Merge pull request #8213 from anand1976/nvm_cache_initial

Allow cache_bench and db_bench to use 3rd party tiered cache

.circleci/config.yml

@@ -17,6 +17,13 @@ commands:
           command: |
             HOMEBREW_NO_AUTO_UPDATE=1 brew install pyenv
 
+  install-cmake-on-macos:
+    steps:
+      - run:
+          name: Install cmake on macos
+          command: |
+            HOMEBREW_NO_AUTO_UPDATE=1 brew install cmake
+
   increase-max-open-files-on-macos:
     steps:
       - run:
@@ -27,10 +34,14 @@ commands:
             sudo launchctl limit maxfiles 1048576
 
   pre-steps:
+    parameters:
+      python-version:
+        default: "3.5.9"
+        type: string
     steps:
       - checkout
-      - run: pyenv install --skip-existing 3.5.9
-      - run: pyenv global 3.5.9
+      - run: pyenv install --skip-existing <<parameters.python-version>>
+      - run: pyenv global <<parameters.python-version>>
       - run:
           name: Setup Environment Variables
           command: |
@@ -39,6 +50,11 @@ commands:
             echo "export SKIP_FORMAT_BUCK_CHECKS=1" >> $BASH_ENV
             echo "export PRINT_PARALLEL_OUTPUTS=1" >> $BASH_ENV
 
+  pre-steps-macos:
+      steps:
+        - pre-steps:
+            python-version: "3.6.0"
+
   post-steps:
     steps:
       - slack/status: *notify-on-master-failure
@@ -46,6 +62,14 @@ commands:
           path: /tmp/test-results
       - store_artifacts: # store LOG for debugging if there's any
           path: LOG
+      - run: # on fail, compress Test Logs for diagnosing the issue
+         name: Compress Test Logs
+         command: tar -cvzf t.tar.gz t
+         when: on_fail
+      - store_artifacts: # on fail, store Test Logs for diagnosing the issue
+          path: t.tar.gz
+          destination: test_logs
+          when: on_fail
 
   install-clang-10:
     steps:
@@ -63,7 +87,7 @@ commands:
           name: Install gflags
           command: |
             sudo apt-get update -y && sudo apt-get install -y libgflags-dev
-  
+
   install-gflags-on-macos:
     steps:
       - run:
@@ -77,7 +101,14 @@ commands:
           name: Install gtest-parallel
           command: |
             git clone --single-branch --branch master --depth 1 https://github.com/google/gtest-parallel.git ~/gtest-parallel
-            echo "export PATH=$HOME/gtest-parallel:$PATH" >> $BASH_ENV
+            echo 'export PATH=$HOME/gtest-parallel:$PATH' >> $BASH_ENV
+
+  install-compression-libs:
+    steps:
+      - run:
+          name: Install compression libs
+          command: |
+            sudo apt-get update -y && sudo apt-get install -y libsnappy-dev zlib1g-dev libbz2-dev liblz4-dev libzstd-dev
 
 executors:
   windows-2xlarge:
@@ -89,15 +120,27 @@ executors:
 jobs:
   build-macos:
     macos:
-      xcode: 11.3.0
+      xcode: 9.4.1
     steps:
       - increase-max-open-files-on-macos
       - install-pyenv-on-macos
-      - pre-steps
       - install-gflags-on-macos
+      - pre-steps-macos
       - run: ulimit -S -n 1048576 && OPT=-DCIRCLECI make V=1 J=32 -j32 check | .circleci/cat_ignore_eagain
       - post-steps
 
+  build-macos-cmake:
+    macos:
+      xcode: 9.4.1
+    steps:
+      - increase-max-open-files-on-macos
+      - install-pyenv-on-macos
+      - install-cmake-on-macos
+      - install-gflags-on-macos
+      - pre-steps-macos
+      - run: ulimit -S -n 1048576 && (mkdir build && cd build && cmake -DWITH_GFLAGS=0 .. && make V=1 -j32) | .circleci/cat_ignore_eagain
+      - post-steps
+
   build-linux:
     machine:
       image: ubuntu-1604:202007-01
@@ -108,6 +151,26 @@ jobs:
       - run: make V=1 J=32 -j32 check | .circleci/cat_ignore_eagain
       - post-steps
 
+  build-linux-mem-env:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - run: MEM_ENV=1 make V=1 J=32 -j32 check | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-linux-encrypted-env:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - run: ENCRYPTED_ENV=1 make V=1 J=32 -j32 check | .circleci/cat_ignore_eagain
+      - post-steps
+
   build-linux-shared_lib-alt_namespace-status_checked:
     machine:
       image: ubuntu-1604:202007-01
@@ -115,19 +178,19 @@ jobs:
     steps:
       - pre-steps
       - install-gflags
-      - run: ASSERT_STATUS_CHECKED=1 TEST_UINT128_COMPAT=1 LIB_MODE=shared OPT="-DROCKSDB_NAMESPACE=alternative_rocksdb_ns" make V=1 -j32 all check_some | .circleci/cat_ignore_eagain
+      - run: ASSERT_STATUS_CHECKED=1 TEST_UINT128_COMPAT=1 ROCKSDB_MODIFY_NPHASH=1 LIB_MODE=shared OPT="-DROCKSDB_NAMESPACE=alternative_rocksdb_ns" make V=1 -j32 check | .circleci/cat_ignore_eagain
       - post-steps
 
   build-linux-release:
     machine:
       image: ubuntu-1604:202007-01
-    resource_class: 2xlarge
+    resource_class: large
     steps:
       - checkout # check out the code in the project directory
-      - run: make V=1 -j32 release | .circleci/cat_ignore_eagain
+      - run: make V=1 -j8 release | .circleci/cat_ignore_eagain
       - run: if ./db_stress --version; then false; else true; fi # ensure without gflags
       - install-gflags
-      - run: make V=1 -j32 release | .circleci/cat_ignore_eagain
+      - run: make V=1 -j8 release | .circleci/cat_ignore_eagain
       - run: ./db_stress --version # ensure with gflags
       - post-steps
 
@@ -138,11 +201,11 @@ jobs:
     steps:
       - checkout # check out the code in the project directory
       - run: make clean
-      - run: USE_RTTI=1 DEBUG_LEVEL=0 make V=1 -j16 static_lib tools db_bench | .circleci/cat_ignore_eagain
+      - run: USE_RTTI=1 DEBUG_LEVEL=0 make V=1 -j8 static_lib tools db_bench | .circleci/cat_ignore_eagain
       - run: if ./db_stress --version; then false; else true; fi # ensure without gflags
       - run: sudo apt-get update -y && sudo apt-get install -y libgflags-dev
       - run: make clean
-      - run: USE_RTTI=1 DEBUG_LEVEL=0 make V=1 -j16 static_lib tools db_bench | .circleci/cat_ignore_eagain
+      - run: USE_RTTI=1 DEBUG_LEVEL=0 make V=1 -j8 static_lib tools db_bench | .circleci/cat_ignore_eagain
       - run: ./db_stress --version # ensure with gflags
 
   build-linux-lite:
@@ -161,21 +224,21 @@ jobs:
     resource_class: large
     steps:
       - checkout # check out the code in the project directory
-      - run: LITE=1 make V=1 -j32 release | .circleci/cat_ignore_eagain
+      - run: LITE=1 make V=1 -j8 release | .circleci/cat_ignore_eagain
       - run: if ./db_stress --version; then false; else true; fi # ensure without gflags
       - install-gflags
-      - run: LITE=1 make V=1 -j32 release | .circleci/cat_ignore_eagain
+      - run: LITE=1 make V=1 -j8 release | .circleci/cat_ignore_eagain
       - run: ./db_stress --version # ensure with gflags
       - post-steps
 
-  build-linux-clang-no-test:
+  build-linux-clang-no_test_run:
     machine:
       image: ubuntu-1604:202007-01
-    resource_class: 2xlarge
+    resource_class: xlarge
     steps:
       - checkout # check out the code in the project directory
       - run: sudo apt-get update -y && sudo apt-get install -y clang libgflags-dev
-      - run: CC=clang CXX=clang++ USE_CLANG=1 PORTABLE=1 make V=1 -j32 all | .circleci/cat_ignore_eagain
+      - run: CC=clang CXX=clang++ USE_CLANG=1 PORTABLE=1 make V=1 -j16 all | .circleci/cat_ignore_eagain
       - post-steps
 
   build-linux-clang10-asan:
@@ -223,33 +286,63 @@ jobs:
       - run: CC=clang-10 CXX=clang++-10 ROCKSDB_DISABLE_ALIGNED_NEW=1 CLANG_ANALYZER="/usr/bin/clang++-10" CLANG_SCAN_BUILD=scan-build-10 USE_CLANG=1 make V=1 -j32 analyze | .circleci/cat_ignore_eagain # aligned new doesn't work for reason we haven't figured out. For unknown, reason passing "clang++-10" as CLANG_ANALYZER doesn't work, and we need a full path.
       - post-steps
 
-  build-linux-cmake:
+  build-linux-cmake-no_test_run:
     machine:
       image: ubuntu-1604:202007-01
-    resource_class: 2xlarge
+    resource_class: large
     steps:
       - checkout # check out the code in the project directory
-      - run: (mkdir build && cd build && cmake -DWITH_GFLAGS=0 .. && make V=1 -j32) | .circleci/cat_ignore_eagain
+      - run: (mkdir build && cd build && cmake -DWITH_GFLAGS=0 .. && make V=1 -j8) | .circleci/cat_ignore_eagain
       - post-steps
 
   build-linux-unity:
     docker: # executor type
       - image: gcc:latest
-    resource_class: xlarge
+    resource_class: large
     steps:
       - checkout # check out the code in the project directory
       - run: apt-get update -y && apt-get install -y libgflags-dev
-      - run: TEST_TMPDIR=/dev/shm && make V=1 -j16 unity_test | .circleci/cat_ignore_eagain
+      - run: TEST_TMPDIR=/dev/shm && make V=1 -j8 unity_test | .circleci/cat_ignore_eagain
       - post-steps
 
-  build-linux-gcc-4-8:
+  build-linux-gcc-4_8-no_test_run:
     machine:
       image: ubuntu-1604:202007-01
     resource_class: large
     steps:
       - pre-steps
       - run: sudo apt-get update -y && sudo apt-get install gcc-4.8 g++-4.8 libgflags-dev
-      - run: CC=gcc-4.8 CXX=g++-4.8 V=1 SKIP_LINK=1 make -j4 all | .circleci/cat_ignore_eagain # Linking broken because libgflags compiled with newer ABI
+      - run: CC=gcc-4.8 CXX=g++-4.8 V=1 SKIP_LINK=1 make -j8 all | .circleci/cat_ignore_eagain # Linking broken because libgflags compiled with newer ABI
+      - post-steps
+
+  build-linux-gcc-8-no_test_run:
+    machine:
+      image: ubuntu-2004:202010-01
+    resource_class: large
+    steps:
+      - pre-steps
+      - run: sudo apt-get update -y && sudo apt-get install gcc-8 g++-8 libgflags-dev
+      - run: CC=gcc-8 CXX=g++-8 V=1 SKIP_LINK=1 make -j8 all | .circleci/cat_ignore_eagain # Linking broken because libgflags compiled with newer ABI
+      - post-steps
+
+  build-linux-gcc-9-no_test_run:
+    machine:
+      image: ubuntu-2004:202010-01
+    resource_class: large
+    steps:
+      - pre-steps
+      - run: sudo apt-get update -y && sudo apt-get install gcc-9 g++-9 libgflags-dev
+      - run: CC=gcc-9 CXX=g++-9 V=1 SKIP_LINK=1 make -j8 all | .circleci/cat_ignore_eagain # Linking broken because libgflags compiled with newer ABI
+      - post-steps
+
+  build-linux-gcc-10-cxx20-no_test_run:
+    machine:
+      image: ubuntu-2004:202010-01
+    resource_class: xlarge
+    steps:
+      - pre-steps
+      - run: sudo apt-get update -y && sudo apt-get install gcc-10 g++-10 libgflags-dev
+      - run: CC=gcc-10 CXX=g++-10 V=1 SKIP_LINK=1 ROCKSDB_CXX_STANDARD=c++20 make -j16 all | .circleci/cat_ignore_eagain # Linking broken because libgflags compiled with newer ABI
       - post-steps
 
   build-windows:
@@ -278,9 +371,13 @@ jobs:
       - run:
           name: "Setup VS"
           command: |
-            if [[ "${VS_YEAR}" == "2017" ]]; then
+            if [[ "${VS_YEAR}" == "2019" ]]; then
+              echo "VS2019 already present."
+            elif [[ "${VS_YEAR}" == "2017" ]]; then
+              echo "Installing VS2017..."
               powershell .circleci/vs2017_install.ps1
             elif [[ "${VS_YEAR}" == "2015" ]]; then
+              echo "Installing VS2015..."
               powershell .circleci/vs2015_install.ps1
             fi
       - run:
@@ -306,6 +403,7 @@ jobs:
             cd build
             ${CMAKE_BIN} -G "${CMAKE_GENERATOR}" -DCMAKE_BUILD_TYPE=Debug -DOPTDBG=1 -DPORTABLE=1 -DSNAPPY=1 -DJNI=1 << parameters.extra_cmake_opt >> ..
             cd ..
+            echo "Building with VS version: ${CMAKE_GENERATOR}"
             msbuild.exe build/rocksdb.sln -maxCpuCount -property:Configuration=Debug -property:Platform=x64
       - run:
           name: "Test RocksDB"
@@ -316,25 +414,102 @@ jobs:
   build-linux-java:
     machine:
       image: ubuntu-1604:202007-01
-    resource_class: 2xlarge
+    resource_class: large
+    environment:
+      JAVA_HOME: /usr/lib/jvm/java-1.8.0-openjdk-amd64
     steps:
       - pre-steps
       - install-gflags
       - run:
-          name: "Build RocksDBJava"
+          name: "Set Java Environment"
           command: |
-            export JAVA_HOME=/usr/lib/jvm/java-1.8.0-openjdk-amd64
-            export PATH=$JAVA_HOME/bin:$PATH
             echo "JAVA_HOME=${JAVA_HOME}"
+            echo 'export PATH=$JAVA_HOME/bin:$PATH' >> $BASH_ENV
             which java && java -version
             which javac && javac -version
-            make V=1 J=32 -j32 rocksdbjava jtest | .circleci/cat_ignore_eagain
+      - run:
+          name: "Build RocksDBJava Shared Library"
+          command: make V=1 J=8 -j8 rocksdbjava | .circleci/cat_ignore_eagain
+      - run:
+          name: "Test RocksDBJava"
+          command: make V=1 J=8 -j8 jtest | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-linux-java-static:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: large
+    environment:
+      JAVA_HOME: /usr/lib/jvm/java-1.8.0-openjdk-amd64
+    steps:
+      - pre-steps
+      - install-gflags
+      - run:
+          name: "Set Java Environment"
+          command: |
+            echo "JAVA_HOME=${JAVA_HOME}"
+            echo 'export PATH=$JAVA_HOME/bin:$PATH' >> $BASH_ENV
+            which java && java -version
+            which javac && javac -version
+      - run:
+          name: "Build RocksDBJava Static Library"
+          command: make V=1 J=8 -j8 rocksdbjavastatic | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-macos-java:
+    macos:
+      xcode: 9.4.1
+    resource_class: medium
+    environment:
+      JAVA_HOME: /Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home
+    steps:
+      - increase-max-open-files-on-macos
+      - install-pyenv-on-macos
+      - install-gflags-on-macos
+      - pre-steps-macos
+      - run:
+          name: "Set Java Environment"
+          command: |
+            echo "JAVA_HOME=${JAVA_HOME}"
+            echo 'export PATH=$JAVA_HOME/bin:$PATH' >> $BASH_ENV
+            which java && java -version
+            which javac && javac -version
+      - run:
+          name: "Build RocksDBJava Shared Library"
+          command: make V=1 J=8 -j8 rocksdbjava | .circleci/cat_ignore_eagain
+      - run:
+          name: "Test RocksDBJava"
+          command: make V=1 J=8 -j8 jtest | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-macos-java-static:
+    macos:
+      xcode: 9.4.1
+    resource_class: medium
+    environment:
+      JAVA_HOME: /Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home
+    steps:
+      - increase-max-open-files-on-macos
+      - install-pyenv-on-macos
+      - install-gflags-on-macos
+      - install-cmake-on-macos
+      - pre-steps-macos
+      - run:
+          name: "Set Java Environment"
+          command: |
+            echo "JAVA_HOME=${JAVA_HOME}"
+            echo 'export PATH=$JAVA_HOME/bin:$PATH' >> $BASH_ENV
+            which java && java -version
+            which javac && javac -version
+      - run:
+          name: "Build RocksDBJava Static Library"
+          command: make V=1 J=8 -j8 rocksdbjavastatic | .circleci/cat_ignore_eagain
       - post-steps
 
   build-examples:
     machine:
       image: ubuntu-1604:202007-01
-    resource_class: medium
+    resource_class: large
     steps:
       - pre-steps
       - install-gflags
@@ -344,6 +519,29 @@ jobs:
             OPT=-DTRAVIS V=1 make -j4 static_lib && cd examples && make -j4 | ../.circleci/cat_ignore_eagain
       - post-steps
 
+  build-cmake-mingw:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - run: sudo apt-get update -y && sudo apt-get install -y mingw-w64
+      - run: sudo update-alternatives --set x86_64-w64-mingw32-g++ /usr/bin/x86_64-w64-mingw32-g++-posix
+      - run:
+          name: "Build cmake-mingw"
+          command: |
+            sudo apt-get install snapd && sudo snap install cmake --beta --classic
+            export PATH=/snap/bin:$PATH
+            sudo apt-get install -y openjdk-8-jdk
+            export JAVA_HOME=/usr/lib/jvm/java-8-openjdk-amd64
+            export PATH=$JAVA_HOME/bin:$PATH
+            echo "JAVA_HOME=${JAVA_HOME}"
+            which java && java -version
+            which javac && javac -version
+            mkdir build && cd build && cmake -DJNI=1 -DWITH_GFLAGS=OFF .. -DCMAKE_C_COMPILER=x86_64-w64-mingw32-gcc -DCMAKE_CXX_COMPILER=x86_64-w64-mingw32-g++ -DCMAKE_SYSTEM_NAME=Windows && make -j4 rocksdb rocksdbjni
+      - post-steps
+
   build-linux-non-shm:
     machine:
       image: ubuntu-1604:202007-01
@@ -371,13 +569,89 @@ jobs:
             sed -i 's/[[:space:]]*$//; s/ / \.\//g; s/.*/.\/&/' /tmp/test_list
             cat /tmp/test_list
             export TEST_TMPDIR=/tmp/rocksdb_test_tmp
-            /usr/bin/python ../gtest-parallel/gtest-parallel $(</tmp/test_list) --output_dir=/tmp | cat  # pipe to cat to continuously output status on circleci UI. Otherwise, no status will be printed while the job is running.
+            gtest-parallel $(</tmp/test_list) --output_dir=/tmp | cat  # pipe to cat to continuously output status on circleci UI. Otherwise, no status will be printed while the job is running.
+      - post-steps
+
+  build-linux-arm-test-full:
+    machine:
+      image: ubuntu-2004:202101-01
+    resource_class: arm.large
+    steps:
+      - pre-steps
+      - install-gflags
+      - run: make V=1 J=4 -j4 check | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-linux-arm:
+    machine:
+      image: ubuntu-2004:202101-01
+    resource_class: arm.large
+    steps:
+      - pre-steps
+      - install-gflags
+      - run: ROCKSDBTESTS_PLATFORM_DEPENDENT=only make V=1 J=4 -j4 all_but_some_tests check_some | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-linux-arm-cmake-no_test_run:
+    machine:
+      image: ubuntu-2004:202101-01
+    resource_class: arm.large
+    environment:
+      JAVA_HOME: /usr/lib/jvm/java-8-openjdk-arm64
+    steps:
+      - pre-steps
+      - run:
+          name: "Set Java Environment"
+          command: |
+            echo "JAVA_HOME=${JAVA_HOME}"
+            echo 'export PATH=$JAVA_HOME/bin:$PATH' >> $BASH_ENV
+            which java && java -version
+            which javac && javac -version
+      - run:
+          name: "Build with cmake"
+          command: |
+            mkdir build
+            cd build
+            cmake -DCMAKE_BUILD_TYPE=Release -DWITH_TESTS=0 -DWITH_GFLAGS=0 -DWITH_BENCHMARK_TOOLS=0 -DWITH_TOOLS=0 -DWITH_CORE_TOOLS=1 ..
+            make -j4
+      - run:
+          name: "Build Java with cmake"
+          command: |
+            rm -rf build
+            mkdir build
+            cd build
+            cmake -DJNI=1 -DCMAKE_BUILD_TYPE=Release -DWITH_GFLAGS=0 ..
+            make -j4 rocksdb rocksdbjni
+      - post-steps
+
+  build-format-compatible:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - install-compression-libs
+      - run:
+          name: "test"
+          command: |
+            export TEST_TMPDIR=/dev/shm/rocksdb
+            rm -rf /dev/shm/rocksdb
+            mkdir /dev/shm/rocksdb
+            tools/check_format_compatible.sh
       - post-steps
 
 workflows:
+  version: 2
   build-linux:
     jobs:
       - build-linux
+  build-linux-mem-env:
+    jobs:
+      - build-linux-mem-env
+  build-linux-encrypted-env:
+    jobs:
+      - build-linux-encrypted-env
   build-linux-shared_lib-alt_namespace-status_checked:
     jobs:
       - build-linux-shared_lib-alt_namespace-status_checked
@@ -393,9 +667,6 @@ workflows:
   build-linux-lite-release:
     jobs:
       - build-linux-lite-release
-  build-linux-clang-no-test:
-    jobs:
-      - build-linux-clang-no-test
   build-linux-clang10-asan:
     jobs:
       - build-linux-clang10-asan
@@ -408,32 +679,36 @@ workflows:
   build-linux-clang10-clang-analyze:
     jobs:
       - build-linux-clang10-clang-analyze
-  build-linux-cmake:
-    jobs:
-      - build-linux-cmake
   build-linux-unity:
     jobs:
       - build-linux-unity
-  build-windows:
+  build-windows-vs2019:
     jobs:
-      - build-windows
+      - build-windows:
+          name: "build-windows-vs2019"
+  build-windows-vs2019-cxx20:
+    jobs:
+      - build-windows:
+          name: "build-windows-vs2019-cxx20"
+          extra_cmake_opt: -DCMAKE_CXX_STANDARD=20
   build-windows-vs2017:
     jobs:
       - build-windows:
+          name: "build-windows-vs2017"
           vs_year: "2017"
           cmake_generator: "Visual Studio 15 Win64"
   build-windows-vs2015:
     jobs:
       - build-windows:
+          name: "build-windows-vs2015"
           vs_year: "2015"
           cmake_generator: "Visual Studio 14 Win64"
-  build-windows-cxx20:
-    jobs:
-      - build-windows:
-          extra_cmake_opt: -DCMAKE_CXX_STANDARD=20
   build-java:
     jobs:
       - build-linux-java
+      - build-linux-java-static
+      - build-macos-java
+      - build-macos-java-static
   build-examples:
     jobs:
       - build-examples
@@ -451,9 +726,35 @@ workflows:
       - build-linux-non-shm:
           start_test: "statistics_test" # make sure unique in src.mk
           end_test: ""
-  build-linux-gcc-4-8:
+  build-linux-compilers-no_test_run:
     jobs:
-      - build-linux-gcc-4-8
+      - build-linux-clang-no_test_run
+      - build-linux-cmake-no_test_run
+      - build-linux-gcc-4_8-no_test_run
+      - build-linux-gcc-8-no_test_run
+      - build-linux-gcc-9-no_test_run
+      - build-linux-gcc-10-cxx20-no_test_run
+      - build-linux-arm-cmake-no_test_run
   build-macos:
     jobs:
       - build-macos
+  build-macos-cmake:
+    jobs:
+      - build-macos-cmake
+  build-cmake-mingw:
+    jobs:
+      - build-cmake-mingw
+  build-linux-arm:
+    jobs:
+      - build-linux-arm
+  nightly:
+    triggers:
+      - schedule:
+          cron: "0 0 * * *"
+          filters:
+            branches:
+              only:
+                - master
+    jobs:
+      - build-format-compatible
+      - build-linux-arm-test-full

.circleci/vs2015_install.ps1

@@ -10,7 +10,7 @@ $process = Start-Process "${PWD}\vs_installer.exe" -ArgumentList $VS_INSTALL_ARG
 Remove-Item -Path vs_installer.exe -Force
 $exitCode = $process.ExitCode
 if (($exitCode -ne 0) -and ($exitCode -ne 3010)) {
-    echo "VS 2017 installer exited with code $exitCode, which should be one of [0, 3010]."
+    echo "VS 2015 installer exited with code $exitCode, which should be one of [0, 3010]."
     curl.exe --retry 3 -kL $COLLECT_DOWNLOAD_LINK --output Collect.exe
     if ($LASTEXITCODE -ne 0) {
         echo "Download of the VS Collect tool failed."
@@ -21,3 +21,4 @@ if (($exitCode -ne 0) -and ($exitCode -ne 3010)) {
     Copy-Item -Path "C:\Users\circleci\AppData\Local\Temp\vslogs.zip" -Destination "C:\w\build-results\"
     exit 1
 }
+echo "VS 2015 installed."

.circleci/vs2017_install.ps1

@@ -32,3 +32,4 @@ if (($exitCode -ne 0) -and ($exitCode -ne 3010)) {
     Copy-Item -Path "C:\Users\circleci\AppData\Local\Temp\vslogs.zip" -Destination "C:\w\build-results\"
     exit 1
 }
+echo "VS 2017 installed."

.github/workflows/sanity_check.yml

@@ -32,7 +32,7 @@ jobs:
     - name: Download clang-format-diff.py
       uses: wei/wget@v1
       with:
-        args: https://raw.githubusercontent.com/llvm-mirror/clang/master/tools/clang-format/clang-format-diff.py
+        args: https://raw.githubusercontent.com/llvm/llvm-project/main/clang/tools/clang-format/clang-format-diff.py
 
     - name: Check format
       run: VERBOSE_CHECK=1 make check-format

.gitignore

@@ -8,6 +8,7 @@ rocksdb.pc
 *.gcda
 *.gcno
 *.o
+*.o.tmp
 *.so
 *.so.*
 *_test
@@ -89,3 +90,8 @@ buckifier/__pycache__
 compile_commands.json
 clang-format-diff.py
 .py3/
+
+fuzz/proto/gen/
+fuzz/crash-*
+
+cmake-build-*

.travis.yml

@@ -2,20 +2,18 @@ dist: xenial
 language: cpp
 os:
   - linux
-  - osx
 arch:
-  - amd64
   - arm64
   - ppc64le
 compiler:
   - clang
   - gcc
-osx_image: xcode9.4
 cache:
   - ccache
 
 addons:
   apt:
+    update: true
     sources:
       - ubuntu-toolchain-r-test
     packages:
@@ -26,15 +24,6 @@ addons:
       - liblzma-dev  # xv
       - libzstd-dev
       - zlib1g-dev
-  homebrew:
-    update: true
-    packages:
-      - ccache
-      - gflags
-      - lz4
-      - snappy
-      - xz
-      - zstd
 
 env:
   - TEST_GROUP=platform_dependent # 16-18 minutes
@@ -58,20 +47,6 @@ env:
 
 matrix:
   exclude:
-  - os: osx
-    env: JOB_NAME=cmake-gcc8
-  - os: osx
-    env: JOB_NAME=cmake-gcc9
-  - os: osx
-    env: JOB_NAME=cmake-gcc9-c++20
-  - os: osx
-    env: JOB_NAME=cmake-mingw
-  - os: osx
-    env: JOB_NAME=make-gcc4.8
-  - os: osx
-    arch: ppc64le
-  - os: osx
-    compiler: gcc
   - os : linux
     arch: arm64
     env: JOB_NAME=cmake-mingw
@@ -86,73 +61,46 @@ matrix:
     env: JOB_NAME=make-gcc4.8
   - os: linux
     compiler: clang
-  # Exclude all but most unique cmake variants for pull requests, but build all in branches
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: arm64
+    env: TEST_GROUP=platform_dependent
   - if: type = pull_request AND commit_message !~ /FULL_CI/
     os : linux
-    arch: amd64
+    arch: arm64
+    env: TEST_GROUP=1
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: TEST_GROUP=1
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: TEST_GROUP=2
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: TEST_GROUP=2
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: TEST_GROUP=3
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: TEST_GROUP=3
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: TEST_GROUP=4
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: TEST_GROUP=4
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
     env: JOB_NAME=cmake
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os : linux
-    arch: amd64
-    env: JOB_NAME=cmake-gcc8
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os : linux
-    arch: amd64
-    env: JOB_NAME=cmake-gcc9
-  # Exclude most osx, arm64 and ppc64le tests for pull requests, but build in branches
-  # Temporarily disable ppc64le cmake test while snapd is broken
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os: linux
-    arch: ppc64le
-    env: JOB_NAME=cmake
-  # NB: the cmake build is a partial java test
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os: osx
-    env: TEST_GROUP=1
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os : linux
-    arch: arm64
-    env: TEST_GROUP=1
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os: linux
-    arch: ppc64le
-    env: TEST_GROUP=1
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os: osx
-    env: TEST_GROUP=2
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os : linux
-    arch: arm64
-    env: TEST_GROUP=2
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os: linux
-    arch: ppc64le
-    env: TEST_GROUP=2
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os: osx
-    env: TEST_GROUP=3
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os : linux
-    arch: arm64
-    env: TEST_GROUP=3
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os: linux
-    arch: ppc64le
-    env: TEST_GROUP=3
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os: osx
-    env: TEST_GROUP=4
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os : linux
-    arch: arm64
-    env: TEST_GROUP=4
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os: linux
-    arch: ppc64le
-    env: TEST_GROUP=4
-  - if: type = pull_request AND commit_message !~ /FULL_CI/ AND commit_message !~ /java/
-    os : osx
-    env: JOB_NAME=java_test
   - if: type = pull_request AND commit_message !~ /FULL_CI/ AND commit_message !~ /java/
     os : linux
     arch: arm64
@@ -161,9 +109,6 @@ matrix:
     os: linux
     arch: ppc64le
     env: JOB_NAME=java_test
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os : osx
-    env: JOB_NAME=lite_build
   - if: type = pull_request AND commit_message !~ /FULL_CI/
     os : linux
     arch: arm64
@@ -172,9 +117,6 @@ matrix:
     os: linux
     arch: ppc64le
     env: JOB_NAME=lite_build
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os : osx
-    env: JOB_NAME=examples
   - if: type = pull_request AND commit_message !~ /FULL_CI/
     os : linux
     arch: arm64
@@ -207,9 +149,6 @@ matrix:
     os: linux
     arch: ppc64le
     env: JOB_NAME=cmake-gcc9-c++20
-  - if: type = pull_request AND commit_message !~ /FULL_CI/
-    os : osx
-    env: JOB_NAME=status_checked
   - if: type = pull_request AND commit_message !~ /FULL_CI/
     os : linux
     arch: arm64
@@ -220,9 +159,6 @@ matrix:
     env: JOB_NAME=status_checked
 
 install:
-  - if [ "${TRAVIS_OS_NAME}" == osx ]; then
-      PATH=$PATH:/usr/local/opt/ccache/libexec;
-    fi
   - if [ "${JOB_NAME}" == cmake-gcc8 ]; then
       sudo apt-get install -y g++-8 || exit $?;
       CC=gcc-8 && CXX=g++-8;
@@ -238,22 +174,21 @@ install:
       sudo apt-get install -y g++-4.8 || exit $?;
       CC=gcc-4.8 && CXX=g++-4.8;
     fi
-  - if [[ "${JOB_NAME}" == cmake* ]] && [ "${TRAVIS_OS_NAME}" == linux ]; then
-      sudo apt-get install snapd && sudo snap install cmake --beta --classic || exit $?;
-      export PATH=/snap/bin:$PATH;
+  - |
+    if [[ "${JOB_NAME}" == cmake* ]]; then
+      sudo apt-get remove -y cmake cmake-data
+      export CMAKE_DEB="cmake-3.14.5-Linux-$(uname -m).deb"
+      export CMAKE_DEB_URL="https://rocksdb-deps.s3-us-west-2.amazonaws.com/cmake/${CMAKE_DEB}"
+      curl --silent --fail --show-error --location --output "${CMAKE_DEB}" "${CMAKE_DEB_URL}" || exit $?
+      sudo dpkg -i "${CMAKE_DEB}" || exit $?
+      which cmake && cmake --version
     fi
   - |
     if [[ "${JOB_NAME}" == java_test || "${JOB_NAME}" == cmake* ]]; then
       # Ensure JDK 8
-      if [ "${TRAVIS_OS_NAME}" == osx ]; then
-        brew tap AdoptOpenJDK/openjdk || exit $?
-        brew cask install adoptopenjdk8 || exit $?
-        export JAVA_HOME=$(/usr/libexec/java_home)
-      else
-        sudo apt-get install -y openjdk-8-jdk || exit $?
-        export PATH=/usr/lib/jvm/java-8-openjdk-$(dpkg --print-architecture)/bin:$PATH
-        export JAVA_HOME=/usr/lib/jvm/java-8-openjdk-$(dpkg --print-architecture)
-      fi
+      sudo apt-get install -y openjdk-8-jdk || exit $?
+      export PATH=/usr/lib/jvm/java-8-openjdk-$(dpkg --print-architecture)/bin:$PATH
+      export JAVA_HOME=/usr/lib/jvm/java-8-openjdk-$(dpkg --print-architecture)
       echo "JAVA_HOME=${JAVA_HOME}"
       which java && java -version
       which javac && javac -version
@@ -269,30 +204,30 @@ script:
   - if [ `command -v ccache` ]; then ccache -C; fi
   - case $TEST_GROUP in
     platform_dependent)
-      OPT=-DTRAVIS V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=only make -j4 all_but_some_tests check_some
+      OPT=-DTRAVIS LIB_MODE=shared V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=only make -j4 all_but_some_tests check_some
       ;;
     1)
-      OPT=-DTRAVIS V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=exclude ROCKSDBTESTS_END=backupable_db_test make -j4 check_some
+      OPT=-DTRAVIS LIB_MODE=shared V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=exclude ROCKSDBTESTS_END=backupable_db_test make -j4 check_some
       ;;
     2)
-      OPT="-DTRAVIS -DROCKSDB_NAMESPACE=alternative_rocksdb_ns" V=1 make -j4 tools && OPT="-DTRAVIS -DROCKSDB_NAMESPACE=alternative_rocksdb_ns" V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=exclude ROCKSDBTESTS_START=backupable_db_test ROCKSDBTESTS_END=db_universal_compaction_test make -j4 check_some
+      OPT="-DTRAVIS -DROCKSDB_NAMESPACE=alternative_rocksdb_ns" LIB_MODE=shared V=1 make -j4 tools && OPT="-DTRAVIS -DROCKSDB_NAMESPACE=alternative_rocksdb_ns" LIB_MODE=shared V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=exclude ROCKSDBTESTS_START=backupable_db_test ROCKSDBTESTS_END=db_universal_compaction_test make -j4 check_some
       ;;
     3)
       OPT=-DTRAVIS LIB_MODE=shared V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=exclude ROCKSDBTESTS_START=db_universal_compaction_test ROCKSDBTESTS_END=table_properties_collector_test make -j4 check_some
       ;;
     4)
-      OPT=-DTRAVIS V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=exclude ROCKSDBTESTS_START=table_properties_collector_test make -j4 check_some
+      OPT=-DTRAVIS LIB_MODE=shared V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=exclude ROCKSDBTESTS_START=table_properties_collector_test make -j4 check_some
       ;;
     esac
   - case $JOB_NAME in
     java_test)
-      OPT=-DTRAVIS V=1 make rocksdbjava jtest
+      OPT=-DTRAVIS LIB_MODE=shared V=1 make rocksdbjava jtest
       ;;
     lite_build)
-      OPT='-DTRAVIS -DROCKSDB_LITE' V=1 make -j4 all
+      OPT='-DTRAVIS -DROCKSDB_LITE' LIB_MODE=shared V=1 make -j4 all
       ;;
     examples)
-      OPT=-DTRAVIS V=1 make -j4 static_lib && cd examples && make -j4
+      OPT=-DTRAVIS LIB_MODE=shared V=1 make -j4 static_lib && cd examples && make -j4
       ;;
     cmake-mingw)
       sudo update-alternatives --set x86_64-w64-mingw32-g++ /usr/bin/x86_64-w64-mingw32-g++-posix;
@@ -308,10 +243,10 @@ script:
       mkdir build && cd build && cmake -DCMAKE_BUILD_TYPE=Release -DWITH_TESTS=0 -DWITH_GFLAGS=0 -DWITH_BENCHMARK_TOOLS=0 -DWITH_TOOLS=0 -DWITH_CORE_TOOLS=1 .. && make -j4 && cd  .. && rm -rf build && mkdir build && cd build && cmake -DJNI=1 .. -DCMAKE_BUILD_TYPE=Release $OPT && make -j4 rocksdb rocksdbjni
       ;;
     make-gcc4.8)
-      OPT=-DTRAVIS V=1 SKIP_LINK=1 make -j4 all && [ "Linking broken because libgflags compiled with newer ABI" ]
+      OPT=-DTRAVIS LIB_MODE=shared V=1 SKIP_LINK=1 make -j4 all && [ "Linking broken because libgflags compiled with newer ABI" ]
       ;;
     status_checked)
-      OPT=-DTRAVIS V=1 ASSERT_STATUS_CHECKED=1 make -j4 check_some
+      OPT=-DTRAVIS LIB_MODE=shared V=1 ASSERT_STATUS_CHECKED=1 make -j4 check_some
       ;;
     esac
 notifications:

CMakeLists.txt

@@ -88,10 +88,9 @@ if( NOT DEFINED CMAKE_CXX_STANDARD )
 endif()
 
 include(CMakeDependentOption)
-CMAKE_DEPENDENT_OPTION(WITH_GFLAGS "build with GFlags" ON
-  "NOT MSVC;NOT MINGW" OFF)
 
 if(MSVC)
+  option(WITH_GFLAGS "build with GFlags" OFF)
   option(WITH_XPRESS "build with windows built in compression" OFF)
   include(${CMAKE_CURRENT_SOURCE_DIR}/thirdparty.inc)
 else()
@@ -107,6 +106,11 @@ else()
     endif()
   endif()
 
+  if(MINGW)
+    option(WITH_GFLAGS "build with GFlags" OFF)
+  else()
+    option(WITH_GFLAGS "build with GFlags" ON)
+  endif()
   set(GFLAGS_LIB)
   if(WITH_GFLAGS)
     # Config with namespace available since gflags 2.2.2
@@ -170,23 +174,25 @@ else()
   endif()
 endif()
 
-string(TIMESTAMP TS "%Y/%m/%d %H:%M:%S" UTC)
-set(GIT_DATE_TIME "${TS}" CACHE STRING "the time we first built rocksdb")
+string(TIMESTAMP TS "%Y-%m-%d %H:%M:%S" UTC)
+set(BUILD_DATE "${TS}" CACHE STRING "the time we first built rocksdb")
 
 find_package(Git)
 
 if(GIT_FOUND AND EXISTS "${CMAKE_CURRENT_SOURCE_DIR}/.git")
-  if(WIN32)
-    execute_process(COMMAND $ENV{COMSPEC} /C ${GIT_EXECUTABLE} -C ${CMAKE_CURRENT_SOURCE_DIR} rev-parse HEAD OUTPUT_VARIABLE GIT_SHA)
-  else()
-    execute_process(COMMAND ${GIT_EXECUTABLE} -C ${CMAKE_CURRENT_SOURCE_DIR} rev-parse HEAD OUTPUT_VARIABLE GIT_SHA)
+  execute_process(WORKING_DIRECTORY "${CMAKE_CURRENT_SOURCE_DIR}" OUTPUT_VARIABLE GIT_SHA COMMAND "${GIT_EXECUTABLE}" rev-parse HEAD )
+  execute_process(WORKING_DIRECTORY "${CMAKE_CURRENT_SOURCE_DIR}" RESULT_VARIABLE GIT_MOD COMMAND "${GIT_EXECUTABLE}" diff-index HEAD --quiet)
+  execute_process(WORKING_DIRECTORY "${CMAKE_CURRENT_SOURCE_DIR}" OUTPUT_VARIABLE GIT_DATE COMMAND "${GIT_EXECUTABLE}" log -1 --date=format:"%Y-%m-%d %T" --format="%ad")
+  execute_process(WORKING_DIRECTORY "${CMAKE_CURRENT_SOURCE_DIR}" OUTPUT_VARIABLE GIT_TAG RESULT_VARIABLE rv COMMAND "${GIT_EXECUTABLE}" symbolic-ref -q --short HEAD OUTPUT_STRIP_TRAILING_WHITESPACE)
+  if (rv AND NOT rv EQUAL 0)
+    execute_process(WORKING_DIRECTORY "${CMAKE_CURRENT_SOURCE_DIR}" OUTPUT_VARIABLE GIT_TAG COMMAND "${GIT_EXECUTABLE}" describe --tags --exact-match OUTPUT_STRIP_TRAILING_WHITESPACE)
   endif()
 else()
   set(GIT_SHA 0)
+  set(GIT_MOD 1)
 endif()
-
-string(REGEX REPLACE "[^0-9a-f]+" "" GIT_SHA "${GIT_SHA}")
-
+string(REGEX REPLACE "[^0-9a-fA-F]+" "" GIT_SHA "${GIT_SHA}")
+string(REGEX REPLACE "[^0-9: /-]+" "" GIT_DATE "${GIT_DATE}")
 
 option(WITH_MD_LIBRARY "build with MD" ON)
 if(WIN32 AND MSVC)
@@ -199,15 +205,16 @@ endif()
 
 set(BUILD_VERSION_CC ${CMAKE_BINARY_DIR}/build_version.cc)
 configure_file(util/build_version.cc.in ${BUILD_VERSION_CC} @ONLY)
-add_library(build_version OBJECT ${BUILD_VERSION_CC})
-target_include_directories(build_version PRIVATE
-  ${CMAKE_CURRENT_SOURCE_DIR}/util)
+
 if(MSVC)
   set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /Zi /nologo /EHsc /GS /Gd /GR /GF /fp:precise /Zc:wchar_t /Zc:forScope /errorReport:queue")
   set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /FC /d2Zi+ /W4 /wd4127 /wd4800 /wd4996 /wd4351 /wd4100 /wd4204 /wd4324")
 else()
   set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -W -Wextra -Wall")
   set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wsign-compare -Wshadow -Wno-unused-parameter -Wno-unused-variable -Woverloaded-virtual -Wnon-virtual-dtor -Wno-missing-field-initializers -Wno-strict-aliasing")
+  if(CMAKE_SYSTEM_PROCESSOR MATCHES "x86_64")
+    set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Wstrict-prototypes")
+  endif()
   if(MINGW)
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-format -fno-asynchronous-unwind-tables")
     add_definitions(-D_POSIX_C_SOURCE=1)
@@ -243,14 +250,14 @@ if(CMAKE_SYSTEM_PROCESSOR MATCHES "^(powerpc|ppc)64")
   endif(HAS_ALTIVEC)
 endif(CMAKE_SYSTEM_PROCESSOR MATCHES "^(powerpc|ppc)64")
 
-if(CMAKE_SYSTEM_PROCESSOR MATCHES "aarch64|AARCH64")
+if(CMAKE_SYSTEM_PROCESSOR MATCHES "arm64|aarch64|AARCH64")
         CHECK_C_COMPILER_FLAG("-march=armv8-a+crc+crypto" HAS_ARMV8_CRC)
   if(HAS_ARMV8_CRC)
     message(STATUS " HAS_ARMV8_CRC yes")
     set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -march=armv8-a+crc+crypto -Wno-unused-function")
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -march=armv8-a+crc+crypto -Wno-unused-function")
   endif(HAS_ARMV8_CRC)
-endif(CMAKE_SYSTEM_PROCESSOR MATCHES "aarch64|AARCH64")
+endif(CMAKE_SYSTEM_PROCESSOR MATCHES "arm64|aarch64|AARCH64")
 
 option(PORTABLE "build a portable binary" OFF)
 option(FORCE_SSE42 "force building with SSE4.2, even when PORTABLE=ON" OFF)
@@ -289,6 +296,7 @@ else()
 endif()
 
 include(CheckCXXSourceCompiles)
+set(OLD_CMAKE_REQUIRED_FLAGS ${CMAKE_REQUIRED_FLAGS})
 if(NOT MSVC)
   set(CMAKE_REQUIRED_FLAGS "-msse4.2 -mpclmul")
 endif()
@@ -305,7 +313,6 @@ int main() {
   auto d = _mm_cvtsi128_si64(c);
 }
 " HAVE_SSE42)
-unset(CMAKE_REQUIRED_FLAGS)
 if(HAVE_SSE42)
   add_definitions(-DHAVE_SSE42)
   add_definitions(-DHAVE_PCLMUL)
@@ -313,18 +320,50 @@ elseif(FORCE_SSE42)
   message(FATAL_ERROR "FORCE_SSE42=ON but unable to compile with SSE4.2 enabled")
 endif()
 
+# Check if -latomic is required or not
+if (NOT MSVC)
+  set(CMAKE_REQUIRED_FLAGS "--std=c++11")
+  CHECK_CXX_SOURCE_COMPILES("
+#include <atomic>
+std::atomic<uint64_t> x(0);
+int main() {
+  uint64_t i = x.load(std::memory_order_relaxed);
+  bool b = x.is_lock_free();
+  return 0;
+}
+" BUILTIN_ATOMIC)
+if (NOT BUILTIN_ATOMIC)
+  #TODO: Check if -latomic exists
+  list(APPEND THIRDPARTY_LIBS atomic)
+endif()
+endif()
+
+# Reset the required flags
+set(CMAKE_REQUIRED_FLAGS ${OLD_CMAKE_REQUIRED_FLAGS})
+
 CHECK_CXX_SOURCE_COMPILES("
 #if defined(_MSC_VER) && !defined(__thread)
 #define __thread __declspec(thread)
 #endif
 int main() {
   static __thread int tls;
+  (void)tls;
 }
 " HAVE_THREAD_LOCAL)
 if(HAVE_THREAD_LOCAL)
   add_definitions(-DROCKSDB_SUPPORT_THREAD_LOCAL)
 endif()
 
+option(WITH_IOSTATS_CONTEXT "Enable IO stats context" ON)
+if (NOT WITH_IOSTATS_CONTEXT)
+  add_definitions(-DNIOSTATS_CONTEXT)
+endif()
+
+option(WITH_PERF_CONTEXT "Enable perf context" ON)
+if (NOT WITH_PERF_CONTEXT)
+  add_definitions(-DNPERF_CONTEXT)
+endif()
+
 option(FAIL_ON_WARNINGS "Treat compile warnings as errors" ON)
 if(FAIL_ON_WARNINGS)
   if(MSVC)
@@ -466,11 +505,6 @@ if(CMAKE_SYSTEM_NAME MATCHES "Cygwin")
   add_definitions(-fno-builtin-memcmp -DCYGWIN)
 elseif(CMAKE_SYSTEM_NAME MATCHES "Darwin")
   add_definitions(-DOS_MACOSX)
-  if(CMAKE_SYSTEM_PROCESSOR MATCHES arm)
-    add_definitions(-DIOS_CROSS_COMPILE -DROCKSDB_LITE)
-    # no debug info for IOS, that will make our library big
-    add_definitions(-DNDEBUG)
-  endif()
 elseif(CMAKE_SYSTEM_NAME MATCHES "Linux")
   add_definitions(-DOS_LINUX)
 elseif(CMAKE_SYSTEM_NAME MATCHES "SunOS")
@@ -571,6 +605,7 @@ set(SOURCES
         db/arena_wrapped_db_iter.cc
         db/blob/blob_file_addition.cc
         db/blob/blob_file_builder.cc
+        db/blob/blob_file_cache.cc
         db/blob/blob_file_garbage.cc
         db/blob/blob_file_meta.cc
         db/blob/blob_file_reader.cc
@@ -580,7 +615,6 @@ set(SOURCES
         db/builder.cc
         db/c.cc
         db/column_family.cc
-        db/compacted_db_impl.cc
         db/compaction/compaction.cc
         db/compaction/compaction_iterator.cc
         db/compaction/compaction_picker.cc
@@ -591,6 +625,7 @@ set(SOURCES
         db/compaction/sst_partitioner.cc
         db/convenience.cc
         db/db_filesnapshot.cc
+        db/db_impl/compacted_db_impl.cc
         db/db_impl/db_impl.cc
         db/db_impl/db_impl_write.cc
         db/db_impl/db_impl_compaction_flush.cc
@@ -641,17 +676,20 @@ set(SOURCES
         db/write_batch_base.cc
         db/write_controller.cc
         db/write_thread.cc
+        env/composite_env.cc
         env/env.cc
         env/env_chroot.cc
         env/env_encryption.cc
         env/env_hdfs.cc
         env/file_system.cc
         env/file_system_tracer.cc
+        env/fs_remap.cc
         env/mock_env.cc
         file/delete_scheduler.cc
         file/file_prefetch_buffer.cc
         file/file_util.cc
         file/filename.cc
+        file/line_file_reader.cc
         file/random_access_file_reader.cc
         file/read_write_util.cc
         file/readahead_raf.cc
@@ -686,6 +724,7 @@ set(SOURCES
         monitoring/thread_status_util_debug.cc
         options/cf_options.cc
         options/configurable.cc
+        options/customizable.cc
         options/db_options.cc
         options/options.cc
         options/options_helper.cc
@@ -764,6 +803,7 @@ set(SOURCES
         util/murmurhash.cc
         util/random.cc
         util/rate_limiter.cc
+        util/ribbon_config.cc
         util/slice.cc
         util/file_checksum_helper.cc
         util/status.cc
@@ -809,8 +849,11 @@ set(SOURCES
         utilities/simulator_cache/sim_cache.cc
         utilities/table_properties_collectors/compact_on_deletion_collector.cc
         utilities/trace/file_trace_reader_writer.cc
-        utilities/transactions/lock/lock_tracker.cc
-        utilities/transactions/lock/point_lock_tracker.cc
+        utilities/transactions/lock/lock_manager.cc
+        utilities/transactions/lock/point/point_lock_tracker.cc
+        utilities/transactions/lock/point/point_lock_manager.cc
+        utilities/transactions/lock/range/range_tree/range_tree_lock_manager.cc
+        utilities/transactions/lock/range/range_tree/range_tree_lock_tracker.cc
         utilities/transactions/optimistic_transaction_db_impl.cc
         utilities/transactions/optimistic_transaction.cc
         utilities/transactions/pessimistic_transaction.cc
@@ -818,7 +861,6 @@ set(SOURCES
         utilities/transactions/snapshot_checker.cc
         utilities/transactions/transaction_base.cc
         utilities/transactions/transaction_db_mutex_impl.cc
-        utilities/transactions/transaction_lock_mgr.cc
         utilities/transactions/transaction_util.cc
         utilities/transactions/write_prepared_txn.cc
         utilities/transactions/write_prepared_txn_db.cc
@@ -826,8 +868,21 @@ set(SOURCES
         utilities/transactions/write_unprepared_txn_db.cc
         utilities/ttl/db_ttl_impl.cc
         utilities/write_batch_with_index/write_batch_with_index.cc
-        utilities/write_batch_with_index/write_batch_with_index_internal.cc
-        $<TARGET_OBJECTS:build_version>)
+        utilities/write_batch_with_index/write_batch_with_index_internal.cc)
+
+list(APPEND SOURCES
+  utilities/transactions/lock/range/range_tree/lib/locktree/concurrent_tree.cc
+  utilities/transactions/lock/range/range_tree/lib/locktree/keyrange.cc
+  utilities/transactions/lock/range/range_tree/lib/locktree/lock_request.cc
+  utilities/transactions/lock/range/range_tree/lib/locktree/locktree.cc
+  utilities/transactions/lock/range/range_tree/lib/locktree/manager.cc
+  utilities/transactions/lock/range/range_tree/lib/locktree/range_buffer.cc
+  utilities/transactions/lock/range/range_tree/lib/locktree/treenode.cc
+  utilities/transactions/lock/range/range_tree/lib/locktree/txnid_set.cc
+  utilities/transactions/lock/range/range_tree/lib/locktree/wfg.cc
+  utilities/transactions/lock/range/range_tree/lib/standalone_port.cc
+  utilities/transactions/lock/range/range_tree/lib/util/dbt.cc
+  utilities/transactions/lock/range/range_tree/lib/util/memarena.cc)
 
 if(HAVE_SSE42 AND NOT MSVC)
   set_source_files_properties(
@@ -904,12 +959,12 @@ else()
   set(SYSTEM_LIBS ${CMAKE_THREAD_LIBS_INIT})
 endif()
 
-add_library(${ROCKSDB_STATIC_LIB} STATIC ${SOURCES})
+add_library(${ROCKSDB_STATIC_LIB} STATIC ${SOURCES} ${BUILD_VERSION_CC})
 target_link_libraries(${ROCKSDB_STATIC_LIB} PRIVATE
   ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
 
 if(ROCKSDB_BUILD_SHARED)
-  add_library(${ROCKSDB_SHARED_LIB} SHARED ${SOURCES})
+  add_library(${ROCKSDB_SHARED_LIB} SHARED ${SOURCES} ${BUILD_VERSION_CC})
   target_link_libraries(${ROCKSDB_SHARED_LIB} PRIVATE
     ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
 
@@ -927,7 +982,7 @@ if(ROCKSDB_BUILD_SHARED)
                           LINKER_LANGUAGE CXX
                           VERSION ${rocksdb_VERSION}
                           SOVERSION ${rocksdb_VERSION_MAJOR}
-                          OUTPUT_NAME "rocksdb")
+                          OUTPUT_NAME "rocksdb${ARTIFACT_SUFFIX}")
   endif()
 endif()
 
@@ -1044,8 +1099,12 @@ if(WITH_TESTS)
         cache/lru_cache_test.cc
         db/blob/blob_file_addition_test.cc
         db/blob/blob_file_builder_test.cc
+        db/blob/blob_file_cache_test.cc
         db/blob/blob_file_garbage_test.cc
         db/blob/blob_file_reader_test.cc
+        db/blob/db_blob_basic_test.cc
+        db/blob/db_blob_compaction_test.cc
+        db/blob/db_blob_corruption_test.cc
         db/blob/db_blob_index_test.cc
         db/column_family_test.cc
         db/compact_files_test.cc
@@ -1068,6 +1127,7 @@ if(WITH_TESTS)
         db/db_iter_test.cc
         db/db_iter_stress_test.cc
         db/db_iterator_test.cc
+        db/db_kv_checksum_test.cc
         db/db_log_iter_test.cc
         db/db_memtable_test.cc
         db/db_merge_operator_test.cc
@@ -1075,7 +1135,7 @@ if(WITH_TESTS)
         db/db_options_test.cc
         db/db_properties_test.cc
         db/db_range_del_test.cc
-        db/db_impl/db_secondary_test.cc
+        db/db_secondary_test.cc
         db/db_sst_test.cc
         db/db_statistics_test.cc
         db/db_table_properties_test.cc
@@ -1139,6 +1199,7 @@ if(WITH_TESTS)
         monitoring/statistics_test.cc
         monitoring/stats_history_test.cc
         options/configurable_test.cc
+        options/customizable_test.cc
         options/options_settable_test.cc
         options/options_test.cc
         table/block_based/block_based_filter_block_test.cc
@@ -1155,6 +1216,7 @@ if(WITH_TESTS)
         table/table_test.cc
         table/block_fetcher_test.cc
         test_util/testutil_test.cc
+        trace_replay/block_cache_tracer_test.cc
         trace_replay/io_tracer_test.cc
         tools/block_cache_analyzer/block_cache_trace_analyzer_test.cc
         tools/io_tracer_parser_test.cc
@@ -1175,6 +1237,7 @@ if(WITH_TESTS)
         util/random_test.cc
         util/rate_limiter_test.cc
         util/repeatable_thread_test.cc
+        util/ribbon_test.cc
         util/slice_test.cc
         util/slice_transform_test.cc
         util/timer_queue_test.cc
@@ -1201,9 +1264,10 @@ if(WITH_TESTS)
         utilities/table_properties_collectors/compact_on_deletion_collector_test.cc
         utilities/transactions/optimistic_transaction_test.cc
         utilities/transactions/transaction_test.cc
-        utilities/transactions/transaction_lock_mgr_test.cc
+        utilities/transactions/lock/point/point_lock_manager_test.cc
         utilities/transactions/write_prepared_transaction_test.cc
         utilities/transactions/write_unprepared_transaction_test.cc
+        utilities/transactions/lock/range/range_locking_test.cc
         utilities/ttl/ttl_test.cc
         utilities/write_batch_with_index/write_batch_with_index_test.cc
     )
@@ -1277,40 +1341,41 @@ if(WITH_TESTS)
 endif()
 
 if(WITH_BENCHMARK_TOOLS)
-  add_executable(db_bench
+  add_executable(db_bench${ARTIFACT_SUFFIX}
     tools/db_bench.cc
     tools/db_bench_tool.cc)
-  target_link_libraries(db_bench
+  target_link_libraries(db_bench${ARTIFACT_SUFFIX}
     ${ROCKSDB_LIB} ${THIRDPARTY_LIBS})
 
-  add_executable(cache_bench
-    cache/cache_bench.cc)
-  target_link_libraries(cache_bench
+  add_executable(cache_bench${ARTIFACT_SUFFIX}
+    cache/cache_bench.cc
+    cache/cache_bench_tool.cc)
+  target_link_libraries(cache_bench${ARTIFACT_SUFFIX}
     ${ROCKSDB_LIB} ${GFLAGS_LIB})
 
-  add_executable(memtablerep_bench
+  add_executable(memtablerep_bench${ARTIFACT_SUFFIX}
     memtable/memtablerep_bench.cc)
-  target_link_libraries(memtablerep_bench
+  target_link_libraries(memtablerep_bench${ARTIFACT_SUFFIX}
     ${ROCKSDB_LIB} ${GFLAGS_LIB})
 
-  add_executable(range_del_aggregator_bench
+  add_executable(range_del_aggregator_bench${ARTIFACT_SUFFIX}
     db/range_del_aggregator_bench.cc)
-  target_link_libraries(range_del_aggregator_bench
+  target_link_libraries(range_del_aggregator_bench${ARTIFACT_SUFFIX}
     ${ROCKSDB_LIB} ${GFLAGS_LIB})
 
-  add_executable(table_reader_bench
+  add_executable(table_reader_bench${ARTIFACT_SUFFIX}
     table/table_reader_bench.cc)
-  target_link_libraries(table_reader_bench
+  target_link_libraries(table_reader_bench${ARTIFACT_SUFFIX}
     ${ROCKSDB_LIB} testharness ${GFLAGS_LIB})
 
-  add_executable(filter_bench
+  add_executable(filter_bench${ARTIFACT_SUFFIX}
     util/filter_bench.cc)
-  target_link_libraries(filter_bench
+  target_link_libraries(filter_bench${ARTIFACT_SUFFIX}
     ${ROCKSDB_LIB} ${GFLAGS_LIB})
 
-  add_executable(hash_table_bench
+  add_executable(hash_table_bench${ARTIFACT_SUFFIX}
     utilities/persistent_cache/hash_table_bench.cc)
-  target_link_libraries(hash_table_bench
+  target_link_libraries(hash_table_bench${ARTIFACT_SUFFIX}
     ${ROCKSDB_LIB} ${GFLAGS_LIB})
 endif()
 

