**** BEGIN LOGGING AT Thu May 26 23:59:56 2005 May 27 00:12:14 aahhh, that feels good: gpe-image-h1940-20050526145924.rootfs.tar.bz2 May 27 00:51:10 morning May 27 00:52:34 good morning all May 27 00:55:12 good evening May 27 01:05:51 morning all May 27 01:06:06 hey RP May 27 01:06:51 koen: Made up your mind about guadec? May 27 01:08:14 yeah May 27 01:08:22 I *want* to go May 27 01:08:43 But I have a progress report to hand in on tuesday May 27 01:09:21 I'm sure you could write it today ;-) May 27 01:09:33 The report is already done May 27 01:09:47 I'm doing the progress bit right now May 27 01:10:47 Ah, Its you'e making the progress match the report? :) May 27 01:11:04 yes May 27 01:11:14 rewriting a whole batch of stinking matlab code May 27 01:11:21 * XorA chuckles May 27 01:12:16 it beats rewriting java :) May 27 01:13:20 Im not sure about that, I've used matlab once May 27 01:14:40 I've never got on with java or matlab May 27 01:18:23 RP: Java is kinda like C except your not allowed to do anything without some extra brackets May 27 01:19:37 although the day gcj is Java 1.5 compatible I will rejoice May 27 01:21:37 XorA: :) May 27 01:21:41 morning all May 27 01:21:47 morning ade|desk May 27 01:30:54 hey May 27 01:30:58 ERROR: Unable to parse conf/bitbake.conf May 27 01:31:07 Whats this about? I'm following the getting started docs in the wiki May 27 01:31:35 df00z: your BBPATH is wrong May 27 01:31:38 RP: http://www.linuxdevices.com/files/article058/sld018.html May 27 01:32:02 BBPATH? May 27 01:32:04 i have a BBFILES May 27 01:32:06 no BBPATH May 27 01:32:16 oh oops May 27 01:32:21 missed the setup environment section May 27 01:32:30 wait no i didnt May 27 01:32:46 diefool@veracity ~/stuff/build $ echo $BBPATH /home/diefool/stuff/build:/home/diefool/stuff/openembedded May 27 01:33:47 hmm... what was the default root password of bootstrap-image again? I finally got it running, and now I can't login :( May 27 01:34:19 it isn't set May 27 01:34:35 My BBPATH isnt set? May 27 01:34:38 and unless DISTRO_TYPE=DEBUG ssh will not allow that May 27 01:34:44 koen: how can I set it? :) May 27 01:34:55 Arjan: serial console May 27 01:35:10 XorA: What should I do? May 27 01:35:22 df00z: follow instructions and set it May 27 01:35:27 don't have that, I'll do it manually :( May 27 01:35:39 XorA: its set.. May 27 01:35:41 diefool@veracity ~/stuff/build $ echo $BBPATH May 27 01:35:41 /home/diefool/stuff/build:/home/diefool/stuff/openembedded May 27 01:36:16 ~/stuff/build theres the conf dir it told me to make ~/stuff/openembedded theres conf files there May 27 01:36:50 koen: mallum beat you to it! :) May 27 01:36:58 damn May 27 01:37:15 so now what?> May 27 01:37:50 koen: is the /etc/passwd file on OE using the same algorythm as /etc/shadow on a normal (x86) system? May 27 01:38:09 I'm not sure it uses shadow passwds May 27 01:38:15 XorA: it says its unable to PARSE it May 27 01:38:20 not open it... May 27 01:38:36 hello all May 27 01:38:47 hey hrw|work May 27 01:38:51 df00z: no idea then May 27 01:39:18 time to rebuild kernel and hostap - tomorrow I will be in wifi area May 27 01:39:59 XorA: python version 2.3.5 ok? May 27 01:40:07 df00z: should be May 27 01:40:37 Well then..what the heck? May 27 01:41:15 moving bitbake.conf makes it say it cant open bitbake.conf May 27 01:41:27 but leaving it as is says it cant parse it..any way to get debug info? May 27 01:41:32 wow.. someone want to use OE for building stuff for m68k May 27 01:42:06 df00z: can you cat it? May 27 01:42:14 My bitbake.conf? May 27 01:42:21 (not to here, but, just, look in it?) May 27 01:42:50 yes i can May 27 01:43:05 what are the perms? May 27 01:43:13 hrw|work: AmigaOE :-) May 27 01:43:20 -rw-r--r-- 1 diefool users 14756 May 27 16:11 bitbake.conf May 27 01:43:22 hrw|work: got to be better than AmigaDE May 27 01:43:25 read. May 27 01:43:36 XorA: did you ever used AmigaDE SDK? May 27 01:43:51 hrw|work: for about 20 secs May 27 01:44:07 XorA: ;) May 27 01:44:13 df00z: OK, then I don't know either, permissions look ok (filesize matches too) May 27 01:44:52 ... May 27 01:44:53 *sigh* May 27 01:45:04 is there any way to make it be more verbose? May 27 01:45:30 oh wait whats this May 27 01:45:32 df00z: bitbake -D -D -D -D -D -D May 27 01:45:39 DEBUG: CONF file 'conf/site.conf' not found May 27 01:45:39 DEBUG: CONF file 'conf/auto.conf' not found May 27 01:45:51 DEBUG: CONF reading /home/diefool/stuff/openembedded/conf/bitbake.conf May 27 01:45:56 it did find that but not the other two? May 27 01:46:14 is that normal? May 27 01:46:24 site.conf and auto.conf are optional May 27 01:46:55 it isnt being very verbose..it just says it cant parse it May 27 01:47:01 DEBUG: CONF including /home/diefool/stuff/build/conf/local.conf May 27 01:47:01 ERROR: Unable to parse conf/bitbake.conf May 27 01:48:47 koen: whats your bitbake version # May 27 01:48:53 BitBake Build Tool Core version 1.3.0, bitbake version 1.2.9 May 27 01:49:02 df00z: latest svn May 27 01:51:17 k May 27 01:51:18 that worked May 27 01:51:22 must have been a bug in gentoos ebuild May 27 01:51:25 nothing new *cough* May 27 01:51:41 unless "new" covers the past year May 27 01:52:16 Gentoo used to work well May 27 01:52:17 ... May 27 01:52:27 NOTE: Not using a cache. Set CACHE = to enable. ERROR: no files to build. May 27 01:52:27 NOTE: Parsing finished. 0 cached, 0 parsed, 0 skipped, 0 masked. May 27 01:52:27 ERROR: Nothing provides pico May 27 01:52:33 hi May 27 01:52:38 trying to bitbake nano..jesus christ May 27 01:52:43 df00z: now that would be a BBFILES thing May 27 01:52:43 df00z: set a CACHE or it will eat RAM May 27 01:52:44 be too May 27 01:52:44 df00z: pico? May 27 01:52:47 hi Snoogie May 27 01:52:52 hi have a pb with OE May 27 01:52:56 i read the starting page May 27 01:53:03 Snoogie: Phil B? May 27 01:53:05 luke-jr_: is 12 mb eating ram? May 27 01:53:10 Phil B ? May 27 01:53:14 df00z: Note: 0 cached, 0 parsed, 0 skipped, 0 masked. May 27 01:53:14 ERROR: Nothing provides pico May 27 01:53:19 Snoogie: "pb" May 27 01:53:22 koen: 12 MB is with cache enabled, no? May 27 01:53:26 No, nano May 27 01:53:30 sorry i have a problem :) May 27 01:53:33 i tried pico too May 27 01:53:37 everything is installed and configured May 27 01:53:41 Snoogie: go ahead May 27 01:53:42 df00z: the important part is the 0s May 27 01:53:44 luke-jr_: I actually haven't tested without a cache to be honest May 27 01:53:49 luke-jr: 0s? May 27 01:53:55 i clean all (tmp/source folder) May 27 01:53:58 df00z: 0 cached + 0 parsed + etc = 0 May 27 01:53:59 df00z: BBFILES are wrong May 27 01:54:05 and now i do a : May 27 01:54:14 opt/build]$ bitbake nano May 27 01:54:15 NOTE: Using cache in '/o May 27 01:54:26 i always got a gcc error May 27 01:54:31 /home/diefool/stuff/openembedded/packages/*/*.bb ? May 27 01:54:38 koen: I had to enable cache to get the non-RAM-eating behaviour May 27 01:54:38 nobody speak french ? :) May 27 01:54:39 RP: you have to promiss to wave to some camera's a guadec so I can spot the embedded clan :) May 27 01:54:51 merde May 27 01:55:14 hrw|work: that looks right? May 27 01:55:22 koen : tu parles francais ? May 27 01:55:25 df00z: try it with 'ls' May 27 01:55:37 Snoogie: non May 27 01:55:37 ls /home/diefool/stuff/openembedded/packages/*/*.bb <-- should spam files at you May 27 01:55:48 argument list too long May 27 01:55:49 lol May 27 01:56:00 but anyway its there May 27 01:56:02 sounds right, then May 27 01:56:10 Right, im sure of that May 27 01:56:16 that's BBFILES? May 27 01:56:19 yes May 27 01:56:24 should it be BBFILES := May 27 01:56:26 sure it's not redefined later? May 27 01:56:26 or = May 27 01:56:43 I have := May 27 01:56:48 k May 27 01:56:51 in local.conf in my build dir? May 27 01:56:55 thats what it is... May 27 01:57:21 03koen 07 * r1.3423 10openembedded/packages/maemo/osso-thumbnail_0.3.1-1.bb: osso-thumbnail_0.3.1-1.bb: fix staging May 27 01:57:22 do you set BBFILES more than once maybe? May 27 01:57:28 did I? no. May 27 01:57:34 grep only finds one instance in local.conf May 27 01:57:38 hmm May 27 01:57:44 How do i set a CACHE? May 27 01:57:45 use absolute paths and no symlinks when possible May 27 01:57:49 export CACHE to whatever? May 27 01:57:55 Right. Not using any syms May 27 01:58:03 df00z: cache is default May 27 01:58:09 df00z: you disable it by setting CACHE="" May 27 01:58:10 gooooooooooood morning mickeyl May 27 01:58:11 NOTE: Not using a cache. Set CACHE = to enable. May 27 01:58:26 df00z: in local.conf May 27 01:58:28 someone can tell me where i can subscribe to OE ML ? May 27 01:58:33 # CACHE = "" May 27 