Merge branch 'main' into wip-ofriedma-snap-optimization

.github/workflows/build_and_publish_jar.yml

@@ -33,25 +33,25 @@ jobs:
         patch=$(( $(grep '_PATCH\s\+[0-9]\+' "$VERSION_FILE" | sed 's/[^0-9]\+//') + 1 )) && echo $patch
         echo "verSion=$major.$minor.$patch"  >> $GITHUB_OUTPUT
        
-  Mac_i86_Bld:
+  Mac_i86:
     needs: pre_build
     uses: ./.github/workflows/ci_macos.yml
     with:
       verSion: ${{ needs.pre_build.outputs.out1 }}     
 
-  Mac_ARM_Bld:
+  Mac_ARM:
     needs: pre_build
     uses: ./.github/workflows/ci_macos_ARM.yml
     with:
       verSion: ${{ needs.pre_build.outputs.out1 }}  
 
-  Ubu_ARM_Bld:
+  Ubuntu_ARM:
     needs: pre_build
-    uses: ./.github/workflows/bld_java_ubu_arm.yml
+    uses: ./.github/workflows/ci_ubuntu_arm.yml
     with:
       verSion: ${{ needs.pre_build.outputs.out1 }}  
 
-  Windows_Bld:
+  Windows:
     needs: pre_build
     uses: ./.github/workflows/ci_windows.yml
     with:
@@ -60,9 +60,9 @@ jobs:
       AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }}
       AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
       AWS_BUCKET: ${{ secrets.AWS_BUCKET }}
-
+   
   Build_and_upload:
-    needs: [pre_build, Mac_i86_Bld, Windows_Bld, Ubu_ARM_Bld]
+    needs: [pre_build, Mac_i86, Windows, Ubuntu_ARM]
     runs-on: ubu4mvn 
     env:
       VERSION_FILE: speedb/version.h
@@ -86,10 +86,12 @@ jobs:
         export JAVA_HOME="$(jrunscript -e 'java.lang.System.out.println(java.lang.System.getProperty("java.home"));')"
         export LIB_JAVA_VERSION=11.0.17
         export the_version=${{ steps.find_version.outputs.verSion }}
-        PORTABLE=1 DEBUG_LEVEL=0 make -j $(nproc) rocksdbjavastatic
+        export SPDB_LIB_DIR=~/spdb_lib &&  mkdir -p $SPDB_LIB_DIR
+        docker run --rm -v $(readlink -f ${SPDB_LIB_DIR}):/out -i speedb-centos-builder ${{ github.ref_name }}
         cd java
         mkdir src/main/resources
-        cp target/libspeedbjni-linux64.so src/main/resources
+        cp $SPDB_LIB_DIR/libspeedbjni-linux64.so src/main/resources
+        mv $SPDB_LIB_DIR/libspeedbjni-linux64.so{,_$(date '+%d_%m_%Y__%H_%M_%S')}
         echo "aws s3 --profile nd7 cp --recursive s3://spdb-builder/jar_test/v$VERSION/ java/src/main/resources/"
         sleep 180
         aws s3 --profile nd7 cp --recursive s3://spdb-builder/jar_test/v$VERSION/ src/main/resources/

.github/workflows/bld_java_ubu_arm.yml → .github/workflows/ci_ubuntu_arm.yml

@@ -23,9 +23,11 @@ jobs:
 
     - name: 'build'
       run: |
-        export JAVA_HOME=$(jrunscript -e 'java.lang.System.out.println(java.lang.System.getProperty("java.home"));')
-        PORTABLE=1 DEBUG_LEVEL=0 make -j $(nproc) rocksdbjavastatic
+        export SPDB_LIB_DIR=~/spdb_lib &&  mkdir -p $SPDB_LIB_DIR
+        docker run --rm -v $(readlink -f ${SPDB_LIB_DIR}):/out -i speedb-centos-builder ${{ github.ref_name }}
     
