@nxtedition/rocksdb 8.0.0 → 8.0.2

package/deps/rocksdb/rocksdb/db/db_impl/db_impl.cc

@@ -1059,16 +1059,31 @@ void DBImpl::DumpStats() {
     return;
   }
 
+  // Also probe block cache(s) for problems, dump to info log
+  UnorderedSet<Cache*> probed_caches;
   TEST_SYNC_POINT("DBImpl::DumpStats:StartRunning");
   {
     InstrumentedMutexLock l(&mutex_);
     for (auto cfd : versions_->GetRefedColumnFamilySet()) {
-      if (cfd->initialized()) {
-        // Release DB mutex for gathering cache entry stats. Pass over all
-        // column families for this first so that other stats are dumped
-        // near-atomically.
-        InstrumentedMutexUnlock u(&mutex_);
-        cfd->internal_stats()->CollectCacheEntryStats(/*foreground=*/false);
+      if (!cfd->initialized()) {
+        continue;
+      }
+
+      // Release DB mutex for gathering cache entry stats. Pass over all
+      // column families for this first so that other stats are dumped
+      // near-atomically.
+      InstrumentedMutexUnlock u(&mutex_);
+      cfd->internal_stats()->CollectCacheEntryStats(/*foreground=*/false);
+
+      // Probe block cache for problems (if not already via another CF)
+      if (immutable_db_options_.info_log) {
+        auto* table_factory = cfd->ioptions()->table_factory.get();
+        assert(table_factory != nullptr);
+        Cache* cache =
+            table_factory->GetOptions<Cache>(TableFactory::kBlockCacheOpts());
+        if (cache && probed_caches.insert(cache).second) {
+          cache->ReportProblems(immutable_db_options_.info_log);
+        }
       }
     }
 
@@ -1555,21 +1570,31 @@ Status DBImpl::ApplyWALToManifest(VersionEdit* synced_wals) {
 }
 
 Status DBImpl::LockWAL() {
-  log_write_mutex_.Lock();
-  auto cur_log_writer = logs_.back().writer;
-  IOStatus status = cur_log_writer->WriteBuffer();
-  if (!status.ok()) {
-    ROCKS_LOG_ERROR(immutable_db_options_.info_log, "WAL flush error %s",
-                    status.ToString().c_str());
-    // In case there is a fs error we should set it globally to prevent the
-    // future writes
-    WriteStatusCheck(status);
+  {
+    InstrumentedMutexLock lock(&mutex_);
+    WriteThread::Writer w;
+    write_thread_.EnterUnbatched(&w, &mutex_);
+    WriteThread::Writer nonmem_w;
+    if (two_write_queues_) {
+      nonmem_write_thread_.EnterUnbatched(&nonmem_w, &mutex_);
+    }
+
+    lock_wal_write_token_ = write_controller_.GetStopToken();
+
+    if (two_write_queues_) {
+      nonmem_write_thread_.ExitUnbatched(&nonmem_w);
+    }
+    write_thread_.ExitUnbatched(&w);
   }
-  return static_cast<Status>(status);
+  return FlushWAL(/*sync=*/false);
 }
 
 Status DBImpl::UnlockWAL() {
-  log_write_mutex_.Unlock();
+  {
+    InstrumentedMutexLock lock(&mutex_);
+    lock_wal_write_token_.reset();
+  }
+  bg_cv_.SignalAll();
   return Status::OK();
 }
 
@@ -1808,7 +1833,8 @@ InternalIterator* DBImpl::NewInternalIterator(
   MergeIteratorBuilder merge_iter_builder(
       &cfd->internal_comparator(), arena,
       !read_options.total_order_seek &&
-          super_version->mutable_cf_options.prefix_extractor != nullptr);
+          super_version->mutable_cf_options.prefix_extractor != nullptr,
+      read_options.iterate_upper_bound);
   // Collect iterator for mutable memtable
   auto mem_iter = super_version->mem->NewIterator(read_options, arena);
   Status s;
@@ -5302,6 +5328,7 @@ Status DBImpl::IngestExternalFiles(
     // Run ingestion jobs.
     if (status.ok()) {
       for (size_t i = 0; i != num_cfs; ++i) {
+        mutex_.AssertHeld();
         status = ingestion_jobs[i].Run();
         if (!status.ok()) {
           break;
@@ -5506,6 +5533,7 @@ Status DBImpl::CreateColumnFamilyWithImport(
 
       num_running_ingest_file_++;
       assert(!cfd->IsDropped());
+      mutex_.AssertHeld();
       status = import_job.Run();
 