HISTORY.md

@@ -1,13 +1,169 @@
 # Rocksdb Change Log
 ## Unreleased
 ### Bug Fixes
+* Fixed a bug in handling file rename error in distributed/network file systems when the server succeeds but client returns error. The bug can cause CURRENT file to point to non-existing MANIFEST file, thus DB cannot be opened.
+* Fixed a bug where ingested files were written with incorrect boundary key metadata. In rare cases this could have led to a level's files being wrongly ordered and queries for the boundary keys returning wrong results.
+* Fixed a data race between insertion into memtables and the retrieval of the DB properties `rocksdb.cur-size-active-mem-table`, `rocksdb.cur-size-all-mem-tables`, and `rocksdb.size-all-mem-tables`.
+* Fixed the false-positive alert when recovering from the WAL file. Avoid reporting "SST file is ahead of WAL" on a newly created empty column family, if the previous WAL file is corrupted.
+
+### Behavior Changes
+* Due to the fix of false-postive alert of "SST file is ahead of WAL", all the CFs with no SST file (CF empty) will bypass the consistency check. We fixed a false-positive, but introduced a very rare true-negative which will be triggered in the following conditions: A CF with some delete operations in the last a few queries which will result in an empty CF (those are flushed to SST file and a compaction triggered which combines this file and all other SST files and generates an empty CF, or there is another reason to write a manifest entry for this CF after a flush that generates no SST file from an empty CF). The deletion entries are logged in a WAL and this WAL was corrupted, while the CF's log number points to the next WAL (due to the flush). Therefore, the DB can only recover to the point without these trailing deletions and cause the inconsistent DB status.
+
+### New Features
+* Add new option allow_stall passed during instance creation of WriteBufferManager. When allow_stall is set, WriteBufferManager will stall all writers shared across multiple DBs and columns if memory usage goes beyond specified WriteBufferManager::buffer_size (soft limit). Stall will be cleared when memory is freed after flush and memory usage goes down below buffer_size.
+
+## 6.20.0 (04/16/2021)
+### Behavior Changes
+* `ColumnFamilyOptions::sample_for_compression` now takes effect for creation of all block-based tables. Previously it only took effect for block-based tables created by flush.
+* `CompactFiles()` can no longer compact files from lower level to up level, which has the risk to corrupt DB (details: #8063). The validation is also added to all compactions.
+* Fixed some cases in which DB::OpenForReadOnly() could write to the filesystem. If you want a Logger with a read-only DB, you must now set DBOptions::info_log yourself, such as using CreateLoggerFromOptions().
+* get_iostats_context() will never return nullptr. If thread-local support is not available, and user does not opt-out iostats context, then compilation will fail. The same applies to perf context as well.
+
+### Bug Fixes
+* Use thread-safe `strerror_r()` to get error messages.
+* Fixed a potential hang in shutdown for a DB whose `Env` has high-pri thread pool disabled (`Env::GetBackgroundThreads(Env::Priority::HIGH) == 0`)
+* Made BackupEngine thread-safe and added documentation comments to clarify what is safe for multiple BackupEngine objects accessing the same backup directory.
+* Fixed crash (divide by zero) when compression dictionary is applied to a file containing only range tombstones.
+* Fixed a backward iteration bug with partitioned filter enabled: not including the prefix of the last key of the previous filter partition in current filter partition can cause wrong iteration result.
+* Fixed a bug that allowed `DBOptions::max_open_files` to be set with a non-negative integer with `ColumnFamilyOptions::compaction_style = kCompactionStyleFIFO`.
+
+### Performance Improvements
+* On ARM platform, use `yield` instead of `wfe` to relax cpu to gain better performance.
+
+### Public API change
+* Added `TableProperties::slow_compression_estimated_data_size` and `TableProperties::fast_compression_estimated_data_size`. When `ColumnFamilyOptions::sample_for_compression > 0`, they estimate what `TableProperties::data_size` would have been if the "fast" or "slow" (see `ColumnFamilyOptions::sample_for_compression` API doc for definitions) compression had been used instead.
+* Update DB::StartIOTrace and remove Env object from the arguments as its redundant and DB already has Env object that is passed down to IOTracer::StartIOTrace
+* Added `FlushReason::kWalFull`, which is reported when a memtable is flushed due to the WAL reaching its size limit; those flushes were previously reported as `FlushReason::kWriteBufferManager`. Also, changed the reason for flushes triggered by the write buffer manager to `FlushReason::kWriteBufferManager`; they were previously reported as `FlushReason::kWriteBufferFull`.
+* Extend file_checksum_dump ldb command and DB::GetLiveFilesChecksumInfo API for IntegratedBlobDB and get checksum of blob files along with SST files.
+
+### New Features
+* Added the ability to open BackupEngine backups as read-only DBs, using BackupInfo::name_for_open and env_for_open provided by BackupEngine::GetBackupInfo() with include_file_details=true.
+* Added BackupEngine support for integrated BlobDB, with blob files shared between backups when table files are shared. Because of current limitations, blob files always use the kLegacyCrc32cAndFileSize naming scheme, and incremental backups must read and checksum all blob files in a DB, even for files that are already backed up.
+* Added an optional output parameter to BackupEngine::CreateNewBackup(WithMetadata) to return the BackupID of the new backup.
+* Added BackupEngine::GetBackupInfo / GetLatestBackupInfo for querying individual backups.
+* Made the Ribbon filter a long-term supported feature in terms of the SST schema(compatible with version >= 6.15.0) though the API for enabling it is expected to change.
+
+
+## 6.19.0 (03/21/2021)
+### Bug Fixes
+* Fixed the truncation error found in APIs/tools when dumping block-based SST files in a human-readable format. After fix, the block-based table can be fully dumped as a readable file.
+* When hitting a write slowdown condition, no write delay (previously 1 millisecond) is imposed until `delayed_write_rate` is actually exceeded, with an initial burst allowance of 1 millisecond worth of bytes. Also, beyond the initial burst allowance, `delayed_write_rate` is now more strictly enforced, especially with multiple column families.
+
+### Public API change
+* Changed default `BackupableDBOptions::share_files_with_checksum` to `true` and deprecated `false` because of potential for data loss. Note that accepting this change in behavior can temporarily increase backup data usage because files are not shared between backups using the two different settings. Also removed obsolete option kFlagMatchInterimNaming.
+* Add a new option BlockBasedTableOptions::max_auto_readahead_size. RocksDB does auto-readahead for iterators on noticing more than two reads for a table file if user doesn't provide readahead_size. The readahead starts at 8KB and doubles on every additional read upto max_auto_readahead_size and now max_auto_readahead_size can be configured dynamically as well. Found that 256 KB readahead size provides the best performance, based on experiments, for auto readahead. Experiment data is in PR #3282. If value is set 0 then no automatic prefetching will be done by rocksdb. Also changing the value will only affect files opened after the change.
+* Add suppport to extend DB::VerifyFileChecksums API to also verify blob files checksum.
+* When using the new BlobDB, the amount of data written by flushes/compactions is now broken down into table files and blob files in the compaction statistics; namely, Write(GB) denotes the amount of data written to table files, while Wblob(GB) means the amount of data written to blob files.
+* New default BlockBasedTableOptions::format_version=5 to enable new Bloom filter implementation by default, compatible with RocksDB versions >= 6.6.0.
+* Add new SetBufferSize API to WriteBufferManager to allow dynamic management of memory allotted to all write buffers.  This allows user code to adjust memory monitoring provided by WriteBufferManager as process memory needs change datasets grow and shrink.
+* Clarified the required semantics of Read() functions in FileSystem and Env APIs. Please ensure any custom implementations are compliant.
+* For the new integrated BlobDB implementation, compaction statistics now include the amount of data read from blob files during compaction (due to garbage collection or compaction filters). Write amplification metrics have also been extended to account for data read from blob files.
+* Add EqualWithoutTimestamp() to Comparator.
+* Extend support to track blob files in SSTFileManager whenever a blob file is created/deleted. Blob files will be scheduled to delete via SSTFileManager and SStFileManager will now take blob files in account while calculating size and space limits along with SST files.
+* Add new Append and PositionedAppend API with checksum handoff to legacy Env.
+
+### New Features
+* Support compaction filters for the new implementation of BlobDB. Add `FilterBlobByKey()` to `CompactionFilter`. Subclasses can override this method so that compaction filters can determine whether the actual blob value has to be read during compaction. Use a new `kUndetermined` in `CompactionFilter::Decision` to indicated that further action is necessary for compaction filter to make a decision.
+* Add support to extend retrieval of checksums for blob files from the MANIFEST when checkpointing. During backup, rocksdb can detect corruption in blob files  during file copies.
+* Add new options for db_bench --benchmarks: flush, waitforcompaction, compact0, compact1.
+* Add an option to BackupEngine::GetBackupInfo to include the name and size of each backed-up file. Especially in the presence of file sharing among backups, this offers detailed insight into backup space usage.
+* Enable backward iteration on keys with user-defined timestamps.
+* Add statistics and info log for error handler: counters for bg error, bg io error, bg retryable io error, auto resume count, auto resume total retry number, and auto resume sucess; Histogram for auto resume retry count in each recovery call. Note that, each auto resume attempt will have one or multiple retries.
+
+### Behavior Changes
+* During flush, only WAL sync retryable IO error is mapped to hard error, which will stall the writes. When WAL is used but only SST file write has retryable IO error, it will be mapped to soft error and write will not be affected.
+
+## 6.18.0 (02/19/2021)
+### Behavior Changes
+* When retryable IO error occurs during compaction, it is mapped to soft error and set the BG error. However, auto resume is not called to clean the soft error since compaction will reschedule by itself. In this change, When retryable IO error occurs during compaction, BG error is not set. User will be informed the error via EventHelper.
+* Introduce a new trace file format for query tracing and replay and trace file version is bump up to 0.2. A payload map is added as the first portion of the payload. We will not have backward compatible issues when adding new entries to trace records. Added the iterator_upper_bound and iterator_lower_bound in Seek and SeekForPrev tracing function. Added them as the new payload member for iterator tracing.
+
+### New Features
+* Add support for key-value integrity protection in live updates from the user buffers provided to `WriteBatch` through the write to RocksDB's in-memory update buffer (memtable). This is intended to detect some cases of in-memory data corruption, due to either software or hardware errors. Users can enable protection by constructing their `WriteBatch` with `protection_bytes_per_key == 8`.
+* Add support for updating `full_history_ts_low` option in manual compaction, which is for old timestamp data GC.
+* Add a mechanism for using Makefile to build external plugin code into the RocksDB libraries/binaries. This intends to simplify compatibility and distribution for plugins (e.g., special-purpose `FileSystem`s) whose source code resides outside the RocksDB repo. See "plugin/README.md" for developer details, and "PLUGINS.md" for a listing of available plugins.
+* Added memory pre-fetching for experimental Ribbon filter, which especially optimizes performance with batched MultiGet.
+* A new, experimental version of BlobDB (key-value separation) is now available. The new implementation is integrated into the RocksDB core, i.e. it is accessible via the usual `rocksdb::DB` API, as opposed to the separate `rocksdb::blob_db::BlobDB` interface used by the earlier version, and can be configured on a per-column family basis using the configuration options `enable_blob_files`, `min_blob_size`, `blob_file_size`, `blob_compression_type`, `enable_blob_garbage_collection`, and `blob_garbage_collection_age_cutoff`. It extends RocksDB's consistency guarantees to blobs, and offers more features and better performance. Note that some features, most notably `Merge`, compaction filters, and backup/restore are not yet supported, and there is no support for migrating a database created by the old implementation.
+
+### Bug Fixes
+* Since 6.15.0, `TransactionDB` returns error `Status`es from calls to `DeleteRange()` and calls to `Write()` where the `WriteBatch` contains a range deletion. Previously such operations may have succeeded while not providing the expected transactional guarantees. There are certain cases where range deletion can still be used on such DBs; see the API doc on `TransactionDB::DeleteRange()` for details.
+* `OptimisticTransactionDB` now returns error `Status`es from calls to `DeleteRange()` and calls to `Write()` where the `WriteBatch` contains a range deletion. Previously such operations may have succeeded while not providing the expected transactional guarantees.
+* Fix `WRITE_PREPARED`, `WRITE_UNPREPARED` TransactionDB `MultiGet()` may return uncommitted data with snapshot.
+* In DB::OpenForReadOnly, if any error happens while checking Manifest file path, it was overridden by Status::NotFound. It has been fixed and now actual error is returned.
+
+### Public API Change
+* Added a "only_mutable_options" flag to the ConfigOptions.  When this flag is "true", the Configurable functions and convenience methods (such as GetDBOptionsFromString) will only deal with options that are marked as mutable.  When this flag is true, only options marked as mutable can be configured (a Status::InvalidArgument will be returned) and options not marked as mutable will not be returned or compared.  The default is "false", meaning to compare all options.
+* Add new Append and PositionedAppend APIs to FileSystem to bring the data verification information (data checksum information) from upper layer (e.g., WritableFileWriter) to the storage layer. In this way, the customized FileSystem is able to verify the correctness of data being written to the storage on time. Add checksum_handoff_file_types to DBOptions. User can use this option to control which file types (Currently supported file tyes: kWALFile, kTableFile, kDescriptorFile.) should use the new Append and PositionedAppend APIs to handoff the verification information. Currently, RocksDB only use crc32c to calculate the checksum for write handoff.
+* Add an option, `CompressionOptions::max_dict_buffer_bytes`, to limit the in-memory buffering for selecting samples for generating/training a dictionary. The limit is currently loosely adhered to.
+
+
+## 6.17.0 (01/15/2021)
+### Behavior Changes
+* When verifying full file checksum with `DB::VerifyFileChecksums()`, we now fail with `Status::InvalidArgument` if the name of the checksum generator used for verification does not match the name of the checksum generator used for protecting the file when it was created.
+* Since RocksDB does not continue write the same file if a file write fails for any reason, the file scope write IO error is treated the same as retryable IO error. More information about error handling of file scope IO error is included in `ErrorHandler::SetBGError`.
+
+### Bug Fixes
+* Version older than 6.15 cannot decode VersionEdits `WalAddition` and `WalDeletion`, fixed this by changing the encoded format of them to be ignorable by older versions.
+* Fix a race condition between DB startups and shutdowns in managing the periodic background worker threads. One effect of this race condition could be the process being terminated.
+
+### Public API Change
+* Add a public API WriteBufferManager::dummy_entries_in_cache_usage() which reports the size of dummy entries stored in cache (passed to WriteBufferManager). Dummy entries are used to account for DataBlocks.
+* Add a SystemClock class that contains the time-related methods from Env.  The original methods in Env may be deprecated in a future release.  This class will allow easier testing, development, and expansion of time-related features.
+* Add a public API GetRocksBuildProperties and GetRocksBuildInfoAsString to get properties about the current build.  These properties may include settings related to the GIT settings (branch, timestamp).  This change also sets the "build date" based on the GIT properties, rather than the actual build time, thereby enabling more reproducible builds.
+
+## 6.16.0 (12/18/2020)
+### Behavior Changes
+* Attempting to write a merge operand without explicitly configuring `merge_operator` now fails immediately, causing the DB to enter read-only mode. Previously, failure was deferred until the `merge_operator` was needed by a user read or a background operation.
+
+### Bug Fixes
+* Truncated WALs ending in incomplete records can no longer produce gaps in the recovered data when `WALRecoveryMode::kPointInTimeRecovery` is used. Gaps are still possible when WALs are truncated exactly on record boundaries; for complete protection, users should enable `track_and_verify_wals_in_manifest`.
+* Fix a bug where compressed blocks read by MultiGet are not inserted into the compressed block cache when use_direct_reads = true.
+* Fixed the issue of full scanning on obsolete files when there are too many outstanding compactions with ConcurrentTaskLimiter enabled.
+* Fixed the logic of populating native data structure for `read_amp_bytes_per_bit` during OPTIONS file parsing on big-endian architecture. Without this fix, original code introduced in PR7659, when running on big-endian machine, can mistakenly store read_amp_bytes_per_bit (an uint32) in little endian format. Future access to `read_amp_bytes_per_bit` will give wrong values. Little endian architecture is not affected.
+* Fixed prefix extractor with timestamp issues.
+* Fixed a bug in atomic flush: in two-phase commit mode, the minimum WAL log number to keep is incorrect.
+* Fixed a bug related to checkpoint in PR7789: if there are multiple column families, and the checkpoint is not opened as read only, then in rare cases, data loss may happen in the checkpoint. Since backup engine relies on checkpoint, it may also be affected.
+* When ldb --try_load_options is used with the --column_family option, the ColumnFamilyOptions for the specified column family was not loaded from the OPTIONS file. Fix it so its loaded from OPTIONS and then overridden with command line overrides.
+
+### New Features
+* User defined timestamp feature supports `CompactRange` and `GetApproximateSizes`.
+* Support getting aggregated table properties (kAggregatedTableProperties and kAggregatedTablePropertiesAtLevel) with DB::GetMapProperty, for easier access to the data in a structured format.
+* Experimental option BlockBasedTableOptions::optimize_filters_for_memory now works with experimental Ribbon filter (as well as Bloom filter).
+
+### Public API Change
+* Deprecated public but rarely-used FilterBitsBuilder::CalculateNumEntry, which is replaced with ApproximateNumEntries taking a size_t parameter and returning size_t.
+* To improve portability the functions `Env::GetChildren` and `Env::GetChildrenFileAttributes` will no longer return entries for the special directories `.` or `..`.
+* Added a new option `track_and_verify_wals_in_manifest`. If `true`, the log numbers and sizes of the synced WALs are tracked in MANIFEST, then during DB recovery, if a synced WAL is missing from disk, or the WAL's size does not match the recorded size in MANIFEST, an error will be reported and the recovery will be aborted. Note that this option does not work with secondary instance.
+* `rocksdb_approximate_sizes` and `rocksdb_approximate_sizes_cf` in the C API now requires an error pointer (`char** errptr`) for receiving any error.
+* All overloads of DB::GetApproximateSizes now return Status, so that any failure to obtain the sizes is indicated to the caller.
+
+## 6.15.0 (11/13/2020)
+### Bug Fixes
 * Fixed a bug in the following combination of features: indexes with user keys (`format_version >= 3`), indexes are partitioned (`index_type == kTwoLevelIndexSearch`), and some index partitions are pinned in memory (`BlockBasedTableOptions::pin_l0_filter_and_index_blocks_in_cache`). The bug could cause keys to be truncated when read from the index leading to wrong read results or other unexpected behavior.
 * Fixed a bug when indexes are partitioned (`index_type == kTwoLevelIndexSearch`), some index partitions are pinned in memory (`BlockBasedTableOptions::pin_l0_filter_and_index_blocks_in_cache`), and partitions reads could be mixed between block cache and directly from the file (e.g., with `enable_index_compression == 1` and `mmap_read == 1`, partitions that were stored uncompressed due to poor compression ratio would be read directly from the file via mmap, while partitions that were stored compressed would be read from block cache). The bug could cause index partitions to be mistakenly considered empty during reads leading to wrong read results.
 * Since 6.12, memtable lookup should report unrecognized value_type as corruption (#7121).
 * Since 6.14, fix false positive flush/compaction `Status::Corruption` failure when `paranoid_file_checks == true` and range tombstones were written to the compaction output files.
+* Since 6.14, fix a bug that could cause a stalled write to crash with mixed of slowdown and no_slowdown writes (`WriteOptions.no_slowdown=true`).
+* Fixed a bug which causes hang in closing DB when refit level is set in opt build. It was because ContinueBackgroundWork() was called in assert statement which is a no op. It was introduced in 6.14.
+* Fixed a bug which causes Get() to return incorrect result when a key's merge operand is applied twice. This can occur if the thread performing Get() runs concurrently with a background flush thread and another thread writing to the MANIFEST file (PR6069).
+* Reverted a behavior change silently introduced in 6.14.2, in which the effects of the `ignore_unknown_options` flag (used in option parsing/loading functions) changed.
+* Reverted a behavior change silently introduced in 6.14, in which options parsing/loading functions began returning `NotFound` instead of `InvalidArgument` for option names not available in the present version.
+* Fixed MultiGet bugs it doesn't return valid data with user defined timestamp.
+* Fixed a potential bug caused by evaluating `TableBuilder::NeedCompact()` before `TableBuilder::Finish()` in compaction job. For example, the `NeedCompact()` method of `CompactOnDeletionCollector` returned by built-in `CompactOnDeletionCollectorFactory` requires `BlockBasedTable::Finish()` to return the correct result. The bug can cause a compaction-generated file not to be marked for future compaction based on deletion ratio.
+* Fixed a seek issue with prefix extractor and timestamp.
+* Fixed a bug of encoding and parsing BlockBasedTableOptions::read_amp_bytes_per_bit as a 64-bit integer.
+* Fixed a bug of a recovery corner case, details in PR7621.
 
 ### Public API Change
 * Deprecate `BlockBasedTableOptions::pin_l0_filter_and_index_blocks_in_cache` and `BlockBasedTableOptions::pin_top_level_index_and_filter`. These options still take effect until users migrate to the replacement APIs in `BlockBasedTableOptions::metadata_cache_options`. Migration guidance can be found in the API comments on the deprecated options.
+* Add new API `DB::VerifyFileChecksums` to verify SST file checksum with corresponding entries in the MANIFEST if present. Current implementation requires scanning and recomputing file checksums.
+
+### Behavior Changes
+* The dictionary compression settings specified in `ColumnFamilyOptions::compression_opts` now additionally affect files generated by flush and compaction to non-bottommost level. Previously those settings at most affected files generated by compaction to bottommost level, depending on whether `ColumnFamilyOptions::bottommost_compression_opts` overrode them. Users who relied on dictionary compression settings in `ColumnFamilyOptions::compression_opts` affecting only the bottommost level can keep the behavior by moving their dictionary settings to `ColumnFamilyOptions::bottommost_compression_opts` and setting its `enabled` flag.
+* When the `enabled` flag is set in `ColumnFamilyOptions::bottommost_compression_opts`, those compression options now take effect regardless of the value in `ColumnFamilyOptions::bottommost_compression`. Previously, those compression options only took effect when `ColumnFamilyOptions::bottommost_compression != kDisableCompressionOption`. Now, they additionally take effect when `ColumnFamilyOptions::bottommost_compression == kDisableCompressionOption` (such a setting causes bottommost compression type to fall back to `ColumnFamilyOptions::compression_per_level` if configured, and otherwise fall back to `ColumnFamilyOptions::compression`).
+
+### New Features
+* An EXPERIMENTAL new Bloom alternative that saves about 30% space compared to Bloom filters, with about 3-4x construction time and similar query times is available using NewExperimentalRibbonFilterPolicy.
 
 ## 6.14 (10/09/2020)
 ### Bug fixes
@@ -25,10 +181,13 @@
 * The settings of the DBOptions and ColumnFamilyOptions are now managed by Configurable objects (see New Features).  The same convenience methods to configure these options still exist but the backend implementation has been unified under a common implementation.
 
 ### New Features
+
 * Methods to configure serialize, and compare -- such as TableFactory -- are exposed directly through the Configurable base class (from which these objects inherit).  This change will allow for better and more thorough configuration management and retrieval in the future.  The options for a Configurable object can be set via the ConfigureFromMap, ConfigureFromString, or ConfigureOption method.  The serialized version of the options of an object can be retrieved via the GetOptionString, ToString, or GetOption methods.  The list of options supported by an object can be obtained via the GetOptionNames method.  The "raw" object (such as the BlockBasedTableOption) for an option may be retrieved via the GetOptions method.  Configurable options can be compared via the AreEquivalent method.  The settings within a Configurable object may be validated via the ValidateOptions method.  The object may be intialized (at which point only mutable options may be updated) via the PrepareOptions method.
 * Introduce options.check_flush_compaction_key_order with default value to be true. With this option, during flush and compaction, key order will be checked when writing to each SST file. If the order is violated, the flush or compaction will fail.
 * Added is_full_compaction to CompactionJobStats, so that the information is available through the EventListener interface.
 * Add more stats for MultiGet in Histogram to get number of data blocks, index blocks, filter blocks and sst files read from file system per level.
+* SST files have a new table property called db_host_id, which is set to the hostname by default. A new option in DBOptions, db_host_id, allows the property value to be overridden with a user specified string, or disable it completely by making the option string empty.
+* Methods to create customizable extensions -- such as TableFactory -- are exposed directly through the Customizable base class (from which these objects inherit).  This change will allow these Customizable classes to be loaded and configured in a standard way (via CreateFromString).  More information on how to write and use Customizable classes is in the customizable.h header file.
 
 ## 6.13 (09/12/2020)
 ### Bug fixes

INSTALL.md

@@ -43,6 +43,8 @@ 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.
 
+* `make check` will also check code formatting, which requires [clang-format](https://clang.llvm.org/docs/ClangFormat.html)
+
 * If you wish to build the RocksJava static target, then cmake is required for building Snappy.
 
 ## Supported platforms
@@ -94,12 +96,21 @@ to build a portable binary, add `PORTABLE=1` before your make commands, like thi
               sudo yum install libasan
 
     * Install zstandard:
+        * With [EPEL](https://fedoraproject.org/wiki/EPEL):
 
-             wget https://github.com/facebook/zstd/archive/v1.1.3.tar.gz
-             mv v1.1.3.tar.gz zstd-1.1.3.tar.gz
-             tar zxvf zstd-1.1.3.tar.gz
-             cd zstd-1.1.3
-             make && sudo make install
+              sudo yum install libzstd-devel
+
+        * With CentOS 8:
+
+              sudo dnf install libzstd-devel
+
+        * From source:
+
+              wget https://github.com/facebook/zstd/archive/v1.1.3.tar.gz
+              mv v1.1.3.tar.gz zstd-1.1.3.tar.gz
+              tar zxvf zstd-1.1.3.tar.gz
+              cd zstd-1.1.3
+              make && sudo make install
 
 * **OS X**:
     * Install latest C++ compiler that supports C++ 11:

Makefile

@@ -55,60 +55,33 @@ DEBUG_LEVEL?=1
 # Set the default LIB_MODE to static
 LIB_MODE?=static
 
-ifeq ($(MAKECMDGOALS),dbg)
+# OBJ_DIR is where the object files reside.  Default to the current directory
+OBJ_DIR?=.
+
+# Check the MAKECMDGOALS to set the DEBUG_LEVEL and LIB_MODE appropriately
+
+ifneq ($(filter clean release install, $(MAKECMDGOALS)),)
+	DEBUG_LEVEL=0
+endif
+ifneq ($(filter dbg, $(MAKECMDGOALS)),)
 	DEBUG_LEVEL=2
-endif
-
-ifeq ($(MAKECMDGOALS),clean)
+else ifneq ($(filter shared_lib install-shared, $(MAKECMDGOALS)),)
 	DEBUG_LEVEL=0
-endif
-
-ifeq ($(MAKECMDGOALS),release)
-	DEBUG_LEVEL=0
-endif
-
-ifeq ($(MAKECMDGOALS),shared_lib)
 	LIB_MODE=shared
-	DEBUG_LEVEL=0
-endif
-
-ifeq ($(MAKECMDGOALS),install-shared)
-	LIB_MODE=shared
-	DEBUG_LEVEL=0
-endif
-
-ifeq ($(MAKECMDGOALS),static_lib)
+else ifneq ($(filter static_lib install-static, $(MAKECMDGOALS)),)
 	DEBUG_LEVEL=0
 	LIB_MODE=static
-endif
-
-ifeq ($(MAKECMDGOALS),install-static)
-	DEBUG_LEVEL=0
-	LIB_MODE=static
-endif
-
-ifeq ($(MAKECMDGOALS),install)
-	DEBUG_LEVEL=0
-endif
-
-
-ifneq ($(findstring jtest, $(MAKECMDGOALS)),)
+else ifneq ($(filter jtest rocksdbjava%, $(MAKECMDGOALS)),)
 	OBJ_DIR=jl
 	LIB_MODE=shared
-endif
-
-ifneq ($(findstring rocksdbjava, $(MAKECMDGOALS)),)
-	LIB_MODE=shared
-        ifneq ($(findstring rocksdbjavastatic, $(MAKECMDGOALS)),)
+	ifneq ($(findstring rocksdbjavastatic, $(MAKECMDGOALS)),)
 		OBJ_DIR=jls
-	        ifneq ($(DEBUG_LEVEL),2)
-	            DEBUG_LEVEL=0
-                endif
-                ifeq ($(MAKECMDGOALS),rocksdbjavastaticpublish)
-	            DEBUG_LEVEL=0
-                endif
-	else
-		OBJ_DIR=jl
+		ifneq ($(DEBUG_LEVEL),2)
+			DEBUG_LEVEL=0
+		endif
+		ifeq ($(MAKECMDGOALS),rocksdbjavastaticpublish)
+			DEBUG_LEVEL=0
+		endif
 	endif
 endif
 
@@ -190,6 +163,11 @@ else
 endif
 
 ifdef ASSERT_STATUS_CHECKED
+# For ASC, turn off constructor elision, preventing the case where a constructor returned
+# by a method may pass the ASC check if the status is checked in the inner method.  Forcing
+# the copy constructor to be invoked disables the optimization and will cause the calling method
+# to check the status in order to prevent an error from being raised.
+PLATFORM_CXXFLAGS += -fno-elide-constructors
 ifeq ($(filter -DROCKSDB_ASSERT_STATUS_CHECKED,$(OPT)),)
 	OPT += -DROCKSDB_ASSERT_STATUS_CHECKED
 endif
@@ -254,6 +232,8 @@ AM_SHARE = $(AM_V_CCLD) $(CXX) $(PLATFORM_SHARED_LDFLAGS)$@ -L. $(patsubst lib%.
 # Export some common variables that might have been passed as Make variables
 # instead of environment variables.
 dummy := $(shell (export ROCKSDB_ROOT="$(CURDIR)"; \
+                  export CXXFLAGS="$(EXTRA_CXXFLAGS)"; \
+                  export LDFLAGS="$(EXTRA_LDFLAGS)"; \
                   export COMPILE_WITH_ASAN="$(COMPILE_WITH_ASAN)"; \
                   export COMPILE_WITH_TSAN="$(COMPILE_WITH_TSAN)"; \
                   export COMPILE_WITH_UBSAN="$(COMPILE_WITH_UBSAN)"; \
@@ -264,6 +244,12 @@ dummy := $(shell (export ROCKSDB_ROOT="$(CURDIR)"; \
 # this file is generated by the previous line to set build flags and sources
 include make_config.mk
 
+ROCKSDB_PLUGIN_MKS = $(foreach plugin, $(ROCKSDB_PLUGINS), plugin/$(plugin)/*.mk)
+include $(ROCKSDB_PLUGIN_MKS)
+ROCKSDB_PLUGIN_SOURCES = $(foreach plugin, $(ROCKSDB_PLUGINS), $(foreach source, $($(plugin)_SOURCES), plugin/$(plugin)/$(source)))
+ROCKSDB_PLUGIN_HEADERS = $(foreach plugin, $(ROCKSDB_PLUGINS), $(foreach header, $($(plugin)_HEADERS), plugin/$(plugin)/$(header)))
+PLATFORM_LDFLAGS += $(foreach plugin, $(ROCKSDB_PLUGINS), $($(plugin)_LDFLAGS))
+
 export JAVAC_ARGS
 CLEAN_FILES += make_config.mk rocksdb.pc
 
@@ -414,6 +400,10 @@ ifdef TEST_UINT128_COMPAT
   PLATFORM_CCFLAGS += -DTEST_UINT128_COMPAT=1
   PLATFORM_CXXFLAGS += -DTEST_UINT128_COMPAT=1
 endif
+ifdef ROCKSDB_MODIFY_NPHASH
+  PLATFORM_CCFLAGS += -DROCKSDB_MODIFY_NPHASH=1
+  PLATFORM_CXXFLAGS += -DROCKSDB_MODIFY_NPHASH=1
+endif
 
 # This (the first rule) must depend on "all".
 default: all
@@ -421,6 +411,10 @@ default: all
 WARNING_FLAGS = -W -Wextra -Wall -Wsign-compare -Wshadow \
   -Wunused-parameter
 
+ifeq (,$(filter amd64, $(MACHINE)))
+	C_WARNING_FLAGS = -Wstrict-prototypes
+endif
+
 ifdef USE_CLANG
 	# Used by some teams in Facebook
 	WARNING_FLAGS += -Wshift-sign-overflow
@@ -465,38 +459,13 @@ ifeq ($(NO_THREEWAY_CRC32C), 1)
 	CXXFLAGS += -DNO_THREEWAY_CRC32C
 endif
 
-CFLAGS += $(WARNING_FLAGS) -I. -I./include $(PLATFORM_CCFLAGS) $(OPT)
+CFLAGS += $(C_WARNING_FLAGS) $(WARNING_FLAGS) -I. -I./include $(PLATFORM_CCFLAGS) $(OPT)
 CXXFLAGS += $(WARNING_FLAGS) -I. -I./include $(PLATFORM_CXXFLAGS) $(OPT) -Woverloaded-virtual -Wnon-virtual-dtor -Wno-missing-field-initializers
 
 LDFLAGS += $(PLATFORM_LDFLAGS)
 
-# If NO_UPDATE_BUILD_VERSION is set we don't update util/build_version.cc, but
-# the file needs to already exist or else the build will fail
-ifndef NO_UPDATE_BUILD_VERSION
-date := $(shell date +%F)
-ifdef FORCE_GIT_SHA
-	git_sha := $(FORCE_GIT_SHA)
-else
-	git_sha := $(shell git rev-parse HEAD 2>/dev/null)
-endif
-gen_build_version = sed -e s/@@GIT_SHA@@/$(git_sha)/ -e s/@@GIT_DATE_TIME@@/$(date)/ util/build_version.cc.in
-
-# Record the version of the source that we are compiling.
-# We keep a record of the git revision in this file.  It is then built
-# as a regular source file as part of the compilation process.
-# One can run "strings executable_filename | grep _build_" to find
-# the version of the source that we used to build the executable file.
-FORCE:
-util/build_version.cc: FORCE
-	$(AM_V_GEN)rm -f $@-t
-	$(AM_V_at)$(gen_build_version) > $@-t
-	$(AM_V_at)if test -f $@; then					\
-	  cmp -s $@-t $@ && rm -f $@-t || mv -f $@-t $@;		\
-	else mv -f $@-t $@; fi
-endif
-
-OBJ_DIR?=.
 LIB_OBJECTS = $(patsubst %.cc, $(OBJ_DIR)/%.o, $(LIB_SOURCES))
+LIB_OBJECTS += $(patsubst %.cc, $(OBJ_DIR)/%.o, $(ROCKSDB_PLUGIN_SOURCES))
 ifeq ($(HAVE_POWER8),1)
 LIB_OBJECTS += $(patsubst %.c, $(OBJ_DIR)/%.o, $(LIB_SOURCES_C))
 LIB_OBJECTS += $(patsubst %.S, $(OBJ_DIR)/%.o, $(LIB_SOURCES_ASM))
@@ -506,6 +475,12 @@ ifeq ($(USE_FOLLY_DISTRIBUTED_MUTEX),1)
   LIB_OBJECTS += $(patsubst %.cpp, $(OBJ_DIR)/%.o, $(FOLLY_SOURCES))
 endif
 
+# range_tree is not compatible with non GNU libc on ppc64
+# see https://jira.percona.com/browse/PS-7559
+ifneq ($(PPC_LIBC_IS_GNU),0)
+  LIB_OBJECTS += $(patsubst %.cc, $(OBJ_DIR)/%.o, $(RANGE_TREE_SOURCES))
+endif
+
 GTEST = $(OBJ_DIR)/$(GTEST_DIR)/gtest/gtest-all.o
 TESTUTIL = $(OBJ_DIR)/test_util/testutil.o
 TESTHARNESS = $(OBJ_DIR)/test_util/testharness.o $(TESTUTIL) $(GTEST)
@@ -516,12 +491,14 @@ VALGRIND_OPTS = --error-exitcode=$(VALGRIND_ERROR) --leak-check=full
 
 TEST_OBJECTS = $(patsubst %.cc, $(OBJ_DIR)/%.o, $(TEST_LIB_SOURCES) $(MOCK_LIB_SOURCES)) $(GTEST)
 BENCH_OBJECTS = $(patsubst %.cc, $(OBJ_DIR)/%.o, $(BENCH_LIB_SOURCES))
+CACHE_BENCH_OBJECTS = $(patsubst %.cc, $(OBJ_DIR)/%.o, $(CACHE_BENCH_LIB_SOURCES))
 TOOL_OBJECTS = $(patsubst %.cc, $(OBJ_DIR)/%.o, $(TOOL_LIB_SOURCES))
 ANALYZE_OBJECTS = $(patsubst %.cc, $(OBJ_DIR)/%.o, $(ANALYZER_LIB_SOURCES))
 STRESS_OBJECTS =  $(patsubst %.cc, $(OBJ_DIR)/%.o, $(STRESS_LIB_SOURCES))
 
+# Exclude build_version.cc -- a generated source file -- from all sources.  Not needed for dependencies
 ALL_SOURCES  = $(LIB_SOURCES) $(TEST_LIB_SOURCES) $(MOCK_LIB_SOURCES) $(GTEST_DIR)/gtest/gtest-all.cc
-ALL_SOURCES += $(TOOL_LIB_SOURCES) $(BENCH_LIB_SOURCES) $(ANALYZER_LIB_SOURCES) $(STRESS_LIB_SOURCES)
+ALL_SOURCES += $(TOOL_LIB_SOURCES) $(BENCH_LIB_SOURCES) $(CACHE_BENCH_LIB_SOURCES) $(ANALYZER_LIB_SOURCES) $(STRESS_LIB_SOURCES)
 ALL_SOURCES += $(TEST_MAIN_SOURCES) $(TOOL_MAIN_SOURCES) $(BENCH_MAIN_SOURCES)
 
 TESTS = $(patsubst %.cc, %, $(notdir $(TEST_MAIN_SOURCES)))
@@ -532,161 +509,38 @@ ifeq ($(USE_FOLLY_DISTRIBUTED_MUTEX),1)
 	ALL_SOURCES += third-party/folly/folly/synchronization/test/DistributedMutexTest.cc
 endif
 
-PARALLEL_TEST = \
-	backupable_db_test \
-	db_bloom_filter_test \
-	db_compaction_filter_test \
-	db_compaction_test \
-	db_merge_operator_test \
-	db_sst_test \
-	db_test \
-	db_test2 \
-	db_universal_compaction_test \
-	db_wal_test \
-	column_family_test \
-	external_sst_file_test \
-	import_column_family_test \
-	fault_injection_test \
-	file_reader_writer_test \
-	inlineskiplist_test \
-	manual_compaction_test \
-	persistent_cache_test \
-	table_test \
-	transaction_test \
-	transaction_lock_mgr_test \
-	write_prepared_transaction_test \
-	write_unprepared_transaction_test \
-
-ifeq ($(USE_FOLLY_DISTRIBUTED_MUTEX),1)
-	TESTS += folly_synchronization_distributed_mutex_test
-	PARALLEL_TEST += folly_synchronization_distributed_mutex_test
-	TESTS_PASSING_ASC = folly_synchronization_distributed_mutex_test
-endif
-
 # options_settable_test doesn't pass with UBSAN as we use hack in the test
 ifdef COMPILE_WITH_UBSAN
         TESTS := $(shell echo $(TESTS) | sed 's/\boptions_settable_test\b//g')
 endif
 ifdef ASSERT_STATUS_CHECKED
-# This is a new check for which we will add support incrementally. This
-# list can be removed once support is fully added.
-	TESTS_PASSING_ASC = \
-		arena_test \
-		autovector_test \
-		cache_test \
-		lru_cache_test \
-		blob_file_addition_test \
-		blob_file_builder_test \
-		blob_file_garbage_test \
-		blob_file_reader_test \
-		bloom_test \
-		cassandra_format_test \
-		cassandra_row_merge_test \
-		cassandra_serialize_test \
-		cleanable_test \
-		coding_test \
-		crc32c_test \
-		dbformat_test \
-		db_basic_test \
-		db_flush_test \
-		db_with_timestamp_basic_test \
-		db_with_timestamp_compaction_test \
-		db_options_test \
-		db_properties_test \
-		db_secondary_test \
-		options_file_test \
-		defer_test \
-		filename_test \
-		dynamic_bloom_test \
-		env_basic_test \
-		env_test \
-		env_logger_test \
-		event_logger_test \
-		error_handler_fs_test \
-		auto_roll_logger_test \
-		file_indexer_test \
-		flush_job_test \
-		hash_table_test \
-		hash_test \
-		heap_test \
-		histogram_test \
-		inlineskiplist_test \
-		io_posix_test \
-		iostats_context_test \
-		ldb_cmd_test \
-		memkind_kmem_allocator_test \
-		merger_test \
-		mock_env_test \
-		object_registry_test \
-		prefix_test \
-		repair_test \
-		configurable_test \
-		options_settable_test \
-		options_test \
-		random_test \
-		range_del_aggregator_test \
-		sst_file_reader_test \
-		range_tombstone_fragmenter_test \
-		repeatable_thread_test \
-		skiplist_test \
-		slice_test \
-		sst_dump_test \
-		statistics_test \
-		stats_history_test \
-		thread_local_test \
-		trace_analyzer_test \
-		env_timed_test \
-		filelock_test \
-		timer_queue_test \
-		timer_test \
-		options_util_test \
-		persistent_cache_test \
-		util_merge_operators_test \
-		block_cache_trace_analyzer_test \
-		block_cache_tracer_test \
-		cache_simulator_test \
-		sim_cache_test \
-		version_builder_test \
-		version_edit_test \
-		work_queue_test \
-		write_controller_test \
-		compaction_iterator_test \
-		compaction_job_test \
-		compaction_job_stats_test \
-	        io_tracer_test \
-		merge_helper_test \
-		memtable_list_test \
-		flush_job_test \
-		block_based_filter_block_test \
-		block_fetcher_test \
-		full_filter_block_test \
-		partitioned_filter_block_test \
-		column_family_test \
-		file_reader_writer_test \
-		corruption_test \
-		db_universal_compaction_test \
-		import_column_family_test \
-		memory_test \
-		table_test \
+	# TODO: finish fixing all tests to pass this check
+	TESTS_FAILING_ASC = \
+		db_test \
+		db_test2 \
+		range_locking_test \
+		testutil_test \
 
-ifeq ($(USE_FOLLY_DISTRIBUTED_MUTEX),1)
-TESTS_PASSING_ASC += folly_synchronization_distributed_mutex_test
+	# Since we have very few ASC exclusions left, excluding them from
+	# the build is the most convenient way to exclude them from testing
+	TESTS := $(filter-out $(TESTS_FAILING_ASC),$(TESTS))
 endif
 
-	# Enable building all unit tests, but use check_some to run only tests
-	# known to pass ASC (ASSERT_STATUS_CHECKED)
-	ROCKSDBTESTS_SUBSET ?= $(TESTS_PASSING_ASC)
-	# Alternate: only build unit tests known to pass ASC, and run them
-	# with make check
-	#TESTS := $(filter $(TESTS_PASSING_ASC),$(TESTS))
-	#PARALLEL_TEST := $(filter $(TESTS_PASSING_ASC),$(PARALLEL_TEST))
-else
-	ROCKSDBTESTS_SUBSET ?= $(TESTS)
-endif
+ROCKSDBTESTS_SUBSET ?= $(TESTS)
+
+# env_test - suspicious use of test::TmpDir
+# deletefile_test - serial because it generates giant temporary files in
+#   its various tests. Parallel can fill up your /dev/shm
+NON_PARALLEL_TEST = \
+	env_test \
+	deletefile_test \
+
+PARALLEL_TEST = $(filter-out $(NON_PARALLEL_TEST), $(TESTS))
+
 # Not necessarily well thought out or up-to-date, but matches old list
 TESTS_PLATFORM_DEPENDENT := \
 	db_basic_test \
-	db_with_timestamp_basic_test \
+	db_blob_basic_test \
 	db_encryption_test \
 	db_test2 \
 	external_sst_file_basic_test \
@@ -704,6 +558,7 @@ TESTS_PLATFORM_DEPENDENT := \
 	io_posix_test \
 	hash_test \
 	random_test \
+	ribbon_test \
 	thread_local_test \
 	work_queue_test \
 	rate_limiter_test \
@@ -769,13 +624,44 @@ else
 LIBRARY=$(STATIC_LIBRARY)
 TEST_LIBRARY=$(STATIC_TEST_LIBRARY)
 TOOLS_LIBRARY=$(STATIC_TOOLS_LIBRARY)
-STRESS_LIBRARY=$(STATIC_STRESS_LIBRARY)
 endif
+STRESS_LIBRARY=$(STATIC_STRESS_LIBRARY)
 
 ROCKSDB_MAJOR = $(shell egrep "ROCKSDB_MAJOR.[0-9]" include/rocksdb/version.h | cut -d ' ' -f 3)
 ROCKSDB_MINOR = $(shell egrep "ROCKSDB_MINOR.[0-9]" include/rocksdb/version.h | cut -d ' ' -f 3)
 ROCKSDB_PATCH = $(shell egrep "ROCKSDB_PATCH.[0-9]" include/rocksdb/version.h | cut -d ' ' -f 3)
 
+# If NO_UPDATE_BUILD_VERSION is set we don't update util/build_version.cc, but
+# the file needs to already exist or else the build will fail
+ifndef NO_UPDATE_BUILD_VERSION
+
+# By default, use the current date-time as the date.  If there are no changes,
+# we will use the last commit date instead.
+build_date := $(shell date "+%Y-%m-%d %T")
+
+ifdef FORCE_GIT_SHA
+	git_sha := $(FORCE_GIT_SHA)
+	git_mod := 1
+	git_date := $(build_date)
+else
+	git_sha := $(shell git rev-parse HEAD 2>/dev/null)
+	git_tag  := $(shell git symbolic-ref -q --short HEAD 2> /dev/null || git describe --tags --exact-match 2>/dev/null)
+	git_mod  := $(shell git diff-index HEAD --quiet 2>/dev/null; echo $$?)
+	git_date := $(shell git log -1 --date=format:"%Y-%m-%d %T" --format="%ad" 2>/dev/null)
+endif
+gen_build_version = sed -e s/@GIT_SHA@/$(git_sha)/ -e s:@GIT_TAG@:"$(git_tag)": -e s/@GIT_MOD@/"$(git_mod)"/ -e s/@BUILD_DATE@/"$(build_date)"/ -e s/@GIT_DATE@/"$(git_date)"/ util/build_version.cc.in
+
+# Record the version of the source that we are compiling.
+# We keep a record of the git revision in this file.  It is then built
+# as a regular source file as part of the compilation process.
+# One can run "strings executable_filename | grep _build_" to find
+# the version of the source that we used to build the executable file.
+util/build_version.cc: $(filter-out $(OBJ_DIR)/util/build_version.o, $(LIB_OBJECTS)) util/build_version.cc.in
+	$(AM_V_GEN)rm -f $@-t
+	$(AM_V_at)$(gen_build_version) > $@
+endif
+CLEAN_FILES += util/build_version.cc
+
 default: all
 
 #-----------------------------------------------
@@ -823,7 +709,8 @@ endif  # PLATFORM_SHARED_EXT
 	analyze tools tools_lib \
 	blackbox_crash_test_with_atomic_flush whitebox_crash_test_with_atomic_flush  \
 	blackbox_crash_test_with_txn whitebox_crash_test_with_txn \
-	blackbox_crash_test_with_best_efforts_recovery
+	blackbox_crash_test_with_best_efforts_recovery \
+	blackbox_crash_test_with_ts whitebox_crash_test_with_ts
 
 
 all: $(LIBRARY) $(BENCHMARKS) tools tools_lib test_libs $(TESTS)
@@ -939,7 +826,7 @@ gen_parallel_tests:
 # 107.816 PASS t/DBTest.EncodeDecompressedBlockSizeTest
 #
 slow_test_regexp = \
-	^.*SnapshotConcurrentAccessTest.*$$|^t/run-table_test-HarnessTest.Randomized$$|^t/run-db_test-.*(?:FileCreationRandomFailure|EncodeDecompressedBlockSizeTest)$$|^.*RecoverFromCorruptedWALWithoutFlush$$
+	^.*SnapshotConcurrentAccessTest.*$$|^.*SeqAdvanceConcurrentTest.*$$|^t/run-table_test-HarnessTest.Randomized$$|^t/run-db_test-.*(?:FileCreationRandomFailure|EncodeDecompressedBlockSizeTest)$$|^.*RecoverFromCorruptedWALWithoutFlush$$
 prioritize_long_running_tests =						\
   perl -pe 's,($(slow_test_regexp)),100 $$1,'				\
     | sort -k1,1gr							\
@@ -1061,6 +948,8 @@ crash_test_with_txn: whitebox_crash_test_with_txn blackbox_crash_test_with_txn
 
 crash_test_with_best_efforts_recovery: blackbox_crash_test_with_best_efforts_recovery
 
+crash_test_with_ts: whitebox_crash_test_with_ts blackbox_crash_test_with_ts
+
 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)
@@ -1074,6 +963,9 @@ blackbox_crash_test_with_txn: db_stress
 blackbox_crash_test_with_best_efforts_recovery: db_stress
 	$(PYTHON) -u tools/db_crashtest.py --test_best_efforts_recovery blackbox $(CRASH_TEST_EXT_ARGS)
 
+blackbox_crash_test_with_ts: db_stress
+	$(PYTHON) -u tools/db_crashtest.py --enable_ts blackbox $(CRASH_TEST_EXT_ARGS)
+
 ifeq ($(CRASH_TEST_KILL_ODD),)
   CRASH_TEST_KILL_ODD=888887
 endif
@@ -1092,6 +984,10 @@ whitebox_crash_test_with_txn: db_stress
 	$(PYTHON) -u tools/db_crashtest.py --txn whitebox --random_kill_odd \
       $(CRASH_TEST_KILL_ODD) $(CRASH_TEST_EXT_ARGS)
 
+whitebox_crash_test_with_ts: db_stress
+	$(PYTHON) -u tools/db_crashtest.py --enable_ts whitebox --random_kill_odd \
+      $(CRASH_TEST_KILL_ODD) $(CRASH_TEST_EXT_ARGS)
+
 asan_check: clean
 	COMPILE_WITH_ASAN=1 $(MAKE) check -j32
 	$(MAKE) clean
@@ -1237,8 +1133,9 @@ analyze_incremental:
 		$(MAKE) dbg
 
 CLEAN_FILES += unity.cc
-unity.cc: Makefile
+unity.cc: Makefile util/build_version.cc.in
 	rm -f $@ $@-t
+	$(AM_V_at)$(gen_build_version) > util/build_version.cc
 	for source_file in $(LIB_SOURCES); do \
 		echo "#include \"$$source_file\"" >> $@-t; \
 	done
@@ -1318,11 +1215,11 @@ $(STATIC_TEST_LIBRARY): $(TEST_OBJECTS)
 	$(AM_V_AR)rm -f $@ $(SHARED_TEST_LIBRARY)
 	$(AM_V_at)$(AR) $(ARFLAGS) $@ $^
 
-$(STATIC_TOOLS_LIBRARY): $(BENCH_OBJECTS) $(TOOL_OBJECTS)
+$(STATIC_TOOLS_LIBRARY): $(TOOL_OBJECTS)
 	$(AM_V_AR)rm -f $@ $(SHARED_TOOLS_LIBRARY)
 	$(AM_V_at)$(AR) $(ARFLAGS) $@ $^
 
-$(STATIC_STRESS_LIBRARY): $(ANALYZE_OBJECTS) $(STRESS_OBJECTS)
+$(STATIC_STRESS_LIBRARY): $(ANALYZE_OBJECTS) $(STRESS_OBJECTS) $(TESTUTIL)
 	$(AM_V_AR)rm -f $@ $(SHARED_STRESS_LIBRARY)
 	$(AM_V_at)$(AR) $(ARFLAGS) $@ $^
 
@@ -1334,7 +1231,7 @@ $(SHARED_TOOLS_LIBRARY): $(TOOL_OBJECTS) $(SHARED1)
 	$(AM_V_AR)rm -f $@ $(STATIC_TOOLS_LIBRARY)
 	$(AM_SHARE)
 
-$(SHARED_STRESS_LIBRARY): $(ANALYZE_OBJECTS) $(STRESS_OBJECTS) $(SHARED_TOOLS_LIBRARY) $(SHARED1)
+$(SHARED_STRESS_LIBRARY): $(ANALYZE_OBJECTS) $(STRESS_OBJECTS) $(TESTUTIL) $(SHARED_TOOLS_LIBRARY) $(SHARED1)
 	$(AM_V_AR)rm -f $@ $(STATIC_STRESS_LIBRARY)
 	$(AM_SHARE)
 
@@ -1356,7 +1253,7 @@ folly_synchronization_distributed_mutex_test: $(OBJ_DIR)/third-party/folly/folly
 	$(AM_LINK)
 endif
 
-cache_bench: $(OBJ_DIR)/cache/cache_bench.o $(LIBRARY)
+cache_bench: $(OBJ_DIR)/cache/cache_bench.o $(CACHE_BENCH_OBJECTS) $(LIBRARY)
 	$(AM_LINK)
 
 persistent_cache_bench: $(OBJ_DIR)/utilities/persistent_cache/persistent_cache_bench.o $(LIBRARY)
@@ -1416,6 +1313,9 @@ hash_test: $(OBJ_DIR)/util/hash_test.o $(TEST_LIBRARY) $(LIBRARY)
 random_test: $(OBJ_DIR)/util/random_test.o  $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
+ribbon_test: $(OBJ_DIR)/util/ribbon_test.o $(TEST_LIBRARY) $(LIBRARY)
+	$(AM_LINK)
+
 option_change_migration_test: $(OBJ_DIR)/utilities/option_change_migration/option_change_migration_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
@@ -1461,6 +1361,12 @@ slice_transform_test: $(OBJ_DIR)/util/slice_transform_test.o $(TEST_LIBRARY) $(L
 db_basic_test: $(OBJ_DIR)/db/db_basic_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
+db_blob_basic_test: $(OBJ_DIR)/db/blob/db_blob_basic_test.o $(TEST_LIBRARY) $(LIBRARY)
+	$(AM_LINK)
+
+db_blob_compaction_test: $(OBJ_DIR)/db/blob/db_blob_compaction_test.o $(TEST_LIBRARY) $(LIBRARY)
+	$(AM_LINK)
+
 db_with_timestamp_basic_test: $(OBJ_DIR)/db/db_with_timestamp_basic_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
@@ -1509,6 +1415,9 @@ db_inplace_update_test: $(OBJ_DIR)/db/db_inplace_update_test.o $(TEST_LIBRARY) $
 db_iterator_test: $(OBJ_DIR)/db/db_iterator_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
+db_kv_checksum_test: $(OBJ_DIR)/db/db_kv_checksum_test.o $(TEST_LIBRARY) $(LIBRARY)
+	$(AM_LINK)
+
 db_memtable_test: $(OBJ_DIR)/db/db_memtable_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
@@ -1784,6 +1693,9 @@ compact_files_test: $(OBJ_DIR)/db/compact_files_test.o $(TEST_LIBRARY) $(LIBRARY
 configurable_test: options/configurable_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
+customizable_test: options/customizable_test.o $(TEST_LIBRARY) $(LIBRARY)
+	$(AM_LINK)
+
 options_test: $(OBJ_DIR)/options/options_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
@@ -1835,7 +1747,7 @@ write_callback_test: $(OBJ_DIR)/db/write_callback_test.o $(TEST_LIBRARY) $(LIBRA
 heap_test: $(OBJ_DIR)/util/heap_test.o $(GTEST)
 	$(AM_LINK)
 
-transaction_lock_mgr_test: utilities/transactions/transaction_lock_mgr_test.o $(TEST_LIBRARY) $(LIBRARY)
+point_lock_manager_test: utilities/transactions/lock/point/point_lock_manager_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
 transaction_test: $(OBJ_DIR)/utilities/transactions/transaction_test.o $(TEST_LIBRARY) $(LIBRARY)
@@ -1889,13 +1801,16 @@ blob_db_test: $(OBJ_DIR)/utilities/blob_db/blob_db_test.o $(TEST_LIBRARY) $(LIBR
 repeatable_thread_test: $(OBJ_DIR)/util/repeatable_thread_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
+range_locking_test: utilities/transactions/lock/range/range_locking_test.o $(TEST_LIBRARY) $(LIBRARY)
+	$(AM_LINK)
+
 range_tombstone_fragmenter_test: $(OBJ_DIR)/db/range_tombstone_fragmenter_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
 sst_file_reader_test: $(OBJ_DIR)/table/sst_file_reader_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
-db_secondary_test: $(OBJ_DIR)/db/db_impl/db_secondary_test.o $(TEST_LIBRARY) $(LIBRARY)
+db_secondary_test: $(OBJ_DIR)/db/db_secondary_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
 block_cache_tracer_test: $(OBJ_DIR)/trace_replay/block_cache_tracer_test.o $(TEST_LIBRARY) $(LIBRARY)
@@ -1913,6 +1828,9 @@ blob_file_addition_test: $(OBJ_DIR)/db/blob/blob_file_addition_test.o $(TEST_LIB
 blob_file_builder_test: $(OBJ_DIR)/db/blob/blob_file_builder_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
+blob_file_cache_test: $(OBJ_DIR)/db/blob/blob_file_cache_test.o $(TEST_LIBRARY) $(LIBRARY)
+	$(AM_LINK)
+
 blob_file_garbage_test: $(OBJ_DIR)/db/blob/blob_file_garbage_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
@@ -1940,6 +1858,11 @@ io_tracer_parser_test: $(OBJ_DIR)/tools/io_tracer_parser_test.o $(OBJ_DIR)/tools
 io_tracer_parser: $(OBJ_DIR)/tools/io_tracer_parser.o $(TOOLS_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
+db_blob_corruption_test: $(OBJ_DIR)/db/blob/db_blob_corruption_test.o $(TEST_LIBRARY) $(LIBRARY)
+	$(AM_LINK)
+
+db_write_buffer_manager_test: $(OBJ_DIR)/db/db_write_buffer_manager_test.o $(TEST_LIBRARY) $(LIBRARY)
+	$(AM_LINK)
 #-------------------------------------------------
 # make install related stuff
 PREFIX ?= /usr/local
@@ -1964,6 +1887,10 @@ install-headers: gen-pc
 	for header in `$(FIND) "include/rocksdb" -type f -name *.h`; do \
 		install -C -m 644 $$header $(DESTDIR)/$(PREFIX)/$$header; \
 	done
+	for header in $(ROCKSDB_PLUGIN_HEADERS); do \
+		install -d $(DESTDIR)/$(PREFIX)/include/rocksdb/`dirname $$header`; \
+		install -C -m 644 $$header $(DESTDIR)/$(PREFIX)/include/rocksdb/$$header; \
+	done
 	install -C -m 644 rocksdb.pc $(INSTALL_LIBDIR)/pkgconfig/rocksdb.pc
 
 install-static: install-headers $(LIBRARY)
@@ -2047,11 +1974,11 @@ BZIP2_DOWNLOAD_BASE ?= https://sourceware.org/pub/bzip2
 SNAPPY_VER ?= 1.1.8
 SNAPPY_SHA256 ?= 16b677f07832a612b0836178db7f374e414f94657c138e6993cbfc5dcc58651f
 SNAPPY_DOWNLOAD_BASE ?= https://github.com/google/snappy/archive
-LZ4_VER ?= 1.9.2
-LZ4_SHA256 ?= 658ba6191fa44c92280d4aa2c271b0f4fbc0e34d249578dd05e50e76d0e5efcc
+LZ4_VER ?= 1.9.3
+LZ4_SHA256 ?= 030644df4611007ff7dc962d981f390361e6c97a34e5cbc393ddfbe019ffe2c1
 LZ4_DOWNLOAD_BASE ?= https://github.com/lz4/lz4/archive
-ZSTD_VER ?= 1.4.4
-ZSTD_SHA256 ?= a364f5162c7d1a455cc915e8e3cf5f4bd8b75d09bc0f53965b0c9ca1383c52c8
+ZSTD_VER ?= 1.4.9
+ZSTD_SHA256 ?= acf714d98e3db7b876e5b540cbf6dee298f60eb3c0723104f6d3f065cd60d6a8
 ZSTD_DOWNLOAD_BASE ?= https://github.com/facebook/zstd/archive
 CURL_SSL_OPTS ?= --tlsv1
 
@@ -2083,80 +2010,80 @@ ifeq ($(PLATFORM), OS_AIX)
 	SNAPPY_MAKE_TARGET = libsnappy.la
 endif
 ifeq ($(PLATFORM), OS_OPENBSD)
-        JAVA_INCLUDE = -I$(JAVA_HOME)/include -I$(JAVA_HOME)/include/openbsd
+	JAVA_INCLUDE = -I$(JAVA_HOME)/include -I$(JAVA_HOME)/include/openbsd
 	ROCKSDBJNILIB = librocksdbjni-openbsd$(ARCH).so
-        ROCKSDB_JAR = rocksdbjni-$(ROCKSDB_JAVA_VERSION)-openbsd$(ARCH).jar
+	ROCKSDB_JAR = rocksdbjni-$(ROCKSDB_JAVA_VERSION)-openbsd$(ARCH).jar
 endif
 
-libz.a:
-	-rm -rf zlib-$(ZLIB_VER)
-ifeq (,$(wildcard ./zlib-$(ZLIB_VER).tar.gz))
+zlib-$(ZLIB_VER).tar.gz:
 	curl --fail --output zlib-$(ZLIB_VER).tar.gz --location ${ZLIB_DOWNLOAD_BASE}/zlib-$(ZLIB_VER).tar.gz
-endif
 	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\"; \
 		exit 1; \
 	fi
+
+libz.a: zlib-$(ZLIB_VER).tar.gz
+	-rm -rf zlib-$(ZLIB_VER)
 	tar xvzf zlib-$(ZLIB_VER).tar.gz
-	cd zlib-$(ZLIB_VER) && CFLAGS='-fPIC ${EXTRA_CFLAGS}' LDFLAGS='${EXTRA_LDFLAGS}' ./configure --static && $(MAKE)
+	cd zlib-$(ZLIB_VER) && CFLAGS='-fPIC ${JAVA_STATIC_DEPS_CCFLAGS} ${EXTRA_CFLAGS}' LDFLAGS='${JAVA_STATIC_DEPS_LDFLAGS} ${EXTRA_LDFLAGS}' ./configure --static && $(MAKE)
 	cp zlib-$(ZLIB_VER)/libz.a .
 
