T2 IRC Log: 2007-06-20
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 Jun 20 00:00:22 2007
00:01 -!- kic_ [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
00:05 -!- DeeKey [n=denis@84-50-16-45-dsl.kjj.estpak.ee] has joined #t2
00:08 < DeeKey> hello
00:13 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Read error: 110 (Connection timed out)]
00:20 -!- kic_ [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
00:21 < Ragnarin> re
00:21 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
00:31 -!- jnewell [n=jnewell@pr-66-150-46-254.wgate.com] has joined #t2
00:31 -!- DeeKey [n=denis@84-50-16-45-dsl.kjj.estpak.ee] has quit [Read error: 110 (Connection timed out)]
00:33 -!- jnewell [n=jnewell@pr-66-150-46-254.wgate.com] has quit [Remote closed the connection]
00:49 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
00:49 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
01:00 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
01:00 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
01:10 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
01:10 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
01:20 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
01:20 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
01:30 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
01:31 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
01:31 -!- mtr_ [n=Michael@pD9E13447.dip0.t-ipconnect.de] has joined #t2
01:45 -!- mtr [n=Michael@pD9E1111A.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)]
01:45 -!- mtr_ is now known as mtr
01:50 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
01:51 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
02:01 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
02:01 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
04:57 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Read error: 113 (No route to host)]
06:38 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has joined #t2
06:38 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has quit [Client Quit]
08:16 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has joined #t2
08:16 < tri> moin
08:22 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has quit [Read error: 104 (Connection reset by peer)]
08:22 -!- tr1 [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has joined #t2
08:27 < Fatal> morning
08:27 -!- tr1 [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has quit [Read error: 104 (Connection reset by peer)]
08:37 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has joined #t2
08:57 < rxr> re
08:57 < rxr> moin all
08:58 < Fatal> good morning
08:58 < tri> moin
08:58 < tri> rxr: have a problem to build the iwlwifi modul
08:59 < tri> if i build the kernel the log says something about "could not find mac80211 headers" in /lib/modules/....
09:01 < tri> MAC80211_INC=/lib/modules/2....-dist/source/include/net <---
09:01 < tri> in the kernel build log
09:01 < tri> ragnarin told me yesterday to change change MAC80211_INC to be $kerneldir/include/net
09:02 < tri> but without success same error message again
09:04 -!- tr1 [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has joined #t2
09:04 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has quit [Read error: 104 (Connection reset by peer)]
09:04 < tr1> re
09:04 < tr1> hm damm internet connection....
09:05 < rxr> tr1: hm - do you use the 7.0-rc release or trunk ?
09:05 < tr1> trunk
09:06 < rxr> hm - I thought I fixed that
09:06 < rxr> probably time for a test-build on my side
09:06 < tr1> in the iwlwifi postlinux.conf there is MAC80211_INC=$moduledir/source/include/net
09:07 < rxr> if you want to take a look yourself, as Ragnarin said it is probably just a matter of setting some make variable to let it pick up on cross build
09:07 < rxr> btw. on which stage does it fail? 1 (cross build) or 5 (native build)?
09:07 < tr1> native build
09:07 < tr1> i do emerg-pkg -force linux26
09:07 < rxr> ah - ok, stage 9 is native build as well
09:08 < tr1> hm where are the mac80211 kernel headers
09:08 < rxr> (emerges run in stage 9, which is last thing of native builds)
09:08 < rxr> they are in the kernel and the third party module usually wants to find them via the temp. kernel source or module dir
09:08 < Ragnarin> btw I tried it again, builds fine for me
09:09 < Ragnarin> does the mac80211 patch apply cleanly earlier?
09:10 < tr1> ah hi ragnarin ;-)
09:10 < tr1> have go to sleep yesterday i hope your lunch was good
09:11 < Ragnarin> yea it was
09:11 < Ragnarin> you do have mac80211 in your pkgsel?
09:11 < tr1> ah no !
09:11 < Ragnarin> then that may very well be the reason :D
09:11 < rxr> lol
09:12 < rxr> heh - I build an update here now anyway, good for security in any case .-)
09:12 < tr1> hm but at least the error was found...
09:12 < tr1> ah ok why using SETI@home if you can build the kernel again...
09:13 < rxr> in T2 you can set a debug flag to just build the modules
09:13 < rxr> (not the kernel)
09:13 < rxr> (and use ccache anyway :-)
09:13 < tr1> how to set such a debug flag ?
09:13 < rxr> SDEDEBUG_LINUX_SUBMODULES_ONLY=1 ./scripts/Emerge ... linux26
09:13 < tr1> is it enough to edit the pkgsel with vim or do i have to run any "newconfig" command ?
09:14 < rxr> he target/.../pkgsel ?
09:14 < rxr> ^-the
09:14 < rxr> after you edit the pkgsel you need to Config -oldconfig
09:14 < rxr> but hey, wait
09:15 < rxr> the Bulid-Target already runs the Config -oldconfig automagically on-demand
09:15 < rxr> I think we should fixed Emerge-Pkg to do the same
09:15 < mtr> moin
09:15 < tr1> hi mtr
09:16 < rxr> moin moin mtr
09:16 < CIA-9> susan * r24647 /trunk/package/graphic/dcraw/dcraw.desc: * updated dcraw (8.74 -> 8.75)
09:16 < CIA-9> susan * r24648 /trunk/package/network/ddclient/ddclient.desc: * updated ddclient (3.7.1 -> 3.7.2)
09:19 < CIA-9> mtr * r24649 /trunk/package/www/firefox/firefox.conf:
09:19 < CIA-9> * fixed firefox to actually install the nss header files
09:19 < CIA-9> (folder name changed)
09:21 < rxr> oh - that might be the reason why nautilus and openoffice-org did not found those when I wanted to build them against firefox. Good catch!
09:25 < mtr> evolution depends on the mozilla nss libs, too
09:27 < CIA-9> rene * r24650 /trunk/package/mail/msmtp/msmtp.desc: * updated msmtp (1.4.11 -> 1.4.12)
09:27 < CIA-9> rene * r24651 /trunk/package/network/ndiswrapper/ndiswrapper.desc: * updated ndiswrapper (1.46 -> 1.47)
09:28 < CIA-9> rene * r24652 /trunk/package/graphic/cowsay/ (cowsay.conf cowsay.desc): * de-custmained cowsay
09:31 < CIA-9> rene * r24653 /trunk/package/ (8 files in 4 dirs): * fixed vamps, p0f, wipe, (the classic) hotplug, to cross build
09:55 < tr1> hm i build the kernel now
09:55 < tr1> but modprobe iwl3945 said module iwl3945 not found
09:56 < tr1> build was fine this time
09:57 < tr1> i also have the firmware in /lib/firmware
09:57 < tr1> and have build it into the initrd
09:59 < rxr> is the module in /lib/modules/... ?
10:00 < tr1> ... ?
10:00 < tr1> hm in KMISC=$moduledir/kernel/drivers/net/wireles
10:00 < tr1> shouldn't it be wireless ?
10:01 < tr1> http://svn.exactcode.de/t2/trunk/package/network/iwlwifi/postlinux.conf
10:01 < tr1> is there a typo ?
10:01 < rxr> probably, but it shoudl work nontheless
10:02 < rxr> the tools too not care about a name at all
10:03 < CIA-9> rene * r24654 /trunk/package/network/iwlwifi/postlinux.conf: * fixed wireles -> wirele_ss_ typo in iwlwifi/postlinux.conf
10:03 < tr1> yes did that
10:05 < mtr> rxr: firefox uses a GCC_WRAPPER_INSERT to globally inject X11 libs, I want to replace this by a simple one-liner patch:
10:05 < mtr> + ifdef MOZ_ENABLE_GTK2
10:05 < mtr> + EXTRA_DSO_LDOPTS += $(MOZ_GTK2_LIBS) \
10:05 < mtr> ++ $(XLDFLAGS) -lX11 -lXrender \
10:05 < mtr> + $(NULL)
10:05 < mtr> + endif
10:06 < rxr> it does build ?
10:07 < mtr> yes
10:07 < rxr> well - sure
10:07 < mtr> for x86
10:07 < rxr> but I think we have this in most newer mozilla a-like packages
10:07 < rxr> thunderbird and co as well
10:07 < mtr> but XLDFLAGS should cover x86_64 as well
10:07 < rxr> did you fix those as well ?
10:07 < rxr> btw. I wonder why it builds on other distributions without the patch
10:07 < mtr> untested
10:08 < rxr> if you find the time to patch the other as well it woudl be appreciated
10:08 < rxr> I can test build them on the ref tester
10:08 < mtr> ok
10:11 < CIA-9> rene * r24655 /trunk/package/network/nginx/nginx.desc: * updated nginx (0.5.24 -> 0.5.25)
10:11 < CIA-9> rene * r24656 /trunk/package/develop/hte/hte.desc: * updated hte (2.0.6 -> 2.0.7)
10:11 < CIA-9> rene * r24657 /trunk/package/base/pcre/pcre.desc: * updated pcre (7.1 -> 7.2)
10:11 < CIA-9> rene * r24658 /trunk/package/audio/libgpod/libgpod.desc: * updated libgpod (0.4.2 -> 0.5.0)
10:12 < CIA-9> rene * r24659 /trunk/package/filesystem/udev/udev.desc: * updated udev (111 -> 112)
10:13 < CIA-9> rene * r24660 /trunk/package/base/device-mapper/device-mapper.desc: * updated device-mapper (1.02.19 -> 1.02.20)
10:13 < CIA-9> rene * r24661 /trunk/package/kde/kalarm/kalarm.desc: * updated kalarm (1.4.11 -> 1.4.13)
10:13 < CIA-9> rene * r24662 /trunk/package/kde/digikam/digikam.desc: * updated digikam (0.9.1 -> 0.9.2)
10:28 < mtr> weird ... or normal? ... do we generally have a stripped /lib/libpthread.so?
10:29 < rxr> if you do not built with debugging enabled it probably will be stripped, yes
10:29 < mtr> seems that libthread_db cannot resolve some symbols for debugging
10:30 < mtr> how can I disable stripping, without SDECFG_DEBUG?
10:33 < rxr> yes
10:36 < sepp> moin :)
10:36 < tr1> hi sepp
10:36 < sepp> hi tr1
10:37 < rxr> moin sepp
10:37 < mtr> hi sepp
10:37 < sepp> hi rxr, mtr :)
10:37 [Users #t2]
10:37 [@ChanServ] [ Dallur] [ LMJ ] [ rxr ] [ TobiX ]
10:37 [ cbrake ] [ emte ] [ mtr ] [ sepp ] [ tr1 ]
10:37 [ CIA-9 ] [ Fatal ] [ Ragnarin] [ Stelz] [ valentin]
10:37 -!- Irssi: #t2: Total of 15 nicks [1 ops, 0 halfops, 0 voices, 14 normal]
11:00 < tr1> ah now i have the ipw3945 module in /lib/ ...
