IRC log for #oe on 20090211

00:11.16*** join/#oe kergoth (n=kergoth@ip68-3-163-239.ph.ph.cox.net)
00:12.45*** part/#oe mturquette (n=mturquet@CC0F872D.ptr.provps.com)
00:22.42CIA-203Michael 'Mickey' Lauer <mickey@vanille-media.de> 07org.openembedded.dev * r6626f2c113 10openembedded.git/conf/distro/include/preferred-om-2008-versions.inc: preferred-om-2008-versions.inc: bump libtool and automake
00:22.43CIA-203Tim 'timtim' Ellis <tim.ellis@foonas.org> 07org.openembedded.dev * r4a9452f983 10openembedded.git/ (4 files in 3 dirs): linux-linkstationppc: Fix last commit, 2.6.27.10 => 2.6.28.4, add usb webcam modules
00:22.43CIA-203Tim 'timtim' Ellis <tim.ellis@foonas.org> 07org.openembedded.dev * r822cebf2a9 10openembedded.git/ (2 files in 2 dirs): checksums.ini: Add linux patch-2.6.28.4
00:23.06CIA-203Tim 'timtim' Ellis <tim.ellis@foonas.org> 07org.openembedded.dev * r822cebf2a9 10openembedded.git/ (2 files in 2 dirs): checksums.ini: Add linux patch-2.6.28.4
00:23.07CIA-203Tim 'timtim' Ellis <tim.ellis@foonas.org> 07org.openembedded.dev * r4a9452f983 10openembedded.git/ (4 files in 3 dirs): linux-linkstationppc: Fix last commit, 2.6.27.10 => 2.6.28.4, add usb webcam modules
00:23.08CIA-203Tim 'timtim' Ellis <tim.ellis@foonas.org> 07org.openembedded.dev * rc800819c4e 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev
00:23.12timtimrederm.
00:24.11timtimredok. since im seeing double, its off to sleep.
00:26.15mwesterno, it's correct -- that's your irc nick.
00:26.37timtimredblinks
00:37.22*** join/#oe guillaum1 (n=gl@AMontsouris-153-1-4-134.w86-212.abo.wanadoo.fr)
00:42.48*** join/#oe kergoth (n=kergoth@ip68-3-163-239.ph.ph.cox.net)
00:48.30*** join/#oe brolin (n=brolin@190.157.13.75)
01:18.20scruggsis there a setup where I can use oe to build binaries for a stock nokia n810? I see the nokia800 machine file but I'm not sure if that is what I need?
01:18.43scruggsI'm not real big on scratchbox :)
01:19.09*** join/#oe HellDragon (n=jd@Wikipedia/HellDragon)
01:21.38TartarusHmm
01:22.01TartarusPutting the modules.* junk into the kernel-modules package shows that KERNEL_OBJECT_SUFFIX isn't being setup right at all
01:29.41Tartarusmodule_strip.bbclass ends up doing:
01:29.54Tartarusfind ... -name "*"
01:31.04BusErroraww. I moved my oe tree, technicaly the path is the same (bar a symlink) but it can't find my 'tmp'
01:32.23*** join/#oe mnabil_ (n=mnabil@82.201.174.77)
01:33.10mwesterno sym links allowed.
01:50.11*** join/#oe leslie (n=leslie@116.228.58.74)
01:53.42*** join/#oe rsalveti (n=salveti@189.70.50.209)
01:59.28TartarusOK, this is it
01:59.32Tartarusmodule-base isn't inherited anywhere
02:01.49TartarusFixing both of these problems now
02:05.03CIA-203Tom Rini <trini@embeddedalley.com> 07org.openembedded.dev * r25b281668e 10openembedded.git/classes/module-base.bbclass:
02:05.03CIA-2module-base: Fix the test on KERNEL_OBJECT_SUFFIX
02:05.03CIA-2We were saying if we're newer than 2.6.0, use ".o" and otherwise use ".ko",
02:05.03CIA-2which is backwards.
02:06.26*** join/#oe BenLauDC (n=benlau@221.125.8.105)
02:09.54*** join/#oe waite (n=bwaite@206.83.81.178.ptr.us.xo.net)
02:15.49*** join/#oe GunsNRose (n=GunsNRos@222.212.61.76)
02:16.02*** join/#oe AvengerMoJo (n=AvengerM@61.14.130.209)
02:43.30*** join/#oe leslie_ (n=leslie@116.228.58.74)
03:10.07*** join/#oe mhnoyes (n=mhnoyes@adsl-99-147-151-219.dsl.pltn13.sbcglobal.net)
03:12.33*** join/#oe GunsNRose_ (n=GunsNRos@222.212.61.76)
03:23.50CIA-203Tom Rini <trini@embeddedalley.com> 07org.openembedded.dev * rf9167aed6e 10openembedded.git/classes/module_strip.bbclass: (log message trimmed)
03:23.50CIA-2module_strip: When picking module suffix, use KERNEL_MAJOR_VERSION.
03:23.50CIA-2With the previously broken and unused KERNEL_OBJECT_SUFFIX, this actually
03:23.50CIA-2evaluted to "" so we always did a find path -name *, which is why we need
03:23.50CIA-2the -d test later on. Switch to testing for KERNEL_OBJECT_SUFFIX==2.6
03:23.53CIA-2and else'ing for 2.4 and doing .ko or .o. Leave KERNEL_OBJECT_SUFFIX in-tact
03:23.55CIA-2for non-mainline users. This was easier than getting kernel.bbclass to
03:40.39*** join/#oe leslie__ (n=leslie@116.228.58.74)
04:17.50*** join/#oe leslie (n=leslie@116.228.58.74)
04:31.56*** join/#oe pigeon_ (n=pigeon@60-241-137-179.static.tpgi.com.au)
04:41.32leslieHi, anyone konws how to rebuild all packages that a image.bb depends?
04:44.40*** join/#oe leslie_ (n=leslie@116.228.58.74)
05:09.22*** join/#oe archae0pteryx (n=snewman@207.47.42.130.static.nextweb.net)
05:09.22*** join/#oe cedric (n=cedric@enlightenment/developer/cedric) [NETSPLIT VICTIM]
05:09.22*** join/#oe Timelord0 (n=TL@16.8c.d12c.cidr.airmail.net)
05:09.22*** join/#oe tmbinc (i=abcd@83.141.3.59) [NETSPLIT VICTIM]
05:09.22*** join/#oe Ironnads1 (n=Ironnads@host86-135-181-230.range86-135.btcentralplus.com)
05:09.22*** join/#oe Tartarus (i=trini@crackaddict.com) [NETSPLIT VICTIM]
05:09.22*** join/#oe ynezz (i=ynezz@ibawizard.net) [NETSPLIT VICTIM]
05:09.35*** join/#oe archae0ptery1 (n=snewman@207.47.42.130.static.nextweb.net)
05:15.10*** join/#oe broonie (i=broonie@cassiel.sirena.org.uk)
05:23.22*** join/#oe Sleep_Walker (n=Sleep@193.179.96.131) [NETSPLIT VICTIM]
05:23.22*** join/#oe guillaum1 (n=gl@AMontsouris-153-1-4-134.w86-212.abo.wanadoo.fr) [NETSPLIT VICTIM]
05:23.22*** join/#oe tomimo (n=kurre@xdsl-83-150-88-111.nebulazone.fi)
05:23.22*** join/#oe |miska| (n=|miska|@uyulala.karlin.mff.cuni.cz)
05:23.23*** join/#oe piroko (n=jeremy@pohl.ececs.uc.edu) [NETSPLIT VICTIM]
05:23.23*** join/#oe borg_ (n=olaf@80.149.17.21) [NETSPLIT VICTIM]
05:23.23*** join/#oe darkschneider (n=gab@213-140-15-160.fastres.net)
05:43.53*** join/#oe Sup3rkiddo (n=sudharsh@unaffiliated/sudharsh)
05:44.12*** join/#oe colin_ (n=colin@88-109-240-183.dynamic.dsl.as9105.com)
05:44.52*** join/#oe leslie__ (n=leslie@116.228.58.74)
05:47.54*** join/#oe AvengerMoJo (n=AvengerM@61.14.130.209)
05:51.07CIA-203Tom Rini <trini@embeddedalley.com> 07org.openembedded.dev * rd420f4df43 10openembedded.git/ (2 files in 2 dirs): mingw-make-canadian-sdk: Add this so we have a 'make' for mingw
05:57.31leslie__when I do configure, it has an error like this: crt1.o: No such file:   how to solve this?
