**** BEGIN LOGGING AT Fri Nov 20 03:00:07 2009
Nov 20 03:23:44 <rphillips> Hi all... I'm build a titan (sh4) angstrom-2008.1 helloworld image on ubuntu and getting the following error:
Nov 20 03:23:46 <rphillips> CROSS COMPILE Badness: /usr/include in INCLUDEPATH: /usr/include
Nov 20 03:23:58 <rphillips> on the gcc-cross-initial package
Nov 20 03:24:02 <rphillips> anybody see this before?
Nov 20 03:40:54 <rphillips> full output if it helps http://pastebin.com/m6d9ee203
Nov 20 06:28:31 <CIA-80> 03Klaus Kurzmann <mok@fluxnetz.de> 07org.openembedded.dev * re86672acc5 10openembedded.git/: Merge branch 'org.openembedded.dev' of git+ssh://git.openembedded.net/openembedded into org.openembedded.dev
Nov 20 06:28:31 <CIA-80> 03Klaus Kurzmann <mok@fluxnetz.de> 07org.openembedded.dev * r64d85e2e60 10openembedded.git/conf/distro/include/sane-srcrevs-fso.inc: 
Nov 20 06:28:31 <CIA-80> sane-srcrevs-fso.inc: bump rev for fsousaged
Nov 20 06:28:31 <CIA-80> Signed-off-by: Klaus Kurzmann <mok@fluxnetz.de>
Nov 20 06:59:50 <recalcati> good morning
Nov 20 08:00:34 <CIA-80> 03Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * r4bdcd55396 10openembedded.git/recipes/nmap/nmap_3.81.bb: nmap: disable parallel make
Nov 20 08:00:35 <CIA-80> 03Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * re7e587f6e5 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.org:openembedded into org.openembedded.dev
Nov 20 08:03:14 <eFfeM> morning everyone
Nov 20 08:20:09 <JaMa> morning
Nov 20 08:20:35 <spaetz> morning :)
Nov 20 08:40:05 <hrw> morning
Nov 20 08:45:43 <pb__> hi zecke
Nov 20 08:47:30 <zecke> pb__: morning, I will most likely work on the waterfall on saturday :)
Nov 20 09:01:07 <CIA-80> 03Alex Ferguson <thoughtmonster@gmail.com> 07org.openembedded.dev * r9bbe96095a 10openembedded.git/conf/machine/sgh-i900.conf: Add machine configuration for the Samsung Omnia SGH-i900.
Nov 20 09:01:08 <CIA-80> 03Alex Ferguson <thoughtmonster@gmail.com> 07org.openembedded.dev * rc2e3f4896e 10openembedded.git/ (4 files in 3 dirs): Add kernel recipe for the Samsung Omnia SGH-i900.
Nov 20 09:01:09 <CIA-80> 03Alex Ferguson <thoughtmonster@gmail.com> 07org.openembedded.dev * re25e476a32 10openembedded.git/recipes/ (2 files in 2 dirs): Add support files for the Samsung Omnia SGH-i900 in tslib and pointercal.
Nov 20 09:01:09 <CIA-80> 03Alex Ferguson <thoughtmonster@gmail.com> 07org.openembedded.dev * rf25438c7b9 10openembedded.git/recipes/firmwares/marvell-gspi-fw_9.70.3-p37.bb: Symlink helper_gspi.bin to gspi8686_hlp.bin for the libertas_spi kernel driver in the Samsung Omnia SGH-i900.
Nov 20 09:03:05 <cdbot2> * * OE Bug 5335 has been RESOLVED (FIXED) by stefan(AT)datenfreihafen.org
Nov 20 09:03:07 <cdbot2> * * Initial support for the Samsung Omnia (SGH-i900)
Nov 20 09:03:09 <cdbot2> * * http://bugs.openembedded.net/show_bug.cgi?id=5335
Nov 20 09:13:20 <recalcati> I'm now at qt 4.6 linking. I don't link libglib-2.0.so.0, instead when I compile its headers are seen correctly. How can I find it ? thx
Nov 20 09:15:23 <JaMa> pb__: eglibc/eglibc-package.bbclass:  rm -rf ${D}${datadir}/zoneinfo, eglibc is only libc without zoneinfo files (iso3166.tab, zone.tab aren't installed with tzdata in oe.dev)
Nov 20 09:16:15 <JaMa> pb__: should I check if LIBC == "eglibc", or can be LIBC behavior united somehow?
Nov 20 09:25:43 <CIA-80> 03Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * r603e1cf2f9 10openembedded.git/classes/package.bbclass: package bbclass: give better diagnostic when bb.copyfile fails
Nov 20 09:29:49 <thebohemian> hey OEers
Nov 20 09:30:04 <thebohemian> I've just sent the application for a stand with 2 tables at OEDEM
Nov 20 09:30:07 <hrw> hi Robert
Nov 20 09:30:12 <hrw> cool
Nov 20 09:30:15 <zecke> thebohemian: hey!
Nov 20 09:30:20 <thebohemian> I'll sent an email to the dev list as well, so everybody is in the know
Nov 20 09:31:08 <hrw> ok
Nov 20 09:31:13 <thebohemian> sorry, that I appeared to have been buried. after OEDEM I was on vacation in England (was great btw)
Nov 20 09:31:30 <hrw> thebohemian: how was concert?
Nov 20 09:31:52 <thebohemian> hrw: I was at two gigs. kasabian in newcastle and arctic monkeys in liverpool
Nov 20 09:32:11 <hrw> enjoyed?
Nov 20 09:32:12 <thebohemian> hrw: I liked kasabian more since it was more intense
Nov 20 09:32:32 <thebohemian> hrw: 3rd album of arctic monkeys is more like a walk in the park
Nov 20 09:32:52 <thebohemian> hrw: which is quite the opposite of the 3rd album of kasabian
Nov 20 09:33:06 <thebohemian> hrw: both gigs had the 3rd album of each of the bands as the main topic
Nov 20 09:33:12 <hrw> nice
Nov 20 09:34:12 <florian> good morning
Nov 20 09:34:47 <hrw> hi florian 
Nov 20 09:34:59 <florian> thebohemian: sounds good - many thanks!
Nov 20 09:35:18 * florian feels like vacation too
Nov 20 09:42:58 <thebohemian> hrw: but the english should get rid of that ugly habit of throwing their pee around at gigs ...
Nov 20 09:43:18 <thebohemian> hrw: I am going to write a travel report soon ... maybe the weekend
Nov 20 09:47:13 <zecke> thebohemian: pee?
Nov 20 09:47:28 <thebohemian> zecke: piss
Nov 20 09:48:31 <hrw> urinate
Nov 20 09:48:35 <thebohemian> instead of going to the toilet some just pee into the cup of beer they just had and then throw it around ...
Nov 20 09:48:46 <CIA-80> 03Martin Jansa <Martin.Jansa@gmail.com> 07org.openembedded.dev * rfe7f68d534 10openembedded.git/conf/checksums.ini: checksums: add checksum for tzdata
Nov 20 09:49:30 <thebohemian> this was done so often at the arctic monkeys gig that the hall smelled like a toilet afterwards ... 
Nov 20 09:49:31 <zecke> thebohemian: wow... canibals
Nov 20 09:49:46 <recalcati> I had to add manually -lglib-2.0 to a linking command. Any idea about the correct way to do it with, I remember, pgk-config ?
Nov 20 09:49:54 <thebohemian> zecke: :)
Nov 20 09:50:02 <zecke> recalcati: for Qt?
Nov 20 09:50:11 <zecke> recalcati: use pkg-config...
Nov 20 09:50:27 <zecke> recalcati: or patch the qmake.conf to add "defaults" for glib linking
Nov 20 09:50:43 <zecke> thebohemian: how are you?
Nov 20 09:51:23 <thebohemian> throwing piss seems to be a controversial topic in the UK at least:
Nov 20 09:51:23 <thebohemian> http://live4ever.proboards.com/index.cgi?board=DBTT&action=display&thread=36094&page=1
Nov 20 09:51:55 <recalcati> ok, I'll try. then I can do only -c compile ?
Nov 20 09:51:58 <recalcati> thx
Nov 20 09:53:57 <thebohemian> zecke: fine now. had a good shower after the liverpool gig :)
Nov 20 09:54:38 <zecke> thebohemian: next time in the UK I will just piss everywhere and see what happens
Nov 20 09:54:51 <zecke> I will not even wear any pants *yeah* *freedom*
Nov 20 09:56:54 <XorA> thebohemian: you did go and see two bands supported by the lowest echelons of UK society
Nov 20 09:56:59 <XorA> thebohemian: otherwise known as Chavs
Nov 20 10:00:17 <RP> morning all
Nov 20 10:00:26 <thebohemian> XorA: good to know. I Germany this sort of music is listened to by academics ;)
Nov 20 10:00:35 <thebohemian> s/I/in/
Nov 20 10:00:48 * XorA has never been covered in piss at a concert
Nov 20 10:01:17 <CIA-80> 03Martin Jansa <Martin.Jansa@gmail.com> 07org.openembedded.dev * r117485f14d 10openembedded.git/ (2 files in 2 dirs): 
Nov 20 10:01:17 <CIA-80> linux-openmoko-shr-drm-devel: switch back to drm-tracking branch
Nov 20 10:01:17 <CIA-80> Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Nov 20 10:01:18 <CIA-80> 03Martin Jansa <Martin.Jansa@gmail.com> 07org.openembedded.dev * re8ade1114b 10openembedded.git/recipes/tzdata/ (tzdata.inc tzdata_2009s.bb): 
Nov 20 10:01:20 <florian> hi RP 
Nov 20 10:01:20 <CIA-80> tzdata: add 2009s, fix typo in do_compile, install zone.tab and iso1366.tab for eglibc
Nov 20 10:01:22 <CIA-80> * there was ${WORKDIR}}
Nov 20 10:01:24 <CIA-80> * zone.tab, iso1366.tab is removed in eglibc/eglibc-package.bbclass
Nov 20 10:06:04 <XorA> thebohemian: also send us German bands more, we like them here :-D
Nov 20 10:06:29 <pb_> JaMa: I think probably all the C libraries should behave the same
Nov 20 10:06:43 <thebohemian> XorA: are you sure? to everyone I talked to nobody knew any non-UK indie band (Mando Diao, Shout Out Louds, The Ark etc)
Nov 20 10:07:26 <XorA> thebohemian: Rammstein, Nachtmar, Die Krupts, Front 242 etc etc always sell out here :-)
Nov 20 10:07:37 <thebohemian> :)
Nov 20 10:08:04 <XorA> I thinks we listen to different music though
Nov 20 10:08:21 <zecke> XorA: Tokyo Hotel?
