annotate Lib/fftw-3.2.1/cell/spu/spu_n2fv_10.spuc @ 0:25bf17994ef1

First commit. VS2013, Codeblocks and Mac OSX configuration

/*
 * 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_notw_c -standalone -fma -reorder-insns -simd -compact -variables 100000 -with-ostride 2 -include fftw-spu.h -store-multiple 2 -n 10 -name X(spu_n2fv_10) */

/*
 * This function contains 42 FP additions, 22 FP multiplications,
 * (or, 24 additions, 4 multiplications, 18 fused multiply/add),
 * 59 stack variables, 4 constants, and 25 memory accesses
 */
#include "fftw-spu.h"

void X(spu_n2fv_10) (const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs) {
     DVK(KP559016994, +0.559016994374947424102293417182819058860154590);
     DVK(KP250000000, +0.250000000000000000000000000000000000000000000);
     DVK(KP618033988, +0.618033988749894848204586834365638117720309180);
     DVK(KP951056516, +0.951056516295153572116439333379382143405698634);
     INT i;
     const R *xi;
     R *xo;
     xi = ri;
     xo = ro;
     for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
	  V Tr, T3, Tm, Tn, TD, TC, TA, Ty, Ti, Tk, T1, T2, Ts, T6, Tw;
	  V Tg, Tt, T9, Tv, Td, T4, T5, Te, Tf, T7, T8, Tb, Tc, Ta, Th;
	  V Tu, Tx, TH, TI, TK, TL, TM, Tq, To, Tp, Tl, Tj, TJ, TG, TE;
	  V TF, TB, Tz, TN, TO, TP, TQ;
	  T1 = LD(&(xi[0]), ivs, &(xi[0]));
	  T2 = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
	  Tr = VADD(T1, T2);
	  T3 = VSUB(T1, T2);
	  T4 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
	  T5 = LD(&(xi[WS(is, 7)]), ivs, &(xi[WS(is, 1)]));
	  Ts = VADD(T4, T5);
	  T6 = VSUB(T4, T5);
	  Te = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
	  Tf = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
	  Tw = VADD(Te, Tf);
	  Tg = VSUB(Te, Tf);
	  T7 = LD(&(xi[WS(is, 8)]), ivs, &(xi[0]));
	  T8 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
	  Tt = VADD(T7, T8);
	  T9 = VSUB(T7, T8);
	  Tb = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
	  Tc = LD(&(xi[WS(is, 9)]), ivs, &(xi[WS(is, 1)]));
	  Tv = VADD(Tb, Tc);
	  Td = VSUB(Tb, Tc);
	  Ta = VADD(T6, T9);
	  Tm = VSUB(T6, T9);
	  Tn = VSUB(Td, Tg);
	  Th = VADD(Td, Tg);
	  Tu = VADD(Ts, Tt);
	  TD = VSUB(Ts, Tt);
	  TC = VSUB(Tv, Tw);
	  Tx = VADD(Tv, Tw);
	  TA = VSUB(Tu, Tx);
	  Ty = VADD(Tu, Tx);
	  Ti = VADD(Ta, Th);
	  Tk = VSUB(Ta, Th);
	  TH = VADD(T3, Ti);
	  STM2(&(xo[10]), TH, ovs, &(xo[2]));
	  TI = VADD(Tr, Ty);
	  STM2(&(xo[0]), TI, ovs, &(xo[0]));
	  Tq = VMUL(LDK(KP951056516), VFNMS(LDK(KP618033988), Tm, Tn));
	  To = VMUL(LDK(KP951056516), VFMA(LDK(KP618033988), Tn, Tm));
	  Tj = VFNMS(LDK(KP250000000), Ti, T3);
	  Tp = VFNMS(LDK(KP559016994), Tk, Tj);
	  Tl = VFMA(LDK(KP559016994), Tk, Tj);
	  TJ = VFNMSI(To, Tl);
	  STM2(&(xo[2]), TJ, ovs, &(xo[2]));
	  STN2(&(xo[0]), TI, TJ, ovs);
	  TK = VFMAI(Tq, Tp);
	  STM2(&(xo[14]), TK, ovs, &(xo[2]));
	  TL = VFMAI(To, Tl);
	  STM2(&(xo[18]), TL, ovs, &(xo[2]));
	  TM = VFNMSI(Tq, Tp);
	  STM2(&(xo[6]), TM, ovs, &(xo[2]));
	  TG = VMUL(LDK(KP951056516), VFMA(LDK(KP618033988), TC, TD));
	  TE = VMUL(LDK(KP951056516), VFNMS(LDK(KP618033988), TD, TC));
	  Tz = VFNMS(LDK(KP250000000), Ty, Tr);
	  TF = VFMA(LDK(KP559016994), TA, Tz);
	  TB = VFNMS(LDK(KP559016994), TA, Tz);
	  TN = VFMAI(TE, TB);
	  STM2(&(xo[4]), TN, ovs, &(xo[0]));
	  STN2(&(xo[4]), TN, TM, ovs);
	  TO = VFNMSI(TG, TF);
	  STM2(&(xo[12]), TO, ovs, &(xo[0]));
	  STN2(&(xo[12]), TO, TK, ovs);
	  TP = VFNMSI(TE, TB);
	  STM2(&(xo[16]), TP, ovs, &(xo[0]));
	  STN2(&(xo[16]), TP, TL, ovs);
	  TQ = VFMAI(TG, TF);
	  STM2(&(xo[8]), TQ, ovs, &(xo[0]));
	  STN2(&(xo[8]), TQ, TH, ovs);
     }
}