpe-calendar-locale-de is installable, so ipkg installs that one too. Jan 19 14:33:03 - user, presumably, continues to live happily ever after Jan 19 14:33:07 Yeah, thanks for explaining. Now it makes sense :) Jan 19 14:33:51 the only magic bit is the "foo*" dependency notation, which means "all the packages providing whose dependencies are already satisfied". Jan 19 14:35:55 hrw: are you using gpe as root? Jan 19 14:36:04 law_: as normal user Jan 19 14:36:14 hmm Jan 19 14:36:33 1386 root 56 S /usr/sbin/atd /var/spool/at Jan 19 14:36:39 is this normal? Jan 19 14:36:51 or should this run as normal user Jan 19 14:37:08 I think running as root is correct. Jan 19 14:37:12 atd is a global service, not per user Jan 19 14:38:11 hmm i can call scripts from /var/spool/at (as normal user) but they dont start automatic Jan 19 14:48:45 re greentux Jan 19 14:49:03 re hrw Jan 19 14:59:12 hi [lala Jan 19 15:03:08 <[lala]> yo Jan 19 15:04:52 lala bugreport hciattch? where should i do that? bluez or kernel? Jan 19 15:05:40 <[lala]> i think kernel since it is in mainline Jan 19 15:13:58 greentux: As its a kernel oops, its a problem in the kernel. Its something mainline developers should probably be made aware of as well though (not jsut a Zaurus problem) Jan 19 15:14:42 RP: yes i know, but i thought that bluez developer are the right place :) Jan 19 15:23:29 <[lala]> greentux: btw the boostrap image worked nearly perfect now Jan 19 15:23:50 [lala] fine... 2.6.15 final... Jan 19 15:23:54 <[lala]> building a new opie image over night Jan 19 15:24:12 <[lala]> yes Jan 19 15:27:18 [lala] over night :) how long does it take? Jan 19 15:27:48 someone here use 2.4.20 on cxx00? Jan 19 15:27:57 pH5: Wow, crazy and super cool news. Got to run to catch the tram. Will talk to you later. Jan 19 15:28:21 Did you need to make any changes? Jan 19 15:28:44 <[lala]> greentux: it depends ... opie components are build from cvs ... so every day a fresh checkout. Jan 19 15:30:17 hrw: what is the advantage of 2.4 ob clamshell? Jan 19 15:30:22 Laibsch: i have used some patches from debian and I also needed to make a -native package. I'll upload it to oe. Jan 19 15:30:34 [lala_]: use fixed SRCDATE or even use opie 1.2.1 - most of changes are backported to 1.2.1/OE Jan 19 15:30:42 Cool. Jan 19 15:30:46 Gotta run. Jan 19 15:30:49 BBL. Jan 19 15:31:02 greentux: I need to know does cxx00 has apm problems with 2.4.20 or not Jan 19 15:31:52 hrw: ok. i see tha problem that nearly nobody will test 2.4... we can check some features for you. but not very long... Jan 19 15:32:17 <[lala]> hrw: yes, i already thought about that ... can i set the srcdate for all packages at once or do i have to list all packages and the version in the conf file? Jan 19 15:32:47 [lala_]: SRCDATE=YYYYMMDD for all or SRCDATE_package=YYYYMMDD for one package Jan 19 15:33:47 greentux: we have 3.5.4 release with 2.4.20 kernel - if I can look into apm problem then I need tester brave enough ;) Jan 19 15:34:09 hm.. does anyone know the stat of .dev? Is it stable enough to build packages? Jan 19 15:34:22 *state Jan 19 15:34:39 NAbyss: build and if something is wrong then report Jan 19 15:34:52 <[lala]> hrw: with the first method i would not only catch opie packages i suppose Jan 19 15:35:28 hrw: I've got issues with libmimetool-native.. won't build. Jan 19 15:35:49 [lala_]: OPIE_VERSION="1.2.1" and include conf/distro/preferred-opie-version.inc will do opie stuff for you Jan 19 15:35:55 0http://pastebin.com/513769 Jan 19 15:35:56 hrw: i understand that, but thats the problem with two kernels for same device (both are stable) Jan 19 15:36:16 <[lala]> hrw: that sounds better :) Jan 19 15:36:21 er, libtool-native Jan 19 15:36:23 greentux: stable like stable - but are they tested? Jan 19 15:36:39 It's too early in the morning. Jan 19 15:36:50 NAbyss: 23:37 here ;) Jan 19 15:37:02 hrw: @fosdem i will get back the borzoi from lrg... perhaps you can use it . Jan 19 15:37:10 hrw: 0937 here.. I usually get into work at 1030 :) Jan 19 15:38:01 greentux: first