T2 IRC Log: 2005-02-03
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 03 00:00:48 2005
00:02 < CIA-8> rene * r5945 /trunk/package/editors/elvis-tiny/elvis-tiny.desc:
00:02 < CIA-8> * marked elvis-tiny to build in stage 5 (not 1+3) - I shuffle the
00:02 < CIA-8> build stages in the psion-pda target now as I wish
00:15 < rxr> -rwxr-xr-x 1 root root 252356 Feb 3 00:14 rsync
00:15 < rxr> ^- arm dietlibc cross ;-)
00:17 < mnemoc> rsync??!
00:19 < rxr> yeah - thought it might be a nice option to sync my notes over the IRDA port ...
00:19 < rxr> but it looks far too huge ...
00:19 < rxr> but at least it is dietlibc able now ...
00:20 < _Ragnar_> static link?
00:21 < rxr> yep
00:21 < _Ragnar_> ah okay
00:21 < mnemoc> .oO( dietlibc use-it package selection will get loooooonger )o
00:21 < mnemoc> rxr: is it too expensive to have it dynamic?
00:21 < rxr> mnemoc: yep - why .oO ?
00:22 < rxr> when we (at ROCK times) adapted to diet it was static only
00:22 < rxr> only later x86 dynamic support was added
00:22 < rxr> then arm and now x86-64
00:22 < mnemoc> on Config we select one by one which files do we want to build againts diet
00:22 < rxr> I think the other arches still have no dynamic support
00:22 < mnemoc> s/files/packages/
00:22 < rxr> yep - I know
00:22 < rxr> why .oO?
00:23 < mnemoc> the list is long
00:23 < mnemoc> and you are improving it to be even longer
00:23 < rxr> yep ;-)
00:23 < rxr> dietlibc is soo cute - lowfat and clean ;-)
00:23 < CIA-8> rene * r5946 /trunk/package/ (2 files in 2 dirs):
00:23 < CIA-8> * added patch to support building rsync with dietlibc - and marked
00:23 < CIA-8> it DIETLIBC ready
00:24 < rxr> I want dietlibc only systems soonder than later ..
00:24 < rxr> *g*
00:24 < mnemoc> rxr: that's why i prefer pure systems, LIBC=dietlibc not mixed s.th
00:24 < rxr> when I claned the dietlibc stuff (read reimplemented to be less hacky and ugly)
00:24 < _Ragnar_> what keeps you?
00:24 < rxr> lack of packages building with it ;-)
00:25 < rxr> when I reimplemented the diet stuff more packages should build without hacking - maybe we would then rather define tested uclibc and dietlibc selection templates - in contrast to adding FLAGS over all the pacag
00:25 < rxr> es
00:25 < mnemoc> rxr: to be static is just an option with can be forced for some archs ;)
00:26 < rxr> sure
00:26 < rxr> there already is that option in the config ...
00:26 < _Ragnar_> it doesn't work for some packages tho
00:26 < mnemoc> rxr: _Ragnar_ is trying to build everything against uclibc and it seems to work
00:26 < mnemoc> :(
00:26 < _Ragnar_> I'm not building static
00:26 < mnemoc> i have to go
00:27 < mnemoc> i know
00:27 < rxr> cu mnemoc
00:27 < _Ragnar_> awww kay
00:27 < rxr> ucibc is full of compat cut'n paste
00:27 < rxr> although neither POSIX nor any other standard
00:27 < rxr> such stuff is not in dietlibc - and if it is it is #ifdef'ed _GNU_SOURCE _BSD_SOURCE
00:27 < rxr> thus the dietlibc/parse-config is full of
00:27 < rxr> nvi)
00:28 < rxr> var_append GCC_OPTIONS_APPEND " " "-D_GNU_SOURCE"
00:28 < rxr> and co to get those packags build
00:28 < rxr> this is why more stuff just builds with uclibc and needs at least some define injections with dietlibc
00:29 < rxr> also C++ stuff might not yet fully work with dietlibc - so dietlibc only systems will be some bigger task not finished tomorrow ...
00:30 < rxr> _Ragnar_: what is the uclibc target triplet?
00:30 < rxr> i386-linux-t2-uclibc or so ?
00:30 < _Ragnar_> ix86-t2-linux-uclibc
00:30 < rxr> ix86 ?
00:31 < _Ragnar_> i686 in my case
00:31 < rxr> ah - ok ;-)
00:43 < rxr> UPN Officially Cancels 'Star Trek: Enterprise' ???
00:43 < rxr> didn't the announced last year that they have signed the full 8 seasons ...
00:56 < rxr> my PDA target has this buildtime:
00:56 < rxr> real 20m39.704s
00:56 < rxr> user 14m7.435s
00:56 < rxr> sys 4m4.398s
00:56 < rxr> 17 builds total, 17 completed fine, 0 with errors.
00:56 < rxr> on a 2.5Ghz Athlon ;-)
00:56 < rxr> n8 all
00:57 < _Ragnar_> gn8
02:27 < mnemoc> re
02:28 < _Ragnar_> re
02:29 < _Ragnar_> hmm does newer gcc/g++ have a different exception handling?
02:29 < mnemoc> no idea
02:29 < mnemoc> why?
02:29 * _Ragnar_ has a problem with cross-compiling an app with mingw
02:30 < _Ragnar_> my exceptions don't get caught at all
02:37 < mnemoc> o_O
04:35 -!- rxr_ [~rene@p213.54.201.251.tisdip.tiscali.de] has joined #t2
04:35 -!- Topic for #t2: T2 | 2.1.0-beta2 RELEASED | The next generation of System Development Enviroments (SDE) | http://www.t2-project.org/ | CIA, don't ever leave us again!
04:35 -!- Topic set by mnemoc [] [Wed Jan 12 03:04:43 2005]
04:35 [Users #t2]
04:35 [ _Ragnar_] [ daja77] [ mnemoc] [ nzg ] [ praenti] [ rxr_ ]
04:35 [ CIA-8 ] [ jsaw ] [ mtr ] [ ooo_] [ rxr ] [ valentin]
04:35 -!- Irssi: #t2: Total of 12 nicks [0 ops, 0 halfops, 0 voices, 12 normal]
04:35 -!- Channel #t2 created Sun Aug 8 21:15:33 2004
04:35 -!- rxr [~rene@p213.54.197.108.tisdip.tiscali.de] has quit [Read error: 104 (Connection reset by peer)]
04:35 -!- [freenode-info] why register and identify? your IRC nick is how people know you. http://freenode.net/faq.shtml#nicksetup
04:35 -!- Irssi: Join to #t2 was synced in 10 secs
04:43 -!- mtr_ [~michael@Ha22a.h.pppool.de] has joined #t2
04:44 -!- ideal [~idealm@211.100.227.107] has joined #t2
04:52 -!- mtr [~michael@H8616.h.pppool.de] has quit [Read error: 60 (Operation timed out)]
05:36 -!- ideal [~idealm@211.100.227.107] has quit [Client Quit]
05:49 -!- ideal [~idealm@211.100.227.107] has joined #t2
07:17 -!- mnemoc_ [~amery@200.75.27.43] has joined #t2
07:17 -!- mnemoc [~amery@200.75.27.32] has quit [Read error: 60 (Operation timed out)]
08:07 -!- Netsplit over, joins: praenti
08:07 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: valentin, daja77, mnemoc_, _Ragnar_, nzg, ideal
08:08 -!- mnemoc [~amery@200.75.27.43] has joined #t2
08:09 -!- _Ragnar_ [loki@66-146-166-62.skyriver.net] has joined #t2
08:09 -!- daja77 [[68kg0qZ1V@odoaker.hrz.tu-chemnitz.de] has joined #t2
08:12 -!- ideal [~idealm@211.100.227.107] has joined #t2
08:13 -!- valentin [~valentin@port-212-202-42-228.dynamic.qsc.de] has joined #t2
08:13 -!- nzg [~tschmidt@xchangecenter.de] has joined #t2
08:15 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: daja77, _Ragnar_
08:16 -!- Netsplit over, joins: _Ragnar_, daja77
08:17 < rxr_> moin
08:17 -!- You're now known as rxr
08:33 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: mnemoc
08:33 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: valentin, nzg, ideal
08:36 -!- Netsplit over, joins: nzg
08:36 -!- Netsplit over, joins: mnemoc
08:36 -!- Netsplit over, joins: valentin
08:42 -!- madtux [~mike@65.182.14.165] has joined #t2
08:42 < madtux> re
08:42 < rxr> hi madtux
08:43 < madtux> guten morgen mein freund :)
08:44 < rxr> good morning over there ;-)
08:45 < madtux> well its just 1:45am here.. but thanks :)
08:48 < _Ragnar_> hi
08:50 -!- ideal [~idealm@211.100.227.107] has joined #t2
08:50 < rxr> madtux: hehe - sorry I'm so sleepy and writing some article stuff that I forgot about that ...
08:50 < rxr> moin _Ragnar_
08:50 < rxr> welcome ideal
08:51 < _Ragnar_> ;)
08:51 < _Ragnar_> moin
08:52 < madtux> helo _Ragnar_ and ideal
08:52 < madtux> rxr, its ok.. i'm currently downloading some packages..
08:52 < madtux> just got 2 new servers which are not yet going live for production
08:52 < madtux> so i will use them for building processes
08:56 < _Ragnar_> ;)
08:57 < madtux> 2x 2.8GHZ xeon, 2x 73 15k rpm scsi, 4gb mem
08:57 < madtux> that shall be enough
08:57 < madtux> ah well thats it for tonight i must sleep
08:57 < madtux> night all
08:58 -!- madtux [~mike@65.182.14.165] has quit ["Leaving"]
09:02 -!- mtr_ is now known as mtr
09:02 < mtr> hi
09:04 < mtr> ncurses/cross-build.patch does interfere with a uclibc ncurses patch
09:04 < mtr> !> Apply patch /work/rock/t2-work/package/base/uclibc/patch ..
09:04 < mtr> !> patching file ncurses/Makefile.in
09:04 < mtr> !> Hunk #1 FAILED at 93.
09:04 < _Ragnar_> ;p
09:07 < mtr> hmm, eventually the ncurses cross-build.patch should not be applied in general?
09:08 < _Ragnar_> that's why I tried to confine any changed to the uclibc/patches, instead of putting patches into the packages themselves
09:08 < _Ragnar_> changes*
09:09 < rxr> hm
09:09 * rxr looking what this uclibc patch does. ..
09:09 < rxr> maybe s.th. quite simillar ...
09:10 < rxr> hm - ic
09:11 < rxr> r5661 | mnemoc | 2005-01-19 04:08:31 +0100 (Wed, 19 Jan 2005) | 2 lines
09:11 < rxr> * added uclibc patch and hack to build ncurses
09:11 < rxr> hm
09:13 < rxr> _Ragnar_: mtr can you try if ncurses builds with uclibc when you remove the non applying patch from uclibc ?
09:13 < _Ragnar_> uhm
09:13 < mtr> hmm
09:13 < _Ragnar_> but including your patch?
