irclog2html for #oe on 20050725

00:00.25JustinPLuke-Jr: the only libtool in work or stamps is libtool-native
00:00.25JustinPbaking libtool-cross
00:00.43JustinP(of course it *may* still be picking up -cross form staging or something....)
00:05.47JustinPrenoo_busy: nope, rebuild of libtool-cross didn't help
00:07.34reenoo_busywhat does this give you? "grep PB tmp/work/pcre-4.4-r1/pcre-4.4/arm-linux-libtool"
00:14.53*** join/#oe hufnus (~slonsiki@mfa1336d0.tmodns.net)
00:15.58JustinPsorry, will need a few minutes, build machine is being very slow ATM
00:16.13reenoo_busynp
00:16.43*** join/#oe Ken|JLime (~Ken@h242n2fls31o1105.telia.com)
00:22.03*** join/#oe Timelord (~TL@66.84.189.72)
00:25.44JustinPrenoo_busy: that gives me the same thing: # This breaks install into our staging area.  -PB
00:26.07reenoo_busyok. so it's using the patched libtool
00:26.41reenoo_busybest reply to the bug with a description of your findings
00:27.13reenoo_busymaybe one of our resident libtool gurus (pb_ and kergoth) will have a look at it
00:27.46reenoo_busyI've run out of ideas why it might work here but not for you ;)
00:31.16JustinPthanks for trying
00:31.21JustinPI'm composing a reply right now
00:31.37JustinPmy fix still does work...at least for now
00:31.47JustinPcould you test and see if it breaks things on your system?
00:32.42reenoo_busythe patch looked rather safe
00:32.44JustinPI realize it's brute force and a horrible hack, but it works...
00:33.07reenoo_busybut it's not the right thing to do as this might be a systematic error
00:33.13JustinPI agree
00:33.32JustinPhowever it would be good to have the hack in there for other people until the real problem is sniffed out
00:34.49reenoo_busyhmm. I'll think about it
00:34.56reenoo_busyoff to bed for now
00:34.58reenoo_busy'night all
00:52.17CIA-203daka 07org.oe.nslu2-linux * r5bbc6ec4... 10/packages/meta/openslug-native.bb: Add gcc-symlinks, fix the files libpthread.so and libc.so from libc6-dev (that really should be fixed in there)
00:53.35*** part/#oe _husam_ (~kvirc@217.206.111.105)
00:53.47*** join/#oe _husam_ (~kvirc@217.206.111.105)
02:52.50*** join/#oe raster (~raster@vagw.valinux.co.jp)
03:12.43CIA-203daka 07org.oe.nslu2-linux * r2eeab934... 10/packages/meta/openslug-native.bb: Add binutils-symlinks, remove the libc.so and libpthread.so fix since things work without it
03:12.47CIA-203daka 07org.oe.nslu2-linux * r27ad08fd... 10/packages/busybox/busybox_1.00.bb:
03:12.47CIA-2Check for existance of 'test' before assuming it isn't there and alias test to 'busybox test'.
03:12.47CIA-2When building a rootfs the previos implementation cased a few serious (non-fatal) errors showing up with tinderbox.
03:19.13*** join/#oe jbowler-away (~jbowler@jbowler.nslu2-linux)
03:28.20*** join/#oe raster (~raster@vagw.valinux.co.jp)
03:56.33*** join/#oe andersee (~andersee@codepoet.org)
05:08.46*** join/#oe minipanda (~zhanght@221.122.51.131)
05:35.39*** join/#oe aquadran (pablo@aquadran.undead.scummvm)
05:54.43*** join/#oe [cc]smart (~smart@gw.ptr-62-65-149-158.customer.ch.netstream.com)
05:58.38*** join/#oe __law__ (~law@mail.stiftadmont.at)
06:12.13kooblahi all
06:15.54*** join/#oe eumel (~chatzilla@p548327C1.dip0.t-ipconnect.de)
06:17.54*** part/#oe minipanda (~zhanght@221.122.51.131)
06:22.09CoreDump|homemorning
06:43.29*** join/#oe Bam-BamIR (~n0n3y4@71.108.73.115)
06:44.02Bam-BamIRmorning
06:44.55Bam-BamIRanyone got wellenrieter to work on a 3100+dlink card?
06:47.32CoreDump|homewell, it works w/ OZ 3.5.3 on a SL-C1000 (which is close enough to a 3100)
06:47.48CoreDump|homenot w/ a dlink card though ;)
06:48.10Bam-BamIRhmmm, maybe my network settings..
06:50.10*** part/#oe dyoung-zzzz (~dyoung@dyoung.nslu2-linux)
07:12.24JustinPwellenreiter workks fine on a C3000 with Ambicom card
07:21.01*** join/#oe dyoung_ (~dyoung@dyoung.nslu2-linux)
07:28.46Bam-BamIRjustin whats your network config for using it?
07:35.40XorAmorning
07:42.26*** join/#oe BigAl (~bigal@dsl-202-45-99-179.ACT.netspace.net.au)
07:57.44*** join/#oe mithro (~tim@lester.mithis.com)
07:59.41mithrohey zecke, you about?
08:02.02*** join/#oe ade|desk (~adavey@194.200.143.249)
08:04.26*** join/#oe __law__ (~law@mail.stiftadmont.at)
10:17.46*** join/#oe ibot (ibot@apt.bot.TimRiker.active.supporter.pdpc)
10:17.46*** topic/#oe is OpenEmbedded Developer Lounge | http://bugs.openembedded.org/ (l/p guest/guest) | This is not a distribution support channel | The OE mailing list is back, but everyone needs to resubscribe. | We have a bitbake-dev mailing list for discussions on the core. If you're interested in BitBake-Ng, please subscribe. | LCA 2005 Embedded Miniconf - http://www.openembedded.org/miniconf
10:18.07koenpb_: http://ewi546.ewi.utwente.nl/tmp/viewmtn/revision.psp?id=f7914a228f685b9a242fcd8550ccf8d2059a7c71 -> 'tar archive'
10:18.26florian_kcyeah... we need a machine with much of cpu power for this ;-)
10:18.29koenthat doesn't have incremental updates, but it's a start
10:18.31pb_florian_kc: heh, right
10:18.47mickeylkoen: how about changing that link to point to headofbranch.psp ?
10:18.49florian_kcthat's nothing for my escala :-/
10:18.57mickeylit won't work when we have two heads, but this is rarely
10:19.15pb_florian_kc: I guess we should ask for donations to buy a dual opteron or something.
10:19.17mickeyls/rarely/rare/
10:20.04florian_kc:-)
10:20.11BigAlmickeyl: I've been working on the keyboard driver a bit, but it's not playing nice right now.. Not giving me interrupts when I press keys.
10:22.01hrwberlios subversion server need more ram
10:24.47kooblai do hourly anonymous pulls of oe.dev - use wall3.soft.uni-linz.ac.at as a mirror if you want
10:27.17BigAlHas anybody managed to get a gpe-image working on a spitz?
10:28.47*** join/#oe Crofton__|laptop (~balister@66-207-66-26.black.dmt.ntelos.net)
10:30.31*** join/#oe zinga (~arnaudb@186.80-203-227.nextgentel.com)
10:33.12*** join/#oe zecke_ (fakeuser@p54BD2BB8.dip0.t-ipconnect.de)
10:33.22mickeylhey zecke
10:33.25mickeylmithro was chasing you
10:33.39koenmoin zecke_
10:34.07hrwhi zecke
10:40.09zecke_damn
10:40.29zecke_koen: could set my password on ewi to something?
10:40.47zecke_I normally use public key authentication...
10:40.54zecke_koen: or could you install a key for me?
10:40.57pb_hi zecke_
10:41.11zecke_http://ewi546.ewi.utwente.nl/OE_qa/cgi-bin//tinder.cgi?tree=OpenEmbeddedBuild&start-time=1122265801&display-hours=48
10:41.23koenzecke_: you can install a key yourself
10:41.23zecke_doesn't look too bad, actually it looks promising
10:41.31hrwbitbake -i does not allow to 'build world' - but 'build *' works
10:41.44zecke_koen: yes if I would have access to ewi..
