**** BEGIN LOGGING AT Tue Aug 30 02:59:56 2005 Aug 30 03:01:33 hi, isthere a way to build oe toolchain as 32bit instead if 64, cos bluez package always fails to build on my 64bit machine, whereas it all builds ok on my laptop ehich is 32bit Aug 30 03:02:08 ~lart bitbake Aug 30 03:02:08 * ibot hooks into a hydrant and hoses bitbake down Aug 30 03:02:09 ~lart bitbake shell Aug 30 03:02:10 * ibot declares bitbake shell a moron Aug 30 03:02:28 dan2003_: 'linux32 bitbake foo' Aug 30 03:02:45 bbsh lack "build -k", bitbake lack "*some*" build feature Aug 30 03:03:08 koen, thanks, i'll give that a whirl Aug 30 03:03:40 dan2003_: that will make uname pretend to be on a 32 bit system Aug 30 03:05:26 i have to do ranlib on a ppp lib and start over at the ppp stage aslo, is that normal? Aug 30 03:05:41 it carroes on after that Aug 30 03:25:14 morning Aug 30 03:26:56 hey XorA Aug 30 03:27:27 XorA: hi Aug 30 03:29:24 dan2003_: you mean bluez-utils-nodbus which errors out with LONG_MAX undefined? Aug 30 03:30:27 Loial, i do indeed Aug 30 03:31:14 just had the linux32 attempt die in a whole diiferent way, will try again after removing .ccache Aug 30 03:31:53 /tmp/ccyCMp9w.s:779: Error: invalid constant -- `bic r3,r3,#-4294967296' Aug 30 03:31:53 and a few more like it Aug 30 03:32:22 dan2003_: have the same, can't get my head around that error, so far it's the only package that fails. ASSUME_PROVIDED does wonders tho :) Aug 30 03:35:21 i'll give that a try then ;) Aug 30 03:50:29 hi ! Aug 30 04:42:09 hi to all Aug 30 04:53:19 2.6.13 released.. I'm waiting for 2.6.13-mm1 then Aug 30 04:53:21 hi lardman Aug 30 04:53:29 hi hrw|work Aug 30 04:53:38 hi RP Aug 30 04:54:11 hi hrw|work Aug 30 04:54:24 I flashed a GPE image from yesterday and the system info applet tells me I have 32MB memory on my C750 Aug 30 04:54:45 what does /proc/meminfo say? Aug 30 04:55:01 I was just about to ask what I should look at, hang on a tick Aug 30 04:55:04 RP: Hi. Aug 30 04:55:27 hi sirfred Aug 30 04:55:32 koen|slug: slight problem, left my Z charging on my desk at home :( Aug 30 04:56:09 heh Aug 30 04:56:16 hey RP Aug 30 04:56:36 Hopefully I'm back to some kind of server stability. No idea what got into it... Aug 30 04:56:37 RP: your last kernel change said something abotu a c7x0 charging bug? any more details Aug 30 04:56:49 koen|slug: I may pop home this arvo, so I'll take a look then. BTW what's a slug? Aug 30 04:57:34 lardman: http://www.ehomeupgrade.com/images/linksys_nslu2.gif Aug 30 04:57:41 see http://openzaurus.org/wordpress/newsletters/ooo-newsletter-200508/ Aug 30 04:57:48 XorA: mickeyl was able to have it autosuspend when he pulled the charger out due to a nasty race bug - probably only affected shepherd but I've fixed it for all of us and removed the pointless autosuspend Aug 30 04:58:16 koen|slug: Cool :) Aug 30 04:58:17 RP: ah, I foudn that on c860, didnt get chance to tell you Aug 30 04:58:46 XorA: Its fixed now anyway. Funny my devices didn't show it up... Aug 30 04:59:19 RP: Its something that the old sharp kernel used to do, and did it to yesterday, but as I was shutting down for the night Aug 30 05:00:36 XorA: Its due to the way the status interrupts are generated by the charging controller the devices use. We can work around it in software though Aug 30 05:04:12 CoreDump|afk: Did you see my message about akita 2.6 kernel testers? Aug 30 05:05:54 Out of interest, does Perl build these days? Aug 30 05:07:29 lardman: it does Aug 30 05:07:33 and it even works! Aug 30 05:07:36 koen|slug: Thanks Aug 30 05:07:43 5.8.7 is the one Aug 30 05:31:33 yep, perl (5.8.7) works nowdays, but the rdepends between the inidvidual 850+ module packages are a bit flawed, the ones that are there are correct, just a few missing on some Aug 30 05:32:05 * DaKa is working on that Aug 30 05:34:42 damn, we are missing a step, pdaXrom people think we sell OE built distribs Aug 30 05:35:50 morning mickeyl Aug 30 05:35:58 morning mickeyl Aug 30 05:38:04 morning folks Aug 30 05:40:33 morning mickeyl Aug 30 06:02:05 <[cc]smart> hi, shall remove AS and CFG qio schedulers to react on bug #41 ? Aug 30 06:02:05 hi mickeyl Aug 30 06:02:45 XorA: where? Aug 30 06:02:52 hi mickeyl Aug 30 06:04:08 [cc]smart: We're currently using CFQ as default aren't we? Aug 30 06:04:15 <[cc]smart> can bugs #110 and #111 get closed ? Aug 30 06:04:26 <[cc]smart> would be surprised.... Aug 30 06:04:30 <[cc]smart> lemme see Aug 30 06:05:40 <[cc]smart> all are compiled in actually, so prolly it's AS Aug 30 06:05:54 <[cc]smart> some of your running OZ kernel 2.6 ? Aug 30 06:07:05 <[cc]smart> in any case, reduction would be sensible. Aug 30 06:07:48 <[cc]smart> since AS and CFQ are heavyweight compared to DEADLINE and i would not expect any advantage on flash from the others it hink DEADLINE is the best choice Aug 30 06:07:54 <[cc]smart> we used it on slug Aug 30 06:08:36 <[cc]smart> s/used/use/ Aug 30 06:08:48 hi pb, hrw Aug 30 06:09:26 mickeyl: seen that RSS reader for qtopia? http://corsair.sourceforge.net/ Aug 30 06:10:03 <[cc]smart> ok, other way round. who would complain if i'd remove AS and CFQ ? Aug 30 06:10:09 hrw|work: same company, not the IT Manager Aug 30 06:10:10 [cc]smart: 110 and 111 can be closed Aug 30 06:10:18 hrw|work: s/not/now Aug 30 06:10:55 [cc]smart: if RP has no objection, I'd say lets try to remove all but one and see if we chose the right sheduler :) Aug 30 06:11:09 XorA: where they write that? Aug 30 06:11:34 hrw|work: never seen that. interesting... Aug 30 06:11:45 hrw|work: no source though Aug 30 06:12:16 hrw|work: ah, it's in the sf.net cvs. want to have a go at adding it? Aug 30 06:12:46 <[cc]smart> mickeyl: ah, wouldn't suggest that. leave NOOP and DEADLINE in and it should be fine. Aug 30 06:12:59 [cc]smart: ok with me Aug 30 06:13:04 mickeyl: maybe will look into Aug 30 06:13:08 hrw|work: I think I am confused between channels Aug 30 06:13:31 <[cc]smart> RP: your word Aug 30 06:13:33 * mickeyl kicks qmake to make it know about .edj's Aug 30 06:13:56 hrw|work:http://www.oesf.org/forums/index.php?showtopic=14427&st=75 near end of thread Aug 30 06:15:09 XorA: thx Aug 30 06:15:25 mickeyl: can bbsh get "build -k" command? Aug 30 06:16:15 hrw|work: add it to bugzilla, i'll think about it Aug 30 06:16:30 ok Aug 30 06:17:12 XorA: also note how they immediately confuse OE and OZ Aug 30 06:17:30 [cc]smart: I don't know enough about it have a