01:58:35 its commented out May 27 01:58:45 hi mickeyl May 27 01:58:46 ok i found May 27 01:58:57 morning koen,hrw|work,RP, et. al. May 27 01:59:04 df00z: is OEROOT set? May 27 01:59:11 where is it set? May 27 01:59:15 local.conf May 27 01:59:19 Snoogie: http://openembedded.org/contact.php May 27 01:59:27 df00z: dunno, but my local.conf says # Uncomment this to disable the parse cache (not recommended). May 27 01:59:35 OEROOT is not set May 27 01:59:40 i dont see an option to set it May 27 01:59:48 are you sure you're having the right version? May 27 01:59:59 right version of? May 27 02:00:06 df00z: bitbake May 27 02:00:06 I just got OE from BK May 27 02:00:13 bitbake using latest May 27 02:00:15 hmm... OECONF might be a var I made up XD May 27 02:00:24 OEROOT isnt set May 27 02:00:27 is that important? May 27 02:00:32 OEROOT* May 27 02:00:39 Right May 27 02:00:41 its not set May 27 02:00:42 Oh. May 27 02:00:43 lol May 27 02:00:51 I think I made it up to use in the other vars May 27 02:01:15 echo $OEROOT May 27 02:01:16 there is no such thing as OE* May 27 02:01:18 it's an env var May 27 02:01:22 mickeyl: do we need all hostap* versions? May 27 02:01:24 uh, the scripts this uses doesnt rely on stuff like filesystems returning the ./ and ../ files in certain orders when like trying to find files in a dir do they? May 27 02:02:07 hrw|work: 0.2.x and 0.1.x can go away May 27 02:02:42 So now what? May 27 02:03:07 mickeyl: 0.3.2 too? May 27 02:03:19 yeah May 27 02:03:40 mickeyl: so leave 0.3.7 only? (0.3.5 are also not needed as 0.3.7 works ok rather) May 27 02:03:54 NOTE: Parsing finished. 0 cached, 0 parsed, 0 skipped, 0 masked. DEBUG: parsing complete May 27 02:04:00 bitbake doesnt find anything May 27 02:04:43 hrw|work: 0.3.7 seems very stable May 27 02:04:58 hrw|work: yank everything else. I'll add 0.4.x devel series soon May 27 02:05:01 df00z: what is your working directory? May 27 02:05:06 /stuff/build May 27 02:05:13 err /home/diefool/stuff/build May 27 02:05:22 where is your local.conf? May 27 02:05:32 mickeyl: ok. 0.4.1 released May 27 02:05:33 /home/diefool/stuff/build/conf May 27 02:05:52 df00z: make sure you're not defining TOPDIR May 27 02:06:06 anywhere May 27 02:06:09 in local.conf? May 27 02:06:09 k May 27 02:06:52 How does bitbake get a list of bb files? May 27 02:07:11 df00z: from BBFILES May 27 02:07:19 Right, but HOW May 27 02:07:25 functions, kernel command, python command May 27 02:07:51 Python's glob func, I thinkj May 27 02:08:31 How do i disable psycho May 27 02:08:34 maybe its that May 27 02:08:44 bitbake -h May 27 02:08:46 but don't bother May 27 02:08:47 it's not that May 27 02:09:07 haha nope May 27 02:09:20 03koen 07 * r1.3424 10openembedded/conf/distro/preferred-gpe-versions.inc: use newer, but still old API dbus in preferred-gpe-versions May 27 02:09:55 make.collect_bbfiles ( self.myProgressCallback ) May 27 02:09:58 whats collect_bbfiles May 27 02:10:05 Im beginning to think its the script May 27 02:10:46 def collect_bbfiles( progressCallback ): May 27 02:10:46 """Collect all available .bb build files""" May 27 02:11:10 Whered you see that? May 27 02:11:13 in bitbake? May 27 02:11:19 bitbake/lib/bb/make.py May 27 02:12:07 03hrw 07 * r1.3422.1.1 10openembedded/BitKeeper/deleted/ (13 files): removed all hostap-* older then 0.3.7 May 27 02:12:34 ahha wow python is cryptic to me May 27 02:12:54 Anyone running reiser4 here? May 27 02:13:08 I really doubt it's the filesystem May 27 02:13:16 You'd be surprized May 27 02:13:17 df00z: once tried - bitbake/oe failed to build on it so removed May 27 02:13:24 Laffo. May 27 02:13:32 reiserfs3 works ok May 27 02:13:34 it's not finding your cache dir either, for some reason May 27 02:13:41 reiser4 doesnt return ./ and ../ when searching for files in a dir w/ posix apis May 27 02:13:55 its not standard when to return ./ and ../ but lots of programs expect it to be first(or last..was it) May 27 02:14:03 i ran into this problem with wine and FindNextFileA function May 27 02:14:14 python is broken :mad: May 27 02:15:09 Google doesn't seem to think so... o.o May 27 02:15:16 Trust me, ask the wine devs May 27 02:15:28 ask wine devs about python? May 27 02:15:36 no, about the reiser4 "bug" May 27 02:15:38 you could always do a loopback ext3 mount =p May 27 02:15:44 Hows that work? May 27 02:15:47 use a sparse file May 27 02:16:06 Hm? May 27 02:16:15 do you have qemu installed? May 27 02:16:19 oh god May 27 02:16:21 no May 27 02:16:24 hmm May 27 02:16:30 * luke-jr_ ponders how else to make a sparse file May 27 02:16:32 I dont like where you're going with that anyway haha May 27 02:16:37 whats an sparse file May 27 02:16:37 oh May 27 02:16:44 sparse file is basically May 27 02:16:57 you have a 4 TB file that doesn't use any space until you write the data May 27 02:17:09 Ugghh... May 27 02:17:15 or any other size May 27 02:17:20 i could always upgrade to python CVS May 27 02:17:58 so you'd do simply: May 27 02:18:06 1. make-sparse-file my-oe-loopback May 27 02:18:24 2. mkfs.ext3 my-oe-loopback May 27 02:18:36 3. mount -o loop my-oe-loopback stuff2 May 27 02:18:41 4. mv stuff/* stuff2 May 27 02:18:42 etc May 27 02:19:02 ah May 27 02:19:07 God, thats annoying. May 27 02:19:38 better get separate partition for builds May 27 02:19:55 hrw|work: sparse file works just as well May 27 02:20:07 i could always learn a bit of python and perhaps write a test case...yell at the python devs and make them fix it May 27 02:20:34 df00z: or yell at reiser4 devs May 27 02:21:50 dd if=/dev/zero bs=1 count=1 seek=100000000000 of=my-oe-loopback May 27 02:21:55 that'll make a 94 GB sparse file :) May 27 02:22:11 luke-jr: wont that make a file that large? May 27 02:22:12 or no? May 27 02:22:12 how long take the compilation of nano the first time ? May 27 02:23:01 currently in "NOTE: package gettext-native-0.14.1-r3: task do_configure: started" for some minutes ... is it normal ? May 27 02:23:02 df00z: exactly May 27 02:23:08 df00z: but it will only use 4 K of disk space May 27 02:23:23 df00z: until you put files in it May 27 02:23:28 Interesting May 27 02:23:32 df00z: du -hs my-oe-loopback May 27 02:23:38 to see disk usage May 27 02:26:15 someone with c7x0, 2.6.12-rc and wifi card? May 27 02:26:27 hrw|work: yep May 27 02:26:41 Hertog: you have pcmcia-cs or pcmciautils? May 27 02:26:54 err... cs I think May 27 02:27:28 I know quite sure :) that's what I've been browsing at lately :) May 27 02:27:41 ok. so you got modules loaded after insert May 27 02:28:48 hrw|work: yes, hostap, _cs and _crypt_web May 27 02:29:28 ok. I have to trace what is going wrong here May 27 02:30:03 hmm... I need to remember I can't run kismet over ssh-over-wifi... May 27 02:30:19 Hm May 27 02:30:21 luke-jr_: or add another card... somehow :) May 27 02:31:42 dual cf jacket May 27 02:31:53 o wait, zauri aren't expandable May 27 02:32:36 ~lart libtool May 27 02:32:36 * ibot slaps libtool around with a large trout May 27 02:34:23 03koen 07 * r1.3424.1.1 10openembedded/packages/maemo/osso-thumbnail_0.3.1-1.bb: osso-thumbnail_0.3.1-1.bb: finally get it right so stuff actually links May 27 02:36:27 Hertog: 2.6.12-rc4-mm2 you are using? May 27 02:37:06 luke-jr_: heh heh May 27 02:37:12 hrw|work: in about 30 seconds I will be (again) flashing bootstrap-image now May 27 02:37:21 hey May 27 02:38:02 XorA: you are c7x0 user... have wifi card? May 27 02:38:48 http://pastebin.com/290622 May 27 02:38:56 hrw|work: I've got orinoco May 27 02:39:07 If I do this in the python interpreter, i get a bunch of [] [] [] [] [] [] [] printed out each on one line May 27 02:39:15 luke-jr_: pcmcia-cs or pcmciautils? May 27 02:39:35 contents is ok though, see the dir listings May 27 02:39:44 hrw|work: -cs May 27 02:39:52 guys i have this error : Connexion vers www.oesources.org[192.216.230.225]:80...connecté. May 27 02:39:52 requête HTTP transmise, en attente de la réponse...404 Not Found May 27 02:40:12 hrw|work: at prompt now, what is it you want to know? May 27 02:40:26 morning May 27 02:41:01 Snoogie: use LC_ALL=C when building - that way each one here will understand error messages May 27 02:41:05 hi CoreDump May 27 02:41:15 thanks May 27 02:41:28 hey hrw|work May 27 02:41:31 Hertog: can you install pcmciautils instead of pcmcia-cs and look does it load modules after card insert? May 27 02:42:49 hrw|work, in fact script try to grab http://www.oesources.org/source/current/flex-2.5.31.tar.bz2 bu doesn't find it May 27 02:42:52 hey May 27 02:42:55 i check and there is a new version May 27 02:42:57 available May 27 02:42:59 ehr... that's gonna take a while I'm