Nov 20 10:08:39 <XorA> zecke: not familiar
Nov 20 10:08:47 <zecke> thebohemian: I'm currently in Iceland.. and they have RTL2 via satellite... I was schocked...
Nov 20 10:08:50 <zecke> XorA: good for you!
Nov 20 10:08:53 <thebohemian> XorA: zecke was just kidding
Nov 20 10:09:10 * XorA goes to google that then
Nov 20 10:09:29 <hrw> zecke: you travel to strange places ;D
Nov 20 10:10:05 <thebohemian> XorA: "bunch of children abused to appeal to the masses (of youth)"
Nov 20 10:10:16 <zecke> thebohemian: I love them...
Nov 20 10:10:23 <thebohemian> zecke: ???
Nov 20 10:11:03 <XorA> ah so S Club 7 types
Nov 20 10:11:31 <zecke> thebohemian: just kidding...
Nov 20 10:12:57 <XorA> thebohemian: in the UK Indie became more mainstream than mainstream, so unless the bands are mega famous "Indie" fans just wont have heard of them
Nov 20 10:13:11 <XorA> thebohemian: one wierd musical paradox :-D
Nov 20 10:17:17 <CIA-80> 03Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * r9fd535b0de 10openembedded.git/classes/package.bbclass: 
Nov 20 10:17:17 <CIA-80> package bbclass: drop ret = 0 check for bb.copyfile since it will return either False or the new mtime
Nov 20 10:17:17 <CIA-80> * Yes, I have files that claim to be from 19700101 :)
Nov 20 10:17:28 <CIA-80> 03Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * r84f302c25b 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.org:openembedded into org.openembedded.dev
Nov 20 10:18:22 <thebohemian> XorA: yeah, in the UK Indie is sort of 'folk' music. everyone listens to it and it can be played at every place. it was quite a different thing than what I am used to in Germany but then this is what a journey is all about :)
Nov 20 10:18:33 <thebohemian> XorA: I *really* enjoyed my trip btw
Nov 20 10:18:47 <XorA> thebohemian: I remeber when indie used to mean white label LPs
Nov 20 10:18:56 <XorA> thebohemian: but then the suits bought it
Nov 20 10:18:59 <zecke> thebohemian: how long did you stay?
Nov 20 10:19:11 <thebohemian> zecke: a week (9th - 15th)
Nov 20 10:19:46 <thebohemian> and I was in London(Camden Town of course), Newcastle, Manchester and Liverpool
Nov 20 10:20:21 <thebohemian> btw. I applied for an OE devroom as well
Nov 20 10:20:34 <XorA> thebohemian: pity you didnt come to nice places :-D
Nov 20 10:20:41 <thebohemian> XorA: ;)
Nov 20 10:21:13 <XorA> thebohemian: how is cambden, last I was there, they were making it all commercial and shiney and spoiling it
Nov 20 10:22:51 <thebohemian> XorA: I don't know how it was before so I cannot compare. It is touristy for sure but I liked to be there and walk around
Nov 20 10:23:22 <thebohemian> XorA: there is a nice towpath between Camden and Islington. 
Nov 20 10:24:21 <XorA> thebohemian: yeah there is
Nov 20 10:24:39 <XorA> thebohemian: It used to be a bigger area, but new offices and the big fire took their toll recently
Nov 20 10:26:24 <thebohemian> I like it when there is a big modern city and you can still find secluded places like the towpath. I know places like that in Berlin, too :)
Nov 20 10:41:25 <Crofton|work> thebohemian, ping
Nov 20 10:42:10 <Crofton|work> thebohemian, devroom or stand?
Nov 20 10:44:44 <Crofton|work> don't worry, I think you answered everything :)
Nov 20 10:47:44 <CIA-80> 03Sebastian Spaeth <Sebastian@SSpaeth.de> 07org.openembedded.dev * r68ff3b4cef 10openembedded.git/recipes/freesmartphone/fso-apm_git.bb: 
Nov 20 10:47:44 <CIA-80> fso-apm: add correct DEPENDS and bump PV to what configure.ac is saying (2.0.0)
Nov 20 10:47:44 <CIA-80> Signed-off-by: Sebastian Spaeth <Sebastian@SSpaeth.de>
Nov 20 10:49:47 <thebohemian> Crofton|work: both
Nov 20 10:50:01 <CIA-80> 03Klaus Kurzmann <mok@fluxnetz.de> 07org.openembedded.dev * r0658a9bfbe 10openembedded.git/conf/distro/include/sane-srcrevs.inc: 
Nov 20 10:50:01 <CIA-80> sane-srcrevs.inc: bump rev for mokonnect
Nov 20 10:50:01 <CIA-80> Signed-off-by: Klaus Kurzmann <mok@fluxnetz.de>
Nov 20 11:20:30 <CIA-80> 03Tom <tom@stosb.com> 07org.openembedded.dev * r8c82fd53f2 10openembedded.git/recipes/tasks/task-x11-illume.bb: ILLUME_CONFIG is now also set by a variable
Nov 20 11:20:40 <CIA-80> 03Tom <tom@stosb.com> 07org.openembedded.dev * r8d577b7454 10openembedded.git/conf/distro/shr.conf: added SHR specific choices for default illume theme and config
Nov 20 11:33:21 <Crofton|work> what is guile?
Nov 20 11:34:03 <Crofton|work> thanks google
Nov 20 11:34:24 <hrw> Crofton|work:  Guile is a Scheme implementation designed for real world programming,
Nov 20 11:39:35 <hrw> florian: xserver-common 1.33?
Nov 20 11:40:46 <Crofton|work> any objections to applying rday's patch to upgrade it?
Nov 20 11:43:24 <CIA-80> 03Martin Jansa <Martin.Jansa@gmail.com> 07org.openembedded.dev * r2eb7235a4b 10openembedded.git/recipes/tasks/task-shr-feed.bb: 
Nov 20 11:43:24 <CIA-80> task-shr-feed: remove all stuff depending on bluez-libs
Nov 20 11:43:24 <CIA-80> Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Nov 20 11:45:18 <CIA-80> 03Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * r2adccab21c 10openembedded.git/recipes/webkit/webkit-efl_git.bb: webkit-efl: run thru oe-stylize.py
Nov 20 11:45:30 <XorA> Crofton|work: go for it, the sooner he stops sending emails the better for my sanity
Nov 20 11:46:09 <Crofton|work> heh
Nov 20 11:46:16 <Crofton|work> his patches are easy to apply :)
Nov 20 11:46:30 <Crofton|work> once I figured out to get the mbox from pathchwork
Nov 20 11:48:07 <hrw> Crofton|work: contrib/patchwork/pw-am.sh NUMBER_FROM_PATCHWORK D:
Nov 20 11:51:23 <Crofton|work> hrw, aweome
Nov 20 11:52:47 <hrw> Crofton|work: remember to mark patch as applied and archive it
Nov 20 11:54:07 <Crofton|work> yep
Nov 20 11:56:33 <florian> hrw: oh... new patches?!
Nov 20 12:05:16 <Crofton> ah, the -s adds signed-off by me?
Nov 20 12:05:30 <woglinde> yes
Nov 20 12:05:31 <woglinde> sir
Nov 20 12:06:09 <Crofton> He didn't get the commit message quite right this time, but close enough
Nov 20 12:06:28 <Crofton> also, are we putting checksum in recipes yet?
Nov 20 12:06:36 <woglinde> hm
Nov 20 12:06:42 <woglinde> I would say no
Nov 20 12:06:55 <hrw> florian: no, xmodmap dir
Nov 20 12:07:01 <pb_> yeah, I have started putting checksums in recipes for things that I add
Nov 20 12:07:05 <Crofton> ok, just checking if we need to start commenting on that in patches
Nov 20 12:07:07 <Crofton> ok
Nov 20 12:07:10 <pb_> you can do it either way, it doesn't really matter
Nov 20 12:07:22 <woglinde> pb ah
Nov 20 12:07:25 <Crofton> I should figure out the syntax and start commenting on patches though
Nov 20 12:08:01 <pb_> curl_7.19.7 has an example, that's one that I added yesterday
Nov 20 12:10:42 <woglinde> jo mickey
Nov 20 12:10:55 <mickey|sofa> morsce
Nov 20 12:11:58 <florian> hi mickey|sofa 
Nov 20 12:12:23 <pb_> hi mickey|sofa 
Nov 20 12:13:28 <mickey|sofa> hey florian, pb_ 
Nov 20 12:14:32 <CIA-80> 03Robert P. J. Day <rpjday@crashcourse.ca> 07org.openembedded.dev * r0efbdb1b33 10openembedded.git/ (3 files in 2 dirs): 
Nov 20 12:14:32 <CIA-80> Upgrade guile from 1.8.6 to 1.8.7.
Nov 20 12:14:32 <CIA-80> Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Nov 20 12:14:32 <CIA-80> Signed-off-by: Philip Balister <philip@balister.org>
Nov 20 12:27:37 <pb_> Crofton: if you are thinking of Wookey, it is his "last" name (i.e. he doesn't have a first name)
Nov 20 12:28:26 <woglinde> didnt know we live in the star wars universe
Nov 20 12:30:05 <hrw> woglinde: never met Wookey?