11:01 < tr1> just one reboot away from modprobe it
11:01 -!- idealm [n=ideal@c58-107-18-66.belrs2.nsw.optusnet.com.au] has joined #t2
11:02 < sepp> why do you need to reboot to do that?
11:03 < tr1> because it is build for the new kernel
11:03 < tr1> hm after i do emerge-pkg linux26
11:03 < tr1> i also have to change grub settings ?
11:03 < tr1> or is this done automaticly ?
11:04 < sepp> if it is a new version, you can use stone
11:04 < CIA-9> rene * r24663 /trunk/package/ (17 files in 8 dirs):
11:04 < CIA-9> * fixed stunnel, metamail, transconnect, nudelstats, xskat, wakkabox,
11:04 < CIA-9> prngd, and mplayer-fonts to cross build
11:05 < tr1> can i load a new linux kernel without rebooting ?
11:05 < sepp> no
11:05 < rxr> yes with kexec
11:07 < tr1> hm some time ago i read something about that option - i will have a look at kexec just for intrest
11:07 < tr1> till now i edited the grub menu with vim but stone is nice to edit it: it founds all kernels...
11:08 < tr1> tada: modprobe iwl3945
11:08 < tr1> and now i have a new device wmaster0 and wlan0
11:08 < rxr> I think kexec is not yet enabled in T2 default kernels
11:08 < rxr> experiemntal anyway
11:09 < tr1> but also a warning: driver for device wlan0 was been compiled with version 22 of wireless extension
11:09 < tr1> while this programm supports up to version 20 . some things may be broken
11:09 < rxr> you can ingore that ...
11:09 < tr1> ah ok
11:09 < tr1> that's from the iwlwifi driver ?
11:09 < rxr> I think we should patch this nonsense message away
11:09 < rxr> no, I think the wireless tools print that out
11:10 < rxr> I bet SuSE and RedHat patch that away ...
11:10 < rxr> I never found anything broken either ...
11:11 -!- sepp_ [n=sepp@Z574e.z.pppool.de] has joined #t2
11:14 < tr1> yes sure SuSE (or is it typed S.U.S.E or suse or SusE) and RedHat patch a lot of the packages...
11:15 -!- Dallur [n=earl@fire-out.ru.is] has quit ["Leaving."]
11:16 -!- Dallur [n=earl@fire-out.ru.is] has joined #t2
11:19 < tr1> from the rt wiki: "Controlling a laser with Linux is crazy, but everyone in this room is crazy in his own way. So if you want to use Linux to control an industrial welding laser, I have no problem with your using PREEMPT_RT." -- Linus Torvalds
11:21 < rxr> heh
11:21 < tr1> kexec is not part of t2 yet ?
11:22 < rxr> it's part of recent kernels
11:22 < rxr> you just have to enable it in the .config
11:22 < rxr> t2 does not do that, yes, as far as I know
11:22 < rxr> despite if the default state for that CONFIG_KEXEC stuff changed and it is on by default :_)
11:23 < tr1> aren't the kexec-tools needed ?
11:23 < Dallur> tr1: I'm controlling a 2.5m x 6.5m Plasma cutting robot with linux myself, laser is actually simpler and easier to do than plasma
11:24 < rxr> tr1: probably, maybe submit a patch adding those
11:24 < Dallur> tr1: http://linuxcnc.org/
11:26 < CIA-9> mtr * r24664 /trunk/package/www/firefox/ (firefox.conf gtk2-x11-libs.patch): * remove firefox build workaround, use a makefile patch instead to add X11 libs
11:26 < tr1> rxr: yip, i will do that
11:26 < Fatal> heh, shame I don't get enough time to port our system to t2, or I would have kexec stuff added already :)
11:26 < Fatal> goodie, one less thing I'll have to worry about \o/
11:26 < CIA-9> mtr * r24665 /trunk/package/www/seamonkey/ (gtk2-x11-libs.patch seamonkey.conf): * remove seamonkey build workaround, use a makefile patch instead to add X11 libs
11:28 -!- sepp [n=sepp@Z4174.z.pppool.de] has quit [Read error: 110 (Connection timed out)]
11:29 -!- sepp_ is now known as sepp
11:32 < mtr> we do have /usr/bin/perl setuid, is ths intentional?
11:36 < tr1> hm: There may be times you would want to run perl scripts as non-root users. This offers more security to your system and this guide will show you how to enable perl to use setuid.
11:37 < tr1> from *bsd wiki
11:37 < tr1> maybe it is a good idea to have "ENABLE_SUIDPERL=true"
11:44 < rxr> mtr: this is historic
11:44 < rxr> mtr: I noticed this caues some fooomatic-rip stufff to error out just because of the setuid bit
11:47 < mtr> and some aclocal regressions:
11:47 < mtr> Insecure $ENV{PATH} while running setuid at /usr/lib/perl5/5.8.8/i686-t2-linux-gnu/IO/File.pm line 192.
11:47 < mtr> make: *** [aclocal.m4] Fehler 1
11:52 < tr1> what is the way if there are more the one author use two or three [A]
11:52 < rxr> yeah - that is also like what some foomatic stuff spits into the log for certain printer rasterizers ..
11:52 < tr1> or just put them all in one [A] ?
11:53 < rxr> tr1: yes, multiple [A]'s is just fine
11:53 < rxr> many packages ahve that already
12:48 -!- idealm [n=ideal@c58-107-18-66.belrs2.nsw.optusnet.com.au] has quit [Read error: 110 (Connection timed out)]
12:49 -!- idealm [n=ideal@58.105.232.64] has joined #t2
12:54 -!- tr1 [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has quit [Read error: 104 (Connection reset by peer)]
13:13 -!- idealm [n=ideal@58.105.232.64] has quit [Read error: 110 (Connection timed out)]
13:19 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has joined #t2
13:23 < tri> re
13:23 < rxr> re
13:32 < sepp> er
13:32 < tri> http://ubuntuforums.org/showthread.php?t=475709
13:35 < tri> do i have to give the full download url in the [D]
13:35 < tri> or can i X filename http://aewfawefawef/dasdf/ ?
13:36 < sepp> the later
13:46 < tri> what is $confdir ?
13:55 < tri> how to build a package which don't have the ./configure make make install thing
13:55 < tri> but ah tar.gz
13:55 < tri> where a install.sh script is int ?
13:55 < tri> s/int/in
13:56 < tri> and that should be executed ?
13:56 < tri> i tried:
13:56 < tri> makeopt=
13:56 < tri> makeinstopt=
13:56 < tri> hook_add postmake 5 'install.sh /'
13:56 < tri> but that didn't work ?
13:58 -!- idealm [n=ideal@58.105.232.64] has joined #t2
14:08 < tri> what is the var for the extracted package dir ?
14:09 < tri> like builddir for the builddir is there also a var for "builddir/extrackt_package/ ?
14:18 < tri> ah
14:18 < tri> no i have used in postmake 5 '$builddir/$pkg/install.sh /'
14:18 < tri> no/now
14:18 < tri> is this the right way ?
14:19 < rxr> re
14:20 < rxr> well - this will break for cross builds
14:20 < rxr> you need at least $root/
14:20 < rxr> but what kind of package is that? just installing files? or compiling something ?=
14:21 < tri> just installing files
14:22 < tri> postmake 5 '$builddir/$pkg/install.sh /' break cross builds ?
14:22 < tri> how to do it "better" ?
14:24 < sepp> use $root/ as target
14:24 < tri> ah ok not "/" ?
14:25 < sepp> no ;)
14:25 < tri> ?
14:25 < sepp> not just "/"
14:25 < tri> ah yes not i understand it
14:25 < rxr> just using / will place the new files int the system you build on, not in the new sandbox
14:26 < tri> cross build isn't installing any files here
14:26 < rxr> though T2 "should" scream about this before you mess your system up
14:26 < tri> but for the tar.gz package which is created
14:26 < tri> for the crossbuild
14:26 < tri> right ?
14:26 < tri> ah ok
14:26 < tri> sorry i'm not that family with t2 but it is getting a bit better...
14:27 < tri> hm i allready wondered why build-pkg installed the files...
14:28 < tri> hm build package was working
14:28 < tri> but emerg-pkg said Error: No such a package
14:28 < tri> the package is just in my target might that be the problem ?
14:28 < sepp> it was not yet in config/.../packages
14:29 < tri> how to get it into config/.../packages ?
14:30 < sepp> rerun Config with -oldconfig
14:30 < rxr> rerun Config with -oldconfig
14:30 < rxr> now that was synchron typing
14:30 < sepp> hehe :)
14:30 < tri> :-)
14:30 < tri> thx both ;-)
14:30 < tri> i try both commands...
14:31 < tri> hm it said Configuration finished but emerg-pkg still not working
14:32 < sepp> it is not enabled? X ... ?
14:33 < tri> hm not it is not in there
14:33 < tri> in the package file
14:41 < tri> ?
14:43 < rxr> in the [P]riority tag of the package file
14:43 < rxr> maybe upload the .desc and we take a look
14:43 < tri> thx :-)
14:43 < tri> P X-1---5---9 100.000
14:44 < rxr> missing space
14:44 < rxr> [P] X -1---5---9 100.000
14:45 < rxr> and probably the patckage should not be built in stage 1 forced, as stage is is "cross build what is needed for a new freestanding system
14:45 < rxr> "
14:49 < tri> http://imi.st/mkinitcpio.desc and http://imi.st/mkinitcpio.conf
14:53 < rxr> oh
14:53 < tri> ?
14:54 < rxr> (just wander if we need such a package)
14:54 < rxr> was hopking for kexec tools or so
14:54 < rxr> what is this script doing? calling cpio ?
14:54 < tri> jop it is building a initrd
14:55 < rxr> the scripts look ok so far
14:55 < rxr> where did you place those ?
14:55 < tri> kexec-tools will follow at the evening
14:55 < tri> place those ?
14:55 < rxr> where in package/ ?
14:55 < tri> ah ok
14:56 < tri> in target/.../package/mkinitcpio
14:56 < rxr> oh
14:56 < tri> hm oh's are good ?
14:56 < rxr> I think this will cause all sorts of horror happening
14:56 < rxr> you can only extend packages that exist
14:56 < rxr> place it in package/contrib/mkinitcpio
14:56 < tri> ah ok
14:57 < tri> what is contrib for (the dir) ?
14:57 < rxr> probably we should harden some scripts here and there for such cases
14:57 < rxr> package/contrib/
14:57 < rxr> for stuff that is not yet decided where to place
14:57 < tri> ah ok
15:00 < tri> so it is not the right way to "test" package in my own target
15:00 < tri> target's are more for special config options package selections etc ?
15:01 < tri> hm still not
15:01 < tri> i moved it but emerge-pkg still don' know the package
15:01 < tri> i also did -oldconfig
15:01 < tri> build-pkg still works
15:01 -!- idealm [n=ideal@58.105.232.64] has quit [Read error: 110 (Connection timed out)]
15:03 < tri> hm but now it is int packages file
15:04 < rxr> k# ./scripts/Emerge-Pkg mkinitcpio
15:04 < rxr> checking mkinitcpio ... Not installed. Added.