affraid. I accidentally trashed my build/tmp (yeah I know... don't do rm -rf without looking very closely) May 27 02:43:02 how i can fix it ? May 27 02:43:31 Snoogie: bk mv old to_new_version, make it compile, send patches May 27 02:43:47 hi btw May 27 02:43:55 moin zecke May 27 02:44:48 whats this section supposed to do in bitbake? May 27 02:44:50 http://pastebin.com/290622 May 27 02:45:51 df00z: why not use make.collect_bbfiles from lib/bb? May 27 02:46:10 zecke: apparently it breaks with reiser4 May 27 02:46:20 zecke, i don't understand :( sorry May 27 02:46:42 i cannot update my openembedded folder ? May 27 02:46:44 zecke: its broken May 27 02:46:47 for reiser4 May 27 02:46:54 and im trying to fix it without any knowledge of python :P May 27 02:46:57 what is broken for reiserfs? May 27 02:47:06 03koen 07 * r1.3427 10openembedded/packages/maemo/hildon-fm_0.9.1-2.bb: Add hildon-fm, part of the maemo.org suite. May 27 02:47:09 the bitbake tool May 27 02:47:14 finding bb files May 27 02:47:31 df00z: I doubt that bitbake can break reiserfs4 May 27 02:47:41 hi pb May 27 02:47:49 hey pb_ May 27 02:47:52 zecke: nah, r4 breaks bb =p May 27 02:48:02 zecke: Sure it can, if anything relies on the filesystem listing "./" and "../" in a specific order May 27 02:48:05 luke-jr_: probably reiser4 is just broken? May 27 02:48:20 zecke: depends on the APIs and standards May 27 02:48:21 ./ and ../ can be returned at any point in a directory May 27 02:48:34 mickey|meeting: hostap-modules 0.4.1 does not need makefile.patch May 27 02:48:37 and i think it might be related to that May 27 02:48:43 hi zecke May 27 02:48:43 df00z: and where do we expect the order of . and .. in bitbake? May 27 02:48:52 I dont know. It might be an internal python thing May 27 02:48:53 zecke: Python's glob might? May 27 02:48:54 df00z: set your BBFILES correctly and don't blame a great fs May 27 02:49:05 zecke: you have r4 working? May 27 02:49:06 zecke my BB files is proper May 27 02:49:07 luke-jr_: hm May 27 02:49:14 we went over this for like 1/2 an hour in this chan May 27 02:49:30 df00z: try the sparse ext3 loop mount May 27 02:49:32 df00z: just to be sure May 27 02:49:33 df00z: did you ever parsed files before on reiserfs? May 27 02:49:39 zecke: no. May 27 02:49:44 Last time I used OE we had oemake May 27 02:49:46 and oebuild May 27 02:49:51 and it ate all my ram so i had to stop using it :( haha May 27 02:49:52 Snoogie: what is your problem with flex? May 27 02:50:15 zecke : my current script want v2.5.31 and only 2.5.4 is available May 27 02:50:31 Snoogie: 2.5.31 is available on sf.net May 27 02:50:33 zecke, i erase tmp and openembedded folder, now i do a bk clone bk://openembedded.bkbits.net/openembedded and i ll try again May 27 02:50:43 Snoogie: comment CVS_TARBALL_STASH and retry? May 27 02:51:10 Snoogie: how does your failure look? sf.net loves to bulshit you May 27 02:51:38 df00z: ok you think glob is broken? May 27 02:51:47 Likely May 27 02:51:50 wine was broken similar May 27 02:51:54 hrw|work, i try May 27 02:51:59 hold on working on it May 27 02:52:51 df00z: what happens if you put some files with absolute path name into the BBFILES? May 27 02:53:00 df00z: and how does your BBPATH look May 27 02:53:14 Snoogie: throwing away stuff and hope it magically works next time is not a solution May 27 02:53:22 Snoogie: what is your exact error with flex? May 27 02:53:34 hold on May 27 02:53:37 i know but yesterday nothing work and today everything seem to works except this error May 27 02:53:42 i really thing glob is broken May 27 02:53:47 zecke, i understand my error :) but i need to reinstall all May 27 02:53:54 next time i ll ask u before do this :) May 27 02:53:59 gimme some minutes :) May 27 02:54:01 yes it really seems it May 27 02:54:09 last time i reinstall all ;) May 27 02:55:11 df00z: from a quick view on glob I see nothing wrong with it May 27 02:55:24 df00z: pattern[0]!='.' is on the filename May 27 02:55:38 zecke: http://pastebin.com/290628 May 27 02:55:40 am i doing this right May 27 02:56:00 RP: ping May 27 02:56:13 * luke-jr_ also looked at glob's code and didn't see anything bad May 27 02:56:40 That code I pasted May 27 02:56:51 am i doing anything wrong in it? May 27 02:56:53 df00z: so check your BBPATH, conf/local.conf May 27 02:57:33 BBFILES? May 27 02:57:34 or BBPATH? May 27 02:57:49 both May 27 02:57:56 i dont have a BB path May 27 02:58:12 zecke: the weird thing is, his local.conf doesn't set CACHE, but cache is disabled... May 27 02:58:17 df00z: BBPATH is env var May 27 02:58:32 Where/what sets it? May 27 02:58:34 do I set it? May 27 02:58:44 yes May 27 02:58:45 *smile* May 27 02:59:06 BBPATH $PWD:openembedded:bitbake May 27 02:59:19 wait what May 27 02:59:22 what is BBPATH May 27 02:59:33 i have bitbake in my path May 27 02:59:35 /usr/bin May 27 02:59:52 zecke: BBPATH $PWD:openembedded should be enough iirc May 27 03:00:11 echo $BBPATH May 27 03:00:11 /home/diefool/stuff/openembedded/packages May 27 03:00:14 I already have that May 27 03:00:19 df00z: and it is wrong May 27 03:00:23 what do I put? May 27 03:00:26 or do you've packages/conf/* May 27 03:00:27 im just following the docs... May 27 03:00:30 or do you've packages/class/* May 27 03:00:37 df00z: BBPATH=builddir:OEdir May 27 03:00:51 What? May 27 03:01:04 /home/diefool/stuff/build and /home/diefool/stuff/openembedded May 27 03:01:09 yes May 27 03:01:10 build/conf has local.conf May 27 03:01:14 whats wrong with my bbpath May 27 03:01:16 export OEPATH=/home/diefool/stuff/build:/home/diefool/stuff/openembedded May 27 03:01:18 BBPATH=/home/diefool/stuff/build:/home/diefool/stuff/openembedded May 27 03:01:20 df00z: READ May 27 03:01:28 err... BBPATH* May 27 03:01:40 err wait May 27 03:01:48 no wait. read. =p May 27 03:02:03 echo $BBPATH May 27 03:02:03 /home/diefool/stuff/build:/home/diefool/stuff/openembedded May 27 03:02:03 df00z: how should bitbake know to look in build/conf for local.conf if you don't tell it is available? May 27 03:02:22 ERROR: Unable to parse conf/bitbake.con May 27 03:02:27 ERROR: Unable to parse conf/bitbake.conf May 27 03:02:32 * zecke 's ignore-list just grew again May 27 03:02:47 ...what May 27 03:03:12 anyway May 27 03:03:15 luke-jr: what am i doing wrong now? May 27 03:03:19 * luke-jr_ heads to bed before he bricks his Z... May 27 03:03:24 Argh. May 27 03:03:33 http://pastebin.ca/12769 - my setup.sh script May 27 03:03:35 df00z: not sure; my brain is going out on me =p May 27 03:03:49 I need to stop messing w/ stuff and sleep May 27 03:03:53 ah May 27 03:04:09 hrw|work: ill try this May 27 03:04:27 we should start a "buildsystem Hall of Shame" May 27 03:04:36 heh.. it looks like 2.6.12-rc4-mm2 does want to play with my wifi card.. tried hostap 0.3.7 and 0.4.1 May 27 03:04:50 koen: and add quotes from #oe there May 27 03:05:38 I go down.. to 2.6.11 May 27 03:05:54 hrw|work: indeed May 27 03:06:02 hrw|work: what card is it? May 27 03:06:16 hrw|work: we can let kergoth write the "why" :) May 27 03:06:23 * luke-jr_ is thinking about going down to 2.4 so his mic works again =p May 27 03:06:40 Hertog: dlink dcf-660w May 27 03:06:59 known issues: bitbake does not scan disk slices for conf/local.conf and only uses BBPATH May 27 03:07:23 ok May 27 03:07:25 now its back to this May 27 03:07:29 zecke: I think maybe the expectation was the working dir, actually May 27 03:07:30 NOTE: Parsing finished. 0 cached, 0 parsed, 0 skipped, 0 masked. May 27 03:07:38 zecke: since bitbake *does* use the working dir for tmp May 27 03:08:03 luke-jr_: no it uses TOPDIR May 27 03:08:03 luke-jr_: only if you don't set TMPDIR in local.conf May 27 03:08:16 zecke: by default, I mean... -.