preference one way or another. We should probably consider that not all of our IO is flash based. We also access CF, SD and microdrives... Aug 30 06:18:09 koen|slug: as always Aug 30 06:18:24 koen|slug: rename to ozXrom and it be simpler Aug 30 06:19:08 XorA: chmod -w / -R Aug 30 06:19:10 <[cc]smart> RP: DEADLINE supports these, too. What you could say to describe the difference in between e.g. AS and DEADLINE is that DEADLINE risks more headmovement but under non concurrent load reacts faster. Aug 30 06:19:34 <[cc]smart> AS tries to reduce headmovement Aug 30 06:19:39 RP: psst: sd and cf are flash too Aug 30 06:19:54 koen|slug: cf can be HD Aug 30 06:20:08 XorA: those are called microdrives Aug 30 06:20:22 and RP mentioned them separately Aug 30 06:20:24 <[cc]smart> on the slug it's quite common to attach USB HD and nobody complained about deadline scheduler :) Aug 30 06:20:25 cvs [checkout aborted]: connect to cvs.sourceforge.net(66.35.250.207):2401 failed: Connection refused Aug 30 06:20:25 koen|slug: or pcmcia sleds Aug 30 06:20:28 ARGH Aug 30 06:20:41 XorA: or usb drives Aug 30 06:21:09 koen|slug: At the driver level SD and CF are significantly different to raw flash Aug 30 06:21:17 I know Aug 30 06:21:22 but it still flash Aug 30 06:21:23 hrw|work: just retry. anon connections are limited Aug 30 06:21:27 yep Aug 30 06:22:54 <[cc]smart> RP: so do, don't, or you want more info ? Aug 30 06:23:45 can the powers that be modify http://oe.handhelds.org/success.php with the info in http://oe.handhelds.org/cgi-bin/moin.cgi/SuccesStories ? Aug 30 06:23:56 [cc]smart: let me try and research it for 10 minutes Aug 30 06:24:04 <[cc]smart> i can help you find URLs Aug 30 06:24:07 <[cc]smart> on me way Aug 30 06:27:22 <[cc]smart> ah now that i see it.. it's important to use more relatively recent comparisons in between schedulers... say age < 1 year Aug 30 06:28:35 ~lart cvs.sf.net Aug 30 06:28:35 * ibot puts cvs.sf.net through a wood chipper Aug 30 06:31:18 koen: success stories now links to the wiki. Aug 30 06:31:42 <[cc]smart> heh... have a very fast downloading URL here: /usr/src/linux/Documentation/block Aug 30 06:36:52 [cc]smart: Ok, I see no benefit from the schedulers for any device except spitz. Make as and cfq modular for those platforms. For spitz, we leave them in as in theory, AS does bring some benefits... Aug 30 06:37:11 (spitz has an internal microdrive) Aug 30 06:38:40 <[cc]smart> will do Aug 30 06:39:26 how do we switch @ runtime between the schedulers? Aug 30 06:39:57 <[cc]smart> there has been a patch not so long ago to allow that, but iirc you need /sys for that Aug 30 06:40:50 <[cc]smart> that aptch is in kernel Aug 30 06:40:52 mickeyl: look in /sys/block/someblockdev/queue/scheduler Aug 30 06:41:21 [cc]smart: We have sysfs and can change at will and per device Aug 30 06:44:42 cool Aug 30 06:45:28 * koen|slug reads about the death of jffs3 Aug 30 06:47:16 jffs3 was a mad bandwagon everyone jumped on. It was only an experiment Aug 30 06:47:49 the good news is jffs2 will see some improvements Aug 30 06:48:57 mickeyl: any news from the CELF people? Aug 30 06:49:26 koen|slug: no, i haven't been in contact with 'em yet Aug 30 06:49:32 in fact, never. Aug 30 06:50:43 * koen|slug tries to remember who mentioned CELF before Aug 30 06:50:58 There was a post on the oe mailing list from them Aug 30 06:51:43 http://tree.celinuxforum.org/pubwiki/moin.cgi/TestFarmProject Aug 30 06:51:48 mithro and pb_ Aug 30 06:52:36 "Open Embedded may represent the closest thing there is to a de-facto standard embedded distribution of Linux." Aug 30 06:52:42 (http://tree.celinuxforum.org/pubwiki/moin.cgi/OpenEmbedded) Aug 30 06:57:01 not bad, apart from the fact that OE is not a distribtion Aug 30 06:58:23 "OpenEmbedded is the name of a project which seeks to provide embedded linux distributions, built from scratch." Aug 30 06:58:26 sounds pretty accurate to me Aug 30 06:58:49 obviously that's not _all_ that openembedded provides, but still. Aug 30 07:01:25 morning Aug 30 07:01:41 morning kergoth Aug 30 07:01:54 hi kergoth Aug 30 07:07:13 anyone know how to create an ISO image from a "broken" cd under linux? Aug 30 07:07:17 dd doesn't work :/ Aug 30 07:07:41 mithro: maybe with cdparanoia? Aug 30 07:07:58 thats for music? Aug 30 07:08:11 I've only used it on audio cds Aug 30 07:08:20 no idea if it works on data cds Aug 30 07:09:01 <[cc]smart> RP: shall i in this case on spitz make AS compiled in and the others modular ? Aug 30 07:11:09 [cc]smart: No, just leave it for now - I'll try and do some benchmarking on that device Aug 30 07:11:32 [cc]smart: Mark the bug closed though - I'll just keep in mind I need to sort spitz out Aug 30 07:17:37 <[cc]smart> will close it after it runs :) Aug 30 07:27:17 hey...if i found a typo in a .bb file, what do I do? Aug 30 07:27:52 like, do submit a bug report, or what? Aug 30 07:28:39 LoDown: exactly - with a patch Aug 30 07:29:33 ok, well...i'll figure out how to make a patch then..it was kidna a stupid error, a patch file name was referenced wrong... Aug 30 07:31:54 wow, monotone sure is slow Aug 30 07:36:35 gb2: indeed Aug 30 07:38:29 If I got a dollar everytime people said that.... Aug 30 07:38:39 heh Aug 30 07:38:42 * koen|slug writes a businessplan for that Aug 30 07:38:42 heh Aug 30 07:39:05 Hey, maybe that *is* their plan Aug 30 07:41:57 03koen 07org.oe.dev * rdeeec139... 10/packages/gaim/ (19 files in 2 dirs): gaim: update packages/gaim as per bug #273 Aug 30 07:42:47 LoDown: monotone pull :) Aug 30 07:43:13 03koen 07org.oe.dev * r57a2e56a... 10/packages/gpe-session-scripts/gpe-session-scripts_0.65.bb: gpe-session-scripts 0.65: fix patch URI, closes bug #275 Aug 30 07:45:17 mickeyl: if we raise $500/month we can lock a russian dude in closet to do fulltime OE dev ;) Aug 30 07:45:25 haha Aug 30 07:45:53 i'm in :-D Aug 30 07:46:20 500 a month? Aug 30 07:46:44 baah!! what muppet env vars have i got casuing this? | /home/mango/oe/oetmp/cross/lib/gcc/arm-linux/3.4.4/../../../../arm-linux/bin/ld: skipping incompatible /usr/lib/libc.so when searching for -lc Aug 30 07:46:58 I can swing 500 a month Aug 30 07:47:07 /usr/lib/libc.a: could not read symbols: File format not recognized Aug 30 07:48:08 CosmicPenguin: that's what the pdaX dude is asking for Aug 30 07:48:16 otherwise he has to get a job to pay for food Aug 30 07:48:50 heh - too bad I don't care about pdaX development, or he would have a gig Aug 30 07:49:40 if only they werent such isolationists Aug 30 07:49:58 i have a question... Aug 30 07:50:13 03florian 07org.oe.dev * r0b1e74a1... 