--- Log opened Tue Nov 25 00:00:01 2008
00:05 -!- tri [n=tri@p4FCF3D01.dip0.t-ipconnect.de] has left #t2 []
01:08 -!- wtracy [n=wtracy@natusers1.stennerglenstudents.com] has joined #t2
02:26 -!- wtracy [n=wtracy@natusers1.stennerglenstudents.com] has quit [Remote closed the connection]
04:34 -!- wtracy [n=wtracy@natusers1.stennerglenstudents.com] has joined #t2
07:46 -!- wtracy [n=wtracy@natusers1.stennerglenstudents.com] has quit [Remote closed the connection]
08:43 < rxr> re
08:43 [Users #t2]
08:43 [@ChanServ] [ Cyda ] [ koan] [ mqueiros] [ rxr    ] [ TobiX] 
08:43 [ CIA-4   ] [ dsoul] [ LMJ ] [ mtr     ] [ Stealth] 
08:43 -!- Irssi: #t2: Total of 11 nicks [1 ops, 0 halfops, 0 voices, 10 normal]
10:27 -!- mpp [n=mpp@i53877F68.versanet.de] has joined #t2
10:27 < mpp> moinsen
10:32 < rxr> moin moin mpp 
10:33 < mpp> moinmoin
10:34 < mpp> got new on one of my probs with building and segfaulting
10:34 < mpp> it is all related to amd architecture
10:35 < mpp> and some motherboards can't handle recent amd phenom cpu's when under heavy multicore load
10:35 < mpp> amd cpu that is
10:35 < mpp> tweaking around bios settings is necesary to have a stable system 
10:36 < mpp> heavy gcc seems to trouble those cpu's especially the multicore ones
10:36 < mpp> p4 xeon core-duo seems to do just fine on the other hand
10:36 < mpp> :-(
10:36 < rxr> :-(
10:37 < mpp> just letting know others with segault and compile errors
10:37 < mpp> these are clearly not related to the kernel , t2  or gcc.
10:37 < rxr> good to know
10:37 < mpp> took me some days/weeks to figure this s**t out
10:38 < mpp> the only segfault repr on my side with high certainty is the gcc segfault on conftest - other than that is a hardware issue
10:38 < rxr> though after years on Intel (while I historically was a "avoid this evil Intel monopoly and support the tinier /we are not evil/ AMD fellow" I was just about to get a decent AMD system for testing into the office :-(
10:39 < mpp> turns out t2 is a good hw stress tool
10:39 < mpp> you should have an amd system
10:39 < rxr> yes, it bootstrapping the whole system always was
10:39 < mpp> as many people also have them
10:39 < rxr> that where funny times back in the days I started with ROCK Linux on a WinChip 200++ MHz box ...
10:40 < mpp> some mobo are flacky some are not
10:40 < mpp> e.g. one can turn off some cores on amd phenom
10:40 < mpp> helps tracking down bios bugs
10:40 < rxr> yeah - we have too much Apple hardware, this is how all this Intel "crap" sneaked into our company :-(
10:41 < mpp> poor poor rxr :-)
10:41 < mpp> anyways - got a stable amd running now !!! :-)
10:41 < mpp> ill write down bios settings particular troublesome
10:41 < mpp> we could have them attached to the handbook maybe .. ?
10:42 < rxr> bios settings in the handbook ?
10:42 < mpp> k bad idea
10:42 < rxr> but I know a good place
10:42 < mpp> maybe ill just paste them into the irc
10:42 < mpp> like what ?
