**** BEGIN LOGGING AT Wed Aug 02 02:59:57 2006 Aug 02 04:20:57 03mwester 07org.oe.dev * r0b224bfa... 10/ (1 packages/images/unslung-image.bb): Unslung: unslung-image.bb - remove cpio and findutils records from ipkg database, and some other ipkg database cleanup. Aug 02 06:36:48 salut ! hello Aug 02 07:32:21 morning Aug 02 07:33:51 morning Aug 02 07:38:05 good morning all Aug 02 07:39:19 hey koen hrw|work Aug 02 07:39:53 my home machine drives me mad Aug 02 07:43:41 46°C inside of case == machine unable to use for >30minut in the evening ;( Aug 02 07:47:34 hrw|work: an amd? Aug 02 07:48:48 http://julien.danjou.info/blog/index.php/2006/08/01/312-etch-song Aug 02 07:49:00 XorA: athlonxp 2200+ (1.8GHz) to be exact Aug 02 07:49:09 XorA: cpu has up to 60°C Aug 02 07:49:17 ~lart amd for not being ntelligent like Intel with P4 Aug 02 07:49:17 * ibot flings poo at amd for not being ntelligent like Intel with P4 Aug 02 07:49:44 hey mikearthur Aug 02 07:49:57 morning koen Aug 02 07:49:57 hrw|work: my CPU reaches 60 sometimes, though its broken fan control in linux that causes it Aug 02 07:50:03 :-( Aug 02 07:58:37 ~lart fancontrol n linux Aug 02 07:58:38 * ibot readies the nuke launcher and fires some rounds at fancontrol n linux Aug 02 07:59:24 so does this summary stuff just make first mount faster, or all mounts? Aug 02 08:02:01 all Aug 02 08:07:32 is that 10M added to the image real? Aug 02 08:08:44 XorA: yes Aug 02 08:08:47 that's the catch Aug 02 08:09:16 so if you say "screw boottime", just use the regular jffs2 Aug 02 08:09:41 but if you are a former wince user that reboots every hour -> summary Aug 02 08:10:01 hrw|work: SRC_URI_append_h1940 doesn't work :( Aug 02 08:10:32 koen: ouch Aug 02 08:10:57 koen: although my wince machine gets rebooted about as often as my Z Aug 02 08:11:18 XorA: I expect to see improvements real soon(TM) since the OLPC ships with half a gig of NAND Aug 02 08:11:47 koen: heh heh Aug 02 08:12:04 * XorA wonders how to get hold of a OLPC Aug 02 08:12:27 http://wiki.laptop.org/go/Developers%27_program <- devboards Aug 02 08:13:34 ArjanS: look at packages/linux/linux-openzaurus_2.6.17.bb Aug 02 08:13:52 hrw|work: how's your monotone migration coming along? Aug 02 08:15:11 koen: going ok, pushed first things Aug 02 08:17:04 a nekkid chick has just stood in the window across street from my office, score :-) Aug 02 08:19:54 hrw|work: ah great, thank you! I didn't know I needed an extra backslash :) Aug 02 08:20:40 I wish Z had half a gig of NAND Aug 02 08:28:12 hey mallum Aug 02 08:29:23 hey XorA Aug 02 08:37:22 hey mallum Aug 02 08:37:31 hey ko Aug 02 08:37:33 en Aug 02 08:42:57 good morning Aug 02 08:43:56 hey florian_kc Aug 02 08:43:57 yo florian_kc Aug 02 08:45:02 god Aug 02 08:45:04 svn sucks Aug 02 08:45:20 it can't do 'svn add --unknown' Aug 02 08:45:39 koen: oooh, controversial opinion :-) Aug 02 08:46:33 now all the gtk+ experts are around, is it gtk+ 2.10.X that is causing a massive slowdown on application load on my device. Or are the performance problems in gtk+ in the widget rendering? Aug 02 08:46:37 i try to debug a framebuffer driver , what are your favorites test to do that ? Aug 02 08:46:51 XorA: pangoft Aug 02 08:47:08 koen: so pango is taking a LONG time to initialise Aug 02 08:47:09 ? Aug 02 08:47:29 XorA: no pango has lots of floats and does those per char iirc Aug 02 08:47:52 * koen looks at mallum Aug 02 08:47:59 koen: ok, so I wonder why suddenly my device is working like a win3.11 machine on a novell network Aug 02 08:48:06 XorA: o-hand has been busy profiling gtk+ and making pretty graphs Aug 02 08:48:19 koen: I was readng the mailing list yesterday Aug 02 08:48:38 XorA: compared to which other gtk version? Aug 02 08:48:46 florian_kc: 2.8 Aug 02 08:49:00 2.8.9 sorry Aug 02 08:49:39 unlikely imho... compared to 2.6 maybe but i don't think that much has changed from 2.8 to 2.10 Aug 02 08:49:58 XorA: svn checkout svn://projects.linuxtogo.org/svn/openembedded Aug 02 08:50:10 florian_kc: ok, I am looking for something else to blame then Aug 02 08:51:43 XorA: :-) Aug 02 08:52:17 XorA: basically lots of performance rendering issues in > GTK+ 2.6 Aug 02 08:52:50 we could apply all those 'no-cairo' patches Aug 02 08:52:54 mallum: cool, this is what I thought, I am working on elimination Aug 02 08:54:21 koen: the no cairo packages are in poky if you want to grab which help a bit Aug 02 08:56:07 XorA: http://svn.o-hand.com/view/poky?rev=519&view=rev <- try those Aug 02 08:56:53 I'm building from the current openzaurus-3.5.4.1 Aug 02 08:57:05 jorn is gonna hack some fixed stuff when he is back from hols Aug 02 08:57:16 just thought someone would know that xext-0.0cvs20060102: failed badly Aug 02 09:00:47 mallum: how does +#ifdef USE_CAIRO_INTERNALLY get selected? Aug 02 09:03:23 koen: I dont think it is :/ Aug 02 09:03:36 koen: Im not sure how jorn was building those Aug 02 09:03:51 mallum: the .bb doesn't mention it Aug 02 09:04:06 hrm Aug 02 09:04:39 kergoth: hi Aug 02 09:04:50 ~nslookup feeds.openzaurus.org Aug 02 09:04:53 I can't find feeds.openzaurus.org in DNS Aug 02 09:04:56 ;( Aug 02 09:07:32 er, forgot about it Aug 02 09:07:35 i need periodic pestering Aug 02 09:07:51 couldve sworn i added that Aug 02 09:08:16 kergoth: iirc mickeyl needed some OE dns changes as well :) Aug 02 09:08:21 dunno if he asked you yet Aug 02 09:08:42 echo 'echo "remember about CNAME"|mail kergoth'>>crontab Aug 02 09:08:47 have him email me, dont think i've seen anything from him Aug 02 09:08:48 hahah Aug 02 09:09:27 ops.. forgot about header... '13 * * * *' :D Aug 02 09:10:01 :) Aug 02 09:10:56 the one risk is that I could end in blacklist/spam folder Aug 02 