**** BEGIN LOGGING AT Thu Mar 09 10:59:57 2006 Mar 09 11:03:39 sharp-aticore-oss does not like make 3.81 ;( Mar 09 11:03:45 good morning Mar 09 11:07:12 file-native is failing to compile on my system. Mar 09 11:07:30 | lt-file: could not find any magic files! Mar 09 11:07:30 | make[2]: *** [magic.mgc] Error 255 Mar 09 11:07:39 r1? Mar 09 11:08:04 ashak: did you update your tree? Mar 09 11:08:12 hrw|work: it seems that I just fumbled myself. Compiling and linking was done against the correct "special" libraries, but since the nano was built with dynamic linking, on running ldd it showed bindings to my host libs because the "special" libraries were not visible. Running ldconfig and tuning ld.so.conf would change that Mar 09 11:08:17 katossi: about an hour ago Mar 09 11:08:52 hrw|work: file-native-4.16-r1 Mar 09 11:09:22 argh. Mar 09 11:10:21 ashak: your tmp stuff must probably be clean, hrw said before how to do it Mar 09 11:10:43 katossi: which parts? Mar 09 11:10:59 or is is simpler to nuke the whole lot? Mar 09 11:11:05 ashak: the ones related to 'file' Mar 09 11:11:35 ashak: cleaning everything obviously will work, but you will start everything from scratch then Mar 09 11:11:47 tmp/work/i686-linux/file-native ? Mar 09 11:12:22 tmp/staging/HOST_SYS/bin/file Mar 09 11:13:02 * ashak tries again then... Mar 09 11:13:40 stuff like this, is there anywhere it is mentioned other than in here? Mar 09 11:13:41 who can help me understand why opie-image does not even try to build opie packages which it has to install? up-to-date OE .dev, up-to-date bitbake Mar 09 11:14:25 hrw|work: I only had that problem when I wasn't using the svn bitbake on my gentoo system. Mar 09 11:14:59 god I wish I understood this modern glibc stuff, like why it add __THROW to all function calls when included from c++ Mar 09 11:21:33 hrw|work: may I suggest a default file for /etc/default/opd_args so that bluetooth file reception works (even if it isn't integrated with opie yet)? Mar 09 11:22:27 I've just been fighting it, and got it to work without much trouble Mar 09 11:23:00 now if we had a "obex incomming folder" it would be even nicer Mar 09 11:23:33 Bah, file-native is still failing Mar 09 11:31:57 Bernardo: you know - patches are welcome and bugzilla is working Mar 09 11:32:19 Bernardo: I'm nearly two months back when it comes to .dev Mar 09 11:35:27 hrw|work, I found the problem with linux-libc-headers -- my TARGET_ARCH was ppc, leading to an undefined ARCH variable, which caused the script to stop Mar 09 11:36:33 koobla: ppc<>powerpc naming problem? Mar 09 11:37:16 right Mar 09 11:37:24 hrw|work: ok, I've sent to the list but I do need to start using bugzilla... Mar 09 11:48:10 hrw|work: just added to bugzilla Mar 09 11:55:12 hi lardman Mar 09 11:55:52 hi hrw|work, how's things? Mar 09 11:58:53 lardman: not bad Mar 09 11:59:29 lardman: I reserved half of day during weekend for oz release (if it will be build in that time) Mar 09 12:00:17 hrw|work: Cool, I'll flash it then and prepare for user support on the forums ;) Mar 09 12:00:24 lardman: ;) Mar 09 12:01:29 and I have problem with my lips - weather ;( Mar 09 12:02:08 it's strangely warm here (tho' raining as standard), Spring must have started Mar 09 12:03:26 here it has to snow in next days a bit Mar 09 12:04:13 koen: do you know where angstrom homepage will be hosted? Mar 09 12:04:44 hrw|work: yes Mar 09 12:05:07 I feel pain each time when I have to go to openzaurus.org... Mar 09 12:05:15 ~kill sf.net project webserver Mar 09 12:05:17 * ibot shoots a excited meson gun at sf.net project webserver Mar 09 12:11:14 koen: now we know why it isn't sf.net ;-) Mar 09 12:12:44 I feel that bitbake/OE.dev is broken in nasty way... Mar 09 12:13:15 build opie-image and let it fail on one of depends.. next run will ignore that fail and skip all depends Mar 09 12:14:01 and then s/opie-image/meta-opie/ and you will get moment where it will not take care od buildtime depends Mar 09 12:15:45 any ideas? Mar 09 12:18:05 morning Mar 09 12:19:19 hi Rene Mar 09 12:19:31 hi hrw|work Mar 09 12:23:21 hey reenoo Mar 09 12:23:33 hrw|work: do you want me to build anything? Mar 09 12:23:41 I'll have my inlaws here, but... Mar 09 12:23:57 hi koen Mar 09 12:24:09 I already left my machine building meta-opie, iqnotes, prboom and opie-kdepim-image for the akita Mar 09 12:24:24 Bernardo: no - I would rather want why that happen Mar 09 12:24:49 ? Mar 09 12:24:56 reenoo: you latest change to glibc seems to work on my config Mar 09 12:24:59 Bernardo: I probably can build that all by hand - but want to know why automatic way does not work Mar 09 12:25:01 reenoo: thanks for fixing it Mar 09 12:25:34 ~lart recent OE (.dev/bitbake) for breaking Mar 09 12:25:34 * ibot puts recent OE (.dev/bitbake) through a wood chipper for breaking Mar 09 12:27:17 hrw|work: I meant for the launch Mar 09 12:27:35 Bernardo: mickeyl machine is working on building Mar 09 12:27:49 ok, if you need anything just tell me Mar 09 12:27:54 sure Mar 09 12:27:56 koen: np Mar 09 12:28:13 Bernardo: currently