15:04 < rxr> Packages scheduled to build: mkinitcpio
15:04 < rxr> works for me
15:04 < rxr> so far
15:05 < tri> ah i have to use -force in emerg-pkg
15:05 < tri> sorry for the noise works here to...
15:06 < tri> :-(
15:06 < rxr> hm ?
15:06 < rxr> maybe not selectd in your target or so ...
15:14 < rxr> tri: should I commit your package, now ?
15:17 < CIA-9> rene * r24666 /trunk/package/contrib/mkinitcpio/ (. mkinitcpio.cache mkinitcpio.conf mkinitcpio.desc):
15:17 < CIA-9> Tobias Richter:
15:17 < CIA-9> * added mkinitcpio (0.5.9) - mkinitcpio
15:27 < tri> wups that was fast ;-)
15:30 < CIA-9> rene * r24667 /trunk/package/contrib/mkinitcpio/mkinitcpio.desc:
15:30 < CIA-9> * imroved mkinitcpio to have a trailing dot in the Text tag, a
15:30 < CIA-9> checksun, and not build it in stage 1 by default
15:35 < tri> no need to add this packages but thx that's nice ;-)
15:35 < tri> i only want to make a few tests with mkinitcpio
15:36 < rxr> we can still have it as an option and/or remove it later on anyway :_)
15:36 < rxr> "release early release often"
15:38 < tri> yes, nice :-)
15:42 < tri> and i like this option even for "burning apps" there are a few different packages
15:42 < rxr> packages missing ?
15:43 < tri> no i mean there are min. 5 different burning apps why not use 2 different initrd shell scripts ?
15:43 < tri> s/use/provide
15:45 < rxr> ah - sure
15:52 < CIA-9> rene * r24668 /trunk/package/base/gpm/ (gpm.conf gpm.desc):
15:52 < CIA-9> * fixed gpm to cross build in way it is also functional thereafter,
15:52 < CIA-9> and not leak stray files in /etc/$root :-)
16:16 -!- kic- [n=kicstah@adsl-074-238-088-206.sip.bct.bellsouth.net] has joined #t2
16:17 < kic-> hi
16:17 < tri> hi kic-
16:17 < rxr> hi kic-
16:17 < kic-> hi there
16:18 < kic-> rene
16:18 < kic-> we are looking at possibly the week of July 2nd
16:18 < kic-> can i /msg you?
16:19 < rxr> kic-: yes, just msg me always :-)
16:19 < kic-> k
16:20 < rxr> sepp: do you have the latest viewcvs of e17 handy ?
16:24 < sepp> i think there is none since they left sf.net
16:27 < rxr> :-(
16:30 < sepp> http://web.enlightenment.org/viewvc/e17/ ahh
16:30 < sepp> there it is
16:31 < rxr> oh - thank you so much
16:31 < tri> http://e.kevb.net/cgi-bin/viewvc.cgi/
16:31 < tri> would also be a option ;-)
16:32 < sepp> ohh
16:34 < sepp> which one was the outdated one?
16:34 < rxr> thanks - found what I was looking for
16:34 < sepp> evas ?
16:35 < rxr> no the evas demos, apparently outsourced into expedite, now
16:35 < rxr> ok - gotta go - back in 1h
16:36 < tri> cu rxr
16:52 < tri> bye
16:52 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has left #t2 []
16:54 -!- keztoo [n=root@adsl-074-238-088-206.sip.bct.bellsouth.net] has joined #t2
17:37 < rxr> re
17:38 < rxr> welcome keztoo
17:38 [Users #t2]
17:38 [@ChanServ] [ Dallur] [ keztoo] [ mtr ] [ sepp ] [ valentin]
17:38 [ cbrake ] [ emte ] [ kic- ] [ Ragnarin] [ Stelz]
17:38 [ CIA-9 ] [ Fatal ] [ LMJ ] [ rxr ] [ TobiX]
17:38 -!- Irssi: #t2: Total of 16 nicks [1 ops, 0 halfops, 0 voices, 15 normal]
17:39 -!- sepp_ [n=sepp@Z46ef.z.pppool.de] has joined #t2
17:43 < keztoo> hi
17:43 < keztoo> thank you
17:44 < keztoo> rxr i have a question when you have a minute
17:46 < rxr> sure
17:46 < keztoo> from a high level
17:47 < keztoo> when doing a cross build the system sometimes tries to chroot and run a target tool
17:47 < keztoo> but if the host machine doesnt support the target instruction set it will always fail
17:47 < CIA-9> rene * r24669 /trunk/package/e17/expedite/ (. expedite.cache expedite.desc): * added expedite (2007-06-10) - A test and benchmark for the Evas canvas library
17:48 < keztoo> kind of like 1.1.1 x11 where it tries to run makekeys but it cant run 64 bit on a 32 bit host
17:48 < rxr> keztoo: in T2 on a cross build nothing does like chroot
17:48 < rxr> keztoo: some packages (that we did not already fix) build helper tools with the cross compiler and try to run them directly
17:48 < rxr> keztoo: just like ./gen-table ... foo bar
17:48 < keztoo> yes, that is what i have seen
17:48 < rxr> those need to be fixed to build the helper with the host compiler, not the cross one
17:49 < rxr> we'll successive correct such instances
17:49 < rxr> we could also integrate a "target system emulator" to run those helper apps, and could build more automatically
17:49 < keztoo> so for example with x11 makekeys the make has to be changed to run the 32 (or host) version , right ?
17:49 < rxr> yes,
17:50 < rxr> but I'm under the impression we can build X.org stuff cross already
17:50 < keztoo> ok, just wanted to make sure what i was seeing was what was happening
17:50 < rxr> do you use trunk ?
17:50 < rxr> trunk:HEAD
17:50 < keztoo> i somehow got a 1.1.1 pkg but i noticed the latest is 1.1.2
17:50 < keztoo> no, i used the standard mirrors
17:50 < keztoo> i am trying to get the head w svs now
17:51 < keztoo> i also tried to change the package to grab 1.1.2 but of course since i didnt fix all the patches i got invalid chunks
17:53 < keztoo> i will grab the latest head now and try again, thank you for your input
17:53 < rxr> for really complex stuff we later could also support running qemu on those cross built binaries, but we want to cross build as much as possible first, as that is the cleanest, most portable and also fasted solution
17:54 < rxr> keztoo: your're welcome
17:57 -!- sepp [n=sepp@Z574e.z.pppool.de] has quit [Read error: 110 (Connection timed out)]
18:00 -!- kic- [n=kicstah@adsl-074-238-088-206.sip.bct.bellsouth.net] has quit [Read error: 104 (Connection reset by peer)]
18:00 -!- keztoo [n=root@adsl-074-238-088-206.sip.bct.bellsouth.net] has quit [Read error: 104 (Connection reset by peer)]
18:02 -!- keztoo [n=root@adsl-074-238-088-206.sip.bct.bellsouth.net] has joined #t2
18:05 < CIA-9> rene * r24670 /trunk/package/develop/libatomic_ops/ (. libatomic_ops.cache libatomic_ops.desc): * added libatomic_ops (1.1) - The atomic_ops project
18:06 -!- kic- [n=kicstah@adsl-074-238-088-206.sip.bct.bellsouth.net] has joined #t2
18:10 < keztoo> to get the trunk:head is this the correct svn cmd ...
18:11 < keztoo> svn co http://svn.exactcode.de/t2/trunk t2-head ??
18:13 < rxr> yes
18:13 < rxr> and then run svn up inside the directory to update from the version control server
18:15 < keztoo> i just executed svn co http://svn.exactcode.de/tr/trunk t2-trunk
18:15 < keztoo> this should also be ok , right ??
18:15 < keztoo> as long as i do the update after it downloads?
18:16 < rxr> t2/trunk
18:16 < rxr> when it downloaded it is up-to-date
18:16 < keztoo> ok, so i did the right one
18:16 < keztoo> ty
18:17 < rxr> the "svn up" is just for like in 8 hours or tomorrow when you want to receive the delta of "what changed between now and the time you run it"
18:17 < keztoo> got it, ty
18:38 < kic-> brb, switching network
18:38 -!- kic- [n=kicstah@adsl-074-238-088-206.sip.bct.bellsouth.net] has quit []
18:45 -!- PAVLINUX [n=pavel@ppp85-140-133-188.pppoe.mtu-net.ru] has joined #t2
18:46 < PAVLINUX> Hello World
18:47 < rxr> hello PAV
18:48 < keztoo> hi
18:48 [Users #t2]
18:48 [@ChanServ] [ Dallur] [ keztoo] [ PAVLINUX] [ sepp_] [ valentin]
18:48 [ cbrake ] [ emte ] [ LMJ ] [ Ragnarin] [ Stelz]
18:48 [ CIA-9 ] [ Fatal ] [ mtr ] [ rxr ] [ TobiX]
18:48 -!- Irssi: #t2: Total of 16 nicks [1 ops, 0 halfops, 0 voices, 15 normal]
18:48 < mtr> hi PAVLINUX
18:49 < rxr> hi mtr
18:50 < PAVLINUX> What do you think about this - http://www.stardust.webpages.pl/files/handbook/handbook-en-0.3-rc1.pdf
18:51 < rxr> oeh - puh
18:51 < rxr> that is a long read - what particualry do you have in mind ?
19:03 < CIA-9> rene * r24671 /trunk/package/ (4 files in 3 dirs):
19:03 < CIA-9> * fixed libsndfile to cross build (patched configure.ac fragment did
19:03 < CIA-9> not expand correctly)
19:03 < CIA-9> * marked liboil, libsamplerate to cross build
19:04 -!- sparc-kly_ [n=mubex@adsl-64-237-248-104.prtc.net] has joined #t2
19:04 < CIA-9> rene * r24672 /trunk/package/audio/pulseaudio/ (. pulseaudio.cache pulseaudio.desc): * added pulseaudio (0.9.6) - Yet another sound server for Unix
19:05 [Users #t2]
19:05 [@ChanServ] [ Dallur] [ keztoo] [ PAVLINUX] [ sepp_ ] [ TobiX ]
19:05 [ cbrake ] [ emte ] [ LMJ ] [ Ragnarin] [ sparc-kly_] [ valentin]
19:05 [ CIA-9 ] [ Fatal ] [ mtr ] [ rxr ] [ Stelz ]
19:05 -!- Irssi: #t2: Total of 17 nicks [1 ops, 0 halfops, 0 voices, 16 normal]
19:06 < keztoo> thats a big document PAV, i think i'll curl up in front of the fireplace with a bottle of wine and a loved one to read it
19:06 < keztoo> :-)
19:24 < PAVLINUX> 66 pages! It's BIG DOCUMENT ;-\
19:24 -!- PAVLINUX [n=pavel@ppp85-140-133-188.pppoe.mtu-net.ru] has quit [Remote closed the connection]
19:30 < CIA-9> rene * r24673 /trunk/package/ (4 files in 3 dirs):
19:30 < CIA-9> * fixed libfame to cross build for $arch != x86
19:30 < CIA-9> * marked libiriverdb and ruby to cross compile
19:31 < CIA-9> rene * r24674 /trunk/package/gnome2/evince/evince.desc: * updated evince (0.9.0 -> 0.9.1)
19:31 < CIA-9> rene * r24675 /trunk/package/doc/howtos/howtos.desc: * updated howtos (20070609 -> 20070616)
19:44 < rxr> ok - cu all then
19:44 < keztoo> by rxr and ty
19:48 -!- keztoo [n=root@adsl-074-238-088-206.sip.bct.bellsouth.net] has quit ["using sirc version 2.211+KSIRC/1.3.12"]
19:48 -!- PAVLINUX [n=pavel@ppp85-140-133-188.pppoe.mtu-net.ru] has joined #t2
19:58 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has joined #t2
19:59 < misl> hi folks
20:08 < mtr> hi misl
20:31 -!- kic- [i=kicstah@adsl-074-238-088-201.sip.bct.bellsouth.net] has joined #t2
20:31 < kic-> any known build problems with perl & gdbm?