09:11:52 .libs/libXext_la-DPMS.o: file not recognized: File format not recognized Aug 02 09:11:56 that is a wierd error Aug 02 09:11:58 any ideas? Aug 02 09:14:17 I'd fling poo at libtool Aug 02 09:14:49 mike@steyraug ~/oe $ libtool < poo Aug 02 09:14:49 bash: poo: No such file or directory Aug 02 09:14:53 didn't work Aug 02 09:16:11 heh Aug 02 09:16:35 hecklerundkoch was too long for a hostname? Aug 02 09:16:41 any non-poo related ideas? Aug 02 09:16:42 mike@steyraug ~/oe $ libtool < poo Aug 02 09:16:42 bash: poo: No such file or directory Aug 02 09:16:45 woops, sorrt Aug 02 09:16:49 sorry Aug 02 09:18:51 morning liam Aug 02 09:19:00 morning mike, all Aug 02 09:19:28 hey lrg Aug 02 09:21:13 hey liam Aug 02 09:22:18 cute, the new mindstorms is an arm7 Aug 02 09:27:54 koen: arse, the massive slowdown is ph5's change to fontconfig Aug 02 09:28:26 XorA: the one that stops the crashes? Aug 02 09:28:32 koen: yes Aug 02 09:28:42 koen: it must run round that loop longer than it used to Aug 02 09:28:54 arse Aug 02 09:33:08 koen: no ideas? Aug 02 09:33:25 hi Aug 02 09:33:48 mikearthur: nope, try asking OZ master hrw|work Aug 02 09:33:58 hrw|work: ping Aug 02 09:35:12 pong mikearthur Aug 02 09:37:10 .libs/libXext_la-DPMS.o: file not recognized: File format not recognized Aug 02 09:37:41 mikearthur: ? Aug 02 09:37:45 on compilingsorry Aug 02 09:37:50 on compiling Aug 02 09:38:02 xext-0.0cvs20060102 Aug 02 09:38:31 returned by ld Aug 02 09:39:13 mikearthur: if you want to say something to someone on irc you have to prepend sentence with nick of target person - otherwise you are talking to all Aug 02 09:39:22 ok, sorry Aug 02 09:39:43 hrw|work: ok, sorry, will in future :) Aug 02 09:39:43 mikearthur: I currently have 16 windows in irc client open so its hard to follow without that Aug 02 09:39:57 hrw|work: fair enough, sorry! Aug 02 09:40:45 hrw|work: any ideas would be welcome Aug 02 09:40:57 mikearthur: no idea Aug 02 09:41:03 sucks Aug 02 09:41:22 and few days will pass before I will find time to do any builds Aug 02 09:41:28 no worries Aug 02 09:41:50 hrw|work: rebuilding libtool-native Aug 02 09:41:53 mikearthur: do an objdump on the file Aug 02 09:42:01 lrg: ok Aug 02 09:46:02 lrg: I'm relatively unfamiliar with objdump, what should I be looking for? Aug 02 09:46:48 * kergoth pondders Aug 02 09:47:35 mikearthur: objdump will tell you if you have something that is at least valid (for a .o file) Aug 02 09:47:47 mikearthur: also try 'file' Aug 02 09:48:01 file just says data Aug 02 09:48:12 and objdump -x looks very wierd Aug 02 09:48:27 lrg: memory allocations are all 0 :s Aug 02 09:48:42 lrg: and unknown arch and the like Aug 02 09:49:35 mikearthur: file should say ELF 32-bit LSB relocatable, ARM, ..... and not data Aug 02 09:49:46 lrg: I suspected as much Aug 02 09:50:46 mikearthur: although, does the _la not mean something to libtool ? Aug 02 09:52:18 lrg: no idea Aug 02 10:21:10 hrw|work: what version of X does openzaurus .oz354 use? Aug 02 10:24:51 mikearthur: kdrive 0.0cvs20050207 Aug 02 10:29:39 hrw|work: should I try a version of libext from the .dev branch? Aug 02 10:30:46 morning all Aug 02 10:30:53 hey Dirk Aug 02 10:30:54 morninge do13_ Aug 02 10:31:02 hi dirk Aug 02 10:31:09 hey koen, mikearthur, Marcin Aug 02 10:31:21 do13_ got a question for you already :D Aug 02 10:31:43 mikearthur: ok Aug 02 10:32:18 mikearthur: which? Aug 02 10:32:42 do13_: .libs/libXext_la-DPMS.o: file not recognized: File format not recognized on compiling xext-0.0cvs20060102-r1? Aug 02 10:33:21 uff. tosa from .oz354x? Aug 02 10:36:36 ~lart tao ntent Aug 02 10:36:36 * ibot DoSes tao ntent Aug 02 10:39:04 hey Dirk Aug 02 10:39:43 hey Liam Aug 02 10:43:14 mikearthur: are you building gpe-image for tosa from .oz354x branch with distro=oz-3.5.4.1? Aug 02 10:43:27 do13_: exactly Aug 02 10:44:47 and it breaks on xext_cvs.bb? Aug 02 10:51:27 mikearthur: file Xext/.libs/libXext_la-DPMS.o: Aug 02 10:51:35 Xext/.libs/libXext_la-DPMS.o: ELF 32-bit LSB relocatable, ARM, version 1 (ARM), not stripped Aug 02 10:51:49 can't reproduce your prob Aug 02 10:52:52 do13_: sucks Aug 02 10:52:57 gcc -v? Aug 02 10:53:15 do13_: I'm on gcc version 4.0.3 Aug 02 10:54:15 gcc-Version 4.0.3 (Ubuntu 4.0.3-1ubuntu5) here Aug 02 10:54:25 wierd Aug 02 10:54:29 *g* Aug 02 10:54:32 I guess another rebuild may be in order :s Aug 02 10:54:49 ubuntu is an african word for "can't configure debian" *g* Aug 02 10:54:58 :) Aug 02 10:55:11 woglinde: or have better things to do that configure debian :p Aug 02 10:55:34 good point Aug 02 10:56:02 woglinde: I moved from Gentoo, the ultimate in having to configure everything, because I was wasting valuable coding time doing basic system setup Aug 02 10:56:16 mikearthur, :) Aug 02 10:56:19 mikearthur hehe Aug 02 10:56:22 mom Aug 02 10:56:43 do13_: reckon a rebuild is worth it? Aug 02 10:56:59 hehe read this -> http://www-128.ibm.com/developerworks/library/l-dist1.html Aug 02 10:57:09 did you changed the configs? Aug 02 10:57:21 not while building this one Aug 02 10:57:55 do13_: I'll post it, you can have a look Aug 02 10:59:09 do13_: http://mikearthur.co.uk/local.conf Aug 02 10:59:13 can you post the logs building xext? Aug 02 10:59:17 sure Aug 02 11:00:52 local.conf looks sane Aug 02 11:01:01 do13_: http://mikearthur.co.uk/xextcompile.log Aug 02 11:06:54 mikearthur: gentoo is for pansies, use LFS Aug 02 11:07:14 koen: one of my friends uses it as his major distribution Aug 02 11:07:22 koen: building Xorg is FUN Aug 02 11:07:40 mikearthur: I used LFS for years Aug 02 11:07:57 till I forgot to update samba and got r00ted Aug 02 11:08:03 debian ever since :) Aug 02 11:08:08 thats the thing Aug 02 