Sorry, this is my fault.
I was fixing some bugs involving the umbrella sampling code ignoring the
-O flag. The root of the problems I was trying to solve was that there
were two different wrappers for open (amopen and opnmrg) , each with their
own interface. To reduce redundancy, confusion and code size, I eliminated
opnmrg and changed all calls to opnmrg into calls to amopen. amopen does
handle errors internally (with an error message and exit)
I thought I cleaned up all the code that used to interface with opnmrg,
but apparently I missed at least one spot. Sorry about the extensive
debugging time.
John
On Fri, 29 Oct 2004, Robert Duke wrote:
> Mike -
> I have amber 8 and 9 trees, and it looks like someone changed the
opnmrg()
> call to an amopen() call; since amopen() returns no status, ierr is
> uninitialized (opnmrg() used to set ierr). I presume the open error is
now
> handled in the guts of amopen() but did not look; anyway, ierr sure
looks
> obsolete to me, and the code related to it should have been wiped when
> opnmrg() was changed.
> Regards - Bob
>
> ----- Original Message -----
> From: "Michael Crowley" <crowley.scripps.edu>
> To: <amber-developers.scripps.edu>
> Sent: Friday, October 29, 2004 3:39 PM
> Subject: amber-developers: code problem, anybody know about this one?
>
>
> > Dear Developers,
> > Nice to see all that came to the meeting at Stony Brook,
> > Many thanks to Carlos!!!
> >
> > Here is a bit of code that is clearly problematic wrt ierr variable.
> > The intended check shows that there
> > is code missing that would have set ierr to something, I think.
> > I am going to delete the ierr stuff, but if someone knows whether it
> > needs to be there with whatever would have set it, please inform.
> >
> > On IBM/AIX/mpxlf90_r, ierr is 1 sometimes, and causes the rdc/dip test
to
> > bomb in a way that has taken me a ridiculous amount of time to debug.
> >
> > THanks,
> > Mike
> >
> > src/sander/align.f
> >
> >
!+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> > !+ [Enter a one-line description of subroutine alignread here]
> > subroutine alignread(natom,x)
> >
> > implicit none
> > integer natom
> > _REAL_ x(*)
> > # include "nmr.h"
> > # include "files.h"
> > # include "md.h"
> > character(len=80) line
> > logical freezemol
> > integer ierr, iin, ifind, i, iof
> > namelist /align/ ndip, id, jd, dobsu, dobsl, dataset, &
> > num_datasets, s11,s22,s12,s13,s23,dcut,gigj,dij,dwt, &
> > freezemol
> >
> > ! If restraint input has been redirected, open the appropriate file
> >
> > call amopen(37,redir(8)(1:iredir(8)),'O','F','R')
> > iin = 37
> > write(6,10) redir(8)(1:iredir(8))
> > 10 format(' Alignment info will be read from file: ',a)
> > if (ierr == 1) call mexit(6,1)
> >
> >
> >
> >
> >
> > -----------------------------------------------------------------
> > Physical mail: Dr. Michael F. Crowley
> > Department of Molecular Biology, TPC6
> > The Scripps Research Institute
> > 10550 North Torrey Pines Road
> > La Jolla, California 92037
> >
> > Electronic mail: crowley.scripps.edu
> > Telephone: 858/784-9290
> > Fax: 858/784-8688
> > -----------------------------------------------------------------
> >
> >
> >
>
>
>
Received on Wed Apr 05 2006 - 23:50:00 PDT