10:43 < mpp> as irc is logged and googled
10:43 < rxr> http://www.t2-project.org/hardware/
10:43 < rxr> add a hardware entry for each mobo / system you got there and want to take notes
10:43 < rxr> feeding google quite nicely with possible landing pages on http://t2-project.org/ :-)
10:43 < mpp> k ;-) 
10:44 < mpp> pimp my google
10:47 < CIA-4> rene * r31352 /trunk/package/gnome2/glib/glib.desc: * updated glib (2.18.2 -> 2.18.3)
10:47 < CIA-4> rene * r31353 /trunk/package/develop/ctalk/ctalk.desc: * updated ctalk (0.0.87a -> 0.0.88a)
10:47 < CIA-4> rene * r31355 /trunk/package/gnome2/libsoup/libsoup.desc: * updated libsoup (2.24.1 -> 2.24.2)
10:47 < CIA-4> rene * r31354 /trunk/package/gnome2/gvfs/gvfs.desc: * updated gvfs (1.0.2 -> 1.0.3)
10:47 < CIA-4> rene * r31356 /trunk/package/network/crossroads/crossroads.desc: * updated crossroads (2.37 -> 2.38)
10:48 < mpp> i even had cpu soft lock ups
10:48 < mpp> which is really bad **s
10:49 < mpp> k so ill be afk some hours
10:49 < rxr> cu
10:49 < rxr> strange
10:49 -!- mpp is now known as mpp_afk
10:49 < rxr> since when does libidl no longer build
10:49 < rxr> hm
10:52 < koan> rxr: yes I had that problem too some days ago
10:55 < koan> didn't have the time then to investigate it
10:56 < CIA-4> rene * r31357 /trunk/package/graphic/libexif/libexif.desc: * updated libexif (0.6.16 -> 0.6.17)
10:56 < CIA-4> rene * r31358 /trunk/package/gnome2/gnome-settings-daemon/gnome-settings-daemon.desc: * updated gnome-settings-daemon (2.24.0 -> 2.24.1)
10:56 < CIA-4> rene * r31359 /trunk/package/gnome2/gnome-speech/gnome-speech.desc: * updated gnome-speech (0.4.21 -> 0.4.22)
10:57 < CIA-4> rene * r31361 /trunk/package/gnome2/gucharmap/gucharmap.desc: * updated gucharmap (2.24.1.1 -> 2.24.2)
10:57 < CIA-4> rene * r31360 /trunk/package/gnome2/gnome-games/gnome-games.desc: * updated gnome-games (2.24.1.1 -> 2.24.2)
12:07 < CIA-4> rene * r31362 /trunk/package/gnome2/gtk+/gtk+.desc: * updated gtk+ (2.14.4 -> 2.14.5)
12:58 -!- mpp_afk_ [n=mpp@i53876033.versanet.de] has joined #t2
13:14 -!- mpp_afk [n=mpp@i53877F68.versanet.de] has quit [Read error: 110 (Connection timed out)]
14:26 -!- bovnet [n=steve@87-194-185-180.bethere.co.uk] has joined #t2
14:26 < bovnet> where or how does the copying stone.d happen in the build
14:36 < rxr> from a quick grep: package/base/stone/parse-config
14:36 < rxr> there is also misc/tail/ for various other copying / post-processing
14:42 < bovnet> thanks ive got the parse-config
14:43 < bovnet> must be your 2nd suggestions im looking for
14:43 < rxr> the stone files should be copied there IIRC
14:44 < rxr> if pkginstalled stone; then
14:44 < rxr>                 echo "Installing $mfile to /$stoned/ ..."
14:44 < rxr>                 cp -vf $file $root/$stoned/$mfile
14:47 < bovnet> the following in the build console
14:47 < bovnet> -> Copy /etc/fstab.
14:47 < bovnet> -> Copy stone.d.
14:55 < bovnet> ive made a new installer directory in /package/base and added it to the target pkgsel
14:56 < rxr> you mean you do not want to use stone ?
14:56 < bovnet> yes trying to add new installer
14:56 < bovnet> but my package doesnt seem to get called or added
14:57 < bovnet> i expected the build to fail but doesnt
14:58 < rxr> it's not built at all ?
14:58 < rxr> you might have to select it, and maybe deselect stone if you do not want stone at all
15:00 < bovnet> yeah id just left stone in at the moment, id figured as during install its executed by command 'stone'
15:00 < bovnet> i could then opt to try mine
15:26 -!- rconstruct [n=hal9000@bl5-96-45.dsl.telepac.pt] has joined #t2
15:26 < rconstruct> hi
15:26 < rxr> hey rconstruct !
