T2 IRC Log: 2009-02-27
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 Feb 27 00:00:13 2009
00:01 -!- koan [n=koan@unaffiliated/koan] has quit ["Lost terminal"]
00:06 -!- koan [n=koan@unaffiliated/koan] has joined #t2
05:29 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: @ChanServ
05:30 -!- ChanServ [ChanServ@services.] has joined #t2
05:30 -!- ServerMode/#t2 [+o ChanServ] by irc.freenode.net
09:10 -!- T-One [n=T-One@62.218.122.66] has joined #t2
10:43 < CIA-23> rene * r32452 /trunk/package/scientific/mpfr/mpfr.desc: * updated mpfr (2.4.0 -> 2.4.1)
10:43 < CIA-23> rene * r32453 /trunk/package/develop/ginac/ginac.desc: * updated ginac (1.5.0 -> 1.5.1)
10:43 < CIA-23> rene * r32451 /trunk/package/xorg/xf86-input-hyperpen/xf86-input-hyperpen.desc: * updated xf86-input-hyperpen (1.2.0 -> 1.3.0)
10:43 < CIA-23> rene * r32454 /trunk/package/x11/fvwm/fvwm.desc: * updated fvwm (2.5.26 -> 2.5.27)
10:47 < CIA-23> rene * r32455 /trunk/package/network/mldonkey/mldonkey.desc: * updated mldonkey (2.9.7 -> 3.0.0)
10:47 -!- digitaloktay [n=debian@dslb-084-057-249-069.pools.arcor-ip.net] has joined #t2
10:54 < CIA-23> rene * r32456 /trunk/package/www/midori/ (midori.conf midori.desc): * updated midori (0.20 -> 0.1.3)
14:01 < CIA-23> rene * r32457 /trunk/package/develop/php/php.desc: * updated php (5.2.8 -> 5.2.9)
14:27 -!- T-One [n=T-One@62.218.122.66] has quit ["Verlassend"]
14:28 < CIA-23> rene * r32458 /trunk/package/base/man-pages/man-pages.desc: * updated man-pages (3.18 -> 3.19)
14:30 < CIA-23> rene * r32459 /trunk/package/xfce4/xfce4-dict-plugin/xfce4-dict-plugin.cache: * added xfce4-dict-plugin
15:05 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has joined #t2
15:05 < Baldzius> moin
15:06 < Baldzius> rxr: are you doing xfce4 update?
15:06 < Baldzius> as I have it mostly done
15:27 < Baldzius> ok I have it all done
15:36 < rxr> Baldzius: I only did a quick curl | sed | Update-Pkg run
15:36 < rxr> not much built yet .-)
15:36 < rxr> just commit .-)
15:39 < Baldzius> cool
15:43 < Baldzius> crap
15:44 < Baldzius> I've deleted libxfce4mcs xfce-mcs-plugins and xfce-mcs-manager and forgot that I had them already compiled on my ref build
15:44 < Baldzius> dammit
15:44 < Baldzius> they all went into new cache files
15:44 < rxr> sed //d -i ...
15:45 < Baldzius> but I want to make sure that xfce4 definently builts without them
15:45 < Baldzius> how can i remove package from build?
