void MY_MMult( int m, int n, int k, double *a, int lda,
double *b, int ldb,
double *c, int ldc )
{
int i, j;
for ( j=0; j<n; j+=4 ){ /* Loop over the columns of C, unrolled by 4 */
for ( i=0; i<m; i+=4 ){ /* Loop over the rows of C */
/* Update C( i,j ), C( i,j+1 ), C( i,j+2 ), and C( i,j+3 ) in
one routine (four inner products) */
AddDot4x4( k, &A( i,0 ), lda, &B( 0,j ), ldb, &C( i,j ), ldc );
}
}
}
void InnerKernel( int m, int n, int k, double *a, int lda,
double *b, int ldb,
double *c, int ldc, int first_time )
{
int i, j;
double
packedA[ m * k ];
static double
packedB[ kc*nb ]; /* Note: using a static buffer is not thread safe... */
for ( j=0; j<n; j+=4 ){ /* Loop over the columns of C, unrolled by 4 */
if ( first_time )
PackMatrixB( k, &B( 0, j ), ldb, &packedB[ j*k ] );
for ( i=0; i<m; i+=4 ){ /* Loop over the rows of C */
/* Update C( i,j ), C( i,j+1 ), C( i,j+2 ), and C( i,j+3 ) in
one routine (four inner products) */
if ( j == 0 )
PackMatrixA( k, &A( i, 0 ), lda, &packedA[ i*k ] );
AddDot4x4( k, &packedA[ i*k ], 4, &packedB[ j*k ], k, &C( i,j ), ldc );
}
}
}