Nov 20 12:30:29 <woglinde> hrw only read about him in the internet
Nov 20 12:31:11 * XorA could legally change his name to XorA these days
Nov 20 12:31:19 <rwhitby> I met Wookey once for 5 minutes
Nov 20 12:32:12 <broonie> XorA: You always could in Scotland.
Nov 20 12:32:34 <XorA> broonie: only after people have known me as XorA for a couple of years
Nov 20 12:32:42 <woglinde> chrome ose
Nov 20 12:35:12 <woglinde> hm intressting
Nov 20 12:35:33 <woglinde> each chrome os package has a debian/ dir
Nov 20 12:37:37 <broonie> Seems reasonable given that Debian and Ubuntu are basically the only traditional distros for ARM.
Nov 20 12:38:13 <Crofton|work> yes i am thinking of Wookie :)
Nov 20 12:38:21 <Crofton|work> or . Wookie on linkedin 
Nov 20 12:38:55 <pb_> ah, wookie would be someone different.  the wookey I am thinking of spells his name with an "ey"
Nov 20 12:39:19 <broonie> XXXXXXXXXXXXXXX Wookie on his passport :)
Nov 20 12:39:24 <woglinde> *g*
Nov 20 12:39:24 <broonie> ey even
Nov 20 12:39:35 <broonie> Crofton|work just mistyped it above.
Nov 20 12:40:18 * Crofton|work is not doing well this morning
Nov 20 12:40:36 <woglinde> crofton go to sleep again
Nov 20 12:40:46 * Crofton|work wants to go to Austria caving with Wookey and the Cambridge guys one day
Nov 20 12:40:56 <Crofton|work> althoug I suspect I would fester alot
Nov 20 12:41:30 <woglinde> austria or australian?
Nov 20 12:42:32 <Crofton> Austria
Nov 20 12:42:55 <woglinde> vienna?
Nov 20 12:43:10 <Crofton> just east of there
Nov 20 12:44:02 <Crofton> http://www.srcf.ucam.org/caving/wiki/Expo
Nov 20 12:45:07 <CIA-80> 03Martin Jansa <martin.jansa@gmail.com> 07org.openembedded.dev * r72a77d6b4c 10openembedded.git/recipes/jamvm/jamvm-initial_1.5.0.bb: 
Nov 20 12:45:07 <CIA-80> jamvm-initial: add libffi-native dependency
Nov 20 12:45:07 <CIA-80> Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Nov 20 12:47:14 <spaetz> anyone responsible for gnuradio here?
Nov 20 12:47:26 <spaetz> it's got a _git.bb but they only have SVN
Nov 20 12:47:53 <spaetz> no git there at all. And parsing that broken SRC_URI fails with using SRCPV
Nov 20 12:48:07 <Crofton|work> yeah
Nov 20 12:48:29 <Crofton|work> my understanding is SRCPV is not in dev
Nov 20 12:48:30 <spaetz> so I'd either remove the recipe, get rid of SRCPV, or have someone to adapt it to using proper svn SRC_URIs
Nov 20 12:48:35 <Crofton|work> and it is git now
Nov 20 12:48:46 <Crofton|work> ?
Nov 20 12:48:53 <Crofton|work> the recipe works for me
Nov 20 12:49:08 <spaetz> now, the _git.bb points to a git location, but they seem to have there code in svn
Nov 20 12:49:19 <Crofton|work> they are migrating
Nov 20 12:49:47 <Crofton|work> very slowly
Nov 20 12:49:47 <Crofton|work> the website needs updating, 
Nov 20 12:49:52 <spaetz> ahh, ok
Nov 20 12:49:53 <Crofton|work> the SRCPV will get fixed whenever the stuff lands in dev
Nov 20 12:49:58 <spaetz> yep
Nov 20 12:50:01 <spaetz> thanks
Nov 20 12:50:07 <Crofton|work> np
Nov 20 12:50:19 <Crofton|work> the version selection will stay in the bb fuile
Nov 20 12:50:24 <JaMa> spaetz: it works for me too with patched bitbake..
Nov 20 12:50:29 <Crofton|work> too much of a pain to keep it in a seperate file
Nov 20 12:50:41 <spaetz> Crofton|work: no problem with that :)
Nov 20 12:51:12 <JaMa> spaetz: maybe just git clone/git rev-list doesn't work for this repo?
Nov 20 12:51:34 <spaetz> JaMa: I don't have a patched bitbake as vanilla bitbake works fine as long as BB_GIT_CLONE_FOR_SRCREV is commented out
Nov 20 12:52:39 <woglinde> Crofton did you test gnuradio with newer boost yet?
Nov 20 12:53:09 <Crofton> not yet
Nov 20 12:53:21 <Crofton> they broke the usb autofoo AGAIN
Nov 20 12:53:32 <woglinde> ??????
Nov 20 12:53:34 <Crofton> I need to fix that so I can move the version to soemthing more recent
Nov 20 12:53:40 <woglinde> gnuradio
Nov 20 12:53:43 <woglinde> or boost?
Nov 20 12:53:48 <Crofton> it builds on macos, mingw, and unix
Nov 20 12:53:51 <Crofton> gnuradio
Nov 20 12:54:23 <Crofton> I need to fix the gnuradio source
Nov 20 12:54:35 <Crofton> without breaking the other cases
Nov 20 12:54:39 <Crofton> hmm
Nov 20 13:37:38 <CIA-80> 03Thomas Zimmermann <ml@vdm-design.de> 07org.openembedded.dev * r1bfa409a3a 10openembedded.git/ (3 files in 2 dirs): om-gta0{1, 2}.conf task-shr-minimal: remove apm from machine features and add fso-apm to task-shr-minimal-base
Nov 20 14:00:59 <noglitch> Hi all !
Nov 20 14:01:23 <noglitch> Someone experienced a pure an simple hang of bitbake ?
Nov 20 14:01:37 <noglitch> without a line of error...
Nov 20 14:01:39 <thebohemian> you got the hang of bitbake :)
Nov 20 14:02:59 <noglitch> nothing more to tell: it hang and I have to kill -9 it...
Nov 20 14:03:17 <noglitch> python seems to be well installed...
Nov 20 14:03:39 <noglitch> for example bbimage shows:
Nov 20 14:03:40 <noglitch> bbimage -v
Nov 20 14:03:40 <noglitch> BitBake Build Tool Core version 1.8.12
Nov 20 14:03:40 <noglitch> BBImage version 1.1
Nov 20 14:04:52 <recalcati> how can I compile using the same environment of openembedded?    I mean... I fix something.. I go in work dir and I do make,  but I wish to have all variables set .   it is possible?  thz
Nov 20 14:05:00 <recalcati> thx
Nov 20 14:09:20 <CIA-80> 03Sebastian Spaeth <Sebastian@SSpaeth.de> 07org.openembedded.dev * r1f782e83f3 10openembedded.git/recipes/tasks/task-shr-minimal.bb: 
Nov 20 14:09:20 <CIA-80> task-shr-minimal: Remove fso-apm and e-wm-config-illume-shr
Nov 20 14:09:20 <CIA-80> * fso-apm is pulled by task-fso2-compliance
Nov 20 14:09:20 <CIA-80> * e-wm-config-illume-shr is pulled by SHR distro setting ILLUME_THEME in task-x11-illume
Nov 20 14:09:22 <CIA-80> Signed-off-by: Sebastian Spaeth <Sebastian@SSpaeth.de>
Nov 20 14:12:17 <pb_> recalcati: see devshell.bbclass
Nov 20 14:12:35 <recalcati> pb_: thx
Nov 20 14:15:43 <XorA> hrmmmmm for some reason pam under uclibc stages wronf
Nov 20 14:15:45 <XorA> wrong
Nov 20 14:17:11 <recalcati> bitbake devshell; MACHINE=dm365-evm bitbake  -b recipes/qt4/qt4-embedded_4.6.bb -c devshell     .     fantastique . A new window appears and I'm inside the qt-4.6 work dir. 
Nov 20 14:17:25 <cbrake> any ideas what this error might mean: /angstrom-2008.1/tmp/cross/armv5te/arm-angstrom-linux-gnueabi/bin/as: unrecognized option '-Qy'
Nov 20 14:17:35 <cbrake> while building gcc-cross-initial
Nov 20 14:18:00 <zecke> cbrake: wow... no... do you have CC set to something weird?
Nov 20 14:18:33 <CIA-80> 03Thomas Zimmermann <zimmermann@vdm-design.de> 07org.openembedded.dev * r66d1101a0a 10openembedded.git/recipes/python/python-epsilon_svn.bb: 
Nov 20 14:18:33 <CIA-80> python-epsilon: Add new SRC_URI
Nov 20 14:18:33 <CIA-80> Signed-off-by: Sebastian Spaeth <Sebastian@SSpaeth.de>
Nov 20 14:18:45 <cbrake> zecke: Joe is building on a 64-bit vmware Ubuntu 9.10 machine and getting this
Nov 20 14:18:51 <cbrake> zecke: works fine for me
Nov 20 14:18:56 <cbrake> zecke: on my native 64-bit system
Nov 20 14:19:50 <hrw> -Qy... check mailing list archive
Nov 20 14:19:55 <cbrake> zecke: from the run.do_compile: export CC="gcc"
Nov 20 14:19:59 <cbrake> hrw: ok, will do
Nov 20 14:22:32 <cbrake> so it should be calling the host as, not arm as ...
Nov 20 14:31:38 <cbrake> found a potential solution: http://bugs.openembedded.org/show_bug.cgi?id=31022~
Nov 20 14:31:51 <hrw> noglitch: never saw that
Nov 20 14:31:54 <cbrake> http://bugs.openembedded.org/show_bug.cgi?id=3102 (with trailing garbage)
Nov 20 14:33:07 <noglitch> hrw: I must say that I am in a virtual machine (512MB or ram) with filesystem on NFS...