15:46 < rxr> mine -rf -R build/... package-name
15:49 < Baldzius> need to put it down
15:49 < Baldzius> I think I had it before but lost :)
15:50 < Baldzius> s/put/write/
15:50 < rxr> best would be if we extend Create-ErrList -newremove / -newdelete to honor delete packages and wipe the in-build-sandbox package content accordingly
15:57 < Baldzius> sounds good
16:12 < CIA-23> aldas * r32460 /trunk/package/ (57 files in 37 dirs): (log message trimmed)
16:12 < CIA-23> * updated xfce4 (4.4.2 -> 4.6.0), including:
16:12 < CIA-23> updated xfce4-terminal (0.2.8 -> 0.2.10)
16:12 < CIA-23> updated libexo (0.3.4 -> 0.3.100)
16:12 < CIA-23> changed libexo build priority (700.000 -> 223.400)
16:12 < CIA-23> updated xfce-utils (4.4.2 -> 4.6.0)
16:12 < Baldzius> should be fine
16:12 < CIA-23> added libxfce4menu (4.6.0) - Xfce4 desktop menu library
16:12 < Baldzius> evertyhing was compiled with those stupid packages
16:12 < Baldzius> tried couple of them without and it didn't fail
16:13 < Baldzius> so hopefully it's fine
16:16 < CIA-23> rene * r32461 /trunk/architecture/blackfin/package/uclibc/error_print_progname.patch:
16:16 < CIA-23> Michael Iedema:
16:16 < CIA-23> * fixed uclibc for blackfin
16:20 -!- digitaloktay [n=debian@dslb-084-057-249-069.pools.arcor-ip.net] has quit ["Verlassend"]
17:25 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has quit ["Leaving"]
18:38 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has joined #t2
18:45 < rxr> http://rene.rebe.name/2009-02-27/a-companion-for-your-sub-notebook-and-yourself/
18:45 [Users #t2]
18:45 [@ChanServ] [ dsoul] [ mtr] [ SerWou ] [ Stealth ] [ TobiX]
18:45 [ CIA-23 ] [ koan ] [ rxr] [ Skaperen] [ synchris] [ tri ]
18:45 -!- Irssi: #t2: Total of 12 nicks [1 ops, 0 halfops, 0 voices, 11 normal]
18:51 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has quit ["Leaving."]
19:10 < koan> rxr: that looks like a woman's handbag :p
19:11 < rxr> don't think so :-)
19:13 < rxr> http://rene.rebe.name/2009-02-27/the-state-of-gpgpu/
19:13 < rxr> maybe you enjoy that more :-)
19:16 < koan> good post
19:17 < koan> so you don't believe OpenCL will change the situation for GPGPU?
19:17 < rxr> as long as the drivers are such a messy mess ...
19:17 < rxr> and we are still quite away from an open source OpenCL implementation
19:17 < rxr> (, anyway)
19:18 < rxr> I think the GPUs are not really up to really accelerate general purpose programs anyway
19:18 < rxr> for that a setup liek the PS3's cell's would be way more flexible
19:18 < rxr> and able to accelerate more, theoretically even the javascript and rendering in your browser
19:19 < rxr> that's nearly impossible to implement / accelerate on the current GPU kind of design architecture
19:19 < koan> indeed, but I think OpenCL is not only interesting for the acceleration part, but also because it gives the programmers one interface to program such a wide range of processor architectures (cpu, gpu, cell, larrabee, ...)
19:20 < rxr> yes, indeed
19:20 < rxr> but I only looked briefly at it (too much TODO, though I even have Snow Leopard somewhere around, though not the latest seed our Apple Developer Connect membership expired recently :-)
19:21 < rxr> to be renewed the next time we need to bug them :-)
19:21 < rxr> (about their bugs :-9
19:21 < koan> but it's not suitable for every programmer, I looked into the specification and it's rather low-level and you need much 'administration' code
19:22 < rxr> well - on the high level you device like "throwing this cind of matrix over this kind of layed out bit patter (int, float, triple, tuple, etc.)
19:22 < rxr> but this is mostly intersting for image processing, physics
19:23 < rxr> stuff like the cell or larrabee (what that spelled correctly?) could potentially even accelerate your (now much hyped) java script or vector (SVG, ...) rendering
19:23 < rxr> I do not see current (classic) GPUs doing that
19:24 < koan> what's the problem with GPU's that they can't do this?
19:24 < rxr> well - the do not have that flexible control flow and conditional stuff - and a too high CPU to GPU latency
19:25 < rxr> they are quite strictly architectured for vector operations "only"
19:25 < rxr> like pixel shading and moving some matrix kernel over some abritrary set of vector data
19:39 < rxr> so - have to leave - cu then
20:57 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has joined #t2
21:17 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has quit ["Leaving."]
21:56 -!- RayMan72 [n=radjis@dhcp-077-251-249-082.chello.nl] has joined #t2
22:29 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has joined #t2
23:09 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has quit [Remote closed the connection]
23:22 < rxr> http://www.t2-project.org/hardware/portable/Acer/Aspire-One-150/
23:22 < rxr> (yes web day)
23:22 < rxr> (just joking)
23:23 < koan> that's the netbook you bought?
