Chris@82: /*
Chris@82:  * Copyright (c) 2003, 2007-14 Matteo Frigo
Chris@82:  * Copyright (c) 2003, 2007-14 Massachusetts Institute of Technology
Chris@82:  *
Chris@82:  * This program is free software; you can redistribute it and/or modify
Chris@82:  * it under the terms of the GNU General Public License as published by
Chris@82:  * the Free Software Foundation; either version 2 of the License, or
Chris@82:  * (at your option) any later version.
Chris@82:  *
Chris@82:  * This program is distributed in the hope that it will be useful,
Chris@82:  * but WITHOUT ANY WARRANTY; without even the implied warranty of
Chris@82:  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
Chris@82:  * GNU General Public License for more details.
Chris@82:  *
Chris@82:  * You should have received a copy of the GNU General Public License
Chris@82:  * along with this program; if not, write to the Free Software
Chris@82:  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
Chris@82:  *
Chris@82:  */
Chris@82: 
Chris@82: /* This file was automatically generated --- DO NOT EDIT */
Chris@82: /* Generated on Thu May 24 08:06:42 EDT 2018 */
Chris@82: 
Chris@82: #include "rdft/codelet-rdft.h"
Chris@82: 
Chris@82: #if defined(ARCH_PREFERS_FMA) || defined(ISA_EXTENSION_PREFERS_FMA)
Chris@82: 
Chris@82: /* Generated by: ../../../genfft/gen_r2cf.native -fma -compact -variables 4 -pipeline-latency 4 -n 6 -name r2cfII_6 -dft-II -include rdft/scalar/r2cfII.h */
Chris@82: 
Chris@82: /*
Chris@82:  * This function contains 13 FP additions, 6 FP multiplications,
Chris@82:  * (or, 7 additions, 0 multiplications, 6 fused multiply/add),
Chris@82:  * 15 stack variables, 2 constants, and 12 memory accesses
Chris@82:  */
Chris@82: #include "rdft/scalar/r2cfII.h"
Chris@82: 
Chris@82: static void r2cfII_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs)
Chris@82: {
Chris@82:      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
Chris@82:      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
Chris@82:      {
Chris@82: 	  INT i;
Chris@82: 	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
Chris@82: 	       E T1, T9, T2, T3, T4, Tc, T8, Ta, T6, T7, T5, Tb;
Chris@82: 	       T1 = R0[0];
Chris@82: 	       T9 = R1[WS(rs, 1)];
Chris@82: 	       T2 = R0[WS(rs, 2)];
Chris@82: 	       T3 = R0[WS(rs, 1)];
Chris@82: 	       T4 = T3 - T2;
Chris@82: 	       Tc = T2 + T3;
Chris@82: 	       T6 = R1[WS(rs, 2)];
Chris@82: 	       T7 = R1[0];
Chris@82: 	       T8 = T6 - T7;
Chris@82: 	       Ta = T6 + T7;
Chris@82: 	       Ci[WS(csi, 1)] = T9 - Ta;
Chris@82: 	       Cr[WS(csr, 1)] = T1 + T2 - T3;
Chris@82: 	       T5 = FMA(KP500000000, T4, T1);
Chris@82: 	       Cr[0] = FNMS(KP866025403, T8, T5);
Chris@82: 	       Cr[WS(csr, 2)] = FMA(KP866025403, T8, T5);
Chris@82: 	       Tb = FMA(KP500000000, Ta, T9);
Chris@82: 	       Ci[0] = -(FMA(KP866025403, Tc, Tb));
Chris@82: 	       Ci[WS(csi, 2)] = FMS(KP866025403, Tc, Tb);
Chris@82: 	  }
Chris@82:      }
Chris@82: }
Chris@82: 
Chris@82: static const kr2c_desc desc = { 6, "r2cfII_6", {7, 0, 6, 0}, &GENUS };
Chris@82: 
Chris@82: void X(codelet_r2cfII_6) (planner *p) {
Chris@82:      X(kr2c_register) (p, r2cfII_6, &desc);
Chris@82: }
Chris@82: 
Chris@82: #else
Chris@82: 
Chris@82: /* Generated by: ../../../genfft/gen_r2cf.native -compact -variables 4 -pipeline-latency 4 -n 6 -name r2cfII_6 -dft-II -include rdft/scalar/r2cfII.h */
Chris@82: 
Chris@82: /*
Chris@82:  * This function contains 13 FP additions, 4 FP multiplications,
Chris@82:  * (or, 11 additions, 2 multiplications, 2 fused multiply/add),
Chris@82:  * 14 stack variables, 2 constants, and 12 memory accesses
Chris@82:  */
Chris@82: #include "rdft/scalar/r2cfII.h"
Chris@82: 
Chris@82: static void r2cfII_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs)
Chris@82: {
Chris@82:      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
Chris@82:      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
Chris@82:      {
Chris@82: 	  INT i;
Chris@82: 	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
Chris@82: 	       E Ta, T7, T9, T1, T3, T2, T8, T4, T5, T6, Tb;
Chris@82: 	       Ta = R1[WS(rs, 1)];
Chris@82: 	       T5 = R1[WS(rs, 2)];
Chris@82: 	       T6 = R1[0];
Chris@82: 	       T7 = KP866025403 * (T5 - T6);
Chris@82: 	       T9 = T5 + T6;
Chris@82: 	       T1 = R0[0];
Chris@82: 	       T3 = R0[WS(rs, 1)];
Chris@82: 	       T2 = R0[WS(rs, 2)];
Chris@82: 	       T8 = KP866025403 * (T2 + T3);
Chris@82: 	       T4 = FMA(KP500000000, T3 - T2, T1);
Chris@82: 	       Cr[0] = T4 - T7;
Chris@82: 	       Cr[WS(csr, 2)] = T4 + T7;
Chris@82: 	       Ci[WS(csi, 1)] = Ta - T9;
Chris@82: 	       Cr[WS(csr, 1)] = T1 + T2 - T3;
Chris@82: 	       Tb = FMA(KP500000000, T9, Ta);
Chris@82: 	       Ci[0] = -(T8 + Tb);
Chris@82: 	       Ci[WS(csi, 2)] = T8 - Tb;
Chris@82: 	  }
Chris@82:      }
Chris@82: }
Chris@82: 
Chris@82: static const kr2c_desc desc = { 6, "r2cfII_6", {11, 2, 2, 0}, &GENUS };
Chris@82: 
Chris@82: void X(codelet_r2cfII_6) (planner *p) {
Chris@82:      X(kr2c_register) (p, r2cfII_6, &desc);
Chris@82: }
Chris@82: 
Chris@82: #endif