2
* Copyright (c) 2003, 2006 Matteo Frigo
3
* Copyright (c) 2003, 2006 Massachusetts Institute of Technology
5
* This program is free software; you can redistribute it and/or modify
6
* it under the terms of the GNU General Public License as published by
7
* the Free Software Foundation; either version 2 of the License, or
8
* (at your option) any later version.
10
* This program is distributed in the hope that it will be useful,
11
* but WITHOUT ANY WARRANTY; without even the implied warranty of
12
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13
* GNU General Public License for more details.
15
* You should have received a copy of the GNU General Public License
16
* along with this program; if not, write to the Free Software
17
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
21
/* This file was automatically generated --- DO NOT EDIT */
22
/* Generated on Sat Jul 1 13:58:08 EDT 2006 */
24
#include "codelet-dft.h"
28
/* Generated by: ../../../genfft/gen_twiddle -fma -reorder-insns -schedule-for-pipeline -compact -variables 4 -pipeline-latency 4 -n 2 -name t1_2 -include t.h */
31
* This function contains 6 FP additions, 4 FP multiplications,
32
* (or, 4 additions, 2 multiplications, 2 fused multiply/add),
33
* 11 stack variables, and 8 memory accesses
37
* $Id: algsimp.ml,v 1.9 2006-02-12 23:34:12 athena Exp $
38
* $Id: fft.ml,v 1.4 2006-01-05 03:04:27 stevenj Exp $
39
* $Id: gen_twiddle.ml,v 1.24 2006-02-12 23:34:12 athena Exp $
44
static const R *t1_2(R *ri, R *ii, const R *W, stride ios, INT m, INT dist)
47
for (i = m; i > 0; i = i - 1, ri = ri + dist, ii = ii + dist, W = W + 2, MAKE_VOLATILE_STRIDE(ios)) {
48
E T1, Ta, T3, T6, T2, T5;
59
T9 = FNMS(T5, T3, T8);
62
ii[WS(ios, 1)] = Ta - T9;
64
ri[WS(ios, 1)] = T1 - T7;
70
static const tw_instr twinstr[] = {
75
static const ct_desc desc = { 2, "t1_2", twinstr, &GENUS, {4, 2, 2, 0}, 0, 0, 0 };
77
void X(codelet_t1_2) (planner *p) {
78
X(kdft_dit_register) (p, t1_2, &desc);
82
/* Generated by: ../../../genfft/gen_twiddle -compact -variables 4 -pipeline-latency 4 -n 2 -name t1_2 -include t.h */
85
* This function contains 6 FP additions, 4 FP multiplications,
86
* (or, 4 additions, 2 multiplications, 2 fused multiply/add),
87
* 9 stack variables, and 8 memory accesses
91
* $Id: algsimp.ml,v 1.9 2006-02-12 23:34:12 athena Exp $
92
* $Id: fft.ml,v 1.4 2006-01-05 03:04:27 stevenj Exp $
93
* $Id: gen_twiddle.ml,v 1.24 2006-02-12 23:34:12 athena Exp $
98
static const R *t1_2(R *ri, R *ii, const R *W, stride ios, INT m, INT dist)
101
for (i = m; i > 0; i = i - 1, ri = ri + dist, ii = ii + dist, W = W + 2, MAKE_VOLATILE_STRIDE(ios)) {
111
T6 = FMA(T2, T3, T4 * T5);
112
T7 = FNMS(T4, T3, T2 * T5);
114
ri[WS(ios, 1)] = T1 - T6;
115
ii[WS(ios, 1)] = T8 - T7;
122
static const tw_instr twinstr[] = {
127
static const ct_desc desc = { 2, "t1_2", twinstr, &GENUS, {4, 2, 2, 0}, 0, 0, 0 };
129
void X(codelet_t1_2) (planner *p) {
130
X(kdft_dit_register) (p, t1_2, &desc);
132
#endif /* HAVE_FMA */