T2 IRC Log: 2008-10-24
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 Fri Oct 24 00:00:19 2008
01:27 -!- mpp [n=user@i5387553E.versanet.de] has quit ["good night - good fight"]
03:35 -!- TheBeast [n=id@picard.blurgle.ca] has joined #t2
03:35 -!- TheBeast [n=id@picard.blurgle.ca] has left #t2 []
09:52 < rxr> re
09:52 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has joined #t2
10:11 < CIA-4> rene * r30986 /trunk/package/x86/nasm/nasm.desc: * updated nasm (2.04 -> 2.05)
10:29 < CIA-4> rene * r30987 /trunk/package/xorg/xproto/xproto.desc: * updated xproto (7.0.13 -> 7.0.14)
11:06 < CIA-4> aldas * r30988 /trunk/package/multimedia/blender/blender.desc: * updated blender (2.48 -> 2.48a)
11:14 < CIA-4> rene * r30989 /trunk/package/gnome2/file-roller/file-roller.desc: * updated file-roller (2.24.0 -> 2.24.1)
11:14 < CIA-4> rene * r30990 /trunk/package/graphic/asymptote/asymptote.desc: * updated asymptote (1.46 -> 1.47)
11:14 < CIA-4> rene * r30991 /trunk/package/develop/mono/mono.desc: * updated mono (2.0 -> 2.0.1)
11:19 < CIA-4> rene * r30992 /trunk/package/network/tftp-hpa/tftp-hpa.desc: * updated tftp-hpa (0.48 -> 0.49)
11:19 < CIA-4> rene * r30993 /trunk/package/develop/clisp/clisp.desc: * updated clisp (2.46 -> 2.47)
12:09 < CIA-4> mtr * r30994 /trunk/package/gnome2/enchant/enchant.desc: * changed enchant build priority (153.900 -> 153.100), required by gtkhtml
12:11 < CIA-4> mtr * r30995 /trunk/package/develop/nspr/nspr.conf: * fixed nspr pkgconfig file: copy&paste led to wrong includedir
12:27 < koan> why does emacs get built when I build qingy?
12:28 < rxr> no idea :-)
12:28 < rxr> ah - you mean as dependency ?
12:29 < koan> yes
12:29 < koan> seems strange, I don't need an editor for running a login manager :-)
12:29 < rxr> build/reference-8.0-trunk-reference-x86-pentiumpro/var/adm/dep-debug/qingy
12:29 < rxr> ^ - from this dep debug output the package really runs xemacs during the build
12:30 < rxr> from qingy.log:
12:30 < rxr> checking for emacs... no
12:30 < rxr> checking for xemacs... xemacs
12:30 < rxr> checking where .elc files should go... ${datadir}/emacs/site-lisp
12:30 < rxr> and then it's never used again it looks
12:30 < rxr> maybe they copied their configure script from another project and have this stray thing in ther e?
12:30 < koan> hm
12:30 < koan> could be
12:30 < rxr> at least the t2 auto-dep is correct for what the package does ...
12:31 < rxr> guess it would also build with Emerge-Pkg -deps=none qingy :-)
12:31 < koan> ok, then I'll bother the qingy peoplke :)
12:31 < koan> -k
12:31 < rxr> yeah - we could also patch this away in the configure on our own and send upstream
12:36 < koan> ok I sent an email to the developer with a question for clarification
12:37 -!- mpp [n=user@i53875697.versanet.de] has joined #t2
12:52 -!- Baldzius [n=Baldzius@87.198.192.226] has joined #t2
12:52 < Baldzius> moin
13:02 < mpp> hey Baldzius
13:02 < mpp> moinsen
13:02 < Baldzius> hey mpp
13:07 -!- mqueiros_ [n=mqueiros@217.70.71.152] has quit [Client Quit]
13:15 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has quit [Read error: 110 (Connection timed out)]
14:16 -!- hwinkel [n=hwinkel@213.187.75.85] has joined #t2
14:37 -!- hwinkel [n=hwinkel@213.187.75.85] has quit [Read error: 113 (No route to host)]
14:55 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has joined #t2
15:02 -!- hwinkel1 [n=hwinkel@port-87-193-170-219.static.qsc.de] has joined #t2
15:03 -!- DuckFault [n=DuckFaul@rrcs-71-43-244-114.se.biz.rr.com] has joined #t2
15:04 -!- DuckFault [n=DuckFaul@rrcs-71-43-244-114.se.biz.rr.com] has left #t2 []
15:17 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has quit [Read error: 110 (Connection timed out)]
15:22 < CIA-4> rene * r30996 /trunk/package/xorg/xf86-input-vmmouse/xf86-input-vmmouse.desc: * updated xf86-input-vmmouse (12.5.2 -> 12.6.1)
15:37 < mpp> gtk+12 package builds outside root dir
15:38 < mpp> anyone else to confirm this ?...
