**** BEGIN LOGGING AT Thu May 09 02:59:58 2013 May 09 07:39:41 good morning May 09 07:48:26 hi mckoan, all May 09 07:49:35 it looks like here in EU we only work in IT and UK today... May 09 07:50:17 * ant_work trying now to reach FR customer May 09 07:54:29 oh, is today VE day or something? May 09 07:56:15 pb its called himmelfahrt or vater/fatherday in germany May 09 07:57:18 ant_work: :-O May 09 07:57:46 yes, frog eaters seems to have holidays May 09 07:57:48 it's Europe Day :D May 09 07:59:36 morning all May 09 07:59:56 gm bluelightning, all May 09 08:00:15 hi mckoan May 09 08:01:26 hi bluelightning May 09 08:01:33 hi woglinde May 09 08:07:58 hi bluelightning May 09 08:08:02 hi pb_ May 09 08:09:01 ant_work: ah yes, so it is. apparently VE day was yesterday for everyone except the Russians and Ukrainians. May 09 08:10:29 I see May 09 08:10:33 I guess we don't have Europe Day as a holiday here because there are already two public holidays in May and that would be too much. :-} May 09 08:10:49 heh May 09 08:10:54 gm bluelightning btw May 09 08:11:06 morning ant_work May 09 08:12:23 see, I may have found a relict yesterday: core-image-minimal-mtdutils May 09 08:13:06 IMAGE_INSTALL += "mtd-utils" won't do the catch anymore after package-split refactoring May 09 08:13:33 dunno if send a fix or a patch to remove that imahe May 09 08:14:55 ant_work: I'd say RFC its removal... I've not heard of anyone using it recently FWIW May 09 08:16:00 ok May 09 08:21:55 good morning May 09 08:22:53 hi apelete May 09 08:24:01 hi bluelightning, mckoan, pb_, ant_work, woglinde, ynezz May 09 08:37:08 hello May 09 08:41:38 hi hrw May 09 08:48:46 and v2 of gdb sent May 09 09:19:14 morning all, apelete, bluelightning, mckoan May 09 09:27:51 hi silvio_, hrw May 09 15:05:59 just pulled master oe-core and bitbake, but get some python errors: File "sanity.bbclass", line 14, in check_gcc_march(sanity_data=) May 09 15:05:59 AttributeError: 'module' object has no attribute 'getstatusoutput May 09 15:11:20 interesting because a from commands import getstatusoutput manually in the python interpreter works.. May 09 15:17:14 aha... oe-core now requires python3, since subprocess lacks getstatusoutput in python 2.x May 09 15:17:27 roric: oh? May 09 15:18:27 erbo, sanity.bbclass uses subprocess.getstatusoutput, seems to be python3, python2 had a getstatusoutput in commands (used by bitbake) May 09 15:18:49 eren, meant to send to you :) May 09 15:19:32 roric: ah, okkie :) May 09 15:19:39 I was happy with python 2.7 though May 09 15:19:45 pb_: your comment about LEAD_SONAME made me smile :) May 09 15:19:47 eren, this commit seem to break it: May 09 15:19:48 e2e1dcd7 meta/classes/sanity.bbclass (Richard Purdie 2013-05-07 13:56:05 +0100 345) import subprocess May 09 15:19:48 sooner or later, we all need to switch to py3k May 09 15:19:55 pb_: We should kill that variable May 09 15:20:09 eren, maybe you have an older checkout than 2 days ago? May 09 15:20:25 roric: pull again as that was just fixed May 09 15:20:44 RP, ah :) great May 09 15:20:49 roric: I haven't touched my OE repositories for 3 days May 09 15:20:58 I'm waiting for uni sysadmins to open a VPN account for me May 09 15:20:59 eren, wow working on old stuff ;-) hehe May 09 15:21:21 they are so slow, and they suck at managing the uni network May 09 15:21:27 anyway, this is off-topic : May 09 15:21:29 :) May 09 15:21:42 RP, have you really pushed it? May 09 15:22:21 roric: er, no, try now, sorry May 09 15:22:36 one of those days :/ May 09 15:23:22 RP, national holiday in your country too? ;) May 09 15:23:48 RP, it works with python2.7 again May 09 15:24:01 roric: Things just not going quite right May 09 15:24:09 roric: great May 09 15:26:28 RP: yeah, I think I added that one in the first place and I have regretted it ever since. May 09 15:27:43 I have a feeling that the reason for adding it was that, at the time, I was under the mistaken impression that we should package glibc the way Debian did (i.e. all in one .deb) and I needed a way to get the shared library autonaming to come out right. May 09 15:28:59 pb_: I think we all have times we've done that... May 09 16:18:27 I pushed hrw/libtirpc branch to oecore-contrib repo. It was rejected but maybe someone find it useful for some reasons. May 09 16:27:43 hrw: thanks May 09 16:28:19 bluelightning: part of cleanups which I do related to my work end May 09 16:28:29 hrw: right, I guessed May 09 16:28:37 bluelightning: is