-libbz2.a:
-	-rm -rf bzip2-$(BZIP2_VER)
-ifeq (,$(wildcard ./bzip2-$(BZIP2_VER).tar.gz))
+bzip2-$(BZIP2_VER).tar.gz:
 	curl --fail --output bzip2-$(BZIP2_VER).tar.gz --location ${CURL_SSL_OPTS} ${BZIP2_DOWNLOAD_BASE}/bzip2-$(BZIP2_VER).tar.gz
-endif
 	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\"; \
 		exit 1; \
 	fi
+
+libbz2.a: bzip2-$(BZIP2_VER).tar.gz
+	-rm -rf bzip2-$(BZIP2_VER)
 	tar xvzf bzip2-$(BZIP2_VER).tar.gz
-	cd bzip2-$(BZIP2_VER) && $(MAKE) CFLAGS='-fPIC -O2 -g -D_FILE_OFFSET_BITS=64 ${EXTRA_CFLAGS}' AR='ar ${EXTRA_ARFLAGS}'
+	cd bzip2-$(BZIP2_VER) && $(MAKE) CFLAGS='-fPIC -O2 -g -D_FILE_OFFSET_BITS=64 ${JAVA_STATIC_DEPS_CCFLAGS} ${EXTRA_CFLAGS}' LDFLAGS='${JAVA_STATIC_DEPS_LDFLAGS} ${EXTRA_LDFLAGS}' AR='ar ${EXTRA_ARFLAGS}'
 	cp bzip2-$(BZIP2_VER)/libbz2.a .
 
-libsnappy.a:
-	-rm -rf snappy-$(SNAPPY_VER)
-ifeq (,$(wildcard ./snappy-$(SNAPPY_VER).tar.gz))
+snappy-$(SNAPPY_VER).tar.gz:
 	curl --fail --output snappy-$(SNAPPY_VER).tar.gz --location ${CURL_SSL_OPTS} ${SNAPPY_DOWNLOAD_BASE}/$(SNAPPY_VER).tar.gz
-endif
 	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
+
+libsnappy.a: snappy-$(SNAPPY_VER).tar.gz
+	-rm -rf snappy-$(SNAPPY_VER)
 	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) ${SNAPPY_MAKE_TARGET}
+	cd snappy-$(SNAPPY_VER)/build && CFLAGS='${JAVA_STATIC_DEPS_CCFLAGS} ${EXTRA_CFLAGS}' CXXFLAGS='${JAVA_STATIC_DEPS_CXXFLAGS} ${EXTRA_CXXFLAGS}' LDFLAGS='${JAVA_STATIC_DEPS_LDFLAGS} ${EXTRA_LDFLAGS}' cmake -DCMAKE_POSITION_INDEPENDENT_CODE=ON ${PLATFORM_CMAKE_FLAGS} .. && $(MAKE) ${SNAPPY_MAKE_TARGET}
 	cp snappy-$(SNAPPY_VER)/build/libsnappy.a .
 
-liblz4.a:
-	-rm -rf lz4-$(LZ4_VER)
-ifeq (,$(wildcard ./lz4-$(LZ4_VER).tar.gz))
+lz4-$(LZ4_VER).tar.gz:
 	curl --fail --output lz4-$(LZ4_VER).tar.gz --location ${CURL_SSL_OPTS} ${LZ4_DOWNLOAD_BASE}/v$(LZ4_VER).tar.gz
-endif
 	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\"; \
 		exit 1; \
 	fi
+
+liblz4.a: lz4-$(LZ4_VER).tar.gz
+	-rm -rf lz4-$(LZ4_VER)
 	tar xvzf lz4-$(LZ4_VER).tar.gz
-	cd lz4-$(LZ4_VER)/lib && $(MAKE) CFLAGS='-fPIC -O2 ${EXTRA_CFLAGS}' all
+	cd lz4-$(LZ4_VER)/lib && $(MAKE) CFLAGS='-fPIC -O2 ${JAVA_STATIC_DEPS_CCFLAGS} ${EXTRA_CFLAGS}' LDFLAGS='${JAVA_STATIC_DEPS_LDFLAGS} ${EXTRA_LDFLAGS}' all
 	cp lz4-$(LZ4_VER)/lib/liblz4.a .
 
-libzstd.a:
-	-rm -rf zstd-$(ZSTD_VER)
-ifeq (,$(wildcard ./zstd-$(ZSTD_VER).tar.gz))
+zstd-$(ZSTD_VER).tar.gz:
 	curl --fail --output zstd-$(ZSTD_VER).tar.gz --location ${CURL_SSL_OPTS} ${ZSTD_DOWNLOAD_BASE}/v$(ZSTD_VER).tar.gz
-endif
 	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\"; \
 		exit 1; \
 	fi
+
+libzstd.a: zstd-$(ZSTD_VER).tar.gz
+	-rm -rf zstd-$(ZSTD_VER)
 	tar xvzf zstd-$(ZSTD_VER).tar.gz
-	cd zstd-$(ZSTD_VER)/lib && DESTDIR=. PREFIX= $(MAKE) CFLAGS='-fPIC -O2 ${EXTRA_CFLAGS}' install
+	cd zstd-$(ZSTD_VER)/lib && DESTDIR=. PREFIX= $(MAKE) CFLAGS='-fPIC -O2 ${JAVA_STATIC_DEPS_CCFLAGS} ${EXTRA_CFLAGS}' LDFLAGS='${JAVA_STATIC_DEPS_LDFLAGS} ${EXTRA_LDFLAGS}' libzstd.a
 	cp zstd-$(ZSTD_VER)/lib/libzstd.a .
 
 # A version of each $(LIB_OBJECTS) compiled with -fPIC and a fixed set of static compression libraries
@@ -2165,14 +2092,23 @@ JAVA_COMPRESSIONS = libz.a libbz2.a libsnappy.a liblz4.a libzstd.a
 endif
 
 JAVA_STATIC_FLAGS = -DZLIB -DBZIP2 -DSNAPPY -DLZ4 -DZSTD
-JAVA_STATIC_INCLUDES = -I./zlib-$(ZLIB_VER) -I./bzip2-$(BZIP2_VER) -I./snappy-$(SNAPPY_VER) -I./lz4-$(LZ4_VER)/lib -I./zstd-$(ZSTD_VER)/lib/include
-ifneq ($(findstring rocksdbjavastatic, $(MAKECMDGOALS)),)
+JAVA_STATIC_INCLUDES = -I./zlib-$(ZLIB_VER) -I./bzip2-$(BZIP2_VER) -I./snappy-$(SNAPPY_VER) -I./snappy-$(SNAPPY_VER)/build -I./lz4-$(LZ4_VER)/lib -I./zstd-$(ZSTD_VER)/lib -I./zstd-$(ZSTD_VER)/lib/dictBuilder
+
+ifneq ($(findstring rocksdbjavastatic, $(filter-out rocksdbjavastatic_deps, $(MAKECMDGOALS))),)
 CXXFLAGS += $(JAVA_STATIC_FLAGS) $(JAVA_STATIC_INCLUDES)
-CFLAGS +=  $(JAVA_STATIC_FLAGS) $(JAVA_STATIC_INCLUDES)
+CFLAGS += $(JAVA_STATIC_FLAGS) $(JAVA_STATIC_INCLUDES)
 endif
-rocksdbjavastatic: $(LIB_OBJECTS) $(JAVA_COMPRESSIONS)
-	cd java;$(MAKE) javalib;
-	rm -f ./java/target/$(ROCKSDBJNILIB)
+rocksdbjavastatic:
+ifeq ($(JAVA_HOME),)
+	$(error JAVA_HOME is not set)
+endif
+	$(MAKE) rocksdbjavastatic_deps
+	$(MAKE) rocksdbjavastatic_libobjects
+	$(MAKE) rocksdbjavastatic_javalib
+
+rocksdbjavastatic_javalib:
+	cd java; SHA256_CMD='$(SHA256_CMD)' $(MAKE) javalib
+	rm -f java/target/$(ROCKSDBJNILIB)
 	$(CXX) $(CXXFLAGS) -I./java/. $(JAVA_INCLUDE) -shared -fPIC \
 	  -o ./java/target/$(ROCKSDBJNILIB) $(JNI_NATIVE_SOURCES) \
 	  $(LIB_OBJECTS) $(COVERAGEFLAGS) \
@@ -2189,6 +2125,10 @@ rocksdbjavastatic: $(LIB_OBJECTS) $(JAVA_COMPRESSIONS)
 	openssl sha1 java/target/$(ROCKSDB_JAVADOCS_JAR) | sed 's/.*= \([0-9a-f]*\)/\1/' > java/target/$(ROCKSDB_JAVADOCS_JAR).sha1
 	openssl sha1 java/target/$(ROCKSDB_SOURCES_JAR) | sed 's/.*= \([0-9a-f]*\)/\1/' > java/target/$(ROCKSDB_SOURCES_JAR).sha1
 
+rocksdbjavastatic_deps: $(JAVA_COMPRESSIONS)
+
+rocksdbjavastatic_libobjects: $(LIB_OBJECTS)
+
 rocksdbjavastaticrelease: rocksdbjavastatic
 	cd java/crossbuild && (vagrant destroy -f || true) && vagrant up linux32 && vagrant halt linux32 && vagrant up linux64 && vagrant halt linux64 && vagrant up linux64-musl && vagrant halt linux64-musl
 	cd java;jar -cf target/$(ROCKSDB_JAR_ALL) HISTORY*.md
@@ -2204,7 +2144,7 @@ rocksdbjavastaticreleasedocker: rocksdbjavastatic rocksdbjavastaticdockerx86 roc
 
 rocksdbjavastaticdockerx86:
 	mkdir -p java/target
-	docker run --rm --name rocksdb_linux_x86-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:centos6_x86-be /rocksdb-host/java/crossbuild/docker-build-linux-centos.sh
+	docker run --rm --name rocksdb_linux_x86-be --platform linux/386 --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:centos6_x86-be /rocksdb-host/java/crossbuild/docker-build-linux-centos.sh
 
 rocksdbjavastaticdockerx86_64:
 	mkdir -p java/target
@@ -2220,7 +2160,7 @@ rocksdbjavastaticdockerarm64v8:
 
 rocksdbjavastaticdockerx86musl:
 	mkdir -p java/target
-	docker run --rm --name rocksdb_linux_x86-musl-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:alpine3_x86-be /rocksdb-host/java/crossbuild/docker-build-linux-centos.sh
+	docker run --rm --name rocksdb_linux_x86-musl-be --platform linux/386 --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:alpine3_x86-be /rocksdb-host/java/crossbuild/docker-build-linux-centos.sh
 
 rocksdbjavastaticdockerx86_64musl:
 	mkdir -p java/target
@@ -2258,7 +2198,10 @@ jl/%.o: %.cc
 	$(AM_V_CC)mkdir -p $(@D) && $(CXX) $(CXXFLAGS) -fPIC -c $< -o $@ $(COVERAGEFLAGS)
 
 rocksdbjava: $(LIB_OBJECTS)
-	$(AM_V_GEN)cd java;$(MAKE) javalib;
+ifeq ($(JAVA_HOME),)
+	$(error JAVA_HOME is not set)
+endif
+	$(AM_V_GEN)cd java; SHA256_CMD='$(SHA256_CMD)' $(MAKE) javalib;
 	$(AM_V_at)rm -f ./java/target/$(ROCKSDBJNILIB)
 	$(AM_V_at)$(CXX) $(CXXFLAGS) -I./java/. $(JAVA_INCLUDE) -shared -fPIC -o ./java/target/$(ROCKSDBJNILIB) $(JNI_NATIVE_SOURCES) $(LIB_OBJECTS) $(JAVA_LDFLAGS) $(COVERAGEFLAGS)
 	$(AM_V_at)cd java;jar -cf target/$(ROCKSDB_JAR) HISTORY*.md
@@ -2270,13 +2213,13 @@ jclean:
 	cd java;$(MAKE) clean;
 
 jtest_compile: rocksdbjava
-	cd java;$(MAKE) java_test
+	cd java; SHA256_CMD='$(SHA256_CMD)' $(MAKE) java_test
 
 jtest_run:
 	cd java;$(MAKE) run_test
 
 jtest: rocksdbjava
-	cd java;$(MAKE) sample;$(MAKE) test;
+	cd java;$(MAKE) sample; SHA256_CMD='$(SHA256_CMD)' $(MAKE) test;
 	$(PYTHON) tools/check_all_python.py # TODO peterd: find a better place for this check in CI targets
 
 jdb_bench:
@@ -2335,12 +2278,14 @@ endif
 # ---------------------------------------------------------------------------
 #  	Source files dependencies detection
 # ---------------------------------------------------------------------------
-
+# If skip dependencies is ON, skip including the dep files
+ifneq ($(SKIP_DEPENDS), 1)
 DEPFILES = $(patsubst %.cc, $(OBJ_DIR)/%.cc.d, $(ALL_SOURCES))
 DEPFILES+ = $(patsubst %.c, $(OBJ_DIR)/%.c.d, $(LIB_SOURCES_C) $(TEST_MAIN_SOURCES_C))
 ifeq ($(USE_FOLLY_DISTRIBUTED_MUTEX),1)
   DEPFILES +=$(patsubst %.cpp, $(OBJ_DIR)/%.cpp.d, $(FOLLY_SOURCES))
 endif
+endif
 
 # Add proper dependency support so changing a .h file forces a .cc file to
 # rebuild.
@@ -2380,20 +2325,9 @@ endif
 build_subset_tests: $(ROCKSDBTESTS_SUBSET)
 	$(AM_V_GEN)if [ -n "$${ROCKSDBTESTS_SUBSET_TESTS_TO_FILE}" ]; then echo "$(ROCKSDBTESTS_SUBSET)" > "$${ROCKSDBTESTS_SUBSET_TESTS_TO_FILE}"; else echo "$(ROCKSDBTESTS_SUBSET)"; fi
 
-# if the make goal is either "clean" or "format", we shouldn't
-# try to import the *.d files.
-# TODO(kailiu) The unfamiliarity of Make's conditions leads to the ugly
-# working solution.
-ifneq ($(MAKECMDGOALS),clean)
-ifneq ($(MAKECMDGOALS),format)
-ifneq ($(MAKECMDGOALS),jclean)
-ifneq ($(MAKECMDGOALS),jtest)
-ifneq ($(MAKECMDGOALS),package)
-ifneq ($(MAKECMDGOALS),analyze)
+# Remove the rules for which dependencies should not be generated and see if any are left.
+#If so, include the dependencies; if not, do not include the dependency files
+ROCKS_DEP_RULES=$(filter-out clean format check-format check-buck-targets jclean jtest package analyze tags rocksdbjavastatic% unity.% unity_test, $(MAKECMDGOALS))
+ifneq ("$(ROCKS_DEP_RULES)", "")
 -include $(DEPFILES)
 endif
-endif
-endif
-endif
-endif
-endif

PLUGINS.md

@@ -0,0 +1,4 @@
+This is the list of all known third-party plugins for RocksDB. If something is missing, please open a pull request to add it.
+
+* [Dedupfs](https://github.com/ajkr/dedupfs): an example for plugin developers to reference
+* [ZenFS](https://github.com/westerndigitalcorporation/zenfs): a file system for zoned block devices

README.md

@@ -3,7 +3,7 @@
 [![CircleCI Status](https://circleci.com/gh/facebook/rocksdb.svg?style=svg)](https://circleci.com/gh/facebook/rocksdb)
 [![TravisCI Status](https://travis-ci.org/facebook/rocksdb.svg?branch=master)](https://travis-ci.org/facebook/rocksdb)
 [![Appveyor Build status](https://ci.appveyor.com/api/projects/status/fbgfu0so3afcno78/branch/master?svg=true)](https://ci.appveyor.com/project/Facebook/rocksdb/branch/master)
-[![PPC64le Build Status](http://140.211.168.68:8080/buildStatus/icon?job=Rocksdb)](http://140.211.168.68:8080/job/Rocksdb)
+[![PPC64le Build Status](http://140-211-168-68-openstack.osuosl.org:8080/buildStatus/icon?job=rocksdb&style=plastic)](http://140-211-168-68-openstack.osuosl.org: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)

USERS.md

@@ -26,6 +26,9 @@ Learn more about those use cases in a Tech Talk by Ankit Gupta and Naveen Somasu
 ## Yahoo
 Yahoo is using RocksDB as a storage engine for their biggest distributed data store Sherpa. Learn more about it here: http://yahooeng.tumblr.com/post/120730204806/sherpa-scales-new-heights
 
+## Baidu
+[Apache Doris](http://doris.apache.org/master/en/) is a MPP analytical database engine released by Baidu. It [uses RocksDB](http://doris.apache.org/master/en/administrator-guide/operation/tablet-meta-tool.html) to manage its tablet's metadata.
+
 ## CockroachDB
 CockroachDB is an open-source geo-replicated transactional database. They are using RocksDB as their storage engine. Check out their github: https://github.com/cockroachdb/cockroach
 
@@ -44,7 +47,7 @@ Tango is using RocksDB as a graph storage to store all users' connection data an
 Turn is using RocksDB as a storage layer for their key/value store, serving at peak 2.4MM QPS out of different datacenters.
 Check out our RocksDB Protobuf merge operator at: https://github.com/vladb38/rocksdb_protobuf
 
-## Santanader UK/Cloudera Profession Services
+## Santander UK/Cloudera Profession Services
 Check out their blog post: http://blog.cloudera.com/blog/2015/08/inside-santanders-near-real-time-data-ingest-architecture/
 
 ## Airbnb
@@ -67,7 +70,7 @@ Pinterest's Object Retrieval System uses RocksDB for storage: https://www.youtub
 [VWO's](https://vwo.com/) Smart Code checker and URL helper uses RocksDB to store all the URLs where VWO's Smart Code is installed.
 
 ## quasardb
-[quasardb](https://www.quasardb.net) is a high-performance, distributed, transactional key-value database that integrates well with in-memory analytics engines such as Apache Spark. 
+[quasardb](https://www.quasardb.net) is a high-performance, distributed, transactional key-value database that integrates well with in-memory analytics engines such as Apache Spark.
 quasardb uses a heavily tuned RocksDB as its persistence layer.
 
 ## Netflix
@@ -86,7 +89,7 @@ quasardb uses a heavily tuned RocksDB as its persistence layer.
 [Uber](http://eng.uber.com/cherami/) uses RocksDB as a durable and scalable task queue.
 
 ## 360 Pika
-[360](http://www.360.cn/) [Pika](https://github.com/Qihoo360/pika) is a nosql compatible with redis. With the huge amount of data stored, redis may suffer for a capacity bottleneck, and pika was born for solving it. It has widely been widely used in many company
+[360](http://www.360.cn/) [Pika](https://github.com/Qihoo360/pika) is a nosql compatible with redis. With the huge amount of data stored, redis may suffer for a capacity bottleneck, and pika was born for solving it. It has widely been used in many companies.
 
 ## LzLabs
 LzLabs is using RocksDB as a storage engine in their multi-database distributed framework to store application configuration and user data.
@@ -96,16 +99,19 @@ LzLabs is using RocksDB as a storage engine in their multi-database distributed
 
 ## 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.
- 
+
 ## Avrio Project
  [Avrio Project](http://avrio-project.github.io/avrio.network/) is using RocksDB in [Avrio ](https://github.com/avrio-project/avrio) to store blocks, account balances and data and other blockchain-releated data. Avrio is a multiblockchain decentralized cryptocurrency empowering monetary transactions.
- 
+
 ## Crux
 [Crux](https://github.com/juxt/crux) is a document database that uses RocksDB for local [EAV](https://en.wikipedia.org/wiki/Entity%E2%80%93attribute%E2%80%93value_model) index storage to enable point-in-time bitemporal Datalog queries. The "unbundled" architecture uses Kafka to provide horizontal scalability.
 
 ## Nebula Graph
-
 [Nebula Graph](https://github.com/vesoft-inc/nebula) is a distributed, scalable, lightning-fast, open source graph database capable of hosting super large scale graphs with dozens of billions of vertices (nodes) and trillions of edges, with milliseconds of latency.
 
 ## YugabyteDB
 [YugabyteDB](https://www.yugabyte.com/) is an open source, high performance, distributed SQL database that uses RocksDB as its storage layer. For more information, please see https://github.com/yugabyte/yugabyte-db/.
+
+## ArangoDB
+[ArangoDB](https://www.arangodb.com/) is a native multi-model database with flexible data models for documents, graphs, and key-values, for building high performance applications using a convenient SQL-like query language or JavaScript extensions. It uses RocksDB as its sotrage engine.
+

buckifier/buckify_rocksdb.py

@@ -24,10 +24,10 @@ from util import ColorString
 # (This generates a TARGET file without user-specified dependency for unit
 # tests.)
 # $python3 buckifier/buckify_rocksdb.py \
-#        '{"fake": { \
-#                      "extra_deps": [":test_dep", "//fakes/module:mock1"],  \
-#                      "extra_compiler_flags": ["-DROCKSDB_LITE", "-Os"], \
-#                  } \
+#        '{"fake": {
+#                      "extra_deps": [":test_dep", "//fakes/module:mock1"],
+#                      "extra_compiler_flags": ["-DROCKSDB_LITE", "-Os"]
+#                  }
 #         }'
 # (Generated TARGETS file has test_dep and mock1 as dependencies for RocksDB
 # unit tests, and will use the extra_compiler_flags to compile the unit test
@@ -69,25 +69,25 @@ def get_cc_files(repo_path):
     return cc_files
 
 
-# Get parallel tests from Makefile
-def get_parallel_tests(repo_path):
+# Get non_parallel tests from Makefile
+def get_non_parallel_tests(repo_path):
     Makefile = repo_path + "/Makefile"
 
     s = set({})
 
-    found_parallel_tests = False
+    found_non_parallel_tests = False
     for line in open(Makefile):
         line = line.strip()
-        if line.startswith("PARALLEL_TEST ="):
-            found_parallel_tests = True
-        elif found_parallel_tests:
+        if line.startswith("NON_PARALLEL_TEST ="):
+            found_non_parallel_tests = True
+        elif found_non_parallel_tests:
             if line.endswith("\\"):
                 # remove the trailing \
                 line = line[:-1]
                 line = line.strip()
                 s.add(line)
             else:
-                # we consumed all the parallel tests
+                # we consumed all the non_parallel tests
                 break
 
     return s
@@ -123,23 +123,33 @@ def generate_targets(repo_path, deps_map):
     src_mk = parse_src_mk(repo_path)
     # get all .cc files
     cc_files = get_cc_files(repo_path)
-    # get parallel tests from Makefile
-    parallel_tests = get_parallel_tests(repo_path)
+    # get non_parallel tests from Makefile
+    non_parallel_tests = get_non_parallel_tests(repo_path)
 
-    if src_mk is None or cc_files is None or parallel_tests is None:
+    if src_mk is None or cc_files is None or non_parallel_tests is None:
         return False
 
-    TARGETS = TARGETSBuilder("%s/TARGETS" % repo_path)
+    extra_argv = ""
+    if len(sys.argv) >= 2:
+        # Heuristically quote and canonicalize whitespace for inclusion
+        # in how the file was generated.
+        extra_argv = " '{0}'".format(" ".join(sys.argv[1].split()))
+
+    TARGETS = TARGETSBuilder("%s/TARGETS" % repo_path, extra_argv)
 
     # rocksdb_lib
     TARGETS.add_library(
         "rocksdb_lib",
         src_mk["LIB_SOURCES"] +
+        # always add range_tree, it's only excluded on ppc64, which we don't use internally
+        src_mk["RANGE_TREE_SOURCES"] +
         src_mk["TOOL_LIB_SOURCES"])
     # rocksdb_whole_archive_lib
     TARGETS.add_library(
         "rocksdb_whole_archive_lib",
         src_mk["LIB_SOURCES"] +
+        # always add range_tree, it's only excluded on ppc64, which we don't use internally
+        src_mk["RANGE_TREE_SOURCES"] +
         src_mk["TOOL_LIB_SOURCES"],
         deps=None,
         headers=None,
@@ -163,6 +173,11 @@ def generate_targets(repo_path, deps_map):
         src_mk.get("ANALYZER_LIB_SOURCES", []) +
         ["test_util/testutil.cc"],
         [":rocksdb_lib"])
+    # rocksdb_cache_bench_tools_lib
+    TARGETS.add_library(
+        "rocksdb_cache_bench_tools_lib",
+        src_mk.get("CACHE_BENCH_LIB_SOURCES", []),
+        [":rocksdb_lib"])
     # rocksdb_stress_lib
     TARGETS.add_rocksdb_library(
         "rocksdb_stress_lib",
@@ -201,7 +216,7 @@ def generate_targets(repo_path, deps_map):
             TARGETS.register_test(
                 test_target_name,
                 test_src,
-                test in parallel_tests,
+                test not in non_parallel_tests,
                 json.dumps(deps['extra_deps']),
                 json.dumps(deps['extra_compiler_flags']))
 

buckifier/targets_builder.py

@@ -25,10 +25,11 @@ def pretty_list(lst, indent=8):
 
 
 class TARGETSBuilder(object):
-    def __init__(self, path):
+    def __init__(self, path, extra_argv):
         self.path = path
         self.targets_file = open(path, 'wb')
-        header = targets_cfg.rocksdb_target_header_template
+        header = targets_cfg.rocksdb_target_header_template.format(
+            extra_argv=extra_argv)
         self.targets_file.write(header.encode("utf-8"))
         self.total_lib = 0
         self.total_bin = 0
@@ -79,26 +80,25 @@ class TARGETSBuilder(object):
 
     def add_c_test(self):
         self.targets_file.write(b"""
-if not is_opt_mode:
-    cpp_binary(
-        name = "c_test_bin",
-        srcs = ["db/c_test.c"],
-        arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
-        os_preprocessor_flags = ROCKSDB_OS_PREPROCESSOR_FLAGS,
-        compiler_flags = ROCKSDB_COMPILER_FLAGS,
-        preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
-        deps = [":rocksdb_test_lib"],
-    )
+cpp_binary(
+    name = "c_test_bin",
+    srcs = ["db/c_test.c"],
+    arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
+    os_preprocessor_flags = ROCKSDB_OS_PREPROCESSOR_FLAGS,
+    compiler_flags = ROCKSDB_COMPILER_FLAGS,
+    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+    include_paths = ROCKSDB_INCLUDE_PATHS,
+    deps = [":rocksdb_test_lib"],
+) if not is_opt_mode else None
 
-if not is_opt_mode:
-    custom_unittest(
-        "c_test",
-        command = [
-            native.package_name() + "/buckifier/rocks_test_runner.sh",
-            "$(location :{})".format("c_test_bin"),
-        ],
-        type = "simple",
-    )
+custom_unittest(
+    name = "c_test",
+    command = [
+        native.package_name() + "/buckifier/rocks_test_runner.sh",
+        "$(location :{})".format("c_test_bin"),
+    ],
+    type = "simple",
+) if not is_opt_mode else None
 """)
 
     def register_test(self,

buckifier/targets_cfg.py

@@ -5,7 +5,8 @@ from __future__ import print_function
 from __future__ import unicode_literals
 
 rocksdb_target_header_template = \
-    """# This file \100generated by `python3 buckifier/buckify_rocksdb.py`
+    """# This file \100generated by:
+#$ python3 buckifier/buckify_rocksdb.py{extra_argv}
 # --> DO NOT EDIT MANUALLY <--
 # This file is a Facebook-specific integration for buck builds, so can
 # only be validated by Facebook employees.
@@ -31,7 +32,7 @@ ROCKSDB_EXTERNAL_DEPS = [
     ("zlib", None, "z"),
     ("gflags", None, "gflags"),
     ("lz4", None, "lz4"),
-    ("zstd", None),
+    ("zstd", None, "zstd"),
 ]
 
 ROCKSDB_OS_DEPS = [
@@ -93,17 +94,19 @@ ROCKSDB_PREPROCESSOR_FLAGS = [
 
     # Added missing flags from output of build_detect_platform
     "-DROCKSDB_BACKTRACE",
-
-    # Directories with files for #include
-    "-I" + REPO_PATH + "include/",
-    "-I" + REPO_PATH,
 ]
 
-ROCKSDB_ARCH_PREPROCESSOR_FLAGS = {
+# Directories with files for #include
+ROCKSDB_INCLUDE_PATHS = [
+    "",
+    "include",
+]
+
+ROCKSDB_ARCH_PREPROCESSOR_FLAGS = {{
     "x86_64": [
         "-DHAVE_PCLMUL",
     ],
-}
+}}
 
 build_mode = read_config("fbcode", "build_mode")
 
@@ -144,6 +147,7 @@ cpp_library(
     os_deps = ROCKSDB_OS_DEPS,
     os_preprocessor_flags = ROCKSDB_OS_PREPROCESSOR_FLAGS,
     preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+    include_paths = ROCKSDB_INCLUDE_PATHS,
     deps = [{deps}],
     external_deps = ROCKSDB_EXTERNAL_DEPS{extra_external_deps},
     link_whole = {link_whole},
@@ -160,6 +164,7 @@ cpp_library(
     os_deps = ROCKSDB_OS_DEPS,
     os_preprocessor_flags = ROCKSDB_OS_PREPROCESSOR_FLAGS,
     preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+    include_paths = ROCKSDB_INCLUDE_PATHS,
     deps = ROCKSDB_LIB_DEPS,
     external_deps = ROCKSDB_EXTERNAL_DEPS,
 )
@@ -172,6 +177,7 @@ cpp_binary(
     arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
     compiler_flags = ROCKSDB_COMPILER_FLAGS,
     preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+    include_paths = ROCKSDB_INCLUDE_PATHS,
     deps = [{deps}],
     external_deps = ROCKSDB_EXTERNAL_DEPS,
 )
@@ -202,6 +208,7 @@ ROCKS_TESTS = [
         os_preprocessor_flags = ROCKSDB_OS_PREPROCESSOR_FLAGS,
         compiler_flags = ROCKSDB_COMPILER_FLAGS + extra_compiler_flags,
         preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+        include_paths = ROCKSDB_INCLUDE_PATHS,
         deps = [":rocksdb_test_lib"] + extra_deps,
         external_deps = ROCKSDB_EXTERNAL_DEPS + [
             ("googletest", None, "gtest"),

build_tools/build_detect_platform

@@ -45,8 +45,13 @@ if test -z "$OUTPUT"; then
   exit 1
 fi
 
-# we depend on C++11
-PLATFORM_CXXFLAGS="-std=c++11"
+# we depend on C++11, but should be compatible with newer standards
+if [ "$ROCKSDB_CXX_STANDARD" ]; then
+  PLATFORM_CXXFLAGS="-std=$ROCKSDB_CXX_STANDARD"
+else
+  PLATFORM_CXXFLAGS="-std=c++11"
+fi
+
 # we currently depend on POSIX platform
 COMMON_FLAGS="-DROCKSDB_PLATFORM_POSIX -DROCKSDB_LIB_IO_POSIX"
 
@@ -172,7 +177,7 @@ case "$TARGET_OS" in
         PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lpthread -lrt -ldl"
         if test $ROCKSDB_USE_IO_URING; then
             # check for liburing
-            $CXX $CFLAGS -x c++ - -luring -o /dev/null 2>/dev/null  <<EOF
+            $CXX $PLATFORM_CXXFLAGS -x c++ - -luring -o /dev/null 2>/dev/null  <<EOF
               #include <liburing.h>
               int main() {
                 struct io_uring ring;
@@ -283,7 +288,7 @@ if [ "$CROSS_COMPILE" = "true" -o "$FBCODE_BUILD" = "true" ]; then
 else
     if ! test $ROCKSDB_DISABLE_FALLOCATE; then
         # Test whether fallocate is available
-        $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <fcntl.h>
           #include <linux/falloc.h>
           int main() {
@@ -299,7 +304,7 @@ EOF
     if ! test $ROCKSDB_DISABLE_SNAPPY; then
         # Test whether Snappy library is installed
         # http://code.google.com/p/snappy/
-        $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <snappy.h>
           int main() {}
 EOF
@@ -314,7 +319,7 @@ EOF
         # Test whether gflags library is installed
         # http://gflags.github.io/gflags/
         # check if the namespace is gflags
-        if $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
+        if $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
           #include <gflags/gflags.h>
           using namespace GFLAGS_NAMESPACE;
           int main() {}
@@ -323,7 +328,7 @@ EOF
           COMMON_FLAGS="$COMMON_FLAGS -DGFLAGS=1"
           PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lgflags"
         # check if namespace is gflags
-        elif $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
+        elif $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
             #include <gflags/gflags.h>
             using namespace gflags;
             int main() {}
@@ -332,7 +337,7 @@ EOF
           COMMON_FLAGS="$COMMON_FLAGS -DGFLAGS=1 -DGFLAGS_NAMESPACE=gflags"
           PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lgflags"
         # check if namespace is google
-        elif $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
+        elif $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
             #include <gflags/gflags.h>
             using namespace google;
             int main() {}
@@ -345,7 +350,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_ZLIB; then
         # Test whether zlib library is installed
-        $CXX $CFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <zlib.h>
           int main() {}
 EOF
@@ -358,7 +363,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_BZIP; then
         # Test whether bzip library is installed
-        $CXX $CFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <bzlib.h>
           int main() {}
 EOF
@@ -371,7 +376,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_LZ4; then
         # Test whether lz4 library is installed
-        $CXX $CFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <lz4.h>
           #include <lz4hc.h>
           int main() {}
@@ -385,7 +390,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_ZSTD; then
         # Test whether zstd library is installed
-        $CXX $CFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <zstd.h>
           int main() {}
 EOF
@@ -398,7 +403,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_NUMA; then
         # Test whether numa is available
-        $CXX $CFLAGS -x c++ - -o /dev/null -lnuma 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null -lnuma 2>/dev/null  <<EOF
           #include <numa.h>
           #include <numaif.h>
           int main() {}
@@ -412,7 +417,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_TBB; then
         # Test whether tbb is available
-        $CXX $CFLAGS $LDFLAGS -x c++ - -o /dev/null -ltbb 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS $LDFLAGS -x c++ - -o /dev/null -ltbb 2>/dev/null  <<EOF
           #include <tbb/tbb.h>
           int main() {}
 EOF
@@ -425,7 +430,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_JEMALLOC; then
         # Test whether jemalloc is available
-        if echo 'int main() {}' | $CXX $CFLAGS -x c++ - -o /dev/null -ljemalloc \
+        if echo 'int main() {}' | $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null -ljemalloc \
           2>/dev/null; then
             # This will enable some preprocessor identifiers in the Makefile
             JEMALLOC=1
@@ -446,7 +451,7 @@ EOF
     fi
     if ! test $JEMALLOC && ! test $ROCKSDB_DISABLE_TCMALLOC; then
         # jemalloc is not available. Let's try tcmalloc
-        if echo 'int main() {}' | $CXX $CFLAGS -x c++ - -o /dev/null \
+        if echo 'int main() {}' | $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null \
           -ltcmalloc 2>/dev/null; then
             PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -ltcmalloc"
             JAVA_LDFLAGS="$JAVA_LDFLAGS -ltcmalloc"
@@ -455,7 +460,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_MALLOC_USABLE_SIZE; then
         # Test whether malloc_usable_size is available
-        $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <malloc.h>
           int main() {
             size_t res = malloc_usable_size(0);
@@ -470,7 +475,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_MEMKIND; then
         # Test whether memkind library is installed
-        $CXX $CFLAGS $COMMON_FLAGS -lmemkind -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS $COMMON_FLAGS -lmemkind -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <memkind.h>
           int main() {
             memkind_malloc(MEMKIND_DAX_KMEM, 1024);
@@ -486,7 +491,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_PTHREAD_MUTEX_ADAPTIVE_NP; then
         # Test whether PTHREAD_MUTEX_ADAPTIVE_NP mutex type is available
-        $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <pthread.h>
           int main() {
             int x = PTHREAD_MUTEX_ADAPTIVE_NP;
@@ -501,7 +506,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_BACKTRACE; then
         # Test whether backtrace is available
-        $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <execinfo.h>
           int main() {
             void* frames[1];
@@ -513,7 +518,7 @@ EOF
             COMMON_FLAGS="$COMMON_FLAGS -DROCKSDB_BACKTRACE"
         else
             # Test whether execinfo library is installed
-            $CXX $CFLAGS -lexecinfo -x c++ - -o /dev/null 2>/dev/null  <<EOF
+            $CXX $PLATFORM_CXXFLAGS -lexecinfo -x c++ - -o /dev/null 2>/dev/null  <<EOF
               #include <execinfo.h>
               int main() {
                 void* frames[1];
@@ -530,7 +535,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_PG; then
         # Test if -pg is supported
-        $CXX $CFLAGS -pg -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS -pg -x c++ - -o /dev/null 2>/dev/null  <<EOF
           int main() {
             return 0;
           }
@@ -542,7 +547,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_SYNC_FILE_RANGE; then
         # Test whether sync_file_range is supported for compatibility with an old glibc
-        $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <fcntl.h>
           int main() {
             int fd = open("/dev/null", 0);
@@ -556,7 +561,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_SCHED_GETCPU; then
         # Test whether sched_getcpu is supported
-        $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <sched.h>
           int main() {
             int cpuid = sched_getcpu();
@@ -570,7 +575,7 @@ EOF
 
     if ! test $ROCKSDB_DISABLE_AUXV_GETAUXVAL; then
         # Test whether getauxval is supported
-        $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+        $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <sys/auxv.h>
           int main() {
             uint64_t auxv = getauxval(AT_HWCAP);
@@ -598,7 +603,7 @@ fi
 # -Wshorten-64-to-32 breaks compilation on FreeBSD i386
 if ! [ "$TARGET_OS" = FreeBSD -a "$TARGET_ARCHITECTURE" = i386 ]; then
   # Test whether -Wshorten-64-to-32 is available
-  $CXX $CFLAGS -x c++ - -o /dev/null -Wshorten-64-to-32 2>/dev/null  <<EOF
+  $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null -Wshorten-64-to-32 2>/dev/null  <<EOF
     int main() {}
 EOF
   if [ "$?" = 0 ]; then
@@ -655,6 +660,28 @@ else
     COMMON_FLAGS="$COMMON_FLAGS -mmacosx-version-min=10.12"
     PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -mmacosx-version-min=10.12"
     PLATFORM_SHARED_LDFLAGS="$PLATFORM_SHARED_LDFLAGS -mmacosx-version-min=10.12"
+    PLATFORM_CMAKE_FLAGS="-DCMAKE_OSX_DEPLOYMENT_TARGET=10.12"
+    JAVA_STATIC_DEPS_COMMON_FLAGS="-mmacosx-version-min=10.12"
+    JAVA_STATIC_DEPS_LDFLAGS="$JAVA_STATIC_DEPS_COMMON_FLAGS"
+    JAVA_STATIC_DEPS_CCFLAGS="$JAVA_STATIC_DEPS_COMMON_FLAGS"
+    JAVA_STATIC_DEPS_CXXFLAGS="$JAVA_STATIC_DEPS_COMMON_FLAGS"
+  fi
+fi
+
+if test -n "`echo $TARGET_ARCHITECTURE | grep ^ppc64`"; then
+  # check for GNU libc on ppc64
+  $CXX -x c++ - -o /dev/null 2>/dev/null <<EOF
+    #include <stdio.h>
+    #include <stdlib.h>
+    #include <gnu/libc-version.h>
+
+    int main(int argc, char *argv[]) {
+      printf("GNU libc version: %s\n", gnu_get_libc_version());
+      return 0;
+    }
+EOF
+  if [ "$?" != 0 ]; then
+      PPC_LIBC_IS_GNU=0
   fi
 fi
 
@@ -812,8 +839,12 @@ echo "CXX=$CXX" >> "$OUTPUT"
 echo "AR=$AR" >> "$OUTPUT"
 echo "PLATFORM=$PLATFORM" >> "$OUTPUT"
 echo "PLATFORM_LDFLAGS=$PLATFORM_LDFLAGS" >> "$OUTPUT"
+echo "PLATFORM_CMAKE_FLAGS=$PLATFORM_CMAKE_FLAGS" >> "$OUTPUT"
 echo "JAVA_LDFLAGS=$JAVA_LDFLAGS" >> "$OUTPUT"
 echo "JAVA_STATIC_LDFLAGS=$JAVA_STATIC_LDFLAGS" >> "$OUTPUT"
+echo "JAVA_STATIC_DEPS_CCFLAGS=$JAVA_STATIC_DEPS_CCFLAGS" >> "$OUTPUT"
+echo "JAVA_STATIC_DEPS_CXXFLAGS=$JAVA_STATIC_DEPS_CXXFLAGS" >> "$OUTPUT"
+echo "JAVA_STATIC_DEPS_LDFLAGS=$JAVA_STATIC_DEPS_LDFLAGS" >> "$OUTPUT"
 echo "JAVAC_ARGS=$JAVAC_ARGS" >> "$OUTPUT"
 echo "VALGRIND_VER=$VALGRIND_VER" >> "$OUTPUT"
 echo "PLATFORM_CCFLAGS=$PLATFORM_CCFLAGS" >> "$OUTPUT"
@@ -847,3 +878,6 @@ echo "LUA_PATH=$LUA_PATH" >> "$OUTPUT"
 if test -n "$USE_FOLLY_DISTRIBUTED_MUTEX"; then
   echo "USE_FOLLY_DISTRIBUTED_MUTEX=$USE_FOLLY_DISTRIBUTED_MUTEX" >> "$OUTPUT"
 fi
+if test -n "$PPC_LIBC_IS_GNU"; then
+  echo "PPC_LIBC_IS_GNU=$PPC_LIBC_IS_GNU" >> "$OUTPUT"
+fi

build_tools/format-diff.sh

@@ -52,15 +52,16 @@ else
     else
       echo "You didn't have clang-format-diff.py and/or clang-format available in your computer!"
       echo "You can download clang-format-diff.py by running: "
-      echo "    curl --location http://goo.gl/iUW1u2 -o ${CLANG_FORMAT_DIFF}"
+      echo "    curl --location https://raw.githubusercontent.com/llvm/llvm-project/main/clang/tools/clang-format/clang-format-diff.py -o ${REPO_ROOT}/clang-format-diff.py"
+      echo "You should make sure the downloaded script is not compromised."
       echo "You can download clang-format by running:"
       echo "    brew install clang-format"
       echo "  Or"
       echo "    apt install clang-format"
       echo "  This might work too:"
       echo "    yum install git-clang-format"
-      echo "Then, move both files (i.e. ${CLANG_FORMAT_DIFF} and clang-format) to some directory within PATH=${PATH}"
-      echo "and make sure ${CLANG_FORMAT_DIFF} is executable."
+      echo "Then make sure clang-format is available and executable from \$PATH:"
+      echo "    clang-format --version"
       exit 128
     fi
     # Check argparse pre-req on interpreter, or it will fail
@@ -75,17 +76,16 @@ else
       exit 129
     fi
     # Unfortunately, some machines have a Python2 clang-format-diff.py
-    # installed but only a Python3 interpreter installed. Rather than trying
-    # different Python versions that might be installed, we can try migrating
-    # the code to Python3 if it looks like Python2
+    # installed but only a Python3 interpreter installed. Unfortunately,
+    # automatic 2to3 migration is insufficient, so suggest downloading latest.
     if grep -q "print '" "$CFD_PATH" && \
        ${PYTHON:-python3} --version | grep -q 'ython 3'; then
-      if [ ! -f "$REPO_ROOT/.py3/clang-format-diff.py" ]; then
-        echo "Migrating $CFD_PATH to Python3 in a hidden file"
-        mkdir -p "$REPO_ROOT/.py3"
-        ${PYTHON:-python3} -m lib2to3 -w -n -o "$REPO_ROOT/.py3" "$CFD_PATH" > /dev/null || exit 128
-      fi
-      CFD_PATH="$REPO_ROOT/.py3/clang-format-diff.py"
+      echo "You have clang-format-diff.py for Python 2 but are using a Python 3"
+      echo "interpreter (${PYTHON:-python3})."
+      echo "You can download clang-format-diff.py for Python 3 by running: "
+      echo "    curl --location https://raw.githubusercontent.com/llvm/llvm-project/main/clang/tools/clang-format/clang-format-diff.py -o ${REPO_ROOT}/clang-format-diff.py"
+      echo "You should make sure the downloaded script is not compromised."
+      exit 130
     fi
     CLANG_FORMAT_DIFF="${PYTHON:-python3} $CFD_PATH"
     # This had better work after all those checks
@@ -136,9 +136,11 @@ then
   FORMAT_UPSTREAM_MERGE_BASE="$(git merge-base "$FORMAT_UPSTREAM" HEAD)"
   # Get the differences
   diffs=$(git diff -U0 "$FORMAT_UPSTREAM_MERGE_BASE" | $CLANG_FORMAT_DIFF -p 1)
+  echo "Checking format of changes not yet in $FORMAT_UPSTREAM..."
 else
   # Check the format of uncommitted lines,
   diffs=$(git diff -U0 HEAD | $CLANG_FORMAT_DIFF -p 1)
+  echo "Checking format of uncommitted changes..."
 fi
 
 if [ -z "$diffs" ]

build_tools/rocksdb-lego-determinator

@@ -3,7 +3,7 @@
 # to determine next steps to run
 
 # Usage:
-# EMAIL=<email> ONCALL=<email> TRIGGER=<trigger> SUBSCRIBER=<email> rocks_ci.py <test-name>
+# EMAIL=<email> ONCALL=<email> TRIGGER=<trigger> SUBSCRIBER=<email> WORKINGDIR=<working_dir> rocksdb-lego-determinator <test-name>
 #
 # Input         Value
 # -------------------------------------------------------------------------
@@ -11,7 +11,7 @@
 # ONCALL        Email address to raise a task on failure
 # TRIGGER       Trigger conditions for email. Valid values are fail, warn, all
 # SUBSCRIBER    Email addresss to add as subscriber for task
-#
+# WORKINGDIR    Working directory
 
 #
 # Report configuration
@@ -53,13 +53,19 @@ if [[ ! -z $REPORT_EMAIL || ! -z $CREATE_TASK ]]; then
   ]"
 fi
 
+# Working directory for the following command, default to current directory
+WORKING_DIR=.
+if [ ! -z $WORKINGDIR ]; then
+  WORKING_DIR=$WORKINGDIR
+fi
+
 #
 # Helper variables
 #
 CLEANUP_ENV="
 {
     'name':'Cleanup environment',
-    'shell':'rm -rf /dev/shm/rocksdb && mkdir /dev/shm/rocksdb && (chmod +t /dev/shm || true)  && make clean',
+    'shell':'cd $WORKING_DIR; rm -rf /dev/shm/rocksdb && mkdir /dev/shm/rocksdb && (chmod +t /dev/shm || true) && make clean',
     'user':'root'
 }"
 
@@ -112,6 +118,7 @@ SETUP_JAVA_ENV="export $HTTP_PROXY; export JAVA_HOME=/usr/local/jdk-8u60-64/; ex
 PARSER="'parser':'python build_tools/error_filter.py $1'"
 
 CONTRUN_NAME="ROCKSDB_CONTRUN_NAME"
+SKIP_FORMAT_CHECKS="SKIP_FORMAT_BUCK_CHECKS=1"
 
 # This code is getting called under various scenarios. What we care about is to
 # understand when it's called from nightly contruns because in that case we'll
@@ -153,7 +160,7 @@ UNIT_TEST_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build and test RocksDB debug version',
-                'shell':'$SHM $DEBUG make $PARALLELISM check || $CONTRUN_NAME=check $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $SKIP_FORMAT_CHECKS make $PARALLELISM check || $CONTRUN_NAME=check $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -176,7 +183,7 @@ UNIT_TEST_NON_SHM_COMMANDS="[
             {
                 'name':'Build and test RocksDB debug version',
                 'timeout': 86400,
-                'shell':'$NON_SHM $DEBUG make $PARALLELISM check || $CONTRUN_NAME=non_shm_check $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $NON_SHM $DEBUG $SKIP_FORMAT_CHECKS make $PARALLELISM check || $CONTRUN_NAME=non_shm_check $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -197,7 +204,7 @@ RELEASE_BUILD_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build RocksDB release',
-                'shell':'make $PARALLEL_j release || $CONTRUN_NAME=release $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; make $PARALLEL_j release || $CONTRUN_NAME=release $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -218,7 +225,7 @@ UNIT_TEST_COMMANDS_481="[
             $CLEANUP_ENV,
             {
                 'name':'Build and test RocksDB debug version',
-                'shell':'$SHM $GCC_481 $DEBUG make $PARALLELISM check || $CONTRUN_NAME=unit_gcc_481_check $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $GCC_481 $DEBUG $SKIP_FORMAT_CHECKS make $PARALLELISM check || $CONTRUN_NAME=unit_gcc_481_check $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -239,7 +246,7 @@ RELEASE_BUILD_COMMANDS_481="[
             $CLEANUP_ENV,
             {
                 'name':'Build RocksDB release on GCC 4.8.1',
-                'shell':'$GCC_481 make $PARALLEL_j release || $CONTRUN_NAME=release_gcc481 $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $GCC_481 make $PARALLEL_j release || $CONTRUN_NAME=release_gcc481 $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -260,7 +267,7 @@ CLANG_UNIT_TEST_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build and test RocksDB debug',
-                'shell':'$CLANG $SHM $DEBUG make $PARALLELISM check || $CONTRUN_NAME=clang_check $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $CLANG $SHM $DEBUG $SKIP_FORMAT_CHECKS make $PARALLELISM check || $CONTRUN_NAME=clang_check $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -281,7 +288,7 @@ CLANG_RELEASE_BUILD_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build RocksDB release',
-                'shell':'$CLANG make $PARALLEL_j release|| $CONTRUN_NAME=clang_release $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $CLANG make $PARALLEL_j release|| $CONTRUN_NAME=clang_release $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -302,7 +309,7 @@ CLANG_ANALYZE_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'RocksDB build and analyze',
-                'shell':'$CLANG $SHM $DEBUG make $PARALLEL_j analyze || $CONTRUN_NAME=clang_analyze $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $CLANG $SHM $DEBUG make $PARALLEL_j analyze || $CONTRUN_NAME=clang_analyze $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -323,7 +330,7 @@ CODE_COV_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build, test and collect code coverage info',
-                'shell':'$SHM $DEBUG make $PARALLELISM coverage || $CONTRUN_NAME=coverage $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $SKIP_FORMAT_CHECKS make $PARALLELISM coverage || $CONTRUN_NAME=coverage $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -344,7 +351,7 @@ UNITY_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build, test unity test',
-                'shell':'$SHM $DEBUG V=1 make J=1 unity_test || $CONTRUN_NAME=unity_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG V=1 make J=1 unity_test || $CONTRUN_NAME=unity_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -365,7 +372,7 @@ 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':'cd $WORKING_DIR; $SKIP_FORMAT_CHECKS make J=1 LITE=1 all check || $CONTRUN_NAME=lite $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -385,7 +392,7 @@ REPORT_LITE_BINARY_SIZE_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Report RocksDB Lite binary size to scuba',
-                'shell':'tools/report_lite_binary_size.sh',
+                'shell':'cd $WORKING_DIR; tools/report_lite_binary_size.sh',
                 'user':'root',
             },
         ],
@@ -404,14 +411,14 @@ STRESS_CRASH_TEST_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build and run RocksDB debug stress tests',
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
             {
                 'name':'Build and run RocksDB debug crash tests',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 crash_test || $CONTRUN_NAME=crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH make J=1 crash_test || $CONTRUN_NAME=crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -434,14 +441,14 @@ BLACKBOX_STRESS_CRASH_TEST_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build and run RocksDB debug stress tests',
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
             {
                 'name':'Build and run RocksDB debug blackbox crash tests',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 blackbox_crash_test || $CONTRUN_NAME=blackbox_crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH make J=1 blackbox_crash_test || $CONTRUN_NAME=blackbox_crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -464,14 +471,14 @@ WHITEBOX_STRESS_CRASH_TEST_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build and run RocksDB debug stress tests',
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
             {
                 'name':'Build and run RocksDB debug whitebox crash tests',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 whitebox_crash_test || $CONTRUN_NAME=whitebox_crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH make J=1 whitebox_crash_test || $CONTRUN_NAME=whitebox_crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -494,14 +501,14 @@ STRESS_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build and run RocksDB debug stress tests',
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $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',
+                'shell':'cd $WORKING_DIR; $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
             },
@@ -524,14 +531,44 @@ STRESS_CRASH_TEST_WITH_TXN_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build and run RocksDB debug stress tests',
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
             {
                 'name':'Build and run RocksDB debug crash tests with txn',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 crash_test_with_txn || $CONTRUN_NAME=crash_test_with_txn $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH make J=1 crash_test_with_txn || $CONTRUN_NAME=crash_test_with_txn $TASK_CREATION_TOOL',
+                'user':'root',
+                $PARSER
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
+#
+# RocksDB stress/crash test with timestamp
+#
+STRESS_CRASH_TEST_WITH_TS_COMMANDS="[
+    {
+        'name':'Rocksdb Stress and Crash Test with ts',
+        'oncall':'$ONCALL',
+        'executeLocal': 'true',
+        'timeout': 86400,
+        'steps': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug stress tests',
+                'shell':'cd $WORKING_DIR; $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 ts',
+                'timeout': 86400,
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH make J=1 crash_test_with_ts || $CONTRUN_NAME=crash_test_with_ts $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -553,7 +590,7 @@ WRITE_STRESS_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build and run RocksDB write stress tests',
-                'shell':'make write_stress && python tools/write_stress_runner.py --runtime_sec=3600 --db=/tmp/rocksdb_write_stress || $CONTRUN_NAME=write_stress $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; make write_stress && python tools/write_stress_runner.py --runtime_sec=3600 --db=/tmp/rocksdb_write_stress || $CONTRUN_NAME=write_stress $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             }
@@ -576,7 +613,7 @@ ASAN_TEST_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Test RocksDB debug under ASAN',
-'shell':'set -o pipefail && ($SHM $ASAN $DEBUG make $PARALLELISM asan_check || $CONTRUN_NAME=asan_check $TASK_CREATION_TOOL) |& /usr/facebook/ops/scripts/asan_symbolize.py -d',
+'shell':'cd $WORKING_DIR; set -o pipefail && ($SHM $ASAN $DEBUG $SKIP_FORMAT_CHECKS make $PARALLELISM asan_check || $CONTRUN_NAME=asan_check $TASK_CREATION_TOOL) |& /usr/facebook/ops/scripts/asan_symbolize.py -d',
                 'user':'root',
                 $PARSER
             }
@@ -599,7 +636,7 @@ ASAN_CRASH_TEST_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug asan_crash_test',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 asan_crash_test || $CONTRUN_NAME=asan_crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH $SKIP_FORMAT_CHECKS make J=1 asan_crash_test || $CONTRUN_NAME=asan_crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -623,7 +660,7 @@ ASAN_BLACKBOX_CRASH_TEST_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug blackbox asan_crash_test',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 blackbox_asan_crash_test || $CONTRUN_NAME=blackbox_asan_crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH $SKIP_FORMAT_CHECKS make J=1 blackbox_asan_crash_test || $CONTRUN_NAME=blackbox_asan_crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -647,7 +684,7 @@ ASAN_WHITEBOX_CRASH_TEST_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug whitebox asan_crash_test',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 whitebox_asan_crash_test || $CONTRUN_NAME=whitebox_asan_crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH $SKIP_FORMAT_CHECKS make J=1 whitebox_asan_crash_test || $CONTRUN_NAME=whitebox_asan_crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -671,7 +708,7 @@ ASAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug asan_crash_test_with_atomic_flush',
                 'timeout': 86400,
-                'shell':'$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',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH $SKIP_FORMAT_CHECKS make J=1 asan_crash_test_with_atomic_flush || $CONTRUN_NAME=asan_crash_test_with_atomic_flush $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -695,7 +732,7 @@ ASAN_CRASH_TEST_WITH_TXN_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug asan_crash_test_with_txn',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 asan_crash_test_with_txn || $CONTRUN_NAME=asan_crash_test_with_txn $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH $SKIP_FORMAT_CHECKS make J=1 asan_crash_test_with_txn || $CONTRUN_NAME=asan_crash_test_with_txn $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -717,7 +754,7 @@ UBSAN_TEST_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Test RocksDB debug under UBSAN',
-                'shell':'set -o pipefail && $SHM $UBSAN $CLANG $DEBUG make $PARALLELISM ubsan_check || $CONTRUN_NAME=ubsan_check $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; set -o pipefail && $SHM $UBSAN $CLANG $DEBUG $SKIP_FORMAT_CHECKS make $PARALLELISM ubsan_check || $CONTRUN_NAME=ubsan_check $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             }
@@ -740,7 +777,7 @@ UBSAN_CRASH_TEST_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug ubsan_crash_test',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH $CLANG make J=1 ubsan_crash_test || $CONTRUN_NAME=ubsan_crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH $CLANG $SKIP_FORMAT_CHECKS make J=1 ubsan_crash_test || $CONTRUN_NAME=ubsan_crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -764,7 +801,7 @@ UBSAN_BLACKBOX_CRASH_TEST_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug blackbox ubsan_crash_test',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH $CLANG make J=1 blackbox_ubsan_crash_test || $CONTRUN_NAME=blackbox_ubsan_crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH $CLANG $SKIP_FORMAT_CHECKS make J=1 blackbox_ubsan_crash_test || $CONTRUN_NAME=blackbox_ubsan_crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -788,7 +825,7 @@ UBSAN_WHITEBOX_CRASH_TEST_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug whitebox ubsan_crash_test',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH $CLANG make J=1 whitebox_ubsan_crash_test || $CONTRUN_NAME=whitebox_ubsan_crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH $CLANG $SKIP_FORMAT_CHECKS make J=1 whitebox_ubsan_crash_test || $CONTRUN_NAME=whitebox_ubsan_crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -812,7 +849,7 @@ UBSAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug ubsan_crash_test_with_atomic_flush',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH $CLANG make J=1 ubsan_crash_test_with_atomic_flush || $CONTRUN_NAME=ubsan_crash_test_with_atomic_flush $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH $CLANG $SKIP_FORMAT_CHECKS make J=1 ubsan_crash_test_with_atomic_flush || $CONTRUN_NAME=ubsan_crash_test_with_atomic_flush $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -836,7 +873,7 @@ UBSAN_CRASH_TEST_WITH_TXN_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug ubsan_crash_test_with_txn',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH $CLANG make J=1 ubsan_crash_test_with_txn || $CONTRUN_NAME=ubsan_crash_test_with_txn $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG $NON_TSAN_CRASH $CLANG $SKIP_FORMAT_CHECKS make J=1 ubsan_crash_test_with_txn || $CONTRUN_NAME=ubsan_crash_test_with_txn $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -860,7 +897,7 @@ VALGRIND_TEST_COMMANDS="[
             {
                 'name':'Run RocksDB debug unit tests',
                 'timeout': 86400,
-                'shell':'$SHM $DEBUG make $PARALLELISM valgrind_test || $CONTRUN_NAME=valgrind_check $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SHM $DEBUG make $PARALLELISM valgrind_test || $CONTRUN_NAME=valgrind_check $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -883,7 +920,7 @@ TSAN_UNIT_TEST_COMMANDS="[
             {
                 'name':'Run RocksDB debug unit test',
                 'timeout': 86400,
-                'shell':'set -o pipefail && $SHM $DEBUG $TSAN make $PARALLELISM check || $CONTRUN_NAME=tsan_check $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; set -o pipefail && $SHM $DEBUG $TSAN $SKIP_FORMAT_CHECKS make $PARALLELISM check || $CONTRUN_NAME=tsan_check $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -906,7 +943,7 @@ TSAN_CRASH_TEST_COMMANDS="[
             {
                 '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 || $CONTRUN_NAME=tsan_crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; set -o pipefail && $SHM $DEBUG $TSAN $TSAN_CRASH CRASH_TEST_KILL_ODD=1887 make J=1 crash_test || $CONTRUN_NAME=tsan_crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -930,7 +967,7 @@ TSAN_BLACKBOX_CRASH_TEST_COMMANDS="[
             {
                 'name':'Compile and run',
                 'timeout': 86400,
-                'shell':'set -o pipefail && $SHM $DEBUG $TSAN $TSAN_CRASH CRASH_TEST_KILL_ODD=1887 make J=1 blackbox_crash_test || $CONTRUN_NAME=tsan_blackbox_crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; set -o pipefail && $SHM $DEBUG $TSAN $TSAN_CRASH CRASH_TEST_KILL_ODD=1887 make J=1 blackbox_crash_test || $CONTRUN_NAME=tsan_blackbox_crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -954,7 +991,7 @@ TSAN_WHITEBOX_CRASH_TEST_COMMANDS="[
             {
                 'name':'Compile and run',
                 'timeout': 86400,
-                'shell':'set -o pipefail && $SHM $DEBUG $TSAN $TSAN_CRASH CRASH_TEST_KILL_ODD=1887 make J=1 whitebox_crash_test || $CONTRUN_NAME=tsan_whitebox_crash_test $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; set -o pipefail && $SHM $DEBUG $TSAN $TSAN_CRASH CRASH_TEST_KILL_ODD=1887 make J=1 whitebox_crash_test || $CONTRUN_NAME=tsan_whitebox_crash_test $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -978,7 +1015,7 @@ TSAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
             {
                 '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',
+                'shell':'cd $WORKING_DIR; 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
             },
@@ -1002,7 +1039,7 @@ TSAN_CRASH_TEST_WITH_TXN_COMMANDS="[
             {
                 '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_txn || $CONTRUN_NAME=tsan_crash_test_with_txn $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; set -o pipefail && $SHM $DEBUG $TSAN $TSAN_CRASH CRASH_TEST_KILL_ODD=1887 make J=1 crash_test_with_txn || $CONTRUN_NAME=tsan_crash_test_with_txn $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -1036,7 +1073,7 @@ FORMAT_COMPATIBLE_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Run RocksDB debug unit test',
-                'shell':'build_tools/rocksdb-lego-determinator run_format_compatible || $CONTRUN_NAME=run_format_compatible $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; build_tools/rocksdb-lego-determinator run_format_compatible || $CONTRUN_NAME=run_format_compatible $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -1058,6 +1095,7 @@ run_no_compression()
   mv .tmp.fbcode_config.sh build_tools/fbcode_config.sh
   cat Makefile | grep -v tools/ldb_test.py > .tmp.Makefile
   mv .tmp.Makefile Makefile
+  export $SKIP_FORMAT_CHECKS
   make $DEBUG J=1 check
 }
 