I have to send 1G sd card to rmk Jan 19 15:38:28 have to write some rules about card return, print it, add to envelope so he will fax it to me Jan 19 15:38:41 hrw: ah ok... i have the kingston 1 GB card on te todo list Jan 19 15:39:52 hrw: you dont need that burocracy for that card. forget that you got it from me... Jan 19 15:40:01 greentux: I hope that even if rmk does not get that card working then he will atleast find something in sd/mmc subsystem to update Jan 19 15:40:19 greentux: ;) Jan 19 15:40:47 hrw: the mmc/sd subsystem.. hope the sdio board will open the protocols for sdio ... Jan 19 15:41:15 greentux: Wishful thinking... i don't think they'll want to give up their royalites. Jan 19 15:41:25 *royalties Jan 19 15:41:35 greentux: personally I do not care about sdio - most of sdio cards are too expensive and have cheaper cf replacements Jan 19 15:41:55 hrw: but there are device w/o cf Jan 19 15:42:12 greentux: yep. Jan 19 15:45:26 ok. cu tom. Jan 19 15:57:16 good night Jan 19 16:17:46 * france is back (gone 77:52:28) Jan 19 16:17:54 zecke: you rang? Jan 19 16:18:15 france: not really Jan 19 16:18:37 zecke: ok, the blue light light up when my name is mentioned. Jan 19 16:18:45 zecke: I have been a way for a few days Jan 19 16:18:48 france: I wanted to check if I can have a directory on h7 to do an initial svn test Jan 19 16:19:02 zecke: sure, or on cvs2.handhelds.org? Jan 19 16:19:14 france: whatever you think is best suited Jan 19 16:19:34 zecke: the load average is getting a bit on h7 Jan 19 16:19:40 s/bit/bit high/ Jan 19 16:19:41 france meant: zecke: the load average is getting a bit high on h7 Jan 19 16:20:13 france: also it would be nice if you could tell me what an admin interface would need in regards to svn Jan 19 16:20:52 zecke: everything is going into a mysql database, as there is now a mysql module for everything Jan 19 16:21:08 pam, postfix, smtp auth, etc.... Jan 19 16:21:23 sorry I need to go to bed now Jan 19 16:21:47 zecke: no worries, have a good night. I will be on at 4am EST, for an hour or so, then back on Saturday. Jan 19 16:22:07 zecke: I will send you an e-mail... Jan 19 16:23:22 thanks Jan 19 16:24:24 shit. looks like ewi died again ;( Jan 19 16:25:04 hrw: I'm still logged in Jan 19 16:25:19 Wtf does links want libpng for? Jan 19 16:25:25 hrw: koen is using monotone + cc1plus Jan 19 16:25:30 I'm logged too but cannot event touch it.. ah. Jan 19 16:25:36 DataBeaver: for graphics Jan 19 16:25:51 I thought links was a textmode browseR? Jan 19 16:25:53 hrw: and I'm running monotone+monotone+git+hg Jan 19 16:26:09 Anyway, it's failing and claims that "Your libpng is too old" Jan 19 16:26:13 good nite Jan 19 16:26:21 zecke: night night Jan 19 16:26:42 hehe Jan 19 16:27:21 hi dirk Jan 19 16:36:55 since ph5 is gone now, can anybody tell me if he compiled anthy or nunome? He said he would upload it oe. any activity in this respect? Where could I check myself? Jan 19 16:37:42 Anything wrong with the monotone servers. The pull does not start for ages for vanille and koen's server. Jan 19 16:37:57 overloaded Jan 19 16:42:17 Stacked Git sounds interesting... Jan 19 16:45:44 hrw: how come? Jan 19 16:45:52 Any major activity? Jan 19 17:03:46 what do people think of this -> https://secure.mithis.com/~tim/tp-poster-small.png ? Jan 19 17:04:41 hey guys, does OZ have GNU octave in it? Jan 19 17:17:50 Hhhhh, is it in the packages directory? Jan 19 17:18:04 I dunno Jan 19 17:18:14 did you look? Jan 19 17:18:26 no Jan 19 17:18:32 where can I see what's in OZ? Jan 19 17:18:44 try the oz website or channel Jan 19 17:19:02 oe is for the build system only Jan 19 17:19:06 k Jan 19 17:19:16 or wait for the sequel's.. sequel Jan 19 17:19:21 wizard of oz 2 was a joke Jan 19 17:19:28 most people don't even know it was made Jan 19 17:19:29 lol Jan 19 17:19:48 sure we do Jan 19 17:19:54 she went to a mental institution Jan 19 17:21:14 you didn't watch it did you? =-p Jan 19 17:21:35 Meh: | configure: error: source directory already configured; run "make distclean" there