Update history and version for 5.16.6 release.

Fix WriteBatchWithIndex's SeekForPrev() (#4559 )
Summary: WriteBatchWithIndex's SeekForPrev() has a bug that we internally place the position just before the seek key rather than after. This makes the iterator to miss the result that is the same as the seek key. Fix it by position the iterator equal or smaller. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4559 Differential Revision: D10468534 Pulled By: siying fbshipit-source-id: 2fb371ae809c561b60a1c11cef71e1c66fea1f19
2026-11-07 11:32:34 -05:00 · 2018-11-12 11:55:03 -08:00 · 2018-10-19 15:44:58 -07:00 · 2018-10-16 10:41:36 -07:00 · 2018-10-16 10:40:21 -07:00 · 2018-10-16 10:33:02 -07:00
830 changed files with 33947 additions and 63484 deletions
@@ -90,7 +90,7 @@ script:
      OPT=-DTRAVIS V=1 ROCKSDBTESTS_START=db_block_cache_test ROCKSDBTESTS_END=full_filter_block_test make -j4 check_some
      ;;
    2)
-      OPT=-DTRAVIS V=1 make -j4 tools && OPT=-DTRAVIS V=1 ROCKSDBTESTS_START=full_filter_block_test ROCKSDBTESTS_END=write_batch_with_index_test make -j4 check_some
+      OPT=-DTRAVIS V=1 ROCKSDBTESTS_START=full_filter_block_test ROCKSDBTESTS_END=write_batch_with_index_test make -j4 check_some
      ;;
    3)
      OPT=-DTRAVIS V=1 ROCKSDBTESTS_START=write_batch_with_index_test ROCKSDBTESTS_END=write_prepared_transaction_test make -j4 check_some
@@ -49,10 +49,6 @@ option(WITH_SNAPPY "build with SNAPPY" OFF)
 option(WITH_LZ4 "build with lz4" OFF)
 option(WITH_ZLIB "build with zlib" OFF)
 option(WITH_ZSTD "build with zstd" OFF)
-option(WITH_WINDOWS_UTF8_FILENAMES "use UTF8 as characterset for opening files, regardles of the system code page" OFF)
-if (WITH_WINDOWS_UTF8_FILENAMES)
-  add_definitions(-DROCKSDB_WINDOWS_UTF8_FILENAMES)
-endif()
 if(MSVC)
  # Defaults currently different for GFLAGS.
  #  We will address find_package work a little later
@@ -92,15 +88,9 @@ else()
  endif()

  if(WITH_ZLIB)
-    find_package(ZLIB REQUIRED)
+    find_package(zlib REQUIRED)
    add_definitions(-DZLIB)
-    if(ZLIB_INCLUDE_DIRS)
-      # CMake 3
-      include_directories(${ZLIB_INCLUDE_DIRS})
-    else()
-      # CMake 2
-      include_directories(${ZLIB_INCLUDE_DIR})
-    endif()
+    include_directories(${ZLIB_INCLUDE_DIR})
    list(APPEND THIRDPARTY_LIBS ${ZLIB_LIBRARIES})
  endif()

@@ -317,10 +307,6 @@ if(DISABLE_STALL_NOTIF)
  add_definitions(-DROCKSDB_DISABLE_STALL_NOTIFICATION)
 endif()

-option(WITH_DYNAMIC_EXTENSION "build with dynamic extension support" OFF)
-if(NOT WITH_DYNAMIC_EXTENSION)
-  add_definitions(-DROCKSDB_NO_DYNAMIC_EXTENSION)
-endif()

 if(DEFINED USE_RTTI)
  if(USE_RTTI)
@@ -350,18 +336,12 @@ endif()

 # Used to run CI build and tests so we can run faster
 option(OPTDBG "Build optimized debug build with MSVC" OFF)
-option(WITH_RUNTIME_DEBUG "build with debug version of runtime library" ON)
 if(MSVC)
  if(OPTDBG)
    message(STATUS "Debug optimization is enabled")
-    set(CMAKE_CXX_FLAGS_DEBUG "/Oxt")
+    set(CMAKE_CXX_FLAGS_DEBUG "/Oxt /${RUNTIME_LIBRARY}d")
  else()
-    set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} /Od /RTC1 /Gm")
-  endif()
-  if(WITH_RUNTIME_DEBUG)
-    set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} /${RUNTIME_LIBRARY}d")
-  else()
-    set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} /${RUNTIME_LIBRARY}")
+    set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} /Od /RTC1 /Gm /${RUNTIME_LIBRARY}d")
  endif()
  set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} /Oxt /Zp8 /Gm- /Gy /${RUNTIME_LIBRARY}")

@@ -376,7 +356,7 @@ endif()
 option(ROCKSDB_LITE "Build RocksDBLite version" OFF)
 if(ROCKSDB_LITE)
  add_definitions(-DROCKSDB_LITE)
-  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fno-exceptions -Os")
+  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fno-exceptions")
 endif()

 if(CMAKE_SYSTEM_NAME MATCHES "Cygwin")
@@ -420,7 +400,7 @@ if(WITH_FALLOCATE)
 #include <linux/falloc.h>
 int main() {
 int fd = open(\"/dev/null\", 0);
- fallocate(fd, FALLOC_FL_KEEP_SIZE, 0, 1024);
+ fallocate(fd, FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE, 0, 1024);
 }
 " HAVE_FALLOCATE)
  if(HAVE_FALLOCATE)
@@ -465,6 +445,8 @@ include_directories(${PROJECT_SOURCE_DIR}/include)
 include_directories(SYSTEM ${PROJECT_SOURCE_DIR}/third-party/gtest-1.7.0/fused-src)
 find_package(Threads REQUIRED)

+add_subdirectory(third-party/gtest-1.7.0/fused-src/gtest)
+
 # Main library source code

 set(SOURCES
@@ -479,7 +461,6 @@ set(SOURCES
        db/compaction_iterator.cc
        db/compaction_job.cc
        db/compaction_picker.cc
-        db/compaction_picker_fifo.cc
        db/compaction_picker_universal.cc
        db/convenience.cc
        db/db_filesnapshot.cc
@@ -491,7 +472,6 @@ set(SOURCES
        db/db_impl_debug.cc
        db/db_impl_experimental.cc
        db/db_impl_readonly.cc
-        db/db_impl_secondary.cc
        db/db_info_dumper.cc
        db/db_iter.cc
        db/dbformat.cc
@@ -504,7 +484,6 @@ set(SOURCES
        db/flush_scheduler.cc
        db/forward_iterator.cc
        db/internal_stats.cc
-        db/in_memory_stats_history.cc
        db/logs_with_prep_tracker.cc
        db/log_reader.cc
        db/log_writer.cc
@@ -514,7 +493,6 @@ set(SOURCES
        db/merge_helper.cc
        db/merge_operator.cc
        db/range_del_aggregator.cc
-        db/range_tombstone_fragmenter.cc
        db/repair.cc
        db/snapshot_impl.cc
        db/table_cache.cc
@@ -534,6 +512,7 @@ set(SOURCES
        env/env_hdfs.cc
        env/mock_env.cc
        memtable/alloc_tracker.cc
+        memtable/hash_cuckoo_rep.cc
        memtable/hash_linklist_rep.cc
        memtable/hash_skiplist_rep.cc
        memtable/skiplistrep.cc
@@ -587,7 +566,6 @@ set(SOURCES
        table/plain_table_index.cc
        table/plain_table_key_coding.cc
        table/plain_table_reader.cc
-        table/sst_file_reader.cc
        table/sst_file_writer.cc
        table/table_properties.cc
        table/two_level_iterator.cc
@@ -605,7 +583,6 @@ set(SOURCES
        util/comparator.cc
        util/compression_context_cache.cc
        util/concurrent_arena.cc
-        util/concurrent_task_limiter_impl.cc
        util/crc32c.cc
        util/delete_scheduler.cc
        util/dynamic_bloom.cc
@@ -615,7 +592,6 @@ set(SOURCES
        util/filename.cc
        util/filter_policy.cc
        util/hash.cc
-        util/jemalloc_nodump_allocator.cc
        util/log_buffer.cc
        util/murmurhash.cc
        util/random.cc
@@ -646,11 +622,20 @@ set(SOURCES
        utilities/cassandra/format.cc
        utilities/cassandra/merge_operator.cc
        utilities/checkpoint/checkpoint_impl.cc
+        utilities/col_buf_decoder.cc
+        utilities/col_buf_encoder.cc
+        utilities/column_aware_encoding_util.cc
        utilities/compaction_filters/remove_emptyvalue_compactionfilter.cc
+        utilities/date_tiered/date_tiered_db_impl.cc
        utilities/debug.cc
+        utilities/document/document_db.cc
+        utilities/document/json_document.cc
+        utilities/document/json_document_builder.cc
        utilities/env_mirror.cc
        utilities/env_timed.cc
+        utilities/geodb/geodb_impl.cc
        utilities/leveldb_options/leveldb_options.cc
+        utilities/lua/rocks_lua_compaction_filter.cc
        utilities/memory/memory_util.cc
        utilities/merge_operators/bytesxor.cc
        utilities/merge_operators/max.cc
@@ -665,7 +650,9 @@ set(SOURCES
        utilities/persistent_cache/block_cache_tier_metadata.cc
        utilities/persistent_cache/persistent_cache_tier.cc
        utilities/persistent_cache/volatile_tier_impl.cc
+        utilities/redis/redis_lists.cc
        utilities/simulator_cache/sim_cache.cc
+        utilities/spatialdb/spatial_db.cc
        utilities/table_properties_collectors/compact_on_deletion_collector.cc
        utilities/trace/file_trace_reader_writer.cc
        utilities/transactions/optimistic_transaction_db_impl.cc
@@ -686,10 +673,12 @@ set(SOURCES
        utilities/write_batch_with_index/write_batch_with_index_internal.cc
        $<TARGET_OBJECTS:build_version>)

-if(HAVE_SSE42 AND NOT MSVC)
-  set_source_files_properties(
-    util/crc32c.cc
-    PROPERTIES COMPILE_FLAGS "-msse4.2 -mpclmul")
+if(HAVE_SSE42 AND NOT FORCE_SSE42)
+  if(NOT MSVC)
+set_source_files_properties(
+  util/crc32c.cc
+      PROPERTIES COMPILE_FLAGS "-msse4.2 -mpclmul")
+  endif()
 endif()

 if(HAVE_POWER8)
@@ -821,7 +810,6 @@ if(NOT WIN32 OR ROCKSDB_INSTALL_ON_WINDOWS)
    TARGETS ${ROCKSDB_SHARED_LIB}
    EXPORT RocksDBTargets
    COMPONENT runtime
-    ARCHIVE DESTINATION "${CMAKE_INSTALL_LIBDIR}"
    RUNTIME DESTINATION "${CMAKE_INSTALL_BINDIR}"
    LIBRARY DESTINATION "${CMAKE_INSTALL_LIBDIR}"
    INCLUDES DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}"
@@ -845,7 +833,6 @@ endif()

 option(WITH_TESTS "build with tests" ON)
 if(WITH_TESTS)
-  add_subdirectory(third-party/gtest-1.7.0/fused-src/gtest)
  set(TESTS
        cache/cache_test.cc
        cache/lru_cache_test.cc
@@ -877,7 +864,6 @@ if(WITH_TESTS)
        db/db_options_test.cc
        db/db_properties_test.cc
        db/db_range_del_test.cc
-        db/db_secondary_test.cc
        db/db_sst_test.cc
        db/db_statistics_test.cc
        db/db_table_properties_test.cc
@@ -907,8 +893,6 @@ if(WITH_TESTS)
        db/perf_context_test.cc
        db/plain_table_db_test.cc
        db/prefix_test.cc
-        db/range_del_aggregator_test.cc
-        db/range_tombstone_fragmenter_test.cc
        db/repair_test.cc
        db/table_properties_collector_test.cc
        db/version_builder_test.cc
@@ -937,7 +921,6 @@ if(WITH_TESTS)
        table/data_block_hash_index_test.cc
        table/full_filter_block_test.cc
        table/merger_test.cc
-        table/sst_file_reader_test.cc
        table/table_test.cc
        tools/ldb_cmd_test.cc
        tools/reduce_levels_test.cc
@@ -957,7 +940,6 @@ if(WITH_TESTS)
        util/hash_test.cc
        util/heap_test.cc
        util/rate_limiter_test.cc
-        util/repeatable_thread_test.cc
        util/slice_transform_test.cc
        util/timer_queue_test.cc
        util/thread_list_test.cc
@@ -969,6 +951,12 @@ if(WITH_TESTS)
        utilities/cassandra/cassandra_row_merge_test.cc
        utilities/cassandra/cassandra_serialize_test.cc
        utilities/checkpoint/checkpoint_test.cc
+        utilities/column_aware_encoding_test.cc
+        utilities/date_tiered/date_tiered_test.cc
+        utilities/document/document_db_test.cc
+        utilities/document/json_document_test.cc
+        utilities/geodb/geodb_test.cc
+        utilities/lua/rocks_lua_test.cc
        utilities/memory/memory_test.cc
        utilities/merge_operators/string_append/stringappend_test.cc
        utilities/object_registry_test.cc
@@ -976,6 +964,8 @@ if(WITH_TESTS)
        utilities/options/options_util_test.cc
        utilities/persistent_cache/hash_table_test.cc
        utilities/persistent_cache/persistent_cache_test.cc
+        utilities/redis/redis_lists_test.cc
+        utilities/spatialdb/spatial_db_test.cc
        utilities/simulator_cache/sim_cache_test.cc
        utilities/table_properties_collectors/compact_on_deletion_collector_test.cc
        utilities/transactions/optimistic_transaction_test.cc
@@ -992,9 +982,9 @@ if(WITH_TESTS)
  set(BENCHMARKS
    cache/cache_bench.cc
    memtable/memtablerep_bench.cc
-    db/range_del_aggregator_bench.cc
    tools/db_bench.cc
    table/table_reader_bench.cc
+    utilities/column_aware_encoding_exp.cc
    utilities/persistent_cache/hash_table_bench.cc)
  add_library(testharness OBJECT util/testharness.cc)
  foreach(sourcefile ${BENCHMARKS})
@@ -1,153 +1,17 @@
 # Rocksdb Change Log
-## 6.2.4 (9/18/2019)
+
+# 5.16.6 (11/24/2018)
 ### Bug Fixes
-* Disable snap_refresh_nanos by default. The feature is to be deprecated in the next release.
+* Fix the bug that WriteBatchWithIndex's SeekForPrev() doesn't see the entries with the same key.

-## 6.2.3 (9/3/2019)
+## 5.16.5 (10/16/2018)
 ### Bug Fixes
-* Fix a bug in file ingestion caused by incorrect file number allocation when the number of column families involved in the ingestion exceeds 2.
-
-## 6.2.2 (6/7/2019)
-### Bug Fixes
-* Disable dynamic extension support by default for CMake.
-
-## 6.2.1 (6/4/2019)
-### Bug Fixes
-* Fix flush's/compaction's merge processing logic which allowed `Put`s covered by range tombstones to reappear. Note `Put`s may exist even if the user only ever called `Merge()` due to an internal conversion during compaction to the bottommost level.
-
-## 6.2.0 (4/30/2019)
-### New Features
-* Add an option `strict_bytes_per_sync` that causes a file-writing thread to block rather than exceed the limit on bytes pending writeback specified by `bytes_per_sync` or `wal_bytes_per_sync`.
-* Improve range scan performance by avoiding per-key upper bound check in BlockBasedTableIterator.
-* Introduce Periodic Compaction for Level style compaction. Files are re-compacted periodically and put in the same level.
-* Block-based table index now contains exact highest key in the file, rather than an upper bound. This may improve Get() and iterator Seek() performance in some situations, especially when direct IO is enabled and block cache is disabled. A setting BlockBasedTableOptions::index_shortening is introduced to control this behavior. Set it to kShortenSeparatorsAndSuccessor to get the old behavior.
-* When reading from option file/string/map, customized envs can be filled according to object registry.
-* Add an option `snap_refresh_nanos` (default to 0.5s) to periodically refresh the snapshot list in compaction jobs. Assign to 0 to disable the feature.
-* Improve range scan performance when using explicit user readahead by not creating new table readers for every iterator.
-
-### Public API Change
-* Change the behavior of OptimizeForPointLookup(): move away from hash-based block-based-table index, and use whole key memtable filtering.
-* Change the behavior of OptimizeForSmallDb(): use a 16MB block cache, put index and filter blocks into it, and cost the memtable size to it. DBOptions.OptimizeForSmallDb() and ColumnFamilyOptions.OptimizeForSmallDb() start to take an optional cache object.
-* Added BottommostLevelCompaction::kForceOptimized to avoid double compacting newly compacted files in the bottommost level compaction of manual compaction. Note this option may prohibit the manual compaction to produce a single file in the bottommost level.
-
-### Bug Fixes
-* Adjust WriteBufferManager's dummy entry size to block cache from 1MB to 256KB.
-* Fix a race condition between WritePrepared::Get and ::Put with duplicate keys.
-* Fix crash when memtable prefix bloom is enabled and read/write a key out of domain of prefix extractor.
-* Close a WAL file before another thread deletes it.
-* Fix an assertion failure `IsFlushPending() == true` caused by one bg thread releasing the db mutex in ~ColumnFamilyData and another thread clearing `flush_requested_` flag.
-
-## 6.1.1 (4/9/2019)
-### New Features
-* When reading from option file/string/map, customized comparators and/or merge operators can be filled according to object registry.
-
-### Public API Change
-
-### Bug Fixes
-* Fix a bug in 2PC where a sequence of txn prepare, memtable flush, and crash could result in losing the prepared transaction.
-* Fix a bug in Encryption Env which could cause encrypted files to be read beyond file boundaries.
-
-## 6.1.0 (3/27/2019)
-### New Features
-* Introduce two more stats levels, kExceptHistogramOrTimers and kExceptTimers.
-* Added a feature to perform data-block sampling for compressibility, and report stats to user.
-* Add support for trace filtering.
-* Add DBOptions.avoid_unnecessary_blocking_io. If true, we avoid file deletion when destorying ColumnFamilyHandle and Iterator. Instead, a job is scheduled to delete the files in background.
-
-### Public API Change
-* Remove bundled fbson library.
-* statistics.stats_level_ becomes atomic. It is preferred to use statistics.set_stats_level() and statistics.get_stats_level() to access it.
-* Introduce a new IOError subcode, PathNotFound, to indicate trying to open a nonexistent file or directory for read.
-* Add initial support for multiple db instances sharing the same data in single-writer, multi-reader mode.
-* Removed some "using std::xxx" from public headers.
-
-### Bug Fixes
-* Fix JEMALLOC_CXX_THROW macro missing from older Jemalloc versions, causing build failures on some platforms.
-* Fix SstFileReader not able to open file ingested with write_glbal_seqno=true.
-
-## 6.0.0 (2/19/2019)
-### New Features
-* Enabled checkpoint on readonly db (DBImplReadOnly).
-* Make DB ignore dropped column families while committing results of atomic flush.
-* RocksDB may choose to preopen some files even if options.max_open_files != -1. This may make DB open slightly longer.
-* For users of dictionary compression with ZSTD v0.7.0+, we now reuse the same digested dictionary when compressing each of an SST file's data blocks for faster compression speeds.
-* For all users of dictionary compression who set `cache_index_and_filter_blocks == true`, we now store dictionary data used for decompression in the block cache for better control over memory usage. For users of ZSTD v1.1.4+ who compile with -DZSTD_STATIC_LINKING_ONLY, this includes a digested dictionary, which is used to increase decompression speed.
-* Add support for block checksums verification for external SST files before ingestion.
-* Introduce stats history which periodically saves Statistics snapshots and added `GetStatsHistory` API to retrieve these snapshots.
-* Add a place holder in manifest which indicate a record from future that can be safely ignored.
-* Add support for trace sampling.
-* Enable properties block checksum verification for block-based tables.
-* For all users of dictionary compression, we now generate a separate dictionary for compressing each bottom-level SST file. Previously we reused a single dictionary for a whole compaction to bottom level. The new approach achieves better compression ratios; however, it uses more memory and CPU for buffering/sampling data blocks and training dictionaries.
-* Add whole key bloom filter support in memtable.
-* Files written by `SstFileWriter` will now use dictionary compression if it is configured in the file writer's `CompressionOptions`.
-
-### Public API Change
-* Disallow CompactionFilter::IgnoreSnapshots() = false, because it is not very useful and the behavior is confusing. The filter will filter everything if there is no snapshot declared by the time the compaction starts. However, users can define a snapshot after the compaction starts and before it finishes and this new snapshot won't be repeatable, because after the compaction finishes, some keys may be dropped.
-* CompactionPri = kMinOverlappingRatio also uses compensated file size, which boosts file with lots of tombstones to be compacted first.
-* Transaction::GetForUpdate is extended with a do_validate parameter with default value of true. If false it skips validating the snapshot before doing the read. Similarly ::Merge, ::Put, ::Delete, and ::SingleDelete are extended with assume_tracked with default value of false. If true it indicates that call is assumed to be after a ::GetForUpdate.
-* `TableProperties::num_entries` and `TableProperties::num_deletions` now also account for number of range tombstones.
-* Remove geodb, spatial_db, document_db, json_document, date_tiered_db, and redis_lists.
-* With "ldb ----try_load_options", when wal_dir specified by the option file doesn't exist, ignore it.
-* Change time resolution in FileOperationInfo.
-* Deleting Blob files also go through SStFileManager.
-* Remove CuckooHash memtable.
-* The counter stat `number.block.not_compressed` now also counts blocks not compressed due to poor compression ratio.
-* Remove ttl option from `CompactionOptionsFIFO`. The option has been deprecated and ttl in `ColumnFamilyOptions` is used instead.
-* Support SST file ingestion across multiple column families via DB::IngestExternalFiles. See the function's comment about atomicity.
-* Remove Lua compaction filter.
-
-### Bug Fixes
-* Fix a deadlock caused by compaction and file ingestion waiting for each other in the event of write stalls.
-* Fix a memory leak when files with range tombstones are read in mmap mode and block cache is enabled
-* Fix handling of corrupt range tombstone blocks such that corruptions cannot cause deleted keys to reappear
-* Lock free MultiGet
-* Fix incorrect `NotFound` point lookup result when querying the endpoint of a file that has been extended by a range tombstone.
-* Fix with pipelined write, write leaders's callback failure lead to the whole write group fail.
-
-### Change Default Options
-* Change options.compaction_pri's default to kMinOverlappingRatio
-
-## 5.18.0 (11/30/2018)
-### New Features
-* Introduced `JemallocNodumpAllocator` memory allocator. When being use, block cache will be excluded from core dump.
-* Introduced `PerfContextByLevel` as part of `PerfContext` which allows storing perf context at each level. Also replaced `__thread` with `thread_local` keyword for perf_context. Added per-level perf context for bloom filter and `Get` query.
-* With level_compaction_dynamic_level_bytes = true, level multiplier may be adjusted automatically when Level 0 to 1 compaction is lagged behind.
-* Introduced DB option `atomic_flush`. If true, RocksDB supports flushing multiple column families and atomically committing the result to MANIFEST. Useful when WAL is disabled.
-* Added `num_deletions` and `num_merge_operands` members to `TableProperties`.
-* Added "rocksdb.min-obsolete-sst-number-to-keep" DB property that reports the lower bound on SST file numbers that are being kept from deletion, even if the SSTs are obsolete.
-* Add xxhash64 checksum support
-* Introduced `MemoryAllocator`, which lets the user specify custom memory allocator for block based table.
-* Improved `DeleteRange` to prevent read performance degradation. The feature is no longer marked as experimental.
-
-### Public API Change
-* `DBOptions::use_direct_reads` now affects reads issued by `BackupEngine` on the database's SSTs.
-* `NO_ITERATORS` is divided into two counters `NO_ITERATOR_CREATED` and `NO_ITERATOR_DELETE`. Both of them are only increasing now, just as other counters.
+* Fix slow flush/compaction when DB contains many snapshots. The problem became noticeable to us in DBs with 100,000+ snapshots, though it will affect others at different thresholds.
+* Properly set the stop key for a truncated manual CompactRange

+## 5.16.4 (10/10/2018)
 ### Bug Fixes
 * Fix corner case where a write group leader blocked due to write stall blocks other writers in queue with WriteOptions::no_slowdown set.
-* Fix in-memory range tombstone truncation to avoid erroneously covering newer keys at a lower level, and include range tombstones in compacted files whose largest key is the range tombstone's start key.
-* Properly set the stop key for a truncated manual CompactRange
-* Fix slow flush/compaction when DB contains many snapshots. The problem became noticeable to us in DBs with 100,000+ snapshots, though it will affect others at different thresholds.
-* Fix the bug that WriteBatchWithIndex's SeekForPrev() doesn't see the entries with the same key.
-* Fix the bug where user comparator was sometimes fed with InternalKey instead of the user key. The bug manifests when during GenerateBottommostFiles.
-* Fix a bug in WritePrepared txns where if the number of old snapshots goes beyond the snapshot cache size (128 default) the rest will not be checked when evicting a commit entry from the commit cache.
-* Fixed Get correctness bug in the presence of range tombstones where merge operands covered by a range tombstone always result in NotFound.
-* Start populating `NO_FILE_CLOSES` ticker statistic, which was always zero previously.
-* The default value of NewBloomFilterPolicy()'s argument use_block_based_builder is changed to false. Note that this new default may cause large temp memory usage when building very large SST files.
-
-## 5.17.0 (10/05/2018)
-### Public API Change
-* `OnTableFileCreated` will now be called for empty files generated during compaction. In that case, `TableFileCreationInfo::file_path` will be "(nil)" and `TableFileCreationInfo::file_size` will be zero.
-* Add `FlushOptions::allow_write_stall`, which controls whether Flush calls start working immediately, even if it causes user writes to stall, or will wait until flush can be performed without causing write stall (similar to `CompactRangeOptions::allow_write_stall`). Note that the default value is false, meaning we add delay to Flush calls until stalling can be avoided when possible. This is behavior change compared to previous RocksDB versions, where Flush calls didn't check if they might cause stall or not.
-* Application using PessimisticTransactionDB is expected to rollback/commit recovered transactions before starting new ones. This assumption is used to skip concurrency control during recovery.
-* Expose column family id to `OnCompactionCompleted`.
-
-### New Features
-* TransactionOptions::skip_concurrency_control allows pessimistic transactions to skip the overhead of concurrency control. Could be used for optimizing certain transactions or during recovery.
-
-### Bug Fixes
-* Avoid creating empty SSTs and subsequently deleting them in certain cases during compaction.
-* Sync CURRENT file contents during checkpoint.

 ## 5.16.3 (10/1/2018)
 ### Bug Fixes
@@ -165,9 +29,11 @@

 ## 5.16.0 (8/21/2018)
 ### Public API Change
+* `OnTableFileCreated` will now be called for empty files generated during compaction. In that case, `TableFileCreationInfo::file_path` will be "(nil)" and `TableFileCreationInfo::file_size` will be zero.
+* Add `FlushOptions::allow_write_stall`, which controls whether Flush calls start working immediately, even if it causes user writes to stall, or will wait until flush can be performed without causing write stall (similar to `CompactRangeOptions::allow_write_stall`). Note that the default value is false, meaning we add delay to Flush calls until stalling can be avoided when possible. This is behavior change compared to previous RocksDB versions, where Flush calls didn't check if they might cause stall or not.
+
 * The merge operands are passed to `MergeOperator::ShouldMerge` in the reversed order relative to how they were merged (passed to FullMerge or FullMergeV2) for performance reasons
 * GetAllKeyVersions() to take an extra argument of `max_num_ikeys`.
-* Using ZSTD dictionary trainer (i.e., setting `CompressionOptions::zstd_max_train_bytes` to a nonzero value) now requires ZSTD version 1.1.3 or later.

 ### New Features
 * Changes the format of index blocks by delta encoding the index values, which are the block handles. This saves the encoding of BlockHandle::offset of the non-head index entries in each restart interval. The feature is backward compatible but not forward compatible. It is disabled by default unless format_version 4 or above is used.
@@ -176,6 +42,7 @@

 ### Bug Fixes
 * Fix a bug in misreporting the estimated partition index size in properties block.
+* Avoid creating empty SSTs and subsequently deleting them in certain cases during compaction.

 ## 5.15.0 (7/17/2018)
 ### Public API Change
@@ -311,8 +178,7 @@
 * `BackupableDBOptions::max_valid_backups_to_open == 0` now means no backups will be opened during BackupEngine initialization. Previously this condition disabled limiting backups opened.
 * `DBOptions::preserve_deletes` is a new option that allows one to specify that DB should not drop tombstones for regular deletes if they have sequence number larger than what was set by the new API call `DB::SetPreserveDeletesSequenceNumber(SequenceNumber seqnum)`. Disabled by default.
 * API call `DB::SetPreserveDeletesSequenceNumber(SequenceNumber seqnum)` was added, users who wish to preserve deletes are expected to periodically call this function to advance the cutoff seqnum (all deletes made before this seqnum can be dropped by DB). It's user responsibility to figure out how to advance the seqnum in the way so the tombstones are kept for the desired period of time, yet are eventually processed in time and don't eat up too much space.
-* `ReadOptions::iter_start_seqnum` was added;
-if set to something > 0 user will see 2 changes in iterators behavior 1) only keys written with sequence larger than this parameter would be returned and 2) the `Slice` returned by iter->key() now points to the memory that keep User-oriented representation of the internal key, rather than user key. New struct `FullKey` was added to represent internal keys, along with a new helper function `ParseFullKey(const Slice& internal_key, FullKey* result);`.
+* `ReadOptions::iter_start_seqnum` was added; if set to something > 0 user will see 2 changes in iterators behavior 1) only keys written with sequence larger than this parameter would be returned and 2) the `Slice` returned by iter->key() now points to the memory that keep User-oriented representation of the internal key, rather than user key. New struct `FullKey` was added to represent internal keys, along with a new helper function `ParseFullKey(const Slice& internal_key, FullKey* result);`.
 * Deprecate trash_dir param in NewSstFileManager, right now we will rename deleted files to <name>.trash instead of moving them to trash directory
 * Allow setting a custom trash/DB size ratio limit in the SstFileManager, after which files that are to be scheduled for deletion are deleted immediately, regardless of any delete ratelimit.
 * Return an error on write if write_options.sync = true and write_options.disableWAL = true to warn user of inconsistent options. Previously we will not write to WAL and not respecting the sync options in this case.
@@ -43,8 +43,6 @@ to build a portable binary, add `PORTABLE=1` before your make commands, like thi
      command line flags processing. You can compile rocksdb library even
      if you don't have gflags installed.

-* If you wish to build the RocksJava static target, then cmake is required for building Snappy.
-
 ## Supported platforms

 * **Linux - Ubuntu**
@@ -1,9 +1,7 @@
 This is the list of all known third-party language bindings for RocksDB. If something is missing, please open a pull request to add it.

 * Java - https://github.com/facebook/rocksdb/tree/master/java
-* Python
-    * http://python-rocksdb.readthedocs.io/en/latest/
-    * http://pyrocksdb.readthedocs.org/en/latest/ (unmaintained)
+* Python - http://pyrocksdb.readthedocs.org/en/latest/
 * Perl - https://metacpan.org/pod/RocksDB
 * Node.js - https://npmjs.org/package/rocksdb
 * Go - https://github.com/tecbot/gorocksdb
@@ -93,38 +93,19 @@ ifeq ($(MAKECMDGOALS),rocksdbjavastaticpublish)
 	DEBUG_LEVEL=0
 endif

-# Lite build flag.
-LITE ?= 0
-ifeq ($(LITE), 0)
-ifneq ($(filter -DROCKSDB_LITE,$(OPT)),)
-  # Be backward compatible and support older format where OPT=-DROCKSDB_LITE is
-  # specified instead of LITE=1 on the command line.
-  LITE=1
-endif
-else ifeq ($(LITE), 1)
-ifeq ($(filter -DROCKSDB_LITE,$(OPT)),)
-	OPT += -DROCKSDB_LITE
-endif
-endif
-
-# Figure out optimize level.
-ifneq ($(DEBUG_LEVEL), 2)
-ifeq ($(LITE), 0)
-	OPT += -O2
-else
-	OPT += -Os
-endif
-endif
-
 # compile with -O2 if debug level is not 2
 ifneq ($(DEBUG_LEVEL), 2)
-OPT += -fno-omit-frame-pointer
+OPT += -O2 -fno-omit-frame-pointer
 # Skip for archs that don't support -momit-leaf-frame-pointer
 ifeq (,$(shell $(CXX) -fsyntax-only -momit-leaf-frame-pointer -xc /dev/null 2>&1))
 OPT += -momit-leaf-frame-pointer
 endif
 endif

+ifeq (,$(shell $(CXX) -fsyntax-only -faligned-new -xc++ /dev/null 2>&1))
+CXXFLAGS += -faligned-new -DHAVE_ALIGNED_NEW
+endif
+
 ifeq (,$(shell $(CXX) -fsyntax-only -maltivec -xc /dev/null 2>&1))
 CXXFLAGS += -DHAS_ALTIVEC
 CFLAGS += -DHAS_ALTIVEC
@@ -137,12 +118,6 @@ CFLAGS +=  -DHAVE_POWER8
 HAVE_POWER8=1
 endif

-ifeq (,$(shell $(CXX) -fsyntax-only -march=armv8-a+crc -xc /dev/null 2>&1))
-CXXFLAGS += -march=armv8-a+crc
-CFLAGS += -march=armv8-a+crc
-ARMCRC_SOURCE=1
-endif
-
 # if we're compiling for release, compile without debug code (-DNDEBUG)
 ifeq ($(DEBUG_LEVEL),0)
 OPT += -DNDEBUG
@@ -204,7 +179,7 @@ include make_config.mk
 CLEAN_FILES += make_config.mk

 missing_make_config_paths := $(shell				\
-	grep "\./\S*\|/\S*" -o $(CURDIR)/make_config.mk | 	\
+	grep "\/\S*" -o $(CURDIR)/make_config.mk | 		\
 	while read path;					\
 		do [ -e $$path ] || echo $$path; 		\
 	done | sort | uniq)
@@ -347,7 +322,7 @@ endif
 ifeq ("$(wildcard $(LUA_LIB))", "") # LUA_LIB does not exist
 $(error $(LUA_LIB) does not exist.  Try to specify both LUA_PATH and LUA_LIB manually)
 endif
-EXEC_LDFLAGS += $(LUA_LIB)
+LDFLAGS += $(LUA_LIB)

 endif

@@ -409,7 +384,7 @@ BENCHTOOLOBJECTS = $(BENCH_LIB_SOURCES:.cc=.o) $(LIBOBJECTS) $(TESTUTIL)

 ANALYZETOOLOBJECTS = $(ANALYZER_LIB_SOURCES:.cc=.o)

-EXPOBJECTS = $(LIBOBJECTS) $(TESTUTIL)
+EXPOBJECTS = $(EXP_LIB_SOURCES:.cc=.o) $(LIBOBJECTS) $(TESTUTIL)

 TESTS = \
 	db_basic_test \
@@ -449,7 +424,6 @@ TESTS = \
 	db_merge_operator_test \
 	db_options_test \
 	db_range_del_test \
-	db_secondary_test \
 	db_sst_test \
 	db_tailing_iter_test \
 	db_io_failure_test \
@@ -489,6 +463,7 @@ TESTS = \
 	merger_test \
 	util_merge_operators_test \
 	options_file_test \
+	redis_test \
 	reduce_levels_test \
 	plain_table_db_test \
 	comparator_db_test \
@@ -502,8 +477,12 @@ TESTS = \
 	cassandra_row_merge_test \
 	cassandra_serialize_test \
 	ttl_test \
+	date_tiered_test \
 	backupable_db_test \
+	document_db_test \
+	json_document_test \
 	sim_cache_test \
+	spatial_db_test \
 	version_edit_test \
 	version_set_test \
 	compaction_picker_test \
@@ -515,6 +494,7 @@ TESTS = \
 	deletefile_test \
 	obsolete_files_test \
 	table_test \
+	geodb_test \
 	delete_scheduler_test \
 	options_test \
 	options_settable_test \
@@ -531,6 +511,7 @@ TESTS = \
 	compaction_job_test \
 	thread_list_test \
 	sst_dump_test \
+	column_aware_encoding_test \
 	compact_files_test \
 	optimistic_transaction_test \
 	write_callback_test \
@@ -542,6 +523,8 @@ TESTS = \
 	ldb_cmd_test \
 	persistent_cache_test \
 	statistics_test \
+	lua_test \
+	range_del_aggregator_test \
 	lru_cache_test \
 	object_registry_test \
 	repair_test \
@@ -550,11 +533,6 @@ TESTS = \
 	write_unprepared_transaction_test \
 	db_universal_compaction_test \
 	trace_analyzer_test \
-	repeatable_thread_test \
-	range_tombstone_fragmenter_test \
-	range_del_aggregator_test \
-	sst_file_reader_test \
-	db_secondary_test \

 PARALLEL_TEST = \
 	backupable_db_test \
@@ -604,7 +582,7 @@ TEST_LIBS = \
 	librocksdb_env_basic_test.a

 # TODO: add back forward_iterator_bench, after making it build in all environemnts.
-BENCHMARKS = db_bench table_reader_bench cache_bench memtablerep_bench persistent_cache_bench range_del_aggregator_bench
+BENCHMARKS = db_bench table_reader_bench cache_bench memtablerep_bench column_aware_encoding_exp persistent_cache_bench

 # if user didn't config LIBNAME, set the default
 ifeq ($(LIBNAME),)
@@ -694,8 +672,7 @@ endif  # PLATFORM_SHARED_EXT
 .PHONY: blackbox_crash_test check clean coverage crash_test ldb_tests package \
 	release tags tags0 valgrind_check whitebox_crash_test format static_lib shared_lib all \
 	dbg rocksdbjavastatic rocksdbjava install install-static install-shared uninstall \
-	analyze tools tools_lib \
-	blackbox_crash_test_with_atomic_flush whitebox_crash_test_with_atomic_flush
+	analyze tools tools_lib


 all: $(LIBRARY) $(BENCHMARKS) tools tools_lib test_libs $(TESTS)
@@ -825,6 +802,7 @@ J ?= 100%
 # Use this regexp to select the subset of tests whose names match.
 tests-regexp = .

+t_run = $(wildcard t/run-*)
 .PHONY: check_0
 check_0:
 	$(AM_V_GEN)export TEST_TMPDIR=$(TMPD); \
@@ -834,7 +812,7 @@ check_0:
 	test -t 1 && eta=--eta || eta=; \
 	{ \
 		printf './%s\n' $(filter-out $(PARALLEL_TEST),$(TESTS)); \
-		find t -name 'run-*' -print; \
+		printf '%s\n' $(t_run); \
 	} \
 	  | $(prioritize_long_running_tests)				\
 	  | grep -E '$(tests-regexp)'					\
@@ -851,7 +829,7 @@ valgrind_check_0:
 	test -t 1 && eta=--eta || eta=;					\
 	{								\
 	  printf './%s\n' $(filter-out $(PARALLEL_TEST) %skiplist_test options_settable_test, $(TESTS));		\
-	  find t -name 'run-*' -print; \
+	  printf '%s\n' $(t_run);					\
 	}								\
 	  | $(prioritize_long_running_tests)				\
 	  | grep -E '$(tests-regexp)'					\
@@ -903,15 +881,10 @@ ldb_tests: ldb

 crash_test: whitebox_crash_test blackbox_crash_test

-crash_test_with_atomic_flush: whitebox_crash_test_with_atomic_flush blackbox_crash_test_with_atomic_flush
-
 blackbox_crash_test: db_stress
 	python -u tools/db_crashtest.py --simple blackbox $(CRASH_TEST_EXT_ARGS)
 	python -u tools/db_crashtest.py blackbox $(CRASH_TEST_EXT_ARGS)

-blackbox_crash_test_with_atomic_flush: db_stress
-	python -u tools/db_crashtest.py --enable_atomic_flush blackbox $(CRASH_TEST_EXT_ARGS)
-
 ifeq ($(CRASH_TEST_KILL_ODD),)
  CRASH_TEST_KILL_ODD=888887
 endif
@@ -922,10 +895,6 @@ whitebox_crash_test: db_stress
 	python -u tools/db_crashtest.py whitebox  --random_kill_odd \
      $(CRASH_TEST_KILL_ODD) $(CRASH_TEST_EXT_ARGS)

-whitebox_crash_test_with_atomic_flush: db_stress
-	python -u tools/db_crashtest.py --enable_atomic_flush whitebox  --random_kill_odd \
-      $(CRASH_TEST_KILL_ODD) $(CRASH_TEST_EXT_ARGS)
-
 asan_check:
 	$(MAKE) clean
 	COMPILE_WITH_ASAN=1 $(MAKE) check -j32
@@ -936,11 +905,6 @@ asan_crash_test:
 	COMPILE_WITH_ASAN=1 $(MAKE) crash_test
 	$(MAKE) clean

-asan_crash_test_with_atomic_flush:
-	$(MAKE) clean
-	COMPILE_WITH_ASAN=1 $(MAKE) crash_test_with_atomic_flush
-	$(MAKE) clean
-
 ubsan_check:
 	$(MAKE) clean
 	COMPILE_WITH_UBSAN=1 $(MAKE) check -j32
@@ -951,11 +915,6 @@ ubsan_crash_test:
 	COMPILE_WITH_UBSAN=1 $(MAKE) crash_test
 	$(MAKE) clean

-ubsan_crash_test_with_atomic_flush:
-	$(MAKE) clean
-	COMPILE_WITH_UBSAN=1 $(MAKE) crash_test_with_atomic_flush
-	$(MAKE) clean
-
 valgrind_test:
 	ROCKSDB_VALGRIND_RUN=1 DISABLE_JEMALLOC=1 $(MAKE) valgrind_check

@@ -1043,10 +1002,8 @@ unity.a: unity.o
 	$(AM_V_AR)rm -f $@
 	$(AM_V_at)$(AR) $(ARFLAGS) $@ unity.o

-
-TOOLLIBOBJECTS = $(TOOL_LIB_SOURCES:.cc=.o)
 # try compiling db_test with unity
-unity_test: db/db_test.o db/db_test_util.o $(TESTHARNESS) $(TOOLLIBOBJECTS) unity.a
+unity_test: db/db_test.o db/db_test_util.o $(TESTHARNESS) unity.a
 	$(AM_LINK)
 	./unity_test

@@ -1169,6 +1126,9 @@ cassandra_row_merge_test: utilities/cassandra/cassandra_row_merge_test.o utiliti
 cassandra_serialize_test: utilities/cassandra/cassandra_serialize_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

+redis_test: utilities/redis/redis_lists_test.o $(LIBOBJECTS) $(TESTHARNESS)
+	$(AM_LINK)
+
 hash_table_test: utilities/persistent_cache/hash_table_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

@@ -1307,9 +1267,18 @@ backupable_db_test: utilities/backupable/backupable_db_test.o $(LIBOBJECTS) $(TE
 checkpoint_test: utilities/checkpoint/checkpoint_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

+document_db_test: utilities/document/document_db_test.o $(LIBOBJECTS) $(TESTHARNESS)
+	$(AM_LINK)
+
+json_document_test: utilities/document/json_document_test.o $(LIBOBJECTS) $(TESTHARNESS)
+	$(AM_LINK)
+
 sim_cache_test: utilities/simulator_cache/sim_cache_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

+spatial_db_test: utilities/spatialdb/spatial_db_test.o $(LIBOBJECTS) $(TESTHARNESS)
+	$(AM_LINK)
+
 env_mirror_test: utilities/env_mirror_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

@@ -1327,6 +1296,9 @@ object_registry_test: utilities/object_registry_test.o $(LIBOBJECTS) $(TESTHARNE
 ttl_test: utilities/ttl/ttl_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

+date_tiered_test: utilities/date_tiered/date_tiered_test.o $(LIBOBJECTS) $(TESTHARNESS)
+	$(AM_LINK)
+
 write_batch_with_index_test: utilities/write_batch_with_index/write_batch_with_index_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

@@ -1453,6 +1425,9 @@ deletefile_test: db/deletefile_test.o $(LIBOBJECTS) $(TESTHARNESS)
 obsolete_files_test: db/obsolete_files_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

+geodb_test: utilities/geodb/geodb_test.o $(LIBOBJECTS) $(TESTHARNESS)
+	$(AM_LINK)
+
 rocksdb_dump: tools/dump/rocksdb_dump.o $(LIBOBJECTS)
 	$(AM_LINK)

@@ -1501,6 +1476,9 @@ timer_queue_test: util/timer_queue_test.o $(LIBOBJECTS) $(TESTHARNESS)
 sst_dump_test: tools/sst_dump_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

+column_aware_encoding_test: utilities/column_aware_encoding_test.o $(TESTHARNESS) $(EXPOBJECTS)
+	$(AM_LINK)
+
 optimistic_transaction_test: utilities/transactions/optimistic_transaction_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

@@ -1540,6 +1518,9 @@ sst_dump: tools/sst_dump.o $(LIBOBJECTS)
 blob_dump: tools/blob_dump.o $(LIBOBJECTS)
 	$(AM_LINK)

+column_aware_encoding_exp: utilities/column_aware_encoding_exp.o $(EXPOBJECTS)
+	$(AM_LINK)
+
 repair_test: db/repair_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

@@ -1561,27 +1542,15 @@ statistics_test: monitoring/statistics_test.o $(LIBOBJECTS) $(TESTHARNESS)
 lru_cache_test: cache/lru_cache_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

+lua_test: utilities/lua/rocks_lua_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
+	$(AM_LINK)
+
 range_del_aggregator_test: db/range_del_aggregator_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

-range_del_aggregator_bench: db/range_del_aggregator_bench.o $(LIBOBJECTS) $(TESTUTIL)
-	$(AM_LINK)
-
 blob_db_test: utilities/blob_db/blob_db_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)

-repeatable_thread_test: util/repeatable_thread_test.o $(LIBOBJECTS) $(TESTHARNESS)
-	$(AM_LINK)
-
-range_tombstone_fragmenter_test: db/range_tombstone_fragmenter_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
-	$(AM_LINK)
-
-sst_file_reader_test: table/sst_file_reader_test.o $(LIBOBJECTS) $(TESTHARNESS)
-	$(AM_LINK)
-
-db_secondary_test: db/db_secondary_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
-	$(AM_LINK)
-
 #-------------------------------------------------
 # make install related stuff
 INSTALL_PATH ?= /usr/local
@@ -1653,14 +1622,14 @@ ZLIB_DOWNLOAD_BASE ?= http://zlib.net
 BZIP2_VER ?= 1.0.6
 BZIP2_SHA256 ?= a2848f34fcd5d6cf47def00461fcb528a0484d8edef8208d6d2e2909dc61d9cd
 BZIP2_DOWNLOAD_BASE ?= https://web.archive.org/web/20180624184835/http://www.bzip.org
-SNAPPY_VER ?= 1.1.7
-SNAPPY_SHA256 ?= 3dfa02e873ff51a11ee02b9ca391807f0c8ea0529a4924afa645fbf97163f9d4
-SNAPPY_DOWNLOAD_BASE ?= https://github.com/google/snappy/archive
-LZ4_VER ?= 1.8.3
-LZ4_SHA256 ?= 33af5936ac06536805f9745e0b6d61da606a1f8b4cc5c04dd3cbaca3b9b4fc43
+SNAPPY_VER ?= 1.1.4
+SNAPPY_SHA256 ?= 134bfe122fd25599bb807bb8130e7ba6d9bdb851e0b16efcb83ac4f5d0b70057
+SNAPPY_DOWNLOAD_BASE ?= https://github.com/google/snappy/releases/download
+LZ4_VER ?= 1.8.0
+LZ4_SHA256 ?= 2ca482ea7a9bb103603108b5a7510b7592b90158c151ff50a28f1ca8389fccf6
 LZ4_DOWNLOAD_BASE ?= https://github.com/lz4/lz4/archive
-ZSTD_VER ?= 1.3.7
-ZSTD_SHA256 ?= 5dd1e90eb16c25425880c8a91327f63de22891ffed082fcc17e5ae84fce0d5fb
+ZSTD_VER ?= 1.3.3
+ZSTD_SHA256 ?= a77c47153ee7de02626c5b2a097005786b71688be61e9fb81806a011f90b297b
 ZSTD_DOWNLOAD_BASE ?= https://github.com/facebook/zstd/archive
 CURL_SSL_OPTS ?= --tlsv1

@@ -1699,9 +1668,7 @@ endif

 libz.a:
 	-rm -rf zlib-$(ZLIB_VER)
-ifeq (,$(wildcard ./zlib-$(ZLIB_VER).tar.gz))
-	curl --output zlib-$(ZLIB_VER).tar.gz -L ${ZLIB_DOWNLOAD_BASE}/zlib-$(ZLIB_VER).tar.gz
-endif
+	curl -O -L ${ZLIB_DOWNLOAD_BASE}/zlib-$(ZLIB_VER).tar.gz
 	ZLIB_SHA256_ACTUAL=`$(SHA256_CMD) zlib-$(ZLIB_VER).tar.gz | cut -d ' ' -f 1`; \
 	if [ "$(ZLIB_SHA256)" != "$$ZLIB_SHA256_ACTUAL" ]; then \
 		echo zlib-$(ZLIB_VER).tar.gz checksum mismatch, expected=\"$(ZLIB_SHA256)\" actual=\"$$ZLIB_SHA256_ACTUAL\"; \
@@ -1713,9 +1680,7 @@ endif

 libbz2.a:
 	-rm -rf bzip2-$(BZIP2_VER)
-ifeq (,$(wildcard ./bzip2-$(BZIP2_VER).tar.gz))
-	curl --output bzip2-$(BZIP2_VER).tar.gz -L ${BZIP2_DOWNLOAD_BASE}/$(BZIP2_VER)/bzip2-$(BZIP2_VER).tar.gz
-endif
+	curl -O -L ${BZIP2_DOWNLOAD_BASE}/$(BZIP2_VER)/bzip2-$(BZIP2_VER).tar.gz
 	BZIP2_SHA256_ACTUAL=`$(SHA256_CMD) bzip2-$(BZIP2_VER).tar.gz | cut -d ' ' -f 1`; \
 	if [ "$(BZIP2_SHA256)" != "$$BZIP2_SHA256_ACTUAL" ]; then \
 		echo bzip2-$(BZIP2_VER).tar.gz checksum mismatch, expected=\"$(BZIP2_SHA256)\" actual=\"$$BZIP2_SHA256_ACTUAL\"; \
@@ -1727,24 +1692,21 @@ endif

 libsnappy.a:
 	-rm -rf snappy-$(SNAPPY_VER)
-ifeq (,$(wildcard ./snappy-$(SNAPPY_VER).tar.gz))
-	curl --output snappy-$(SNAPPY_VER).tar.gz -L ${CURL_SSL_OPTS} ${SNAPPY_DOWNLOAD_BASE}/$(SNAPPY_VER).tar.gz
-endif
+	curl -O -L ${CURL_SSL_OPTS} ${SNAPPY_DOWNLOAD_BASE}/$(SNAPPY_VER)/snappy-$(SNAPPY_VER).tar.gz
 	SNAPPY_SHA256_ACTUAL=`$(SHA256_CMD) snappy-$(SNAPPY_VER).tar.gz | cut -d ' ' -f 1`; \
 	if [ "$(SNAPPY_SHA256)" != "$$SNAPPY_SHA256_ACTUAL" ]; then \
 		echo snappy-$(SNAPPY_VER).tar.gz checksum mismatch, expected=\"$(SNAPPY_SHA256)\" actual=\"$$SNAPPY_SHA256_ACTUAL\"; \
 		exit 1; \
 	fi
 	tar xvzf snappy-$(SNAPPY_VER).tar.gz
-	mkdir snappy-$(SNAPPY_VER)/build
-	cd snappy-$(SNAPPY_VER)/build && CFLAGS='${EXTRA_CFLAGS}' CXXFLAGS='${EXTRA_CXXFLAGS}' LDFLAGS='${EXTRA_LDFLAGS}' cmake -DCMAKE_POSITION_INDEPENDENT_CODE=ON .. && $(MAKE) ${SNAPPY_MAKE_TARGET}
-	cp snappy-$(SNAPPY_VER)/build/libsnappy.a .
+	cd snappy-$(SNAPPY_VER) && CFLAGS='${EXTRA_CFLAGS}' CXXFLAGS='${EXTRA_CXXFLAGS}' LDFLAGS='${EXTRA_LDFLAGS}' ./configure --with-pic --enable-static --disable-shared
+	cd snappy-$(SNAPPY_VER) && $(MAKE) ${SNAPPY_MAKE_TARGET}
+	cp snappy-$(SNAPPY_VER)/.libs/libsnappy.a .

 liblz4.a:
 	-rm -rf lz4-$(LZ4_VER)
-ifeq (,$(wildcard ./lz4-$(LZ4_VER).tar.gz))
-	curl --output lz4-$(LZ4_VER).tar.gz -L ${CURL_SSL_OPTS} ${LZ4_DOWNLOAD_BASE}/v$(LZ4_VER).tar.gz
-endif
+	curl -O -L ${CURL_SSL_OPTS} ${LZ4_DOWNLOAD_BASE}/v$(LZ4_VER).tar.gz
+	mv v$(LZ4_VER).tar.gz lz4-$(LZ4_VER).tar.gz
 	LZ4_SHA256_ACTUAL=`$(SHA256_CMD) lz4-$(LZ4_VER).tar.gz | cut -d ' ' -f 1`; \
 	if [ "$(LZ4_SHA256)" != "$$LZ4_SHA256_ACTUAL" ]; then \
 		echo lz4-$(LZ4_VER).tar.gz checksum mismatch, expected=\"$(LZ4_SHA256)\" actual=\"$$LZ4_SHA256_ACTUAL\"; \
@@ -1756,9 +1718,8 @@ endif

 libzstd.a:
 	-rm -rf zstd-$(ZSTD_VER)
-ifeq (,$(wildcard ./zstd-$(ZSTD_VER).tar.gz))
-	curl --output zstd-$(ZSTD_VER).tar.gz -L ${CURL_SSL_OPTS} ${ZSTD_DOWNLOAD_BASE}/v$(ZSTD_VER).tar.gz
-endif
+	curl -O -L ${CURL_SSL_OPTS} ${ZSTD_DOWNLOAD_BASE}/v$(ZSTD_VER).tar.gz
+	mv v$(ZSTD_VER).tar.gz zstd-$(ZSTD_VER).tar.gz
 	ZSTD_SHA256_ACTUAL=`$(SHA256_CMD) zstd-$(ZSTD_VER).tar.gz | cut -d ' ' -f 1`; \
 	if [ "$(ZSTD_SHA256)" != "$$ZSTD_SHA256_ACTUAL" ]; then \
 		echo zstd-$(ZSTD_VER).tar.gz checksum mismatch, expected=\"$(ZSTD_SHA256)\" actual=\"$$ZSTD_SHA256_ACTUAL\"; \
@@ -1927,8 +1888,7 @@ commit_prereq: build_tools/rocksdb-lego-determinator \
 ifeq ($(PLATFORM), IOS)
 # For iOS, create universal object files to be used on both the simulator and
 # a device.
-XCODEROOT=$(shell xcode-select -print-path)
-PLATFORMSROOT=$(XCODEROOT)/Platforms
+PLATFORMSROOT=/Applications/Xcode.app/Contents/Developer/Platforms
 SIMULATORROOT=$(PLATFORMSROOT)/iPhoneSimulator.platform/Developer
 DEVICEROOT=$(PLATFORMSROOT)/iPhoneOS.platform/Developer
 IOSVERSION=$(shell defaults read $(PLATFORMSROOT)/iPhoneOS.platform/version CFBundleShortVersionString)
@@ -1965,7 +1925,7 @@ endif
 #  	Source files dependencies detection
 # ---------------------------------------------------------------------------

-all_sources = $(LIB_SOURCES) $(MAIN_SOURCES) $(MOCK_LIB_SOURCES) $(TOOL_LIB_SOURCES) $(BENCH_LIB_SOURCES) $(TEST_LIB_SOURCES) $(ANALYZER_LIB_SOURCES)
+all_sources = $(LIB_SOURCES) $(MAIN_SOURCES) $(MOCK_LIB_SOURCES) $(TOOL_LIB_SOURCES) $(BENCH_LIB_SOURCES) $(TEST_LIB_SOURCES) $(EXP_LIB_SOURCES) $(ANALYZER_LIB_SOURCES)
 DEPFILES = $(all_sources:.cc=.cc.d)

 # Add proper dependency support so changing a .h file forces a .cc file to
@@ -5,7 +5,7 @@
 [![PPC64le Build Status](http://140.211.168.68:8080/buildStatus/icon?job=Rocksdb)](http://140.211.168.68:8080/job/Rocksdb)

 RocksDB is developed and maintained by Facebook Database Engineering Team.
-It is built on earlier work on [LevelDB](https://github.com/google/leveldb) by Sanjay Ghemawat (sanjay@google.com)
+It is built on earlier work on LevelDB by Sanjay Ghemawat (sanjay@google.com)
 and Jeff Dean (jeff@google.com)

 This code is a library that forms the core building block for a fast
@@ -1,10 +1,12 @@
 load("@fbcode_macros//build_defs:auto_headers.bzl", "AutoHeaders")
-load("@fbcode_macros//build_defs:cpp_library.bzl", "cpp_library")
-load(":defs.bzl", "test_binary")

 REPO_PATH = package_name() + "/"

-ROCKSDB_COMPILER_FLAGS = [
+BUCK_BINS = "buck-out/gen/" + REPO_PATH
+
+TEST_RUNNER = REPO_PATH + "buckifier/rocks_test_runner.sh"
+
+rocksdb_compiler_flags = [
    "-fno-builtin-memcmp",
    "-DROCKSDB_PLATFORM_POSIX",
    "-DROCKSDB_LIB_IO_POSIX",
@@ -20,7 +22,6 @@ ROCKSDB_COMPILER_FLAGS = [
    "-DBZIP2",
    "-DLZ4",
    "-DZSTD",
-    "-DZSTD_STATIC_LINKING_ONLY",
    "-DGFLAGS=gflags",
    "-DNUMA",
    "-DTBB",
@@ -32,7 +33,7 @@ ROCKSDB_COMPILER_FLAGS = [
    "-Wnarrowing",
 ]

-ROCKSDB_EXTERNAL_DEPS = [
+rocksdb_external_deps = [
    ("bzip2", None, "bz2"),
    ("snappy", None, "snappy"),
    ("zlib", None, "z"),
@@ -44,13 +45,13 @@ ROCKSDB_EXTERNAL_DEPS = [
    ("googletest", None, "gtest"),
 ]

-ROCKSDB_PREPROCESSOR_FLAGS = [
+rocksdb_preprocessor_flags = [
    # Directories with files for #include
    "-I" + REPO_PATH + "include/",
    "-I" + REPO_PATH,
 ]

-ROCKSDB_ARCH_PREPROCESSOR_FLAGS = {
+rocksdb_arch_preprocessor_flags = {
    "x86_64": [
        "-DHAVE_SSE42",
        "-DHAVE_PCLMUL",
@@ -63,15 +64,8 @@ is_opt_mode = build_mode.startswith("opt")

 # -DNDEBUG is added by default in opt mode in fbcode. But adding it twice
 # doesn't harm and avoid forgetting to add it.
-ROCKSDB_COMPILER_FLAGS += (["-DNDEBUG"] if is_opt_mode else [])
-
-sanitizer = read_config("fbcode", "sanitizer")
-
-# Do not enable jemalloc if sanitizer presents. RocksDB will further detect
-# whether the binary is linked with jemalloc at runtime.
-ROCKSDB_COMPILER_FLAGS += (["-DROCKSDB_JEMALLOC"] if sanitizer == "" else [])
-
-ROCKSDB_EXTERNAL_DEPS += ([("jemalloc", None, "headers")] if sanitizer == "" else [])
+if is_opt_mode:
+    rocksdb_compiler_flags.append("-DNDEBUG")

 cpp_library(
    name = "rocksdb_lib",
@@ -87,7 +81,6 @@ cpp_library(
        "db/compaction_iterator.cc",
        "db/compaction_job.cc",
        "db/compaction_picker.cc",
-        "db/compaction_picker_fifo.cc",
        "db/compaction_picker_universal.cc",
        "db/convenience.cc",
        "db/db_filesnapshot.cc",
@@ -98,7 +91,6 @@ cpp_library(
        "db/db_impl_files.cc",
        "db/db_impl_open.cc",
        "db/db_impl_readonly.cc",
-        "db/db_impl_secondary.cc",
        "db/db_impl_write.cc",
        "db/db_info_dumper.cc",
        "db/db_iter.cc",
@@ -111,7 +103,6 @@ cpp_library(
        "db/flush_job.cc",
        "db/flush_scheduler.cc",
        "db/forward_iterator.cc",
-        "db/in_memory_stats_history.cc",
        "db/internal_stats.cc",
        "db/log_reader.cc",
        "db/log_writer.cc",
@@ -122,7 +113,6 @@ cpp_library(
        "db/merge_helper.cc",
        "db/merge_operator.cc",
        "db/range_del_aggregator.cc",
-        "db/range_tombstone_fragmenter.cc",
        "db/repair.cc",
        "db/snapshot_impl.cc",
        "db/table_cache.cc",
@@ -144,6 +134,7 @@ cpp_library(
        "env/io_posix.cc",
        "env/mock_env.cc",
        "memtable/alloc_tracker.cc",
+        "memtable/hash_cuckoo_rep.cc",
        "memtable/hash_linklist_rep.cc",
        "memtable/hash_skiplist_rep.cc",
        "memtable/skiplistrep.cc",
@@ -199,7 +190,6 @@ cpp_library(
        "table/plain_table_index.cc",
        "table/plain_table_key_coding.cc",
        "table/plain_table_reader.cc",
-        "table/sst_file_reader.cc",
        "table/sst_file_writer.cc",
        "table/table_properties.cc",
        "table/two_level_iterator.cc",
@@ -216,7 +206,6 @@ cpp_library(
        "util/comparator.cc",
        "util/compression_context_cache.cc",
        "util/concurrent_arena.cc",
-        "util/concurrent_task_limiter_impl.cc",
        "util/crc32c.cc",
        "util/delete_scheduler.cc",
        "util/dynamic_bloom.cc",
@@ -226,7 +215,6 @@ cpp_library(
        "util/filename.cc",
        "util/filter_policy.cc",
        "util/hash.cc",
-        "util/jemalloc_nodump_allocator.cc",
        "util/log_buffer.cc",
        "util/murmurhash.cc",
        "util/random.cc",
@@ -258,10 +246,16 @@ cpp_library(
        "utilities/checkpoint/checkpoint_impl.cc",
        "utilities/compaction_filters/remove_emptyvalue_compactionfilter.cc",
        "utilities/convenience/info_log_finder.cc",
+        "utilities/date_tiered/date_tiered_db_impl.cc",
        "utilities/debug.cc",
+        "utilities/document/document_db.cc",
+        "utilities/document/json_document.cc",
+        "utilities/document/json_document_builder.cc",
        "utilities/env_mirror.cc",
        "utilities/env_timed.cc",
+        "utilities/geodb/geodb_impl.cc",
        "utilities/leveldb_options/leveldb_options.cc",
+        "utilities/lua/rocks_lua_compaction_filter.cc",
        "utilities/memory/memory_util.cc",
        "utilities/merge_operators/bytesxor.cc",
        "utilities/merge_operators/max.cc",
@@ -276,7 +270,9 @@ cpp_library(
        "utilities/persistent_cache/block_cache_tier_metadata.cc",
        "utilities/persistent_cache/persistent_cache_tier.cc",
        "utilities/persistent_cache/volatile_tier_impl.cc",
+        "utilities/redis/redis_lists.cc",
        "utilities/simulator_cache/sim_cache.cc",
+        "utilities/spatialdb/spatial_db.cc",
        "utilities/table_properties_collectors/compact_on_deletion_collector.cc",
        "utilities/trace/file_trace_reader_writer.cc",
        "utilities/transactions/optimistic_transaction.cc",
@@ -296,12 +292,12 @@ cpp_library(
        "utilities/write_batch_with_index/write_batch_with_index.cc",
        "utilities/write_batch_with_index/write_batch_with_index_internal.cc",
    ],
-    auto_headers = AutoHeaders.RECURSIVE_GLOB,
-    arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
-    compiler_flags = ROCKSDB_COMPILER_FLAGS,
-    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+    headers = AutoHeaders.RECURSIVE_GLOB,
+    arch_preprocessor_flags = rocksdb_arch_preprocessor_flags,
+    compiler_flags = rocksdb_compiler_flags,
+    preprocessor_flags = rocksdb_preprocessor_flags,
    deps = [],
-    external_deps = ROCKSDB_EXTERNAL_DEPS,
+    external_deps = rocksdb_external_deps,
 )

 cpp_library(
@@ -314,39 +310,41 @@ cpp_library(
        "util/testharness.cc",
        "util/testutil.cc",
        "utilities/cassandra/test_utils.cc",
+        "utilities/col_buf_decoder.cc",
+        "utilities/col_buf_encoder.cc",
+        "utilities/column_aware_encoding_util.cc",
    ],
-    auto_headers = AutoHeaders.RECURSIVE_GLOB,
-    arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
-    compiler_flags = ROCKSDB_COMPILER_FLAGS,
-    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+    headers = AutoHeaders.RECURSIVE_GLOB,
+    arch_preprocessor_flags = rocksdb_arch_preprocessor_flags,
+    compiler_flags = rocksdb_compiler_flags,
+    preprocessor_flags = rocksdb_preprocessor_flags,
    deps = [":rocksdb_lib"],
-    external_deps = ROCKSDB_EXTERNAL_DEPS,
+    external_deps = rocksdb_external_deps,
 )

 cpp_library(
    name = "rocksdb_tools_lib",
    srcs = [
        "tools/db_bench_tool.cc",
-        "tools/trace_analyzer_tool.cc",
        "util/testutil.cc",
    ],
-    auto_headers = AutoHeaders.RECURSIVE_GLOB,
-    arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
-    compiler_flags = ROCKSDB_COMPILER_FLAGS,
-    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+    headers = AutoHeaders.RECURSIVE_GLOB,
+    arch_preprocessor_flags = rocksdb_arch_preprocessor_flags,
+    compiler_flags = rocksdb_compiler_flags,
+    preprocessor_flags = rocksdb_preprocessor_flags,
    deps = [":rocksdb_lib"],
-    external_deps = ROCKSDB_EXTERNAL_DEPS,
+    external_deps = rocksdb_external_deps,
 )

 cpp_library(
    name = "env_basic_test_lib",
    srcs = ["env/env_basic_test.cc"],
-    auto_headers = AutoHeaders.RECURSIVE_GLOB,
-    arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
-    compiler_flags = ROCKSDB_COMPILER_FLAGS,
-    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+    headers = AutoHeaders.RECURSIVE_GLOB,
+    arch_preprocessor_flags = rocksdb_arch_preprocessor_flags,
+    compiler_flags = rocksdb_compiler_flags,
+    preprocessor_flags = rocksdb_preprocessor_flags,
    deps = [":rocksdb_test_lib"],
-    external_deps = ROCKSDB_EXTERNAL_DEPS,
+    external_deps = rocksdb_external_deps,
 )

 # [test_name, test_src, test_type]
@@ -436,6 +434,11 @@ ROCKS_TESTS = [
        "util/coding_test.cc",
        "serial",
    ],
+    [
+        "column_aware_encoding_test",
+        "utilities/column_aware_encoding_test.cc",
+        "serial",
+    ],
    [
        "column_family_test",
        "db/column_family_test.cc",
@@ -506,6 +509,11 @@ ROCKS_TESTS = [
        "table/data_block_hash_index_test.cc",
        "serial",
    ],
+    [
+        "date_tiered_test",
+        "utilities/date_tiered/date_tiered_test.cc",
+        "serial",
+    ],
    [
        "db_basic_test",
        "db/db_basic_test.cc",
@@ -606,11 +614,6 @@ ROCKS_TESTS = [
        "db/db_range_del_test.cc",
        "serial",
    ],
-    [
-        "db_secondary_test",
-        "db/db_secondary_test.cc",
-        "serial",
-    ],
    [
        "db_sst_test",
        "db/db_sst_test.cc",
@@ -671,6 +674,11 @@ ROCKS_TESTS = [
        "db/deletefile_test.cc",
        "serial",
    ],
+    [
+        "document_db_test",
+        "utilities/document/document_db_test.cc",
+        "serial",
+    ],
    [
        "dynamic_bloom_test",
        "util/dynamic_bloom_test.cc",
@@ -746,6 +754,11 @@ ROCKS_TESTS = [
        "table/full_filter_block_test.cc",
        "serial",
    ],
+    [
+        "geodb_test",
+        "utilities/geodb/geodb_test.cc",
+        "serial",
+    ],
    [
        "hash_table_test",
        "utilities/persistent_cache/hash_table_test.cc",
@@ -776,6 +789,11 @@ ROCKS_TESTS = [
        "monitoring/iostats_context_test.cc",
        "serial",
    ],
+    [
+        "json_document_test",
+        "utilities/document/json_document_test.cc",
+        "serial",
+    ],
    [
        "ldb_cmd_test",
        "tools/ldb_cmd_test.cc",
@@ -901,11 +919,6 @@ ROCKS_TESTS = [
        "db/range_del_aggregator_test.cc",
        "serial",
    ],
-    [
-        "range_tombstone_fragmenter_test",
-        "db/range_tombstone_fragmenter_test.cc",
-        "serial",
-    ],
    [
        "rate_limiter_test",
        "util/rate_limiter_test.cc",
@@ -921,11 +934,6 @@ ROCKS_TESTS = [
        "db/repair_test.cc",
        "serial",
    ],
-    [
-        "repeatable_thread_test",
-        "util/repeatable_thread_test.cc",
-        "serial",
-    ],
    [
        "sim_cache_test",
        "utilities/simulator_cache/sim_cache_test.cc",
@@ -942,13 +950,13 @@ ROCKS_TESTS = [
        "serial",
    ],
    [
-        "sst_dump_test",
-        "tools/sst_dump_test.cc",
+        "spatial_db_test",
+        "utilities/spatialdb/spatial_db_test.cc",
        "serial",
    ],
    [
-        "sst_file_reader_test",
-        "table/sst_file_reader_test.cc",
+        "sst_dump_test",
+        "tools/sst_dump_test.cc",
        "serial",
    ],
    [
@@ -1066,16 +1074,26 @@ ROCKS_TESTS = [
 # Generate a test rule for each entry in ROCKS_TESTS
 # Do not build the tests in opt mode, since SyncPoint and other test code
 # will not be included.
-[
-    test_binary(
-        parallelism = parallelism,
-        rocksdb_arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
-        rocksdb_compiler_flags = ROCKSDB_COMPILER_FLAGS,
-        rocksdb_external_deps = ROCKSDB_EXTERNAL_DEPS,
-        rocksdb_preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
-        test_cc = test_cc,
-        test_name = test_name,
-    )
-    for test_name, test_cc, parallelism in ROCKS_TESTS
-    if not is_opt_mode
-]
+if not is_opt_mode:
+    for test_cfg in ROCKS_TESTS:
+        test_name = test_cfg[0]
+        test_cc = test_cfg[1]
+        ttype = "gtest" if test_cfg[2] == "parallel" else "simple"
+        test_bin = test_name + "_bin"
+
+        cpp_binary (
+          name = test_bin,
+          srcs = [test_cc],
+          deps = [":rocksdb_test_lib"],
+          preprocessor_flags = rocksdb_preprocessor_flags,
+          arch_preprocessor_flags = rocksdb_arch_preprocessor_flags,
+          compiler_flags = rocksdb_compiler_flags,
+          external_deps = rocksdb_external_deps,
+        )
+
+        custom_unittest(
+          name = test_name,
+          type = ttype,
+          deps = [":" + test_bin],
+          command = [TEST_RUNNER, BUCK_BINS + test_bin]
+        )
@@ -12,7 +12,6 @@ At Facebook, we use RocksDB as storage engines in multiple data management servi
 6. LogDevice -- a distributed data store for logs [2]

 [1] https://research.facebook.com/publications/realtime-data-processing-at-facebook/
-
 [2] https://code.facebook.com/posts/357056558062811/logdevice-a-distributed-data-store-for-logs/

 ## LinkedIn
@@ -89,6 +88,3 @@ LzLabs is using RocksDB as a storage engine in their multi-database distributed

 ## ProfaneDB
 [ProfaneDB](https://profanedb.gitlab.io/) is a database for Protocol Buffers, and uses RocksDB for storage. It is accessible via gRPC, and the schema is defined using directly `.proto` files.
-
-## IOTA Foundation
- [IOTA Foundation](https://www.iota.org/) is using RocksDB in the [IOTA Reference Implementation (IRI)](https://github.com/iotaledger/iri) to store the local state of the Tangle. The Tangle is the first open-source distributed ledger powering the future of the Internet of Things.
@@ -43,7 +43,7 @@ We plan to use this port for our business purposes here at Bing and this provide

 * Certain headers that are not present and not necessary on Windows were simply `#ifndef OS_WIN` in a few places (`unistd.h`)
 * All posix specific headers were replaced to port/port.h which worked well
-* Replaced `dirent.h` for `port/port_dirent.h` (very few places) with the implementation of the relevant interfaces within `rocksdb::port` namespace
+* Replaced `dirent.h` for `port/dirent.h` (very few places) with the implementation of the relevant interfaces within `rocksdb::port` namespace
 * Replaced `sys/time.h` to `port/sys_time.h` (few places) implemented equivalents within `rocksdb::port`
 * `printf %z` specification is not supported on Windows. To imitate existing standards we came up with a string macro `ROCKSDB_PRIszt` which expands to `zu` on posix systems and to `Iu` on windows.
 * in class member initialization were moved to a __ctors in some cases
@@ -3,7 +3,7 @@ image: Visual Studio 2017
 before_build:
 - md %APPVEYOR_BUILD_FOLDER%\build
 - cd %APPVEYOR_BUILD_FOLDER%\build
- cmake -G "Visual Studio 15 Win64" -DOPTDBG=1 -DWITH_XPRESS=1 -DPORTABLE=1 -DJNI=1 ..
+- cmake -G "Visual Studio 15 Win64" -DOPTDBG=1 -DWITH_XPRESS=1 -DPORTABLE=1 ..
 - cd ..
 build:
  project: build\rocksdb.sln
@@ -1,4 +1,3 @@
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
@@ -82,7 +81,7 @@ def get_tests(repo_path):
            else:
                # we consumed all the parallel tests
                break
-
+    
    return tests


@@ -110,14 +109,12 @@ def generate_targets(repo_path):
        "rocksdb_test_lib",
        src_mk.get("MOCK_LIB_SOURCES", []) +
        src_mk.get("TEST_LIB_SOURCES", []) +
-        src_mk.get("EXP_LIB_SOURCES", []) +
-        src_mk.get("ANALYZER_LIB_SOURCES", []),
+        src_mk.get("EXP_LIB_SOURCES", []),
        [":rocksdb_lib"])
    # rocksdb_tools_lib
    TARGETS.add_library(
        "rocksdb_tools_lib",
        src_mk.get("BENCH_LIB_SOURCES", []) +
-        src_mk.get("ANALYZER_LIB_SOURCES", []) +
        ["util/testutil.cc"],
        [":rocksdb_lib"])

@@ -1,5 +1,4 @@
 #!/usr/bin/env bash
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 # Create a tmp directory for the test to use
 TEST_DIR=$(mktemp -d /dev/shm/fbcode_rocksdb_XXXXXXX)
 # shellcheck disable=SC2068
@@ -1,4 +1,3 @@
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
@@ -11,7 +10,7 @@ def pretty_list(lst, indent=8):

    if len(lst) == 1:
        return "\"%s\"" % lst[0]
-
+    
    separator = "\",\n%s\"" % (" " * indent)
    res = separator.join(sorted(lst))
    res = "\n" + (" " * indent) + "\"" + res + "\",\n" + (" " * (indent - 4))
@@ -32,16 +31,13 @@ class TARGETSBuilder:
        self.targets_file.close()

    def add_library(self, name, srcs, deps=None, headers=None):
-        headers_attr_prefix = ""
        if headers is None:
-            headers_attr_prefix = "auto_"
            headers = "AutoHeaders.RECURSIVE_GLOB"
-        self.targets_file.write(targets_cfg.library_template.format(
-            name=name,
-            srcs=pretty_list(srcs),
-            headers_attr_prefix=headers_attr_prefix,
-            headers=headers,
-            deps=pretty_list(deps)))
+        self.targets_file.write(targets_cfg.library_template % (
+            name,
+            pretty_list(srcs),
+            headers,
+            pretty_list(deps)))
        self.total_lib = self.total_lib + 1

    def add_binary(self, name, srcs, deps=None):
@@ -1,15 +1,17 @@
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 from __future__ import unicode_literals
-rocksdb_target_header = """load("@fbcode_macros//build_defs:auto_headers.bzl", "AutoHeaders")
-load("@fbcode_macros//build_defs:cpp_library.bzl", "cpp_library")
-load(":defs.bzl", "test_binary")
+rocksdb_target_header = """
+load("@fbcode_macros//build_defs:auto_headers.bzl", "AutoHeaders")

 REPO_PATH = package_name() + "/"

-ROCKSDB_COMPILER_FLAGS = [
+BUCK_BINS = "buck-out/gen/" + REPO_PATH
+
+TEST_RUNNER = REPO_PATH + "buckifier/rocks_test_runner.sh"
+
+rocksdb_compiler_flags = [
    "-fno-builtin-memcmp",
    "-DROCKSDB_PLATFORM_POSIX",
    "-DROCKSDB_LIB_IO_POSIX",
@@ -25,7 +27,6 @@ ROCKSDB_COMPILER_FLAGS = [
    "-DBZIP2",
    "-DLZ4",
    "-DZSTD",
-    "-DZSTD_STATIC_LINKING_ONLY",
    "-DGFLAGS=gflags",
    "-DNUMA",
    "-DTBB",
@@ -37,7 +38,7 @@ ROCKSDB_COMPILER_FLAGS = [
    "-Wnarrowing",
 ]

-ROCKSDB_EXTERNAL_DEPS = [
+rocksdb_external_deps = [
    ("bzip2", None, "bz2"),
    ("snappy", None, "snappy"),
    ("zlib", None, "z"),
@@ -49,13 +50,13 @@ ROCKSDB_EXTERNAL_DEPS = [
    ("googletest", None, "gtest"),
 ]

-ROCKSDB_PREPROCESSOR_FLAGS = [
+rocksdb_preprocessor_flags = [
    # Directories with files for #include
    "-I" + REPO_PATH + "include/",
    "-I" + REPO_PATH,
 ]

-ROCKSDB_ARCH_PREPROCESSOR_FLAGS = {
+rocksdb_arch_preprocessor_flags = {
    "x86_64": [
        "-DHAVE_SSE42",
        "-DHAVE_PCLMUL",
@@ -68,28 +69,21 @@ is_opt_mode = build_mode.startswith("opt")

 # -DNDEBUG is added by default in opt mode in fbcode. But adding it twice
 # doesn't harm and avoid forgetting to add it.
-ROCKSDB_COMPILER_FLAGS += (["-DNDEBUG"] if is_opt_mode else [])
-
-sanitizer = read_config("fbcode", "sanitizer")
-
-# Do not enable jemalloc if sanitizer presents. RocksDB will further detect
-# whether the binary is linked with jemalloc at runtime.
-ROCKSDB_COMPILER_FLAGS += (["-DROCKSDB_JEMALLOC"] if sanitizer == "" else [])
-
-ROCKSDB_EXTERNAL_DEPS += ([("jemalloc", None, "headers")] if sanitizer == "" else [])
+if is_opt_mode:
+    rocksdb_compiler_flags.append("-DNDEBUG")
 """


 library_template = """
 cpp_library(
-    name = "{name}",
-    srcs = [{srcs}],
-    {headers_attr_prefix}headers = {headers},
-    arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
-    compiler_flags = ROCKSDB_COMPILER_FLAGS,
-    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
-    deps = [{deps}],
-    external_deps = ROCKSDB_EXTERNAL_DEPS,
+    name = "%s",
+    srcs = [%s],
+    headers = %s,
+    arch_preprocessor_flags = rocksdb_arch_preprocessor_flags,
+    compiler_flags = rocksdb_compiler_flags,
+    preprocessor_flags = rocksdb_preprocessor_flags,
+    deps = [%s],
+    external_deps = rocksdb_external_deps,
 )
 """

@@ -97,11 +91,11 @@ binary_template = """
 cpp_binary(
    name = "%s",
    srcs = [%s],
-    arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
-    compiler_flags = ROCKSDB_COMPILER_FLAGS,
-    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+    arch_preprocessor_flags = rocksdb_arch_preprocessor_flags,
+    compiler_flags = rocksdb_compiler_flags,
+    preprocessor_flags = rocksdb_preprocessor_flags,
    deps = [%s],
-    external_deps = ROCKSDB_EXTERNAL_DEPS,
+    external_deps = rocksdb_external_deps,
 )
 """

@@ -120,17 +114,28 @@ ROCKS_TESTS = [
 # Generate a test rule for each entry in ROCKS_TESTS
 # Do not build the tests in opt mode, since SyncPoint and other test code
 # will not be included.
-[
-    test_binary(
-        parallelism = parallelism,
-        rocksdb_arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
-        rocksdb_compiler_flags = ROCKSDB_COMPILER_FLAGS,
-        rocksdb_external_deps = ROCKSDB_EXTERNAL_DEPS,
-        rocksdb_preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
-        test_cc = test_cc,
-        test_name = test_name,
-    )
-    for test_name, test_cc, parallelism in ROCKS_TESTS
-    if not is_opt_mode
-]
+if not is_opt_mode:
+    for test_cfg in ROCKS_TESTS:
+        test_name = test_cfg[0]
+        test_cc = test_cfg[1]
+        ttype = "gtest" if test_cfg[2] == "parallel" else "simple"
+        test_bin = test_name + "_bin"
+
+        cpp_binary (
+          name = test_bin,
+          srcs = [test_cc],
+          deps = [":rocksdb_test_lib"],
+          preprocessor_flags = rocksdb_preprocessor_flags,
+          arch_preprocessor_flags = rocksdb_arch_preprocessor_flags,
+          compiler_flags = rocksdb_compiler_flags,
+          external_deps = rocksdb_external_deps,
+        )
+
+        custom_unittest(
+          name = test_name,
+          type = ttype,
+          deps = [":" + test_bin],
+          command = [TEST_RUNNER, BUCK_BINS + test_bin]
+        )
+
 """
@@ -1,4 +1,3 @@
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 """
 This module keeps commonly used components.
 """
@@ -7,12 +7,12 @@
 // Name of the environment variables which need to be set by the entity which
 // triggers continuous runs so that code at the end of the file gets executed
 // and Sandcastle run starts.
-const ENV_POST_RECEIVE_HOOK = "POST_RECEIVE_HOOK";
-const ENV_HTTPS_APP_VALUE = "HTTPS_APP_VALUE";
-const ENV_HTTPS_TOKEN_VALUE = "HTTPS_TOKEN_VALUE";
+define("ENV_POST_RECEIVE_HOOK", "POST_RECEIVE_HOOK");
+define("ENV_HTTPS_APP_VALUE", "HTTPS_APP_VALUE");
+define("ENV_HTTPS_TOKEN_VALUE", "HTTPS_TOKEN_VALUE");

-const PRIMARY_TOKEN_FILE = '/home/krad/.sandcastle';
-const CONT_RUN_ALIAS = "leveldb";
+define("PRIMARY_TOKEN_FILE", '/home/krad/.sandcastle');
+define("CONT_RUN_ALIAS", "leveldb");

 //////////////////////////////////////////////////////////////////////
 /*  Run tests in sandcastle */
@@ -97,7 +97,7 @@ function getSteps($applyDiff, $diffID, $username, $test) {
  }

  // fbcode is a sub-repo. We cannot patch until we add it to ignore otherwise
-  // Git thinks it is an uncommitted change.
+  // Git thinks it is an uncommited change.
  $fix_git_ignore = array(
    "name" => "Fix git ignore",
    "shell" => "echo fbcode >> .git/info/exclude",
@@ -1,5 +1,4 @@
 #!/usr/bin/python
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.