09:13 < rxr> yes
09:14 < rxr> the one placed in the ncurses package
09:18 < _Ragnar_> both the patch and the .conf fix, or just the patch?
09:22 * _Ragnar_ pokes rxr
09:22 < mtr> defining _GNU_SOURCE and NDEBUG should be safe, and -I../include is set in cross-build.patch, too
09:23 < mtr> oh, wait
09:23 < mtr> == 02/03/05 09:21:24 =[1]=> Aborted building package ncurses
09:23 < mtr> just removing the uclibc patch does not work
09:25 < jsaw> re
09:25 * mtr disabling now the other patch ...
09:25 < jsaw> rxr: 5am??? you were a bit late, ain't you?
09:25 < mtr> hi jsaw
09:25 < jsaw> hi mtr
09:26 < jsaw> hi _Ragnar_
09:26 < _Ragnar_> hi jsaw
09:28 < mtr> _Ragnar_, i have two more packages for uclibc_fix_configsub, do you mind if i commit ?
09:29 < _Ragnar_> mtr: hmmm I was going to remove most of the cases for that and substitute with a general find for configure/config.sub
09:30 < mtr> yes, i read about this, therefore i asked
09:30 < _Ragnar_> which seems to work pretty well
09:31 < _Ragnar_> but I really need to go through the build logs first and see if there are more hidden errors
09:31 < _Ragnar_> before I commit that
09:31 < mtr> so, just waiting for your commit
09:33 < _Ragnar_> okay
09:33 < _Ragnar_> what packages/target are you building
09:33 < _Ragnar_> ?
09:34 < mtr> some packages for my dvb-harddisc-recorder system, like mplayer vdr libdvdnnn and such
09:34 -!- mnemoc [~amery@200.75.27.43] has quit [Read error: 113 (No route to host)]
09:35 < mtr> == 02/03/05 09:33:55 =[1]=> Finished building package ncurses
09:36 < rxr> do you svn rm the ncurses patch and commit?
09:36 < mtr> do you want me to do this?
09:38 < _Ragnar_> which patch?
09:39 < _Ragnar_> imho, the uclibc one shouldn't be removed until it's worked around
09:40 < _Ragnar_> but I need to get to bed
09:41 < _Ragnar_> rxr: does the uclibc ncurses patch fix the general crossbuild case?
09:42 < _Ragnar_> anyway, nini
09:43 < mtr> ok, my uboot target continues to build, afk
09:43 < mtr> gn8 _Ragnar_
09:43 < _Ragnar_> gn8
09:44 < rxr> _Ragnar_: the patch I comitted fixes it in general
09:45 < rxr> I had no idea that uclibc tinker s.th. abroad ...
09:47 < CIA-8> rene * r5947 /trunk/package/base/uclibc/patches/ (ncurses-100-buildccflags.patch ncurses.conf):
09:47 < CIA-8> * removed uclibc ncurses patches (should be fixed by my generic cross
09:47 < CIA-8> build fixes to ncurses yesterday - the .patch removal was tested,
09:47 < CIA-8> however the .conf was not - but I'm confident it will work without)
09:47 -!- mnemoc [~amery@200.75.27.6] has joined #t2
09:48 < rxr> moin mnemoc
10:24 < rxr> Sun vergibt Service-Vertrag für interne Softwarewartung an CSC
10:24 < rxr> ???
11:03 < rxr> hm - my good old U5 still running 32bit userspace from ROCK times just fucked up ..
11:03 < rxr> I was no longer to write to any filesystem
11:03 < rxr> before the U160 IBM SCSI disk made some strange noice
11:03 < rxr> a reboot seems to cured that ..
11:03 * rxr headshake
11:04 < rxr> or not
11:04 < rxr> damn
11:04 < rxr> ah
11:04 < rxr> the usual / 100% filled
11:04 < rxr> damn - why can't programs have more sensible error messages ...
11:07 -!- ideal [~idealm@211.100.227.107] has quit ["Leaving"]
11:12 -!- mtr [~michael@Ha22a.h.pppool.de] has quit [Read error: 60 (Operation timed out)]
11:23 -!- _Ragnar__ [loki@66-146-166-62.skyriver.net] has joined #t2
11:23 -!- _Ragnar_ [loki@66-146-166-62.skyriver.net] has quit [Read error: 104 (Connection reset by peer)]
11:44 < rxr> oh - it seems I got a 1.3 in assembly hackery ... *g*
12:01 < CIA-8> rene * r5948 /trunk/package/x11/rxvt-unicode/rxvt-unicode.desc: * updated rxvt-unicode (4.7 -> 4.9)
12:30 < rxr> /home/rene/myrevo/a 2745 1226 1369 48% /mnt/misc
12:32 < rxr> 758467 Feb 3 12:45 initrd
12:47 < rxr> ok - the ARM binaries work
12:48 * rxr in a ash in this PDA ...
12:48 < rxr> no digicam - so no pics ...
13:28 < CIA-8> rene * r5949 /trunk/package/base/irda-utils/irda-utils.conf: * fixed irda-utils for cross builds
13:33 < CIA-8> rene * r5950 /trunk/package/base/dietlibc/parse-config-9:
13:33 < CIA-8> * fixed the dietlibc arch_target redefine to be -dietlibc instead of
13:33 < CIA-8> the ugly -gnulibc1. No idea how this could have be commited in the
13:33 < CIA-8> first place ... My Psion PDA testcase builds with the new one.
13:35 < CIA-8> rene * r5951 /trunk/target/psion-pda/ (config.in parse-config):
13:35 < CIA-8> * added iproute2 and irda-utils to the psion-pda target, as well
13:35 < CIA-8> as a overwrite of install_init to prevent init scripts from beeing
13:35 < CIA-8> instaslled
13:51 -!- mtr [~michael@H9b1f.h.pppool.de] has joined #t2
14:00 < rxr> dietlibc static vs. dynamic:
14:00 < rxr> -rwxr-xr-x 1 root root 2336 Feb 3 14:00 hello
14:00 < rxr> -rwxr-xr-x 1 root root 7908 Feb 3 14:00 hello-s
14:02 < rxr> vs. glibc:
14:02 < rxr> -rwxr-xr-x 1 root root 2804 Feb 3 14:01 hello-glibc
14:02 < rxr> -rwxr-xr-x 1 root root 315792 Feb 3 14:01 hello-glibc-s
14:02 < rxr> *g*
14:35 < mnemoc> moin!
14:35 < rxr> moin!
14:36 < mnemoc> "fixed the dietlibc arch_target redefine to be -dietlibc" <--- this impact every config.sub file, and binutils mainly
14:42 < rxr> nothing does use dietlibc
14:42 < rxr> and I see now regression
14:43 < rxr> using one totally messed name with some normal official name should not screw anything up ...
14:43 < mnemoc> let see how does it go here
14:43 < rxr> mnemoc: can you remove the dietlibc/sincostan.patch before you start the build
14:43 < mnemoc> ok
14:43 < rxr> it does not work on x86 and I need to take a look how to only inject it for non x86 builds ...
14:43 < rxr> well - ok I have some idea .. but - test without the patch first ...
14:44 < mnemoc> do you want me to test a pure minimal/dietlibc or a bootdisk?
14:44 < rxr> bootdisk
14:44 < rxr> pure minimal is highly borked
14:44 < rxr> most probably did not worked for over a year already
14:44 < rxr> but I'm on the track to reimplement those bit in a way one would like to maintain
14:45 < rxr> the rock inherited dietlibc junk - is well - quite ugly ...
14:45 < mnemoc> the progress on T2/uclibc hit you :)
14:46 < rxr> yeah - and looking at the code (I guess esden injected) knocked me from the chair ...
14:47 < mnemoc> :)
14:49 < mnemoc> package/base/uclibc/patches/ncurses-100-buildccflags.patch <--- uhm?
14:49 < mnemoc> why deleted?
14:50 < mnemoc> log read
14:52 < mnemoc> Upgrade Patch for postfix (package/mail/postfix): 2.0.20 -> 2.1.5
14:52 < mnemoc> err
14:52 < mnemoc> sorry
14:55 < rxr> mnemoc: conflict with my general cross fix ..
14:55 < mnemoc> rxr: do you know if iptables-1.3 is backward compatible
14:55 < rxr> should build without
14:55 < rxr> no
14:56 < rxr> never heared of that one ...
14:56 < mnemoc> don't know or it's not?
14:56 < mnemoc> ok
14:56 < rxr> is it new?
14:56 < rxr> (than I can earn money posting the news ...)
14:56 < mnemoc> 2005-Feb-01: iptables-1.3.0rc1
14:56 < mnemoc> Finally, the first 1.3.0 release candidate is released. If you've been looking for tremendous ruleset update performance improvements, this is the way to go.
14:57 < rxr> no idea what is new ...
14:58 < rxr> arm-t2-linux-gnu/bin/ld: cannot find -lc
14:59 < rxr> ^- dietlibc dynamic for arm
15:05 < mnemoc> uhm
15:11 < rxr> ok - our wrappers fult
15:11 < rxr> fault
15:11 < rxr> s.th. is inserting static
15:11 < rxr> and then gcc only looks for libc.a ....
15:12 < mnemoc> ximina evolution died?
15:12 < rxr> what ?
15:13 < mnemoc> www.ximian.com/products/ximian_evolution/ redirects to a 'buy it' at novell
15:13 < rxr> hehe
15:15 < rxr> so - when my dietlibc builds dynamic for arm now - I'll be off for lunch
15:15 < rxr> yeah
15:15 < rxr> == 02/03/05 15:15:45 =[1]=> Finished building package dietlibc.
15:16 < rxr> ok - that looks dynamic:
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libc.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libc.so
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libcompat.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libcompat.so
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libcrypt.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libdl.so
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/liblatin1.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libm.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libm.so
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libpthread.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libpthread.so
15:18 < mnemoc> :)
15:19 < CIA-8> rene * r5952 /trunk/package/base/dietlibc/dietlibc.conf:
15:19 < CIA-8> * fixed dietlibc to also build the -dyn wrapper (using the system libc)
15:19 < CIA-8> * removed some more || true nonone wants in a .conf
15:19 < CIA-8> * fixed a variable name typo I introduced a few weeks ago and refixed
15:19 < CIA-8> build accoringly
15:19 < CIA-8> * indenting cleanups
15:19 < CIA-8> (sorry - only tested to cross build with the dynamic part on ARM)
15:19 < rxr> == 02/03/05 15:19:08 =[1]=> Finished building package gzip.
15:19 < rxr> ^- dynamic arm
15:22 < CIA-8> rene * r5953 /trunk/package/shells/zsh/zsh.desc: * updated zsh (4.2.1 -> 4.2.4)
15:26 < rxr> ouhm
15:26 < rxr> seems I should quickly do some libm work for dietlibc:
15:26 < rxr> undefined reference to `log'
15:26 < rxr> undefined reference to `sqrt'
15:26 < rxr> undefined reference to `floor'
15:26 < rxr> undefined reference to `fmod'
15:26 < rxr> undefined reference to `__half'
15:26 < rxr> undefined reference to `exp'
15:28 * rxr off - lunch - and doing a test bootdisk x86 build to see what breaks ...
