**** BEGIN LOGGING AT Tue Jul 11 02:59:57 2006 Jul 11 03:43:33 03mwester 07org.oe.dev * r759e9a24... 10/packages/ (7 files in 5 dirs): Unslung: netconsole debugging support Jul 11 04:42:38 nifty, i'm getting "/usr/lib/libstdc++.so: could not read symbols: File in wrong format" when cross-compiling one of my libraries Jul 11 04:42:57 i wonder what i have done to break that :/ Jul 11 05:13:25 03mwester 07org.oe.dev * rff99c83e... 10/packages/ (2 files in 2 dirs): Unslung: netconsole - fix module parameter and change linuxrc to match. Jul 11 06:19:54 good morning all Jul 11 06:20:12 morning Jul 11 06:38:51 <_law_> morning koen Jul 11 06:38:59 hey _law_ Jul 11 07:00:06 morning all Jul 11 07:19:14 hey dirk Jul 11 07:19:33 hey koen Jul 11 07:22:14 heh Jul 11 07:22:36 looks like lilo pushed the wrong button *again* Jul 11 07:24:30 morning Jul 11 07:24:36 hey XorA Jul 11 07:24:49 poor RP Jul 11 07:31:33 hi Jul 11 07:31:52 hey hrw|work Jul 11 07:31:52 hey hrw|work Jul 11 07:34:18 hey Marcin, XorA Jul 11 07:49:10 rmrfchik:/cr Jul 11 07:49:54 XorA: http://www.oesf.org/forums/index.php?showtopic=20418 - you have such card iirc? Jul 11 07:50:31 hrw|work: network probs? Jul 11 07:51:57 do13_: as usual at work Jul 11 07:53:54 hrw|work: at least our network is stable at work Jul 11 07:55:25 hrw|work: and the best: today 2 hours work left, then we have a works outing :) Jul 11 07:57:35 hrw|work: I really should dig that card out :-) Jul 11 07:57:43 hrw|work: anyway, posted what I used to use to run it Jul 11 07:59:31 do13_: nice Jul 11 07:59:36 do13_: http://www.hrw.one.pl/2006/07/10/tosa-and-26-kernel/ Jul 11 08:01:59 hrw|work: I wished the Z bootloader would support something like this: http://dominion.kabel.utwente.nl/koen/cms/getting-back-to-ep93xx-testing Jul 11 08:02:35 hrw|work: sometimes battery info report 0%: is this only after resume? Jul 11 08:03:43 koen: I saw that Jul 11 08:03:46 do13_: yes Jul 11 08:04:09 hrw|work: thx. I know the reason:) Jul 11 08:08:26 do13_: any news on the ts<>lcd sync? Jul 11 08:11:00 koen: I had no time last weekend Jul 11 08:13:42 grr fontconfig suchs Jul 11 08:13:46 s/h/k Jul 11 08:42:34 so, can anyone give me hints about why my build would be picking up the native libstdc++ during a cross-compile? Jul 11 08:43:01 i'm somewhat baffled by the current bug Jul 11 08:43:52 zwelch: here do you see this ? Jul 11 08:44:35 my BBFILES are at http://svn.minisplat.org/svn/src/trunk/build.d/dist/oe/ and the problem child is libsplaudio-codecs Jul 11 08:45:53 ade|desk: does that answer your question? ;) (i'd referenced this in the channel earlier) Jul 11 08:47:49 does your package know that it is to be cross-compiled ? Jul 11 08:47:53 about four hours ago, i said "nifty, i'm getting "/usr/lib/libstdc++.so: could not read symbols: File in wrong format" when cross-compiling one of my libraries" Jul 11 08:48:08 well, i'd had success with all of the other libraries until this one Jul 11 08:48:23 your classes don't say anything about how it is to be complied Jul 11 08:48:25 i am guessing the problem lies in the fact that this is the first plug-in module that it's tried to build Jul 11 08:48:29 compiled Jul 11 08:48:35 autotools/pkgconfig Jul 11 08:49:02 and where is that inherited from ? Jul 11 08:49:10 zwelch: could I persuade you to look into upgrading/fixing glibmm and gtkmm? Jul 11 08:49:16 koen: yes Jul 11 08:49:21 :) Jul 11 08:49:37 if by "persuade" you mean "authorize" Jul 11 08:49:41 I suspect you inherited the deps on g***mm from minisip Jul 11 08:49:51 i have inherited quite a few deps Jul 11 08:49:57 and... i have other packages to add Jul 11 08:50:42 ade|desk: ${URI_ABOVE}/classes/minisplat.bbclass Jul 11 08:50:49 well, almost... one sec Jul 11 08:53:06 okay, i just committed my update to the minisplat.bbclass to inherit from pkgconfig Jul 11 08:53:13 as i should have done to start Jul 11 08:53:13 you inherit minisplat-svn and then ... ?/? nothing Jul 11 08:53:58 but the codec doesn't inherit minisplat but minisplat-svn Jul 11 08:54:16 libsplaudio-codecs_svn.bb -> include libsplaudio-codecs.inc -> inherit minisplat Jul 11 08:54:30 look again Jul 11 08:54:36 ditto Jul 11 08:55:03 am I the only one that since pulling repo this morning cant build a lot of stuff as bitbake spits out a load of backtrace? Jul 11 08:55:46 XorA: which repo exactly? Jul 11 08:55:52 zwelch: .dev Jul 11 08:56:10 i just rebuild fresh a few hours ago Jul 11 08:56:16 zwelch: have you tried since your change ? Jul 11 08:57:22 ade|desk: tried what exactly? Jul 11 08:57:37 bitbake packges ? Jul 11 08:58:16 howzat? Jul 11 08:58:27 just build it Jul 11 08:58:39 since your change from 358 to 359 Jul 11 08:59:09 oh. the change was in my tree, just not committed Jul 11 08:59:21 it's not really the problem Jul 11 09:00:21 or rather, i don't think it was the solution, because, yes, i believe i had rebuilt all the libspl* stuff with that change included Jul 11 09:00:56 (that just didn't make it less correct, just not the solution to my strange cross-compile problem) Jul 11 09:01:28 have you looked in the configure log to see if it has picked up any of the cross -toolchain ? Jul 11 09:01:29 otherwise, i'd have experienced problems before libsplaudio-codecs. again, i think it's a problem with the fact that it's a plug-in Jul 11 09:01:44 well, everything works up until the link Jul 11 09:02:00 and the link of the first plug-in, afaik (though i haven't actually verified that) Jul 11 09:02:01 works or compiles ? Jul 11 09:02:36 heh, nothing "works" yet for me with OE/bb, if we aren't counting compiling :) Jul 11 09:03:25 ~lart pastebin Jul 11 09:03:25 * ibot pours hot grits down the front of pastebin's pants Jul 11 09:04:06 this is what bitbake is throwing at me http://pastebin.ca/84763 Jul 11 09:04:11 ade|desk: this error is blocking me from adding the remaining bbfiles for my packages Jul 11 09:04:30 zwelch: check to see if your other stuff is compiled with the correct toolchain Jul 11 09:04:33 so, i haven't even had a chance to compile everything that i need to even bother testing to see if it "works" :) Jul 11 09:04:34 XorA: BBMASK? Jul 11 09:04:55 hrw|work: BBMASK = "" Jul 11 09:05:03 XorA: other regexp used? Jul 11 09:05:07 XorA: http://bugs.openembedded.org/show_bug.cgi?id=1124 Jul 11 09:06:04 nifty :) Jul 11 09:06:45 is there a proposed solution for that? Jul 11 09:06:56 Is there a way to make text/icon and so on bigger in Opie. The default out of OE is build for 3.5" 240x320 LCD, but we have 3.6" 480x640 LCD so the text/icons is kind of small before i adjust the apperance. Jul 11 09:07:34 I want it to come out of OE so it is usable at firsrt boot. Jul 11 09:07:52 hrw|work: nope, and nothing that I have changed recently Jul 11 09:08:33 well, it's a bug in compile Jul 11 09:08:47 so, it's compiling some file (presumably one that you updated) Jul 11 09:09:08 * zwelch points at the VCS log and shrugs Jul 11 09:09:19 look for the bad reg ex :) Jul 11 09:09:43 goxboxlive: try #opie Jul 11 09:10:11 ok koen. hi btw Jul 11 09:10:21 XorA: also, what version of bitbake? Jul 11 09:10:48 zwelch: latest stable branch Jul 11 09:10:54 and more generally, does that make a difference yet? Jul 11 09:12:40 ade|desk: fwiw, my minisplat repo has scripts for building to mingw32 too, so i know the system builds with cross-compilers Jul 11 09:13:02 it just doesn't build with every possible packaging/build system Jul 11 09:13:34 (and probably not for every archtecture targetable by cross-compilers, but that's an aside) Jul 11 09:19:14 is it safe to run two instances of bitbake on the same tree? :) Jul 11 09:20:05 zwelch: well, we won't see a nuclear fallout Jul 11 09:20:12 zwelch: but you could have issues :) Jul 11 09:20:34 yeah, kinda figured that would be non-trivial to acheive Jul 11 09:21:56 incidentally, that's gotta be a FAQ (if only in the minds of developers), and that's a good answer to quote in the manual ;) Jul 11 09:24:04 zwelch: e.g. if you have built the base toolchain Jul 11 09:24:15 zwelch: you could build most of opie and most of gpe in parallel Jul 11 09:24:36 heh, "most of" being the operative word Jul 11 09:26:05 the best solution would be a producer/consumer model, where all active instances communicate to let each other know which packages they are building Jul 11 09:26:37 of course, you also get to throw in loads of locking that slows down the comon single instance case Jul 11 09:26:44 * zwelch shrugs Jul 11 09:27:13 it's all good; developers should know how to ensure they always win in these kinds of races ;) Jul 11 09:27:35 one strategy being to avoid them altogether :D Jul 11 09:28:22 bye Jul 11 09:28:31 do13_: have fun! Jul 11 09:28:36 cu dirk Jul 11 09:28:43 thx Jul 11 09:29:29 zwelch: OE doesn't do locking of builds, so you could have two instances building the same package Jul 11 09:30:08 * zwelch loves races Jul 11 09:30:12 (which would be bad) Jul 11 09:30:14 I would start one task to fetch (bitbake -cfetch targetstuff) and second to build it Jul 11 09:30:18 they are definitely better than deadlocks Jul 11 09:30:31 but it can goes wrong if builds are faster then fetching Jul 11 09:30:40 a-yup :) Jul 11 09:30:54 with a bit of common sense it should work Jul 11 09:31:17 my bigger concern would be corruption of the cache Jul 11 