 # amalgamate.py creates an amalgamation from a unity build.
 # It can be run with either Python 2 or 3.
@@ -53,13 +53,11 @@ if [ -z "$ROCKSDB_NO_FBCODE" -a -d /mnt/gvfs/third-party ]; then
    FBCODE_BUILD="true"
    # If we're compiling with TSAN we need pic build
    PIC_BUILD=$COMPILE_WITH_TSAN
-    if [ -n "$ROCKSDB_FBCODE_BUILD_WITH_481" ]; then
-      # we need this to build with MySQL. Don't use for other purposes.
-      source "$PWD/build_tools/fbcode_config4.8.1.sh"
-    elif [ -n "$ROCKSDB_FBCODE_BUILD_WITH_5xx" ]; then
+    if [ -z "$ROCKSDB_FBCODE_BUILD_WITH_481" ]; then
      source "$PWD/build_tools/fbcode_config.sh"
    else
-      source "$PWD/build_tools/fbcode_config_platform007.sh"
+      # we need this to build with MySQL. Don't use for other purposes.
+      source "$PWD/build_tools/fbcode_config4.8.1.sh"
    fi
 fi

@@ -234,7 +232,7 @@ else
          #include <linux/falloc.h>
          int main() {
      int fd = open("/dev/null", 0);
-      fallocate(fd, FALLOC_FL_KEEP_SIZE, 0, 1024);
+      fallocate(fd, FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE, 0, 1024);
          }
 EOF
        if [ "$?" = 0 ]; then
@@ -370,16 +368,6 @@ EOF
            # JEMALLOC can be enabled either using the flag (like here) or by
            # providing direct link to the jemalloc library
            WITH_JEMALLOC_FLAG=1
-            # check for JEMALLOC installed with HomeBrew
-            if [ "$PLATFORM" == "OS_MACOSX" ]; then
-                if hash brew 2>/dev/null && brew ls --versions jemalloc > /dev/null; then
-                    JEMALLOC_VER=$(brew ls --versions jemalloc | tail -n 1 | cut -f 2 -d ' ')
-                    JEMALLOC_INCLUDE="-I/usr/local/Cellar/jemalloc/${JEMALLOC_VER}/include"
-                    JEMALLOC_LIB="/usr/local/Cellar/jemalloc/${JEMALLOC_VER}/lib/libjemalloc_pic.a"
-                    PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS $JEMALLOC_LIB"
-                    JAVA_STATIC_LDFLAGS="$JAVA_STATIC_LDFLAGS $JEMALLOC_LIB"
-                fi
-            fi
        fi
    fi
    if ! test $JEMALLOC && ! test $ROCKSDB_DISABLE_TCMALLOC; then
@@ -486,17 +474,6 @@ EOF
            COMMON_FLAGS="$COMMON_FLAGS -DROCKSDB_SCHED_GETCPU_PRESENT"
        fi
    fi
-
-    if ! test $ROCKSDB_DISABLE_ALIGNED_NEW; then
-        # Test whether c++17 aligned-new is supported
-        $CXX $PLATFORM_CXXFLAGS -faligned-new -x c++ - -o /dev/null 2>/dev/null <<EOF
-            struct alignas(1024) t {int a;};
-            int main() {}
-EOF
-        if [ "$?" = 0 ]; then
-            PLATFORM_CXXFLAGS="$PLATFORM_CXXFLAGS -faligned-new -DHAVE_ALIGNED_NEW"
-        fi
-    fi
 fi

 # TODO(tec): Fix -Wshorten-64-to-32 errors on FreeBSD and enable the warning.
@@ -518,8 +495,8 @@ if test "$USE_HDFS"; then
    echo "JAVA_HOME has to be set for HDFS usage."
    exit 1
  fi
-  HDFS_CCFLAGS="$HDFS_CCFLAGS -I$JAVA_HOME/include -I$JAVA_HOME/include/linux -DUSE_HDFS -I$HADOOP_HOME/include"
-  HDFS_LDFLAGS="$HDFS_LDFLAGS -lhdfs -L$JAVA_HOME/jre/lib/amd64 -L$HADOOP_HOME/lib/native"
+  HDFS_CCFLAGS="$HDFS_CCFLAGS -I$JAVA_HOME/include -I$JAVA_HOME/include/linux -DUSE_HDFS"
+  HDFS_LDFLAGS="$HDFS_LDFLAGS -lhdfs -L$JAVA_HOME/jre/lib/amd64"
  HDFS_LDFLAGS="$HDFS_LDFLAGS -L$JAVA_HOME/jre/lib/amd64/server -L$GLIBC_RUNTIME_PATH/lib"
  HDFS_LDFLAGS="$HDFS_LDFLAGS -ldl -lverify -ljava -ljvm"
  COMMON_FLAGS="$COMMON_FLAGS $HDFS_CCFLAGS"
@@ -1,4 +1,3 @@
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 GCC_BASE=/mnt/gvfs/third-party2/gcc/112ec378fec7002ad3e09afde022e656049f7191/5.x/centos7-native/c447969
 CLANG_BASE=/mnt/gvfs/third-party2/llvm-fb/04999bdb3ce81a11073535dcb00b5e13dc1cbaf5/stable/centos7-native/c9f9104
 LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/92b0c8e5c8eecc71eb042594ce1ab3413799b385/5.x/gcc-5-glibc-2.23/339d858
@@ -1,5 +1,4 @@
 # shellcheck disable=SC2148
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 GCC_BASE=/mnt/gvfs/third-party2/gcc/cf7d14c625ce30bae1a4661c2319c5a283e4dd22/4.8.1/centos6-native/cc6c9dc
 CLANG_BASE=/mnt/gvfs/third-party2/llvm-fb/8598c375b0e94e1448182eb3df034704144a838d/stable/centos6-native/3f16ddd
 LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/d6e0a7da6faba45f5e5b1638f9edd7afc2f34e7d/4.8.1/gcc-4.8.1-glibc-2.17/8aac7fc
@@ -1,19 +0,0 @@
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
-GCC_BASE=/mnt/gvfs/third-party2/gcc/6e8e715624fd15256a7970073387793dfcf79b46/7.x/centos7-native/b2ef2b6
-CLANG_BASE=/mnt/gvfs/third-party2/llvm-fb/ef37e1faa1c29782abfac1ae65a291b9b7966f6d/stable/centos7-native/c9f9104
-LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/c67031f0f739ac61575a061518d6ef5038f99f90/7.x/platform007/5620abc
-GLIBC_BASE=/mnt/gvfs/third-party2/glibc/60d6f124a78798b73944f5ba87c2306ae3460153/2.26/platform007/f259413
-SNAPPY_BASE=/mnt/gvfs/third-party2/snappy/7f9bdaada18f59bc27ec2b0871eb8a6144343aef/1.1.3/platform007/ca4da3d
-ZLIB_BASE=/mnt/gvfs/third-party2/zlib/22c2d65676fb7c23cfa797c4f6937f38b026f3cf/1.2.8/platform007/ca4da3d
-BZIP2_BASE=/mnt/gvfs/third-party2/bzip2/dc49a21c5fceec6456a7a28a94dcd16690af1337/1.0.6/platform007/ca4da3d
-LZ4_BASE=/mnt/gvfs/third-party2/lz4/907b498203d297947f3bb70b9466f47e100f1873/r131/platform007/ca4da3d
-ZSTD_BASE=/mnt/gvfs/third-party2/zstd/3ee276cbacfad3074e3f07bf826ac47f06970f4e/1.3.5/platform007/15a3614
-GFLAGS_BASE=/mnt/gvfs/third-party2/gflags/0b9929d2588991c65a57168bf88aff2db87c5d48/2.2.0/platform007/ca4da3d
-JEMALLOC_BASE=/mnt/gvfs/third-party2/jemalloc/9c910d36d6235cc40e8ff559358f1833452300ca/master/platform007/5b0f53e
-NUMA_BASE=/mnt/gvfs/third-party2/numa/9cbf2460284c669ed19c3ccb200a71f7dd7e53c7/2.0.11/platform007/ca4da3d
-LIBUNWIND_BASE=/mnt/gvfs/third-party2/libunwind/bf3d7497fe4e6d007354f0adffa16ce3003f8338/1.3/platform007/6f3e0a9
-TBB_BASE=/mnt/gvfs/third-party2/tbb/ff4e0b093534704d8abab678a4fd7f5ea7b094c7/2018_U5/platform007/ca4da3d
-KERNEL_HEADERS_BASE=/mnt/gvfs/third-party2/kernel-headers/b5c4a61a5c483ba24722005ae07895971a2ac707/fb/platform007/da39a3e
-BINUTILS_BASE=/mnt/gvfs/third-party2/binutils/92ff90349e2f43ea0a8246d8b1cf17b6869013e3/2.29.1/centos7-native/da39a3e
-VALGRIND_BASE=/mnt/gvfs/third-party2/valgrind/f3f697a28122e6bcd513273dd9c1ff23852fc59f/3.13.0/platform007/ca4da3d
-LUA_BASE=/mnt/gvfs/third-party2/lua/f0cd714433206d5139df61659eb7b28b1dea6683/5.3.4/platform007/5007832
@@ -1,3 +1,2 @@
 #!/usr/bin/env bash
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 docker run -v $PWD:/rocks -w /rocks buildpack-deps make
@@ -132,14 +132,11 @@ _TEST_NAME_TO_PARSERS = {
    'lite': [CompilerErrorParser],
    'lite_test': [CompilerErrorParser, GTestErrorParser],
    'stress_crash': [CompilerErrorParser, DbCrashErrorParser],
-    'stress_crash_with_atomic_flush': [CompilerErrorParser, DbCrashErrorParser],
    'write_stress': [CompilerErrorParser, WriteStressErrorParser],
    'asan': [CompilerErrorParser, GTestErrorParser, AsanErrorParser],
    'asan_crash': [CompilerErrorParser, AsanErrorParser, DbCrashErrorParser],
-    'asan_crash_with_atomic_flush': [CompilerErrorParser, AsanErrorParser, DbCrashErrorParser],
    'ubsan': [CompilerErrorParser, GTestErrorParser, UbsanErrorParser],
    'ubsan_crash': [CompilerErrorParser, UbsanErrorParser, DbCrashErrorParser],
-    'ubsan_crash_with_atomic_flush': [CompilerErrorParser, UbsanErrorParser, DbCrashErrorParser],
    'valgrind': [CompilerErrorParser, GTestErrorParser, ValgrindErrorParser],
    'tsan': [CompilerErrorParser, GTestErrorParser, TsanErrorParser],
    'format_compatible': [CompilerErrorParser, CompatErrorParser],
@@ -1,5 +1,5 @@
 #!/bin/sh
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
+
 # fail early
 set -e

@@ -1,5 +1,4 @@
 #!/bin/sh
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 #
 # Set environment variables so that we can compile rocksdb using
 # fbcode settings.  It uses the latest g++ and clang compilers and also
@@ -52,7 +51,7 @@ if test -z $PIC_BUILD; then
 else
  ZSTD_LIBS=" $ZSTD_BASE/lib/libzstd_pic.a"
 fi
-CFLAGS+=" -DZSTD -DZSTD_STATIC_LINKING_ONLY"
+CFLAGS+=" -DZSTD"

 # location of gflags headers and libraries
 GFLAGS_INCLUDE=" -I $GFLAGS_BASE/include/"
@@ -1,5 +1,4 @@
 #!/bin/sh
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 #
 # Set environment variables so that we can compile rocksdb using
 # fbcode settings.  It uses the latest g++ compiler and also
@@ -1,158 +0,0 @@
-#!/bin/sh
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
-#
-# Set environment variables so that we can compile rocksdb using
-# fbcode settings.  It uses the latest g++ and clang compilers and also
-# uses jemalloc
-# Environment variables that change the behavior of this script:
-# PIC_BUILD -- if true, it will only take pic versions of libraries from fbcode. libraries that don't have pic variant will not be included
-
-
-BASEDIR=`dirname $BASH_SOURCE`
-source "$BASEDIR/dependencies_platform007.sh"
-
-CFLAGS=""
-
-# libgcc
-LIBGCC_INCLUDE="$LIBGCC_BASE/include/c++/7.3.0"
-LIBGCC_LIBS=" -L $LIBGCC_BASE/lib"
-
-# glibc
-GLIBC_INCLUDE="$GLIBC_BASE/include"
-GLIBC_LIBS=" -L $GLIBC_BASE/lib"
-
-# snappy
-SNAPPY_INCLUDE=" -I $SNAPPY_BASE/include/"
-if test -z $PIC_BUILD; then
-  SNAPPY_LIBS=" $SNAPPY_BASE/lib/libsnappy.a"
-else
-  SNAPPY_LIBS=" $SNAPPY_BASE/lib/libsnappy_pic.a"
-fi
-CFLAGS+=" -DSNAPPY"
-
-if test -z $PIC_BUILD; then
-  # location of zlib headers and libraries
-  ZLIB_INCLUDE=" -I $ZLIB_BASE/include/"
-  ZLIB_LIBS=" $ZLIB_BASE/lib/libz.a"
-  CFLAGS+=" -DZLIB"
-
-  # location of bzip headers and libraries
-  BZIP_INCLUDE=" -I $BZIP2_BASE/include/"
-  BZIP_LIBS=" $BZIP2_BASE/lib/libbz2.a"
-  CFLAGS+=" -DBZIP2"
-
-  LZ4_INCLUDE=" -I $LZ4_BASE/include/"
-  LZ4_LIBS=" $LZ4_BASE/lib/liblz4.a"
-  CFLAGS+=" -DLZ4"
-fi
-
-ZSTD_INCLUDE=" -I $ZSTD_BASE/include/"
-if test -z $PIC_BUILD; then
-  ZSTD_LIBS=" $ZSTD_BASE/lib/libzstd.a"
-else
-  ZSTD_LIBS=" $ZSTD_BASE/lib/libzstd_pic.a"
-fi
-CFLAGS+=" -DZSTD"
-
-# location of gflags headers and libraries
-GFLAGS_INCLUDE=" -I $GFLAGS_BASE/include/"
-if test -z $PIC_BUILD; then
-  GFLAGS_LIBS=" $GFLAGS_BASE/lib/libgflags.a"
-else
-  GFLAGS_LIBS=" $GFLAGS_BASE/lib/libgflags_pic.a"
-fi
-CFLAGS+=" -DGFLAGS=gflags"
-
-# location of jemalloc
-JEMALLOC_INCLUDE=" -I $JEMALLOC_BASE/include/"
-JEMALLOC_LIB=" $JEMALLOC_BASE/lib/libjemalloc.a"
-
-if test -z $PIC_BUILD; then
-  # location of numa
-  NUMA_INCLUDE=" -I $NUMA_BASE/include/"
-  NUMA_LIB=" $NUMA_BASE/lib/libnuma.a"
-  CFLAGS+=" -DNUMA"
-
-  # location of libunwind
-  LIBUNWIND="$LIBUNWIND_BASE/lib/libunwind.a"
-fi
-
-# location of TBB
-TBB_INCLUDE=" -isystem $TBB_BASE/include/"
-if test -z $PIC_BUILD; then
-  TBB_LIBS="$TBB_BASE/lib/libtbb.a"
-else
-  TBB_LIBS="$TBB_BASE/lib/libtbb_pic.a"
-fi
-CFLAGS+=" -DTBB"
-
-# use Intel SSE support for checksum calculations
-export USE_SSE=1
-export PORTABLE=1
-
-BINUTILS="$BINUTILS_BASE/bin"
-AR="$BINUTILS/ar"
-
-DEPS_INCLUDE="$SNAPPY_INCLUDE $ZLIB_INCLUDE $BZIP_INCLUDE $LZ4_INCLUDE $ZSTD_INCLUDE $GFLAGS_INCLUDE $NUMA_INCLUDE $TBB_INCLUDE"
-
-STDLIBS="-L $GCC_BASE/lib64"
-
-CLANG_BIN="$CLANG_BASE/bin"
-CLANG_LIB="$CLANG_BASE/lib"
-CLANG_SRC="$CLANG_BASE/../../src"
-
-CLANG_ANALYZER="$CLANG_BIN/clang++"
-CLANG_SCAN_BUILD="$CLANG_SRC/llvm/tools/clang/tools/scan-build/bin/scan-build"
-
-if [ -z "$USE_CLANG" ]; then
-  # gcc
-  CC="$GCC_BASE/bin/gcc"
-  CXX="$GCC_BASE/bin/g++"
-
-  CFLAGS+=" -B$BINUTILS/gold"
-  CFLAGS+=" -isystem $LIBGCC_INCLUDE"
-  CFLAGS+=" -isystem $GLIBC_INCLUDE"
-  JEMALLOC=1
-else
-  # clang
-  CLANG_INCLUDE="$CLANG_LIB/clang/stable/include"
-  CC="$CLANG_BIN/clang"
-  CXX="$CLANG_BIN/clang++"
-
-  KERNEL_HEADERS_INCLUDE="$KERNEL_HEADERS_BASE/include"
-
-  CFLAGS+=" -B$BINUTILS/gold -nostdinc -nostdlib"
-  CFLAGS+=" -isystem $LIBGCC_BASE/include/c++/7.x "
-  CFLAGS+=" -isystem $LIBGCC_BASE/include/c++/7.x/x86_64-facebook-linux "
-  CFLAGS+=" -isystem $GLIBC_INCLUDE"
-  CFLAGS+=" -isystem $LIBGCC_INCLUDE"
-  CFLAGS+=" -isystem $CLANG_INCLUDE"
-  CFLAGS+=" -isystem $KERNEL_HEADERS_INCLUDE/linux "
-  CFLAGS+=" -isystem $KERNEL_HEADERS_INCLUDE "
-  CFLAGS+=" -Wno-expansion-to-defined "
-  CXXFLAGS="-nostdinc++"
-fi
-
-CFLAGS+=" $DEPS_INCLUDE"
-CFLAGS+=" -DROCKSDB_PLATFORM_POSIX -DROCKSDB_LIB_IO_POSIX -DROCKSDB_FALLOCATE_PRESENT -DROCKSDB_MALLOC_USABLE_SIZE -DROCKSDB_RANGESYNC_PRESENT -DROCKSDB_SCHED_GETCPU_PRESENT -DROCKSDB_SUPPORT_THREAD_LOCAL -DHAVE_SSE42"
-CXXFLAGS+=" $CFLAGS"
-
-EXEC_LDFLAGS=" $SNAPPY_LIBS $ZLIB_LIBS $BZIP_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS $NUMA_LIB $TBB_LIBS"
-EXEC_LDFLAGS+=" -B$BINUTILS/gold"
-EXEC_LDFLAGS+=" -Wl,--dynamic-linker,/usr/local/fbcode/platform007/lib/ld.so"
-EXEC_LDFLAGS+=" $LIBUNWIND"
-EXEC_LDFLAGS+=" -Wl,-rpath=/usr/local/fbcode/platform007/lib"
-# required by libtbb
-EXEC_LDFLAGS+=" -ldl"
-
-PLATFORM_LDFLAGS="$LIBGCC_LIBS $GLIBC_LIBS $STDLIBS -lgcc -lstdc++"
-
-EXEC_LDFLAGS_SHARED="$SNAPPY_LIBS $ZLIB_LIBS $BZIP_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS $TBB_LIBS"
-
-VALGRIND_VER="$VALGRIND_BASE/bin/"
-
-# lua not supported because it's on track for deprecation, I think
-LUA_PATH=
-LUA_LIB=
-
-export CC CXX AR CFLAGS CXXFLAGS EXEC_LDFLAGS EXEC_LDFLAGS_SHARED VALGRIND_VER JEMALLOC_LIB JEMALLOC_INCLUDE CLANG_ANALYZER CLANG_SCAN_BUILD LUA_PATH LUA_LIB
@@ -1,5 +1,4 @@
 #!/usr/bin/env bash
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 # If clang_format_diff.py command is not specfied, we assume we are able to
 # access directly without any path.
 if [ -z $CLANG_FORMAT_DIFF ]
@@ -5082,8 +5082,8 @@ sub openoutputfiles {
 	# Set reading FD if using --group (--ungroup does not need)
 	for my $fdno (1,2) {
 	    # Re-open the file for reading
-	    # so fdw can be closed separately
-	    # and fdr can be seeked separately (for --line-buffer)
+	    # so fdw can be closed seperately
+	    # and fdr can be seeked seperately (for --line-buffer)
 	    open(my $fdr,"<", $self->fh($fdno,'name')) ||
 		::die_bug("fdr: Cannot open ".$self->fh($fdno,'name'));
 	    $self->set_fh($fdno,'r',$fdr);
@@ -1,6 +1,5 @@
 # shellcheck disable=SC1113
 #/usr/bin/env bash
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.

 set -e

@@ -1,5 +1,4 @@
 #!/usr/local/fbcode/gcc-4.9-glibc-2.20-fb/bin/python2.7
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.

 from __future__ import absolute_import
 from __future__ import division
@@ -1,5 +1,4 @@
 #!/usr/bin/env bash
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.

 set -e

@@ -85,9 +85,8 @@ NON_SHM="TMPD=/tmp/rocksdb_test_tmp"
 GCC_481="ROCKSDB_FBCODE_BUILD_WITH_481=1"
 ASAN="COMPILE_WITH_ASAN=1"
 CLANG="USE_CLANG=1"
-# in gcc-5 there are known problems with TSAN like https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71090.
-# using platform007 gives us gcc-8 or higher which has that bug fixed.
-TSAN="ROCKSDB_FBCODE_BUILD_WITH_PLATFORM007=1 COMPILE_WITH_TSAN=1"
+LITE="OPT=\"-DROCKSDB_LITE -g\""
+TSAN="COMPILE_WITH_TSAN=1"
 UBSAN="COMPILE_WITH_UBSAN=1"
 TSAN_CRASH='CRASH_TEST_EXT_ARGS="--compression_type=zstd --log2_keys_per_lock=22"'
 NON_TSAN_CRASH="CRASH_TEST_EXT_ARGS=--compression_type=zstd"
@@ -346,7 +345,7 @@ LITE_BUILD_COMMANDS="[
            $CLEANUP_ENV,
            {
                'name':'Build RocksDB debug version',
-                'shell':'make J=1 LITE=1 all check || $CONTRUN_NAME=lite $TASK_CREATION_TOOL',
+                'shell':'$LITE make J=1 all check || $CONTRUN_NAME=lite $TASK_CREATION_TOOL',
                'user':'root',
                $PARSER
            },
@@ -400,35 +399,6 @@ STRESS_CRASH_TEST_COMMANDS="[
    }
 ]"

-#
-# RocksDB stress/crash test with atomic flush
-#
-STRESS_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
-    {
-        'name':'Rocksdb Stress/Crash Test (atomic flush)',
-        'oncall':'$ONCALL',
-        'timeout': 86400,
-        'steps': [
-            $CLEANUP_ENV,
-            {
-                'name':'Build and run RocksDB debug stress tests',
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
-                'user':'root',
-                $PARSER
-            },
-            {
-                'name':'Build and run RocksDB debug crash tests with atomic flush',
-                'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 crash_test_with_atomic_flush || $CONTRUN_NAME=crash_test_with_atomic_flush $TASK_CREATION_TOOL',
-                'user':'root',
-                $PARSER
-            }
-        ],
-        $ARTIFACTS,
-        $REPORT
-    }
-]"
-
 # RocksDB write stress test.
 # We run on disk device on purpose (i.e. no $SHM)
 # because we want to add some randomness to fsync commands
@@ -493,28 +463,6 @@ ASAN_CRASH_TEST_COMMANDS="[
    }
 ]"

-#
-# RocksDB crash testing with atomic flush under address sanitizer
-#
-ASAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
-    {
-        'name':'Rocksdb crash test (atomic flush) under ASAN',
-        'oncall':'$ONCALL',
-        'timeout': 86400,
-        'steps': [
-            $CLEANUP_ENV,
-            {
-                'name':'Build and run RocksDB debug asan_crash_test_with_atomic_flush',
-                'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 asan_crash_test_with_atomic_flush || $CONTRUN_NAME=asan_crash_test_with_atomic_flush $TASK_CREATION_TOOL',
-                'user':'root',
-                $PARSER
-            },
-        ],
-        $REPORT
-    }
-]"
-
 #
 # RocksDB test under undefined behavior sanitizer
 #
@@ -557,28 +505,6 @@ UBSAN_CRASH_TEST_COMMANDS="[
    }
 ]"

-#
-# RocksDB crash testing with atomic flush under undefined behavior sanitizer
-#
-UBSAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
-    {
-        'name':'Rocksdb crash test (atomic flush) under UBSAN',
-        'oncall':'$ONCALL',
-        'timeout': 86400,
-        'steps': [
-            $CLEANUP_ENV,
-            {
-                'name':'Build and run RocksDB debug ubsan_crash_test_with_atomic_flush',
-                'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 ubsan_crash_test_with_atomic_flush || $CONTRUN_NAME=ubsan_crash_test_with_atomic_flush $TASK_CREATION_TOOL',
-                'user':'root',
-                $PARSER
-            },
-        ],
-        $REPORT
-    }
-]"
-
 #
 # RocksDB unit test under valgrind
 #
@@ -645,28 +571,6 @@ TSAN_CRASH_TEST_COMMANDS="[
    }
 ]"

-#
-# RocksDB crash test with atomic flush under TSAN
-#
-TSAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
-    {
-        'name':'Rocksdb Crash Test with atomic flush under TSAN',
-        'oncall':'$ONCALL',
-        'timeout': 86400,
-        'steps': [
-            $CLEANUP_ENV,
-            {
-                'name':'Compile and run',
-                'timeout': 86400,
-                'shell':'set -o pipefail && $SHM $DEBUG $TSAN $TSAN_CRASH CRASH_TEST_KILL_ODD=1887 make J=1 crash_test_with_atomic_flush || $CONTRUN_NAME=tsan_crash_test_with_atomic_flush $TASK_CREATION_TOOL',
-                'user':'root',
-                $PARSER
-            },
-        ],
-        $REPORT
-    }
-]"
-
 #
 # RocksDB format compatible
 #
@@ -706,7 +610,7 @@ run_no_compression()
  rm -rf /dev/shm/rocksdb
  mkdir /dev/shm/rocksdb
  make clean
-  cat build_tools/fbcode_config.sh | grep -iv dzstd | grep -iv dzlib | grep -iv dlz4 | grep -iv dsnappy | grep -iv dbzip2 > .tmp.fbcode_config.sh
+  cat build_tools/fbcode_config.sh | grep -iv dzlib | grep -iv dlz4 | grep -iv dsnappy | grep -iv dbzip2 > .tmp.fbcode_config.sh
  mv .tmp.fbcode_config.sh build_tools/fbcode_config.sh
  cat Makefile | grep -v tools/ldb_test.py > .tmp.Makefile
  mv .tmp.Makefile Makefile
@@ -759,12 +663,12 @@ run_regression()

  # === lite build ===
  make clean
-  make LITE=1 -j$(nproc) static_lib
+  OPT=-DROCKSDB_LITE make -j$(nproc) static_lib
  send_size_to_ods static_lib_lite $(stat --printf="%s" librocksdb.a)
  strip librocksdb.a
  send_size_to_ods static_lib_lite_stripped $(stat --printf="%s" librocksdb.a)

-  make LITE=1 -j$(nproc) shared_lib
+  OPT=-DROCKSDB_LITE make -j$(nproc) shared_lib
  send_size_to_ods shared_lib_lite $(stat --printf="%s" `readlink -f librocksdb.so`)
  strip `readlink -f librocksdb.so`
  send_size_to_ods shared_lib_lite_stripped $(stat --printf="%s" `readlink -f librocksdb.so`)
@@ -848,9 +752,6 @@ case $1 in
  stress_crash)
    echo $STRESS_CRASH_TEST_COMMANDS
    ;;
-  stress_crash_with_atomic_flush)
-    echo $STRESS_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS
-    ;;
  write_stress)
    echo $WRITE_STRESS_COMMANDS
    ;;
@@ -860,18 +761,12 @@ case $1 in
  asan_crash)
    echo $ASAN_CRASH_TEST_COMMANDS
    ;;
-  asan_crash_with_atomic_flush)
-    echo $ASAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS
-    ;;
  ubsan)
    echo $UBSAN_TEST_COMMANDS
    ;;
  ubsan_crash)
    echo $UBSAN_CRASH_TEST_COMMANDS
    ;;
-  ubsan_crash_with_atomic_flush)
-    echo $UBSAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS
-    ;;
  valgrind)
    echo $VALGRIND_TEST_COMMANDS
    ;;
@@ -881,9 +776,6 @@ case $1 in
  tsan_crash)
    echo $TSAN_CRASH_TEST_COMMANDS
    ;;
-  tsan_crash_with_atomic_flush)
-    echo $TSAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS
-    ;;
  format_compatible)
    echo $FORMAT_COMPATIBLE_COMMANDS
    ;;
@@ -1,4 +1,3 @@
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 # This script enables you running RocksDB tests by running
 # All the tests concurrently and utilizing all the cores
 Param(
@@ -1,5 +1,4 @@
 #!/bin/bash
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 set -e

 ROCKSDB_VERSION="5.10.3"
@@ -1,5 +1,4 @@
 #!/bin/sh
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 #
 # Update dependencies.sh file with the latest avaliable versions

@@ -54,45 +53,6 @@ function get_lib_base()
  log_variable $__res_var
 }

-###########################################################
-#                platform007 dependencies                 #
-###########################################################
-
-OUTPUT="$BASEDIR/dependencies_platform007.sh"
-
-rm -f "$OUTPUT"
-touch "$OUTPUT"
-
-echo "Writing dependencies to $OUTPUT"
-
-# Compilers locations
-GCC_BASE=`readlink -f $TP2_LATEST/gcc/7.x/centos7-native/*/`
-CLANG_BASE=`readlink -f $TP2_LATEST/llvm-fb/stable/centos7-native/*/`
-
-log_variable GCC_BASE
-log_variable CLANG_BASE
-
-# Libraries locations
-get_lib_base libgcc     7.x     platform007
-get_lib_base glibc      2.26    platform007
-get_lib_base snappy     LATEST  platform007
-get_lib_base zlib       LATEST  platform007
-get_lib_base bzip2      LATEST  platform007
-get_lib_base lz4        LATEST  platform007
-get_lib_base zstd       LATEST  platform007
-get_lib_base gflags     LATEST  platform007
-get_lib_base jemalloc   LATEST  platform007
-get_lib_base numa       LATEST  platform007
-get_lib_base libunwind  LATEST  platform007
-get_lib_base tbb        LATEST  platform007
-
-get_lib_base kernel-headers fb platform007
-get_lib_base binutils   LATEST centos7-native
-get_lib_base valgrind   LATEST platform007
-get_lib_base lua        5.3.4  platform007
-
-git diff $OUTPUT
-
 ###########################################################
 #                   5.x dependencies                      #
 ###########################################################
@@ -1,5 +1,4 @@
 #!/usr/bin/env bash
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 if [ "$#" = "0" ]; then
  echo "Usage: $0 major|minor|patch|full"
  exit 1
@@ -64,8 +64,8 @@ class CacheTest : public testing::TestWithParam<std::string> {

  std::vector<int> deleted_keys_;
  std::vector<int> deleted_values_;
-  std::shared_ptr<Cache> cache_;
-  std::shared_ptr<Cache> cache2_;
+  shared_ptr<Cache> cache_;
+  shared_ptr<Cache> cache2_;

  CacheTest()
      : cache_(NewCache(kCacheSize, kNumShardBits, false)),
@@ -73,7 +73,8 @@ class CacheTest : public testing::TestWithParam<std::string> {
    current_ = this;
  }

-  ~CacheTest() override {}
+  ~CacheTest() {
+  }

  std::shared_ptr<Cache> NewCache(size_t capacity) {
    auto type = GetParam();
@@ -98,7 +99,7 @@ class CacheTest : public testing::TestWithParam<std::string> {
    return nullptr;
  }

-  int Lookup(std::shared_ptr<Cache> cache, int key) {
+  int Lookup(shared_ptr<Cache> cache, int key) {
    Cache::Handle* handle = cache->Lookup(EncodeKey(key));
    const int r = (handle == nullptr) ? -1 : DecodeValue(cache->Value(handle));
    if (handle != nullptr) {
@@ -107,16 +108,16 @@ class CacheTest : public testing::TestWithParam<std::string> {
    return r;
  }

-  void Insert(std::shared_ptr<Cache> cache, int key, int value,
-              int charge = 1) {
+  void Insert(shared_ptr<Cache> cache, int key, int value, int charge = 1) {
    cache->Insert(EncodeKey(key), EncodeValue(value), charge,
                  &CacheTest::Deleter);
  }

-  void Erase(std::shared_ptr<Cache> cache, int key) {
+  void Erase(shared_ptr<Cache> cache, int key) {
    cache->Erase(EncodeKey(key));
  }

+
  int Lookup(int key) {
    return Lookup(cache_, key);
  }
@@ -144,7 +145,7 @@ class CacheTest : public testing::TestWithParam<std::string> {
 CacheTest* CacheTest::current_;

 TEST_P(CacheTest, UsageTest) {
-  // cache is std::shared_ptr and will be automatically cleaned up.
+  // cache is shared_ptr and will be automatically cleaned up.
  const uint64_t kCapacity = 100000;
  auto cache = NewCache(kCapacity, 8, false);

@@ -172,7 +173,7 @@ TEST_P(CacheTest, UsageTest) {
 }

 TEST_P(CacheTest, PinnedUsageTest) {
-  // cache is std::shared_ptr and will be automatically cleaned up.
+  // cache is shared_ptr and will be automatically cleaned up.
  const uint64_t kCapacity = 100000;
  auto cache = NewCache(kCapacity, 8, false);

@@ -306,7 +307,7 @@ TEST_P(CacheTest, EvictionPolicy) {
  Insert(200, 201);

  // Frequently used entry must be kept around
-  for (int i = 0; i < kCacheSize + 200; i++) {
+  for (int i = 0; i < kCacheSize + 100; i++) {
    Insert(1000+i, 2000+i);
    ASSERT_EQ(101, Lookup(100));
  }
@@ -359,7 +360,7 @@ TEST_P(CacheTest, EvictionPolicyRef) {
  Insert(303, 104);

  // Insert entries much more than Cache capacity
-  for (int i = 0; i < kCacheSize + 200; i++) {
+  for (int i = 0; i < kCacheSize + 100; i++) {
    Insert(1000 + i, 2000 + i);
  }

@@ -687,8 +688,7 @@ TEST_P(CacheTest, DefaultShardBits) {
 }

 #ifdef SUPPORT_CLOCK_CACHE
-std::shared_ptr<Cache> (*new_clock_cache_func)(size_t, int,
-                                               bool) = NewClockCache;
+shared_ptr<Cache> (*new_clock_cache_func)(size_t, int, bool) = NewClockCache;
 INSTANTIATE_TEST_CASE_P(CacheTestInstance, CacheTest,
                        testing::Values(kLRU, kClock));
 #else
@@ -234,35 +234,38 @@ struct CleanupContext {
 };

 // A cache shard which maintains its own CLOCK cache.
-class ClockCacheShard final : public CacheShard {
+class ClockCacheShard : public CacheShard {
 public:
  // Hash map type.
  typedef tbb::concurrent_hash_map<CacheKey, CacheHandle*, CacheKey> HashTable;

  ClockCacheShard();
-  ~ClockCacheShard() override;
+  ~ClockCacheShard();

  // Interfaces
-  void SetCapacity(size_t capacity) override;
-  void SetStrictCapacityLimit(bool strict_capacity_limit) override;
-  Status Insert(const Slice& key, uint32_t hash, void* value, size_t charge,
-                void (*deleter)(const Slice& key, void* value),
-                Cache::Handle** handle, Cache::Priority priority) override;
-  Cache::Handle* Lookup(const Slice& key, uint32_t hash) override;
+  virtual void SetCapacity(size_t capacity) override;
+  virtual void SetStrictCapacityLimit(bool strict_capacity_limit) override;
+  virtual Status Insert(const Slice& key, uint32_t hash, void* value,
+                        size_t charge,
+                        void (*deleter)(const Slice& key, void* value),
+                        Cache::Handle** handle,
+                        Cache::Priority priority) override;
+  virtual Cache::Handle* Lookup(const Slice& key, uint32_t hash) override;
  // If the entry in in cache, increase reference count and return true.
  // Return false otherwise.
  //
  // Not necessary to hold mutex_ before being called.
-  bool Ref(Cache::Handle* handle) override;
-  bool Release(Cache::Handle* handle, bool force_erase = false) override;
-  void Erase(const Slice& key, uint32_t hash) override;
+  virtual bool Ref(Cache::Handle* handle) override;
+  virtual bool Release(Cache::Handle* handle,
+                       bool force_erase = false) override;
+  virtual void Erase(const Slice& key, uint32_t hash) override;
  bool EraseAndConfirm(const Slice& key, uint32_t hash,
                       CleanupContext* context);
-  size_t GetUsage() const override;
-  size_t GetPinnedUsage() const override;
-  void EraseUnRefEntries() override;
-  void ApplyToAllCacheEntries(void (*callback)(void*, size_t),
-                              bool thread_safe) override;
+  virtual size_t GetUsage() const override;
+  virtual size_t GetPinnedUsage() const override;
+  virtual void EraseUnRefEntries() override;
+  virtual void ApplyToAllCacheEntries(void (*callback)(void*, size_t),
+                                      bool thread_safe) override;

 private:
  static const uint32_t kInCacheBit = 1;
@@ -672,7 +675,7 @@ void ClockCacheShard::EraseUnRefEntries() {
  Cleanup(context);
 }

-class ClockCache final : public ShardedCache {
+class ClockCache : public ShardedCache {
 public:
  ClockCache(size_t capacity, int num_shard_bits, bool strict_capacity_limit)
      : ShardedCache(capacity, num_shard_bits, strict_capacity_limit) {
@@ -682,31 +685,31 @@ class ClockCache final : public ShardedCache {
    SetStrictCapacityLimit(strict_capacity_limit);
  }

-  ~ClockCache() override { delete[] shards_; }
+  virtual ~ClockCache() { delete[] shards_; }

-  const char* Name() const override { return "ClockCache"; }
+  virtual const char* Name() const override { return "ClockCache"; }

-  CacheShard* GetShard(int shard) override {
+  virtual CacheShard* GetShard(int shard) override {
    return reinterpret_cast<CacheShard*>(&shards_[shard]);
  }

-  const CacheShard* GetShard(int shard) const override {
+  virtual const CacheShard* GetShard(int shard) const override {
    return reinterpret_cast<CacheShard*>(&shards_[shard]);
  }

-  void* Value(Handle* handle) override {
+  virtual void* Value(Handle* handle) override {
    return reinterpret_cast<const CacheHandle*>(handle)->value;
  }

-  size_t GetCharge(Handle* handle) const override {
+  virtual size_t GetCharge(Handle* handle) const override {
    return reinterpret_cast<const CacheHandle*>(handle)->charge;
  }

-  uint32_t GetHash(Handle* handle) const override {
+  virtual uint32_t GetHash(Handle* handle) const override {
    return reinterpret_cast<const CacheHandle*>(handle)->hash;
  }

-  void DisownData() override { shards_ = nullptr; }
+  virtual void DisownData() override { shards_ = nullptr; }

 private:
  ClockCacheShard* shards_;
@@ -100,16 +100,14 @@ void LRUHandleTable::Resize() {
 }

 LRUCacheShard::LRUCacheShard(size_t capacity, bool strict_capacity_limit,
-                             double high_pri_pool_ratio,
-                             bool use_adaptive_mutex)
+                             double high_pri_pool_ratio)
    : capacity_(0),
      high_pri_pool_usage_(0),
      strict_capacity_limit_(strict_capacity_limit),
      high_pri_pool_ratio_(high_pri_pool_ratio),
      high_pri_pool_capacity_(0),
      usage_(0),
-      lru_usage_(0),
-      mutex_(use_adaptive_mutex) {
+      lru_usage_(0) {
  // Make empty circular linked list
  lru_.next = &lru_;
  lru_.prev = &lru_;
@@ -463,19 +461,15 @@ std::string LRUCacheShard::GetPrintableOptions() const {
 }

 LRUCache::LRUCache(size_t capacity, int num_shard_bits,
-                   bool strict_capacity_limit, double high_pri_pool_ratio,
-                   std::shared_ptr<MemoryAllocator> allocator,
-                   bool use_adaptive_mutex)
-    : ShardedCache(capacity, num_shard_bits, strict_capacity_limit,
-                   std::move(allocator)) {
+                   bool strict_capacity_limit, double high_pri_pool_ratio)
+    : ShardedCache(capacity, num_shard_bits, strict_capacity_limit) {
  num_shards_ = 1 << num_shard_bits;
  shards_ = reinterpret_cast<LRUCacheShard*>(
      port::cacheline_aligned_alloc(sizeof(LRUCacheShard) * num_shards_));
  size_t per_shard = (capacity + (num_shards_ - 1)) / num_shards_;
  for (int i = 0; i < num_shards_; i++) {
    new (&shards_[i])
-        LRUCacheShard(per_shard, strict_capacity_limit, high_pri_pool_ratio,
-            use_adaptive_mutex);
+        LRUCacheShard(per_shard, strict_capacity_limit, high_pri_pool_ratio);
  }
 }

@@ -543,16 +537,12 @@ double LRUCache::GetHighPriPoolRatio() {
 std::shared_ptr<Cache> NewLRUCache(const LRUCacheOptions& cache_opts) {
  return NewLRUCache(cache_opts.capacity, cache_opts.num_shard_bits,
                     cache_opts.strict_capacity_limit,
-                     cache_opts.high_pri_pool_ratio,
-                     cache_opts.memory_allocator,
-                     cache_opts.use_adaptive_mutex);
+                     cache_opts.high_pri_pool_ratio);
 }

-std::shared_ptr<Cache> NewLRUCache(
-    size_t capacity, int num_shard_bits, bool strict_capacity_limit,
-    double high_pri_pool_ratio,
-    std::shared_ptr<MemoryAllocator> memory_allocator,
-    bool use_adaptive_mutex) {
+std::shared_ptr<Cache> NewLRUCache(size_t capacity, int num_shard_bits,
+                                   bool strict_capacity_limit,
+                                   double high_pri_pool_ratio) {
  if (num_shard_bits >= 20) {
    return nullptr;  // the cache cannot be sharded into too many fine pieces
  }
@@ -564,9 +554,7 @@ std::shared_ptr<Cache> NewLRUCache(
    num_shard_bits = GetDefaultCacheShardBits(capacity);
  }
  return std::make_shared<LRUCache>(capacity, num_shard_bits,
-                                    strict_capacity_limit, high_pri_pool_ratio,
-                                    std::move(memory_allocator),
-                                    use_adaptive_mutex);
+                                    strict_capacity_limit, high_pri_pool_ratio);
 }

 }  // namespace rocksdb
@@ -55,18 +55,10 @@ struct LRUHandle {
                     // cache itself is counted as 1

  // Include the following flags:
-  //   IN_CACHE:         whether this entry is referenced by the hash table.
-  //   IS_HIGH_PRI:      whether this entry is high priority entry.
-  //   IN_HIGH_PRI_POOL: whether this entry is in high-pri pool.
-  //   HAS_HIT:          whether this entry has had any lookups (hits).
-  enum Flags : uint8_t {
-    IN_CACHE = (1 << 0),
-    IS_HIGH_PRI = (1 << 1),
-    IN_HIGH_PRI_POOL = (1 << 2),
-    HAS_HIT = (1 << 3),
-  };
-
-  uint8_t flags;
+  //   in_cache:    whether this entry is referenced by the hash table.
+  //   is_high_pri: whether this entry is high priority entry.
+  //   in_high_pri_pool: whether this entry is in high-pri pool.
+  char flags;

  uint32_t hash;     // Hash of key(); used for fast sharding and comparisons

@@ -82,36 +74,36 @@ struct LRUHandle {
    }
  }

-  bool InCache() const { return flags & IN_CACHE; }
-  bool IsHighPri() const { return flags & IS_HIGH_PRI; }
-  bool InHighPriPool() const { return flags & IN_HIGH_PRI_POOL; }
-  bool HasHit() const { return flags & HAS_HIT; }
+  bool InCache() { return flags & 1; }
+  bool IsHighPri() { return flags & 2; }
+  bool InHighPriPool() { return flags & 4; }
+  bool HasHit() { return flags & 8; }

  void SetInCache(bool in_cache) {
    if (in_cache) {
-      flags |= IN_CACHE;
+      flags |= 1;
    } else {
-      flags &= ~IN_CACHE;
+      flags &= ~1;
    }
  }

  void SetPriority(Cache::Priority priority) {
    if (priority == Cache::Priority::HIGH) {
-      flags |= IS_HIGH_PRI;
+      flags |= 2;
    } else {
-      flags &= ~IS_HIGH_PRI;
+      flags &= ~2;
    }
  }

  void SetInHighPriPool(bool in_high_pri_pool) {
    if (in_high_pri_pool) {
-      flags |= IN_HIGH_PRI_POOL;
+      flags |= 4;
    } else {
-      flags &= ~IN_HIGH_PRI_POOL;
+      flags &= ~4;
    }
  }

-  void SetHit() { flags |= HAS_HIT; }
+  void SetHit() { flags |= 8; }

  void Free() {
    assert((refs == 1 && InCache()) || (refs == 0 && !InCache()));
@@ -165,10 +157,10 @@ class LRUHandleTable {
 };

 // A single shard of sharded cache.
-class ALIGN_AS(CACHE_LINE_SIZE) LRUCacheShard final : public CacheShard {
+class ALIGN_AS(CACHE_LINE_SIZE) LRUCacheShard : public CacheShard {
 public:
  LRUCacheShard(size_t capacity, bool strict_capacity_limit,
-                double high_pri_pool_ratio, bool use_adaptive_mutex);
+                double high_pri_pool_ratio);
  virtual ~LRUCacheShard();

  // Separate from constructor so caller can easily make an array of LRUCache
@@ -284,16 +276,10 @@ class ALIGN_AS(CACHE_LINE_SIZE) LRUCacheShard final : public CacheShard {
  mutable port::Mutex mutex_;
 };

-class LRUCache
-#ifdef NDEBUG
-    final
-#endif
-    : public ShardedCache {
+class LRUCache : public ShardedCache {
 public:
  LRUCache(size_t capacity, int num_shard_bits, bool strict_capacity_limit,
-           double high_pri_pool_ratio,
-           std::shared_ptr<MemoryAllocator> memory_allocator = nullptr,
-           bool use_adaptive_mutex = kDefaultToAdaptiveMutex);
+           double high_pri_pool_ratio);
  virtual ~LRUCache();
  virtual const char* Name() const override { return "LRUCache"; }
  virtual CacheShard* GetShard(int shard) override;
@@ -15,7 +15,7 @@ namespace rocksdb {
 class LRUCacheTest : public testing::Test {
 public:
  LRUCacheTest() {}
-  ~LRUCacheTest() override { DeleteCache(); }
+  ~LRUCacheTest() { DeleteCache(); }

  void DeleteCache() {
    if (cache_ != nullptr) {
@@ -25,13 +25,12 @@ class LRUCacheTest : public testing::Test {
    }
  }

-  void NewCache(size_t capacity, double high_pri_pool_ratio = 0.0,
-                bool use_adaptive_mutex = kDefaultToAdaptiveMutex) {
+  void NewCache(size_t capacity, double high_pri_pool_ratio = 0.0) {
    DeleteCache();
    cache_ = reinterpret_cast<LRUCacheShard*>(
        port::cacheline_aligned_alloc(sizeof(LRUCacheShard)));
    new (cache_) LRUCacheShard(capacity, false /*strict_capcity_limit*/,
-                               high_pri_pool_ratio, use_adaptive_mutex);
+                               high_pri_pool_ratio);
  }

  void Insert(const std::string& key,
@@ -20,10 +20,8 @@
 namespace rocksdb {

 ShardedCache::ShardedCache(size_t capacity, int num_shard_bits,
-                           bool strict_capacity_limit,
-                           std::shared_ptr<MemoryAllocator> allocator)
-    : Cache(std::move(allocator)),
-      num_shard_bits_(num_shard_bits),
+                           bool strict_capacity_limit)
+    : num_shard_bits_(num_shard_bits),
      capacity_(capacity),
      strict_capacity_limit_(strict_capacity_limit),
      last_id_(1) {}
@@ -144,9 +142,6 @@ std::string ShardedCache::GetPrintableOptions() const {
             strict_capacity_limit_);
    ret.append(buffer);
  }
-  snprintf(buffer, kBufferSize, "    memory_allocator : %s\n",
-           memory_allocator() ? memory_allocator()->Name() : "None");
-  ret.append(buffer);
  ret.append(GetShard(0)->GetPrintableOptions());
  return ret;
 }
@@ -47,8 +47,7 @@ class CacheShard {
 // Keys are sharded by the highest num_shard_bits bits of hash value.
 class ShardedCache : public Cache {
 public:
-  ShardedCache(size_t capacity, int num_shard_bits, bool strict_capacity_limit,
-               std::shared_ptr<MemoryAllocator> memory_allocator = nullptr);
+  ShardedCache(size_t capacity, int num_shard_bits, bool strict_capacity_limit);
  virtual ~ShardedCache() = default;
  virtual const char* Name() const override = 0;
  virtual CacheShard* GetShard(int shard) = 0;
@@ -83,7 +82,7 @@ class ShardedCache : public Cache {

 private:
  static inline uint32_t HashSlice(const Slice& s) {
-    return static_cast<uint32_t>(GetSliceNPHash64(s));
+    return Hash(s.data(), s.size(), 0);
  }

  uint32_t Shard(uint32_t hash) {
@@ -0,0 +1,21 @@
+# - Find zlib
+# Find the zlib compression library and includes
+#
+# ZLIB_INCLUDE_DIR - where to find zlib.h, etc.
+# ZLIB_LIBRARIES - List of libraries when using zlib.
+# ZLIB_FOUND - True if zlib found.
+
+find_path(ZLIB_INCLUDE_DIR
+  NAMES zlib.h
+  HINTS ${ZLIB_ROOT_DIR}/include)
+
+find_library(ZLIB_LIBRARIES
+  NAMES z
+  HINTS ${ZLIB_ROOT_DIR}/lib)
+
+include(FindPackageHandleStandardArgs)
+find_package_handle_standard_args(zlib DEFAULT_MSG ZLIB_LIBRARIES ZLIB_INCLUDE_DIR)
+
+mark_as_advanced(
+  ZLIB_LIBRARIES
+  ZLIB_INCLUDE_DIR)
@@ -1,5 +1,4 @@
 #!/usr/bin/env bash
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.

 # Exit on error.
 set -e
@@ -1,4 +1,3 @@
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 import re
 import sys

@@ -18,7 +18,6 @@
 #include "db/event_helpers.h"
 #include "db/internal_stats.h"
 #include "db/merge_helper.h"
-#include "db/range_del_aggregator.h"
 #include "db/table_cache.h"
 #include "db/version_edit.h"
 #include "monitoring/iostats_context_imp.h"
@@ -47,20 +46,18 @@ TableBuilder* NewTableBuilder(
        int_tbl_prop_collector_factories,
    uint32_t column_family_id, const std::string& column_family_name,
    WritableFileWriter* file, const CompressionType compression_type,
-    uint64_t sample_for_compression, const CompressionOptions& compression_opts,
-    int level, const bool skip_filters, const uint64_t creation_time,
-    const uint64_t oldest_key_time, const uint64_t target_file_size,
-    const uint64_t file_creation_time) {
+    const CompressionOptions& compression_opts, int level,
+    const std::string* compression_dict, const bool skip_filters,
+    const uint64_t creation_time, const uint64_t oldest_key_time) {
  assert((column_family_id ==
          TablePropertiesCollectorFactory::Context::kUnknownColumnFamily) ==
         column_family_name.empty());
  return ioptions.table_factory->NewTableBuilder(
      TableBuilderOptions(ioptions, moptions, internal_comparator,
                          int_tbl_prop_collector_factories, compression_type,
-                          sample_for_compression, compression_opts,
-                          skip_filters, column_family_name, level,
-                          creation_time, oldest_key_time, target_file_size,
-                          file_creation_time),
+                          compression_opts, compression_dict, skip_filters,
+                          column_family_name, level, creation_time,
+                          oldest_key_time),
      column_family_id, file);
 }

@@ -68,21 +65,19 @@ Status BuildTable(
    const std::string& dbname, Env* env, const ImmutableCFOptions& ioptions,
    const MutableCFOptions& mutable_cf_options, const EnvOptions& env_options,
    TableCache* table_cache, InternalIterator* iter,
-    std::vector<std::unique_ptr<FragmentedRangeTombstoneIterator>>
-        range_del_iters,
-    FileMetaData* meta, const InternalKeyComparator& internal_comparator,
+    std::unique_ptr<InternalIterator> range_del_iter, FileMetaData* meta,
+    const InternalKeyComparator& internal_comparator,
    const std::vector<std::unique_ptr<IntTblPropCollectorFactory>>*
        int_tbl_prop_collector_factories,
    uint32_t column_family_id, const std::string& column_family_name,
    std::vector<SequenceNumber> snapshots,
    SequenceNumber earliest_write_conflict_snapshot,
    SnapshotChecker* snapshot_checker, const CompressionType compression,
-    uint64_t sample_for_compression, const CompressionOptions& compression_opts,
-    bool paranoid_file_checks, InternalStats* internal_stats,
-    TableFileCreationReason reason, EventLogger* event_logger, int job_id,
-    const Env::IOPriority io_priority, TableProperties* table_properties,
-    int level, const uint64_t creation_time, const uint64_t oldest_key_time,
-    Env::WriteLifeTimeHint write_hint, const uint64_t file_creation_time) {
+    const CompressionOptions& compression_opts, bool paranoid_file_checks,
+    InternalStats* internal_stats, TableFileCreationReason reason,
+    EventLogger* event_logger, int job_id, const Env::IOPriority io_priority,
+    TableProperties* table_properties, int level, const uint64_t creation_time,
+    const uint64_t oldest_key_time, Env::WriteLifeTimeHint write_hint) {
  assert((column_family_id ==
          TablePropertiesCollectorFactory::Context::kUnknownColumnFamily) ==
         column_family_name.empty());
@@ -91,10 +86,12 @@ Status BuildTable(
  Status s;
  meta->fd.file_size = 0;
  iter->SeekToFirst();
-  std::unique_ptr<CompactionRangeDelAggregator> range_del_agg(
-      new CompactionRangeDelAggregator(&internal_comparator, snapshots));
-  for (auto& range_del_iter : range_del_iters) {
-    range_del_agg->AddTombstones(std::move(range_del_iter));
+  std::unique_ptr<RangeDelAggregator> range_del_agg(
+      new RangeDelAggregator(internal_comparator, snapshots));
+  s = range_del_agg->AddTombstones(std::move(range_del_iter));
+  if (!s.ok()) {
+    // may be non-ok if a range tombstone key is unparsable
+    return s;
  }

  std::string fname = TableFileName(ioptions.cf_paths, meta->fd.GetNumber(),
@@ -107,14 +104,9 @@ Status BuildTable(

  if (iter->Valid() || !range_del_agg->IsEmpty()) {
    TableBuilder* builder;
-    std::unique_ptr<WritableFileWriter> file_writer;
-    // Currently we only enable dictionary compression during compaction to the
-    // bottommost level.
-    CompressionOptions compression_opts_for_flush(compression_opts);
-    compression_opts_for_flush.max_dict_bytes = 0;
-    compression_opts_for_flush.zstd_max_train_bytes = 0;
+    unique_ptr<WritableFileWriter> file_writer;
    {
-      std::unique_ptr<WritableFile> file;
+      unique_ptr<WritableFile> file;
 #ifndef NDEBUG
      bool use_direct_writes = env_options.use_direct_writes;
      TEST_SYNC_POINT_CALLBACK("BuildTable:create_file", &use_direct_writes);
@@ -129,16 +121,14 @@ Status BuildTable(
      file->SetIOPriority(io_priority);
      file->SetWriteLifeTimeHint(write_hint);

-      file_writer.reset(
-          new WritableFileWriter(std::move(file), fname, env_options, env,
-                                 ioptions.statistics, ioptions.listeners));
+      file_writer.reset(new WritableFileWriter(std::move(file), env_options,
+                                               ioptions.statistics));
      builder = NewTableBuilder(
          ioptions, mutable_cf_options, internal_comparator,
          int_tbl_prop_collector_factories, column_family_id,
-          column_family_name, file_writer.get(), compression,
-          sample_for_compression, compression_opts_for_flush, level,
-          false /* skip_filters */, creation_time, oldest_key_time,
-          0 /*target_file_size*/, file_creation_time);
+          column_family_name, file_writer.get(), compression, compression_opts,
+          level, nullptr /* compression_dict */, false /* skip_filters */,
+          creation_time, oldest_key_time);
    }

    MergeHelper merge(env, internal_comparator.user_comparator(),
@@ -167,10 +157,8 @@ Status BuildTable(
      }
    }

-    auto range_del_it = range_del_agg->NewIterator();
-    for (range_del_it->SeekToFirst(); range_del_it->Valid();
-         range_del_it->Next()) {
-      auto tombstone = range_del_it->Tombstone();
+    for (auto it = range_del_agg->NewIterator(); it->Valid(); it->Next()) {
+      auto tombstone = it->Tombstone();
      auto kv = tombstone.Serialize();
      builder->Add(kv.first.Encode(), kv.second);
      meta->UpdateBoundariesForRange(kv.first, tombstone.SerializeEndKey(),
@@ -9,7 +9,6 @@
 #include <string>
 #include <utility>
 #include <vector>
-#include "db/range_tombstone_fragmenter.h"
 #include "db/table_properties_collector.h"
 #include "options/cf_options.h"
 #include "rocksdb/comparator.h"
@@ -40,6 +39,8 @@ class InternalStats;
 // @param column_family_name Name of the column family that is also identified
 //    by column_family_id, or empty string if unknown. It must outlive the
 //    TableBuilder returned by this function.
+// @param compression_dict Data for presetting the compression library's
+//    dictionary, or nullptr.
 TableBuilder* NewTableBuilder(
    const ImmutableCFOptions& options, const MutableCFOptions& moptions,
    const InternalKeyComparator& internal_comparator,
@@ -47,11 +48,10 @@ TableBuilder* NewTableBuilder(
        int_tbl_prop_collector_factories,
    uint32_t column_family_id, const std::string& column_family_name,
    WritableFileWriter* file, const CompressionType compression_type,
-    const uint64_t sample_for_compression,
    const CompressionOptions& compression_opts, int level,
+    const std::string* compression_dict = nullptr,
    const bool skip_filters = false, const uint64_t creation_time = 0,
-    const uint64_t oldest_key_time = 0, const uint64_t target_file_size = 0,
-    const uint64_t file_creation_time = 0);
+    const uint64_t oldest_key_time = 0);

 // Build a Table file from the contents of *iter.  The generated file
 // will be named according to number specified in meta. On success, the rest of
@@ -65,23 +65,20 @@ extern Status BuildTable(
    const std::string& dbname, Env* env, const ImmutableCFOptions& options,
    const MutableCFOptions& mutable_cf_options, const EnvOptions& env_options,
    TableCache* table_cache, InternalIterator* iter,
-    std::vector<std::unique_ptr<FragmentedRangeTombstoneIterator>>
-        range_del_iters,
-    FileMetaData* meta, const InternalKeyComparator& internal_comparator,
+    std::unique_ptr<InternalIterator> range_del_iter, FileMetaData* meta,
+    const InternalKeyComparator& internal_comparator,
    const std::vector<std::unique_ptr<IntTblPropCollectorFactory>>*
        int_tbl_prop_collector_factories,
    uint32_t column_family_id, const std::string& column_family_name,
    std::vector<SequenceNumber> snapshots,
    SequenceNumber earliest_write_conflict_snapshot,
    SnapshotChecker* snapshot_checker, const CompressionType compression,
-    const uint64_t sample_for_compression,
    const CompressionOptions& compression_opts, bool paranoid_file_checks,
    InternalStats* internal_stats, TableFileCreationReason reason,
    EventLogger* event_logger = nullptr, int job_id = 0,
    const Env::IOPriority io_priority = Env::IO_HIGH,
    TableProperties* table_properties = nullptr, int level = -1,
    const uint64_t creation_time = 0, const uint64_t oldest_key_time = 0,
-    Env::WriteLifeTimeHint write_hint = Env::WLTH_NOT_SET,
-    const uint64_t file_creation_time = 0);
+    Env::WriteLifeTimeHint write_hint = Env::WLTH_NOT_SET);

 }  // namespace rocksdb
@@ -33,7 +33,6 @@
 #include "rocksdb/utilities/backupable_db.h"
 #include "rocksdb/utilities/checkpoint.h"
 #include "rocksdb/utilities/db_ttl.h"
-#include "rocksdb/utilities/memory_util.h"
 #include "rocksdb/utilities/optimistic_transaction_db.h"
 #include "rocksdb/utilities/transaction.h"
 #include "rocksdb/utilities/transaction_db.h"
@@ -42,10 +41,6 @@
 #include "rocksdb/perf_context.h"
 #include "utilities/merge_operators.h"

-#include <vector>
-#include <unordered_set>
-#include <map>
-
 using rocksdb::BytewiseComparator;
 using rocksdb::Cache;
 using rocksdb::ColumnFamilyDescriptor;
@@ -113,12 +108,8 @@ using rocksdb::TransactionLogIterator;
 using rocksdb::BatchResult;
 using rocksdb::PerfLevel;
 using rocksdb::PerfContext;
-using rocksdb::MemoryUtil;

 using std::shared_ptr;
-using std::vector;
-using std::unordered_set;
-using std::map;

 extern "C" {

@@ -151,20 +142,14 @@ struct rocksdb_writablefile_t    { WritableFile*     rep; };
 struct rocksdb_wal_iterator_t { TransactionLogIterator* rep; };
 struct rocksdb_wal_readoptions_t { TransactionLogIterator::ReadOptions rep; };
 struct rocksdb_filelock_t        { FileLock*         rep; };
-struct rocksdb_logger_t {
-  std::shared_ptr<Logger> rep;
-};
-struct rocksdb_cache_t {
-  std::shared_ptr<Cache> rep;
-};
+struct rocksdb_logger_t          { shared_ptr<Logger>  rep; };
+struct rocksdb_cache_t           { shared_ptr<Cache>   rep; };
 struct rocksdb_livefiles_t       { std::vector<LiveFileMetaData> rep; };
 struct rocksdb_column_family_handle_t  { ColumnFamilyHandle* rep; };
 struct rocksdb_envoptions_t      { EnvOptions        rep; };
 struct rocksdb_ingestexternalfileoptions_t  { IngestExternalFileOptions rep; };
 struct rocksdb_sstfilewriter_t   { SstFileWriter*    rep; };
-struct rocksdb_ratelimiter_t {
-  std::shared_ptr<RateLimiter> rep;
-};
+struct rocksdb_ratelimiter_t     { shared_ptr<RateLimiter>      rep; };
 struct rocksdb_perfcontext_t     { PerfContext*      rep; };
 struct rocksdb_pinnableslice_t {
  PinnableSlice rep;
@@ -208,10 +193,13 @@ struct rocksdb_compactionfilter_t : public CompactionFilter {
  const char* (*name_)(void*);
  unsigned char ignore_snapshots_;

-  ~rocksdb_compactionfilter_t() override { (*destructor_)(state_); }
+  virtual ~rocksdb_compactionfilter_t() {
+    (*destructor_)(state_);
+  }

-  bool Filter(int level, const Slice& key, const Slice& existing_value,
-              std::string* new_value, bool* value_changed) const override {
+  virtual bool Filter(int level, const Slice& key, const Slice& existing_value,
+                      std::string* new_value,
+                      bool* value_changed) const override {
    char* c_new_value = nullptr;
    size_t new_value_length = 0;
    unsigned char c_value_changed = 0;
@@ -228,9 +216,9 @@ struct rocksdb_compactionfilter_t : public CompactionFilter {
    return result;
  }

-  const char* Name() const override { return (*name_)(state_); }
+  virtual const char* Name() const override { return (*name_)(state_); }

-  bool IgnoreSnapshots() const override { return ignore_snapshots_; }
+  virtual bool IgnoreSnapshots() const override { return ignore_snapshots_; }
 };

 struct rocksdb_compactionfilterfactory_t : public CompactionFilterFactory {
@@ -240,9 +228,9 @@ struct rocksdb_compactionfilterfactory_t : public CompactionFilterFactory {
      void*, rocksdb_compactionfiltercontext_t* context);
  const char* (*name_)(void*);

-  ~rocksdb_compactionfilterfactory_t() override { (*destructor_)(state_); }
+  virtual ~rocksdb_compactionfilterfactory_t() { (*destructor_)(state_); }

-  std::unique_ptr<CompactionFilter> CreateCompactionFilter(
+  virtual std::unique_ptr<CompactionFilter> CreateCompactionFilter(
      const CompactionFilter::Context& context) override {
    rocksdb_compactionfiltercontext_t ccontext;
    ccontext.rep = context;
@@ -250,7 +238,7 @@ struct rocksdb_compactionfilterfactory_t : public CompactionFilterFactory {
    return std::unique_ptr<CompactionFilter>(cf);
  }

-  const char* Name() const override { return (*name_)(state_); }
+  virtual const char* Name() const override { return (*name_)(state_); }
 };

 struct rocksdb_comparator_t : public Comparator {
@@ -262,17 +250,20 @@ struct rocksdb_comparator_t : public Comparator {
      const char* b, size_t blen);
  const char* (*name_)(void*);

-  ~rocksdb_comparator_t() override { (*destructor_)(state_); }
+  virtual ~rocksdb_comparator_t() {
+    (*destructor_)(state_);
+  }

-  int Compare(const Slice& a, const Slice& b) const override {
+  virtual int Compare(const Slice& a, const Slice& b) const override {
    return (*compare_)(state_, a.data(), a.size(), b.data(), b.size());
  }

-  const char* Name() const override { return (*name_)(state_); }
+  virtual const char* Name() const override { return (*name_)(state_); }

  // No-ops since the C binding does not support key shortening methods.
-  void FindShortestSeparator(std::string*, const Slice&) const override {}
-  void FindShortSuccessor(std::string* /*key*/) const override {}
+  virtual void FindShortestSeparator(std::string*,
+                                     const Slice&) const override {}
+  virtual void FindShortSuccessor(std::string* /*key*/) const override {}
 };

 struct rocksdb_filterpolicy_t : public FilterPolicy {
@@ -292,11 +283,14 @@ struct rocksdb_filterpolicy_t : public FilterPolicy {
      void*,
      const char* filter, size_t filter_length);

-  ~rocksdb_filterpolicy_t() override { (*destructor_)(state_); }
+  virtual ~rocksdb_filterpolicy_t() {
+    (*destructor_)(state_);
+  }

-  const char* Name() const override { return (*name_)(state_); }
+  virtual const char* Name() const override { return (*name_)(state_); }

-  void CreateFilter(const Slice* keys, int n, std::string* dst) const override {
+  virtual void CreateFilter(const Slice* keys, int n,
+                            std::string* dst) const override {
    std::vector<const char*> key_pointers(n);
    std::vector<size_t> key_sizes(n);
    for (int i = 0; i < n; i++) {
@@ -314,7 +308,8 @@ struct rocksdb_filterpolicy_t : public FilterPolicy {
    }
  }

-  bool KeyMayMatch(const Slice& key, const Slice& filter) const override {
+  virtual bool KeyMayMatch(const Slice& key,
+                           const Slice& filter) const override {
    return (*key_match_)(state_, key.data(), key.size(),
                         filter.data(), filter.size());
  }
@@ -339,12 +334,14 @@ struct rocksdb_mergeoperator_t : public MergeOperator {
      void*,
      const char* value, size_t value_length);

-  ~rocksdb_mergeoperator_t() override { (*destructor_)(state_); }
+  virtual ~rocksdb_mergeoperator_t() {
+    (*destructor_)(state_);
+  }

-  const char* Name() const override { return (*name_)(state_); }
+  virtual const char* Name() const override { return (*name_)(state_); }

-  bool FullMergeV2(const MergeOperationInput& merge_in,
-                   MergeOperationOutput* merge_out) const override {
+  virtual bool FullMergeV2(const MergeOperationInput& merge_in,
+                           MergeOperationOutput* merge_out) const override {
    size_t n = merge_in.operand_list.size();
    std::vector<const char*> operand_pointers(n);
    std::vector<size_t> operand_sizes(n);
@@ -378,10 +375,10 @@ struct rocksdb_mergeoperator_t : public MergeOperator {
    return success;
  }

-  bool PartialMergeMulti(const Slice& key,
-                         const std::deque<Slice>& operand_list,
-                         std::string* new_value,
-                         Logger* /*logger*/) const override {
+  virtual bool PartialMergeMulti(const Slice& key,
+                                 const std::deque<Slice>& operand_list,
+                                 std::string* new_value,
+                                 Logger* /*logger*/) const override {
    size_t operand_count = operand_list.size();
    std::vector<const char*> operand_pointers(operand_count);
    std::vector<size_t> operand_sizes(operand_count);
@@ -432,21 +429,23 @@ struct rocksdb_slicetransform_t : public SliceTransform {
      void*,
      const char* key, size_t length);

-  ~rocksdb_slicetransform_t() override { (*destructor_)(state_); }
+  virtual ~rocksdb_slicetransform_t() {
+    (*destructor_)(state_);
+  }

-  const char* Name() const override { return (*name_)(state_); }
+  virtual const char* Name() const override { return (*name_)(state_); }

-  Slice Transform(const Slice& src) const override {
+  virtual Slice Transform(const Slice& src) const override {
    size_t len;
    char* dst = (*transform_)(state_, src.data(), src.size(), &len);
    return Slice(dst, len);
  }

-  bool InDomain(const Slice& src) const override {
+  virtual bool InDomain(const Slice& src) const override {
    return (*in_domain_)(state_, src.data(), src.size());
  }

-  bool InRange(const Slice& src) const override {
+  virtual bool InRange(const Slice& src) const override {
    return (*in_range_)(state_, src.data(), src.size());
  }
 };
@@ -1070,18 +1069,6 @@ int rocksdb_property_int(
  }
 }

-int rocksdb_property_int_cf(
-    rocksdb_t* db,
-    rocksdb_column_family_handle_t* column_family,
-    const char* propname,
-    uint64_t *out_val) {
-  if (db->rep->GetIntProperty(column_family->rep, Slice(propname), out_val)) {
-    return 0;
-  } else {
-    return -1;
-  }
-}
-
 char* rocksdb_property_value_cf(
    rocksdb_t* db,
    rocksdb_column_family_handle_t* column_family,
@@ -1195,14 +1182,6 @@ void rocksdb_flush(
  SaveError(errptr, db->rep->Flush(options->rep));
 }

-void rocksdb_flush_cf(
-    rocksdb_t* db,
-    const rocksdb_flushoptions_t* options,
-    rocksdb_column_family_handle_t* column_family,
-    char** errptr) {
-  SaveError(errptr, db->rep->Flush(options->rep, column_family->rep));
-}
-
 void rocksdb_disable_file_deletions(
    rocksdb_t* db,
    char** errptr) {
@@ -1501,10 +1480,10 @@ class H : public WriteBatch::Handler {
  void* state_;
  void (*put_)(void*, const char* k, size_t klen, const char* v, size_t vlen);
  void (*deleted_)(void*, const char* k, size_t klen);
-  void Put(const Slice& key, const Slice& value) override {
+  virtual void Put(const Slice& key, const Slice& value) override {
    (*put_)(state_, key.data(), key.size(), value.data(), value.size());
  }
-  void Delete(const Slice& key) override {
+  virtual void Delete(const Slice& key) override {
    (*deleted_)(state_, key.data(), key.size());
  }
 };
@@ -1790,11 +1769,11 @@ rocksdb_iterator_t* rocksdb_writebatch_wi_create_iterator_with_base(
 }

 rocksdb_iterator_t* rocksdb_writebatch_wi_create_iterator_with_base_cf(
-    rocksdb_writebatch_wi_t* wbwi, rocksdb_iterator_t* base_iterator,
+    rocksdb_writebatch_wi_t* wbwi,
+    rocksdb_iterator_t* base_iterator,
    rocksdb_column_family_handle_t* column_family) {
  rocksdb_iterator_t* result = new rocksdb_iterator_t;
-  result->rep =
-      wbwi->rep->NewIteratorWithBase(column_family->rep, base_iterator->rep);
+  result->rep = wbwi->rep->NewIteratorWithBase(column_family->rep, base_iterator->rep);
  delete base_iterator;
  return result;
 }
@@ -2073,15 +2052,6 @@ void rocksdb_set_options(
            db->rep->SetOptions(options_map));
    }

-void rocksdb_set_options_cf(
-    rocksdb_t* db, rocksdb_column_family_handle_t* handle, int count, const char* const keys[], const char* const values[], char** errptr) {
-        std::unordered_map<std::string, std::string> options_map;
-        for (int i=0; i<count; i++)
-            options_map[keys[i]] = values[i];
-        SaveError(errptr,
-            db->rep->SetOptions(handle->rep, options_map));
-    }
-
 rocksdb_options_t* rocksdb_options_create() {
  return new rocksdb_options_t;
 }
@@ -2226,11 +2196,6 @@ void rocksdb_options_set_max_bytes_for_level_base(
  opt->rep.max_bytes_for_level_base = n;
 }

-void rocksdb_options_set_snap_refresh_nanos(rocksdb_options_t* opt,
-                                            uint64_t n) {
-  opt->rep.snap_refresh_nanos = n;
-}
-
 void rocksdb_options_set_level_compaction_dynamic_level_bytes(
    rocksdb_options_t* opt, unsigned char v) {
  opt->rep.level_compaction_dynamic_level_bytes = v;
@@ -2437,7 +2402,7 @@ void rocksdb_options_set_bytes_per_sync(

 void rocksdb_options_set_writable_file_max_buffer_size(rocksdb_options_t* opt,
                                                       uint64_t v) {
-  opt->rep.writable_file_max_buffer_size = static_cast<size_t>(v);
+  opt->rep.writable_file_max_buffer_size = v;
 }

 void rocksdb_options_set_allow_concurrent_memtable_write(rocksdb_options_t* opt,
@@ -2468,13 +2433,11 @@ void rocksdb_options_set_max_write_buffer_number_to_maintain(
  opt->rep.max_write_buffer_number_to_maintain = n;
 }

-void rocksdb_options_set_enable_pipelined_write(rocksdb_options_t* opt,
-                                                unsigned char v) {
+void rocksdb_options_set_enable_pipelined_write(rocksdb_options_t* opt, unsigned char v) {
  opt->rep.enable_pipelined_write = v;
 }

-void rocksdb_options_set_max_subcompactions(rocksdb_options_t* opt,
-                                            uint32_t n) {
+void rocksdb_options_set_max_subcompactions(rocksdb_options_t* opt, uint32_t n) {
  opt->rep.max_subcompactions = n;
 }

@@ -2877,7 +2840,7 @@ rocksdb_compactionfilter_t* rocksdb_compactionfilter_create(
  result->state_ = state;
  result->destructor_ = destructor;
  result->filter_ = filter;
-  result->ignore_snapshots_ = true;
+  result->ignore_snapshots_ = false;
  result->name_ = name;
  return result;
 }
@@ -2977,7 +2940,7 @@ rocksdb_filterpolicy_t* rocksdb_filterpolicy_create_bloom_format(int bits_per_ke
  // supplied C functions.
  struct Wrapper : public rocksdb_filterpolicy_t {
    const FilterPolicy* rep_;
-    ~Wrapper() override { delete rep_; }
+    ~Wrapper() { delete rep_; }
    const char* Name() const override { return rep_->Name(); }
    void CreateFilter(const Slice* keys, int n,
                      std::string* dst) const override {
@@ -3432,7 +3395,7 @@ void rocksdb_slicetransform_destroy(rocksdb_slicetransform_t* st) {

 struct Wrapper : public rocksdb_slicetransform_t {
  const SliceTransform* rep_;
-  ~Wrapper() override { delete rep_; }
+  ~Wrapper() { delete rep_; }
  const char* Name() const override { return rep_->Name(); }
  Slice Transform(const Slice& src) const override {
    return rep_->Transform(src);
@@ -3570,18 +3533,6 @@ const char* rocksdb_livefiles_largestkey(
  return lf->rep[index].largestkey.data();
 }

-uint64_t rocksdb_livefiles_entries(
-    const rocksdb_livefiles_t* lf,
-    int index) {
-  return lf->rep[index].num_entries;
-}
-
-uint64_t rocksdb_livefiles_deletions(
-    const rocksdb_livefiles_t* lf,
-    int index) {
-  return lf->rep[index].num_deletions;
-}
-
 extern void rocksdb_livefiles_destroy(
  const rocksdb_livefiles_t* lf) {
  delete lf;
@@ -3727,38 +3678,6 @@ rocksdb_transactiondb_t* rocksdb_transactiondb_open(
  return result;
 }

-rocksdb_transactiondb_t* rocksdb_transactiondb_open_column_families(
-    const rocksdb_options_t* options,
-    const rocksdb_transactiondb_options_t* txn_db_options, const char* name,
-    int num_column_families, const char** column_family_names,
-    const rocksdb_options_t** column_family_options,
-    rocksdb_column_family_handle_t** column_family_handles, char** errptr) {
-  std::vector<ColumnFamilyDescriptor> column_families;
-  for (int i = 0; i < num_column_families; i++) {
-    column_families.push_back(ColumnFamilyDescriptor(
-        std::string(column_family_names[i]),
-        ColumnFamilyOptions(column_family_options[i]->rep)));
-  }
-
-  TransactionDB* txn_db;
-  std::vector<ColumnFamilyHandle*> handles;
-  if (SaveError(errptr, TransactionDB::Open(options->rep, txn_db_options->rep,
-                                            std::string(name), column_families,
-                                            &handles, &txn_db))) {
-    return nullptr;
-  }
-
-  for (size_t i = 0; i < handles.size(); i++) {
-    rocksdb_column_family_handle_t* c_handle =
-        new rocksdb_column_family_handle_t;
-    c_handle->rep = handles[i];
-    column_family_handles[i] = c_handle;
-  }
-  rocksdb_transactiondb_t* result = new rocksdb_transactiondb_t;
-  result->rep = txn_db;
-  return result;
-}
-
 const rocksdb_snapshot_t* rocksdb_transactiondb_create_snapshot(
    rocksdb_transactiondb_t* txn_db) {
  rocksdb_snapshot_t* result = new rocksdb_snapshot_t;
@@ -3879,26 +3798,6 @@ char* rocksdb_transaction_get_for_update(rocksdb_transaction_t* txn,
  return result;
 }

-char* rocksdb_transaction_get_for_update_cf(
-    rocksdb_transaction_t* txn, const rocksdb_readoptions_t* options,
-    rocksdb_column_family_handle_t* column_family, const char* key, size_t klen,
-    size_t* vlen, unsigned char exclusive, char** errptr) {
-  char* result = nullptr;
-  std::string tmp;
-  Status s = txn->rep->GetForUpdate(options->rep, column_family->rep,
-                                    Slice(key, klen), &tmp, exclusive);
-  if (s.ok()) {
-    *vlen = tmp.size();
-    result = CopyString(tmp);
-  } else {
-    *vlen = 0;
-    if (!s.IsNotFound()) {
-      SaveError(errptr, s);
-    }
-  }
-  return result;
-}
-
 // Read a key outside a transaction
 char* rocksdb_transactiondb_get(
    rocksdb_transactiondb_t* txn_db,
@@ -3991,14 +3890,6 @@ void rocksdb_transaction_merge(rocksdb_transaction_t* txn, const char* key,
  SaveError(errptr, txn->rep->Merge(Slice(key, klen), Slice(val, vlen)));
 }

-void rocksdb_transaction_merge_cf(rocksdb_transaction_t* txn,
-                                  rocksdb_column_family_handle_t* column_family,
-                                  const char* key, size_t klen, const char* val,
-                                  size_t vlen, char** errptr) {
-  SaveError(errptr, txn->rep->Merge(column_family->rep, Slice(key, klen),
-                                    Slice(val, vlen)));
-}
-
 // Merge a key outside a transaction
 void rocksdb_transactiondb_merge(rocksdb_transactiondb_t* txn_db,
                                 const rocksdb_writeoptions_t* options,
@@ -4008,14 +3899,6 @@ void rocksdb_transactiondb_merge(rocksdb_transactiondb_t* txn_db,
                                       Slice(val, vlen)));
 }

-void rocksdb_transactiondb_merge_cf(
-    rocksdb_transactiondb_t* txn_db, const rocksdb_writeoptions_t* options,
-    rocksdb_column_family_handle_t* column_family, const char* key, size_t klen,
-    const char* val, size_t vlen, char** errptr) {
-  SaveError(errptr, txn_db->rep->Merge(options->rep, column_family->rep,
-                                       Slice(key, klen), Slice(val, vlen)));
-}
-
 // Delete a key inside a transaction
 void rocksdb_transaction_delete(rocksdb_transaction_t* txn, const char* key,
                                size_t klen, char** errptr) {
@@ -4068,14 +3951,6 @@ rocksdb_iterator_t* rocksdb_transactiondb_create_iterator(
  return result;
 }

-rocksdb_iterator_t* rocksdb_transactiondb_create_iterator_cf(
-    rocksdb_transactiondb_t* txn_db, const rocksdb_readoptions_t* options,
-    rocksdb_column_family_handle_t* column_family) {
-  rocksdb_iterator_t* result = new rocksdb_iterator_t;
-  result->rep = txn_db->rep->NewIterator(options->rep, column_family->rep);
-  return result;
-}
-
 void rocksdb_transactiondb_close(rocksdb_transactiondb_t* txn_db) {
  delete txn_db->rep;
  delete txn_db;
@@ -4224,98 +4099,6 @@ const char* rocksdb_pinnableslice_value(const rocksdb_pinnableslice_t* v,
  *vlen = v->rep.size();
  return v->rep.data();
 }
-
-// container to keep databases and caches in order to use rocksdb::MemoryUtil
-struct rocksdb_memory_consumers_t {
-  std::vector<rocksdb_t*> dbs;
-  std::unordered_set<rocksdb_cache_t*> caches;
-};
-
-// initializes new container of memory consumers
-rocksdb_memory_consumers_t* rocksdb_memory_consumers_create() {
-  return new rocksdb_memory_consumers_t;
-}
-
-// adds datatabase to the container of memory consumers
-void rocksdb_memory_consumers_add_db(rocksdb_memory_consumers_t* consumers,
-                                     rocksdb_t* db) {
-  consumers->dbs.push_back(db);
-}
-
-// adds cache to the container of memory consumers
-void rocksdb_memory_consumers_add_cache(rocksdb_memory_consumers_t* consumers,
-                                        rocksdb_cache_t* cache) {
-  consumers->caches.insert(cache);
-}
-
-// deletes container with memory consumers
-void rocksdb_memory_consumers_destroy(rocksdb_memory_consumers_t* consumers) {
-  delete consumers;
-}
-
-// contains memory usage statistics provided by rocksdb::MemoryUtil
-struct rocksdb_memory_usage_t {
-  uint64_t mem_table_total;
-  uint64_t mem_table_unflushed;
-  uint64_t mem_table_readers_total;
-  uint64_t cache_total;
-};
-
-// estimates amount of memory occupied by consumers (dbs and caches)
-rocksdb_memory_usage_t* rocksdb_approximate_memory_usage_create(
-    rocksdb_memory_consumers_t* consumers, char** errptr) {
-
-  vector<DB*> dbs;
-  for (auto db : consumers->dbs) {
-    dbs.push_back(db->rep);
-  }
-
-  unordered_set<const Cache*> cache_set;
-  for (auto cache : consumers->caches) {
-    cache_set.insert(const_cast<const Cache*>(cache->rep.get()));
-  }
-
-  std::map<rocksdb::MemoryUtil::UsageType, uint64_t> usage_by_type;
-
-  auto status = MemoryUtil::GetApproximateMemoryUsageByType(dbs, cache_set,
-                                                            &usage_by_type);
-  if (SaveError(errptr, status)) {
-    return nullptr;
-  }
-
-  auto result = new rocksdb_memory_usage_t;
-  result->mem_table_total = usage_by_type[MemoryUtil::kMemTableTotal];
-  result->mem_table_unflushed = usage_by_type[MemoryUtil::kMemTableUnFlushed];
-  result->mem_table_readers_total = usage_by_type[MemoryUtil::kTableReadersTotal];
-  result->cache_total = usage_by_type[MemoryUtil::kCacheTotal];
-  return result;
-}
-
-uint64_t rocksdb_approximate_memory_usage_get_mem_table_total(
-    rocksdb_memory_usage_t* memory_usage) {
-  return memory_usage->mem_table_total;
-}
-
-uint64_t rocksdb_approximate_memory_usage_get_mem_table_unflushed(
-    rocksdb_memory_usage_t* memory_usage) {
-  return memory_usage->mem_table_unflushed;
-}
-
-uint64_t rocksdb_approximate_memory_usage_get_mem_table_readers_total(
-    rocksdb_memory_usage_t* memory_usage) {
-  return memory_usage->mem_table_readers_total;
-}
-
-uint64_t rocksdb_approximate_memory_usage_get_cache_total(
-    rocksdb_memory_usage_t* memory_usage) {
-  return memory_usage->cache_total;
-}
-
-// deletes container with memory usage estimates
-void rocksdb_approximate_memory_usage_destroy(rocksdb_memory_usage_t* usage) {
-  delete usage;
-}
-
 }  // end extern "C"

 #endif  // !ROCKSDB_LITE
@@ -1,7 +1,6 @@
 /* Copyright (c) 2011 The LevelDB Authors. All rights reserved.
   Use of this source code is governed by a BSD-style license that can be
   found in the LICENSE file. See the AUTHORS file for names of contributors. */
-// Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.