- May 27 03:08:16 hrw|work: hmmpf, no problems here with linksys WCF12, with pcmcia-cs, but that shouldn't matter I think May 27 03:09:00 pb_: I've only one issue left in my VC_svn.pm if nothing changed in the respository I get an error May 27 03:09:19 pb_: and I do not know how to ignore it :} eval sadly didn't help May 27 03:09:33 zecke: do you have a page with pretty coloured output yet? May 27 03:09:53 koen: http://127.0.0.1/tinder/status.html May 27 03:10:07 zecke: heh May 27 03:10:16 I actually had to think one second before not clicking that :) May 27 03:10:17 Hertog: 2.6.12-rc4-mm2? May 27 03:10:25 hehe May 27 03:10:33 zecke: what causes the error? May 27 03:10:37 hrw|work: yes May 27 03:10:43 zecke: ERROR: Unable to parse conf/bitbake.conf whats this? May 27 03:10:54 It can OPEN it, it cant pARSE it May 27 03:11:01 pb_: I remember what the last revision I saw was May 27 03:11:12 pb_: and next time I start with +1 May 27 03:11:27 ah, I see, so it blows up if that revision doesn't exist? May 27 03:12:08 Hertog: so I have no luck ;9 May 27 03:12:18 hrw|work: aahrg, scrap that. (yes I am on 2.6.12-rc4-mm2, and yes the modules do autoload, but no, no wifi0 or wlan0 do show up, very very sorry May 27 03:12:29 BBFILES := "/home/diefool/stuff/openembedded/packages/*/*.bb" in local.conf May 27 03:12:29 BBPATH is /home/diefool/stuff/build:/home/diefool/stuff/openembedded May 27 03:12:29 Hertog: can you pastebin /lib/modules/2.6.12-rc4-mm2/modules.alias? May 27 03:12:43 ERROR: Unable to parse conf/bitbake.conf May 27 03:14:18 pb_: eval {} helps a bit but tinder.cgi still prints garbage May 27 03:14:28 hrw|work: am i doing anything wrong _NOW_ ? May 27 03:14:35 hrw|work: I must have had too little coffee, no wonder no wlan shows up, have my bt card in it, they do show up.. will try to paste, wait a sec May 27 03:14:58 pb_: tonight I will play with my exim and procmail and create a python based tinder client May 27 03:15:30 zecke: ? May 27 03:15:38 Hertog: ok May 27 03:15:39 zecke: cool May 27 03:15:48 Hertog: heh May 27 03:15:53 ddf00z: you use reiser4 May 27 03:16:05 Yes? zecke told me not to blame reiser4 May 27 03:16:13 and then apparently blocked me May 27 03:17:15 ddf00z: if you can then try on other FS May 27 03:17:21 Ok i will May 27 03:17:30 I only once tried r4 and it failed for me May 27 03:17:53 Well if I switch to ext3 and it works can you tell zecke :sigh: May 27 03:18:11 hrw|work: didn't setting BBPATH help him? May 27 03:18:29 I already had BBPATh set i had like two terminals open trying different things May 27 03:18:53 hrw|work: http://pastebin.ca/12770 May 27 03:20:12 * CoreDump|home twitches when he hears the word "reiserfs" May 27 03:20:15 oh, btw, hrw: is something up with wifi cards? Like, they show up but dont work. dmesg shows orinoco_cs causing a kernel oops in the latest OZ. thats why im trying to build a new kernel/modules w/ OE May 27 03:24:00 bitbake works now May 27 03:24:04 err wait May 27 03:24:06 zecke: should help yes. but I remember my build problems when I tried reiser4... May 27 03:24:27 Hertog: thx May 27 03:24:37 ddf00z: what machine? May 27 03:24:43 in on the ext3 now May 27 03:25:02 ERROR: Unable to parse conf/bitbake.conf still getting that what the... May 27 03:25:15 ext3 mounted as loopback May 27 03:25:27 /home/diefool/stuff/build:/home/diefool/stuff/openembedded May 27 03:25:31 thats my BBPATH May 27 03:25:53 koen: according to terminal talk yesterday... hp2623a was old (years 70') text terminal working in 80x24 mode May 27 03:26:12 /home/mhentges/OpenEmbedded/bitbake/openembedded:/home/mhentges/OpenEmbedded/bitbake/openembedded:/home/mhentges/OpenEmbedded/bitbake <- that's mine, just for reference May 27 03:28:59 hrw|work: I only have those fancy hp Xterminals at home May 27 03:29:34 well May 27 03:29:39 its still mounted on a reiser partition... May 27 03:29:44 its an ext3 loopback in my home dir May 27 03:31:43 hrw|work: can you send me your bitbake conf file perhaps? May 27 03:32:22 ddf00z: bitbake.conf or local.conf? May 27 03:32:27 bitbake.conf May 27 03:32:49 ddf00z: I dont have own bitbake.conf - I use OE one May 27 03:32:57 right May 27 03:32:58 so do i May 27 03:33:03 But maybe by OE one is broken? May 27 03:33:28 DEBUG: CONF including /home/diefool/stuff/openembedded/conf/local.conf May 27 03:33:28 ERROR: Unable to parse conf/bitbake.conf May 27 03:34:08 Can you ask zecke to unblock me? This is his area it seems ._. May 27 03:34:41 looks like a b0rked BBPATH to me May 27 03:34:56 export BBPATH=/home/diefool/stuff/build:/home/diefool/stuff/openembedded May 27 03:35:14 openembedded/conf/bitbake.conf is there May 27 03:35:25 if i move it it says unable to OPEN bitbake.conf May 27 03:35:31 by default it says unable to parse May 27 03:35:59 ddf00z: where did you get OE from? May 27 03:36:49 BK May 27 03:36:52 you should have a bitbake/openembedded directory, not stuff/openembedded?? Where's your bitbake checkout? May 27 03:36:55 http://openembedded.org/cgi-bin/moin.cgi/GettingStarted?action=highlight&value=Getting+Started#head-01d3a3e1e5fc0d7e496a7b6af39a9e8785d65957 May 27 03:37:13 ddf00z: ok, not some wierd character conversion problem then May 27 03:37:31 CoreDump: bitbake checkout? i ran setup.py May 27 03:37:34 and installed it into my path May 27 03:38:04 dunno, does setup.py pull the svn bitbake stuff? May 27 03:38:38 i did that manually May 27 03:38:51 svn co svn://svn.berlios.de/bitbake/trunk/bitbake May 27 03:38:59 right May 27 03:40:20 Can you send me your bitbake.conf? May 27 03:40:23 FWIW, my "openembedded" directory lies withing the "bitbake" directory. Dunno if that is a requirement though May 27 03:40:34 I dont think that is as long as BBPATH is right May 27 03:41:04 my bitbake.conf has to somehow be bad... May 27 03:41:40 ddf00z: "bitbake" has an own conf/bitbake.conf. Maybe OE is expecting this file instead of openembedded/conf/bitbake.conf? May 27 03:42:27 Hmm, maybe May 27 03:42:51 YES :) I got dropbear<>desktop connection using public keys May 27 03:44:30 ok May 27 03:44:31 hrw|work: :) May 27 03:45:42 grrr. May 27 03:46:07 ~lart gdb May 27 03:46:07 * ibot hauls gdb up by the scruff of the neck and spanks him until he waddles May 27 03:46:16 odd May 27 03:48:41 CoreDump: it can read bitbake.conf from bitbake.confs dir May 27 03:48:44 but it doesnt work with OE May 27 03:48:50 it cant parse OE's bitbake.conf file May 27 03:49:43 well, I dunno what would cause that, sorry. If you are not on dialup, I'd sugest trashing OE and bitbake and start over May 27 03:50:01 ok i can do that May 27 03:52:13 this will only take a few mins May 27 03:59:26 * zecke considers getting naked in the office May 27 03:59:39 heh May 27 03:59:52 thank good my "office" has an a/c May 27 04:00:27 CoreDump|home: we've P4 systems here the Anti A/C May 27 04:00:39 :) May 27 04:01:00 at least you have something to play with. I'm stuck w/ XP at work May 27 04:01:24 on crappy Dell machines I might add ;) May 27 04:02:23 luckily we've more than enough PXA boards here May 27 04:02:37 * zecke considers installing Mosix on them May 27 04:03:42 zecke can i pm u ? May 27 04:04:00 Snoogie: what does pm mean? May 27 04:04:07 Snoogie: but yes just shoot May 27 04:04:36 thanks May 27 04:05:00 --13:00:15-- http://www.oesources.org/source/current/flex-2.5.31.tar.bz2 May 27 04:05:00 => `/home/snoogie/JLime/sources/flex-2.5.31.tar.bz2' May 27 04:05:00 Resolving www.oesources.org... 192.216.230.225 May 27 04:05:00 Connecting to www.oesources.org[192.216.230.225]:80... connected. May 27 04:05:00 HTTP request sent, awaiting response... 404 Not Found May 27 04:05:02 13:00:16 ERROR 404: Not Found. May 27 04:05:04 NOTE: Task failed: Fetch failed: http://aleron.dl.sourceforge.net/sourceforge/lex/flex-2.5.31.tar.bz2 May 27 04:05:06 NOTE: package flex-native-2.5.31-r2: task do_fetch: failed May 27 04:05:08 ERROR: TaskFailed event exception, aborting May 27 04:05:10 NOTE: package flex-native-2.5.31: failed May 27 04:05:12 ERROR: Build of nano failed May 27 04:07:04 http://openzaurus.org/wordpress/not-in-menu/passwordless-ssh/ - is it readable? May 27 04:07:39 RFC: moving default SF mirror to heanet (which works better then aleron) May 27 04:09:06 hrw|work: looks good May 27 04:09:27 hrw|work: looks fine to me too May 27 04:10:25 users suxx May 27 04:10:33 you opie dudes finally discover keybased auth? May 27 04:11:00 * CoreDump|home is using key-auth for years on his Z May 27 04:11:07 koen: no - it is response to one user post May 27 04:11:21 "The reason for removing dropbear is that dropbear cannot generate ssh keys which are important for easy syncing later." May 27 04:11:24 to bad it's an opie only tutorial May 27 04:11:31 koen: really? May 27 04:11:35 ROOT@c7x0 May 27 04:11:47 koen: feel free to change it May 27 04:11:50 bleh May 27 04:12:11 koen: forgot about brainless OZ users? May 27 04:12:25 if a user doesn't know how to use this doc for another user than root, then he is just beeing an idiot May 27 04:12:50 koen: How does the ipaq-pxa-2.6 handle the touchscreen definitions? May 27 04:13:12 RP: using detect-stylus (at least it should) May 27 04:14:21 koen: I'm trying to get it running on the hx2750 but something isn't working... May 27 04:14:21 hrw|work: "not allowed to edit this post May 27 04:14:37 koen: moment.. May 27 04:14:40 RP: it should run detect-stylus --device May 27 04:14:58 koen: now allowed May 27 04:15:57 wow.. rboot author finally posted root-on-sd howto (which will be obsolete with next release) May 27 04:16:06 