11:08:20 mikearthur, i had debian for one year, then 3 years of gentoo now Aug 02 11:08:20 its just so obtrusive to keep distros like gentoo/lfs up to date Aug 02 11:08:30 I have debian for 10 years Aug 02 11:08:32 I used gentoo for 3 years Aug 02 11:08:39 and ubuntu for a few months now Aug 02 11:08:44 but my parents have been on it for a year Aug 02 11:08:48 well, kubuntu for me Aug 02 11:08:54 i got ubuntu on my notebook where i dont have time waiting for it to compile Aug 02 11:08:57 ten years of smooth updating via network Aug 02 11:09:06 :) Aug 02 11:09:20 the thing I came to the conclusion is Aug 02 11:09:21 problem about debian is the out-of-dateness in my pov Aug 02 11:09:30 ubuntu is better here with the half year cycle Aug 02 11:09:33 mikearthur: why is your arm-linux-gcc invoked without -march=armv5te -mtune=xscale ? Aug 02 11:09:43 do13: I wouldn't know Aug 02 11:09:47 thecan but I works Aug 02 11:09:54 is stable Aug 02 11:10:00 there ubuntu have some problems Aug 02 11:10:20 woglinde, well gentoo is also stable and works for me, didnt really have many problems in those 3 years Aug 02 11:10:42 but compiling is such a waste of time Aug 02 11:10:46 problem is just updating takes some time (well compiling..) Aug 02 11:10:56 thecan but company you dont want to compile all the time Aug 02 11:11:01 +in a Aug 02 11:11:03 yes Aug 02 11:11:49 do13_: I don't know, is that a big problem? Aug 02 11:12:03 and the performance benefit is negligable Aug 02 11:12:19 you can use emerge -k i think Aug 02 11:12:41 mikearthur: this is the only diff I see in our logs Aug 02 11:13:06 do13_: wierd Aug 02 11:13:17 TheCan: yeh, only if you have a binhost, and there aren't many Aug 02 11:13:34 oh ok Aug 02 11:13:35 didnt try Aug 02 11:13:47 why do you compile from source? Aug 02 11:13:49 mikearthur: unless you have to count a lot of leading zeros Aug 02 11:13:51 out of interest Aug 02 11:14:01 koen: heh Aug 02 11:14:08 is there any other compiling that from source? Aug 02 11:14:12 mikearthur, well gentoo has more uptodate packages etc. Aug 02 11:14:23 normally source-based distros are more recent than binary-based ones Aug 02 11:14:30 that?s the biggest benefit Aug 02 11:14:49 compiling doesnt you give much speed imho Aug 02 11:14:59 apart from that, when you change architecture you got big problems Aug 02 11:15:07 you ever tried debian unstable? Aug 02 11:15:08 thats why i run my amd64 with i686 simply Aug 02 11:15:16 yes i had debian unstable in the past Aug 02 11:15:23 too many senseless updates Aug 02 11:15:30 and some breakages Aug 02 11:15:41 apart from that still not up to date Aug 02 11:15:42 do you use x86 or ~x86? Aug 02 11:15:49 as some packages are unmaintained Aug 02 11:16:01 i normally dont use masked packages Aug 02 11:16:24 i rather try to build myself (without gentoo?s help) then or get a binary package somewhere Aug 02 11:16:49 fair enough Aug 02 11:16:56 what i dont like is when you can?t do emerge -v world anymore Aug 02 11:17:06 this is the situation for me now btw :) Aug 02 11:17:18 but using this etc/portage/* stuff can help Aug 02 11:17:59 Hello, I'm trying to do 'bitbake intltool-native', but it fails with non-presence of perl XML::Parser Aug 02 11:18:39 strange is that I bitbake'd XML::Parser, but it did not install anything in staging directory for my arch... Aug 02 11:19:04 mmp: ah, i guess you need this installed on your local system Aug 02 11:19:42 florian_kc: It is also installed on local system; I had a look into libxml-parser-perl bbfile and it seems to invoke perl from staging directory Aug 02 11:20:40 strange Aug 02 11:21:10 maybe I'm wrong Aug 02 11:22:46 let me try it once more; maybe I just messed some things up in my head:) Aug 02 11:24:05 No, I did not:( Aug 02 11:45:58 03koen 07org.oe.packaged-staging * r5549e516... 10/ (1 classes/packaged-staging.bbclass): classes/packaged-staging.bbclass: start adding support for cross/ Aug 02 11:46:03 03koen 07org.oe.packaged-staging * r2cab1957... 10/ (1 classes/packaged-staging.bbclass): classes/packaged-staging.bbclass: create CROSS_DIR Aug 02 11:46:07 03koen 07org.oe.packaged-staging * re2c985cb... 10/ (1 classes/packaged-staging.bbclass): classes/packaged-staging.bbclass: only package cross if it exists to avoid bogus packages Aug 02 11:46:11 03koen 07org.oe.packaged-staging * rb47b53c4... 10/ (1 classes/packaged-staging.bbclass): classes/packaged-staging.bbclass: fix logic for crossdir Aug 02 11:50:13 cute: http://lkml.org/lkml/2006/7/23/93 Aug 02 11:50:34 RMK post same to lakml Aug 02 11:50:43 I know Aug 02 11:50:48 that's where I got the link from Aug 02 11:55:02 * lrg gets lots of private support that he has to answer all the time Aug 02 11:55:30 * XorA hides from lrg for making hm busy Aug 02 11:55:58 hey pH5 Aug 02 11:56:09 koen: no plans for kernel-discuss@lists.ltg? Aug 02 11:56:24 hey koen, hrw|work Aug 02 11:56:32 hi pH5 Aug 02 11:56:38 hrw|work: no, since ltg doesn't have "it's own" kernel Aug 02 11:56:59 hrw|work: ltg is happy to host kernel projects, though Aug 02 11:57:17 pH5: XorA had some problems with the one-j-too-many patch Aug 02 11:57:39 XorA: it's not you. :) Aug 02 11:58:03 lrg: let me guess, they are at GMT+11 :-) Aug 02 11:58:35 Quickie question: is glibc-initial supposed to be compiled native, or cross (for target)? Aug 02 11:58:41 XorA: what kind of problems? Aug 02 11:58:58 pH5: it takes 5 seconds to start an app with that patch Aug 02 11:59:13 pH5: as compared to instant without Aug 02 11:59:48 XorA: that's bad. Aug 02 12:00:42 pH5: guess that wierd loop was an optimisation Aug 02 12:00:51 ~lart my I key Aug 02 12:00:51 * ibot decapitates my I key conan the destroyer style Aug 02 12:00:53 https://bugs.freedesktop.org/show_bug.cgi?id=7640 Aug 02 12:00:59 nobody commented on that yet Aug 02 12:03:29 XorA: from my (admittedly poor) understanding of the code, I have no clue why that change would make such a difference :-( Aug 02 12:03:49 pH5: no idea either, the change looks so simple Aug 02 12:04:31 03florian 07org.oe.oz354x * r58b31b4e... 10/ (1 packages/binutils/binutils-cross-sdk_2.15.94.0.1.bb): binutils-cross-sdk: Add custom install stage to fix several beakages. Aug 02 12:04:38 03florian 07org.oe.oz354x * rb767895b... 10/ (1 packages/meta/meta-sdk.bb): meta-sdk: Add some missing and useful libraries. Aug 02 12:04:43 03florian 07org.oe.oz354x * rf2119017... 