#emc-devel | Logs for 2007-04-26

[02:49:59] <skunkworks> so what is going on wednesday night that is more important than #emc
[02:52:00] <skunkworks> ?
[02:56:40] <jmkasunich> hi guys
[02:56:47] <jmkasunich> jepler: around?
[02:58:36] <cradek> hi
[02:59:42] <jmkasunich> hi
[02:59:54] <skunkworks> * skunkworks ment that to be on #emc
[03:00:06] <jmkasunich> jepler mentioned my name earlier today
[03:00:15] <jmkasunich> said I reverted something and busted 64bit
[03:01:20] <cradek> ah
[03:03:15] <skunkworks> cradek: btw - the dual processor didn't restart x with either kernel. (running the latency test and glxgears)
[03:05:07] <skunkworks> but that video card (ati rage 128) runs the glxgears smooth.
[03:06:10] <skunkworks> Smoothest I have ever seen it :)
[08:54:06] <jmk-solo> hmmm, we seem to have some small issues with parallel make...
[08:54:37] <jmk-solo> using "make -j" I got:
[08:54:38] <jmk-solo> Copying python script pumagui
[08:54:38] <jmk-solo> Copying python script scaragui
[08:54:38] <jmk-solo> Linking ../rtlib/abs.so
[08:54:38] <jmk-solo> Linking ../rtlib/and2.so
[08:54:38] <jmk-solo> ld: no input files
[08:54:39] <jmk-solo> make: *** [../rtlib/abs.so] Error 1
[08:54:42] <jmk-solo> make: *** Waiting for unfinished jobs....
[08:54:44] <jmk-solo> Compiling emc/rs274ngc/gcodemodule.cc
[08:54:46] <jmk-solo> ld: no input files
[08:54:48] <jmk-solo> make: *** [../rtlib/and2.so] Error 1
[08:54:49] <jmk-solo> Copying python script debuglevel
[08:54:58] <jmk-solo> following it up with a plain "make" successfully finished the build
[08:55:37] <jmk-solo> total real time 35 seconds ;-)
[09:06:02] <jmk-solo> jmk-solo is now known as jmk-core2
[11:21:42] <cradek> wow I hope you don't plan on going to work today, being still up at 5am
[11:24:24] <alex_joni> morning chris
[11:28:34] <cradek> hi
[11:28:39] <cradek> man it's early
[11:30:36] <jepler> jmkasunich: it was also a rather old distribution (fedora core 2 or something) so I wouldn't sweat it as long as more modern 64 bits work
[11:30:39] <alex_joni> I can understand
[11:30:45] <alex_joni> but it seems you people find no sleep :D
[11:31:44] <jepler> jmkasunich: hm I am surprised at your "make -j" error; I did that a lot (when I first got my dual-core) but maybe some things have changed since then
[11:35:39] <jepler> I *do* get the error now, though
[11:46:54] <jepler> actually I get it without -j
[11:46:59] <jepler> I get it after a 'clean'
[14:07:26] <cradek> hi ray! haven't seen you around for a while
[14:07:30] <cradek> hope all is well
[20:21:55] <lerman_> lerman_ is now known as lerman
[22:47:58] <mschuhmacher> Hi
[22:48:15] <mschuhmacher> is anyone around, who knows something about this: http://www.linuxcnc.org/experimental/
[22:48:35] <skunkworks> what is your question?
[22:48:57] <mschuhmacher> when I start the rtai my system will lock
[22:49:33] <skunkworks> is this a dual processor machine?
[22:49:43] <skunkworks> are you using the rtai tests from the wiki?
[22:50:04] <mschuhmacher> yes
[22:50:15] <mschuhmacher> it is a 4x Xeon
[22:50:23] <skunkworks> 4 processor?
[22:50:44] <mschuhmacher> http://www.filpro.de/Photo/Screenshot.png
[22:52:11] <mschuhmacher> I there are 4 logical processors
[22:52:22] <skunkworks> I think that it is only compiled for a dual processor machine.
[22:52:25] <skunkworks> try this
[22:52:26] <skunkworks> sudo mknod /dev/rtf3 c 150 3; cd /usr/realtime-2.6.17-rtai3.5/testsuite/kern/latency; ./run
[22:55:39] <mschuhmacher> System Locked
[22:56:04] <mschuhmacher> reboot
[22:56:36] <skunkworks> hmm - yah - I am pretty sure cradek said that it would require tweeking for more than 2 processors.
[23:15:20] <cradek> those packages are a failed experiment. sadly, making one build of rtai work on many machines is not a goal for the rtai team, so our hands are pretty tied when it comes to using anything but the most basic rtai configuration
[23:16:02] <cradek> of course you can build rtai to whatever specs you like for your machine, and I bet it would work fine
[23:53:43] <jmkasunich> jepler: thanks for fixing make -j