**** BEGIN LOGGING AT Wed Apr 19 09:59:56 2006 Apr 19 10:03:02 This one 'vim /usr/src/universal//deploy/ipk/Package' gives me an empty file. And 'ls -la | grep bootstrap' gives me nothing. Apr 19 10:07:42 Sorry it gave me alot of packages, but nothing about bootstrap Apr 19 10:11:40 goxboxlive, did you build task-bootstrap.bb yet ? Apr 19 10:12:18 ipkg complains in that way as it "installs" the builded packages directly into the rootfs Apr 19 10:12:34 and it misses task-bootstrap.bb Apr 19 10:12:38 goxboxlive: did you try to manually build it? Apr 19 10:12:58 goxboxlive: have you in your log some reference to building task-bootstrap? Apr 19 10:13:09 goxboxlive: e.g Not creating empty package for task-bootstrap Apr 19 10:15:16 "The related bug has been reported to the GNU make maintainers in 1988 but Apr 19 10:15:16 as the bug has not yet been fixed, it seems that GNU make is unmaintained :-( Apr 19 10:15:16 A working highly portable make program is at ftp://ftp.berlios.de/pub/smake" Apr 19 10:15:20 I love Joerg Apr 19 10:20:13 when i run this, everything works just fine: bitbake -b ../org.openembedded.oz354fam083/packages/meta/task-bootstrap.bb Apr 19 10:20:37 goxboxlive: clean it, make it again Apr 19 10:20:52 goxboxlive: see if you have a task-bootstrap.ipk Apr 19 10:22:02 But when i run this one, it failes: bitbake -b ../org.openembedded.oz354fam083/packages/linux/xanadux-un-2.6_2.6.12.bb se http://pastebin.com/668936 Apr 19 10:22:23 goxboxlive: relation to the ipkg problem? Apr 19 10:24:09 The task-bootstrap says it builds, but it is not in //deploy/ipk directory. Apr 19 10:25:04 it seems like your xanadu/HTC configuration is simply wrong... Apr 19 10:25:06 NOTE: exceptions.NameError:name 'linux_module_packages' is not defined while evaluating: Apr 19 10:25:11 as it sais in line 6 Apr 19 10:25:53 yes i know i can see that, but have no clue how to fix it. Do you :-/ Apr 19 10:26:21 goxboxlive: pastebin your local.conf, xanadux-un...bb and the machine configuration Apr 19 10:27:14 OK I'll do that. Apr 19 10:27:52 goxboxlive: and remove that trailing slash too many from your path... Apr 19 10:28:01 goxboxlive: //deploy looks like a depot (svk,p4) Apr 19 10:28:48 lthis is local.conf http://pastebin.com/668940 Apr 19 10:29:39 Question: # Don't use symlinks in in the path to avoid problems Apr 19 10:29:40 TMPDIR = /usr/src/universal/ Apr 19 10:29:45 is that a symlink? Apr 19 10:30:41 no i dont have any symlinks. htcuniversal.conf http://pastebin.com/668941 Apr 19 10:32:13 This is myeThis is my xanadu...... http://pastebin.com/668943 Apr 19 10:33:02 *dinner* Apr 19 10:33:06 wait a bit plase Apr 19 10:34:03 yes of course. Bon apetite Apr 19 10:38:11 goxboxlive, btw maybe i will get a universal soon Apr 19 10:38:15 morning Apr 19 10:39:25 Rob_w : I think you should. It's a cool gadget ( but way to expensive) . Apr 19 10:40:09 goxboxlive, yeah but i did some work for a friend and he now looks into getting me that cheap Apr 19 10:41:08 ANd it will be so great when the linux port is finished. Apr 19 10:46:04 yeah ,that is the goal Apr 19 10:46:08 brb Apr 19 10:59:40 hi all Apr 19 11:06:51 koen|work: hey Apr 19 11:06:52 re Apr 19 11:06:58 morning all Apr 19 11:07:06 zecke: Good news on the release :) Apr 19 11:07:28 hey zecke & RP Apr 19 11:07:58 RP: I'm glad we were able to end the instability that quickly Apr 19 11:08:16 goxboxlive: just a rule of thumb Apr 19 11:08:24 zecke: Yes, its nice to have the resolved Apr 19 11:08:26 goxboxlive: replace the 'include' with require Apr 19 11:08:42 in htcuniversal.conf? Apr 19 11:08:52 yes Apr 19 11:08:57 zecke: Is there an easy way to disable COW in bitbake btw? I'm wondering what happens to speed if we do... Apr 19 11:09:34 RP: and to do what instead? Apr 19 11:09:37 RP: deepcopy? Apr 19 11:09:44 zecke: yes Apr 19 11:09:53 hi Apr 19 11:10:00 hi Bernardo Apr 19 11:10:01 RP: will be slower Apr 19 11:10:09 RP: you can test that easily Apr 19 11:10:13 zecke: ok, I just wondered Apr 19 11:10:39 zecke: I also wondered if you'd tried storing content and the other var flags in separate dicts? Apr 19 11:11:03 RP: no that :) Apr 19 11:11:12 That wasn't clear - content in one dict, all the other flags in another Apr 19 11:11:23 zecke: I was thinking it might make lookups faster Apr 19 11:11:39 RP: I can prepare patch for you (to dict_smart.py) Apr 19 11:11:49 RP: you will change the createCopy method to do a deepcopy Apr 19 11:12:03 RP: and maybe remove some code, or just inherit dict Apr 19 11:12:35 zecke: I haven't time to experiment now but I think I know what I need to do to test it - I'll give it a go at some point... Apr 19 11:13:22 zecke: The main question is whether COW for speed or memory usage? Apr 19 11:13:34 I don't think we'd have the memory usage problem now... Apr 19 11:13:50 RP: both Apr 19 11:14:06 RP: accessing can get a bit slower Apr 19 11:14:26 RP: but you save the deepcopy Apr 19 11:14:33 but creation is faster, right Apr 19 11:14:44 RP: see the 'override helpers' Apr 19 11:14:52 RP: with cow you will save another minute there Apr 19 11:16:28 zecke: that didn't help either. You meant the include conf/machine/include/handheld-common.conf to require conf/machine/include/handheld-common.conf and the same for the X-scale? Apr 19 11:17:00 yes Apr 19 11:18:07 Well it didnt help. This is so strange, for a couple of days a go it worked, but now it refuse. Apr 19 11:18:26 goxboxlive: okay, we know that these file exists Apr 19 11:18:35 goxboxlive: the kernel file itself looks correct as well Apr 19 11:19:11 goxboxlive: just make sure, using 'S' in that way is not too smart Apr 19 11:20:29 What do u mean about a caption s ? Apr 19 11:21:06 goxboxlive: you want to use your local directory as SOURCE directory Apr 19 11:21:15 goxboxlive: this looks somehow abusive ;) Apr 19 11:22:00 later all Apr 19 11:22:23 Yes , because i do some changes to it, and it har worked before. But one thing. Shall i make the defconfig and modules my self or will bitbake do that for me? Apr 19 11:24:03 ( in the cvs linux kernel) Apr 19 11:24:26 goxboxlive: it will handle .config Apr 19 11:24:34 goxboxlive: and it will package all modules you built Apr 19 11:26:34 goxboxlive: Update bitbake, update OE Apr 19 11:26:50 goxboxlive: then run bitbake -DDD .../task-bootstrap.bb -cclean Apr 19 11:26:52 goxboxlive: then run bitbake -DDD .../task-bootstrap.bb Apr 19 11:26:58 goxboxlive: pastebin the logs Apr 19 11:31:16 Zecke: I'll try now with the S pointing to cvs kernel. OE is updated to day, bitbake is not. I'll do that when the cvs is downloaded. I realy appreciate your help here. Apr 19 11:34:51 goxboxlive: you are welcome Apr 19 11:56:31 mickeyl: morning Apr 19 11:56:38 HI TheMasterMind1 Apr 19 11:56:59 zecke: congrats for the release. HEAD is open for dirty hacking again? Apr 19 12:02:35 mickeyl: I'm not sure Apr 19 12:02:50 mickeyl: maybe we will have branches/dirty-hacking/zecke Apr 19 12:02:54 mickeyl: maybe we will have branches/dirty-hacking/mickeyl Apr 19 12:02:58 mickeyl: and then merge it to trunk Apr 19 12:03:59 xxf86dgaproto Apr 19 12:04:06 what should hat be? Apr 19 12:04:11 meta-sdl is horrible broken Apr 19 12:04:23 only one x i think Apr 19 12:07:52 TheMasterMind1: no these packages actually exist Apr 19 12:07:58 http://packages.debian.org/unstable/libdevel/libxxf86dga-dev Apr 19 12:08:01 03freyther 07org.oe.dev * rea299360... 10/packages/ (frodo/frodo_4.2.bb uae/e-uae_0.8.27.bb): (log message trimmed) Apr 19 12:08:01 packages/frodo_4.2.bb: Apr 19 12:08:01 Use the new CVS home of Frodo. He left uni and now Apr 19 12:08:01 is in a power plant... The package should be renamed Apr 19 12:08:01 _cvs.bb as well Apr 19 12:08:02 packages/uae/e-uae_0.8.27.bb: Apr 19 12:08:04 0.8.27 is a old release, the SRC_URI changed to match Apr 19 12:09:39 TheMasterMind1: right, thanks Apr 19 12:12:48 Zecke: I'back, i tried the bitbake -DDD .../task-bootstrap.bb (after -cclean) alot of things happend, but the only error i got whas this: ERROR: Nothing provides dependency ../org.openembedded.oz354fam083/packages/meta/task-bootstrap.bb Apr 19 12:13:19 Shouldn't i also have a '-b' in there? Apr 19 12:15:30 goxboxlive: sure ;) Apr 19 12:18:03 zecke: heres the output : http://pastebin.com/669070 Apr 19 12:19:24 goxboxlive: clean it first ;) Apr 19 12:19:45 like this: bitbake -DDD .../task-bootstrap.bb -cclean Apr 19 12:19:56 yes Apr 19 12:20:00 -b as well Apr 19 12:20:48 zecke: well, it's not _that_ dirty or is it? isn't HEAD for current development? :) Apr 19 12:21:12 ok like this then 'bitbake -b -DDD .../task-bootstrap.bb -cclean Apr 19 12:21:14 mickeyl: well, what do you want to do? Apr 19 12:22:24 mickeyl: BitBake should reach a higher level Apr 19 12:22:44 mickeyl: print out the "Request of HEAD change" paper Apr 19 12:22:47 mickeyl: and fax it to me Apr 19 12:22:53 (filled out) Apr 19 12:23:16 mickeyl: I will handle your request according to our 'Request of HEAD change'-process Apr 19 12:23:50 zecke: this gave me no errors: bitbake -DDDb ../org.openembedded.oz354fam083/packages/meta/task-bootstrap.bb What do u want me to do now? Apr 19 12:24:03 mickeyl: seriously: more radical changes should go to a branch Apr 19 12:24:12 goxboxlive: I want to see the output Apr 19 12:24:24 goxboxlive: clean it, build it, show me the logs (pastebin) Apr 19 12:24:42 goxboxlive: why are you not running what zecke told you to? Apr 19 12:26:22 zecke: hmm... isn't that backwards? I thought branches were for stable changes, HEAD for unstable ones? (potentially) Apr 19 12:26:32 I'm find with whatever, of course Apr 19 12:26:36 s/find/fine/ Apr 19 12:26:47 mickeyl: sure we have a stable branch Apr 19 12:26:58 mickeyl: which is absolutely stable ;) Apr 19 12:27:07 mickeyl: and we have HEAD (multiple ones) Apr 19 12:27:15 hey we should use monotone for bitbake as well Apr 19 12:28:11 mickeyl: Currently, I simply want to avoid bigger breakage Apr 19 12:28:56 goxboxlive: oh well, wait a second Apr 19 12:28:58 zecke here: http://pastebin.com/669081 Apr 19 12:30:08 XorA: goxboxlive: why are you not running what zecke told you to? Because we i should build it, and i mistace it missed and -b to build. Apr 19 12:30:14 goxboxlive: do you generate ipackages at all? Apr 19 12:30:41 mickeyl: did I kill monotone on vanille? Apr 19 12:31:16 it is repsonding now Apr 19 12:31:23 zecke: I used to, but it looks like it doesnt now. Apr 19 12:31:56 goxboxlive: this explains why it can't install a task-bootstrap ;) Apr 19 12:32:46 zecke: I c, but do i generate them? Whay to edit? Apr 19 12:32:55 goxboxlive: conf/local.conf Apr 19 12:36:23 zecke: Ok, i had a look there, but there where nothing about ipkg at all. What should i write in the local.conf to generate ipkg? Apr 19 12:36:56 03freyther 07org.oe.dev * r15d10e01... 