**** BEGIN LOGGING AT Fri Dec 30 03:00:01 2005 Dec 30 03:00:08 hi all Dec 30 03:00:33 hey alan|laptop Dec 30 03:01:17 hey hey lrg|home Dec 30 03:01:35 hey alan|laptop Dec 30 03:02:05 yop koen Dec 30 03:03:47 morning Dec 30 03:03:54 hey hrw|work Dec 30 03:04:32 morning hrw|work Dec 30 03:07:18 ~lart knowit builsystem Dec 30 03:07:18 * ibot installs a bad bootloader on knowit builsystem and turns knowit builsystem into a brick Dec 30 03:28:12 koen: can I use http://ewi/tmp/hrw/ for own OZ related images etc? Dec 30 03:28:30 koen: or other dir on ewi Dec 30 03:29:06 hrw|work: sure Dec 30 03:30:48 thx Dec 30 03:45:58 * france is away: Away Dec 30 04:02:19 hi there. who's responsible for kernel related things in OE? Dec 30 04:03:13 lots of people Dec 30 04:03:27 each kernel .bb should have a MAINTAINER entry Dec 30 04:04:16 koen: good hint. thank you Dec 30 04:27:30 morning all Dec 30 04:27:59 hey RP Dec 30 04:30:34 pezi: which machine kernel? Dec 30 04:39:47 hey RP Dec 30 04:55:26 hey Dec 30 04:55:31 mithro: ping Dec 30 04:55:36 mickey|working: ping Dec 30 04:57:20 zecke: where did you end up ? Dec 30 04:59:36 mithro: room three? where did you go? Dec 30 05:00:42 saal2 Dec 30 05:07:31 how full is your room? Dec 30 05:07:34 my room is really empty Dec 30 05:08:18 morning zecke Dec 30 05:09:45 hey pb_ Dec 30 05:09:49 you going to Fosdem? Dec 30 05:10:17 dunno, I haven't really planned that far ahead Dec 30 05:19:45 cyphunk: hi Dec 30 05:19:53 cyphunk: in which 'saal' are you currently? Dec 30 05:22:43 not Dec 30 05:22:47 im in the hack center Dec 30 05:23:06 why? Dec 30 05:26:57 hey cyphunk, i'm at 22C3 too, so is Zecke Dec 30 05:27:54 ~lart iqnotes author for one .pro for desktop and pda Dec 30 05:27:54 * ibot chops iqnotes author for one .pro in half with a free AOL CD for desktop and pda Dec 30 05:27:57 ahah... cool. i might be adding some junk to one of the oe projects and just have basic questions. it'd be easiest to just show you might current working tree and then have one of you tell me what should change Dec 30 05:29:01 cyphunk: zecke might be able to help Dec 30 05:29:20 mithro, thanks. Dec 30 05:29:56 i might be able to help, but zecke r0ck0z Dec 30 05:29:56 zecke, will you have time at any point today? it will probably take all of 15 minutes Dec 30 05:30:07 cyphunk: sure Dec 30 05:30:21 cyphunk: I'm currently in the Paper Prototyping lecture Dec 30 05:30:27 okay Dec 30 05:30:43 and I will move to the "Saal 1" afterwards and will stay there until the end of the congress ;) Dec 30 05:31:16 zecke, ahh, okay, so i wil move there as well and catch you therel Dec 30 05:31:30 morning Dec 30 05:32:01 cyphunk: oh well I do not find the 14-15h talks too interesting. I could come down to the hackcenter as well Dec 30 05:32:17 hello, do someone know what is the package name for bitbaking the kernel in org.openembedded.dev ? Dec 30 05:32:27 theturtle: bitbake virtual/kernel Dec 30 05:32:30 zecke, nah, im mobile... saal1 is fine Dec 30 05:32:31 theturtle: virtual/kernel? Dec 30 05:32:38 cyphunk: okay Dec 30 05:32:39 thanks guys =) Dec 30 05:32:59 koen: quilt is finally building on OS X (again) ;) Dec 30 05:34:58 zecke: cool Dec 30 05:35:26 koen: somehow parsing is dog slooow... I don't know why though Dec 30 05:35:44 * hrw|work will do something which will does not like to qt(/e) guys in iqnotes 2.1.0-rc1 recipe Dec 30 05:35:50 zecke: I know: python Dec 30 05:35:54 zecke: i think it's more closer to dead snail slow :P Dec 30 05:36:31 koen: If I'm lucky it is parsing a recipe within two seconds Dec 30 05:36:47 hmm Dec 30 05:37:00 iirc my slug is faster than that Dec 30 05:37:39 zecke: maybe it's falling back to regex code in pure python? Dec 30 05:38:17 