/* * Copyright (C) 2005-2019 Apple Inc. All rights reserved. * Copyright (C) 2008 David Levin * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Library General Public * License as published by the Free Software Foundation; either * version 2 of the License, or (at your option) any later version. * * This library is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * Library General Public License for more details. * * You should have received a copy of the GNU Library General Public License * along with this library; see the file COPYING.LIB. If not, write to * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, * Boston, MA 02110-1301, USA. * */ #pragma once #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #define DUMP_HASHTABLE_STATS 0 #define DUMP_HASHTABLE_STATS_PER_TABLE 0 #if DUMP_HASHTABLE_STATS_PER_TABLE #include #endif namespace WTF { DECLARE_ALLOCATOR_WITH_HEAP_IDENTIFIER(HashTable); // Enables internal WTF consistency checks that are invoked automatically. Non-WTF callers can call checkTableConsistency() even if internal checks are disabled. #define CHECK_HASHTABLE_CONSISTENCY 0 #ifdef NDEBUG #define CHECK_HASHTABLE_ITERATORS 0 #define CHECK_HASHTABLE_USE_AFTER_DESTRUCTION 0 #else #define CHECK_HASHTABLE_ITERATORS 1 #define CHECK_HASHTABLE_USE_AFTER_DESTRUCTION 1 #endif #if DUMP_HASHTABLE_STATS struct HashTableStats { // The following variables are all atomically incremented when modified. WTF_EXPORT_PRIVATE static std::atomic numAccesses; WTF_EXPORT_PRIVATE static std::atomic numRehashes; WTF_EXPORT_PRIVATE static std::atomic numRemoves; WTF_EXPORT_PRIVATE static std::atomic numReinserts; // The following variables are only modified in the recordCollisionAtCount method within a mutex. WTF_EXPORT_PRIVATE static unsigned maxCollisions; WTF_EXPORT_PRIVATE static unsigned numCollisions; WTF_EXPORT_PRIVATE static unsigned collisionGraph[4096]; WTF_EXPORT_PRIVATE static void recordCollisionAtCount(unsigned count); WTF_EXPORT_PRIVATE static void dumpStats(); }; #endif template class HashTableIterator; template class HashTableConstIterator; template void addIterator(const HashTableType*, HashTableConstIterator*); template void removeIterator(HashTableConstIterator*); template void invalidateIterators(const HashTableType*); #if !CHECK_HASHTABLE_ITERATORS template inline void addIterator(const HashTableType*, HashTableConstIterator*) { } template inline void removeIterator(HashTableConstIterator*) { } template void invalidateIterators(const HashTableType*) { } #endif typedef enum { HashItemKnownGood } HashItemKnownGoodTag; template class HashTableConstIterator : public std::iterator { WTF_MAKE_FAST_ALLOCATED; private: using HashTableType = HashTable; typedef HashTableIterator iterator; typedef HashTableConstIterator const_iterator; typedef Value ValueType; typedef const ValueType& ReferenceType; typedef const ValueType* PointerType; friend HashTableType; friend iterator; void skipEmptyBuckets() { while (m_position != m_endPosition && HashTableType::isEmptyOrDeletedBucket(*m_position)) ++m_position; } HashTableConstIterator(const HashTableType* table, PointerType position, PointerType endPosition) : m_position(position), m_endPosition(endPosition) { addIterator(table, this); skipEmptyBuckets(); } HashTableConstIterator(const HashTableType* table, PointerType position, PointerType endPosition, HashItemKnownGoodTag) : m_position(position), m_endPosition(endPosition) { addIterator(table, this); } public: HashTableConstIterator() { addIterator(static_cast(0), this); } // default copy, assignment and destructor are OK if CHECK_HASHTABLE_ITERATORS is 0 #if CHECK_HASHTABLE_ITERATORS ~HashTableConstIterator() { removeIterator(this); } HashTableConstIterator(const const_iterator& other) : m_position(other.m_position), m_endPosition(other.m_endPosition) { addIterator(other.m_table, this); } const_iterator& operator=(const const_iterator& other) { m_position = other.m_position; m_endPosition = other.m_endPosition; removeIterator(this); addIterator(other.m_table, this); return *this; } #endif PointerType get() const { checkValidity(); return m_position; } ReferenceType operator*() const { return *get(); } PointerType operator->() const { return get(); } const_iterator& operator++() { checkValidity(); ASSERT(m_position != m_endPosition); ++m_position; skipEmptyBuckets(); return *this; } // postfix ++ intentionally omitted // Comparison. bool operator==(const const_iterator& other) const { checkValidity(other); return m_position == other.m_position; } bool operator!=(const const_iterator& other) const { checkValidity(other); return m_position != other.m_position; } bool operator==(const iterator& other) const { return *this == static_cast(other); } bool operator!