10/ (1 packages/meta/meta-sdk.bb): meta-sdk: Remove libhandoff which isn't released yet. Aug 02 12:04:52 03florian 07org.oe.oz354x * rf241c59e... 10/ (1 packages/gnome/libsoup_2.2.93.bb): libsoup: add Aug 02 12:12:43 florian_kc: thx Aug 02 12:14:40 hrw|work: yw, i'll run a test with the latest modifications later today... need to wait for another image to finish. Aug 02 12:17:14 florian_kc: I need to test native sdk - probably will upgrade to binutils 2.16 Aug 02 12:21:34 03koen 07org.oe.dev * rd9402618... 10/ (1 packages/gpe-conf/gpe-conf_0.2.0.bb): gpe-conf: update to 0.2.0 Aug 02 12:21:44 03koen 07org.oe.dev * re7213895... 10/ (6 files in 2 dirs): gpe-conf: drop old versions Aug 02 12:22:16 hrm... i just ran into this freetype libz issue Aug 02 12:22:17 lrg: thought you were in this avo? Aug 02 12:22:42 florian_kc: disable zlib in freetype Aug 02 12:23:55 mikearthur: my bag hasn't arrived yet. I called and it's with the couriers somewhere in east Scotland. Aug 02 12:25:37 lrg: noobs Aug 02 12:25:57 koen: ok, thanks Aug 02 12:26:02 brb, food Aug 02 13:02:00 florian_kc: the XML::Parser packages was installed in OEROOT/tmp/staging/local/lib/perl5..., instead of the x86_64 directory. I'm going to rebuild perl to see if there is not something wrong with it Aug 02 13:02:21 s/packages/package/ Aug 02 13:18:33 03xora 07org.oe.dev * rf0e88c30... 10/ (1 packages/sylpheed/sylpheed-claws_2.4.0.bb): sylpheed-claws_2.4.0.bb : new version Aug 02 13:18:43 03koen 07org.oe.dev * rf008d20c... 10/ (1 packages/xorg-font/font-util_X11R7.1-1.0.1.bb): font-util_X11R7.1-1.0.1: fix RDEPENDS Aug 02 13:19:06 djeez Aug 02 13:19:20 if kdrive finds fonts it suddenly insist on loading a cursos font Aug 02 13:19:26 cursor* Aug 02 13:19:34 lol Aug 02 13:20:18 and exits if you don't have it Aug 02 13:20:31 but if you don't have any fonts *at all* it will happily load Aug 02 13:22:43 gpe-login still crashes, but the strace log now has less errors Aug 02 13:23:02 it works fine over remote-X, so kdrive is buggy Aug 02 13:23:32 koen: dillo kills my kdrive :-) Aug 02 13:24:54 koen: do you also have problems with kdrive interpreting the ts wrong? on my (recent) .dev image, it handles every first tap as 'focus this field', and a second tap on the same place is the actual mouse click Aug 02 13:27:09 ArjanS: haven't noticed that Aug 02 13:27:13 mmp: sounds like a bug Aug 02 13:28:01 florian_kc: now I debugged few things; it is installed there, but perl looks for its modules also in that directory, so this should be ok Aug 02 13:29:06 I need to debug a bit that bbfile; when running OE's version of native perl manually and requiring XML::Parser, it complains that it can't find libexpat.so.0. I wonder if this same happens also in bbfile Aug 02 13:31:53 koen: okay... can you perhaps provide some pointers as in where I can look for bugs? the touchscreen works perfectly in old gpe images, opie and ts_test Aug 02 13:32:13 ArjanS: kdrive rotation? Aug 02 13:38:44 koen: sorry, didn't understand what you mean Aug 02 13:41:21 mmp: perl-native in this case Aug 02 13:41:30 mmp: and it is complaining about a missing libexpat.so Aug 02 13:41:44 mmp: could you use ldd on the perl c module to check where it is checking for libexpat Aug 02 13:42:00 hmm, gpesyncd fails during linking with undefined references to libz stuff (deflate, inflate, etc.) Aug 02 13:42:42 * hrw|work wait for a day when OE autobuilder will be possible Aug 02 13:43:07 hrw|work: let us use SUSE buildservice Aug 02 13:43:54 zecke: do you think Expat.so? This one just tells me "not found" for libexpat.so.0 Aug 02 13:44:10 zecke: currently OE does not support "bitbake --from-scratch gpesyncd" in acceptable time Aug 02 13:45:07 pH5: hmm... nasty Aug 02 13:45:28 pH5: i'm just building from the scratch... i'll wait for the same to happen here. Aug 02 13:45:57 zecke: but I need to prepend LD_LIBRARY_PATH=/path/to/oe/libdir to perl call to make it succeed. Aug 02 13:46:45 mmp: okay, this needs to be changed :) Aug 02 13:48:00 zecke: bitbake -e does not show anything containing LD_LIBRARY_PATH... Aug 02 13:50:50 XorA: try installing preling (and prelink -u /bin/busybox afterwards) Aug 02 13:51:54 koen: what kind of mess s that gong to get me in :-) Aug 02 13:52:17 XorA: apps startup a lot faster Aug 02 13:57:12 zecke: definitely confirming the original problem was also caused by this - "LD_LIBRARY_PATH=$OEROOT/build/tmp/staging/x86_64-linux/lib/ bitbake intltool-native" build intltool-native correctly Aug 02 13:58:26 koen: gelf_getverdaux.os gelf_getrel.os gelf_getrela.os gelf_update_rel.os gelf_update Aug 02 13:58:29 _rela.os gelf_getdyn.os gelf_update_dyn.os gelf_getmove.os gelf_update_move.os g Aug 02 13:58:32 elf_getsyminfo.os gelf_update_syminfo.os gelf_xlatetof.os gelf_xlatetom.os nlist Aug 02 13:58:34 .os gelf_getsymshndx.os gelf_update_symshndx.os gelf_update_versym.os gelf_updat Aug 02 13:58:37 e_verneed.os gelf_update_vernaux.os gelf_update_verdef.os gelf_update_verdaux.os Aug 02 