Nov 20 14:33:22 <noglitch> hrw: ... not so usual
Nov 20 14:34:09 <hrw> hmm.. maybe thats why
Nov 20 14:34:30 <noglitch> but CPU is idle... some memory remains. Python does not seem to work at all
Nov 20 14:36:23 <noglitch> python 2.6 / psyco / ubuntu 9.10...
Nov 20 14:39:59 <noglitch> here is the log of verbose python -vvv: http://paste.debian.net/51976/
Nov 20 14:40:25 <noglitch> BTW: I am on stable/2009
Nov 20 14:40:33 <CIA-80> 03Sebastian Spaeth <Sebastian@SSpaeth.de> 07org.openembedded.dev * r849cd3aa31 10openembedded.git/ (3 files in 2 dirs): 
Nov 20 14:40:33 <CIA-80> Revert "om-gta0{1, 2}.conf task-shr-minimal: remove apm from machine
Nov 20 14:40:33 <CIA-80> features and add fso-apm to task-shr-minimal-base"
Nov 20 14:40:33 <CIA-80> * Will do it distro specific....
Nov 20 14:40:33 <CIA-80> * This reverts commit 1bfa409a3a28cee60c4800d3caf33ce3f100d4ca.
Nov 20 14:40:59 <recalcati> I need a 16 core .. at least .. and DDR3 ram.  anybody can point me to very fast compiling servers ?   for oe, but also for other compilations.
Nov 20 14:42:39 <recalcati> because my work team is wasting too much time waiting their PCs
Nov 20 14:44:30 <zecke> recalcati: have you used icecream before?
Nov 20 14:44:49 <recalcati> icecream: mean eat one and wait without hurry
Nov 20 14:45:27 <recalcati> apt-get ... installed 
Nov 20 14:45:56 <recalcati> no, it is about streaming
Nov 20 14:46:55 <zecke> recalcati: icecc and see the opensuse wiki about it
Nov 20 14:53:33 <cbrake>  zecke hrw: appears the "as: unrecognized option '-Qy'" problem was caused by user adding './' to the PATH
Nov 20 14:54:12 <hrw> cbrake: great. can you add it to wiki?
Nov 20 14:54:30 <cbrake> zecke: wonder if this is something we should check for ...
Nov 20 14:55:12 <mckoan> good morning
Nov 20 14:55:15 <cbrake> hrw: yes -- looking for a good place ...
Nov 20 15:03:22 <mickey|sofa> XorA: you seem to have some serious problems with me or why do you take every chance to disagree? i wonder what i did to you.
Nov 20 15:03:28 <mickey|sofa> *shrug*
Nov 20 15:07:25 <spaetz> hrw: thanks for the MACHINE_distro suggestion. I should have thought of that in the first place
Nov 20 15:09:04 <XorA> mickey|sofa: I dont disagree, I just think you exposed a bug in FSO
Nov 20 15:09:25 <XorA> mickey|sofa: maybe I been speaking to koen too much and come over too abrupt sorry
Nov 20 15:09:26 <hrw> hi mckoan mickey|sofa 
Nov 20 15:09:28 <hrw> spaetz: np
Nov 20 15:10:05 <mickey|sofa> XorA: yeah, i seem to imagine a pattern of increased rudeness and agressivity on your side against me and am really wondering 
Nov 20 15:10:26 <mickey|sofa> anyways, wrt. the technical issue see my other mail
Nov 20 15:11:59 <spaetz> Just out of interest, how can I see who else is basing a distro on that machine type?
Nov 20 15:12:16 <spaetz> Not that it's really important to know...
Nov 20 15:12:42 <hrw> spaetz: my out-of-tree umbaumba 31.4 distro uses om-gta01
Nov 20 15:12:59 <spaetz> hrw: ahh ok.
Nov 20 15:13:07 <mickey|sofa> you can't
Nov 20 15:13:11 <Crofton|work> we all have crappy days, I sure was having one yesterday
Nov 20 15:14:02 <florian> hrw: "GPE: fboor * r9972 /trunk/base/xserver-common/ (ChangeLog Makefile): Release 1.33"
Nov 20 15:14:36 <hrw> florian: does it gives same result as 1.30 from OE?
Nov 20 15:15:55 <florian> hrw: I hope its an improvement :-)
Nov 20 15:16:46 <florian> would have to check if its really the same 
Nov 20 15:19:44 <hrw> 11 minutes for building rootfs...
Nov 20 15:19:45 <hrw> suxx
Nov 20 15:23:07 <florian> ask for compiling Qt ;)
Nov 20 15:23:37 <mickey|sofa> hrw: glad to see the return of umbaumba. it's been long :))
Nov 20 15:25:10 <florian> heh
Nov 20 15:25:18 <hrw> florian: I meant 11 minutes in do_rootfs task 
Nov 20 15:25:32 * XorA creates an umbaumba.conf
Nov 20 15:25:34 <hrw> mickey|sofa: 17.4, 18.1, 31.4
Nov 20 15:26:53 <florian> hrw: that's how I read it. I speak about the qt-x11-free do_compile task :)
Nov 20 15:27:58 <hrw> florian: ok - then look at glibc.do_package;D
Nov 20 15:29:00 <florian> heh
Nov 20 15:31:25 <florian> hrw: have you ever noticed a difference when you build setting $MACHINE in the environment in contrast to have it in a config file?
Nov 20 15:32:11 <hrw> no
Nov 20 15:32:28 <hrw> I usually have ?=at91sam9263ek in auto.conf and manually set others
Nov 20 15:33:24 <florian> I need to configure the bot to built latest stuff too. What I currently built is somewhat outdated from the OE perspective.
Nov 20 15:33:32 * DJWillis grumbles about people failing to sort checksum.ini when commiting changes.
Nov 20 15:34:18 <florian> I have seen random failures to build the machine specific packages getting $MACHINE from environment.
Nov 20 15:34:31 * florian waits for checksums.ini to die
Nov 20 15:37:18 <XorA> DJWillis: write a commit hook for checking that :-D
Nov 20 15:40:26 * likewise puts a bullet in checksums.ini
Nov 20 15:40:49 <noglitch> hrw: I am totally stuck here: http://paste.debian.net/51976/ did you experience issues with python2.6 ?
Nov 20 15:42:01 <kergoth> meh
Nov 20 15:42:06 <kergoth> glad this week is over
Nov 20 15:42:11 <kergoth> well, nearly
Nov 20 15:42:12 <hrw> I use 2.5 mostly but no problems with 2.6
Nov 20 15:42:31 <RP> morning kergoth :)
Nov 20 15:42:54 <kergoth> speaking of which, why the hell can't i find a python2.6 package in sid anymore
Nov 20 15:42:57 <kergoth> hrmph
Nov 20 15:43:01 <kergoth> morning RP 
Nov 20 15:43:15 <broonie> kergoth: The python maintainer is a fuckwit.
Nov 20 15:43:20 <kergoth> ah, right
Nov 20 15:43:23 <kergoth> thatd do it
Nov 20 15:43:36 <hrw> kergoth: look at experimental
Nov 20 15:43:41 <broonie> It's in experimental.
Nov 20 15:43:47 <kergoth> yeah, i saw it there, still annoying :)
Nov 20 15:44:15 <noglitch> it seems to be related to python2.6 in karmic
Nov 20 15:44:37 <noglitch> <bb>: inconsistent use of tabs and spaces in indentation
Nov 20 15:45:10 <noglitch> for "sre_constants"...
Nov 20 15:45:50 <hrw> kergoth: and expect problems with it
Nov 20 15:46:19 <kergoth> someday I'm going to remember how to do apt pinning / distro selection without having to look it up.  someday.. :)
Nov 20 15:46:30 <kergoth> heh, experimental is aptly named, eh
Nov 20 15:57:02 <pb_> DJWillis: heh, just forget about checksums.ini
Nov 20 16:02:24 * kergoth sighs and starts trying to sync up the mvl6 recipes with OE
Nov 20 16:04:43 <zecke> cbrake: good question
Nov 20 16:05:08 <zecke> wow meta-toolchain is broken again
Nov 20 16:05:47 <pb_> zecke: doh, again?
Nov 20 16:05:58 <pb_> if only we had some kind of waterfall for oestats we could monitor this :-}
Nov 20 16:06:12 <zecke> "*  glibc-utils *  localedef *"... missing dependencies
Nov 20 16:06:14 <zecke> any idea?
Nov 20 16:06:58 <kergoth> i love git-clean
Nov 20 16:07:05 <pb_> I don't quite know what that message means.  I guess from the * decorations that this is opkg, right?
Nov 20 16:07:09 <hrw> "git clean -df ." rox
Nov 20 16:07:14 <pb_> is it saying that opkg-utils depends on localedef?
Nov 20 16:07:22 <kergoth> hrw: yup
Nov 20 16:07:41 <kergoth> git clean -fdX is a good one too, just cleans the ignored files
Nov 20 16:08:16 <zecke> pb_: task-sdk-bare is depending on these two
Nov 20 16:08:32 <zecke> pb_: and as a bad user I only pasted one of the two lines...
Nov 20 16:08:35 <pb_> ah, that does seem bogus.  those are both target executables, you wouldn't want them in an sdk
Nov 20 16:09:18 <zecke> pb_: well, it might get installed in the target area (still bogus)... I just wonder if they used to be present and are gone now, or a new
Nov 20 16:10:26 <zecke> and with cheer luck it points to work of RP again... I fear to tell him :}
Nov 20 16:11:19 <pb_> afaik, the packages do still exist (they're both generated as part of glibc).  so, either they have changed architecture or otherwise become unavailable to the sdk, or the dependency on them is new.  I don't know which of those is likely to be the case.