=(const iterator& other) const { return *this != static_cast(other); } private: void checkValidity() const { #if CHECK_HASHTABLE_ITERATORS ASSERT(m_table); #endif } #if CHECK_HASHTABLE_ITERATORS void checkValidity(const const_iterator& other) const { ASSERT(m_table); ASSERT_UNUSED(other, other.m_table); ASSERT(m_table == other.m_table); } #else void checkValidity(const const_iterator&) const { } #endif PointerType m_position { nullptr }; PointerType m_endPosition { nullptr }; #if CHECK_HASHTABLE_ITERATORS public: // Any modifications of the m_next or m_previous of an iterator that is in a linked list of a HashTable::m_iterator, // should be guarded with m_table->m_mutex. mutable const HashTableType* m_table; mutable const_iterator* m_next; mutable const_iterator* m_previous; #endif }; template class HashTableIterator : public std::iterator { WTF_MAKE_FAST_ALLOCATED; private: using HashTableType = HashTable; typedef HashTableIterator iterator; typedef HashTableConstIterator const_iterator; typedef Value ValueType; typedef ValueType& ReferenceType; typedef ValueType* PointerType; friend HashTableType; HashTableIterator(HashTableType* table, PointerType pos, PointerType end) : m_iterator(table, pos, end) { } HashTableIterator(HashTableType* table, PointerType pos, PointerType end, HashItemKnownGoodTag tag) : m_iterator(table, pos, end, tag) { } public: HashTableIterator() { } // default copy, assignment and destructor are OK PointerType get() const { return const_cast(m_iterator.get()); } ReferenceType operator*() const { return *get(); } PointerType operator->() const { return get(); } iterator& operator++() { ++m_iterator; return *this; } // postfix ++ intentionally omitted // Comparison. bool operator==(const iterator& other) const { return m_iterator == other.m_iterator; } bool operator!=(const iterator& other) const { return m_iterator != other.m_iterator; } bool operator==(const const_iterator& other) const { return m_iterator == other; } bool operator!=(const const_iterator& other) const { return m_iterator != other; } operator const_iterator() const { return m_iterator; } private: const_iterator m_iterator; }; template class IdentityHashTranslator { public: template static unsigned hash(const T& key) { return HashFunctions::hash(key); } template static bool equal(const T& a, const U& b) { return HashFunctions::equal(a, b); } template static void translate(T& location, const U&, V&& value) { ValueTraits::assignToEmpty(location, std::forward(value)); } }; template struct HashTableAddResult { HashTableAddResult() : isNewEntry(false) { } HashTableAddResult(IteratorType iter, bool isNewEntry) : iterator(iter), isNewEntry(isNewEntry) { } IteratorType iterator; bool isNewEntry; explicit operator bool() const { return isNewEntry; } }; // HashTableCapacityForSize computes the upper power of two capacity to hold the size parameter. // This is done at compile time to initialize the HashTraits. template struct HashTableCapacityForSize { // Load-factor for small table is 75%. static constexpr unsigned smallMaxLoadNumerator = 3; static constexpr unsigned smallMaxLoadDenominator = 4; // Load-factor for large table is 50%. static constexpr unsigned largeMaxLoadNumerator = 1; static constexpr unsigned largeMaxLoadDenominator = 2; static constexpr unsigned maxSmallTableCapacity = 1024; static constexpr unsigned minLoad = 6; static constexpr bool shouldExpand(uint64_t keyAndDeleteCount, uint64_t tableSize) { if (tableSize <= maxSmallTableCapacity) return keyAndDeleteCount * smallMaxLoadDenominator >= tableSize * smallMaxLoadNumerator; return keyAndDeleteCount * largeMaxLoadDenominator >= tableSize * largeMaxLoadNumerator; } static constexpr unsigned capacityForSize(uint32_t sizeArg) { if (!sizeArg) return 0; constexpr unsigned maxCapacity = 1U << 31; UNUSED_PARAM(maxCapacity); ASSERT_UNDER_CONSTEXPR_CONTEXT(sizeArg <= maxCapacity); uint32_t capacity = roundUpToPowerOfTwo(sizeArg); ASSERT_UNDER_CONSTEXPR_CONTEXT(capacity <= maxCapacity); if (shouldExpand(sizeArg, capacity)) { ASSERT_UNDER_CONSTEXPR_CONTEXT((static_cast(capacity) * 2) <= maxCapacity); return capacity * 2; } return capacity; } static constexpr unsigned value = capacityForSize(size); static_assert(size > 0, "HashTableNonZeroMinimumCapacity"); static_assert(!static_cast(value >> 31), "HashTableNoCapacityOverflow"); }; template class HashTable { public: using HashTableType = HashTable; typedef HashTableIterator iterator; typedef HashTableConstIterator const_iterator; typedef Traits ValueTraits; typedef Key KeyType; typedef Value ValueType; typedef IdentityHashTranslator IdentityTranslatorType; typedef HashTableAddResult AddResult; using HashTableSizePolicy = HashTableCapacityForSize<1>; #if DUMP_HASHTABLE_STATS_PER_TABLE struct Stats { WTF_MAKE_STRUCT_FAST_ALLOCATED; Stats() : numAccesses(0) , numRehashes(0) , numRemoves(0) , numReinserts(0) , maxCollisions(0) , numCollisions(0) , collisionGraph() { } unsigned numAccesses; unsigned numRehashes; unsigned numRemoves; unsigned numReinserts; unsigned maxCollisions; unsigned numCollisions; unsigned collisionGraph[4096]; void recordCollisionAtCount(unsigned count) { if (count > maxCollisions) maxCollisions = count; numCollisions++; collisionGraph[count]++; } void dumpStats() { dataLogF("\nWTF::HashTable::Stats dump\n\n"); dataLogF("%d accesses\n", numAccesses); dataLogF("%d total collisions, average %.2f probes per access\n", numCollisions, 1.0 * (numAccesses + numCollisions) / numAccesses); dataLogF("longest collision chain: %d\n", maxCollisions); for (unsigned i = 1; i <= maxCollisions; i++) { dataLogF(" %d lookups with exactly %d collisions (%.2f%% , %.2f%% with this many or more)\n", collisionGraph[i], i, 100.0 * (collisionGraph[i] - collisionGraph[i+1]) / numAccesses, 100.0 * collisionGraph[i] / numAccesses); } dataLogF("%d rehashes\n", numRehashes); dataLogF("%d reinserts\n", numReinserts); } }; #endif HashTable(); ~HashTable() { invalidateIterators(this); if (m_table) deallocateTable(m_table); #if CHECK_HASHTABLE_USE_AFTER_DESTRUCTION m_table = (ValueType*)(uintptr_t)0xbbadbeef; #endif } HashTable(const HashTable&); void swap(HashTable&); HashTable& operator=(const HashTable&); HashTable(HashTable&&); HashTable& operator=(HashTable&&); // When the hash table is empty, just return the same iterator for end as for begin. // This is more efficient because we don't have to skip all the empty and deleted // buckets, and iterating an empty table is a common case that's worth optimizing. iterator begin() { return isEmpty() ? end() : makeIterator(m_table); } iterator end() { return makeKnownGoodIterator(m_table + tableSize()); } const_iterator begin() const { return isEmpty() ? end() : makeConstIterator(m_table); } const_iterator end() const { return makeKnownGoodConstIterator(m_table + tableSize()); } iterator random() { if (isEmpty()) return end(); while (true) { auto& bucket = m_table[weakRandomUint32() & tableSizeMask()]; if (!isEmptyOrDeletedBucket(bucket)) return makeKnownGoodIterator(&bucket); }; } const_iterator random() const { return static_cast(const_cast(this)->random()); } unsigned size() const { return keyCount(); } unsigned capacity() const { return tableSize(); } bool isEmpty() const { return !keyCount(); } void reserveInitialCapacity(unsigned keyCount) { ASSERT(!m_table); ASSERT(!tableSize()); unsigned minimumTableSize = KeyTraits::minimumTableSize; unsigned newTableSize = std::max(minimumTableSize, computeBestTableSize(keyCount)); m_table = allocateTable(newTableSize); setTableSize(newTableSize); setTableSizeMask(newTableSize - 1); setDeletedCount(0); setKeyCount(0); } AddResult add(const ValueType& value) { return add(Extractor::extract(value), value); } AddResult add(ValueType&& value) { return add(Extractor::extract(value), WTFMove(value)); } // A special version of add() that finds the object by hashing and comparing // with some other type, to avoid the cost of type conversion if the object is already // in the table. template AddResult add(T&& key, Extra&&); template AddResult addPassingHashCode(T&& key, Extra&&); iterator find(const KeyType& key) { return find(key); } const_iterator find(const KeyType& key) const { return find(key); } bool contains(const KeyType& key) const { return contains(key); } template iterator find(const T&); template const_iterator find(const T&) const; template bool contains(const T&) const; void remove(const KeyType&); void remove(iterator); void removeWithoutEntryConsistencyCheck(iterator); void removeWithoutEntryConsistencyCheck(const_iterator); template bool removeIf(const Functor&); void clear(); static bool isEmptyBucket(const ValueType& value) { return isHashTraitsEmptyValue(Extractor::extract(value)); } static bool isReleasedWeakBucket(const ValueType& value) { return isHashTraitsReleasedWeakValue(Extractor::extract(value)); } static bool isDeletedBucket(const ValueType& value) { return KeyTraits::isDeletedValue(Extractor::extract(value)); } static bool isEmptyOrDeletedBucket(const ValueType& value) { return isEmptyBucket(value) || isDeletedBucket(value); } ValueType* lookup(const Key& key) { return lookup(key); } template ValueType* lookup(const T&); template ValueType* inlineLookup(const T&); ALWAYS_INLINE bool isNullStorage() const { return !m_table; } #if ASSERT_ENABLED void checkTableConsistency() const; #else static void checkTableConsistency() { } #endif #if CHECK_HASHTABLE_CONSISTENCY void internalCheckTableConsistency() const { checkTableConsistency(); } void internalCheckTableConsistencyExceptSize() const { checkTableConsistencyExceptSize(); } #else static void internalCheckTableConsistencyExceptSize() { } static void internalCheckTableConsistency() { } #endif private: static ValueType* allocateTable(unsigned size); static void deallocateTable(ValueType* table); typedef std::pair LookupType; typedef std::pair FullLookupType; LookupType lookupForWriting(const Key& key) { return lookupForWriting(key); }; template FullLookupType fullLookupForWriting(const T&); template LookupType lookupForWriting(const T&); template void addUniqueForInitialization(T&& key, Extra&&); template void checkKey(const T&); void removeAndInvalidateWithoutEntryConsistencyCheck(ValueType*); void removeAndInvalidate(ValueType*); void remove(ValueType*); static constexpr unsigned computeBestTableSize(unsigned keyCount); bool shouldExpand() const { return HashTableSizePolicy::shouldExpand(keyCount() + deletedCount(), tableSize()); } bool mustRehashInPlace() const { return keyCount() * minLoad < tableSize() * 2; } bool shouldShrink() const { return keyCount() * minLoad < tableSize() && tableSize() > KeyTraits::minimumTableSize; } ValueType* expand(ValueType* entry = nullptr); void shrink() { rehash(tableSize() / 2, nullptr); } void shrinkToBestSize(); void deleteReleasedWeakBuckets(); ValueType* rehash(unsigned newTableSize, ValueType* entry); ValueType* reinsert(ValueType&&); static void initializeBucket(ValueType& bucket); static