0
0
Fork 0
mirror of https://github.com/bitcoin/bitcoin.git synced 2025-03-09 15:37:00 -04:00
bitcoin-core/src/dbwrapper.cpp
fanquake b565485c24
Merge bitcoin/bitcoin#28186: kernel: Prune leveldb headers
d8f1222ac5 refactor: Correct dbwrapper key naming (TheCharlatan)
be8f159ac5 build: Remove leveldb from BITCOIN_INCLUDES (TheCharlatan)
c95b37d641 refactor: Move CDBWrapper leveldb members to their own context struct (TheCharlatan)
c534a615e9 refactor: Split dbwrapper CDBWrapper::EstimateSize implementation (TheCharlatan)
586448888b refactor: Move HandleError to dbwrapper implementation (TheCharlatan)
dede0eef7a refactor: Split dbwrapper CDBWrapper::Exists implementation (TheCharlatan)
a5c2eb5748 refactor: Fix logging.h includes (TheCharlatan)
84058e0eed refactor: Split dbwrapper CDBWrapper::Read implementation (TheCharlatan)
e4af2408f2 refactor: Pimpl leveldb::Iterator for CDBIterator (TheCharlatan)
ef941ff128 refactor: Split dbwrapper CDBIterator::GetValue implementation (TheCharlatan)
b7a1ab5cb4 refactor: Split dbwrapper CDBIterator::GetKey implementation (TheCharlatan)
d7437908cd refactor: Split dbwrapper CDBIterator::Seek implementation (TheCharlatan)
ea8135de7e refactor: Pimpl leveldb::batch for CDBBatch (TheCharlatan)
b9870c920d refactor: Split dbwrapper CDBatch::Erase implementation (TheCharlatan)
532ee812a4 refactor: Split dbwrapper CDBBatch::Write implementation (TheCharlatan)
afc534df9a refactor: Wrap DestroyDB in dbwrapper helper (TheCharlatan)

Pull request description:

  Leveldb headers are currently included in the `dbwrapper.h` file and thus available to many of Bitcoin Core's source files. However, leveldb-specific functionality should be abstracted by the `dbwrapper` and does not need to be available to the rest of the code. Having leveldb included in a widely-used header such as `dbwrapper.h` bloats the entire project's header tree.

  The `dbwrapper` is a key component of the libbitcoinkernel library. Future users of this library would not want to contend with  having the leveldb headers exposed and potentially polluting their project's namespace.

  For these reasons, the leveldb headers are removed from the `dbwrapper` by moving leveldb-specific code to the implementation file and creating a [pimpl](https://en.cppreference.com/w/cpp/language/pimpl) where leveldb member variables are indispensable. As a final step, the leveldb include flags are removed from the `BITCOIN_INCLUDES` and moved to places where the dbwrapper is compiled.

  ---

  This pull request is part of the [libbitcoinkernel project](https://github.com/bitcoin/bitcoin/issues/27587), and more specifically its stage 1 step 3 "Decouple most non-consensus headers from libbitcoinkernel".

ACKs for top commit:
  stickies-v:
    re-ACK d8f1222ac5
  MarcoFalke:
    ACK d8f1222ac5  🔠

Tree-SHA512: 0f58309be165af0162e648233451cd80fda88726fc10c0da7bfe4ec2ffa9afe63fbf7ffae9493698d3f39653b4ad870c372eee652ecc90ab1c29d86c387070f3
2023-08-07 22:31:46 +02:00

