leveldb::DBImpl Class Reference

#include <db_impl.h>

Inheritance diagram for leveldb::DBImpl:

Collaboration diagram for leveldb::DBImpl:

Classes
struct	CompactionState
struct	CompactionStats
struct	ManualCompaction
struct	Writer

Public Member Functions
	DBImpl (const Options &options, const std::string &dbname)
virtual	~DBImpl ()
virtual Status	Put (const WriteOptions &, const Slice &key, const Slice &value)
virtual Status	Delete (const WriteOptions &, const Slice &key)
virtual Status	Write (const WriteOptions &options, WriteBatch *updates)
virtual Status	Get (const ReadOptions &options, const Slice &key, std::string *value)
virtual Iterator *	NewIterator (const ReadOptions &)
virtual const Snapshot *	GetSnapshot ()
virtual void	ReleaseSnapshot (const Snapshot *snapshot)
virtual bool	GetProperty (const Slice &property, std::string *value)
virtual void	GetApproximateSizes (const Range *range, int n, uint64_t *sizes)
virtual void	CompactRange (const Slice *begin, const Slice *end)
void	TEST_CompactRange (int level, const Slice *begin, const Slice *end)
Status	TEST_CompactMemTable ()
Iterator *	TEST_NewInternalIterator ()
int64_t	TEST_MaxNextLevelOverlappingBytes ()
void	RecordReadSample (Slice key)
Public Member Functions inherited from leveldb::DB
	DB ()
virtual	~DB ()

Private Member Functions
Iterator *	NewInternalIterator (const ReadOptions &, SequenceNumber *latest_snapshot, uint32_t *seed)
Status	NewDB ()
Status	Recover (VersionEdit *edit, bool *save_manifest) EXCLUSIVE_LOCKS_REQUIRED(mutex_)
void	MaybeIgnoreError (Status *s) const
void	DeleteObsoleteFiles ()
void	CompactMemTable () EXCLUSIVE_LOCKS_REQUIRED(mutex_)
Status	RecoverLogFile (uint64_t log_number, bool last_log, bool *save_manifest, VersionEdit *edit, SequenceNumber *max_sequence) EXCLUSIVE_LOCKS_REQUIRED(mutex_)
Status	WriteLevel0Table (MemTable *mem, VersionEdit *edit, Version *base) EXCLUSIVE_LOCKS_REQUIRED(mutex_)
Status	MakeRoomForWrite (bool force) EXCLUSIVE_LOCKS_REQUIRED(mutex_)
WriteBatch *	BuildBatchGroup (Writer **last_writer)
void	RecordBackgroundError (const Status &s)
void	MaybeScheduleCompaction () EXCLUSIVE_LOCKS_REQUIRED(mutex_)
void	BackgroundCall ()
void	BackgroundCompaction () EXCLUSIVE_LOCKS_REQUIRED(mutex_)
void	CleanupCompaction (CompactionState *compact) EXCLUSIVE_LOCKS_REQUIRED(mutex_)
Status	DoCompactionWork (CompactionState *compact) EXCLUSIVE_LOCKS_REQUIRED(mutex_)
Status	OpenCompactionOutputFile (CompactionState *compact)
Status	FinishCompactionOutputFile (CompactionState *compact, Iterator *input)
Status	InstallCompactionResults (CompactionState *compact) EXCLUSIVE_LOCKS_REQUIRED(mutex_)
	DBImpl (const DBImpl &)
void	operator= (const DBImpl &)
const Comparator *	user_comparator () const

Static Private Member Functions
static void	BGWork (void *db)

Private Attributes
Env *const	env_
const InternalKeyComparator	internal_comparator_
const InternalFilterPolicy	internal_filter_policy_
const Options	options_
bool	owns_info_log_
bool	owns_cache_
const std::string	dbname_
TableCache *	table_cache_
FileLock *	db_lock_
port::Mutex	mutex_
port::AtomicPointer	shutting_down_
port::CondVar	bg_cv_
MemTable *	mem_
MemTable *	imm_
port::AtomicPointer	has_imm_
WritableFile *	logfile_
uint64_t	logfile_number_
log::Writer *	log_
uint32_t	seed_
std::deque< Writer * >	writers_
WriteBatch *	tmp_batch_
SnapshotList	snapshots_
std::set< uint64_t >	pending_outputs_
bool	bg_compaction_scheduled_
ManualCompaction *	manual_compaction_
VersionSet *	versions_
Status	bg_error_
CompactionStats	stats_ [config::kNumLevels]

Friends
class	DB

Additional Inherited Members
Static Public Member Functions inherited from leveldb::DB
static Status	Open (const Options &options, const std::string &name, DB **dbptr)

Detailed Description

Definition at line 26 of file db_impl.h.

---

Class Documentation

leveldb::DBImpl::ManualCompaction

struct leveldb::DBImpl::ManualCompaction

Definition at line 162 of file db_impl.h.

Collaboration diagram for leveldb::DBImpl::ManualCompaction:

Class Members
const InternalKey *	begin
bool	done
const InternalKey *	end
int	level
InternalKey	tmp_storage

Constructor & Destructor Documentation

DBImpl() [1/2]

leveldb::DBImpl::DBImpl	(	const Options &	options,
		const std::string &	dbname
	)

Definition at line 117 of file db_impl.cc.

    : env_(raw_options.env),
      internal_comparator_(raw_options.comparator),
      internal_filter_policy_(raw_options.filter_policy),
      options_(SanitizeOptions(dbname, &internal_comparator_,
                               &internal_filter_policy_, raw_options)),
      owns_info_log_(options_.info_log != raw_options.info_log),
      owns_cache_(options_.block_cache != raw_options.block_cache),
      dbname_(dbname),
      db_lock_(NULL),
      shutting_down_(NULL),
      bg_cv_(&mutex_),
      mem_(NULL),
      imm_(NULL),
      logfile_(NULL),
      logfile_number_(0),
      log_(NULL),
      seed_(0),
      tmp_batch_(new WriteBatch),
      bg_compaction_scheduled_(false),
      manual_compaction_(NULL) {
  has_imm_.Release_Store(NULL);

  // Reserve ten files or so for other uses and give the rest to TableCache.
  const int table_cache_size = options_.max_open_files - kNumNonTableCacheFiles;
  table_cache_ = new TableCache(dbname_, &options_, table_cache_size);

  versions_ = new VersionSet(dbname_, &options_, table_cache_,
                             &internal_comparator_);
}

Here is the caller graph for this function:

~DBImpl()

leveldb::DBImpl::~DBImpl ( )

virtual

Definition at line 148 of file db_impl.cc.

                {
  // Wait for background work to finish
  mutex_.Lock();
  shutting_down_.Release_Store(this);  // Any non-NULL value is ok
  while (bg_compaction_scheduled_) {
    bg_cv_.Wait();
  }
  mutex_.Unlock();

  if (db_lock_ != NULL) {
    env_->UnlockFile(db_lock_);
  }

  delete versions_;
  if (mem_ != NULL) mem_->Unref();
  if (imm_ != NULL) imm_->Unref();
  delete tmp_batch_;
  delete log_;
  delete logfile_;
  delete table_cache_;

  if (owns_info_log_) {
    delete options_.info_log;
  }
  if (owns_cache_) {
    delete options_.block_cache;
  }
}

Here is the call graph for this function:

DBImpl() [2/2]

leveldb::DBImpl::DBImpl ( const DBImpl &  )

private

Member Function Documentation

BackgroundCall()

void leveldb::DBImpl::BackgroundCall ( )

private

Definition at line 667 of file db_impl.cc.