10/packages/xlibs/ (libxxf86dga_1.0.0.bb libxxf86vm_1.0.0.bb): Apr 19 12:36:56 packages/xlibs: Apr 19 12:36:56 Remove one 'x' to have something that is buildable. Apr 19 12:36:56 The CVS versions do not suffer this problem. This fix Apr 19 12:36:56 makes it possible to build snes9x again. Apr 19 12:37:02 goxboxlive: see local.conf.sample Apr 19 12:37:05 goxboxlive: grep for ipk Apr 19 12:51:02 zecke: It's going forward. The DISTRO= was wrong in my local.conf. But now it ask's for this. : NOTE: exceptions.NameError:name 'linux_module_packages' is not defined while evaluating: How do i define it? Apr 19 12:55:50 hello ! Apr 19 12:57:46 i am trying to modify ntp.bb in order to install ntpdate in rc2.d infront of ntpd (which is S20) Apr 19 12:57:56 i am using something like this Apr 19 12:58:14 pkg_postinst_ntpdate_owmnr() { Apr 19 12:58:15 update-rc.d -s ntpdate defaults 19 Apr 19 12:58:15 } Apr 19 12:58:31 but it does not seem to work Apr 19 12:58:57 thats not the way to do it Apr 19 12:59:08 look at the ntpd package and see how it does it Apr 19 12:59:32 you inherit update-rc.d and use INITSCRIPT_NAME Apr 19 13:00:58 its allready inherited in the bb. Apr 19 13:01:17 i used the Nylon as an example Apr 19 13:02:03 http://pastebin.com/669135 Apr 19 13:02:29 it's defined like that inside the ntp.bb Apr 19 13:09:40 hmmm strange.... If i copy the packet over to the target machine and install it, it does install the init script Apr 19 13:09:58 the rootfs though in /tmp/rootfs does not Apr 19 13:10:14 containing it Apr 19 13:20:16 zecke: There are also others who have the same problem as i have for building to their blueangel. So i guess it's something with the branch or something. Maybe it's not my configs at all. Apr 19 13:31:05 goxboxlive: where can one get the branch? Apr 19 13:44:38 zecke what do u mean? do u mean org.openembedded.oz354fam083? Apr 19 13:46:12 goxboxlive: what do I need to build it/produce this issue? Apr 19 13:47:25 zecke: Was there anything on my list that you wanted kept? Apr 19 13:48:22 treke|home: haven't looked too carefully Apr 19 13:50:35 doesnt look like it's anything interesting. Just the big groups of cvs builds Apr 19 13:50:53 hrw|gone: hey Apr 19 13:51:21 hrw|gone: did you move the packages already? Apr 19 13:51:27 treke|home: did you move the packages already? Apr 19 13:51:42 zecke: took them offline. haven't deleted them yet Apr 19 13:51:46 treke|home: otherwise I could do a -cfetch worl and see what is mirrored Apr 19 13:51:49 treke|home: ah okay Apr 19 13:54:42 treke|home: could you wait another 12 hours? Apr 19 13:54:59 sure Apr 19 13:54:59 treke|home: I want to do a -c fetch world and look for missing stuff Apr 19 13:55:10 hrw|gone: where do you keep the source for OpenZaurus.org? Apr 19 13:55:23 I just went over the list again and so far it looks really uninteresting Apr 19 13:55:34 all that's on it are the cvs snapshots Apr 19 13:55:35 treke|home: I think hrw if away for a few days Apr 19 13:55:51 RP: ah okay Apr 19 13:55:57 honeymoon already? Apr 19 13:56:20 zecke: I don't think so, not yet... Apr 19 13:56:53 we should consider making a joined present for him Apr 19 14:01:22 he said he be back tomorrow if I recall right Apr 19 14:02:47 zecke: thats a good plan Apr 19 14:03:31 hey folks Apr 19 14:03:36 i need a bit of arm asm help Apr 19 14:03:42 hey mickeyl Apr 19 14:03:45 could you provide a translation for http://oe.pastebin.com/669250 ? Apr 19 14:05:21 mickeyl: to ARM? Apr 19 14:05:34 mickeyl: no atomic increment in armv4, armv5 IIRC Apr 19 14:05:54 mickeyl: specially no lock operation ;) Apr 19 14:05:56 hmm Apr 19 14:06:11 so i will just subtitute that with ++i for arm ? Apr 19 14:06:16 (and the like) Apr 19 14:06:28 mickeyl: all you can rely on SA1100 and PXA (ARMv4 >) is swp Apr 19 14:06:37 mickeyl: take a look at Qt Apr 19 14:08:28 arm gcc defines __ARM__ ? Apr 19 14:09:22 i think so, or __arm__ Apr 19 14:09:27 right __arm__, thanks Apr 19 14:09:31 * mickeyl removes that x86 asm Apr 19 14:09:32 feierabend Apr 19 14:09:37 Prost Apr 19 14:16:37 mickeyl: You could look at the atomic ops code in the kernel Apr 19 14:17:21 RP: yeah, will do that Apr 19 14:17:28 ~lart x86isms Apr 19 14:17:28 * ibot hurls dozens of incontinent, insomniac, hungry kittens with tiny little razor-sharp claws and a wide variety of contagious intestinal parasites at x86isms Apr 19 14:18:41 mickeyl: Sadly, as zecke said, there is no locking on arm so the kernel versions use irq_save which isn't an option in userspace Apr 19 14:20:12 mickeyl: well use swp ;) Apr 19 14:20:19 is somebody compile something against libsdl-x11? Apr 19 14:20:37 mickeyl: static /* olatile */ char my_lock = 0; Apr 19 14:20:50 urhgs Apr 19 14:20:52 mickeyl: and use 'swp' to simulate a test_and_lock.. Apr 19 14:20:52 more x86isms Apr 19 14:20:54 #define jassertfalse { asm("int $3"); } Apr 19 14:21:01 *sigh* Apr 19 14:21:30 mickeyl: and use something like while(my_swp(&var, ~0)!=0); Apr 19 14:21:34 mickeyl: now your increment Apr 19 14:21:42 my_lock = 0; Apr 19 14:21:43 return Apr 19 14:21:54 *nod* Apr 19 14:22:28 I thin Dijkstra invented it Apr 19 14:22:35 though I have not seen a proof for that yet Apr 19 14:22:52 finally, cya Apr 19 14:30:03 just those 3 files i posted at pastebin. Thats all Apr 19 14:58:46 mickeyl: you alive? Apr 19 14:58:55 or somebody with commit access to OE? Apr 19 15:01:43 yep, what's up? Apr 19 15:01:56 does OE have the compat libs in it? Apr 19 15:02:11 i seem to remeber it does Apr 19 15:02:16 yes, they're downloaded from somewhere Apr 19 15:03:02 http://openzaurus.org/mirror/oz-compat_0.5.tar.gz Apr 19 15:03:15 the .bb is in sharp-binary-only/ Apr 19 15:03:32 okay can you add a test into the file so that if you are compiling with SOFTFLOAT they will die horribly Apr 19 15:03:52 preferably insulting the person Apr 19 15:04:10 i could, but you know that they're not entirely useless even with soft float Apr 19 15:05:13 i know - but the user hassle is not worth it - they do use floating point operations and the fact they don't segfault at all is a mirical Apr 19 15:06:41 well, since they're not actually getting compiled, wouldn't it be better to tell the distro maintainers not to include them in feeds instead of having the .bb bail out? Apr 19 15:07:26 mickeyl: no - the oz-compat is NOT SUPPORTED and should NOT be used when using soft-float Apr 19 15:08:35 uhm... yes, but i don't see how preventing them to build will have anything to do with them actually getting used. Apr 19 15:09:05 i can do that if it makes you happy, but I seriously question the effectiveness of such a step Apr 19 15:09:10 *shrug* Apr 19 15:09:25 mickeyl: well it stops accidently inclusion in SOFT-FLOAT distros Apr 19 15:10:11 mithro: in fact, i don't think they're getting accidently included... Apr 19 15:10:21 i think hrw knows exactly about the problems Apr 19 15:10:38 i know it won't stop people from using it - but it makes sense anyway for things which are known to not work on SOFTFLOAT not actually be able to compiled for SOFTFLOAT Apr 19 15:11:02 ok, for the casual bb user Apr 19 15:11:19 bbl, wandering homme Apr 19 15:11:22 home, as well Apr 19 15:11:34 mickeyl: okay thanks Apr 19 15:11:45 just sent a rant to the list about the libraries Apr 19 15:12:27 there was a reason I removed them from my site in the first place :P Apr 19 15:14:26 really the only app I ever used them for was opera, once that new swanking konq-e makes it into OZ I doubt people want to run opera anymore Apr 19 15:15:12 whats the new konq-e like? Apr 19 15:16:10 mithro: no idea personally, I run GPE these days Apr 19 15:16:22 whats the browser of choice on GPE? Apr 19 15:16:32 mithro: for me, links/firefox Apr 19 15:16:46 mithro: or for quick vews of sites gpe-mini-browser Apr 19 15:16:49 firefox run okay on the zaurus? Apr 19 15:17:02 mithro: a touch slow, but usable on 400Mhz machines Apr 19 15:18:11 i wonder what ephiphany would be like Apr 19 15:18:43 feel free to make a .bb for me to test :-D Apr 19 15:19:49 C parser of bitbake is more probably more importnat Apr 19 15:21:52 bye Apr 19 15:38:08 time to do my first ever monotone checkout Apr 19 15:40:22 how intrepid of you Apr 19 15:41:48 Required software page doesnt mention a version number for monotone, is the protocol pretty stable? Apr 19 15:43:03 I think you need a fairly recent version. The one in debian unstable seems to be OK. Apr 19 15:43:08 0.24 or 0.25 or something Apr 19 15:43:21 hehe. hopefully 0.23 works Apr 19 15:47:05 ggilbert: I suspect you need 0.24 Apr 19 15:47:48 hehehe. Guess it's backport time :p Apr 19 16:42:12 hi all Apr 19 16:42:23 ~stab gprs Apr 19 16:42:29 * ibot runs at gprs with an origami Swiss Army knife, and inflicts a nasty paper cut. Apr 19 16:49:19 hi koen Apr 19 16:49:53 hey mickeyl Apr 19 16:50:20 my connection will be gone as soon as the refridgerator starts cooling again... Apr 19 16:51:36 freeze! Apr 19 16:52:53 heh Apr 19 17:04:05 XorA: do you use links2 in the console or with -g? Apr 19 17:04:14 pgas: -g Apr 19 17:05:09 after pre14 it allows for copy pasting ...I'm a links2 lover :) Apr 19 17:23:55 Woo. I finally got a gpe-image build to complete. Apr 19 17:24:15 Necronom: congrats Apr 19 17:26:42 koen: e-image-core biulds with angstrom distro ^_^ Apr 19 17:26:56 koen: of course it doesn't even work for OZ in .dev right now, so..... Apr 19 17:27:08 JustinP: yay! Apr 19 17:28:04 I think Xfbdev is segfaulting for some reason.....don't know why yet Apr 19 17:29:21 I'm trying to build e-image now Apr 19 17:31:15 do we have any OE devs that can read/write CJK? Apr 19 17:32:11 hmmm. I could have swordn someone lived in korea Apr 19 17:33:04 It seems asia (and china in specific) is booming