10/packages/gpe-session-scripts/gpe-session-scripts_0.65.bb: gpe-session-scripts: Fix typo. Aug 30 07:50:15 if you perform a pull in the middle of a compilation...say something breaks like happened to me... Aug 30 07:50:28 how much (if any) consequences will that have? Aug 30 07:51:05 LoDown: not much Aug 30 07:51:15 i found another problem... Aug 30 07:51:17 bitbake caches the metadate before starting the build Aug 30 07:51:27 03koen 07org.oe.dev * r80321922... 10/conf/distro/preferred-gpe-versions-2.7.inc: preferred-gpe-versions-2.7.inc: update gpe-login and gpe-session-scripts Aug 30 07:51:37 right, so do I need to blow my cache for the new bb's to take effect? Aug 30 07:51:50 or 'touch conf/local.conf' Aug 30 07:51:57 ok Aug 30 07:52:02 that is easier... Aug 30 07:52:11 anybody know why pcre is trying to link against my system libs? ie /usr/lib/libc.so Aug 30 07:52:15 i found a problem with the header paths for mininet... Aug 30 07:52:25 should I submit a bug report? Aug 30 07:53:07 yes Aug 30 07:53:12 ok :-D Aug 30 07:57:47 ok...another bug :-D Aug 30 08:16:55 what would happen if RMK decided to run the arm-linux site on an OE based distro? Aug 30 08:17:53 <[cc]smart> the installation would become famous for the fastet site ever DOSd by normal operation i guess Aug 30 08:17:55 <[cc]smart> :) Aug 30 08:18:34 <[cc]smart> methinks you need some piece of HW for that Aug 30 08:18:41 any of you use opera? Aug 30 08:19:12 emte: I - already free registered Aug 30 08:19:40 yeah, wasnt sure if anyone else knew they are giving away free reg today Aug 30 08:19:57 http://my.opera.com/community/party/reg.dml Aug 30 08:20:01 [cc]smart: heh, I don't think the arm-linux site is particularly high traffic. I imagine it would run quite happily on an nslu2. Aug 30 08:20:03 for the people who didnt Aug 30 08:26:03 i know that this isn't directly related to OE, but bitbake is trying to DL a file from CVS, and it says: cvs [login aborted]: reading from server: Connection reset by peer. Aug 30 08:26:11 anyone have an idea what is going on? Aug 30 08:26:23 i get the same message when I try to log into it manually... Aug 30 08:28:44 LoDown: fetch from cvs.sf.net? Aug 30 08:28:54 ya Aug 30 08:29:10 LoDown: sf.net is fscking slow last months Aug 30 08:29:22 it is trying to get gtk-webcore... Aug 30 08:29:31 ok..so just keep trying, in other words? Aug 30 08:29:34 03mickeyl 07org.oe.dev * rfaeb226f... 10/ (packages/pcmcia-cs/pcmcia-cs-3.2.8/borzoi/ide.opts .mt-attrs): pcmcia: ship dedicated ide.opts for borzoi Aug 30 08:30:19 LoDown: yep, keep trying Aug 30 08:30:36 ok...just wanted to make sure I wasn't doing anything wrong Aug 30 08:30:39 thanks :-D Aug 30 08:42:03 http://handhelds.org/scap/port.17244.png - why scap suxx so badly on my husky? Aug 30 08:42:43 Package: scap Aug 30 08:42:43 Version: 0.1-r1 Aug 30 08:43:30 upgrading to r4 (maybe this will help) Aug 30 08:45:10 nope... Aug 30 08:45:21 koen|slug: which version of scap util works at all? Aug 30 08:45:25 hrw|work: you need to ask some guy to not rotate the image produced off a 2.6 husky, I forget the name though. I meant to ask but got busy Aug 30 08:45:37 XorA: nelson Aug 30 08:45:44 hrw|work: they all work for me Aug 30 08:45:45 hrw|work: there you go Aug 30 08:45:46 ha-ha! Aug 30 08:45:59 CosmicPenguin: no, not mr muntz ;) Aug 30 08:46:08 koen|slug: its because scap works around sharp crapness, this workaround is no longer required Aug 30 08:46:22 XorA: it works on my 2.6 corgi Aug 30 08:46:32 code taken from top of /scap/ also does not work when started from shell Aug 30 08:46:36 http://handhelds.org/scap/port.17903.png Aug 30 08:46:36 koen|slug: in landscape? Aug 30 08:46:45 yep Aug 30 08:46:55 root@c7x0:~# /sbin/fbset Aug 30 08:46:56 -sh: /sbin/fbset: not found Aug 30 08:46:56 koen|slug: bizarre, I always see same result as hrw|work Aug 30 08:47:04 thats why probably? Aug 30 08:47:34 yes Aug 30 08:47:43 it uses fbset to get the orientation Aug 30 08:47:43 koen|slug: you should reply here: http://www.oesf.org/forums/index.php?showtopic=14427&st=75 with the single word 'Perhaps'. Aug 30 08:48:09 XorA, hrw|work : http://handhelds.org/scap/port.17973.png Aug 30 08:48:19 root@c7x0:~# uname -a Aug 30 08:48:19 Linux c7x0 2.6.12-mm1 #1 Tue Jun 21 23:09:47 CEST 2005 armv5tel unknown Aug 30 08:48:33 Linux c7x0 2.6.13-rc6-mm2 #1 PREEMPT Fri Aug 26 12:39:05 CEST 2005 armv5tel unknown Aug 30 08:48:45 koen|slug: 'which fbset' Aug 30 08:49:05 hmm.. fbset is not linked to busybox.. Aug 30 08:49:39 koen|slug: http://handhelds.org/scap/port.18230.png Aug 30 08:52:13 * koen|slug stares blankly at the latest post to oe@ Aug 30 08:52:53 http://handhelds.org/scap/port.18525.png - finally (but with shell code from hh.org/scap/ not with scap) Aug 30 08:53:35 * XorA joins koen|slug in blankness Aug 30 09:41:33 cu Aug 30 10:05:03 ~oe snapshots Aug 30 10:05:06 methinks oe snapshots is see http://ewi546.ewi.utwente.nl/OE/ Aug 30 10:20:39 trying to build gpe-image, I get: " Package libxsettings-client was not found in the pkg-config search path. | Perhaps you should add the directory containing `libxsettings-client.pc' | to the PKG_CONFIG_PATH environment variable | No package 'libxsettings-client' found..." Any ideas? Aug 30 10:21:50 I have libxsettings-client-dev_0.1[34]-r0_arm.ipk and libxsettings-client0_0.1[34]-r0_arm.ipk in build/tmp/deploy/ipk Aug 30 10:22:16 check whether a) this error message is right b) libxsettings has been built c) libxsettings inherits pkgconfig d) PKG_CONFIG_PATH is properly set Aug 30 10:22:29 if anything doesn't apply then it sounds like a bug in bitbake Aug 30 10:22:31 no Aug 30 10:22:35 oe, of course Aug 30 10:22:59 http://handhelds.org/hypermail/oe-commits/12/1253.html Aug 30 10:23:12 rebuild offending packages to correctly stage the .pc file Aug 30 10:24:51 thanks, I'll try that Aug 30 10:25:41 monotone: network error: failed to connect: Connection refused Aug 30 10:25:52 monotone --db=oe.db --quiet pull monotone.vanille.de org.openembedded.dev Aug 30 10:26:03 try one of the other servers Aug 30 10:26:18 hi mickeyl Aug 30 10:26:22 ewi546.ewi.utwente.nl or monotone.nslu2-linux.org Aug 30 10:26:27 mtd-snapshot-20050715.tar.bz2 is no longer in exsitance