13:58:39 elf_getshnum.os elf_getshstrndx.os gelf_checksum.os elf32_checksum.os elf64_che Aug 02 13:58:42 cksum.os gelf_rawchunk.os gelf_freechunk.os libelf_crc32.os libelf_next_prime.os Aug 02 13:58:44 elf_clone.os gelf_getlib.os gelf_update_lib.os Aug 02 13:58:46 arm-linux-ranlib libelf_pic.a Aug 02 13:58:49 arm-linux-gcc -march=armv5te -mtune=xscale -shared -o libelf.so -Wl,--whole-arch Aug 02 13:58:52 ive,libelf_pic.a,--no-whole-archive \ Aug 02 13:58:54 -Wl,--version-script,/home/dp/zaurus/build-gcc/tmp/work/armv5te-li Aug 02 13:58:57 nux/elfutils-0.108-r0/elfutils-0.108/libelf/libelf.map,--no-undefined \ Aug 02 13:58:59 -Wl,--soname,libelf.so.1,-z,-defs Aug 02 13:59:02 damnation sorry people, wrong key in wndows Aug 02 13:59:04 koen: http://pastebin.ca/111866 Aug 02 13:59:08 XorA: lol Aug 02 13:59:14 lame excuse Aug 02 13:59:24 ~lart me for being a lamer Aug 02 13:59:24 * ibot offers xora some herring for being a lamer Aug 02 13:59:41 mmmm, herring.... Aug 02 14:00:28 XorA: it works for me(TM), which is all the advice I can offer Aug 02 14:07:31 "mov r0,r0" discussion is fun Aug 02 14:10:43 hrw|work: you got src and sest backwards :-) Aug 02 14:10:51 s/sest/dest/ Aug 02 14:13:24 koen: bleh I understand not this elf magic Aug 02 14:14:01 XorA: dwarf magic would be better anyway ;) Aug 02 14:14:28 koen: Dark Elves have better babes Aug 02 14:16:01 XorA: if you aren't turned on by beards, that is. Aug 02 14:20:00 pH5: if your into that come to the backwoods of scotland :-) Aug 02 14:21:19 * mwester wonders at the unexpected things one learns on #oe :-D Aug 02 14:30:03 note to self: check md5sums of all sd/mmc cards with 2.6.17 Aug 02 14:35:34 ~lart libpng,.so Aug 02 14:35:34 * ibot --purges libpng,.so Aug 02 14:35:57 haha Aug 02 14:40:04 bonjour alan|philo :) Aug 02 14:40:22 koen, did I get added to the commit list for oe (philip@balister.org) Aug 02 14:40:45 Crofton: which list? Aug 02 14:40:52 commit list for montone Aug 02 14:41:02 I sent my key to you and mickeyl Aug 02 14:41:04 Crofton: if you were, you'd have a mail from mickeyl in your inbox Aug 02 14:41:07 last week Aug 02 14:41:10 ok Aug 02 14:41:13 I don't :) Aug 02 14:41:22 haeh? don't we have a mailman interface? Aug 02 14:41:38 zecke: for adding keys? Aug 02 14:43:20 bonjour Genesis Aug 02 14:43:25 hi all Aug 02 14:44:17 hi alan Aug 02 14:44:18 koen: does Angstrom image work yet? Aug 02 14:45:01 Crofton: you confuse me. By commit list you mean mailinglist? Aug 02 14:45:32 no Aug 02 14:45:38 monotone access :) Aug 02 14:53:21 afternoon Aug 02 14:54:52 XorA: apart from gpe-login crashing, yes Aug 02 15:15:27 morning Aug 02 15:16:30 morning chouimat Aug 02 15:24:33 after spending some time with my work enviroment I feel why OLPC devteam care about memory usage so much Aug 02 15:24:41 Mem: 775752k total, 755920k used, 19832k free, 82308k buffers Aug 02 15:24:41 Swap: 522072k total, 457896k used, 64176k free, 321880k cached Aug 02 15:27:54 well - duh Aug 02 15:29:15 I could understand so big usage with java apps but I'm getting scared of that usage Aug 02 15:30:32 hehe Aug 02 15:33:22 hi all Aug 02 15:51:47 cu all Aug 02 16:29:34 bye Aug 02 17:07:20 RP: ping Aug 02 17:08:19 JustinP: iirc RP is on holiday for this week Aug 02 17:09:02 ah.... Aug 02 17:09:18 I don't suppoose anyone else know how to get the state of the hinge in the 2.6 kernels.... Aug 02 17:09:40 JustinP: it should be in the zaurusd source Aug 02 17:09:50 and there's the chkhinge26 in o-hand svn Aug 02 17:10:06 zaurusd didn't help me Aug 02 17:10:13 I copied all of the relevant code and always get 0 Aug 02 17:11:58 and with chkhinge26? Aug 02 17:12:06 I don't know.....haven't found it yet Aug 02 17:12:59 http://svn.o-hand.com/view/misc/trunk/ Aug 02 17:13:00 all I see is matchbox on http://svn.o-hand.com/view/matchbox/ Aug 02 17:13:08 argh Aug 02 17:13:13 missed the drop-down Aug 02 17:16:07 kergoth: What's the state of tslib? Is Russel's CVS still the repository for it? Aug 02 17:16:19 jg_: yes Aug 02 17:16:20 jg_: it is Aug 02 17:16:33 jg_: it works, is stable and stagnates :) Aug 02 17:17:09 zecke: thanks. Is it worth bothering with? Aug 02 17:18:16 bbiab Aug 02 17:22:16 jg_: theres not really any alternatives Aug 02 17:22:36 jg_: and dispite being cvs only, its pretty heavily used Aug 02 17:33:04 jg_: sure. What else would you use? Aug 02 17:33:32 jg_: it is stable, stable api, it doesn't leak and has good debugging tools while being flexible :) Aug 02 17:33:39 zecke: one can always write code from scratch ;-) Aug 02 17:34:03 hehe Aug 02 17:34:39 jg_: it is good enough for Nokia (well they have some patches), good enough for the familiar and zaurus users Aug 02 17:36:09 jg_: kdrive support too, with calibration Aug 02 17:36:40 kergoth: you really should do a tslib release ;) Aug 02 17:36:57 mallum: hehe Aug 02 17:37:24 mallum: using full up X.org release on OLPC. And technically, we don't have a touch screen, but a new dual mode touchpad/tablet device. Aug 02 17:37:26 zecke|food: nobody knows about it / trusts it as cvs only :( Aug 02 17:38:02 mallum: no release before its time: tslib looks last touched a year or two ago in CVS. Aug 02 17:38:23 jg_: but it is in heavy use Aug 02 17:38:24 jg_: right, theres probably been no need Aug 02 17:38:52 jg_: custom XInput driver then maybe ? Aug 02 17:38:53 not after fixing 59 occurences of the unsigned char problem a while back Aug 02 17:39:17 mallum: yes, on evdev, rather than antique brokenness of X.org input nonesense. Aug 02 17:39:24 I remember someone made a patch to add tslib support to Xorg Aug 02 17:39:39 mallum: warp is doing the work for us under contract. Aug 02 17:40:22 jg_: the driver ? Aug 02 