with embedded startups Apr 19 17:37:49 oh wait, that was an opie developer Apr 19 17:41:09 is it safe to ignore errors like this when parsing: ERROR: Error Method already seen: get_glibc_fpu_setting in' glibc_2.2.5.bb' now in 'glibc_2.4.bb' Apr 19 17:41:25 mreimer: it is safe to ignore them Apr 19 17:41:30 thanks koen Apr 19 17:42:45 hi mreimer Apr 19 17:42:48 hi pb_ Apr 19 17:43:07 pb_: apmd working well for you now? I hope to try your patch when I get a chance Apr 19 17:43:41 mreimer: seems to be working fairly well, though I haven't been using it very intensively Apr 19 17:44:10 I hope we can get that change into 0.8.4-rc2, though, so that it gets a bit more exposure. Apr 19 17:44:18 agreed Apr 19 17:44:45 the other two major items I have for -rc2 are the hh41 kernel and that ipkg patch for l10n. Apr 19 17:44:53 * koen is still disappointed that 0.8.4 won't ship with full support for h2200, hx4700 and h6300 Apr 19 17:45:03 koen: what's lacking? Apr 19 17:45:49 mreimer: they were demoted to "technology demos" instead of "supported" Apr 19 17:45:59 really? why? Apr 19 17:46:21 no idea Apr 19 17:46:41 what's missing? Apr 19 17:46:43 hw support? Apr 19 17:46:46 nothing that I know of Apr 19 17:46:47 sw support? Apr 19 17:46:54 everything works on h2200 Apr 19 17:46:59 just documentation and testing, as far as I know. Apr 19 17:46:59 someone taking the time to test and put everything together neatly? Apr 19 17:47:00 almost everything on hx4700 Apr 19 17:47:04 pb: hm Apr 19 17:47:12 pb: i guess that's sortof good :/ Apr 19 17:47:13 I've no idea what the situation is with h6300, but the other two platforms seem to be fairly well supported Apr 19 17:47:21 the only h2200 problem I know of is flaky battery monitoring on some h2200 boards Apr 19 17:47:36 last time I checked, though, there was no install manual for hx4700. Apr 19 17:47:52 there's that wiki page Apr 19 17:48:00 which is all h2200 has too Apr 19 17:48:12 yeah, that's probably not quite good enough for "supported" Apr 19 17:48:31 particularly if it still says "DANGER, THESE INSTRUCTIONS ARE UNTESTED" or something at the top, which the hx4700 one used to Apr 19 17:48:41 heh, h2200 still has that too :-) Apr 19 17:48:44 time to remove it. Apr 19 17:49:06 what else needs to be added to those docs to consider them "supported"? Apr 19 17:49:09 mreimer: udev is too old, the kernel probably won't compile in the branch, etc Apr 19 17:49:28 we're clear to upgrade udev Apr 19 17:49:44 what's the "kernel probably won't compile in the branch" problem? Apr 19 17:49:59 iirc I instructed it to build the _cvs kernel Apr 19 17:50:09 mreimer: not sure, I'd need to review them. The ideal thing would be to have hx4700 and h2200 support integrated into the existing manual, but the next best thing would be a clear step-by-step procedure on a webpage. Apr 19 17:50:22 if it compiles it will be 2.6.ancient, otherwise you end up with a half broken 2.6.16 Apr 19 17:50:52 koen: I have my h2200.conf patched locally to use 2.6.15-hh0, which I'd recommend for 0.8.4 Apr 19 17:51:15 pb_: I'm willing to do the work to get h2200 instructions ready to integrate into the existing manual Apr 19 17:51:22 mreimer: excellent Apr 19 17:51:27 just tell me what needs to happen other than copy and paste Apr 19 17:52:40 I would guess hx4700 hw support will be complete in the next month or two. wifi is almost there, and power consumption will likely be fixed very shortly, as people with documentation/schematics are involved now Apr 19 17:52:44 basically, start with http://familiar.handhelds.org/releases/v0.8.2/install/ and identify the parts that need changing for the h2200. Apr 19 17:52:44 hm. Apr 19 17:53:02 anyone here use either csound or timidity, on arm? Apr 19 17:53:04 depending on what the differences are, it might either make sense to merge the h2200 bits into those pages, or create parallel pages with the h2200 steps on Apr 19 17:54:29 pb_: steps 4 and 5 will be completely different, since we don't use bootldr Apr 19 17:54:31 evening Apr 19 17:54:40 hey reenoo Apr 19 17:54:47 hey mreimer Apr 19 17:55:04 didn't get very far building angstrom: /opt/oe/build-angstrom/tmp/work/quilt-native-0.42-r0/temp/run.do_patchcmd.938: line 493: unexpected EOF while looking for matching ``' Apr 19 17:55:38 mreimer: okay. in that case, it sounds like h2200 should have a separate workflow for the installation. Apr 19 17:56:06 in which case, we basically just need a new set of webpages in the same approximate style. Apr 19 17:56:32 I'd like to retain the "walkthrough" kind of approach, since this seems to have worked well in the past. Apr 19 17:56:40 hi reenoo Apr 19 17:56:47 hey pb_ Apr 19 17:57:12 pb_: ok. we also have multiple installation options: rootfs on CF, SD, internal flash, plus one can either boot from haret (for the dual-booters) or from LAB installed in flash Apr 19 17:57:54 mreimer: okay Apr 19 17:58:25 mreimer: I guess it isn't strictly necessary to document all the permutations, just the most common or useful one(s) Apr 19 17:58:46 for example, if you decided that the only "supported" approach for 0.8.4 was to install into internal flash with LAB, that would be fine Apr 19 17:59:08 pb_: yeah, for time I'll probably have to do that Apr 19 17:59:25 pb_: but the speed increase of not having to use jffs2 tempts me to document the others Apr 19 17:59:48 I suppose I can always footnote the wiki page Apr 19 17:59:55 yeah Apr 19 18:01:45 by the way, does anybody know if netfilter has a "random drop" match option? I thought it did but I can't find one now. Apr 19 18:02:20 * JustinP hugs altboot Apr 19 18:02:24 need to simulate various kinds of packet loss conditions for testing Apr 19 18:08:45 reenoo: I have a few h2200 patches for 0.8.4 to test, then I'll commit to .oz354fam083 Apr 19 18:09:23 mreimer: excellent. please go ahead. Apr 19 18:10:12 reenoo: lamikr approves of using udev 0.84, so I don't think there is any reason not to include it in fam 0.8.4 Apr 19 18:11:42 mreimer: yeah, I saw that, just haven't got around to replying yet Apr 19 18:12:02 pb_: is the familiar installation manual in CVS somewhere? Apr 19 18:12:32 mreimer: yeah, I think so. let me check Apr 19 18:13:16 it's in familiar/website Apr 19 18:14:03 thanks pb_ Apr 19 18:14:03 heh Apr 19 18:14:08 not according to france Apr 19 18:14:51 hmm, cvs only has 0.8.0 Apr 19 18:15:21 ah, so it does Apr 19 18:15:33 that's a bit weird, but I guess it hasn't changed much since 0.8.0 Apr 19 18:15:39 let me diff the 0.8.0 version against 0.8.2 Apr 19 18:16:53 mreimer: http://handhelds.org/~pb/install.diff is the differences Apr 19 18:17:01 thanks pb_ Apr 19 18:17:16 hi from Brazil Apr 19 18:17:37 hail zecke Apr 19 18:17:48 hey Apr 19 18:17:54 hey obergix[work] Apr 19 18:17:55 obergix[work]: FISL? Apr 19 18:18:28 hey zecke Apr 19 18:19:20 koen: yep Apr 19 18:19:38 koen: error compiling gcc-4.1.0 (not cross) Apr 19 18:19:48 at least 3 state governors in VIP room now... lucky me... libre software is really hype here Apr 19 18:20:03 goxboxlive: hey Apr 19 18:20:21 goxboxlive: could you send me all the files you are using and which are not in OE? Apr 19 18:20:28 JustinP: there's a patch for that in bugzilla Apr 19 18:21:08 JustinP: http://bugs.openembedded.org/show_bug.cgi?id=851 Apr 19 18:22:46 JustinP: if the lack of response on the bugs stays the same I'll apply the patch in the weekend Apr 19 18:23:35 koen: ok. I'll apply it locally for now so I can keep building ^_^ Apr 19 18:24:22 * JustinP abhors pasted patches Apr 19 18:24:26 03rw 07org.oe.oz354fam083 * r4c4f092f... 10/packages/detect-stylus/ (8 files in 3 dirs): detect-stylus-nox: add variant of detect-stylus without X support Apr 19 18:25:03 zecke: seems you were too slow in forking that code Apr 19 18:25:27 pb_: maybe I control reenoos mind Apr 19 18:25:51 oh, hadn't thought of that Apr 19 18:26:13 maybe reenoo is just a script written in nickel? Apr 19 18:26:16 Hello Apr 19 18:26:26 just so long as you don't control _my_ mind... Apr 19 18:26:47 pb_: I have forked it, and had adequately bashings in the header Apr 19 18:26:57 pb_: though I didn't came around pushing it ;) Apr 19 18:27:39 Do you know any free software for data aquisition systems? I mean, a software that reads data from serial port (or another port) and presents some usefull features like graphs, logs and etc? Or anything that can read and send control signals back? Apr 19 18:28:00 fromvega: | :) Apr 19 18:28:13 zecke: | ? Apr 19 18:28:14 fromvega: GNU plot, occtave, microcom/cu Apr 19 18:28:20 zecke: hum.... Apr 19 18:30:38 zecke: is that I'm developing a simple data aquisition and control system, where I will read data from sensors and send back some commands for attached devices. I want to design a software for that, where the user can create some control routines without programming. Apr 19 18:32:03 zecke: what's occtave and microcom/cu ? Apr 19 18:32:21 fromvega: google Apr 19 18:32:28 ~google octave Apr 19 18:32:30 fromvega: cu -> get data from the serial Apr 19 18:32:59 zecke: it's oCtave or oCCtave? Apr 19 18:33:40 fromvega: what do you think? Apr 19 18:34:14 :) Apr 19 18:34:27 fromvega: I normally follow google's 'Did you mean...' Apr 19 18:34:40 zecke: does octave handle softfpa correctly yet? Apr 19 18:34:42 zecke: ok, is that you have written with double CC so I get confused. But google found some occtave Apr 19 18:34:55 hehe Apr 19 18:36:47 zecke: and how about microcom/cu ? google does not found much. Apr 19 18:37:05 that's microcom or cu Apr 19 18:37:10 fromvega: microcom or cu Apr 19 18:37:40 or minicom, even Apr 19 18:39:07 WTF is up with sf.net and gtk-webcore Apr 19 18:39:20 koen: have you done any gpe builds lately with angstrom? Apr 19 18:39:32 same thing that's always up with sourceforge's cvs tree, I guess Apr 19 18:39:35 koen: perhaps you have a copy of JavaScriptCore_cvs.sourceforge.net__20060212.tar.gz I could borrow? Apr 19 18:39:51 JustinP: I have a diff from pb_, but I can't build librsvg to test the change? Apr 19 18:39:55 I haven't been able to check out the jscore module for a while... Apr 19 18:39:58 you can use the ones from the familiar 0.8.3 source stash, probably Apr 19 18:39:59 JustinP: try the familiar oesources.org Apr 19 