comparison src/fftw-3.3.3/dft/simd/common/t2fv_2.c @ 10:37bf6b4a2645

Add FFTW3

+/*
+ * Copyright (c) 2003, 2007-11 Matteo Frigo
+ * Copyright (c) 2003, 2007-11 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 Sun Nov 25 07:38:35 EST 2012 */
+
+#include "codelet-dft.h"
+
+#ifdef HAVE_FMA
+
+/* Generated by: ../../../genfft/gen_twiddle_c.native -fma -reorder-insns -schedule-for-pipeline -simd -compact -variables 4 -pipeline-latency 8 -n 2 -name t2fv_2 -include t2f.h */
+
+/*
+ * This function contains 3 FP additions, 2 FP multiplications,
+ * (or, 3 additions, 2 multiplications, 0 fused multiply/add),
+ * 5 stack variables, 0 constants, and 4 memory accesses
+ */
+#include "t2f.h"
+
+static void t2fv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
+{
+     {
+	  INT m;
+	  R *x;
+	  x = ri;
+	  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)) {
+	       V T1, T2, T3;
+	       T1 = LD(&(x[0]), ms, &(x[0]));
+	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
+	       T3 = BYTWJ(&(W[0]), T2);
+	       ST(&(x[0]), VADD(T1, T3), ms, &(x[0]));
+	       ST(&(x[WS(rs, 1)]), VSUB(T1, T3), ms, &(x[WS(rs, 1)]));
+	  }
+     }
+     VLEAVE();
+}
+
+static const tw_instr twinstr[] = {
+     VTW(0, 1),
+     {TW_NEXT, VL, 0}
+};
+
+static const ct_desc desc = { 2, XSIMD_STRING("t2fv_2"), twinstr, &GENUS, {3, 2, 0, 0}, 0, 0, 0 };
+
+void XSIMD(codelet_t2fv_2) (planner *p) {
+     X(kdft_dit_register) (p, t2fv_2, &desc);
+}
+#else				/* HAVE_FMA */
+
+/* Generated by: ../../../genfft/gen_twiddle_c.native -simd -compact -variables 4 -pipeline-latency 8 -n 2 -name t2fv_2 -include t2f.h */
+
+/*
+ * This function contains 3 FP additions, 2 FP multiplications,
+ * (or, 3 additions, 2 multiplications, 0 fused multiply/add),
+ * 5 stack variables, 0 constants, and 4 memory accesses
+ */
+#include "t2f.h"
+
+static void t2fv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
+{
+     {
+	  INT m;
+	  R *x;
+	  x = ri;
+	  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)) {
+	       V T1, T3, T2;
+	       T1 = LD(&(x[0]), ms, &(x[0]));
+	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
+	       T3 = BYTWJ(&(W[0]), T2);
+	       ST(&(x[WS(rs, 1)]), VSUB(T1, T3), ms, &(x[WS(rs, 1)]));
+	       ST(&(x[0]), VADD(T1, T3), ms, &(x[0]));
+	  }
+     }
+     VLEAVE();
+}
+
+static const tw_instr twinstr[] = {
+     VTW(0, 1),
+     {TW_NEXT, VL, 0}
+};
+
+static const ct_desc desc = { 2, XSIMD_STRING("t2fv_2"), twinstr, &GENUS, {3, 2, 0, 0}, 0, 0, 0 };
+
+void XSIMD(codelet_t2fv_2) (planner *p) {
+     X(kdft_dit_register) (p, t2fv_2, &desc);
+}
+#endif				/* HAVE_FMA */