427 lines
15 KiB
C++

// Copyright (c) 2012-2022 The Bitcoin Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <dbwrapper.h>
#include <clientversion.h>
#include <logging.h>
#include <random.h>
#include <serialize.h>
#include <span.h>
#include <streams.h>
#include <util/fs.h>
#include <util/fs_helpers.h>
#include <util/strencodings.h>
#include <algorithm>
#include <cassert>
#include <cstdarg>
#include <cstdint>
#include <cstdio>
#include <leveldb/cache.h>
#include <leveldb/db.h>
#include <leveldb/env.h>
#include <leveldb/filter_policy.h>
#include <leveldb/helpers/memenv/memenv.h>
#include <leveldb/iterator.h>
#include <leveldb/options.h>
#include <leveldb/slice.h>
#include <leveldb/status.h>
#include <leveldb/write_batch.h>
#include <memory>
#include <optional>
#include <utility>
static auto CharCast(const std::byte* data) { return reinterpret_cast<const char*>(data); }
bool DestroyDB(const std::string& path_str)
{
return leveldb::DestroyDB(path_str, {}).ok();
}
/** Handle database error by throwing dbwrapper_error exception.
*/
static void HandleError(const leveldb::Status& status)
{
if (status.ok())
return;
const std::string errmsg = "Fatal LevelDB error: " + status.ToString();
LogPrintf("%s\n", errmsg);
LogPrintf("You can use -debug=leveldb to get more complete diagnostic messages\n");
throw dbwrapper_error(errmsg);
}
class CBitcoinLevelDBLogger : public leveldb::Logger {
public:
// This code is adapted from posix_logger.h, which is why it is using vsprintf.
// Please do not do this in normal code
void Logv(const char * format, va_list ap) override {
if (!LogAcceptCategory(BCLog::LEVELDB, BCLog::Level::Debug)) {
return;
}
char buffer[500];
for (int iter = 0; iter < 2; iter++) {
char* base;
int bufsize;
if (iter == 0) {
bufsize = sizeof(buffer);
base = buffer;
}
else {
bufsize = 30000;
base = new char[bufsize];
}
char* p = base;
char* limit = base + bufsize;
// Print the message
if (p < limit) {
va_list backup_ap;
va_copy(backup_ap, ap);
// Do not use vsnprintf elsewhere in bitcoin source code, see above.
p += vsnprintf(p, limit - p, format, backup_ap);
va_end(backup_ap);
}
// Truncate to available space if necessary
if (p >= limit) {
if (iter == 0) {
continue; // Try again with larger buffer
}
else {
p = limit - 1;
}
}
// Add newline if necessary
if (p == base || p[-1] != '\n') {
*p++ = '\n';
}
assert(p <= limit);
base[std::min(bufsize - 1, (int)(p - base))] = '\0';
LogPrintLevel(BCLog::LEVELDB, BCLog::Level::Debug, "%s", base); // NOLINT(bitcoin-unterminated-logprintf)
if (base != buffer) {
delete[] base;
}
break;
}
}
};
static void SetMaxOpenFiles(leveldb::Options *options) {
// On most platforms the default setting of max_open_files (which is 1000)
// is optimal. On Windows using a large file count is OK because the handles
// do not interfere with select() loops. On 64-bit Unix hosts this value is
// also OK, because up to that amount LevelDB will use an mmap
// implementation that does not use extra file descriptors (the fds are
// closed after being mmap'ed).
//
// Increasing the value beyond the default is dangerous because LevelDB will
// fall back to a non-mmap implementation when the file count is too large.
// On 32-bit Unix host we should decrease the value because the handles use
// up real fds, and we want to avoid fd exhaustion issues.
//
// See PR #12495 for further discussion.
int default_open_files = options->max_open_files;
#ifndef WIN32
if (sizeof(void*) < 8) {
options->max_open_files = 64;
}
#endif
LogPrint(BCLog::LEVELDB, "LevelDB using max_open_files=%d (default=%d)\n",
options->max_open_files, default_open_files);
}
static leveldb::Options GetOptions(size_t nCacheSize)
{
leveldb::Options options;
options.block_cache = leveldb::NewLRUCache(nCacheSize / 2);
options.write_buffer_size = nCacheSize / 4; // up to two write buffers may be held in memory simultaneously
options.filter_policy = leveldb::NewBloomFilterPolicy(10);
options.compression = leveldb::kNoCompression;
options.info_log = new CBitcoinLevelDBLogger();
if (leveldb::kMajorVersion > 1 || (leveldb::kMajorVersion == 1 && leveldb::kMinorVersion >= 16)) {
// LevelDB versions before 1.16 consider short writes to be corruption. Only trigger error
// on corruption in later versions.
options.paranoid_checks = true;
}
SetMaxOpenFiles(&options);
return options;
}
struct CDBBatch::WriteBatchImpl {
