n1bv_5.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 20:43:58 EST 2008 */

#include "codelet-dft.h"

#ifdef HAVE_FMA

/* Generated by: ../../../genfft/gen_notw_c -fma -reorder-insns -schedule-for-pipeline -simd -compact -variables 4 -pipeline-latency 8 -sign 1 -n 5 -name n1bv_5 -include n1b.h */

/*
 * This function contains 16 FP additions, 11 FP multiplications,
 * (or, 7 additions, 2 multiplications, 9 fused multiply/add),
 * 23 stack variables, 4 constants, and 10 memory accesses
 */
#include "n1b.h"

static void n1bv_5(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 = ii;
     xo = io;
     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 T1, T2, T3, T5, T6;
	  T1 = LD(&(xi[0]), ivs, &(xi[0]));
	  T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
	  T3 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
	  T5 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
	  T6 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
	  {
	       V Tc, T4, Td, T7;
	       Tc = VSUB(T2, T3);
	       T4 = VADD(T2, T3);
	       Td = VSUB(T5, T6);
	       T7 = VADD(T5, T6);
	       {
		    V Tg, Te, Ta, T8, T9, Tf, Tb;
		    Tg = VMUL(LDK(KP951056516), VFNMS(LDK(KP618033988), Tc, Td));
		    Te = VMUL(LDK(KP951056516), VFMA(LDK(KP618033988), Td, Tc));
		    Ta = VSUB(T4, T7);
		    T8 = VADD(T4, T7);
		    T9 = VFNMS(LDK(KP250000000), T8, T1);
		    ST(&(xo[0]), VADD(T1, T8), ovs, &(xo[0]));
		    Tf = VFNMS(LDK(KP559016994), Ta, T9);
		    Tb = VFMA(LDK(KP559016994), Ta, T9);
		    ST(&(xo[WS(os, 2)]), VFNMSI(Tg, Tf), ovs, &(xo[0]));
		    ST(&(xo[WS(os, 3)]), VFMAI(Tg, Tf), ovs, &(xo[WS(os, 1)]));
		    ST(&(xo[WS(os, 4)]), VFNMSI(Te, Tb), ovs, &(xo[0]));
		    ST(&(xo[WS(os, 1)]), VFMAI(Te, Tb), ovs, &(xo[WS(os, 1)]));
	       }
	  }
     }
}

static const kdft_desc desc = { 5, "n1bv_5", {7, 2, 9, 0}, &GENUS, 0, 0, 0, 0 };
void X(codelet_n1bv_5) (planner *p) {
     X(kdft_register) (p, n1bv_5, &desc);
}

#else				/* HAVE_FMA */

/* Generated by: ../../../genfft/gen_notw_c -simd -compact -variables 4 -pipeline-latency 8 -sign 1 -n 5 -name n1bv_5 -include n1b.h */

/*
 * This function contains 16 FP additions, 6 FP multiplications,
 * (or, 13 additions, 3 multiplications, 3 fused multiply/add),
 * 18 stack variables, 4 constants, and 10 memory accesses
 */
#include "n1b.h"

static void n1bv_5(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
     DVK(KP250000000, +0.250000000000000000000000000000000000000000000);
     DVK(KP587785252, +0.587785252292473129168705954639072768597652438);
     DVK(KP951056516, +0.951056516295153572116439333379382143405698634);
     DVK(KP559016994, +0.559016994374947424102293417182819058860154590);
     INT i;
     const R *xi;
     R *xo;
     xi = ii;
     xo = io;
     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 Tb, T3, Tc, T6, Ta;
	  Tb = LD(&(xi[0]), ivs, &(xi[0]));
	  {
	       V T1, T2, T8, T4, T5, T9;
	       T1 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
	       T2 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
	       T8 = VADD(T1, T2);
	       T4 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
	       T5 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
	       T9 = VADD(T4, T5);
	       T3 = VSUB(T1, T2);
	       Tc = VADD(T8, T9);
	       T6 = VSUB(T4, T5);
	       Ta = VMUL(LDK(KP559016994), VSUB(T8, T9));
	  }
	  ST(&(xo[0]), VADD(Tb, Tc), ovs, &(xo[0]));
	  {
	       V T7, Tf, Te, Tg, Td;
	       T7 = VBYI(VFMA(LDK(KP951056516), T3, VMUL(LDK(KP587785252), T6)));
	       Tf = VBYI(VFNMS(LDK(KP951056516), T6, VMUL(LDK(KP587785252), T3)));
	       Td = VFNMS(LDK(KP250000000), Tc, Tb);
	       Te = VADD(Ta, Td);
	       Tg = VSUB(Td, Ta);
	       ST(&(xo[WS(os, 1)]), VADD(T7, Te), ovs, &(xo[WS(os, 1)]));
	       ST(&(xo[WS(os, 3)]), VSUB(Tg, Tf), ovs, &(xo[WS(os, 1)]));
	       ST(&(xo[WS(os, 4)]), VSUB(Te, T7), ovs, &(xo[0]));
	       ST(&(xo[WS(os, 2)]), VADD(Tf, Tg), ovs, &(xo[0]));
	  }
     }
}

static const kdft_desc desc = { 5, "n1bv_5", {13, 3, 3, 0}, &GENUS, 0, 0, 0, 0 };
void X(codelet_n1bv_5) (planner *p) {
     X(kdft_register) (p, n1bv_5, &desc);
}

#endif				/* HAVE_FMA */