Mercurial > hg > sv-dependency-builds
comparison src/fftw-3.3.8/rdft/scalar/r2cb/r2cbIII_6.c @ 167:bd3cc4d1df30
Add FFTW 3.3.8 source, and a Linux build
author | Chris Cannam <cannam@all-day-breakfast.com> |
---|---|
date | Tue, 19 Nov 2019 14:52:55 +0000 |
parents | |
children |
comparison
equal
deleted
inserted
replaced
166:cbd6d7e562c7 | 167:bd3cc4d1df30 |
---|---|
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 Thu May 24 08:07:42 EDT 2018 */ | |
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_r2cb.native -fma -compact -variables 4 -pipeline-latency 4 -sign 1 -n 6 -name r2cbIII_6 -dft-III -include rdft/scalar/r2cbIII.h */ | |
29 | |
30 /* | |
31 * This function contains 12 FP additions, 8 FP multiplications, | |
32 * (or, 6 additions, 2 multiplications, 6 fused multiply/add), | |
33 * 15 stack variables, 2 constants, and 12 memory accesses | |
34 */ | |
35 #include "rdft/scalar/r2cbIII.h" | |
36 | |
37 static void r2cbIII_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs) | |
38 { | |
39 DK(KP1_732050807, +1.732050807568877293527446341505872366942805254); | |
40 DK(KP2_000000000, +2.000000000000000000000000000000000000000000000); | |
41 { | |
42 INT i; | |
43 for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) { | |
44 E T1, T8, T4, Ta, T7, Tc, T9, Tb; | |
45 T1 = Cr[WS(csr, 1)]; | |
46 T8 = Ci[WS(csi, 1)]; | |
47 { | |
48 E T2, T3, T5, T6; | |
49 T2 = Cr[WS(csr, 2)]; | |
50 T3 = Cr[0]; | |
51 T4 = T2 + T3; | |
52 Ta = T2 - T3; | |
53 T5 = Ci[WS(csi, 2)]; | |
54 T6 = Ci[0]; | |
55 T7 = T5 + T6; | |
56 Tc = T5 - T6; | |
57 } | |
58 R0[0] = KP2_000000000 * (T1 + T4); | |
59 R1[WS(rs, 1)] = KP2_000000000 * (T8 - T7); | |
60 T9 = FMA(KP2_000000000, T8, T7); | |
61 R1[0] = -(FMA(KP1_732050807, Ta, T9)); | |
62 R1[WS(rs, 2)] = FMS(KP1_732050807, Ta, T9); | |
63 Tb = FNMS(KP2_000000000, T1, T4); | |
64 R0[WS(rs, 1)] = FMA(KP1_732050807, Tc, Tb); | |
65 R0[WS(rs, 2)] = FMS(KP1_732050807, Tc, Tb); | |
66 } | |
67 } | |
68 } | |
69 | |
70 static const kr2c_desc desc = { 6, "r2cbIII_6", {6, 2, 6, 0}, &GENUS }; | |
71 | |
72 void X(codelet_r2cbIII_6) (planner *p) { | |
73 X(kr2c_register) (p, r2cbIII_6, &desc); | |
74 } | |
75 | |
76 #else | |
77 | |
78 /* Generated by: ../../../genfft/gen_r2cb.native -compact -variables 4 -pipeline-latency 4 -sign 1 -n 6 -name r2cbIII_6 -dft-III -include rdft/scalar/r2cbIII.h */ | |
79 | |
80 /* | |
81 * This function contains 12 FP additions, 6 FP multiplications, | |
82 * (or, 10 additions, 4 multiplications, 2 fused multiply/add), | |
83 * 15 stack variables, 2 constants, and 12 memory accesses | |
84 */ | |
85 #include "rdft/scalar/r2cbIII.h" | |
86 | |
87 static void r2cbIII_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi, INT v, INT ivs, INT ovs) | |
88 { | |
89 DK(KP2_000000000, +2.000000000000000000000000000000000000000000000); | |
90 DK(KP1_732050807, +1.732050807568877293527446341505872366942805254); | |
91 { | |
92 INT i; | |
93 for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) { | |
94 E T1, T6, T4, T5, T9, Tb, Ta, Tc; | |
95 T1 = Cr[WS(csr, 1)]; | |
96 T6 = Ci[WS(csi, 1)]; | |
97 { | |
98 E T2, T3, T7, T8; | |
99 T2 = Cr[WS(csr, 2)]; | |
100 T3 = Cr[0]; | |
101 T4 = T2 + T3; | |
102 T5 = KP1_732050807 * (T2 - T3); | |
103 T7 = Ci[WS(csi, 2)]; | |
104 T8 = Ci[0]; | |
105 T9 = T7 + T8; | |
106 Tb = KP1_732050807 * (T7 - T8); | |
107 } | |
108 R0[0] = KP2_000000000 * (T1 + T4); | |
109 R1[WS(rs, 1)] = KP2_000000000 * (T6 - T9); | |
110 Ta = FMA(KP2_000000000, T6, T9); | |
111 R1[0] = -(T5 + Ta); | |
112 R1[WS(rs, 2)] = T5 - Ta; | |
113 Tc = FMS(KP2_000000000, T1, T4); | |
114 R0[WS(rs, 1)] = Tb - Tc; | |
115 R0[WS(rs, 2)] = Tc + Tb; | |
116 } | |
117 } | |
118 } | |
119 | |
120 static const kr2c_desc desc = { 6, "r2cbIII_6", {10, 4, 2, 0}, &GENUS }; | |
121 | |
122 void X(codelet_r2cbIII_6) (planner *p) { | |
123 X(kr2c_register) (p, r2cbIII_6, &desc); | |
124 } | |
125 | |
126 #endif |