leveldb::WriteBatch batch;
};
CDBBatch::CDBBatch(const CDBWrapper& _parent) : parent(_parent),
m_impl_batch{std::make_unique<CDBBatch::WriteBatchImpl>()},
ssValue(SER_DISK, CLIENT_VERSION){};
CDBBatch::~CDBBatch() = default;
void CDBBatch::Clear()
{
m_impl_batch->batch.Clear();
size_estimate = 0;
}
void CDBBatch::WriteImpl(Span<const std::byte> key, CDataStream& ssValue)
{
leveldb::Slice slKey(CharCast(key.data()), key.size());
ssValue.Xor(dbwrapper_private::GetObfuscateKey(parent));
leveldb::Slice slValue(CharCast(ssValue.data()), ssValue.size());
m_impl_batch->batch.Put(slKey, slValue);
// LevelDB serializes writes as:
// - byte: header
// - varint: key length (1 byte up to 127B, 2 bytes up to 16383B, ...)
// - byte[]: key
// - varint: value length
// - byte[]: value
// The formula below assumes the key and value are both less than 16k.
size_estimate += 3 + (slKey.size() > 127) + slKey.size() + (slValue.size() > 127) + slValue.size();
}
void CDBBatch::EraseImpl(Span<const std::byte> key)
{
leveldb::Slice slKey(CharCast(key.data()), key.size());
m_impl_batch->batch.Delete(slKey);
// LevelDB serializes erases as:
// - byte: header
// - varint: key length
// - byte[]: key
// The formula below assumes the key is less than 16kB.
size_estimate += 2 + (slKey.size() > 127) + slKey.size();
}
struct LevelDBContext {
//! custom environment this database is using (may be nullptr in case of default environment)
leveldb::Env* penv;
//! database options used
leveldb::Options options;
//! options used when reading from the database
leveldb::ReadOptions readoptions;
//! options used when iterating over values of the database
leveldb::ReadOptions iteroptions;
//! options used when writing to the database
leveldb::WriteOptions writeoptions;
//! options used when sync writing to the database
leveldb::WriteOptions syncoptions;
//! the database itself
leveldb::DB* pdb;
};
CDBWrapper::CDBWrapper(const DBParams& params)
: m_db_context{std::make_unique<LevelDBContext>()}, m_name{fs::PathToString(params.path.stem())}, m_path{params.path}, m_is_memory{params.memory_only}
{
DBContext().penv = nullptr;
DBContext().readoptions.verify_checksums = true;
DBContext().iteroptions.verify_checksums = true;
DBContext().iteroptions.fill_cache = false;
DBContext().syncoptions.sync = true;
DBContext().options = GetOptions(params.cache_bytes);
DBContext().options.create_if_missing = true;
if (params.memory_only) {
DBContext().penv = leveldb::NewMemEnv(leveldb::Env::Default());
DBContext().options.env = DBContext().penv;
} else {
if (params.wipe_data) {
LogPrintf("Wiping LevelDB in %s\n", fs::PathToString(params.path));
leveldb::Status result = leveldb::DestroyDB(fs::PathToString(params.path), DBContext().options);
HandleError(result);
}
TryCreateDirectories(params.path);
LogPrintf("Opening LevelDB in %s\n", fs::PathToString(params.path));
}
// PathToString() return value is safe to pass to leveldb open function,
// because on POSIX leveldb passes the byte string directly to ::open(), and
// on Windows it converts from UTF-8 to UTF-16 before calling ::CreateFileW
// (see env_posix.cc and env_windows.cc).
leveldb::Status status = leveldb::DB::Open(DBContext().options, fs::PathToString(params.path), &DBContext().pdb);
HandleError(status);
LogPrintf("Opened LevelDB successfully\n");
if (params.options.force_compact) {
LogPrintf("Starting database compaction of %s\n", fs::PathToString(params.path));
DBContext().pdb->CompactRange(nullptr, nullptr);
LogPrintf("Finished database compaction of %s\n", fs::PathToString(params.path));
}
// The base-case obfuscation key, which is a noop.
obfuscate_key = std::vector<unsigned char>(OBFUSCATE_KEY_NUM_BYTES, '\000');
bool key_exists = Read(OBFUSCATE_KEY_KEY, obfuscate_key);
if (!key_exists && params.obfuscate && IsEmpty()) {
// Initialize non-degenerate obfuscation if it won't upset
// existing, non-obfuscated data.
std::vector<unsigned char> new_key = CreateObfuscateKey();
// Write `new_key` so we don't obfuscate the key with itself
Write(OBFUSCATE_KEY_KEY, new_key);
obfuscate_key = new_key;
LogPrintf("Wrote new obfuscate key for %s: %s\n", fs::PathToString(params.path), HexStr(obfuscate_key));
}
LogPrintf("Using obfuscation key for %s: %s\n", fs::PathToString(params.path), HexStr(obfuscate_key));
}
CDBWrapper::~CDBWrapper()
{
delete DBContext().pdb;