void deleteBucket(ValueType& bucket) { hashTraitsDeleteBucket(bucket); } FullLookupType makeLookupResult(ValueType* position, bool found, unsigned hash) { return FullLookupType(LookupType(position, found), hash); } iterator makeIterator(ValueType* pos) { return iterator(this, pos, m_table + tableSize()); } const_iterator makeConstIterator(ValueType* pos) const { return const_iterator(this, pos, m_table + tableSize()); } iterator makeKnownGoodIterator(ValueType* pos) { return iterator(this, pos, m_table + tableSize(), HashItemKnownGood); } const_iterator makeKnownGoodConstIterator(ValueType* pos) const { return const_iterator(this, pos, m_table + tableSize(), HashItemKnownGood); } #if ASSERT_ENABLED void checkTableConsistencyExceptSize() const; #else static void checkTableConsistencyExceptSize() { } #endif // Load-factor for small table is 75%. static constexpr unsigned smallMaxLoadNumerator = HashTableSizePolicy::smallMaxLoadNumerator; static constexpr unsigned smallMaxLoadDenominator = HashTableSizePolicy::smallMaxLoadDenominator; // Load-factor for large table is 50%. static constexpr unsigned largeMaxLoadNumerator = HashTableSizePolicy::largeMaxLoadNumerator; static constexpr unsigned largeMaxLoadDenominator = HashTableSizePolicy::largeMaxLoadDenominator; static constexpr unsigned maxSmallTableCapacity = HashTableSizePolicy::maxSmallTableCapacity; static constexpr unsigned minLoad = HashTableSizePolicy::minLoad; static constexpr int tableSizeOffset = -1; static constexpr int tableSizeMaskOffset = -2; static constexpr int keyCountOffset = -3; static constexpr int deletedCountOffset = -4; static constexpr unsigned metadataSize = 4 * sizeof(unsigned); unsigned tableSize() const { return m_table ? reinterpret_cast_ptr(m_table)[tableSizeOffset] : 0; } void setTableSize(unsigned size) const { ASSERT(m_table); reinterpret_cast_ptr(m_table)[tableSizeOffset] = size; } unsigned tableSizeMask() const { ASSERT(m_table); return m_table ? reinterpret_cast_ptr(m_table)[tableSizeMaskOffset] : 0; } void setTableSizeMask(unsigned mask) { ASSERT(m_table); reinterpret_cast_ptr(m_table)[tableSizeMaskOffset] = mask; } unsigned keyCount() const { return m_table ? reinterpret_cast_ptr(m_table)[keyCountOffset] : 0; } void setKeyCount(unsigned count) const { ASSERT(m_table); reinterpret_cast_ptr(m_table)[keyCountOffset] = count; } unsigned deletedCount() const { ASSERT(m_table); return reinterpret_cast_ptr(m_table)[deletedCountOffset]; } void setDeletedCount(unsigned count) const { ASSERT(m_table); reinterpret_cast_ptr(m_table)[deletedCountOffset] = count; } union { ValueType* m_table { nullptr }; unsigned* m_tableForLLDB; }; #if CHECK_HASHTABLE_ITERATORS public: // All access to m_iterators should be guarded with m_mutex. mutable const_iterator* m_iterators; // Use std::unique_ptr so HashTable can still be memmove'd or memcpy'ed. mutable std::unique_ptr m_mutex; #endif #if DUMP_HASHTABLE_STATS_PER_TABLE public: mutable std::unique_ptr m_stats; #endif }; template inline HashTable::HashTable() : m_table(nullptr) #if CHECK_HASHTABLE_ITERATORS , m_iterators(0) , m_mutex(makeUnique()) #endif #if DUMP_HASHTABLE_STATS_PER_TABLE , m_stats(makeUnique()) #endif { } inline unsigned doubleHash(unsigned key) { key = ~key + (key >> 23); key ^= (key << 12); key ^= (key >> 7); key ^= (key << 2); key ^= (key >> 20); return key; } #if !ASSERT_ENABLED template template inline void HashTable::checkKey(const T&) { } #else // ASSERT_ENABLED template template void HashTable::checkKey(const T& key) { if (!HashFunctions::safeToCompareToEmptyOrDeleted) return; ASSERT(!HashTranslator::equal(KeyTraits::emptyValue(), key)); typename std::aligned_storage::value>::type deletedValueBuffer; ValueType* deletedValuePtr = reinterpret_cast_ptr(&deletedValueBuffer); ValueType& deletedValue = *deletedValuePtr; Traits::constructDeletedValue(deletedValue); ASSERT(!HashTranslator::equal(Extractor::extract(deletedValue), key)); } #endif // ASSERT_ENABLED template template inline auto HashTable::lookup(const T& key) -> ValueType* { return inlineLookup(key); } template template ALWAYS_INLINE auto HashTable::inlineLookup(const T& key) -> ValueType* { static_assert(sizeof(Value) <= 128, "Your HashTable types are too big to efficiently move when rehashing. Consider using UniqueRef instead"); checkKey(key); unsigned k = 0; ValueType* table = m_table; if (!table) return nullptr; unsigned sizeMask = tableSizeMask(); unsigned h = HashTranslator::hash(key); unsigned i = h & sizeMask; #if DUMP_HASHTABLE_STATS ++HashTableStats::numAccesses; unsigned probeCount = 0; #endif #if DUMP_HASHTABLE_STATS_PER_TABLE ++m_stats->numAccesses; #endif while (true) { ValueType* entry = table + i; // we count on the compiler to optimize out this branch if (HashFunctions::safeToCompareToEmptyOrDeleted) { if (HashTranslator::equal(Extractor::extract(*entry), key)) return entry; if (isEmptyBucket(*entry)) return nullptr; } else { if (isEmptyBucket(*entry)) return nullptr; if (!isDeletedBucket(*entry) && HashTranslator::equal(Extractor::extract(*entry), key)) return entry; } #if DUMP_HASHTABLE_STATS ++probeCount; HashTableStats::recordCollisionAtCount(probeCount); #endif #if DUMP_HASHTABLE_STATS_PER_TABLE