T2 IRC Log: 2009-08-05
This is the log as captured by an IRC bot in the channel. The statements are those of the individual people and might not neccessarily reflect the policy and legal rules as set forth by the T2 SDE Project.
--- Log opened Wed Aug 05 00:00:38 2009
00:49 -!- mqueiros [n=mqueiros@217.70.75.151] has quit ["Ex-Chat"]
01:38 -!- mqueiros_ [n=mqueiros@217.70.75.151] has joined #t2
02:15 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: bitdance-vacatio
02:18 -!- Netsplit over, joins: bitdance-vacatio
08:26 < mjungwirth> moin
08:29 -!- mpp [n=mpp@2a01:48:251:301:21a:6bff:fe35:820a] has joined #t2
08:33 < mpp> moinsen
08:39 < mjungwirth> can i make a 64Bit System without /lib64 dir?
09:10 -!- T-One [n=T-One@62.218.122.66] has joined #t2
09:42 < rxr> moin
09:42 < rxr> mjungwirth: you mean without 32bit libs in /lib ?
09:43 < mjungwirth> yes
09:43 < mjungwirth> right
09:43 < rxr> with some drawbacks, yes
09:43 < rxr> e.g. you could not have grub as boot loader
09:43 < rxr> lilo would work
09:43 < mjungwirth> hmm i use syslinux
09:43 < rxr> and you might have to patch some hardcoded packages
09:43 < rxr> well
09:43 < rxr> it's some config settings but it's possible
09:44 < rxr> some tinkering and trial'n error
09:44 < rxr> I do some simillar tricks for sparc64, e.g. build with a pure 32bit system by default with only very few 64bit binaries
09:44 < rxr> (like the other way round the normal on amd64
09:44 < rxr> )
09:45 < mjungwirth> ok i use the /lib64 :
09:45 < mjungwirth> )
09:46 < rxr> http://www.t2-project.org/packages/dosemu.html
09:46 < rxr> http://www.t2-project.org/packages/dosbox.html
09:52 < mjungwirth> hmm i build a 64 Bit System. On Stage 1 "gcc package" i get the following error: crti.o: No such file: No such file or directory
09:52 < mjungwirth> any solution?
09:52 < rxr> it's decades since I got something like that
09:52 < rxr> maybe no glibc32 or so ?
09:53 < rxr> you try a pure 64bit build ?
09:53 < rxr> you need to force disable multilib for that case
09:53 < mjungwirth> yes without glibc32
09:53 < rxr> as I said: a pure 64bit build is possible with some trial'n error to touch the right places
09:53 < rxr> is this for some product or so? to minimize space ?
09:54 < mjungwirth> ok i test it with glibc32
10:01 < mjungwirth> yea i received my new server only for t2-system builds (Intel I7, 8GB RAM, 4x256GB SSD Samsung with Adaptec Raid Controler)
10:02 < mjungwirth> what is the best distribution for the t2 project ?
10:02 < rxr> you mean to build t2 ?
10:02 < mjungwirth> yes
10:02 < mjungwirth> i currently use the ubuntu server
10:02 < mjungwirth> LTS
10:02 < rxr> well, I mostly use T2 itself to build T2 :-)
10:02 < mjungwirth> :)
10:03 < rxr> btw.
10:03 < rxr> berliner morgenpost today has a windows 7 article
10:03 < rxr> thes had a photo of a Dell running "windows 7", though ti was KDE 4!
10:03 < rxr> with the sub-line like "die visuallen veränderungen sah man schon seit der ersten beta" :-)
10:03 < mjungwirth> :)
10:04 < rxr> I have the article with me, should scan it :-)
10:04 < mjungwirth> yea send it to me please :) i hang it out on our firm-black-table
10:06 < rxr> ok - should do some work first
10:06 < rxr> will do later
10:07 < mjungwirth> ok thanks
10:07 < rxr> probably have to get a a3 scanner off the shelf
10:08 < mjungwirth> btw. after i add the glibc32 gcc build correctly. Thanks
10:08 < rxr> your're welcoem
10:08 < rxr> if you want a lib(32) less build you can disable the multilib option
10:09 < rxr> it's default is automatically the "sane choice" per architecture
10:09 < rxr> so you can not change it in the UI, and have to overwrite it in a target config
10:09 < rxr> some years ago it worked to build a pure 64bit environment
10:09 < rxr> without the stuff that requires 32bit though
10:09 < rxr> like grub, wine, maybe some dos-emu stuff etc.
10:13 < rxr> (lilo and others work, though)
10:25 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has joined #t2
10:25 < Baldzius> moin
10:25 < rxr> moin Baldzius
10:25 < Baldzius> hey rxr
10:25 < Baldzius> rxr , you've started a fresh build recently right?
10:26 < rxr> yes
10:26 < rxr> but I had to fsck my volume over night
10:26 < rxr> becuase I could not wipe one build with some magic defect file
10:26 < Baldzius> gettext passes on cross stage for you?
10:26 < rxr> I guess you want to ask about attr ?
10:26 < rxr> oh - gettext, yes
10:26 < rxr> but I have some changes to the package lingering
10:26 < Baldzius> ic
10:27 < Baldzius> as it fails with autoconf-2.64
10:27 < rxr> oh
10:27 < Baldzius> you've updated yesterday I think
10:27 < rxr> I patched to not autoregen
10:27 < rxr> anyway, ages ago
10:27 < Baldzius> not sure
10:27 < Baldzius> hm
10:27 < rxr> because it took too much CPU and broke powerpc
10:27 < rxr> I can commit now
10:27 < Baldzius> just was wondering
10:27 < Baldzius> which autoconf version you have on host?
10:27 < Baldzius> cause this might be a reason
10:27 < rxr> oh no idea
10:28 < rxr> # mine -q autoconf
10:28 < rxr> autoconf 2.59 8.0-trunk
10:28 < rxr> but I can commit this change anyway
10:28 < rxr> as it shoudl fix powerpc and lower cpu useage
10:28 < rxr> by not endlessly rebuilding so much auto* junk
10:28 < Baldzius> so it's not the reason
10:28 < Baldzius> I have 2.61
10:28 < Baldzius> hm
10:28 < Baldzius> if you commit
10:28 < Baldzius> I would test it
10:28 < rxr> done
10:29 < rxr> Committed revision 33780.