15:42 < mtr> hi mpp
15:43 < mtr> sorry, gtk+12 yesterday build fine here
15:43 < mtr> 311830 23. Okt 05:13 var/adm/logs/5-gtk+12.log
15:48 < rxr> mtr: you cross build again ß
15:48 < rxr> mpp: you cross build again?
15:49 < rxr> (btw. I rarely have gtk+12 in any build anyway due to the shear age and non antialised fonts etc.)
15:49 < rxr> it's only there for very old backward compat. software ...
16:05 < CIA-4> mtr * r30997 /trunk/package/ (x11/mesa/mesa.desc xorg/xorg-server/xorg-server.desc):
16:05 < CIA-4> * fixed mesa and xorg-server priority, build order lbdrm - mesa - xorg-server
16:05 < CIA-4> * changed xorg-server build priority (112.446 -> 112.447)
16:05 < CIA-4> * changed mesa build priority (112.400 -> 112.446)
16:06 < mpp> ah nope
16:06 < mpp> since i have all the hosts on 64bit running no need for crossbuild anymore :-)
16:08 < mtr> rxr: mesa has changed its prefix, its now in /usr/X11R7, maybe we should revert this
16:08 < rxr> hm
16:08 < rxr> indeed some stuff might not check carefully and fail to find libGL
16:08 < mtr> its causing some pkgs to fail, like those linking with qt4-opengl (kdebindings)
16:09 < mtr> or gpac
16:09 < rxr> yes, the old mesa.conf has some symlink hackery
16:09 < mtr> they do not have the right include path
16:09 < rxr> we had it in both for backward compat in usr and usr/X11 and symlinks from one place to the other
16:09 < mtr> GL/gl.h nout found
16:10 < rxr> maybe restore the .conf and reuse the former install hook as "symlink adding" only hook
16:10 < rxr> btw. I carefully hacked mesa to cross build just some months ago
16:10 < rxr> I have not yet checked if the new autoconfiscated mesa still cross build :-(((
16:12 < mtr> ah yes, found the symlink hackery in mesa.conf
16:14 < CIA-4> aldas * r30998 /trunk/package/develop/subversion/subversion.desc: * updated subversion (1.5.3 -> 1.5.4)
16:31 < rxr> or we try to just have it in /usr - probably that works for (nearly?) all stuff today
16:36 < mtr> the symlinks are fine
16:36 < mtr> and quite useful when installing 3rd party GL libraries (nvidia, ati, eg)
16:38 < mtr> we might restore the symlinks with xcfgt2 if the system gpu changes ... ;-)
16:43 < rxr> http://images.appleinsider.com/macminifarm-081023.jpg
16:43 < rxr> (no it's not our office *g*)
16:45 < mpp> nice one :-)
16:45 < mpp> wheres the mini warp fusion reactor hiding ?
16:45 < mpp> :-O
16:45 < mpp> all your electricity are belong to us !
16:50 < rxr> well - the mac mini is laptop hardware inside
16:50 < rxr> probably quite efficient server cabinte you can create with it
16:51 < rxr> claimed to be from this company: http://www.macminicolo.net/
16:51 < rxr> macmini based hosting :-)
16:51 < mpp> btw rxr
16:52 < mpp> had an idea for a nice cpu benchmark
16:52 < rxr> http://www.macminicolo.net/facility.html
16:52 < mpp> i compiled povray37 wich has smp built in
16:52 < mpp> it also streeses fpu and stuff
16:53 < mpp> was courious how your 8way xeon performs
16:53 < rxr> I could check, yes
16:53 < mpp> if interested i'll post it as package - it needs an older version of boost
16:54 < mpp> but i managed to have it running from a single source directory without messing system libs
16:54 < mpp> just out of curiosity
16:56 < koan> hm how do you get german keyboard layout in X? my regular xorg.conf voodoo doesn't work
16:56 < rxr> you need the new xkeyboard-config
16:56 < rxr> and then it is just as it was before
16:56 < rxr> we should probably svn rm the former xkbdata to avoid confusion
16:56 < rxr> looks dead to me
16:57 < koan> I have xkeyboard-config
16:57 < rxr> then it should just work
16:57 < koan> can you pastebin your xorg.conf then?
