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 08:06:26 EDT 2018 */
Chris@82: 
Chris@82: #include "rdft/codelet-rdft.h"
Chris@82: 
Chris@82: #if defined(ARCH_PREFERS_FMA) || defined(ISA_EXTENSION_PREFERS_FMA)
Chris@82: 
Chris@82: /* Generated by: ../../../genfft/gen_r2cf.native -fma -compact -variables 4 -pipeline-latency 4 -n 13 -name r2cf_13 -include rdft/scalar/r2cf.h */
Chris@82: 
Chris@82: /*
Chris@82:  * This function contains 76 FP additions, 51 FP multiplications,
Chris@82:  * (or, 31 additions, 6 multiplications, 45 fused multiply/add),
Chris@82:  * 58 stack variables, 23 constants, and 26 memory accesses
Chris@82:  */
Chris@82: #include "rdft/scalar/r2cf.h"
Chris@82: 
Chris@82: static void r2cf_13(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs)
Chris@82: {
Chris@82:      DK(KP300462606, +0.300462606288665774426601772289207995520941381);
Chris@82:      DK(KP516520780, +0.516520780623489722840901288569017135705033622);
Chris@82:      DK(KP859542535, +0.859542535098774820163672132761689612766401925);
Chris@82:      DK(KP581704778, +0.581704778510515730456870384989698884939833902);
Chris@82:      DK(KP514918778, +0.514918778086315755491789696138117261566051239);
Chris@82:      DK(KP769338817, +0.769338817572980603471413688209101117038278899);
Chris@82:      DK(KP686558370, +0.686558370781754340655719594850823015421401653);
Chris@82:      DK(KP226109445, +0.226109445035782405468510155372505010481906348);
Chris@82:      DK(KP251768516, +0.251768516431883313623436926934233488546674281);
Chris@82:      DK(KP503537032, +0.503537032863766627246873853868466977093348562);
Chris@82:      DK(KP301479260, +0.301479260047709873958013540496673347309208464);
Chris@82:      DK(KP083333333, +0.083333333333333333333333333333333333333333333);
Chris@82:      DK(KP904176221, +0.904176221990848204433795481776887926501523162);
Chris@82:      DK(KP575140729, +0.575140729474003121368385547455453388461001608);
Chris@82:      DK(KP522026385, +0.522026385161275033714027226654165028300441940);
Chris@82:      DK(KP957805992, +0.957805992594665126462521754605754580515587217);
Chris@82:      DK(KP600477271, +0.600477271932665282925769253334763009352012849);
Chris@82:      DK(KP853480001, +0.853480001859823990758994934970528322872359049);
Chris@82:      DK(KP612264650, +0.612264650376756543746494474777125408779395514);
Chris@82:      DK(KP038632954, +0.038632954644348171955506895830342264440241080);
Chris@82:      DK(KP302775637, +0.302775637731994646559610633735247973125648287);
Chris@82:      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
Chris@82:      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
Chris@82:      {
Chris@82: 	  INT i;
Chris@82: 	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(52, rs), MAKE_VOLATILE_STRIDE(52, csr), MAKE_VOLATILE_STRIDE(52, csi)) {
Chris@82: 	       E TN, TA, TD, TO, TR, TS, TZ, T12, Tu, Tx, Tj, Tw, TW, T13;
Chris@82: 	       TN = R0[0];
