DSS Error When Used in Subroutine

[Brad_Doudican]

I am attempting to get DSS to operate through Visual Studio 2008 to solve the sparse matrix problem Ax=B where A=GLK and B=BL. The code first converts A=GLK to CDR format, assigns BL, then calls Solver to deploy the DSS functions. The program runs fine through the call to DSS_CREATE, then exits with an error at line DSS_DEFINE_STRUCTURE. Error and code are below.

Since the first DSS function executes, I am assuming compiler configurations for MKL library are fine. Is there a way to debug to find the problem in execution of DSS_DEFINE_STRUCTURE? How do we interpret code -8?

Many thanks for any help.

**********ERROR*************

The program '[5004] doudican_current.exe: Native' has exited with code -8 (0xfffffff8).

*******CODE***********

!***********************************************************************

!

SUBROUTINE DOUDICAN_SOLVE (GLK, BL, SOLUTION)

!

! THIS SUBROUTINE

!

USE all

!

implicit none

!

REAL(KIND=DBL), INTENT(IN), DIMENSION(NEQ, NEQ) :: GLK

REAL(KIND=DBL), INTENT(IN), DIMENSION(NEQ) :: BL

REAL(KIND=DBL), INTENT(OUT), DIMENSION(NEQ) :: SOLUTION

REAL(KIND=DBL), ALLOCATABLE, DIMENSION(:) :: RHS, VALUES

INTEGER, ALLOCATABLE, DIMENSION(:) :: COLUMNS

INTEGER, DIMENSION(NEQ+1) :: ROWINDEX

INTEGER ::I, J, KOUNT,NROWS

!

! ALTERNATVE DIRECT SPARSE SOLVER

!

KOUNT=COUNT(GLK/=0d0)

ALLOCATE ( VALUES(KOUNT), COLUMNS(KOUNT) )

KOUNT = 0

ROWINDEX(1)=1

DO J=1,NEQ

ROWINDEX(J+1)=ROWINDEX(J)

DO I=1,NEQ

IF (GLK(J,I)/=0) THEN

ROWINDEX(J+1)=ROWINDEX(J+1)+1

KOUNT=KOUNT+1

VALUES(KOUNT)=GLK(J,I)

COLUMNS(KOUNT)=I

END IF

END DO

END DO

!

NROWS = NEQ

ALLOCATE ( RHS(NROWS) )

RHS = BL

!

CALL SOLVER ( VALUES, COLUMNS, ROWINDEX, RHS, KOUNT, NROWS, SOLUTION )

!

RETURN

END SUBROUTINE DOUDICAN_SOLVE

!

!***********************************************************************

!***********************************************************************

SUBROUTINE SOLVER ( VALUES, COLUMNS, ROWINDEX, RHS, KOUNT, NROWS, SOLUTION )

!

USE ALL

use mkl_dss

!

IMPLICIT NONE

INTEGER, PARAMETER :: dp = KIND(1.0D0)

INTEGER :: error

INTEGER :: i, KOUNT

INTEGER, PARAMETER :: bufLen = 20

! Define the data arrays and the solution and rhs vectors.

INTEGER, DIMENSION(KOUNT) :: columns

INTEGER :: nCols

INTEGER :: nNonZeros

INTEGER :: nRhs

INTEGER :: nRows

REAL(KIND=DP), DIMENSION(NROWS) :: rhs

INTEGER, DIMENSION(NEQ+1) :: rowIndex

REAL(KIND=DBL), DIMENSION(NEQ) :: solution

REAL(KIND=DP), DIMENSION(KOUNT) :: values

TYPE(MKL_DSS_HANDLE) :: handle ! Allocate storage for the solver handle.

REAL(KIND=DP),ALLOCATABLE::statOUt( : )

CHARACTER*15 statIn

INTEGER perm(1)

INTEGER buff(bufLen)

!

! Set the problem to be solved.

!

nCols = 5

nNonZeros = KOUNT

nRhs = 1

perm(1) = 0

!

! Initialize the solver.

error = DSS_CREATE( handle, MKL_DSS_DEFAULTS )

IF (error /= MKL_DSS_SUCCESS) GOTO 999

!

! Define the non-zero structure of the matrix.

!

! *NOTE: ERROR OCCURS IN NEXT LINE*

!

error = DSS_DEFINE_STRUCTURE( handle, MKL_DSS_SYMMETRIC, rowIndex, nRows, &

& nCols, columns, nNonZeros )

IF (error /= MKL_DSS_SUCCESS) GOTO 999

!

! Reorder the matrix.

error = DSS_REORDER( handle, MKL_DSS_DEFAULTS, perm )

IF (error /= MKL_DSS_SUCCESS) GOTO 999

!

! Factor the matrix.

error = DSS_FACTOR_REAL( handle, MKL_DSS_DEFAULTS, values )

IF (error /= MKL_DSS_SUCCESS) GOTO 999

!

! Solve the problem.

error = DSS_SOLVE_REAL(handle, MKL_DSS_DEFAULTS, rhs, nRhs, solution )

IF (error /= MKL_DSS_SUCCESS) GOTO 999

!

! Deallocate solver storage and various local arrays.

error = DSS_DELETE( handle, MKL_DSS_DEFAULTS )

IF (error /= MKL_DSS_SUCCESS ) GOTO 999

!

! Print the solution vector, deallocate it and exit

WRITE(*,"('Solution Array: '(5F10.3))") ( solution(i), i = 1, nCols )

GOTO 1000

!

! Print an error message and exit

999 WRITE(*,*) "Solver returned error code ", error

STOP 1

!

1000 CONTINUE

pause

END SUBROUTINE SOLVER

!***********************************************************************

[Gennady_F_Intel]

that's mean your input matrix is not square:

DSS software operates only on square matrices, sonRowsmust be equal tonCols.

see mkl_dss.h file as an example: #define MKL_DSS_NOT_SQUARE -8

--Gennady

[mecej4]

In Subroutine Solver you have "nCols = 5"; should that have been "ncols=NEQ" ?