Nov 20 16:11:27 <pb_> ah, poor RP
Nov 20 16:11:39 <zecke> I will have to figure this out on the weekend
Nov 20 16:12:13 <pb_> well, if he gets angry he can always vent some steam on the mailing list, that seems to be the standard modus operandi at present
Nov 20 16:13:25 <zecke> pb_: any idea how to improve the situation?
Nov 20 16:13:25 <zecke> pb_: when koen asked you to use --author I wanted to reply but he even used "please"
Nov 20 16:14:01 <pb_> zecke: yeah, I wasn't too bothered about koen asking me to use --author.  I think that was a reasonable request.
Nov 20 16:15:08 <pb_> there does seem to be a general highly-strung atmosphere on the mailing list at present, though, and unfortunately I have no idea how to improve that situation.
Nov 20 16:15:26 <florian> He had a good day obvoiously... the comment about second names was good :)
Nov 20 16:15:43 <Crofton|work> the holidays are coming up
Nov 20 16:15:53 <Crofton|work> and we are coming off a dev is broken phase
Nov 20 16:16:29 <florian> yes I guess people who are just using .dev are having a hard time
Nov 20 16:16:53 <hrw> thats why I mostly spend time with Poky last weeks
Nov 20 16:16:58 <RP> zecke: hmm. Is there a localedef package?
Nov 20 16:17:39 * mwester didn't do it!
Nov 20 16:18:40 * florian too... mostly because we try to get our low budget distro ready
Nov 20 16:19:25 * pb_ go home now
Nov 20 16:19:30 <pb_> babysitting duties call
Nov 20 16:19:31 <pb_> later all
Nov 20 16:19:38 <florian> pb_: enjoy
Nov 20 16:20:14 <zecke> RP: sorry, thanks for the help but right now I'm unable to take it. I will spend my saturday working on oestats and figuring it out
Nov 20 16:21:32 <florian> zecke: oh... I was tempted to hack on oestats too.
Nov 20 16:23:02 <RP> zecke: I changed glibc's packaging and moved a lot of things into a .bbclass. Looking at it I can't immediately see why that would break the localedef package but I can look into it. Its a wider problem than meta-toolchain
Nov 20 16:24:47 <RP> zecke: localedef is in PACKAGES and a package is attempted but its empty :/
Nov 20 16:26:48 <zecke> RP: but the FILES_localedef looked fine
Nov 20 16:27:11 <RP> zecke: yes, which is puzzling
Nov 20 16:27:32 <RP> zecke: something odd is going on :/
Nov 20 16:28:40 <zecke> RP: don't look at it, I will have time on saturday
Nov 20 16:30:52 <zecke> florian: what do you want to hack on?
Nov 20 16:32:19 <florian> zecke: I miss a column with the build target in the lists.
Nov 20 16:32:48 <zecke> ah
Nov 20 16:32:53 <zecke> I will try to give pb_ a waterfall
Nov 20 16:34:47 <florian> cool
Nov 20 16:35:28 <florian> zecke: if we extend it to schedule builds we have a pretty cool tool.
Nov 20 16:36:43 <zecke> florian: why would you want it to schedule builds? it makes the setup more complicated :)
Nov 20 16:38:10 <florian> zecke: to avoid setting up buildbot. but yes you are right, the easy setup procedure is nice.
Nov 20 16:38:36 <RP> zecke: I dream of an oestats and buildbot merged together
Nov 20 16:38:55 <RP> zecke: builtbot it so ununsuited to our needs its not funny :(
Nov 20 16:39:10 <RP> zecke: but it also does certain things well
Nov 20 16:39:34 <zecke> RP: the scheduling of things are great
Nov 20 16:39:40 <zecke> but I can't stand twisted
Nov 20 16:41:07 <RP> zecke: :)
Nov 20 16:42:37 <florian> zecke: it should be possible to make the setup easy even if you intend to control builds but its likely to be some effort
Nov 20 16:44:45 <zecke> florian: the problem is that... companies need to put these servers in a DMZ...
Nov 20 16:46:23 <florian> zecke: hum well, this could even be an internal server
Nov 20 16:46:39 <hrw> florian: oestats in one way service. how you want to schedule it?
Nov 20 16:47:59 <florian> hrw: cureently it is, yes
Nov 20 16:48:22 <hrw> florian: I wonder how it can be twoways
Nov 20 16:49:17 <noglitch> hrw: seems to work better with python2.5...
Nov 20 16:56:15 <recalcati> I have qt 4.6 runniing on dm365 evm
Nov 20 16:57:06 <recalcati> I'll try to arrive to the real end of bitbake compilation and then I'll send the recipe
Nov 20 16:57:25 <recalcati> by now there is some trick to be done
Nov 20 16:58:44 <RP> zecke: The problem is FILES_glibc interacts badly with FILES_${PN} :/
Nov 20 17:00:29 <kergoth> recipe_sanity checks for that
Nov 20 17:00:38 <kergoth> it notifies you when the expansion of a variable name overwrites an existing variable
Nov 20 17:00:40 <zecke> RP: hey, I asked you to not look into it... makes sense you would want to have a "+="?
Nov 20 17:01:05 <RP> zecke: I'm bothered the whole of OE is broken due to it :/
Nov 20 17:01:17 <hrw> hmm..
Nov 20 17:01:17 <RP> zecke: += is not appropriate :/
Nov 20 17:01:24 <hrw> "learned" or "learnt"?
Nov 20 17:02:14 <RP> hrw: learnt
Nov 20 17:02:31 <RP> hrw: Well, depends on the context :)
Nov 20 17:02:35 <zecke> RP: the nihilist in me says everything is broken anyway :)
Nov 20 17:02:39 <zecke> (or needs to be broken)
Nov 20 17:02:50 <hrw> http://www.askoxford.com/asktheexperts/faq/aboutspelling/learnt?view=uk
Nov 20 17:03:02 <hrw> RP: looks like both are proper even in British English
Nov 20 17:03:39 <RP> hrw: I'd have said learnt is past tense, learned can describe someone
Nov 20 17:03:51 <RP> hrw: looks like americans would disagree with me ;-)
Nov 20 17:04:45 <hrw> k
Nov 20 17:04:46 <recalcati> zecke: thank you very much. I have to finish well the recipe, anyway. I'll do it monday
Nov 20 17:05:05 <RP> zecke: We don't have a way to unset something in the metadata do we (other than a python delvar call) :/
Nov 20 17:05:08 <recalcati> I have also learned devshell
Nov 20 17:06:04 <recalcati> good night. byebye
Nov 20 17:08:24 * RP realises how to solve this
Nov 20 17:11:50 <kergoth> why does gmane's search suck so much ass, when their thread browsing interface is so awesome?
Nov 20 17:11:53 <kergoth> it confuses me
Nov 20 17:14:19 <zecke> RP: not really :(
Nov 20 17:20:28 <hrw> ~hail git-svn
Nov 20 17:20:29 * ibot bows down to git-svn and chants, "I'M NOT WORTHY!!"
Nov 20 17:22:23 <CIA-80> 03Richard Purdie <rpurdie@linux.intel.com> 07org.openembedded.dev * r2b05f41b11 10openembedded.git/classes/glibc-package.bbclass: 
Nov 20 17:22:23 <CIA-80> glibc-package.bbclass: Add support for precompiled glibc trees (from Poky)
Nov 20 17:22:23 <CIA-80> Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
Nov 20 17:22:34 <CIA-80> 03Richard Purdie <rpurdie@linux.intel.com> 07org.openembedded.dev * rf0af65955d 10openembedded.git/classes/glibc-package.bbclass: 
Nov 20 17:22:34 <CIA-80> glibc-package.bbclass: Fix case PN = glibc and corrupts FILES variables
Nov 20 17:22:34 <CIA-80> Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
Nov 20 17:22:34 <kergoth> precompiled glibc trees?
Nov 20 17:22:34 <RP> zecke: I've just pushed a fix, see if that helps...
Nov 20 17:22:38 <kergoth> hmm
Nov 20 17:22:52 <hrw> RP: csl-glibc?
Nov 20 17:22:56 <RP> kergoth: ?
Nov 20 17:23:20 * kergoth points at the above commit notifications
Nov 20 17:23:42 <RP> kergoth: precompiled binary locale files would be a better description
Nov 20 17:23:47 <RP> kergoth: like CSL toolchains
Nov 20 17:23:57 <kergoth> ah, right, makes more sense that way :)
Nov 20 17:24:01 <mario-goulart> Hi.  I'm trying to cross compile using Chicken on OE.  Chicken is a Scheme to C compiler and can generate native code by compiling Scheme to C then calling a C compiler behind the scenes to compile C to native code.  The cross compilation process is described at http://chicken.wiki.br/cross-compilation
Nov 20 17:24:42 <RP> kergoth: I hate recipes like this: http://git.pokylinux.org/cgit.cgi/poky/tree/meta/packages/meta/external-csl-toolchain_2008q3-72.bb
Nov 20 17:25:23 <kergoth> yech
Nov 20 17:25:36 * kergoth wonders how else one would manage that, though
Nov 20 17:25:44 <mario-goulart> Firstly, it seems that chicken uses a different meaning for "native".  It seems that chicken's "native" means a binary for the target. According to http://docs.openembedded.org/usermanual/html/recipes_variables.html, OE's meaning for native is "things built to run natively on the build host itself."
Nov 20 17:26:00 <RP> kergoth: "inherit glibc-package" now?
Nov 20 17:26:09 <RP> kergoth: Removes a large chunk of that file
Nov 20 17:26:17 <kergoth> ah, yes
Nov 20 17:26:20 <kergoth> cool
Nov 20 17:26:32 * RP needs to fix it up and push
Nov 20 17:26:39 <mario-goulart> Am I missing something or the concepts for "native" are really different?
Nov 20 17:31:14 * florian -> home, brb
Nov 20 17:35:51 * hrw -> off in few minutes
Nov 20 17:36:06 <hrw> just finished blog post for tomorrow with few OE/Poky tricks
Nov 20 17:37:38 <hrw> have a nice weekend guys
Nov 20 17:37:46 <woglinde> bye hrw
Nov 20 17:39:01 <RP> hi woglinde 
Nov 20 17:39:45 <zecke> RP: would bumping the PR make sense?