06:02.11*** join/#oe pvanhoof (n=pvanhoof@host-85-27-48-181.brutele.be)
06:02.31*** join/#oe waite (n=bwaite@206.83.81.178.ptr.us.xo.net)
06:12.33*** join/#oe daurnimator (n=daurn@ppp118-208-160-14.lns10.mel4.internode.on.net)
06:45.29*** join/#oe leslie (n=leslie@116.228.58.74)
06:48.22*** join/#oe garren|work (n=chatzill@mail.dm.co.za)
06:54.15*** join/#oe pleemans (n=toi@d54C2AAB7.access.telenet.be)
07:00.51*** join/#oe Sleep-Walker (n=Sleep@193.179.96.131)
07:03.53*** join/#oe leslie_ (n=leslie@116.228.58.74)
07:10.18*** join/#oe leslie__ (n=leslie@116.228.58.74)
07:13.31*** join/#oe DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk)
07:15.19*** join/#oe gremlin[it] (n=gremlin@host209-103-static.32-85-b.business.telecomitalia.it)
07:19.42gremlin[it]morning all !
07:31.16*** join/#oe vu3rdd (n=vu3rdd@164.164.250.10)
07:37.13*** join/#oe Timelord (n=TL@16.8c.d12c.cidr.airmail.net)
07:40.35*** join/#oe Jay7 (n=jay@93.81.136.171)
07:46.48*** join/#oe Timelord0 (n=TL@16.8c.d12c.cidr.airmail.net)
07:47.09*** join/#oe gilligan_ (n=gilligan@p5B17EB4E.dip.t-dialin.net)
08:04.38*** join/#oe digital-ex (n=alex@99.242.233.201)
08:05.30*** join/#oe Timelord (n=TL@16.8c.d12c.cidr.airmail.net)
08:08.46CIA-203Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * ra340f93f67 10openembedded.git/ (conf/checksums.ini packages/ogre/wgois_1.2.0.bb): wgois: add 1.2.0 release
08:08.49CIA-203Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * rb7c5e97b62 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev
08:10.38*** join/#oe boris_OmegA (n=boris@194.51.141.172)
08:19.59gilligan_anyone happens to know the URL to cbrake's OE example setup ? he hosts some Makefile / sample config files somewhere at bec-systems.com ..
08:26.10*** join/#oe BET_frogger (n=mkl@metis.ext.pengutronix.de)
08:36.56mckoanBET_frogger: moin
08:38.19BET_froggermoin mckoan
08:40.19CIA-203Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * r1eda21e337 10openembedded.git/packages/ogre/ogre-egl_svn.bb: ogre-egl: build samples against a modern toolkit and package them
08:40.19leslie__Hi, while bitbake building a image, in do_rootfs, it reports "Cannot find package xserver-kdrive" and "Cannot find package xf86-video-fbdev", but I do build these two packages, what is this error?
08:43.39*** join/#oe rob_w (n=bill@217.237.177.190)
08:55.54*** join/#oe Xilent (n=xk@unaffiliated/xilent)
09:00.07*** join/#oe cbrake_away (n=cbrake@oh-69-34-21-229.sta.embarqhsd.net)
09:09.44rob_wanybody have an idea with this ? http://pastebin.com/m116b6887
09:10.32*** join/#oe booxter (n=booxter@cpmsq.epam.com)
09:11.17*** join/#oe booxter (n=booxter@cpmsq.epam.com)
09:13.00*** join/#oe Timelord (n=TL@16.8c.d12c.cidr.airmail.net)
09:13.24*** join/#oe erbo (n=erik@c-7b7de455.115-16-64736c14.cust.bredbandsbolaget.se)
09:22.27gremlin[it]hi mckoan !
09:24.59*** join/#oe Timelord0 (n=TL@16.8c.d12c.cidr.airmail.net)
09:33.15*** join/#oe ||Leto|| (n=tomboy64@e178059177.adsl.alicedsl.de)
09:36.12hrwmorning
09:36.12*** join/#oe florian_kc (n=fuchs@217.146.132.69)
09:37.05gremlin[it]morning hrw
09:37.13*** join/#oe cyberdeck (n=mnemonic@iss60.vlsi.informatik.tu-darmstadt.de)
09:41.06*** join/#oe mithro (n=tim@unaffiliated/mithro)
09:49.05*** join/#oe LetoThe2nd (n=LetoThe2@unaffiliated/letothe2nd)
09:57.19CIA-203Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * r4c097d8570 10openembedded.git/packages/clutter/ (aaina_svn.bb clutter_0.8+svn.bb):
09:57.19CIA-2clutter 0.8+svn: move to git
09:57.19CIA-2* this slowly brings us closer to poky
09:57.29*** join/#oe Roman_Khimov (n=rik@office.altell.ru)
09:58.25*** join/#oe slapin (n=slapin@iap-pxy-mow1.siemens.ru)
09:58.33slapinhi, all!
10:00.34slapinwith relocation truncated to fit error with ARM_INSTRUCTION_SET = "thumb", and none with ... = "arm", what is the best thing to do - just put ARM_INSTRUCTION_SET = "arm" into .bb or there's some better solution possible?
10:02.20floriangood monring
10:03.22*** join/#oe Timelord (n=TL@16.8c.d12c.cidr.airmail.net)
10:13.29*** join/#oe sicu (n=n_a@cm-84.208.86.111.getinternet.no)
10:14.24*** join/#oe Gnutoo (n=gnutoo@host221-133-dynamic.31-79-r.retail.telecomitalia.it)
10:16.13slapinflorian, good morning!
10:16.53hrwgot another set of bug linux changes merged into Poky ;)
10:17.04Sleep-Walkermorning, slapin
10:17.27slapinSleep-Walker, and to you!
10:17.54Sleep-Walkerslapin: I've got newly old LD - I'll overbother Marex now ;)
10:18.20slapinSleep-Walker, that's great
10:19.44Jay7slapin: according to util-linux/util-linux_2.12q.bb - yes, just ARM_INSTRUCTION_SET
10:19.57Jay7but FIXME :)
10:20.48slapinSleep-Walker, I try to recover memory of all palm things I did before to dig what can go into .30 and what can't
10:22.35slapinJay7, well, I just want to find out true reason for this and whatever it is fixable with some compiler option, because otherwise it will need to be this way with all big .so files, which is weird.
10:23.02Sleep-Walkerslapin: I'm quite busy with school and work, but when I finish my projects, I'll try to finish Treo680, retest T3, play a bit with LD and create some new releases for that, with up to date kernels and newer rootfs
10:23.30*** join/#oe Sleep_Walker (n=SleepWal@193.179.96.131)
10:24.37Jay7slapin: grep -r thumb */*.bb can provide you some ideas :)
10:25.11Jay7e.g. some interesting options for libmad
10:25.32Jay7but I'm not a gcc/arm/thumb guru :)
10:27.29*** join/#oe Timelord (n=TL@16.8c.d12c.cidr.airmail.net)
10:29.29*** join/#oe zedstar (n=john@fsf/member/zedstar)
10:34.25*** join/#oe otavio (n=otavio@debian/developer/otavio)
10:34.25pb_florian: good morning
10:35.11pb_slapin: what the right solution is depends on what exactly is the code causing the problem.
10:37.06Jay7OSError: [Errno 22] Invalid argument: '/bin/sh'
10:37.12Jay7hm..
10:38.10Jay7i have pulled now and trying to start bitbake
10:39.18Jay7it fails with http://www.pastebin.ca/1333885
10:39.19*** join/#oe timtimwork (i=54e9fcf8@gateway/web/ajax/mibbit.com/x-0a1447a90a2a99d2)
10:40.11Jay7bitbake 1.8.13 (rev. 1150)
10:42.03slapinpb_, it is unable to get to libgcc.a functions with message 'relocation truncated to fit, *unwind* functions. So I think that's size issue.
10:42.45pb_how big is your object?
10:44.16pb_(also, should you not be using libgcc_s.so rather than libgcc.a?)
10:48.07*** join/#oe likewise (n=likewise@82-171-51-231.ip.telfort.nl)
10:49.02likewisegm
10:50.35Jay7do anybody have some clue about this: http://www.pastebin.ca/1333885 ?
10:52.40pb_what does line 89 of the file in question look like?
10:52.47hrwif os.path.basename(os.readlink('/bin/sh')) == 'dash':
10:52.52hrwthats probably this
10:53.31hrwJay7: show "ls -l /bin/sh" please
10:53.38Jay7/bin/sh --version says GNU bash, version 3.2.33(1)-release (i586-alt-linux-gnu)
10:53.46pb_oh right, yeah
10:53.56pb_that would be bogus, os.readlink() will fail if the target isn't a link
10:53.56Jay7hrw: -rwxr-xr-x 1 root root 403236 íÁÒ 27  2008 /bin/sh*
10:54.06hrwthx
10:54.11Jay7that was ls -laF
10:54.15hrwwill have to change code then
10:54.24hrwlrwxrwxrwx 1 root root 4 maj 13  2008 /bin/sh -> bash*
10:54.58Jay7hrw: will you commit changes into bitbake's svn?
10:55.06Jay7or I should create link for now?