16:57 < rxr> the t2 default one does work
16:57 < koan> hm
16:57 < rxr> maybe you have something about "no xkb" in the xorg log?
16:57 < rxr> /var/log/X.0.log or so
16:58 < rxr> what does setxkmap de # say ?
16:58 < rxr> xkbmap even
16:58 < rxr> a customer had to noitice however, that the swiss variants changed
16:58 < rxr> if you are "over there"
16:58 < rxr> setxkmap ch -variant de ; setxkbmap ch -variant fr # it's now
16:59 < koan> it's with the kbd driver?
16:59 < rxr> yes, I use the old style drivers
16:59 < rxr> the new hal dbus xevent stuff is too messy for me
16:59 < rxr> breaks more often then it works ...
17:01 < koan> ok it works (for be, I asked about de because I knew you were german and had the same need for non-us keyboard)
17:01 < koan> but
17:02 < koan> if I push alt gr X seems to handle it like an enter button
17:02 < koan> so I can't input | etc
17:03 < rxr> maybe a typo in the new (XML?) data
17:03 < rxr> is that the only issue you have?
17:03 < rxr> if so then it's probably some upstream typo
17:03 < koan> yes it seems so
17:03 < koan> and Xorx.0.log has an error message
17:03 < koan> (EE) XKB: Couldn't open rules file /usr/X11R7/share/X11/xkb/rules/base
17:03 < rxr> all other "alt gr" combos work ?
17:03 < rxr> ok - that's not good
17:03 < koan> euh no
17:04 < rxr> how old is your xkeyboard-config emerge ?
17:04 < koan> none, "alt gr" = "enter" here
17:04 < koan> two days ago
17:04 < rxr> maybe try emering the xkeyboarc-config package
17:04 < rxr> strange
17:04 < rxr> we fixed this rules/base thing some 4 weeks ago
17:04 < rxr> you are on trunk or 7.0 ?
17:04 < koan> trunk
17:04 < rxr> hm - should work, strange
17:05 < rxr> ah - ah
17:05 < rxr> maybe you got an xorg.conf with this base while' we changed it back to xorg
17:05 < koan> hm
17:05 < rxr> is this "/base" string in your xorg.conf ?
17:05 < rxr> try make it "/xorg" if it is
17:06 < koan> no,I have rules xorg, model pc105 and layout be
17:06 < rxr> all strange
17:07 < rxr> apparently I do not specify fules
17:07 < rxr> Section "InputDevice"
17:07 < rxr> Identifier "Keyboard1"
17:07 < rxr> Driver "keyboard"
17:07 < rxr> Option "AutoRepeat" "250 30"
17:07 < rxr> Option "XkbModel" "pc105"
17:07 < rxr> Option "XkbLayout" "de"
17:07 < rxr> EndSection
17:08 < rxr> but this is on some "mixed as usualy, over the yeares merged together" laptop
17:08 < rxr> but a trunk build worked just some days ago in some other project which re-based to trunk's xorg:HEAD
17:09 < koan> still the same
17:09 < koan> but I have in Xorg.0.log an xinput message, it seems he finds the keyboard two times: one time from my xorg.conf and one time from hal?
17:09 < rxr> maybe that's the culprit
17:10 < koan> http://www.vervloesem.eu/temp/Xorg.0.log
17:10 < koan> at the bottom
17:11 < rxr> do you have
17:11 < rxr> xkeyboard-config: usr/X11R7/share/X11/xkb/rules/base
17:11 < rxr> xkeyboard-config: usr/X11R7/share/X11/xkb/rules/xorg
17:12 < koan> I have no base...
17:12 < rxr> hm
17:12 < koan> only xorg
17:12 < rxr> maybe try svn up ; mine -rf xkeyboard-config ; Emerge-Pkg xkeyboard-config just to be sure
17:12 < rxr> my (I hope t2:HEAD) system has base with xorg being a symlink to base
17:12 < koan> ok
17:18 < koan> ok now I have base and xorg
17:19 < koan> let's try again :-)
17:19 < rxr> maybe we fixed it not soo long ago as I remember ?
17:19 < koan> hm now the layout is querty here, also in xdm
17:20 < rxr> r30310 | mtr | 2008-09-02 21:50:54 +0200 (Tue, 02 Sep 2008) | 2 lines
17:20 < rxr> * fixed xkeyboard-config to install a default rule set "xorg"
17:20 < koan> but no errors in Xorg.0.log
17:20 < rxr> you mean us querty ?
