--- Log opened Wed Jan 04 00:00:42 2006
00:16 < mnemoc> rehi!
00:17 < karasz> Ohayoo gozaimasu mnemoc san
00:17 < morfoh> wb mnemoc :)
00:18 < mnemoc> karasz: :)
00:18 < mnemoc> thx morfoh
00:26 < karasz> oki build in stage [5] now
00:26 < karasz> so i guess i'll go to sleep
00:27 < karasz> tomorrow i will make a bootdisk build and isos
00:27 < karasz> iff all goes well by tomorrow i can move in :D
00:27 < morfoh> karasz: sleep well ... cu tomorrow ;)
00:27 < karasz> arigato morfoh san
00:28 < karasz> same 2u2
00:28 < karasz> :P
00:28 < morfoh> :p
00:28 < karasz> cu guys
00:28 < karasz> bye
00:28 < morfoh> cu karasz san :p
00:39 < mnemoc> OT: is mingw a gcc toolchain for un*x to build win32 binaries, or a gcc toolchain to run on win32 platform?
01:00 -!- nzg [n=tschmidt@p54AFDBE1.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)]
01:14 < CIA-1> amery * r14465 /trunk/package/archiver/bzip2/ (bzip2.conf bzip2.desc fix_shared.patch makefile.patch): * fixed libbzip2 under dietlibc, and copyright notes updated
01:15 < morfoh> thx ;)
01:16 < demian> bye
01:16 < morfoh> cu demian
01:17 < mnemoc> cu demian
02:19 < mnemoc> # ls | sed -e "s/f//g" | wc -l
02:19 < mnemoc> 21
02:19 < mnemoc> # ls | sed -e "s/^f//g" | wc -l
02:19 < mnemoc> sed: -e expression #1, char 7: Memory exhausted
02:19 < mnemoc> 0
02:19 < mnemoc> sweet
02:19 < morfoh> # ls | sed -e "s/f//g" | wc -l
02:19 < morfoh> 18
02:19 < morfoh> dietlibc + minised
02:20 < mnemoc> and ^f ?
02:20 < morfoh> ls | sed -e "s/^f//g" | wc -l
02:20 < morfoh> 18
02:20 < mnemoc> and normal sed?
02:21 < morfoh> any other wishes ? :p
02:21 < mnemoc> nope
02:21 < mnemoc> just that
02:21 < morfoh> haha
02:21 < mnemoc> :)
02:22 < morfoh> :)
02:23 < morfoh> ... waiting ...
02:24 < morfoh> # ls | sed -e "s/f//g" | wc -l
02:24 < morfoh> 18
02:24 < morfoh> # ls | sed -e "s/^f//g" | wc -l
02:24 < morfoh> sed: -e expression #1, char 7: Memory exhausted
02:24 < morfoh> 0
02:25 < morfoh> on embedded target from trunk HEAD
02:25 < mnemoc> so it's not me
02:26 < morfoh> unfort. not :p
02:26 < mnemoc> :)
02:27 < morfoh> minised ?
02:28 < mnemoc> doens't sound like the right fix
02:29 < morfoh> even not for a dietlibc target ?
02:31 < CIA-1> amery * r14466 /trunk/package/mail/vmailmgr/ (6 files): * added vmailmgr (0.97) - Simple virtualizing for qmail
02:31 < mnemoc> dietlibc doesn't mean 'embedded'
02:31 < mnemoc> it can be 'just light'
02:31 < mnemoc> anyway i doubt minised is ready for all our sed expressions
02:32 < mnemoc> like i doubt ash is ready for running out scripts
02:33 < morfoh> god s(h)ave bash :p
02:35 < mnemoc> F* strace doesn't build for dietlibc
02:39 < morfoh> same here
02:44 < mnemoc> try this: for x in 1 2 3 4 5 6 7; do echo $x; done | sort -n
02:45 < morfoh> no sort
02:46 < mnemoc> embedded boy :(
02:47 < morfoh> skript kiddie :p
02:50 < mnemoc> we need 'sort' to enter stage 2
02:50 < mnemoc> it's not a matter of targets
02:50 < mnemoc> or package selection
02:51 < morfoh> so we depend on coreutils
02:52 < mnemoc> yep
02:54 < mnemoc> cool.... and gdb doesn't build on stage1
03:05 < morfoh> sort: write error: Bad file number
03:06 < mnemoc> yeah!
03:06 < mnemoc> i love it :)
03:06 * mnemoc trying to get strace built
03:06 < morfoh> I already recognize it ;)
03:10 < mnemoc> stupid fefe's ML
03:10 < mnemoc> it doesn't accept plain text attachments
04:13 -!- ojh [n=omer@71-32-220-34.eugn.qwest.net] has joined #t2
04:16 < ojh> Hi. I seem to be having some hardware detection problems on 2.1.0
04:19 < ojh> Error from messages 'Jan 3 20:13:39 localhost hal.hotplug[2838]: DEVPATH is not set (subsystem pci)'
04:20 < mnemoc> uhm
04:22 < ojh> mnemoc: The thing that is not working is my BTTV capture card. The modules are loaded in the lernel but nothing is in /dev
04:22 < ojh> mnemoc: This error from HotPlug looks like the problem.
04:22 < mnemoc> how 'nothing'?
04:23 < ojh> mnemoc: No /dev/v4l/ . . . nothing.
04:24 < ojh> mnemoc: But if I 'lsmod' everything seems to be loaded just fine.
04:25 < mnemoc> uhm
04:25 < ojh> mnemoc: And it does set up all automagicly on ROCK 2.0.0
04:25 < mnemoc> uhm^2
04:26 < mnemoc> what target?
04:26 < ojh> mnemoc: Installed from the normal T2 Desktop ISOs
04:26 < mnemoc> uhm^3
04:26 < ojh> LOL
04:27 < ojh> mnemoc: All my other hardware works perfectly :-/
04:28 < mnemoc> can you pastebin (http://www.rafb.net/paste/) your var/log/message ? i can't review now (too sleepy) but tomorrow
04:29 < ojh> Sure
04:29 -!- morfoh [n=jeru@p54BEE216.dip.t-dialin.net] has quit ["using sirc version 2.211+KSIRC/1.3.11"]
04:38 < ojh> mnemoc: Thanks. You are a big part of the support success of T2, this IRC channel has been invaluable when I have a problem.
04:38 < mnemoc> :)
04:38 < mnemoc> but i'm a tired part
04:41 < ojh> mnemoc: Get enough users of T2 and send all the IRC logs to O'Riley so they can write a book or three on T2. What animal would that have on the cover?!?!?! :-)
04:41 < mnemoc> a caterpillar :)
04:41 < ojh> :-)
05:26 -!- ojh [n=omer@71-32-220-34.eugn.qwest.net] has left #t2 []
05:49 -!- mipe [n=mipe@dsl10040.japo.fi] has joined #t2
05:56 < mipe> hmm, any news about libstdc++.la thing? does the patch really work... libstdc++.la looks bit empty on dependency_libs section
06:45 < mipe> hmm, again i get build/desktop-*-desktopusr/
06:48 < mipe> and the dietlibc posix utime et all fix breaks dietlibc for atleast x64
07:10 < rxr> re
07:11 < rxr> mipe: there is nothing libstdc++ should need more
07:11 < rxr> mipe: looks good so far
07:11 < rxr> I have not yet seen dietlibc failing but I'll give it a spin
07:16 < mipe> all i have in libstdc++.la (dependency_libs) is -lm, but yes, its working so far...
07:16 < rxr> lm is the only thing needed ... -lc and -lgcc and such are default anyway ...
07:17 < rxr> that is you do not hand them to gcc when compiling anyway ...
07:17 < rxr> thus this is fine
07:17 < mipe> rxr: after removing the dietlibc posixfy_utime_execl_execle.patch it builds normally...
07:17 < mipe> yep i know
07:17 < mipe> dietlibc error was some conflicting types for 'utime
07:18 < mipe> +s'
07:20 < rxr> hm
07:20 < rxr> !> libcompat/utimes.c:10: error: conflicting types for 'uti ..
07:20 < rxr> !> include/sys/time.h:78: error: previous declaration of 'u ..
07:20 < LMJ> morning rxr
07:21 < mipe> also shadow failing on stage 3, but thats because some pam include files missing...
07:22 < rxr> patch is tiny
07:22 < rxr> just one more const
07:22 < rxr> mipe: yes - I noticed this last nicht
07:22 < rxr> night
07:22 < rxr> jsaw did a in-system emerge to test
07:23 < rxr> the new pam does install includes wrong
07:23 < rxr> woudl be cool if you cold take a look
07:24 < rxr> the dietlibc fix gets commited in a minute when it pased a test build
07:27 < CIA-1> rene * r14467 /trunk/package/base/dietlibc/posixfy_utime_execl_execle.patch:
07:27 < CIA-1> * fixed dietlibc/posixfy_utime_execl_execle.patch by correcting another
07:27 < CIA-1> define of utimes
07:55 < rxr> cu
08:25 < valentin> moin
08:26 < LMJ> moin valentin
08:26 < valentin> hi LMJ
08:26 < LMJ> how are you ?
09:00 < rxr> re
09:02 < rxr> moin valentin
09:06 < valentin> moin rxr
09:06 < valentin> LMJ: just fine
09:06 < valentin> how are you rxr and LMJ
09:07 < LMJ> fine thx here
09:08 < valentin> so, i'll take a short look what's wrong with the fonts....
09:12 < rxr> heh
09:12 < rxr> look at the google logo today
09:12 < LMJ> hehe, i was watching it too
09:12 < LMJ> you was faster rxr ;)
09:13 < rxr> valentin: what kinda font problem do you have ?
09:13 < valentin> no font seems to work
09:14 < valentin> i only have [] everywhere in X
09:14 < rxr> the system font path is missing in /etc/font/fonts.conf then ..
09:14 < valentin> yes, that is what i thought
09:14 < rxr> /usr/share/fonts
09:14 < rxr>
09:14 < rxr> ~/.fonts
09:14 < rxr> the empty line should read
09:14 < rxr> /usr/X11/lib/X11/fonts
09:14 < rxr> or so
09:14 < rxr> need to review why this happens ...
09:15 < rxr> bug or so i the modular fontconfig/configure or so ...
09:15 < rxr> add the line and run the 80-xorg or so postinstall or just
09:15 < rxr> fc-cache -vs
--- Log opened Wed Jan 04 09:22:22 2006
09:22 -!- Irssi: Join to #t2 was synced in 13 secs
09:22 < rxr> so
09:22 < rxr> damn - this cable was loose ...
09:22 < rxr> should not manage to do this more often ...
09:23 < mipe> scripts/functions.in: line 266: /usr/lib/security/pam_xauth.la: No such file or
09:23 < mipe> directory
09:23 < mipe> Verifing the .la files ...
09:23 < mipe> grep: lib/libpam.la: No such file or directory
09:24 < rxr> hm
09:24 < mipe> pam should be working now, i could commit the patch soon... fixing include directories
09:24 < mipe> except for those above errors
09:25 < mipe> eeer warnings even
09:25 < rxr> mipe: did you fix pam to install into the correct directory again ?
09:25 < mipe> include files, I didnt notice that other files where in incorrect directory?
09:26 < rxr> mipe: yeah I just wanted to know whether you tweaked the #include directives or the file installation ...
09:35 < valentin> *brunch*
09:41 -!- menomc [n=amery@200.75.27.14] has joined #t2
09:42 -!- mnemoc [n=amery@200.75.27.99] has quit [Nick collision from services.]
09:42 < mipe> hmm
09:43 < mipe> pam fix might be just simpler to change includedir=$root/usr/include/security, so all the header files are there. so no need to patch all Makefile.in in pam subdirs
09:43 < mipe> and looks like from my running system all the pam headers are there
09:44 -!- menomc is now known as mnemoc
09:45 < mipe> eek
09:47 < mipe> svn segfauls on commit
09:47 < mipe> 2.1.1 from iso few days ago
09:52 < rxr> ouh
09:54 < mipe> does subversion really need java jdk?
09:54 < mipe> is listed on cache
09:54 < rxr> optional dep on java bindings ?
09:55 < mipe> hmm, might be
10:01 < CIA-1> rene * r14468 /trunk/package/x11/freetype1/freetype1.desc:
10:01 < CIA-1> * marked freetype1 AUTOMAKE-QUIRK, at least powerpc64 is not in the
10:01 < CIA-1> outdated config.{guess,sub}
10:18 < rxr> http://www.kernel.org looks forbidden ...
10:18 < Baldzius> :)
10:18 < Baldzius> good morning rxr :)
10:20 < rxr> hi Baldzius
10:38 < jsaw> re (just for a second)
10:39 < jsaw> rxr: mozilla builds on x86-64, but is heavily unstable
10:39 < jsaw> :(
10:40 < jsaw> rxr: apropos fonts.conf, /usr/X11/share/fonts is what I used in xproto.conf yesterday, should it have been /usr/share/fonts ?
10:40 < rxr> jsaw: well other dists use /usr/share/fonts ...
10:40 < rxr> IIRC
10:41 < rxr> maybe we should migrate to that as well ... ?
10:41 < rxr> jsaw: in the past mozilla on x86-64 worked normally ...
10:41 < rxr> how do you define a unstable mozilla ?
10:43 < jsaw> segfaulting, even on t2 homepage...
10:43 < jsaw> but now always
10:43 < jsaw> s/now/not/
10:44 < jsaw> bbl
10:46 < rxr> hm
10:46 < rxr> ok - I have not tried firefox 1.5 on x86-64 yet ...
10:46 < rxr> maybe it is due that ...
10:51 < rxr> eiiiiii
10:51 < rxr> damn shilly
10:56 < rxr> /var/adm/flists/cdrtools:cdrtools: usr/include/align.h
10:56 < rxr> wonderful
10:56 < rxr> not to mention his makefiles rules stink ...
10:56 < rxr> and yes I try to cross build it right now ...
11:03 < rxr> and the makefiles just finishes without a proper exit code due his hackery ...
11:03 < rxr> even on errors ...
11:05 < rxr> and his code uses functions I never saw before ...
11:05 < rxr> gcvt - convert a floating-point number to a string
11:05 < rxr> NOTES
11:05 < rxr> This function is obsolete. Instead, sprintf() is recommended.
11:05 < rxr> bummer ..
11:26 < mipe> anyone up and running 2.1/2.1.1 on amd64 with subversion (commit) working?
