T2 IRC Log: 2008-02-21
This is the log as captured by an IRC bot in the channel. The statements are those of the individual people and might not neccessarily reflect the policy and legal rules as set forth by the T2 SDE Project.
--- Log opened Thu Feb 21 00:00:06 2008
01:40 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has quit [Read error: 110 (Connection timed out)]
04:12 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has joined #t2
06:07 -!- emte [n=emte@66.183.8.167] has joined #T2
07:12 -!- emte [n=emte@66.183.8.167] has quit [Read error: 104 (Connection reset by peer)]
07:31 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: Enqlave
07:32 -!- Netsplit over, joins: Enqlave
08:25 -!- jeffdameth [n=jeff@dyndsl-091-096-058-028.ewe-ip-backbone.de] has joined #t2
08:25 < jeffdameth> rxr: i got the kernel to build. now it fails at binutils with no more errormessages than
08:25 < jeffdameth> make[1]: Leaving directory `/home/jeff/src/t2-trunk/src.binutils.default.20080221.081953.9241.rappel/binutils-2.18.50.0.3/objdir'
08:25 < jeffdameth> make: *** [all] Error 2
08:27 < jeffdameth> should i paste my linux26.desc somewhere, so you could try?
08:37 < mtr> hi jeffdameth
08:37 < jeffdameth> hello
08:38 < mtr> pasting the binutils error log would help, in src.binutils.default.*/ERROR.LOG
08:38 < mtr> you changed the linux26.desc?
08:40 < jeffdameth> yep. openmoko changed the patches. or at least one would have to use the old branch and not trunk to fetch the patches
08:40 < jeffdameth> i've now updated the desc from 2.6.21 to the .24 branch
08:41 < mtr> ah, you build the neo target
08:43 < mtr> but I doubt that the updated kernel is related to the binutils failure
08:44 < jeffdameth> ok i guess this is the cause floatformat.c:(.text+0x2ec): undefined reference to `frexp'
08:48 < jeffdameth> hmm floating point stuff. maybe it needs a wrapper or something since arm has no floating point afaik
08:51 < mtr> does it fail in stage 0 or in the cross-build stage 1?
08:52 < mtr> the former would indicate a problem with the host system, maybe missing libraries or such
08:53 < mtr> please paste or publish the error log, maybe there are some more hints in the logfile
08:56 < jeffdameth> http://staff.get-e.org/~jeffdameth/ERROR-LOG
08:58 < jeffdameth> how to figure out the stage? does =[1]=> mean stage one ?
09:01 < mtr> yes
09:05 < mtr> I wonder if the uClibc libm for arm has frexp ...
09:09 < jeffdameth> i found a hint here http://www.mail-archive.com/t2@t2-project.org/msg00488.html
09:09 < jeffdameth> but how to 'use the T2 compiler wraper to inject "-lm": ' isnt explained there
09:11 < mtr> yep ,seems that the arm binutils gas needs to link explicit with -lm
09:15 < mtr> injecting the -lm with the SDE wrapper is one option, but maybe we should add the patch from
09:15 < mtr> http://www.busybox.net/lists/uclibc/2006-August/016088.html
09:17 < jeffdameth> ok thanks. got it working with the injection now
09:19 -!- jeffdamet1 [n=jeff@dyndsl-091-096-058-183.ewe-ip-backbone.de] has joined #t2
09:23 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has joined #t2
09:33 -!- Baldzius [n=Baldzius@87.198.192.226] has joined #t2
09:33 < Baldzius> moin
09:36 -!- jeffdameth [n=jeff@dyndsl-091-096-058-028.ewe-ip-backbone.de] has quit [Read error: 110 (Connection timed out)]
09:51 < jeffdamet1> it seems alsa has now python as default dependency, but it isnt in alsa-lib.cache and i dont want to build python. in which file do i change the configure options ?
09:52 < jeffdamet1> *isnt set as dependency in
09:53 < jeffdamet1> ok nvm. but mybe one should fix the alsa-lib.cache
10:17 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has quit [Read error: 110 (Connection timed out)]
10:32 < mtr> jeffdamet1: alsa-lib does build fine without python, it does not depend on it
10:33 < mtr> jeffdamet1: if it doesn't work you may even explicitly disable it with --disable-python
10:33 < mtr> just add it to the confopt string
10:34 < jeffdamet1> but it has the explicit --without-python option so somehow it assumed that python was installed. maybe their configure doesnt check for python correctly oO
10:34 < jeffdamet1> mtr: yea i got it working. having trouble with xorg-server now
10:35 < rxr> re
10:36 < mtr> hi rxr
10:36 < rxr> hi mtr jeffdamet1 Baldzius et al.
10:37 < rxr> 7n
10:37 [Users #t2]
10:37 [@ChanServ] [ CIA-8 ] [ Fatal ] [ mtr ] [ TobiX ]
10:37 [ __init__] [ Dallu1 ] [ hwinkel ] [ nelix ] [ valentin]
10:37 [ axionix_] [ dsoul ] [ jeffdamet1] [ Ragnar|away]
10:37 [ Baldzius] [ Enqlave] [ LMJ ] [ rxr ]
10:37 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
10:39 < jeffdamet1> rxr: hi
10:42 < jeffdamet1> i had to add pixman to the xorg deps. how do i make it to check the deps again?
10:45 < mtr> jeffdamet1: the .cache files are not evaluated for cross-builds, they are mainly used to emerge packages into a running system
10:45 < mtr> you have to add pixman to the neo pkgsel file
10:46 < jeffdamet1> mtr: ah ok
10:46 < mtr> its a static package list without automatic dependency resolution
10:46 < rxr> also because most targets developers know what they do and want full control
10:47 < rxr> however we plan to add dynamic resolution (as we have for Emerge-Pkg) also for the target pkgsel
10:47 < rxr> but we'll probably have to start manually (or so) tagging optional dependencies before this is really useful
10:48 < rxr> roadmap is here: http://www.t2-project.org/releases/ and nice to do stuff here: http://www.t2-project.org/developer/soc.html
10:50 < Baldzius> hi rxr
10:55 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has joined #t2
10:55 [Users #t2]
10:55 [@ChanServ] [ CIA-8 ] [ Fatal ] [ mqueiros ] [ rxr ]
10:55 [ __init__] [ Dallu1 ] [ hwinkel ] [ mtr ] [ TobiX ]
10:55 [ axionix_] [ dsoul ] [ jeffdamet1] [ nelix ] [ valentin]
10:55 [ Baldzius] [ Enqlave] [ LMJ ] [ Ragnar|away]
10:55 -!- Irssi: #t2: Total of 19 nicks [1 ops, 0 halfops, 0 voices, 18 normal]
11:08 < jeffdamet1> what could cause ucs2any (or any tool) to not find a file even if it's there?
