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 5 */
27
28 /*
29 * This function contains 64 FP additions, 40 FP multiplications,
30 * (or, 44 additions, 20 multiplications, 20 fused multiply/add),
31 * 27 stack variables, and 40 memory accesses
32 */
33 static const fftw_real K250000000 =
34 FFTW_KONST(+0.250000000000000000000000000000000000000000000);
35 static const fftw_real K559016994 =
36 FFTW_KONST(+0.559016994374947424102293417182819058860154590);
37 static const fftw_real K587785252 =
38 FFTW_KONST(+0.587785252292473129168705954639072768597652438);
39 static const fftw_real K951056516 =
40 FFTW_KONST(+0.951056516295153572116439333379382143405698634);
41
42 /*
43 * Generator Id's :
44 * $Id: exprdag.ml,v 1.43 2003/03/16 23:43:46 stevenj Exp $
45 * $Id: fft.ml,v 1.44 2003/03/16 23:43:46 stevenj Exp $
46 * $Id: to_c.ml,v 1.26 2003/03/16 23:43:46 stevenj Exp $
47 */
48
fftw_hc2hc_forward_5(fftw_real * A,const fftw_complex * W,int iostride,int m,int dist)49 void fftw_hc2hc_forward_5(fftw_real *A, const fftw_complex *W,
50 int iostride, int m, int dist)
51 {
52 int i;
53 fftw_real *X;
54 fftw_real *Y;
55 X = A;
56 Y = A + (5 * iostride);
57 {
58 fftw_real tmp70;
59 fftw_real tmp67;
60 fftw_real tmp68;
61 fftw_real tmp63;
62 fftw_real tmp71;
63 fftw_real tmp66;
64 fftw_real tmp69;
65 fftw_real tmp72;
66 ASSERT_ALIGNED_DOUBLE;
67 tmp70 = X[0];
68 {
69 fftw_real tmp61;
70 fftw_real tmp62;
71 fftw_real tmp64;
72 fftw_real tmp65;
73 ASSERT_ALIGNED_DOUBLE;
74 tmp61 = X[4 * iostride];
75 tmp62 = X[iostride];
76 tmp67 = tmp61 + tmp62;
77 tmp64 = X[2 * iostride];
78 tmp65 = X[3 * iostride];
79 tmp68 = tmp64 + tmp65;
80 tmp63 = tmp61 - tmp62;
81 tmp71 = tmp67 + tmp68;
82 tmp66 = tmp64 - tmp65;
83 }
84 Y[-iostride] = (K951056516 * tmp63) - (K587785252 * tmp66);
85 Y[-2 * iostride] = (K587785252 * tmp63) + (K951056516 * tmp66);
86 X[0] = tmp71 + tmp70;
87 tmp69 = K559016994 * (tmp67 - tmp68);
88 tmp72 = tmp70 - (K250000000 * tmp71);
89 X[iostride] = tmp69 + tmp72;
90 X[2 * iostride] = tmp72 - tmp69;
91 }
92 X = X + dist;
93 Y = Y - dist;
94 for (i = 2; i < m; i = i + 2, X = X + dist, Y = Y - dist, W = W + 4) {
95 fftw_real tmp13;
96 fftw_real tmp52;
97 fftw_real tmp42;
98 fftw_real tmp45;
99 fftw_real tmp49;
100 fftw_real tmp50;
101 fftw_real tmp51;
102 fftw_real tmp54;
103 fftw_real tmp53;
104 fftw_real tmp24;
105 fftw_real tmp35;
106 fftw_real tmp36;
107 ASSERT_ALIGNED_DOUBLE;
108 tmp13 = X[0];
109 tmp52 = Y[-4 * iostride];
110 {
111 fftw_real tmp18;
112 fftw_real tmp40;
113 fftw_real tmp34;
114 fftw_real tmp44;
115 fftw_real tmp23;
116 fftw_real tmp41;
117 fftw_real tmp29;
118 fftw_real tmp43;
119 ASSERT_ALIGNED_DOUBLE;
120 {
121 fftw_real tmp15;
122 fftw_real tmp17;
123 fftw_real tmp14;
124 fftw_real tmp16;
125 ASSERT_ALIGNED_DOUBLE;
126 tmp15 = X[iostride];
127 tmp17 = Y[-3 * iostride];
128 tmp14 = c_re(W[0]);
129 tmp16 = c_im(W[0]);
130 tmp18 = (tmp14 * tmp15) - (tmp16 * tmp17);
131 tmp40 = (tmp16 * tmp15) + (tmp14 * tmp17);
132 }
133 {
134 fftw_real tmp31;
135 fftw_real tmp33;
136 fftw_real tmp30;
137 fftw_real tmp32;
138 ASSERT_ALIGNED_DOUBLE;
139 tmp31 = X[3 * iostride];
140 tmp33 = Y[-iostride];
141 tmp30 = c_re(W[2]);
142 tmp32 = c_im(W[2]);
143 tmp34 = (tmp30 * tmp31) - (tmp32 * tmp33);