11:26 < mipe> rebuilded subversion+deps, but didnt help... :(
11:27 < mipe> and does someone have t2 image that could be used for usb booting and later installing t2
11:29 < rxr> mipe: the latest initrd stuff supports booting from USB
11:29 < rxr> the new install/ stuff is tested to load from USB fine
11:29 < rxr> the archivista target as well
11:29 < rxr> (at least weeks before it got broken by all the latest changes)
11:29 < mipe> ok thanks
11:30 < rxr> the initrd generated is able to wait for usb root devices to show up and udevd will create /dev notes for USB on the fly as need (inside the initrd)
11:30 < rxr> my geode even boots/runs from a usb flash
11:30 < rxr> though technically it is not different from usb hard-disk
11:31 < mipe> rxr: have you runned svn commit in system under 2.1-branch?
11:32 < rxr> I only built 2.1 on the box
11:33 < karasz> moin all
11:33 < rxr> I run trunk
11:33 < rxr> trunk commit works ...
11:39 < karasz> bild not finished...??
11:41 < rxr> karasz: x86-64 on my or your side ?
11:41 < karasz> mine
11:42 < karasz> i build a stable desktop...
11:42 < rxr> from my 1000 packages in a desktop build only 19 failed so far
11:42 < rxr> trunk that is
11:43 < karasz> i have 2 failed on stable...
11:44 < karasz> so far
11:44 < karasz> where is the dld failures file?
11:48 -!- LMJ [n=serwou@laf31-2-82-224-107-105.fbx.proxad.net] has quit [Read error: 104 (Connection reset by peer)]
11:48 < karasz> transfig and transset failed...
11:48 < karasz> but i guess it is a x86-64 issue
11:49 -!- LMJ [n=serwou@laf31-2-82-224-107-105.fbx.proxad.net] has joined #t2
11:49 < karasz> where is the download failures log?
11:57 < CIA-1> rene * r14469 /trunk/scripts/config.in:
11:57 < CIA-1> * removed the cpus*2 that we only had to get an inital testing of
11:57 < CIA-1> parallel build on all builds, now that nearly all NOPARALLEL packages
11:57 < CIA-1> should be marked, just using a PARALLEL_MAX=$cpus is more optimal
12:01 -!- mtr_ [n=Michael@p54AF8B9A.dip0.t-ipconnect.de] has joined #t2
12:09 < mipe> hmm, anyone know if pinnacle (digi)-tv cards work on linux?
12:11 < rxr> depends on the chip
12:11 < rxr> guess some work and some don't ...
12:14 < mipe> hmm, easy to way specify svn commit log file?
12:15 < rxr> you mean the message to use ?
12:15 < mipe> ah, found it
12:15 < rxr> --file
12:15 < mipe> yep
12:15 -!- mtr [n=Michael@p54AFC4C6.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)]
12:21 < karasz> Downloading transset:download/mirror/t/transset-2005-04-01.tar.bz2 ...
12:21 < karasz> Finished downloading 0 bytes in 0.236 seconds (0.000 bytes/sec).
12:21 < karasz> curl: (22) The requested URL returned error: 404
12:21 < karasz> ERROR: CURL Returned Error 22. Please read the curl manpage.
12:21 < karasz> INFO: download from mirror failed, trying original URL.
12:21 < karasz> cvs -z4 -Q -d :pserver:anoncvs@freedesktop.org:/cvs/xapps co -P -D 2005-04-01 transset
12:21 < karasz> Error during checkout.
12:21 < karasz> ???
12:24 < rxr> debug-cdrtools:[i386-linux-cc]# file cdrecord
12:24 < rxr> cdrecord: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), statically linked, stripped
12:24 < rxr> -rwxr-xr-x 1 root root 285269 Jan 4 13:23 cdrecord
12:25 < rxr> `- dietlibc with grave manual hacks cross compiled ...
12:33 -!- rxr_ [n=rene@e178171255.adsl.alicedsl.de] has joined #t2
12:33 -!- Topic for #t2: T2 | 2.1.1 TAGGED and 2.2.0-epsilon RELEASED! | The System Development Environment (SDE) | http://www.t2-project.org/ | Give us a greeting and don't hesitate to ask any questions you have :-)
12:33 -!- Topic set by mnemoc [] [Mon Jan 2 15:21:42 2006]
12:33 [Users #t2]
12:33 [@ChanServ] [ CIA-1 ] [ karasz] [ mnemoc] [ rxr ] [ zod ]
12:33 [ axion ] [ demian] [ LMJ ] [ mtr_ ] [ rxr_ ] [ _Ragnar_]
12:33 [ Baldzius] [ jsaw ] [ mipe ] [ owl ] [ valentin]
12:33 -!- Irssi: #t2: Total of 17 nicks [1 ops, 0 halfops, 0 voices, 16 normal]
12:33 -!- Channel #t2 created Sun Aug 8 19:15:33 2004
12:33 -!- [freenode-info] if you need to send private messages, please register: http://freenode.net/faq.shtml#privmsg
12:33 -!- Irssi: Join to #t2 was synced in 31 secs
12:48 -!- rxr [n=rene@e178164226.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)]
12:51 < jsaw> re
12:55 < karasz> moin jsaw
12:55 < jsaw> hi karasz
12:55 < rxr_> hi jsaw
12:55 < karasz> moin new rxr_
12:56 -!- madtux [i=miguel@pf0.hostarica.com] has joined #t2
12:57 < rxr_> hola miguel
12:57 -!- You're now known as rxr
13:01 -!- rxr_ [n=rene@e178178050.adsl.alicedsl.de] has joined #t2
13:01 -!- Topic for #t2: T2 | 2.1.1 TAGGED and 2.2.0-epsilon RELEASED! | The System Development Environment (SDE) | http://www.t2-project.org/ | Give us a greeting and don't hesitate to ask any questions you have :-)
13:01 -!- Topic set by mnemoc [] [Mon Jan 2 15:21:42 2006]
13:01 [Users #t2]
13:01 [@ChanServ] [ CIA-1 ] [ karasz] [ mipe ] [ owl ] [ valentin]
13:01 [ axion ] [ demian] [ LMJ ] [ mnemoc] [ rxr ] [ zod ]
13:01 [ Baldzius] [ jsaw ] [ madtux] [ mtr_ ] [ rxr_] [ _Ragnar_]
13:01 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
13:01 -!- [freenode-info] please register your nickname...don't forget to auto-identify! http://freenode.net/faq.shtml#nicksetup
13:01 -!- Channel #t2 created Sun Aug 8 19:15:33 2004
13:02 -!- Irssi: Join to #t2 was synced in 32 secs
13:02 < rxr_> re
13:03 < jsaw> hi rxr_
13:07 -!- rxr [n=rene@e178171255.adsl.alicedsl.de] has quit [Read error: 60 (Operation timed out)]
13:11 -!- You're now known as rxr
13:15 < rxr> jsaw: did you build with as-needed?
13:15 < jsaw> no
13:15 < rxr> ah
13:15 < jsaw> (given up on it)
13:15 < rxr> can you test if mozilla is more stable ?
13:15 < jsaw> you mean, with as-needed?
13:15 < rxr> you saied firefox is instable
13:16 < rxr> just meant is mozilla more stable?
13:16 < jsaw> oh, firefox doesn't build :(
13:16 < jsaw> or didn't
13:16 < rxr> oh - so mozilla is unstable ...
13:17 < jsaw> sorry, yes, mozilla segfaults
13:17 < rxr> hm - libtool still fails
13:17 < rxr> despite fixed gcc ...
13:18 < jsaw> ?
13:18 < rxr> in my ppc64 build
13:18 < rxr> pcre sill fails linking against the wrong libstdc++
13:18 < jsaw> O.O
13:19 < rxr> on your side on x86-64 as well ?
13:19 < jsaw> I just started a fresh build this morning, I can tell in a few hours
13:19 < jsaw> updating shadow now...
13:19 < rxr> h
13:19 < rxr> m
13:19 < rxr> on my x86-64 box libtool built thru ...
13:22 < karasz> usualy libtool builds...
13:23 < rxr> pcre that is ...
13:23 < karasz> when my pcre failed i have not problems buildin libtool
13:23 < rxr> yes I meant pcre ..
13:24 < rxr> libtool just slippted into the channel because I typed it a lot these days ...
13:24 < Baldzius> :)
13:26 < jsaw> fgrep '+++' shadow-4.0.13-4.0.14.diff
13:26 < jsaw> +++ shadow-4.0.14/ChangeLog 2006-01-03 17:49:30.000000000 +0100
13:26 < jsaw> +++ shadow-4.0.14/Makefile.in 2006-01-03 17:12:34.000000000 +0100
13:26 < jsaw> +++ shadow-4.0.14/NEWS 2006-01-03 17:45:47.000000000 +0100
13:26 < jsaw> +++ shadow-4.0.14/README 2005-12-14 18:10:07.000000000 +0100
13:26 < jsaw> +++ shadow-4.0.14/aclocal.m4 2006-01-03 17:12:16.000000000 +0100
13:26 < jsaw> +++ shadow-4.0.14/configure 2006-01-03 17:12:36.000000000 +0100
13:26 < jsaw> +++ shadow-4.0.14/configure.in 2006-01-03 00:51:44.000000000 +0100
13:26 < jsaw> tststs
13:31 < rxr> this is all ?
13:31 < jsaw> yeah...
13:32 < rxr> btw. do you know diffstat ? or just not installed yet ? ,-)
13:32 < jsaw> the latter :)
13:33 < jsaw> i was shocked myself, no lsdiff around...
13:34 < jsaw> ohoh - too early in the morning -
13:34 * jsaw is stupid
13:35 < jsaw> note to self: diff doesn't recurse into subdirs unless you tell it to do so
13:35 < rxr> ah - yep
13:36 * jsaw buries his hat (without taking it off)
13:37 < Baldzius> :)
13:38 < jsaw> rxr:
13:38 < jsaw> -#include
13:38 < jsaw> -#include
13:38 < jsaw> +#include "pam_defs.h"
13:38 < jsaw> seems like the new shadow might solve the compilation problems... (me hoping)
13:39 < rxr> those pld people stink
13:39 < jsaw> ?
13:39 < jsaw> you can smell that even in Berlin?
13:39 < rxr> why the hack do they need to alter install location in minor versions ????
13:40 < rxr> and thy the hell do they need to hack those two packages to death ?
13:40 < jsaw> because somebody had a good idea!
13:41 < rxr> is there s.th. in the NEWS or so file about this ?
13:41 < jsaw> or not...
13:41 < jsaw> sec
13:42 < jsaw> not really
13:43 < rxr> and in pam ?
13:45 < jsaw> not really...
13:50 < jsaw> hmmm. now 1-linux26 doesn't build anymore
13:50 < mipe> i just added includedir=$root/usr/include/security pam.conf to get shadow built
13:50 < mipe> on my running system all pam headers are there
13:51 < jsaw> new shadow doesn't need that...
13:51 < jsaw> probably ;|
13:51 < jsaw> I hope
13:51 < mipe> damm glibc fails cross :(
13:52 < CIA-1> rene * r14470 /trunk/package/base/libtool/parse-config: * commented and copyrighted the libtool/parse-config
13:52 < rxr> mipe: built here last night for x86-64 and ppc64 ...
13:52 < rxr> mipe: which arch which t2 flavour ?
13:55 < jsaw> okay, shadow needs the pam-changed-include-path patch :(
13:55 < jsaw> gr
13:57 < jsaw> hmmmm....
13:57 < mipe> rxr: trunk generic-minimal-desktop from x86-64 to via c3-2
13:57 < rxr> hm
13:57 < rxr> what is the build error ?
13:57 < rxr> jsaw: maybe better patch pam to install the files where they have been since ages ?
13:57 < jsaw> actually, linux-pam is not wrong, it's the installation that screws it
13:58 < jsaw> rxr: linux-pam didn't change it
13:58 < jsaw> there's some makefile interference...
13:58 < rxr> ah - liek libdir= or so ?
13:58 < mipe> rxr:conflicting types for __lll_lock_wait
13:58 < rxr> suggested that last night ...
13:58 < rxr> mipe: hm
13:58 < jsaw> rxr: trying to read the autocrap genereted makefile....
13:58 < rxr> most probably just that one optmization setting that screws it ...
13:59 < rxr> .oO
13:59 < CIA-1> rene * r14471 /trunk/package/base/libtool/parse-config:
13:59 < CIA-1> * forgot that libtool/parse-config is a shell script not a patch,
13:59 < CIA-1> fixed comments to be properly quoted ...
13:59 < rxr> and why I should test comment changes ...
13:59 < rxr> my oh my ...
14:02 < mipe> rxr:any suggestions what is wrong with cross build?
14:05 < jsaw> I take mipe solution for pam... it's actually a thinko about how Makefile.am works (imo)
14:05 < jsaw> *gah*
14:05 < CIA-1> rene * r14472 /trunk/target/archivista/rootfs/home/archivista/login-setup.sh:
14:05 < CIA-1> * fixed archivista/login-setup.sh to stop/start apache instead of restart so
14:05 < CIA-1> mod_perl is not reloaded otherwise
14:10 < jsaw> rxr: I should remind you testing the input-match.cc change ;)
14:11 < CIA-1> rene * r14473 /trunk/target/archivista/rootfs/home/archivista/login-setup.sh:
14:11 < CIA-1> * fixed archivista/login-setup.sh to properly exit on cancel and not
14:11 < CIA-1> silently write empty strings into the config
14:12 < rxr> jsaw: yes and the slides
14:12 < jsaw> :)
14:12 < jsaw> hotplug is more important ;)
14:12 < jsaw> right now
14:12 < jsaw> I mean
14:13 < rxr> doing some archivista work to make sure they are happy and I have money also ...
14:13 < rxr> too bad I can not reproduce some of their filled bugs ...
14:14 < CIA-1> jsaw * r14474 /trunk/package/base/pam/ (pam.conf pam.desc):
14:14 < CIA-1> * fix include installation (following mipe's suggestion, setting includedir)
14:14 < CIA-1> * add documentation installation
14:20 < mipe> huoh, glibc bug tracking page had a report of lll_ll* bug and was marked as duplicate report, but dammed no proper duplicate bug id
14:21 < mnemoc> moin
14:21 < karasz> moin mnemoc
14:22 < mipe> YES!! pcre built
14:23 < mnemoc> moin karasz, mipe, rxr, ks
14:23 < mnemoc> s/ks/jsaw/
14:23 < karasz> mnemoc, i succeded to break stable... :D
14:25 < mipe> hmm, looks like nptl doesnt build for via... thats the reason for lll_ll_wait crap
14:25 < jsaw> *gaaaaaaaaaaaaaaaaaaaah*
14:25 < mnemoc> karasz: kudos!
14:25 < karasz> :D
14:25 < karasz> that's all?
14:26 < mipe> nptl is for i486 or higher only... hmmm
14:26 < mnemoc> rxr: coreutils' sed fails on '^'
14:26 < karasz> i thought that at least i will get slapped or some...
14:27 < rxr> karasz: ?