15:28 < mtr> hi
15:29 < mnemoc> hi mtr
15:29 < mtr> hi mnemoc
15:35 < mnemoc> On Mon, Jan 31, 2005 at 11:48:47AM -0300, Alejandro Mery wrote:
15:35 < mnemoc> >> hi, i'm trying to build a non-NLS system which of course doesn't have
15:35 < mnemoc> >> iconv.h. but buffer.h wants it.
15:35 < mnemoc> I'd just install libiconv. If you really want to build lftp without it,
15:35 < mnemoc> add a bunch of #ifdef's to sources, then send me the resulting patch.
15:35 < mnemoc> -- Alexander.
15:35 < mnemoc> snif snif
15:54 -!- madtux [~mike@196.40.44.28] has joined #t2
15:54 < madtux> hi
15:54 < mnemoc> hi mightytux
16:14 < madtux> hi mnemoc
17:19 -!- _Ragnar__ is now known as _Ragnar_
17:23 < mnemoc> _Ragnar_: OT question, what 'to be fly' means?
17:23 < _Ragnar_> no 'on the fly'?
17:23 < mnemoc> nope
17:24 < mnemoc> "i'm preaty fly for a white guy", on a song
17:24 < _Ragnar_> no clue
17:24 < mnemoc> offspring
17:25 < daja77> ;)
17:26 < mnemoc> _Ragnar_: isn't that your native language? :(
17:26 < mnemoc> "And all the girlies say I'm pretty fly for a white guy"
17:26 < mnemoc> ive it to me baby, uh huh, uh huh
17:26 < mnemoc> Give it to me baby, uh huh, uh huh
17:26 < mnemoc> :\
17:26 < mnemoc> music
17:27 < _Ragnar_> wordnets says: fly (n):
17:27 < _Ragnar_> 4: (baseball) a hit that flies up in the air [syn: {fly ball}]
17:28 < mnemoc> how that definition applies to the sense??
17:29 < _Ragnar_> like 'I'm a hit'
17:29 < _Ragnar_> tho I kinda doubt offspring would do such play on words
17:29 < _Ragnar_> you sure they don't sing 'sly' ?
17:30 < mnemoc> http://www.lyrics007.com/Offspring%20Lyrics/Pretty%20Fly%20(For%20A%20White%20Guy)%20Lyrics.html
17:31 < _Ragnar_> *shrugs*
17:31 < _Ragnar_> no clue then aside from the one I already gave
17:34 < mnemoc> what 'to be a hit' means?
17:52 < rxr> rehi
17:52 < mnemoc> hi rxr
17:55 < CIA-8> rene * r5954 /trunk/package/base/linux24/arch-sparc64-atomic.patch: * well - arch-sparc64-atomic.patch does not apply to linux24 ...
18:04 -!- mnemoc [~amery@200.75.27.6] has quit [Read error: 104 (Connection reset by peer)]
18:30 -!- madtux [~mike@196.40.44.28] has quit ["Lost terminal"]
19:02 < _Ragnar_> hi
19:02 < rxr> == 02/03/05 19:00:21 =[1]=> Finished building package ncurses.
19:03 < rxr> ^- dietlibc arm - libm hacked ncurses ...
19:03 < rxr> libm of dietlibc hacked for ncurses that is ...
19:04 < _Ragnar_> uh huh
19:08 < _Ragnar_> 1-ncurses still fails with uclibc
19:08 < rxr> I did nothing to it so far
19:08 < rxr> can you provide the full error log to me ?
19:09 < _Ragnar_> http://66.146.166.62/~loki/1-ncurses.err
19:14 < rxr> this happens in the ncurses rebuid with wide char support
19:14 < rxr> can you take a look why the type is undefined there and if you have in from uclibc ?
19:15 < rxr> maybe it just build accidently before because the former patched crutially redefined some CCPFLAGS stuff maybe resulting in the 2nd build beeing a normal (not wide but narrow) ncurses build ...
19:15 < rxr> if you do not have this type we could just disable the ncursesw build for uclibc as I did for dietlibc ...
19:17 < CIA-8> rene * r5955 /trunk/package/security/openssl/openssl.conf: * fixed openssl build for arm (added target to the case ...)
19:18 < _Ragnar_> uhm okay
19:20 < _Ragnar_> rxr: ncurses should define that type itself
19:20 < rxr> ouhm ok
19:21 < _Ragnar_> so this is more likely to mean wrong header files get included...
19:21 < rxr> i take a look
19:22 < _Ragnar_> it depends on _XOPEN_SOURCE_EXTENDED being defined
19:23 < _Ragnar_> which should get defined with _GNU_SOURCE
19:46 * rxr soon taking a lookg
19:46 < rxr> my brother phoned and then some commerial s.th. wanted to sell newspaper abos ...
19:46 < rxr> busy hour ...
19:46 < rxr> when openssl builds I'll take a look into ncurses
19:50 < rxr> yeah
19:50 < rxr> openssl built for arm with dynamic dietlibc ...
19:50 < rxr> + it is a cross build ...
19:51 < rxr> so - now I test some openssl.conf cleanup
20:32 < _Ragnar_> rxr: did you fix the ncurses thingy?
20:32 < rxr> not et
20:33 < rxr> cleaning openssl up
20:34 < _Ragnar_> k
20:35 < rxr> oh my god is openssl huge: 947116
20:36 < rxr> ^- dynamic dietlibc arm
20:36 < rxr> 1188416 <- dynamic glibc x86
20:36 < rxr> so no ssl on my psion ;_)
20:36 < rxr> but good to know we can cross build openssl now ;-)
20:37 < rxr> and the .conf is not as amazingly ugly as it was before ...
20:37 < rxr> == 02/03/05 20:37:16 =[1]=> Finished building package openssl.
20:39 < jsaw> re
20:39 < rxr> hi jsaw
20:39 < jsaw> hi rxr!
20:40 < _Ragnar_> did you patch that for dietlibc, or in general again (will it break uclibc)?
20:41 < rxr> I cleaned it up significantly ...
20:41 < rxr> it is cross build able now
20:42 < rxr> it should not break with other libc's ...
20:42 < rxr> in general I of course do not mangle packages due to dietlibc - all dietlibc junk will go into packge/base/dietlibc/....
20:42 < _Ragnar_> okay ;)
20:43 < rxr> it is just that ncurses was broken in general - and openssl.conf so ugly that I could not sleep well with that in t2
20:45 < _Ragnar_> *lol* I see
20:45 < rxr> as in you just looked into it ?
20:48 < _Ragnar_> what's so ugly about it?
20:48 < rxr> custmain
20:48 < rxr> ugly /etc/ssl symlinking to get files into /usr
20:48 < rxr> and then the stuff it not root / DESTDIR save and just overwrites the bulid system's /usr and so on
20:49 < _Ragnar_> yeah I see now
20:49 < _Ragnar_> that sucks
20:50 < rxr> .oO
20:50 < rxr> we did not install the openssl man files properly before?
20:50 < rxr> !> Found shared files with other packages:
20:50 < rxr> !> usr/share/man/man1/passwd.1: openssl shadow
20:50 < rxr> !> usr/share/man/man3/buffer.3: libowfat openssl
20:50 < rxr> !> usr/share/man/man3/err.3: man-pages openssl
20:50 < rxr> !> usr/share/man/man3/rand.3: man-pages openssl
20:50 < rxr> !> usr/share/man/man3/threads.3: openssl perl
20:51 < rxr> *lol*
20:51 < _Ragnar_> uhm
20:51 < rxr> _Ragnar_: that we did not install the openssl manpages ? *g*
20:51 < _Ragnar_> sure?
20:51 < rxr> openssl: etc/ssl/man
20:51 < rxr> openssl: usr/share/doc/openssl/man
20:51 < rxr> openssl: usr/share/doc/openssl/man/man7
20:51 < rxr> openssl: usr/share/doc/openssl/man/man7/Modes of DES.7
20:51 < rxr> that is openssl on ROCK Linux 2.0.1 (2004/03/13)
20:52 < _Ragnar_> oh *lol*
20:53 < rxr> in my T2 working copy I get dozens installed
20:53 < rxr> and the above conflicts ...
20:54 < rxr> .o(why did I started to clean up openssl - why ...)
20:55 < _Ragnar_> =) good luck
20:56 -!- mnemoc [~amery@200.75.28.126] has joined #t2
20:57 < mnemoc> back
20:57 < rxr> _Ragnar_: know that I spent so much time with openssl - might you liek to review ncurses ?
20:57 < mnemoc> did i miss too much?
20:57 < rxr> _Ragnar_: e.g. review if the uclibc patch added some -I or so that we might like to add
20:57 < rxr> mnemoc: some bashing of the ROCK inherited openssl package and such - not much ...
20:58 < mnemoc> rxr: ncurses patch removes CFLAGS only
20:58 < rxr> my cross all and dietlibc making progress
20:58 < rxr> mnemoc: didn't it hardset some other option, too
20:58 < rxr> I'm sure I saw s.th. like that
20:58 < mnemoc> nope
20:58 < mnemoc> maybe mom
20:59 < _Ragnar_> mnemoc: the .conf does
20:59 < _Ragnar_> but that shouldn't affect it
20:59 < rxr> does or did *g* ?
20:59 < mnemoc> # svn merge -r5947:5946 .
20:59 < mnemoc> does
21:00 < mnemoc> # configure want BUILD_CPPFLAGS=CPPFLAGS which includes $xroot/usr/include, harmfull for BUILDCC
21:00 < mnemoc> export BUILD_CPPFLAGS="-I../include -D_GNU_SOURCE -DNDEBUG"
21:00 < _Ragnar_> yes that defines GNU_SOURCE
21:00 < _Ragnar_> which we need her3e
21:00 < _Ragnar_> here*
21:01 < mnemoc> 7 -BUILD_CCFLAGS = -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) @BUILD_CPPFLAGS@ @BUILD_CFLAGS@
21:01 < rxr> can you rephrase: what need to be readded to get it compile ?
21:01 < mnemoc> 8 +BUILD_CCFLAGS = -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) @BUILD_CPPFLAGS@
21:01 < mnemoc> they are complementary
21:01 < mnemoc> some stuff on @BUILD_CFLAGS@ is harmfull
21:01 < mnemoc> some is necesary
21:02 < mnemoc> so i added the necesary stuff on BUILD_CPPFLAGS and removed BUILD_CFLAGS
21:03 < mnemoc> the comment tells the real problem
21:03 < rxr> can you point me to the exact variable that after my cleanup is a) present that shouod not be - or b) missing ?
21:03 < rxr> == 02/03/05 21:00:31 =[9]=> Finished building package openssl.
