T2 IRC Log: 2010-04-28

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.

« prev | next »

--- Log opened Wed Apr 28 00:00:03 2010
04:21 -!- digitaloktay2 [~arch@dslb-084-056-231-021.pools.arcor-ip.net] has joined #t2
04:25 -!- digitaloktay [~arch@unaffiliated/digitaloktay] has quit [Ping timeout: 245 seconds]
06:47 -!- digitaloktay2 [~arch@dslb-084-056-231-021.pools.arcor-ip.net] has quit [Quit: bin mal langer!]
09:38 < CIA-32> rene * r36166 /trunk/package/filesystem/fuse/fuse.desc: * updated fuse (2.8.3 -> 2.8.4)
09:40 < rxr> http://www.t2-project.org/packages/root.html
09:44 < CIA-32> rene * r36167 /trunk/package/network/yp-tools/yp-tools.desc: * updated yp-tools (2.11 -> 2.12)
09:47 < CIA-32> rene * r36168 /trunk/package/xorg/xf86-video-siliconmotion/xf86-video-siliconmotion.desc: * updated xf86-video-siliconmotion (1.7.3 -> 1.7.4)
10:00 < CIA-32> rene * r36169 /trunk/package/x11/xdaliclock/xdaliclock.desc: * updated xdaliclock (2.32 -> 2.33)
10:00 < CIA-32> rene * r36170 /trunk/package/network/cherokee/cherokee.desc: * updated cherokee (0.99.45 -> 0.99.47)
10:00 < CIA-32> rene * r36171 /trunk/package/gnome2/gst-plugins-good/gst-plugins-good.desc: * updated gst-plugins-good (0.10.18 -> 0.10.22)
10:00 < CIA-32> rene * r36172 /trunk/package/gnome2/gnome-icon-theme/gnome-icon-theme.desc: * updated gnome-icon-theme (2.30.1 -> 2.30.2)
10:41 -!- digitaloktay [~arch@unaffiliated/digitaloktay] has joined #t2
10:51 -!- Baldzius [~aldas@87-198-133-94.ptr.magnet.ie] has joined #t2
10:51 < Baldzius> moin
11:18 < rxr> hi Baldzius
11:18 < Baldzius> hey rxr
11:57 < rxr> you use your vm today?
11:57 < rxr> I might like to reboot for an update, later, if I find the spare time
11:58 < Baldzius> reboot whenever you like
11:58 < rxr> I could snapshot and resume, though I have never tested that yet ...
11:58 < rxr> ok
11:58 < Baldzius> that's fine with me
11:58 < rxr> not now anyway, later, if I find a free time slot
11:59 < Baldzius> cool
11:59 < rxr> trunk build goes fine?
11:59 < Baldzius> I am testing some stuff anyway
11:59 < Baldzius> I am playing with gcc's libffi
11:59 < Baldzius> want to get external one
12:00 < Baldzius> upcoming guile 2.0 will require it
12:00 < Baldzius> so preparing in advance
12:01 < rxr> yeah - I wrote the other day, or night, that I saw some packages failing because they could not find libffi anyway
12:01 < Baldzius> but we shouldn't hurry with guile update, might break some stuff around
12:01 < Baldzius> yep, it should fix it
12:01 < rxr> yeah, want to tag a new major version anyway
12:01 < Baldzius> like g-wrap and some others
12:01 < rxr> should do so soon, before upstream updates break something majorly, yet again
12:02 < Baldzius> I need to fix guile for gcc 4.5 now
12:02 < rxr> you build with gcc-4.5 already? trunk looks fine for you, too?
12:02 < Baldzius> have patch , will commit later
12:02 < Baldzius> I've not progressed too far
12:02 < Baldzius> got stuck at guile and was playing with libffi
12:03 < Baldzius> I should commit something by the end of the day
12:03 < Baldzius> only lvm fails in stage 1
12:03 < rxr> yeah I kno9w
12:03 < rxr> should disable llvm stage 1 for now
12:03 < Baldzius> but that was with pre-release 1
12:03 < rxr> :-)
12:03 < Baldzius> oh, ok
12:03 < rxr> yeah, I have not yet patched it up to cross build properly
12:03 < Baldzius> ic
12:04 < rxr> not important for the next T2 release
12:04 < rxr> but clang is nice to have for native builds for now anyway
12:04 < rxr> pretty damn good code base
12:04 < rxr> http://rene.rebe.de/2010-04-26/every-linker-needs-to-behave-differently/
12:05 < Baldzius> so only orbit2 is one annoying thing in trunk
12:05 < rxr> in stage 0 or so?
12:05 < Baldzius> you can't get it build on stage 1 on minimal
12:05 < Baldzius> or stage 0
12:05 < rxr> ic
12:05 < rxr> we should indeed only build orbit2, or commonc++ or wht it is in stage 0 if they are enabled for cross builds
12:05 < Baldzius> yep, 0
12:05 < rxr> otherwise they have no value anyway
12:06 < rxr> I only had to add them to the toolchain stage to get projects using it to cross build, to have the matching native tools to run, ...
12:06 < rxr> needs either custom code, or config code, to archive the conditional enabling in only when cross build
12:07 < rxr> if we want this generically we could specially mark the stages in the Priority tag, e.g. using a questionmark (?) to mark those stages as "only when cross build"
12:07 < rxr> or use an special tag, but that would certainly be less elegant
12:07 < rxr> if we do not want further special config stuff, we could just add custom code to the individual packages to toggle this in their config.in
12:07 < rxr> less pretty, less fast, ...
12:11 < rxr> Baldzius: btw. if you are lucky I have soon a 6-core VM container for some more distributed CPU clock cycles here
