**** BEGIN LOGGING AT Tue Sep 12 02:59:57 2006 Sep 12 04:24:50 03pfalcon 07org.oe.dev * rda740e6f... 10/ (1 classes/package.bbclass): Sep 12 04:24:50 package.bbclass: Make "strippedness" check overridable. Sep 12 04:24:50 * Factor out "strippedness" substring as FILE_UNSTRIPPED_MATCH. Sep 12 04:24:50 * Allow FILE_UNSTRIPPED_MATCH to be overriden, useful to support Sep 12 04:24:50 other executable formats. Sep 12 04:24:51 * Also, don't complain if .debug directory already exist Sep 12 04:24:53 (can heppen if manually run BB tasks, e.g. for debugging). Sep 12 04:26:55 * v8jlene is back Sep 12 05:18:17 on x86_64, what variable(s) eval to "x86_64-pc-linux-gnu" ? Sep 12 05:32:45 eno: I'd guess that's a string with arch, vendor and os.. so if it's your build system it'd be HOST_SYS (look at the "Architecture-dependent build variables" in bitbake.conf) Sep 12 05:33:17 sorry, BUILD_SYS not HOST_SYS Sep 12 05:33:50 v8jlene: thx, will look at that Sep 12 06:00:59 hi Sep 12 06:04:41 http://www.hrw.one.pl/2006/09/12/haerwu-created/ Sep 12 06:09:48 hi hrw Sep 12 06:29:41 03dirk 07org.oe.oz354x * r01f1f031... 10/ (3 files in 2 dirs): gconf-dbus: add missing xml-backend patches from dev Sep 12 06:48:32 cu Sep 12 07:18:00 koen|away: my macbook is not totally broken Sep 12 07:19:11 hi, folks. Sep 12 07:19:13 if I want build opie, which distro do you preferred ? Sep 12 07:25:51 leoncamel: distro on host computer? Sep 12 07:26:44 den-ros, on pxa27x or pxa25x CPU. Sep 12 07:27:17 den-ros, xscale's series architecture. Sep 12 07:29:02 leoncamel: on host i use SuSE - contain full and complete developmend system. You plan to build opie on PDA with XScale? Sep 12 07:30:03 den-ros, OH. sorry, what I said "distro" is the one of the configure in local.conf. Sep 12 07:30:35 den-ros, you can get all acceptable distros in 'ls org.openembedded.dev/conf/distro' Sep 12 07:30:40 s/in/by Sep 12 07:30:47 ok :) Sep 12 07:31:54 den-ros, and how about yours ? Sep 12 07:32:31 den-ros, which "distro" and "machine" ? Sep 12 07:32:32 leoncamel: sorry - i did not yet start use OE Sep 12 07:32:45 den-ros, OK. Sep 12 07:33:08 den-ros, hmm, how about your SuSE ? which version are you using ? Sep 12 07:33:20 leoncamel: 10.0 Sep 12 07:33:44 den-ros, OH. are you using XGL now ? Sep 12 07:34:02 leoncamel: not yet too, friend Sep 12 07:34:24 den-ros, :). it is very cool with XGL. Sep 12 07:34:44 morning Sep 12 07:35:14 leoncamel: i did not yet need XGL - but i agree it can be make GUI nice look :) Sep 12 07:35:24 hrw|work, here is 15:50 now . :). good afternoon . Sep 12 07:35:30 leoncamel: openzaurus-unstable should work Sep 12 07:35:31 ~ugt Sep 12 07:35:34 ugt is, like, Universial Greeting Time. Created in #mipslinux, it is a rule that states that whenever somebody enters an IRC channel it is always morning, and it is always late when the person leaves. The local time of any other people in the channel, including the greeter, is irrelevant. http://www.total-knowledge.com/~ilya/mips/ugt.html Sep 12 07:36:10 morning hrw Sep 12 07:36:19 morning Sep 12 07:37:19 hrw|work: hey Sep 12 07:37:47 hrw|work, Haha.. very intersting ugt. :) Sep 12 07:38:20 den-ros: anyway xgl on nvidia/ati need non-free drivers ;( Sep 12 07:41:56 http://www.hrw.one.pl/2006/09/12/haerwu-created/ - have to write to ML.. Sep 12 07:42:37 hrw|work: nothing wrong with the nvidia drivers, the ati drivers on the other hand are a whole different ball gamer Sep 12 07:42:59 leoncamel: on my main work notebook Intel video Sep 12 07:42:59 XorA: I use my mx440 with free driver Sep 12 07:43:20 hrw|work: I like my gfx accelleration :-) Sep 12 07:44:02 hrw|work: congratulation, where is the ogg pronouncing HaeRWu? Sep 12 07:44:40 looks like a sneeze to me Sep 12 07:45:05 morning all Sep 12 07:45:27 RP: hi Sep 12 07:45:56 hey RP Sep 12 07:47:51 hi RP Sep 12 07:48:12 zecke: you will hear me during oedem Sep 12 07:49:21 I'm looking forward Sep 12 07:54:33 morning folks Sep 12 07:54:37 hi mickeyl Sep 12 07:54:38 morning mickeyl Sep 12 07:55:00 zecke: hey, how about the occam garni for devdays? Sep 12 07:55:09 hey mickeyl Sep 12 07:55:21 zecke: this is 1.3 km from the venue Sep 12 07:57:59 mickeyl: what is "occam garni"? Sep 12 07:58:03 :} Sep 12 07:58:04 heh Sep 12 07:58:08 zecke: a hotel Sep 12 07:58:19 mickeyl: you are bringing your wife? Sep 12 07:58:26 zecke: to where? Sep 12 07:58:42 zecke: there's a chance Sabine will come to oedem. this is why i had to book the berliner bär Sep 12 07:58:43 morning all Sep 12 07:58:51 mickeyl: ah, now I get it :) Sep 12 07:59:00 be back in a couple of minutes Sep 12 07:59:06 ok, no hurry Sep 12 07:59:32 morning do13 Sep 12 07:59:48 hmm... no milk for coffee Sep 12 07:59:50 darn Sep 12 08:00:11 hey mickeyl Sep 12 08:01:14 mickeyl: real coffee is pure black coffee Sep 12 08:01:18 hi dirk Sep 12 08:01:20 hehe Sep 12 08:01:34 I sometimes to one exception and add sugar Sep 12 08:01:45 hi Marcin Sep 12 08:02:54 i stopped using sugars in _any_ drinks 10 years ago Sep 12 08:04:02 need to call apple now :} Sep 12 08:04:08 zecke: why's that? Sep 12 08:04:20 mickeyl: Random Shutdown Syndrome (RSS) Sep 12 08:04:28 argh :/ i heard about that one Sep 12 08:04:31 mickeyl: if I'm lucky it stay on a minute Sep 12 08:04:38 bummer. Sep 12 08:04:47 oh apropos... mac expo in france today... Sep 12 08:05:10 hehe Sep 12 08:05:19 * mickeyl still waiting for Merom in MBP Sep 12 08:05:52 new alsa 1.0.12 Sep 12 08:08:41 mickeyl: at least they have an auto callback service Sep 12 08:09:17 ya, that's good Sep 12 08:11:26 http://www3.hrs.de/search.do%3Bjsessionid=D9C07818597850CD4C1E96DA64C44F9E?clientId=ZGU-&cid=3-4&randomID=-9223372036834882822&activity=showSearchResult Sep 12 08:11:44 2nd one from the top. only 2 stars, but hey... we just want to sleep there, don't we? Sep 12 08:11:54 or do you