anywahere, and bitbake wants it ;( Aug 30 10:26:59 sounds like your oe tree is old Aug 30 10:26:59 pulling from ewi Aug 30 10:27:18 in my oe tree we have upgraded to 20050801 Aug 30 10:27:29 i just did pull Aug 30 10:27:45 should i delete and restart the build? Aug 30 10:27:59 ahh, i have to do the checkout thing t complete the update? Aug 30 10:28:01 2 secs Aug 30 10:28:04 of course Aug 30 10:28:25 that should fix it ! Aug 30 10:29:43 anyone recall how to change where the @handhelds.org email address forwards to? Aug 30 10:30:04 send $bribe to france or jamey Aug 30 10:30:14 since the admin interface is still down afaik Aug 30 10:30:27 k Aug 30 10:30:52 otherwise https://admin.handhelds.org/ would be the place (iirc) Aug 30 10:33:50 monotone: warning: missing packages/efl/esmart_0.9.0.004.bb \n monotone: misuse: 1 missing files Aug 30 10:34:08 do I need to do anything about that, or will it take care of itself? Aug 30 10:35:44 koen, I've pull the latest from ewi, but classes/pkgconfig.bbclass doesn't have florian's commit from yesterday. Is ewi not up-to-date? Aug 30 10:36:26 it is Aug 30 10:36:44 but you have to rebuild a packages using pkgconfig Aug 30 10:36:54 all* Aug 30 10:37:14 since OE thinks they built (staged) ok Aug 30 10:37:27 yeah, I'll rebuild the packages, but shouldn't I see the change in the commit you mentioned above in classes/pkgconfig.bbclass? Aug 30 10:37:38 you should Aug 30 10:37:43 did you do a monotone update? Aug 30 10:37:55 yes Aug 30 10:38:19 unless monotone update bailed out after printing the error I mentioned Aug 30 10:38:52 monotone: warning: missing packages/efl/esmart_0.9.0.004.bb \n monotone: misuse: 1 missing files Aug 30 10:40:20 mt drop packages/efl/esmart_0.9.004.bb Aug 30 10:40:24 then try again Aug 30 10:40:53 or 'monotone list missing | xargs monotone drop' for all files Aug 30 10:41:22 about OE blowing up on gentoo: Aug 30 10:41:23 19:38 < peteru-home> Arghh!!! Effing gentoo patches just cover up the broken expand command by patching the testsuite to not run tests on expand. Aug 30 10:41:33 heh Aug 30 10:42:08 people how emerged world and use OE might have some troubles Aug 30 10:43:03 /show/who/ Aug 30 10:47:06 http://my.opera.com/community/party/reg.dml Aug 30 10:47:27 koen, mickeyl, thanks, that helped Aug 30 10:47:58 mreimer: koen@bitbake:~/OE/monotone/openembedded/packages$ cat tslib/tslib/ts.conf-h2200 module_raw inputmodule linear_h2200 Aug 30 10:48:15 shouldn't that be 'module_raw linear_h2200' ? Aug 30 10:48:24 koen|slug: yes Aug 30 10:48:31 * koen|slug fixes Aug 30 10:50:01 thanks koen Aug 30 10:51:08 uh oh, I mt drop the wrong file. how do I get it back? Aug 30 10:51:14 mt revert Aug 30 10:51:19 thanks :-) Aug 30 10:51:39 I did the same 30 minutes ago :) Aug 30 10:52:23 03koen 07org.oe.dev * r1c3f5d84... 10/packages/tslib/ (tslib/ts.conf-h2200 tslib_cvs.bb): tslib: change ts.conf-h2200 to use module_raw for h2200 to include it in the calibration, bump PR Aug 30 10:53:05 oops Aug 30 10:53:22 s/ h2200/ linux-h2200/ Aug 30 10:53:42 linear* Aug 30 10:54:18 the module name uses an underscore, right? if that's what you're referring to Aug 30 10:54:47 module_raw linear_h2200 Aug 30 10:55:27 yes Aug 30 10:55:33 simple patch, hard commit message Aug 30 10:55:52 that should fix the ts problem a bit Aug 30 10:57:20 yes, it helps a bit--at least it's possible to tap on the panel Aug 30 10:58:39 if that build succeeds, could you upload it to hh.org/~mreimer so I can test it on a sd card? Aug 30 10:58:54 sure. Aug 30 10:59:19 Do you want a pristine gpe-image, or would the one I built the other day be good enough? Aug 30 10:59:39 which in is the best? Aug 30 11:00:12 03mickeyl 07org.oe.dev * rbf33034c... 10/conf/machine/zaurus-clamshell-2.6.conf: zaurus-clamshell-2.6.conf: revert accidant commit that outcommented several modules Aug 30 11:00:17 Well, I know the gpe-image from the other day works because I tested it. Aug 30 11:01:02 I'll go ahead and upload that one for now, since this build is going to take a long time Aug 30 11:01:52 koen|slug: so you want to boot from SD? I'll build a kernel + modules for you to do that Aug 30 11:02:45 I'm preparing a sd card right now Aug 30 11:03:01 * mreimer enjoys having lots of disk space now, and "make O=..." Aug 30 11:03:02 should the ext2 partition be on p1 or p2? Aug 30 11:03:33 koen|slug: doesn't matter, as you can specify which in your kernel args (root=/dev/mmcblk0p2 or whatever) Aug 30 11:03:46 nice Aug 30 11:03:55 koen|slug: are you going to boot from haret? Aug 30 11:04:17 I think so Aug 30 11:04:23 that's simplest Aug 30 11:04:27 unless there is an easy way to restore wince Aug 30 11:05:06 yes, very simple if you have a big enough CF card, but not really necessary if you just want to try GPE and keep wince functional Aug 30 11:05:53 I'll wait for gpe 2.8 before inflicting the linux horror on my girlfriend ;) Aug 30 11:06:13 :-) Aug 30 11:07:13 koen|slug: kernel still building (competing for CPU with bitbake parse) Aug 30 11:14:43 wb RP Aug 30 11:16:23 mickeyl: hi. Did I miss anything? (my server is still behaving in an odd fashion) Aug 30 11:19:30 nothing important ;) Aug 30 11:19:43 mickeyl: I've had one report of the 2.6 C1000 kernel booting which is nice to know Aug 30 11:22:37 ya, that's great! Aug 30 11:23:37 btw., there's one thing i don't like in the decision to use the switch event interface for the hinge Aug 30 11:24:13 since you can't filter events on /dev/input/event, i have to listen to all events and discard everything but the switches Aug 30 11:24:16 kind of suboptimal Aug 30 11:24:55 bye all Aug 30 11:25:40 koen|slug: http://handhelds.org/~mreimer/mmc/20050830/ Aug 30 11:25:50 cool, thanks Aug 30 11:25:51 perhaps a third device reporting nothing but switches would make more sense Aug 30 11:25:54 * koen|slug wgets Aug 30 11:27:12 mickeyl: That makes no sense to the kernel people... Aug 30 11:27:18 What events do you have to ignore btw? Aug 30 11:27:35 all keys Aug 30 11:28:43 mickeyl: Can you not get them from the same source as the keyboard driver? I guess that abstraction doesn't make sense... Aug 30 11:29:04 no, the keyboard driver works on a different level (qte core) and listens on /dev/tty Aug 30 11:29:46 Ah, I thought it had switched to the input event layer... Aug 30 11:30:00 I don't think there is anything we can do I'm afraid... Aug 30 11:30:26 The only alternative would be