OTC hiring OE devs? May 09 16:28:56 hrw: not atm that I am aware of, but I'll let you know if that changes May 09 16:29:20 bluelightning: thanks, just looking around ;) May 09 16:29:43 hrw@linux.intel.com again would be good May 09 16:31:25 hrw hm May 09 16:55:10 OTC? May 09 16:57:39 ah, intel... May 09 16:58:06 ok, long email sent to oecore and linaro-dev with info about Linaro layers May 09 16:58:16 mr_science: Open Technology Center May 09 16:58:37 mr_science: which is not fully open as they have also 'cursed MS Windows team' May 09 16:58:44 or had 4.5 year ago May 09 17:02:07 * mr_science tries to come up with a metaphor for "windows and open source" May 09 17:02:10 hrw: OTC doesn't include Windows anymore, FWIW May 09 17:02:32 bluelightning: yay! May 09 17:02:32 (assuming it did, I wasn't here when it did so) May 09 17:05:58 hrw: we have a new site at 01.org FYI May 09 17:06:48 bluelightning: I know :) powertop is there and I have few OTC tshirts May 09 17:06:58 very nice cotton ;D May 09 17:07:15 hrw: yeah that t-shirt is one of my favourites, mostly for the material :) May 09 17:08:54 could do with having some OE t-shirts really May 09 17:09:52 bluelightning: there was just one OE tshirt. at 2006 oedem May 09 17:10:15 ah ok... time for a new one maybe May 09 17:25:00 i like tshirts... May 09 17:27:10 I'd suggest setting up a cafepress for OE but the last t-shirt I had from them (a while ago) was quite poor quality May 09 17:29:24 yeah, it would be a neat idea to have some oe t-shirts done. I think last time we tried that, the stumbling block was getting some decent artwork to go on them. May 09 17:29:48 if the current oe logo is available somewhere as a vector file then that should be fine though May 09 17:30:59 how would you distribute it? May 09 17:32:37 what, the artwork? May 09 17:32:44 no, the t-shirt May 09 17:33:22 well, I guess we can cover about 90% of developers by sending a pallet-load to intel. presumably they have an internal mail system of some sort :-} May 09 17:33:47 otherwise, no problem, you can pick them up from my office. May 09 17:34:09 more seriously though, traditionally we distributed those sorts of things at fosdems, linuxtags and the like. May 09 17:34:13 and oedems, of course, if we have them anymore May 09 17:34:15 hehe, thanks but it would mean at least 11-hours flight for me + $1.200 :P May 09 17:34:33 pb_: if only it were that simple :) we had some interesting issues transporting quantities of yocto project shirts to different countries May 09 17:34:41 heh May 09 17:35:11 I will probably be in Germany 1 month later, I'm not sure if there is any linux conferences between June - September May 09 17:36:12 mm, dunno. to be honest, the oe presence in germany has faded a bit of late since Florian and Mickey grew up and had children. May 09 17:37:05 hm May 09 17:37:31 there are a bunch of developers (me included) for whom jumping on a plane and going to a conference on a whim is no longer quite such a trivial undertaking as it used to be. May 09 17:37:53 the OE logo is available here FWIW: http://cgit.openembedded.org/openembedded/tree/contrib/artwork May 09 17:38:00 ah right, very good May 09 17:38:10 reminds me, we really need a new repo for that kind of "collateral" stuff May 09 17:38:25 well, if I have some time and enthusiasm tomorrow, I might see about having a few printed up. send me your size requests if appropriate. May 09 17:40:03 pb_: ah, decline in German presence is sad May 09 17:41:12 probably I'm the only one participating from Turkey. I'm not currently actively engaged due to my finals and other things, I'm thinking of speeding up later May 09 17:41:44 eren: thanks for submitting meta-hamradio :) hopefully beth gets some time to contribute to it at some point May 09 17:42:11 crumbs, is turkey really an 11-hour flight away? May 09 17:42:13 bluelightning: my pleasure :) only aprx is in there right now. I will package svxlink and some logging stuff May 09 17:44:17 pb_: well, accordingly to Turkish Airlines, LTBA (Istanbul) to LAX takes 13:45 May 09 17:44:23 non-stop May 09 17:44:42 ah, well, LAX is about 12 hours from here as well. :-} May 09 17:44:48 how long does LTBA to LHR take? May 09 17:51:03 pb_: 4:10 May 09 17:51:27 pb_: btw, the other name of the airport should be IST May 09 17:51:38 IATA, ICAO difference :) May 09 17:54:16 hm, the tickets are quite expensive. May 09 18:15:10 if someone has time for nonsense stuff (building on BSD) would he be so kind to shed a bit of light on building the cross compiler..? May 09 18:16:34 in the cross compiler work dir are 3 folders now, binutils-cross, gcc-cross-initial and linux libc headers.. May 09 18:17:10 This seems like 3 recipes in one to me.. May 09 18:18:15 the crosscompiler build process consists of a number of recipes, yes May 09 18:18:26 I would have expected to first build gcc-cross-initial, then cross-libc, folllowed by complete gcc-cross and then binutils-cross (the native is already there).. May 09 18:19:11 right, so I should start with gcc-cross-initial I suspose...? May 09 18:19:14 why bintuils-cross at the end? May 09 18:21:30 then there is a full blown compiler, the native owns are already compiled... (but perhaps there is arch specific code in there..?) but elf utils are in general the same I guess... May 09 18:24:47 s/owns/ones May 09 18:28:00 jeroen_: it builds what it has to. and you don't need to 'start' anywhere, the tools build a toolchain if and when it needs one May 09 18:29:14 kergoth: not here ;).. I guess I need to spell it out for him.. May 09 18:29:47 ? May 09 18:29:52 bitbake gcc-cross May 09 18:29:53 the 'spelling out' is already done, in the recipes themselves, as dependencies May 09 18:29:56 no clue what you're talking about there May 09 18:30:08 if you have a problem, then post the errors you're seeing somewhere May 09 18:30:09 and you have the cross compiler May 09 18:30:31 and I bet fbsd breaks on bintuils May 09 18:36:09 woglinde: I has no reason to, bitbake -c devshell pseudo May 09 18:36:30 which automake May 09 18:36:42 /usr/opt/poky/bpp3/tmp-eglibc/sysroots/i386-freebsd/usr/bin/automake May 09 18:37:10 woglinde: unless oe is not as self contained as it claims to be.. May 09 18:38:15 woglinde: ow, binutils, nevermind, I have autotools in my mind as a big problem... May 09 18:39:35 what does automake or pseudo have to do with anything? May 09 18:42:18 kergoth: there is no port yet for pseudo for FreeBSD... May 09 18:42:52 kergoth: automake is ancient on FreeBSD and an pain in the *** with versioning May 09 19:35:04 JaMa, did I send anything adding x86-64 support to numpy for meta-oe? May 09 19:36:41 <_rink> automake on freebsd is ancient? May 09 19:36:57 <_rink> jeroen_, you do know that they like to suffix it with the version? i.e. automake-1.11 etc May 09 19:38:02 jeroen_, doesnt ports have newer vers May 09 19:38:04 and a wrapper even May 09 19:39:36 <_rink> Daemon404, automake isn't in the base system May 09 19:39:47 yeah i know May 09 19:39:49 hence i said ports. May 09 19:40:20 <_rink> pretty sure ports would have newer version; maybe a portsnap fetch update will fix his problems :D May 09 19:42:17 daemon404@bb-nas:~$ pkg_version -v | cut -d" " -f1 | grep auto May 09 19:42:18 autoconf-2.69 May 09 19:42:18 autoconf-wrapper-20101119 May 09 19:42:18 automake-1.12.6 May 09 19:42:18 automake-wrapper-20101119 May 09 19:42:20 ^ current May 09 19:42:28 in ports i mean May 09 19:45:23 Crofton: I don't know, if it isn't in patchwork queue then I don't remember May 09 19:54:24 I do not think I did May 09 19:54:29 I jsut sent it May 09 19:54:40 it was sitting on my disk for a while, not checked in May 09 19:57:04 _dv_: how confident you're about your gstreamer-1.0 recipes? I would like to test newer webkit-efl to see if it builds correctly with gcc-4.8 but revisions newer then last 3 months require gstreamer-1.0 May 09 19:57:27 so I'm wondering if enabling your gst layer will help me to test it quickly or bring more issues :) May 09 20:02:50 gcc upgrades are always fun... May 09 20:59:02 JaMa: is it just -efl that requires gst 1.0? what about other ports? May 09 21:02:56 hollisb: other too May 09 21:03:25 hollisb: support for gst-0.10 was removed completely in 142756 (0.11.5) May 09 21:04:57 hollisb: but you can disable gst support in webkit if you want May 09 21:07:56 JaMa: interesting May 09 21:08:18 JaMa: 0.10 seems to be clinging to life longer than I'd expect :-) May 09 21:09:15 so it's interesting to see it finally disappearing May 09 21:19:01 Daemon404, _rink: until now the automake isn't the biggest problem.. May 09 21:19:20 mostly sed's in the recipes.. May 09 21:20:00 and eyes, ports, has up 2 date version, question is if oe pulls them in... May 09 21:26:12 Daemon404, _rink: more tomorrow, the whole point is oe should be self contained and not be that dependend on os / clang etc **** ENDING LOGGING AT Fri May 10 02:59:58 2013