need wifi? ;) Sep 12 08:12:43 mickeyl: If I need net I will go to Haidenau Platz and use the blue cable ;) Sep 12 08:13:13 heheh, righto Sep 12 08:13:16 let Sep 12 08:13:21 mickeyl: they present me the query page Sep 12 08:14:46 woah they're starting @ 0800 ! Sep 12 08:14:52 (devday agenda) Sep 12 08:15:08 i wonder whether we should arrive one day earlier Sep 12 08:15:15 what do you think? Sep 12 08:15:51 mickeyl: hi (shugar needed to brain work... :) ) Sep 12 08:17:29 zecke: yeah, i definitly think we need to be there at 10.10. Sep 12 08:17:34 den-ros: right :) Sep 12 08:20:09 mickeyl: (but i use shugar and know programming at low level - and You did not use shugar and know programming wery well - so conclusion - that knowing programming stop brain work :) ) Sep 12 08:20:12 morning all Sep 12 08:21:41 ade|desk: morning Sep 12 08:25:00 mickeyl: yes we need to Sep 12 08:28:06 zecke: ok, let me just reserve this thing. if anything goes wrong, we can cancel it without any fees until 10.10., 12h Sep 12 08:29:40 mickeyl: I have not seen it, is brekfast included? Sep 12 08:29:59 zecke: it is. we probably don't need it because TT has a light breakfast as well Sep 12 08:30:22 twice doesn't hurt anyway ;) Sep 12 08:30:28 mickeyl: I bet they mean: Love is your food so let us hug each other Sep 12 08:30:33 hehehe Sep 12 08:31:57 mickeyl: hmm, ideas how to force support to send you a new macbook? Sep 12 08:32:13 mickeyl: I really don't want to wait 8 days for repair Sep 12 08:32:25 zecke: you say you need it for important work otherwise you lose money and want them to pay Sep 12 08:32:38 * mickeyl books occam garni (strange name, btw) Sep 12 08:32:50 mickeyl: I have not outlined the consequences... hmm Sep 12 08:33:00 mickeyl: to the name, it might just be a japanese sleeping 'cabin' Sep 12 08:33:08 mickeyl: I have seen them in TV Sep 12 08:33:25 mickeyl: I hope that you will get better 'garni' then I had year ago in slovakia Sep 12 08:33:43 hehe Sep 12 08:34:04 hrw|work: how was it? Sep 12 08:34:32 mickeyl: awful, dusty, hard-to-reach power outlets, balls of dust Sep 12 08:34:47 mickeyl: I will bring a vacum cleaner Sep 12 08:34:53 zecke: that will do :D Sep 12 08:36:17 zecke: and power cords? Sep 12 08:36:31 or rather: (ELEKTR) extension lead (BRIT), extension cord (US) Sep 12 08:36:44 hrw|work: I have a macbook, I do not need power Sep 12 08:37:05 brb Sep 12 08:37:12 mickeyl: ah funny, I will leave germany and I would have to leave without notebook Sep 12 08:37:47 oh right. akademy Sep 12 08:38:01 mickeyl: okay, need to try harder Sep 12 08:39:00 mickeyl: do you know python 'stat' well? Sep 12 08:39:14 mickeyl: I would like to have something like the find -perm '-' option Sep 12 08:39:33 zecke: -perm - is 000? Sep 12 08:40:17 mickeyl: no, '-' says to match set bits Sep 12 08:40:25 ah Sep 12 08:40:37 well, os.stat only takes one argument Sep 12 08:40:39 which is the name Sep 12 08:41:02 mickeyl: and ST_MODE gives you the mode Sep 12 08:41:07 and I could use & myself Sep 12 08:41:20 yep Sep 12 08:44:13 later then :) Sep 12 08:44:18 righto Sep 12 08:44:22 * mickeyl needs to buy milk Sep 12 08:45:27 03freyther 07org.oe.dev * r4377a116... 10/ (1 classes/insane.bbclass): classes/insane.bbclass: Run the first test of the insanity.bbclass Sep 12 08:45:34 03freyther 07org.oe.dev * r9fa7d7e5... 10/ (1 classes/icecc.bbclass): Sep 12 08:45:34 classes/icecc.bbclass: Use split so we get the real version instead of a single charachter Sep 12 08:45:34 [2] will get us the 2nd charachter of the first line. This Sep 12 08:45:34 is not what we wanted. We want the version number which is the Sep 12 08:45:35 '2nd' column of the first line. Sep 12 08:45:40 03freyther 07org.oe.dev * rc2682ec1... 10/ (1 classes/insane.bbclass): Sep 12 08:45:40 classes/insanity.bbclass: Check for possible insecure RPATH, make it work with the funy BitBake scope Sep 12 08:45:41 The def methods are in a different scope that the PACKAGEFUNC and the Sep 12 08:45:43 result is only the packagefunc is able to use 'd' so we need to make Sep 12 08:45:45 this a parameter for the other methods. Sep 12 08:45:49 Use the installed scanelf to check for the RPATH of the files Sep 12 08:48:30 whats insanity.bbclass? Sep 12 08:55:07 XorA: it is a bbclass detecting broken packages Sep 12 08:55:28 XorA: there is not much it can do now. It is inspured by ebuild QA checks and will do the same checks Sep 12 08:55:30 _arm package with host binaries etc? Sep 12 08:55:53 XorA: it will check for broken permissions, dangerous RPATH, if PACKAGE_ARCH matches the ELF e_machine Sep 12 08:56:21 XorA: if installed .la and .pc files look correctly, if staged .la and .pc files do not carry 'installed' and do not point to WORKDIR Sep 12 08:57:36 zecke: nice idea - will run Sep 12 08:57:39 XorA: and I only needed about a year and now it can check for dangerous rpath Sep 12 08:58:02 hrw|work: With my current workload, speed, I will need another year to implement the tests :} Sep 12 08:58:56 I'm waiting for day when 768M ram will be enough for my system usage.. Sep 12 09:00:08 time to dig for smallest (ram usage) possible gui (qt/kde/gtk2) player with streaming, playlist... Sep 12 09:00:13 amarok take too much Sep 12 09:10:00 ~lart joomla for not allowing to multiple logins on same userid Sep 12 09:10:01 * ibot farts in joomla's general direction for not allowing to multiple logins on same userid Sep 12 09:16:31 ~lart mtn Sep 12 09:16:31 * ibot whips out a hot clue gun and makes sure that mtn is stuck to the floor Sep 12 09:16:47 invariant ;( Sep 12 09:17:22 no OE building today - need to get newer mtn, migrate db etc.. Sep 12 09:17:31 zecke: cool Sep 12 09:24:50 argh Sep 12 09:24:58 ~lart monotone again Sep 12 09:24:59 * ibot whacks monotone again upside the head Sep 12 09:25:36 mtn: NOTE: this