11:08 < rxr> err
11:09 < rxr> potentially our file log wrapper, but we have it for years and it survives 3000 package builds on 8-core machines, so ...
11:11 < jeffdamet1> this fails /usr/bin/ucs2any 4x6.bdf /usr/X11R7/lib/X11/fonts/util/map-ISO8859-1 ISO8859-1
11:11 < rxr> hm
11:11 < jeffdamet1> but cat ./build/default-8.0-trunk-neo1973-arm-gnueabi-arm920t-cross/usr/X11R7/lib/X11/fonts/util/map-ISO8859-1 exists
11:11 < rxr> maybe thats related to the absoulte path
11:12 < rxr> maybe it's accessing files in your host system?
11:12 < jeffdamet1> i thought the build runs in chroot, doesnt it?
11:13 < jeffdamet1> .. but i also had the feeling, since otherwise alsa-lib shouldnt try to build with python
11:14 < rxr> jeffdamet1: no only native builds run in the chroot
11:14 < jeffdamet1> only strange would it be then that so many other packages build c
11:14 < rxr> which font package is it failing there ?
11:14 < jeffdamet1> font-misc-misc
11:15 < rxr> can you paste the whole svn diff of your current tree somewhere so I can merge fixes?
11:15 < jeffdamet1> sounds as if it is not absolutly needed :p
11:15 < rxr> yeah
11:15 < rxr> if you can paste your binutils and kernel patch fixes I could build and take a look on the font stuff
11:27 < jeffdamet1> yep, http://staff.get-e.org/~jeffdameth/package.patch
11:27 < jeffdamet1> http://staff.get-e.org/~jeffdameth/target-neo1973.patch
11:29 < jeffdamet1> i commented out some kernel patches that didnt apply and didnt looked too important (those with # ?), to see if i get it working at all first
11:30 < jeffdamet1> the binutils patch is wrong i just added the option to make it work.
11:30 < jeffdamet1> this one should be the correct fix (09:15:24 AM) mtr: http://www.busybox.net/lists/uclibc/2006-August/016088.html
11:43 < jeffdamet1> hmm i strongly suspect that that configure sources my root pkgcfg stuff: now libglib is searched for in build/.../usr/lib but was installed to build/.../opt/gnome2/lib/libglib-2.0.so.0
11:46 < jeffdamet1> nup that wasnt it
11:49 < rxr> the auto and libtool stuff is highly fragile
11:49 < rxr> we have quite some automatics to guide and fix it, ...
11:53 < CIA-8> rene * r28007 /trunk/package/xorg/xorg-server/xorg-server.cache:
11:53 < CIA-8> * updated xorg-server.cache from latest reference build, mostly or pixman
11:53 < CIA-8> dependency
11:55 < rxr> jeffdamet1: the xorg-server change related to the embedded text piping and EOT vs EOF marker, was it really needed?
11:56 < rxr> that is did it error out for you ?
11:57 < jeffdamet1> not the EOF, but the other part. without it there was no such file or dir error
11:57 < rxr> hm
11:57 < rxr> interesting
11:57 < CIA-8> rene * r28008 /trunk/package/base/uclibc/patches/binutils.conf: * fixed uclibc builds of binutils to explicitly link against libm
11:58 < rxr> I choose the libm injection
11:58 < rxr> would have to rediff the patch and this way it works even with the next binutils update
11:59 < rxr> I guess it is ok when I commit the neo target update with your name as submitter ?
12:01 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has quit []
12:01 < jeffdamet1> sure :)
12:01 < jeffdamet1> hannes janetzek is my name btw
12:01 < rxr> the one retured from IRC /name ...
12:01 < rxr> ok, matches /name :-)
12:04 < CIA-8> rene * r28009 /trunk/target/neo1973/ (5 files in 2 dirs):
12:04 < CIA-8> Hannes Janetzek:
12:04 < CIA-8> * updated neo1974 target: xorg-server patch, kernel patches and config
12:04 < CIA-8> and added new *pixman dependencies
12:05 < mtr> hmm, now we have -lm for all uclibc builds :( I dont want to be pedantic, but all other archs build without it, just for arm ...
12:05 < rxr> mtr: the other mailing list patch would have added the same
12:05 < rxr> just thru the configure file
12:05 < CIA-8> rene * r28010 /trunk/package/network/snownews/snownews.desc: * updated snownews (1.5.8 -> 1.5.9)
12:06 < rxr> additionally binutils is usually not run on your target system anyway
12:06 < rxr> actually default cross builds could avoid building the native binutils alltogether
12:06 < rxr> btw. I saw this build failture a few times already, but then it never was important enough to spent my workday fixing it :-(
12:07 < jeffdamet1> may i add --prefix=/usr to the configopt or will that install it to my root dir. or what is the env var of the build root?
12:07 < rxr> hm
12:07 < rxr> normally packages should use DESTDIR
12:08 < rxr> but for some packages that do not have DESTDIR we auto-workaround that by configureing with --prefix=${root}/$prefix or so ...
12:08 < rxr> is something installed somewhere you do not like it to be or why do you ask ?