                            {
  MutexLock l(&mutex_);
  assert(bg_compaction_scheduled_);
  if (shutting_down_.Acquire_Load()) {
    // No more background work when shutting down.
  } else if (!bg_error_.ok()) {
    // No more background work after a background error.
  } else {
    BackgroundCompaction();
  }

  bg_compaction_scheduled_ = false;

  // Previous compaction may have produced too many files in a level,
  // so reschedule another compaction if needed.
  MaybeScheduleCompaction();
  bg_cv_.SignalAll();
}

Here is the call graph for this function:

Here is the caller graph for this function:

BackgroundCompaction()

void leveldb::DBImpl::BackgroundCompaction ( )

private

Definition at line 686 of file db_impl.cc.

                                  {
  mutex_.AssertHeld();

  if (imm_ != NULL) {
    CompactMemTable();
    return;
  }

  Compaction* c;
  bool is_manual = (manual_compaction_ != NULL);
  InternalKey manual_end;
  if (is_manual) {
    ManualCompaction* m = manual_compaction_;
    c = versions_->CompactRange(m->level, m->begin, m->end);
    m->done = (c == NULL);
    if (c != NULL) {
      manual_end = c->input(0, c->num_input_files(0) - 1)->largest;
    }
    Log(options_.info_log,
        "Manual compaction at level-%d from %s .. %s; will stop at %s\n",
        m->level,
        (m->begin ? m->begin->DebugString().c_str() : "(begin)"),
        (m->end ? m->end->DebugString().c_str() : "(end)"),
        (m->done ? "(end)" : manual_end.DebugString().c_str()));
  } else {
    c = versions_->PickCompaction();
  }

  Status status;
  if (c == NULL) {
    // Nothing to do
  } else if (!is_manual && c->IsTrivialMove()) {
    // Move file to next level
    assert(c->num_input_files(0) == 1);
    FileMetaData* f = c->input(0, 0);
    c->edit()->DeleteFile(c->level(), f->number);
    c->edit()->AddFile(c->level() + 1, f->number, f->file_size,
                       f->smallest, f->largest);
    status = versions_->LogAndApply(c->edit(), &mutex_);
    if (!status.ok()) {
      RecordBackgroundError(status);
    }
    VersionSet::LevelSummaryStorage tmp;
    Log(options_.info_log, "Moved #%lld to level-%d %lld bytes %s: %s\n",
        static_cast<unsigned long long>(f->number),
        c->level() + 1,
        static_cast<unsigned long long>(f->file_size),
        status.ToString().c_str(),
        versions_->LevelSummary(&tmp));
  } else {
    CompactionState* compact = new CompactionState(c);
    status = DoCompactionWork(compact);
    if (!status.ok()) {
      RecordBackgroundError(status);
    }
    CleanupCompaction(compact);
    c->ReleaseInputs();
    DeleteObsoleteFiles();
  }
  delete c;

  if (status.ok()) {
    // Done
  } else if (shutting_down_.Acquire_Load()) {
    // Ignore compaction errors found during shutting down
  } else {
    Log(options_.info_log,
        "Compaction error: %s", status.ToString().c_str());
  }

  if (is_manual) {
    ManualCompaction* m = manual_compaction_;
    if (!status.ok()) {
      m->done = true;
    }
    if (!m->done) {
      // We only compacted part of the requested range.  Update *m
      // to the range that is left to be compacted.
      m->tmp_storage = manual_end;
      m->begin = &m->tmp_storage;
    }
    manual_compaction_ = NULL;
  }
}

Here is the call graph for this function:

Here is the caller graph for this function:

BGWork()

void leveldb::DBImpl::BGWork ( void *  db )

staticprivate

Definition at line 663 of file db_impl.cc.

                            {
  reinterpret_cast<DBImpl*>(db)->BackgroundCall();
}

Here is the call graph for this function:

Here is the caller graph for this function:

BuildBatchGroup()

WriteBatch * leveldb::DBImpl::BuildBatchGroup ( Writer **  last_writer )

private

Definition at line 1269 of file db_impl.cc.

                                                        {
  assert(!writers_.empty());
  Writer* first = writers_.front();
  WriteBatch* result = first->batch;
  assert(result != NULL);

  size_t size = WriteBatchInternal::ByteSize(first->batch);

  // Allow the group to grow up to a maximum size, but if the
  // original write is small, limit the growth so we do not slow
  // down the small write too much.
  size_t max_size = 1 << 20;
  if (size <= (128<<10)) {
    max_size = size + (128<<10);
  }

  *last_writer = first;
  std::deque<Writer*>::iterator iter = writers_.begin();
  ++iter;  // Advance past "first"
  for (; iter != writers_.end(); ++iter) {
    Writer* w = *iter;
    if (w->sync && !first->sync) {
      // Do not include a sync write into a batch handled by a non-sync write.
      break;
    }

    if (w->batch != NULL) {
      size += WriteBatchInternal::ByteSize(w->batch);
      if (size > max_size) {
        // Do not make batch too big
        break;
      }

      // Append to *result
      if (result == first->batch) {
        // Switch to temporary batch instead of disturbing caller's batch
        result = tmp_batch_;
        assert(WriteBatchInternal::Count(result) == 0);
        WriteBatchInternal::Append(result, first->batch);
      }
      WriteBatchInternal::Append(result, w->batch);
    }
    *last_writer = w;
  }
  return result;
}

Here is the call graph for this function:

Here is the caller graph for this function:

CleanupCompaction()

void leveldb::DBImpl::CleanupCompaction ( CompactionState *  compact )

private

Definition at line 771 of file db_impl.cc.

                                                       {
  mutex_.AssertHeld();
  if (compact->builder != NULL) {
    // May happen if we get a shutdown call in the middle of compaction
    compact->builder->Abandon();
    delete compact->builder;
  } else {
    assert(compact->outfile == NULL);
  }
  delete compact->outfile;
  for (size_t i = 0; i < compact->outputs.size(); i++) {
    const CompactionState::Output& out = compact->outputs[i];
    pending_outputs_.erase(out.number);
  }
  delete compact;
}

Here is the call graph for this function:

Here is the caller graph for this function:

CompactMemTable()

void leveldb::DBImpl::CompactMemTable ( )

private

Definition at line 534 of file db_impl.cc.

                             {
  mutex_.AssertHeld();
  assert(imm_ != NULL);

  // Save the contents of the memtable as a new Table
  VersionEdit edit;
  Version* base = versions_->current();
  base->Ref();
  Status s = WriteLevel0Table(imm_, &edit, base);
  base->Unref();

  if (s.ok() && shutting_down_.Acquire_Load()) {
    s = Status::IOError("Deleting DB during memtable compaction");
  }

  // Replace immutable memtable with the generated Table
  if (s.ok()) {
    edit.SetPrevLogNumber(0);
    edit.SetLogNumber(logfile_number_);  // Earlier logs no longer needed
    s = versions_->LogAndApply(&edit, &mutex_);
  }

  if (s.ok()) {
    // Commit to the new state
    imm_->Unref();
    imm_ = NULL;
    has_imm_.Release_Store(NULL);
    DeleteObsoleteFiles();
  } else {
    RecordBackgroundError(s);
  }
}

Here is the call graph for this function:

Here is the caller graph for this function:

CompactRange()

void leveldb::DBImpl::CompactRange ( const Slice *  begin, const Slice *  end  )

virtual

Implements leveldb::DB.