reenoo_: probably... I will find that out later Dec 30 05:42:21 let us start writing a ppc backend for psyco Dec 30 05:51:52 zecke: i'll see you in Saal 1 soon? Dec 30 05:54:08 yes Dec 30 05:55:34 NOTE: package iqnotes-2.1.0rc1-r0: task do_build: completed Dec 30 05:55:55 hrw|work: good ;) Dec 30 05:56:29 crap Dec 30 05:56:55 the neighbourkiddies have discovered their firework stashes Dec 30 05:57:15 hehe Dec 30 05:57:29 ./libschedule/libschedule_0.15.bb:inherit autotools libtool Dec 30 05:57:36 what is wrong in this line? Dec 30 05:57:54 inherit libtool? Dec 30 05:57:58 right Dec 30 05:59:04 I'm moving upstairs now Dec 30 06:00:44 zecke: good? palmtop.bbclass or something other stinks.. I had to patch .pro file... Dec 30 06:05:46 ibot: test Dec 30 06:05:46 hrw|work: due stdc++? Dec 30 06:05:48 Test Failed! Dec 30 06:08:56 NOTE: package iqnotes-2.0.99-2.1.0rc1-r0: task do_build: completed Dec 30 06:09:01 now with better PV Dec 30 06:16:23 03koen 07org.oe.dev * r5bc7da91... 10/packages/cairo/cairo_1.0.2.bb: cairo: update to 1.0.2 Dec 30 06:16:27 03koen 07org.oe.dev * rb15a6354... 10/packages/libschedule/libschedule_0.15.bb: libschedule 0.15: clean up .bb a bit Dec 30 06:22:42 hrw|work: did you have a issue with stdc++? Dec 30 06:26:08 cyphunk: where now in "Saal 1" Dec 30 06:26:28 hmmm Dec 30 06:26:29 xrdb-0.0cvsnow-r2 Dec 30 06:27:35 hehe Dec 30 06:27:36 wow. that's specific Dec 30 06:27:55 you set CVSDATE, what did you expect? :P Dec 30 06:32:42 zecke: yeah, im in saal1 Dec 30 06:32:53 zecke: close to south entrance Dec 30 06:35:56 cyphunk: where is south? i'm near the back left mic and so is zecke Dec 30 06:38:06 mithro: yeah, im near the back right mic.... im sitting at the tables.... second table to the front of the very end with a small IBM x40 12"laptop ;) Dec 30 06:38:28 i've got a very small Panasonic laptop Dec 30 06:38:38 i think it's most probably smaller then your IBM :P Dec 30 06:38:50 brat Dec 30 06:39:03 are you at the tables or in the seats? Dec 30 06:39:22 i'm sitting on the stairs atm Dec 30 06:39:31 waiting for a seat near zecke to clear up Dec 30 06:39:55 ahh, okay... got it... so at the break I will find you or zecke Dec 30 06:40:16 yeah Dec 30 06:46:00 03koen 07org.oe.dev * r9cda4c5f... 10/conf/distro/familiar-unstable.conf: familiar-unstable: don't use cvsdate=now because it produces stuff like this: xrdb-0.0cvsnow-r2 Dec 30 06:51:43 zecke: issue is with one .pro file for both pda and desktop Dec 30 07:00:37 cyphunk: I will get some food now Dec 30 07:00:53 zecke: mithro going with you? Dec 30 07:01:08 cyphunk: haven't asked him yet ;) Dec 30 07:01:11 dunno yet :P Dec 30 07:01:15 I will either get a Boulette or Bratwurst ;) Dec 30 07:01:57 cyphunk: I could drop by before getting food, or afterwars Dec 30 07:02:03 afterwards even Dec 30 07:03:40 Hi, Dec 30 07:04:39 mmmmm.... bratwurst Dec 30 07:05:03 Does the most recent self-compiled version of OE support altboot? I read that it will likely be included in 3.5.4. Can I install the bleeding edge and use altboot? Dec 30 07:05:14 koen: you like bratwurst? Dec 30 07:06:04 Laibsch: http://ewi546.ewi.utwente.nl/tmp/hrw/oz-3.5.4-rc-collie/ - opie-image + kernel 64-0 which have altboot Dec 30 07:06:28 hrw|work: Cool! Dec 30 07:07:09 Can I use altboot to boot into Sharp ROM? I still need it for a few things. Dec 30 07:07:56 Laibsch: you probably can try but iirc noone get it working Dec 30 07:08:26 Laibsch: who doesn't like bratwurst? Dec 30 07:08:44 koen: Depends on the Wurst, I'd say ;-) Dec 30 07:09:14 koen: How do you guys say? "smakelijk eeten" or some such? Dec 30 07:09:32 one 'e' less :) Dec 30 07:09:36 ~bon appetit Dec 30 07:09:41 i guess bon appetit is smacznego. Guten Appetit. Eet