17:21 < koan> I have an azerty keyboard
17:21 < koan> but X chooses querty now
17:21 < rxr> maybe it needs some layout supplied for the auto matched device?
17:21 < koan> he still seems to find the keyboard two times, one time from the xorg.conf and one time by evdev/hal?
17:22 < rxr> yes, this hal stuff is new :-(
17:22 < rxr> I have not yet really looked at it
17:22 < koan> how can I suppress loading the evdev module?
17:22 < rxr> no idea .-(
17:22 < koan> it's not in xorg.conf, but it's still loaded
17:22 < rxr> I would query google, llikewise
17:23 < koan> so then I would have to write a hal rule?
17:24 < rxr> I hope not
17:24 < rxr> I hope there is some evdev or global flag to tell xorg not to auto-config stuff
17:25 < koan> I have done mine -r xf86-input-evdev
17:25 < rxr> heh - that helps ?
17:25 < koan> I have a warning in Xorg.0.log now, but I have back azerty layout
17:25 < koan> but still not working alt gr
17:25 < rxr> cool *g*
17:25 < rxr> :-(
17:26 < koan> I'll look into that, maybe something wrong with the keycodes I can fix with xkbindkeys or so
17:28 < koan> hm do you get Alt_R as keysym for your alt gr in xev?
17:28 < rxr> or in xkeyboard-config itself
17:28 < rxr> I have apple keyboard withotu two altgr
17:29 < rxr> and a t2 specific kernel tweak to remap stupid apple decided keycodes ...
17:29 < koan> :-)
17:29 < rxr> but my alt gr works the way I want .-)
17:30 < rxr> appears I get ISO_Level3_Shift for my altgr key
17:30 < koan> yes here on my mac too
17:30 < rxr> ah :-)
17:30 < koan> but on the T2 system (AMD64) Alt_R
17:31 < rxr> ok - so a normal pc like keyboard your alt does not yet work?
17:31 < rxr> or an apple keyboard ?
17:31 < koan> it's a normal pc (laptop) keyboard for which alt gr isn't working
17:33 < rxr> ah - ok - so it should not be my apple keyboard patch present in t2 :-)
17:33 < koan> no no
17:33 < rxr> have you checked if other people report something like this in other distributions bug trackers ?
17:33 < koan> I'm looking for it now
17:58 < CIA-4> mtr * r30999 /trunk/package/x11/mesa/mesa.conf: * reactived mesa compatibility symlinks in /usr/{include,lib}
17:58 < mtr> one more to go ...
18:00 < rxr> we're so quick these days
18:00 < rxr> but well - with all the many packages all the auto-notified updates are quite contributing to it
18:01 < mtr> yep, but I have to leave now
18:01 < rxr> cu - have fun
18:01 < rxr> I have to leave in a sec, too
18:02 < mtr> even if xine-lib and kdebindings build error annoys me much
18:02 < rxr> have fun later :-)
18:02 < mtr> ffmpeg update introduces a lot of regressions, the folks changed include folder names
18:02 < mtr> cu :)
18:03 < rxr> cu
18:08 < Baldzius> hm, xine-lib was fine for me
18:08 < rxr> maybe due to a previous build ?
18:09 < rxr> (of ffmpeg and the old files still being in place)
18:09 < rxr> so - have to leave, quickly - cu then
18:10 < Baldzius> cu
18:10 < Baldzius> maybe, i'll check during the weekend
18:10 < Baldzius> with a fresh build
19:04 -!- mpp [n=user@i53875697.versanet.de] has quit ["good night - good fight"]
19:07 -!- hwinkel1 [n=hwinkel@port-87-193-170-219.static.qsc.de] has quit [Read error: 110 (Connection timed out)]
19:20 -!- axionix [n=axion@cpe-67-242-94-6.nycap.res.rr.com] has quit [Read error: 60 (Operation timed out)]
19:20 -!- axionix [n=axion@cpe-67-242-94-6.nycap.res.rr.com] has joined #t2
20:05 -!- Cyda [n=cyda@c-66-229-118-244.hsd1.fl.comcast.net] has joined #t2
21:07 -!- Baldzius [n=Baldzius@87.198.192.226] has quit [Remote closed the connection]
21:46 -!- gw [n=gw@2001:44b8:62:7f0:21b:63ff:fe01:f18d] has quit [Read error: 113 (No route to host)]
23:09 -!- gw [n=gw@2001:44b8:62:7f0:21b:63ff:fe01:f18d] has joined #t2
--- Log closed Sat Oct 25 00:00:25 2008
01:27 -!- mpp [n=user@i5387553E.versanet.de] has quit ["good night - good fight"]
