fcb8d040e8
Instead of lengthy blurbs, switch to single-line, machine-readable standardized (https://spdx.dev) license identifiers. The Linux kernel switched long ago, so there is strong precedent. Three cases are handled: AGPL-only, Apache-only, and dual licensed. For the latter case, I chose (AGPL-3.0-or-later and Apache-2.0), reasoning that our changes are extensive enough to apply our license. The changes we applied mechanically with a script, except to licenses/README.md. Closes #9937
261 lines
9.1 KiB
C++
261 lines
9.1 KiB
C++
/*
|
|
* Copyright (C) 2017-present ScyllaDB
|
|
*
|
|
*/
|
|
|
|
/*
|
|
* SPDX-License-Identifier: AGPL-3.0-or-later
|
|
*/
|
|
|
|
#pragma once
|
|
|
|
#include <boost/intrusive/list.hpp>
|
|
#include <limits>
|
|
#include <seastar/core/bitops.hh>
|
|
#include "seastarx.hh"
|
|
|
|
namespace bi = boost::intrusive;
|
|
|
|
// Returns largest N such that 2^N <= v.
|
|
// Undefined for v == 0.
|
|
inline size_t pow2_rank(size_t v) {
|
|
return std::numeric_limits<size_t>::digits - 1 - count_leading_zeros(v);
|
|
}
|
|
|
|
|
|
// Returns largest N such that 2^N <= v.
|
|
// Undefined for v == 0.
|
|
inline constexpr size_t pow2_rank_constexpr(size_t v) {
|
|
return v <= 1 ? 1 : 1 + pow2_rank_constexpr(v >> 1);
|
|
}
|
|
|
|
// Configures log_heap.
|
|
// Only values <= max_size can be inserted into the histogram.
|
|
// log_heap::contains_above_min() returns true if and only if the histogram contains any value >= min_size.
|
|
struct log_heap_options {
|
|
const size_t min_size;
|
|
const size_t sub_bucket_shift;
|
|
const size_t max_size;
|
|
|
|
constexpr log_heap_options(const size_t min_size, size_t sub_bucket_shift, size_t max_size)
|
|
: min_size(min_size)
|
|
, sub_bucket_shift(sub_bucket_shift)
|
|
, max_size(max_size) {
|
|
}
|
|
|
|
size_t bucket_of(size_t value) const {
|
|
#ifdef SANITIZE
|
|
// ubsan will otherwise complain about pow2_rank(0)
|
|
if (value < min_size) {
|
|
return 0;
|
|
}
|
|
#endif
|
|
const auto min_mask = -size_t(value >= min_size); // 0 when below min_size, all bits on otherwise
|
|
value = value - min_size + 1;
|
|
const auto pow2_index = pow2_rank(value);
|
|
const auto bucket = (pow2_index + 1) & min_mask;
|
|
const auto unmasked_sub_bucket_index = (value << sub_bucket_shift) >> pow2_index;
|
|
const auto mask = ((1 << sub_bucket_shift) - 1) & min_mask;
|
|
const auto sub_bucket_index = unmasked_sub_bucket_index & mask;
|
|
return (bucket << sub_bucket_shift) - mask + sub_bucket_index;
|
|
}
|
|
|
|
constexpr size_t number_of_buckets() const {
|
|
const auto min_mask = -size_t(max_size >= min_size); // 0 when below min_size, all bits on otherwise
|
|
const auto value = max_size - min_size + 1;
|
|
const auto pow2_index = pow2_rank_constexpr(value);
|
|
const auto bucket = (pow2_index + 1) & min_mask;
|
|
const auto unmasked_sub_bucket_index = (value << sub_bucket_shift) >> pow2_index;
|
|
const auto mask = ((1 << sub_bucket_shift) - 1) & min_mask;
|
|
const auto sub_bucket_index = unmasked_sub_bucket_index & mask;
|
|
const auto ret = (bucket << sub_bucket_shift) - mask + sub_bucket_index;
|
|
return ret + 1;
|
|
}
|
|
};
|
|
|
|
template<const log_heap_options& opts>
|
|
struct log_heap_bucket_index {
|
|
using type = std::conditional_t<(opts.number_of_buckets() > ((1 << 16) - 1)), uint32_t,
|
|
std::conditional_t<(opts.number_of_buckets() > ((1 << 8) - 1)), uint16_t, uint8_t>>;
|
|
};
|
|
|
|
template<const log_heap_options& opts>
|
|
struct log_heap_hook : public bi::list_base_hook<> {
|
|
typename log_heap_bucket_index<opts>::type cached_bucket;
|
|
};
|
|
|
|
template<typename T>
|
|
size_t hist_key(const T&);
|
|
|
|