18:40:04 http://familiar.handhelds.org/source/ Apr 19 18:40:53 this version is only used if you're compiling with gcc4 Apr 19 18:40:59 zecke: Hi there, i am back. Yes can have those three files fromemen, but i have managed to build task-boostrap now. Struggling with the bootsrap-image because of this: http://pastebin.com/669816 Apr 19 18:41:53 pb_, zecke: yep, the one up there is the older version Apr 19 18:41:55 goxboxlive: well Apr 19 18:42:01 zecke: http://ewi546.ewi.utwente.nl/tmp/sources/ Apr 19 18:42:03 ehm Apr 19 18:42:05 JustinP: http://ewi546.ewi.utwente.nl/tmp/sources/ Apr 19 18:42:07 I need someone who is compiling with GCC4 to give me the snapshot... Apr 19 18:42:24 koen: thx much Apr 19 18:42:26 goxboxlive: well Apr 19 18:42:28 oh, right. the checkout version depends on your compiler version? that's kind of funky Apr 19 18:42:31 koen: will probably need that nrcit one as well Apr 19 18:42:38 goxboxlive: you build the stuff without packaging support ;) Apr 19 18:42:50 goxboxlive: this means even if you have built pciutils, usbutils Apr 19 18:42:59 goxboxlive: you do not have ipk packages ;) Apr 19 18:43:06 JustinP: all my sf.net checkouts are in that location now Apr 19 18:43:24 I can't wait for the next congres... Apr 19 18:43:31 I will mirror sf.net :} Apr 19 18:43:50 koen: cool , thanks :-) Apr 19 18:45:07 zecke: But, when i made the task-bootsrap it mad a whole lot of *.ipk's More than i can count in an hour :-) Apr 19 18:46:31 goxboxlive: it did? well, you can only install stuff into a ipk image, where you have ipks ;) Apr 19 18:48:49 Zecke': Yes i ran bitbake task-bootstrap and it got build : ls -la /usr/src/universal/deploy/ipk/task* Apr 19 18:48:49 -rw-r--r-- 1 jrs src 932 2006-04-19 17:52 /usr/src/universal/deploy/ipk/task-bootstrap_1.0-r25_htcuniversal.ipk Apr 19 18:49:10 goxboxlive: try to read Apr 19 18:49:18 goxboxlive: reading is awesome, it helps you out Apr 19 18:49:37 NOTE: package edje-native-0.5.0.025-r3: task do_fetch: started Apr 19 18:49:38 NOTE: Update cvs://anonymous@anoncvs.enlightenment.org/var/cvs/e;module=e17/libs/edje/m4;date=20060101 Apr 19 18:49:40 cvs [update aborted]: connect to thinktux.net(213.239.215.151):2401 failed: Connection refused Apr 19 18:49:44 hrm Apr 19 18:50:29 JustinP: shouldn't it be using the new cvs repo? Apr 19 18:50:29 | linux-hotplug: unsatisfied recommendation for pciutils Apr 19 18:50:29 | linux-hotplug: unsatisfied recommendation for usbutils Apr 19 18:50:35 no bugs though Apr 19 18:50:47 goxboxlive: kernel-module-htcuniversal-phone-2.6 kernel-module-htcuniversal-wlan-2.6 udev Apr 19 18:50:48 | zecke: Apr 19 18:50:57 | zecke: unsatisfied recommendation for bugs Apr 19 18:50:57 goxboxlive: there are no packages for that Apr 19 18:51:57 zecke: I know, but the starnge thing is that i have removed them from the htcuniversal.conf file. Apr 19 18:52:36 goxboxlive: and changing that file, changes the contents of the ipk? Apr 19 18:52:55 koen: yeah.... Apr 19 18:53:48 koen: strange thing, though: cvs://anonymous@anoncvs.enlightenment.org connect to thinktux.net Apr 19 18:54:23 aargh Apr 19 18:54:25 * koen can't read Apr 19 18:54:32 I noticed that earlier too Apr 19 18:54:56 I also looked in my sources dir and I had the tarballs for these....but they had a ":" after the host so they weren't picked up Apr 19 18:55:10 I really shouldn't be pulling that stuff from cvs but...::shrug:: Apr 19 18:55:26 zecke: Actually i have no idea. Maybe this is way to hig for me, actually i dont get it. But i'll try to build pciutils ,usbutils and udev now. Apr 19 18:56:21 koen: rm -rf CVS in your sources. reading Root from CVS always gets preferred Apr 19 18:56:23 bbl, tv Apr 19 18:56:54 ah.... Apr 19 18:58:46 goxboxlive: read again Apr 19 18:59:08 * koen rms Apr 19 18:59:11 goxboxlive: you change the DEPENDENCIES of a package, rebuild the package Apr 19 18:59:25 mickey|tv: and another point why cvs sucks donkey ass Apr 19 19:01:55 goxboxlive: rebuild task-bootstrap Apr 19 19:02:24 goxboxlive: your EXTRA_DEPENDS in htc...conf get put into the 'Depends line' of the package Apr 19 19:02:41 goxboxlive: once the ipk is built, bitbake will not rebuild it Apr 19 19:03:36 zecke: ok i'll try :-/ Apr 19 19:03:40 thx Apr 19 19:04:56 goxboxlive: bitbake will only rebuild a package if Apr 19 19:05:11 goxboxlive: the PV or PR changes Apr 19 19:08:09 hey Pigi Apr 19 19:08:48 hi koen ! Apr 19 19:09:00 pgas: new links-x11 version just for you Apr 19 19:09:21 03xora 07org.oe.dev * rbc59ba05... 10/packages/links/ (4 files): links-x11_2.1pre21.bb, links_2.1pre21.bb : new versions Apr 19 19:10:32 XorA: well, I hope for you too, I'm not using Oz atm ;) Apr 19 19:10:53 thanks anyway Apr 19 19:11:11 mickey|tv: what is the state of QtopiaCore in OE? Apr 19 19:11:12 pgas: whats your current poison then? Apr 19 19:11:25 pdaxrom of course :) Apr 19 19:11:42 ah, 2.4-crappix Apr 19 19:11:47 pgas: heh heh, Im sure with 10 secs of .conf file making you can get OE ti build packages Apr 19 19:12:26 XorA: every time we suggest that sash puts his fingers in his ears and goes "la la la la la la" Apr 19 19:13:00 koen: Im not suggesting sash use it, but as far as I recall, pdaXrom uses similar glibc and gcc to use and basic difference is the ipkg format Apr 19 19:13:15 yeah, note that he knows his build system from A to Z so it's understandable Apr 19 19:13:16 XorA: glibc 2.ancient Apr 19 19:13:16 koen: but it should be fairly simple to get OE to output packages for pdaX Apr 19 19:13:41 XorA: yeah, just assume_provided gcc and glibc and you should be good to go Apr 19 19:13:53 and some packages in the pdaxrom builder are missing from oe Apr 19 19:14:02 pgas: Oh, do tell Apr 19 19:14:04 (and vice versa of course_ Apr 19 19:14:13 * XorA likes missing packages Apr 19 19:14:25 I think more packages are missing from the builder than the reverse Apr 19 19:14:38 with 2 or 3 orders of magnitude Apr 19 19:15:20 reenoo: http://pastebin.com/669901 Apr 19 19:15:27 ("not in OZ feed" != "not in OE", fwiw) Apr 19 19:15:27 koen: don't you get it? Apr 19 19:15:41 koen: I know Apr 19 19:15:44 koen: people want to be reliable on one person alone, single point of failure Apr 19 19:15:46 zecke: no, I'm pretty thick Apr 19 19:15:48 koen: cope with that Apr 19 19:15:59 rely even... Apr 19 19:16:23 zecke: and demand $500/month Apr 19 19:18:45 koen do you have a problem with money? Apr 19 19:19:09 yeah, he spends it all :) Apr 19 19:20:10 Should bitbake boostrap-image work with the latest bitbake from svn? Apr 19 19:20:23 Crofton: depends on your machine/distro Apr 19 19:20:28 Crofton: but generally: yes Apr 19 19:20:30 Mine fails after parsing with the sanity check message Apr 19 19:20:34 hmm Apr 19 19:20:49 koen: do you know any issue? Apr 19 19:21:12 machine would be omap5912osk and distro is openomap Apr 19 19:21:31 zecke: issue with what? Apr 19 19:21:31 Crofton: and openomap.conf exists in conf/distro? Apr 19 19:21:36 yeah Apr 19 19:21:43 koen: building images with a new bitbake Apr 19 19:21:47 kergoth made it when I started fooling iththis stuff Apr 19 19:21:54 I suspect it is a copy if something Apr 19 19:21:59 zecke: I haven't found any yet Apr 19 19:22:01 njs: I have question regarding metaclasses? Apr 19 19:22:02 Crofton: I got the sanity check message as well, but then the second time I ran bitbake, I did not get it. Apr 19 19:22:13 ok Apr 19 19:22:17 zecke: or something went wrong with short term memory -> long term memory Apr 19 19:22:19 I'll try that :) Apr 19 19:22:31 now I am building Apr 19 19:22:55 I assume I should look into all the errors during parsing when I have free time ? Apr 19 19:23:32 no Apr 19 19:23:33 do it now! Apr 19 19:23:35 do it! Apr 19 19:24:07 heh Apr 19 19:24:18 overload/multi-tasking is occuring Apr 19 19:24:24 to many threads of execution Apr 19 19:24:54 wow, bitbake sure is fast now -- takes less than a second if the data is cached -- very nice :-) Apr 19 19:25:41 Yeah, its slick isn't it? Apr 19 19:26:05 opssss ... i update my monotone to 0.26 ... oe will upgrade to 0.26 ??? Apr 19 19:28:35 i got the message >>> mtn: misuse: wait for a leader to migrate project data, ... <<< Apr 19 19:28:50 later all Apr 19 19:29:19 gremlin[it]: I think you want to stay at 0.25 Apr 19 19:30:03 gremlin[it]: there is big changes between 0.25 and 0.26 which implies to completly rebuild the project history Apr 19 19:30:20 cbrake too late ... i upgrage monotone (easy to revert) and update the oe database (i suppose less easy to revert) :( :( :( Apr 19 19:30:39 gremlin[it]: :-( Apr 19 19:30:39 gremlin[it]: someone here told me OE will probably switch with the 0.27 release Apr 19 19:31:25 * cbrake makes a note to stick monotone at 0.25 in his gentoo portage config ... Apr 19 19:31:43 gremlin[it]: because we all need to switch together Apr 19 19:32:39 XorA: hey man , i built mplayer-atty apparently the story is different than oz's Apr 19 19:33:24 cyrilRomain yes i understand just asking if 0.25 -> 0.26 was a work in progess ... Apr 19 19:33:56 gremlin[it]: don't know... Apr 19 19:35:27 Is there a page somewhere describing the strong point of bitbake/openembedded ? Apr 19 19:35:37 s/point/points ? Apr 19 19:35:52 ie why one would want to use OE? Apr 19 19:36:35 when i'm missing some app i just build it with oe Apr 19 19:37:21 later all Apr 19 19:37:22 Because it has koen's personal guarantee of personal satisfaction or your money back Apr 19 19:39:57 evening Apr 19 19:40:37 pgas: not on the wiki in my knowledge, but I think a strong point is patches inclusion for openzaurus, a tool to create cross-toolchains, and of course cross-compiling Apr 19 19:40:40 RP: which config should I set for the LEDs? Apr 19 19:41:24 pgas, the need for OE marketing literature has been identified :) Apr 19 19:42:56 yeah, knowing what it does could be usefull to convince people to use it..patches inclusion, cross-toolchains and cross compiling is what I would expect of such a system Apr 19 19:44:40 03rw 07org.oe.oz354fam083 * r915bafbe... 10/packages/detect-stylus/ (3 files in 2 dirs): rename detect-stylus-nox to detect-tsdevice and make it not RCONFLICT detect-stylus any more Apr 19 19:44:50 03rw 07org.oe.oz354fam083 * r60982d22... 