21:04 < mnemoc> me? no... i should start a new uclibc build
21:04 < mnemoc> and complete uboot
21:04 < mnemoc> i need some hours to get that point
21:05 < rxr> mnemoc: ok ok
21:05 < rxr> _Ragnar_: ?? *g*
21:05 < mnemoc> :p
21:05 < rxr> this one is big ->
21:05 < CIA-8> rene * r5956 /trunk/package/security/openssl/openssl.conf: (log message trimmed)
21:05 < CIA-8> * deep openssl.conf cleanup:
21:05 < CIA-8> * do not use custmain
21:05 < CIA-8> * do not do ugly symlinking
21:05 < CIA-8> * really install the manpages
21:05 < CIA-8> * be cross copmilable
21:05 < CIA-8> * the forgotten bits
21:06 < rxr> ^- please test ;-)
21:06 < _Ragnar_> I can't right nw
21:06 < _Ragnar_> now*
21:06 < rxr> ok - no problem at all
21:06 < _Ragnar_> rxr: but what seems to be definitely missing from your flags is -D_GNU_SOURCE
21:07 < rxr> _Ragnar_: does it work when you ineject it to the patch
21:07 < rxr> mom
21:08 < mnemoc> rxr: is your patch postconf too?
21:08 < rxr> postconf ?
21:08 < rxr> Makefile.in ...
21:08 < mnemoc> great :p
21:08 < mnemoc> i patched Makefile directly :D
21:08 < rxr> _Ragnar_: does this work:
21:08 < rxr> --- cross-build.patch (revision 5938)
21:08 < rxr> +++ cross-build.patch (working copy)
21:08 < rxr> ...
21:09 < mnemoc> err
21:09 < rxr> -+BUILD_CCFLAGS = -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) -I. -I../include
21:09 < rxr> ++BUILD_CCFLAGS = -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) -I. -I../include -D_GNU_SOURCE
21:09 < mnemoc> rxr: i confused it with ncurses/bash problem
21:10 < _Ragnar_> building...
21:14 < mnemoc> any objections to imap_headercache patch on mutt?
21:14 < jsaw> ?
21:14 < mnemoc> hi jsaw
21:14 < jsaw> hi mnemoc
21:15 < mnemoc> i want to add that third party patch, objections?
21:15 < jsaw> i don't know it - so i simply trust you ;)
21:15 < mnemoc> :)
21:16 < _Ragnar_> rxr: ncurses builds with that patch
21:16 < mnemoc> :)
21:16 < jsaw> ack.
21:16 < jsaw> hi _Ragnar_ btw.
21:17 < rxr> hm -- shoudl I simply commit that hardcoding ... *g* ;-)
21:17 < jsaw> yep.
21:17 < _Ragnar_> yes
21:17 < _Ragnar_> hi jsaw
21:18 < _Ragnar_> does the -D_GNU_SOURCE break dietlibc or glibc?
21:18 < rxr> _Ragnar_: we'll see ;-)
21:18 < CIA-8> rene * r5957 /trunk/package/base/ncurses/cross-build.patch: * define -D_GNU_SOURCE in ncurses BUILD_CCFLAGS to make uclibc happy
21:18 < jsaw> rxr: we have squashfs-tools, but no squashfs in kernel...?
21:19 < rxr> no - we could inject it if we would like to. ...
21:19 < rxr> jsaw: do you play with it ?
21:19 < CIA-8> mnemoc * r5958 /trunk/package/mail/mutt/ (mutt.conf mutt.desc no_autogen_for_headercache.patch):
21:19 < CIA-8> * updated mutt (1.4.2i -> 1.4.2.1i)
21:19 < CIA-8> * improved mutt with imap_headercache patch
21:19 < jsaw> no yet but soon. First I try to get it running with cramfs...
21:19 < rxr> on which arch and hardware ?
21:20 < jsaw> target/distcc, x86/x86
21:21 < jsaw> maybe I should call ist distcccd...
21:21 < jsaw> -s
21:21 < rxr> distcd ;-)
21:21 < rxr> dist-tripple-c-d
21:21 < jsaw> distc^3d
21:21 < jsaw> ....
21:21 < jsaw> distcd is good! I like that.
21:22 < jsaw> I only have to insert an "u" for uclibc somewhere... *g*
21:22 < rxr> so the target name will be distkajdfghlksdjfhcd in the end
21:23 < rxr> encoding any package capital ...
21:23 < jsaw> rxr: btw, reading the squashfs page, seems like it's speed is quite good.
21:23 < jsaw> well I mean the comparison page
21:23 < rxr> it is read-only, you know that ?
21:23 < jsaw> wait...
21:23 < rxr> oehrm
21:24 < jsaw> looks like readonly
21:24 < rxr> jsaw: i only read about it in read-only ontect
21:25 < rxr> the compression write would be quite expensive for the squashfs design ...
21:25 < jsaw> right...
21:26 < rxr> I heavily patch dietlibc right now ...
21:27 < rxr> jsaw: how many mails are on the dietlibc list?
21:27 < rxr> I subscribed yesterday and got none so far ...
21:27 < jsaw> mails are very rare...
21:27 < rxr> .oO
21:28 < rxr> and the archive never worked ... :-(
21:28 < jsaw> last one was 1 Feb
21:28 < rxr> ok - thanks
21:28 < jsaw> hm, now gawk fails.
21:28 < CIA-8> mnemoc * r5959 /trunk/package/ (network/ipvsadm/ security/ipvsadm/): * moved ipvsadm from security/ to network/ because basicly it's a tool to set routes
21:29 < rxr> there is some more stuff to sort
21:29 < rxr> e.g. the extra/ stuff as well as quite a lot of base that is not base ...
21:29 < rxr> == 02/03/05 21:29:14 =[1]=> Finished building package embutils.
21:29 < rxr> ^- ok - more dietlibc dyn fixed (failed due to pivot_root before)
21:29 < rxr> so /me evening dinner
21:29 < rxr> and then maybe preparing patches for fefe
21:29 < rxr> and then there is this libm stuff I should inject into dietlibc ...
21:30 < rxr> and - well - this money earning scanner driver thing :-(
21:30 < _Ragnar_> what did you inject into libm?
21:30 < rxr> dietlibc
21:30 < mnemoc> question: ipvsadm-1.21 is for 2.4 kernels only, ipvsadm-1.24 is for 2.6 kernels only. both kernels include ipvs code. our ipsvadm package has 1.24 and i got a patch to make 1.24 support .old as modutils. what do you propose?
21:30 < rxr> some infrastructure bits to get my arm binaries linked ...
21:31 < mnemoc> ipvsadm-1.24 as ipvsadm2 package? ipvsadm-1.21 as ipvsadm-$foo package?
21:31 < rxr> well - no idea
21:31 < mnemoc> both on ipsvadm package?
21:31 < rxr> I do not even know what the package does ...
21:31 < rxr> mnemoc: you are the maintainer? much fun throwing a coin ;-)
21:31 < mnemoc> :\
21:32 < rxr> if you need both - package both ;-)
21:32 < mnemoc> no more voting?
21:32 < rxr> configure:6152: checking compiler and flags for sanity
21:32 < rxr> configure:6156: error: cannot run test program while cross compiling
21:32 < rxr> ^- this is what openssl screams about ?!?!?!
21:32 < rxr> f*cking package
21:32 * rxr off for evening dinner ...
21:32 < mnemoc> LOL
21:35 < rxr> ok - stupid test patched away ...
21:35 < rxr> ok - zlib missing ...
21:37 < _Ragnar_> brb, food
21:38 < rxr> checking for zlib 1.1.4 or greater... configure: error: cannot run test program
21:38 < rxr> while cross compiling
21:39 < rxr> ok - finally food
21:44 < CIA-8> mnemoc * r5960 /trunk/package/network/ipvsadm2/: * added ipvsadm2 package as a copy of ipvsadm
21:48 < CIA-8> mnemoc * r5961 /trunk/package/network/ipvsadm2/ (6 files): * renamed ipvsadm2 files according to new $pkg name
22:04 -!- mnemoc [~amery@200.75.28.126] has quit [Read error: 110 (Connection timed out)]
22:07 -!- mnemoc [~amery@200.75.27.22] has joined #t2
22:08 < mnemoc> am i really here?
22:08 < daja77> seems so
22:08 < mnemoc> :D
22:09 < mnemoc> good
22:09 < mnemoc> hi daja77
22:09 < mnemoc> rxr: here?
22:09 < daja77> hi mnemoc
22:12 < mnemoc> rxr: i have a postlinux.conf dilema. ipvsadm needs _only_ net/ip_vs.h from kernel sources which is not on /usr/include. i) postlinux.conf to only copy that header? ii) ipvsadm/parse-config to copy that header iii) linux24.conf hardcoded copy of that header. ip_vs.{o,ko} is built always but both kernels.
22:21 -!- ShinGouki [~bla@me-lewiston-pw-627.mint.adelphia.net] has joined #t2
22:21 -!- ShinGouki [~bla@me-lewiston-pw-627.mint.adelphia.net] has left #t2 []
22:41 < _Ragnar_> re
22:42 < mnemoc> re _Ragnar_
22:43 < _Ragnar_> what did I miss?
22:44 < mnemoc> no idea, my connection sticks today
22:46 < rxr> mnemoc: ouohm - can I think about that later - e.g. tomorrow ?
22:46 < CIA-8> mnemoc * r5962 /trunk/package/network/ipvsadm2/pkg_linux-header_pre.conf: * added hook to install kernel's net/ip_vs.h if ipvsadm2 will be installed
22:47 < mnemoc> rxr: ok
22:47 < rxr> I see you already decided ... :-)
22:47 < mnemoc> the method can be changed
22:47 < mnemoc> i only want it building
22:52 < rxr> why the h*ll does openssh want to run so many test programs ...
22:52 * rxr headshake
22:57 < rxr> I give up to cross compile openssh shit right now
22:57 < rxr> I would not have imagined that this OpenBSD prime time project has such an ugly configure ...
22:58 < mnemoc> ipvsadm for 2.4 doens't build against 2.6 headers :(
22:59 < _Ragnar_> *g*
22:59 < rxr> inject the needed headers into the ipvsadm package ...
22:59 < rxr> using -I $PWD/include
22:59 < rxr> and drop the needed headers into it ...
22:59 < mnemoc> ok
23:00 < mnemoc> i'll do that for 2.4
23:00 < mnemoc> keepalived also needs it so i prefer 2.6's on /usr/include
23:00 -!- praenti [~praenti@82.139.198.99] has quit [Read error: 104 (Connection reset by peer)]
23:00 -!- praenti [~praenti@82.139.198.99] has joined #t2
23:17 < CIA-8> mnemoc * r5963 /trunk/package/network/ipvsadm/ (inject_2.4.29_ip_vs.h.patch ipvsadm.desc): * downdated ipvsadm (1.24 -> 1.21-11) to be ipvsadm for 2.4 kernels
23:39 < CIA-8> mnemoc * r5964 /trunk/package/network/ (6 files in 2 dirs): * improved ipvsadm and ipvsadm2 to cohexist using .old suffix on ipvsadm tools when ipvsadm2 is installed
23:59 < CIA-8> mnemoc * r5965 /trunk/package/ (network/keepalived/ security/keepalived/): * moved keepalived from security/ to network/
--- Log closed Fri Feb 04 00:00:48 2005
00:02 < CIA-8> rene * r5945 /trunk/package/editors/elvis-tiny/elvis-tiny.desc:
00:02 < CIA-8> * marked elvis-tiny to build in stage 5 (not 1+3) - I shuffle the
00:02 < CIA-8> build stages in the psion-pda target now as I wish
00:15 < rxr> -rwxr-xr-x 1 root root 252356 Feb 3 00:14 rsync
00:15 < rxr> ^- arm dietlibc cross ;-)
00:17 < mnemoc> rsync??!
