t1bv_12.c

快速fft变换

/*
 * 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
 *
 */

/* This file was automatically generated --- DO NOT EDIT */
/* Generated on Sat Nov 15 20:51:05 EST 2008 */

#include "codelet-dft.h"

#ifdef HAVE_FMA

/* Generated by: ../../../genfft/gen_twiddle_c -fma -reorder-insns -schedule-for-pipeline -simd -compact -variables 4 -pipeline-latency 8 -n 12 -name t1bv_12 -include t1b.h -sign 1 */

/*
 * This function contains 59 FP additions, 42 FP multiplications,
 * (or, 41 additions, 24 multiplications, 18 fused multiply/add),
 * 41 stack variables, 2 constants, and 24 memory accesses
 */
#include "t1b.h"

static void t1bv_12(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
{
     DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
     DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
     INT m;
     R *x;
     x = ii;
     for (m = mb, W = W + (mb * ((TWVL / VL) * 22)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(rs)) {
	  V TI, Ti, TA, T7, Tm, TE, Tw, Tk, Tf, TB, TU, TM;
	  {
	       V T9, TK, Tj, TL, Te;
	       {
		    V T1, T4, T2, Tp, Tt, Tr;
		    T1 = LD(&(x[0]), ms, &(x[0]));
		    T4 = LD(&(x[WS(rs, 8)]), ms, &(x[0]));
		    T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
		    Tp = LD(&(x[WS(rs, 9)]), ms, &(x[WS(rs, 1)]));
		    Tt = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
		    Tr = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
		    {
			 V T5, T3, Tq, Tu, Ts, Td, Tb, T8, Tc, Ta;
			 T8 = LD(&(x[WS(rs, 6)]), ms, &(x[0]));
			 Tc = LD(&(x[WS(rs, 2)]), ms, &(x[0]));
			 Ta = LD(&(x[WS(rs, 10)]), ms, &(x[0]));
			 T5 = BYTW(&(W[TWVL * 14]), T4);
			 T3 = BYTW(&(W[TWVL * 6]), T2);
			 Tq = BYTW(&(W[TWVL * 16]), Tp);
			 Tu = BYTW(&(W[TWVL * 8]), Tt);
			 Ts = BYTW(&(W[0]), Tr);
			 T9 = BYTW(&(W[TWVL * 10]), T8);
			 Td = BYTW(&(W[TWVL * 2]), Tc);
			 Tb = BYTW(&(W[TWVL * 18]), Ta);
			 {
			      V Th, T6, Tl, Tv;
			      Th = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
			      TK = VSUB(T3, T5);
			      T6 = VADD(T3, T5);
			      Tl = LD(&(x[WS(rs, 11)]), ms, &(x[WS(rs, 1)]));
			      Tv = VADD(Ts, Tu);
			      TI = VSUB(Tu, Ts);
			      Tj = LD(&(x[WS(rs, 7)]), ms, &(x[WS(rs, 1)]));
			      TL = VSUB(Tb, Td);
			      Te = VADD(Tb, Td);
			      Ti = BYTW(&(W[TWVL * 4]), Th);
			      TA = VFNMS(LDK(KP500000000), T6, T1);
			      T7 = VADD(T1, T6);
			      Tm = BYTW(&(W[TWVL * 20]), Tl);
			      TE = VFNMS(LDK(KP500000000), Tv, Tq);
			      Tw = VADD(Tq, Tv);
			 }
		    }
	       }
	       Tk = BYTW(&(W[TWVL * 12]), Tj);
	       Tf = VADD(T9, Te);
	       TB = VFNMS(LDK(KP500000000), Te, T9);
	       TU = VSUB(TK, TL);
	       TM = VADD(TK, TL);
	  }
	  {
	       V Tn, TH, TC, TQ, Ty, Tg;
	       Tn = VADD(Tk, Tm);
	       TH = VSUB(Tk, Tm);
	       TC = VADD(TA, TB);
	       TQ = VSUB(TA, TB);
	       Ty = VADD(T7, Tf);
	       Tg = VSUB(T7, Tf);
	       {
		    V To, TD, TJ, TR;
		    To = VADD(Ti, Tn);
		    TD = VFNMS(LDK(KP500000000), Tn, Ti);
		    TJ = VSUB(TH, TI);
		    TR = VADD(TH, TI);
		    {
			 V TP, TN, TW, TS, TO, TG, TX, TV;
			 {
			      V Tz, Tx, TF, TT;
			      Tz = VADD(To, Tw);
			      Tx = VSUB(To, Tw);
			      TF = VADD(TD, TE);
			      TT = VSUB(TD, TE);
			      TP = VMUL(LDK(KP866025403), VADD(TM, TJ));
			      TN = VMUL(LDK(KP866025403), VSUB(TJ, TM));
			      TW = VFMA(LDK(KP866025403), TR, TQ);
			      TS = VFNMS(LDK(KP866025403), TR, TQ);
			      ST(&(x[WS(rs, 6)]), VSUB(Ty, Tz), ms, &(x[0]));
			      ST(&(x[0]), VADD(Ty, Tz), ms, &(x[0]));
			      ST(&(x[WS(rs, 9)]), VFMAI(Tx, Tg), ms, &(x[WS(rs, 1)]));
			      ST(&(x[WS(rs, 3)]), VFNMSI(Tx, Tg), ms, &(x[WS(rs, 1)]));
			      TO = VADD(TC, TF);
			      TG = VSUB(TC, TF);
			      TX = VFNMS(LDK(KP866025403), TU, TT);
			      TV = VFMA(LDK(KP866025403), TU, TT);
			 }
			 ST(&(x[WS(rs, 8)]), VFNMSI(TP, TO), ms, &(x[0]));
			 ST(&(x[WS(rs, 4)]), VFMAI(TP, TO), ms, &(x[0]));
			 ST(&(x[WS(rs, 2)]), VFMAI(TN, TG), ms, &(x[0]));
			 ST(&(x[WS(rs, 10)]), VFNMSI(TN, TG), ms, &(x[0]));
			 ST(&(x[WS(rs, 5)]), VFMAI(TX, TW), ms, &(x[WS(rs, 1)]));
			 ST(&(x[WS(rs, 7)]), VFNMSI(TX, TW), ms, &(x[WS(rs, 1)]));
			 ST(&(x[WS(rs, 11)]), VFNMSI(TV, TS), ms, &(x[WS(rs, 1)]));
			 ST(&(x[WS(rs, 1)]), VFMAI(TV, TS), ms, &(x[WS(rs, 1)]));
		    }
	       }
	  }
     }
}

static const tw_instr twinstr[] = {
     VTW(0, 1),
     VTW(0, 2),
     VTW(0, 3),
     VTW(0, 4),
     VTW(0, 5),
     VTW(0, 6),
     VTW(0, 7),
     VTW(0, 8),
     VTW(0, 9),
     VTW(0, 10),
     VTW(0, 11),
     {TW_NEXT, VL, 0}
};

static const ct_desc desc = { 12, "t1bv_12", twinstr, &GENUS, {41, 24, 18, 0}, 0, 0, 0 };

void X(codelet_t1bv_12) (planner *p) {
     X(kdft_dit_register) (p, t1bv_12, &desc);
}
#else				/* HAVE_FMA */

/* Generated by: ../../../genfft/gen_twiddle_c -simd -compact -variables 4 -pipeline-latency 8 -n 12 -name t1bv_12 -include t1b.h -sign 1 */

/*
 * This function contains 59 FP additions, 30 FP multiplications,
 * (or, 55 additions, 26 multiplications, 4 fused multiply/add),
 * 28 stack variables, 2 constants, and 24 memory accesses
 */
#include "t1b.h"

static void t1bv_12(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
{
     DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
     DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
     INT m;
     R *x;
     x = ii;
     for (m = mb, W = W + (mb * ((TWVL / VL) * 22)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(rs)) {
	  V T1, Tt, T6, T7, TB, Tq, TC, TD, T9, Tu, Te, Tf, Tx, Tl, Ty;
	  V Tz;
	  {
	       V T5, T3, T4, T2;
	       T1 = LD(&(x[0]), ms, &(x[0]));
	       T4 = LD(&(x[WS(rs, 8)]), ms, &(x[0]));
	       T5 = BYTW(&(W[TWVL * 14]), T4);
	       T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
	       T3 = BYTW(&(W[TWVL * 6]), T2);
	       Tt = VSUB(T3, T5);
	       T6 = VADD(T3, T5);
	       T7 = VFNMS(LDK(KP500000000), T6, T1);
	  }
	  {
	       V Tn, Tp, Tm, TA, To;
	       Tm = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
	       Tn = BYTW(&(W[0]), Tm);
	       TA = LD(&(x[WS(rs, 9)]), ms, &(x[WS(rs, 1)]));
	       TB = BYTW(&(W[TWVL * 16]), TA);
	       To = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
	       Tp = BYTW(&(W[TWVL * 8]), To);
	       Tq = VSUB(Tn, Tp);
	       TC = VADD(Tn, Tp);
	       TD = VFNMS(LDK(KP500000000), TC, TB);
	  }
	  {
	       V Td, Tb, T8, Tc, Ta;
	       T8 = LD(&(x[WS(rs, 6)]), ms, &(x[0]));
	       T9 = BYTW(&(W[TWVL * 10]), T8);
	       Tc = LD(&(x[WS(rs, 2)]), ms, &(x[0]));
	       Td = BYTW(&(W[TWVL * 2]), Tc);
	       Ta = LD(&(x[WS(rs, 10)]), ms, &(x[0]));
	       Tb = BYTW(&(W[TWVL * 18]), Ta);
	       Tu = VSUB(Tb, Td);
	       Te = VADD(Tb, Td);
	       Tf = VFNMS(LDK(KP500000000), Te, T9);
	  }
	  {
	       V Ti, Tk, Th, Tw, Tj;
	       Th = LD(&(x[WS(rs, 7)]), ms, &(x[WS(rs, 1)]));
	       Ti = BYTW(&(W[TWVL * 12]), Th);
	       Tw = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
	       Tx = BYTW(&(W[TWVL * 4]), Tw);
	       Tj = LD(&(x[WS(rs, 11)]), ms, &(x[WS(rs, 1)]));
	       Tk = BYTW(&(W[TWVL * 20]), Tj);
	       Tl = VSUB(Ti, Tk);
	       Ty = VADD(Ti, Tk);
	       Tz = VFNMS(LDK(KP500000000), Ty, Tx);
	  }
	  {
	       V Ts, TG, TF, TH;
	       {
		    V Tg, Tr, Tv, TE;
		    Tg = VSUB(T7, Tf);
		    Tr = VMUL(LDK(KP866025403), VSUB(Tl, Tq));
		    Ts = VSUB(Tg, Tr);
		    TG = VADD(Tg, Tr);
		    Tv = VMUL(LDK(KP866025403), VSUB(Tt, Tu));
		    TE = VSUB(Tz, TD);
		    TF = VBYI(VADD(Tv, TE));
		    TH = VBYI(VSUB(TE, Tv));
	       }
	       ST(&(x[WS(rs, 11)]), VSUB(Ts, TF), ms, &(x[WS(rs, 1)]));
	       ST(&(x[WS(rs, 5)]), VADD(TG, TH), ms, &(x[WS(rs, 1)]));
	       ST(&(x[WS(rs, 1)]), VADD(Ts, TF), ms, &(x[WS(rs, 1)]));
	       ST(&(x[WS(rs, 7)]), VSUB(TG, TH), ms, &(x[WS(rs, 1)]));
	  }
	  {
	       V TS, TW, TV, TX;
	       {
		    V TQ, TR, TT, TU;
		    TQ = VADD(T1, T6);
		    TR = VADD(T9, Te);
		    TS = VSUB(TQ, TR);
		    TW = VADD(TQ, TR);
		    TT = VADD(Tx, Ty);
		    TU = VADD(TB, TC);
		    TV = VBYI(VSUB(TT, TU));
		    TX = VADD(TT, TU);
	       }
	       ST(&(x[WS(rs, 3)]), VSUB(TS, TV), ms, &(x[WS(rs, 1)]));
	       ST(&(x[0]), VADD(TW, TX), ms, &(x[0]));
	       ST(&(x[WS(rs, 9)]), VADD(TS, TV), ms, &(x[WS(rs, 1)]));
	       ST(&(x[WS(rs, 6)]), VSUB(TW, TX), ms, &(x[0]));
	  }
	  {
	       V TK, TO, TN, TP;
	       {
		    V TI, TJ, TL, TM;
		    TI = VADD(Tl, Tq);
		    TJ = VADD(Tt, Tu);
		    TK = VBYI(VMUL(LDK(KP866025403), VSUB(TI, TJ)));
		    TO = VBYI(VMUL(LDK(KP866025403), VADD(TJ, TI)));
		    TL = VADD(T7, Tf);
		    TM = VADD(Tz, TD);
		    TN = VSUB(TL, TM);
		    TP = VADD(TL, TM);
	       }
	       ST(&(x[WS(rs, 2)]), VADD(TK, TN), ms, &(x[0]));
	       ST(&(x[WS(rs, 8)]), VSUB(TP, TO), ms, &(x[0]));
	       ST(&(x[WS(rs, 10)]), VSUB(TN, TK), ms, &(x[0]));
	       ST(&(x[WS(rs, 4)]), VADD(TO, TP), ms, &(x[0]));
	  }
     }
}

static const tw_instr twinstr[] = {
     VTW(0, 1),
     VTW(0, 2),
     VTW(0, 3),
     VTW(0, 4),
     VTW(0, 5),
     VTW(0, 6),
     VTW(0, 7),
     VTW(0, 8),
     VTW(0, 9),
     VTW(0, 10),
     VTW(0, 11),
     {TW_NEXT, VL, 0}
};

static const ct_desc desc = { 12, "t1bv_12", twinstr, &GENUS, {55, 26, 4, 0}, 0, 0, 0 };

void X(codelet_t1bv_12) (planner *p) {
     X(kdft_dit_register) (p, t1bv_12, &desc);
}
#endif				/* HAVE_FMA */