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
|
// Copyright (c) 2017-2023, University of Tennessee. All rights reserved.
// SPDX-License-Identifier: BSD-3-Clause
// This program is free software: you can redistribute it and/or modify it under
// the terms of the BSD 3-Clause license. See the accompanying LICENSE file.
#include "test.hh"
#include "lapack.hh"
#include "lapack/flops.hh"
#include "print_matrix.hh"
#include "error.hh"
#include "lapacke_wrappers.hh"
#include <vector>
// -----------------------------------------------------------------------------
template< typename scalar_t >
void test_lantr_work( Params& params, bool run )
{
using real_t = blas::real_type< scalar_t >;
// get & mark input values
lapack::Norm norm = params.norm();
lapack::Uplo uplo = params.uplo();
lapack::Diag diag = params.diag();
int64_t m = params.dim.m();
int64_t n = params.dim.n();
int64_t align = params.align();
int64_t verbose = params.verbose();
params.matrix.mark();
// mark non-standard output values
params.ref_time();
//params.ref_gflops();
//params.gflops();
params.msg();
if (! run)
return;
// Any m, n actually works, despite the LAPACK documentation.
// ---------- setup
int64_t lda = roundup( blas::max( m, 1 ), align );
size_t size_A = (size_t) lda * n;
std::vector< scalar_t > A( size_A );
lapack::generate_matrix( params.matrix, m, n, &A[0], lda );
if (verbose >= 1) {
printf( "\n"
"A m=%5lld, n=%5lld, lda=%5lld\n",
llong( m ), llong( n ), llong( lda ) );
}
if (verbose >= 2) {
printf( "A = " ); print_matrix( m, n, &A[0], lda );
}
// ---------- run test
testsweeper::flush_cache( params.cache() );
double time = testsweeper::get_wtime();
real_t norm_tst = lapack::lantr( norm, uplo, diag, m, n, &A[0], lda );
time = testsweeper::get_wtime() - time;
params.time() = time;
//double gflop = lapack::Gflop< scalar_t >::lantr( norm, diag, m, n );
//params.gflops() = gflop / time;
if (verbose >= 1) {
printf( "norm_tst = %.8e\n", norm_tst );
}
if (params.ref() == 'y' || params.check() == 'y') {
// ---------- run reference
testsweeper::flush_cache( params.cache() );
time = testsweeper::get_wtime();
real_t norm_ref = LAPACKE_lantr( to_char( norm ), to_char( uplo ), to_char( diag ), m, n, &A[0], lda );
time = testsweeper::get_wtime() - time;
params.ref_time() = time;
//params.ref_gflops() = gflop / time;
if (verbose >= 1) {
printf( "norm_ref = %.8e\n", norm_ref );
}
// ---------- check error compared to reference
real_t tol = 3 * std::numeric_limits< real_t >::epsilon();
real_t normalize = 1;
if (norm == lapack::Norm::Max && ! blas::is_complex< scalar_t >::value) {
// max-norm depends on only one element, so in real there should be
// zero error, but in complex there's error in abs().
tol = 0;
}
else if (norm == lapack::Norm::One)
normalize = sqrt( real_t(m) );
else if (norm == lapack::Norm::Inf)
normalize = sqrt( real_t(n) );
else if (norm == lapack::Norm::Fro)
normalize = sqrt( real_t(m)*n );
real_t error = std::abs( norm_tst - norm_ref ) / normalize;
if (norm_ref != 0)
error /= norm_ref;
params.error() = error;
params.okay() = (error <= tol);
}
}
// -----------------------------------------------------------------------------
void test_lantr( Params& params, bool run )
{
switch (params.datatype()) {
case testsweeper::DataType::Single:
test_lantr_work< float >( params, run );
break;
case testsweeper::DataType::Double:
test_lantr_work< double >( params, run );
break;
case testsweeper::DataType::SingleComplex:
test_lantr_work< std::complex<float> >( params, run );
break;
case testsweeper::DataType::DoubleComplex:
test_lantr_work< std::complex<double> >( params, run );
break;
default:
throw std::runtime_error( "unknown datatype" );
break;
}
}
|