14:27 < mnemoc> i have no know what you did first :)
14:27 * mnemoc doesn't slap for free
14:27 < rxr> mnemoc: ah - thanks - acked
14:27 < rxr> tested some single sed expresions in the morning and they worked
14:27 < rxr> indeed ^ is a culprit
14:28 < rxr> let's see if I find some time to track this today ...
14:28 < mnemoc> but sort fails on all my tests
14:28 < mnemoc> do you have a patch to get strace built?
14:28 < rxr> mnemoc: with ugly hackery cdrools can be cross build - even with dietlibc ..
14:28 < karasz> mnemoc,
14:28 < karasz> install -c fig2dev /usr/X11R6/bin/fig2dev
14:28 < karasz> + mkdir -p /usr/lib64/X11/xfig/bitmaps
14:28 < karasz> mkdir: cannot create directory `/usr/lib64/X11': File exists
14:28 < karasz> Copying bitmap files for tk to /usr/lib64/X11/xfig/bitmaps
14:28 < karasz> install: cannot create regular file `/usr/lib64/X11/xfig/bitmaps': No such file or directory
14:28 < karasz> make[1]: *** [install] Error 1
14:28 < karasz> make[1]: Leaving directory `/TOOLCHAIN/src.transfig.desktop.20060104.061851.8136.karasz/transfig.3.2.4/fig2dev'
14:28 < karasz> make: *** [install] Error 2
14:28 < mnemoc> rxr: :D
14:28 < karasz> Due to previous errors, no 5-transfig.log file!
14:28 < rxr> mnemoc: nope - not tried yet ...
14:28 < karasz> transfig
14:28 < rxr> mnemoc: but cdtools is ugly - I do not commit hackery for it yet
14:29 < rxr> well - at least not the evil bits ...
14:29 < mnemoc> rxr: np
14:29 < mnemoc> karasz: file build/desktop-*/usr/lib64/X11
14:30 < CIA-1> rene * r14475 /trunk/package/base/cdrtools/cdrtools.conf:
14:30 < CIA-1> * fixed cdrtools to use the specified compiler for linking (not
14:30 < CIA-1> gcc hardcoded) - improves cross compiling, but due to ugly config
14:30 < CIA-1> neightmare tried to build with a wrong config - fixes for that at a
14:30 < CIA-1> later time
14:30 < karasz> build/desktop-2.1-stable-x86-64-athlon64-desktop/usr/lib64/X11: symbolic link to `../X11/lib64/X11'
14:31 -!- morfoh [n=jeru@p54BEF73E.dip.t-dialin.net] has joined #t2
14:31 < karasz> but the error that intrigues me the most is transet\
14:31 < karasz> even transset
14:31 < mnemoc> karasz: that symlink is broken!
14:31 < mnemoc> build/desktop-2.1-stable-x86-64-athlon64-desktop/usr/lib64/X11/../X11/lib64/X11
14:31 < morfoh> hi *
14:31 < mnemoc> hi morfoh
14:31 < karasz> moin morfoh
14:32 < morfoh> hi mnemoc + karasz ;)
14:32 < mnemoc> karasz: grep 'usr/lib64/X11$' build/desktop-2.1-stable-x86-64-athlon64-desktop/var/adm/flist/*
14:32 < mnemoc> rxr: did you try dtach?
14:32 < karasz> grep: build/desktop-2.1-stable-x86-64-athlon64-desktop/var/adm/flist/*: No such file or directory
14:32 < mnemoc> flists
14:33 < karasz> build/desktop-2.1-stable-x86-64-athlon64-desktop/var/adm/flists/xorg:xorg: usr/lib64/X11
14:33 < mnemoc> rm build/desktop-2.1-stable-x86-64-athlon64-desktop/usr/lib64/X11
14:33 < mnemoc> mkdir build/desktop-2.1-stable-x86-64-athlon64-desktop/usr/lib64/X11
14:33 < mnemoc> until i find the bug
14:34 < karasz> but transset ...
14:34 < karasz> Downloading transset:download/mirror/t/transset-2005-04-01.tar.bz2 ...
14:34 < karasz> Finished downloading 0 bytes in 0.236 seconds (0.000 bytes/sec).
14:34 < karasz> curl: (22) The requested URL returned error: 404
14:34 < karasz> ERROR: CURL Returned Error 22. Please read the curl manpage.
14:34 < karasz> INFO: download from mirror failed, trying original URL.
14:34 < karasz> cvs -z4 -Q -d :pserver:anoncvs@freedesktop.org:/cvs/xapps co -P -D 2005-04-01 transset
14:34 < karasz> Error during checkout.
14:35 < karasz> i even checked out manualy t2 and the file is not there...
14:36 < morfoh> karasz: I thought you are running a stable system now ... what the hack you've done to screw it up again :p
14:36 < karasz> :D)))
14:36 < karasz> just buildin' stable
14:36 < morfoh> oh! you just smiled at your computer ?
14:36 < karasz> and broke it as usual...:D)))
14:37 < mnemoc> karasz: karasz UTFM
14:37 < karasz> UTFM?
14:38 < mnemoc> use the fine mirrors
14:38 < morfoh> perhaps mnemoc broke it ... and you're just the little test mouse ...
14:38 * karasz renames itself: karasz The Breaker
14:38 < karasz> i was using gsmp...
14:40 < mnemoc> http://t2.geeks.cl/mirror/2.1-stable/t/transset-2005-04-01.tar.bz2
14:41 < rxr> mnemoc: I have not tried detach
14:41 < rxr> should I ?
14:41 < mnemoc> rxr: you were looking for a simpler 'screen'
14:41 < karasz> mnemoc, and what about the chksum?
14:41 < karasz> i guess i shouyld make it manualy or some...
14:41 < CIA-1> rene * r14476 /trunk/package/network/dhcp/rocknet_dhcp.sh:
14:41 < CIA-1> * fixed dhcp rocknet module to properly flush the ip address when the
14:41 < CIA-1> interface goes down, so it does not interfere with a new one when it is
14:41 < CIA-1> takes up again (reported by Archivista)
14:41 < karasz> or just dld in the right location using wget?
14:41 < mnemoc> dtach, which is != detach, it's almost the same without TERM hackery or off-line buffer
14:42 < mnemoc> karasz: you could see my poor box as your mirror if gsmp doesn't have 2.1 files
14:42 < rxr> mnemoc: nah
14:42 < rxr> let us fix missing files
14:43 < karasz> oki i will wait for missing file
14:43 < rxr> karasz: I inject it right now
14:43 < mnemoc> :)
14:43 < mnemoc> much better
14:43 < karasz> mnemoc, your upload is better spent with gsmp than me ;)
14:43 < mnemoc> :)
14:44 < rxr> rm mozilla-1.7.11-source.tar.bz2.cksum-err
14:44 < rxr> rm: remove regular file `mozilla-1.7.11-source.tar.bz2.cksum-err'
14:44 < rxr> ?
14:44 < mnemoc> .oO( "you could see"??? what am i writting these days )o
14:44 < rxr> so - what was it I wanted to do right now ...
14:44 < rxr> damn
14:45 < rxr> there was someting important to remeber ...
14:45 < mnemoc> coreutils on dietlibc :)
14:45 < rxr> haha
14:45 < rxr> no %$$$ related
14:45 < mnemoc> rxr: 2.1 has .8 and trunk .12
14:46 < rxr> ah
14:46 < mnemoc> coreutils is lovely
14:49 < karasz> guys an advice about partitioning...
14:49 < karasz> i usualy do a /boot swap / in that order...
14:49 < karasz> should i change my ways?
14:50 < mnemoc> i do boot + big lvm2
14:50 < rxr> I do no boot
14:51 < mnemoc> i mean root
14:51 < rxr> and I place swap first, because normal HDs are fastest at the beginning ...
14:51 < karasz> hmmm o do not think my .boot is used...
14:51 < mnemoc> and swap, home, var, usr, opt, srv on lvm2
14:52 < karasz> for now i guess i will do just swap and / and throw the rest to vserver
14:52 < karasz> how does it sound?
14:52 < mnemoc> yes, HDs are faster at the begining, but how cares :p
14:52 < mnemoc> lvm flexibility pays
14:52 < rxr> hehe - goo that network/dhcp/rocknet_dhcp.sh was not renamed
14:53 < mnemoc> renamed?
14:53 < karasz> another one:
14:53 < rxr> updated from r10937 to r14476 on the carefully selected archivista anspshot ...
14:53 < rxr> snapshot ...
14:53 < karasz> i have some very small apps that do not have a Makefile...
14:53 < rxr> woah!
14:53 < karasz> am i suposed to provide one for T2?
14:53 < rxr> kdebase on the hammer notebook: ccache 663 hits 1 misses !!!
14:54 < mnemoc> karasz: vserver use /vservers, and i that case i use a logical volume for each client in there
14:54 < mnemoc> karasz: grep '^makeopt=$' package/*/*/*.conf
14:54 < rxr> karasz: apps of yourself or other open source sites ?
14:55 < karasz> other
14:55 < karasz> i do not code C :(
14:55 < karasz> mnemoc,
14:55 < mnemoc> not all software is written on C
14:55 < karasz> package/java/eclipse/eclipse.conf:makeopt=
14:55 < karasz> package/java/java-dirtree/java-dirtree.conf:makeopt=
14:55 < karasz> package/mail/qmqtool/qmqtool.conf:makeopt=
14:55 < karasz> package/multimedia/flashplayer/flashplayer.conf:makeopt=
14:55 < karasz> package/multimedia/w32codec/w32codec.conf:makeopt=
14:55 < karasz> package/printing/cups-samba/cups-samba.conf:makeopt=
14:55 < karasz> package/security/vserver/vserver.conf:makeopt=
14:55 < karasz> package/textproc/text2pdf/text2pdf.conf:makeopt=
14:55 < mnemoc> karasz: not for me!
14:55 < mnemoc> karasz: for _you_
14:56 < karasz> `makeopt=` ?
14:56 < mnemoc> karasz: those packages don't use `make all`, so you can take them as reference
14:56 < karasz> oki
14:56 < karasz> now i get it
14:56 < rxr> karasz: how is the software built ?
14:57 < karasz> a simple gcc call
14:57 < karasz> gcc -o foo foo.c
14:57 < karasz> single C files mostly
14:58 < mnemoc> if it's more than one .c, probably patching in a Makefile would be better
14:58 < karasz> just one C
14:58 < mnemoc> makeopt=
14:58 < mnemoc> makeinstopt=
14:58 < karasz> very very small apps
14:58 < mnemoc> hook_add inmake 5 '$CC -o foo foo.c'
14:58 < karasz> hook_add inmake 5 "$CC -o $bindir/text2pdf $( match_source_file -p text2pdf )"
14:59 < karasz> oki
14:59 < mnemoc> hook_add postmake 5 'cp -v foo $sbindir'
14:59 < mnemoc> yes, text2pdf is a single .c :)
14:59 < mnemoc> but that inmake is dirty
14:59 < mnemoc> compiling and installing on the same line ... yuck
14:59 < mnemoc> .oO( who did that? )o
15:01 < rxr> mnemoc: it is even so for rc - which could potentionlly crash an admins rc while emerging ...
15:01 < mnemoc> rxr: ack, install shall be as atomic as possible
15:02 < rxr> when you have a THz box maybe assembling looks atomic for a crawling snake ... ,-)
15:03 < rxr> == 01/04/06 16:02:24 =[5]=> Finished building package kdelibs.
15:03 < rxr> so - finally on the ppc64 box ..
15:03 < rxr> can't await to have a real browser ...
15:03 < mnemoc> hehe
15:04 * mnemoc can't wait for a normal dietlibc-based target
15:04 < mnemoc> do you have something to build strace for dietlibc?
15:04 < rxr> I feared corrupted select on ppc*, posix incorrect declares and crashing sort and sed would annoy you ...
15:04 < rxr> mnemoc: not off hand
15:04 < mnemoc> oh
15:05 < rxr> until archivista calls back I look onto coreutils and strace for some minutes ...
15:05 < mnemoc> i'll do money-work for now too
15:06 < rxr> well I have more $$$ on the TODO but I need something relaxing for some minutes ..
15:06 < rxr> we could declare this lunch break ,-)
15:07 < mnemoc> coreutils is good for relaxing :)
15:09 < rxr> 17068 root 18 0 589m 312m 1192 R 1.7 72.1 0:18.85 cc1plus
15:09 < rxr> ^- hamer sub-notebook swapping to death ...
15:09 < rxr> at least better now that I build with PARALLEL_MAX=1 ...
15:10 < rxr> but I guess I need 1GB RAM ...
15:10 < CIA-1> jsaw * r14477 /trunk/package/base/shadow/ (nullok.patch shadow.conf shadow.desc):
15:10 < CIA-1> * update shadow (4.0.13 -> 4.0.14)
15:10 < CIA-1> * disable selinux module usage on non-selinux systems
15:10 < CIA-1> (otherwise pam fails to login :| )
15:10 < CIA-1> * FIXME: created a currently empty etc/pam.d/system-auth file
15:13 < CIA-1> jsaw * r14478 /trunk/TODO:
15:13 < CIA-1> * added note for PAM review
15:13 < CIA-1> * removed two items (gcc update to 4.0, boot code refactoring)
15:14 < mnemoc> jsaw: i thought you were using pwdutils
15:14 < jsaw> for now I try to get a running system before starting experiments... :|
15:14 < mnemoc> valid :)
15:15 < CIA-1> jsaw * r14479 /trunk/package/base/cracklib/cracklib.conf: * install the simple dictionary (others should be added probably also)
15:16 < jsaw> okay, no more pam_console etc. errors, but now pam complains about missing prelude client...
15:16 < jsaw> I'll ignore that for now...
15:16 < mnemoc> pam _require_ prelude??
15:17 < jsaw> no, it doesn't, but the new pam.d/files enable more than one usually has installed ...
15:17 < jsaw> I already had to add an selinux fix
15:18 < mnemoc> oh
15:18 < jsaw> Sooner or later I'll add our own properly tweaked pam.d/files
15:18 < mnemoc> btw, do you plan to package the rest of http://www.thkukuk.de/pam/ ?
15:20 < jsaw> I just forgot the url, but now I added it to my todo list...
15:20 < mnemoc> :)
15:20 < rxr> 17068 root 18 0 704m 324m 3644 T 0.0 75.0 0:29.91 cc1plus
15:20 < rxr> ...
15:20 < jsaw> O.O
15:20 < jsaw> kde?
15:21 < mnemoc> what else? :)
15:22 < jsaw> what'd you mean with what else?
15:22 < mnemoc> what else could those levels of perfect resource consume?
15:23 < mnemoc> could reach*
15:23 < rxr> mnemoc: what else? pixie or boost are candidates ..
15:24 < mnemoc> pixie is kde :)
15:24 < rxr> nope
15:24 < rxr> a 3d renderer
15:24 < rxr> pixie++ is the kde paint thing ..
