diff options
author | scuri <scuri> | 2008-10-17 06:10:15 +0000 |
---|---|---|
committer | scuri <scuri> | 2008-10-17 06:10:15 +0000 |
commit | 5a422aba704c375a307a902bafe658342e209906 (patch) | |
tree | 5005011e086bb863d8fb587ad3319bbec59b2447 /src/fftw3/dft/codelets/standard/n1_10.c |
First commit - moving from LuaForge to SourceForge
Diffstat (limited to 'src/fftw3/dft/codelets/standard/n1_10.c')
-rw-r--r-- | src/fftw3/dft/codelets/standard/n1_10.c | 195 |
1 files changed, 195 insertions, 0 deletions
diff --git a/src/fftw3/dft/codelets/standard/n1_10.c b/src/fftw3/dft/codelets/standard/n1_10.c new file mode 100644 index 0000000..95fb984 --- /dev/null +++ b/src/fftw3/dft/codelets/standard/n1_10.c @@ -0,0 +1,195 @@ +/* + * 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 21:29:32 EDT 2003 */ + +#include "codelet-dft.h" + +/* Generated by: /homee/stevenj/cvs/fftw3.0.1/genfft/gen_notw -compact -variables 4 -n 10 -name n1_10 -include n.h */ + +/* + * This function contains 84 FP additions, 24 FP multiplications, + * (or, 72 additions, 12 multiplications, 12 fused multiply/add), + * 41 stack variables, and 40 memory accesses + */ +/* + * Generator Id's : + * $Id: n1_10.c,v 1.1 2008/10/17 06:11:09 scuri Exp $ + * $Id: n1_10.c,v 1.1 2008/10/17 06:11:09 scuri Exp $ + * $Id: n1_10.c,v 1.1 2008/10/17 06:11:09 scuri Exp $ + */ + +#include "n.h" + +static void n1_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, int v, int ivs, int ovs) +{ + DK(KP250000000, +0.250000000000000000000000000000000000000000000); + DK(KP559016994, +0.559016994374947424102293417182819058860154590); + DK(KP587785252, +0.587785252292473129168705954639072768597652438); + DK(KP951056516, +0.951056516295153572116439333379382143405698634); + int i; + for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs) { + E T3, Tj, TQ, T1e, TU, TV, T1c, T1b, Tm, Tp, Tq, Ta, Th, Ti, TA; + E TH, T17, T14, T1f, T1g, T1h, TL, TM, TR; + { + E T1, T2, TO, TP; + T1 = ri[0]; + T2 = ri[WS(is, 5)]; + T3 = T1 - T2; + Tj = T1 + T2; + TO = ii[0]; + TP = ii[WS(is, 5)]; + TQ = TO - TP; + T1e = TO + TP; + } + { + E T6, Tk, Tg, To, T9, Tl, Td, Tn; + { + E T4, T5, Te, Tf; + T4 = ri[WS(is, 2)]; + T5 = ri[WS(is, 7)]; + T6 = T4 - T5; + Tk = T4 + T5; + Te = ri[WS(is, 6)]; + Tf = ri[WS(is, 1)]; + Tg = Te - Tf; + To = Te + Tf; + } + { + E T7, T8, Tb, Tc; + T7 = ri[WS(is, 8)]; + T8 = ri[WS(is, 3)]; + T9 = T7 - T8; + Tl = T7 + T8; + Tb = ri[WS(is, 4)]; + Tc = ri[WS(is, 9)]; + Td = Tb - Tc; + Tn = Tb + Tc; + } + TU = T6 - T9; + TV = Td - Tg; + T1c = Tk - Tl; + T1b = Tn - To; + Tm = Tk + Tl; + Tp = Tn + To; + Tq = Tm + Tp; + Ta = T6 + T9; + Th = Td + Tg; + Ti = Ta + Th; + } + { + E Tw, T15, TG, T13, Tz, T16, TD, T12; + { + E Tu, Tv, TE, TF; + Tu = ii[WS(is, 2)]; + Tv = ii[WS(is, 7)]; + Tw = Tu - Tv; + T15 = Tu + Tv; + TE = ii[WS(is, 6)]; + TF = ii[WS(is, 1)]; + TG = TE - TF; + T13 = TE + TF; + } + { + E Tx, Ty, TB, TC; + Tx = ii[WS(is, 8)]; + Ty = ii[WS(is, 3)]; + Tz = Tx - Ty; + T16 = Tx + Ty; + TB = ii[WS(is, 4)]; + TC = ii[WS(is, 9)]; + TD = TB - TC; + T12 = TB + TC; + } + TA = Tw - Tz; + TH = TD - TG; + T17 = T15 - T16; + T14 = T12 - T13; + T1f = T15 + T16; + T1g = T12 + T13; + T1h = T1f + T1g; + TL = Tw + Tz; + TM = TD + TG; + TR = TL + TM; + } + ro[WS(os, 5)] = T3 + Ti; + io[WS(os, 5)] = TQ + TR; + ro[0] = Tj + Tq; + io[0] = T1e + T1h; + { + E TI, TK, Tt, TJ, Tr, Ts; + TI = FMA(KP951056516, TA, KP587785252 * TH); + TK = FNMS(KP587785252, TA, KP951056516 * TH); + Tr = KP559016994 * (Ta - Th); + Ts = FNMS(KP250000000, Ti, T3); + Tt = Tr + Ts; + TJ = Ts - Tr; + ro[WS(os, 9)] = Tt - TI; + ro[WS(os, 3)] = TJ + TK; + ro[WS(os, 1)] = Tt + TI; + ro[WS(os, 7)] = TJ - TK; + } + { + E TW, TY, TT, TX, TN, TS; + TW = FMA(KP951056516, TU, KP587785252 * TV); + TY = FNMS(KP587785252, TU, KP951056516 * TV); + TN = KP559016994 * (TL - TM); + TS = FNMS(KP250000000, TR, TQ); + TT = TN + TS; + TX = TS - TN; + io[WS(os, 1)] = TT - TW; + io[WS(os, 7)] = TY + TX; + io[WS(os, 9)] = TW + TT; + io[WS(os, 3)] = TX - TY; + } + { + E T18, T1a, T11, T19, TZ, T10; + T18 = FNMS(KP587785252, T17, KP951056516 * T14); + T1a = FMA(KP951056516, T17, KP587785252 * T14); + TZ = FNMS(KP250000000, Tq, Tj); + T10 = KP559016994 * (Tm - Tp); + T11 = TZ - T10; + T19 = T10 + TZ; + ro[WS(os, 2)] = T11 - T18; + ro[WS(os, 6)] = T19 + T1a; + ro[WS(os, 8)] = T11 + T18; + ro[WS(os, 4)] = T19 - T1a; + } + { + E T1d, T1l, T1k, T1m, T1i, T1j; + T1d = FNMS(KP587785252, T1c, KP951056516 * T1b); + T1l = FMA(KP951056516, T1c, KP587785252 * T1b); + T1i = FNMS(KP250000000, T1h, T1e); + T1j = KP559016994 * (T1f - T1g); + T1k = T1i - T1j; + T1m = T1j + T1i; + io[WS(os, 2)] = T1d + T1k; + io[WS(os, 6)] = T1m - T1l; + io[WS(os, 8)] = T1k - T1d; + io[WS(os, 4)] = T1l + T1m; + } + } +} + +static const kdft_desc desc = { 10, "n1_10", {72, 12, 12, 0}, &GENUS, 0, 0, 0, 0 }; +void X(codelet_n1_10) (planner *p) { + X(kdft_register) (p, n1_10, &desc); +} |