upgrade cleared the roster cache Sep 12 09:25:36 mtn: you should now run 'mtn db regenerate_rosters' Sep 12 09:25:49 mtn: this will take rest of your day Sep 12 09:27:58 good morning Sep 12 09:28:03 eeks Sep 12 09:28:03 hrw|work: sounds painful Sep 12 09:28:19 XorA: you will get same with 0.30 Sep 12 09:28:48 hrw|work: but by then Ill have a 4 Core CPU to run it on :-) Sep 12 09:29:55 or 4 CPU cores to put it right way round Sep 12 09:30:54 XorA: intel has core quad already Sep 12 09:31:03 hrw|work: 2x Dual Core Sep 12 09:31:19 XorA: in one chip Sep 12 09:31:19 hrw|work: is apparently the machine Liam ordered for me Sep 12 09:31:26 XorA: I know that Sep 12 09:31:36 hrw|work: two chips Sep 12 09:31:55 XorA: I know that - Liam told that config on irc Sep 12 09:32:30 hrw|work: then I have misunderstood you :-) Sep 12 09:32:38 np Sep 12 09:32:51 for home Id love an amd64 4x4 :-) Sep 12 09:33:05 XorA: Id any amd64 Sep 12 09:33:33 hrw|work: dont get a shuttle one Sep 12 09:33:56 XorA: if I will buy amd64 then it will be 'SkÅ‚adaQ' (homemade) Sep 12 09:34:53 hrw|work: just make sure plenty of airflow in the case Sep 12 09:34:59 hrw|work: thats where mine fails Sep 12 09:35:14 hrw|work: and Kergoth sees the same problems on his Sep 12 09:37:16 anyway time has come to catch train into office Sep 12 09:37:30 XorA: Scythe Kama Bay on front (12mm in 3x5,25" bays), 80mm on rear, 80mm on top Sep 12 10:13:07 mail to ML sent. Sep 12 10:13:08 http://www.youtube.com/watch?v=Hjhi_FHxY8k Sep 12 10:14:25 hrw account is taken on youtube ;( Sep 12 10:14:55 ~lart arts Sep 12 10:14:55 * ibot wallops arts with a main rotation server that needs rehubbing. It won't take long Sep 12 10:14:59 ~lart flash plugin Sep 12 10:14:59 * ibot stuffs flash plugin into a shiny new tin can and vacuum seals it Sep 12 10:15:02 ~lart adobe Sep 12 10:15:03 * ibot readies the nuke launcher and fires some rounds at adobe Sep 12 10:21:15 morning Sep 12 10:21:27 hi chouimat Sep 12 10:21:52 ~lart mtn for changing db format again Sep 12 10:21:52 * ibot chops mtn in half with a free Solaris 7 CD for changing db format again Sep 12 10:24:31 hrw|work: did you at least view the video? Sep 12 10:25:43 zecke: no - had to close it to release some swap Sep 12 10:26:03 * zecke sends some love to hrw|work Sep 12 10:26:20 768M is too low for today desktops ;( Sep 12 10:26:51 17181 hrw 15 0 198m 93m 21m R 0.7 12.4 11:27.29 firefox-bin 6209 hrw 15 0 102m 69m 26m S 0.0 9.1 2:47.42 konqueror Sep 12 10:26:51 hrw|work: I will talk about this at aKademy but fear to blow up Sep 12 10:27:11 zecke: amarok >100M, konq >100M Sep 12 10:27:24 kopete nearly 100M :( Sep 12 10:28:00 sooner or later people will switch back to mpg321 + elinks Sep 12 10:28:33 18818 hrw 15 0 48452 10m 8832 S 0.0 1.4 0:00.22 kio_imap4 Sep 12 10:28:33 19019 hrw 15 0 47960 10m 8492 S 0.0 1.3 0:00.12 kio_pop3 Sep 12 10:28:57 another mem eaters Sep 12 10:31:49 hrw|work: why do you need amarok for dev work? Sep 12 10:35:06 do13_: music + headphones == nicer work atmosphere Sep 12 10:35:26 especially with nectarine streaming Sep 12 10:36:54 hehe Sep 12 10:38:18 http://www.tomshardware.com/2006/09/10/four_cores_on_the_rampage/ Sep 12 10:38:20 do13_: 5 people in room with different music preferred... from time to time discussions about projects which I do not have to care Sep 12 10:38:28 chouimat: nice read it is Sep 12 10:39:04 hrw|work: I want a motherboard with 8 of those ... should be ok while rebuilding anything ;) Sep 12 10:39:33 chouimat: and flash based harddisks to get fast i/o Sep 12 10:40:17 hrw|work: is openembedded-devel a marketplace for consulting work? ;-) Sep 12 10:40:47 do13_: not quite but its one of places Sep 12 10:41:45 hrw|work: good luck for you business. Do you have first contracts? Sep 12 10:49:45 ~lart kopete Sep 12 10:49:46 * ibot steals kopete's mojo Sep 12 10:49:49 morning Sep 12 10:57:16 do13_: discussed with some Sep 12 10:59:06 hrw|work: I did this step a few weeks ago. My first contract was payed kernel work. Sep 12 11:00:19 ~lart mtn again Sep 12 11:00:19 * ibot resizes mtn again's terminal to 40x24 Sep 12 11:01:15 mtn 0.30 migration on this machine will take 3-4h... Sep 12 11:03:27 hrw|work: they change things again? Sep 12 11:03:48 mtn 0.30 is out? Sep 12 11:04:09 do13_: next week - I use version which koen got after 'invariant' bug Sep 12 11:04:21 thx Sep 12 11:04:48 which version runs on the server? Sep 12 11:05:12 0.29 or 0.28 Sep 12 11:28:36 mtn: 4700/10176 after 2h ;( Sep 12 11:29:11 hrw|work: converting to 0.30? Sep 12 11:30:06 yes Sep 12 11:38:04 http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=6fe9febb8af2f82f9caace1aa1c00cd8de7469ac Sep 12 11:38:17 2G SD/MMC cards pushed into mainline Sep 12 11:38:35 Finally Sep 12 11:45:56 morning all Sep 12 11:47:24 I'm stumped as why when I build oz 3.5.4.2 image why the /lib/modules/2.6.17 directory is empty Sep 12 11:47:42 jerrybme: ? Sep 12 11:48:12 I can build the gpe image in dev fine Sep 12 11:48:39 local config is basically the same Sep 12 11:49:34 I even tried removing the entire build enviroment & recreating, same result Sep 12 11:50:06 jerrybme: tomorrow I will be able to run new build and test Sep 12 11:51:39 are others building oz3.5.4.2rc2 gpe images that work? or is it just me? Sep 12 11:52:54 jerrybme: florian built one yesterday Sep 12 11:53:27 my local conf if anyone is interested: http://ninethsense.com/pastebin/index.php?p=213 Sep 12 11:54:32 should be ok Sep 12 11:54:55 anyone know how to redirect both stderr and stdout to the same file with busybox? Sep 12 11:55:19 Cwiiis: 2>&1 >file Sep 12 11:55:26 Cwiiis: should work Sep 12 11:55:59 hrw|work: It doesn't - it ignores the redirect of stderr to stdout Sep 12 11:56:30 ;( Sep 12 11:56:33 ah, 1>file 2>&1 works though Sep 12 11:56:43 ;) Sep 12 11:56:54 brb Sep 