Definition at line 567 of file db_impl.cc.

                                                              {
  int max_level_with_files = 1;
  {
    MutexLock l(&mutex_);
    Version* base = versions_->current();
    for (int level = 1; level < config::kNumLevels; level++) {
      if (base->OverlapInLevel(level, begin, end)) {
        max_level_with_files = level;
      }
    }
  }
  TEST_CompactMemTable(); // TODO(sanjay): Skip if memtable does not overlap
  for (int level = 0; level < max_level_with_files; level++) {
    TEST_CompactRange(level, begin, end);
  }
}

Here is the call graph for this function:

Here is the caller graph for this function:

Delete()

Status leveldb::DBImpl::Delete ( const WriteOptions &  options, const Slice &  key  )

virtual

Implements leveldb::DB.

Definition at line 1190 of file db_impl.cc.

                                                                   {
  return DB::Delete(options, key);
}

Here is the call graph for this function:

DeleteObsoleteFiles()

void leveldb::DBImpl::DeleteObsoleteFiles ( )

private

Definition at line 218 of file db_impl.cc.

                                 {
  if (!bg_error_.ok()) {
    // After a background error, we don't know whether a new version may
    // or may not have been committed, so we cannot safely garbage collect.
    return;
  }

  // Make a set of all of the live files
  std::set<uint64_t> live = pending_outputs_;
  versions_->AddLiveFiles(&live);

  std::vector<std::string> filenames;
  env_->GetChildren(dbname_, &filenames); // Ignoring errors on purpose
  uint64_t number;
  FileType type;
  for (size_t i = 0; i < filenames.size(); i++) {
    if (ParseFileName(filenames[i], &number, &type)) {
      bool keep = true;
      switch (type) {
        case kLogFile:
          keep = ((number >= versions_->LogNumber()) ||
                  (number == versions_->PrevLogNumber()));
          break;
        case kDescriptorFile:
          // Keep my manifest file, and any newer incarnations'
          // (in case there is a race that allows other incarnations)
          keep = (number >= versions_->ManifestFileNumber());
          break;
        case kTableFile:
          keep = (live.find(number) != live.end());
          break;
        case kTempFile:
          // Any temp files that are currently being written to must
          // be recorded in pending_outputs_, which is inserted into "live"
          keep = (live.find(number) != live.end());
          break;
        case kCurrentFile:
        case kDBLockFile:
        case kInfoLogFile:
          keep = true;
          break;
      }

      if (!keep) {
        if (type == kTableFile) {
          table_cache_->Evict(number);
        }
        Log(options_.info_log, "Delete type=%d #%lld\n",
            int(type),
            static_cast<unsigned long long>(number));
        env_->DeleteFile(dbname_ + "/" + filenames[i]);
      }
    }
  }
}

Here is the call graph for this function:

Here is the caller graph for this function:

DoCompactionWork()

Status leveldb::DBImpl::DoCompactionWork ( CompactionState *  compact )

private

Definition at line 887 of file db_impl.cc.

                                                        {
  const uint64_t start_micros = env_->NowMicros();
  int64_t imm_micros = 0;  // Micros spent doing imm_ compactions

  Log(options_.info_log,  "Compacting %d@%d + %d@%d files",
      compact->compaction->num_input_files(0),
      compact->compaction->level(),
      compact->compaction->num_input_files(1),
      compact->compaction->level() + 1);

  assert(versions_->NumLevelFiles(compact->compaction->level()) > 0);
  assert(compact->builder == NULL);
  assert(compact->outfile == NULL);
  if (snapshots_.empty()) {
    compact->smallest_snapshot = versions_->LastSequence();
  } else {
    compact->smallest_snapshot = snapshots_.oldest()->number_;
  }

  // Release mutex while we're actually doing the compaction work
  mutex_.Unlock();

  Iterator* input = versions_->MakeInputIterator(compact->compaction);
  input->SeekToFirst();
  Status status;
  ParsedInternalKey ikey;
  std::string current_user_key;
  bool has_current_user_key = false;
  SequenceNumber last_sequence_for_key = kMaxSequenceNumber;
  for (; input->Valid() && !shutting_down_.Acquire_Load(); ) {
    // Prioritize immutable compaction work
    if (has_imm_.NoBarrier_Load() != NULL) {
      const uint64_t imm_start = env_->NowMicros();
      mutex_.Lock();
      if (imm_ != NULL) {
        CompactMemTable();
        bg_cv_.SignalAll();  // Wakeup MakeRoomForWrite() if necessary
      }
      mutex_.Unlock();
      imm_micros += (env_->NowMicros() - imm_start);
    }

    Slice key = input->key();
    if (compact->compaction->ShouldStopBefore(key) &&
        compact->builder != NULL) {
      status = FinishCompactionOutputFile(compact, input);
      if (!status.ok()) {
        break;
      }
    }

    // Handle key/value, add to state, etc.
    bool drop = false;
    if (!ParseInternalKey(key, &ikey)) {
      // Do not hide error keys
      current_user_key.clear();
      has_current_user_key = false;
      last_sequence_for_key = kMaxSequenceNumber;
    } else {
      if (!has_current_user_key ||
          user_comparator()->Compare(ikey.user_key,
                                     Slice(current_user_key)) != 0) {
        // First occurrence of this user key
        current_user_key.assign(ikey.user_key.data(), ikey.user_key.size());
        has_current_user_key = true;
        last_sequence_for_key = kMaxSequenceNumber;
      }

      if (last_sequence_for_key <= compact->smallest_snapshot) {
        // Hidden by an newer entry for same user key
        drop = true;    // (A)
      } else if (ikey.type == kTypeDeletion &&
                 ikey.sequence <= compact->smallest_snapshot &&
                 compact->compaction->IsBaseLevelForKey(ikey.user_key)) {
        // For this user key:
        // (1) there is no data in higher levels
        // (2) data in lower levels will have larger sequence numbers
        // (3) data in layers that are being compacted here and have
        //     smaller sequence numbers will be dropped in the next
        //     few iterations of this loop (by rule (A) above).
        // Therefore this deletion marker is obsolete and can be dropped.
        drop = true;
      }

      last_sequence_for_key = ikey.sequence;
    }
#if 0
    Log(options_.info_log,
        "  Compact: %s, seq %d, type: %d %d, drop: %d, is_base: %d, "
        "%d smallest_snapshot: %d",
        ikey.user_key.ToString().c_str(),
        (int)ikey.sequence, ikey.type, kTypeValue, drop,
        compact->compaction->IsBaseLevelForKey(ikey.user_key),
        (int)last_sequence_for_key, (int)compact->smallest_snapshot);
#endif

    if (!drop) {
      // Open output file if necessary
      if (compact->builder == NULL) {
        status = OpenCompactionOutputFile(compact);
        if (!status.ok()) {
          break;
        }
      }
      if (compact->builder->NumEntries() == 0) {
        compact->current_output()->smallest.DecodeFrom(key);
      }
      compact->current_output()->largest.DecodeFrom(key);
      compact->builder->Add(key, input->value());