ewi is enough for me if I have to build something Mar 09 12:29:55 koen: I'm about to fix the h3900 tune-*.conf situation and would like it to default to tune-strongarm. would you mind adding a variable to angstrom.conf to override that? Mar 09 12:30:35 (H3900_TUNING = "xscale" to be precise) Mar 09 12:31:22 I wouldn't mind Mar 09 12:31:23 TUNING_h3900 = "xscale" maybe? Mar 09 12:31:48 so machine override comes for free Mar 09 12:31:49 what is that supposed to do? use some custom flags for gcc? Mar 09 12:32:00 ~lart EABI for being armv5 only Mar 09 12:32:01 * ibot lowers EABI's priority for being armv5 only Mar 09 12:32:06 katossi: look at conf/machine/includes/tune-* Mar 09 12:32:12 hrw|work: no. think of the H3900_ as a namespace prefix. Mar 09 12:32:13 aha Mar 09 12:32:23 reenoo: ok Mar 09 12:33:44 reenoo: how will this be handles in other armv5 machines? Mar 09 12:34:10 hrw|work: ok it's as I thought, but, is it really worth of it? what's the real gain? Mar 09 12:34:21 koen: dunno. I guess they could adopt the same strategy Mar 09 12:34:45 wouldn't it be smarter to propose it to oe@ and implement it in a day or to? Mar 09 12:38:34 katossi: depends Mar 09 12:38:57 katossi: rumours say that xscale gain few % of speed Mar 09 12:38:59 koen: the amount of time I have these days is limited and I'm just adding a variable... Mar 09 12:39:17 hrw|work: 10% is ok to me, 1% is worthless Mar 09 12:39:47 reenoo: I'm sort of abusing oe@ for documentation purposes Mar 09 12:40:56 koen: if you look at pH5's recent post you'll see how people forget what's discussed/announced there very soon... Mar 09 12:41:41 true Mar 09 12:42:06 hrw|work: this SD timeout problem, is it supposed to give an output in dmesg? Mar 09 12:42:12 koen: I'll comment on it in the commit message and forward that to oe@ Mar 09 12:42:24 reenoo: thanks Mar 09 12:42:29 XorA: no Mar 09 12:42:57 hrw|work: how do you know when you have hit the problem then? Mar 09 12:43:15 XorA: just test 'while true;do md5sum /dev/mmcblk0;done' - if results differ then you got it Mar 09 12:43:43 laptop for sale 500$ want it gone today. price includes shipping, case, wireless router. message me if interested on aim at ogd443 or msn at mcsltd2@hotmail.com Mar 09 12:43:45 XorA: http://marc.theaimsgroup.com/?l=linux-arm-kernel&m=114164723821285&w=2 Mar 09 12:44:01 hrw|work: hmmm, a mv to /dev/mmcblk0p1 has just entered D state for me Mar 09 12:44:22 XorA: dmesg full of i/o errors? Mar 09 12:44:38 highfi: shipping to any part of world in price? nice Mar 09 12:44:46 hrw|work: nope, hence why I was checking Mar 09 12:45:31 hrw|work: I suspect I have found why my root on SD crashed occasionally Mar 09 12:45:56 XorA: timeout problem? Mar 09 12:46:21 hrw|work: I have no errors to tell any form of problem Mar 09 12:46:41 koen: should I add H3900_TUNING = "xscale" to angstrom.conf so it's all in the same commit or do you want to do that? Mar 09 12:46:42 XorA: try that md5sum Mar 09 12:47:01 reenoo: go ahead and add it if you want Mar 09 12:47:15 koen: ok, will do. thanks Mar 09 12:47:24 XorA: it failed on any bad card - my 64M/256M, 1G from greentux, 256M transcend (oz user) Mar 09 12:47:50 hrw|work: hmm, md5sum immediately enters D state, mmc driver must be fscked Mar 09 12:48:03 XorA: dmesg gives? Mar 09 12:48:03 * XorA alt-sysreqs Mar 09 12:48:13 hrw|work: nothing Mar 09 12:49:24 strange Mar 09 12:49:43 XorA: so linux-oz r11 will not work for you probably Mar 09 12:50:39 hrw|work: this problem is older than r11 Mar 09 12:51:02 hrw|work: but with my / on SD it used to just lock up device, I moved / back to jffs2 and I can now see why I get lockups Mar 09 12:51:06 XorA: I know - but r11 got rmk patch instead of my old one Mar 09 12:51:11 hrw|work: its been bugging me for a while Mar 09 12:52:00 XorA: I first got it serious in December Mar 09 12:52:18 hrw|work: this afternoon I will try md5sum and see if this is same problem or a different mmc bug Mar 09 12:56:15 is there any known problem with gnome-vfs-dbus and samba > 3.1.14? why is that specific version required? Mar 09 13:12:46 hmm, I think there is some problem with the toolchain which I builded with OE. Mar 09 13:13:40 the problem is very strange. It always crashed like this : http://rafb.net/paste/results/I1u23266.html Mar 09 13:15:52 I rebuild a toolchain with crosstool. It is configured with "softfloat, gcc-3.4" . Then I build a busybox with static option. Mar 09 13:16:43 and It will not crash then. Mar 09 13:16:48 any idea ? Mar 09 13:17:08 leoncamel: arm? Mar 09 13:17:41 hrw|work, yes. Mar 09 13:17:54 leoncamel: checked linux-arm-kernel archives? Mar 09 13:18:18 hrw|work, ok. I will check it now . Mar 09 13:19:02 hrw|work, But I don't know which keyword should I use .. :( Mar 09 13:19:22 does cvs have a recursive add? Mar 09 13:19:43 koen: no Mar 09 13:19:48 hrm Mar 09 13:19:58 sure -- 'find -print0 | xargs -0 cvs add' ;-) Mar 09 13:20:13 (err... or maybe that does horrible things to existing CVS dirs.) Mar 09 13:20:16 leoncamel: kernel init hang Mar 09 13:20:30 hrw|work, :). OK. Mar 