template<typename T, const log_heap_options& opts, bool = std::is_base_of<log_heap_hook<opts>, T>::value>
|
|
struct log_heap_element_traits {
|
|
using bucket_type = bi::list<T, bi::constant_time_size<false>>;
|
|
static void cache_bucket(T& v, typename log_heap_bucket_index<opts>::type b) {
|
|
v.cached_bucket = b;
|
|
}
|
|
static size_t cached_bucket(const T& v) {
|
|
return v.cached_bucket;
|
|
}
|
|
static size_t hist_key(const T& v) {
|
|
return ::hist_key<T>(v);
|
|
}
|
|
};
|
|
|
|
template<typename T, const log_heap_options& opts>
|
|
struct log_heap_element_traits<T, opts, false> {
|
|
using bucket_type = typename T::bucket_type;
|
|
static void cache_bucket(T&, typename log_heap_bucket_index<opts>::type);
|
|
static size_t cached_bucket(const T&);
|
|
static size_t hist_key(const T&);
|
|
};
|
|
|
|
/*
|
|
* Histogram that stores elements in different buckets according to their size.
|
|
* Values are mapped to a sequence of power-of-two ranges that are split in
|
|
* 1 << opts.sub_bucket_shift sub-buckets. Values less than opts.min_size
|
|
* are placed in bucket 0, whereas values bigger than opts.max_size are
|
|
* not admitted. The histogram gives bigger precision to smaller values, with
|
|
* precision decreasing as values get larger.
|
|
*/
|
|
template<typename T, const log_heap_options& opts>
|
|
requires requires() {
|
|
typename log_heap_element_traits<T, opts>;
|
|
}
|
|
class log_heap final {
|
|
// Ensure that (value << sub_bucket_index) in bucket_of() doesn't overflow
|
|
static_assert(pow2_rank_constexpr(opts.max_size - opts.min_size + 1) + opts.sub_bucket_shift < std::numeric_limits<size_t>::digits, "overflow");
|
|
private:
|
|
using traits = log_heap_element_traits<T, opts>;
|
|
using bucket = typename traits::bucket_type;
|
|
|
|
struct hist_size_less_compare {
|
|
inline bool operator()(const T& v1, const T& v2) const {
|
|
return traits::hist_key(v1) < traits::hist_key(v2);
|
|
}
|
|
};
|
|
|
|
std::array<bucket, opts.number_of_buckets()> _buckets;
|
|
ssize_t _watermark = -1;
|
|
public:
|
|
template <bool IsConst>
|
|
class hist_iterator {
|
|
public:
|
|
using iterator_category = std::input_iterator_tag;
|
|
using value_type = std::conditional_t<IsConst, const T, T>;
|
|
using difference_type = std::ptrdiff_t;
|
|
using pointer = std::conditional_t<IsConst, const T, T>*;
|
|
using reference = std::conditional_t<IsConst, const T, T>&;
|
|
private:
|
|
using hist_type = std::conditional_t<IsConst, const log_heap, log_heap>;
|
|
using iterator_type = std::conditional_t<IsConst, typename bucket::const_iterator, typename bucket::iterator>;
|
|
|
|
hist_type& _h;
|
|
ssize_t _b;
|
|
iterator_type _it;
|
|
public:
|
|
struct end_tag {};
|
|
hist_iterator(hist_type& h)
|
|
: _h(h)
|
|
, _b(h._watermark)
|
|
, _it(_b >= 0 ? h._buckets[_b].begin() : h._buckets[0].end()) {
|
|
}
|
|
hist_iterator(hist_type& h, end_tag)
|
|
: _h(h)
|
|
, _b(-1)
|
|
, _it(h._buckets[0].end()) {
|
|
}
|
|
std::conditional_t<IsConst, const T, T>& operator*() {
|
|
return *_it;
|
|
}
|
|
hist_iterator& operator++() {
|
|
if (++_it == _h._buckets[_b].end()) {
|
|
do {
|
|
--_b;
|
|
} while (_b >= 0 && (_it = _h._buckets[_b].begin()) == _h._buckets[_b].end());
|
|
}
|
|
return *this;
|
|
}
|
|
bool operator==(const hist_iterator& other) const {
|
|
return _b == other._b && _it == other._it;
|
|
}
|
|
bool operator!=(const hist_iterator& other) const {
|
|
return !(*this == other);
|
|
}
|
|
};
|
|
using iterator = hist_iterator<false>;
|
|
using const_iterator = hist_iterator<true>;
|
|
public:
|
|
bool empty() const {
|
|
return _watermark == -1;
|
|
}
|
|
// Returns true if and only if contains any value >= opts.min_size.
