/* * Copyright (c) 1997-1999, 2003 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 Mon Mar 24 02:07:48 EST 2003 */ #include "fftw-int.h" #include "fftw.h" /* Generated by: /homee/stevenj/cvs/fftw/gensrc/genfft -magic-alignment-check -magic-twiddle-load-all -magic-variables 4 -magic-loopi -hc2hc-forward 3 */ /* * This function contains 25 FP additions, 16 FP multiplications, * (or, 17 additions, 8 multiplications, 8 fused multiply/add), * 15 stack variables, and 24 memory accesses */ static const fftw_real K866025403 = FFTW_KONST(+0.866025403784438646763723170752936183471402627); static const fftw_real K500000000 = FFTW_KONST(+0.500000000000000000000000000000000000000000000); /* * Generator Id's : * $Id: exprdag.ml,v 1.43 2003/03/16 23:43:46 stevenj Exp $ * $Id: fft.ml,v 1.44 2003/03/16 23:43:46 stevenj Exp $ * $Id: to_c.ml,v 1.26 2003/03/16 23:43:46 stevenj Exp $ */ void fftw_hc2hc_forward_3(fftw_real *A, const fftw_complex *W, int iostride, int m, int dist) { int i; fftw_real *X; fftw_real *Y; X = A; Y = A + (3 * iostride); { fftw_real tmp24; fftw_real tmp25; fftw_real tmp26; fftw_real tmp27; ASSERT_ALIGNED_DOUBLE; tmp24 = X[0]; tmp25 = X[iostride]; tmp26 = X[2 * iostride]; tmp27 = tmp25 + tmp26; X[iostride] = tmp24 - (K500000000 * tmp27); X[0] = tmp24 + tmp27; Y[-iostride] = K866025403 * (tmp26 - tmp25); } X = X + dist; Y = Y - dist; for (i = 2; i < m; i = i + 2, X = X + dist, Y = Y - dist, W = W + 2) { fftw_real tmp4; fftw_real tmp21; fftw_real tmp9; fftw_real tmp17; fftw_real tmp14; fftw_real tmp18; fftw_real tmp15; fftw_real tmp20; ASSERT_ALIGNED_DOUBLE; tmp4 = X[0]; tmp21 = Y[-2 * iostride]; { fftw_real tmp6; fftw_real tmp8; fftw_real tmp5; fftw_real tmp7; ASSERT_ALIGNED_DOUBLE; tmp6 = X[iostride]; tmp8 = Y[-iostride]; tmp5 = c_re(W[0]); tmp7 = c_im(W[0]); tmp9 = (tmp5 * tmp6) - (tmp7 * tmp8); tmp17 = (tmp7 * tmp6) + (tmp5 * tmp8); } { fftw_real tmp11; fftw_real tmp13; fftw_real tmp10; fftw_real tmp12; ASSERT_ALIGNED_DOUBLE; tmp11 = X[2 * iostride]; tmp13 = Y[0]; tmp10 = c_re(W[1]); tmp12 = c_im(W[1]); tmp14 = (tmp10 * tmp11) - (tmp12 * tmp13); tmp18 = (tmp12 * tmp11) + (tmp10 * tmp13); } tmp15 = tmp9 + tmp14; tmp20 = tmp17 + tmp18; { fftw_real tmp16; fftw_real tmp19; fftw_real tmp22; fftw_real tmp23; ASSERT_ALIGNED_DOUBLE; X[0] = tmp4 + tmp15; tmp16 = tmp4 - (K500000000 * tmp15); tmp19 = K866025403 * (tmp17 - tmp18); Y[-2 * iostride] = tmp16 - tmp19; X[iostride] = tmp16 + tmp19; Y[0] = tmp20 + tmp21; tmp22 = tmp21 - (K500000000 * tmp20); tmp23 = K866025403 * (tmp14 - tmp9); X[2 * iostride] = -(tmp22 - tmp23); Y[-iostride] = tmp23 + tmp22; } } if (i == m) { fftw_real tmp1; fftw_real tmp3; fftw_real tmp2; ASSERT_ALIGNED_DOUBLE; tmp1 = X[iostride]; tmp3 = X[0]; tmp2 = X[2 * iostride]; Y[0] = -(K866025403 * (tmp1 + tmp2)); X[iostride] = tmp2 + tmp3 - tmp1; X[0] = tmp3 - (K500000000 * (tmp2 - tmp1)); } } static const int twiddle_order[] = { 1, 2 }; fftw_codelet_desc fftw_hc2hc_forward_3_desc = { "fftw_hc2hc_forward_3", (void (*)()) fftw_hc2hc_forward_3, 3, FFTW_FORWARD, FFTW_HC2HC, 69, 2, twiddle_order, };