10:55.13hrwJay7: its not bitbake but oe
10:55.20Jay7ah, well
10:55.32Jay7I'll wait for pull :)
10:57.06CIA-203Sergey Lapin <slapin@ossfans.org> 07ieee80215 * ra848a9c61d 10openembedded.git/packages/lowpan-utils/lowpan-utils_git.bb:
10:57.06CIA-2lowpan-utils: fixed binaries locations
10:57.07CIA-2Signed-off-by: Sergey Lapin <slapin@ossfans.org>
10:57.09CIA-203Sergey Lapin <slapin@ossfans.org> 07ieee80215 * r7f012b9002 10openembedded.git/packages/wireshark/wireshark_1.0.5.bb:
10:57.09CIA-2wireshark: ARM_INSTRUCTION_SET = "arm" due to relocation truncated to fit
10:57.12CIA-2Signed-off-by: Sergey Lapin <slapin@ossfans.org>
10:57.20slapinpb_, that's automake-generated stuff. Look into wireshark_1.0.5.bb in ieee80215 branch, there's no anything special about this package, except size.
10:57.33hrwJay7: http://www.pastebin.ca/1333915 - apply and check
10:57.55hrwslapin: no need to add S-o-b in commits if they are yours
10:58.12slapinhrw, sorry, just a habit
10:58.59*** join/#oe marex (n=marex@vasut.kolej.mff.cuni.cz)
11:00.32Jay7hrw: seems working
11:00.43hrwthx
11:00.51*** join/#oe mr_nice (n=nice@77-21-130-108-dynip.superkabel.de)
11:05.56Jay7hrw: imho better is execute /bin/sh --version and check for dash
11:06.18Jay7inside try-except block :)
11:06.26hrwpatches are welcome
11:06.47Jay7I don't know python :)
11:06.57Jay7at least yet
11:08.04hrwneither do I
11:10.16*** join/#oe frank7d (n=frank@91.2.3.213.fix.bluewin.ch)
11:11.26Jay7well, away for ~1-2hrs
11:13.44methrilgood morning
11:14.26*** join/#oe alecrim (n=alecrim@189.2.128.130)
11:18.26pb_Jay7: well, heh, better would be to fix oe to actually work with dash :-}
11:19.22*** join/#oe Gnutoo (n=gnutoo@host221-133-dynamic.31-79-r.retail.telecomitalia.it)
11:19.43*** join/#oe mario-goulart (n=user@201-40-162-47.cable.viacabocom.com.br)
11:19.46DJWillispb_: it is often not OE at fault but the cruff some projects pop into there build systems. I think all the bash'isms are long gone in core OE now.
11:21.54*** join/#oe gilligan_ (n=gilligan@p5B17EB4E.dip.t-dialin.net)
11:21.55gilligan_hi
11:22.58hrwDJWillis: nope, gtk+ guys just added new ones
11:24.25gilligan_I just built some custom image (uclibc) for the compulab target ..  now when trying to build a quite similar image (glibc tho) I get an Error message about "information not available for target arm-INVALID" ? I think I'm supposed to set a variable somewhere.. could someone help me out on this ?
11:25.33mattfaceis there a bitbake dev. in the house?
11:26.06hrwwhats a problem mattface?
11:26.47mattfacei cant seem to put ${VARIABLE} in my BBFILES any longer with 1.8.12
11:26.52mattfaceworks fine with 1.8.10
11:27.16hrwwhere ${VARIABLE} is set in environment not in config files?
11:27.22gilligan_mattface, maybe because bitbake is cleaning env variables ?
11:27.25mattfacecorrect
11:27.31mattfacehrw: correct
11:27.52gilligan_so you have to use that BB_ENV_EXTRAWHITE = VARIABLE solution in your .bashrc
11:28.05hrwlike gilligan_ said
11:28.31mattfacerighto, i'll give that a whirl
11:29.26mattfaceyup that works, thx
11:29.52gilligan_hrw, any idea about that arm error message ?
11:30.22hrwno
11:30.49*** join/#oe pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB)
11:31.27gilligan_hm..weird
11:40.46*** join/#oe Heinervdm (n=thomas@pD9E14AD5.dip.t-dialin.net)
11:48.22mattfacei think i'm going to go back to bb 1.8.10, it's much faster
11:49.35Crofton|workmattface, I do not think that would be a good idea ...
11:50.28mattfacei only graded due to the new sanity checking suggested it
11:51.50mattfaceit seems to hang at 87% every time when "Handling BitBake Files" and then again for a few seconds before doing any building
11:54.28DJWillismattface: if you use GIT repros for packages then anything < 1.8.12 is a bad idea ;-)
11:56.42gilligan_aehm.. great.. now I actually can't build anything at all anymore because I always run into this 'arm-INVALID' error .. http://pastebin.com/m260dc293
11:56.51mattfaceDJWillis: it is mainly for our own packages but suggestion noted
11:57.29nik0ni have a runtime problem with the grub package. when i invoke grub-install it prints many segfaults but tells me that it was installed successfully.
11:57.44nik0nbut after a reboot it hangs at stage 1.5 any ideas?
12:01.03gilligan_for some reason the cmx270 is not propagating arm/endianess anymore or so.. (wild guessing.. not really into this..)
12:01.06DJWillismattface: by 'GIT repros' it should be noted I mean building packages that store there sourcecode in GIT. The GIT fetcher got a lot of work in the run up to 1.8.12 and will often die in interesting ways in older versions.
12:01.57Crofton|workgilligan_, the /varli/lib/pythin stuff looks funny
12:02.06mattfaceDJWillis: ahh i see. We're only CVS at the moment (don't laugh)
12:03.04gilligan_Crofton|work, well the weirdest thing is that this configuration worked 10 mins ago .. what I did in between was updating ubuntu packages ... x_X
12:03.16Crofton|workurg
12:03.32gilligan_something is broken ... x_X
12:04.45pb_DJWillis: if it's just the git fetcher then presumably it would be easy enough to backport those changes to 1.8.10.  the fetchers are pretty self-contained for the most part.
12:06.37gilligan_ah..hm.. conf/bitbake.conf:615: including conf/target/arm-INVALID.conf
12:10.04pb_mattface: fwiw, I think you can turn off the crazy environment filtering in 1.8.12 using BB_PRESERVE_ENV.
12:10.43pb_it's possible that 1.8.12 might still suck for other reasons though.  I don't know what the deal would be with the extra slowness.
12:11.26mattfacepb_: i'd rather not just turn it off, it does seem to be a good idea
12:11.39mattfacethat said, these hangs are just too annoying
12:13.33*** part/#oe DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk)
12:13.41*** join/#oe DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk)
12:19.21Croftonpb_, There have been cases where env vars from the environment change/break peoples builds
12:31.30StanchoooHi all. How should I create a patch for one package in OE and then a developer to update it in the repository (the .bb file)
12:32.01gilligan_well for some unexplainable reason TARGET_ARCH set in conf/machines/cm-x270.conf gets lost here and falls back to INVALID
12:33.14gilligan_TARGET_ARCH is also set to arm in local.conf .. how the hell can it be reverted to INVALID when there is "TARGET_ARCH ?= INVALID" in bitbake.conf ??
12:38.56*** join/#oe thesing (n=tkunze@p57BB749F.dip.t-dialin.net)
12:42.17*** join/#oe mnabil (n=mnabil@196.218.15.202)
12:48.26gilligan_looks like I found what was causing the troubles.. I absolutely don't understand it though .. the old local.conf file had DISTRO set to angstrom2007.1 -  there is no conf file for 2007 anymore ..
12:48.57gilligan_when I changed it to 2008.1 the problems where resolved .. *however* : how on earth could it succeed the first time around ?
12:49.33gilligan_i successfully built an image with the exact same config before
12:56.04Jay7Stanchooo: create patch and send it to oe-devel to review and commit
12:58.13*** join/#oe pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB)
12:59.04*** join/#oe birunko (n=birunko@200.184.118.130)
12:59.40*** join/#oe rsalveti (n=salveti@200.184.118.130)
13:00.07*** join/#oe thesing (n=tkunze@p57BB749F.dip.t-dialin.net)
13:02.42*** join/#oe leonardo__ (n=cabarnes@64-198-47-2.ip.mcleodusa.net)
13:03.40beagle-dev-543I've done "bitbake meta-toolchain-gpe" to build a toolchain that other people can use, but when I install it and try to build my application the compilation fails due to the fact it can't find the definition of _IOWR
13:04.32leonardo__i'm trying to get the 'runpy' library into my OE build but I don't know what package it's in.  Is there some way for me to know which of the python library packages it's in?
13:05.09*** join/#oe drw (n=drw@cpe-76-187-64-213.tx.res.rr.com)
13:10.53beagle-dev-543Following the headers the app includes <inux/ioctl.h> -> <bits/ioctls.h> -> <asm/ioctls.h>
13:13.11beagle-dev-543What I find strange is that I can't work out where it gets asm/ioctls.h from as the toolchain OE built doesn't have any system headers in it. The app does add my Linux kernel to it's system path (with -isystem) but there is no ioctls.h in the kernel include directory since that's now moved up to arch/arm/include/asm
13:13.46beagle-dev-543But shouldn't the toolchain be build with the kernel headers installed in it anyway?
13:15.02BET_froggerbeagle-dev-543: nope, that's old-fashioned
13:15.39beagle-dev-543So where do the kernel headers come from?