17:41:03 mallum: the driver both kernel and X. Aug 02 17:41:16 jg_: there is an evdev touchscreen driver for X Aug 02 17:42:08 warp's been trying to drag X into the modern world of evdev; this is a way to help that happen. The old X.org stuff is so broken, I want nothing to do with it. Aug 02 17:42:14 jg_: http://opensource.idealcorp.com/evdev/ Aug 02 17:44:19 jg_: do you have a url for warp ? Aug 02 17:44:45 warp at debian.org is one of his mail addresses Aug 02 17:45:05 His name is Zephaniah Hull. Aug 02 17:46:42 zecke I am in the mood for nfs u2 Aug 02 17:46:45 icculus guy ? Aug 02 17:47:13 woglinde: get a copy, come here to play Aug 02 17:47:23 sorry cant Aug 02 17:47:34 maybee sunday? Aug 02 17:48:04 woglinde: any day but wednesday or thursday as I will be in munich again Aug 02 17:48:12 hm http://www.linuxdevices.com/news/NS4539563903.html Aug 02 17:48:20 next week? Aug 02 17:48:25 yes Aug 02 17:48:35 okay so whats with sunday? Aug 02 17:49:23 ouch the phone has tv-out Aug 02 17:49:50 woglinde: get a copy, come to my flat whenever you want Aug 02 17:50:01 your flat? Aug 02 17:50:07 did I missed something? Aug 02 17:50:21 woglinde: hehe, flat, house, hat, room whatever Aug 02 17:50:34 woglinde: steve gave me another shark Aug 02 17:50:43 now you have a cluster? Aug 02 17:51:01 2.4 kernel only that suckz Aug 02 17:51:20 woglinde: mizi? Aug 02 17:51:23 jo Aug 02 17:52:47 woglinde: alternate is sloow. I wait for my IDE disk... Aug 02 17:56:02 hm why did you order at alternate? Aug 02 17:56:13 mindfactory is cheaper and faster Aug 02 18:16:56 "morning" Aug 02 18:17:11 moin mickeyl Aug 02 18:26:19 ~lart mesa Aug 02 18:26:20 * ibot urinates on mesa Aug 02 18:26:26 ibot: botsnack Aug 02 18:26:27 aw, gee, koen Aug 02 18:30:18 hi all Aug 02 18:31:26 hey mr_nice_slacker Aug 02 18:31:38 hi koen Aug 02 18:32:17 any one know a good chanel for gcc first stage cross compiler building problems (chicken and egg problem with libc) ? Aug 02 18:32:45 OE builds them for you :) Aug 02 18:33:14 yes, but I want to ger able to build them myself ;) Aug 02 18:33:38 why? Aug 02 18:33:50 there's a reason why people started OE Aug 02 18:34:45 Just for technical studies - or something like that Aug 02 18:35:12 anyone built diet-x11 lately? Aug 02 18:35:15 mr_nice_slacker: search developerworks of IBM Aug 02 18:35:21 it doesn't find its patches :/ Aug 02 18:35:37 ~lart people commiting moves and renames without testing Aug 02 18:35:38 * ibot squeezes people commiting moves and renames without testing till people commiting moves and renames without testing turns blue like papa smurf Aug 02 18:35:49 Well this document brought me to that chicken egg problem .. Aug 02 18:35:58 any idea why diet-x11 wouldn't find xextensions although it has been built? Aug 02 18:36:04 mr_nice why? Aug 02 18:36:47 it uses --with-newlib to ship arround but it brings me to crto stuff Aug 02 18:39:15 mickeyl: the old diet-x11 is missing its patches iirc Aug 02 18:39:26 mickeyl: see bugzilla Aug 02 18:39:49 mr_nice_slacker: It is the chicken and egg problem Aug 02 18:40:11 mr_nice_slacker: see ghc as another incarnation of the problem Aug 02 18:40:28 I never ran into a chichen/egg problem when building a cross-compiler... Aug 02 18:41:00 JoeSchmo: please tell me the secret ;) Aug 02 18:41:12 trac.cross-lfs.org :) Aug 02 18:41:56 dev == canadian cross, 2.0 == sysroot style (very young) Aug 02 18:42:13 joeschmo nono you dont understand Aug 02 18:42:22 guess not Aug 02 18:42:39 mr_nice problem is, where gcc firs or libc first Aug 02 18:43:16 linux-headers, binutils, glibc-headers, gcc-static, glibc, gcc-final Aug 02 18:43:23 that's my standard build order Aug 02 18:43:35 nono Aug 02 18:43:37 not now Aug 02 18:43:41 back in the old times Aug 02 18:43:50 where stallman has to leave mit Aug 02 18:43:52 when Aug 02 18:45:22 koen: I've readded the patches, but it doesn't find xextensions now. Aug 02 18:45:26 * mickeyl checks bugtracker Aug 02 18:45:33 where's ciabot btw? Aug 02 18:46:07 mickeyl: 'screen -Ux' Aug 02 18:46:22 Cannot open your terminal '/dev/pts/0' - please check. Aug 02 18:46:39 did you use sudo or something? Aug 02 18:46:42 ya Aug 02 18:46:49 don't :) Aug 02 18:46:51 hmm, k Aug 02 18:47:14 ~/ciabot would be my guess Aug 02 18:47:36 djeez Aug 02 18:47:41 thx anyone - I will start to read a bit more documentation now Aug 02 18:48:13 xserver-xorg insist on doing a cross-compile unsafe check for building does even if you tell to *not* build docs Aug 02 18:48:21 who wrote crap like that? Aug 02 18:48:43 ah Aug 02 18:48:45 dnl Copyright © 2003-2005 Keith Packard, Daniel Ston Aug 02 18:48:53 uh Aug 02 18:49:00 they should know Aug 02 18:49:04 jesus Aug 02 18:49:53 and you can't remove that extra test, since that's the one used to *actually* enable/disable docs Aug 02 18:51:38 koen: send me a patch and I'll commit it Aug 02 18:51:44 03xora 07org.oe.dev * r5f3f04b4... 10/ (8 files in 2 dirs): packages/sylpheed/* : drop old versions Aug 02 18:51:56 03xora 07org.oe.dev * rec203c2d... 10/ (4 files in 2 dirs): Aug 02 18:51:56 sylpheed-claws-plugin-gtkhtml2-viewer-0.10.2_2.4.0.bb Aug 02 18:51:56 sylpheed-claws-plugin-maildir-0.24.1_2.4.0.bb Aug 02 18:51:56 sylpheed-claws-plugin-rssyl-0.6_2.4.0.bb Aug 02 18:51:56 New versions for newer 2.4.0 sylpheed-claws Aug 02 18:51:58 mallum: still trying to figure out how to solve it Aug 02 18:52:07 03mickeyl 07org.oe.dev * ra5338f1e... 10/ (12 files in 3 dirs): distro configurations: PREFERRED_PROVIDER_libx11 -> PREFERRED_PROVIDER_virtual/libx11 Aug 02 18:52:10 03mickeyl 07org.oe.dev * r82b524f1... 