10:46.46mickeylhrw: yeah,
10:46.52mickeylhrw: world is a bitbake special case
10:47.00zecke_koen: nevermind, remember my password...
10:47.57*** join/#oe pH5 (~ph5@p5485D0E1.dip.t-dialin.net)
10:50.09koenrwhitby: where is the 2.3 beta firmware hiding?
10:51.43koenrwhitby: http://www.openslug.org/downloads.php doesn't list it
10:53.02*** join/#oe NAiL (~repvik@banan.dataparty-mn.biz)
10:55.53zecke_koen: cp /data/tinderbox2/bin/processmail_builds /data/tinderbox2/cgi-bin/process_builds.cgi
10:55.59zecke_koen: and same for tb1 please
10:56.48koendone
10:57.11koenis that http or email code? or timezone stuff?
10:57.43zecke_koen: it is for receiving build logs per http
10:57.50koencool
10:57.51zecke_koen: as mail is so freaking delayed...
10:58.13koenDeferred: 450 4.3.2 {mx020} Too many mails (mail bomb)
10:58.24zecke_lol
10:58.31koenthat is what gmx had to say about my tinderclient
10:58.40koenand yes, I got bounces for every mail
10:58.57zecke_lol
10:59.04zecke_we could turn off verbose reporting...
10:59.21zecke_or we could install a jabberd on ewi... :}
10:59.31zecke_wow xkb is working here... that feels soooo good
11:00.41zecke_svn up
11:00.41zecke_zecke123@svn.berlios.de's password:
11:00.41zecke_svn: Berkeley DB error while opening 'nodes' table for filesystem /svnroot/repos/bitbake/db:
11:00.44zecke_Cannot allocate memory
11:00.47zecke_svn: bdb: Logging region out of memory; you may need to increase its size
11:00.49zecke_lol
11:00.59zecke_maybe Mr. Schilling switched berlios.de to Schillinx?
11:01.36koenI'd say running a solaris kernel would improve stuff
11:03.40zecke_mickey|lunch: so it was mithro that called me?
11:06.19*** join/#oe Jenna_zzz (~cherryRed@62.116.121.242)
11:06.34*** part/#oe Jenna_zzz (~cherryRed@62.116.121.242)
11:13.12*** join/#oe reenoo_ (~r@84.137.215.90)
11:15.04*** part/#oe Camelot (~Anon@ppp1434.adelaide.on.net.au)
11:16.21hrwre
11:16.34*** part/#oe pH5 (~ph5@p5485D0E1.dip.t-dialin.net)
11:16.37*** join/#oe pH5 (~ph5@p5485D0E1.dip.t-dialin.net)
11:17.38hrwFSCK...
11:17.46hrwreiserfs on / is corrupt..
11:19.59hrwtime to buy dvd writer and/or better powersupply
11:21.05*** join/#oe stevenh (~xev@65.167.23.2)
11:21.17*** join/#oe qfh (~qfh@ppp-82-135-1-94.mnet-online.de)
11:23.45rwhitbykoen: OpenSlug 2.3 binary image has not been released yet
11:23.52XorAhrw: nasty
11:24.12hrwXorA: I installed woody on swap partition to get reiserfsck
11:24.19koenrwhitby: I guessed as much, but everything talks about 2.3 and point to the openslug site
11:24.31koenrwhitby: but I flashed 2.0 for the time being
11:24.35XorAhrw: I abandoned reiser after 3rd time it corrupted my disks
11:24.53rwhitbykoen: I can put a 2.4 binary somewhere for you to download if you like.
11:24.57hrwXorA: I user reiserfs since it landed in 2.4
11:25.12XorAhrw: reiser and amd64 dont seem to mix
11:25.42koenrwhitby: ok, I'll ask that when I get my switch back so I don't have to plug it in to my workstation anymore
11:26.30hrwXorA: that problem is not mine yet
11:31.38hrwuf.. reiserfsck --rebuild-tree finished
11:32.53reenoo_morning
11:33.38hrwbrb
11:34.15reenoo_last time I checked (the first, last, and only time I used reiserfs) reiserfsck was the equivalent of an rm -rf /
11:35.18tmbinci used it twice in my life, and in both cases, an rm -rf / would have been more effective and less time-consuming.
11:35.48koenafter two(!) changes off the on disk format I cursed Hans as a corporate whore and installing the ext2 journaling patches
11:35.50reenoo_hehe
11:35.50pb_XorA: I think what you mean is "reiser and files don't seem to mix"
11:36.24koenexplicitly stating "I break compat for money" doesn't win me over
11:38.21*** join/#oe mithro (~tim@lester.mithis.com)
11:40.32mithrohey zecke_, you alive?
11:45.07rwhitbydid something change in glibc in the last 24 hours?   Unslung depends on glibc-2.2.5 (yes an old version for compatibility reasons), and today it doesn't build.
11:45.35pb_what's the error?
11:45.42rwhitbyand does anyone know how to find out the answer to that question by interrogating monotone?
11:46.27rwhitby"/home/slug/unslung/tmp/work/glibc-2.2.5-r5/build-armeb-linux/csu/crti.S: Assembler messages:
11:46.27rwhitby"/home/slug/unslung/tmp/work/glibc-2.2.5-r5/build-armeb-linux/csu/crti.S:96: Error: can't resolve `_GLOBAL_OFFSET_TABLE_' {*UND* section} - `.L6' {.text section}
11:46.52rwhitbypb_: if you need more info I can pastebin it
11:46.55pb_that is odd
11:47.13pb_what does the offending line of that file say?
11:48.01rwhitbyalso "/home/slug/unslung/tmp/work/glibc-2.2.5-r5/build-armeb-linux/csu/crti.S:61: Error: internal_relocation (type: OFFSET_IMM) not fixed up"
11:48.18rwhitbyand three more instances of that error
11:48.41pb_what versions of binutils and gcc do you have?
11:48.50rwhitby<PROTECTED>
11:49.11rwhitby<PROTECTED>
11:49.21rwhitbybuild host versions?
11:49.28pb_no, cross tool versions
11:49.28koenrwhitby: openslug 2.0 is working pretty good so far
11:50.02rwhitbypb_: we don't fix those to a version, so whatever latest OE versions are
11:50.09pb_rwhitby: are .L6 and .L7 defined in the .S file?
11:50.13pb_.L6 should be very close to line 96
11:50.39pb_actually, maybe not that close, but it should be somewhere in the preceding function
11:50.41rwhitbyL6 is one line 63 and 64
11:50.52rwhitbyL7 is on 95
11:51.21rwhitbybinutils-cross-2.15.94.0.1-r0
11:51.29rwhitbygcc-cross-3.4.4-r0
11:51.56rwhitbydo we need to rollback gcc-cross perhaps?
11:52.08*** join/#oe Virusmaster (~Thomas@p549CF203.dip.t-dialin.net)
11:53.43rwhitbyrwhitby@nudi:/home/unslung/firmware/tmp/work$ ls -ld gcc*
11:53.43rwhitbydrwxrwsr-x    6 rwhitby  users        4096 Apr  2 03:14 gcc-cross-3.4.3-r10
11:53.43rwhitbydrwxrwsr-x    6 rwhitby  users        4096 Jun 23 19:30 gcc-cross-3.4.4-r0
11:53.56rwhitbylooks like up until Jun 23 we were using 3.4.3 and that was working.
11:54.58rwhitbyHmm - looks like the build on Jun 24 with 3.4.4 worked too.
11:56.20koenrwhitby: where should I make openslug feature requests like "could you enable ipv6 as a kernel modules?" ?
11:56.27pb_rwhitby: it sounds more like a binutils bug to me.
11:56.40pb_those two lines look quite valid.
11:57.21rwhitbykoen: as an OE core developer, you'd ask in #openslug whether it's ok to just do it yourself :-)
11:58.14rwhitbypb_: our binutils version hasn't changed since at least Apr 2.