     - name: 'upload artifacts'    #This step executed only when this workflow is called by another and a version is provided
       if: inputs.verSion != ' '
-      run: aws s3 cp java/target/libspeedbjni-linux-aarch64.so s3://spdb-builder/jar_test/v${{ inputs.verSion }}/libspeedbjni-linux-aarch64.so
+      run: |
+        aws s3 cp ~/spdb_lib/libspeedbjni-linux-aarch64.so s3://spdb-builder/jar_test/v${{ inputs.verSion }}/libspeedbjni-linux-aarch64.so
+        mv ~/spdb_lib/libspeedbjni-linux-aarch64.so{,_$(date '+%d_%m_%Y__%H_%M_%S')}

.github/workflows/ci_windows.yml

@@ -50,4 +50,4 @@ jobs:
         AWS_REGION:  us-east-1
         SOURCE_MODE: FILE
         SOURCE_PATH: runner_bld\java\Release\speedbjni-shared.dll
-        DEST_FILE: jar_test/v${{ inputs.verSion }}/speedbjni-shared.dll
+        DEST_FILE: jar_test/v${{ inputs.verSion }}/libspeedbjni-win64.dll 

.github/workflows/sanity_check.yml

@@ -1,19 +1,19 @@
 name: Check buck targets and code format
-on: [push, workflow_call, workflow_dispatch]
+on: [push, workflow_call, workflow_dispatch, pull_request_target]
 permissions:
   contents: read
 
 jobs:
   check:
     name: Check TARGETS file and code format
-    #runs-on: ubuntu:20.04
-    runs-on: [self-hosted, ubuntu, asrunner]
-    container:
-      image: ubuntu:focal
+    runs-on: ubuntu-20.04
+    #runs-on: [self-hosted, ubuntu, asrunner]
+    #container:
+    #  image: ubuntu:focal
     steps:
 
     - name: pre
-      run: apt update && apt install -y sudo git make clang build-essential clang-format wget
+      run: sudo apt update && sudo apt install -y git make clang build-essential clang-format wget
 
     - name: Checkout feature branch
       uses: actions/checkout@v2
@@ -36,10 +36,10 @@ jobs:
         architecture: 'x64'
 
     - name: Install Dependencies
-      run: python -m pip install --upgrade pip
+      run: sudo python -m pip install --upgrade pip
 
     - name: Install argparse
-      run: pip install argparse
+      run: sudo pip install argparse
 
     - name: Download clang-format-diff.py
       run: |

CMakeLists.txt

@@ -869,6 +869,7 @@ set(SOURCES
         table/block_based/partitioned_index_iterator.cc
         table/block_based/partitioned_index_reader.cc
         table/block_based/reader_common.cc
+        table/block_based/table_pinning_policy.cc
         table/block_based/uncompression_dict_reader.cc
         table/block_fetcher.cc
         table/cuckoo/cuckoo_table_builder.cc

HISTORY.md

@@ -4,12 +4,19 @@
 
 ### New Features
 * Snapshot optimization - The most important information inside a snapshot is its Sequence number, which allows the compaction to know if the key-value should be deleted or not. The sequence number is being changed when modification happens in the db. This feature allows the db to take a snapshot without acquiring db mutex when the last snapshot has the same sequence number as a new one. In transactional db with mostly read operations, it should improve performance when used with multithreaded environment and as well other scenarios of taking large amount of snapshots with mostly read operations.
+* Add a TablePinningPolicy to the BlockBasedTableOptions.  This class controls when blocks should be pinned in memory for a block based table.  The default behavior uses the MetadataCacheOptions to control pinning and behaves identical to the previous releases. 
 
 ### Enhancements
 * db_bench: add estimate-table-readers-mem benchmark which prints these stats.
 
 ### Bug Fixes
 * unit tests: fix GlobalWriteControllerTest.GlobalAndWBMSetupDelay by waiting for the memtable memory release.
+* spdb memtable: use_seek_parallel_threshold option parameter mishandled (#570)
+* build: Plug memtable global switch memtable stuck fix.  (#606)
+* build: Windows compilation fix (#568).
+* Logger: fix Block cache stats trace by spacing it from the last trace (#578).
+* WriteController: move the class to public interface which should have been done under #346.
+* unit tests: fix DBCompactionTest.DisableMultiManualCompaction by blocking all bg compaction threads which increased by default to 8 in #194. 
 