       // Install job edit [Mutex will be unlocked here]

package/deps/rocksdb/rocksdb/db/db_impl/db_impl.h

@@ -1161,7 +1161,7 @@ class DBImpl : public DB {
   int TEST_BGCompactionsAllowed() const;
   int TEST_BGFlushesAllowed() const;
   size_t TEST_GetWalPreallocateBlockSize(uint64_t write_buffer_size) const;
-  void TEST_WaitForPeridicTaskRun(std::function<void()> callback) const;
+  void TEST_WaitForPeriodicTaskRun(std::function<void()> callback) const;
   SeqnoToTimeMapping TEST_GetSeqnoToTimeMapping() const;
   size_t TEST_EstimateInMemoryStatsHistorySize() const;
 
@@ -2680,6 +2680,10 @@ class DBImpl : public DB {
   // seqno_time_mapping_ stores the sequence number to time mapping, it's not
   // thread safe, both read and write need db mutex hold.
   SeqnoToTimeMapping seqno_time_mapping_;
+
+  // stop write token that is acquired when LockWal() is called. Destructed
+  // when UnlockWal() is called.
+  std::unique_ptr<WriteControllerToken> lock_wal_write_token_;
 };
 
 class GetWithTimestampReadCallback : public ReadCallback {

package/deps/rocksdb/rocksdb/db/db_impl/db_impl_compaction_flush.cc

@@ -1344,18 +1344,8 @@ Status DBImpl::CompactFilesImpl(
     }
   }
 
-  SequenceNumber earliest_mem_seqno = kMaxSequenceNumber;
-  if (cfd->mem() != nullptr) {
-    earliest_mem_seqno =
-        std::min(cfd->mem()->GetEarliestSequenceNumber(), earliest_mem_seqno);
-  }
-  if (cfd->imm() != nullptr && cfd->imm()->current() != nullptr) {
-    earliest_mem_seqno =
-        std::min(cfd->imm()->current()->GetEarliestSequenceNumber(false),
-                 earliest_mem_seqno);
-  }
   Status s = cfd->compaction_picker()->SanitizeCompactionInputFiles(
-      &input_set, cf_meta, output_level, earliest_mem_seqno);
+      &input_set, cf_meta, output_level);
   if (!s.ok()) {
     return s;
   }
@@ -1697,14 +1687,15 @@ Status DBImpl::ReFitLevel(ColumnFamilyData* cfd, int level, int target_level) {
 
     VersionEdit edit;
     edit.SetColumnFamily(cfd->GetID());
+
     for (const auto& f : vstorage->LevelFiles(level)) {
       edit.DeleteFile(level, f->fd.GetNumber());
       edit.AddFile(
           to_level, f->fd.GetNumber(), f->fd.GetPathId(), f->fd.GetFileSize(),
           f->smallest, f->largest, f->fd.smallest_seqno, f->fd.largest_seqno,
           f->marked_for_compaction, f->temperature, f->oldest_blob_file_number,
-          f->oldest_ancester_time, f->file_creation_time, f->file_checksum,
-          f->file_checksum_func_name, f->unique_id);
+          f->oldest_ancester_time, f->file_creation_time, f->epoch_number,
+          f->file_checksum, f->file_checksum_func_name, f->unique_id);
     }
     ROCKS_LOG_DEBUG(immutable_db_options_.info_log,
                     "[%s] Apply version edit:\n%s", cfd->GetName().c_str(),
@@ -3344,8 +3335,8 @@ Status DBImpl::BackgroundCompaction(bool* made_progress,
             f->fd.GetFileSize(), f->smallest, f->largest, f->fd.smallest_seqno,
             f->fd.largest_seqno, f->marked_for_compaction, f->temperature,
             f->oldest_blob_file_number, f->oldest_ancester_time,
-            f->file_creation_time, f->file_checksum, f->file_checksum_func_name,
-            f->unique_id);
+            f->file_creation_time, f->epoch_number, f->file_checksum,
+            f->file_checksum_func_name, f->unique_id);
 