09 13:20:37 (like, end up with CVS/Entries files in the repo, or something equally horribly stupid.) Mar 09 13:21:01 njs: cvs will refuse to add that type of thing Mar 09 13:21:38 reenoo: I probably have the details wrong, I've mostly managed to suppress most of my cvs scar tissue these days. I remember _something_ like that... I wonder what. Mar 09 13:21:40 apart from the cvs add thing, I have a RO monotone-cvs mirror online :) Mar 09 13:22:13 who made gnome-vfs-dbus bb file? Mar 09 13:23:06 katossi: mt log? Mar 09 13:24:17 gnmoe-vfs-dbus: depend on an older samba to fix build Mar 09 13:24:19 koen Mar 09 13:24:49 katossi: so test with current samba Mar 09 13:24:58 katossi: Revision: ff8482dffaa2fde2376b39d9666e0598a8beae83 Mar 09 13:25:27 its over 0.5h old change Mar 09 13:25:33 s/h /y Mar 09 13:29:27 do13: hey Mar 09 13:29:39 hi zecke Mar 09 13:29:40 do13: where are your konq bb files again? Mar 09 13:29:55 hi zecke ;-) Mar 09 13:30:00 zecke: another bitbake/OE.dev problem found ;( Mar 09 13:30:29 zecke: build opie-image and let it fail on one of depends.. next run will ignore that fail and skip all depends Mar 09 13:30:31 hrw|work: that is not funny any more :( (for both of us) Mar 09 13:30:49 hrw|work: that one is new Mar 09 13:31:15 zecke: I know - but I was out of .dev for month(s) so more things are new for me Mar 09 13:31:41 zecke: here it was sharp-aticore-oss failed so qte dont build so no opie at all Mar 09 13:31:50 hrw|work: depends in DEPENDS or in RDEPENDS? Mar 09 13:32:20 zecke: opie-image rdepends Mar 09 13:35:54 hi zecke! Mar 09 13:37:43 koobla: found the issue? Mar 09 13:37:51 yes Mar 09 13:37:58 koobla: what was it? Mar 09 13:38:02 BBFILES contained 2 paths Mar 09 13:38:14 moved my packages into tree -> now it works Mar 09 13:38:28 so much for the symptoms :) Mar 09 13:38:31 BB>> getvar BBFILES Mar 09 13:38:31 /home/hrw/devel/oe/org.openembedded/packages/*/*.bb /home/hrw/devel/oe/hrw-packages/*/*.bb Mar 09 13:38:34 and it works Mar 09 13:38:41 hrw|work: hmm I wonder how to reproduce it Mar 09 13:38:43 hmmm. cured it for me at least Mar 09 13:39:09 i looked into the bb cache - the function names are corrupt Mar 09 13:39:17 zecke: install make 3.81 and build opie-image/c7x0 from scratch - it will fail on sharp-aticore-oss Mar 09 13:40:02 hrw|work: that would explain why OE suddenly built glibc 2.3.2 Mar 09 13:40:16 hrw|work: Are you seeing this problem under the bitbake shell or from the commandline? Mar 09 13:40:26 RP: shell Mar 09 13:40:31 cmdline Mar 09 13:40:38 will try cmdline now Mar 09 13:40:57 XorA: I did read the discussion about the mixer yesterday - I think its a bug somewhere in ASoC by the looks of it Mar 09 13:41:14 koen: You're seeing the same issue? Mar 09 13:41:17 RP: cool, just making sure you saw it when someone else complains Mar 09 13:41:26 RP: I did yesterday Mar 09 13:42:34 Next time someone sees the RDEPENDS issue I'd appreciate a -D -D -D -v output log from bitbake to try and track the problem down with... Mar 09 13:42:45 XorA: thanks :) Mar 09 13:42:54 RP: ok Mar 09 13:43:21 RP: Ive got a horrible workaround involving alsamixer and alt tab :-( Mar 09 13:43:51 also 'buffers' in bbshell is broken Mar 09 13:44:01 XorA: We'll get there eventually... Mar 09 13:44:02 it renumbers the buffers the wrong way Mar 09 13:44:18 * RP is clueless about bbshell Mar 09 13:45:40 Oh my Mar 09 13:46:09 I was able to build a working familiar & gpe for 3630 Mar 09 13:47:11 morning Mar 09 13:47:40 RP: hmm.. now it works.. Mar 09 13:50:34 http://pastebin.com/592549 <- cvs mirror script for OE Mar 09 13:55:22 koen: THX for http://www.oesf.org/forums/index.php?showtopic=18098&view=findpost&p=117517 Mar 09 13:57:22 http://ewi546.ewi.utwente.nl/tmp/hrw/mmc_plus_2gb.diff will go into 3.5.4 first upgrade... Mar 09 13:57:30 <_law_> hrw|work, removed file from staging dirs, but i always get the same error :-( Mar 09 13:58:14 argh.. Mar 09 13:58:35 I will not touch non-maintained stuff in next few weeks Mar 09 14:00:23 hey zecke_ Mar 09 14:05:00 re Mar 09 14:06:24 zecke_: http://pastebin.com/592549 Mar 09 14:06:36 hi all Mar 09 14:06:43 he do13_ Mar 09 14:06:49 hi dirk Mar 09 14:06:56 hey koen, Marcin Mar 09 14:07:41 hi do13 Mar 09 14:07:49 zecke told me, you are building embedded konqueror Mar 09 14:08:03 hey ulit1, yep Mar 09 14:08:06 do oyu have any recent bb files? Mar 09 14:08:16 hey uli Mar 09 14:08:21 hey zecke Mar 09 14:08:27 hey zecke Mar 09 14:09:16 RP: rmk timeout patch works better for you then our old version? Mar 09 14:09:34 hrw|work: It works just the same but is a proper fix Mar 09 14:09:50 do13: maybe, i should introduce myself :-) Mar 09 14:10:09 ulit1: hehe, good thinking Mar 09 14:10:10 RP: ok. I'm adding one fix for 3.5.4 c7x0 kernel now Mar 09 14:10:10 ulit1: sorry. I need some cleanups Mar 09 14:10:18 I am working for Road, The company that is sponsoring the port Mar 09 14:10:25 My Name is Uli Mar 09 14:10:36 Hey Uli :) Mar 09 14:10:51 And I am trying to get it built with bitbake Mar 09 14:11:22 ulit1: wait a sec Mar 09 14:12:43 mickey|dinner: ping Mar 09 14:12:57 do13_: was thar javascript error a compile time or runtime error? Mar 09 14:13:52 zecke_: runtime error Mar 09 14:14:08 did you get pcre to linbk cleanly? Mar 09 14:15:32 NOTE: package linux-openzaurus-2.6.14-rc1: started Mar 09 14:15:49 ulit1: with a trick. Try this one http://www.do13.de/oz/temp/konqueror-embedded_svn.bb Mar 09 14:16:36 03hrw 07org.oe.oz354fam083 * r2ebbef46... 