@@ -1070,7 +1108,7 @@ NO_COMPRESSION_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Run RocksDB debug unit test',
-                'shell':'build_tools/rocksdb-lego-determinator run_no_compression || $CONTRUN_NAME=run_no_compression $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; build_tools/rocksdb-lego-determinator run_no_compression || $CONTRUN_NAME=run_no_compression $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -1129,7 +1167,7 @@ REGRESSION_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Make and run script',
-                'shell':'build_tools/rocksdb-lego-determinator run_regression || $CONTRUN_NAME=run_regression $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; build_tools/rocksdb-lego-determinator run_regression || $CONTRUN_NAME=run_regression $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -1150,7 +1188,7 @@ JAVA_BUILD_TEST_COMMANDS="[
             $CLEANUP_ENV,
             {
                 'name':'Build RocksDB for Java',
-                'shell':'$SETUP_JAVA_ENV; $SHM make rocksdbjava || $CONTRUN_NAME=rocksdbjava $TASK_CREATION_TOOL',
+                'shell':'cd $WORKING_DIR; $SETUP_JAVA_ENV; $SHM make rocksdbjava || $CONTRUN_NAME=rocksdbjava $TASK_CREATION_TOOL',
                 'user':'root',
                 $PARSER
             },
@@ -1212,6 +1250,9 @@ case $1 in
   stress_crash_with_txn)
     echo $STRESS_CRASH_TEST_WITH_TXN_COMMANDS
     ;;
+  stress_crash_with_ts)
+    echo $STRESS_CRASH_TEST_WITH_TS_COMMANDS
+    ;;
   write_stress)
     echo $WRITE_STRESS_COMMANDS
     ;;

build_tools/run_ci_db_test.ps1

@@ -68,7 +68,7 @@ $BinariesFolder = -Join($RootFolder, "\build\Debug\")
 
 if($WorkFolder -eq "") {
 
-    # If TEST_TMPDIR is set use it    
+    # If TEST_TMPDIR is set use it
     [string]$var = $Env:TEST_TMPDIR
     if($var -eq "") {
         $WorkFolder = -Join($RootFolder, "\db_tests\")
@@ -93,7 +93,7 @@ $ExcludeCasesSet = New-Object System.Collections.Generic.HashSet[string]
 if($ExcludeCases -ne "") {
     Write-Host "ExcludeCases: $ExcludeCases"
     $l = $ExcludeCases -split ' '
-    ForEach($t in $l) { 
+    ForEach($t in $l) {
       $ExcludeCasesSet.Add($t) | Out-Null
     }
 }
@@ -102,7 +102,7 @@ $ExcludeExesSet = New-Object System.Collections.Generic.HashSet[string]
 if($ExcludeExes -ne "") {
     Write-Host "ExcludeExe: $ExcludeExes"
     $l = $ExcludeExes -split ' '
-    ForEach($t in $l) { 
+    ForEach($t in $l) {
       $ExcludeExesSet.Add($t) | Out-Null
     }
 }
@@ -118,6 +118,10 @@ if($ExcludeExes -ne "") {
 #   MultiThreaded/MultiThreadedDBTest.
 #     MultiThreaded/0  # GetParam() = 0
 #     MultiThreaded/1  # GetParam() = 1
+#   RibbonTypeParamTest/0.  # TypeParam = struct DefaultTypesAndSettings
+#     CompactnessAndBacktrackAndFpRate
+#     Extremes
+#     FindOccupancyForSuccessRate
 #
 # into this:
 #
@@ -125,6 +129,9 @@ if($ExcludeExes -ne "") {
 #   DBTest.WriteEmptyBatch
 #   MultiThreaded/MultiThreadedDBTest.MultiThreaded/0
 #   MultiThreaded/MultiThreadedDBTest.MultiThreaded/1
+#   RibbonTypeParamTest/0.CompactnessAndBacktrackAndFpRate
+#   RibbonTypeParamTest/0.Extremes
+#   RibbonTypeParamTest/0.FindOccupancyForSuccessRate
 #
 # Output into the parameter in a form TestName -> Log File Name
 function ExtractTestCases([string]$GTestExe, $HashTable) {
@@ -138,6 +145,8 @@ function ExtractTestCases([string]$GTestExe, $HashTable) {
 
     ForEach( $l in $Tests) {
 
+      # remove trailing comment if any
+      $l = $l -replace '\s+\#.*',''
       # Leading whitespace is fine
       $l = $l -replace '^\s+',''
       # Trailing dot is a test group but no whitespace
@@ -146,8 +155,7 @@ function ExtractTestCases([string]$GTestExe, $HashTable) {
       }  else {
         # Otherwise it is a test name, remove leading space
         $test = $l
-        # remove trailing comment if any and create a log name
-        $test = $test -replace '\s+\#.*',''
+        # create a log name
         $test = "$Group$test"
 
         if($ExcludeCasesSet.Contains($test)) {
@@ -253,7 +261,7 @@ if($Run -ne "") {
 
   $DiscoveredExe = @()
   dir -Path $search_path | ForEach-Object {
-     $DiscoveredExe += ($_.Name)     
+     $DiscoveredExe += ($_.Name)
   }
 
   # Remove exclusions
@@ -293,7 +301,7 @@ if($SuiteRun -ne "") {
 
   $ListOfExe = @()
   dir -Path $search_path | ForEach-Object {
-     $ListOfExe += ($_.Name)     
+     $ListOfExe += ($_.Name)
   }
 
   # Exclude those in RunOnly from running as suites
@@ -348,7 +356,7 @@ function RunJobs($Suites, $TestCmds, [int]$ConcurrencyVal)
 
     # Wait for all to finish and get the results
     while(($JobToLog.Count -gt 0) -or
-          ($TestCmds.Count -gt 0) -or 
+          ($TestCmds.Count -gt 0) -or
            ($Suites.Count -gt 0)) {
 
         # Make sure we have maximum concurrent jobs running if anything
@@ -468,8 +476,8 @@ RunJobs -Suites $CasesToRun -TestCmds $TestExes -ConcurrencyVal $Concurrency
 
 $EndDate = (Get-Date)
 
-New-TimeSpan -Start $StartDate -End $EndDate | 
-  ForEach-Object { 
+New-TimeSpan -Start $StartDate -End $EndDate |
+  ForEach-Object {
     "Elapsed time: {0:g}" -f $_
   }
 
@@ -484,4 +492,4 @@ if(!$script:success) {
 
  exit 0
 
- 
+

cache/cache_bench.cc

@@ -1,7 +1,11 @@
-//  Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
+//  Copyright (c) 2013-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.
 
 #ifndef GFLAGS
 #include <cstdio>
@@ -10,372 +14,8 @@ int main() {
   return 1;
 }
 #else
-
-#include <stdio.h>
-#include <sys/types.h>
-#include <cinttypes>
-#include <limits>
-
-#include "port/port.h"
-#include "rocksdb/cache.h"
-#include "rocksdb/db.h"
-#include "rocksdb/env.h"
-#include "util/coding.h"
-#include "util/gflags_compat.h"
-#include "util/hash.h"
-#include "util/mutexlock.h"
-#include "util/random.h"
-
-using GFLAGS_NAMESPACE::ParseCommandLineFlags;
-
-static constexpr uint32_t KiB = uint32_t{1} << 10;
-static constexpr uint32_t MiB = KiB << 10;
-static constexpr uint64_t GiB = MiB << 10;
-
-DEFINE_uint32(threads, 16, "Number of concurrent threads to run.");
-DEFINE_uint64(cache_size, 1 * GiB,
-              "Number of bytes to use as a cache of uncompressed data.");
-DEFINE_uint32(num_shard_bits, 6, "shard_bits.");
-
-DEFINE_double(resident_ratio, 0.25,
-              "Ratio of keys fitting in cache to keyspace.");
-DEFINE_uint64(ops_per_thread, 0,
-              "Number of operations per thread. (Default: 5 * keyspace size)");
-DEFINE_uint32(value_bytes, 8 * KiB, "Size of each value added.");
-
-DEFINE_uint32(skew, 5, "Degree of skew in key selection");
-DEFINE_bool(populate_cache, true, "Populate cache before operations");
-
-DEFINE_uint32(lookup_insert_percent, 87,
-              "Ratio of lookup (+ insert on not found) to total workload "
-              "(expressed as a percentage)");
-DEFINE_uint32(insert_percent, 2,
-              "Ratio of insert to total workload (expressed as a percentage)");
-DEFINE_uint32(lookup_percent, 10,
-              "Ratio of lookup to total workload (expressed as a percentage)");
-DEFINE_uint32(erase_percent, 1,
-              "Ratio of erase to total workload (expressed as a percentage)");
-
-DEFINE_bool(use_clock_cache, false, "");
-
-namespace ROCKSDB_NAMESPACE {
-
-class CacheBench;
-namespace {
-// State shared by all concurrent executions of the same benchmark.
-class SharedState {
- public:
-  explicit SharedState(CacheBench* cache_bench)
-      : cv_(&mu_),
-        num_initialized_(0),
-        start_(false),
-        num_done_(0),
-        cache_bench_(cache_bench) {}
-
-  ~SharedState() {}
-
-  port::Mutex* GetMutex() {
-    return &mu_;
-  }
-
-  port::CondVar* GetCondVar() {
-    return &cv_;
-  }
-
-  CacheBench* GetCacheBench() const {
-    return cache_bench_;
-  }
-
-  void IncInitialized() {
-    num_initialized_++;
-  }
-
-  void IncDone() {
-    num_done_++;
-  }
-
-  bool AllInitialized() const { return num_initialized_ >= FLAGS_threads; }
-
-  bool AllDone() const { return num_done_ >= FLAGS_threads; }
-
-  void SetStart() {
-    start_ = true;
-  }
-
-  bool Started() const {
-    return start_;
-  }
-
- private:
-  port::Mutex mu_;
-  port::CondVar cv_;
-
-  uint64_t num_initialized_;
-  bool start_;
-  uint64_t num_done_;
-
-  CacheBench* cache_bench_;
-};
-
-// Per-thread state for concurrent executions of the same benchmark.
-struct ThreadState {
-  uint32_t tid;
-  Random64 rnd;
-  SharedState* shared;
-
-  ThreadState(uint32_t index, SharedState* _shared)
-      : tid(index), rnd(1000 + index), shared(_shared) {}
-};
-
-struct KeyGen {
-  char key_data[27];
-
-  Slice GetRand(Random64& rnd, uint64_t max_key) {
-    uint64_t raw = rnd.Next();
-    // Skew according to setting
-    for (uint32_t i = 0; i < FLAGS_skew; ++i) {
-      raw = std::min(raw, rnd.Next());
-    }
-    uint64_t key = FastRange64(raw, max_key);
-    // Variable size and alignment
-    size_t off = key % 8;
-    key_data[0] = char{42};
-    EncodeFixed64(key_data + 1, key);
-    key_data[9] = char{11};
-    EncodeFixed64(key_data + 10, key);
-    key_data[18] = char{4};
-    EncodeFixed64(key_data + 19, key);
-    return Slice(&key_data[off], sizeof(key_data) - off);
-  }
-};
-
-char* createValue(Random64& rnd) {
-  char* rv = new char[FLAGS_value_bytes];
-  // Fill with some filler data, and take some CPU time
-  for (uint32_t i = 0; i < FLAGS_value_bytes; i += 8) {
-    EncodeFixed64(rv + i, rnd.Next());
-  }
-  return rv;
-}
-
-void deleter(const Slice& /*key*/, void* value) {
-  delete[] static_cast<char*>(value);
-}
-}  // namespace
-
-class CacheBench {
-  static constexpr uint64_t kHundredthUint64 =
-      std::numeric_limits<uint64_t>::max() / 100U;
-
- public:
-  CacheBench()
-      : max_key_(static_cast<uint64_t>(FLAGS_cache_size / FLAGS_resident_ratio /
-                                       FLAGS_value_bytes)),
-        lookup_insert_threshold_(kHundredthUint64 *
-                                 FLAGS_lookup_insert_percent),
-        insert_threshold_(lookup_insert_threshold_ +
-                          kHundredthUint64 * FLAGS_insert_percent),
-        lookup_threshold_(insert_threshold_ +
-                          kHundredthUint64 * FLAGS_lookup_percent),
-        erase_threshold_(lookup_threshold_ +
-                         kHundredthUint64 * FLAGS_erase_percent) {
-    if (erase_threshold_ != 100U * kHundredthUint64) {
-      fprintf(stderr, "Percentages must add to 100.\n");
-      exit(1);
-    }
-    if (FLAGS_use_clock_cache) {
-      cache_ = NewClockCache(FLAGS_cache_size, FLAGS_num_shard_bits);
-      if (!cache_) {
-        fprintf(stderr, "Clock cache not supported.\n");
-        exit(1);
-      }
-    } else {
-      cache_ = NewLRUCache(FLAGS_cache_size, FLAGS_num_shard_bits);
-    }
-    if (FLAGS_ops_per_thread == 0) {
-      FLAGS_ops_per_thread = 5 * max_key_;
-    }
-  }
-
-  ~CacheBench() {}
-
-  void PopulateCache() {
-    Random64 rnd(1);
-    KeyGen keygen;
-    for (uint64_t i = 0; i < 2 * FLAGS_cache_size; i += FLAGS_value_bytes) {
-      cache_->Insert(keygen.GetRand(rnd, max_key_), createValue(rnd),
-                     FLAGS_value_bytes, &deleter);
-    }
-  }
-
-  bool Run() {
-    ROCKSDB_NAMESPACE::Env* env = ROCKSDB_NAMESPACE::Env::Default();
-
-    PrintEnv();
-    SharedState shared(this);
-    std::vector<std::unique_ptr<ThreadState> > threads(FLAGS_threads);
-    for (uint32_t i = 0; i < FLAGS_threads; i++) {
-      threads[i].reset(new ThreadState(i, &shared));
-      env->StartThread(ThreadBody, threads[i].get());
-    }
-    {
-      MutexLock l(shared.GetMutex());
-      while (!shared.AllInitialized()) {
-        shared.GetCondVar()->Wait();
-      }
-      // Record start time
-      uint64_t start_time = env->NowMicros();
-
-      // Start all threads
-      shared.SetStart();
-      shared.GetCondVar()->SignalAll();
-
-      // Wait threads to complete
-      while (!shared.AllDone()) {
-        shared.GetCondVar()->Wait();
-      }
-
-      // Record end time
-      uint64_t end_time = env->NowMicros();
-      double elapsed = static_cast<double>(end_time - start_time) * 1e-6;
-      uint32_t qps = static_cast<uint32_t>(
-          static_cast<double>(FLAGS_threads * FLAGS_ops_per_thread) / elapsed);
-      fprintf(stdout, "Complete in %.3f s; QPS = %u\n", elapsed, qps);
-    }
-    return true;
-  }
-
- private:
-  std::shared_ptr<Cache> cache_;
-  const uint64_t max_key_;
-  // Cumulative thresholds in the space of a random uint64_t
-  const uint64_t lookup_insert_threshold_;
-  const uint64_t insert_threshold_;
-  const uint64_t lookup_threshold_;
-  const uint64_t erase_threshold_;
-
-  static void ThreadBody(void* v) {
-    ThreadState* thread = static_cast<ThreadState*>(v);
-    SharedState* shared = thread->shared;
-
-    {
-      MutexLock l(shared->GetMutex());
-      shared->IncInitialized();
-      if (shared->AllInitialized()) {
-        shared->GetCondVar()->SignalAll();
-      }
-      while (!shared->Started()) {
-        shared->GetCondVar()->Wait();
-      }
-    }
-    thread->shared->GetCacheBench()->OperateCache(thread);
-
-    {
-      MutexLock l(shared->GetMutex());
-      shared->IncDone();
-      if (shared->AllDone()) {
-        shared->GetCondVar()->SignalAll();
-      }
-    }
-  }
-
-  void OperateCache(ThreadState* thread) {
-    // To use looked-up values
-    uint64_t result = 0;
-    // To hold handles for a non-trivial amount of time
-    Cache::Handle* handle = nullptr;
-    KeyGen gen;
-    for (uint64_t i = 0; i < FLAGS_ops_per_thread; i++) {
-      Slice key = gen.GetRand(thread->rnd, max_key_);
-      uint64_t random_op = thread->rnd.Next();
-      if (random_op < lookup_insert_threshold_) {
-        if (handle) {
-          cache_->Release(handle);
-          handle = nullptr;
-        }
-        // do lookup
-        handle = cache_->Lookup(key);
-        if (handle) {
-          // do something with the data
-          result += NPHash64(static_cast<char*>(cache_->Value(handle)),
-                             FLAGS_value_bytes);
-        } else {
-          // do insert
-          cache_->Insert(key, createValue(thread->rnd), FLAGS_value_bytes,
-                         &deleter, &handle);
-        }
-      } else if (random_op < insert_threshold_) {
-        if (handle) {
-          cache_->Release(handle);
-          handle = nullptr;
-        }
-        // do insert
-        cache_->Insert(key, createValue(thread->rnd), FLAGS_value_bytes,
-                       &deleter, &handle);
-      } else if (random_op < lookup_threshold_) {
-        if (handle) {
-          cache_->Release(handle);
-          handle = nullptr;
-        }
-        // do lookup
-        handle = cache_->Lookup(key);
-        if (handle) {
-          // do something with the data
-          result += NPHash64(static_cast<char*>(cache_->Value(handle)),
-                             FLAGS_value_bytes);
-        }
-      } else if (random_op < erase_threshold_) {
-        // do erase
-        cache_->Erase(key);
-      } else {
-        // Should be extremely unlikely (noop)
-        assert(random_op >= kHundredthUint64 * 100U);
-      }
-    }
-    if (handle) {
-      cache_->Release(handle);
-      handle = nullptr;
-    }
-  }
-
-  void PrintEnv() const {
-    printf("RocksDB version     : %d.%d\n", kMajorVersion, kMinorVersion);
-    printf("Number of threads   : %u\n", FLAGS_threads);
-    printf("Ops per thread      : %" PRIu64 "\n", FLAGS_ops_per_thread);
-    printf("Cache size          : %" PRIu64 "\n", FLAGS_cache_size);
-    printf("Num shard bits      : %u\n", FLAGS_num_shard_bits);
-    printf("Max key             : %" PRIu64 "\n", max_key_);
-    printf("Resident ratio      : %g\n", FLAGS_resident_ratio);
-    printf("Skew degree         : %u\n", FLAGS_skew);
-    printf("Populate cache      : %d\n", int{FLAGS_populate_cache});
-    printf("Lookup+Insert pct   : %u%%\n", FLAGS_lookup_insert_percent);
-    printf("Insert percentage   : %u%%\n", FLAGS_insert_percent);
-    printf("Lookup percentage   : %u%%\n", FLAGS_lookup_percent);
-    printf("Erase percentage    : %u%%\n", FLAGS_erase_percent);
-    printf("----------------------------\n");
-  }
-};
-}  // namespace ROCKSDB_NAMESPACE
-
+#include <rocksdb/cache_bench_tool.h>
 int main(int argc, char** argv) {
-  ParseCommandLineFlags(&argc, &argv, true);
-
-  if (FLAGS_threads <= 0) {
-    fprintf(stderr, "threads number <= 0\n");
-    exit(1);
-  }
-
-  ROCKSDB_NAMESPACE::CacheBench bench;
-  if (FLAGS_populate_cache) {
-    bench.PopulateCache();
-    printf("Population complete\n");
-    printf("----------------------------\n");
-  }
-  if (bench.Run()) {
-    return 0;
-  } else {
-    return 1;
-  }
+  return ROCKSDB_NAMESPACE::cache_bench_tool(argc, argv);
 }
-
 #endif  // GFLAGS

cache/cache_bench_tool.cc

@@ -0,0 +1,504 @@
+//  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).
+
+#ifdef GFLAGS
+#include "rocksdb/cache_bench_tool.h"
+
+#include <sys/types.h>
+
+#include <cinttypes>
+#include <cstdio>
+#include <limits>
+
+#include "options/configurable_helper.h"
+#include "port/port.h"
+#include "rocksdb/cache.h"
+#include "rocksdb/db.h"
+#include "rocksdb/env.h"
+#include "rocksdb/system_clock.h"
+#include "rocksdb/tiered_cache.h"
+#include "rocksdb/utilities/object_registry.h"
+#include "rocksdb/utilities/options_type.h"
+#include "util/coding.h"
+#include "util/gflags_compat.h"
+#include "util/hash.h"
+#include "util/mutexlock.h"
+#include "util/random.h"
+#include "util/stop_watch.h"
+
+using GFLAGS_NAMESPACE::ParseCommandLineFlags;
+
+static constexpr uint32_t KiB = uint32_t{1} << 10;
+static constexpr uint32_t MiB = KiB << 10;
+static constexpr uint64_t GiB = MiB << 10;
+
+DEFINE_uint32(threads, 16, "Number of concurrent threads to run.");
+DEFINE_uint64(cache_size, 1 * GiB,
+              "Number of bytes to use as a cache of uncompressed data.");
+DEFINE_uint32(num_shard_bits, 6, "shard_bits.");
+
+DEFINE_double(resident_ratio, 0.25,
+              "Ratio of keys fitting in cache to keyspace.");
+DEFINE_uint64(ops_per_thread, 0,
+              "Number of operations per thread. (Default: 5 * keyspace size)");
+DEFINE_uint32(value_bytes, 8 * KiB, "Size of each value added.");
+
+DEFINE_uint32(skew, 5, "Degree of skew in key selection");
+DEFINE_bool(populate_cache, true, "Populate cache before operations");
+
+DEFINE_uint32(lookup_insert_percent, 87,
+              "Ratio of lookup (+ insert on not found) to total workload "
+              "(expressed as a percentage)");
+DEFINE_uint32(insert_percent, 2,
+              "Ratio of insert to total workload (expressed as a percentage)");
+DEFINE_uint32(lookup_percent, 10,
+              "Ratio of lookup to total workload (expressed as a percentage)");
+DEFINE_uint32(erase_percent, 1,
+              "Ratio of erase to total workload (expressed as a percentage)");
+
+DEFINE_bool(use_clock_cache, false, "");
+
+DEFINE_bool(skewed, false, "If true, skew the key access distribution");
+#ifndef ROCKSDB_LITE
+DEFINE_string(tiered_cache_uri, "",
+              "Full URI for creating a custom NVM cache object");
+static class std::shared_ptr<ROCKSDB_NAMESPACE::TieredCache> tiered_cache;
+#endif
+
+namespace ROCKSDB_NAMESPACE {
+class CacheBench;
+
+namespace {
+// State shared by all concurrent executions of the same benchmark.
+class SharedState {
+ public:
+  explicit SharedState(CacheBench* cache_bench)
+      : cv_(&mu_),
+        num_initialized_(0),
+        start_(false),
+        num_done_(0),
+        cache_bench_(cache_bench) {}
+
+  ~SharedState() {}
+
+  port::Mutex* GetMutex() { return &mu_; }
+
+  port::CondVar* GetCondVar() { return &cv_; }
+
+  CacheBench* GetCacheBench() const { return cache_bench_; }
+
+  void IncInitialized() { num_initialized_++; }
+
+  void IncDone() { num_done_++; }
+
+  bool AllInitialized() const { return num_initialized_ >= FLAGS_threads; }
+
+  bool AllDone() const { return num_done_ >= FLAGS_threads; }
+
+  void SetStart() { start_ = true; }
+
+  bool Started() const { return start_; }
+
+ private:
+  port::Mutex mu_;
+  port::CondVar cv_;
+
+  uint64_t num_initialized_;
+  bool start_;
+  uint64_t num_done_;
+
+  CacheBench* cache_bench_;
+};
+
+class Stats {
+ private:
+  uint64_t hits;
+  uint64_t misses;
+  uint64_t lookup_us;
+  uint64_t insert_us;
+
+ public:
+  Stats() : hits(0), misses(0), lookup_us(0), insert_us(0) {}
+
+  void AddHits(uint64_t inc) { hits += inc; }
+  void AddMisses(uint64_t inc) { misses += inc; }
+  void AddLookupUs(uint64_t lookup) { lookup_us += lookup; }
+  void AddInsertUs(uint64_t insert) { insert_us += insert; }
+  void Merge(const Stats& other) {
+    hits += other.hits;
+    misses += other.misses;
+    lookup_us += other.lookup_us;
+    insert_us += other.insert_us;
+  }
+  void Report() {
+    fprintf(stdout, "%llu hits, %llu misses, %llu lookup_us, %llu insert_us\n",
+            (unsigned long long)hits, (unsigned long long)misses,
+            (unsigned long long)lookup_us, (unsigned long long)insert_us);
+    fflush(stdout);
+  }
+};
+
+// Per-thread state for concurrent executions of the same benchmark.
+struct ThreadState {
+  uint32_t tid;
+  Random64 rnd;
+  Stats stats;
+  SharedState* shared;
+
+  ThreadState(uint32_t index, SharedState* _shared)
+      : tid(index), rnd(1000 + index), shared(_shared) {}
+};
+
+struct KeyGen {
+  char key_data[27];
+
+  Slice GetRand(Random64& rnd, uint64_t max_key, int max_log) {
+    uint64_t key = 0;
+    if (!FLAGS_skewed) {
+      uint64_t raw = rnd.Next();
+      // Skew according to setting
+      for (uint32_t i = 0; i < FLAGS_skew; ++i) {
+        raw = std::min(raw, rnd.Next());
+      }
+      key = FastRange64(raw, max_key);
+    } else {
+      key = rnd.Skewed(max_log);
+      if (key > max_key) {
+        key -= max_key;
+      }
+    }
+    // Variable size and alignment
+    size_t off = key % 8;
+    key_data[0] = char{42};
+    EncodeFixed64(key_data + 1, key);
+    key_data[9] = char{11};
+    EncodeFixed64(key_data + 10, key);
+    key_data[18] = char{4};
+    EncodeFixed64(key_data + 19, key);
+    return Slice(&key_data[off], sizeof(key_data) - off);
+  }
+};
+
+char* createValue(Random64& rnd) {
+  char* rv = new char[FLAGS_value_bytes];
+  // Fill with some filler data, and take some CPU time
+  for (uint32_t i = 0; i < FLAGS_value_bytes; i += 8) {
+    EncodeFixed64(rv + i, rnd.Next());
+  }
+  return rv;
+}
+
+void helperCallback(Cache::SizeCallback* size_cb,
+                    Cache::SaveToCallback* save_cb,
+                    Cache::DeletionCallback* del_cb) {
+  if (size_cb) {
+    *size_cb = [](void* /*obj*/) -> size_t { return FLAGS_value_bytes; };
+  }
+  if (save_cb) {
+    *save_cb = [](void* obj, size_t /*offset*/, size_t size,
+                  void* out) -> Status {
+      memcpy(out, obj, size);
+      return Status::OK();
+    };
+  }
+  if (del_cb) {
+    *del_cb = [](const Slice& /*key*/, void* obj) -> void {
+      delete[] static_cast<char*>(obj);
+    };
+  }
+}
+}  // namespace
+
+class CacheBench {
+  static constexpr uint64_t kHundredthUint64 =
+      std::numeric_limits<uint64_t>::max() / 100U;
+
+ public:
+  CacheBench()
+      : max_key_(static_cast<uint64_t>(FLAGS_cache_size / FLAGS_resident_ratio /
+                                       FLAGS_value_bytes)),
+        lookup_insert_threshold_(kHundredthUint64 *
+                                 FLAGS_lookup_insert_percent),
+        insert_threshold_(lookup_insert_threshold_ +
+                          kHundredthUint64 * FLAGS_insert_percent),
+        lookup_threshold_(insert_threshold_ +
+                          kHundredthUint64 * FLAGS_lookup_percent),
+        erase_threshold_(lookup_threshold_ +
+                         kHundredthUint64 * FLAGS_erase_percent),
+        skewed_(FLAGS_skewed) {
+    if (erase_threshold_ != 100U * kHundredthUint64) {
+      fprintf(stderr, "Percentages must add to 100.\n");
+      exit(1);
+    }
+
+    max_log_ = 0;
+    if (skewed_) {
+      uint64_t max_key = max_key_;
+      while (max_key >>= 1) max_log_++;
+      if (max_key > (1u << max_log_)) max_log_++;
+    }
+
+    if (FLAGS_use_clock_cache) {
+      cache_ = NewClockCache(FLAGS_cache_size, FLAGS_num_shard_bits);
+      if (!cache_) {
+        fprintf(stderr, "Clock cache not supported.\n");
+        exit(1);
+      }
+    } else {
+      LRUCacheOptions opts(FLAGS_cache_size, FLAGS_num_shard_bits, false, 0.5);
+#ifndef ROCKSDB_LITE
+      if (!FLAGS_tiered_cache_uri.empty()) {
+        Status s = ObjectRegistry::NewInstance()->NewSharedObject<TieredCache>(
+            FLAGS_tiered_cache_uri, &tiered_cache);
+        if (tiered_cache == nullptr) {
+          fprintf(stderr,
+                  "No tiered cache registered matching string: %s status=%s\n",
+                  FLAGS_tiered_cache_uri.c_str(), s.ToString().c_str());
+          exit(1);
+        }
+        opts.tiered_cache = tiered_cache;
+      }
+#endif
+
+      cache_ = NewLRUCache(opts);
+    }
+    if (FLAGS_ops_per_thread == 0) {
+      FLAGS_ops_per_thread = 5 * max_key_;
+    }
+  }
+
+  ~CacheBench() {}
+
+  void PopulateCache() {
+    Random64 rnd(1);
+    KeyGen keygen;
+    for (uint64_t i = 0; i < 10 * FLAGS_cache_size; i += FLAGS_value_bytes) {
+      cache_->Insert(keygen.GetRand(rnd, max_key_, max_log_), createValue(rnd),
+                     helperCallback, FLAGS_value_bytes);
+    }
+  }
+
+  bool Run() {
+    ROCKSDB_NAMESPACE::Env* env = ROCKSDB_NAMESPACE::Env::Default();
+    const auto& clock = env->GetSystemClock();
+
+    PrintEnv();
+    SharedState shared(this);
+    std::vector<std::unique_ptr<ThreadState> > threads(FLAGS_threads);
+    for (uint32_t i = 0; i < FLAGS_threads; i++) {
+      threads[i].reset(new ThreadState(i, &shared));
+      env->StartThread(ThreadBody, threads[i].get());
+    }
+    {
+      MutexLock l(shared.GetMutex());
+      while (!shared.AllInitialized()) {
+        shared.GetCondVar()->Wait();
+      }
+      // Record start time
+      uint64_t start_time = clock->NowMicros();
+
+      // Start all threads
+      shared.SetStart();
+      shared.GetCondVar()->SignalAll();
+
+      // Wait threads to complete
+      while (!shared.AllDone()) {
+        shared.GetCondVar()->Wait();
+      }
+
+      // Record end time
+      uint64_t end_time = clock->NowMicros();
+      double elapsed = static_cast<double>(end_time - start_time) * 1e-6;
+      uint32_t qps = static_cast<uint32_t>(
+          static_cast<double>(FLAGS_threads * FLAGS_ops_per_thread) / elapsed);
+      fprintf(stdout, "Complete in %.3f s; QPS = %u\n", elapsed, qps);
+
+      Stats merge_stats;
+      for (uint32_t i = 0; i < FLAGS_threads; ++i) {
+        merge_stats.Merge(threads[i]->stats);
+      }
+      merge_stats.Report();
+    }
+    return true;
+  }
+
+ private:
+  std::shared_ptr<Cache> cache_;
+  const uint64_t max_key_;
+  // Cumulative thresholds in the space of a random uint64_t
+  const uint64_t lookup_insert_threshold_;
+  const uint64_t insert_threshold_;
+  const uint64_t lookup_threshold_;
+  const uint64_t erase_threshold_;
+  const bool skewed_;
+  int max_log_;
+
+  static void ThreadBody(void* v) {
+    ThreadState* thread = static_cast<ThreadState*>(v);
+    SharedState* shared = thread->shared;
+
+    {
+      MutexLock l(shared->GetMutex());
+      shared->IncInitialized();
+      if (shared->AllInitialized()) {
+        shared->GetCondVar()->SignalAll();
+      }
+      while (!shared->Started()) {
+        shared->GetCondVar()->Wait();
+      }
+    }
+    thread->shared->GetCacheBench()->OperateCache(thread);
+
+    {
+      MutexLock l(shared->GetMutex());
+      shared->IncDone();
+      if (shared->AllDone()) {
+        shared->GetCondVar()->SignalAll();
+      }
+    }
+  }
+
+  void OperateCache(ThreadState* thread) {
+    // To use looked-up values
+    uint64_t result = 0;
+    // To hold handles for a non-trivial amount of time
+    Cache::Handle* handle = nullptr;
+    KeyGen gen;
+    ROCKSDB_NAMESPACE::Env* env = ROCKSDB_NAMESPACE::Env::Default();
+    SystemClock* clock = env->GetSystemClock().get();
+    for (uint64_t i = 0; i < FLAGS_ops_per_thread; i++) {
+      Slice key = gen.GetRand(thread->rnd, max_key_, max_log_);
+      uint64_t random_op = thread->rnd.Next();
+      Cache::CreateCallback create_cb =
+          [](void* buf, size_t size, void** out_obj, size_t* charge) -> Status {
+        *out_obj = reinterpret_cast<void*>(new char[size]);
+        memcpy(*out_obj, buf, size);
+        *charge = size;
+        return Status::OK();
+      };
+
+      if (random_op < lookup_insert_threshold_) {
+        if (handle) {
+          cache_->Release(handle);
+          handle = nullptr;
+        }
+        // do lookup
+        {
+          uint64_t elapsed = 0;
+          {
+            StopWatch sw(clock, nullptr, 0, &elapsed);
+            handle = cache_->Lookup(key, helperCallback, create_cb,
+                                    Cache::Priority::LOW, true);
+          }
+          thread->stats.AddLookupUs(elapsed);
+        }
+        if (handle) {
+          thread->stats.AddHits(1);
+          // do something with the data
+          result += NPHash64(static_cast<char*>(cache_->Value(handle)),
+                             FLAGS_value_bytes);
+        } else {
+          thread->stats.AddMisses(1);
+          // do insert
+          {
+            uint64_t elapsed = 0;
+            {
+              StopWatch sw(clock, nullptr, 0, &elapsed);
+              cache_->Insert(key, createValue(thread->rnd), helperCallback,
+                             FLAGS_value_bytes, &handle);
+            }
+            thread->stats.AddInsertUs(elapsed);
+          }
+        }
+      } else if (random_op < insert_threshold_) {
+        if (handle) {
+          cache_->Release(handle);
+          handle = nullptr;
+        }
+        // do insert
+        {
+          uint64_t elapsed = 0;
+          {
+            StopWatch sw(clock, nullptr, 0, &elapsed);
+            cache_->Insert(key, createValue(thread->rnd), helperCallback,
+                           FLAGS_value_bytes, &handle);
+          }
+          thread->stats.AddInsertUs(elapsed);
+        }
+      } else if (random_op < lookup_threshold_) {
+        if (handle) {
+          cache_->Release(handle);
+          handle = nullptr;
+        }
+        // do lookup
+        {
+          uint64_t elapsed = 0;
+          {
+            StopWatch sw(clock, nullptr, 0, &elapsed);
+            handle = cache_->Lookup(key, helperCallback, create_cb,
+                                    Cache::Priority::LOW, true);
+          }
+          thread->stats.AddLookupUs(elapsed);
+        }
+        if (handle) {
+          thread->stats.AddHits(1);
+          // do something with the data
+          result += NPHash64(static_cast<char*>(cache_->Value(handle)),
+                             FLAGS_value_bytes);
+        }
+      } else if (random_op < erase_threshold_) {
+        // do erase
+        cache_->Erase(key);
+      } else {
+        // Should be extremely unlikely (noop)
+        assert(random_op >= kHundredthUint64 * 100U);
+      }
+    }
+    if (handle) {
+      cache_->Release(handle);
+      handle = nullptr;
+    }
+  }
+
+  void PrintEnv() const {
+    printf("RocksDB version     : %d.%d\n", kMajorVersion, kMinorVersion);
+    printf("Number of threads   : %u\n", FLAGS_threads);
+    printf("Ops per thread      : %" PRIu64 "\n", FLAGS_ops_per_thread);
+    printf("Cache size          : %" PRIu64 "\n", FLAGS_cache_size);
+    printf("Num shard bits      : %u\n", FLAGS_num_shard_bits);
+    printf("Max key             : %" PRIu64 "\n", max_key_);
+    printf("Resident ratio      : %g\n", FLAGS_resident_ratio);
+    printf("Skew degree         : %u\n", FLAGS_skew);
+    printf("Populate cache      : %d\n", int{FLAGS_populate_cache});
+    printf("Lookup+Insert pct   : %u%%\n", FLAGS_lookup_insert_percent);
+    printf("Insert percentage   : %u%%\n", FLAGS_insert_percent);
+    printf("Lookup percentage   : %u%%\n", FLAGS_lookup_percent);
+    printf("Erase percentage    : %u%%\n", FLAGS_erase_percent);
+    printf("----------------------------\n");
+  }
+};
+
+int cache_bench_tool(int argc, char** argv) {
+  ParseCommandLineFlags(&argc, &argv, true);
+
+  if (FLAGS_threads <= 0) {
+    fprintf(stderr, "threads number <= 0\n");
+    exit(1);
+  }
+
+  ROCKSDB_NAMESPACE::CacheBench bench;
+  if (FLAGS_populate_cache) {
+    bench.PopulateCache();
+    printf("Population complete\n");
+    printf("----------------------------\n");
+  }
+  if (bench.Run()) {
+    return 0;
+  } else {
+    return 1;
+  }
+}
+}  // namespace ROCKSDB_NAMESPACE
+#endif

cache/cache_helpers.h

@@ -0,0 +1,114 @@
+//  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
+
+#include <cassert>
+
+#include "rocksdb/cache.h"
+#include "rocksdb/rocksdb_namespace.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+// Returns the cached value given a cache handle.
+template <typename T>
+T* GetFromCacheHandle(Cache* cache, Cache::Handle* handle) {
+  assert(cache);
+  assert(handle);
+
+  return static_cast<T*>(cache->Value(handle));
+}
+
+// Simple generic deleter for Cache (to be used with Cache::Insert).
+template <typename T>
+void DeleteCacheEntry(const Slice& /* key */, void* value) {
+  delete static_cast<T*>(value);
+}
+
+// Turns a T* into a Slice so it can be used as a key with Cache.
+template <typename T>
+Slice GetSlice(const T* t) {
+  return Slice(reinterpret_cast<const char*>(t), sizeof(T));
+}
+
+// Generic resource management object for cache handles that releases the handle
+// when destroyed. Has unique ownership of the handle, so copying it is not
+// allowed, while moving it transfers ownership.
+template <typename T>
+class CacheHandleGuard {
+ public:
+  CacheHandleGuard() = default;
+
+  CacheHandleGuard(Cache* cache, Cache::Handle* handle)
+      : cache_(cache),
+        handle_(handle),
+        value_(GetFromCacheHandle<T>(cache, handle)) {
+    assert(cache_ && handle_ && value_);
+  }
+
+  CacheHandleGuard(const CacheHandleGuard&) = delete;
+  CacheHandleGuard& operator=(const CacheHandleGuard&) = delete;
+
+  CacheHandleGuard(CacheHandleGuard&& rhs) noexcept
+      : cache_(rhs.cache_), handle_(rhs.handle_), value_(rhs.value_) {
+    assert((!cache_ && !handle_ && !value_) || (cache_ && handle_ && value_));
+
+    rhs.ResetFields();
+  }
+
+  CacheHandleGuard& operator=(CacheHandleGuard&& rhs) noexcept {
+    if (this == &rhs) {
+      return *this;
+    }
+
+    ReleaseHandle();
+
+    cache_ = rhs.cache_;
+    handle_ = rhs.handle_;
+    value_ = rhs.value_;
+
+    assert((!cache_ && !handle_ && !value_) || (cache_ && handle_ && value_));
+
+    rhs.ResetFields();
+
+    return *this;
+  }
+
+  ~CacheHandleGuard() { ReleaseHandle(); }
+
+  bool IsEmpty() const { return !handle_; }
+
+  Cache* GetCache() const { return cache_; }
+  Cache::Handle* GetCacheHandle() const { return handle_; }
+  T* GetValue() const { return value_; }
+
+  void Reset() {
+    ReleaseHandle();
+    ResetFields();
+  }
+
+ private:
+  void ReleaseHandle() {
+    if (IsEmpty()) {
+      return;
+    }
+
+    assert(cache_);
+    cache_->Release(handle_);
+  }
+
+  void ResetFields() {
+    cache_ = nullptr;
+    handle_ = nullptr;
+    value_ = nullptr;
+  }
+
+ private:
+  Cache* cache_ = nullptr;
+  Cache::Handle* handle_ = nullptr;
+  T* value_ = nullptr;
+};
+
+}  // namespace ROCKSDB_NAMESPACE

cache/clock_cache.cc

@@ -271,7 +271,27 @@ class ClockCacheShard final : public CacheShard {
   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;
+  Status Insert(const Slice& key, uint32_t hash, void* value,
+                Cache::CacheItemHelperCallback helper_cb, size_t charge,
+                Cache::Handle** handle, Cache::Priority priority) override {
+    Cache::DeletionCallback delete_cb;
+    (*helper_cb)(nullptr, nullptr, &delete_cb);
+    return Insert(key, hash, value, charge, delete_cb, handle, priority);
+  }
   Cache::Handle* Lookup(const Slice& key, uint32_t hash) override;
+  Cache::Handle* Lookup(const Slice& key, uint32_t hash,
+                        Cache::CacheItemHelperCallback /*helper_cb*/,
+                        const Cache::CreateCallback& /*create_cb*/,
+                        Cache::Priority /*priority*/, bool /*wait*/) override {
+    return Lookup(key, hash);
+  }
+  bool Release(Cache::Handle* handle, bool /*useful*/,
+               bool force_erase) override {
+    return Release(handle, force_erase);
+  }
+  bool isReady(Cache::Handle* /*handle*/) override { return true; }
+  void Wait(Cache::Handle* /*handle*/) override {}
+
   // If the entry in in cache, increase reference count and return true.
   // Return false otherwise.
   //
@@ -748,6 +768,8 @@ class ClockCache final : public ShardedCache {
 
   void DisownData() override { shards_ = nullptr; }
 
+  void WaitAll(std::vector<Handle*>& /*handles*/) override {}
+
  private:
   ClockCacheShard* shards_;
 };

cache/lru_cache.cc

@@ -9,9 +9,8 @@
 
 #include "cache/lru_cache.h"
 
-#include <assert.h>
-#include <stdio.h>
-#include <stdlib.h>
+#include <cassert>
+#include <cstdio>
 #include <string>
 
 #include "util/mutexlock.h"
@@ -98,7 +97,8 @@ void LRUHandleTable::Resize() {
 LRUCacheShard::LRUCacheShard(size_t capacity, bool strict_capacity_limit,
                              double high_pri_pool_ratio,
                              bool use_adaptive_mutex,
-                             CacheMetadataChargePolicy metadata_charge_policy)
+                             CacheMetadataChargePolicy metadata_charge_policy,
+                             const std::shared_ptr<TieredCache>& tiered_cache)
     : capacity_(0),
       high_pri_pool_usage_(0),
       strict_capacity_limit_(strict_capacity_limit),
@@ -106,7 +106,8 @@ LRUCacheShard::LRUCacheShard(size_t capacity, bool strict_capacity_limit,
       high_pri_pool_capacity_(0),
       usage_(0),
       lru_usage_(0),
-      mutex_(use_adaptive_mutex) {
+      mutex_(use_adaptive_mutex),
+      tiered_cache_(tiered_cache) {
   set_metadata_charge_policy(metadata_charge_policy);
   // Make empty circular linked list
   lru_.next = &lru_;
@@ -257,8 +258,14 @@ void LRUCacheShard::SetCapacity(size_t capacity) {
     EvictFromLRU(0, &last_reference_list);
   }
 
+  // Try to insert the evicted entries into tiered cache
   // Free the entries outside of mutex for performance reasons
   for (auto entry : last_reference_list) {
+    if (tiered_cache_ && entry->IsTieredCacheCompatible() &&
+        !entry->IsPromoted()) {
+      tiered_cache_->Insert(entry->key(), entry->value, entry->info_.helper_cb)
+          .PermitUncheckedError();
+    }
     entry->Free();
   }
 }
@@ -268,17 +275,126 @@ void LRUCacheShard::SetStrictCapacityLimit(bool strict_capacity_limit) {
   strict_capacity_limit_ = strict_capacity_limit;
 }
 
-Cache::Handle* LRUCacheShard::Lookup(const Slice& key, uint32_t hash) {
-  MutexLock l(&mutex_);
-  LRUHandle* e = table_.Lookup(key, hash);
-  if (e != nullptr) {
-    assert(e->InCache());
-    if (!e->HasRefs()) {
-      // The entry is in LRU since it's in hash and has no external references
-      LRU_Remove(e);
+Status LRUCacheShard::InsertItem(LRUHandle* e, Cache::Handle** handle) {
+  Status s = Status::OK();
+  autovector<LRUHandle*> last_reference_list;
+  size_t total_charge = e->CalcTotalCharge(metadata_charge_policy_);
+
+  {
+    MutexLock l(&mutex_);
+
+    // Free the space following strict LRU policy until enough space
+    // is freed or the lru list is empty
+    EvictFromLRU(total_charge, &last_reference_list);
+
+    if ((usage_ + total_charge) > capacity_ &&
+        (strict_capacity_limit_ || handle == nullptr)) {
+      if (handle == nullptr) {
+        // Don't insert the entry but still return ok, as if the entry inserted
+        // into cache and get evicted immediately.
+        e->SetInCache(false);
+        last_reference_list.push_back(e);
+      } else {
+        delete[] reinterpret_cast<char*>(e);
+        *handle = nullptr;
+        s = Status::Incomplete("Insert failed due to LRU cache being full.");
+      }
+    } else {
+      // Insert into the cache. Note that the cache might get larger than its
+      // capacity if not enough space was freed up.
+      LRUHandle* old = table_.Insert(e);
+      usage_ += total_charge;
+      if (old != nullptr) {
+        s = Status::OkOverwritten();
+        assert(old->InCache());
+        old->SetInCache(false);
+        if (!old->HasRefs()) {
+          // old is on LRU because it's in cache and its reference count is 0
+          LRU_Remove(old);
+          size_t old_total_charge =
+              old->CalcTotalCharge(metadata_charge_policy_);
+          assert(usage_ >= old_total_charge);
+          usage_ -= old_total_charge;
+          last_reference_list.push_back(old);
+        }
+      }
+      if (handle == nullptr) {
+        LRU_Insert(e);
+      } else {
+        e->Ref();
+        *handle = reinterpret_cast<Cache::Handle*>(e);
+      }
+    }
+  }
+
+  // Try to insert the evicted entries into NVM cache
+  // Free the entries here outside of mutex for performance reasons
+  for (auto entry : last_reference_list) {
+    if (tiered_cache_ && entry->IsTieredCacheCompatible() &&
+        !entry->IsPromoted()) {
+      tiered_cache_->Insert(entry->key(), entry->value, entry->info_.helper_cb)
+          .PermitUncheckedError();
+    }
+    entry->Free();
+  }
+
+  return s;
+}
+
+Cache::Handle* LRUCacheShard::Lookup(
+    const Slice& key, uint32_t hash,
+    ShardedCache::CacheItemHelperCallback helper_cb,
+    const ShardedCache::CreateCallback& create_cb, Cache::Priority priority,
+    bool wait) {
+  LRUHandle* e = nullptr;
+  {
+    MutexLock l(&mutex_);
+    e = table_.Lookup(key, hash);
+    if (e != nullptr) {
+      assert(e->InCache());
+      if (!e->HasRefs()) {
+        // The entry is in LRU since it's in hash and has no external references
+        LRU_Remove(e);
+      }
+      e->Ref();
+      e->SetHit();
+    }
+  }
+
+  // If handle table lookup failed, then allocate a handle outside the
+  // mutex if we're going to lookup in the NVM cache
+  // Only support synchronous for now
+  // TODO: Support asynchronous lookup in NVM cache
+  if (!e && tiered_cache_ && helper_cb && wait) {
+    assert(create_cb);
+    std::unique_ptr<TieredCacheHandle> tiered_handle =
+        tiered_cache_->Lookup(key, create_cb, wait);
+    if (tiered_handle != nullptr) {
+      e = reinterpret_cast<LRUHandle*>(
+          new char[sizeof(LRUHandle) - 1 + key.size()]);
+
+      e->flags = 0;
+      e->SetPromoted(true);
+      e->SetTieredCacheCompatible(true);
+      e->info_.helper_cb = helper_cb;
+      e->key_length = key.size();
+      e->hash = hash;
+      e->refs = 0;
+      e->next = e->prev = nullptr;
+      e->SetInCache(true);
+      e->SetPriority(priority);
+      memcpy(e->key_data, key.data(), key.size());
+
+      e->value = tiered_handle->Value();
+      e->charge = tiered_handle->Size();
+
+      // This call could nullify e if the cache is over capacity and
+      // strict_capacity_limit_ is true. In such a case, the caller will try
+      // to insert later, which might again fail, but its ok as this should
+      // not be common
+      InsertItem(e, reinterpret_cast<Cache::Handle**>(&e))
+          .PermitUncheckedError();
     }
-    e->Ref();
-    e->SetHit();
   }
   return reinterpret_cast<Cache::Handle*>(e);
 }
@@ -339,81 +455,32 @@ bool LRUCacheShard::Release(Cache::Handle* handle, bool force_erase) {
 Status LRUCacheShard::Insert(const Slice& key, uint32_t hash, void* value,
                              size_t charge,
                              void (*deleter)(const Slice& key, void* value),
+                             Cache::CacheItemHelperCallback helper_cb,
                              Cache::Handle** handle, Cache::Priority priority) {
   // Allocate the memory here outside of the mutex
   // If the cache is full, we'll have to release it
   // It shouldn't happen very often though.
   LRUHandle* e = reinterpret_cast<LRUHandle*>(
       new char[sizeof(LRUHandle) - 1 + key.size()]);
-  Status s = Status::OK();
-  autovector<LRUHandle*> last_reference_list;
 
   e->value = value;
-  e->deleter = deleter;
+  e->flags = 0;
+  if (helper_cb) {
+    e->SetTieredCacheCompatible(true);
+    e->info_.helper_cb = helper_cb;
+  } else {
+    e->info_.deleter = deleter;
+  }
   e->charge = charge;
   e->key_length = key.size();
-  e->flags = 0;
   e->hash = hash;
   e->refs = 0;
   e->next = e->prev = nullptr;
   e->SetInCache(true);
   e->SetPriority(priority);
   memcpy(e->key_data, key.data(), key.size());
-  size_t total_charge = e->CalcTotalCharge(metadata_charge_policy_);
 
-  {
-    MutexLock l(&mutex_);
-
-    // Free the space following strict LRU policy until enough space
-    // is freed or the lru list is empty
-    EvictFromLRU(total_charge, &last_reference_list);
-
-    if ((usage_ + total_charge) > capacity_ &&
-        (strict_capacity_limit_ || handle == nullptr)) {
-      if (handle == nullptr) {
-        // Don't insert the entry but still return ok, as if the entry inserted
-        // into cache and get evicted immediately.
-        e->SetInCache(false);
-        last_reference_list.push_back(e);
-      } else {
-        delete[] reinterpret_cast<char*>(e);
-        *handle = nullptr;
-        s = Status::Incomplete("Insert failed due to LRU cache being full.");
-      }
-    } else {
-      // Insert into the cache. Note that the cache might get larger than its
-      // capacity if not enough space was freed up.
-      LRUHandle* old = table_.Insert(e);
-      usage_ += total_charge;
-      if (old != nullptr) {
-        s = Status::OkOverwritten();
-        assert(old->InCache());
-        old->SetInCache(false);
-        if (!old->HasRefs()) {
-          // old is on LRU because it's in cache and its reference count is 0
-          LRU_Remove(old);
-          size_t old_total_charge =
-              old->CalcTotalCharge(metadata_charge_policy_);
-          assert(usage_ >= old_total_charge);
-          usage_ -= old_total_charge;
-          last_reference_list.push_back(old);
-        }
-      }
-      if (handle == nullptr) {
-        LRU_Insert(e);
-      } else {
-        e->Ref();
-        *handle = reinterpret_cast<Cache::Handle*>(e);
-      }
-    }
-  }
-
-  // Free the entries here outside of mutex for performance reasons
-  for (auto entry : last_reference_list) {
-    entry->Free();
-  }
-
-  return s;
+  return InsertItem(e, handle);
 }
 
 void LRUCacheShard::Erase(const Slice& key, uint32_t hash) {
@@ -469,7 +536,8 @@ 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,
-                   CacheMetadataChargePolicy metadata_charge_policy)
+                   CacheMetadataChargePolicy metadata_charge_policy,
+                   const std::shared_ptr<TieredCache>& tiered_cache)
     : ShardedCache(capacity, num_shard_bits, strict_capacity_limit,
                    std::move(allocator)) {
   num_shards_ = 1 << num_shard_bits;
@@ -479,7 +547,7 @@ LRUCache::LRUCache(size_t capacity, int num_shard_bits,
   for (int i = 0; i < num_shards_; i++) {
     new (&shards_[i])
         LRUCacheShard(per_shard, strict_capacity_limit, high_pri_pool_ratio,
-                      use_adaptive_mutex, metadata_charge_policy);
+                      use_adaptive_mutex, metadata_charge_policy, tiered_cache);
   }
 }
 
@@ -544,19 +612,12 @@ double LRUCache::GetHighPriPoolRatio() {
   return result;
 }
 
-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.metadata_charge_policy);
-}
-
 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,
-    CacheMetadataChargePolicy metadata_charge_policy) {
+    CacheMetadataChargePolicy metadata_charge_policy,
+    const std::shared_ptr<TieredCache>& tiered_cache) {
   if (num_shard_bits >= 20) {
     return nullptr;  // the cache cannot be sharded into too many fine pieces
   }
@@ -569,7 +630,25 @@ std::shared_ptr<Cache> NewLRUCache(
   }
   return std::make_shared<LRUCache>(
       capacity, num_shard_bits, strict_capacity_limit, high_pri_pool_ratio,
-      std::move(memory_allocator), use_adaptive_mutex, metadata_charge_policy);
+      std::move(memory_allocator), use_adaptive_mutex, metadata_charge_policy,
+      tiered_cache);
 }
 
+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.metadata_charge_policy, cache_opts.tiered_cache);
+}
+
+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,
+    CacheMetadataChargePolicy metadata_charge_policy) {
+  return NewLRUCache(capacity, num_shard_bits, strict_capacity_limit,
+                     high_pri_pool_ratio, memory_allocator, use_adaptive_mutex,
+                     metadata_charge_policy, nullptr);
+}
 }  // namespace ROCKSDB_NAMESPACE

cache/lru_cache.h

@@ -11,9 +11,9 @@
 #include <string>
 
 #include "cache/sharded_cache.h"
-
 #include "port/malloc.h"
 #include "port/port.h"
+#include "rocksdb/tiered_cache.h"
 #include "util/autovector.h"
 
 namespace ROCKSDB_NAMESPACE {
@@ -49,7 +49,14 @@ namespace ROCKSDB_NAMESPACE {
 
 struct LRUHandle {
   void* value;
-  void (*deleter)(const Slice&, void* value);
+  union Info {
+    Info() {}
+    ~Info() {}
+    void (*deleter)(const Slice&, void* value);
+    ShardedCache::CacheItemHelperCallback helper_cb;
+    // This needs to be explicitly constructed and destructed
+    std::unique_ptr<TieredCacheHandle> tiered_handle;
+  } info_;
   LRUHandle* next_hash;
   LRUHandle* next;
   LRUHandle* prev;
@@ -67,8 +74,14 @@ struct LRUHandle {
     IS_HIGH_PRI = (1 << 1),
     // Whether this entry is in high-pri pool.
     IN_HIGH_PRI_POOL = (1 << 2),
-    // Wwhether this entry has had any lookups (hits).
+    // Whether this entry has had any lookups (hits).
     HAS_HIT = (1 << 3),
+    // Can this be inserted into the tiered cache
+    IS_TIERED_CACHE_COMPATIBLE = (1 << 4),
+    // Is the handle still being read from a lower tier
+    IS_PENDING = (1 << 5),
+    // Has the item been promoted from a lower tier
+    IS_PROMOTED = (1 << 6),
   };
 
   uint8_t flags;
@@ -95,6 +108,11 @@ struct LRUHandle {
   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 IsTieredCacheCompatible() const {
+    return flags & IS_TIERED_CACHE_COMPATIBLE;
+  }
+  bool IsPending() const { return flags & IS_PENDING; }
+  bool IsPromoted() const { return flags & IS_PROMOTED; }
 
   void SetInCache(bool in_cache) {
     if (in_cache) {
@@ -122,15 +140,43 @@ struct LRUHandle {
 
   void SetHit() { flags |= HAS_HIT; }
 
+  void SetTieredCacheCompatible(bool tiered) {
+    if (tiered) {
+      flags |= IS_TIERED_CACHE_COMPATIBLE;
+    } else {
+      flags &= ~IS_TIERED_CACHE_COMPATIBLE;
+    }
+  }
+
+  void SetIncomplete(bool incomp) {
+    if (incomp) {
+      flags |= IS_PENDING;
+    } else {
+      flags &= ~IS_PENDING;
+    }
+  }
+
+  void SetPromoted(bool promoted) {
+    if (promoted) {
+      flags |= IS_PROMOTED;
+    } else {
+      flags &= ~IS_PROMOTED;
+    }
+  }
+
   void Free() {
     assert(refs == 0);
-    if (deleter) {
-      (*deleter)(key(), value);
+    if (!IsTieredCacheCompatible() && info_.deleter) {
+      (*info_.deleter)(key(), value);
+    } else if (IsTieredCacheCompatible()) {
+      ShardedCache::DeletionCallback del_cb;
+      (*info_.helper_cb)(nullptr, nullptr, &del_cb);
+      (*del_cb)(key(), value);
     }
     delete[] reinterpret_cast<char*>(this);
   }
 
