First commit - moving from LuaForge to SourceForge

1 files changed, 203 insertions, 0 deletions

diff --git a/src/fftw3/rdft/codelets/r2hc/hf_8.c b/src/fftw3/rdft/codelets/r2hc/hf_8.c
new file mode 100644
index 0000000..21dedce
--- /dev/null
+++ b/src/fftw3/rdft/codelets/r2hc/hf_8.c
@@ -0,0 +1,203 @@
+/*
+ * Copyright (c) 2003 Matteo Frigo
+ * Copyright (c) 2003 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., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+/* This file was automatically generated --- DO NOT EDIT */
+/* Generated on Sat Jul  5 21:56:58 EDT 2003 */
+
+#include "codelet-rdft.h"
+
+/* Generated by: /homee/stevenj/cvs/fftw3.0.1/genfft/gen_hc2hc -compact -variables 4 -n 8 -dit -name hf_8 -include hf.h */
+
+/*
+ * This function contains 66 FP additions, 32 FP multiplications,
+ * (or, 52 additions, 18 multiplications, 14 fused multiply/add),
+ * 28 stack variables, and 32 memory accesses
+ */
+/*
+ * Generator Id's : 
+ * $Id: hf_8.c,v 1.1 2008/10/17 06:12:34 scuri Exp $
+ * $Id: hf_8.c,v 1.1 2008/10/17 06:12:34 scuri Exp $
+ * $Id: hf_8.c,v 1.1 2008/10/17 06:12:34 scuri Exp $
+ */
+
+#include "hf.h"
+
+static const R *hf_8(R *rio, R *iio, const R *W, stride ios, int m, int dist)
+{
+     DK(KP707106781, +0.707106781186547524400844362104849039284835938);
+     int i;
+     for (i = m - 2; i > 0; i = i - 2, rio = rio + dist, iio = iio - dist, W = W + 14) {
+	  E T7, T1e, TH, T19, TF, T13, TR, TU, Ti, T1f, TK, T16, Tu, T12, TM;
+	  E TP;
+	  {
+	       E T1, T18, T6, T17;
+	       T1 = rio[0];
+	       T18 = iio[-WS(ios, 7)];
+	       {
+		    E T3, T5, T2, T4;
+		    T3 = rio[WS(ios, 4)];
+		    T5 = iio[-WS(ios, 3)];
+		    T2 = W[6];
+		    T4 = W[7];
+		    T6 = FMA(T2, T3, T4 * T5);
+		    T17 = FNMS(T4, T3, T2 * T5);
+	       }
+	       T7 = T1 + T6;
+	       T1e = T18 - T17;
+	       TH = T1 - T6;
+	       T19 = T17 + T18;
+	  }
+	  {
+	       E Tz, TS, TE, TT;
+	       {
+		    E Tw, Ty, Tv, Tx;
+		    Tw = rio[WS(ios, 7)];
+		    Ty = iio[0];
+		    Tv = W[12];
+		    Tx = W[13];
+		    Tz = FMA(Tv, Tw, Tx * Ty);
+		    TS = FNMS(Tx, Tw, Tv * Ty);
+	       }
+	       {
+		    E TB, TD, TA, TC;
+		    TB = rio[WS(ios, 3)];
+		    TD = iio[-WS(ios, 4)];
+		    TA = W[4];
+		    TC = W[5];
+		    TE = FMA(TA, TB, TC * TD);
+		    TT = FNMS(TC, TB, TA * TD);
+	       }
+	       TF = Tz + TE;
+	       T13 = TS + TT;
+	       TR = Tz - TE;
+	       TU = TS - TT;
+	  }
+	  {
+	       E Tc, TI, Th, TJ;
+	       {
+		    E T9, Tb, T8, Ta;
+		    T9 = rio[WS(ios, 2)];
+		    Tb = iio[-WS(ios, 5)];