      // Close output file if it is big enough
      if (compact->builder->FileSize() >=
          compact->compaction->MaxOutputFileSize()) {
        status = FinishCompactionOutputFile(compact, input);
        if (!status.ok()) {
          break;
        }
      }
    }

    input->Next();
  }

  if (status.ok() && shutting_down_.Acquire_Load()) {
    status = Status::IOError("Deleting DB during compaction");
  }
  if (status.ok() && compact->builder != NULL) {
    status = FinishCompactionOutputFile(compact, input);
  }
  if (status.ok()) {
    status = input->status();
  }
  delete input;
  input = NULL;

  CompactionStats stats;
  stats.micros = env_->NowMicros() - start_micros - imm_micros;
  for (int which = 0; which < 2; which++) {
    for (int i = 0; i < compact->compaction->num_input_files(which); i++) {
      stats.bytes_read += compact->compaction->input(which, i)->file_size;
    }
  }
  for (size_t i = 0; i < compact->outputs.size(); i++) {
    stats.bytes_written += compact->outputs[i].file_size;
  }

  mutex_.Lock();
  stats_[compact->compaction->level() + 1].Add(stats);

  if (status.ok()) {
    status = InstallCompactionResults(compact);
  }
  if (!status.ok()) {
    RecordBackgroundError(status);
  }
  VersionSet::LevelSummaryStorage tmp;
  Log(options_.info_log,
      "compacted to: %s", versions_->LevelSummary(&tmp));
  return status;
}

Here is the call graph for this function:

Here is the caller graph for this function:

FinishCompactionOutputFile()

Status leveldb::DBImpl::FinishCompactionOutputFile ( CompactionState *  compact, Iterator *  input  )

private

Definition at line 813 of file db_impl.cc.

                                                           {
  assert(compact != NULL);
  assert(compact->outfile != NULL);
  assert(compact->builder != NULL);

  const uint64_t output_number = compact->current_output()->number;
  assert(output_number != 0);

  // Check for iterator errors
  Status s = input->status();
  const uint64_t current_entries = compact->builder->NumEntries();
  if (s.ok()) {
    s = compact->builder->Finish();
  } else {
    compact->builder->Abandon();
  }
  const uint64_t current_bytes = compact->builder->FileSize();
  compact->current_output()->file_size = current_bytes;
  compact->total_bytes += current_bytes;
  delete compact->builder;
  compact->builder = NULL;

  // Finish and check for file errors
  if (s.ok()) {
    s = compact->outfile->Sync();
  }
  if (s.ok()) {
    s = compact->outfile->Close();
  }
  delete compact->outfile;
  compact->outfile = NULL;

  if (s.ok() && current_entries > 0) {
    // Verify that the table is usable
    Iterator* iter = table_cache_->NewIterator(ReadOptions(),
                                               output_number,
                                               current_bytes);
    s = iter->status();
    delete iter;
    if (s.ok()) {
      Log(options_.info_log,
          "Generated table #%llu@%d: %lld keys, %lld bytes",
          (unsigned long long) output_number,
          compact->compaction->level(),
          (unsigned long long) current_entries,
          (unsigned long long) current_bytes);
    }
  }
  return s;
}

Here is the call graph for this function:

Here is the caller graph for this function:

Get()

Status leveldb::DBImpl::Get ( const ReadOptions &  options, const Slice &  key, std::string *  value  )

virtual

Implements leveldb::DB.

Definition at line 1109 of file db_impl.cc.

                                     {
  Status s;
  MutexLock l(&mutex_);
  SequenceNumber snapshot;
  if (options.snapshot != NULL) {
    snapshot = reinterpret_cast<const SnapshotImpl*>(options.snapshot)->number_;
  } else {
    snapshot = versions_->LastSequence();
  }

  MemTable* mem = mem_;
  MemTable* imm = imm_;
  Version* current = versions_->current();
  mem->Ref();
  if (imm != NULL) imm->Ref();
  current->Ref();

  bool have_stat_update = false;
  Version::GetStats stats;

  // Unlock while reading from files and memtables
  {
    mutex_.Unlock();
    // First look in the memtable, then in the immutable memtable (if any).
    LookupKey lkey(key, snapshot);
    if (mem->Get(lkey, value, &s)) {
      // Done
    } else if (imm != NULL && imm->Get(lkey, value, &s)) {
      // Done
    } else {
      s = current->Get(options, lkey, value, &stats);
      have_stat_update = true;
    }
    mutex_.Lock();
  }

  if (have_stat_update && current->UpdateStats(stats)) {
    MaybeScheduleCompaction();
  }
  mem->Unref();
  if (imm != NULL) imm->Unref();
  current->Unref();
  return s;
}

Here is the call graph for this function:

GetApproximateSizes()

void leveldb::DBImpl::GetApproximateSizes ( const Range *  range, int  n, uint64_t *  sizes  )

virtual

Implements leveldb::DB.

Definition at line 1448 of file db_impl.cc.

                     {
  // TODO(opt): better implementation
  Version* v;
  {
    MutexLock l(&mutex_);
    versions_->current()->Ref();
    v = versions_->current();
  }

  for (int i = 0; i < n; i++) {
    // Convert user_key into a corresponding internal key.
    InternalKey k1(range[i].start, kMaxSequenceNumber, kValueTypeForSeek);
    InternalKey k2(range[i].limit, kMaxSequenceNumber, kValueTypeForSeek);
    uint64_t start = versions_->ApproximateOffsetOf(v, k1);
    uint64_t limit = versions_->ApproximateOffsetOf(v, k2);
    sizes[i] = (limit >= start ? limit - start : 0);
  }

  {
    MutexLock l(&mutex_);
    v->Unref();
  }
}

Here is the call graph for this function:

GetProperty()

bool leveldb::DBImpl::GetProperty ( const Slice &  property, std::string *  value  )

virtual

Implements leveldb::DB.

Definition at line 1381 of file db_impl.cc.

                                                                {
  value->clear();

  MutexLock l(&mutex_);
  Slice in = property;
  Slice prefix("leveldb.");
  if (!in.starts_with(prefix)) return false;
  in.remove_prefix(prefix.size());

  if (in.starts_with("num-files-at-level")) {
    in.remove_prefix(strlen("num-files-at-level"));
    uint64_t level;
    bool ok = ConsumeDecimalNumber(&in, &level) && in.empty();
    if (!ok || level >= config::kNumLevels) {
      return false;
    } else {
      char buf[100];
      snprintf(buf, sizeof(buf), "%d",
               versions_->NumLevelFiles(static_cast<int>(level)));
      *value = buf;
      return true;
    }
  } else if (in == "stats") {
    char buf[200];
    snprintf(buf, sizeof(buf),
             "                               Compactions\n"
             "Level  Files Size(MB) Time(sec) Read(MB) Write(MB)\n"
             "--------------------------------------------------\n"
             );
    value->append(buf);
    for (int level = 0; level < config::kNumLevels; level++) {
      int files = versions_->NumLevelFiles(level);
      if (stats_[level].micros > 0 || files > 0) {
        snprintf(
            buf, sizeof(buf),
            "%3d %8d %8.0f %9.0f %8.0f %9.0f\n",
            level,
            files,
            versions_->NumLevelBytes(level) / 1048576.0,
            stats_[level].micros / 1e6,
            stats_[level].bytes_read / 1048576.0,
            stats_[level].bytes_written / 1048576.0);
        value->append(buf);
      }
    }
    return true;
  } else if (in == "sstables") {
    *value = versions_->current()->DebugString();
    return true;
  } else if (in == "approximate-memory-usage") {
    size_t total_usage = options_.block_cache->TotalCharge();
    if (mem_) {
      total_usage += mem_->ApproximateMemoryUsage();
    }
    if (imm_) {
      total_usage += imm_->ApproximateMemoryUsage();
    }
    char buf[50];
    snprintf(buf, sizeof(buf), "%llu",
             static_cast<unsigned long long>(total_usage));
    value->append(buf);
    return true;
  }

  return false;
}

Here is the call graph for this function:

GetSnapshot()

const Snapshot * leveldb::DBImpl::GetSnapshot ( )

virtual

Implements leveldb::DB.

