root / src / pauxil / HPL_dlaswp00N.c @ 1
Historique | Voir | Annoter | Télécharger (8,26 ko)
| 1 |
/*
|
|---|---|
| 2 |
* -- High Performance Computing Linpack Benchmark (HPL)
|
| 3 |
* HPL - 2.0 - September 10, 2008
|
| 4 |
* Antoine P. Petitet
|
| 5 |
* University of Tennessee, Knoxville
|
| 6 |
* Innovative Computing Laboratory
|
| 7 |
* (C) Copyright 2000-2008 All Rights Reserved
|
| 8 |
*
|
| 9 |
* -- Copyright notice and Licensing terms:
|
| 10 |
*
|
| 11 |
* Redistribution and use in source and binary forms, with or without
|
| 12 |
* modification, are permitted provided that the following conditions
|
| 13 |
* are met:
|
| 14 |
*
|
| 15 |
* 1. Redistributions of source code must retain the above copyright
|
| 16 |
* notice, this list of conditions and the following disclaimer.
|
| 17 |
*
|
| 18 |
* 2. Redistributions in binary form must reproduce the above copyright
|
| 19 |
* notice, this list of conditions, and the following disclaimer in the
|
| 20 |
* documentation and/or other materials provided with the distribution.
|
| 21 |
*
|
| 22 |
* 3. All advertising materials mentioning features or use of this
|
| 23 |
* software must display the following acknowledgement:
|
| 24 |
* This product includes software developed at the University of
|
| 25 |
* Tennessee, Knoxville, Innovative Computing Laboratory.
|
| 26 |
*
|
| 27 |
* 4. The name of the University, the name of the Laboratory, or the
|
| 28 |
* names of its contributors may not be used to endorse or promote
|
| 29 |
* products derived from this software without specific written
|
| 30 |
* permission.
|
| 31 |
*
|
| 32 |
* -- Disclaimer:
|
| 33 |
*
|
| 34 |
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
| 35 |
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
| 36 |
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
| 37 |
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE UNIVERSITY
|
| 38 |
* OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
| 39 |
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
| 40 |
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
| 41 |
* DATA OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
| 42 |
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
| 43 |
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
| 44 |
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
| 45 |
* ---------------------------------------------------------------------
|
| 46 |
*/
|
| 47 |
/*
|
| 48 |
* Include files
|
| 49 |
*/
|
| 50 |
#include "hpl.h" |
| 51 |
/*
|
| 52 |
* Define default value for unrolling factor
|
| 53 |
*/
|
| 54 |
#ifndef HPL_LASWP00N_DEPTH
|
| 55 |
#define HPL_LASWP00N_DEPTH 32 |
| 56 |
#define HPL_LASWP00N_LOG2_DEPTH 5 |
| 57 |
#endif
|
| 58 |
|
| 59 |
#ifdef STDC_HEADERS
|
| 60 |
void HPL_dlaswp00N
|
| 61 |
( |
| 62 |
const int M, |
| 63 |
const int N, |
| 64 |
double * A,
|
| 65 |
const int LDA, |
| 66 |
const int * IPIV |
| 67 |
) |
| 68 |
#else
|
| 69 |
void HPL_dlaswp00N
|
| 70 |
( M, N, A, LDA, IPIV ) |
| 71 |
const int M; |
| 72 |
const int N; |
| 73 |
double * A;
|
| 74 |
const int LDA; |
| 75 |
const int * IPIV; |
| 76 |
#endif
|
| 77 |
{
|
| 78 |
/*
|
| 79 |
* Purpose
|
| 80 |
* =======
|
| 81 |
*
|
| 82 |
* HPL_dlaswp00N performs a series of local row interchanges on a matrix
|
| 83 |
* A. One row interchange is initiated for rows 0 through M-1 of A.
|
| 84 |
*
|
| 85 |
* Arguments
|
| 86 |
* =========
|
| 87 |
*
|
| 88 |
* M (local input) const int
|
| 89 |
* On entry, M specifies the number of rows of the array A to be
|
| 90 |
* interchanged. M must be at least zero.