|
|
bool contains_above_min() const {
|
|
return _watermark > 0;
|
|
}
|
|
const_iterator begin() const {
|
|
return const_iterator(*this);
|
|
}
|
|
const_iterator end() const {
|
|
return const_iterator(*this, typename const_iterator::end_tag());
|
|
}
|
|
iterator begin() {
|
|
return iterator(*this);
|
|
}
|
|
iterator end() {
|
|
return iterator(*this, typename iterator::end_tag());
|
|
}
|
|
// Returns a range of buckets starting from that with the smaller values.
|
|
// Each bucket is a range of const T&.
|
|
const auto& buckets() const {
|
|
return _buckets;
|
|
}
|
|
// Pops one of the largest elements in the histogram.
|
|
void pop_one_of_largest() {
|
|
_buckets[_watermark].pop_front();
|
|
maybe_adjust_watermark();
|
|
}
|
|
// Returns one of the largest elements in the histogram.
|
|
const T& one_of_largest() const {
|
|
return _buckets[_watermark].front();
|
|
}
|
|
// Returns one of the largest elements in the histogram.
|
|
T& one_of_largest() {
|
|
return _buckets[_watermark].front();
|
|
}
|
|
// Pushes a new element onto the histogram.
|
|
void push(T& v) {
|
|
auto b = opts.bucket_of(traits::hist_key(v));
|
|
traits::cache_bucket(v, b);
|
|
_buckets[b].push_front(v);
|
|
_watermark = std::max(ssize_t(b), _watermark);
|
|
}
|
|
// Adjusts the histogram when the specified element becomes larger.
|
|
void adjust_up(T& v) {
|
|
auto b = traits::cached_bucket(v);
|
|
auto nb = opts.bucket_of(traits::hist_key(v));
|
|
if (nb != b) {
|
|
traits::cache_bucket(v, nb);
|
|
_buckets[nb].splice(_buckets[nb].begin(), _buckets[b], _buckets[b].iterator_to(v));
|
|
_watermark = std::max(ssize_t(nb), _watermark);
|
|
}
|
|
}
|
|
// Removes the specified element from the histogram.
|
|
void erase(T& v) {
|
|
auto& b = _buckets[traits::cached_bucket(v)];
|
|
b.erase(b.iterator_to(v));
|
|
maybe_adjust_watermark();
|
|
}
|
|
// Merges the specified histogram, moving all elements from it into this.
|
|
void merge(log_heap& other) {
|
|
for (size_t i = 0; i < opts.number_of_buckets(); ++i) {
|
|
_buckets[i].splice(_buckets[i].begin(), other._buckets[i]);
|
|
}
|
|
_watermark = std::max(_watermark, other._watermark);
|
|
other._watermark = -1;
|
|
}
|
|
private:
|
|
void maybe_adjust_watermark() {
|
|
while (_buckets[_watermark].empty() && --_watermark >= 0) ;
|
|
}
|
|
};
|