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 08:07:31 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_hc2hc.native -fma -compact -variables 4 -pipeline-latency 4 -sign 1 -n 2 -dif -name hb_2 -include rdft/scalar/hb.h */
cannam@167: 
cannam@167: /*
cannam@167:  * This function contains 6 FP additions, 4 FP multiplications,
cannam@167:  * (or, 4 additions, 2 multiplications, 2 fused multiply/add),
cannam@167:  * 11 stack variables, 0 constants, and 8 memory accesses
cannam@167:  */
cannam@167: #include "rdft/scalar/hb.h"
cannam@167: 
cannam@167: static void hb_2(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
cannam@167: {
cannam@167:      {
cannam@167: 	  INT m;
cannam@167: 	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 2, MAKE_VOLATILE_STRIDE(4, rs)) {
cannam@167: 	       E T1, T2, T6, T3, T4, T9;
cannam@167: 	       T1 = cr[0];
cannam@167: 	       T2 = ci[0];
cannam@167: 	       T6 = T1 - T2;
cannam@167: 	       T3 = ci[WS(rs, 1)];
cannam@167: 	       T4 = cr[WS(rs, 1)];
cannam@167: 	       T9 = T3 + T4;
cannam@167: 	       cr[0] = T1 + T2;
cannam@167: 	       ci[0] = T3 - T4;
cannam@167: 	       {
cannam@167: 		    E T5, T7, T8, Ta;
cannam@167: 		    T5 = W[0];
cannam@167: 		    T7 = T5 * T6;
cannam@167: 		    T8 = W[1];
cannam@167: 		    Ta = T8 * T6;
cannam@167: 		    cr[WS(rs, 1)] = FNMS(T8, T9, T7);
cannam@167: 		    ci[WS(rs, 1)] = FMA(T5, T9, Ta);
cannam@167: 	       }
cannam@167: 	  }
cannam@167:      }
cannam@167: }
cannam@167: 
cannam@167: static const tw_instr twinstr[] = {
cannam@167:      {TW_FULL, 1, 2},
cannam@167:      {TW_NEXT, 1, 0}
cannam@167: };
cannam@167: 
cannam@167: static const hc2hc_desc desc = { 2, "hb_2", twinstr, &GENUS, {4, 2, 2, 0} };
cannam@167: 
cannam@167: void X(codelet_hb_2) (planner *p) {
cannam@167:      X(khc2hc_register) (p, hb_2, &desc);
cannam@167: }
cannam@167: #else
cannam@167: 
cannam@167: /* Generated by: ../../../genfft/gen_hc2hc.native -compact -variables 4 -pipeline-latency 4 -sign 1 -n 2 -dif -name hb_2 -include rdft/scalar/hb.h */
cannam@167: 
cannam@167: /*
cannam@167:  * This function contains 6 FP additions, 4 FP multiplications,
cannam@167:  * (or, 4 additions, 2 multiplications, 2 fused multiply/add),
cannam@167:  * 9 stack variables, 0 constants, and 8 memory accesses
cannam@167:  */
cannam@167: #include "rdft/scalar/hb.h"
cannam@167: 
cannam@167: static void hb_2(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
cannam@167: {
cannam@167:      {
cannam@167: 	  INT m;
cannam@167: 	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 2, MAKE_VOLATILE_STRIDE(4, rs)) {
cannam@167: 	       E T1, T2, T6, T3, T4, T8, T5, T7;
cannam@167: 	       T1 = cr[0];
cannam@167: 	       T2 = ci[0];
cannam@167: 	       T6 = T1 - T2;
cannam@167: 	       T3 = ci[WS(rs, 1)];
cannam@167: 	       T4 = cr[WS(rs, 1)];
cannam@167: 	       T8 = T3 + T4;
cannam@167: 	       cr[0] = T1 + T2;
cannam@167: 	       ci[0] = T3 - T4;
cannam@167: 	       T5 = W[0];
cannam@167: 	       T7 = W[1];
cannam@167: 	       cr[WS(rs, 1)] = FNMS(T7, T8, T5 * T6);
cannam@167: 	       ci[WS(rs, 1)] = FMA(T7, T6, T5 * T8);
cannam@167: 	  }
cannam@167:      }
cannam@167: }
cannam@167: 
cannam@167: static const tw_instr twinstr[] = {
cannam@167:      {TW_FULL, 1, 2},
cannam@167:      {TW_NEXT, 1, 0}
cannam@167: };
cannam@167: 
cannam@167: static const hc2hc_desc desc = { 2, "hb_2", twinstr, &GENUS, {4, 2, 2, 0} };
cannam@167: 
cannam@167: void X(codelet_hb_2) (planner *p) {
cannam@167:      X(khc2hc_register) (p, hb_2, &desc);
cannam@167: }
cannam@167: #endif