03:35 -!- TheBeast [n=id@picard.blurgle.ca] has joined #t2
03:35 -!- TheBeast [n=id@picard.blurgle.ca] has left #t2 []
09:52 < rxr> re
09:52 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has joined #t2
10:11 < CIA-4> rene * r30986 /trunk/package/x86/nasm/nasm.desc: * updated nasm (2.04 -> 2.05)
10:29 < CIA-4> rene * r30987 /trunk/package/xorg/xproto/xproto.desc: * updated xproto (7.0.13 -> 7.0.14)
11:06 < CIA-4> aldas * r30988 /trunk/package/multimedia/blender/blender.desc: * updated blender (2.48 -> 2.48a)
11:14 < CIA-4> rene * r30989 /trunk/package/gnome2/file-roller/file-roller.desc: * updated file-roller (2.24.0 -> 2.24.1)
11:14 < CIA-4> rene * r30990 /trunk/package/graphic/asymptote/asymptote.desc: * updated asymptote (1.46 -> 1.47)
11:14 < CIA-4> rene * r30991 /trunk/package/develop/mono/mono.desc: * updated mono (2.0 -> 2.0.1)
11:19 < CIA-4> rene * r30992 /trunk/package/network/tftp-hpa/tftp-hpa.desc: * updated tftp-hpa (0.48 -> 0.49)
11:19 < CIA-4> rene * r30993 /trunk/package/develop/clisp/clisp.desc: * updated clisp (2.46 -> 2.47)
12:09 < CIA-4> mtr * r30994 /trunk/package/gnome2/enchant/enchant.desc: * changed enchant build priority (153.900 -> 153.100), required by gtkhtml
12:11 < CIA-4> mtr * r30995 /trunk/package/develop/nspr/nspr.conf: * fixed nspr pkgconfig file: copy&paste led to wrong includedir
12:27 < koan> why does emacs get built when I build qingy?
12:28 < rxr> no idea :-)
12:28 < rxr> ah - you mean as dependency ?
12:29 < koan> yes
12:29 < koan> seems strange, I don't need an editor for running a login manager :-)
12:29 < rxr> build/reference-8.0-trunk-reference-x86-pentiumpro/var/adm/dep-debug/qingy
12:29 < rxr> ^ - from this dep debug output the package really runs xemacs during the build
12:30 < rxr> from qingy.log:
12:30 < rxr> checking for emacs... no
12:30 < rxr> checking for xemacs... xemacs
12:30 < rxr> checking where .elc files should go... ${datadir}/emacs/site-lisp
12:30 < rxr> and then it's never used again it looks
12:30 < rxr> maybe they copied their configure script from another project and have this stray thing in ther e?
12:30 < koan> hm
12:30 < koan> could be
12:30 < rxr> at least the t2 auto-dep is correct for what the package does ...
12:31 < rxr> guess it would also build with Emerge-Pkg -deps=none qingy :-)
12:31 < koan> ok, then I'll bother the qingy peoplke :)
12:31 < koan> -k
12:31 < rxr> yeah - we could also patch this away in the configure on our own and send upstream
12:36 < koan> ok I sent an email to the developer with a question for clarification
12:37 -!- mpp [n=user@i53875697.versanet.de] has joined #t2
12:52 -!- Baldzius [n=Baldzius@87.198.192.226] has joined #t2
12:52 < Baldzius> moin
13:02 < mpp> hey Baldzius
13:02 < mpp> moinsen
13:02 < Baldzius> hey mpp
13:07 -!- mqueiros_ [n=mqueiros@217.70.71.152] has quit [Client Quit]
13:15 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has quit [Read error: 110 (Connection timed out)]
14:16 -!- hwinkel [n=hwinkel@213.187.75.85] has joined #t2
14:37 -!- hwinkel [n=hwinkel@213.187.75.85] has quit [Read error: 113 (No route to host)]
14:55 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has joined #t2
15:02 -!- hwinkel1 [n=hwinkel@port-87-193-170-219.static.qsc.de] has joined #t2
15:03 -!- DuckFault [n=DuckFaul@rrcs-71-43-244-114.se.biz.rr.com] has joined #t2
15:04 -!- DuckFault [n=DuckFaul@rrcs-71-43-244-114.se.biz.rr.com] has left #t2 []
15:17 -!- hwinkel [n=hwinkel@port-87-193-170-219.static.qsc.de] has quit [Read error: 110 (Connection timed out)]
15:22 < CIA-4> rene * r30996 /trunk/package/xorg/xf86-input-vmmouse/xf86-input-vmmouse.desc: * updated xf86-input-vmmouse (12.5.2 -> 12.6.1)
15:37 < mpp> gtk+12 package builds outside root dir
15:38 < mpp> anyone else to confirm this ?...