23:24 < koan> I have the A110, but didn't try T2 yet
--- Log closed Sat Feb 28 00:00:13 2009
00:01 -!- koan [n=koan@unaffiliated/koan] has quit ["Lost terminal"]
00:06 -!- koan [n=koan@unaffiliated/koan] has joined #t2
05:29 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: @ChanServ
05:30 -!- ChanServ [ChanServ@services.] has joined #t2
05:30 -!- ServerMode/#t2 [+o ChanServ] by irc.freenode.net
09:10 -!- T-One [n=T-One@62.218.122.66] has joined #t2
10:43 < CIA-23> rene * r32452 /trunk/package/scientific/mpfr/mpfr.desc: * updated mpfr (2.4.0 -> 2.4.1)
10:43 < CIA-23> rene * r32453 /trunk/package/develop/ginac/ginac.desc: * updated ginac (1.5.0 -> 1.5.1)
10:43 < CIA-23> rene * r32451 /trunk/package/xorg/xf86-input-hyperpen/xf86-input-hyperpen.desc: * updated xf86-input-hyperpen (1.2.0 -> 1.3.0)
10:43 < CIA-23> rene * r32454 /trunk/package/x11/fvwm/fvwm.desc: * updated fvwm (2.5.26 -> 2.5.27)
10:47 < CIA-23> rene * r32455 /trunk/package/network/mldonkey/mldonkey.desc: * updated mldonkey (2.9.7 -> 3.0.0)
10:47 -!- digitaloktay [n=debian@dslb-084-057-249-069.pools.arcor-ip.net] has joined #t2
10:54 < CIA-23> rene * r32456 /trunk/package/www/midori/ (midori.conf midori.desc): * updated midori (0.20 -> 0.1.3)
14:01 < CIA-23> rene * r32457 /trunk/package/develop/php/php.desc: * updated php (5.2.8 -> 5.2.9)
14:27 -!- T-One [n=T-One@62.218.122.66] has quit ["Verlassend"]
14:28 < CIA-23> rene * r32458 /trunk/package/base/man-pages/man-pages.desc: * updated man-pages (3.18 -> 3.19)
14:30 < CIA-23> rene * r32459 /trunk/package/xfce4/xfce4-dict-plugin/xfce4-dict-plugin.cache: * added xfce4-dict-plugin
15:05 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has joined #t2
15:05 < Baldzius> moin
15:06 < Baldzius> rxr: are you doing xfce4 update?
15:06 < Baldzius> as I have it mostly done
15:27 < Baldzius> ok I have it all done
15:36 < rxr> Baldzius: I only did a quick curl | sed | Update-Pkg run
15:36 < rxr> not much built yet .-)
15:36 < rxr> just commit .-)
15:39 < Baldzius> cool
15:43 < Baldzius> crap
15:44 < Baldzius> I've deleted libxfce4mcs xfce-mcs-plugins and xfce-mcs-manager and forgot that I had them already compiled on my ref build
15:44 < Baldzius> dammit
15:44 < Baldzius> they all went into new cache files
15:44 < rxr> sed //d -i ...
15:45 < Baldzius> but I want to make sure that xfce4 definently builts without them
15:45 < Baldzius> how can i remove package from build?
15:46 < rxr> mine -rf -R build/... package-name
15:49 < Baldzius> need to put it down
15:49 < Baldzius> I think I had it before but lost :)
15:50 < Baldzius> s/put/write/
15:50 < rxr> best would be if we extend Create-ErrList -newremove / -newdelete to honor delete packages and wipe the in-build-sandbox package content accordingly
15:57 < Baldzius> sounds good
16:12 < CIA-23> aldas * r32460 /trunk/package/ (57 files in 37 dirs): (log message trimmed)
16:12 < CIA-23> * updated xfce4 (4.4.2 -> 4.6.0), including:
16:12 < CIA-23> updated xfce4-terminal (0.2.8 -> 0.2.10)
16:12 < CIA-23> updated libexo (0.3.4 -> 0.3.100)
16:12 < CIA-23> changed libexo build priority (700.000 -> 223.400)
16:12 < CIA-23> updated xfce-utils (4.4.2 -> 4.6.0)
16:12 < Baldzius> should be fine
16:12 < CIA-23> added libxfce4menu (4.6.0) - Xfce4 desktop menu library
16:12 < Baldzius> evertyhing was compiled with those stupid packages
16:12 < Baldzius> tried couple of them without and it didn't fail
16:13 < Baldzius> so hopefully it's fine
16:16 < CIA-23> rene * r32461 /trunk/architecture/blackfin/package/uclibc/error_print_progname.patch:
16:16 < CIA-23> Michael Iedema
16:16 < CIA-23> * fixed uclibc for blackfin
16:20 -!- digitaloktay [n=debian@dslb-084-057-249-069.pools.arcor-ip.net] has quit ["Verlassend"]
17:25 -!- Baldzius [n=aldas@87-198-133-94.ptr.magnet.ie] has quit ["Leaving"]
18:38 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has joined #t2
18:45 < rxr> http://rene.rebe.name/2009-02-27/a-companion-for-your-sub-notebook-and-yourself/
18:45 [Users #t2]
18:45 [@ChanServ] [ dsoul] [ mtr] [ SerWou ] [ Stealth ] [ TobiX]
18:45 [ CIA-23 ] [ koan ] [ rxr] [ Skaperen] [ synchris] [ tri ]
18:45 -!- Irssi: #t2: Total of 12 nicks [1 ops, 0 halfops, 0 voices, 11 normal]
18:51 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has quit ["Leaving."]