10:29 < rxr> CIA-69: ?
10:29 < CIA-69> rene * r33780 /trunk/package/base/gettext/ (gettext-no_long_long.patch gettext.conf gettext.desc): * do not auto-rebuild gettext, unbreaks powerpc, saves CPU cycles
10:29 < Baldzius> thanks
10:29 < Baldzius> will check it
10:29 < rxr> thanks
10:31 < rxr> I intend to commit the next glibc major update later, likewis
10:31 < rxr> e
10:33 < Baldzius> 2.10.1?
10:33 < rxr> yep
10:33 < rxr> * updated glibc (2.9 -> 2.10.1)
10:33 < rxr> * updated glibc32 (2.9 -> 2.10.1)
10:57 < CIA-69> rene * r33781 /trunk/package/archiver/cpio/ (cpio.desc no_rmt.patch): * updated cpio (2.9 -> 2.10)
11:02 < Baldzius> thanks rxr, your commit fixed gettext
11:02 < rxr> nice
11:15 < mjungwirth> rxr: would you like to extend you hardware/server section on the t2-project.org?
11:16 < rxr> sure
11:16 < mjungwirth> hmm i have the hp dl360 and dl380 (g4, g5, g6) servers tested
11:18 < rxr> nice
11:19 < rxr> if you want to write the entries :-)
11:19 < mjungwirth> i send you a mail :)
11:20 < mjungwirth> i get a new hp blade g6 system at 01.10.2009. i think i must test it with t2 :)
11:26 -!- jncneo [n=alex@91.195.10.75] has joined #t2
11:27 < jncneo> hi all!
11:27 -!- _| [n=joeDeuce@pdpc/supporter/student/joeDeuce] has quit [Read error: 110 (Connection timed out)]
11:27 < jncneo> i have a problem building uclibc in embedded configuration. could anyone help me?
11:29 -!- _| [n=joeDeuce@pdpc/supporter/student/joeDeuce] has joined #T2
11:33 < rxr> maybe
11:33 < rxr> mjungwirth: thanks
11:34 < CIA-69> rene * r33782 /trunk/package/xorg/xf86-video-siliconmotion/xf86-video-siliconmotion.desc: * updated xf86-video-siliconmotion (1.7.2 -> 1.7.3)
11:34 < CIA-69> rene * r33783 /trunk/package/xorg/xf86-video-trident/xf86-video-trident.desc: * updated xf86-video-trident (1.3.2 -> 1.3.3)
11:39 < jncneo> here is log: http://jncneo.nm.ru/log
11:41 < CIA-69> rene * r33784 /trunk/package/graphic/gocr/gocr.desc: * updated gocr (0.47 -> 0.48)
11:41 < CIA-69> rene * r33785 /trunk/package/gnome2/gstreamer/gstreamer.desc: * updated gstreamer (0.10.23 -> 0.10.24)
11:43 < CIA-69> rene * r33786 /trunk/package/emulators/dosemu/dosemu.desc: * updated dosemu (1.3.4 -> 1.4.0)
11:48 -!- regomodo [n=kvirc@host86-158-97-148.range86-158.btcentralplus.com] has joined #t2
11:53 [Users #t2]
11:53 [@ChanServ ] [ CIA-69] [ mjungwirth] [ oliv` ] [ T-One]
11:53 [ _| ] [ dsoul ] [ mpp ] [ regomodo] [ TobiX]
11:53 [ Baldzius ] [ jncneo] [ mqueiros_ ] [ rxr ]
11:53 [ bitdance-vacatio] [ koan ] [ mtr ] [ Stealth ]
11:53 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
11:53 < CIA-69> rene * r33787 /trunk/package/network/dhcp/dhcp.desc:
11:53 < CIA-69> Mario Jungwirth:
11:53 < CIA-69> * updated dhcp (4.1.0p1 -> 4.1.1b1)
11:55 < CIA-69> rene * r33788 /trunk/package/emulators/dosbox/ (dosbox.desc gcc43.patch): * updated dosbox (0.72 -> 0.73)
11:55 < CIA-69> rene * r33789 /trunk/package/audio/gnupod/gnupod.desc: * updated gnupod (0.99.7 -> 0.99.8)
11:55 < CIA-69> rene * r33790 /trunk/package/network/duplicity/duplicity.desc: * updated duplicity (0.6.02 -> 0.6.04)
12:05 < CIA-69> rene * r33791 /trunk/package/base/ (glibc/glibc.desc glibc32/glibc32.desc):
12:05 < CIA-69> * updated glibc (2.9 -> 2.10.1)
12:05 < CIA-69> * updated glibc32 (2.9 -> 2.10.1)
12:08 < CIA-69> rene * r33792 /trunk/package/python/gst-python/gst-python.desc: * updated gst-python (0.10.15 -> 0.10.16)
12:08 < CIA-69> rene * r33794 /trunk/package/x11/dmenu/dmenu.desc: * updated dmenu (3.6 -> 4.0)
12:09 < CIA-69> rene * r33793 /trunk/package/x11/dwm/dwm.desc: * updated dwm (5.6 -> 5.6.1)
12:10 < CIA-69> aldas * r33795 /trunk/package/base/coreutils/ (4 files): * updated coreutils (7.2 -> 7.4)
12:10 < Baldzius> rxr: could you test coreutils
12:11 < rxr> ouh
12:11 < rxr> yes, will
12:11 < Baldzius> I've removed both cross patches as it compiles fine
12:11 < Baldzius> should be fine I guess
12:12 < rxr> oh I hope too
12:12 < rxr> btw. regarding attr/acl
12:12 < rxr> I vaguely remeber that they once broke with some utils update
12:12 < rxr> like coreutils
12:12 < rxr> maybe you get an error that I don't with those too packages because some basic util update ?