-  // Caclculate the memory usage by metadata
+  // Calculate the memory usage by metadata
   inline size_t CalcTotalCharge(
       CacheMetadataChargePolicy metadata_charge_policy) {
     size_t meta_charge = 0;
@@ -193,7 +239,8 @@ class ALIGN_AS(CACHE_LINE_SIZE) LRUCacheShard final : public CacheShard {
  public:
   LRUCacheShard(size_t capacity, bool strict_capacity_limit,
                 double high_pri_pool_ratio, bool use_adaptive_mutex,
-                CacheMetadataChargePolicy metadata_charge_policy);
+                CacheMetadataChargePolicy metadata_charge_policy,
+                const std::shared_ptr<TieredCache>& tiered_cache);
   virtual ~LRUCacheShard() override = default;
 
   // Separate from constructor so caller can easily make an array of LRUCache
@@ -212,8 +259,32 @@ class ALIGN_AS(CACHE_LINE_SIZE) LRUCacheShard final : public CacheShard {
                         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;
+                        Cache::Priority priority) override {
+    return Insert(key, hash, value, charge, deleter, nullptr, handle, priority);
+  }
+  virtual Status Insert(const Slice& key, uint32_t hash, void* value,
+                        Cache::CacheItemHelperCallback helper_cb, size_t charge,
+                        Cache::Handle** handle,
+                        Cache::Priority priority) override {
+    return Insert(key, hash, value, charge, nullptr, helper_cb, handle,
+                  priority);
+  }
+  // If helper_cb is null, the values of the following arguments don't
+  // matter
+  virtual Cache::Handle* Lookup(const Slice& key, uint32_t hash,
+                                ShardedCache::CacheItemHelperCallback helper_cb,
+                                const ShardedCache::CreateCallback& create_cb,
+                                ShardedCache::Priority priority,
+                                bool wait) override;
+  virtual Cache::Handle* Lookup(const Slice& key, uint32_t hash) override {
+    return Lookup(key, hash, nullptr, nullptr, Cache::Priority::LOW, true);
+  }
+  virtual bool Release(Cache::Handle* handle, bool /*useful*/,
+                       bool force_erase) override {
+    return Release(handle, force_erase);
+  }
+  virtual bool isReady(Cache::Handle* /*handle*/) override { return true; }
+  virtual void Wait(Cache::Handle* /*handle*/) override {}
   virtual bool Ref(Cache::Handle* handle) override;
   virtual bool Release(Cache::Handle* handle,
                        bool force_erase = false) override;
@@ -239,10 +310,15 @@ class ALIGN_AS(CACHE_LINE_SIZE) LRUCacheShard final : public CacheShard {
   //  not threadsafe
   size_t TEST_GetLRUSize();
 
-  //  Retrives high pri pool ratio
+  //  Retrieves high pri pool ratio
   double GetHighPriPoolRatio();
 
  private:
+  Status InsertItem(LRUHandle* item, Cache::Handle** handle);
+  Status Insert(const Slice& key, uint32_t hash, void* value, size_t charge,
+                void (*deleter)(const Slice& key, void* value),
+                Cache::CacheItemHelperCallback helper_cb,
+                Cache::Handle** handle, Cache::Priority priority);
   void LRU_Remove(LRUHandle* e);
   void LRU_Insert(LRUHandle* e);
 
@@ -303,6 +379,8 @@ class ALIGN_AS(CACHE_LINE_SIZE) LRUCacheShard final : public CacheShard {
   // We don't count mutex_ as the cache's internal state so semantically we
   // don't mind mutex_ invoking the non-const actions.
   mutable port::Mutex mutex_;
+
+  std::shared_ptr<TieredCache> tiered_cache_;
 };
 
 class LRUCache
@@ -316,7 +394,8 @@ class LRUCache
            std::shared_ptr<MemoryAllocator> memory_allocator = nullptr,
            bool use_adaptive_mutex = kDefaultToAdaptiveMutex,
            CacheMetadataChargePolicy metadata_charge_policy =
-               kDontChargeCacheMetadata);
+               kDontChargeCacheMetadata,
+           const std::shared_ptr<TieredCache>& tiered_cache = nullptr);
   virtual ~LRUCache();
   virtual const char* Name() const override { return "LRUCache"; }
   virtual CacheShard* GetShard(int shard) override;
@@ -325,10 +404,11 @@ class LRUCache
   virtual size_t GetCharge(Handle* handle) const override;
   virtual uint32_t GetHash(Handle* handle) const override;
   virtual void DisownData() override;
+  virtual void WaitAll(std::vector<Handle*>& /*handles*/) override {}
 
   //  Retrieves number of elements in LRU, for unit test purpose only
   size_t TEST_GetLRUSize();
-  //  Retrives high pri pool ratio
+  //  Retrieves high pri pool ratio
   double GetHighPriPoolRatio();
 
  private:

cache/lru_cache_test.cc

@@ -7,8 +7,12 @@
 
 #include <string>
 #include <vector>
+
 #include "port/port.h"
+#include "rocksdb/cache.h"
 #include "test_util/testharness.h"
+#include "util/coding.h"
+#include "util/random.h"
 
 namespace ROCKSDB_NAMESPACE {
 
@@ -30,9 +34,9 @@ class LRUCacheTest : public testing::Test {
     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,
-                               kDontChargeCacheMetadata);
+    new (cache_) LRUCacheShard(
+        capacity, false /*strict_capcity_limit*/, high_pri_pool_ratio,
+        use_adaptive_mutex, kDontChargeCacheMetadata, nullptr /*tiered_cache*/);
   }
 
   void Insert(const std::string& key,
@@ -191,6 +195,175 @@ TEST_F(LRUCacheTest, EntriesWithPriority) {
   ValidateLRUList({"e", "f", "g", "Z", "d"}, 2);
 }
 
+class TestTieredCache : public TieredCache {
+ public:
+  TestTieredCache(size_t capacity) : num_inserts_(0), num_lookups_(0) {
+    cache_ = NewLRUCache(capacity, 0, false, 0.5, nullptr,
+                         kDefaultToAdaptiveMutex, kDontChargeCacheMetadata);
+  }
+  ~TestTieredCache() { cache_.reset(); }
+
+  std::string Name() override { return "TestTieredCache"; }
+
+  Status Insert(const Slice& key, void* value,
+                Cache::CacheItemHelperCallback helper_cb) override {
+    Cache::SizeCallback size_cb;
+    Cache::SaveToCallback save_cb;
+    size_t size;
+    char* buf;
+    Status s;
+
+    num_inserts_++;
+    (*helper_cb)(&size_cb, &save_cb, nullptr);
+    size = (*size_cb)(value);
+    buf = new char[size + sizeof(uint64_t)];
+    EncodeFixed64(buf, size);
+    s = (*save_cb)(value, 0, size, buf + sizeof(uint64_t));
+    EXPECT_OK(s);
+    return cache_->Insert(key, buf, size,
+                          [](const Slice& /*key*/, void* val) -> void {
+                            delete[] reinterpret_cast<char*>(val);
+                          });
+  }
+
+  std::unique_ptr<TieredCacheHandle> Lookup(
+      const Slice& key, const Cache::CreateCallback& create_cb,
+      bool /*wait*/) override {
+    std::unique_ptr<TieredCacheHandle> tiered_handle;
+    Cache::Handle* handle = cache_->Lookup(key);
+    num_lookups_++;
+    if (handle) {
+      void* value;
+      size_t charge;
+      char* ptr = (char*)cache_->Value(handle);
+      size_t size = DecodeFixed64(ptr);
+      ptr += sizeof(uint64_t);
+      Status s = create_cb(ptr, size, &value, &charge);
+      EXPECT_OK(s);
+      tiered_handle.reset(
+          new TestTieredCacheHandle(cache_.get(), handle, value, charge));
+    }
+    return tiered_handle;
+  }
+
+  void Erase(const Slice& /*key*/) override {}
+
+  void WaitAll(std::vector<TieredCacheHandle*> /*handles*/) override {}
+
+  std::string GetPrintableOptions() const override { return ""; }
+
+  uint32_t num_inserts() { return num_inserts_; }
+
+  uint32_t num_lookups() { return num_lookups_; }
+
+ private:
+  class TestTieredCacheHandle : public TieredCacheHandle {
+   public:
+    TestTieredCacheHandle(Cache* cache, Cache::Handle* handle, void* value,
+                          size_t size)
+        : cache_(cache), handle_(handle), value_(value), size_(size) {}
+    ~TestTieredCacheHandle() { cache_->Release(handle_); }
+
+    bool isReady() override { return true; }
+
+    void Wait() override {}
+
+    void* Value() override { return value_; }
+
+    size_t Size() override { return size_; }
+
+   private:
+    Cache* cache_;
+    Cache::Handle* handle_;
+    void* value_;
+    size_t size_;
+  };
+
+  std::shared_ptr<Cache> cache_;
+  uint32_t num_inserts_;
+  uint32_t num_lookups_;
+};
+
+TEST_F(LRUCacheTest, TestTieredCache) {
+  LRUCacheOptions opts(1024, 0, false, 0.5, nullptr, kDefaultToAdaptiveMutex,
+                       kDontChargeCacheMetadata);
+  std::shared_ptr<TestTieredCache> tiered_cache(new TestTieredCache(2048));
+  opts.tiered_cache = tiered_cache;
+  std::shared_ptr<Cache> cache = NewLRUCache(opts);
+
+  class TestItem {
+   public:
+    TestItem(const char* buf, size_t size) : buf_(new char[size]), size_(size) {
+      memcpy(buf_.get(), buf, size);
+    }
+    ~TestItem() {}
+
+    char* Buf() { return buf_.get(); }
+    size_t Size() { return size_; }
+
+   private:
+    std::unique_ptr<char[]> buf_;
+    size_t size_;
+  };
+
+  Cache::CacheItemHelperCallback helper_cb =
+      [](Cache::SizeCallback* size_cb, Cache::SaveToCallback* saveto_cb,
+         Cache::DeletionCallback* del_cb) -> void {
+    if (size_cb) {
+      *size_cb = [](void* obj) -> size_t {
+        return reinterpret_cast<TestItem*>(obj)->Size();
+      };
+    }
+    if (saveto_cb) {
+      *saveto_cb = [](void* obj, size_t offset, size_t size,
+                      void* out) -> Status {
+        TestItem* item = reinterpret_cast<TestItem*>(obj);
+        char* buf = item->Buf();
+        EXPECT_EQ(size, item->Size());
+        EXPECT_EQ(offset, 0);
+        memcpy(out, buf, size);
+        return Status::OK();
+      };
+    }
+    if (del_cb) {
+      *del_cb = [](const Slice& /*key*/, void* obj) -> void {
+        delete reinterpret_cast<TestItem*>(obj);
+      };
+    }
+  };
+
+  int create_count = 0;
+  Cache::CreateCallback test_item_creator =
+      [&create_count](void* buf, size_t size, void** out_obj,
+                      size_t* charge) -> Status {
+    create_count++;
+    *out_obj = reinterpret_cast<void*>(new TestItem((char*)buf, size));
+    *charge = size;
+    return Status::OK();
+  };
+
+  Random rnd(301);
+  std::string str1 = rnd.RandomString(1020);
+  TestItem* item1 = new TestItem(str1.data(), str1.length());
+  ASSERT_OK(cache->Insert("k1", item1, helper_cb, str1.length()));
+  std::string str2 = rnd.RandomString(1020);
+  TestItem* item2 = new TestItem(str2.data(), str2.length());
+  // k2 should be demoted to NVM
+  ASSERT_OK(cache->Insert("k2", item2, helper_cb, str2.length()));
+
+  Cache::Handle* handle;
+  handle = cache->Lookup("k2", helper_cb, test_item_creator,
+                         Cache::Priority::LOW, true);
+  ASSERT_NE(handle, nullptr);
+  cache->Release(handle);
+  // This lookup should promote k1 and demote k2
+  handle = cache->Lookup("k1", helper_cb, test_item_creator,
+                         Cache::Priority::LOW, true);
+  ASSERT_NE(handle, nullptr);
+  cache->Release(handle);
+  ASSERT_EQ(tiered_cache->num_inserts(), 2u);
+  ASSERT_EQ(tiered_cache->num_lookups(), 1u);
+}
 }  // namespace ROCKSDB_NAMESPACE
 
 int main(int argc, char** argv) {

cache/sharded_cache.cc

@@ -51,11 +51,39 @@ Status ShardedCache::Insert(const Slice& key, void* value, size_t charge,
       ->Insert(key, hash, value, charge, deleter, handle, priority);
 }
 
+Status ShardedCache::Insert(const Slice& key, void* value,
+                            CacheItemHelperCallback helper_cb, size_t charge,
+                            Handle** handle, Priority priority) {
+  uint32_t hash = HashSlice(key);
+  return GetShard(Shard(hash))
+      ->Insert(key, hash, value, helper_cb, charge, handle, priority);
+}
+
 Cache::Handle* ShardedCache::Lookup(const Slice& key, Statistics* /*stats*/) {
   uint32_t hash = HashSlice(key);
   return GetShard(Shard(hash))->Lookup(key, hash);
 }
 
+Cache::Handle* ShardedCache::Lookup(const Slice& key,
+                                    CacheItemHelperCallback helper_cb,
+                                    const CreateCallback& create_cb,
+                                    Priority priority, bool wait,
+                                    Statistics* /*stats*/) {
+  uint32_t hash = HashSlice(key);
+  return GetShard(Shard(hash))
+      ->Lookup(key, hash, helper_cb, create_cb, priority, wait);
+}
+
+bool ShardedCache::isReady(Handle* handle) {
+  uint32_t hash = GetHash(handle);
+  return GetShard(Shard(hash))->isReady(handle);
+}
+
+void ShardedCache::Wait(Handle* handle) {
+  uint32_t hash = GetHash(handle);
+  GetShard(Shard(hash))->Wait(handle);
+}
+
 bool ShardedCache::Ref(Handle* handle) {
   uint32_t hash = GetHash(handle);
   return GetShard(Shard(hash))->Ref(handle);
@@ -66,6 +94,11 @@ bool ShardedCache::Release(Handle* handle, bool force_erase) {
   return GetShard(Shard(hash))->Release(handle, force_erase);
 }
 
+bool ShardedCache::Release(Handle* handle, bool useful, bool force_erase) {
+  uint32_t hash = GetHash(handle);
+  return GetShard(Shard(hash))->Release(handle, useful, force_erase);
+}
+
 void ShardedCache::Erase(const Slice& key) {
   uint32_t hash = HashSlice(key);
   GetShard(Shard(hash))->Erase(key, hash);

cache/sharded_cache.h

@@ -28,9 +28,20 @@ class CacheShard {
                         size_t charge,
                         void (*deleter)(const Slice& key, void* value),
                         Cache::Handle** handle, Cache::Priority priority) = 0;
+  virtual Status Insert(const Slice& key, uint32_t hash, void* value,
+                        Cache::CacheItemHelperCallback helper_cb, size_t charge,
+                        Cache::Handle** handle, Cache::Priority priority) = 0;
   virtual Cache::Handle* Lookup(const Slice& key, uint32_t hash) = 0;
+  virtual Cache::Handle* Lookup(const Slice& key, uint32_t hash,
+                                Cache::CacheItemHelperCallback helper_cb,
+                                const Cache::CreateCallback& create_cb,
+                                Cache::Priority priority, bool wait) = 0;
+  virtual bool Release(Cache::Handle* handle, bool useful,
+                       bool force_erase) = 0;
+  virtual bool isReady(Cache::Handle* handle) = 0;
+  virtual void Wait(Cache::Handle* handle) = 0;
   virtual bool Ref(Cache::Handle* handle) = 0;
-  virtual bool Release(Cache::Handle* handle, bool force_erase = false) = 0;
+  virtual bool Release(Cache::Handle* handle, bool force_erase) = 0;
   virtual void Erase(const Slice& key, uint32_t hash) = 0;
   virtual void SetCapacity(size_t capacity) = 0;
   virtual void SetStrictCapacityLimit(bool strict_capacity_limit) = 0;
@@ -62,6 +73,7 @@ class ShardedCache : public Cache {
   virtual const CacheShard* GetShard(int shard) const = 0;
   virtual void* Value(Handle* handle) override = 0;
   virtual size_t GetCharge(Handle* handle) const override = 0;
+  virtual void WaitAll(std::vector<Handle*>& handles) override = 0;
 
   virtual uint32_t GetHash(Handle* handle) const = 0;
   virtual void DisownData() override = 0;
@@ -72,7 +84,18 @@ class ShardedCache : public Cache {
   virtual 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,
+                        CacheItemHelperCallback helper_cb, size_t chargge,
+                        Handle** handle = nullptr,
+                        Priority priority = Priority::LOW) override;
   virtual Handle* Lookup(const Slice& key, Statistics* stats) override;
+  virtual Handle* Lookup(const Slice& key, CacheItemHelperCallback helper_cb,
+                         const CreateCallback& create_cb, Priority priority,
+                         bool wait, Statistics* stats = nullptr) override;
+  virtual bool Release(Handle* handle, bool useful,
+                       bool force_erase = false) override;
+  virtual bool isReady(Handle* handle) override;
+  virtual void Wait(Handle* handle) override;
   virtual bool Ref(Handle* handle) override;
   virtual bool Release(Handle* handle, bool force_erase = false) override;
   virtual void Erase(const Slice& key) override;

db/arena_wrapped_db_iter.cc

@@ -30,20 +30,19 @@ Status ArenaWrappedDBIter::GetProperty(std::string prop_name,
   return db_iter_->GetProperty(prop_name, prop);
 }
 
-void ArenaWrappedDBIter::Init(Env* env, const ReadOptions& read_options,
-                              const ImmutableCFOptions& cf_options,
-                              const MutableCFOptions& mutable_cf_options,
-                              const SequenceNumber& sequence,
-                              uint64_t max_sequential_skip_in_iteration,
-                              uint64_t version_number,
-                              ReadCallback* read_callback, DBImpl* db_impl,
-                              ColumnFamilyData* cfd, bool allow_blob,
-                              bool allow_refresh) {
+void ArenaWrappedDBIter::Init(
+    Env* env, const ReadOptions& read_options,
+    const ImmutableCFOptions& cf_options,
+    const MutableCFOptions& mutable_cf_options, const Version* version,
+    const SequenceNumber& sequence, uint64_t max_sequential_skip_in_iteration,
+    uint64_t version_number, ReadCallback* read_callback, DBImpl* db_impl,
+    ColumnFamilyData* cfd, bool expose_blob_index, bool allow_refresh) {
   auto mem = arena_.AllocateAligned(sizeof(DBIter));
-  db_iter_ = new (mem) DBIter(env, read_options, cf_options, mutable_cf_options,
-                              cf_options.user_comparator, nullptr, sequence,
-                              true, max_sequential_skip_in_iteration,
-                              read_callback, db_impl, cfd, allow_blob);
+  db_iter_ =
+      new (mem) DBIter(env, read_options, cf_options, mutable_cf_options,
+                       cf_options.user_comparator, /* iter */ nullptr, version,
+                       sequence, true, max_sequential_skip_in_iteration,
+                       read_callback, db_impl, cfd, expose_blob_index);
   sv_number_ = version_number;
   read_options_ = read_options;
   allow_refresh_ = allow_refresh;
@@ -72,8 +71,9 @@ Status ArenaWrappedDBIter::Refresh() {
       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_,
+         sv->current, latest_seq,
+         sv->mutable_cf_options.max_sequential_skip_in_iterations,
+         cur_sv_number, read_callback_, db_impl_, cfd_, expose_blob_index_,
          allow_refresh_);
 
     InternalIterator* internal_iter = db_impl_->NewInternalIterator(
@@ -90,16 +90,16 @@ Status ArenaWrappedDBIter::Refresh() {
 ArenaWrappedDBIter* NewArenaWrappedDbIterator(
     Env* env, const ReadOptions& read_options,
     const ImmutableCFOptions& cf_options,
-    const MutableCFOptions& mutable_cf_options, const SequenceNumber& sequence,
-    uint64_t max_sequential_skip_in_iterations, uint64_t version_number,
-    ReadCallback* read_callback, DBImpl* db_impl, ColumnFamilyData* cfd,
-    bool allow_blob, bool allow_refresh) {
+    const MutableCFOptions& mutable_cf_options, const Version* version,
+    const SequenceNumber& sequence, uint64_t max_sequential_skip_in_iterations,
+    uint64_t version_number, ReadCallback* read_callback, DBImpl* db_impl,
+    ColumnFamilyData* cfd, bool expose_blob_index, bool allow_refresh) {
   ArenaWrappedDBIter* iter = new ArenaWrappedDBIter();
-  iter->Init(env, read_options, cf_options, mutable_cf_options, sequence,
-             max_sequential_skip_in_iterations, version_number, read_callback,
-             db_impl, cfd, allow_blob, allow_refresh);
+  iter->Init(env, read_options, cf_options, mutable_cf_options, version,
+             sequence, max_sequential_skip_in_iterations, version_number,
+             read_callback, db_impl, cfd, expose_blob_index, allow_refresh);
   if (db_impl != nullptr && cfd != nullptr && allow_refresh) {
-    iter->StoreRefreshInfo(db_impl, cfd, read_callback, allow_blob);
+    iter->StoreRefreshInfo(db_impl, cfd, read_callback, expose_blob_index);
   }
 
   return iter;

db/arena_wrapped_db_iter.h

@@ -23,6 +23,7 @@
 namespace ROCKSDB_NAMESPACE {
 
 class Arena;
+class Version;
 
 // A wrapper iterator which wraps DB Iterator and the arena, with which the DB
 // iterator is supposed to be allocated. This class is used as an entry point of
@@ -72,20 +73,20 @@ class ArenaWrappedDBIter : public Iterator {
 
   void Init(Env* env, const ReadOptions& read_options,
             const ImmutableCFOptions& cf_options,
-            const MutableCFOptions& mutable_cf_options,
+            const MutableCFOptions& mutable_cf_options, const Version* version,
             const SequenceNumber& sequence,
             uint64_t max_sequential_skip_in_iterations, uint64_t version_number,
             ReadCallback* read_callback, DBImpl* db_impl, ColumnFamilyData* cfd,
-            bool allow_blob, bool allow_refresh);
+            bool expose_blob_index, bool allow_refresh);
 
   // Store some parameters so we can refresh the iterator at a later point
   // with these same params
   void StoreRefreshInfo(DBImpl* db_impl, ColumnFamilyData* cfd,
-                        ReadCallback* read_callback, bool allow_blob) {
+                        ReadCallback* read_callback, bool expose_blob_index) {
     db_impl_ = db_impl;
     cfd_ = cfd;
     read_callback_ = read_callback;
-    allow_blob_ = allow_blob;
+    expose_blob_index_ = expose_blob_index;
   }
 
  private:
@@ -96,7 +97,7 @@ class ArenaWrappedDBIter : public Iterator {
   DBImpl* db_impl_ = nullptr;
   ReadOptions read_options_;
   ReadCallback* read_callback_;
-  bool allow_blob_ = false;
+  bool expose_blob_index_ = false;
   bool allow_refresh_ = true;
 };
 
@@ -106,9 +107,9 @@ class ArenaWrappedDBIter : public Iterator {
 extern ArenaWrappedDBIter* NewArenaWrappedDbIterator(
     Env* env, const ReadOptions& read_options,
     const ImmutableCFOptions& cf_options,
-    const MutableCFOptions& mutable_cf_options, const SequenceNumber& sequence,
-    uint64_t max_sequential_skip_in_iterations, uint64_t version_number,
-    ReadCallback* read_callback, DBImpl* db_impl = nullptr,
-    ColumnFamilyData* cfd = nullptr, bool allow_blob = false,
-    bool allow_refresh = true);
+    const MutableCFOptions& mutable_cf_options, const Version* version,
+    const SequenceNumber& sequence, uint64_t max_sequential_skip_in_iterations,
+    uint64_t version_number, ReadCallback* read_callback,
+    DBImpl* db_impl = nullptr, ColumnFamilyData* cfd = nullptr,
+    bool expose_blob_index = false, bool allow_refresh = true);
 }  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_builder.cc

@@ -8,6 +8,7 @@
 #include <cassert>
 
 #include "db/blob/blob_file_addition.h"
+#include "db/blob/blob_file_completion_callback.h"
 #include "db/blob/blob_index.h"
 #include "db/blob/blob_log_format.h"
 #include "db/blob/blob_log_writer.h"
@@ -17,39 +18,44 @@
 #include "file/writable_file_writer.h"
 #include "logging/logging.h"
 #include "options/cf_options.h"
+#include "options/options_helper.h"
 #include "rocksdb/slice.h"
 #include "rocksdb/status.h"
 #include "test_util/sync_point.h"
+#include "trace_replay/io_tracer.h"
 #include "util/compression.h"
 
 namespace ROCKSDB_NAMESPACE {
 
 BlobFileBuilder::BlobFileBuilder(
-    VersionSet* versions, Env* env, FileSystem* fs,
+    VersionSet* versions, FileSystem* fs,
     const ImmutableCFOptions* immutable_cf_options,
     const MutableCFOptions* mutable_cf_options, const FileOptions* file_options,
     int job_id, uint32_t column_family_id,
     const std::string& column_family_name, Env::IOPriority io_priority,
     Env::WriteLifeTimeHint write_hint,
+    const std::shared_ptr<IOTracer>& io_tracer,
+    BlobFileCompletionCallback* blob_callback,
     std::vector<std::string>* blob_file_paths,
     std::vector<BlobFileAddition>* blob_file_additions)
-    : BlobFileBuilder([versions]() { return versions->NewFileNumber(); }, env,
-                      fs, immutable_cf_options, mutable_cf_options,
-                      file_options, job_id, column_family_id,
-                      column_family_name, io_priority, write_hint,
-                      blob_file_paths, blob_file_additions) {}
+    : BlobFileBuilder([versions]() { return versions->NewFileNumber(); }, fs,
+                      immutable_cf_options, mutable_cf_options, file_options,
+                      job_id, column_family_id, column_family_name, io_priority,
+                      write_hint, io_tracer, blob_callback, blob_file_paths,
+                      blob_file_additions) {}
 
 BlobFileBuilder::BlobFileBuilder(
-    std::function<uint64_t()> file_number_generator, Env* env, FileSystem* fs,
+    std::function<uint64_t()> file_number_generator, FileSystem* fs,
     const ImmutableCFOptions* immutable_cf_options,
     const MutableCFOptions* mutable_cf_options, const FileOptions* file_options,
     int job_id, uint32_t column_family_id,
     const std::string& column_family_name, Env::IOPriority io_priority,
     Env::WriteLifeTimeHint write_hint,
+    const std::shared_ptr<IOTracer>& io_tracer,
+    BlobFileCompletionCallback* blob_callback,
     std::vector<std::string>* blob_file_paths,
     std::vector<BlobFileAddition>* blob_file_additions)
     : file_number_generator_(std::move(file_number_generator)),
-      env_(env),
       fs_(fs),
       immutable_cf_options_(immutable_cf_options),
       min_blob_size_(mutable_cf_options->min_blob_size),
@@ -61,12 +67,13 @@ BlobFileBuilder::BlobFileBuilder(
       column_family_name_(column_family_name),
       io_priority_(io_priority),
       write_hint_(write_hint),
+      io_tracer_(io_tracer),
+      blob_callback_(blob_callback),
       blob_file_paths_(blob_file_paths),
       blob_file_additions_(blob_file_additions),
       blob_count_(0),
       blob_bytes_(0) {
   assert(file_number_generator_);
-  assert(env_);
   assert(fs_);
   assert(immutable_cf_options_);
   assert(file_options_);
@@ -157,11 +164,12 @@ Status BlobFileBuilder::OpenBlobFileIfNeeded() {
   std::unique_ptr<FSWritableFile> file;
 
   {
-    TEST_SYNC_POINT("BlobFileBuilder::OpenBlobFileIfNeeded:NewWritableFile");
-
     assert(file_options_);
-    const Status s =
-        NewWritableFile(fs_, blob_file_path, &file, *file_options_);
+    Status s = NewWritableFile(fs_, blob_file_path, &file, *file_options_);
+
+    TEST_SYNC_POINT_CALLBACK(
+        "BlobFileBuilder::OpenBlobFileIfNeeded:NewWritableFile", &s);
+
     if (!s.ok()) {
       return s;
     }
@@ -176,17 +184,20 @@ Status BlobFileBuilder::OpenBlobFileIfNeeded() {
   assert(file);
   file->SetIOPriority(io_priority_);
   file->SetWriteLifeTimeHint(write_hint_);
-
+  FileTypeSet tmp_set = immutable_cf_options_->checksum_handoff_file_types;
   Statistics* const statistics = immutable_cf_options_->statistics;
-
   std::unique_ptr<WritableFileWriter> file_writer(new WritableFileWriter(
-      std::move(file), blob_file_paths_->back(), *file_options_, env_,
-      nullptr /*IOTracer*/, statistics, immutable_cf_options_->listeners,
-      immutable_cf_options_->file_checksum_gen_factory));
+      std::move(file), blob_file_paths_->back(), *file_options_,
+      immutable_cf_options_->clock, io_tracer_, statistics,
+      immutable_cf_options_->listeners,
+      immutable_cf_options_->file_checksum_gen_factory.get(),
+      tmp_set.Contains(FileType::kBlobFile)));
 
-  std::unique_ptr<BlobLogWriter> blob_log_writer(
-      new BlobLogWriter(std::move(file_writer), env_, statistics,
-                        blob_file_number, immutable_cf_options_->use_fsync));
+  constexpr bool do_flush = false;
+
+  std::unique_ptr<BlobLogWriter> blob_log_writer(new BlobLogWriter(
+      std::move(file_writer), immutable_cf_options_->clock, statistics,
+      blob_file_number, immutable_cf_options_->use_fsync, do_flush));
 
   constexpr bool has_ttl = false;
   constexpr ExpirationRange expiration_range;
@@ -195,9 +206,11 @@ Status BlobFileBuilder::OpenBlobFileIfNeeded() {
                        expiration_range);
 
   {
-    TEST_SYNC_POINT("BlobFileBuilder::OpenBlobFileIfNeeded:WriteHeader");
+    Status s = blob_log_writer->WriteHeader(header);
+
+    TEST_SYNC_POINT_CALLBACK(
+        "BlobFileBuilder::OpenBlobFileIfNeeded:WriteHeader", &s);
 
-    const Status s = blob_log_writer->WriteHeader(header);
     if (!s.ok()) {
       return s;
     }
@@ -247,9 +260,10 @@ Status BlobFileBuilder::WriteBlobToFile(const Slice& key, const Slice& blob,
 
   uint64_t key_offset = 0;
 
-  TEST_SYNC_POINT("BlobFileBuilder::WriteBlobToFile:AddRecord");
+  Status s = writer_->AddRecord(key, blob, &key_offset, blob_offset);
+
+  TEST_SYNC_POINT_CALLBACK("BlobFileBuilder::WriteBlobToFile:AddRecord", &s);
 
-  const Status s = writer_->AddRecord(key, blob, &key_offset, blob_offset);
   if (!s.ok()) {
     return s;
   }
@@ -271,10 +285,10 @@ Status BlobFileBuilder::CloseBlobFile() {
   std::string checksum_method;
   std::string checksum_value;
 
-  TEST_SYNC_POINT("BlobFileBuilder::WriteBlobToFile:AppendFooter");
+  Status s = writer_->AppendFooter(footer, &checksum_method, &checksum_value);
+
+  TEST_SYNC_POINT_CALLBACK("BlobFileBuilder::WriteBlobToFile:AppendFooter", &s);
 
-  const Status s =
-      writer_->AppendFooter(footer, &checksum_method, &checksum_value);
   if (!s.ok()) {
     return s;
   }
@@ -292,12 +306,15 @@ Status BlobFileBuilder::CloseBlobFile() {
                  " total blobs, %" PRIu64 " total bytes",
                  column_family_name_.c_str(), job_id_, blob_file_number,
                  blob_count_, blob_bytes_);
+  if (blob_callback_) {
+    s = blob_callback_->OnBlobFileCompleted(blob_file_paths_->back());
+  }
 
   writer_.reset();
   blob_count_ = 0;
   blob_bytes_ = 0;
 
-  return Status::OK();
+  return s;
 }
 
 Status BlobFileBuilder::CloseBlobFileIfNeeded() {
@@ -313,4 +330,20 @@ Status BlobFileBuilder::CloseBlobFileIfNeeded() {
   return CloseBlobFile();
 }
 
+void BlobFileBuilder::Abandon() {
+  if (!IsBlobFileOpen()) {
+    return;
+  }
+
+  if (blob_callback_) {
+    // BlobFileBuilder::Abandon() is called because of error while writing to
+    // Blob files. So we can ignore the below error.
+    blob_callback_->OnBlobFileCompleted(blob_file_paths_->back())
+        .PermitUncheckedError();
+  }
+
+  writer_.reset();
+  blob_count_ = 0;
+  blob_bytes_ = 0;
+}
 }  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_builder.h

@@ -18,6 +18,7 @@ namespace ROCKSDB_NAMESPACE {
 
 class VersionSet;
 class FileSystem;
+class SystemClock;
 struct ImmutableCFOptions;
 struct MutableCFOptions;
 struct FileOptions;
@@ -25,10 +26,12 @@ class BlobFileAddition;
 class Status;
 class Slice;
 class BlobLogWriter;
+class IOTracer;
+class BlobFileCompletionCallback;
 
 class BlobFileBuilder {
  public:
-  BlobFileBuilder(VersionSet* versions, Env* env, FileSystem* fs,
+  BlobFileBuilder(VersionSet* versions, FileSystem* fs,
                   const ImmutableCFOptions* immutable_cf_options,
                   const MutableCFOptions* mutable_cf_options,
                   const FileOptions* file_options, int job_id,
@@ -36,10 +39,12 @@ class BlobFileBuilder {
                   const std::string& column_family_name,
                   Env::IOPriority io_priority,
                   Env::WriteLifeTimeHint write_hint,
+                  const std::shared_ptr<IOTracer>& io_tracer,
+                  BlobFileCompletionCallback* blob_callback,
                   std::vector<std::string>* blob_file_paths,
                   std::vector<BlobFileAddition>* blob_file_additions);
 
-  BlobFileBuilder(std::function<uint64_t()> file_number_generator, Env* env,
+  BlobFileBuilder(std::function<uint64_t()> file_number_generator,
                   FileSystem* fs,
                   const ImmutableCFOptions* immutable_cf_options,
                   const MutableCFOptions* mutable_cf_options,
@@ -48,6 +53,8 @@ class BlobFileBuilder {
                   const std::string& column_family_name,
                   Env::IOPriority io_priority,
                   Env::WriteLifeTimeHint write_hint,
+                  const std::shared_ptr<IOTracer>& io_tracer,
+                  BlobFileCompletionCallback* blob_callback,
                   std::vector<std::string>* blob_file_paths,
                   std::vector<BlobFileAddition>* blob_file_additions);
 
@@ -58,6 +65,7 @@ class BlobFileBuilder {
 
   Status Add(const Slice& key, const Slice& value, std::string* blob_index);
   Status Finish();
+  void Abandon();
 
  private:
   bool IsBlobFileOpen() const;
@@ -69,7 +77,6 @@ class BlobFileBuilder {
   Status CloseBlobFileIfNeeded();
 
   std::function<uint64_t()> file_number_generator_;
-  Env* env_;
   FileSystem* fs_;
   const ImmutableCFOptions* immutable_cf_options_;
   uint64_t min_blob_size_;
@@ -81,6 +88,8 @@ class BlobFileBuilder {
   std::string column_family_name_;
   Env::IOPriority io_priority_;
   Env::WriteLifeTimeHint write_hint_;
+  std::shared_ptr<IOTracer> io_tracer_;
+  BlobFileCompletionCallback* blob_callback_;
   std::vector<std::string>* blob_file_paths_;
   std::vector<BlobFileAddition>* blob_file_additions_;
   std::unique_ptr<BlobLogWriter> writer_;

db/blob/blob_file_builder_test.cc

@@ -15,7 +15,6 @@
 #include "db/blob/blob_index.h"
 #include "db/blob/blob_log_format.h"
 #include "db/blob/blob_log_sequential_reader.h"
-#include "env/composite_env_wrapper.h"
 #include "env/mock_env.h"
 #include "file/filename.h"
 #include "file/random_access_file_reader.h"
@@ -40,7 +39,10 @@ class TestFileNumberGenerator {
 
 class BlobFileBuilderTest : public testing::Test {
  protected:
-  BlobFileBuilderTest() : mock_env_(Env::Default()), fs_(&mock_env_) {}
+  BlobFileBuilderTest() : mock_env_(Env::Default()) {
+    fs_ = mock_env_.GetFileSystem().get();
+    clock_ = mock_env_.GetSystemClock().get();
+  }
 
   void VerifyBlobFile(uint64_t blob_file_number,
                       const std::string& blob_file_path,
@@ -54,14 +56,13 @@ class BlobFileBuilderTest : public testing::Test {
     std::unique_ptr<FSRandomAccessFile> file;
     constexpr IODebugContext* dbg = nullptr;
     ASSERT_OK(
-        fs_.NewRandomAccessFile(blob_file_path, file_options_, &file, dbg));
+        fs_->NewRandomAccessFile(blob_file_path, file_options_, &file, dbg));
 
     std::unique_ptr<RandomAccessFileReader> file_reader(
-        new RandomAccessFileReader(std::move(file), blob_file_path,
-                                   &mock_env_));
+        new RandomAccessFileReader(std::move(file), blob_file_path, clock_));
 
     constexpr Statistics* statistics = nullptr;
-    BlobLogSequentialReader blob_log_reader(std::move(file_reader), &mock_env_,
+    BlobLogSequentialReader blob_log_reader(std::move(file_reader), clock_,
                                             statistics);
 
     BlobLogHeader header;
@@ -108,7 +109,8 @@ class BlobFileBuilderTest : public testing::Test {
   }
 
   MockEnv mock_env_;
-  LegacyFileSystemWrapper fs_;
+  FileSystem* fs_;
+  SystemClock* clock_;
   FileOptions file_options_;
 };
 
@@ -125,6 +127,7 @@ TEST_F(BlobFileBuilderTest, BuildAndCheckOneFile) {
                             "BlobFileBuilderTest_BuildAndCheckOneFile"),
       0);
   options.enable_blob_files = true;
+  options.env = &mock_env_;
 
   ImmutableCFOptions immutable_cf_options(options);
   MutableCFOptions mutable_cf_options(options);
@@ -138,10 +141,11 @@ TEST_F(BlobFileBuilderTest, BuildAndCheckOneFile) {
   std::vector<std::string> blob_file_paths;
   std::vector<BlobFileAddition> blob_file_additions;
 
-  BlobFileBuilder builder(TestFileNumberGenerator(), &mock_env_, &fs_,
-                          &immutable_cf_options, &mutable_cf_options,
-                          &file_options_, job_id, column_family_id,
-                          column_family_name, io_priority, write_hint,
+  BlobFileBuilder builder(TestFileNumberGenerator(), fs_, &immutable_cf_options,
+                          &mutable_cf_options, &file_options_, job_id,
+                          column_family_id, column_family_name, io_priority,
+                          write_hint, nullptr /*IOTracer*/,
+                          nullptr /*BlobFileCompletionCallback*/,
                           &blob_file_paths, &blob_file_additions);
 
   std::vector<std::pair<std::string, std::string>> expected_key_value_pairs(
@@ -208,6 +212,7 @@ TEST_F(BlobFileBuilderTest, BuildAndCheckMultipleFiles) {
       0);
   options.enable_blob_files = true;
   options.blob_file_size = value_size;
+  options.env = &mock_env_;
 
   ImmutableCFOptions immutable_cf_options(options);
   MutableCFOptions mutable_cf_options(options);
@@ -221,10 +226,11 @@ TEST_F(BlobFileBuilderTest, BuildAndCheckMultipleFiles) {
   std::vector<std::string> blob_file_paths;
   std::vector<BlobFileAddition> blob_file_additions;
 
-  BlobFileBuilder builder(TestFileNumberGenerator(), &mock_env_, &fs_,
-                          &immutable_cf_options, &mutable_cf_options,
-                          &file_options_, job_id, column_family_id,
-                          column_family_name, io_priority, write_hint,
+  BlobFileBuilder builder(TestFileNumberGenerator(), fs_, &immutable_cf_options,
+                          &mutable_cf_options, &file_options_, job_id,
+                          column_family_id, column_family_name, io_priority,
+                          write_hint, nullptr /*IOTracer*/,
+                          nullptr /*BlobFileCompletionCallback*/,
                           &blob_file_paths, &blob_file_additions);
 
   std::vector<std::pair<std::string, std::string>> expected_key_value_pairs(
@@ -293,6 +299,7 @@ TEST_F(BlobFileBuilderTest, InlinedValues) {
       0);
   options.enable_blob_files = true;
   options.min_blob_size = 1024;
+  options.env = &mock_env_;
 
   ImmutableCFOptions immutable_cf_options(options);
   MutableCFOptions mutable_cf_options(options);
@@ -306,10 +313,11 @@ TEST_F(BlobFileBuilderTest, InlinedValues) {
   std::vector<std::string> blob_file_paths;
   std::vector<BlobFileAddition> blob_file_additions;
 
-  BlobFileBuilder builder(TestFileNumberGenerator(), &mock_env_, &fs_,
-                          &immutable_cf_options, &mutable_cf_options,
-                          &file_options_, job_id, column_family_id,
-                          column_family_name, io_priority, write_hint,
+  BlobFileBuilder builder(TestFileNumberGenerator(), fs_, &immutable_cf_options,
+                          &mutable_cf_options, &file_options_, job_id,
+                          column_family_id, column_family_name, io_priority,
+                          write_hint, nullptr /*IOTracer*/,
+                          nullptr /*BlobFileCompletionCallback*/,
                           &blob_file_paths, &blob_file_additions);
 
   for (size_t i = 0; i < number_of_blobs; ++i) {
@@ -345,6 +353,7 @@ TEST_F(BlobFileBuilderTest, Compression) {
       test::PerThreadDBPath(&mock_env_, "BlobFileBuilderTest_Compression"), 0);
   options.enable_blob_files = true;
   options.blob_compression_type = kSnappyCompression;
+  options.env = &mock_env_;
 
   ImmutableCFOptions immutable_cf_options(options);
   MutableCFOptions mutable_cf_options(options);
@@ -358,10 +367,11 @@ TEST_F(BlobFileBuilderTest, Compression) {
   std::vector<std::string> blob_file_paths;
   std::vector<BlobFileAddition> blob_file_additions;
 
-  BlobFileBuilder builder(TestFileNumberGenerator(), &mock_env_, &fs_,
-                          &immutable_cf_options, &mutable_cf_options,
-                          &file_options_, job_id, column_family_id,
-                          column_family_name, io_priority, write_hint,
+  BlobFileBuilder builder(TestFileNumberGenerator(), fs_, &immutable_cf_options,
+                          &mutable_cf_options, &file_options_, job_id,
+                          column_family_id, column_family_name, io_priority,
+                          write_hint, nullptr /*IOTracer*/,
+                          nullptr /*BlobFileCompletionCallback*/,
                           &blob_file_paths, &blob_file_additions);
 
   const std::string key("1");
@@ -427,7 +437,7 @@ TEST_F(BlobFileBuilderTest, CompressionError) {
       0);
   options.enable_blob_files = true;
   options.blob_compression_type = kSnappyCompression;
-
+  options.env = &mock_env_;
   ImmutableCFOptions immutable_cf_options(options);
   MutableCFOptions mutable_cf_options(options);
 
@@ -440,10 +450,11 @@ TEST_F(BlobFileBuilderTest, CompressionError) {
   std::vector<std::string> blob_file_paths;
   std::vector<BlobFileAddition> blob_file_additions;
 
-  BlobFileBuilder builder(TestFileNumberGenerator(), &mock_env_, &fs_,
-                          &immutable_cf_options, &mutable_cf_options,
-                          &file_options_, job_id, column_family_id,
-                          column_family_name, io_priority, write_hint,
+  BlobFileBuilder builder(TestFileNumberGenerator(), fs_, &immutable_cf_options,
+                          &mutable_cf_options, &file_options_, job_id,
+                          column_family_id, column_family_name, io_priority,
+                          write_hint, nullptr /*IOTracer*/,
+                          nullptr /*BlobFileCompletionCallback*/,
                           &blob_file_paths, &blob_file_additions);
 
   SyncPoint::GetInstance()->SetCallBack("CompressData:TamperWithReturnValue",
@@ -504,6 +515,7 @@ TEST_F(BlobFileBuilderTest, Checksum) {
   options.enable_blob_files = true;
   options.file_checksum_gen_factory =
       std::make_shared<DummyFileChecksumGenFactory>();
+  options.env = &mock_env_;
 
   ImmutableCFOptions immutable_cf_options(options);
   MutableCFOptions mutable_cf_options(options);
@@ -517,10 +529,11 @@ TEST_F(BlobFileBuilderTest, Checksum) {
   std::vector<std::string> blob_file_paths;
   std::vector<BlobFileAddition> blob_file_additions;
 
-  BlobFileBuilder builder(TestFileNumberGenerator(), &mock_env_, &fs_,
-                          &immutable_cf_options, &mutable_cf_options,
-                          &file_options_, job_id, column_family_id,
-                          column_family_name, io_priority, write_hint,
+  BlobFileBuilder builder(TestFileNumberGenerator(), fs_, &immutable_cf_options,
+                          &mutable_cf_options, &file_options_, job_id,
+                          column_family_id, column_family_name, io_priority,
+                          write_hint, nullptr /*IOTracer*/,
+                          nullptr /*BlobFileCompletionCallback*/,
                           &blob_file_paths, &blob_file_additions);
 
   const std::string key("1");
@@ -570,13 +583,11 @@ class BlobFileBuilderIOErrorTest
  protected:
   BlobFileBuilderIOErrorTest()
       : mock_env_(Env::Default()),
-        fault_injection_env_(&mock_env_),
-        fs_(&fault_injection_env_),
+        fs_(mock_env_.GetFileSystem().get()),
         sync_point_(GetParam()) {}
 
   MockEnv mock_env_;
-  FaultInjectionTestEnv fault_injection_env_;
-  LegacyFileSystemWrapper fs_;
+  FileSystem* fs_;
   FileOptions file_options_;
   std::string sync_point_;
 };
@@ -597,11 +608,11 @@ TEST_P(BlobFileBuilderIOErrorTest, IOError) {
 
   Options options;
   options.cf_paths.emplace_back(
-      test::PerThreadDBPath(&fault_injection_env_,
-                            "BlobFileBuilderIOErrorTest_IOError"),
+      test::PerThreadDBPath(&mock_env_, "BlobFileBuilderIOErrorTest_IOError"),
       0);
   options.enable_blob_files = true;
   options.blob_file_size = value_size;
+  options.env = &mock_env_;
 
   ImmutableCFOptions immutable_cf_options(options);
   MutableCFOptions mutable_cf_options(options);
@@ -615,15 +626,18 @@ TEST_P(BlobFileBuilderIOErrorTest, IOError) {
   std::vector<std::string> blob_file_paths;
   std::vector<BlobFileAddition> blob_file_additions;
 
-  BlobFileBuilder builder(TestFileNumberGenerator(), &fault_injection_env_,
-                          &fs_, &immutable_cf_options, &mutable_cf_options,
-                          &file_options_, job_id, column_family_id,
-                          column_family_name, io_priority, write_hint,
+  BlobFileBuilder builder(TestFileNumberGenerator(), fs_, &immutable_cf_options,
+                          &mutable_cf_options, &file_options_, job_id,
+                          column_family_id, column_family_name, io_priority,
+                          write_hint, nullptr /*IOTracer*/,
+                          nullptr /*BlobFileCompletionCallback*/,
                           &blob_file_paths, &blob_file_additions);
 
-  SyncPoint::GetInstance()->SetCallBack(sync_point_, [this](void* /* arg */) {
-    fault_injection_env_.SetFilesystemActive(false,
-                                             Status::IOError(sync_point_));
+  SyncPoint::GetInstance()->SetCallBack(sync_point_, [this](void* arg) {
+    Status* const s = static_cast<Status*>(arg);
+    assert(s);
+
+    (*s) = Status::IOError(sync_point_);
   });
   SyncPoint::GetInstance()->EnableProcessing();
 

db/blob/blob_file_cache.cc

@@ -0,0 +1,102 @@
+//  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/blob/blob_file_cache.h"
+
+#include <cassert>
+#include <memory>
+
+#include "db/blob/blob_file_reader.h"
+#include "options/cf_options.h"
+#include "rocksdb/cache.h"
+#include "rocksdb/slice.h"
+#include "test_util/sync_point.h"
+#include "trace_replay/io_tracer.h"
+#include "util/hash.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+BlobFileCache::BlobFileCache(Cache* cache,
+                             const ImmutableCFOptions* immutable_cf_options,
+                             const FileOptions* file_options,
+                             uint32_t column_family_id,
+                             HistogramImpl* blob_file_read_hist,
+                             const std::shared_ptr<IOTracer>& io_tracer)
+    : cache_(cache),
+      mutex_(kNumberOfMutexStripes, kGetSliceNPHash64UnseededFnPtr),
+      immutable_cf_options_(immutable_cf_options),
+      file_options_(file_options),
+      column_family_id_(column_family_id),
+      blob_file_read_hist_(blob_file_read_hist),
+      io_tracer_(io_tracer) {
+  assert(cache_);
+  assert(immutable_cf_options_);
+  assert(file_options_);
+}
+
+Status BlobFileCache::GetBlobFileReader(
+    uint64_t blob_file_number,
+    CacheHandleGuard<BlobFileReader>* blob_file_reader) {
+  assert(blob_file_reader);
+  assert(blob_file_reader->IsEmpty());
+
+  const Slice key = GetSlice(&blob_file_number);
+
+  assert(cache_);
+
+  Cache::Handle* handle = cache_->Lookup(key);
+  if (handle) {
+    *blob_file_reader = CacheHandleGuard<BlobFileReader>(cache_, handle);
+    return Status::OK();
+  }
+
+  TEST_SYNC_POINT("BlobFileCache::GetBlobFileReader:DoubleCheck");
+
+  // Check again while holding mutex
+  MutexLock lock(mutex_.get(key));
+
+  handle = cache_->Lookup(key);
+  if (handle) {
+    *blob_file_reader = CacheHandleGuard<BlobFileReader>(cache_, handle);
+    return Status::OK();
+  }
+
+  assert(immutable_cf_options_);
+  Statistics* const statistics = immutable_cf_options_->statistics;
+
+  RecordTick(statistics, NO_FILE_OPENS);
+
+  std::unique_ptr<BlobFileReader> reader;
+
+  {
+    assert(file_options_);
+    const Status s = BlobFileReader::Create(
+        *immutable_cf_options_, *file_options_, column_family_id_,
+        blob_file_read_hist_, blob_file_number, io_tracer_, &reader);
+    if (!s.ok()) {
+      RecordTick(statistics, NO_FILE_ERRORS);
+      return s;
+    }
+  }
+
+  {
+    constexpr size_t charge = 1;
+
+    const Status s = cache_->Insert(key, reader.get(), charge,
+                                    &DeleteCacheEntry<BlobFileReader>, &handle);
+    if (!s.ok()) {
+      RecordTick(statistics, NO_FILE_ERRORS);
+      return s;
+    }
+  }
+
+  reader.release();
+
+  *blob_file_reader = CacheHandleGuard<BlobFileReader>(cache_, handle);
+
+  return Status::OK();
+}
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_cache.h

@@ -0,0 +1,52 @@
+//  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
+
+#include <cinttypes>
+
+#include "cache/cache_helpers.h"
+#include "rocksdb/rocksdb_namespace.h"
+#include "util/mutexlock.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class Cache;
+struct ImmutableCFOptions;
+struct FileOptions;
+class HistogramImpl;
+class Status;
+class BlobFileReader;
+class Slice;
+class IOTracer;
+
+class BlobFileCache {
+ public:
+  BlobFileCache(Cache* cache, const ImmutableCFOptions* immutable_cf_options,
+                const FileOptions* file_options, uint32_t column_family_id,
+                HistogramImpl* blob_file_read_hist,
+                const std::shared_ptr<IOTracer>& io_tracer);
+
+  BlobFileCache(const BlobFileCache&) = delete;
+  BlobFileCache& operator=(const BlobFileCache&) = delete;
+
+  Status GetBlobFileReader(uint64_t blob_file_number,
+                           CacheHandleGuard<BlobFileReader>* blob_file_reader);
+
+ private:
+  Cache* cache_;
+  // Note: mutex_ below is used to guard against multiple threads racing to open
+  // the same file.
+  Striped<port::Mutex, Slice> mutex_;
+  const ImmutableCFOptions* immutable_cf_options_;
+  const FileOptions* file_options_;
+  uint32_t column_family_id_;
+  HistogramImpl* blob_file_read_hist_;
+  std::shared_ptr<IOTracer> io_tracer_;
+
+  static constexpr size_t kNumberOfMutexStripes = 1 << 7;
+};
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_cache_test.cc

@@ -0,0 +1,268 @@
+//  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/blob/blob_file_cache.h"
+
+#include <cassert>
+#include <string>
+
+#include "db/blob/blob_log_format.h"
+#include "db/blob/blob_log_writer.h"
+#include "env/mock_env.h"
+#include "file/filename.h"
+#include "file/read_write_util.h"
+#include "file/writable_file_writer.h"
+#include "options/cf_options.h"
+#include "rocksdb/cache.h"
+#include "rocksdb/env.h"
+#include "rocksdb/file_system.h"
+#include "rocksdb/options.h"
+#include "rocksdb/statistics.h"
+#include "test_util/sync_point.h"
+#include "test_util/testharness.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+namespace {
+
+// Creates a test blob file with a single blob in it.
+void WriteBlobFile(uint32_t column_family_id,
+                   const ImmutableCFOptions& immutable_cf_options,
+                   uint64_t blob_file_number) {
+  assert(!immutable_cf_options.cf_paths.empty());
+
+  const std::string blob_file_path = BlobFileName(
+      immutable_cf_options.cf_paths.front().path, blob_file_number);
+
+  std::unique_ptr<FSWritableFile> file;
+  ASSERT_OK(NewWritableFile(immutable_cf_options.fs, blob_file_path, &file,
+                            FileOptions()));
+
+  std::unique_ptr<WritableFileWriter> file_writer(
+      new WritableFileWriter(std::move(file), blob_file_path, FileOptions(),
+                             immutable_cf_options.clock));
+
+  constexpr Statistics* statistics = nullptr;
+  constexpr bool use_fsync = false;
+  constexpr bool do_flush = false;
+
+  BlobLogWriter blob_log_writer(std::move(file_writer),
+                                immutable_cf_options.clock, statistics,
+                                blob_file_number, use_fsync, do_flush);
+
+  constexpr bool has_ttl = false;
+  constexpr ExpirationRange expiration_range;
+
+  BlobLogHeader header(column_family_id, kNoCompression, has_ttl,
+                       expiration_range);
+
+  ASSERT_OK(blob_log_writer.WriteHeader(header));
+
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+
+  std::string compressed_blob;
+
+  uint64_t key_offset = 0;
+  uint64_t blob_offset = 0;
+
+  ASSERT_OK(blob_log_writer.AddRecord(key, blob, &key_offset, &blob_offset));
+
+  BlobLogFooter footer;
+  footer.blob_count = 1;
+  footer.expiration_range = expiration_range;
+
+  std::string checksum_method;
+  std::string checksum_value;
+
+  ASSERT_OK(
+      blob_log_writer.AppendFooter(footer, &checksum_method, &checksum_value));
+}
+
+}  // anonymous namespace
+
+class BlobFileCacheTest : public testing::Test {
+ protected:
+  BlobFileCacheTest() : mock_env_(Env::Default()) {}
+
+  MockEnv mock_env_;
+};
+
+TEST_F(BlobFileCacheTest, GetBlobFileReader) {
+  Options options;
+  options.env = &mock_env_;
+  options.statistics = CreateDBStatistics();
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_, "BlobFileCacheTest_GetBlobFileReader"),
+      0);
+  options.enable_blob_files = true;
+
+  constexpr uint32_t column_family_id = 1;
+  ImmutableCFOptions immutable_cf_options(options);
+  constexpr uint64_t blob_file_number = 123;
+
+  WriteBlobFile(column_family_id, immutable_cf_options, blob_file_number);
+
+  constexpr size_t capacity = 10;
+  std::shared_ptr<Cache> backing_cache = NewLRUCache(capacity);
+
+  FileOptions file_options;
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  BlobFileCache blob_file_cache(backing_cache.get(), &immutable_cf_options,
+                                &file_options, column_family_id,
+                                blob_file_read_hist, nullptr /*IOTracer*/);
+
+  // First try: reader should be opened and put in cache
+  CacheHandleGuard<BlobFileReader> first;
+
+  ASSERT_OK(blob_file_cache.GetBlobFileReader(blob_file_number, &first));
+  ASSERT_NE(first.GetValue(), nullptr);
+  ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_OPENS), 1);
+  ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_ERRORS), 0);
+
+  // Second try: reader should be served from cache
+  CacheHandleGuard<BlobFileReader> second;
+
+  ASSERT_OK(blob_file_cache.GetBlobFileReader(blob_file_number, &second));
+  ASSERT_NE(second.GetValue(), nullptr);
+  ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_OPENS), 1);
+  ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_ERRORS), 0);
+
+  ASSERT_EQ(first.GetValue(), second.GetValue());
+}
+
+TEST_F(BlobFileCacheTest, GetBlobFileReader_Race) {
+  Options options;
+  options.env = &mock_env_;
+  options.statistics = CreateDBStatistics();
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_,
+                            "BlobFileCacheTest_GetBlobFileReader_Race"),
+      0);
+  options.enable_blob_files = true;
+
+  constexpr uint32_t column_family_id = 1;
+  ImmutableCFOptions immutable_cf_options(options);
+  constexpr uint64_t blob_file_number = 123;
+
+  WriteBlobFile(column_family_id, immutable_cf_options, blob_file_number);
+
+  constexpr size_t capacity = 10;
+  std::shared_ptr<Cache> backing_cache = NewLRUCache(capacity);
+
+  FileOptions file_options;
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  BlobFileCache blob_file_cache(backing_cache.get(), &immutable_cf_options,
+                                &file_options, column_family_id,
+                                blob_file_read_hist, nullptr /*IOTracer*/);
+
+  CacheHandleGuard<BlobFileReader> first;
+  CacheHandleGuard<BlobFileReader> second;
+
+  SyncPoint::GetInstance()->SetCallBack(
+      "BlobFileCache::GetBlobFileReader:DoubleCheck", [&](void* /* arg */) {
+        // Disabling sync points to prevent infinite recursion
+        SyncPoint::GetInstance()->DisableProcessing();
+
+        ASSERT_OK(blob_file_cache.GetBlobFileReader(blob_file_number, &second));
+        ASSERT_NE(second.GetValue(), nullptr);
+        ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_OPENS), 1);
+        ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_ERRORS), 0);
+      });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  ASSERT_OK(blob_file_cache.GetBlobFileReader(blob_file_number, &first));
+  ASSERT_NE(first.GetValue(), nullptr);
+  ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_OPENS), 1);
+  ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_ERRORS), 0);
+
+  ASSERT_EQ(first.GetValue(), second.GetValue());
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+TEST_F(BlobFileCacheTest, GetBlobFileReader_IOError) {
+  Options options;
+  options.env = &mock_env_;
+  options.statistics = CreateDBStatistics();
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_,
+                            "BlobFileCacheTest_GetBlobFileReader_IOError"),
+      0);
+  options.enable_blob_files = true;
+
+  constexpr size_t capacity = 10;
+  std::shared_ptr<Cache> backing_cache = NewLRUCache(capacity);
+
+  ImmutableCFOptions immutable_cf_options(options);
+  FileOptions file_options;
+  constexpr uint32_t column_family_id = 1;
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  BlobFileCache blob_file_cache(backing_cache.get(), &immutable_cf_options,
+                                &file_options, column_family_id,
+                                blob_file_read_hist, nullptr /*IOTracer*/);
+
+  // Note: there is no blob file with the below number
+  constexpr uint64_t blob_file_number = 123;
+
+  CacheHandleGuard<BlobFileReader> reader;
+
+  ASSERT_TRUE(
+      blob_file_cache.GetBlobFileReader(blob_file_number, &reader).IsIOError());
+  ASSERT_EQ(reader.GetValue(), nullptr);
+  ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_OPENS), 1);
+  ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_ERRORS), 1);
+}
+
+TEST_F(BlobFileCacheTest, GetBlobFileReader_CacheFull) {
+  Options options;
+  options.env = &mock_env_;
+  options.statistics = CreateDBStatistics();
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_,
+                            "BlobFileCacheTest_GetBlobFileReader_CacheFull"),
+      0);
+  options.enable_blob_files = true;
+
+  constexpr uint32_t column_family_id = 1;
+  ImmutableCFOptions immutable_cf_options(options);
+  constexpr uint64_t blob_file_number = 123;
+
+  WriteBlobFile(column_family_id, immutable_cf_options, blob_file_number);
+
+  constexpr size_t capacity = 0;
+  constexpr int num_shard_bits = -1;  // determined automatically
+  constexpr bool strict_capacity_limit = true;
+  std::shared_ptr<Cache> backing_cache =
+      NewLRUCache(capacity, num_shard_bits, strict_capacity_limit);
+
+  FileOptions file_options;
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  BlobFileCache blob_file_cache(backing_cache.get(), &immutable_cf_options,
+                                &file_options, column_family_id,
+                                blob_file_read_hist, nullptr /*IOTracer*/);
+
+  // Insert into cache should fail since it has zero capacity and
+  // strict_capacity_limit is set
+  CacheHandleGuard<BlobFileReader> reader;
+
+  ASSERT_TRUE(blob_file_cache.GetBlobFileReader(blob_file_number, &reader)
+                  .IsIncomplete());
+  ASSERT_EQ(reader.GetValue(), nullptr);
+  ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_OPENS), 1);
+  ASSERT_EQ(options.statistics->getTickerCount(NO_FILE_ERRORS), 1);
+}
+
+}  // namespace ROCKSDB_NAMESPACE
+
+int main(int argc, char** argv) {
+  ::testing::InitGoogleTest(&argc, argv);
+  return RUN_ALL_TESTS();
+}

db/blob/blob_file_completion_callback.h

@@ -0,0 +1,57 @@
+//  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
+
+#include "db/error_handler.h"
+#include "file/sst_file_manager_impl.h"
+#include "rocksdb/status.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class BlobFileCompletionCallback {
+ public:
+#ifdef ROCKSDB_LITE
+  BlobFileCompletionCallback(SstFileManager* /*sst_file_manager*/,
+                             InstrumentedMutex* /*mutex*/,
+                             ErrorHandler* /*error_handler*/) {}
+  Status OnBlobFileCompleted(const std::string& /*file_name*/) {
+    return Status::OK();
+  }
+#else
+  BlobFileCompletionCallback(SstFileManager* sst_file_manager,
+                             InstrumentedMutex* mutex,
+                             ErrorHandler* error_handler)
+      : sst_file_manager_(sst_file_manager),
+        mutex_(mutex),
+        error_handler_(error_handler) {}
+
+  Status OnBlobFileCompleted(const std::string& file_name) {
+    Status s;
+    auto sfm = static_cast<SstFileManagerImpl*>(sst_file_manager_);
+    if (sfm) {
+      // Report new blob files to SstFileManagerImpl
+      s = sfm->OnAddFile(file_name);
+      if (sfm->IsMaxAllowedSpaceReached()) {
+        s = Status::SpaceLimit("Max allowed space was reached");
+        TEST_SYNC_POINT(
+            "BlobFileCompletionCallback::CallBack::MaxAllowedSpaceReached");
+        InstrumentedMutexLock l(mutex_);
+        error_handler_->SetBGError(s, BackgroundErrorReason::kFlush);
+      }
+    }
+    return s;
+  }
+
+ private:
+  SstFileManager* sst_file_manager_;
+  InstrumentedMutex* mutex_;
+  ErrorHandler* error_handler_;
+#endif  // ROCKSDB_LITE
+};
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_reader.cc