13:15.42BET_froggertoday you do "make headers_install"
13:15.46BET_froggerin the kernel
13:16.10BET_froggeryou'll get the sanitized headers
13:16.15*** join/#oe mnabil_ (n=mnabil@62.135.96.220)
13:16.40beagle-dev-543I assume that "bitbake meta-toolchain-gpe" should have done that
13:17.23beagle-dev-543so you get a tarball which I untar into / and should be ready to go, but that isn't the case as I'm missing headers in the toolchain it builds
13:17.31BET_froggerbeagle-dev-543: why does you application include "linux/ioctl.h" in the first place
13:17.49BET_froggermy manpage says: "#include <stropts.h>"
13:17.52beagle-dev-543as it wants to speak to a device driver
13:18.17BET_frogger<-- posix manpage
13:18.25BET_frogger"#include <sys/ioctl.h>" <--- glibc manpage
13:18.39BET_froggerdon't include _linux_/ioctl.h
13:19.37*** join/#oe pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB)
13:19.56beagle-dev-543The app includes <sys/linux.h> which then includes <linux/ioctl.h> further down the include chain
13:20.28beagle-dev-543sorry your right!
13:21.29*** join/#oe [X]Spot (n=stancho@78.90.99.168)
13:31.42beagle-dev-543<PROTECTED>
13:31.55CIA-203stanislav <s.popov@rsc.bg> 07org.openembedded.dev * r1abbf0b0f7 10openembedded.git/packages/vlc/ (vlc.inc vlc_0.9.8a.bb): vlc: --enable-realrtsp option added
13:33.15beagle-dev-543Looks like the compiler is build with headers installed into /usr/local/angstrom/arm/arm-angstrom-linux-gnueabi/usr/local but the compiler default search path looks in/usr/local/angstrom/arm/arm-angstrom-linux-gnueabi/usr/
13:33.41beagle-dev-543Strangely copying the files across doesn't seem to work :/
13:43.34*** join/#oe XorA (n=XorA@94-193-93-235.zone7.bethere.co.uk)
13:43.57gremlin[it]hi XorA !
13:44.03XorAhey gremlin[it]
13:45.23likewiseyo
13:45.40likewisegremlin[it]: I found the kernel internal api - let me find it again :-)
13:46.01beagle-dev-543OK, being silly. The header gets included from both userland apps and kernel drivers so need to select between <sys/ioctl.h> and <linux/ioctl.h> depending on what your building
13:46.34likewisegremlin[it]: you need two things: the high res api, and setting your kernel thread to real-time priority. this does the first only: http://lwn.net/Articles/167897/
13:48.29*** join/#oe aloisiojr (n=aloisio@200.184.118.130)
13:51.21gremlin[it]likewise, wow thanks !!!
13:51.31thesingXorA: hi
13:51.45likewisegremlin[it]: did your driver use a thread?
13:51.49thesingXorA, if you have time, could you build current collie images and test them?
13:52.02XorAthesing: just waiting for fsck to finish
13:52.03gremlin[it]at now no cause if for 2.4 kernel
13:52.06XorAthesing: then I can
13:52.25thesingXorA, disable hwclock in /etc/init.d/bootmisc if it doesn't boot.
13:53.36beagle-dev-543The userland app now uses <sys/ioctl.h> but I still have the same problem as <asm/ioctls.h> doesn't exist in my sdk toolchain
13:56.41gremlin[it]likewise, no at now, is a driver for a 2.4 kernel, but i'm planing to rewrite it all for 2.6 ...
13:57.41BET_froggerbeagle-dev-543: you toolchain is probably broken: in my toolchain, there is asm/ioctl.h http://www.pastebin.ca/1334039
13:57.55BET_froggerbeagle-dev-543: its ioctl.h w/o "s"
13:58.18*** join/#oe Sup3rkiddo (n=sudharsh@unaffiliated/sudharsh)
13:59.49beagle-dev-543The problem is that I used OE to build that toolchain.
14:00.39beagle-dev-543The one that it builds and uses internally is fine, but then one it builds with  "bitbake meta-toolchain-gpe" is broken
14:01.12beagle-dev-543So I can't distribute my toolchain to the other devs :(
14:01.26CIA-203Bernhard Guillon <Bernhard.Guillon@opensimpad.org> 07org.openembedded.dev * r5082491d1d 10openembedded.git/packages/aufs/ (aufs/aufs_create_no_manpage.patch aufs_cvs.bb): aufs: fix cross compiling, delete old patch and bump PR
14:01.51BET_froggerbeagle-dev-543: seems that you have to blame the meta-toolchain-gpe author :)
14:02.27slapinIs there some way to have both angstrom feeds and custom feeds in the same generated image?
14:02.35hrwsure
14:02.48hrwput your custom ones in /etc/opkg/ dir
14:03.37likewisegremlin[it]: ok, even without a thread, you can assign a priority to workqueue's also.
14:04.16gremlin[it]likewise, ah ok ... i'll check what would be better :)
14:05.42slapinhrw, I want to have the same structure as Angstrom but with custom URL. I want for opkg to take my packages when they are built for particular architecture, otherwise get Angstrom ones. I could do the same as with angstrom-feed-configs.bb, but I'd like to reduce copying...
14:07.41hrwslapin: I always did it by hand..
14:09.00XorAhey mr BET_frogger long time no see
14:12.11BET_froggerXorA: heyhey
14:12.24XorABET_frogger: hows things going?
14:13.06BET_froggergood, besides hacking prehistoric MV 2.6.21 kernels
14:13.08BET_frogger;P
14:13.38BET_froggerI met liam at FOSDEM, have you been there, too?
14:14.01beagle-dev-543Looking at the meta-toolchain issue, I think it's a generic problem as all meta-toolchains depend on meta-toolchain
14:21.16XorABET_frogger: I was there, liam shoulda pointed you out
14:21.41XorAdamn, time for driving lesson, better go get some shoes
14:22.06XorA|car-lessonlooks less like a name now :-)
14:23.38*** join/#oe brolin (n=brolin@iner.udea.edu.co)
14:24.44beagle-dev-543Hmm, I guess  bitbake meta-toolchain-gpe and bitbake meta-toolchain do produce different header files layout for the compiler
14:25.02beagle-dev-543the meta-toolchain gets past the iotcl problem
14:25.44Crofton|workcar lesson?
14:26.10BET_froggerXorA|car-lesson: unfortunately liam had to leave after his talk quite fast
14:26.12beagle-dev-543I now get "/usr/local/angstrom/arm/arm-angstrom-linux-gnueabi/usr/include/limits.h:125:26: error: no include path in which to search for limits.h"
14:26.17gremlin[it]Carl Esson ... ghghghgh ;)
14:26.35Crofton|workis scared for the people of Scotland
14:26.57beagle-dev-543The problem line in limits.h being "# include_next <limits.h>"
14:28.23likewiseCrofton|work: they gotta learn to drive on the right side of the road, which they think is left
14:28.41beagle-dev-543It is!
14:28.49likewiseCrofton|work: and the bike repair shop just called they lost the keys to my motor cycle.
14:29.00likewisebeagle-dev-543: I agree.
14:29.15Crofton|workhmm, sounds like you need a ne repiar shop
14:29.39likewiseCrofton|work: well, got sick of the previous one, I thought this was the new one :-)
14:30.06*** join/#oe Sup3rkiddo (n=sudharsh@unaffiliated/sudharsh)
14:30.34Crofton|workgrass is always greener on the other side of the fence
14:32.16BET_froggerlikewise: reminds me of: "If people in Britain want to buy a pet, they go to a pet shop. If they want to buy a pet shop, they go to a pet shop shop. If they want to buy a pet shop shop, then they're just being silly."
14:34.53likewiseBET_frogger: we should have met at fosdem too.
14:35.04likewiseBET_frogger: but we didn't. :-/
14:35.08*** join/#oe pleemans (n=toi@d54C2AAB7.access.telenet.be)
14:35.11beagle-dev-543The problem is that limits.h tries to include the next limits.h which should be from the compiler, but the compilers limits.h isn't in my toolchain
14:35.57beagle-dev-543From libc's limits.h http://www.mibbit.com/pb/rM9zmK
14:36.15BET_froggerlikewise: i went along the OE booth 2x times, hurrying to the next talk or to the fritten bude
14:37.47likewiseon sunday or saturday?
14:38.32florianBET_frogger: You are one of the people who think OE is selling geek keyboards? ;)
14:38.32BET_froggerflorian: aren't they?
14:38.32BET_frogger:P
14:38.32florian:-)
14:39.29BET_froggerlikewise: sunday
14:41.56likewiseBET_frogger: ok, was there only on saturday. Better luck next time.