         ROCKS_LOG_BUFFER(
             log_buffer,

package/deps/rocksdb/rocksdb/db/db_impl/db_impl_debug.cc

@@ -290,7 +290,7 @@ size_t DBImpl::TEST_GetWalPreallocateBlockSize(
 }
 
 #ifndef ROCKSDB_LITE
-void DBImpl::TEST_WaitForPeridicTaskRun(std::function<void()> callback) const {
+void DBImpl::TEST_WaitForPeriodicTaskRun(std::function<void()> callback) const {
   periodic_task_scheduler_.TEST_WaitForRun(callback);
 }
 

package/deps/rocksdb/rocksdb/db/db_impl/db_impl_experimental.cc

@@ -136,7 +136,7 @@ Status DBImpl::PromoteL0(ColumnFamilyHandle* column_family, int target_level) {
                    f->fd.smallest_seqno, f->fd.largest_seqno,
                    f->marked_for_compaction, f->temperature,
                    f->oldest_blob_file_number, f->oldest_ancester_time,
-                   f->file_creation_time, f->file_checksum,
+                   f->file_creation_time, f->epoch_number, f->file_checksum,
                    f->file_checksum_func_name, f->unique_id);
     }
 

package/deps/rocksdb/rocksdb/db/db_impl/db_impl_files.cc

@@ -315,6 +315,7 @@ void DBImpl::FindObsoleteFiles(JobContext* job_context, bool force,
     }
     log_write_mutex_.Unlock();
     mutex_.Unlock();
+    TEST_SYNC_POINT_CALLBACK("FindObsoleteFiles::PostMutexUnlock", nullptr);
     log_write_mutex_.Lock();
     while (!logs_.empty() && logs_.front().number < min_log_number) {
       auto& log = logs_.front();
@@ -360,6 +361,8 @@ void DBImpl::DeleteObsoleteFileImpl(int job_id, const std::string& fname,
   }
   TEST_SYNC_POINT_CALLBACK("DBImpl::DeleteObsoleteFileImpl:AfterDeletion",
                            &file_deletion_status);
+  TEST_SYNC_POINT_CALLBACK("DBImpl::DeleteObsoleteFileImpl:AfterDeletion2",
+                           const_cast<std::string*>(&fname));
   if (file_deletion_status.ok()) {
     ROCKS_LOG_DEBUG(immutable_db_options_.info_log,
                     "[JOB %d] Delete %s type=%d #%" PRIu64 " -- %s\n", job_id,

package/deps/rocksdb/rocksdb/db/db_impl/db_impl_open.cc

@@ -1515,7 +1515,7 @@ Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
         .PermitUncheckedError();  // ignore error
     const uint64_t current_time = static_cast<uint64_t>(_current_time);
     meta.oldest_ancester_time = current_time;
-
+    meta.epoch_number = cfd->NewEpochNumber();
     {
       auto write_hint = cfd->CalculateSSTWriteHint(0);
       mutex_.Unlock();
@@ -1583,13 +1583,13 @@ Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
   constexpr int level = 0;
 
   if (s.ok() && has_output) {
-    edit->AddFile(level, meta.fd.GetNumber(), meta.fd.GetPathId(),
-                  meta.fd.GetFileSize(), meta.smallest, meta.largest,
-                  meta.fd.smallest_seqno, meta.fd.largest_seqno,
-                  meta.marked_for_compaction, meta.temperature,
-                  meta.oldest_blob_file_number, meta.oldest_ancester_time,
-                  meta.file_creation_time, meta.file_checksum,
-                  meta.file_checksum_func_name, meta.unique_id);
+    edit->AddFile(
+        level, meta.fd.GetNumber(), meta.fd.GetPathId(), meta.fd.GetFileSize(),
+        meta.smallest, meta.largest, meta.fd.smallest_seqno,
+        meta.fd.largest_seqno, meta.marked_for_compaction, meta.temperature,
+        meta.oldest_blob_file_number, meta.oldest_ancester_time,
+        meta.file_creation_time, meta.epoch_number, meta.file_checksum,
+        meta.file_checksum_func_name, meta.unique_id);
 
     for (const auto& blob : blob_file_additions) {
       edit->AddBlobFile(blob);

package/deps/rocksdb/rocksdb/db/db_impl/db_impl_write.cc

@@ -924,6 +924,15 @@ Status DBImpl::WriteImplWALOnly(
       write_thread->ExitAsBatchGroupLeader(write_group, status);
       return status;
     }
+  } else {
+    InstrumentedMutexLock lock(&mutex_);
+    Status status = DelayWrite(/*num_bytes=*/0ull, write_options);
+    if (!status.ok()) {
+      WriteThread::WriteGroup write_group;
+      write_thread->EnterAsBatchGroupLeader(&w, &write_group);
+      write_thread->ExitAsBatchGroupLeader(write_group, status);
+      return status;
+    }
   }
 