@@ -24,6 +24,7 @@ Status BlobFileReader::Create(
     const ImmutableCFOptions& immutable_cf_options,
     const FileOptions& file_options, uint32_t column_family_id,
     HistogramImpl* blob_file_read_hist, uint64_t blob_file_number,
+    const std::shared_ptr<IOTracer>& io_tracer,
     std::unique_ptr<BlobFileReader>* blob_file_reader) {
   assert(blob_file_reader);
   assert(!*blob_file_reader);
@@ -34,7 +35,7 @@ Status BlobFileReader::Create(
   {
     const Status s =
         OpenFile(immutable_cf_options, file_options, blob_file_read_hist,
-                 blob_file_number, &file_size, &file_reader);
+                 blob_file_number, io_tracer, &file_size, &file_reader);
     if (!s.ok()) {
       return s;
     }
@@ -68,8 +69,8 @@ Status BlobFileReader::Create(
 Status BlobFileReader::OpenFile(
     const ImmutableCFOptions& immutable_cf_options,
     const FileOptions& file_opts, HistogramImpl* blob_file_read_hist,
-    uint64_t blob_file_number, uint64_t* file_size,
-    std::unique_ptr<RandomAccessFileReader>* file_reader) {
+    uint64_t blob_file_number, const std::shared_ptr<IOTracer>& io_tracer,
+    uint64_t* file_size, std::unique_ptr<RandomAccessFileReader>* file_reader) {
   assert(file_size);
   assert(file_reader);
 
@@ -117,10 +118,10 @@ Status BlobFileReader::OpenFile(
   }
 
   file_reader->reset(new RandomAccessFileReader(
-      std::move(file), blob_file_path, immutable_cf_options.env,
-      std::shared_ptr<IOTracer>(), immutable_cf_options.statistics,
-      BLOB_DB_BLOB_FILE_READ_MICROS, blob_file_read_hist,
-      immutable_cf_options.rate_limiter, immutable_cf_options.listeners));
+      std::move(file), blob_file_path, immutable_cf_options.clock, io_tracer,
+      immutable_cf_options.statistics, BLOB_DB_BLOB_FILE_READ_MICROS,
+      blob_file_read_hist, immutable_cf_options.rate_limiter.get(),
+      immutable_cf_options.listeners));
 
   return Status::OK();
 }
@@ -269,7 +270,8 @@ Status BlobFileReader::GetBlob(const ReadOptions& read_options,
                                const Slice& user_key, uint64_t offset,
                                uint64_t value_size,
                                CompressionType compression_type,
-                               PinnableSlice* value) const {
+                               PinnableSlice* value,
+                               uint64_t* bytes_read) const {
   assert(value);
 
   const uint64_t key_size = user_key.size();
@@ -292,6 +294,9 @@ Status BlobFileReader::GetBlob(const ReadOptions& read_options,
           : 0;
   assert(offset >= adjustment);
 
+  const uint64_t record_offset = offset - adjustment;
+  const uint64_t record_size = value_size + adjustment;
+
   Slice record_slice;
   Buffer buf;
   AlignedBuf aligned_buf;
@@ -299,9 +304,6 @@ Status BlobFileReader::GetBlob(const ReadOptions& read_options,
   {
     TEST_SYNC_POINT("BlobFileReader::GetBlob:ReadFromFile");
 
-    const uint64_t record_offset = offset - adjustment;
-    const uint64_t record_size = value_size + adjustment;
-
     const Status s = ReadFromFile(file_reader_.get(), record_offset,
                                   static_cast<size_t>(record_size),
                                   &record_slice, &buf, &aligned_buf);
@@ -330,6 +332,10 @@ Status BlobFileReader::GetBlob(const ReadOptions& read_options,
     }
   }
 
+  if (bytes_read) {
+    *bytes_read = record_size;
+  }
+
   return Status::OK();
 }
 

db/blob/blob_file_reader.h

@@ -29,6 +29,7 @@ class BlobFileReader {
                        uint32_t column_family_id,
                        HistogramImpl* blob_file_read_hist,
                        uint64_t blob_file_number,
+                       const std::shared_ptr<IOTracer>& io_tracer,
                        std::unique_ptr<BlobFileReader>* reader);
 
   BlobFileReader(const BlobFileReader&) = delete;
@@ -38,7 +39,8 @@ class BlobFileReader {
 
   Status GetBlob(const ReadOptions& read_options, const Slice& user_key,
                  uint64_t offset, uint64_t value_size,
-                 CompressionType compression_type, PinnableSlice* value) const;
+                 CompressionType compression_type, PinnableSlice* value,
+                 uint64_t* bytes_read) const;
 
  private:
   BlobFileReader(std::unique_ptr<RandomAccessFileReader>&& file_reader,
@@ -47,7 +49,9 @@ class BlobFileReader {
   static Status OpenFile(const ImmutableCFOptions& immutable_cf_options,
                          const FileOptions& file_opts,
                          HistogramImpl* blob_file_read_hist,
-                         uint64_t blob_file_number, uint64_t* file_size,
+                         uint64_t blob_file_number,
+                         const std::shared_ptr<IOTracer>& io_tracer,
+                         uint64_t* file_size,
                          std::unique_ptr<RandomAccessFileReader>* file_reader);
 
   static Status ReadHeader(const RandomAccessFileReader* file_reader,

db/blob/blob_file_reader_test.cc

@@ -18,6 +18,7 @@
 #include "rocksdb/env.h"
 #include "rocksdb/file_system.h"
 #include "rocksdb/options.h"
+#include "test_util/sync_point.h"
 #include "test_util/testharness.h"
 #include "util/compression.h"
 #include "utilities/fault_injection_env.h"
@@ -49,14 +50,15 @@ void WriteBlobFile(const ImmutableCFOptions& immutable_cf_options,
 
   std::unique_ptr<WritableFileWriter> file_writer(
       new WritableFileWriter(std::move(file), blob_file_path, FileOptions(),
-                             immutable_cf_options.env));
+                             immutable_cf_options.clock));
 
   constexpr Statistics* statistics = nullptr;
   constexpr bool use_fsync = false;
+  constexpr bool do_flush = false;
 
   BlobLogWriter blob_log_writer(std::move(file_writer),
-                                immutable_cf_options.env, statistics,
-                                blob_file_number, use_fsync);
+                                immutable_cf_options.clock, statistics,
+                                blob_file_number, use_fsync, do_flush);
 
   BlobLogHeader header(column_family_id, compression_type, has_ttl,
                        expiration_range_header);
@@ -140,9 +142,9 @@ TEST_F(BlobFileReaderTest, CreateReaderAndGetBlob) {
 
   std::unique_ptr<BlobFileReader> reader;
 
-  ASSERT_OK(BlobFileReader::Create(immutable_cf_options, FileOptions(),
-                                   column_family_id, blob_file_read_hist,
-                                   blob_file_number, &reader));
+  ASSERT_OK(BlobFileReader::Create(
+      immutable_cf_options, FileOptions(), column_family_id,
+      blob_file_read_hist, blob_file_number, nullptr /*IOTracer*/, &reader));
 
   // Make sure the blob can be retrieved with and without checksum verification
   ReadOptions read_options;
@@ -150,83 +152,103 @@ TEST_F(BlobFileReaderTest, CreateReaderAndGetBlob) {
 
   {
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
     ASSERT_OK(reader->GetBlob(read_options, key, blob_offset, blob_size,
-                              kNoCompression, &value));
+                              kNoCompression, &value, &bytes_read));
     ASSERT_EQ(value, blob);
+    ASSERT_EQ(bytes_read, blob_size);
   }
 
   read_options.verify_checksums = true;
 
   {
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
     ASSERT_OK(reader->GetBlob(read_options, key, blob_offset, blob_size,
-                              kNoCompression, &value));
+                              kNoCompression, &value, &bytes_read));
     ASSERT_EQ(value, blob);
+
+    constexpr uint64_t key_size = sizeof(key) - 1;
+    ASSERT_EQ(bytes_read,
+              BlobLogRecord::CalculateAdjustmentForRecordHeader(key_size) +
+                  blob_size);
   }
 
   // Invalid offset (too close to start of file)
   {
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
     ASSERT_TRUE(reader
                     ->GetBlob(read_options, key, blob_offset - 1, blob_size,
-                              kNoCompression, &value)
+                              kNoCompression, &value, &bytes_read)
                     .IsCorruption());
+    ASSERT_EQ(bytes_read, 0);
   }
 
   // Invalid offset (too close to end of file)
   {
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
     ASSERT_TRUE(reader
                     ->GetBlob(read_options, key, blob_offset + 1, blob_size,
-                              kNoCompression, &value)
+                              kNoCompression, &value, &bytes_read)
                     .IsCorruption());
+    ASSERT_EQ(bytes_read, 0);
   }
 
   // Incorrect compression type
   {
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
-    ASSERT_TRUE(
-        reader
-            ->GetBlob(read_options, key, blob_offset, blob_size, kZSTD, &value)
-            .IsCorruption());
+    ASSERT_TRUE(reader
+                    ->GetBlob(read_options, key, blob_offset, blob_size, kZSTD,
+                              &value, &bytes_read)
+                    .IsCorruption());
+    ASSERT_EQ(bytes_read, 0);
   }
 
   // Incorrect key size
   {
     constexpr char shorter_key[] = "k";
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
     ASSERT_TRUE(reader
                     ->GetBlob(read_options, shorter_key,
                               blob_offset - (sizeof(key) - sizeof(shorter_key)),
-                              blob_size, kNoCompression, &value)
+                              blob_size, kNoCompression, &value, &bytes_read)
                     .IsCorruption());
+    ASSERT_EQ(bytes_read, 0);
   }
 
   // Incorrect key
   {
     constexpr char incorrect_key[] = "foo";
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
     ASSERT_TRUE(reader
                     ->GetBlob(read_options, incorrect_key, blob_offset,
-                              blob_size, kNoCompression, &value)
+                              blob_size, kNoCompression, &value, &bytes_read)
                     .IsCorruption());
+    ASSERT_EQ(bytes_read, 0);
   }
 
   // Incorrect value size
   {
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
     ASSERT_TRUE(reader
                     ->GetBlob(read_options, key, blob_offset, blob_size + 1,
-                              kNoCompression, &value)
+                              kNoCompression, &value, &bytes_read)
                     .IsCorruption());
+    ASSERT_EQ(bytes_read, 0);
   }
 }
 
@@ -258,14 +280,15 @@ TEST_F(BlobFileReaderTest, Malformed) {
 
     std::unique_ptr<WritableFileWriter> file_writer(
         new WritableFileWriter(std::move(file), blob_file_path, FileOptions(),
-                               immutable_cf_options.env));
+                               immutable_cf_options.clock));
 
     constexpr Statistics* statistics = nullptr;
     constexpr bool use_fsync = false;
+    constexpr bool do_flush = false;
 
     BlobLogWriter blob_log_writer(std::move(file_writer),
-                                  immutable_cf_options.env, statistics,
-                                  blob_file_number, use_fsync);
+                                  immutable_cf_options.clock, statistics,
+                                  blob_file_number, use_fsync, do_flush);
 
     BlobLogHeader header(column_family_id, kNoCompression, has_ttl,
                          expiration_range);
@@ -279,7 +302,8 @@ TEST_F(BlobFileReaderTest, Malformed) {
 
   ASSERT_TRUE(BlobFileReader::Create(immutable_cf_options, FileOptions(),
                                      column_family_id, blob_file_read_hist,
-                                     blob_file_number, &reader)
+                                     blob_file_number, nullptr /*IOTracer*/,
+                                     &reader)
                   .IsCorruption());
 }
 
@@ -312,7 +336,8 @@ TEST_F(BlobFileReaderTest, TTL) {
 
   ASSERT_TRUE(BlobFileReader::Create(immutable_cf_options, FileOptions(),
                                      column_family_id, blob_file_read_hist,
-                                     blob_file_number, &reader)
+                                     blob_file_number, nullptr /*IOTracer*/,
+                                     &reader)
                   .IsCorruption());
 }
 
@@ -350,7 +375,8 @@ TEST_F(BlobFileReaderTest, ExpirationRangeInHeader) {
 
   ASSERT_TRUE(BlobFileReader::Create(immutable_cf_options, FileOptions(),
                                      column_family_id, blob_file_read_hist,
-                                     blob_file_number, &reader)
+                                     blob_file_number, nullptr /*IOTracer*/,
+                                     &reader)
                   .IsCorruption());
 }
 
@@ -388,7 +414,8 @@ TEST_F(BlobFileReaderTest, ExpirationRangeInFooter) {
 
   ASSERT_TRUE(BlobFileReader::Create(immutable_cf_options, FileOptions(),
                                      column_family_id, blob_file_read_hist,
-                                     blob_file_number, &reader)
+                                     blob_file_number, nullptr /*IOTracer*/,
+                                     &reader)
                   .IsCorruption());
 }
 
@@ -426,7 +453,7 @@ TEST_F(BlobFileReaderTest, IncorrectColumnFamily) {
   ASSERT_TRUE(BlobFileReader::Create(immutable_cf_options, FileOptions(),
                                      incorrect_column_family_id,
                                      blob_file_read_hist, blob_file_number,
-                                     &reader)
+                                     nullptr /*IOTracer*/, &reader)
                   .IsCorruption());
 }
 
@@ -457,9 +484,9 @@ TEST_F(BlobFileReaderTest, BlobCRCError) {
 
   std::unique_ptr<BlobFileReader> reader;
 
-  ASSERT_OK(BlobFileReader::Create(immutable_cf_options, FileOptions(),
-                                   column_family_id, blob_file_read_hist,
-                                   blob_file_number, &reader));
+  ASSERT_OK(BlobFileReader::Create(
+      immutable_cf_options, FileOptions(), column_family_id,
+      blob_file_read_hist, blob_file_number, nullptr /*IOTracer*/, &reader));
 
   SyncPoint::GetInstance()->SetCallBack(
       "BlobFileReader::VerifyBlob:CheckBlobCRC", [](void* arg) {
@@ -472,11 +499,13 @@ TEST_F(BlobFileReaderTest, BlobCRCError) {
   SyncPoint::GetInstance()->EnableProcessing();
 
   PinnableSlice value;
+  uint64_t bytes_read = 0;
 
   ASSERT_TRUE(reader
                   ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
-                            kNoCompression, &value)
+                            kNoCompression, &value, &bytes_read)
                   .IsCorruption());
+  ASSERT_EQ(bytes_read, 0);
 
   SyncPoint::GetInstance()->DisableProcessing();
   SyncPoint::GetInstance()->ClearAllCallBacks();
@@ -513,9 +542,9 @@ TEST_F(BlobFileReaderTest, Compression) {
 
   std::unique_ptr<BlobFileReader> reader;
 
-  ASSERT_OK(BlobFileReader::Create(immutable_cf_options, FileOptions(),
-                                   column_family_id, blob_file_read_hist,
-                                   blob_file_number, &reader));
+  ASSERT_OK(BlobFileReader::Create(
+      immutable_cf_options, FileOptions(), column_family_id,
+      blob_file_read_hist, blob_file_number, nullptr /*IOTracer*/, &reader));
 
   // Make sure the blob can be retrieved with and without checksum verification
   ReadOptions read_options;
@@ -523,20 +552,28 @@ TEST_F(BlobFileReaderTest, Compression) {
 
   {
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
     ASSERT_OK(reader->GetBlob(read_options, key, blob_offset, blob_size,
-                              kSnappyCompression, &value));
+                              kSnappyCompression, &value, &bytes_read));
     ASSERT_EQ(value, blob);
+    ASSERT_EQ(bytes_read, blob_size);
   }
 
   read_options.verify_checksums = true;
 
   {
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
     ASSERT_OK(reader->GetBlob(read_options, key, blob_offset, blob_size,
-                              kSnappyCompression, &value));
+                              kSnappyCompression, &value, &bytes_read));
     ASSERT_EQ(value, blob);
+
+    constexpr uint64_t key_size = sizeof(key) - 1;
+    ASSERT_EQ(bytes_read,
+              BlobLogRecord::CalculateAdjustmentForRecordHeader(key_size) +
+                  blob_size);
   }
 }
 
@@ -573,9 +610,9 @@ TEST_F(BlobFileReaderTest, UncompressionError) {
 
   std::unique_ptr<BlobFileReader> reader;
 
-  ASSERT_OK(BlobFileReader::Create(immutable_cf_options, FileOptions(),
-                                   column_family_id, blob_file_read_hist,
-                                   blob_file_number, &reader));
+  ASSERT_OK(BlobFileReader::Create(
+      immutable_cf_options, FileOptions(), column_family_id,
+      blob_file_read_hist, blob_file_number, nullptr /*IOTracer*/, &reader));
 
   SyncPoint::GetInstance()->SetCallBack(
       "BlobFileReader::UncompressBlobIfNeeded:TamperWithResult", [](void* arg) {
@@ -589,11 +626,13 @@ TEST_F(BlobFileReaderTest, UncompressionError) {
   SyncPoint::GetInstance()->EnableProcessing();
 
   PinnableSlice value;
+  uint64_t bytes_read = 0;
 
   ASSERT_TRUE(reader
                   ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
-                            kSnappyCompression, &value)
+                            kSnappyCompression, &value, &bytes_read)
                   .IsCorruption());
+  ASSERT_EQ(bytes_read, 0);
 
   SyncPoint::GetInstance()->DisableProcessing();
   SyncPoint::GetInstance()->ClearAllCallBacks();
@@ -658,9 +697,9 @@ TEST_P(BlobFileReaderIOErrorTest, IOError) {
 
   std::unique_ptr<BlobFileReader> reader;
 
-  const Status s = BlobFileReader::Create(immutable_cf_options, FileOptions(),
-                                          column_family_id, blob_file_read_hist,
-                                          blob_file_number, &reader);
+  const Status s = BlobFileReader::Create(
+      immutable_cf_options, FileOptions(), column_family_id,
+      blob_file_read_hist, blob_file_number, nullptr /*IOTracer*/, &reader);
 
   const bool fail_during_create =
       (sync_point_ != "BlobFileReader::GetBlob:ReadFromFile");
@@ -671,11 +710,13 @@ TEST_P(BlobFileReaderIOErrorTest, IOError) {
     ASSERT_OK(s);
 
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
     ASSERT_TRUE(reader
                     ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
-                              kNoCompression, &value)
+                              kNoCompression, &value, &bytes_read)
                     .IsIOError());
+    ASSERT_EQ(bytes_read, 0);
   }
 
   SyncPoint::GetInstance()->DisableProcessing();
@@ -738,9 +779,9 @@ TEST_P(BlobFileReaderDecodingErrorTest, DecodingError) {
 
   std::unique_ptr<BlobFileReader> reader;
 
-  const Status s = BlobFileReader::Create(immutable_cf_options, FileOptions(),
-                                          column_family_id, blob_file_read_hist,
-                                          blob_file_number, &reader);
+  const Status s = BlobFileReader::Create(
+      immutable_cf_options, FileOptions(), column_family_id,
+      blob_file_read_hist, blob_file_number, nullptr /*IOTracer*/, &reader);
 
   const bool fail_during_create =
       sync_point_ != "BlobFileReader::GetBlob:TamperWithResult";
@@ -751,11 +792,13 @@ TEST_P(BlobFileReaderDecodingErrorTest, DecodingError) {
     ASSERT_OK(s);
 
     PinnableSlice value;
+    uint64_t bytes_read = 0;
 
     ASSERT_TRUE(reader
                     ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
-                              kNoCompression, &value)
+                              kNoCompression, &value, &bytes_read)
                     .IsCorruption());
+    ASSERT_EQ(bytes_read, 0);
   }
 
   SyncPoint::GetInstance()->DisableProcessing();

db/blob/blob_log_format.cc

@@ -95,10 +95,6 @@ Status BlobLogFooter::DecodeFrom(Slice src) {
   return Status::OK();
 }
 
-uint64_t BlobLogRecord::CalculateAdjustmentForRecordHeader(uint64_t key_size) {
-  return key_size + kHeaderSize;
-}
-
 void BlobLogRecord::EncodeHeaderTo(std::string* dst) {
   assert(dst != nullptr);
   dst->clear();

db/blob/blob_log_format.h

@@ -107,7 +107,9 @@ struct BlobLogRecord {
   // Note that the offset field of BlobIndex actually points to the blob value
   // as opposed to the start of the blob record. The following method can
   // be used to calculate the adjustment needed to read the blob record header.
-  static uint64_t CalculateAdjustmentForRecordHeader(uint64_t key_size);
+  static uint64_t CalculateAdjustmentForRecordHeader(uint64_t key_size) {
+    return key_size + kHeaderSize;
+  }
 
   uint64_t key_size = 0;
   uint64_t value_size = 0;

db/blob/blob_log_sequential_reader.cc

@@ -6,8 +6,6 @@
 
 #include "db/blob/blob_log_sequential_reader.h"
 
-#include <algorithm>
-
 #include "file/random_access_file_reader.h"
 #include "monitoring/statistics.h"
 #include "util/stop_watch.h"
@@ -15,10 +13,10 @@
 namespace ROCKSDB_NAMESPACE {
 
 BlobLogSequentialReader::BlobLogSequentialReader(
-    std::unique_ptr<RandomAccessFileReader>&& file_reader, Env* env,
+    std::unique_ptr<RandomAccessFileReader>&& file_reader, SystemClock* clock,
     Statistics* statistics)
     : file_(std::move(file_reader)),
-      env_(env),
+      clock_(clock),
       statistics_(statistics),
       next_byte_(0) {}
 
@@ -29,7 +27,7 @@ Status BlobLogSequentialReader::ReadSlice(uint64_t size, Slice* slice,
   assert(slice);
   assert(file_);
 
-  StopWatch read_sw(env_, statistics_, BLOB_DB_BLOB_FILE_READ_MICROS);
+  StopWatch read_sw(clock_, statistics_, BLOB_DB_BLOB_FILE_READ_MICROS);
   Status s = file_->Read(IOOptions(), next_byte_, static_cast<size_t>(size),
                          slice, buf, nullptr);
   next_byte_ += size;

db/blob/blob_log_sequential_reader.h

@@ -16,6 +16,7 @@ class RandomAccessFileReader;
 class Env;
 class Statistics;
 class Status;
+class SystemClock;
 
 /**
  * BlobLogSequentialReader is a general purpose log stream reader
@@ -35,7 +36,7 @@ class BlobLogSequentialReader {
 
   // Create a reader that will return log records from "*file_reader".
   BlobLogSequentialReader(std::unique_ptr<RandomAccessFileReader>&& file_reader,
-                          Env* env, Statistics* statistics);
+                          SystemClock* clock, Statistics* statistics);
 
   // No copying allowed
   BlobLogSequentialReader(const BlobLogSequentialReader&) = delete;
@@ -63,7 +64,8 @@ class BlobLogSequentialReader {
   Status ReadSlice(uint64_t size, Slice* slice, char* buf);
 
   const std::unique_ptr<RandomAccessFileReader> file_;
-  Env* env_;
+  SystemClock* clock_;
+
   Statistics* statistics_;
 
   Slice buffer_;

db/blob/blob_log_writer.cc

@@ -11,7 +11,7 @@
 #include "db/blob/blob_log_format.h"
 #include "file/writable_file_writer.h"
 #include "monitoring/statistics.h"
-#include "rocksdb/env.h"
+#include "rocksdb/system_clock.h"
 #include "test_util/sync_point.h"
 #include "util/coding.h"
 #include "util/stop_watch.h"
@@ -19,14 +19,16 @@
 namespace ROCKSDB_NAMESPACE {
 
 BlobLogWriter::BlobLogWriter(std::unique_ptr<WritableFileWriter>&& dest,
-                             Env* env, Statistics* statistics,
-                             uint64_t log_number, bool use_fs, uint64_t boffset)
+                             SystemClock* clock, Statistics* statistics,
+                             uint64_t log_number, bool use_fs, bool do_flush,
+                             uint64_t boffset)
     : dest_(std::move(dest)),
-      env_(env),
+      clock_(clock),
       statistics_(statistics),
       log_number_(log_number),
       block_offset_(boffset),
       use_fsync_(use_fs),
+      do_flush_(do_flush),
       last_elem_type_(kEtNone) {}
 
 BlobLogWriter::~BlobLogWriter() = default;
@@ -34,7 +36,7 @@ BlobLogWriter::~BlobLogWriter() = default;
 Status BlobLogWriter::Sync() {
   TEST_SYNC_POINT("BlobLogWriter::Sync");
 
-  StopWatch sync_sw(env_, statistics_, BLOB_DB_BLOB_FILE_SYNC_MICROS);
+  StopWatch sync_sw(clock_, statistics_, BLOB_DB_BLOB_FILE_SYNC_MICROS);
   Status s = dest_->Sync(use_fsync_);
   RecordTick(statistics_, BLOB_DB_BLOB_FILE_SYNCED);
   return s;
@@ -49,7 +51,9 @@ Status BlobLogWriter::WriteHeader(BlobLogHeader& header) {
   Status s = dest_->Append(Slice(str));
   if (s.ok()) {
     block_offset_ += str.size();
-    s = dest_->Flush();
+    if (do_flush_) {
+      s = dest_->Flush();
+    }
   }
   last_elem_type_ = kEtFileHdr;
   RecordTick(statistics_, BLOB_DB_BLOB_FILE_BYTES_WRITTEN,
@@ -144,7 +148,7 @@ Status BlobLogWriter::EmitPhysicalRecord(const std::string& headerbuf,
                                          const Slice& key, const Slice& val,
                                          uint64_t* key_offset,
                                          uint64_t* blob_offset) {
-  StopWatch write_sw(env_, statistics_, BLOB_DB_BLOB_FILE_WRITE_MICROS);
+  StopWatch write_sw(clock_, statistics_, BLOB_DB_BLOB_FILE_WRITE_MICROS);
   Status s = dest_->Append(Slice(headerbuf));
   if (s.ok()) {
     s = dest_->Append(key);
@@ -152,7 +156,7 @@ Status BlobLogWriter::EmitPhysicalRecord(const std::string& headerbuf,
   if (s.ok()) {
     s = dest_->Append(val);
   }
-  if (s.ok()) {
+  if (do_flush_ && s.ok()) {
     s = dest_->Flush();
   }
 

db/blob/blob_log_writer.h

@@ -9,7 +9,6 @@
 #include <string>
 
 #include "db/blob/blob_log_format.h"
-#include "rocksdb/env.h"
 #include "rocksdb/slice.h"
 #include "rocksdb/statistics.h"
 #include "rocksdb/status.h"
@@ -18,7 +17,7 @@
 namespace ROCKSDB_NAMESPACE {
 
 class WritableFileWriter;
-
+class SystemClock;
 /**
  * BlobLogWriter is the blob log stream writer. It provides an append-only
  * abstraction for writing blob data.
@@ -32,9 +31,9 @@ class BlobLogWriter {
   // Create a writer that will append data to "*dest".
   // "*dest" must be initially empty.
   // "*dest" must remain live while this BlobLogWriter is in use.
-  BlobLogWriter(std::unique_ptr<WritableFileWriter>&& dest, Env* env,
+  BlobLogWriter(std::unique_ptr<WritableFileWriter>&& dest, SystemClock* clock,
                 Statistics* statistics, uint64_t log_number, bool use_fsync,
-                uint64_t boffset = 0);
+                bool do_flush, uint64_t boffset = 0);
   // No copying allowed
   BlobLogWriter(const BlobLogWriter&) = delete;
   BlobLogWriter& operator=(const BlobLogWriter&) = delete;
@@ -69,11 +68,12 @@ class BlobLogWriter {
 
  private:
   std::unique_ptr<WritableFileWriter> dest_;
-  Env* env_;
+  SystemClock* clock_;
   Statistics* statistics_;
   uint64_t log_number_;
   uint64_t block_offset_;  // Current offset in block
   bool use_fsync_;
+  bool do_flush_;
 
  public:
   enum ElemType { kEtNone, kEtFileHdr, kEtRecord, kEtFileFooter };

db/blob/db_blob_basic_test.cc

@@ -0,0 +1,460 @@
+//  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 <array>
+
+#include "db/blob/blob_index.h"
+#include "db/db_test_util.h"
+#include "port/stack_trace.h"
+#include "test_util/sync_point.h"
+#include "utilities/fault_injection_env.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class DBBlobBasicTest : public DBTestBase {
+ protected:
+  DBBlobBasicTest()
+      : DBTestBase("/db_blob_basic_test", /* env_do_fsync */ false) {}
+};
+
+TEST_F(DBBlobBasicTest, GetBlob) {
+  Options options = GetDefaultOptions();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+
+  Reopen(options);
+
+  constexpr char key[] = "key";
+  constexpr char blob_value[] = "blob_value";
+
+  ASSERT_OK(Put(key, blob_value));
+
+  ASSERT_OK(Flush());
+
+  ASSERT_EQ(Get(key), blob_value);
+
+  // Try again with no I/O allowed. The table and the necessary blocks should
+  // already be in their respective caches; however, the blob itself can only be
+  // read from the blob file, so the read should return Incomplete.
+  ReadOptions read_options;
+  read_options.read_tier = kBlockCacheTier;
+
+  PinnableSlice result;
+  ASSERT_TRUE(db_->Get(read_options, db_->DefaultColumnFamily(), key, &result)
+                  .IsIncomplete());
+}
+
+TEST_F(DBBlobBasicTest, MultiGetBlobs) {
+  constexpr size_t min_blob_size = 6;
+
+  Options options = GetDefaultOptions();
+  options.enable_blob_files = true;
+  options.min_blob_size = min_blob_size;
+
+  Reopen(options);
+
+  // Put then retrieve three key-values. The first value is below the size limit
+  // and is thus stored inline; the other two are stored separately as blobs.
+  constexpr size_t num_keys = 3;
+
+  constexpr char first_key[] = "first_key";
+  constexpr char first_value[] = "short";
+  static_assert(sizeof(first_value) - 1 < min_blob_size,
+                "first_value too long to be inlined");
+
+  ASSERT_OK(Put(first_key, first_value));
+
+  constexpr char second_key[] = "second_key";
+  constexpr char second_value[] = "long_value";
+  static_assert(sizeof(second_value) - 1 >= min_blob_size,
+                "second_value too short to be stored as blob");
+
+  ASSERT_OK(Put(second_key, second_value));
+
+  constexpr char third_key[] = "third_key";
+  constexpr char third_value[] = "other_long_value";
+  static_assert(sizeof(third_value) - 1 >= min_blob_size,
+                "third_value too short to be stored as blob");
+
+  ASSERT_OK(Put(third_key, third_value));
+
+  ASSERT_OK(Flush());
+
+  ReadOptions read_options;
+
+  std::array<Slice, num_keys> keys{{first_key, second_key, third_key}};
+
+  {
+    std::array<PinnableSlice, num_keys> values;
+    std::array<Status, num_keys> statuses;
+
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys, &keys[0],
+                  &values[0], &statuses[0]);
+
+    ASSERT_OK(statuses[0]);
+    ASSERT_EQ(values[0], first_value);
+
+    ASSERT_OK(statuses[1]);
+    ASSERT_EQ(values[1], second_value);
+
+    ASSERT_OK(statuses[2]);
+    ASSERT_EQ(values[2], third_value);
+  }
+
+  // Try again with no I/O allowed. The table and the necessary blocks should
+  // already be in their respective caches. The first (inlined) value should be
+  // successfully read; however, the two blob values could only be read from the
+  // blob file, so for those the read should return Incomplete.
+  read_options.read_tier = kBlockCacheTier;
+
+  {
+    std::array<PinnableSlice, num_keys> values;
+    std::array<Status, num_keys> statuses;
+
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys, &keys[0],
+                  &values[0], &statuses[0]);
+
+    ASSERT_OK(statuses[0]);
+    ASSERT_EQ(values[0], first_value);
+
+    ASSERT_TRUE(statuses[1].IsIncomplete());
+
+    ASSERT_TRUE(statuses[2].IsIncomplete());
+  }
+}
+
+TEST_F(DBBlobBasicTest, GetBlob_CorruptIndex) {
+  Options options = GetDefaultOptions();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+
+  Reopen(options);
+
+  constexpr char key[] = "key";
+
+  // Fake a corrupt blob index.
+  const std::string blob_index("foobar");
+
+  WriteBatch batch;
+  ASSERT_OK(WriteBatchInternal::PutBlobIndex(&batch, 0, key, blob_index));
+  ASSERT_OK(db_->Write(WriteOptions(), &batch));
+
+  ASSERT_OK(Flush());
+
+  PinnableSlice result;
+  ASSERT_TRUE(db_->Get(ReadOptions(), db_->DefaultColumnFamily(), key, &result)
+                  .IsCorruption());
+}
+
+TEST_F(DBBlobBasicTest, GetBlob_InlinedTTLIndex) {
+  constexpr uint64_t min_blob_size = 10;
+
+  Options options = GetDefaultOptions();
+  options.enable_blob_files = true;
+  options.min_blob_size = min_blob_size;
+
+  Reopen(options);
+
+  constexpr char key[] = "key";
+  constexpr char blob[] = "short";
+  static_assert(sizeof(short) - 1 < min_blob_size,
+                "Blob too long to be inlined");
+
+  // Fake an inlined TTL blob index.
+  std::string blob_index;
+
+  constexpr uint64_t expiration = 1234567890;
+
+  BlobIndex::EncodeInlinedTTL(&blob_index, expiration, blob);
+
+  WriteBatch batch;
+  ASSERT_OK(WriteBatchInternal::PutBlobIndex(&batch, 0, key, blob_index));
+  ASSERT_OK(db_->Write(WriteOptions(), &batch));
+
+  ASSERT_OK(Flush());
+
+  PinnableSlice result;
+  ASSERT_TRUE(db_->Get(ReadOptions(), db_->DefaultColumnFamily(), key, &result)
+                  .IsCorruption());
+}
+
+TEST_F(DBBlobBasicTest, GetBlob_IndexWithInvalidFileNumber) {
+  Options options = GetDefaultOptions();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+
+  Reopen(options);
+
+  constexpr char key[] = "key";
+
+  // Fake a blob index referencing a non-existent blob file.
+  std::string blob_index;
+
+  constexpr uint64_t blob_file_number = 1000;
+  constexpr uint64_t offset = 1234;
+  constexpr uint64_t size = 5678;
+
+  BlobIndex::EncodeBlob(&blob_index, blob_file_number, offset, size,
+                        kNoCompression);
+
+  WriteBatch batch;
+  ASSERT_OK(WriteBatchInternal::PutBlobIndex(&batch, 0, key, blob_index));
+  ASSERT_OK(db_->Write(WriteOptions(), &batch));
+
+  ASSERT_OK(Flush());
+
+  PinnableSlice result;
+  ASSERT_TRUE(db_->Get(ReadOptions(), db_->DefaultColumnFamily(), key, &result)
+                  .IsCorruption());
+}
+
+#ifndef ROCKSDB_LITE
+TEST_F(DBBlobBasicTest, GenerateIOTracing) {
+  Options options = GetDefaultOptions();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+  std::string trace_file = dbname_ + "/io_trace_file";
+
+  Reopen(options);
+  {
+    // Create IO trace file
+    std::unique_ptr<TraceWriter> trace_writer;
+    ASSERT_OK(
+        NewFileTraceWriter(env_, EnvOptions(), trace_file, &trace_writer));
+    ASSERT_OK(db_->StartIOTrace(TraceOptions(), std::move(trace_writer)));
+
+    constexpr char key[] = "key";
+    constexpr char blob_value[] = "blob_value";
+
+    ASSERT_OK(Put(key, blob_value));
+    ASSERT_OK(Flush());
+    ASSERT_EQ(Get(key), blob_value);
+
+    ASSERT_OK(db_->EndIOTrace());
+    ASSERT_OK(env_->FileExists(trace_file));
+  }
+  {
+    // Parse trace file to check file operations related to blob files are
+    // recorded.
+    std::unique_ptr<TraceReader> trace_reader;
+    ASSERT_OK(
+        NewFileTraceReader(env_, EnvOptions(), trace_file, &trace_reader));
+    IOTraceReader reader(std::move(trace_reader));
+
+    IOTraceHeader header;
+    ASSERT_OK(reader.ReadHeader(&header));
+    ASSERT_EQ(kMajorVersion, static_cast<int>(header.rocksdb_major_version));
+    ASSERT_EQ(kMinorVersion, static_cast<int>(header.rocksdb_minor_version));
+
+    // Read records.
+    int blob_files_op_count = 0;
+    Status status;
+    while (true) {
+      IOTraceRecord record;
+      status = reader.ReadIOOp(&record);
+      if (!status.ok()) {
+        break;
+      }
+      if (record.file_name.find("blob") != std::string::npos) {
+        blob_files_op_count++;
+      }
+    }
+    // Assuming blob files will have Append, Close and then Read operations.
+    ASSERT_GT(blob_files_op_count, 2);
+  }
+}
+#endif  // !ROCKSDB_LITE
+
+TEST_F(DBBlobBasicTest, BestEffortsRecovery_MissingNewestBlobFile) {
+  Options options = GetDefaultOptions();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+  options.create_if_missing = true;
+  Reopen(options);
+
+  ASSERT_OK(dbfull()->DisableFileDeletions());
+  constexpr int kNumTableFiles = 2;
+  for (int i = 0; i < kNumTableFiles; ++i) {
+    for (char ch = 'a'; ch != 'c'; ++ch) {
+      std::string key(1, ch);
+      ASSERT_OK(Put(key, "value" + std::to_string(i)));
+    }
+    ASSERT_OK(Flush());
+  }
+
+  Close();
+
+  std::vector<std::string> files;
+  ASSERT_OK(env_->GetChildren(dbname_, &files));
+  std::string blob_file_path;
+  uint64_t max_blob_file_num = kInvalidBlobFileNumber;
+  for (const auto& fname : files) {
+    uint64_t file_num = 0;
+    FileType type;
+    if (ParseFileName(fname, &file_num, /*info_log_name_prefix=*/"", &type) &&
+        type == kBlobFile) {
+      if (file_num > max_blob_file_num) {
+        max_blob_file_num = file_num;
+        blob_file_path = dbname_ + "/" + fname;
+      }
+    }
+  }
+  ASSERT_OK(env_->DeleteFile(blob_file_path));
+
+  options.best_efforts_recovery = true;
+  Reopen(options);
+  std::string value;
+  ASSERT_OK(db_->Get(ReadOptions(), "a", &value));
+  ASSERT_EQ("value" + std::to_string(kNumTableFiles - 2), value);
+}
+
+class DBBlobBasicIOErrorTest : public DBBlobBasicTest,
+                               public testing::WithParamInterface<std::string> {
+ protected:
+  DBBlobBasicIOErrorTest() : sync_point_(GetParam()) {
+    fault_injection_env_.reset(new FaultInjectionTestEnv(env_));
+  }
+  ~DBBlobBasicIOErrorTest() { Close(); }
+
+  std::unique_ptr<FaultInjectionTestEnv> fault_injection_env_;
+  std::string sync_point_;
+};
+
+INSTANTIATE_TEST_CASE_P(DBBlobBasicTest, DBBlobBasicIOErrorTest,
+                        ::testing::ValuesIn(std::vector<std::string>{
+                            "BlobFileReader::OpenFile:NewRandomAccessFile",
+                            "BlobFileReader::GetBlob:ReadFromFile"}));
+
+TEST_P(DBBlobBasicIOErrorTest, GetBlob_IOError) {
+  Options options;
+  options.env = fault_injection_env_.get();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+
+  Reopen(options);
+
+  constexpr char key[] = "key";
+  constexpr char blob_value[] = "blob_value";
+
+  ASSERT_OK(Put(key, blob_value));
+
+  ASSERT_OK(Flush());
+
+  SyncPoint::GetInstance()->SetCallBack(sync_point_, [this](void* /* arg */) {
+    fault_injection_env_->SetFilesystemActive(false,
+                                              Status::IOError(sync_point_));
+  });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  PinnableSlice result;
+  ASSERT_TRUE(db_->Get(ReadOptions(), db_->DefaultColumnFamily(), key, &result)
+                  .IsIOError());
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+TEST_P(DBBlobBasicIOErrorTest, MultiGetBlobs_IOError) {
+  Options options = GetDefaultOptions();
+  options.env = fault_injection_env_.get();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+
+  Reopen(options);
+
+  constexpr size_t num_keys = 2;
+
+  constexpr char first_key[] = "first_key";
+  constexpr char first_value[] = "first_value";
+
+  ASSERT_OK(Put(first_key, first_value));
+
+  constexpr char second_key[] = "second_key";
+  constexpr char second_value[] = "second_value";
+
+  ASSERT_OK(Put(second_key, second_value));
+
+  ASSERT_OK(Flush());
+
+  std::array<Slice, num_keys> keys{{first_key, second_key}};
+  std::array<PinnableSlice, num_keys> values;
+  std::array<Status, num_keys> statuses;
+
+  SyncPoint::GetInstance()->SetCallBack(sync_point_, [this](void* /* arg */) {
+    fault_injection_env_->SetFilesystemActive(false,
+                                              Status::IOError(sync_point_));
+  });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  db_->MultiGet(ReadOptions(), db_->DefaultColumnFamily(), num_keys, &keys[0],
+                &values[0], &statuses[0]);
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+
+  ASSERT_TRUE(statuses[0].IsIOError());
+  ASSERT_TRUE(statuses[1].IsIOError());
+}
+
+namespace {
+
+class ReadBlobCompactionFilter : public CompactionFilter {
+ public:
+  ReadBlobCompactionFilter() = default;
+  const char* Name() const override {
+    return "rocksdb.compaction.filter.read.blob";
+  }
+  CompactionFilter::Decision FilterV2(
+      int /*level*/, const Slice& /*key*/, ValueType value_type,
+      const Slice& existing_value, std::string* new_value,
+      std::string* /*skip_until*/) const override {
+    if (value_type != CompactionFilter::ValueType::kValue) {
+      return CompactionFilter::Decision::kKeep;
+    }
+    assert(new_value);
+    new_value->assign(existing_value.data(), existing_value.size());
+    return CompactionFilter::Decision::kChangeValue;
+  }
+};
+
+}  // anonymous namespace
+
+TEST_P(DBBlobBasicIOErrorTest, CompactionFilterReadBlob_IOError) {
+  Options options = GetDefaultOptions();
+  options.env = fault_injection_env_.get();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+  options.create_if_missing = true;
+  std::unique_ptr<CompactionFilter> compaction_filter_guard(
+      new ReadBlobCompactionFilter);
+  options.compaction_filter = compaction_filter_guard.get();
+
+  DestroyAndReopen(options);
+  constexpr char key[] = "foo";
+  constexpr char blob_value[] = "foo_blob_value";
+  ASSERT_OK(Put(key, blob_value));
+  ASSERT_OK(Flush());
+
+  SyncPoint::GetInstance()->SetCallBack(sync_point_, [this](void* /* arg */) {
+    fault_injection_env_->SetFilesystemActive(false,
+                                              Status::IOError(sync_point_));
+  });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  ASSERT_TRUE(db_->CompactRange(CompactRangeOptions(), /*begin=*/nullptr,
+                                /*end=*/nullptr)
+                  .IsIOError());
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+}  // namespace ROCKSDB_NAMESPACE
+
+int main(int argc, char** argv) {
+  ROCKSDB_NAMESPACE::port::InstallStackTraceHandler();
+  ::testing::InitGoogleTest(&argc, argv);
+  return RUN_ALL_TESTS();
+}

db/blob/db_blob_compaction_test.cc

@@ -0,0 +1,399 @@
+//  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/blob/blob_index.h"
+#include "db/db_test_util.h"
+#include "port/stack_trace.h"
+#include "test_util/sync_point.h"
+#include "utilities/fault_injection_env.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class DBBlobCompactionTest : public DBTestBase {
+ public:
+  explicit DBBlobCompactionTest()
+      : DBTestBase("/db_blob_compaction_test", /*env_do_fsync=*/false) {}
+
+#ifndef ROCKSDB_LITE
+  const std::vector<InternalStats::CompactionStats>& GetCompactionStats() {
+    VersionSet* const versions = dbfull()->TEST_GetVersionSet();
+    assert(versions);
+    assert(versions->GetColumnFamilySet());
+
+    ColumnFamilyData* const cfd = versions->GetColumnFamilySet()->GetDefault();
+    assert(cfd);
+
+    const InternalStats* const internal_stats = cfd->internal_stats();
+    assert(internal_stats);
+
+    return internal_stats->TEST_GetCompactionStats();
+  }
+#endif  // ROCKSDB_LITE
+};
+
+namespace {
+
+class FilterByKeyLength : public CompactionFilter {
+ public:
+  explicit FilterByKeyLength(size_t len) : length_threshold_(len) {}
+  const char* Name() const override {
+    return "rocksdb.compaction.filter.by.key.length";
+  }
+  CompactionFilter::Decision FilterBlobByKey(
+      int /*level*/, const Slice& key, std::string* /*new_value*/,
+      std::string* /*skip_until*/) const override {
+    if (key.size() < length_threshold_) {
+      return CompactionFilter::Decision::kRemove;
+    }
+    return CompactionFilter::Decision::kKeep;
+  }
+
+ private:
+  size_t length_threshold_;
+};
+
+class BadBlobCompactionFilter : public CompactionFilter {
+ public:
+  explicit BadBlobCompactionFilter(std::string prefix,
+                                   CompactionFilter::Decision filter_by_key,
+                                   CompactionFilter::Decision filter_v2)
+      : prefix_(std::move(prefix)),
+        filter_blob_by_key_(filter_by_key),
+        filter_v2_(filter_v2) {}
+  const char* Name() const override { return "rocksdb.compaction.filter.bad"; }
+  CompactionFilter::Decision FilterBlobByKey(
+      int /*level*/, const Slice& key, std::string* /*new_value*/,
+      std::string* /*skip_until*/) const override {
+    if (key.size() >= prefix_.size() &&
+        0 == strncmp(prefix_.data(), key.data(), prefix_.size())) {
+      return CompactionFilter::Decision::kUndetermined;
+    }
+    return filter_blob_by_key_;
+  }
+  CompactionFilter::Decision FilterV2(
+      int /*level*/, const Slice& /*key*/, ValueType /*value_type*/,
+      const Slice& /*existing_value*/, std::string* /*new_value*/,
+      std::string* /*skip_until*/) const override {
+    return filter_v2_;
+  }
+
+ private:
+  const std::string prefix_;
+  const CompactionFilter::Decision filter_blob_by_key_;
+  const CompactionFilter::Decision filter_v2_;
+};
+
+class ValueBlindWriteFilter : public CompactionFilter {
+ public:
+  explicit ValueBlindWriteFilter(std::string new_val)
+      : new_value_(std::move(new_val)) {}
+  const char* Name() const override {
+    return "rocksdb.compaction.filter.blind.write";
+  }
+  CompactionFilter::Decision FilterBlobByKey(
+      int level, const Slice& key, std::string* new_value,
+      std::string* skip_until) const override;
+
+ private:
+  const std::string new_value_;
+};
+
+CompactionFilter::Decision ValueBlindWriteFilter::FilterBlobByKey(
+    int /*level*/, const Slice& /*key*/, std::string* new_value,
+    std::string* /*skip_until*/) const {
+  assert(new_value);
+  new_value->assign(new_value_);
+  return CompactionFilter::Decision::kChangeValue;
+}
+
+class ValueMutationFilter : public CompactionFilter {
+ public:
+  explicit ValueMutationFilter(std::string padding)
+      : padding_(std::move(padding)) {}
+  const char* Name() const override {
+    return "rocksdb.compaction.filter.value.mutation";
+  }
+  CompactionFilter::Decision FilterV2(int level, const Slice& key,
+                                      ValueType value_type,
+                                      const Slice& existing_value,
+                                      std::string* new_value,
+                                      std::string* skip_until) const override;
+
+ private:
+  const std::string padding_;
+};
+
+CompactionFilter::Decision ValueMutationFilter::FilterV2(
+    int /*level*/, const Slice& /*key*/, ValueType value_type,
+    const Slice& existing_value, std::string* new_value,
+    std::string* /*skip_until*/) const {
+  assert(CompactionFilter::ValueType::kBlobIndex != value_type);
+  if (CompactionFilter::ValueType::kValue != value_type) {
+    return CompactionFilter::Decision::kKeep;
+  }
+  assert(new_value);
+  new_value->assign(existing_value.data(), existing_value.size());
+  new_value->append(padding_);
+  return CompactionFilter::Decision::kChangeValue;
+}
+
+class AlwaysKeepFilter : public CompactionFilter {
+ public:
+  explicit AlwaysKeepFilter() = default;
+  const char* Name() const override {
+    return "rocksdb.compaction.filter.always.keep";
+  }
+  CompactionFilter::Decision FilterV2(
+      int /*level*/, const Slice& /*key*/, ValueType /*value_type*/,
+      const Slice& /*existing_value*/, std::string* /*new_value*/,
+      std::string* /*skip_until*/) const override {
+    return CompactionFilter::Decision::kKeep;
+  }
+};
+}  // anonymous namespace
+
+class DBBlobBadCompactionFilterTest
+    : public DBBlobCompactionTest,
+      public testing::WithParamInterface<
+          std::tuple<std::string, CompactionFilter::Decision,
+                     CompactionFilter::Decision>> {
+ public:
+  explicit DBBlobBadCompactionFilterTest()
+      : compaction_filter_guard_(new BadBlobCompactionFilter(
+            std::get<0>(GetParam()), std::get<1>(GetParam()),
+            std::get<2>(GetParam()))) {}
+
+ protected:
+  std::unique_ptr<CompactionFilter> compaction_filter_guard_;
+};
+
+INSTANTIATE_TEST_CASE_P(
+    BadCompactionFilter, DBBlobBadCompactionFilterTest,
+    testing::Combine(
+        testing::Values("a"),
+        testing::Values(CompactionFilter::Decision::kChangeBlobIndex,
+                        CompactionFilter::Decision::kIOError),
+        testing::Values(CompactionFilter::Decision::kUndetermined,
+                        CompactionFilter::Decision::kChangeBlobIndex,
+                        CompactionFilter::Decision::kIOError)));
+
+TEST_F(DBBlobCompactionTest, FilterByKeyLength) {
+  Options options = GetDefaultOptions();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+  options.create_if_missing = true;
+  constexpr size_t kKeyLength = 2;
+  std::unique_ptr<CompactionFilter> compaction_filter_guard(
+      new FilterByKeyLength(kKeyLength));
+  options.compaction_filter = compaction_filter_guard.get();
+
+  constexpr char short_key[] = "a";
+  constexpr char long_key[] = "abc";
+  constexpr char blob_value[] = "value";
+
+  DestroyAndReopen(options);
+  ASSERT_OK(Put(short_key, blob_value));
+  ASSERT_OK(Put(long_key, blob_value));
+  ASSERT_OK(Flush());
+  CompactRangeOptions cro;
+  ASSERT_OK(db_->CompactRange(cro, /*begin=*/nullptr, /*end=*/nullptr));
+  std::string value;
+  ASSERT_TRUE(db_->Get(ReadOptions(), short_key, &value).IsNotFound());
+  value.clear();
+  ASSERT_OK(db_->Get(ReadOptions(), long_key, &value));
+  ASSERT_EQ("value", value);
+
+#ifndef ROCKSDB_LITE
+  const auto& compaction_stats = GetCompactionStats();
+  ASSERT_GE(compaction_stats.size(), 2);
+
+  // Filter decides between kKeep and kRemove solely based on key;
+  // this involves neither reading nor writing blobs
+  ASSERT_EQ(compaction_stats[1].bytes_read_blob, 0);
+  ASSERT_EQ(compaction_stats[1].bytes_written_blob, 0);
+#endif  // ROCKSDB_LITE
+
+  Close();
+}
+
+TEST_F(DBBlobCompactionTest, BlindWriteFilter) {
+  Options options = GetDefaultOptions();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+  options.create_if_missing = true;
+  constexpr char new_blob_value[] = "new_blob_value";
+  std::unique_ptr<CompactionFilter> compaction_filter_guard(
+      new ValueBlindWriteFilter(new_blob_value));
+  options.compaction_filter = compaction_filter_guard.get();
+  DestroyAndReopen(options);
+  const std::vector<std::string> keys = {"a", "b", "c"};
+  const std::vector<std::string> values = {"a_value", "b_value", "c_value"};
+  assert(keys.size() == values.size());
+  for (size_t i = 0; i < keys.size(); ++i) {
+    ASSERT_OK(Put(keys[i], values[i]));
+  }
+  ASSERT_OK(Flush());
+  ASSERT_OK(db_->CompactRange(CompactRangeOptions(), /*begin=*/nullptr,
+                              /*end=*/nullptr));
+  for (const auto& key : keys) {
+    ASSERT_EQ(new_blob_value, Get(key));
+  }
+
+#ifndef ROCKSDB_LITE
+  const auto& compaction_stats = GetCompactionStats();
+  ASSERT_GE(compaction_stats.size(), 2);
+
+  // Filter unconditionally changes value in FilterBlobByKey;
+  // this involves writing but not reading blobs
+  ASSERT_EQ(compaction_stats[1].bytes_read_blob, 0);
+  ASSERT_GT(compaction_stats[1].bytes_written_blob, 0);
+#endif  // ROCKSDB_LITE
+
+  Close();
+}
+
+TEST_P(DBBlobBadCompactionFilterTest, BadDecisionFromCompactionFilter) {
+  Options options = GetDefaultOptions();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+  options.create_if_missing = true;
+  options.compaction_filter = compaction_filter_guard_.get();
+  DestroyAndReopen(options);
+  ASSERT_OK(Put("b", "value"));
+  ASSERT_OK(Flush());
+  ASSERT_TRUE(db_->CompactRange(CompactRangeOptions(), /*begin=*/nullptr,
+                                /*end=*/nullptr)
+                  .IsNotSupported());
+  Close();
+
+  DestroyAndReopen(options);
+  std::string key(std::get<0>(GetParam()));
+  ASSERT_OK(Put(key, "value"));
+  ASSERT_OK(Flush());
+  ASSERT_TRUE(db_->CompactRange(CompactRangeOptions(), /*begin=*/nullptr,
+                                /*end=*/nullptr)
+                  .IsNotSupported());
+  Close();
+}
+
+TEST_F(DBBlobCompactionTest, CompactionFilter_InlinedTTLIndex) {
+  Options options = GetDefaultOptions();
+  options.create_if_missing = true;
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+  std::unique_ptr<CompactionFilter> compaction_filter_guard(
+      new ValueMutationFilter(""));
+  options.compaction_filter = compaction_filter_guard.get();
+  DestroyAndReopen(options);
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+  // Fake an inlined TTL blob index.
+  std::string blob_index;
+  constexpr uint64_t expiration = 1234567890;
+  BlobIndex::EncodeInlinedTTL(&blob_index, expiration, blob);
+  WriteBatch batch;
+  ASSERT_OK(WriteBatchInternal::PutBlobIndex(&batch, 0, key, blob_index));
+  ASSERT_OK(db_->Write(WriteOptions(), &batch));
+  ASSERT_OK(Flush());
+  ASSERT_TRUE(db_->CompactRange(CompactRangeOptions(), /*begin=*/nullptr,
+                                /*end=*/nullptr)
+                  .IsCorruption());
+  Close();
+}
+
+TEST_F(DBBlobCompactionTest, CompactionFilter) {
+  Options options = GetDefaultOptions();
+  options.create_if_missing = true;
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+  constexpr char padding[] = "_delta";
+  std::unique_ptr<CompactionFilter> compaction_filter_guard(
+      new ValueMutationFilter(padding));
+  options.compaction_filter = compaction_filter_guard.get();
+  DestroyAndReopen(options);
+  const std::vector<std::pair<std::string, std::string>> kvs = {
+      {"a", "a_value"}, {"b", "b_value"}, {"c", "c_value"}};
+  for (const auto& kv : kvs) {
+    ASSERT_OK(Put(kv.first, kv.second));
+  }
+  ASSERT_OK(Flush());
+  ASSERT_OK(db_->CompactRange(CompactRangeOptions(), /*begin=*/nullptr,
+                              /*end=*/nullptr));
+  for (const auto& kv : kvs) {
+    ASSERT_EQ(kv.second + std::string(padding), Get(kv.first));
+  }
+
+#ifndef ROCKSDB_LITE
+  const auto& compaction_stats = GetCompactionStats();
+  ASSERT_GE(compaction_stats.size(), 2);
+
+  // Filter changes the value using the previous value in FilterV2;
+  // this involves reading and writing blobs
+  ASSERT_GT(compaction_stats[1].bytes_read_blob, 0);
+  ASSERT_GT(compaction_stats[1].bytes_written_blob, 0);
+#endif  // ROCKSDB_LITE
+
+  Close();
+}
+
+TEST_F(DBBlobCompactionTest, CorruptedBlobIndex) {
+  Options options = GetDefaultOptions();
+  options.create_if_missing = true;
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+  std::unique_ptr<CompactionFilter> compaction_filter_guard(
+      new ValueMutationFilter(""));
+  options.compaction_filter = compaction_filter_guard.get();
+  DestroyAndReopen(options);
+  // Mock a corrupted blob index
+  constexpr char key[] = "key";
+  std::string blob_idx("blob_idx");
+  WriteBatch write_batch;
+  ASSERT_OK(WriteBatchInternal::PutBlobIndex(&write_batch, 0, key, blob_idx));
+  ASSERT_OK(db_->Write(WriteOptions(), &write_batch));
+  ASSERT_OK(Flush());
+  ASSERT_TRUE(db_->CompactRange(CompactRangeOptions(), /*begin=*/nullptr,
+                                /*end=*/nullptr)
+                  .IsCorruption());
+  Close();
+}
+
+TEST_F(DBBlobCompactionTest, CompactionFilterReadBlobAndKeep) {
+  Options options = GetDefaultOptions();
+  options.create_if_missing = true;
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+  std::unique_ptr<CompactionFilter> compaction_filter_guard(
+      new AlwaysKeepFilter());
+  options.compaction_filter = compaction_filter_guard.get();
+  DestroyAndReopen(options);
+  ASSERT_OK(Put("foo", "foo_value"));
+  ASSERT_OK(Flush());
+  std::vector<uint64_t> blob_files = GetBlobFileNumbers();
+  ASSERT_EQ(1, blob_files.size());
+  ASSERT_OK(db_->CompactRange(CompactRangeOptions(), /*begin=*/nullptr,
+                              /*end=*/nullptr));
+  ASSERT_EQ(blob_files, GetBlobFileNumbers());
+
+#ifndef ROCKSDB_LITE
+  const auto& compaction_stats = GetCompactionStats();
+  ASSERT_GE(compaction_stats.size(), 2);
+
+  // Filter decides to keep the existing value in FilterV2;
+  // this involves reading but not writing blobs
+  ASSERT_GT(compaction_stats[1].bytes_read_blob, 0);
+  ASSERT_EQ(compaction_stats[1].bytes_written_blob, 0);
+#endif  // ROCKSDB_LITE
+
+  Close();
+}
+
+}  // namespace ROCKSDB_NAMESPACE
+
+int main(int argc, char** argv) {
+  ROCKSDB_NAMESPACE::port::InstallStackTraceHandler();
+  ::testing::InitGoogleTest(&argc, argv);
+  return RUN_ALL_TESTS();
+}

db/blob/db_blob_corruption_test.cc

@@ -0,0 +1,81 @@
+//  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_test_util.h"
+#include "port/stack_trace.h"
+#include "test_util/sync_point.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class DBBlobCorruptionTest : public DBTestBase {
+ protected:
+  DBBlobCorruptionTest()
+      : DBTestBase("/db_blob_corruption_test", /* env_do_fsync */ false) {}
+
+  void Corrupt(FileType filetype, int offset, int bytes_to_corrupt) {
+    // Pick file to corrupt
+    std::vector<std::string> filenames;
+    ASSERT_OK(env_->GetChildren(dbname_, &filenames));
+    uint64_t number;
+    FileType type;
+    std::string fname;
+    uint64_t picked_number = kInvalidBlobFileNumber;
+    for (size_t i = 0; i < filenames.size(); i++) {
+      if (ParseFileName(filenames[i], &number, &type) && type == filetype &&
+          number > picked_number) {  // Pick latest file
+        fname = dbname_ + "/" + filenames[i];
+        picked_number = number;
+      }
+    }
+    ASSERT_TRUE(!fname.empty()) << filetype;
+    ASSERT_OK(test::CorruptFile(env_, fname, offset, bytes_to_corrupt));
+  }
+};
+
+#ifndef ROCKSDB_LITE
+TEST_F(DBBlobCorruptionTest, VerifyWholeBlobFileChecksum) {
+  Options options = GetDefaultOptions();
+  options.enable_blob_files = true;
+  options.min_blob_size = 0;
+  options.create_if_missing = true;
+  options.file_checksum_gen_factory =
+      ROCKSDB_NAMESPACE::GetFileChecksumGenCrc32cFactory();
+  Reopen(options);
+
+  ASSERT_OK(Put(Slice("key_1"), Slice("blob_value_1")));
+  ASSERT_OK(Flush());
+  ASSERT_OK(Put(Slice("key_2"), Slice("blob_value_2")));
+  ASSERT_OK(Flush());
+  ASSERT_OK(db_->VerifyFileChecksums(ReadOptions()));
+  Close();
+
+  Corrupt(kBlobFile, 0, 2);
+
+  ASSERT_OK(TryReopen(options));
+
+  int count{0};
+  SyncPoint::GetInstance()->SetCallBack(
+      "DBImpl::VerifyFullFileChecksum:mismatch", [&](void* arg) {
+        const Status* s = static_cast<Status*>(arg);
+        ASSERT_NE(s, nullptr);
+        ++count;
+        ASSERT_NOK(*s);
+      });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  ASSERT_TRUE(db_->VerifyFileChecksums(ReadOptions()).IsCorruption());
+  ASSERT_EQ(1, count);
+
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+#endif  // !ROCKSDB_LITE
+}  // namespace ROCKSDB_NAMESPACE
+
+int main(int argc, char** argv) {
+  ROCKSDB_NAMESPACE::port::InstallStackTraceHandler();
+  ::testing::InitGoogleTest(&argc, argv);
+  return RUN_ALL_TESTS();
+}

db/blob/db_blob_index_test.cc

@@ -73,6 +73,9 @@ class DBBlobIndexTest : public DBTestBase {
     if (s.IsNotFound()) {
       return "NOT_FOUND";
     }
+    if (s.IsCorruption()) {
+      return "CORRUPTION";
+    }
     if (s.IsNotSupported()) {
       return "NOT_SUPPORTED";
     }
@@ -95,11 +98,12 @@ class DBBlobIndexTest : public DBTestBase {
   ArenaWrappedDBIter* GetBlobIterator() {
     return dbfull()->NewIteratorImpl(
         ReadOptions(), cfd(), dbfull()->GetLatestSequenceNumber(),
-        nullptr /*read_callback*/, true /*allow_blob*/);
+        nullptr /*read_callback*/, true /*expose_blob_index*/);
   }
 