12:12 < Baldzius> I will try that
12:12 < rxr> but I'll see with my all new builds here, too
12:12 < Baldzius> host has now 7.2 and it doesn't help
12:13 < Baldzius> I need to play with it a little bit
12:15 < jncneo> have anyone ever built uclibc from trunk in embedded-minimal configuration?
12:15 < rxr> probably, some weeks ago
12:15 < jncneo> was it ok?
12:16 < rxr> sure, we continously adapt stuff to keep building in regards to upstream changse
12:16 < rxr> I wanted to view your log, but your server did not mark it as content encoding text, so my browser wanted to save
12:16 < rxr> and I was too lazy to save and open an ext. editor when I usually view such stuff directly in my browser ...
12:16 < jncneo> oh, i'll check it now
12:17 < rxr> but usually breakage is easy to fix / adapt if one has a little source, edit, compile experience
12:17 < jncneo> i have a little experience in other distributions but i'm using T2 first time
12:18 < rxr> yeah - I meant "general experience" :-)
12:19 < jncneo> http://jncneo.nm.ru/log.txt not it's ok in my firefox
12:20 < rxr> ok - that are "just" shared files, installed by glibc and uclibc
12:20 < rxr> which config did you build?
12:21 < jncneo> embedded minimal
12:21 < rxr> with no option in the Config changed ?
12:23 < jncneo> aslo i've checked optimization to AMD Geode LX and "Print Build-Output to terminal while building"
12:23 < rxr> ok - they are unproblematic, but nothing else touched in the "Config" ?
12:24 < jncneo> nothing
12:24 < rxr> ok will testbuild a little later
12:24 < jncneo> thanks, i'll wait
12:27 < rxr> == 08/05/09 12:35:55 =[1]=> Finished building package coreutils.
12:29 < Baldzius> cool
12:30 < rxr> ^- x86
12:30 < rxr> == 08/05/09 12:38:05 =[3]=> Finished building package coreutils.
12:30 < rxr> ^- x86-64
12:31 < rxr> mostly fresh from scratch builds, with only todays updates intermixed
12:33 < rxr> will also start fresh ppc builds before I go for lunch
13:06 < regomodo> When was t2-7.0 released?
13:32 < regomodo> Hi, i'm doing a crosscompile trunk minimal build and I am stuck at apr-utils. The error is "checking for Berkeley DB 4.4 in /usr/include/db4... directory not found" but I have BerkeleyDB installed?
13:37 < rxr> hm
13:37 < rxr> maybe it needs some adaption to cross compile again
13:40 < rxr> Baldzius: so, fully fresh x86-64 build is in stage 5
13:55 < regomodo> rxr: I got rid of the configure error by creating a symlink between /usr/include/4.6 and /usr/include/db4 but that brings up another error
13:57 < regomodo> i686-t2-linux-gnu-gcc: /home/jon/t2sde/t2-trunk/build/current-8.0-trunk-generic-x86-i686-cross/usr/lib64/libapr-1.so: No such file or directory
13:57 < regomodo> make[1]: *** [libaprutil-1.la] Error 1"
14:02 < regomodo> Has anybody got a x-compile to work ever? I can't get one to work in 7.0 or trunk even with the minimal profile/
14:04 < rxr> regomodo: sure, regularly using it
14:05 < regomodo> Lucky you. I've been thinkng maybe it's the host's toolchain so i've tried Gentoo x86/amd64, Fedora x86, Ubuntu x86 and Debian amd64. I'm about to give up and go back to LFS.
14:08 < rxr> what is your problem?
14:08 < rxr> what do you need aprutil for in a normal build? not really X related ...
14:10 < regomodo> rxr: I've chosen the minimal profile, without X, and it's being pulled in.
14:10 < rxr> aha
14:10 < rxr> so you just want X?
14:10 < rxr> maybe try with the currently cross broken offenders disabled? like this aprutil ?
14:10 < rxr> though soonder or later I'll probably also hit it and fix
14:12 < regomodo> rxr: I don't want X as i've found I can't get that to build (libxcb issues) so I just want something that builds and installs
14:12 < rxr> I forgot, why do you cross build at all?
14:13 < regomodo> I'm on Gentoo amd64 and wan't to build an i686 install
14:13 < rxr> if you build for common and fast silicon you have a magnitutes easier life with a native build
14:13 < rxr> yeah - but still you can native build that
14:13 < rxr> your x86-64 kernel will run the 32 bit binaries
14:13 < regomodo> Really?
14:13 < rxr> (usually, at least, unless gentoo disabled the COMPAT32 thing in the linux kernel by default)
14:13 < rxr> sure
14:14 < regomodo> wow, i'm off to try that
14:14 < rxr> in the "non cross" case T2 only cross builds the bare minimum toolchain and native builds the rest in the chroot sandbox
14:14 < rxr> as you noticed some stupid upstream changes sometimes break (regress) some of the more advanced packages to cross build
14:14 < regomodo> so i'd use x-compile between x86 and ppc thern?
14:14 < rxr> yep
14:14 < rxr> or arm, sparc etc.
14:15 < regomodo> damn
14:15 < rxr> only if your CPU silicon can not run the resulting binaries
14:18 < regomodo> i'm sure a c2d should be able to run i686 binaries. I just checked my .config and "IA32_EMULATION" is enabled
14:18 < rxr> was that the name?
14:18 < rxr> remebered some with COMPAT
14:18 < rxr> whatever
14:19 < rxr> CONFIG_IA32_EMULATION=y
14:19 < rxr> indeed
14:19 < rxr> strange
14:24 < mjungwirth> rxr: i have found a bug (build by me) in the vsftpd.conf. (Build abort on 64Bit system). i send you the correct vsftpd.conf
14:37 < CIA-69> rene * r33796 /trunk/package/network/vsftpd/vsftpd.conf:
14:37 < CIA-69> Mario Jungwirth:
14:37 < CIA-69> * fixed vsftpd for lib64 builds
14:37 < mjungwirth> thanks
14:38 < jncneo> rxr: have you checked uclibc? I've made few more tries but with no results :(
14:39 < rxr> I will now
14:39 < jncneo> thanks
14:41 < rxr> jncneo: what exactly did you select?