Nov 20 17:40:30 <RP> zecke: yes, most likely since people probably have bust glibcs now
Nov 20 17:41:12 <zecke> RP: do you need a hand on the PR bump?
Nov 20 17:41:24 <RP> zecke: You are more than welcome to push one :)
Nov 20 17:44:05 <zecke> RP: it will take until sunday... if one can wait that long I'm happy to do it
Nov 20 17:45:30 <RP> zecke: by then I'd imagine most people will have bitbake glibc -c clean 'd anyway :/
Nov 20 17:45:33 <woglinde> jo re
Nov 20 17:47:11 <zecke> RP: did you bump PR when doing the .bbclass addition?
Nov 20 17:49:13 <pb__> well, luckily I think the bustage is probably going to be minor for most folks.  glibc-utils is a fairly fringe interest (except, weirdly, for the sdk-* tasks which seem to like it) and you only need localedef if you want to build locales on the target.
Nov 20 17:49:25 <pb__> so, bumping PR is probably a good thing but I don't think it's super-urgent.
Nov 20 17:50:55 <woglinde> hes
Nov 20 17:50:57 <woglinde> args
Nov 20 17:51:12 <woglinde> rp I will now test your uclibc patches
Nov 20 17:52:56 <DJWillis> pb__: it may be nice to tie up the PR bump to go in before the upcomming Angstrom DISTRO_PR bump (ok, it's one distro but you know what I mean)
Nov 20 17:53:36 <RP> zecke: no, it wasn't meant to change anything 
Nov 20 17:54:15 <pb__> DJWillis: I doubt it makes a lot of difference.  if angstrom is going to bump DISTRO_PR then they will be recompiling anyway and the question of whether the package PR is bumped or not will be irrelevant for them.  plus, I don't think any/many of the angstrom targets will be affected by the lack of those two packages anyway.
Nov 20 17:54:50 <DJWillis> pb__: doh!, of course, i'll crawl back to my cave now.
Nov 20 17:55:54 <pb__> DJWillis: sorry, that wasn't meant to be dismissive.
Nov 20 17:56:42 <DJWillis> pb__: no offense taken at all :-)
Nov 20 17:56:53 <pb__> ah, jolly good :-)
Nov 20 17:57:34 <zecke> RP: now configure failed... hmm
Nov 20 17:59:13 <florian> re
Nov 20 17:59:29 <woglinde> jo florian
Nov 20 17:59:40 <zecke> RP: sorry, i can't look into this right now :(
Nov 20 17:59:40 <RP> zecke: bitbake glibc-initial -c clean
Nov 20 17:59:59 <RP> zecke: I'd bet config.log says it can't find limits.h
Nov 20 18:00:22 <zecke> RP: ah maybe
Nov 20 18:00:32 <zecke> RP: it says mips*cpp -E fails sanity check
Nov 20 18:00:42 <mario-goulart> Regarding to the Chicken topic, it seems that the cross compiler needs the target libraries to generate target binaries (except for static binaries, everything is linked against libchicken.so, which needs to be from target).  Where should OE keep the target libs for this case?  Under tmp/cross or tmp/staging?
Nov 20 18:00:42 <woglinde> hm building uclibc
Nov 20 18:00:47 <woglinde> gets me
Nov 20 18:00:48 <woglinde> ERROR: Error Method already seen: get_glibc_fpu_setting in' glibc-package.bbclass' now in 'glibc-package.inc'
Nov 20 18:02:00 <RP> woglinde: DId I leave two copies? :/
Nov 20 18:02:11 <woglinde> rp of what?
Nov 20 18:02:28 <woglinde> the fpu method?
Nov 20 18:03:11 <RP> woglinde: yes, I don't seem to have done
Nov 20 18:03:34 <RP> woglinde: I wonder if your cache needs cleaning or something :/
Nov 20 18:03:39 <pb__> seems a bit weird that uclibc would be going near glibc-package.inc at all
Nov 20 18:03:48 <woglinde> rp I removed whole oetmp
Nov 20 18:04:08 <zecke> pb__: that is printed during parse
Nov 20 18:04:46 <RP> pb__: I think uclibc copied and pasted a libc function
Nov 20 18:05:33 <pb__> zecke: oh right, yeah, of course
Nov 20 18:07:13 <woglinde> rp ah hm
Nov 20 18:07:47 <RP> woglinde: I can't find any reference to that in my tree other than in glibc-package.inc
Nov 20 18:08:05 <RP> woglinde: Do you have any local changes referencing it?
Nov 20 18:08:29 <woglinde> just grepping
Nov 20 18:08:38 <woglinde> but I am sure I never touched this
Nov 20 18:08:56 <woglinde> hm I should run with -DDD to see which package it is
Nov 20 18:11:45 * kergoth tries to wrap his head around the rebase --preserve-merges option
Nov 20 18:18:47 * RP calls it a day
Nov 20 18:18:48 <zecke> RP: nice hint... do you know what happens?
Nov 20 18:18:52 <zecke> ah take care!
Nov 20 18:19:21 <woglinde> Rp again?
Nov 20 18:20:21 <RP> zecke: glibc's headers overwrite those from glibc-initial. You clean glibc and then it has no headers to build against
Nov 20 18:20:59 <RP> zecke: By building glibc-initial, you get headers for glibc so its happy again
Nov 20 18:21:04 * RP really gone
Nov 20 18:22:04 <woglinde> bye RP
Nov 20 18:22:07 <woglinde> take a rest
Nov 20 18:46:41 <zecke> RP: meta-toolchain built again
Nov 20 18:47:47 <kergoth> hmm.
Nov 20 18:51:53 <woglinde> hm
Nov 20 18:52:00 <kergoth> the branch setup the git guys use is cute, i like it
Nov 20 18:52:04 <woglinde> # Need to overwrite the version from -initial
Nov 20 18:52:04 <woglinde>         if [ ! -e ${D}${libdir}/libc.so ]; then
Nov 20 18:52:04 <woglinde>                 ln -s ../../lib/libc.so.0 ${D}${libdir}/libc.so
Nov 20 18:52:05 <woglinde>         fi
Nov 20 18:52:19 <woglinde> this gives me some headaches
Nov 20 18:52:45 <kergoth> we should think about using a method like theirs for bitbake
Nov 20 18:52:46 <kergoth> hmm
Nov 20 18:53:08 <woglinde> dont know why rp made this
Nov 20 18:53:35 <kristoffer> kergoth, I did another test today. Still got same issue, getting no tar package at all. bitbake -e still returns package_tar
Nov 20 18:53:43 <woglinde> but it seems utterly wrong
Nov 20 18:53:44 <kergoth> that's very odd
Nov 20 18:54:16 <woglinde> hm but maybee I have to check the staging-dir
Nov 20 18:55:28 <pb__> woglinde: what goes wrong with that?  it doesn't look as bogus as all that.
Nov 20 18:56:09 <woglinde> chmod: cannot operate on dangling symlink `/devel/arm/oetmp-ang/work/beagleboard-oe-linux-uclibceabi/uclibc-0.9.30.1-r32.1/image//lib/libc.so' 
Nov 20 18:56:17 <woglinde> later in the patch
Nov 20 18:56:34 <woglinde> chmod +x ${D}/${base_libdir}/*
Nov 20 18:56:36 <woglinde> hm ah
Nov 20 18:56:41 <woglinde> was in before
Nov 20 18:56:49 <pb__> oh. this is micro?
Nov 20 18:56:59 <woglinde> nope uclibc recipes
Nov 20 18:57:01 <pb__> i guess you need to adjust the path there
Nov 20 18:57:13 <pb__> right, but what distro?
Nov 20 18:57:17 <woglinde> ah yes I am buildn micro
Nov 20 18:57:18 <woglinde> distro
Nov 20 18:57:18 <woglinde> sorry
Nov 20 18:57:30 <kristoffer> kergoth, when I look further I see that package_tar actually tries to create a package but skips through everything with "not creating empty package for...". It does this for all folders (even though those that should contain atleast binaries)
Nov 20 18:58:02 <pb__> yeah, it needs to be different in that case.  "../../lib/libc.so.0" assumes libdir is in /usr
Nov 20 18:58:06 <kergoth> huh, sounds like it may just be broken.. open a bug? :)
Nov 20 18:58:24 <kristoffer> kergoth, yeppers, will do
Nov 20 18:58:41 <woglinde> pb hm 
Nov 20 18:58:47 <woglinde> how to solve this
Nov 20 18:58:57 <pb__> might be easier to use an absolute path
Nov 20 18:59:13 <woglinde> pb WORKDIR?
Nov 20 18:59:19 <pb__> oh, no, that still won't work right
Nov 20 18:59:44 <pb__> hm, tricky.  what did uclibc do before rp's recent chanes?
Nov 20 18:59:52 <pb__> changes, even
Nov 20 19:00:10 <woglinde> install -m 644 lib/libc.so ${UCLIBC_STAGE_PREFIX}/lib
Nov 20 19:00:15 * pb__ typing with one hand, my daughter is occupying the other arm
Nov 20 19:00:27 <woglinde> ah mom
Nov 20 19:00:35 <kergoth> ooh, i didn't know about git stash --patch
Nov 20 19:00:39 <kergoth> er, git stash save --patch
Nov 20 19:00:44 <Crofton|work> wew, the patch I applied from rday did the right thing with author
Nov 20 19:00:47 <woglinde> pb in do_install install -m 644 lib/libc.so ${D}${libdir}/
Nov 20 19:00:51 <zecke> bye bye
Nov 20 19:01:05 <Crofton|work> gn zecke 
Nov 20 19:01:05 <woglinde> args this was inital
Nov 20 19:01:12 <XorA> mickey|sofa: florian: awake?
Nov 20 19:01:28 <mickey|sofa> somewhat
Nov 20 19:01:33 <woglinde> pb hm nothing
Nov 20 19:01:36 <mickey|sofa> shallwe join #oe-ev?
Nov 20 19:01:36 <Crofton|work> oh yeah
Nov 20 19:01:44 <mickey|sofa> or do ithere?
Nov 20 19:01:45 <XorA> mickey|sofa: good plan
Nov 20 19:01:47 <mickey|sofa> okies
Nov 20 19:02:23 <woglinde> pb ah found
Nov 20 19:02:35 <woglinde> uclibc directly manipulated staging
Nov 20 19:02:42 <pb__> ah
Nov 20 19:03:00 <mickey|sofa> florian: ping
Nov 20 19:03:16 <woglinde> but I wonder because in staging is no libc.so yet
Nov 20 19:03:25 <kergoth> hm
Nov 20 19:03:26 <woglinde> but maybee it got deleted
Nov 20 19:03:43 <pb__> in cross?
Nov 20 19:04:01 <woglinde> cross is nothing too
Nov 20 19:04:04 <pb__> hm
Nov 20 19:04:06 <woglinde> just checked it
Nov 20 19:04:13 <woglinde> but nobody complained
Nov 20 19:04:22 <mickey|sofa> florian: PING
Nov 20 19:04:32 <woglinde> othetwise gcc-inter might have some problems
Nov 20 19:04:46 <kergoth> http://gitready.com/ cute little site
Nov 20 19:05:16 <florian> mickey|sofa: pong
Nov 20 19:05:46 <woglinde> pb the goal is that libc.so points to libc.so.0  right?
Nov 20 19:07:05 <cdbot2> * * OE Bug 5339 has been created by kristoffer.ericson(AT)gmail.com
Nov 20 19:07:07 <cdbot2> * * package_tar no longer produces packages
Nov 20 19:07:09 <cdbot2> * * http://bugs.openembedded.org/show_bug.cgi?id=5339
Nov 20 19:14:33 * kergoth wonders how painful a collection would be that used git-submodule with separate repositories for the recipes
Nov 20 19:22:33 <CIA-80> 03Kristoffer Ericson <kristoffer@boggieman.(none)> 07org.openembedded.dev * rc3660ee99a 10openembedded.git/conf/distro/jlime-2009.1.conf: update jlime-2009.1 distro file and check so it builds
Nov 20 19:22:33 <CIA-80> 03Kristoffer Ericson <kristoffer@boggieman.(none)> 07org.openembedded.dev * rc84b4b9dd9 10openembedded.git/MAINTAINERS: update Maintainer information
Nov 20 19:22:34 <CIA-80> 03Kristoffer Ericson <kristoffer@boggieman.(none)> 07org.openembedded.dev * rfbdd91d7a6 10openembedded.git/conf/distro/jlime-2009.1.conf: Change config into something that actually builds.
Nov 20 19:22:36 <CIA-80> 03Kristoffer Ericson <kristoffer@boggieman.(none)> 07org.openembedded.dev * rb0d5120892 10openembedded.git/conf/machine/ben-nanonote.conf: Add machine file for Qi-Hardware's ben nanonote.
Nov 20 19:22:38 <CIA-80> 03Kristoffer Ericson <kristoffer@boggieman.(none)> 07org.openembedded.dev * reb0bcf7a3b 10openembedded.git/recipes/linux/ (22 files in 2 dirs): 
Nov 20 19:22:41 <CIA-80> Add preferred kernel for the ben-nanonote and merge that
Nov 20 19:22:43 <CIA-80> into linux_2.6.31.bb. Patches and config are grabbed
Nov 20 19:22:45 <CIA-80> from qi-hardware.com. fix_mips_vmlinux.lds.patch is
Nov 20 19:22:47 <CIA-80> added to fix problem in 2.6.31.5 vanilla.
Nov 20 19:26:48 <woglinde> pb I think an easy solution is ln -s ../../lib/libc.so.0 ${D}${libdir}/libc.so
Nov 20 19:26:50 <woglinde> args
Nov 20 19:27:02 <woglinde> pb I think an easy solution is ln -s ../${prefix}/lib/libc.so.0 ${D}${libdir}/libc.so
Nov 20 19:27:10 <woglinde> that  should do the trick
Nov 20 19:28:18 <woglinde> hms
Nov 20 19:30:06 <woglinde> hm nope
Nov 20 19:32:03 <woglinde> hm seems to be right
Nov 20 19:32:09 <woglinde> hm hmh hmh
Nov 20 19:32:12 <woglinde> *g*
Nov 20 19:43:21 <woglinde> ~seen khem
Nov 20 19:43:24 <ibot> khem <n=khem@99-57-141-118.lightspeed.sntcca.sbcglobal.net> was last seen on IRC in channel #oe, 7d 21h 42m 59s ago, saying: 'hmm autofs works much better with tcp for mountproto'.
Nov 20 19:45:03 <woglinde> Kristoffer Ericson <kristoffer@boggieman.(none)>
Nov 20 19:45:12 <woglinde> hms
Nov 20 19:46:14 <kristoffer> woglinde, yeah I noticed it after I sent them. Fixed it in git config now
Nov 20 20:06:25 <woglinde> okay
Nov 20 20:14:22 <florian> args... all ran away 
Nov 20 20:27:19 <rsalveti> just a quick question, what's actually the meaning of the PE variable?
Nov 20 20:27:34 <kergoth> epoch.
Nov 20 20:27:52 <kergoth> PE > PV > PR
Nov 20 20:28:13 <kergoth> a recipe with a PE of 1 and a PV of 2 is newer than a recipe with a PE of 0 and a PV of 5
Nov 20 20:28:39 <kergoth> i need to break this habit of ending my git commit message summaries with a period
Nov 20 20:28:58 <rsalveti> kergoth: hm, guess I got it
Nov 20 20:29:04 <kergoth> it's most commonly used when someone screws up
Nov 20 20:29:28 <kergoth> if they release something newer thats accidentally older in the version comparison, you can bump the epoch to make sure its newer
Nov 20 20:29:34 <kergoth> never use it unless you have a good reason to do so
Nov 20 20:30:50 <rsalveti> kergoth: oh, now I found the reason why :-)
Nov 20 20:30:53 <rsalveti> kergoth: thanks
Nov 20 20:30:58 <kergoth> np
Nov 20 20:31:08 <rsalveti> kergoth: so, when creating a new recipe should I add PE = 1 or something like that?
Nov 20 20:31:14 <kergoth> its not unique to oe/bitbake, debian and rpm have it too
Nov 20 20:31:19 <kergoth> no
Nov 20 20:31:19 <kergoth> a
Nov 20 20:31:19 <kergoth> gain
Nov 20 20:31:23 <kergoth> dont set it unless you have a reason to
Nov 20 20:31:24 <kergoth> at all.
Nov 20 20:31:54 <rsalveti> kergoth: hehe, got it :-)
Nov 20 20:32:31 <kergoth> and there generally isn't a need to define something unless you're setting it differently than the defaults defined in bitbake.conf.  the only real exception to that is PR, which people add regardless, to make revision bumps later on slightly quickly
Nov 20 20:32:34 <kergoth> s/ly$/er/
Nov 20 20:33:18 <rsalveti> yep, true
Nov 20 20:42:08 <pb__> woglinde: no, that won't work, you've got it backwards.  it's the number of "../" components that needs to change.
Nov 20 20:43:03 <pb__> you're trying to create a symlink in ${libdir} which points to a file in ${base_libdir}.  in a normal distro, the former is /usr/lib and the latter is /lib, so "../../lib" (or "../../${base_libdir}") is the appropriate source path.
Nov 20 20:43:51 <pb__> a cheap hack would just be to just add a straight conditional for the case where libdir and base_libdir are the same.  that'd fix micro, though it would still not work in the completely general case.
Nov 20 20:44:48 <pb__> for a completely general case, you need a function to compute the relative path to ${base_libdir} starting from ${libdir}, which is not completely trivial but ought to be fairly straightforward to do in python.
Nov 20 20:45:01 <pb__> probably a bit of a pita to implement in shell although I'm sure you could do it there if you really wanted
Nov 20 20:49:12 <woglinde> re
Nov 20 20:49:18 <woglinde> pb hm I just looked up
Nov 20 20:49:31 <woglinde> in the old recipe we dont have libc.so anyway
Nov 20 20:49:56 <woglinde> in the builddir it is a linkerscript as for glibc
Nov 20 20:50:09 <woglinde> dont know why rp is linking it to libc.so.6
Nov 20 20:50:28 <woglinde> so I am little bit puzzeld
Nov 20 20:50:38 <pb__> oh, that is a bit weird
Nov 20 20:50:56 <pb__> maybe that chunk is just bogus then.  what happens if you delete it and allow uclibc to install the linker script?
Nov 20 20:51:02 <woglinde> but it doesnt seem to make it to the image
Nov 20 20:51:28 <woglinde> the question is do the we really need libc.so
Nov 20 20:52:04 <woglinde> as I said I cannt find it in staging or packages for old way
Nov 20 20:54:46 <woglinde> hm I found something from the uclibc self for install
Nov 20 20:55:11 <woglinde> http://pastebin.com/m30f8b9db
Nov 20 20:56:13 <pb__> you definitely need libc.so in some form, or you will have nothing to link against.
Nov 20 20:56:36 <pb__> however, uclibc itself might be installing that as part of "make install", it might not be explicit in the recipes.
Nov 20 20:57:37 <kergoth> hmm
Nov 20 20:57:55 <pb__> it wouldn't surprise me too much if the -dev packages (and sdk) for uclibc were broken before, I suspect nobody was using them.  but staging must have been correct, or nearly so, otherwise nothing would have worked.