12:08 < rxr> the probably less error prone way would be to set the prefix variable in your target
12:09 < rxr> mtr: in the latest trunk auto regtest most uclibc 1-binutils failed
12:09 < jeffdamet1> ok pango and glib wer installed to /opt/gnome2 and then not found by gtk and other
12:10 < rxr> yes, that is the default due our regular "GNU/Linux desktop builds"
12:10 < rxr> to not clutter /usr that much
12:10 < rxr> that can be controlled even with the scripts/Config
12:10 < rxr> i the expert section GNOME 2.x prefix
12:10 < rxr> and the Neo target could just auto-enforce /usr for that option if we prefer that one the phone
12:10 < rxr> no need to manually inject --prefix= option :-)
12:11 < rxr> mtr: only these built:
12:11 < rxr> avr32-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> blackfin-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> mips-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> mipsel-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> powerpc-glibc.../var/adm/logs/1-binutils.log
12:11 < rxr> powerpc-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> sparc64-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> superh-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> x86-64-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> x86-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> and yes, our 8-core spends many days cross building them all
12:12 < rxr> actually I should finalize a pretty web view of the results so that the utilized energy is worth something :-)
12:12 < rxr> oh, I did not paste the error list? here it is:
12:12 < rxr> alpha-uclibc.../var/adm/logs/1-binutils.err
12:12 < rxr> arm-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> arm-uclibc.../var/adm/logs/1-binutils.err
12:12 < rxr> armeb-uclibc.../var/adm/logs/1-binutils.err
12:12 < rxr> m68k-uclibc.../var/adm/logs/1-binutils.err
12:12 < rxr> mips-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> powerpc-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> powerpc64-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> sparc-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> sparc-uclibc.../var/adm/logs/1-binutils.err
12:12 < rxr> sparc64-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> x86-64-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> x86-dietlibc.../var/adm/logs/1-binutils.err
12:14 < rxr> though not all fail due to -lm
12:14 < rxr> on some some headers are not uclibc friendly (e.g. alpha .-)
12:15 < rxr> sorry for the log spam :-)
12:18 < rxr> Apply patch /mnt/builder/t2-trunk-cross/download/mirror/g/gta01-power_control.patch ...
12:18 < rxr> patching file arch/arm/common/Makefile
12:18 < rxr> Hunk #1 FAILED at 17.
12:20 < nelix> its t2 stable on sgimips/
12:20 < nelix> (i have an o2 and an octane, if anyone needs testing)
12:21 < rxr> you already are on t2/mips or want ot be ?
12:23 < nelix> i might want to be
12:23 < rxr> ah
12:23 < nelix> depending on how likly it is that it will work
12:24 < rxr> well - in theory it is stable - running on access points and such
12:24 < rxr> however on the big, old workstations Linux had some stability issues on my Octane
12:24 < nelix> does anyone run run t2 on sgi o2 or octane?
12:24 < nelix> currently gentoo is running, reasonably
12:24 < nelix> but not perfectly
12:24 < nelix> and gentoo is not so much my first choice
12:24 < rxr> on the o2 only on the older, less thak 10k or so ones due to some MMU speculation whatever exception handling not yet implemented for those CPUs under linux
12:24 < nelix> i am considering netbsd, as they have a sound driver now!
12:25 < rxr> I had t2 on my octane like 2y ago
12:25 < nelix> maybe a project for this weekend :D
12:25 < rxr> but the kernel froze every now and then and eventually I had to much work since then to work on that .-)
12:26 < rxr> if gentoo runs you can at least collect the same patches and just built t2
12:26 < rxr> we also have not yeat an easy installation on those boxes due to the proprietary nature of those beast back then
12:26 < rxr> also when I worked on that last there where not that great boot loaders, I think that changed now
12:26 < rxr> and I know gentoo even has a live cd for those :-(
12:27 < rxr> t2 runs on any mips just fine, as I outlined just for sgi kernel patch and instalation wise it is some patch collection and low level fiddeling to get it on the box
12:29 < nelix> i am trying to follow the mips-linux mailing list
12:29 < nelix> so i can keep up with patchs
12:29 < rxr> :-)
12:29 < CIA-8> rene * r28012 /trunk/package/base/sudo/sudo.desc: * updated sudo (1.6.9p12 -> 1.6.9p13)
12:29 < rxr> well - any sgi patches welcome :-)
12:30 < nelix> maybe i can make a live cd for sgi boxs for t2 one day
12:30 < nelix> i am pretty uncommited though...
12:30 < nelix> i'm trying to learn erlang
12:30 < CIA-8> rene * r28014 /trunk/package/python/setuptools/setuptools.desc: * updated setuptools (0.6c7 -> 0.6c8)
12:31 < CIA-8> rene * r28011 /trunk/package/network/mod_security/mod_security.desc: * updated mod_security (2.1.2 -> 2.1.6)
12:31 < CIA-8> rene * r28013 /trunk/package/develop/squirrel/squirrel.desc: * updated squirrel (2.1.1_stable -> 2.2)
12:31 < CIA-8> rene * r28015 /trunk/package/gnome2/gst-plugins-good/ (compile-fix.patch gst-plugins-good.desc): * updated gst-plugins-good (0.10.6 -> 0.10.7)
12:36 < mtr> rxr: the *-dietlibc cross-builds are not affected by the uclibc binutils build fix
12:37 < rxr> oh - did I list the
12:37 < mtr> therefore the statistic is a bit misleading
12:37 < rxr> sorry, that was not intentionally ..
12:38 < mtr> leaving alpha, arm and m68k as non-building platforms
12:38 < rxr> mtr: I have no problem making the line conditional on arm
12:38 < rxr> or do you rather want the patch?
12:38 < mtr> if its fixed for those, too
12:38 < rxr> but the patch would add it for the other architectures anyway, internall thru the makefile
12:38 < jeffdamet1> rxr: how can i fix this: Created file outside allowed top-level dir: config.cache ?
12:39 < rxr> jeffdamet1: for wich package ?
12:39 < jeffdamet1> happens with base/sysfiles
12:39 < jeffdamet1> gotta eat. brb
12:39 < rxr> mtr: you mostly care about needless linking of binutils binaries against libm ?
12:44 < mtr> I think the patch would be more obvious, the wrapper-fix hides the problem, but never mind, if it fixes it for the other arch's as well its alright for me
12:45 < rxr> mtr: cut'n pasting did not apply and I did not wanted to rediff it for 10m
12:45 < rxr> but the discussion now was longer anyway *sigh*
12:45 < mtr> :)
12:45 < rxr> eithe way is fine for me I just wanted to save a little time :-)
12:45 < rxr> maybe you exchange it with the patch *g* ?
