mirror of
https://github.com/tildearrow/furnace.git
synced 2024-12-02 01:07:25 +00:00
128 lines
4.7 KiB
C
128 lines
4.7 KiB
C
/*
|
|
* Copyright (c) 2003, 2007-14 Matteo Frigo
|
|
* Copyright (c) 2003, 2007-14 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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
|
|
*
|
|
*/
|
|
|
|
/* This file was automatically generated --- DO NOT EDIT */
|
|
/* Generated on Tue Sep 14 10:46:47 EDT 2021 */
|
|
|
|
#include "rdft/codelet-rdft.h"
|
|
|
|
#if defined(ARCH_PREFERS_FMA) || defined(ISA_EXTENSION_PREFERS_FMA)
|
|
|
|
/* Generated by: ../../../genfft/gen_r2cb.native -fma -compact -variables 4 -pipeline-latency 4 -sign 1 -n 5 -name r2cb_5 -include rdft/scalar/r2cb.h */
|
|
|
|
/*
|
|
* This function contains 12 FP additions, 10 FP multiplications,
|
|
* (or, 2 additions, 0 multiplications, 10 fused multiply/add),
|
|
* 18 stack variables, 5 constants, and 10 memory accesses
|
|
*/
|
|
#include "rdft/scalar/r2cb.h"
|
|
|
|
static void r2cb_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs)
|
|
{
|
|
DK(KP1_902113032, +1.902113032590307144232878666758764286811397268);
|
|
DK(KP1_118033988, +1.118033988749894848204586834365638117720309180);
|
|
DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
|
|
DK(KP500000000, +0.500000000000000000000000000000000000000000000);
|
|
DK(KP618033988, +0.618033988749894848204586834365638117720309180);
|
|
{
|
|
INT i;
|
|
for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) {
|
|
E Ta, Tc, T1, T4, T5, T6, Tb, T7;
|
|
{
|
|
E T8, T9, T2, T3;
|
|
T8 = Ci[WS(csi, 1)];
|
|
T9 = Ci[WS(csi, 2)];
|
|
Ta = FMA(KP618033988, T9, T8);
|
|
Tc = FMS(KP618033988, T8, T9);
|
|
T1 = Cr[0];
|
|
T2 = Cr[WS(csr, 1)];
|
|
T3 = Cr[WS(csr, 2)];
|
|
T4 = T2 + T3;
|
|
T5 = FNMS(KP500000000, T4, T1);
|
|
T6 = T2 - T3;
|
|
}
|
|
R0[0] = FMA(KP2_000000000, T4, T1);
|
|
Tb = FNMS(KP1_118033988, T6, T5);
|
|
R0[WS(rs, 1)] = FNMS(KP1_902113032, Tc, Tb);
|
|
R1[WS(rs, 1)] = FMA(KP1_902113032, Tc, Tb);
|
|
T7 = FMA(KP1_118033988, T6, T5);
|
|
R1[0] = FNMS(KP1_902113032, Ta, T7);
|
|
R0[WS(rs, 2)] = FMA(KP1_902113032, Ta, T7);
|
|
}
|
|
}
|
|
}
|
|
|
|
static const kr2c_desc desc = { 5, "r2cb_5", { 2, 0, 10, 0 }, &GENUS };
|
|
|
|
void X(codelet_r2cb_5) (planner *p) { X(kr2c_register) (p, r2cb_5, &desc);
|
|
}
|
|
|
|
#else
|
|
|
|
/* Generated by: ../../../genfft/gen_r2cb.native -compact -variables 4 -pipeline-latency 4 -sign 1 -n 5 -name r2cb_5 -include rdft/scalar/r2cb.h */
|
|
|
|
/*
|
|
* This function contains 12 FP additions, 7 FP multiplications,
|
|
* (or, 8 additions, 3 multiplications, 4 fused multiply/add),
|
|
* 18 stack variables, 5 constants, and 10 memory accesses
|
|
*/
|
|
#include "rdft/scalar/r2cb.h"
|
|
|
|
static void r2cb_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs)
|
|
{
|
|
DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
|
|
DK(KP1_118033988, +1.118033988749894848204586834365638117720309180);
|
|
DK(KP500000000, +0.500000000000000000000000000000000000000000000);
|
|
DK(KP1_902113032, +1.902113032590307144232878666758764286811397268);
|
|
DK(KP1_175570504, +1.175570504584946258337411909278145537195304875);
|
|
{
|
|
INT i;
|
|
for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) {
|
|
E Ta, Tc, T1, T4, T5, T6, Tb, T7;
|
|
{
|
|
E T8, T9, T2, T3;
|
|
T8 = Ci[WS(csi, 1)];
|
|
T9 = Ci[WS(csi, 2)];
|
|
Ta = FNMS(KP1_902113032, T9, KP1_175570504 * T8);
|
|
Tc = FMA(KP1_902113032, T8, KP1_175570504 * T9);
|
|
T1 = Cr[0];
|
|
T2 = Cr[WS(csr, 1)];
|
|
T3 = Cr[WS(csr, 2)];
|
|
T4 = T2 + T3;
|
|
T5 = FNMS(KP500000000, T4, T1);
|
|
T6 = KP1_118033988 * (T2 - T3);
|
|
}
|
|
R0[0] = FMA(KP2_000000000, T4, T1);
|
|
Tb = T6 + T5;
|
|
R1[0] = Tb - Tc;
|
|
R0[WS(rs, 2)] = Tb + Tc;
|
|
T7 = T5 - T6;
|
|
R0[WS(rs, 1)] = T7 - Ta;
|
|
R1[WS(rs, 1)] = T7 + Ta;
|
|
}
|
|
}
|
|
}
|
|
|
|
static const kr2c_desc desc = { 5, "r2cb_5", { 8, 3, 4, 0 }, &GENUS };
|
|
|
|
void X(codelet_r2cb_5) (planner *p) { X(kr2c_register) (p, r2cb_5, &desc);
|
|
}
|
|
|
|
#endif
|