10/packages/linux/ (2 files in 2 dirs): Mar 09 14:16:38 linux-openzaurus 2.6.14-rc1: [MMC] Support MMC version 4 cards Mar 09 14:16:40 Version 4 of the MMC specification increased the version number of the Mar 09 14:16:42 CID structure. None of the fields changed though so the only required Mar 09 14:16:44 change is adding '4' to the approved list. Mar 09 14:16:46 patch taken from 2.6.15 Mar 09 14:17:17 <_law_> NOTE: package file-native-4.16: completed :-) Mar 09 14:17:33 zecke_: On javascript pages konqe eats 99% cpu load and crashed after a minute Mar 09 14:17:42 do13_: thanks Mar 09 14:17:47 i'll give it a try Mar 09 14:18:00 zecke_: But on my pc with qvfb it works Mar 09 14:18:18 ulit1: np Mar 09 14:18:23 do13_: did you try in a qemu image yet? Mar 09 14:18:39 03pH5 07org.oe.dev * rea1850c8... 10/conf/distro/ (4 files): distro/*.conf: change PREFERRED_VERSION_x11 to PREFERRED_VERSION_libx11 Mar 09 14:18:44 ulit1: thanks to Richard we can create qemu images Mar 09 14:18:50 zecke_: not yet Mar 09 14:19:42 hrw|work: how did we sort out the file-native issue? Mar 09 14:19:53 zecke_: using just-build file Mar 09 14:20:08 zecke: we should extend qemu to emulate our peripherals Mar 09 14:20:57 zecke_: but it still fails for many people ;( Mar 09 14:21:05 I will not touch non-maintained stuff in next few weeks Mar 09 14:21:17 hrw|work: yes it still fails :( Mar 09 14:21:46 that explains why OE builds dbus 0.23 and glibmm 2.4 Mar 09 14:21:49 hrw|work: well, somehow it complains that no magic files were found :( Mar 09 14:23:20 RP: Does opie works with qemu yet? Mar 09 14:23:53 03pH5 07org.oe.dev * r235af9cf... 10/packages/avahi/avahi_0.6.9.bb: avahi: add 0.6.9 Mar 09 14:26:47 file-native compiled fine for me, after I totally removed the tmp dir :) Mar 09 14:28:11 03pH5 07org.oe.dev * ra8fc8d79... 10/packages/keylaunch/ (files/makefile-fix.patch keylaunch_2.0.7.bb): keylaunch-2.0.7: LDFLAGS Makefile fix Mar 09 14:32:22 03pH5 07org.oe.dev * reb48ce1d... 10/packages/login-manager/files/login-manager: login-manager: simplify get_valid_lms, fix for kbdless devices with only one dm installed Mar 09 14:34:31 do13, how did you get konqueror not to link against "-L/home/luckas/svn-slimfast/phone-oe-arm.clean/oe-build/tmp/staging/arm-linux/lib" :-) Mar 09 14:34:46 there is a major bug in the makefile Mar 09 14:35:12 which makes the response of "pcre-config" a dependency and nto only a linker flag Mar 09 14:35:46 ~lart win2k server and snmp agents Mar 09 14:35:46 * ibot whips out a sword and chops win2k server and snmp agents in half Mar 09 14:37:36 ulit1: PCRE_CONFIG="invalid" and the configure script use the default values. This is a nasty hack Mar 09 14:38:11 ulit1: btw. my bb-file contains a copy and paste error. I put a new one online Mar 09 14:38:36 do13: which caused my builds to link against the hosts libpcre :-( Mar 09 14:38:52 i'll give your bb a try with PCRE_CONFIG="invalid" Mar 09 14:43:28 hi ulit1 Mar 09 14:43:54 ulit1: Adding support for more arm cpus/peripherals to qemu would be good :) Mar 09 14:44:59 ulit1: NOTE: package konqueror-embedded-0.0+svn20060307-r0: task do_compile: completed Mar 09 14:45:16 :-) Mar 09 14:45:25 NOTE: package konqueror-embedded-20060309: completed Mar 09 14:45:25 Build statistics: Mar 09 14:45:32 :-) Mar 09 14:45:35 'mornin all Mar 09 14:45:40 let's put it on the device ... Mar 09 14:45:48 hey HopsNBarley Mar 09 14:45:57 hi koen! Mar 09 14:46:06 * koen wonders what the magic trick is to embed an png into a svg Mar 09 14:46:12 scp: ./konqueror-embedded_20060309-r0_armv5te.ipk: No space left on device Mar 09 14:46:13 :-( Mar 09 14:47:36 ulit1: you need a bigger flash :) Mar 09 14:49:16 ulit1: do you have seen problems with javascript? Mar 09 14:49:21 oh. thanks do13. i was already wondering how to solve the scp issue ;-) Mar 09 14:49:53 * ashak sulks, wanted to get myself a pda, some nice ipaq h5500's on ebay, but they're going for far too much money Mar 09 14:49:55 this is my first try on the arm platform. JS on x86 seems fine Mar 09 15:03:10 soon we will have a origami.conf :) Mar 09 15:05:16 koen|away, why would you embed a png into an svg? they are both vector graphics, you should be able to convert easily Mar 09 15:05:30 emte: png is bitmap not vector Mar 09 15:05:52 really? Mar 09 15:06:07 it was vector when i fist worked with it Mar 09 15:06:12 hi emte Mar 09 15:06:28 but that was ... 