Could someone please check if the patch below fixes the problem? I do not
have this version of gfortran around to try it myself.
Index: mt19937.f
===================================================================
RCS file: /cvsroot/amber11/src/sander/mt19937.f,v
retrieving revision 10.1
diff -u -r10.1 mt19937.f
--- mt19937.f 7 May 2008 20:50:54 -0000 10.1
+++ mt19937.f 4 Nov 2008 03:55:22 -0000
.. -269,7 +269,10 ..
data matrix_a /z'9908B0DF'/
data matrix_b /z'0'/
data temper_a /z'9D2C5680'/
- data temper_b /z'EFC60000'/
+! data temper_b /z'EFC60000'/
+
+ temper_b = z'EFC6' ! makes gfortran happier
+ temper_b = ieor(ishft(temper_b, 16), z'0000')
if ( self%mti > n ) then ! generate N words at one time
if ( .not. self%mtinit ) call init_by_seed(self, seed_d) ! if
init_genrand() has not been called, a default initial seed is used
Received on Fri Dec 05 2008 - 10:36:17 PST