11:58.53pb_rwhitby: odd
11:58.54rwhitbyglibc, glibc-cross were the latest to change on Jun23
11:59.06rwhitby2.2.5-r4 to 2.2.5-r5
11:59.25rwhitbys/glibc-cross/gcc-cross/
12:00.05rwhitbygcc also went from 3.4.3 to 3.4.4-r0 on Jun23 (I'm just looking back at tmp/work in an old dir that I know built fine on May 27 then Jun 23)
12:00.29*** part/#oe koobla (~chatzilla@koobla.active.supporter.pdpc)
12:01.55pb_if you have an old tmp directory, can you diff crti.S between the version that works and the version that fails?
12:03.06rwhitbywill do.  Also trying a rebuild in that old dir of glibc-2.2.5-r5
12:03.24*** join/#oe jamey (~jamey@192.208.47.196)
12:03.30pb_okay, cool
12:03.31pb_hi jamey
12:03.47jameyhi pb_
12:04.22*** join/#oe mithro (~tim@lester.mithis.com)
12:04.27mickeyl~lart CoreDump|afk
12:04.42mickeylCoreDump|afk: hey, any special reason to destroy my work? ;)
12:04.44mickeyl-# autoload modules
12:04.44mickeyl-module_autoload_usb-ohci-pxa27x = "usb-ohci-pxa27x"
12:04.44mickeyl-
12:05.03mickeylI've readded it
12:06.53*** join/#oe tux_mike (~mike@zelda.tuxnami.org)
12:08.39rwhitbypb_: lots of differences between old crti.S and new one
12:08.58rwhitbywant them both in a pastebin?
12:11.18mithromickeyl: do you have zecke's phone number?
12:13.02rwhitbypb_: the glibc-2.2.5-r5 in the old dir rebuilt fine.  now to track down what is different ...
12:13.38*** join/#oe hrw (szczepan@blue.stone.pl)
12:13.45hrwre
12:15.23mickeylmithro: yes, i /msg'ed it to you some hours ago
12:15.29mickeyli can do it again
12:15.50mickeyldone
12:20.44rwhitbypb_: the only packages that have changed betwen the good old build and the bad new build are: (flex-native-2.5.31-r1 -> r2, gnu-config-native-0.1cvs20050616-r3 -> 0.1cvs20050701-r4, libtool-native-1.5.10-r1 -> r3, and linux-libc-headers-2.6.11.1-r0 -> r1).  Could any of them cause a difference in crti.S ?
12:20.52zecke_mithro: hey
12:20.57zecke_sorry
12:23.31mithrozecke_: thats okay
12:23.38mithroso should I give you a phone call now?
12:23.52mithromickey|meeting: my internet is horribly unreliable so I didn't get it till now
12:27.29zecke_mithro: do you have my cell phone number? is it expensive to call it?
12:27.39koenzecke_: does the OE tinderclient use httppost already?
12:27.57zecke_nope
12:28.19mithrocurrently trying to organise how to get to WhatTheHack
12:28.51*** join/#oe koobla (~chatzilla@koobla.active.supporter.pdpc)
12:34.00XorAah interesting, NIST has come clean about their familiar usage http://www.nist.gov/public_affairs/techbeat/tb2004_1110.htm#software
12:38.24jameyx-mobilecard is a smartcard in SD formfactor?
12:38.35XorAjamey: yes
12:38.53XorAjamey: well smartcard + 64Meg of flash
12:38.54*** join/#oe NAiL (~repvik@84.16.208.26)
12:38.56XorAjamey: in its current form
12:39.04jameyinteresting
12:39.06*** join/#oe hrw (szczepan@195.205.148.100)
12:39.07koen"Linux PC/SC Lite driver for Renesas XMC mobile card on IPAQ H3800. This bundle contains source code and binaries for kernel version 2.4.19-rmk6-pxa1-hh30 (Familiar 0.7.2). Additional information and Renesas' developer program can be found at X-MobileCard.com
12:39.12koen"
12:39.31reenoo_-hh30
12:39.31hrwre
12:39.36reenoo_oh well
12:39.38jameyfound it opaque
12:39.52hrwfreebsd 4.9 suxx badly
12:40.09jameyit would probably apply to current 2.4 kernel
12:40.21jameywould need to be updated to 2.6 mmc framework
12:40.26jameyfor 2.6
12:40.54XorAjamey: irritatingly, thats my code on the X-MobileCard and the idiots never thought to ask me anything :-(
12:44.55reenoo_heh
12:45.27jameyXorA:  I'm sorry to hear that.  I'm also disappointed they did not post their code to handhelds.org
12:45.44koen~emulate zautrix
12:45.44ibotthis is exactly what the GPL is about. grab the stuff and run away
12:46.04jameyibot: zautrix
12:46.28jameygrab the stuff, distribute source and changes, and run away
12:46.35jameythey did all the things required by GPL
12:46.48jameyjust not the community building things
12:47.45zecke_jamey: zautrix is the forker at pi-sync.net
12:48.05jameyah
12:49.20zecke_and sadly he gets paid for Qt and Qtopia development...
12:51.15XorAjamey: unfortuneately, NIST are 2 customers down the chain from me ;-(
12:53.55jameywhat can you do
12:54.05jameyah, you could post a patch to handhelds.org and/or kernel.org
12:54.31*** join/#oe zinga (~arnaudb@186.80-203-227.nextgentel.com)
12:54.55RPXorA: The hostap problem is caused by a change in the latest -mm kernels
12:55.13XorARP: so im not mad :-)
12:55.54robtaylorlo all :)
12:56.32RPXorA: No. I had wierd files leftover in staging which was breaking my builds...
12:56.48RPXorA: I'm not sure what the correct fix is :-/
12:58.00robtaylorI've just been pondering on how to do gracful fails if a user pulls power in the middle of a package upgrade.. does anyone know of any schemes that do this?
12:58.31*** join/#oe drw (~drw@c-67-172-219-167.hsd1.tx.comcast.net)
12:58.41pb_rwhitby: none of those packages sound likely to cause the build failure
12:58.53hrwcu
12:59.27*** join/#oe pH5 (~ph5@p5485D0E1.dip.t-dialin.net)
12:59.36rwhitbypb_: I'm leaving it for someone else to fix.  I'm not the toolchain guy :-)  Thanks for your assistance anyway.
13:00.24RPXorA: Your patch looks like the correct fix. I'd check it in. It certainly doesn't hurt anything including non-mm kernels
13:01.49XorARP: should technically wait for mickey|meeting
13:03.02koenjamey: did you hear back from the miplv6 guys?
13:03.51RPXorA: Your call. Looks fine to me though and I kind of maintain the kernel stuff ;-)
13:04.08XorARP: what the hell, we have an SCM :-)
13:04.18RPFeel free to blame me if mickeyl objects ;-)
13:04.27jameykoen:  have not heard from miplv6 guys
13:04.44koenjamey: I just notice the are still on 2.6.8.1 with their patches
13:05.01jameykoen:  I think they posted a later one, but they have been very infrequent
13:05.09jameyvery disappointing
13:05.37koenyeah
13:05.48koenI'm thinking of going the tinc (http://www.tinc-vpn.org/) route
13:06.46*** join/#oe mrz80 (1000@caledonia.cns.ufl.edu)
13:10.47pH5Hi! I made a patch to add hx4700 configuration to tslib: http://pastebin.ca/18449. is this any good?
13:12.52RPpH5: Without a hx4700 machine type, that isn't going to do much :-/
13:13.48RPI'm not sure what the policy is for adding the hx4700 - ideally, we need to encapsulate it with the hx2750 and any other pxa270 based machines...
13:14.08koenconf/machine/ipaq-pxa270.conf
13:15.19RPkoen: The header on the file is a touch misleading ;-)
13:18.49pH5i started from aboeglin's patches. so hx4700 shouldn't get its own machine type?