|
| 91 |
*
|
| 92 |
* N (local input) const int
|
| 93 |
* On entry, N specifies the number of columns of the array A.
|
| 94 |
* N must be at least zero.
|
| 95 |
*
|
| 96 |
* A (local input/output) double *
|
| 97 |
* On entry, A points to an array of dimension (LDA,N) to which
|
| 98 |
* the row interchanges will be applied. On exit, the permuted
|
| 99 |
* matrix.
|
| 100 |
*
|
| 101 |
* LDA (local input) const int
|
| 102 |
* On entry, LDA specifies the leading dimension of the array A.
|
| 103 |
* LDA must be at least MAX(1,M).
|
| 104 |
*
|
| 105 |
* IPIV (local input) const int *
|
| 106 |
* On entry, IPIV is an array of size M that contains the
|
| 107 |
* pivoting information. For k in [0..M), IPIV[k]=IROFF + l
|
| 108 |
* implies that local rows k and l are to be interchanged.
|
| 109 |
*
|
| 110 |
* ---------------------------------------------------------------------
|
| 111 |
*/
|
| 112 |
/*
|
| 113 |
* .. Local Variables ..
|
| 114 |
*/
|
| 115 |
register double r; |
| 116 |
double * a0, * a1;
|
| 117 |
const int incA = (int)( (unsigned int)(LDA) << |
| 118 |
HPL_LASWP00N_LOG2_DEPTH ); |
| 119 |
int ip, nr, nu;
|
| 120 |
register int i, j; |
| 121 |
/* ..
|
| 122 |
* .. Executable Statements ..
|
| 123 |
*/
|
| 124 |
if( ( M <= 0 ) || ( N <= 0 ) ) return; |
| 125 |
|
| 126 |
nr = N - ( nu = (int)( ( (unsigned int)(N) >> HPL_LASWP00N_LOG2_DEPTH ) |
| 127 |
<< HPL_LASWP00N_LOG2_DEPTH ) ); |
| 128 |
|
| 129 |
for( j = 0; j < nu; j += HPL_LASWP00N_DEPTH, A += incA ) |
| 130 |
{
|
| 131 |
for( i = 0; i < M; i++ ) |
| 132 |
{
|
| 133 |
if( i != ( ip = IPIV[i] ) )
|
| 134 |
{
|
| 135 |
a0 = A + i; a1 = A + ip; |
| 136 |
|
| 137 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 138 |
#if ( HPL_LASWP00N_DEPTH > 1 ) |
| 139 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 140 |
#endif
|
| 141 |
#if ( HPL_LASWP00N_DEPTH > 2 ) |
| 142 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 143 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 144 |
#endif
|
| 145 |
#if ( HPL_LASWP00N_DEPTH > 4 ) |
| 146 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 147 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 148 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 149 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 150 |
#endif
|
| 151 |
#if ( HPL_LASWP00N_DEPTH > 8 ) |
| 152 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 153 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 154 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 155 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 156 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 157 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 158 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 159 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 160 |
#endif
|
| 161 |
#if ( HPL_LASWP00N_DEPTH > 16 ) |
| 162 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 163 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 164 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 165 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 166 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 167 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 168 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 169 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 170 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 171 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 172 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 173 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 174 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 175 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 176 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 177 |
r = *a0; *a0 = *a1; *a1 = r; a0 += LDA; a1 += LDA; |
| 178 |
#endif
|
| 179 |
} |
| 180 |
} |
| 181 |
} |
| 182 |
|
| 183 |
if( nr > 0 ) |
| 184 |
{
|
| 185 |
for( i = 0; i < M; i++ ) |
| 186 |
{
|
| 187 |
if( i != ( ip = IPIV[i] ) )
|
| 188 |
{
|
| 189 |
a0 = A + i; a1 = A + ip; |
| 190 |
for( j = 0; j < nr; j++, a0 += LDA, a1 += LDA ) |
| 191 |
{ r = *a0; *a0 = *a1; *a1 = r; }
|
| 192 |
} |
| 193 |
} |
| 194 |
} |
| 195 |
/*
|
| 196 |
* End of HPL_dlaswp00N
|
| 197 |
*/
|
| 198 |
} |