   Options GetTestOptions() {
     Options options;
+    options.env = CurrentOptions().env;
     options.create_if_missing = true;
     options.num_levels = 2;
     options.disable_auto_compactions = true;
@@ -153,8 +157,13 @@ TEST_F(DBBlobIndexTest, Write) {
   }
 }
 
-// Get should be able to return blob index if is_blob_index is provided,
-// otherwise return Status::NotSupported status.
+// Note: the following test case pertains to the StackableDB-based BlobDB
+// implementation. Get should be able to return blob index if is_blob_index is
+// provided, otherwise it should return Status::NotSupported (when reading from
+// memtable) or Status::Corruption (when reading from SST). Reading from SST
+// returns Corruption because we can't differentiate between the application
+// accidentally opening the base DB of a stacked BlobDB and actual corruption
+// when using the integrated BlobDB.
 TEST_F(DBBlobIndexTest, Get) {
   for (auto tier : kAllTiers) {
     DestroyAndReopen(GetTestOptions());
@@ -171,15 +180,22 @@ TEST_F(DBBlobIndexTest, Get) {
     ASSERT_EQ("value", GetImpl("key", &is_blob_index));
     ASSERT_FALSE(is_blob_index);
     // Verify blob index
-    ASSERT_TRUE(Get("blob_key", &value).IsNotSupported());
-    ASSERT_EQ("NOT_SUPPORTED", GetImpl("blob_key"));
+    if (tier <= kImmutableMemtables) {
+      ASSERT_TRUE(Get("blob_key", &value).IsNotSupported());
+      ASSERT_EQ("NOT_SUPPORTED", GetImpl("blob_key"));
+    } else {
+      ASSERT_TRUE(Get("blob_key", &value).IsCorruption());
+      ASSERT_EQ("CORRUPTION", GetImpl("blob_key"));
+    }
     ASSERT_EQ("blob_index", GetImpl("blob_key", &is_blob_index));
     ASSERT_TRUE(is_blob_index);
   }
 }
 
-// Get should NOT return Status::NotSupported if blob index is updated with
-// a normal value.
+// Note: the following test case pertains to the StackableDB-based BlobDB
+// implementation. Get should NOT return Status::NotSupported/Status::Corruption
+// if blob index is updated with a normal value. See the test case above for
+// more details.
 TEST_F(DBBlobIndexTest, Updated) {
   for (auto tier : kAllTiers) {
     DestroyAndReopen(GetTestOptions());
@@ -206,7 +222,11 @@ TEST_F(DBBlobIndexTest, Updated) {
       ASSERT_EQ("blob_index", GetBlobIndex("key" + ToString(i), snapshot));
     }
     ASSERT_EQ("new_value", Get("key1"));
-    ASSERT_EQ("NOT_SUPPORTED", GetImpl("key2"));
+    if (tier <= kImmutableMemtables) {
+      ASSERT_EQ("NOT_SUPPORTED", GetImpl("key2"));
+    } else {
+      ASSERT_EQ("CORRUPTION", GetImpl("key2"));
+    }
     ASSERT_EQ("NOT_FOUND", Get("key3"));
     ASSERT_EQ("NOT_FOUND", Get("key4"));
     ASSERT_EQ("a,b,c", GetImpl("key5"));
@@ -218,8 +238,11 @@ TEST_F(DBBlobIndexTest, Updated) {
   }
 }
 
-// Iterator should get blob value if allow_blob flag is set,
-// otherwise return Status::NotSupported status.
+// Note: the following test case pertains to the StackableDB-based BlobDB
+// implementation. When a blob iterator is used, it should set the
+// expose_blob_index flag for the underlying DBIter, and retrieve/return the
+// corresponding blob value. If a regular DBIter is created (i.e.
+// expose_blob_index is not set), it should return Status::Corruption.
 TEST_F(DBBlobIndexTest, Iterate) {
   const std::vector<std::vector<ValueType>> data = {
       /*00*/ {kTypeValue},
@@ -282,6 +305,7 @@ TEST_F(DBBlobIndexTest, Iterate) {
                     std::function<void(Iterator*)> extra_check = nullptr) {
     // Seek
     auto* iterator = create_iterator();
+    ASSERT_OK(iterator->status());
     ASSERT_OK(iterator->Refresh());
     iterator->Seek(get_key(index));
     check_iterator(iterator, expected_status, forward_value);
@@ -295,6 +319,7 @@ TEST_F(DBBlobIndexTest, Iterate) {
     ASSERT_OK(iterator->Refresh());
     iterator->Seek(get_key(index - 1));
     ASSERT_TRUE(iterator->Valid());
+    ASSERT_OK(iterator->status());
     iterator->Next();
     check_iterator(iterator, expected_status, forward_value);
     if (extra_check) {
@@ -304,6 +329,7 @@ TEST_F(DBBlobIndexTest, Iterate) {
 
     // SeekForPrev
     iterator = create_iterator();
+    ASSERT_OK(iterator->status());
     ASSERT_OK(iterator->Refresh());
     iterator->SeekForPrev(get_key(index));
     check_iterator(iterator, expected_status, backward_value);
@@ -316,6 +342,7 @@ TEST_F(DBBlobIndexTest, Iterate) {
     iterator = create_iterator();
     iterator->Seek(get_key(index + 1));
     ASSERT_TRUE(iterator->Valid());
+    ASSERT_OK(iterator->status());
     iterator->Prev();
     check_iterator(iterator, expected_status, backward_value);
     if (extra_check) {
@@ -353,7 +380,7 @@ TEST_F(DBBlobIndexTest, Iterate) {
             ASSERT_OK(Write(&batch));
             break;
           default:
-            assert(false);
+            FAIL();
         };
       }
       snapshots.push_back(dbfull()->GetSnapshot());
@@ -364,8 +391,8 @@ TEST_F(DBBlobIndexTest, Iterate) {
     MoveDataTo(tier);
 
     // Normal iterator
-    verify(1, Status::kNotSupported, "", "", create_normal_iterator);
-    verify(3, Status::kNotSupported, "", "", create_normal_iterator);
+    verify(1, Status::kCorruption, "", "", create_normal_iterator);
+    verify(3, Status::kCorruption, "", "", create_normal_iterator);
     verify(5, Status::kOk, get_value(5, 0), get_value(5, 0),
            create_normal_iterator);
     verify(7, Status::kOk, get_value(8, 0), get_value(6, 0),

db/builder.cc

@@ -23,11 +23,13 @@
 #include "db/range_del_aggregator.h"
 #include "db/table_cache.h"
 #include "db/version_edit.h"
+#include "file/file_util.h"
 #include "file/filename.h"
 #include "file/read_write_util.h"
 #include "file/writable_file_writer.h"
 #include "monitoring/iostats_context_imp.h"
 #include "monitoring/thread_status_util.h"
+#include "options/options_helper.h"
 #include "rocksdb/db.h"
 #include "rocksdb/env.h"
 #include "rocksdb/iterator.h"
@@ -50,8 +52,8 @@ 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 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 std::string& db_id,
     const std::string& db_session_id) {
@@ -61,16 +63,16 @@ TableBuilder* NewTableBuilder(
   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, db_id, db_session_id),
+                          compression_opts, skip_filters, column_family_name,
+                          level, creation_time, oldest_key_time,
+                          target_file_size, file_creation_time, db_id,
+                          db_session_id),
       column_family_id, file);
 }
 
 Status BuildTable(
-    const std::string& dbname, VersionSet* versions, Env* env, FileSystem* fs,
-    const ImmutableCFOptions& ioptions,
+    const std::string& dbname, VersionSet* versions,
+    const ImmutableDBOptions& db_options, const ImmutableCFOptions& ioptions,
     const MutableCFOptions& mutable_cf_options, const FileOptions& file_options,
     TableCache* table_cache, InternalIterator* iter,
     std::vector<std::unique_ptr<FragmentedRangeTombstoneIterator>>
@@ -83,15 +85,15 @@ Status BuildTable(
     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, IOStatus* io_status,
-    const std::shared_ptr<IOTracer>& io_tracer, EventLogger* event_logger,
-    int job_id, const Env::IOPriority io_priority,
+    const CompressionOptions& compression_opts, bool paranoid_file_checks,
+    InternalStats* internal_stats, TableFileCreationReason reason,
+    IOStatus* io_status, const std::shared_ptr<IOTracer>& io_tracer,
+    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 std::string& db_id,
-    const std::string& db_session_id) {
+    const std::string& db_session_id, const std::string* full_history_ts_low,
+    BlobFileCompletionCallback* blob_callback) {
   assert((column_family_id ==
           TablePropertiesCollectorFactory::Context::kUnknownColumnFamily) ==
          column_family_name.empty());
@@ -120,15 +122,15 @@ Status BuildTable(
   EventHelpers::NotifyTableFileCreationStarted(
       ioptions.listeners, dbname, column_family_name, fname, job_id, reason);
 #endif  // !ROCKSDB_LITE
+  Env* env = db_options.env;
+  assert(env);
+  FileSystem* fs = db_options.fs.get();
+  assert(fs);
+
   TableProperties tp;
   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;
     {
       std::unique_ptr<FSWritableFile> file;
 #ifndef NDEBUG
@@ -148,35 +150,35 @@ Status BuildTable(
             file_checksum, file_checksum_func_name);
         return s;
       }
+      FileTypeSet tmp_set = ioptions.checksum_handoff_file_types;
       file->SetIOPriority(io_priority);
       file->SetWriteLifeTimeHint(write_hint);
-
       file_writer.reset(new WritableFileWriter(
-          std::move(file), fname, file_options, env, io_tracer,
+          std::move(file), fname, file_options, ioptions.clock, io_tracer,
           ioptions.statistics, ioptions.listeners,
-          ioptions.file_checksum_gen_factory));
+          ioptions.file_checksum_gen_factory.get(),
+          tmp_set.Contains(FileType::kTableFile)));
 
       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,
+          column_family_name, file_writer.get(), compression, compression_opts,
+          level, false /* skip_filters */, creation_time, oldest_key_time,
           0 /*target_file_size*/, file_creation_time, db_id, db_session_id);
     }
 
     MergeHelper merge(env, internal_comparator.user_comparator(),
-                      ioptions.merge_operator, nullptr, ioptions.info_log,
+                      ioptions.merge_operator.get(), nullptr, ioptions.info_log,
                       true /* internal key corruption is not ok */,
                       snapshots.empty() ? 0 : snapshots.back(),
                       snapshot_checker);
 
     std::unique_ptr<BlobFileBuilder> blob_file_builder(
         (mutable_cf_options.enable_blob_files && blob_file_additions)
-            ? new BlobFileBuilder(versions, env, fs, &ioptions,
-                                  &mutable_cf_options, &file_options, job_id,
-                                  column_family_id, column_family_name,
-                                  io_priority, write_hint, &blob_file_paths,
+            ? new BlobFileBuilder(versions, fs, &ioptions, &mutable_cf_options,
+                                  &file_options, job_id, column_family_id,
+                                  column_family_name, io_priority, write_hint,
+                                  io_tracer, blob_callback, &blob_file_paths,
                                   blob_file_additions)
             : nullptr);
 
@@ -185,7 +187,11 @@ Status BuildTable(
         &snapshots, earliest_write_conflict_snapshot, snapshot_checker, env,
         ShouldReportDetailedTime(env, ioptions.statistics),
         true /* internal key corruption is not ok */, range_del_agg.get(),
-        blob_file_builder.get(), ioptions.allow_data_in_errors);
+        blob_file_builder.get(), ioptions.allow_data_in_errors,
+        /*compaction=*/nullptr,
+        /*compaction_filter=*/nullptr, /*shutting_down=*/nullptr,
+        /*preserve_deletes_seqnum=*/0, /*manual_compaction_paused=*/nullptr,
+        db_options.info_log, full_history_ts_low);
 
     c_iter.SeekToFirst();
     for (; c_iter.Valid(); c_iter.Next()) {
@@ -212,6 +218,7 @@ Status BuildTable(
     } else if (!c_iter.status().ok()) {
       s = c_iter.status();
     }
+
     if (s.ok()) {
       auto range_del_it = range_del_agg->NewIterator();
       for (range_del_it->SeekToFirst(); range_del_it->Valid();
@@ -222,10 +229,6 @@ Status BuildTable(
         meta->UpdateBoundariesForRange(kv.first, tombstone.SerializeEndKey(),
                                        tombstone.seq_, internal_comparator);
       }
-
-      if (blob_file_builder) {
-        s = blob_file_builder->Finish();
-      }
     }
 
     TEST_SYNC_POINT("BuildTable:BeforeFinishBuildTable");
@@ -254,7 +257,7 @@ Status BuildTable(
     // Finish and check for file errors
     TEST_SYNC_POINT("BuildTable:BeforeSyncTable");
     if (s.ok() && !empty) {
-      StopWatch sw(env, ioptions.statistics, TABLE_SYNC_MICROS);
+      StopWatch sw(ioptions.clock, ioptions.statistics, TABLE_SYNC_MICROS);
       *io_status = file_writer->Sync(ioptions.use_fsync);
     }
     TEST_SYNC_POINT("BuildTable:BeforeCloseTableFile");
@@ -273,6 +276,15 @@ Status BuildTable(
       s = *io_status;
     }
 
+    if (blob_file_builder) {
+      if (s.ok()) {
+        s = blob_file_builder->Finish();
+      } else {
+        blob_file_builder->Abandon();
+      }
+      blob_file_builder.reset();
+    }
+
     // TODO Also check the IO status when create the Iterator.
 
     if (s.ok() && !empty) {
@@ -318,6 +330,8 @@ Status BuildTable(
   }
 
   if (!s.ok() || meta->fd.GetFileSize() == 0) {
+    TEST_SYNC_POINT("BuildTable:BeforeDeleteFile");
+
     constexpr IODebugContext* dbg = nullptr;
 
     Status ignored = fs->DeleteFile(fname, IOOptions(), dbg);
@@ -327,11 +341,11 @@ Status BuildTable(
 
     if (blob_file_additions) {
       for (const std::string& blob_file_path : blob_file_paths) {
-        ignored = fs->DeleteFile(blob_file_path, IOOptions(), dbg);
+        ignored = DeleteDBFile(&db_options, blob_file_path, dbname,
+                               /*force_bg=*/false, /*force_fg=*/false);
         ignored.PermitUncheckedError();
+        TEST_SYNC_POINT("BuildTable::AfterDeleteFile");
       }
-
-      blob_file_additions->clear();
     }
   }
 

db/builder.h

@@ -38,6 +38,7 @@ class VersionEdit;
 class TableBuilder;
 class WritableFileWriter;
 class InternalStats;
+class BlobFileCompletionCallback;
 
 // @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
@@ -49,7 +50,6 @@ 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 bool skip_filters = false, const uint64_t creation_time = 0,
     const uint64_t oldest_key_time = 0, const uint64_t target_file_size = 0,
@@ -65,8 +65,8 @@ TableBuilder* NewTableBuilder(
 // @param column_family_name Name of the column family that is also identified
 //    by column_family_id, or empty string if unknown.
 extern Status BuildTable(
-    const std::string& dbname, VersionSet* versions, Env* env, FileSystem* fs,
-    const ImmutableCFOptions& options,
+    const std::string& dbname, VersionSet* versions,
+    const ImmutableDBOptions& db_options, const ImmutableCFOptions& options,
     const MutableCFOptions& mutable_cf_options, const FileOptions& file_options,
     TableCache* table_cache, InternalIterator* iter,
     std::vector<std::unique_ptr<FragmentedRangeTombstoneIterator>>
@@ -79,7 +79,6 @@ extern Status BuildTable(
     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,
     IOStatus* io_status, const std::shared_ptr<IOTracer>& io_tracer,
@@ -89,6 +88,8 @@ extern Status BuildTable(
     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, const std::string& db_id = "",
-    const std::string& db_session_id = "");
+    const std::string& db_session_id = "",
+    const std::string* full_history_ts_low = nullptr,
+    BlobFileCompletionCallback* blob_callback = nullptr);
 
 }  // namespace ROCKSDB_NAMESPACE

db/c.cc

@@ -11,7 +11,11 @@
 
 #include "rocksdb/c.h"
 
-#include <stdlib.h>
+#include <cstdlib>
+#include <map>
+#include <unordered_set>
+#include <vector>
+
 #include "port/port.h"
 #include "rocksdb/cache.h"
 #include "rocksdb/compaction_filter.h"
@@ -24,6 +28,7 @@
 #include "rocksdb/memtablerep.h"
 #include "rocksdb/merge_operator.h"
 #include "rocksdb/options.h"
+#include "rocksdb/perf_context.h"
 #include "rocksdb/rate_limiter.h"
 #include "rocksdb/slice_transform.h"
 #include "rocksdb/statistics.h"
@@ -39,13 +44,8 @@
 #include "rocksdb/utilities/transaction_db.h"
 #include "rocksdb/utilities/write_batch_with_index.h"
 #include "rocksdb/write_batch.h"
-#include "rocksdb/perf_context.h"
 #include "utilities/merge_operators.h"
 
-#include <vector>
-#include <unordered_set>
-#include <map>
-
 using ROCKSDB_NAMESPACE::BackupableDBOptions;
 using ROCKSDB_NAMESPACE::BackupEngine;
 using ROCKSDB_NAMESPACE::BackupID;
@@ -60,7 +60,6 @@ using ROCKSDB_NAMESPACE::ColumnFamilyDescriptor;
 using ROCKSDB_NAMESPACE::ColumnFamilyHandle;
 using ROCKSDB_NAMESPACE::ColumnFamilyOptions;
 using ROCKSDB_NAMESPACE::CompactionFilter;
-using ROCKSDB_NAMESPACE::CompactionFilterContext;
 using ROCKSDB_NAMESPACE::CompactionFilterFactory;
 using ROCKSDB_NAMESPACE::CompactionOptionsFIFO;
 using ROCKSDB_NAMESPACE::CompactRangeOptions;
@@ -80,9 +79,10 @@ using ROCKSDB_NAMESPACE::IngestExternalFileOptions;
 using ROCKSDB_NAMESPACE::Iterator;
 using ROCKSDB_NAMESPACE::LiveFileMetaData;
 using ROCKSDB_NAMESPACE::Logger;
+using ROCKSDB_NAMESPACE::LRUCacheOptions;
+using ROCKSDB_NAMESPACE::MemoryAllocator;
 using ROCKSDB_NAMESPACE::MemoryUtil;
 using ROCKSDB_NAMESPACE::MergeOperator;
-using ROCKSDB_NAMESPACE::MergeOperators;
 using ROCKSDB_NAMESPACE::NewBloomFilterPolicy;
 using ROCKSDB_NAMESPACE::NewGenericRateLimiter;
 using ROCKSDB_NAMESPACE::NewLRUCache;
@@ -115,10 +115,8 @@ using ROCKSDB_NAMESPACE::WriteBatch;
 using ROCKSDB_NAMESPACE::WriteBatchWithIndex;
 using ROCKSDB_NAMESPACE::WriteOptions;
 
-using std::shared_ptr;
 using std::vector;
 using std::unordered_set;
-using std::map;
 
 extern "C" {
 
@@ -154,6 +152,12 @@ struct rocksdb_filelock_t        { FileLock*         rep; };
 struct rocksdb_logger_t {
   std::shared_ptr<Logger> rep;
 };
+struct rocksdb_lru_cache_options_t {
+  LRUCacheOptions rep;
+};
+struct rocksdb_memory_allocator_t {
+  std::shared_ptr<MemoryAllocator> rep;
+};
 struct rocksdb_cache_t {
   std::shared_ptr<Cache> rep;
 };
@@ -181,6 +185,9 @@ struct rocksdb_transaction_options_t {
 struct rocksdb_transaction_t {
   Transaction* rep;
 };
+struct rocksdb_backupable_db_options_t {
+  BackupableDBOptions rep;
+};
 struct rocksdb_checkpoint_t {
   Checkpoint* rep;
 };
@@ -549,6 +556,18 @@ rocksdb_backup_engine_t* rocksdb_backup_engine_open(
   return result;
 }
 
+rocksdb_backup_engine_t* rocksdb_backup_engine_open_opts(
+    const rocksdb_backupable_db_options_t* options, rocksdb_env_t* env,
+    char** errptr) {
+  BackupEngine* be;
+  if (SaveError(errptr, BackupEngine::Open(options->rep, env->rep, &be))) {
+    return nullptr;
+  }
+  rocksdb_backup_engine_t* result = new rocksdb_backup_engine_t;
+  result->rep = be;
+  return result;
+}
+
 void rocksdb_backup_engine_create_new_backup(rocksdb_backup_engine_t* be,
                                              rocksdb_t* db,
                                              char** errptr) {
@@ -645,6 +664,128 @@ void rocksdb_backup_engine_close(rocksdb_backup_engine_t* be) {
   delete be;
 }
 
+rocksdb_backupable_db_options_t* rocksdb_backupable_db_options_create(
+    const char* backup_dir) {
+  return new rocksdb_backupable_db_options_t{
+      BackupableDBOptions(std::string(backup_dir))};
+}
+
+void rocksdb_backupable_db_options_set_backup_dir(
+    rocksdb_backupable_db_options_t* options, const char* backup_dir) {
+  options->rep.backup_dir = std::string(backup_dir);
+}
+
+void rocksdb_backupable_db_options_set_env(
+    rocksdb_backupable_db_options_t* options, rocksdb_env_t* env) {
+  options->rep.backup_env = (env ? env->rep : nullptr);
+}
+
+void rocksdb_backupable_db_options_set_share_table_files(
+    rocksdb_backupable_db_options_t* options, unsigned char val) {
+  options->rep.share_table_files = val;
+}
+
+unsigned char rocksdb_backupable_db_options_get_share_table_files(
+    rocksdb_backupable_db_options_t* options) {
+  return options->rep.share_table_files;
+}
+
+void rocksdb_backupable_db_options_set_sync(
+    rocksdb_backupable_db_options_t* options, unsigned char val) {
+  options->rep.sync = val;
+}
+
+unsigned char rocksdb_backupable_db_options_get_sync(
+    rocksdb_backupable_db_options_t* options) {
+  return options->rep.sync;
+}
+
+void rocksdb_backupable_db_options_set_destroy_old_data(
+    rocksdb_backupable_db_options_t* options, unsigned char val) {
+  options->rep.destroy_old_data = val;
+}
+
+unsigned char rocksdb_backupable_db_options_get_destroy_old_data(
+    rocksdb_backupable_db_options_t* options) {
+  return options->rep.destroy_old_data;
+}
+
+void rocksdb_backupable_db_options_set_backup_log_files(
+    rocksdb_backupable_db_options_t* options, unsigned char val) {
+  options->rep.backup_log_files = val;
+}
+
+unsigned char rocksdb_backupable_db_options_get_backup_log_files(
+    rocksdb_backupable_db_options_t* options) {
+  return options->rep.backup_log_files;
+}
+
+void rocksdb_backupable_db_options_set_backup_rate_limit(
+    rocksdb_backupable_db_options_t* options, uint64_t limit) {
+  options->rep.backup_rate_limit = limit;
+}
+
+uint64_t rocksdb_backupable_db_options_get_backup_rate_limit(
+    rocksdb_backupable_db_options_t* options) {
+  return options->rep.backup_rate_limit;
+}
+
+void rocksdb_backupable_db_options_set_restore_rate_limit(
+    rocksdb_backupable_db_options_t* options, uint64_t limit) {
+  options->rep.restore_rate_limit = limit;
+}
+
+uint64_t rocksdb_backupable_db_options_get_restore_rate_limit(
+    rocksdb_backupable_db_options_t* options) {
+  return options->rep.restore_rate_limit;
+}
+
+void rocksdb_backupable_db_options_set_max_background_operations(
+    rocksdb_backupable_db_options_t* options, int val) {
+  options->rep.max_background_operations = val;
+}
+
+int rocksdb_backupable_db_options_get_max_background_operations(
+    rocksdb_backupable_db_options_t* options) {
+  return options->rep.max_background_operations;
+}
+
+void rocksdb_backupable_db_options_set_callback_trigger_interval_size(
+    rocksdb_backupable_db_options_t* options, uint64_t size) {
+  options->rep.callback_trigger_interval_size = size;
+}
+
+uint64_t rocksdb_backupable_db_options_get_callback_trigger_interval_size(
+    rocksdb_backupable_db_options_t* options) {
+  return options->rep.callback_trigger_interval_size;
+}
+
+void rocksdb_backupable_db_options_set_max_valid_backups_to_open(
+    rocksdb_backupable_db_options_t* options, int val) {
+  options->rep.max_valid_backups_to_open = val;
+}
+
+int rocksdb_backupable_db_options_get_max_valid_backups_to_open(
+    rocksdb_backupable_db_options_t* options) {
+  return options->rep.max_valid_backups_to_open;
+}
+
+void rocksdb_backupable_db_options_set_share_files_with_checksum_naming(
+    rocksdb_backupable_db_options_t* options, int val) {
+  options->rep.share_files_with_checksum_naming =
+      static_cast<BackupableDBOptions::ShareFilesNaming>(val);
+}
+
+int rocksdb_backupable_db_options_get_share_files_with_checksum_naming(
+    rocksdb_backupable_db_options_t* options) {
+  return static_cast<int>(options->rep.share_files_with_checksum_naming);
+}
+
+void rocksdb_backupable_db_options_destroy(
+    rocksdb_backupable_db_options_t* options) {
+  delete options;
+}
+
 rocksdb_checkpoint_t* rocksdb_checkpoint_object_create(rocksdb_t* db,
                                                        char** errptr) {
   Checkpoint* checkpoint;
@@ -1255,34 +1396,39 @@ char* rocksdb_property_value_cf(
   }
 }
 
-void rocksdb_approximate_sizes(
-    rocksdb_t* db,
-    int num_ranges,
-    const char* const* range_start_key, const size_t* range_start_key_len,
-    const char* const* range_limit_key, const size_t* range_limit_key_len,
-    uint64_t* sizes) {
+void rocksdb_approximate_sizes(rocksdb_t* db, int num_ranges,
+                               const char* const* range_start_key,
+                               const size_t* range_start_key_len,
+                               const char* const* range_limit_key,
+                               const size_t* range_limit_key_len,
+                               uint64_t* sizes, char** errptr) {
   Range* ranges = new Range[num_ranges];
   for (int i = 0; i < num_ranges; i++) {
     ranges[i].start = Slice(range_start_key[i], range_start_key_len[i]);
     ranges[i].limit = Slice(range_limit_key[i], range_limit_key_len[i]);
   }
-  db->rep->GetApproximateSizes(ranges, num_ranges, sizes);
+  Status s = db->rep->GetApproximateSizes(ranges, num_ranges, sizes);
+  if (!s.ok()) {
+    SaveError(errptr, s);
+  }
   delete[] ranges;
 }
 
 void rocksdb_approximate_sizes_cf(
-    rocksdb_t* db,
-    rocksdb_column_family_handle_t* column_family,
-    int num_ranges,
-    const char* const* range_start_key, const size_t* range_start_key_len,
-    const char* const* range_limit_key, const size_t* range_limit_key_len,
-    uint64_t* sizes) {
+    rocksdb_t* db, rocksdb_column_family_handle_t* column_family,
+    int num_ranges, const char* const* range_start_key,
+    const size_t* range_start_key_len, const char* const* range_limit_key,
+    const size_t* range_limit_key_len, uint64_t* sizes, char** errptr) {
   Range* ranges = new Range[num_ranges];
   for (int i = 0; i < num_ranges; i++) {
     ranges[i].start = Slice(range_start_key[i], range_start_key_len[i]);
     ranges[i].limit = Slice(range_limit_key[i], range_limit_key_len[i]);
   }
-  db->rep->GetApproximateSizes(column_family->rep, ranges, num_ranges, sizes);
+  Status s = db->rep->GetApproximateSizes(column_family->rep, ranges,
+                                          num_ranges, sizes);
+  if (!s.ok()) {
+    SaveError(errptr, s);
+  }
   delete[] ranges;
 }
 
@@ -2544,6 +2690,59 @@ unsigned char rocksdb_options_get_skip_checking_sst_file_sizes_on_db_open(
   return opt->rep.skip_checking_sst_file_sizes_on_db_open;
 }
 
+/* Blob Options Settings */
+void rocksdb_options_set_enable_blob_files(rocksdb_options_t* opt,
+                                           unsigned char val) {
+  opt->rep.enable_blob_files = val;
+}
+extern ROCKSDB_LIBRARY_API unsigned char rocksdb_options_get_enable_blob_files(
+    rocksdb_options_t* opt) {
+  return opt->rep.enable_blob_files;
+}
+
+void rocksdb_options_set_min_blob_size(rocksdb_options_t* opt, uint64_t val) {
+  opt->rep.min_blob_size = val;
+}
+
+uint64_t rocksdb_options_get_min_blob_size(rocksdb_options_t* opt) {
+  return opt->rep.min_blob_size;
+}
+
+void rocksdb_options_set_blob_file_size(rocksdb_options_t* opt, uint64_t val) {
+  opt->rep.blob_file_size = val;
+}
+
+uint64_t rocksdb_options_get_blob_file_size(rocksdb_options_t* opt) {
+  return opt->rep.blob_file_size;
+}
+
+void rocksdb_options_set_blob_compression_type(rocksdb_options_t* opt,
+                                               int val) {
+  opt->rep.blob_compression_type = static_cast<CompressionType>(val);
+}
+
+int rocksdb_options_get_blob_compression_type(rocksdb_options_t* opt) {
+  return opt->rep.blob_compression_type;
+}
+
+void rocksdb_options_set_enable_blob_gc(rocksdb_options_t* opt,
+                                        unsigned char val) {
+  opt->rep.enable_blob_garbage_collection = val;
+}
+
+unsigned char rocksdb_options_get_enable_blob_gc(rocksdb_options_t* opt) {
+  return opt->rep.enable_blob_garbage_collection;
+}
+
+void rocksdb_options_set_blob_gc_age_cutoff(rocksdb_options_t* opt,
+                                            double val) {
+  opt->rep.blob_garbage_collection_age_cutoff = val;
+}
+
+double rocksdb_options_get_blob_gc_age_cutoff(rocksdb_options_t* opt) {
+  return opt->rep.blob_garbage_collection_age_cutoff;
+}
+
 void rocksdb_options_set_num_levels(rocksdb_options_t* opt, int n) {
   opt->rep.num_levels = n;
 }
@@ -2636,6 +2835,14 @@ void rocksdb_options_set_bottommost_compression_options_zstd_max_train_bytes(
   opt->rep.bottommost_compression_opts.enabled = enabled;
 }
 
+void rocksdb_options_set_bottommost_compression_options_max_dict_buffer_bytes(
+    rocksdb_options_t* opt, uint64_t max_dict_buffer_bytes,
+    unsigned char enabled) {
+  opt->rep.bottommost_compression_opts.max_dict_buffer_bytes =
+      max_dict_buffer_bytes;
+  opt->rep.bottommost_compression_opts.enabled = enabled;
+}
+
 void rocksdb_options_set_compression_options(rocksdb_options_t* opt, int w_bits,
                                              int level, int strategy,
                                              int max_dict_bytes) {
@@ -2650,6 +2857,11 @@ void rocksdb_options_set_compression_options_zstd_max_train_bytes(
   opt->rep.compression_opts.zstd_max_train_bytes = zstd_max_train_bytes;
 }
 
+void rocksdb_options_set_compression_options_max_dict_buffer_bytes(
+    rocksdb_options_t* opt, uint64_t max_dict_buffer_bytes) {
+  opt->rep.compression_opts.max_dict_buffer_bytes = max_dict_buffer_bytes;
+}
+
 void rocksdb_options_set_prefix_extractor(
     rocksdb_options_t* opt, rocksdb_slicetransform_t* prefix_extractor) {
   opt->rep.prefix_extractor.reset(prefix_extractor);
@@ -2808,6 +3020,8 @@ void rocksdb_options_set_access_hint_on_compaction_start(
       opt->rep.access_hint_on_compaction_start =
           ROCKSDB_NAMESPACE::Options::WILLNEED;
       break;
+    default:
+      assert(0);
   }
 }
 
@@ -3798,6 +4012,25 @@ unsigned char rocksdb_readoptions_get_ignore_range_deletions(
   return opt->rep.ignore_range_deletions;
 }
 
+void rocksdb_readoptions_set_deadline(rocksdb_readoptions_t* opt,
+                                      uint64_t microseconds) {
+  opt->rep.deadline = std::chrono::microseconds(microseconds);
+}
+
+uint64_t rocksdb_readoptions_get_deadline(rocksdb_readoptions_t* opt) {
+  return opt->rep.deadline.count();
+}
+
+void rocksdb_readoptions_set_io_timeout(rocksdb_readoptions_t* opt,
+                                        uint64_t microseconds) {
+  opt->rep.io_timeout = std::chrono::microseconds(microseconds);
+}
+
+extern ROCKSDB_LIBRARY_API uint64_t
+rocksdb_readoptions_get_io_timeout(rocksdb_readoptions_t* opt) {
+  return opt->rep.io_timeout.count();
+}
+
 rocksdb_writeoptions_t* rocksdb_writeoptions_create() {
   return new rocksdb_writeoptions_t;
 }
@@ -3930,16 +4163,58 @@ unsigned char rocksdb_flushoptions_get_wait(rocksdb_flushoptions_t* opt) {
   return opt->rep.wait;
 }
 
+rocksdb_memory_allocator_t* rocksdb_jemalloc_nodump_allocator_create(
+    char** errptr) {
+  rocksdb_memory_allocator_t* allocator = new rocksdb_memory_allocator_t;
+  ROCKSDB_NAMESPACE::JemallocAllocatorOptions options;
+  SaveError(errptr, ROCKSDB_NAMESPACE::NewJemallocNodumpAllocator(
+                        options, &allocator->rep));
+  return allocator;
+}
+
+void rocksdb_memory_allocator_destroy(rocksdb_memory_allocator_t* allocator) {
+  delete allocator;
+}
+
+rocksdb_lru_cache_options_t* rocksdb_lru_cache_options_create() {
+  return new rocksdb_lru_cache_options_t;
+}
+
+void rocksdb_lru_cache_options_destroy(rocksdb_lru_cache_options_t* opt) {
+  delete opt;
+}
+
+void rocksdb_lru_cache_options_set_capacity(rocksdb_lru_cache_options_t* opt,
+                                            size_t capacity) {
+  opt->rep.capacity = capacity;
+}
+
+void rocksdb_lru_cache_options_set_memory_allocator(
+    rocksdb_lru_cache_options_t* opt, rocksdb_memory_allocator_t* allocator) {
+  opt->rep.memory_allocator = allocator->rep;
+}
+
 rocksdb_cache_t* rocksdb_cache_create_lru(size_t capacity) {
   rocksdb_cache_t* c = new rocksdb_cache_t;
   c->rep = NewLRUCache(capacity);
   return c;
 }
 
+rocksdb_cache_t* rocksdb_cache_create_lru_opts(
+    rocksdb_lru_cache_options_t* opt) {
+  rocksdb_cache_t* c = new rocksdb_cache_t;
+  c->rep = NewLRUCache(opt->rep);
+  return c;
+}
+
 void rocksdb_cache_destroy(rocksdb_cache_t* cache) {
   delete cache;
 }
 
+void rocksdb_cache_disown_data(rocksdb_cache_t* cache) {
+  cache->rep->DisownData();
+}
+
 void rocksdb_cache_set_capacity(rocksdb_cache_t* cache, size_t capacity) {
   cache->rep->SetCapacity(capacity);
 }
@@ -3985,20 +4260,36 @@ void rocksdb_env_set_background_threads(rocksdb_env_t* env, int n) {
   env->rep->SetBackgroundThreads(n);
 }
 
+int rocksdb_env_get_background_threads(rocksdb_env_t* env) {
+  return env->rep->GetBackgroundThreads();
+}
+
 void rocksdb_env_set_bottom_priority_background_threads(rocksdb_env_t* env,
                                                         int n) {
   env->rep->SetBackgroundThreads(n, Env::BOTTOM);
 }
 
+int rocksdb_env_get_bottom_priority_background_threads(rocksdb_env_t* env) {
+  return env->rep->GetBackgroundThreads(Env::BOTTOM);
+}
+
 void rocksdb_env_set_high_priority_background_threads(rocksdb_env_t* env, int n) {
   env->rep->SetBackgroundThreads(n, Env::HIGH);
 }
 
+int rocksdb_env_get_high_priority_background_threads(rocksdb_env_t* env) {
+  return env->rep->GetBackgroundThreads(Env::HIGH);
+}
+
 void rocksdb_env_set_low_priority_background_threads(rocksdb_env_t* env,
                                                      int n) {
   env->rep->SetBackgroundThreads(n, Env::LOW);
 }
 
+int rocksdb_env_get_low_priority_background_threads(rocksdb_env_t* env) {
+  return env->rep->GetBackgroundThreads(Env::LOW);
+}
+
 void rocksdb_env_join_all_threads(rocksdb_env_t* env) {
   env->rep->WaitForJoin();
 }
@@ -4226,32 +4517,62 @@ void rocksdb_universal_compaction_options_set_size_ratio(
   uco->rep->size_ratio = ratio;
 }
 
+int rocksdb_universal_compaction_options_get_size_ratio(
+    rocksdb_universal_compaction_options_t* uco) {
+  return uco->rep->size_ratio;
+}
+
 void rocksdb_universal_compaction_options_set_min_merge_width(
   rocksdb_universal_compaction_options_t* uco, int w) {
   uco->rep->min_merge_width = w;
 }
 
+int rocksdb_universal_compaction_options_get_min_merge_width(
+    rocksdb_universal_compaction_options_t* uco) {
+  return uco->rep->min_merge_width;
+}
+
 void rocksdb_universal_compaction_options_set_max_merge_width(
   rocksdb_universal_compaction_options_t* uco, int w) {
   uco->rep->max_merge_width = w;
 }
 
+int rocksdb_universal_compaction_options_get_max_merge_width(
+    rocksdb_universal_compaction_options_t* uco) {
+  return uco->rep->max_merge_width;
+}
+
 void rocksdb_universal_compaction_options_set_max_size_amplification_percent(
   rocksdb_universal_compaction_options_t* uco, int p) {
   uco->rep->max_size_amplification_percent = p;
 }
 
+int rocksdb_universal_compaction_options_get_max_size_amplification_percent(
+    rocksdb_universal_compaction_options_t* uco) {
+  return uco->rep->max_size_amplification_percent;
+}
+
 void rocksdb_universal_compaction_options_set_compression_size_percent(
   rocksdb_universal_compaction_options_t* uco, int p) {
   uco->rep->compression_size_percent = p;
 }
 
+int rocksdb_universal_compaction_options_get_compression_size_percent(
+    rocksdb_universal_compaction_options_t* uco) {
+  return uco->rep->compression_size_percent;
+}
+
 void rocksdb_universal_compaction_options_set_stop_style(
   rocksdb_universal_compaction_options_t* uco, int style) {
   uco->rep->stop_style =
       static_cast<ROCKSDB_NAMESPACE::CompactionStopStyle>(style);
 }
 
+int rocksdb_universal_compaction_options_get_stop_style(
+    rocksdb_universal_compaction_options_t* uco) {
+  return static_cast<int>(uco->rep->stop_style);
+}
+
 void rocksdb_universal_compaction_options_destroy(
   rocksdb_universal_compaction_options_t* uco) {
   delete uco->rep;
@@ -4269,6 +4590,11 @@ void rocksdb_fifo_compaction_options_set_max_table_files_size(
   fifo_opts->rep.max_table_files_size = size;
 }
 
+uint64_t rocksdb_fifo_compaction_options_get_max_table_files_size(
+    rocksdb_fifo_compaction_options_t* fifo_opts) {
+  return fifo_opts->rep.max_table_files_size;
+}
+
 void rocksdb_fifo_compaction_options_destroy(
     rocksdb_fifo_compaction_options_t* fifo_opts) {
   delete fifo_opts;
@@ -4567,7 +4893,10 @@ void rocksdb_transaction_destroy(rocksdb_transaction_t* txn) {
 
 const rocksdb_snapshot_t* rocksdb_transaction_get_snapshot(
     rocksdb_transaction_t* txn) {
-  rocksdb_snapshot_t* result = new rocksdb_snapshot_t;
+  // This will be freed later on using free, so use malloc here to avoid a
+  // mismatch
+  rocksdb_snapshot_t* result =
+      (rocksdb_snapshot_t*)malloc(sizeof(rocksdb_snapshot_t));
   result->rep = txn->rep->GetSnapshot();
   return result;
 }

db/c_test.c

@@ -988,7 +988,9 @@ int main(int argc, char** argv) {
                   &err);
       CheckNoError(err);
     }
-    rocksdb_approximate_sizes(db, 2, start, start_len, limit, limit_len, sizes);
+    rocksdb_approximate_sizes(db, 2, start, start_len, limit, limit_len, sizes,
+                              &err);
+    CheckNoError(err);
     CheckCondition(sizes[0] > 0);
     CheckCondition(sizes[1] > 0);
   }
@@ -1117,9 +1119,8 @@ int main(int argc, char** argv) {
         // Essentially a fingerprint of the block-based Bloom schema
         CheckCondition(hits == 241);
       } else {
-        // Essentially a fingerprint of the full Bloom schema(s),
-        // format_version < 5, which vary for three different CACHE_LINE_SIZEs
-        CheckCondition(hits == 224 || hits == 180 || hits == 125);
+        // Essentially a fingerprint of full Bloom schema, format_version=5
+        CheckCondition(hits == 188);
       }
       CheckCondition(
           (keys_to_query - hits) ==
@@ -1758,6 +1759,25 @@ int main(int argc, char** argv) {
     rocksdb_options_set_atomic_flush(o, 1);
     CheckCondition(1 == rocksdb_options_get_atomic_flush(o));
 
+    /* Blob Options */
+    rocksdb_options_set_enable_blob_files(o, 1);
+    CheckCondition(1 == rocksdb_options_get_enable_blob_files(o));
+
+    rocksdb_options_set_min_blob_size(o, 29);
+    CheckCondition(29 == rocksdb_options_get_min_blob_size(o));
+
+    rocksdb_options_set_blob_file_size(o, 30);
+    CheckCondition(30 == rocksdb_options_get_blob_file_size(o));
+
+    rocksdb_options_set_blob_compression_type(o, 4);
+    CheckCondition(4 == rocksdb_options_get_blob_compression_type(o));
+
+    rocksdb_options_set_enable_blob_gc(o, 1);
+    CheckCondition(1 == rocksdb_options_get_enable_blob_gc(o));
+
+    rocksdb_options_set_blob_gc_age_cutoff(o, 0.75);
+    CheckCondition(0.75 == rocksdb_options_get_blob_gc_age_cutoff(o));
+
     // Create a copy that should be equal to the original.
     rocksdb_options_t* copy;
     copy = rocksdb_options_create_copy(o);
@@ -2279,6 +2299,12 @@ int main(int argc, char** argv) {
     rocksdb_readoptions_set_ignore_range_deletions(ro, 1);
     CheckCondition(1 == rocksdb_readoptions_get_ignore_range_deletions(ro));
 
+    rocksdb_readoptions_set_deadline(ro, 300);
+    CheckCondition(300 == rocksdb_readoptions_get_deadline(ro));
+
+    rocksdb_readoptions_set_io_timeout(ro, 400);
+    CheckCondition(400 == rocksdb_readoptions_get_io_timeout(ro));
+
     rocksdb_readoptions_destroy(ro);
   }
 
@@ -2355,6 +2381,155 @@ int main(int argc, char** argv) {
     rocksdb_cache_destroy(co);
   }
 
+  StartPhase("jemalloc_nodump_allocator");
+  {
+    rocksdb_memory_allocator_t* allocator;
+    allocator = rocksdb_jemalloc_nodump_allocator_create(&err);
+    if (err != NULL) {
+      // not supported on all platforms, allow unsupported error
+      const char* ni = "Not implemented: ";
+      size_t ni_len = strlen(ni);
+      size_t err_len = strlen(err);
+
+      CheckCondition(err_len >= ni_len);
+      CheckCondition(memcmp(ni, err, ni_len) == 0);
+      Free(&err);
+    } else {
+      rocksdb_cache_t* co;
+      rocksdb_lru_cache_options_t* copts;
+
+      copts = rocksdb_lru_cache_options_create();
+
+      rocksdb_lru_cache_options_set_capacity(copts, 100);
+      rocksdb_lru_cache_options_set_memory_allocator(copts, allocator);
+
+      co = rocksdb_cache_create_lru_opts(copts);
+      CheckCondition(100 == rocksdb_cache_get_capacity(co));
+
+      rocksdb_cache_destroy(co);
+      rocksdb_lru_cache_options_destroy(copts);
+    }
+    rocksdb_memory_allocator_destroy(allocator);
+  }
+
+  StartPhase("env");
+  {
+    rocksdb_env_t* e;
+    e = rocksdb_create_default_env();
+
+    rocksdb_env_set_background_threads(e, 10);
+    CheckCondition(10 == rocksdb_env_get_background_threads(e));
+
+    rocksdb_env_set_high_priority_background_threads(e, 20);
+    CheckCondition(20 == rocksdb_env_get_high_priority_background_threads(e));
+
+    rocksdb_env_set_low_priority_background_threads(e, 30);
+    CheckCondition(30 == rocksdb_env_get_low_priority_background_threads(e));
+
+    rocksdb_env_set_bottom_priority_background_threads(e, 40);
+    CheckCondition(40 == rocksdb_env_get_bottom_priority_background_threads(e));
+
+    rocksdb_env_destroy(e);
+  }
+
+  StartPhase("universal_compaction_options");
+  {
+    rocksdb_universal_compaction_options_t* uco;
+    uco = rocksdb_universal_compaction_options_create();
+
+    rocksdb_universal_compaction_options_set_size_ratio(uco, 5);
+    CheckCondition(5 ==
+                   rocksdb_universal_compaction_options_get_size_ratio(uco));
+
+    rocksdb_universal_compaction_options_set_min_merge_width(uco, 15);
+    CheckCondition(
+        15 == rocksdb_universal_compaction_options_get_min_merge_width(uco));
+
+    rocksdb_universal_compaction_options_set_max_merge_width(uco, 25);
+    CheckCondition(
+        25 == rocksdb_universal_compaction_options_get_max_merge_width(uco));
+
+    rocksdb_universal_compaction_options_set_max_size_amplification_percent(uco,
+                                                                            35);
+    CheckCondition(
+        35 ==
+        rocksdb_universal_compaction_options_get_max_size_amplification_percent(
+            uco));
+
+    rocksdb_universal_compaction_options_set_compression_size_percent(uco, 45);
+    CheckCondition(
+        45 ==
+        rocksdb_universal_compaction_options_get_compression_size_percent(uco));
+
+    rocksdb_universal_compaction_options_set_stop_style(uco, 1);
+    CheckCondition(1 ==
+                   rocksdb_universal_compaction_options_get_stop_style(uco));
+
+    rocksdb_universal_compaction_options_destroy(uco);
+  }
+
+  StartPhase("fifo_compaction_options");
+  {
+    rocksdb_fifo_compaction_options_t* fco;
+    fco = rocksdb_fifo_compaction_options_create();
+
+    rocksdb_fifo_compaction_options_set_max_table_files_size(fco, 100000);
+    CheckCondition(
+        100000 ==
+        rocksdb_fifo_compaction_options_get_max_table_files_size(fco));
+
+    rocksdb_fifo_compaction_options_destroy(fco);
+  }
+
+  StartPhase("backupable_db_option");
+  {
+    rocksdb_backupable_db_options_t* bdo;
+    bdo = rocksdb_backupable_db_options_create("path");
+
+    rocksdb_backupable_db_options_set_share_table_files(bdo, 1);
+    CheckCondition(1 ==
+                   rocksdb_backupable_db_options_get_share_table_files(bdo));
+
+    rocksdb_backupable_db_options_set_sync(bdo, 1);
+    CheckCondition(1 == rocksdb_backupable_db_options_get_sync(bdo));
+
+    rocksdb_backupable_db_options_set_destroy_old_data(bdo, 1);
+    CheckCondition(1 ==
+                   rocksdb_backupable_db_options_get_destroy_old_data(bdo));
+
+    rocksdb_backupable_db_options_set_backup_log_files(bdo, 1);
+    CheckCondition(1 ==
+                   rocksdb_backupable_db_options_get_backup_log_files(bdo));
+
+    rocksdb_backupable_db_options_set_backup_rate_limit(bdo, 123);
+    CheckCondition(123 ==
+                   rocksdb_backupable_db_options_get_backup_rate_limit(bdo));
+
+    rocksdb_backupable_db_options_set_restore_rate_limit(bdo, 37);
+    CheckCondition(37 ==
+                   rocksdb_backupable_db_options_get_restore_rate_limit(bdo));
+
+    rocksdb_backupable_db_options_set_max_background_operations(bdo, 20);
+    CheckCondition(
+        20 == rocksdb_backupable_db_options_get_max_background_operations(bdo));
+
+    rocksdb_backupable_db_options_set_callback_trigger_interval_size(bdo, 9000);
+    CheckCondition(
+        9000 ==
+        rocksdb_backupable_db_options_get_callback_trigger_interval_size(bdo));
+
+    rocksdb_backupable_db_options_set_max_valid_backups_to_open(bdo, 40);
+    CheckCondition(
+        40 == rocksdb_backupable_db_options_get_max_valid_backups_to_open(bdo));
+
+    rocksdb_backupable_db_options_set_share_files_with_checksum_naming(bdo, 2);
+    CheckCondition(
+        2 == rocksdb_backupable_db_options_get_share_files_with_checksum_naming(
+                 bdo));
+
+    rocksdb_backupable_db_options_destroy(bdo);
+  }
+
   StartPhase("iterate_upper_bound");
   {
     // Create new empty database
@@ -2744,6 +2919,7 @@ int main(int argc, char** argv) {
   rocksdb_readoptions_destroy(roptions);
   rocksdb_writeoptions_destroy(woptions);
   rocksdb_compactoptions_destroy(coptions);
+  rocksdb_cache_disown_data(cache);
   rocksdb_cache_destroy(cache);
   rocksdb_comparator_destroy(cmp);
   rocksdb_dbpath_destroy(dbpath);
@@ -2755,7 +2931,7 @@ int main(int argc, char** argv) {
 
 #else
 
-int main() {
+int main(void) {
   fprintf(stderr, "SKIPPED\n");
   return 0;
 }

db/column_family.cc

@@ -16,6 +16,7 @@
 #include <string>
 #include <vector>
 
+#include "db/blob/blob_file_cache.h"
 #include "db/compaction/compaction_picker.h"
 #include "db/compaction/compaction_picker_fifo.h"
 #include "db/compaction/compaction_picker_level.h"
@@ -32,6 +33,7 @@
 #include "monitoring/thread_status_util.h"
 #include "options/options_helper.h"
 #include "port/port.h"
+#include "rocksdb/convenience.h"
 #include "rocksdb/table.h"
 #include "table/merging_iterator.h"
 #include "util/autovector.h"
@@ -334,7 +336,9 @@ ColumnFamilyOptions SanitizeOptions(const ImmutableDBOptions& db_options,
   // was not used)
   auto sfm = static_cast<SstFileManagerImpl*>(db_options.sst_file_manager.get());
   for (size_t i = 0; i < result.cf_paths.size(); i++) {
-    DeleteScheduler::CleanupDirectory(db_options.env, sfm, result.cf_paths[i].path);
+    DeleteScheduler::CleanupDirectory(db_options.env, sfm,
+                                      result.cf_paths[i].path)
+        .PermitUncheckedError();
   }
 #endif
 
@@ -448,9 +452,7 @@ void SuperVersion::Cleanup() {
     to_delete.push_back(m);
   }
   current->Unref();
-  if (cfd->Unref()) {
-    delete cfd;
-  }
+  cfd->UnrefAndTryDelete(this);
 }
 
 void SuperVersion::Init(ColumnFamilyData* new_cfd, MemTable* new_mem,
@@ -556,9 +558,13 @@ ColumnFamilyData::ColumnFamilyData(
   // if _dummy_versions is nullptr, then this is a dummy column family.
   if (_dummy_versions != nullptr) {
     internal_stats_.reset(
-        new InternalStats(ioptions_.num_levels, db_options.env, this));
+        new InternalStats(ioptions_.num_levels, ioptions_.clock, this));
     table_cache_.reset(new TableCache(ioptions_, file_options, _table_cache,
                                       block_cache_tracer, io_tracer));
+    blob_file_cache_.reset(
+        new BlobFileCache(_table_cache, ioptions(), soptions(), id_,
+                          internal_stats_->GetBlobFileReadHist(), io_tracer));
+
     if (ioptions_.compaction_style == kCompactionStyleLevel) {
       compaction_picker_.reset(
           new LevelCompactionPicker(ioptions_, &internal_comparator_));
@@ -627,7 +633,7 @@ ColumnFamilyData::~ColumnFamilyData() {
 
   if (dummy_versions_ != nullptr) {
     // List must be empty
-    assert(dummy_versions_->TEST_Next() == dummy_versions_);
+    assert(dummy_versions_->Next() == dummy_versions_);
     bool deleted __attribute__((__unused__));
     deleted = dummy_versions_->Unref();
     assert(deleted);
@@ -655,7 +661,7 @@ ColumnFamilyData::~ColumnFamilyData() {
   }
 }
 
-bool ColumnFamilyData::UnrefAndTryDelete() {
+bool ColumnFamilyData::UnrefAndTryDelete(SuperVersion* sv_under_cleanup) {
   int old_refs = refs_.fetch_sub(1);
   assert(old_refs > 0);
 
@@ -665,7 +671,11 @@ bool ColumnFamilyData::UnrefAndTryDelete() {
     return true;
   }
 