12:48 * mtr took a note for the next uclibc test build ... ;-)
12:52 < rxr> yeah - like my XML notes .-)
12:52 < rxr> so - gotta lunch ... cu
12:52 < mtr> oh yes, the docbook issues ...
12:52 < mtr> cu, enjoy you rmeal
13:02 < jeffdamet1> rxr: the error was caused by having xtrace enabled
13:09 < jeffdamet1> wooot! finished compiling :)
13:09 < jeffdamet1> now lets see how to get it on the device
14:37 < rxr> I suggest transfering the rootfs for a test
14:37 < rxr> or USB / NFS mounting .-)
14:38 < CIA-8> rene * r28016 /trunk/package/mail/libetpan/libetpan.desc: * updated libetpan (0.53 -> 0.54)
14:40 < CIA-8> rene * r28017 /trunk/package/audio/alsa-lib/alsa-lib.conf: * guide latest alsa-lib to not pickup host python while cross compiling
14:59 < rxr> hm, interesting this new xorg-server cross tweak I asked about before (<<-EOT related) now fails here, likewise
14:59 < rxr> but I'll simpliy & clean it up in another way
15:04 < CIA-8> rene * r28018 /trunk/package/xorg/xorg-server/xorg-server.conf: * simplified xorg-server config.cache tweak as it failed with syntax errors
15:05 < rxr> jeffdamet1: how's your t2 impression so far ?
15:08 < jeffdamet1> rxr: yea it's nice to work with. i should have tried a mature target first i think :)
15:08 < rxr> yeah
15:08 < rxr> though the neo target compiled impressively well after just lingering around for some months :-)
15:08 < jeffdamet1> but still definitly nicer than get into bitbake and openembedded :)
15:09 < jeffdamet1> the problem now is to try the kernel. this is a beast of a 4MB image
15:10 < jeffdamet1> i could transfer it into memory, but it wont boot
15:10 < jeffdamet1> will try to boot from sd now but this should only work with some version of uboot and i wont flash the bootloader since it's not my device and i dont have the dev kit
15:11 < rxr> yeah
15:11 < rxr> I have the advanced collecting dust here (I mented I have too much workload?) I could however try and get the t2 built kernel booting
16:08 < CIA-8> rene * r28019 /trunk/package/base/sysfiles/sysfiles.conf:
16:08 < CIA-8> * reverted r27830: set runconf=0 in sysfiles to prevent created file
16:08 < CIA-8> outside basedir, as it could not had effect, it probably just built
16:08 < CIA-8> on the second try anyway
16:25 < CIA-8> rene * r28020 /trunk/scripts/functions.in:
16:25 < CIA-8> * added a note to the "Hook $x is still marked as dirty ..." message
16:25 < CIA-8> to indicate it's forced to run at that point
16:28 < CIA-8> rene * r28021 /trunk/package/xorg/xorg-server/parse-config:
16:28 < CIA-8> * work arounded xorg-server/parse-config to only tweak the config.cache
16:28 < CIA-8> if it exists (as it might be run in a non-build dir after the build,
16:28 < CIA-8> enforced in during the hook check and be located in another system
16:28 < CIA-8> directory, such as the top-level system root during sysfiles
17:14 < CIA-8> rene * r28022 /trunk/package/archiver/bzip2/ (bzip2.conf destdir.patch makefile.patch):
17:14 < CIA-8> * improved bzip2 to install with correct symlinks for alternative
17:14 < CIA-8> program names when cross building (added DESTDIR patch due this)
17:14 < CIA-8> and use install for installing the shared object
17:16 < rxr> just fallout spotted in my neo1973 testbuild :-)
17:16 < rxr> so - back to other work - cu then
17:24 -!- Baldzius [n=Baldzius@87.198.192.226] has quit ["Leaving"]
17:24 < jeffdamet1> rxr: means it didnt work?
17:27 < rxr> look the cia
17:27 < rxr> the sysfiles outside basedir you saw as well for example
17:27 < rxr> and the symlinks of bzip2 where wrong, yes
18:03 < CIA-8> rene * r28023 /trunk/package/archiver/bzip2/makefile.patch:
18:03 < CIA-8> * removed some outdated chmod +x removal from the bzip2/makefile.patch
18:03 < CIA-8> so that those scripts are marked executable, again
19:03 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has quit [Read error: 104 (Connection reset by peer)]
19:03 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has joined #t2
19:42 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has quit [Read error: 110 (Connection timed out)]
20:08 < CIA-8> rene * r28024 /trunk/target/neo1973/config.in:
20:08 < CIA-8> * set the gnome(gtk+) prefix to just /usr (instead the
20:08 < CIA-8> /opt/gnome2 default) for the neo1973 target
20:17 -!- hwinkel [n=hwinkel@p54B6967A.dip0.t-ipconnect.de] has joined #t2
21:19 -!- jeffdameth [n=jeff@dyndsl-091-096-057-066.ewe-ip-backbone.de] has joined #t2
21:24 [Users #t2]
21:24 [@ChanServ] [ CIA-8 ] [ Enqlave] [ jeffdamet1] [ mtr ] [ rxr ]
21:24 [ __init__] [ Dallu1] [ Fatal ] [ jeffdameth] [ nelix ] [ TobiX ]
21:24 [ axionix_] [ dsoul ] [ hwinkel] [ LMJ ] [ Ragnar|away] [ valentin]
21:24 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
21:34 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has joined #t2
21:35 -!- jeffdamet1 [n=jeff@dyndsl-091-096-058-183.ewe-ip-backbone.de] has quit [Read error: 110 (Connection timed out)]
22:17 -!- hwinkel [n=hwinkel@p54B6967A.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)]
22:18 -!- hwinkel [n=hwinkel@p54B6967A.dip0.t-ipconnect.de] has joined #t2
22:55 -!- shevy [n=shevy@chello080108103172.36.11.univie.teleweb.at] has joined #t2
23:16 -!- hwinkel [n=hwinkel@p54B6967A.dip0.t-ipconnect.de] has left #t2 []
--- Log closed Fri Feb 22 00:00:06 2008
01:40 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has quit [Read error: 110 (Connection timed out)]
04:12 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has joined #t2
06:07 -!- emte [n=emte@66.183.8.167] has joined #T2
07:12 -!- emte [n=emte@66.183.8.167] has quit [Read error: 104 (Connection reset by peer)]
07:31 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: Enqlave
07:32 -!- Netsplit over, joins: Enqlave
08:25 -!- jeffdameth [n=jeff@dyndsl-091-096-058-028.ewe-ip-backbone.de] has joined #t2
08:25 < jeffdameth> rxr: i got the kernel to build. now it fails at binutils with no more errormessages than
08:25 < jeffdameth> make[1]: Leaving directory `/home/jeff/src/t2-trunk/src.binutils.default.20080221.081953.9241.rappel/binutils-2.18.50.0.3/objdir'
08:25 < jeffdameth> make: *** [all] Error 2
08:27 < jeffdameth> should i paste my linux26.desc somewhere, so you could try?
08:37 < mtr> hi jeffdameth
08:37 < jeffdameth> hello
08:38 < mtr> pasting the binutils error log would help, in src.binutils.default.*/ERROR.LOG
08:38 < mtr> you changed the linux26.desc?
08:40 < jeffdameth> yep. openmoko changed the patches. or at least one would have to use the old branch and not trunk to fetch the patches
08:40 < jeffdameth> i've now updated the desc from 2.6.21 to the .24 branch
08:41 < mtr> ah, you build the neo target
08:43 < mtr> but I doubt that the updated kernel is related to the binutils failure
08:44 < jeffdameth> ok i guess this is the cause floatformat.c:(.text+0x2ec): undefined reference to `frexp'
08:48 < jeffdameth> hmm floating point stuff. maybe it needs a wrapper or something since arm has no floating point afaik
08:51 < mtr> does it fail in stage 0 or in the cross-build stage 1?
08:52 < mtr> the former would indicate a problem with the host system, maybe missing libraries or such
08:53 < mtr> please paste or publish the error log, maybe there are some more hints in the logfile
08:56 < jeffdameth> http://staff.get-e.org/~jeffdameth/ERROR-LOG
08:58 < jeffdameth> how to figure out the stage? does =[1]=> mean stage one ?
09:01 < mtr> yes
09:05 < mtr> I wonder if the uClibc libm for arm has frexp ...
09:09 < jeffdameth> i found a hint here http://www.mail-archive.com/t2@t2-project.org/msg00488.html
09:09 < jeffdameth> but how to 'use the T2 compiler wraper to inject "-lm": ' isnt explained there
09:11 < mtr> yep ,seems that the arm binutils gas needs to link explicit with -lm
09:15 < mtr> injecting the -lm with the SDE wrapper is one option, but maybe we should add the patch from
09:15 < mtr> http://www.busybox.net/lists/uclibc/2006-August/016088.html
09:17 < jeffdameth> ok thanks. got it working with the injection now
09:19 -!- jeffdamet1 [n=jeff@dyndsl-091-096-058-183.ewe-ip-backbone.de] has joined #t2
09:23 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has joined #t2
09:33 -!- Baldzius [n=Baldzius@87.198.192.226] has joined #t2
09:33 < Baldzius> moin
09:36 -!- jeffdameth [n=jeff@dyndsl-091-096-058-028.ewe-ip-backbone.de] has quit [Read error: 110 (Connection timed out)]
09:51 < jeffdamet1> it seems alsa has now python as default dependency, but it isnt in alsa-lib.cache and i dont want to build python. in which file do i change the configure options ?
09:52 < jeffdamet1> *isnt set as dependency in
09:53 < jeffdamet1> ok nvm. but mybe one should fix the alsa-lib.cache
10:17 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has quit [Read error: 110 (Connection timed out)]
10:32 < mtr> jeffdamet1: alsa-lib does build fine without python, it does not depend on it
10:33 < mtr> jeffdamet1: if it doesn't work you may even explicitly disable it with --disable-python
10:33 < mtr> just add it to the confopt string
10:34 < jeffdamet1> but it has the explicit --without-python option so somehow it assumed that python was installed. maybe their configure doesnt check for python correctly oO
10:34 < jeffdamet1> mtr: yea i got it working. having trouble with xorg-server now
10:35 < rxr> re
10:36 < mtr> hi rxr
10:36 < rxr> hi mtr jeffdamet1 Baldzius et al.
10:37 < rxr> 7n
10:37 [Users #t2]
10:37 [@ChanServ] [ CIA-8 ] [ Fatal ] [ mtr ] [ TobiX ]
10:37 [ __init__] [ Dallu1 ] [ hwinkel ] [ nelix ] [ valentin]
10:37 [ axionix_] [ dsoul ] [ jeffdamet1] [ Ragnar|away]
10:37 [ Baldzius] [ Enqlave] [ LMJ ] [ rxr ]
10:37 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
10:39 < jeffdamet1> rxr: hi
10:42 < jeffdamet1> i had to add pixman to the xorg deps. how do i make it to check the deps again?
10:45 < mtr> jeffdamet1: the .cache files are not evaluated for cross-builds, they are mainly used to emerge packages into a running system
10:45 < mtr> you have to add pixman to the neo pkgsel file
10:46 < jeffdamet1> mtr: ah ok
10:46 < mtr> its a static package list without automatic dependency resolution
10:46 < rxr> also because most targets developers know what they do and want full control
10:47 < rxr> however we plan to add dynamic resolution (as we have for Emerge-Pkg) also for the target pkgsel
10:47 < rxr> but we'll probably have to start manually (or so) tagging optional dependencies before this is really useful
10:48 < rxr> roadmap is here: http://www.t2-project.org/releases/ and nice to do stuff here: http://www.t2-project.org/developer/soc.html
10:50 < Baldzius> hi rxr
10:55 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has joined #t2
10:55 [Users #t2]
10:55 [@ChanServ] [ CIA-8 ] [ Fatal ] [ mqueiros ] [ rxr ]
10:55 [ __init__] [ Dallu1 ] [ hwinkel ] [ mtr ] [ TobiX ]
10:55 [ axionix_] [ dsoul ] [ jeffdamet1] [ nelix ] [ valentin]
10:55 [ Baldzius] [ Enqlave] [ LMJ ] [ Ragnar|away]
10:55 -!- Irssi: #t2: Total of 19 nicks [1 ops, 0 halfops, 0 voices, 18 normal]
11:08 < jeffdamet1> what could cause ucs2any (or any tool) to not find a file even if it's there?