12:11 < Baldzius> I am happy with vm I have right now :)
12:11 < Baldzius> it's pretty good
12:11 < Baldzius> no complaint
12:11 < Baldzius> s
12:11 < rxr> yeah, just the 8-core xeon in the corner is sucking ...
12:11 < rxr> need to replace it, with AMD cpus :-)
12:12 < rxr> It's suprising how much more virtualization performance the way tinier, and cheaper AMD box delivers
12:12 < rxr> and that is not even decent silion, that is just the most energy efficient silicon, ...
12:15 < Baldzius> I never liked AMD cpus until they released Athlon
12:15 < Baldzius> back in 2001 I guess or so
12:16 < rxr> well in the mid 90s and mid 2000s AMD was rocking Intel majorly
12:16 < rxr> also the K6 was already pretty good, Athon was the crown, though
12:16 < rxr> also the K10 amd64 design was awesome, number crucnhing any Intel CPU at the time
12:17 < rxr> Intel catched up pretty good the last 5 years, though
12:17 < rxr> Inte's worst CPU ever, was certainly the P4
12:18 -!- mjungwirth2 [~mj-firm-w@213.174.234.68] has joined #t2
12:19 < rxr> with the fail of Intel's IA64, aka as Itanium you can see how much fail Intel was at that time
12:19 < Baldzius> yeah
12:20 < Baldzius> the strange this is that there are people who don't want anything else but Intel
12:20 < rxr> yeah
12:20 < rxr> I also got too much intel silicon into the house due to those Apple computers :-(
12:21 < rxr> but as the latest devices don't suite me anymore (either too big, hot, or the tinier ones too few ports, and too many glossy display) that's currently stagnating :-)
12:22 < rxr> http://rene.rebe.de/2009-10-29/yes-we-can-mac-os-x-on-nokia-booklet-3g/
12:24 -!- Netsplit *.net <-> *.split quits: mjungwirth_, mjungwirth
12:28 -!- Netsplit over, joins: mjungwirth_
13:57 -!- arete74 [~arete@93-44-144-189.ip97.fastwebnet.it] has joined #t2
14:00 -!- arete74 [~arete@93-44-144-189.ip97.fastwebnet.it] has quit [Client Quit]
14:09 < CIA-32> rene * r36173 /trunk/package/develop/llvm/llvm.desc: * do not try to cross build llvm, for now
14:14 < CIA-32> aldas * r36174 /trunk/package/database/mysql/mysql.desc: * updated mysql (5.1.45 -> 5.1.46)
14:14 < CIA-32> aldas * r36175 /trunk/package/develop/subversion/subversion.desc: * updated subversion (1.6.9 -> 1.6.11)
14:15 < CIA-32> aldas * r36176 /trunk/package/develop/guile/gcc45.patch: * added guile/gcc45.patch
14:25 < rxr> Baldzius: btw. if couchdb from the mailing list builds for you feel free to apply
14:25 < rxr> my build fails because I had no erlang (from the nightly buiulder, ...)
14:26 < Baldzius> my build is going to be slow and painful
14:28 < rxr> why slow and painful?
14:29 < Baldzius> because it will stop on every single error :)
14:29 < rxr> oh
14:30 < Baldzius> except junk packages I've already disabled
14:30 < rxr> ok, sound good, too
14:30 < rxr> in that case I will try to get erlang for couchdb soon myself
14:31 < Baldzius> this will be much faster :)
14:32 < rxr> thanks for your careful work, too!
14:37 < Baldzius> you're welcome
15:22 < CIA-32> aldas * r36177 /trunk/package/develop/libunistring/ (. libunistring.cache libunistring.desc): * added libunistring (0.9.2.1) - Library for manipulating Unicode strings and C strings according to the Unicode standard
15:27 -!- arete74 [~arete@93-44-144-189.ip97.fastwebnet.it] has joined #t2
15:28 < arete74> quit
15:28 < rxr> hi arete74
15:28 < rxr> for your couchdb I first have to get erlang to compile for me
15:28 < rxr> or I just commit it as it :-)
15:30 < arete74> for me work, but is better that test erlang compilation
15:30 < arete74> i have an error on build llvm on stage 0
15:30 < arete74> error is on bindig occlm
15:31 < rxr> oh
15:31 < rxr> ok, maybe we just also disable llvm in stage 0 for now, too
15:31 < rxr> not much value yet until we fixup it's cross build a little more, ...
15:34 < arete74> but the prev version 2.5 work fine
15:45 < rxr> you do not really use llvm / clang, yet, do you?
15:48 < arete74> i not use llvm/clang, but i tested it for future use!
15:49 < arete74> you have an idea for fix error on build abiword on 64 bit?
15:50 < Baldzius> that one is pretty ugly
15:50 < Baldzius> not sure where it came from
15:51 < Baldzius> that's basically the reason I am doing new build with gcc-4.5
16:15 -!- arete74 [~arete@93-44-144-189.ip97.fastwebnet.it] has quit [Quit: leaving]
16:43 -!- arete74 [~arete@93-44-144-189.ip97.fastwebnet.it] has joined #t2
16:45 -!- Baldzius [~aldas@87-198-133-94.ptr.magnet.ie] has quit [Remote host closed the connection]
16:50 -!- arete74 [~arete@93-44-144-189.ip97.fastwebnet.it] has quit [Quit: leaving]
17:42 < rxr> I just have to stress valgrind is the most awesome debugging tool of the decade
18:37 < rxr> http://www.t2-project.org/packages/valgrind.html
--- Log closed Thu Apr 29 00:00:03 2010