09:32:12 * zwelch doesn't know if that's possible though Jul 11 09:40:22 ~lart nwfpe Jul 11 09:40:22 * ibot beats nwfpe over the head with a microkernel Jul 11 09:40:27 heh Jul 11 09:40:30 hehe Jul 11 09:40:50 koen: just assume how much more floating point emulation would suck on a HURD system Jul 11 09:41:32 nwfpe: bad structure size Jul 11 09:42:03 oh Jul 11 09:42:06 on EABI? Jul 11 09:42:12 yes Jul 11 09:42:15 anyway, torture starts Jul 11 09:42:20 I sent a patch for that to LAK Jul 11 09:42:32 where is that stupid fellow student... Jul 11 09:42:41 -} floatx80; Jul 11 09:42:41 +} __attribute__ ((packed,aligned(4))) floatx80; Jul 11 09:42:45 ~lart EABI Jul 11 09:42:45 * ibot stabs EABI Jul 11 09:42:52 ibot: botsnack Jul 11 09:42:52 koen: :) Jul 11 09:51:42 yes i was thinking the same thing Jul 11 09:51:49 does anybody has experience with VFP? Jul 11 09:52:01 with the 3.4.4 csl gcc? Jul 11 09:52:21 giel_: real vfp or softvfp? Jul 11 09:52:40 i _think_ softvfp, but i'm not sure :S Jul 11 09:53:16 well, i use -mfloat-abi=soft because gcc was complaining it couldn't do hardfloat in combination with vfp Jul 11 09:54:05 is it just me, or is oe.pastebin.com not working properly? Jul 11 09:54:44 zwelch: it isn't Jul 11 09:54:52 zwelch: try rafb.net/paste Jul 11 09:55:10 yes, well... how can i tell bitbake to use it for 'pastelog' Jul 11 09:55:17 ehm.... Jul 11 09:55:21 hm, let me try one more compile run, and if that doesn't work i'll tell elaborately what my problem is ;) Jul 11 09:55:22 * koen looks at mickey|bbl Jul 11 09:55:34 heh Jul 11 09:55:40 well, i can hack it myself... Jul 11 09:56:10 but that presumes a compatible pastebin protocol (request/response) Jul 11 10:00:03 koen: hurry up with your SoC :-) Jul 11 10:00:14 :) Jul 11 10:00:30 XorA: I'm writing down a flowchart in my google notebook Jul 11 10:03:14 I must admit I'm looking forward to seeing the results of the SoC project :) Jul 11 10:06:29 * _law_ is away: meeting Jul 11 10:06:37 hngr Jul 11 10:06:52 * koen cycles to the helpdesk to reset his uni password Jul 11 10:24:11 does glibc 2.3.5 build with binutils 2.17? Jul 11 10:27:09 koen, mickey|bbl: try this patch: http://rafb.net/paste/results/BVzqDk32.html Jul 11 10:27:23 incidentally, it was placed there by that patch :) Jul 11 10:28:24 * koen is in "wait a day for active directory to sync your accounts" hell Jul 11 10:28:47 hmmm Jul 11 10:29:02 koen: a day? Jul 11 10:29:09 okay, to come back to my problem Jul 11 10:29:11 | /home/oe/build/tmp/cross/lib/gcc/arm-linux/3.4.4/../../../../arm-linux/bin/ld: ERROR: /home/oe/build/tmp/cross/lib/gcc/arm-linux/3.4.4/libgcc.a(_ashldi3.o) uses FPA instructions, whereas /home/oe/build/tmp/work/glibc-intermediate-2.4-r5/build-arm-linux/elf/librtld.map.o does not Jul 11 10:29:16 lots of those Jul 11 10:29:46 giel: that means you switch params in the middle of a build Jul 11 10:29:52 switched* Jul 11 10:29:59 koen: i started a clean build Jul 11 10:30:18 hrm Jul 11 10:32:15 so, i have in my TARGET_CC_ARCH -mcpu=-march=armv5te -mtune=arm926ej-s -mfpu=vfp -mfloat-abi=soft Jul 11 10:32:49 shouldn't that be 'softfp' ? Jul 11 10:32:59 tried that, same effect Jul 11 10:33:33 right now you're telling gcc to use the vfp coprocessor Jul 11 10:33:44 ehm Jul 11 10:33:46 actually no Jul 11 10:34:55 source of the problem is that i'm not completely sure what i'm doing, i hate that ;( Jul 11 10:35:24 but i have to link against some libs i have here, which i compiled out of tree and they didn't compile without vfp Jul 11 10:35:44 and now my ffmpeg (in-tree) won't link against these libs if i compile it without vfp Jul 11 10:37:07 okay, i'll just fiddle on, if anyone has some great idea plz let me know ;) Jul 11 10:37:51 if all else fails: compile statically Jul 11 10:38:16 I suspect you might need some patches from crosstool/buildroot to get softvfp to work with gcc Jul 11 10:38:26 hmz Jul 11 10:38:37 might gcc 4.1 or something like that help? Jul 11 10:38:56 I sure hope so :) Jul 11 10:39:08 then i will also try to switch to that Jul 11 10:39:09 gcc 4.1.1 should have EABI (=softvfp) support Jul 11 10:39:23 so it should be able to handle sovftp without eabi as well Jul 11 10:39:25 but 3.4.4 csl should have as well, isn't it? Jul 11 10:39:30 no idea Jul 11 10:39:40 3.4.4 csl == patched to hell Jul 11 10:39:43 hm Jul 11 10:39:45 true Jul 11 10:39:57 okay, that's on my to-try list then, using 4.1.1 Jul 11 10:40:17 gm Jul 11 10:41:41 hey Crofton Jul 11 10:41:52 hi Jul 11 10:41:59 just back from Italy Jul 11 10:42:02 good vacation Jul 11 10:42:09 that's good to hear Jul 11 10:42:27 I like the kids playgrounds in beer gardens in Germany :) Jul 11 10:42:34 need to tell my friends with kids Jul 11 10:45:35 hey mickey|thesis Jul 11 10:45:41 for space reasons, i want that .pyo-files (optimized compiled python files) are not included in the image, as we aren't using python -O. what's the easiest way to do this? some package.bbclass hack or is there a nicer way? Jul 11 10:45:49 morning mickey|thesis Jul 11 10:49:16 hi Jul 11 10:51:33 goxboxlive: that's no problem. those settings are specified by qpe.conf. you should really look into the OE manual one of these days to learn about OVERRIDES :D Jul 11 10:51:43 goxboxlive: I'm commiting your large icons in a second Jul 11 10:52:00 mickey i have done it allready Jul 11 10:52:10 not in the OE but just localy Jul 11 10:52:38 But i need some help with keyboard. Can we move obver to #OPIE? Jul 11 10:52:50 *sigh* still your keyboard Jul 11 10:52:55 ya Jul 11 10:57:40 mickey|thesis: i'd be interested in hearing your thoughts about my pastebin patch ;) (no pressure) Jul 11 10:58:51 zwelch: first glance looks good. does it work? :D Jul 11 10:59:13 i'm guessing that's a rhetorical question, given my comment above :D Jul 11 10:59:23 right Jul 11 10:59:35 * mickey|thesis applies Jul 11 11:00:06 the only improvement might be to get the name of the buffer Jul 11 11:00:29 i.e. 'buffers' shows the name of the file, and i added the description to be passed along... might as well use it if we can still get at it Jul 11 11:00:48 that's right. Jul 11 11:01:07 i just did what i could do locally Jul 11 11:01:19 i.e. without more knowledge of the bb code Jul 11 11:03:17 I haven't had a chance to improve the shell lately. I plan to return to it in a couple of months. Jul 11 11:03:27 + stacking a GUI on it Jul 11 11:03:53 i'll probably continue to submit patches Jul 11 11:04:22 excellent! Jul 11 11:05:18 zwelch: my glibc build died for the same reason, even after adding the most sensible PREFERRED_PROVIDERS :( Jul 11 11:05:37 Kerwood_: doh! Jul 11 11:05:50 i wish i had better advice for you :/ Jul 11 11:05:54 zwelch: I need to go to other system where it worked and study the differences Jul 11 11:06:00 oooooh! Jul 11 11:06:09 I think I may have a dirty PATH Jul 11 11:06:10 wait... i just remembered what i think the problem is Jul 11 11:06:38 what version of binutils-cross are you getting? Jul 11 11:07:18 zwelch: 2.17 Jul 11 11:07:45 okay, i think that might be it; you'll note i added a PREFERRED_VERSION for 2.16 in my conf Jul 11 11:07:59 sorry. i guess i steered you to the wrong PREFERRED_THINGY the first time :( Jul 11 11:08:22 * zwelch should probably take notes about these things; his memory isn't what it used to be Jul 11 11:09:06 zwelch: Sure enough! Wow, this might be it! Thanks -- I'll give it a whirl now. Jul 11 11:09:07 it's worth noting that 2.16 is what you want, not the 2.16.91.* versions that are also there Jul 11 11:09:18 03mickeyl 07org.oe.dev * rc0df5bfe... 10/conf/distro/generic.conf: generic.conf include sane srcdates Jul 11 11:09:23 03mickeyl 07org.oe.dev * rf8ca5d54... 10/conf/distro/generic-unstable.conf: generic-unstable.conf detach from generic, since we don't want to include sane-srcdates here Jul 11 11:09:27 03mickeyl 07org.oe.dev * rfa3fb685... 10/packages/opie-taskbar/ (3 files in 2 dirs): opie-taskbar 1.2.2 add dedicated qpe.conf for htcuniversal Jul 11 11:09:39 zwelch: Roger dat Jul 11 11:10:56 zwelch: angstrom is using th 91.