15:24 < mnemoc> aha
15:24 < rxr> mnemoc: btw. kde only consumes so much memory because we --enable-final which let's kde preprocess all modules files into one file so gcc can do better inter procedural optimizations ...
15:25 < rxr> this of course leads to big C++ code chunks ...
15:25 < mnemoc> yes yes... kde is not the crap i like to think
15:26 < mnemoc> woh! pixie is nice
15:26 * mnemoc wants it
15:26 < rxr> pixie freezed or killed or so my athlon ...
15:26 < rxr> just building reference
15:27 < rxr> I hope the 1.5GB dual ppc does survive a pixie build ...
15:27 < karasz> --enable-final is also requested by konversation...
15:27 < mnemoc> outch
15:28 < rxr> but right now I hope to reach kdebase soon to have a browser
15:28 < karasz> i mean that --enable-final must be romoved for konversation
15:28 < karasz> not that it matters very much
15:29 < mnemoc> karasz: var_remove confopt ' ' '--enable-final'
15:29 < karasz> it is on konversation.conf
15:29 < karasz> already
15:29 < mnemoc> rxr: pixar really use pixie?
15:29 < rxr> no idea ...
15:30 < rxr> I only stumbled over it when I wanted to find out what code freezed my box every week reference hit it ...
15:30 < mnemoc> :)
15:31 < mnemoc> Maintainer: Sebastian Czech <--- sepp?
15:31 < rxr> IIRC yes
15:32 < mnemoc> Build time (on reference hardware): 52117% (relative to binutils) <-- *G*
15:32 < rxr> I fear the cache is outdated
15:32 < rxr> because it never built on my box ...
15:35 < rxr> sed is so complex ...
15:35 < mnemoc> pixie 1.4.4 built on 2.1
15:35 < rxr> no wonder minised is 10 times faster ...
15:36 < mnemoc> but t2 is too complex for minised
15:36 < rxr> what code I dive thru for sed to compile the simple s/^a/b/ substitude ...
15:36 < rxr> mnemoc: there is not much missing ..
15:36 < rxr> so far I know two things ...
15:36 < rxr> could be easily tweaked in minised ..
15:37 < mnemoc> without bloating?
15:37 < rxr> though one is a gnu extension - we could alter t2 instead ...
15:37 < rxr> the other is a bug I think
15:37 < rxr> I could write an minisort as well ..
15:37 < rxr> the coreutils sort looked, well, very bloated ...
15:37 < rxr> n
15:38 < mnemoc> fullfeatured :)
15:38 < rxr> nah bloated
15:38 < mnemoc> *g*
15:38 < rxr> minised is also mostly posxliy full-featured and not bloated ..
15:39 < rxr> well minised's biggest restriction is a fixed byte-compiled expression space - but I could remove that in an afternoon ...
15:39 < mnemoc> we should avoid non-posix
15:40 < mnemoc> and sort? also disable coreutils' ?
15:40 < rxr> hehe
15:40 < rxr> yes ...
15:40 < rxr> joking ...
15:40 < mnemoc> :)
15:40 < rxr> I still dive thru sed ...
15:40 < rxr> 9th breakpoint already ...
15:40 < mnemoc> outch
15:40 < rxr> 10th
15:41 < mnemoc> sadly we can enforce posix sed on t2, but not on packages
15:41 < rxr> well - configure is so complex bacause they have so many fallbacks ...
15:42 < rxr> but instead writing posix conform scripts at the first place they generate all this complex fluff in the hope one variant will run ...
15:42 < rxr> well - more or less
15:43 < rxr> 12th breakpoint
15:43 < rxr> damn is that deep and complex ...
15:43 < rxr> I just asked for s/^a/b/ and not one milions backrefs ...
15:43 < rxr> my ...
15:43 < mnemoc> memory exhausted
15:44 < mnemoc> but is this done on sed itself or libc?
15:44 < rxr> 13th
15:45 < mnemoc> what are you using? gdb?
15:45 < rxr> yep
15:46 < mipe> NOOOOOOOOOOOOOOooooooooooooooo
15:46 < rxr> mipe: ?
15:47 < mipe> gfx card from my laptop is now officially broken
15:47 < rxr> ?
15:47 < mipe> there went ATI 9000
15:47 < rxr> >
15:47 < rxr> >
15:47 < rxr> ? even ... ?
15:47 < mipe> lots of distrortion and now only blank screen
15:47 < mnemoc> .oO( why gdb didn't build for me? )o
15:47 < rxr> mnemoc: I first debugged on my full system ...
15:47 < mipe> pixels went all mad :(
15:48 < rxr> of course I first rebuild with debug symbols ...
15:48 < rxr> no need to build dietlibc'ed bdb ..
15:48 < rxr> but now I debug a temp build sed on my ppc64 box since my hammer is too busy swapping ...
15:48 < rxr> mipe: ouhm
15:51 < rxr> mnemoc: I fear this is just because dietlibc does not allow mallo (0) ?
15:51 < mnemoc> what malloc(0) should do?
15:51 < rxr> undefined
15:51 < rxr> I saw someone commited a allow mallo (0) for uclibc some days ago
15:52 < mnemoc> sounds easy to implement
15:52 < rxr> on glibc it retunrs a pointer to 0 lenth storage ...
15:52 < rxr> depens on your memory model
15:52 < rxr> depends
15:52 < jsaw> If size is 0, either a null pointer or a unique pointer that can be successfully passed to free() will be returned.
15:52 < mnemoc> "If the size of the space requested is 0, the behavior is implementation-defined: the value returned shall be either a null pointer or a unique pointer."
15:52 < mnemoc> ^--- POSIX
15:53 < jsaw> http://www.opengroup.org/onlinepubs/007908799/xsh/malloc.html
15:53 < rxr> ack
15:53 < mnemoc> http://www.opengroup.org/onlinepubs/009695399/functions/malloc.html :)
15:53 < rxr> dietlibc let the kernel do the work
15:53 < rxr> and let the kernel handle all the mapping
15:53 < rxr> should scale well on linux I think
15:53 < rxr> might not be easy to add 0 lenth hackery there ...
15:53 < rxr> maybe better fix sed ,-)
15:53 < mnemoc> really?
15:54 < rxr> #ifdef WANT_MALLOC_ZERO
15:54 < rxr> aha!
15:54 < rxr> ok - it might be as easy as ... ,-)
15:54 < jsaw> :)
15:54 < rxr> let me see what we choose to build dietlibc with ... ,-)
15:54 < rxr> dietfeatures.h:/* #define WANT_MALLOC_ZERO */
15:54 < rxr> mnemoc: maybe at least sed is easy to get ...
15:55 < rxr> let's see how hard sort will be (if it does not go away with this change ...)
15:55 * rxr hoping ...
15:56 < mnemoc> jsaw: any particular reason to use posix 1997 instead of posix 2004?
15:56 < rxr> mnemoc: inside the build ?
15:57 < jsaw> mnemoc: which one is which ?
15:57 < jsaw> oh, ic
15:57 < rxr> ?
15:57 < jsaw> I justed picked the first hit while search opengroup
15:58 < mnemoc> jsaw: :)
15:58 < mnemoc> rxr: as reference, i have a bookmark to posix 2004... but jsaw pasted a link from posix 1997 :)
15:59 < rxr> ok
15:59 < mnemoc> and i thought he has reasons to prefer it
16:00 < jsaw> good ol times, ya mean?
16:00 < rxr> I thought you talk about s.th. inside t2 ...
16:00 < rxr> posix 1989 was so nice ... had so many beloved functions deprecated these days ...
16:01 < rxr> # ./sed s/^a/b/ /proc/cpuinfo
16:01 < rxr> processor : 0
16:01 < rxr> works now (tm)
16:01 < mnemoc> WANT_MALLOC_ZERO?
16:02 < rxr> # ./sed s/^p/b/ /proc/cpuinfo
16:02 < rxr> brocessor : 0
16:02 < rxr> mnemoc: ack
16:02 < mnemoc> sort too?
16:02 < rxr> mnemoc: -ENOTTESTEDYET
16:02 < CIA-1> rene * r14480 /trunk/package/base/dietlibc/dietlibc.conf: * enabled WANT_MALLOC_ZERO in dietlibc since GNU sed depends on it
16:04 < rxr> btw. the xomplexity of gnu sed is rather disgusting ...
16:04 < mnemoc> rxr: what's the trick to ask dietlibc to not warn about bloat on malloc and stdio?
16:04 < mnemoc> coreutils is rather disgusting
16:05 < rxr> mnemoc: I think it is a compile time switch - and I would rather like to keep it ...
16:06 < mnemoc> it's not nice when Build-Tools
16:06 < rxr> is it just cosmetic ?
16:07 < mnemoc> and, also... i get an issue with start.o... minute
16:07 < mnemoc> let me rebuild dietlibc
16:10 < rxr> nope - sort is not yet solved ...
16:11 < mnemoc> http://store.sun.com/CMTemplate/CEServlet?process=SunStore&cmdViewProduct_CP&catid=132680 <--- come on... how can that cost 9k dollars??!
16:11 < mnemoc> almost 10k
16:12 < rxr> hm - what is it ?
16:12 < mnemoc> a laptop
16:12 < mnemoc> 17"
16:12 < rxr> a sparc ?
16:12 < mnemoc> yes
16:12 < mnemoc> but 10k dollars?!
16:13 < rxr> because here is no competition ?
16:13 < jsaw> har har, bincimap fails on x86-64
16:13 < rxr> they could sell it for 20k and the people would buy it ...
16:13 < mnemoc> o.o
16:13 < morfoh> mnemoc: any idea what I need to do if I want to switch from devfs -> udev on syne ? I want to test a newer kernel on 2.1
16:14 < mnemoc> morfoh: build mnemosyne (without X) on 2.2
16:14 < karasz> hmm should i build bootcd in paralel with desktop?
16:14 < rxr> hm
16:15 < rxr> it appears to be just on exit stuff bailing out ...
16:15 < morfoh> mnemoc: what is the missing piece on 2.1 ?
16:15 < mnemoc> karasz: 2.1? one build of bootdisk target and one build of desktop target
16:15 < karasz> 2.1
16:15 < mnemoc> morfoh: udev/initrd/hotplug++
16:15 < karasz> desktop is buildin
16:15 < karasz> i wonder if i may launch boot too...
16:16 < mnemoc> karasz: you could build bootdisk ant the same time
16:16 < morfoh> mnemoc: so I assume updating the "3" above on 2.1 and include them would work too ?
16:17 < mnemoc> morfoh: i doubt
16:17 < mnemoc> morfoh: it's not just the packages, it's the integration
16:17 < mnemoc> morfoh: sysfiles and mkinitrd
16:18 < rxr> mnemoc: sort "is just freaking up in the atexit handlers ..."
16:18 < morfoh> mnemoc: hmmmm ...
16:18 < mnemoc> o.o
16:18 < rxr> though in theory I have no time to tinker ...
16:18 < rxr> let's see what it take to fix ...
16:19 < demian> just to know.. are there radio FM devices, so that i could tune my radio stations from a serial cable, or sth else ?
16:20 * demian wants to listen local FM radio stations and tune them from the PC..
16:21 < mnemoc> rxr: FYI: http://www.rafb.net/paste/results/eD5agq71.html
16:23 < rxr> but it builds fine, right ?
16:23 < mnemoc> Config thing, no
16:23 < mnemoc> tools, yes
16:23 < rxr> Config in dietlibc ?
16:24 < mnemoc> but 0.28 doesn't handle malloc_zero :(
16:24 < rxr> mnemoc: ouhm
16:24 < mnemoc> rxr: see line 5
16:24 < morfoh> mnemoc: it was added to 0.29
16:24 < rxr> well - then advanced dietlibc stuff is for trunk ,-)
16:24 < mnemoc> rxr: well... feature_malloc_zero.patch :p
16:25 < rxr> heh
16:25 < karasz> cu in a while
16:25 < rxr> or update dietlibc once on 2.1 ...
16:26 < mnemoc> doesn't 0.29 have regressions?
16:27 < rxr> need to review ...
16:27 < rxr> well if you want you can back the zero malloc stuff out ...
16:33 < mnemoc> it doesn't pay if sort can't be fixed :)
16:33 < morfoh> mnemoc: why did you proposed building a syne on 2.2 without X ?
16:34 < morfoh> mnemoc: not modular-x aware ?
16:34 < mnemoc> yep
16:35 < mnemoc> but a module would be needed for udev anyway
16:35 < mnemoc> one .all for rockplug if on 2.1, and one .all for udev+hotplug++ if on 2.2
16:36 < morfoh> mnemoc: hmmm ... yes
16:36 < mnemoc> using SDECFG_SHORTID
16:37 < mnemoc> but, mnemosyne.pl should be fixed to support non-module dependencies and patterns on the match
16:39 < morfoh> mnemoc: so you mean, that we can inject "SDECFG_FOO" as a dependency
16:40 < mnemoc> yes
16:41 < mnemoc> but currently [ ] is used, but if it containts '*' we should use [[ ]]
16:42 < mnemoc> and i don't rember if the call to tsort is ignoring deps with ^SDECFG_
16:43 < mnemoc> but that's the idea
16:46 < jsaw> gotta go
16:46 < jsaw> cu guys
16:46 < mnemoc> cu dr.j
16:46 < morfoh> cu jsaw
16:46 < mnemoc> .oO( that's like a hip hop singer )o
16:47 < morfoh> mnemoc: Dr. "Jam Master" J. :)
16:51 < morfoh> mnemoc: if it would be also interesting for you, I could try to get udev/hotplug++ support running on 2.1 in favour of rockplug ... perhaps that could be interesting for future migration scenarios too ? ...
16:51 < mnemoc> no time for that currently :\
16:53 < morfoh> mnemoc: I know that you have no time for that, but I didn't asked you to do it ... I just wanted to know your thoughts about it ;)
16:53 < mnemoc> i don't want udev on 2.1
16:53 < mnemoc> udev is 2.2 field
16:54 < mnemoc> as modular kernel and modular X
16:54 < morfoh> ok! that's a word
16:54 < CIA-1> rene * r14481 /trunk/package/base/dietlibc/pkg_patch/pkg_coreutils.patch:
16:54 < CIA-1> * extended dietlibc/pkg_coreutils.patch to not crash with dietlibc
16:54 < CIA-1> (the __fpending code is strange and I wonder if it is POSIX conform
16:54 < CIA-1> to flose stdout ...)
16:54 < mnemoc> all the rest can be merged back
16:56 < rxr> mnemoc: there is a udev package on 2.1 ,-)
16:56 < mnemoc> yes, but unsupported
16:56 < rxr> heh
16:56 < mnemoc> or you want me to replicate all your features on 2.1 ,-)
16:56 < rxr> of course not
16:57 < rxr> that was more of a "maybe we should svn rm it on 2.1 ..."
