Re: [AMBER-Developers] Current AMBER11 CVS make test.serial failures:

From: Volodymyr Babin <vbabin.ncsu.edu>
Date: Mon, 16 Mar 2009 17:27:33 +0000

>< nstlim = 1000, nscm = 999999999, nrespa = 1
> ---
>> nstlim = 1000, nscm = 1001, nrespa = 1

I changed 'nscm' related line or two sometime ago and the difference
9999999999 vs 1001 seems to be because of that. Specifically, I did
the following:

Index: mdread.f
===================================================================
RCS file: /cvsroot/amber11/src/sander/mdread.f,v
retrieving revision 10.13
retrieving revision 10.14
diff -u -r10.13 -r10.14
--- mdread.f 3 Dec 2008 05:37:33 -0000 10.13
+++ mdread.f 28 Jan 2009 19:33:09 -0000 10.14
.@ -745,6 +745,9 @@
       ndfmin = 0
    endif

+ if (ntr.eq.1) &
+ nscm = nstlim + 1
+
    nsolut = nrp
    if ( nscm > 0 .and. ntb == 0 ) then
       ndfmin = 6 ! both translation and rotation com motion removed

which I believe is absolutely correct.

> > ** NCSU-Error ** : expected list value for key 'i', got '<EMPTY>'
> instead

I believe this is gfortran 4.1.2 bug.

Have a great day,

Volodymyr


_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Wed Mar 18 2009 - 01:10:19 PDT
Custom Search