From: Wenyu Zhong (wyzhong78_at_163.com)
Date: Thu Apr 15 2004 - 06:35:08 CDT
Another question in compiling i686-icc version namd.
I tried to compile i686-tcp-smp version NAMD, and after resolving several problems, I got it.
But when I tried to run it,I meet the below errors:
error while loading shared libraries: libimf.so: cannot open shared object file: No such file or directory
libm.so.6 => /lib/tls/libm.so.6 (0xb75c0000)
libpthread.so.0 => /lib/tls/libpthread.so.0 (0xb75b0000)
libdl.so.2 => /lib/libdl.so.2 (0xb75ad000)
libz.so.1 => /usr/lib/libz.so.1 (0xb759f000)
libimf.so => not found
libcprts.so.5 => not found
libcxa.so.5 => not found
libunwind.so.5 => not found
libc.so.6 => /lib/tls/libc.so.6 (0xb7466000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0xb75eb000)
Charmrun rsh(compute-0-21.local.0)> Exiting with error code 1
----- Original Message -----
From: "Wenyu Zhong" <wyzhong78_at_163.com>
Sent: Thursday, April 15, 2004 2:42 PM
Subject: namd-l: Compiling NAMD on amd64
> I have tried compiling NAMD 2.5 on opteron 242(1.6G), Rocks3.1.0 x86_64(RedHat 2.1EL).I compiled a Linux-amd64-g++ version successfully and it achived about 5.4s/step in running apoa1 benchmark .it is little slower than 5.1s/step got by the standard NAMD_2.5_Linux binary running on Xeon 2.4G.
> How do you think about the performance of the Linux-amd64-g++ version? Is there any optimization I should do? Do you think other compliers like pgcc will do much more better than gcc?
> Best regards!
> Wenyu Zhong
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:37:31 CST