m_stats->recordCollisionAtCount(probeCount); #endif if (k == 0) k = 1 | doubleHash(h); i = (i + k) & sizeMask; } } template template inline auto HashTable::lookupForWriting(const T& key) -> LookupType { ASSERT(m_table); checkKey(key); unsigned k = 0; ValueType* table = m_table; unsigned sizeMask = tableSizeMask(); unsigned h = HashTranslator::hash(key); unsigned i = h & sizeMask; #if DUMP_HASHTABLE_STATS ++HashTableStats::numAccesses; unsigned probeCount = 0; #endif #if DUMP_HASHTABLE_STATS_PER_TABLE ++m_stats->numAccesses; #endif ValueType* deletedEntry = nullptr; while (true) { ValueType* entry = table + i; // we count on the compiler to optimize out this branch if (HashFunctions::safeToCompareToEmptyOrDeleted) { if (isEmptyBucket(*entry)) return LookupType(deletedEntry ? deletedEntry : entry, false); if (HashTranslator::equal(Extractor::extract(*entry), key)) return LookupType(entry, true); if (isDeletedBucket(*entry)) deletedEntry = entry; } else { if (isEmptyBucket(*entry)) return LookupType(deletedEntry ? deletedEntry : entry, false); if (isDeletedBucket(*entry)) deletedEntry = entry; else if (HashTranslator::equal(Extractor::extract(*entry), key)) return LookupType(entry, true); } #if DUMP_HASHTABLE_STATS ++probeCount; HashTableStats::recordCollisionAtCount(probeCount); #endif #if DUMP_HASHTABLE_STATS_PER_TABLE m_stats->recordCollisionAtCount(probeCount); #endif if (k == 0) k = 1 | doubleHash(h); i = (i + k) & sizeMask; } } template template inline auto HashTable::fullLookupForWriting(const T& key) -> FullLookupType { ASSERT(m_table); checkKey(key); unsigned k = 0; ValueType* table = m_table; unsigned sizeMask = tableSizeMask(); unsigned h = HashTranslator::hash(key); unsigned i = h & sizeMask; #if DUMP_HASHTABLE_STATS ++HashTableStats::numAccesses; unsigned probeCount = 0; #endif #if DUMP_HASHTABLE_STATS_PER_TABLE ++m_stats->numAccesses; #endif ValueType* deletedEntry = nullptr; while (true) { ValueType* entry = table + i; // we count on the compiler to optimize out this branch if (HashFunctions::safeToCompareToEmptyOrDeleted) { if (isEmptyBucket(*entry)) return makeLookupResult(deletedEntry ? deletedEntry : entry, false, h); if (HashTranslator::equal(Extractor::extract(*entry), key)) return makeLookupResult(entry, true, h); if (isDeletedBucket(*entry)) deletedEntry = entry; } else { if (isEmptyBucket(*entry)) return makeLookupResult(deletedEntry ? deletedEntry : entry, false, h); if (isDeletedBucket(*entry)) deletedEntry = entry; else if (HashTranslator::equal(Extractor::extract(*entry), key)) return makeLookupResult(entry, true, h); } #if DUMP_HASHTABLE_STATS ++probeCount; HashTableStats::recordCollisionAtCount(probeCount); #endif #if DUMP_HASHTABLE_STATS_PER_TABLE m_stats->recordCollisionAtCount(probeCount); #endif if (k == 0) k = 1 | doubleHash(h); i = (i + k) & sizeMask; } } template template ALWAYS_INLINE void HashTable::addUniqueForInitialization(T&& key, Extra&& extra) { ASSERT(m_table); checkKey(key); invalidateIterators(this); internalCheckTableConsistency(); unsigned k = 0; ValueType* table = m_table; unsigned sizeMask = tableSizeMask(); unsigned h = HashTranslator::hash(key); unsigned i = h & sizeMask; #if DUMP_HASHTABLE_STATS ++HashTableStats::numAccesses; unsigned probeCount = 0; #endif #if DUMP_HASHTABLE_STATS_PER_TABLE ++m_stats->numAccesses; #endif ValueType* entry; while (true) { entry = table + i; if (isEmptyBucket(*entry)) break; #if DUMP_HASHTABLE_STATS ++probeCount; HashTableStats::recordCollisionAtCount(probeCount); #endif #if DUMP_HASHTABLE_STATS_PER_TABLE m_stats->recordCollisionAtCount(probeCount); #endif if (k == 0) k = 1 | doubleHash(h); i = (i + k) & sizeMask; } HashTranslator::translate(*entry, std::forward(key), std::forward(extra)); internalCheckTableConsistency(); } template struct HashTableBucketInitializer; template<> struct HashTableBucketInitializer { template static void initialize(Value& bucket) { Traits::template constructEmptyValue(bucket); } }; template<> struct HashTableBucketInitializer { template static void initialize(Value& bucket) { // This initializes the bucket without copying the empty value. // That makes it possible to use this with types that don't support copying. // The memset to 0 looks like a slow operation but is optimized by the compilers. memset(static_cast(std::addressof(bucket)), 0, sizeof(bucket)); } }; template inline void HashTable::initializeBucket(ValueType& bucket) { HashTableBucketInitializer::template initialize(bucket); } template template ALWAYS_INLINE auto HashTable::add(T&& key, Extra&& extra) -> AddResult { checkKey(key); invalidateIterators(this); if (!m_table) expand(nullptr); internalCheckTableConsistency(); ASSERT(m_table); unsigned k = 0; ValueType* table = m_table; unsigned sizeMask = tableSizeMask(); unsigned h = HashTranslator::hash(key); unsigned i = h & sizeMask; #if DUMP_HASHTABLE_STATS ++HashTableStats::numAccesses; unsigned probeCount = 0; #endif #if DUMP_HASHTABLE_STATS_PER_TABLE ++m_stats->numAccesses; #endif ValueType* deletedEntry = nullptr; ValueType* entry; while (true) { entry = table + i; // we count on the compiler to optimize out this branch if (HashFunctions::safeToCompareToEmptyOrDeleted) { if (isEmptyBucket(*entry)) break; if (HashTranslator::equal(Extractor::extract(*entry), key)) return AddResult(makeKnownGoodIterator(entry), false); if (isDeletedBucket(*entry)) deletedEntry = entry; } else { if (isEmptyBucket(*entry)) break; if (isDeletedBucket(*entry)) deletedEntry = entry; else if (HashTranslator::equal(Extractor::extract(*entry), key)) return AddResult(makeKnownGoodIterator(entry), false); } #if DUMP_HASHTABLE_STATS ++probeCount; HashTableStats::recordCollisionAtCount(probeCount); #endif #if DUMP_HASHTABLE_STATS_PER_TABLE m_stats->recordCollisionAtCount(probeCount); #endif if (k == 0) k = 1 | doubleHash(h); i = (i + k) & sizeMask; } if (deletedEntry) { initializeBucket(*deletedEntry); entry = deletedEntry; setDeletedCount(deletedCount() - 1); } HashTranslator::translate(*entry, std::forward(key), std::forward(extra)); setKeyCount(keyCount() + 1); if (shouldExpand()) entry = expand(entry); internalCheckTableConsistency(); return AddResult(makeKnownGoodIterator(entry), true); } template template inline auto HashTable::addPassingHashCode(T&& key, Extra&& extra) -> AddResult { checkKey(key); invalidateIterators(this); if (!m_table) expand(); internalCheckTableConsistency(); FullLookupType lookupResult = fullLookupForWriting(key); ValueType* entry = lookupResult.first.first; bool found = lookupResult.first.second; unsigned h = lookupResult.second; if (found) return AddResult(makeKnownGoodIterator(entry), false); if (isDeletedBucket(*entry)) { initializeBucket(*entry); setDeletedCount(deletedCount() - 1); } HashTranslator::translate(*entry, std::forward(key), std::forward(extra), h); setKeyCount(keyCount() + 1); if (shouldExpand()) entry = expand(entry); internalCheckTableConsistency(); return AddResult(makeKnownGoodIterator(entry), true); } template inline auto HashTable::reinsert(ValueType&& entry) -> ValueType* { ASSERT(m_table); ASSERT(!lookupForWriting(Extractor::extract(entry)).second); ASSERT(!isDeletedBucket(*(lookupForWriting(Extractor::extract(entry)).first))); #if DUMP_HASHTABLE_STATS ++HashTableStats::numReinserts; #endif #if DUMP_HASHTABLE_STATS_PER_TABLE ++m_stats->numReinserts; #endif Value* newEntry = lookupForWriting(Extractor::extract(entry)).first; newEntry->~Value(); new (NotNull, newEntry) ValueType(WTFMove(entry)); return newEntry; } template template auto HashTable::find(const T& key) -> iterator { if (!m_table) return end(); ValueType* entry = lookup(key); if (!entry) return end(); return makeKnownGoodIterator(entry); } template template auto HashTable::find(const T& key) const -> const_iterator { if (!m_table) return end(); ValueType* entry = const_cast(this)->lookup(key); if (!entry) return end(); return makeKnownGoodConstIterator(entry); } template template bool HashTable::contains(const T& key) const { if (!m_table) return false; return const_cast(this)->lookup(key); } template void HashTable::removeAndInvalidateWithoutEntryConsistencyCheck(ValueType* pos) { invalidateIterators(this); remove(pos); } template void HashTable::removeAndInvalidate(ValueType* pos) { invalidateIterators(this); internalCheckTableConsistency(); remove(pos); } template void HashTable::remove(ValueType* pos) { #if DUMP_HASHTABLE_STATS ++HashTableStats::numRemoves; #endif #if DUMP_HASHTABLE_STATS_PER_TABLE ++m_stats->numRemoves; #endif deleteBucket(*pos); setDeletedCount(deletedCount() + 1); setKeyCount(keyCount() - 1); if (shouldShrink()) shrink(); internalCheckTableConsistency(); } template inline void HashTable::remove(iterator it) { if (it == end()) return; removeAndInvalidate(const_cast(it.m_iterator.m_position)); } template inline void HashTable::removeWithoutEntryConsistencyCheck(iterator it) { if (it == end()) return; removeAndInvalidateWithoutEntryConsistencyCheck(const_cast(it.m_iterator.m_position)); } template inline void HashTable::removeWithoutEntryConsistencyCheck(const_iterator it) { if (it == end()) return; removeAndInvalidateWithoutEntryConsistencyCheck(const_cast(it.m_position)); } template inline void HashTable::remove(const KeyType& key) { remove(find(key)); } template template inline bool HashTable::removeIf(const Functor& functor) { // We must use local copies in case "functor" or "deleteBucket" // make a function call, which prevents the compiler from keeping // the values in register. unsigned removedBucketCount = 0; ValueType* table = m_table; for (unsigned i = tableSize(); i--;) { ValueType& bucket = table[i]; if (isEmptyOrDeletedBucket(bucket)) continue; if (!functor(bucket)) continue; deleteBucket(bucket); ++removedBucketCount; } if (removedBucketCount) { setDeletedCount(deletedCount() + removedBucketCount); setKeyCount(keyCount() - removedBucketCount); } if (shouldShrink()) shrinkToBestSize(); internalCheckTableConsistency(); return removedBucketCount; } template auto HashTable::allocateTable(unsigned size) -> ValueType* { static_assert(!(metadataSize % alignof(ValueType))); // would use a template member function with explicit specializations here, but // gcc doesn't appear to support that if (Traits::emptyValueIsZero) return reinterpret_cast_ptr(static_cast(HashTableMalloc::zeroedMalloc(metadataSize + size * sizeof(ValueType))) + metadataSize); ValueType* result = reinterpret_cast_ptr(static_cast(HashTableMalloc::malloc(metadataSize + size * sizeof(ValueType))) + metadataSize); for (unsigned i = 0; i < size; i++) initializeBucket(result[i]); return result; } template void HashTable::deallocateTable(ValueType* table) { unsigned size = reinterpret_cast_ptr(table)[tableSizeOffset]; for (unsigned i = 0; i < size; ++i) { if (!isDeletedBucket(table[i])) table[i].