/*
 * 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 3 -name X(spu_t1fv_3) */

/*
 * This function contains 8 FP additions, 8 FP multiplications,
 * (or, 5 additions, 5 multiplications, 3 fused multiply/add),
 * 15 stack variables, 2 constants, and 6 memory accesses
 */
#include "fftw-spu.h"

void X(spu_t1fv_3) (R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms) {
     DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
     DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
     INT m;
     R *x;
     x = ri;
     for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
	  V T1, T6, T9, T3, T5, T2, T4, T8, T7, Tb, Ta;
	  T1 = LD(&(x[0]), ms, &(x[0]));
	  T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
	  T3 = BYTWJ(&(W[0]), T2);
	  T4 = LD(&(x[WS(rs, 2)]), ms, &(x[0]));
	  T5 = BYTWJ(&(W[TWVL * 2]), T4);
	  T6 = VADD(T3, T5);
	  T9 = VMUL(LDK(KP866025403), VSUB(T5, T3));
	  T7 = VADD(T1, T6);
	  T8 = VFNMS(LDK(KP500000000), T6, T1);
	  ST(&(x[0]), T7, ms, &(x[0]));
	  Tb = VFMAI(T9, T8);
	  Ta = VFNMSI(T9, T8);
	  ST(&(x[WS(rs, 2)]), Ta, ms, &(x[0]));
	  ST(&(x[WS(rs, 1)]), Tb, ms, &(x[WS(rs, 1)]));
     }
}