cannam@167: /* cannam@167: * Copyright (c) 2003, 2007-14 Matteo Frigo cannam@167: * Copyright (c) 2003, 2007-14 Massachusetts Institute of Technology cannam@167: * cannam@167: * This program is free software; you can redistribute it and/or modify cannam@167: * it under the terms of the GNU General Public License as published by cannam@167: * the Free Software Foundation; either version 2 of the License, or cannam@167: * (at your option) any later version. cannam@167: * cannam@167: * This program is distributed in the hope that it will be useful, cannam@167: * but WITHOUT ANY WARRANTY; without even the implied warranty of cannam@167: * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the cannam@167: * GNU General Public License for more details. cannam@167: * cannam@167: * You should have received a copy of the GNU General Public License cannam@167: * along with this program; if not, write to the Free Software cannam@167: * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA cannam@167: * cannam@167: */ cannam@167: cannam@167: /* This file was automatically generated --- DO NOT EDIT */ cannam@167: /* Generated on Thu May 24 08:06:42 EDT 2018 */ cannam@167: cannam@167: #include "rdft/codelet-rdft.h" cannam@167: cannam@167: #if defined(ARCH_PREFERS_FMA) || defined(ISA_EXTENSION_PREFERS_FMA) cannam@167: cannam@167: /* Generated by: ../../../genfft/gen_r2cf.native -fma -compact -variables 4 -pipeline-latency 4 -n 5 -name r2cfII_5 -dft-II -include rdft/scalar/r2cfII.h */ cannam@167: cannam@167: /* cannam@167: * This function contains 12 FP additions, 7 FP multiplications, cannam@167: * (or, 7 additions, 2 multiplications, 5 fused multiply/add), cannam@167: * 17 stack variables, 4 constants, and 10 memory accesses cannam@167: */ cannam@167: #include "rdft/scalar/r2cfII.h" cannam@167: cannam@167: static void r2cfII_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs) cannam@167: { cannam@167: DK(KP559016994, +0.559016994374947424102293417182819058860154590); cannam@167: DK(KP250000000, +0.250000000000000000000000000000000000000000000); cannam@167: DK(KP618033988, +0.618033988749894848204586834365638117720309180); cannam@167: DK(KP951056516, +0.951056516295153572116439333379382143405698634); cannam@167: { cannam@167: INT i; cannam@167: for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) { cannam@167: E T1, T4, T7, T8, Tc, Tb, T9, Ta; cannam@167: T1 = R0[0]; cannam@167: { cannam@167: E T2, T3, T5, T6; cannam@167: T2 = R0[WS(rs, 1)]; cannam@167: T3 = R1[WS(rs, 1)]; cannam@167: T4 = T2 - T3; cannam@167: T5 = R0[WS(rs, 2)]; cannam@167: T6 = R1[0]; cannam@167: T7 = T5 - T6; cannam@167: T8 = T4 + T7; cannam@167: Tc = T5 + T6; cannam@167: Tb = T2 + T3; cannam@167: } cannam@167: Cr[WS(csr, 2)] = T1 + T8; cannam@167: Ci[WS(csi, 1)] = -(KP951056516 * (FNMS(KP618033988, Tb, Tc))); cannam@167: Ci[0] = -(KP951056516 * (FMA(KP618033988, Tc, Tb))); cannam@167: T9 = FNMS(KP250000000, T8, T1); cannam@167: Ta = T4 - T7; cannam@167: Cr[0] = FMA(KP559016994, Ta, T9); cannam@167: Cr[WS(csr, 1)] = FNMS(KP559016994, Ta, T9); cannam@167: } cannam@167: } cannam@167: } cannam@167: cannam@167: static const kr2c_desc desc = { 5, "r2cfII_5", {7, 2, 5, 0}, &GENUS }; cannam@167: cannam@167: void X(codelet_r2cfII_5) (planner *p) { cannam@167: X(kr2c_register) (p, r2cfII_5, &desc); cannam@167: } cannam@167: cannam@167: #else cannam@167: cannam@167: /* Generated by: ../../../genfft/gen_r2cf.native -compact -variables 4 -pipeline-latency 4 -n 5 -name r2cfII_5 -dft-II -include rdft/scalar/r2cfII.h */ cannam@167: cannam@167: /* cannam@167: * This function contains 12 FP additions, 6 FP multiplications, cannam@167: * (or, 9 additions, 3 multiplications, 3 fused multiply/add), cannam@167: * 17 stack variables, 4 constants, and 10 memory accesses cannam@167: */ cannam@167: #include "rdft/scalar/r2cfII.h" cannam@167: cannam@167: static void r2cfII_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs) cannam@167: { cannam@167: DK(KP250000000, +0.250000000000000000000000000000000000000000000); cannam@167: DK(KP559016994, +0.559016994374947424102293417182819058860154590); cannam@167: DK(KP951056516, +0.951056516295153572116439333379382143405698634); cannam@167: DK(KP587785252, +0.587785252292473129168705954639072768597652438); cannam@167: { cannam@167: INT i; cannam@167: for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) { cannam@167: E T8, T3, T6, T9, Tc, Tb, T7, Ta; cannam@167: T8 = R0[0]; cannam@167: { cannam@167: E T1, T2, T4, T5; cannam@167: T1 = R0[WS(rs, 1)]; cannam@167: T2 = R1[WS(rs, 1)]; cannam@167: T3 = T1 - T2; cannam@167: T4 = R0[WS(rs, 2)]; cannam@167: T5 = R1[0]; cannam@167: T6 = T4 - T5; cannam@167: T9 = T3 + T6; cannam@167: Tc = T4 + T5; cannam@167: Tb = T1 + T2; cannam@167: } cannam@167: Cr[WS(csr, 2)] = T8 + T9; cannam@167: Ci[WS(csi, 1)] = FNMS(KP951056516, Tc, KP587785252 * Tb); cannam@167: Ci[0] = -(FMA(KP951056516, Tb, KP587785252 * Tc)); cannam@167: T7 = KP559016994 * (T3 - T6); cannam@167: Ta = FNMS(KP250000000, T9, T8); cannam@167: Cr[0] = T7 + Ta; cannam@167: Cr[WS(csr, 1)] = Ta - T7; cannam@167: } cannam@167: } cannam@167: } cannam@167: cannam@167: static const kr2c_desc desc = { 5, "r2cfII_5", {9, 3, 3, 0}, &GENUS }; cannam@167: cannam@167: void X(codelet_r2cfII_5) (planner *p) { cannam@167: X(kr2c_register) (p, r2cfII_5, &desc); cannam@167: } cannam@167: cannam@167: #endif