14:42.26likewisewe sell them for EUR 300
14:44.38beagle-dev-543Found the problem. /usr/local/angstrom/arm/lib/gcc/arm-angstrom-linux-gnueabi/4.3.1/include-fixed exists in my tmp/work/i686-armv7a-sdk-angstrom-linux-gnueabi/gcc-cross-sdk-4.3.1-r12/image directory but not in the sdk tar ball or in the install directory in tmp/work/i686-armv7a-sdk-angstrom-linux-gnueabi/gcc-cross-sdk-4.3.1-r12/install
14:44.53*** join/#oe BenLauDC (n=benlau@221.125.8.105)
14:45.44florian:-)
14:45.51beagle-dev-543I assume that for each package the "image" directory is where the "make install" gets done and the "install" directory is where it ipkg install happens?
14:47.21beagle-dev-543If so then the package is being stripped of needed headerfiles during the creation of the ipkg
14:48.43*** join/#oe waite (n=bwaite@206.83.81.178.ptr.us.xo.net)
14:56.32beagle-dev-543Doing a manual copy of include-fixed solves the problem.
14:58.16*** join/#oe kergoth (n=kergoth@65.200.49.156)
14:59.57*** join/#oe jimsheldon (n=jim@c-98-216-198-0.hsd1.nh.comcast.net)
14:59.58kergothmorning
15:00.05beagle-dev-543I'd like to fix the issue properly, but I've not modified any bitbake files. I'll have a look around to see if it's anything I might be able to fix, if not is there a bug tracker for OE?
15:00.14pb_hi kergoth
15:00.17jimsheldoncbrake: ping?
15:02.58*** join/#oe chouimat|work (n=dieu@209.217.106.98)
15:07.10timtimworkthe left side is the right side? :)
15:08.27beagle-dev-543yes, so if a person on his horse comes at you you have your sword had to fend him off
15:09.10timtimworksword? :)
15:09.59beagle-dev-543Yes
15:10.00timtimworki like to confuse things, and drive a LHD car on the left, which is the right side.
15:10.19timtimwork:)
15:10.40beagle-dev-543:)
15:11.31timtimworkhas a 30 year old vette :)
15:11.40timtimworkoh, by beagle case arrived. excellent.
15:14.03likewisekergoth: morning
15:14.37likewisedidn't want to insult anyone
15:15.06likewisebeagle-dev-543: so it was assumed everyone was right handed
15:16.27*** join/#oe Sleep_Walker (n=SleepWal@1.57.broadband2.iol.cz)
15:16.38gremlin[it]wait for C2 beagle board
15:17.13BusErroranyone has problems building rxvt-unicode ? it fails with a "permission denied" on /etc/terminfo/r...
15:17.18beagle-dev-543You got a C1 beagleboard?
15:17.29BusErrorI think it tries to touch the host's
15:17.37gremlin[it]beagle-dev-543, talk with me ?
15:20.44cbrakejimsheldon: hello Jim
15:22.29*** join/#oe thomasg (n=thomasg@85.131.189.115)
15:23.08jimsheldoncbrake: do you have any experience with using the framebuffer on the cm-x270?
15:23.56jimsheldoncbrake: my company is considering the cm-x270 for a touchscreen device using Qt
15:24.38*** join/#oe ScriptRipper (n=martin@85.233.34.53.dynamic.cablesurf.de)
15:25.53booxterwhat should I have for FILES_${PN} to pack all installed files into the real package. I have NOTEs on building my package which tells me that my files are not installed in any package
15:26.00BusErrorI had to disable /usr/bin/tic to get rxvt-unicode to build, for anyone interested
15:26.40*** join/#oe Ojg (n=org@ojg-laptop.tekproj.bth.se)
15:30.26hrwre
15:30.32hrwBusError: ? why?
15:30.46CIA-203Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * rb0f7bd2117 10openembedded.git/packages/powervr-drivers/libgles-omap3_3.00.00.06.bb: libgles-omap3: make 06 default now
15:30.47CIA-203Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * r92550d769a 10openembedded.git/packages/clutter/clutter_0.8+git.bb: clutter: add 0.8 file
15:31.24*** join/#oe AvengerMoJo (n=AvengerM@124.126.138.24)
15:37.13BusErrorhrw well, if tic is presetn on the host, rxvt tries to use it... on the host. and fails
15:37.45gremlin[it]not sure but if i remeber correct i report a bug about it long time ago
15:37.53BusErrorshould never have updated git, now rootfs fails in eglibc, "rm: command not found"
15:38.25BusErrorI did a mv /usr/bin/tic /usr/bin/tic.dis -- and rxvt built
15:38.27[X]SpotIs this an error?
15:38.28[X]Spothttp://stancho.com/pastebin/1230
15:38.31[X]Spot?
15:38.45[X]Spotis it critical for something ?
15:39.36hrw[X]Spot: safe
15:39.48hrw~curse users with [!@#!%$#%^*(&@$^(@! etc in nicks
15:39.48ibotMay you be reincarnated as a Windows XP administrator, users with [!@#!%$#%^*(&@$^(@! etc in nicks !
15:40.02hrwBusError: show exact bug please
15:41.01cbrakejimsheldon: yes, I'm working on a project now that uses the cm-x270 with a VGA display
15:41.10cbrakejimsheldon: what size display are you looking at using?
15:43.09OjgLo all, don't know if this is the right channel for this, but i'm trying to build a very small embedded linux system with the openembedded tools. My distro only needs wifi,SDL and upnp support. Ive read the guide and tried a few times without success, so what should i choose for this build?, can i use and existing distro config file or do i need to make a new one and what should i tell bitbake to build?
15:43.42booxterOjg: build angstrom for start
15:43.58*** join/#oe pvanhoof (n=pvanhoof@host-85-27-48-181.brutele.be)
15:44.18booxterfor minimal-image or console-image bb package
15:44.32booxterand then populate your image with the needed packages
15:45.10CIA-203Bernhard Guillon <Bernhard.Guillon@opensimpad.org> 07org.openembedded.dev * r893ea81815 10openembedded.git/contrib/angstrom/build-feeds.sh: build-feeds.sh: add task-xqtlauncher
15:45.51Ojgtried to build with angstrom as distro before today and told bitbake to build base-image and i got an error when it was building the kernel
15:47.14*** join/#oe kergoth (n=kergoth@65.200.49.156)
15:47.43Ojgdon't have a pastebin on it but there was some kind of problem with the sumversion.c
15:47.58*** join/#oe zecke (n=ich@92.117.34.12)
15:48.25jimsheldoncbrake: actually the details of the screen aren't decided yet, I'm going to send you an email so we can talk that way
15:48.30Ojgbut anyway when its done building minimal-image ill just tell bitbake to build the other packages i need what should i tell bitbake todo after that to build my image i can use to boot with ?
15:49.28jimsheldoncbrake: email sent :)
15:49.37BusErroreglibc install failing, in do_rootfs, from the current git .dev head http://pastebin.com/m4bd5fd04
15:51.23*** join/#oe CosmicPenguin (n=nobody@66-162-99-237.static.twtelecom.net)
15:52.13*** join/#oe steliosk (n=Stelios@ipa107.2.tellas.gr)
15:52.18BusErrorthe amount of garbage installed by just the eglibc is rather incredible
16:08.47*** join/#oe scruggs (n=chris@72-161-105-25.dyn.centurytel.net)
16:09.00*** join/#oe robtow (n=rtow@64.62.142.114)
16:15.23*** join/#oe mrc3 (n=mrc3@nat/ti/x-8f2959626e1a17d8)
16:22.24XorABET_frogger: if you went to OE stand you must have met me :-)
16:22.31XorABET_frogger: I was there all day sat and sunday :-)
16:23.19BET_froggerXorA: I just got along the booth
16:23.33BET_froggerand I don't know how you look like
16:25.01hrwBET_frogger: black fingernails, black tshirt, black shoes, black..., only skin white
16:25.19BET_froggerhrw: like 75% of the other geeks?
16:26.02hrwI had grey sweter 'D
16:26.23XorABET_frogger: I was one of the very few with black fingernails
16:26.29gremlin[it]except nails and hairs me and  XorA have the same colours :}
16:26.30XorABET_frogger: the laser keyboard doesnt work for me :-)
16:27.21BET_froggerbtw: the plant next to my desk looks quite sad, my co-workes say it's due to _my_ lack of sunlight
16:27.34XorAhehe
16:29.00BET_froggernext time i'll look at all OE guy's fingernails
16:30.05gremlin[it]next year XorA maybe change color :)
16:31.28XorABET_frogger: http://viewmorepics.myspace.com/index.cfm?fuseaction=viewImage&friendID=94611883&albumID=1348929&imageID=38228412
16:32.06BET_froggerXorA: there ain't to nailpolish on this picture
16:32.15timtimworkshines sunlight at the goths
16:32.17timtimwork:)
16:32.50BET_froggerhides
16:32.59XorABET_frogger: rare day I couldnt be bothered :-)
16:33.13Crofton|workhttp://www.flickr.com/photos/32615155@N00/3260848665/in/set-72157613444405963/
16:33.46BET_froggerCrofton|work: can't see the nails on this pic
16:34.09eFfeMCrofton|work guess this was early morning :-)
16:34.24timtimworkthe goth thing is ok, but painting nails is for girls :)
16:34.29timtimworkhides
16:35.02BET_froggereFfeM: exif says:  2009:02:07 05:37:52
16:35.39eFfeMguessed that
16:35.49eFfeMnext time i'll try to be there for the beer event too
16:38.03gremlin[it]beagle-dev-543,
16:38.57eFfeMthese are all flickr pics tagged fosdem2009
16:39.06eFfeMhttp://www.flickr.com/photos/tags/fosdem2009/
16:39.27eFfeMis wondering why pic #9 is not #1 when it is about most interesting
16:42.10florianmickeyl: good morning!
