// Copyright (c) 2011-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 #include #include "rocksdb/slice.h" #include "port/port.h" #include "util/hash.h" #include namespace rocksdb { class Slice; class Allocator; class Logger; // A legacy Bloom filter implementation used by Plain Table db format, for // schema backward compatibility. Not for use in new filter applications. class PlainTableBloomV1 { public: // allocator: pass allocator to bloom filter, hence trace the usage of memory // total_bits: fixed total bits for the bloom // num_probes: number of hash probes for a single key // locality: If positive, optimize for cache line locality, 0 otherwise. // hash_func: customized hash function // huge_page_tlb_size: if >0, try to allocate bloom bytes from huge page TLB // within this page size. Need to reserve huge pages for // it to be allocated, like: // sysctl -w vm.nr_hugepages=20 // See linux doc Documentation/vm/hugetlbpage.txt explicit PlainTableBloomV1(uint32_t num_probes = 6); void SetTotalBits(Allocator* allocator, uint32_t total_bits, uint32_t locality, size_t huge_page_tlb_size, Logger* logger); ~PlainTableBloomV1() {} // Assuming single threaded access to this function. void AddHash(uint32_t hash); // Multithreaded access to this function is OK bool MayContainHash(uint32_t hash) const; void Prefetch(uint32_t hash); uint32_t GetNumBlocks() const { return kNumBlocks; } Slice GetRawData() const { return Slice(reinterpret_cast(data_), GetTotalBits() / 8); } void SetRawData(unsigned char* raw_data, uint32_t total_bits, uint32_t num_blocks = 0); uint32_t GetTotalBits() const { return kTotalBits; } bool IsInitialized() const { return kNumBlocks > 0 || kTotalBits > 0; } private: uint32_t kTotalBits; uint32_t kNumBlocks; const uint32_t kNumProbes; uint8_t* data_; }; #if defined(_MSC_VER) #pragma warning(push) // local variable is initialized but not referenced #pragma warning(disable : 4189) #endif inline void PlainTableBloomV1::Prefetch(uint32_t h) { if (kNumBlocks != 0) { uint32_t b = ((h >> 11 | (h << 21)) % kNumBlocks) * (CACHE_LINE_SIZE * 8); PREFETCH(&(data_[b / 8]), 0, 3); } } #if defined(_MSC_VER) #pragma warning(pop) #endif inline bool PlainTableBloomV1::MayContainHash(uint32_t h) const { assert(IsInitialized()); const uint32_t delta = (h >> 17) | (h << 15); // Rotate right 17 bits if (kNumBlocks != 0) { uint32_t b = ((h >> 11 | (h << 21)) % kNumBlocks) * (CACHE_LINE_SIZE * 8); for (uint32_t i = 0; i < kNumProbes; ++i) { // Since CACHE_LINE_SIZE is defined as 2^n, this line will be optimized // to a simple and operation by compiler. const uint32_t bitpos = b + (h % (CACHE_LINE_SIZE * 8)); if ((data_[bitpos / 8] & (1 << (bitpos % 8))) == 0) { return false; } // Rotate h so that we don't reuse the same bytes. h = h / (CACHE_LINE_SIZE * 8) + (h % (CACHE_LINE_SIZE * 8)) * (0x20000000U / CACHE_LINE_SIZE); h += delta; } } else { for (uint32_t i = 0; i < kNumProbes; ++i) { const uint32_t bitpos = h % kTotalBits; if ((data_[bitpos / 8] & (1 << (bitpos % 8))) == 0) { return false; } h += delta; } } return true; } inline void PlainTableBloomV1::AddHash(uint32_t h) { assert(IsInitialized()); const uint32_t delta = (h >> 17) | (h << 15); // Rotate right 17 bits if (kNumBlocks != 0) { uint32_t b = ((h >> 11 | (h << 21)) % kNumBlocks) * (CACHE_LINE_SIZE * 8); for (uint32_t i = 0; i < kNumProbes; ++i) { // Since CACHE_LINE_SIZE is defined as 2^n, this line will be optimized // to a simple and operation by compiler. const uint32_t bitpos = b + (h % (CACHE_LINE_SIZE * 8)); data_[bitpos / 8] |= (1 << (bitpos % 8)); // Rotate h so that we don't reuse the same bytes. h = h / (CACHE_LINE_SIZE * 8) + (h % (CACHE_LINE_SIZE * 8)) * (0x20000000U / CACHE_LINE_SIZE); h += delta; } } else { for (uint32_t i = 0; i < kNumProbes; ++i) { const uint32_t bitpos = h % kTotalBits; data_[bitpos / 8] |= (1 << (bitpos % 8)); h += delta; } } } class BloomBlockBuilder { public: static const std::string kBloomBlock; explicit BloomBlockBuilder(uint32_t num_probes = 6) : bloom_(num_probes) {} void SetTotalBits(Allocator* allocator, uint32_t total_bits, uint32_t locality, size_t huge_page_tlb_size, Logger* logger) { bloom_.SetTotalBits(allocator, total_bits, locality, huge_page_tlb_size, logger); } uint32_t GetNumBlocks() const { return bloom_.GetNumBlocks(); } void AddKeysHashes(const std::vector& keys_hashes); Slice Finish(); private: PlainTableBloomV1 bloom_; }; }; // namespace rocksdb