15:42 < mtr> hi mpp
15:43 < mtr> sorry, gtk+12 yesterday build fine here
15:43 < mtr> 311830 23. Okt 05:13 var/adm/logs/5-gtk+12.log
15:48 < rxr> mtr: you cross build again ß
15:48 < rxr> mpp: you cross build again?
15:49 < rxr> (btw. I rarely have gtk+12 in any build anyway due to the shear age and non antialised fonts etc.)
15:49 < rxr> it's only there for very old backward compat. software ...
16:05 < CIA-4> mtr * r30997 /trunk/package/ (x11/mesa/mesa.desc xorg/xorg-server/xorg-server.desc):
16:05 < CIA-4> * fixed mesa and xorg-server priority, build order lbdrm - mesa - xorg-server
16:05 < CIA-4> * changed xorg-server build priority (112.446 -> 112.447)
16:05 < CIA-4> * changed mesa build priority (112.400 -> 112.446)
16:06 < mpp> ah nope
16:06 < mpp> since i have all the hosts on 64bit running no need for crossbuild anymore :-)
16:08 < mtr> rxr: mesa has changed its prefix, its now in /usr/X11R7, maybe we should revert this
16:08 < rxr> hm
16:08 < rxr> indeed some stuff might not check carefully and fail to find libGL
16:08 < mtr> its causing some pkgs to fail, like those linking with qt4-opengl (kdebindings)
16:09 < mtr> or gpac
16:09 < rxr> yes, the old mesa.conf has some symlink hackery
16:09 < mtr> they do not have the right include path
16:09 < rxr> we had it in both for backward compat in usr and usr/X11 and symlinks from one place to the other
16:09 < mtr> GL/gl.h nout found
16:10 < rxr> maybe restore the .conf and reuse the former install hook as "symlink adding" only hook
16:10 < rxr> btw. I carefully hacked mesa to cross build just some months ago
16:10 < rxr> I have not yet checked if the new autoconfiscated mesa still cross build :-(((
16:12 < mtr> ah yes, found the symlink hackery in mesa.conf
16:14 < CIA-4> aldas * r30998 /trunk/package/develop/subversion/subversion.desc: * updated subversion (1.5.3 -> 1.5.4)
16:31 < rxr> or we try to just have it in /usr - probably that works for (nearly?) all stuff today
16:36 < mtr> the symlinks are fine
16:36 < mtr> and quite useful when installing 3rd party GL libraries (nvidia, ati, eg)
16:38 < mtr> we might restore the symlinks with xcfgt2 if the system gpu changes ... ;-)
16:43 < rxr> http://images.appleinsider.com/macminifarm-081023.jpg
16:43 < rxr> (no it's not our office *g*)
16:45 < mpp> nice one :-)
16:45 < mpp> wheres the mini warp fusion reactor hiding ?
16:45 < mpp> :-O
16:45 < mpp> all your electricity are belong to us !
16:50 < rxr> well - the mac mini is laptop hardware inside
16:50 < rxr> probably quite efficient server cabinte you can create with it
16:51 < rxr> claimed to be from this company: http://www.macminicolo.net/
16:51 < rxr> macmini based hosting :-)
16:51 < mpp> btw rxr
16:52 < mpp> had an idea for a nice cpu benchmark
16:52 < rxr> http://www.macminicolo.net/facility.html
16:52 < mpp> i compiled povray37 wich has smp built in
16:52 < mpp> it also streeses fpu and stuff
16:53 < mpp> was courious how your 8way xeon performs
16:53 < rxr> I could check, yes
16:53 < mpp> if interested i'll post it as package - it needs an older version of boost
16:54 < mpp> but i managed to have it running from a single source directory without messing system libs
16:54 < mpp> just out of curiosity
16:56 < koan> hm how do you get german keyboard layout in X? my regular xorg.conf voodoo doesn't work
16:56 < rxr> you need the new xkeyboard-config
16:56 < rxr> and then it is just as it was before
16:56 < rxr> we should probably svn rm the former xkbdata to avoid confusion
16:56 < rxr> looks dead to me
16:57 < koan> I have xkeyboard-config
16:57 < rxr> then it should just work
16:57 < koan> can you pastebin your xorg.conf then?
16:57 < rxr> the t2 default one does work
16:57 < koan> hm
16:57 < rxr> maybe you have something about "no xkb" in the xorg log?
16:57 < rxr> /var/log/X.0.log or so
16:58 < rxr> what does setxkmap de # say ?
16:58 < rxr> xkbmap even
16:58 < rxr> a customer had to noitice however, that the swiss variants changed
16:58 < rxr> if you are "over there"
16:58 < rxr> setxkmap ch -variant de ; setxkbmap ch -variant fr # it's now
16:59 < koan> it's with the kbd driver?
16:59 < rxr> yes, I use the old style drivers
16:59 < rxr> the new hal dbus xevent stuff is too messy for me
16:59 < rxr> breaks more often then it works ...
17:01 < koan> ok it works (for be, I asked about de because I knew you were german and had the same need for non-us keyboard)
17:01 < koan> but
17:02 < koan> if I push alt gr X seems to handle it like an enter button
17:02 < koan> so I can't input | etc
17:03 < rxr> maybe a typo in the new (XML?) data
17:03 < rxr> is that the only issue you have?
17:03 < rxr> if so then it's probably some upstream typo
17:03 < koan> yes it seems so
17:03 < koan> and Xorx.0.log has an error message
17:03 < koan> (EE) XKB: Couldn't open rules file /usr/X11R7/share/X11/xkb/rules/base
17:03 < rxr> all other "alt gr" combos work ?
17:03 < rxr> ok - that's not good
17:03 < koan> euh no
17:04 < rxr> how old is your xkeyboard-config emerge ?
17:04 < koan> none, "alt gr" = "enter" here
17:04 < koan> two days ago
17:04 < rxr> maybe try emering the xkeyboarc-config package
17:04 < rxr> strange
17:04 < rxr> we fixed this rules/base thing some 4 weeks ago
17:04 < rxr> you are on trunk or 7.0 ?
17:04 < koan> trunk
17:04 < rxr> hm - should work, strange
17:05 < rxr> ah - ah
17:05 < rxr> maybe you got an xorg.conf with this base while' we changed it back to xorg
17:05 < koan> hm
17:05 < rxr> is this "/base" string in your xorg.conf ?
17:05 < rxr> try make it "/xorg" if it is
17:06 < koan> no,I have rules xorg, model pc105 and layout be
17:06 < rxr> all strange
17:07 < rxr> apparently I do not specify fules
17:07 < rxr> Section "InputDevice"
17:07 < rxr> Identifier "Keyboard1"
17:07 < rxr> Driver "keyboard"
17:07 < rxr> Option "AutoRepeat" "250 30"
17:07 < rxr> Option "XkbModel" "pc105"
17:07 < rxr> Option "XkbLayout" "de"
17:07 < rxr> EndSection
17:08 < rxr> but this is on some "mixed as usualy, over the yeares merged together" laptop
17:08 < rxr> but a trunk build worked just some days ago in some other project which re-based to trunk's xorg:HEAD
17:09 < koan> still the same
17:09 < koan> but I have in Xorg.0.log an xinput message, it seems he finds the keyboard two times: one time from my xorg.conf and one time from hal?
17:09 < rxr> maybe that's the culprit
17:10 < koan> http://www.vervloesem.eu/temp/Xorg.0.log
17:10 < koan> at the bottom
17:11 < rxr> do you have
17:11 < rxr> xkeyboard-config: usr/X11R7/share/X11/xkb/rules/base
17:11 < rxr> xkeyboard-config: usr/X11R7/share/X11/xkb/rules/xorg
17:12 < koan> I have no base...
17:12 < rxr> hm
17:12 < koan> only xorg
17:12 < rxr> maybe try svn up ; mine -rf xkeyboard-config ; Emerge-Pkg xkeyboard-config just to be sure
17:12 < rxr> my (I hope t2:HEAD) system has base with xorg being a symlink to base
17:12 < koan> ok
17:18 < koan> ok now I have base and xorg
17:19 < koan> let's try again :-)
17:19 < rxr> maybe we fixed it not soo long ago as I remember ?
17:19 < koan> hm now the layout is querty here, also in xdm
17:20 < rxr> r30310 | mtr | 2008-09-02 21:50:54 +0200 (Tue, 02 Sep 2008) | 2 lines
17:20 < rxr> * fixed xkeyboard-config to install a default rule set "xorg"
17:20 < koan> but no errors in Xorg.0.log
17:20 < rxr> you mean us querty ?