* version Jul 11 11:11:02 zwelch: patch applied in R546, thanks Jul 11 11:11:33 giel: i'm just reporting my experience... i had to rewind to 2.16 to get it working again Jul 11 11:11:59 zwelch: okay, i will note that if i get any problems Jul 11 11:12:06 zwelch: When do you plan to sleep, btw? Jul 11 11:12:13 and... i suppose that i only tried to .0.7 build Jul 11 11:12:24 Kerwood_: sleep is for the week... and the cubicle farmers Jul 11 11:12:41 zwelch: =:O Jul 11 11:12:44 * zwelch notes that the typo in the last sentence may be fatigue induced ;) Jul 11 11:12:59 zwelch: that's what angstrom is using as well, and i'm trying to build it right now Jul 11 11:14:37 Kerwood_: my patch record (as documented in the minisplat repo logs) shows my sleeping patterns can be erratic to non-existance Jul 11 11:15:04 * zwelch thinks of sleep as an unfortunately necessity rather than a habit Jul 11 11:15:22 zwelch: Well, as long as you aren't cheating with chemicals, I guess there's no harm Jul 11 11:15:38 zwelch: you can try the uberman sleep schedule Jul 11 11:15:39 wait, when did chemicals become cheating? Jul 11 11:15:52 heh Jul 11 11:16:12 * zwelch enjoys his caffeine, sugar, and nicotine trifecta Jul 11 11:20:52 * Kerwood_ is off to work. l8r... Jul 11 11:22:13 03zecke123 * r544 10bitbake/lib/bb/parse/parse_c/ (BBHandler.py Makefile bitbakec.pyx): bitbake c parser: first steps torwards a working parser Jul 11 11:22:58 there we go Jul 11 11:23:30 03zecke123 * r544 10bitbake/lib/bb/parse/parse_c/ (BBHandler.py Makefile bitbakec.pyx): bitbake c parser: first steps torwards a working parser Jul 11 11:23:45 hmm, thats odd Jul 11 11:23:51 kergoth: cool Jul 11 11:24:29 03zecke123 * r544 10bitbake/lib/bb/parse/parse_c/ (BBHandler.py Makefile bitbakec.pyx): bitbake c parser: first steps torwards a working parser Jul 11 11:25:09 if i were a betting man, i would guess that is not the last we are going to hear about r544 Jul 11 11:26:31 * zwelch wonders if predict bugs always works so efficiently for ensuring they aren't actually there Jul 11 11:26:40 03kergoth * r547 10bitbake/a: Testing CIA notification. Jul 11 11:26:53 03kergoth * r548 10bitbake/a: Testing CIA notification, again. Jul 11 11:26:59 there, all good Jul 11 11:27:01 RP_: what's the functional status of pxa270 cpufreq? Jul 11 11:35:46 <[g2]> has the monotone update occured yet ? Jul 11 11:35:59 [g2]: not yet Jul 11 11:36:06 <[g2]> morning koen Jul 11 11:36:15 <[g2]> do you know when that's to occur ? Jul 11 11:36:19 yes Jul 11 11:36:20 see http://www.openembedded.org/openembedded-migration-from-monotone-0-25-to-0-27 Jul 11 11:36:29 <[g2]> THX Jul 11 11:43:23 hello Jul 11 11:43:33 hi all Jul 11 11:43:50 hey cbrake & eFfeM Jul 11 11:44:20 koen: There isn't any patch I'd trust Jul 11 11:44:20 zwelch: so you got your bbclass to work :-) Jul 11 11:44:57 RP_: I see Jul 11 11:47:31 i really dont think cpufreq is sufficient, especially after messing with omap's Jul 11 11:49:33 kergoth: cpufreq + dyntick would make a nice combo Jul 11 11:51:22 is there any experience of angstrom working in akitas? Jul 11 11:51:40 last time I tried it was freezing pretty bad Jul 11 12:07:05 ~change hkd 61 to eur Jul 11 12:07:07 that doesn't look right Jul 11 12:07:22 ~change 61 hkd to eur Jul 11 12:07:27 61.00 Dollar (HKD) makes 6.15700 Euro (EUR) (from http://www.xe.com/) Jul 11 12:07:32 yeah, figured that out Jul 11 12:07:53 NAbyss, tnx Jul 11 12:08:03 np Jul 11 12:08:43 got my mp3 player from ebay, paid eur 14 s&h it had hkd 61 on it, was curious on how much that actually was Jul 11 12:09:27 ~change 31.30 hkd to eur Jul 11 12:09:32 31.30 Dollar (HKD) makes 3.15921 Euro (EUR) (from http://www.xe.com/) Jul 11 12:09:34 misread teh stamkps Jul 11 12:10:44 pH5: busybox 1.2.0 seems to work with eabi Jul 11 12:11:06 koen: finally booted something? Jul 11 12:11:31 oz 3.5.4.1 has been released but I only update to rc4, what's wrong? Jul 11 12:11:42 hrw|work: yep, on my c700 Jul 11 12:12:11 katossi_uni: ? Jul 11 12:12:47 ok maybe I should try deleting all the temps Jul 11 12:12:48 koen: cool. so no need for coreutils anymore? Jul 11 12:13:18 pH5: it does freeze just after syslog, but that might be due to angstrom-bootstrap-image not shipping some stuff Jul 11 12:13:31 pH5: but editing the udev script to run busybox works Jul 11 12:13:39 pH5: no weird errors Jul 11 12:13:57 pH5: I have to note that this is an armv4t rootfs Jul 11 12:14:45 hrw|work: what I meant is I pulled the sources from .oz354 but I could only see rc4 stuff Jul 11 12:15:30 katossi_uni: http://www.openembedded.org/repo/org.openembedded.oz354x/conf/distro/openzaurus-3.5.4.1.conf say 3.5.4.1 Jul 11 12:15:53 nops Jul 11 12:16:00 it says rc4 Jul 11 12:16:06 damn Jul 11 12:16:59 but I'm doing the pull and update! Jul 11 12:18:23 Anyone here using distcc with OE? Is it troublefree? Jul 11 12:19:47 inherit += "distcc"? Jul 11 12:21:07 Laibsch: try icecream Jul 11 12:21:23 Laibsch: no it is not troublefree, you have no influence about which toolchain the other clients are using Jul 11 12:26:34 hrw|work: despite zecke's warning (which I will investigate), would it be as easy as that simple line to activate distcc for openembedded? Jul 11 12:26:57 Laibsch: dont know - I have one machine only Jul 11 12:28:15 hrw|work: you could throw in you zaurii ;-) Jul 11 12:28:33 your Jul 11 12:28:43 Laibsch: all 4 ones? Jul 11 12:29:17 yes, don't be shy ;-) Jul 11 12:29:22 hrw|work: try OpenMosix on them :} Jul 11 12:29:27 You will have enormous computing power. Jul 11 12:29:41 zecke: Interesting project indeed. Jul 11 12:30:11 is it still around? Jul 11 12:30:14 zecke: Actually one that I was considering when first trying out OE. and you still needed 1+G of RAM which I did not have. Jul 11 12:30:29 Laibsch: trust me, you want icecream not distcc Jul 11 12:30:34 everybody loves icecream Jul 11 12:30:36 but OE will not run in parallel on openmosix since it is just a single thread. Jul 11 12:30:42 zecke: I trust you. Jul 11 12:30:59 Laibsch: RP will break the barrier soon (soon as in Free Software 'soon') Jul 11 12:32:06 * Laibsch does "apt-get install icecc" Jul 11 12:32:23 zecke: what barrier? Jul 11 12:32:29 the single-thread one? Jul 11 12:33:28 i see some bb files that have a DEPENDS on eg libxml2, shouldn't that be libxml2-dev ? Jul 11 12:34:28 Laibsch: yes, the single threaded one Jul 11 12:35:08 eFfeM: no Jul 11 12:35:29 eFfeM: libxml2 is recipe PN, libxml2-dev is result of build Jul 11 12:36:01 zecke: Do you use icecc with OE? What Desktop distro? Jul 11 12:36:16 Docs are more than sketchy for it on ubuntu. Jul 11 12:36:31 Laibsch: I used to use it Jul 11 12:36:44 Why did you abandon it? Jul 11 12:36:47 Laibsch: but that was was at my internship where we had more than one computer Jul 11 12:37:07 Everybody just single computer here, eh? Jul 11 12:37:57 Laibsch: we either live in more poor countries, or are just poor Jul 11 12:38:21 I have lots of old one lying around. And one main desktop/server and laptop. The latter are the only powerful ones by todays standard and the ones I use regularly. Jul 11 12:39:03 i have loads of computers, just old and not connected by a lan :( Jul 11 12:39:11 only my server is not >5 years old Jul 11 12:39:20 but that's because the mobo died 3 years ago Jul 11 12:39:33 all the other x86 crap is pretty old Jul 11 12:39:34 zecke: Well, for me it it the other way round. I stopped buying new computer a while ago but took the computers from my relatives who screwed up their Windows system on it and thought a new computer would be the cure-all Jul 11 12:39:46 hehe Jul 11 12:40:00 ;)) Jul 11 12:40:02 Laibsch: well, I wouldn't be able to pay the power bill for old computers... Jul 11 12:40:30 my machine is ~3 years old: athlonxp2200+, 1G ram, 120G hdd Jul 11 12:40:44 zecke: I would not want the power-bill for today's computer.s Jul 11 12:40:51 i was offered a MIPS beast, was well up for it, but then i found i needed 3-phase power for it :( Jul 11 12:41:21 ade|desk: a friend of mine used his onyx as a space heater in the winter Jul 11 12:41:33 hehe Jul 11 12:41:36 koen: ;D Jul 11 12:41:41 thats small fry Jul 11 12:42:18 this one was r4k but just had hundreads of the things Jul 11 12:42:31 zecke: power consumption per CPU horsepower has been pretty stable, Google complained lately. Jul 11 12:43:20 Thus faster computers waste more unless utilized for 100% all the time. Jul 11 12:43:35 get a T1 :) Jul 11 12:44:17 What for? To compile more? Jul 11 12:44:17 /me not trying to set a compile record here, just get stuff done ;-) Jul 11 12:44:36 hrw|work don't understand the -dev part completely, especially when working native I need to have the headers installed (but if I have a pacakge that does not cross compile, I still would like to use cross-compiled packages from other versions Jul 11 12:44:59 Laibsch: i thought that new servers reduced clock cycles if not needed, least HP Proliant ones do Jul 11 12:45:11 zecke: back to icecc. What did you do to make OE use it? I guess just inherit will not be sufficient, will it? Jul 11 12:45:19 YES at last. I have gotten the keyboard working in the HTC Universal right out of OE Jul 11 12:45:38 Laibsch: I have not used it in a while, it could be a bit rotted Jul 11 12:45:45 Laibsch: but INHERIT += "icecc" should do Jul 11 12:45:48 eFfeM: OE/bitbake know that "libxml2" exist. they do not know that result will be 'libxml2-dev' (depends on version of oe/bitbake) Jul 11 12:46:05 Wow, OE is absolutely incredible, unbelievable. Jul 11 12:46:08 zecke and hrw : cool right? finaly Jul 11 12:46:28 goxboxlive: awesome :) Jul 11 12:48:12 hrw|work, to make it more difficult: there is meta/slugos-native.bb which aims at installing a complete native toolchain, had to add DEPENDS=libc6-dev to get crt1.o in (and libc6 is from glibc) Jul 11 12:48:52 can we rename that to meta/build-essential.bb ? Jul 11 12:48:55 eFfeM: it should work with DEPENDS="glibc" RDEPENDS="libc6-dev" iirc Jul 11 12:49:01 koen: NO! Jul 11 12:49:21 koen: openslug-native has tooo much stuff inside to be called 'build-essential' Jul 11 12:49:45 koen: Depends: libc6-dev | libc-dev, gcc (>= 4:4.0.2-2), g++ (>= 4:4.0.2-2), make, dpkg-dev (>= 1.13.5) Jul 11 12:49:55 thats build-essential from debian Jul 11 12:50:19 i'd rather leave it to rwhitby and koen to decide would be happy if there is a build-essential.bb that I could include Jul 11 12:53:10 how to kill gvim: gvimdiff dump.sql dump2.sql (49k and 57k lines) Jul 11 12:59:55 can someone help me out with dropbear? Jul 11 13:00:04 gandhijee: Anything in particular? Jul 11 13:00:33 yeah i can't seem to login Jul 11 13:00:41 is get this Jul 11 13:00:41 authpriv.info dropbear[204]: Child connection from 192.168.1.2:20858 Jul 11 13:00:49 authpriv.warn dropbear[204]: login attempt for nonexistent user from 192.168.1.2:20858 Jul 11 13:01:00 but i am login in as root w/ passwd Jul 11 13:01:27 Hmm. Are you doing ssh root@device-ip? Jul 11 13:01:37 i am using putty Jul 11 13:01:43 Try it from the device itself.. ssh root@localhost Jul 11 13:01:45 so when i get there it ask for my user name Jul 11 13:01:59 one sec, i have to build dbclient Jul 11 13:03:11 ls Jul 11 13:05:25 do i have to setup another file with authorized IPs? Jul 11 13:05:38 03mickeyl 07org.oe.dev * ra853b21c... 10/packages/pulseaudio/pulseaudio_0.9.2.bb: pulseaudio 0.9.2 improve packaging Jul 11 13:06:14 gandhijee: Not AFAIK Jul 11 13:07:15 dbclient: connection to root@127.0.0.1:22 exited: Failed to get homedir Jul 11 13:07:27 one sec Jul 11 13:07:31 i have to fix somethings Jul 11 13:08:06 umm Jul 11 13:08:15 i can't even ping localhost... Jul 11 13:08:28 it doesn't pull that info from my hosts file. Jul 11 13:08:35 Try just using the IP Jul 11 13:08:43 yeah i did that Jul 11 13:08:52 dbclient: connection to root@127.0.0.1:22 exited: Failed to get homedir Jul 11 13:11:38 03mickeyl 07org.oe.dev * r74324a04... 