Chris@82: 	       {
Chris@82: 		    E T3, TP, Th, TB, Tp, Te, TC, Tm, T6, Tr, T9, Ts, Ta, TQ, T1;
Chris@82: 		    E T2;
Chris@82: 		    T1 = R0[WS(rs, 4)];
Chris@82: 		    T2 = R1[WS(rs, 2)];
Chris@82: 		    T3 = T1 - T2;
Chris@82: 		    TP = T1 + T2;
Chris@82: 		    {
Chris@82: 			 E Tn, Tf, Tg, To;
Chris@82: 			 Tn = R0[WS(rs, 6)];
Chris@82: 			 Tf = R0[WS(rs, 5)];
Chris@82: 			 Tg = R0[WS(rs, 2)];
Chris@82: 			 To = Tf + Tg;
Chris@82: 			 Th = Tf - Tg;
Chris@82: 			 TB = Tn + To;
Chris@82: 			 Tp = FMS(KP500000000, To, Tn);
Chris@82: 		    }
Chris@82: 		    {
Chris@82: 			 E Tk, Tc, Td, Tl;
Chris@82: 			 Tk = R1[0];
Chris@82: 			 Tc = R1[WS(rs, 4)];
Chris@82: 			 Td = R1[WS(rs, 1)];
Chris@82: 			 Tl = Td + Tc;
Chris@82: 			 Te = Tc - Td;
Chris@82: 			 TC = Tk + Tl;
Chris@82: 			 Tm = FNMS(KP500000000, Tl, Tk);
Chris@82: 		    }
Chris@82: 		    {
Chris@82: 			 E T4, T5, T7, T8;
Chris@82: 			 T4 = R1[WS(rs, 5)];
Chris@82: 			 T5 = R0[WS(rs, 3)];
Chris@82: 			 T6 = T4 - T5;
Chris@82: 			 Tr = T4 + T5;
Chris@82: 			 T7 = R1[WS(rs, 3)];
Chris@82: 			 T8 = R0[WS(rs, 1)];
Chris@82: 			 T9 = T7 - T8;
Chris@82: 			 Ts = T7 + T8;
Chris@82: 		    }
Chris@82: 		    Ta = T6 + T9;
Chris@82: 		    TQ = Tr + Ts;
Chris@82: 		    TA = T3 + Ta;
Chris@82: 		    TD = TB - TC;
Chris@82: 		    TO = TC + TB;
Chris@82: 		    TR = TP + TQ;
Chris@82: 		    TS = TO + TR;
Chris@82: 		    {
Chris@82: 			 E TX, TY, Tq, Tt;
Chris@82: 			 TX = Tm - Tp;
Chris@82: 			 TY = FNMS(KP500000000, TQ, TP);
Chris@82: 			 TZ = TX + TY;
Chris@82: 			 T12 = TX - TY;
Chris@82: 			 Tq = Tm + Tp;
Chris@82: 			 Tt = Tr - Ts;
Chris@82: 			 Tu = FMA(KP866025403, Tt, Tq);
Chris@82: 			 Tx = FNMS(KP866025403, Tt, Tq);
Chris@82: 		    }
Chris@82: 		    {
Chris@82: 			 E Tb, Ti, TU, TV;
Chris@82: 			 Tb = FNMS(KP500000000, Ta, T3);
Chris@82: 			 Ti = Te + Th;
Chris@82: 			 Tj = FMA(KP866025403, Ti, Tb);
Chris@82: 			 Tw = FNMS(KP866025403, Ti, Tb);
Chris@82: 			 TU = Th - Te;
Chris@82: 			 TV = T6 - T9;
Chris@82: 			 TW = TU + TV;
Chris@82: 			 T13 = TU - TV;
Chris@82: 		    }
Chris@82: 	       }
Chris@82: 	       Cr[0] = TN + TS;
Chris@82: 	       {
Chris@82: 		    E TE, TI, Tz, TK, TH, TM, TJ, TL;
Chris@82: 		    TE = FMA(KP302775637, TD, TA);
Chris@82: 		    TI = FNMS(KP302775637, TA, TD);
Chris@82: 		    {
Chris@82: 			 E Tv, Ty, TF, TG;
Chris@82: 			 Tv = FMA(KP038632954, Tu, Tj);
Chris@82: 			 Ty = FMA(KP612264650, Tx, Tw);
Chris@82: 			 Tz = FNMS(KP853480001, Ty, Tv);
Chris@82: 			 TK = FMA(KP853480001, Ty, Tv);
Chris@82: 			 TF = FNMS(KP038632954, Tj, Tu);
Chris@82: 			 TG = FNMS(KP612264650, Tw, Tx);
Chris@82: 			 TH = FNMS(KP853480001, TG, TF);
Chris@82: 			 TM = FMA(KP853480001, TG, TF);
Chris@82: 		    }
Chris@82: 		    Ci[WS(csi, 1)] = KP600477271 * (FMA(KP957805992, TE, Tz));
Chris@82: 		    Ci[WS(csi, 5)] = -(KP600477271 * (FNMS(KP957805992, TI, TH)));
Chris@82: 		    TJ = FMA(KP522026385, TH, TI);
