/src/fftw3/rdft/scalar/r2cf/r2cf_6.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 Sep 8 06:41:09 UTC 2024 */ |
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 6 -name r2cf_6 -include rdft/scalar/r2cf.h */ |
29 | | |
30 | | /* |
31 | | * This function contains 14 FP additions, 4 FP multiplications, |
32 | | * (or, 12 additions, 2 multiplications, 2 fused multiply/add), |
33 | | * 17 stack variables, 2 constants, and 12 memory accesses |
34 | | */ |
35 | | #include "rdft/scalar/r2cf.h" |
36 | | |
37 | | static void r2cf_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs) |
38 | | { |
39 | | DK(KP500000000, +0.500000000000000000000000000000000000000000000); |
40 | | DK(KP866025403, +0.866025403784438646763723170752936183471402627); |
41 | | { |
42 | | INT i; |
43 | | for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) { |
44 | | E T3, Td, T9, Tc, T6, Tb, T1, T2, Ta, Te; |
45 | | T1 = R0[0]; |
46 | | T2 = R1[WS(rs, 1)]; |
47 | | T3 = T1 - T2; |
48 | | Td = T1 + T2; |
49 | | { |
50 | | E T7, T8, T4, T5; |
51 | | T7 = R0[WS(rs, 2)]; |
52 | | T8 = R1[0]; |
53 | | T9 = T7 - T8; |
54 | | Tc = T7 + T8; |
55 | | T4 = R0[WS(rs, 1)]; |
56 | | T5 = R1[WS(rs, 2)]; |
57 | | T6 = T4 - T5; |
58 | | Tb = T4 + T5; |
59 | | } |
60 | | Ci[WS(csi, 1)] = KP866025403 * (T9 - T6); |
61 | | Ta = T6 + T9; |
62 | | Cr[WS(csr, 1)] = FNMS(KP500000000, Ta, T3); |
63 | | Cr[WS(csr, 3)] = T3 + Ta; |
64 | | Ci[WS(csi, 2)] = KP866025403 * (Tb - Tc); |
65 | | Te = Tb + Tc; |
66 | | Cr[WS(csr, 2)] = FNMS(KP500000000, Te, Td); |
67 | | Cr[0] = Td + Te; |
68 | | } |
69 | | } |
70 | | } |
71 | | |
72 | | static const kr2c_desc desc = { 6, "r2cf_6", { 12, 2, 2, 0 }, &GENUS }; |
73 | | |
74 | | void X(codelet_r2cf_6) (planner *p) { X(kr2c_register) (p, r2cf_6, &desc); |
75 | | } |
76 | | |
77 | | #else |
78 | | |
79 | | /* Generated by: ../../../genfft/gen_r2cf.native -compact -variables 4 -pipeline-latency 4 -n 6 -name r2cf_6 -include rdft/scalar/r2cf.h */ |
80 | | |
81 | | /* |
82 | | * This function contains 14 FP additions, 4 FP multiplications, |
83 | | * (or, 12 additions, 2 multiplications, 2 fused multiply/add), |
84 | | * 17 stack variables, 2 constants, and 12 memory accesses |
85 | | */ |
86 | | #include "rdft/scalar/r2cf.h" |
87 | | |
88 | | static void r2cf_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs) |
89 | 0 | { |
90 | 0 | DK(KP500000000, +0.500000000000000000000000000000000000000000000); |
91 | 0 | DK(KP866025403, +0.866025403784438646763723170752936183471402627); |
92 | 0 | { |
93 | 0 | INT i; |
94 | 0 | for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) { |
95 | 0 | E T3, Td, T9, Tc, T6, Tb, T1, T2, Ta, Te; |
96 | 0 | T1 = R0[0]; |
97 | 0 | T2 = R1[WS(rs, 1)]; |
98 | 0 | T3 = T1 - T2; |
99 | 0 | Td = T1 + T2; |
100 | 0 | { |
101 | 0 | E T7, T8, T4, T5; |
102 | 0 | T7 = R0[WS(rs, 2)]; |
103 | 0 | T8 = R1[0]; |
104 | 0 | T9 = T7 - T8; |
105 | 0 | Tc = T7 + T8; |
106 | 0 | T4 = R0[WS(rs, 1)]; |
107 | 0 | T5 = R1[WS(rs, 2)]; |
108 | 0 | T6 = T4 - T5; |
109 | 0 | Tb = T4 + T5; |
110 | 0 | } |
111 | 0 | Ci[WS(csi, 1)] = KP866025403 * (T9 - T6); |
112 | 0 | Ta = T6 + T9; |
113 | 0 | Cr[WS(csr, 1)] = FNMS(KP500000000, Ta, T3); |
114 | 0 | Cr[WS(csr, 3)] = T3 + Ta; |
115 | 0 | Ci[WS(csi, 2)] = KP866025403 * (Tb - Tc); |
116 | 0 | Te = Tb + Tc; |
117 | 0 | Cr[WS(csr, 2)] = FNMS(KP500000000, Te, Td); |
118 | 0 | Cr[0] = Td + Te; |
119 | 0 | } |
120 | 0 | } |
121 | 0 | } |
122 | | |
123 | | static const kr2c_desc desc = { 6, "r2cf_6", { 12, 2, 2, 0 }, &GENUS }; |
124 | | |
125 | 1 | void X(codelet_r2cf_6) (planner *p) { X(kr2c_register) (p, r2cf_6, &desc); |
126 | 1 | } |
127 | | |
128 | | #endif |