@nxtedition/rocksdb 7.1.30 → 7.1.32

package/deps/rocksdb/rocksdb/db/compaction/compaction_picker.h

@@ -51,15 +51,24 @@ class CompactionPicker {
   virtual ~CompactionPicker();
 
   // Pick level and inputs for a new compaction.
+  //
+  // `earliest_mem_seqno` is the earliest seqno of unflushed memtables.
+  // It is needed to compare with compaction input SST files' largest seqnos
+  // in order to exclude those of seqnos potentially overlap with memtables'
+  // seqnos when doing compaction to L0. This will avoid creating a SST files in
+  // L0 newer than a unflushed memtable. Such SST file can exist in the first
+  // place when it's ingested or resulted from compaction involving files
+  // ingested.
+  //
   // Returns nullptr if there is no compaction to be done.
   // Otherwise returns a pointer to a heap-allocated object that
   // describes the compaction.  Caller should delete the result.
   virtual Compaction* PickCompaction(
       const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
       const MutableDBOptions& mutable_db_options, VersionStorageInfo* vstorage,
-      LogBuffer* log_buffer,
-      SequenceNumber earliest_memtable_seqno = kMaxSequenceNumber) = 0;
+      LogBuffer* log_buffer, const SequenceNumber earliest_mem_seqno) = 0;
 
+  // `earliest_mem_seqno`: see PickCompaction() API
   // Return a compaction object for compacting the range [begin,end] in
   // the specified level.  Returns nullptr if there is nothing in that
   // level that overlaps the specified range.  Caller should delete
@@ -78,7 +87,8 @@ class CompactionPicker {
       const CompactRangeOptions& compact_range_options,
       const InternalKey* begin, const InternalKey* end,
       InternalKey** compaction_end, bool* manual_conflict,
-      uint64_t max_file_num_to_ignore, const std::string& trim_ts);
+      uint64_t max_file_num_to_ignore, const std::string& trim_ts,
+      const SequenceNumber earliest_mem_seqno);
 
   // The maximum allowed output level.  Default value is NumberLevels() - 1.
   virtual int MaxOutputLevel() const { return NumberLevels() - 1; }
@@ -91,10 +101,18 @@ class CompactionPicker {
 // files.  If it's not possible to conver an invalid input_files
 // into a valid one by adding more files, the function will return a
 // non-ok status with specific reason.
+//
+// Cases of returning non-ok status include but not limited to:
+// - When output_level == 0 and input_files contains sst files
+// of largest seqno greater than `earliest_mem_seqno`. This will
+// avoid creating a SST files in L0 newer than a unflushed memtable.
+// Such SST file can exist in the first place when it's ingested or
+// resulted from compaction involving files ingested.
 #ifndef ROCKSDB_LITE
-  Status SanitizeCompactionInputFiles(std::unordered_set<uint64_t>* input_files,
-                                      const ColumnFamilyMetaData& cf_meta,
-                                      const int output_level) const;
+  Status SanitizeCompactionInputFiles(
+      std::unordered_set<uint64_t>* input_files,
+      const ColumnFamilyMetaData& cf_meta, const int output_level,
+      const SequenceNumber earliest_mem_seqno) const;
 #endif  // ROCKSDB_LITE
 
   // Free up the files that participated in a compaction
@@ -182,7 +200,8 @@ class CompactionPicker {
   // Returns true if the key range that `inputs` files cover overlap with the
   // key range of a currently running compaction.
   bool FilesRangeOverlapWithCompaction(
-      const std::vector<CompactionInputFiles>& inputs, int level) const;
+      const std::vector<CompactionInputFiles>& inputs, int level,
+      int penultimate_level) const;
 
   bool SetupOtherInputs(const std::string& cf_name,
                         const MutableCFOptions& mutable_cf_options,
@@ -229,7 +248,8 @@ class CompactionPicker {
 #ifndef ROCKSDB_LITE
   virtual Status SanitizeCompactionInputFilesForAllLevels(
       std::unordered_set<uint64_t>* input_files,
-      const ColumnFamilyMetaData& cf_meta, const int output_level) const;
+      const ColumnFamilyMetaData& cf_meta, const int output_level,
+      const SequenceNumber earliest_mem_seqno) const;
 #endif  // ROCKSDB_LITE
 
   // Keeps track of all compactions that are running on Level0.
@@ -259,23 +279,22 @@ class NullCompactionPicker : public CompactionPicker {
       const MutableCFOptions& /*mutable_cf_options*/,
       const MutableDBOptions& /*mutable_db_options*/,
       VersionStorageInfo* /*vstorage*/, LogBuffer* /* log_buffer */,
-      SequenceNumber /* earliest_memtable_seqno */) override {
+      const SequenceNumber /* earliest_mem_seqno */) override {
     return nullptr;
   }
 