10/packages/tslib/ (7 files in 7 dirs): tslib: use detect-tsdevice rather than detect-stylus and fix dependencies Apr 19 19:45:09 CSMan: in what way? Apr 19 19:46:21 mreimer: hmm. isn't gpe-nmf 0.22 picked by default anyway? or do you just want it pinned to a specific version? Apr 19 19:46:32 cedric: I'll have a look. Apr 19 19:46:55 reenoo: didn't preferred-versions specify 0.21 before? Apr 19 19:47:26 XorA: it's not playing , like it gets stuck and cpu goes to 100% Apr 19 19:47:38 cedric: CONFIG_NEW_LEDS, CONFIG_LEDS_CLASS, CONFIG_LED_TRIGGERS, CONFIG_LED_TRIGGER* and CONFIG_LEDS_LOCOMO Apr 19 19:48:03 CSMan: did you resize video to 320x240 or 240x320 depending on which way your native framebuffer goes? Apr 19 19:48:07 mreimer: unless the parser is ignoring the leading #, no, doesn't seem like it Apr 19 19:48:21 XorA: i'm just trying to play an mp3 file which mpg123 can do without a problem Apr 19 19:48:23 CSMan: also you might want to try -ac mad, if mplayer-config didnt work for some reason Apr 19 19:48:38 RP: thx dude Apr 19 19:49:03 reenoo: d'oh. ok Apr 19 19:49:10 XorA: vc -mad worked just fine ! what codec was I using ? Apr 19 19:49:24 CSMan: the floating point on, mp3lib Apr 19 19:49:30 OMG Apr 19 19:49:55 XorA: what do you mean mplayer-config may have not worked? Apr 19 19:50:25 CSMan: Coredump added a mplayer-config.bb which install a config that is supposed to make -ac mad default Apr 19 19:50:52 CSMan: mplayer-common sorry Apr 19 19:51:08 XorA: so if i install mplayer-common that should do the job Apr 19 19:51:23 CSMan: mplayer-atty should Depend on it Apr 19 19:51:29 njs: wow metaclasses are sexy Apr 19 19:51:38 CSMan: as should mplayer Apr 19 19:52:06 i don't have it in my repository Apr 19 19:52:15 CSMan: you on a branch Apr 19 19:52:20 anyway, i'll just tweak the .mplayer/config Apr 19 19:52:31 XorA: .dev Apr 19 19:52:37 CSMan: yeah, thats what I have, my mplayer is so old :-) Apr 19 19:52:38 is that a branch ? Apr 19 19:52:45 CSMan: no thats mainline Apr 19 19:53:10 ah, i find bitbake convenient but a bit mixed up with all those branches Apr 19 19:54:13 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:14 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:14 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:21 yeah baby Apr 19 19:54:21 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:22 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:24 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:28 yeah give it to me Apr 19 19:54:31 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:32 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:36 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:40 WTF Apr 19 19:54:44 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:44 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:48 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:54:54 this is annoying Apr 19 19:55:02 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:03 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:03 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:08 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:09 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:13 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:17 what a loser Apr 19 19:55:19 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:21 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:25 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:27 kajon: come hack my server, please, i beg you Apr 19 19:55:30 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:33 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:37 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 19:55:53 bizarre Apr 19 19:56:02 thats not even effective spam Apr 19 19:56:04 Wheres a channel op when you need them? Apr 19 19:56:16 /ignore kajon ALL Apr 19 19:56:19 ehe, been like this all day Apr 19 19:56:21 works wonders Apr 19 19:57:00 is that one of those boot all the *doze users off irc hacks or something? Apr 19 19:57:04 You would think that the MOTD would actually tell us how to contact somebody in charge Apr 19 19:57:15 ChanServ claims OE is registered Apr 19 19:57:35 CosmicPenguin: try lilo failing all else Apr 19 19:57:40 heh Apr 19 19:57:56 I wonder who has ops here? Apr 19 19:57:57 CosmicCANUCK Apr 19 19:58:04 since I am kinda busy and might not be paying attention Apr 19 19:58:11 if that spamming idiot comes back let me know in /msg Apr 19 19:58:20 I can have my friends make his life very miserable Apr 19 19:58:22 dude - I wouldn't talk to you if you owned the joint Apr 19 19:58:29 :( Apr 19 19:59:26 CosmicPenguin: how should one understand the last sentence? Apr 19 19:59:49 not sure Apr 19 19:59:55 I just sort of typed it off the cuff Apr 19 20:00:40 * fishhead bites CosmicPenguin Apr 19 20:00:42 well, sounds like 'tension' Apr 19 20:00:43 meanie :( Apr 19 20:00:54 but well, I haven't said anything Apr 19 20:09:20 03lamikr 07org.oe.dev * r334808b0... 10/packages/openobex/ (files/libusb_crosscompile_check.patch openobex_1.2.bb): openobex: Openobex failed to compiler due to error in checking libusb.pc in the crosscompiler environment. Apr 19 20:14:03 RP: Registered led device: locomo:amber Apr 19 20:14:03 Registered led device: locomo:green Apr 19 20:17:09 RP: how can I test it? Apr 19 20:17:54 http://www.digitimes.com/bits_chips/a20060419PR204.html Apr 19 20:17:55 mmmm Apr 19 20:18:27 RP: echo 1 > brightness Apr 19 20:18:41 RP: works for both leds... :-) Apr 19 20:20:32 ~praise RP Apr 19 20:20:33 All hail RP! Apr 19 20:20:56 Hi Apr 19 20:21:10 nite all Apr 19 20:22:52 Any idea how to remove/prevent empty /etc/modules.conf file to appear to gpe-images automatically? After I removed that, the update-modules configured the /etc/modules file correctly in 084-prebuild in the first boot? Apr 19 20:26:54 lamikr: Which package does ipkg think it came from? Apr 19 20:27:29 pb_: Hmm, can I check that easily? Apr 19 20:27:35 "ipkg search /etc/modules.conf" Apr 19 20:27:44 or inspect the .list files in /usr/lib/ipkg/info by hand Apr 19 20:28:31 pb_: I will check that once I get my kernel again to compilable state... Apr 19 20:30:10 okay Apr 19 20:30:32 if it's in a package, obviously you would need to inspect that package to find out why it's there. Apr 19 20:30:50 if it isn't in any package, it's probably coming from a malfunctioning postinst script. Apr 19 20:31:22 in that latter case, you'd need to use grep in usr/lib/ipkg/info to identify likely culprits Apr 19 20:33:21 I will try :-) Apr 19 20:34:27 RP: I have a hacky solution now Apr 19 20:36:34 how can i again migrate the db ? Apr 19 20:36:58 rob_w: monotone db? Apr 19 20:37:08 yes Apr 19 20:37:15 rob_w: monotone --db OE.db db migrate ? Apr 19 20:37:24 rob_w: or how to rosterify for monotone 0.26? Apr 19 20:38:08 yeah i ve a new host and had missong monotoe .. so took the binary 0.26 and now it wants that Apr 19 20:38:31 rob_w: are you aware of the implications of monotone 0.26? Apr 19 20:38:48 rob_w: if you use it, you can not connect to ewi etc... Apr 19 20:39:05 cbrake: did you mail me earlier this week? Apr 19 20:39:12 err Apr 19 20:39:24 zecke: yes Apr 19 20:39:38 cbrake: sorry, I still need to respond :} Apr 19 20:39:55 cbrake: it is still in my inbox, and awaiting action Apr 19 20:40:11 zecke: no problem. Apr 19 20:40:50 zecke how about dominion ? Apr 19 20:41:02 rob_w: all, they use monotone 0.25 Apr 19 20:41:13 i see .. Apr 19 20:41:16 rob_w: 0.26 is a more dramatic (to the better) change of monotone Apr 19 20:41:29 rob_w: but we need to have a migration plan for it first Apr 19 20:41:30 zecke: Hacky solution to which problem? Apr 19 20:41:35 RP: cow Apr 19 20:41:43 so why does it then implicate that horrible outcome Apr 19 20:41:53 zecke: yay :) Apr 19 20:42:02 rob_w: new core, new way of storing the data ;) Apr 19 20:42:03 zecke: It gives the 1 minute speed improvement? Apr 19 20:42:09 rob_w: IANA assigned network port? Apr 19 20:42:13 i imagine then i cannot build if i miss those servers , as i also saw that anon access to vanille is off Apr 19 20:42:15 RP: dunno Apr 19 20:42:26 rob_w: it is off? Apr 19 20:42:34 rob_w: monotone 0.26 connects at another port ;) Apr 19 20:42:46 rob_w: simples solution, get monotone 0.25 Apr 19 20:42:47 zecke, err not vanille but some 3rd one Apr 19 20:43:04 rob_w: and wait for monotone 0.26+ Apr 19 20:43:08 alrighto .. will it merge back or do i need a new oe.db ? Apr 19 20:43:22 rob_w: dunno, I think a new one Apr 19 20:43:59 RP: I hate when I forget 'time' Apr 19 20:45:14 zecke ,, hehe i wonder when the day comes when i get the gettingstarted stuff by head .. not today Apr 19 20:46:00 rob_w: the day we improve our docu? Apr 19 20:47:07 mreimer: while you're working on h2200 stuff... could you add an fstab for h2200? Apr 19 20:47:21 reenoo: sure. where? Apr 19 20:47:31 i.e. what package? Apr 19 20:47:33 reenoo: thanks for removing the last lines from detect stylus :) Apr 19 20:47:42 mreimer: base-files Apr 19 20:47:57 reenoo: what is lacking? Apr 19 20:48:35 Any suggestions on what to do while doing update?... Read a book? Apr 19 20:49:02 mreimer: check the ones of the other ipaqs. a tmpfs entry for /var would be most important I think Apr 19 20:49:09 reenoo: ok Apr 19 20:49:33 zecke, probably Apr 19 20:49:45 mreimer: thanks Apr 19 20:50:26 RP: could you parse on a slower machine for me? Apr 19 20:50:33 zecke: heh. I just patched it up a bit. pb_ is going to split the code into two separate apps properly. Apr 19 20:51:02 error , my mozilla breaks on the link to the db snapshots .. wtf ?!? Apr 19 20:51:16 rob_w: I updated the getttingStarted two days ago, because I'm trying to dive into OE. I'm completly new to OE and don't have enough experience to make docu more clear and more complete, sorry :-/ Apr 19 20:51:23 reenoo: I weren't able to argh.. 