13:18.53koenRP: it was in my bk repo, but it was not pushed to bkbits
13:19.31RPpH5: No, its part of the ipaq-pxa270 machine type
13:19.37reenoo_koen: "OpenSSL is used to encrypt the traffic". OpenSSL and GPL apps don't mix well from a distribution POV
13:19.57koenreenoo_: que?
13:20.06reenoo_koen: quoting the tinc website
13:20.09RPkoen: Any chance you can resurect any of the changes you'd made. I might them be persuaded to fill in any gaps ;-)
13:20.46koenreenoo_: ok, I'll only use tinc personally
13:21.17koenRP: I was getting at the fact that ipaq-pxa270 doesn't exist on bkbits.net :)
13:22.21RPkoen: ah, right. So its here kind of by accident? :)
13:22.37koenahead of time :)
13:22.41CIA-203coredump 07org.oe.dev * rf7914a22... 10/packages/linux/ (3 files in 2 dirs): Akita/Spitz: Use OE fuer setting CMDLINE, not defconfig.
13:22.45CIA-203mickeyl 07org.oe.dev * r9862905f... 10/packages/linux/ (2 files):
13:22.45CIA-2openzaurus-pxa27x:
13:22.45CIA-2- restore module_autoload
13:22.45CIA-2- reorder metadata to be more comparable to openzaurus-pxa-2.4.18
13:22.49CIA-203xora 07org.oe.dev * r1b091d0a... 10/packages/hostap/ (2 files in 2 dirs):
13:22.49CIA-2hostap-modules: kernel 2.6.13-rc3-mm1 has made changes that make the
13:22.49CIA-2KERNEL_VERSION macro not available to hostap without explicitly including
13:22.49CIA-2the utsname.h file.
13:22.51CIA-2This patch should be harmless for other kernel versions but includes the
13:22.53CIA-2needed file if using 2.6.13-rc3-mm1
13:23.50koenRP: http://ewi546.ewi.utwente.nl/tmp/viewmtn/revision.psp?id=d8f77bc6335858610b8f332521ddee863aa71db6
13:25.21*** join/#oe [g2] (~g2@g2.nslu2-linux)
13:25.48RPkoen: thanks :)
13:25.55mickeylXorA: thanks
13:25.57mickeyl(hostap)
13:26.09XorAmickeyl: no probs, been talking to RP since 9am
13:26.11mickeylCoreDump|afk: please don't use Umlaute in commit messages, this breaks CIA
13:26.22mickeylbbl, wandering home
13:28.04CIA-203koen 07org.oe.dev * r24d3b1da... 10/conf/machine/ipaq-pxa270.conf: clean up override mess and header
13:28.20*** join/#oe repvik (~repvik@repvik-active-pdpc.nslu2-linux)
13:29.03zecke_mickeyl: cya
13:29.14koenRP: when I wanted to put in the defconfigs the great soc->platform-dev conversions was going on
13:30.20RPkoen: I guess my own kernel tree providing the hx2750 kernel is going to be a bit tricky :-/
13:30.37koenRP: I thought you updated hh.org?
13:31.04koenRP: but mainline + hx2xxx +hx47xx would be fine by me
13:31.21RPkoen: not yet. The tree in oe is always going to be more uptodate...
13:31.31RPOf course we don't have hx47xx support in that tree...
13:33.26pH5RP, koen: thanks. i guess i should try to rebuild with ipaq-pxa270 first
13:33.52*** part/#oe koobla (~chatzilla@koobla.active.supporter.pdpc)
13:34.29*** join/#oe koobla (~chatzilla@koobla.active.supporter.pdpc)
13:35.18RPpH5: I think some key support may be missing from that machine type. Patches welcome to fill in the blanks though ;-)
13:35.28koenpH5: that won't work at the moment, but it shouldn't be hard to adapt alex' patches
13:38.23pH5ok, back to fiddling :)
13:38.25*** part/#oe pH5 (~ph5@p5485D0E1.dip.t-dialin.net)
13:43.26CIA-203nail 07org.oe.nslu2-linux * r5eed6a33... 10/packages/nonworking/screen/ (6 files in 4 dirs): Removed screen from nonworking
13:46.17repvikanyone know why ncurses is still in nonworking? I compile ncurses a couple of times a day :)
13:46.44koenNAiL: feel free to remove them
13:50.30NAiLdone
13:53.05*** join/#oe marcan (1337@83.213.70.253)
13:53.43CIA-203nail 07org.oe.nslu2-linux * r21252c4c... 10/packages/nonworking/ncurses/ncurses_4.2.bb: Removed ncurses from nonworking
14:10.15*** join/#oe CosmicPenguin (~nobody@svl-ext-proxy01.amd.com)
14:19.31*** join/#oe Crofton__|laptop (~balister@h80ad92dd.dhcp.vt.edu)
14:21.06*** join/#oe Crofton__|laptop (~balister@h80ad92dd.dhcp.vt.edu)
14:24.08CIA-203rwhitby 07org.oe.nslu2-linux * r9c450302... 10/packages/meta/ (openslug-native.bb openslug-packages.bb): Added gzip to openslug-native and openslug-packages
14:54.55hrwRP: something (using related) changed between 2.6.12-mm1 and 2.6.13-rc3-mm1 on c7x0?
14:54.56CIA-203nail 07org.oe.nslu2-linux * raf9c8fe6... 10/packages/meta/openslug-packages.bb: Added procps
14:55.30*** join/#oe Bernardo (~jbs@Bernardo.sourcemage)
14:57.58RPhrw: In what way? They should be similar as far as I know...
14:58.35mickeylany idea why ipkg-collateral is no longer in the dependencies of a typical spitz image?
14:59.07mickeylaha
14:59.08mickeylnm
14:59.12hrwmickeyl: none of ipkg stuff probably?
14:59.26mickeylno, the problem is something BOOTSTRAP_EXTRA_RDEPENDS =
14:59.28mickeylinstead of +=
14:59.37mickeylhence overriding the stuff inherited through package_ipk.bbclass
14:59.45hrwRP: you trace kernel more then I so I asked you
15:15.26*** join/#oe pH5 (~ph5@p5485D0E1.dip.t-dialin.net)
15:23.43*** join/#oe tigrux (~tigrux@dsl-200-67-236-216.prod-empresarial.com.mx)
15:25.56*** join/#oe ar_ (~ar@port-ip-213-211-233-149.reverse.mdcc-fun.de)
15:30.21NAiLhrw: yeah, removed screen and ncurses from nonworking
15:31.56hrwcool
15:32.29hrwNAiL: nonworking is from old times when we tried to add something and failed
15:32.36NAiLah
15:32.54hrwDL_DIR = 1.5G here... not counting CVS_DIR..
15:32.56NAiLthe rest of the nonworking stuff is stuff I ain't touching so I don't know status
15:33.26hrwNAiL: want arpwatch?
15:33.48RPhrw: No, I don't know of any major changes. A few things even seem slightly more stable (like the ejection of cf memory cards)
15:34.21hrwlooks like I need to upgrade soon
15:34.38NAiLhrw: not that I know. Why?
15:35.02RPhrw: Its not totally tested in that I haven't tried wifi yet. I can't see why things would break though...
15:36.28hrwNAiL: you (nslu people) use more network stuff then I - I have arpwatch buildable but didnt pushed it yet as I dont finished it (but recently got into point where I dont need it)
15:36.59NAiLhrw: aha. It does look like a nice tool
15:37.10RPOne major bug with this is "cardctl eject" still leaves the disk mounted! :-/
15:37.27NAiLhrw: what is missing from it?
15:38.13hrwNAiL: in debian arpwatch runs as 'arpwatch' user, which is not added/removed in package etc
15:38.25XorARP: is that a kernel bug, or is hotplug just not doing right thing?