~ValueType(); } HashTableMalloc::free(reinterpret_cast(table) - metadataSize); } template auto HashTable::expand(ValueType* entry) -> ValueType* { if (KeyTraits::hasIsReleasedWeakValueFunction) deleteReleasedWeakBuckets(); unsigned newSize; unsigned oldSize = tableSize(); if (!oldSize) newSize = KeyTraits::minimumTableSize; else if (mustRehashInPlace()) newSize = oldSize; else newSize = oldSize * 2; return rehash(newSize, entry); } template constexpr unsigned HashTable::computeBestTableSize(unsigned keyCount) { unsigned bestTableSize = WTF::roundUpToPowerOfTwo(keyCount); if (HashTableSizePolicy::shouldExpand(keyCount, bestTableSize)) bestTableSize *= 2; auto aboveThresholdForEagerExpansion = [](double loadFactor, unsigned keyCount, unsigned tableSize) { // Here is the rationale behind this calculation, using 3/4 load-factor. // With maxLoad at 3/4 and minLoad at 1/6, our average load is 11/24. // If we are getting half-way between 11/24 and 3/4, we double the size // to avoid being too close to loadMax and bring the ratio close to 11/24. This // give us a load in the bounds [9/24, 15/24). double maxLoadRatio = loadFactor; double minLoadRatio = 1.0 / minLoad; double averageLoadRatio = (maxLoadRatio + minLoadRatio) / 2; double halfWayBetweenAverageAndMaxLoadRatio = (averageLoadRatio + maxLoadRatio) / 2; return keyCount >= tableSize * halfWayBetweenAverageAndMaxLoadRatio; }; if (bestTableSize <= maxSmallTableCapacity) { constexpr double smallLoadFactor = static_cast(smallMaxLoadNumerator) / smallMaxLoadDenominator; if (aboveThresholdForEagerExpansion(smallLoadFactor, keyCount, bestTableSize)) bestTableSize *= 2; } else { constexpr double largeLoadFactor = static_cast(largeMaxLoadNumerator) / largeMaxLoadDenominator; if (aboveThresholdForEagerExpansion(largeLoadFactor, keyCount, bestTableSize)) bestTableSize *= 2; } unsigned minimumTableSize = KeyTraits::minimumTableSize; return std::max(bestTableSize, minimumTableSize); } template void HashTable::shrinkToBestSize() { unsigned minimumTableSize = KeyTraits::minimumTableSize; rehash(std::max(minimumTableSize, computeBestTableSize(keyCount())), nullptr); } template void HashTable::deleteReleasedWeakBuckets() { unsigned tableSize = this->tableSize(); for (unsigned i = 0; i < tableSize; ++i) { auto& entry = m_table[i]; if (isReleasedWeakBucket(entry)) { deleteBucket(entry); setDeletedCount(deletedCount() + 1); setKeyCount(keyCount() - 1); } } } template auto HashTable::rehash(unsigned newTableSize, ValueType* entry) -> ValueType* { internalCheckTableConsistencyExceptSize(); unsigned oldTableSize = tableSize(); ValueType* oldTable = m_table; #if DUMP_HASHTABLE_STATS if (oldTableSize != 0) ++HashTableStats::numRehashes; #endif #if DUMP_HASHTABLE_STATS_PER_TABLE if (oldTableSize != 0) ++m_stats->numRehashes; #endif unsigned oldKeyCount = keyCount(); m_table = allocateTable(newTableSize); setTableSize(newTableSize); setTableSizeMask(newTableSize - 1); setDeletedCount(0); setKeyCount(oldKeyCount); Value* newEntry = nullptr; for (unsigned i = 0; i != oldTableSize; ++i) { auto& oldEntry = oldTable[i]; if (isDeletedBucket(oldEntry)) { ASSERT(std::addressof(oldEntry) != entry); continue; } if (isEmptyBucket(oldEntry)) { ASSERT(std::addressof(oldEntry) != entry); oldTable[i].~ValueType(); continue; } if (isReleasedWeakBucket(oldEntry)) { ASSERT(std::addressof(oldEntry) != entry); oldEntry.~ValueType(); setKeyCount(keyCount() - 1); continue; } Value* reinsertedEntry = reinsert(WTFMove(oldEntry)); oldEntry.~ValueType(); if (std::addressof(oldEntry) == entry) { ASSERT(!newEntry); newEntry = reinsertedEntry; } } if (oldTable) HashTableMalloc::free(reinterpret_cast(oldTable) - metadataSize); internalCheckTableConsistency(); return newEntry; } template void HashTable::clear() { invalidateIterators(this); if (!m_table) return; deallocateTable(std::exchange(m_table, nullptr)); } template HashTable::HashTable(const HashTable& other) : m_table(nullptr) #if CHECK_HASHTABLE_ITERATORS , m_iterators(nullptr) , m_mutex(makeUnique()) #endif #if DUMP_HASHTABLE_STATS_PER_TABLE , m_stats(makeUnique(*other.m_stats)) #endif { unsigned otherKeyCount = other.size(); if (!otherKeyCount) return; unsigned bestTableSize = computeBestTableSize(otherKeyCount); m_table = allocateTable(bestTableSize); setTableSize(bestTableSize); setTableSizeMask(bestTableSize - 1); setKeyCount(otherKeyCount); setDeletedCount(0); for (const auto& otherValue : other) addUniqueForInitialization(Extractor::extract(otherValue), otherValue); } template void HashTable::swap(HashTable& other) { invalidateIterators(this); invalidateIterators(&other); std::swap(m_table, other.m_table); #if DUMP_HASHTABLE_STATS_PER_TABLE m_stats.swap(other.m_stats); #endif } template auto HashTable::operator=(const HashTable& other) -> HashTable& { HashTable tmp(other); swap(tmp); return *this; } template inline HashTable::HashTable(HashTable&& other) #if