11:08 < rxr> err
11:09 < rxr> potentially our file log wrapper, but we have it for years and it survives 3000 package builds on 8-core machines, so ...
11:11 < jeffdamet1> this fails /usr/bin/ucs2any 4x6.bdf /usr/X11R7/lib/X11/fonts/util/map-ISO8859-1 ISO8859-1
11:11 < rxr> hm
11:11 < jeffdamet1> but cat ./build/default-8.0-trunk-neo1973-arm-gnueabi-arm920t-cross/usr/X11R7/lib/X11/fonts/util/map-ISO8859-1 exists
11:11 < rxr> maybe thats related to the absoulte path
11:12 < rxr> maybe it's accessing files in your host system?
11:12 < jeffdamet1> i thought the build runs in chroot, doesnt it?
11:13 < jeffdamet1> .. but i also had the feeling, since otherwise alsa-lib shouldnt try to build with python
11:14 < rxr> jeffdamet1: no only native builds run in the chroot
11:14 < jeffdamet1> only strange would it be then that so many other packages build c
11:14 < rxr> which font package is it failing there ?
11:14 < jeffdamet1> font-misc-misc
11:15 < rxr> can you paste the whole svn diff of your current tree somewhere so I can merge fixes?
11:15 < jeffdamet1> sounds as if it is not absolutly needed :p
11:15 < rxr> yeah
11:15 < rxr> if you can paste your binutils and kernel patch fixes I could build and take a look on the font stuff
11:27 < jeffdamet1> yep, http://staff.get-e.org/~jeffdameth/package.patch
11:27 < jeffdamet1> http://staff.get-e.org/~jeffdameth/target-neo1973.patch
11:29 < jeffdamet1> i commented out some kernel patches that didnt apply and didnt looked too important (those with # ?), to see if i get it working at all first
11:30 < jeffdamet1> the binutils patch is wrong i just added the option to make it work.
11:30 < jeffdamet1> this one should be the correct fix (09:15:24 AM) mtr: http://www.busybox.net/lists/uclibc/2006-August/016088.html
11:43 < jeffdamet1> hmm i strongly suspect that that configure sources my root pkgcfg stuff: now libglib is searched for in build/.../usr/lib but was installed to build/.../opt/gnome2/lib/libglib-2.0.so.0
11:46 < jeffdamet1> nup that wasnt it
11:49 < rxr> the auto and libtool stuff is highly fragile
11:49 < rxr> we have quite some automatics to guide and fix it, ...
11:53 < CIA-8> rene * r28007 /trunk/package/xorg/xorg-server/xorg-server.cache:
11:53 < CIA-8> * updated xorg-server.cache from latest reference build, mostly or pixman
11:53 < CIA-8> dependency
11:55 < rxr> jeffdamet1: the xorg-server change related to the embedded text piping and EOT vs EOF marker, was it really needed?
11:56 < rxr> that is did it error out for you ?
11:57 < jeffdamet1> not the EOF, but the other part. without it there was no such file or dir error
11:57 < rxr> hm
11:57 < rxr> interesting
11:57 < CIA-8> rene * r28008 /trunk/package/base/uclibc/patches/binutils.conf: * fixed uclibc builds of binutils to explicitly link against libm
11:58 < rxr> I choose the libm injection
11:58 < rxr> would have to rediff the patch and this way it works even with the next binutils update
11:59 < rxr> I guess it is ok when I commit the neo target update with your name as submitter ?
12:01 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has quit []
12:01 < jeffdamet1> sure :)
12:01 < jeffdamet1> hannes janetzek is my name btw
12:01 < rxr> the one retured from IRC /name ...
12:01 < rxr> ok, matches /name :-)
12:04 < CIA-8> rene * r28009 /trunk/target/neo1973/ (5 files in 2 dirs):
12:04 < CIA-8> Hannes Janetzek:
12:04 < CIA-8> * updated neo1974 target: xorg-server patch, kernel patches and config
12:04 < CIA-8> and added new *pixman dependencies
12:05 < mtr> hmm, now we have -lm for all uclibc builds :( I dont want to be pedantic, but all other archs build without it, just for arm ...
12:05 < rxr> mtr: the other mailing list patch would have added the same
12:05 < rxr> just thru the configure file
12:05 < CIA-8> rene * r28010 /trunk/package/network/snownews/snownews.desc: * updated snownews (1.5.8 -> 1.5.9)
12:06 < rxr> additionally binutils is usually not run on your target system anyway
12:06 < rxr> actually default cross builds could avoid building the native binutils alltogether
12:06 < rxr> btw. I saw this build failture a few times already, but then it never was important enough to spent my workday fixing it :-(
12:07 < jeffdamet1> may i add --prefix=/usr to the configopt or will that install it to my root dir. or what is the env var of the build root?
12:07 < rxr> hm
12:07 < rxr> normally packages should use DESTDIR
12:08 < rxr> but for some packages that do not have DESTDIR we auto-workaround that by configureing with --prefix=${root}/$prefix or so ...
12:08 < rxr> is something installed somewhere you do not like it to be or why do you ask ?