14:41 < rxr> when I select embedded / uclibc then I get no glibc selected
14:43 < jncneo> i've seleced: embedded-minimal, install, x85, geodelx, grub, cross-build, "Continue on package error after stage 4", Keep src dirs on build errors, (1 ) Maximum number of jobs to be executed in parallel,Use ANSI colors for terminal output and Print Build-Output to terminal while building
14:43 < jncneo> there were no checkbox with glibc or uclibc...
14:44 < rxr> ah! embedded-minimal! one second
14:44 < mjungwirth> rxr: can you add my samba fix_samba_mtab patch to the svn?
14:45 < jncneo> ok
14:46 < mjungwirth> i can't build the new samba version 3.4.0! The patchfile "configure.patch.cross" has the incorrect patch-directory for version 3.4.0
14:46 < mjungwirth> i fixed it and send per mail
14:49 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has quit [Read error: 104 (Connection reset by peer)]
14:50 < rxr> mjungwirth: when?
14:51 < mjungwirth> moment my mail-server is angry :)
14:52 < mjungwirth> ok is out
14:52 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has joined #t2
14:54 < CIA-69> rene * r33797 /trunk/package/network/samba/configure.patch.cross:
14:54 < CIA-69> Mario Jungwirth:
14:54 < CIA-69> * updated samba cross compile patch
14:55 < rxr> mjungwirth: you will get a conflict, I added the proper directory name and version for us humans as reference
14:56 < CIA-69> rene * r33798 /trunk/target/embedded-minimal/ (config.in pkgsel):
14:56 < CIA-69> * fixed embedded-minimal by selecting the system libc with the proper
14:56 < CIA-69> config option to avoid implicitly selcting another (like glibc)
14:56 < rxr> jncneo: the last commit was for your config case
14:56 < mjungwirth> but the directory has changed from source to source3
14:56 < rxr> mjungwirth: yes, but you had ./ as prefix
14:56 < rxr> I changed the . to the samba-3.4.x/
14:57 < rxr> to better reflect which version the patch was against
14:57 < jncneo> understood. i'll check right now
14:57 < rxr> so it should all be well, apply and such
14:57 < rxr> the first directory (your . my samba-$ver) will be thrown away by patch -p1 anyway ...
14:57 < rxr> this is just academic information ...
14:57 < mjungwirth> i will remember
14:58 < rxr> (though I also sometimes add ./ patches ... :-)
14:58 < mjungwirth> :)
14:58 < rxr> it's always nicer to have the proper directory prefix
14:58 < mjungwirth> i'm not alone :)
14:58 < rxr> though I'm not linus torvalds and ask for a patch resend with proper signed-off line :-)
14:58 < rxr> I edit in-place and commit :-)
14:59 < rxr> even accept inline, attachment, .zip, .tar, replacement files etc. :-)
15:00 < rxr> jncneo: you need to start a clean build to get rid of the conflicting glibc files
15:00 < rxr> jncneo: my test build is still running though - only verified the resulting .config before the commit
15:01 < jncneo> understood, thanks. i'll also make test and i'll say you if smth wrong
15:33 < rxr> initial builds with the new glibc and trunk:HEAD look fine on ppc, ppc64, likewise
15:33 < koan> maybe I should try T2 again on my G5
15:34 < koan> will you publish a new build soon?
15:57 < rxr> I should :-)
15:58 < rxr> cpu : PPC970MP, altivec supported
15:58 < rxr> clock : 2000.000000MHz
15:58 < rxr> revision : 1.0 (pvr 0044 0100)
15:58 < rxr> model : PowerMac11,2
16:04 < rxr> http://www.t2-project.org/packages/emacs.html
16:31 -!- mpp [n=mpp@2a01:48:251:301:21a:6bff:fe35:820a] has quit ["Verlassend"]
17:21 < jncneo> rxr: i have one more error during configuring dropbear:
17:21 < jncneo> checking for deflate in -lz... no
17:21 < jncneo> configure: error: *** zlib missing - install first or check config.log ***
17:21 < jncneo> Due to previous errors, no 1-dropbear.log file!
17:22 < jncneo> but i have installed zlib on my system...
17:27 -!- |regomodo| [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has joined #t2
17:43 -!- regomodo [n=kvirc@host86-158-97-148.range86-158.btcentralplus.com] has quit [Read error: 110 (Connection timed out)]
17:54 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has quit ["Leaving"]
18:49 < |regomodo|> Hi, I've finally built a minimal install but when I go to build the iso I get this error:
18:49 < |regomodo|> [19:44:36] Creating iso/current_cd1.iso ...
18:49 < |regomodo|> Bad Option '-q' (error -1 BADFLAG).
18:49 < |regomodo|>
18:52 < |regomodo|> Hmm, I went into the Create-ISO script and got rid of the "-q" flag OPT on line 326
19:31 < |regomodo|> I'm guessing "-q" is for quiet output. Just checked the man pages and that doesn't exist. You have to use "-quiet"
19:40 < koan> jncneo: maybe you need zlib-dev?
19:52 -!- PassingStranger [n=cat@p54BF7B9D.dip.t-dialin.net] has joined #t2
19:56 < |regomodo|> I've just tried to build a minimal+x install in trunk and although it says it's completed it has skipped around 90 binary packages. Should I just ignore that or is that pretty bad?
21:02 -!- PassingStranger [n=cat@p54BF7B9D.dip.t-dialin.net] has quit ["Leaving"]
21:13 -!- |regomodo| [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has quit [Remote closed the connection]
21:31 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has joined #t2
21:32 -!- koan [n=koan@unaffiliated/koan] has quit ["Lost terminal"]
21:32 -!- koan [n=koan@unaffiliated/koan] has joined #t2
22:35 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has quit ["Changing server..."]