Definition at line 1175 of file db_impl.cc.

                                    {
  MutexLock l(&mutex_);
  return snapshots_.New(versions_->LastSequence());
}

Here is the call graph for this function:

InstallCompactionResults()

Status leveldb::DBImpl::InstallCompactionResults ( CompactionState *  compact )

private

Definition at line 866 of file db_impl.cc.

                                                                {
  mutex_.AssertHeld();
  Log(options_.info_log,  "Compacted %d@%d + %d@%d files => %lld bytes",
      compact->compaction->num_input_files(0),
      compact->compaction->level(),
      compact->compaction->num_input_files(1),
      compact->compaction->level() + 1,
      static_cast<long long>(compact->total_bytes));

  // Add compaction outputs
  compact->compaction->AddInputDeletions(compact->compaction->edit());
  const int level = compact->compaction->level();
  for (size_t i = 0; i < compact->outputs.size(); i++) {
    const CompactionState::Output& out = compact->outputs[i];
    compact->compaction->edit()->AddFile(
        level + 1,
        out.number, out.file_size, out.smallest, out.largest);
  }
  return versions_->LogAndApply(compact->compaction->edit(), &mutex_);
}

Here is the call graph for this function:

Here is the caller graph for this function:

MakeRoomForWrite()

Status leveldb::DBImpl::MakeRoomForWrite ( bool  force )

private

Definition at line 1318 of file db_impl.cc.

                                          {
  mutex_.AssertHeld();
  assert(!writers_.empty());
  bool allow_delay = !force;
  Status s;
  while (true) {
    if (!bg_error_.ok()) {
      // Yield previous error
      s = bg_error_;
      break;
    } else if (
        allow_delay &&
        versions_->NumLevelFiles(0) >= config::kL0_SlowdownWritesTrigger) {
      // We are getting close to hitting a hard limit on the number of
      // L0 files.  Rather than delaying a single write by several
      // seconds when we hit the hard limit, start delaying each
      // individual write by 1ms to reduce latency variance.  Also,
      // this delay hands over some CPU to the compaction thread in
      // case it is sharing the same core as the writer.
      mutex_.Unlock();
      env_->SleepForMicroseconds(1000);
      allow_delay = false;  // Do not delay a single write more than once
      mutex_.Lock();
    } else if (!force &&
               (mem_->ApproximateMemoryUsage() <= options_.write_buffer_size)) {
      // There is room in current memtable
      break;
    } else if (imm_ != NULL) {
      // We have filled up the current memtable, but the previous
      // one is still being compacted, so we wait.
      Log(options_.info_log, "Current memtable full; waiting...\n");
      bg_cv_.Wait();
    } else if (versions_->NumLevelFiles(0) >= config::kL0_StopWritesTrigger) {
      // There are too many level-0 files.
      Log(options_.info_log, "Too many L0 files; waiting...\n");
      bg_cv_.Wait();
    } else {
      // Attempt to switch to a new memtable and trigger compaction of old
      assert(versions_->PrevLogNumber() == 0);
      uint64_t new_log_number = versions_->NewFileNumber();
      WritableFile* lfile = NULL;
      s = env_->NewWritableFile(LogFileName(dbname_, new_log_number), &lfile);
      if (!s.ok()) {
        // Avoid chewing through file number space in a tight loop.
        versions_->ReuseFileNumber(new_log_number);
        break;
      }
      delete log_;
      delete logfile_;
      logfile_ = lfile;
      logfile_number_ = new_log_number;
      log_ = new log::Writer(lfile);
      imm_ = mem_;
      has_imm_.Release_Store(imm_);
      mem_ = new MemTable(internal_comparator_);
      mem_->Ref();
      force = false;   // Do not force another compaction if have room
      MaybeScheduleCompaction();
    }
  }
  return s;
}

Here is the call graph for this function:

Here is the caller graph for this function:

MaybeIgnoreError()

void leveldb::DBImpl::MaybeIgnoreError ( Status *  s ) const

private

Definition at line 209 of file db_impl.cc.

                                             {
  if (s->ok() || options_.paranoid_checks) {
    // No change needed
  } else {
    Log(options_.info_log, "Ignoring error %s", s->ToString().c_str());
    *s = Status::OK();
  }
}

Here is the call graph for this function:

Here is the caller graph for this function:

MaybeScheduleCompaction()

void leveldb::DBImpl::MaybeScheduleCompaction ( )

private

Definition at line 645 of file db_impl.cc.

                                     {
  mutex_.AssertHeld();
  if (bg_compaction_scheduled_) {
    // Already scheduled
  } else if (shutting_down_.Acquire_Load()) {
    // DB is being deleted; no more background compactions
  } else if (!bg_error_.ok()) {
    // Already got an error; no more changes
  } else if (imm_ == NULL &&
             manual_compaction_ == NULL &&
             !versions_->NeedsCompaction()) {
    // No work to be done
  } else {
    bg_compaction_scheduled_ = true;
    env_->Schedule(&DBImpl::BGWork, this);
  }
}

Here is the call graph for this function:

Here is the caller graph for this function:

NewDB()

Status leveldb::DBImpl::NewDB ( )

private

Definition at line 177 of file db_impl.cc.

                     {
  VersionEdit new_db;
  new_db.SetComparatorName(user_comparator()->Name());
  new_db.SetLogNumber(0);
  new_db.SetNextFile(2);
  new_db.SetLastSequence(0);

  const std::string manifest = DescriptorFileName(dbname_, 1);
  WritableFile* file;
  Status s = env_->NewWritableFile(manifest, &file);
  if (!s.ok()) {
    return s;
  }
  {
    log::Writer log(file);
    std::string record;
    new_db.EncodeTo(&record);
    s = log.AddRecord(record);
    if (s.ok()) {
      s = file->Close();
    }
  }
  delete file;
  if (s.ok()) {
    // Make "CURRENT" file that points to the new manifest file.
    s = SetCurrentFile(env_, dbname_, 1);
  } else {
    env_->DeleteFile(manifest);
  }
  return s;
}

Here is the call graph for this function:

Here is the caller graph for this function:

NewInternalIterator()

Iterator * leveldb::DBImpl::NewInternalIterator ( const ReadOptions &  options, SequenceNumber *  latest_snapshot, uint32_t *  seed  )

private

Definition at line 1067 of file db_impl.cc.

