1 /*
2 * Copyright (c) 1997-1999, 2003 Massachusetts Institute of Technology
3 *
4 * This program is free software; you can redistribute it and/or modify
5 * it under the terms of the GNU General Public License as published by
6 * the Free Software Foundation; either version 2 of the License, or
7 * (at your option) any later version.
8 *
9 * This program is distributed in the hope that it will be useful,
10 * but WITHOUT ANY WARRANTY; without even the implied warranty of
11 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 * GNU General Public License for more details.
13 *
14 * You should have received a copy of the GNU General Public License
15 * along with this program; if not, write to the Free Software
16 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17 *
18 */
19
20 /* This file was automatically generated --- DO NOT EDIT */
21 /* Generated on Mon Mar 24 02:07:49 EST 2003 */
22
23 #include "fftw-int.h"
24 #include "fftw.h"
25
26 /* Generated by: /homee/stevenj/cvs/fftw/gensrc/genfft -magic-alignment-check -magic-twiddle-load-all -magic-variables 4 -magic-loopi -hc2hc-forward 4 */
27
28 /*
29 * This function contains 34 FP additions, 14 FP multiplications,
30 * (or, 28 additions, 8 multiplications, 6 fused multiply/add),
31 * 15 stack variables, and 32 memory accesses
32 */
33 static const fftw_real K707106781 =
34 FFTW_KONST(+0.707106781186547524400844362104849039284835938);
35
36 /*
37 * Generator Id's :
38 * $Id: exprdag.ml,v 1.43 2003/03/16 23:43:46 stevenj Exp $
39 * $Id: fft.ml,v 1.44 2003/03/16 23:43:46 stevenj Exp $
40 * $Id: to_c.ml,v 1.26 2003/03/16 23:43:46 stevenj Exp $
41 */
42
fftw_hc2hc_forward_4(fftw_real * A,const fftw_complex * W,int iostride,int m,int dist)43 void fftw_hc2hc_forward_4(fftw_real *A, const fftw_complex *W,
44 int iostride, int m, int dist)
45 {
46 int i;
47 fftw_real *X;
48 fftw_real *Y;
49 X = A;
50 Y = A + (4 * iostride);
51 {
52 fftw_real tmp35;
53 fftw_real tmp36;
54 fftw_real tmp37;
55 fftw_real tmp38;
56 fftw_real tmp39;
57 fftw_real tmp40;
58 ASSERT_ALIGNED_DOUBLE;
59 tmp35 = X[0];
60 tmp36 = X[2 * iostride];
61 tmp37 = tmp35 + tmp36;
62 tmp38 = X[iostride];
63 tmp39 = X[3 * iostride];
64 tmp40 = tmp38 + tmp39;
65 X[iostride] = tmp35 - tmp36;
66 Y[-iostride] = -(tmp38 - tmp39);
67 X[2 * iostride] = tmp37 - tmp40;
68 X[0] = tmp37 + tmp40;
69 }
70 X = X + dist;
71 Y = Y - dist;
72 for (i = 2; i < m; i = i + 2, X = X + dist, Y = Y - dist, W = W + 3) {
73 fftw_real tmp7;
74 fftw_real tmp30;
75 fftw_real tmp12;
76 fftw_real tmp29;
77 fftw_real tmp18;
78 fftw_real tmp26;
79 fftw_real tmp23;
80 fftw_real tmp27;
81 ASSERT_ALIGNED_DOUBLE;
82 tmp7 = X[0];
83 tmp30 = Y[-3 * iostride];
84 {
85 fftw_real tmp9;
86 fftw_real tmp11;
87 fftw_real tmp8;
88 fftw_real tmp10;
89 ASSERT_ALIGNED_DOUBLE;
90 tmp9 = X[2 * iostride];
91 tmp11 = Y[-iostride];
92 tmp8 = c_re(W[1]);
93 tmp10 = c_im(W[1]);
94 tmp12 = (tmp8 * tmp9) - (tmp10 * tmp11);
95 tmp29 = (tmp10 * tmp9) + (tmp8 * tmp11);
96 }
97 {
98 fftw_real tmp15;
99 fftw_real tmp17;
100 fftw_real tmp14;
101 fftw_real tmp16;
102 ASSERT_ALIGNED_DOUBLE;
103 tmp15 = X[iostride];
104 tmp17 = Y[-2 * iostride];
105 tmp14 = c_re(W[0]);
106 tmp16 = c_im(W[0]);
107 tmp18 = (tmp14 * tmp15) - (tmp16 * tmp17);
108 tmp26 = (tmp16 * tmp15) + (tmp14 * tmp17);
109 }
110 {
111 fftw_real tmp20;
112 fftw_real tmp22;
113 fftw_real tmp19;
114 fftw_real tmp21;
115 ASSERT_ALIGNED_DOUBLE;
116 tmp20 = X[3 * iostride];
117 tmp22 = Y[0];
118 tmp19 = c_re(W[2]);
119 tmp21 = c_im(W[2]);
120 tmp23 = (tmp19 * tmp20) - (tmp21 * tmp22);
121 tmp27 = (tmp21 * tmp20) + (tmp19 * tmp22);
122 }
123 {
124 fftw_real tmp13;
125 fftw_real tmp24;
126 fftw_real tmp33;
127 fftw_real tmp34;
128 ASSERT_ALIGNED_DOUBLE;
129 tmp13 = tmp7 + tmp12;
130 tmp24 = tmp18 + tmp23;
131 Y[-2 * iostride] = tmp13 - tmp24;
132 X[0] = tmp13 + tmp24;
133 tmp33 = tmp18 - tmp23;
134 tmp34 = tmp30 - tmp29;
135 X[3 * iostride] = -(tmp33 + tmp34);
136 Y[-iostride] = tmp34 - tmp33;
137 }
138 {
139 fftw_real tmp31;
140 fftw_real tmp32;
141 fftw_real tmp25;
142 fftw_real tmp28;
143 ASSERT_ALIGNED_DOUBLE;
144 tmp31 = tmp29 + tmp30;
145 tmp32 = tmp26 + tmp27;
146 X[2 * iostride] = -(tmp31 - tmp32);
147 Y[0] = tmp32 + tmp31;
148 tmp25 = tmp7 - tmp12;
149 tmp28 = tmp26 - tmp27;
150 Y[-3 * iostride] = tmp25 - tmp28;
151 X[iostride] = tmp25 + tmp28;
152 }
153 }
154 if (i == m) {
155 fftw_real tmp1;
156 fftw_real tmp6;
157 fftw_real tmp4;
158 fftw_real tmp5;
159 fftw_real tmp2;
160 fftw_real tmp3;
161 ASSERT_ALIGNED_DOUBLE;
162 tmp1 = X[0];
163 tmp6 = X[2 * iostride];
164 tmp2 = X[iostride];
165 tmp3 = X[3 * iostride];
166 tmp4 = K707106781 * (tmp2 - tmp3);
167 tmp5 = K707106781 * (tmp2 + tmp3);
168 X[iostride] = tmp1 - tmp4;
169 X[0] = tmp1 + tmp4;
170 Y[0] = -(tmp5 + tmp6);
171 Y[-iostride] = tmp6 - tmp5;
172 }
173 }
174
175 static const int twiddle_order[] = { 1, 2, 3 };
176 fftw_codelet_desc fftw_hc2hc_forward_4_desc = {
177 "fftw_hc2hc_forward_4",
178 (void (*)()) fftw_hc2hc_forward_4,
179 4,
180 FFTW_FORWARD,
181 FFTW_HC2HC,
182 91,
183 3,
184 twiddle_order,
185 };
186