16:57 < mnemoc> too late for that imo
16:57 < mnemoc> or.... you think noone is using it?
16:58 < rxr> yes
16:58 < rxr> but just leave it there ...
16:58 < rxr> no work no trouble ...
16:59 < mnemoc> have the hope we are more that the 27 fellows on frappr
17:02 < rxr> == 17:58:26 =[5]=> Building kde/kdebase [3.5.0 2.2.0-alpha].
17:02 < rxr> `- ppc64
17:03 < mnemoc> does fefe has a public svn/cvs ?
17:03 < rxr> cvs
17:04 < mnemoc> oh, here it is.... /me blind
17:07 < CIA-1> rene * r14482 /trunk/target/embedded/config.in:
17:07 < CIA-1> * fixed embedded target pkg format locking for the type rename
17:07 < CIA-1> * let the embedded target version the tarballs
17:08 < mnemoc> versioned tarballs are good :)
17:08 < rxr> yes
17:08 < rxr> formerly unversioned was for the simple extractor code
17:08 < rxr> now we have other, more advanced image creation
17:08 < rxr> the tarballs are not needed anymore at all
17:09 < rxr> but for reference builder with automatic publishing they are nice ...
17:09 < rxr> build the next arch every night and auto-publish the files leaving the old versions for reference ...
17:09 < rxr> no ?
17:09 < mnemoc> uhm, MALLOC_ZERO thing was not introduced on 0.29
17:10 < mnemoc> rxr: ehm.... i think you will need a package.db file there
17:10 < mnemoc> rxr: to see checksum changes
17:10 < mnemoc> rxr: for small changes
17:10 < rxr> you mean auto publish ?
17:10 < mnemoc> yes
17:11 < mnemoc> don't you hate upstreamers changes tarballs without changing names?
17:11 < rxr> sure, but for the just for convience, publication ?
17:12 < rxr> if you want that, ... we could implement revisioned tarballs ...
17:12 < mnemoc> no please no
17:13 < rxr> I just wanted rsync build results to known example cross build location keeping old version files ...
17:13 < mnemoc> publish, but publish with an index file containing the checksums
17:13 < rxr> will someone look into the checksum ....
17:13 < mnemoc> automatized scripts... like sam :p
17:16 < morfoh> poor "sam" :p
17:16 < mnemoc> :)
17:17 < CIA-1> rene * r14483 /trunk/target/ (6 files in 6 dirs):
17:17 < CIA-1> * fixed remaining PKGFILE_TYPEs, set desktop to tar.bz2 now that
17:17 < CIA-1> we have a patched mine that can handle it and archivista, livecd,
17:17 < CIA-1> psion-pda and dreamcast to none
17:22 < morfoh> rxr: what do you think ... will I get a working desktop out of trunk or what tree do you propose ?
17:23 < mnemoc> minimal-desktop builds almost flawless
17:23 < mnemoc> on turnk
17:26 < morfoh> mnemoc: thx ;)
17:26 < morfoh> mnemoc: but if you can imagine I'd like to keep my syne :p
17:27 < mnemoc> make it 2.2 aware
17:27 < morfoh> that's the idea
17:27 < mnemoc> you could pkgenable udev and hotplug++ on a .conf for now
17:27 < morfoh> aha
17:27 < mnemoc> well... hotplug++ could go with rockplug on the same .all because the are only available on one tree
17:28 < mnemoc> and only pkgenable udev
17:28 < mnemoc> modular X is also safe on a multitree target
17:28 < mnemoc> because the names don't mix
17:28 < mnemoc> if there is package that match, you are on the tree that needs it
17:29 < morfoh> ok
17:29 < mnemoc> for desktop i would love you to try azazel
17:30 < mnemoc> i want to make it prelude/ldap/bacula ready
17:30 < morfoh> mnemoc: what ? you want to love me ? no way ... I've go outa here :p
17:31 < mnemoc> :)
17:31 < morfoh> mnemoc: ok ... I take a look on aza*what
17:31 < mnemoc> azazel
17:31 < morfoh> yeah ... right this one
17:31 < mnemoc> http://en.wikipedia.org/wiki/Azazel
17:32 < morfoh> I feared that you would provide a link
17:32 < mnemoc> :(
17:32 * mnemoc is predictable
17:32 < morfoh> don't worry
17:34 < mnemoc> azazel is a deamon which teach humans about the good things of life... like seduction, make up and wepons
17:36 < morfoh> but I cannot see a connection between the story and a desktop target
17:37 < mnemoc> azazel is a mnemosyne-based desktop target
17:38 < mnemoc> just that
17:39 < mnemoc> rxr: http://www.rafb.net/paste/results/bzEo5T47.html <--- does it really work for you? 0.28 has MALLOC_ZERO thing
17:40 < mnemoc> rxr: btw, sort is working :)
17:40 < mnemoc> rxr: thanks a lot
17:43 < morfoh> mnemoc: so the diet build is proceeding ?
17:43 < mnemoc> ehm uhm... welll... yes
17:44 < mnemoc> with pasted result
17:54 < mnemoc> i'll see on trunk now
17:56 < mnemoc> # svn pget svn:externals ../t2-2.1/target/ | svn pset svn:externals -F - target/
17:56 < mnemoc> svn: Reading from stdin is currently broken, so disabled
17:56 < mnemoc> snif snif
17:57 < mnemoc> # svn pget svn:externals ../t2-2.1/target/ | svn pset svn:externals -F /dev/stdin target/
17:57 < mnemoc> property 'svn:externals' set on 'target'
17:57 < mnemoc> yeah yeah
17:57 < mnemoc> long life to lazyness!
17:57 < mnemoc> yeah yeah
17:58 < morfoh> :)
17:59 < morfoh> property 'svn:externals' set on 'target' <-- thanks Mr. Lazyness ;)
18:00 < rxr> mnemoc: I think it works here
18:01 < rxr> have you rebuild sed ?
18:01 < mnemoc> # svn pget svn:externals target/ | sed '/^$/d;' | wc -l
18:01 < mnemoc> 6
18:01 < mnemoc> rxr: yes
18:02 < mnemoc> i'll start the same build on trunk to compare
18:02 < mnemoc> rxr: http://www.rafb.net/paste/results/eD5agq71.html <--- did you see start.o thing on top?
18:03 * mnemoc doing stress testing over t2's dietlibc integration :p
18:05 < rxr> I rebuild and trest
18:05 < rxr> == 01/04/06 19:04:11 =[5]=> Finished building package kdebase.
18:05 < rxr> ah no
18:05 < rxr> missed that
18:08 -!- nzg [n=tschmidt@p54AFF4BE.dip.t-dialin.net] has joined #t2
18:08 < mnemoc> wb nzg
18:10 < nzg> thx mnemoc
18:10 < nzg> moin
18:10 < rxr> cool
18:10 < rxr> konqueror works ..
18:10 < morfoh> moin nzg
18:10 < rxr> on ppc64 ...
18:10 < rxr> in contrast to segfaulting mozilla and not compiling firefox ...
18:11 < rxr> not to mention I even hat to patch mozilla not to try assembling x86 assembly ...
18:17 < rxr> damn, how good a really browser feels under the fingertips ..
18:17 < mnemoc> :)
18:17 < rxr> some dasys without any and some more on links -g make you consider killing for one ...
18:17 < mnemoc> links -g is better tahn dillo :)
18:18 < rxr> not when it has major graphic glitches due 64bit or big endian ...
18:18 < rxr> ... or both at once
18:18 < rxr> mnemoc: sed works now
18:18 < mnemoc> now?
18:18 < rxr> rebuild ..
18:18 < rxr> that is last commit did it ...
18:18 < mnemoc> i have done that
18:18 < rxr> $ sed --version
18:18 < rxr> GNU sed version 4.1.4
18:18 < mnemoc> last ocmmit last commit...
18:19 < mnemoc> PKGFILE_TYPEs?
18:19 < mnemoc> # sed --version
18:19 < mnemoc> GNU sed version 4.1.4
18:20 < mnemoc> # ls -1 | sed 's,^d,f,g'
18:20 < mnemoc> sed: -e expression #1, char 8: Memory exhausted
18:20 < mnemoc> with MALLOC_ZERO revision on dietlibc
18:20 < rxr> $ ls -1 | sed 's,^d,f,g'
18:20 < rxr> rene
18:20 < rxr> vol1
18:20 < rxr> vol2
18:20 < mnemoc> :(
18:21 < rxr> feel like paste your dietlibc log ?
18:21 < mnemoc> 1-dietlibc.log ?
18:21 < rxr> ack
18:21 < mnemoc> minute
18:21 < rxr> maybe some WANT_MALLOC_ZERO redefined ...
18:21 < rxr> # ./scripts/Config -nobashmod
18:21 < rxr> Creating configuration script.
18:21 < rxr> T2 2.2.0-alpha configuration ...
18:21 < rxr> unknown returncode: 127
18:22 < rxr> ouhm
18:22 < rxr> but I won't hunt that now
18:22 < mnemoc> 0.28-0.29 has nothing related
18:22 < rxr> weee
18:23 * rxr in a Config of a full diet system ...
18:23 < rxr> even crossbuilt ...
18:23 < rxr> I even tweaked gcc specs file to use dietlibc by default ... ,-)
18:23 < mnemoc> http://t2.geeks.cl/1-dietlibc.log.txt
18:24 < karasz> damn this is takin' long...
18:24 < mnemoc> gcc specs file.... wasn't that prohibit by mr. core architect?
18:26 < rxr> hm
18:26 < rxr> you log looks good
18:26 < mnemoc> yep
18:26 < mnemoc> == 15:25:57 =[1]=> Building base/dietlibc [0.29 2.2.0-alpha]. <--- same config
18:27 * rxr trying to build subversion with system dietlibc manually ...
18:29 < mnemoc> == 15:28:17 =[1]=> Building perl/perl [5.8.7 2.2.0-alpha]. /me praying
18:30 < rxr> configure went thru on the diet only box ...
18:30 < morfoh> == 19:27:58 =[1]=> Building perl/perl [5.8.7 2.2.0-alpha].
18:30 < rxr> let's see if dietlibc is mature enough for svn and the gcc on diet stable enough to survive it all
18:30 < morfoh> mnemoc: I think perl will build
18:31 * mnemoc prefer pure Build-Target instead of nasty emerges
18:31 < mnemoc> == 01/04/06 15:30:38 =[1]=> Aborted building package perl.
18:31 < morfoh> :/
18:31 < morfoh> == 01/04/06 19:30:20 =[1]=> Aborted building package perl.
18:32 < mnemoc> same time to fail :)
18:32 < morfoh> allmost :)
18:33 < mnemoc> what mess was jsaw doing on 1-perl?
18:33 < morfoh> reentr.h:84:26: error: crypt.h: No such file or directory
18:34 < mnemoc> good, our dietlibc doesn't have crypt.h
18:37 < rxr> subversion building .... and building ...
18:37 < rxr> oh
18:37 < rxr> stop
18:38 < rxr> hm - neon config.h not found ...
18:39 < mnemoc> found. perl is a lier
18:40 < mnemoc> lib64?
18:40 < mnemoc> ehm, no... ignore that
18:51 < CIA-1> rene * r14484 /trunk/package/base/ (dietlibc/parse-config-9 file/file.desc):
18:51 < CIA-1> * added file to the BSD compat section of dietlibc and marked it
18:51 < CIA-1> DIETLIBC
18:52 < mnemoc> rxr: does perl build on your dietlibc?
18:53 < rxr> no ide - I fight other stuff
18:54 < rxr> idea
18:54 < mnemoc> can linux be built without perl these days?
18:54 < mipe> huh, i guess laptop works now, after cleaning up the gfx cooler from dust :)
18:55 < mnemoc> ehm
18:55 < mipe> was enough dust to block it from spinning
18:55 < mnemoc> dirty mipe
18:58 < mipe> mnemoc: something to test, svn commit segfaults on 2.1.1
18:58 < rxr> maybe x86-64 only
18:58 < mipe> mnemoc:tryed to rebuild svn and deps from 2.1 branch but didnt help
18:58 < mipe> or just my host only
18:59 < mnemoc> strace ?
19:00 < mipe> i could run it through strace and post the output somewhere...
19:01 < mnemoc> yep
19:02 < mnemoc> rxr: perl's Configure check host to know what is installed, it see crypt.h on host... (cross)compiler doesn't
19:02 < mnemoc> rxr: sugestions?
19:02 < rxr> yeah - such stuff is dirty ...
19:02 < rxr> aside hacking configure not too much suggestions ...
19:04 < mnemoc> to look at $root or $root/TOOLCHAIN/$toolsdir ?
19:04 < mnemoc> (1-perl)
19:04 < rxr> depends how the configure is hacked if it can be guided
19:04 < rxr> otherwise some generic mechanism need to be added to force it ...
19:06 < mipe> mnemoc: http://koti.japo.fi/~pmika/strace-svn.log
19:07 < mipe> looks like some memory issues
19:07 < mnemoc> perl's Configure is far^3 uglier than conf-* and install.c
19:08 < mipe> so 1001 builds total, 529 completed fine, 2 with errors.
19:09 < mipe> trunk desktop x64
19:09 < mnemoc> usrinc!
19:09 < mipe> but have to switch to bloody thesis work... till tomorrow
19:09 -!- mipe [n=mipe@dsl10040.japo.fi] has quit ["leaving"]
19:12 < mnemoc> rxr: libpth="/usr/local/${libdir##*/} /${libdir##*/} /usr/${libdir##*/}" <--- why no $root ?
19:13 < rxr> oehm
19:13 < rxr> do not remeber off hand
19:14 < rxr> when we continue at this pace we will have a dietlibc only system not at christmas 2006 but cebit ...
19:14 < mnemoc> :D
19:15 < mnemoc> who planned it for xmas 2006?
19:16 < rxr> my estimation some days ago ...
19:17 < CIA-1> rene * r14485 /trunk/target/embedded/config.in:
19:17 < CIA-1> * added automake and libtool to stage 0 of embedded, we need
19:17 < CIA-1> those for certain broken packages that have a *-QUIRK mark
19:17 < CIA-1> * added file, diffutils, mktemp and time to embedded, just builds with
19:17 < CIA-1> dietlibc and handy if the resulting system is used for development
19:23 < mnemoc> rxr: any know reason to not use -Dusecrosscompile=yes on 1-perl?
19:23 < rxr> nope
19:23 < rxr> the reson is that I did not spend a day to research this braindamge ..
19:23 < rxr> I only hacked it only it build for our usual bootstrap
19:24 < mnemoc> :)
19:25 < morfoh> mnemoc: did -Dusecrosscompile=yes on 1-perl worked ?
19:25 < rxr> so - let see if my diet gcc builds libxml ...