 #include <stdio.h>

@@ -20,8 +19,11 @@

 // Can not use port/port.h macros as this is a c file
 #ifdef OS_WIN
+
 #include <windows.h>

+#define snprintf _snprintf
+
 // Ok for uniqueness
 int geteuid() {
  int result = 0;
@@ -32,11 +34,6 @@ int geteuid() {
  return result;
 }

-// VS < 2015
-#if defined(_MSC_VER) && (_MSC_VER < 1900)
-#define snprintf _snprintf
-#endif
-
 #endif

 const char* phase = "";
@@ -50,19 +47,12 @@ static void StartPhase(const char* name) {
  fprintf(stderr, "=== Test %s\n", name);
  phase = name;
 }
-#ifdef _MSC_VER
-#pragma warning(push)
-#pragma warning (disable: 4996) // getenv security warning
-#endif
 static const char* GetTempDir(void) {
    const char* ret = getenv("TEST_TMPDIR");
    if (ret == NULL || ret[0] == '\0')
        ret = "/tmp";
    return ret;
 }
-#ifdef _MSC_VER
-#pragma warning(pop)
-#endif

 #define CheckNoError(err)                                               \
  if ((err) != NULL) {                                                  \
@@ -653,7 +643,7 @@ int main(int argc, char** argv) {
    rocksdb_sstfilewriter_t* writer =
        rocksdb_sstfilewriter_create(env_opt, io_options);

-    remove(sstfilename);
+    unlink(sstfilename);
    rocksdb_sstfilewriter_open(writer, sstfilename, &err);
    CheckNoError(err);
    rocksdb_sstfilewriter_put(writer, "sstk1", 5, "v1", 2, &err);
@@ -674,7 +664,7 @@ int main(int argc, char** argv) {
    CheckGet(db, roptions, "sstk2", "v2");
    CheckGet(db, roptions, "sstk3", "v3");

-    remove(sstfilename);
+    unlink(sstfilename);
    rocksdb_sstfilewriter_open(writer, sstfilename, &err);
    CheckNoError(err);
    rocksdb_sstfilewriter_put(writer, "sstk2", 5, "v4", 2, &err);
@@ -918,8 +908,7 @@ int main(int argc, char** argv) {
    rocksdb_writebatch_wi_t* wbi = rocksdb_writebatch_wi_create(0, 1);
    rocksdb_writebatch_wi_put(wbi, "bar", 3, "b", 1);
    rocksdb_writebatch_wi_delete(wbi, "foo", 3);
-    rocksdb_iterator_t* iter =
-        rocksdb_writebatch_wi_create_iterator_with_base(wbi, base_iter);
+    rocksdb_iterator_t* iter = rocksdb_writebatch_wi_create_iterator_with_base(wbi, base_iter);
    CheckCondition(!rocksdb_iter_valid(iter));
    rocksdb_iter_seek_to_first(iter);
    CheckCondition(rocksdb_iter_valid(iter));
@@ -1188,12 +1177,6 @@ int main(int argc, char** argv) {
    rocksdb_put_cf(db, woptions, handles[1], "foo", 3, "hello", 5, &err);
    CheckNoError(err);

-    rocksdb_flushoptions_t *flush_options = rocksdb_flushoptions_create();
-    rocksdb_flushoptions_set_wait(flush_options, 1);
-    rocksdb_flush_cf(db, flush_options, handles[1], &err);
-    CheckNoError(err)
-    rocksdb_flushoptions_destroy(flush_options);
-
    CheckGetCF(db, roptions, handles[1], "foo", "hello");
    CheckPinGetCF(db, roptions, handles[1], "foo", "hello");

@@ -1351,47 +1334,6 @@ int main(int argc, char** argv) {
    rocksdb_destroy_db(options, dbname, &err);
  }

-  // Check memory usage stats
-  StartPhase("approximate_memory_usage");
-  {
-    // Create database
-    db = rocksdb_open(options, dbname, &err);
-    CheckNoError(err);
-
-    rocksdb_memory_consumers_t* consumers;
-    consumers = rocksdb_memory_consumers_create();
-    rocksdb_memory_consumers_add_db(consumers, db);
-    rocksdb_memory_consumers_add_cache(consumers, cache);
-
-    // take memory usage report before write-read operation
-    rocksdb_memory_usage_t* mu1;
-    mu1 = rocksdb_approximate_memory_usage_create(consumers, &err);
-    CheckNoError(err);
-
-    // Put data (this should affect memtables)
-    rocksdb_put(db, woptions, "memory", 6, "test", 4, &err);
-    CheckNoError(err);
-    CheckGet(db, roptions, "memory", "test");
-
-    // take memory usage report after write-read operation
-    rocksdb_memory_usage_t* mu2;
-    mu2 = rocksdb_approximate_memory_usage_create(consumers, &err);
-    CheckNoError(err);
-
-    // amount of memory used within memtables should grow
-    CheckCondition(rocksdb_approximate_memory_usage_get_mem_table_total(mu2) >=
-                   rocksdb_approximate_memory_usage_get_mem_table_total(mu1));
-    CheckCondition(rocksdb_approximate_memory_usage_get_mem_table_unflushed(mu2) >=
-                   rocksdb_approximate_memory_usage_get_mem_table_unflushed(mu1));
-
-    rocksdb_memory_consumers_destroy(consumers);
-    rocksdb_approximate_memory_usage_destroy(mu1);
-    rocksdb_approximate_memory_usage_destroy(mu2);
-    rocksdb_close(db);
-    rocksdb_destroy_db(options, dbname, &err);
-    CheckNoError(err);
-  }
-
  StartPhase("cuckoo_options");
  {
    rocksdb_cuckoo_table_options_t* cuckoo_options;
@@ -1535,7 +1477,7 @@ int main(int argc, char** argv) {
    const rocksdb_snapshot_t* snapshot;
    snapshot = rocksdb_transactiondb_create_snapshot(txn_db);
    rocksdb_readoptions_set_snapshot(roptions, snapshot);
-
+  
    rocksdb_transactiondb_put(txn_db, woptions, "foo", 3, "hey", 3,  &err);
    CheckNoError(err);

@@ -1733,7 +1675,7 @@ int main(int argc, char** argv) {
    db = rocksdb_open(options, dbname, &err);
    CheckNoError(err);
  }
-
+  
  StartPhase("cleanup");
  rocksdb_close(db);
  rocksdb_options_destroy(options);
@@ -20,12 +20,10 @@
 #include <limits>

 #include "db/compaction_picker.h"
-#include "db/compaction_picker_fifo.h"
 #include "db/compaction_picker_universal.h"
 #include "db/db_impl.h"
 #include "db/internal_stats.h"
 #include "db/job_context.h"
-#include "db/range_del_aggregator.h"
 #include "db/table_properties_collector.h"
 #include "db/version_set.h"
 #include "db/write_controller.h"
@@ -63,24 +61,12 @@ ColumnFamilyHandleImpl::~ColumnFamilyHandleImpl() {
    JobContext job_context(0);
    mutex_->Lock();
    if (cfd_->Unref()) {
-      bool dropped = cfd_->IsDropped();
-
      delete cfd_;
-
-      if (dropped) {
-        db_->FindObsoleteFiles(&job_context, false, true);
-      }
    }
+    db_->FindObsoleteFiles(&job_context, false, true);
    mutex_->Unlock();
    if (job_context.HaveSomethingToDelete()) {
-      bool defer_purge =
-          db_->immutable_db_options().avoid_unnecessary_blocking_io;
-      db_->PurgeObsoleteFiles(job_context, defer_purge);
-      if (defer_purge) {
-        mutex_->Lock();
-        db_->SchedulePurge();
-        mutex_->Unlock();
-      }
+      db_->PurgeObsoleteFiles(job_context);
    }
    job_context.Clean();
  }
@@ -119,6 +105,9 @@ void GetIntTblPropCollectorFactory(
    int_tbl_prop_collector_factories->emplace_back(
        new UserKeyTablePropertiesCollectorFactory(collector_factories[i]));
  }
+  // Add collector to collect internal key statistics
+  int_tbl_prop_collector_factories->emplace_back(
+      new InternalKeyPropertiesCollectorFactory);
 }

 Status CheckCompressionSupported(const ColumnFamilyOptions& cf_options) {
@@ -141,10 +130,14 @@ Status CheckCompressionSupported(const ColumnFamilyOptions& cf_options) {
    }
  }
  if (cf_options.compression_opts.zstd_max_train_bytes > 0) {
-    if (!ZSTD_TrainDictionarySupported()) {
+    if (!CompressionTypeSupported(CompressionType::kZSTD)) {
+      // Dictionary trainer is available since v0.6.1, but ZSTD was marked
+      // stable only since v0.8.0. For now we enable the feature in stable
+      // versions only.
      return Status::InvalidArgument(
-          "zstd dictionary trainer cannot be used because ZSTD 1.1.3+ "
-          "is not linked with the binary.");
+          "zstd dictionary trainer cannot be used because " +
+          CompressionTypeToString(CompressionType::kZSTD) +
+          " is not linked with the binary.");
    }
    if (cf_options.compression_opts.max_dict_bytes == 0) {
      return Status::InvalidArgument(
@@ -952,16 +945,26 @@ Status ColumnFamilyData::RangesOverlapWithMemtables(
  super_version->imm->AddIterators(read_opts, &merge_iter_builder);
  ScopedArenaIterator memtable_iter(merge_iter_builder.Finish());

-  auto read_seq = super_version->current->version_set()->LastSequence();
-  ReadRangeDelAggregator range_del_agg(&internal_comparator_, read_seq);
+  std::vector<InternalIterator*> memtable_range_del_iters;
  auto* active_range_del_iter =
-      super_version->mem->NewRangeTombstoneIterator(read_opts, read_seq);
-  range_del_agg.AddTombstones(
-      std::unique_ptr<FragmentedRangeTombstoneIterator>(active_range_del_iter));
-  super_version->imm->AddRangeTombstoneIterators(read_opts, nullptr /* arena */,
-                                                 &range_del_agg);
-
+      super_version->mem->NewRangeTombstoneIterator(read_opts);
+  if (active_range_del_iter != nullptr) {
+    memtable_range_del_iters.push_back(active_range_del_iter);
+  }
+  super_version->imm->AddRangeTombstoneIterators(read_opts,
+                                                 &memtable_range_del_iters);
+  RangeDelAggregator range_del_agg(internal_comparator_, {} /* snapshots */,
+                                   false /* collapse_deletions */);
  Status status;
+  {
+    std::unique_ptr<InternalIterator> memtable_range_del_iter(
+        NewMergingIterator(&internal_comparator_,
+                           memtable_range_del_iters.empty()
+                               ? nullptr
+                               : &memtable_range_del_iters[0],
+                           static_cast<int>(memtable_range_del_iters.size())));
+    status = range_del_agg.AddTombstones(std::move(memtable_range_del_iter));
+  }
  for (size_t i = 0; i < ranges.size() && status.ok() && !*overlap; ++i) {
    auto* vstorage = super_version->current->storage_info();
    auto* ucmp = vstorage->InternalComparator()->user_comparator();
@@ -994,14 +997,13 @@ const int ColumnFamilyData::kCompactToBaseLevel = -2;

 Compaction* ColumnFamilyData::CompactRange(
    const MutableCFOptions& mutable_cf_options, int input_level,
-    int output_level, const CompactRangeOptions& compact_range_options,
+    int output_level, uint32_t output_path_id, uint32_t max_subcompactions,
    const InternalKey* begin, const InternalKey* end,
-    InternalKey** compaction_end, bool* conflict,
-    uint64_t max_file_num_to_ignore) {
+    InternalKey** compaction_end, bool* conflict) {
  auto* result = compaction_picker_->CompactRange(
      GetName(), mutable_cf_options, current_->storage_info(), input_level,
-      output_level, compact_range_options, begin, end, compaction_end, conflict,
-      max_file_num_to_ignore);
+      output_level, output_path_id, max_subcompactions, begin, end,
+      compaction_end, conflict);
  if (result != nullptr) {
    result->SetInputVersion(current_);
  }
@@ -296,10 +296,9 @@ class ColumnFamilyData {
  // REQUIRES: DB mutex held
  Compaction* CompactRange(const MutableCFOptions& mutable_cf_options,
                           int input_level, int output_level,
-                           const CompactRangeOptions& compact_range_options,
+                           uint32_t output_path_id, uint32_t max_subcompactions,
                           const InternalKey* begin, const InternalKey* end,
-                           InternalKey** compaction_end, bool* manual_conflict,
-                           uint64_t max_file_num_to_ignore);
+                           InternalKey** compaction_end, bool* manual_conflict);

  CompactionPicker* compaction_picker() { return compaction_picker_.get(); }
  // thread-safe
@@ -384,8 +383,6 @@ class ColumnFamilyData {

  Directory* GetDataDir(size_t path_id) const;

-  ThreadLocalPtr* TEST_GetLocalSV() { return local_sv_.get(); }
-
 private:
  friend class ColumnFamilySet;
  ColumnFamilyData(uint32_t id, const std::string& name,
@@ -14,7 +14,6 @@

 #include "db/db_impl.h"
 #include "db/db_test_util.h"
-#include "memtable/hash_skiplist_rep.h"
 #include "options/options_parser.h"
 #include "port/port.h"
 #include "rocksdb/db.h"
@@ -48,7 +47,7 @@ class EnvCounter : public EnvWrapper {
  int GetNumberOfNewWritableFileCalls() {
    return num_new_writable_file_;
  }
-  Status NewWritableFile(const std::string& f, std::unique_ptr<WritableFile>* r,
+  Status NewWritableFile(const std::string& f, unique_ptr<WritableFile>* r,
                         const EnvOptions& soptions) override {
    ++num_new_writable_file_;
    return EnvWrapper::NewWritableFile(f, r, soptions);
@@ -69,7 +68,7 @@ class ColumnFamilyTestBase : public testing::Test {
    DestroyDB(dbname_, Options(db_options_, column_family_options_));
  }

-  ~ColumnFamilyTestBase() override {
+  virtual ~ColumnFamilyTestBase() {
    std::vector<ColumnFamilyDescriptor> column_families;
    for (auto h : handles_) {
      ColumnFamilyDescriptor cfdescriptor;
@@ -487,9 +486,9 @@ class ColumnFamilyTestBase : public testing::Test {
  void CopyFile(const std::string& source, const std::string& destination,
                uint64_t size = 0) {
    const EnvOptions soptions;
-    std::unique_ptr<SequentialFile> srcfile;
+    unique_ptr<SequentialFile> srcfile;
    ASSERT_OK(env_->NewSequentialFile(source, &srcfile, soptions));
-    std::unique_ptr<WritableFile> destfile;
+    unique_ptr<WritableFile> destfile;
    ASSERT_OK(env_->NewWritableFile(destination, &destfile, soptions));

    if (size == 0) {
@@ -514,16 +513,6 @@ class ColumnFamilyTestBase : public testing::Test {
    return static_cast<int>(files.size());
  }

-  void RecalculateWriteStallConditions(ColumnFamilyData* cfd,
-      const MutableCFOptions& mutable_cf_options)  {
-    // add lock to avoid race condition between
-    // `RecalculateWriteStallConditions` which writes to CFStats and
-    // background `DBImpl::DumpStats()` threads which read CFStats
-    dbfull()->TEST_LockMutex();
-    cfd->RecalculateWriteStallConditions(mutable_cf_options);
-    dbfull()-> TEST_UnlockMutex();
-  }
-
  std::vector<ColumnFamilyHandle*> handles_;
  std::vector<std::string> names_;
  std::vector<std::set<std::string>> keys_;
@@ -1207,32 +1196,29 @@ TEST_P(ColumnFamilyTest, DifferentWriteBufferSizes) {
 }
 #endif  // !ROCKSDB_LITE

-// The test is commented out because we want to test that snapshot is
-// not created for memtables not supported it, but There isn't a memtable
-// that doesn't support snapshot right now. If we have one later, we can
-// re-enable the test.
-//
-// #ifndef ROCKSDB_LITE  // Cuckoo is not supported in lite
-//   TEST_P(ColumnFamilyTest, MemtableNotSupportSnapshot) {
-//   db_options_.allow_concurrent_memtable_write = false;
-//   Open();
-//   auto* s1 = dbfull()->GetSnapshot();
-//   ASSERT_TRUE(s1 != nullptr);
-//   dbfull()->ReleaseSnapshot(s1);
+#ifndef ROCKSDB_LITE  // Cuckoo is not supported in lite
+TEST_P(ColumnFamilyTest, MemtableNotSupportSnapshot) {
+  db_options_.allow_concurrent_memtable_write = false;
+  Open();
+  auto* s1 = dbfull()->GetSnapshot();
+  ASSERT_TRUE(s1 != nullptr);
+  dbfull()->ReleaseSnapshot(s1);

-//   // Add a column family that doesn't support snapshot
-//   ColumnFamilyOptions first;
-//   first.memtable_factory.reset(new DummyMemtableNotSupportingSnapshot());
-//   CreateColumnFamilies({"first"}, {first});
-//   auto* s2 = dbfull()->GetSnapshot();
-//   ASSERT_TRUE(s2 == nullptr);
+  // Add a column family that doesn't support snapshot
+  ColumnFamilyOptions first;
+  first.memtable_factory.reset(NewHashCuckooRepFactory(1024 * 1024));
+  CreateColumnFamilies({"first"}, {first});
+  auto* s2 = dbfull()->GetSnapshot();
+  ASSERT_TRUE(s2 == nullptr);

-//   // Add a column family that supports snapshot. Snapshot stays not
-//   supported. ColumnFamilyOptions second; CreateColumnFamilies({"second"},
-//   {second}); auto* s3 = dbfull()->GetSnapshot(); ASSERT_TRUE(s3 == nullptr);
-//   Close();
-// }
-// #endif  // !ROCKSDB_LITE
+  // Add a column family that supports snapshot. Snapshot stays not supported.
+  ColumnFamilyOptions second;
+  CreateColumnFamilies({"second"}, {second});
+  auto* s3 = dbfull()->GetSnapshot();
+  ASSERT_TRUE(s3 == nullptr);
+  Close();
+}
+#endif  // !ROCKSDB_LITE

 class TestComparator : public Comparator {
  int Compare(const rocksdb::Slice& /*a*/,
@@ -2046,7 +2032,7 @@ TEST_P(ColumnFamilyTest, SameCFAutomaticManualCompactions) {
 }
 #endif  // !ROCKSDB_LITE

-#ifndef ROCKSDB_LITE  // Tailing iterator not supported
+#ifndef ROCKSDB_LITE  // Tailing interator not supported
 namespace {
 std::string IterStatus(Iterator* iter) {
  std::string result;
@@ -2514,139 +2500,139 @@ TEST_P(ColumnFamilyTest, WriteStallSingleColumnFamily) {
  mutable_cf_options.disable_auto_compactions = false;

  vstorage->TEST_set_estimated_compaction_needed_bytes(50);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());

  vstorage->TEST_set_estimated_compaction_needed_bytes(201);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate, GetDbDelayedWriteRate());
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->TEST_set_estimated_compaction_needed_bytes(400);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25, GetDbDelayedWriteRate());
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->TEST_set_estimated_compaction_needed_bytes(500);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25 / 1.25, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(450);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(205);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(202);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(201);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(198);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());

  vstorage->TEST_set_estimated_compaction_needed_bytes(399);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(599);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(2001);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->TEST_set_estimated_compaction_needed_bytes(3001);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());

  vstorage->TEST_set_estimated_compaction_needed_bytes(390);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(100);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());

  vstorage->set_l0_delay_trigger_count(100);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate, GetDbDelayedWriteRate());
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->set_l0_delay_trigger_count(101);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25, GetDbDelayedWriteRate());

  vstorage->set_l0_delay_trigger_count(0);
  vstorage->TEST_set_estimated_compaction_needed_bytes(300);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25 / 1.25, GetDbDelayedWriteRate());

  vstorage->set_l0_delay_trigger_count(101);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25 / 1.25 / 1.25, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(200);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25 / 1.25, GetDbDelayedWriteRate());

  vstorage->set_l0_delay_trigger_count(0);
  vstorage->TEST_set_estimated_compaction_needed_bytes(0);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());

  mutable_cf_options.disable_auto_compactions = true;
  dbfull()->TEST_write_controler().set_delayed_write_rate(kBaseRate);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());

  vstorage->set_l0_delay_trigger_count(50);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(0, GetDbDelayedWriteRate());
@@ -2654,7 +2640,7 @@ TEST_P(ColumnFamilyTest, WriteStallSingleColumnFamily) {

  vstorage->set_l0_delay_trigger_count(60);
  vstorage->TEST_set_estimated_compaction_needed_bytes(300);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(0, GetDbDelayedWriteRate());
@@ -2663,14 +2649,14 @@ TEST_P(ColumnFamilyTest, WriteStallSingleColumnFamily) {
  mutable_cf_options.disable_auto_compactions = false;
  vstorage->set_l0_delay_trigger_count(70);
  vstorage->TEST_set_estimated_compaction_needed_bytes(500);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate, GetDbDelayedWriteRate());

  vstorage->set_l0_delay_trigger_count(71);
  vstorage->TEST_set_estimated_compaction_needed_bytes(501);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25, GetDbDelayedWriteRate());
@@ -2695,31 +2681,31 @@ TEST_P(ColumnFamilyTest, CompactionSpeedupSingleColumnFamily) {
  mutable_cf_options.hard_pending_compaction_bytes_limit = 2000;

  vstorage->TEST_set_estimated_compaction_needed_bytes(40);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->TEST_set_estimated_compaction_needed_bytes(50);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->TEST_set_estimated_compaction_needed_bytes(300);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->TEST_set_estimated_compaction_needed_bytes(45);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->set_l0_delay_trigger_count(7);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->set_l0_delay_trigger_count(9);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->set_l0_delay_trigger_count(6);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());

  // Speed up threshold = min(4 * 2, 4 + (12 - 4)/4) = 6
@@ -2728,15 +2714,15 @@ TEST_P(ColumnFamilyTest, CompactionSpeedupSingleColumnFamily) {
  mutable_cf_options.level0_stop_writes_trigger = 30;

  vstorage->set_l0_delay_trigger_count(5);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->set_l0_delay_trigger_count(7);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->set_l0_delay_trigger_count(3);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());
 }

@@ -2763,53 +2749,53 @@ TEST_P(ColumnFamilyTest, WriteStallTwoColumnFamilies) {
  mutable_cf_options1.soft_pending_compaction_bytes_limit = 500;

  vstorage->TEST_set_estimated_compaction_needed_bytes(50);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());

  vstorage1->TEST_set_estimated_compaction_needed_bytes(201);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());

  vstorage1->TEST_set_estimated_compaction_needed_bytes(600);
-  RecalculateWriteStallConditions(cfd1, mutable_cf_options);
+  cfd1->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(70);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate, GetDbDelayedWriteRate());

  vstorage1->TEST_set_estimated_compaction_needed_bytes(800);
-  RecalculateWriteStallConditions(cfd1, mutable_cf_options);
+  cfd1->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(300);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25 / 1.25, GetDbDelayedWriteRate());

  vstorage1->TEST_set_estimated_compaction_needed_bytes(700);
-  RecalculateWriteStallConditions(cfd1, mutable_cf_options);
+  cfd1->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25, GetDbDelayedWriteRate());

  vstorage->TEST_set_estimated_compaction_needed_bytes(500);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25 / 1.25, GetDbDelayedWriteRate());

  vstorage1->TEST_set_estimated_compaction_needed_bytes(600);
-  RecalculateWriteStallConditions(cfd1, mutable_cf_options);
+  cfd1->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25, GetDbDelayedWriteRate());
@@ -2842,41 +2828,41 @@ TEST_P(ColumnFamilyTest, CompactionSpeedupTwoColumnFamilies) {
  mutable_cf_options1.level0_slowdown_writes_trigger = 16;

  vstorage->TEST_set_estimated_compaction_needed_bytes(40);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->TEST_set_estimated_compaction_needed_bytes(60);
-  RecalculateWriteStallConditions(cfd1, mutable_cf_options);
+  cfd1->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage1->TEST_set_estimated_compaction_needed_bytes(30);
-  RecalculateWriteStallConditions(cfd1, mutable_cf_options);
+  cfd1->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage1->TEST_set_estimated_compaction_needed_bytes(70);
-  RecalculateWriteStallConditions(cfd1, mutable_cf_options);
+  cfd1->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->TEST_set_estimated_compaction_needed_bytes(20);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage1->TEST_set_estimated_compaction_needed_bytes(3);
-  RecalculateWriteStallConditions(cfd1, mutable_cf_options);
+  cfd1->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->set_l0_delay_trigger_count(9);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage1->set_l0_delay_trigger_count(2);
-  RecalculateWriteStallConditions(cfd1, mutable_cf_options);
+  cfd1->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());

  vstorage->set_l0_delay_trigger_count(0);
-  RecalculateWriteStallConditions(cfd, mutable_cf_options);
+  cfd->RecalculateWriteStallConditions(mutable_cf_options);
  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());
 }

@@ -35,9 +35,9 @@ class CompactFilesTest : public testing::Test {
 class FlushedFileCollector : public EventListener {
 public:
  FlushedFileCollector() {}
-  ~FlushedFileCollector() override {}
+  ~FlushedFileCollector() {}

-  void OnFlushCompleted(DB* /*db*/, const FlushJobInfo& info) override {
+  virtual void OnFlushCompleted(DB* /*db*/, const FlushJobInfo& info) override {
    std::lock_guard<std::mutex> lock(mutex_);
    flushed_files_.push_back(info.file_path);
  }
@@ -256,9 +256,9 @@ TEST_F(CompactFilesTest, CapturingPendingFiles) {
 TEST_F(CompactFilesTest, CompactionFilterWithGetSv) {
  class FilterWithGet : public CompactionFilter {
   public:
-    bool Filter(int /*level*/, const Slice& /*key*/, const Slice& /*value*/,
-                std::string* /*new_value*/,
-                bool* /*value_changed*/) const override {
+    virtual bool Filter(int /*level*/, const Slice& /*key*/,
+                        const Slice& /*value*/, std::string* /*new_value*/,
+                        bool* /*value_changed*/) const override {
      if (db_ == nullptr) {
        return true;
      }
@@ -271,7 +271,7 @@ TEST_F(CompactFilesTest, CompactionFilterWithGetSv) {
      db_ = db;
    }

-    const char* Name() const override { return "FilterWithGet"; }
+    virtual const char* Name() const override { return "FilterWithGet"; }

   private:
    DB* db_;
@@ -309,14 +309,6 @@ TEST_F(CompactFilesTest, CompactionFilterWithGetSv) {
 }

 TEST_F(CompactFilesTest, SentinelCompressionType) {
-  if (!Zlib_Supported()) {
-    fprintf(stderr, "zlib compression not supported, skip this test\n");
-    return;
-  }
-  if (!Snappy_Supported()) {
-    fprintf(stderr, "snappy compression not supported, skip this test\n");
-    return;
-  }
  // Check that passing `CompressionType::kDisableCompressionOption` to
  // `CompactFiles` causes it to use the column family compression options.
  for (auto compaction_style :
@@ -357,51 +349,6 @@ TEST_F(CompactFilesTest, SentinelCompressionType) {
  }
 }

-TEST_F(CompactFilesTest, GetCompactionJobInfo) {
-  Options options;
-  options.create_if_missing = true;
-  // Disable RocksDB background compaction.
-  options.compaction_style = kCompactionStyleNone;
-  options.level0_slowdown_writes_trigger = 1000;
-  options.level0_stop_writes_trigger = 1000;
-  options.write_buffer_size = 65536;
-  options.max_write_buffer_number = 2;
-  options.compression = kNoCompression;
-  options.max_compaction_bytes = 5000;
-
-  // Add listener
-  FlushedFileCollector* collector = new FlushedFileCollector();
-  options.listeners.emplace_back(collector);
-
-  DB* db = nullptr;
-  DestroyDB(db_name_, options);
-  Status s = DB::Open(options, db_name_, &db);
-  assert(s.ok());
-  assert(db);
-
-  // create couple files
-  for (int i = 0; i < 500; ++i) {
-    db->Put(WriteOptions(), ToString(i), std::string(1000, 'a' + (i % 26)));
-  }
-  reinterpret_cast<DBImpl*>(db)->TEST_WaitForFlushMemTable();
-  auto l0_files_1 = collector->GetFlushedFiles();
-  CompactionOptions co;
-  co.compression = CompressionType::kLZ4Compression;
-  CompactionJobInfo compaction_job_info;
-  ASSERT_OK(
-      db->CompactFiles(co, l0_files_1, 0, -1, nullptr, &compaction_job_info));
-  ASSERT_EQ(compaction_job_info.base_input_level, 0);
-  ASSERT_EQ(compaction_job_info.cf_id, db->DefaultColumnFamily()->GetID());
-  ASSERT_EQ(compaction_job_info.cf_name, db->DefaultColumnFamily()->GetName());
-  ASSERT_EQ(compaction_job_info.compaction_reason,
-            CompactionReason::kManualCompaction);
-  ASSERT_EQ(compaction_job_info.compression, CompressionType::kLZ4Compression);
-  ASSERT_EQ(compaction_job_info.output_level, 0);
-  ASSERT_OK(compaction_job_info.status);
-  // no assertion failure
-  delete db;
-}
-
 }  // namespace rocksdb

 int main(int argc, char** argv) {
@@ -25,12 +25,22 @@ CompactedDBImpl::~CompactedDBImpl() {
 }

 size_t CompactedDBImpl::FindFile(const Slice& key) {
+  size_t left = 0;
  size_t right = files_.num_files - 1;
-  auto cmp = [&](const FdWithKeyRange& f, const Slice& k) -> bool {
-    return user_comparator_->Compare(ExtractUserKey(f.largest_key), k) < 0;
-  };
-  return static_cast<size_t>(std::lower_bound(files_.files,
-                            files_.files + right, key, cmp) - files_.files);
+  while (left < right) {
+    size_t mid = (left + right) >> 1;
+    const FdWithKeyRange& f = files_.files[mid];
+    if (user_comparator_->Compare(ExtractUserKey(f.largest_key), key) < 0) {
+      // Key at "mid.largest" is < "target".  Therefore all
+      // files at or before "mid" are uninteresting.
+      left = mid + 1;
+    } else {
+      // Key at "mid.largest" is >= "target".  Therefore all files
+      // after "mid" are uninteresting.
+      right = mid;
+    }
+  }
+  return right;
 }

 Status CompactedDBImpl::Get(const ReadOptions& options, ColumnFamilyHandle*,
@@ -147,7 +157,6 @@ Status CompactedDBImpl::Open(const Options& options,
  std::unique_ptr<CompactedDBImpl> db(new CompactedDBImpl(db_options, dbname));
  Status s = db->Init(options);
  if (s.ok()) {
-    db->StartTimedTasks();
    ROCKS_LOG_INFO(db->immutable_db_options_.info_log,
                   "Opened the db as fully compacted mode");
    LogFlush(db->immutable_db_options_.info_log);
@@ -67,11 +67,10 @@ class CompactedDBImpl : public DBImpl {
  virtual Status EnableFileDeletions(bool /*force*/) override {
    return Status::NotSupported("Not supported in compacted db mode.");
  }
-  virtual Status GetLiveFiles(std::vector<std::string>& ret,
-                              uint64_t* manifest_file_size,
-                              bool /*flush_memtable*/) override {
-    return DBImpl::GetLiveFiles(ret, manifest_file_size,
-                                false /* flush_memtable */);
+  virtual Status GetLiveFiles(std::vector<std::string>&,
+                              uint64_t* /*manifest_file_size*/,
+                              bool /*flush_memtable*/ = true) override {
+    return Status::NotSupported("Not supported in compacted db mode.");
  }
  using DBImpl::Flush;
  virtual Status Flush(const FlushOptions& /*options*/,
@@ -23,43 +23,6 @@

 namespace rocksdb {

-const uint64_t kRangeTombstoneSentinel =
-    PackSequenceAndType(kMaxSequenceNumber, kTypeRangeDeletion);
-
-int sstableKeyCompare(const Comparator* user_cmp, const InternalKey& a,
-                      const InternalKey& b) {
-  auto c = user_cmp->Compare(a.user_key(), b.user_key());
-  if (c != 0) {
-    return c;
-  }
-  auto a_footer = ExtractInternalKeyFooter(a.Encode());
-  auto b_footer = ExtractInternalKeyFooter(b.Encode());
-  if (a_footer == kRangeTombstoneSentinel) {
-    if (b_footer != kRangeTombstoneSentinel) {
-      return -1;
-    }
-  } else if (b_footer == kRangeTombstoneSentinel) {
-    return 1;
-  }
-  return 0;
-}
-
-int sstableKeyCompare(const Comparator* user_cmp, const InternalKey* a,
-                      const InternalKey& b) {
-  if (a == nullptr) {
-    return -1;
-  }
-  return sstableKeyCompare(user_cmp, *a, b);
-}
-
-int sstableKeyCompare(const Comparator* user_cmp, const InternalKey& a,
-                      const InternalKey* b) {
-  if (b == nullptr) {
-    return -1;
-  }
-  return sstableKeyCompare(user_cmp, a, *b);
-}
-
 uint64_t TotalFileSize(const std::vector<FileMetaData*>& files) {
  uint64_t sum = 0;
  for (size_t i = 0; i < files.size() && files[i]; i++) {
@@ -118,49 +81,6 @@ void Compaction::GetBoundaryKeys(
  }
 }

-std::vector<CompactionInputFiles> Compaction::PopulateWithAtomicBoundaries(
-    VersionStorageInfo* vstorage, std::vector<CompactionInputFiles> inputs) {
-  const Comparator* ucmp = vstorage->InternalComparator()->user_comparator();
-  for (size_t i = 0; i < inputs.size(); i++) {
-    if (inputs[i].level == 0 || inputs[i].files.empty()) {
-      continue;
-    }
-    inputs[i].atomic_compaction_unit_boundaries.reserve(inputs[i].files.size());
-    AtomicCompactionUnitBoundary cur_boundary;
-    size_t first_atomic_idx = 0;
-    auto add_unit_boundary = [&](size_t to) {
-      if (first_atomic_idx == to) return;
-      for (size_t k = first_atomic_idx; k < to; k++) {
-        inputs[i].atomic_compaction_unit_boundaries.push_back(cur_boundary);
-      }
-      first_atomic_idx = to;
-    };
-    for (size_t j = 0; j < inputs[i].files.size(); j++) {
-      const auto* f = inputs[i].files[j];
-      if (j == 0) {
-        // First file in a level.
-        cur_boundary.smallest = &f->smallest;
-        cur_boundary.largest = &f->largest;
-      } else if (sstableKeyCompare(ucmp, *cur_boundary.largest, f->smallest) ==
-                 0) {
-        // SSTs overlap but the end key of the previous file was not
-        // artificially extended by a range tombstone. Extend the current
-        // boundary.
-        cur_boundary.largest = &f->largest;
-      } else {
-        // Atomic compaction unit has ended.
-        add_unit_boundary(j);
-        cur_boundary.smallest = &f->smallest;
-        cur_boundary.largest = &f->largest;
-      }
-    }
-    add_unit_boundary(inputs[i].files.size());
-    assert(inputs[i].files.size() ==
-           inputs[i].atomic_compaction_unit_boundaries.size());
-  }
-  return inputs;
-}
-
 // helper function to determine if compaction is creating files at the
 // bottommost level
 bool Compaction::IsBottommostLevel(
@@ -235,7 +155,7 @@ Compaction::Compaction(VersionStorageInfo* vstorage,
      output_compression_(_compression),
      output_compression_opts_(_compression_opts),
      deletion_compaction_(_deletion_compaction),
-      inputs_(PopulateWithAtomicBoundaries(vstorage, std::move(_inputs))),
+      inputs_(std::move(_inputs)),
      grandparents_(std::move(_grandparents)),
      score_(_score),
      bottommost_level_(IsBottommostLevel(output_level_, vstorage, inputs_)),
@@ -250,12 +170,6 @@ Compaction::Compaction(VersionStorageInfo* vstorage,
  if (max_subcompactions_ == 0) {
    max_subcompactions_ = immutable_cf_options_.max_subcompactions;
  }
-  if (!bottommost_level_) {
-    // Currently we only enable dictionary compression during compaction to the
-    // bottommost level.
-    output_compression_opts_.max_dict_bytes = 0;
-    output_compression_opts_.zstd_max_train_bytes = 0;
-  }

 #ifndef NDEBUG
  for (size_t i = 1; i < inputs_.size(); ++i) {
@@ -379,8 +293,8 @@ bool Compaction::KeyNotExistsBeyondOutputLevel(
        if (user_cmp->Compare(user_key, f->largest.user_key()) <= 0) {
          // We've advanced far enough
          if (user_cmp->Compare(user_key, f->smallest.user_key()) >= 0) {
-            // Key falls in this file's range, so it may
-            // exist beyond output level
+            // Key falls in this file's range, so definitely
+            // exists beyond output level
            return false;
          }
          break;
@@ -417,14 +331,12 @@ const char* Compaction::InputLevelSummary(
    if (!is_first) {
      len +=
          snprintf(scratch->buffer + len, sizeof(scratch->buffer) - len, " + ");
-      len = std::min(len, static_cast<int>(sizeof(scratch->buffer)));
    } else {
      is_first = false;
    }
    len += snprintf(scratch->buffer + len, sizeof(scratch->buffer) - len,
                    "%" ROCKSDB_PRIszt "@%d", input_level.size(),
                    input_level.level);
-    len = std::min(len, static_cast<int>(sizeof(scratch->buffer)));
  }
  snprintf(scratch->buffer + len, sizeof(scratch->buffer) - len,
           " files to L%d", output_level());
@@ -15,43 +15,11 @@

 namespace rocksdb {

-// Utility for comparing sstable boundary keys. Returns -1 if either a or b is
-// null which provides the property that a==null indicates a key that is less
-// than any key and b==null indicates a key that is greater than any key. Note
-// that the comparison is performed primarily on the user-key portion of the
-// key. If the user-keys compare equal, an additional test is made to sort
-// range tombstone sentinel keys before other keys with the same user-key. The
-// result is that 2 user-keys will compare equal if they differ purely on
-// their sequence number and value, but the range tombstone sentinel for that
-// user-key will compare not equal. This is necessary because the range
-// tombstone sentinel key is set as the largest key for an sstable even though
-// that key never appears in the database. We don't want adjacent sstables to
-// be considered overlapping if they are separated by the range tombstone
-// sentinel.
-int sstableKeyCompare(const Comparator* user_cmp, const InternalKey& a,
-                      const InternalKey& b);
-int sstableKeyCompare(const Comparator* user_cmp, const InternalKey* a,
-                      const InternalKey& b);
-int sstableKeyCompare(const Comparator* user_cmp, const InternalKey& a,
-                      const InternalKey* b);
-
-// An AtomicCompactionUnitBoundary represents a range of keys [smallest,
-// largest] that exactly spans one ore more neighbouring SSTs on the same
-// level. Every pair of  SSTs in this range "overlap" (i.e., the largest
-// user key of one file is the smallest user key of the next file). These
-// boundaries are propagated down to RangeDelAggregator during compaction
-// to provide safe truncation boundaries for range tombstones.
-struct AtomicCompactionUnitBoundary {
-  const InternalKey* smallest = nullptr;
-  const InternalKey* largest = nullptr;
-};
-
 // The structure that manages compaction input files associated
 // with the same physical level.
 struct CompactionInputFiles {
  int level;
  std::vector<FileMetaData*> files;
-  std::vector<AtomicCompactionUnitBoundary> atomic_compaction_unit_boundaries;
  inline bool empty() const { return files.empty(); }
  inline size_t size() const { return files.size(); }
  inline void clear() { files.clear(); }
@@ -128,12 +96,6 @@ class Compaction {
    return inputs_[compaction_input_level][i];
  }

-  const std::vector<AtomicCompactionUnitBoundary>* boundaries(
-      size_t compaction_input_level) const {
-    assert(compaction_input_level < inputs_.size());
-    return &inputs_[compaction_input_level].atomic_compaction_unit_boundaries;
-  }
-
  // Returns the list of file meta data of the specified compaction
  // input level.
  // REQUIREMENT: "compaction_input_level" must be >= 0 and
@@ -300,13 +262,6 @@ class Compaction {
                              const std::vector<CompactionInputFiles>& inputs,
                              Slice* smallest_key, Slice* largest_key);

-  // Get the atomic file boundaries for all files in the compaction. Necessary
-  // in order to avoid the scenario described in
-  // https://github.com/facebook/rocksdb/pull/4432#discussion_r221072219 and plumb
-  // down appropriate key boundaries to RangeDelAggregator during compaction.
-  static std::vector<CompactionInputFiles> PopulateWithAtomicBoundaries(
-      VersionStorageInfo* vstorage, std::vector<CompactionInputFiles> inputs);
-
  // helper function to determine if compaction with inputs and storage is
  // bottommost
  static bool IsBottommostLevel(
@@ -9,23 +9,6 @@
 #include "port/likely.h"
 #include "rocksdb/listener.h"
 #include "table/internal_iterator.h"
-#include "util/sync_point.h"
-
-#define DEFINITELY_IN_SNAPSHOT(seq, snapshot)                       \
-  ((seq) <= (snapshot) &&                                           \
-   (snapshot_checker_ == nullptr ||                                 \
-    LIKELY(snapshot_checker_->CheckInSnapshot((seq), (snapshot)) == \
-           SnapshotCheckerResult::kInSnapshot)))
-
-#define DEFINITELY_NOT_IN_SNAPSHOT(seq, snapshot)                     \
-  ((seq) > (snapshot) ||                                              \
-   (snapshot_checker_ != nullptr &&                                   \
-    UNLIKELY(snapshot_checker_->CheckInSnapshot((seq), (snapshot)) == \
-             SnapshotCheckerResult::kNotInSnapshot)))
-
-#define IN_EARLIEST_SNAPSHOT(seq) \
-  ((seq) <= earliest_snapshot_ && \
-   (snapshot_checker_ == nullptr || LIKELY(IsInEarliestSnapshot(seq))))

 namespace rocksdb {

@@ -35,19 +18,17 @@ CompactionIterator::CompactionIterator(
    SequenceNumber earliest_write_conflict_snapshot,
    const SnapshotChecker* snapshot_checker, Env* env,
    bool report_detailed_time, bool expect_valid_internal_key,
-    CompactionRangeDelAggregator* range_del_agg, const Compaction* compaction,
-    const CompactionFilter* compaction_filter,
+    RangeDelAggregator* range_del_agg,
+    const Compaction* compaction, const CompactionFilter* compaction_filter,
    const std::atomic<bool>* shutting_down,
-    const SequenceNumber preserve_deletes_seqnum,
-    SnapshotListFetchCallback* snap_list_callback)
+    const SequenceNumber preserve_deletes_seqnum)
    : CompactionIterator(
          input, cmp, merge_helper, last_sequence, snapshots,
          earliest_write_conflict_snapshot, snapshot_checker, env,
          report_detailed_time, expect_valid_internal_key, range_del_agg,
          std::unique_ptr<CompactionProxy>(
              compaction ? new CompactionProxy(compaction) : nullptr),
-          compaction_filter, shutting_down, preserve_deletes_seqnum,
-          snap_list_callback) {}
+          compaction_filter, shutting_down, preserve_deletes_seqnum) {}

 CompactionIterator::CompactionIterator(
    InternalIterator* input, const Comparator* cmp, MergeHelper* merge_helper,
@@ -55,12 +36,11 @@ CompactionIterator::CompactionIterator(
    SequenceNumber earliest_write_conflict_snapshot,
    const SnapshotChecker* snapshot_checker, Env* env,
    bool report_detailed_time, bool expect_valid_internal_key,
-    CompactionRangeDelAggregator* range_del_agg,
+    RangeDelAggregator* range_del_agg,
    std::unique_ptr<CompactionProxy> compaction,
    const CompactionFilter* compaction_filter,
    const std::atomic<bool>* shutting_down,
-    const SequenceNumber preserve_deletes_seqnum,
-    SnapshotListFetchCallback* snap_list_callback)
+    const SequenceNumber preserve_deletes_seqnum)
    : input_(input),
      cmp_(cmp),
      merge_helper_(merge_helper),
@@ -75,21 +55,42 @@ CompactionIterator::CompactionIterator(
      compaction_filter_(compaction_filter),
      shutting_down_(shutting_down),
      preserve_deletes_seqnum_(preserve_deletes_seqnum),
+      ignore_snapshots_(false),
      current_user_key_sequence_(0),
      current_user_key_snapshot_(0),
      merge_out_iter_(merge_helper_),
-      current_key_committed_(false),
-      snap_list_callback_(snap_list_callback) {
+      current_key_committed_(false) {
  assert(compaction_filter_ == nullptr || compaction_ != nullptr);
-  assert(snapshots_ != nullptr);
  bottommost_level_ =
      compaction_ == nullptr ? false : compaction_->bottommost_level();
  if (compaction_ != nullptr) {
    level_ptrs_ = std::vector<size_t>(compaction_->number_levels(), 0);
  }
-  ProcessSnapshotList();
+
+  if (snapshots_->size() == 0) {
+    // optimize for fast path if there are no snapshots
+    visible_at_tip_ = true;
+    earliest_snapshot_ = kMaxSequenceNumber;
+    latest_snapshot_ = 0;
+  } else {
+    visible_at_tip_ = false;
+    earliest_snapshot_ = snapshots_->at(0);
+    latest_snapshot_ = snapshots_->back();
+  }
+#ifndef NDEBUG
+  // findEarliestVisibleSnapshot assumes this ordering.
+  for (size_t i = 1; i < snapshots_->size(); ++i) {
+    assert(snapshots_->at(i - 1) <= snapshots_->at(i));
+  }
+#endif
+  if (compaction_filter_ != nullptr) {
+    if (compaction_filter_->IgnoreSnapshots()) {
+      ignore_snapshots_ = true;
+    }
+  } else {
+    ignore_snapshots_ = false;
+  }
  input_->SetPinnedItersMgr(&pinned_iters_mgr_);
-  TEST_SYNC_POINT_CALLBACK("CompactionIterator:AfterInit", compaction_.get());
 }

 CompactionIterator::~CompactionIterator() {
@@ -159,7 +160,12 @@ void CompactionIterator::Next() {
 void CompactionIterator::InvokeFilterIfNeeded(bool* need_skip,
                                              Slice* skip_until) {
  if (compaction_filter_ != nullptr &&
-      (ikey_.type == kTypeValue || ikey_.type == kTypeBlobIndex)) {
+      (ikey_.type == kTypeValue || ikey_.type == kTypeBlobIndex) &&
+      (visible_at_tip_ || ignore_snapshots_ ||
+       ikey_.sequence > latest_snapshot_ ||
+       (snapshot_checker_ != nullptr &&
+        UNLIKELY(!snapshot_checker_->IsInSnapshot(ikey_.sequence,
+                                                  latest_snapshot_))))) {
    // If the user has specified a compaction filter and the sequence
    // number is greater than any external snapshot, then invoke the
    // filter. If the return value of the compaction filter is true,
@@ -209,28 +215,6 @@ void CompactionIterator::InvokeFilterIfNeeded(bool* need_skip,
  }
 }

-void CompactionIterator::ProcessSnapshotList() {
-#ifndef NDEBUG
-  // findEarliestVisibleSnapshot assumes this ordering.
-  for (size_t i = 1; i < snapshots_->size(); ++i) {
-    assert(snapshots_->at(i - 1) < snapshots_->at(i));
-  }
-#endif
-  if (snapshots_->size() == 0) {
-    // optimize for fast path if there are no snapshots
-    visible_at_tip_ = true;
-    earliest_snapshot_iter_ = snapshots_->end();
-    earliest_snapshot_ = kMaxSequenceNumber;
-    latest_snapshot_ = 0;
-  } else {
-    visible_at_tip_ = false;
-    earliest_snapshot_iter_ = snapshots_->begin();
-    earliest_snapshot_ = snapshots_->at(0);
-    latest_snapshot_ = snapshots_->back();
-  }
-  released_snapshots_.clear();
-}
-
 void CompactionIterator::NextFromInput() {
  at_next_ = false;
  valid_ = false;
@@ -257,7 +241,6 @@ void CompactionIterator::NextFromInput() {
      valid_ = true;
      break;
    }
-    TEST_SYNC_POINT_CALLBACK("CompactionIterator:ProcessKV", &ikey_);

    // Update input statistics
    if (ikey_.type == kTypeDeletion || ikey_.type == kTypeSingleDeletion) {
@@ -278,13 +261,6 @@ void CompactionIterator::NextFromInput() {
    // compaction filter). ikey_.user_key is pointing to the copy.
    if (!has_current_user_key_ ||
        !cmp_->Equal(ikey_.user_key, current_user_key_)) {
-      num_keys_++;
-      // Use num_keys_ to reduce the overhead of reading current time
-      if (snap_list_callback_ && snapshots_->size() &&
-          snap_list_callback_->TimeToRefresh(num_keys_)) {
-        snap_list_callback_->Refresh(snapshots_, latest_snapshot_);
-        ProcessSnapshotList();
-      }
      // First occurrence of this user key
      // Copy key for output
      key_ = current_key_.SetInternalKey(key_, &ikey_);
@@ -293,7 +269,9 @@ void CompactionIterator::NextFromInput() {
      has_outputted_key_ = false;
      current_user_key_sequence_ = kMaxSequenceNumber;
      current_user_key_snapshot_ = 0;
-      current_key_committed_ = KeyCommitted(ikey_.sequence);
+      current_key_committed_ =
+          (snapshot_checker_ == nullptr ||
+           snapshot_checker_->IsInSnapshot(ikey_.sequence, kMaxSequenceNumber));

      // Apply the compaction filter to the first committed version of the user
      // key.
@@ -315,7 +293,8 @@ void CompactionIterator::NextFromInput() {
      // to query snapshot_checker_ in that case.
      if (UNLIKELY(!current_key_committed_)) {
        assert(snapshot_checker_ != nullptr);
-        current_key_committed_ = KeyCommitted(ikey_.sequence);
+        current_key_committed_ =
+            snapshot_checker_->IsInSnapshot(ikey_.sequence, kMaxSequenceNumber);
        // Apply the compaction filter to the first committed version of the
        // user key.
        if (current_key_committed_) {
@@ -399,8 +378,10 @@ void CompactionIterator::NextFromInput() {
          cmp_->Equal(ikey_.user_key, next_ikey.user_key)) {
        // Check whether the next key belongs to the same snapshot as the
        // SingleDelete.
-        if (prev_snapshot == 0 ||
-            DEFINITELY_NOT_IN_SNAPSHOT(next_ikey.sequence, prev_snapshot)) {
+        if (prev_snapshot == 0 || next_ikey.sequence > prev_snapshot ||
+            (snapshot_checker_ != nullptr &&
+             UNLIKELY(!snapshot_checker_->IsInSnapshot(next_ikey.sequence,
+                                                       prev_snapshot)))) {
          if (next_ikey.type == kTypeSingleDeletion) {
            // We encountered two SingleDeletes in a row.  This could be due to
            // unexpected user input.
@@ -412,8 +393,11 @@ void CompactionIterator::NextFromInput() {
            ++iter_stats_.num_record_drop_obsolete;
            ++iter_stats_.num_single_del_mismatch;
          } else if (has_outputted_key_ ||
-                     DEFINITELY_IN_SNAPSHOT(
-                         ikey_.sequence, earliest_write_conflict_snapshot_)) {
+                     (ikey_.sequence <= earliest_write_conflict_snapshot_ &&
+                      (snapshot_checker_ == nullptr ||
+                       LIKELY(snapshot_checker_->IsInSnapshot(
+                           ikey_.sequence,
+                           earliest_write_conflict_snapshot_))))) {
            // Found a matching value, we can drop the single delete and the
            // value.  It is safe to drop both records since we've already
            // outputted a key in this snapshot, or there is no earlier
@@ -423,8 +407,7 @@ void CompactionIterator::NextFromInput() {
            // is an unexpected Merge or Delete.  We will compact it out
            // either way. We will maintain counts of how many mismatches
            // happened
-            if (next_ikey.type != kTypeValue &&
-                next_ikey.type != kTypeBlobIndex) {
+            if (next_ikey.type != kTypeValue) {
              ++iter_stats_.num_single_del_mismatch;
            }

@@ -461,7 +444,10 @@ void CompactionIterator::NextFromInput() {
        // iteration. If the next key is corrupt, we return before the
        // comparison, so the value of has_current_user_key does not matter.
        has_current_user_key_ = false;
-        if (compaction_ != nullptr && IN_EARLIEST_SNAPSHOT(ikey_.sequence) &&
+        if (compaction_ != nullptr && ikey_.sequence <= earliest_snapshot_ &&
+            (snapshot_checker_ == nullptr ||
+             LIKELY(snapshot_checker_->IsInSnapshot(ikey_.sequence,
+                                                    earliest_snapshot_))) &&
            compaction_->KeyNotExistsBeyondOutputLevel(ikey_.user_key,
                                                       &level_ptrs_)) {
          // Key doesn't exist outside of this range.
@@ -480,37 +466,24 @@ void CompactionIterator::NextFromInput() {
      if (valid_) {
        at_next_ = true;
      }
-    } else if (last_snapshot == current_user_key_snapshot_ ||
-               (last_snapshot > 0 &&
-                last_snapshot < current_user_key_snapshot_)) {
+    } else if (last_snapshot == current_user_key_snapshot_) {
      // If the earliest snapshot is which this key is visible in
      // is the same as the visibility of a previous instance of the
      // same key, then this kv is not visible in any snapshot.
      // Hidden by an newer entry for same user key
+      // TODO(noetzli): why not > ?
      //
      // Note: Dropping this key will not affect TransactionDB write-conflict
      // checking since there has already been a record returned for this key
      // in this snapshot.
      assert(last_sequence >= current_user_key_sequence_);
-
-      // Note2: if last_snapshot < current_user_key_snapshot, it can only
-      // mean last_snapshot is released between we process last value and
-      // this value, and findEarliestVisibleSnapshot returns the next snapshot
-      // as current_user_key_snapshot. In this case last value and current
-      // value are both in current_user_key_snapshot currently.
-      // Although last_snapshot is released we might still get a definitive
-      // response when key sequence number changes, e.g., when seq is determined
-      // too old and visible in all snapshots.
-      assert(last_snapshot == current_user_key_snapshot_ ||
-             (snapshot_checker_ != nullptr &&
-              snapshot_checker_->CheckInSnapshot(current_user_key_sequence_,
-                                                 last_snapshot) !=
-                  SnapshotCheckerResult::kNotInSnapshot));
-
      ++iter_stats_.num_record_drop_hidden;  // (A)
      input_->Next();
    } else if (compaction_ != nullptr && ikey_.type == kTypeDeletion &&
-               IN_EARLIEST_SNAPSHOT(ikey_.sequence) &&
+               ikey_.sequence <= earliest_snapshot_ &&
+               (snapshot_checker_ == nullptr ||
+                LIKELY(snapshot_checker_->IsInSnapshot(ikey_.sequence,
+                                                       earliest_snapshot_))) &&
               ikeyNotNeededForIncrementalSnapshot() &&
               compaction_->KeyNotExistsBeyondOutputLevel(ikey_.user_key,
                                                          &level_ptrs_)) {
@@ -538,28 +511,6 @@ void CompactionIterator::NextFromInput() {
        ++iter_stats_.num_optimized_del_drop_obsolete;
      }
      input_->Next();
-    } else if ((ikey_.type == kTypeDeletion) && bottommost_level_ &&
-               ikeyNotNeededForIncrementalSnapshot()) {
-      // Handle the case where we have a delete key at the bottom most level
-      // We can skip outputting the key iff there are no subsequent puts for this
-      // key
-      ParsedInternalKey next_ikey;
-      input_->Next();
-      // Skip over all versions of this key that happen to occur in the same snapshot
-      // range as the delete
-      while (input_->Valid() && ParseInternalKey(input_->key(), &next_ikey) &&
-             cmp_->Equal(ikey_.user_key, next_ikey.user_key) &&
-             (prev_snapshot == 0 ||
-              DEFINITELY_NOT_IN_SNAPSHOT(next_ikey.sequence, prev_snapshot))) {
-        input_->Next();
-      }
-      // If you find you still need to output a row with this key, we need to output the
-      // delete too
-      if (input_->Valid() && ParseInternalKey(input_->key(), &next_ikey) &&
-          cmp_->Equal(ikey_.user_key, next_ikey.user_key)) {
-        valid_ = true;
-        at_next_ = true;
-      }
    } else if (ikey_.type == kTypeMerge) {
      if (!merge_helper_->HasOperator()) {
        status_ = Status::InvalidArgument(
@@ -641,9 +592,14 @@ void CompactionIterator::PrepareOutput() {
  //
  // Can we do the same for levels above bottom level as long as
  // KeyNotExistsBeyondOutputLevel() return true?
-  if ((compaction_ != nullptr && !compaction_->allow_ingest_behind()) &&
-      ikeyNotNeededForIncrementalSnapshot() && bottommost_level_ && valid_ &&
-      IN_EARLIEST_SNAPSHOT(ikey_.sequence) && ikey_.type != kTypeMerge) {
+  if ((compaction_ != nullptr &&
+      !compaction_->allow_ingest_behind()) &&
+      ikeyNotNeededForIncrementalSnapshot() &&
+      bottommost_level_ && valid_ && ikey_.sequence <= earliest_snapshot_ &&
+      (snapshot_checker_ == nullptr || LIKELY(snapshot_checker_->IsInSnapshot(
+        ikey_.sequence, earliest_snapshot_))) &&
+      ikey_.type != kTypeMerge &&
+      !cmp_->Equal(compaction_->GetLargestUserKey(), ikey_.user_key)) {
    assert(ikey_.type != kTypeDeletion && ikey_.type != kTypeSingleDeletion);
    ikey_.sequence = 0;
    current_key_.UpdateInternalKey(0, ikey_.type);
@@ -661,23 +617,12 @@ inline SequenceNumber CompactionIterator::findEarliestVisibleSnapshot(
    *prev_snapshot = *std::prev(snapshots_iter);
    assert(*prev_snapshot < in);
  }
-  if (snapshot_checker_ == nullptr) {
-    return snapshots_iter != snapshots_->end()
-      ? *snapshots_iter : kMaxSequenceNumber;
-  }
-  bool has_released_snapshot = !released_snapshots_.empty();
  for (; snapshots_iter != snapshots_->end(); ++snapshots_iter) {
    auto cur = *snapshots_iter;
    assert(in <= cur);
-    // Skip if cur is in released_snapshots.
-    if (has_released_snapshot && released_snapshots_.count(cur) > 0) {
-      continue;
-    }
-    auto res = snapshot_checker_->CheckInSnapshot(in, cur);
-    if (res == SnapshotCheckerResult::kInSnapshot) {
+    if (snapshot_checker_ == nullptr ||
+        snapshot_checker_->IsInSnapshot(in, cur)) {
      return cur;
-    } else if (res == SnapshotCheckerResult::kSnapshotReleased) {
-      released_snapshots_.insert(cur);
    }
    *prev_snapshot = cur;
  }
@@ -691,30 +636,4 @@ inline bool CompactionIterator::ikeyNotNeededForIncrementalSnapshot() {
         (ikey_.sequence < preserve_deletes_seqnum_);
 }

-bool CompactionIterator::IsInEarliestSnapshot(SequenceNumber sequence) {
-  assert(snapshot_checker_ != nullptr);
-  assert(earliest_snapshot_ == kMaxSequenceNumber ||
-         (earliest_snapshot_iter_ != snapshots_->end() &&
-          *earliest_snapshot_iter_ == earliest_snapshot_));
-  auto in_snapshot =
-      snapshot_checker_->CheckInSnapshot(sequence, earliest_snapshot_);
-  while (UNLIKELY(in_snapshot == SnapshotCheckerResult::kSnapshotReleased)) {
-    // Avoid the the current earliest_snapshot_ being return as
-    // earliest visible snapshot for the next value. So if a value's sequence
-    // is zero-ed out by PrepareOutput(), the next value will be compact out.
-    released_snapshots_.insert(earliest_snapshot_);
-    earliest_snapshot_iter_++;
-
-    if (earliest_snapshot_iter_ == snapshots_->end()) {
-      earliest_snapshot_ = kMaxSequenceNumber;
-    } else {
-      earliest_snapshot_ = *earliest_snapshot_iter_;
-    }
-    in_snapshot =
-        snapshot_checker_->CheckInSnapshot(sequence, earliest_snapshot_);
-  }
-  assert(in_snapshot != SnapshotCheckerResult::kSnapshotReleased);
-  return in_snapshot == SnapshotCheckerResult::kInSnapshot;
-}
-
 }  // namespace rocksdb
@@ -7,7 +7,6 @@
 #include <algorithm>
 #include <deque>
 #include <string>
-#include <unordered_set>
 #include <vector>

 #include "db/compaction.h"
@@ -21,54 +20,6 @@

 namespace rocksdb {

-// This callback can be used to refresh the snapshot list from the db. It
-// includes logics to exponentially decrease the refresh rate to limit the
-// overhead of refresh.
-class SnapshotListFetchCallback {
- public:
-  SnapshotListFetchCallback(Env* env, uint64_t snap_refresh_nanos,
-                            size_t every_nth_key = 1024)
-      : timer_(env, /*auto restart*/ true),
-        snap_refresh_nanos_(snap_refresh_nanos),
-        every_nth_key_minus_one_(every_nth_key - 1) {
-    assert(every_nth_key > 0);
-    assert((ceil(log2(every_nth_key)) == floor(log2(every_nth_key))));
-  }
-  // Refresh the snapshot list. snapshots will bre replacted with the new list.
-  // max is the upper bound. Note: this function will acquire the db_mutex_.
-  virtual void Refresh(std::vector<SequenceNumber>* snapshots,
-                       SequenceNumber max) = 0;
-  inline bool TimeToRefresh(const size_t key_index) {
-    assert(snap_refresh_nanos_ != 0);
-    // skip the key if key_index % every_nth_key (which is of power 2) is not 0.
-    if ((key_index & every_nth_key_minus_one_) != 0) {
-      return false;
-    }
-    const uint64_t elapsed = timer_.ElapsedNanos();
-    auto ret = elapsed > snap_refresh_nanos_;
-    // pre-compute the next time threshold
-    if (ret) {
-      // inc next refresh period exponentially (by x4)
-      auto next_refresh_threshold = snap_refresh_nanos_ << 2;
-      // make sure the shift has not overflown the highest 1 bit
-      snap_refresh_nanos_ =
-          std::max(snap_refresh_nanos_, next_refresh_threshold);
-    }
-    return ret;
-  }
-  static constexpr SnapshotListFetchCallback* kDisabled = nullptr;
-
-  virtual ~SnapshotListFetchCallback() {}
-
- private:
-  // Time since the callback was created
-  StopWatchNano timer_;
-  // The delay before calling ::Refresh. To be increased exponentially.
-  uint64_t snap_refresh_nanos_;
-  // Skip evey nth key. Number n if of power 2. The math will require n-1.
-  const uint64_t every_nth_key_minus_one_;
-};
-
 class CompactionIterator {
 public:
  // A wrapper around Compaction. Has a much smaller interface, only what
@@ -113,12 +64,11 @@ class CompactionIterator {
                     SequenceNumber earliest_write_conflict_snapshot,
                     const SnapshotChecker* snapshot_checker, Env* env,
                     bool report_detailed_time, bool expect_valid_internal_key,
-                     CompactionRangeDelAggregator* range_del_agg,
+                     RangeDelAggregator* range_del_agg,
                     const Compaction* compaction = nullptr,
                     const CompactionFilter* compaction_filter = nullptr,
                     const std::atomic<bool>* shutting_down = nullptr,
-                     const SequenceNumber preserve_deletes_seqnum = 0,
-                     SnapshotListFetchCallback* snap_list_callback = nullptr);
+                     const SequenceNumber preserve_deletes_seqnum = 0);

  // Constructor with custom CompactionProxy, used for tests.
  CompactionIterator(InternalIterator* input, const Comparator* cmp,
@@ -127,12 +77,11 @@ class CompactionIterator {
                     SequenceNumber earliest_write_conflict_snapshot,
                     const SnapshotChecker* snapshot_checker, Env* env,
                     bool report_detailed_time, bool expect_valid_internal_key,
-                     CompactionRangeDelAggregator* range_del_agg,
+                     RangeDelAggregator* range_del_agg,
                     std::unique_ptr<CompactionProxy> compaction,
                     const CompactionFilter* compaction_filter = nullptr,
                     const std::atomic<bool>* shutting_down = nullptr,
-                     const SequenceNumber preserve_deletes_seqnum = 0,
-                     SnapshotListFetchCallback* snap_list_callback = nullptr);
+                     const SequenceNumber preserve_deletes_seqnum = 0);

  ~CompactionIterator();

@@ -160,8 +109,6 @@ class CompactionIterator {
 private:
  // Processes the input stream to find the next output
  void NextFromInput();
-  // Process snapshots_ and assign related variables
-  void ProcessSnapshotList();

  // Do last preparations before presenting the output to the callee. At this
  // point this only zeroes out the sequence number if possible for better
@@ -185,31 +132,16 @@ class CompactionIterator {
  // or seqnum be zero-ed out even if all other conditions for it are met.
  inline bool ikeyNotNeededForIncrementalSnapshot();

-  inline bool KeyCommitted(SequenceNumber sequence) {
-    return snapshot_checker_ == nullptr ||
-           snapshot_checker_->CheckInSnapshot(sequence, kMaxSequenceNumber) ==
-               SnapshotCheckerResult::kInSnapshot;
-  }
-
-  bool IsInEarliestSnapshot(SequenceNumber sequence);
-
  InternalIterator* input_;
  const Comparator* cmp_;
  MergeHelper* merge_helper_;
-  std::vector<SequenceNumber>* snapshots_;
-  // List of snapshots released during compaction.
-  // findEarliestVisibleSnapshot() find them out from return of
-  // snapshot_checker, and make sure they will not be returned as
-  // earliest visible snapshot of an older value.
-  // See WritePreparedTransactionTest::ReleaseSnapshotDuringCompaction3.
-  std::unordered_set<SequenceNumber> released_snapshots_;
-  std::vector<SequenceNumber>::const_iterator earliest_snapshot_iter_;
+  const std::vector<SequenceNumber>* snapshots_;
  const SequenceNumber earliest_write_conflict_snapshot_;
  const SnapshotChecker* const snapshot_checker_;
  Env* env_;
  bool report_detailed_time_;
  bool expect_valid_internal_key_;
-  CompactionRangeDelAggregator* range_del_agg_;
+  RangeDelAggregator* range_del_agg_;
  std::unique_ptr<CompactionProxy> compaction_;
  const CompactionFilter* compaction_filter_;
  const std::atomic<bool>* shutting_down_;
@@ -219,6 +151,7 @@ class CompactionIterator {
  bool visible_at_tip_;
  SequenceNumber earliest_snapshot_;
  SequenceNumber latest_snapshot_;
+  bool ignore_snapshots_;

  // State
  //
@@ -271,9 +204,6 @@ class CompactionIterator {
  // Used to avoid purging uncommitted values. The application can specify
  // uncommitted values by providing a SnapshotChecker object.
  bool current_key_committed_;
-  SnapshotListFetchCallback* snap_list_callback_;
-  // number of distinct keys processed
-  size_t num_keys_ = 0;

  bool IsShuttingDown() {
    // This is a best-effort facility, so memory_order_relaxed is sufficient.
@@ -112,39 +112,39 @@ class LoggingForwardVectorIterator : public InternalIterator {
    assert(keys_.size() == values_.size());
  }

-  bool Valid() const override { return current_ < keys_.size(); }
+  virtual bool Valid() const override { return current_ < keys_.size(); }

-  void SeekToFirst() override {
+  virtual void SeekToFirst() override {
    log.emplace_back(Action::Type::SEEK_TO_FIRST);
    current_ = 0;
  }
-  void SeekToLast() override { assert(false); }
+  virtual void SeekToLast() override { assert(false); }

-  void Seek(const Slice& target) override {
+  virtual void Seek(const Slice& target) override {
    log.emplace_back(Action::Type::SEEK, target.ToString());
    current_ = std::lower_bound(keys_.begin(), keys_.end(), target.ToString()) -
               keys_.begin();
  }

-  void SeekForPrev(const Slice& /*target*/) override { assert(false); }
+  virtual void SeekForPrev(const Slice& /*target*/) override { assert(false); }

-  void Next() override {
+  virtual void Next() override {
    assert(Valid());
    log.emplace_back(Action::Type::NEXT);
    current_++;
  }
-  void Prev() override { assert(false); }
+  virtual void Prev() override { assert(false); }

-  Slice key() const override {
+  virtual Slice key() const override {
    assert(Valid());
    return Slice(keys_[current_]);
  }
-  Slice value() const override {
+  virtual Slice value() const override {
    assert(Valid());
    return Slice(values_[current_]);
  }

-  Status status() const override { return Status::OK(); }
+  virtual Status status() const override { return Status::OK(); }

  std::vector<Action> log;

@@ -158,20 +158,22 @@ class FakeCompaction : public CompactionIterator::CompactionProxy {
 public:
  FakeCompaction() = default;

-  int level(size_t /*compaction_input_level*/) const override { return 0; }
-  bool KeyNotExistsBeyondOutputLevel(
+  virtual int level(size_t /*compaction_input_level*/) const override {
+    return 0;
+  }
+  virtual bool KeyNotExistsBeyondOutputLevel(
      const Slice& /*user_key*/,
      std::vector<size_t>* /*level_ptrs*/) const override {
    return is_bottommost_level || key_not_exists_beyond_output_level;
  }
-  bool bottommost_level() const override { return is_bottommost_level; }
-  int number_levels() const override { return 1; }
-  Slice GetLargestUserKey() const override {
+  virtual bool bottommost_level() const override { return is_bottommost_level; }
+  virtual int number_levels() const override { return 1; }
+  virtual Slice GetLargestUserKey() const override {
    return "\xff\xff\xff\xff\xff\xff\xff\xff\xff";
  }
-  bool allow_ingest_behind() const override { return false; }
+  virtual bool allow_ingest_behind() const override { return false; }

-  bool preserve_deletes() const override { return false; }
+  virtual bool preserve_deletes() const override { return false; }

  bool key_not_exists_beyond_output_level = false;

@@ -188,17 +190,13 @@ class TestSnapshotChecker : public SnapshotChecker {
      : last_committed_sequence_(last_committed_sequence),
        snapshots_(snapshots) {}

-  SnapshotCheckerResult CheckInSnapshot(
-      SequenceNumber seq, SequenceNumber snapshot_seq) const override {
+  bool IsInSnapshot(SequenceNumber seq,
+                    SequenceNumber snapshot_seq) const override {
    if (snapshot_seq == kMaxSequenceNumber) {
-      return seq <= last_committed_sequence_
-                 ? SnapshotCheckerResult::kInSnapshot
-                 : SnapshotCheckerResult::kNotInSnapshot;
+      return seq <= last_committed_sequence_;
    }
    assert(snapshots_.count(snapshot_seq) > 0);
-    return seq <= snapshots_.at(snapshot_seq)
-               ? SnapshotCheckerResult::kInSnapshot
-               : SnapshotCheckerResult::kNotInSnapshot;
+    return seq <= snapshots_.at(snapshot_seq);
  }

 private:
@@ -223,15 +221,10 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
      MergeOperator* merge_op = nullptr, CompactionFilter* filter = nullptr,
      bool bottommost_level = false,
      SequenceNumber earliest_write_conflict_snapshot = kMaxSequenceNumber) {
-    std::unique_ptr<InternalIterator> unfragmented_range_del_iter(
+    std::unique_ptr<InternalIterator> range_del_iter(
        new test::VectorIterator(range_del_ks, range_del_vs));
-    auto tombstone_list = std::make_shared<FragmentedRangeTombstoneList>(
-        std::move(unfragmented_range_del_iter), icmp_);
-    std::unique_ptr<FragmentedRangeTombstoneIterator> range_del_iter(
-        new FragmentedRangeTombstoneIterator(tombstone_list, icmp_,
-                                             kMaxSequenceNumber));
-    range_del_agg_.reset(new CompactionRangeDelAggregator(&icmp_, snapshots_));
-    range_del_agg_->AddTombstones(std::move(range_del_iter));
+    range_del_agg_.reset(new RangeDelAggregator(icmp_, snapshots_));
+    ASSERT_OK(range_del_agg_->AddTombstones(std::move(range_del_iter)));

    std::unique_ptr<CompactionIterator::CompactionProxy> compaction;
    if (filter || bottommost_level) {
@@ -254,8 +247,9 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
    c_iter_.reset(new CompactionIterator(
        iter_.get(), cmp_, merge_helper_.get(), last_sequence, &snapshots_,
        earliest_write_conflict_snapshot, snapshot_checker_.get(),
-        Env::Default(), false /* report_detailed_time */, false,
-        range_del_agg_.get(), std::move(compaction), filter, &shutting_down_));
+        Env::Default(), false /* report_detailed_time */,
+        false, range_del_agg_.get(), std::move(compaction), filter,
+        &shutting_down_));
  }

  void AddSnapshot(SequenceNumber snapshot,
@@ -299,7 +293,7 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
  std::unique_ptr<MergeHelper> merge_helper_;
  std::unique_ptr<LoggingForwardVectorIterator> iter_;
  std::unique_ptr<CompactionIterator> c_iter_;
-  std::unique_ptr<CompactionRangeDelAggregator> range_del_agg_;
+  std::unique_ptr<RangeDelAggregator> range_del_agg_;
  std::unique_ptr<SnapshotChecker> snapshot_checker_;
  std::atomic<bool> shutting_down_{false};
  FakeCompaction* compaction_proxy_;
@@ -375,9 +369,10 @@ TEST_P(CompactionIteratorTest, RangeDeletionWithSnapshots) {

 TEST_P(CompactionIteratorTest, CompactionFilterSkipUntil) {
  class Filter : public CompactionFilter {
-    Decision FilterV2(int /*level*/, const Slice& key, ValueType t,
-                      const Slice& existing_value, std::string* /*new_value*/,
-                      std::string* skip_until) const override {
+    virtual Decision FilterV2(int /*level*/, const Slice& key, ValueType t,
+                              const Slice& existing_value,
+                              std::string* /*new_value*/,
+                              std::string* skip_until) const override {
      std::string k = key.ToString();
      std::string v = existing_value.ToString();
      // See InitIterators() call below for the sequence of keys and their
@@ -557,9 +552,10 @@ TEST_P(CompactionIteratorTest, ShuttingDownInMerge) {

 TEST_P(CompactionIteratorTest, SingleMergeOperand) {
  class Filter : public CompactionFilter {
-    Decision FilterV2(int /*level*/, const Slice& key, ValueType t,
-                      const Slice& existing_value, std::string* /*new_value*/,
-                      std::string* /*skip_until*/) const override {
+    virtual Decision FilterV2(int /*level*/, const Slice& key, ValueType t,
+                              const Slice& existing_value,
+                              std::string* /*new_value*/,
+                              std::string* /*skip_until*/) const override {
      std::string k = key.ToString();
      std::string v = existing_value.ToString();

@@ -676,12 +672,8 @@ TEST_P(CompactionIteratorTest, ZeroOutSequenceAtBottomLevel) {
 TEST_P(CompactionIteratorTest, RemoveDeletionAtBottomLevel) {
  AddSnapshot(1);
  RunTest({test::KeyStr("a", 1, kTypeDeletion),
-           test::KeyStr("b", 3, kTypeDeletion),
-           test::KeyStr("b", 1, kTypeValue)},
-          {"", "", ""},
-          {test::KeyStr("b", 3, kTypeDeletion),
-           test::KeyStr("b", 0, kTypeValue)},
-          {"", ""},
+           test::KeyStr("b", 2, kTypeDeletion)},
+          {"", ""}, {test::KeyStr("b", 2, kTypeDeletion)}, {""},
          kMaxSequenceNumber /*last_commited_seq*/, nullptr /*merge_operator*/,
          nullptr /*compaction_filter*/, true /*bottommost_level*/);
 }
@@ -850,26 +842,13 @@ TEST_F(CompactionIteratorWithSnapshotCheckerTest,
      {test::KeyStr("a", 1, kTypeDeletion), test::KeyStr("b", 2, kTypeDeletion),
       test::KeyStr("c", 3, kTypeDeletion)},
      {"", "", ""},
-      {},
+      {test::KeyStr("b", 2, kTypeDeletion),
+       test::KeyStr("c", 3, kTypeDeletion)},
      {"", ""}, kMaxSequenceNumber /*last_commited_seq*/,
      nullptr /*merge_operator*/, nullptr /*compaction_filter*/,
      true /*bottommost_level*/);
 }

-TEST_F(CompactionIteratorWithSnapshotCheckerTest,
-       NotRemoveDeletionIfValuePresentToEarlierSnapshot) {
-  AddSnapshot(2,1);
-  RunTest(
-      {test::KeyStr("a", 4, kTypeDeletion), test::KeyStr("a", 1, kTypeValue),
-          test::KeyStr("b", 3, kTypeValue)},
-      {"", "", ""},
-      {test::KeyStr("a", 4, kTypeDeletion), test::KeyStr("a", 0, kTypeValue),
-            test::KeyStr("b", 3, kTypeValue)},
-      {"", "", ""}, kMaxSequenceNumber /*last_commited_seq*/,
-      nullptr /*merge_operator*/, nullptr /*compaction_filter*/,
-      true /*bottommost_level*/);
-}
-
 TEST_F(CompactionIteratorWithSnapshotCheckerTest,
       NotRemoveSingleDeletionIfNotVisibleToEarliestSnapshot) {
  AddSnapshot(2, 1);
@@ -36,7 +36,6 @@
 #include "db/memtable_list.h"
 #include "db/merge_context.h"
 #include "db/merge_helper.h"
-#include "db/range_del_aggregator.h"
 #include "db/version_set.h"
 #include "monitoring/iostats_context_imp.h"
 #include "monitoring/perf_context_imp.h"
@@ -97,8 +96,6 @@ const char* GetCompactionReasonString(CompactionReason compaction_reason) {
      return "Flush";
    case CompactionReason::kExternalSstIngestion:
      return "ExternalSstIngestion";
-    case CompactionReason::kPeriodicCompaction:
-      return "PeriodicCompaction";
    case CompactionReason::kNumOfReasons:
      // fall through
    default:
@@ -159,6 +156,7 @@ struct CompactionJob::SubcompactionState {
  uint64_t overlapped_bytes = 0;
  // A flag determine whether the key has been seen in ShouldStopBefore()
  bool seen_key = false;
+  std::string compression_dict;

  SubcompactionState(Compaction* c, Slice* _start, Slice* _end,
                     uint64_t size = 0)
@@ -174,7 +172,8 @@ struct CompactionJob::SubcompactionState {
        approx_size(size),
        grandparent_index(0),
        overlapped_bytes(0),
-        seen_key(false) {
+        seen_key(false),
+        compression_dict() {
    assert(compaction != nullptr);
  }

@@ -197,10 +196,11 @@ struct CompactionJob::SubcompactionState {
    grandparent_index = std::move(o.grandparent_index);
    overlapped_bytes = std::move(o.overlapped_bytes);
    seen_key = std::move(o.seen_key);
+    compression_dict = std::move(o.compression_dict);
    return *this;
  }

-  // Because member std::unique_ptrs do not have these.
+  // Because member unique_ptrs do not have these.
  SubcompactionState(const SubcompactionState&) = delete;

  SubcompactionState& operator=(const SubcompactionState&) = delete;
@@ -314,8 +314,7 @@ CompactionJob::CompactionJob(
    SequenceNumber earliest_write_conflict_snapshot,
    const SnapshotChecker* snapshot_checker, std::shared_ptr<Cache> table_cache,
    EventLogger* event_logger, bool paranoid_file_checks, bool measure_io_stats,
-    const std::string& dbname, CompactionJobStats* compaction_job_stats,
-    Env::Priority thread_pri, SnapshotListFetchCallback* snap_list_callback)
+    const std::string& dbname, CompactionJobStats* compaction_job_stats)
    : job_id_(job_id),
      compact_(new CompactionState(compaction)),
      compaction_job_stats_(compaction_job_stats),
@@ -336,7 +335,6 @@ CompactionJob::CompactionJob(
      db_mutex_(db_mutex),
      db_error_handler_(db_error_handler),
      existing_snapshots_(std::move(existing_snapshots)),
-      snap_list_callback_(snap_list_callback),
      earliest_write_conflict_snapshot_(earliest_write_conflict_snapshot),
      snapshot_checker_(snapshot_checker),
      table_cache_(std::move(table_cache)),
@@ -344,8 +342,7 @@ CompactionJob::CompactionJob(
      bottommost_level_(false),
      paranoid_file_checks_(paranoid_file_checks),
      measure_io_stats_(measure_io_stats),
-      write_hint_(Env::WLTH_NOT_SET),
-      thread_pri_(thread_pri) {
+      write_hint_(Env::WLTH_NOT_SET) {
  assert(log_buffer_ != nullptr);
  const auto* cfd = compact_->compaction->column_family_data();
  ThreadStatusUtil::SetColumnFamily(cfd, cfd->ioptions()->env,
@@ -419,10 +416,11 @@ void CompactionJob::Prepare() {
  bottommost_level_ = c->bottommost_level();

  if (c->ShouldFormSubcompactions()) {
-    {
-      StopWatch sw(env_, stats_, SUBCOMPACTION_SETUP_TIME);
-      GenSubcompactionBoundaries();
-    }
+    const uint64_t start_micros = env_->NowMicros();
+    GenSubcompactionBoundaries();
+    MeasureTime(stats_, SUBCOMPACTION_SETUP_TIME,
+                env_->NowMicros() - start_micros);
+
    assert(sizes_.size() == boundaries_.size() + 1);

    for (size_t i = 0; i <= boundaries_.size(); i++) {
@@ -430,8 +428,8 @@ void CompactionJob::Prepare() {
      Slice* end = i == boundaries_.size() ? nullptr : &boundaries_[i];
      compact_->sub_compact_states.emplace_back(c, start, end, sizes_[i]);
    }
-    RecordInHistogram(stats_, NUM_SUBCOMPACTIONS_SCHEDULED,
-                      compact_->sub_compact_states.size());
+    MeasureTime(stats_, NUM_SUBCOMPACTIONS_SCHEDULED,
+                compact_->sub_compact_states.size());
  } else {
    compact_->sub_compact_states.emplace_back(c, nullptr, nullptr);
  }
@@ -513,10 +511,7 @@ void CompactionJob::GenSubcompactionBoundaries() {
  // size of data covered by keys in that range
  uint64_t sum = 0;
  std::vector<RangeWithSize> ranges;
-  // Get input version from CompactionState since it's already referenced
-  // earlier in SetInputVersioCompaction::SetInputVersion and will not change
-  // when db_mutex_ is released below
-  auto* v = compact_->compaction->input_version();
+  auto* v = cfd->current();
  for (auto it = bounds.begin();;) {
    const Slice a = *it;
    it++;
@@ -526,13 +521,7 @@ void CompactionJob::GenSubcompactionBoundaries() {
    }

    const Slice b = *it;
-
-    // ApproximateSize could potentially create table reader iterator to seek
-    // to the index block and may incur I/O cost in the process. Unlock db
-    // mutex to reduce contention
-    db_mutex_->Unlock();
    uint64_t size = versions_->ApproximateSize(v, a, b, start_lvl, out_lvl + 1);
-    db_mutex_->Lock();
    ranges.emplace_back(a, b, size);
    sum += size;
  }
@@ -604,16 +593,12 @@ Status CompactionJob::Run() {
    thread.join();
  }