12 11:57:41 hrw|work: I see on the list you're offering your services for oe paid support btw - good luck :) Sep 12 12:00:43 ~lart mtn again Sep 12 12:00:44 * ibot takes a rusty axe and swings it violently, taking mtn again's head off Sep 12 12:07:13 sounds like good SD Card news .. Sep 12 12:09:00 rob_w: yes Sep 12 12:09:04 hi Rene Sep 12 12:09:28 afternoon Sep 12 12:09:31 reenoo: hey Sep 12 12:09:32 hey Marcin Sep 12 12:09:36 rob_w: we have that patch in OZ kernels for long time Sep 12 12:09:42 reenoo: how are you? Sep 12 12:11:44 zecke: hi! I'm fine although still working my way through the ~3000 messages that had accumulated in my inbox while I was on vacation. you? Sep 12 12:18:27 hi Liam Sep 12 12:18:38 hey Marcin, all Sep 12 12:18:47 hi Liam Sep 12 12:20:10 XorA: 2 weeks to go Sep 12 12:20:28 lrg: yes Sep 12 12:20:35 lrg: Ive never been so busy here :-) Sep 12 12:21:09 XorA: Dinner next week, I don't know the day yet. Sep 12 12:21:14 lrg: ok Sep 12 12:21:31 lrg: prefer some notice if its thursday so I can cancel Sep 12 12:21:32 XorA: will find out by Thu/Fri. All free (inc beer) :) Sep 12 12:21:42 lrg: cool, plenty of notice then Sep 12 12:22:12 XorA: I'll ask around today, see if anyone knows... Sep 12 12:32:09 XorA: next Tues/Wed Sep 12 12:32:24 lrg: sorted Sep 12 13:26:57 * mrz80 is away: Metaphysically I'm still here; physically, I'm elsewhere Sep 12 13:31:52 reenoo: I'm fine as well. Sep 12 14:15:36 Did an apples to apples test of Big vs Little Endian performance on the NSLU2. The BE NSLU took 100s to download a 871MB file to /dev/null while the LE one took 140s. Both running 2.6.17 kernels. Sep 12 14:17:38 which version of bb do i want to use for .dev? Sep 12 14:17:57 florian_kc: I'd use 1.6.x Sep 12 14:18:05 joshin: Did you run any other benchmarks? Sep 12 14:18:30 hrm... my 1.6.0 throws error messages Sep 12 14:18:32 RP, not yet. Any suggestions? Sep 12 14:19:21 joshin: Not really. I was thinking of tests for things like graphics, raw processing power and memory access speed Sep 12 14:19:37 RP: graphics? Sep 12 14:19:39 RP, the NSLU has no graphics capabilities. Sep 12 14:19:52 right :) Sep 12 14:20:03 see how fast that LED can flash :-) Sep 12 14:20:05 memory access speed might be a better way to put that :) Sep 12 14:20:08 florian_kc: 1.6.1 works ok Sep 12 14:20:17 florian_kc: What kind of errors? Sep 12 14:20:22 florian_kc: where 1.6.1 = svn bitbake-1.6 branch Sep 12 14:20:34 joshin: At present our test just proves the network interface driver might prefer big endian mode Sep 12 14:21:06 ah well, looks like this is a different issue... Sep 12 14:21:20 i just trued to build openzaurus-unstable for the 770 Sep 12 14:21:27 Yup. Which is why I wanted to see if anyone had ideas for other tests. Sep 12 14:21:42 uf... nearly 5h but finally monotone migrated Sep 12 14:22:16 RP: that migth be related to your changes... "base_contains" is not defined. Sep 12 14:23:09 RP: i guess OZ or the 770 machine config need some updates Sep 12 14:23:25 florian_kc: That does sound like my changes. Was this in .dev? Sep 12 14:23:38 RP: yes Sep 12 14:24:06 florian_kc: base_contains is defined in base.bbclass :-/ Sep 12 14:24:32 florian_kc: Can you pastebin the full error? Sep 12 14:24:36 sure Sep 12 14:26:57 * florian_kc waits for pastebin Sep 12 14:27:21 There are several different ones... Sep 12 14:27:24 ~pastebin Sep 12 14:27:26 somebody said pastebin was a place to paste your stuff without flooding the channel - try http://pastebin.com/ (BROKEN AND SUCKING NUTS), or http://pastebin.ca, or http://channels.debian.net/paste, or for #oe use http://oe.pastebin.com, or http://bzflag.pastebin.ca/, or http://paste.lisp.org/ for the lisp/scheme nuts, or http://paste-it.net Sep 12 14:29:34 It is kinda funny that the ~pastebin result from ibot is long enough to require a pastebin of its own Sep 12 14:32:31 Doing a series of "time cat /dev/mem > /dev/null" tests shows no significant differences between LE & BE both take 0.70-0.71s (32M) Sep 12 14:33:48 joshin: we share ibot :) Sep 12 14:33:55 joshin: That probably means its the network driver thats big endian sensitive :) Sep 12 14:34:10 joshin: That's a relief to those of us using LE ;-) Sep 12 14:34:13 joshin: and we have started adding oe.pastebin.com and then everyone added theirs Sep 12 14:35:35 tim should make it channel dependent :) Sep 12 14:36:45 RP, Yes. Though I have not found a CPU test program yet. I would expect CPU performance not have more than a couple % difference. I'm just finding it interesting since its been something I've wondered about for a while. Sep 12 14:37:08 joshin: Ethernet is BE iirc Sep 12 14:37:58 * lrg also thinks Ethernet is BE Sep 12 14:38:27 Most networking protocols are BE. Sep 12 14:39:28 Most audio is LE Sep 12 14:40:11 Since the NSLU2 doesn't have audio, I can't test that... Sep 12 14:40:25 joshin: but it can have.. Sep 12 14:40:57 I know. It is just not a test that I can do. Sep 12 14:41:06 RP: http://linuxtogo.org/~florian/err.txt Sep 12 14:41:17 And I don't have any idea if a BE openzaurus would run on my 5500. :) Sep 12 14:41:34 joshin: could be interesting ;) Sep 12 14:42:13 "Interesting" usually keeps me up at night. I like boring. :) Sep 12 14:47:51 sigh - sugar artwork is going to be a pita Sep 12 14:49:15 why is gnome so dang cross-compile unfriendly? Sep 12 14:53:02 Hi! Sep 12 14:53:37 zecke: I've found bug in do_package() handling. Any hints on which level this should be fixed? Sep 12 14:53:53 psokolovsky_: which one? Sep 12 14:54:04 anyone in the US here? Sep 12 14:54:22 RP, It does look to be the ethernet driver. I did a "scp foo 127.0.0.1:/dev/null" where foo is a 7M text file in RAM disk and the results were pretty close. 