16:43.27mickeylhey florian
16:45.05gremlin[it]hi mickeyl, i suppose u aren't in europe if u wakeup now :)
16:45.23timtimworkhe could just be really lazy :)
16:45.27timtimworkhey micky :)
16:45.44gremlin[it]i hope for him ;) ...
16:45.46*** join/#oe darion76 (i=darion@94-248-12-187.dynamic.peoplenet.ua)
16:46.08XorAkicks timtimwork in the stones
16:46.49timtimworkhmm which comment was that for? :)
16:47.04XorAgirls one
16:47.18*** join/#oe thomasg (n=thomasg@85.131.189.115)
16:47.50timtimworkheh. its a good thing it was a kick - dont want nail polish on me :)
16:50.09mwesteris glad he didn't speak aloud that he agrees with timtimwork on that ;)
16:50.09mickeylgremlin[it]: heh
16:50.15mickeyli need to work less
16:50.21mickeylat least on Om
16:50.28*** join/#oe Ojg (n=org@ojg-laptop.tekproj.bth.se)
16:50.32mickeylso trying to do some more fun things
16:50.36hrwmickeyl: still doing openmoko things?
16:50.38mickeylpicking up my music
16:50.44mickeyland xelatex typesetting :)
16:50.57mickeylhrw: yes, and no. freesmartphone.org, but that's 100% funded by Om, so... yes.
16:51.21OjgBug while building kernel 2.6.21, bitbake minimal-image distro = angstrom      http://pastebin.com/f711632f0
16:52.30mickeyli need some serious time to pick up the 'minimal' distro again
16:53.22gremlin[it]mickey : yesssss i was talking about at FOSDEM too ... it's something more than one feel is missing
16:53.43Ojgis there a stable git ? because i assume the git.openembedded.net/openembedded is a dev branch ?
16:53.56mickeylgremlin[it]: right
16:54.01mickeylangstrom is great, _BUT_
16:54.06mickeylit completely hides problems in our metadata
16:54.10mickeyllike... sane default versions and settings
16:54.17mickeylwe need to act sooner than later
16:54.31mickeylelse we won't be able to build anything else than angstrom
16:54.52gremlin[it]it's a nice distro ... but not a good starting point for a custom project (distro) with minimal dependencies
16:55.05mickeylok, plan:
16:55.11mickeyli try to come up with a proposal and a call for hands
16:55.18mickeylwill send it to the oe mailing list next week
16:55.20Tartarustimtimwork, you're the samba guy, yeah?
16:55.24mickeyland then lets see whether we get some people for helping us
16:55.38timtimworki am that samba guy indeed
16:55.39hrwmickeyl: I have angstrom based distro
16:55.51mwesteragrees -- it is a constant struggle to keep SlugOS building because SlugOS != angstrom.
16:55.58mickeyltoo much magic
16:55.59mickeylreally
16:56.02mickeylless magic
16:56.04Tartarustimtimwork, k :) smbfs should just pull in stuff for mounting samba filesystems, ya?
16:56.04mickeylbetter defaults
16:56.05mickeylit's all that
16:56.20timtimworkyes, although you should probably use cifs these days
16:56.29gremlin[it]i go home, see u later
16:56.31mwesterCan't build krb (for example) because of some angstrom-specific sensitivies :(
16:56.40hrwstandard problem is that those devs which have knowledge lack time
16:56.45Tartarustimtimwork, ok, is there a cifsfs package too?
16:56.48Tartarusshould check, heh
16:56.50*** join/#oe scruggs_ (n=chris@72-161-105-25.dyn.centurytel.net)
16:56.52hrwTartarus: cifs is kernel
16:56.59hrwTartarus: cifs is in kernel (module)
16:57.02timtimworkTartarus: i think so. im at work though atm and away from my OE workstation
16:57.06TartarusFILES_cifs = "${base_sbindir}/mount.cifs ${base_sbindir}/umount.cifs"
16:57.15timtimworkin fact im 90%% certain
16:57.29Tartarusso yes, there's a package
16:57.38timtimworkyes kernel module and mount utilities
16:57.45timtimworkideally
16:57.52timtimworksmbfs is deprecated for cifs
16:57.55*** join/#oe asarcoin1 (n=asarcoin@139.219.89-79.rev.gaoland.net)
16:57.59BusErroranyone who knows why eglibc is suddenly borken at install time ? I posted a log some time ago. I;m stuck on that, I have no idea where to look or how to force it to contimue making the rootfs
16:58.11BusError<- eglibc install failing, in do_rootfs, from the current git .dev head http://pastebin.com/m4bd5fd04
16:58.12Tartarustimtimwork, i'll have to keep that in mind, thanks
16:58.22timtimworkat some point this/next week i'll remove DEFAULT_PREFERENCE -1 on 3.3.0
16:58.31Ojgwhat distro should i use in my oe config if i want to build a minimal system someone said angstorm before but as i wrote earlier it gives problems with sumversion.c while compiling the kernel?
16:58.42timtimworkits set to that as I havnt tested it enough and it upgrades any tdbs in a non downgradeable manner
16:59.06timtimworkwill add something to 3.3.0 to back them up somewhere just in case before i do that
16:59.48timtimworkright back to work...
16:59.49timtimwork:)
17:00.08TartarusOjg, that's a kernel problem, if it's the failure i'm thinking off
17:00.23Tartarusneed to patch the kernel (not really kernel, but part of the kernel's config stuff)
17:00.30*** join/#oe xjqian (n=gordon@mscitspubwlgw.wustl.edu)
17:01.05Ojgk thought bitbake would handle all the patching for me
17:01.11Ojghttp://pastebin.com/f711632f0 this is my problem
17:01.40Ojgtrying to build a minimal-image with angstrom as distro
17:02.00Tartarusyes, bitbake handles all of the patching that exists
17:02.04*** join/#oe pcgeil (n=steffen@p549E45F8.dip.t-dialin.net)
17:02.04Tartarusyou need to create a patch :)
17:02.18Ojgsigh k
17:02.25Tartarusit's fixed upstream alrady
17:02.26Tartarus*already
17:02.32Tartarus2.6.21 is just old
17:02.42Ojgyeah
17:03.09Ojgshould i specify another version in the local.conf or where do i change this ?
17:03.31Tartaruswhat's your MACHINE?
17:03.50Tartarusbitbake should be picking the newest version that supports that machine
17:04.24Ojgx86
17:04.44Ojggoing to rebuild the machine as a arm later when i have wrote and tested my code
17:04.52hrwwe need to push x86 to 2.6.28
17:05.26Ojgor well arm is a arch so nvm
17:05.32Ojghrw please do =D
17:06.05hrwbut this will add more work for me as I have 2.6.28 changes to be merged...
17:06.06hrw;D
17:06.18Ojg=(
17:07.04Ojgany other suggestions ? i need a minimal embeddedsystem with wifi,SDL and upnp for my project
17:07.48hrwdefine minimal
17:08.02hrwI would build angstrom and use small image as base
17:08.10TartarusYou can hack a fix in for that real quick too
17:08.16TartarusJust add the missing include directly
17:08.20Tartarusand get the build going again
17:10.53Ojgminimal as in the absolute necessary to run the support i need, going to build my program that is going to run in the machine with c++ so some libs for that as well no compiler on the system would prefer cross-compiling my code to it as well
17:11.47hrwOjg: minimal  == 8MB of flash? 256MB of flash?
17:12.12Ojg10-15MB max
17:12.48mwesterOjg:  what processor?
17:13.02TartarusYou'll probably end up defining a custom busybox config file
17:13.30Ojghardware aint 100% decided but it seems like a arm right now but was thinking of building a test machine for i386 first
17:14.48OjgTartarus read about busybox when i tried to build my system with LFS seems promising =)
17:15.38Ojgim not new to linux been running it for many years but haven't built a system from scratch like this before =) so a lot of new stuff as well
17:16.49hrwOjg: MACHINE=qemux86 DISTRO=angstrom-2008.1 and then 'bitbake console-base-image'
17:17.21hrwOjg: you can run this in qemu or on PC. check how it works, add extra stuff from angstrom repositories
17:18.44Ojgwas it qemu that could emulate other cpu arches ?
17:18.49Ojgmaybe i should build it for arm directly
17:20.16Ojgdo i need to remove by tmp folder aswell ? and start from scratch ?
17:20.23hrwOjg: MACHINE=qemuarm then
17:20.54hrwqemu is able to emulate x86, arm (v4/5/6/7a), ppc
17:21.42Ojghrw kk thanks for the help =) do i need to remove my tmp folder aswell or ill bitbake overwrite it ?