first Jan 19 17:21:42 (libtool-native-1.5.10-r3) Jan 19 17:22:25 a long time ago once on disney i think Jan 19 17:22:49 not much to watch when you live somewhere that only has one channel Jan 19 17:25:41 anyone had a chance to look at bug # 399 yet? Jan 19 17:31:17 * Necronom notes that the oe monotone database can't be migrated to rosters. Jan 19 18:07:20 Ahh. Bug 591. Jan 19 18:10:11 Yep... it's a prbolem I'm having. Jan 19 18:11:21 Err. Is there another mtn server besides monotone.vanille.de? It seems to be MIA. Jan 19 18:14:11 monotone.nslu2-linux.org has the org.openembedded.dev branch on it Jan 19 18:14:31 Danke. Jan 19 18:14:48 ewi564.ewi.utwente.nl is another server Jan 19 18:17:30 at least one of the above is usually up :-) Jan 19 18:41:16 does anyone else have trouble compiling mono (with bitbake? Jan 19 20:24:46 ALSA causes a kernel oops in 2.6.15 on Akita again... Jan 19 20:31:56 * france is away: Away Jan 20 00:15:12 Err, Links wants some unreleased version of libpng? 1.2.8 isn't new enough for it, even though it's the most recent release. Jan 20 00:20:28 It would seem that the configure script of Links is flawed... It's not finding a certain function in libpng, even though it is there. Jan 20 00:21:43 Okay, it's actually finding the function, but it isn't finding libsdl. Jan 20 00:21:49 Because I haven't built that. Jan 20 00:23:35 "Your libpng is too old" is quite misleading in this case Jan 20 00:24:32 | /* Define png_get_sRGB to an innocuous variant, in case declares png_get_sRGB. Jan 20 00:24:32 | For example, HP-UX 11i declares gettimeofday. */ Jan 20 00:24:34 wtf? Jan 20 00:33:23 Yeaha, it compiled after I build libsdl-x11 first. Jan 20 00:33:40 interesting Jan 20 00:33:56 I just built links and don't have libsdl-x11 built yet... Jan 20 00:35:03 Find png_get_sRGB from config.log and see what command it's using to build the test program? Jan 20 00:39:34 in config.log their are two references to png_get_sRGB: "configure:17044: checking for png_get_sRGB Jan 20 00:39:34 " and "ac_cv_func_png_get_sRGB" Jan 20 00:41:24 you mean the line below "checking for png_get_sRGB" that starts with ccache arm-linux-gccc...? Jan 20 00:41:43 Yeah Jan 20 00:41:47 Is there -lSDL on it? Jan 20 00:42:19 http://pastebin.com/514278 Jan 20 00:42:24 nope Jan 20 00:42:59 Hm, strange. Jan 20 00:44:20 Is there "checking for sdl-config" in your config.log? Jan 20 00:45:24 it checked and failed Jan 20 00:45:43 but kept going... Jan 20 00:46:22 Hm. Jan 20 00:46:39 Do you have libsdl installed on the host system? Jan 20 00:47:09 nope Jan 20 00:47:17 yeah, that would make sense then Jan 20 00:47:21 That might explain it. Jan 20 00:47:42 I have libsdl installed, so it's probably finding sdl-config in /usr/bin Jan 20 01:15:21 jbowler: Those bitbake changes are in svn now Jan 20 01:23:38 RP: You helped me with alsa on 2.6.15-rc7 on an Akita the other night...I just updated to 2.6.15 and the Oops came back Jan 20 01:23:57 I applied the same change you told me the first time and all is good again... Jan 20 01:25:28 RP: thx Jan 20 01:28:46 hi Jan 20 01:50:05 morning Jan 20 01:54:47 pb__: the branch is gited completely Jan 20 01:55:13 pb__: only issue it contains only the objects that were explictely comitted to this branch Jan 20 02:00:33 ah, so not the stuff that existed in the trunk at the point of branching? Jan 20 02:09:54 pb__: no, you could pull that from the org.oe.dev branch :} Jan 20 02:10:06 pb__: and see how it woes Jan 20 02:10:11 (actually it should work) Jan 20 02:10:46 pb__: I like git and darcs because it does only need a simple webserver (that servers a directory) Jan 20 02:21:26 re Jan 20 02:23:29 zecke: right. sounds excellent. Jan 20 02:23:32 * pb__ go to the office now Jan 20 02:23:32 bbiab Jan 20 02:29:04 johnX: I still need to apply that fix to the kernel in OE, sorry about that. I was sort of expecting an ALSA update before now Jan 20 02:29:16 ah Jan 20 02:29:23 just mentioning it in case you forgot or somethin **** ENDING LOGGING AT Fri Jan 20 02:59:58 2006