-  compaction_stats_.micros = env_->NowMicros() - start_micros;
-  compaction_stats_.cpu_micros = 0;
-  for (size_t i = 0; i < compact_->sub_compact_states.size(); i++) {
-    compaction_stats_.cpu_micros +=
-        compact_->sub_compact_states[i].compaction_job_stats.cpu_micros;
+  if (output_directory_) {
+    output_directory_->Fsync();
  }

-  RecordTimeToHistogram(stats_, COMPACTION_TIME, compaction_stats_.micros);
-  RecordTimeToHistogram(stats_, COMPACTION_CPU_TIME,
-                        compaction_stats_.cpu_micros);
+  compaction_stats_.micros = env_->NowMicros() - start_micros;
+  MeasureTime(stats_, COMPACTION_TIME, compaction_stats_.micros);

  TEST_SYNC_POINT("CompactionJob::Run:BeforeVerify");

@@ -626,10 +611,6 @@ Status CompactionJob::Run() {
    }
  }

-  if (status.ok() && output_directory_) {
-    status = output_directory_->Fsync();
-  }
-
  if (status.ok()) {
    thread_pool.clear();
    std::vector<const FileMetaData*> files_meta;
@@ -722,7 +703,7 @@ Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options) {
  Status status = compact_->status;
  ColumnFamilyData* cfd = compact_->compaction->column_family_data();
  cfd->internal_stats()->AddCompactionStats(
-      compact_->compaction->output_level(), thread_pri_, compaction_stats_);
+      compact_->compaction->output_level(), compaction_stats_);

  if (status.ok()) {
    status = InstallCompactionResults(mutable_cf_options);
@@ -776,7 +757,6 @@ Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options) {
  stream << "job" << job_id_ << "event"
         << "compaction_finished"
         << "compaction_time_micros" << compaction_stats_.micros
-         << "compaction_time_cpu_micros" << compaction_stats_.cpu_micros
         << "output_level" << compact_->compaction->output_level()
         << "num_output_files" << compact_->NumOutputFiles()
         << "total_output_size" << compact_->total_bytes << "num_input_records"
@@ -814,35 +794,11 @@ Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options) {

 void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
  assert(sub_compact != nullptr);
-
-  uint64_t prev_cpu_micros = env_->NowCPUNanos() / 1000;
-
  ColumnFamilyData* cfd = sub_compact->compaction->column_family_data();
-
-  // Create compaction filter and fail the compaction if
-  // IgnoreSnapshots() = false because it is not supported anymore
-  const CompactionFilter* compaction_filter =
-      cfd->ioptions()->compaction_filter;
-  std::unique_ptr<CompactionFilter> compaction_filter_from_factory = nullptr;
-  if (compaction_filter == nullptr) {
-    compaction_filter_from_factory =
-        sub_compact->compaction->CreateCompactionFilter();
-    compaction_filter = compaction_filter_from_factory.get();
-  }
-  if (compaction_filter != nullptr && !compaction_filter->IgnoreSnapshots()) {
-    sub_compact->status = Status::NotSupported(
-        "CompactionFilter::IgnoreSnapshots() = false is not supported "
-        "anymore.");
-    return;
-  }
-
-  CompactionRangeDelAggregator range_del_agg(&cfd->internal_comparator(),
-                                             existing_snapshots_);
-
-  // Although the v2 aggregator is what the level iterator(s) know about,
-  // the AddTombstones calls will be propagated down to the v1 aggregator.
+  std::unique_ptr<RangeDelAggregator> range_del_agg(
+      new RangeDelAggregator(cfd->internal_comparator(), existing_snapshots_));
  std::unique_ptr<InternalIterator> input(versions_->MakeInputIterator(
-      sub_compact->compaction, &range_del_agg, env_optiosn_for_read_));
+      sub_compact->compaction, range_del_agg.get(), env_optiosn_for_read_));

  AutoThreadOperationStageUpdater stage_updater(
      ThreadStatus::STAGE_COMPACTION_PROCESS_KV);
@@ -854,19 +810,58 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
  uint64_t prev_fsync_nanos = 0;
  uint64_t prev_range_sync_nanos = 0;
  uint64_t prev_prepare_write_nanos = 0;
-  uint64_t prev_cpu_write_nanos = 0;
-  uint64_t prev_cpu_read_nanos = 0;
  if (measure_io_stats_) {
    prev_perf_level = GetPerfLevel();
-    SetPerfLevel(PerfLevel::kEnableTimeAndCPUTimeExceptForMutex);
+    SetPerfLevel(PerfLevel::kEnableTime);
    prev_write_nanos = IOSTATS(write_nanos);
    prev_fsync_nanos = IOSTATS(fsync_nanos);
    prev_range_sync_nanos = IOSTATS(range_sync_nanos);
    prev_prepare_write_nanos = IOSTATS(prepare_write_nanos);
-    prev_cpu_write_nanos = IOSTATS(cpu_write_nanos);
-    prev_cpu_read_nanos = IOSTATS(cpu_read_nanos);
  }

+  const MutableCFOptions* mutable_cf_options =
+      sub_compact->compaction->mutable_cf_options();
+
+  // To build compression dictionary, we sample the first output file, assuming
+  // it'll reach the maximum length. We optionally pass these samples through
+  // zstd's dictionary trainer, or just use them directly. Then, the dictionary
+  // is used for compressing subsequent output files in the same subcompaction.
+  const bool kUseZstdTrainer =
+      sub_compact->compaction->output_compression_opts().zstd_max_train_bytes >
+      0;
+  const size_t kSampleBytes =
+      kUseZstdTrainer
+          ? sub_compact->compaction->output_compression_opts()
+                .zstd_max_train_bytes
+          : sub_compact->compaction->output_compression_opts().max_dict_bytes;
+  const int kSampleLenShift = 6;  // 2^6 = 64-byte samples
+  std::set<size_t> sample_begin_offsets;
+  if (bottommost_level_ && kSampleBytes > 0) {
+    const size_t kMaxSamples = kSampleBytes >> kSampleLenShift;
+    const size_t kOutFileLen =
+        static_cast<size_t>(MaxFileSizeForLevel(*mutable_cf_options,
+            compact_->compaction->output_level(),
+            cfd->ioptions()->compaction_style,
+            compact_->compaction->GetInputBaseLevel(),
+            cfd->ioptions()->level_compaction_dynamic_level_bytes));
+    if (kOutFileLen != port::kMaxSizet) {
+      const size_t kOutFileNumSamples = kOutFileLen >> kSampleLenShift;
+      Random64 generator{versions_->NewFileNumber()};
+      for (size_t i = 0; i < kMaxSamples; ++i) {
+        sample_begin_offsets.insert(
+            static_cast<size_t>(generator.Uniform(kOutFileNumSamples))
+            << kSampleLenShift);
+      }
+    }
+  }
+
+  auto compaction_filter = cfd->ioptions()->compaction_filter;
+  std::unique_ptr<CompactionFilter> compaction_filter_from_factory = nullptr;
+  if (compaction_filter == nullptr) {
+    compaction_filter_from_factory =
+        sub_compact->compaction->CreateCompactionFilter();
+    compaction_filter = compaction_filter_from_factory.get();
+  }
  MergeHelper merge(
      env_, cfd->user_comparator(), cfd->ioptions()->merge_operator,
      compaction_filter, db_options_.info_log.get(),
@@ -892,10 +887,8 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
      input.get(), cfd->user_comparator(), &merge, versions_->LastSequence(),
      &existing_snapshots_, earliest_write_conflict_snapshot_,
      snapshot_checker_, env_, ShouldReportDetailedTime(env_, stats_), false,
-      &range_del_agg, sub_compact->compaction, compaction_filter,
-      shutting_down_, preserve_deletes_seqnum_,
-      // Currently range_del_agg is incompatible with snapshot refresh feature.
-      range_del_agg.IsEmpty() ? snap_list_callback_ : nullptr));
+      range_del_agg.get(), sub_compact->compaction, compaction_filter,
+      shutting_down_, preserve_deletes_seqnum_));
  auto c_iter = sub_compact->c_iter.get();
  c_iter->SeekToFirst();
  if (c_iter->Valid() && sub_compact->compaction->output_level() != 0) {
@@ -906,6 +899,12 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
                                  sub_compact->current_output_file_size);
  }
  const auto& c_iter_stats = c_iter->iter_stats();
+  auto sample_begin_offset_iter = sample_begin_offsets.cbegin();
+  // data_begin_offset and dict_sample_data are only valid while generating
+  // dictionary from the first output file.
+  size_t data_begin_offset = 0;
+  std::string dict_sample_data;
+  dict_sample_data.reserve(kSampleBytes);

  while (status.ok() && !cfd->IsDropped() && c_iter->Valid()) {
    // Invariant: c_iter.status() is guaranteed to be OK if c_iter->Valid()
@@ -941,6 +940,55 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
        key, c_iter->ikey().sequence);
    sub_compact->num_output_records++;

+    if (sub_compact->outputs.size() == 1) {  // first output file
+      // Check if this key/value overlaps any sample intervals; if so, appends
+      // overlapping portions to the dictionary.
+      for (const auto& data_elmt : {key, value}) {
+        size_t data_end_offset = data_begin_offset + data_elmt.size();
+        while (sample_begin_offset_iter != sample_begin_offsets.cend() &&
+               *sample_begin_offset_iter < data_end_offset) {
+          size_t sample_end_offset =
+              *sample_begin_offset_iter + (1 << kSampleLenShift);
+          // Invariant: Because we advance sample iterator while processing the
+          // data_elmt containing the sample's last byte, the current sample
+          // cannot end before the current data_elmt.
+          assert(data_begin_offset < sample_end_offset);
+
+          size_t data_elmt_copy_offset, data_elmt_copy_len;
+          if (*sample_begin_offset_iter <= data_begin_offset) {
+            // The sample starts before data_elmt starts, so take bytes starting
+            // at the beginning of data_elmt.
+            data_elmt_copy_offset = 0;
+          } else {
+            // data_elmt starts before the sample starts, so take bytes starting
+            // at the below offset into data_elmt.
+            data_elmt_copy_offset =
+                *sample_begin_offset_iter - data_begin_offset;
+          }
+          if (sample_end_offset <= data_end_offset) {
+            // The sample ends before data_elmt ends, so take as many bytes as
+            // needed.
+            data_elmt_copy_len =
+                sample_end_offset - (data_begin_offset + data_elmt_copy_offset);
+          } else {
+            // data_elmt ends before the sample ends, so take all remaining
+            // bytes in data_elmt.
+            data_elmt_copy_len =
+                data_end_offset - (data_begin_offset + data_elmt_copy_offset);
+          }
+          dict_sample_data.append(&data_elmt.data()[data_elmt_copy_offset],
+                                  data_elmt_copy_len);
+          if (sample_end_offset > data_end_offset) {
+            // Didn't finish sample. Try to finish it with the next data_elmt.
+            break;
+          }
+          // Next sample may require bytes from same data_elmt.
+          sample_begin_offset_iter++;
+        }
+        data_begin_offset = data_end_offset;
+      }
+    }
+
    // Close output file if it is big enough. Two possibilities determine it's
    // time to close it: (1) the current key should be this file's last key, (2)
    // the next key should not be in this file.
@@ -977,11 +1025,23 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
        next_key = &c_iter->key();
      }
      CompactionIterationStats range_del_out_stats;
-      status =
-          FinishCompactionOutputFile(input_status, sub_compact, &range_del_agg,
-                                     &range_del_out_stats, next_key);
+      status = FinishCompactionOutputFile(input_status, sub_compact,
+                                          range_del_agg.get(),
+                                          &range_del_out_stats, next_key);
      RecordDroppedKeys(range_del_out_stats,
                        &sub_compact->compaction_job_stats);
+      if (sub_compact->outputs.size() == 1) {
+        // Use samples from first output file to create dictionary for
+        // compression of subsequent files.
+        if (kUseZstdTrainer) {
+          sub_compact->compression_dict = ZSTD_TrainDictionary(
+              dict_sample_data, kSampleLenShift,
+              sub_compact->compaction->output_compression_opts()
+                  .max_dict_bytes);
+        } else {
+          sub_compact->compression_dict = std::move(dict_sample_data);
+        }
+      }
    }
  }

@@ -1017,7 +1077,8 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
  }

  if (status.ok() && sub_compact->builder == nullptr &&
-      sub_compact->outputs.size() == 0 && !range_del_agg.IsEmpty()) {
+      sub_compact->outputs.size() == 0 &&
+      !range_del_agg->IsEmpty()) {
    // handle subcompaction containing only range deletions
    status = OpenCompactionOutputFile(sub_compact);
  }
@@ -1026,17 +1087,14 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
  // close the output file.
  if (sub_compact->builder != nullptr) {
    CompactionIterationStats range_del_out_stats;
-    Status s = FinishCompactionOutputFile(status, sub_compact, &range_del_agg,
-                                          &range_del_out_stats);
+    Status s = FinishCompactionOutputFile(
+        status, sub_compact, range_del_agg.get(), &range_del_out_stats);
    if (status.ok()) {
      status = s;
    }
    RecordDroppedKeys(range_del_out_stats, &sub_compact->compaction_job_stats);
  }

-  sub_compact->compaction_job_stats.cpu_micros =
-      env_->NowCPUNanos() / 1000 - prev_cpu_micros;
-
  if (measure_io_stats_) {
    sub_compact->compaction_job_stats.file_write_nanos +=
        IOSTATS(write_nanos) - prev_write_nanos;
@@ -1046,11 +1104,7 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
        IOSTATS(range_sync_nanos) - prev_range_sync_nanos;
    sub_compact->compaction_job_stats.file_prepare_write_nanos +=
        IOSTATS(prepare_write_nanos) - prev_prepare_write_nanos;
-    sub_compact->compaction_job_stats.cpu_micros -=
-        (IOSTATS(cpu_write_nanos) - prev_cpu_write_nanos +
-         IOSTATS(cpu_read_nanos) - prev_cpu_read_nanos) /
-        1000;
-    if (prev_perf_level != PerfLevel::kEnableTimeAndCPUTimeExceptForMutex) {
+    if (prev_perf_level != PerfLevel::kEnableTime) {
      SetPerfLevel(prev_perf_level);
    }
  }
@@ -1099,7 +1153,7 @@ void CompactionJob::RecordDroppedKeys(

 Status CompactionJob::FinishCompactionOutputFile(
    const Status& input_status, SubcompactionState* sub_compact,
-    CompactionRangeDelAggregator* range_del_agg,
+    RangeDelAggregator* range_del_agg,
    CompactionIterationStats* range_del_out_stats,
    const Slice* next_table_min_key /* = nullptr */) {
  AutoThreadOperationStageUpdater stage_updater(
@@ -1123,12 +1177,10 @@ Status CompactionJob::FinishCompactionOutputFile(
    Slice lower_bound_guard, upper_bound_guard;
    std::string smallest_user_key;
    const Slice *lower_bound, *upper_bound;
-    bool lower_bound_from_sub_compact = false;
    if (sub_compact->outputs.size() == 1) {
      // For the first output table, include range tombstones before the min key
      // but after the subcompaction boundary.
      lower_bound = sub_compact->start;
-      lower_bound_from_sub_compact = true;
    } else if (meta->smallest.size() > 0) {
      // For subsequent output tables, only include range tombstones from min
      // key onwards since the previous file was extended to contain range
@@ -1140,19 +1192,10 @@ Status CompactionJob::FinishCompactionOutputFile(
      lower_bound = nullptr;
    }
    if (next_table_min_key != nullptr) {
-      // This may be the last file in the subcompaction in some cases, so we
-      // need to compare the end key of subcompaction with the next file start
-      // key. When the end key is chosen by the subcompaction, we know that
-      // it must be the biggest key in output file. Therefore, it is safe to
-      // use the smaller key as the upper bound of the output file, to ensure
-      // that there is no overlapping between different output files.
+      // This isn't the last file in the subcompaction, so extend until the next
+      // file starts.
      upper_bound_guard = ExtractUserKey(*next_table_min_key);
-      if (sub_compact->end != nullptr &&
-          ucmp->Compare(upper_bound_guard, *sub_compact->end) >= 0) {
-        upper_bound = sub_compact->end;
-      } else {
-        upper_bound = &upper_bound_guard;
-      }
+      upper_bound = &upper_bound_guard;
    } else {
      // This is the last file in the subcompaction, so extend until the
      // subcompaction ends.
@@ -1162,45 +1205,18 @@ Status CompactionJob::FinishCompactionOutputFile(
    if (existing_snapshots_.size() > 0) {
      earliest_snapshot = existing_snapshots_[0];
    }
-    bool has_overlapping_endpoints;
-    if (upper_bound != nullptr && meta->largest.size() > 0) {
-      has_overlapping_endpoints =
-          ucmp->Compare(meta->largest.user_key(), *upper_bound) == 0;
-    } else {
-      has_overlapping_endpoints = false;
-    }
-
-    // The end key of the subcompaction must be bigger or equal to the upper
-    // bound. If the end of subcompaction is null or the upper bound is null,
-    // it means that this file is the last file in the compaction. So there
-    // will be no overlapping between this file and others.
-    assert(sub_compact->end == nullptr ||
-           upper_bound == nullptr ||
-           ucmp->Compare(*upper_bound , *sub_compact->end) <= 0);
-    auto it = range_del_agg->NewIterator(lower_bound, upper_bound,
-                                         has_overlapping_endpoints);
-    // Position the range tombstone output iterator. There may be tombstone
-    // fragments that are entirely out of range, so make sure that we do not
-    // include those.
+    auto it = range_del_agg->NewIterator();
    if (lower_bound != nullptr) {
      it->Seek(*lower_bound);
-    } else {
-      it->SeekToFirst();
    }
    for (; it->Valid(); it->Next()) {
      auto tombstone = it->Tombstone();
-      if (upper_bound != nullptr) {
-        int cmp = ucmp->Compare(*upper_bound, tombstone.start_key_);
-        if ((has_overlapping_endpoints && cmp < 0) ||
-            (!has_overlapping_endpoints && cmp <= 0)) {
-          // Tombstones starting after upper_bound only need to be included in
-          // the next table. If the current SST ends before upper_bound, i.e.,
-          // `has_overlapping_endpoints == false`, we can also skip over range
-          // tombstones that start exactly at upper_bound. Such range tombstones
-          // will be included in the next file and are not relevant to the point
-          // keys or endpoints of the current file.
-          break;
-        }
+      if (upper_bound != nullptr &&
+          ucmp->Compare(*upper_bound, tombstone.start_key_) <= 0) {
+        // Tombstones starting at upper_bound or later only need to be included
+        // in the next table. Break because subsequent tombstones will start
+        // even later.
+        break;
      }

      if (bottommost_level_ && tombstone.seq_ <= earliest_snapshot) {
@@ -1212,8 +1228,6 @@ Status CompactionJob::FinishCompactionOutputFile(
      }

      auto kv = tombstone.Serialize();
-      assert(lower_bound == nullptr ||
-             ucmp->Compare(*lower_bound, kv.second) < 0);
      sub_compact->builder->Add(kv.first.Encode(), kv.second);
      InternalKey smallest_candidate = std::move(kv.first);
      if (lower_bound != nullptr &&
@@ -1222,24 +1236,11 @@ Status CompactionJob::FinishCompactionOutputFile(
        // (the max key in the previous table or subcompaction) in order for
        // files to appear key-space partitioned.
        //
-        // When lower_bound is chosen by a subcompaction, we know that
-        // subcompactions over smaller keys cannot contain any keys at
-        // lower_bound. We also know that smaller subcompactions exist, because
-        // otherwise the subcompaction woud be unbounded on the left. As a
-        // result, we know that no other files on the output level will contain
-        // actual keys at lower_bound (an output file may have a largest key of
-        // lower_bound@kMaxSequenceNumber, but this only indicates a large range
-        // tombstone was truncated). Therefore, it is safe to use the
-        // tombstone's sequence number, to ensure that keys at lower_bound at
-        // lower levels are covered by truncated tombstones.
-        //
-        // If lower_bound was chosen by the smallest data key in the file,
-        // choose lowest seqnum so this file's smallest internal key comes after
-        // the previous file's largest. The fake seqnum is OK because the read
-        // path's file-picking code only considers user key.
-        smallest_candidate = InternalKey(
-            *lower_bound, lower_bound_from_sub_compact ? tombstone.seq_ : 0,
-            kTypeRangeDeletion);
+        // Choose lowest seqnum so this file's smallest internal key comes
+        // after the previous file's/subcompaction's largest. The fake seqnum
+        // is OK because the read path's file-picking code only considers user
+        // key.
+        smallest_candidate = InternalKey(*lower_bound, 0, kTypeRangeDeletion);
      }
      InternalKey largest_candidate = tombstone.SerializeEndKey();
      if (upper_bound != nullptr &&
@@ -1261,23 +1262,9 @@ Status CompactionJob::FinishCompactionOutputFile(
        largest_candidate =
            InternalKey(*upper_bound, kMaxSequenceNumber, kTypeRangeDeletion);
      }
-#ifndef NDEBUG
-      SequenceNumber smallest_ikey_seqnum = kMaxSequenceNumber;
-      if (meta->smallest.size() > 0) {
-        smallest_ikey_seqnum = GetInternalKeySeqno(meta->smallest.Encode());
-      }
-#endif
      meta->UpdateBoundariesForRange(smallest_candidate, largest_candidate,
                                     tombstone.seq_,
                                     cfd->internal_comparator());
-
-      // The smallest key in a file is used for range tombstone truncation, so
-      // it cannot have a seqnum of 0 (unless the smallest data key in a file
-      // has a seqnum of 0). Otherwise, the truncated tombstone may expose
-      // deleted keys at lower levels.
-      assert(smallest_ikey_seqnum == 0 ||
-             ExtractInternalKeyFooter(meta->smallest.Encode()) !=
-                 PackSequenceAndType(0, kTypeRangeDeletion));
    }
    meta->marked_for_compaction = sub_compact->builder->NeedCompact();
  }
@@ -1322,7 +1309,9 @@ Status CompactionJob::FinishCompactionOutputFile(
    // VersionEdit.
    assert(!sub_compact->outputs.empty());
    sub_compact->outputs.pop_back();
-    meta = nullptr;
+    sub_compact->builder.reset();
+    sub_compact->current_output_file_size = 0;
+    return s;
  }

  if (s.ok() && (current_entries > 0 || tp.num_range_deletions > 0)) {
@@ -1355,7 +1344,10 @@ Status CompactionJob::FinishCompactionOutputFile(
  auto sfm =
      static_cast<SstFileManagerImpl*>(db_options_.sst_file_manager.get());
  if (sfm && meta != nullptr && meta->fd.GetPathId() == 0) {
-    sfm->OnAddFile(fname);
+    auto fn =
+        TableFileName(sub_compact->compaction->immutable_cf_options()->cf_paths,
+                      meta->fd.GetNumber(), meta->fd.GetPathId());
+    sfm->OnAddFile(fn);
    if (sfm->IsMaxAllowedSpaceReached()) {
      // TODO(ajkr): should we return OK() if max space was reached by the final
      // compaction output file (similarly to how flush works when full)?
@@ -1441,7 +1433,7 @@ Status CompactionJob::OpenCompactionOutputFile(
      TableFileCreationReason::kCompaction);
 #endif  // !ROCKSDB_LITE
  // Make the output file
-  std::unique_ptr<WritableFile> writable_file;
+  unique_ptr<WritableFile> writable_file;
 #ifndef NDEBUG
  bool syncpoint_arg = env_options_.use_direct_writes;
  TEST_SYNC_POINT_CALLBACK("CompactionJob::OpenCompactionOutputFile",
@@ -1473,11 +1465,8 @@ Status CompactionJob::OpenCompactionOutputFile(
  writable_file->SetWriteLifeTimeHint(write_hint_);
  writable_file->SetPreallocationBlockSize(static_cast<size_t>(
      sub_compact->compaction->OutputFilePreallocationSize()));
-  const auto& listeners =
-      sub_compact->compaction->immutable_cf_options()->listeners;
-  sub_compact->outfile.reset(
-      new WritableFileWriter(std::move(writable_file), fname, env_options_,
-                             env_, db_options_.statistics.get(), listeners));
+  sub_compact->outfile.reset(new WritableFileWriter(
+      std::move(writable_file), env_options_, db_options_.statistics.get()));

  // If the Column family flag is to only optimize filters for hits,
  // we can skip creating filters if this is the bottommost_level where
@@ -1485,20 +1474,20 @@ Status CompactionJob::OpenCompactionOutputFile(
  bool skip_filters =
      cfd->ioptions()->optimize_filters_for_hits && bottommost_level_;

-  int64_t temp_current_time = 0;
-  auto get_time_status = env_->GetCurrentTime(&temp_current_time);
-  // Safe to proceed even if GetCurrentTime fails. So, log and proceed.
-  if (!get_time_status.ok()) {
-    ROCKS_LOG_WARN(db_options_.info_log,
-                   "Failed to get current time. Status: %s",
-                   get_time_status.ToString().c_str());
-  }
-  uint64_t current_time = static_cast<uint64_t>(temp_current_time);
-
-  uint64_t latest_key_time =
+  uint64_t output_file_creation_time =
      sub_compact->compaction->MaxInputFileCreationTime();
-  if (latest_key_time == 0) {
-    latest_key_time = current_time;
+  if (output_file_creation_time == 0) {
+    int64_t _current_time = 0;
+    auto status = db_options_.env->GetCurrentTime(&_current_time);
+    // Safe to proceed even if GetCurrentTime fails. So, log and proceed.
+    if (!status.ok()) {
+      ROCKS_LOG_WARN(
+          db_options_.info_log,
+          "Failed to get current time to populate creation_time property. "
+          "Status: %s",
+          status.ToString().c_str());
+    }
+    output_file_creation_time = static_cast<uint64_t>(_current_time);
  }

  sub_compact->builder.reset(NewTableBuilder(
@@ -1506,11 +1495,9 @@ Status CompactionJob::OpenCompactionOutputFile(
      cfd->internal_comparator(), cfd->int_tbl_prop_collector_factories(),
      cfd->GetID(), cfd->GetName(), sub_compact->outfile.get(),
      sub_compact->compaction->output_compression(),
-      0 /*sample_for_compression */,
      sub_compact->compaction->output_compression_opts(),
-      sub_compact->compaction->output_level(), skip_filters, latest_key_time,
-      0 /* oldest_key_time */, sub_compact->compaction->max_output_file_size(),
-      current_time));
+      sub_compact->compaction->output_level(), &sub_compact->compression_dict,
+      skip_filters, output_file_creation_time));
  LogFlush(db_options_.info_log);
  return s;
 }
@@ -29,8 +29,8 @@
 #include "db/version_edit.h"
 #include "db/write_controller.h"
 #include "db/write_thread.h"
-#include "options/cf_options.h"
 #include "options/db_options.h"
+#include "options/cf_options.h"
 #include "port/port.h"
 #include "rocksdb/compaction_filter.h"
 #include "rocksdb/compaction_job_stats.h"
@@ -57,20 +57,22 @@ class VersionSet;

 class CompactionJob {
 public:
-  CompactionJob(
-      int job_id, Compaction* compaction, const ImmutableDBOptions& db_options,
-      const EnvOptions env_options, VersionSet* versions,
-      const std::atomic<bool>* shutting_down,
-      const SequenceNumber preserve_deletes_seqnum, LogBuffer* log_buffer,
-      Directory* db_directory, Directory* output_directory, Statistics* stats,
-      InstrumentedMutex* db_mutex, ErrorHandler* db_error_handler,
-      std::vector<SequenceNumber> existing_snapshots,
-      SequenceNumber earliest_write_conflict_snapshot,
-      const SnapshotChecker* snapshot_checker,
-      std::shared_ptr<Cache> table_cache, EventLogger* event_logger,
-      bool paranoid_file_checks, bool measure_io_stats,
-      const std::string& dbname, CompactionJobStats* compaction_job_stats,
-      Env::Priority thread_pri, SnapshotListFetchCallback* snap_list_callback);
+  CompactionJob(int job_id, Compaction* compaction,
+                const ImmutableDBOptions& db_options,
+                const EnvOptions env_options, VersionSet* versions,
+                const std::atomic<bool>* shutting_down,
+                const SequenceNumber preserve_deletes_seqnum,
+                LogBuffer* log_buffer,
+                Directory* db_directory, Directory* output_directory,
+                Statistics* stats, InstrumentedMutex* db_mutex,
+                ErrorHandler* db_error_handler,
+                std::vector<SequenceNumber> existing_snapshots,
+                SequenceNumber earliest_write_conflict_snapshot,
+                const SnapshotChecker* snapshot_checker,
+                std::shared_ptr<Cache> table_cache, EventLogger* event_logger,
+                bool paranoid_file_checks, bool measure_io_stats,
+                const std::string& dbname,
+                CompactionJobStats* compaction_job_stats);

  ~CompactionJob();

@@ -102,7 +104,7 @@ class CompactionJob {

  Status FinishCompactionOutputFile(
      const Status& input_status, SubcompactionState* sub_compact,
-      CompactionRangeDelAggregator* range_del_agg,
+      RangeDelAggregator* range_del_agg,
      CompactionIterationStats* range_del_out_stats,
      const Slice* next_table_min_key = nullptr);
  Status InstallCompactionResults(const MutableCFOptions& mutable_cf_options);
@@ -150,7 +152,6 @@ class CompactionJob {
  // entirely within s1 and s2, then the earlier version of k1 can be safely
  // deleted because that version is not visible in any snapshot.
  std::vector<SequenceNumber> existing_snapshots_;
-  SnapshotListFetchCallback* snap_list_callback_;

  // This is the earliest snapshot that could be used for write-conflict
  // checking by a transaction.  For any user-key newer than this snapshot, we
@@ -171,7 +172,6 @@ class CompactionJob {
  // Stores the approx size of keys covered in the range of each subcompaction
  std::vector<uint64_t> sizes_;
  Env::WriteLifeTimeHint write_hint_;
-  Env::Priority thread_pri_;
 };

 }  // namespace rocksdb
@@ -113,7 +113,7 @@ class CompactionJobStatsTest : public testing::Test,
    Reopen(options);
  }

-  ~CompactionJobStatsTest() override {
+  ~CompactionJobStatsTest() {
    rocksdb::SyncPoint::GetInstance()->DisableProcessing();
    rocksdb::SyncPoint::GetInstance()->LoadDependency({});
    rocksdb::SyncPoint::GetInstance()->ClearAllCallBacks();
@@ -426,7 +426,7 @@ class CompactionJobStatsChecker : public EventListener {
  // Once a compaction completed, this function will verify the returned
  // CompactionJobInfo with the oldest CompactionJobInfo added earlier
  // in "expected_stats_" which has not yet being used for verification.
-  void OnCompactionCompleted(DB* /*db*/, const CompactionJobInfo& ci) override {
+  virtual void OnCompactionCompleted(DB* /*db*/, const CompactionJobInfo& ci) {
    if (verify_next_comp_io_stats_) {
      ASSERT_GT(ci.stats.file_write_nanos, 0);
      ASSERT_GT(ci.stats.file_range_sync_nanos, 0);
@@ -523,7 +523,7 @@ class CompactionJobDeletionStatsChecker : public CompactionJobStatsChecker {
 public:
  // Verifies whether two CompactionJobStats match.
  void Verify(const CompactionJobStats& current_stats,
-              const CompactionJobStats& stats) override {
+              const CompactionJobStats& stats) {
    ASSERT_EQ(
      current_stats.num_input_deletion_records,
      stats.num_input_deletion_records);
@@ -5,13 +5,7 @@

 #ifndef ROCKSDB_LITE

-#ifndef __STDC_FORMAT_MACROS
-#define __STDC_FORMAT_MACROS
-#endif
-
-#include <inttypes.h>
 #include <algorithm>
-#include <array>
 #include <map>
 #include <string>
 #include <tuple>
@@ -85,7 +79,7 @@ class CompactionJobTest : public testing::Test {
        shutting_down_(false),
        preserve_deletes_seqnum_(0),
        mock_table_factory_(new mock::MockTableFactory()),
-        error_handler_(nullptr, db_options_, &mutex_) {
+        error_handler_(db_options_, &mutex_) {
    EXPECT_OK(env_->CreateDirIfMissing(dbname_));
    db_options_.db_paths.emplace_back(dbname_,
                                      std::numeric_limits<uint64_t>::max());
@@ -178,7 +172,7 @@ class CompactionJobTest : public testing::Test {
        // This is how the key will look like once it's written in bottommost
        // file
        InternalKey bottommost_internal_key(
-            key, 0, kTypeValue);
+            key, (key == "9999") ? sequence_number : 0, kTypeValue);

        if (corrupt_id(k)) {
          test::CorruptKeyType(&internal_key);
@@ -200,25 +194,18 @@ class CompactionJobTest : public testing::Test {
  }

  void NewDB() {
-    DestroyDB(dbname_, Options());
-    EXPECT_OK(env_->CreateDirIfMissing(dbname_));
-    versions_.reset(new VersionSet(dbname_, &db_options_, env_options_,
-                                   table_cache_.get(), &write_buffer_manager_,
-                                   &write_controller_));
-    compaction_job_stats_.Reset();
-
    VersionEdit new_db;
    new_db.SetLogNumber(0);
    new_db.SetNextFile(2);
    new_db.SetLastSequence(0);

    const std::string manifest = DescriptorFileName(dbname_, 1);
-    std::unique_ptr<WritableFile> file;
+    unique_ptr<WritableFile> file;
    Status s = env_->NewWritableFile(
        manifest, &file, env_->OptimizeForManifestWrite(env_options_));
    ASSERT_OK(s);
-    std::unique_ptr<WritableFileWriter> file_writer(
-        new WritableFileWriter(std::move(file), manifest, env_options_));
+    unique_ptr<WritableFileWriter> file_writer(
+        new WritableFileWriter(std::move(file), env_options_));
    {
      log::Writer log(std::move(file_writer), 0, false);
      std::string record;
@@ -243,10 +230,7 @@ class CompactionJobTest : public testing::Test {
      const std::vector<std::vector<FileMetaData*>>& input_files,
      const stl_wrappers::KVMap& expected_results,
      const std::vector<SequenceNumber>& snapshots = {},
-      SequenceNumber earliest_write_conflict_snapshot = kMaxSequenceNumber,
-      int output_level = 1, bool verify = true,
-      SnapshotListFetchCallback* snapshot_fetcher =
-          SnapshotListFetchCallback::kDisabled) {
+      SequenceNumber earliest_write_conflict_snapshot = kMaxSequenceNumber) {
    auto cfd = versions_->GetColumnFamilySet()->GetDefault();

    size_t num_input_files = 0;
@@ -263,7 +247,7 @@ class CompactionJobTest : public testing::Test {

    Compaction compaction(cfd->current()->storage_info(), *cfd->ioptions(),
                          *cfd->GetLatestMutableCFOptions(),
-                          compaction_input_files, output_level, 1024 * 1024,
+                          compaction_input_files, 1, 1024 * 1024,
                          10 * 1024 * 1024, 0, kNoCompression,
                          cfd->ioptions()->compression_opts, 0, {}, true);
    compaction.SetInputVersion(cfd->current());
@@ -278,8 +262,7 @@ class CompactionJobTest : public testing::Test {
        &shutting_down_, preserve_deletes_seqnum_, &log_buffer, nullptr,
        nullptr, nullptr, &mutex_, &error_handler_, snapshots,
        earliest_write_conflict_snapshot, snapshot_checker, table_cache_,
-        &event_logger, false, false, dbname_, &compaction_job_stats_,
-        Env::Priority::USER, snapshot_fetcher);
+        &event_logger, false, false, dbname_, &compaction_job_stats_);
    VerifyInitializationOfCompactionJobStats(compaction_job_stats_);

    compaction_job.Prepare();
@@ -291,17 +274,15 @@ class CompactionJobTest : public testing::Test {
    ASSERT_OK(compaction_job.Install(*cfd->GetLatestMutableCFOptions()));
    mutex_.Unlock();

-    if (verify) {
-      if (expected_results.size() == 0) {
-        ASSERT_GE(compaction_job_stats_.elapsed_micros, 0U);
-        ASSERT_EQ(compaction_job_stats_.num_input_files, num_input_files);
-        ASSERT_EQ(compaction_job_stats_.num_output_files, 0U);
-      } else {
-        ASSERT_GE(compaction_job_stats_.elapsed_micros, 0U);
-        ASSERT_EQ(compaction_job_stats_.num_input_files, num_input_files);
-        ASSERT_EQ(compaction_job_stats_.num_output_files, 1U);
-        mock_table_factory_->AssertLatestFile(expected_results);
-      }
+    if (expected_results.size() == 0) {
+      ASSERT_GE(compaction_job_stats_.elapsed_micros, 0U);
+      ASSERT_EQ(compaction_job_stats_.num_input_files, num_input_files);
+      ASSERT_EQ(compaction_job_stats_.num_output_files, 0U);
+    } else {
+      ASSERT_GE(compaction_job_stats_.elapsed_micros, 0U);
+      ASSERT_EQ(compaction_job_stats_.num_input_files, num_input_files);
+      ASSERT_EQ(compaction_job_stats_.num_output_files, 1U);
+      mock_table_factory_->AssertLatestFile(expected_results);
    }
  }

@@ -398,7 +379,7 @@ TEST_F(CompactionJobTest, SimpleOverwrite) {

  auto expected_results =
      mock::MakeMockFile({{KeyStr("a", 0U, kTypeValue), "val2"},
-                          {KeyStr("b", 0U, kTypeValue), "val3"}});
+                          {KeyStr("b", 4U, kTypeValue), "val3"}});

  SetLastSequence(4U);
  auto files = cfd_->current()->storage_info()->LevelFiles(0);
@@ -451,7 +432,7 @@ TEST_F(CompactionJobTest, SimpleMerge) {

  auto expected_results =
      mock::MakeMockFile({{KeyStr("a", 0U, kTypeValue), "3,4,5"},
-                          {KeyStr("b", 0U, kTypeValue), "1,2"}});
+                          {KeyStr("b", 2U, kTypeValue), "1,2"}});

  SetLastSequence(5U);
  auto files = cfd_->current()->storage_info()->LevelFiles(0);
@@ -475,7 +456,8 @@ TEST_F(CompactionJobTest, NonAssocMerge) {

  auto expected_results =
      mock::MakeMockFile({{KeyStr("a", 0U, kTypeValue), "3,4,5"},
-                          {KeyStr("b", 0U, kTypeValue), "1,2"}});
+                          {KeyStr("b", 2U, kTypeMerge), "2"},
+                          {KeyStr("b", 1U, kTypeMerge), "1"}});

  SetLastSequence(5U);
  auto files = cfd_->current()->storage_info()->LevelFiles(0);
@@ -502,7 +484,7 @@ TEST_F(CompactionJobTest, MergeOperandFilter) {

  auto expected_results =
      mock::MakeMockFile({{KeyStr("a", 0U, kTypeValue), test::EncodeInt(8U)},
-                          {KeyStr("b", 0U, kTypeValue), test::EncodeInt(2U)}});
+                          {KeyStr("b", 2U, kTypeMerge), test::EncodeInt(2U)}});

  SetLastSequence(5U);
  auto files = cfd_->current()->storage_info()->LevelFiles(0);
@@ -765,7 +747,7 @@ TEST_F(CompactionJobTest, SingleDeleteZeroSeq) {
  AddMockFile(file2);

  auto expected_results = mock::MakeMockFile({
-      {KeyStr("dummy", 0U, kTypeValue), "val2"},
+      {KeyStr("dummy", 5U, kTypeValue), "val2"},
  });

  SetLastSequence(22U);
@@ -949,112 +931,13 @@ TEST_F(CompactionJobTest, CorruptionAfterDeletion) {
      mock::MakeMockFile({{test::KeyStr("A", 0U, kTypeValue), "val3"},
                          {test::KeyStr("a", 0U, kTypeValue, true), "val"},
                          {test::KeyStr("b", 0U, kTypeValue, true), "val"},
-                          {test::KeyStr("c", 0U, kTypeValue), "val2"}});
+                          {test::KeyStr("c", 1U, kTypeValue), "val2"}});

  SetLastSequence(6U);
  auto files = cfd_->current()->storage_info()->LevelFiles(0);
  RunCompaction({files}, expected_results);
 }

-// Test the snapshot fetcher in compaction
-TEST_F(CompactionJobTest, SnapshotRefresh) {
-  uint64_t time_seed = env_->NowMicros();
-  printf("time_seed is %" PRIu64 "\n", time_seed);  // would help to reproduce
-  Random64 rand(time_seed);
-  std::vector<SequenceNumber> db_snapshots;
-  class SnapshotListFetchCallbackTest : public SnapshotListFetchCallback {
-   public:
-    SnapshotListFetchCallbackTest(Env* env, Random64& rand,
-                                  std::vector<SequenceNumber>* snapshots)
-        : SnapshotListFetchCallback(env, 1 /*short time delay*/,
-                                    1 /*fetch after each key*/),
-          rand_(rand),
-          snapshots_(snapshots) {}
-    virtual void Refresh(std::vector<SequenceNumber>* snapshots,
-                         SequenceNumber) override {
-      assert(snapshots->size());
-      assert(snapshots_->size());
-      assert(snapshots_->size() == snapshots->size());
-      if (rand_.OneIn(2)) {
-        uint64_t release_index = rand_.Uniform(snapshots_->size());
-        snapshots_->erase(snapshots_->begin() + release_index);
-        *snapshots = *snapshots_;
-      }
-    }
-
-   private:
-    Random64 rand_;
-    std::vector<SequenceNumber>* snapshots_;
-  } snapshot_fetcher(env_, rand, &db_snapshots);
-
-  std::vector<std::pair<const std::string, std::string>> file1_kvs, file2_kvs;
-  std::array<ValueType, 4> types = {kTypeValue, kTypeDeletion,
-                                    kTypeSingleDeletion};
-  SequenceNumber last_seq = 0;
-  for (int i = 1; i < 100; i++) {
-    SequenceNumber seq = last_seq + 1;
-    last_seq = seq;
-    if (rand.OneIn(2)) {
-      auto type = types[rand.Uniform(types.size())];
-      file1_kvs.push_back(
-          {test::KeyStr("k" + ToString(i), seq, type), "v" + ToString(i)});
-    }
-  }
-  auto file1 = mock::MakeMockFile(file1_kvs);
-  for (int i = 1; i < 100; i++) {
-    SequenceNumber seq = last_seq + 1;
-    last_seq++;
-    if (rand.OneIn(2)) {
-      auto type = types[rand.Uniform(types.size())];
-      file2_kvs.push_back(
-          {test::KeyStr("k" + ToString(i), seq, type), "v" + ToString(i)});
-    }
-  }
-  auto file2 = mock::MakeMockFile(file2_kvs);
-  for (SequenceNumber i = 1; i < last_seq + 1; i++) {
-    if (rand.OneIn(5)) {
-      db_snapshots.push_back(i);
-    }
-  }
-
-  const bool kVerify = true;
-  const int output_level_0 = 0;
-  NewDB();
-  AddMockFile(file1);
-  AddMockFile(file2);
-  SetLastSequence(last_seq);
-  auto files = cfd_->current()->storage_info()->LevelFiles(0);
-  // put the output on L0 since it is easier to feed them again to the 2nd
-  // compaction
-  RunCompaction({files}, file1, db_snapshots, kMaxSequenceNumber,
-                output_level_0, !kVerify, &snapshot_fetcher);
-
-  // Now db_snapshots are changed. Run the compaction again without snapshot
-  // fetcher but with the updated snapshot list.
-  compaction_job_stats_.Reset();
-  files = cfd_->current()->storage_info()->LevelFiles(0);
-  RunCompaction({files}, file1, db_snapshots, kMaxSequenceNumber,
-                output_level_0 + 1, !kVerify);
-  // The result should be what we get if we run compaction without snapshot
-  // fetcher on the updated list of snapshots
-  auto expected = mock_table_factory_->output();
-
-  NewDB();
-  AddMockFile(file1);
-  AddMockFile(file2);
-  SetLastSequence(last_seq);
-  files = cfd_->current()->storage_info()->LevelFiles(0);
-  RunCompaction({files}, expected, db_snapshots, kMaxSequenceNumber,
-                output_level_0, !kVerify);
-  // The 2nd compaction above would get rid of useless delete markers. To get
-  // the output here exactly as what we got above after two compactions, we also
-  // run the compaction for 2nd time.
-  compaction_job_stats_.Reset();
-  files = cfd_->current()->storage_info()->LevelFiles(0);
-  RunCompaction({files}, expected, db_snapshots, kMaxSequenceNumber,
-                output_level_0 + 1, !kVerify);
-}
-
 }  // namespace rocksdb

 int main(int argc, char** argv) {
@@ -18,7 +18,6 @@
 #include <queue>
 #include <string>
 #include <utility>
-#include <vector>
 #include "db/column_family.h"
 #include "monitoring/statistics.h"
 #include "util/filename.h"
@@ -37,7 +36,6 @@ uint64_t TotalCompensatedFileSize(const std::vector<FileMetaData*>& files) {
  }
  return sum;
 }
-}  // anonymous namespace

 bool FindIntraL0Compaction(const std::vector<FileMetaData*>& level_files,
                           size_t min_files_to_compact,
@@ -51,7 +49,7 @@ bool FindIntraL0Compaction(const std::vector<FileMetaData*>& level_files,
  // increasing.
  size_t new_compact_bytes_per_del_file = 0;
  for (span_len = 1; span_len < level_files.size(); ++span_len) {
-    compact_bytes += static_cast<size_t>(level_files[span_len]->fd.file_size);
+    compact_bytes += level_files[span_len]->fd.file_size;
    new_compact_bytes_per_del_file = compact_bytes / span_len;
    if (level_files[span_len]->being_compacted ||
        new_compact_bytes_per_del_file > compact_bytes_per_del_file) {
@@ -71,6 +69,7 @@ bool FindIntraL0Compaction(const std::vector<FileMetaData*>& level_files,
  }
  return false;
 }
+}  // anonymous namespace

 // Determine compression type, based on user options, level of the output
 // file and whether compression is disabled.
@@ -543,9 +542,9 @@ void CompactionPicker::GetGrandparents(
 Compaction* CompactionPicker::CompactRange(
    const std::string& cf_name, const MutableCFOptions& mutable_cf_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) {
+    uint32_t output_path_id, uint32_t max_subcompactions,
+    const InternalKey* begin, const InternalKey* end,
+    InternalKey** compaction_end, bool* manual_conflict) {
  // CompactionPickerFIFO has its own implementation of compact range
  assert(ioptions_.compaction_style != kCompactionStyleFIFO);

@@ -609,13 +608,11 @@ Compaction* CompactionPicker::CompactRange(
        output_level,
        MaxFileSizeForLevel(mutable_cf_options, output_level,
                            ioptions_.compaction_style),
-        /* max_compaction_bytes */ LLONG_MAX,
-        compact_range_options.target_path_id,
+        /* max_compaction_bytes */ LLONG_MAX, output_path_id,
        GetCompressionType(ioptions_, vstorage, mutable_cf_options,
                           output_level, 1),
        GetCompressionOptions(ioptions_, vstorage, output_level),
-        compact_range_options.max_subcompactions, /* grandparents */ {},
-        /* is manual */ true);
+        max_subcompactions, /* grandparents */ {}, /* is manual */ true);
    RegisterCompaction(c);
    return c;
  }
@@ -660,43 +657,7 @@ Compaction* CompactionPicker::CompactRange(
      }
    }
  }
-  assert(compact_range_options.target_path_id <
-         static_cast<uint32_t>(ioptions_.cf_paths.size()));
-
-  // for BOTTOM LEVEL compaction only, use max_file_num_to_ignore to filter out
-  // files that are created during the current compaction.
-  if (compact_range_options.bottommost_level_compaction ==
-          BottommostLevelCompaction::kForceOptimized &&
-      max_file_num_to_ignore != port::kMaxUint64) {
-    assert(input_level == output_level);
-    // inputs_shrunk holds a continuous subset of input files which were all
-    // created before the current manual compaction
-    std::vector<FileMetaData*> inputs_shrunk;
-    size_t skip_input_index = inputs.size();
-    for (size_t i = 0; i < inputs.size(); ++i) {
-      if (inputs[i]->fd.GetNumber() < max_file_num_to_ignore) {
-        inputs_shrunk.push_back(inputs[i]);
-      } else if (!inputs_shrunk.empty()) {
-        // inputs[i] was created during the current manual compaction and
-        // need to be skipped
-        skip_input_index = i;
-        break;
-      }
-    }
-    if (inputs_shrunk.empty()) {
-      return nullptr;
-    }
-    if (inputs.size() != inputs_shrunk.size()) {
-      inputs.files.swap(inputs_shrunk);
-    }
-    // set covering_the_whole_range to false if there is any file that need to
-    // be compacted in the range of inputs[skip_input_index+1, inputs.size())
-    for (size_t i = skip_input_index + 1; i < inputs.size(); ++i) {
-      if (inputs[i]->fd.GetNumber() < max_file_num_to_ignore) {
-        covering_the_whole_range = false;
-      }
-    }
-  }
+  assert(output_path_id < static_cast<uint32_t>(ioptions_.cf_paths.size()));

  InternalKey key_storage;
  InternalKey* next_smallest = &key_storage;
@@ -762,12 +723,11 @@ Compaction* CompactionPicker::CompactRange(
      MaxFileSizeForLevel(mutable_cf_options, output_level,
                          ioptions_.compaction_style, vstorage->base_level(),
                          ioptions_.level_compaction_dynamic_level_bytes),
-      mutable_cf_options.max_compaction_bytes,
-      compact_range_options.target_path_id,
+      mutable_cf_options.max_compaction_bytes, output_path_id,
      GetCompressionType(ioptions_, vstorage, mutable_cf_options, output_level,
                         vstorage->base_level()),
      GetCompressionOptions(ioptions_, vstorage, output_level),
-      compact_range_options.max_subcompactions, std::move(grandparents),
+      max_subcompactions, std::move(grandparents),
      /* is manual compaction */ true);

  TEST_SYNC_POINT_CALLBACK("CompactionPicker::CompactRange:Return", compaction);
@@ -983,8 +943,8 @@ Status CompactionPicker::SanitizeCompactionInputFiles(
  // any currently-existing files.
  for (auto file_num : *input_files) {
    bool found = false;
-    for (const auto& level_meta : cf_meta.levels) {
-      for (const auto& file_meta : level_meta.files) {
+    for (auto level_meta : cf_meta.levels) {
+      for (auto file_meta : level_meta.files) {
        if (file_num == TableFileNameToNumber(file_meta.name)) {
          if (file_meta.being_compacted) {
            return Status::Aborted("Specified compaction input file " +
@@ -1112,9 +1072,6 @@ bool LevelCompactionPicker::NeedsCompaction(
  if (!vstorage->ExpiredTtlFiles().empty()) {
    return true;
  }
-  if (!vstorage->FilesMarkedForPeriodicCompaction().empty()) {
-    return true;
-  }
  if (!vstorage->BottommostFilesMarkedForCompaction().empty()) {
    return true;
  }
@@ -1183,8 +1140,6 @@ class LevelCompactionBuilder {

  void PickExpiredTtlFiles();

-  void PickFilesMarkedForPeriodicCompaction();
-
  const std::string& cf_name_;
  VersionStorageInfo* vstorage_;
  CompactionPicker* compaction_picker_;
@@ -1247,39 +1202,6 @@ void LevelCompactionBuilder::PickExpiredTtlFiles() {
  start_level_inputs_.files.clear();
 }

-void LevelCompactionBuilder::PickFilesMarkedForPeriodicCompaction() {
-  if (vstorage_->FilesMarkedForPeriodicCompaction().empty()) {
-    return;
-  }
-
-  auto continuation = [&](std::pair<int, FileMetaData*> level_file) {
-    // If it's being compacted it has nothing to do here.
-    // If this assert() fails that means that some function marked some
-    // files as being_compacted, but didn't call ComputeCompactionScore()
-    assert(!level_file.second->being_compacted);
-    output_level_ = start_level_ = level_file.first;
-
-    if (start_level_ == 0 &&
-        !compaction_picker_->level0_compactions_in_progress()->empty()) {
-      return false;
-    }
-
-    start_level_inputs_.files = {level_file.second};
-    start_level_inputs_.level = start_level_;
-    return compaction_picker_->ExpandInputsToCleanCut(cf_name_, vstorage_,
-                                                      &start_level_inputs_);
-  };
-
-  for (auto& level_file : vstorage_->FilesMarkedForPeriodicCompaction()) {
-    if (continuation(level_file)) {
-      // found the compaction!
-      return;
-    }
-  }
-
-  start_level_inputs_.files.clear();
-}
-
 void LevelCompactionBuilder::SetupInitialFiles() {
  // Find the compactions by size on all levels.
  bool skipped_l0_to_base = false;
@@ -1333,6 +1255,7 @@ void LevelCompactionBuilder::SetupInitialFiles() {
  if (start_level_inputs_.empty()) {
    parent_index_ = base_index_ = -1;

+    // PickFilesMarkedForCompaction();
    compaction_picker_->PickFilesMarkedForCompaction(
        cf_name_, vstorage_, &start_level_, &output_level_, &start_level_inputs_);
    if (!start_level_inputs_.empty()) {
@@ -1340,10 +1263,7 @@ void LevelCompactionBuilder::SetupInitialFiles() {
      compaction_reason_ = CompactionReason::kFilesMarkedForCompaction;
      return;
    }
-  }

-  // Bottommost Files Compaction on deleting tombstones
-  if (start_level_inputs_.empty()) {
    size_t i;
    for (i = 0; i < vstorage_->BottommostFilesMarkedForCompaction().size();
         ++i) {
@@ -1364,23 +1284,11 @@ void LevelCompactionBuilder::SetupInitialFiles() {
      compaction_reason_ = CompactionReason::kBottommostFiles;
      return;
    }
-  }

-  // TTL Compaction
-  if (start_level_inputs_.empty()) {
+    assert(start_level_inputs_.empty());
    PickExpiredTtlFiles();
    if (!start_level_inputs_.empty()) {
      compaction_reason_ = CompactionReason::kTtl;
-      return;
-    }
-  }
-
-  // Periodic Compaction
-  if (start_level_inputs_.empty()) {
-    PickFilesMarkedForPeriodicCompaction();
-    if (!start_level_inputs_.empty()) {
-      compaction_reason_ = CompactionReason::kPeriodicCompaction;
-      return;
    }
  }
 }
@@ -1639,4 +1547,217 @@ Compaction* LevelCompactionPicker::PickCompaction(
  return builder.PickCompaction();
 }

+#ifndef ROCKSDB_LITE
+bool FIFOCompactionPicker::NeedsCompaction(
+    const VersionStorageInfo* vstorage) const {
+  const int kLevel0 = 0;
+  return vstorage->CompactionScore(kLevel0) >= 1;
+}
+
+namespace {
+uint64_t GetTotalFilesSize(
+    const std::vector<FileMetaData*>& files) {
+  uint64_t total_size = 0;
+  for (const auto& f : files) {
+    total_size += f->fd.file_size;
+  }
+  return total_size;
+}
+}  // anonymous namespace
+
+Compaction* FIFOCompactionPicker::PickTTLCompaction(
+    const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
+    VersionStorageInfo* vstorage, LogBuffer* log_buffer) {
+  assert(mutable_cf_options.compaction_options_fifo.ttl > 0);
+
+  const int kLevel0 = 0;
+  const std::vector<FileMetaData*>& level_files = vstorage->LevelFiles(kLevel0);
+  uint64_t total_size = GetTotalFilesSize(level_files);
+
+  int64_t _current_time;
+  auto status = ioptions_.env->GetCurrentTime(&_current_time);
+  if (!status.ok()) {
+    ROCKS_LOG_BUFFER(log_buffer,
+                     "[%s] FIFO compaction: Couldn't get current time: %s. "
+                     "Not doing compactions based on TTL. ",
+                     cf_name.c_str(), status.ToString().c_str());
+    return nullptr;
+  }
+  const uint64_t current_time = static_cast<uint64_t>(_current_time);
+
+  std::vector<CompactionInputFiles> inputs;
+  inputs.emplace_back();
+  inputs[0].level = 0;
+
+  // avoid underflow
+  if (current_time > mutable_cf_options.compaction_options_fifo.ttl) {
+    for (auto ritr = level_files.rbegin(); ritr != level_files.rend(); ++ritr) {
+      auto f = *ritr;
+      if (f->fd.table_reader != nullptr &&
+          f->fd.table_reader->GetTableProperties() != nullptr) {
+        auto creation_time =
+            f->fd.table_reader->GetTableProperties()->creation_time;
+        if (creation_time == 0 ||
+            creation_time >= (current_time -
+                              mutable_cf_options.compaction_options_fifo.ttl)) {
+          break;
+        }
+        total_size -= f->compensated_file_size;
+        inputs[0].files.push_back(f);
+      }
+    }
+  }
+
+  // Return a nullptr and proceed to size-based FIFO compaction if:
+  // 1. there are no files older than ttl OR
+  // 2. there are a few files older than ttl, but deleting them will not bring
+  //    the total size to be less than max_table_files_size threshold.
+  if (inputs[0].files.empty() ||
+      total_size >
+          mutable_cf_options.compaction_options_fifo.max_table_files_size) {
+    return nullptr;
+  }
+
+  for (const auto& f : inputs[0].files) {
+    ROCKS_LOG_BUFFER(log_buffer,
+                     "[%s] FIFO compaction: picking file %" PRIu64
+                     " with creation time %" PRIu64 " for deletion",
+                     cf_name.c_str(), f->fd.GetNumber(),
+                     f->fd.table_reader->GetTableProperties()->creation_time);
+  }
+
+  Compaction* c = new Compaction(
+      vstorage, ioptions_, mutable_cf_options, std::move(inputs), 0, 0, 0, 0,
+      kNoCompression, ioptions_.compression_opts, /* max_subcompactions */ 0,
+      {}, /* is manual */ false, vstorage->CompactionScore(0),
+      /* is deletion compaction */ true, CompactionReason::kFIFOTtl);
+  return c;
+}
+
+Compaction* FIFOCompactionPicker::PickSizeCompaction(
+    const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
+    VersionStorageInfo* vstorage, LogBuffer* log_buffer) {
+  const int kLevel0 = 0;
+  const std::vector<FileMetaData*>& level_files = vstorage->LevelFiles(kLevel0);
+  uint64_t total_size = GetTotalFilesSize(level_files);
+
+  if (total_size <=
+          mutable_cf_options.compaction_options_fifo.max_table_files_size ||
+      level_files.size() == 0) {
+    // total size not exceeded
+    if (mutable_cf_options.compaction_options_fifo.allow_compaction &&
+        level_files.size() > 0) {
+      CompactionInputFiles comp_inputs;
+      // try to prevent same files from being compacted multiple times, which
+      // could produce large files that may never TTL-expire. Achieve this by
+      // disallowing compactions with files larger than memtable (inflate its
+      // size by 10% to account for uncompressed L0 files that may have size
+      // slightly greater than memtable size limit).
+      size_t max_compact_bytes_per_del_file =
+          static_cast<size_t>(MultiplyCheckOverflow(
+              static_cast<uint64_t>(mutable_cf_options.write_buffer_size),
+              1.1));
+      if (FindIntraL0Compaction(
+              level_files,
+              mutable_cf_options
+                  .level0_file_num_compaction_trigger /* min_files_to_compact */
+              ,
+              max_compact_bytes_per_del_file, &comp_inputs)) {
+        Compaction* c = new Compaction(
+            vstorage, ioptions_, mutable_cf_options, {comp_inputs}, 0,
+            16 * 1024 * 1024 /* output file size limit */,
+            0 /* max compaction bytes, not applicable */,
+            0 /* output path ID */, mutable_cf_options.compression,
+            ioptions_.compression_opts, 0 /* max_subcompactions */, {},
+            /* is manual */ false, vstorage->CompactionScore(0),
+            /* is deletion compaction */ false,
+            CompactionReason::kFIFOReduceNumFiles);
+        return c;
+      }
+    }
+
+    ROCKS_LOG_BUFFER(
+        log_buffer,
+        "[%s] FIFO compaction: nothing to do. Total size %" PRIu64
+        ", max size %" PRIu64 "\n",
+        cf_name.c_str(), total_size,
+        mutable_cf_options.compaction_options_fifo.max_table_files_size);
+    return nullptr;
+  }
+
+  if (!level0_compactions_in_progress_.empty()) {
+    ROCKS_LOG_BUFFER(
+        log_buffer,
+        "[%s] FIFO compaction: Already executing compaction. No need "
+        "to run parallel compactions since compactions are very fast",
+        cf_name.c_str());
+    return nullptr;
+  }
+
+  std::vector<CompactionInputFiles> inputs;
+  inputs.emplace_back();
+  inputs[0].level = 0;
+
+  for (auto ritr = level_files.rbegin(); ritr != level_files.rend(); ++ritr) {
+    auto f = *ritr;
+    total_size -= f->compensated_file_size;
+    inputs[0].files.push_back(f);
+    char tmp_fsize[16];
+    AppendHumanBytes(f->fd.GetFileSize(), tmp_fsize, sizeof(tmp_fsize));
+    ROCKS_LOG_BUFFER(log_buffer,
+                     "[%s] FIFO compaction: picking file %" PRIu64
+                     " with size %s for deletion",
+                     cf_name.c_str(), f->fd.GetNumber(), tmp_fsize);
+    if (total_size <=
+        mutable_cf_options.compaction_options_fifo.max_table_files_size) {
+      break;
+    }
+  }
+
+  Compaction* c = new Compaction(
+      vstorage, ioptions_, mutable_cf_options, std::move(inputs), 0, 0, 0, 0,
+      kNoCompression, ioptions_.compression_opts, /* max_subcompactions */ 0,
+      {}, /* is manual */ false, vstorage->CompactionScore(0),
+      /* is deletion compaction */ true, CompactionReason::kFIFOMaxSize);
+  return c;
+}
+
+Compaction* FIFOCompactionPicker::PickCompaction(
+    const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
+    VersionStorageInfo* vstorage, LogBuffer* log_buffer) {
+  assert(vstorage->num_levels() == 1);
+
+  Compaction* c = nullptr;
+  if (mutable_cf_options.compaction_options_fifo.ttl > 0) {
+    c = PickTTLCompaction(cf_name, mutable_cf_options, vstorage, log_buffer);
+  }
+  if (c == nullptr) {
+    c = PickSizeCompaction(cf_name, mutable_cf_options, vstorage, log_buffer);
+  }
+  RegisterCompaction(c);
+  return c;
+}
+
+Compaction* FIFOCompactionPicker::CompactRange(
+    const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
+    VersionStorageInfo* vstorage, int input_level, int output_level,
+    uint32_t /*output_path_id*/, uint32_t /*max_subcompactions*/,
+    const InternalKey* /*begin*/, const InternalKey* /*end*/,
+    InternalKey** compaction_end, bool* /*manual_conflict*/) {
+#ifdef NDEBUG
+  (void)input_level;
+  (void)output_level;
+#endif
+  assert(input_level == 0);
+  assert(output_level == 0);
+  *compaction_end = nullptr;
+  LogBuffer log_buffer(InfoLogLevel::INFO_LEVEL, ioptions_.info_log);
+  Compaction* c =
+      PickCompaction(cf_name, mutable_cf_options, vstorage, &log_buffer);
+  log_buffer.FlushBufferToLog();
+  return c;
+}
+
+#endif  // !ROCKSDB_LITE
+
 }  // namespace rocksdb
@@ -58,10 +58,9 @@ class CompactionPicker {
  virtual Compaction* CompactRange(
      const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
      VersionStorageInfo* vstorage, int input_level, int output_level,
-      const CompactRangeOptions& compact_range_options,
+      uint32_t output_path_id, uint32_t max_subcompactions,
      const InternalKey* begin, const InternalKey* end,
-      InternalKey** compaction_end, bool* manual_conflict,
-      uint64_t max_file_num_to_ignore);
+      InternalKey** compaction_end, bool* manual_conflict);

  // The maximum allowed output level.  Default value is NumberLevels() - 1.
  virtual int MaxOutputLevel() const { return NumberLevels() - 1; }
@@ -236,6 +235,42 @@ class LevelCompactionPicker : public CompactionPicker {
 };

 #ifndef ROCKSDB_LITE
+class FIFOCompactionPicker : public CompactionPicker {
+ public:
+  FIFOCompactionPicker(const ImmutableCFOptions& ioptions,
+                       const InternalKeyComparator* icmp)
+      : CompactionPicker(ioptions, icmp) {}
+
+  virtual Compaction* PickCompaction(const std::string& cf_name,
+                                     const MutableCFOptions& mutable_cf_options,
+                                     VersionStorageInfo* version,
+                                     LogBuffer* log_buffer) override;
+
+  virtual Compaction* CompactRange(
+      const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
+      VersionStorageInfo* vstorage, int input_level, int output_level,
+      uint32_t output_path_id, uint32_t max_subcompactions,
+      const InternalKey* begin, const InternalKey* end,
+      InternalKey** compaction_end, bool* manual_conflict) override;
+
+  // The maximum allowed output level.  Always returns 0.
+  virtual int MaxOutputLevel() const override { return 0; }
+
+  virtual bool NeedsCompaction(
+      const VersionStorageInfo* vstorage) const override;
+
+ private:
+  Compaction* PickTTLCompaction(const std::string& cf_name,
+                                const MutableCFOptions& mutable_cf_options,
+                                VersionStorageInfo* version,
+                                LogBuffer* log_buffer);
+
+  Compaction* PickSizeCompaction(const std::string& cf_name,
+                                 const MutableCFOptions& mutable_cf_options,
+                                 VersionStorageInfo* version,
+                                 LogBuffer* log_buffer);
+};
+
 class NullCompactionPicker : public CompactionPicker {
 public:
  NullCompactionPicker(const ImmutableCFOptions& ioptions,
@@ -256,12 +291,12 @@ class NullCompactionPicker : public CompactionPicker {
                           const MutableCFOptions& /*mutable_cf_options*/,
                           VersionStorageInfo* /*vstorage*/,
                           int /*input_level*/, int /*output_level*/,
-                           const CompactRangeOptions& /*compact_range_options*/,
+                           uint32_t /*output_path_id*/,
+                           uint32_t /*max_subcompactions*/,
                           const InternalKey* /*begin*/,
                           const InternalKey* /*end*/,
                           InternalKey** /*compaction_end*/,
-                           bool* /*manual_conflict*/,
-                           uint64_t /*max_file_num_to_ignore*/) override {
+                           bool* /*manual_conflict*/) override {
    return nullptr;
  }

@@ -273,11 +308,6 @@ class NullCompactionPicker : public CompactionPicker {
 };
 #endif  // !ROCKSDB_LITE

-bool FindIntraL0Compaction(const std::vector<FileMetaData*>& level_files,
-                           size_t min_files_to_compact,
-                           uint64_t max_compact_bytes_per_del_file,
-                           CompactionInputFiles* comp_inputs);
-
 CompressionType GetCompressionType(const ImmutableCFOptions& ioptions,
                                   const VersionStorageInfo* vstorage,
                                   const MutableCFOptions& mutable_cf_options,
@@ -1,235 +0,0 @@
-//  Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
-//  This source code is licensed under both the GPLv2 (found in the
-//  COPYING file in the root directory) and Apache 2.0 License
-//  (found in the LICENSE.Apache file in the root directory).
-//
-// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
-// Use of this source code is governed by a BSD-style license that can be
-// found in the LICENSE file. See the AUTHORS file for names of contributors.
-
-#include "db/compaction_picker_fifo.h"
-#ifndef ROCKSDB_LITE
-
-#ifndef __STDC_FORMAT_MACROS
-#define __STDC_FORMAT_MACROS
-#endif
-
-#include <inttypes.h>
-#include <string>
-#include <vector>
-#include "db/column_family.h"
-#include "util/log_buffer.h"
-#include "util/string_util.h"
-
-namespace rocksdb {
-namespace {
-uint64_t GetTotalFilesSize(const std::vector<FileMetaData*>& files) {
-  uint64_t total_size = 0;
-  for (const auto& f : files) {
-    total_size += f->fd.file_size;
-  }
-  return total_size;
-}
-}  // anonymous namespace
-
-bool FIFOCompactionPicker::NeedsCompaction(
-    const VersionStorageInfo* vstorage) const {
-  const int kLevel0 = 0;
-  return vstorage->CompactionScore(kLevel0) >= 1;
-}
-
-Compaction* FIFOCompactionPicker::PickTTLCompaction(
-    const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
-    VersionStorageInfo* vstorage, LogBuffer* log_buffer) {
-  assert(mutable_cf_options.ttl > 0);
-
-  const int kLevel0 = 0;
-  const std::vector<FileMetaData*>& level_files = vstorage->LevelFiles(kLevel0);
-  uint64_t total_size = GetTotalFilesSize(level_files);
-
-  int64_t _current_time;
-  auto status = ioptions_.env->GetCurrentTime(&_current_time);
-  if (!status.ok()) {
-    ROCKS_LOG_BUFFER(log_buffer,
-                     "[%s] FIFO compaction: Couldn't get current time: %s. "
-                     "Not doing compactions based on TTL. ",
-                     cf_name.c_str(), status.ToString().c_str());
-    return nullptr;
-  }
-  const uint64_t current_time = static_cast<uint64_t>(_current_time);
-
-  std::vector<CompactionInputFiles> inputs;
-  inputs.emplace_back();
-  inputs[0].level = 0;
-
-  // avoid underflow
-  if (current_time > mutable_cf_options.ttl) {
-    for (auto ritr = level_files.rbegin(); ritr != level_files.rend(); ++ritr) {
-      auto f = *ritr;
-      if (f->fd.table_reader != nullptr &&
-          f->fd.table_reader->GetTableProperties() != nullptr) {
-        auto creation_time =
-            f->fd.table_reader->GetTableProperties()->creation_time;
-        if (creation_time == 0 ||
-            creation_time >= (current_time - mutable_cf_options.ttl)) {
-          break;
-        }
-        total_size -= f->compensated_file_size;
-        inputs[0].files.push_back(f);
-      }
-    }
-  }
-
-  // Return a nullptr and proceed to size-based FIFO compaction if:
-  // 1. there are no files older than ttl OR
-  // 2. there are a few files older than ttl, but deleting them will not bring
-  //    the total size to be less than max_table_files_size threshold.
-  if (inputs[0].files.empty() ||
-      total_size >
-          mutable_cf_options.compaction_options_fifo.max_table_files_size) {
-    return nullptr;
-  }
-
-  for (const auto& f : inputs[0].files) {
-    ROCKS_LOG_BUFFER(log_buffer,
-                     "[%s] FIFO compaction: picking file %" PRIu64
-                     " with creation time %" PRIu64 " for deletion",
-                     cf_name.c_str(), f->fd.GetNumber(),
-                     f->fd.table_reader->GetTableProperties()->creation_time);
-  }
-
-  Compaction* c = new Compaction(
-      vstorage, ioptions_, mutable_cf_options, std::move(inputs), 0, 0, 0, 0,
-      kNoCompression, ioptions_.compression_opts, /* max_subcompactions */ 0,
-      {}, /* is manual */ false, vstorage->CompactionScore(0),
-      /* is deletion compaction */ true, CompactionReason::kFIFOTtl);
-  return c;
-}
-
-Compaction* FIFOCompactionPicker::PickSizeCompaction(
-    const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
-    VersionStorageInfo* vstorage, LogBuffer* log_buffer) {
-  const int kLevel0 = 0;
-  const std::vector<FileMetaData*>& level_files = vstorage->LevelFiles(kLevel0);
-  uint64_t total_size = GetTotalFilesSize(level_files);
-
-  if (total_size <=
-          mutable_cf_options.compaction_options_fifo.max_table_files_size ||
-      level_files.size() == 0) {
-    // total size not exceeded
-    if (mutable_cf_options.compaction_options_fifo.allow_compaction &&
-        level_files.size() > 0) {
-      CompactionInputFiles comp_inputs;
-      // try to prevent same files from being compacted multiple times, which
-      // could produce large files that may never TTL-expire. Achieve this by
-      // disallowing compactions with files larger than memtable (inflate its
-      // size by 10% to account for uncompressed L0 files that may have size
-      // slightly greater than memtable size limit).
-      size_t max_compact_bytes_per_del_file =
-          static_cast<size_t>(MultiplyCheckOverflow(
-              static_cast<uint64_t>(mutable_cf_options.write_buffer_size),
-              1.1));
-      if (FindIntraL0Compaction(
-              level_files,
-              mutable_cf_options
-                  .level0_file_num_compaction_trigger /* min_files_to_compact */
-              ,
-              max_compact_bytes_per_del_file, &comp_inputs)) {
-        Compaction* c = new Compaction(
-            vstorage, ioptions_, mutable_cf_options, {comp_inputs}, 0,
-            16 * 1024 * 1024 /* output file size limit */,
-            0 /* max compaction bytes, not applicable */,
-            0 /* output path ID */, mutable_cf_options.compression,
-            ioptions_.compression_opts, 0 /* max_subcompactions */, {},
-            /* is manual */ false, vstorage->CompactionScore(0),
-            /* is deletion compaction */ false,
-            CompactionReason::kFIFOReduceNumFiles);
-        return c;
-      }
-    }
-
-    ROCKS_LOG_BUFFER(
-        log_buffer,
-        "[%s] FIFO compaction: nothing to do. Total size %" PRIu64
-        ", max size %" PRIu64 "\n",
-        cf_name.c_str(), total_size,
-        mutable_cf_options.compaction_options_fifo.max_table_files_size);
-    return nullptr;
-  }
-
-  if (!level0_compactions_in_progress_.empty()) {
-    ROCKS_LOG_BUFFER(
-        log_buffer,
-        "[%s] FIFO compaction: Already executing compaction. No need "
-        "to run parallel compactions since compactions are very fast",
-        cf_name.c_str());
-    return nullptr;
-  }
-
-  std::vector<CompactionInputFiles> inputs;
-  inputs.emplace_back();
-  inputs[0].level = 0;
-
-  for (auto ritr = level_files.rbegin(); ritr != level_files.rend(); ++ritr) {
-    auto f = *ritr;
-    total_size -= f->compensated_file_size;
-    inputs[0].files.push_back(f);
-    char tmp_fsize[16];
-    AppendHumanBytes(f->fd.GetFileSize(), tmp_fsize, sizeof(tmp_fsize));
-    ROCKS_LOG_BUFFER(log_buffer,
-                     "[%s] FIFO compaction: picking file %" PRIu64
-                     " with size %s for deletion",
-                     cf_name.c_str(), f->fd.GetNumber(), tmp_fsize);
-    if (total_size <=
-        mutable_cf_options.compaction_options_fifo.max_table_files_size) {
-      break;
-    }
-  }
-
-  Compaction* c = new Compaction(
-      vstorage, ioptions_, mutable_cf_options, std::move(inputs), 0, 0, 0, 0,
-      kNoCompression, ioptions_.compression_opts, /* max_subcompactions */ 0,
-      {}, /* is manual */ false, vstorage->CompactionScore(0),
-      /* is deletion compaction */ true, CompactionReason::kFIFOMaxSize);
-  return c;
-}
-
-Compaction* FIFOCompactionPicker::PickCompaction(
-    const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
-    VersionStorageInfo* vstorage, LogBuffer* log_buffer) {
-  assert(vstorage->num_levels() == 1);
-
-  Compaction* c = nullptr;
-  if (mutable_cf_options.ttl > 0) {
-    c = PickTTLCompaction(cf_name, mutable_cf_options, vstorage, log_buffer);
-  }
-  if (c == nullptr) {
-    c = PickSizeCompaction(cf_name, mutable_cf_options, vstorage, log_buffer);
-  }
-  RegisterCompaction(c);
-  return c;
-}
-
-Compaction* FIFOCompactionPicker::CompactRange(
-    const std::string& cf_name, const MutableCFOptions& mutable_cf_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*/) {
-#ifdef NDEBUG
-  (void)input_level;
-  (void)output_level;
-#endif
-  assert(input_level == 0);
-  assert(output_level == 0);
-  *compaction_end = nullptr;
-  LogBuffer log_buffer(InfoLogLevel::INFO_LEVEL, ioptions_.info_log);
-  Compaction* c =
-      PickCompaction(cf_name, mutable_cf_options, vstorage, &log_buffer);
-  log_buffer.FlushBufferToLog();
-  return c;
-}
-
-}  // namespace rocksdb
-#endif  // !ROCKSDB_LITE
@@ -1,53 +0,0 @@
-//  Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
-//  This source code is licensed under both the GPLv2 (found in the
-//  COPYING file in the root directory) and Apache 2.0 License
-//  (found in the LICENSE.Apache file in the root directory).
-//
-// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
-// Use of this source code is governed by a BSD-style license that can be
-// found in the LICENSE file. See the AUTHORS file for names of contributors.
-
-#pragma once
-#ifndef ROCKSDB_LITE
-
-#include "db/compaction_picker.h"
-
-namespace rocksdb {
-class FIFOCompactionPicker : public CompactionPicker {
- public:
-  FIFOCompactionPicker(const ImmutableCFOptions& ioptions,
-                       const InternalKeyComparator* icmp)
-      : CompactionPicker(ioptions, icmp) {}
-
-  virtual Compaction* PickCompaction(const std::string& cf_name,
-                                     const MutableCFOptions& mutable_cf_options,
-                                     VersionStorageInfo* version,
-                                     LogBuffer* log_buffer) override;
-
-  virtual Compaction* CompactRange(
-      const std::string& cf_name, const MutableCFOptions& mutable_cf_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) override;
-
-  // The maximum allowed output level.  Always returns 0.
-  virtual int MaxOutputLevel() const override { return 0; }
-
-  virtual bool NeedsCompaction(
-      const VersionStorageInfo* vstorage) const override;
-
- private:
-  Compaction* PickTTLCompaction(const std::string& cf_name,
-                                const MutableCFOptions& mutable_cf_options,
-                                VersionStorageInfo* version,
-                                LogBuffer* log_buffer);
-
-  Compaction* PickSizeCompaction(const std::string& cf_name,
-                                 const MutableCFOptions& mutable_cf_options,
-                                 VersionStorageInfo* version,
-                                 LogBuffer* log_buffer);
-};
-}  // namespace rocksdb
-#endif  // !ROCKSDB_LITE
@@ -4,12 +4,10 @@
 //  (found in the LICENSE.Apache file in the root directory).