15:38.59NAiLhrw: aha. I am supposed to do a bbclass to add users and stuff. I haven't gotten around to that yet
15:39.30NAiLSo far, I have no idea how a bbclass actually works
15:40.09hrwNAiL: bbclass is shell or python code
15:41.05NAiLin this case, preferrably shell
15:42.35NAiLAre there any docs or good examples I can look at? Because there are a few packages that needs users added
15:42.51NAiLeg. mysql, openntpd
15:42.58hrwopenembedded/classes/*.bbclass
15:44.33mickeyltry to model it after the update-rc.d bbclass
15:44.40mickeylyou need to think about what variables you need
15:44.43mickeylfind good names
15:44.47mickeylthen pass them into the bbclass
15:44.55mickeylthat's more or less all
15:45.08NAiLOk, I think I'll need a crash course in python :-P
15:45.32NAiLupdate-rc.d looks like a good starting point
15:46.57hrw"bitbake -i;fetch *" is easiest way to test source availability
15:47.19*** join/#oe marcan_ (1337@83.213.70.253)
15:54.54NAiLaha
15:55.04*** join/#oe ljp (~lpotter@203.94.178.46)
16:01.01*** join/#oe NAiL__ (~repvik@banan.dataparty-mn.biz)
16:10.39RPhrw: I'm not sure :-/. I'm sure the kernel used to handle it but am not 100% sure
16:13.03*** join/#oe treke (~ggilbert@tinman.treke.net)
16:15.27hrwmickeyl: can you post on openzaurus.org (and maybe on oesf) that 3.5.4 will be without poodle support if noone join?
16:17.35pb_yah, update-rc.d.bbclass and/or update-alternatives.bbclass are pretty good as simple examples
16:17.48pb_package.bbclass is a good example of more complex stuff
16:26.06CIA-203rwhitby 07org.oe.nslu2-linux * ra1fa598b... 10/packages/meta/openslug-native.bb: Added python to openslug-native
16:26.10CIA-203rwhitby 07org.oe.nslu2-linux * rb6cba454... 10/packages/meta/openslug-native.bb: Added more python packages to openslug-native
16:27.58*** join/#oe jipi (~jipi@202.156.187.184)
16:34.21hrwhi keturn
16:34.24hrwkergoth: hi
16:34.28kergothhey
16:34.40RPAnd in two posts, rmk makes any chance of progress on collie approximately equal to zero :-/
16:35.03kergoth:(
16:35.39RPIn summary he is the original driver author and therefore any changes to the driver need to be presented to him for his approval before any further progress can be made...
16:35.55RPDespite his own admission he has no clue about how to get it into the kernel!
16:42.07hrwRP: which posts?
16:43.27RPhttp://lkml.org/ Look down the list for rmk
16:44.11hrwok
16:44.48hrwwhole thread?
16:44.57koenRP: is "send them to me" the same as "cat them to /dev/null" ?
16:45.29koenin RNK's world
16:45.33koens/N/M/
16:48.00RPkoen: yes, in my experience its exactly equivalent
16:48.18RPIt would also appear to be where peronally addressed emails go :-/
16:48.24RPDon't get me started...
16:48.40koenI assume replying with "sending the patches to /dev/null amounts to the same" will piss of RMK?
16:48.54koenalong with a list of "pending" RMK stuff
16:48.58RPIt would :)
16:49.41koen"Hello Russel, here is a list of other stuff you promised to look at, and never got done:"
16:49.45koen<big list>
16:49.52koen"regards, Richard"
16:50.09RPHe'd just winge about lack of time/help and blame everyone else...
16:51.22koenone could suggest "we want to take some load of your shoulders, <sucking up>, let us handle it for you"
16:52.26RPI'm not going there. I suspect he seriously dislikes me by now...
16:53.07RPHe seems to prefer John in that he referred to "John or someone" the other day :)
16:53.29RPDespite 50% of the code he was talking about was written by me and I was in the cc:...
16:56.06*** join/#oe france_ (~france@dhcp-208-55.crl.hpl.hp.com)
16:58.25koenhey france_
16:59.50*** join/#oe dkey (nobody@193.170.48.236)
17:03.20hrw~lart elfutils
17:18.37hrwany reasons to bump libxml to 2.6.20?
17:26.18*** join/#oe mithro (~tim@83-169-171-16-dynip.superkabel.de)
17:27.06mithrohowdy people
17:27.08mithroat zecke's house today
17:27.30RPhi mithro
17:27.49RPYou seem to be getting around :) Having fun?
17:36.07mithroyeah
17:41.11france_koen: hi
17:41.44CosmicPenguingrrr...
17:41.54CosmicPenguinmonotone always finds new ways to piss me off
17:42.15JustinPwhat happened now?
17:42.29CosmicPenguinI'm getting the read from fd 4 bullshit
17:42.50CosmicPenguinI know its my own damn fault for not trying montone before the July 1 deadline, so I won't complain
17:43.01JustinPwhat version do you have?
17:43.08CosmicPenguin.20
17:43.18JustinPjust making sure
17:43.27CosmicPenguinIt worked last monday.. :)
17:43.42JustinPtry the backup server
17:43.54CosmicPenguinHow do you switch servers with an existing tree?
17:44.07koenthe tree doesn't know about a server
17:44.21koenin fact, monotone doesn't know servers at all
17:44.40CosmicPenguinBut when I do a pull, it knows about montone.vanille.de
17:44.54koenit remembers you last used vanille as a peer
17:44.59CosmicPenguinahh
17:45.08CosmicPenguinso I just do a mt pull foo.bar and I'm good?
17:45.12koenyes
17:45.44hrw~lart libxml
17:46.22CosmicPenguinAlright - whats the backup server de-jour?
17:47.17koenewi546.ewi.utwente.nl, monotone.nslu2-linux.org or dominion.kabel.utwente.nl
17:47.35CosmicPenguinWhich one will piss off the most people? :)
17:47.35koenor any other people running monotone serve
17:47.40koen:)
17:48.22CosmicPenguinNope
17:48.26CosmicPenguinfailure on all three
17:49.08koenmonotone: fd 5 (peer 199.45.160.146:58537) exchanged goodbyes and flushed output, disconnecting
17:49.11koenis that you?
17:49.37*** join/#oe NAiL__ (~repvik@217-13-18-99.dd.nextgentel.com)
17:49.56CosmicPenguinThats me
17:50.23CosmicPenguinSo you're saying its my fault, huh?
17:50.38koenit doesn't seem to throw errors
17:51.13koenyou could download http://ewi546.ewi.utwente.nl/OE/OE.db.bz2, extract add pubkey and take it from there
17:51.23CosmicPenguinYeah, but I have a damn DB
17:51.29CosmicPenguinthis needs to work
17:54.13CosmicPenguinSo yu have to specify the full string, it appears
17:54.19CosmicPenguinjust --db <pull> doesn't work
17:54.37koenah yes
17:54.48koenmonotone --db=foo.db pull hostname branches
17:55.06CosmicPenguinDer wiki needs to be updated
17:56.00koenhelp yourself :)
17:56.40CosmicPenguinCan do you have to do a org.openembedded/.* always, or just on the first pull?
17:56.47CosmicPenguinerr - remove that slash
17:56.59koenfirst pull, it remembers it after that
17:57.39CosmicPenguinWow - yeah, thats better
17:57.52*** join/#oe mithro (~tim@83-169-171-16-dynip.superkabel.de)
17:58.25CosmicPenguin.* had 406 revs, .dev just had 37
18:01.15*** join/#oe Timelord (~TL@66.209.31.29)
18:06.28*** join/#oe W8TVI (~duck@mg.acegroup.cc)
18:10.14mickeylre
18:11.12hrw|gonere mickeyl
18:11.15hrw|gone~seen p_
18:11.15ibothrw|gone: i haven't seen 'p_'
18:11.16hrw|gone~seen pb_
18:11.16ibotpb_ <~pb@2002:5168:d38c:1:a00:1fff:fe06:93c> was last seen on IRC in channel #gpe, 47d 19h 23m 1s ago, saying: 'later all'.