22:36 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has joined #t2
22:37 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has quit [Client Quit]
22:38 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has joined #t2
22:40 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has quit [Client Quit]
23:17 -!- regomodo [n=jon@host81-159-18-101.range81-159.btcentralplus.com] has joined #t2
23:36 -!- regomodo [n=jon@host81-159-18-101.range81-159.btcentralplus.com] has quit [Remote closed the connection]
--- Log closed Thu Aug 06 00:00:38 2009
00:49 -!- mqueiros [n=mqueiros@217.70.75.151] has quit ["Ex-Chat"]
01:38 -!- mqueiros_ [n=mqueiros@217.70.75.151] has joined #t2
02:15 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: bitdance-vacatio
02:18 -!- Netsplit over, joins: bitdance-vacatio
08:26 < mjungwirth> moin
08:29 -!- mpp [n=mpp@2a01:48:251:301:21a:6bff:fe35:820a] has joined #t2
08:33 < mpp> moinsen
08:39 < mjungwirth> can i make a 64Bit System without /lib64 dir?
09:10 -!- T-One [n=T-One@62.218.122.66] has joined #t2
09:42 < rxr> moin
09:42 < rxr> mjungwirth: you mean without 32bit libs in /lib ?
09:43 < mjungwirth> yes
09:43 < mjungwirth> right
09:43 < rxr> with some drawbacks, yes
09:43 < rxr> e.g. you could not have grub as boot loader
09:43 < rxr> lilo would work
09:43 < mjungwirth> hmm i use syslinux
09:43 < rxr> and you might have to patch some hardcoded packages
09:43 < rxr> well
09:43 < rxr> it's some config settings but it's possible
09:44 < rxr> some tinkering and trial'n error
09:44 < rxr> I do some simillar tricks for sparc64, e.g. build with a pure 32bit system by default with only very few 64bit binaries
09:44 < rxr> (like the other way round the normal on amd64
09:44 < rxr> )
09:45 < mjungwirth> ok i use the /lib64 :
09:45 < mjungwirth> )
09:46 < rxr> http://www.t2-project.org/packages/dosemu.html
09:46 < rxr> http://www.t2-project.org/packages/dosbox.html
09:52 < mjungwirth> hmm i build a 64 Bit System. On Stage 1 "gcc package" i get the following error: crti.o: No such file: No such file or directory
09:52 < mjungwirth> any solution?
09:52 < rxr> it's decades since I got something like that
09:52 < rxr> maybe no glibc32 or so ?
09:53 < rxr> you try a pure 64bit build ?
09:53 < rxr> you need to force disable multilib for that case
09:53 < mjungwirth> yes without glibc32
09:53 < rxr> as I said: a pure 64bit build is possible with some trial'n error to touch the right places
09:53 < rxr> is this for some product or so? to minimize space ?
09:54 < mjungwirth> ok i test it with glibc32
10:01 < mjungwirth> yea i received my new server only for t2-system builds (Intel I7, 8GB RAM, 4x256GB SSD Samsung with Adaptec Raid Controler)
10:02 < mjungwirth> what is the best distribution for the t2 project ?
10:02 < rxr> you mean to build t2 ?
10:02 < mjungwirth> yes
10:02 < mjungwirth> i currently use the ubuntu server
10:02 < mjungwirth> LTS
10:02 < rxr> well, I mostly use T2 itself to build T2 :-)
10:02 < mjungwirth> :)
10:03 < rxr> btw.
10:03 < rxr> berliner morgenpost today has a windows 7 article
10:03 < rxr> thes had a photo of a Dell running "windows 7", though ti was KDE 4!
10:03 < rxr> with the sub-line like "die visuallen veränderungen sah man schon seit der ersten beta" :-)
10:03 < mjungwirth> :)
10:04 < rxr> I have the article with me, should scan it :-)
10:04 < mjungwirth> yea send it to me please :) i hang it out on our firm-black-table
10:06 < rxr> ok - should do some work first
10:06 < rxr> will do later
10:07 < mjungwirth> ok thanks
10:07 < rxr> probably have to get a a3 scanner off the shelf
10:08 < mjungwirth> btw. after i add the glibc32 gcc build correctly. Thanks
10:08 < rxr> your're welcoem
10:08 < rxr> if you want a lib(32) less build you can disable the multilib option
10:09 < rxr> it's default is automatically the "sane choice" per architecture
10:09 < rxr> so you can not change it in the UI, and have to overwrite it in a target config
10:09 < rxr> some years ago it worked to build a pure 64bit environment
10:09 < rxr> without the stuff that requires 32bit though
10:09 < rxr> like grub, wine, maybe some dos-emu stuff etc.
10:13 < rxr> (lilo and others work, though)
10:25 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has joined #t2
10:25 < Baldzius> moin
10:25 < rxr> moin Baldzius
10:25 < Baldzius> hey rxr
10:25 < Baldzius> rxr , you've started a fresh build recently right?
10:26 < rxr> yes
10:26 < rxr> but I had to fsck my volume over night
10:26 < rxr> becuase I could not wipe one build with some magic defect file
10:26 < Baldzius> gettext passes on cross stage for you?
10:26 < rxr> I guess you want to ask about attr ?
10:26 < rxr> oh - gettext, yes
10:26 < rxr> but I have some changes to the package lingering
10:26 < Baldzius> ic
10:27 < Baldzius> as it fails with autoconf-2.64
10:27 < rxr> oh
10:27 < Baldzius> you've updated yesterday I think
10:27 < rxr> I patched to not autoregen
10:27 < rxr> anyway, ages ago
10:27 < Baldzius> not sure
10:27 < Baldzius> hm
10:27 < rxr> because it took too much CPU and broke powerpc
10:27 < rxr> I can commit now
10:27 < Baldzius> just was wondering
10:27 < Baldzius> which autoconf version you have on host?
10:27 < Baldzius> cause this might be a reason
10:27 < rxr> oh no idea
10:28 < rxr> # mine -q autoconf
10:28 < rxr> autoconf 2.59 8.0-trunk
10:28 < rxr> but I can commit this change anyway
10:28 < rxr> as it shoudl fix powerpc and lower cpu useage
10:28 < rxr> by not endlessly rebuilding so much auto* junk
10:28 < Baldzius> so it's not the reason
10:28 < Baldzius> I have 2.61
10:28 < Baldzius> hm
10:28 < Baldzius> if you commit
10:28 < Baldzius> I would test it
10:28 < rxr> done
10:29 < rxr> Committed revision 33780.