00:19 < rxr> yeah - thought it might be a nice option to sync my notes over the IRDA port ...
00:19 < rxr> but it looks far too huge ...
00:19 < rxr> but at least it is dietlibc able now ...
00:20 < _Ragnar_> static link?
00:21 < rxr> yep
00:21 < _Ragnar_> ah okay
00:21 < mnemoc> .oO( dietlibc use-it package selection will get loooooonger )o
00:21 < mnemoc> rxr: is it too expensive to have it dynamic?
00:21 < rxr> mnemoc: yep - why .oO ?
00:22 < rxr> when we (at ROCK times) adapted to diet it was static only
00:22 < rxr> only later x86 dynamic support was added
00:22 < rxr> then arm and now x86-64
00:22 < mnemoc> on Config we select one by one which files do we want to build againts diet
00:22 < rxr> I think the other arches still have no dynamic support
00:22 < mnemoc> s/files/packages/
00:22 < rxr> yep - I know
00:22 < rxr> why .oO?
00:23 < mnemoc> the list is long
00:23 < mnemoc> and you are improving it to be even longer
00:23 < rxr> yep ;-)
00:23 < rxr> dietlibc is soo cute - lowfat and clean ;-)
00:23 < CIA-8> rene * r5946 /trunk/package/ (2 files in 2 dirs):
00:23 < CIA-8> * added patch to support building rsync with dietlibc - and marked
00:23 < CIA-8> it DIETLIBC ready
00:24 < rxr> I want dietlibc only systems soonder than later ..
00:24 < rxr> *g*
00:24 < mnemoc> rxr: that's why i prefer pure systems, LIBC=dietlibc not mixed s.th
00:24 < rxr> when I claned the dietlibc stuff (read reimplemented to be less hacky and ugly)
00:24 < _Ragnar_> what keeps you?
00:24 < rxr> lack of packages building with it ;-)
00:25 < rxr> when I reimplemented the diet stuff more packages should build without hacking - maybe we would then rather define tested uclibc and dietlibc selection templates - in contrast to adding FLAGS over all the pacag
00:25 < rxr> es
00:25 < mnemoc> rxr: to be static is just an option with can be forced for some archs ;)
00:26 < rxr> sure
00:26 < rxr> there already is that option in the config ...
00:26 < _Ragnar_> it doesn't work for some packages tho
00:26 < mnemoc> rxr: _Ragnar_ is trying to build everything against uclibc and it seems to work
00:26 < mnemoc> :(
00:26 < _Ragnar_> I'm not building static
00:26 < mnemoc> i have to go
00:27 < mnemoc> i know
00:27 < rxr> cu mnemoc
00:27 < _Ragnar_> awww kay
00:27 < rxr> ucibc is full of compat cut'n paste
00:27 < rxr> although neither POSIX nor any other standard
00:27 < rxr> such stuff is not in dietlibc - and if it is it is #ifdef'ed _GNU_SOURCE _BSD_SOURCE
00:27 < rxr> thus the dietlibc/parse-config is full of
00:27 < rxr> nvi)
00:28 < rxr> var_append GCC_OPTIONS_APPEND " " "-D_GNU_SOURCE"
00:28 < rxr> and co to get those packags build
00:28 < rxr> this is why more stuff just builds with uclibc and needs at least some define injections with dietlibc
00:29 < rxr> also C++ stuff might not yet fully work with dietlibc - so dietlibc only systems will be some bigger task not finished tomorrow ...
00:30 < rxr> _Ragnar_: what is the uclibc target triplet?
00:30 < rxr> i386-linux-t2-uclibc or so ?
00:30 < _Ragnar_> ix86-t2-linux-uclibc
00:30 < rxr> ix86 ?
00:31 < _Ragnar_> i686 in my case
00:31 < rxr> ah - ok ;-)
00:43 < rxr> UPN Officially Cancels 'Star Trek: Enterprise' ???
00:43 < rxr> didn't the announced last year that they have signed the full 8 seasons ...
00:56 < rxr> my PDA target has this buildtime:
00:56 < rxr> real 20m39.704s
00:56 < rxr> user 14m7.435s
00:56 < rxr> sys 4m4.398s
00:56 < rxr> 17 builds total, 17 completed fine, 0 with errors.
00:56 < rxr> on a 2.5Ghz Athlon ;-)
00:56 < rxr> n8 all
00:57 < _Ragnar_> gn8
02:27 < mnemoc> re
02:28 < _Ragnar_> re
02:29 < _Ragnar_> hmm does newer gcc/g++ have a different exception handling?
02:29 < mnemoc> no idea
02:29 < mnemoc> why?
02:29 * _Ragnar_ has a problem with cross-compiling an app with mingw
02:30 < _Ragnar_> my exceptions don't get caught at all
02:37 < mnemoc> o_O
04:35 -!- rxr_ [~rene@p213.54.201.251.tisdip.tiscali.de] has joined #t2
04:35 -!- Topic for #t2: T2 | 2.1.0-beta2 RELEASED | The next generation of System Development Enviroments (SDE) | http://www.t2-project.org/ | CIA, don't ever leave us again!
04:35 -!- Topic set by mnemoc [] [Wed Jan 12 03:04:43 2005]
04:35 [Users #t2]
04:35 [ _Ragnar_] [ daja77] [ mnemoc] [ nzg ] [ praenti] [ rxr_ ]
04:35 [ CIA-8 ] [ jsaw ] [ mtr ] [ ooo_] [ rxr ] [ valentin]
04:35 -!- Irssi: #t2: Total of 12 nicks [0 ops, 0 halfops, 0 voices, 12 normal]
04:35 -!- Channel #t2 created Sun Aug 8 21:15:33 2004
04:35 -!- rxr [~rene@p213.54.197.108.tisdip.tiscali.de] has quit [Read error: 104 (Connection reset by peer)]
04:35 -!- [freenode-info] why register and identify? your IRC nick is how people know you. http://freenode.net/faq.shtml#nicksetup
04:35 -!- Irssi: Join to #t2 was synced in 10 secs
04:43 -!- mtr_ [~michael@Ha22a.h.pppool.de] has joined #t2
04:44 -!- ideal [~idealm@211.100.227.107] has joined #t2
04:52 -!- mtr [~michael@H8616.h.pppool.de] has quit [Read error: 60 (Operation timed out)]
05:36 -!- ideal [~idealm@211.100.227.107] has quit [Client Quit]
05:49 -!- ideal [~idealm@211.100.227.107] has joined #t2
07:17 -!- mnemoc_ [~amery@200.75.27.43] has joined #t2
07:17 -!- mnemoc [~amery@200.75.27.32] has quit [Read error: 60 (Operation timed out)]
08:07 -!- Netsplit over, joins: praenti
08:07 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: valentin, daja77, mnemoc_, _Ragnar_, nzg, ideal
08:08 -!- mnemoc [~amery@200.75.27.43] has joined #t2
08:09 -!- _Ragnar_ [loki@66-146-166-62.skyriver.net] has joined #t2
08:09 -!- daja77 [[68kg0qZ1V@odoaker.hrz.tu-chemnitz.de] has joined #t2
08:12 -!- ideal [~idealm@211.100.227.107] has joined #t2
08:13 -!- valentin [~valentin@port-212-202-42-228.dynamic.qsc.de] has joined #t2
08:13 -!- nzg [~tschmidt@xchangecenter.de] has joined #t2
08:15 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: daja77, _Ragnar_
08:16 -!- Netsplit over, joins: _Ragnar_, daja77
08:17 < rxr_> moin
08:17 -!- You're now known as rxr
08:33 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: mnemoc
08:33 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: valentin, nzg, ideal
08:36 -!- Netsplit over, joins: nzg
08:36 -!- Netsplit over, joins: mnemoc
08:36 -!- Netsplit over, joins: valentin
08:42 -!- madtux [~mike@65.182.14.165] has joined #t2
08:42 < madtux> re
08:42 < rxr> hi madtux
08:43 < madtux> guten morgen mein freund :)
08:44 < rxr> good morning over there ;-)
08:45 < madtux> well its just 1:45am here.. but thanks :)
08:48 < _Ragnar_> hi
08:50 -!- ideal [~idealm@211.100.227.107] has joined #t2
08:50 < rxr> madtux: hehe - sorry I'm so sleepy and writing some article stuff that I forgot about that ...
08:50 < rxr> moin _Ragnar_
08:50 < rxr> welcome ideal
08:51 < _Ragnar_> ;)
08:51 < _Ragnar_> moin
08:52 < madtux> helo _Ragnar_ and ideal
08:52 < madtux> rxr, its ok.. i'm currently downloading some packages..
08:52 < madtux> just got 2 new servers which are not yet going live for production
08:52 < madtux> so i will use them for building processes
08:56 < _Ragnar_> ;)
08:57 < madtux> 2x 2.8GHZ xeon, 2x 73 15k rpm scsi, 4gb mem
08:57 < madtux> that shall be enough
08:57 < madtux> ah well thats it for tonight i must sleep
08:57 < madtux> night all
08:58 -!- madtux [~mike@65.182.14.165] has quit ["Leaving"]
09:02 -!- mtr_ is now known as mtr
09:02 < mtr> hi
09:04 < mtr> ncurses/cross-build.patch does interfere with a uclibc ncurses patch
09:04 < mtr> !> Apply patch /work/rock/t2-work/package/base/uclibc/patch ..
09:04 < mtr> !> patching file ncurses/Makefile.in
09:04 < mtr> !> Hunk #1 FAILED at 93.
09:04 < _Ragnar_> ;p
09:07 < mtr> hmm, eventually the ncurses cross-build.patch should not be applied in general?
09:08 < _Ragnar_> that's why I tried to confine any changed to the uclibc/patches, instead of putting patches into the packages themselves
09:08 < _Ragnar_> changes*
09:09 < rxr> hm
09:09 * rxr looking what this uclibc patch does. ..
09:09 < rxr> maybe s.th. quite simillar ...
09:10 < rxr> hm - ic
09:11 < rxr> r5661 | mnemoc | 2005-01-19 04:08:31 +0100 (Wed, 19 Jan 2005) | 2 lines
09:11 < rxr> * added uclibc patch and hack to build ncurses
09:11 < rxr> hm
09:13 < rxr> _Ragnar_: mtr can you try if ncurses builds with uclibc when you remove the non applying patch from uclibc ?
09:13 < _Ragnar_> uhm
09:13 < mtr> hmm
09:13 < _Ragnar_> but including your patch?
09:13 < rxr> yes
09:14 < rxr> the one placed in the ncurses package
09:18 < _Ragnar_> both the patch and the .conf fix, or just the patch?
