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 08:07:43 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_r2cb.native -fma -compact -variables 4 -pipeline-latency 4 -sign 1 -n 8 -name r2cbIII_8 -dft-III -include rdft/scalar/r2cbIII.h */ cannam@167: cannam@167: /* cannam@167: * This function contains 22 FP additions, 12 FP multiplications, cannam@167: * (or, 18 additions, 8 multiplications, 4 fused multiply/add), cannam@167: * 19 stack variables, 4 constants, and 16 memory accesses cannam@167: */ cannam@167: #include "rdft/scalar/r2cbIII.h" cannam@167: cannam@167: static void r2cbIII_8(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(KP1_414213562, +1.414213562373095048801688724209698078569671875); cannam@167: DK(KP414213562, +0.414213562373095048801688724209698078569671875); cannam@167: DK(KP1_847759065, +1.847759065022573512256366378793576573644833252); cannam@167: DK(KP2_000000000, +2.000000000000000000000000000000000000000000000); cannam@167: { cannam@167: INT i; cannam@167: for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(32, rs), MAKE_VOLATILE_STRIDE(32, csr), MAKE_VOLATILE_STRIDE(32, csi)) { cannam@167: E T3, T7, Tf, Tl, T6, Tc, Ta, Tk, Tb, Tg; cannam@167: { cannam@167: E T1, T2, Td, Te; cannam@167: T1 = Cr[0]; cannam@167: T2 = Cr[WS(csr, 3)]; cannam@167: T3 = T1 + T2; cannam@167: T7 = T1 - T2; cannam@167: Td = Ci[0]; cannam@167: Te = Ci[WS(csi, 3)]; cannam@167: Tf = Td + Te; cannam@167: Tl = Te - Td; cannam@167: } cannam@167: { cannam@167: E T4, T5, T8, T9; cannam@167: T4 = Cr[WS(csr, 2)]; cannam@167: T5 = Cr[WS(csr, 1)]; cannam@167: T6 = T4 + T5; cannam@167: Tc = T4 - T5; cannam@167: T8 = Ci[WS(csi, 2)]; cannam@167: T9 = Ci[WS(csi, 1)]; cannam@167: Ta = T8 + T9; cannam@167: Tk = T8 - T9; cannam@167: } cannam@167: R0[0] = KP2_000000000 * (T3 + T6); cannam@167: R0[WS(rs, 2)] = KP2_000000000 * (Tl - Tk); cannam@167: Tb = T7 - Ta; cannam@167: Tg = Tc + Tf; cannam@167: R1[0] = KP1_847759065 * (FNMS(KP414213562, Tg, Tb)); cannam@167: R1[WS(rs, 2)] = -(KP1_847759065 * (FMA(KP414213562, Tb, Tg))); cannam@167: { cannam@167: E Th, Ti, Tj, Tm; cannam@167: Th = Tc - Tf; cannam@167: Ti = T7 + Ta; cannam@167: R1[WS(rs, 1)] = KP1_847759065 * (FMA(KP414213562, Ti, Th)); cannam@167: R1[WS(rs, 3)] = -(KP1_847759065 * (FNMS(KP414213562, Th, Ti))); cannam@167: Tj = T3 - T6; cannam@167: Tm = Tk + Tl; cannam@167: R0[WS(rs, 1)] = KP1_414213562 * (Tj + Tm); cannam@167: R0[WS(rs, 3)] = KP1_414213562 * (Tm - Tj); cannam@167: } cannam@167: } cannam@167: } cannam@167: } cannam@167: cannam@167: static const kr2c_desc desc = { 8, "r2cbIII_8", {18, 8, 4, 0}, &GENUS }; cannam@167: cannam@167: void X(codelet_r2cbIII_8) (planner *p) { cannam@167: X(kr2c_register) (p, r2cbIII_8, &desc); cannam@167: } cannam@167: cannam@167: #else cannam@167: cannam@167: /* Generated by: ../../../genfft/gen_r2cb.native -compact -variables 4 -pipeline-latency 4 -sign 1 -n 8 -name r2cbIII_8 -dft-III -include rdft/scalar/r2cbIII.h */ cannam@167: cannam@167: /* cannam@167: * This function contains 22 FP additions, 12 FP multiplications, cannam@167: * (or, 18 additions, 8 multiplications, 4 fused multiply/add), cannam@167: * 19 stack variables, 4 constants, and 16 memory accesses cannam@167: */ cannam@167: #include "rdft/scalar/r2cbIII.h" cannam@167: cannam@167: static void r2cbIII_8(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(KP1_414213562, +1.414213562373095048801688724209698078569671875); cannam@167: DK(KP765366864, +0.765366864730179543456919968060797733522689125); cannam@167: DK(KP1_847759065, +1.847759065022573512256366378793576573644833252); cannam@167: DK(KP2_000000000, +2.000000000000000000000000000000000000000000000); cannam@167: { cannam@167: INT i; cannam@167: for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(32, rs), MAKE_VOLATILE_STRIDE(32, csr), MAKE_VOLATILE_STRIDE(32, csi)) { cannam@167: E T3, T7, Tf, Tl, T6, Tc, Ta, Tk, Tb, Tg; cannam@167: { cannam@167: E T1, T2, Td, Te; cannam@167: T1 = Cr[0]; cannam@167: T2 = Cr[WS(csr, 3)]; cannam@167: T3 = T1 + T2; cannam@167: T7 = T1 - T2; cannam@167: Td = Ci[0]; cannam@167: Te = Ci[WS(csi, 3)]; cannam@167: Tf = Td + Te; cannam@167: Tl = Te - Td; cannam@167: } cannam@167: { cannam@167: E T4, T5, T8, T9; cannam@167: T4 = Cr[WS(csr, 2)]; cannam@167: T5 = Cr[WS(csr, 1)]; cannam@167: T6 = T4 + T5; cannam@167: Tc = T4 - T5; cannam@167: T8 = Ci[WS(csi, 2)]; cannam@167: T9 = Ci[WS(csi, 1)]; cannam@167: Ta = T8 + T9; cannam@167: Tk = T8 - T9; cannam@167: } cannam@167: R0[0] = KP2_000000000 * (T3 + T6); cannam@167: R0[WS(rs, 2)] = KP2_000000000 * (Tl - Tk); cannam@167: Tb = T7 - Ta; cannam@167: Tg = Tc + Tf; cannam@167: R1[0] = FNMS(KP765366864, Tg, KP1_847759065 * Tb); cannam@167: R1[WS(rs, 2)] = -(FMA(KP765366864, Tb, KP1_847759065 * Tg)); cannam@167: { cannam@167: E Th, Ti, Tj, Tm; cannam@167: Th = T7 + Ta; cannam@167: Ti = Tc - Tf; cannam@167: R1[WS(rs, 1)] = FMA(KP765366864, Th, KP1_847759065 * Ti); cannam@167: R1[WS(rs, 3)] = FNMS(KP1_847759065, Th, KP765366864 * Ti); cannam@167: Tj = T3 - T6; cannam@167: Tm = Tk + Tl; cannam@167: R0[WS(rs, 1)] = KP1_414213562 * (Tj + Tm); cannam@167: R0[WS(rs, 3)] = KP1_414213562 * (Tm - Tj); cannam@167: } cannam@167: } cannam@167: } cannam@167: } cannam@167: cannam@167: static const kr2c_desc desc = { 8, "r2cbIII_8", {18, 8, 4, 0}, &GENUS }; cannam@167: cannam@167: void X(codelet_r2cbIII_8) (planner *p) { cannam@167: X(kr2c_register) (p, r2cbIII_8, &desc); cannam@167: } cannam@167: cannam@167: #endif