19:25 < rxr> where I removed a big chunk of C++ prerocessor checks I do not have and libxml does not seem to need ...
19:25 < mnemoc> morfoh: not yet
19:25 < mnemoc> Targethost not defined.
19:25 < mnemoc> Cannot continue, aborting.
19:26 < CIA-1> rene * r14486 /trunk/package/editors/zile/zile.desc: * makred zile DIETLIBC, just builds
19:27 < mnemoc> as good software should always do
19:32 < mnemoc> ehm... we don't want -Dusecrosscompile=yes, it connects to a remote machine to do the job!
19:33 < morfoh> what ?
19:33 < mnemoc> and you said djb-style was ugly
19:33 < mnemoc> 2709 case "$targetrun" in
19:33 < mnemoc> 2710 '') targetrun=ssh ;;
19:34 < mnemoc> 2755 $targetrun -l $targetuser $targethost "cd \$cwd && ./\$exe \$@"
19:34 < mnemoc> you don't want to see the rest
19:36 < madtux> yes he does
19:36 < mnemoc> hi madtux
19:36 < morfoh> hola madtux ;)
19:37 < owl> /nick ice-owl
19:37 < owl> hi
19:37 < morfoh> hi owl ... seems your hands are that frozen that you can't type anymore :)
19:37 -!- YeT2 [n=perdiT2@i53878831.versanet.de] has joined #t2
19:38 < YeT2> hi
19:38 < owl> nah. just to lazy to rename and rename agin ;)
19:38 < owl> hi morfoh ;)
19:38 < karasz> hi owl
19:38 < owl> hi YeT2
19:38 < owl> hi karasz *g*
19:38 < mnemoc> hi frigid-owl
19:38 < mnemoc> hi YeT2
19:38 < karasz> *g* ?
19:38 < morfoh> mnemoc: uhhhhh ... what a nice way to say hello
19:38 < owl> hi mnemoc :)
19:38 < owl> *g* because of multi-hi
19:39 < madtux> greetings freaks.
19:39 < karasz> err...
19:39 < owl> hi madtux :)
19:39 < madtux> i'm original :)
19:39 < karasz> hi madtux
19:39 * karasz hides
19:39 < YeT2> i joined just in time
19:40 < jsaw> ls
19:40 < jsaw> ups....
19:40 < jsaw> :)
19:40 < madtux> bash: ls: command not found
19:40 < YeT2> you mean dir :)
19:40 < jsaw> hehe
19:40 < mnemoc> madtux's box doesn't have 'ls' :)
19:41 < madtux> mnemoc, scrollback :)
19:41 < YeT2> what for ... i from my side know everything "by heart"
19:42 < madtux> YeT2, by heart huh?
19:42 < madtux> :)
19:42 < YeT2> mnemoc can confirm ... my questions are much less now (FOR THE MOMENT!)
19:43 < mnemoc> o.o
19:46 < rxr> Wine 0.9.5
19:46 < mnemoc> CIA?
19:49 < rxr> not commited - just pasted from my shiny new konqueror ...
19:49 < mnemoc> :)
19:49 < rxr> GLib 2.8.5
19:49 < rxr> Tor 0.1.0.16
19:51 < mnemoc> how can someone write such a pathetic configure script... (perl)
19:54 < mnemoc> 21 # $Id: Head.U,v 3.0.1.9 1997/02/28 15:02:09 ram Exp $
19:54 < mnemoc> 22 #
19:54 < mnemoc> 23 # Generated on Mon Apr 11 12:33:25 CEST 2005 [metaconfig 3.0 PL70]
19:54 < mnemoc> 24 # (with additional metaconfig patches by perlbug@perl.org)
19:55 < CIA-1> jsaw * r14487 /trunk/package/base/ (pam/etc_pamd_system-auth.txt shadow/shadow.conf):
19:55 < CIA-1> * need a correct system-auth file (or local login won't work...)
19:55 < CIA-1> -> move it to pam package
20:02 < karasz> mnemoc,
20:02 < karasz> database/bdb-java
20:02 < karasz> checking if gcj -C works... configure: error: The Java c ..
20:02 < karasz> database/bdb33-java
20:02 < karasz> checking for javac... nojavac
20:02 < karasz> !> configure: error: no javac compiler in PATH
20:02 < karasz> another 2 packages failed :D
20:03 < mnemoc> expectable considering kaffe doesn't build on x86-64
20:03 < karasz> boot target failed also at bash
20:03 < mnemoc> ?
20:03 < karasz> == 22:53:04 =[3]=> Building shells/bash [3.0 2.1-stable].
20:03 < karasz> -> Parallel build using 6 jobs enabled.
20:03 < karasz> -> Compiler Cache enabled: 4.76% cache hits so far.
20:03 < karasz> -> Reading package configuration from package directory.
20:03 < karasz> -> Preparing build in src.bash.boot.20060104.225302.11579.karasz
20:03 < karasz> -> Building. Writing output to $root/var/adm/logs/3-bash.out
20:03 < karasz> -> Compiler Cache Hits while pkg build: 85.43% (170 hits, 29 misses)
20:03 < karasz> -> Cleaning up ccache ...
20:03 < karasz> !> gcc -c -I. -I../.. -I../.. -I../../lib -I../../include ..
20:03 < karasz> !> rm -f libsh.a
20:03 < karasz> !> ar cr libsh.a clktck.o clock.o getenv.o oslib.o setlineb ..
20:04 < karasz> !> test -n "ranlib" && ranlib libsh.a
20:04 < karasz> !> make[1]: Leaving directory `/TOOLCHAIN/src.bash.boot.200 ..
20:04 < karasz> !> Due to previous errors, no 3-bash.log file!
20:04 < karasz> !> (Try enabling xtrace in the config to track an error ins ..
20:04 < karasz> -> $root/var/adm/logs/3-bash.out -> 3-bash.err
20:04 < karasz> == 01/04/06 22:54:54 =[3]=> Aborted building package bash.
20:04 < karasz> -> Unmounting loop mounts ...
20:04 < karasz> regression in a moment
20:04 < mnemoc> uhm
20:05 < CIA-1> rene * r14488 /trunk/package/ (base/dietlibc/parse-config-9 textproc/libxml/libxml.desc): * added libxml to the BSD compat section of dietlibc and marked libxml
20:06 < karasz> hmmm why takes so long to regression?
20:07 < rxr> so - xine on the ppc64, now I only need support for the sound hardware to make me less productive ...
20:07 < mnemoc> :)
20:07 < rxr> oh and kmail as well
20:07 < rxr> guess I can start creating a rene account now ... ,-)
20:08 < rxr> hm
20:08 < rxr> when we hack subversion to crosss build we could provide a static client for all t2 arches as weeks regression test ...
20:09 < karasz> still not finished the regression....
20:09 < karasz> ok
20:10 < karasz> http://www.x5.ro/regression/boot/regressions.html
20:11 < karasz> hmmm rm mbutil.o then tryin to ar cr ... mbutil.o
20:13 < karasz> but wait a minute
20:13 < karasz> in desktop target bash built fine...
20:13 < karasz> O_0
20:14 < mnemoc> in everywhere bash build fine, but your boot
20:14 -!- misl [n=chatzill@82-217-66-150.cable.quicknet.nl] has joined #t2
20:14 < mnemoc> rebuild it
20:14 < mnemoc> wb misl
20:14 < rxr> hi misl
20:14 < karasz> but how is my boot diferrent?
20:15 < karasz> create-errlist ...-remove bash?
20:15 < karasz> then build?
20:15 < misl> hi folks
20:15 < CIA-1> rene * r14489 /trunk/package/base/ (automake/parse-config libtool/parse-config): * fixed AUTOMAKE-QUIRK code and cleaned LIBTOOL-QUIRK accordingly
20:15 < karasz> and why should bash build then?
20:15 < rxr> paralle issue ?
20:15 < karasz> hello minto
20:15 < rxr> l
20:15 < karasz> me rxr?
20:16 < rxr> karasz: yes
20:16 < mnemoc> karasz: -job 3-bash
20:16 < rxr> try if it builds on a second run - if so it is due PARALLEL or sop
20:16 < karasz> i check if i have bash marked noparallel
20:16 < karasz> mnemoc, ?????
20:16 < mnemoc> at this step we will have to change NOPARALLEL flag with PARALLEL
20:16 < karasz> -job 3-bash?
20:16 < mnemoc> Build-Target -cfg boot -job 3-bash
20:17 < rxr> mnemoc: ?
20:17 < mnemoc> rxr: ?
20:18 < rxr> 20:16 < mnemoc> at this step we will have to change NOPARALLEL flag with
20:18 < rxr> PARALLEL
20:18 < karasz> what does -job do?
20:18 -!- YeT2 [n=perdiT2@i53878831.versanet.de] has quit ["User pushed the X - because it's Xtra, baby"]
20:18 < rxr> karasz: build a specific job now
20:18 < mnemoc> rxr: the amount of NOPARALLEL % on large packages
20:18 < rxr> $stagelevel-$pkg
20:19 < rxr> mnemoc: I find the amount reasonable ...
20:19 < mnemoc> wth 1-perl builds on 2.1 but not on 2.2 on the same cross-compiling context
20:20 < rxr> the ones we know are broken could be fixed in the future ...
20:20 < mnemoc> yes, i know... it was a joke
20:20 < rxr> ouhm
20:20 < rxr> it is too cold here to get your hot jokes ..
20:21 < mnemoc> :)
20:21 < CIA-1> rene * r14490 /trunk/package/network/neon/neon.desc: * marked neon DIETLIBC - just builds
20:21 < karasz> 01/04/06 23:14:25 =[3]=> Finished building package bash.
20:21 < karasz> should i relaunch target build?
20:21 < rxr> ack
20:21 < mnemoc> wow, diff between perl.conf at 2.1 and 2.2 is important
20:22 < mnemoc> so it was a parallelism issue
20:22 < karasz> maybe
20:22 < karasz> should i relaunch target build?
20:22 < rxr> [F] DIETLIBC NOPARALLEL
20:22 < rxr> on trunk bash already is NOPARALLEL
20:22 < rxr> on 2.1 not ?
20:23 < mnemoc> you marked it on 3.1 iirc
20:23 < mnemoc> but i'll merge
20:23 < mnemoc> rxr: i think the whole issue is the removal of perl_cross_bootstrap
20:23 < rxr> removal ?
20:24 < rxr> I think it was introduced on trunk, no ?
20:24 < rxr> so absense in 2.1 ...
20:24 < jsaw> re
20:24 < mnemoc> oh, true.... wrong diff
20:24 < mnemoc> so it's add
20:24 * mnemoc stunned
20:25 < CIA-1> rene * r14491 /trunk/target/embedded/config.in:
20:25 < CIA-1> * added zile to embedded, to have an emacs a-like editor for the other
20:25 < CIA-1> half of the world and elvis-tiny as a static edit buffer of 500kB,
20:25 < CIA-1> zile has either a dynamic or it is just much bigger
20:26 < mnemoc> aaaah! .... i can't find it :\
20:27 < rxr> mnemoc: find ?
20:28 < mnemoc> the source of the difference
20:29 < mnemoc> it builds perfect on 2.1, but it's big pain on 2.2
20:29 < rxr> ouh
20:30 < mnemoc> i can't find why on 2.1 doesn't look for includes on host
20:30 < rxr> what do you try right now ?
20:30 < rxr> corss build dietperl or what was it ?
20:30 < mnemoc> pass 1-perl on _native_ dietlibc minimal at trunk
20:30 < rxr> mnemoc: you have installed a dietlibc (in a vserver or so) and want to build?
20:31 < rxr> or glibc host and just normal dietlibc based build ?
20:31 < mnemoc> no
20:31 < mnemoc> yes
20:31 < karasz> but mnemoc when i make my isos should i care about those java packages?
20:31 < mnemoc> karasz: no
20:31 < karasz> so i do not have to do anything about them?
20:31 < karasz> like remove from iso.txt or some?
20:32 < mnemoc> karasz: uhm... i would just ignore them
20:32 < rxr> pure diet libc system is quite promissing, more stable than 64bit user-space on sparc64 ...
20:32 < karasz> in the end i do not really need java
20:32 < rxr> karasz: do you build desktop ?
20:32 < karasz> yes
20:32 < karasz> 2.1.0 desktop
20:32 < rxr> desktop has no commercial jave packages
20:33 < karasz> or is 2.1.1?
20:33 < rxr> what do you worry about ?
20:33 < rxr> java-dirtree ?
20:33 < karasz> i am not worried, just wondering if my cd's will be usable...
20:33 < rxr> other java bits like jikes, gcj or kaffe ar efree ...
20:34 < karasz> rxr i do not need java at all
20:34 < rxr> karasz: well then disable some more packages
20:34 < karasz> maybe just the plugin for firefox...
20:34 < rxr> but the desktop does not incldue java bits that are non-free ...
20:34 < karasz> nah
20:34 < karasz> let them build
20:34 < rxr> at least not if mnemoc has not messed it up ...
20:34 < karasz> i have time now...
20:34 < mnemoc> just kaffe
20:34 -!- madtux [i=miguel@pf0.hostarica.com] has quit [Read error: 104 (Connection reset by peer)]
20:34 < karasz> and patience :D
20:35 < karasz> maybe i will provide some unknown info about some breakage...
20:35 < mnemoc> i failed concurrent programming last year because i refused to install sun-java
20:35 < rxr> ouh
20:36 < karasz> in the end, more builds means more knowledge for everyone, doen't it?
20:37 < rxr> yep testing more is always better
20:38 < mnemoc> every machine is different
20:39 < rxr> mnemoc: I failed uni because I refused to waste too much time on crap and too much on rock linux and later t2 and other commercial work ...
20:40 < mnemoc> i still have the hope of returning university some day.... i have 6y approved, only 1 left :(
20:40 < mnemoc> but it's too expensive
20:40 < karasz> mnemoc, do you care to explai the last part of hook_add inmake 5 "$CC -o $bindir/text2pdf $( match_source_file -p text2pdf )"
20:40 < karasz> ?
20:40 < rxr> but this time I install Adobe products for my design study to experience how crappy they are and why I won't spent kEUR on this stuff when I have better open source around that I even can debug on a issue ...
20:41 -!- madtux [i=miguel@pf0.hostarica.com] has joined #t2
20:41 < mnemoc> karasz: it returns the location on download/ of the entries matching 'text2pdf' on [D] fields of your .desc
20:41 < rxr> .oO
20:41 < rxr> I fear dietlibc misses too much pthread stuff ...
20:41 < karasz> hmm
20:41 < karasz> why does it have to do that?