koen: Its doesn't seem to setup a serial console either so I'm a bit stuck :) May 27 04:16:31 * RP is tempted just to use a c7x0 image which works ;-) May 27 04:17:22 hrw|work: http://www.hentges.net/misc/openzaurus/docs/altboot.shtml May 27 04:17:24 ;) May 27 04:17:26 later all May 27 04:18:10 RP: hacks are always faster, doing it "the right way" is 99% of the times slower May 27 04:18:35 koen: true :-/ May 27 04:18:41 Which package sets up inittab? May 27 04:18:52 CoreDump|afk: you should consider takeover collie maintaince May 27 04:19:49 RP: initscripts iirc May 27 04:20:44 RP: do you have hostap_cs based card? my dlink does not work with 2.6.12-rc4-mm2-r6 (after manually loading hostap_cs no interfaces appear) May 27 04:21:20 hrw|work: I think I have - Is the Buffalo hostap? May 27 04:21:41 RP: dont know May 27 04:21:53 hrw|work: There have been a number of changes to the kernel pcmcia layer - it may be your device's id has got lost :-/ May 27 04:21:56 hrw|work: I had my hostap working (ish) in 2.6.12-rc4-mm2 May 27 04:22:26 Was it just -r6 which broke things? May 27 04:22:40 RP: I rarely use that card - lat time was with 2.6.11 May 27 04:23:24 RP: where I should look for it? drivers/pcmcia/ somewhere? May 27 04:23:43 XorA: could you paste "(pc)cardctl ident"? May 27 04:23:58 hostap uses an external module doesn't it? May 27 04:24:07 hrw|work: I went back to 2.6.11, CF was giving me huge hassle on 2.6.12 May 27 04:24:18 RP: yes May 27 04:24:39 RP: I on the other hand have no problems with hostap May 27 04:25:10 XorA: anyway - could you? May 27 04:25:15 XorA: I'd be interested in a list of ways to make 2.6.12 oops - with the move to the driver model, its much more likely we can get fixes merged now May 27 04:25:41 wtf May 27 04:25:47 where did that page go? May 27 04:25:50 (hmm, thinking of it, didn't hostap change versions in OE recently?) May 27 04:26:11 RP: cardctl eject, ls /media/cf is a favourite May 27 04:26:28 koen: login into admin area, manage, pages, bottom of page May 27 04:26:32 RP: hostap didnt oops, but after a while the card would just stop taling wireless May 27 04:26:40 talking May 27 04:26:45 XorA: If you wait about 5 seconds between the commands does it do that? May 27 04:27:18 RP: not sure, Im busy weekens from now till June 20th so dont get much time to experiement May 27 04:27:21 hrw|work: With hostap being external, that makes the none functionality with 2.6.12 more of an issue :-/ May 27 04:27:33 hrw|work: Ah, I see, it reset the title May 27 04:27:36 http://openzaurus.org/wordpress/not-in-menu/passwordless-ssh/ May 27 04:27:45 koen: remove 'not-in-menu' part May 27 04:27:49 XorA: I understand. I haven't much time atm myself... May 27 04:27:57 RP: wedding in 2 weeks :-) May 27 04:28:11 XorA: That won't help either :) May 27 04:28:48 hrw|work: ok, moved to howto May 27 04:28:56 RP: so Im afraid I went for the old faithful until hecktickness is over May 27 04:29:45 koen: one root@c7x0 left ;) and that #... May 27 04:30:56 koen: Its sysvinit btw May 27 04:31:12 hrw|work: I left the '#' on purpose May 27 04:31:18 ;) May 27 04:31:29 and the remaining root should only be in your example :) May 27 04:32:33 generate public key May 27 04:32:33 root@c7x0:~/.ssh# dropbearkey -y -f db_key > db_key.pub May 27 04:32:38 5. point May 27 04:34:07 D'oh May 27 04:48:56 hi koen May 27 04:49:04 hey greg72 May 27 04:49:59 03mickeyl 07 * r1.3425.1.1 10openembedded/ (7 files in 3 dirs): clean up the flex packages May 27 04:51:05 yes, buffallo is a prism2 card using hostap May 27 04:52:45 mickeyl: flex 2.5.31 is available on sf.net May 27 04:52:59 I will push one fscking change to OE... s/aleron/heanet/ May 27 04:53:07 urghs May 27 04:53:09 aleron looks like many files missing May 27 04:53:18 don't say 2.5.31 is newer than 2.5.4 ? May 27 04:53:30 I would say so May 27 04:53:31 ~lart people using decimal subversions May 27 04:53:31 * ibot eats people using decimal subversions's liver with some fava beans and a nice chianti May 27 04:53:33 it is May 27 04:53:39 31>4 May 27 04:53:48 downloading flex-2.5.31: Select Preferred Mirror heanet (IE) ovh (FR) switch (CH) peterhost (ru) citkit (RU) nchc (TW) puzzle (CH) kent (UK) mesh (DE) easynews (US) jaist (JP) internap (US) ufpr (BR) umn (US) none May 27 04:53:52 i read it as 2.5.3.1 May 27 04:53:55 no aleron here May 27 04:54:03 will readd when a new release comes out May 27 04:54:27 so s/aleron/internap/ May 27 04:54:54 old flex (2.5.4) has some other quirks then 2.5.31 - some things are changed May 27 04:55:00 koen: why not heanet? May 27 04:55:17 don't know how that performs for our US buddies May 27 04:55:50 they already probably have configured other then aleron in local.conf May 27 04:56:36 mickeyl: 2.5.4 and 2.5.31 are not always compatible - in debian there is flex-old (2.5.4) and flex (2.5.31) because some soft need older version May 27 04:56:52 hrw|work: i see. well, then lets see what breaks next :) May 27 04:57:26 mickeyl: build flex 2.5.4 and then rebuild modutils - they are patched to build with 2.5.31 May 27 04:57:39 let me check May 27 04:57:55 Hmmm. Looks like ipaq-pxa-2.6 needs the disable-apm.patch we use on the c7x0... May 27 04:58:57 hrw|work: builds fine here May 27 04:59:04 tch, someone should just fix the darn 2.6 apm emulation. May 27 04:59:29 disabling APM in the xserver will break suspend/resume on the h5500, so that wouldn't be acceptable. May 27 04:59:30 pb_: If we knew what was wrong with it, perhaps one of us would... May 27 04:59:49 I'm a bit surprised that suspend/resume works on the zaurus without it, actually. May 27 05:00:10 pb_: I suspect it's already broken for the h5500 to be honest May 27 05:00:52 RP: possibly, but that doesn't mean that we should break it even more. May 27 05:01:05 mickeyl: anyway I would leave flex 2.5.31 and changed default sf mirror May 27 05:01:19 pb_: Providing the h5500 uses a 2.6 kernel, I don't see why X needs to mess around with the vt for a suspend/resume May 27 05:01:33 hrw|work: ya, i will ressurect it May 27 05:01:45 I think its been fixed in the kernel and X just gets comfused now... May 27 05:01:50 I'm not sure what you mean by "mess around with the vt", but X needs to know about suspend/resume so that it can repaint the contents of the framebuffer. May 27 05:02:05 What do you think has been fixed in the kernel? May 27 05:02:45 pb_: 2.6 kernels now switch to an unused vt just before suspending. X doesn't know about that and gets confused (seemingly) May 27 05:02:56 pushing sf.net switch to heanet May 27 05:03:04 X has no need to repaint the framebuffer May 27 05:03:46 03hrw 07 * r1.3425.2.1 10openembedded/conf/bitbake.conf: May 27 05:03:47 bitbake.conf: May 27 05:03:47 switched SF mirror to heanet as aleron (previous default one) lack many files) May 27 05:04:37 ah, the kernel synthesizes a pair of VT switches around suspend/resume? I guess that would indeed make it work. May 27 05:05:33 pb_: correct. But X gets confused so one workaround it to disable apm as it doesn't need to know anymore... May 27 05:06:02 Quite why everything gets so confused needs investigation... May 27 05:06:58 I see. May 27 05:18:40 bluelightning: hey May 27 05:18:46 hi zecke May 27 05:19:46 hmm, OE/conf/machine/zaurus-clamshell-2.6.conf has PREFERRED_VERSION_orinoco-modules = "0.13e-26" in it, which isn't available anymore May 27 05:20:19 Hertog: It can be changed to 0.15rc1 quite safely May 27 05:20:27 I'll fix that May 27 05:20:49 RP: I don't need the orinoco stuff I think, I just noticed it May 27 05:23:10 03mickeyl 07 * r1.3428.1.1 10openembedded/packages/flex/ (5 files): reintroduce flex(-native) 2.5.31 May 27 05:26:20 pb_: want the deatils of the X change? May 27 05:31:10 RP: I can still remeber all the details of the why and how X failed before that nasty hack May 27 05:38:23 hrw|work: is using pcmciautils done by replacing pcmcia-cs for pcmciautils in zaurus-clamshell-2.6.conf ? May 27 05:40:46 XorA: It might be an idea to put that into a summary email somewhere so someone with some X knowledge can sort the problem out May 27 05:41:07 Hertog: I'd expect so, if pcmciautils will build. I'm not sure its been tested May 27 05:41:18 it does build May 27 05:42:14 hrw|work: hmm, needs hot and coldplug I see, we'll try again :) May 27 05:42:40 RP: it is, on oe@handhelds.org May 27 05:42:52 03RP 07 * r1.3429.1.1 10openembedded/conf/machine/zaurus-clamshell-2.6.conf: Update zaurus orinoco-modules version May 27 05:42:56 XorA: Ok, remind me next week to do something about that :) May 27 05:43:09 Hertog: ignore that coldplug thing - it 