/* * 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 22:11:12 EDT 2003 */ #include "codelet-rdft.h" /* Generated by: /homee/stevenj/cvs/fftw3.0.1/genfft/gen_hc2r -compact -variables 4 -sign 1 -n 13 -name hc2r_13 -include hc2r.h */ /* * This function contains 76 FP additions, 35 FP multiplications, * (or, 56 additions, 15 multiplications, 20 fused multiply/add), * 56 stack variables, and 26 memory accesses */ /* * Generator Id's : * $Id: hc2r_13.c,v 1.1 2008/10/17 06:12:08 scuri Exp $ * $Id: hc2r_13.c,v 1.1 2008/10/17 06:12:08 scuri Exp $ * $Id: hc2r_13.c,v 1.1 2008/10/17 06:12:08 scuri Exp $ */ #include "hc2r.h" static void hc2r_13(const R *ri, const R *ii, R *O, stride ris, stride iis, stride os, int v, int ivs, int ovs) { DK(KP1_007074065, +1.007074065727533254493747707736933954186697125); DK(KP227708958, +0.227708958111581597949308691735310621069285120); DK(KP531932498, +0.531932498429674575175042127684371897596660533); DK(KP774781170, +0.774781170935234584261351932853525703557550433); DK(KP265966249, +0.265966249214837287587521063842185948798330267); DK(KP516520780, +0.516520780623489722840901288569017135705033622); DK(KP151805972, +0.151805972074387731966205794490207080712856746); DK(KP503537032, +0.503537032863766627246873853868466977093348562); DK(KP166666666, +0.166666666666666666666666666666666666666666667); DK(KP600925212, +0.600925212577331548853203544578415991041882762); DK(KP500000000, +0.500000000000000000000000000000000000000000000); DK(KP256247671, +0.256247671582936600958684654061725059144125175); DK(KP156891391, +0.156891391051584611046832726756003269660212636); DK(KP348277202, +0.348277202304271810011321589858529485233929352); DK(KP1_150281458, +1.150281458948006242736771094910906776922003215); DK(KP300238635, +0.300238635966332641462884626667381504676006424); DK(KP011599105, +0.011599105605768290721655456654083252189827041); DK(KP1_732050807, +1.732050807568877293527446341505872366942805254); DK(KP2_000000000, +2.000000000000000000000000000000000000000000000); int i; for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, O = O + ovs) { E TG, TS, TR, T15, TJ, TT, T1, Tm, Tc, Td, Tg, Tj, Tk, Tn, To; E Tp; { E Ts, Tv, Tw, TE, TC, TB, Tz, TD, TA, TF; { E Tt, Tu, Tx, Ty; Ts = ii[WS(iis, 1)]; Tt = ii[WS(iis, 3)]; Tu = ii[WS(iis, 4)]; Tv = Tt - Tu; Tw = FMS(KP2_000000000, Ts, Tv); TE = KP1_732050807 * (Tt + Tu); TC = ii[WS(iis, 5)]; Tx = ii[WS(iis, 6)]; Ty = ii[WS(iis, 2)]; TB = Tx + Ty; Tz = KP1_732050807 * (Tx - Ty); TD = FNMS(KP2_000000000, TC, TB); } TA = Tw + Tz; TF = TD - TE; TG = FMA(KP011599105, TA, KP300238635 * TF); TS = FNMS(KP011599105, TF, KP300238635 * TA); { E TP, TQ, TH, TI; TP = Ts + Tv; TQ = TB + TC; TR = FNMS(KP348277202, TQ, KP1_150281458 * TP); T15 = FMA(KP348277202, TP, KP1_150281458 * TQ); TH = Tw - Tz; TI = TE + TD; TJ = FMA(KP156891391, TH, KP256247671 * TI); TT = FNMS(KP256247671, TH, KP156891391 * TI); } } { E Tb, Ti, Tf, T6, Th, Te; T1 = ri[0]; { E T7, T8, T9, Ta; T7 = ri[WS(ris, 5)]; T8 = ri[WS(ris, 2)]; T9 = ri[WS(ris, 6)]; Ta = T8 + T9; Tb = T7 + Ta; Ti = FNMS(KP500000000, Ta, T7); Tf = T8 - T9; } { E T2, T3, T4, T5; T2 = ri[WS(ris, 1)]; T3 = ri[WS(ris, 3)]; T4 = ri[WS(ris, 4)]; T5 = T3 + T4; T6 = T2 + T5; Th = FNMS(KP500000000, T5, T2); Te = T3 - T4; } Tm = KP600925212 * (T6 - Tb); Tc = T6 + Tb; Td = FNMS(KP166666666, Tc, T1); Tg = Te + Tf; Tj = Th + Ti; Tk = FMA(KP503537032, Tg, KP151805972 * Tj); Tn = Th - Ti; To = Te - Tf; Tp = FNMS(KP265966249, To, KP516520780 * Tn); } O[0] = FMA(KP2_000000000, Tc, T1); { E TK, T1b, TV, T12, T16, T18, TO, T1a, Tr, T17, T11, T13; { E TU, T14, TM, TN; TK = KP1_732050807 * (TG + TJ); T1b = KP1_732050807 * (TS - TT); TU = TS + TT; TV = TR - TU; T12 = FMA(KP2_000000000, TU, TR); T14 = TG - TJ; T16 = FMS(KP2_000000000, T14, T15); T18 = T14 + T15; TM = FMA(KP774781170, To, KP531932498 * Tn); TN = FNMS(KP1_007074065, Tj, KP227708958 * Tg); TO = TM - TN; T1a = TM + TN; { E Tl, Tq, TZ, T10; Tl = Td - Tk; Tq = Tm - Tp; Tr = Tl - Tq; T17 = Tq + Tl; TZ = FMA(KP2_000000000, Tk, Td); T10 = FMA(KP2_000000000, Tp, Tm); T11 = TZ - T10; T13 = T10 + TZ; } } O[WS(os, 5)] = T11 - T12; O[WS(os, 12)] = T13 - T16; O[WS(os, 1)] = T13 + T16; O[WS(os, 8)] = T11 + T12; { E TL, TW, T19, T1c; TL = Tr - TK; TW = TO - TV; O[WS(os, 7)] = TL - TW; O[WS(os, 2)] = TL + TW; T19 = T17 - T18; T1c = T1a + T1b; O[WS(os, 3)] = T19 - T1c; O[WS(os, 9)] = T1c + T19; } { E T1d, T1e, TX, TY; T1d = T1a - T1b; T1e = T17 + T18; O[WS(os, 4)] = T1d + T1e; O[WS(os, 10)] = T1e - T1d; TX = Tr + TK; TY = TO + TV; O[WS(os, 6)] = TX - TY; O[WS(os, 11)] = TX + TY; } } } } static const khc2r_desc desc = { 13, "hc2r_13", {56, 15, 20, 0}, &GENUS, 0, 0, 0, 0, 0 }; void X(codelet_hc2r_13) (planner *p) { X(khc2r_register) (p, hc2r_13, &desc); }