a seperate input device but the kernel people are unlikely to go for that... Aug 30 11:30:37 nope. i don't expect anything as well, i was just pointing out that it's a bit suboptimal. it'll work fine anyway. Aug 30 11:31:02 mreimer: any recommended version of haret? Aug 30 11:31:27 I'm using 0.3.4. I tried fluffy's version, but it wouldn't run on my version of wince Aug 30 11:31:30 mickeyl: Yes, I hadn't thought of it like that. We can't have everything I guess ;-) Aug 30 11:31:37 koen|slug: do you need haret too? Aug 30 11:32:01 mreimer: I can't seem to find in in my homedir Aug 30 11:32:08 koen|slug: just a sec... Aug 30 11:33:07 koen|slug: http://handhelds.org/~mreimer/haret.exe Aug 30 11:35:26 <[cc]smart> arg Aug 30 11:35:31 <[cc]smart> ERROR: see log in /data/work/nslu2-linux/openzaurus/tmp/work/gcc-snapshot-cross-20050830-r0/temp/log.do_configure.11742 Aug 30 11:35:45 <[cc]smart> The following requested languages could not be built: f77 Aug 30 11:35:45 <[cc]smart> Recognised languages are: c,ada,c++,f95,java,objc,obj-c++,treelang Aug 30 11:35:54 i think gcc-snapshot is pretty much unmaintained Aug 30 11:36:13 better use 3.4.x Aug 30 11:36:15 <[cc]smart> then i'm building the wrong target prolly Aug 30 11:36:36 <[cc]smart> just did my "initial setup" for openzaurus Aug 30 11:36:56 <[cc]smart> never tried that before if i don't count sth like... 2 years before ? Aug 30 11:37:04 sounds like DISTRO is wrong then or something Aug 30 11:37:22 <[cc]smart> DISTRO="openzaurus" Aug 30 11:37:23 03justinp 07org.oe.dev * r8457b71b... 10/packages/musicpd/ (glurp_0.11.3.bb glurp_0.11.6.bb): glurp: update version, switch to SF mirror Aug 30 11:37:26 <[cc]smart> MACHINE="collie" Aug 30 11:37:29 that's not recommended Aug 30 11:37:34 read the distro configs Aug 30 11:37:39 openzaurus has everything free floating Aug 30 11:37:52 better use openzaurus-3.5.4.conf Aug 30 11:38:06 which nails down fixed versions of tools and such Aug 30 11:38:49 <[cc]smart> ok, thx Aug 30 11:38:59 np Aug 30 11:45:21 hmmm Aug 30 11:45:35 the switches don't seem to be included in the global key state Aug 30 11:45:41 that's not good Aug 30 11:45:52 * kergoth wonders what snmp bits are in oe Aug 30 11:45:59 mickeyl: There is a different ioctl for switch states Aug 30 11:46:04 koen|slug: also, I put evtest / ts_test / ts_calibrate in /home/root Aug 30 11:46:17 RP: ah, a new one. do you have that link for the patch again, please? Aug 30 11:46:39 * koen|slug directs people attention to http://bugs.openembedded.org/show_bug.cgi?id=232 Aug 30 11:47:14 mickeyl: http://www.rpsys.net/openzaurus/patches/corgikbd_switch-r0.patch Aug 30 11:47:23 koen: i vote for the compat link Aug 30 11:47:27 RP: thanks Aug 30 11:47:50 mickeyl: Sadly we can't do one kernel supports all for cxx00's if we boot off CF on spitz as the cmdlines are different. How annoying :-/ Aug 30 11:48:50 RP: yeah. Aug 30 11:49:02 * mickeyl looks around for the guys who talked about a new zaurus bootloader and then vanished Aug 30 11:50:32 I'm back Aug 30 11:50:51 re my shepherd only having 30MB of memory, I've got /proc/meminfo Aug 30 11:51:34 http://pastebin.ca/21622 Aug 30 11:51:53 lardman: What does /proc/cpuinfo say your device is? Aug 30 11:53:33 Which bit, I've not got a connection so I'll have to type it by hand Aug 30 11:53:50 lardman: Does it say its a Sharp Corgi or a Sharp Shepherd? Aug 30 11:54:35 actually hang on, I can do some CF card shuffling - anything else you want to see? Aug 30 11:54:50 It says Corgi Aug 30 11:55:04 uh oh Aug 30 11:55:12 this is kernel 2.6 ? Aug 30 11:55:13 lardman: and its a C750? That's the problem then - its being misdetected :-/ Aug 30 11:55:18 How recent is the kernel? Aug 30 11:55:22 yes that's right Aug 30 11:55:26 This is with: Linux c7x0 2.6.13-rc5-mm1 #1 PREEMPT Mon Aug 29 16:07:52 BST armvtel unknown Aug 30 11:55:44 oops Aug 30 11:55:46 righto Aug 30 11:55:47 Rats. I've broken the detection scripts :-( Aug 30 11:55:47 I pulled & updated just before I built it Aug 30 11:55:52 here as well Aug 30 11:56:01 glad it's not just me :) Aug 30 11:56:10 I tested on corgi and husky... Aug 30 11:56:13 hehe Aug 30 11:56:41 I'll be about tomorrow for testing if needs be Aug 30 11:56:52 Switch global status: Aug 30 11:56:52 Switch (bit 0) active Aug 30 11:56:52 Switch (bit 1) active Aug 30 11:56:58 hmm, yes. Aug 30 11:58:32 Anyway night all, I've also got to get back to my PhD Aug 30 11:58:39 about how long should it take to build gpe-image? Aug 30 11:58:58 lardman: I'll try and track the problem down and will let you know the result Aug 30 12:01:39 lardman, mickeyl : The bug is actually one I found and thought I'd fixed but forgot to bump the patch number. Increasing the sharsl_detection patch version to -r1 will fix this - sorry :-/ Aug 30 12:01:58 heh, no worries Aug 30 12:02:06 bbl Aug 30 12:04:33 monotone.vanilla.de is still in maintenence ? Aug 30 12:04:38 hi Aug 30 12:07:31 hi hrw Aug 30 12:18:34 building gconf-dbus: "/opt/oe/build/tmp/staging/arm-linux/lib/libpangox-1.0.so: undefined reference to `_pango_engine_shape_covers' | collect2: ld returned 1 exit status". I rebuilt pango, but that didn't help. Ideas? Aug 30 12:18:48 03rpurdie 07org.oe.dev * r6332ba3a... 10/packages/linux/linux-openzaurus_2.6.13-rc5-mm1.bb: linux-oz-2.6: Update detection routines to correctly detect corgi and shepherd. Very nasty to get this wrongMT: ---------------------------------------------------------------------- Aug 30 12:28:38 RP: how much does c7x0 differ? not counting pxa250/255 and flashsize Aug 30 12:28:40 what the crap? I'm getting a 503 from sf.net....no wonder bitbake can't DL from there Aug 30 12:31:33 <[cc]smart> hm, seems iw rongly assumed all bug entries/comments would go to the mailinglist Aug 30 12:31:43 <[cc]smart> so i'm missing comments on bug # 272 Aug 30 12:32:01 only stuff assigned to oe@ goes to oe@ Aug 30 12:32:13 <[cc]smart> today, this became obvious :) Aug 30 12:33:27 <[cc]smart> is it possible to post- deliver to oe@ ? Aug 30 12:53:13 building gconf-dbus: "/opt/oe/build/tmp/staging/arm-linux/lib/libpangox-1.0.so: undefined reference to `_pango_engine_shape_covers' | collect2: ld returned 1 exit status". I rebuilt pango, but that didn't help. Ideas? Aug 30 12:54:19 remove that file and rebuild again? Aug 30 12:55:09 I rebuilt it. I'll try removing and rebuild again Aug 30 12:58:17 hrw: Different charging hardware Aug 