Smakelijk. God Appetitt. Buon Appetito. Buen apetito Bom Apetite. buen apetito Dec 30 07:10:02 this bot knows too much ;-) Dec 30 07:10:22 ~too much Dec 30 07:10:32 noooooo Dec 30 07:10:39 Laibsch: that bot does not know too much ;) Dec 30 07:10:48 LOOL Dec 30 07:11:34 hrw|work: IOW, booting into other ROMs works fine but not into a Sharp ROM? Dec 30 07:11:55 I really need to get my ass back up and work on the Japanese input methods for OE. Dec 30 07:12:07 Laibsch: you know roms for collie other then sharprom based? Dec 30 07:12:09 ~rom Dec 30 07:12:10 rumour has it, rom is Read Only Memory Dec 30 07:12:43 OZ ;-) Dec 30 07:12:53 You could boot into a different version. Dec 30 07:12:58 OZ isn't a ROM Dec 30 07:13:24 OK, OK Dec 30 07:13:40 Don't mean to argue ;-) Dec 30 07:17:40 03koen 07org.oe.dev * rdeb38864... 10/ (2 files in 2 dirs): h2200: machine and bootloader fixes Dec 30 07:57:00 cu all - HNY Dec 30 07:57:26 cu hrw|work Dec 30 09:17:35 What is the rootfs file in http://ewi546.ewi.utwente.nl/tmp/hrw/oz-3.5.4-rc-collie/ for? Dec 30 09:17:44 Do I need to flash to OZ? Dec 30 09:39:24 heh Dec 30 09:39:26 alsa-utils: unsatisfied recommendation for ${ALSA_PACKAGES} Dec 30 09:39:50 funny to see ipkg use OE variables Dec 30 09:44:56 man autoconf hell... Dec 30 09:45:08 anyone mind if I replace pkgconfig with the C++ rewrite? Dec 30 09:45:37 on mac glib1.2.8 doesn't know what value could be used as a gint32 Dec 30 09:54:35 Good evening! Dec 30 09:54:40 hey uv1 Dec 30 09:54:42 kergoth: _m4_divert_diversion <- fun? right? Dec 30 09:55:38 I just builded a fresh opie-image for CL860 from scratch (fresh monotone database (.dev);rm -rf tmp;bitbake opie-image) ... Dec 30 09:55:51 An got an error building the image ... Dec 30 09:56:03 that's opie ;) Dec 30 09:56:13 ERROR: Cannot satisfy the following dependencies for task-bootstrap: Dec 30 09:56:13 | + '[' -f /data/scdev2/build-boxer-mono/tmp/rootfs/usr/lib/ipkg/i Dec 30 09:56:48 koen: OK opie ;-) Dec 30 09:58:00 hmmm Dec 30 09:58:09 alsamixer isn't linked to libgcc Dec 30 09:59:03 hi, while installing alsa-utils (on an nslu2/openslug) I got a recommends for $(ALSA_PACKAGES) Dec 30 09:59:18 and indeed alsa-utils_1.0.10.bb contains: Dec 30 09:59:33 RRECOMMENDS_alsa-utils += "${ALSA_PACKAGES}" Dec 30 09:59:50 17:39 < koen> heh Dec 30 09:59:50 17:39 < koen> alsa-utils: unsatisfied recommendation for ${ALSA_PACKAGES} Dec 30 09:59:51 17:40 < koen> funny to see ipkg use OE variables Dec 30 09:59:52 but there is no ALSA_PACKAGES macro in the bb file, only PACKAGES Dec 30 10:00:07 * koen just saw that error as well Dec 30 10:00:19 lol, what a coincidence Dec 30 10:01:02 koen, i'm still trying to get my sound card to run, it used to in the past, decided to reinstall all Dec 30 10:01:54 ~weather eheh Dec 30 10:01:58 Eindhoven, Netherlands; (EHEH) 51-27N 005-25E 28M; last updated: 2005.12.30 1643 UTC; Dew Point: 26 F (-3 C); Pressure (altimeter): 29.62 in. Hg (1003 hPa); Relative Humidity: 92%; Sky conditions: obscured; Temperature: 28 F (-2 C); Visibility: less than 1 mile; Weather: Low drifting snow; Wind: from the SSE (160 degrees) at 20 MPH (17 KT) gusting to 31 MPH (27 ... Dec 30 10:02:14 started to snow here ... Dec 30 10:02:31 * koen looks outside Dec 30 10:02:32 excellent name for a weather station Dec 30 10:02:36 still dry here Dec 30 10:03:52 * eFfeM looks NNE, but can't see koen any more due to the snow .... Dec 30 10:04:44 need a reboot, back in a few mins Dec 30 10:04:45 zecke: glib 1.2.8? sounds like you are stuck in some kind of timewarp Dec 30 10:16:58 pb_: no pkgconfig contains a fork of glib-1.2.8 ;) Dec 30 10:17:14 crumbs Dec 30 10:17:51 I think I will patch out glib-1.2.8 and use glib-1.2.10-native