Centre Blaise Pascal » HPL sur GPU

/*

 * -- High Performance Computing Linpack Benchmark (HPL)

 *    HPL - 2.0 - September 10, 2008

 *    Antoine P. Petitet

 *    University of Tennessee, Knoxville

 *    Innovative Computing Laboratory

 *    (C) Copyright 2000-2008 All Rights Reserved

 *

 * -- Copyright notice and Licensing terms:

 *

 * Redistribution  and  use in  source and binary forms, with or without

 * modification, are  permitted provided  that the following  conditions

 * are met:

 *

 * 1. Redistributions  of  source  code  must retain the above copyright

 * notice, this list of conditions and the following disclaimer.

 *

 * 2. Redistributions in binary form must reproduce  the above copyright

 * notice, this list of conditions,  and the following disclaimer in the

 * documentation and/or other materials provided with the distribution.

 *

 * 3. All  advertising  materials  mentioning  features  or  use of this

 * software must display the following acknowledgement:

 * This  product  includes  software  developed  at  the  University  of

 * Tennessee, Knoxville, Innovative Computing Laboratory.

 *

 * 4. The name of the  University,  the name of the  Laboratory,  or the

 * names  of  its  contributors  may  not  be used to endorse or promote

 * products  derived   from   this  software  without  specific  written

 * permission.

 *

 * -- Disclaimer:

 *

 * THIS  SOFTWARE  IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS

 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES,  INCLUDING,  BUT NOT

 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR

 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE UNIVERSITY

 * OR  CONTRIBUTORS  BE  LIABLE FOR ANY  DIRECT,  INDIRECT,  INCIDENTAL,

 * SPECIAL,  EXEMPLARY,  OR  CONSEQUENTIAL DAMAGES  (INCLUDING,  BUT NOT

 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,

 * DATA OR PROFITS; OR BUSINESS INTERRUPTION)  HOWEVER CAUSED AND ON ANY

 * THEORY OF LIABILITY, WHETHER IN CONTRACT,  STRICT LIABILITY,  OR TORT

 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE

 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

 * ---------------------------------------------------------------------

 */

/*

 * Include files

 */

#include "hpl.h"

#ifndef HPL_dgemv

#ifdef HPL_CALL_VSIPL

#ifdef STDC_HEADERS

static void HPL_dgemv0

(

   const enum HPL_TRANS       TRANS,

   const int                  M,

   const int                  N,

   const double               ALPHA,

   const double               * A,

   const int                  LDA,

   const double               * X,

   const int                  INCX,

   const double               BETA,

   double                     * Y,

   const int                  INCY

)

#else

static void HPL_dgemv0( TRANS, M, N, ALPHA, A, LDA, X, INCX, BETA, Y, INCY )

   const enum HPL_TRANS       TRANS;

   const int                  INCX, INCY, LDA, M, N;

   const double               ALPHA, BETA;

   const double               * A, * X;

   double                     * Y;

#endif

{

/*

 * .. Local Variables ..

 */

   int                        i, iaij, ix, iy, j, jaj, jx, jy;

   register double            t0;

/* ..

 * .. Executable Statements ..

 */

   if( ( M == 0 ) || ( N == 0 ) ||

       ( ( ALPHA == HPL_rzero ) && ( BETA == HPL_rone  ) ) ) return;

   if( ALPHA == HPL_rzero ) { HPL_dscal( M, BETA, Y, INCY ); return; }

   if( TRANS == HplNoTrans )

   {

      HPL_dscal( M, BETA, Y, INCY );

      for( j = 0, jaj  = 0, jx = 0; j < N; j++, jaj += LDA, jx += INCX )

      {

         t0 = ALPHA * X[jx];

         for( i = 0, iaij = jaj, iy = 0; i < M; i++, iaij += 1, iy += INCY )

         { Y[iy] += A[iaij] * t0; }

      }

   }

   else

   {

      for( j = 0, jaj  = 0, jy  = 0; j < N; j++, jaj += LDA, jy += INCY )

      {

         t0 = HPL_rzero;

         for( i = 0, iaij = jaj, ix = 0; i < M; i++, iaij += 1, ix += INCX )

         { t0 += A[iaij] * X[ix]; }

         if( BETA == HPL_rzero ) Y[jy] = ALPHA * t0;

         else                    Y[jy] = BETA * Y[jy] + ALPHA * t0;