12:08 < rxr> the probably less error prone way would be to set the prefix variable in your target
12:09 < rxr> mtr: in the latest trunk auto regtest most uclibc 1-binutils failed
12:09 < jeffdamet1> ok pango and glib wer installed to /opt/gnome2 and then not found by gtk and other
12:10 < rxr> yes, that is the default due our regular "GNU/Linux desktop builds"
12:10 < rxr> to not clutter /usr that much
12:10 < rxr> that can be controlled even with the scripts/Config
12:10 < rxr> i the expert section GNOME 2.x prefix
12:10 < rxr> and the Neo target could just auto-enforce /usr for that option if we prefer that one the phone
12:10 < rxr> no need to manually inject --prefix= option :-)
12:11 < rxr> mtr: only these built:
12:11 < rxr> avr32-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> blackfin-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> mips-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> mipsel-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> powerpc-glibc.../var/adm/logs/1-binutils.log
12:11 < rxr> powerpc-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> sparc64-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> superh-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> x86-64-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> x86-uclibc.../var/adm/logs/1-binutils.log
12:11 < rxr> and yes, our 8-core spends many days cross building them all
12:12 < rxr> actually I should finalize a pretty web view of the results so that the utilized energy is worth something :-)
12:12 < rxr> oh, I did not paste the error list? here it is:
12:12 < rxr> alpha-uclibc.../var/adm/logs/1-binutils.err
12:12 < rxr> arm-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> arm-uclibc.../var/adm/logs/1-binutils.err
12:12 < rxr> armeb-uclibc.../var/adm/logs/1-binutils.err
12:12 < rxr> m68k-uclibc.../var/adm/logs/1-binutils.err
12:12 < rxr> mips-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> powerpc-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> powerpc64-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> sparc-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> sparc-uclibc.../var/adm/logs/1-binutils.err
12:12 < rxr> sparc64-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> x86-64-dietlibc.../var/adm/logs/1-binutils.err
12:12 < rxr> x86-dietlibc.../var/adm/logs/1-binutils.err
12:14 < rxr> though not all fail due to -lm
12:14 < rxr> on some some headers are not uclibc friendly (e.g. alpha .-)
12:15 < rxr> sorry for the log spam :-)
12:18 < rxr> Apply patch /mnt/builder/t2-trunk-cross/download/mirror/g/gta01-power_control.patch ...
12:18 < rxr> patching file arch/arm/common/Makefile
12:18 < rxr> Hunk #1 FAILED at 17.
12:20 < nelix> its t2 stable on sgimips/
12:20 < nelix> (i have an o2 and an octane, if anyone needs testing)
12:21 < rxr> you already are on t2/mips or want ot be ?
12:23 < nelix> i might want to be
12:23 < rxr> ah
12:23 < nelix> depending on how likly it is that it will work
12:24 < rxr> well - in theory it is stable - running on access points and such
12:24 < rxr> however on the big, old workstations Linux had some stability issues on my Octane
12:24 < nelix> does anyone run run t2 on sgi o2 or octane?
12:24 < nelix> currently gentoo is running, reasonably
12:24 < nelix> but not perfectly
12:24 < nelix> and gentoo is not so much my first choice
12:24 < rxr> on the o2 only on the older, less thak 10k or so ones due to some MMU speculation whatever exception handling not yet implemented for those CPUs under linux
12:24 < nelix> i am considering netbsd, as they have a sound driver now!
12:25 < rxr> I had t2 on my octane like 2y ago
12:25 < nelix> maybe a project for this weekend :D
12:25 < rxr> but the kernel froze every now and then and eventually I had to much work since then to work on that .-)
12:26 < rxr> if gentoo runs you can at least collect the same patches and just built t2
12:26 < rxr> we also have not yeat an easy installation on those boxes due to the proprietary nature of those beast back then
12:26 < rxr> also when I worked on that last there where not that great boot loaders, I think that changed now
12:26 < rxr> and I know gentoo even has a live cd for those :-(
12:27 < rxr> t2 runs on any mips just fine, as I outlined just for sgi kernel patch and instalation wise it is some patch collection and low level fiddeling to get it on the box
12:29 < nelix> i am trying to follow the mips-linux mailing list
12:29 < nelix> so i can keep up with patchs
12:29 < rxr> :-)
12:29 < CIA-8> rene * r28012 /trunk/package/base/sudo/sudo.desc: * updated sudo (1.6.9p12 -> 1.6.9p13)
12:29 < rxr> well - any sgi patches welcome :-)
12:30 < nelix> maybe i can make a live cd for sgi boxs for t2 one day
12:30 < nelix> i am pretty uncommited though...
12:30 < nelix> i'm trying to learn erlang
12:30 < CIA-8> rene * r28014 /trunk/package/python/setuptools/setuptools.desc: * updated setuptools (0.6c7 -> 0.6c8)
12:31 < CIA-8> rene * r28011 /trunk/package/network/mod_security/mod_security.desc: * updated mod_security (2.1.2 -> 2.1.6)
12:31 < CIA-8> rene * r28013 /trunk/package/develop/squirrel/squirrel.desc: * updated squirrel (2.1.1_stable -> 2.2)
12:31 < CIA-8> rene * r28015 /trunk/package/gnome2/gst-plugins-good/ (compile-fix.patch gst-plugins-good.desc): * updated gst-plugins-good (0.10.6 -> 0.10.7)
12:36 < mtr> rxr: the *-dietlibc cross-builds are not affected by the uclibc binutils build fix
12:37 < rxr> oh - did I list the
12:37 < mtr> therefore the statistic is a bit misleading
12:37 < rxr> sorry, that was not intentionally ..
12:38 < mtr> leaving alpha, arm and m68k as non-building platforms
12:38 < rxr> mtr: I have no problem making the line conditional on arm
12:38 < rxr> or do you rather want the patch?
12:38 < mtr> if its fixed for those, too
12:38 < rxr> but the patch would add it for the other architectures anyway, internall thru the makefile
12:38 < jeffdamet1> rxr: how can i fix this: Created file outside allowed top-level dir: config.cache ?
12:39 < rxr> jeffdamet1: for wich package ?
12:39 < jeffdamet1> happens with base/sysfiles
12:39 < jeffdamet1> gotta eat. brb
12:39 < rxr> mtr: you mostly care about needless linking of binutils binaries against libm ?
12:44 < mtr> I think the patch would be more obvious, the wrapper-fix hides the problem, but never mind, if it fixes it for the other arch's as well its alright for me
12:45 < rxr> mtr: cut'n pasting did not apply and I did not wanted to rediff it for 10m
12:45 < rxr> but the discussion now was longer anyway *sigh*
12:45 < mtr> :)
12:45 < rxr> eithe way is fine for me I just wanted to save a little time :-)
12:45 < rxr> maybe you exchange it with the patch *g* ?
