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 08:06:01 EDT 2018 */
Chris@82: 
Chris@82: #include "dft/codelet-dft.h"
Chris@82: 
Chris@82: #if defined(ARCH_PREFERS_FMA) || defined(ISA_EXTENSION_PREFERS_FMA)
Chris@82: 
Chris@82: /* Generated by: ../../../genfft/gen_twiddle_c.native -fma -simd -compact -variables 4 -pipeline-latency 8 -n 2 -name t2bv_2 -include dft/simd/t2b.h -sign 1 */
Chris@82: 
Chris@82: /*
Chris@82:  * This function contains 3 FP additions, 2 FP multiplications,
Chris@82:  * (or, 3 additions, 2 multiplications, 0 fused multiply/add),
Chris@82:  * 5 stack variables, 0 constants, and 4 memory accesses
Chris@82:  */
Chris@82: #include "dft/simd/t2b.h"
Chris@82: 
Chris@82: static void t2bv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
Chris@82: {
Chris@82:      {
Chris@82: 	  INT m;
Chris@82: 	  R *x;
Chris@82: 	  x = ii;
Chris@82: 	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
Chris@82: 	       V T1, T3, T2;
Chris@82: 	       T1 = LD(&(x[0]), ms, &(x[0]));
Chris@82: 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
Chris@82: 	       T3 = BYTW(&(W[0]), T2);
Chris@82: 	       ST(&(x[WS(rs, 1)]), VSUB(T1, T3), ms, &(x[WS(rs, 1)]));
Chris@82: 	       ST(&(x[0]), VADD(T1, T3), ms, &(x[0]));
Chris@82: 	  }
Chris@82:      }
Chris@82:      VLEAVE();
Chris@82: }
Chris@82: 
Chris@82: static const tw_instr twinstr[] = {
Chris@82:      VTW(0, 1),
Chris@82:      {TW_NEXT, VL, 0}
Chris@82: };
Chris@82: 
Chris@82: static const ct_desc desc = { 2, XSIMD_STRING("t2bv_2"), twinstr, &GENUS, {3, 2, 0, 0}, 0, 0, 0 };
Chris@82: 
Chris@82: void XSIMD(codelet_t2bv_2) (planner *p) {
Chris@82:      X(kdft_dit_register) (p, t2bv_2, &desc);
Chris@82: }
Chris@82: #else
Chris@82: 
Chris@82: /* Generated by: ../../../genfft/gen_twiddle_c.native -simd -compact -variables 4 -pipeline-latency 8 -n 2 -name t2bv_2 -include dft/simd/t2b.h -sign 1 */
Chris@82: 
Chris@82: /*
Chris@82:  * This function contains 3 FP additions, 2 FP multiplications,
Chris@82:  * (or, 3 additions, 2 multiplications, 0 fused multiply/add),
Chris@82:  * 5 stack variables, 0 constants, and 4 memory accesses
Chris@82:  */
Chris@82: #include "dft/simd/t2b.h"
Chris@82: 
Chris@82: static void t2bv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
Chris@82: {
Chris@82:      {
Chris@82: 	  INT m;
Chris@82: 	  R *x;
Chris@82: 	  x = ii;
Chris@82: 	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
Chris@82: 	       V T1, T3, T2;
Chris@82: 	       T1 = LD(&(x[0]), ms, &(x[0]));
Chris@82: 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
Chris@82: 	       T3 = BYTW(&(W[0]), T2);
Chris@82: 	       ST(&(x[WS(rs, 1)]), VSUB(T1, T3), ms, &(x[WS(rs, 1)]));
Chris@82: 	       ST(&(x[0]), VADD(T1, T3), ms, &(x[0]));
Chris@82: 	  }
Chris@82:      }
Chris@82:      VLEAVE();
Chris@82: }
Chris@82: 
Chris@82: static const tw_instr twinstr[] = {
Chris@82:      VTW(0, 1),
Chris@82:      {TW_NEXT, VL, 0}
Chris@82: };
Chris@82: 
Chris@82: static const ct_desc desc = { 2, XSIMD_STRING("t2bv_2"), twinstr, &GENUS, {3, 2, 0, 0}, 0, 0, 0 };
Chris@82: 
Chris@82: void XSIMD(codelet_t2bv_2) (planner *p) {
Chris@82:      X(kdft_dit_register) (p, t2bv_2, &desc);
Chris@82: }
Chris@82: #endif