'works' without it May 27 05:43:17 hrw|work: hmm, bb file needs some work, it doesn't install the binaries May 27 05:43:21 hrw|work: Ken|JLime May 27 05:43:28 euh, tab too many May 27 05:43:43 RP: X suspend problems on 2.6 c7x0 devices May 27 06:01:19 XorA: sure May 27 06:08:22 pb_: Well if you can search OE list for the email I detailed above that was a quick summary of the two problems May 27 06:14:33 XorA: your (1) sounds like it must be a kernel bug. May 27 06:14:42 pb_: that is what I thought May 27 06:15:04 (2) does sound like an X server problem. I'll take a look at that one when I have a spare moment. May 27 06:15:51 pb_: 2 is a couple of missing if(NULL) statements I think, but I was told X had a new patch to fix a lot of that, but I could not test as newer xorg builds dont work with tslib May 27 06:16:05 xorg? May 27 06:16:19 pb_: kdrive sorry May 27 06:16:25 ah, okay May 27 06:16:39 pb_: fingers off on their own journey again May 27 06:16:43 yeah, I gather there is some problem with the tslib interface at th emoment May 27 06:17:25 where 'some' is completely not working May 27 06:17:40 we could revert the tslib change with a patch to a recent cvs May 27 06:17:51 * Hertog was not made for this weather May 27 06:17:59 but that's ugly and messy May 27 06:18:08 koen: so is my kludge May 27 06:18:47 mallum and I got to the point it detect the ts right, but it failed in kmousereadbyte or something May 27 06:19:02 pb_: its a 1 line fix I think May 27 06:19:16 pb_: just finding where that 1 line is :/ May 27 06:19:18 mallum: aha. do you know which line? May 27 06:19:20 oh, heh May 27 06:19:38 mallum: and add the mouseadd() thingie to the tslib part May 27 06:19:57 koen: yeah but I dont know if that was right May 27 06:20:21 koen, pb_: its just a case of follwing the code through with gdb on something with tslib May 27 06:20:32 right May 27 06:21:24 there is an array of 'mice' of which something is getting trashed, not init'd right May 27 06:21:51 okay May 27 06:21:58 I'll see if I can find a suitable machine and have a look at it over the weekend May 27 06:22:33 pb_: aren't you attending guadec? May 27 06:22:40 nope May 27 06:24:18 pb_: :( May 27 06:53:12 pb_: I was looking forward to meeting as well :( May 27 06:53:46 RP: heh, I didn't know you were a gnome person May 27 06:55:09 pb_: 66% of the company is and I suspect this is a conversion exercise ;-) May 27 06:57:05 ah, like one of those corporate bonding things? very good. May 27 06:58:18 night all May 27 06:58:27 'night bluelightning May 27 06:58:40 'night bluelightning May 27 06:59:50 I'm still battling with this MACHINE=ipaq-pxa-2.6 image. detect-stylus --device returns the correct /dev/input/event1 May 27 07:00:03 If I cat /dev/input/event1 I have ts data May 27 07:00:26 and the ts works under the c7x0 image May 27 07:00:39 Yes the calibration util doesn't respond... May 27 07:01:02 There seems to be a suspicious lack of tslib config files... May 27 07:01:15 ipaq-pxa-2.6 is just a joke. I recommend you don't use that MACHINE. May 27 07:01:23 You probably want MACHINE=h3900 instead. May 27 07:01:35 that one needs a complete reworking too May 27 07:01:44 Why is it a joke? May 27 07:01:46 most stuff in there need a _kernel24 appended May 27 07:02:18 that might be true as well, but I think it's a better starting point than the ipaq-pxa-2.6 one. May 27 07:02:41 RP: well, not a joke in the sense of actually being funny, but it doesn't really work at all. May 27 07:02:47 I'm extremely close to using the c7x0 image as that just works... May 27 07:03:07 the plus of ipaq-pxa-2.6 is that it doesn't include ipaq-common, which is a collection of 2.4 only apps May 27 07:03:56 eh? May 27 07:05:15 In fact I can't see and reason for not using the c7x0 2.6 image as there's very little Zaurus specific about that May 27 07:05:23 BOOTSTRAP_EXTRA_DEPENDS += "kbdd bl hostap-modules orinoco-modules mipv6 atmelwlandriver" May 27 07:05:23 I think bl is the only one of those that's 2.4-only, and it would be harmless to have it in a 2.6 image. May 27 07:05:23 miplv6 comes to mind May 27 07:05:41 Doesn't MIPL work on 2.6? May 27 07:05:41 I'm fairly sure that recent mipl releases do support those kernels. Maybe our package needs updating or something. May 27 07:05:42 mipv6 want to compile the mipv6 kernel modules, which are 2.4 only May 27 07:05:49 What do the ipaq images gain over the zaurus ones? May 27 07:06:15 Dunno. You'd have to diff the config files. May 27 07:06:17 Probably not a whole lot. May 27 07:06:50 koen: oh, right. in that case, mipv6.bb is just broken and should be fixed. May 27 07:06:56 (I'm already using distro=oz as I want to use the same ipks as my Z's) May 27 07:07:08 the whole mipb6 is broken on 2.6 May 27 07:07:29 couldn't get the patch to apply to vanilla 2.6 May 27 07:07:54 and after it applied it turns out the kconfig is messed up May 27 07:07:57 Its just going to be a bit rough when OZ announces support for an ipaq :) May 27 07:08:15 I thought we agreed not to do that May 27 07:08:35 I'm trying my best not to... May 27 07:08:48 i see we have to merge sooner than later May 27 07:09:00 RP: well, distro and machine are already orthogonal. You've been able to build an openzaurus image for h3900 for months, in the same way that you can build familiar for collie. May 27 07:10:22 koen: sure, but irrespective of how messed-up the kernel portion is, it should not be lumped in with the userspace applications (which is what mipv6.bb ought to be building). May 27 07:10:44 afaik the userspace apps depends on the module May 27 07:10:59 in what way? you can't link an application against a module. May 27 07:11:01 Maybe we need a third option to go with MACHINE and DISTRO which sets certain image parameters May 27 07:11:11 RP: what parameters would that be ? May 27 07:11:20 pb_: it doesn't work without the module May 27 07:11:34 we could make the module a RRECOMMEND May 27 07:11:38 koen: right. what's your point? May 27 07:11:59 mickeyl: Basically something which would allow a 2.6-pxa rootfs to be shared between the hx2750, c7x0 and any other 2.6 pxa device which still having the fine control over the machine setting May 27 07:12:07 lots of applications won't run without the corresponding modules, but that doesn't mean they need to build those modules from their own .bb files. May 27 07:12:57 RP: what sort of "fine control" are you thinking of? May 27 07:13:05 pb_: specific kernels May 27 07:14:14 or specific modules May 27 07:14:20 RP: I think you should probably just have one target that builds and ships all the kernels. May 27 07:14:38 have a nice weekend guys May 27 07:14:43 you too, hrw May 27 07:14:56 hrw|work: you too :) May 27 07:15:34 I will look does Z will connect to neighbours wifi network ;) May 27 07:15:45 pb_: I disagree with that. I think there is a need to be able to specify a basic image type and then allow machine specific tweaks.... May 27 07:15:53 cu May 27 07:17:41 RP: if you have "machine specific tweaks", the image isn't going to be sharable any more. May 27 07:17:41 shareable, even May 27 07:17:41 I guess a sharable image would be something quite different May 27 07:18:13 The c7x0 2.6 rootfs image is near enough shareable as it stands May 27 07:18:42 I agree that a shareable rootfs is a desirable goal: it's certainly the intention that an h3900 rootfs should be usable on h3900, h5400, hx4700, h4100 and h1910. May 27 07:19:05 (plus axim, himalaya and no doubt other machines I've forgotten about) May 27 07:19:07 and h2210 May 27 07:19:12 oh yeah, and h2210 May 27 07:19:34 I think you could lump the c7x0, cx000, poodle hx2750, hx4700 and probably others as pxa-2.6 May 27 07:19:46 They're just the ones I know... May 27 07:20:05 You could probably include a load of Spyro's toshibas as well May 27 07:20:31 ah yeah, in fact I think we do have a pile of toshiba bits in the h3900 images. May 27 07:20:38 I kind of lose track of that stuff. May 27 07:21:03 What I'm thinking is you define a rootfs "type" and then a list of the machines you want it to include in the build May 27 07:21:27 Each of the machine specific files explains which kernel to build and what extra modules that machine needs May 27 07:22:09 As things stand, installtion of a kernel on the c7x0, cx000, poodle and hx2750 is a seperate process anyway May 27 07:22:25 (The hx2750 is probably always going to use the wince bootloader) May 27 07:22:55 jtagging it isn't going to be easy as I can't trace the connector... May 27 07:23:10 can the rootfs be flashed yet? May 27 07:23:40 koen: You could do it but if it goes wrong you're