10/ (4 files in 3 dirs): diet-x11: move/copy patches to make it compile again Aug 02 18:52:21 which diet-x11 is recommended to build nowadays? Aug 02 18:52:39 mickeyl: the X11R7 one Aug 02 18:52:50 ah. now to find out why 6.2.1 is chosen per defaul Aug 02 18:54:59 that brings me back to that bitbake feature wish ... Aug 02 18:55:01 reverse lookup Aug 02 18:55:06 of variables Aug 02 18:55:48 mallum: the XORG_CHECK_LINUXDOC macro is causing the grief Aug 02 18:56:07 aha Aug 02 18:56:10 mallum: AM_CONDITIONAL(BUILDDOCS, test "x$BUILDDOCS" = xyes) should take care of that Aug 02 18:56:13 preferred-gpe-versions-2.6.inc:PREFERRED_VERSION_diet-x11 ?= "6.2.1" Aug 02 18:56:13 preferred-gpe-versions-2.7.inc:PREFERRED_VERSION_diet-x11 ?= "6.2.1" Aug 02 18:56:13 preferred-gpe-versions.inc:PREFERRED_VERSION_diet-x11 ?= "6.2.1" Aug 02 18:56:25 is that still correct? Aug 02 18:56:38 mickeyl: for old style x, yes Aug 02 18:57:05 unsupported and obsolete X, that is Aug 02 18:57:13 aha. who will feel responsible for bringing it up to the new generation Aug 02 18:57:14 ? Aug 02 18:57:26 we now have buggy and might-be-supported 7.1 now in .dev Aug 02 18:57:38 shouldn't we remove 6.2.1 completely if it's unsupported and obsolete? Aug 02 18:57:42 mickeyl: gpe 2.6 and 2.7 are already released Aug 02 18:57:58 I don't feel like doing a stalinistic history tampering Aug 02 18:58:16 me neither, but then we should at least try to keep the old versions working Aug 02 18:58:23 yeah Aug 02 18:58:38 wanna help out with getting X into shape? Aug 02 18:58:56 R6, R7 and R7.1 are all a bit messy in .dev Aug 02 18:59:07 not if I can keep out of it. I'm busy enough with all the Qt/E, Opie and Python stuff Aug 02 18:59:40 i had hoped there were enough X lovers around Aug 02 19:00:05 they aren't building from .dev Aug 02 19:00:09 in fact the reason why I'm stumbling over that is because python depends on libx11 Aug 02 19:00:17 the gpe people are using .oz354x and o-hand has poky Aug 02 19:00:18 and i was trying to fix some python things Aug 02 19:00:30 mickeyl why core python depends on x? Aug 02 19:00:33 dont get it Aug 02 19:00:37 easy Aug 02 19:00:41 because python builds tkinter Aug 02 19:00:47 which relies on tk Aug 02 19:00:49 which relies on libx11 Aug 02 19:01:06 tk and tcl is ugly and somewhat dead Aug 02 19:01:28 sure other people dont belevie this and using it Aug 02 19:01:43 koen: that's exactly why I hate branches in general. do you think there is a way to drag some of the gpe guys back into .dev? Aug 02 19:02:00 small groups just can't afford branches Aug 02 19:02:01 :/ Aug 02 19:02:12 mickeyl cant you disable tkinter? Aug 02 19:02:15 s/branches/long lived branches/ Aug 02 19:02:40 mickeyl: I think OE would benefit greatly from small feature based branches that will be merged back Aug 02 19:02:51 koen: do you just want to add a if BUILDDOC DOC_DIR=dir endif to toplevel Makefile.am ? Aug 02 19:02:58 mickeyl: instead of the commit-break-fix-commit we are doing now Aug 02 19:03:01 woglinde: i can, but X not building bothers me because I tend to feel responsible for the OE quality in general. If we don't have a simple libx11 working in .dev, then OE simply sucks. Aug 02 19:03:24 mickeyl okay Aug 02 19:03:59 mallum: that wouldn't help configure from bailing out, right? Aug 02 19:04:02 *sigh* if we just had those crazy pdaXrom guys here *sigh* Aug 02 19:04:06 ~lart crosstool Aug 02 19:04:07 * ibot stuffs crosstool into a shiny new tin can and vacuum seals it Aug 02 19:04:07 ~lart crosstool again Aug 02 19:04:08 * ibot sends a legion of lawyers after crosstool again's head Aug 02 19:04:19 mickeyl: pdaX is still using Xfree86 Aug 02 19:04:28 ok, so those wouldn't help in that case? Aug 02 19:04:30 mickeyl: the X situation would actually be worse L( Aug 02 19:04:38 s/L/:/ Aug 02 19:04:51 i see Aug 02 19:05:10 mallum: AC_DEFUN([XORG_CHECK_LINUXDOC],[AC_CHECK_FILE( [$prefix/share/X11/sgml/defs.ent], Aug 02 19:05:16 mallum: that's the problem Aug 02 19:06:32 oh great. pkg-config SIGSEGVs here Aug 02 19:07:00 guess it's one of those days where I won't get anywhere Aug 02 19:07:05 * mickeyl gives up for now Aug 02 19:07:06 l8er Aug 02 19:07:13 night mickeyl Aug 02 19:07:19 mallum: would your makefile.am patch 'fix' stuff if I just remove XORG_CHECK_LINUXDOC from configure.ac? Aug 02 19:07:45 mickey|tv: DISTRO=angstrom-2006.9 solves a lot of the problems you are seeing Aug 02 19:09:11 mickey|tv: latwe Aug 02 19:09:30 koen: ah its XORG_CHECK_LINUXDOC breaking ? Aug 02 19:09:38 koen: let us replace fink and darwinports Aug 02 19:09:41 mallum: yes Aug 02 19:10:15 koen: wrap XORG_CHECK_LINUXDOC if a check too Aug 02 19:11:22 koen: hmm actually there are docs all over the place :/ Aug 02 19:11:50 mallum: hence 'who wrote this crap?' Aug 02 19:12:51 koen: though that mostly seems to have the write checks Aug 02 19:13:15 koen: so maybe just wrapping XORG_CHECK_LINUXDOC with if builddocs is enough Aug 02 19:14:54 mallum: that should have the same effect as removing it, right? Aug 02 19:14:57 (just to test) Aug 02 19:15:25 in theory Aug 02 19:19:35 mallum: that give me this: http://ewi546.ewi.utwente.nl/tmp/xorg-breakage.txt Aug 02 19:20:16 koen: oh fuckaduck Aug 02 19:20:55 koen hm more doc-tags to delete Aug 02 19:21:41 koen: I think XORG_CHECK_LINUXDOC needs fixing Aug 02 19:22:06 woglinde: the bright side is that xserver-xorg uses xorg-macros.m4 instead of duplicating it in aclocal.m4 Aug 02 19:22:14 (like libxfixes and others do) Aug 02 19:24:04 koen: fink sucks soo much Aug 02 19:25:06 zecke|food: yes Aug 02 19:25:15 zecke|food: but darwinports sucks more Aug 02 19:25:25 koen: right, let us make OE suck less Aug 02 19:26:13 bleh Aug 02 19:26:21 koen: ? Aug 02 19:26:30 and I only want to build xorg to cross check a bug