Nov 20 20:58:25 <woglinde> pb jupp
Nov 20 20:58:59 <woglinde> hm so the right solution I see let uclibc install the libc.so
Nov 20 20:59:08 <pb__> yeah, probably
Nov 20 20:59:08 <Crofton|work> pb__, what situation do you need to use --author in?
Nov 20 20:59:12 <woglinde> and we have to fix the sed's out again
Nov 20 20:59:42 <Crofton|work> diff not applied via "git am"?
Nov 20 20:59:44 <woglinde> the linkin to libc.so.0 is wrong
Nov 20 20:59:49 <woglinde> in my eyes
Nov 20 21:00:01 <pb__> Crofton|work: I think koen was upset about 5a8ab317885f234a411dce0ef1c76ef74583b299
Nov 20 21:00:17 <pb__> and/or a7affee8d4ce7f514648d07a1c59b05032d37778, I can't quite remember which was which
Nov 20 21:00:58 <Crofton|work> I just realized I hadn't used --author when I commited rday's guile patches, but the author is good on them 
Nov 20 21:01:06 <pb__> woglinde: yeah, if uclibc has a _nonshared.a then making a direct link from .so.0 -> .so is wrong
Nov 20 21:01:27 <pb__> Crofton|work: if you apply the patch with git am then it will set the author automatically
Nov 20 21:02:12 <pb__> it's only if you run git commit by hand that it becomes an issue.
Nov 20 21:02:35 <Crofton|work> yeah
Nov 20 21:03:07 <pb__> I think one or other of those patches needed some manual adjustment: either it wouldn't apply at all with git am or the result would not have been right, so I did it manually.
Nov 20 21:03:08 <Crofton|work> I'm tempted to say make the patches apply via git am, but it is still good to give people credit :)
Nov 20 21:03:13 <Crofton|work> yeah
Nov 20 21:03:34 <Crofton|work> I had troble saving from mail, but downlaoding from patchwork did work
Nov 20 21:03:41 <pb__> the other one I probably could have applied with git am but I just used the same method out of habit.
Nov 20 21:03:43 <woglinde> pb__ it has
Nov 20 21:04:21 <Crofton|work> there is also the script in contrib/patchwork that makes it super easy, with bonus of picking up all the ack's etc
Nov 20 21:05:08 <woglinde> GROUP ( uclibc_nonshared.a libc.so.0 AS_NEEDED ( ld-uClibc.so.0 ) )
Nov 20 21:11:36 <pb__> woglinde: that looks about right
Nov 20 21:17:12 <woglinde> damn
Nov 20 21:17:24 <woglinde> somehow the install dont work
Nov 20 21:18:58 <woglinde> seems something is wrong with the snippet I posted
Nov 20 21:25:12 <woglinde> hm the uclibc install stuff seems broken anyway
Nov 20 21:30:52 <woglinde> hm a seems a dependency problem
Nov 20 21:31:00 <woglinde> dev libs are installed before runtime
Nov 20 21:31:24 <woglinde> so if [ -f ./lib/libc.so -a -f /devel/arm/oetmp-ang/work/beagleboard-oe-linux-uclibceabi/uclibc-0.9.30.1-r32.1/image/lib/libc.so.0 ] ;
Nov 20 21:31:36 <woglinde> goes wrong because there is no libc.so.0
Nov 20 21:31:37 <woglinde> hm
Nov 20 21:31:40 <woglinde> let me test it
Nov 20 21:32:18 <woglinde> ah
Nov 20 21:32:21 <woglinde> parallel make
Nov 20 21:32:40 <woglinde> install: install_runtime install_dev
Nov 20 21:32:54 <woglinde> and install_dev has no dep on install_runtime
Nov 20 21:34:13 <woglinde> good catch
Nov 20 21:37:24 <woglinde> yes it was
Nov 20 21:39:31 * mwester wonders what percentage of the oft-used packages have been forced to disable parallel makes...
Nov 20 21:40:04 <woglinde> mwester *g*
Nov 20 21:40:11 <woglinde> sometimes its easy to fix
Nov 20 21:40:14 <woglinde> as in this case
Nov 20 21:41:40 <woglinde> wonder why khem didnt figure this out
Nov 20 21:42:15 <pb__> woglinde: ah, very good
Nov 20 21:42:37 <woglinde> args
Nov 20 21:42:45 <woglinde> NOTE: the following files were installed but not shipped in any package:
Nov 20 21:42:45 <woglinde> NOTE:   /usr/bin/ldd
Nov 20 21:43:42 <pb__> doh
Nov 20 21:43:43 <pb__> silly uclibc
Nov 20 21:44:05 <woglinde> *g*
Nov 20 21:44:24 <pb__> it must have a hard coded /usr in its makefile somewhere
Nov 20 21:44:31 <woglinde> nope
Nov 20 21:44:39 <woglinde> it worked before richards stuff
Nov 20 21:45:16 <woglinde> ah okay
Nov 20 21:45:23 <woglinde> nasty hack in uclibc.inc
Nov 20 21:45:51 <woglinde> oh god I should lart khem for this
Nov 20 21:46:23 <pb__> ./utils/Makefile.in:	$(Q)$(INSTALL) -D -m 755 $(utils_OUT)/ldd $(PREFIX)$(RUNTIME_PREFIX)usr/bin/ldd
Nov 20 21:46:43 <pb__> that seems fairly clearly broken
Nov 20 21:46:46 <woglinde> jupp
Nov 20 22:00:51 <CIA-80> 03Philip Balister <philip@balister.org> 07org.openembedded.dev * r9712678939 10openembedded.git/recipes/gnuradio/gnuradio_git.bb: 
Nov 20 22:00:51 <CIA-80> gnuradio : Update git recipe to build from current HEAD.
Nov 20 22:00:51 <CIA-80> * Update recipe to allow building from dev repos and branches.
Nov 20 22:04:37 <CIA-80> 03Philip Balister <philip@balister.org> 07org.openembedded.dev * rf0583a9040 10openembedded.git/recipes/gnuradio/gnuradio_git.bb: gnuradio : Restore DEFAULT_PREFERENCE = "-1".
Nov 20 22:10:16 <woglinde> jo ant
Nov 20 22:11:16 <CIA-80> 03Philip Balister <philip@balister.org> 07org.openembedded.dev * r238eb99969 10openembedded.git/recipes/gnuradio/gnuradio_git.bb: gnuradio : Fix stupid typo. Sorry for the noise.
Nov 20 22:12:00 <Crofton|work> shoot me
Nov 20 22:12:59 <woglinde> ~shoot crofton
Nov 20 22:13:01 * ibot shoots crofton in the foot with a cork gun!
Nov 20 22:13:03 <Crofton|work> heh
Nov 20 22:13:14 <woglinde> as you wish master ;)
Nov 20 22:13:22 <Crofton|work> not mmy finest hour there ....
Nov 20 22:14:34 <ant__> woglinde: hallo
Nov 20 22:22:19 <woglinde> hm
Nov 20 22:22:21 <woglinde> NOTE: Not creating empty archive for libopie2-1.2.4-r2
Nov 20 22:22:24 <woglinde> not good
Nov 20 22:23:21 <ant__> woglinde: about opie issue on the ML, bluelightning is informed. He's just studying the best way to make the fix compatible with older kernels
Nov 20 22:23:32 <woglinde> ah
Nov 20 22:23:38 <woglinde> thats the an other error
Nov 20 22:43:59 <kurtis99> hello
Nov 20 22:53:15 * ant__ will touch the rough metal launching another BFS form armv5te tonite
Nov 20 22:54:03 <ant__> woglinde: kexec-tools guys finally added arm detection upstream
Nov 20 22:54:13 <ant__> they even added uImage support
Nov 20 22:54:25 <mwester> cool!
Nov 20 22:54:32 <ant__> there is not yet a release but perhaps we could import th epčatch...
Nov 20 22:56:25 <ant__> mwester: FYI we already have arm patches in OE, even for klibc-statical builds  :)
Nov 20 22:58:49 <ant__> mwester: we would badly need your help in a parallel project: kexecboot. Kernels are now passing the cmdline at the new location so 'old' kernels cannot find the bootparams. iirc you did a patch to relocate these...?
Nov 20 23:18:25 <RP> woglinde: hi
Nov 20 23:18:39 <RP> woglinde: I see you've been cursing me a bit :/
Nov 20 23:19:05 <RP> woglinde: The reason for creating that was to overwrite libc.so installed by uclibc-initial
Nov 20 23:19:20 <RP> woglinde: I suspect you're onto a better solution...
Nov 20 23:28:38 <woglinde> rp yes
Nov 20 23:28:55 <woglinde> problem was in install from uclibc
Nov 20 23:29:01 <woglinde> not parallelsafe
Nov 20 23:29:35 <woglinde> fixed locally
Nov 20 23:29:45 <woglinde> and so the libc.so link can go away
Nov 20 23:30:10 <woglinde> the orginal libc.so will overwrite the libc.so from inital
Nov 21 01:18:12 <NvrBst> Ahh question.  I have a pre-built shared object I want to copy to the board during install.  I made a recipie, add the "file://libME.so.0" to SRC_URI, add the "install -m 0755 ${WORKDIR}/libME.so.0 ${D}${libdir}" to do_install.
Nov 21 01:18:12 <NvrBst> But I get a "No GNU_HASH in the elf binary".  Is there a way around this?  The shared object should never be re-build, and works on the board, so I didn't want to bother making a new recpie to build; just edit an old recpie that already copies some files over to board.
Nov 21 01:18:17 <NvrBst> I also tried adding the "-Wl,--hash-style=gnu" or "--hash-style=gnu" to my external shared object build but it says "unrecognized command line option '-W1,--hash-style=gnu'".
Nov 21 01:27:13 <NvrBst> I found the "INSANE_SKIP_${PN} = True" thing which makes it work :)  If there is a more approprate way please feel free to head me in the right direction ;)
**** ENDING LOGGING AT Sat Nov 21 02:59:57 2009