30 13:09:20 RP: ah.. Aug 30 13:10:45 hrw: and memory size Aug 30 13:11:30 My server's hdd is dieing, hence all the werid problems I'm been seeing. Its SMART status is still fine though (but the driver fails its own self tests) Aug 30 13:20:43 And finally Aug 30 13:21:23 an initial version of my mediadistribution is released... If you want to check what openembedded pkg's it has get a copy from http://www.chassi.se Aug 30 13:25:20 ldc: screenshots! screenshots! Aug 30 13:25:24 =) Aug 30 13:25:28 it's just running freevo Aug 30 13:25:32 freevo.sourceforge.net Aug 30 13:25:46 * koen|slug hints ldc at http://oe.handhelds.org/cgi-bin/moin.cgi/SuccesStories Aug 30 13:26:22 yea Aug 30 13:26:24 feel free to attach your changes to bugzilla Aug 30 13:27:05 just that i dont have a name for it yet :) Aug 30 13:27:26 the ldc-noname-via-epia-sp-mediadist-deluxe :) Aug 30 13:27:30 kergoth: converting you new boss to the OE religion? Aug 30 13:27:55 koen|slug: hehe, thats the -long- term plan. short term has more mundane things like actually using source control ;) Aug 30 13:28:09 ldc: tiny UxIXOS Aug 30 13:28:20 koen, well, there's just to many for me to manage... and i'm not really shure that i have followed any OE recommendations when writing the pkg's either Aug 30 13:28:58 that's why we can comment on bugzilla ;) Aug 30 13:29:13 koen, hmm... got to have an f'n in there to say it's running freevo also :) Aug 30 13:29:48 koen, hehe... well... that many comments will just get me down... Aug 30 13:30:47 koen, and also, i have been forced to adjust some pkg's to fit this configuration... But i think there is one thing that you just can yank out of my tree, and that's the xorg-6.8.99.16 pkg Aug 30 13:31:30 that one got support for via's CN400 chip Aug 30 13:31:41 deleting libpango* and rebuilding pango didn't help; I still get "/opt/oe/build/tmp/staging/arm-linux/lib/libpangox-1.0.so: undefined reference to `_pango_engine_shape_covers'" Aug 30 13:34:11 re Aug 30 13:42:39 argh... <1kbps to my server ;( Aug 30 13:43:14 :( Aug 30 13:46:05 probably will mirror archive on oz.org - but then people will shout 'why it is so fscking slow?' Aug 30 13:48:27 ~hail hh.org Aug 30 13:48:29 * ibot bows down to hh.org and chants, "I'M NOT WORTHY!!" Aug 30 13:51:20 ~hail chassi.se Aug 30 13:51:22 * ibot bows down to chassi.se and chants, "I'M NOT WORTHY!!" Aug 30 13:51:24 just had to try it :) Aug 30 13:52:22 and after a couple of well-deserved beer i'm easily amused...=) Aug 30 13:52:47 uh oh Aug 30 13:52:53 * koen|slug looks behind him Aug 30 13:53:08 yep, that guinness has been sitting there for a while Aug 30 13:53:31 ~hail Guinness Aug 30 13:53:33 * ibot bows down to Guinness and chants, "I'M NOT WORTHY!!" Aug 30 13:58:13 urgh.. I need a oe cheatsheet... what's /var in oe-isn? localstatedir? Aug 30 14:00:00 export localstatedir = "/var" Aug 30 14:00:09 cat conf/biktbake.conf > cheatsheet Aug 30 14:00:12 when in doubt, grep conf/bitbake.conf :) Aug 30 14:01:33 thx Aug 30 14:01:57 03hrw 07org.oe.dev * r080cfc56... 10/conf/machine/zaurus-clamshell-2.6.conf: fixed typo in conf/zaurus-clamshell-2.6.conf Aug 30 14:02:01 03hrw 07org.oe.dev * r0cce57df... 10/packages/ttf-fonts/ttf-dejavu_1.13-20050820.bb: ttf-dejavu: archive mirrored on my account on handhelds.org - close bug#272 Aug 30 14:02:33 ~lart sf.net Aug 30 14:02:33 * ibot turns sf.net into a lifesized tux doll Aug 30 14:02:42 ~lart sharp Aug 30 14:02:43 * ibot sends a legion of lawyers after sharp's head Aug 30 14:04:53 That won't work - thats where lawyers are bred Aug 30 14:06:04 yay! Aug 30 14:06:05 http://handhelds.org/scap/land.2419.png Aug 30 14:06:17 03koen 07org.oe.dev * r8fc2532e... 10/packages/gpesyncd/gpesyncd_cvs.bb: Aug 30 14:06:17 gpesyncd: add cvs version of gpesyncd till it sees a release Aug 30 14:06:17 NOTE: it needs libmimedir_cvs, but OE doesn't have a clean way to specify that Aug 30 14:06:17 ~lart OE Aug 30 14:06:17 * ibot frags OE with his BFG9000 Aug 30 14:06:29 Heh Aug 30 14:06:32 Now thats funny Aug 30 14:07:01 I thought cute comments were out Aug 30 14:07:04 somebody is going to go after you Aug 30 14:07:24 only grouchy reenoo Aug 30 14:07:24 it is worth it :) Aug 30 14:07:39 or grouchy groucho Aug 30 14:07:53 lol Aug 30 14:08:44 koen|slug: DEPENDS = "libmimedir-${CVSDATE}" :) Aug 30 14:09:52 ~lart sf.net once more Aug 30 14:09:52 * ibot gets a hotmal account and SPAMs sf.net once more Aug 30 14:10:15 I go sleep - when I wake up then maybe page from sf.net will load.... Aug 30 14:10:16 cu Aug 30 14:15:54 koen|slug: we should really set up a round-robin dns Aug 30 14:16:05 koen|slug: monotone.openembedded.org -> (vanille.de, etc) Aug 30 14:16:31 kergoth: ::shudder::....hwo often would those sync? we'd have hydras all the time Aug 30 14:16:39 I was thinking the same Aug 30 14:16:42 well maybe not all the time, but more often I would think Aug 30 14:16:43 and see monotone blowup on changed keys for hostnames Aug 30 14:16:49 heh, yep Aug 30 14:17:06 if monotone blows up that easily, we should be using something less unstable Aug 30 14:17:24 replace your sshd key at home Aug 30 14:17:28 ssh will blow up too Aug 30 14:17:29 See now thats just crazy talk Aug 30 14:18:04 unless you think the whole key based auth is unstable Aug 30 14:18:25 03nail 07org.oe.dev * r7241b1d9... 10/packages/postfix/postfix_2.0.20.bb: Fix hardcoded paths, and create dirs in image before trying to copy files there Aug 30 14:19:03 btw... shouldn't the build stamp be CET or something? Aug 30 14:19:03 NOTE: build 200508302317: started Aug 30 14:19:05 with ssh, i could easily replace the host key on the host end if i was setting up a round robin system Aug 30 14:19:18 another option would be to share the same privkey on all dbs Aug 30 14:19:19 not that i'd set that up with ssh, but you get the point Aug 30 14:20:36 but anyway, the backup servers are listed in the wiki Aug 30 14:21:05 if people didn't read GettingStarted I don't feel sorry if they get stuck Aug 30 14:21:36 clearly you've forgotten who you're talking to. you think i give a flying fuck about people who dont read the instructions? Aug 30 14:21:42 heh Aug 30 14:21:55 its still a good idea to do what we can to improve ease of use Aug 30 14:21:56 ~hail kergoth Aug 30 14:21:58 * ibot bows down to kergoth and chants, "I'M NOT WORTHY!!" Aug 30 14:22:21 whee, pull for the first time in a week or so Aug 30 14:22:23 * kergoth taps foot Aug 30 14:22:50 a bittorrent like pull would be nice Aug 30 14:23:08 have a list of servers, compare the head revision and pull away Aug 30 14:23:37 since everything has been digitally signed it shouldn't be too hard to implement it Aug 30 14:28:14 noooooooooo Aug 30 14:28:26 rday discovered the familiar@ list Aug 30 14:28:32 haha Aug 30 14:28:52 'i'm assuming that the "familiar" stuff is more Aug 30 14:28:53 up to date' Aug 30 14:29:05 to have something completely out of context Aug 30 14:30:53 he's one stupid question away from a procmail rule to dev/null Aug 30 14:31:39 * koen|slug plays some othello to get happy again Aug 30 14:39:43 lol Aug 30 14:40:16 just got a post on my forum.... asking if i'm distributing the sourcecode for all the pkg's in my OE buildroot Aug 30 14:41:32 ldc: they are all in the DL_DIR Aug 30 14:41:40 yea... i know... Aug 30 14:42:06 you know more than most people :) Aug 30 14:42:10 but it's a bit hard to fit 2.6Gb worth of sources on a 200Mb hosting ... =) Aug 30 14:42:28 you only have to provide them if they ask for it Aug 30 14:42:40 it's perfectly reasonable to burn a DVD and mail it Aug 30 14:42:42 nope... since i dont provide any binary images Aug 30 14:42:50 heh Aug 30 14:42:54 atleast yet... Aug 30 14:46:04 gonna distribute an image later on when i get some type of installation-configuration or web-configuration for the machine..... ... maybe just put out a bittorrent download for the sources... Aug 30 14:50:44 im receiving this error when trying to use bitbake build nano: /usr/local/bin/autoreconf: unrecognized option `--exclude=autopoint' Aug 30 14:50:55 is my version of automake too old? (v1.9) Aug 30 14:51:27 what version of nano? Aug 30 14:51:35 1.3.5 Aug 30 14:52:01 this is my first attempt at building anything with bitbake+oe. let me try something else to see if i still have this problem Aug 30 14:53:19 you have built the crosscompilers and sutch? Aug 30 14:53:44 nano works perfectly here.. Aug 30 14:54:11 no i didnt. i didnt see that in the getting started section on the Wiki. where can i find directions on building the cross compiling environment? Aug 30 14:54:23 OE builds it for you Aug 30 14:54:56 your version of autoreconf isn't the correct one Aug 30 14:55:10 koen, well, it depends on how you type the command :) Aug 30 14:55:21 yeah Aug 30 15:10:28 03nail 07org.oe.dev * r104db762... 10/packages/wakelan/wakelan_1.1.bb: Fix hardcoded path trying to set permissions on /usr/bin/ Aug 30 15:13:30 ldc: how did you invoke bitbake to compile nano? Aug 30 15:13:57 JustinTodd, "bitbake nano" Aug 30 15:18:22 arg, yea if i dont tell bitbake explicitly where the bb file is with -b it'll just scroll for ever saying ERROR: [Errno 2] No such file or directory: '/justin/oe/org.openembedded.dev/packages/[x]/[x].bb\n' even though my BBPATH (env) and BBFILES (local.conf) params are set correctly Aug 30 15:18:44 off to work l8r Aug 30 15:19:12 JustinTodd, well, it seems like it's not correctly set... Aug 30 15:19:55 OEROOT = "/home/bitbake/mediastation" Aug 30 15:20:05 BBFILES = "${OEROOT}/openembedded/packages/*/*.bb" Aug 30 15:20:26 JustinTodd: if you give it a specifit bb file it doesn't compile dependencies Aug 30 15:20:36 JustinTodd: you normally don't want to do that Aug 30 15:20:41 as an example in your $OEROOT/build/conf/local.conf Aug 30 15:21:07 ok. I havent set OEROOT. that might be it Aug 30 15:21:11 and when standing in $OEROOT execute "bitbake nano" Aug 30 15:23:22 excellent!!!! but SLLLLOOOOOOOOOOOWWWW Aug 30 15:23:36 well, time for bed here...... if you just follow the howto step by step it will work... got it working here on my first try... Aug 30 15:23:52 JustinTodd, it's slow the first time, next time it will use a cache that will speed it up a bit Aug 30 15:25:36 Ok. im still not entirely sure how this whole system works. Is everything done locally? ie. the source tarball is wget'd and cross compiled on the local machine? Aug 30 15:26:01 yes, all it downloaded and compiled on your system Aug 30 15:26:21 *gone* Aug 30 15:26:42 well this is an absolutely amazing tool. thanks for your help everyone Aug 30 15:26:48 ~lart boost Aug 30 15:26:48 * ibot takes a big bite out of boost's jugular vein Aug 30 15:27:42 i had troubles compiling boost aswell. i just used the monotone binary distrib Aug 30 15:36:10 Hey guys. I am trying to do a fresh checkout of org.openembedded.dev and am getting the following message: Aug 30 15:36:13 monotone: misuse: branch org.openembedded.dev has multiple heads Aug 30 15:37:02 I'd say monotone pull again Aug 30 15:37:25 since http://monotone.vanille.de/viewmtn/headofbranch.psp?branch=org.openembedded.dev only shows one head Aug 30 15:37:48 I tried that (delete the db, do a fresh, full pull) Aug 30 15:39:12 I'll try downloading the snapshot, update and checkout from that Aug 30 15:47:52 koen|slug: worked that time, thanks Aug 30 15:48:56 * koen|slug secretly refreshed the db snapshot Aug 30 15:50:23 yeah I noticed, I didn't have to wait for too many revision updates :) Aug 30 15:52:20 * CosmicPenguin abuses himself by doing a brand new refresh Aug 30 16:12:49 drw|bbl: did you ever do a commit locally? Aug 30 16:44:03 i have a problem...bitbake can't download the files for mtd-utils-native Aug 30 16:44:36 it trys two different mirrors, one is ftp and never connects, and the other is http, but does not have the mtd-snapshot Aug 30 16:45:36 maybe you can just search for it on google and place it in the sources directory? Aug 30 16:45:50 thats what I'm trying right now... Aug 30 16:46:02 didn't know if anybody knew of any good mirrors... Aug 30 16:47:24 LoDown, is it 200507 sumin its lookin for? Aug 30 16:47:42 20060801 Aug 30 16:47:48 *20050801 Aug 30 16:48:23 oh ok Aug 30 16:49:58 i can't even find MTD project home page...whatever MTD is :-D Aug 30 16:50:35 ~google for linux mtd Aug 30 16:50:57 duh...i was googling for everything but that Aug 30 16:50:58 thanks Aug 30 16:51:28 heh Aug 30 16:51:30 np Aug 30 16:54:21 03nail 07org.oe.dev * r13c59a81... 