in kdrive Aug 02 19:26:39 hehe Aug 02 19:32:34 * koen gives up for today Aug 02 19:32:43 koen: I have found a monkey to convert the remaining wiki pages Aug 02 19:32:47 too much autofoo and gdb Aug 02 19:32:53 zecke: ah, cool Aug 02 19:32:58 koen: I will show him what to do next week in munich Aug 02 19:33:20 zecke: I was secretly happy with the decrufting Aug 02 19:34:28 koen: well, either we do a selection of the files (work on your site) Aug 02 19:34:47 koen: or we ask the guy to mark them as old (if drupal permits) Aug 02 19:35:52 old -> unpublish them Aug 02 19:36:14 oh I may push my branches again :) Aug 02 19:37:19 zecke: does svn have some form of a --missing switch? Aug 02 19:37:35 svn status? Aug 02 19:37:49 + sed + cut +tr Aug 02 19:37:57 zecke: 'svn add --missing' and 'svn delete --missing' Aug 02 19:38:02 ahh Aug 02 19:38:12 koen: dunno, don't waste your time Aug 02 19:38:19 koen: point people to tailor to check how to do it Aug 02 19:38:49 zecke: mickeyl is trying to hook up the svn mirror to ciabot Aug 02 19:46:12 wb Aug 02 19:57:45 * zecke throws fink into the garbage... Aug 02 20:50:59 Hey whats the maximum root image size I can upload to my C1000? Aug 02 20:51:10 Zaurus Aug 02 21:30:43 Is anybody in here? Aug 02 21:30:59 yes Aug 02 21:31:26 do you use openzaurus or OE? Aug 02 21:32:00 nope Aug 02 21:32:05 aeh I use OE Aug 02 21:32:06 dtx: wrong question Aug 02 21:32:09 but not opnezaurus Aug 02 21:32:11 sorry Aug 02 21:32:26 dtx: OpenZaurus uses OE to create the distribution Aug 02 21:32:31 sorry Aug 02 21:32:34 I knew that Aug 02 21:32:36 misphrasing Aug 02 21:32:55 I was just wondering if anyone was familiar with building OE for openzaurus Aug 02 21:33:08 more specifically if they knew what the maximum initrd.bin size is Aug 02 21:33:14 for a Zaurus C1000 Aug 02 21:33:21 OZ uses OE to build ipks and images Aug 02 21:33:28 you don't "build OE" Aug 02 21:33:37 and yes, many of us are familiar with it Aug 02 21:34:02 not meaning to get into semantics, but you have a .Net type of an issue with the OpenEmbedded name Aug 02 21:34:22 ? Aug 02 21:34:28 .net? Aug 02 21:34:30 I understand OpenEmbedded supplies the facilities to build OpenZaurus and familiar and a bunch of other stuff Aug 02 21:34:37 ;-) Aug 02 21:34:40 ok Aug 02 21:34:40 well Aug 02 21:34:43 try asking your question Aug 02 21:34:49 justin he has Aug 02 21:35:10 dtx: sorry I don't know the max size of init.rd Aug 02 21:35:11 what the max initrd.bin size is for sharp zaurus c1000 Aug 02 21:35:27 or maybee how he can changed i Aug 02 21:35:28 t Aug 02 21:39:52 ah...heh, sorry, yes he did Aug 02 21:40:04 dtx: it's in the conf/machine/spitz.conf file IIRC Aug 02 22:22:33 hello Aug 02 22:23:13 are there any IDEs that can drive/manage the bitbake/openembedded build system and metadata? Aug 02 22:24:03 have followed the oe gettingstarted guide, but i get the following type of errors: "Unable to generate local paths for SRC_URI due to malformed uri" when running bitbake on any package Aug 02 22:24:11 am i missing something? Aug 02 22:24:33 that's odd. try running with -D or -vvv? Aug 02 22:24:56 i.e. see if you can coax more information out of it Aug 02 22:26:39 bitbake doesn't need to be run as root, does it? Aug 02 22:26:42 will try now Aug 02 22:26:45 also, i've found the bitbake shell's pastebin functionaility to be useful; run 'shell bitbake ' (where the id can be determined from buffers) Aug 02 22:26:54 no, you don't need root Aug 02 22:28:16 anyway, the example above is how one can use bitbake to trivial share the output from a problem with bitbake; as long as the shell functionality works, other bugs in bitbake can still be reported with bitbake :) Aug 02 22:30:05 running `bitbake -vvv nano`, first error is "ERROR: dictionary changed size during iteration while parsing ..." Aug 02 22:36:18 furlongm: perhaps you didn't set up yout BBPATH correctly? Aug 02 22:37:59 hmm Aug 02 22:38:38 ~/projects/oe/build $ echo $BBPATH Aug 02 22:38:38 /home/furlongm/projects/oe/build:/home/furlongm/projects/oe/org.openembedded.dev Aug 02 22:39:33 seems ok to me Aug 02 22:39:48 is OEPATH needed? (found it on google) Aug 02 22:46:49 * mr_nice_slacker Prises the Elfes - Elfes are nice little creatures which cares about the suffered Cross Compiler newbies ;) Aug 02 22:47:12 * furlongm finds a symlink he made earlier from conf/local.conf to conf/bitbake.conf and solved the mystery Aug 02 22:51:33 * reenoo heads off to bed Aug 02 22:51:36 'night all Aug 02 22:55:22 furlongm: yep, that would be it Aug 02 22:55:31 furlongm: (no OEPATH, that's very old) Aug 03 00:59:40 dtx: if you're running out of space on boot you need to remove some programs from your image, yes Aug 03 01:02:02 What is the difference between .summary.img and .img? Aug 03 01:02:11 no idea Aug 03 01:02:19 hmmm Aug 03 01:02:22 never seen a .summary.img AFAIK Aug 03 01:02:39 I'm building on the .dev branch I don't know how different that is Aug 03 01:03:07 that's why you're getting different results Aug 03 01:03:17 OZ is released from release branches Aug 03 01:03:21 right now it's .oz354x Aug 03 01:03:26 When I build an image it puts three files in the deploy/imaqes folder: the .img, .summary.img (which is about 10MB larger than .img), and the .tar.bz2 Aug 03 01:03:35 * JustinP shrugs Aug 03 01:03:50 you can always try the tarball and altboot to test Aug 03 01:03:51 on an SD card or something Aug 03 01:04:21 altboot? Aug 03 01:05:01 So what files does it build for yours? Aug 03 01:05:06 when you build an image Aug 03 01:06:46 And are the .dev and .oz354x branches really that different? Aug 03 01:12:39 mmm hmm... **** ENDING LOGGING AT Thu Aug 03 02:59:57 2006