DBContext().pdb = nullptr;
delete DBContext().options.filter_policy;
DBContext().options.filter_policy = nullptr;
delete DBContext().options.info_log;
DBContext().options.info_log = nullptr;
delete DBContext().options.block_cache;
DBContext().options.block_cache = nullptr;
delete DBContext().penv;
DBContext().options.env = nullptr;
}
bool CDBWrapper::WriteBatch(CDBBatch& batch, bool fSync)
{
const bool log_memory = LogAcceptCategory(BCLog::LEVELDB, BCLog::Level::Debug);
double mem_before = 0;
if (log_memory) {
mem_before = DynamicMemoryUsage() / 1024.0 / 1024;
}
leveldb::Status status = DBContext().pdb->Write(fSync ? DBContext().syncoptions : DBContext().writeoptions, &batch.m_impl_batch->batch);
HandleError(status);
if (log_memory) {
double mem_after = DynamicMemoryUsage() / 1024.0 / 1024;
LogPrint(BCLog::LEVELDB, "WriteBatch memory usage: db=%s, before=%.1fMiB, after=%.1fMiB\n",
m_name, mem_before, mem_after);
}
return true;
}
size_t CDBWrapper::DynamicMemoryUsage() const
{
std::string memory;
std::optional<size_t> parsed;
if (!DBContext().pdb->GetProperty("leveldb.approximate-memory-usage", &memory) || !(parsed = ToIntegral<size_t>(memory))) {
LogPrint(BCLog::LEVELDB, "Failed to get approximate-memory-usage property\n");
return 0;
}
return parsed.value();
}
// Prefixed with null character to avoid collisions with other keys
//
// We must use a string constructor which specifies length so that we copy
// past the null-terminator.
const std::string CDBWrapper::OBFUSCATE_KEY_KEY("\000obfuscate_key", 14);
const unsigned int CDBWrapper::OBFUSCATE_KEY_NUM_BYTES = 8;
/**
* Returns a string (consisting of 8 random bytes) suitable for use as an
* obfuscating XOR key.
*/
std::vector<unsigned char> CDBWrapper::CreateObfuscateKey() const
{
std::vector<uint8_t> ret(OBFUSCATE_KEY_NUM_BYTES);
GetRandBytes(ret);
return ret;
}
std::optional<std::string> CDBWrapper::ReadImpl(Span<const std::byte> key) const
{
leveldb::Slice slKey(CharCast(key.data()), key.size());
std::string strValue;
leveldb::Status status = DBContext().pdb->Get(DBContext().readoptions, slKey, &strValue);
if (!status.ok()) {
if (status.IsNotFound())
return std::nullopt;
LogPrintf("LevelDB read failure: %s\n", status.ToString());
HandleError(status);
}
return strValue;
}
bool CDBWrapper::ExistsImpl(Span<const std::byte> key) const
{
leveldb::Slice slKey(CharCast(key.data()), key.size());
std::string strValue;
leveldb::Status status = DBContext().pdb->Get(DBContext().readoptions, slKey, &strValue);
if (!status.ok()) {
if (status.IsNotFound())
return false;
LogPrintf("LevelDB read failure: %s\n", status.ToString());
HandleError(status);
}
return true;
}
size_t CDBWrapper::EstimateSizeImpl(Span<const std::byte> key1, Span<const std::byte> key2) const
{
leveldb::Slice slKey1(CharCast(key1.data()), key1.size());
leveldb::Slice slKey2(CharCast(key2.data()), key2.size());
uint64_t size = 0;
leveldb::Range range(slKey1, slKey2);
DBContext().pdb->GetApproximateSizes(&range, 1, &size);
return size;
}
bool CDBWrapper::IsEmpty()
{
std::unique_ptr<CDBIterator> it(NewIterator());
it->SeekToFirst();
return !(it->Valid());
}
struct CDBIterator::IteratorImpl {
const std::unique_ptr<leveldb::Iterator> iter;
explicit IteratorImpl(leveldb::Iterator* _iter) : iter{_iter} {}
};
CDBIterator::CDBIterator(const CDBWrapper& _parent, std::unique_ptr<IteratorImpl> _piter) : parent(_parent),
m_impl_iter(std::move(_piter)) {}
CDBIterator* CDBWrapper::NewIterator()
{
return new CDBIterator{*this, std::make_unique<CDBIterator::IteratorImpl>(DBContext().pdb->NewIterator(DBContext().iteroptions))};
}
void CDBIterator::SeekImpl(Span<const std::byte> key)
{
leveldb::Slice slKey(CharCast(key.data()), key.size());
m_impl_iter->iter->Seek(slKey);
}
Span<const std::byte> CDBIterator::GetKeyImpl() const
{
return MakeByteSpan(m_impl_iter->iter->key());
}
Span<const std::byte> CDBIterator::GetValueImpl() const
{
return MakeByteSpan(m_impl_iter->iter->value());
}
CDBIterator::~CDBIterator() = default;
bool CDBIterator::Valid() const { return m_impl_iter->iter->Valid(); }
void CDBIterator::SeekToFirst() { m_impl_iter->iter->SeekToFirst(); }
void CDBIterator::Next() { m_impl_iter->iter->Next(); }
namespace dbwrapper_private {
const std::vector<unsigned char>& GetObfuscateKey(const CDBWrapper &w)
{
return w.obfuscate_key;
}
} // namespace dbwrapper_private