Dec 30 10:18:14 yeah, good idea Dec 30 10:19:08 pb_: the best part about it, I can use the fink patch :} Dec 30 10:19:54 heh Dec 30 10:19:55 ideal Dec 30 10:22:20 pb_: what does acpaths = "-I ." is a correct syntax for the autotools class? Dec 30 10:22:48 zecke: sorry, I cannot parse that sentence Dec 30 10:25:59 pb_: sorry I'm getting tired Dec 30 10:26:07 pb_: let us reorder it then Dec 30 10:27:35 pb_: I figure acpaths = "-I ." will add the current directory to some search path? Dec 30 10:27:49 is the 'space' between -I and '.' correct? Dec 30 10:28:05 yeah, I think so Dec 30 10:28:13 iirc, the space is optional and useless, but harmless Dec 30 10:29:06 lol Dec 30 10:29:16 glib native doesn't know gint32 as well Dec 30 10:29:19 doh Dec 30 10:32:27 hey zecke Dec 30 10:32:58 mithro: hey Dec 30 10:33:08 | glib.h:502: error: parse error before "gssize" Dec 30 10:33:12 fun fun fun Dec 30 10:35:37 mithro: RIM is awesome Dec 30 10:37:30 zecke: i'm going Saal 1 now Dec 30 10:37:32 where are you? Dec 30 10:44:41 hey zecke where are you? Dec 30 10:46:47 pb_: Do you have any experience with the udc in the pxa27x? Dec 30 10:49:35 mithro: where we where before Dec 30 10:54:45 RP: nope, sorry Dec 30 11:04:06 What package do I need to access my SD card? Is SD support currently broken? Dec 30 11:11:42 that sounds like an issue for your distribution support channel. Dec 30 11:20:29 http://pastebin.com/484382 Dec 30 11:20:29 fun Dec 30 11:22:48 heh Dec 30 11:24:09 pb_: can I make autoconf emit the conftest.c code somewhere? Dec 30 11:24:15 ah well it is embedded in the configure Dec 30 11:24:17 *tired* Dec 30 11:26:41 indeed it is Dec 30 11:28:03 static const char x[256]; Dec 30 11:28:04 ; Dec 30 11:28:04 return 0; Dec 30 11:28:04 } Dec 30 11:28:10 looks like something is missing here Dec 30 11:28:34 Hi again ... Dec 30 11:28:48 uv1: use pastebin for the complete log Dec 30 11:28:51 Still investigating toe opie-image build problem ... Dec 30 11:29:04 Configuring wireless-tools Dec 30 11:29:05 | Collected errors: Dec 30 11:29:05 | ERROR: Cannot satisfy the following dependencies for task-bootstrap: Dec 30 11:29:05 | virtual/pcmcia-manager Dec 30 11:29:32 Hmm.. anyone around who can help with a glibc build problem? Dec 30 11:29:46 How is the dependency arrange to virtual/pcmcia-manager ? Dec 30 11:30:48 zecke: that certainly is a minimalist testcase Dec 30 11:31:13 pb_: sorry too tired... Dec 30 11:32:06 uv1: pull+update Dec 30 11:34:55 reenoo_: Done before, but no change. Trying again ... Dec 30 11:35:45 reenoo_: During update I get a note 'branch 'org.openembedded.dev' has multiple heads'. Is that OK ? Dec 30 11:36:04 no Dec 30 11:36:49 reenoo_: Dammed. That was a initial pull yesterday. Took 20 ! hours. Any hint how to go on ? Dec 30 11:37:49 pb_: why does this code get generated? Dec 30 11:38:06 uv1: did you do a commit? Dec 30 11:38:11 pester whoever is responsible for the multiple heads to merge them and pull that change (only that change, i.e. not from scratch) Dec 30 11:38:15 uv1: use merge or check if it is upstream Dec 30 11:38:30 uv1: and then do what reenoo said Dec 30 11:39:04 Tried the suggested "monotone merge". Got 'monotone: help required for 3-way merge' on distro/slugos.conf. Dec 30 11:39:42 zecke: Did not do a commit. Dec 30 11:40:25 I think it was koen ? At least merge says 'monotone: common ancestor 999aa50f7b3865183e757616bd277b34096c7f73 koen@openembedded.org 2005-12-09T10:40:03 found' Dec 30 11:41:01 Nevertheless. This does not have anything to do with my problem. That still exists after the last pull. Dec 30 11:41:36 zecke: dunno. if you can show me the configure.in source and the generated configure script, I can probably figure