VMD-L Mailing List

From: Justin Gullingsrud (justin)
Date: Sat Jun 24 2000 - 22:37:13 CDT

Just for kicks I decided to try compiling VMD on the T3E at the
Pittsburgh Supercomputer Center. I started with OSF1 and made a few
hacks, including:

 - change "/usr/bin/env perl" to "/bin/env perl"
 - add /usr/local/include and /usr/local/lib to INCDIR and LIBDIR
 - remove -DVMDTCLX, since it doesn't seem to be installed at PSC,
   and commented out the "signal" and "rename help" lines in vmdinit.tcl.
 - had a weird bug in MSMSInterface with the linker telling me that the
   symbol nsendmsg(Inform &) was undefined; I commented out all sendmsg
   lines and it linked fine. Not that we have MSMS for T3E anyway...

Anyway, I can load molecules, do atom selections, and run simple scripts.
I haven't tried to compile Mesa yet, I guess that's the next thing to try.
Also, pdbload fails because 'random' is apparently a TclX command or
something. Still, having a text-only version of VMD on the T3E might be
nice for running analysis scripts or something...

Justin
  

-- 
Justin Gullingsrud         3111 Beckman Institute
H: (217) 384-4220
W: (217) 244-8946          The time of purification is at hand...