20:32 < kic-> tried to build minimal desktop with xorg target, and errors out when building perl with ld: cannot find -lgdbm
20:32 < sepp_> gdbm not enabled?
20:33 < kic-> how do you enable/disable it
20:34 < sepp_> grep gdbm config/.../packages
20:34 < sepp_> when the line does not start iwth an X is it not enabled
20:36 < kic-> hrm
20:36 < kic-> it started with O
20:36 < kic-> so i can just edit that file, and change it to an X right?
20:36 < kic-> and re-run the Build-Target
20:36 < sepp_> then it will never find gdbm
20:36 < kic-> ?
20:36 < sepp_> no
20:36 < kic-> hrm
20:36 < sepp_> run Config
20:36 < kic-> ok
20:36 < sepp_> expert options
20:36 < sepp_> custom package selection
20:36 < sepp_> X gdbm
20:37 < kic-> hrm
20:37 < kic-> ok
20:37 < sepp_> that happens with minimal-xorg ?
20:38 < kic-> yep
20:38 < kic-> minimul xorg, with x86-64, core2 optimize, generic system
20:38 < sepp_> ok, will try later and add it to the real pkgsel :)
20:39 < kic-> cool
20:40 < kic-> changed config, trying rebuild now
20:41 < kic-> hrm
20:41 < kic-> same error
20:41 < kic-> any caches i have to clear before re-building?
20:41 < sepp_> no
20:41 < sepp_> has it build gdbm 1st?
20:41 < kic-> cause Build-Target didn't actually start downloading gdbm and building it before perl
20:41 < kic-> it just started right at perl again
20:42 < sepp_> 1-perl ?
20:42 < kic-> yeah
20:42 < sepp_> well, that should build without gdb
20:42 < sepp_> i think
20:43 < sepp_> gdbm
20:43 < kic-> it gets all the way to the bottom of perl compile, and tries to finally link a binary with gcc -o miniperl [more junk] -lgdbm
20:43 < kic-> hrm
20:43 < kic-> anyway i can tell the perl build not to use -lgdbm
20:43 < kic-> ?
20:45 < sepp_> you could remove -lgdbm from the gcc wrapper
20:46 < PAVLINUX> tar -xvjf perl.tar.bz && cd perl && grep -R lgdbm ./*
20:46 < kic-> where can i edit the gcc wrapper?
20:46 < PAVLINUX> vi configure; /gdbm dd
20:48 < sepp_> var_append GCC_WRAPPER_REMOVE ' ' '-lgdbm'
20:49 < sepp_> in package/perl/perl/perl.conf
20:49 < kic-> hrm
20:49 < kic-> k
20:49 < sepp_> after # build a native miniperl
20:51 < kic-> k, re-running build
20:53 < kic-> woot, that worked
20:53 < kic-> on to autoconf now
20:53 < kic-> yay
20:53 < sepp_> :)
20:53 < kic-> i should bring my questions in here more often
20:53 < kic-> heh
20:54 < kic-> can't be to many errors with this one
20:54 < kic-> its just a default build for the most part
20:55 < kic-> kernel patches and compile is one of the last builds right?
20:56 < kic-> its downloading 2.6.21.tar.bz2 at the moment
20:56 < sepp_> no, it builds in several stages
20:56 < sepp_> and linux* is quite early
20:57 < kic-> i saw that it already built linux 2*17 or something
20:57 < kic-> previously
20:57 < kic-> how far into it, is udev?
20:57 < sepp_> very early
20:57 < kic-> i haven't seen what stage its in, cause i walked away earlier
20:57 < kic-> aww
20:57 < kic-> ok
20:57 < kic-> got a while to go then i guess
20:57 < kic-> :D
20:57 < sepp_> that will be a long walk
20:57 < sepp_> :)
20:58 < kic-> hehe
20:58 < kic-> the machine i'm building on has some horse power though
20:58 < kic-> dual xeon 3ghz, with core2 duo, 2GB RAM, 1GB SWAP
21:13 < sepp_> still takes some h
21:16 < kic-> does it build the modules multiple times also?
21:16 < sepp_> modules?
21:16 < kic-> kernel modules
21:17 < sepp_> yes, it builds an complete toolchain 1st, glibc binutils gcc, all that stuff
21:17 < kic-> hrm, neat
21:18 < sepp_> and later it builds the real stuff with the toolchain in the sandbox ...
21:20 < kic-> ahh
21:20 < kic-> thats where some previous builds i was trying were erroring out then
21:21 < kic-> was getting chroot: error bash exec format error
21:21 < kic-> or soemthing like that
21:21 < kic-> yesterday
21:21 < sepp_> that would mean your cpu can not run the binaries you created - for example because it is for an different cpu
21:22 < kic-> erm
21:22 < kic-> that was a for a cross compile though
21:22 < sepp_> like building ppc on some x86*
21:22 < kic-> but you should be able to do that with cross compiling
21:22 < kic-> right?
21:23 < sepp_> no, there is fast growing support for cross compilation but if some package wants to run some thing - it needs a patch
21:24 < sepp_> but you can already crosscompile quite a lot
21:24 < kic-> all of our cross compiling would be for a small embedded system
21:24 < kic-> with just minimal requirements..
21:25 < sepp_> that should be fine
21:25 < kic-> kernel/couple of specific modules, and couple of specific utility packages
21:25 < sepp_> but you dont get kde or gnome yet ...
21:25 < kic-> thats fine
21:26 < kic-> can't run it anyway on a 128mb flash disk, and no vga port
21:26 < kic-> :d
21:26 < sepp_> ;)
21:29 < kic-> wonder how much bandwidth the mirrors use
21:29 < kic-> i'm putting up a T2 mirror in new york
21:29 < kic-> on a 10mb ethernet
21:29 < sepp_> ask rxr
21:29 < kic-> he's probably in bed by now i think
21:30 < sepp_> ask him later ...
21:30 < kic-> yah, i will
21:52 -!- kic_ [n=kic^^@adsl-074-246-112-050.sip.bct.bellsouth.net] has joined #t2
21:52 < kic_> sepp
21:52 < kic_> you still here?
21:53 < kic_> or anyone that might know how to disable a couple of modules being built
21:54 < kic_> i had another build running, and it was almost done, got to created bzImage, etc
21:54 < kic_> but build errored out on module build error on a couple of modules that we don't need
21:56 -!- kic_ [n=kic^^@adsl-074-246-112-050.sip.bct.bellsouth.net] has quit [Client Quit]
21:57 < kic-> heh
22:01 < mtr> disable kernel modules in the same way as the packages itself, add a rule like "O iwlwifi" (or whatever fails to build) to your custom package selection
22:03 < kic-> k, thanks
22:03 < sepp_> kic-: yes
22:04 < sepp_> :)
22:04 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has quit [Read error: 110 (Connection timed out)]
22:07 < sepp_> librtld.os: In function `_dl_lookup_symbol_x' undefined reference to `__sync_fetch_and_add_4'
22:08 < sepp_> where would that nicely named function come from?
22:08 < sepp_> at 1-glibc32 ...
22:09 < sepp_> ohh, ok
22:12 -!- PAVLINUX [n=pavel@ppp85-140-133-188.pppoe.mtu-net.ru] has quit ["using sirc version 2.211+KSIRC/1.3.12"]
22:12 < mtr> sepp_: ?
22:13 < sepp_> needs at least -march=i486 ...
22:13 < mtr> the 4.2.0 stdc++ headers seem to need a proper "-march=" option
22:13 < mtr> :)
22:13 < sepp_> why are all removed and none appended on a multilib x86-64 build ?
22:16 < kic-> hrm
22:16 < kic-> heres a question
22:16 < kic-> our developers have created our application, as a loadable kernel module
22:17 < kic-> is there a way to put our kernel module in so that it compiles against the kernel source during Build-Target
22:17 < kic-> ?
22:17 < sepp_> yes
22:17 < kic-> or even Build-Pkg linux26
22:17 < kic-> ahh, whew
22:17 < kic-> was gonna say, thats a show stopper for us if we can't do it
22:17 < kic-> :D
22:17 < kic-> wouldn't happen to know how we can do that do ya?
22:17 < kic-> :D
22:17 < sepp_> 3rd party kernel mods are always build during linux*
22:18 -!- sparc-kly__ [n=mubex@adsl-64-237-248-104.prtc.net] has joined #t2
22:18 < sepp_> look at package/network/acx100/
22:19 < sepp_> there are a few examples - ls package/*/*/pkg_linux26_post.conf
22:20 < kic-> ok
22:20 < sepp_> if you are lucky you just need to append a makeopt
22:20 < sepp_> so that the package knows where the kernel source is
22:20 < mtr> sepp_: the x86-64 march option is set in architecture/x86-64/config.in, should be non-NULL for non-Generic
22:22 < sepp_> :)
22:23 < sepp_> i was too lazy to select something, just wanted to see if 1-perl really fails with gdbm ...
22:24 < kic-> :D
22:24 < mtr> :)
22:24 < kic-> did it
22:24 < kic-> ?
22:25 < sepp_> never reached that 1-perl this way
22:25 < sepp_> lol
22:25 < kic-> ah
22:25 < kic-> :D
22:26 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has joined #t2
22:28 < sepp_> now i start from 0 ...
22:29 < sepp_> maybe it should not default to non working generic in the 1st place
22:32 < kic-> :P
22:35 -!- sparc-kly_ [n=mubex@adsl-64-237-248-104.prtc.net] has quit [Read error: 110 (Connection timed out)]
23:17 -!- roman [n=roman@85-127-134-3.dynamic.adsl-line.inode.at] has joined #t2
23:29 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has quit ["ChatZilla 0.9.78.1 [Firefox 2.0.0.4/2007051502]"]
23:48 < sepp_> == 06/20/07 23:45:01 =[1]=> Finished building package perl.
