mirror of
https://github.com/facebook/rocksdb.git
synced 2024-11-30 22:41:48 +00:00
f4e4039f00
Summary: BottomNBits() - there is a single fast instruction for this on x86 since BMI2, but testing with godbolt indicates you need at least GCC 10 for the compiler to choose that instruction from the obvious C++ code. https://godbolt.org/z/5a7Ysd41h BitwiseAnd() - this is a convenience function that works around the language flaw that the type of the result of x & y is the larger of the two input types, when it should be the smaller. This can save some ugly static_cast. I expect to use both of these in coming HyperClockCache developments, and have applied them in a couple of places in existing code. Pull Request resolved: https://github.com/facebook/rocksdb/pull/11660 Test Plan: unit tests added Reviewed By: jowlyzhang Differential Revision: D47935531 Pulled By: pdillinger fbshipit-source-id: d148c43a1e51df4a1c549b93aaf2725a3f8d3bd6
86 lines
2.5 KiB
C++
86 lines
2.5 KiB
C++
// Copyright (c) 2017-present, Facebook, Inc. All rights reserved.
|
|
// This source code is licensed under both the GPLv2 (found in the
|
|
// COPYING file in the root directory) and Apache 2.0 License
|
|
// (found in the LICENSE.Apache file in the root directory).
|
|
|
|
#pragma once
|
|
|
|
#include <cassert>
|
|
#include <cstddef>
|
|
#include <thread>
|
|
#include <utility>
|
|
#include <vector>
|
|
|
|
#include "port/likely.h"
|
|
#include "port/port.h"
|
|
#include "util/math.h"
|
|
#include "util/random.h"
|
|
|
|
namespace ROCKSDB_NAMESPACE {
|
|
|
|
// An array of core-local values. Ideally the value type, T, is cache aligned to
|
|
// prevent false sharing.
|
|
template <typename T>
|
|
class CoreLocalArray {
|
|
public:
|
|
CoreLocalArray();
|
|
|
|
size_t Size() const;
|
|
// returns pointer to the element corresponding to the core that the thread
|
|
// currently runs on.
|
|
T* Access() const;
|
|
// same as above, but also returns the core index, which the client can cache
|
|
// to reduce how often core ID needs to be retrieved. Only do this if some
|
|
// inaccuracy is tolerable, as the thread may migrate to a different core.
|
|
std::pair<T*, size_t> AccessElementAndIndex() const;
|
|
// returns pointer to element for the specified core index. This can be used,
|
|
// e.g., for aggregation, or if the client caches core index.
|
|
T* AccessAtCore(size_t core_idx) const;
|
|
|
|
private:
|
|
std::unique_ptr<T[]> data_;
|
|
int size_shift_;
|
|
};
|
|
|
|
template <typename T>
|
|
CoreLocalArray<T>::CoreLocalArray() {
|
|
int num_cpus = static_cast<int>(std::thread::hardware_concurrency());
|
|
// find a power of two >= num_cpus and >= 8
|
|
size_shift_ = 3;
|
|
while (1 << size_shift_ < num_cpus) {
|
|
++size_shift_;
|
|
}
|
|
data_.reset(new T[static_cast<size_t>(1) << size_shift_]);
|
|
}
|
|
|
|
template <typename T>
|
|
size_t CoreLocalArray<T>::Size() const {
|
|
return static_cast<size_t>(1) << size_shift_;
|
|
}
|
|
|
|
template <typename T>
|
|
T* CoreLocalArray<T>::Access() const {
|
|
return AccessElementAndIndex().first;
|
|
}
|
|
|
|
template <typename T>
|
|
std::pair<T*, size_t> CoreLocalArray<T>::AccessElementAndIndex() const {
|
|
int cpuid = port::PhysicalCoreID();
|
|
size_t core_idx;
|
|
if (UNLIKELY(cpuid < 0)) {
|
|
// cpu id unavailable, just pick randomly
|
|
core_idx = Random::GetTLSInstance()->Uniform(1 << size_shift_);
|
|
} else {
|
|
core_idx = static_cast<size_t>(BottomNBits(cpuid, size_shift_));
|
|
}
|
|
return {AccessAtCore(core_idx), core_idx};
|
|
}
|
|
|
|
template <typename T>
|
|
T* CoreLocalArray<T>::AccessAtCore(size_t core_idx) const {
|
|
assert(core_idx < static_cast<size_t>(1) << size_shift_);
|
|
return &data_[core_idx];
|
|
}
|
|
|
|
} // namespace ROCKSDB_NAMESPACE
|