/*
 * Copyright (c) 2003, 2007-8 Matteo Frigo
 * Copyright (c) 2003, 2007-8 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
 *
 */
/* Generated by: ../../genfft/gen_twiddle_c -standalone -fma -reorder-insns -simd -compact -variables 100000 -include fftw-spu.h -trivial-stores -n 8 -name X(spu_t1fv_8) */

/*
 * This function contains 33 FP additions, 24 FP multiplications,
 * (or, 23 additions, 14 multiplications, 10 fused multiply/add),
 * 44 stack variables, 1 constants, and 16 memory accesses
 */
#include "fftw-spu.h"

void X(spu_t1fv_8) (R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms) {
     DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
     INT m;
     R *x;
     x = ri;
     for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
	  V T4, Tu, Tl, Tv, Tx, Ty, Tf, Tm, T1, T3, T2, Ti, Tk, Th, Tj;
	  V T6, Td, T8, Tb, T5, Tc, T7, Ta, T9, Te, Tq, Tr, Tw, Tz, To;
	  V Tp, Tg, Tn, TE, TF, TC, TD, Ts, Tt, TA, TB;
	  T1 = LD(&(x[0]), ms, &(x[0]));
	  T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
	  T3 = BYTWJ(&(W[TWVL * 6]), T2);
	  T4 = VSUB(T1, T3);
	  Tu = VADD(T1, T3);
	  Th = LD(&(x[WS(rs, 2)]), ms, &(x[0]));
	  Ti = BYTWJ(&(W[TWVL * 2]), Th);
	  Tj = LD(&(x[WS(rs, 6)]), ms, &(x[0]));
	  Tk = BYTWJ(&(W[TWVL * 10]), Tj);
	  Tl = VSUB(Ti, Tk);
	  Tv = VADD(Ti, Tk);
	  T5 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
	  T6 = BYTWJ(&(W[0]), T5);
	  Tc = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
	  Td = BYTWJ(&(W[TWVL * 4]), Tc);
	  T7 = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
	  T8 = BYTWJ(&(W[TWVL * 8]), T7);
	  Ta = LD(&(x[WS(rs, 7)]), ms, &(x[WS(rs, 1)]));
	  Tb = BYTWJ(&(W[TWVL * 12]), Ta);
	  Tx = VADD(T6, T8);
	  T9 = VSUB(T6, T8);
	  Te = VSUB(Tb, Td);
	  Ty = VADD(Tb, Td);
	  Tf = VADD(T9, Te);
	  Tm = VSUB(Te, T9);
	  Tg = VFMA(LDK(KP707106781), Tf, T4);
	  Tq = VFNMS(LDK(KP707106781), Tf, T4);
	  Tr = VFMA(LDK(KP707106781), Tm, Tl);
	  Tn = VFNMS(LDK(KP707106781), Tm, Tl);
	  To = VFNMSI(Tn, Tg);
	  Tp = VFMAI(Tn, Tg);
	  ST(&(x[WS(rs, 1)]), To, ms, &(x[WS(rs, 1)]));
	  ST(&(x[WS(rs, 7)]), Tp, ms, &(x[WS(rs, 1)]));
	  TC = VSUB(Tu, Tv);
	  Tw = VADD(Tu, Tv);
	  Tz = VADD(Tx, Ty);
	  TD = VSUB(Ty, Tx);
	  TE = VFNMSI(TD, TC);
	  TF = VFMAI(TD, TC);
	  ST(&(x[WS(rs, 6)]), TE, ms, &(x[0]));
	  ST(&(x[WS(rs, 2)]), TF, ms, &(x[0]));
	  Ts = VFNMSI(Tr, Tq);
	  Tt = VFMAI(Tr, Tq);
	  ST(&(x[WS(rs, 5)]), Ts, ms, &(x[WS(rs, 1)]));
	  ST(&(x[WS(rs, 3)]), Tt, ms, &(x[WS(rs, 1)]));
	  TA = VSUB(Tw, Tz);
	  TB = VADD(Tw, Tz);
	  ST(&(x[WS(rs, 4)]), TA, ms, &(x[0]));
	  ST(&(x[0]), TB, ms, &(x[0]));
     }
}