12:48 * mtr took a note for the next uclibc test build ... ;-)
12:52 < rxr> yeah - like my XML notes .-)
12:52 < rxr> so - gotta lunch ... cu
12:52 < mtr> oh yes, the docbook issues ...
12:52 < mtr> cu, enjoy you rmeal
13:02 < jeffdamet1> rxr: the error was caused by having xtrace enabled
13:09 < jeffdamet1> wooot! finished compiling :)
13:09 < jeffdamet1> now lets see how to get it on the device
14:37 < rxr> I suggest transfering the rootfs for a test
14:37 < rxr> or USB / NFS mounting .-)
14:38 < CIA-8> rene * r28016 /trunk/package/mail/libetpan/libetpan.desc: * updated libetpan (0.53 -> 0.54)
14:40 < CIA-8> rene * r28017 /trunk/package/audio/alsa-lib/alsa-lib.conf: * guide latest alsa-lib to not pickup host python while cross compiling
14:59 < rxr> hm, interesting this new xorg-server cross tweak I asked about before (<<-EOT related) now fails here, likewise
14:59 < rxr> but I'll simpliy & clean it up in another way
15:04 < CIA-8> rene * r28018 /trunk/package/xorg/xorg-server/xorg-server.conf: * simplified xorg-server config.cache tweak as it failed with syntax errors
15:05 < rxr> jeffdamet1: how's your t2 impression so far ?
15:08 < jeffdamet1> rxr: yea it's nice to work with. i should have tried a mature target first i think :)
15:08 < rxr> yeah
15:08 < rxr> though the neo target compiled impressively well after just lingering around for some months :-)
15:08 < jeffdamet1> but still definitly nicer than get into bitbake and openembedded :)
15:09 < jeffdamet1> the problem now is to try the kernel. this is a beast of a 4MB image
15:10 < jeffdamet1> i could transfer it into memory, but it wont boot
15:10 < jeffdamet1> will try to boot from sd now but this should only work with some version of uboot and i wont flash the bootloader since it's not my device and i dont have the dev kit
15:11 < rxr> yeah
15:11 < rxr> I have the advanced collecting dust here (I mented I have too much workload?) I could however try and get the t2 built kernel booting
16:08 < CIA-8> rene * r28019 /trunk/package/base/sysfiles/sysfiles.conf:
16:08 < CIA-8> * reverted r27830: set runconf=0 in sysfiles to prevent created file
16:08 < CIA-8> outside basedir, as it could not had effect, it probably just built
16:08 < CIA-8> on the second try anyway
16:25 < CIA-8> rene * r28020 /trunk/scripts/functions.in:
16:25 < CIA-8> * added a note to the "Hook $x is still marked as dirty ..." message
16:25 < CIA-8> to indicate it's forced to run at that point
16:28 < CIA-8> rene * r28021 /trunk/package/xorg/xorg-server/parse-config:
16:28 < CIA-8> * work arounded xorg-server/parse-config to only tweak the config.cache
16:28 < CIA-8> if it exists (as it might be run in a non-build dir after the build,
16:28 < CIA-8> enforced in during the hook check and be located in another system
16:28 < CIA-8> directory, such as the top-level system root during sysfiles
17:14 < CIA-8> rene * r28022 /trunk/package/archiver/bzip2/ (bzip2.conf destdir.patch makefile.patch):
17:14 < CIA-8> * improved bzip2 to install with correct symlinks for alternative
17:14 < CIA-8> program names when cross building (added DESTDIR patch due this)
17:14 < CIA-8> and use install for installing the shared object
17:16 < rxr> just fallout spotted in my neo1973 testbuild :-)
17:16 < rxr> so - back to other work - cu then
17:24 -!- Baldzius [n=Baldzius@87.198.192.226] has quit ["Leaving"]
17:24 < jeffdamet1> rxr: means it didnt work?
17:27 < rxr> look the cia
17:27 < rxr> the sysfiles outside basedir you saw as well for example
17:27 < rxr> and the symlinks of bzip2 where wrong, yes
18:03 < CIA-8> rene * r28023 /trunk/package/archiver/bzip2/makefile.patch:
18:03 < CIA-8> * removed some outdated chmod +x removal from the bzip2/makefile.patch
18:03 < CIA-8> so that those scripts are marked executable, again
19:03 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has quit [Read error: 104 (Connection reset by peer)]
19:03 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has joined #t2
19:42 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has quit [Read error: 110 (Connection timed out)]
20:08 < CIA-8> rene * r28024 /trunk/target/neo1973/config.in:
20:08 < CIA-8> * set the gnome(gtk+) prefix to just /usr (instead the
20:08 < CIA-8> /opt/gnome2 default) for the neo1973 target
20:17 -!- hwinkel [n=hwinkel@p54B6967A.dip0.t-ipconnect.de] has joined #t2
21:19 -!- jeffdameth [n=jeff@dyndsl-091-096-057-066.ewe-ip-backbone.de] has joined #t2
21:24 [Users #t2]
21:24 [@ChanServ] [ CIA-8 ] [ Enqlave] [ jeffdamet1] [ mtr ] [ rxr ]
21:24 [ __init__] [ Dallu1] [ Fatal ] [ jeffdameth] [ nelix ] [ TobiX ]
21:24 [ axionix_] [ dsoul ] [ hwinkel] [ LMJ ] [ Ragnar|away] [ valentin]
21:24 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
21:34 -!- mqueiros [n=mqueiros@c-217-70-68-61.bragatel.pt] has joined #t2
21:35 -!- jeffdamet1 [n=jeff@dyndsl-091-096-058-183.ewe-ip-backbone.de] has quit [Read error: 110 (Connection timed out)]
22:17 -!- hwinkel [n=hwinkel@p54B6967A.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)]
22:18 -!- hwinkel [n=hwinkel@p54B6967A.dip0.t-ipconnect.de] has joined #t2
22:55 -!- shevy [n=shevy@chello080108103172.36.11.univie.teleweb.at] has joined #t2
23:16 -!- hwinkel [n=hwinkel@p54B6967A.dip0.t-ipconnect.de] has left #t2 []
--- Log closed Fri Feb 22 00:00:06 2008