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 02:08:53 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-backward 6 */
27
28 /*
29 * This function contains 72 FP additions, 38 FP multiplications,
30 * (or, 54 additions, 20 multiplications, 18 fused multiply/add),
31 * 25 stack variables, and 48 memory accesses
32 */
33 static const fftw_real K500000000 =
34 FFTW_KONST(+0.500000000000000000000000000000000000000000000);
35 static const fftw_real K866025403 =
36 FFTW_KONST(+0.866025403784438646763723170752936183471402627);
37 static const fftw_real K2_000000000 =
38 FFTW_KONST(+2.000000000000000000000000000000000000000000000);
39 static const fftw_real K1_732050807 =
40 FFTW_KONST(+1.732050807568877293527446341505872366942805254);
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_backward_6(fftw_real * A,const fftw_complex * W,int iostride,int m,int dist)49 void fftw_hc2hc_backward_6(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 + (6 * iostride);
57 {
58 fftw_real tmp71;
59 fftw_real tmp75;
60 fftw_real tmp80;
61 fftw_real tmp82;
62 fftw_real tmp74;
63 fftw_real tmp76;
64 fftw_real tmp69;
65 fftw_real tmp70;
66 fftw_real tmp77;
67 fftw_real tmp81;
68 ASSERT_ALIGNED_DOUBLE;
69 tmp69 = X[0];
70 tmp70 = X[3 * iostride];
71 tmp71 = tmp69 - tmp70;
72 tmp75 = tmp69 + tmp70;
73 {
74 fftw_real tmp78;
75 fftw_real tmp79;
76 fftw_real tmp72;
77 fftw_real tmp73;
78 ASSERT_ALIGNED_DOUBLE;
79 tmp78 = Y[-2 * iostride];
80 tmp79 = Y[-iostride];
81 tmp80 = K1_732050807 * (tmp78 + tmp79);
82 tmp82 = K1_732050807 * (tmp78 - tmp79);
83 tmp72 = X[2 * iostride];
84 tmp73 = X[iostride];
85 tmp74 = tmp72 - tmp73;
86 tmp76 = tmp72 + tmp73;
87 }
88 X[3 * iostride] = tmp71 + (K2_000000000 * tmp74);
89 tmp77 = tmp71 - tmp74;
90 X[iostride] = tmp77 - tmp80;
91 X[5 * iostride] = tmp77 + tmp80;
92 X[0] = tmp75 + (K2_000000000 * tmp76);
93 tmp81 = tmp75 - tmp76;
94 X[2 * iostride] = tmp81 + tmp82;
95 X[4 * iostride] = tmp81 - tmp82;
96 }
97 X = X + dist;
98 Y = Y - dist;
99 for (i = 2; i < m; i = i + 2, X = X + dist, Y = Y - dist, W = W + 5) {
100 fftw_real tmp15;
101 fftw_real tmp46;
102 fftw_real tmp25;
103 fftw_real tmp52;
104 fftw_real tmp22;
105 fftw_real tmp35;
106 fftw_real tmp49;
107 fftw_real tmp62;
108 fftw_real tmp32;
109 fftw_real tmp39;
110 fftw_real tmp55;
111 fftw_real tmp59;
112 ASSERT_ALIGNED_DOUBLE;
113 {
114 fftw_real tmp13;
115 fftw_real tmp14;
116 fftw_real tmp23;
117 fftw_real tmp24;
118 ASSERT_ALIGNED_DOUBLE;
119 tmp13 = X[0];
120 tmp14 = Y[-3 * iostride];
121 tmp15 = tmp13 + tmp14;
122 tmp46 = tmp13 - tmp14;
123 tmp23 = Y[0];
124 tmp24 = X[3 * iostride];
125 tmp25 = tmp23 - tmp24;
126 tmp52 = tmp23 + tmp24;
127 }
128 {
129 fftw_real tmp18;
130 fftw_real tmp47;
131 fftw_real tmp21;
132 fftw_real tmp48;
133 ASSERT_ALIGNED_DOUBLE;
134 {
135 fftw_real tmp16;
136 fftw_real tmp17;
137 fftw_real tmp19;
138 fftw_real tmp20;
139 ASSERT_ALIGNED_DOUBLE;
140 tmp16 = X[2 * iostride];
141 tmp17 = Y[-5 * iostride];
142 tmp18 = tmp16 + tmp17;
143 tmp47 = tmp16 - tmp17;
144 tmp19 = Y[-4 * iostride];
145 tmp20 = X[iostride];
146 tmp21 = tmp19 + tmp20;
147 tmp48 = tmp19 - tmp20;
148 }
149 tmp22 = tmp18 + tmp21;
150 tmp35 = K866025403 * (tmp18 - tmp21);
151 tmp49 = tmp47 + tmp48;
152 tmp62 = K866025403 * (tmp47 - tmp48);
153 }
154 {
155 fftw_real tmp28;
156 fftw_real tmp54;
157 fftw_real tmp31;
158 fftw_real tmp53;
159 ASSERT_ALIGNED_DOUBLE;
160 {
161 fftw_real tmp26;
162 fftw_real tmp27;
163 fftw_real tmp29;
164 fftw_real tmp30;
165 ASSERT_ALIGNED_DOUBLE;
166 tmp26 = Y[-2 * iostride];