23:49 < sepp_> ^- with minimal-xorg on x86-64 (without changes)
--- Log closed Thu Jun 21 00:00:17 2007
00:01 -!- kic_ [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
00:05 -!- DeeKey [n=denis@84-50-16-45-dsl.kjj.estpak.ee] has joined #t2
00:08 < DeeKey> hello
00:13 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Read error: 110 (Connection timed out)]
00:20 -!- kic_ [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
00:21 < Ragnarin> re
00:21 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
00:31 -!- jnewell [n=jnewell@pr-66-150-46-254.wgate.com] has joined #t2
00:31 -!- DeeKey [n=denis@84-50-16-45-dsl.kjj.estpak.ee] has quit [Read error: 110 (Connection timed out)]
00:33 -!- jnewell [n=jnewell@pr-66-150-46-254.wgate.com] has quit [Remote closed the connection]
00:49 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
00:49 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
01:00 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
01:00 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
01:10 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
01:10 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
01:20 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
01:20 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
01:30 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
01:31 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
01:31 -!- mtr_ [n=Michael@pD9E13447.dip0.t-ipconnect.de] has joined #t2
01:45 -!- mtr [n=Michael@pD9E1111A.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)]
01:45 -!- mtr_ is now known as mtr
01:50 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
01:51 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
02:01 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Remote closed the connection]
02:01 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has joined #t2
04:57 -!- kic- [n=kicstah@adsl-074-238-088-194.sip.bct.bellsouth.net] has quit [Read error: 113 (No route to host)]
06:38 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has joined #t2
06:38 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has quit [Client Quit]
08:16 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has joined #t2
08:16 < tri> moin
08:22 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has quit [Read error: 104 (Connection reset by peer)]
08:22 -!- tr1 [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has joined #t2
08:27 < Fatal> morning
08:27 -!- tr1 [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has quit [Read error: 104 (Connection reset by peer)]
08:37 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has joined #t2
08:57 < rxr> re
08:57 < rxr> moin all
08:58 < Fatal> good morning
08:58 < tri> moin
08:58 < tri> rxr: have a problem to build the iwlwifi modul
08:59 < tri> if i build the kernel the log says something about "could not find mac80211 headers" in /lib/modules/....
09:01 < tri> MAC80211_INC=/lib/modules/2....-dist/source/include/net <---
09:01 < tri> in the kernel build log
09:01 < tri> ragnarin told me yesterday to change change MAC80211_INC to be $kerneldir/include/net
09:02 < tri> but without success same error message again
09:04 -!- tr1 [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has joined #t2
09:04 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has quit [Read error: 104 (Connection reset by peer)]
09:04 < tr1> re
09:04 < tr1> hm damm internet connection....
09:05 < rxr> tr1: hm - do you use the 7.0-rc release or trunk ?
09:05 < tr1> trunk
09:06 < rxr> hm - I thought I fixed that
09:06 < rxr> probably time for a test-build on my side
09:06 < tr1> in the iwlwifi postlinux.conf there is MAC80211_INC=$moduledir/source/include/net
09:07 < rxr> if you want to take a look yourself, as Ragnarin said it is probably just a matter of setting some make variable to let it pick up on cross build
09:07 < rxr> btw. on which stage does it fail? 1 (cross build) or 5 (native build)?
09:07 < tr1> native build
09:07 < tr1> i do emerg-pkg -force linux26
09:07 < rxr> ah - ok, stage 9 is native build as well
09:08 < tr1> hm where are the mac80211 kernel headers
09:08 < rxr> (emerges run in stage 9, which is last thing of native builds)
09:08 < rxr> they are in the kernel and the third party module usually wants to find them via the temp. kernel source or module dir
09:08 < Ragnarin> btw I tried it again, builds fine for me
09:09 < Ragnarin> does the mac80211 patch apply cleanly earlier?
09:10 < tr1> ah hi ragnarin ;-)
09:10 < tr1> have go to sleep yesterday i hope your lunch was good
09:11 < Ragnarin> yea it was
09:11 < Ragnarin> you do have mac80211 in your pkgsel?
09:11 < tr1> ah no !
09:11 < Ragnarin> then that may very well be the reason :D
09:11 < rxr> lol
09:12 < rxr> heh - I build an update here now anyway, good for security in any case .-)
09:12 < tr1> hm but at least the error was found...
09:12 < tr1> ah ok why using SETI@home if you can build the kernel again...
09:13 < rxr> in T2 you can set a debug flag to just build the modules
09:13 < rxr> (not the kernel)
09:13 < rxr> (and use ccache anyway :-)
09:13 < tr1> how to set such a debug flag ?
09:13 < rxr> SDEDEBUG_LINUX_SUBMODULES_ONLY=1 ./scripts/Emerge ... linux26
09:13 < tr1> is it enough to edit the pkgsel with vim or do i have to run any "newconfig" command ?
09:14 < rxr> he target/.../pkgsel ?
09:14 < rxr> ^-the
09:14 < rxr> after you edit the pkgsel you need to Config -oldconfig
09:14 < rxr> but hey, wait
09:15 < rxr> the Bulid-Target already runs the Config -oldconfig automagically on-demand
09:15 < rxr> I think we should fixed Emerge-Pkg to do the same
09:15 < mtr> moin
09:15 < tr1> hi mtr
09:16 < rxr> moin moin mtr
09:16 < CIA-9> susan * r24647 /trunk/package/graphic/dcraw/dcraw.desc: * updated dcraw (8.74 -> 8.75)
09:16 < CIA-9> susan * r24648 /trunk/package/network/ddclient/ddclient.desc: * updated ddclient (3.7.1 -> 3.7.2)
09:19 < CIA-9> mtr * r24649 /trunk/package/www/firefox/firefox.conf:
09:19 < CIA-9> * fixed firefox to actually install the nss header files
09:19 < CIA-9> (folder name changed)
09:21 < rxr> oh - that might be the reason why nautilus and openoffice-org did not found those when I wanted to build them against firefox. Good catch!
09:25 < mtr> evolution depends on the mozilla nss libs, too
09:27 < CIA-9> rene * r24650 /trunk/package/mail/msmtp/msmtp.desc: * updated msmtp (1.4.11 -> 1.4.12)
09:27 < CIA-9> rene * r24651 /trunk/package/network/ndiswrapper/ndiswrapper.desc: * updated ndiswrapper (1.46 -> 1.47)
09:28 < CIA-9> rene * r24652 /trunk/package/graphic/cowsay/ (cowsay.conf cowsay.desc): * de-custmained cowsay
09:31 < CIA-9> rene * r24653 /trunk/package/ (8 files in 4 dirs): * fixed vamps, p0f, wipe, (the classic) hotplug, to cross build
09:55 < tr1> hm i build the kernel now
09:55 < tr1> but modprobe iwl3945 said module iwl3945 not found
09:56 < tr1> build was fine this time
09:57 < tr1> i also have the firmware in /lib/firmware
09:57 < tr1> and have build it into the initrd
09:59 < rxr> is the module in /lib/modules/... ?
10:00 < tr1> ... ?
10:00 < tr1> hm in KMISC=$moduledir/kernel/drivers/net/wireles
10:00 < tr1> shouldn't it be wireless ?
10:01 < tr1> http://svn.exactcode.de/t2/trunk/package/network/iwlwifi/postlinux.conf
10:01 < tr1> is there a typo ?
10:01 < rxr> probably, but it shoudl work nontheless
10:02 < rxr> the tools too not care about a name at all
10:03 < CIA-9> rene * r24654 /trunk/package/network/iwlwifi/postlinux.conf: * fixed wireles -> wirele_ss_ typo in iwlwifi/postlinux.conf
10:03 < tr1> yes did that
10:05 < mtr> rxr: firefox uses a GCC_WRAPPER_INSERT to globally inject X11 libs, I want to replace this by a simple one-liner patch:
10:05 < mtr> + ifdef MOZ_ENABLE_GTK2
10:05 < mtr> + EXTRA_DSO_LDOPTS += $(MOZ_GTK2_LIBS) \
10:05 < mtr> ++ $(XLDFLAGS) -lX11 -lXrender \
10:05 < mtr> + $(NULL)
10:05 < mtr> + endif
10:06 < rxr> it does build ?
10:07 < mtr> yes
10:07 < rxr> well - sure
10:07 < mtr> for x86
10:07 < rxr> but I think we have this in most newer mozilla a-like packages
10:07 < rxr> thunderbird and co as well
10:07 < mtr> but XLDFLAGS should cover x86_64 as well
10:07 < rxr> did you fix those as well ?
10:07 < rxr> btw. I wonder why it builds on other distributions without the patch
10:07 < mtr> untested
10:08 < rxr> if you find the time to patch the other as well it woudl be appreciated
10:08 < rxr> I can test build them on the ref tester
10:08 < mtr> ok
10:11 < CIA-9> rene * r24655 /trunk/package/network/nginx/nginx.desc: * updated nginx (0.5.24 -> 0.5.25)
10:11 < CIA-9> rene * r24656 /trunk/package/develop/hte/hte.desc: * updated hte (2.0.6 -> 2.0.7)
10:11 < CIA-9> rene * r24657 /trunk/package/base/pcre/pcre.desc: * updated pcre (7.1 -> 7.2)
10:11 < CIA-9> rene * r24658 /trunk/package/audio/libgpod/libgpod.desc: * updated libgpod (0.4.2 -> 0.5.0)
10:12 < CIA-9> rene * r24659 /trunk/package/filesystem/udev/udev.desc: * updated udev (111 -> 112)
10:13 < CIA-9> rene * r24660 /trunk/package/base/device-mapper/device-mapper.desc: * updated device-mapper (1.02.19 -> 1.02.20)
10:13 < CIA-9> rene * r24661 /trunk/package/kde/kalarm/kalarm.desc: * updated kalarm (1.4.11 -> 1.4.13)
10:13 < CIA-9> rene * r24662 /trunk/package/kde/digikam/digikam.desc: * updated digikam (0.9.1 -> 0.9.2)
10:28 < mtr> weird ... or normal? ... do we generally have a stripped /lib/libpthread.so?
10:29 < rxr> if you do not built with debugging enabled it probably will be stripped, yes
10:29 < mtr> seems that libthread_db cannot resolve some symbols for debugging
10:30 < mtr> how can I disable stripping, without SDECFG_DEBUG?
10:33 < rxr> yes
10:36 < sepp> moin :)
10:36 < tr1> hi sepp
10:36 < sepp> hi tr1
10:37 < rxr> moin sepp
10:37 < mtr> hi sepp
10:37 < sepp> hi rxr, mtr :)
10:37 [Users #t2]
10:37 [@ChanServ] [ Dallur] [ LMJ ] [ rxr ] [ TobiX ]
10:37 [ cbrake ] [ emte ] [ mtr ] [ sepp ] [ tr1 ]
10:37 [ CIA-9 ] [ Fatal ] [ Ragnarin] [ Stelz] [ valentin]
10:37 -!- Irssi: #t2: Total of 15 nicks [1 ops, 0 halfops, 0 voices, 14 normal]
11:00 < tr1> ah now i have the ipw3945 module in /lib/ ...