09:22 * _Ragnar_ pokes rxr
09:22 < mtr> defining _GNU_SOURCE and NDEBUG should be safe, and -I../include is set in cross-build.patch, too
09:23 < mtr> oh, wait
09:23 < mtr> == 02/03/05 09:21:24 =[1]=> Aborted building package ncurses
09:23 < mtr> just removing the uclibc patch does not work
09:25 < jsaw> re
09:25 * mtr disabling now the other patch ...
09:25 < jsaw> rxr: 5am??? you were a bit late, ain't you?
09:25 < mtr> hi jsaw
09:25 < jsaw> hi mtr
09:26 < jsaw> hi _Ragnar_
09:26 < _Ragnar_> hi jsaw
09:28 < mtr> _Ragnar_, i have two more packages for uclibc_fix_configsub, do you mind if i commit ?
09:29 < _Ragnar_> mtr: hmmm I was going to remove most of the cases for that and substitute with a general find for configure/config.sub
09:30 < mtr> yes, i read about this, therefore i asked
09:30 < _Ragnar_> which seems to work pretty well
09:31 < _Ragnar_> but I really need to go through the build logs first and see if there are more hidden errors
09:31 < _Ragnar_> before I commit that
09:31 < mtr> so, just waiting for your commit
09:33 < _Ragnar_> okay
09:33 < _Ragnar_> what packages/target are you building
09:33 < _Ragnar_> ?
09:34 < mtr> some packages for my dvb-harddisc-recorder system, like mplayer vdr libdvdnnn and such
09:34 -!- mnemoc [~amery@200.75.27.43] has quit [Read error: 113 (No route to host)]
09:35 < mtr> == 02/03/05 09:33:55 =[1]=> Finished building package ncurses
09:36 < rxr> do you svn rm the ncurses patch and commit?
09:36 < mtr> do you want me to do this?
09:38 < _Ragnar_> which patch?
09:39 < _Ragnar_> imho, the uclibc one shouldn't be removed until it's worked around
09:40 < _Ragnar_> but I need to get to bed
09:41 < _Ragnar_> rxr: does the uclibc ncurses patch fix the general crossbuild case?
09:42 < _Ragnar_> anyway, nini
09:43 < mtr> ok, my uboot target continues to build, afk
09:43 < mtr> gn8 _Ragnar_
09:43 < _Ragnar_> gn8
09:44 < rxr> _Ragnar_: the patch I comitted fixes it in general
09:45 < rxr> I had no idea that uclibc tinker s.th. abroad ...
09:47 < CIA-8> rene * r5947 /trunk/package/base/uclibc/patches/ (ncurses-100-buildccflags.patch ncurses.conf):
09:47 < CIA-8> * removed uclibc ncurses patches (should be fixed by my generic cross
09:47 < CIA-8> build fixes to ncurses yesterday - the .patch removal was tested,
09:47 < CIA-8> however the .conf was not - but I'm confident it will work without)
09:47 -!- mnemoc [~amery@200.75.27.6] has joined #t2
09:48 < rxr> moin mnemoc
10:24 < rxr> Sun vergibt Service-Vertrag für interne Softwarewartung an CSC
10:24 < rxr> ???
11:03 < rxr> hm - my good old U5 still running 32bit userspace from ROCK times just fucked up ..
11:03 < rxr> I was no longer to write to any filesystem
11:03 < rxr> before the U160 IBM SCSI disk made some strange noice
11:03 < rxr> a reboot seems to cured that ..
11:03 * rxr headshake
11:04 < rxr> or not
11:04 < rxr> damn
11:04 < rxr> ah
11:04 < rxr> the usual / 100% filled
11:04 < rxr> damn - why can't programs have more sensible error messages ...
11:07 -!- ideal [~idealm@211.100.227.107] has quit ["Leaving"]
11:12 -!- mtr [~michael@Ha22a.h.pppool.de] has quit [Read error: 60 (Operation timed out)]
11:23 -!- _Ragnar__ [loki@66-146-166-62.skyriver.net] has joined #t2
11:23 -!- _Ragnar_ [loki@66-146-166-62.skyriver.net] has quit [Read error: 104 (Connection reset by peer)]
11:44 < rxr> oh - it seems I got a 1.3 in assembly hackery ... *g*
12:01 < CIA-8> rene * r5948 /trunk/package/x11/rxvt-unicode/rxvt-unicode.desc: * updated rxvt-unicode (4.7 -> 4.9)
12:30 < rxr> /home/rene/myrevo/a 2745 1226 1369 48% /mnt/misc
12:32 < rxr> 758467 Feb 3 12:45 initrd
12:47 < rxr> ok - the ARM binaries work
12:48 * rxr in a ash in this PDA ...
12:48 < rxr> no digicam - so no pics ...
13:28 < CIA-8> rene * r5949 /trunk/package/base/irda-utils/irda-utils.conf: * fixed irda-utils for cross builds
13:33 < CIA-8> rene * r5950 /trunk/package/base/dietlibc/parse-config-9:
13:33 < CIA-8> * fixed the dietlibc arch_target redefine to be -dietlibc instead of
13:33 < CIA-8> the ugly -gnulibc1. No idea how this could have be commited in the
13:33 < CIA-8> first place ... My Psion PDA testcase builds with the new one.
13:35 < CIA-8> rene * r5951 /trunk/target/psion-pda/ (config.in parse-config):
13:35 < CIA-8> * added iproute2 and irda-utils to the psion-pda target, as well
13:35 < CIA-8> as a overwrite of install_init to prevent init scripts from beeing
13:35 < CIA-8> instaslled
13:51 -!- mtr [~michael@H9b1f.h.pppool.de] has joined #t2
14:00 < rxr> dietlibc static vs. dynamic:
14:00 < rxr> -rwxr-xr-x 1 root root 2336 Feb 3 14:00 hello
14:00 < rxr> -rwxr-xr-x 1 root root 7908 Feb 3 14:00 hello-s
14:02 < rxr> vs. glibc:
14:02 < rxr> -rwxr-xr-x 1 root root 2804 Feb 3 14:01 hello-glibc
14:02 < rxr> -rwxr-xr-x 1 root root 315792 Feb 3 14:01 hello-glibc-s
14:02 < rxr> *g*
14:35 < mnemoc> moin!
14:35 < rxr> moin!
14:36 < mnemoc> "fixed the dietlibc arch_target redefine to be -dietlibc" <--- this impact every config.sub file, and binutils mainly
14:42 < rxr> nothing does use dietlibc
14:42 < rxr> and I see now regression
14:43 < rxr> using one totally messed name with some normal official name should not screw anything up ...
14:43 < mnemoc> let see how does it go here
14:43 < rxr> mnemoc: can you remove the dietlibc/sincostan.patch before you start the build
14:43 < mnemoc> ok
14:43 < rxr> it does not work on x86 and I need to take a look how to only inject it for non x86 builds ...
14:43 < rxr> well - ok I have some idea .. but - test without the patch first ...
14:44 < mnemoc> do you want me to test a pure minimal/dietlibc or a bootdisk?
14:44 < rxr> bootdisk
14:44 < rxr> pure minimal is highly borked
14:44 < rxr> most probably did not worked for over a year already
14:44 < rxr> but I'm on the track to reimplement those bit in a way one would like to maintain
14:45 < rxr> the rock inherited dietlibc junk - is well - quite ugly ...
14:45 < mnemoc> the progress on T2/uclibc hit you :)
14:46 < rxr> yeah - and looking at the code (I guess esden injected) knocked me from the chair ...
14:47 < mnemoc> :)
14:49 < mnemoc> package/base/uclibc/patches/ncurses-100-buildccflags.patch <--- uhm?
14:49 < mnemoc> why deleted?
14:50 < mnemoc> log read
14:52 < mnemoc> Upgrade Patch for postfix (package/mail/postfix): 2.0.20 -> 2.1.5
14:52 < mnemoc> err
14:52 < mnemoc> sorry
14:55 < rxr> mnemoc: conflict with my general cross fix ..
14:55 < mnemoc> rxr: do you know if iptables-1.3 is backward compatible
14:55 < rxr> should build without
14:55 < rxr> no
14:56 < rxr> never heared of that one ...
14:56 < mnemoc> don't know or it's not?
14:56 < mnemoc> ok
14:56 < rxr> is it new?
14:56 < rxr> (than I can earn money posting the news ...)
14:56 < mnemoc> 2005-Feb-01: iptables-1.3.0rc1
14:56 < mnemoc> Finally, the first 1.3.0 release candidate is released. If you've been looking for tremendous ruleset update performance improvements, this is the way to go.
14:57 < rxr> no idea what is new ...
14:58 < rxr> arm-t2-linux-gnu/bin/ld: cannot find -lc
14:59 < rxr> ^- dietlibc dynamic for arm
15:05 < mnemoc> uhm
15:11 < rxr> ok - our wrappers fult
15:11 < rxr> fault
15:11 < rxr> s.th. is inserting static
15:11 < rxr> and then gcc only looks for libc.a ....
15:12 < mnemoc> ximina evolution died?
15:12 < rxr> what ?
15:13 < mnemoc> www.ximian.com/products/ximian_evolution/ redirects to a 'buy it' at novell
15:13 < rxr> hehe
15:15 < rxr> so - when my dietlibc builds dynamic for arm now - I'll be off for lunch
15:15 < rxr> yeah
15:15 < rxr> == 02/03/05 15:15:45 =[1]=> Finished building package dietlibc.
15:16 < rxr> ok - that looks dynamic:
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libc.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libc.so
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libcompat.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libcompat.so
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libcrypt.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libdl.so
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/liblatin1.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libm.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libm.so
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libpthread.a
15:16 < rxr> dietlibc: usr/dietlibc/lib-arm/libpthread.so
15:18 < mnemoc> :)
15:19 < CIA-8> rene * r5952 /trunk/package/base/dietlibc/dietlibc.conf:
15:19 < CIA-8> * fixed dietlibc to also build the -dyn wrapper (using the system libc)
15:19 < CIA-8> * removed some more || true nonone wants in a .conf
15:19 < CIA-8> * fixed a variable name typo I introduced a few weeks ago and refixed
15:19 < CIA-8> build accoringly
15:19 < CIA-8> * indenting cleanups
15:19 < CIA-8> (sorry - only tested to cross build with the dynamic part on ARM)
15:19 < rxr> == 02/03/05 15:19:08 =[1]=> Finished building package gzip.
15:19 < rxr> ^- dynamic arm
15:22 < CIA-8> rene * r5953 /trunk/package/shells/zsh/zsh.desc: * updated zsh (4.2.1 -> 4.2.4)
15:26 < rxr> ouhm
15:26 < rxr> seems I should quickly do some libm work for dietlibc:
15:26 < rxr> undefined reference to `log'
15:26 < rxr> undefined reference to `sqrt'
15:26 < rxr> undefined reference to `floor'
15:26 < rxr> undefined reference to `fmod'
15:26 < rxr> undefined reference to `__half'
15:26 < rxr> undefined reference to `exp'
15:28 * rxr off - lunch - and doing a test bootdisk x86 build to see what breaks ...