it out. Dec 30 11:42:09 Anyone aware of which package provides virtual/pcmcia-manager ? Dec 30 11:42:26 Can somebody explain how this virtual stuff works ? Dec 30 11:42:26 pcmcia-cs, I suppose. Dec 30 11:43:03 uv1: virtual/pcmcia-manager is long gone. there's a reason why I tell you to update your tree Dec 30 11:43:17 pb_: that is a nice offer Dec 30 11:44:32 reenoo_: Did it ! But still complained. Which package to clean ? And why is pcmcia-manager still in pcmcia-cs bb ? Dec 30 11:45:10 uv1: you can't update from a branch with multiple heads Dec 30 11:46:58 reenoo_: Shit, so we I have to wait ? Dec 30 11:47:09 Thanks so far ... Dec 30 11:47:10 in other news, I just pulled from vanille.de and .dev doesn't have multiple heads Dec 30 11:48:53 reenoo_: So whats wrong here ? Did a pull from vanille.de and and update of .dev and still have that multiple heads message. Dec 30 11:49:01 Hi koen! Dec 30 11:49:57 pb_: I have mailed you the files Dec 30 11:50:09 uv1: local changes? broken db? I don't know Dec 30 11:50:38 koen: I have a multiple heads message while updating from vanille.de. The merge says its distro/slugos.conf and you probably did something here. At least your name is mentioned during merge. Dec 30 11:50:49 reenoo_: no local changes ! Dec 30 11:51:16 03hrw 07org.oe.dev * r5d1c592a... 10/packages/iqnotes/ (files/pro.patch iqnotes_2.0.99-2.1.0rc1.bb): Dec 30 11:51:16 iqnotes: added 2.1.0-rc1 version Dec 30 11:51:16 - I do not know how to get it working without patching .pro file ;( Dec 30 11:51:21 03koen 07org.oe.dev * r3429ddf9... 10/packages/linux/handhelds-pxa-2.6/h2200/defconfig: h2200 defconfig: update to latest working version Dec 30 11:51:25 03arjan 07org.oe.dev * raba8f56e... 10/packages/mtd/ (mtd-utils/fix-ignoreerrors.patch mtd-utils_20050801.bb): mtd-utils: add patch to make nanddump --ignoreerrors actually do something different Dec 30 11:51:35 Dont know, whether I have a broken db and would not like to pull initial again just as a try! Dec 30 11:52:00 uv1: download a snapshot and use that as a starting point Dec 30 11:52:40 pb_: should I send you the original pkgconfig data? Dec 30 11:53:58 sure Dec 30 11:55:52 Have to leave now. Thanks and see you tomorrow ... Dec 30 11:56:06 oh, I see, that's rather cute. Dec 30 11:56:27 zecke: it declares this static array x[], then invokes "size" to figure out how large the elements are. Dec 30 11:56:56 pb_: but all the test cases return 0; and do not write a configval file ;) Dec 30 11:57:11 pb_: the ones from the original pkgconfig look fine and work on OS X Dec 30 11:57:19 zecke: yah, the testcases are not meant to be executed Dec 30 11:57:26 pb_: but after reconf the tests are totally fucked Dec 30 11:57:36 the configure script just compiles them and then inspects the resulting binary Dec 30 11:57:54 I suspect what might be happening is that your gcc is optimizing out the (apparently unused) variable, so the binaries always end up with zero data in them Dec 30 11:58:13 pb_: haha Dec 30 11:58:18 pb_: thanks a lot, I did not know size Dec 30 12:00:10 pb_: ah the one that works contains -g in the gcc call Dec 30 12:11:48 wow, libjava is still compiling Dec 30 12:29:00 wtf. how can I execute gcc without any optimisation without messing with CPPFLAGS anf CFLAGS... Dec 30 12:30:56 pb_: is that gcc HEAD you're working on? Dec 30 12:31:12 no, 4.0 at the moment Dec 30 12:31:43 ah, right, for debian Dec 30 12:32:08 right Dec 30 12:33:18 zecke: you can't. if you must change the optimisation, you will have to save CFLAGS, clobber it, run your compilation, then put CFLAGS back the way it was. Dec 30 12:33:20 ~lart optimizing compilers... Dec 30 12:33:20 * ibot readies the nuke launcher and fires some