8/9 yrs ago ... Mar 09 15:06:44 whne macromedia first introduced it Mar 09 15:06:53 hey chouimat Mar 09 15:07:18 all line and color feild stuff Mar 09 15:07:35 for director and flash Mar 09 15:07:39 ...are you sure that is the same png? Mar 09 15:07:47 not anymore Mar 09 15:08:39 i wonder what happens if i load my old stuff now ... Mar 09 15:08:48 hmm, wikipedia says that "macromedia fireworks" uses PNG's extensibility to store more its own custom info in PNGs Mar 09 15:09:07 lol Mar 09 15:09:22 there i go i guess Mar 09 15:10:10 as far as you can trust wikipedia now that it was dubbed the most flawed referance availible Mar 09 15:12:34 Is there any way to setup vnc within gpe, like you can within Opie? Mar 09 15:14:08 ashak: install x11vnc Mar 09 15:15:41 ashak: you could use teleport? teleport the applications to your linux desktop Mar 09 15:17:04 Can someone tell me what the difference is exactly between IPKG_INSTALL, DEPENDS, and RDEPENDS. And which ones resolve their dependencies automatically? Mar 09 15:17:57 zecke: teleport? sounds fun... *investigates* Mar 09 15:18:59 ashak: I don't know if matchbox desktop is supporting it Mar 09 15:18:59 <[lala_]> do13_: are you planning to put your konq bb file into dev? Mar 09 15:19:18 [lala_]: as I judge do13, he tries to sell it to lineo first Mar 09 15:19:34 <[lala_]> hehe Mar 09 15:19:52 dtx: IPKG_INSTALL = packages which will be in image, DEPENDS = build depends, RDEPENDS = on-device depends (and they are autosolved for libs) Mar 09 15:20:27 [lala_]: lineo track his work on patches... Mar 09 15:20:57 have you seen the new Zaurus already? Mar 09 15:21:44 zecke_: slc3200? Mar 09 15:21:47 * france is back (gone 33:06:08) Mar 09 15:21:47 yes Mar 09 15:22:08 zecke_: sharp out of 4G uDrives so released new machine Mar 09 15:22:22 zecke_: changed colors to hide that it is same as before ;) Mar 09 15:24:53 * france is away: Away Mar 09 15:30:11 This sucks, I have no idea what i'm doing :) Mar 09 15:31:58 NOTE: package opie-image-1.0: completed Mar 09 15:32:01 finally Mar 09 15:51:37 http://www.iral.com/~albertr/linux/zaurus/wireless/ nice hacks to cxx00 Mar 09 15:53:02 [lala_]: If the last dependencies are solved I'll publish my bb- files Mar 09 15:54:14 off topic i know, but batteries, the zaurus batteries, is the lio-ion charge within the pack? or is this just the cell protection circuit? if so why the third terminal? does naybody know?l Mar 09 15:54:25 do13: please do. We do need a half-decent web browser Mar 09 15:54:51 do13_: we (opie users) will get very nice browser Mar 09 15:55:04 cu - english lesson Mar 09 15:55:07 hrw|work: yep. Mar 09 15:55:13 I widh to build n external charger for my spare Cells, im quite awra eof the lio-ion charge cycle and have a propper li-ion charger IC, but i am unsure about how the cicruit within the z battery packs works Mar 09 16:01:31 interesting, scope reveals its a serial comunications line! Mar 09 16:03:12 do13:thanks for your help. got konqui running on my device with the same problems you reported. Mar 09 16:04:42 i'll keep you updated on further improvements (try adding " -DENABLE_BOOKMARKS" to your *_extraflags for example) Mar 09 16:08:43 hi Mar 09 16:08:51 ulit1: thx Mar 09 16:12:53 ulit1: I hope we get an discount for your phone :) Mar 09 16:13:31 * Bernardo plans to buy a decent smartphone (with 2Mp photo cam) this year... any suggestions? Mar 09 16:16:14 * Bernardo just found out iqnotes is now on berlios... have to try tonight to update the bb to 2.1.0rc1 and see if it builds Mar 09 16:21:38 do13: :-) Mar 09 16:22:05 do13: does having in-berlin as a provider mean you are living in berlin? Mar 09 16:22:20 ulit1: no near frankfurt Mar 09 16:29:06 hrw|gone: someone is ill with mmc in -r11 :-( Mar 09 16:42:34 Bernardo: iqnotes 2.1.0rc1 is in OE iirc Mar 09 16:43:15 Bernardo: iqnotes_2.0.99-2.1.0rc1.bb (.dev) Mar 09 16:43:16 cu Mar 09 16:45:16 03koen 07org.oe.dev * rec8212ce... 10/packages/gpe-bootsplash/files/angstrom/splash-l.svg: gpe-bootsplash: start adding some Angstrom branding Mar 09 16:50:19 hrw|gone: thanks. That's what happens for only checking oz354... Mar 09 16:57:50 ERROR: Nothing provides runtime dependency diffstat-native <- is anyone else getting this? There is indeed no -native .bb and diffstat.bb doesn't build it either Mar 09 16:59:19 CoreDump|home: hey Mar 09 16:59:32 CoreDump|home: we now have an ASSUME_PROVIDED for required native packages Mar 09 16:59:40 err Mar 09 16:59:43 CoreDump|home: and bitbake.conf sets it, make sure you do not overwrite it Mar 09 17:00:04 I didn't touch bitbake.conf and svn upped yesterday Mar 09 17:00:18 CoreDump|home: sorry for my phrasing, Mar 09 17:00:29 CoreDump|home: "Du überschreibst ASSUME_PROVIDED!" Mar 09 17:00:35 ahh Mar 09 17:00:45 probably in my local.conf Mar 09 17:00:45 CoreDump|home: You need to do ASSUME_PROVIDE += instead of '=' Mar 09 17:00:51 * CoreDump|home checks Mar 09 17:01:53 zecke: thanks =) local.conf indeed overwrote that var Mar 09 17:02:28 * CoreDump|home patiently waits another 10 minutes to reparse Mar 09 17:02:54 *food* Mar 09 17:04:15 ~lart nfs-kernel-server Mar 09 17:04:15 * ibot does a little 'dpkg -P nfs-kernel-server' action Mar 09 17:04:26 heh how fitting Mar 09 17:07:18 zecke: I have a stupid question about the ASSUME_PROVIDED