10/packages/libsndfile/ (libsndfile1_1.0.11.bb libsndfile1_1.0.16.bb): libsndfile 1.0.11 -> 1.0.16 Jul 11 13:11:54 Hmm. Jul 11 13:12:02 Make sure root's ~ is set correctly? Jul 11 13:13:02 i have it set as the / dir Jul 11 13:13:18 What happens if you login as root.. what's pwd? Jul 11 13:13:30 "/" Jul 11 13:13:52 do i have to set a $HOME var? Jul 11 13:14:07 my embedded systems book doesn't really go in to too much detail Jul 11 13:14:14 That should be set by the login process, not your login script Jul 11 13:14:23 i.e. ssd Jul 11 13:14:26 er, sshd Jul 11 13:15:26 i don't have a login scipt Jul 11 13:15:34 Yeah, that shouldn't matter Jul 11 13:15:36 and i use busybox for everything else Jul 11 13:15:42 sshd should do it all.. Jul 11 13:15:58 i am using dropbear though Jul 11 13:16:10 do i need to symlink dropbear to sshd??? Jul 11 13:16:37 You shouldn't need ot Jul 11 13:17:08 man Jul 11 13:17:26 then the only thing i can say is dropbear needs better documentation Jul 11 13:17:31 do i need pty's?? Jul 11 13:17:36 Hehe, yeah, most things do.. Jul 11 13:17:38 ok, I am living in wierd world :-) Jul 11 13:17:41 :-( Jul 11 13:17:50 I'd only be guessing, but, yes, I'd say ptys would be necessary Jul 11 13:18:06 hmm Jul 11 13:19:19 the Legacy (BSD) PTY support in the kernel?? Jul 11 13:22:05 morning Jul 11 13:24:01 XorA: yes? Jul 11 13:26:07 gandhijee: that is not necessary (legacy stuff) Jul 11 13:27:17 busybox option then? Jul 11 13:27:20 [cc]smart, Jul 11 13:27:23 whoop Jul 11 13:28:00 i can't seem to get dropbear to let me in to my system Jul 11 13:31:36 i always get this: (none) authpriv.warn dropbear[220]: login attempt for nonexistent user from 192.168.1.2:20932 Jul 11 13:32:29 can mtn query the history of a single file ? Jul 11 13:32:45 ade|desk: mtn log --no-merges | more Jul 11 13:32:57 ade|desk: that tracks renames as well Jul 11 13:33:07 cool cheers Jul 11 13:35:37 hm... now i have different EABI versions, my own gcc is apparently version 4, while one of my other libraries (of which i don;t have the source) appears to be version 0 Jul 11 13:35:54 anyone know why dropbear would give me that error? Jul 11 13:36:11 NOTE: Applying patch 'zecke-no-host-includes.patch' Jul 11 13:36:11 NOTE: package gcc-4.1.1-r4: task do_patch: completed Jul 11 13:36:24 gandhijee: A long shot, if you can't work it out otherwise.. run dropbear through strace, or in debug mode? Jul 11 13:36:47 ah, those are guarded by ifdefs Jul 11 13:36:49 gandhijee: can u login from any other user? Jul 11 13:37:00 thats the only user on the machine. Jul 11 13:37:06 its an embedded style system Jul 11 13:37:12 IXP465 Jul 11 13:37:13 are you sure you have not set shell to /bin/false? :) Jul 11 13:37:19 yeah Jul 11 13:37:50 root::0:0:root:/:/bin/ash Jul 11 13:38:05 i can login VIA console Jul 11 13:39:39 maybe dropbear denies root logins? duh Jul 11 13:39:54 with a very confusing message Jul 11 13:40:18 koen: yes? Jul 11 13:40:47 no, i have it setup to allow root logins Jul 11 13:47:27 and today's phrase, "jesus, wtf just happend?" Jul 11 13:54:06 bitbake randomly choosing the wrong packages version to those set in the distro conf, but renaming tmp and then it chooses the correct ones Jul 11 13:54:20 wtf Jul 11 14:03:07 ade|desk: wtf, indeed Jul 11 14:03:47 ade|desk: which bitbake? Jul 11 14:04:09 ade|desk: Ive been fighting bb madness without hope all morning, then it just magically fixed itself Jul 11 14:05:05 1.4.2 Jul 11 14:05:09 shit 1.4.3 Jul 11 14:05:16 Hello Jul 11 14:05:36 XorA: voodoo magic Jul 11 14:06:42 i got fontconfig to run by cheating, replaced the Makefiles for fc-{case,lang,glphname} for local stuff, works ok now :) Jul 11 14:07:44 * Kerwood is comforted that others are seeing the same bb weirdness, too... Jul 11 14:07:49 ade|desk: must be your voodoo that fixed it Jul 11 14:08:22 firefox 2 seems to be nearing release Jul 11 14:08:29 any volunteers for adding it to IE? Jul 11 14:08:31 ehm Jul 11 14:08:33 OE Jul 11 14:09:05 koen: firefox 1.5.0.4 doesnt build, 1.5.0.1 builds but doesnt run, so I dont hold hope of 2.0 working :-( Jul 11 14:09:29 maybe they fixed all that Jul 11 14:09:31 *cough* Jul 11 14:09:51 koen: looking at the list of patches distros add to firefox I doubt it :-) Jul 11 14:09:57 lol Jul 11 14:15:55 hi ... Jul 11 14:16:18 i've have a problem during bitbake meta-opie Jul 11 14:17:39 it stops at intltool-native-0.31-r0 , it complains about missing XML::Parser Jul 11 14:18:08 pxa270: apt-get install libxml-perl or something similar Jul 11 14:18:25 i've already build libxml-parser-perl Jul 11 14:18:43 pxa270: but not a 15:54 < pxa270> i've already build libxml-parser-perl Jul 11 14:18:57 but not a libxml-parser-perl-native Jul 11 14:19:08 apologies for bogus line Jul 11 14:20:55 i've build both of them.. Jul 11 14:21:16 libxml-parser-perl and libxml-parser-perl-native Jul 11 14:21:40 pxa270: hrmm, no idea then Jul 11 14:25:32 mickey|thesis: The PACKAGES_DYNAMIC = "pulseaudio-module-**" breaks bitbake :} Jul 11 14:27:40 XorA: here's my error log - http://paste.plone.org/5206 Jul 11 14:31:41 pxa270: I guess it doesnt find the one in staging for some reason Jul 11 14:33:39 XorA: ok .. thanx.. Jul 11 14:33:39 03koen 07org.oe.dev * r17de3227... 10/packages/monotone/monotone-6_0.27.bb: monotone: add 0.27 Jul 11 14:34:35 pxa270: Im afraid Im just about to run away from office, so I cant look deeper Jul 11 14:35:17 XorA: no worries... i'll try and fix it... thanx.... Jul 11 15:07:56 * zecke hopes italy gets disqualified Jul 11 15:09:09 zecke: what? over the cup? Jul 11 15:09:56 thejapa: yes Jul 11 15:10:14 thejapa: since march of this year teams can be disqualified Jul 11 15:10:30 thejapa: if players say something discriminating or racist Jul 11 15:10:41 thejapa: which obviously has happened during the match Jul 11 15:10:53 so: Red Card for Zidane, Disqualification of Italy Jul 11 15:11:00 zecke: like calling zidane "you idiot playing like brazilian" Jul 11 15:11:01 World Champion is France :) Jul 11 15:11:02 hehe Jul 11 15:11:11 hi Jul 11 15:11:18 hi Jul 11 15:11:41 looks like your coach is parreira Jul 11 15:11:48 you are fat like ronaldo Jul 11 15:11:53 thejapa: I hope FIFA will investivage Jul 11 15:12:22 * koen <- food Jul 11 15:12:38 http://fun.drno.de/flash/games/zidane.swf Jul 11 15:13:01 zecke: just for fun :) Jul 11 15:13:14 thejapa: I have played it already :) Jul 11 15:13:18 k Jul 11 15:13:19 dillo don t do flash :( Jul 11 15:13:31 neither amd64 Jul 11 15:16:54 there is gnash for flash on amd64 Jul 11 15:17:13 I consider lack of flash a feature, not a bug.. Jul 11 15:17:35 hehe Jul 11 15:17:38 but a privilege :) Jul 11 15:17:48 NAbyss: no flash animation for the survey results? Jul 11 15:18:12 zecke: I wouldn't be able to author it, let alone view it.. Jul 11 15:19:48 Anyways, time for sleep Jul 11 15:40:42 cu Jul 11 15:45:07 is there any documentation on how to define new distro and machine files? I tried defining them to build the ti-davinci linux, but it fails saying it can satisfy a depmon dependency, which i dont include Jul 11 15:54:59 jasuris, the only docs i know of are by example Jul 11 15:55:26 just look and follow Jul 11 15:55:30 jasuris: it's in the usermanual Jul 11 15:55:43 k thanks Jul 11 15:55:45 which user manual? Jul 11 15:56:04 make sure that your kernel .bb file has a version number. Jul 11 15:57:07 * emte wonders why noone says anything until i take a guess ... Jul 11 15:57:44 probably because it's your guess which is inspiring the others Jul 11 15:58:06 emte: openembedded.org -> documentation -> user manual Jul 11 16:00:07 the manual section on building the distro files is empty Jul 11 16:00:11 yeah Jul 11 16:00:16 same is true for the machine Jul 11 16:00:17 i just saw that myself Jul 11 16:01:03 well, there is still my idea :) Jul 11 16:01:37 see http://www.openembedded.org/user-manual&dpage=ch05s05#id2468081 Jul 11 16:02:09 anyone have any idea which