16.8s for LE and 16.20 for BE (5 runs averaged) Sep 12 14:54:27 mithro, sir yes sir Sep 12 14:54:37 zecke: It is error for a package name to appear twice in PACKAGES, leads to overwrite. need to catch, warn, and unique it. Sep 12 14:54:53 psokolovsky_: ah okay Sep 12 14:54:54 * joshin puts on his asbestos underwear Sep 12 14:55:17 psokolovsky_: last time someone told that such PACKAGES are invalid and metadata should be rather fixed Sep 12 14:56:14 i'm looking for somebody to ship high speed ADC's samples to who can them forward them to me Sep 12 14:56:44 Hello Sep 12 14:56:45 hrw|work, I guess, but right now it's silently ignored. I'm ok to error, not warn on it ;-) Sep 12 14:56:46 mithro, as long as there's no $ out of pocket for me, I could do that. Sep 12 14:57:07 zecke, so where that would be handled - in OE classes or in BB? Sep 12 14:58:13 well, there's already check ;-\ Sep 12 14:58:35 psokolovsky_: I think the package bbclass should refuse to work Sep 12 14:59:32 zecke: ok Sep 12 14:59:54 is pulling working for all atm ? Sep 12 15:00:14 ade|desk: see mailing list Sep 12 15:00:17 ade|desk: not for me Sep 12 15:01:42 zecke: rought date/time ? Sep 12 15:01:45 -t Sep 12 15:02:12 florian_kc: An error like that usually means the quoting is wrong somewhere Sep 12 15:02:32 florian_kc: I'd look at task-base.bb and start commenting bits out - see if you can isolate the problem further Sep 12 15:03:36 psokolovsky_: Last time this was mentioned, I said the packages class should error on dulpicates :) Sep 12 15:04:27 RP: well, there's a duplicate check, but somehow it doesn't work. investigating Sep 12 15:04:32 ade|desk: ignore me :) Sep 12 15:04:41 ade|desk: it is not working for me as well Sep 12 15:04:48 How do I create a bb file for something that is installed via setup.py? Sep 12 15:04:54 zecke: ok sure np Sep 12 15:04:55 What do I need to inherit? Sep 12 15:05:10 RP, zecke: any hints on deactivating BB cache freshness check? ;-) my first try didn't work Sep 12 15:05:15 RP: ok Sep 12 15:05:31 florian_kc: It also might indicate a variable wasn't set Sep 12 15:05:59 psokolovsky_: lib/bb/cache.py ;-) Sep 12 15:06:02 Laibsch: distutils iirc Sep 12 15:06:10 Laibsch: look at packages/python/*.bb Sep 12 15:06:10 hrw|work: Thanks. Sep 12 15:06:15 OK Sep 12 15:06:35 psokolovsky_: Test with bitbake -b instead though, then it will reparse :) Sep 12 15:06:56 RP: oh, right ;-) Sep 12 15:07:06 hello Sep 12 15:07:40 i'm following the doc : http://www.openembedded.org/user-manual&dpage=ch03s02 in order to get oe ; but when i do : Sep 12 15:08:19 mtn --mtn=OE.mtn co -b org.openembedded.dev .... it tells me : mtn: mauvais usage : unknown option mtn Sep 12 15:08:49 i checked the man, but i'm not sure which option i should put to replace "--mtn" Sep 12 15:09:17 flux: mtn --db Sep 12 15:09:29 ok :) Sep 12 15:09:42 and the server is not responding atm Sep 12 15:09:54 so no pulls Sep 12 15:13:26 ok Sep 12 15:13:51 RP, hrw|work: Well, python can be unsafe. "pkgs += pkg". Just you wonder, what will be in pkgs if pkg is string? ;-) Sep 12 15:14:54 zecke: ^ Sep 12 15:16:32 psokolovsky_: I've seen this kind of issue before :) Sep 12 15:16:40 lol Sep 12 15:17:07 RP, hrw|work, zecke: So, are you ok to make this fatal? So far, it spews error, but continues. Sep 12 15:17:45 psokolovsky_: That's fine with me Sep 12 15:18:17 psokolovsky_: ok with it Sep 12 15:19:42 * mrz80 is back (gone 01:52:46) Sep 12 15:21:16 We should probably have some kind of exception to raise in these cases so bitbake in "no abort" mode doesn't get bothered by them Sep 12 15:23:27 RP, hrw|work: well, yes, on a second thought, that may be just too tough on users who may have no idea how to fix that. Let's start with error for now, fixing metadata is our task, not theirs. Sep 12 15:24:37 * CosmicPenguin sighs Sep 12 15:24:45 something is pulling in diet-x11, and I don't know what! Sep 12 15:25:32 psokolovsky_: Ideally, we error out and let bitbake decide what to do. From that PoV, OE should never call fatal. Perhaps we should change the behaviour of fatal for outside callers?... Sep 12 15:25:54 CosmicPenguin: python? Sep 12 15:26:21 CosmicPenguin: Suggestions on how to make bitbakes depdenceny tree readable welcome :) Sep 12 15:26:28 hrw|work: no, its something in the chain of libmatchbox Sep 12 15:26:36 CosmicPenguin: want real X11 or just don't want the X at all Sep 12 15:26:42 RP: yes, I guess. a separate task to consider though Sep 12 15:26:47 CosmicPenguin: rm diet-x11.bb then see what .bb bitbake refuses to build Sep 12 15:26:59 heh Sep 12 15:27:09 CosmicPenguin: that how I solve those problems Sep 12 15:27:17 zecke: How do you think we should handle fatal? Sep 12 15:29:44 XorA: heh - annoyingly, nothing is complaining.. :) Sep 12 15:30:06 CosmicPenguin: It used full x11 instead? Sep 12 15:30:13 looks like it is Sep 12 15:32:02 * CosmicPenguin shakes his head Sep 12 15:32:11 bbl Sep 12 15:45:50 03pfalcon 07org.oe.dev * r35850347... 10/ (1 classes/package.bbclass): (log message trimmed) Sep 12 15:45:50 package.bbclass: Detect duplicates in PACKAGES properly. Sep 12 15:45:50 * Fix bug which caused non-detection. Sep 12 15:45:50 * Make visible error message for such condition. Sep 12 15:45:50 * But still recover and continue for now, while we don't Sep 12 15:45:51 have all occurances fixed in metadata. Note that while Sep 12 15:45:53 content of the package will be ok, metadata can be wrong. Sep 12 15:49:44 03pfalcon 07org.oe.dev * r3e2ce2d7... 