thingy. Wasn't the whole point of OE to have a unified devel environment which is self-contained? Ie: it compiles its own buildroot and (almost) all dependencies required for crosscompilation. Mar 09 17:07:50 Using ASSUME_PROVIDED kinda defeats this purpose Mar 09 17:08:10 well, you have to stop at some point Mar 09 17:08:32 CoreDump|home, it does, but it does help if you are too lazy to figure out why a native build desn't work for you ... Mar 09 17:09:32 CoreDump|home: i.e. compiling a compiler without a compiler won't work. the same goes for patching the sources of patch without patch. etc. etc. Mar 09 17:09:43 I do understand that it would be quite some work to include the "-native" stuff but I always "assumed" (hehe) that this was a sortf of principle of OE Mar 09 17:09:54 * CoreDump|home is not complaining Mar 09 17:10:25 CoreDump|home: yes true. But how do you compile a compiler? Mar 09 17:10:36 CoreDump|home: well you need one to compile -> ASSUME_PROVIDED Mar 09 17:10:45 CoreDump|home: how to run bitbake -> python is present Mar 09 17:10:54 you don't, at a very early stage you have to use pre-installed stuff of course Mar 09 17:11:12 CoreDump|home: quilt needs patch as RDEPENBDS, but do build quilt we need to patch it Mar 09 17:11:22 CoreDump|home: patch -> ASSUME_PROVIDED Mar 09 17:11:44 CoreDump|home: the minimal set of absolute requirements will land in ASSUME_PROVIDED Mar 09 17:11:56 zecke: ah ok Mar 09 17:12:41 CoreDump|home: the only bigger package sin ASSUME_PROVIDED are python and perl Mar 09 17:16:34 thanks for clearing this up =) Mar 09 17:22:02 CoreDump|home: do you have an idea on how to document that better? Mar 09 17:22:56 as the change was very recent, maybe add it to /topic? Mar 09 17:23:35 or add a "Important Notes" features to BB which prints out this kind of stuff on the first run after an update =) Mar 09 17:28:19 * zecke ordered a NSLU2 today Mar 09 17:31:29 zecke: nice! Mar 09 17:39:30 zecke: I have a sanity.conf and sanity.bbclass which detects the ASSUME_PROVIDED error now :) Mar 09 17:41:02 RP: you check if the know bits are still present? Mar 09 17:41:11 RP: you might do require instead of include? Mar 09 17:41:49 zecke: you mean for sanity.conf? Mar 09 17:42:06 yes (don't tell me you do that already) Mar 09 17:42:30 its an include at the moment but a require would be fine Mar 09 17:42:44 as long as people can override (which they can) Mar 09 17:43:02 zecke: It also checks for a minimum bitbake version Mar 09 17:43:58 zecke: http://www.rpsys.net/openzaurus/temp/oe-sanity.patch Mar 09 17:44:19 I posted to oe@ about including it. I'm tempted to just push it... Mar 09 17:44:44 the bikeshed should be yellow Mar 09 17:44:46 RP: I have seen it, use require conf/sanity.conf Mar 09 17:44:51 gah Mar 09 17:44:56 ~lart nickserv Mar 09 17:44:56 * ibot chops nickserv in half with a free Solaris 7 CD Mar 09 17:46:51 RP: you could just require all files Mar 09 17:47:11 RP: and kergoth can just use touch (once) Mar 09 17:50:04 zecke: I don't see the point in requiring all files. Running without a distro set is valid... But I agree sanity should be required which will catch a different set of misconfigurations :) Mar 09 19:01:27 CoreDump|busy: did file-native build for you? Mar 09 19:01:56 CoreDump|busy: for me opie-image, meta-opie, meta-opie-all are building, meta-gpe fails on sylpheed Mar 09 19:03:02 zecke: 0.8.3? I just today built sylpheed 2.2.2 in .dev Mar 09 19:03:15 hi zecke Mar 09 19:03:45 pH5: 1.9.12 :} (whatever my config is) Mar 09 19:04:00 pH5: it fails to link, see #730 Mar 09 19:04:44 RP: I have one question for your sanity.bbclass, how often is the quick return used? Mar 09 19:08:25 good evening Mar 09 19:08:31 evening mickeyl Mar 09 19:08:44 hi reenoo Mar 09 19:10:59 hmm, whats changed in the .bb format since i was last involved? addition of 'require', what else? Mar 09 19:11:04 * kergoth updating his vim syntax highlight file :P Mar 09 19:11:29 hi mickeyl Mar 09 19:11:30 hi kergoth Mar 09 19:11:37 kergoth: see the ChangeLog :} Mar 09 19:11:40 hey gang. i'm not looking for a tutorial, but some tips on bringing up a new architecture would be amazingly helpful. Mar 09 19:11:41 heh :) Mar 09 19:14:04 i'd start by co'ing the bitbake and openembedded stuff. Mar 09 19:14:25 hi pb_, hi kergoth Mar 09 19:14:25 there is a distro (lite5200) that's similar. Mar 09 19:14:55 i'm lost on how much of the nslu2 stuff to use... Mar 09 19:15:23 HopsNBarley, just enought to make it work, not so much as to bloat it :) Mar 09 19:18:55 Crofton, that's always the trick! Mar 09 19:19:00 :) Mar 09 19:19:19 zecke: I suspect there might be issues there - we could perhaps do with a different event type which let you access the data after the configuration reading stage or something... Mar 09 19:19:58 zecke: It did work when I was testing but that was probably as I was only working on a single package (so it does only run once per package)... Mar 09 19:20:28 RP: I just guess it might not work Mar 09 19:21:00 zecke: It does to an extent but perhaps only partially :-/ Mar 09 19:21:27 RP: ATM I'm trying to fix - aeh understand - the most horrible c code ever Mar 09 19:21:35 ~lart mad_fcl Mar 09 19:21:35 * ibot nukes mad_fcl with a single large nuke Mar 09 19:22:35 zecke: :-/ Mar 09 19:22:42 * RP -> back shortly Mar 09 19:30:37 <_law_> hmm gpe doesnt boot since last update :-( Mar 09 19:30:57 <_law_> /etc/X11/login-managers/ Permission denied Mar 09 19:31:47 oh dear Mar 09 19:32:47 <_law_> Can?t open display (null) Mar 09 19:35:13 which version of gpe-dm is that? I thought I had got rid of the login-manager dependency. Mar 09 19:36:08 <_law_> gpe-dm 0.48-r2 Mar 09 19:36:27 _law_: do you have e17 installed? Mar 09 19:36:34 <_law_> no gpe Mar 09 19:36:40 if not, just removing login-manager should fix this Mar 09 19:36:57 <_law_> der isnt anything unter /etc/X11/login-managers/ Mar 09 19:37:26 _law_: that's because the change in gpe-dm that moved /etc/init.d/gpe-dm to /etc/X11/login-managers was reverted Mar 09 19:38:10 kergoth: ping Mar 09 19:38:12 <_law_> pH5: starting gpe-dm manually works :-) Mar 09 19:38:33 <_law_> so i only should remove login-manager from rc5.d ? Mar 09 19:39:08 _law_: ipkg remove login-manager Mar 09 19:39:16 <_law_> okay Mar 09 19:39:42 * reenoo ponders RCONFLICTing on it Mar 09 19:40:59 pb_: does ipkg grok something like Conflicts: foo << 1.2 ? Mar 09 19:41:17 not sure Mar 09 19:41:27 actually < will do Mar 09 19:42:17 err Mar 09 19:42:38 that was supposed to read foo (< 1.2) of course ;) Mar 09 19:46:47 Zero_Chaos: pong Mar 09 19:47:09 kergoth: out of curiousity, would you have time to upgrade nmap to 4.01 in OE? Mar 09 19:47:52 doubtful, i havent touched oe much in like a year Mar 09 19:47:57 feel free to send a patch to the list Mar 09 19:48:30 kergoth: I've been trying, apparently it's not that easy :-) I'll get it eventually I guess. Mar 09 19:48:59 hehe. well if i have some spare time i'll take a look, i've been meaning to get back into it some anyway Mar 09 19:49:22 kergoth: well, I'll submit it if I get it to work :-) Mar 09 19:49:30 good good Mar 09 20:02:22 <_law_> hmm gpe-dm only start manually init scripts always says Cant open display (null) Mar 09 20:02:47 <_law_> xmodmap: unable to open display localhost_0.0 Mar 09 20:02:53 <_law_> localhost:0.0 Mar 09 20:03:08 <_law_> starting it as root works fine Mar 09 20:03:09 hehe, http://unix.rulez.org/~calver/pictures/curves.jpg Mar 09 20:04:50 hi all Mar 09 20:05:37 hey Mar 09 20:05:49 hey koen|train Mar 09 20:06:27 pH5: I wonder why glib 2.10 crashes gpe-bluetooth Mar 09 20:07:54 pH5: gconf-merge-tree is also a nice thing to do Mar 09 20:08:12 especially on jffs2 Mar 09 20:11:35 koen|train: because it uses threads. last time I tried against some glib 2.9.x, it crashed deep inside glib :-/ Mar 09 20:11:43 * pH5 rebuilds with DEBUG_BUILD=1 Mar 09 20:12:04 <_law_> is it nomal that ther is no gpe-dm in rc5? Mar 09 20:12:08 <_law_> only in rc3 Mar 09 20:12:34 <_law_> rc2 Mar 09 20:12:43 rc1 Mar 09 20:12:45 blast off! Mar 09 20:13:15 <_law_> koen|train: remove gpe-dm from rc 1 ? Mar 09 20:13:33 nah, I was just simulating a count down Mar 09 20:14:21 <_law_> koen|train: ok Mar 09 20:18:08 ~stab inkscapeOSX Mar 09 20:18:09 * ibot runs at inkscapeOSX with an origami Swiss Army knife, and inflicts a nasty paper cut. Mar 09 20:19:08 Wow - iboit is up on the times if he already has an oragami Mar 09 20:19:20 heh Mar 09 20:19:30 1 kilo of winXP death Mar 09 20:21:45 <_law_> shouldnt gpe-dm also be in rc5 ? Mar 09 20:22:15 it should be Mar 09 20:22:23 unless you fell victim to the login-manager thing Mar 09 20:23:33 <_law_> ok so i?ll add in manualy Mar 09 20:28:55 re Mar 09 20:32:18 <_law_> thats the clue Mar 09 20:58:09 * RP returns Mar 09 20:59:48 e image is failing Mar 09 21:00:15 JustinP: http://ewi546.ewi.utwente.nl/tinderbox/showlog.pl?machine_id=4&logfile=20060309200900.log Mar 09 21:37:01 no one have problem with file-native package ??? Mar 09 21:37:47 <_law_> gremlin[it]: whats your problem? Mar 09 21:38:09 don-t compile Mar 09 21:38:24 | ../src/file -C -m /mnt/hda/oe/build/tmp/work/x86_64-linux/file-native-4.16-r1/file-4.16/magic/magic.mime Mar 09 21:38:24 | ../src/file -C -m magic Mar 09 21:38:24 | magic, 8022: Warning type `search/0x140 \x0\x40\x1\x0' invalid Mar 09 21:38:36 and a lot more of errors after Mar 09 21:38:54 <_law_> removedtmp/staging/$HOST_SYSTEM/bin/files ? Mar 09 21:39:25 <_law_> remove alse: tmp/staging/$HOST_SYSTEM/*magic* stuff Mar 09 21:39:29 <_law_> also Mar 09 21:40:26 <_law_> i found some magic stuff in lib/ share/ ... Mar 09 21:41:27 trying Mar 09 21:44:53 I'm offline for about four days (holiday) - see you then! Mar 09 21:45:27 <_law_> rwhitby: wish you nice holidays Mar 09 21:45:38 ok now work ... mainly seem was the magic library of previous versine to cause the toubles Mar 09 21:45:47 thanks _law_ Mar 09 21:45:52 <_law_> gremlin[it]: np Mar 09 21:46:12 <_law_> gremlin[it]: ran into same problem :-) Mar 09 22:02:51 hell Mar 09 22:07:05 those aren't e-image failures..... Mar 09 22:11:45 ~ibot Mar 09 22:11:46 extra, extra, read all about it, ibot is a blootbot written in perl run by TimRiker on his server. logs on http://ibot.rikers.org// , ibot, jbot, apt are all the same process. It uses sqlite, but mysql or other SQL storage is also supported. Mar 09 22:13:43 ~seen uli Mar 09 22:13:47 uli was last seen on IRC in channel #debian, 165d 8h 13m 50s ago, saying: 'i'm new in this forum'. Mar 09 22:53:50 <_law_> n8 all Mar 09 23:45:04 hi. does anyone know of a CF card that uses a chipset for which I can find source code drivers, to use the control and address registers? Mar 09 23:46:34 like umm Mar 09 23:46:36 ATHEROS? Mar 10 00:07:44 maybe try mad-wifi for atheros Mar 10 00:10:43 yeah, thanks Mar 10 00:10:51 looks like I'm stuck with agere Mar 10 00:11:12 Linksys made a wifi CF card up to 802.11g Mar 10 00:11:22 that's what I'll have to use Mar 10 00:16:09 03rw 07org.oe.dev * r8bcba3a9... 10/packages/xkbd/ (xkbd-0.8.15/fix-circumkey.patch xkbd_0.8.15.bb): xkbd: fix circum key mapping. fixes (reopened) hh.org Bug #1512. Mar 10 00:43:14 Anyone have a Nokia 770? Mar 10 00:53:34 Would you tell me what is package name for opie-devel? Mar 10 00:53:54 I want opie header files to compile my opie apps Mar 10 00:58:47 I want a million dollars Mar 10 01:29:44 code coverage is soo cool Mar 10 01:58:16 great deal : http://shop1.outpost.com/product/4300115?site=sr:SEARCH:MAIN_RSLT_PG Mar 10 02:13:48 must go today 2 laptops price 500 each message me if interested on msn at mcsltd2@hotmail.com, yahoo at thishastogotoday or aim at ogd443. Mar 10 03:27:46 anyone has cross compile for h1940 on OE? Mar 10 03:36:45 * france is back (gone 12:11:56) Mar 10 05:16:58 * france is away: Away Mar 10 06:40:58 has anyone sucessfully used a semi-recent version of firefox compiled from OE on something running GPE? Mar 10 06:42:15 It compiles fine for me but says "GConf Error: No D-BUS daemon running" and just sits there soaking up 100% CPU and doesn't display a window or anything... Mar 10 08:22:03 monring Mar 10 08:22:06 *morning Mar 10 08:42:42 koen: http://www.mathcaddy.com/windowsxpbootsonamac!!!!1/ <- finally Mar 10 08:44:03 ahaha...I was almost worried I was on the wrong channel for a second Mar 10 08:44:07 morning Mar 10 08:44:16 hehe Mar 10 08:44:36 mornin' Mar 10 08:44:42 it might be wise to put me on the ignore list, too much OT output from me Mar 10 08:48:05 morning Mar 10 08:48:21 hrw|work: put my 1G card through 16 iterations last night Mar 10 08:50:43 16 iterations of what? Mar 10 08:51:21 johnX: the read timeout test Mar 10 08:51:27 XorA: and? Mar 10 08:51:36 hrw|work: no problems with -r11 Mar 10 08:52:36 XorA: ok - with older you had problem? Mar 10 08:53:00 hrw|work: no, never had read problems with that card Mar 10 08:53:24 its just the lockup on write :-( Mar 10 08:53:28 ok Mar 10 08:54:33 what slap head in gpg signed a security upgrade patch with gpg? Mar 10 09:10:43 ~lart vim on windows Mar 10 09:10:52 XorA: vim6 or vim7? Mar 10 09:11:04 hrw|work: vim6.4 Mar 10 09:13:10 VikiMode is erroring but I cant actually see the error for more than a millisecond Mar 10 09:32:15 hi liam Mar 10 09:32:31 hey Marcin Mar 10 09:35:49 NOTE: package dropbear-0.48-r0: task do_build: completed Mar 10 09:37:07 good morning Mar 10 09:37:11 hi Bernardo Mar 10 09:37:15 ~lart hostap Mar 10 09:37:16 * ibot sends a legion of lawyers after hostap's head Mar 10 09:37:19 ~lart apm Mar 10 09:37:20 * ibot accelerates a free AOL cd to 50,000 rpm and lets apm feel it Mar 10 09:38:31 anyone else having problems with last update to hostap and apm (on the akita, oz354)? Mar 10 09:39:45 with sunday's version everything worked, updated last night, rebuilt, and now my wifi card won't work, and my akita won't resume from suspend Mar 10 09:42:02 Bernardo: 354 on cxx00 == borken Mar 10 09:43:32 :( Mar 10 09:43:55 the version I built sunday worked so well Mar 10 09:47:37 Time to revert... Thank god I kept the images Mar 10 09:49:49 cleaned dropbear/ dir Mar 10 10:18:36 ~lart our monotone servers Mar 10 10:18:37 * ibot calls our monotone servers on the phone ... the lights are on but nobody's home Mar 10 10:19:26 hrw|work: they off air? Mar 10 10:19:49 * Bernardo finds out he doesn't have the next latest image anymore... only a Feb20 image Mar 10 10:19:51 connecting..... Mar 10 10:33:34 hrw|work: dominion if up Mar 10 10:34:21 XorA: I push to vanille or ewi Mar 10 10:34:32 XorA: and do not use dominion Mar 10 10:34:57 hrw|work: and dont forget nslug2 Mar 10 10:35:19 XorA: nslu monotone server is their server Mar 10 10:35:42 OE official ones are vanille, ewi + dominion (which is on adsl or something like that) Mar 10 10:35:46 hrw|work: koen said all 4 had the same keys installed Mar 10 10:36:20 ok - but I do not feel a rush on pull/merge/update/push dropbear 0.48 and changes Mar 10 10:39:24 hrw|work: damn we need to upgrade gpg as well Mar 10 10:41:53 and we have maintainer for it.. Mar 10 10:48:19 hrw|work: Pawel Osiczko apparently Mar 10 10:50:59 yep Mar 10 10:51:11 hrw|work: working on it now **** ENDING LOGGING AT Fri Mar 10 10:59:57 2006