// 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). // // Copyright (c) 2011 The LevelDB Authors. All rights reserved. // Use of this source code is governed by a BSD-style license that can be // found in the LICENSE file. See the AUTHORS file for names of contributors. // // Endian-neutral encoding: // * Fixed-length numbers are encoded with least-significant byte first // * In addition we support variable length "varint" encoding // * Strings are encoded prefixed by their length in varint format #pragma once #include #include #include #include #include "rocksdb/write_batch.h" #include "port/port.h" // Some processors does not allow unaligned access to memory #if defined(__sparc) #define PLATFORM_UNALIGNED_ACCESS_NOT_ALLOWED #endif namespace rocksdb { // The maximum length of a varint in bytes for 64-bit. const unsigned int kMaxVarint64Length = 10; // Standard Put... routines append to a string extern void PutFixed32(std::string* dst, uint32_t value); extern void PutFixed64(std::string* dst, uint64_t value); extern void PutVarint32(std::string* dst, uint32_t value); extern void PutVarint32Varint32(std::string* dst, uint32_t value1, uint32_t value2); extern void PutVarint32Varint32Varint32(std::string* dst, uint32_t value1, uint32_t value2, uint32_t value3); extern void PutVarint64(std::string* dst, uint64_t value); extern void PutVarint64Varint64(std::string* dst, uint64_t value1, uint64_t value2); extern void PutVarint32Varint64(std::string* dst, uint32_t value1, uint64_t value2); extern void PutVarint32Varint32Varint64(std::string* dst, uint32_t value1, uint32_t value2, uint64_t value3); extern void PutLengthPrefixedSlice(std::string* dst, const Slice& value); extern void PutLengthPrefixedSliceParts(std::string* dst, const SliceParts& slice_parts); // Standard Get... routines parse a value from the beginning of a Slice // and advance the slice past the parsed value. extern bool GetFixed64(Slice* input, uint64_t* value); extern bool GetFixed32(Slice* input, uint32_t* value); extern bool GetVarint32(Slice* input, uint32_t* value); extern bool GetVarint64(Slice* input, uint64_t* value); extern bool GetLengthPrefixedSlice(Slice* input, Slice* result); // This function assumes data is well-formed. extern Slice GetLengthPrefixedSlice(const char* data); extern Slice GetSliceUntil(Slice* slice, char delimiter); // Pointer-based variants of GetVarint... These either store a value // in *v and return a pointer just past the parsed value, or return // nullptr on error. These routines only look at bytes in the range // [p..limit-1] extern const char* GetVarint32Ptr(const char* p,const char* limit, uint32_t* v); extern const char* GetVarint64Ptr(const char* p,const char* limit, uint64_t* v); // Returns the length of the varint32 or varint64 encoding of "v" extern int VarintLength(uint64_t v); // Lower-level versions of Put... that write directly into a character buffer // REQUIRES: dst has enough space for the value being written extern void EncodeFixed32(char* dst, uint32_t value); extern void EncodeFixed64(char* dst, uint64_t value); // Lower-level versions of Put... that write directly into a character buffer // and return a pointer just past the last byte written. // REQUIRES: dst has enough space for the value being written extern char* EncodeVarint32(char* dst, uint32_t value); extern char* EncodeVarint64(char* dst, uint64_t value); // Lower-level versions of Get... that read directly from a character buffer // without any bounds checking. inline uint32_t DecodeFixed32(const char* ptr) { if (port::kLittleEndian) { // Load the raw bytes uint32_t result; memcpy(&result, ptr, sizeof(result)); // gcc optimizes this to a plain load return