in extreme trouble - no jtag May 27 07:24:03 I've not been prepared to try jtag as I'll probably damage the device trying to connect to the pins May 27 07:24:20 I need to trace the connector May 27 07:24:33 I suspect that qte is going to continue being a problem for the whole shared rootfs thing May 27 07:24:49 ggilbert: Why? May 27 07:25:08 ggilbert: Can it not be configured at runtime? May 27 07:25:13 ggilbert: it seems qte is history anyway, so I guess that doesn't matter too much. May 27 07:25:14 heh May 27 07:25:34 RP: there are compile time differences between some of the clients May 27 07:25:36 but whatever May 27 07:25:43 if you guys want to drop opie support do it May 27 07:25:48 RP: not the bootloader, but the rootfs May 27 07:25:55 ggilbert: opie/x11 May 27 07:26:03 koen: You can't reflash from the bootloader May 27 07:26:18 opie/x11 is basically non functional May 27 07:26:19 RP: damn May 27 07:26:22 it was never finished May 27 07:26:29 koen: So if you break the rootfs, you break the device May 27 07:26:45 ggilbert: I have no intention of dropping support for opie May 27 07:26:46 RP: does jamey not have any jtag details for those machines? May 27 07:27:08 pb_: I know where the jtag pins are. I just can't get a connector to tap into them May 27 07:28:01 pb_: http://www.handhelds.org/moin/moin.cgi/HpIpaqHx2000Connector May 27 07:28:03 oh, I see. May 27 07:28:22 pb_: http://www.rpsys.net/ipaq/pictures/DSCF0920.JPG May 27 07:28:38 The connector in question is on the RHS, just below the wifi May 27 07:28:45 (white thing) May 27 07:29:18 If someone knows what it is that'd be good, a source for them would be even better :) May 27 07:29:19 whow, that's one crowded board :) May 27 07:29:23 ah, one of those little board-to-board connectors? May 27 07:29:40 pb_: Yes, the keypad usually plugs in there May 27 07:30:18 yah, most of the ipaqs have a connector a bit like that. May 27 07:30:40 hi, does the gpe calibration use another device than X itself? when X is starting (just before the calibration), I can move the pointer, but when it asks me to touch the crosshairs it doesn't do anything May 27 07:30:45 pb_: Do you have a manufacturer? May 27 07:31:11 pb_: or better still a part number? :) May 27 07:31:35 The problem is the beveled edges on it... May 27 07:31:44 The h5400 jtag is also on the keypad connector, which is the same kind of affair. May 27 07:34:02 RP: not based on that picture, but the four people I would go to are Molex, Samtec, Hirose and JST. I think all of them make connectors like that. May 27 07:35:52 I guess you already looked at Foxconn, right? May 27 07:39:15 pb_: I couldn't find anything at Foxconn, no... May 27 07:46:38 pb_: Its 0.5mm pitch if that helps narrow it down May 27 07:54:08 RP: thats one dinky WiFi card May 27 07:54:28 RP: if you send it to me I can solder onto that May 27 07:55:08 re May 27 07:55:24 I can do you a setup like my toshi's - with the wires routed out the headphone jack May 27 07:55:59 Spyro: I can solder onto it myself. Its just the keyboard will never work again :-/ May 27 07:56:25 I need that connector intact :-/ May 27 07:57:33 RP: no May 27 07:57:37 Can you not solder to the pins where the connector mounts on the board? May 27 07:57:41 RP: I mean I can leave the conector intact May 27 07:59:16 It may be possible to manage that. There really isn't a lot of space though and so far, I've not like to risk breaking it May 27 08:00:04 have you tried using a multimeter to beep out where the jtag pins from that header go? I wouldnt mind betting they come out on one of those testpoints too May 27 08:00:12 anyhow, my offer stands May 27 08:01:37 Spyro: I'll keep it in mind, thanks :). I am quite reasonable at soldering myself - I just don't want to break it... May 27 08:02:14 There are a load of test points and that's a good idea. I'll try that perhaps next week May 27 08:03:24 np May 27 08:03:50 that wifi card really is tiny though :) May 27 08:04:05 I thought my e800 had a small wifi card... that one takes the cake... May 27 08:04:22 pb_: For the record, all the above suppliers turn up negative for a 0.5mm pitch connector with that type of bevel May 27 08:04:38 It is rather neat :) May 27 08:04:57 RP: just a thought but cant you scrounge a dead PDA? May 27 08:05:17 then use a heatgun to remove the connector, solder on jtag, and araldite it for longevity May 27 08:05:59 if you ask in a forum or such, you could probably scrounge one for <30ukp or free May 27 08:24:31 <[g2]> mickeyl, bb -i is looking sweet :) May 27 08:29:36 :) May 27 08:31:53 zecke_: ping May 27 08:37:35 03koen 07 * r1.3427.1.2 10openembedded/packages/maemo/hildon-fm_0.9.1-2.bb: hildon-fm_0.9.1-2.bb: stage libs May 27 08:37:40 03koen 07 * r1.3427.1.1 10openembedded/packages/maemo/hildon-fm_0.9.1-2.bb: hildon-fm_0.9.1-2.bb: stage headers May 27 08:42:31 Spyro: The pda is that recent, I doubt there will be any broken ones around. If you know somewhere to look, please let me know :) May 27 08:42:35 * chouimat is away: lunch and cleaning May 27 08:44:19 what model ? May 27 08:44:23 RP? May 27 08:44:40 Spyro: ipaq hx2750 (hx2700 and hx2450 are similar) May 27 08:45:08 the one with the fingerprint sensor May 27 08:46:20 Yes, how similar the ones without the fingerprint sensor are, I don't know. I'd guess they'd leave it disconnected... May 27 09:00:55 03mickeyl * r224 10bitbake/lib/bb/shell.py: May 27 09:00:55 - add command specific tab completion May 27 09:00:55 - add expert mode - an embedded python interpreter May 27 09:00:55 - add exec command May 27 09:04:58 fun on a bun May 27 09:11:16 ibot: botmail to zecke: update_data in bitbake breaks DEPENDS. the standard DEPENDS as shown by bbread (i.e. glibc and a compiler) are no longer present then. May 27 09:43:59 koen: ping May 27 09:44:08 pong May 27 09:44:31 sorry I'm asking you again... don't know someone else who might know it... May 27 09:44:54 I've been testing with GPE on my h1940. it boots using NFS now May 27 09:45:01 cool May 27 09:45:19 when X starts, I can move the pointer, but then the calibration starts, and I can't interact with it May 27 09:45:31 does the calibration use another mouse configuration than the X server? May 27 09:45:45 it shouldn't May 27 09:46:15 try installing ts_callibrate and calibrate it on the cmdline May 27 09:46:17 guys i trying to compile OE May 27 09:46:26 i do exactly what written on starting guide May 27 09:46:31 and i get this error May 27 09:46:33 koen: I'll try that May 27 09:46:33 make[2]: *** [/opt/build/tmp/work/glibc-2.3.3+cvs20050221-r1/build-sh3-linux/tcb-offsets.h] Error 1 May 27 09:46:50 i have the file tcb-offsets.h.dT May 27 09:46:55 but not the .h file May 27 09:46:59 idea ? May 27 09:50:06 i got this error too : /opt/build/tmp/cross/sh3-linux/include/asm/ptrace.h:4:21: asm/ubc.h: No such file or directory May 27 09:55:18 koen: where do I find ts_calibrate and ts_test? *feels stupid* May 27 09:55:32 bitbake tslib May 27 09:55:39 that will give you those packages May 27 09:56:01 otherwise check ipkgfind.handhelds.org May 27 09:56:38 someone can help me ? May 27 09:56:45 the file ptrace.h use ubs.h May 27 09:56:46 ah, I see. thank you. May 27 09:56:48 have a good weekend May 27 10:33:03 anyone who can help me with a compilation error in OE ? May 27 12:12:40 morning May 27 12:31:27 * schurig has got his second child, Sarah, born on 25th with 2880 Gramm and 48 cm :-) May 27 12:31:42 schurig: congrats! May 27 12:32:58 Gratz :D May 27 12:39:41 schurig: Félicitations. May 27 12:50:40 Hi guys. May 27 12:51:09 whats the current state of play wrt the bitbake version of kdrive? May 27 12:51:24 does tslib work properly now? May 27 12:51:34 it doesn't May 27 12:51:38 doh May 27 12:51:45 how did it get broken ? May 27 12:51:45 use the 20050207 one May 27 12:51:53 Spyro: keithp 'fixed' it May 27 12:51:56 ok next question May 27 12:52:01 and probably didn't test it May 27 12:52:24 does the 20050207 one work across suspend/resume on 2.6 ? May 27 12:52:37 no, put use the c7x0 patch in the .bb May 27 12:52:42 s/put/but/ May 27 12:52:47 koen|tv: cheers May 27 12:52:49 that kind of fixes it May 27 12:53:06 ok next next question... I believe a LOT of improvements happened to bb lately May 27 12:53:15 pb_ said he'd try to look into it this weekend May 27 12:53:22 is it ok to just upgrade bitbake by replacing the bitbake dir ? May 27 12:53:27 yes May 27 12:53:30 that's whay I do May 27 12:53:31 cool May 27 12:53:42 I run it out of the svn checkout May 27 12:54:01 ok I have over 150 terminals open right now (7 desktops, two monitors each) so Im gonna