1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
|
/*
-- MAGMA (version 2.9.0) --
Univ. of Tennessee, Knoxville
Univ. of California, Berkeley
Univ. of Colorado, Denver
@date January 2025
@generated from testing/testing_zhemv_batched.cpp, normal z -> c, Wed Jan 22 14:40:41 2025
@author Mark Gates
@author Azzam Haidar
@author Tingxing Dong
*/
// includes, system
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <math.h>
// includes, project
#include "flops.h"
#include "magma_v2.h"
#include "magma_lapack.h"
#include "testings.h"
#if defined(_OPENMP)
#include <omp.h>
#include "../control/magma_threadsetting.h" // internal header
#endif
/* ////////////////////////////////////////////////////////////////////////////
-- Testing chemv_batched
*/
int main( int argc, char** argv)
{
TESTING_CHECK( magma_init() );
magma_print_environment();
real_Double_t gflops, magma_perf, magma_time, cpu_perf, cpu_time;
float error, magma_error, normalize, work[1];
magma_int_t N, lda, ldda;
magma_int_t sizeA, sizeX, sizeY;
magma_int_t incx = 1;
magma_int_t incy = 1;
magma_int_t ione = 1;
magma_int_t ISEED[4] = {0,0,0,1};
int status = 0;
magma_int_t batchCount;
magmaFloatComplex *h_A, *h_X, *h_Y, *h_Ymagma;
magmaFloatComplex *d_A, *d_X, *d_Y;
magmaFloatComplex **d_A_array = NULL;
magmaFloatComplex **d_X_array = NULL;
magmaFloatComplex **d_Y_array = NULL;
magmaFloatComplex c_neg_one = MAGMA_C_NEG_ONE;
magmaFloatComplex alpha = MAGMA_C_MAKE( 0.29, -0.86 );
magmaFloatComplex beta = MAGMA_C_MAKE( -0.48, 0.38 );
magma_opts opts( MagmaOptsBatched );
opts.parse_opts( argc, argv );
opts.lapack |= opts.check;
batchCount = opts.batchcount;
float *Anorm, *Xnorm, *Ynorm;
TESTING_CHECK( magma_smalloc_cpu( &Anorm, batchCount ));
TESTING_CHECK( magma_smalloc_cpu( &Xnorm, batchCount ));
TESTING_CHECK( magma_smalloc_cpu( &Ynorm, batchCount ));
TESTING_CHECK( magma_malloc( (void**) &d_A_array, batchCount * sizeof(magmaFloatComplex*) ));
TESTING_CHECK( magma_malloc( (void**) &d_X_array, batchCount * sizeof(magmaFloatComplex*) ));
TESTING_CHECK( magma_malloc( (void**) &d_Y_array, batchCount * sizeof(magmaFloatComplex*) ));
// See testing_cgemm about tolerance.
float eps = lapackf77_slamch("E");
float tol = 3*eps;
printf("%% uplo = %s\n", lapack_uplo_const(opts.uplo) );
printf("%% BatchCount N MAGMA Gflop/s (ms) CPU Gflop/s (ms) MAGMA error\n");
printf("%%=======================================================================\n");
for( int itest = 0; itest < opts.ntest; ++itest ) {
for( int iter = 0; iter < opts.niter; ++iter ) {
N = opts.nsize[itest];
lda = N;
ldda = magma_roundup( N, opts.align ); // multiple of 32 by default
gflops = FLOPS_CHEMV( N ) / 1e9 * batchCount;
sizeA = lda*N*batchCount;
sizeX = incx*N*batchCount;
sizeY = incy*N*batchCount;
TESTING_CHECK( magma_cmalloc_cpu( &h_A, sizeA ));
TESTING_CHECK( magma_cmalloc_cpu( &h_X, sizeX ));
TESTING_CHECK( magma_cmalloc_cpu( &h_Y, sizeY ));
TESTING_CHECK( magma_cmalloc_cpu( &h_Ymagma, sizeY ));
TESTING_CHECK( magma_cmalloc( &d_A, ldda*N*batchCount ));
TESTING_CHECK( magma_cmalloc( &d_X, sizeX ));
TESTING_CHECK( magma_cmalloc( &d_Y, sizeY ));
/* Initialize the matrices */
lapackf77_clarnv( &ione, ISEED, &sizeA, h_A );
lapackf77_clarnv( &ione, ISEED, &sizeX, h_X );
lapackf77_clarnv( &ione, ISEED, &sizeY, h_Y );
/* set the opposite triangular part to NAN to check */
magma_int_t N1 = N-1;
for(magma_int_t i = 0; i < batchCount; i++){
magmaFloatComplex* Ai = h_A + i * N * lda;
if ( opts.uplo == MagmaUpper ) {
lapackf77_claset( "Lower", &N1, &N1, &MAGMA_C_NAN, &MAGMA_C_NAN, &Ai[1], &lda );
}
else {
lapackf77_claset( "Upper", &N1, &N1, &MAGMA_C_NAN, &MAGMA_C_NAN, &Ai[lda], &lda );