10/packages/bluez-utils/bluez-utils-common_2.18.inc: Add support for the broadcom firmware loader Aug 30 16:56:03 http://raphael.slinckx.net/blog/index.php/2005-08-31/the-wonders-of-pastebin Aug 30 17:01:49 in case you all that have been in here all day haven't figured out...i'm really new to the whole bitbake thing Aug 30 17:02:00 so I was wondering if I could ask some questions... Aug 30 17:02:31 1) how long is it supposed to take (on an average system) to build gpe-image...cause I want to make sure I didn't mis configure something Aug 30 17:02:34 koen|slug: Brilliant :D Aug 30 17:02:56 2) does it have to do a complete rebuild to build a new image, or does it just update the files that have changed since the last build Aug 30 17:03:33 * koen|slug builds a gpe image in 3 hours Aug 30 17:03:34 LoDown: it will take some hours to finish the first build, even on a fast machine Aug 30 17:03:53 3 hrs? Aug 30 17:03:56 yes Aug 30 17:03:57 LoDown: if you don't clean out the tmpdir then further compiled will just update things Aug 30 17:03:59 on http://ewi546.ewi.utwente.nl/phpsysinfo/ Aug 30 17:04:13 i have a 2.8 ghz xeon with 1 gig memory...and it has taken me a LOT longer ( nd I am still not done) Aug 30 17:04:26 NAiL: did you see the mail to oe@hh about commit messages? Aug 30 17:04:28 thanks JustinP, i was getting worried :-D Aug 30 17:04:41 JustinP: no, not subscribed Aug 30 17:04:47 03nail 07org.oe.dev * rcde1c3e7... 10/packages/linux/openslug-kernel-2.6.12.2/defconfig: Add broadcom bluetooth module to openslug-kernel Aug 30 17:04:51 03nail 07org.oe.dev * ra13ce7ef... 10/packages/linux/nslu2-kernel_2.6.12.2.bb: Bump PR Aug 30 17:04:52 koen|slug: pastebin requires X? Aug 30 17:05:11 JustinP: what was it about? Aug 30 17:05:20 NAiL: ah. well basically it said that all commit messages should be prefixed with "package name:"...and some other stuff Aug 30 17:05:25 let me see if I can find the mail Aug 30 17:06:37 shouldn't the monotone->CIA thingy be updated to remove the filenames modified then? Aug 30 17:07:04 no Aug 30 17:07:30 people tend to forget to include filenames sometimes Aug 30 17:07:42 and it keeps people honest :) Aug 30 17:07:46 heh :P Aug 30 17:07:50 fair nuff Aug 30 17:08:07 koen|slug: but why does pastebin want gtk? :( Aug 30 17:08:54 there we go Aug 30 17:09:01 http://marc.theaimsgroup.com/?l=openembedded-discuss&m=112500399319378&w=2 Aug 30 17:09:06 http://pastebin.ca/21673 <-- It breaks on my non-X box Aug 30 17:10:20 how do I build a single program with debugging symbols? Aug 30 17:10:45 INHIBIT_PACKAGE_STRIP =1 in conf/local.conf Aug 30 17:11:02 that prevents stripping Aug 30 17:11:07 ah, thanks Aug 30 17:11:08 no, that doesnt build with debugging symbols, that just doesnt strip Aug 30 17:11:14 yeah Aug 30 17:11:17 you'll want to add -g to the cflags as well as inhibiting hte strip Aug 30 17:11:22 * koen|slug is searching on how to do that Aug 30 17:11:23 iirc. Aug 30 17:11:32 isnt this sitting in local.conf.sample? Aug 30 17:11:34 i'm pretty sure it is. Aug 30 17:11:36 with a comment about it Aug 30 17:11:43 EXTRA_CFLAGS=" -g " ? Aug 30 17:11:56 eh? Aug 30 17:12:18 in that package .bb Aug 30 17:12:38 kergoth: yep Aug 30 17:12:40 in local.conf it'll affect everything, no? Aug 30 17:12:41 i know this doesn't seem like much to you guys (seasoned vets and all ;) ) but MY IMAGE IS DONE!!! Aug 30 17:12:43 yay Aug 30 17:12:45 kergoth: thanks.... Aug 30 17:12:46 :) Aug 30 17:12:48 ~lart JustinP Aug 30 17:12:49 * ibot farts in JustinP's general direction Aug 30 17:12:55 thanks for all your help Aug 30 17:12:59 ~cheer Aug 30 17:13:01 shift to the left, unshift to the right, push down, pop up, byte byte byte. Aug 30 17:13:02 NAiL: yes, which is why you only do it during hte build of that package. modifying .bb files locally unnecessarily tends to cause problems Aug 30 17:13:03 awww.... Aug 30 17:13:10 oh, it worked Aug 30 17:13:23 haha Aug 30 17:13:25 or write a bbclass thatll let you do it per package, similar to the cvsdate logic Aug 30 17:13:28 hah Aug 30 17:13:55 damnit, my inbound email is down :( Aug 30 17:14:03 in a few days i'll no doubt get thousands all at once Aug 30 17:14:04 fun fun Aug 30 17:15:50 'night all Aug 30 17:17:05 night koen Aug 30 17:17:56 gnite **** ENDING LOGGING AT Tue Aug 30 19:27:42 2005 **** BEGIN LOGGING AT Tue Aug 30 19:27:42 2005 **** ENDING LOGGING AT Tue Aug 30 19:28:05 2005 **** BEGIN LOGGING AT Tue Aug 30 19:28:34 2005 **** ENDING LOGGING AT Tue Aug 30 19:36:34 2005 **** BEGIN LOGGING AT Tue Aug 30 19:36:34 2005 Aug 30 19:58:24 hey guys where would i findd a list of known bug for oz 3.5.4?? found a few probs on my collie & gpe/3.5.4 Aug 30 20:00:19 ZonE-HaKin9|wrk: bugs.openembedded.org Aug 30 20:05:19 JustinP: D'oh ThanX Aug 30 20:47:27 anyone here know how to force oe & bitbake to retrive only from http?? i can't svn or cvs out from work :( Aug 30 20:47:46 use a proxy **** BEGIN LOGGING AT Tue Aug 30 21:02:13 2005 Aug 30 23:39:29 i'm off home guys l8r Aug 31 00:31:46 morning all Aug 31 00:33:33 morning Aug 31 00:33:34 morning RP Aug 31 00:48:28 hi Aug 31 00:49:33 hi Aug 31 01:04:15 hi hrw|work Aug 31 01:10:24 RP: I'm curious when will be 2.6.13-mm1 Aug 31 01:14:31 hrw|work: I'd expect one soon Aug 31 01:15:20 rc5-mm1 as stable and 13-mm1 as development? Aug 31 01:16:48 hrw|work: Maybe, yes although there probably won't be much difference between the versions Aug 31 01:17:32 RP: its nice to see that cxk models are recognised now Aug 31 01:19:46 hrw|work: Better still that the hacked ioexp driver works. Someone just needs to write a proper i2c driver for it now... Aug 31 01:20:31 yep Aug 31 01:21:07 We still need to confirm Borzoi works as well... Aug 31 01:22:04 what about adding post into oz.org about it? maybe someone does not read oz-users/oz-devel but want to test it? Aug 31 01:23:23 It could be an idea. We're going to need some testers... Aug 31 01:23:41 for all cxk probably Aug 31 01:24:07 yes Aug 31 01:27:55 good morning all Aug 31 01:28:03 hi koen|slu1 Aug 31 01:29:03 morning all Aug 31 01:29:28 so whats been happening for the last couple of weeks ?? Aug 31 01:31:14 morning koen|slug Aug 31 01:31:44 ade|desk: back from honeymoon? Aug 31 01:33:05 morning Aug 31 01:33:23 hi XorA Aug 31 01:34:26 morning ade|desk, XorA Aug 31 01:36:57 koen|slug: yyup :) Aug 31 02:54:11 morning Aug 31 02:55:43 I've just built octave 2.1.71 by renaming the bb file. The build was fine and I'll test it later, but I don't imagine there'll be any issues - do I need to submit a patch to bump the filename in monotone? Aug 31 02:57:03 lardman|work: submit a patch to bugzilla or change it yourself Aug 31 02:58:05 lardman|work: TARGET_FPU=soft, check the gcc .incs **** ENDING LOGGING AT Wed Aug 31 02:59:56 2005