rounds at optimizing compilers... Dec 30 12:33:30 but even this is dangerous, since CFLAGS might include options like "-m64". Dec 30 12:34:16 pb_: even with -0O and -g... Dec 30 12:34:31 ah, heh Dec 30 12:34:36 your compiler must be too sophisticated Dec 30 12:34:57 maybe you can trick it by adding some statement like "x[0]++;" to the testcase. Dec 30 12:35:14 $CC -v "Zeckes Zero Size Output Compiler" Dec 30 12:35:43 hehe Dec 30 12:36:54 I'm compiling that code by hand now... to see the size Dec 30 12:38:47 or is BSD size != GNU size? Dec 30 12:39:37 should be the same, I think Dec 30 12:40:01 __TEXT __DATA __OBJC others dec hex Dec 30 12:40:02 8192 4096 0 12288 24576 6000 Dec 30 12:40:16 that looks good Dec 30 12:41:27 oh, no, it doesn't Dec 30 12:41:29 in fact, that looks bad Dec 30 12:42:27 I wonder why they have not dumped the sizeof() number to a file... Dec 30 12:42:37 yeah, that would seem better Dec 30 12:43:06 indeed, that seems to be what the GLIB_SIZEOF macro in that acglib.m4 that you sent me does. Dec 30 12:43:08 GLIB_SIZEOF does that... Dec 30 12:43:23 yeah Dec 30 12:43:26 right. I do not find the other test in any .m4 file Dec 30 12:43:32 I wonder where it comes from Dec 30 12:43:32 I guess you could install gnu binutils, but that seems a bit much. Dec 30 12:44:09 fun :) Dec 30 12:44:14 Ciao all Dec 30 12:44:42 welcome Dec 30 12:46:34 hi pigi Dec 30 12:46:51 hi pb_ ! Dec 30 12:46:58 pb_: oh that size output is really bad Dec 30 12:47:21 yeah, it's the worst Dec 30 12:49:12 could that size of test be 'built into' autoconf or m4? Dec 30 12:49:23 not m4. might be part of autoconf. Dec 30 12:49:37 m4 itself is just a macro processor, it doesn't know about that sort of thing. Dec 30 12:50:07 oh well... Dec 30 12:50:18 that size test makes complete sense for cross compiling Dec 30 12:51:12 hehe I could run configure and make, make install myself without using the autotools class.. Dec 30 12:51:29 yeah, that'd work Dec 30 12:51:55 looking at your configure output, I think those tests are the expansion of the AC_CHECK_SIZEOF() lines in configure.in. Dec 30 12:52:02 which would suggest that they are indeed an autoconf builtin thing Dec 30 12:52:13 (or patched by our autoconf) Dec 30 12:52:50 ah yeah Dec 30 12:53:16 any people can i help me in germany by a oe problem ? Dec 30 12:53:22 sure enough, there it is in sizeof_types.patch Dec 30 12:53:27 heh Dec 30 12:53:40 pb_: what make you believ that? Dec 30 12:53:58 what makes you believe that... Dec 30 12:54:02 zecke: these lines on my monitor: Dec 30 12:54:03 +AC_COMPILE_IFELSE([AC_LANG_PROGRAM([AC_INCLUDES_DEFAULT([$3])], Dec 30 12:54:03 + [static const $1 x[[256]];])], Dec 30 12:54:03 + [ Dec 30 12:54:03 + AS_TR_SH([ac_cv_sizeof_$1])=`$SIZE conftest.$ac_objext | tail -n 1 | $AWK '{print [$]3/256}'` Dec 30 12:54:04 I'm about to switch off my device Dec 30 12:54:18 Nix-niX: if you are at the 22C3 you might get help Dec 30 12:54:21 pb_: maybe you are tired as well? Dec 30 12:54:24 but that is finishing in about 4 minutes Dec 30 12:55:12 zecke: probably Dec 30 12:55:24 pb_: can I make that check conditional on $cross_compile Dec 30 12:55:47 yeah, good plan Dec 30 12:56:10 mithro, 22c3 ? Dec 30 12:56:14 this does not help me with cross compiling later on :} Dec 30 12:56:22 Nix-niX: Congress ;) Dec 30 12:56:29 well, when you're cross compiling you won't have a problem Dec 30 12:56:35 assuming you are using gnu binutils for your cross toolchain Dec 30 12:56:46 pb_: I'm a bit hesistant to use binutils ;) (Even if I love GNU) Dec 30 12:56:59 you'd only lose if you were somehow using bsd tools as your cross toolchain Dec 30 12:57:10 cya ;) Dec 30 12:57:18 later Dec 30 12:58:11 pb_: ah right. Lets hope