CHECK_HASHTABLE_ITERATORS : m_iterators(nullptr) , m_mutex(makeUnique()) #endif { invalidateIterators(&other); m_table = std::exchange(other.m_table, nullptr); #if DUMP_HASHTABLE_STATS_PER_TABLE m_stats = WTFMove(other.m_stats); other.m_stats = nullptr; #endif } template inline auto HashTable::operator=(HashTable&& other) -> HashTable& { HashTable temp = WTFMove(other); swap(temp); return *this; } #if ASSERT_ENABLED template void HashTable::checkTableConsistency() const { checkTableConsistencyExceptSize(); ASSERT(!m_table || !shouldExpand()); ASSERT(!shouldShrink()); } template void HashTable::checkTableConsistencyExceptSize() const { if (!m_table) return; unsigned count = 0; unsigned deletedCount = 0; unsigned tableSize = this->tableSize(); for (unsigned j = 0; j < tableSize; ++j) { ValueType* entry = m_table + j; if (isEmptyBucket(*entry)) continue; if (isDeletedBucket(*entry)) { ++deletedCount; continue; } auto& key = Extractor::extract(*entry); const_iterator it = find(key); ASSERT(entry == it.m_position); ++count; ValueCheck::checkConsistency(key); } ASSERT(count == keyCount()); ASSERT(deletedCount == this->deletedCount()); ASSERT(this->tableSize() >= KeyTraits::minimumTableSize); ASSERT(tableSizeMask()); ASSERT(this->tableSize() == tableSizeMask() + 1); } #endif // ASSERT_ENABLED #if CHECK_HASHTABLE_ITERATORS template void invalidateIterators(const HashTableType* table) { Locker locker { *table->m_mutex }; typename HashTableType::const_iterator* next; for (typename HashTableType::const_iterator* p = table->m_iterators; p; p = next) { next = p->m_next; p->m_table = nullptr; p->m_next = nullptr; p->m_previous = nullptr; } table->m_iterators = nullptr; } template void addIterator(const HashTableType* table, HashTableConstIterator* it) { it->m_table = table; it->m_previous = nullptr; // Insert iterator at head of doubly-linked list of iterators. if (!table) { it->m_next = nullptr; } else { Locker locker { *table->m_mutex }; ASSERT(table->m_iterators != it); it->m_next = table->m_iterators; table->m_iterators = it; if (it->m_next) { ASSERT(!it->m_next->m_previous); it->m_next->m_previous = it; } } } template void removeIterator(HashTableConstIterator* it) { // Delete iterator from doubly-linked list of iterators. if (!it->m_table) { ASSERT(!it->m_next); ASSERT(!it->m_previous); } else { Locker locker { *it->m_table->m_mutex }; if (it->m_next) { ASSERT(it->m_next->m_previous == it); it->m_next->m_previous = it->m_previous; } if (it->m_previous) { ASSERT(it->m_table->m_iterators != it); ASSERT(it->m_previous->m_next == it); it->m_previous->m_next = it->m_next; } else { ASSERT(it->m_table->m_iterators == it); it->m_table->m_iterators = it->m_next; } } it->m_table = nullptr; it->m_next = nullptr; it->m_previous = nullptr; } #endif // CHECK_HASHTABLE_ITERATORS struct HashTableTraits { template using TableType = HashTable; }; // iterator adapters template struct HashTableConstIteratorAdapter : public std::iterator { HashTableConstIteratorAdapter() {} HashTableConstIteratorAdapter(const typename HashTableType::const_iterator& impl) : m_impl(impl) {} const ValueType* get() const { return (const ValueType*)m_impl.get(); } const ValueType& operator*() const { return *get(); } const ValueType* operator->() const { return get(); } HashTableConstIteratorAdapter& operator++() { ++m_impl; return *this; } // postfix ++ intentionally omitted typename HashTableType::const_iterator m_impl; }; template struct HashTableIteratorAdapter : public std::iterator { HashTableIteratorAdapter() {} HashTableIteratorAdapter(const typename HashTableType::iterator& impl) : m_impl(impl) {} ValueType* get() const { return (ValueType*)m_impl.get(); } ValueType& operator*() const { return *get(); } ValueType* operator->() const { return get(); } HashTableIteratorAdapter& operator++() { ++m_impl; return *this; } // postfix ++ intentionally omitted operator HashTableConstIteratorAdapter() { typename HashTableType::const_iterator i = m_impl; return i; } typename HashTableType::iterator m_impl; }; template inline bool operator==(const HashTableConstIteratorAdapter& a, const HashTableConstIteratorAdapter& b) { return a.m_impl == b.m_impl; } template inline bool operator!=(const HashTableConstIteratorAdapter& a, const HashTableConstIteratorAdapter& b) { return a.m_impl != b.m_impl; } template inline bool operator==(const HashTableIteratorAdapter& a, const HashTableIteratorAdapter& b) { return a.m_impl == b.m_impl; } template inline bool operator!=(const HashTableIteratorAdapter& a, const HashTableIteratorAdapter& b) { return a.m_impl != b.m_impl; } // All 4 combinations of ==, != and Const,non const. template inline bool operator==(const HashTableConstIteratorAdapter& a, const HashTableIteratorAdapter& b) { return a.m_impl == b.m_impl; } template inline bool operator!=(const HashTableConstIteratorAdapter& a, const HashTableIteratorAdapter& b) { return a.m_impl != b.m_impl; } template inline bool operator==(const HashTableIteratorAdapter& a, const HashTableConstIteratorAdapter& b) { return a.m_impl == b.m_impl; } template inline bool operator!=(const HashTableIteratorAdapter& a, const HashTableConstIteratorAdapter& b) { return a.m_impl != b.m_impl; } } // namespace WTF #include