20:42 < karasz> wait
20:42 < karasz> i get it
20:42 < rxr> to compile the downloaded file
20:42 < karasz> ok
20:42 < karasz> ok
20:42 -!- YeT2 [n=root@i53878831.versanet.de] has joined #t2
20:42 < karasz> i got it in the end
20:42 < karasz> but my fingers were more rapid than my brain
20:43 < rxr> nah ! dietlibc has it - strike
20:43 < CIA-1> aldas * r14492 /trunk/package/mail/postfix/ (postfix.conf postfix.desc postfix.init postsysfiles.in): * updated postfix (2.2.7 -> 2.2.8)
20:43 < karasz> so "$CC -o $bindir/text2pdf $( match_source_file -p text2pdf )" is clear, it is only the command to compile
20:43 < karasz> hook_add inmake 5?
20:44 < karasz> 5 is the stage?
20:44 < mnemoc> the build process has many 'hooks'
20:44 < mnemoc> inmake is between make all and make install
20:44 < karasz> one is `inmake`?
20:44 < mnemoc> yes
20:44 < karasz> make all does not include install?
20:44 < karasz> ofcourse not
20:44 < mnemoc> and 5 is just a 'middle' for the list of actions to run on 'inmake'
20:45 < karasz> ahm
20:45 < rxr> wow - the first static dietsvn binaries start to materialize
20:47 < karasz> where can i find thoose hooks?
20:47 < mnemoc> scripts/functions.in
20:47 < rxr> 1507444 Jan 4 20:47 sv
20:48 < rxr> `- now that is a fat binary ...
20:48 < CIA-1> aldas * r14493 /trunk/package/network/mod_security/ (mod_security.conf mod_security.desc): * updated mod_security (1.8.7 -> 1.9.1)
20:48 < mnemoc> rxr: static?
20:49 < rxr> dietlibc static
20:49 < rxr> all inclusive
20:49 < rxr> and not functional
20:49 < mnemoc> *G*
20:49 < rxr> # ./svn
20:49 < rxr> svn: error: cannot set LC_ALL locale
20:49 < rxr> svn: error: environment variable LANG is not set
20:49 < rxr> svn: error: please check that your locale name is correct
20:49 < misl> mnemoc: have you tried building the latest kaffe on noisy? Seems like the dual processor issues is solved. I unfortunately lack the time to give it a spin :(
20:50 < rxr> misl: what is the power on time of noisy currently, up at night or down?
20:50 < misl> at night down
20:50 < rxr> btw. have I wished a happy new year to you and your family alread ?
20:51 < mnemoc> misl: i'm waiting for rc1, planned for tonight
20:51 < rxr> have new year over there if I missed it so far ,-)!
20:51 < misl> :)
20:51 < mnemoc> misl: --with-system-zlib and --with-fastjar
20:51 < misl> mnemoc: ooehh, exiting
20:51 < mnemoc> promising :)
20:51 < CIA-1> aldas * r14494 /trunk/package/database/postgresql/ (5 files): * updated postgresql (8.0.4 -> 8.0.5)
20:51 < misl> rxr: noisy is on from about 7:30 till about 19:30
20:52 < misl> CET that is.
20:52 < misl> exiting = exciting
20:52 < misl> hehe, funny typeo
20:53 < misl> hmm :( another one
20:53 < karasz> Baldzius, ....
20:54 < Baldzius> a?
20:54 < mnemoc> misl: i undertood it :)
20:55 < misl> rxr: we mentioned this new year already. :) However I wish you and all t2-ers a happy and prosperous new year
20:55 < misl> hehe: undertood == undertoad?
20:55 < mnemoc> :)
20:55 < rxr> misl: all fine on your side ?
20:56 < misl> seems like this year in the year of the broken keyboards :)
20:56 < misl> rxr: no I am overworked
20:56 < mnemoc> rebel fingers
20:56 < rxr> misl: oh :-(
20:56 < misl> mnemoc: always blame something or someone else :)
20:56 < rxr> # ./svn
20:56 < rxr> Type 'svn help' for usage.
20:57 < rxr> # svn st
20:57 < rxr> Segmentation fault
20:57 < rxr> well, ok
20:57 < misl> ahh, stupid me, you blamed your fingers.
20:57 < rxr> not reallyready
20:57 < mnemoc> neon?
20:57 < rxr> # svn co http://svn.exactcode.de/hotplug++/trunk hotplug++-trunk
20:57 < rxr> Segmentation fault
20:57 < rxr> hm
20:58 < rxr> ok
20:58 < rxr> -ENOCARE for now ...
20:58 < rxr> mnemoc: dietsvn ...
20:58 < mnemoc> :)
20:58 < misl> rxr: overworking is not really a problem. The old project will soon be finished. But the new one is waiting in line already.
20:58 < rxr> misl: nice, money for the living wise ,-)
20:59 < misl> :)
20:59 -!- madtux [i=miguel@pf0.hostarica.com] has quit ["Leaving"]
20:59 < misl> I am off to bed now.
21:00 < misl> sleep well all.
21:00 < Baldzius> gn8 misl :)
21:00 -!- misl [n=chatzill@82-217-66-150.cable.quicknet.nl] has quit ["Chatzilla 0.9.68a [Firefox 1.0.6/20050717]"]
21:00 < Baldzius> karasz ... ?
21:00 < rxr> /we off into bed as well
21:00 < Baldzius> :)
21:00 < Baldzius> gn8 rxr :)
21:01 < karasz> Baldzius, why 8.0.5?
21:01 < karasz> cu rxr?
21:01 < CIA-1> rene * r14495 /trunk/package/develop/ (apr/apr.desc subversion/subversion.desc):
21:01 < CIA-1> * marked apr and subversion DIETLIBC, they build for me, manually -
21:01 < CIA-1> though svn segfaulted
21:01 < Baldzius> karasz, you supposed to see 8.1.1 ?
21:01 < karasz> yup :D
21:02 < karasz> it is trunk isn't it...
21:02 < karasz> anyway 8.1.1 was on my todo
21:02 < Baldzius> i'am preparing here masive BASE update like apache 2.2 subversion 1.3.0 mysql 5.0.18 postgresql 8.1.1 and so on ......
21:02 < karasz> cause i have a slightly different view on postgresql than mnemoc
21:03 < morfoh> karasz: but it's nice from Baldzius doing so because it's better for the stable tree maintainer ;)
21:03 < karasz> morfoh, ?
21:04 < mnemoc> what's my view on postgresql?
21:04 < morfoh> merging packages
21:04 < karasz> i explained to you sometimes that i am not very happy with that sysvinit script
21:04 < mnemoc> merging what?
21:05 < morfoh> mnemoc: nothing
21:05 < Baldzius> :)
21:05 < mnemoc> karasz: i don't give a sh* about .init files, they suck by design
21:05 < karasz> i know that mnemoc
21:05 < karasz> but some people still use it... :D
21:05 < mnemoc> and when i get time i'll remove them all
21:06 < karasz> replacing with what?
21:06 < karasz> runit?
21:06 < mnemoc> some multi-init declaration format
21:06 < mnemoc> aka 'tail'
21:06 < morfoh> karasz: I just said that it is nice from Baldzius doing updates in small steps even if there are more recent versions out there, because it will ease the process of merging and testing the package on a stable tree
21:07 < karasz> ack morfoh
21:07 < mnemoc> eyah
21:07 < mnemoc> yeah
21:07 < morfoh> now I shut up :p
21:08 < mnemoc> good
21:08 < Baldzius> :)))))))))))
21:08 < mnemoc> *g*
21:09 < mnemoc> Baldzius is a well trained updater :)
21:09 < karasz> o.o
21:09 < mnemoc> karasz... i'm waiting for his firsts commit
21:09 < Baldzius> thanks ;)
21:09 < karasz> karasz... i'm waiting for his firsts commit ?
21:09 < karasz> Baldzius's?
21:10 < mnemoc> karasz's
21:10 < karasz> i think he jhas some already
21:10 < karasz> patience mnemoc
21:10 < karasz> small steps :D
21:10 < karasz> as you (and usualy I) like it
21:11 < mnemoc> yes, you can fix typos on .desc files at the beginning :)
21:11 < karasz> :D))
21:12 < karasz> `nothing is too small if you are humble`
21:12 < karasz> (TM)
21:13 < mnemoc> so, i'm waiting for your first commit
21:14 < karasz> yup
21:14 < karasz> will come
21:14 < karasz> first night after i got installed
21:15 < karasz> should thoose utils go to /multimedia?
21:15 < mnemoc> .oO( what utils? )o
21:15 < karasz> they transform various images to iso files...
21:15 < karasz> nrg's, ccd's and so on to .iso
21:15 < mnemoc> ehm uhm... filesystem i guess
21:16 < karasz> i thought multimedia because cdrecord is there too...
21:16 < mnemoc> ehm
21:16 < mnemoc> i missed that
21:16 < mnemoc> i didn't know cdrecord played media
21:17 < karasz> checkin' right now
21:18 < karasz> cdrtools is in base
21:20 < rxr> cdrtools is in base
21:20 < rxr> which is wrong
21:20 < rxr> I would like to have it in filesystem or so
21:20 < mnemoc> :)
21:20 < mnemoc> openldap is at base on rock now :)
21:21 < rxr> other stuff should go out of base as welll
21:21 < rxr> rocknet -> network
21:21 < rxr> and we should open s.th. like system or hardware
21:21 < rxr> hotplug* and all the raid and so on utils should go out of base ...
21:21 < rxr> and gcc into devel, no?
21:22 < karasz> hmm desktop finished
21:22 < mnemoc> don't mix categories with repositories
21:22 < mnemoc> hardware, yes
21:22 < mnemoc> but system? what is that?
21:23 < karasz> mnemoc, regression in a minute
21:23 < karasz> :)
21:23 < karasz> i am only concerned by iptables...
21:24 < mnemoc> uhm?
21:24 < karasz> but we'll see
21:25 < YeT2> something i still cannot understand : what about rock ... or: what is the reason for the new development fork T2
21:25 < CIA-1> aldas * r14496 /trunk/package/gnome2/glib/ (6 files): * updated glib (2.8.4 -> 2.8.5)
21:26 < YeT2> pls don't misunderstand me ... i like T2 ... but whats wrong with rock ?
21:26 < mnemoc> basicly missleading
21:26 < rxr> YeT2: there is a lot of reference about this on our site and the list, irc logs, ...
21:26 * rxr going to bed right now, otherwise I would even feel like repating it once again ,-)
21:26 < rxr> maybe mnemoc wants ,-)
21:27 < Baldzius> :)
21:27 < mnemoc> nah, i'm having lot of fun with 1-perl for dietlibc
21:28 < karasz> http://www.x5.ro/regression/desktop/regressions.html
21:28 < karasz> mnemoc, ...
21:28 < rxr> n8 all
21:29 < nzg> n8 rxr
21:29 < mnemoc> gn8 rxr
21:29 < karasz> night rxr
21:29 < YeT2> :) ... n8
21:29 < mnemoc> karasz: mark iptables NOPARALLEL and rebuild
21:30 < karasz> in .desc
21:30 < karasz> [?] NOPARALLEL...
21:31 < mnemoc> F
21:32 < karasz> did transset arived on gsmp?
21:33 < mnemoc> i think so
21:33 < karasz> i will -remove it also...
21:33 < mnemoc> ok
21:33 < mnemoc> but no need
21:33 < mnemoc> -job is enough
21:33 < karasz> right now i accidentaly typed a regression :(
21:34 < karasz> and waitin for it to finish...
21:36 < karasz> hmmm dld -required does not download transset...
21:36 < karasz> it acts like it should not be dlded...
21:37 < mnemoc> Download transset
21:37 < mnemoc> explicitcly
21:37 < mnemoc> when using -required remember to set -cfg as first arg
21:39 < YeT2> curl and shadow only available over original url not on mirror ?!
21:40 < YeT2> ok ... i suppose not replicated yet ..
21:40 < mnemoc> if they were updated today, yes.
21:41 < karasz> mnemoc, i learned that the hard way :)
21:41 < karasz> transset built
21:41 < CIA-1> aldas * r14497 /trunk/package/database/phppgadmin/ (phppgadmin.conf phppgadmin.desc): * updated phppgadmin (3.5.6 -> 4.0.1)
21:44 < karasz> iptables too
21:44 < karasz> now transfig hopefully
21:44 < karasz> good
21:44 < karasz> transfig also
21:45 < karasz> 5 database/bdb-java
21:45 < karasz> 5 database/bdb33-java
21:45 < karasz> thoose are all that failed
21:45 < mnemoc> karasz: there is any other way?
21:45 < karasz> looks that i know no other :)
21:45 < karasz> so mnemoc
21:45 < karasz> 5 database/bdb33-java
21:45 < karasz> 5 database/bdb-java
21:45 < mnemoc> don't worry about them, or pkgsel a closed java
21:45 < karasz> pkgsel a closed java?
21:46 < karasz> dummy's please
21:46 < mnemoc> there are non-free java implementation which can be used
21:46 < mnemoc> peek one, and pkgsel it in
21:46 < mnemoc> if you want a java
21:46 < karasz> but it won't prevent me
21:46 < karasz> a good install?
21:46 < karasz> i do not want jave
21:46 < mnemoc> nope
21:47 < karasz> just for firefox
21:47 < mnemoc> that is a yes
21:47 < karasz> oki
21:47 < mnemoc> the plugin need the fully non-free java crap
21:47 < karasz> waitin for boot to finish :)
21:47 < karasz> :(
21:48 < karasz> so what should i do?
21:50 < karasz> pkgsel what package?
21:50 < mnemoc> i wont suggest any non-free java
21:51 < mnemoc> take a look into package/java
21:51 < mnemoc> and pick one jdk which support your arch
21:52 < karasz> i hate java
21:52 < mnemoc> so, ignore it
21:53 < karasz> i'll make a try wit sun thou
21:53 < karasz> jdk or jre?
21:54 < karasz> oki
21:55 < karasz> second read allways is more conclusive
21:56 < YeT2> karasz : if you intend to run apache tomcat go for jdk 1.5 (jre does not suffice) !
21:56 < mnemoc> to build bdb-java you need a jdk too
21:57 < karasz> nice
21:57 < karasz> Downloading sun-jdk-150:download/local/j/jdk-1_5_0_04-linux-amd64.bin ...
21:57 < karasz> The file jdk-1_5_0_04-linux-amd64.bin can not be fetched automatically
21:57 < karasz> please visit: http://java.sun.com/j2se/1.5.0/download.jsp
21:57 < karasz> and download it manually into /root or somewhere else using -alt-dir
21:57 < karasz> please pay us a visit because we are sellin' clicks or something...
21:57 < YeT2> :) ... yeah ... very common to me tooo
22:02 < YeT2> mnemoc : there are still dependencies to sysfiles in package descriptions (eg . firmware/ipw2200) ... problem ?
22:02 < mnemoc> YeT2: sysfiles has to be always installed anyway
22:03 < YeT2> hmmm ... i understood it is split now (and blacklisted ?)