10:29 < rxr> CIA-69: ?
10:29 < CIA-69> rene * r33780 /trunk/package/base/gettext/ (gettext-no_long_long.patch gettext.conf gettext.desc): * do not auto-rebuild gettext, unbreaks powerpc, saves CPU cycles
10:29 < Baldzius> thanks
10:29 < Baldzius> will check it
10:29 < rxr> thanks
10:31 < rxr> I intend to commit the next glibc major update later, likewis
10:31 < rxr> e
10:33 < Baldzius> 2.10.1?
10:33 < rxr> yep
10:33 < rxr> * updated glibc (2.9 -> 2.10.1)
10:33 < rxr> * updated glibc32 (2.9 -> 2.10.1)
10:57 < CIA-69> rene * r33781 /trunk/package/archiver/cpio/ (cpio.desc no_rmt.patch): * updated cpio (2.9 -> 2.10)
11:02 < Baldzius> thanks rxr, your commit fixed gettext
11:02 < rxr> nice
11:15 < mjungwirth> rxr: would you like to extend you hardware/server section on the t2-project.org?
11:16 < rxr> sure
11:16 < mjungwirth> hmm i have the hp dl360 and dl380 (g4, g5, g6) servers tested
11:18 < rxr> nice
11:19 < rxr> if you want to write the entries :-)
11:19 < mjungwirth> i send you a mail :)
11:20 < mjungwirth> i get a new hp blade g6 system at 01.10.2009. i think i must test it with t2 :)
11:26 -!- jncneo [n=alex@91.195.10.75] has joined #t2
11:27 < jncneo> hi all!
11:27 -!- _| [n=joeDeuce@pdpc/supporter/student/joeDeuce] has quit [Read error: 110 (Connection timed out)]
11:27 < jncneo> i have a problem building uclibc in embedded configuration. could anyone help me?
11:29 -!- _| [n=joeDeuce@pdpc/supporter/student/joeDeuce] has joined #T2
11:33 < rxr> maybe
11:33 < rxr> mjungwirth: thanks
11:34 < CIA-69> rene * r33782 /trunk/package/xorg/xf86-video-siliconmotion/xf86-video-siliconmotion.desc: * updated xf86-video-siliconmotion (1.7.2 -> 1.7.3)
11:34 < CIA-69> rene * r33783 /trunk/package/xorg/xf86-video-trident/xf86-video-trident.desc: * updated xf86-video-trident (1.3.2 -> 1.3.3)
11:39 < jncneo> here is log: http://jncneo.nm.ru/log
11:41 < CIA-69> rene * r33784 /trunk/package/graphic/gocr/gocr.desc: * updated gocr (0.47 -> 0.48)
11:41 < CIA-69> rene * r33785 /trunk/package/gnome2/gstreamer/gstreamer.desc: * updated gstreamer (0.10.23 -> 0.10.24)
11:43 < CIA-69> rene * r33786 /trunk/package/emulators/dosemu/dosemu.desc: * updated dosemu (1.3.4 -> 1.4.0)
11:48 -!- regomodo [n=kvirc@host86-158-97-148.range86-158.btcentralplus.com] has joined #t2
11:53 [Users #t2]
11:53 [@ChanServ ] [ CIA-69] [ mjungwirth] [ oliv` ] [ T-One]
11:53 [ _| ] [ dsoul ] [ mpp ] [ regomodo] [ TobiX]
11:53 [ Baldzius ] [ jncneo] [ mqueiros_ ] [ rxr ]
11:53 [ bitdance-vacatio] [ koan ] [ mtr ] [ Stealth ]
11:53 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
11:53 < CIA-69> rene * r33787 /trunk/package/network/dhcp/dhcp.desc:
11:53 < CIA-69> Mario Jungwirth
11:53 < CIA-69> * updated dhcp (4.1.0p1 -> 4.1.1b1)
11:55 < CIA-69> rene * r33788 /trunk/package/emulators/dosbox/ (dosbox.desc gcc43.patch): * updated dosbox (0.72 -> 0.73)
11:55 < CIA-69> rene * r33789 /trunk/package/audio/gnupod/gnupod.desc: * updated gnupod (0.99.7 -> 0.99.8)
11:55 < CIA-69> rene * r33790 /trunk/package/network/duplicity/duplicity.desc: * updated duplicity (0.6.02 -> 0.6.04)
12:05 < CIA-69> rene * r33791 /trunk/package/base/ (glibc/glibc.desc glibc32/glibc32.desc):
12:05 < CIA-69> * updated glibc (2.9 -> 2.10.1)
12:05 < CIA-69> * updated glibc32 (2.9 -> 2.10.1)
12:08 < CIA-69> rene * r33792 /trunk/package/python/gst-python/gst-python.desc: * updated gst-python (0.10.15 -> 0.10.16)
12:08 < CIA-69> rene * r33794 /trunk/package/x11/dmenu/dmenu.desc: * updated dmenu (3.6 -> 4.0)
12:09 < CIA-69> rene * r33793 /trunk/package/x11/dwm/dwm.desc: * updated dwm (5.6 -> 5.6.1)
12:10 < CIA-69> aldas * r33795 /trunk/package/base/coreutils/ (4 files): * updated coreutils (7.2 -> 7.4)
12:10 < Baldzius> rxr: could you test coreutils
12:11 < rxr> ouh
12:11 < rxr> yes, will
12:11 < Baldzius> I've removed both cross patches as it compiles fine
12:11 < Baldzius> should be fine I guess
12:12 < rxr> oh I hope too
12:12 < rxr> btw. regarding attr/acl
12:12 < rxr> I vaguely remeber that they once broke with some utils update
12:12 < rxr> like coreutils
12:12 < rxr> maybe you get an error that I don't with those too packages because some basic util update ?
12:12 < Baldzius> I will try that
12:12 < rxr> but I'll see with my all new builds here, too
12:12 < Baldzius> host has now 7.2 and it doesn't help
12:13 < Baldzius> I need to play with it a little bit
12:15 < jncneo> have anyone ever built uclibc from trunk in embedded-minimal configuration?