that $SIZE expands to arm-linux-size Dec 30 12:58:16 anyway cya later Dec 30 12:58:26 welp the 22c3 is over Dec 30 12:58:33 we are packing u now Dec 30 12:59:38 i make in my bb-file FILES_${PN} += " ${datadir}/neutrino ${D}/var ${libdir}/neutrino " and oe not install in rootfs Dec 30 13:01:36 in tmp/work/neutrinong-1.0-r0/image/usr/neutrino is install Dec 30 13:02:37 well, that would be a problem Dec 30 13:02:58 none of the things you are adding to FILES_${PN} will match that directory. Dec 30 13:04:05 why Dec 30 13:04:22 ${D} is tmp/work/neutrinong-1.0-r0/image Dec 30 13:05:54 indeed Dec 30 13:32:04 hi Dec 30 13:39:55 join #openslug Dec 30 14:38:33 re Dec 30 14:39:40 zecke: wb Dec 30 15:27:38 it's that time.... Dec 30 15:28:15 good year to all, and that the new year would be better than the one that is finishing, for all of you ( and me too ) Dec 30 15:28:18 g'nite Dec 30 15:40:52 zdhehe Dec 30 15:41:07 it seems I'm still in some of the kernel packages as the maintainer :) Dec 30 15:41:39 heh, congratulations Dec 30 15:42:45 ah well. I directed Peter Halbmayer to talk to you folks :) Dec 30 16:05:30 reenoo_: thanks Dec 30 16:07:30 03jbowler 07org.oe.dev * r4cd08291... 10/classes/kernel.bbclass: Dec 30 16:07:31 kernel.bbclass: prevent self-referential asm links in kernel.bbclass Dec 30 16:07:31 - ln -sf asm- .../include/asm works fine first time, but Dec 30 16:07:31 - second time round it creates a link in asm to asm- Dec 30 16:07:31 - called asm-, this breaks the klibc build Dec 30 16:07:35 03jbowler 07org.oe.dev * ra8433080... 10/packages/klibc/ (files/install.patch klibc_1.1.1.bb klibc.inc): Dec 30 16:07:35 klibc: release 1.1.1 Dec 30 16:07:36 - klibc.inc now works with 1.1.1, builds packages and stages the Dec 30 16:07:38 - klibc to the host staging directory. klcc can be used to build Dec 30 16:07:40 - packages against klibc Dec 30 16:09:22 pb_: run! Dec 30 16:09:32 pb_: too late I'm going to abuse you know Dec 30 16:09:42 ! Dec 30 16:09:43 pb_: how does the 'site' think work with autoconf? Dec 30 16:09:57 the oe site file, you mean? Dec 30 16:09:58 pb_: I can make it not run a test with this site file Dec 30 16:10:02 pb_: yes Dec 30 16:10:14 it's basically the same as config.cache: it stores a pile of precalculated autoconf results. Dec 30 16:10:30 pb_: and we have results for the size of type already Dec 30 16:10:32 ahh Dec 30 16:10:56 well we do not need the sizeof patch for autoconf, our site scripts takes care of that already? Dec 30 16:11:09 yeah, that's probably true Dec 30 16:11:42 the sizes are (presumably) fixed for any given machine, so taking them from the site file ought to be fine Dec 30 16:11:52 I will be adventurous Dec 30 16:12:09 good luck Dec 30 16:13:02 pb_: luck won't be enough for autotools Dec 30 16:15:36 pb_: how do we fed the site script to autoconf? Dec 30 16:15:53 zecke: there's some voodoo in autotools.bbclass Dec 30 16:15:57 feed even.. and the other spelling/grammar mistakes should be fixed as well Dec 30 16:16:23 what else... Dec 30 16:17:55 oh, actually, it's defined in bitbake.conf Dec 30 16:18:07 "export CONFIG_SITE ..." is the important bit Dec 30 16:24:16 pb_: if the patch is needed I could put a powerpc site file with the right sizes there Dec 30 16:27:46 pb_: sadly we will need this patch :( Dec 30 16:27:58 ah Dec 30 16:28:04 why is that? Dec 30 16:28:19 pb_: some arch. I do not have access to do not define all sizes Dec 30 16:28:28 oh well Dec 30 16:28:32 pb_: some do not have float, double, ssize, size Dec 30 16:28:59 yeah, I guess people only fill in the site files as needed Dec 30 16:29:32 pb_: and I value mips more than building on OS X :} Dec 30 16:29:41 :-} Dec 30 16:29:56 pb_: I will write a site