clean up a bit ;-) May 27 12:56:08 <[g2]-away> on the bitbake update @ 224 May 27 12:56:24 <[g2]-away> bb -DD openslug-image May 27 12:56:24 <[g2]-away> DEBUG: CONF reading /home/tom/dev/bb_1x/openembedded/conf/bitbake.conf May 27 12:56:24 <[g2]-away> DEBUG: CONF file 'conf/site.conf' not found May 27 12:56:24 <[g2]-away> DEBUG: CONF file 'conf/auto.conf' not found May 27 12:56:24 <[g2]-away> DEBUG: CONF including /home/tom/dev/bb_1x/conf/local.conf May 27 12:57:11 <[g2]-away> I'm guessing site.conf and auto.conf are new May 27 12:57:26 no May 27 12:57:32 they have been there for a while May 27 12:58:25 <[g2]-away> hmmm... it finishes with May 27 12:58:26 I use auto.conf to hold MACHINE May 27 12:58:34 so I can switch it without using vi May 27 12:58:37 <[g2]-away> ERROR: Unable to parse conf/bitbake.conf May 27 12:59:13 *drool* ANTEX do a 0.12mm bit May 27 13:02:58 <[g2]-away> ok let me rephrase this. I"m getting a new error on the parsing how can I find out more information other than "bitbake -DD package" ? May 27 13:14:47 <[g2]-away> OK.... it's something with my local.conf May 27 14:04:54 03koen 07 * r1.3433 10openembedded/packages/maemo/osso-af-utils_0.4.bb: Add osso-af-utils, part of the maemo.org suite. May 27 14:30:50 Hey. BitBake can't parse the openembedded/conf/bitbake.conf file, It can open it, however. ERROR: Unable to parse conf/bitbake.conf May 27 14:31:20 I'm gonna guess your BBPATH is wrong. May 27 14:31:32 /home/diefool/stuff/build:/home/diefool/stuff/openembedded: May 27 14:31:54 in build theres a conf dir, in stuff/openembedded theres packages and all those other dirs May 27 14:32:58 If I move bitbake.conf from the openembedded dir, it says it cant open it May 27 14:33:20 [g2]-away was complaining about something similar before May 27 14:33:32 if i copy the bitbake.conf from the bitbake conf dir to the openembedded conf dir, backuping the old one May 27 14:33:36 it works but it cant find any bb files May 27 14:33:37 I guess it would really help if we can have some sort of idea what the parse eeror is May 27 14:33:44 Yeah May 27 14:35:00 * chouimat is away: stuff May 27 14:35:26 hmm May 27 14:39:54 so its just not gonna work? :\ May 27 14:40:08 bot unless you can figure out what the problem is. May 27 14:40:13 hmm May 27 14:46:08 03koen 07 * r1.3434 10openembedded/packages/maemo/osso-gwobex_0.26.1.bb: osso-gwobex_0.26.1.bb: fix staging, damnit May 27 14:46:37 if I add code to BBHandler.py in the lib/bb directory May 27 14:46:44 should it do it if i run bin/bitbake ? May 27 14:47:54 its totally ignoring code i put in there May 27 14:49:14 How rude May 27 14:50:08 heheh May 27 14:50:16 usually that means you are putting it in the wrong place May 27 14:50:23 Yeah usually May 27 14:50:44 Im trying to find bb.parse.handle May 27 14:50:50 i thought its in lib/bb/BBHandler.py May 27 14:51:06 try parse.py. May 27 14:51:26 no such python file? May 27 14:52:00 then look in the parse subdiretory May 27 14:52:25 BBHandler.py ConfHandler.py __init__.py May 27 14:52:35 start with __init__.py. May 27 14:52:41 work from there. May 27 14:53:52 theres a def handle there too May 27 14:53:56 in __init__.py May 27 14:54:08 adding debug messages there..still ignoring May 27 14:54:29 then start with bin/bitbake and work your way through the code May 27 14:54:31 line by line May 27 14:54:47 eventually you'll find the right spot May 27 14:55:25 evening May 27 15:04:05 Grr. I cant find the right parse. May 27 15:06:31 wait..its failing all the way at the bottom of the bitbake.conf file after include conf/local.conf May 27 15:06:39 build/${BUILD_SYS}.conf May 27 15:11:44 Are binaries stripped by default? And if so.. how to disable this on a per-package basis? May 27 15:12:16 hertog yes strip is on per default May 27 15:12:26 hertog look at various .bb files May 27 15:12:45 woglinde: thought so, saw it pass by in the logs. May 27 15:15:16 Ok fixed it May 27 15:15:33 You're not allowed to have " INHERIT =" or any spaces before stuff in local.conf May 27 15:19:01 Can OE build working orinoco/hostap/etc modules? May 27 15:19:19 In OZ 3.5.3, everything is broken and the ipks are broken in the feed. they make the kernel oops in dmesg May 27 15:19:40 OH! and fixed point math May 27 15:19:43 is that working? May 27 15:19:59 ddf00z: did you upgrade the kernel or just ipkg upgrade? May 27 15:20:20 hertog: just used the ipk..theres a kernel upgrade? May 27 15:20:29 in upgrade feed May 27 15:20:30 I upgraded the kernel and initrd to the latest oz May 27 15:20:47 oh maybe thats it then May 27 15:21:21 ddf00z: just put update.sh and kernelimage on sd/cf and do the 'ok'-start trick May 27 15:21:40 does OE build w/ fixed point math support in glib now? May 27 15:24:49 fixed point math? May 27 15:25:19 is that glib or glibc? May 27 15:25:29 Yeah, like...there was a feature that was going to be built in OE that makes glibc use fixed point May 27 15:25:31 glibc sorry May 27 15:25:48 Was that ever built in/working? May 27 15:26:02 there isn't such a feature May 27 15:26:11 you are probably referring to softfloat May 27 15:26:34 The kernel feature? May 27 15:26:40 Im not referring to the FPU emulator in the kernel May 27 15:26:41 OZ >3.5.0 is built with that May 27 15:27:05 How does it work? May 27 15:27:27 use google May 27 15:27:29 automatically does say a program compiled with the compiler and softfloat convert 3.2/3.5 to fixed point? May 27 15:27:42 it doesn't use fixed-point May 27 15:28:32 I can keep repeating myself, but it would be better if you just use google to look it up May 27 15:31:02 Hm May 27 15:31:32 If I build with soft-float, does my kernel need to include a floating point emulator still? May 27 15:33:12 google May 27 15:33:23 Google what? Im googling soft-float... May 27 15:34:30 i can't google "will soft-float take precedence over kernel floating point emulator" May 27 15:36:29 Aye. May 27 15:37:41 lol May 27 15:38:03 :( haha May 27 15:39:40 http://www.uclibc.org/lists/uclibc/2003-June/006380.html May 27 15:54:42 03mickeyl 07 * r1.3430.1.3 10openembedded/conf/distro/openzaurus-3.5.4.conf: bump oz-3.5.4 CVSDATE May 27 15:54:45 03mickeyl 07 * r1.3430.1.2 10openembedded/packages/wlan-ng/wlan-ng-modules_0.2.1-pre26.bb: fix wlan-ng-modules build for tosa May 27 15:54:48 03mickeyl 07 * r1.3430.1.1 10openembedded/packages/ (3 files in 3 dirs): apply hack to fix wlan-ng-modules compiling against oz-pxa-2.4.18-embedix May 27 16:01:45 03reenoo 07 * r1.3434.1.1 10openembedded/packages/gtk-webcore/ (4 files): gtk-webcore: switch to a fixed CVSDATE for now (current CVS is known broken) May 27 16:18:07 mickeyl: is something up with wlan modules lately in OE/OZ? May 27 16:50:30 InfoZIP has released unzip 5.52 and therefore deleted 5.51 from their download site. I did 'bk mv unzip_551.bb unzip_552.bb' and 'bk mv unzip-native_551.bb unzip-native_552.bb' and editted the 5.51 in unzip_552.bb to 5.52. Seems to build fine. May 27 16:50:40 I guess that INHIBIT_PACKAGE_STRIP = "1" prevents the 'auto'-stripping of a package? May 27 17:06:01 03reenoo 07 * r1.3437 10openembedded/packages/gtk-webcore/ (osb-nrcit_20050430.bb files/gdk-colorspace.diff): gtk-webcore/osb-nrcit: fix typo. apply patch by Kimmo Kinnunen to fix bitdepths mismatches. May 27 17:25:46 arm-linux-gcc-2.95: No such file or directory something stil relies on 2.95? May 27 17:26:10 it's trying to build the kernel May 27 17:26:48 <[g2]> thanks to whoever fixed flex-native May 27 17:27:09 <[g2]> however, the src_uri also needs to be changed in flex :) May 27 17:27:57 <[g2]> flex-2.5.31-r0 that is May 27 17:30:16 ddf00z: http://www.openembedded.org/cgi-bin/moin.cgi/ZaurusKernels May 27 17:32:36 Yeah I just saw that May 27 17:32:39 Thanks May 27 17:39:18 hertog: will orinoco-modules also build with 2.95? May 27 17:39:33 the ipks in the OZ feed cause the kernel to oops, thats why im trying to build them myself May 27 17:39:58 ddf00z: no clue. don't have such a card May 27 17:40:01 ah May 27 17:40:15 ddf00z: nor such a kernel ;) May 27 17:40:22 heheh May 27 17:40:26 Anyone know of a bug causing the touchscreen to work badly with xserver-kdrive? (movement and position getting wrongly translated) May 27 17:47:15 'night all May 27 17:49:42 HAHA sweet May 27 17:49:44 it works now May 27 17:49:48 AWESSSOMMEE May 27 17:51:00 Awesome. Orinoco works now, I guess the OZ package is no good. May 27 17:54:18 ddf00z: http://bugs.openembedded.org/ ;) file it May 27 17:59:41 its an OZ bug not OE May 27 18:49:49 * chouimat is back. May 27 20:48:01 ddf00z: 2.6 compiles w/ GCC 3 **** ENDING LOGGING AT Fri May 27 23:59:56 2005