12:15 < rxr> probably, some weeks ago
12:15 < jncneo> was it ok?
12:16 < rxr> sure, we continously adapt stuff to keep building in regards to upstream changse
12:16 < rxr> I wanted to view your log, but your server did not mark it as content encoding text, so my browser wanted to save
12:16 < rxr> and I was too lazy to save and open an ext. editor when I usually view such stuff directly in my browser ...
12:16 < jncneo> oh, i'll check it now
12:17 < rxr> but usually breakage is easy to fix / adapt if one has a little source, edit, compile experience
12:17 < jncneo> i have a little experience in other distributions but i'm using T2 first time
12:18 < rxr> yeah - I meant "general experience" :-)
12:19 < jncneo> http://jncneo.nm.ru/log.txt not it's ok in my firefox
12:20 < rxr> ok - that are "just" shared files, installed by glibc and uclibc
12:20 < rxr> which config did you build?
12:21 < jncneo> embedded minimal
12:21 < rxr> with no option in the Config changed ?
12:23 < jncneo> aslo i've checked optimization to AMD Geode LX and "Print Build-Output to terminal while building"
12:23 < rxr> ok - they are unproblematic, but nothing else touched in the "Config" ?
12:24 < jncneo> nothing
12:24 < rxr> ok will testbuild a little later
12:24 < jncneo> thanks, i'll wait
12:27 < rxr> == 08/05/09 12:35:55 =[1]=> Finished building package coreutils.
12:29 < Baldzius> cool
12:30 < rxr> ^- x86
12:30 < rxr> == 08/05/09 12:38:05 =[3]=> Finished building package coreutils.
12:30 < rxr> ^- x86-64
12:31 < rxr> mostly fresh from scratch builds, with only todays updates intermixed
12:33 < rxr> will also start fresh ppc builds before I go for lunch
13:06 < regomodo> When was t2-7.0 released?
13:32 < regomodo> Hi, i'm doing a crosscompile trunk minimal build and I am stuck at apr-utils. The error is "checking for Berkeley DB 4.4 in /usr/include/db4... directory not found" but I have BerkeleyDB installed?
13:37 < rxr> hm
13:37 < rxr> maybe it needs some adaption to cross compile again
13:40 < rxr> Baldzius: so, fully fresh x86-64 build is in stage 5
13:55 < regomodo> rxr: I got rid of the configure error by creating a symlink between /usr/include/4.6 and /usr/include/db4 but that brings up another error
13:57 < regomodo> i686-t2-linux-gnu-gcc: /home/jon/t2sde/t2-trunk/build/current-8.0-trunk-generic-x86-i686-cross/usr/lib64/libapr-1.so: No such file or directory
13:57 < regomodo> make[1]: *** [libaprutil-1.la] Error 1"
14:02 < regomodo> Has anybody got a x-compile to work ever? I can't get one to work in 7.0 or trunk even with the minimal profile/
14:04 < rxr> regomodo: sure, regularly using it
14:05 < regomodo> Lucky you. I've been thinkng maybe it's the host's toolchain so i've tried Gentoo x86/amd64, Fedora x86, Ubuntu x86 and Debian amd64. I'm about to give up and go back to LFS.
14:08 < rxr> what is your problem?
14:08 < rxr> what do you need aprutil for in a normal build? not really X related ...
14:10 < regomodo> rxr: I've chosen the minimal profile, without X, and it's being pulled in.
14:10 < rxr> aha
14:10 < rxr> so you just want X?
14:10 < rxr> maybe try with the currently cross broken offenders disabled? like this aprutil ?
14:10 < rxr> though soonder or later I'll probably also hit it and fix
14:12 < regomodo> rxr: I don't want X as i've found I can't get that to build (libxcb issues) so I just want something that builds and installs
14:12 < rxr> I forgot, why do you cross build at all?
14:13 < regomodo> I'm on Gentoo amd64 and wan't to build an i686 install
14:13 < rxr> if you build for common and fast silicon you have a magnitutes easier life with a native build
14:13 < rxr> yeah - but still you can native build that
14:13 < rxr> your x86-64 kernel will run the 32 bit binaries
14:13 < regomodo> Really?
14:13 < rxr> (usually, at least, unless gentoo disabled the COMPAT32 thing in the linux kernel by default)
14:13 < rxr> sure
14:14 < regomodo> wow, i'm off to try that
14:14 < rxr> in the "non cross" case T2 only cross builds the bare minimum toolchain and native builds the rest in the chroot sandbox
14:14 < rxr> as you noticed some stupid upstream changes sometimes break (regress) some of the more advanced packages to cross build
14:14 < regomodo> so i'd use x-compile between x86 and ppc thern?
14:14 < rxr> yep
14:14 < rxr> or arm, sparc etc.
14:15 < regomodo> damn
14:15 < rxr> only if your CPU silicon can not run the resulting binaries
14:18 < regomodo> i'm sure a c2d should be able to run i686 binaries. I just checked my .config and "IA32_EMULATION" is enabled
14:18 < rxr> was that the name?
14:18 < rxr> remebered some with COMPAT
14:18 < rxr> whatever
14:19 < rxr> CONFIG_IA32_EMULATION=y
14:19 < rxr> indeed
14:19 < rxr> strange
14:24 < mjungwirth> rxr: i have found a bug (build by me) in the vsftpd.conf. (Build abort on 64Bit system). i send you the correct vsftpd.conf
14:37 < CIA-69> rene * r33796 /trunk/package/network/vsftpd/vsftpd.conf:
14:37 < CIA-69> Mario Jungwirth
14:37 < CIA-69> * fixed vsftpd for lib64 builds
14:37 < mjungwirth> thanks
14:38 < jncneo> rxr: have you checked uclibc? I've made few more tries but with no results :(
14:39 < rxr> I will now
14:39 < jncneo> thanks
14:41 < rxr> jncneo: what exactly did you select?