package provides libx11? Jul 11 16:02:31 tkp: a wild guess: libx11? Jul 11 16:02:57 koen: there is no such package Jul 11 16:03:18 powerbook-wlan:~/Projects/OpenEmbedded/org.openembedded.dev/packages/xlibs koen$ ls | grep libx11 Jul 11 16:03:18 libx11 Jul 11 16:03:19 libx11-native_6.2.1.bb Jul 11 16:03:19 libx11-native_cvs.bb Jul 11 16:03:19 libx11_20050226.bb Jul 11 16:03:19 libx11_6.2.1.bb Jul 11 16:03:21 libx11_X11R7.0-1.0.0.bb Jul 11 16:03:23 libx11_X11R7.1-1.0.1.bb Jul 11 16:03:25 libx11_cvs.bb Jul 11 16:03:27 libx11_git.bb Jul 11 16:03:30 hmm Jul 11 16:03:31 and diet-x11 of course :) Jul 11 16:03:42 fair enough... missed them somehow Jul 11 16:04:11 ahhh. xlibs :) Jul 11 16:09:52 mickey|thesis: adding the base kernel version to the bb file did the trick, thanks Jul 11 16:10:01 np :) Jul 11 16:10:26 if you want to understand why, then have a look at kernel.bbclass Jul 11 16:10:30 which contains Jul 11 16:10:31 DEPENDS += "virtual/${TARGET_PREFIX}depmod-${@get_kernelmajorversion('${PV}')} Jul 11 16:10:57 * mickey|thesis notes to write stuff like that into the book and manual Jul 11 16:11:16 hi mickey|thesis I fixed the HTC universal keyboard today. Sp now it works roght out of OE :-) Jul 11 16:11:22 goxboxlive: woot! Jul 11 16:11:27 goxboxlive: how did you manage? Jul 11 16:11:55 I disbaled a lot of patches in qte-common.inc and voila it worked Jul 11 16:12:08 oh Jul 11 16:12:40 not quite a solution we can apply, but may give us a hint Jul 11 16:13:03 it seems like key.patch and the machine definitions in it messed things up. Jul 11 16:13:23 So now iam starting to build with miltuthread support Jul 11 16:13:41 multithread Jul 11 16:14:30 is there a way to have a machine related qte-common.inc? Jul 11 16:14:59 Like i put it in a directory named htcuniversal ( like i do for inittab) ? Jul 11 16:20:40 zecke: I have a great idea Jul 11 16:21:06 zecke: let's add 'this chapter is a stub, please help us expand it' to the empty manual sections! Jul 11 16:21:39 hi I'm trying to follow the tutorial on the oe webpage. I'm stuck on the step of downloading the monotone database (OE.db.bz2), there is no such file in that directory, can some one direct me to the right place (file)? Jul 11 16:22:29 koen: and get launchpad integration done? Jul 11 16:22:38 zecke: heh Jul 11 16:22:39 koen: so people can just edit content :} Jul 11 16:22:55 YoG: works fine over here Jul 11 16:23:27 YoG: just follow http://www.openembedded.org/wiki/GettingStarted Jul 11 16:23:28 I'm getting an error bitbaking tslib... "ERROR: function do_patchcleancmd failed" Jul 11 16:23:29 koen: is that: "wget http://ewi546.ewi.utwente.nl/OE/OE.db.bz2"? Jul 11 16:23:37 no Jul 11 16:24:06 tmp/work/arm-linux/tslib-0.0cvs20060102-r39/temp/run.do_patchcleancmd.19882: line 497: syntax error near unexpected token `(' Jul 11 16:24:18 bronson: check your local.conf for spurious " or ' chars Jul 11 16:24:46 koen, I've already baked all the kernels and a buch of packages up til now. But will do. Jul 11 16:24:46 YoG: just follow http://www.openembedded.org/wiki/GettingStarted Jul 11 16:25:25 zecke, you maen like the simple familiar one? Jul 11 16:25:28 mean* Jul 11 16:25:38 koen: ok, but what's the sense of putting a tutorial...? Jul 11 16:25:54 YoG: what tutorial? Jul 11 16:26:02 gettingstarted is correct Jul 11 16:26:05 koen: just a sec... Jul 11 16:26:27 koen: http://www.openembedded.org/screencasts/org.openembedded.htm Jul 11 16:26:47 Here's the (shortened) offending line, fwiw: "install -m 0644 ...tslib-0.0cvs20060102-r39/ts-bb.data.getVar("ZKERNEL_VERSION", d, 1).conf ...image/usr/share/tslib/ts.conf" Jul 11 16:26:50 zecke: your screencast is broken Jul 11 16:27:16 goxboxlive: no, overrides don't apply to include/require Jul 11 16:27:33 goxboxlive: despite that, we want our qte to work with USB keyboards out of the box Jul 11 16:27:39 koen, zecke: this tutorial is a great idea, pitty that it doesn't work.... Jul 11 16:27:40 so we need to fix it. Jul 11 16:27:48 mickey|thesis: machine specific anti-patch Jul 11 16:27:58 koen: nah. no more hacks to qte Jul 11 16:29:11 koen: why? Jul 11 16:29:27 wrong url Jul 11 16:29:35 for the snapshot Jul 11 16:30:31 nice hostname btw :P Jul 11 16:31:00 which snapshot url? Jul 11 16:31:05 koen, I'm pretty sure that all quotes match fine. Jul 11 16:31:23 OE.db Jul 11 16:31:42 ah, I scrolled up Jul 11 16:32:09 what is the right url? Jul 11 16:32:20 YoG, openembedded.org Jul 11 16:32:24 YoG: see GettingStarted :) Jul 11 16:32:30 ok ok... Jul 11 16:32:30 YoG: I don't remember the right url Jul 11 16:33:15 just for the record: wget http://www.openembedded.org/snapshots/OE.db.bz2 Jul 11 16:33:27 thanx Jul 11 16:35:34 * emte hates xauth Jul 11 16:36:24 Can anyone tell me why '(' is an unexpected token? Jul 11 16:36:29 bronson: usually it are stray quotes that trigger that error, but it could be something else Jul 11 16:36:46 Is getVar("ZKERNEL_VERSION", d, 1) supposed to be in this file? Jul 11 16:39:15 Bash is definitely complaining about the getVar... Is that supposed to be expanded before this file is generated? Jul 11 16:39:33 Specificially, the "getVar(" Jul 11 16:41:49 sounds like something pollutes your shell script with python Jul 11 16:42:10 interesting choice of words Jul 11 16:42:19 heh Jul 11 16:42:24 coming from mickeyl 'python' lauer Jul 11 16:42:33 that's right Jul 11 16:42:40 the only excuse is being a non-native speaker Jul 11 16:42:48 s/pollutes/mixes/ Jul 11 16:44:35 so getVar is not supposed to be in this file? Any ideas on how to track this further? Jul 11 16:46:07 pastebin and look into your local.conf file Jul 11 16:47:39 tslib_cvs.bb: Jul 11 16:47:39 install -m 0644 ${WORKDIR}/ts-${ZKERNEL_VERSION}.conf ${D}${datadir}/tslib/ts.conf Jul 11 16:47:50 sounds like your bitbake is broken if that mangles to the line you pasted Jul 11 16:48:49 mickey|thesis: aehm Jul 11 16:49:58 no? install -m 0644 ...tslib-0.0cvs20060102-r39/ts-bb.data.getVar("ZKERNEL_VERSION", d, 1).conf looks pretty broken to me in shell :) Jul 11 16:50:49 getVar() is a python function is it not? Jul 11 16:50:57 mickey|thesis: well, if ZKERNEL_VERSION expands to that? Jul 11 16:51:23 zecke: hmm... interesting theory. in that case, something else would be broken Jul 11 16:51:31 emte: yeah. Jul 11 16:52:02 koen: now you need to upload the patch? Jul 11 16:53:18 zecke: which patch? Jul 11 16:53:20 mickey|thesis: should ZKERNEL_VERSION bes something? Jul 11 16:53:26 koen: NWFPE Jul 11 16:53:37 to OE? Jul 11 16:53:37 zecke: in .oz35x it's supposed to be "2.4" or "2.6" Jul 11 16:53:44 koen: to the patchtracker Jul 11 16:53:45 in .dev this variable no longer exists Jul 11 16:53:47 in .dev as well Jul 11 16:53:52 zecke: no idea Jul 11 16:54:05 bronson|build: which version did you set? Jul 11 16:54:48 mickey|thesis: http://www.linuxdevices.com/files/misc/CerfBoard-270-display-linux.jpg Jul 11 16:55:16 zecke: coool :D Jul 11 16:55:34 looks like OpenSIMpad Jul 11 16:56:22 mickey|thesis: the answer from sisihaus made me happy Jul 11 16:56:32 zecke: did it? hmm Jul 11 16:56:40 it was nice to see anyone supporting us, yeah Jul 11 16:56:48 but it didn't bring anything forward or does it? Jul 11 16:56:54 mickey|thesis: it doesn't Jul 11 16:57:05 isn't he working @ TT nowadays? Jul 11 16:57:13 mickey|thesis: yupp Jul 11 16:57:22 mickey|thesis: he is doing scribe Jul 11 16:57:31 full screen hw iM? Jul 11 16:57:50 mickey|thesis: no, Scribe the QTextWidget, Layouting, etc... Jul 11 16:57:53 oh right Jul 11 16:58:06 ~lart namespaces Jul 11 16:58:06 * ibot calls namespaces on the phone ... the lights are on but nobody's home Jul 11 16:58:19 any shaped pages with flowing text ... Jul 11 16:59:02 it would be about time that a couple of Oslo guys complain about Brisbane's way of handling things Jul 11 16:59:29 mickey|thesis: they are mostly unaware Jul 11 16:59:34 :( Jul 11 17:00:02 mickey|thesis: "TT supports Free Software, they hired me" Jul 11 17:00:10 mickey|thesis: is quite common Jul 11 17:00:40 mickey|thesis: it looks like this common Free Software liking didn't swap over to Brisbane Jul 11 17:00:59 lpotter: please correct me if I'm wrong, it is not meant to be FUD Jul 11 17:01:45 prayer-wheel-mantra Jul 11 17:01:47 lpotter: it is like HURD. We have a small window to make it happen. If we (emphasis>we) fail, Qtopia will be obsolete Jul 11 17:01:50 won't say anything else Jul 11 17:02:15 mickey|thesis: Most of them don't want to know that such things happen within their company Jul 11 17:02:21 mickey|thesis: they simply can't believe Jul 11 17:02:42 i remember. even hogne was surprised we didn't already have the qtopia 2 source code Jul 11 17:02:57 * mickey|thesis ponders whether to really buy stocks Jul 11 17:03:15 mickey|thesis: well, did you see woglindes SUSE buy out plan? Jul 11 17:03:20 mickey|thesis: we could adept it Jul 11 17:03:23 zecke: yeah. a good plan. Jul 11 17:04:17 hmmm? Jul 11 17:04:23 and hi * Jul 11 17:04:31 fear not. we are young enough. at least you are. we have some more time to make things move. Jul 11 17:04:35 :D Jul 11 17:04:55 Sorry -- unexpected phone call. Jul 11 17:05:06 * mickey|thesis orders Panzarotti Jul 11 17:05:10 zecke, which version? You mean kernel version? Jul 11 17:05:32 bronson|build: this variable, is that set to anything? Jul 11 