/src/fftw3/rdft/scalar/r2cf/r2cf_5.c
Line | Count | Source (jump to first uncovered line) |
1 | | /* |
2 | | * Copyright (c) 2003, 2007-14 Matteo Frigo |
3 | | * Copyright (c) 2003, 2007-14 Massachusetts Institute of Technology |
4 | | * |
5 | | * This program is free software; you can redistribute it and/or modify |
6 | | * it under the terms of the GNU General Public License as published by |
7 | | * the Free Software Foundation; either version 2 of the License, or |
8 | | * (at your option) any later version. |
9 | | * |
10 | | * This program is distributed in the hope that it will be useful, |
11 | | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
12 | | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
13 | | * GNU General Public License for more details. |
14 | | * |
15 | | * You should have received a copy of the GNU General Public License |
16 | | * along with this program; if not, write to the Free Software |
17 | | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
18 | | * |
19 | | */ |
20 | | |
21 | | /* This file was automatically generated --- DO NOT EDIT */ |
22 | | /* Generated on Sun Jun 22 06:43:29 UTC 2025 */ |
23 | | |
24 | | #include "rdft/codelet-rdft.h" |
25 | | |
26 | | #if defined(ARCH_PREFERS_FMA) || defined(ISA_EXTENSION_PREFERS_FMA) |
27 | | |
28 | | /* Generated by: ../../../genfft/gen_r2cf.native -fma -compact -variables 4 -pipeline-latency 4 -n 5 -name r2cf_5 -include rdft/scalar/r2cf.h */ |
29 | | |
30 | | /* |
31 | | * This function contains 12 FP additions, 7 FP multiplications, |
32 | | * (or, 7 additions, 2 multiplications, 5 fused multiply/add), |
33 | | * 17 stack variables, 4 constants, and 10 memory accesses |
34 | | */ |
35 | | #include "rdft/scalar/r2cf.h" |
36 | | |
37 | | static void r2cf_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs) |
38 | | { |
39 | | DK(KP559016994, +0.559016994374947424102293417182819058860154590); |
40 | | DK(KP250000000, +0.250000000000000000000000000000000000000000000); |
41 | | DK(KP618033988, +0.618033988749894848204586834365638117720309180); |
42 | | DK(KP951056516, +0.951056516295153572116439333379382143405698634); |
43 | | { |
44 | | INT i; |
45 | | 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)) { |
46 | | E T7, T8, T9, T3, Ta, T6, Tb, Tc; |
47 | | T7 = R0[0]; |
48 | | { |
49 | | E T1, T2, T4, T5; |
50 | | T1 = R0[WS(rs, 2)]; |
51 | | T2 = R1[0]; |
52 | | T8 = T2 + T1; |
53 | | T4 = R0[WS(rs, 1)]; |
54 | | T5 = R1[WS(rs, 1)]; |
55 | | T9 = T4 + T5; |
56 | | T3 = T1 - T2; |
57 | | Ta = T8 + T9; |
58 | | T6 = T4 - T5; |
59 | | } |
60 | | Ci[WS(csi, 1)] = KP951056516 * (FNMS(KP618033988, T6, T3)); |
61 | | Ci[WS(csi, 2)] = KP951056516 * (FMA(KP618033988, T3, T6)); |
62 | | Cr[0] = T7 + Ta; |
63 | | Tb = FNMS(KP250000000, Ta, T7); |
64 | | Tc = T8 - T9; |
65 | | Cr[WS(csr, 1)] = FMA(KP559016994, Tc, Tb); |
66 | | Cr[WS(csr, 2)] = FNMS(KP559016994, Tc, Tb); |
67 | | } |
68 | | } |
69 | | } |
70 | | |
71 | | static const kr2c_desc desc = { 5, "r2cf_5", { 7, 2, 5, 0 }, &GENUS }; |
72 | | |
73 | | void X(codelet_r2cf_5) (planner *p) { X(kr2c_register) (p, r2cf_5, &desc); |
74 | | } |
75 | | |
76 | | #else |
77 | | |
78 | | /* Generated by: ../../../genfft/gen_r2cf.native -compact -variables 4 -pipeline-latency 4 -n 5 -name r2cf_5 -include rdft/scalar/r2cf.h */ |
79 | | |
80 | | /* |
81 | | * This function contains 12 FP additions, 6 FP multiplications, |
82 | | * (or, 9 additions, 3 multiplications, 3 fused multiply/add), |
83 | | * 17 stack variables, 4 constants, and 10 memory accesses |
84 | | */ |
85 | | #include "rdft/scalar/r2cf.h" |
86 | | |
87 | | static void r2cf_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs) |
88 | 0 | { |
89 | 0 | DK(KP250000000, +0.250000000000000000000000000000000000000000000); |
90 | 0 | DK(KP559016994, +0.559016994374947424102293417182819058860154590); |
91 | 0 | DK(KP587785252, +0.587785252292473129168705954639072768597652438); |
92 | 0 | DK(KP951056516, +0.951056516295153572116439333379382143405698634); |
93 | 0 | { |
94 | 0 | INT i; |
95 | 0 | 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)) { |
96 | 0 | E Ta, T7, T8, T3, Tb, T6, T9, Tc; |
97 | 0 | Ta = R0[0]; |
98 | 0 | { |
99 | 0 | E T1, T2, T4, T5; |
100 | 0 | T1 = R0[WS(rs, 2)]; |
101 | 0 | T2 = R1[0]; |
102 | 0 | T7 = T2 + T1; |
103 | 0 | T4 = R0[WS(rs, 1)]; |
104 | 0 | T5 = R1[WS(rs, 1)]; |
105 | 0 | T8 = T4 + T5; |
106 | 0 | T3 = T1 - T2; |
107 | 0 | Tb = T7 + T8; |
108 | 0 | T6 = T4 - T5; |
109 | 0 | } |
110 | 0 | Ci[WS(csi, 1)] = FNMS(KP587785252, T6, KP951056516 * T3); |
111 | 0 | Ci[WS(csi, 2)] = FMA(KP587785252, T3, KP951056516 * T6); |
112 | 0 | Cr[0] = Ta + Tb; |
113 | 0 | T9 = KP559016994 * (T7 - T8); |
114 | 0 | Tc = FNMS(KP250000000, Tb, Ta); |
115 | 0 | Cr[WS(csr, 1)] = T9 + Tc; |
116 | 0 | Cr[WS(csr, 2)] = Tc - T9; |
117 | 0 | } |
118 | 0 | } |
119 | 0 | } |
120 | | |
121 | | static const kr2c_desc desc = { 5, "r2cf_5", { 9, 3, 3, 0 }, &GENUS }; |
122 | | |
123 | 1 | void X(codelet_r2cf_5) (planner *p) { X(kr2c_register) (p, r2cf_5, &desc); |
124 | 1 | } |
125 | | |
126 | | #endif |