   // Always return "nullptr"
-  Compaction* CompactRange(const std::string& /*cf_name*/,
-                           const MutableCFOptions& /*mutable_cf_options*/,
-                           const MutableDBOptions& /*mutable_db_options*/,
-                           VersionStorageInfo* /*vstorage*/,
-                           int /*input_level*/, int /*output_level*/,
-                           const CompactRangeOptions& /*compact_range_options*/,
-                           const InternalKey* /*begin*/,
-                           const InternalKey* /*end*/,
-                           InternalKey** /*compaction_end*/,
-                           bool* /*manual_conflict*/,
-                           uint64_t /*max_file_num_to_ignore*/,
-                           const std::string& /*trim_ts*/) override {
+  Compaction* CompactRange(
+      const std::string& /*cf_name*/,
+      const MutableCFOptions& /*mutable_cf_options*/,
+      const MutableDBOptions& /*mutable_db_options*/,
+      VersionStorageInfo* /*vstorage*/, int /*input_level*/,
+      int /*output_level*/,
+      const CompactRangeOptions& /*compact_range_options*/,
+      const InternalKey* /*begin*/, const InternalKey* /*end*/,
+      InternalKey** /*compaction_end*/, bool* /*manual_conflict*/,
+      uint64_t /*max_file_num_to_ignore*/, const std::string& /*trim_ts*/,
+      const SequenceNumber /* earliest_mem_seqno */) override {
     return nullptr;
   }
 
@@ -302,12 +321,14 @@ class NullCompactionPicker : public CompactionPicker {
 //                                        initialized with corresponding input
 //                                        files. Cannot be nullptr.
 //
+// @param earliest_mem_seqno              See PickCompaction() API
 // @return                                true iff compaction was found.
-bool FindIntraL0Compaction(
-    const std::vector<FileMetaData*>& level_files, size_t min_files_to_compact,
-    uint64_t max_compact_bytes_per_del_file, uint64_t max_compaction_bytes,
-    CompactionInputFiles* comp_inputs,
-    SequenceNumber earliest_mem_seqno = kMaxSequenceNumber);
+bool FindIntraL0Compaction(const std::vector<FileMetaData*>& level_files,
+                           size_t min_files_to_compact,
+                           uint64_t max_compact_bytes_per_del_file,
+                           uint64_t max_compaction_bytes,
+                           CompactionInputFiles* comp_inputs,
+                           const SequenceNumber earliest_mem_seqno);
 
 CompressionType GetCompressionType(const VersionStorageInfo* vstorage,
                                    const MutableCFOptions& mutable_cf_options,

package/deps/rocksdb/rocksdb/db/compaction/compaction_picker_fifo.cc

@@ -139,7 +139,7 @@ Compaction* FIFOCompactionPicker::PickTTLCompaction(
 Compaction* FIFOCompactionPicker::PickSizeCompaction(
     const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
     const MutableDBOptions& mutable_db_options, VersionStorageInfo* vstorage,
-    LogBuffer* log_buffer) {
+    LogBuffer* log_buffer, const SequenceNumber earliest_mem_seqno) {
   // compute the total size and identify the last non-empty level
   int last_level = 0;
   uint64_t total_size = 0;
@@ -176,7 +176,8 @@ Compaction* FIFOCompactionPicker::PickSizeCompaction(
                   .level0_file_num_compaction_trigger /* min_files_to_compact */
               ,
               max_compact_bytes_per_del_file,
-              mutable_cf_options.max_compaction_bytes, &comp_inputs)) {
+              mutable_cf_options.max_compaction_bytes, &comp_inputs,
+              earliest_mem_seqno)) {
         Compaction* c = new Compaction(
             vstorage, ioptions_, mutable_cf_options, mutable_db_options,
             {comp_inputs}, 0, 16 * 1024 * 1024 /* output file size limit */,
@@ -275,7 +276,8 @@ Compaction* FIFOCompactionPicker::PickSizeCompaction(
 Compaction* FIFOCompactionPicker::PickCompactionToWarm(
     const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
     const MutableDBOptions& mutable_db_options, VersionStorageInfo* vstorage,
-    LogBuffer* log_buffer) {
+    LogBuffer* log_buffer, const SequenceNumber earliest_mem_seqno) {
+  TEST_SYNC_POINT("PickCompactionToWarm");
   if (mutable_cf_options.compaction_options_fifo.age_for_warm == 0) {
     return nullptr;
   }
@@ -299,6 +301,8 @@ Compaction* FIFOCompactionPicker::PickCompactionToWarm(
                      cf_name.c_str(), status.ToString().c_str());
     return nullptr;
   }
+  TEST_SYNC_POINT_CALLBACK("PickCompactionToWarm::BeforeGetCurrentTime",
+                           &_current_time);
   const uint64_t current_time = static_cast<uint64_t>(_current_time);
 