 ## Fig v2.5.0 (06/14/2023)
 Based on RocksDB 8.1.1
@@ -51,6 +58,7 @@ Also switch to waiting a sec on the CV each time. This is required since a bg er
 * spdb memtable use after free bug (#501)
 * db_bench: Create a WBM once for all db-s regardless of their use in different groups (#550)
 * Tompstone unit test faiure (#560)
+* build: Remove unused variables in unit tests (#581)
 
 ### Miscellaneous
 * disable failing unit tests and paired bloom filter stress testing

TARGETS

@@ -201,6 +201,7 @@ cpp_library_wrapper(name="rocksdb_lib", srcs=[
         "table/block_based/partitioned_index_iterator.cc",
         "table/block_based/partitioned_index_reader.cc",
         "table/block_based/reader_common.cc",
+        "table/block_based/table_pinning_policy.cc",
         "table/block_based/uncompression_dict_reader.cc",
         "table/block_fetcher.cc",
         "table/compaction_merging_iterator.cc",
@@ -550,6 +551,7 @@ cpp_library_wrapper(name="rocksdb_whole_archive_lib", srcs=[
         "table/block_based/partitioned_index_iterator.cc",
         "table/block_based/partitioned_index_reader.cc",
         "table/block_based/reader_common.cc",
+        "table/block_based/table_pinning_policy.cc",
         "table/block_based/uncompression_dict_reader.cc",
         "table/block_fetcher.cc",
         "table/compaction_merging_iterator.cc",

db/column_family.cc

@@ -29,14 +29,14 @@
 #include "db/range_del_aggregator.h"
 #include "db/table_properties_collector.h"
 #include "db/version_set.h"
-#include "db/write_controller.h"
 #include "file/sst_file_manager_impl.h"
 #include "logging/logging.h"
 #include "monitoring/thread_status_util.h"
 #include "options/options_helper.h"
 #include "port/port.h"
 #include "rocksdb/convenience.h"
 #include "rocksdb/table.h"
+#include "rocksdb/write_controller.h"
 #include "table/merging_iterator.h"
 #include "util/autovector.h"
 #include "util/cast_util.h"

db/column_family.h

@@ -19,12 +19,12 @@
 #include "db/table_cache.h"
 #include "db/table_properties_collector.h"
 #include "db/write_batch_internal.h"
-#include "db/write_controller.h"
 #include "options/cf_options.h"
 #include "rocksdb/compaction_job_stats.h"
 #include "rocksdb/db.h"
 #include "rocksdb/env.h"
 #include "rocksdb/options.h"
+#include "rocksdb/write_controller.h"
 #include "trace_replay/block_cache_tracer.h"
 #include "util/hash_containers.h"
 #include "util/thread_local.h"

db/compaction/compaction_job.h

@@ -29,7 +29,6 @@
 #include "db/range_del_aggregator.h"
 #include "db/seqno_to_time_mapping.h"
 #include "db/version_edit.h"
-#include "db/write_controller.h"
 #include "db/write_thread.h"
 #include "logging/event_logger.h"
 #include "options/cf_options.h"
@@ -41,6 +40,7 @@
 #include "rocksdb/env.h"
 #include "rocksdb/memtablerep.h"
 #include "rocksdb/transaction_log.h"
+#include "rocksdb/write_controller.h"
 #include "table/scoped_arena_iterator.h"
 #include "util/autovector.h"
 #include "util/stop_watch.h"

db/db_compaction_test.cc

@@ -8618,9 +8618,13 @@ TEST_F(DBCompactionTest, DisableMultiManualCompaction) {
   MoveFilesToLevel(1);
 
   // Block compaction queue
-  test::SleepingBackgroundTask sleeping_task_low;
-  env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask, &sleeping_task_low,
-                 Env::Priority::LOW);
+  std::vector<test::SleepingBackgroundTask> sleeping_task_low(
+      std::max(1, env_->GetBackgroundThreads(Env::Priority::LOW)));
+
+  for (auto& sleeping_task : sleeping_task_low) {
+    env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask, &sleeping_task,
+                   Env::Priority::LOW);
+  }
 
   port::Thread compact_thread1([&]() {
     CompactRangeOptions cro;
@@ -8651,8 +8655,11 @@ TEST_F(DBCompactionTest, DisableMultiManualCompaction) {
   compact_thread1.join();
   compact_thread2.join();
 
