Re: can somebody verify this C program which calls dsaupd_ ? (longish)
- From: glen herrmannsfeldt <gah@xxxxxxxxxxxxxxxx>
- Date: Sat, 02 Apr 2005 19:07:06 -0800
H. S. wrote:
Apparently, _Madhusudan Singh_, on 02/04/05 19:05,typed:
H. S. wrote:
(terribly sorry for this long post which includes my two source files, but I could't find out how put the files online using my Yahoo account)
The problem is not with the length of your post but with the relevance to
the newsgroup.
Since I was requesting somebody who already has the FORTRAN libraries installed and works with them to test if the program does call dsaupd_ properly without getting info=-9999 as error, I do not see how the post is NOT relevant to FORTRAN language newsgroup.
(from previous post)
> The remaining parameters to the function calls are as follows:
> dsaupd(int n, int nev, double *Evals) > dsaupd(int n, int nev, doubel *Evals, double **Evecs)
I would be very surprised to see a Fortran program expecting a double**.
It would be usual to expect a double* that is big enough for the whole
matrix, as it would be for C if declared as a 2D matrix. It is, of course, system dependent, but to satisfy some of the Fortran rules for
calls, it isn't easy otherwise.
If dsaupd is not a Fortran program, but a C interface to Fortran then I would have to agree that the question is 100% C. (Actually, C++ as there are many non-C constructs used.)
Otherwise, post the generated assembly code for the smallest Fortran subroutine expecting a two-dimensional array argument so we can see the expected calling sequence. How about an in-place transpose of a 2D matrix.
-- glen
-- glen
.
- Follow-Ups:
- Prev by Date: Re: Prime numbers
- Next by Date: Re: A question on Newton's Method
- Previous by thread: Re: Prime numbers
- Next by thread: Re: can somebody verify this C program which calls dsaupd_ ? (longish)
- Index(es):
Relevant Pages
|
Loading