22:03 < mnemoc> YeT2: on the next .cache syncronization it may disappear
22:03 < mnemoc> YeT2: sysfiles is not emergeable, because it's too risky
22:03 < YeT2> uh.. ok then
22:04 < mnemoc> YeT2: sysfiles exists, but now new packages do part of it's previous job
22:04 < mnemoc> YeT2: mkinitrd and rocknet are packages, and rc was moved to sysvinit
22:04 < YeT2> pheeww ... (i was wondering if i had to go through fixing hell ...)
22:04 < mnemoc> YeT2: those are emergeables, and then updateable
22:05 < CIA-1> jsaw * r14498 /trunk/package/mail/bincimap/ (6 files):
22:05 < CIA-1> * add a patch that fixes x86-64 compilation
22:05 < CIA-1> * add a patch for some unitialized values (maybe send upstream?)
22:05 < CIA-1> * copyright update
22:05 < YeT2> menmoc: thx !
22:05 < mnemoc> :)
22:05 < mnemoc> F* 1-perl!
22:05 < mnemoc> F* perl!
22:06 < mnemoc> aaaarg....
22:07 < jsaw> somebody might want to send the uninitialized-fix.patch in bincimap upstream (I can't currently)
22:07 < mnemoc> ok, i do
22:08 < mnemoc> jsaw: and the 64bits one?
22:09 < jsaw> oh yeah, that also
22:10 < jsaw> mnemoc: the unitialized-fix is not serious as far as I can see, still it's more correct
22:14 < jsaw> btw, bincimap does the ssl on its own
22:14 < morfoh> jsaw: good to know ;)
22:14 < mnemoc> jsaw: TLS
22:14 < jsaw> now I only have to setup my own CA ...
22:15 < mnemoc> jsaw: www.cacert.org
22:16 < mnemoc> jsaw: i get my certs from there
22:16 < jsaw> how much ?
22:16 < mnemoc> 0
22:16 < jsaw> O.O
22:16 < jsaw> oh, really ?
22:16 < jsaw> but only for private use, or?
22:16 < mnemoc> of course
22:16 < mnemoc> nah
22:17 < morfoh> not restricted
22:17 < mnemoc> free world man
22:17 < morfoh> :)
22:17 < mnemoc> :)
22:17 < morfoh> mnemoc: I like it if you're sounding revolutionary ! :p
22:17 < jsaw> :)
22:17 < mnemoc> but people like morfoh still using self-signed certs
22:18 < mnemoc> morfoh: :p
22:18 < morfoh> thanks ... for the reminder :p
22:19 < jsaw> viva la revolucion...
22:19 < mnemoc> :)
22:19 < morfoh> mnemoc: but you're right ... it's a shame on my side :/
22:19 < mnemoc> i can forgive jsaw because he didn't know it, but you..... a shame
22:20 < morfoh> mnemoc: yeah! beat me again master ... mea culpa
22:21 * mnemoc hits morfoh, again
22:21 < mnemoc> jsaw: http://www.rafb.net/paste/results/DYJwFf19.html
22:22 < mnemoc> morfoh: http://www.rafb.net/paste/results/DYJwFf19.html
22:23 < mnemoc> but i can't get f* 1-perl building!
22:23 < mnemoc> (on trunk)
22:24 < karasz> mnemoc, do i need java-dirtree?
22:24 < mnemoc> karasz: for java, yes
22:25 < karasz> for my lousy firefox plugin...
22:25 < mnemoc> yes
22:25 < karasz> nice
22:25 < karasz> now i remembered why i hate java
22:30 < karasz> hmmm
22:31 < karasz> now i see that i do not have firefox selected nor konqueror...
22:31 < mnemoc> jsaw: mail sent
22:31 < jsaw> :)
22:31 < jsaw> click
22:31 < morfoh> cu later ... have to reboot
22:31 < karasz> konqueror is in some other package?
22:31 -!- morfoh [n=jeru@p54BEF73E.dip.t-dialin.net] has quit ["using sirc version 2.211+KSIRC/1.3.11"]
22:32 < mnemoc> karasz: kdebloat
22:32 < karasz> ?
22:33 < karasz> i grepped through /logs...
22:33 < karasz> no 5-konqueror.log
22:34 < mnemoc> kdebase
22:34 < jsaw> mnemoc: strace from segfault?
22:34 < mnemoc> i'm trying to find the line first :)
22:34 < mnemoc> Re-running configuration:
22:34 < mnemoc> -> Creating configuration script.
22:34 < mnemoc> -> T2 2.1-stable configuration ...
22:34 < mnemoc> -> Building src/config_helper.so.
22:34 < mnemoc> /usr/lib/gcc/i686-t2-linux-gnudietlibc/3.4.4/../../../../i686-t2-linux-gnudietlibc/bin/ld: start.o: No such file: No such file or directory
22:34 < jsaw> mnemoc: things I don't like are the implicit function declarations
22:34 < mnemoc> collect2: ld returned 1 exit status
22:34 < mnemoc> ^--- these from rebuilds is also fun
22:34 < mnemoc> this*
22:35 < mnemoc> jsaw: which implicit function declarations?
22:35 < jsaw> strchnul
22:35 < jsaw> basename
22:35 < jsaw> and get_current_dir_name
22:35 < jsaw> looks like gnu extensions
22:36 < mnemoc> and senseless
22:36 < mnemoc> +++ ccache -M 1000M
22:36 < mnemoc> Segmentation fault
22:37 < jsaw> basename is POSIX
22:37 -!- morfoh [n=jeru@p54BEF73E.dip.t-dialin.net] has joined #t2
22:37 < jsaw> get_current_ is GNU ext
22:37 < morfoh> hell
22:37 < jsaw> and there's not even a man page for strchrnul
22:38 < mnemoc> debian has to have one :)
22:38 < morfoh> seems my machine is "laming" :/
22:38 < mnemoc> as the owner
22:39 < morfoh> mnemoc: of course :/
22:40 < jsaw> mnemoc: I'll change the fl_wrapper now
22:41 < mnemoc> thanks
22:41 < jsaw> ********GGGGGGAAAAAAAAAAAAAAAAAAAAAAAHHHHHHHH*********
22:41 < jsaw> when I see sprintf without length check, my eyes start hurting
22:42 < mnemoc> :D
22:42 < jsaw> ah, not even PATH_MAX is used
22:44 < mnemoc> chars economy
22:44 < mnemoc> ==[apply_patchfiles:674 (last $?=1)> patch -bfp1 -z .orig
22:44 < mnemoc> Segmentation fault
22:44 < mnemoc> yucj
22:44 < mnemoc> yuck
22:44 < jsaw> furuncles growing on my eyes
22:44 < mnemoc> :)
22:47 -!- sparc-kly [n=mubex@64.237.252.223] has joined #t2
22:55 < jsaw> more furuncles growing on my eyes
22:57 < karasz> burn them jsaw
22:57 < karasz> :)
22:57 * jsaw renames fl_wrapper to furuncle_wrapper
22:57 < jsaw> and burnes it
22:57 < jsaw> -e
22:58 < mnemoc> :D
22:58 < mnemoc> yuck, no strace for dietlibc :(
23:02 < CIA-1> aldas * r14499 /trunk/package/gnome2/evolution/ (evolution.conf evolution.desc): * updated evolution (2.4.2 -> 2.4.2.1)
23:04 < YeT2> mnemoc: still having trouble with 1-perl ? ... anything special with your target ? ... just built on trunk with pentium M opt !
23:05 < morfoh> YeT2: mnemoc is trying to build on a dietlibc based target
23:05 < YeT2> ahh yess ... now i understand :)
23:13 < morfoh> 27658301 <-- nice size for a linux kernel image :p
23:13 < jsaw> O.O
23:13 < morfoh> 2680040 <-- after stripping it :p
23:13 < jsaw> ...
23:14 < morfoh> usermode linux kernel
23:14 < karasz> -> Forgotten packages summary:
23:14 < karasz> !> `- glibc32
23:14 < karasz> is it good or bad?
23:14 < karasz> -> `- Extracting glibc.tar.bz2 ...
23:14 < karasz> !> `- Not found in $package_map: glibc32
23:14 < karasz> !> ... fix target/bootdisk/build_stage2.sh
23:14 < karasz> ??
23:14 < karasz> mnemoc
23:16 < CIA-1> aldas * r14500 /trunk/package/gnome2/gob/gob.desc: * updated gob (2.0.12 -> 2.0.13)
23:16 < karasz> guys...
23:17 < jsaw> you should add it...
23:17 < jsaw> but don't think it hurts for bootdisk
23:17 < jsaw> but _I_
23:17 < karasz> more exactly...?
23:18 < jsaw> grep for "package_map"
23:18 < jsaw> in target/bootdisk/build_stage2.sh
23:18 -!- YeT2 [n=root@i53878831.versanet.de] has left #t2 ["Konversation terminated!"]
23:19 -!- YeT2 [n=root@i53878831.versanet.de] has joined #t2
23:19 < karasz> i found it jsaw
23:19 < karasz> i ad a +glibc32?
23:19 < jsaw> yep
23:19 < karasz> where?
23:20 < karasz> after +glibc?
23:20 < jsaw> where ever, but that's a good place
23:21 < karasz> done
23:21 < karasz> now?
23:21 < karasz> run build again?
23:26 < karasz> [02:20:01] Creating ISO-MD5 tools ...
23:26 < karasz> make: Entering directory `/usr/src/2.1.1/src/isomd5sum'
23:26 < karasz> gcc -c -O -fPIC -D_FILE_OFFSET_BITS=64 -O -Wall -Werror -o implantisomd5.o implantisomd5.c
23:26 < karasz> gcc -c -O -fPIC -D_FILE_OFFSET_BITS=64 -O -Wall -Werror -o md5.o md5.c
23:26 < karasz> gcc -c -O -fPIC -D_FILE_OFFSET_BITS=64 -O -Wall -Werror -o libimplantisomd5.o libimplantisomd5.c
23:26 < karasz> cc1: warnings being treated as errors
23:26 < karasz> libimplantisomd5.c: In function 'implantISOFile':
23:26 < karasz> libimplantisomd5.c:122: warning: pointer targets in passing argument 2 of 'parsepvd' differ in signedness
23:26 < karasz> libimplantisomd5.c:189: warning: pointer targets in passing argument 1 of 'snprintf' differ in signedness
23:26 < karasz> libimplantisomd5.c:190: warning: pointer targets in passing argument 2 of 'writeAppData' differ in signedness
23:26 < karasz> make: *** [libimplantisomd5.o] Error 1
23:26 < karasz> make: Leaving directory `/usr/src/2.1.1/src/isomd5sum'
23:28 < karasz> anyway
23:29 < karasz> i have 2 isos :D
23:29 < karasz> tomorrow is installin' day
23:29 < karasz> i hope they work...
23:29 < karasz> should i save my config, or some?
23:30 < karasz> if i want to use emerge latter?
23:30 < jsaw> karasz: are you on 2.1 ?
23:30 < karasz> now it is 22
23:30 < karasz> it will be 2.1
23:30 < jsaw> ???
23:30 < karasz> tomorrow
23:30 < jsaw> I thought I fixed the md5 sum thing...
23:31 < YeT2> jsaw: maybe in trunk ?
23:31 < karasz> it is a 2.1 build
23:31 < karasz> my installed system is 2.2.0 -epsilon i guess...
23:31 < jsaw> oh
23:32 < jsaw> mnemoc: you have to merge the isomd5 thing from trunk for x86-64
23:34 < YeT2> karasz: you are running an installed 2.2.0-epsilon ... but building for 2.1.0 ??? special reason for this ?
23:34 < karasz> i cannot get a desktop in 2.2.0
23:36 < karasz> that is because i am a little grashopper :)
23:36 < YeT2> ahh yess ... was my problem too ... therefore i (finally after trying different hosts) am building on a Knoppix 4.02 DVD installed with checkedout trunk !
23:37 < karasz> nah
23:37 < karasz> i wanted the real thingie
23:37 < karasz> T2 built on T2
23:38 < morfoh> VFS: Mounted root (squashfs filesystem) readonly. :)
23:38 < karasz> wakarimasen morSAN
23:38 < karasz> wakarimasen morfoh san even
23:39 < YeT2> well ... desktop is still in stage 1 ... yeah ... the real thingie ws my first attemp ... but i wanted a USB-DISK capable bootstrap system !
23:40 < YeT2> so ... I WILL HAVE MY REAL DESKTOP SYSTEM SOON (i hope)
23:40 -!- ojh [n=omer@70-56-27-70.eugn.qwest.net] has joined #t2
23:41 < YeT2> (another build is running on a gentoo ;) )
23:42 < karasz> kore wa nani `VFS: Mounted root (squashfs filesystem) readonly. :)` morfoh san
23:43 < karasz> nanika mondai aru?
23:43 < morfoh> karasz: what ? :)
23:44 < morfoh> karasz: I just test some things ;)
23:44 < karasz> nah, i thought you don't wanna play :)
23:44 < YeT2> ok ... i am not using mIrc !
23:45 < karasz> if you are testing morfoh then `zutto tsuzukete`
23:46 < YeT2> hmmm ... i don't speak swaheli :)
23:46 < karasz> me neider
23:46 < karasz> but i would like to...
23:47 < karasz> and YeT2 it is not swaheli
23:47 < karasz> it is japanese in romanji ;)
23:48 < YeT2> ooops ... sophisticated !
23:48 < YeT2> so zutto tsuzukete ?
23:49 < YeT2> what does it mean ?
23:49 < karasz> keep going
23:49 < morfoh> zutto tsuzukete karasz san ... then :p
23:49 < karasz> nah enough for today
23:49 < karasz> i am waitin for my cd's to burn...
23:50 < morfoh> karasz: yeah ... burn them ... :)
23:50 < karasz> but anyway arigato gozaimas morfoh san
23:51 < YeT2> f ... ?
23:51 < jsaw> can somebody check if uclibc has "strchrnul" function?
23:52 < morfoh> jsaw: mom
23:52 -!- YeT2 [n=root@i53878831.versanet.de] has quit ["Leaving..."]
23:53 < morfoh> jsaw: /usr/include/string.h
23:53 < jsaw> thx
23:54 < morfoh> jsaw: yw
23:54 < mnemoc> rehi
23:55 < morfoh> wb companero
23:55 < karasz> rehi mnemoc
23:55 < morfoh> oh! sorry ... I mean companero san! ... hei!
23:56 < karasz> even dou shiteru?
23:56 < mnemoc> :)
23:56 -!- YeT2 [n=chmo@i53878831.versanet.de] has joined #t2
23:57 < karasz> fingers crossed rebooting black-box...:)
--- Log closed Thu Jan 05 00:00:33 2006