MATMUL_SPARSE.f90 File Reference

Go to the source code of this file.

Functions/Subroutines
subroutine	matmul_sparse (as, nnz, jsp, isp, vet_out, n, vet_in,
	Matrix-vector multiplication for sparse matrices (RCS format).

Function/Subroutine Documentation

matmul_sparse()

subroutine matmul_sparse	(	real*8, dimension(nnz)	as,
		integer*4	nnz,
		integer*4, dimension(nnz)	jsp,
		integer*4, dimension(0:n)	isp,
		real*8, dimension(n)	vet_out,
		integer*4	n,
		real*8, dimension(m)	vet_in
	)

Matrix-vector multiplication for sparse matrices (RCS format).

Author

Ilario Mazzieri

Date

September, 2013

Version

1.0

Parameters

[in]	nnz	number of nonzero elements
[in]	As	sparse matrix
[in]	Jsp	pointer for sparsity pattern
[in]	Isp	pointer for sparsity pattern
[in]	n	row of the matrix
[in]	m	column of the matrix
[in]	vet_in	input vector for the multiplication As*vet_in
[in]	error	control parameter (dummy)
[out]	vet_out	result of the multiplication As*vet_in

Definition at line 33 of file MATMUL_SPARSE.f90.

 
 
        implicit none
        
        integer*4 :: nnz, i, j, n, m, error   
        integer*4 :: Jsp(nnz), Isp(0:n)  
  
        real*8 :: as(nnz), vet_in(m), vet_out(n)
 
        vet_out = 0.d0
 
              
        do i = 1, n 
           do j = isp(i-1) + 1, isp(i)
              vet_out(i) = vet_out(i) + as(j)*vet_in(jsp(j))
           enddo
        enddo
        
 
        return

Referenced by compute_energy_error().

Here is the caller graph for this function: