hc2cfdftv_6.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 21:14:12 EST 2008 */

#include "codelet-rdft.h"

#ifdef HAVE_FMA

/* Generated by: ../../../genfft/gen_hc2cdft_c -fma -reorder-insns -schedule-for-pipeline -simd -compact -variables 4 -pipeline-latency 8 -trivial-stores -variables 32 -no-generate-bytw -n 6 -dit -name hc2cfdftv_6 -include hc2cfv.h */

/*
 * This function contains 29 FP additions, 30 FP multiplications,
 * (or, 17 additions, 18 multiplications, 12 fused multiply/add),
 * 38 stack variables, 2 constants, and 12 memory accesses
 */
#include "hc2cfv.h"

static void hc2cfdftv_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, INT me, INT ms)
{
     DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
     DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
     INT m;
     for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 10)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
	  V T5, T6, T3, Tj, T4, T9, Te, Th, T1, T2, Ti, Tc, Td, Tb, Tg;
	  V T7, Ta, Tt, Tk, Tr, T8, Ts, Tf, Tx, Tu, To, Tl, Tw, Tv, Tn;
	  V Tm, Tz, Ty, Tp, Tq;
	  T1 = LD(&(Rp[0]), ms, &(Rp[0]));
	  T2 = LD(&(Rm[0]), -ms, &(Rm[0]));
	  Ti = LDW(&(W[0]));
	  Tc = LD(&(Rp[WS(rs, 2)]), ms, &(Rp[0]));
	  Td = LD(&(Rm[WS(rs, 2)]), -ms, &(Rm[0]));
	  Tb = LDW(&(W[TWVL * 8]));
	  Tg = LDW(&(W[TWVL * 6]));
	  T5 = LD(&(Rp[WS(rs, 1)]), ms, &(Rp[WS(rs, 1)]));
	  T6 = LD(&(Rm[WS(rs, 1)]), -ms, &(Rm[WS(rs, 1)]));
	  T3 = VFMACONJ(T2, T1);
	  Tj = VZMULIJ(Ti, VFNMSCONJ(T2, T1));
	  T4 = LDW(&(W[TWVL * 4]));
	  T9 = LDW(&(W[TWVL * 2]));
	  Te = VZMULIJ(Tb, VFNMSCONJ(Td, Tc));
	  Th = VZMULJ(Tg, VFMACONJ(Td, Tc));
	  T7 = VZMULIJ(T4, VFNMSCONJ(T6, T5));
	  Ta = VZMULJ(T9, VFMACONJ(T6, T5));
	  Tt = VADD(Tj, Th);
	  Tk = VSUB(Th, Tj);
	  Tr = VADD(T3, T7);
	  T8 = VSUB(T3, T7);
	  Ts = VADD(Ta, Te);
	  Tf = VSUB(Ta, Te);
	  Tx = VMUL(LDK(KP866025403), VSUB(Tt, Ts));
	  Tu = VADD(Ts, Tt);
	  To = VMUL(LDK(KP866025403), VSUB(Tk, Tf));
	  Tl = VADD(Tf, Tk);
	  Tw = VFNMS(LDK(KP500000000), Tu, Tr);
	  Tv = VCONJ(VMUL(LDK(KP500000000), VADD(Tr, Tu)));
	  Tn = VFNMS(LDK(KP500000000), Tl, T8);
	  Tm = VMUL(LDK(KP500000000), VADD(T8, Tl));
	  Tz = VMUL(LDK(KP500000000), VFMAI(Tx, Tw));
	  Ty = VCONJ(VMUL(LDK(KP500000000), VFNMSI(Tx, Tw)));
	  ST(&(Rm[WS(rs, 2)]), Tv, -ms, &(Rm[0]));
	  Tp = VMUL(LDK(KP500000000), VFNMSI(To, Tn));
	  Tq = VCONJ(VMUL(LDK(KP500000000), VFMAI(To, Tn)));
	  ST(&(Rp[0]), Tm, ms, &(Rp[0]));
	  ST(&(Rp[WS(rs, 1)]), Tz, ms, &(Rp[WS(rs, 1)]));
	  ST(&(Rm[0]), Ty, -ms, &(Rm[0]));
	  ST(&(Rm[WS(rs, 1)]), Tq, -ms, &(Rm[WS(rs, 1)]));
	  ST(&(Rp[WS(rs, 2)]), Tp, ms, &(Rp[0]));
     }
}

static const tw_instr twinstr[] = {
     VTW(1, 1),
     VTW(1, 2),
     VTW(1, 3),
     VTW(1, 4),
     VTW(1, 5),
     {TW_NEXT, VL, 0}
};

static const hc2c_desc desc = { 6, "hc2cfdftv_6", twinstr, &GENUS, {17, 18, 12, 0} };

void X(codelet_hc2cfdftv_6) (planner *p) {
     X(khc2c_register) (p, hc2cfdftv_6, &desc, HC2C_VIA_DFT);
}
#else				/* HAVE_FMA */

/* Generated by: ../../../genfft/gen_hc2cdft_c -simd -compact -variables 4 -pipeline-latency 8 -trivial-stores -variables 32 -no-generate-bytw -n 6 -dit -name hc2cfdftv_6 -include hc2cfv.h */

/*
 * This function contains 29 FP additions, 20 FP multiplications,
 * (or, 27 additions, 18 multiplications, 2 fused multiply/add),
 * 42 stack variables, 3 constants, and 12 memory accesses
 */
#include "hc2cfv.h"

static void hc2cfdftv_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, INT me, INT ms)
{
     DVK(KP250000000, +0.250000000000000000000000000000000000000000000);
     DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
     DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
     INT m;
     for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 10)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
	  V Ta, Tu, Tn, Tw, Ti, Tv, T1, T8, Tg, Tf, T7, T3, Te, T6, T2;
	  V T4, T9, T5, Tk, Tm, Tj, Tl, Tc, Th, Tb, Td, Tr, Tp, Tq, To;
	  V Tt, Ts, TA, Ty, Tz, Tx, TC, TB;
	  T1 = LD(&(Rp[0]), ms, &(Rp[0]));
	  T8 = LD(&(Rp[WS(rs, 1)]), ms, &(Rp[WS(rs, 1)]));
	  Tg = LD(&(Rp[WS(rs, 2)]), ms, &(Rp[0]));
	  Te = LD(&(Rm[WS(rs, 2)]), -ms, &(Rm[0]));
	  Tf = VCONJ(Te);
	  T6 = LD(&(Rm[WS(rs, 1)]), -ms, &(Rm[WS(rs, 1)]));
	  T7 = VCONJ(T6);
	  T2 = LD(&(Rm[0]), -ms, &(Rm[0]));
	  T3 = VCONJ(T2);
	  T4 = VADD(T1, T3);
	  T5 = LDW(&(W[TWVL * 4]));
	  T9 = VZMULIJ(T5, VSUB(T7, T8));
	  Ta = VADD(T4, T9);
	  Tu = VSUB(T4, T9);
	  Tj = LDW(&(W[0]));
	  Tk = VZMULIJ(Tj, VSUB(T3, T1));
	  Tl = LDW(&(W[TWVL * 6]));
	  Tm = VZMULJ(Tl, VADD(Tf, Tg));
	  Tn = VADD(Tk, Tm);
	  Tw = VSUB(Tm, Tk);
	  Tb = LDW(&(W[TWVL * 2]));
	  Tc = VZMULJ(Tb, VADD(T7, T8));
	  Td = LDW(&(W[TWVL * 8]));
	  Th = VZMULIJ(Td, VSUB(Tf, Tg));
	  Ti = VADD(Tc, Th);
	  Tv = VSUB(Tc, Th);
	  Tr = VMUL(LDK(KP500000000), VBYI(VMUL(LDK(KP866025403), VSUB(Tn, Ti))));
	  To = VADD(Ti, Tn);
	  Tp = VMUL(LDK(KP500000000), VADD(Ta, To));
	  Tq = VFNMS(LDK(KP250000000), To, VMUL(LDK(KP500000000), Ta));
	  ST(&(Rp[0]), Tp, ms, &(Rp[0]));
	  Tt = VCONJ(VADD(Tq, Tr));
	  ST(&(Rm[WS(rs, 1)]), Tt, -ms, &(Rm[WS(rs, 1)]));
	  Ts = VSUB(Tq, Tr);
	  ST(&(Rp[WS(rs, 2)]), Ts, ms, &(Rp[0]));
	  TA = VMUL(LDK(KP500000000), VBYI(VMUL(LDK(KP866025403), VSUB(Tw, Tv))));
	  Tx = VADD(Tv, Tw);
	  Ty = VCONJ(VMUL(LDK(KP500000000), VADD(Tu, Tx)));
	  Tz = VFNMS(LDK(KP250000000), Tx, VMUL(LDK(KP500000000), Tu));
	  ST(&(Rm[WS(rs, 2)]), Ty, -ms, &(Rm[0]));
	  TC = VADD(Tz, TA);
	  ST(&(Rp[WS(rs, 1)]), TC, ms, &(Rp[WS(rs, 1)]));
	  TB = VCONJ(VSUB(Tz, TA));
	  ST(&(Rm[0]), TB, -ms, &(Rm[0]));
     }
}

static const tw_instr twinstr[] = {
     VTW(1, 1),
     VTW(1, 2),
     VTW(1, 3),
     VTW(1, 4),
     VTW(1, 5),
     {TW_NEXT, VL, 0}
};

static const hc2c_desc desc = { 6, "hc2cfdftv_6", twinstr, &GENUS, {27, 18, 2, 0} };

void X(codelet_hc2cfdftv_6) (planner *p) {
     X(khc2c_register) (p, hc2cfdftv_6, &desc, HC2C_VIA_DFT);
}
#endif				/* HAVE_FMA */