                                                      {
  IterState* cleanup = new IterState;
  mutex_.Lock();
  *latest_snapshot = versions_->LastSequence();

  // Collect together all needed child iterators
  std::vector<Iterator*> list;
  list.push_back(mem_->NewIterator());
  mem_->Ref();
  if (imm_ != NULL) {
    list.push_back(imm_->NewIterator());
    imm_->Ref();
  }
  versions_->current()->AddIterators(options, &list);
  Iterator* internal_iter =
      NewMergingIterator(&internal_comparator_, &list[0], list.size());
  versions_->current()->Ref();

  cleanup->mu = &mutex_;
  cleanup->mem = mem_;
  cleanup->imm = imm_;
  cleanup->version = versions_->current();
  internal_iter->RegisterCleanup(CleanupIteratorState, cleanup, NULL);

  *seed = ++seed_;
  mutex_.Unlock();
  return internal_iter;
}

Here is the call graph for this function:

Here is the caller graph for this function:

NewIterator()

Iterator * leveldb::DBImpl::NewIterator ( const ReadOptions &  options )

virtual

Implements leveldb::DB.

Definition at line 1156 of file db_impl.cc.

                                                        {
  SequenceNumber latest_snapshot;
  uint32_t seed;
  Iterator* iter = NewInternalIterator(options, &latest_snapshot, &seed);
  return NewDBIterator(
      this, user_comparator(), iter,
      (options.snapshot != NULL
       ? reinterpret_cast<const SnapshotImpl*>(options.snapshot)->number_
       : latest_snapshot),
      seed);
}

Here is the call graph for this function:

OpenCompactionOutputFile()

Status leveldb::DBImpl::OpenCompactionOutputFile ( CompactionState *  compact )

private

Definition at line 788 of file db_impl.cc.

                                                                {
  assert(compact != NULL);
  assert(compact->builder == NULL);
  uint64_t file_number;
  {
    mutex_.Lock();
    file_number = versions_->NewFileNumber();
    pending_outputs_.insert(file_number);
    CompactionState::Output out;
    out.number = file_number;
    out.smallest.Clear();
    out.largest.Clear();
    compact->outputs.push_back(out);
    mutex_.Unlock();
  }

  // Make the output file
  std::string fname = TableFileName(dbname_, file_number);
  Status s = env_->NewWritableFile(fname, &compact->outfile);
  if (s.ok()) {
    compact->builder = new TableBuilder(options_, compact->outfile);
  }
  return s;
}

Here is the call graph for this function:

Here is the caller graph for this function:

operator=()

void leveldb::DBImpl::operator= ( const DBImpl &  )

private

Put()

Status leveldb::DBImpl::Put ( const WriteOptions &  o, const Slice &  key, const Slice &  value  )

virtual

Implements leveldb::DB.

Definition at line 1186 of file db_impl.cc.

                                                                            {
  return DB::Put(o, key, val);
}

Here is the call graph for this function:

RecordBackgroundError()

void leveldb::DBImpl::RecordBackgroundError ( const Status &  s )

private

Definition at line 637 of file db_impl.cc.

                                                  {
  mutex_.AssertHeld();
  if (bg_error_.ok()) {
    bg_error_ = s;
    bg_cv_.SignalAll();
  }
}

Here is the call graph for this function:

Here is the caller graph for this function:

RecordReadSample()

void leveldb::DBImpl::RecordReadSample

(

Slice

key

)

Definition at line 1168 of file db_impl.cc.

                                       {
  MutexLock l(&mutex_);
  if (versions_->current()->RecordReadSample(key)) {
    MaybeScheduleCompaction();
  }
}

Here is the call graph for this function:

Recover()

Status leveldb::DBImpl::Recover ( VersionEdit *  edit, bool *  save_manifest  )

private

Definition at line 274 of file db_impl.cc.

                                                             {
  mutex_.AssertHeld();

  // Ignore error from CreateDir since the creation of the DB is
  // committed only when the descriptor is created, and this directory
  // may already exist from a previous failed creation attempt.
  env_->CreateDir(dbname_);
  assert(db_lock_ == NULL);
  Status s = env_->LockFile(LockFileName(dbname_), &db_lock_);
  if (!s.ok()) {
    return s;
  }

  if (!env_->FileExists(CurrentFileName(dbname_))) {
    if (options_.create_if_missing) {
      s = NewDB();
      if (!s.ok()) {
        return s;
      }
    } else {
      return Status::InvalidArgument(
          dbname_, "does not exist (create_if_missing is false)");
    }
  } else {
    if (options_.error_if_exists) {
      return Status::InvalidArgument(
          dbname_, "exists (error_if_exists is true)");
    }
  }

  s = versions_->Recover(save_manifest);
  if (!s.ok()) {
    return s;
  }
  SequenceNumber max_sequence(0);

  // Recover from all newer log files than the ones named in the
  // descriptor (new log files may have been added by the previous
  // incarnation without registering them in the descriptor).
  //
  // Note that PrevLogNumber() is no longer used, but we pay
  // attention to it in case we are recovering a database
  // produced by an older version of leveldb.
  const uint64_t min_log = versions_->LogNumber();
  const uint64_t prev_log = versions_->PrevLogNumber();
  std::vector<std::string> filenames;
  s = env_->GetChildren(dbname_, &filenames);
  if (!s.ok()) {
    return s;
  }
  std::set<uint64_t> expected;
  versions_->AddLiveFiles(&expected);
  uint64_t number;
  FileType type;
  std::vector<uint64_t> logs;
  for (size_t i = 0; i < filenames.size(); i++) {
    if (ParseFileName(filenames[i], &number, &type)) {
      expected.erase(number);
      if (type == kLogFile && ((number >= min_log) || (number == prev_log)))
        logs.push_back(number);
    }
  }
  if (!expected.empty()) {
    char buf[50];
    snprintf(buf, sizeof(buf), "%d missing files; e.g.",
             static_cast<int>(expected.size()));
    return Status::Corruption(buf, TableFileName(dbname_, *(expected.begin())));
  }

  // Recover in the order in which the logs were generated
  std::sort(logs.begin(), logs.end());
  for (size_t i = 0; i < logs.size(); i++) {
    s = RecoverLogFile(logs[i], (i == logs.size() - 1), save_manifest, edit,
                       &max_sequence);
    if (!s.ok()) {
      return s;
    }

    // The previous incarnation may not have written any MANIFEST
    // records after allocating this log number.  So we manually
    // update the file number allocation counter in VersionSet.
    versions_->MarkFileNumberUsed(logs[i]);
  }

  if (versions_->LastSequence() < max_sequence) {
    versions_->SetLastSequence(max_sequence);
  }

  return Status::OK();
}

Here is the call graph for this function:

Here is the caller graph for this function:

RecoverLogFile()

Status leveldb::DBImpl::RecoverLogFile ( uint64_t  log_number, bool  last_log, bool *  save_manifest, VersionEdit *  edit, SequenceNumber *  max_sequence  )

private

Definition at line 365 of file db_impl.cc.