      }

   }

}

#endif

#ifdef STDC_HEADERS

void HPL_dgemv

(

   const enum HPL_ORDER             ORDER,

   const enum HPL_TRANS             TRANS,

   const int                        M,

   const int                        N,

   const double                     ALPHA,

   const double *                   A,

   const int                        LDA,

   const double *                   X,

   const int                        INCX,

   const double                     BETA,

   double *                         Y,

   const int                        INCY

)

#else

void HPL_dgemv

( ORDER, TRANS, M, N, ALPHA, A, LDA, X, INCX, BETA, Y, INCY )

   const enum HPL_ORDER             ORDER;

   const enum HPL_TRANS             TRANS;

   const int                        M;

   const int                        N;

   const double                     ALPHA;

   const double *                   A;

   const int                        LDA;

   const double *                   X;

   const int                        INCX;

   const double                     BETA;

   double *                         Y;

   const int                        INCY;

#endif

{

/*

 * Purpose

 * =======

 *

 * HPL_dgemv performs one of the matrix-vector operations

 *

 *     y := alpha * op( A ) * x + beta * y,

 *

 *  where op( X ) is one of

 *

 *     op( X ) = X   or   op( X ) = X^T.

 *

 * where alpha and beta are scalars, x and y are vectors and  A  is an m

 * by n matrix.

 *

 * Arguments

 * =========

 *

 * ORDER   (local input)                 const enum HPL_ORDER

 *         On entry, ORDER  specifies the storage format of the operands

 *         as follows:

 *            ORDER = HplRowMajor,

 *            ORDER = HplColumnMajor.

 *

 * TRANS   (local input)                 const enum HPL_TRANS

 *         On entry,  TRANS  specifies the  operation to be performed as

 *         follows:

 *            TRANS = HplNoTrans y := alpha*A  *x + beta*y,

 *            TRANS = HplTrans   y := alpha*A^T*x + beta*y.

 *

 * M       (local input)                 const int

 *         On entry,  M  specifies  the number of rows of  the matrix A.

 *         M must be at least zero.

 *

 * N       (local input)                 const int

 *         On entry, N  specifies the number of columns of the matrix A.

 *         N must be at least zero.

 *

 * ALPHA   (local input)                 const double

 *         On entry, ALPHA specifies the scalar alpha.   When  ALPHA  is

 *         supplied as zero then  A and X  need not be set on input.

 *

 * A       (local input)                 const double *

 *         On entry,  A  points  to an array of size equal to or greater

 *         than LDA * n.  Before  entry, the leading m by n part  of the

 *         array  A  must contain the matrix coefficients.

 *

 * LDA     (local input)                 const int

 *         On entry,  LDA  specifies  the  leading  dimension  of  A  as

 *         declared  in  the  calling  (sub) program.  LDA  must  be  at

 *         least MAX(1,m).

 *

 * X       (local input)                 const double *

 *         On entry,  X  is an incremented array of dimension  at  least

 *         ( 1 + ( n - 1 ) * abs( INCX ) )  that  contains the vector x.

 *

 * INCX    (local input)                 const int

 *         On entry, INCX specifies the increment for the elements of X.

 *         INCX must not be zero.

 *

 * BETA    (local input)                 const double

 *         On entry, BETA  specifies the scalar beta.    When  ALPHA  is

 *         supplied as zero then  Y  need not be set on input.

 *

 * Y       (local input/output)          double *

 *         On entry,  Y  is an incremented array of dimension  at  least

 *         ( 1 + ( n - 1 ) * abs( INCY ) )  that  contains the vector y.

 *         Before entry with BETA non-zero, the incremented array Y must

 *         contain the vector  y.  On exit,  Y  is  overwritten  by  the

 *         updated vector y.

 *

 * INCY    (local input)                 const int

 *         On entry, INCY specifies the increment for the elements of Y.

 *         INCY must not be zero.