15:26 < rconstruct> hi Rene :)
15:26 < rconstruct> cleaning things, makign diffs, etc... getting everything ready
15:27 < rconstruct> just found a tricky situation regarding openexr
15:27 < rconstruct> there are 2 "versions" of exr
15:27 < rconstruct> both official
15:27 < rconstruct> 1 with CTL (color transformation language)
15:27 < rconstruct> provided by ampas, the other the no-CTL, which is the one we currently use
15:27 < rconstruct> i was thinking in making a config option, to enable/disable CTL
15:28 < rconstruct> in case CTL is enabled, pkg ctl is available to download, and ilmbase + openexr, use the ctl branches/patches
15:28 < rconstruct> i'm not sure if this is doable though
15:29 < rconstruct> hmm, the .desc can just use the [O] . foo/bar/*/*/ctl-config.in
15:29 < rconstruct> and i can do the voodoo here in ctl-config.in
15:30 < rxr> hm
15:30 < rxr> I'll scroll over it and could help to take a look at those more complex ones
15:30 < rconstruct> ok, in the mean time, i already have some things ready
15:31 < rconstruct> going to work in 30m (offset printing house...)
15:31 < rxr> I fear we're burried under patches for days to commit all your stuff :-)
15:31 < rconstruct> no hurries
15:31 < rconstruct> and no worries, my main change is
15:31 < rconstruct> well, littlecms, argyllcms, vips, etc... these 3 pkgs alone
15:31 < rconstruct> provide over 200 binaries
15:31 < rconstruct> there are more...
15:32 < rconstruct> so i'm installing this in /opt/graphics
15:32 < rconstruct> made a menu cfg to choose prefix for these things
15:32 < rconstruct> defaults to /opt/graphics
15:32 < rconstruct> which also means i had to add a small fix to gimp, scribus, etc... for them to be able to locate lcms, since the configure scripts don't use pkg-config to detect lcms
15:33 < rconstruct> i just added a line to their conf, pkginstalled lcms && var_append GCC_WRAPPER_APPEND ' ' "-I$(pkgprefix includedir lcms) -L$(pkgprefix libdir lcms)"
15:33 < rconstruct> and such
15:33 < rconstruct> koffice too, krita can use color management..., the pkgs that needed this small update were gimp, scribus, imagemagick, koffice (for krita), ufraw
15:34 < rconstruct> etc...
15:34 < rxr> I'm curious to that moster diff :-)
15:34 < rconstruct> it basically revolves around this
15:34 < rconstruct> avoiding overcrowding /usr/bin
15:34 < rconstruct> vips, for instance, image editing app, has over 200 binaries
15:34 < rconstruct> has a gui, gtk, called nip2
15:35 < rconstruct> made some new pkgs too, rawstudio, ale, pfstools, pfstmo, pfscalibration, qtpfsgui
15:35 < rconstruct> ivygenerator, fv
15:35 < rconstruct> want me to mail these already?
15:35 < rconstruct> or you prefer a monster tarball ?
15:35 < rconstruct> :)
15:36 < rxr> well - best is single patches for each topic / package
15:36 < rxr> but given your enormous kind of work I would also be willing to sort a moster tarball .-)
15:36 < rconstruct> lol ok
15:36 < rconstruct> i have the diffs in their categories directories, graphics, audio, development, etc..
15:37 < rxr> I'm super relaxed to contributions anyway, like unpackig zips or rars, Linux kernel folk would not do that, just reply come back with a plain/text patch dude :-)
15:37 < rxr> I'm rather worried about lost efford, so maybe just flush all of it so different eyes can take a look on it and get it merged soonish .-)
15:38 < rxr> so multiple peoeple can sort it
15:38 < rconstruct> i can send mail by mail too
15:38 < rconstruct> no problem
15:38 < rxr> but if you take your time to sort it into mergable patches that's of course great, likewise
15:38 < rxr> list is good, indeed
15:38 < rconstruct> just prepare yourself for a daily dosis of +10 patches
15:38 < rconstruct> eheheh
15:39 < rconstruct> everything seems ok so far, my only concern is priorities, i've been checking that
15:39 < rconstruct> it seems ok, but i could use someone rechecking that
15:39 < rconstruct> ok, better to mail 1 by 1, makes it easier to check this
15:40 < rconstruct> i'll leave the CTL issue for later, this will require some well thought planning
15:40 < rconstruct> about the /opt/graphics
15:40 < rconstruct> it's not for all graphics related stuff, i mean
15:40 < rconstruct> gimp, inkscape, etc... use their standard prefixes
15:40 < rconstruct> i just created this for really very specific packages
15:41 < rconstruct> related to manipulation of color profiles, color management, luts
15:41 < rconstruct> the type of stuff you won't normally see in a linux distro (except in studios)
15:46 < rconstruct> you want the diff attached to the mail, or included in the actual body?