Chris@82: 		    Ci[WS(csi, 2)] = KP575140729 * (FNMS(KP904176221, TK, TJ));
Chris@82: 		    Ci[WS(csi, 6)] = KP575140729 * (FMA(KP904176221, TK, TJ));
Chris@82: 		    TL = FNMS(KP522026385, Tz, TE);
Chris@82: 		    Ci[WS(csi, 3)] = KP575140729 * (FNMS(KP904176221, TM, TL));
Chris@82: 		    Ci[WS(csi, 4)] = -(KP575140729 * (FMA(KP904176221, TM, TL)));
Chris@82: 	       }
Chris@82: 	       {
Chris@82: 		    E T11, T17, T1c, T1e, T16, T18, TT, T10, T19, T1d;
Chris@82: 		    TT = FNMS(KP083333333, TS, TN);
Chris@82: 		    T10 = FMA(KP301479260, TZ, TW);
Chris@82: 		    T11 = FMA(KP503537032, T10, TT);
Chris@82: 		    T17 = FNMS(KP251768516, T10, TT);
Chris@82: 		    {
Chris@82: 			 E T1a, T1b, T14, T15;
Chris@82: 			 T1a = FNMS(KP226109445, TW, TZ);
Chris@82: 			 T1b = FMA(KP686558370, T12, T13);
Chris@82: 			 T1c = FNMS(KP769338817, T1b, T1a);
Chris@82: 			 T1e = FMA(KP769338817, T1b, T1a);
Chris@82: 			 T14 = FNMS(KP514918778, T13, T12);
Chris@82: 			 T15 = TO - TR;
Chris@82: 			 T16 = FMA(KP581704778, T15, T14);
Chris@82: 			 T18 = FNMS(KP859542535, T14, T15);
Chris@82: 		    }
Chris@82: 		    Cr[WS(csr, 5)] = FNMS(KP516520780, T16, T11);
Chris@82: 		    Cr[WS(csr, 1)] = FMA(KP516520780, T16, T11);
Chris@82: 		    T19 = FMA(KP300462606, T18, T17);
Chris@82: 		    Cr[WS(csr, 4)] = FNMS(KP503537032, T1c, T19);
Chris@82: 		    Cr[WS(csr, 3)] = FMA(KP503537032, T1c, T19);
Chris@82: 		    T1d = FNMS(KP300462606, T18, T17);
Chris@82: 		    Cr[WS(csr, 6)] = FNMS(KP503537032, T1e, T1d);
Chris@82: 		    Cr[WS(csr, 2)] = FMA(KP503537032, T1e, T1d);
Chris@82: 	       }
Chris@82: 	  }
Chris@82:      }
Chris@82: }
Chris@82: 
Chris@82: static const kr2c_desc desc = { 13, "r2cf_13", {31, 6, 45, 0}, &GENUS };
Chris@82: 
Chris@82: void X(codelet_r2cf_13) (planner *p) {
Chris@82:      X(kr2c_register) (p, r2cf_13, &desc);
Chris@82: }
Chris@82: 
Chris@82: #else
Chris@82: 
Chris@82: /* Generated by: ../../../genfft/gen_r2cf.native -compact -variables 4 -pipeline-latency 4 -n 13 -name r2cf_13 -include rdft/scalar/r2cf.h */
Chris@82: 
Chris@82: /*
Chris@82:  * This function contains 76 FP additions, 34 FP multiplications,
Chris@82:  * (or, 57 additions, 15 multiplications, 19 fused multiply/add),
Chris@82:  * 55 stack variables, 20 constants, and 26 memory accesses
Chris@82:  */
Chris@82: #include "rdft/scalar/r2cf.h"
Chris@82: 
Chris@82: static void r2cf_13(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs)
Chris@82: {
Chris@82:      DK(KP083333333, +0.083333333333333333333333333333333333333333333);
Chris@82:      DK(KP075902986, +0.075902986037193865983102897245103540356428373);
Chris@82:      DK(KP251768516, +0.251768516431883313623436926934233488546674281);
Chris@82:      DK(KP503537032, +0.503537032863766627246873853868466977093348562);
Chris@82:      DK(KP113854479, +0.113854479055790798974654345867655310534642560);
Chris@82:      DK(KP265966249, +0.265966249214837287587521063842185948798330267);