 #include "db/compaction_picker.h"
-
 #include <limits>
 #include <string>
 #include <utility>
 #include "db/compaction.h"
-#include "db/compaction_picker_fifo.h"
 #include "db/compaction_picker_universal.h"

 #include "util/logging.h"
@@ -22,7 +20,9 @@ namespace rocksdb {
 class CountingLogger : public Logger {
 public:
  using Logger::Logv;
-  void Logv(const char* /*format*/, va_list /*ap*/) override { log_count++; }
+  virtual void Logv(const char* /*format*/, va_list /*ap*/) override {
+    log_count++;
+  }
  size_t log_count;
 };

@@ -57,16 +57,14 @@ class CompactionPickerTest : public testing::Test {
        log_buffer_(InfoLogLevel::INFO_LEVEL, &logger_),
        file_num_(1),
        vstorage_(nullptr) {
-    // ioptions_.compaction_pri = kMinOverlappingRatio has its own set of
-    // tests to cover.
-    ioptions_.compaction_pri = kByCompensatedSize;
    fifo_options_.max_table_files_size = 1;
    mutable_cf_options_.RefreshDerivedOptions(ioptions_);
    ioptions_.cf_paths.emplace_back("dummy",
                                    std::numeric_limits<uint64_t>::max());
  }

-  ~CompactionPickerTest() override {}
+  ~CompactionPickerTest() {
+  }

  void NewVersionStorage(int num_levels, CompactionStyle style) {
    DeleteVersionStorage();
@@ -85,8 +83,8 @@ class CompactionPickerTest : public testing::Test {

  void Add(int level, uint32_t file_number, const char* smallest,
           const char* largest, uint64_t file_size = 1, uint32_t path_id = 0,
-           SequenceNumber smallest_seq = 100, SequenceNumber largest_seq = 100,
-           size_t compensated_file_size = 0) {
+           SequenceNumber smallest_seq = 100,
+           SequenceNumber largest_seq = 100) {
    assert(level < vstorage_->num_levels());
    FileMetaData* f = new FileMetaData;
    f->fd = FileDescriptor(file_number, path_id, file_size);
@@ -94,8 +92,7 @@ class CompactionPickerTest : public testing::Test {
    f->largest = InternalKey(largest, largest_seq, kTypeValue);
    f->fd.smallest_seqno = smallest_seq;
    f->fd.largest_seqno = largest_seq;
-    f->compensated_file_size =
-        (compensated_file_size != 0) ? compensated_file_size : file_size;
+    f->compensated_file_size = file_size;
    f->refs = 0;
    vstorage_->AddFile(level, f);
    files_.emplace_back(f);
@@ -618,35 +615,6 @@ TEST_F(CompactionPickerTest, CompactionPriMinOverlapping3) {
  ASSERT_EQ(8U, compaction->input(0, 0)->fd.GetNumber());
 }

-TEST_F(CompactionPickerTest, CompactionPriMinOverlapping4) {
-  NewVersionStorage(6, kCompactionStyleLevel);
-  ioptions_.compaction_pri = kMinOverlappingRatio;
-  mutable_cf_options_.max_bytes_for_level_base = 10000000;
-  mutable_cf_options_.max_bytes_for_level_multiplier = 10;
-
-  // file 7 and 8 over lap with the same file, but file 8 is smaller so
-  // it will be picked.
-  // Overlaps with file 26, 27. And the file is compensated so will be
-  // picked up.
-  Add(2, 6U, "150", "167", 60000000U, 0, 100, 100, 180000000U);
-  Add(2, 7U, "168", "169", 60000000U);  // Overlaps with file 27
-  Add(2, 8U, "201", "300", 61000000U);  // Overlaps with file 28
-
-  Add(3, 26U, "160", "165", 60000000U);
-  // Boosted file size in output level is not considered.
-  Add(3, 27U, "166", "170", 60000000U, 0, 100, 100, 260000000U);
-  Add(3, 28U, "180", "400", 60000000U);
-  Add(3, 29U, "401", "500", 60000000U);
-  UpdateVersionStorageInfo();
-
-  std::unique_ptr<Compaction> compaction(level_compaction_picker.PickCompaction(
-      cf_name_, mutable_cf_options_, vstorage_.get(), &log_buffer_));
-  ASSERT_TRUE(compaction.get() != nullptr);
-  ASSERT_EQ(1U, compaction->num_input_files(0));
-  // Picking file 8 because overlapping ratio is the biggest.
-  ASSERT_EQ(6U, compaction->input(0, 0)->fd.GetNumber());
-}
-
 // This test exhibits the bug where we don't properly reset parent_index in
 // PickCompaction()
 TEST_F(CompactionPickerTest, ParentIndexResetBug) {
@@ -395,8 +395,8 @@ Compaction* UniversalCompactionPicker::PickCompaction(
  }
 #endif
  // update statistics
-  RecordInHistogram(ioptions_.statistics, NUM_FILES_IN_SINGLE_COMPACTION,
-                    c->inputs(0)->size());
+  MeasureTime(ioptions_.statistics, NUM_FILES_IN_SINGLE_COMPACTION,
+              c->inputs(0)->size());

  RegisterCompaction(c);
  vstorage->ComputeCompactionScore(ioptions_, mutable_cf_options);
@@ -27,24 +27,24 @@ class KVIter : public Iterator {
 public:
  explicit KVIter(const stl_wrappers::KVMap* map)
      : map_(map), iter_(map_->end()) {}
-  bool Valid() const override { return iter_ != map_->end(); }
-  void SeekToFirst() override { iter_ = map_->begin(); }
-  void SeekToLast() override {
+  virtual bool Valid() const override { return iter_ != map_->end(); }
+  virtual void SeekToFirst() override { iter_ = map_->begin(); }
+  virtual void SeekToLast() override {
    if (map_->empty()) {
      iter_ = map_->end();
    } else {
      iter_ = map_->find(map_->rbegin()->first);
    }
  }
-  void Seek(const Slice& k) override {
+  virtual void Seek(const Slice& k) override {
    iter_ = map_->lower_bound(k.ToString());
  }
-  void SeekForPrev(const Slice& k) override {
+  virtual void SeekForPrev(const Slice& k) override {
    iter_ = map_->upper_bound(k.ToString());
    Prev();
  }
-  void Next() override { ++iter_; }
-  void Prev() override {
+  virtual void Next() override { ++iter_; }
+  virtual void Prev() override {
    if (iter_ == map_->begin()) {
      iter_ = map_->end();
      return;
@@ -52,9 +52,9 @@ class KVIter : public Iterator {
    --iter_;
  }

-  Slice key() const override { return iter_->first; }
-  Slice value() const override { return iter_->second; }
-  Status status() const override { return Status::OK(); }
+  virtual Slice key() const override { return iter_->first; }
+  virtual Slice value() const override { return iter_->second; }
+  virtual Status status() const override { return Status::OK(); }

 private:
  const stl_wrappers::KVMap* const map_;
@@ -171,9 +171,9 @@ class DoubleComparator : public Comparator {
 public:
  DoubleComparator() {}

-  const char* Name() const override { return "DoubleComparator"; }
+  virtual const char* Name() const override { return "DoubleComparator"; }

-  int Compare(const Slice& a, const Slice& b) const override {
+  virtual int Compare(const Slice& a, const Slice& b) const override {
 #ifndef CYGWIN
    double da = std::stod(a.ToString());
    double db = std::stod(b.ToString());
@@ -189,19 +189,19 @@ class DoubleComparator : public Comparator {
      return -1;
    }
  }
-  void FindShortestSeparator(std::string* /*start*/,
-                             const Slice& /*limit*/) const override {}
+  virtual void FindShortestSeparator(std::string* /*start*/,
+                                     const Slice& /*limit*/) const override {}

-  void FindShortSuccessor(std::string* /*key*/) const override {}
+  virtual void FindShortSuccessor(std::string* /*key*/) const override {}
 };

 class HashComparator : public Comparator {
 public:
  HashComparator() {}

-  const char* Name() const override { return "HashComparator"; }
+  virtual const char* Name() const override { return "HashComparator"; }

-  int Compare(const Slice& a, const Slice& b) const override {
+  virtual int Compare(const Slice& a, const Slice& b) const override {
    uint32_t ha = Hash(a.data(), a.size(), 66);
    uint32_t hb = Hash(b.data(), b.size(), 66);
    if (ha == hb) {
@@ -212,19 +212,19 @@ class HashComparator : public Comparator {
      return -1;
    }
  }
-  void FindShortestSeparator(std::string* /*start*/,
-                             const Slice& /*limit*/) const override {}
+  virtual void FindShortestSeparator(std::string* /*start*/,
+                                     const Slice& /*limit*/) const override {}

-  void FindShortSuccessor(std::string* /*key*/) const override {}
+  virtual void FindShortSuccessor(std::string* /*key*/) const override {}
 };

 class TwoStrComparator : public Comparator {
 public:
  TwoStrComparator() {}

-  const char* Name() const override { return "TwoStrComparator"; }
+  virtual const char* Name() const override { return "TwoStrComparator"; }

-  int Compare(const Slice& a, const Slice& b) const override {
+  virtual int Compare(const Slice& a, const Slice& b) const override {
    assert(a.size() >= 2);
    assert(b.size() >= 2);
    size_t size_a1 = static_cast<size_t>(a[0]);
@@ -244,10 +244,10 @@ class TwoStrComparator : public Comparator {
    }
    return a2.compare(b2);
  }
-  void FindShortestSeparator(std::string* /*start*/,
-                             const Slice& /*limit*/) const override {}
+  virtual void FindShortestSeparator(std::string* /*start*/,
+                                     const Slice& /*limit*/) const override {}

-  void FindShortSuccessor(std::string* /*key*/) const override {}
+  virtual void FindShortSuccessor(std::string* /*key*/) const override {}
 };
 }  // namespace

@@ -272,7 +272,7 @@ class ComparatorDBTest
    EXPECT_OK(DestroyDB(dbname_, last_options_));
  }

-  ~ComparatorDBTest() override {
+  ~ComparatorDBTest() {
    delete db_;
    EXPECT_OK(DestroyDB(dbname_, last_options_));
    comparator = BytewiseComparator();
@@ -35,7 +35,7 @@ Status DeleteFilesInRanges(DB* db, ColumnFamilyHandle* column_family,
 Status VerifySstFileChecksum(const Options& options,
                             const EnvOptions& env_options,
                             const std::string& file_path) {
-  std::unique_ptr<RandomAccessFile> file;
+  unique_ptr<RandomAccessFile> file;
  uint64_t file_size;
  InternalKeyComparator internal_comparator(options.comparator);
  ImmutableCFOptions ioptions(options);
@@ -46,7 +46,7 @@ Status VerifySstFileChecksum(const Options& options,
  } else {
    return s;
  }
-  std::unique_ptr<TableReader> table_reader;
+  unique_ptr<TableReader> table_reader;
  std::unique_ptr<RandomAccessFileReader> file_reader(
      new RandomAccessFileReader(std::move(file), file_path));
  const bool kImmortal = true;
@@ -24,8 +24,6 @@
 #include "rocksdb/env.h"
 #include "rocksdb/table.h"
 #include "rocksdb/write_batch.h"
-#include "table/block_based_table_builder.h"
-#include "table/meta_blocks.h"
 #include "util/filename.h"
 #include "util/string_util.h"
 #include "util/testharness.h"
@@ -39,7 +37,7 @@ class CorruptionTest : public testing::Test {
 public:
  test::ErrorEnv env_;
  std::string dbname_;
-  std::shared_ptr<Cache> tiny_cache_;
+  shared_ptr<Cache> tiny_cache_;
  Options options_;
  DB* db_;

@@ -62,9 +60,9 @@ class CorruptionTest : public testing::Test {
    options_.create_if_missing = false;
  }

-  ~CorruptionTest() override {
-    delete db_;
-    DestroyDB(dbname_, Options());
+  ~CorruptionTest() {
+     delete db_;
+     DestroyDB(dbname_, Options());
  }

  void CloseDb() {
@@ -469,39 +467,6 @@ TEST_F(CorruptionTest, UnrelatedKeys) {
  ASSERT_EQ(Value(1000, &tmp2).ToString(), v);
 }

-TEST_F(CorruptionTest, RangeDeletionCorrupted) {
-  ASSERT_OK(
-      db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(), "a", "b"));
-  ASSERT_OK(db_->Flush(FlushOptions()));
-  std::vector<LiveFileMetaData> metadata;
-  db_->GetLiveFilesMetaData(&metadata);
-  ASSERT_EQ(static_cast<size_t>(1), metadata.size());
-  std::string filename = dbname_ + metadata[0].name;
-
-  std::unique_ptr<RandomAccessFile> file;
-  ASSERT_OK(options_.env->NewRandomAccessFile(filename, &file, EnvOptions()));
-  std::unique_ptr<RandomAccessFileReader> file_reader(
-      new RandomAccessFileReader(std::move(file), filename));
-
-  uint64_t file_size;
-  ASSERT_OK(options_.env->GetFileSize(filename, &file_size));
-
-  BlockHandle range_del_handle;
-  ASSERT_OK(FindMetaBlock(
-      file_reader.get(), file_size, kBlockBasedTableMagicNumber,
-      ImmutableCFOptions(options_), kRangeDelBlock, &range_del_handle));
-
-  ASSERT_OK(TryReopen());
-  CorruptFile(filename, static_cast<int>(range_del_handle.offset()), 1);
-  // The test case does not fail on TryReopen because failure to preload table
-  // handlers is not considered critical.
-  ASSERT_OK(TryReopen());
-  std::string val;
-  // However, it does fail on any read involving that file since that file
-  // cannot be opened with a corrupt range deletion meta-block.
-  ASSERT_TRUE(db_->Get(ReadOptions(), "a", &val).IsCorruption());
-}
-
 TEST_F(CorruptionTest, FileSystemStateCorrupted) {
  for (int iter = 0; iter < 2; ++iter) {
    Options options;
@@ -520,7 +485,7 @@ TEST_F(CorruptionTest, FileSystemStateCorrupted) {
    db_ = nullptr;

    if (iter == 0) {  // corrupt file size
-      std::unique_ptr<WritableFile> file;
+      unique_ptr<WritableFile> file;
      env_.NewWritableFile(filename, &file, EnvOptions());
      file->Append(Slice("corrupted sst"));
      file.reset();
@@ -31,7 +31,7 @@ class CuckooTableDBTest : public testing::Test {
    Reopen();
  }

-  ~CuckooTableDBTest() override {
+  ~CuckooTableDBTest() {
    delete db_;
    EXPECT_OK(DestroyDB(dbname_, Options()));
  }
@@ -6,7 +6,6 @@
 // Copyright (c) 2011 The LevelDB Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file. See the AUTHORS file for names of contributors.
-// #include <iostream>
 #include "db/db_test_util.h"
 #include "port/stack_trace.h"
 #include "rocksdb/perf_context.h"
@@ -215,11 +214,11 @@ TEST_F(DBBasicTest, PutSingleDeleteGet) {
    ASSERT_EQ("v2", Get(1, "foo2"));
    ASSERT_OK(SingleDelete(1, "foo"));
    ASSERT_EQ("NOT_FOUND", Get(1, "foo"));
-    // Ski FIFO and universal compaction because they do not apply to the test
-    // case. Skip MergePut because single delete does not get removed when it
-    // encounters a merge.
-  } while (ChangeOptions(kSkipFIFOCompaction | kSkipUniversalCompaction |
-                         kSkipMergePut));
+    // Skip HashCuckooRep as it does not support single delete. FIFO and
+    // universal compaction do not apply to the test case. Skip MergePut
+    // because single delete does not get removed when it encounters a merge.
+  } while (ChangeOptions(kSkipHashCuckoo | kSkipFIFOCompaction |
+                         kSkipUniversalCompaction | kSkipMergePut));
 }

 TEST_F(DBBasicTest, EmptyFlush) {
@@ -237,11 +236,11 @@ TEST_F(DBBasicTest, EmptyFlush) {
    ASSERT_OK(Flush(1));

    ASSERT_EQ("[ ]", AllEntriesFor("a", 1));
-    // Skip FIFO and  universal compaction as they do not apply to the test
-    // case. Skip MergePut because merges cannot be combined with single
-    // deletions.
-  } while (ChangeOptions(kSkipFIFOCompaction | kSkipUniversalCompaction |
-                         kSkipMergePut));
+    // Skip HashCuckooRep as it does not support single delete. FIFO and
+    // universal compaction do not apply to the test case. Skip MergePut
+    // because merges cannot be combined with single deletions.
+  } while (ChangeOptions(kSkipHashCuckoo | kSkipFIFOCompaction |
+                         kSkipUniversalCompaction | kSkipMergePut));
 }

 TEST_F(DBBasicTest, GetFromVersions) {
@@ -265,6 +264,11 @@ TEST_F(DBBasicTest, GetSnapshot) {
      std::string key = (i == 0) ? std::string("foo") : std::string(200, 'x');
      ASSERT_OK(Put(1, key, "v1"));
      const Snapshot* s1 = db_->GetSnapshot();
+      if (option_config_ == kHashCuckoo) {
+        // Unsupported case.
+        ASSERT_TRUE(s1 == nullptr);
+        break;
+      }
      ASSERT_OK(Put(1, key, "v2"));
      ASSERT_EQ("v2", Get(1, key));
      ASSERT_EQ("v1", Get(1, key, s1));
@@ -505,7 +509,7 @@ TEST_F(DBBasicTest, Snapshot) {
    ASSERT_EQ(0U, GetNumSnapshots());
    ASSERT_EQ("0v4", Get(0, "foo"));
    ASSERT_EQ("1v4", Get(1, "foo"));
-  } while (ChangeOptions());
+  } while (ChangeOptions(kSkipHashCuckoo));
 }

 #endif  // ROCKSDB_LITE
@@ -561,7 +565,8 @@ TEST_F(DBBasicTest, CompactBetweenSnapshots) {
                           nullptr);
    ASSERT_EQ("sixth", Get(1, "foo"));
    ASSERT_EQ(AllEntriesFor("foo", 1), "[ sixth ]");
-  } while (ChangeOptions(kSkipFIFOCompaction));
+    // skip HashCuckooRep as it does not support snapshot
+  } while (ChangeOptions(kSkipHashCuckoo | kSkipFIFOCompaction));
 }

 TEST_F(DBBasicTest, DBOpen_Options) {
@@ -789,7 +794,7 @@ TEST_F(DBBasicTest, ChecksumTest) {
  BlockBasedTableOptions table_options;
  Options options = CurrentOptions();
  // change when new checksum type added
-  int max_checksum = static_cast<int>(kxxHash64);
+  int max_checksum = static_cast<int>(kxxHash);
  const int kNumPerFile = 2;

  // generate one table with each type of checksum
@@ -804,7 +809,7 @@ TEST_F(DBBasicTest, ChecksumTest) {
  }

  // verify data with each type of checksum
-  for (int i = 0; i <= kxxHash64; ++i) {
+  for (int i = 0; i <= kxxHash; ++i) {
    table_options.checksum = static_cast<ChecksumType>(i);
    options.table_factory.reset(NewBlockBasedTableFactory(table_options));
    Reopen(options);
@@ -852,17 +857,18 @@ class TestEnv : public EnvWrapper {
      public:
        using Logger::Logv;
        TestLogger(TestEnv *env_ptr) : Logger() { env = env_ptr; }
-        ~TestLogger() override {
+        ~TestLogger() {
          if (!closed_) {
            CloseHelper();
          }
        }
-        void Logv(const char* /*format*/, va_list /*ap*/) override{};
+        virtual void Logv(const char* /*format*/, va_list /*ap*/) override{};

       protected:
-        Status CloseImpl() override { return CloseHelper(); }
-
-       private:
+        virtual Status CloseImpl() override {
+          return CloseHelper();
+        }
+      private:
        Status CloseHelper() {
          env->CloseCountInc();;
          return Status::IOError();
@@ -874,8 +880,8 @@ class TestEnv : public EnvWrapper {

    int GetCloseCount() { return close_count; }

-    Status NewLogger(const std::string& /*fname*/,
-                     std::shared_ptr<Logger>* result) override {
+    virtual Status NewLogger(const std::string& /*fname*/,
+                             shared_ptr<Logger>* result) {
      result->reset(new TestLogger(this));
      return Status::OK();
    }
@@ -950,340 +956,6 @@ TEST_F(DBBasicTest, DBCloseFlushError) {
  Destroy(options);
 }

-TEST_F(DBBasicTest, MultiGetMultiCF) {
-  Options options = CurrentOptions();
-  CreateAndReopenWithCF({"pikachu", "ilya", "muromec", "dobrynia", "nikitich",
-                         "alyosha", "popovich"},
-                        options);
-
-  for (int i = 0; i < 8; ++i) {
-    ASSERT_OK(Put(i, "cf" + std::to_string(i) + "_key",
-                  "cf" + std::to_string(i) + "_val"));
-  }
-
-  int get_sv_count = 0;
-  rocksdb::DBImpl* db = reinterpret_cast<DBImpl*>(db_);
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
-      "DBImpl::MultiGet::AfterRefSV", [&](void* /*arg*/) {
-        if (++get_sv_count == 2) {
-          // After MultiGet refs a couple of CFs, flush all CFs so MultiGet
-          // is forced to repeat the process
-          for (int i = 0; i < 8; ++i) {
-            ASSERT_OK(Flush(i));
-            ASSERT_OK(Put(i, "cf" + std::to_string(i) + "_key",
-                          "cf" + std::to_string(i) + "_val2"));
-          }
-        }
-        if (get_sv_count == 11) {
-          for (int i = 0; i < 8; ++i) {
-            auto* cfd = reinterpret_cast<ColumnFamilyHandleImpl*>(
-                            db->GetColumnFamilyHandle(i))
-                            ->cfd();
-            ASSERT_EQ(cfd->TEST_GetLocalSV()->Get(), SuperVersion::kSVInUse);
-          }
-        }
-      });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-
-  std::vector<int> cfs;
-  std::vector<std::string> keys;
-  std::vector<std::string> values;
-
-  for (int i = 0; i < 8; ++i) {
-    cfs.push_back(i);
-    keys.push_back("cf" + std::to_string(i) + "_key");
-  }
-
-  values = MultiGet(cfs, keys, nullptr);
-  ASSERT_EQ(values.size(), 8);
-  for (unsigned int j = 0; j < values.size(); ++j) {
-    ASSERT_EQ(values[j], "cf" + std::to_string(j) + "_val2");
-  }
-  for (int i = 0; i < 8; ++i) {
-    auto* cfd = reinterpret_cast<ColumnFamilyHandleImpl*>(
-                    reinterpret_cast<DBImpl*>(db_)->GetColumnFamilyHandle(i))
-                    ->cfd();
-    ASSERT_NE(cfd->TEST_GetLocalSV()->Get(), SuperVersion::kSVInUse);
-    ASSERT_NE(cfd->TEST_GetLocalSV()->Get(), SuperVersion::kSVObsolete);
-  }
-}
-
-TEST_F(DBBasicTest, MultiGetMultiCFMutex) {
-  Options options = CurrentOptions();
-  CreateAndReopenWithCF({"pikachu", "ilya", "muromec", "dobrynia", "nikitich",
-                         "alyosha", "popovich"},
-                        options);
-
-  for (int i = 0; i < 8; ++i) {
-    ASSERT_OK(Put(i, "cf" + std::to_string(i) + "_key",
-                  "cf" + std::to_string(i) + "_val"));
-  }
-
-  int get_sv_count = 0;
-  int retries = 0;
-  bool last_try = false;
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
-      "DBImpl::MultiGet::LastTry", [&](void* /*arg*/) {
-        last_try = true;
-        rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-      });
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
-      "DBImpl::MultiGet::AfterRefSV", [&](void* /*arg*/) {
-        if (last_try) {
-          return;
-        }
-        if (++get_sv_count == 2) {
-          ++retries;
-          get_sv_count = 0;
-          for (int i = 0; i < 8; ++i) {
-            ASSERT_OK(Flush(i));
-            ASSERT_OK(Put(
-                i, "cf" + std::to_string(i) + "_key",
-                "cf" + std::to_string(i) + "_val" + std::to_string(retries)));
-          }
-        }
-      });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-
-  std::vector<int> cfs;
-  std::vector<std::string> keys;
-  std::vector<std::string> values;
-
-  for (int i = 0; i < 8; ++i) {
-    cfs.push_back(i);
-    keys.push_back("cf" + std::to_string(i) + "_key");
-  }
-
-  values = MultiGet(cfs, keys, nullptr);
-  ASSERT_TRUE(last_try);
-  ASSERT_EQ(values.size(), 8);
-  for (unsigned int j = 0; j < values.size(); ++j) {
-    ASSERT_EQ(values[j],
-              "cf" + std::to_string(j) + "_val" + std::to_string(retries));
-  }
-  for (int i = 0; i < 8; ++i) {
-    auto* cfd = reinterpret_cast<ColumnFamilyHandleImpl*>(
-                    reinterpret_cast<DBImpl*>(db_)->GetColumnFamilyHandle(i))
-                    ->cfd();
-    ASSERT_NE(cfd->TEST_GetLocalSV()->Get(), SuperVersion::kSVInUse);
-  }
-}
-
-TEST_F(DBBasicTest, MultiGetMultiCFSnapshot) {
-  Options options = CurrentOptions();
-  CreateAndReopenWithCF({"pikachu", "ilya", "muromec", "dobrynia", "nikitich",
-                         "alyosha", "popovich"},
-                        options);
-
-  for (int i = 0; i < 8; ++i) {
-    ASSERT_OK(Put(i, "cf" + std::to_string(i) + "_key",
-                  "cf" + std::to_string(i) + "_val"));
-  }
-
-  int get_sv_count = 0;
-  rocksdb::DBImpl* db = reinterpret_cast<DBImpl*>(db_);
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
-      "DBImpl::MultiGet::AfterRefSV", [&](void* /*arg*/) {
-        if (++get_sv_count == 2) {
-          for (int i = 0; i < 8; ++i) {
-            ASSERT_OK(Flush(i));
-            ASSERT_OK(Put(i, "cf" + std::to_string(i) + "_key",
-                          "cf" + std::to_string(i) + "_val2"));
-          }
-        }
-        if (get_sv_count == 8) {
-          for (int i = 0; i < 8; ++i) {
-            auto* cfd = reinterpret_cast<ColumnFamilyHandleImpl*>(
-                            db->GetColumnFamilyHandle(i))
-                            ->cfd();
-            ASSERT_TRUE(
-                (cfd->TEST_GetLocalSV()->Get() == SuperVersion::kSVInUse) ||
-                (cfd->TEST_GetLocalSV()->Get() == SuperVersion::kSVObsolete));
-          }
-        }
-      });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-
-  std::vector<int> cfs;
-  std::vector<std::string> keys;
-  std::vector<std::string> values;
-
-  for (int i = 0; i < 8; ++i) {
-    cfs.push_back(i);
-    keys.push_back("cf" + std::to_string(i) + "_key");
-  }
-
-  const Snapshot* snapshot = db_->GetSnapshot();
-  values = MultiGet(cfs, keys, snapshot);
-  db_->ReleaseSnapshot(snapshot);
-  ASSERT_EQ(values.size(), 8);
-  for (unsigned int j = 0; j < values.size(); ++j) {
-    ASSERT_EQ(values[j], "cf" + std::to_string(j) + "_val");
-  }
-  for (int i = 0; i < 8; ++i) {
-    auto* cfd = reinterpret_cast<ColumnFamilyHandleImpl*>(
-                    reinterpret_cast<DBImpl*>(db_)->GetColumnFamilyHandle(i))
-                    ->cfd();
-    ASSERT_NE(cfd->TEST_GetLocalSV()->Get(), SuperVersion::kSVInUse);
-  }
-}
-
-TEST_F(DBBasicTest, MultiGetBatchedSimpleUnsorted) {
-  do {
-    CreateAndReopenWithCF({"pikachu"}, CurrentOptions());
-    SetPerfLevel(kEnableCount);
-    ASSERT_OK(Put(1, "k1", "v1"));
-    ASSERT_OK(Put(1, "k2", "v2"));
-    ASSERT_OK(Put(1, "k3", "v3"));
-    ASSERT_OK(Put(1, "k4", "v4"));
-    ASSERT_OK(Delete(1, "k4"));
-    ASSERT_OK(Put(1, "k5", "v5"));
-    ASSERT_OK(Delete(1, "no_key"));
-
-    get_perf_context()->Reset();
-
-    std::vector<Slice> keys({"no_key", "k5", "k4", "k3", "k2", "k1"});
-    std::vector<PinnableSlice> values(keys.size());
-    std::vector<ColumnFamilyHandle*> cfs(keys.size(), handles_[1]);
-    std::vector<Status> s(keys.size());
-
-    db_->MultiGet(ReadOptions(), handles_[1], keys.size(), keys.data(),
-                  values.data(), s.data(), false);
-
-    ASSERT_EQ(values.size(), keys.size());
-    ASSERT_EQ(std::string(values[5].data(), values[5].size()), "v1");
-    ASSERT_EQ(std::string(values[4].data(), values[4].size()), "v2");
-    ASSERT_EQ(std::string(values[3].data(), values[3].size()), "v3");
-    ASSERT_EQ(std::string(values[1].data(), values[1].size()), "v5");
-    // four kv pairs * two bytes per value
-    ASSERT_EQ(8, (int)get_perf_context()->multiget_read_bytes);
-
-    ASSERT_TRUE(s[0].IsNotFound());
-    ASSERT_OK(s[1]);
-    ASSERT_TRUE(s[2].IsNotFound());
-    ASSERT_OK(s[3]);
-    ASSERT_OK(s[4]);
-    ASSERT_OK(s[5]);
-
-    SetPerfLevel(kDisable);
-  } while (ChangeCompactOptions());
-}
-
-TEST_F(DBBasicTest, MultiGetBatchedSimpleSorted) {
-  do {
-    CreateAndReopenWithCF({"pikachu"}, CurrentOptions());
-    SetPerfLevel(kEnableCount);
-    ASSERT_OK(Put(1, "k1", "v1"));
-    ASSERT_OK(Put(1, "k2", "v2"));
-    ASSERT_OK(Put(1, "k3", "v3"));
-    ASSERT_OK(Put(1, "k4", "v4"));
-    ASSERT_OK(Delete(1, "k4"));
-    ASSERT_OK(Put(1, "k5", "v5"));
-    ASSERT_OK(Delete(1, "no_key"));
-
-    get_perf_context()->Reset();
-
-    std::vector<Slice> keys({"k1", "k2", "k3", "k4", "k5", "no_key"});
-    std::vector<PinnableSlice> values(keys.size());
-    std::vector<ColumnFamilyHandle*> cfs(keys.size(), handles_[1]);
-    std::vector<Status> s(keys.size());
-
-    db_->MultiGet(ReadOptions(), handles_[1], keys.size(), keys.data(),
-                  values.data(), s.data(), true);
-
-    ASSERT_EQ(values.size(), keys.size());
-    ASSERT_EQ(std::string(values[0].data(), values[0].size()), "v1");
-    ASSERT_EQ(std::string(values[1].data(), values[1].size()), "v2");
-    ASSERT_EQ(std::string(values[2].data(), values[2].size()), "v3");
-    ASSERT_EQ(std::string(values[4].data(), values[4].size()), "v5");
-    // four kv pairs * two bytes per value
-    ASSERT_EQ(8, (int)get_perf_context()->multiget_read_bytes);
-
-    ASSERT_OK(s[0]);
-    ASSERT_OK(s[1]);
-    ASSERT_OK(s[2]);
-    ASSERT_TRUE(s[3].IsNotFound());
-    ASSERT_OK(s[4]);
-    ASSERT_TRUE(s[5].IsNotFound());
-
-    SetPerfLevel(kDisable);
-  } while (ChangeCompactOptions());
-}
-
-TEST_F(DBBasicTest, MultiGetBatchedMultiLevel) {
-  Options options = CurrentOptions();
-  options.disable_auto_compactions = true;
-  Reopen(options);
-  int num_keys = 0;
-
-  for (int i = 0; i < 128; ++i) {
-    ASSERT_OK(Put("key_" + std::to_string(i), "val_l2_" + std::to_string(i)));
-    num_keys++;
-    if (num_keys == 8) {
-      Flush();
-      num_keys = 0;
-    }
-  }
-  if (num_keys > 0) {
-    Flush();
-    num_keys = 0;
-  }
-  MoveFilesToLevel(2);
-
-  for (int i = 0; i < 128; i += 3) {
-    ASSERT_OK(Put("key_" + std::to_string(i), "val_l1_" + std::to_string(i)));
-    num_keys++;
-    if (num_keys == 8) {
-      Flush();
-      num_keys = 0;
-    }
-  }
-  if (num_keys > 0) {
-    Flush();
-    num_keys = 0;
-  }
-  MoveFilesToLevel(1);
-
-  for (int i = 0; i < 128; i += 5) {
-    ASSERT_OK(Put("key_" + std::to_string(i), "val_l0_" + std::to_string(i)));
-    num_keys++;
-    if (num_keys == 8) {
-      Flush();
-      num_keys = 0;
-    }
-  }
-  if (num_keys > 0) {
-    Flush();
-    num_keys = 0;
-  }
-  ASSERT_EQ(0, num_keys);
-
-  for (int i = 0; i < 128; i += 9) {
-    ASSERT_OK(Put("key_" + std::to_string(i), "val_mem_" + std::to_string(i)));
-  }
-
-  std::vector<std::string> keys;
-  std::vector<std::string> values;
-
-  for (int i = 64; i < 80; ++i) {
-    keys.push_back("key_" + std::to_string(i));
-  }
-
-  values = MultiGet(keys, nullptr);
-  ASSERT_EQ(values.size(), 16);
-  for (unsigned int j = 0; j < values.size(); ++j) {
-    int key = j + 64;
-    if (key % 9 == 0) {
-      ASSERT_EQ(values[j], "val_mem_" + std::to_string(key));
-    } else if (key % 5 == 0) {
-      ASSERT_EQ(values[j], "val_l0_" + std::to_string(key));
-    } else if (key % 3 == 0) {
-      ASSERT_EQ(values[j], "val_l1_" + std::to_string(key));
-    } else {
-      ASSERT_EQ(values[j], "val_l2_" + std::to_string(key));
-    }
-  }
-}
 }  // namespace rocksdb

 int main(int argc, char** argv) {
@@ -173,7 +173,7 @@ TEST_F(DBBlockCacheTest, TestWithoutCompressedBlockCache) {
  delete iter;
  iter = nullptr;

-  // Release iterators and access cache again.
+  // Release interators and access cache again.
  for (size_t i = 0; i < kNumBlocks - 1; i++) {
    iterators[i].reset();
    CheckCacheCounters(options, 0, 0, 0, 0);
@@ -346,14 +346,14 @@ TEST_F(DBBlockCacheTest, IndexAndFilterBlocksStats) {
  options.statistics = rocksdb::CreateDBStatistics();
  BlockBasedTableOptions table_options;
  table_options.cache_index_and_filter_blocks = true;
-  // 500 bytes are enough to hold the first two blocks
-  std::shared_ptr<Cache> cache = NewLRUCache(500, 0, false);
+  // 200 bytes are enough to hold the first two blocks
+  std::shared_ptr<Cache> cache = NewLRUCache(200, 0, false);
  table_options.block_cache = cache;
-  table_options.filter_policy.reset(NewBloomFilterPolicy(20, true));
+  table_options.filter_policy.reset(NewBloomFilterPolicy(20));
  options.table_factory.reset(new BlockBasedTableFactory(table_options));
  CreateAndReopenWithCF({"pikachu"}, options);

-  ASSERT_OK(Put(1, "longer_key", "val"));
+  ASSERT_OK(Put(1, "key", "val"));
  // Create a new table
  ASSERT_OK(Flush(1));
  size_t index_bytes_insert =
@@ -367,9 +367,7 @@ TEST_F(DBBlockCacheTest, IndexAndFilterBlocksStats) {
  cache->SetCapacity(index_bytes_insert + filter_bytes_insert);
  ASSERT_EQ(TestGetTickerCount(options, BLOCK_CACHE_INDEX_BYTES_EVICT), 0);
  ASSERT_EQ(TestGetTickerCount(options, BLOCK_CACHE_FILTER_BYTES_EVICT), 0);
-  // Note that the second key needs to be no longer than the first one.
-  // Otherwise the second index block may not fit in cache.
-  ASSERT_OK(Put(1, "key", "val"));
+  ASSERT_OK(Put(1, "key2", "val"));
  // Create a new table
  ASSERT_OK(Flush(1));
  // cache evicted old index and block entries
@@ -397,9 +395,9 @@ class MockCache : public LRUCache {
                 false /*strict_capacity_limit*/, 0.0 /*high_pri_pool_ratio*/) {
  }

-  Status Insert(const Slice& key, void* value, size_t charge,
-                void (*deleter)(const Slice& key, void* value), Handle** handle,
-                Priority priority) override {
+  virtual Status Insert(const Slice& key, void* value, size_t charge,
+                        void (*deleter)(const Slice& key, void* value),
+                        Handle** handle, Priority priority) override {
    if (priority == Priority::LOW) {
      low_pri_insert_count++;
    } else {
@@ -633,74 +631,6 @@ TEST_F(DBBlockCacheTest, CompressedCache) {
  }
 }

-TEST_F(DBBlockCacheTest, CacheCompressionDict) {
-  const int kNumFiles = 4;
-  const int kNumEntriesPerFile = 128;
-  const int kNumBytesPerEntry = 1024;
-
-  // Try all the available libraries that support dictionary compression
-  std::vector<CompressionType> compression_types;
-#ifdef ZLIB
-  compression_types.push_back(kZlibCompression);
-#endif  // ZLIB
-#if LZ4_VERSION_NUMBER >= 10400
-  compression_types.push_back(kLZ4Compression);
-  compression_types.push_back(kLZ4HCCompression);
-#endif  // LZ4_VERSION_NUMBER >= 10400
-#if ZSTD_VERSION_NUMBER >= 500
-  compression_types.push_back(kZSTD);
-#endif  // ZSTD_VERSION_NUMBER >= 500
-  Random rnd(301);
-  for (auto compression_type : compression_types) {
-    Options options = CurrentOptions();
-    options.compression = compression_type;
-    options.compression_opts.max_dict_bytes = 4096;
-    options.create_if_missing = true;
-    options.num_levels = 2;
-    options.statistics = rocksdb::CreateDBStatistics();
-    options.target_file_size_base = kNumEntriesPerFile * kNumBytesPerEntry;
-    BlockBasedTableOptions table_options;
-    table_options.cache_index_and_filter_blocks = true;
-    table_options.block_cache.reset(new MockCache());
-    options.table_factory.reset(new BlockBasedTableFactory(table_options));
-    DestroyAndReopen(options);
-
-    for (int i = 0; i < kNumFiles; ++i) {
-      ASSERT_EQ(i, NumTableFilesAtLevel(0, 0));
-      for (int j = 0; j < kNumEntriesPerFile; ++j) {
-        std::string value = RandomString(&rnd, kNumBytesPerEntry);
-        ASSERT_OK(Put(Key(j * kNumFiles + i), value.c_str()));
-      }
-      ASSERT_OK(Flush());
-    }
-    dbfull()->TEST_WaitForCompact();
-    ASSERT_EQ(0, NumTableFilesAtLevel(0));
-    ASSERT_EQ(kNumFiles, NumTableFilesAtLevel(1));
-
-    // Seek to a key in a file. It should cause the SST's dictionary meta-block
-    // to be read.
-    RecordCacheCounters(options);
-    ASSERT_EQ(0,
-              TestGetTickerCount(options, BLOCK_CACHE_COMPRESSION_DICT_MISS));
-    ASSERT_EQ(0, TestGetTickerCount(options, BLOCK_CACHE_COMPRESSION_DICT_ADD));
-    ASSERT_EQ(
-        TestGetTickerCount(options, BLOCK_CACHE_COMPRESSION_DICT_BYTES_INSERT),
-        0);
-    ReadOptions read_options;
-    ASSERT_NE("NOT_FOUND", Get(Key(kNumFiles * kNumEntriesPerFile - 1)));
-    // Two blocks missed/added: dictionary and data block
-    // One block hit: index since it's prefetched
-    CheckCacheCounters(options, 2 /* expected_misses */, 1 /* expected_hits */,
-                       2 /* expected_inserts */, 0 /* expected_failures */);
-    ASSERT_EQ(1,
-              TestGetTickerCount(options, BLOCK_CACHE_COMPRESSION_DICT_MISS));
-    ASSERT_EQ(1, TestGetTickerCount(options, BLOCK_CACHE_COMPRESSION_DICT_ADD));
-    ASSERT_GT(
-        TestGetTickerCount(options, BLOCK_CACHE_COMPRESSION_DICT_BYTES_INSERT),
-        0);
-  }
-}
-
 #endif  // ROCKSDB_LITE

 }  // namespace rocksdb
@@ -32,7 +32,7 @@ class DBBloomFilterTestWithParam
 public:
  DBBloomFilterTestWithParam() : DBTestBase("/db_bloom_filter_tests") {}

-  ~DBBloomFilterTestWithParam() override {}
+  ~DBBloomFilterTestWithParam() {}

  void SetUp() override {
    use_block_based_filter_ = std::get<0>(GetParam());
@@ -147,7 +147,6 @@ TEST_F(DBBloomFilterTest, GetFilterByPrefixBloomCustomPrefixExtractor) {
    options.prefix_extractor =
        std::make_shared<SliceTransformLimitedDomainGeneric>();
    options.statistics = rocksdb::CreateDBStatistics();
-    get_perf_context()->EnablePerLevelPerfContext();
    BlockBasedTableOptions bbto;
    bbto.filter_policy.reset(NewBloomFilterPolicy(10, false));
    if (partition_filters) {
@@ -172,39 +171,20 @@ TEST_F(DBBloomFilterTest, GetFilterByPrefixBloomCustomPrefixExtractor) {

    ASSERT_EQ("foo", Get("barbarbar"));
    ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 0);
-    ASSERT_EQ(
-        0,
-        (*(get_perf_context()->level_to_perf_context))[0].bloom_filter_useful);
    ASSERT_EQ("foo2", Get("barbarbar2"));
    ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 0);
-    ASSERT_EQ(
-        0,
-        (*(get_perf_context()->level_to_perf_context))[0].bloom_filter_useful);
    ASSERT_EQ("NOT_FOUND", Get("barbarbar3"));
    ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 0);
-    ASSERT_EQ(
-        0,
-        (*(get_perf_context()->level_to_perf_context))[0].bloom_filter_useful);

    ASSERT_EQ("NOT_FOUND", Get("barfoofoo"));
    ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 1);
-    ASSERT_EQ(
-        1,
-        (*(get_perf_context()->level_to_perf_context))[0].bloom_filter_useful);

    ASSERT_EQ("NOT_FOUND", Get("foobarbar"));
    ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 2);
-    ASSERT_EQ(
-        2,
-        (*(get_perf_context()->level_to_perf_context))[0].bloom_filter_useful);

    ro.total_order_seek = true;
    ASSERT_TRUE(db_->Get(ro, "foobarbar", &value).IsNotFound());
    ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 2);
-    ASSERT_EQ(
-        2,
-        (*(get_perf_context()->level_to_perf_context))[0].bloom_filter_useful);
-    get_perf_context()->Reset();
  }
 }

@@ -213,7 +193,6 @@ TEST_F(DBBloomFilterTest, GetFilterByPrefixBloom) {
    Options options = last_options_;
    options.prefix_extractor.reset(NewFixedPrefixTransform(8));
    options.statistics = rocksdb::CreateDBStatistics();
-    get_perf_context()->EnablePerLevelPerfContext();
    BlockBasedTableOptions bbto;
    bbto.filter_policy.reset(NewBloomFilterPolicy(10, false));
    if (partition_filters) {
@@ -252,10 +231,6 @@ TEST_F(DBBloomFilterTest, GetFilterByPrefixBloom) {
    ro.total_order_seek = true;
    ASSERT_TRUE(db_->Get(ro, "foobarbar", &value).IsNotFound());
    ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 2);
-    ASSERT_EQ(
-        2,
-        (*(get_perf_context()->level_to_perf_context))[0].bloom_filter_useful);
-    get_perf_context()->Reset();
  }
 }

@@ -264,7 +239,6 @@ TEST_F(DBBloomFilterTest, WholeKeyFilterProp) {
    Options options = last_options_;
    options.prefix_extractor.reset(NewFixedPrefixTransform(3));
    options.statistics = rocksdb::CreateDBStatistics();
-    get_perf_context()->EnablePerLevelPerfContext();

    BlockBasedTableOptions bbto;
    bbto.filter_policy.reset(NewBloomFilterPolicy(10, false));
@@ -412,14 +386,6 @@ TEST_F(DBBloomFilterTest, WholeKeyFilterProp) {
    ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 12);
    ASSERT_EQ("bar", Get("barfoo"));
    ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 12);
-    uint64_t bloom_filter_useful_all_levels = 0;
-    for (auto& kv : (*(get_perf_context()->level_to_perf_context))) {
-      if (kv.second.bloom_filter_useful > 0) {
-        bloom_filter_useful_all_levels += kv.second.bloom_filter_useful;
-      }
-    }
-    ASSERT_EQ(12, bloom_filter_useful_all_levels);
-    get_perf_context()->Reset();
  }
 }

@@ -499,7 +465,6 @@ TEST_P(DBBloomFilterTestWithParam, BloomFilter) {
  } while (ChangeCompactOptions());
 }

-#ifndef ROCKSDB_VALGRIND_RUN
 INSTANTIATE_TEST_CASE_P(
    FormatDef, DBBloomFilterTestDefFormatVersion,
    ::testing::Values(std::make_tuple(true, false, test::kDefaultFormatVersion),
@@ -520,13 +485,11 @@ INSTANTIATE_TEST_CASE_P(
                      std::make_tuple(false, true, test::kLatestFormatVersion),
                      std::make_tuple(false, false,
                                      test::kLatestFormatVersion)));
-#endif  // ROCKSDB_VALGRIND_RUN

 TEST_F(DBBloomFilterTest, BloomFilterRate) {
  while (ChangeFilterOptions()) {
    Options options = CurrentOptions();
    options.statistics = rocksdb::CreateDBStatistics();
-    get_perf_context()->EnablePerLevelPerfContext();
    CreateAndReopenWithCF({"pikachu"}, options);

    const int maxKey = 10000;
@@ -548,10 +511,6 @@ TEST_F(DBBloomFilterTest, BloomFilterRate) {
      ASSERT_EQ("NOT_FOUND", Get(1, Key(i + 33333)));
    }
    ASSERT_GE(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), maxKey * 0.98);
-    ASSERT_GE(
-        (*(get_perf_context()->level_to_perf_context))[0].bloom_filter_useful,
-        maxKey * 0.98);
-    get_perf_context()->Reset();
  }
 }

@@ -642,7 +601,7 @@ class WrappedBloom : public FilterPolicy {
  explicit WrappedBloom(int bits_per_key)
      : filter_(NewBloomFilterPolicy(bits_per_key)), counter_(0) {}

-  ~WrappedBloom() override { delete filter_; }
+  ~WrappedBloom() { delete filter_; }

  const char* Name() const override { return "WrappedRocksDbFilterPolicy"; }

@@ -786,75 +745,6 @@ TEST_F(DBBloomFilterTest, PrefixExtractorBlockFilter) {
  delete iter;
 }

-TEST_F(DBBloomFilterTest, MemtableWholeKeyBloomFilter) {
-  // regression test for #2743. the range delete tombstones in memtable should
-  // be added even when Get() skips searching due to its prefix bloom filter
-  const int kMemtableSize = 1 << 20;              // 1MB
-  const int kMemtablePrefixFilterSize = 1 << 13;  // 8KB
-  const int kPrefixLen = 4;
-  Options options = CurrentOptions();
-  options.memtable_prefix_bloom_size_ratio =
-      static_cast<double>(kMemtablePrefixFilterSize) / kMemtableSize;
-  options.prefix_extractor.reset(rocksdb::NewFixedPrefixTransform(kPrefixLen));
-  options.write_buffer_size = kMemtableSize;
-  options.memtable_whole_key_filtering = false;
-  Reopen(options);
-  std::string key1("AAAABBBB");
-  std::string key2("AAAACCCC");  // not in DB
-  std::string key3("AAAADDDD");
-  std::string key4("AAAAEEEE");
-  std::string value1("Value1");
-  std::string value3("Value3");
-  std::string value4("Value4");
-
-  ASSERT_OK(Put(key1, value1, WriteOptions()));
-
-  // check memtable bloom stats
-  ASSERT_EQ("NOT_FOUND", Get(key2));
-  ASSERT_EQ(0, get_perf_context()->bloom_memtable_miss_count);
-  // same prefix, bloom filter false positive
-  ASSERT_EQ(1, get_perf_context()->bloom_memtable_hit_count);
-
-  // enable whole key bloom filter
-  options.memtable_whole_key_filtering = true;
-  Reopen(options);
-  // check memtable bloom stats
-  ASSERT_OK(Put(key3, value3, WriteOptions()));
-  ASSERT_EQ("NOT_FOUND", Get(key2));
-  // whole key bloom filter kicks in and determines it's a miss
-  ASSERT_EQ(1, get_perf_context()->bloom_memtable_miss_count);
-  ASSERT_EQ(1, get_perf_context()->bloom_memtable_hit_count);
-
-  // verify whole key filtering does not depend on prefix_extractor
-  options.prefix_extractor.reset();
-  Reopen(options);
-  // check memtable bloom stats
-  ASSERT_OK(Put(key4, value4, WriteOptions()));
-  ASSERT_EQ("NOT_FOUND", Get(key2));
-  // whole key bloom filter kicks in and determines it's a miss
-  ASSERT_EQ(2, get_perf_context()->bloom_memtable_miss_count);
-  ASSERT_EQ(1, get_perf_context()->bloom_memtable_hit_count);
-}
-
-TEST_F(DBBloomFilterTest, MemtablePrefixBloomOutOfDomain) {
-  constexpr size_t kPrefixSize = 8;
-  const std::string kKey = "key";
-  assert(kKey.size() < kPrefixSize);
-  Options options = CurrentOptions();
-  options.prefix_extractor.reset(NewFixedPrefixTransform(kPrefixSize));
-  options.memtable_prefix_bloom_size_ratio = 0.25;
-  Reopen(options);
-  ASSERT_OK(Put(kKey, "v"));
-  ASSERT_EQ("v", Get(kKey));
-  std::unique_ptr<Iterator> iter(dbfull()->NewIterator(ReadOptions()));
-  iter->Seek(kKey);
-  ASSERT_TRUE(iter->Valid());
-  ASSERT_EQ(kKey, iter->key());
-  iter->SeekForPrev(kKey);
-  ASSERT_TRUE(iter->Valid());
-  ASSERT_EQ(kKey, iter->key());
-}
-
 #ifndef ROCKSDB_LITE
 class BloomStatsTestWithParam
    : public DBBloomFilterTest,
@@ -892,7 +782,7 @@ class BloomStatsTestWithParam
    DestroyAndReopen(options_);
  }

-  ~BloomStatsTestWithParam() override {
+  ~BloomStatsTestWithParam() {
    get_perf_context()->Reset();
    Destroy(options_);
  }
@@ -966,7 +856,7 @@ TEST_P(BloomStatsTestWithParam, BloomStatsTestWithIter) {
  ASSERT_OK(Put(key1, value1, WriteOptions()));
  ASSERT_OK(Put(key3, value3, WriteOptions()));

-  std::unique_ptr<Iterator> iter(dbfull()->NewIterator(ReadOptions()));
+  unique_ptr<Iterator> iter(dbfull()->NewIterator(ReadOptions()));

  // check memtable bloom stats
  iter->Seek(key1);
@@ -1004,16 +894,13 @@ TEST_P(BloomStatsTestWithParam, BloomStatsTestWithIter) {
  ASSERT_OK(iter->status());
  ASSERT_TRUE(iter->Valid());
  ASSERT_EQ(value3, iter->value().ToString());
-  // The seek doesn't check block-based bloom filter because last index key
-  // starts with the same prefix we're seeking to.
-  uint64_t expected_hits = use_block_based_builder_ ? 1 : 2;
-  ASSERT_EQ(expected_hits, get_perf_context()->bloom_sst_hit_count);
+  ASSERT_EQ(2, get_perf_context()->bloom_sst_hit_count);

  iter->Seek(key2);
  ASSERT_OK(iter->status());
  ASSERT_TRUE(!iter->Valid());
  ASSERT_EQ(1, get_perf_context()->bloom_sst_miss_count);
-  ASSERT_EQ(expected_hits, get_perf_context()->bloom_sst_hit_count);
+  ASSERT_EQ(2, get_perf_context()->bloom_sst_hit_count);
 }

 INSTANTIATE_TEST_CASE_P(BloomStatsTestWithParam, BloomStatsTestWithParam,
@@ -1145,7 +1032,6 @@ TEST_F(DBBloomFilterTest, OptimizeFiltersForHits) {
  options.table_factory.reset(NewBlockBasedTableFactory(bbto));
  options.optimize_filters_for_hits = true;
  options.statistics = rocksdb::CreateDBStatistics();
-  get_perf_context()->EnablePerLevelPerfContext();
  CreateAndReopenWithCF({"mypikachu"}, options);

  int numkeys = 200000;
@@ -1192,14 +1078,6 @@ TEST_F(DBBloomFilterTest, OptimizeFiltersForHits) {
  // no bloom filter. Most keys be checked bloom filters twice.
  ASSERT_GT(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 65000 * 2);
  ASSERT_LT(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 120000 * 2);
-  uint64_t bloom_filter_useful_all_levels = 0;
-  for (auto& kv : (*(get_perf_context()->level_to_perf_context))) {
-    if (kv.second.bloom_filter_useful > 0) {
-      bloom_filter_useful_all_levels += kv.second.bloom_filter_useful;
-    }
-  }
-  ASSERT_GT(bloom_filter_useful_all_levels, 65000 * 2);
-  ASSERT_LT(bloom_filter_useful_all_levels, 120000 * 2);

  for (int i = 0; i < numkeys; i += 2) {
    ASSERT_EQ(Get(1, Key(i)), "val");
@@ -1309,7 +1187,6 @@ TEST_F(DBBloomFilterTest, OptimizeFiltersForHits) {
  ASSERT_EQ(0, TestGetTickerCount(options, BLOCK_CACHE_FILTER_HIT));
  ASSERT_EQ(2 /* index and data block */,
            TestGetTickerCount(options, BLOCK_CACHE_ADD));
-  get_perf_context()->Reset();
 }

 int CountIter(std::unique_ptr<Iterator>& iter, const Slice& key) {
@@ -1338,8 +1215,6 @@ TEST_F(DBBloomFilterTest, DynamicBloomFilterUpperBound) {
    table_options.cache_index_and_filter_blocks = true;
    table_options.filter_policy.reset(
        NewBloomFilterPolicy(10, use_block_based_builder));
-    table_options.index_shortening = BlockBasedTableOptions::
-        IndexShorteningMode::kShortenSeparatorsAndSuccessor;
    options.table_factory.reset(NewBlockBasedTableFactory(table_options));
    DestroyAndReopen(options);

@@ -63,33 +63,33 @@ INSTANTIATE_TEST_CASE_P(DBTestCompactionFilterWithCompactOption,

 class KeepFilter : public CompactionFilter {
 public:
-  bool Filter(int /*level*/, const Slice& /*key*/, const Slice& /*value*/,
-              std::string* /*new_value*/,
-              bool* /*value_changed*/) const override {
+  virtual bool Filter(int /*level*/, const Slice& /*key*/,
+                      const Slice& /*value*/, std::string* /*new_value*/,
+                      bool* /*value_changed*/) const override {
    cfilter_count++;
    return false;
  }

-  const char* Name() const override { return "KeepFilter"; }
+  virtual const char* Name() const override { return "KeepFilter"; }
 };

 class DeleteFilter : public CompactionFilter {
 public:
-  bool Filter(int /*level*/, const Slice& /*key*/, const Slice& /*value*/,
-              std::string* /*new_value*/,
-              bool* /*value_changed*/) const override {
+  virtual bool Filter(int /*level*/, const Slice& /*key*/,
+                      const Slice& /*value*/, std::string* /*new_value*/,
+                      bool* /*value_changed*/) const override {
    cfilter_count++;
    return true;
  }

-  const char* Name() const override { return "DeleteFilter"; }
+  virtual const char* Name() const override { return "DeleteFilter"; }
 };

 class DeleteISFilter : public CompactionFilter {
 public:
-  bool Filter(int /*level*/, const Slice& key, const Slice& /*value*/,
-              std::string* /*new_value*/,
-              bool* /*value_changed*/) const override {
+  virtual bool Filter(int /*level*/, const Slice& key, const Slice& /*value*/,
+                      std::string* /*new_value*/,
+                      bool* /*value_changed*/) const override {
    cfilter_count++;
    int i = std::stoi(key.ToString());
    if (i > 5 && i <= 105) {
@@ -98,18 +98,20 @@ class DeleteISFilter : public CompactionFilter {
    return false;
  }

-  bool IgnoreSnapshots() const override { return true; }
+  virtual bool IgnoreSnapshots() const override { return true; }

-  const char* Name() const override { return "DeleteFilter"; }
+  virtual const char* Name() const override { return "DeleteFilter"; }
 };

 // Skip x if floor(x/10) is even, use range skips. Requires that keys are
 // zero-padded to length 10.
 class SkipEvenFilter : public CompactionFilter {
 public:
-  Decision FilterV2(int /*level*/, const Slice& key, ValueType /*value_type*/,
-                    const Slice& /*existing_value*/, std::string* /*new_value*/,
-                    std::string* skip_until) const override {
+  virtual Decision FilterV2(int /*level*/, const Slice& key,
+                            ValueType /*value_type*/,
+                            const Slice& /*existing_value*/,
+                            std::string* /*new_value*/,
+                            std::string* skip_until) const override {
    cfilter_count++;
    int i = std::stoi(key.ToString());
    if (i / 10 % 2 == 0) {
@@ -122,22 +124,22 @@ class SkipEvenFilter : public CompactionFilter {
    return Decision::kKeep;
  }

-  bool IgnoreSnapshots() const override { return true; }
+  virtual bool IgnoreSnapshots() const override { return true; }

-  const char* Name() const override { return "DeleteFilter"; }
+  virtual const char* Name() const override { return "DeleteFilter"; }
 };

 class DelayFilter : public CompactionFilter {
 public:
  explicit DelayFilter(DBTestBase* d) : db_test(d) {}
-  bool Filter(int /*level*/, const Slice& /*key*/, const Slice& /*value*/,
-              std::string* /*new_value*/,
-              bool* /*value_changed*/) const override {
+  virtual bool Filter(int /*level*/, const Slice& /*key*/,
+                      const Slice& /*value*/, std::string* /*new_value*/,
+                      bool* /*value_changed*/) const override {
    db_test->env_->addon_time_.fetch_add(1000);
    return true;
  }

-  const char* Name() const override { return "DelayFilter"; }
+  virtual const char* Name() const override { return "DelayFilter"; }

 private:
  DBTestBase* db_test;
@@ -147,13 +149,13 @@ class ConditionalFilter : public CompactionFilter {
 public:
  explicit ConditionalFilter(const std::string* filtered_value)
      : filtered_value_(filtered_value) {}
-  bool Filter(int /*level*/, const Slice& /*key*/, const Slice& value,
-              std::string* /*new_value*/,
-              bool* /*value_changed*/) const override {
+  virtual bool Filter(int /*level*/, const Slice& /*key*/, const Slice& value,
+                      std::string* /*new_value*/,
+                      bool* /*value_changed*/) const override {
    return value.ToString() == *filtered_value_;
  }

-  const char* Name() const override { return "ConditionalFilter"; }
+  virtual const char* Name() const override { return "ConditionalFilter"; }

 private:
  const std::string* filtered_value_;
@@ -163,15 +165,16 @@ class ChangeFilter : public CompactionFilter {
 public:
  explicit ChangeFilter() {}

-  bool Filter(int /*level*/, const Slice& /*key*/, const Slice& /*value*/,
-              std::string* new_value, bool* value_changed) const override {
+  virtual bool Filter(int /*level*/, const Slice& /*key*/,
+                      const Slice& /*value*/, std::string* new_value,
+                      bool* value_changed) const override {
    assert(new_value != nullptr);
    *new_value = NEW_VALUE;
    *value_changed = true;
    return false;
  }

-  const char* Name() const override { return "ChangeFilter"; }
+  virtual const char* Name() const override { return "ChangeFilter"; }
 };

 class KeepFilterFactory : public CompactionFilterFactory {
@@ -182,7 +185,7 @@ class KeepFilterFactory : public CompactionFilterFactory {
        check_context_cf_id_(check_context_cf_id),
        compaction_filter_created_(false) {}

-  std::unique_ptr<CompactionFilter> CreateCompactionFilter(
+  virtual std::unique_ptr<CompactionFilter> CreateCompactionFilter(
      const CompactionFilter::Context& context) override {
    if (check_context_) {
      EXPECT_EQ(expect_full_compaction_.load(), context.is_full_compaction);
@@ -197,7 +200,7 @@ class KeepFilterFactory : public CompactionFilterFactory {

  bool compaction_filter_created() const { return compaction_filter_created_; }

-  const char* Name() const override { return "KeepFilterFactory"; }
+  virtual const char* Name() const override { return "KeepFilterFactory"; }
  bool check_context_;
  bool check_context_cf_id_;
  std::atomic_bool expect_full_compaction_;
@@ -208,7 +211,7 @@ class KeepFilterFactory : public CompactionFilterFactory {

 class DeleteFilterFactory : public CompactionFilterFactory {
 public:
-  std::unique_ptr<CompactionFilter> CreateCompactionFilter(
+  virtual std::unique_ptr<CompactionFilter> CreateCompactionFilter(
      const CompactionFilter::Context& context) override {
    if (context.is_manual_compaction) {
      return std::unique_ptr<CompactionFilter>(new DeleteFilter());
@@ -217,13 +220,13 @@ class DeleteFilterFactory : public CompactionFilterFactory {
    }
  }

-  const char* Name() const override { return "DeleteFilterFactory"; }
+  virtual const char* Name() const override { return "DeleteFilterFactory"; }
 };

 // Delete Filter Factory which ignores snapshots
 class DeleteISFilterFactory : public CompactionFilterFactory {
 public:
-  std::unique_ptr<CompactionFilter> CreateCompactionFilter(
+  virtual std::unique_ptr<CompactionFilter> CreateCompactionFilter(
      const CompactionFilter::Context& context) override {
    if (context.is_manual_compaction) {
      return std::unique_ptr<CompactionFilter>(new DeleteISFilter());
@@ -232,12 +235,12 @@ class DeleteISFilterFactory : public CompactionFilterFactory {
    }
  }

-  const char* Name() const override { return "DeleteFilterFactory"; }
+  virtual const char* Name() const override { return "DeleteFilterFactory"; }
 };

 class SkipEvenFilterFactory : public CompactionFilterFactory {
 public:
-  std::unique_ptr<CompactionFilter> CreateCompactionFilter(
+  virtual std::unique_ptr<CompactionFilter> CreateCompactionFilter(
      const CompactionFilter::Context& context) override {
    if (context.is_manual_compaction) {
      return std::unique_ptr<CompactionFilter>(new SkipEvenFilter());
@@ -246,18 +249,18 @@ class SkipEvenFilterFactory : public CompactionFilterFactory {
    }
  }

-  const char* Name() const override { return "SkipEvenFilterFactory"; }
+  virtual const char* Name() const override { return "SkipEvenFilterFactory"; }
 };

 class DelayFilterFactory : public CompactionFilterFactory {
 public:
  explicit DelayFilterFactory(DBTestBase* d) : db_test(d) {}
-  std::unique_ptr<CompactionFilter> CreateCompactionFilter(
+  virtual std::unique_ptr<CompactionFilter> CreateCompactionFilter(
      const CompactionFilter::Context& /*context*/) override {
    return std::unique_ptr<CompactionFilter>(new DelayFilter(db_test));
  }

-  const char* Name() const override { return "DelayFilterFactory"; }
+  virtual const char* Name() const override { return "DelayFilterFactory"; }

 private:
  DBTestBase* db_test;
@@ -268,13 +271,15 @@ class ConditionalFilterFactory : public CompactionFilterFactory {
  explicit ConditionalFilterFactory(const Slice& filtered_value)
      : filtered_value_(filtered_value.ToString()) {}

-  std::unique_ptr<CompactionFilter> CreateCompactionFilter(
+  virtual std::unique_ptr<CompactionFilter> CreateCompactionFilter(
      const CompactionFilter::Context& /*context*/) override {
    return std::unique_ptr<CompactionFilter>(
        new ConditionalFilter(&filtered_value_));
  }

-  const char* Name() const override { return "ConditionalFilterFactory"; }
+  virtual const char* Name() const override {
+    return "ConditionalFilterFactory";
+  }

 private:
  std::string filtered_value_;
@@ -284,12 +289,12 @@ class ChangeFilterFactory : public CompactionFilterFactory {
 public:
  explicit ChangeFilterFactory() {}

-  std::unique_ptr<CompactionFilter> CreateCompactionFilter(
+  virtual std::unique_ptr<CompactionFilter> CreateCompactionFilter(
      const CompactionFilter::Context& /*context*/) override {
    return std::unique_ptr<CompactionFilter>(new ChangeFilter());
  }

-  const char* Name() const override { return "ChangeFilterFactory"; }
+  virtual const char* Name() const override { return "ChangeFilterFactory"; }
 };

 #ifndef ROCKSDB_LITE
@@ -335,10 +340,9 @@ TEST_F(DBTestCompactionFilter, CompactionFilter) {
  Arena arena;
  {
    InternalKeyComparator icmp(options.comparator);
-    ReadRangeDelAggregator range_del_agg(&icmp,
-                                         kMaxSequenceNumber /* upper_bound */);
-    ScopedArenaIterator iter(dbfull()->NewInternalIterator(
-        &arena, &range_del_agg, kMaxSequenceNumber, handles_[1]));
+    RangeDelAggregator range_del_agg(icmp, {} /* snapshots */);
+    ScopedArenaIterator iter(
+        dbfull()->NewInternalIterator(&arena, &range_del_agg, handles_[1]));
    iter->SeekToFirst();
    ASSERT_OK(iter->status());
    while (iter->Valid()) {
@@ -352,7 +356,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilter) {
    }
  }
  ASSERT_EQ(total, 100000);
-  ASSERT_EQ(count, 0);
+  ASSERT_EQ(count, 1);

  // overwrite all the 100K keys once again.
  for (int i = 0; i < 100000; i++) {
@@ -425,10 +429,9 @@ TEST_F(DBTestCompactionFilter, CompactionFilter) {
  count = 0;
  {
    InternalKeyComparator icmp(options.comparator);
-    ReadRangeDelAggregator range_del_agg(&icmp,
-                                         kMaxSequenceNumber /* upper_bound */);
-    ScopedArenaIterator iter(dbfull()->NewInternalIterator(
-        &arena, &range_del_agg, kMaxSequenceNumber, handles_[1]));
+    RangeDelAggregator range_del_agg(icmp, {} /* snapshots */);
+    ScopedArenaIterator iter(
+        dbfull()->NewInternalIterator(&arena, &range_del_agg, handles_[1]));
    iter->SeekToFirst();
    ASSERT_OK(iter->status());
    while (iter->Valid()) {
@@ -643,10 +646,9 @@ TEST_F(DBTestCompactionFilter, CompactionFilterContextManual) {
    int total = 0;
    Arena arena;
    InternalKeyComparator icmp(options.comparator);
-    ReadRangeDelAggregator range_del_agg(&icmp,
-                                         kMaxSequenceNumber /* snapshots */);
-    ScopedArenaIterator iter(dbfull()->NewInternalIterator(
-        &arena, &range_del_agg, kMaxSequenceNumber));
+    RangeDelAggregator range_del_agg(icmp, {} /* snapshots */);
+    ScopedArenaIterator iter(
+        dbfull()->NewInternalIterator(&arena, &range_del_agg));
    iter->SeekToFirst();
    ASSERT_OK(iter->status());
    while (iter->Valid()) {
@@ -659,7 +661,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilterContextManual) {
      iter->Next();
    }
    ASSERT_EQ(total, 700);
-    ASSERT_EQ(count, 0);
+    ASSERT_EQ(count, 1);
  }
 }
 #endif  // ROCKSDB_LITE
@@ -694,7 +696,44 @@ TEST_F(DBTestCompactionFilter, CompactionFilterContextCfId) {
 }

 #ifndef ROCKSDB_LITE
-// Compaction filters aplies to all records, regardless snapshots.
+// Compaction filters should only be applied to records that are newer than the
+// latest snapshot. This test inserts records and applies a delete filter.
+TEST_F(DBTestCompactionFilter, CompactionFilterSnapshot) {
+  Options options = CurrentOptions();
+  options.compaction_filter_factory = std::make_shared<DeleteFilterFactory>();
+  options.disable_auto_compactions = true;
+  options.create_if_missing = true;
+  DestroyAndReopen(options);
+
+  // Put some data.
+  const Snapshot* snapshot = nullptr;
+  for (int table = 0; table < 4; ++table) {
+    for (int i = 0; i < 10; ++i) {
+      Put(ToString(table * 100 + i), "val");
+    }
+    Flush();
+
+    if (table == 0) {
+      snapshot = db_->GetSnapshot();
+    }
+  }
+  assert(snapshot != nullptr);
+
+  cfilter_count = 0;
+  ASSERT_OK(db_->CompactRange(CompactRangeOptions(), nullptr, nullptr));
+  // The filter should delete 10 records.
+  ASSERT_EQ(30U, cfilter_count);
+
+  // Release the snapshot and compact again -> now all records should be
+  // removed.
+  db_->ReleaseSnapshot(snapshot);
+  ASSERT_OK(db_->CompactRange(CompactRangeOptions(), nullptr, nullptr));
+  ASSERT_EQ(0U, CountLiveFiles());
+}
+
+// Compaction filters should only be applied to records that are newer than the
+// latest snapshot. However, if the compaction filter asks to ignore snapshots
+// records newer than the snapshot will also be processed
 TEST_F(DBTestCompactionFilter, CompactionFilterIgnoreSnapshot) {
  std::string five = ToString(5);
  Options options = CurrentOptions();
@@ -809,7 +848,7 @@ TEST_F(DBTestCompactionFilter, SkipUntilWithBloomFilter) {
  DestroyAndReopen(options);

  Put("0000000010", "v10");
-  Put("0000000020", "v20");  // skipped
+  Put("0000000020", "v20"); // skipped
  Put("0000000050", "v50");
  Flush();

@@ -832,38 +871,6 @@ TEST_F(DBTestCompactionFilter, SkipUntilWithBloomFilter) {
  EXPECT_EQ("v50", val);
 }

-class TestNotSupportedFilter : public CompactionFilter {
- public:
-  bool Filter(int /*level*/, const Slice& /*key*/, const Slice& /*value*/,
-              std::string* /*new_value*/,
-              bool* /*value_changed*/) const override {
-    return true;
-  }
-
-  const char* Name() const override { return "NotSupported"; }
-  bool IgnoreSnapshots() const override { return false; }
-};
-
-TEST_F(DBTestCompactionFilter, IgnoreSnapshotsFalse) {
-  Options options = CurrentOptions();
-  options.compaction_filter = new TestNotSupportedFilter();
-  DestroyAndReopen(options);
-
-  Put("a", "v10");
-  Put("z", "v20");
-  Flush();
-
-  Put("a", "v10");
-  Put("z", "v20");
-  Flush();
-
-  // Comapction should fail because IgnoreSnapshots() = false
-  EXPECT_TRUE(db_->CompactRange(CompactRangeOptions(), nullptr, nullptr)
-                  .IsNotSupported());
-
-  delete options.compaction_filter;
-}
-
 }  // namespace rocksdb

 int main(int argc, char** argv) {
@@ -8,15 +8,11 @@
 // found in the LICENSE file. See the AUTHORS file for names of contributors.

 #include "db/db_test_util.h"
-#include "port/port.h"
 #include "port/stack_trace.h"
-#include "rocksdb/concurrent_task_limiter.h"
+#include "port/port.h"
 #include "rocksdb/experimental.h"
 #include "rocksdb/utilities/convenience.h"
-#include "util/concurrent_task_limiter_impl.h"
-#include "util/fault_injection_test_env.h"
 #include "util/sync_point.h"
-
 namespace rocksdb {

 // SYNC_POINT is not supported in released Windows mode.
@@ -55,9 +51,9 @@ namespace {
 class FlushedFileCollector : public EventListener {
 public:
  FlushedFileCollector() {}
-  ~FlushedFileCollector() override {}
+  ~FlushedFileCollector() {}

-  void OnFlushCompleted(DB* /*db*/, const FlushJobInfo& info) override {
+  virtual void OnFlushCompleted(DB* /*db*/, const FlushJobInfo& info) override {
    std::lock_guard<std::mutex> lock(mutex_);
    flushed_files_.push_back(info.file_path);
  }
@@ -87,23 +83,24 @@ public:
    }
  }

-  ~CompactionStatsCollector() override {}
+  ~CompactionStatsCollector() {}

-  void OnCompactionCompleted(DB* /* db */,
-                             const CompactionJobInfo& info) override {
+  virtual void OnCompactionCompleted(DB* /* db */,
+      const CompactionJobInfo& info) override {
    int k = static_cast<int>(info.compaction_reason);
    int num_of_reasons = static_cast<int>(CompactionReason::kNumOfReasons);
    assert(k >= 0 && k < num_of_reasons);
    compaction_completed_[k]++;
  }

-  void OnExternalFileIngested(
-      DB* /* db */, const ExternalFileIngestionInfo& /* info */) override {
+  virtual void OnExternalFileIngested(DB* /* db */,
+      const ExternalFileIngestionInfo& /* info */) override {
    int k = static_cast<int>(CompactionReason::kExternalSstIngestion);
    compaction_completed_[k]++;
  }

-  void OnFlushCompleted(DB* /* db */, const FlushJobInfo& /* info */) override {
+  virtual void OnFlushCompleted(DB* /* db */,
+      const FlushJobInfo& /* info */) override {
    int k = static_cast<int>(CompactionReason::kFlush);
    compaction_completed_[k]++;
  }
@@ -123,12 +120,13 @@ class SstStatsCollector : public EventListener {
 public:
  SstStatsCollector() : num_ssts_creation_started_(0) {}

-  void OnTableFileCreationStarted(
-      const TableFileCreationBriefInfo& /* info */) override {
+  void OnTableFileCreationStarted(const TableFileCreationBriefInfo& /* info */) override {
    ++num_ssts_creation_started_;
  }

-  int num_ssts_creation_started() { return num_ssts_creation_started_; }
+  int num_ssts_creation_started() {
+    return num_ssts_creation_started_;
+  }

 private:
  std::atomic<int> num_ssts_creation_started_;
@@ -353,8 +351,7 @@ TEST_P(DBCompactionTestWithParam, CompactionsPreserveDeletes) {
    CompactRangeOptions cro;
    cro.change_level = true;
    cro.target_level = 2;
-    cro.bottommost_level_compaction =
-        BottommostLevelCompaction::kForceOptimized;
+    cro.bottommost_level_compaction = BottommostLevelCompaction::kForce;

    dbfull()->TEST_WaitForFlushMemTable();
    dbfull()->CompactRange(cro, nullptr, nullptr);
@@ -414,7 +411,6 @@ TEST_F(DBCompactionTest, SkipStatsUpdateTest) {

  // Reopen the DB with stats-update disabled
  options.skip_stats_update_on_db_open = true;
-  options.max_open_files = 20;
  env_->random_file_open_counter_.store(0);
  Reopen(options);

@@ -440,7 +436,7 @@ TEST_F(DBCompactionTest, TestTableReaderForCompaction) {
  Options options = CurrentOptions();
  options.env = env_;
  options.new_table_reader_for_compaction_inputs = true;
-  options.max_open_files = 20;
+  options.max_open_files = 100;
  options.level0_file_num_compaction_trigger = 3;
  DestroyAndReopen(options);
  Random rnd(301);
@@ -469,16 +465,15 @@ TEST_F(DBCompactionTest, TestTableReaderForCompaction) {
      Flush();
      dbfull()->TEST_WaitForCompact();
      // preloading iterator issues one table cache lookup and create
-      // a new table reader, if not preloaded.
-      int old_num_table_cache_lookup = num_table_cache_lookup;
-      ASSERT_GE(num_table_cache_lookup, 1);
+      // a new table reader.
+      ASSERT_EQ(num_table_cache_lookup, 1);
      ASSERT_EQ(num_new_table_reader, 1);

      num_table_cache_lookup = 0;
      num_new_table_reader = 0;
      ASSERT_EQ(Key(k), Get(Key(k)));
      // lookup iterator from table cache and no need to create a new one.
-      ASSERT_EQ(old_num_table_cache_lookup + num_table_cache_lookup, 2);
+      ASSERT_EQ(num_table_cache_lookup, 1);
      ASSERT_EQ(num_new_table_reader, 0);
    }
  }
@@ -491,10 +486,7 @@ TEST_F(DBCompactionTest, TestTableReaderForCompaction) {
  // a new table reader. One file is created for flush and one for compaction.
  // Compaction inputs make no table cache look-up for data/range deletion
  // iterators
-  // May preload table cache too.
-  ASSERT_GE(num_table_cache_lookup, 2);
-  int old_num_table_cache_lookup2 = num_table_cache_lookup;
-
+  ASSERT_EQ(num_table_cache_lookup, 2);
  // Create new iterator for:
  // (1) 1 for verifying flush results
  // (2) 3 for compaction input files
@@ -504,7 +496,7 @@ TEST_F(DBCompactionTest, TestTableReaderForCompaction) {
  num_table_cache_lookup = 0;
  num_new_table_reader = 0;
  ASSERT_EQ(Key(1), Get(Key(1)));
-  ASSERT_EQ(num_table_cache_lookup + old_num_table_cache_lookup2, 3);
+  ASSERT_EQ(num_table_cache_lookup, 1);
  ASSERT_EQ(num_new_table_reader, 0);

  num_table_cache_lookup = 0;
@@ -512,20 +504,18 @@ TEST_F(DBCompactionTest, TestTableReaderForCompaction) {
  CompactRangeOptions cro;
  cro.change_level = true;
  cro.target_level = 2;
-  cro.bottommost_level_compaction = BottommostLevelCompaction::kForceOptimized;
+  cro.bottommost_level_compaction = BottommostLevelCompaction::kForce;
  db_->CompactRange(cro, nullptr, nullptr);
  // Only verifying compaction outputs issues one table cache lookup
  // for both data block and range deletion block).
-  // May preload table cache too.
-  ASSERT_GE(num_table_cache_lookup, 1);
-  old_num_table_cache_lookup2 = num_table_cache_lookup;
+  ASSERT_EQ(num_table_cache_lookup, 1);
  // One for compaction input, one for verifying compaction results.
  ASSERT_EQ(num_new_table_reader, 2);

  num_table_cache_lookup = 0;
  num_new_table_reader = 0;
  ASSERT_EQ(Key(1), Get(Key(1)));
-  ASSERT_EQ(num_table_cache_lookup + old_num_table_cache_lookup2, 2);
+  ASSERT_EQ(num_table_cache_lookup, 1);
  ASSERT_EQ(num_new_table_reader, 0);

  rocksdb::SyncPoint::GetInstance()->ClearAllCallBacks();
@@ -1603,8 +1593,7 @@ TEST_F(DBCompactionTest, DeleteFileRange) {
  // Note that we don't delete level 0 files
  compact_options.change_level = true;
  compact_options.target_level = 1;
-  ASSERT_OK(db_->CompactRange(compact_options, nullptr, nullptr));
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(dbfull()->TEST_CompactRange(0, nullptr, nullptr));

  ASSERT_OK(
      DeleteFilesInRange(db_, db_->DefaultColumnFamily(), nullptr, nullptr));
@@ -2261,8 +2250,6 @@ TEST_P(DBCompactionTestWithParam, ConvertCompactionStyle) {
  CompactRangeOptions compact_options;
  compact_options.change_level = true;
  compact_options.target_level = 0;
-  // cannot use kForceOptimized here because the compaction here is expected
-  // to generate one output file
  compact_options.bottommost_level_compaction =
      BottommostLevelCompaction::kForce;
  compact_options.exclusive_manual_compaction = exclusive_manual_compaction_;
@@ -2491,7 +2478,6 @@ TEST_P(DBCompactionTestWithParam, ManualLevelCompactionOutputPathId) {

    // Compaction range overlaps files
    Compact(1, "p1", "p9", 1);
-    ASSERT_OK(dbfull()->TEST_WaitForCompact());
    ASSERT_EQ("0,1", FilesPerLevel(1));
    ASSERT_EQ(1, GetSstFileCount(options.db_paths[1].path));
    ASSERT_EQ(0, GetSstFileCount(options.db_paths[0].path));
@@ -2507,7 +2493,6 @@ TEST_P(DBCompactionTestWithParam, ManualLevelCompactionOutputPathId) {

    // Compact just the new range
    Compact(1, "b", "f", 1);
-    ASSERT_OK(dbfull()->TEST_WaitForCompact());
    ASSERT_EQ("0,2", FilesPerLevel(1));
    ASSERT_EQ(2, GetSstFileCount(options.db_paths[1].path));
    ASSERT_EQ(0, GetSstFileCount(options.db_paths[0].path));
@@ -2524,7 +2509,6 @@ TEST_P(DBCompactionTestWithParam, ManualLevelCompactionOutputPathId) {
    compact_options.target_path_id = 1;
    compact_options.exclusive_manual_compaction = exclusive_manual_compaction_;
    db_->CompactRange(compact_options, handles_[1], nullptr, nullptr);
-    ASSERT_OK(dbfull()->TEST_WaitForCompact());

    ASSERT_EQ("0,1", FilesPerLevel(1));
    ASSERT_EQ(1, GetSstFileCount(options.db_paths[1].path));
@@ -2967,12 +2951,6 @@ TEST_F(DBCompactionTest, SuggestCompactRangeNoTwoLevel0Compactions) {
  dbfull()->TEST_WaitForCompact();
 }

-static std::string ShortKey(int i) {
-  assert(i < 10000);
-  char buf[100];
-  snprintf(buf, sizeof(buf), "key%04d", i);
-  return std::string(buf);
-}

 TEST_P(DBCompactionTestWithParam, ForceBottommostLevelCompaction) {
  int32_t trivial_move = 0;
@@ -2985,28 +2963,10 @@ TEST_P(DBCompactionTestWithParam, ForceBottommostLevelCompaction) {
      [&](void* /*arg*/) { non_trivial_move++; });
  rocksdb::SyncPoint::GetInstance()->EnableProcessing();

-  // The key size is guaranteed to be <= 8
-  class ShortKeyComparator : public Comparator {
-    int Compare(const rocksdb::Slice& a,
-                const rocksdb::Slice& b) const override {
-      assert(a.size() <= 8);
-      assert(b.size() <= 8);
-      return BytewiseComparator()->Compare(a, b);
-    }
-    const char* Name() const override { return "ShortKeyComparator"; }
-    void FindShortestSeparator(std::string* start,
-                               const rocksdb::Slice& limit) const override {
-      return BytewiseComparator()->FindShortestSeparator(start, limit);
-    }
-    void FindShortSuccessor(std::string* key) const override {
-      return BytewiseComparator()->FindShortSuccessor(key);
-    }
-  } short_key_cmp;
  Options options = CurrentOptions();
  options.target_file_size_base = 100000000;
  options.write_buffer_size = 100000000;
  options.max_subcompactions = max_subcompactions_;
-  options.comparator = &short_key_cmp;
  DestroyAndReopen(options);

  int32_t value_size = 10 * 1024;  // 10 KB
@@ -3016,7 +2976,7 @@ TEST_P(DBCompactionTestWithParam, ForceBottommostLevelCompaction) {
  // File with keys [ 0 => 99 ]
  for (int i = 0; i < 100; i++) {
    values.push_back(RandomString(&rnd, value_size));
-    ASSERT_OK(Put(ShortKey(i), values[i]));
+    ASSERT_OK(Put(Key(i), values[i]));
  }
  ASSERT_OK(Flush());

@@ -3033,7 +2993,7 @@ TEST_P(DBCompactionTestWithParam, ForceBottommostLevelCompaction) {
  // File with keys [ 100 => 199 ]
  for (int i = 100; i < 200; i++) {
    values.push_back(RandomString(&rnd, value_size));
-    ASSERT_OK(Put(ShortKey(i), values[i]));
+    ASSERT_OK(Put(Key(i), values[i]));
  }
  ASSERT_OK(Flush());

@@ -3042,7 +3002,7 @@ TEST_P(DBCompactionTestWithParam, ForceBottommostLevelCompaction) {
  // then compacte the bottommost level L3=>L3 (non trivial move)
  compact_options = CompactRangeOptions();
  compact_options.bottommost_level_compaction =
-      BottommostLevelCompaction::kForceOptimized;
+      BottommostLevelCompaction::kForce;
  ASSERT_OK(db_->CompactRange(compact_options, nullptr, nullptr));
  ASSERT_EQ("0,0,0,1", FilesPerLevel(0));
  ASSERT_EQ(trivial_move, 4);
@@ -3051,7 +3011,7 @@ TEST_P(DBCompactionTestWithParam, ForceBottommostLevelCompaction) {
  // File with keys [ 200 => 299 ]
  for (int i = 200; i < 300; i++) {
    values.push_back(RandomString(&rnd, value_size));
-    ASSERT_OK(Put(ShortKey(i), values[i]));
+    ASSERT_OK(Put(Key(i), values[i]));
  }
  ASSERT_OK(Flush());

@@ -3069,7 +3029,7 @@ TEST_P(DBCompactionTestWithParam, ForceBottommostLevelCompaction) {
  ASSERT_EQ(non_trivial_move, 0);

  for (int i = 0; i < 300; i++) {
-    ASSERT_EQ(Get(ShortKey(i)), values[i]);
+    ASSERT_EQ(Get(Key(i)), values[i]);
  }

  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
@@ -3351,7 +3311,7 @@ TEST_F(DBCompactionTest, CompactBottomLevelFilesWithDeletions) {
  options.level0_file_num_compaction_trigger = kNumLevelFiles;
  // inflate it a bit to account for key/metadata overhead
  options.target_file_size_base = 120 * kNumKeysPerFile * kValueSize / 100;
-  CreateAndReopenWithCF({"one"}, options);
+  Reopen(options);

  Random rnd(301);
  const Snapshot* snapshot = nullptr;
@@ -3382,12 +3342,10 @@ TEST_F(DBCompactionTest, CompactBottomLevelFilesWithDeletions) {
  // just need to bump seqnum so ReleaseSnapshot knows the newest key in the SST
  // files does not need to be preserved in case of a future snapshot.
  ASSERT_OK(Put(Key(0), "val"));
-  ASSERT_NE(kMaxSequenceNumber, dbfull()->bottommost_files_mark_threshold_);
  // release snapshot and wait for compactions to finish. Single-file
  // compactions should be triggered, which reduce the size of each bottom-level
  // file without changing file count.
  db_->ReleaseSnapshot(snapshot);
-  ASSERT_EQ(kMaxSequenceNumber, dbfull()->bottommost_files_mark_threshold_);
  rocksdb::SyncPoint::GetInstance()->SetCallBack(
      "LevelCompactionPicker::PickCompaction:Return", [&](void* arg) {
        Compaction* compaction = reinterpret_cast<Compaction*>(arg);
@@ -3520,234 +3478,6 @@ TEST_F(DBCompactionTest, LevelCompactExpiredTtlFiles) {
  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
 }