11:01 < tr1> just one reboot away from modprobe it
11:01 -!- idealm [n=ideal@c58-107-18-66.belrs2.nsw.optusnet.com.au] has joined #t2
11:02 < sepp> why do you need to reboot to do that?
11:03 < tr1> because it is build for the new kernel
11:03 < tr1> hm after i do emerge-pkg linux26
11:03 < tr1> i also have to change grub settings ?
11:03 < tr1> or is this done automaticly ?
11:04 < sepp> if it is a new version, you can use stone
11:04 < CIA-9> rene * r24663 /trunk/package/ (17 files in 8 dirs):
11:04 < CIA-9> * fixed stunnel, metamail, transconnect, nudelstats, xskat, wakkabox,
11:04 < CIA-9> prngd, and mplayer-fonts to cross build
11:05 < tr1> can i load a new linux kernel without rebooting ?
11:05 < sepp> no
11:05 < rxr> yes with kexec
11:07 < tr1> hm some time ago i read something about that option - i will have a look at kexec just for intrest
11:07 < tr1> till now i edited the grub menu with vim but stone is nice to edit it: it founds all kernels...
11:08 < tr1> tada: modprobe iwl3945
11:08 < tr1> and now i have a new device wmaster0 and wlan0
11:08 < rxr> I think kexec is not yet enabled in T2 default kernels
11:08 < rxr> experiemntal anyway
11:09 < tr1> but also a warning: driver for device wlan0 was been compiled with version 22 of wireless extension
11:09 < tr1> while this programm supports up to version 20 . some things may be broken
11:09 < rxr> you can ingore that ...
11:09 < tr1> ah ok
11:09 < tr1> that's from the iwlwifi driver ?
11:09 < rxr> I think we should patch this nonsense message away
11:09 < rxr> no, I think the wireless tools print that out
11:10 < rxr> I bet SuSE and RedHat patch that away ...
11:10 < rxr> I never found anything broken either ...
11:11 -!- sepp_ [n=sepp@Z574e.z.pppool.de] has joined #t2
11:14 < tr1> yes sure SuSE (or is it typed S.U.S.E or suse or SusE) and RedHat patch a lot of the packages...
11:15 -!- Dallur [n=earl@fire-out.ru.is] has quit ["Leaving."]
11:16 -!- Dallur [n=earl@fire-out.ru.is] has joined #t2
11:19 < tr1> from the rt wiki: "Controlling a laser with Linux is crazy, but everyone in this room is crazy in his own way. So if you want to use Linux to control an industrial welding laser, I have no problem with your using PREEMPT_RT." -- Linus Torvalds
11:21 < rxr> heh
11:21 < tr1> kexec is not part of t2 yet ?
11:22 < rxr> it's part of recent kernels
11:22 < rxr> you just have to enable it in the .config
11:22 < rxr> t2 does not do that, yes, as far as I know
11:22 < rxr> despite if the default state for that CONFIG_KEXEC stuff changed and it is on by default :_)
11:23 < tr1> aren't the kexec-tools needed ?
11:23 < Dallur> tr1: I'm controlling a 2.5m x 6.5m Plasma cutting robot with linux myself, laser is actually simpler and easier to do than plasma
11:24 < rxr> tr1: probably, maybe submit a patch adding those
11:24 < Dallur> tr1: http://linuxcnc.org/
11:26 < CIA-9> mtr * r24664 /trunk/package/www/firefox/ (firefox.conf gtk2-x11-libs.patch): * remove firefox build workaround, use a makefile patch instead to add X11 libs
11:26 < tr1> rxr: yip, i will do that
11:26 < Fatal> heh, shame I don't get enough time to port our system to t2, or I would have kexec stuff added already :)
11:26 < Fatal> goodie, one less thing I'll have to worry about \o/
11:26 < CIA-9> mtr * r24665 /trunk/package/www/seamonkey/ (gtk2-x11-libs.patch seamonkey.conf): * remove seamonkey build workaround, use a makefile patch instead to add X11 libs
11:28 -!- sepp [n=sepp@Z4174.z.pppool.de] has quit [Read error: 110 (Connection timed out)]
11:29 -!- sepp_ is now known as sepp
11:32 < mtr> we do have /usr/bin/perl setuid, is ths intentional?
11:36 < tr1> hm: There may be times you would want to run perl scripts as non-root users. This offers more security to your system and this guide will show you how to enable perl to use setuid.
11:37 < tr1> from *bsd wiki
11:37 < tr1> maybe it is a good idea to have "ENABLE_SUIDPERL=true"
11:44 < rxr> mtr: this is historic
11:44 < rxr> mtr: I noticed this caues some fooomatic-rip stufff to error out just because of the setuid bit
11:47 < mtr> and some aclocal regressions:
11:47 < mtr> Insecure $ENV{PATH} while running setuid at /usr/lib/perl5/5.8.8/i686-t2-linux-gnu/IO/File.pm line 192.
11:47 < mtr> make: *** [aclocal.m4] Fehler 1
11:52 < tr1> what is the way if there are more the one author use two or three [A]
11:52 < rxr> yeah - that is also like what some foomatic stuff spits into the log for certain printer rasterizers ..
11:52 < tr1> or just put them all in one [A] ?
11:53 < rxr> tr1: yes, multiple [A]'s is just fine
11:53 < rxr> many packages ahve that already
12:48 -!- idealm [n=ideal@c58-107-18-66.belrs2.nsw.optusnet.com.au] has quit [Read error: 110 (Connection timed out)]
12:49 -!- idealm [n=ideal@58.105.232.64] has joined #t2
12:54 -!- tr1 [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has quit [Read error: 104 (Connection reset by peer)]
13:13 -!- idealm [n=ideal@58.105.232.64] has quit [Read error: 110 (Connection timed out)]
13:19 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has joined #t2
13:23 < tri> re
13:23 < rxr> re
13:32 < sepp> er
13:32 < tri> http://ubuntuforums.org/showthread.php?t=475709
13:35 < tri> do i have to give the full download url in the [D]
13:35 < tri> or can i X filename http://aewfawefawef/dasdf/ ?
13:36 < sepp> the later
13:46 < tri> what is $confdir ?
13:55 < tri> how to build a package which don't have the ./configure make make install thing
13:55 < tri> but ah tar.gz
13:55 < tri> where a install.sh script is int ?
13:55 < tri> s/int/in
13:56 < tri> and that should be executed ?
13:56 < tri> i tried:
13:56 < tri> makeopt=
13:56 < tri> makeinstopt=
13:56 < tri> hook_add postmake 5 'install.sh /'
13:56 < tri> but that didn't work ?
13:58 -!- idealm [n=ideal@58.105.232.64] has joined #t2
14:08 < tri> what is the var for the extracted package dir ?
14:09 < tri> like builddir for the builddir is there also a var for "builddir/extrackt_package/ ?
14:18 < tri> ah
14:18 < tri> no i have used in postmake 5 '$builddir/$pkg/install.sh /'
14:18 < tri> no/now
14:18 < tri> is this the right way ?
14:19 < rxr> re
14:20 < rxr> well - this will break for cross builds
14:20 < rxr> you need at least $root/
14:20 < rxr> but what kind of package is that? just installing files? or compiling something ?=
14:21 < tri> just installing files
14:22 < tri> postmake 5 '$builddir/$pkg/install.sh /' break cross builds ?
14:22 < tri> how to do it "better" ?
14:24 < sepp> use $root/ as target
14:24 < tri> ah ok not "/" ?
14:25 < sepp> no ;)
14:25 < tri> ?
14:25 < sepp> not just "/"
14:25 < tri> ah yes not i understand it
14:25 < rxr> just using / will place the new files int the system you build on, not in the new sandbox
14:26 < tri> cross build isn't installing any files here
14:26 < rxr> though T2 "should" scream about this before you mess your system up
14:26 < tri> but for the tar.gz package which is created
14:26 < tri> for the crossbuild
14:26 < tri> right ?
14:26 < tri> ah ok
14:26 < tri> sorry i'm not that family with t2 but it is getting a bit better...
14:27 < tri> hm i allready wondered why build-pkg installed the files...
14:28 < tri> hm build package was working
14:28 < tri> but emerg-pkg said Error: No such a package
14:28 < tri> the package is just in my target might that be the problem ?
14:28 < sepp> it was not yet in config/.../packages
14:29 < tri> how to get it into config/.../packages ?
14:30 < sepp> rerun Config with -oldconfig
14:30 < rxr> rerun Config with -oldconfig
14:30 < rxr> now that was synchron typing
14:30 < sepp> hehe :)
14:30 < tri> :-)
14:30 < tri> thx both ;-)
14:30 < tri> i try both commands...
14:31 < tri> hm it said Configuration finished but emerg-pkg still not working
14:32 < sepp> it is not enabled? X ... ?
14:33 < tri> hm not it is not in there
14:33 < tri> in the package file
14:41 < tri> ?
14:43 < rxr> in the [P]riority tag of the package file
14:43 < rxr> maybe upload the .desc and we take a look
14:43 < tri> thx :-)
14:43 < tri> P X-1---5---9 100.000
14:44 < rxr> missing space
14:44 < rxr> [P] X -1---5---9 100.000
14:45 < rxr> and probably the patckage should not be built in stage 1 forced, as stage is is "cross build what is needed for a new freestanding system
14:45 < rxr> "
14:49 < tri> http://imi.st/mkinitcpio.desc and http://imi.st/mkinitcpio.conf
14:53 < rxr> oh
14:53 < tri> ?
14:54 < rxr> (just wander if we need such a package)
14:54 < rxr> was hopking for kexec tools or so
14:54 < rxr> what is this script doing? calling cpio ?
14:54 < tri> jop it is building a initrd
14:55 < rxr> the scripts look ok so far
14:55 < rxr> where did you place those ?
14:55 < tri> kexec-tools will follow at the evening
14:55 < tri> place those ?
14:55 < rxr> where in package/ ?
14:55 < tri> ah ok
14:56 < tri> in target/.../package/mkinitcpio
14:56 < rxr> oh
14:56 < tri> hm oh's are good ?
14:56 < rxr> I think this will cause all sorts of horror happening
14:56 < rxr> you can only extend packages that exist
14:56 < rxr> place it in package/contrib/mkinitcpio
14:56 < tri> ah ok
14:57 < tri> what is contrib for (the dir) ?
14:57 < rxr> probably we should harden some scripts here and there for such cases
14:57 < rxr> package/contrib/
14:57 < rxr> for stuff that is not yet decided where to place
14:57 < tri> ah ok
15:00 < tri> so it is not the right way to "test" package in my own target
15:00 < tri> target's are more for special config options package selections etc ?
15:01 < tri> hm still not
15:01 < tri> i moved it but emerge-pkg still don' know the package
15:01 < tri> i also did -oldconfig
15:01 < tri> build-pkg still works
15:01 -!- idealm [n=ideal@58.105.232.64] has quit [Read error: 110 (Connection timed out)]
15:03 < tri> hm but now it is int packages file
15:04 < rxr> k# ./scripts/Emerge-Pkg mkinitcpio
15:04 < rxr> checking mkinitcpio ... Not installed. Added.