   WriteThread::WriteGroup write_group;
@@ -1762,6 +1771,7 @@ uint64_t DBImpl::GetMaxTotalWalSize() const {
 // REQUIRES: this thread is currently at the front of the writer queue
 Status DBImpl::DelayWrite(uint64_t num_bytes,
                           const WriteOptions& write_options) {
+  mutex_.AssertHeld();
   uint64_t time_delayed = 0;
   bool delayed = false;
   {

package/deps/rocksdb/rocksdb/db/db_iter.cc

@@ -158,6 +158,8 @@ void DBIter::Next() {
 
   local_stats_.next_count_++;
   if (ok && iter_.Valid()) {
+    ClearSavedValue();
+
     if (prefix_same_as_start_) {
       assert(prefix_extractor_ != nullptr);
       const Slice prefix = prefix_.GetUserKey();
@@ -544,8 +546,7 @@ bool DBIter::MergeValuesNewToOld() {
       // hit a put, merge the put value with operands and store the
       // final result in saved_value_. We are done!
       const Slice val = iter_.value();
-      Status s = Merge(&val, ikey.user_key);
-      if (!s.ok()) {
+      if (!Merge(&val, ikey.user_key)) {
         return false;
       }
       // iter_ is positioned after put
@@ -574,8 +575,7 @@ bool DBIter::MergeValuesNewToOld() {
         return false;
       }
       valid_ = true;
-      Status s = Merge(&blob_value_, ikey.user_key);
-      if (!s.ok()) {
+      if (!Merge(&blob_value_, ikey.user_key)) {
         return false;
       }
 
@@ -589,11 +589,18 @@ bool DBIter::MergeValuesNewToOld() {
       }
       return true;
     } else if (kTypeWideColumnEntity == ikey.type) {
-      // TODO: support wide-column entities
-      status_ = Status::NotSupported(
-          "Merge currently not supported for wide-column entities");
-      valid_ = false;
-      return false;
+      if (!MergeEntity(iter_.value(), ikey.user_key)) {
+        return false;
+      }
+
+      // iter_ is positioned after put
+      iter_.Next();
+      if (!iter_.status().ok()) {
+        valid_ = false;
+        return false;
+      }
+
+      return true;
     } else {
       valid_ = false;
       status_ = Status::Corruption(
@@ -612,8 +619,7 @@ bool DBIter::MergeValuesNewToOld() {
   // a deletion marker.
   // feed null as the existing value to the merge operator, such that
   // client can differentiate this scenario and do things accordingly.
-  Status s = Merge(nullptr, saved_key_.GetUserKey());
-  if (!s.ok()) {
+  if (!Merge(nullptr, saved_key_.GetUserKey())) {
     return false;
   }
   assert(status_.ok());
@@ -636,6 +642,8 @@ void DBIter::Prev() {
     }
   }
   if (ok) {
+    ClearSavedValue();
+
     Slice prefix;
     if (prefix_same_as_start_) {
       assert(prefix_extractor_ != nullptr);
@@ -960,8 +968,7 @@ bool DBIter::FindValueForCurrentKey() {
       if (last_not_merge_type == kTypeDeletion ||
           last_not_merge_type == kTypeSingleDeletion ||
           last_not_merge_type == kTypeDeletionWithTimestamp) {
-        s = Merge(nullptr, saved_key_.GetUserKey());
-        if (!s.ok()) {
+        if (!Merge(nullptr, saved_key_.GetUserKey())) {
           return false;
         }
         return true;
@@ -976,8 +983,7 @@ bool DBIter::FindValueForCurrentKey() {
           return false;
         }
         valid_ = true;
-        s = Merge(&blob_value_, saved_key_.GetUserKey());
-        if (!s.ok()) {
+        if (!Merge(&blob_value_, saved_key_.GetUserKey())) {
           return false;
         }
 