-TEST_F(DBCompactionTest, LevelPeriodicCompaction) {
-  const int kNumKeysPerFile = 32;
-  const int kNumLevelFiles = 2;
-  const int kValueSize = 100;
-
-  Options options = CurrentOptions();
-  options.periodic_compaction_seconds = 48 * 60 * 60;  // 2 days
-  options.max_open_files = -1;       // needed for ttl compaction
-  env_->time_elapse_only_sleep_ = false;
-  options.env = env_;
-
-  env_->addon_time_.store(0);
-  DestroyAndReopen(options);
-
-  int periodic_compactions = 0;
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
-      "LevelCompactionPicker::PickCompaction:Return", [&](void* arg) {
-        Compaction* compaction = reinterpret_cast<Compaction*>(arg);
-        auto compaction_reason = compaction->compaction_reason();
-        if (compaction_reason == CompactionReason::kPeriodicCompaction) {
-          periodic_compactions++;
-        }
-      });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-
-  Random rnd(301);
-  for (int i = 0; i < kNumLevelFiles; ++i) {
-    for (int j = 0; j < kNumKeysPerFile; ++j) {
-      ASSERT_OK(
-          Put(Key(i * kNumKeysPerFile + j), RandomString(&rnd, kValueSize)));
-    }
-    Flush();
-  }
-  dbfull()->TEST_WaitForCompact();
-
-  ASSERT_EQ("2", FilesPerLevel());
-  ASSERT_EQ(0, periodic_compactions);
-
-  // Add 50 hours and do a write
-  env_->addon_time_.fetch_add(50 * 60 * 60);
-  ASSERT_OK(Put("a", "1"));
-  Flush();
-  dbfull()->TEST_WaitForCompact();
-  // Assert that the files stay in the same level
-  ASSERT_EQ("3", FilesPerLevel());
-  // The two old files go through the periodic compaction process
-  ASSERT_EQ(2, periodic_compactions);
-
-  MoveFilesToLevel(1);
-  ASSERT_EQ("0,3", FilesPerLevel());
-
-  // Add another 50 hours and do another write
-  env_->addon_time_.fetch_add(50 * 60 * 60);
-  ASSERT_OK(Put("b", "2"));
-  Flush();
-  dbfull()->TEST_WaitForCompact();
-  ASSERT_EQ("1,3", FilesPerLevel());
-  // The three old files now go through the periodic compaction process. 2 + 3.
-  ASSERT_EQ(5, periodic_compactions);
-
-  // Add another 50 hours and do another write
-  env_->addon_time_.fetch_add(50 * 60 * 60);
-  ASSERT_OK(Put("c", "3"));
-  Flush();
-  dbfull()->TEST_WaitForCompact();
-  ASSERT_EQ("2,3", FilesPerLevel());
-  // The four old files now go through the periodic compaction process. 5 + 4.
-  ASSERT_EQ(9, periodic_compactions);
-
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-}
-
-TEST_F(DBCompactionTest, LevelPeriodicCompactionWithOldDB) {
-  // This test makes sure that periodic compactions are working with a DB
-  // where file_creation_time of some files is 0.
-  // After compactions the new files are created with a valid file_creation_time
-
-  const int kNumKeysPerFile = 32;
-  const int kNumFiles = 4;
-  const int kValueSize = 100;
-
-  Options options = CurrentOptions();
-  options.max_open_files = -1;  // needed for ttl compaction
-  env_->time_elapse_only_sleep_ = false;
-  options.env = env_;
-
-  env_->addon_time_.store(0);
-  DestroyAndReopen(options);
-
-  int periodic_compactions = 0;
-  bool set_file_creation_time_to_zero = true;
-  bool set_creation_time_to_zero = true;
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
-      "LevelCompactionPicker::PickCompaction:Return", [&](void* arg) {
-        Compaction* compaction = reinterpret_cast<Compaction*>(arg);
-        auto compaction_reason = compaction->compaction_reason();
-        if (compaction_reason == CompactionReason::kPeriodicCompaction) {
-          periodic_compactions++;
-        }
-      });
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
-      "PropertyBlockBuilder::AddTableProperty:Start", [&](void* arg) {
-        TableProperties* props = reinterpret_cast<TableProperties*>(arg);
-        if (set_file_creation_time_to_zero) {
-          props->file_creation_time = 0;
-        }
-        if (set_creation_time_to_zero) {
-          props->creation_time = 0;
-        }
-      });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-
-  Random rnd(301);
-  for (int i = 0; i < kNumFiles; ++i) {
-    for (int j = 0; j < kNumKeysPerFile; ++j) {
-      ASSERT_OK(
-          Put(Key(i * kNumKeysPerFile + j), RandomString(&rnd, kValueSize)));
-    }
-    Flush();
-    // Move the first two files to L2.
-    if (i == 1) {
-      MoveFilesToLevel(2);
-      set_creation_time_to_zero = false;
-    }
-  }
-  ASSERT_OK(dbfull()->TEST_WaitForCompact());
-
-  ASSERT_EQ("2,0,2", FilesPerLevel());
-  ASSERT_EQ(0, periodic_compactions);
-
-  Close();
-
-  set_file_creation_time_to_zero = false;
-  // Forward the clock by 2 days.
-  env_->addon_time_.fetch_add(2 * 24 * 60 * 60);
-  options.periodic_compaction_seconds = 1 * 24 * 60 * 60;  // 1 day
-
-  Reopen(options);
-  ASSERT_OK(dbfull()->TEST_WaitForCompact());
-  ASSERT_EQ("2,0,2", FilesPerLevel());
-  // Make sure that all files go through periodic compaction.
-  ASSERT_EQ(kNumFiles, periodic_compactions);
-
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-}
-
-TEST_F(DBCompactionTest, LevelPeriodicAndTtlCompaction) {
-  const int kNumKeysPerFile = 32;
-  const int kNumLevelFiles = 2;
-  const int kValueSize = 100;
-
-  Options options = CurrentOptions();
-  options.ttl = 10 * 60 * 60;  // 10 hours
-  options.periodic_compaction_seconds = 48 * 60 * 60;  // 2 days
-  options.max_open_files = -1;   // needed for both periodic and ttl compactions
-  env_->time_elapse_only_sleep_ = false;
-  options.env = env_;
-
-  env_->addon_time_.store(0);
-  DestroyAndReopen(options);
-
-  int periodic_compactions = 0;
-  int ttl_compactions = 0;
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
-      "LevelCompactionPicker::PickCompaction:Return", [&](void* arg) {
-        Compaction* compaction = reinterpret_cast<Compaction*>(arg);
-        auto compaction_reason = compaction->compaction_reason();
-        if (compaction_reason == CompactionReason::kPeriodicCompaction) {
-          periodic_compactions++;
-        } else if (compaction_reason == CompactionReason::kTtl) {
-          ttl_compactions++;
-        }
-      });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-
-  Random rnd(301);
-  for (int i = 0; i < kNumLevelFiles; ++i) {
-    for (int j = 0; j < kNumKeysPerFile; ++j) {
-      ASSERT_OK(
-          Put(Key(i * kNumKeysPerFile + j), RandomString(&rnd, kValueSize)));
-    }
-    Flush();
-  }
-  dbfull()->TEST_WaitForCompact();
-
-  MoveFilesToLevel(3);
-
-  ASSERT_EQ("0,0,0,2", FilesPerLevel());
-  ASSERT_EQ(0, periodic_compactions);
-  ASSERT_EQ(0, ttl_compactions);
-
-  // Add some time greater than periodic_compaction_time.
-  env_->addon_time_.fetch_add(50 * 60 * 60);
-  ASSERT_OK(Put("a", "1"));
-  Flush();
-  dbfull()->TEST_WaitForCompact();
-  // Files in the bottom level go through periodic compactions.
-  ASSERT_EQ("1,0,0,2", FilesPerLevel());
-  ASSERT_EQ(2, periodic_compactions);
-  ASSERT_EQ(0, ttl_compactions);
-
-  // Add a little more time than ttl
-  env_->addon_time_.fetch_add(11 * 60 * 60);
-  ASSERT_OK(Put("b", "1"));
-  Flush();
-  dbfull()->TEST_WaitForCompact();
-  // Notice that the previous file in level 1 falls down to the bottom level
-  // due to ttl compactions, one level at a time.
-  // And bottom level files don't get picked up for ttl compactions.
-  ASSERT_EQ("1,0,0,3", FilesPerLevel());
-  ASSERT_EQ(2, periodic_compactions);
-  ASSERT_EQ(3, ttl_compactions);
-
-  // Add some time greater than periodic_compaction_time.
-  env_->addon_time_.fetch_add(50 * 60 * 60);
-  ASSERT_OK(Put("c", "1"));
-  Flush();
-  dbfull()->TEST_WaitForCompact();
-  // Previous L0 file falls one level at a time to bottom level due to ttl.
-  // And all 4 bottom files go through periodic compactions.
-  ASSERT_EQ("1,0,0,4", FilesPerLevel());
-  ASSERT_EQ(6, periodic_compactions);
-  ASSERT_EQ(6, ttl_compactions);
-
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-}
-
-
 TEST_F(DBCompactionTest, CompactRangeDelayedByL0FileCount) {
  // Verify that, when `CompactRangeOptions::allow_write_stall == false`, manual
  // compaction only triggers flush after it's sure stall won't be triggered for
@@ -4132,192 +3862,6 @@ TEST_F(DBCompactionTest, CompactionHasEmptyOutput) {
  ASSERT_EQ(2, collector->num_ssts_creation_started());
 }

-TEST_F(DBCompactionTest, CompactionLimiter) {
-  const int kNumKeysPerFile = 10;
-  const int kMaxBackgroundThreads = 64;
-
-  struct CompactionLimiter {
-    std::string name;
-    int limit_tasks;
-    int max_tasks;
-    int tasks;
-    std::shared_ptr<ConcurrentTaskLimiter> limiter;
-  };
-
-  std::vector<CompactionLimiter> limiter_settings;
-  limiter_settings.push_back({"limiter_1", 1, 0, 0, nullptr});
-  limiter_settings.push_back({"limiter_2", 2, 0, 0, nullptr});
-  limiter_settings.push_back({"limiter_3", 3, 0, 0, nullptr});
-
-  for (auto& ls : limiter_settings) {
-    ls.limiter.reset(NewConcurrentTaskLimiter(ls.name, ls.limit_tasks));
-  }
-
-  std::shared_ptr<ConcurrentTaskLimiter> unique_limiter(
-    NewConcurrentTaskLimiter("unique_limiter", -1));
-
-  const char* cf_names[] = {"default", "0", "1", "2", "3", "4", "5",
-    "6", "7", "8", "9", "a", "b", "c", "d", "e", "f" };
-  const int cf_count = sizeof cf_names / sizeof cf_names[0];
-
-  std::unordered_map<std::string, CompactionLimiter*> cf_to_limiter;
-
-  Options options = CurrentOptions();
-  options.write_buffer_size = 110 * 1024;  // 110KB
-  options.arena_block_size = 4096;
-  options.num_levels = 3;
-  options.level0_file_num_compaction_trigger = 4;
-  options.level0_slowdown_writes_trigger = 64;
-  options.level0_stop_writes_trigger = 64;
-  options.max_background_jobs = kMaxBackgroundThreads; // Enough threads
-  options.memtable_factory.reset(new SpecialSkipListFactory(kNumKeysPerFile));
-  options.max_write_buffer_number = 10; // Enough memtables
-  DestroyAndReopen(options);
-
-  std::vector<Options> option_vector;
-  option_vector.reserve(cf_count);
-
-  for (int cf = 0; cf < cf_count; cf++) {
-    ColumnFamilyOptions cf_opt(options);
-    if (cf == 0) {
-      // "Default" CF does't use compaction limiter
-      cf_opt.compaction_thread_limiter = nullptr;
-    } else if (cf == 1) {
-      // "1" CF uses bypass compaction limiter
-      unique_limiter->SetMaxOutstandingTask(-1);
-      cf_opt.compaction_thread_limiter = unique_limiter;
-    } else {
-      // Assign limiter by mod
-      auto& ls = limiter_settings[cf % 3];
-      cf_opt.compaction_thread_limiter = ls.limiter;
-      cf_to_limiter[cf_names[cf]] = &ls;
-    }
-    option_vector.emplace_back(DBOptions(options), cf_opt);
-  }
-
-  for (int cf = 1; cf < cf_count; cf++) {
-    CreateColumnFamilies({cf_names[cf]}, option_vector[cf]);
-  }
-
-  ReopenWithColumnFamilies(std::vector<std::string>(cf_names,
-                                                    cf_names + cf_count),
-                           option_vector);
-
-  port::Mutex mutex;
-
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
-    "DBImpl::BackgroundCompaction:BeforeCompaction", [&](void* arg) {
-      const auto& cf_name = static_cast<ColumnFamilyData*>(arg)->GetName();
-      auto iter = cf_to_limiter.find(cf_name);
-      if (iter != cf_to_limiter.end()) {
-        MutexLock l(&mutex);
-        ASSERT_GE(iter->second->limit_tasks, ++iter->second->tasks);
-        iter->second->max_tasks = std::max(iter->second->max_tasks,
-              iter->second->limit_tasks);
-      }
-    });
-
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
-    "DBImpl::BackgroundCompaction:AfterCompaction", [&](void* arg) {
-      const auto& cf_name = static_cast<ColumnFamilyData*>(arg)->GetName();
-      auto iter = cf_to_limiter.find(cf_name);
-      if (iter != cf_to_limiter.end()) {
-        MutexLock l(&mutex);
-        ASSERT_GE(--iter->second->tasks, 0);
-      }
-    });
-
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-
-  // Block all compact threads in thread pool.
-  const size_t kTotalFlushTasks = kMaxBackgroundThreads / 4;
-  const size_t kTotalCompactTasks = kMaxBackgroundThreads - kTotalFlushTasks;
-  env_->SetBackgroundThreads((int)kTotalFlushTasks, Env::HIGH);
-  env_->SetBackgroundThreads((int)kTotalCompactTasks, Env::LOW);
-
-  test::SleepingBackgroundTask sleeping_compact_tasks[kTotalCompactTasks];
-
-  // Block all compaction threads in thread pool.
-  for (size_t i = 0; i < kTotalCompactTasks; i++) {
-    env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask,
-                   &sleeping_compact_tasks[i], Env::LOW);
-    sleeping_compact_tasks[i].WaitUntilSleeping();
-  }
-
-  int keyIndex = 0;
-
-  for (int n = 0; n < options.level0_file_num_compaction_trigger; n++) {
-    for (int cf = 0; cf < cf_count; cf++) {
-      for (int i = 0; i < kNumKeysPerFile; i++) {
-        ASSERT_OK(Put(cf, Key(keyIndex++), ""));
-      }
-      // put extra key to trigger flush
-      ASSERT_OK(Put(cf, "", ""));
-    }
-
-    for (int cf = 0; cf < cf_count; cf++) {
-      dbfull()->TEST_WaitForFlushMemTable(handles_[cf]);
-    }
-  }
-
-  // Enough L0 files to trigger compaction
-  for (int cf = 0; cf < cf_count; cf++) {
-    ASSERT_EQ(NumTableFilesAtLevel(0, cf),
-      options.level0_file_num_compaction_trigger);
-  }
-
-  // Create more files for one column family, which triggers speed up
-  // condition, all compactions will be scheduled.
-  for (int num = 0; num < options.level0_file_num_compaction_trigger; num++) {
-    for (int i = 0; i < kNumKeysPerFile; i++) {
-      ASSERT_OK(Put(0, Key(i), ""));
-    }
-    // put extra key to trigger flush
-    ASSERT_OK(Put(0, "", ""));
-    dbfull()->TEST_WaitForFlushMemTable(handles_[0]);
-    ASSERT_EQ(options.level0_file_num_compaction_trigger + num + 1,
-              NumTableFilesAtLevel(0, 0));
-  }
-
-  // All CFs are pending compaction
-  ASSERT_EQ(cf_count, env_->GetThreadPoolQueueLen(Env::LOW));
-
-  // Unblock all compaction threads
-  for (size_t i = 0; i < kTotalCompactTasks; i++) {
-    sleeping_compact_tasks[i].WakeUp();
-    sleeping_compact_tasks[i].WaitUntilDone();
-  }
-
-  for (int cf = 0; cf < cf_count; cf++) {
-    dbfull()->TEST_WaitForFlushMemTable(handles_[cf]);
-  }
-
-  ASSERT_OK(dbfull()->TEST_WaitForCompact());
-
-  // Max outstanding compact tasks reached limit
-  for (auto& ls : limiter_settings) {
-    ASSERT_EQ(ls.limit_tasks, ls.max_tasks);
-    ASSERT_EQ(0, ls.limiter->GetOutstandingTask());
-  }
-
-  // test manual compaction under a fully throttled limiter
-  int cf_test = 1;
-  unique_limiter->SetMaxOutstandingTask(0);
-
-  // flush one more file to cf 1
-  for (int i = 0; i < kNumKeysPerFile; i++) {
-      ASSERT_OK(Put(cf_test, Key(keyIndex++), ""));
-  }
-  // put extra key to trigger flush
-  ASSERT_OK(Put(cf_test, "", ""));
-
-  dbfull()->TEST_WaitForFlushMemTable(handles_[cf_test]);
-  ASSERT_EQ(1, NumTableFilesAtLevel(0, cf_test));
-
-  Compact(cf_test, Key(0), Key(keyIndex));
-  ASSERT_OK(dbfull()->TEST_WaitForCompact());
-}
-
 INSTANTIATE_TEST_CASE_P(DBCompactionTestWithParam, DBCompactionTestWithParam,
                        ::testing::Values(std::make_tuple(1, true),
                                          std::make_tuple(1, false),
@@ -4419,14 +3963,14 @@ class NoopMergeOperator : public MergeOperator {
 public:
  NoopMergeOperator() {}

-  bool FullMergeV2(const MergeOperationInput& /*merge_in*/,
-                   MergeOperationOutput* merge_out) const override {
+  virtual bool FullMergeV2(const MergeOperationInput& /*merge_in*/,
+                           MergeOperationOutput* merge_out) const override {
    std::string val("bar");
    merge_out->new_value = val;
    return true;
  }

-  const char* Name() const override { return "Noop"; }
+  virtual const char* Name() const override { return "Noop"; }
 };

 TEST_F(DBCompactionTest, PartialManualCompaction) {
@@ -4455,179 +3999,10 @@ TEST_F(DBCompactionTest, PartialManualCompaction) {
      {{"max_compaction_bytes", std::to_string(max_compaction_bytes)}}));

  CompactRangeOptions cro;
-  cro.bottommost_level_compaction = BottommostLevelCompaction::kForceOptimized;
+  cro.bottommost_level_compaction = BottommostLevelCompaction::kForce;
  dbfull()->CompactRange(cro, nullptr, nullptr);
 }

-TEST_F(DBCompactionTest, ManualCompactionFailsInReadOnlyMode) {
-  // Regression test for bug where manual compaction hangs forever when the DB
-  // is in read-only mode. Verify it now at least returns, despite failing.
-  const int kNumL0Files = 4;
-  std::unique_ptr<FaultInjectionTestEnv> mock_env(
-      new FaultInjectionTestEnv(Env::Default()));
-  Options opts = CurrentOptions();
-  opts.disable_auto_compactions = true;
-  opts.env = mock_env.get();
-  DestroyAndReopen(opts);
-
-  Random rnd(301);
-  for (int i = 0; i < kNumL0Files; ++i) {
-    // Make sure files are overlapping in key-range to prevent trivial move.
-    Put("key1", RandomString(&rnd, 1024));
-    Put("key2", RandomString(&rnd, 1024));
-    Flush();
-  }
-  ASSERT_EQ(kNumL0Files, NumTableFilesAtLevel(0));
-
-  // Enter read-only mode by failing a write.
-  mock_env->SetFilesystemActive(false);
-  // Make sure this is outside `CompactRange`'s range so that it doesn't fail
-  // early trying to flush memtable.
-  ASSERT_NOK(Put("key3", RandomString(&rnd, 1024)));
-
-  // In the bug scenario, the first manual compaction would fail and forget to
-  // unregister itself, causing the second one to hang forever due to conflict
-  // with a non-running compaction.
-  CompactRangeOptions cro;
-  cro.exclusive_manual_compaction = false;
-  Slice begin_key("key1");
-  Slice end_key("key2");
-  ASSERT_NOK(dbfull()->CompactRange(cro, &begin_key, &end_key));
-  ASSERT_NOK(dbfull()->CompactRange(cro, &begin_key, &end_key));
-
-  // Close before mock_env destruct.
-  Close();
-}
-
-// ManualCompactionBottomLevelOptimization tests the bottom level manual
-// compaction optimization to skip recompacting files created by Ln-1 to Ln
-// compaction
-TEST_F(DBCompactionTest, ManualCompactionBottomLevelOptimized) {
-  Options opts = CurrentOptions();
-  opts.num_levels = 3;
-  opts.level0_file_num_compaction_trigger = 5;
-  opts.compression = kNoCompression;
-  opts.merge_operator.reset(new NoopMergeOperator());
-  opts.target_file_size_base = 1024;
-  opts.max_bytes_for_level_multiplier = 2;
-  opts.disable_auto_compactions = true;
-  DestroyAndReopen(opts);
-  ColumnFamilyHandleImpl* cfh =
-      static_cast<ColumnFamilyHandleImpl*>(dbfull()->DefaultColumnFamily());
-  ColumnFamilyData* cfd = cfh->cfd();
-  InternalStats* internal_stats_ptr = cfd->internal_stats();
-  ASSERT_NE(internal_stats_ptr, nullptr);
-
-  Random rnd(301);
-  for (auto i = 0; i < 8; ++i) {
-    for (auto j = 0; j < 10; ++j) {
-      ASSERT_OK(
-          Put("foo" + std::to_string(i * 10 + j), RandomString(&rnd, 1024)));
-    }
-    Flush();
-  }
-
-  MoveFilesToLevel(2);
-
-  for (auto i = 0; i < 8; ++i) {
-    for (auto j = 0; j < 10; ++j) {
-      ASSERT_OK(
-          Put("bar" + std::to_string(i * 10 + j), RandomString(&rnd, 1024)));
-    }
-    Flush();
-  }
-  const std::vector<InternalStats::CompactionStats>& comp_stats =
-      internal_stats_ptr->TEST_GetCompactionStats();
-  int num = comp_stats[2].num_input_files_in_output_level;
-  ASSERT_EQ(num, 0);
-
-  CompactRangeOptions cro;
-  cro.bottommost_level_compaction = BottommostLevelCompaction::kForceOptimized;
-  dbfull()->CompactRange(cro, nullptr, nullptr);
-
-  const std::vector<InternalStats::CompactionStats>& comp_stats2 =
-      internal_stats_ptr->TEST_GetCompactionStats();
-  num = comp_stats2[2].num_input_files_in_output_level;
-  ASSERT_EQ(num, 0);
-}
-
-// FixFileIngestionCompactionDeadlock tests and verifies that compaction and
-// file ingestion do not cause deadlock in the event of write stall triggered
-// by number of L0 files reaching level0_stop_writes_trigger.
-TEST_P(DBCompactionTestWithParam, FixFileIngestionCompactionDeadlock) {
-  const int kNumKeysPerFile = 100;
-  // Generate SST files.
-  Options options = CurrentOptions();
-
-  // Generate an external SST file containing a single key, i.e. 99
-  std::string sst_files_dir = dbname_ + "/sst_files/";
-  test::DestroyDir(env_, sst_files_dir);
-  ASSERT_OK(env_->CreateDir(sst_files_dir));
-  SstFileWriter sst_writer(EnvOptions(), options);
-  const std::string sst_file_path = sst_files_dir + "test.sst";
-  ASSERT_OK(sst_writer.Open(sst_file_path));
-  ASSERT_OK(sst_writer.Put(Key(kNumKeysPerFile - 1), "value"));
-  ASSERT_OK(sst_writer.Finish());
-
-  SyncPoint::GetInstance()->DisableProcessing();
-  SyncPoint::GetInstance()->ClearAllCallBacks();
-  SyncPoint::GetInstance()->LoadDependency({
-      {"DBImpl::IngestExternalFile:AfterIncIngestFileCounter",
-       "BackgroundCallCompaction:0"},
-  });
-  SyncPoint::GetInstance()->EnableProcessing();
-
-  options.write_buffer_size = 110 << 10;  // 110KB
-  options.level0_file_num_compaction_trigger =
-      options.level0_stop_writes_trigger;
-  options.max_subcompactions = max_subcompactions_;
-  options.memtable_factory.reset(new SpecialSkipListFactory(kNumKeysPerFile));
-  DestroyAndReopen(options);
-  Random rnd(301);
-
-  // Generate level0_stop_writes_trigger L0 files to trigger write stop
-  for (int i = 0; i != options.level0_file_num_compaction_trigger; ++i) {
-    for (int j = 0; j != kNumKeysPerFile; ++j) {
-      ASSERT_OK(Put(Key(j), RandomString(&rnd, 990)));
-    }
-    if (0 == i) {
-      // When we reach here, the memtables have kNumKeysPerFile keys. Note that
-      // flush is not yet triggered. We need to write an extra key so that the
-      // write path will call PreprocessWrite and flush the previous key-value
-      // pairs to e flushed. After that, there will be the newest key in the
-      // memtable, and a bunch of L0 files. Since there is already one key in
-      // the memtable, then for i = 1, 2, ..., we do not have to write this
-      // extra key to trigger flush.
-      ASSERT_OK(Put("", ""));
-    }
-    dbfull()->TEST_WaitForFlushMemTable();
-    ASSERT_EQ(NumTableFilesAtLevel(0 /*level*/, 0 /*cf*/), i + 1);
-  }
-  // When we reach this point, there will be level0_stop_writes_trigger L0
-  // files and one extra key (99) in memory, which overlaps with the external
-  // SST file. Write stall triggers, and can be cleared only after compaction
-  // reduces the number of L0 files.
-
-  // Compaction will also be triggered since we have reached the threshold for
-  // auto compaction. Note that compaction may begin after the following file
-  // ingestion thread and waits for ingestion to finish.
-
-  // Thread to ingest file with overlapping key range with the current
-  // memtable. Consequently ingestion will trigger a flush. The flush MUST
-  // proceed without waiting for the write stall condition to clear, otherwise
-  // deadlock can happen.
-  port::Thread ingestion_thr([&]() {
-    IngestExternalFileOptions ifo;
-    Status s = db_->IngestExternalFile({sst_file_path}, ifo);
-    ASSERT_OK(s);
-  });
-
-  // More write to trigger write stop
-  ingestion_thr.join();
-  ASSERT_OK(dbfull()->TEST_WaitForCompact());
-  Close();
-}
-
 #endif // !defined(ROCKSDB_LITE)
 }  // namespace rocksdb

@@ -27,7 +27,7 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase) {
    return;
  }
  // Use InMemoryEnv, or it would be too slow.
-  std::unique_ptr<Env> env(new MockEnv(env_));
+  unique_ptr<Env> env(new MockEnv(env_));

  const int kNKeys = 1000;
  int keys[kNKeys];
@@ -125,7 +125,6 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
  int kMaxKey = 1000000;

  Options options = CurrentOptions();
-  options.compression = kNoCompression;
  options.create_if_missing = true;
  options.write_buffer_size = 20480;
  options.max_write_buffer_number = 2;
@@ -168,8 +167,8 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
  ASSERT_TRUE(db_->GetIntProperty("rocksdb.base-level", &int_prop));
  ASSERT_EQ(4U, int_prop);

-  // Insert extra about 28K to L0. After they are compacted to L4, the base
-  // level should be changed to L3.
+  // Insert extra about 28K to L0. After they are compacted to L4, base level
+  // should be changed to L3.
  ASSERT_OK(dbfull()->SetOptions({
      {"disable_auto_compactions", "true"},
  }));
@@ -190,7 +189,13 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
  ASSERT_TRUE(db_->GetProperty("rocksdb.num-files-at-level2", &str_prop));
  ASSERT_EQ("0", str_prop);

-  // Write even more data while leaving the base level at L3.
+  // Trigger parallel compaction, and the first one would change the base
+  // level.
+  // Hold compaction jobs to make sure
+  rocksdb::SyncPoint::GetInstance()->SetCallBack(
+      "CompactionJob::Run():Start",
+      [&](void* /*arg*/) { env_->SleepForMicroseconds(100000); });
+  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
  ASSERT_OK(dbfull()->SetOptions({
      {"disable_auto_compactions", "true"},
  }));
@@ -203,12 +208,18 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
      {"disable_auto_compactions", "false"},
  }));
  Flush();
+  // Wait for 200 milliseconds before proceeding compactions to make sure two
+  // parallel ones are executed.
+  env_->SleepForMicroseconds(200000);
  dbfull()->TEST_WaitForCompact();
  ASSERT_TRUE(db_->GetIntProperty("rocksdb.base-level", &int_prop));
  ASSERT_EQ(3U, int_prop);
+  rocksdb::SyncPoint::GetInstance()->DisableProcessing();

-  // Fill up L0, and then run an (auto) L0->Lmax compaction to raise the base
-  // level to 2.
+  // Trigger a condition that the compaction changes base level and L0->Lbase
+  // happens at the same time.
+  // We try to make last levels' targets to be 40K, 160K, 640K, add triggers
+  // another compaction from 40K->160K.
  ASSERT_OK(dbfull()->SetOptions({
      {"disable_auto_compactions", "true"},
  }));
@@ -218,31 +229,23 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
    ASSERT_OK(Put(Key(static_cast<int>(rnd.Uniform(kMaxKey))),
                  RandomString(&rnd, 380)));
  }
-
-  // Make sure that the compaction starts before the last bit of data is
-  // flushed, so that the base level isn't raised to L1.
-  rocksdb::SyncPoint::GetInstance()->LoadDependency({
-      {"CompactionJob::Run():Start", "DynamicLevelMaxBytesBase2:0"},
-  });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-
  ASSERT_OK(dbfull()->SetOptions({
      {"disable_auto_compactions", "false"},
  }));
-
-  TEST_SYNC_POINT("DynamicLevelMaxBytesBase2:0");
  Flush();
  dbfull()->TEST_WaitForCompact();
  ASSERT_TRUE(db_->GetIntProperty("rocksdb.base-level", &int_prop));
  ASSERT_EQ(2U, int_prop);
+
+  // A manual compaction will trigger the base level to become L2
+  // Keep Writing data until base level changed 2->1. There will be L0->L2
+  // compaction going on at the same time.
  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
  rocksdb::SyncPoint::GetInstance()->ClearAllCallBacks();

-  // Write more data until the base level changes to L1. There will be
-  // a manual compaction going on at the same time.
  rocksdb::SyncPoint::GetInstance()->LoadDependency({
-      {"CompactionJob::Run():Start", "DynamicLevelMaxBytesBase2:1"},
-      {"DynamicLevelMaxBytesBase2:2", "CompactionJob::Run():End"},
+      {"CompactionJob::Run():Start", "DynamicLevelMaxBytesBase2:0"},
+      {"DynamicLevelMaxBytesBase2:1", "CompactionJob::Run():End"},
      {"DynamicLevelMaxBytesBase2:compact_range_finish",
       "FlushJob::WriteLevel0Table"},
  });
@@ -254,12 +257,12 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
    TEST_SYNC_POINT("DynamicLevelMaxBytesBase2:compact_range_finish");
  });

-  TEST_SYNC_POINT("DynamicLevelMaxBytesBase2:1");
+  TEST_SYNC_POINT("DynamicLevelMaxBytesBase2:0");
  for (int i = 0; i < 2; i++) {
    ASSERT_OK(Put(Key(static_cast<int>(rnd.Uniform(kMaxKey))),
                  RandomString(&rnd, 380)));
  }
-  TEST_SYNC_POINT("DynamicLevelMaxBytesBase2:2");
+  TEST_SYNC_POINT("DynamicLevelMaxBytesBase2:1");

  Flush();

@@ -40,7 +40,7 @@ TEST_F(DBEncryptionTest, CheckEncrypted) {
      continue;
    }
    auto filePath = dbname_ + "/" + *it;
-    std::unique_ptr<SequentialFile> seqFile;
+    unique_ptr<SequentialFile> seqFile;
    auto envOptions = EnvOptions(CurrentOptions());
    status = defaultEnv->NewSequentialFile(filePath, &seqFile, envOptions);
    ASSERT_OK(status);
@@ -44,7 +44,7 @@ Status DBImpl::EnableFileDeletions(bool force) {
  // Job id == 0 means that this is not our background process, but rather
  // user thread
  JobContext job_context(0);
-  bool file_deletion_enabled = false;
+  bool should_purge_files = false;
  {
    InstrumentedMutexLock l(&mutex_);
    if (force) {
@@ -54,18 +54,19 @@ Status DBImpl::EnableFileDeletions(bool force) {
      --disable_delete_obsolete_files_;
    }
    if (disable_delete_obsolete_files_ == 0)  {
-      file_deletion_enabled = true;
+      ROCKS_LOG_INFO(immutable_db_options_.info_log, "File Deletions Enabled");
+      should_purge_files = true;
      FindObsoleteFiles(&job_context, true);
      bg_cv_.SignalAll();
+    } else {
+      ROCKS_LOG_WARN(
+          immutable_db_options_.info_log,
+          "File Deletions Enable, but not really enabled. Counter: %d",
+          disable_delete_obsolete_files_);
    }
  }
-  if (file_deletion_enabled) {
-    ROCKS_LOG_INFO(immutable_db_options_.info_log, "File Deletions Enabled");
+  if (should_purge_files)  {
    PurgeObsoleteFiles(job_context);
-  } else {
-    ROCKS_LOG_WARN(immutable_db_options_.info_log,
-                   "File Deletions Enable, but not really enabled. Counter: %d",
-                   disable_delete_obsolete_files_);
  }
  job_context.Clean();
  LogFlush(immutable_db_options_.info_log);
@@ -86,28 +87,19 @@ Status DBImpl::GetLiveFiles(std::vector<std::string>& ret,
  if (flush_memtable) {
    // flush all dirty data to disk.
    Status status;
-    if (immutable_db_options_.atomic_flush) {
-      autovector<ColumnFamilyData*> cfds;
-      SelectColumnFamiliesForAtomicFlush(&cfds);
+    for (auto cfd : *versions_->GetColumnFamilySet()) {
+      if (cfd->IsDropped()) {
+        continue;
+      }
+      cfd->Ref();
      mutex_.Unlock();
-      status = AtomicFlushMemTables(cfds, FlushOptions(),
-                                    FlushReason::kGetLiveFiles);
+      status = FlushMemTable(cfd, FlushOptions(), FlushReason::kGetLiveFiles);
+      TEST_SYNC_POINT("DBImpl::GetLiveFiles:1");
+      TEST_SYNC_POINT("DBImpl::GetLiveFiles:2");
      mutex_.Lock();
-    } else {
-      for (auto cfd : *versions_->GetColumnFamilySet()) {
-        if (cfd->IsDropped()) {
-          continue;
-        }
-        cfd->Ref();
-        mutex_.Unlock();
-        status = FlushMemTable(cfd, FlushOptions(), FlushReason::kGetLiveFiles);
-        TEST_SYNC_POINT("DBImpl::GetLiveFiles:1");
-        TEST_SYNC_POINT("DBImpl::GetLiveFiles:2");
-        mutex_.Lock();
-        cfd->Unref();
-        if (!status.ok()) {
-          break;
-        }
+      cfd->Unref();
+      if (!status.ok()) {
+        break;
      }
    }
    versions_->GetColumnFamilySet()->FreeDeadColumnFamilies();
@@ -134,7 +126,7 @@ Status DBImpl::GetLiveFiles(std::vector<std::string>& ret,

  // create names of the live files. The names are not absolute
  // paths, instead they are relative to dbname_;
-  for (const auto& live_file : live) {
+  for (auto live_file : live) {
    ret.push_back(MakeTableFileName("", live_file.GetNumber()));
  }

@@ -25,12 +25,6 @@ class DBFlushDirectIOTest : public DBFlushTest,
  DBFlushDirectIOTest() : DBFlushTest() {}
 };

-class DBAtomicFlushTest : public DBFlushTest,
-                          public ::testing::WithParamInterface<bool> {
- public:
-  DBAtomicFlushTest() : DBFlushTest() {}
-};
-
 // We had issue when two background threads trying to flush at the same time,
 // only one of them get committed. The test verifies the issue is fixed.
 TEST_F(DBFlushTest, FlushWhileWritingManifest) {
@@ -46,7 +40,7 @@ TEST_F(DBFlushTest, FlushWhileWritingManifest) {
  SyncPoint::GetInstance()->LoadDependency(
      {{"VersionSet::LogAndApply:WriteManifest",
        "DBFlushTest::FlushWhileWritingManifest:1"},
-       {"MemTableList::TryInstallMemtableFlushResults:InProgress",
+       {"MemTableList::InstallMemtableFlushResults:InProgress",
        "VersionSet::LogAndApply:WriteManifestDone"}});
  SyncPoint::GetInstance()->EnableProcessing();

@@ -62,6 +56,7 @@ TEST_F(DBFlushTest, FlushWhileWritingManifest) {
 #endif  // ROCKSDB_LITE
 }

+#ifndef TRAVIS
 // Disable this test temporarily on Travis as it fails intermittently.
 // Github issue: #4151
 TEST_F(DBFlushTest, SyncFail) {
@@ -72,15 +67,11 @@ TEST_F(DBFlushTest, SyncFail) {
  options.env = fault_injection_env.get();

  SyncPoint::GetInstance()->LoadDependency(
-      {{"DBFlushTest::SyncFail:GetVersionRefCount:1",
-        "DBImpl::FlushMemTableToOutputFile:BeforePickMemtables"},
-       {"DBImpl::FlushMemTableToOutputFile:AfterPickMemtables",
-        "DBFlushTest::SyncFail:GetVersionRefCount:2"},
-       {"DBFlushTest::SyncFail:1", "DBImpl::SyncClosedLogs:Start"},
+      {{"DBFlushTest::SyncFail:1", "DBImpl::SyncClosedLogs:Start"},
       {"DBImpl::SyncClosedLogs:Failed", "DBFlushTest::SyncFail:2"}});
  SyncPoint::GetInstance()->EnableProcessing();

-  CreateAndReopenWithCF({"pikachu"}, options);
+  Reopen(options);
  Put("key", "value");
  auto* cfd =
      reinterpret_cast<ColumnFamilyHandleImpl*>(db_->DefaultColumnFamily())
@@ -91,10 +82,6 @@ TEST_F(DBFlushTest, SyncFail) {
  // Flush installs a new super-version. Get the ref count after that.
  auto current_before = cfd->current();
  int refs_before = cfd->current()->TEST_refs();
-  TEST_SYNC_POINT("DBFlushTest::SyncFail:GetVersionRefCount:1");
-  TEST_SYNC_POINT("DBFlushTest::SyncFail:GetVersionRefCount:2");
-  int refs_after_picking_memtables = cfd->current()->TEST_refs();
-  ASSERT_EQ(refs_before + 1, refs_after_picking_memtables);
  fault_injection_env->SetFilesystemActive(false);
  TEST_SYNC_POINT("DBFlushTest::SyncFail:1");
  TEST_SYNC_POINT("DBFlushTest::SyncFail:2");
@@ -109,30 +96,7 @@ TEST_F(DBFlushTest, SyncFail) {
  ASSERT_EQ(refs_before, cfd->current()->TEST_refs());
  Destroy(options);
 }
-
-TEST_F(DBFlushTest, SyncSkip) {
-  Options options = CurrentOptions();
-
-  SyncPoint::GetInstance()->LoadDependency(
-      {{"DBFlushTest::SyncSkip:1", "DBImpl::SyncClosedLogs:Skip"},
-       {"DBImpl::SyncClosedLogs:Skip", "DBFlushTest::SyncSkip:2"}});
-  SyncPoint::GetInstance()->EnableProcessing();
-
-  Reopen(options);
-  Put("key", "value");
-
-  FlushOptions flush_options;
-  flush_options.wait = false;
-  ASSERT_OK(dbfull()->Flush(flush_options));
-
-  TEST_SYNC_POINT("DBFlushTest::SyncSkip:1");
-  TEST_SYNC_POINT("DBFlushTest::SyncSkip:2");
-
-  // Now the background job will do the flush; wait for it.
-  dbfull()->TEST_WaitForFlushMemTable();
-
-  Destroy(options);
-}
+#endif  // TRAVIS

 TEST_F(DBFlushTest, FlushInLowPriThreadPool) {
  // Verify setting an empty high-pri (flush) thread pool causes flushes to be
@@ -250,306 +214,9 @@ TEST_F(DBFlushTest, FlushError) {
  ASSERT_NE(s, Status::OK());
 }

-TEST_F(DBFlushTest, ManualFlushFailsInReadOnlyMode) {
-  // Regression test for bug where manual flush hangs forever when the DB
-  // is in read-only mode. Verify it now at least returns, despite failing.
-  Options options;
-  std::unique_ptr<FaultInjectionTestEnv> fault_injection_env(
-      new FaultInjectionTestEnv(env_));
-  options.env = fault_injection_env.get();
-  options.max_write_buffer_number = 2;
-  Reopen(options);
-
-  // Trigger a first flush but don't let it run
-  ASSERT_OK(db_->PauseBackgroundWork());
-  ASSERT_OK(Put("key1", "value1"));
-  FlushOptions flush_opts;
-  flush_opts.wait = false;
-  ASSERT_OK(db_->Flush(flush_opts));
-
-  // Write a key to the second memtable so we have something to flush later
-  // after the DB is in read-only mode.
-  ASSERT_OK(Put("key2", "value2"));
-
-  // Let the first flush continue, hit an error, and put the DB in read-only
-  // mode.
-  fault_injection_env->SetFilesystemActive(false);
-  ASSERT_OK(db_->ContinueBackgroundWork());
-  dbfull()->TEST_WaitForFlushMemTable();
-#ifndef ROCKSDB_LITE
-  uint64_t num_bg_errors;
-  ASSERT_TRUE(db_->GetIntProperty(DB::Properties::kBackgroundErrors,
-                                  &num_bg_errors));
-  ASSERT_GT(num_bg_errors, 0);
-#endif  // ROCKSDB_LITE
-
-  // In the bug scenario, triggering another flush would cause the second flush
-  // to hang forever. After the fix we expect it to return an error.
-  ASSERT_NOK(db_->Flush(FlushOptions()));
-
-  Close();
-}
-
-TEST_P(DBAtomicFlushTest, ManualAtomicFlush) {
-  Options options = CurrentOptions();
-  options.create_if_missing = true;
-  options.atomic_flush = GetParam();
-  options.write_buffer_size = (static_cast<size_t>(64) << 20);
-
-  CreateAndReopenWithCF({"pikachu", "eevee"}, options);
-  size_t num_cfs = handles_.size();
-  ASSERT_EQ(3, num_cfs);
-  WriteOptions wopts;
-  wopts.disableWAL = true;
-  for (size_t i = 0; i != num_cfs; ++i) {
-    ASSERT_OK(Put(static_cast<int>(i) /*cf*/, "key", "value", wopts));
-  }
-  std::vector<int> cf_ids;
-  for (size_t i = 0; i != num_cfs; ++i) {
-    cf_ids.emplace_back(static_cast<int>(i));
-  }
-  ASSERT_OK(Flush(cf_ids));
-  for (size_t i = 0; i != num_cfs; ++i) {
-    auto cfh = static_cast<ColumnFamilyHandleImpl*>(handles_[i]);
-    ASSERT_EQ(0, cfh->cfd()->imm()->NumNotFlushed());
-    ASSERT_TRUE(cfh->cfd()->mem()->IsEmpty());
-  }
-}
-
-TEST_P(DBAtomicFlushTest, AtomicFlushTriggeredByMemTableFull) {
-  Options options = CurrentOptions();
-  options.create_if_missing = true;
-  options.atomic_flush = GetParam();
-  // 4KB so that we can easily trigger auto flush.
-  options.write_buffer_size = 4096;
-
-  SyncPoint::GetInstance()->LoadDependency(
-      {{"DBImpl::BackgroundCallFlush:FlushFinish:0",
-        "DBAtomicFlushTest::AtomicFlushTriggeredByMemTableFull:BeforeCheck"}});
-  SyncPoint::GetInstance()->EnableProcessing();
-
-  CreateAndReopenWithCF({"pikachu", "eevee"}, options);
-  size_t num_cfs = handles_.size();
-  ASSERT_EQ(3, num_cfs);
-  WriteOptions wopts;
-  wopts.disableWAL = true;
-  for (size_t i = 0; i != num_cfs; ++i) {
-    ASSERT_OK(Put(static_cast<int>(i) /*cf*/, "key", "value", wopts));
-  }
-  // Keep writing to one of them column families to trigger auto flush.
-  for (int i = 0; i != 4000; ++i) {
-    ASSERT_OK(Put(static_cast<int>(num_cfs) - 1 /*cf*/,
-                  "key" + std::to_string(i), "value" + std::to_string(i),
-                  wopts));
-  }
-
-  TEST_SYNC_POINT(
-      "DBAtomicFlushTest::AtomicFlushTriggeredByMemTableFull:BeforeCheck");
-  if (options.atomic_flush) {
-    for (size_t i = 0; i != num_cfs - 1; ++i) {
-      auto cfh = static_cast<ColumnFamilyHandleImpl*>(handles_[i]);
-      ASSERT_EQ(0, cfh->cfd()->imm()->NumNotFlushed());
-      ASSERT_TRUE(cfh->cfd()->mem()->IsEmpty());
-    }
-  } else {
-    for (size_t i = 0; i != num_cfs - 1; ++i) {
-      auto cfh = static_cast<ColumnFamilyHandleImpl*>(handles_[i]);
-      ASSERT_EQ(0, cfh->cfd()->imm()->NumNotFlushed());
-      ASSERT_FALSE(cfh->cfd()->mem()->IsEmpty());
-    }
-  }
-  SyncPoint::GetInstance()->DisableProcessing();
-}
-
-TEST_P(DBAtomicFlushTest, AtomicFlushRollbackSomeJobs) {
-  bool atomic_flush = GetParam();
-  if (!atomic_flush) {
-    return;
-  }
-  std::unique_ptr<FaultInjectionTestEnv> fault_injection_env(
-      new FaultInjectionTestEnv(env_));
-  Options options = CurrentOptions();
-  options.create_if_missing = true;
-  options.atomic_flush = atomic_flush;
-  options.env = fault_injection_env.get();
-  SyncPoint::GetInstance()->DisableProcessing();
-  SyncPoint::GetInstance()->LoadDependency(
-      {{"DBImpl::AtomicFlushMemTablesToOutputFiles:SomeFlushJobsComplete:1",
-        "DBAtomicFlushTest::AtomicFlushRollbackSomeJobs:1"},
-       {"DBAtomicFlushTest::AtomicFlushRollbackSomeJobs:2",
-        "DBImpl::AtomicFlushMemTablesToOutputFiles:SomeFlushJobsComplete:2"}});
-  SyncPoint::GetInstance()->EnableProcessing();
-
-  CreateAndReopenWithCF({"pikachu", "eevee"}, options);
-  size_t num_cfs = handles_.size();
-  ASSERT_EQ(3, num_cfs);
-  WriteOptions wopts;
-  wopts.disableWAL = true;
-  for (size_t i = 0; i != num_cfs; ++i) {
-    int cf_id = static_cast<int>(i);
-    ASSERT_OK(Put(cf_id, "key", "value", wopts));
-  }
-  FlushOptions flush_opts;
-  flush_opts.wait = false;
-  ASSERT_OK(dbfull()->Flush(flush_opts, handles_));
-  TEST_SYNC_POINT("DBAtomicFlushTest::AtomicFlushRollbackSomeJobs:1");
-  fault_injection_env->SetFilesystemActive(false);
-  TEST_SYNC_POINT("DBAtomicFlushTest::AtomicFlushRollbackSomeJobs:2");
-  for (auto* cfh : handles_) {
-    dbfull()->TEST_WaitForFlushMemTable(cfh);
-  }
-  for (size_t i = 0; i != num_cfs; ++i) {
-    auto cfh = static_cast<ColumnFamilyHandleImpl*>(handles_[i]);
-    ASSERT_EQ(1, cfh->cfd()->imm()->NumNotFlushed());
-    ASSERT_TRUE(cfh->cfd()->mem()->IsEmpty());
-  }
-  fault_injection_env->SetFilesystemActive(true);
-  Destroy(options);
-}
-
-TEST_P(DBAtomicFlushTest, FlushMultipleCFs_DropSomeBeforeRequestFlush) {
-  bool atomic_flush = GetParam();
-  if (!atomic_flush) {
-    return;
-  }
-  Options options = CurrentOptions();
-  options.create_if_missing = true;
-  options.atomic_flush = atomic_flush;
-  SyncPoint::GetInstance()->DisableProcessing();
-  SyncPoint::GetInstance()->ClearAllCallBacks();
-  SyncPoint::GetInstance()->EnableProcessing();
-
-  CreateAndReopenWithCF({"pikachu", "eevee"}, options);
-  size_t num_cfs = handles_.size();
-  ASSERT_EQ(3, num_cfs);
-  WriteOptions wopts;
-  wopts.disableWAL = true;
-  std::vector<int> cf_ids;
-  for (size_t i = 0; i != num_cfs; ++i) {
-    int cf_id = static_cast<int>(i);
-    ASSERT_OK(Put(cf_id, "key", "value", wopts));
-    cf_ids.push_back(cf_id);
-  }
-  ASSERT_OK(dbfull()->DropColumnFamily(handles_[1]));
-  ASSERT_TRUE(Flush(cf_ids).IsShutdownInProgress());
-  Destroy(options);
-}
-
-TEST_P(DBAtomicFlushTest,
-       FlushMultipleCFs_DropSomeAfterScheduleFlushBeforeFlushJobRun) {
-  bool atomic_flush = GetParam();
-  if (!atomic_flush) {
-    return;
-  }
-  Options options = CurrentOptions();
-  options.create_if_missing = true;
-  options.atomic_flush = atomic_flush;
-
-  CreateAndReopenWithCF({"pikachu", "eevee"}, options);
-
-  SyncPoint::GetInstance()->DisableProcessing();
-  SyncPoint::GetInstance()->ClearAllCallBacks();
-  SyncPoint::GetInstance()->LoadDependency(
-      {{"DBImpl::AtomicFlushMemTables:AfterScheduleFlush",
-        "DBAtomicFlushTest::BeforeDropCF"},
-       {"DBAtomicFlushTest::AfterDropCF",
-        "DBImpl::BackgroundCallFlush:start"}});
-  SyncPoint::GetInstance()->EnableProcessing();
-
-  size_t num_cfs = handles_.size();
-  ASSERT_EQ(3, num_cfs);
-  WriteOptions wopts;
-  wopts.disableWAL = true;
-  for (size_t i = 0; i != num_cfs; ++i) {
-    int cf_id = static_cast<int>(i);
-    ASSERT_OK(Put(cf_id, "key", "value", wopts));
-  }
-  port::Thread user_thread([&]() {
-    TEST_SYNC_POINT("DBAtomicFlushTest::BeforeDropCF");
-    ASSERT_OK(dbfull()->DropColumnFamily(handles_[1]));
-    TEST_SYNC_POINT("DBAtomicFlushTest::AfterDropCF");
-  });
-  FlushOptions flush_opts;
-  flush_opts.wait = true;
-  ASSERT_OK(dbfull()->Flush(flush_opts, handles_));
-  user_thread.join();
-  for (size_t i = 0; i != num_cfs; ++i) {
-    int cf_id = static_cast<int>(i);
-    ASSERT_EQ("value", Get(cf_id, "key"));
-  }
-
-  ReopenWithColumnFamilies({kDefaultColumnFamilyName, "eevee"}, options);
-  num_cfs = handles_.size();
-  ASSERT_EQ(2, num_cfs);
-  for (size_t i = 0; i != num_cfs; ++i) {
-    int cf_id = static_cast<int>(i);
-    ASSERT_EQ("value", Get(cf_id, "key"));
-  }
-  Destroy(options);
-}
-
-TEST_P(DBAtomicFlushTest, TriggerFlushAndClose) {
-  bool atomic_flush = GetParam();
-  if (!atomic_flush) {
-    return;
-  }
-  const int kNumKeysTriggerFlush = 4;
-  Options options = CurrentOptions();
-  options.create_if_missing = true;
-  options.atomic_flush = atomic_flush;
-  options.memtable_factory.reset(
-      new SpecialSkipListFactory(kNumKeysTriggerFlush));
-  CreateAndReopenWithCF({"pikachu"}, options);
-
-  for (int i = 0; i != kNumKeysTriggerFlush; ++i) {
-    ASSERT_OK(Put(0, "key" + std::to_string(i), "value" + std::to_string(i)));
-  }
-  SyncPoint::GetInstance()->DisableProcessing();
-  SyncPoint::GetInstance()->ClearAllCallBacks();
-  SyncPoint::GetInstance()->EnableProcessing();
-  ASSERT_OK(Put(0, "key", "value"));
-  Close();
-
-  ReopenWithColumnFamilies({kDefaultColumnFamilyName, "pikachu"}, options);
-  ASSERT_EQ("value", Get(0, "key"));
-}
-
-TEST_P(DBAtomicFlushTest, PickMemtablesRaceWithBackgroundFlush) {
-  bool atomic_flush = GetParam();
-  Options options = CurrentOptions();
-  options.create_if_missing = true;
-  options.atomic_flush = atomic_flush;
-  options.max_write_buffer_number = 4;
-  // Set min_write_buffer_number_to_merge to be greater than 1, so that
-  // a column family with one memtable in the imm will not cause IsFlushPending
-  // to return true when flush_requested_ is false.
-  options.min_write_buffer_number_to_merge = 2;
-  CreateAndReopenWithCF({"pikachu"}, options);
-  ASSERT_EQ(2, handles_.size());
-  ASSERT_OK(dbfull()->PauseBackgroundWork());
-  ASSERT_OK(Put(0, "key00", "value00"));
-  ASSERT_OK(Put(1, "key10", "value10"));
-  FlushOptions flush_opts;
-  flush_opts.wait = false;
-  ASSERT_OK(dbfull()->Flush(flush_opts, handles_));
-  ASSERT_OK(Put(0, "key01", "value01"));
-  // Since max_write_buffer_number is 4, the following flush won't cause write
-  // stall.
-  ASSERT_OK(dbfull()->Flush(flush_opts));
-  ASSERT_OK(dbfull()->DropColumnFamily(handles_[1]));
-  ASSERT_OK(dbfull()->DestroyColumnFamilyHandle(handles_[1]));
-  handles_[1] = nullptr;
-  ASSERT_OK(dbfull()->ContinueBackgroundWork());
-  ASSERT_OK(dbfull()->TEST_WaitForFlushMemTable(handles_[0]));
-  delete handles_[0];
-  handles_.clear();
-}
-
 INSTANTIATE_TEST_CASE_P(DBFlushDirectIOTest, DBFlushDirectIOTest,
                        testing::Bool());

-INSTANTIATE_TEST_CASE_P(DBAtomicFlushTest, DBAtomicFlushTest, testing::Bool());
-
 }  // namespace rocksdb

 int main(int argc, char** argv) {
@@ -22,16 +22,15 @@
 #include "db/column_family.h"
 #include "db/compaction_job.h"
 #include "db/dbformat.h"
+#include "db/external_sst_file_ingestion_job.h"
 #include "db/error_handler.h"
 #include "db/event_helpers.h"
-#include "db/external_sst_file_ingestion_job.h"
 #include "db/flush_job.h"
 #include "db/flush_scheduler.h"
 #include "db/internal_stats.h"
 #include "db/log_writer.h"
 #include "db/logs_with_prep_tracker.h"
 #include "db/pre_release_callback.h"
-#include "db/range_del_aggregator.h"
 #include "db/read_callback.h"
 #include "db/snapshot_checker.h"
 #include "db/snapshot_impl.h"
@@ -54,7 +53,6 @@
 #include "util/autovector.h"
 #include "util/event_logger.h"
 #include "util/hash.h"
-#include "util/repeatable_thread.h"
 #include "util/stop_watch.h"
 #include "util/thread_local.h"
 #include "util/trace_replay.h"
@@ -63,10 +61,8 @@ namespace rocksdb {

 class Arena;
 class ArenaWrappedDBIter;
-class InMemoryStatsHistoryIterator;
 class MemTable;
 class TableCache;
-class TaskLimiterToken;
 class Version;
 class VersionEdit;
 class VersionSet;
@@ -124,25 +120,6 @@ class DBImpl : public DB {
      const std::vector<Slice>& keys,
      std::vector<std::string>* values) override;

-  // This MultiGet is a batched version, which may be faster than calling Get
-  // multiple times, especially if the keys have some spatial locality that
-  // enables them to be queried in the same SST files/set of files. The larger
-  // the batch size, the more scope for batching and performance improvement
-  // The values and statuses parameters are arrays with number of elements
-  // equal to keys.size(). This allows the storage for those to be alloacted
-  // by the caller on the stack for small batches
-  virtual void MultiGet(const ReadOptions& options,
-                        ColumnFamilyHandle* column_family,
-                        const size_t num_keys, const Slice* keys,
-                        PinnableSlice* values, Status* statuses,
-                        const bool sorted_input = false) override;
-
-  void MultiGetImpl(
-      const ReadOptions& options, ColumnFamilyHandle* column_family,
-      autovector<KeyContext, MultiGetContext::MAX_BATCH_SIZE>& key_context,
-      bool sorted_input, ReadCallback* callback = nullptr,
-      bool* is_blob_index = nullptr);
-
  virtual Status CreateColumnFamily(const ColumnFamilyOptions& cf_options,
                                    const std::string& column_family,
                                    ColumnFamilyHandle** handle) override;
@@ -197,9 +174,10 @@ class DBImpl : public DB {
  virtual bool GetAggregatedIntProperty(const Slice& property,
                                        uint64_t* aggregated_value) override;
  using DB::GetApproximateSizes;
-  virtual void GetApproximateSizes(
-      ColumnFamilyHandle* column_family, const Range* range, int n,
-      uint64_t* sizes, uint8_t include_flags = INCLUDE_FILES) override;
+  virtual void GetApproximateSizes(ColumnFamilyHandle* column_family,
+                                   const Range* range, int n, uint64_t* sizes,
+                                   uint8_t include_flags
+                                   = INCLUDE_FILES) override;
  using DB::GetApproximateMemTableStats;
  virtual void GetApproximateMemTableStats(ColumnFamilyHandle* column_family,
                                           const Range& range,
@@ -211,13 +189,13 @@ class DBImpl : public DB {
                              const Slice* begin, const Slice* end) override;

  using DB::CompactFiles;
-  virtual Status CompactFiles(
-      const CompactionOptions& compact_options,
-      ColumnFamilyHandle* column_family,
-      const std::vector<std::string>& input_file_names, const int output_level,
-      const int output_path_id = -1,
-      std::vector<std::string>* const output_file_names = nullptr,
-      CompactionJobInfo* compaction_job_info = nullptr) override;
+  virtual Status CompactFiles(const CompactionOptions& compact_options,
+                              ColumnFamilyHandle* column_family,
+                              const std::vector<std::string>& input_file_names,
+                              const int output_level,
+                              const int output_path_id = -1,
+                              std::vector<std::string>* const output_file_names
+                              = nullptr) override;

  virtual Status PauseBackgroundWork() override;
  virtual Status ContinueBackgroundWork() override;
@@ -249,23 +227,11 @@ class DBImpl : public DB {
  using DB::Flush;
  virtual Status Flush(const FlushOptions& options,
                       ColumnFamilyHandle* column_family) override;
-  virtual Status Flush(
-      const FlushOptions& options,
-      const std::vector<ColumnFamilyHandle*>& column_families) override;
  virtual Status FlushWAL(bool sync) override;
-  bool TEST_WALBufferIsEmpty(bool lock = true);
+  bool TEST_WALBufferIsEmpty();
  virtual Status SyncWAL() override;
-  virtual Status LockWAL() override;
-  virtual Status UnlockWAL() override;

  virtual SequenceNumber GetLatestSequenceNumber() const override;
-  virtual SequenceNumber GetLastPublishedSequence() const {
-    if (last_seq_same_as_publish_seq_) {
-      return versions_->LastSequence();
-    } else {
-      return versions_->LastPublishedSequence();
-    }
-  }
  // REQUIRES: joined the main write queue if two_write_queues is disabled, and
  // the second write queue otherwise.
  virtual void SetLastPublishedSequence(SequenceNumber seq);
@@ -289,9 +255,9 @@ class DBImpl : public DB {
  virtual Status GetSortedWalFiles(VectorLogPtr& files) override;

  virtual Status GetUpdatesSince(
-      SequenceNumber seq_number, std::unique_ptr<TransactionLogIterator>* iter,
-      const TransactionLogIterator::ReadOptions& read_options =
-          TransactionLogIterator::ReadOptions()) override;
+      SequenceNumber seq_number, unique_ptr<TransactionLogIterator>* iter,
+      const TransactionLogIterator::ReadOptions&
+          read_options = TransactionLogIterator::ReadOptions()) override;
  virtual Status DeleteFile(std::string name) override;
  Status DeleteFilesInRanges(ColumnFamilyHandle* column_family,
                             const RangePtr* ranges, size_t n,
@@ -304,8 +270,9 @@ class DBImpl : public DB {
  // Status::NotFound() will be returned if the current DB does not have
  // any column family match the specified name.
  // TODO(yhchiang): output parameter is placed in the end in this codebase.
-  virtual void GetColumnFamilyMetaData(ColumnFamilyHandle* column_family,
-                                       ColumnFamilyMetaData* metadata) override;
+  virtual void GetColumnFamilyMetaData(
+      ColumnFamilyHandle* column_family,
+      ColumnFamilyMetaData* metadata) override;

  Status SuggestCompactRange(ColumnFamilyHandle* column_family,
                             const Slice* begin, const Slice* end) override;
@@ -366,10 +333,6 @@ class DBImpl : public DB {
      const std::vector<std::string>& external_files,
      const IngestExternalFileOptions& ingestion_options) override;

-  using DB::IngestExternalFiles;
-  virtual Status IngestExternalFiles(
-      const std::vector<IngestExternalFileArg>& args) override;
-
  virtual Status VerifyChecksum() override;

  using DB::StartTrace;
@@ -395,21 +358,18 @@ class DBImpl : public DB {

  virtual Status GetDbIdentity(std::string& identity) const override;

-  // max_file_num_to_ignore allows bottom level compaction to filter out newly
-  // compacted SST files. Setting max_file_num_to_ignore to kMaxUint64 will
-  // disable the filtering
  Status RunManualCompaction(ColumnFamilyData* cfd, int input_level,
-                             int output_level,
-                             const CompactRangeOptions& compact_range_options,
+                             int output_level, uint32_t output_path_id,
+                             uint32_t max_subcompactions,
                             const Slice* begin, const Slice* end,
-                             bool exclusive, bool disallow_trivial_move,
-                             uint64_t max_file_num_to_ignore);
+                             bool exclusive,
+                             bool disallow_trivial_move = false);

  // Return an internal iterator over the current state of the database.
  // The keys of this iterator are internal keys (see format.h).
  // The returned iterator should be deleted when no longer needed.
  InternalIterator* NewInternalIterator(
-      Arena* arena, RangeDelAggregator* range_del_agg, SequenceNumber sequence,
+      Arena* arena, RangeDelAggregator* range_del_agg,
      ColumnFamilyHandle* column_family = nullptr);

  LogsWithPrepTracker* logs_with_prep_tracker() {
@@ -449,8 +409,8 @@ class DBImpl : public DB {

  // Return the maximum overlapping data (in bytes) at next level for any
  // file at a level >= 1.
-  int64_t TEST_MaxNextLevelOverlappingBytes(
-      ColumnFamilyHandle* column_family = nullptr);
+  int64_t TEST_MaxNextLevelOverlappingBytes(ColumnFamilyHandle* column_family =
+                                                nullptr);

  // Return the current manifest file no.
  uint64_t TEST_Current_Manifest_FileNo();
@@ -504,11 +464,6 @@ class DBImpl : public DB {

  int TEST_BGCompactionsAllowed() const;
  int TEST_BGFlushesAllowed() const;
-  size_t TEST_GetWalPreallocateBlockSize(uint64_t write_buffer_size) const;
-  void TEST_WaitForDumpStatsRun(std::function<void()> callback) const;
-  void TEST_WaitForPersistStatsRun(std::function<void()> callback) const;
-  bool TEST_IsPersistentStatsEnabled() const;
-  size_t TEST_EstiamteStatsHistorySize() const;

 #endif  // NDEBUG

@@ -530,14 +485,6 @@ class DBImpl : public DB {

  uint64_t MinLogNumberToKeep();

-  // Returns the lower bound file number for SSTs that won't be deleted, even if
-  // they're obsolete. This lower bound is used internally to prevent newly
-  // created flush/compaction output files from being deleted before they're
-  // installed. This technique avoids the need for tracking the exact numbers of
-  // files pending creation, although it prevents more files than necessary from
-  // being deleted.
-  uint64_t MinObsoleteSstNumberToKeep();
-
  // Returns the list of live files in 'live' and the list
  // of all files in the filesystem in 'candidate_files'.
  // If force == false and the last call was less than
@@ -561,13 +508,6 @@ class DBImpl : public DB {

  const SnapshotList& snapshots() const { return snapshots_; }

-  void LoadSnapshots(std::vector<SequenceNumber>* snap_vector,
-                     SequenceNumber* oldest_write_conflict_snapshot,
-                     const SequenceNumber& max_seq) const {
-    InstrumentedMutexLock l(mutex());
-    snapshots().GetAll(snap_vector, oldest_write_conflict_snapshot, max_seq);
-  }
-
  const ImmutableDBOptions& immutable_db_options() const {
    return immutable_db_options_;
  }
@@ -604,8 +544,7 @@ class DBImpl : public DB {
  ColumnFamilyHandle* GetColumnFamilyHandle(uint32_t column_family_id);

  // Same as above, should called without mutex held and not on write thread.
-  std::unique_ptr<ColumnFamilyHandle> GetColumnFamilyHandleUnlocked(
-      uint32_t column_family_id);
+  ColumnFamilyHandle* GetColumnFamilyHandleUnlocked(uint32_t column_family_id);

  // Returns the number of currently running flushes.
  // REQUIREMENT: mutex_ must be held when calling this function.
@@ -623,9 +562,11 @@ class DBImpl : public DB {

  const WriteController& write_controller() { return write_controller_; }

-  InternalIterator* NewInternalIterator(
-      const ReadOptions&, ColumnFamilyData* cfd, SuperVersion* super_version,
-      Arena* arena, RangeDelAggregator* range_del_agg, SequenceNumber sequence);
+  InternalIterator* NewInternalIterator(const ReadOptions&,
+                                        ColumnFamilyData* cfd,
+                                        SuperVersion* super_version,
+                                        Arena* arena,
+                                        RangeDelAggregator* range_del_agg);

  // hollow transactions shell used for recovery.
  // these will then be passed to TransactionDB so that
@@ -737,16 +678,10 @@ class DBImpl : public DB {

  void SetSnapshotChecker(SnapshotChecker* snapshot_checker);

-  // Fill JobContext with snapshot information needed by flush and compaction.
-  void GetSnapshotContext(JobContext* job_context,
-                          std::vector<SequenceNumber>* snapshot_seqs,
-                          SequenceNumber* earliest_write_conflict_snapshot,
-                          SnapshotChecker** snapshot_checker);
-
  // Not thread-safe.
  void SetRecoverableStatePreReleaseCallback(PreReleaseCallback* callback);

-  InstrumentedMutex* mutex() const { return &mutex_; }
+  InstrumentedMutex* mutex() { return &mutex_; }

  Status NewDB();

@@ -761,21 +696,10 @@ class DBImpl : public DB {
  static Status CreateAndNewDirectory(Env* env, const std::string& dirname,
                                      std::unique_ptr<Directory>* directory);

-  // Given a time window, return an iterator for accessing stats history
-  Status GetStatsHistory(
-      uint64_t start_time, uint64_t end_time,
-      std::unique_ptr<StatsHistoryIterator>* stats_iterator) override;
-
-  // find stats map from stats_history_ with smallest timestamp in
-  // the range of [start_time, end_time)
-  bool FindStatsByTime(uint64_t start_time, uint64_t end_time,
-                       uint64_t* new_time,
-                       std::map<std::string, uint64_t>* stats_map);
-
 protected:
  Env* const env_;
  const std::string dbname_;
-  std::unique_ptr<VersionSet> versions_;
+  unique_ptr<VersionSet> versions_;
  // Flag to check whether we allocated and own the info log file
  bool own_info_log_;
  const DBOptions initial_db_options_;
@@ -787,46 +711,6 @@ class DBImpl : public DB {
  std::unique_ptr<Tracer> tracer_;
  InstrumentedMutex trace_mutex_;

-  // State below is protected by mutex_
-  // With two_write_queues enabled, some of the variables that accessed during
-  // WriteToWAL need different synchronization: log_empty_, alive_log_files_,
-  // logs_, logfile_number_. Refer to the definition of each variable below for
-  // more description.
-  mutable InstrumentedMutex mutex_;
-
-  ColumnFamilyHandleImpl* default_cf_handle_;
-  InternalStats* default_cf_internal_stats_;
-
-  // only used for dynamically adjusting max_total_wal_size. it is a sum of
-  // [write_buffer_size * max_write_buffer_number] over all column families
-  uint64_t max_total_in_memory_state_;
-  // If true, we have only one (default) column family. We use this to optimize
-  // some code-paths
-  bool single_column_family_mode_;
-
-  // The options to access storage files
-  const EnvOptions env_options_;
-
-  // Additonal options for compaction and flush
-  EnvOptions env_options_for_compaction_;
-
-  std::unique_ptr<ColumnFamilyMemTablesImpl> column_family_memtables_;
-
-  // Increase the sequence number after writing each batch, whether memtable is
-  // disabled for that or not. Otherwise the sequence number is increased after
-  // writing each key into memtable. This implies that when disable_memtable is
-  // set, the seq is not increased at all.
-  //
-  // Default: false
-  const bool seq_per_batch_;
-  // This determines during recovery whether we expect one writebatch per
-  // recovered transaction, or potentially multiple writebatches per
-  // transaction. For WriteUnprepared, this is set to false, since multiple
-  // batches can exist per transaction.
-  //
-  // Default: true
-  const bool batch_per_txn_;
-
  // Except in DB::Open(), WriteOptionsFile can only be called when:
  // Persist options to options file.
  // If need_mutex_lock = false, the method will lock DB mutex.
@@ -847,12 +731,8 @@ class DBImpl : public DB {
                              const MutableCFOptions& mutable_cf_options,
                              int job_id, TableProperties prop);

-  void NotifyOnCompactionBegin(ColumnFamilyData* cfd, Compaction* c,
-                               const Status& st,
-                               const CompactionJobStats& job_stats, int job_id);
-
-  void NotifyOnCompactionCompleted(ColumnFamilyData* cfd, Compaction* c,
-                                   const Status& st,
+  void NotifyOnCompactionCompleted(ColumnFamilyData* cfd,
+                                   Compaction *c, const Status &st,
                                   const CompactionJobStats& job_stats,
                                   int job_id);
  void NotifyOnMemTableSealed(ColumnFamilyData* cfd,
@@ -913,17 +793,8 @@ class DBImpl : public DB {
  // Actual implementation of Close()
  Status CloseImpl();

-  // Recover the descriptor from persistent storage.  May do a significant
-  // amount of work to recover recently logged updates.  Any changes to
-  // be made to the descriptor are added to *edit.
-  virtual Status Recover(
-      const std::vector<ColumnFamilyDescriptor>& column_families,
-      bool read_only = false, bool error_if_log_file_exist = false,
-      bool error_if_data_exists_in_logs = false);
-
 private:
  friend class DB;
-  friend class ErrorHandler;
  friend class InternalStats;
  friend class PessimisticTransaction;
  friend class TransactionBaseImpl;
@@ -941,7 +812,6 @@ class DBImpl : public DB {
  friend class CompactedDBImpl;
  friend class DBTest_ConcurrentFlushWAL_Test;
  friend class DBTest_MixedSlowdownOptionsStop_Test;
-  friend class DBCompactionTest_CompactBottomLevelFilesWithDeletions_Test;
 #ifndef NDEBUG
  friend class DBTest2_ReadCallbackTest_Test;
  friend class WriteCallbackTest_WriteWithCallbackTest_Test;
@@ -969,7 +839,12 @@ class DBImpl : public DB {
  struct PrepickedCompaction;
  struct PurgeFileInfo;

-  Status ResumeImpl();
+  // Recover the descriptor from persistent storage.  May do a significant
+  // amount of work to recover recently logged updates.  Any changes to
+  // be made to the descriptor are added to *edit.
+  Status Recover(const std::vector<ColumnFamilyDescriptor>& column_families,
+                 bool read_only = false, bool error_if_log_file_exist = false,
+                 bool error_if_data_exists_in_logs = false);

  void MaybeIgnoreError(Status* s) const;

@@ -1010,58 +885,15 @@ class DBImpl : public DB {
  Status SyncClosedLogs(JobContext* job_context);

  // Flush the in-memory write buffer to storage.  Switches to a new
-  // log-file/memtable and writes a new descriptor iff successful. Then
-  // installs a new super version for the column family.
-  Status FlushMemTableToOutputFile(
-      ColumnFamilyData* cfd, const MutableCFOptions& mutable_cf_options,
-      bool* madeProgress, JobContext* job_context,
-      SuperVersionContext* superversion_context,
-      std::vector<SequenceNumber>& snapshot_seqs,
-      SequenceNumber earliest_write_conflict_snapshot,
-      SnapshotChecker* snapshot_checker, LogBuffer* log_buffer,
-      Env::Priority thread_pri);
-
-  // Argument required by background flush thread.
-  struct BGFlushArg {
-    BGFlushArg()
-        : cfd_(nullptr), max_memtable_id_(0), superversion_context_(nullptr) {}
-    BGFlushArg(ColumnFamilyData* cfd, uint64_t max_memtable_id,
-               SuperVersionContext* superversion_context)
-        : cfd_(cfd),
-          max_memtable_id_(max_memtable_id),
-          superversion_context_(superversion_context) {}
-
-    // Column family to flush.
-    ColumnFamilyData* cfd_;
-    // Maximum ID of memtable to flush. In this column family, memtables with
-    // IDs smaller than this value must be flushed before this flush completes.
-    uint64_t max_memtable_id_;
-    // Pointer to a SuperVersionContext object. After flush completes, RocksDB
-    // installs a new superversion for the column family. This operation
-    // requires a SuperVersionContext object (currently embedded in JobContext).
-    SuperVersionContext* superversion_context_;
-  };
-
-  // Argument passed to flush thread.
-  struct FlushThreadArg {
-    DBImpl* db_;
-
-    Env::Priority thread_pri_;
-  };
-
-  // Flush the memtables of (multiple) column families to multiple files on
-  // persistent storage.
-  Status FlushMemTablesToOutputFiles(
-      const autovector<BGFlushArg>& bg_flush_args, bool* made_progress,
-      JobContext* job_context, LogBuffer* log_buffer, Env::Priority thread_pri);
-
-  Status AtomicFlushMemTablesToOutputFiles(
-      const autovector<BGFlushArg>& bg_flush_args, bool* made_progress,
-      JobContext* job_context, LogBuffer* log_buffer, Env::Priority thread_pri);
+  // log-file/memtable and writes a new descriptor iff successful.
+  Status FlushMemTableToOutputFile(ColumnFamilyData* cfd,
+                                   const MutableCFOptions& mutable_cf_options,
+                                   bool* madeProgress, JobContext* job_context,
+                                   LogBuffer* log_buffer);

  // REQUIRES: log_numbers are sorted in ascending order
-  virtual Status RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
-                                 SequenceNumber* next_sequence, bool read_only);
+  Status RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
+                         SequenceNumber* next_sequence, bool read_only);

  // The following two methods are used to flush a memtable to
  // storage. The first one is used at database RecoveryTime (when the
@@ -1071,12 +903,6 @@ class DBImpl : public DB {
  Status WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
                                     MemTable* mem, VersionEdit* edit);

-  // Restore alive_log_files_ and total_log_size_ after recovery.
-  // It needs to run only when there's no flush during recovery
-  // (e.g. avoid_flush_during_recovery=true). May also trigger flush
-  // in case total_log_size > max_total_wal_size.
-  Status RestoreAliveLogFiles(const std::vector<uint64_t>& log_numbers);
-
  // num_bytes: for slowdown case, delay time is calculated based on
  //            `num_bytes` going through.
  Status DelayWrite(uint64_t num_bytes, const WriteOptions& write_options);
@@ -1086,19 +912,13 @@ class DBImpl : public DB {

  Status ScheduleFlushes(WriteContext* context);

-  Status SwitchMemtable(ColumnFamilyData* cfd, WriteContext* context);
-
-  void SelectColumnFamiliesForAtomicFlush(autovector<ColumnFamilyData*>* cfds);
+  Status SwitchMemtable(ColumnFamilyData* cfd, WriteContext* context,
+                        FlushReason flush_reason = FlushReason::kOthers);

  // Force current memtable contents to be flushed.
  Status FlushMemTable(ColumnFamilyData* cfd, const FlushOptions& options,
                       FlushReason flush_reason, bool writes_stopped = false);

-  Status AtomicFlushMemTables(
-      const autovector<ColumnFamilyData*>& column_family_datas,
-      const FlushOptions& options, FlushReason flush_reason,
-      bool writes_stopped = false);
-
  // Wait until flushing this column family won't stall writes
  Status WaitUntilFlushWouldNotStallWrites(ColumnFamilyData* cfd,
                                           bool* flush_needed);
@@ -1107,22 +927,8 @@ class DBImpl : public DB {
  // If flush_memtable_id is non-null, wait until the memtable with the ID
  // gets flush. Otherwise, wait until the column family don't have any
  // memtable pending flush.
-  // resuming_from_bg_err indicates whether the caller is attempting to resume
-  // from background error.
  Status WaitForFlushMemTable(ColumnFamilyData* cfd,
-                              const uint64_t* flush_memtable_id = nullptr,
-                              bool resuming_from_bg_err = false) {
-    return WaitForFlushMemTables({cfd}, {flush_memtable_id},
-                                 resuming_from_bg_err);
-  }
-  // Wait for memtables to be flushed for multiple column families.
-  Status WaitForFlushMemTables(
-      const autovector<ColumnFamilyData*>& cfds,
-      const autovector<const uint64_t*>& flush_memtable_ids,
-      bool resuming_from_bg_err);
-
-  // REQUIRES: mutex locked and in write thread.
-  void AssignAtomicFlushSeq(const autovector<ColumnFamilyData*>& cfds);
+                              const uint64_t* flush_memtable_id = nullptr);

  // REQUIRES: mutex locked
  Status SwitchWAL(WriteContext* write_context);
@@ -1163,8 +969,7 @@ class DBImpl : public DB {
                          const std::vector<std::string>& input_file_names,
                          std::vector<std::string>* const output_file_names,
                          const int output_level, int output_path_id,
-                          JobContext* job_context, LogBuffer* log_buffer,
-                          CompactionJobInfo* compaction_job_info);
+                          JobContext* job_context, LogBuffer* log_buffer);

  // Wait for current IngestExternalFile() calls to finish.
  // REQUIRES: mutex_ held
@@ -1179,20 +984,7 @@ class DBImpl : public DB {
  ColumnFamilyData* GetColumnFamilyDataByName(const std::string& cf_name);

  void MaybeScheduleFlushOrCompaction();
-
-  // A flush request specifies the column families to flush as well as the
-  // largest memtable id to persist for each column family. Once all the
-  // memtables whose IDs are smaller than or equal to this per-column-family
-  // specified value, this flush request is considered to have completed its
-  // work of flushing this column family. After completing the work for all
-  // column families in this request, this flush is considered complete.
-  typedef std::vector<std::pair<ColumnFamilyData*, uint64_t>> FlushRequest;
-
-  void GenerateFlushRequest(const autovector<ColumnFamilyData*>& cfds,
-                            FlushRequest* req);
-
-  void SchedulePendingFlush(const FlushRequest& req, FlushReason flush_reason);
-
+  void SchedulePendingFlush(ColumnFamilyData* cfd, FlushReason flush_reason);
  void SchedulePendingCompaction(ColumnFamilyData* cfd);
  void SchedulePendingPurge(std::string fname, std::string dir_to_sync,
                            FileType type, uint64_t number, int job_id);
@@ -1200,50 +992,31 @@ class DBImpl : public DB {
  // Runs a pre-chosen universal compaction involving bottom level in a
  // separate, bottom-pri thread pool.
  static void BGWorkBottomCompaction(void* arg);
-  static void BGWorkFlush(void* arg);
+  static void BGWorkFlush(void* db);
  static void BGWorkPurge(void* arg);
-  static void UnscheduleCompactionCallback(void* arg);
-  static void UnscheduleFlushCallback(void* arg);
+  static void UnscheduleCallback(void* arg);
  void BackgroundCallCompaction(PrepickedCompaction* prepicked_compaction,
-                                Env::Priority thread_pri);
-  void BackgroundCallFlush(Env::Priority thread_pri);
+                                Env::Priority bg_thread_pri);
+  void BackgroundCallFlush();
  void BackgroundCallPurge();
  Status BackgroundCompaction(bool* madeProgress, JobContext* job_context,
                              LogBuffer* log_buffer,
-                              PrepickedCompaction* prepicked_compaction,
-                              Env::Priority thread_pri);
+                              PrepickedCompaction* prepicked_compaction);
  Status BackgroundFlush(bool* madeProgress, JobContext* job_context,
-                         LogBuffer* log_buffer, FlushReason* reason,
-                         Env::Priority thread_pri);
+                         LogBuffer* log_buffer);

-  bool EnoughRoomForCompaction(ColumnFamilyData* cfd,
-                               const std::vector<CompactionInputFiles>& inputs,
+  bool EnoughRoomForCompaction(const std::vector<CompactionInputFiles>& inputs,
                               bool* sfm_bookkeeping, LogBuffer* log_buffer);

-  // Request compaction tasks token from compaction thread limiter.
-  // It always succeeds if force = true or limiter is disable.
-  bool RequestCompactionToken(ColumnFamilyData* cfd, bool force,
-                              std::unique_ptr<TaskLimiterToken>* token,
-                              LogBuffer* log_buffer);
-
-  // Schedule background tasks
-  void StartTimedTasks();
-
  void PrintStatistics();

-  size_t EstiamteStatsHistorySize() const;
-
-  // persist stats to column family "_persistent_stats"
-  void PersistStats();
-
  // dump rocksdb.stats to LOG
-  void DumpStats();
+  void MaybeDumpStats();

  // Return the minimum empty level that could hold the total data in the
  // input level. Return the input level, if such level could not be found.
  int FindMinimumEmptyLevelFitting(ColumnFamilyData* cfd,
-                                   const MutableCFOptions& mutable_cf_options,
-                                   int level);
+      const MutableCFOptions& mutable_cf_options, int level);

  // Move the files in the input level to the target level.
  // If target_level < 0, automatically calculate the minimum level that could
@@ -1253,17 +1026,13 @@ class DBImpl : public DB {
  // helper functions for adding and removing from flush & compaction queues
  void AddToCompactionQueue(ColumnFamilyData* cfd);
  ColumnFamilyData* PopFirstFromCompactionQueue();
-  FlushRequest PopFirstFromFlushQueue();
-
-  // Pick the first unthrottled compaction with task token from queue.
-  ColumnFamilyData* PickCompactionFromQueue(
-      std::unique_ptr<TaskLimiterToken>* token, LogBuffer* log_buffer);
+  void AddToFlushQueue(ColumnFamilyData* cfd, FlushReason flush_reason);
+  ColumnFamilyData* PopFirstFromFlushQueue();

  // helper function to call after some of the logs_ were synced
  void MarkLogsSynced(uint64_t up_to, bool synced_dir, const Status& status);

-  SnapshotImpl* GetSnapshotImpl(bool is_write_conflict_boundary,
-                                bool lock = true);
+  SnapshotImpl* GetSnapshotImpl(bool is_write_conflict_boundary);

  uint64_t GetMaxTotalWalSize() const;

@@ -1271,21 +1040,23 @@ class DBImpl : public DB {

  Status CloseHelper();

-  void WaitForBackgroundWork();
-
  // table_cache_ provides its own synchronization
  std::shared_ptr<Cache> table_cache_;

  // Lock over the persistent DB state.  Non-nullptr iff successfully acquired.
  FileLock* db_lock_;

-  // In addition to mutex_, log_write_mutex_ protected writes to stats_history_
-  InstrumentedMutex stats_history_mutex_;
  // In addition to mutex_, log_write_mutex_ protected writes to logs_ and
  // logfile_number_. With two_write_queues it also protects alive_log_files_,
  // and log_empty_. Refer to the definition of each variable below for more
  // details.
  InstrumentedMutex log_write_mutex_;
+  // State below is protected by mutex_
+  // With two_write_queues enabled, some of the variables that accessed during
+  // WriteToWAL need different synchronization: log_empty_, alive_log_files_,
+  // logs_, logfile_number_. Refer to the definition of each variable below for
+  // more description.
+  mutable InstrumentedMutex mutex_;

  std::atomic<bool> shutting_down_;
  // This condition variable is signaled on these conditions:
@@ -1317,9 +1088,12 @@ class DBImpl : public DB {
  // read and writes are protected by log_write_mutex_ instead. This is to avoid
  // expesnive mutex_ lock during WAL write, which update log_empty_.
  bool log_empty_;
-
+  ColumnFamilyHandleImpl* default_cf_handle_;
+  InternalStats* default_cf_internal_stats_;
+  unique_ptr<ColumnFamilyMemTablesImpl> column_family_memtables_;
  struct LogFileNumberSize {
-    explicit LogFileNumberSize(uint64_t _number) : number(_number) {}
+    explicit LogFileNumberSize(uint64_t _number)
+        : number(_number) {}
    void AddSize(uint64_t new_size) { size += new_size; }
    uint64_t number;
    uint64_t size = 0;
@@ -1344,7 +1118,7 @@ class DBImpl : public DB {

    uint64_t number;
    // Visual Studio doesn't support deque's member to be noncopyable because
-    // of a std::unique_ptr as a member.
+    // of a unique_ptr as a member.
    log::Writer* writer;  // own
    // true for some prefix of logs_
    bool getting_synced = false;
@@ -1382,19 +1156,18 @@ class DBImpl : public DB {
  WriteBatch cached_recoverable_state_;
  std::atomic<bool> cached_recoverable_state_empty_ = {true};
  std::atomic<uint64_t> total_log_size_;
-
+  // only used for dynamically adjusting max_total_wal_size. it is a sum of
+  // [write_buffer_size * max_write_buffer_number] over all column families
+  uint64_t max_total_in_memory_state_;
+  // If true, we have only one (default) column family. We use this to optimize
+  // some code-paths
+  bool single_column_family_mode_;
  // If this is non-empty, we need to delete these log files in background
  // threads. Protected by db mutex.
  autovector<log::Writer*> logs_to_free_;

  bool is_snapshot_supported_;