17:05:45 zecke: yeah, I'm now registering with rmk and all Jul 11 17:05:53 mickey|bbl: EFL++, but this drug called Qt is just too good Jul 11 17:06:10 zecke: unbelievable how much work it is to get a one-liner into the kernel Jul 11 17:06:17 koen: yes :) Jul 11 17:06:26 koen: now you need to accept the faq, enable cookies Jul 11 17:06:35 koen: then don't forget the Signed-Off lines... Jul 11 17:06:37 zecke, right now KERNEL_VERSION is commented out... I thought machine/collie would set that. I'll look deeper. Jul 11 17:07:03 bronson|build: it might do, I'm clueless in this regard Jul 11 17:07:17 zecke: I have the signed off lines, but thunderbird messes up my patch Jul 11 17:07:18 more clueful than me though! Jul 11 17:07:34 zecke: but anyway, I need to wait for rmk to ack my account for the patch tracker Jul 11 17:08:43 zecke, mickey|bbl: make qt bindings for efl :) Jul 11 17:09:26 efl? Jul 11 17:09:44 zecke: if TT's claims are true then Qt/E is the way to go with an own interface on top. Forget about Linux standard mobile phone UI, that time frame is gone. We will have to build something original and that will increase fragmentation, but the best system will probably survive. Jul 11 17:09:58 s/probably/hopefully/ Jul 11 17:10:13 mickey|bbl: oh well Jul 11 17:10:13 ~efl Jul 11 17:10:16 zecke, explicitly setting KERNEL_VERSION="2.4" didn't help. :( Jul 11 17:10:18 i guess efl is the Enlightenment Foundation Libraries. See http://enlightenment.org/index.php?id=18 The core libraries of the EFL are: Evas, Imlib2, Ecore, EDB, EET, Embryo, Edje, Esmart, Etox, and EWL. Jul 11 17:10:55 ibot: no, efl is the Enlightenment Foundation Libraries. See http://enlightenment.org/index.php?id=18 The core libraries of the EFL are: Evas, Ecore, EET, Embryo, Edje, Esmart, EWL, and ETK. Jul 11 17:10:58 mickey|bbl: okay Jul 11 17:11:22 bronson|build: did you omit the 'Z' Jul 11 17:11:22 mickey|bbl: well, we could start using this MOTO GSM 07.11 code and create a phoneserver on top Jul 11 17:11:22 mickey|bbl: no imlib2? Jul 11 17:11:52 imlib2 is no longer a necessity Jul 11 17:11:56 it helps though Jul 11 17:12:18 the size of qte is distracting though Jul 11 17:12:49 zecke, intentionally. KERNEL_VERSION is in build.conf. I'll try to manually set ZKERNEL_VERSION but that just seems wrong somehow... Jul 11 17:12:54 * mickey|bbl picks up the Panzarotti Jul 11 17:16:07 zecke, manually setting ZKERNEL_VERSION fails too. Jul 11 17:16:31 oh well Jul 11 17:16:40 then spot for other errors in your local.conf :} Jul 11 17:16:58 bronson|build: which version of bitbake do you use? Jul 11 17:17:21 zecke, 1.4.2 Jul 11 17:17:32 that is fine :) Jul 11 17:18:07 hmm did the ISP's nameserver crash :} Jul 11 17:19:04 Can anyone pastebin a local.conf that works for oz? Jul 11 17:19:48 I've been over this file 10 times and it's starting to make me crosseyed...\ Jul 11 17:24:02 for collie? Jul 11 17:25:20 zecke, for any machine. I'm happy to get it working, then adapt it to collie. Jul 11 17:25:31 I'd just like something to finish without error! Jul 11 17:25:43 bronson|build: the one from the earlier screencast fails? Jul 11 17:26:07 er... earlier screencast? Jul 11 17:26:24 bronson|build: http://www.openembedded.org/screencasts/org.openembedded.htm Jul 11 17:27:04 * bronson|build gives it a look... Jul 11 17:27:27 Drat. Flash. Jul 11 17:28:04 bronson|build: ;) Jul 11 17:49:14 hi Jul 11 17:49:33 I'm trying to write a bb package that pulls some source from svn Jul 11 17:49:43 the url is svn://svn.freevo.org/kaa/trunk Jul 11 17:50:01 and I'm trying like: SRC_URI = "svn://svn.freevo.org/kaa;module=trunk" Jul 11 17:50:02 tkp: see the bitbake manual Jul 11 17:50:24 it downloads it, but then complains that setup.py cant be found Jul 11 17:50:43 tkp: so svn is a red herring? Jul 11 17:51:01 tkp: you might need to set S properly Jul 11 17:51:02 seems to pull the entire repo rather than just the trunk Jul 11 17:51:20 I have a kaa dir, with branches, tags and trunk Jul 11 17:53:25 ah hang on.. forgot to bitbake -c clean :) Jul 11 17:53:54 I fucking hate Windows XP Jul 11 17:53:54 tkp: see the bitbake manual Jul 11 17:53:54 tkp: we might not have much documentation, but the documentation that has been written Jul 11 17:53:54 tkp: needs to be read and correct Jul 11 17:54:34 tkp: and then you likely need to set S Jul 11 17:54:43 I set S = "${WORKDIR}/${SRCNAME}-${PV}" Jul 11 17:54:47 and inherit distutils Jul 11 17:55:02 yet it's calling: Jul 11 17:55:05 /home/tom/stv/oe-stv/build/tmp/staging/x86_64-linux/bin/python setup.py build Jul 11 18:00:14 seems to be ignoring S Jul 11 18:00:22 or I'm doing it wrong Jul 11 18:01:17 tkp: well, from where is it calling setup.py is the question Jul 11 18:01:17 tkp: and well, look into the tmp dir Jul 11 18:01:21 try looking at the work dir Jul 11 18:01:46 hmm. has thr dirs in it: Jul 11 18:01:50 kaa-1.0/ temp/ trunk/ Jul 11 18:02:13 * koen suspects it will look in kaa-1.0 Jul 11 18:02:26 so S = "${WORKDIR}/trunk" Jul 11 18:02:29 that ones empty Jul 11 18:02:33 all the sources are in trunk Jul 11 18:02:43 right Jul 11 18:03:08 then suprise, suprise, your S is wrong Jul 11 18:03:16 yup Jul 11 18:03:57 I hate windows... Jul 11 18:04:10 cygwin is finding the files, for the rest of window the dir is empty Jul 11 18:04:13 don't we all! Jul 11 18:04:43 ~lart indows Jul 11 18:04:43 * ibot turns indows into a lifesized tux doll Jul 11 18:04:50 oops Jul 11 18:05:13 take that you stupid indows! Jul 11 18:06:10 ah, CosmicPenguin ... wanted to ask a couple of questions about the bootimg stuff Jul 11 18:06:33 ok Jul 11 18:06:39 you have 6 minutes or so Jul 11 18:06:39 you know if it's possible to reimage a live system with imgloader Jul 11 18:06:51 its probably not a good idea currently Jul 11 18:07:02 I tried, and it seems to die at about 98% on the main partition Jul 11 18:07:09 imgloader walks over everything Jul 11 18:07:17 so by the time you get to where your image is stored, you're probably going to be in trouble Jul 11 18:07:22 right Jul 11 18:07:34 you know any other ways to image a live system like that Jul 11 18:07:47 ie. to provide firmware-type updates Jul 11 18:08:02 The best way to do it would be to sync everything, remount as readonly, put the image on a temp filesystem and then do it from there Jul 11 18:08:06 03koen 07org.oe.documentation * rab21466d... 10/usermanual/usermanual.xml: usermanual.xml: add stubs Jul 11 18:08:22 imgloader just wasn't designed for live system updates - though any other method you try will have the same problems Jul 11 18:08:29 hmm Jul 11 18:08:40 My favorite way to do that sort of thing is with an initrd Jul 11 18:09:04 an initrd... I'll look into into it Jul 11 18:09:28 imgloader (or a command line friendly clone) would be perfect in that case Jul 11 18:09:50 not too hot on initrd's Jul 11 18:09:55 * tkp need to do some reading Jul 11 18:09:55 in fact, imgloader has built in network support for just that situation Jul 11 18:10:07 anyway, I have to go Jul 11 18:10:13 cool Jul 11 18:10:21 damn, a new day and I mixed up CoreDump and CP again :} Jul 11 18:10:35 btw, zecke: get anyware further with dosfstools? Jul 11 18:10:42 tkp: no :} Jul 11 18:11:03 tkp: week started, uni+'research' project occupies time again Jul 11 18:11:15 ahh... np :) Jul 11 18:11:29 tkp: I really wonder. This is fedora core? Jul 11 18:11:37 yeah. 5 Jul 11 18:11:51 tkp: could you search me a SRC RPM (or whatever source packages are called nowadays) Jul 11 18:13:07 ~lart cisco training Jul 11 18:13:07 * ibot slams cisco training against a large cement Tux Jul 11 18:13:09 http://download.fedora.redhat.com/pub/fedora/linux/core/updates/5/x86_64/dosfstools-2.11-5.FC5.x86_64.rpm Jul 11 18:13:34 tkp: that is the source code? Jul 11 18:13:53 umm.. hmm Jul 11 18:13:57 no, prolly not Jul 11 18:14:09 OT: Anyone got some expierence with Fujitsu Siemens Desktop Systems? Jul 11 18:14:35 http://mirror.web-ster.com/fedora/core/5/source/SRPMS/dosfstools-2.11-4.2.src.rpm Jul 11 18:14:36 ? Jul 11 18:15:18 tkp: yes Jul 11 18:15:24 cool Jul 11 18:22:43 tkp: need to build that 'rpm' thing first Jul 11 18:41:09 zecke, looks like DISTRO="openzaurus-3.5.4.1" (instead of openzaurus-3.5.4) fixes it. Jul 11 18:41:16 Well, "fixes" it. :) Jul 11 18:54:39 bronson, you could use the new setup as well with DISTRO and DISTRO_VERSION Jul 11 18:58:49 bronson|build: hehe Jul 11 18:59:01 bronson|build: do openzaurus-3.5.4.conf doesn't exist? Jul 11 19:03:59 zecke, both exist. Jul 11 19:04:07 zecke: I didnt see any answer from tronical Jul 11 19:05:23 lpotter: you were on the CC Jul 11 19:05:28 agree with you Holger. Even a message would be better than nothing at all. Jul 11 19:05:32 eek Jul 11 19:06:10 zecke, and from 3.5.4.1, this appears to be the important line: ZKERNEL_VERSION_collie = "2.4" Jul 11 19:13:57 ~seen RP Jul 11 19:14:10 rp is currently on #gpe (11h 47m 43s) #oe (11h 47m 43s) #openzaurus (11h 47m 43s) #handhelds.org (11h 47m 43s), last said: 'hrw: I guess so, maybe cc the pcmcia maintainer'. Jul 11 19:15:01 RP: Any luck finding som e time to mess with poodle audio? Jul 11 19:22:09 ~mornings Jul 11 19:22:10 Mornings MUST be destroyed! (see also http://www.destroymornings.com/) Jul 11 19:24:03 hi all Jul 11 19:42:22 RP: http://dominion.kabel.utwente.nl/koen/cms/working-native-eabi-toolchain