17:21 < koan> I have an azerty keyboard
17:21 < koan> but X chooses querty now
17:21 < rxr> maybe it needs some layout supplied for the auto matched device?
17:21 < koan> he still seems to find the keyboard two times, one time from the xorg.conf and one time by evdev/hal?
17:22 < rxr> yes, this hal stuff is new :-(
17:22 < rxr> I have not yet really looked at it
17:22 < koan> how can I suppress loading the evdev module?
17:22 < rxr> no idea .-(
17:22 < koan> it's not in xorg.conf, but it's still loaded
17:22 < rxr> I would query google, llikewise
17:23 < koan> so then I would have to write a hal rule?
17:24 < rxr> I hope not
17:24 < rxr> I hope there is some evdev or global flag to tell xorg not to auto-config stuff
17:25 < koan> I have done mine -r xf86-input-evdev
17:25 < rxr> heh - that helps ?
17:25 < koan> I have a warning in Xorg.0.log now, but I have back azerty layout
17:25 < koan> but still not working alt gr
17:25 < rxr> cool *g*
17:25 < rxr> :-(
17:26 < koan> I'll look into that, maybe something wrong with the keycodes I can fix with xkbindkeys or so
17:28 < koan> hm do you get Alt_R as keysym for your alt gr in xev?
17:28 < rxr> or in xkeyboard-config itself
17:28 < rxr> I have apple keyboard withotu two altgr
17:29 < rxr> and a t2 specific kernel tweak to remap stupid apple decided keycodes ...
17:29 < koan> :-)
17:29 < rxr> but my alt gr works the way I want .-)
17:30 < rxr> appears I get ISO_Level3_Shift for my altgr key
17:30 < koan> yes here on my mac too
17:30 < rxr> ah :-)
17:30 < koan> but on the T2 system (AMD64) Alt_R
17:31 < rxr> ok - so a normal pc like keyboard your alt does not yet work?
17:31 < rxr> or an apple keyboard ?
17:31 < koan> it's a normal pc (laptop) keyboard for which alt gr isn't working
17:33 < rxr> ah - ok - so it should not be my apple keyboard patch present in t2 :-)
17:33 < koan> no no
17:33 < rxr> have you checked if other people report something like this in other distributions bug trackers ?
17:33 < koan> I'm looking for it now
17:58 < CIA-4> mtr * r30999 /trunk/package/x11/mesa/mesa.conf: * reactived mesa compatibility symlinks in /usr/{include,lib}
17:58 < mtr> one more to go ...
18:00 < rxr> we're so quick these days
18:00 < rxr> but well - with all the many packages all the auto-notified updates are quite contributing to it
18:01 < mtr> yep, but I have to leave now
18:01 < rxr> cu - have fun
18:01 < rxr> I have to leave in a sec, too
18:02 < mtr> even if xine-lib and kdebindings build error annoys me much
18:02 < rxr> have fun later :-)
18:02 < mtr> ffmpeg update introduces a lot of regressions, the folks changed include folder names
18:02 < mtr> cu :)
18:03 < rxr> cu
18:08 < Baldzius> hm, xine-lib was fine for me
18:08 < rxr> maybe due to a previous build ?
18:09 < rxr> (of ffmpeg and the old files still being in place)
18:09 < rxr> so - have to leave, quickly - cu then
18:10 < Baldzius> cu
18:10 < Baldzius> maybe, i'll check during the weekend
18:10 < Baldzius> with a fresh build
19:04 -!- mpp [n=user@i53875697.versanet.de] has quit ["good night - good fight"]
19:07 -!- hwinkel1 [n=hwinkel@port-87-193-170-219.static.qsc.de] has quit [Read error: 110 (Connection timed out)]
19:20 -!- axionix [n=axion@cpe-67-242-94-6.nycap.res.rr.com] has quit [Read error: 60 (Operation timed out)]
19:20 -!- axionix [n=axion@cpe-67-242-94-6.nycap.res.rr.com] has joined #t2
20:05 -!- Cyda [n=cyda@c-66-229-118-244.hsd1.fl.comcast.net] has joined #t2
21:07 -!- Baldzius [n=Baldzius@87.198.192.226] has quit [Remote closed the connection]
21:46 -!- gw [n=gw@2001:44b8:62:7f0:21b:63ff:fe01:f18d] has quit [Read error: 113 (No route to host)]
23:09 -!- gw [n=gw@2001:44b8:62:7f0:21b:63ff:fe01:f18d] has joined #t2
--- Log closed Sat Oct 25 00:00:25 2008