}
}
// Compute norms for error computation
for (int s = 0; s < batchCount; ++s) {
Anorm[s] = safe_lapackf77_clanhe( "F", lapack_uplo_const(opts.uplo), &N, &h_A[s*lda*N], &lda, work );
Xnorm[s] = lapackf77_clange( "F", &ione, &N, &h_X[s*N*incx], &incx, work );
Ynorm[s] = lapackf77_clange( "F", &ione, &N, &h_Y[s*N*incy], &incy, work );
}
/* =====================================================================
Performs operation using MAGMABLAS
=================================================================== */
magma_csetmatrix( N, N*batchCount, h_A, lda, d_A, ldda, opts.queue );
magma_csetvector( N*batchCount, h_X, incx, d_X, incx, opts.queue );
magma_csetvector( N*batchCount, h_Y, incy, d_Y, incy, opts.queue );
magma_cset_pointer( d_A_array, d_A, ldda, 0, 0, ldda*N, batchCount, opts.queue );
magma_cset_pointer( d_X_array, d_X, 1, 0, 0, incx*N, batchCount, opts.queue );
magma_cset_pointer( d_Y_array, d_Y, 1, 0, 0, incy*N, batchCount, opts.queue );
magma_time = magma_sync_wtime( opts.queue );
magmablas_chemv_batched(opts.uplo, N,
alpha, d_A_array, ldda,
d_X_array, incx,
beta, d_Y_array, incy, batchCount, opts.queue);
magma_time = magma_sync_wtime( opts.queue ) - magma_time;
magma_perf = gflops / magma_time;
magma_cgetvector( N*batchCount, d_Y, incy, h_Ymagma, incy, opts.queue );
/* =====================================================================
Performs operation using CPU BLAS
=================================================================== */
if ( opts.lapack ) {
cpu_time = magma_wtime();
#if !defined (BATCHED_DISABLE_PARCPU) && defined(_OPENMP)
magma_int_t nthreads = magma_get_lapack_numthreads();
magma_set_lapack_numthreads(1);
magma_set_omp_numthreads(nthreads);
#pragma omp parallel for schedule(dynamic)
#endif
for (int i=0; i < batchCount; i++)
{
blasf77_chemv( lapack_uplo_const(opts.uplo), &N,
&alpha, h_A + i*lda*N, &lda,
h_X + i*N*incx, &incx,
&beta, h_Y + i*N*incy, &incy );
}
#if !defined (BATCHED_DISABLE_PARCPU) && defined(_OPENMP)
magma_set_lapack_numthreads(nthreads);
#endif
cpu_time = magma_wtime() - cpu_time;
cpu_perf = gflops / cpu_time;
}
/* =====================================================================
Check the result
=================================================================== */
if ( opts.lapack ) {
// compute error compared lapack
// error = |dY - Y| / (gamma_{k+2}|A||X| + gamma_2|Yin|); k = m
magma_error = 0;
for (int s=0; s < batchCount; s++){
normalize = sqrt(float(N+2))*Anorm[s]*Xnorm[s] + 2*Ynorm[s];
if (normalize == 0)
normalize = 1;
blasf77_caxpy( &N, &c_neg_one, &h_Y[s*N*incy], &incy, &h_Ymagma[s*N*incy], &incy );
error = lapackf77_clange( "F", &ione, &N, &h_Ymagma[s*N*incy], &incy, work )
/ normalize;
magma_error = magma_max_nan( error, magma_error );
}
bool okay = (magma_error < tol);
status += ! okay;
printf(" %10lld %5lld %7.2f (%7.2f) %7.2f (%7.2f) %8.2e %s\n",
(long long) batchCount, (long long) N,
magma_perf, 1000.*magma_time,
cpu_perf, 1000.*cpu_time,
magma_error, (okay ? "ok" : "failed"));
}
else {
printf(" %10lld %5lld %7.2f (%7.2f) --- ( --- ) ---\n",
(long long) batchCount,(long long) N,
magma_perf, 1000.*magma_time);
}
magma_free_cpu( h_A );
magma_free_cpu( h_X );
magma_free_cpu( h_Y );
magma_free_cpu( h_Ymagma );
magma_free( d_A );
magma_free( d_X );
magma_free( d_Y );
fflush( stdout);
}
if ( opts.niter > 1 ) {
printf( "\n" );
}
}
magma_free_cpu( Anorm );
magma_free_cpu( Xnorm );
magma_free_cpu( Ynorm );
magma_free( d_A_array );
magma_free( d_X_array );
magma_free( d_Y_array );
opts.cleanup();
TESTING_CHECK( magma_finalize() );
return status;
}
|