                                                            {
  struct LogReporter : public log::Reader::Reporter {
    Env* env;
    Logger* info_log;
    const char* fname;
    Status* status;  // NULL if options_.paranoid_checks==false
    virtual void Corruption(size_t bytes, const Status& s) {
      Log(info_log, "%s%s: dropping %d bytes; %s",
          (this->status == NULL ? "(ignoring error) " : ""),
          fname, static_cast<int>(bytes), s.ToString().c_str());
      if (this->status != NULL && this->status->ok()) *this->status = s;
    }
  };

  mutex_.AssertHeld();

  // Open the log file
  std::string fname = LogFileName(dbname_, log_number);
  SequentialFile* file;
  Status status = env_->NewSequentialFile(fname, &file);
  if (!status.ok()) {
    MaybeIgnoreError(&status);
    return status;
  }

  // Create the log reader.
  LogReporter reporter;
  reporter.env = env_;
  reporter.info_log = options_.info_log;
  reporter.fname = fname.c_str();
  reporter.status = (options_.paranoid_checks ? &status : NULL);
  // We intentionally make log::Reader do checksumming even if
  // paranoid_checks==false so that corruptions cause entire commits
  // to be skipped instead of propagating bad information (like overly
  // large sequence numbers).
  log::Reader reader(file, &reporter, true/*checksum*/,
                     0/*initial_offset*/);
  Log(options_.info_log, "Recovering log #%llu",
      (unsigned long long) log_number);

  // Read all the records and add to a memtable
  std::string scratch;
  Slice record;
  WriteBatch batch;
  int compactions = 0;
  MemTable* mem = NULL;
  while (reader.ReadRecord(&record, &scratch) &&
         status.ok()) {
    if (record.size() < 12) {
      reporter.Corruption(
          record.size(), Status::Corruption("log record too small"));
      continue;
    }
    WriteBatchInternal::SetContents(&batch, record);

    if (mem == NULL) {
      mem = new MemTable(internal_comparator_);
      mem->Ref();
    }
    status = WriteBatchInternal::InsertInto(&batch, mem);
    MaybeIgnoreError(&status);
    if (!status.ok()) {
      break;
    }
    const SequenceNumber last_seq =
        WriteBatchInternal::Sequence(&batch) +
        WriteBatchInternal::Count(&batch) - 1;
    if (last_seq > *max_sequence) {
      *max_sequence = last_seq;
    }

    if (mem->ApproximateMemoryUsage() > options_.write_buffer_size) {
      compactions++;
      *save_manifest = true;
      status = WriteLevel0Table(mem, edit, NULL);
      mem->Unref();
      mem = NULL;
      if (!status.ok()) {
        // Reflect errors immediately so that conditions like full
        // file-systems cause the DB::Open() to fail.
        break;
      }
    }
  }

  delete file;

  // See if we should keep reusing the last log file.
  if (status.ok() && options_.reuse_logs && last_log && compactions == 0) {
    assert(logfile_ == NULL);
    assert(log_ == NULL);
    assert(mem_ == NULL);
    uint64_t lfile_size;
    if (env_->GetFileSize(fname, &lfile_size).ok() &&
        env_->NewAppendableFile(fname, &logfile_).ok()) {
      Log(options_.info_log, "Reusing old log %s \n", fname.c_str());
      log_ = new log::Writer(logfile_, lfile_size);
      logfile_number_ = log_number;
      if (mem != NULL) {
        mem_ = mem;
        mem = NULL;
      } else {
        // mem can be NULL if lognum exists but was empty.
        mem_ = new MemTable(internal_comparator_);
        mem_->Ref();
      }
    }
  }

  if (mem != NULL) {
    // mem did not get reused; compact it.
    if (status.ok()) {
      *save_manifest = true;
      status = WriteLevel0Table(mem, edit, NULL);
    }
    mem->Unref();
  }

  return status;
}

Here is the call graph for this function:

Here is the caller graph for this function:

ReleaseSnapshot()

void leveldb::DBImpl::ReleaseSnapshot ( const Snapshot *  snapshot )

virtual

Implements leveldb::DB.

Definition at line 1180 of file db_impl.cc.

                                              {
  MutexLock l(&mutex_);
  snapshots_.Delete(reinterpret_cast<const SnapshotImpl*>(s));
}

Here is the call graph for this function:

TEST_CompactMemTable()

Status leveldb::DBImpl::TEST_CompactMemTable

(

)

Definition at line 621 of file db_impl.cc.

                                    {
  // NULL batch means just wait for earlier writes to be done
  Status s = Write(WriteOptions(), NULL);
  if (s.ok()) {
    // Wait until the compaction completes
    MutexLock l(&mutex_);
    while (imm_ != NULL && bg_error_.ok()) {
      bg_cv_.Wait();
    }
    if (imm_ != NULL) {
      s = bg_error_;
    }
  }
  return s;
}

Here is the call graph for this function:

Here is the caller graph for this function:

TEST_CompactRange()

void leveldb::DBImpl::TEST_CompactRange	(	int	level,
		const Slice *	begin,
		const Slice *	end
	)

Definition at line 584 of file db_impl.cc.

                                                                             {
  assert(level >= 0);
  assert(level + 1 < config::kNumLevels);

  InternalKey begin_storage, end_storage;

  ManualCompaction manual;
  manual.level = level;
  manual.done = false;
  if (begin == NULL) {
    manual.begin = NULL;
  } else {
    begin_storage = InternalKey(*begin, kMaxSequenceNumber, kValueTypeForSeek);
    manual.begin = &begin_storage;
  }
  if (end == NULL) {
    manual.end = NULL;
  } else {
    end_storage = InternalKey(*end, 0, static_cast<ValueType>(0));
    manual.end = &end_storage;
  }

  MutexLock l(&mutex_);
  while (!manual.done && !shutting_down_.Acquire_Load() && bg_error_.ok()) {
    if (manual_compaction_ == NULL) {  // Idle
      manual_compaction_ = &manual;
      MaybeScheduleCompaction();
    } else {  // Running either my compaction or another compaction.
      bg_cv_.Wait();
    }
  }
  if (manual_compaction_ == &manual) {
    // Cancel my manual compaction since we aborted early for some reason.
    manual_compaction_ = NULL;
  }
}

Here is the call graph for this function:

Here is the caller graph for this function:

TEST_MaxNextLevelOverlappingBytes()

int64_t leveldb::DBImpl::TEST_MaxNextLevelOverlappingBytes

(

)

Definition at line 1104 of file db_impl.cc.

                                                  {
  MutexLock l(&mutex_);
  return versions_->MaxNextLevelOverlappingBytes();
}

Here is the call graph for this function:

TEST_NewInternalIterator()

Iterator * leveldb::DBImpl::TEST_NewInternalIterator

(

)

Definition at line 1098 of file db_impl.cc.

                                           {
  SequenceNumber ignored;
  uint32_t ignored_seed;
  return NewInternalIterator(ReadOptions(), &ignored, &ignored_seed);
}

Here is the call graph for this function:

user_comparator()

const Comparator* leveldb::DBImpl::user_comparator ( ) const

inlineprivate

Definition at line 197 of file db_impl.h.

                                            {
    return internal_comparator_.user_comparator();
  }

Here is the call graph for this function:

Here is the caller graph for this function:

Write()

Status leveldb::DBImpl::Write ( const WriteOptions &  options, WriteBatch *  updates  )

virtual

Implements leveldb::DB.

Definition at line 1194 of file db_impl.cc.