15:46 -!- mpp_afk_ [n=mpp@i53876033.versanet.de] has quit ["ChatZilla 0.9.84 [Firefox 3.0.3/2008092510]"]
15:58 < rxr> re
15:58 < rconstruct> wb
15:59 < rconstruct> sent lcms, diff in the mail body, i can send the diffs as attachments, which you prefer?
16:22 -!- DigitalOktay [n=DigitalO@dslb-084-057-229-215.pools.arcor-ip.net] has joined #t2
16:23 < rconstruct> need to go to work, i'll mail them in the body
16:37 < rconstruct> ok, i'll bbl, work
16:37 -!- rconstruct [n=hal9000@bl5-96-45.dsl.telepac.pt] has quit ["Leaving"]
17:08 < bovnet> rxr can now see my package being built so i can work on that
17:08 < bovnet> but still can see where this is called
17:08 < bovnet> -> `- Found usr/lib/libparted-1.8.so.8.
17:08 < bovnet> -> `- Found usr/lib/libreadline.so.5.
17:08 < bovnet> -> Copy /etc/fstab.
17:08 < bovnet> -> Copy stone.d.
17:08 < bovnet> -> Creating links for identical files.
17:10 < bovnet> cant find the copy stone.d to see whats going on
18:02 -!- DigitalOktay [n=DigitalO@dslb-084-057-229-215.pools.arcor-ip.net] has quit []
18:33 -!- CIA-4 [n=CIA@208.69.182.149] has quit [Remote closed the connection]
18:59 -!- d_oilen [n=iulian@89.18.29.87] has joined #t2
18:59 < d_oilen> hello
19:09 < d_oilen> question...
19:10 < d_oilen> in the 7.0 version from svn the kernel version is 2.6.29.x and in the default download tarball 7.0-rc2 is 2.6.21 why?
19:11 < d_oilen> in the 7.0 version from svn the kernel version is 2.6.19.x and in the default download tarball 7.0-rc2 is 2.6.21 why?
19:28 -!- wtracy [n=wtracy@natusers1.stennerglenstudents.com] has joined #t2
19:30 < d_oilen> stone package management does not work on the installed target
19:30 < d_oilen> how to manually install a package?
19:39 < d_oilen> tlx does not exist in 7.0 svn. where to download it?
19:52 -!- d_oilen [n=iulian@89.18.29.87] has quit ["Leaving"]
19:57 -!- mqueiros [n=mqueiros@217.70.71.193] has quit []
19:59 -!- CIA-10 [n=CIA@208.69.182.149] has joined #t2
20:01 -!- wtracy [n=wtracy@natusers1.stennerglenstudents.com] has quit [Remote closed the connection]
21:14 < koan> rxr: libexif doesn't build, there's a patch mismatch
21:22 < koan> is it normal that Eterm is installed in /opt ?
21:22 < koan> (package eterm16)
21:30 -!- tri [n=tri@p4FCF3B1A.dip0.t-ipconnect.de] has joined #t2
21:33 < tri> re
21:40 < rxr> we install kde, gnomei, mozilla*, xfce, enightenment* by default into /opt to have some little more grouping than on other linux systems
21:41 < rxr> so eterm is enlightenment 16 and is in /opt/e16 my default IIRC
21:41 < koan> ok
23:01 -!- tri [n=tri@p4FCF3B1A.dip0.t-ipconnect.de] has left #t2 []
--- Log closed Wed Nov 26 00:00:07 2008