Jul 11 19:54:47 ~eabi Jul 11 20:06:56 koen have u 5 minutes to answer some (simple) questions about eabi ? Jul 11 20:07:15 sure Jul 11 20:07:16 hey gremlin[it], how goes? Jul 11 20:07:22 I hope I have the answers :) Jul 11 20:07:37 weel i start read arm doc about eabi ... what i uderstand is : Jul 11 20:08:03 gremlin[it]: most usefull stuff is in http://wiki.debian.org/ArmEabiPort (if I'm being un-intelligible) Jul 11 20:08:11 koen: you should start a business on EABI consulting Jul 11 20:08:26 1) is (just) a standard and well define way on how to prosuce executables, stared library and pass parameters to/from function calls Jul 11 20:08:37 koen: "Will EABI be able to order pizza" "No, that is out of scope, but BitBake can do" Jul 11 20:09:06 ooo Jul 11 20:09:13 i make bitbake be pretty :) Jul 11 20:09:21 2) i suppose this involve (only) compiler, dinamic loader (ld.so) and kernel Jul 11 20:09:52 gremlin[it]: and it standardizes what floatingpoint format to use Jul 11 20:09:55 2b) kernel is involved about systems calls and load/execute of executables ... Jul 11 20:10:29 so basically it involve libc + gcc + kernel .. right ? Jul 11 20:11:40 gremlin[it]: binutils Jul 11 20:12:13 zecke yes ... all involver in creation of executables / shader libraries Jul 11 20:12:39 now all seem more easy :) ... Jul 11 20:12:58 emte ... lot of work on ep93xx platform :) Jul 11 20:15:25 as there is already that topic about libc ... which one should i prefere on arm? g-,uc-,diet-libc or newlib? Jul 11 20:17:40 mr_nice_slacker: glibc is the right thing for most people Jul 11 20:18:46 koen:ok Jul 11 20:27:06 cbrake: yes. your feedback made me start looking outside of the bbfiles/bbclass themselves, and i discovered that my BBPATH was borked Jul 11 20:29:03 for anyone following my saga, the problem with my libsplaudio-codecs package getting the native (and incompatible) libstdc++ was a result of the speex.m4 macro needing to be fed --with options, and it was giving me a bad -L Jul 11 20:29:45 ain't crosscompiling fun? Jul 11 20:29:47 as it happens, i know that I am the first OE/bb consumer of the speex.m4 macro, since i added a patch to bugzilla to fix the staging of that file Jul 11 20:30:13 koen: if by "fun" you mean "challenging", yes :) Jul 11 20:31:09 build systems like bitbake are even more "fun" :) Jul 11 20:31:17 yup Jul 11 20:31:31 but once things are fixed, they work for lots of platforms Jul 11 20:31:46 Did the xerces-c problem from last week get resolved? Jul 11 20:35:59 zwelch: Remarkably, the version of binutils chosen by BB on the other computer in work was...wait for it... 2.16. All with no prompting by me. Jul 11 20:36:49 Kerwood: well, i had it even weirder. my one computer chose 2.16 to start... then it decided one day (with no prompting by me) to upgrade to 2.17 Jul 11 20:37:03 i was lucky that i noticed it and knew to think to regress it Jul 11 20:37:17 zwelch: indeed. I'm waiting for the same to happen to me. Jul 11 20:37:22 if i had missed that action, i would have been scratching my head Jul 11 20:46:41 koen: you asked me to look at the gtkmm/glademm packages, yeah? Jul 11 20:47:20 * zwelch discovered today that his new irc client's scrollback was one order of magnitude too small Jul 11 20:48:00 zwelch: yes Jul 11 20:48:09 did you have anything particular in mind? Jul 11 20:48:47 moreover, is there a general purpose strategy in place for bumping and regression testing packages like these? Jul 11 20:48:48 maybe update them to newer (better?) versions Jul 11 20:48:58 but basically, just check if they work Jul 11 20:49:08 it's been a loooooong time since I used minisip Jul 11 20:49:15 (and hence gtkmm) Jul 11 20:49:29 well, i can assure you the existing bb files wil not build from svn Jul 11 20:49:41 at the very least, they moved the URL of their repository recently Jul 11 20:50:36 fortunately for them, my work on minisplat's build system will allow them to basically copy/paste Jul 11 20:51:16 I was actually meaning released versions Jul 11 20:51:30 *_svn.bb are a nightmare to maintain Jul 11 20:51:38 why? Jul 11 20:51:58 * zwelch has a number of questions in that regard Jul 11 20:52:43 bitrot Jul 11 20:52:43 i suppose i just don't see it, because i just spent a few days designing a set of classes that remove all duplication from my release and svn bbfiles Jul 11 20:52:56 so, that will not happen to my files Jul 11 20:53:06 and after that people with stupid firewalls Jul 11 20:53:16 however, i do not plan to maintain more than a single release version in my repository either Jul 11 20:53:32 anyway Jul 11 20:53:37 * koen -> sleep now Jul 11 20:53:42 hehe Jul 11 20:54:03 that stupid firewall comment is rather invalid Jul 11 20:54:30 if they cant get snv then most likly they cant get to the monotone repo anyway Jul 11 20:54:45 i mean, his point about released files is somewhat valid though; really, users should be using those, and not the svn version Jul 11 20:55:07 but i love using the svn versions. Jul 11 20:55:09 more and more projects are not using "releases" Jul 11 20:55:37 don't you mean "less and less projects are adhering to formal engineering practices"? Jul 11 20:55:56 * zwelch does not see the lack of releases as an evolution in engineering Jul 11 20:56:11 no, i mean it the other way around Jul 11 20:56:25 maintaining CVS snapshot versions of e17 was a nightmare Jul 11 20:56:28 hence I removed them Jul 11 20:56:45 right, having to snapshot is bad Jul 11 20:56:45 trying to figure out a date which works, ficing issues, etc....:-P Jul 11 20:56:58 there are more and more projects that have no structured development or professional guidance Jul 11 20:57:02 * zwelch is still frustrated by ffmpeg for this... has that project released 0.5.0 yet? :) Jul 11 20:57:09 using the "release" versions (which are essentially a snapshot of CVS by the e17 devs) is far easier Jul 11 20:57:30 emte: if that's the case, they should not really be used by other projects that do have such Jul 11 20:57:52 thats a rather idealistic view Jul 11 20:58:20 (unless those disciplined individuals are willing to bring their skills to the table... where they will find their ideas are not necessarily welcomed) Jul 11 21:00:17 emte: there need to be standards of development quality Jul 11 21:00:59 i'm not _even_ going to try to define what those might be, but i hope you can ack that such must exist at some nebulous level Jul 11 21:01:31 for professional stuff, but you have to remember most open projects start as someone casually playing around on weekends, and then suddely a huge amount of people gain interest and start using it ... Jul 11 21:01:52 suddenly* Jul 11 21:01:54 * Kerwood is headin' home...l8r... Jul 11 21:03:59 it's definitely a complex issue, but i don't think it's one that we should carry on discussing here :) Jul 11 21:04:21 everyone else went to sleep anyway :) Jul 11 21:05:37 hmm, this discussion mirrors the source of my depression Jul 11 21:06:36 hrm? Jul 11 21:06:55 your having issues maintaining your project? Jul 11 21:06:57 I am working on an open source software defined radio framework Jul 11 21:07:05 at my school Jul 11 21:07:12 ah Jul 11 21:07:19 and people are starting to get interested Jul 11 21:07:29 and I have no idea how we will actually get work done Jul 11 21:07:36 rf stuff gets touchy and expensive Jul 11 21:08:03 of course, we have the problem of only half the people being paid to do work on it are actually doing work that needs to be done Jul 11 21:08:26 that sounds typical Jul 11 21:08:42 yeah Jul 11 21:08:50 I end up doing an ass load of work Jul 11 21:09:08 assloads are the worse loads Jul 11 21:09:16 yeah Jul 11 21:09:27 you could push for a commit based eval ... Jul 11 21:09:34 i know a few places that use it Jul 11 21:09:37 I looked at the trac log from the week I was on vacation Jul 11 21:09:49 next to nothing actually happened Jul 11 21:11:07 Crofton: you are not alone Jul 11 21:11:09 are you lead? Jul 11 21:11:18 not really Jul 11 21:11:35 effectively yes Jul 11 21:11:41 but not really Jul 11 21:12:07 well eventually the lead will get called in and then things come down Jul 11 21:12:23 the lead is not really effective Jul 11 21:12:51 * Crofton hates academia right now Jul 11 21:12:55 if its funded, they will when issues come down Jul 11 21:13:14 they will just yell at me Jul 11 21:13:20 this is a problem that reaches into every kind organization; it sounds like you have basic management problems Jul 11 21:13:26 hahhahahaha Jul 11 21:13:32 exactly Jul 11 21:13:50 if you do not already understand how good management is suppose to work, i would recommend reading a good book on the subject Jul 11 21:14:10 at this point, we need management Jul 11 21:14:17 specifically, the most empowered worker is one that understands how the chain of command manages authority, responsibility, and accountability Jul 11 21:14:18 (slightly different topic) would anyone like to test e-images on some Zaurii? Jul 11 21:14:32 someone actually beating people and making sure they do the work they are paid for Jul 11 21:14:36 heh Jul 11 21:14:58 then, you can document your co-workers (or leaders) failures in those areas, and present them as simple facts to their superiors - let those people draw their own