CUDA-accelerated NAMD does not support NBFIX parameters

From: Héctor Urbina (hurbinas_at_gmail.com)
Date: Sun Sep 13 2009 - 13:47:33 CDT

Dear All,

I am trying to get NAMD CUDA working in my laptop, a Dell Vostro 1520,
with a GeForce 9300M GS nVidia GPU. I'm using Arch Linux, kernel
2.6.30.6-1, having cuda 2.3-1 and nvidia drivers 190.32-1 installed.

I have downloaded and compiled several cvs versions of namd, the last
one was the August-31 source code.

I am using the icc compiler to build charm net-linux-icc and NAMD
Linux-x86-icc. Then, the command line: "/opt/namd_cvs_cuda/charmrun
+p2 /opt/namd_cvs_cuda/namd2 +idlepoll apoa1.namd" ends with the
following error message:

FATAL ERROR: CUDA-accelerated NAMD does not support NFBIX parameters.

I understand NFBIX actually refers to NBFIX, but I don't understand
what exactly should CUDA-accelerated NAMD support.

Am I missing something during the build process? or it is an error
related to my GPU, or to my cuda installed version...?

Google search doesn't help, until now.

Thanks for your consideration

-- 
Héctor Urbina S.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:51:28 CST