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
|
/*
* Copyright (c) 2003, 2006 Matteo Frigo
* Copyright (c) 2003, 2006 Massachusetts Institute of Technology
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*
*/
/* This file was automatically generated --- DO NOT EDIT */
/* Generated on Sat Jul 1 22:36:34 EDT 2006 */
#include "codelet-dft.h"
#ifdef HAVE_FMA
/* Generated by: ../../../genfft/gen_twidsq_c -fma -reorder-insns -schedule-for-pipeline -simd -compact -variables 4 -pipeline-latency 8 -n 2 -dif -name q1fv_2 -include q1f.h */
/*
* This function contains 6 FP additions, 4 FP multiplications,
* (or, 6 additions, 4 multiplications, 0 fused multiply/add),
* 8 stack variables, and 8 memory accesses
*/
/*
* Generator Id's :
* $Id: algsimp.ml,v 1.9 2006-02-12 23:34:12 athena Exp $
* $Id: fft.ml,v 1.4 2006-01-05 03:04:27 stevenj Exp $
* $Id: gen_twidsq_c.ml,v 1.8 2006-02-12 23:34:12 athena Exp $
*/
#include "q1f.h"
static const R *q1fv_2(R *ri, R *ii, const R *W, stride is, stride vs, INT m, INT dist)
{
INT i;
R *x;
x = ri;
for (i = 0; i < m; i = i + VL, x = x + (VL * dist), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(vs)) {
V T1, T2, T4, T5, T3, T6;
T1 = LD(&(x[0]), dist, &(x[0]));
T2 = LD(&(x[WS(is, 1)]), dist, &(x[WS(is, 1)]));
T4 = LD(&(x[WS(vs, 1)]), dist, &(x[WS(vs, 1)]));
T5 = LD(&(x[WS(vs, 1) + WS(is, 1)]), dist, &(x[WS(vs, 1) + WS(is, 1)]));
ST(&(x[0]), VADD(T1, T2), dist, &(x[0]));
T3 = BYTWJ(&(W[0]), VSUB(T1, T2));
ST(&(x[WS(is, 1)]), VADD(T4, T5), dist, &(x[WS(is, 1)]));
T6 = BYTWJ(&(W[0]), VSUB(T4, T5));
ST(&(x[WS(vs, 1)]), T3, dist, &(x[WS(vs, 1)]));
ST(&(x[WS(vs, 1) + WS(is, 1)]), T6, dist, &(x[WS(vs, 1) + WS(is, 1)]));
}
return W;
}
static const tw_instr twinstr[] = {
VTW(1),
{TW_NEXT, VL, 0}
};
static const ct_desc desc = { 2, "q1fv_2", twinstr, &GENUS, {6, 4, 0, 0}, 0, 0, 0 };
void X(codelet_q1fv_2) (planner *p) {
X(kdft_difsq_register) (p, q1fv_2, &desc);
}
#else /* HAVE_FMA */
/* Generated by: ../../../genfft/gen_twidsq_c -simd -compact -variables 4 -pipeline-latency 8 -n 2 -dif -name q1fv_2 -include q1f.h */
/*
* This function contains 6 FP additions, 4 FP multiplications,
* (or, 6 additions, 4 multiplications, 0 fused multiply/add),
* 8 stack variables, and 8 memory accesses
*/
/*
* Generator Id's :
* $Id: algsimp.ml,v 1.9 2006-02-12 23:34:12 athena Exp $
* $Id: fft.ml,v 1.4 2006-01-05 03:04:27 stevenj Exp $
* $Id: gen_twidsq_c.ml,v 1.8 2006-02-12 23:34:12 athena Exp $
*/
#include "q1f.h"
static const R *q1fv_2(R *ri, R *ii, const R *W, stride is, stride vs, INT m, INT dist)
{
INT i;
R *x;
x = ri;
for (i = 0; i < m; i = i + VL, x = x + (VL * dist), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(vs)) {
V T1, T2, T3, T4, T5, T6;
T1 = LD(&(x[0]), dist, &(x[0]));
T2 = LD(&(x[WS(is, 1)]), dist, &(x[WS(is, 1)]));
T3 = BYTWJ(&(W[0]), VSUB(T1, T2));
T4 = LD(&(x[WS(vs, 1)]), dist, &(x[WS(vs, 1)]));
T5 = LD(&(x[WS(vs, 1) + WS(is, 1)]), dist, &(x[WS(vs, 1) + WS(is, 1)]));
T6 = BYTWJ(&(W[0]), VSUB(T4, T5));
ST(&(x[WS(vs, 1)]), T3, dist, &(x[WS(vs, 1)]));
ST(&(x[WS(vs, 1) + WS(is, 1)]), T6, dist, &(x[WS(vs, 1) + WS(is, 1)]));
ST(&(x[0]), VADD(T1, T2), dist, &(x[0]));
ST(&(x[WS(is, 1)]), VADD(T4, T5), dist, &(x[WS(is, 1)]));
}
return W;
}
static const tw_instr twinstr[] = {
VTW(1),
{TW_NEXT, VL, 0}
};
static const ct_desc desc = { 2, "q1fv_2", twinstr, &GENUS, {6, 4, 0, 0}, 0, 0, 0 };
void X(codelet_q1fv_2) (planner *p) {
X(kdft_difsq_register) (p, q1fv_2, &desc);
}
#endif /* HAVE_FMA */
|