15:28 < mtr> hi
15:29 < mnemoc> hi mtr
15:29 < mtr> hi mnemoc
15:35 < mnemoc> On Mon, Jan 31, 2005 at 11:48:47AM -0300, Alejandro Mery wrote:
15:35 < mnemoc> >> hi, i'm trying to build a non-NLS system which of course doesn't have
15:35 < mnemoc> >> iconv.h. but buffer.h wants it.
15:35 < mnemoc> I'd just install libiconv. If you really want to build lftp without it,
15:35 < mnemoc> add a bunch of #ifdef's to sources, then send me the resulting patch.
15:35 < mnemoc> -- Alexander.
15:35 < mnemoc> snif snif
15:54 -!- madtux [~mike@196.40.44.28] has joined #t2
15:54 < madtux> hi
15:54 < mnemoc> hi mightytux
16:14 < madtux> hi mnemoc
17:19 -!- _Ragnar__ is now known as _Ragnar_
17:23 < mnemoc> _Ragnar_: OT question, what 'to be fly' means?
17:23 < _Ragnar_> no 'on the fly'?
17:23 < mnemoc> nope
17:24 < mnemoc> "i'm preaty fly for a white guy", on a song
17:24 < _Ragnar_> no clue
17:24 < mnemoc> offspring
17:25 < daja77> ;)
17:26 < mnemoc> _Ragnar_: isn't that your native language? :(
17:26 < mnemoc> "And all the girlies say I'm pretty fly for a white guy"
17:26 < mnemoc> ive it to me baby, uh huh, uh huh
17:26 < mnemoc> Give it to me baby, uh huh, uh huh
17:26 < mnemoc> :\
17:26 < mnemoc> music
17:27 < _Ragnar_> wordnets says: fly (n):
17:27 < _Ragnar_> 4: (baseball) a hit that flies up in the air [syn: {fly ball}]
17:28 < mnemoc> how that definition applies to the sense??
17:29 < _Ragnar_> like 'I'm a hit'
17:29 < _Ragnar_> tho I kinda doubt offspring would do such play on words
17:29 < _Ragnar_> you sure they don't sing 'sly' ?
17:30 < mnemoc> http://www.lyrics007.com/Offspring%20Lyrics/Pretty%20Fly%20(For%20A%20White%20Guy)%20Lyrics.html
17:31 < _Ragnar_> *shrugs*
17:31 < _Ragnar_> no clue then aside from the one I already gave
17:34 < mnemoc> what 'to be a hit' means?
17:52 < rxr> rehi
17:52 < mnemoc> hi rxr
17:55 < CIA-8> rene * r5954 /trunk/package/base/linux24/arch-sparc64-atomic.patch: * well - arch-sparc64-atomic.patch does not apply to linux24 ...
18:04 -!- mnemoc [~amery@200.75.27.6] has quit [Read error: 104 (Connection reset by peer)]
18:30 -!- madtux [~mike@196.40.44.28] has quit ["Lost terminal"]
19:02 < _Ragnar_> hi
19:02 < rxr> == 02/03/05 19:00:21 =[1]=> Finished building package ncurses.
19:03 < rxr> ^- dietlibc arm - libm hacked ncurses ...
19:03 < rxr> libm of dietlibc hacked for ncurses that is ...
19:04 < _Ragnar_> uh huh
19:08 < _Ragnar_> 1-ncurses still fails with uclibc
19:08 < rxr> I did nothing to it so far
19:08 < rxr> can you provide the full error log to me ?
19:09 < _Ragnar_> http://66.146.166.62/~loki/1-ncurses.err
19:14 < rxr> this happens in the ncurses rebuid with wide char support
19:14 < rxr> can you take a look why the type is undefined there and if you have in from uclibc ?
19:15 < rxr> maybe it just build accidently before because the former patched crutially redefined some CCPFLAGS stuff maybe resulting in the 2nd build beeing a normal (not wide but narrow) ncurses build ...
19:15 < rxr> if you do not have this type we could just disable the ncursesw build for uclibc as I did for dietlibc ...
19:17 < CIA-8> rene * r5955 /trunk/package/security/openssl/openssl.conf: * fixed openssl build for arm (added target to the case ...)
19:18 < _Ragnar_> uhm okay
19:20 < _Ragnar_> rxr: ncurses should define that type itself
19:20 < rxr> ouhm ok
19:21 < _Ragnar_> so this is more likely to mean wrong header files get included...
19:21 < rxr> i take a look
19:22 < _Ragnar_> it depends on _XOPEN_SOURCE_EXTENDED being defined
19:23 < _Ragnar_> which should get defined with _GNU_SOURCE
19:46 * rxr soon taking a lookg
19:46 < rxr> my brother phoned and then some commerial s.th. wanted to sell newspaper abos ...
19:46 < rxr> busy hour ...
19:46 < rxr> when openssl builds I'll take a look into ncurses
19:50 < rxr> yeah
19:50 < rxr> openssl built for arm with dynamic dietlibc ...
19:50 < rxr> + it is a cross build ...
19:51 < rxr> so - now I test some openssl.conf cleanup
20:32 < _Ragnar_> rxr: did you fix the ncurses thingy?
20:32 < rxr> not et
20:33 < rxr> cleaning openssl up
20:34 < _Ragnar_> k
20:35 < rxr> oh my god is openssl huge: 947116
20:36 < rxr> ^- dynamic dietlibc arm
20:36 < rxr> 1188416 <- dynamic glibc x86
20:36 < rxr> so no ssl on my psion ;_)
20:36 < rxr> but good to know we can cross build openssl now ;-)
20:37 < rxr> and the .conf is not as amazingly ugly as it was before ...
20:37 < rxr> == 02/03/05 20:37:16 =[1]=> Finished building package openssl.
20:39 < jsaw> re
20:39 < rxr> hi jsaw
20:39 < jsaw> hi rxr!
20:40 < _Ragnar_> did you patch that for dietlibc, or in general again (will it break uclibc)?
20:41 < rxr> I cleaned it up significantly ...
20:41 < rxr> it is cross build able now
20:42 < rxr> it should not break with other libc's ...
20:42 < rxr> in general I of course do not mangle packages due to dietlibc - all dietlibc junk will go into packge/base/dietlibc/....
20:42 < _Ragnar_> okay ;)
20:43 < rxr> it is just that ncurses was broken in general - and openssl.conf so ugly that I could not sleep well with that in t2
20:45 < _Ragnar_> *lol* I see
20:45 < rxr> as in you just looked into it ?
20:48 < _Ragnar_> what's so ugly about it?
20:48 < rxr> custmain
20:48 < rxr> ugly /etc/ssl symlinking to get files into /usr
20:48 < rxr> and then the stuff it not root / DESTDIR save and just overwrites the bulid system's /usr and so on
20:49 < _Ragnar_> yeah I see now
20:49 < _Ragnar_> that sucks
20:50 < rxr> .oO
20:50 < rxr> we did not install the openssl man files properly before?
20:50 < rxr> !> Found shared files with other packages:
20:50 < rxr> !> usr/share/man/man1/passwd.1: openssl shadow
20:50 < rxr> !> usr/share/man/man3/buffer.3: libowfat openssl
20:50 < rxr> !> usr/share/man/man3/err.3: man-pages openssl
20:50 < rxr> !> usr/share/man/man3/rand.3: man-pages openssl
20:50 < rxr> !> usr/share/man/man3/threads.3: openssl perl
20:51 < rxr> *lol*
20:51 < _Ragnar_> uhm
20:51 < rxr> _Ragnar_: that we did not install the openssl manpages ? *g*
20:51 < _Ragnar_> sure?
20:51 < rxr> openssl: etc/ssl/man
20:51 < rxr> openssl: usr/share/doc/openssl/man
20:51 < rxr> openssl: usr/share/doc/openssl/man/man7
20:51 < rxr> openssl: usr/share/doc/openssl/man/man7/Modes of DES.7
20:51 < rxr> that is openssl on ROCK Linux 2.0.1 (2004/03/13)
20:52 < _Ragnar_> oh *lol*
20:53 < rxr> in my T2 working copy I get dozens installed
20:53 < rxr> and the above conflicts ...
20:54 < rxr> .o(why did I started to clean up openssl - why ...)
20:55 < _Ragnar_> =) good luck
20:56 -!- mnemoc [~amery@200.75.28.126] has joined #t2
20:57 < mnemoc> back
20:57 < rxr> _Ragnar_: know that I spent so much time with openssl - might you liek to review ncurses ?
20:57 < mnemoc> did i miss too much?
20:57 < rxr> _Ragnar_: e.g. review if the uclibc patch added some -I or so that we might like to add
20:57 < rxr> mnemoc: some bashing of the ROCK inherited openssl package and such - not much ...
20:58 < mnemoc> rxr: ncurses patch removes CFLAGS only
20:58 < rxr> my cross all and dietlibc making progress
20:58 < rxr> mnemoc: didn't it hardset some other option, too
20:58 < rxr> I'm sure I saw s.th. like that
20:58 < mnemoc> nope
20:58 < mnemoc> maybe mom
20:59 < _Ragnar_> mnemoc: the .conf does
20:59 < _Ragnar_> but that shouldn't affect it
20:59 < rxr> does or did *g* ?
20:59 < mnemoc> # svn merge -r5947:5946 .
20:59 < mnemoc> does
21:00 < mnemoc> # configure want BUILD_CPPFLAGS=CPPFLAGS which includes $xroot/usr/include, harmfull for BUILDCC
21:00 < mnemoc> export BUILD_CPPFLAGS="-I../include -D_GNU_SOURCE -DNDEBUG"
21:00 < _Ragnar_> yes that defines GNU_SOURCE
21:00 < _Ragnar_> which we need her3e
21:00 < _Ragnar_> here*
21:01 < mnemoc> 7 -BUILD_CCFLAGS = -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) @BUILD_CPPFLAGS@ @BUILD_CFLAGS@
21:01 < rxr> can you rephrase: what need to be readded to get it compile ?
21:01 < mnemoc> 8 +BUILD_CCFLAGS = -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) @BUILD_CPPFLAGS@
21:01 < mnemoc> they are complementary
21:01 < mnemoc> some stuff on @BUILD_CFLAGS@ is harmfull
21:01 < mnemoc> some is necesary
21:02 < mnemoc> so i added the necesary stuff on BUILD_CPPFLAGS and removed BUILD_CFLAGS
21:03 < mnemoc> the comment tells the real problem
21:03 < rxr> can you point me to the exact variable that after my cleanup is a) present that shouod not be - or b) missing ?
21:03 < rxr> == 02/03/05 21:00:31 =[9]=> Finished building package openssl.