result; } else { return ((static_cast(static_cast(ptr[0]))) | (static_cast(static_cast(ptr[1])) << 8) | (static_cast(static_cast(ptr[2])) << 16) | (static_cast(static_cast(ptr[3])) << 24)); } } inline uint64_t DecodeFixed64(const char* ptr) { if (port::kLittleEndian) { // Load the raw bytes uint64_t result; memcpy(&result, ptr, sizeof(result)); // gcc optimizes this to a plain load return result; } else { uint64_t lo = DecodeFixed32(ptr); uint64_t hi = DecodeFixed32(ptr + 4); return (hi << 32) | lo; } } // Internal routine for use by fallback path of GetVarint32Ptr extern const char* GetVarint32PtrFallback(const char* p, const char* limit, uint32_t* value); inline const char* GetVarint32Ptr(const char* p, const char* limit, uint32_t* value) { if (p < limit) { uint32_t result = *(reinterpret_cast(p)); if ((result & 128) == 0) { *value = result; return p + 1; } } return GetVarint32PtrFallback(p, limit, value); } // -- Implementation of the functions declared above inline void EncodeFixed32(char* buf, uint32_t value) { if (port::kLittleEndian) { memcpy(buf, &value, sizeof(value)); } else { buf[0] = value & 0xff; buf[1] = (value >> 8) & 0xff; buf[2] = (value >> 16) & 0xff; buf[3] = (value >> 24) & 0xff; } } inline void EncodeFixed64(char* buf, uint64_t value) { if (port::kLittleEndian) { memcpy(buf, &value, sizeof(value)); } else { buf[0] = value & 0xff; buf[1] = (value >> 8) & 0xff; buf[2] = (value >> 16) & 0xff; buf[3] = (value >> 24) & 0xff; buf[4] = (value >> 32) & 0xff; buf[5] = (value >> 40) & 0xff; buf[6] = (value >> 48) & 0xff; buf[7] = (value >> 56) & 0xff; } } // Pull the last 8 bits and cast it to a character inline void PutFixed32(std::string* dst, uint32_t value) { if (port::kLittleEndian) { dst->append(const_cast(reinterpret_cast(&value)), sizeof(value)); } else { char buf[sizeof(value)]; EncodeFixed32(buf, value); dst->append(buf, sizeof(buf)); } } inline void PutFixed64(std::string* dst, uint64_t value) { if (port::kLittleEndian) { dst->append(const_cast(reinterpret_cast(&value)), sizeof(value)); } else { char buf[sizeof(value)]; EncodeFixed64(buf, value); dst->append(buf, sizeof(buf)); } } inline void PutVarint32(std::string* dst, uint32_t v) { char buf[5]; char* ptr = EncodeVarint32(buf, v); dst->append(buf, static_cast(ptr - buf)); } inline void PutVarint32Varint32(std::string* dst, uint32_t v1, uint32_t v2) { char buf[10]; char* ptr = EncodeVarint32(buf, v1); ptr = EncodeVarint32(ptr, v2); dst->append(buf, static_cast(ptr - buf)); } inline void PutVarint32Varint32Varint32(std::string* dst, uint32_t v1, uint32_t v2, uint32_t v3) { char buf[15]; char* ptr = EncodeVarint32(buf, v1); ptr = EncodeVarint32(ptr, v2); ptr = EncodeVarint32(ptr, v3); dst->append(buf, static_cast(ptr - buf)); } inline char* EncodeVarint64(char* dst, uint64_t v) { static const unsigned int B = 128; unsigned char* ptr = reinterpret_cast(dst); while (v >= B) { *(ptr++) = (v & (B - 1)) | B; v >>= 7; } *(ptr++) = static_cast(v); return reinterpret_cast(ptr); } inline void PutVarint64(std::string* dst, uint64_t v) { char buf[10]; char* ptr = EncodeVarint64(buf, v); dst->append(buf, static_cast(ptr - buf)); } inline void PutVarint64Varint64(std::string* dst, uint64_t v1, uint64_t v2) { char buf[20]; char* ptr = EncodeVarint64(buf, v1); ptr = EncodeVarint64(ptr, v2); dst->append(buf, static_cast(ptr - buf)); } inline void PutVarint32Varint64(std::string* dst, uint32_t v1, uint64_t v2) { char buf[15]; char* ptr = EncodeVarint32(buf, v1); ptr = EncodeVarint64(ptr, v2); dst->append(buf, static_cast(ptr - buf)); } inline void PutVarint32Varint32Varint64(std::string* dst, uint32_t v1, uint32_t v2, uint64_t v3) { char buf[20]; char* ptr = EncodeVarint32(buf, v1); ptr = EncodeVarint32(ptr, v2); ptr = EncodeVarint64(ptr, v3); dst->append(buf, static_cast(ptr - buf)); } inline void PutLengthPrefixedSlice(std::string* dst, const Slice& value) { PutVarint32(dst, static_cast(value.size())); dst->append(value.data(), value.size()); } inline void PutLengthPrefixedSliceParts(std::string* dst, const SliceParts& slice_parts) { size_t total_bytes = 0; for (int i = 0; i < slice_parts.num_parts; ++i) { total_bytes += slice_parts.parts[i].size(); } PutVarint32(dst, static_cast(total_bytes)); for (int i = 0; i < slice_parts.num_parts; ++i) { dst->append(slice_parts.parts[i].data(), slice_parts.parts[i].size()); } } inline int VarintLength(uint64_t v) { int len = 1; while (v >= 128) { v >>= 7; len++; } return len; } inline bool GetFixed64(Slice* input, uint64_t* value) { if (input->size() < sizeof(uint64_t)) { return false; } *value = DecodeFixed64(input->data()); input->remove_prefix(sizeof(uint64_t)); return true; } inline bool GetFixed32(Slice* input, uint32_t* value) { if (input->size() < sizeof(uint32_t)) { return false; } *value = DecodeFixed32(input->data()); input->remove_prefix(sizeof(uint32_t)); return true; } inline bool GetVarint32(Slice* input, uint32_t* value) { const char* p = input->data(); const char* limit = p + input->size(); const char* q = GetVarint32Ptr(p, limit, value); if (q == nullptr) { return false; } else { *input = Slice(q, static_cast(limit - q)); return true; } } inline bool GetVarint64(Slice* input, uint64_t* value) { const char* p = input->data(); const char* limit = p + input->size(); const char* q = GetVarint64Ptr(p, limit, value); if (q == nullptr) { return false; } else { *input = Slice(q, static_cast(limit - q)); return true; } } // Provide an interface for platform independent endianness transformation inline uint64_t EndianTransform(uint64_t input, size_t size) { char* pos = reinterpret_cast(&input); uint64_t ret_val = 0; for (size_t i = 0; i < size; ++i) { ret_val |= (static_cast(static_cast(pos[i])) << ((size - i - 1) << 3)); } return ret_val; } inline bool GetLengthPrefixedSlice(Slice* input, Slice* result) { uint32_t len = 0; if (GetVarint32(input, &len) && input->size() >= len) { *result = Slice(input->data(), len); input->remove_prefix(len); return true; } else { return false; } } inline Slice GetLengthPrefixedSlice(const char* data) { uint32_t len = 0; // +5: we assume "data" is not corrupted // unsigned char is 7 bits, uint32_t is 32 bits, need 5 unsigned char auto p = GetVarint32Ptr(data, data + 5 /* limit */, &len); return Slice(p, len); } inline Slice GetSliceUntil(Slice* slice, char delimiter) { uint32_t len = 0; for (len = 0; len < slice->size() && slice->data()[len] != delimiter; ++len) { // nothing } Slice ret(slice->data(), len); slice->remove_prefix(len + ((len < slice->size()) ? 1 : 0)); return ret; } template #ifdef ROCKSDB_UBSAN_RUN #if defined(__clang__) __attribute__((__no_sanitize__("alignment"))) #elif defined(__GNUC__) __attribute__((__no_sanitize_undefined__)) #endif #endif inline void PutUnaligned(T *memory, const T &value) { #if defined(PLATFORM_UNALIGNED_ACCESS_NOT_ALLOWED) char *nonAlignedMemory = reinterpret_cast(memory); memcpy(nonAlignedMemory, reinterpret_cast(&value), sizeof(T)); #else *memory = value; #endif } template #ifdef ROCKSDB_UBSAN_RUN #if defined(__clang__) __attribute__((__no_sanitize__("alignment"))) #elif defined(__GNUC__) __attribute__((__no_sanitize_undefined__)) #endif #endif inline void GetUnaligned(const T *memory, T *value) { #if defined(PLATFORM_UNALIGNED_ACCESS_NOT_ALLOWED) char *nonAlignedMemory = reinterpret_cast(value); memcpy(nonAlignedMemory, reinterpret_cast(memory), sizeof(T)); #else *value = *memory; #endif } } // namespace rocksdb