1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
|
// Copyright 2022 The Abseil Authors.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// https://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
#ifndef ABSL_BASE_INTERNAL_PREFETCH_H_
#define ABSL_BASE_INTERNAL_PREFETCH_H_
#include "absl/base/config.h"
// Compatibility wrappers around __builtin_prefetch, to prefetch data
// for read if supported by the toolchain.
// Move data into the cache before it is read, or "prefetch" it.
//
// The value of `addr` is the address of the memory to prefetch. If
// the target and compiler support it, data prefetch instructions are
// generated. If the prefetch is done some time before the memory is
// read, it may be in the cache by the time the read occurs.
//
// The function names specify the temporal locality heuristic applied,
// using the names of Intel prefetch instructions:
//
// T0 - high degree of temporal locality; data should be left in as
// many levels of the cache possible
// T1 - moderate degree of temporal locality
// T2 - low degree of temporal locality
// Nta - no temporal locality, data need not be left in the cache
// after the read
//
// Incorrect or gratuitous use of these functions can degrade
// performance, so use them only when representative benchmarks show
// an improvement.
//
// Example usage:
//
// absl::base_internal::PrefetchT0(addr);
//
// Currently, the different prefetch calls behave on some Intel
// architectures as follows:
//
// SNB..SKL SKX
// PrefetchT0() L1/L2/L3 L1/L2
// PrefetchT1() L2/L3 L2
// PrefetchT2() L2/L3 L2
// PrefetchNta() L1/--/L3 L1*
//
// * On SKX PrefetchNta() will bring the line into L1 but will evict
// from L3 cache. This might result in surprising behavior.
//
// SNB = Sandy Bridge, SKL = Skylake, SKX = Skylake Xeon.
//
namespace absl {
ABSL_NAMESPACE_BEGIN
namespace base_internal {
void PrefetchT0(const void* addr);
void PrefetchT1(const void* addr);
void PrefetchT2(const void* addr);
void PrefetchNta(const void* addr);
// Implementation details follow.
#if ABSL_HAVE_BUILTIN(__builtin_prefetch) || defined(__GNUC__)
// See __builtin_prefetch:
// https://gcc.gnu.org/onlinedocs/gcc/Other-Builtins.html.
//
// These functions speculatively load for read only. This is
// safe for all currently supported platforms. However, prefetch for
// store may have problems depending on the target platform.
//
inline void PrefetchT0(const void* addr) {
// Note: this uses prefetcht0 on Intel.
__builtin_prefetch(addr, 0, 3);
}
inline void PrefetchT1(const void* addr) {
// Note: this uses prefetcht1 on Intel.
__builtin_prefetch(addr, 0, 2);
}
inline void PrefetchT2(const void* addr) {
// Note: this uses prefetcht2 on Intel.
__builtin_prefetch(addr, 0, 1);
}
inline void PrefetchNta(const void* addr) {
// Note: this uses prefetchtnta on Intel.
__builtin_prefetch(addr, 0, 0);
}
#else
inline void PrefetchT0(const void*) {}
inline void PrefetchT1(const void*) {}
inline void PrefetchT2(const void*) {}
inline void PrefetchNta(const void*) {}
#endif
} // namespace base_internal
ABSL_NAMESPACE_END
} // namespace absl
#endif // ABSL_BASE_INTERNAL_PREFETCH_H_
|