for darwin-ppp ;) Dec 30 16:30:32 good plan Dec 30 16:32:08 pb_: I will you abuse even more. do I find an site script in autom4te? Dec 30 16:32:44 don't think so Dec 30 16:32:50 afaik, autom4te is just a wrapper around m4 Dec 30 16:51:32 pb_: updating CONFIG_SITE in native.bbclass does not sound wrong?! Dec 30 16:53:53 it does sound a bit wrong. Why do you want to do that? Dec 30 16:54:18 to point to powerpc-darwin ;) Dec 30 16:54:26 otherwise I can not inject my values? Dec 30 16:58:30 okay I will try to update the sizeof patch to take cross compile into account Dec 30 17:13:55 surely in the native case the site file will already point to powerpc-darwin Dec 30 17:14:10 if it doesn't, that's definitely a bug Dec 30 17:18:43 using peek of the bitbake shell I see it points to arm-linux :} Dec 30 17:21:03 crumbs Dec 30 17:21:35 it shouldn't do: it's defined in terms of HOST_ARCH and HOST_OS, which are (should be) set by native.bbclass Dec 30 17:22:50 right Dec 30 17:24:07 hmmm Dec 30 17:24:19 now it evaluates to something empty... Dec 30 17:28:41 'night all Dec 30 17:46:59 koen: http://pastebin.com/484845 Dec 30 18:00:00 * pb_ zzz now Dec 30 18:00:01 night night Dec 30 18:01:17 pb_: cya Dec 30 19:02:52 lo Dec 30 20:33:40 Are there any known issues with glibc in OE? Keep getting a "Machine byte order unknown." #error on build. Dec 30 21:20:50 Nabyss explain Dec 30 21:22:12 gerwinin: I'm grabbing the sources yet again at present, but when attempting to do a 'bitbake pivotboot-image' for a zaurus (borzoi), glibc compilation fails with an #error Dec 30 21:23:17 I'm guessing that there's something probably missing from my local.conf with regards to target machine, but I've got MACHINE="borzoi" and TARGET_ARCH="arm" Dec 30 21:23:54 Can you put the error on pastbin Dec 30 21:23:58 So I can have a look Dec 30 21:24:50 When you specify machine than you do not need to specify target_arch Dec 30 21:24:52 Yep, in a couple of minutes.. I'm pulling down the libc source again, so once it (attempts to) build, I'll chuck it up Dec 30 21:24:57 ah, okies Dec 30 21:25:08 I've tried it with and without target_arch.. no luck both ways Dec 30 21:41:16 http://pastebin.com/484993 Dec 30 21:45:00 I think this is some kind of error that can be fixed with an arm specific thing Dec 30 21:45:06 Did you try to google ? Dec 30 21:45:45 Yeah, zilch Dec 30 21:46:10 But what I do know is that sometimes changing the prefered provide for glibc works Dec 30 21:47:07 So that for example you grab a previous version of glibc Dec 30 21:48:26 ah, okies.. where do I change that.. the .bb file? Dec 30 21:50:09 No in the local.conf Dec 30 21:52:34 I'm still a bit lost with what to put in there.. any pointers? Dec 30 22:00:02 Okay when my compile is finished I will give you an example Dec 30 22:00:08 Okies, thanks Dec 30 22:09:55 NAbyss, files have been moved around in the glibc CVS that OE depends on. Dec 30 22:11:25 apply this patch to the openembedded stuff and it should work: http://zoo.weinigel.se/n30/downloads/glibc.patch Dec 30 22:11:37 wingel_away: Thanks Dec 30 22:11:43 cd org.openembedded.oz354fam083 Dec 30 22:11:51 patch -p1 Yep, succeeded Dec 30 22:12:51 I'll see how the build goes Dec 30 22:24:19 Well, that solved part of the issue.. now it's terminating with this: Dec 30 22:24:34 | /mnt/personal/zaurus/tmp/work/glibc-2.3.5+cvs20050627-r1/temp/run.do_compile.20513: line 169: cd: /mnt/personal/zaurus/tmp/work/glibc-2.3.5+cvs20050627-r1/libc/sunrpc/rpcsvc: No such file or directory Dec 30 22:42:20 nm.. seems to be compiling now.. just needed to nuke tmp. Dec 31 02:16:39 good morning Dec 31 02:51:55 Good morning all Dec 31 02:58:13 morning **** ENDING LOGGING AT Sat Dec 31 03:00:44 2005