                                                                      {
  Writer w(&mutex_);
  w.batch = my_batch;
  w.sync = options.sync;
  w.done = false;

  MutexLock l(&mutex_);
  writers_.push_back(&w);
  while (!w.done && &w != writers_.front()) {
    w.cv.Wait();
  }
  if (w.done) {
    return w.status;
  }

  // May temporarily unlock and wait.
  Status status = MakeRoomForWrite(my_batch == NULL);
  uint64_t last_sequence = versions_->LastSequence();
  Writer* last_writer = &w;
  if (status.ok() && my_batch != NULL) {  // NULL batch is for compactions
    WriteBatch* updates = BuildBatchGroup(&last_writer);
    WriteBatchInternal::SetSequence(updates, last_sequence + 1);
    last_sequence += WriteBatchInternal::Count(updates);

    // Add to log and apply to memtable.  We can release the lock
    // during this phase since &w is currently responsible for logging
    // and protects against concurrent loggers and concurrent writes
    // into mem_.
    {
      mutex_.Unlock();
      status = log_->AddRecord(WriteBatchInternal::Contents(updates));
      bool sync_error = false;
      if (status.ok() && options.sync) {
        status = logfile_->Sync();
        if (!status.ok()) {
          sync_error = true;
        }
      }
      if (status.ok()) {
        status = WriteBatchInternal::InsertInto(updates, mem_);
      }
      mutex_.Lock();
      if (sync_error) {
        // The state of the log file is indeterminate: the log record we
        // just added may or may not show up when the DB is re-opened.
        // So we force the DB into a mode where all future writes fail.
        RecordBackgroundError(status);
      }
    }
    if (updates == tmp_batch_) tmp_batch_->Clear();

    versions_->SetLastSequence(last_sequence);
  }

  while (true) {
    Writer* ready = writers_.front();
    writers_.pop_front();
    if (ready != &w) {
      ready->status = status;
      ready->done = true;
      ready->cv.Signal();
    }
    if (ready == last_writer) break;
  }

  // Notify new head of write queue
  if (!writers_.empty()) {
    writers_.front()->cv.Signal();
  }

  return status;
}

Here is the call graph for this function:

Here is the caller graph for this function:

WriteLevel0Table()

Status leveldb::DBImpl::WriteLevel0Table ( MemTable *  mem, VersionEdit *  edit, Version *  base  )

private

Definition at line 488 of file db_impl.cc.

                                               {
  mutex_.AssertHeld();
  const uint64_t start_micros = env_->NowMicros();
  FileMetaData meta;
  meta.number = versions_->NewFileNumber();
  pending_outputs_.insert(meta.number);
  Iterator* iter = mem->NewIterator();
  Log(options_.info_log, "Level-0 table #%llu: started",
      (unsigned long long) meta.number);

  Status s;
  {
    mutex_.Unlock();
    s = BuildTable(dbname_, env_, options_, table_cache_, iter, &meta);
    mutex_.Lock();
  }

  Log(options_.info_log, "Level-0 table #%llu: %lld bytes %s",
      (unsigned long long) meta.number,
      (unsigned long long) meta.file_size,
      s.ToString().c_str());
  delete iter;
  pending_outputs_.erase(meta.number);


  // Note that if file_size is zero, the file has been deleted and
  // should not be added to the manifest.
  int level = 0;
  if (s.ok() && meta.file_size > 0) {
    const Slice min_user_key = meta.smallest.user_key();
    const Slice max_user_key = meta.largest.user_key();
    if (base != NULL) {
      level = base->PickLevelForMemTableOutput(min_user_key, max_user_key);
    }
    edit->AddFile(level, meta.number, meta.file_size,
                  meta.smallest, meta.largest);
  }

  CompactionStats stats;
  stats.micros = env_->NowMicros() - start_micros;
  stats.bytes_written = meta.file_size;
  stats_[level].Add(stats);
  return s;
}

Here is the call graph for this function:

Here is the caller graph for this function:

Friends And Related Function Documentation

DB

friend class DB

friend

Definition at line 68 of file db_impl.h.

Member Data Documentation

bg_compaction_scheduled_

bool leveldb::DBImpl::bg_compaction_scheduled_

private

Definition at line 159 of file db_impl.h.

bg_cv_

port::CondVar leveldb::DBImpl::bg_cv_

private

Definition at line 139 of file db_impl.h.

bg_error_

Status leveldb::DBImpl::bg_error_

private

Definition at line 174 of file db_impl.h.

db_lock_

FileLock* leveldb::DBImpl::db_lock_

private

Definition at line 134 of file db_impl.h.

dbname_

const std::string leveldb::DBImpl::dbname_

private

Definition at line 128 of file db_impl.h.

env_

Env* const leveldb::DBImpl::env_

private

Definition at line 122 of file db_impl.h.

has_imm_

port::AtomicPointer leveldb::DBImpl::has_imm_

private

Definition at line 142 of file db_impl.h.

imm_

MemTable* leveldb::DBImpl::imm_

private

Definition at line 141 of file db_impl.h.

internal_comparator_

const InternalKeyComparator leveldb::DBImpl::internal_comparator_

private

Definition at line 123 of file db_impl.h.

internal_filter_policy_

const InternalFilterPolicy leveldb::DBImpl::internal_filter_policy_

private

Definition at line 124 of file db_impl.h.

log_

log::Writer* leveldb::DBImpl::log_

private

Definition at line 145 of file db_impl.h.

logfile_

WritableFile* leveldb::DBImpl::logfile_

private

Definition at line 143 of file db_impl.h.

logfile_number_

uint64_t leveldb::DBImpl::logfile_number_

private

Definition at line 144 of file db_impl.h.

manual_compaction_

ManualCompaction* leveldb::DBImpl::manual_compaction_

private

Definition at line 169 of file db_impl.h.

mem_

MemTable* leveldb::DBImpl::mem_

private

Definition at line 140 of file db_impl.h.

mutex_

port::Mutex leveldb::DBImpl::mutex_

private

Definition at line 137 of file db_impl.h.

options_

const Options leveldb::DBImpl::options_

private

Definition at line 125 of file db_impl.h.

owns_cache_

bool leveldb::DBImpl::owns_cache_

private

Definition at line 127 of file db_impl.h.

owns_info_log_

bool leveldb::DBImpl::owns_info_log_

private

Definition at line 126 of file db_impl.h.

pending_outputs_

std::set<uint64_t> leveldb::DBImpl::pending_outputs_

private

Definition at line 156 of file db_impl.h.

seed_

uint32_t leveldb::DBImpl::seed_

private

Definition at line 146 of file db_impl.h.

shutting_down_

port::AtomicPointer leveldb::DBImpl::shutting_down_

private

Definition at line 138 of file db_impl.h.

snapshots_

SnapshotList leveldb::DBImpl::snapshots_

private

Definition at line 152 of file db_impl.h.

stats_

CompactionStats leveldb::DBImpl::stats_[config::kNumLevels]

private

Definition at line 191 of file db_impl.h.

table_cache_

TableCache* leveldb::DBImpl::table_cache_

private

Definition at line 131 of file db_impl.h.

tmp_batch_

WriteBatch* leveldb::DBImpl::tmp_batch_

private

Definition at line 150 of file db_impl.h.

versions_

VersionSet* leveldb::DBImpl::versions_

private

Definition at line 171 of file db_impl.h.

writers_

std::deque<Writer*> leveldb::DBImpl::writers_

private

Definition at line 149 of file db_impl.h.

---

The documentation for this class was generated from the following files:

• db/db_impl.h
• db/db_impl.cc