 *

 * ---------------------------------------------------------------------

 */

#ifdef HPL_CALL_CBLAS

   cblas_dgemv( ORDER, TRANS, M, N, ALPHA, A, LDA, X, INCX, BETA, Y, INCY );

#endif

#ifdef HPL_CALL_VSIPL

   if( ORDER == HplColumnMajor )

   {

      HPL_dgemv0( TRANS, M, N, ALPHA, A, LDA, X, INCX, BETA, Y, INCY );

   }

   else

   {

      HPL_dgemv0( ( TRANS == HplNoTrans ? HplTrans : HplNoTrans ),

                  N, M, ALPHA, A, LDA, X, INCX, BETA, Y, INCY );

   }

#endif

#ifdef HPL_CALL_FBLAS

   double                    alpha = ALPHA, beta = BETA;

#ifdef StringSunStyle

#ifdef HPL_USE_F77_INTEGER_DEF

   F77_INTEGER               IONE = 1;

#else

   int                       IONE = 1;

#endif

#endif

#ifdef StringStructVal

   F77_CHAR                  ftran;

#endif

#ifdef StringStructPtr

   F77_CHAR                  ftran;

#endif

#ifdef StringCrayStyle

   F77_CHAR                  ftran;

#endif

#ifdef HPL_USE_F77_INTEGER_DEF

   const F77_INTEGER         F77M    = M,   F77N    = N,

                             F77lda  = LDA, F77incx = INCX, F77incy = INCY;

#else

#define F77M                 M

#define F77N                 N

#define F77lda               LDA

#define F77incx              INCX

#define F77incy              INCY

#endif

   char                      ctran;

   if( ORDER == HplColumnMajor )

   {

      ctran = ( TRANS == HplNoTrans ? 'N' : 'T' );

#ifdef StringSunStyle

      F77dgemv( &ctran, &F77M, &F77N, &alpha, A, &F77lda, X, &F77incx,

                &beta, Y, &F77incy, IONE );

#endif

#ifdef StringCrayStyle

      ftran = HPL_C2F_CHAR( ctran );

      F77dgemv( ftran,  &F77M, &F77N, &alpha, A, &F77lda, X, &F77incx,

                &beta, Y, &F77incy );

#endif

#ifdef StringStructVal

      ftran.len = 1; ftran.cp = &ctran;

      F77dgemv( ftran,  &F77M, &F77N, &alpha, A, &F77lda, X, &F77incx,

                &beta, Y, &F77incy );

#endif

#ifdef StringStructPtr

      ftran.len = 1; ftran.cp = &ctran;

      F77dgemv( &ftran, &F77M, &F77N, &alpha, A, &F77lda, X, &F77incx,

                &beta, Y, &F77incy );

#endif

   }

   else

   {

      ctran = ( TRANS == HplNoTrans ? 'T' : 'N' );

#ifdef StringSunStyle

      F77dgemv( &ctran, &F77N, &F77M, &alpha, A, &F77lda, X, &F77incx,

                &beta, Y, &F77incy, IONE );

#endif

#ifdef StringCrayStyle

      ftran = HPL_C2F_CHAR( ctran );

      F77dgemv( ftran,  &F77N, &F77M, &alpha, A, &F77lda, X, &F77incx,

                &beta, Y, &F77incy );

#endif

#ifdef StringStructVal

      ftran.len = 1; ftran.cp = &ctran;

      F77dgemv( ftran,  &F77N, &F77M, &alpha, A, &F77lda, X, &F77incx,

                &beta, Y, &F77incy );

#endif

#ifdef StringStructPtr

      ftran.len = 1; ftran.cp = &ctran;

      F77dgemv( &ftran, &F77N, &F77M, &alpha, A, &F77lda, X, &F77incx,

                &beta, Y, &F77incy );

#endif

   }

#endif

#ifdef HPL_CALL_CUBLAS

   double                    alpha = ALPHA, beta = BETA;

   int                       IONE = 1;

#define CUBLASM                 M

#define CUBLASN                 N

#define CUBLASlda               LDA

#define CUBLASincx              INCX

#define CUBLASincy              INCY

   char                      ctran;

   if( ORDER == HplColumnMajor )

   {

      ctran = ( TRANS == HplNoTrans ? 'N' : 'T' );

      CUBLAS_DGEMV( &ctran, &CUBLASM, &CUBLASN,

                    &alpha, A, &CUBLASlda, X, &CUBLASincx,

                    &beta, Y, &CUBLASincy, IONE );

   }

   else

   {

      ctran = ( TRANS == HplNoTrans ? 'T' : 'N' );

      CUBLAS_DGEMV( &ctran, &CUBLASN, &CUBLASM,

                    &alpha, A, &CUBLASlda, X, &CUBLASincx,

                    &beta, Y, &CUBLASincy, IONE );

   }

#endif

/*

 * End of HPL_dgemv

 */

}

#endif