10/ (1 packages/ntp/ntp_4.2.0.bb): Sep 12 15:49:44 ntp 4.2.0: If we use += for PACKAGES, ntp is already there, don't add it twice. Sep 12 15:49:44 * Fixes #1413. Sep 12 15:49:45 screw this Sep 12 16:03:09 * joshin offers CosmicPenguin a beer and 6 shots of whiskey Sep 12 16:08:28 heh Sep 12 16:08:33 I got toys Sep 12 16:08:37 no time for OE Sep 12 16:12:01 hi CosmicPenguin Sep 12 16:14:51 toys are good Sep 12 16:43:04 yay mtn is back up and running Sep 12 16:43:31 wow off for a weekend and 348 revs to pull Sep 12 16:44:02 oh well time for home , cya all Sep 12 16:48:21 OE is currently building python-native for me. Is that really necessary since I have the ubuntu python packages installed? Can I tell OE not to build certain native packages. Sep 12 16:48:41 add them to ASSUME_PROVIDED and pray Sep 12 16:48:53 the latter is the most important :) Sep 12 16:51:22 I've bought usb nic which was claimed to be supproted by linux, but neither Z, neigher desktop seem to load any modules... What I may be missing? The device is ID 0a46:9601 Davicom Semiconductor, Inc. Sep 12 16:52:02 monotone-viz + OE.mtn = hardcore Sep 12 16:53:59 hrw|work: Thanks. I was thinking that it was ASSUME_PROVIDED. But the description in conf/local.conf left me thinking this was something else. Sep 12 16:57:25 I go home Sep 12 17:46:23 hrw|gone: congrats on the new biz Sep 12 17:56:17 'evening all Sep 12 17:56:22 I wonder how people so high up in the CELF tree can be so clueless Sep 12 17:56:26 hey cyrilRomain Sep 12 17:56:46 clueless? in what way? Sep 12 17:57:06 'python without x11' Sep 12 17:57:09 Laibsch: remember that the target python package will require the same version of python in native, so make sure you have that version on your system Sep 12 17:57:23 (iirc) Sep 12 17:57:27 the python *package* has never depended on x11 Sep 12 17:57:36 heh Sep 12 17:58:01 only the python-gtk2, python-tkinter and friend depend on x11 Sep 12 17:58:14 wonder what he's talking about then Sep 12 17:58:34 kergoth: 'bitbake python' will build x11 for python-tkinter Sep 12 17:58:39 lots off people get lost there Sep 12 17:58:44 hi Sep 12 17:58:50 and we explain that ~weekly on the mailinglist Sep 12 17:58:55 anyone can help with building ld? Sep 12 17:58:59 heh Sep 12 17:59:22 the errori get is: ld: cannot represent machine `arm:mymachine` Sep 12 17:59:35 that's why I said 'clueless', since it was explained ~yesterday :) Sep 12 17:59:49 hehe Sep 12 18:00:11 so I just posted a link to the python package in angstrom showing no depends on x11 Sep 12 18:00:18 * CoreDump|afk tries a native x86 build w/ .oz354x Sep 12 18:00:31 good luck :) Sep 12 18:00:52 but it's good to see the CELF people trying to communicate again Sep 12 18:01:33 we were a bit surprised to have CosmicPenguin tell us that they demoed Cool Stuff(TM) at their conference last year Sep 12 18:02:20 celf demoed cool stuff? Sep 12 18:02:23 as far as I know, Matthew Locke is running OE built images in their test lab Sep 12 18:02:32 * kergoth had no idea Sep 12 18:02:54 He presented a whole lot of interesting stuff - like web triggered builds Sep 12 18:03:16 yeah Sep 12 18:03:18 <[ich]> hey Sep 12 18:03:26 * kergoth is still somewhat surprised at just how far oe has come since starting it Sep 12 18:03:33 unfortunately, his presentation isn't posted on the CELF website Sep 12 18:03:43 I watched the video Sep 12 18:03:53 oh yeah - it had a video Sep 12 18:03:54 thats right Sep 12 18:04:01 my thought were "turn those *(#*%(#(%#%(*%(* cellphones off" Sep 12 18:04:44 * CosmicPenguin does bitbake sugar and ducks Sep 12 18:07:03 CosmicPenguin: vanilla OE has pretty nice support for all that webstuff Sep 12 18:07:25 CosmicPenguin: it's just missing a php page that does system("bitbake foo") Sep 12 18:08:13 but I hope they hire OE people to get a clue :) Sep 12 18:09:24 If I set my distro to angstrom, I can still build minimal images with no display support? Sep 12 18:09:42 sure Sep 12 18:09:58 that's not a distro thing Sep 12 18:09:59 I think this is what I want to do ... Sep 12 18:10:00 yeah Sep 12 18:10:25 I just need to learn how to write the bb file for "bitbake ossie-sdr" Sep 12 18:10:27 'bitbake angstrom-bootstrap-image' is probablly what you want Sep 12 18:10:45 Crofton: take a peak in packages/angstrom/ Sep 12 18:10:49 I need to work on some other stuff for a few weeks Sep 12 18:10:53 that defines some tasks and images Sep 12 18:10:56 ok Sep 12 18:11:17 I think I have been copying stuff from the angstrom distro file anyway ... Sep 12 18:11:18 having some money for a good tinderbox would be nice :} Sep 12 18:11:45 yep Sep 12 18:11:58 I have some suggestions for OEDEM :) Sep 12 18:12:41 zecke: google fed-ex'ed the final cheque, some of that could be spent on tinderboxen Sep 12 18:12:54 keep your money Sep 12 18:13:16 zecke: and if OE becomes a proper SoC org next year.... Sep 12 18:14:24 what is a proper SoC org? Sep 12 18:16:55 Crofton: google summer of code org Sep 12 18:17:22 yeah Sep 12 18:17:24 i keep thinking system-on-chip when i read that Sep 12 18:17:28 it throws me off Sep 12 18:17:30 but what would make oe proper? Sep 12 18:17:41 I am procrstinating ..... Sep 12 18:17:58 Crofton: by registering it, instead of clinging on to hh.org Sep 12 18:18:01 kergoth: your are not that far off. They have hired old Alpha engineers Sep 12 18:18:06 hehe Sep 12 18:22:23 re Sep 12 18:24:48 need to sync my home image Sep 12 18:24:50 later Sep 12 18:30:15 kergoth: Thanks for the hint. I guess I will rather bite the bullet and wait a little longer and spend less time figuring out why things go haywire .-) Sep 12 18:30:41 s/a litte longer/a little longer compiling/ Sep 12 19:41:02 03pH5 07org.oe.dev * re7d24f7f... 10/ (5 files in 4 dirs): udev-097,100: firmware_helper is now firmware.sh - fix udev.rules Sep 12 19:42:18 <[ich]> ljp: hey, Qtopia GPL download is still 2.2 Sep 12 19:42:35 ljp: hey, Qtopia GPL download is still 2.2 Sep 12 19:43:57 zecke: do you found http://www.linuxdevices.com/news/NS5201088922.html ? Sep 12 19:46:59 that looks like a exploit-rich overpriced mvista-ism Sep 12 19:47:48 hehe Sep 12 19:49:11 I wonder how long it takes till OE supports it and some OE dev will port it to 2.6.recent Sep 12 19:49:51 koen depend on how many of "US" get the phone :) I'm currently low on cash :) Sep 12 19:49:54 udev_097-r2_ipaq-pxa270.ipk ?? Sep 12 19:50:11 I wonder where that ipaq-pxa270 comes from. Sep 12 19:50:11 Hmm 695$ Sep 12 19:50:24 exactly :) Sep 12 19:50:26 pH5: ah, right I think SRC_URI containts machine-isms Sep 12 19:50:37 chouimat: get a cheaper Motorola phone Sep 12 19:50:53 chouimat: isn't TT into the community like nokia and giving them away to devs? Sep 12 19:51:19 koen: yes see the community version for 695$ Sep 12 19:51:23 florian: I'm currently have no need for a cell phone ... since I'm always working from home :) tht might change too Sep 12 19:51:29 koen: only for h2200 Sep 12 19:51:42 pH5: I thing that upsets OE :( Sep 12 19:52:00 pH5: and also highlight a huge missing piece in OE/ipkg Sep 12 19:52:40 I was quite annoyed when i found out that Motorola seems to be the only phone manufacturer caring about the GPL. Sep 12 19:53:21 koen: here is the community version: http://www.trolltech.com/products/qtopia/phone_edition/greenphone/greenphone_pricing Sep 12 19:54:49 ljp: where is the download? Sep 12 19:55:26 so I actually have to pay TT $200 for being able to fix known exploits? Sep 12 19:56:37 koen I think it's better we wait a little more :) Sep 12 19:57:22 koen you have the community edition for gpl developer at 695 Sep 12 19:59:12 Greenphone Community SDK Components: -License to develop open source software :) Sep 12 19:59:30 ain't that great? Sep 12 19:59:36 oh dear Sep 12 19:59:44 I guess we should stop developing now Sep 12 19:59:53 It seems we need to buy a license Sep 12 20:00:30 koen then license is free (GPL app only) the phone is 695 Sep 12 20:01:51 * koen gets some food Sep 12 20:04:58 <:-) Sep 12 20:16:09 zecke: using a patched content_checker I get a strange error on the apt-package.inc: http://rafb.net/paste/results/RN95zN58.html any idea ? Sep 12 20:17:32 chouimat: no download Sep 12 20:17:41 chouimat: download refers to Qtopia2.2 Sep 12 20:18:13 cyrilRomain: I'm off to bed now Sep 12 20:18:39 zecke: ok np, I'll investigate a bit latter Sep 12 20:18:42 koen|away: at least one needs to buy a license and then hope we can redistribute the proprietary bits.. Sep 12 20:19:05 cyrilRomain: hi, that's re[gexp] module exception, it didn't match what it expected Sep 12 20:20:30 zecke: which mean I will wait and see :) Sep 12 20:20:53 * chouimat is away: PUB time !!!! Sep 12 20:21:05 later Sep 12 20:21:38 cyrilRomain: it is likely the fetcher and SRC_URI do not like each othe Sep 12 20:21:39 r Sep 12 20:25:28 zecke: ? I'm afraid not to catch your mind Sep 12 20:26:24 how is SRC_URI_OVERRIDES_PACKAGE_ARCH supposed to work? Sep 12 20:26:38 zecke: do you think SRC_URI interpolation fail ? Sep 12 20:26:58 I don't see why udev's PACKAGE_ARCH gets overwritten by MACHINE_ARCH Sep 12 20:27:56 * cyrilRomain continue the test with apt-package in BBMASK Sep 12 20:29:49 pH5: if its SRC_URI is manipulated in a machine specific manner, then, obviously, the package would be bound to the machine rather than the architecture.. if thats what you're asking Sep 12 20:30:15 cyrilRomain: well, a regexp is failing Sep 12 20:30:29 cyrilRomain: so it is likely a) you altered a fetcher b) a SRC_URI is wrong Sep 12 20:30:53 zecke: does bittest handle BBMASK ? Sep 12 20:31:18 maybe :) Sep 12 20:31:29 check lib/bittest/fileparser.py Sep 12 20:31:34 anyway I'm in bed now Sep 12 20:31:40 check if the SRC_URI is wrong Sep 12 20:31:41 later Sep 12 20:31:43 zecke: ok thanks, 'night Sep 12 20:34:09 Just the shear number of packages needed to do anything in GUI land these days is impressive Sep 12 20:36:00 Laibsch: FYI, you can't get away without building python-native. It contains important patches to stock python to make it cross compile python modules. Your native python won't be able to do that. Sep 12 20:36:02 kergoth: it's just that the only SRC_URI modification in udev is for h2200, and even when I remove everything containing _h2200, bitbake still builds an _ipaq-pxa270 package Sep 12 20:39:19 mickeyl: Thanks for the info. Sep 12 20:39:26 np Sep 12 20:39:44 this is actually the case for more than one -native package in OE Sep 12 20:42:13 pH5: ah, thats odd indeed Sep 12 20:42:33 kergoth: oh, now I understand. somebody added an ipaq-pxa270 specific version of mount.blacklist. Sep 12 20:42:50 pH5: time to cut down that h2200-specific. it's a joke to have specificity there Sep 12 20:43:03 aha Sep 12 20:43:45 I don't like those machine specific mount.blacklist files. Sep 12 20:43:45 | bing.c:155:28: error: netinet/ip_var.h: File not found Sep 12 20:43:52 ~curse programs using obscure headers Sep 12 20:43:54 May the fleas of a thousand camels infest your most sensitive regions, programs using obscure headers ! Sep 12 20:46:35 psokolovsky_: agreed. how is this hostap rule h2200 specific anyway? Sep 12 20:48:25 pH5, well, I mean't q mark at the end really, not push for that. But IMHO, such rules are not really machine specific. So, either workarounds like that would be needed for others too, or should be treated like they are - hacks, and probably go into separate package and RDEPENDed by specific machines. Sep 12 20:54:54 * cyrilRomain running bittest content_checker and source_checker in parallel Sep 12 20:56:26 03mickeyl 07org.oe.dev * r83d0e48a... 10/ (1 packages/bing packages/bing/bing_1.1.3.bb): add bing, a console network performance measurement utility Sep 12 21:01:37 hum, another getVar issue: http://rafb.net/paste/results/bgVT9M53.html Maybe in such case we simply can't check the bb file content... Sep 12 21:02:10 I mean without using bitbake and building stuff Sep 12 21:11:32 * v8jlene is back Sep 12 21:13:06 hey v8jlene Sep 12 21:14:19 Morning Sep 12 21:15:14 * v8jlene needs coffee... Sep 12 21:22:21 cyrilRomain: I'll submit your update now "Fix a hang when processing comment block and fix some erroneous for loop usage." - does that sound right? Sep 12 21:23:17 v8jlene: sure :) Sep 12 21:26:21 mickeyl: Do you remember http://bugs.openembedded.org/show_bug.cgi?id=1188 ? Sep 12 21:26:32 mickeyl: The proposed patch fixed the compile for me. Sep 12 21:26:35 03cyril.romain 07org.oe.dev * r22712c86... 