+		    T8 = W[2];
+		    Ta = W[3];
+		    Tc = FMA(T8, T9, Ta * Tb);
+		    TI = FNMS(Ta, T9, T8 * Tb);
+	       }
+	       {
+		    E Te, Tg, Td, Tf;
+		    Te = rio[WS(ios, 6)];
+		    Tg = iio[-WS(ios, 1)];
+		    Td = W[10];
+		    Tf = W[11];
+		    Th = FMA(Td, Te, Tf * Tg);
+		    TJ = FNMS(Tf, Te, Td * Tg);
+	       }
+	       Ti = Tc + Th;
+	       T1f = Tc - Th;
+	       TK = TI - TJ;
+	       T16 = TI + TJ;
+	  }
+	  {
+	       E To, TN, Tt, TO;
+	       {
+		    E Tl, Tn, Tk, Tm;
+		    Tl = rio[WS(ios, 1)];
+		    Tn = iio[-WS(ios, 6)];
+		    Tk = W[0];
+		    Tm = W[1];
+		    To = FMA(Tk, Tl, Tm * Tn);
+		    TN = FNMS(Tm, Tl, Tk * Tn);
+	       }
+	       {
+		    E Tq, Ts, Tp, Tr;
+		    Tq = rio[WS(ios, 5)];
+		    Ts = iio[-WS(ios, 2)];
+		    Tp = W[8];
+		    Tr = W[9];
+		    Tt = FMA(Tp, Tq, Tr * Ts);
+		    TO = FNMS(Tr, Tq, Tp * Ts);
+	       }
+	       Tu = To + Tt;
+	       T12 = TN + TO;
+	       TM = To - Tt;
+	       TP = TN - TO;
+	  }
+	  {
+	       E Tj, TG, T1b, T1c;
+	       Tj = T7 + Ti;
+	       TG = Tu + TF;
+	       iio[-WS(ios, 4)] = Tj - TG;
+	       rio[0] = Tj + TG;
+	       {
+		    E T15, T1a, T11, T14;
+		    T15 = T12 + T13;
+		    T1a = T16 + T19;
+		    rio[WS(ios, 4)] = T15 - T1a;
+		    iio[0] = T15 + T1a;
+		    T11 = T7 - Ti;
+		    T14 = T12 - T13;
+		    iio[-WS(ios, 6)] = T11 - T14;
+		    rio[WS(ios, 2)] = T11 + T14;
+	       }
+	       T1b = TF - Tu;
+	       T1c = T19 - T16;
+	       rio[WS(ios, 6)] = T1b - T1c;
+	       iio[-WS(ios, 2)] = T1b + T1c;
+	       {
+		    E TX, T1g, T10, T1d, TY, TZ;
+		    TX = TH - TK;
+		    T1g = T1e - T1f;
+		    TY = TP - TM;
+		    TZ = TR + TU;
+		    T10 = KP707106781 * (TY - TZ);
+		    T1d = KP707106781 * (TY + TZ);
+		    iio[-WS(ios, 7)] = TX - T10;
+		    iio[-WS(ios, 1)] = T1d + T1g;
+		    rio[WS(ios, 3)] = TX + T10;
+		    rio[WS(ios, 5)] = T1d - T1g;
+	       }
+	       {
+		    E TL, T1i, TW, T1h, TQ, TV;
+		    TL = TH + TK;
+		    T1i = T1f + T1e;
+		    TQ = TM + TP;
+		    TV = TR - TU;
+		    TW = KP707106781 * (TQ + TV);
+		    T1h = KP707106781 * (TV - TQ);
+		    iio[-WS(ios, 5)] = TL - TW;
+		    iio[-WS(ios, 3)] = T1h + T1i;
+		    rio[WS(ios, 1)] = TL + TW;
+		    rio[WS(ios, 7)] = T1h - T1i;
+	       }
+	  }
+     }
+     return W;
+}
+
+static const tw_instr twinstr[] = {
+     {TW_FULL, 0, 8},
+     {TW_NEXT, 1, 0}
+};
+
+static const hc2hc_desc desc = { 8, "hf_8", twinstr, {52, 18, 14, 0}, &GENUS, 0, 0, 0 };
+
+void X(codelet_hf_8) (planner *p) {
+     X(khc2hc_dit_register) (p, hf_8, &desc);
+}