18:11.28mickeylhrw|gone: ya, i'll write a mail to oz-devel,oz-users about poodle support
18:11.45hrw|gonemickeyl: and put it on website
18:11.57zecke_re
18:12.02mickeylin preparation for angstrom...
18:12.08mickeylI want two people for every machine
18:12.13mickeylpeople that feel commited
18:12.13zecke_CosmicPenguin: if you get such read on... errors
18:12.14hrw|gonelibxml2 2.6.20 builds but libxslt 1.1.14 no
18:12.23mickeyli think we will find two people for all machines
18:12.26mickeylexcept poodle
18:12.29JustinPhas angstrom actually been "started"?
18:12.36zecke_CosmicPenguin: either vanille.de has routing issues or your branch names are wrong
18:12.41trekeso what is angstrom? :p
18:12.47mickeylJustinP: in our heads
18:12.49CosmicPenguinzecke_: the issue that that branch names must be specified
18:13.00CosmicPenguinzecke_: once that is done, life is good
18:13.22JustinPmickeyl: that's about what I thought. ;-) Nice to know work is being done, though. Isn't angstrom waiting for EABI?
18:13.29*** join/#oe hufnus (~slonsiki@m111c36d0.tmodns.net)
18:13.30CosmicPenguinWell, as good as downloading a revision every two minutes can be
18:13.33mickeylJustinP: yes, we're waiting for that
18:13.41zecke_JustinP: war is over if you want ;)
18:13.52zecke_JustinP: EABI is present, but probably not buildable...
18:14.15JustinPah....interesting....why is it not buildable?
18:14.31mickeyltreke: a fresh start :) One feed, one ABI, combines releases, and some improvements to workflow
18:14.32koenglibc doesn't have EABI support yet
18:14.39trekeah nice
18:14.47JustinPah
18:15.02mickeylsupported target hardware: all ipaqs, all zaurii, the simpad, and perhaps more
18:15.14trekemickeyl: this documented anywhere?
18:15.18mickeylalso a new way to drag people who have had reservations to OpenZaurus into :))
18:15.22zecke_mickeyl: kergoth : do you know what is up with berlios?
18:15.48mickeyltreke: not yet. koen and me are working on a roadmap and a policy document to get a discussion base up to angstrom-dev@hh.org
18:15.56trekegah
18:15.57mickeylthen we'll see how things work out
18:15.59trekeanother mailing list
18:16.03mickeylsure
18:16.05mickeylthe idea is
18:16.06mickeyl+1
18:16.07mickeyl-3
18:16.11mickeyltotal reduction: 2
18:16.12mickeyl:D
18:16.32trekehehe
18:16.45*** join/#oe mithro (~tim@83-169-171-16-dynip.superkabel.de)
18:17.47koenmickeyl: let's not create angstrom-users
18:17.49koen~users
18:17.49ibotso this is what we get for sharing our unpaid volunteer work with you? complaints and accuses? <irony> well done, this clearly supports our motivation to continue working on open source projects. </irony>
18:18.00koenmickeyl: we clearly don't want users ;)
18:19.16mickeylheh, no kidding. I really thought about that
18:19.21mickeylbut not because we don't want users
18:19.32mickeylrather because we think all users are kind of developing this distro
18:19.39mickeylno matter what contribution
18:19.40mickeylbugs
18:19.41mickeyldocs
18:19.43mickeylfixes
18:19.46mickeyletc.
18:20.02mickeylheh
18:20.23koenyou do have a point
18:21.11mickeyli think it's worth to try it like that. the distinction between users and developers is too much of a commercial legacy to me
18:21.17koenmickeyl: we just need to have an infrastructure (bugzilla, feeds, QA, lists, maintainers, testers) before going live
18:21.23mickeylkoen: definitly.
18:21.36mickeylkoen: thankfully there's no hurry
18:21.48koenideally angstrom beta wouldn't be called angstrom
18:22.00JustinPwhy's that?
18:22.56koentoo keep people from building prejudices from the firsts beta's
18:22.59mickeyl*nod*
18:23.13mickeylI really would like to have a polished first release this time :)
18:23.15mickeyloh btw.
18:23.28mickeyli think we should join the new age version numbers
18:23.35mickeylAngstrom 2006/01
18:23.36koen2006.1
18:23.39mickeyl*nod*
18:23.46mickeylrather than some more or less arbitrary versioning
18:23.48mickeyljust the date
18:24.12koenwhich is rather a good estimate IMHO
18:24.21mickeylyep. glad we agree on that
18:24.34JustinPthat soon, huh?
18:24.50trekejust don't do  it ubuntu's rather dumb way
18:24.59mickeylhow do they do?
18:25.06JustinPoh, I've been meaning to ask...I heard that familiar wasn't joining up with angstrom. Did I hear wrong?
18:25.16trekeY.M
18:25.21trekeso 4.10
18:25.23treke5.04
18:25.32mickeylhmm 5.04 looks odd
18:25.34trekeI think the next one is 5.10
18:25.35trekeyeah
18:25.39trekeit's the date
18:26.59trekebut in a form that just looks arbitrary
18:27.00mickeylJustinP: delicate topic. I can't make any more comments except a) Angstrom will also be for ipaqs and b) Familiar Linux will continue to exist.
18:27.00mickeyl:)
18:27.00JustinPmickeyl: ok, I'll leave it alone, then
18:27.20CIA-203jbowler 07org.oe.nslu2-linux * r054c1e56... 10/packages/man/man_1.5p.bb:
18:27.20CIA-2Fix the build - under some circumstances the do_install step will end up
18:27.20CIA-2building stuff because of make timestamps (only on build systems with
18:27.20CIA-2nano-second timestamps). Fix by passing the install command the compile
18:27.20CIA-2command DEFS.
18:27.40koenJustinP: familiar is going to be like debian/stable and angstrom like debian/testing (sort of)
18:29.01JustinPkoen: I hope you don't *really* mean that...debian/stable has never been up to date in my experience...I really hated that.
18:30.21trekeJustinP: debian stable hasn't been up to date, but it has usually been reliable :)
18:30.26trekeand that has its advantages
18:30.40*** join/#oe zap (~zap@217.170.93.9)
18:31.30JustinPtreke: I suppose so. I'm generally a developer type, though, so I like up to date and experimental things to try out. :-)
18:32.21trekemost users of a pda like device probably do want it to  work reliably though
18:32.42treke:)
18:32.57koentreke: why do they keep buying wince?
18:33.09trekebecause for the most part it does? :p
18:33.23koen*when* it works ;)
18:34.35koenthink of windows ME
18:34.44koenthrow up
18:34.51koenpet your linux powered PDA
18:37.20reenoo_heh. Gentoo. don't get me started.
18:37.32reenoo_~lart gentoo
18:37.40reenoo_~lart gentoo one more time
18:39.09trekeok
18:39.20trekeon angstrom-dev, but I dont imagine I'll be involved much at all
18:42.47hrwre
18:56.43koenRP: did the nokia people copy damage onto the driver level to avoid the dma penalty?
19:01.25hrwERROR: Nothing provides evas
19:01.26hrwERROR: dependency evas (for edje) not satisfied
19:01.26hrwNOTE: no buildable providers for edje
19:01.30hrwhmm..
19:04.01koeniirc it's virtual/evas
19:04.10koendue to the evas/fb and evas/x11 thing
19:04.29koenotherwise schurig will delete evas because X isn't in his staging
19:04.32RPkoen: It all to do with the weird framebuffer the 770 has...
19:04.48RPnot too much bandwidth to it
19:04.51*** join/#oe NAiL__ (~repvik@84.16.208.26)
19:07.00koenI'm still deciding wether to use the stock maemo image or an OE maemo-image
19:07.00koenseeing how they vandalized FHS
19:09.17*** join/#oe darkschneider (~gab@213-140-6-96.fastres.net)
19:25.01RPIt looks like there's a hotplug/udev/cardmgr race on the cf memory card issue. Something pulls /dev/hda1 before cardmgr can call umount...
19:25.58hrwRP: it looks like udev has to call mount/umount now rather
19:26.49RPhrw: That can't be right - it was always said udev was *not* to do that. It should probably fall to hotplug...