Chris@82:      DK(KP387390585, +0.387390585467617292130675966426762851778775217);
Chris@82:      DK(KP300462606, +0.300462606288665774426601772289207995520941381);
Chris@82:      DK(KP132983124, +0.132983124607418643793760531921092974399165133);
Chris@82:      DK(KP258260390, +0.258260390311744861420450644284508567852516811);
Chris@82:      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
Chris@82:      DK(KP1_732050807, +1.732050807568877293527446341505872366942805254);
Chris@82:      DK(KP300238635, +0.300238635966332641462884626667381504676006424);
Chris@82:      DK(KP011599105, +0.011599105605768290721655456654083252189827041);
Chris@82:      DK(KP156891391, +0.156891391051584611046832726756003269660212636);
Chris@82:      DK(KP256247671, +0.256247671582936600958684654061725059144125175);
Chris@82:      DK(KP174138601, +0.174138601152135905005660794929264742616964676);
Chris@82:      DK(KP575140729, +0.575140729474003121368385547455453388461001608);
Chris@82:      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
Chris@82:      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
Chris@82:      {
Chris@82: 	  INT i;
Chris@82: 	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(52, rs), MAKE_VOLATILE_STRIDE(52, csr), MAKE_VOLATILE_STRIDE(52, csi)) {
Chris@82: 	       E T13, Tb, Tm, TW, TX, T14, TU, T10, Tz, TB, Tu, TC, TR, T11;
Chris@82: 	       T13 = R0[0];
Chris@82: 	       {
Chris@82: 		    E Te, TO, Ta, Tv, To, T5, Tw, Tp, Th, Tr, Tk, Ts, Tl, TP, Tc;
Chris@82: 		    E Td;
Chris@82: 		    Tc = R0[WS(rs, 4)];
Chris@82: 		    Td = R1[WS(rs, 2)];
Chris@82: 		    Te = Tc - Td;
Chris@82: 		    TO = Tc + Td;
Chris@82: 		    {
Chris@82: 			 E T6, T7, T8, T9;
Chris@82: 			 T6 = R1[0];
Chris@82: 			 T7 = R1[WS(rs, 1)];
Chris@82: 			 T8 = R1[WS(rs, 4)];
Chris@82: 			 T9 = T7 + T8;
Chris@82: 			 Ta = T6 + T9;
Chris@82: 			 Tv = T7 - T8;
Chris@82: 			 To = FNMS(KP500000000, T9, T6);
Chris@82: 		    }
Chris@82: 		    {
Chris@82: 			 E T1, T2, T3, T4;
Chris@82: 			 T1 = R0[WS(rs, 6)];
Chris@82: 			 T2 = R0[WS(rs, 5)];
Chris@82: 			 T3 = R0[WS(rs, 2)];
Chris@82: 			 T4 = T2 + T3;
Chris@82: 			 T5 = T1 + T4;
Chris@82: 			 Tw = T2 - T3;
Chris@82: 			 Tp = FNMS(KP500000000, T4, T1);
Chris@82: 		    }
Chris@82: 		    {
Chris@82: 			 E Tf, Tg, Ti, Tj;
Chris@82: 			 Tf = R1[WS(rs, 5)];
Chris@82: 			 Tg = R0[WS(rs, 3)];
Chris@82: 			 Th = Tf - Tg;
Chris@82: 			 Tr = Tf + Tg;
Chris@82: 			 Ti = R1[WS(rs, 3)];
Chris@82: 			 Tj = R0[WS(rs, 1)];
Chris@82: 			 Tk = Ti - Tj;
Chris@82: 			 Ts = Ti + Tj;
Chris@82: 		    }
Chris@82: 		    Tl = Th + Tk;
Chris@82: 		    TP = Tr + Ts;
Chris@82: 		    Tb = T5 - Ta;
Chris@82: 		    Tm = Te + Tl;
Chris@82: 		    TW = Ta + T5;
Chris@82: 		    TX = TO + TP;
Chris@82: 		    T14 = TW + TX;
Chris@82: 		    {
Chris@82: 			 E TS, TT, Tx, Ty;
Chris@82: 			 TS = Tv + Tw;
Chris@82: 			 TT = Th - Tk;
Chris@82: 			 TU = TS - TT;
Chris@82: 			 T10 = TS + TT;