   if (!level0_compactions_in_progress_.empty()) {
@@ -345,7 +349,8 @@ Compaction* FIFOCompactionPicker::PickCompactionToWarm(
         // for warm tier.
         break;
       }
-      if (prev_file != nullptr) {
+      if (prev_file != nullptr &&
+          prev_file->fd.largest_seqno <= earliest_mem_seqno) {
         compaction_size += prev_file->fd.GetFileSize();
         if (compaction_size > mutable_cf_options.max_compaction_bytes) {
           break;
@@ -389,7 +394,7 @@ Compaction* FIFOCompactionPicker::PickCompactionToWarm(
 Compaction* FIFOCompactionPicker::PickCompaction(
     const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
     const MutableDBOptions& mutable_db_options, VersionStorageInfo* vstorage,
-    LogBuffer* log_buffer, SequenceNumber /*earliest_memtable_seqno*/) {
+    LogBuffer* log_buffer, const SequenceNumber earliest_mem_seqno) {
   Compaction* c = nullptr;
   if (mutable_cf_options.ttl > 0) {
     c = PickTTLCompaction(cf_name, mutable_cf_options, mutable_db_options,
@@ -397,11 +402,11 @@ Compaction* FIFOCompactionPicker::PickCompaction(
   }
   if (c == nullptr) {
     c = PickSizeCompaction(cf_name, mutable_cf_options, mutable_db_options,
-                           vstorage, log_buffer);
+                           vstorage, log_buffer, earliest_mem_seqno);
   }
   if (c == nullptr) {
     c = PickCompactionToWarm(cf_name, mutable_cf_options, mutable_db_options,
-                             vstorage, log_buffer);
+                             vstorage, log_buffer, earliest_mem_seqno);
   }
   RegisterCompaction(c);
   return c;
@@ -414,7 +419,8 @@ Compaction* FIFOCompactionPicker::CompactRange(
     const CompactRangeOptions& /*compact_range_options*/,
     const InternalKey* /*begin*/, const InternalKey* /*end*/,
     InternalKey** compaction_end, bool* /*manual_conflict*/,
-    uint64_t /*max_file_num_to_ignore*/, const std::string& /*trim_ts*/) {
+    uint64_t /*max_file_num_to_ignore*/, const std::string& /*trim_ts*/,
+    const SequenceNumber earliest_mem_seqno) {
 #ifdef NDEBUG
   (void)input_level;
   (void)output_level;
@@ -423,8 +429,9 @@ Compaction* FIFOCompactionPicker::CompactRange(
   assert(output_level == 0);
   *compaction_end = nullptr;
   LogBuffer log_buffer(InfoLogLevel::INFO_LEVEL, ioptions_.logger);
-  Compaction* c = PickCompaction(cf_name, mutable_cf_options,
-                                 mutable_db_options, vstorage, &log_buffer);
+  Compaction* c =
+      PickCompaction(cf_name, mutable_cf_options, mutable_db_options, vstorage,
+                     &log_buffer, earliest_mem_seqno);
   log_buffer.FlushBufferToLog();
   return c;
 }

package/deps/rocksdb/rocksdb/db/compaction/compaction_picker_fifo.h

@@ -22,9 +22,12 @@ class FIFOCompactionPicker : public CompactionPicker {
   virtual Compaction* PickCompaction(
       const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
       const MutableDBOptions& mutable_db_options, VersionStorageInfo* version,
-      LogBuffer* log_buffer,
-      SequenceNumber earliest_memtable_seqno = kMaxSequenceNumber) override;
+      LogBuffer* log_buffer, const SequenceNumber earliest_mem_seqno) override;
 
+  // `earliest_mem_seqno`: see PickCompaction() API for more. In FIFO's
+  // implementation of CompactRange(), different from others, we will not return
+  // `nullptr` right away when intput files of compaction to L0 has seqnos
+  // potentially overlapping with memtable's but exlucde those files.
   virtual Compaction* CompactRange(
       const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
       const MutableDBOptions& mutable_db_options, VersionStorageInfo* vstorage,
@@ -32,7 +35,8 @@ class FIFOCompactionPicker : public CompactionPicker {
       const CompactRangeOptions& compact_range_options,
       const InternalKey* begin, const InternalKey* end,
       InternalKey** compaction_end, bool* manual_conflict,
-      uint64_t max_file_num_to_ignore, const std::string& trim_ts) override;
+      uint64_t max_file_num_to_ignore, const std::string& trim_ts,
+      const SequenceNumber earliest_mem_seqno) override;
 
   // The maximum allowed output level.  Always returns 0.
   virtual int MaxOutputLevel() const override { return 0; }
@@ -51,13 +55,15 @@ class FIFOCompactionPicker : public CompactionPicker {
                                  const MutableCFOptions& mutable_cf_options,
                                  const MutableDBOptions& mutable_db_options,
                                  VersionStorageInfo* version,
-                                 LogBuffer* log_buffer);
+                                 LogBuffer* log_buffer,
+                                 SequenceNumber earliest_mem_seqno);
 