-  std::map<uint64_t, std::map<std::string, uint64_t>> stats_history_;
-
-  std::map<std::string, uint64_t> stats_slice_;
-
-  bool stats_slice_initialized_ = false;
-
  // Class to maintain directories for all database paths other than main one.
  class Directories {
   public:
@@ -1431,7 +1204,7 @@ class DBImpl : public DB {

  WriteController write_controller_;

-  std::unique_ptr<RateLimiter> low_pri_write_rate_limiter_;
+  unique_ptr<RateLimiter> low_pri_write_rate_limiter_;

  // Size of the last batch group. In slowdown mode, next write needs to
  // sleep if it uses up the quota.
@@ -1480,14 +1253,14 @@ class DBImpl : public DB {
  // compacted. Consumers of these queues are flush and compaction threads. When
  // column family is put on this queue, we increase unscheduled_flushes_ and
  // unscheduled_compactions_. When these variables are bigger than zero, that
-  // means we need to schedule background threads for flush and compaction.
+  // means we need to schedule background threads for compaction and thread.
  // Once the background threads are scheduled, we decrease unscheduled_flushes_
  // and unscheduled_compactions_. That way we keep track of number of
  // compaction and flush threads we need to schedule. This scheduling is done
  // in MaybeScheduleFlushOrCompaction()
  // invariant(column family present in flush_queue_ <==>
  // ColumnFamilyData::pending_flush_ == true)
-  std::deque<FlushRequest> flush_queue_;
+  std::deque<ColumnFamilyData*> flush_queue_;
  // invariant(column family present in compaction_queue_ <==>
  // ColumnFamilyData::pending_compaction_ == true)
  std::deque<ColumnFamilyData*> compaction_queue_;
@@ -1531,15 +1304,15 @@ class DBImpl : public DB {
    uint32_t output_path_id;
    Status status;
    bool done;
-    bool in_progress;            // compaction request being processed?
-    bool incomplete;             // only part of requested range compacted
-    bool exclusive;              // current behavior of only one manual
-    bool disallow_trivial_move;  // Force actual compaction to run
-    const InternalKey* begin;    // nullptr means beginning of key range
-    const InternalKey* end;      // nullptr means end of key range
-    InternalKey* manual_end;     // how far we are compacting
-    InternalKey tmp_storage;     // Used to keep track of compaction progress
-    InternalKey tmp_storage1;    // Used to keep track of compaction progress
+    bool in_progress;             // compaction request being processed?
+    bool incomplete;              // only part of requested range compacted
+    bool exclusive;               // current behavior of only one manual
+    bool disallow_trivial_move;   // Force actual compaction to run
+    const InternalKey* begin;     // nullptr means beginning of key range
+    const InternalKey* end;       // nullptr means end of key range
+    InternalKey* manual_end;      // how far we are compacting
+    InternalKey tmp_storage;      // Used to keep track of compaction progress
+    InternalKey tmp_storage1;     // Used to keep track of compaction progress
  };
  struct PrepickedCompaction {
    // background compaction takes ownership of `compaction`.
@@ -1547,8 +1320,6 @@ class DBImpl : public DB {
    // caller retains ownership of `manual_compaction_state` as it is reused
    // across background compactions.
    ManualCompactionState* manual_compaction_state;  // nullptr if non-manual
-    // task limiter token is requested during compaction picking.
-    std::unique_ptr<TaskLimiterToken> task_token;
  };
  std::deque<ManualCompactionState*> manual_compaction_dequeue_;

@@ -1588,7 +1359,7 @@ class DBImpl : public DB {
  std::atomic<bool> has_unpersisted_data_;

  // if an attempt was made to flush all column families that
-  // the oldest log depends on but uncommitted data in the oldest
+  // the oldest log depends on but uncommited data in the oldest
  // log prevents the log from being released.
  // We must attempt to free the dependent memtables again
  // at a later time after the transaction in the oldest
@@ -1601,6 +1372,12 @@ class DBImpl : public DB {

  std::string db_absolute_path_;

+  // The options to access storage files
+  const EnvOptions env_options_;
+
+  // Additonal options for compaction and flush
+  EnvOptions env_options_for_compaction_;
+
  // Number of running IngestExternalFile() calls.
  // REQUIRES: mutex held
  int num_running_ingest_file_;
@@ -1624,10 +1401,6 @@ class DBImpl : public DB {
  // Indicate DB was opened successfully
  bool opened_successfully_;

-  // The min threshold to triggere bottommost compaction for removing
-  // garbages, among all column families.
-  SequenceNumber bottommost_files_mark_threshold_ = kMaxSequenceNumber;
-
  LogsWithPrepTracker logs_with_prep_tracker_;

  // Callback for compaction to check if a key is visible to a snapshot.
@@ -1638,14 +1411,6 @@ class DBImpl : public DB {
  // Only to be set during initialization
  std::unique_ptr<PreReleaseCallback> recoverable_state_pre_release_callback_;

-  // handle for scheduling stats dumping at fixed intervals
-  // REQUIRES: mutex locked
-  std::unique_ptr<rocksdb::RepeatableThread> thread_dump_stats_;
-
-  // handle for scheduling stats snapshoting at fixed intervals
-  // REQUIRES: mutex locked
-  std::unique_ptr<rocksdb::RepeatableThread> thread_persist_stats_;
-
  // No copying allowed
  DBImpl(const DBImpl&);
  void operator=(const DBImpl&);
@@ -1659,13 +1424,14 @@ class DBImpl : public DB {
  // state needs flush or compaction.
  void InstallSuperVersionAndScheduleWork(
      ColumnFamilyData* cfd, SuperVersionContext* sv_context,
-      const MutableCFOptions& mutable_cf_options);
+      const MutableCFOptions& mutable_cf_options,
+      FlushReason flush_reason = FlushReason::kOthers);

 #ifndef ROCKSDB_LITE
  using DB::GetPropertiesOfAllTables;
-  virtual Status GetPropertiesOfAllTables(
-      ColumnFamilyHandle* column_family,
-      TablePropertiesCollection* props) override;
+  virtual Status GetPropertiesOfAllTables(ColumnFamilyHandle* column_family,
+                                          TablePropertiesCollection* props)
+      override;
  virtual Status GetPropertiesOfTablesInRange(
      ColumnFamilyHandle* column_family, const Range* range, std::size_t n,
      TablePropertiesCollection* props) override;
@@ -1684,35 +1450,33 @@ class DBImpl : public DB {
  bool ShouldntRunManualCompaction(ManualCompactionState* m);
  bool HaveManualCompaction(ColumnFamilyData* cfd);
  bool MCOverlap(ManualCompactionState* m, ManualCompactionState* m1);
-#ifndef ROCKSDB_LITE
-  void BuildCompactionJobInfo(const ColumnFamilyData* cfd, Compaction* c,
-                              const Status& st,
-                              const CompactionJobStats& compaction_job_stats,
-                              const int job_id, const Version* current,
-                              CompactionJobInfo* compaction_job_info) const;
-  // Reserve the next 'num' file numbers for to-be-ingested external SST files,
-  // and return the current file_number in 'next_file_number'.
-  // Write a version edit to the MANIFEST.
-  Status ReserveFileNumbersBeforeIngestion(
-      ColumnFamilyData* cfd, uint64_t num,
-      std::list<uint64_t>::iterator* pending_output_elem,
-      uint64_t* next_file_number);
-#endif  //! ROCKSDB_LITE

  bool ShouldPurge(uint64_t file_number) const;
  void MarkAsGrabbedForPurge(uint64_t file_number);

  size_t GetWalPreallocateBlockSize(uint64_t write_buffer_size) const;
-  Env::WriteLifeTimeHint CalculateWALWriteHint() { return Env::WLTH_SHORT; }
-
-  Status CreateWAL(uint64_t log_file_num, uint64_t recycle_log_number,
-                   size_t preallocate_block_size, log::Writer** new_log);
+  Env::WriteLifeTimeHint CalculateWALWriteHint() {
+    return Env::WLTH_SHORT;
+  }

  // When set, we use a separate queue for writes that dont write to memtable.
  // In 2PC these are the writes at Prepare phase.
  const bool two_write_queues_;
  const bool manual_wal_flush_;
-
+  // Increase the sequence number after writing each batch, whether memtable is
+  // disabled for that or not. Otherwise the sequence number is increased after
+  // writing each key into memtable. This implies that when disable_memtable is
+  // set, the seq is not increased at all.
+  //
+  // Default: false
+  const bool seq_per_batch_;
+  // This determines during recovery whether we expect one writebatch per
+  // recovered transaction, or potentially multiple writebatches per
+  // transaction. For WriteUnprepared, this is set to false, since multiple
+  // batches can exist per transaction.
+  //
+  // Default: true
+  const bool batch_per_txn_;
  // LastSequence also indicates last published sequence visibile to the
  // readers. Otherwise LastPublishedSequence should be used.
  const bool last_seq_same_as_publish_seq_;
@@ -1720,16 +1484,6 @@ class DBImpl : public DB {
  // flush/compaction and if it is not provided vis SnapshotChecker, we should
  // disable gc to be safe.
  const bool use_custom_gc_;
-  // Flag to indicate that the DB instance shutdown has been initiated. This
-  // different from shutting_down_ atomic in that it is set at the beginning
-  // of shutdown sequence, specifically in order to prevent any background
-  // error recovery from going on in parallel. The latter, shutting_down_,
-  // is set a little later during the shutdown after scheduling memtable
-  // flushes
-  std::atomic<bool> shutdown_initiated_;
-  // Flag to indicate whether sst_file_manager object was allocated in
-  // DB::Open() or passed to us
-  bool own_sfm_;

  // Clients must periodically call SetPreserveDeletesSequenceNumber()
  // to advance this seqnum. Default value is 0 which means ALL deletes are
@@ -1742,20 +1496,10 @@ class DBImpl : public DB {
  bool closed_;

  ErrorHandler error_handler_;
-
-  // Conditional variable to coordinate installation of atomic flush results.
-  // With atomic flush, each bg thread installs the result of flushing multiple
-  // column families, and different threads can flush different column
-  // families. It's difficult to rely on one thread to perform batch
-  // installation for all threads. This is different from the non-atomic flush
-  // case.
-  // atomic_flush_install_cv_ makes sure that threads install atomic flush
-  // results sequentially. Flush results of memtables with lower IDs get
-  // installed to MANIFEST first.
-  InstrumentedCondVar atomic_flush_install_cv_;
 };

-extern Options SanitizeOptions(const std::string& db, const Options& src);
+extern Options SanitizeOptions(const std::string& db,
+                               const Options& src);

 extern DBOptions SanitizeOptions(const std::string& db, const DBOptions& src);

@@ -26,16 +26,10 @@ void DBImpl::TEST_SwitchWAL() {
  SwitchWAL(&write_context);
 }

-bool DBImpl::TEST_WALBufferIsEmpty(bool lock) {
-  if (lock) {
-    log_write_mutex_.Lock();
-  }
+bool DBImpl::TEST_WALBufferIsEmpty() {
+  InstrumentedMutexLock wl(&log_write_mutex_);
  log::Writer* cur_log_writer = logs_.back().writer;
-  auto res = cur_log_writer->TEST_BufferIsEmpty();
-  if (lock) {
-    log_write_mutex_.Unlock();
-  }
-  return res;
+  return cur_log_writer->TEST_BufferIsEmpty();
 }

 int64_t DBImpl::TEST_MaxNextLevelOverlappingBytes(
@@ -93,9 +87,8 @@ Status DBImpl::TEST_CompactRange(int level, const Slice* begin,
       cfd->ioptions()->compaction_style == kCompactionStyleFIFO)
          ? level
          : level + 1;
-  return RunManualCompaction(cfd, level, output_level, CompactRangeOptions(),
-                             begin, end, true, disallow_trivial_move,
-                             port::kMaxUint64 /*max_file_num_to_ignore*/);
+  return RunManualCompaction(cfd, level, output_level, 0, 0, begin, end, true,
+                             disallow_trivial_move);
 }

 Status DBImpl::TEST_SwitchMemtable(ColumnFamilyData* cfd) {
@@ -108,7 +101,7 @@ Status DBImpl::TEST_SwitchMemtable(ColumnFamilyData* cfd) {
 }

 Status DBImpl::TEST_FlushMemTable(bool wait, bool allow_write_stall,
-                                  ColumnFamilyHandle* cfh) {
+    ColumnFamilyHandle* cfh) {
  FlushOptions fo;
  fo.wait = wait;
  fo.allow_write_stall = allow_write_stall;
@@ -130,7 +123,7 @@ Status DBImpl::TEST_WaitForFlushMemTable(ColumnFamilyHandle* column_family) {
    auto cfh = reinterpret_cast<ColumnFamilyHandleImpl*>(column_family);
    cfd = cfh->cfd();
  }
-  return WaitForFlushMemTable(cfd, nullptr, false);
+  return WaitForFlushMemTable(cfd);
 }

 Status DBImpl::TEST_WaitForCompact(bool wait_unscheduled) {
@@ -144,15 +137,19 @@ Status DBImpl::TEST_WaitForCompact(bool wait_unscheduled) {
  while ((bg_bottom_compaction_scheduled_ || bg_compaction_scheduled_ ||
          bg_flush_scheduled_ ||
          (wait_unscheduled && unscheduled_compactions_)) &&
-         (error_handler_.GetBGError() == Status::OK())) {
+         !error_handler_.IsDBStopped()) {
    bg_cv_.Wait();
  }
  return error_handler_.GetBGError();
 }

-void DBImpl::TEST_LockMutex() { mutex_.Lock(); }
+void DBImpl::TEST_LockMutex() {
+  mutex_.Lock();
+}

-void DBImpl::TEST_UnlockMutex() { mutex_.Unlock(); }
+void DBImpl::TEST_UnlockMutex() {
+  mutex_.Unlock();
+}

 void* DBImpl::TEST_BeginWrite() {
  auto w = new WriteThread::Writer();
@@ -240,30 +237,5 @@ SequenceNumber DBImpl::TEST_GetLastVisibleSequence() const {
  }
 }

-size_t DBImpl::TEST_GetWalPreallocateBlockSize(
-    uint64_t write_buffer_size) const {
-  InstrumentedMutexLock l(&mutex_);
-  return GetWalPreallocateBlockSize(write_buffer_size);
-}
-
-void DBImpl::TEST_WaitForDumpStatsRun(std::function<void()> callback) const {
-  if (thread_dump_stats_ != nullptr) {
-    thread_dump_stats_->TEST_WaitForRun(callback);
-  }
-}
-
-void DBImpl::TEST_WaitForPersistStatsRun(std::function<void()> callback) const {
-  if (thread_persist_stats_ != nullptr) {
-    thread_persist_stats_->TEST_WaitForRun(callback);
-  }
-}
-
-bool DBImpl::TEST_IsPersistentStatsEnabled() const {
-  return thread_persist_stats_ && thread_persist_stats_->IsRunning();
-}
-
-size_t DBImpl::TEST_EstiamteStatsHistorySize() const {
-  return EstiamteStatsHistorySize();
-}
 }  // namespace rocksdb
 #endif  // NDEBUG
@@ -20,7 +20,6 @@
 #include "util/sst_file_manager_impl.h"

 namespace rocksdb {
-
 uint64_t DBImpl::MinLogNumberToKeep() {
  if (allow_2pc()) {
    return versions_->min_log_number_to_keep_2pc();
@@ -29,14 +28,6 @@ uint64_t DBImpl::MinLogNumberToKeep() {
  }
 }

-uint64_t DBImpl::MinObsoleteSstNumberToKeep() {
-  mutex_.AssertHeld();
-  if (!pending_outputs_.empty()) {
-    return *pending_outputs_.begin();
-  }
-  return std::numeric_limits<uint64_t>::max();
-}
-
 // * Returns the list of live files in 'sst_live'
 // If it's doing full scan:
 // * Returns the list of all files in the filesystem in
@@ -108,7 +99,7 @@ void DBImpl::FindObsoleteFiles(JobContext* job_context, bool force,
  versions_->AddLiveFiles(&job_context->sst_live);
  if (doing_the_full_scan) {
    InfoLogPrefix info_log_prefix(!immutable_db_options_.db_log_dir.empty(),
-                                  dbname_);
+        dbname_);
    std::set<std::string> paths;
    for (size_t path_id = 0; path_id < immutable_db_options_.db_paths.size();
         path_id++) {
@@ -152,7 +143,8 @@ void DBImpl::FindObsoleteFiles(JobContext* job_context, bool force,
        }

        // TODO(icanadi) clean up this mess to avoid having one-off "/" prefixes
-        job_context->full_scan_candidate_files.emplace_back("/" + file, path);
+        job_context->full_scan_candidate_files.emplace_back(
+            "/" + file, path);
      }
    }

@@ -162,8 +154,8 @@ void DBImpl::FindObsoleteFiles(JobContext* job_context, bool force,
      env_->GetChildren(immutable_db_options_.wal_dir,
                        &log_files);  // Ignore errors
      for (const std::string& log_file : log_files) {
-        job_context->full_scan_candidate_files.emplace_back(
-            log_file, immutable_db_options_.wal_dir);
+        job_context->full_scan_candidate_files.emplace_back(log_file,
+            immutable_db_options_.wal_dir);
      }
    }
    // Add info log files in db_log_dir
@@ -173,8 +165,8 @@ void DBImpl::FindObsoleteFiles(JobContext* job_context, bool force,
      // Ignore errors
      env_->GetChildren(immutable_db_options_.db_log_dir, &info_log_files);
      for (std::string& log_file : info_log_files) {
-        job_context->full_scan_candidate_files.emplace_back(
-            log_file, immutable_db_options_.db_log_dir);
+        job_context->full_scan_candidate_files.emplace_back(log_file,
+            immutable_db_options_.db_log_dir);
      }
    }
  }
@@ -259,14 +251,14 @@ void DBImpl::DeleteObsoleteFileImpl(int job_id, const std::string& fname,
                                    const std::string& path_to_sync,
                                    FileType type, uint64_t number) {
  Status file_deletion_status;
-  if (type == kTableFile || type == kLogFile) {
+  if (type == kTableFile) {
    file_deletion_status =
-        DeleteDBFile(&immutable_db_options_, fname, path_to_sync);
+        DeleteSSTFile(&immutable_db_options_, fname, path_to_sync);
  } else {
    file_deletion_status = env_->DeleteFile(fname);
  }
  TEST_SYNC_POINT_CALLBACK("DBImpl::DeleteObsoleteFileImpl:AfterDeletion",
-                           &file_deletion_status);
+      &file_deletion_status);
  if (file_deletion_status.ok()) {
    ROCKS_LOG_DEBUG(immutable_db_options_.info_log,
                    "[JOB %d] Delete %s type=%d #%" PRIu64 " -- %s\n", job_id,
@@ -321,8 +313,7 @@ void DBImpl::PurgeObsoleteFiles(JobContext& state, bool schedule_only) {
  const char* kDumbDbName = "";
  for (auto& file : state.sst_delete_files) {
    candidate_files.emplace_back(
-        MakeTableFileName(kDumbDbName, file.metadata->fd.GetNumber()),
-        file.path);
+        MakeTableFileName(kDumbDbName, file.metadata->fd.GetNumber()), file.path);
    if (file.metadata->table_reader_handle) {
      table_cache_->Release(file.metadata->table_reader_handle);
    }
@@ -332,7 +323,7 @@ void DBImpl::PurgeObsoleteFiles(JobContext& state, bool schedule_only) {
  for (auto file_num : state.log_delete_files) {
    if (file_num > 0) {
      candidate_files.emplace_back(LogFileName(kDumbDbName, file_num),
-                                   immutable_db_options_.wal_dir);
+          immutable_db_options_.wal_dir);
    }
  }
  for (const auto& filename : state.manifest_delete_files) {
@@ -465,12 +456,7 @@ void DBImpl::PurgeObsoleteFiles(JobContext& state, bool schedule_only) {
    } else {
      dir_to_sync =
          (type == kLogFile) ? immutable_db_options_.wal_dir : dbname_;
-      fname = dir_to_sync +
-              ((!dir_to_sync.empty() && dir_to_sync.back() == '/') ||
-                       (!to_delete.empty() && to_delete.front() == '/')
-                   ? ""
-                   : "/") +
-              to_delete;
+      fname = dir_to_sync + "/" + to_delete;
    }

 #ifndef ROCKSDB_LITE
@@ -481,11 +467,6 @@ void DBImpl::PurgeObsoleteFiles(JobContext& state, bool schedule_only) {
    }
 #endif  // !ROCKSDB_LITE

-    for (const auto w : state.logs_to_free) {
-      // TODO: maybe check the return value of Close.
-      w->Close();
-    }
-
    Status file_deletion_status;
    if (schedule_only) {
      InstrumentedMutexLock guard_lock(&mutex_);
@@ -23,7 +23,8 @@
 #include "util/sync_point.h"

 namespace rocksdb {
-Options SanitizeOptions(const std::string& dbname, const Options& src) {
+Options SanitizeOptions(const std::string& dbname,
+                        const Options& src) {
  auto db_options = SanitizeOptions(dbname, DBOptions(src));
  ImmutableDBOptions immutable_db_options(db_options);
  auto cf_options =
@@ -41,8 +42,6 @@ DBOptions SanitizeOptions(const std::string& dbname, const DBOptions& src) {
      max_max_open_files = 0x400000;
    }
    ClipToRange(&result.max_open_files, 20, max_max_open_files);
-    TEST_SYNC_POINT_CALLBACK("SanitizeOptions::AfterChangeMaxOpenFiles",
-                             &result.max_open_files);
  }

  if (result.info_log == nullptr) {
@@ -57,9 +56,10 @@ DBOptions SanitizeOptions(const std::string& dbname, const DBOptions& src) {
    result.write_buffer_manager.reset(
        new WriteBufferManager(result.db_write_buffer_size));
  }
-  auto bg_job_limits = DBImpl::GetBGJobLimits(
-      result.max_background_flushes, result.max_background_compactions,
-      result.max_background_jobs, true /* parallelize_compactions */);
+  auto bg_job_limits = DBImpl::GetBGJobLimits(result.max_background_flushes,
+                                              result.max_background_compactions,
+                                              result.max_background_jobs,
+                                              true /* parallelize_compactions */);
  result.env->IncBackgroundThreadsIfNeeded(bg_job_limits.max_compactions,
                                           Env::Priority::LOW);
  result.env->IncBackgroundThreadsIfNeeded(bg_job_limits.max_flushes,
@@ -107,12 +107,14 @@ DBOptions SanitizeOptions(const std::string& dbname, const DBOptions& src) {
    result.db_paths.emplace_back(dbname, std::numeric_limits<uint64_t>::max());
  }

-  if (result.use_direct_reads && result.compaction_readahead_size == 0) {
+  if (result.use_direct_reads &&
+      result.compaction_readahead_size == 0) {
    TEST_SYNC_POINT_CALLBACK("SanitizeOptions:direct_io", nullptr);
    result.compaction_readahead_size = 1024 * 1024 * 2;
  }

-  if (result.compaction_readahead_size > 0 || result.use_direct_reads) {
+  if (result.compaction_readahead_size > 0 ||
+      result.use_direct_reads) {
    result.new_table_reader_for_compaction_inputs = true;
  }

@@ -132,15 +134,8 @@ DBOptions SanitizeOptions(const std::string& dbname, const DBOptions& src) {
  for (size_t i = 0; i < result.db_paths.size(); i++) {
    DeleteScheduler::CleanupDirectory(result.env, sfm, result.db_paths[i].path);
  }
-
-  // Create a default SstFileManager for purposes of tracking compaction size
-  // and facilitating recovery from out of space errors.
-  if (result.sst_file_manager.get() == nullptr) {
-    std::shared_ptr<SstFileManager> sst_file_manager(
-        NewSstFileManager(result.env, result.info_log));
-    result.sst_file_manager = sst_file_manager;
-  }
 #endif
+
  return result;
 }

@@ -176,7 +171,7 @@ static Status ValidateOptions(
      return s;
    }

-    if (cfd.options.ttl > 0) {
+    if (cfd.options.ttl > 0 || cfd.options.compaction_options_fifo.ttl > 0) {
      if (db_options.max_open_files != -1) {
        return Status::NotSupported(
            "TTL is only supported when files are always "
@@ -188,20 +183,6 @@ static Status ValidateOptions(
            "TTL is only supported in Block-Based Table format. ");
      }
    }
-
-    if (cfd.options.periodic_compaction_seconds > 0) {
-      if (db_options.max_open_files != -1) {
-        return Status::NotSupported(
-            "Periodic Compaction is only supported when files are always "
-            "kept open (set max_open_files = -1). ");
-      }
-      if (cfd.options.table_factory->Name() !=
-          BlockBasedTableFactory().Name()) {
-        return Status::NotSupported(
-            "Periodic Compaction is only supported in "
-            "Block-Based Table format. ");
-      }
-    }
  }

  if (db_options.db_paths.size() > 4) {
@@ -230,7 +211,7 @@ static Status ValidateOptions(

  return Status::OK();
 }
-}  // namespace
+} // namespace
 Status DBImpl::NewDB() {
  VersionEdit new_db;
  new_db.SetLogNumber(0);
@@ -242,7 +223,7 @@ Status DBImpl::NewDB() {
  ROCKS_LOG_INFO(immutable_db_options_.info_log, "Creating manifest 1 \n");
  const std::string manifest = DescriptorFileName(dbname_, 1);
  {
-    std::unique_ptr<WritableFile> file;
+    unique_ptr<WritableFile> file;
    EnvOptions env_options = env_->OptimizeForManifestWrite(env_options_);
    s = NewWritableFile(env_, manifest, &file, env_options);
    if (!s.ok()) {
@@ -250,9 +231,8 @@ Status DBImpl::NewDB() {
    }
    file->SetPreallocationBlockSize(
        immutable_db_options_.manifest_preallocation_size);
-    std::unique_ptr<WritableFileWriter> file_writer(new WritableFileWriter(
-        std::move(file), manifest, env_options, env_, nullptr /* stats */,
-        immutable_db_options_.listeners));
+    unique_ptr<WritableFileWriter> file_writer(
+        new WritableFileWriter(std::move(file), env_options));
    log::Writer log(std::move(file_writer), 0, false);
    std::string record;
    new_db.EncodeTo(&record);
@@ -270,8 +250,9 @@ Status DBImpl::NewDB() {
  return s;
 }

-Status DBImpl::CreateAndNewDirectory(Env* env, const std::string& dirname,
-                                     std::unique_ptr<Directory>* directory) {
+Status DBImpl::CreateAndNewDirectory(
+    Env* env, const std::string& dirname,
+    std::unique_ptr<Directory>* directory) {
  // We call CreateDirIfMissing() as the directory may already exist (if we
  // are reopening a DB), when this happens we don't want creating the
  // directory to cause an error. However, we need to check if creating the
@@ -352,8 +333,8 @@ Status DBImpl::Recover(
      }
    } else if (s.ok()) {
      if (immutable_db_options_.error_if_exists) {
-        return Status::InvalidArgument(dbname_,
-                                       "exists (error_if_exists is true)");
+        return Status::InvalidArgument(
+            dbname_, "exists (error_if_exists is true)");
      }
    } else {
      // Unexpected error reading file
@@ -373,7 +354,7 @@ Status DBImpl::Recover(
    }
    // Verify compatibility of env_options_ and filesystem
    {
-      std::unique_ptr<RandomAccessFile> idfile;
+      unique_ptr<RandomAccessFile> idfile;
      EnvOptions customized_env(env_options_);
      customized_env.use_direct_reads |=
          immutable_db_options_.use_direct_io_for_flush_and_compaction;
@@ -408,16 +389,6 @@ Status DBImpl::Recover(
      }
    }
  }
-
-  // Initial max_total_in_memory_state_ before recovery logs. Log recovery
-  // may check this value to decide whether to flush.
-  max_total_in_memory_state_ = 0;
-  for (auto cfd : *versions_->GetColumnFamilySet()) {
-    auto* mutable_cf_options = cfd->GetLatestMutableCFOptions();
-    max_total_in_memory_state_ += mutable_cf_options->write_buffer_size *
-                                  mutable_cf_options->max_write_buffer_number;
-  }
-
  if (s.ok()) {
    SequenceNumber next_sequence(kMaxSequenceNumber);
    default_cf_handle_ = new ColumnFamilyHandleImpl(
@@ -435,10 +406,7 @@ Status DBImpl::Recover(
    // produced by an older version of rocksdb.
    std::vector<std::string> filenames;
    s = env_->GetChildren(immutable_db_options_.wal_dir, &filenames);
-    if (s.IsNotFound()) {
-      return Status::InvalidArgument("wal_dir not found",
-                                     immutable_db_options_.wal_dir);
-    } else if (!s.ok()) {
+    if (!s.ok()) {
      return s;
    }

@@ -493,26 +461,12 @@ Status DBImpl::Recover(
    }
  }

-  if (read_only) {
-    // If we are opening as read-only, we need to update options_file_number_
-    // to reflect the most recent OPTIONS file. It does not matter for regular
-    // read-write db instance because options_file_number_ will later be
-    // updated to versions_->NewFileNumber() in RenameTempFileToOptionsFile.
-    std::vector<std::string> file_names;
-    if (s.ok()) {
-      s = env_->GetChildren(GetName(), &file_names);
-    }
-    if (s.ok()) {
-      uint64_t number = 0;
-      uint64_t options_file_number = 0;
-      FileType type;
-      for (const auto& fname : file_names) {
-        if (ParseFileName(fname, &number, &type) && type == kOptionsFile) {
-          options_file_number = std::max(number, options_file_number);
-        }
-      }
-      versions_->options_file_number_ = options_file_number;
-    }
+  // Initial value
+  max_total_in_memory_state_ = 0;
+  for (auto cfd : *versions_->GetColumnFamilySet()) {
+    auto* mutable_cf_options = cfd->GetLatestMutableCFOptions();
+    max_total_in_memory_state_ += mutable_cf_options->write_buffer_size *
+                                  mutable_cf_options->max_write_buffer_number;
  }

  return s;
@@ -526,7 +480,7 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
    Logger* info_log;
    const char* fname;
    Status* status;  // nullptr if immutable_db_options_.paranoid_checks==false
-    void Corruption(size_t bytes, const Status& s) override {
+    virtual void Corruption(size_t bytes, const Status& s) override {
      ROCKS_LOG_WARN(info_log, "%s%s: dropping %d bytes; %s",
                     (this->status == nullptr ? "(ignoring error) " : ""),
                     fname, static_cast<int>(bytes), s.ToString().c_str());
@@ -563,9 +517,10 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
    std::map<std::string, uint32_t> cf_name_id_map;
    std::map<uint32_t, uint64_t> cf_lognumber_map;
    for (auto cfd : *versions_->GetColumnFamilySet()) {
-      cf_name_id_map.insert(std::make_pair(cfd->GetName(), cfd->GetID()));
+      cf_name_id_map.insert(
+        std::make_pair(cfd->GetName(), cfd->GetID()));
      cf_lognumber_map.insert(
-          std::make_pair(cfd->GetID(), cfd->GetLogNumber()));
+        std::make_pair(cfd->GetID(), cfd->GetLogNumber()));
    }

    immutable_db_options_.wal_filter->ColumnFamilyLogNumberMap(cf_lognumber_map,
@@ -594,7 +549,7 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,

    ROCKS_LOG_INFO(immutable_db_options_.info_log,
                   "Recovering log #%" PRIu64 " mode %d", log_number,
-                   static_cast<int>(immutable_db_options_.wal_recovery_mode));
+                   immutable_db_options_.wal_recovery_mode);
    auto logFileDropped = [this, &fname]() {
      uint64_t bytes;
      if (env_->GetFileSize(fname, &bytes).ok()) {
@@ -608,9 +563,9 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
      continue;
    }

-    std::unique_ptr<SequentialFileReader> file_reader;
+    unique_ptr<SequentialFileReader> file_reader;
    {
-      std::unique_ptr<SequentialFile> file;
+      unique_ptr<SequentialFile> file;
      status = env_->NewSequentialFile(fname, &file,
                                       env_->OptimizeForLogRead(env_options_));
      if (!status.ok()) {
@@ -643,7 +598,8 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
    // to be skipped instead of propagating bad information (like overly
    // large sequence numbers).
    log::Reader reader(immutable_db_options_.info_log, std::move(file_reader),
-                       &reporter, true /*checksum*/, log_number);
+                       &reporter, true /*checksum*/, 0 /*initial_offset*/,
+                       log_number);

    // Determine if we should tolerate incomplete records at the tail end of the
    // Read all the records and add to a memtable
@@ -737,8 +693,7 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
                " mode %d log filter %s returned "
                "more records (%d) than original (%d) which is not allowed. "
                "Aborting recovery.",
-                log_number,
-                static_cast<int>(immutable_db_options_.wal_recovery_mode),
+                log_number, immutable_db_options_.wal_recovery_mode,
                immutable_db_options_.wal_filter->Name(), new_count,
                original_count);
            status = Status::NotSupported(
@@ -915,8 +870,8 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
      // VersionSet::next_file_number_ always to be strictly greater than any
      // log number
      versions_->MarkFileNumberUsed(max_log_number + 1);
-      status = versions_->LogAndApply(cfd, *cfd->GetLatestMutableCFOptions(),
-                                      edit, &mutex_);
+      status = versions_->LogAndApply(
+          cfd, *cfd->GetLatestMutableCFOptions(), edit, &mutex_);
      if (!status.ok()) {
        // Recovery failed
        break;
@@ -924,8 +879,18 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
    }
  }

-  if (status.ok() && data_seen && !flushed) {
-    status = RestoreAliveLogFiles(log_numbers);
+  if (data_seen && !flushed) {
+    // Mark these as alive so they'll be considered for deletion later by
+    // FindObsoleteFiles()
+    if (two_write_queues_) {
+      log_write_mutex_.Lock();
+    }
+    for (auto log_number : log_numbers) {
+      alive_log_files_.push_back(LogFileNumberSize(log_number));
+    }
+    if (two_write_queues_) {
+      log_write_mutex_.Unlock();
+    }
  }

  event_logger_.Log() << "job" << job_id << "event"
@@ -934,60 +899,6 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
  return status;
 }

-Status DBImpl::RestoreAliveLogFiles(const std::vector<uint64_t>& log_numbers) {
-  if (log_numbers.empty()) {
-    return Status::OK();
-  }
-  Status s;
-  mutex_.AssertHeld();
-  assert(immutable_db_options_.avoid_flush_during_recovery);
-  if (two_write_queues_) {
-    log_write_mutex_.Lock();
-  }
-  // Mark these as alive so they'll be considered for deletion later by
-  // FindObsoleteFiles()
-  total_log_size_ = 0;
-  log_empty_ = false;
-  for (auto log_number : log_numbers) {
-    LogFileNumberSize log(log_number);
-    std::string fname = LogFileName(immutable_db_options_.wal_dir, log_number);
-    // This gets the appear size of the logs, not including preallocated space.
-    s = env_->GetFileSize(fname, &log.size);
-    if (!s.ok()) {
-      break;
-    }
-    total_log_size_ += log.size;
-    alive_log_files_.push_back(log);
-    // We preallocate space for logs, but then after a crash and restart, those
-    // preallocated space are not needed anymore. It is likely only the last
-    // log has such preallocated space, so we only truncate for the last log.
-    if (log_number == log_numbers.back()) {
-      std::unique_ptr<WritableFile> last_log;
-      Status truncate_status = env_->ReopenWritableFile(
-          fname, &last_log,
-          env_->OptimizeForLogWrite(
-              env_options_,
-              BuildDBOptions(immutable_db_options_, mutable_db_options_)));
-      if (truncate_status.ok()) {
-        truncate_status = last_log->Truncate(log.size);
-      }
-      if (truncate_status.ok()) {
-        truncate_status = last_log->Close();
-      }
-      // Not a critical error if fail to truncate.
-      if (!truncate_status.ok()) {
-        ROCKS_LOG_WARN(immutable_db_options_.info_log,
-                       "Failed to truncate log #%" PRIu64 ": %s", log_number,
-                       truncate_status.ToString().c_str());
-      }
-    }
-  }
-  if (two_write_queues_) {
-    log_write_mutex_.Unlock();
-  }
-  return s;
-}
-
 Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
                                           MemTable* mem, VersionEdit* edit) {
  mutex_.AssertHeld();
@@ -1029,21 +940,14 @@ Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
      if (use_custom_gc_ && snapshot_checker == nullptr) {
        snapshot_checker = DisableGCSnapshotChecker::Instance();
      }
-      std::vector<std::unique_ptr<FragmentedRangeTombstoneIterator>>
-          range_del_iters;
-      auto range_del_iter =
-          mem->NewRangeTombstoneIterator(ro, kMaxSequenceNumber);
-      if (range_del_iter != nullptr) {
-        range_del_iters.emplace_back(range_del_iter);
-      }
      s = BuildTable(
          dbname_, env_, *cfd->ioptions(), mutable_cf_options,
          env_options_for_compaction_, cfd->table_cache(), iter.get(),
-          std::move(range_del_iters), &meta, cfd->internal_comparator(),
+          std::unique_ptr<InternalIterator>(mem->NewRangeTombstoneIterator(ro)),
+          &meta, cfd->internal_comparator(),
          cfd->int_tbl_prop_collector_factories(), cfd->GetID(), cfd->GetName(),
          snapshot_seqs, earliest_write_conflict_snapshot, snapshot_checker,
          GetCompressionFlush(*cfd->ioptions(), mutable_cf_options),
-          mutable_cf_options.sample_for_compression,
          cfd->ioptions()->compression_opts, paranoid_file_checks,
          cfd->internal_stats(), TableFileCreationReason::kRecovery,
          &event_logger_, job_id, Env::IO_HIGH, nullptr /* table_properties */,
@@ -1073,9 +977,9 @@ Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
  stats.micros = env_->NowMicros() - start_micros;
  stats.bytes_written = meta.fd.GetFileSize();
  stats.num_output_files = 1;
-  cfd->internal_stats()->AddCompactionStats(level, Env::Priority::USER, stats);
-  cfd->internal_stats()->AddCFStats(InternalStats::BYTES_FLUSHED,
-                                    meta.fd.GetFileSize());
+  cfd->internal_stats()->AddCompactionStats(level, stats);
+  cfd->internal_stats()->AddCFStats(
+      InternalStats::BYTES_FLUSHED, meta.fd.GetFileSize());
  RecordTick(stats_, COMPACT_WRITE_BYTES, meta.fd.GetFileSize());
  return s;
 }
@@ -1106,45 +1010,6 @@ Status DB::Open(const DBOptions& db_options, const std::string& dbname,
                      !kSeqPerBatch, kBatchPerTxn);
 }

-Status DBImpl::CreateWAL(uint64_t log_file_num, uint64_t recycle_log_number,
-                         size_t preallocate_block_size, log::Writer** new_log) {
-  Status s;
-  std::unique_ptr<WritableFile> lfile;
-
-  DBOptions db_options =
-      BuildDBOptions(immutable_db_options_, mutable_db_options_);
-  EnvOptions opt_env_options =
-      env_->OptimizeForLogWrite(env_options_, db_options);
-  std::string log_fname =
-      LogFileName(immutable_db_options_.wal_dir, log_file_num);
-
-  if (recycle_log_number) {
-    ROCKS_LOG_INFO(immutable_db_options_.info_log,
-                   "reusing log %" PRIu64 " from recycle list\n",
-                   recycle_log_number);
-    std::string old_log_fname =
-        LogFileName(immutable_db_options_.wal_dir, recycle_log_number);
-    s = env_->ReuseWritableFile(log_fname, old_log_fname, &lfile,
-                                opt_env_options);
-  } else {
-    s = NewWritableFile(env_, log_fname, &lfile, opt_env_options);
-  }
-
-  if (s.ok()) {
-    lfile->SetWriteLifeTimeHint(CalculateWALWriteHint());
-    lfile->SetPreallocationBlockSize(preallocate_block_size);
-
-    const auto& listeners = immutable_db_options_.listeners;
-    std::unique_ptr<WritableFileWriter> file_writer(
-        new WritableFileWriter(std::move(lfile), log_fname, opt_env_options,
-                               env_, nullptr /* stats */, listeners));
-    *new_log = new log::Writer(std::move(file_writer), log_file_num,
-                               immutable_db_options_.recycle_log_file_num > 0,
-                               immutable_db_options_.manual_wal_flush);
-  }
-  return s;
-}
-
 Status DBImpl::Open(const DBOptions& db_options, const std::string& dbname,
                    const std::vector<ColumnFamilyDescriptor>& column_families,
                    std::vector<ColumnFamilyHandle*>* handles, DB** dbptr,
@@ -1186,12 +1051,6 @@ Status DBImpl::Open(const DBOptions& db_options, const std::string& dbname,
        break;
      }
    }
-
-    // For recovery from NoSpace() error, we can only handle
-    // the case where the database is stored in a single path
-    if (paths.size() <= 1) {
-      impl->error_handler_.EnableAutoRecovery();
-    }
  }

  if (!s.ok()) {
@@ -1205,23 +1064,38 @@ Status DBImpl::Open(const DBOptions& db_options, const std::string& dbname,
    return s;
  }
  impl->mutex_.Lock();
+  auto write_hint = impl->CalculateWALWriteHint();
  // Handles create_if_missing, error_if_exists
  s = impl->Recover(column_families);
  if (s.ok()) {
    uint64_t new_log_number = impl->versions_->NewFileNumber();
-    log::Writer* new_log = nullptr;
-    const size_t preallocate_block_size =
-        impl->GetWalPreallocateBlockSize(max_write_buffer_size);
-    s = impl->CreateWAL(new_log_number, 0 /*recycle_log_number*/,
-                        preallocate_block_size, &new_log);
+    unique_ptr<WritableFile> lfile;
+    EnvOptions soptions(db_options);
+    EnvOptions opt_env_options =
+        impl->immutable_db_options_.env->OptimizeForLogWrite(
+            soptions, BuildDBOptions(impl->immutable_db_options_,
+                                     impl->mutable_db_options_));
+    s = NewWritableFile(
+        impl->immutable_db_options_.env,
+        LogFileName(impl->immutable_db_options_.wal_dir, new_log_number),
+        &lfile, opt_env_options);
    if (s.ok()) {
-      InstrumentedMutexLock wl(&impl->log_write_mutex_);
-      impl->logfile_number_ = new_log_number;
-      assert(new_log != nullptr);
-      impl->logs_.emplace_back(new_log_number, new_log);
-    }
+      lfile->SetWriteLifeTimeHint(write_hint);
+      lfile->SetPreallocationBlockSize(
+          impl->GetWalPreallocateBlockSize(max_write_buffer_size));
+      {
+        InstrumentedMutexLock wl(&impl->log_write_mutex_);
+        impl->logfile_number_ = new_log_number;
+        unique_ptr<WritableFileWriter> file_writer(
+            new WritableFileWriter(std::move(lfile), opt_env_options));
+        impl->logs_.emplace_back(
+            new_log_number,
+            new log::Writer(
+                std::move(file_writer), new_log_number,
+                impl->immutable_db_options_.recycle_log_file_num > 0,
+                impl->immutable_db_options_.manual_wal_flush));
+      }

-    if (s.ok()) {
      // set column family handles
      for (auto cf : column_families) {
        auto cfd =
@@ -1290,8 +1164,7 @@ Status DBImpl::Open(const DBOptions& db_options, const std::string& dbname,
          !cfd->mem()->IsMergeOperatorSupported()) {
        s = Status::InvalidArgument(
            "The memtable of column family %s does not support merge operator "
-            "its options.merge_operator is non-null",
-            cfd->GetName().c_str());
+            "its options.merge_operator is non-null", cfd->GetName().c_str());
      }
      if (!s.ok()) {
        break;
@@ -1341,20 +1214,11 @@ Status DBImpl::Open(const DBOptions& db_options, const std::string& dbname,
        }
      }
    }
-
-    // Reserve some disk buffer space. This is a heuristic - when we run out
-    // of disk space, this ensures that there is atleast write_buffer_size
-    // amount of free space before we resume DB writes. In low disk space
-    // conditions, we want to avoid a lot of small L0 files due to frequent
-    // WAL write failures and resultant forced flushes
-    sfm->ReserveDiskBuffer(max_write_buffer_size,
-                           impl->immutable_db_options_.db_paths[0].path);
  }
 #endif  // !ROCKSDB_LITE

  if (s.ok()) {
-    ROCKS_LOG_HEADER(impl->immutable_db_options_.info_log, "DB pointer %p",
-                     impl);
+    ROCKS_LOG_INFO(impl->immutable_db_options_.info_log, "DB pointer %p", impl);
    LogFlush(impl->immutable_db_options_.info_log);
    assert(impl->TEST_WALBufferIsEmpty());
    // If the assert above fails then we need to FlushWAL before returning
@@ -1365,9 +1229,6 @@ Status DBImpl::Open(const DBOptions& db_options, const std::string& dbname,
          persist_options_status.ToString());
    }
  }
-  if (s.ok()) {
-    impl->StartTimedTasks();
-  }
  if (!s.ok()) {
    for (auto* h : *handles) {
      delete h;
@@ -9,6 +9,7 @@
 #include "db/db_impl.h"
 #include "db/db_iter.h"
 #include "db/merge_context.h"
+#include "db/range_del_aggregator.h"
 #include "monitoring/perf_context_imp.h"

 namespace rocksdb {
@@ -30,8 +31,6 @@ Status DBImplReadOnly::Get(const ReadOptions& read_options,
                           ColumnFamilyHandle* column_family, const Slice& key,
                           PinnableSlice* pinnable_val) {
  assert(pinnable_val != nullptr);
-  // TODO: stopwatch DB_GET needed?, perf timer needed?
-  PERF_TIMER_GUARD(get_snapshot_time);
  Status s;
  SequenceNumber snapshot = versions_->LastSequence();
  auto cfh = reinterpret_cast<ColumnFamilyHandleImpl*>(column_family);
@@ -44,24 +43,16 @@ Status DBImplReadOnly::Get(const ReadOptions& read_options,
  }
  SuperVersion* super_version = cfd->GetSuperVersion();
  MergeContext merge_context;
-  SequenceNumber max_covering_tombstone_seq = 0;
+  RangeDelAggregator range_del_agg(cfd->internal_comparator(), snapshot);
  LookupKey lkey(key, snapshot);
-  PERF_TIMER_STOP(get_snapshot_time);
  if (super_version->mem->Get(lkey, pinnable_val->GetSelf(), &s, &merge_context,
-                              &max_covering_tombstone_seq, read_options)) {
+                              &range_del_agg, read_options)) {
    pinnable_val->PinSelf();
-    RecordTick(stats_, MEMTABLE_HIT);
  } else {
    PERF_TIMER_GUARD(get_from_output_files_time);
    super_version->current->Get(read_options, lkey, pinnable_val, &s,
-                                &merge_context, &max_covering_tombstone_seq);
-    RecordTick(stats_, MEMTABLE_MISS);
+                                &merge_context, &range_del_agg);
  }
-  RecordTick(stats_, NUMBER_KEYS_READ);
-  size_t size = pinnable_val->size();
-  RecordTick(stats_, BYTES_READ, size);
-  RecordInHistogram(stats_, BYTES_PER_READ, size);
-  PERF_COUNTER_ADD(get_read_bytes, size);
  return s;
 }

@@ -71,20 +62,18 @@ Iterator* DBImplReadOnly::NewIterator(const ReadOptions& read_options,
  auto cfd = cfh->cfd();
  SuperVersion* super_version = cfd->GetSuperVersion()->Ref();
  SequenceNumber latest_snapshot = versions_->LastSequence();
-  SequenceNumber read_seq =
-      read_options.snapshot != nullptr
-          ? reinterpret_cast<const SnapshotImpl*>(read_options.snapshot)
-                ->number_
-          : latest_snapshot;
  ReadCallback* read_callback = nullptr;  // No read callback provided.
  auto db_iter = NewArenaWrappedDbIterator(
      env_, read_options, *cfd->ioptions(), super_version->mutable_cf_options,
-      read_seq,
+      (read_options.snapshot != nullptr
+           ? reinterpret_cast<const SnapshotImpl*>(read_options.snapshot)
+                 ->number_
+           : latest_snapshot),
      super_version->mutable_cf_options.max_sequential_skip_in_iterations,
      super_version->version_number, read_callback);
  auto internal_iter =
      NewInternalIterator(read_options, cfd, super_version, db_iter->GetArena(),
-                          db_iter->GetRangeDelAggregator(), read_seq);
+                          db_iter->GetRangeDelAggregator());
  db_iter->SetIterUnderDBIter(internal_iter);
  return db_iter;
 }
@@ -100,22 +89,21 @@ Status DBImplReadOnly::NewIterators(
  iterators->clear();
  iterators->reserve(column_families.size());
  SequenceNumber latest_snapshot = versions_->LastSequence();
-  SequenceNumber read_seq =
-      read_options.snapshot != nullptr
-          ? reinterpret_cast<const SnapshotImpl*>(read_options.snapshot)
-                ->number_
-          : latest_snapshot;

  for (auto cfh : column_families) {
    auto* cfd = reinterpret_cast<ColumnFamilyHandleImpl*>(cfh)->cfd();
    auto* sv = cfd->GetSuperVersion()->Ref();
    auto* db_iter = NewArenaWrappedDbIterator(
-        env_, read_options, *cfd->ioptions(), sv->mutable_cf_options, read_seq,
+        env_, read_options, *cfd->ioptions(), sv->mutable_cf_options,
+        (read_options.snapshot != nullptr
+             ? reinterpret_cast<const SnapshotImpl*>(read_options.snapshot)
+                   ->number_
+             : latest_snapshot),
        sv->mutable_cf_options.max_sequential_skip_in_iterations,
        sv->version_number, read_callback);
    auto* internal_iter =
        NewInternalIterator(read_options, cfd, sv, db_iter->GetArena(),
-                            db_iter->GetRangeDelAggregator(), read_seq);
+                            db_iter->GetRangeDelAggregator());
    db_iter->SetIterUnderDBIter(internal_iter);
    iterators->push_back(db_iter);
  }
@@ -7,9 +7,9 @@

 #ifndef ROCKSDB_LITE

-#include <string>
-#include <vector>
 #include "db/db_impl.h"
+#include <vector>
+#include <string>

 namespace rocksdb {

@@ -77,8 +77,8 @@ class DBImplReadOnly : public DBImpl {
      ColumnFamilyHandle* /*column_family*/,
      const std::vector<std::string>& /*input_file_names*/,
      const int /*output_level*/, const int /*output_path_id*/ = -1,
-      std::vector<std::string>* const /*output_file_names*/ = nullptr,
-      CompactionJobInfo* /*compaction_job_info*/ = nullptr) override {
+      std::vector<std::string>* const /*output_file_names*/ = nullptr
+      ) override {
    return Status::NotSupported("Not supported operation in read only mode.");
  }

@@ -89,11 +89,10 @@ class DBImplReadOnly : public DBImpl {
  virtual Status EnableFileDeletions(bool /*force*/) override {
    return Status::NotSupported("Not supported operation in read only mode.");
  }
-  virtual Status GetLiveFiles(std::vector<std::string>& ret,
-                              uint64_t* manifest_file_size,
-                              bool /*flush_memtable*/) override {
-    return DBImpl::GetLiveFiles(ret, manifest_file_size,
-                                false /* flush_memtable */);
+  virtual Status GetLiveFiles(std::vector<std::string>&,
+                              uint64_t* /*manifest_file_size*/,
+                              bool /*flush_memtable*/ = true) override {
+    return Status::NotSupported("Not supported operation in read only mode.");
  }

  using DBImpl::Flush;
@@ -122,6 +121,6 @@ class DBImplReadOnly : public DBImpl {
  DBImplReadOnly(const DBImplReadOnly&);
  void operator=(const DBImplReadOnly&);
 };
-}  // namespace rocksdb
+}

 #endif  // !ROCKSDB_LITE
@@ -1,524 +0,0 @@
-//  Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
-//  This source code is licensed under both the GPLv2 (found in the
-//  COPYING file in the root directory) and Apache 2.0 License
-//  (found in the LICENSE.Apache file in the root directory).
-
-#include "db/db_impl_secondary.h"
-
-#ifndef __STDC_FORMAT_MACROS
-#define __STDC_FORMAT_MACROS
-#endif
-#include <inttypes.h>
-
-#include "db/db_iter.h"
-#include "db/merge_context.h"
-#include "monitoring/perf_context_imp.h"
-#include "util/auto_roll_logger.h"
-
-namespace rocksdb {
-
-#ifndef ROCKSDB_LITE
-
-DBImplSecondary::DBImplSecondary(const DBOptions& db_options,
-                                 const std::string& dbname)
-    : DBImpl(db_options, dbname) {
-  ROCKS_LOG_INFO(immutable_db_options_.info_log,
-                 "Opening the db in secondary mode");
-  LogFlush(immutable_db_options_.info_log);
-}
-
-DBImplSecondary::~DBImplSecondary() {}
-
-Status DBImplSecondary::Recover(
-    const std::vector<ColumnFamilyDescriptor>& column_families,
-    bool /*readonly*/, bool /*error_if_log_file_exist*/,
-    bool /*error_if_data_exists_in_logs*/) {
-  mutex_.AssertHeld();
-
-  Status s;
-  s = static_cast<ReactiveVersionSet*>(versions_.get())
-          ->Recover(column_families, &manifest_reader_, &manifest_reporter_,
-                    &manifest_reader_status_);
-  if (!s.ok()) {
-    return s;
-  }
-  if (immutable_db_options_.paranoid_checks && s.ok()) {
-    s = CheckConsistency();
-  }
-  // Initial max_total_in_memory_state_ before recovery logs.
-  max_total_in_memory_state_ = 0;
-  for (auto cfd : *versions_->GetColumnFamilySet()) {
-    auto* mutable_cf_options = cfd->GetLatestMutableCFOptions();
-    max_total_in_memory_state_ += mutable_cf_options->write_buffer_size *
-                                  mutable_cf_options->max_write_buffer_number;
-  }
-  if (s.ok()) {
-    default_cf_handle_ = new ColumnFamilyHandleImpl(
-        versions_->GetColumnFamilySet()->GetDefault(), this, &mutex_);
-    default_cf_internal_stats_ = default_cf_handle_->cfd()->internal_stats();
-    single_column_family_mode_ =
-        versions_->GetColumnFamilySet()->NumberOfColumnFamilies() == 1;
-
-    // Recover from all newer log files than the ones named in the
-    // descriptor.
-    std::vector<std::string> filenames;
-    s = env_->GetChildren(immutable_db_options_.wal_dir, &filenames);
-    if (s.IsNotFound()) {
-      return Status::InvalidArgument("Failed to open wal_dir",
-                                     immutable_db_options_.wal_dir);
-    } else if (!s.ok()) {
-      return s;
-    }
-
-    std::vector<uint64_t> logs;
-    // if log_readers_ is non-empty, it means we have applied all logs with log
-    // numbers smaller than the smallest log in log_readers_, so there is no
-    // need to pass these logs to RecoverLogFiles
-    uint64_t log_number_min = 0;
-    if (log_readers_.size() > 0) {
-      log_number_min = log_readers_.begin()->first;
-    }
-    for (size_t i = 0; i < filenames.size(); i++) {
-      uint64_t number;
-      FileType type;
-      if (ParseFileName(filenames[i], &number, &type) && type == kLogFile &&
-          number >= log_number_min) {
-        logs.push_back(number);
-      }
-    }
-
-    if (!logs.empty()) {
-      // Recover in the order in which the logs were generated
-      std::sort(logs.begin(), logs.end());
-      SequenceNumber next_sequence(kMaxSequenceNumber);
-      s = RecoverLogFiles(logs, &next_sequence, true /*read_only*/);
-    }
-  }
-
-  // TODO: update options_file_number_ needed?
-
-  return s;
-}
-
-// try to find log reader using log_number from log_readers_ map, initialize
-// if it doesn't exist
-Status DBImplSecondary::MaybeInitLogReader(
-    uint64_t log_number, log::FragmentBufferedReader** log_reader) {
-  auto iter = log_readers_.find(log_number);
-  // make sure the log file is still present
-  if (iter == log_readers_.end() ||
-      iter->second->reader_->GetLogNumber() != log_number) {
-    // delete the obsolete log reader if log number mismatch
-    if (iter != log_readers_.end()) {
-      log_readers_.erase(iter);
-    }
-    // initialize log reader from log_number
-    // TODO: min_log_number_to_keep_2pc check needed?
-    // Open the log file
-    std::string fname = LogFileName(immutable_db_options_.wal_dir, log_number);
-    ROCKS_LOG_INFO(immutable_db_options_.info_log,
-                   "Recovering log #%" PRIu64 " mode %d", log_number,
-                   static_cast<int>(immutable_db_options_.wal_recovery_mode));
-
-    std::unique_ptr<SequentialFileReader> file_reader;
-    {
-      std::unique_ptr<SequentialFile> file;
-      Status status = env_->NewSequentialFile(
-          fname, &file, env_->OptimizeForLogRead(env_options_));
-      if (!status.ok()) {
-        *log_reader = nullptr;
-        return status;
-      }
-      file_reader.reset(new SequentialFileReader(std::move(file), fname));
-    }
-
-    // Create the log reader.
-    LogReaderContainer* log_reader_container = new LogReaderContainer(
-        env_, immutable_db_options_.info_log, std::move(fname),
-        std::move(file_reader), log_number);
-    log_readers_.insert(std::make_pair(
-        log_number, std::unique_ptr<LogReaderContainer>(log_reader_container)));
-  }
-  iter = log_readers_.find(log_number);
-  assert(iter != log_readers_.end());
-  *log_reader = iter->second->reader_;
-  return Status::OK();
-}
-
-// After manifest recovery, replay WALs and refresh log_readers_ if necessary
-// REQUIRES: log_numbers are sorted in ascending order
-Status DBImplSecondary::RecoverLogFiles(
-    const std::vector<uint64_t>& log_numbers, SequenceNumber* next_sequence,
-    bool /*read_only*/) {
-  mutex_.AssertHeld();
-  Status status;
-  for (auto log_number : log_numbers) {
-    log::FragmentBufferedReader* reader = nullptr;
-    status = MaybeInitLogReader(log_number, &reader);
-    if (!status.ok()) {
-      return status;
-    }
-    assert(reader != nullptr);
-  }
-  for (auto log_number : log_numbers) {
-    auto it  = log_readers_.find(log_number);
-    assert(it != log_readers_.end());
-    log::FragmentBufferedReader* reader = it->second->reader_;
-    // Manually update the file number allocation counter in VersionSet.
-    versions_->MarkFileNumberUsed(log_number);
-
-    // Determine if we should tolerate incomplete records at the tail end of the
-    // Read all the records and add to a memtable
-    std::string scratch;
-    Slice record;
-    WriteBatch batch;
-
-    while (reader->ReadRecord(&record, &scratch,
-                              immutable_db_options_.wal_recovery_mode) &&
-           status.ok()) {
-      if (record.size() < WriteBatchInternal::kHeader) {
-        reader->GetReporter()->Corruption(
-            record.size(), Status::Corruption("log record too small"));
-        continue;
-      }
-      WriteBatchInternal::SetContents(&batch, record);
-      // do not check sequence number because user may toggle disableWAL
-      // between writes which breaks sequence number continuity guarantee
-
-      // If column family was not found, it might mean that the WAL write
-      // batch references to the column family that was dropped after the
-      // insert. We don't want to fail the whole write batch in that case --
-      // we just ignore the update.
-      // That's why we set ignore missing column families to true
-      // passing null flush_scheduler will disable memtable flushing which is
-      // needed for secondary instances
-      bool has_valid_writes = false;
-      status = WriteBatchInternal::InsertInto(
-          &batch, column_family_memtables_.get(), nullptr /* flush_scheduler */,
-          true, log_number, this, false /* concurrent_memtable_writes */,
-          next_sequence, &has_valid_writes, seq_per_batch_, batch_per_txn_);
-      if (!status.ok()) {
-        // We are treating this as a failure while reading since we read valid
-        // blocks that do not form coherent data
-        reader->GetReporter()->Corruption(record.size(), status);
-        continue;
-      }
-    }
-
-    if (!status.ok()) {
-      return status;
-    }
-
-    auto last_sequence = *next_sequence - 1;
-    if ((*next_sequence != kMaxSequenceNumber) &&
-        (versions_->LastSequence() <= last_sequence)) {
-      versions_->SetLastAllocatedSequence(last_sequence);
-      versions_->SetLastPublishedSequence(last_sequence);
-      versions_->SetLastSequence(last_sequence);
-    }
-  }
-  // remove logreaders from map after successfully recovering the WAL
-  if (log_readers_.size() > 1) {
-    auto eraseIter = log_readers_.begin();
-    std::advance(eraseIter, log_readers_.size() - 1);
-    log_readers_.erase(log_readers_.begin(), eraseIter);
-  }
-  return status;
-}
-
-// Implementation of the DB interface
-Status DBImplSecondary::Get(const ReadOptions& read_options,
-                            ColumnFamilyHandle* column_family, const Slice& key,
-                            PinnableSlice* value) {
-  return GetImpl(read_options, column_family, key, value);
-}
-
-Status DBImplSecondary::GetImpl(const ReadOptions& read_options,
-                                ColumnFamilyHandle* column_family,
-                                const Slice& key, PinnableSlice* pinnable_val) {
-  assert(pinnable_val != nullptr);
-  PERF_CPU_TIMER_GUARD(get_cpu_nanos, env_);
-  StopWatch sw(env_, stats_, DB_GET);
-  PERF_TIMER_GUARD(get_snapshot_time);
-
-  auto cfh = static_cast<ColumnFamilyHandleImpl*>(column_family);
-  ColumnFamilyData* cfd = cfh->cfd();
-  if (tracer_) {
-    InstrumentedMutexLock lock(&trace_mutex_);
-    if (tracer_) {
-      tracer_->Get(column_family, key);
-    }
-  }
-  // Acquire SuperVersion
-  SuperVersion* super_version = GetAndRefSuperVersion(cfd);
-  SequenceNumber snapshot = versions_->LastSequence();
-  MergeContext merge_context;
-  SequenceNumber max_covering_tombstone_seq = 0;
-  Status s;
-  LookupKey lkey(key, snapshot);
-  PERF_TIMER_STOP(get_snapshot_time);
-
-  bool done = false;
-  if (super_version->mem->Get(lkey, pinnable_val->GetSelf(), &s, &merge_context,
-                              &max_covering_tombstone_seq, read_options)) {
-    done = true;
-    pinnable_val->PinSelf();
-    RecordTick(stats_, MEMTABLE_HIT);
-  } else if ((s.ok() || s.IsMergeInProgress()) &&
-             super_version->imm->Get(
-                 lkey, pinnable_val->GetSelf(), &s, &merge_context,
-                 &max_covering_tombstone_seq, read_options)) {
-    done = true;
-    pinnable_val->PinSelf();
-    RecordTick(stats_, MEMTABLE_HIT);
-  }
-  if (!done && !s.ok() && !s.IsMergeInProgress()) {
-    ReturnAndCleanupSuperVersion(cfd, super_version);
-    return s;
-  }
-  if (!done) {
-    PERF_TIMER_GUARD(get_from_output_files_time);
-    super_version->current->Get(read_options, lkey, pinnable_val, &s,
-                                &merge_context, &max_covering_tombstone_seq);
-    RecordTick(stats_, MEMTABLE_MISS);
-  }
-  {
-    PERF_TIMER_GUARD(get_post_process_time);
-    ReturnAndCleanupSuperVersion(cfd, super_version);
-    RecordTick(stats_, NUMBER_KEYS_READ);
-    size_t size = pinnable_val->size();
-    RecordTick(stats_, BYTES_READ, size);
-    RecordTimeToHistogram(stats_, BYTES_PER_READ, size);
-    PERF_COUNTER_ADD(get_read_bytes, size);
-  }
-  return s;
-}
-
-Iterator* DBImplSecondary::NewIterator(const ReadOptions& read_options,
-                                       ColumnFamilyHandle* column_family) {
-  if (read_options.managed) {
-    return NewErrorIterator(
-        Status::NotSupported("Managed iterator is not supported anymore."));
-  }
-  if (read_options.read_tier == kPersistedTier) {
-    return NewErrorIterator(Status::NotSupported(
-        "ReadTier::kPersistedData is not yet supported in iterators."));
-  }
-  Iterator* result = nullptr;
-  auto cfh = reinterpret_cast<ColumnFamilyHandleImpl*>(column_family);
-  auto cfd = cfh->cfd();
-  ReadCallback* read_callback = nullptr;  // No read callback provided.
-  if (read_options.tailing) {
-    return NewErrorIterator(Status::NotSupported(
-        "tailing iterator not supported in secondary mode"));
-  } else if (read_options.snapshot != nullptr) {
-    // TODO (yanqin) support snapshot.
-    return NewErrorIterator(
-        Status::NotSupported("snapshot not supported in secondary mode"));
-  } else {
-    auto snapshot = versions_->LastSequence();
-    result = NewIteratorImpl(read_options, cfd, snapshot, read_callback);
-  }
-  return result;
-}
-
-ArenaWrappedDBIter* DBImplSecondary::NewIteratorImpl(
-    const ReadOptions& read_options, ColumnFamilyData* cfd,
-    SequenceNumber snapshot, ReadCallback* read_callback) {
-  assert(nullptr != cfd);
-  SuperVersion* super_version = cfd->GetReferencedSuperVersion(&mutex_);
-  auto db_iter = NewArenaWrappedDbIterator(
-      env_, read_options, *cfd->ioptions(), super_version->mutable_cf_options,
-      snapshot,
-      super_version->mutable_cf_options.max_sequential_skip_in_iterations,
-      super_version->version_number, read_callback);
-  auto internal_iter =
-      NewInternalIterator(read_options, cfd, super_version, db_iter->GetArena(),
-                          db_iter->GetRangeDelAggregator(), snapshot);
-  db_iter->SetIterUnderDBIter(internal_iter);
-  return db_iter;
-}
-
-Status DBImplSecondary::NewIterators(
-    const ReadOptions& read_options,
-    const std::vector<ColumnFamilyHandle*>& column_families,
-    std::vector<Iterator*>* iterators) {
-  if (read_options.managed) {
-    return Status::NotSupported("Managed iterator is not supported anymore.");
-  }
-  if (read_options.read_tier == kPersistedTier) {
-    return Status::NotSupported(
-        "ReadTier::kPersistedData is not yet supported in iterators.");
-  }
-  ReadCallback* read_callback = nullptr;  // No read callback provided.
-  if (iterators == nullptr) {
-    return Status::InvalidArgument("iterators not allowed to be nullptr");
-  }
-  iterators->clear();
-  iterators->reserve(column_families.size());
-  if (read_options.tailing) {
-    return Status::NotSupported(
-        "tailing iterator not supported in secondary mode");
-  } else if (read_options.snapshot != nullptr) {
-    // TODO (yanqin) support snapshot.
-    return Status::NotSupported("snapshot not supported in secondary mode");
-  } else {
-    SequenceNumber read_seq = versions_->LastSequence();
-    for (auto cfh : column_families) {
-      ColumnFamilyData* cfd = static_cast<ColumnFamilyHandleImpl*>(cfh)->cfd();
-      iterators->push_back(
-          NewIteratorImpl(read_options, cfd, read_seq, read_callback));
-    }
-  }
-  return Status::OK();
-}
-
-Status DBImplSecondary::TryCatchUpWithPrimary() {
-  assert(versions_.get() != nullptr);
-  assert(manifest_reader_.get() != nullptr);
-  Status s;
-  std::unordered_set<ColumnFamilyData*> cfds_changed;
-  InstrumentedMutexLock lock_guard(&mutex_);
-  s = static_cast<ReactiveVersionSet*>(versions_.get())
-          ->ReadAndApply(&mutex_, &manifest_reader_, &cfds_changed);
-  if (s.ok()) {
-    SuperVersionContext sv_context(true /* create_superversion */);
-    for (auto cfd : cfds_changed) {
-      sv_context.NewSuperVersion();
-      cfd->InstallSuperVersion(&sv_context, &mutex_);
-    }
-    sv_context.Clean();
-  }
-  return s;
-}
-
-Status DB::OpenAsSecondary(const Options& options, const std::string& dbname,
-                           const std::string& secondary_path, DB** dbptr) {
-  *dbptr = nullptr;
-
-  DBOptions db_options(options);
-  ColumnFamilyOptions cf_options(options);
-  std::vector<ColumnFamilyDescriptor> column_families;
-  column_families.emplace_back(kDefaultColumnFamilyName, cf_options);
-  std::vector<ColumnFamilyHandle*> handles;
-
-  Status s = DB::OpenAsSecondary(db_options, dbname, secondary_path,
-                                 column_families, &handles, dbptr);
-  if (s.ok()) {
-    assert(handles.size() == 1);
-    delete handles[0];
-  }
-  return s;
-}
-
-Status DB::OpenAsSecondary(
-    const DBOptions& db_options, const std::string& dbname,
-    const std::string& secondary_path,
-    const std::vector<ColumnFamilyDescriptor>& column_families,
-    std::vector<ColumnFamilyHandle*>* handles, DB** dbptr) {
-  *dbptr = nullptr;
-  if (db_options.max_open_files != -1) {
-    // TODO (yanqin) maybe support max_open_files != -1 by creating hard links
-    // on SST files so that db secondary can still have access to old SSTs
-    // while primary instance may delete original.
-    return Status::InvalidArgument("require max_open_files to be -1");
-  }
-
-  DBOptions tmp_opts(db_options);
-  if (nullptr == tmp_opts.info_log) {
-    Env* env = tmp_opts.env;
-    assert(env != nullptr);
-    std::string secondary_abs_path;
-    env->GetAbsolutePath(secondary_path, &secondary_abs_path);
-    std::string fname = InfoLogFileName(secondary_path, secondary_abs_path,
-                                        tmp_opts.db_log_dir);
-
-    env->CreateDirIfMissing(secondary_path);
-    if (tmp_opts.log_file_time_to_roll > 0 || tmp_opts.max_log_file_size > 0) {
-      AutoRollLogger* result = new AutoRollLogger(
-          env, secondary_path, tmp_opts.db_log_dir, tmp_opts.max_log_file_size,
-          tmp_opts.log_file_time_to_roll, tmp_opts.info_log_level);
-      Status s = result->GetStatus();
-      if (!s.ok()) {
-        delete result;
-      } else {
-        tmp_opts.info_log.reset(result);
-      }
-    }
-    if (nullptr == tmp_opts.info_log) {
-      env->RenameFile(
-          fname, OldInfoLogFileName(secondary_path, env->NowMicros(),
-                                    secondary_abs_path, tmp_opts.db_log_dir));
-      Status s = env->NewLogger(fname, &(tmp_opts.info_log));
-      if (tmp_opts.info_log != nullptr) {
-        tmp_opts.info_log->SetInfoLogLevel(tmp_opts.info_log_level);
-      }
-    }
-  }
-
-  assert(tmp_opts.info_log != nullptr);
-
-  handles->clear();
-  DBImplSecondary* impl = new DBImplSecondary(tmp_opts, dbname);
-  impl->versions_.reset(new ReactiveVersionSet(
-      dbname, &impl->immutable_db_options_, impl->env_options_,
-      impl->table_cache_.get(), impl->write_buffer_manager_,
-      &impl->write_controller_));
-  impl->column_family_memtables_.reset(
-      new ColumnFamilyMemTablesImpl(impl->versions_->GetColumnFamilySet()));
-  impl->mutex_.Lock();
-  Status s = impl->Recover(column_families, true, false, false);
-  if (s.ok()) {
-    for (auto cf : column_families) {
-      auto cfd =
-          impl->versions_->GetColumnFamilySet()->GetColumnFamily(cf.name);
-      if (nullptr == cfd) {
-        s = Status::InvalidArgument("Column family not found: ", cf.name);
-        break;
-      }
-      handles->push_back(new ColumnFamilyHandleImpl(cfd, impl, &impl->mutex_));
-    }
-  }
-  SuperVersionContext sv_context(true /* create_superversion */);
-  if (s.ok()) {
-    for (auto cfd : *impl->versions_->GetColumnFamilySet()) {
-      sv_context.NewSuperVersion();
-      cfd->InstallSuperVersion(&sv_context, &impl->mutex_);
-    }
-  }
-  impl->mutex_.Unlock();
-  sv_context.Clean();
-  if (s.ok()) {
-    *dbptr = impl;
-    for (auto h : *handles) {
-      impl->NewThreadStatusCfInfo(
-          reinterpret_cast<ColumnFamilyHandleImpl*>(h)->cfd());
-    }
-  } else {
-    for (auto h : *handles) {
-      delete h;
-    }
-    handles->clear();
-    delete impl;
-  }
-  return s;
-}
-#else   // !ROCKSDB_LITE
-
-Status DB::OpenAsSecondary(const Options& /*options*/,
-                           const std::string& /*name*/,
-                           const std::string& /*secondary_path*/,
-                           DB** /*dbptr*/) {
-  return Status::NotSupported("Not supported in ROCKSDB_LITE.");
-}
-
-Status DB::OpenAsSecondary(
-    const DBOptions& /*db_options*/, const std::string& /*dbname*/,
-    const std::string& /*secondary_path*/,
-    const std::vector<ColumnFamilyDescriptor>& /*column_families*/,
-    std::vector<ColumnFamilyHandle*>* /*handles*/, DB** /*dbptr*/) {
-  return Status::NotSupported("Not supported in ROCKSDB_LITE.");
-}
-#endif  // !ROCKSDB_LITE
-
-}  // namespace rocksdb
@@ -1,212 +0,0 @@
-//  Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
-//  This source code is licensed under both the GPLv2 (found in the
-//  COPYING file in the root directory) and Apache 2.0 License
-//  (found in the LICENSE.Apache file in the root directory).
-
-#pragma once
-
-#ifndef ROCKSDB_LITE
-
-#include <string>
-#include <vector>
-#include "db/db_impl.h"
-
-namespace rocksdb {
-
-class LogReaderContainer {
- public:
-  LogReaderContainer()
-      : reader_(nullptr), reporter_(nullptr), status_(nullptr) {}
-  LogReaderContainer(Env* env, std::shared_ptr<Logger> info_log,
-                     std::string fname,
-                     std::unique_ptr<SequentialFileReader>&& file_reader,
-                     uint64_t log_number) {
-    LogReporter* reporter = new LogReporter();
-    status_ = new Status();
-    reporter->env = env;
-    reporter->info_log = info_log.get();
-    reporter->fname = std::move(fname);
-    reporter->status = status_;
-    reporter_ = reporter;
-    // We intentially 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).
-    reader_ = new log::FragmentBufferedReader(info_log, std::move(file_reader),
-                                              reporter, true /*checksum*/,
-                                              log_number);
-  }
-  log::FragmentBufferedReader* reader_;
-  log::Reader::Reporter* reporter_;
-  Status* status_;
-  ~LogReaderContainer() {
-    delete reader_;
-    delete reporter_;
-    delete status_;
-  }
- private:
-  struct LogReporter : public log::Reader::Reporter {
-    Env* env;
-    Logger* info_log;
-    std::string fname;
-    Status* status;  // nullptr if immutable_db_options_.paranoid_checks==false
-    void Corruption(size_t bytes, const Status& s) override {
-      ROCKS_LOG_WARN(info_log, "%s%s: dropping %d bytes; %s",
-                     (this->status == nullptr ? "(ignoring error) " : ""),
-                     fname.c_str(), static_cast<int>(bytes),
-                     s.ToString().c_str());
-      if (this->status != nullptr && this->status->ok()) {
-        *this->status = s;
-      }
-    }
-  };
-};
-
-class DBImplSecondary : public DBImpl {
- public:
-  DBImplSecondary(const DBOptions& options, const std::string& dbname);
-  ~DBImplSecondary() override;
-
-  Status Recover(const std::vector<ColumnFamilyDescriptor>& column_families,
-                 bool read_only, bool error_if_log_file_exist,
-                 bool error_if_data_exists_in_logs) override;
-
-  // Implementations of the DB interface
-  using DB::Get;
-  Status Get(const ReadOptions& options, ColumnFamilyHandle* column_family,
-             const Slice& key, PinnableSlice* value) override;
-
-  Status GetImpl(const ReadOptions& options, ColumnFamilyHandle* column_family,
-                 const Slice& key, PinnableSlice* value);
-
-  using DBImpl::NewIterator;
-  Iterator* NewIterator(const ReadOptions&,
-                        ColumnFamilyHandle* column_family) override;
-
-  ArenaWrappedDBIter* NewIteratorImpl(const ReadOptions& read_options,
-                                      ColumnFamilyData* cfd,
-                                      SequenceNumber snapshot,
-                                      ReadCallback* read_callback);
-
-  Status NewIterators(const ReadOptions& options,
-                      const std::vector<ColumnFamilyHandle*>& column_families,
-                      std::vector<Iterator*>* iterators) override;
-
-  using DBImpl::Put;
-  Status Put(const WriteOptions& /*options*/,
-             ColumnFamilyHandle* /*column_family*/, const Slice& /*key*/,
-             const Slice& /*value*/) override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  using DBImpl::Merge;
-  Status Merge(const WriteOptions& /*options*/,
-               ColumnFamilyHandle* /*column_family*/, const Slice& /*key*/,
-               const Slice& /*value*/) override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  using DBImpl::Delete;
-  Status Delete(const WriteOptions& /*options*/,
-                ColumnFamilyHandle* /*column_family*/,
-                const Slice& /*key*/) override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  using DBImpl::SingleDelete;
-  Status SingleDelete(const WriteOptions& /*options*/,
-                      ColumnFamilyHandle* /*column_family*/,
-                      const Slice& /*key*/) override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  Status Write(const WriteOptions& /*options*/,
-               WriteBatch* /*updates*/) override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  using DBImpl::CompactRange;
-  Status CompactRange(const CompactRangeOptions& /*options*/,
-                      ColumnFamilyHandle* /*column_family*/,
-                      const Slice* /*begin*/, const Slice* /*end*/) override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  using DBImpl::CompactFiles;
-  Status CompactFiles(
-      const CompactionOptions& /*compact_options*/,
-      ColumnFamilyHandle* /*column_family*/,
-      const std::vector<std::string>& /*input_file_names*/,
-      const int /*output_level*/, const int /*output_path_id*/ = -1,
-      std::vector<std::string>* const /*output_file_names*/ = nullptr,
-      CompactionJobInfo* /*compaction_job_info*/ = nullptr) override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  Status DisableFileDeletions() override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  Status EnableFileDeletions(bool /*force*/) override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  Status GetLiveFiles(std::vector<std::string>&,
-                      uint64_t* /*manifest_file_size*/,
-                      bool /*flush_memtable*/ = true) override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  using DBImpl::Flush;
-  Status Flush(const FlushOptions& /*options*/,
-               ColumnFamilyHandle* /*column_family*/) override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  using DBImpl::SyncWAL;
-  Status SyncWAL() override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  using DB::IngestExternalFile;
-  Status IngestExternalFile(
-      ColumnFamilyHandle* /*column_family*/,
-      const std::vector<std::string>& /*external_files*/,
-      const IngestExternalFileOptions& /*ingestion_options*/) override {
-    return Status::NotSupported("Not supported operation in read only mode.");
-  }
-
-  // Try to catch up with the primary by reading as much as possible from the
-  // log files until there is nothing more to read or encounters an error. If
-  // the amount of information in the log files to process is huge, this
-  // method can take long time due to all the I/O and CPU costs.
-  Status TryCatchUpWithPrimary() override;
-
-  Status MaybeInitLogReader(uint64_t log_number,
-                            log::FragmentBufferedReader** log_reader);
-
- private:
-  friend class DB;
-
-  // No copying allowed
-  DBImplSecondary(const DBImplSecondary&);
-  void operator=(const DBImplSecondary&);
-
-  using DBImpl::Recover;
-
-  Status RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
-                         SequenceNumber* next_sequence,
-                         bool read_only) override;
-
-  std::unique_ptr<log::FragmentBufferedReader> manifest_reader_;
-  std::unique_ptr<log::Reader::Reporter> manifest_reporter_;
-  std::unique_ptr<Status> manifest_reader_status_;
-
-  // cache log readers for each log number, used for continue WAL replay
-  // after recovery
-  std::map<uint64_t, std::unique_ptr<LogReaderContainer>> log_readers_;
-};
-
-}  // namespace rocksdb
-
-#endif  // !ROCKSDB_LITE
@@ -146,6 +146,17 @@ Status DBImpl::WriteImpl(const WriteOptions& write_options,

    if (write_thread_.CompleteParallelMemTableWriter(&w)) {
      // we're responsible for exit batch group
+      for (auto* writer : *(w.write_group)) {
+        if (!writer->CallbackFailed() && writer->pre_release_callback) {
+          assert(writer->sequence != kMaxSequenceNumber);
+          Status ws = writer->pre_release_callback->Callback(writer->sequence,
+                                                             disable_memtable);
+          if (!ws.ok()) {
+            status = ws;
+            break;
+          }
+        }
+      }
      // TODO(myabandeh): propagate status to write_group
      auto last_sequence = w.write_group->last_sequence;
      versions_->SetLastSequence(last_sequence);
@@ -268,7 +279,7 @@ Status DBImpl::WriteImpl(const WriteOptions& write_options,
                        concurrent_update);
      RecordTick(stats_, WRITE_DONE_BY_OTHER, write_done_by_other);
    }
-    RecordInHistogram(stats_, BYTES_PER_WRITE, total_byte_size);
+    MeasureTime(stats_, BYTES_PER_WRITE, total_byte_size);

    if (write_options.disableWAL) {
      has_unpersisted_data_.store(true, std::memory_order_relaxed);
@@ -298,35 +309,6 @@ Status DBImpl::WriteImpl(const WriteOptions& write_options,
    const SequenceNumber current_sequence = last_sequence + 1;
    last_sequence += seq_inc;

-    // PreReleaseCallback is called after WAL write and before memtable write
-    if (status.ok()) {
-      SequenceNumber next_sequence = current_sequence;
-      // Note: the logic for advancing seq here must be consistent with the
-      // logic in WriteBatchInternal::InsertInto(write_group...) as well as
-      // with WriteBatchInternal::InsertInto(write_batch...) that is called on
-      // the merged batch during recovery from the WAL.
-      for (auto* writer : write_group) {
-        if (writer->CallbackFailed()) {
-          continue;
-        }
-        writer->sequence = next_sequence;
-        if (writer->pre_release_callback) {
-          Status ws = writer->pre_release_callback->Callback(
-              writer->sequence, disable_memtable, writer->log_used);
-          if (!ws.ok()) {
-            status = ws;
-            break;
-          }
-        }
-        if (seq_per_batch_) {
-          assert(writer->batch_cnt);
-          next_sequence += writer->batch_cnt;
-        } else if (writer->ShouldWriteToMemtable()) {
-          next_sequence += WriteBatchInternal::Count(writer->batch);
-        }
-      }
-    }
-
    if (status.ok()) {
      PERF_TIMER_GUARD(write_memtable_time);

@@ -338,7 +320,26 @@ Status DBImpl::WriteImpl(const WriteOptions& write_options,
            0 /*recovery_log_number*/, this, parallel, seq_per_batch_,
            batch_per_txn_);
      } else {
+        SequenceNumber next_sequence = current_sequence;
+        // Note: the logic for advancing seq here must be consistent with the
+        // logic in WriteBatchInternal::InsertInto(write_group...) as well as
+        // with WriteBatchInternal::InsertInto(write_batch...) that is called on
+        // the merged batch during recovery from the WAL.
+        for (auto* writer : write_group) {
+          if (writer->CallbackFailed()) {
+            continue;
+          }
+          writer->sequence = next_sequence;
+          if (seq_per_batch_) {
+            assert(writer->batch_cnt);
+            next_sequence += writer->batch_cnt;
+          } else if (writer->ShouldWriteToMemtable()) {
+            next_sequence += WriteBatchInternal::Count(writer->batch);
+          }
+        }
        write_group.last_sequence = last_sequence;
+        write_group.running.store(static_cast<uint32_t>(write_group.size),
+                                  std::memory_order_relaxed);
        write_thread_.LaunchParallelMemTableWriters(&write_group);
        in_parallel_group = true;