10/ (1 contrib/sanitize.py): Sep 12 21:26:35 contrib/sanitize.py: Fix a hang when processing comment block and fix some Sep 12 21:26:35 erroneous for loop usage. Sep 12 21:27:04 thanks v8jlene Sep 12 21:27:07 mickeyl: The trick even works for opie-qcop Sep 12 21:28:35 cyrilRomain: No problem. Good work on it. It makes it a lot easier to get things consistent! And saves my having to think about how to order things ;) Sep 12 21:29:01 If only I didn't have to think about libtool and autoconf ;( Sep 12 21:29:49 ;) Sep 12 21:35:57 hum... I will simply catch exceptions and return "Exception processing the %s file" Sep 12 21:36:48 stefan_schmidt: yes, i remember. i'm still not convinced though why this is necessary Sep 12 21:36:53 it works, of course Sep 12 21:36:59 but i want to know why Sep 12 21:37:09 :) Sep 12 21:38:02 then again, since I really don't care that much about Opie, I might just apply it Sep 12 21:38:11 (nowadays) Sep 12 21:39:52 does anyone want to be part of a test of the 'get-inform-by-mail' feature of bittest content checker ? mickeyl ? Sep 12 21:40:05 s/inform/informed/ Sep 12 21:40:24 cyrilRomain: might as well add a new mailing list for that, don't you think? Sep 12 21:40:48 mickeyl: yes, just want to test it first Sep 12 21:41:02 cyrilRomain: ok, feel free to add mickeyl@linuxtogo.org for a test Sep 12 21:41:28 mickeyl: ok, which package e.g. do you want to check ? Sep 12 21:41:57 qt3, qt4 Sep 12 21:42:15 ok, let's go Sep 12 21:43:50 * cyrilRomain double check his BBFILE variable or this will end up to spamming ;) Sep 12 21:47:42 mickeyl: you should have received three mails Sep 12 21:48:01 cyrilRomain: yep, they're there Sep 12 21:50:09 v8jlene: so are you manually changing the author cert for thos commits? Sep 12 21:53:13 JustinP: Yeah. Using --author to mt commit. Someone pointed this out and asked people to use it when committing someone elses work a while back. Sep 12 21:53:41 mickeyl: At your convenience, can you also review http://bugs.openembedded.org/show_bug.cgi?id=1389 ? Sep 12 21:55:14 psokolovsky_: oh that's a neat idea. Sep 12 21:55:17 i like that patch Sep 12 21:55:25 gets my 'go' Sep 12 21:55:38 mickeyl: Ok, thanks! Sep 12 21:59:10 v8jlene: ah, yep, good idea. I'd noticed strange commits (aith author name capitalized) and was wondering what that was, then saw commits from cyril when he doesn' thave an account. ;-) I looked at the certs and was just curious as to hwo it was done. Sep 12 21:59:25 * JustinP is the typo king Sep 12 21:59:39 ~typo Sep 12 21:59:43 [typo] when someone tries to type really fast without knowing what he is actually doing.... so he makes typing mistakes, or rm -rf / Sep 12 22:00:33 ibot: typoKing is JustinP Sep 12 22:00:35 okay, JustinP Sep 12 22:00:39 ~typoKing Sep 12 22:00:41 somebody said typoking was JustinP Sep 12 22:00:46 &_& Sep 12 22:01:01 lol....interesting typo there Sep 12 22:01:13 * JustinP gets back to work Sep 12 22:03:55 JustinP: Seems like it would be a good idea if mt log also showed who actually commited the work... I guess it must know internally for the signing and stuff but just not show it. Sep 12 22:06:23 JustinP: yes, it appears that OE accounts are email-based (though ones @oe.org are virtual ;-) ), and it's nice idea to use email for --author=, not just first name or something Sep 12 22:16:35 psokolovsky_: well...they're account names, not e-mail. The oe.org ones don't work at all. Would be nice if they would forward (like php.net) Sep 12 22:18:53 JustinP: Yes, at least they look like emails ;-). And for people w/o accounts, using emails is at least look-consistent (plus, they can be reached by it in case of questions). Of course, there's spam problem, but contributors should be ready to leave their coordinates... Anyway, just my 2 cents Sep 12 22:18:57 g'night Sep 12 22:24:04 good night Sep 12 22:31:50 is there a variable I can use to use the host system compiler? Sep 12 22:32:07 (ie, instead of ${CC}, is there a ${HOSTCC}?) Sep 12 22:33:06 03pfalcon 07org.oe.dev * r5db3d555... 10/ (1 classes/base.bbclass): (log message trimmed) Sep 12 22:33:06 base.bbclass(oe_libinstall): Handle shared libs basenames of which were Sep 12 22:33:06 suffixed with version properly. Sep 12 22:33:06 * Sometimes soname of a shlib is overriden to contain extra version. If Sep 12 22:33:06 it's build by libtool, .la still has basename w/o suffix. This fix recognizes Sep 12 22:33:07 such case, and handles shlib installation properly (via libtool). Sep 12 22:33:11 * Example of this is smpeg_0.4.4+0.4.5cvs20040311 which builds libsmpeg-0.4, Sep 12 22:37:51 * CosmicPenguin falls asleep waiting for xulrunner to build Sep 12 22:43:00 03nail 07org.oe.dev * r1ba97399... 10/ (1 packages/ixp4xx/ixp4xx-npe_0.1.bb): ixp4xx-npe: Add initial microcode compiler stuff Sep 12 22:47:41 ~lart busybox Sep 12 22:47:43 * ibot puts busybox through a wood chipper Sep 12 22:48:28 I want to rip mount.blacklist out of udev. It irrirates me no end :-/ Sep 12 22:49:23 'night all Sep 12 22:50:42 NAiL: see the BUILD_ variables Sep 12 22:51:14 inherit native did what I needed ;) Sep 12 23:38:34 http://sourceware.org/ml/systemtap/2006-q3/msg00479.html Sep 12 23:38:55 thats more then slightly interesting to me - if I could compile systemtap modules with OE... that would rule Sep 13 00:21:47 * v8jlene is away (gone cycling) Sep 13 02:17:04 morning all .~! :) **** ENDING LOGGING AT Wed Sep 13 02:59:56 2006