21:04 < mnemoc> me? no... i should start a new uclibc build
21:04 < mnemoc> and complete uboot
21:04 < mnemoc> i need some hours to get that point
21:05 < rxr> mnemoc: ok ok
21:05 < rxr> _Ragnar_: ?? *g*
21:05 < mnemoc> :p
21:05 < rxr> this one is big ->
21:05 < CIA-8> rene * r5956 /trunk/package/security/openssl/openssl.conf: (log message trimmed)
21:05 < CIA-8> * deep openssl.conf cleanup:
21:05 < CIA-8> * do not use custmain
21:05 < CIA-8> * do not do ugly symlinking
21:05 < CIA-8> * really install the manpages
21:05 < CIA-8> * be cross copmilable
21:05 < CIA-8> * the forgotten bits
21:06 < rxr> ^- please test ;-)
21:06 < _Ragnar_> I can't right nw
21:06 < _Ragnar_> now*
21:06 < rxr> ok - no problem at all
21:06 < _Ragnar_> rxr: but what seems to be definitely missing from your flags is -D_GNU_SOURCE
21:07 < rxr> _Ragnar_: does it work when you ineject it to the patch
21:07 < rxr> mom
21:08 < mnemoc> rxr: is your patch postconf too?
21:08 < rxr> postconf ?
21:08 < rxr> Makefile.in ...
21:08 < mnemoc> great :p
21:08 < mnemoc> i patched Makefile directly :D
21:08 < rxr> _Ragnar_: does this work:
21:08 < rxr> --- cross-build.patch (revision 5938)
21:08 < rxr> +++ cross-build.patch (working copy)
21:08 < rxr> ...
21:09 < mnemoc> err
21:09 < rxr> -+BUILD_CCFLAGS = -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) -I. -I../include
21:09 < rxr> ++BUILD_CCFLAGS = -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) -DHAVE_CONFIG_H -I../ncurses -I$(srcdir) -I. -I../include -D_GNU_SOURCE
21:09 < mnemoc> rxr: i confused it with ncurses/bash problem
21:10 < _Ragnar_> building...
21:14 < mnemoc> any objections to imap_headercache patch on mutt?
21:14 < jsaw> ?
21:14 < mnemoc> hi jsaw
21:14 < jsaw> hi mnemoc
21:15 < mnemoc> i want to add that third party patch, objections?
21:15 < jsaw> i don't know it - so i simply trust you ;)
21:15 < mnemoc> :)
21:16 < _Ragnar_> rxr: ncurses builds with that patch
21:16 < mnemoc> :)
21:16 < jsaw> ack.
21:16 < jsaw> hi _Ragnar_ btw.
21:17 < rxr> hm -- shoudl I simply commit that hardcoding ... *g* ;-)
21:17 < jsaw> yep.
21:17 < _Ragnar_> yes
21:17 < _Ragnar_> hi jsaw
21:18 < _Ragnar_> does the -D_GNU_SOURCE break dietlibc or glibc?
21:18 < rxr> _Ragnar_: we'll see ;-)
21:18 < CIA-8> rene * r5957 /trunk/package/base/ncurses/cross-build.patch: * define -D_GNU_SOURCE in ncurses BUILD_CCFLAGS to make uclibc happy
21:18 < jsaw> rxr: we have squashfs-tools, but no squashfs in kernel...?
21:19 < rxr> no - we could inject it if we would like to. ...
21:19 < rxr> jsaw: do you play with it ?
21:19 < CIA-8> mnemoc * r5958 /trunk/package/mail/mutt/ (mutt.conf mutt.desc no_autogen_for_headercache.patch):
21:19 < CIA-8> * updated mutt (1.4.2i -> 1.4.2.1i)
21:19 < CIA-8> * improved mutt with imap_headercache patch
21:19 < jsaw> no yet but soon. First I try to get it running with cramfs...
21:19 < rxr> on which arch and hardware ?
21:20 < jsaw> target/distcc, x86/x86
21:21 < jsaw> maybe I should call ist distcccd...
21:21 < jsaw> -s
21:21 < rxr> distcd ;-)
21:21 < rxr> dist-tripple-c-d
21:21 < jsaw> distc^3d
21:21 < jsaw> ....
21:21 < jsaw> distcd is good! I like that.
21:22 < jsaw> I only have to insert an "u" for uclibc somewhere... *g*
21:22 < rxr> so the target name will be distkajdfghlksdjfhcd in the end
21:23 < rxr> encoding any package capital ...
21:23 < jsaw> rxr: btw, reading the squashfs page, seems like it's speed is quite good.
21:23 < jsaw> well I mean the comparison page
21:23 < rxr> it is read-only, you know that ?
21:23 < jsaw> wait...
21:23 < rxr> oehrm
21:24 < jsaw> looks like readonly
21:24 < rxr> jsaw: i only read about it in read-only ontect
21:25 < rxr> the compression write would be quite expensive for the squashfs design ...
21:25 < jsaw> right...
21:26 < rxr> I heavily patch dietlibc right now ...
21:27 < rxr> jsaw: how many mails are on the dietlibc list?
21:27 < rxr> I subscribed yesterday and got none so far ...
21:27 < jsaw> mails are very rare...
21:27 < rxr> .oO
21:28 < rxr> and the archive never worked ... :-(
21:28 < jsaw> last one was 1 Feb
21:28 < rxr> ok - thanks
21:28 < jsaw> hm, now gawk fails.
21:28 < CIA-8> mnemoc * r5959 /trunk/package/ (network/ipvsadm/ security/ipvsadm/): * moved ipvsadm from security/ to network/ because basicly it's a tool to set routes
21:29 < rxr> there is some more stuff to sort
21:29 < rxr> e.g. the extra/ stuff as well as quite a lot of base that is not base ...
21:29 < rxr> == 02/03/05 21:29:14 =[1]=> Finished building package embutils.
21:29 < rxr> ^- ok - more dietlibc dyn fixed (failed due to pivot_root before)
21:29 < rxr> so /me evening dinner
21:29 < rxr> and then maybe preparing patches for fefe
21:29 < rxr> and then there is this libm stuff I should inject into dietlibc ...
21:30 < rxr> and - well - this money earning scanner driver thing :-(
21:30 < _Ragnar_> what did you inject into libm?
21:30 < rxr> dietlibc
21:30 < mnemoc> question: ipvsadm-1.21 is for 2.4 kernels only, ipvsadm-1.24 is for 2.6 kernels only. both kernels include ipvs code. our ipsvadm package has 1.24 and i got a patch to make 1.24 support .old as modutils. what do you propose?
21:30 < rxr> some infrastructure bits to get my arm binaries linked ...
21:31 < mnemoc> ipvsadm-1.24 as ipvsadm2 package? ipvsadm-1.21 as ipvsadm-$foo package?
21:31 < rxr> well - no idea
21:31 < mnemoc> both on ipsvadm package?
21:31 < rxr> I do not even know what the package does ...
21:31 < rxr> mnemoc: you are the maintainer? much fun throwing a coin ;-)
21:31 < mnemoc> :\
21:32 < rxr> if you need both - package both ;-)
21:32 < mnemoc> no more voting?
21:32 < rxr> configure:6152: checking compiler and flags for sanity
21:32 < rxr> configure:6156: error: cannot run test program while cross compiling
21:32 < rxr> ^- this is what openssl screams about ?!?!?!
21:32 < rxr> f*cking package
21:32 * rxr off for evening dinner ...
21:32 < mnemoc> LOL
21:35 < rxr> ok - stupid test patched away ...
21:35 < rxr> ok - zlib missing ...
21:37 < _Ragnar_> brb, food
21:38 < rxr> checking for zlib 1.1.4 or greater... configure: error: cannot run test program
21:38 < rxr> while cross compiling
21:39 < rxr> ok - finally food
21:44 < CIA-8> mnemoc * r5960 /trunk/package/network/ipvsadm2/: * added ipvsadm2 package as a copy of ipvsadm
21:48 < CIA-8> mnemoc * r5961 /trunk/package/network/ipvsadm2/ (6 files): * renamed ipvsadm2 files according to new $pkg name
22:04 -!- mnemoc [~amery@200.75.28.126] has quit [Read error: 110 (Connection timed out)]
22:07 -!- mnemoc [~amery@200.75.27.22] has joined #t2
22:08 < mnemoc> am i really here?
22:08 < daja77> seems so
22:08 < mnemoc> :D
22:09 < mnemoc> good
22:09 < mnemoc> hi daja77
22:09 < mnemoc> rxr: here?
22:09 < daja77> hi mnemoc
22:12 < mnemoc> rxr: i have a postlinux.conf dilema. ipvsadm needs _only_ net/ip_vs.h from kernel sources which is not on /usr/include. i) postlinux.conf to only copy that header? ii) ipvsadm/parse-config to copy that header iii) linux24.conf hardcoded copy of that header. ip_vs.{o,ko} is built always but both kernels.
22:21 -!- ShinGouki [~bla@me-lewiston-pw-627.mint.adelphia.net] has joined #t2
22:21 -!- ShinGouki [~bla@me-lewiston-pw-627.mint.adelphia.net] has left #t2 []
22:41 < _Ragnar_> re
22:42 < mnemoc> re _Ragnar_
22:43 < _Ragnar_> what did I miss?
22:44 < mnemoc> no idea, my connection sticks today
22:46 < rxr> mnemoc: ouohm - can I think about that later - e.g. tomorrow ?
22:46 < CIA-8> mnemoc * r5962 /trunk/package/network/ipvsadm2/pkg_linux-header_pre.conf: * added hook to install kernel's net/ip_vs.h if ipvsadm2 will be installed
22:47 < mnemoc> rxr: ok
22:47 < rxr> I see you already decided ... :-)
22:47 < mnemoc> the method can be changed
22:47 < mnemoc> i only want it building
22:52 < rxr> why the h*ll does openssh want to run so many test programs ...
22:52 * rxr headshake
22:57 < rxr> I give up to cross compile openssh shit right now
22:57 < rxr> I would not have imagined that this OpenBSD prime time project has such an ugly configure ...
22:58 < mnemoc> ipvsadm for 2.4 doens't build against 2.6 headers :(
22:59 < _Ragnar_> *g*
22:59 < rxr> inject the needed headers into the ipvsadm package ...
22:59 < rxr> using -I $PWD/include
22:59 < rxr> and drop the needed headers into it ...
22:59 < mnemoc> ok
23:00 < mnemoc> i'll do that for 2.4
23:00 < mnemoc> keepalived also needs it so i prefer 2.6's on /usr/include
23:00 -!- praenti [~praenti@82.139.198.99] has quit [Read error: 104 (Connection reset by peer)]
23:00 -!- praenti [~praenti@82.139.198.99] has joined #t2
23:17 < CIA-8> mnemoc * r5963 /trunk/package/network/ipvsadm/ (inject_2.4.29_ip_vs.h.patch ipvsadm.desc): * downdated ipvsadm (1.24 -> 1.21-11) to be ipvsadm for 2.4 kernels
23:39 < CIA-8> mnemoc * r5964 /trunk/package/network/ (6 files in 2 dirs): * improved ipvsadm and ipvsadm2 to cohexist using .old suffix on ipvsadm tools when ipvsadm2 is installed
23:59 < CIA-8> mnemoc * r5965 /trunk/package/ (network/keepalived/ security/keepalived/): * moved keepalived from security/ to network/
--- Log closed Fri Feb 04 00:00:48 2005