@@ -985,15 +991,14 @@ bool DBIter::FindValueForCurrentKey() {
 
         return true;
       } else if (last_not_merge_type == kTypeWideColumnEntity) {
-        // TODO: support wide-column entities
-        status_ = Status::NotSupported(
-            "Merge currently not supported for wide-column entities");
-        valid_ = false;
-        return false;
+        if (!MergeEntity(pinned_value_, saved_key_.GetUserKey())) {
+          return false;
+        }
+
+        return true;
       } else {
         assert(last_not_merge_type == kTypeValue);
-        s = Merge(&pinned_value_, saved_key_.GetUserKey());
-        if (!s.ok()) {
+        if (!Merge(&pinned_value_, saved_key_.GetUserKey())) {
           return false;
         }
         return true;
@@ -1177,8 +1182,7 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
 
     if (ikey.type == kTypeValue) {
       const Slice val = iter_.value();
-      Status s = Merge(&val, saved_key_.GetUserKey());
-      if (!s.ok()) {
+      if (!Merge(&val, saved_key_.GetUserKey())) {
         return false;
       }
       return true;
@@ -1197,8 +1201,7 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
         return false;
       }
       valid_ = true;
-      Status s = Merge(&blob_value_, saved_key_.GetUserKey());
-      if (!s.ok()) {
+      if (!Merge(&blob_value_, saved_key_.GetUserKey())) {
         return false;
       }
 
@@ -1206,11 +1209,11 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
 
       return true;
     } else if (ikey.type == kTypeWideColumnEntity) {
-      // TODO: support wide-column entities
-      status_ = Status::NotSupported(
-          "Merge currently not supported for wide-column entities");
-      valid_ = false;
-      return false;
+      if (!MergeEntity(iter_.value(), saved_key_.GetUserKey())) {
+        return false;
+      }
+
+      return true;
     } else {
       valid_ = false;
       status_ = Status::Corruption(
@@ -1220,8 +1223,7 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
     }
   }
 
-  Status s = Merge(nullptr, saved_key_.GetUserKey());
-  if (!s.ok()) {
+  if (!Merge(nullptr, saved_key_.GetUserKey())) {
     return false;
   }
 
@@ -1244,7 +1246,7 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
   return true;
 }
 
-Status DBIter::Merge(const Slice* val, const Slice& user_key) {
+bool DBIter::Merge(const Slice* val, const Slice& user_key) {
   Status s = MergeHelper::TimedFullMerge(
       merge_operator_, user_key, val, merge_context_.GetOperands(),
       &saved_value_, logger_, statistics_, clock_, &pinned_value_,
@@ -1252,14 +1254,33 @@ Status DBIter::Merge(const Slice* val, const Slice& user_key) {
   if (!s.ok()) {
     valid_ = false;
     status_ = s;
-    return s;
+    return false;
   }
 
   SetValueAndColumnsFromPlain(pinned_value_.data() ? pinned_value_
                                                    : saved_value_);
 
   valid_ = true;
-  return s;
+  return true;
+}
+
+bool DBIter::MergeEntity(const Slice& entity, const Slice& user_key) {
+  Status s = MergeHelper::TimedFullMergeWithEntity(
+      merge_operator_, user_key, entity, merge_context_.GetOperands(),
+      &saved_value_, logger_, statistics_, clock_,
+      /* update_num_ops_stats */ true);
+  if (!s.ok()) {
+    valid_ = false;
+    status_ = s;
+    return false;
+  }
+
+  if (!SetValueAndColumnsFromEntity(saved_value_)) {
+    return false;
+  }
+
+  valid_ = true;
+  return true;
 }
 
 // Move backwards until the key smaller than saved_key_.

package/deps/rocksdb/rocksdb/db/db_iter.h

@@ -318,7 +318,8 @@ class DBIter final : public Iterator {
   }
 
   // If user-defined timestamp is enabled, `user_key` includes timestamp.
-  Status Merge(const Slice* val, const Slice& user_key);
+  bool Merge(const Slice* val, const Slice& user_key);
+  bool MergeEntity(const Slice& entity, const Slice& user_key);
 
   const SliceTransform* prefix_extractor_;
   Env* const env_;