conclusions Jul 11 21:15:33 if your the only one checking in code, its pretty documented who does work Jul 11 21:15:37 of course this would annoy the professors Jul 11 21:15:52 they hate people that rock the boat Jul 11 21:16:26 doesnt everyone? Jul 11 21:16:47 I just got back from vacation and the amount of work I need to do has set in Jul 11 21:16:52 emte: if code was the only metric to successful contributions, i am guessing that this wouldn't be as big a problem ;) Jul 11 21:16:57 I think I will go home and drink Jul 11 21:17:18 go home and drink tea Jul 11 21:17:23 beer Jul 11 21:17:30 no, tea Jul 11 21:17:30 crappy American beer Jul 11 21:17:36 if it were me, i'd just plug in my basic time and let the slacckers suffer Jul 11 21:17:42 Crofton: where are you located? Jul 11 21:17:48 nice japanese green tea Jul 11 21:17:51 Virginia, USA Jul 11 21:17:52 if you were in Colorado, you would be drinking excellent american beer Jul 11 21:17:58 ... Jul 11 21:18:02 I frinnk crappy beer by choice :) Jul 11 21:18:07 there is no such thing as good american beer Jul 11 21:18:11 * zwelch is in Oregon, with the very best American microbrews ;) Jul 11 21:18:12 right Jul 11 21:18:17 cosmicpenguin: if I was in colorado, I might be drinking crappy american tea, but I doubt I'd be drinking beer, american, crappy or otherwise Jul 11 21:18:27 you know what - every single person from Germany that has come over here has said exactly that Jul 11 21:18:36 and every single one has gone home with a completely different attitude Jul 11 21:18:36 tea totaller Jul 11 21:18:53 hah - Oregon wishes they were Colorado Jul 11 21:19:06 emte: i agree with you, actually Jul 11 21:19:18 there is no such thing as good american beer, because there is no such thing as good beer Jul 11 21:19:23 I sure hope tomorrows TdF stage is good Jul 11 21:19:29 i lived in NW NM, and i visited Colorado frequently. Colorado has nothing on Oregon :) Jul 11 21:19:33 i can agree with that Jul 11 21:19:40 ale person myself Jul 11 21:19:40 colorado may have better water Jul 11 21:19:55 I am thinking of moving to Northern Italy and becoming a bum Jul 11 21:20:04 good idea Jul 11 21:20:04 raduga: mmm, that's probably a close call, but you may be right Jul 11 21:20:24 raduga: altough, i'll wager we have more of it ;) Jul 11 21:20:28 http://www.flickr.com/photos/32615155@N00/187334403/ Jul 11 21:20:33 probably Jul 11 21:20:57 zwelch: in california we have just as much water as you do, but ours (like yours) is mostly salt-water Jul 11 21:21:18 of the freshwater, I really can't compare for quality or quantity Jul 11 21:21:29 raduga: well, we get quit a bit of the precipitous form here Jul 11 21:22:20 my personal "green energy" design contribution would only work in areas like the Pacific Northwest, where solar is a sub-optimal solution: microturbines in rain gutter downspouts :) Jul 11 21:22:44 * zwelch cackles wildly, as a proper mad scientist must Jul 11 21:23:27 we keep trying to put through current-trap turbines but the eco-lobbiests keep getting them vetoed Jul 11 21:23:56 "current-trap"? Jul 11 21:24:28 a large vent in a current flow that funnels the water through a turbine Jul 11 21:24:47 efficient and sustainable energy Jul 11 21:25:06 ah, is that what they call what i was talking about? Jul 11 21:25:20 it has many names Jul 11 21:25:33 traps are ment more for ocean and rivers Jul 11 21:25:58 this is getting off-topic; i'd like to know more, so pop into #minisplat Jul 11 21:28:03 what is minisplat Jul 11 21:40:51 minisplat.org Jul 11 21:40:56 @Crofton Jul 11 22:03:26 later Jul 11 22:03:50 g'night Jul 11 22:06:37 03mickeyl 07org.oe.dev * r8bf57f7a... 10/packages/ (3 files in 2 dirs): add python-pyephem. PyEphem provides scientific-grade astronomical computations for the Python programming language. Jul 11 22:11:13 arg. i see why koen asked me to look at gtkmm. just trying to build it crashes bitbake Jul 11 22:12:27 ewww Jul 11 22:12:40 http://rafb.net/paste/results/RfljKk52.html Jul 11 22:14:18 * zwelch backs up and tries to provide more details Jul 11 22:18:33 okay, here is the output from 'bitbake -v gtkmm', with the current .dev branch and bitbake-1.5.0+trunk: http://rafb.net/paste/results/0Yi0By54.html Jul 11 22:20:09 * zwelch notes that he's just fishing for tips to continue debugging the problem Jul 11 22:21:30 here is the output with -DDD too: http://rafb.net/paste/results/cNidxT48.html Jul 11 22:27:06 hmm. seems there's a bug in my new pastebin patch, eh? :) Jul 11 22:28:03 03rpurdie 07org.oe.dev * rc5b3cb5a... 10/packages/linux/ (4 files in 3 dirs): linux-oz-2.6.17 + 2.6.17+git: Add poodle audio patch. Gives working playback via headphones and working mic support. The speaker will probably also work when we fix the corgi speaker bug. Jul 11 22:35:14 ~hail RP Jul 11 22:35:16 * ibot bows down to RP and chants, "I'M NOT WORTHY!!" Jul 11 22:35:57 RP: I just saw your checkin...:) I will need to give this a whirl tonight when I get home :) Jul 11 22:38:33 ibot: you are worthy Jul 11 22:38:34 raduga: what are you talking about? Jul 11 22:38:41 ibot: you Jul 11 22:38:43 hvontres|work: It basically works perfectly. I still can't believe it :) Jul 11 22:38:43 i'm a broken bot Jul 11 22:38:56 ibot: you are a beautiful bot Jul 11 22:38:57 I think you lost me on that one, raduga Jul 11 22:39:01 ~botsnack Jul 11 22:39:01 raduga: aw, gee Jul 11 22:39:10 ibot: enjoy Jul 11 22:39:17 hvontres|work: The internal speaker doesn't but that problem is shared with corgi and should get fixed in the next ASoC verison Jul 11 22:40:27 RP: No biggie, I use my poodle mostly as my "MyPod", so head phones are just great: Jul 11 22:41:11 RP: Thanks for all your hard work on this..I was getting worried that my poor poodle would become abandoned.. Jul 11 22:43:12 RP: I think it works so well because you got the harder of the two wokring first..The Corgi implementation with all those differnet modes looks like a real bear..:) Jul 11 22:44:00 hvontres|work: Thankfully the two are very similar :) Jul 11 22:44:20 hvontres|work: Where does this put poodle 2.6 support - are we missing anything else major? Jul 11 22:46:10 RP: I am not sure..The USB networking seems to work. CoreDump's status page showed some sd card errors, but i have not seen them yet. I have not tried any of the other usb features either. Jul 11 22:46:49 But I think we are close to being ready for an -rc type image, I think. Jul 11 22:47:25 I will try to get a bootstrap image going tonight when I get home. I might know more by tomorow. Jul 11 22:47:50 hvontres|work: The SD errors would be shared by all the other pxa devices as its all common code Jul 11 22:48:13 We need to sort out a default mixer config for Zaurusd but I can take care of that... Jul 11 22:48:24 RP: Ok. Did those ever get fixed? Jul 11 22:48:59 Did what get fixed? Jul 11 22:49:16 RP: The SD issues? Jul 11 22:49:30 I've no idea - I'm not aware of any outstanding ones Jul 11 22:51:03 RP: Maybe they already got fixed. CoreDump's page is getting a bit old. Maybe we should let him know about the audio stuff too. He seems to be really good at fixing userland stuff quickly :) Jul 11 22:51:30 There's a good reason to fixup userland now :) Jul 11 22:52:04 OE userland will need fixing to add sound support but its all plain sailing now the driver itself works :) Jul 11 22:53:01 RP: I think that is true for the clamshells as well, right? Jul 11 22:53:11 no, they just work Jul 11 22:53:32 but for poodle we mainly copy their setup :) Jul 11 22:54:17 anyhow, I'm off to bed now - happy :) Jul 11 22:54:37 RP: Night :) Jul 12 00:12:55 is anyone hosting/targeting OE builds on/to any BSDs? Jul 12 00:13:16 * zwelch is currently working on helping someone port minisplat to openbsd Jul 12 00:17:43 there has been BSD rumbling in the past but nothing solid Jul 12 00:19:42 zwelch: Magically everything is purring on my OE environment at home now, thanks to you Jul 12 00:21:31 Kerwood_: glad to hear it Jul 12 00:22:42 CosmicPenguin: well, in ask about BSD and MinGW32, I'm probing the boundaries of the current implementation's realization of its design goals Jul 12 00:23:03 Bitbake itself can handle it Jul 12 00:23:09 certainly, these things ought to be possible... but they are definitely non-trivial endeavors Jul 12 00:23:15 yea, it's more of an OE thing Jul 12 00:23:26 the only question is how much work it will be to come up with the set of tools needed on BSD as well Jul 12 00:23:29 and really, at a certain point, a project for another metadata repo Jul 12 00:23:38 Well - at least a branch Jul 12 00:23:46 yeah, i guess that makes the most sense Jul 12 00:24:08 or basic protection flags to handle the host OS Jul 12 00:24:43 so, offhand, i am interested in the following branches: bsd, mingw32, and darwin Jul 12 00:25:59 bootstrapping is a bear though :) Jul 12 00:27:25 Kerwood_: incidentally, speaking of magic fixes... my problems with gtk+ have suddenly disappeared, although i think i updated my branch copy at some point before it started working again.... i'm not going to ask why, i'm just going to keep going and pretend it never broke in the first place Jul 12 00:28:48 Precisely so. There seem to be "non-technical" events going on... Jul 12 01:32:25 gn **** ENDING LOGGING AT Wed Jul 12 02:59:57 2006