'ueberzeugen'... I hope you get what I meant... Apr 19 20:52:32 cyrilRomain, then i could imagine other people to worry about this Apr 19 20:52:33 http://ewi546.ewi.utwente.nl/OE/data_cow.patch Apr 19 20:53:05 zecke, is that for me ? Apr 19 20:53:12 no Apr 19 20:53:17 for RP to test parsing Apr 19 20:53:23 I see no difference on ewi Apr 19 20:53:26 sorry Apr 19 20:53:26 * reenoo suddenly remembers schedule for tomorrow.. lecture at 8am... Apr 19 20:53:31 but that is a pretty... Apr 19 20:53:40 reenoo: hehe, mine as well.. Apr 19 20:54:00 reenoo|afk: I hope mine is crowded 200+ so this one gets split up Apr 19 21:01:43 nite Apr 19 21:02:33 zecke: night night Apr 19 21:04:40 hmm. That's what I get for not paying attention... Apr 19 21:20:50 now gst-plugins fails...Package gstreamer-0.8 was not found.....looks like it compiled gstreamer-0.10.4 Apr 19 21:24:34 XorA: well, i got mplayer running, what's a good format to play with ? Apr 19 21:25:13 CSMan: Apr 19 21:25:29 CSMan: http://openzaurus.org/wordpress/howto/encoding-and-playing-video-for-c7x0-devices/ Apr 19 21:25:41 CSMan: that approx what I do Apr 19 21:26:11 now i am AGAIN stuck on boost , argh Apr 19 21:26:23 rob_w: what happened with boost Apr 19 21:27:07 CSMan, ive to rebuild it as iam on a new host .. took over the trees but i need ot reinstall boost and monotone Apr 19 21:27:20 and boost misses now that filesystem stuff Apr 19 21:27:25 rob_w: i remember i had a problem with unit tests, i had to change LANG_ALL to en_US Apr 19 21:27:54 CSMan, where is that ? Apr 19 21:28:27 well, it wasn't boost itself but a dependency : icu Apr 19 21:28:58 icu =? Apr 19 21:29:02 and it was in freebsd Apr 19 21:29:07 oh Apr 19 21:32:44 usr/include/boost/config/compiler/gcc.hpp:92:7: warning: #warning "Unknown compiler version - please run the configure tests and report the results" Apr 19 21:40:00 XorA: appreciated =) Apr 19 21:40:54 ahm , all 291 tests in monotone/testsuite failed ?!? Apr 19 22:02:20 Where should I send new files and patches? Apr 19 22:12:05 Anyone familiar with this error? ERROR: Error Method already seen: get_glibc_fpu_setting in' glibc_2.3.5+cvs20050627.bb' now in 'glibc-package.bbclass' Apr 19 22:17:59 Ken|JLime: You can safely ignore it Apr 19 22:18:16 Ah nice to know Apr 19 22:18:27 Btw, where should I upload new distro file?.. Apr 19 22:20:37 Ken|JLime: Patches should go into the bugzilla Apr 19 22:24:02 Oki, thx Apr 19 22:24:06 shouldn't we do a s,ERROR,WARNING, on that Apr 19 22:24:17 if not supress it Apr 19 22:24:39 TheMasterMind1: Yes and no. It is actually a real bug Apr 19 22:25:03 It just so happens we don't see any error due to it... Apr 19 22:27:37 RP, TheMasterMind1: I think it is worth mention it in the GettingStarted page if you don't remove it Apr 19 22:28:18 cyrilRomain: Its a transient thing - we'll either fix it or remove the warning, once we decide on the correct approach Apr 19 22:30:05 pb_: I think I found the causer of /etc/modules.conf... Just removed modutils-collateral from the h6300.conf Apr 19 22:30:13 tmbinc: ping Apr 19 22:30:18 03lamikr 07org.oe.oz354fam083 * re2d14610... 10/conf/machine/h6300.conf: h6300.conf: remove modutils-collateral to get the update-modules to create /etc/modules correctly during the first boot. Apr 19 22:33:32 tmbinc: http://bugs.openembedded.org/show_bug.cgi?id=880 Apr 19 22:38:57 RP: ok then. Apr 19 22:49:51 lamikr: could you please remove the ENABLE_BINARY_LOCALE_GENERATION/ASSUME_PROVIDED stuff and reinstate the correct include statement? Apr 19 22:50:48 reenoo: have you decided to include udev 0.84, or should I put a PREFERRED_VERSION for it in h2200.conf? Apr 19 22:51:34 mreimer: I'm fine with upgrading udev to 084 Apr 19 22:51:48 ok, I'll leave h2200.conf then Apr 19 22:51:49 mreimer: PREFERRED_VERSION alone wouldn't work btw Apr 19 22:52:11 I've been using PREFERRED_VERSION_udev = "084" in my h2200.conf for a little while now, and it seems to work Apr 19 22:52:23 are you using a different procedure than build gpe-image? Apr 19 22:52:35 yes and no Apr 19 22:52:56 images are built for all target devices Apr 19 22:53:34 if one of those decides to build more recent _arm packages, all devices will get the more recent version Apr 19 22:54:05 oh, because of the way ipkg-index works? Apr 19 22:54:55 reenoo: Hups, sorry I forget those there. I will now send the fix. Apr 19 22:55:22 mreimer: even if ipkg-make-index didn't move stuff to the morgue I'm moderately sure ipkg would pick the more recent version if more than one version is present Apr 19 22:55:44 ah I se Apr 19 22:55:45 e Apr 19 22:56:14 lamikr: thanks. you probably want to move the ENABLE_BINARY_LOCALE_GENERATION/ASSUME_PROVIDED stuff to your local.conf Apr 19 22:58:43 reenoo: Yes, I will do that. I just do not like much from the local.conf as it tends to have uncommitted stuff. But for that it is ok. Apr 19 23:01:03 lamikr: well. you have to distinguish between build specific, site specific, and generally valid data. only the last category may go into the files in conf/, the others must remain in your local configuration files Apr 19 23:02:24 * emte tries to remember ... there was someone else here that uses bitbake in a chroot ... Apr 19 23:02:41 in regards to JustinP 's mailing list message Apr 19 23:03:23 running in a chroot is actually the better approach Apr 19 23:04:08 probably the only way to get fully reproducable builds Apr 19 23:05:23 reenoo: I put by accident the incorrect comment to my oe.db. (same than in earlier h6300.conf commit). I have not pushed it yet. Do you know is there any method to revert the change without pulling the oe.db again from the server? Apr 19 23:07:44 lamikr: you mean you want to change the commit message after the actual commit? no idea, sorry. if the commit just cleans up h6300.conf, it's probably not worth worrying about. Apr 19 23:08:30 Yes, that's what I meant. Well, I am too tired, so I will just push this cleanup. Apr 19 23:09:02 go ahead Apr 19 23:09:16 thanks, done now. Apr 19 23:09:49 reenoo, i agree, i havent tried it since bitbake was renamed. but especially with QEMU you cannot do that Apr 19 23:10:16 03lamikr 07org.oe.oz354fam083 * r9f2e3548... 10/conf/machine/h6300.conf: h6300.conf: remove modutils-collateral to get the update-modules to create /etc/modules correctly during the first boot. Apr 19 23:10:28 i am still getting QEMU accessing root directories and genrerating files outside of the bitbake root Apr 19 23:11:07 good night Apr 19 23:11:08 to qualify that, the only reason i know is that QEMU is not installed on my host and i am finding labled files Apr 19 23:12:04 but maybe that has been fixed now Apr 19 23:12:38 lamikr: 'night Apr 19 23:18:33 /dev/sdb1 112G 6.1G 106G 6% /storage. Yay. Apr 19 23:19:02 woohoo :) Apr 19 23:19:32 ggilbert, you going to setup an access/request tracker? Apr 19 23:19:41 emte: request tracker? Apr 19 23:20:19 file access/request to generate a list of used filed Apr 19 23:20:22 files* Apr 19 23:20:46 then you could use a script to move them to holding for a month and del em after that Apr 19 23:20:52 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:20:53 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:20:53 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:20:55 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:20:59 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:13 What the fuck Apr 19 23:21:14 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:14 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:14 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:16 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:20 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:28 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:28 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:32 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:36 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:40 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:43 Mmm, spam. Apr 19 23:21:44 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:45 emte: ah. I've been meaning to do something like that. Even have a custom bitbake lib app Apr 19 23:21:48 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:48 could someone set the channel +R? Apr 19 23:21:52 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:56 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:21:57 but I doubt it still runs after all this time Apr 19 23:22:00 #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai #fazlamesai Apr 19 23:22:28 ggilbert, you could use a bit fancier tool ... Apr 19 23:22:46 hehe. It was pretty darned fancy :p Apr 19 23:22:48 i am trying to remeber ... there is one taht is actually an apache module ... Apr 19 23:23:22 This one was basically a replacement for using bitbake -cfetch Apr 19 23:23:39 koen|away: we need to fix this gnome-vfs-dbus thing....I keep having to add this to my local.conf: PREFERRED_VERSION_gnome-vfs-dbus = "svn" Apr 19 23:24:29 except it kept track of everything Apr 19 23:25:18 yeah ... Apr 19 23:25:38 i belive i "may" be thinking about webalizer Apr 19 23:25:48 oh, access log stuff Apr 19 23:26:03 I thought you meant for archive maintenance Apr 19 23:26:08 i do Apr 19 23:26:12 based on access Apr 19 23:26:20 which to me makes sense Apr 19 23:26:34 yeah, a bit Apr 19 23:27:23 My primary source of concern is what has packages in the oe repository Apr 19 23:27:43 then after that frequency of access Apr 19 23:27:59 you could setup your scripts taht way ... Apr 19 23:28:19 yeah Apr 19 23:28:30 i would suspect the LICENSE ones would be run all the time Apr 19 23:28:36 access based once a month Apr 19 23:29:24 Well there's really two potential use cases Apr 19 23:29:42 1) some SRC_URI where upstream disappeared Apr 19 23:30:08 yeah, exceptions override list (protected) Apr 19 23:30:42 That stuff really doesnt matter to me much when it comes to pruning Apr 19 23:31:10 2) The floating snapshot oe Apr 19 23:31:25 Those suck up tons of disk space for little benefit Apr 19 23:31:58 you archiving more than two? Apr 19 23:33:11 Right now I've been archiving till full Apr 19 23:33:46 eww Apr 19 23:34:29 Of that 112g drive, about 9g was being eaten by oe temp files. 