167 tmp27 = X[5 * iostride];
168 tmp28 = tmp26 - tmp27;
169 tmp54 = tmp26 + tmp27;
170 tmp29 = Y[-iostride];
171 tmp30 = X[4 * iostride];
172 tmp31 = tmp29 - tmp30;
173 tmp53 = tmp29 + tmp30;
174 }
175 tmp32 = tmp28 + tmp31;
176 tmp39 = K866025403 * (tmp31 - tmp28);
177 tmp55 = tmp53 - tmp54;
178 tmp59 = K866025403 * (tmp54 + tmp53);
179 }
180 X[0] = tmp15 + tmp22;
181 {
182 fftw_real tmp36;
183 fftw_real tmp42;
184 fftw_real tmp40;
185 fftw_real tmp44;
186 fftw_real tmp34;
187 fftw_real tmp38;
188 ASSERT_ALIGNED_DOUBLE;
189 tmp34 = tmp25 - (K500000000 * tmp32);
190 tmp36 = tmp34 - tmp35;
191 tmp42 = tmp35 + tmp34;
192 tmp38 = tmp15 - (K500000000 * tmp22);
193 tmp40 = tmp38 - tmp39;
194 tmp44 = tmp38 + tmp39;
195 {
196 fftw_real tmp33;
197 fftw_real tmp37;
198 fftw_real tmp41;
199 fftw_real tmp43;
200 ASSERT_ALIGNED_DOUBLE;
201 tmp33 = c_re(W[1]);
202 tmp37 = c_im(W[1]);
203 Y[-3 * iostride] = (tmp33 * tmp36) - (tmp37 * tmp40);
204 X[2 * iostride] = (tmp37 * tmp36) + (tmp33 * tmp40);
205 tmp41 = c_re(W[3]);
206 tmp43 = c_im(W[3]);
207 Y[-iostride] = (tmp41 * tmp42) - (tmp43 * tmp44);
208 X[4 * iostride] = (tmp43 * tmp42) + (tmp41 * tmp44);
209 }
210 }
211 Y[-5 * iostride] = tmp25 + tmp32;
212 {
213 fftw_real tmp50;
214 fftw_real tmp56;
215 fftw_real tmp45;
216 fftw_real tmp51;
217 ASSERT_ALIGNED_DOUBLE;
218 tmp50 = tmp46 + tmp49;
219 tmp56 = tmp52 - tmp55;
220 tmp45 = c_re(W[2]);
221 tmp51 = c_im(W[2]);
222 X[3 * iostride] = (tmp45 * tmp50) + (tmp51 * tmp56);
223 Y[-2 * iostride] = (tmp45 * tmp56) - (tmp51 * tmp50);
224 }
225 {
226 fftw_real tmp60;
227 fftw_real tmp66;
228 fftw_real tmp64;
229 fftw_real tmp68;
230 fftw_real tmp58;
231 fftw_real tmp63;
232 ASSERT_ALIGNED_DOUBLE;
233 tmp58 = tmp46 - (K500000000 * tmp49);
234 tmp60 = tmp58 - tmp59;
235 tmp66 = tmp58 + tmp59;
236 tmp63 = tmp52 + (K500000000 * tmp55);
237 tmp64 = tmp62 + tmp63;
238 tmp68 = tmp63 - tmp62;
239 {
240 fftw_real tmp57;
241 fftw_real tmp61;
242 fftw_real tmp65;
243 fftw_real tmp67;
244 ASSERT_ALIGNED_DOUBLE;
245 tmp57 = c_re(W[0]);
246 tmp61 = c_im(W[0]);
247 X[iostride] = (tmp57 * tmp60) + (tmp61 * tmp64);
248 Y[-4 * iostride] = (tmp57 * tmp64) - (tmp61 * tmp60);
249 tmp65 = c_re(W[4]);
250 tmp67 = c_im(W[4]);
251 X[5 * iostride] = (tmp65 * tmp66) + (tmp67 * tmp68);
252 Y[0] = (tmp65 * tmp68) - (tmp67 * tmp66);
253 }
254 }
255 }
256 if (i == m) {
257 fftw_real tmp1;
258 fftw_real tmp6;
259 fftw_real tmp4;
260 fftw_real tmp5;
261 fftw_real tmp9;
262 fftw_real tmp11;
263 fftw_real tmp12;
264 fftw_real tmp10;
265 ASSERT_ALIGNED_DOUBLE;
266 tmp1 = X[iostride];
267 tmp6 = Y[-iostride];
268 {
269 fftw_real tmp2;
270 fftw_real tmp3;
271 fftw_real tmp7;
272 fftw_real tmp8;
273 ASSERT_ALIGNED_DOUBLE;
274 tmp2 = X[2 * iostride];
275 tmp3 = X[0];
276 tmp4 = tmp2 + tmp3;
277 tmp5 = K1_732050807 * (tmp2 - tmp3);
278 tmp7 = Y[-2 * iostride];
279 tmp8 = Y[0];
280 tmp9 = tmp7 + tmp8;
281 tmp11 = K1_732050807 * (tmp7 - tmp8);
282 }
283 X[0] = K2_000000000 * (tmp1 + tmp4);
284 tmp12 = (K2_000000000 * tmp1) - tmp4;
285 X[2 * iostride] = tmp11 - tmp12;
286 X[4 * iostride] = tmp12 + tmp11;
287 X[3 * iostride] = K2_000000000 * (tmp6 - tmp9);
288 tmp10 = (K2_000000000 * tmp6) + tmp9;
289 X[iostride] = -(tmp5 + tmp10);
290 X[5 * iostride] = tmp5 - tmp10;
291 }
292 }
293
294 static const int twiddle_order[] = { 1, 2, 3, 4, 5 };
295 fftw_codelet_desc fftw_hc2hc_backward_6_desc = {
296 "fftw_hc2hc_backward_6",
297 (void (*)()) fftw_hc2hc_backward_6,
298 6,
299 FFTW_BACKWARD,
300 FFTW_HC2HC,
301 146,
302 5,
303 twiddle_order,
304 };
305