n1_8.c

fftw-3.0.1

/*
 * Copyright (c) 2003 Matteo Frigo
 * Copyright (c) 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 Sat Jul  5 21:29:32 EDT 2003 */

#include "codelet-dft.h"

/* Generated by: /homee/stevenj/cvs/fftw3.0.1/genfft/gen_notw -compact -variables 4 -n 8 -name n1_8 -include n.h */

/*
 * This function contains 52 FP additions, 4 FP multiplications,
 * (or, 52 additions, 4 multiplications, 0 fused multiply/add),
 * 28 stack variables, and 32 memory accesses
 */
/*
 * Generator Id's : 
 * $Id: algsimp.ml,v 1.7 2003/03/15 20:29:42 stevenj Exp $
 * $Id: fft.ml,v 1.2 2003/03/15 20:29:42 stevenj Exp $
 * $Id: gen_notw.ml,v 1.22 2003/04/17 11:07:19 athena Exp $
 */

#include "n.h"

static void n1_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, int v, int ivs, int ovs)
{
     DK(KP707106781, +0.707106781186547524400844362104849039284835938);
     int i;
     for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs) {
	  E T3, Tn, Ti, TC, T6, TB, Tl, To, Td, TN, Tz, TH, Ta, TM, Tu;
	  E TG;
	  {
	       E T1, T2, Tj, Tk;
	       T1 = ri[0];
	       T2 = ri[WS(is, 4)];
	       T3 = T1 + T2;
	       Tn = T1 - T2;
	       {
		    E Tg, Th, T4, T5;
		    Tg = ii[0];
		    Th = ii[WS(is, 4)];
		    Ti = Tg + Th;
		    TC = Tg - Th;
		    T4 = ri[WS(is, 2)];
		    T5 = ri[WS(is, 6)];
		    T6 = T4 + T5;
		    TB = T4 - T5;
	       }
	       Tj = ii[WS(is, 2)];
	       Tk = ii[WS(is, 6)];
	       Tl = Tj + Tk;
	       To = Tj - Tk;
	       {
		    E Tb, Tc, Tv, Tw, Tx, Ty;
		    Tb = ri[WS(is, 7)];
		    Tc = ri[WS(is, 3)];
		    Tv = Tb - Tc;
		    Tw = ii[WS(is, 7)];
		    Tx = ii[WS(is, 3)];
		    Ty = Tw - Tx;
		    Td = Tb + Tc;
		    TN = Tw + Tx;
		    Tz = Tv - Ty;
		    TH = Tv + Ty;
	       }
	       {
		    E T8, T9, Tq, Tr, Ts, Tt;
		    T8 = ri[WS(is, 1)];
		    T9 = ri[WS(is, 5)];
		    Tq = T8 - T9;
		    Tr = ii[WS(is, 1)];
		    Ts = ii[WS(is, 5)];
		    Tt = Tr - Ts;
		    Ta = T8 + T9;
		    TM = Tr + Ts;
		    Tu = Tq + Tt;
		    TG = Tt - Tq;
	       }
	  }
	  {
	       E T7, Te, TP, TQ;
	       T7 = T3 + T6;
	       Te = Ta + Td;
	       ro[WS(os, 4)] = T7 - Te;
	       ro[0] = T7 + Te;
	       TP = Ti + Tl;
	       TQ = TM + TN;
	       io[WS(os, 4)] = TP - TQ;
	       io[0] = TP + TQ;
	  }
	  {
	       E Tf, Tm, TL, TO;
	       Tf = Td - Ta;
	       Tm = Ti - Tl;
	       io[WS(os, 2)] = Tf + Tm;
	       io[WS(os, 6)] = Tm - Tf;
	       TL = T3 - T6;
	       TO = TM - TN;
	       ro[WS(os, 6)] = TL - TO;
	       ro[WS(os, 2)] = TL + TO;
	  }
	  {
	       E Tp, TA, TJ, TK;
	       Tp = Tn + To;
	       TA = KP707106781 * (Tu + Tz);
	       ro[WS(os, 5)] = Tp - TA;
	       ro[WS(os, 1)] = Tp + TA;
	       TJ = TC - TB;
	       TK = KP707106781 * (TG + TH);
	       io[WS(os, 5)] = TJ - TK;
	       io[WS(os, 1)] = TJ + TK;
	  }
	  {
	       E TD, TE, TF, TI;
	       TD = TB + TC;
	       TE = KP707106781 * (Tz - Tu);
	       io[WS(os, 7)] = TD - TE;
	       io[WS(os, 3)] = TD + TE;
	       TF = Tn - To;
	       TI = KP707106781 * (TG - TH);
	       ro[WS(os, 7)] = TF - TI;
	       ro[WS(os, 3)] = TF + TI;
	  }
     }
}

static const kdft_desc desc = { 8, "n1_8", {52, 4, 0, 0}, &GENUS, 0, 0, 0, 0 };
void X(codelet_n1_8) (planner *p) {
     X(kdft_register) (p, n1_8, &desc);
}