80 or so by the CVS/Snapshot downloads and 5 by regular source downloads Apr 19 23:34:38 along with the balance being my personal stuff Apr 19 23:35:00 So as it stands, it seems like the key here is just managing the stuff that floats based on the date Apr 19 23:35:45 Because my intent is to keep anything that is hardcoded Apr 19 23:36:38 Maybe keep them for 1 yeard after they leave the repo Apr 19 23:37:20 1yr seems excessive Apr 19 23:40:50 Yeah. Was just picking a safe number Apr 19 23:41:05 the idea is that for stuff like that, I want to leave a bit of wiggle room for people coding off older snapshots Apr 19 23:42:27 the same goes for the CVS_TARBALL_STASH stuff, but it's more expensive to keep and isn't necesarrily useful. So that's where I might use the access logs Apr 19 23:42:57 but I'm not sure how to differentiate something that's a floating cvs build from something that's a hardcoded cvs build Apr 19 23:43:21 all of our opie stuff comes out of CVS ( or did when I last looked) Apr 19 23:43:41 I don't want the source for opie 0.9.3 to disappear just because the script thinks it's 15 or 30 days old Apr 19 23:44:18 CosmicPenguin: btw, I know you guys use the tarball stash, how long of a lag time would AMD find appropriate? Apr 20 00:37:11 what patch command does oe use when applying patches to source? Apr 20 00:38:21 I'm trying to fix up a broken package but the existing patches to it use "patcher". Apr 20 00:38:52 my regular patch program won't apply these patches :( Apr 20 00:58:00 ok Apr 20 00:58:04 so here's a bb question Apr 20 00:58:24 Would it be possible to get bb to tell me where a particular piece of metadata was most recently found? Apr 20 01:13:31 treke|home, that wouldnt be exactly valid Apr 20 01:13:36 i dont think Apr 20 01:13:47 emte: por que? Apr 20 01:13:58 isp restrictions, best path, etc Apr 20 01:14:11 firewalls, routers, etc Apr 20 01:14:18 wnot quire sure what you mean there Apr 20 01:14:45 wow. That's bad spelling. Apr 20 01:15:47 bitbake would only tell you the first valid path for you Apr 20 01:16:01 wich may not be for others Apr 20 01:16:16 Right, but I don't care for anyone else :) Apr 20 01:16:37 Because this is a dirty nasty hack, but it's the best I can think of atm Apr 20 01:22:31 if something is referencing ${CVSDATE}, but ${CVSDATE} was last defined in base.conf, I know I want to treat it differently Apr 20 01:24:11 if I see it was defined in openembedded-3.9.9.conf, then I know I want to treat it is as a static file Apr 20 01:26:12 though maybe I could do if (date == today) {assume_floating) Apr 20 01:30:10 Technically it's wrong, but in most cases it should behave properly. Just fails the case where we set the CVSDATE=today and then the next day change it. That should be fairly rare :) Apr 20 01:34:05 heh Apr 20 01:40:50 treke|home, you could address the repos seperatly Apr 20 01:41:28 since ... theoretically .dev shouldnt have static CVS and the stablizing branches should Apr 20 01:41:47 Has theory become practice? Apr 20 01:42:18 i am not sure it is any better or worse than evaluatin every cvsdate Apr 20 01:42:47 but you could ignore cvsdate for .dev in the second case Apr 20 01:43:16 I'm also not sure I like that approach because it means tracking individual distributions Apr 20 01:43:40 no... there are only 4 branches currently that i know of Apr 20 01:44:17 dev, ozfam, oz, and nlsu Apr 20 01:44:47 hmmm Apr 20 01:45:28 i am not sure if other still exist or not Apr 20 01:45:45 I need to poke around a bit more Apr 20 01:45:49 how can i help w/ regard to the suspend/resume bug and OZ? Apr 20 01:45:58 i've asked this many times, and yes i understand RP is busy Apr 20 01:46:03 I'm thinking I'm only interested in tracking one branch Apr 20 01:46:04 there was talk about more kernel hackers Apr 20 01:46:19 http://oe.handhelds.org/cgi-bin/moin.cgi/DevelopmentBranches Apr 20 01:47:04 treke|home, well the branches you would want to track for CVSDATE would be the stable ones, no? Apr 20 01:47:13 Not necessarilly Apr 20 01:47:26 i cant see why you would cache actuve CVS Apr 20 01:47:31 active* Apr 20 01:47:52 People who can't access cvs Apr 20 01:47:55 (ignoring the fact sf is unreliable) Apr 20 01:48:39 also, CVS is just the most obvious example. There are other things like the qt-embedded-snapshot packages Apr 20 01:48:53 but does that not bring up a timezone skew on the today cvs? Apr 20 01:49:03 emte: not that I know of Apr 20 01:49:16 ie someone 6hours ahead of your server Apr 20 01:49:45 Shouldnt matter, unless things have changed Apr 20 01:50:06 since default was 12:00 am today, in the timezone of the CVS server Apr 20 01:50:13 sf anon cvs is not a priority Apr 20 01:50:22 their normal cvs hasn't gone down in a long time Apr 20 01:50:25 it's very reliable Apr 20 01:50:31 i just mean you wouldnt have a current cvs for someone accross the midnight barrier Apr 20 01:50:37 emte: sure you would Apr 20 01:50:58 your snap on the 19 and thier date the 2oth? Apr 20 01:51:03 20th* Apr 20 01:51:37 jnc, sf was down yesterday Apr 20 01:51:47 or overloaded Apr 20 01:51:47 Does bitbake not handle things in GMT? Apr 20 01:52:00 i do not belive so Apr 20 01:52:14 since requesting a CVSDATE in the future doesn't seem wise Apr 20 01:53:55 not wise and i am not sure if cvs will let you Apr 20 01:54:18 since bitbake would potentially be doing that :p Apr 20 01:55:10 I'd have to dig into the code to look. I'm inclined to say i don't care about that edge case Apr 20 01:56:18 i was just trying to give you a reason to discard current cvs :) Apr 20 01:56:51 hehe Apr 20 01:57:12 Would make live easier, but it's functionality that I know people were using Apr 20 01:57:31 So I'd like to leave it if I can come up with a reasonable solution Apr 20 03:39:29 Is it possible to cross compile using oe & bitbake from cygwin? Apr 20 03:40:02 Because if it's not, that may be my first problem =) Apr 20 03:46:35 zach: Well, I haven't heard of anyone doing it... may as well give it a go and see if it does work :) Apr 20 03:47:07 I keep getting an error when trying to build a package... I'm wondering if that's the cause Apr 20 03:47:33 zach, i highly doubt it will work Apr 20 03:47:59 for one you would have to trust that python will adapt paths properly Apr 20 03:48:43 I installed the cygwin version of python Apr 20 03:48:45 and that all the calls to filesystem functions would be the same in both *nix and windows Apr 20 03:48:53 and made sure it was being called instead of the win32 version Apr 20 03:50:33 be aware that bitbake only marginally works in OSX as far as i know Apr 20 03:50:51 i am not even fully sure it works in BSD Apr 20 03:51:11 (without compatability enabled) Apr 20 03:52:01 it will be interesting to know what will happen when gygwin compiles gcc and then tries to use that gcc to compile the target gcc Apr 20 03:52:06 cygwin* Apr 20 03:55:19 ERROR: Exception:exceptions.OSError Message:[Errno 2] No such file or directory: '/home/zmcgrew/stuff/build/tmp/work/gnu-config-native-0.1cvs20050701-r4/.' Apr 20 03:55:28 Hooray! I'll take that as a no. Apr 20 03:56:09 /home/zmcgrew/stuff/build/tmp/work/gnu-config-native-0.1cvs20050701-r4/ exists, but it doesn't like the '.' Apr 20 03:57:32 i think it appended the . Apr 20 03:57:56 zach, dont you notice an issue with the path? Apr 20 03:58:13 like the fact the slashes are all the wrong direction? Apr 20 03:58:25 cygwin interprets them correctly Apr 20 03:58:32 And windows accepts either way Apr 20 03:58:33 and its not referencing fron c: Apr 20 03:58:47 from* Apr 20 03:58:58 Hmm Apr 20 03:59:25 maybe add C:/home... to your path? Apr 20 04:00:54 That's what I'm going to do Apr 20 04:06:05 I'm so confused... Apr 20 04:06:19 OSError: [Errno 2] No such file or directory: '/home/zmcgrew/stuff/build/tmp/work/gnu-config-native-0.1cvs20050701-r4/.' Apr 20 04:06:52 But... os.path.exists('/home/zmcgrew/stuff/build/tmp/work/gnu-config-native-0.1cvs20050701-r4/.') returns true Apr 20 04:07:55 what is returning the error? Apr 20 04:08:13 i suspect its not bitbake but a toll that it handed off to Apr 20 04:08:16 tool* Apr 20 04:08:50 Seems to be bitbake Apr 20 04:09:02 it looks suspectly like your unpacking failed or it couldnt create the path when it unpacked Apr 20 04:09:23 or it put it somewhere else Apr 20 04:09:41 its hard to tell without more error information Apr 20 04:09:54 I can paste the whole thing, gimme a sec Apr 20 04:11:39 http://rafb.net/paste/results/oW6SDy78.html Apr 20 04:13:09 The thing at the bottom was just me messing around with the cygwin python testing it out Apr 20 04:13:26 ERROR: Error in executing: /home/zmcgrew/stuff/org.openembedded.oz354fam083/pack Apr 20 04:13:26 ages/gnu-config/gnu-config-native_20050701.bb Apr 20 04:13:29 thats your error Apr 20 04:13:57 it cant read the bbfile Apr 20 04:14:18 which is interesting ... i wonder how it parsed it Apr 20 04:14:52 did it actually unpack the file? Apr 20 04:14:55 The file exists Apr 20 04:15:03 and seemsto be ok Apr 20 04:15:30 did it actually unpack the tgz? Apr 20 04:16:22 i am wondering if the windows tar uses different flags Apr 20 04:16:43 tar/gzip/b2zip Apr 20 04:17:34 Seems to have extracted fine Apr 20 04:17:52 But I'm not 100% sure what I'm looking fore Apr 20 04:17:54 for* Apr 20 04:18:03 But I can tar -xzf the file just fine Apr 20 04:18:10 if it unpacked the file Apr 20 04:18:37 look in the /home/zmcgrew/stuff/build/tmp/work/gnu-config-native-0.1cvs20050701-r4/temp/ for a log file Apr 20 04:18:51 or a run file Apr 20 04:19:05 it should contain more details on failure if it could write to it Apr 20 04:19:20 there's no /temp in that folder Apr 20 04:19:25 on a "config" Apr 20 04:20:36 what is all in that folder? Apr 20 04:20:47 CVS ChangeLog Makefile config.guess config.sub testsuite uname Apr 20 04:21:00 ... Apr 20 04:21:06 that doesnt lkook right Apr 20 04:22:12 sec i'll duplicate part of yours Apr 20 04:30:05 ls tmp/work/gnu-config-native-0.1cvs20050701-r4 Apr 20 04:30:13 config/ gnu-configize.in image/ temp/ Apr 20 04:30:34 is how a correctly setup directory looks Apr 20 04:30:44 Hmm Apr 20 04:31:05 so maybe the way tar is extracting a path is getting cut off? Apr 20 04:31:10 $ ls tmp/work/gnu-config-native-0.1cvs20050701-r4/ Apr 20 04:31:10 config Apr 20 04:31:58 Those other files aren't even in the .tar.gz Apr 20 04:32:01 yeah config should contain the src Apr 20 04:32:07 no they arnt Apr 20 04:32:14 bitbake and tools generate them Apr 20 04:32:55 do you have debug uncommented in your local.conf? Apr 20 04:34:57 BBDEBUG = "yes" ? Apr 20 04:38:32 yup Apr 20 04:38:40 just turned it on Apr 20 04:38:44 Rebuilding now Apr 20 04:42:01 http://rafb.net/paste/results/5wkYm264.html