144 tmp44 = (tmp32 * tmp31) + (tmp30 * tmp33);
145 }
146 {
147 fftw_real tmp20;
148 fftw_real tmp22;
149 fftw_real tmp19;
150 fftw_real tmp21;
151 ASSERT_ALIGNED_DOUBLE;
152 tmp20 = X[4 * iostride];
153 tmp22 = Y[0];
154 tmp19 = c_re(W[3]);
155 tmp21 = c_im(W[3]);
156 tmp23 = (tmp19 * tmp20) - (tmp21 * tmp22);
157 tmp41 = (tmp21 * tmp20) + (tmp19 * tmp22);
158 }
159 {
160 fftw_real tmp26;
161 fftw_real tmp28;
162 fftw_real tmp25;
163 fftw_real tmp27;
164 ASSERT_ALIGNED_DOUBLE;
165 tmp26 = X[2 * iostride];
166 tmp28 = Y[-2 * iostride];
167 tmp25 = c_re(W[1]);
168 tmp27 = c_im(W[1]);
169 tmp29 = (tmp25 * tmp26) - (tmp27 * tmp28);
170 tmp43 = (tmp27 * tmp26) + (tmp25 * tmp28);
171 }
172 tmp42 = tmp40 - tmp41;
173 tmp45 = tmp43 - tmp44;
174 tmp49 = tmp40 + tmp41;
175 tmp50 = tmp43 + tmp44;
176 tmp51 = tmp49 + tmp50;
177 tmp54 = tmp29 - tmp34;
178 tmp53 = tmp18 - tmp23;
179 tmp24 = tmp18 + tmp23;
180 tmp35 = tmp29 + tmp34;
181 tmp36 = tmp24 + tmp35;
182 }
183 X[0] = tmp13 + tmp36;
184 {
185 fftw_real tmp46;
186 fftw_real tmp48;
187 fftw_real tmp39;
188 fftw_real tmp47;
189 fftw_real tmp37;
190 fftw_real tmp38;
191 ASSERT_ALIGNED_DOUBLE;
192 tmp46 = (K951056516 * tmp42) + (K587785252 * tmp45);
193 tmp48 = (K951056516 * tmp45) - (K587785252 * tmp42);
194 tmp37 = K559016994 * (tmp24 - tmp35);
195 tmp38 = tmp13 - (K250000000 * tmp36);
196 tmp39 = tmp37 + tmp38;
197 tmp47 = tmp38 - tmp37;
198 Y[-4 * iostride] = tmp39 - tmp46;
199 X[iostride] = tmp39 + tmp46;
200 X[2 * iostride] = tmp47 - tmp48;
201 Y[-3 * iostride] = tmp47 + tmp48;
202 }
203 Y[0] = tmp51 + tmp52;
204 {
205 fftw_real tmp55;
206 fftw_real tmp60;
207 fftw_real tmp58;
208 fftw_real tmp59;
209 fftw_real tmp56;
210 fftw_real tmp57;
211 ASSERT_ALIGNED_DOUBLE;
212 tmp55 = (K951056516 * tmp53) + (K587785252 * tmp54);
213 tmp60 = (K951056516 * tmp54) - (K587785252 * tmp53);
214 tmp56 = K559016994 * (tmp49 - tmp50);
215 tmp57 = tmp52 - (K250000000 * tmp51);
216 tmp58 = tmp56 + tmp57;
217 tmp59 = tmp57 - tmp56;
218 X[4 * iostride] = -(tmp55 + tmp58);
219 Y[-iostride] = tmp58 - tmp55;
220 X[3 * iostride] = -(tmp59 - tmp60);
221 Y[-2 * iostride] = tmp60 + tmp59;
222 }
223 }
224 if (i == m) {
225 fftw_real tmp10;
226 fftw_real tmp7;
227 fftw_real tmp8;
228 fftw_real tmp3;
229 fftw_real tmp11;
230 fftw_real tmp6;
231 fftw_real tmp9;
232 fftw_real tmp12;
233 ASSERT_ALIGNED_DOUBLE;
234 tmp10 = X[0];
235 {
236 fftw_real tmp1;
237 fftw_real tmp2;
238 fftw_real tmp4;
239 fftw_real tmp5;
240 ASSERT_ALIGNED_DOUBLE;
241 tmp1 = X[2 * iostride];
242 tmp2 = X[3 * iostride];
243 tmp7 = tmp1 - tmp2;
244 tmp4 = X[4 * iostride];
245 tmp5 = X[iostride];
246 tmp8 = tmp4 - tmp5;
247 tmp3 = tmp1 + tmp2;
248 tmp11 = tmp7 + tmp8;
249 tmp6 = tmp4 + tmp5;
250 }
251 Y[0] = -((K951056516 * tmp3) + (K587785252 * tmp6));
252 Y[-iostride] = -((K951056516 * tmp6) - (K587785252 * tmp3));
253 X[2 * iostride] = tmp11 + tmp10;
254 tmp9 = K559016994 * (tmp7 - tmp8);
255 tmp12 = tmp10 - (K250000000 * tmp11);
256 X[0] = tmp9 + tmp12;
257 X[iostride] = tmp12 - tmp9;
258 }
259 }
260
261 static const int twiddle_order[] = { 1, 2, 3, 4 };
262 fftw_codelet_desc fftw_hc2hc_forward_5_desc = {
263 "fftw_hc2hc_forward_5",
264 (void (*)()) fftw_hc2hc_forward_5,
265 5,
266 FFTW_FORWARD,
267 FFTW_HC2HC,
268 113,
269 4,
270 twiddle_order,
271 };
272