Chris@82: 			 Tx = KP866025403 * (Tv - Tw);
Chris@82: 			 Ty = FNMS(KP500000000, Tl, Te);
Chris@82: 			 Tz = Tx + Ty;
Chris@82: 			 TB = Ty - Tx;
Chris@82: 		    }
Chris@82: 		    {
Chris@82: 			 E Tq, Tt, TN, TQ;
Chris@82: 			 Tq = To - Tp;
Chris@82: 			 Tt = KP866025403 * (Tr - Ts);
Chris@82: 			 Tu = Tq - Tt;
Chris@82: 			 TC = Tq + Tt;
Chris@82: 			 TN = To + Tp;
Chris@82: 			 TQ = FNMS(KP500000000, TP, TO);
Chris@82: 			 TR = TN - TQ;
Chris@82: 			 T11 = TN + TQ;
Chris@82: 		    }
Chris@82: 	       }
Chris@82: 	       Cr[0] = T13 + T14;
Chris@82: 	       {
Chris@82: 		    E Tn, TG, TE, TF, TJ, TM, TK, TL;
Chris@82: 		    Tn = FNMS(KP174138601, Tm, KP575140729 * Tb);
Chris@82: 		    TG = FMA(KP174138601, Tb, KP575140729 * Tm);
Chris@82: 		    {
Chris@82: 			 E TA, TD, TH, TI;
Chris@82: 			 TA = FNMS(KP156891391, Tz, KP256247671 * Tu);
Chris@82: 			 TD = FNMS(KP300238635, TC, KP011599105 * TB);
Chris@82: 			 TE = TA + TD;
Chris@82: 			 TF = KP1_732050807 * (TD - TA);
Chris@82: 			 TH = FMA(KP300238635, TB, KP011599105 * TC);
Chris@82: 			 TI = FMA(KP256247671, Tz, KP156891391 * Tu);
Chris@82: 			 TJ = TH - TI;
Chris@82: 			 TM = KP1_732050807 * (TI + TH);
Chris@82: 		    }
Chris@82: 		    Ci[WS(csi, 5)] = FMA(KP2_000000000, TE, Tn);
Chris@82: 		    Ci[WS(csi, 1)] = FMA(KP2_000000000, TJ, TG);
Chris@82: 		    TK = TG - TJ;
Chris@82: 		    Ci[WS(csi, 4)] = TF - TK;
Chris@82: 		    Ci[WS(csi, 3)] = TF + TK;
Chris@82: 		    TL = Tn - TE;
Chris@82: 		    Ci[WS(csi, 2)] = TL - TM;
Chris@82: 		    Ci[WS(csi, 6)] = TL + TM;
Chris@82: 	       }
Chris@82: 	       {
Chris@82: 		    E TZ, T1b, T19, T1e, T16, T1a, TV, TY, T1c, T1d;
Chris@82: 		    TV = FNMS(KP132983124, TU, KP258260390 * TR);
Chris@82: 		    TY = KP300462606 * (TW - TX);
Chris@82: 		    TZ = FMA(KP2_000000000, TV, TY);
Chris@82: 		    T1b = TY - TV;
Chris@82: 		    {
Chris@82: 			 E T17, T18, T12, T15;
Chris@82: 			 T17 = FMA(KP387390585, TU, KP265966249 * TR);
Chris@82: 			 T18 = FNMS(KP503537032, T11, KP113854479 * T10);
Chris@82: 			 T19 = T17 - T18;
Chris@82: 			 T1e = T17 + T18;
Chris@82: 			 T12 = FMA(KP251768516, T10, KP075902986 * T11);
Chris@82: 			 T15 = FNMS(KP083333333, T14, T13);
Chris@82: 			 T16 = FMA(KP2_000000000, T12, T15);
Chris@82: 			 T1a = T15 - T12;
Chris@82: 		    }
Chris@82: 		    Cr[WS(csr, 1)] = TZ + T16;
Chris@82: 		    Cr[WS(csr, 5)] = T16 - TZ;
Chris@82: 		    T1c = T1a - T1b;
Chris@82: 		    Cr[WS(csr, 2)] = T19 + T1c;
Chris@82: 		    Cr[WS(csr, 6)] = T1c - T19;
Chris@82: 		    T1d = T1b + T1a;
Chris@82: 		    Cr[WS(csr, 3)] = T1d - T1e;
Chris@82: 		    Cr[WS(csr, 4)] = T1e + T1d;
Chris@82: 	       }
Chris@82: 	  }
Chris@82:      }
Chris@82: }
Chris@82: 
Chris@82: static const kr2c_desc desc = { 13, "r2cf_13", {57, 15, 19, 0}, &GENUS };
Chris@82: 
Chris@82: void X(codelet_r2cf_13) (planner *p) {
Chris@82:      X(kr2c_register) (p, r2cf_13, &desc);
Chris@82: }
Chris@82: 
Chris@82: #endif