-  sleeping_task_low.WakeUp();
-  sleeping_task_low.WaitUntilDone();
+  for (auto& sleeping_task : sleeping_task_low) {
+    sleeping_task.WakeUp();
+    sleeping_task.WaitUntilDone();
+  }
+
   ASSERT_OK(dbfull()->TEST_WaitForCompact(true));
 }
 

db/db_impl/db_impl.h

@@ -45,7 +45,6 @@
 #include "db/trim_history_scheduler.h"
 #include "db/version_edit.h"
 #include "db/wal_manager.h"
-#include "db/write_controller.h"
 #include "db/write_thread.h"
 #include "logging/event_logger.h"
 #include "monitoring/instrumented_mutex.h"
@@ -59,6 +58,7 @@
 #include "rocksdb/transaction_log.h"
 #include "rocksdb/utilities/replayer.h"
 #include "rocksdb/write_buffer_manager.h"
+#include "rocksdb/write_controller.h"
 #include "table/merging_iterator.h"
 #include "table/scoped_arena_iterator.h"
 #include "util/autovector.h"

db/db_test.cc

@@ -3484,10 +3484,8 @@ static bool CompareIterators(int step, DB* model, DB* db,
   options.snapshot = db_snap;
   Iterator* dbiter = db->NewIterator(options);
   bool ok = true;
-  int count = 0;
   for (miter->SeekToFirst(), dbiter->SeekToFirst();
        ok && miter->Valid() && dbiter->Valid(); miter->Next(), dbiter->Next()) {
-    count++;
     if (miter->key().compare(dbiter->key()) != 0) {
       fprintf(stderr, "step %d: Key mismatch: '%s' vs. '%s'\n", step,
               EscapeString(miter->key()).c_str(),

db/external_sst_file_ingestion_job.cc

@@ -679,7 +679,7 @@ Status ExternalSstFileIngestionJob::GetIngestedFileInfo(
           *cfd_->ioptions(), sv->mutable_cf_options.prefix_extractor,
           env_options_, cfd_->internal_comparator(),
           /*skip_filters*/ false, /*immortal*/ false,
-          /*force_direct_prefetch*/ false, /*level*/ -1,
+          /*force_direct_prefetch*/ false, /*level*/ -1, /*bottommost*/ false,
           /*block_cache_tracer*/ nullptr,
           /*max_file_size_for_l0_meta_pin*/ 0, versions_->DbSessionId(),
           /*cur_file_num*/ new_file_number),

db/flush_job.h

@@ -28,7 +28,6 @@
 #include "db/seqno_to_time_mapping.h"
 #include "db/snapshot_impl.h"
 #include "db/version_edit.h"
-#include "db/write_controller.h"
 #include "db/write_thread.h"
 #include "logging/event_logger.h"
 #include "monitoring/instrumented_mutex.h"
@@ -39,6 +38,7 @@
 #include "rocksdb/listener.h"
 #include "rocksdb/memtablerep.h"
 #include "rocksdb/transaction_log.h"
+#include "rocksdb/write_controller.h"
 #include "table/scoped_arena_iterator.h"
 #include "util/autovector.h"
 #include "util/stop_watch.h"

db/global_write_controller_test.cc

@@ -13,8 +13,8 @@
 // limitations under the License.
 
 #include "db/db_test_util.h"
-#include "db/write_controller.h"
 #include "rocksdb/write_buffer_manager.h"
+#include "rocksdb/write_controller.h"
 
 namespace ROCKSDB_NAMESPACE {
 

db/import_column_family_job.cc

@@ -247,7 +247,7 @@ Status ImportColumnFamilyJob::GetIngestedFileInfo(
           *cfd_->ioptions(), sv->mutable_cf_options.prefix_extractor,
           env_options_, cfd_->internal_comparator(),
           /*skip_filters*/ false, /*immortal*/ false,
-          /*force_direct_prefetch*/ false, /*level*/ -1,
+          /*force_direct_prefetch*/ false, /*level*/ -1, /*bottommost*/ false,
           /*block_cache_tracer*/ nullptr,
           /*max_file_size_for_l0_meta_pin*/ 0, versions_->DbSessionId(),
           /*cur_file_num*/ new_file_number),