19:10 < koan> rxr: that looks like a woman's handbag :p
19:11 < rxr> don't think so :-)
19:13 < rxr> http://rene.rebe.name/2009-02-27/the-state-of-gpgpu/
19:13 < rxr> maybe you enjoy that more :-)
19:16 < koan> good post
19:17 < koan> so you don't believe OpenCL will change the situation for GPGPU?
19:17 < rxr> as long as the drivers are such a messy mess ...
19:17 < rxr> and we are still quite away from an open source OpenCL implementation
19:17 < rxr> (, anyway)
19:18 < rxr> I think the GPUs are not really up to really accelerate general purpose programs anyway
19:18 < rxr> for that a setup liek the PS3's cell's would be way more flexible
19:18 < rxr> and able to accelerate more, theoretically even the javascript and rendering in your browser
19:19 < rxr> that's nearly impossible to implement / accelerate on the current GPU kind of design architecture
19:19 < koan> indeed, but I think OpenCL is not only interesting for the acceleration part, but also because it gives the programmers one interface to program such a wide range of processor architectures (cpu, gpu, cell, larrabee, ...)
19:20 < rxr> yes, indeed
19:20 < rxr> but I only looked briefly at it (too much TODO, though I even have Snow Leopard somewhere around, though not the latest seed our Apple Developer Connect membership expired recently :-)
19:21 < rxr> to be renewed the next time we need to bug them :-)
19:21 < rxr> (about their bugs :-9
19:21 < koan> but it's not suitable for every programmer, I looked into the specification and it's rather low-level and you need much 'administration' code
19:22 < rxr> well - on the high level you device like "throwing this cind of matrix over this kind of layed out bit patter (int, float, triple, tuple, etc.)
19:22 < rxr> but this is mostly intersting for image processing, physics
19:23 < rxr> stuff like the cell or larrabee (what that spelled correctly?) could potentially even accelerate your (now much hyped) java script or vector (SVG, ...) rendering
19:23 < rxr> I do not see current (classic) GPUs doing that
19:24 < koan> what's the problem with GPU's that they can't do this?
19:24 < rxr> well - the do not have that flexible control flow and conditional stuff - and a too high CPU to GPU latency
19:25 < rxr> they are quite strictly architectured for vector operations "only"
19:25 < rxr> like pixel shading and moving some matrix kernel over some abritrary set of vector data
19:39 < rxr> so - have to leave - cu then
20:57 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has joined #t2
21:17 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has quit ["Leaving."]
21:56 -!- RayMan72 [n=radjis@dhcp-077-251-249-082.chello.nl] has joined #t2
22:29 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has joined #t2
23:09 -!- tri [n=tri@p4FCF1E73.dip0.t-ipconnect.de] has quit [Remote closed the connection]
23:22 < rxr> http://www.t2-project.org/hardware/portable/Acer/Aspire-One-150/
23:22 < rxr> (yes web day)
23:22 < rxr> (just joking)
23:23 < koan> that's the netbook you bought?
23:24 < koan> I have the A110, but didn't try T2 yet
--- Log closed Sat Feb 28 00:00:13 2009