19:26.56RP(same as mmc)
19:27.41hrwudev 060+ got RUN parametr for rules
19:27.47koenRP: you mean the hotplug that is planned to be obsoleted by udev rules?
19:28.07kergoththats what you'd expect.  hotplug calls udev, udev creates/removes the device.  i'd expect that RUN/dev.d depending on udev version would be the correct solution
19:28.09RPtrue :)
19:28.11kergothheh
19:28.23hrwkergoth: welcome udev masta
19:28.26kergothheh
19:28.48RPI think I've just proven I'm not the person to sort the mess out :)
19:29.03CosmicPenguinpcmcia hotplug will make that much more clear
19:29.15kergothindeed
19:29.17kergothdown with cardmgr!
19:29.22CosmicPenguinno shit
19:29.23hrwpcmcia hotplug is the only way after November 2005
19:29.44kergothanyone tried it out yet?
19:29.45kergothheh
19:29.48RPWe have a pcmcia hotplug now?
19:29.50kergothpcmciautils has everything you need
19:29.53kergothyep
19:29.59hrwI tried
19:30.00RPDoes it compile now? :)
19:30.03CosmicPenguinI haven't updated to the .13-rc yet
19:30.17CosmicPenguinbut everybody should - all the cool shit at OLS is there
19:30.17hrwhostap-modules are not compatible with pcmcia hotplug yet
19:30.53RPI can see some headaches coming up :-/
19:31.23hrwRP: you can see example in my patch to linux-openzaurus_2.6.12+
19:31.52CosmicPenguinHeadaches are acceptable - we know cardmgr sucks
19:32.04CosmicPenguinjust like we knew that devfs sucked, so the udev pain was/is acceptable
19:32.47RPI know. I'm just thinking of the complexities of supporting both this and old style cardmgr in one image...
19:33.25hrwRP: virtual/pcmcia-manager?
19:33.48CosmicPenguinI say rip it right off - like a band-aid
19:34.00RPCosmicPenguin: What about 2.4 kernel users?
19:34.03hrwCosmicPenguin: we have too many 2.4 users
19:34.38RPhostap will have to compile for both etc.
19:34.41hrwand pcmcia-hotplug wont be accepted for 2.4
19:35.17*** join/#oe jh___ (~jason@host-69-144-80-55.bzm-mt.client.bresnan.net)
19:35.20CosmicPenguinit is often not a good policy to ignore new technology in favor of legacy techology
19:35.23CosmicPenguinBut I'm sure you already knew that
19:35.54RPBut at the same time, we can't ignore the legacy users...
19:40.25hrwhmm.. maybe the method for small flash devices would be CoreDump altboot+'cardfs'
19:42.08hrwfrom Zaurus machines all except collie will probably sooner or later go 2.6 (or trash)
19:44.42*** join/#oe emte (emte@d64-180-41-158.bchsia.telus.net)
19:45.52*** join/#oe micropal_ (ident@40.80-202-64.nextgentel.com)
19:46.03*** join/#oe ljp (~lpotter@203.94.178.46)
19:51.44CIA-203hrw 07org.oe.dev * rf6d1ae33... 10/packages/libxslt/libxslt_1.1.7.bb: fixed SRC_URI for libxlst 1.1.7
20:19.34*** join/#oe Pigi (~NoOne@Pigi.active.supporter.pdpc)
20:19.43PigiCiao all
20:20.23hrwalo Pigi
20:20.27Pigihi hrw
20:20.28hrw~lart stage.maemo.org for lack of resume
20:21.01*** join/#oe andersee (~andersee@codepoet.org)
20:21.04hrwok - will fetch that archive and time to sleep
20:23.21Pigiis monotone.vanille.de down for some scheduled maintanance AFAYK ?
20:23.59hrwtry ewi or monotone.nslu2-linux.org
20:24.05Pigithx
20:25.12CosmicPenguinThey syould just schedule uptime
20:25.15CosmicPenguinthat would make it easier
20:25.20Pigiheh
20:25.42PigiI will "time" my pull. Today, after three hours I had to stop it :(
20:26.07CosmicPenguinYou can get some idea of how it will do by counting how many revisions you get in a minute
20:26.35Pigiyes, more or less. But "time"ing it would be more accurate I suppose.
20:28.17Pigiheh, it took 3 minutes for 2 revs.....  I think it will be a long night :)
20:28.35CosmicPenguinhow many revs you grabbing?
20:29.03PigiIf I have understood correctly, "revs in" is 493
20:29.20JustinPhaven't pulled for a while, eh?
20:29.33JustinPin my experience "revs written" doesn't go to the "revs in" number
20:29.43JustinPusually around half of "revs in" or a little over
20:29.48Pigia week or so. I've just been reconnected after my move to new house
20:30.06hrwPigi: so it will take few hours
20:30.19JustinPyep
20:30.34Pigiheh. Fortunally I'm home now, and can leave the pc running all night long.
20:45.01Pigihi koen
20:45.15koenhey Pigi
20:45.35Pigiit's a long time..... I've been finally reconnected... :)
20:50.55*** join/#oe _husam_ (~kvirc@217.206.111.105)
21:06.47*** join/#oe jamie (~jamie@jamie.active.supporter.pdpc)
21:18.09CosmicPenguinkoen: so, when you do a pull the second time, the server name is optional, right?
21:18.22koenCosmicPenguin: it should be
21:20.33CosmicPenguinkoen: updated the wiki then
21:20.43koenok
21:32.33*** join/#oe gerwinin (~gerwin@84.87.179.14)
21:32.41gerwininI cannot get bitbake
21:32.49gerwininI get a connection refusedn
21:33.01gerwininDoes anybody know what the problem could be ?
21:34.15CosmicPenguinI think, but I might not be sure, that berlios is sickly
21:35.21koenmickey_away:has a snapshot on vanille.de and the nslu2 people have it in monotone
21:37.02gerwininKoen hoe installeer ik bitbake dan , ook via svn ?
21:37.21koenhttp://vanille.de/temp/oe/bitbake-snapshot-050728.tar.bz2
21:38.08JustinPgerwinin: I thought that was English for a second and it made my brain hurt....
21:38.28JustinP;-)
21:39.09CosmicPenguinWow
21:39.12gerwininJustinP sorry :) doesn't happen to often you see a fellow dutch person
21:39.20gerwininat oe
21:39.30CosmicPenguinHeh
21:39.36CosmicPenguinThats funny - we see them all the time... :)
21:40.02gerwininI never met somebody from Eindhoven here :)
21:41.59*** join/#oe jamie (~jamie@jamie.active.supporter.pdpc)
21:42.56JustinPgerwinin: it's perfectly all right to use your own language for questions and such, it just confused me for a second :-)
21:45.20JustinPwow, 4 revs in in about 2 seconds. That's nice :-)
21:55.48Pigitime to sleep
21:55.50Piginight all
21:56.09*** join/#oe treke (~ggilbert@tinman.treke.net)
22:15.21gerwiningetting an error building bitbake
22:15.45emteyou dont generally build bitbake, you build thing with bitbake
22:15.50emtethings*
22:15.55gerwininrpm -ba --define _topdir /root/bitbake/build/bdist.linux-i686/rpm --clean build/bdist.linux-i686/rpm/SPECS/bitbake.spec
22:15.55gerwinin-ba: unknown option
22:16.16gerwininI know but I want to build bitbake first to be able to use oe
22:16.35emtesince when have we had an rpm version of bitbake?
22:17.15emteanyway you will have to write the rpm.SPEC file if you want to do it that way
22:18.07CosmicPenguinI had no idea that even was a possiblitiy
22:18.09emtei wasnt aware anyone had ever created an rpm
22:18.25CosmicPenguinI'm shocked - shocked!
