cannam@167: /*
cannam@167:  * Copyright (c) 2003, 2007-14 Matteo Frigo
cannam@167:  * Copyright (c) 2003, 2007-14 Massachusetts Institute of Technology
cannam@167:  *
cannam@167:  * This program is free software; you can redistribute it and/or modify
cannam@167:  * it under the terms of the GNU General Public License as published by
cannam@167:  * the Free Software Foundation; either version 2 of the License, or
cannam@167:  * (at your option) any later version.
cannam@167:  *
cannam@167:  * This program is distributed in the hope that it will be useful,
cannam@167:  * but WITHOUT ANY WARRANTY; without even the implied warranty of
cannam@167:  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
cannam@167:  * GNU General Public License for more details.
cannam@167:  *
cannam@167:  * You should have received a copy of the GNU General Public License
cannam@167:  * along with this program; if not, write to the Free Software
cannam@167:  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
cannam@167:  *
cannam@167:  */
cannam@167: 
cannam@167: /* This file was automatically generated --- DO NOT EDIT */
cannam@167: /* Generated on Thu May 24 08:07:28 EDT 2018 */
cannam@167: 
cannam@167: #include "rdft/codelet-rdft.h"
cannam@167: 
cannam@167: #if defined(ARCH_PREFERS_FMA) || defined(ISA_EXTENSION_PREFERS_FMA)
cannam@167: 
cannam@167: /* Generated by: ../../../genfft/gen_r2cb.native -fma -compact -variables 4 -pipeline-latency 4 -sign 1 -n 6 -name r2cb_6 -include rdft/scalar/r2cb.h */
cannam@167: 
cannam@167: /*
cannam@167:  * This function contains 14 FP additions, 6 FP multiplications,
cannam@167:  * (or, 8 additions, 0 multiplications, 6 fused multiply/add),
cannam@167:  * 17 stack variables, 2 constants, and 12 memory accesses
cannam@167:  */
cannam@167: #include "rdft/scalar/r2cb.h"
cannam@167: 
cannam@167: static void r2cb_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs)
cannam@167: {
cannam@167:      DK(KP1_732050807, +1.732050807568877293527446341505872366942805254);
cannam@167:      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
cannam@167:      {
cannam@167: 	  INT i;
cannam@167: 	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
cannam@167: 	       E T3, T7, Tc, Te, T6, T8, T1, T2, T9, Td;
cannam@167: 	       T1 = Cr[0];
cannam@167: 	       T2 = Cr[WS(csr, 3)];
cannam@167: 	       T3 = T1 + T2;
cannam@167: 	       T7 = T1 - T2;
cannam@167: 	       {
cannam@167: 		    E Ta, Tb, T4, T5;
cannam@167: 		    Ta = Ci[WS(csi, 2)];
cannam@167: 		    Tb = Ci[WS(csi, 1)];
cannam@167: 		    Tc = Ta - Tb;
cannam@167: 		    Te = Ta + Tb;
cannam@167: 		    T4 = Cr[WS(csr, 2)];
cannam@167: 		    T5 = Cr[WS(csr, 1)];
cannam@167: 		    T6 = T4 + T5;
cannam@167: 		    T8 = T5 - T4;
cannam@167: 	       }
cannam@167: 	       R0[0] = FMA(KP2_000000000, T6, T3);
cannam@167: 	       R1[WS(rs, 1)] = FNMS(KP2_000000000, T8, T7);
cannam@167: 	       T9 = T3 - T6;
cannam@167: 	       R0[WS(rs, 2)] = FNMS(KP1_732050807, Tc, T9);
cannam@167: 	       R0[WS(rs, 1)] = FMA(KP1_732050807, Tc, T9);
cannam@167: 	       Td = T7 + T8;
cannam@167: 	       R1[0] = FNMS(KP1_732050807, Te, Td);
cannam@167: 	       R1[WS(rs, 2)] = FMA(KP1_732050807, Te, Td);
cannam@167: 	  }
cannam@167:      }
cannam@167: }
cannam@167: 
cannam@167: static const kr2c_desc desc = { 6, "r2cb_6", {8, 0, 6, 0}, &GENUS };
cannam@167: 
cannam@167: void X(codelet_r2cb_6) (planner *p) {
cannam@167:      X(kr2c_register) (p, r2cb_6, &desc);
cannam@167: }
cannam@167: 
cannam@167: #else
cannam@167: 
cannam@167: /* Generated by: ../../../genfft/gen_r2cb.native -compact -variables 4 -pipeline-latency 4 -sign 1 -n 6 -name r2cb_6 -include rdft/scalar/r2cb.h */
cannam@167: 
cannam@167: /*
cannam@167:  * This function contains 14 FP additions, 4 FP multiplications,
cannam@167:  * (or, 12 additions, 2 multiplications, 2 fused multiply/add),
cannam@167:  * 17 stack variables, 2 constants, and 12 memory accesses
cannam@167:  */
cannam@167: #include "rdft/scalar/r2cb.h"
cannam@167: 
cannam@167: static void r2cb_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs)
cannam@167: {
cannam@167:      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
cannam@167:      DK(KP1_732050807, +1.732050807568877293527446341505872366942805254);
cannam@167:      {
cannam@167: 	  INT i;
cannam@167: 	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
cannam@167: 	       E T3, T7, Tc, Te, T6, T8, T1, T2, T9, Td;
cannam@167: 	       T1 = Cr[0];
cannam@167: 	       T2 = Cr[WS(csr, 3)];
cannam@167: 	       T3 = T1 - T2;
cannam@167: 	       T7 = T1 + T2;
cannam@167: 	       {
cannam@167: 		    E Ta, Tb, T4, T5;
cannam@167: 		    Ta = Ci[WS(csi, 2)];
cannam@167: 		    Tb = Ci[WS(csi, 1)];
cannam@167: 		    Tc = KP1_732050807 * (Ta - Tb);
cannam@167: 		    Te = KP1_732050807 * (Ta + Tb);
cannam@167: 		    T4 = Cr[WS(csr, 2)];
cannam@167: 		    T5 = Cr[WS(csr, 1)];
cannam@167: 		    T6 = T4 - T5;
cannam@167: 		    T8 = T4 + T5;
cannam@167: 	       }
cannam@167: 	       R1[WS(rs, 1)] = FMA(KP2_000000000, T6, T3);
cannam@167: 	       R0[0] = FMA(KP2_000000000, T8, T7);
cannam@167: 	       T9 = T7 - T8;
cannam@167: 	       R0[WS(rs, 2)] = T9 - Tc;
cannam@167: 	       R0[WS(rs, 1)] = T9 + Tc;
cannam@167: 	       Td = T3 - T6;
cannam@167: 	       R1[0] = Td - Te;
cannam@167: 	       R1[WS(rs, 2)] = Td + Te;
cannam@167: 	  }
cannam@167:      }
cannam@167: }
cannam@167: 
cannam@167: static const kr2c_desc desc = { 6, "r2cb_6", {12, 2, 2, 0}, &GENUS };
cannam@167: 
cannam@167: void X(codelet_r2cb_6) (planner *p) {
cannam@167:      X(kr2c_register) (p, r2cb_6, &desc);
cannam@167: }
cannam@167: 
cannam@167: #endif