@@ -389,8 +390,17 @@ Status DBImpl::WriteImpl(const WriteOptions& write_options,
  }
  if (should_exit_batch_group) {
    if (status.ok()) {
-      // Note: if we are to resume after non-OK statuses we need to revisit how
-      // we reacts to non-OK statuses here.
+      for (auto* writer : write_group) {
+        if (!writer->CallbackFailed() && writer->pre_release_callback) {
+          assert(writer->sequence != kMaxSequenceNumber);
+          Status ws = writer->pre_release_callback->Callback(writer->sequence,
+                                                             disable_memtable);
+          if (!ws.ok()) {
+            status = ws;
+            break;
+          }
+        }
+      }
      versions_->SetLastSequence(last_sequence);
    }
    MemTableInsertStatusCheck(w.status);
@@ -463,11 +473,11 @@ Status DBImpl::PipelinedWriteImpl(const WriteOptions& write_options,
    RecordTick(stats_, NUMBER_KEYS_WRITTEN, total_count);
    stats->AddDBStats(InternalStats::BYTES_WRITTEN, total_byte_size);
    RecordTick(stats_, BYTES_WRITTEN, total_byte_size);
-    RecordInHistogram(stats_, BYTES_PER_WRITE, total_byte_size);
+    MeasureTime(stats_, BYTES_PER_WRITE, total_byte_size);

    PERF_TIMER_STOP(write_pre_and_post_process_time);

-    if (w.status.ok() && !write_options.disableWAL) {
+    if (w.ShouldWriteToWAL()) {
      PERF_TIMER_GUARD(write_wal_time);
      stats->AddDBStats(InternalStats::WRITE_DONE_BY_SELF, 1);
      RecordTick(stats_, WRITE_DONE_BY_SELF, 1);
@@ -496,7 +506,7 @@ Status DBImpl::PipelinedWriteImpl(const WriteOptions& write_options,
  WriteThread::WriteGroup memtable_write_group;
  if (w.state == WriteThread::STATE_MEMTABLE_WRITER_LEADER) {
    PERF_TIMER_GUARD(write_memtable_time);
-    assert(w.ShouldWriteToMemtable());
+    assert(w.status.ok());
    write_thread_.EnterAsMemTableWriter(&w, &memtable_write_group);
    if (memtable_write_group.size > 1 &&
        immutable_db_options_.allow_concurrent_memtable_write) {
@@ -594,7 +604,7 @@ Status DBImpl::WriteImplWALOnly(const WriteOptions& write_options,
                      concurrent_update);
    RecordTick(stats_, WRITE_DONE_BY_OTHER, write_done_by_other);
  }
-  RecordInHistogram(stats_, BYTES_PER_WRITE, total_byte_size);
+  MeasureTime(stats_, BYTES_PER_WRITE, total_byte_size);

  PERF_TIMER_STOP(write_pre_and_post_process_time);

@@ -649,8 +659,8 @@ Status DBImpl::WriteImplWALOnly(const WriteOptions& write_options,
      if (!writer->CallbackFailed() && writer->pre_release_callback) {
        assert(writer->sequence != kMaxSequenceNumber);
        const bool DISABLE_MEMTABLE = true;
-        Status ws = writer->pre_release_callback->Callback(
-            writer->sequence, DISABLE_MEMTABLE, writer->log_used);
+        Status ws = writer->pre_release_callback->Callback(writer->sequence,
+                                                           DISABLE_MEMTABLE);
        if (!ws.ok()) {
          status = ws;
          break;
@@ -700,10 +710,6 @@ Status DBImpl::PreprocessWrite(const WriteOptions& write_options,
  assert(write_context != nullptr && need_log_sync != nullptr);
  Status status;

-  if (error_handler_.IsDBStopped()) {
-    status = error_handler_.GetBGError();
-  }
-
  PERF_TIMER_GUARD(write_scheduling_flushes_compactions_time);

  assert(!single_column_family_mode_ ||
@@ -722,6 +728,10 @@ Status DBImpl::PreprocessWrite(const WriteOptions& write_options,
    status = HandleWriteBufferFull(write_context);
  }

+  if (UNLIKELY(status.ok())) {
+    status = error_handler_.GetBGError();
+  }
+
  if (UNLIKELY(status.ok() && !flush_scheduler_.Empty())) {
    status = ScheduleFlushes(write_context);
  }
@@ -868,7 +878,7 @@ Status DBImpl::WriteToWAL(const WriteThread::WriteGroup& write_group,
  status = WriteToWAL(*merged_batch, log_writer, log_used, &log_size);
  if (to_be_cached_state) {
    cached_recoverable_state_ = *to_be_cached_state;
-    cached_recoverable_state_empty_ = false;
+      cached_recoverable_state_empty_ = false;
  }

  if (status.ok() && need_log_sync) {
@@ -944,7 +954,7 @@ Status DBImpl::ConcurrentWriteToWAL(const WriteThread::WriteGroup& write_group,
  status = WriteToWAL(*merged_batch, log_writer, log_used, &log_size);
  if (to_be_cached_state) {
    cached_recoverable_state_ = *to_be_cached_state;
-    cached_recoverable_state_empty_ = false;
+      cached_recoverable_state_empty_ = false;
  }
  log_write_mutex_.Unlock();

@@ -993,9 +1003,8 @@ Status DBImpl::WriteRecoverableState() {
      const bool DISABLE_MEMTABLE = true;
      for (uint64_t sub_batch_seq = seq + 1;
           sub_batch_seq < next_seq && status.ok(); sub_batch_seq++) {
-        uint64_t const no_log_num = 0;
        status = recoverable_state_pre_release_callback_->Callback(
-            sub_batch_seq, !DISABLE_MEMTABLE, no_log_num);
+            sub_batch_seq, !DISABLE_MEMTABLE);
      }
    }
    if (status.ok()) {
@@ -1007,28 +1016,6 @@ Status DBImpl::WriteRecoverableState() {
  return Status::OK();
 }

-void DBImpl::SelectColumnFamiliesForAtomicFlush(
-    autovector<ColumnFamilyData*>* cfds) {
-  for (ColumnFamilyData* cfd : *versions_->GetColumnFamilySet()) {
-    if (cfd->IsDropped()) {
-      continue;
-    }
-    if (cfd->imm()->NumNotFlushed() != 0 || !cfd->mem()->IsEmpty() ||
-        !cached_recoverable_state_empty_.load()) {
-      cfds->push_back(cfd);
-    }
-  }
-}
-
-// Assign sequence number for atomic flush.
-void DBImpl::AssignAtomicFlushSeq(const autovector<ColumnFamilyData*>& cfds) {
-  assert(immutable_db_options_.atomic_flush);
-  auto seq = versions_->LastSequence();
-  for (auto cfd : cfds) {
-    cfd->imm()->AssignAtomicFlushSeq(seq);
-  }
-}
-
 Status DBImpl::SwitchWAL(WriteContext* write_context) {
  mutex_.AssertHeld();
  assert(write_context != nullptr);
@@ -1041,22 +1028,22 @@ Status DBImpl::SwitchWAL(WriteContext* write_context) {
  auto oldest_alive_log = alive_log_files_.begin()->number;
  bool flush_wont_release_oldest_log = false;
  if (allow_2pc()) {
-    auto oldest_log_with_uncommitted_prep =
+    auto oldest_log_with_uncommited_prep =
        logs_with_prep_tracker_.FindMinLogContainingOutstandingPrep();

-    assert(oldest_log_with_uncommitted_prep == 0 ||
-           oldest_log_with_uncommitted_prep >= oldest_alive_log);
-    if (oldest_log_with_uncommitted_prep > 0 &&
-        oldest_log_with_uncommitted_prep == oldest_alive_log) {
+    assert(oldest_log_with_uncommited_prep == 0 ||
+           oldest_log_with_uncommited_prep >= oldest_alive_log);
+    if (oldest_log_with_uncommited_prep > 0 &&
+        oldest_log_with_uncommited_prep == oldest_alive_log) {
      if (unable_to_release_oldest_log_) {
        // we already attempted to flush all column families dependent on
-        // the oldest alive log but the log still contained uncommitted
+        // the oldest alive log but the log still contained uncommited
        // transactions so there is still nothing that we can do.
        return status;
      } else {
        ROCKS_LOG_WARN(
            immutable_db_options_.info_log,
-            "Unable to release oldest log due to uncommitted transaction");
+            "Unable to release oldest log due to uncommited transaction");
        unable_to_release_oldest_log_ = true;
        flush_wont_release_oldest_log = true;
      }
@@ -1065,52 +1052,32 @@ Status DBImpl::SwitchWAL(WriteContext* write_context) {
  if (!flush_wont_release_oldest_log) {
    // we only mark this log as getting flushed if we have successfully
    // flushed all data in this log. If this log contains outstanding prepared
-    // transactions then we cannot flush this log until those transactions are
-    // commited.
+    // transactions then we cannot flush this log until those transactions are commited.
    unable_to_release_oldest_log_ = false;
    alive_log_files_.begin()->getting_flushed = true;
  }

-  ROCKS_LOG_INFO(
-      immutable_db_options_.info_log,
-      "Flushing all column families with data in WAL number %" PRIu64
-      ". Total log size is %" PRIu64 " while max_total_wal_size is %" PRIu64,
-      oldest_alive_log, total_log_size_.load(), GetMaxTotalWalSize());
+  ROCKS_LOG_INFO(immutable_db_options_.info_log,
+                 "Flushing all column families with data in WAL number %" PRIu64
+                 ". Total log size is %" PRIu64
+                 " while max_total_wal_size is %" PRIu64,
+                 oldest_alive_log, total_log_size_.load(), GetMaxTotalWalSize());
  // no need to refcount because drop is happening in write thread, so can't
  // happen while we're in the write thread
-  autovector<ColumnFamilyData*> cfds;
-  if (immutable_db_options_.atomic_flush) {
-    SelectColumnFamiliesForAtomicFlush(&cfds);
-  } else {
-    for (auto cfd : *versions_->GetColumnFamilySet()) {
-      if (cfd->IsDropped()) {
-        continue;
+  for (auto cfd : *versions_->GetColumnFamilySet()) {
+    if (cfd->IsDropped()) {
+      continue;
+    }
+    if (cfd->OldestLogToKeep() <= oldest_alive_log) {
+      status = SwitchMemtable(cfd, write_context);
+      if (!status.ok()) {
+        break;
      }
-      if (cfd->OldestLogToKeep() <= oldest_alive_log) {
-        cfds.push_back(cfd);
-      }
-    }
-  }
-  for (const auto cfd : cfds) {
-    cfd->Ref();
-    status = SwitchMemtable(cfd, write_context);
-    cfd->Unref();
-    if (!status.ok()) {
-      break;
-    }
-  }
-  if (status.ok()) {
-    if (immutable_db_options_.atomic_flush) {
-      AssignAtomicFlushSeq(cfds);
-    }
-    for (auto cfd : cfds) {
      cfd->imm()->FlushRequested();
+      SchedulePendingFlush(cfd, FlushReason::kWriteBufferManager);
    }
-    FlushRequest flush_req;
-    GenerateFlushRequest(cfds, &flush_req);
-    SchedulePendingFlush(flush_req, FlushReason::kWriteBufferManager);
-    MaybeScheduleFlushOrCompaction();
  }
+  MaybeScheduleFlushOrCompaction();
  return status;
 }

@@ -1127,59 +1094,36 @@ Status DBImpl::HandleWriteBufferFull(WriteContext* write_context) {
  ROCKS_LOG_INFO(
      immutable_db_options_.info_log,
      "Flushing column family with largest mem table size. Write buffer is "
-      "using %" ROCKSDB_PRIszt " bytes out of a total of %" ROCKSDB_PRIszt ".",
+      "using %" PRIu64 " bytes out of a total of %" PRIu64 ".",
      write_buffer_manager_->memory_usage(),
      write_buffer_manager_->buffer_size());
  // no need to refcount because drop is happening in write thread, so can't
  // happen while we're in the write thread
-  autovector<ColumnFamilyData*> cfds;
-  if (immutable_db_options_.atomic_flush) {
-    SelectColumnFamiliesForAtomicFlush(&cfds);
-  } else {
-    ColumnFamilyData* cfd_picked = nullptr;
-    SequenceNumber seq_num_for_cf_picked = kMaxSequenceNumber;
+  ColumnFamilyData* cfd_picked = nullptr;
+  SequenceNumber seq_num_for_cf_picked = kMaxSequenceNumber;

-    for (auto cfd : *versions_->GetColumnFamilySet()) {
-      if (cfd->IsDropped()) {
-        continue;
-      }
-      if (!cfd->mem()->IsEmpty()) {
-        // We only consider active mem table, hoping immutable memtable is
-        // already in the process of flushing.
-        uint64_t seq = cfd->mem()->GetCreationSeq();
-        if (cfd_picked == nullptr || seq < seq_num_for_cf_picked) {
-          cfd_picked = cfd;
-          seq_num_for_cf_picked = seq;
-        }
-      }
-    }
-    if (cfd_picked != nullptr) {
-      cfds.push_back(cfd_picked);
-    }
-  }
-
-  for (const auto cfd : cfds) {
-    if (cfd->mem()->IsEmpty()) {
+  for (auto cfd : *versions_->GetColumnFamilySet()) {
+    if (cfd->IsDropped()) {
      continue;
    }
-    cfd->Ref();
-    status = SwitchMemtable(cfd, write_context);
-    cfd->Unref();
-    if (!status.ok()) {
-      break;
+    if (!cfd->mem()->IsEmpty()) {
+      // We only consider active mem table, hoping immutable memtable is
+      // already in the process of flushing.
+      uint64_t seq = cfd->mem()->GetCreationSeq();
+      if (cfd_picked == nullptr || seq < seq_num_for_cf_picked) {
+        cfd_picked = cfd;
+        seq_num_for_cf_picked = seq;
+      }
    }
  }
-  if (status.ok()) {
-    if (immutable_db_options_.atomic_flush) {
-      AssignAtomicFlushSeq(cfds);
+  if (cfd_picked != nullptr) {
+    status = SwitchMemtable(cfd_picked, write_context,
+                            FlushReason::kWriteBufferFull);
+    if (status.ok()) {
+      cfd_picked->imm()->FlushRequested();
+      SchedulePendingFlush(cfd_picked, FlushReason::kWriteBufferFull);
+      MaybeScheduleFlushOrCompaction();
    }
-    for (const auto cfd : cfds) {
-      cfd->imm()->FlushRequested();
-    }
-    FlushRequest flush_req;
-    GenerateFlushRequest(cfds, &flush_req);
-    SchedulePendingFlush(flush_req, FlushReason::kWriteBufferFull);
-    MaybeScheduleFlushOrCompaction();
  }
  return status;
 }
@@ -1229,11 +1173,7 @@ Status DBImpl::DelayWrite(uint64_t num_bytes,
      write_thread_.EndWriteStall();
    }

-    // Don't wait if there's a background error, even if its a soft error. We
-    // might wait here indefinitely as the background compaction may never
-    // finish successfully, resulting in the stall condition lasting
-    // indefinitely
-    while (error_handler_.GetBGError().ok() && write_controller_.IsStopped()) {
+    while (!error_handler_.IsDBStopped() && write_controller_.IsStopped()) {
      if (write_options.no_slowdown) {
        return Status::Incomplete("Write stall");
      }
@@ -1254,19 +1194,7 @@ Status DBImpl::DelayWrite(uint64_t num_bytes,
    RecordTick(stats_, STALL_MICROS, time_delayed);
  }

-  // If DB is not in read-only mode and write_controller is not stopping
-  // writes, we can ignore any background errors and allow the write to
-  // proceed
-  Status s;
-  if (write_controller_.IsStopped()) {
-    // If writes are still stopped, it means we bailed due to a background
-    // error
-    s = Status::Incomplete(error_handler_.GetBGError().ToString());
-  }
-  if (error_handler_.IsDBStopped()) {
-    s = error_handler_.GetBGError();
-  }
-  return s;
+  return error_handler_.GetBGError();
 }

 Status DBImpl::ThrottleLowPriWritesIfNeeded(const WriteOptions& write_options,
@@ -1300,42 +1228,17 @@ Status DBImpl::ThrottleLowPriWritesIfNeeded(const WriteOptions& write_options,
 }

 Status DBImpl::ScheduleFlushes(WriteContext* context) {
-  autovector<ColumnFamilyData*> cfds;
-  if (immutable_db_options_.atomic_flush) {
-    SelectColumnFamiliesForAtomicFlush(&cfds);
-    for (auto cfd : cfds) {
-      cfd->Ref();
-    }
-    flush_scheduler_.Clear();
-  } else {
-    ColumnFamilyData* tmp_cfd;
-    while ((tmp_cfd = flush_scheduler_.TakeNextColumnFamily()) != nullptr) {
-      cfds.push_back(tmp_cfd);
-    }
-  }
-  Status status;
-  for (auto& cfd : cfds) {
-    if (!cfd->mem()->IsEmpty()) {
-      status = SwitchMemtable(cfd, context);
-    }
+  ColumnFamilyData* cfd;
+  while ((cfd = flush_scheduler_.TakeNextColumnFamily()) != nullptr) {
+    auto status = SwitchMemtable(cfd, context, FlushReason::kWriteBufferFull);
    if (cfd->Unref()) {
      delete cfd;
-      cfd = nullptr;
    }
    if (!status.ok()) {
-      break;
+      return status;
    }
  }
-  if (status.ok()) {
-    if (immutable_db_options_.atomic_flush) {
-      AssignAtomicFlushSeq(cfds);
-    }
-    FlushRequest flush_req;
-    GenerateFlushRequest(cfds, &flush_req);
-    SchedulePendingFlush(flush_req, FlushReason::kWriteBufferFull);
-    MaybeScheduleFlushOrCompaction();
-  }
-  return status;
+  return Status::OK();
 }

 #ifndef ROCKSDB_LITE
@@ -1356,7 +1259,8 @@ void DBImpl::NotifyOnMemTableSealed(ColumnFamilyData* /*cfd*/,

 // REQUIRES: mutex_ is held
 // REQUIRES: this thread is currently at the front of the writer queue
-Status DBImpl::SwitchMemtable(ColumnFamilyData* cfd, WriteContext* context) {
+Status DBImpl::SwitchMemtable(ColumnFamilyData* cfd, WriteContext* context,
+                              FlushReason flush_reason) {
  mutex_.AssertHeld();
  WriteThread::Writer nonmem_w;
  if (two_write_queues_) {
@@ -1365,7 +1269,7 @@ Status DBImpl::SwitchMemtable(ColumnFamilyData* cfd, WriteContext* context) {
    nonmem_write_thread_.EnterUnbatched(&nonmem_w, &mutex_);
  }

-  std::unique_ptr<WritableFile> lfile;
+  unique_ptr<WritableFile> lfile;
  log::Writer* new_log = nullptr;
  MemTable* new_mem = nullptr;

@@ -1418,19 +1322,57 @@ Status DBImpl::SwitchMemtable(ColumnFamilyData* cfd, WriteContext* context) {
  // Log this later after lock release. It may be outdated, e.g., if background
  // flush happens before logging, but that should be ok.
  int num_imm_unflushed = cfd->imm()->NumNotFlushed();
+  DBOptions db_options =
+      BuildDBOptions(immutable_db_options_, mutable_db_options_);
  const auto preallocate_block_size =
-      GetWalPreallocateBlockSize(mutable_cf_options.write_buffer_size);
+    GetWalPreallocateBlockSize(mutable_cf_options.write_buffer_size);
+  auto write_hint = CalculateWALWriteHint();
  mutex_.Unlock();
-  if (creating_new_log) {
-    // TODO: Write buffer size passed in should be max of all CF's instead
-    // of mutable_cf_options.write_buffer_size.
-    s = CreateWAL(new_log_number, recycle_log_number, preallocate_block_size,
-                  &new_log);
-  }
-  if (s.ok()) {
-    SequenceNumber seq = versions_->LastSequence();
-    new_mem = cfd->ConstructNewMemtable(mutable_cf_options, seq);
-    context->superversion_context.NewSuperVersion();
+  {
+    if (creating_new_log) {
+      EnvOptions opt_env_opt =
+          env_->OptimizeForLogWrite(env_options_, db_options);
+      if (recycle_log_number) {
+        ROCKS_LOG_INFO(immutable_db_options_.info_log,
+                       "reusing log %" PRIu64 " from recycle list\n",
+                       recycle_log_number);
+        s = env_->ReuseWritableFile(
+            LogFileName(immutable_db_options_.wal_dir, new_log_number),
+            LogFileName(immutable_db_options_.wal_dir, recycle_log_number),
+            &lfile, opt_env_opt);
+      } else {
+        s = NewWritableFile(
+            env_, LogFileName(immutable_db_options_.wal_dir, new_log_number),
+            &lfile, opt_env_opt);
+      }
+      if (s.ok()) {
+        // Our final size should be less than write_buffer_size
+        // (compression, etc) but err on the side of caution.
+
+        // use preallocate_block_size instead
+        // of calling GetWalPreallocateBlockSize()
+        lfile->SetPreallocationBlockSize(preallocate_block_size);
+        lfile->SetWriteLifeTimeHint(write_hint);
+        unique_ptr<WritableFileWriter> file_writer(
+            new WritableFileWriter(std::move(lfile), opt_env_opt));
+        new_log = new log::Writer(
+            std::move(file_writer), new_log_number,
+            immutable_db_options_.recycle_log_file_num > 0, manual_wal_flush_);
+      }
+    }
+
+    if (s.ok()) {
+      SequenceNumber seq = versions_->LastSequence();
+      new_mem = cfd->ConstructNewMemtable(mutable_cf_options, seq);
+      context->superversion_context.NewSuperVersion();
+    }
+
+#ifndef ROCKSDB_LITE
+    // PLEASE NOTE: We assume that there are no failable operations
+    // after lock is acquired below since we are already notifying
+    // client about mem table becoming immutable.
+    NotifyOnMemTableSealed(cfd, memtable_info);
+#endif //ROCKSDB_LITE
  }
  ROCKS_LOG_INFO(immutable_db_options_.info_log,
                 "[%s] New memtable created with log file: #%" PRIu64
@@ -1439,7 +1381,10 @@ Status DBImpl::SwitchMemtable(ColumnFamilyData* cfd, WriteContext* context) {
  mutex_.Lock();
  if (s.ok() && creating_new_log) {
    log_write_mutex_.Lock();
+    logfile_number_ = new_log_number;
    assert(new_log != nullptr);
+    log_empty_ = true;
+    log_dir_synced_ = false;
    if (!logs_.empty()) {
      // Alway flush the buffer of the last log before switching to a new one
      log::Writer* cur_log_writer = logs_.back().writer;
@@ -1447,41 +1392,21 @@ Status DBImpl::SwitchMemtable(ColumnFamilyData* cfd, WriteContext* context) {
      if (!s.ok()) {
        ROCKS_LOG_WARN(immutable_db_options_.info_log,
                       "[%s] Failed to switch from #%" PRIu64 " to #%" PRIu64
-                       "  WAL file\n",
+                       "  WAL file -- %s\n",
                       cfd->GetName().c_str(), cur_log_writer->get_log_number(),
                       new_log_number);
      }
    }
-    if (s.ok()) {
-      logfile_number_ = new_log_number;
-      log_empty_ = true;
-      log_dir_synced_ = false;
-      logs_.emplace_back(logfile_number_, new_log);
-      alive_log_files_.push_back(LogFileNumberSize(logfile_number_));
-    }
+    logs_.emplace_back(logfile_number_, new_log);
+    alive_log_files_.push_back(LogFileNumberSize(logfile_number_));
    log_write_mutex_.Unlock();
  }

  if (!s.ok()) {
    // how do we fail if we're not creating new log?
    assert(creating_new_log);
-    if (new_mem) {
-      delete new_mem;
-    }
-    if (new_log) {
-      delete new_log;
-    }
-    SuperVersion* new_superversion =
-        context->superversion_context.new_superversion.release();
-    if (new_superversion != nullptr) {
-      delete new_superversion;
-    }
-    // We may have lost data from the WritableFileBuffer in-memory buffer for
-    // the current log, so treat it as a fatal error and set bg_error
-    error_handler_.SetBGError(s, BackgroundErrorReason::kMemTable);
-    // Read back bg_error in order to get the right severity
-    s = error_handler_.GetBGError();
-
+    assert(!new_mem);
+    assert(!new_log);
    if (two_write_queues_) {
      nonmem_write_thread_.ExitUnbatched(&nonmem_w);
    }
@@ -1507,14 +1432,7 @@ Status DBImpl::SwitchMemtable(ColumnFamilyData* cfd, WriteContext* context) {
  new_mem->Ref();
  cfd->SetMemtable(new_mem);
  InstallSuperVersionAndScheduleWork(cfd, &context->superversion_context,
-                                     mutable_cf_options);
-#ifndef ROCKSDB_LITE
-  mutex_.Unlock();
-  // Notify client that memtable is sealed, now that we have successfully
-  // installed a new memtable
-  NotifyOnMemTableSealed(cfd, memtable_info);
-  mutex_.Lock();
-#endif  // ROCKSDB_LITE
+                                     mutable_cf_options, flush_reason);
  if (two_write_queues_) {
    nonmem_write_thread_.ExitUnbatched(&nonmem_w);
  }
@@ -1523,13 +1441,13 @@ Status DBImpl::SwitchMemtable(ColumnFamilyData* cfd, WriteContext* context) {

 size_t DBImpl::GetWalPreallocateBlockSize(uint64_t write_buffer_size) const {
  mutex_.AssertHeld();
-  size_t bsize =
-      static_cast<size_t>(write_buffer_size / 10 + write_buffer_size);
+  size_t bsize = static_cast<size_t>(
+    write_buffer_size / 10 + write_buffer_size);
  // Some users might set very high write_buffer_size and rely on
  // max_total_wal_size or other parameters to control the WAL size.
  if (mutable_db_options_.max_total_wal_size > 0) {
-    bsize = std::min<size_t>(
-        bsize, static_cast<size_t>(mutable_db_options_.max_total_wal_size));
+    bsize = std::min<size_t>(bsize, static_cast<size_t>(
+      mutable_db_options_.max_total_wal_size));
  }
  if (immutable_db_options_.db_write_buffer_size > 0) {
    bsize = std::min<size_t>(bsize, immutable_db_options_.db_write_buffer_size);
@@ -42,7 +42,7 @@ void DumpDBFileSummary(const ImmutableDBOptions& options,
          "Error when reading %s dir\n", dbname.c_str());
  }
  std::sort(files.begin(), files.end());
-  for (const std::string& file : files) {
+  for (std::string file : files) {
    if (!ParseFileName(file, &number, &type)) {
      continue;
    }
@@ -85,7 +85,7 @@ void DumpDBFileSummary(const ImmutableDBOptions& options,
        continue;
      }
      std::sort(files.begin(), files.end());
-      for (const std::string& file : files) {
+      for (std::string file : files) {
        if (ParseFileName(file, &number, &type)) {
          if (type == kTableFile && ++file_num < 10) {
            file_info.append(file).append(" ");
@@ -109,7 +109,7 @@ void DumpDBFileSummary(const ImmutableDBOptions& options,
      return;
    }
    wal_info.clear();
-    for (const std::string& file : files) {
+    for (std::string file : files) {
      if (ParseFileName(file, &number, &type)) {
        if (type == kLogFile) {
          env->GetFileSize(options.wal_dir + "/" + file, &file_size);
@@ -88,6 +88,7 @@ TEST_F(DBIOFailureTest, DropWritesFlush) {
    env_->drop_writes_.store(false, std::memory_order_release);
  } while (ChangeCompactOptions());
 }
+#endif  // ROCKSDB_LITE

 // Check that CompactRange() returns failure if there is not enough space left
 // on device
@@ -115,7 +116,6 @@ TEST_F(DBIOFailureTest, NoSpaceCompactRange) {
    env_->no_space_.store(false, std::memory_order_release);
  } while (ChangeCompactOptions());
 }
-#endif  // ROCKSDB_LITE

 TEST_F(DBIOFailureTest, NonWritableFileSystem) {
  do {
@@ -22,14 +22,12 @@
 #include "rocksdb/merge_operator.h"
 #include "rocksdb/options.h"
 #include "table/internal_iterator.h"
-#include "table/iterator_wrapper.h"
 #include "util/arena.h"
 #include "util/filename.h"
 #include "util/logging.h"
 #include "util/mutexlock.h"
 #include "util/string_util.h"
 #include "util/trace_replay.h"
-#include "util/user_comparator_wrapper.h"

 namespace rocksdb {

@@ -119,70 +117,78 @@ class DBIter final: public Iterator {
         uint64_t max_sequential_skip_in_iterations,
         ReadCallback* read_callback, DBImpl* db_impl, ColumnFamilyData* cfd,
         bool allow_blob)
-      : env_(_env),
+      : arena_mode_(arena_mode),
+        env_(_env),
        logger_(cf_options.info_log),
        user_comparator_(cmp),
        merge_operator_(cf_options.merge_operator),
        iter_(iter),
-        read_callback_(read_callback),
        sequence_(s),
+        direction_(kForward),
+        valid_(false),
+        current_entry_is_merged_(false),
        statistics_(cf_options.statistics),
        num_internal_keys_skipped_(0),
        iterate_lower_bound_(read_options.iterate_lower_bound),
        iterate_upper_bound_(read_options.iterate_upper_bound),
-        direction_(kForward),
-        valid_(false),
-        current_entry_is_merged_(false),
        prefix_same_as_start_(read_options.prefix_same_as_start),
        pin_thru_lifetime_(read_options.pin_data),
        total_order_seek_(read_options.total_order_seek),
-        allow_blob_(allow_blob),
-        is_blob_(false),
-        arena_mode_(arena_mode),
-        range_del_agg_(&cf_options.internal_comparator, s),
+        range_del_agg_(cf_options.internal_comparator, s,
+                       true /* collapse_deletions */),
+        read_callback_(read_callback),
        db_impl_(db_impl),
        cfd_(cfd),
+        allow_blob_(allow_blob),
+        is_blob_(false),
        start_seqnum_(read_options.iter_start_seqnum) {
-    RecordTick(statistics_, NO_ITERATOR_CREATED);
+    RecordTick(statistics_, NO_ITERATORS);
    prefix_extractor_ = mutable_cf_options.prefix_extractor.get();
    max_skip_ = max_sequential_skip_in_iterations;
    max_skippable_internal_keys_ = read_options.max_skippable_internal_keys;
    if (pin_thru_lifetime_) {
      pinned_iters_mgr_.StartPinning();
    }
-    if (iter_.iter()) {
-      iter_.iter()->SetPinnedItersMgr(&pinned_iters_mgr_);
+    if (iter_) {
+      iter_->SetPinnedItersMgr(&pinned_iters_mgr_);
    }
  }
-  ~DBIter() override {
+  virtual ~DBIter() {
    // Release pinned data if any
    if (pinned_iters_mgr_.PinningEnabled()) {
      pinned_iters_mgr_.ReleasePinnedData();
    }
-    RecordTick(statistics_, NO_ITERATOR_DELETED);
+    // Compiler warning issue filed:
+    // https://github.com/facebook/rocksdb/issues/3013
+    RecordTick(statistics_, NO_ITERATORS, uint64_t(-1));
    ResetInternalKeysSkippedCounter();
    local_stats_.BumpGlobalStatistics(statistics_);
-    iter_.DeleteIter(arena_mode_);
+    if (!arena_mode_) {
+      delete iter_;
+    } else {
+      iter_->~InternalIterator();
+    }
  }
  virtual void SetIter(InternalIterator* iter) {
-    assert(iter_.iter() == nullptr);
-    iter_.Set(iter);
-    iter_.iter()->SetPinnedItersMgr(&pinned_iters_mgr_);
+    assert(iter_ == nullptr);
+    iter_ = iter;
+    iter_->SetPinnedItersMgr(&pinned_iters_mgr_);
  }
-  virtual ReadRangeDelAggregator* GetRangeDelAggregator() {
+  virtual RangeDelAggregator* GetRangeDelAggregator() {
    return &range_del_agg_;
  }

-  bool Valid() const override { return valid_; }
-  Slice key() const override {
+  virtual bool Valid() const override { return valid_; }
+  virtual Slice key() const override {
    assert(valid_);
    if(start_seqnum_ > 0) {
      return saved_key_.GetInternalKey();
    } else {
      return saved_key_.GetUserKey();
    }
+
  }
-  Slice value() const override {
+  virtual Slice value() const override {
    assert(valid_);
    if (current_entry_is_merged_) {
      // If pinned_value_ is set then the result of merge operator is one of
@@ -191,12 +197,12 @@ class DBIter final: public Iterator {
    } else if (direction_ == kReverse) {
      return pinned_value_;
    } else {
-      return iter_.value();
+      return iter_->value();
    }
  }
-  Status status() const override {
+  virtual Status status() const override {
    if (status_.ok()) {
-      return iter_.status();
+      return iter_->status();
    } else {
      assert(!valid_);
      return status_;
@@ -207,13 +213,14 @@ class DBIter final: public Iterator {
    return is_blob_;
  }

-  Status GetProperty(std::string prop_name, std::string* prop) override {
+  virtual Status GetProperty(std::string prop_name,
+                             std::string* prop) override {
    if (prop == nullptr) {
      return Status::InvalidArgument("prop is nullptr");
    }
    if (prop_name == "rocksdb.iterator.super-version-number") {
      // First try to pass the value returned from inner iterator.
-      return iter_.iter()->GetProperty(prop_name, prop);
+      return iter_->GetProperty(prop_name, prop);
    } else if (prop_name == "rocksdb.iterator.is-key-pinned") {
      if (valid_) {
        *prop = (pin_thru_lifetime_ && saved_key_.IsKeyPinned()) ? "1" : "0";
@@ -225,22 +232,17 @@ class DBIter final: public Iterator {
      *prop = saved_key_.GetUserKey().ToString();
      return Status::OK();
    }
-    return Status::InvalidArgument("Unidentified property.");
+    return Status::InvalidArgument("Undentified property.");
  }

-  inline void Next() final override;
-  inline void Prev() final override;
-  inline void Seek(const Slice& target) final override;
-  inline void SeekForPrev(const Slice& target) final override;
-  inline void SeekToFirst() final override;
-  inline void SeekToLast() final override;
+  virtual void Next() override;
+  virtual void Prev() override;
+  virtual void Seek(const Slice& target) override;
+  virtual void SeekForPrev(const Slice& target) override;
+  virtual void SeekToFirst() override;
+  virtual void SeekToLast() override;
  Env* env() { return env_; }
-  void set_sequence(uint64_t s) {
-    sequence_ = s;
-    if (read_callback_) {
-      read_callback_->Refresh(s);
-    }
-  }
+  void set_sequence(uint64_t s) { sequence_ = s; }
  void set_valid(bool v) { valid_ = v; }

 private:
@@ -254,13 +256,13 @@ class DBIter final: public Iterator {
  bool FindValueForCurrentKeyUsingSeek();
  bool FindUserKeyBeforeSavedKey();
  inline bool FindNextUserEntry(bool skipping, bool prefix_check);
-  inline bool FindNextUserEntryInternal(bool skipping, bool prefix_check);
+  bool FindNextUserEntryInternal(bool skipping, bool prefix_check);
  bool ParseKey(ParsedInternalKey* key);
  bool MergeValuesNewToOld();

  void PrevInternal();
  bool TooManyInternalKeysSkipped(bool increment = true);
-  inline bool IsVisible(SequenceNumber sequence);
+  bool IsVisible(SequenceNumber sequence);

  // CanReseekToSkip() returns whether the iterator can use the optimization
  // where it reseek by sequence number to get the next key when there are too
@@ -268,6 +270,12 @@ class DBIter final: public Iterator {
  // sequence number does not guarantee that it is visible.
  inline bool CanReseekToSkip();

+  // MaxVisibleSequenceNumber() returns the maximum visible sequence number
+  // for this snapshot. This sequence number may be greater than snapshot
+  // seqno because uncommitted data written to DB for write unprepared will
+  // have a higher sequence number.
+  inline SequenceNumber MaxVisibleSequenceNumber();
+
  // Temporarily pin the blocks that we encounter until ReleaseTempPinnedData()
  // is called
  void TempPinData() {
@@ -301,16 +309,15 @@ class DBIter final: public Iterator {
  }

  const SliceTransform* prefix_extractor_;
+  bool arena_mode_;
  Env* const env_;
  Logger* logger_;
-  UserComparatorWrapper user_comparator_;
+  const Comparator* const user_comparator_;
  const MergeOperator* const merge_operator_;
-  IteratorWrapper iter_;
-  ReadCallback* read_callback_;
-  // Max visible sequence number. It is normally the snapshot seq unless we have
-  // uncommitted data in db as in WriteUnCommitted.
+  InternalIterator* iter_;
  SequenceNumber sequence_;

+  Status status_;
  IterKey saved_key_;
  // Reusable internal key data structure. This is only used inside one function
  // and should not be used across functions. Reusing this object can reduce
@@ -318,6 +325,9 @@ class DBIter final: public Iterator {
  ParsedInternalKey ikey_;
  std::string saved_value_;
  Slice pinned_value_;
+  Direction direction_;
+  bool valid_;
+  bool current_entry_is_merged_;
  // for prefix seek mode to support prev()
  Statistics* statistics_;
  uint64_t max_skip_;
@@ -325,29 +335,23 @@ class DBIter final: public Iterator {
  uint64_t num_internal_keys_skipped_;
  const Slice* iterate_lower_bound_;
  const Slice* iterate_upper_bound_;
-
  IterKey prefix_start_buf_;
-
-  Status status_;
  Slice prefix_start_key_;
-  Direction direction_;
-  bool valid_;
-  bool current_entry_is_merged_;
  const bool prefix_same_as_start_;
  // Means that we will pin all data blocks we read as long the Iterator
  // is not deleted, will be true if ReadOptions::pin_data is true
  const bool pin_thru_lifetime_;
  const bool total_order_seek_;
-  bool allow_blob_;
-  bool is_blob_;
-  bool arena_mode_;
  // List of operands for merge operator.
  MergeContext merge_context_;
-  ReadRangeDelAggregator range_del_agg_;
+  RangeDelAggregator range_del_agg_;
  LocalStatistics local_stats_;
  PinnedIteratorsManager pinned_iters_mgr_;
+  ReadCallback* read_callback_;
  DBImpl* db_impl_;
  ColumnFamilyData* cfd_;
+  bool allow_blob_;
+  bool is_blob_;
  // for diff snapshots we want the lower bound on the seqnum;
  // if this value > 0 iterator will return internal keys
  SequenceNumber start_seqnum_;
@@ -358,11 +362,11 @@ class DBIter final: public Iterator {
 };

 inline bool DBIter::ParseKey(ParsedInternalKey* ikey) {
-  if (!ParseInternalKey(iter_.key(), ikey)) {
+  if (!ParseInternalKey(iter_->key(), ikey)) {
    status_ = Status::Corruption("corrupted internal key in DBIter");
    valid_ = false;
    ROCKS_LOG_ERROR(logger_, "corrupted internal key in DBIter: %s",
-                    iter_.key().ToString(true).c_str());
+                    iter_->key().ToString(true).c_str());
    return false;
  } else {
    return true;
@@ -373,30 +377,28 @@ void DBIter::Next() {
  assert(valid_);
  assert(status_.ok());

-  PERF_CPU_TIMER_GUARD(iter_next_cpu_nanos, env_);
  // Release temporarily pinned blocks from last operation
  ReleaseTempPinnedData();
-  local_stats_.skip_count_ += num_internal_keys_skipped_;
-  local_stats_.skip_count_--;
-  num_internal_keys_skipped_ = 0;
+  ResetInternalKeysSkippedCounter();
  bool ok = true;
  if (direction_ == kReverse) {
    if (!ReverseToForward()) {
      ok = false;
    }
-  } else if (!current_entry_is_merged_) {
+  } else if (iter_->Valid() && !current_entry_is_merged_) {
    // If the current value is not a merge, the iter position is the
    // current key, which is already returned. We can safely issue a
    // Next() without checking the current key.
    // If the current key is a merge, very likely iter already points
    // to the next internal position.
-    assert(iter_.Valid());
-    iter_.Next();
+    iter_->Next();
    PERF_COUNTER_ADD(internal_key_skipped_count, 1);
  }

-  local_stats_.next_count_++;
-  if (ok && iter_.Valid()) {
+  if (statistics_ != nullptr) {
+    local_stats_.next_count_++;
+  }
+  if (ok && iter_->Valid()) {
    FindNextUserEntry(true /* skipping the current user key */,
                      prefix_same_as_start_);
  } else {
@@ -428,9 +430,9 @@ inline bool DBIter::FindNextUserEntry(bool skipping, bool prefix_check) {
 }

 // Actual implementation of DBIter::FindNextUserEntry()
-inline bool DBIter::FindNextUserEntryInternal(bool skipping, bool prefix_check) {
+bool DBIter::FindNextUserEntryInternal(bool skipping, bool prefix_check) {
  // Loop until we hit an acceptable entry to yield
-  assert(iter_.Valid());
+  assert(iter_->Valid());
  assert(status_.ok());
  assert(direction_ == kForward);
  current_entry_is_merged_ = false;
@@ -455,7 +457,7 @@ inline bool DBIter::FindNextUserEntryInternal(bool skipping, bool prefix_check)
    }

    if (iterate_upper_bound_ != nullptr &&
-        user_comparator_.Compare(ikey_.user_key, *iterate_upper_bound_) >= 0) {
+        user_comparator_->Compare(ikey_.user_key, *iterate_upper_bound_) >= 0) {
      break;
    }

@@ -470,8 +472,8 @@ inline bool DBIter::FindNextUserEntryInternal(bool skipping, bool prefix_check)
    }

    if (IsVisible(ikey_.sequence)) {
-      if (skipping && user_comparator_.Compare(ikey_.user_key,
-                                               saved_key_.GetUserKey()) <= 0) {
+      if (skipping && user_comparator_->Compare(ikey_.user_key,
+                                                saved_key_.GetUserKey()) <= 0) {
        num_skipped++;  // skip this entry
        PERF_COUNTER_ADD(internal_key_skipped_count, 1);
      } else {
@@ -492,8 +494,8 @@ inline bool DBIter::FindNextUserEntryInternal(bool skipping, bool prefix_check)
              return true;
            } else {
              saved_key_.SetUserKey(
-                  ikey_.user_key, !pin_thru_lifetime_ ||
-                                      !iter_.iter()->IsKeyPinned() /* copy */);
+                ikey_.user_key,
+                !pin_thru_lifetime_ || !iter_->IsKeyPinned() /* copy */);
              skipping = true;
              PERF_COUNTER_ADD(internal_delete_skipped_count, 1);
            }
@@ -513,16 +515,14 @@ inline bool DBIter::FindNextUserEntryInternal(bool skipping, bool prefix_check)
              } else {
                // this key and all previous versions shouldn't be included,
                // skipping
-                saved_key_.SetUserKey(
-                    ikey_.user_key,
-                    !pin_thru_lifetime_ ||
-                        !iter_.iter()->IsKeyPinned() /* copy */);
+                saved_key_.SetUserKey(ikey_.user_key,
+                  !pin_thru_lifetime_ || !iter_->IsKeyPinned() /* copy */);
                skipping = true;
              }
            } else {
              saved_key_.SetUserKey(
-                  ikey_.user_key, !pin_thru_lifetime_ ||
-                                      !iter_.iter()->IsKeyPinned() /* copy */);
+                  ikey_.user_key,
+                  !pin_thru_lifetime_ || !iter_->IsKeyPinned() /* copy */);
              if (range_del_agg_.ShouldDelete(
                      ikey_, RangeDelPositioningMode::kForwardTraversal)) {
                // Arrange to skip all upcoming entries for this key since
@@ -552,7 +552,7 @@ inline bool DBIter::FindNextUserEntryInternal(bool skipping, bool prefix_check)
          case kTypeMerge:
            saved_key_.SetUserKey(
                ikey_.user_key,
-                !pin_thru_lifetime_ || !iter_.iter()->IsKeyPinned() /* copy */);
+                !pin_thru_lifetime_ || !iter_->IsKeyPinned() /* copy */);
            if (range_del_agg_.ShouldDelete(
                    ikey_, RangeDelPositioningMode::kForwardTraversal)) {
              // Arrange to skip all upcoming entries for this key since
@@ -580,13 +580,13 @@ inline bool DBIter::FindNextUserEntryInternal(bool skipping, bool prefix_check)
      // If this happens too many times in a row for the same user key, we want
      // to seek to the target sequence number.
      int cmp =
-          user_comparator_.Compare(ikey_.user_key, saved_key_.GetUserKey());
+          user_comparator_->Compare(ikey_.user_key, saved_key_.GetUserKey());
      if (cmp == 0 || (skipping && cmp <= 0)) {
        num_skipped++;
      } else {
        saved_key_.SetUserKey(
            ikey_.user_key,
-            !iter_.iter()->IsKeyPinned() || !pin_thru_lifetime_ /* copy */);
+            !iter_->IsKeyPinned() || !pin_thru_lifetime_ /* copy */);
        skipping = false;
        num_skipped = 0;
      }
@@ -615,20 +615,20 @@ inline bool DBIter::FindNextUserEntryInternal(bool skipping, bool prefix_check)
                          ParsedInternalKey(saved_key_.GetUserKey(), sequence_,
                                            kValueTypeForSeek));
      }
-      iter_.Seek(last_key);
+      iter_->Seek(last_key);
      RecordTick(statistics_, NUMBER_OF_RESEEKS_IN_ITERATION);
    } else {
-      iter_.Next();
+      iter_->Next();
    }
-  } while (iter_.Valid());
+  } while (iter_->Valid());

  valid_ = false;
-  return iter_.status().ok();
+  return iter_->status().ok();
 }

 // Merge values of the same user key starting from the current iter_ position
 // Scan from the newer entries to older entries.
-// PRE: iter_.key() points to the first merge type entry
+// PRE: iter_->key() points to the first merge type entry
 //      saved_key_ stores the user key
 // POST: saved_value_ has the merged value for the user key
 //       iter_ points to the next entry (or invalid)
@@ -644,19 +644,19 @@ bool DBIter::MergeValuesNewToOld() {
  TempPinData();
  merge_context_.Clear();
  // Start the merge process by pushing the first operand
-  merge_context_.PushOperand(
-      iter_.value(), iter_.iter()->IsValuePinned() /* operand_pinned */);
+  merge_context_.PushOperand(iter_->value(),
+                             iter_->IsValuePinned() /* operand_pinned */);
  TEST_SYNC_POINT("DBIter::MergeValuesNewToOld:PushedFirstOperand");

  ParsedInternalKey ikey;
  Status s;
-  for (iter_.Next(); iter_.Valid(); iter_.Next()) {
+  for (iter_->Next(); iter_->Valid(); iter_->Next()) {
    TEST_SYNC_POINT("DBIter::MergeValuesNewToOld:SteppedToNextOperand");
    if (!ParseKey(&ikey)) {
      return false;
    }

-    if (!user_comparator_.Equal(ikey.user_key, saved_key_.GetUserKey())) {
+    if (!user_comparator_->Equal(ikey.user_key, saved_key_.GetUserKey())) {
      // hit the next user key, stop right here
      break;
    } else if (kTypeDeletion == ikey.type || kTypeSingleDeletion == ikey.type ||
@@ -664,12 +664,12 @@ bool DBIter::MergeValuesNewToOld() {
                   ikey, RangeDelPositioningMode::kForwardTraversal)) {
      // hit a delete with the same user key, stop right here
      // iter_ is positioned after delete
-      iter_.Next();
+      iter_->Next();
      break;
    } else if (kTypeValue == ikey.type) {
      // 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();
+      const Slice val = iter_->value();
      s = MergeHelper::TimedFullMerge(
          merge_operator_, ikey.user_key, &val, merge_context_.GetOperands(),
          &saved_value_, logger_, statistics_, env_, &pinned_value_, true);
@@ -679,8 +679,8 @@ bool DBIter::MergeValuesNewToOld() {
        return false;
      }
      // iter_ is positioned after put
-      iter_.Next();
-      if (!iter_.status().ok()) {
+      iter_->Next();
+      if (!iter_->status().ok()) {
        valid_ = false;
        return false;
      }
@@ -688,8 +688,8 @@ bool DBIter::MergeValuesNewToOld() {
    } else if (kTypeMerge == ikey.type) {
      // hit a merge, add the value as an operand and run associative merge.
      // when complete, add result to operands and continue.
-      merge_context_.PushOperand(
-          iter_.value(), iter_.iter()->IsValuePinned() /* operand_pinned */);
+      merge_context_.PushOperand(iter_->value(),
+                                 iter_->IsValuePinned() /* operand_pinned */);
      PERF_COUNTER_ADD(internal_merge_count, 1);
    } else if (kTypeBlobIndex == ikey.type) {
      if (!allow_blob_) {
@@ -708,7 +708,7 @@ bool DBIter::MergeValuesNewToOld() {
    }
  }

-  if (!iter_.status().ok()) {
+  if (!iter_->status().ok()) {
    valid_ = false;
    return false;
  }
@@ -734,8 +734,6 @@ bool DBIter::MergeValuesNewToOld() {
 void DBIter::Prev() {
  assert(valid_);
  assert(status_.ok());
-
-  PERF_CPU_TIMER_GUARD(iter_prev_cpu_nanos, env_);
  ReleaseTempPinnedData();
  ResetInternalKeysSkippedCounter();
  bool ok = true;
@@ -757,32 +755,33 @@ void DBIter::Prev() {
 }

 bool DBIter::ReverseToForward() {
-  assert(iter_.status().ok());
+  assert(iter_->status().ok());

  // When moving backwards, iter_ is positioned on _previous_ key, which may
  // not exist or may have different prefix than the current key().
  // If that's the case, seek iter_ to current key.
-  if ((prefix_extractor_ != nullptr && !total_order_seek_) || !iter_.Valid()) {
+  if ((prefix_extractor_ != nullptr && !total_order_seek_) || !iter_->Valid()) {
    IterKey last_key;
    last_key.SetInternalKey(ParsedInternalKey(
        saved_key_.GetUserKey(), kMaxSequenceNumber, kValueTypeForSeek));
-    iter_.Seek(last_key.GetInternalKey());
+    iter_->Seek(last_key.GetInternalKey());
  }

  direction_ = kForward;
  // Skip keys less than the current key() (a.k.a. saved_key_).
-  while (iter_.Valid()) {
+  while (iter_->Valid()) {
    ParsedInternalKey ikey;
    if (!ParseKey(&ikey)) {
      return false;
    }
-    if (user_comparator_.Compare(ikey.user_key, saved_key_.GetUserKey()) >= 0) {
+    if (user_comparator_->Compare(ikey.user_key, saved_key_.GetUserKey()) >=
+        0) {
      return true;
    }
-    iter_.Next();
+    iter_->Next();
  }

-  if (!iter_.status().ok()) {
+  if (!iter_->status().ok()) {
    valid_ = false;
    return false;
  }
@@ -792,14 +791,14 @@ bool DBIter::ReverseToForward() {

 // Move iter_ to the key before saved_key_.
 bool DBIter::ReverseToBackward() {
-  assert(iter_.status().ok());
+  assert(iter_->status().ok());

  // When current_entry_is_merged_ is true, iter_ may be positioned on the next
  // key, which may not exist or may have prefix different from current.
  // If that's the case, seek to saved_key_.
  if (current_entry_is_merged_ &&
      ((prefix_extractor_ != nullptr && !total_order_seek_) ||
-       !iter_.Valid())) {
+       !iter_->Valid())) {
    IterKey last_key;
    // Using kMaxSequenceNumber and kValueTypeForSeek
    // (not kValueTypeForSeekForPrev) to seek to a key strictly smaller
@@ -807,15 +806,15 @@ bool DBIter::ReverseToBackward() {
    last_key.SetInternalKey(ParsedInternalKey(
        saved_key_.GetUserKey(), kMaxSequenceNumber, kValueTypeForSeek));
    if (prefix_extractor_ != nullptr && !total_order_seek_) {
-      iter_.SeekForPrev(last_key.GetInternalKey());
+      iter_->SeekForPrev(last_key.GetInternalKey());
    } else {
      // Some iterators may not support SeekForPrev(), so we avoid using it
      // when prefix seek mode is disabled. This is somewhat expensive
      // (an extra Prev(), as well as an extra change of direction of iter_),
      // so we may need to reconsider it later.
-      iter_.Seek(last_key.GetInternalKey());
-      if (!iter_.Valid() && iter_.status().ok()) {
-        iter_.SeekToLast();
+      iter_->Seek(last_key.GetInternalKey());
+      if (!iter_->Valid() && iter_->status().ok()) {
+        iter_->SeekToLast();
      }
    }
  }
@@ -825,10 +824,10 @@ bool DBIter::ReverseToBackward() {
 }

 void DBIter::PrevInternal() {
-  while (iter_.Valid()) {
+  while (iter_->Valid()) {
    saved_key_.SetUserKey(
-        ExtractUserKey(iter_.key()),
-        !iter_.iter()->IsKeyPinned() || !pin_thru_lifetime_ /* copy */);
+        ExtractUserKey(iter_->key()),
+        !iter_->IsKeyPinned() || !pin_thru_lifetime_ /* copy */);

    if (prefix_extractor_ && prefix_same_as_start_ &&
        prefix_extractor_->Transform(saved_key_.GetUserKey())
@@ -839,8 +838,8 @@ void DBIter::PrevInternal() {
    }

    if (iterate_lower_bound_ != nullptr &&
-        user_comparator_.Compare(saved_key_.GetUserKey(),
-                                 *iterate_lower_bound_) < 0) {
+        user_comparator_->Compare(saved_key_.GetUserKey(),
+                                  *iterate_lower_bound_) < 0) {
      // We've iterated earlier than the user-specified lower bound.
      valid_ = false;
      return;
@@ -882,7 +881,7 @@ void DBIter::PrevInternal() {
 // POST: iter_ is positioned on one of the entries equal to saved_key_, or on
 //       the entry just before them, or on the entry just after them.
 bool DBIter::FindValueForCurrentKey() {
-  assert(iter_.Valid());
+  assert(iter_->Valid());
  merge_context_.Clear();
  current_entry_is_merged_ = false;
  // last entry before merge (could be kTypeDeletion, kTypeSingleDeletion or
@@ -894,14 +893,14 @@ bool DBIter::FindValueForCurrentKey() {
  ReleaseTempPinnedData();
  TempPinData();
  size_t num_skipped = 0;
-  while (iter_.Valid()) {
+  while (iter_->Valid()) {
    ParsedInternalKey ikey;
    if (!ParseKey(&ikey)) {
      return false;
    }

    if (!IsVisible(ikey.sequence) ||
-        !user_comparator_.Equal(ikey.user_key, saved_key_.GetUserKey())) {
+        !user_comparator_->Equal(ikey.user_key, saved_key_.GetUserKey())) {
      break;
    }
    if (TooManyInternalKeysSkipped()) {
@@ -924,8 +923,8 @@ bool DBIter::FindValueForCurrentKey() {
          last_key_entry_type = kTypeRangeDeletion;
          PERF_COUNTER_ADD(internal_delete_skipped_count, 1);
        } else {
-          assert(iter_.iter()->IsValuePinned());
-          pinned_value_ = iter_.value();
+          assert(iter_->IsValuePinned());
+          pinned_value_ = iter_->value();
        }
        merge_context_.Clear();
        last_not_merge_type = last_key_entry_type;
@@ -946,8 +945,7 @@ bool DBIter::FindValueForCurrentKey() {
        } else {
          assert(merge_operator_ != nullptr);
          merge_context_.PushOperandBack(
-              iter_.value(),
-              iter_.iter()->IsValuePinned() /* operand_pinned */);
+              iter_->value(), iter_->IsValuePinned() /* operand_pinned */);
          PERF_COUNTER_ADD(internal_merge_count, 1);
        }
        break;
@@ -956,11 +954,11 @@ bool DBIter::FindValueForCurrentKey() {
    }

    PERF_COUNTER_ADD(internal_key_skipped_count, 1);
-    iter_.Prev();
+    iter_->Prev();
    ++num_skipped;
  }

-  if (!iter_.status().ok()) {
+  if (!iter_->status().ok()) {
    valid_ = false;
    return false;
  }
@@ -1040,22 +1038,22 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
  std::string last_key;
  AppendInternalKey(&last_key, ParsedInternalKey(saved_key_.GetUserKey(),
                                                 sequence_, kValueTypeForSeek));
-  iter_.Seek(last_key);
+  iter_->Seek(last_key);
  RecordTick(statistics_, NUMBER_OF_RESEEKS_IN_ITERATION);

  // In case read_callback presents, the value we seek to may not be visible.
  // Find the next value that's visible.
  ParsedInternalKey ikey;
  while (true) {
-    if (!iter_.Valid()) {
+    if (!iter_->Valid()) {
      valid_ = false;
-      return iter_.status().ok();
+      return iter_->status().ok();
    }

    if (!ParseKey(&ikey)) {
      return false;
    }
-    if (!user_comparator_.Equal(ikey.user_key, saved_key_.GetUserKey())) {
+    if (!user_comparator_->Equal(ikey.user_key, saved_key_.GetUserKey())) {
      // No visible values for this key, even though FindValueForCurrentKey()
      // has seen some. This is possible if we're using a tailing iterator, and
      // the entries were discarded in a compaction.
@@ -1067,7 +1065,7 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
      break;
    }

-    iter_.Next();
+    iter_->Next();
  }

  if (ikey.type == kTypeDeletion || ikey.type == kTypeSingleDeletion ||
@@ -1085,8 +1083,8 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
    return false;
  }
  if (ikey.type == kTypeValue || ikey.type == kTypeBlobIndex) {
-    assert(iter_.iter()->IsValuePinned());
-    pinned_value_ = iter_.value();
+    assert(iter_->IsValuePinned());
+    pinned_value_ = iter_->value();
    valid_ = true;
    return true;
  }
@@ -1096,13 +1094,13 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
  assert(ikey.type == kTypeMerge);
  current_entry_is_merged_ = true;
  merge_context_.Clear();
-  merge_context_.PushOperand(
-      iter_.value(), iter_.iter()->IsValuePinned() /* operand_pinned */);
+  merge_context_.PushOperand(iter_->value(),
+                             iter_->IsValuePinned() /* operand_pinned */);
  while (true) {
-    iter_.Next();
+    iter_->Next();

-    if (!iter_.Valid()) {
-      if (!iter_.status().ok()) {
+    if (!iter_->Valid()) {
+      if (!iter_->status().ok()) {
        valid_ = false;
        return false;
      }
@@ -1111,16 +1109,16 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
    if (!ParseKey(&ikey)) {
      return false;
    }
-    if (!user_comparator_.Equal(ikey.user_key, saved_key_.GetUserKey())) {
+    if (!user_comparator_->Equal(ikey.user_key, saved_key_.GetUserKey())) {
      break;
    }

    if (ikey.type == kTypeDeletion || ikey.type == kTypeSingleDeletion ||
        range_del_agg_.ShouldDelete(
-            ikey, RangeDelPositioningMode::kForwardTraversal)) {
+            ikey, RangeDelPositioningMode::kBackwardTraversal)) {
      break;
    } else if (ikey.type == kTypeValue) {
-      const Slice val = iter_.value();
+      const Slice val = iter_->value();
      Status s = MergeHelper::TimedFullMerge(
          merge_operator_, saved_key_.GetUserKey(), &val,
          merge_context_.GetOperands(), &saved_value_, logger_, statistics_,
@@ -1133,8 +1131,8 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
      valid_ = true;
      return true;
    } else if (ikey.type == kTypeMerge) {
-      merge_context_.PushOperand(
-          iter_.value(), iter_.iter()->IsValuePinned() /* operand_pinned */);
+      merge_context_.PushOperand(iter_->value(),
+                                 iter_->IsValuePinned() /* operand_pinned */);
      PERF_COUNTER_ADD(internal_merge_count, 1);
    } else if (ikey.type == kTypeBlobIndex) {
      if (!allow_blob_) {
@@ -1166,13 +1164,13 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
  // Make sure we leave iter_ in a good state. If it's valid and we don't care
  // about prefixes, that's already good enough. Otherwise it needs to be
  // seeked to the current key.
-  if ((prefix_extractor_ != nullptr && !total_order_seek_) || !iter_.Valid()) {
+  if ((prefix_extractor_ != nullptr && !total_order_seek_) || !iter_->Valid()) {
    if (prefix_extractor_ != nullptr && !total_order_seek_) {
-      iter_.SeekForPrev(last_key);
+      iter_->SeekForPrev(last_key);
    } else {
-      iter_.Seek(last_key);
-      if (!iter_.Valid() && iter_.status().ok()) {
-        iter_.SeekToLast();
+      iter_->Seek(last_key);
+      if (!iter_->Valid() && iter_->status().ok()) {
+        iter_->SeekToLast();
      }
    }
    RecordTick(statistics_, NUMBER_OF_RESEEKS_IN_ITERATION);
@@ -1187,13 +1185,13 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
 bool DBIter::FindUserKeyBeforeSavedKey() {
  assert(status_.ok());
  size_t num_skipped = 0;
-  while (iter_.Valid()) {
+  while (iter_->Valid()) {
    ParsedInternalKey ikey;
    if (!ParseKey(&ikey)) {
      return false;
    }

-    if (user_comparator_.Compare(ikey.user_key, saved_key_.GetUserKey()) < 0) {
+    if (user_comparator_->Compare(ikey.user_key, saved_key_.GetUserKey()) < 0) {
      return true;
    }

@@ -1215,19 +1213,19 @@ bool DBIter::FindUserKeyBeforeSavedKey() {
          saved_key_.GetUserKey(), kMaxSequenceNumber, kValueTypeForSeek));
      // It would be more efficient to use SeekForPrev() here, but some
      // iterators may not support it.
-      iter_.Seek(last_key.GetInternalKey());
+      iter_->Seek(last_key.GetInternalKey());
      RecordTick(statistics_, NUMBER_OF_RESEEKS_IN_ITERATION);
-      if (!iter_.Valid()) {
+      if (!iter_->Valid()) {
        break;
      }
    } else {
      ++num_skipped;
    }

-    iter_.Prev();
+    iter_->Prev();
  }

-  if (!iter_.status().ok()) {
+  if (!iter_->status().ok()) {
    valid_ = false;
    return false;
  }
@@ -1248,25 +1246,30 @@ bool DBIter::TooManyInternalKeysSkipped(bool increment) {
 }

 bool DBIter::IsVisible(SequenceNumber sequence) {
-  if (read_callback_ == nullptr) {
-    return sequence <= sequence_;
-  } else {
-    return read_callback_->IsVisible(sequence);
-  }
+  return sequence <= MaxVisibleSequenceNumber() &&
+         (read_callback_ == nullptr || read_callback_->IsVisible(sequence));
 }

 bool DBIter::CanReseekToSkip() {
-  return read_callback_ == nullptr || read_callback_->CanReseekToSkip();
+  return read_callback_ == nullptr ||
+         read_callback_->MaxUnpreparedSequenceNumber() == 0;
+}
+
+SequenceNumber DBIter::MaxVisibleSequenceNumber() {
+  if (read_callback_ == nullptr) {
+    return sequence_;
+  }
+
+  return std::max(sequence_, read_callback_->MaxUnpreparedSequenceNumber());
 }

 void DBIter::Seek(const Slice& target) {
-  PERF_CPU_TIMER_GUARD(iter_seek_cpu_nanos, env_);
  StopWatch sw(env_, statistics_, DB_SEEK);
  status_ = Status::OK();
  ReleaseTempPinnedData();
  ResetInternalKeysSkippedCounter();

-  SequenceNumber seq = sequence_;
+  SequenceNumber seq = MaxVisibleSequenceNumber();
  saved_key_.Clear();
  saved_key_.SetInternalKey(target, seq);

@@ -1277,19 +1280,19 @@ void DBIter::Seek(const Slice& target) {
 #endif  // ROCKSDB_LITE

  if (iterate_lower_bound_ != nullptr &&
-      user_comparator_.Compare(saved_key_.GetUserKey(), *iterate_lower_bound_) <
-          0) {
+      user_comparator_->Compare(saved_key_.GetUserKey(),
+                                *iterate_lower_bound_) < 0) {
    saved_key_.Clear();
    saved_key_.SetInternalKey(*iterate_lower_bound_, seq);
  }

  {
    PERF_TIMER_GUARD(seek_internal_seek_time);
-    iter_.Seek(saved_key_.GetInternalKey());
+    iter_->Seek(saved_key_.GetInternalKey());
    range_del_agg_.InvalidateRangeDelMapPositions();
  }
  RecordTick(statistics_, NUMBER_DB_SEEK);
-  if (iter_.Valid()) {
+  if (iter_->Valid()) {
    if (prefix_extractor_ && prefix_same_as_start_) {
      prefix_start_key_ = prefix_extractor_->Transform(target);
    }
@@ -1318,7 +1321,6 @@ void DBIter::Seek(const Slice& target) {
 }

 void DBIter::SeekForPrev(const Slice& target) {
-  PERF_CPU_TIMER_GUARD(iter_seek_cpu_nanos, env_);
  StopWatch sw(env_, statistics_, DB_SEEK);
  status_ = Status::OK();
  ReleaseTempPinnedData();
@@ -1329,15 +1331,15 @@ void DBIter::SeekForPrev(const Slice& target) {
                            kValueTypeForSeekForPrev);

  if (iterate_upper_bound_ != nullptr &&
-      user_comparator_.Compare(saved_key_.GetUserKey(),
-                               *iterate_upper_bound_) >= 0) {
+      user_comparator_->Compare(saved_key_.GetUserKey(),
+                                *iterate_upper_bound_) >= 0) {
    saved_key_.Clear();
    saved_key_.SetInternalKey(*iterate_upper_bound_, kMaxSequenceNumber);
  }

  {
    PERF_TIMER_GUARD(seek_internal_seek_time);
-    iter_.SeekForPrev(saved_key_.GetInternalKey());
+    iter_->SeekForPrev(saved_key_.GetInternalKey());
    range_del_agg_.InvalidateRangeDelMapPositions();
  }

@@ -1348,7 +1350,7 @@ void DBIter::SeekForPrev(const Slice& target) {
 #endif  // ROCKSDB_LITE

  RecordTick(statistics_, NUMBER_DB_SEEK);
-  if (iter_.Valid()) {
+  if (iter_->Valid()) {
    if (prefix_extractor_ && prefix_same_as_start_) {
      prefix_start_key_ = prefix_extractor_->Transform(target);
    }
@@ -1379,7 +1381,6 @@ void DBIter::SeekToFirst() {
    Seek(*iterate_lower_bound_);
    return;
  }
-  PERF_CPU_TIMER_GUARD(iter_seek_cpu_nanos, env_);
  // Don't use iter_::Seek() if we set a prefix extractor
  // because prefix seek will be used.
  if (prefix_extractor_ != nullptr && !total_order_seek_) {
@@ -1393,15 +1394,15 @@ void DBIter::SeekToFirst() {

  {
    PERF_TIMER_GUARD(seek_internal_seek_time);
-    iter_.SeekToFirst();
+    iter_->SeekToFirst();
    range_del_agg_.InvalidateRangeDelMapPositions();
  }

  RecordTick(statistics_, NUMBER_DB_SEEK);
-  if (iter_.Valid()) {
+  if (iter_->Valid()) {
    saved_key_.SetUserKey(
-        ExtractUserKey(iter_.key()),
-        !iter_.iter()->IsKeyPinned() || !pin_thru_lifetime_ /* copy */);
+        ExtractUserKey(iter_->key()),
+        !iter_->IsKeyPinned() || !pin_thru_lifetime_ /* copy */);
    FindNextUserEntry(false /* not skipping */, false /* no prefix check */);
    if (statistics_ != nullptr) {
      if (valid_) {
@@ -1424,14 +1425,13 @@ void DBIter::SeekToLast() {
  if (iterate_upper_bound_ != nullptr) {
    // Seek to last key strictly less than ReadOptions.iterate_upper_bound.
    SeekForPrev(*iterate_upper_bound_);
-    if (Valid() && user_comparator_.Equal(*iterate_upper_bound_, key())) {
+    if (Valid() && user_comparator_->Equal(*iterate_upper_bound_, key())) {
      ReleaseTempPinnedData();
      PrevInternal();
    }
    return;
  }

-  PERF_CPU_TIMER_GUARD(iter_seek_cpu_nanos, env_);
  // Don't use iter_::Seek() if we set a prefix extractor
  // because prefix seek will be used.
  if (prefix_extractor_ != nullptr && !total_order_seek_) {
@@ -1445,7 +1445,7 @@ void DBIter::SeekToLast() {

  {
    PERF_TIMER_GUARD(seek_internal_seek_time);
-    iter_.SeekToLast();
+    iter_->SeekToLast();
    range_del_agg_.InvalidateRangeDelMapPositions();
  }
  PrevInternal();
@@ -1482,7 +1482,7 @@ Iterator* NewDBIterator(Env* env, const ReadOptions& read_options,

 ArenaWrappedDBIter::~ArenaWrappedDBIter() { db_iter_->~DBIter(); }

-ReadRangeDelAggregator* ArenaWrappedDBIter::GetRangeDelAggregator() {
+RangeDelAggregator* ArenaWrappedDBIter::GetRangeDelAggregator() {
  return db_iter_->GetRangeDelAggregator();
 }

@@ -1552,17 +1552,13 @@ Status ArenaWrappedDBIter::Refresh() {
    new (&arena_) Arena();

    SuperVersion* sv = cfd_->GetReferencedSuperVersion(db_impl_->mutex());
-    if (read_callback_) {
-      read_callback_->Refresh(latest_seq);
-    }
    Init(env, read_options_, *(cfd_->ioptions()), sv->mutable_cf_options,
         latest_seq, sv->mutable_cf_options.max_sequential_skip_in_iterations,
         cur_sv_number, read_callback_, db_impl_, cfd_, allow_blob_,
         allow_refresh_);

    InternalIterator* internal_iter = db_impl_->NewInternalIterator(
-        read_options_, cfd_, sv, &arena_, db_iter_->GetRangeDelAggregator(),
-        latest_seq);
+        read_options_, cfd_, sv, &arena_, db_iter_->GetRangeDelAggregator());
    SetIterUnderDBIter(internal_iter);
  } else {
    db_iter_->set_sequence(latest_seq);
Author	SHA1	Message	Date
Fosco Marotto	cfdea78fde	Update history and version for 5.16.6 release.	2018-11-12 11:55:03 -08:00
Siying Dong	fef38d9477	Fix WriteBatchWithIndex's SeekForPrev() (#4559 ) Summary: WriteBatchWithIndex's SeekForPrev() has a bug that we internally place the position just before the seek key rather than after. This makes the iterator to miss the result that is the same as the seek key. Fix it by position the iterator equal or smaller. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4559 Differential Revision: D10468534 Pulled By: siying fbshipit-source-id: 2fb371ae809c561b60a1c11cef71e1c66fea1f19	2018-10-19 15:44:58 -07:00
Andrew Kryczka	4d461733e4	update HISTORY.md and version number	2018-10-16 10:41:36 -07:00
anand1976	1e7cca20fb	Properly determine a truncated CompactRange stop key (#4496 ) Summary: When a CompactRange() call for a level is truncated before the end key is reached, because it exceeds max_compaction_bytes, we need to properly set the compaction_end parameter to indicate the stop key. The next CompactRange will use that as the begin key. We set it to the smallest key of the next file in the level after expanding inputs to get a clean cut. Previously, we were setting it before expanding inputs. So we could end up recompacting some files. In a pathological case, where a single key has many entries spanning all the files in the level (possibly due to merge operands without a partial merge operator, thus resulting in compaction output identical to the input), this would result in an endless loop over the same set of files. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4496 Differential Revision: D10395026 Pulled By: anand1976 fbshipit-source-id: f0c2f89fee29b4b3be53b6467b53abba8e9146a9	2018-10-16 10:40:21 -07:00
Andrew Kryczka	ab21dc6323	Avoid per-key linear scan over snapshots in compaction (#4495 ) Summary: `CompactionIterator::snapshots_` is ordered by ascending seqnum, just like `DBImpl`'s linked list of snapshots from which it was copied. This PR exploits this ordering to make `findEarliestVisibleSnapshot` do binary search rather than linear scan. This can make flush/compaction significantly faster when many snapshots exist since that function is called on every single key. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4495 Differential Revision: D10386470 Pulled By: ajkr fbshipit-source-id: 29734991631227b6b7b677e156ac567690118a8b	2018-10-16 10:33:02 -07:00
Anand Ananthabhotla	fa37825fce	Update version to 5.16.4 Summary: Test Plan: Reviewers: Subscribers: Tasks: Tags:	2018-10-10 14:10:08 -07:00
Anand Ananthabhotla	4f6da54e8d	Handle mixed slowdown/no_slowdown writer properly (#4475 ) Summary: There is a bug when the write queue leader is blocked on a write delay/stop, and the queue has writers with WriteOptions::no_slowdown set to true. They are not woken up until the write stall is cleared. The fix introduces a dummy writer inserted at the tail to indicate a write stall and prevent further inserts into the queue, and a condition variable that writers who can tolerate slowdown wait on before adding themselves to the queue. The leader calls WriteThread::BeginWriteStall() to add the dummy writer and then walk the queue to fail any writers with no_slowdown set. Once the stall clears, the leader calls WriteThread::EndWriteStall() to remove the dummy writer and signal the condition variable. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4475 Differential Revision: D10285827 Pulled By: anand1976 fbshipit-source-id: 747465e5e7f07a829b1fb0bc1afcd7b93f4ab1a9	2018-10-10 12:59:34 -07:00
Andrew Kryczka	98f11d01bd	Fix CompactFiles support for kDisableCompressionOption (#4438 ) Summary: Previously `CompactFiles` with `CompressionType::kDisableCompressionOption` caused program to crash on assertion failure. This PR fixes the crash by adding support for that setting. Now, that setting will cause RocksDB to choose compression according to the column family's options. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4438 Differential Revision: D10115761 Pulled By: ajkr fbshipit-source-id: a553c6fa76fa5b6f73b0d165d95640da6f454122	2018-10-01 17:05:59 -07:00
Yi Wu	7b378c0de2	Bump version to 5.16.2	2018-09-21 09:13:34 -07:00
Yi Wu	e90493a865	BlobDB: handle IO error on read (#4410 ) Summary: Fix IO error on read not being handle and crashing the DB. With the fix we properly return the error. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4410 Differential Revision: D9979246 Pulled By: yiwu-arbug fbshipit-source-id: 111a85675067a29c03cb60e9a34103f4ff636694	2018-09-21 09:11:16 -07:00
Maysam Yabandeh	e36e75d7c8	Fix bug in partition filters with format_version=4 (#4381 ) Summary: Value delta encoding in format_version 4 requires the differences between the size of two consecutive handles to be sent to BlockBuilder::Add. This applies not only to indexes on blocks but also the indexes on indexes and filters in partitioned indexes and filters respectively. The patch fixes a bug where the partitioned filters would encode the entire size of the handle rather than the difference of the size with the last size. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4381 Differential Revision: D9879505 Pulled By: maysamyabandeh fbshipit-source-id: 27a22e49b482b927fbd5629dc310c46d63d4b6d1	2018-09-21 09:11:04 -07:00
Sagar Vemuri	895a4ad185	Bump version and update history	2018-09-17 12:20:30 -07:00
Sagar Vemuri	377d4d2c31	Fix sync-point comment in Block destructor (#4380 ) Summary: This is a follow up to #4370. The earlier comment is not correct. Thanks to ajkr for pointing this out. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4380 Differential Revision: D9874667 Pulled By: sagar0 fbshipit-source-id: f4e092d86b29c715258210b770643d367e38caae	2018-09-17 12:02:56 -07:00
Sagar Vemuri	5271d0750a	Remove sync point from Block destructor (#4370 ) Summary: AddressSanitizer: heap-use-after-free in std::__atomic_base<bool>::load(std::memory_order) const ==1798517==ABORTING ``` Pull Request resolved: https://github.com/facebook/rocksdb/pull/4370 Differential Revision: D9844146 Pulled By: sagar0 fbshipit-source-id: 18a2970b1d504b4f6c8fb04857f26e0f32124dd1	2018-09-17 10:48:21 -07:00
Yanqin Jin	29713489b6	Fix Makefile target 'jtest' on PowerPC (#4357 ) Summary: Before the fix: On a PowerPC machine, run the following ``` $ make jtest ``` The command will fail due to "undefined symbol: crc32c_ppc". It was caused by 'rocksdbjava' Makefile target not including crc32c_ppc object files when generating the shared lib. The fix is simple. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4357 Differential Revision: D9779474 Pulled By: riversand963 fbshipit-source-id: 3c5ec9068c2b9c796e6500f71cd900267064fd51	2018-09-17 10:47:45 -07:00
Yanqin Jin	abcad80559	Remove warnings caused by unused variables in jni (#4345 ) Summary: Test plan ``` $make clean jclean $make -j32 rocksdbjavastatic $make -j32 rocksdbjava ``` Pull Request resolved: https://github.com/facebook/rocksdb/pull/4345 Differential Revision: D9661256 Pulled By: riversand963 fbshipit-source-id: aed316c53b29d02fbdd3fa1063a3e832b8a66469	2018-09-17 10:47:14 -07:00
Anand Ananthabhotla	b6a8f16b5e	Remove trace_analyzer_tool.cc from rocksdb_lib target Summary: Test Plan: Reviewers: Subscribers: Tasks: Tags:	2018-09-14 18:12:11 -07:00
Maysam Yabandeh	eddc3ebd90	Reduce IndexBlockIter size (#4358 ) Summary: With #3983 the size of IndexBlockIter was increased. This had resulted in a regression on P50 latencies in one of our benchmarks. The patch reduces IndexBlockIter size be eliminating active_comparator_ field from the class. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4358 Differential Revision: D9781737 Pulled By: maysamyabandeh fbshipit-source-id: 71e2b28d90ff0813db9e04b737ae73e185583c52	2018-09-12 10:14:07 -07:00
Anand Ananthabhotla	73fcd6b8ff	Fix a lint error due to unspecified move evaluation order (#4348 ) Summary: In C++ 11, the order of argument and move evaluation in a statement such as below is unspecified - foo(a.b).bar(std::move(a)) The compiler is free to evaluate std::move(a) first, and then a.b is unspecified. In C++ 17, this will be safe if a draft proposal around function chaining rules is accepted. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4348 Differential Revision: D9688810 Pulled By: anand1976 fbshipit-source-id: e4651d0ca03dcf007e50371a0fc72c0d1e710fb4	2018-09-06 16:41:46 -07:00
Philip Jameson	7981e040a7	Grab straggler files to explicitly import AutoHeaders Summary: There were a few files that were missed when AutoHeaders were moved to their own file. Add explicit loads Reviewed By: yfeldblum Differential Revision: D9499942 fbshipit-source-id: 942bf3a683b8961e1b6244136f6337477dcc45af	2018-09-04 13:09:04 -07:00
Mikhail Antonov	5279417e73	Avoiding write stall caused by manual flushes (#4297 ) Summary: Basically at the moment it seems it's possible to cause write stall by calling flush (either manually vis DB::Flush(), or from Backup Engine directly calling FlushMemTable() while background flush may be already happening. One of the ways to fix it is that in DBImpl::CompactRange() we already check for possible stall and delay flush if needed before we actually proceed to call FlushMemTable(). We can simply move this delay logic to separate method and call it from FlushMemTable. This is draft patch, for first look; need to check tests/update SyncPoints and most certainly would need to add allow_write_stall method to FlushOptions(). Pull Request resolved: https://github.com/facebook/rocksdb/pull/4297 Differential Revision: D9420705 Pulled By: mikhail-antonov fbshipit-source-id: f81d206b55e1d7b39e4dc64242fdfbceeea03fcc	2018-08-31 14:18:59 -07:00
Andrew Kryczka	8c7b501d0f	Reduce empty SST creation/deletion in compaction (#4336 ) Summary: This is a followup to #4311. Checking `!RangeDelAggregator::IsEmpty()` before opening a dedicated range tombstone SST did not properly prevent empty SSTs from being generated. That's because it relies on `CollapsedRangeDelMap::Size`, which had an underflow bug when the map was empty. This PR fixes that underflow bug. Also fixed an uninitialized variable in db_stress. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4336 Differential Revision: D9600080 Pulled By: ajkr fbshipit-source-id: bc6980ca79d2cd01b825ebc9dbccd51c1a70cfc7	2018-08-31 12:35:59 -07:00
Fenggang Wu	3c1781d6af	DataBlockHashIndex: avoiding expensive iiter->Next when handling hash kNoEntry (#4296 ) Summary: When returning `kNoEntry` from HashIndex lookup, previously we invalidate the `biter` by set `current_=restarts_`, so that the search can continue to the next block in case the search result may reside in the next block. There is one problem: when we are searching for a missing key, if the search finds a `kNoEntry` and continue the search to the next block, there is also a non-trivial possibility that the HashIndex return `kNoEntry` too, and the expensive index iterator `Next()` will happen several times for nothing. The solution is that if the hash table returns `kNoEntry`, `SeekForGetImpl()` just search the last restart interval for the key. It will stop at the first key that is large than the seek_key, or to the end of the block, and each case will be handled correctly. Microbenchmark script: ``` TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillseq,readtocache,readmissing \ --cache_size=20000000000 --use_data_block_hash_index={true\|false} ``` `readmissing` performance (lower is better): ``` binary: 3.6098 micros/op hash (before applying diff): 4.1048 micros/op hash (after applying diff): 3.3502 micros/op ``` Pull Request resolved: https://github.com/facebook/rocksdb/pull/4296 Differential Revision: D9419159 Pulled By: fgwu fbshipit-source-id: 21e3eedcccbc47a249aa8eb4bf405c9def0b8a05	2018-08-30 15:45:26 -07:00
Fenggang Wu	9d5eb4063c	DataBlockHashIndex: fix comment in NumRestarts() (#4286 ) Summary: Improve the description of the backward compatibility check in NumRestarts() Pull Request resolved: https://github.com/facebook/rocksdb/pull/4286 Differential Revision: D9412490 Pulled By: fgwu fbshipit-source-id: ea7dd5c61d8ff8eacef623b729d4e4fd53cca066	2018-08-30 15:34:53 -07:00
Sagar Vemuri	0e35aa9930	Download bzip2 packages from Internet Archive (#4306 ) Summary: Since bzip.org is no longer maintained, download the bzip2 packages from a snapshot taken by the internet archive until we figure out a more credible source. Fixes issue: #4305 Pull Request resolved: https://github.com/facebook/rocksdb/pull/4306 Differential Revision: D9514868 Pulled By: sagar0 fbshipit-source-id: 57c6a141a62e652f94377efc7ca9916b458e68d5	2018-08-30 15:33:45 -07:00
Yi Wu	023e9bf214	BlobDB: Implement DisableFileDeletions (#4314 ) Summary: `DB::DiableFileDeletions` and `DB::EnableFileDeletions` are used for applications to stop RocksDB background jobs to delete files while they are doing replication. Implement these methods for BlobDB. `DeleteObsolteFiles` now needs to check `disable_file_deletions_` before starting, and will hold `delete_file_mutex_` the whole time while it is running. `DisableFileDeletions` needs to wait on `delete_file_mutex_` for running `DeleteObsolteFiles` job and set `disable_file_deletions_` flag. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4314 Differential Revision: D9501373 Pulled By: yiwu-arbug fbshipit-source-id: 81064c1228f1724eff46da22b50ff765b16292cd	2018-08-27 11:22:31 -07:00
Andrew Kryczka	096e5f5d02	Reduce empty SST creation/deletion during compaction (#4311 ) Summary: I have a PR to start calling `OnTableFileCreated` for empty SSTs: #4307. However, it is a behavior change so should not go into a patch release. This PR adds back a check to make sure range deletions at least exist before starting file creation. This PR should be safe to backport to earlier versions. Pull Request resolved: https://github.com/facebook/rocksdb/pull/4311 Differential Revision: D9493734 Pulled By: ajkr fbshipit-source-id: f0d43cda4cfd904f133cfe3a6eb622f52a9ccbe8	2018-08-24 14:09:39 -07:00