22:18.33emteit is, but not really a functional one since it changes quite a bit
22:19.19koenevery time someone says 'rpm' I get pictures of hours of searching the web for the needed dependencies
22:19.31emtethat was the old system, yes
22:19.35CosmicPenguinOh, and bitbake would be a doozie
22:19.38emtei doubt it
22:19.52CosmicPenguinI've seen people try to update Python on Redhat 9 boxes
22:19.53emtebut FC and RHE dont have taht problem any more
22:19.54CosmicPenguinits not pretty
22:20.11emteboth cleanly follow deps
22:20.22emteand normally use only signed rpms now
22:20.44emtesimular to how debs do it
22:21.14CosmicPenguinyou heard it here first - emte hates debs, loves rpms!
22:21.20emteanyway gerwinin your far better off to pull directly from svn
22:21.31emteactually, i dont mind either
22:21.41emtei hate building both
22:22.02emtemy current workstation is debian and my gf's system is FC
22:22.04gerwininemte I get a connection refused if I do that
22:22.43emtehmm
22:22.45koenemte: berlios.de is pretty much dead
22:22.52koenat least, it smells that way
22:23.08emtemaybe you can convince koen to tar up his and send it to ya :P
22:23.17emtemine is way too far out of date
22:23.31koenlike this: http://vanille.de/temp/oe/bitbake-snapshot-050728.tar.bz2?
22:23.37emteyeah :P
22:23.50koenit's also in the nslu2 monotone repo
22:24.06gerwininI downloaded it and untared it but I keep on having all kind of errors
22:24.14emtewhat errors?
22:24.26emtemost are related to missing onfig/deps
22:24.30emteconfig*
22:25.03gerwininERROR: Unable to open conf/bitbake.conf
22:25.16emtebad config
22:25.21emteenviroment
22:25.33emteecho $BBPATH
22:25.37emteecho $BBFILES
22:25.46koenBBFILES is sent in local.conf
22:26.09emteyeah but i am guessing none of the enviro is setup
22:26.36gerwinin/root/bitbake
22:27.35koeneeks!
22:27.40koendon't run bitbake as root!
22:27.59koenif some package install goes wrong you end up with arm binaries on your host
22:28.18koen(arm/mips/sparc or whatever the target is)
22:28.58gerwininI am not running it as root
22:29.02gerwininCompiling it as root
22:29.24koenbitbake is a buch of python scripts
22:30.15koenuntar bitbake, check out OE repo, seth PATH and BBPATH, edit local.conf, done
22:30.28koenanyway,
22:30.29koennight all
22:31.24JustinPgerwinin: you don't have to compile bitbake
22:32.14emtehe has all of bitbake itself
22:32.20emtehe doesnt have it setup
22:32.27emteand bitbake doesnt get compiled
22:32.38emteits a collection of scripts for the most part
22:32.50emteprimarily written in python
22:33.07gerwininOkay so I don't need to install it emte ?
22:33.20emteyou need to set it up
22:33.42emteyour question at this point is not really valid
22:34.19emtewhen it is setup and working, the first thing it does is setup/install the cross-compile enviroment
22:34.54gerwininOkay
22:36.15emteto "install" bitbake itself all you need to do is configure it's enviroment so it knows where to find everything
22:36.15emtebtw are you reading the setup instructions from the wiki?
22:36.39emtehttp://oe.handhelds.org/cgi-bin/moin.cgi/GettingStarted
22:38.08emtehmm i think i see where the rpm version of bitbake came from
22:39.39emtewhere did i put all my a/d docs
22:40.38gerwininyes
22:43.31gerwininBut it seems svn is not working anymore
22:45.42rwhitbycorrect.  That's why we (nslu2-linux) are moving all our build dependencies (including bitbake) away from berlios.de - they don't seem to be bothered working on the weekend, whereas we usually do.
22:46.14rwhitbyWe have copied bitbake to our monotone repo.  We respectfullly suggest that OE moves bitbake away from svn completely.
22:46.46rwhitbyIf they do, then we will begin syncing it from the new stable location.  In my opinion Berlios.de has shot itself in the foot for the last time.
22:46.49kergothmoving it to monotone gains us absolutely nothing.  bitbake doesnt need what monotone provides, and it being in svn avoids the headaches associated with monotone.
22:47.02rwhitbykergoth: put it somewhere which stays up then.
22:47.05emtedown servers?
22:47.20rwhitbykergoth: nod on the not needing to be in monotone
22:48.19emtei am hoping it was to put it somewhere which stays up
22:49.05rwhitbyemte: I was nodding to kergoth that it doesn't need to be in monotone.
22:49.11emteah
22:49.18emteyeah
22:49.27gerwininbut my question is how do I install it now ?
22:49.31emtebut i was hoping his was to another server :P
22:49.55rwhitbygerwinin: I'm with koen|ewi - never install it - just run it from the source dir.
22:50.54emteyeah... whoever did the newest version of getting started has made it rather difficult to follow
22:52.16emtegerwinin, after unpacking and putting it somewhere for a USER to access
22:52.41emtejump to the "Setup the enviroment" part in getting started
22:53.23emteif your using a dedicated user you could actually edit your .bash_profile to export BBPATH when you login
22:54.06JustinPI just source a shell script
22:54.23JustinP3 simple lines
22:54.23JustinPexport PATH=/usr/local/arm/2.95.3/bin:$PATH
22:54.24JustinPexport BBPATH=/home/papercrane/oe/build:/home/papercrane/oe/com.reversefold.oe
22:54.24JustinPcd /home/papercrane/oe/build
22:54.46JustinP(of course I'm building for a zaurus so I need the PATH line, others dont' need it)
22:54.48emteyeah that looks like a zaurus setup
22:55.29emtedepending on your paranoia level you might want to take it farther
22:55.47JustinPeh?
22:55.52JustinPhow's that?
22:56.14JustinPyou mean if I don't trust the OE/BB developers I should run a user which only does OE building?
22:56.14emtewell if you have ever had a cross-compile escape you get a bit more paranoid
22:56.26JustinPescape?
22:56.35JustinPwell, I'm not running it as root...
22:56.55emteusually a mistake you do as a user once in a while
22:57.12emteand changing some things results in very bad juju
22:58.07emtedepending on what the user has access to, you can do almost as much damage as if it were root
22:58.47*** join/#oe Timelord (~TL@66.84.189.72)
22:58.53emtethere are/have been users who run bitbake in a chroot for just that reason
22:59.23JustinPtrue....perhaps I should switch to another user then...
22:59.43emteits all about your comfort level
23:00.13emtei use a dedicated user myself and have the enviroment load up in my .bash_profile
23:00.25JustinPmakes sense.
23:01.00CosmicPenguinI have had more then one occassion where a bitbake fails because "cannot copy to /usr/bin: permission denied"
23:01.06CosmicPenguinWhew
23:01.25emteyeah those nice hardcoded instructions
23:02.01JustinPMe too. Which is why I don't run it as root
23:05.19emtehttp://oe.pastebin.com/320633
23:05.28emteis my old .bash_profile
23:05.34emtehas a few redundancies
23:06.06emtedont think its even valid for the current bb now
23:08.23emtelol
23:15.22CosmicPenguinMad probs to the first person to do a xen domain just for OE
23:15.34CosmicPenguinwell, mad props too, but I think you'll have your share of probs
23:18.44emtewould be nice to get OE into eclipse ...
23:19.04*** join/#oe ljp (~lpotter@203.94.178.46)
23:19.08emtewell it would make it prettier anyway
23:19.19emtedunno about anything beyond that
23:21.51emtebah i hate it when my printer postscript barfs ...
23:22.09*** join/#oe Geo_KM (~keith@bh02i525f01.au.ibm.com)
23:25.59*** join/#oe hufnus (~slonsiki@m221636d0.tmodns.net)
23:38.23reenoo_'night all
23:40.42*** join/#oe ljp_ (~lpotter@203.94.178.46)
23:44.56chouimathi ljp_
23:54.55gerwininI am getting this error with monotone
23:55.03gerwininmonotone: misuse: monotone book-keeping directory MT

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with blootbot logs, split per channel, etc.