Apr 20 04:42:04 Seems to be the same Apr 20 04:43:43 look in your dir Apr 20 04:43:56 hopefully more files were output Apr 20 04:45:35 same files Apr 20 04:49:11 :( Apr 20 04:50:17 i wonder Apr 20 04:50:25 ? Apr 20 04:52:38 what happens if you bitbake gnu-config ? Apr 20 04:53:13 i am wondering which bb it doesnt like as -native inhertis that file Apr 20 04:54:23 Interesting... Apr 20 04:54:24 getopt: command not found Apr 20 04:54:40 What package installs getopt? Apr 20 04:55:16 getopt ... Apr 20 04:55:20 good question Apr 20 04:55:40 lol Apr 20 04:55:46 i am not sure you want to know Apr 20 04:55:51 glibc Apr 20 04:58:24 I don't see anything similar to that listed Apr 20 04:59:41 http://www.gnu.org/software/libc/manual/html_node/Getopt.html Apr 20 05:36:47 * zach shakes fist cygwin Apr 20 05:36:56 The renamed linux-utils to utils-linux Apr 20 05:37:01 And hid getopt in there Apr 20 05:43:31 I got getopt, but I still get the same error Apr 20 05:43:39 Even when I build gnu-config Apr 20 05:43:58 Which used to give me the getopt error, now it just gives the same error as the last one Apr 20 05:47:01 sounds like your progressing Apr 20 05:47:11 probably failing on another related tool Apr 20 05:48:04 finally ... memory amps take way too long to decode from schematics Apr 20 06:01:18 wtf are memory amps? Apr 20 06:03:34 I get 404's when I try and build... Apr 20 06:03:43 That may be part of my problem... Apr 20 06:05:48 Instead of downloading a .tar.gz, it seems to pull the source from cvs Apr 20 06:07:38 Yeah, it does that... checks oesources for a mirror, then CVSes it Apr 20 06:13:52 if e.errno != 17 Apr 20 06:13:57 What's 17? Apr 20 06:14:10 I'm getting errno 2 Apr 20 06:14:14 And it's nailing the build Apr 20 06:22:05 checking build system type... Invalid configuration `i686-cygwin_nt-5.1': system `5.1' not recognized Apr 20 06:26:41 koen|away: e-image built. I had to disable gpe-nmf because gst-plugins won't compile.... Apr 20 06:26:51 03justinp 07org.oe.dev * r6a34921c... 10/packages/gnome/gnome-vfs-dbus_svn.bb: gnome-vfs-dbus_svn: Add sed hackery to replace missing werror patch Apr 20 06:46:46 Hooray! Yet another error: getopt: unknown option -- n Apr 20 07:12:12 bonjour Apr 20 07:20:53 morning all Apr 20 08:19:12 hey Apr 20 08:19:22 koen|away: I'm going to reply to your Bug-Day Apr 20 08:22:41 morning all Apr 20 08:23:13 hey Apr 20 08:32:33 morning all Apr 20 08:32:58 zecke: http://www.rpsys.net/openzaurus/temp/zecke_cow.txt Apr 20 08:33:57 RP: it saves a minute for you? Apr 20 08:34:11 (data_plain did it work?) Apr 20 08:34:45 what do read pdf on embedded device ? ( on GPE ) Apr 20 08:36:08 hi all Apr 20 08:37:23 hi koen|work Apr 20 08:37:26 Genesis: there are plenty of choiches Apr 20 08:37:32 zecke: It saves about 45 seconds, yes Apr 20 08:37:34 hey RP Apr 20 08:37:39 Genesis: any Gtk+, E17, KDE based viewer will work Apr 20 08:37:50 RP: on ewi it is one second slower Apr 20 08:38:09 zecke: I do find the times can be quite variable... Apr 20 08:38:50 someting integreted with an embedded web browser like gpe-mini-browser ? Apr 20 08:39:00 zecke: I was surprised data_plain made as little difference Apr 20 08:39:11 gpdf for example Apr 20 08:39:22 Genesis: nothing integrated Apr 20 08:39:25 oki Apr 20 08:39:27 Genesis: build KDE for that ;) Apr 20 08:39:49 RP: well, from the last traces Apr 20 08:39:59 RP: most time was spend in the expanding of variables Apr 20 08:40:04 RP: (python_sub) Apr 20 08:40:17 zecke: I guess that's understandable Apr 20 08:40:36 RP: well, it could no, that there are no python vars inside ;) Apr 20 08:40:48 KDE !!!! it's a little device not a xcray :/ Apr 20 08:41:24 zecke: I also tried splitting content from self.dict but it made no real difference Apr 20 08:41:29 Genesis: well, people ran KDE on their iPAQs :} Apr 20 08:41:46 and people run KDE on the Zaurus... Apr 20 08:41:53 RP: I will commit this patch then Apr 20 08:42:31 hum Apr 20 08:42:39 if only i could try Apr 20 08:43:12 Genesis: easier to download the pdf and open it with a viewer... Apr 20 08:43:21 it's not the pb Apr 20 08:43:29 i don't success to flash my card Apr 20 08:45:32 do you think kde / konqueror with kpdf will run on arm 926 ? Apr 20 08:45:51 not knowing arm926, I can't tell Apr 20 08:46:06 Genesis: i think it is unlikely ;) Apr 20 08:46:38 or it'll probably run but as to know if it'll be somewhere near usable... Apr 20 08:54:47 morning Apr 20 08:55:13 hrw|work: hey Apr 20 08:55:18 wb hrw|work! Apr 20 08:55:33 oops :) Apr 20 08:55:40 wb hrw|work ! Apr 20 08:56:17 need to start new builddir for 3.5.4.x work ;( fscked last one Apr 20 08:58:56 ~lart my connection to freenode Apr 20 08:58:56 * ibot squishes my connection to freenode like a bug Apr 20 09:00:07 RP: cow patch: The direct access to .__dict__ is... welll Apr 20 09:01:04 zecke: I didn't look too much at what it was doing ;-) Apr 20 09:02:18 morning Apr 20 09:04:17 zecke: the 770 is arm926 Apr 20 09:04:19 hey hrw|work Apr 20 09:04:31 hey XorA Apr 20 09:08:46 hey koen|work Apr 20 09:13:04 mickey|tv: Help, I need some body, help Apr 20 09:16:44 * koen|work hands zecke some rolling stones recoreds Apr 20 09:17:13 koen|work: no thanks, I need to sticky fingers... Apr 20 09:18:26 * koen|work wonders why mkfs.jffs2 suddenly refuses to assemble images Apr 20 09:19:04 *bored* Apr 20 09:19:58 * koen|work starts reading the flamewar on maemo-dev Apr 20 09:20:51 koen|work: anything interesting in the flames? Apr 20 09:21:12 koen|work: how is it named? Apr 20 09:21:21 XorA: trolltech and thekompany peopling are doing their best to say "qtopia is da bestest" Apr 20 09:21:24 something interesting happened last days? Apr 20 09:21:37 zecke: "maemo 2.0 api" and "too busy..." Apr 20 09:21:38 other then bitbake 1.4.0? Apr 20 09:22:39 koen|work: how many people work at dakompany? Apr 20 09:22:56 zecke: no idea Apr 20 09:23:00 hrw|work: no, koen and me are flithy rich and bitbake was released Apr 20 09:23:19 zecke: we are? Apr 20 09:23:29 * koen|work quits job and goes on holiday Apr 20 09:23:29 zecke: and I got mail from you-know-which company Apr 20 09:25:48 hrw|work: *finally* Apr 20 09:25:53 hrw|work: at least positvely? Apr 20 09:26:25 zecke: yes Apr 20 09:27:05 heh Apr 20 09:27:15 my 'boss'/research master is in munich today, so I might know more tomorrow Apr 20 09:27:23 the topic in #angstrom breaks my non utf8 irssi Apr 20 09:28:49 * ade|desk hugs gaim Apr 20 09:29:26 koen|work: http://maemo.org/pipermail/maemo-developers/2006-April/003709.html <- is actually quite interesting Apr 20 09:32:36 zecke: hrm, were you asking about metaclasses? :-) Apr 20 09:33:28 koen|work: see what you mean Apr 20 09:34:16 njs: yes Apr 20 09:34:34 njs: http://ewi546.ewi.utwente.nl/OE/data_cow.patch Apr 20 09:34:37 www.naughty-xxx-porn-girls.com Apr 20 09:34:41 njs: basicly MetaCow and createMetaCow Apr 20 09:34:56 njs: should I really use new.classobj to create a copy? Apr 20 09:35:17 njs: (and besdies having used class.__dict__ ) I will use getattr,setattr Apr 20 09:35:35 njs: from a quick look, do you see some possible speed ups? Apr 20 09:36:09 zecke: the maemo mail confirms to me that dealing with .jp companies always fails for the same reasons Apr 20 09:36:42 XorA: hehe Apr 20 09:37:03 XorA: I think the janitor in the office in japan is higer ranked Apr 20 09:37:07 zecke: similar situation is bankrupting this company the the .jp's have a paranoia about other countries Apr 20 09:37:11 XorA: than any foreigner in a foreign office ;) Apr 20 09:37:45 zecke: the annoying thing is I flew to .jp to meet the engineers and they are great guys, just people in middle management block any progress Apr 20 09:38:15 zecke: you may find it nicer to use hasattr/getattr/setattr than mucking with __dict__ directly? Apr 20 09:38:24 oh, and delattr Apr 20 09:38:33 njs: yeah, I found out today ;) Apr 20 09:39:10 njs: as you look to have some python knowledge: Where can I find documentation on how to implement the iter protocol? Apr 20 09:39:37 zecke: umm. and wouldn't [True, False][x in ...] be a _teensy_ bit clearer if you spelled it "x not in ..."? Apr 20 09:40:07 njs: people should be able to do this easy reduction in head ;) Apr 20 09:40:12 * zecke looks at the patch Apr 20 09:40:47 njs: lol, you are right ;) Apr 20 09:41:05 njs: damn, I was too keen on using something like the 'cond' operator Apr 20 09:41:13 :-) Apr 20 09:41:51 oh, man, I see what you're doing Apr 20 09:42:00 you're not using metaclasses at all! :-) Apr 20 09:42:06 damn :( Apr 20 09:42:14 I feared that Apr 20 09:42:16 this approach is kind of cute :-) Apr 20 09:42:25 I don't know that you actually need to, the basic approach is the same Apr 20 09:42:45 njs: I need to be able to 'copy' arbitary classes Apr 20 09:42:57 njs: this is why I needed to create new classes :} Apr 20 09:43:23 if the whole point is to do weird things to speed stuff up, then when you benchmark it against the old code, it might make sense to benchmark metaclasses too, though Apr 20 09:43:27 njs: I was not able to find a true documentation on '__metaclass__' Apr 20 09:43:35 the problem is that there are both new- and old-style classes Apr 20 09:44:12 http://www.python.org/doc/newstyle/ Apr 20 09:44:37 thanks for the pointers Apr 20 09:44:59 I don't know what's good to read now, I haven't played with them since they first came out, and while there were some docs then, I still had to read the CPython source to figure out how some stuff worked Apr 20 09:45:07 (I was _really_ abusing them, though) Apr 20 09:46:41 hmm, dir() is a somewhat dubious thing to be calling, its interface contract is "get something that would be sorta useful for debugging" Apr 20 09:47:12 but doing proper filtering might kill any speed improvements you were going for, if keys() is being called a lot... :-) Apr 20 09:47:31 njs: at that point of Battle Start Galactica, I did not want to use __dict__directly ;) Apr 20 09:48:36 zecke: createMeta(base = MetaCow) should perhaps be defined as: return type(base)('MetaCowCopy', (base,), {}) Apr 20 09:48:49 I _think_ that works no matter whether base is a new- or old-style class Apr 20 09:49:25 njs: that looks, weird, I will improve it ;) **** ENDING LOGGING AT Thu Apr 20 09:59:56 2006