17:21.54Ojgwill *
17:22.11hrwOjg: if you did angstrom builds it is safe to keep it. otherwise remove
17:22.22Ojgkk
17:24.01Ojgthanks for the help everyone =) would probably take a while to find this information with the holy grail of the internetz google =P
17:24.42hrwOjg: feel free to edit our wiki
17:25.37Ojgk think i need to get a working system before i write anything
17:27.33Ojghmm, when i build for the qemux86 will to choose a newer kernel version to build or do i still need to fix this somehow ?
17:27.55Ojgwill it *
17:28.47hrwit will use 2.6.26 iirc
17:29.01Ojgkk
17:30.17Ojgwell need to catch the train so cya all later
17:35.33*** join/#oe XorA (n=XorA@94-193-93-235.zone7.bethere.co.uk)
17:41.57hrwwe need better install-native version
17:45.24hrwbye
17:46.04CIA-203Marcin Juszkiewicz <hrw@openembedded.org> 07org.openembedded.dev * r8a46418c79 10openembedded.git/conf/checksums.ini: checksums.ini: added some entries
17:46.04CIA-203Marcin Juszkiewicz <hrw@openembedded.org> 07org.openembedded.dev * r625acc7538 10openembedded.git/classes/sanity.bbclass: sanity.bbclass: do not fail if /bin/sh is not symlink
17:50.40*** join/#oe tomboy64 (n=tomboy64@e178059177.adsl.alicedsl.de)
17:52.19beagle-dev-543I'm trying to fix the missing headers in sdk target issue
17:52.35beagle-dev-543I've added include-fixed to gcc-package-sdk.inc
17:53.26likewiseI'm reading on programming the GPU: The presentation says: "Modify build system accordingly"
17:53.47*** join/#oe valhalla (n=valhalla@81-174-37-40.dynamic.ngi.it)
17:53.54beagle-dev-543and I can see in the install directory in gcc-cross-sdk-4.3.1-r12 that the include-fixed is there, but it's not in the sdk tarball
17:54.18beagle-dev-543Something must be stripping out that directory, any ideas where to look?
17:54.29TartarusOr not including it to start with
17:54.45Tartaruspackages/gcc/gcc-package-sdk.bbclass
17:55.20beagle-dev-543How do you mean? I've added it to gcc-package-sdk.inc FILES list
17:57.08TartarusThen it's a bit more complex, but i'd have thought it's done already..
17:59.17beagle-dev-543I did the last pull a few weeks ago
17:59.20CIA-203Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * ra8fa00b175 10openembedded.git/conf/distro/include/angstrom-uclibc.inc: angstrom 2009.X: fix avr32 build
17:59.45beagle-dev-543how do I find out my version from git?
17:59.52CIA-203Tom Rini <trini@embeddedalley.com> 07org.openembedded.dev * r2dc8f77b23 10openembedded.git/conf/bitbake.conf: bitbake.conf: Add UBI_VOLNAME to allow distros to set the volume name for the ubi volume
18:00.02CIA-203Tom Rini <trini@embeddedalley.com> 07org.openembedded.dev * r5fa90a761b 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev
18:02.17beagle-dev-543Tartarus: I can't see any packages/gcc/gcc-package-sdk.bbclass
18:02.30Tartarusmight be .inc
18:02.38TartarusBut it sounds like you added to the file i was thinking of
18:02.48beagle-dev-543yes, there's the .inc, which I've modified
18:04.03beagle-dev-543like I said, the install directory in my gcc-cross-sdk-4.3.1-r12 now has the include-fixed directory so I assume that there is not problem with that part
18:04.28beagle-dev-543but still the include-fixed does not appear in the sdk toolchain
18:04.55TartarusWell, what package does it end up in?
18:05.03TartarusAnd is that installed by your sdk target?
18:06.21beagle-dev-543It must be as otherwise I would have no compiler
18:07.13TartarusWell, n
18:07.14Tartaruso
18:07.23Tartaruswhat output package did you add it to?
18:07.28TartarusAnd have you confirmed it's in the ipk itself
18:07.33Tartarusand that the ipk in question is installed
18:08.08beagle-dev-543Does meta-toolchain build ipks?
18:08.25TartarusAll of the meta-toolchains-foo's do
18:08.28TartarusWhat one are you using?
18:08.55beagle-dev-543meta-toolchain itself
18:09.16beagle-dev-543I tried the meta-toolchain-gpe version as well, but that has more problems
18:09.33TartarusOK
18:09.42TartarusWell, check the ipk files
18:09.49Tartarusand the build logs to see what packages were installed
18:10.04Tartarusand check the ipk you think you added it to to see if it's in there
18:11.09*** join/#oe wicknix (n=wicked@32.136.202.22)
18:11.44beagle-dev-543can I just untar the .ipk?
18:11.49Tartarusar -x
18:11.54Tartarusthen data.tar.gz
18:12.04denixbeagle-dev-543: would this help: http://blog.denix.org/2008/09/gcc-or-glibc.html
18:12.21*** join/#oe guillaum1 (n=gl@AMontsouris-153-1-4-134.w86-212.abo.wanadoo.fr)
18:14.52beagle-dev-543denix: Thanks, the same issue I've hit
18:15.20beagle-dev-543Tartarus: include-fixed is in the ipk
18:15.47TartarusNow is that ipk installed in the sdk?
18:15.54TartarusCheck tmp/work/.../temp/log.do_populate_sdk
18:17.44gilligan_net-snmp-5.4.1 is missing in checksums.ini
18:17.50beagle-dev-543Hmm, no log.do_populate_sdk
18:18.04gilligan_huh.. no..actually it is not
18:18.53beagle-dev-543wait, justing trying something
18:21.54*** join/#oe e-ffi (n=cybercom@dslb-088-068-185-204.pools.arcor-ip.net)
18:22.18beagle-dev-543got it, forgot to do bitbake -crebuild meta-toolchain
18:24.10beagle-dev-543The link that denix posted shows that this bug has been fixed before so I'm not sure why I've hit it again
18:24.40denixbeagle-dev-543: yeah, I thought it was fixed back then. let me check the logs...
18:26.02beagle-dev-543Head of git shows the same as I've got, no include-fixed (http://cgit.openembedded.net/cgit.cgi?url=openembedded/tree/packages/gcc/gcc-package-sdk.inc)
18:26.05*** join/#oe ClashTheBunny (i=c72ec6e7@gateway/web/ajax/mibbit.com/x-8714d700b1c3eaf8)
18:33.04*** join/#oe pvanhoof (n=pvanhoof@host-85-27-48-181.brutele.be)
18:34.10beagle-dev-543Looks like the patch never made it
18:35.11beagle-dev-543${prefix}/${TARGET_SYS}/usr/include/* was added around that time
18:35.49otavioAny idea why this fails?
18:35.50otavio| running configure fragment for nptl/sysdeps/pthread
18:35.51otavio| configure: error: compiler support for __thread is required
18:35.51otavioNOTE: Task failed: /home/otavio/hacking/ossystems/oe/tmp/osthindesktop-lite/work/i486-osthindesktoplite-linux/glibc-initial-2.6.1-r15/temp/log.do_configure.2273
18:37.16denixbeagle-dev-543: looks like it... the glibc was patched to use include-fixed though
18:39.13denixbeagle-dev-543: here is the diff in question - http://cgit.openembedded.net/cgit.cgi?url=openembedded/commit/&id=2da8451745e6168f888392698258ff8fa1e94f76
18:41.32beagle-dev-543denix: Ah, the the compiler knows to look their but meta-toolchain doesn't copy those files across
18:42.02beagle-dev-543denix: Do you want a patch file :)
18:42.11denixbeagle-dev-543: I guess only gcc-cross was tested
18:42.41denixbeagle-dev-543: and include-fixed is not packaged into SDK tarball, like you said
18:42.44denixbeagle-dev-543: sure
18:44.59*** join/#oe tharvey (n=tharvey@adsl-76-205-222-173.dsl.snlo01.sbcglobal.net)
18:45.09*** join/#oe Sleep_Walker (n=SleepWal@247.153.broadband6.iol.cz)
18:46.09beagle-dev-543http://www.mibbit.com/pb/TLNQaO
18:46.18beagle-dev-543That OK
18:46.20beagle-dev-543?
18:46.21*** join/#oe booxter (n=booxter@night02.sml.by)
18:48.46denixbeagle-dev-543: looks good, let me try it...
18:50.13denixbeagle-dev-543: the only problem is include-fixed is new for gcc 4.x, but gcc-package-sdk is generic to all gcc versions...
18:51.08*** join/#oe vu3rdd (n=user@122.167.87.36)
18:52.57denixI meant new to 4.3.x...