   Compaction* PickCompactionToWarm(const std::string& cf_name,
                                    const MutableCFOptions& mutable_cf_options,
                                    const MutableDBOptions& mutable_db_options,
                                    VersionStorageInfo* version,
-                                   LogBuffer* log_buffer);
+                                   LogBuffer* log_buffer,
+                                   const SequenceNumber earliest_mem_seqno);
 };
 }  // namespace ROCKSDB_NAMESPACE
 #endif  // !ROCKSDB_LITE

package/deps/rocksdb/rocksdb/db/compaction/compaction_picker_level.cc

@@ -50,7 +50,7 @@ class LevelCompactionBuilder {
  public:
   LevelCompactionBuilder(const std::string& cf_name,
                          VersionStorageInfo* vstorage,
-                         SequenceNumber earliest_mem_seqno,
+                         const SequenceNumber earliest_mem_seqno,
                          CompactionPicker* compaction_picker,
                          LogBuffer* log_buffer,
                          const MutableCFOptions& mutable_cf_options,
@@ -122,7 +122,7 @@ class LevelCompactionBuilder {
 
   const std::string& cf_name_;
   VersionStorageInfo* vstorage_;
-  SequenceNumber earliest_mem_seqno_;
+  const SequenceNumber earliest_mem_seqno_;
   CompactionPicker* compaction_picker_;
   LogBuffer* log_buffer_;
   int start_level_ = -1;
@@ -379,7 +379,9 @@ void LevelCompactionBuilder::SetupOtherFilesWithRoundRobinExpansion() {
     if (!compaction_picker_->ExpandInputsToCleanCut(cf_name_, vstorage_,
                                                     &tmp_start_level_inputs) ||
         compaction_picker_->FilesRangeOverlapWithCompaction(
-            {tmp_start_level_inputs}, output_level_)) {
+            {tmp_start_level_inputs}, output_level_,
+            Compaction::EvaluatePenultimateLevel(
+                vstorage_, ioptions_, start_level_, output_level_))) {
       // Constraint 1a
       tmp_start_level_inputs.clear();
       return;
@@ -453,7 +455,9 @@ bool LevelCompactionBuilder::SetupOtherInputsIfNeeded() {
       // (2) AddFile ingest a new file into the LSM tree
       // We need to disallow this from happening.
       if (compaction_picker_->FilesRangeOverlapWithCompaction(
-              compaction_inputs_, output_level_)) {
+              compaction_inputs_, output_level_,
+              Compaction::EvaluatePenultimateLevel(
+                  vstorage_, ioptions_, start_level_, output_level_))) {
         // This compaction output could potentially conflict with the output
         // of a currently running compaction, we cannot run it.
         return false;
@@ -755,7 +759,9 @@ bool LevelCompactionBuilder::PickFileToCompact() {
     if (!compaction_picker_->ExpandInputsToCleanCut(cf_name_, vstorage_,
                                                     &start_level_inputs_) ||
         compaction_picker_->FilesRangeOverlapWithCompaction(
-            {start_level_inputs_}, output_level_)) {
+            {start_level_inputs_}, output_level_,
+            Compaction::EvaluatePenultimateLevel(
+                vstorage_, ioptions_, start_level_, output_level_))) {
       // A locked (pending compaction) input-level file was pulled in due to
       // user-key overlap.
       start_level_inputs_.clear();
@@ -826,7 +832,7 @@ bool LevelCompactionBuilder::PickIntraL0Compaction() {
 Compaction* LevelCompactionPicker::PickCompaction(
     const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
     const MutableDBOptions& mutable_db_options, VersionStorageInfo* vstorage,
-    LogBuffer* log_buffer, SequenceNumber earliest_mem_seqno) {
+    LogBuffer* log_buffer, const SequenceNumber earliest_mem_seqno) {
   LevelCompactionBuilder builder(cf_name, vstorage, earliest_mem_seqno, this,
                                  log_buffer, mutable_cf_options, ioptions_,
                                  mutable_db_options);

package/deps/rocksdb/rocksdb/db/compaction/compaction_picker_level.h

@@ -23,8 +23,7 @@ class LevelCompactionPicker : public CompactionPicker {
   virtual Compaction* PickCompaction(
       const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
       const MutableDBOptions& mutable_db_options, VersionStorageInfo* vstorage,
-      LogBuffer* log_buffer,
-      SequenceNumber earliest_memtable_seqno = kMaxSequenceNumber) override;
+      LogBuffer* log_buffer, const SequenceNumber earliest_mem_seqno) override;
 
   virtual bool NeedsCompaction(
       const VersionStorageInfo* vstorage) const override;