db/memtable_list_test.cc

@@ -11,10 +11,10 @@
 
 #include "db/merge_context.h"
 #include "db/version_set.h"
-#include "db/write_controller.h"
 #include "rocksdb/db.h"
 #include "rocksdb/status.h"
 #include "rocksdb/write_buffer_manager.h"
+#include "rocksdb/write_controller.h"
 #include "test_util/testharness.h"
 #include "test_util/testutil.h"
 #include "util/string_util.h"

db/table_cache.cc

@@ -124,12 +124,13 @@ Status TableCache::GetTableReader(
       file->Hint(FSRandomAccessFile::kRandom);
     }
     StopWatch sw(ioptions_.clock, ioptions_.stats, TABLE_OPEN_IO_MICROS);
+    bool is_bottom = (level == ioptions_.num_levels - 1);
     std::unique_ptr<RandomAccessFileReader> file_reader(
         new RandomAccessFileReader(
             std::move(file), fname, ioptions_.clock, io_tracer_,
             record_read_stats ? ioptions_.stats : nullptr, SST_READ_MICROS,
             file_read_hist, ioptions_.rate_limiter.get(), ioptions_.listeners,
-            file_temperature, level == ioptions_.num_levels - 1));
+            file_temperature, is_bottom));
     UniqueId64x2 expected_unique_id;
     if (ioptions_.verify_sst_unique_id_in_manifest) {
       expected_unique_id = file_meta.unique_id;
@@ -140,7 +141,7 @@ Status TableCache::GetTableReader(
         ro,
         TableReaderOptions(ioptions_, prefix_extractor, file_options,
                            internal_comparator, skip_filters, immortal_tables_,
-                           false /* force_direct_prefetch */, level,
+                           false /* force_direct_prefetch */, level, is_bottom,
                            block_cache_tracer_, max_file_size_for_l0_meta_pin,
                            db_session_id_, file_meta.fd.GetNumber(),
                            expected_unique_id, file_meta.fd.largest_seqno),

db/version_set.h

@@ -45,8 +45,8 @@
 #include "db/table_cache.h"
 #include "db/version_builder.h"
 #include "db/version_edit.h"
-#include "db/write_controller.h"
 #include "env/file_system_tracer.h"
+#include "rocksdb/write_controller.h"
 #if USE_COROUTINES
 #include "folly/experimental/coro/BlockingWait.h"
 #include "folly/experimental/coro/Collect.h"

db/write_controller.cc

@@ -3,7 +3,7 @@
 //  COPYING file in the root directory) and Apache 2.0 License
 //  (found in the LICENSE.Apache file in the root directory).
 
-#include "db/write_controller.h"
+#include "rocksdb/write_controller.h"
 
 #include <algorithm>
 #include <atomic>

db/write_controller_test.cc

@@ -3,7 +3,7 @@
 //  COPYING file in the root directory) and Apache 2.0 License
 //  (found in the LICENSE.Apache file in the root directory).
 //
-#include "db/write_controller.h"
+#include "rocksdb/write_controller.h"
 
 #include <array>
 #include <ratio>

db_stress_tool/db_stress_common.h

@@ -242,6 +242,7 @@ DECLARE_bool(compression_use_zstd_dict_trainer);
 DECLARE_string(checksum_type);
 DECLARE_string(env_uri);
 DECLARE_string(fs_uri);
+DECLARE_string(pinning_policy);
 DECLARE_uint64(ops_per_thread);
 DECLARE_uint64(log2_keys_per_lock);
 DECLARE_uint64(max_manifest_file_size);

db_stress_tool/db_stress_gflags.cc

@@ -876,6 +876,8 @@ DEFINE_string(fs_uri, "",
               " with --env_uri."
               " Creates a default environment with the specified filesystem.");
 
+DEFINE_string(pinning_policy, "", "URI for registry TablePinningPolicy");
+
 DEFINE_uint64(ops_per_thread, 1200000, "Number of operations per thread.");
 static const bool FLAGS_ops_per_thread_dummy __attribute__((__unused__)) =
     RegisterFlagValidator(&FLAGS_ops_per_thread, &ValidateUint32Range);