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
|
/*
-- MAGMA (version 2.9.0) --
Univ. of Tennessee, Knoxville
Univ. of California, Berkeley
Univ. of Colorado, Denver
@date January 2025
@generated from magmablas/zgemv_vbatched.cpp, normal z -> c, Wed Jan 22 14:42:08 2025
@author Ahmad Abdelfattah
*/
#include "magma_internal.h"
#include "commonblas_c.h"
#define PRECISION_c
/******************************************************************************/
extern "C" void
magmablas_cgemv_vbatched_max(
magma_trans_t trans, magma_int_t* m, magma_int_t* n,
magmaFloatComplex alpha,
magmaFloatComplex_ptr dA_array[], magma_int_t* ldda,
magmaFloatComplex_ptr dx_array[], magma_int_t* incx,
magmaFloatComplex beta,
magmaFloatComplex_ptr dy_array[], magma_int_t* incy,
magma_int_t batchCount,
magma_int_t max_m, magma_int_t max_n, magma_queue_t queue)
{
magma_int_t info = 0;
info = magma_gemv_vbatched_checker( trans, m, n, ldda, incx, incy, batchCount, queue );
if (info != 0) {
magma_xerbla( __func__, -(info) );
return;
}
magmablas_cgemv_vbatched_max_nocheck(
trans,
m, n,
alpha, dA_array, ldda,
dx_array, incx,
beta, dy_array, incy,
batchCount, max_m, max_n, queue);
}
/******************************************************************************/
extern "C" void
magmablas_cgemv_vbatched_nocheck(
magma_trans_t trans, magma_int_t* m, magma_int_t* n,
magmaFloatComplex alpha,
magmaFloatComplex_ptr dA_array[], magma_int_t* ldda,
magmaFloatComplex_ptr dx_array[], magma_int_t* incx,
magmaFloatComplex beta,
magmaFloatComplex_ptr dy_array[], magma_int_t* incy,
magma_int_t batchCount, magma_queue_t queue)
{
// compute the max. dimensions
magma_imax_size_2(m, n, batchCount, queue);
magma_int_t max_m, max_n;
magma_igetvector_async(1, &m[batchCount], 1, &max_m, 1, queue);
magma_igetvector_async(1, &n[batchCount], 1, &max_n, 1, queue);
magma_queue_sync( queue );
magmablas_cgemv_vbatched_max_nocheck(
trans,
m, n,
alpha, dA_array, ldda,
dx_array, incx,
beta, dy_array, incy,
batchCount, max_m, max_n, queue);
}
/***************************************************************************//**
Purpose
-------
CGEMV performs one of the matrix-vector operations
y := alpha*A*x + beta*y, or
y := alpha*A**T*x + beta*y, or
y := alpha*A**H*x + beta*y,
where alpha and beta are scalars, x and y are vectors and A is an
m by n matrix.
Arguments
----------
@param[in]
trans magma_trans_t
On entry, TRANS specifies the operation to be performed as
follows:
- = MagmaNoTrans: y := alpha*A *x + beta*y
- = MagmaTrans: y := alpha*A^T*x + beta*y
- = MagmaConjTrans: y := alpha*A^H*x + beta*y
@param[in]
m Array of integers, dimension (batchCount + 1).
On entry, each INTEGER M specifies the number of rows of each matrix A.
The last element of the array is used internally by the routine.
@param[in]
n Array of integers, dimension (batchCount + 1).
On entry, each INTEGER N specifies the number of columns of each matrix A
The last element of the array is used internally by the routine.
@param[in]
alpha COMPLEX
On entry, ALPHA specifies the scalar alpha.
@param[in]
dA_array Array of pointers, dimension (batchCount).
Each is a COMPLEX array A of DIMENSION ( LDDA, N ) on the GPU
@param[in]
ldda Array of integers, dimension (batchCount + 1).
Each INTEGER LDDA specifies the leading dimension of each matrix A.
@param[in]
dx_array Array of pointers, dimension (batchCount).
Each is a COMPLEX array of dimension
N if trans == MagmaNoTrans
M if trans == MagmaTrans or MagmaConjTrans
@param[in]
incx Array of integers, dimension (batchCount + 1).
Each integer specifies the increment for the elements of each vector X.
INCX must not be zero.
The last element of the array is used internally by the routine.
@param[in]
beta COMPLEX
On entry, ALPHA specifies the scalar beta. When BETA is
supplied as zero then Y need not be set on input.
@param[out]
dy_array Array of pointers, dimension (batchCount).
Each is a COMPLEX array of dimension
M if trans == MagmaNoTrans
N if trans == MagmaTrans or MagmaConjTrans
@param[in]
incy Array of integers, dimension (batchCount + 1).
Each integer specifies the increment for the elements of each vector Y.
INCY must not be zero.
The last element of the array is used internally by the routine.
@param[in]
batchCount INTEGER
The number of matrices to operate on.
@param[in]
queue magma_queue_t
Queue to execute in.
@ingroup magma_gemv_batched
*******************************************************************************/
extern "C" void
magmablas_cgemv_vbatched(
magma_trans_t trans, magma_int_t* m, magma_int_t* n,
magmaFloatComplex alpha,
magmaFloatComplex_ptr dA_array[], magma_int_t* ldda,
magmaFloatComplex_ptr dx_array[], magma_int_t* incx,
magmaFloatComplex beta,
magmaFloatComplex_ptr dy_array[], magma_int_t* incy,
magma_int_t batchCount, magma_queue_t queue)
{
magma_int_t info = 0;
info = magma_gemv_vbatched_checker( trans, m, n, ldda, incx, incy, batchCount, queue );
if (info != 0) {
magma_xerbla( __func__, -(info) );
return;
}
// compute the max. dimensions
magma_imax_size_2(m, n, batchCount, queue);
magma_int_t max_m, max_n;
magma_igetvector_async(1, &m[batchCount], 1, &max_m, 1, queue);
magma_igetvector_async(1, &n[batchCount], 1, &max_n, 1, queue);
magma_queue_sync( queue );
magmablas_cgemv_vbatched_max_nocheck(
trans,
m, n,
alpha, dA_array, ldda,
dx_array, incx,
beta, dy_array, incy,
batchCount, max_m, max_n, queue);
}
|