14:41 < rxr> when I select embedded / uclibc then I get no glibc selected
14:43 < jncneo> i've seleced: embedded-minimal, install, x85, geodelx, grub, cross-build, "Continue on package error after stage 4", Keep src dirs on build errors, (1 ) Maximum number of jobs to be executed in parallel,Use ANSI colors for terminal output and Print Build-Output to terminal while building
14:43 < jncneo> there were no checkbox with glibc or uclibc...
14:44 < rxr> ah! embedded-minimal! one second
14:44 < mjungwirth> rxr: can you add my samba fix_samba_mtab patch to the svn?
14:45 < jncneo> ok
14:46 < mjungwirth> i can't build the new samba version 3.4.0! The patchfile "configure.patch.cross" has the incorrect patch-directory for version 3.4.0
14:46 < mjungwirth> i fixed it and send per mail
14:49 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has quit [Read error: 104 (Connection reset by peer)]
14:50 < rxr> mjungwirth: when?
14:51 < mjungwirth> moment my mail-server is angry :)
14:52 < mjungwirth> ok is out
14:52 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has joined #t2
14:54 < CIA-69> rene * r33797 /trunk/package/network/samba/configure.patch.cross:
14:54 < CIA-69> Mario Jungwirth
14:54 < CIA-69> * updated samba cross compile patch
14:55 < rxr> mjungwirth: you will get a conflict, I added the proper directory name and version for us humans as reference
14:56 < CIA-69> rene * r33798 /trunk/target/embedded-minimal/ (config.in pkgsel):
14:56 < CIA-69> * fixed embedded-minimal by selecting the system libc with the proper
14:56 < CIA-69> config option to avoid implicitly selcting another (like glibc)
14:56 < rxr> jncneo: the last commit was for your config case
14:56 < mjungwirth> but the directory has changed from source to source3
14:56 < rxr> mjungwirth: yes, but you had ./ as prefix
14:56 < rxr> I changed the . to the samba-3.4.x/
14:57 < rxr> to better reflect which version the patch was against
14:57 < jncneo> understood. i'll check right now
14:57 < rxr> so it should all be well, apply and such
14:57 < rxr> the first directory (your . my samba-$ver) will be thrown away by patch -p1 anyway ...
14:57 < rxr> this is just academic information ...
14:57 < mjungwirth> i will remember
14:58 < rxr> (though I also sometimes add ./ patches ... :-)
14:58 < mjungwirth> :)
14:58 < rxr> it's always nicer to have the proper directory prefix
14:58 < mjungwirth> i'm not alone :)
14:58 < rxr> though I'm not linus torvalds and ask for a patch resend with proper signed-off line :-)
14:58 < rxr> I edit in-place and commit :-)
14:59 < rxr> even accept inline, attachment, .zip, .tar, replacement files etc. :-)
15:00 < rxr> jncneo: you need to start a clean build to get rid of the conflicting glibc files
15:00 < rxr> jncneo: my test build is still running though - only verified the resulting .config before the commit
15:01 < jncneo> understood, thanks. i'll also make test and i'll say you if smth wrong
15:33 < rxr> initial builds with the new glibc and trunk:HEAD look fine on ppc, ppc64, likewise
15:33 < koan> maybe I should try T2 again on my G5
15:34 < koan> will you publish a new build soon?
15:57 < rxr> I should :-)
15:58 < rxr> cpu : PPC970MP, altivec supported
15:58 < rxr> clock : 2000.000000MHz
15:58 < rxr> revision : 1.0 (pvr 0044 0100)
15:58 < rxr> model : PowerMac11,2
16:04 < rxr> http://www.t2-project.org/packages/emacs.html
16:31 -!- mpp [n=mpp@2a01:48:251:301:21a:6bff:fe35:820a] has quit ["Verlassend"]
17:21 < jncneo> rxr: i have one more error during configuring dropbear:
17:21 < jncneo> checking for deflate in -lz... no
17:21 < jncneo> configure: error: *** zlib missing - install first or check config.log ***
17:21 < jncneo> Due to previous errors, no 1-dropbear.log file!
17:22 < jncneo> but i have installed zlib on my system...
17:27 -!- |regomodo| [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has joined #t2
17:43 -!- regomodo [n=kvirc@host86-158-97-148.range86-158.btcentralplus.com] has quit [Read error: 110 (Connection timed out)]
17:54 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has quit ["Leaving"]
18:49 < |regomodo|> Hi, I've finally built a minimal install but when I go to build the iso I get this error:
18:49 < |regomodo|> [19:44:36] Creating iso/current_cd1.iso ...
18:49 < |regomodo|> Bad Option '-q' (error -1 BADFLAG).
18:49 < |regomodo|>
18:52 < |regomodo|> Hmm, I went into the Create-ISO script and got rid of the "-q" flag OPT on line 326
19:31 < |regomodo|> I'm guessing "-q" is for quiet output. Just checked the man pages and that doesn't exist. You have to use "-quiet"
19:40 < koan> jncneo: maybe you need zlib-dev?
19:52 -!- PassingStranger [n=cat@p54BF7B9D.dip.t-dialin.net] has joined #t2
19:56 < |regomodo|> I've just tried to build a minimal+x install in trunk and although it says it's completed it has skipped around 90 binary packages. Should I just ignore that or is that pretty bad?
21:02 -!- PassingStranger [n=cat@p54BF7B9D.dip.t-dialin.net] has quit ["Leaving"]
21:13 -!- |regomodo| [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has quit [Remote closed the connection]
21:31 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has joined #t2
21:32 -!- koan [n=koan@unaffiliated/koan] has quit ["Lost terminal"]
21:32 -!- koan [n=koan@unaffiliated/koan] has joined #t2
22:35 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has quit ["Changing server..."]
22:36 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has joined #t2
22:37 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has quit [Client Quit]
22:38 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has joined #t2
22:40 -!- regomodo [n=kvirc@host81-159-18-101.range81-159.btcentralplus.com] has quit [Client Quit]
23:17 -!- regomodo [n=jon@host81-159-18-101.range81-159.btcentralplus.com] has joined #t2
23:36 -!- regomodo [n=jon@host81-159-18-101.range81-159.btcentralplus.com] has quit [Remote closed the connection]
--- Log closed Thu Aug 06 00:00:38 2009