NAMD on Linux cluster with Myrinet

From: Alberto Santana (
Date: Sat Oct 09 2004 - 20:42:55 CDT

Hi folks,

We are using version 8.0 of Intel's compiler and GM-2.0.11_Linux. I
downloaded charm-5.8 and built it with ./build charm++ net-linux icc
gm --basedir=/usr/opt/gm -O -DCMK_OPTIMIZE=1. This step went very smooth.
NAMD compiled without a problem. A test run gave the following output:

Charm++: scheduler running in netpoll mode.
Info: NAMD 2.5 for Linux-i686-GM
Info: Please visit
Info: and send feedback or bug reports to
Info: Please cite Kale et al., J. Comp. Phys. 151:283-312 (1999)
Info: in all publications reporting results obtained with NAMD.
Info: Based on Charm++/Converse 0143160 for net-linux-gm-icc
Info: Built Sat Oct 9 20:13:10 EDT 2004 by santana on master
Info: Sending usage information to NAMD developers via UDP. Sent data is:
Info: 1 NAMD 2.5 Linux-i686-GM 4 node3 santana
Info: Running on 4 processors.
Info: 9733 kB of memory in use.
Measuring processor speeds...------------- Processor 1 Exiting: Caught
Signal ------------
Signal: segmentation violation
Suggestion: Try running with '++debug', or linking with '-memory paranoid'.
Fatal error on PE 1> segmentation violation

The question is: What is the secret behind making NAMD run on Myrinet?
Has anyone managed to get this thing right?


This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:37:54 CST