15:04 < rxr> Packages scheduled to build: mkinitcpio
15:04 < rxr> works for me
15:04 < rxr> so far
15:05 < tri> ah i have to use -force in emerg-pkg
15:05 < tri> sorry for the noise works here to...
15:06 < tri> :-(
15:06 < rxr> hm ?
15:06 < rxr> maybe not selectd in your target or so ...
15:14 < rxr> tri: should I commit your package, now ?
15:17 < CIA-9> rene * r24666 /trunk/package/contrib/mkinitcpio/ (. mkinitcpio.cache mkinitcpio.conf mkinitcpio.desc):
15:17 < CIA-9> Tobias Richter
15:17 < CIA-9> * added mkinitcpio (0.5.9) - mkinitcpio
15:27 < tri> wups that was fast ;-)
15:30 < CIA-9> rene * r24667 /trunk/package/contrib/mkinitcpio/mkinitcpio.desc:
15:30 < CIA-9> * imroved mkinitcpio to have a trailing dot in the Text tag, a
15:30 < CIA-9> checksun, and not build it in stage 1 by default
15:35 < tri> no need to add this packages but thx that's nice ;-)
15:35 < tri> i only want to make a few tests with mkinitcpio
15:36 < rxr> we can still have it as an option and/or remove it later on anyway :_)
15:36 < rxr> "release early release often"
15:38 < tri> yes, nice :-)
15:42 < tri> and i like this option even for "burning apps" there are a few different packages
15:42 < rxr> packages missing ?
15:43 < tri> no i mean there are min. 5 different burning apps why not use 2 different initrd shell scripts ?
15:43 < tri> s/use/provide
15:45 < rxr> ah - sure
15:52 < CIA-9> rene * r24668 /trunk/package/base/gpm/ (gpm.conf gpm.desc):
15:52 < CIA-9> * fixed gpm to cross build in way it is also functional thereafter,
15:52 < CIA-9> and not leak stray files in /etc/$root :-)
16:16 -!- kic- [n=kicstah@adsl-074-238-088-206.sip.bct.bellsouth.net] has joined #t2
16:17 < kic-> hi
16:17 < tri> hi kic-
16:17 < rxr> hi kic-
16:17 < kic-> hi there
16:18 < kic-> rene
16:18 < kic-> we are looking at possibly the week of July 2nd
16:18 < kic-> can i /msg you?
16:19 < rxr> kic-: yes, just msg me always :-)
16:19 < kic-> k
16:20 < rxr> sepp: do you have the latest viewcvs of e17 handy ?
16:24 < sepp> i think there is none since they left sf.net
16:27 < rxr> :-(
16:30 < sepp> http://web.enlightenment.org/viewvc/e17/ ahh
16:30 < sepp> there it is
16:31 < rxr> oh - thank you so much
16:31 < tri> http://e.kevb.net/cgi-bin/viewvc.cgi/
16:31 < tri> would also be a option ;-)
16:32 < sepp> ohh
16:34 < sepp> which one was the outdated one?
16:34 < rxr> thanks - found what I was looking for
16:34 < sepp> evas ?
16:35 < rxr> no the evas demos, apparently outsourced into expedite, now
16:35 < rxr> ok - gotta go - back in 1h
16:36 < tri> cu rxr
16:52 < tri> bye
16:52 -!- tri [n=Richter@pd95b2df8.dip0.t-ipconnect.de] has left #t2 []
16:54 -!- keztoo [n=root@adsl-074-238-088-206.sip.bct.bellsouth.net] has joined #t2
17:37 < rxr> re
17:38 < rxr> welcome keztoo
17:38 [Users #t2]
17:38 [@ChanServ] [ Dallur] [ keztoo] [ mtr ] [ sepp ] [ valentin]
17:38 [ cbrake ] [ emte ] [ kic- ] [ Ragnarin] [ Stelz]
17:38 [ CIA-9 ] [ Fatal ] [ LMJ ] [ rxr ] [ TobiX]
17:38 -!- Irssi: #t2: Total of 16 nicks [1 ops, 0 halfops, 0 voices, 15 normal]
17:39 -!- sepp_ [n=sepp@Z46ef.z.pppool.de] has joined #t2
17:43 < keztoo> hi
17:43 < keztoo> thank you
17:44 < keztoo> rxr i have a question when you have a minute
17:46 < rxr> sure
17:46 < keztoo> from a high level
17:47 < keztoo> when doing a cross build the system sometimes tries to chroot and run a target tool
17:47 < keztoo> but if the host machine doesnt support the target instruction set it will always fail
17:47 < CIA-9> rene * r24669 /trunk/package/e17/expedite/ (. expedite.cache expedite.desc): * added expedite (2007-06-10) - A test and benchmark for the Evas canvas library
17:48 < keztoo> kind of like 1.1.1 x11 where it tries to run makekeys but it cant run 64 bit on a 32 bit host
17:48 < rxr> keztoo: in T2 on a cross build nothing does like chroot
17:48 < rxr> keztoo: some packages (that we did not already fix) build helper tools with the cross compiler and try to run them directly
17:48 < rxr> keztoo: just like ./gen-table ... foo bar
17:48 < keztoo> yes, that is what i have seen
17:48 < rxr> those need to be fixed to build the helper with the host compiler, not the cross one
17:49 < rxr> we'll successive correct such instances
17:49 < rxr> we could also integrate a "target system emulator" to run those helper apps, and could build more automatically
17:49 < keztoo> so for example with x11 makekeys the make has to be changed to run the 32 (or host) version , right ?
17:49 < rxr> yes,
17:50 < rxr> but I'm under the impression we can build X.org stuff cross already
17:50 < keztoo> ok, just wanted to make sure what i was seeing was what was happening
17:50 < rxr> do you use trunk ?
17:50 < rxr> trunk:HEAD
17:50 < keztoo> i somehow got a 1.1.1 pkg but i noticed the latest is 1.1.2
17:50 < keztoo> no, i used the standard mirrors
17:50 < keztoo> i am trying to get the head w svs now
17:51 < keztoo> i also tried to change the package to grab 1.1.2 but of course since i didnt fix all the patches i got invalid chunks
17:53 < keztoo> i will grab the latest head now and try again, thank you for your input
17:53 < rxr> for really complex stuff we later could also support running qemu on those cross built binaries, but we want to cross build as much as possible first, as that is the cleanest, most portable and also fasted solution
17:54 < rxr> keztoo: your're welcome
17:57 -!- sepp [n=sepp@Z574e.z.pppool.de] has quit [Read error: 110 (Connection timed out)]
18:00 -!- kic- [n=kicstah@adsl-074-238-088-206.sip.bct.bellsouth.net] has quit [Read error: 104 (Connection reset by peer)]
18:00 -!- keztoo [n=root@adsl-074-238-088-206.sip.bct.bellsouth.net] has quit [Read error: 104 (Connection reset by peer)]
18:02 -!- keztoo [n=root@adsl-074-238-088-206.sip.bct.bellsouth.net] has joined #t2
18:05 < CIA-9> rene * r24670 /trunk/package/develop/libatomic_ops/ (. libatomic_ops.cache libatomic_ops.desc): * added libatomic_ops (1.1) - The atomic_ops project
18:06 -!- kic- [n=kicstah@adsl-074-238-088-206.sip.bct.bellsouth.net] has joined #t2
18:10 < keztoo> to get the trunk:head is this the correct svn cmd ...
18:11 < keztoo> svn co http://svn.exactcode.de/t2/trunk t2-head ??
18:13 < rxr> yes
18:13 < rxr> and then run svn up inside the directory to update from the version control server
18:15 < keztoo> i just executed svn co http://svn.exactcode.de/tr/trunk t2-trunk
18:15 < keztoo> this should also be ok , right ??
18:15 < keztoo> as long as i do the update after it downloads?
18:16 < rxr> t2/trunk
18:16 < rxr> when it downloaded it is up-to-date
18:16 < keztoo> ok, so i did the right one
18:16 < keztoo> ty
18:17 < rxr> the "svn up" is just for like in 8 hours or tomorrow when you want to receive the delta of "what changed between now and the time you run it"
18:17 < keztoo> got it, ty
18:38 < kic-> brb, switching network
18:38 -!- kic- [n=kicstah@adsl-074-238-088-206.sip.bct.bellsouth.net] has quit []
18:45 -!- PAVLINUX [n=pavel@ppp85-140-133-188.pppoe.mtu-net.ru] has joined #t2
18:46 < PAVLINUX> Hello World
18:47 < rxr> hello PAV
18:48 < keztoo> hi
18:48 [Users #t2]
18:48 [@ChanServ] [ Dallur] [ keztoo] [ PAVLINUX] [ sepp_] [ valentin]
18:48 [ cbrake ] [ emte ] [ LMJ ] [ Ragnarin] [ Stelz]
18:48 [ CIA-9 ] [ Fatal ] [ mtr ] [ rxr ] [ TobiX]
18:48 -!- Irssi: #t2: Total of 16 nicks [1 ops, 0 halfops, 0 voices, 15 normal]
18:48 < mtr> hi PAVLINUX
18:49 < rxr> hi mtr
18:50 < PAVLINUX> What do you think about this - http://www.stardust.webpages.pl/files/handbook/handbook-en-0.3-rc1.pdf
18:51 < rxr> oeh - puh
18:51 < rxr> that is a long read - what particualry do you have in mind ?
19:03 < CIA-9> rene * r24671 /trunk/package/ (4 files in 3 dirs):
19:03 < CIA-9> * fixed libsndfile to cross build (patched configure.ac fragment did
19:03 < CIA-9> not expand correctly)
19:03 < CIA-9> * marked liboil, libsamplerate to cross build
19:04 -!- sparc-kly_ [n=mubex@adsl-64-237-248-104.prtc.net] has joined #t2
19:04 < CIA-9> rene * r24672 /trunk/package/audio/pulseaudio/ (. pulseaudio.cache pulseaudio.desc): * added pulseaudio (0.9.6) - Yet another sound server for Unix
19:05 [Users #t2]
19:05 [@ChanServ] [ Dallur] [ keztoo] [ PAVLINUX] [ sepp_ ] [ TobiX ]
19:05 [ cbrake ] [ emte ] [ LMJ ] [ Ragnarin] [ sparc-kly_] [ valentin]
19:05 [ CIA-9 ] [ Fatal ] [ mtr ] [ rxr ] [ Stelz ]
19:05 -!- Irssi: #t2: Total of 17 nicks [1 ops, 0 halfops, 0 voices, 16 normal]
19:06 < keztoo> thats a big document PAV, i think i'll curl up in front of the fireplace with a bottle of wine and a loved one to read it
19:06 < keztoo> :-)
19:24 < PAVLINUX> 66 pages! It's BIG DOCUMENT ;-\
19:24 -!- PAVLINUX [n=pavel@ppp85-140-133-188.pppoe.mtu-net.ru] has quit [Remote closed the connection]
19:30 < CIA-9> rene * r24673 /trunk/package/ (4 files in 3 dirs):
19:30 < CIA-9> * fixed libfame to cross build for $arch != x86
19:30 < CIA-9> * marked libiriverdb and ruby to cross compile
19:31 < CIA-9> rene * r24674 /trunk/package/gnome2/evince/evince.desc: * updated evince (0.9.0 -> 0.9.1)
19:31 < CIA-9> rene * r24675 /trunk/package/doc/howtos/howtos.desc: * updated howtos (20070609 -> 20070616)
19:44 < rxr> ok - cu all then
19:44 < keztoo> by rxr and ty
19:48 -!- keztoo [n=root@adsl-074-238-088-206.sip.bct.bellsouth.net] has quit ["using sirc version 2.211+KSIRC/1.3.12"]
19:48 -!- PAVLINUX [n=pavel@ppp85-140-133-188.pppoe.mtu-net.ru] has joined #t2
19:58 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has joined #t2
19:59 < misl> hi folks
20:08 < mtr> hi misl
20:31 -!- kic- [i=kicstah@adsl-074-238-088-201.sip.bct.bellsouth.net] has joined #t2
20:31 < kic-> any known build problems with perl & gdbm?