18:53.15CIA-203Tom Rini <trini@embeddedalley.com> 07org.openembedded.dev * ref14e3df7a 10openembedded.git/ (8 files in 3 dirs): ltp: Add 20090131 version
18:53.25TartarusAnd with that, man, fuck ltp :|
18:55.01Tartarushmm
18:55.11TartarusNOTE: smbfs contains dangling symlink to /usr/bin/smbmount
18:55.15Tartarusw/ 3.2.8
18:55.21beagle-dev-543denix: Oh. What will happen if you add a directory that doesn't exist to FILES, will it ignore it or bomb out
18:56.28denixbeagle-dev-543: I believe ${gcclibdir}/${TARGET_SYS}/${BINV}/include* should work... I need to try
18:56.38Tartarustimtimwork, jfyi
18:56.52Tartarustimtimwork, i'll see if further poking is needed soon, once i get my image built
18:57.03beagle-dev-543denix: ok
18:57.17Gnutoo~seen woglinde
18:57.30ibotwoglinde <i=woglinde@e178095060.adsl.alicedsl.de> was last seen on IRC in channel #oe, 2d 21h 3m 14s ago, saying: 'bye khem'.
18:58.51CIA-203Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * rb58dface5b 10openembedded.git/packages/busybox/busybox_1.13.2.bb: busybox: bandaid for avr32
18:59.50beagle-dev-543denix: I've got to go now, thanks for your help :)
19:00.21denixbeagle-dev-543: np, I'll test it and then try to push it
19:02.50CIA-203Jeremy Lainé <jeremy.laine@m4x.org> 07org.openembedded.dev * r5df0198331 10openembedded.git/: Merge branch 'org.openembedded.dev' of ssh://git@git.openembedded.net/openembedded into org.openembedded.dev
19:03.01CIA-203Jeremy Lainé <jeremy.laine@m4x.org> 07org.openembedded.dev * r1929da3bd3 10openembedded.git/ (5 files in 3 dirs): dnsmasq: update to version 2.47
19:13.07*** join/#oe thesing (n=tkunze@BAA2631.baa.pppool.de)
19:13.52thesingrehi
19:14.04thesingXorA: did you find time for a test?
19:14.22*** join/#oe dijenerate (n=dijenera@65.48.179.42)
19:22.39*** join/#oe BlindMan (n=othmar@h081217029133.dyn.cm.kabsi.at)
19:27.39*** join/#oe pb___ (n=pb@79-79-208-205.dynamic.dsl.as9105.com)
19:32.54*** join/#oe _hc (n=hanschri@128-122-151-160.DYNAPOOL.NYU.EDU)
19:33.05*** join/#oe garren (n=garren@41.246.4.3)
19:35.32*** join/#oe pvanhoof (n=pvanhoof@host-85-27-48-181.brutele.be)
19:42.09*** join/#oe timtimred (n=meh@92-237-114-21.cable.ubr03.chms.blueyonder.co.uk)
19:42.41*** join/#oe mrmoku_ (n=flux@ppp-62-216-217-132.dynamic.mnet-online.de)
19:45.48slapin_hi, all!
19:46.09slapin_do anybody have linker script example for renaming of symbols?
19:50.01leonardo__i'm looking to add the python library runpy to my OE build.  How do I check which python library to install that includes runpy?  There is no python-runpy.
19:52.19*** join/#oe Gnutoo (n=gnutoo@host221-133-dynamic.31-79-r.retail.telecomitalia.it)
19:52.52garrenif I do_install() for my package do I have to specify where it gets installed?
19:57.09thesinggarren: yes. see most of the recipies.
19:57.33*** join/#oe pvanhoof (n=pvanhoof@host-85-27-48-181.brutele.be)
19:59.02khemslapin_: what do you want to do
20:00.00garrencool thanks
20:16.29slapin_khem, I have a blob working with very old kernel API, I want to write wrapper for it
20:17.10slapin_khem, so to rename all imported and exported globals
20:19.15*** join/#oe stefan_schmidt (n=stefan@p5B035179.dip.t-dialin.net)
20:23.47*** join/#oe gremlin[it] (n=gremlin@ppp-151-188.30-151.libero.it)
20:26.16*** join/#oe Sleep_Walker (n=SleepWal@1.57.broadband2.iol.cz)
20:37.21*** join/#oe _hc (n=hanschri@128-122-151-160.DYNAPOOL.NYU.EDU)
20:37.33CIA-203Frans Meulenbroeks <fransmeulenbroeks@gmail.com> 07org.openembedded.dev * r2f56f3b3b0 10openembedded.git/ (2 files in 2 dirs): python-coherence: moved to svn 1161, added dependency on python-twisted-protocols
20:37.55*** join/#oe Stanchooo (n=stancho@78.90.99.168)
20:44.14*** join/#oe ClashTheBunny (i=c72ec6e6@gateway/web/ajax/mibbit.com/x-9f145ee42b0434fc)
20:44.42*** join/#oe grma (n=gruberm@chello212186029093.tirol.surfer.at)
20:48.08*** join/#oe Gnutoo (n=gnutoo@host221-133-dynamic.31-79-r.retail.telecomitalia.it)
21:00.52*** join/#oe ClashTheBunny (i=c72ec6e8@gateway/web/ajax/mibbit.com/x-ef5bd76a0760b26f)
21:14.39*** join/#oe frank7d (n=frank@p549F1F7D.dip0.t-ipconnect.de)
21:27.53*** join/#oe bin1010 (n=aars@cyclone.ecs.baylor.edu)
21:28.35*** join/#oe [X]Spot (n=stancho@78.90.99.168)
21:31.46*** join/#oe robtow (n=rtow@nat/montavista/x-56fe33ac2f80fd75)
21:32.11*** join/#oe archae0pteryx (n=snewman@207.47.42.130.static.nextweb.net)
21:32.49CIA-203Frans Meulenbroeks <fransmeulenbroeks@gmail.com> 07org.openembedded.dev * r4e1b79d2d8 10openembedded.git/packages/opkg/ (7 files in 2 dirs): opkg: fix bug that occured with file names of exactly 100 chars in the tar file
21:32.59*** join/#oe florian (n=fuchs@f048063220.adsl.alicedsl.de)
21:50.45*** join/#oe kristoffer (n=kristoff@94.191.132.27.bredband.tre.se)
21:52.35*** join/#oe hillct (n=hillct@cpe-069-134-049-165.nc.res.rr.com)
21:53.16*** join/#oe khem` (n=khem@adsl-71-146-3-29.dsl.pltn13.sbcglobal.net)
21:54.40florianre
21:58.31*** join/#oe lrg_ (n=lrg@79-68-17-78.dynamic.dsl.as9105.com)
21:59.06CIA-203Frans Meulenbroeks <fransmeulenbroeks@gmail.com> 07org.openembedded.dev * r5a0e1d473c 10openembedded.git/packages/busybox/ (busybox-1.13.2/get_header_tar.patch busybox_1.13.2.bb): busybox: fix tar problem with filenames that are exactly 100 bytes
21:59.08CIA-203Frans Meulenbroeks <fransmeulenbroeks@gmail.com> 07org.openembedded.dev * rb0a6cab1a2 10openembedded.git/packages/zope/zope_3.3.1.bb:
21:59.08CIA-2zope: removed twisted subdir during install;
21:59.08CIA-2this is a very old version of twisted and it shadows the twisted package
21:59.10CIA-2as such it broke python-coherence
22:16.40*** join/#oe jconnolly (n=jconnoll@firebug.buglabs.net)
22:25.16*** join/#oe bin10101 (n=aars@rrcs-24-153-239-89.sw.biz.rr.com)
22:28.27*** join/#oe Sleep_Walker (n=SleepWal@gprs7.vodafone.cz)
22:39.16CIA-203Michael 'Mickey' Lauer <mickey@vanille-media.de> 07fso/milestone5 * r378ecdc561 10openembedded.git/packages/initscripts/ (initscripts-1.0/checkroot.sh initscripts_1.0.bb): initscripts: fix remounting root partition when booting on a read-only mounted root partition
22:42.35khemslapin_: You can create aliases for symbols if that helps
22:48.57*** join/#oe brolin (n=brolin@190.157.13.75)
22:52.28*** join/#oe archae0ptery1 (n=snewman@207.47.42.130.static.nextweb.net)
22:59.32*** join/#oe pvanhoof (n=pvanhoof@host-85-27-48-181.brutele.be)
23:04.02*** join/#oe archae0pteryx (n=snewman@207.47.42.130.static.nextweb.net)
23:25.08*** join/#oe pvanhoof_ (n=pvanhoof@host-85-27-48-181.brutele.be)
23:46.00*** join/#oe pvanhoof__ (n=pvanhoof@host-85-27-48-181.brutele.be)
23:46.25timtimredyay. i have zoneminder running (appallingly) on a beagle
23:46.58timtimredand a almost finished zoneminder image to commit. be good for an atom/epia i guess.
23:48.01*** join/#oe Sleep_Walker (n=SleepWal@193.179.96.131)
23:49.20timtimredas seen herre http://downloads.buffalo.nas-central.org/Users/timtimred/beagleboard/DSC00683.JPG
23:51.22brolinhello all, exist any difference between have a gcc compiled with support for multiples archs and have multiples gcc_s for each architecture?

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.