20:32 < kic-> tried to build minimal desktop with xorg target, and errors out when building perl with ld: cannot find -lgdbm
20:32 < sepp_> gdbm not enabled?
20:33 < kic-> how do you enable/disable it
20:34 < sepp_> grep gdbm config/.../packages
20:34 < sepp_> when the line does not start iwth an X is it not enabled
20:36 < kic-> hrm
20:36 < kic-> it started with O
20:36 < kic-> so i can just edit that file, and change it to an X right?
20:36 < kic-> and re-run the Build-Target
20:36 < sepp_> then it will never find gdbm
20:36 < kic-> ?
20:36 < sepp_> no
20:36 < kic-> hrm
20:36 < sepp_> run Config
20:36 < kic-> ok
20:36 < sepp_> expert options
20:36 < sepp_> custom package selection
20:36 < sepp_> X gdbm
20:37 < kic-> hrm
20:37 < kic-> ok
20:37 < sepp_> that happens with minimal-xorg ?
20:38 < kic-> yep
20:38 < kic-> minimul xorg, with x86-64, core2 optimize, generic system
20:38 < sepp_> ok, will try later and add it to the real pkgsel :)
20:39 < kic-> cool
20:40 < kic-> changed config, trying rebuild now
20:41 < kic-> hrm
20:41 < kic-> same error
20:41 < kic-> any caches i have to clear before re-building?
20:41 < sepp_> no
20:41 < sepp_> has it build gdbm 1st?
20:41 < kic-> cause Build-Target didn't actually start downloading gdbm and building it before perl
20:41 < kic-> it just started right at perl again
20:42 < sepp_> 1-perl ?
20:42 < kic-> yeah
20:42 < sepp_> well, that should build without gdb
20:42 < sepp_> i think
20:43 < sepp_> gdbm
20:43 < kic-> it gets all the way to the bottom of perl compile, and tries to finally link a binary with gcc -o miniperl [more junk] -lgdbm
20:43 < kic-> hrm
20:43 < kic-> anyway i can tell the perl build not to use -lgdbm
20:43 < kic-> ?
20:45 < sepp_> you could remove -lgdbm from the gcc wrapper
20:46 < PAVLINUX> tar -xvjf perl.tar.bz && cd perl && grep -R lgdbm ./*
20:46 < kic-> where can i edit the gcc wrapper?
20:46 < PAVLINUX> vi configure; /gdbm dd
20:48 < sepp_> var_append GCC_WRAPPER_REMOVE ' ' '-lgdbm'
20:49 < sepp_> in package/perl/perl/perl.conf
20:49 < kic-> hrm
20:49 < kic-> k
20:49 < sepp_> after # build a native miniperl
20:51 < kic-> k, re-running build
20:53 < kic-> woot, that worked
20:53 < kic-> on to autoconf now
20:53 < kic-> yay
20:53 < sepp_> :)
20:53 < kic-> i should bring my questions in here more often
20:53 < kic-> heh
20:54 < kic-> can't be to many errors with this one
20:54 < kic-> its just a default build for the most part
20:55 < kic-> kernel patches and compile is one of the last builds right?
20:56 < kic-> its downloading 2.6.21.tar.bz2 at the moment
20:56 < sepp_> no, it builds in several stages
20:56 < sepp_> and linux* is quite early
20:57 < kic-> i saw that it already built linux 2*17 or something
20:57 < kic-> previously
20:57 < kic-> how far into it, is udev?
20:57 < sepp_> very early
20:57 < kic-> i haven't seen what stage its in, cause i walked away earlier
20:57 < kic-> aww
20:57 < kic-> ok
20:57 < kic-> got a while to go then i guess
20:57 < kic-> :D
20:57 < sepp_> that will be a long walk
20:57 < sepp_> :)
20:58 < kic-> hehe
20:58 < kic-> the machine i'm building on has some horse power though
20:58 < kic-> dual xeon 3ghz, with core2 duo, 2GB RAM, 1GB SWAP
21:13 < sepp_> still takes some h
21:16 < kic-> does it build the modules multiple times also?
21:16 < sepp_> modules?
21:16 < kic-> kernel modules
21:17 < sepp_> yes, it builds an complete toolchain 1st, glibc binutils gcc, all that stuff
21:17 < kic-> hrm, neat
21:18 < sepp_> and later it builds the real stuff with the toolchain in the sandbox ...
21:20 < kic-> ahh
21:20 < kic-> thats where some previous builds i was trying were erroring out then
21:21 < kic-> was getting chroot: error bash exec format error
21:21 < kic-> or soemthing like that
21:21 < kic-> yesterday
21:21 < sepp_> that would mean your cpu can not run the binaries you created - for example because it is for an different cpu
21:22 < kic-> erm
21:22 < kic-> that was a for a cross compile though
21:22 < sepp_> like building ppc on some x86*
21:22 < kic-> but you should be able to do that with cross compiling
21:22 < kic-> right?
21:23 < sepp_> no, there is fast growing support for cross compilation but if some package wants to run some thing - it needs a patch
21:24 < sepp_> but you can already crosscompile quite a lot
21:24 < kic-> all of our cross compiling would be for a small embedded system
21:24 < kic-> with just minimal requirements..
21:25 < sepp_> that should be fine
21:25 < kic-> kernel/couple of specific modules, and couple of specific utility packages
21:25 < sepp_> but you dont get kde or gnome yet ...
21:25 < kic-> thats fine
21:26 < kic-> can't run it anyway on a 128mb flash disk, and no vga port
21:26 < kic-> :d
21:26 < sepp_> ;)
21:29 < kic-> wonder how much bandwidth the mirrors use
21:29 < kic-> i'm putting up a T2 mirror in new york
21:29 < kic-> on a 10mb ethernet
21:29 < sepp_> ask rxr
21:29 < kic-> he's probably in bed by now i think
21:30 < sepp_> ask him later ...
21:30 < kic-> yah, i will
21:52 -!- kic_ [n=kic^^@adsl-074-246-112-050.sip.bct.bellsouth.net] has joined #t2
21:52 < kic_> sepp
21:52 < kic_> you still here?
21:53 < kic_> or anyone that might know how to disable a couple of modules being built
21:54 < kic_> i had another build running, and it was almost done, got to created bzImage, etc
21:54 < kic_> but build errored out on module build error on a couple of modules that we don't need
21:56 -!- kic_ [n=kic^^@adsl-074-246-112-050.sip.bct.bellsouth.net] has quit [Client Quit]
21:57 < kic-> heh
22:01 < mtr> disable kernel modules in the same way as the packages itself, add a rule like "O iwlwifi" (or whatever fails to build) to your custom package selection
22:03 < kic-> k, thanks
22:03 < sepp_> kic-: yes
22:04 < sepp_> :)
22:04 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has quit [Read error: 110 (Connection timed out)]
22:07 < sepp_> librtld.os: In function `_dl_lookup_symbol_x' undefined reference to `__sync_fetch_and_add_4'
22:08 < sepp_> where would that nicely named function come from?
22:08 < sepp_> at 1-glibc32 ...
22:09 < sepp_> ohh, ok
22:12 -!- PAVLINUX [n=pavel@ppp85-140-133-188.pppoe.mtu-net.ru] has quit ["using sirc version 2.211+KSIRC/1.3.12"]
22:12 < mtr> sepp_: ?
22:13 < sepp_> needs at least -march=i486 ...
22:13 < mtr> the 4.2.0 stdc++ headers seem to need a proper "-march=" option
22:13 < mtr> :)
22:13 < sepp_> why are all removed and none appended on a multilib x86-64 build ?
22:16 < kic-> hrm
22:16 < kic-> heres a question
22:16 < kic-> our developers have created our application, as a loadable kernel module
22:17 < kic-> is there a way to put our kernel module in so that it compiles against the kernel source during Build-Target
22:17 < kic-> ?
22:17 < sepp_> yes
22:17 < kic-> or even Build-Pkg linux26
22:17 < kic-> ahh, whew
22:17 < kic-> was gonna say, thats a show stopper for us if we can't do it
22:17 < kic-> :D
22:17 < kic-> wouldn't happen to know how we can do that do ya?
22:17 < kic-> :D
22:17 < sepp_> 3rd party kernel mods are always build during linux*
22:18 -!- sparc-kly__ [n=mubex@adsl-64-237-248-104.prtc.net] has joined #t2
22:18 < sepp_> look at package/network/acx100/
22:19 < sepp_> there are a few examples - ls package/*/*/pkg_linux26_post.conf
22:20 < kic-> ok
22:20 < sepp_> if you are lucky you just need to append a makeopt
22:20 < sepp_> so that the package knows where the kernel source is
22:20 < mtr> sepp_: the x86-64 march option is set in architecture/x86-64/config.in, should be non-NULL for non-Generic
22:22 < sepp_> :)
22:23 < sepp_> i was too lazy to select something, just wanted to see if 1-perl really fails with gdbm ...
22:24 < kic-> :D
22:24 < mtr> :)
22:24 < kic-> did it
22:24 < kic-> ?
22:25 < sepp_> never reached that 1-perl this way
22:25 < sepp_> lol
22:25 < kic-> ah
22:25 < kic-> :D
22:26 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has joined #t2
22:28 < sepp_> now i start from 0 ...
22:29 < sepp_> maybe it should not default to non working generic in the 1st place
22:32 < kic-> :P
22:35 -!- sparc-kly_ [n=mubex@adsl-64-237-248-104.prtc.net] has quit [Read error: 110 (Connection timed out)]
23:17 -!- roman [n=roman@85-127-134-3.dynamic.adsl-line.inode.at] has joined #t2
23:29 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has quit ["ChatZilla 0.9.78.1 [Firefox 2.0.0.4/2007051502]"]
23:48 < sepp_> == 06/20/07 23:45:01 =[1]=> Finished building package perl.
23:49 < sepp_> ^- with minimal-xorg on x86-64 (without changes)
--- Log closed Thu Jun 21 00:00:17 2007