-  if (old_refs == 2 && super_version_ != nullptr) {
+  // If called under SuperVersion::Cleanup, we should not re-enter Cleanup on
+  // the same SuperVersion. (But while installing a new SuperVersion, this
+  // cfd could be referenced only by two SuperVersions.)
+  if (old_refs == 2 && super_version_ != nullptr &&
+      super_version_ != sv_under_cleanup) {
     // Only the super_version_ holds me
     SuperVersion* sv = super_version_;
     super_version_ = nullptr;
@@ -703,9 +713,7 @@ uint64_t ColumnFamilyData::OldestLogToKeep() {
   auto current_log = GetLogNumber();
 
   if (allow_2pc_) {
-    autovector<MemTable*> empty_list;
-    auto imm_prep_log =
-        imm()->PrecomputeMinLogContainingPrepSection(empty_list);
+    auto imm_prep_log = imm()->PrecomputeMinLogContainingPrepSection();
     auto mem_prep_log = mem()->GetMinLogContainingPrepSection();
 
     if (imm_prep_log > 0 && imm_prep_log < current_log) {
@@ -827,7 +835,8 @@ std::pair<WriteStallCondition, ColumnFamilyData::WriteStallCause>
 ColumnFamilyData::GetWriteStallConditionAndCause(
     int num_unflushed_memtables, int num_l0_files,
     uint64_t num_compaction_needed_bytes,
-    const MutableCFOptions& mutable_cf_options) {
+    const MutableCFOptions& mutable_cf_options,
+    const ImmutableCFOptions& immutable_cf_options) {
   if (num_unflushed_memtables >= mutable_cf_options.max_write_buffer_number) {
     return {WriteStallCondition::kStopped, WriteStallCause::kMemtableLimit};
   } else if (!mutable_cf_options.disable_auto_compactions &&
@@ -841,7 +850,9 @@ ColumnFamilyData::GetWriteStallConditionAndCause(
             WriteStallCause::kPendingCompactionBytes};
   } else if (mutable_cf_options.max_write_buffer_number > 3 &&
              num_unflushed_memtables >=
-                 mutable_cf_options.max_write_buffer_number - 1) {
+                 mutable_cf_options.max_write_buffer_number - 1 &&
+             num_unflushed_memtables - 1 >=
+                 immutable_cf_options.min_write_buffer_number_to_merge) {
     return {WriteStallCondition::kDelayed, WriteStallCause::kMemtableLimit};
   } else if (!mutable_cf_options.disable_auto_compactions &&
              mutable_cf_options.level0_slowdown_writes_trigger >= 0 &&
@@ -869,7 +880,8 @@ WriteStallCondition ColumnFamilyData::RecalculateWriteStallConditions(
 
     auto write_stall_condition_and_cause = GetWriteStallConditionAndCause(
         imm()->NumNotFlushed(), vstorage->l0_delay_trigger_count(),
-        vstorage->estimated_compaction_needed_bytes(), mutable_cf_options);
+        vstorage->estimated_compaction_needed_bytes(), mutable_cf_options,
+        *ioptions());
     write_stall_condition = write_stall_condition_and_cause.first;
     auto write_stall_cause = write_stall_condition_and_cause.second;
 
@@ -1083,7 +1095,7 @@ bool ColumnFamilyData::RangeOverlapWithCompaction(
 
 Status ColumnFamilyData::RangesOverlapWithMemtables(
     const autovector<Range>& ranges, SuperVersion* super_version,
-    bool* overlap) {
+    bool allow_data_in_errors, bool* overlap) {
   assert(overlap != nullptr);
   *overlap = false;
   // Create an InternalIterator over all unflushed memtables
@@ -1116,13 +1128,12 @@ Status ColumnFamilyData::RangesOverlapWithMemtables(
     memtable_iter->Seek(range_start.Encode());
     status = memtable_iter->status();
     ParsedInternalKey seek_result;
-    if (status.ok()) {
-      if (memtable_iter->Valid() &&
-          ParseInternalKey(memtable_iter->key(), &seek_result) !=
-              Status::OK()) {
-        status = Status::Corruption("DB have corrupted keys");
-      }
+
+    if (status.ok() && memtable_iter->Valid()) {
+      status = ParseInternalKey(memtable_iter->key(), &seek_result,
+                                allow_data_in_errors);
     }
+
     if (status.ok()) {
       if (memtable_iter->Valid() &&
           ucmp->Compare(seek_result.user_key, ranges[i].limit) <= 0) {
@@ -1335,24 +1346,39 @@ Status ColumnFamilyData::ValidateOptions(
           "Block-Based Table format. ");
     }
   }
+
+  if (cf_options.enable_blob_garbage_collection &&
+      (cf_options.blob_garbage_collection_age_cutoff < 0.0 ||
+       cf_options.blob_garbage_collection_age_cutoff > 1.0)) {
+    return Status::InvalidArgument(
+        "The age cutoff for blob garbage collection should be in the range "
+        "[0.0, 1.0].");
+  }
+
+  if (cf_options.compaction_style == kCompactionStyleFIFO &&
+      db_options.max_open_files != -1 && cf_options.ttl > 0) {
+    return Status::NotSupported(
+        "FIFO compaction only supported with max_open_files = -1.");
+  }
+
   return s;
 }
 
 #ifndef ROCKSDB_LITE
 Status ColumnFamilyData::SetOptions(
-    const DBOptions& db_options,
+    const DBOptions& db_opts,
     const std::unordered_map<std::string, std::string>& options_map) {
-  MutableCFOptions new_mutable_cf_options;
-  Status s =
-      GetMutableOptionsFromStrings(mutable_cf_options_, options_map,
-                                   ioptions_.info_log, &new_mutable_cf_options);
+  ColumnFamilyOptions cf_opts =
+      BuildColumnFamilyOptions(initial_cf_options_, mutable_cf_options_);
+  ConfigOptions config_opts;
+  config_opts.mutable_options_only = true;
+  Status s = GetColumnFamilyOptionsFromMap(config_opts, cf_opts, options_map,
+                                           &cf_opts);
   if (s.ok()) {
-    ColumnFamilyOptions cf_options =
-        BuildColumnFamilyOptions(initial_cf_options_, new_mutable_cf_options);
-    s = ValidateOptions(db_options, cf_options);
+    s = ValidateOptions(db_opts, cf_opts);
   }
   if (s.ok()) {
-    mutable_cf_options_ = new_mutable_cf_options;
+    mutable_cf_options_ = MutableCFOptions(cf_opts);
     mutable_cf_options_.RefreshDerivedOptions(ioptions_);
   }
   return s;

db/column_family.h

@@ -44,6 +44,7 @@ class LogBuffer;
 class InstrumentedMutex;
 class InstrumentedMutexLock;
 struct SuperVersionContext;
+class BlobFileCache;
 
 extern const double kIncSlowdownRatio;
 // This file contains a list of data structures for managing column family
@@ -252,7 +253,7 @@ extern Status CheckCFPathsSupported(const DBOptions& db_options,
 
 extern ColumnFamilyOptions SanitizeOptions(const ImmutableDBOptions& db_options,
                                            const ColumnFamilyOptions& src);
-// Wrap user defined table proproties collector factories `from cf_options`
+// Wrap user defined table properties collector factories `from cf_options`
 // into internal ones in int_tbl_prop_collector_factories. Add a system internal
 // one too.
 extern void GetIntTblPropCollectorFactory(
@@ -278,21 +279,11 @@ class ColumnFamilyData {
   // holding a DB mutex, or as the leader in a write batch group).
   void Ref() { refs_.fetch_add(1); }
 
-  // Unref decreases the reference count, but does not handle deletion
-  // when the count goes to 0.  If this method returns true then the
-  // caller should delete the instance immediately, or later, by calling
-  // FreeDeadColumnFamilies().  Unref() can only be called while holding
-  // a DB mutex, or during single-threaded recovery.
-  bool Unref() {
-    int old_refs = refs_.fetch_sub(1);
-    assert(old_refs > 0);
-    return old_refs == 1;
-  }
-
   // UnrefAndTryDelete() decreases the reference count and do free if needed,
   // return true if this is freed else false, UnrefAndTryDelete() can only
   // be called while holding a DB mutex, or during single-threaded recovery.
-  bool UnrefAndTryDelete();
+  // sv_under_cleanup is only provided when called from SuperVersion::Cleanup.
+  bool UnrefAndTryDelete(SuperVersion* sv_under_cleanup = nullptr);
 
   // SetDropped() can only be called under following conditions:
   // 1) Holding a DB mutex,
@@ -359,6 +350,11 @@ class ColumnFamilyData {
 
   MemTableList* imm() { return &imm_; }
   MemTable* mem() { return mem_; }
+
+  bool IsEmpty() {
+    return mem()->GetFirstSequenceNumber() == 0 && imm()->NumNotFlushed() == 0;
+  }
+
   Version* current() { return current_; }
   Version* dummy_versions() { return dummy_versions_; }
   void SetCurrent(Version* _current);
@@ -381,6 +377,7 @@ class ColumnFamilyData {
                          SequenceNumber earliest_seq);
 
   TableCache* table_cache() const { return table_cache_.get(); }
+  BlobFileCache* blob_file_cache() const { return blob_file_cache_.get(); }
 
   // See documentation in compaction_picker.h
   // REQUIRES: DB mutex held
@@ -404,7 +401,8 @@ class ColumnFamilyData {
   //
   // Thread-safe
   Status RangesOverlapWithMemtables(const autovector<Range>& ranges,
-                                    SuperVersion* super_version, bool* overlap);
+                                    SuperVersion* super_version,
+                                    bool allow_data_in_errors, bool* overlap);
 
   // A flag to tell a manual compaction is to compact all levels together
   // instead of a specific level.
@@ -443,7 +441,7 @@ class ColumnFamilyData {
   // Get SuperVersion stored in thread local storage. If it does not exist,
   // get a reference from a current SuperVersion.
   SuperVersion* GetThreadLocalSuperVersion(DBImpl* db);
-  // Try to return SuperVersion back to thread local storage. Retrun true on
+  // Try to return SuperVersion back to thread local storage. Return true on
   // success and false on failure. It fails when the thread local storage
   // contains anything other than SuperVersion::kSVInUse flag.
   bool ReturnThreadLocalSuperVersion(SuperVersion* sv);
@@ -477,9 +475,11 @@ class ColumnFamilyData {
     kPendingCompactionBytes,
   };
   static std::pair<WriteStallCondition, WriteStallCause>
-  GetWriteStallConditionAndCause(int num_unflushed_memtables, int num_l0_files,
-                                 uint64_t num_compaction_needed_bytes,
-                                 const MutableCFOptions& mutable_cf_options);
+  GetWriteStallConditionAndCause(
+      int num_unflushed_memtables, int num_l0_files,
+      uint64_t num_compaction_needed_bytes,
+      const MutableCFOptions& mutable_cf_options,
+      const ImmutableCFOptions& immutable_cf_options);
 
   // Recalculate some small conditions, which are changed only during
   // compaction, adding new memtable and/or
@@ -506,6 +506,21 @@ class ColumnFamilyData {
 
   FSDirectory* GetDataDir(size_t path_id) const;
 
+  // full_history_ts_low_ can only increase.
+  void SetFullHistoryTsLow(std::string ts_low) {
+    assert(!ts_low.empty());
+    const Comparator* ucmp = user_comparator();
+    assert(ucmp);
+    if (full_history_ts_low_.empty() ||
+        ucmp->CompareTimestamp(ts_low, full_history_ts_low_) > 0) {
+      full_history_ts_low_ = std::move(ts_low);
+    }
+  }
+
+  const std::string& GetFullHistoryTsLow() const {
+    return full_history_ts_low_;
+  }
+
   ThreadLocalPtr* TEST_GetLocalSV() { return local_sv_.get(); }
 
  private:
@@ -543,6 +558,7 @@ class ColumnFamilyData {
   const bool is_delete_range_supported_;
 
   std::unique_ptr<TableCache> table_cache_;
+  std::unique_ptr<BlobFileCache> blob_file_cache_;
 
   std::unique_ptr<InternalStats> internal_stats_;
 
@@ -601,6 +617,8 @@ class ColumnFamilyData {
   std::vector<std::shared_ptr<FSDirectory>> data_dirs_;
 
   bool db_paths_registered_;
+
+  std::string full_history_ts_low_;
 };
 
 // ColumnFamilySet has interesting thread-safety requirements

db/column_family_test.cc

@@ -35,10 +35,10 @@ namespace ROCKSDB_NAMESPACE {
 static const int kValueSize = 1000;
 
 // counts how many operations were performed
-class EnvCounter : public EnvWrapper {
+class EnvCounter : public SpecialEnv {
  public:
   explicit EnvCounter(Env* base)
-      : EnvWrapper(base), num_new_writable_file_(0) {}
+      : SpecialEnv(base), num_new_writable_file_(0) {}
   int GetNumberOfNewWritableFileCalls() {
     return num_new_writable_file_;
   }
@@ -68,11 +68,12 @@ class ColumnFamilyTestBase : public testing::Test {
 #endif  // !ROCKSDB_LITE
     EXPECT_NE(nullptr, base_env);
     env_ = new EnvCounter(base_env);
+    env_->skip_fsync_ = true;
     dbname_ = test::PerThreadDBPath("column_family_test");
     db_options_.create_if_missing = true;
     db_options_.fail_if_options_file_error = true;
     db_options_.env = env_;
-    DestroyDB(dbname_, Options(db_options_, column_family_options_));
+    EXPECT_OK(DestroyDB(dbname_, Options(db_options_, column_family_options_)));
   }
 
   ~ColumnFamilyTestBase() override {
@@ -87,7 +88,6 @@ class ColumnFamilyTestBase : public testing::Test {
 #endif  // ROCKSDB_LITE
       column_families.push_back(cfdescriptor);
     }
-    Close();
     ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
     Destroy(column_families);
     delete env_;
@@ -187,8 +187,8 @@ class ColumnFamilyTestBase : public testing::Test {
     std::vector<ColumnFamilyDescriptor> column_families;
     names_.clear();
     for (size_t i = 0; i < cf.size(); ++i) {
-      column_families.push_back(ColumnFamilyDescriptor(
-          cf[i], options.size() == 0 ? column_family_options_ : options[i]));
+      column_families.emplace_back(
+          cf[i], options.size() == 0 ? column_family_options_ : options[i]);
       names_.push_back(cf[i]);
     }
     return DB::Open(db_options_, dbname_, column_families, &handles_, &db_);
@@ -199,8 +199,8 @@ class ColumnFamilyTestBase : public testing::Test {
     std::vector<ColumnFamilyDescriptor> column_families;
     names_.clear();
     for (size_t i = 0; i < cf.size(); ++i) {
-      column_families.push_back(ColumnFamilyDescriptor(
-          cf[i], options.size() == 0 ? column_family_options_ : options[i]));
+      column_families.emplace_back(
+          cf[i], options.size() == 0 ? column_family_options_ : options[i]);
       names_.push_back(cf[i]);
     }
     return DB::OpenForReadOnly(db_options_, dbname_, column_families, &handles_,
@@ -653,8 +653,8 @@ TEST_P(FlushEmptyCFTestWithParam, FlushEmptyCFTest) {
   // after flushing file B is deleted. At the same time, the min log number of
   // default CF is not written to manifest. Log file A still remains.
   // Flushed to SST file Y.
-  Flush(1);
-  Flush(0);
+  ASSERT_OK(Flush(1));
+  ASSERT_OK(Flush(0));
   ASSERT_OK(Put(1, "bar", "v3"));  // seqID 4
   ASSERT_OK(Put(1, "foo", "v4"));  // seqID 5
   ASSERT_OK(db_->FlushWAL(/*sync=*/false));
@@ -708,15 +708,15 @@ TEST_P(FlushEmptyCFTestWithParam, FlushEmptyCFTest2) {
   // and is set to current. Both CFs' min log number is set to file C so after
   // flushing file B is deleted. Log file A still remains.
   // Flushed to SST file Y.
-  Flush(1);
+  ASSERT_OK(Flush(1));
   ASSERT_OK(Put(0, "bar", "v2"));  // seqID 4
   ASSERT_OK(Put(2, "bar", "v2"));  // seqID 5
   ASSERT_OK(Put(1, "bar", "v3"));  // seqID 6
   // Flushing all column families. This forces all CFs' min log to current. This
   // is written to the manifest file. Log file C is cleared.
-  Flush(0);
-  Flush(1);
-  Flush(2);
+  ASSERT_OK(Flush(0));
+  ASSERT_OK(Flush(1));
+  ASSERT_OK(Flush(2));
   // Write to log file D
   ASSERT_OK(Put(1, "bar", "v4"));  // seqID 7
   ASSERT_OK(Put(1, "bar", "v5"));  // seqID 8
@@ -898,9 +898,7 @@ TEST_P(ColumnFamilyTest, IgnoreRecoveredLog) {
   std::vector<std::string> old_files;
   ASSERT_OK(env_->GetChildren(backup_logs, &old_files));
   for (auto& file : old_files) {
-    if (file != "." && file != "..") {
-      ASSERT_OK(env_->DeleteFile(backup_logs + "/" + file));
-    }
+    ASSERT_OK(env_->DeleteFile(backup_logs + "/" + file));
   }
 
   column_family_options_.merge_operator =
@@ -929,9 +927,7 @@ TEST_P(ColumnFamilyTest, IgnoreRecoveredLog) {
   std::vector<std::string> logs;
   ASSERT_OK(env_->GetChildren(db_options_.wal_dir, &logs));
   for (auto& log : logs) {
-    if (log != ".." && log != ".") {
-      CopyFile(db_options_.wal_dir + "/" + log, backup_logs + "/" + log);
-    }
+    CopyFile(db_options_.wal_dir + "/" + log, backup_logs + "/" + log);
   }
 
   // recover the DB
@@ -956,9 +952,7 @@ TEST_P(ColumnFamilyTest, IgnoreRecoveredLog) {
     if (iter == 0) {
       // copy the logs from backup back to wal dir
       for (auto& log : logs) {
-        if (log != ".." && log != ".") {
-          CopyFile(backup_logs + "/" + log, db_options_.wal_dir + "/" + log);
-        }
+        CopyFile(backup_logs + "/" + log, db_options_.wal_dir + "/" + log);
       }
     }
   }
@@ -985,7 +979,7 @@ TEST_P(ColumnFamilyTest, FlushTest) {
     for (int i = 0; i < 3; ++i) {
       uint64_t max_total_in_memory_state =
           MaxTotalInMemoryState();
-      Flush(i);
+      ASSERT_OK(Flush(i));
       AssertMaxTotalInMemoryState(max_total_in_memory_state);
     }
     ASSERT_OK(Put(1, "foofoo", "bar"));
@@ -1093,7 +1087,7 @@ TEST_P(ColumnFamilyTest, CrashAfterFlush) {
   ASSERT_OK(batch.Put(handles_[0], Slice("foo"), Slice("bar")));
   ASSERT_OK(batch.Put(handles_[1], Slice("foo"), Slice("bar")));
   ASSERT_OK(db_->Write(WriteOptions(), &batch));
-  Flush(0);
+  ASSERT_OK(Flush(0));
   fault_env->SetFilesystemActive(false);
 
   std::vector<std::string> names;
@@ -1103,7 +1097,7 @@ TEST_P(ColumnFamilyTest, CrashAfterFlush) {
     }
   }
   Close();
-  fault_env->DropUnsyncedFileData();
+  ASSERT_OK(fault_env->DropUnsyncedFileData());
   fault_env->ResetState();
   Open(names, {});
 
@@ -2236,7 +2230,7 @@ TEST_P(ColumnFamilyTest, FlushStaleColumnFamilies) {
   // files for column family [one], because it's empty
   AssertCountLiveFiles(4);
 
-  Flush(0);
+  ASSERT_OK(Flush(0));
   ASSERT_EQ(0, dbfull()->TEST_total_log_size());
   Close();
 }
@@ -3040,7 +3034,7 @@ TEST_P(ColumnFamilyTest, IteratorCloseWALFile1) {
   Iterator* it = db_->NewIterator(ReadOptions(), handles_[1]);
   ASSERT_OK(it->status());
   // A flush will make `it` hold the last reference of its super version.
-  Flush(1);
+  ASSERT_OK(Flush(1));
 
   ASSERT_OK(Put(1, "fodor", "mirko"));
   ASSERT_OK(Put(0, "fodor", "mirko"));
@@ -3093,7 +3087,7 @@ TEST_P(ColumnFamilyTest, IteratorCloseWALFile2) {
   Iterator* it = db_->NewIterator(ro, handles_[1]);
   ASSERT_OK(it->status());
   // A flush will make `it` hold the last reference of its super version.
-  Flush(1);
+  ASSERT_OK(Flush(1));
 
   ASSERT_OK(Put(1, "fodor", "mirko"));
   ASSERT_OK(Put(0, "fodor", "mirko"));
@@ -3147,7 +3141,7 @@ TEST_P(ColumnFamilyTest, ForwardIteratorCloseWALFile) {
   CreateColumnFamilies({"one"});
   ASSERT_OK(Put(1, "fodor", "mirko"));
   ASSERT_OK(Put(1, "fodar2", "mirko"));
-  Flush(1);
+  ASSERT_OK(Flush(1));
 
   // Create an iterator holding the current super version, as well as
   // the SST file just flushed.
@@ -3159,7 +3153,7 @@ TEST_P(ColumnFamilyTest, ForwardIteratorCloseWALFile) {
 
   ASSERT_OK(Put(1, "fodor", "mirko"));
   ASSERT_OK(Put(1, "fodar2", "mirko"));
-  Flush(1);
+  ASSERT_OK(Flush(1));
 
   WaitForCompaction();
 
@@ -3232,9 +3226,9 @@ TEST_P(ColumnFamilyTest, LogSyncConflictFlush) {
   ROCKSDB_NAMESPACE::port::Thread thread([&] { ASSERT_OK(db_->SyncWAL()); });
 
   TEST_SYNC_POINT("ColumnFamilyTest::LogSyncConflictFlush:1");
-  Flush(1);
+  ASSERT_OK(Flush(1));
   ASSERT_OK(Put(1, "foo", "bar"));
-  Flush(1);
+  ASSERT_OK(Flush(1));
 
   TEST_SYNC_POINT("ColumnFamilyTest::LogSyncConflictFlush:2");
 
@@ -3256,7 +3250,7 @@ TEST_P(ColumnFamilyTest, DISABLED_LogTruncationTest) {
   Build(0, 100);
 
   // Flush the 0th column family to force a roll of the wal log
-  Flush(0);
+  ASSERT_OK(Flush(0));
 
   // Add some more entries
   Build(100, 100);
@@ -3271,7 +3265,7 @@ TEST_P(ColumnFamilyTest, DISABLED_LogTruncationTest) {
     FileType type;
     if (!(ParseFileName(filenames[i], &number, &type))) continue;
 
-    if (type != kLogFile) continue;
+    if (type != kWalFile) continue;
 
     logfs.push_back(filenames[i]);
   }
@@ -3332,14 +3326,14 @@ TEST_P(ColumnFamilyTest, DefaultCfPathsTest) {
 
   // Fill Column family 1.
   PutRandomData(1, 100, 100);
-  Flush(1);
+  ASSERT_OK(Flush(1));
 
   ASSERT_EQ(1, GetSstFileCount(cf_opt1.cf_paths[0].path));
   ASSERT_EQ(0, GetSstFileCount(dbname_));
 
   // Fill column family 2
   PutRandomData(2, 100, 100);
-  Flush(2);
+  ASSERT_OK(Flush(2));
 
   // SST from Column family 2 should be generated in
   // db_paths which is dbname_ in this case.
@@ -3358,14 +3352,14 @@ TEST_P(ColumnFamilyTest, MultipleCFPathsTest) {
   Reopen({ColumnFamilyOptions(), cf_opt1, cf_opt2});
 
   PutRandomData(1, 100, 100, true /* save */);
-  Flush(1);
+  ASSERT_OK(Flush(1));
 
   // Check that files are generated in appropriate paths.
   ASSERT_EQ(1, GetSstFileCount(cf_opt1.cf_paths[0].path));
   ASSERT_EQ(0, GetSstFileCount(dbname_));
 
   PutRandomData(2, 100, 100, true /* save */);
-  Flush(2);
+  ASSERT_OK(Flush(2));
 
   ASSERT_EQ(1, GetSstFileCount(cf_opt2.cf_paths[0].path));
   ASSERT_EQ(0, GetSstFileCount(dbname_));
@@ -3391,6 +3385,30 @@ TEST_P(ColumnFamilyTest, MultipleCFPathsTest) {
   }
 }
 
+TEST(ColumnFamilyTest, ValidateBlobGCCutoff) {
+  DBOptions db_options;
+
+  ColumnFamilyOptions cf_options;
+  cf_options.enable_blob_garbage_collection = true;
+
+  cf_options.blob_garbage_collection_age_cutoff = -0.5;
+  ASSERT_TRUE(ColumnFamilyData::ValidateOptions(db_options, cf_options)
+                  .IsInvalidArgument());
+
+  cf_options.blob_garbage_collection_age_cutoff = 0.0;
+  ASSERT_OK(ColumnFamilyData::ValidateOptions(db_options, cf_options));
+
+  cf_options.blob_garbage_collection_age_cutoff = 0.5;
+  ASSERT_OK(ColumnFamilyData::ValidateOptions(db_options, cf_options));
+
+  cf_options.blob_garbage_collection_age_cutoff = 1.0;
+  ASSERT_OK(ColumnFamilyData::ValidateOptions(db_options, cf_options));
+
+  cf_options.blob_garbage_collection_age_cutoff = 1.5;
+  ASSERT_TRUE(ColumnFamilyData::ValidateOptions(db_options, cf_options)
+                  .IsInvalidArgument());
+}
+
 }  // namespace ROCKSDB_NAMESPACE
 
 #ifdef ROCKSDB_UNITTESTS_WITH_CUSTOM_OBJECTS_FROM_STATIC_LIBS

db/compact_files_test.cc

@@ -91,9 +91,9 @@ TEST_F(CompactFilesTest, L0ConflictsFiles) {
   // create couple files
   // Background compaction starts and waits in BackgroundCallCompaction:0
   for (int i = 0; i < kLevel0Trigger * 4; ++i) {
-    db->Put(WriteOptions(), ToString(i), "");
-    db->Put(WriteOptions(), ToString(100 - i), "");
-    db->Flush(FlushOptions());
+    ASSERT_OK(db->Put(WriteOptions(), ToString(i), ""));
+    ASSERT_OK(db->Put(WriteOptions(), ToString(100 - i), ""));
+    ASSERT_OK(db->Flush(FlushOptions()));
   }
 
   ROCKSDB_NAMESPACE::ColumnFamilyMetaData meta;
@@ -118,6 +118,78 @@ TEST_F(CompactFilesTest, L0ConflictsFiles) {
   delete db;
 }
 
+TEST_F(CompactFilesTest, MultipleLevel) {
+  Options options;
+  options.create_if_missing = true;
+  options.level_compaction_dynamic_level_bytes = true;
+  options.num_levels = 6;
+  // 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_OK(s);
+  ASSERT_NE(db, nullptr);
+
+  // create couple files in L0, L3, L4 and L5
+  for (int i = 5; i > 2; --i) {
+    collector->ClearFlushedFiles();
+    ASSERT_OK(db->Put(WriteOptions(), ToString(i), ""));
+    ASSERT_OK(db->Flush(FlushOptions()));
+    auto l0_files = collector->GetFlushedFiles();
+    ASSERT_OK(db->CompactFiles(CompactionOptions(), l0_files, i));
+
+    std::string prop;
+    ASSERT_TRUE(
+        db->GetProperty("rocksdb.num-files-at-level" + ToString(i), &prop));
+    ASSERT_EQ("1", prop);
+  }
+  ASSERT_OK(db->Put(WriteOptions(), ToString(0), ""));
+  ASSERT_OK(db->Flush(FlushOptions()));
+
+  ColumnFamilyMetaData meta;
+  db->GetColumnFamilyMetaData(&meta);
+  // Compact files except the file in L3
+  std::vector<std::string> files;
+  for (int i = 0; i < 6; ++i) {
+    if (i == 3) continue;
+    for (auto& file : meta.levels[i].files) {
+      files.push_back(file.db_path + "/" + file.name);
+    }
+  }
+
+  SyncPoint::GetInstance()->LoadDependency({
+      {"CompactionJob::Run():Start", "CompactFilesTest.MultipleLevel:0"},
+      {"CompactFilesTest.MultipleLevel:1", "CompactFilesImpl:3"},
+  });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  std::thread thread([&] {
+    TEST_SYNC_POINT("CompactFilesTest.MultipleLevel:0");
+    ASSERT_OK(db->Put(WriteOptions(), "bar", "v2"));
+    ASSERT_OK(db->Put(WriteOptions(), "foo", "v2"));
+    ASSERT_OK(db->Flush(FlushOptions()));
+    TEST_SYNC_POINT("CompactFilesTest.MultipleLevel:1");
+  });
+
+  // Compaction cannot move up the data to higher level
+  // here we have input file from level 5, so the output level has to be >= 5
+  for (int invalid_output_level = 0; invalid_output_level < 5;
+       invalid_output_level++) {
+    s = db->CompactFiles(CompactionOptions(), files, invalid_output_level);
+    std::cout << s.ToString() << std::endl;
+    ASSERT_TRUE(s.IsInvalidArgument());
+  }
+
+  ASSERT_OK(db->CompactFiles(CompactionOptions(), files, 5));
+  SyncPoint::GetInstance()->DisableProcessing();
+  thread.join();
+
+  delete db;
+}
+
 TEST_F(CompactFilesTest, ObsoleteFiles) {
   Options options;
   // to trigger compaction more easily
@@ -138,18 +210,18 @@ TEST_F(CompactFilesTest, ObsoleteFiles) {
   DB* db = nullptr;
   DestroyDB(db_name_, options);
   Status s = DB::Open(options, db_name_, &db);
-  assert(s.ok());
-  assert(db);
+  ASSERT_OK(s);
+  ASSERT_NE(db, nullptr);
 
   // create couple files
   for (int i = 1000; i < 2000; ++i) {
-    db->Put(WriteOptions(), ToString(i),
-            std::string(kWriteBufferSize / 10, 'a' + (i % 26)));
+    ASSERT_OK(db->Put(WriteOptions(), ToString(i),
+                      std::string(kWriteBufferSize / 10, 'a' + (i % 26))));
   }
 
   auto l0_files = collector->GetFlushedFiles();
   ASSERT_OK(db->CompactFiles(CompactionOptions(), l0_files, 1));
-  static_cast_with_check<DBImpl>(db)->TEST_WaitForCompact();
+  ASSERT_OK(static_cast_with_check<DBImpl>(db)->TEST_WaitForCompact());
 
   // verify all compaction input files are deleted
   for (auto fname : l0_files) {
@@ -182,15 +254,17 @@ TEST_F(CompactFilesTest, NotCutOutputOnLevel0) {
 
   // create couple files
   for (int i = 0; i < 500; ++i) {
-    db->Put(WriteOptions(), ToString(i), std::string(1000, 'a' + (i % 26)));
+    ASSERT_OK(db->Put(WriteOptions(), ToString(i),
+                      std::string(1000, 'a' + (i % 26))));
   }
-  static_cast_with_check<DBImpl>(db)->TEST_WaitForFlushMemTable();
+  ASSERT_OK(static_cast_with_check<DBImpl>(db)->TEST_WaitForFlushMemTable());
   auto l0_files_1 = collector->GetFlushedFiles();
   collector->ClearFlushedFiles();
   for (int i = 0; i < 500; ++i) {
-    db->Put(WriteOptions(), ToString(i), std::string(1000, 'a' + (i % 26)));
+    ASSERT_OK(db->Put(WriteOptions(), ToString(i),
+                      std::string(1000, 'a' + (i % 26))));
   }
-  static_cast_with_check<DBImpl>(db)->TEST_WaitForFlushMemTable();
+  ASSERT_OK(static_cast_with_check<DBImpl>(db)->TEST_WaitForFlushMemTable());
   auto l0_files_2 = collector->GetFlushedFiles();
   ASSERT_OK(db->CompactFiles(CompactionOptions(), l0_files_1, 0));
   ASSERT_OK(db->CompactFiles(CompactionOptions(), l0_files_2, 0));
@@ -213,13 +287,13 @@ TEST_F(CompactFilesTest, CapturingPendingFiles) {
   DB* db = nullptr;
   DestroyDB(db_name_, options);
   Status s = DB::Open(options, db_name_, &db);
-  assert(s.ok());
+  ASSERT_OK(s);
   assert(db);
 
   // Create 5 files.
   for (int i = 0; i < 5; ++i) {
-    db->Put(WriteOptions(), "key" + ToString(i), "value");
-    db->Flush(FlushOptions());
+    ASSERT_OK(db->Put(WriteOptions(), "key" + ToString(i), "value"));
+    ASSERT_OK(db->Flush(FlushOptions()));
   }
 
   auto l0_files = collector->GetFlushedFiles();
@@ -237,8 +311,8 @@ TEST_F(CompactFilesTest, CapturingPendingFiles) {
 
   // In the meantime flush another file.
   TEST_SYNC_POINT("CompactFilesTest.CapturingPendingFiles:0");
-  db->Put(WriteOptions(), "key5", "value");
-  db->Flush(FlushOptions());
+  ASSERT_OK(db->Put(WriteOptions(), "key5", "value"));
+  ASSERT_OK(db->Flush(FlushOptions()));
   TEST_SYNC_POINT("CompactFilesTest.CapturingPendingFiles:1");
 
   compaction_thread.join();
@@ -249,7 +323,7 @@ TEST_F(CompactFilesTest, CapturingPendingFiles) {
 
   // Make sure we can reopen the DB.
   s = DB::Open(options, db_name_, &db);
-  ASSERT_TRUE(s.ok());
+  ASSERT_OK(s);
   assert(db);
   delete db;
 }
@@ -293,8 +367,8 @@ TEST_F(CompactFilesTest, CompactionFilterWithGetSv) {
   cf->SetDB(db);
 
   // Write one L0 file
-  db->Put(WriteOptions(), "K1", "V1");
-  db->Flush(FlushOptions());
+  ASSERT_OK(db->Put(WriteOptions(), "K1", "V1"));
+  ASSERT_OK(db->Flush(FlushOptions()));
 
   // Compact all L0 files using CompactFiles
   ROCKSDB_NAMESPACE::ColumnFamilyMetaData meta;
@@ -337,8 +411,8 @@ TEST_F(CompactFilesTest, SentinelCompressionType) {
     DB* db = nullptr;
     ASSERT_OK(DB::Open(options, db_name_, &db));
 
-    db->Put(WriteOptions(), "key", "val");
-    db->Flush(FlushOptions());
+    ASSERT_OK(db->Put(WriteOptions(), "key", "val"));
+    ASSERT_OK(db->Flush(FlushOptions()));
 
     auto l0_files = collector->GetFlushedFiles();
     ASSERT_EQ(1, l0_files.size());
@@ -377,14 +451,15 @@ TEST_F(CompactFilesTest, GetCompactionJobInfo) {
   DB* db = nullptr;
   DestroyDB(db_name_, options);
   Status s = DB::Open(options, db_name_, &db);
-  assert(s.ok());
+  ASSERT_OK(s);
   assert(db);
 
   // create couple files
   for (int i = 0; i < 500; ++i) {
-    db->Put(WriteOptions(), ToString(i), std::string(1000, 'a' + (i % 26)));
+    ASSERT_OK(db->Put(WriteOptions(), ToString(i),
+                      std::string(1000, 'a' + (i % 26))));
   }
-  static_cast_with_check<DBImpl>(db)->TEST_WaitForFlushMemTable();
+  ASSERT_OK(static_cast_with_check<DBImpl>(db)->TEST_WaitForFlushMemTable());
   auto l0_files_1 = collector->GetFlushedFiles();
   CompactionOptions co;
   co.compression = CompressionType::kLZ4Compression;

db/compaction/compaction.cc

@@ -248,12 +248,6 @@ Compaction::Compaction(VersionStorageInfo* vstorage,
   if (max_subcompactions_ == 0) {
     max_subcompactions_ = _mutable_db_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) {
@@ -383,7 +377,13 @@ bool Compaction::KeyNotExistsBeyondOutputLevel(
         auto* f = files[level_ptrs->at(lvl)];
         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) {
+          // In the presence of user-defined timestamp, we may need to handle
+          // the case in which f->smallest.user_key() (including ts) has the
+          // same user key, but the ts part is smaller. If so,
+          // Compare(user_key, f->smallest.user_key()) returns -1.
+          // That's why we need CompareWithoutTimestamp().
+          if (user_cmp->CompareWithoutTimestamp(user_key,
+                                                f->smallest.user_key()) >= 0) {
             // Key falls in this file's range, so it may
             // exist beyond output level
             return false;
@@ -519,7 +519,7 @@ uint64_t Compaction::OutputFilePreallocationSize() const {
 
   // Over-estimate slightly so we don't end up just barely crossing
   // the threshold
-  // No point to prellocate more than 1GB.
+  // No point to preallocate more than 1GB.
   return std::min(uint64_t{1073741824},
                   preallocation_size + (preallocation_size / 10));
 }

db/compaction/compaction.h

@@ -341,7 +341,7 @@ class Compaction {
   const uint32_t output_path_id_;
   CompressionType output_compression_;
   CompressionOptions output_compression_opts_;
-  // If true, then the comaction can be done by simply deleting input files.
+  // If true, then the compaction can be done by simply deleting input files.
   const bool deletion_compaction_;
 
   // Compaction input files organized by level. Constant after construction

db/compaction/compaction_iteration_stats.h

@@ -34,4 +34,8 @@ struct CompactionIterationStats {
   // Single-Delete diagnostics for exceptional situations
   uint64_t num_single_del_fallthru = 0;
   uint64_t num_single_del_mismatch = 0;
+
+  // Blob related statistics
+  uint64_t num_blobs_read = 0;
+  uint64_t total_blob_bytes_read = 0;
 };

db/compaction/compaction_iterator.cc

@@ -5,9 +5,11 @@
 
 #include "db/compaction/compaction_iterator.h"
 
-#include <cinttypes>
+#include <iterator>
+#include <limits>
 
 #include "db/blob/blob_file_builder.h"
+#include "db/blob/blob_index.h"
 #include "db/snapshot_checker.h"
 #include "port/likely.h"
 #include "rocksdb/listener.h"
@@ -44,16 +46,17 @@ CompactionIterator::CompactionIterator(
     const std::atomic<bool>* shutting_down,
     const SequenceNumber preserve_deletes_seqnum,
     const std::atomic<int>* manual_compaction_paused,
-    const std::shared_ptr<Logger> info_log)
+    const std::shared_ptr<Logger> info_log,
+    const std::string* full_history_ts_low)
     : 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,
           blob_file_builder, allow_data_in_errors,
           std::unique_ptr<CompactionProxy>(
-              compaction ? new CompactionProxy(compaction) : nullptr),
+              compaction ? new RealCompaction(compaction) : nullptr),
           compaction_filter, shutting_down, preserve_deletes_seqnum,
-          manual_compaction_paused, info_log) {}
+          manual_compaction_paused, info_log, full_history_ts_low) {}
 
 CompactionIterator::CompactionIterator(
     InternalIterator* input, const Comparator* cmp, MergeHelper* merge_helper,
@@ -68,7 +71,8 @@ CompactionIterator::CompactionIterator(
     const std::atomic<bool>* shutting_down,
     const SequenceNumber preserve_deletes_seqnum,
     const std::atomic<int>* manual_compaction_paused,
-    const std::shared_ptr<Logger> info_log)
+    const std::shared_ptr<Logger> info_log,
+    const std::string* full_history_ts_low)
     : input_(input),
       cmp_(cmp),
       merge_helper_(merge_helper),
@@ -76,6 +80,7 @@ CompactionIterator::CompactionIterator(
       earliest_write_conflict_snapshot_(earliest_write_conflict_snapshot),
       snapshot_checker_(snapshot_checker),
       env_(env),
+      clock_(env_->GetSystemClock().get()),
       report_detailed_time_(report_detailed_time),
       expect_valid_internal_key_(expect_valid_internal_key),
       range_del_agg_(range_del_agg),
@@ -85,12 +90,17 @@ CompactionIterator::CompactionIterator(
       shutting_down_(shutting_down),
       manual_compaction_paused_(manual_compaction_paused),
       preserve_deletes_seqnum_(preserve_deletes_seqnum),
+      info_log_(info_log),
+      allow_data_in_errors_(allow_data_in_errors),
+      timestamp_size_(cmp_ ? cmp_->timestamp_size() : 0),
+      full_history_ts_low_(full_history_ts_low),
       current_user_key_sequence_(0),
       current_user_key_snapshot_(0),
       merge_out_iter_(merge_helper_),
+      blob_garbage_collection_cutoff_file_number_(
+          ComputeBlobGarbageCollectionCutoffFileNumber(compaction_.get())),
       current_key_committed_(false),
-      info_log_(info_log),
-      allow_data_in_errors_(allow_data_in_errors) {
+      cmp_with_history_ts_low_(0) {
   assert(compaction_filter_ == nullptr || compaction_ != nullptr);
   assert(snapshots_ != nullptr);
   bottommost_level_ = compaction_ == nullptr
@@ -117,13 +127,15 @@ CompactionIterator::CompactionIterator(
   for (size_t i = 1; i < snapshots_->size(); ++i) {
     assert(snapshots_->at(i - 1) < snapshots_->at(i));
   }
+  assert(timestamp_size_ == 0 || !full_history_ts_low_ ||
+         timestamp_size_ == full_history_ts_low_->size());
 #endif
   input_->SetPinnedItersMgr(&pinned_iters_mgr_);
   TEST_SYNC_POINT_CALLBACK("CompactionIterator:AfterInit", compaction_.get());
 }
 
 CompactionIterator::~CompactionIterator() {
-  // input_ Iteartor lifetime is longer than pinned_iters_mgr_ lifetime
+  // input_ Iterator lifetime is longer than pinned_iters_mgr_ lifetime
   input_->SetPinnedItersMgr(nullptr);
 }
 
@@ -151,13 +163,13 @@ void CompactionIterator::Next() {
     if (merge_out_iter_.Valid()) {
       key_ = merge_out_iter_.key();
       value_ = merge_out_iter_.value();
-      Status s = ParseInternalKey(key_, &ikey_);
+      Status s = ParseInternalKey(key_, &ikey_, allow_data_in_errors_);
       // MergeUntil stops when it encounters a corrupt key and does not
       // include them in the result, so we expect the keys here to be valid.
       assert(s.ok());
       if (!s.ok()) {
-        ROCKS_LOG_FATAL(info_log_, "Invalid key (%s) in compaction",
-                        key_.ToString(true).c_str());
+        ROCKS_LOG_FATAL(info_log_, "Invalid key in compaction. %s",
+                        s.getState());
       }
 
       // Keep current_key_ in sync.
@@ -192,72 +204,146 @@ void CompactionIterator::Next() {
 
 bool CompactionIterator::InvokeFilterIfNeeded(bool* need_skip,
                                               Slice* skip_until) {
-  if (compaction_filter_ != nullptr &&
-      (ikey_.type == kTypeValue || ikey_.type == kTypeBlobIndex)) {
-    // 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,
-    // replace the entry with a deletion marker.
-    CompactionFilter::Decision filter;
-    compaction_filter_value_.clear();
-    compaction_filter_skip_until_.Clear();
-    CompactionFilter::ValueType value_type =
-        ikey_.type == kTypeValue ? CompactionFilter::ValueType::kValue
-                                 : CompactionFilter::ValueType::kBlobIndex;
-    // Hack: pass internal key to BlobIndexCompactionFilter since it needs
-    // to get sequence number.
-    Slice& filter_key = ikey_.type == kTypeValue ? ikey_.user_key : key_;
-    {
-      StopWatchNano timer(env_, report_detailed_time_);
+  if (!compaction_filter_ ||
+      (ikey_.type != kTypeValue && ikey_.type != kTypeBlobIndex)) {
+    return true;
+  }
+  bool error = false;
+  // 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,
+  // replace the entry with a deletion marker.
+  CompactionFilter::Decision filter = CompactionFilter::Decision::kUndetermined;
+  compaction_filter_value_.clear();
+  compaction_filter_skip_until_.Clear();
+  CompactionFilter::ValueType value_type =
+      ikey_.type == kTypeValue ? CompactionFilter::ValueType::kValue
+                               : CompactionFilter::ValueType::kBlobIndex;
+  // Hack: pass internal key to BlobIndexCompactionFilter since it needs
+  // to get sequence number.
+  assert(compaction_filter_);
+  Slice& filter_key =
+      (ikey_.type == kTypeValue ||
+       !compaction_filter_->IsStackedBlobDbInternalCompactionFilter())
+          ? ikey_.user_key
+          : key_;
+  {
+    StopWatchNano timer(clock_, report_detailed_time_);
+    if (kTypeBlobIndex == ikey_.type) {
+      blob_value_.Reset();
+      filter = compaction_filter_->FilterBlobByKey(
+          compaction_->level(), filter_key, &compaction_filter_value_,
+          compaction_filter_skip_until_.rep());
+      if (CompactionFilter::Decision::kUndetermined == filter &&
+          !compaction_filter_->IsStackedBlobDbInternalCompactionFilter()) {
+        // For integrated BlobDB impl, CompactionIterator reads blob value.
+        // For Stacked BlobDB impl, the corresponding CompactionFilter's
+        // FilterV2 method should read the blob value.
+        BlobIndex blob_index;
+        Status s = blob_index.DecodeFrom(value_);
+        if (!s.ok()) {
+          status_ = s;
+          valid_ = false;
+          return false;
+        }
+        if (blob_index.HasTTL() || blob_index.IsInlined()) {
+          status_ = Status::Corruption("Unexpected TTL/inlined blob index");
+          valid_ = false;
+          return false;
+        }
+        const Version* const version = compaction_->input_version();
+        assert(version);
+
+        uint64_t bytes_read = 0;
+        s = version->GetBlob(ReadOptions(), ikey_.user_key, blob_index,
+                             &blob_value_, &bytes_read);
+        if (!s.ok()) {
+          status_ = s;
+          valid_ = false;
+          return false;
+        }
+
+        ++iter_stats_.num_blobs_read;
+        iter_stats_.total_blob_bytes_read += bytes_read;
+
+        value_type = CompactionFilter::ValueType::kValue;
+      }
+    }
+    if (CompactionFilter::Decision::kUndetermined == filter) {
       filter = compaction_filter_->FilterV2(
-          compaction_->level(), filter_key, value_type, value_,
-          &compaction_filter_value_, compaction_filter_skip_until_.rep());
-      iter_stats_.total_filter_time +=
-          env_ != nullptr && report_detailed_time_ ? timer.ElapsedNanos() : 0;
+          compaction_->level(), filter_key, value_type,
+          blob_value_.empty() ? value_ : blob_value_, &compaction_filter_value_,
+          compaction_filter_skip_until_.rep());
     }
+    iter_stats_.total_filter_time +=
+        env_ != nullptr && report_detailed_time_ ? timer.ElapsedNanos() : 0;
+  }
 
-    if (filter == CompactionFilter::Decision::kRemoveAndSkipUntil &&
-        cmp_->Compare(*compaction_filter_skip_until_.rep(), ikey_.user_key) <=
-            0) {
-      // Can't skip to a key smaller than the current one.
-      // Keep the key as per FilterV2 documentation.
-      filter = CompactionFilter::Decision::kKeep;
+  if (CompactionFilter::Decision::kUndetermined == filter) {
+    // Should not reach here, since FilterV2 should never return kUndetermined.
+    status_ =
+        Status::NotSupported("FilterV2() should never return kUndetermined");
+    valid_ = false;
+    return false;
+  }
+
+  if (filter == CompactionFilter::Decision::kRemoveAndSkipUntil &&
+      cmp_->Compare(*compaction_filter_skip_until_.rep(), ikey_.user_key) <=
+          0) {
+    // Can't skip to a key smaller than the current one.
+    // Keep the key as per FilterV2 documentation.
+    filter = CompactionFilter::Decision::kKeep;
+  }
+
+  if (filter == CompactionFilter::Decision::kRemove) {
+    // convert the current key to a delete; key_ is pointing into
+    // current_key_ at this point, so updating current_key_ updates key()
+    ikey_.type = kTypeDeletion;
+    current_key_.UpdateInternalKey(ikey_.sequence, kTypeDeletion);
+    // no value associated with delete
+    value_.clear();
+    iter_stats_.num_record_drop_user++;
+  } else if (filter == CompactionFilter::Decision::kChangeValue) {
+    if (ikey_.type == kTypeBlobIndex) {
+      // value transfer from blob file to inlined data
+      ikey_.type = kTypeValue;
+      current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
     }
-
-    if (filter == CompactionFilter::Decision::kRemove) {
-      // convert the current key to a delete; key_ is pointing into
-      // current_key_ at this point, so updating current_key_ updates key()
-      ikey_.type = kTypeDeletion;
-      current_key_.UpdateInternalKey(ikey_.sequence, kTypeDeletion);
-      // no value associated with delete
-      value_.clear();
-      iter_stats_.num_record_drop_user++;
-    } else if (filter == CompactionFilter::Decision::kChangeValue) {
-      if (ikey_.type == kTypeBlobIndex) {
-        // value transfer from blob file to inlined data
-        ikey_.type = kTypeValue;
-        current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
-      }
-      value_ = compaction_filter_value_;
-    } else if (filter == CompactionFilter::Decision::kRemoveAndSkipUntil) {
-      *need_skip = true;
-      compaction_filter_skip_until_.ConvertFromUserKey(kMaxSequenceNumber,
-                                                       kValueTypeForSeek);
-      *skip_until = compaction_filter_skip_until_.Encode();
-    } else if (filter == CompactionFilter::Decision::kChangeBlobIndex) {
-      if (ikey_.type == kTypeValue) {
-        // value transfer from inlined data to blob file
-        ikey_.type = kTypeBlobIndex;
-        current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
-      }
-      value_ = compaction_filter_value_;
-    } else if (filter == CompactionFilter::Decision::kIOError) {
-      status_ =
-          Status::IOError("Failed to access blob during compaction filter");
+    value_ = compaction_filter_value_;
+  } else if (filter == CompactionFilter::Decision::kRemoveAndSkipUntil) {
+    *need_skip = true;
+    compaction_filter_skip_until_.ConvertFromUserKey(kMaxSequenceNumber,
+                                                     kValueTypeForSeek);
+    *skip_until = compaction_filter_skip_until_.Encode();
+  } else if (filter == CompactionFilter::Decision::kChangeBlobIndex) {
+    // Only the StackableDB-based BlobDB impl's compaction filter should return
+    // kChangeBlobIndex. Decision about rewriting blob and changing blob index
+    // in the integrated BlobDB impl is made in subsequent call to
+    // PrepareOutput() and its callees.
+    if (!compaction_filter_->IsStackedBlobDbInternalCompactionFilter()) {
+      status_ = Status::NotSupported(
+          "Only stacked BlobDB's internal compaction filter can return "
+          "kChangeBlobIndex.");
+      valid_ = false;
       return false;
     }
+    if (ikey_.type == kTypeValue) {
+      // value transfer from inlined data to blob file
+      ikey_.type = kTypeBlobIndex;
+      current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
+    }
+    value_ = compaction_filter_value_;
+  } else if (filter == CompactionFilter::Decision::kIOError) {
+    if (!compaction_filter_->IsStackedBlobDbInternalCompactionFilter()) {
+      status_ = Status::NotSupported(
+          "CompactionFilter for integrated BlobDB should not return kIOError");
+      valid_ = false;
+      return false;
+    }
+    status_ = Status::IOError("Failed to access blob during compaction filter");
+    error = true;
   }
-  return true;
+  return !error;
 }
 
 void CompactionIterator::NextFromInput() {
@@ -270,22 +356,14 @@ void CompactionIterator::NextFromInput() {
     value_ = input_->value();
     iter_stats_.num_input_records++;
 
-    Status pikStatus = ParseInternalKey(key_, &ikey_);
-    if (!pikStatus.ok()) {
+    Status pik_status = ParseInternalKey(key_, &ikey_, allow_data_in_errors_);
+    if (!pik_status.ok()) {
       iter_stats_.num_input_corrupt_records++;
 
       // If `expect_valid_internal_key_` is false, return the corrupted key
       // and let the caller decide what to do with it.
-      // TODO(noetzli): We should have a more elegant solution for this.
       if (expect_valid_internal_key_) {
-        std::string msg("Corrupted internal key not expected.");
-        if (allow_data_in_errors_) {
-          msg.append(" Corrupt key: " + ikey_.user_key.ToString(/*hex=*/true) +
-                     ". ");
-          msg.append("key type: " + std::to_string(ikey_.type) + ".");
-          msg.append("seq: " + std::to_string(ikey_.sequence) + ".");
-        }
-        status_ = Status::Corruption(msg.c_str());
+        status_ = pik_status;
         return;
       }
       key_ = current_key_.SetInternalKey(key_);
@@ -298,7 +376,8 @@ void CompactionIterator::NextFromInput() {
     TEST_SYNC_POINT_CALLBACK("CompactionIterator:ProcessKV", &ikey_);
 
     // Update input statistics
-    if (ikey_.type == kTypeDeletion || ikey_.type == kTypeSingleDeletion) {
+    if (ikey_.type == kTypeDeletion || ikey_.type == kTypeSingleDeletion ||
+        ikey_.type == kTypeDeletionWithTimestamp) {
       iter_stats_.num_input_deletion_records++;
     }
     iter_stats_.total_input_raw_key_bytes += key_.size();
@@ -311,19 +390,54 @@ void CompactionIterator::NextFromInput() {
     // merge_helper_->compaction_filter_skip_until_.
     Slice skip_until;
 
+    bool user_key_equal_without_ts = false;
+    int cmp_ts = 0;
+    if (has_current_user_key_) {
+      user_key_equal_without_ts =
+          cmp_->EqualWithoutTimestamp(ikey_.user_key, current_user_key_);
+      // if timestamp_size_ > 0, then curr_ts_ has been initialized by a
+      // previous key.
+      cmp_ts = timestamp_size_ ? cmp_->CompareTimestamp(
+                                     ExtractTimestampFromUserKey(
+                                         ikey_.user_key, timestamp_size_),
+                                     curr_ts_)
+                               : 0;
+    }
+
     // Check whether the user key changed. After this if statement current_key_
     // is a copy of the current input key (maybe converted to a delete by the
     // compaction filter). ikey_.user_key is pointing to the copy.
-    if (!has_current_user_key_ ||
-        !cmp_->Equal(ikey_.user_key, current_user_key_)) {
+    if (!has_current_user_key_ || !user_key_equal_without_ts || cmp_ts != 0) {
       // First occurrence of this user key
       // Copy key for output
       key_ = current_key_.SetInternalKey(key_, &ikey_);
+
+      // If timestamp_size_ > 0, then copy from ikey_ to curr_ts_ for the use
+      // in next iteration to compare with the timestamp of next key.
+      UpdateTimestampAndCompareWithFullHistoryLow();
+
+      // If
+      // (1) !has_current_user_key_, OR
+      // (2) timestamp is disabled, OR
+      // (3) all history will be preserved, OR
+      // (4) user key (excluding timestamp) is different from previous key, OR
+      // (5) timestamp is NO older than *full_history_ts_low_
+      // then current_user_key_ must be treated as a different user key.
+      // This means, if a user key (excluding ts) is the same as the previous
+      // user key, and its ts is older than *full_history_ts_low_, then we
+      // consider this key for GC, e.g. it may be dropped if certain conditions
+      // match.
+      if (!has_current_user_key_ || !timestamp_size_ || !full_history_ts_low_ ||
+          !user_key_equal_without_ts || cmp_with_history_ts_low_ >= 0) {
+        // Initialize for future comparison for rule (A) and etc.
+        current_user_key_sequence_ = kMaxSequenceNumber;
+        current_user_key_snapshot_ = 0;
+        has_current_user_key_ = true;
+      }
       current_user_key_ = ikey_.user_key;
-      has_current_user_key_ = true;
+
       has_outputted_key_ = false;
-      current_user_key_sequence_ = kMaxSequenceNumber;
-      current_user_key_snapshot_ = 0;
+
       current_key_committed_ = KeyCommitted(ikey_.sequence);
 
       // Apply the compaction filter to the first committed version of the user
@@ -381,8 +495,8 @@ void CompactionIterator::NextFromInput() {
       // In the previous iteration we encountered a single delete that we could
       // not compact out.  We will keep this Put, but can drop it's data.
       // (See Optimization 3, below.)
-      assert(ikey_.type == kTypeValue);
-      if (ikey_.type != kTypeValue) {
+      assert(ikey_.type == kTypeValue || ikey_.type == kTypeBlobIndex);
+      if (ikey_.type != kTypeValue && ikey_.type != kTypeBlobIndex) {
         ROCKS_LOG_FATAL(info_log_,
                         "Unexpected key type %d for compaction output",
                         ikey_.type);
@@ -395,6 +509,11 @@ void CompactionIterator::NextFromInput() {
                         current_user_key_snapshot_, last_snapshot);
       }
 
+      if (ikey_.type == kTypeBlobIndex) {
+        ikey_.type = kTypeValue;
+        current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
+      }
+
       value_.clear();
       valid_ = true;
       clear_and_output_next_key_ = false;
@@ -439,7 +558,8 @@ void CompactionIterator::NextFromInput() {
       // Check whether the next key exists, is not corrupt, and is the same key
       // as the single delete.
       if (input_->Valid() &&
-          ParseInternalKey(input_->key(), &next_ikey) == Status::OK() &&
+          ParseInternalKey(input_->key(), &next_ikey, allow_data_in_errors_)
+              .ok() &&
           cmp_->Equal(ikey_.user_key, next_ikey.user_key)) {
         // Check whether the next key belongs to the same snapshot as the
         // SingleDelete.
@@ -543,9 +663,12 @@ void CompactionIterator::NextFromInput() {
                         last_sequence, current_user_key_sequence_);
       }
 
-      ++iter_stats_.num_record_drop_hidden;  // (A)
+      ++iter_stats_.num_record_drop_hidden;  // rule (A)
       input_->Next();
-    } else if (compaction_ != nullptr && ikey_.type == kTypeDeletion &&
+    } else if (compaction_ != nullptr &&
+               (ikey_.type == kTypeDeletion ||
+                (ikey_.type == kTypeDeletionWithTimestamp &&
+                 cmp_with_history_ts_low_ < 0)) &&
                IN_EARLIEST_SNAPSHOT(ikey_.sequence) &&
                ikeyNotNeededForIncrementalSnapshot() &&
                compaction_->KeyNotExistsBeyondOutputLevel(ikey_.user_key,
@@ -569,13 +692,19 @@ void CompactionIterator::NextFromInput() {
       // given that:
       // (1) The deletion is earlier than earliest_write_conflict_snapshot, and
       // (2) No value exist earlier than the deletion.
+      //
+      // Note also that a deletion marker of type kTypeDeletionWithTimestamp
+      // will be treated as a different user key unless the timestamp is older
+      // than *full_history_ts_low_.
       ++iter_stats_.num_record_drop_obsolete;
       if (!bottommost_level_) {
         ++iter_stats_.num_optimized_del_drop_obsolete;
       }
       input_->Next();
-    } else if ((ikey_.type == kTypeDeletion) && bottommost_level_ &&
-               ikeyNotNeededForIncrementalSnapshot()) {
+    } else if ((ikey_.type == kTypeDeletion ||
+                (ikey_.type == kTypeDeletionWithTimestamp &&
+                 cmp_with_history_ts_low_ < 0)) &&
+               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
@@ -583,12 +712,17 @@ void CompactionIterator::NextFromInput() {
                                  ikey_.user_key, &level_ptrs_));
       ParsedInternalKey next_ikey;
       input_->Next();
-      // Skip over all versions of this key that happen to occur in the same snapshot
-      // range as the delete
+      // Skip over all versions of this key that happen to occur in the same
+      // snapshot range as the delete.
+      //
+      // Note that a deletion marker of type kTypeDeletionWithTimestamp will be
+      // considered to have a different user key unless the timestamp is older
+      // than *full_history_ts_low_.
       while (!IsPausingManualCompaction() && !IsShuttingDown() &&
              input_->Valid() &&
-             (ParseInternalKey(input_->key(), &next_ikey) == Status::OK()) &&
-             cmp_->Equal(ikey_.user_key, next_ikey.user_key) &&
+             (ParseInternalKey(input_->key(), &next_ikey, allow_data_in_errors_)
+                  .ok()) &&
+             cmp_->EqualWithoutTimestamp(ikey_.user_key, next_ikey.user_key) &&
              (prev_snapshot == 0 ||
               DEFINITELY_NOT_IN_SNAPSHOT(next_ikey.sequence, prev_snapshot))) {
         input_->Next();
@@ -596,8 +730,9 @@ void CompactionIterator::NextFromInput() {
       // 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) == Status::OK()) &&
-          cmp_->Equal(ikey_.user_key, next_ikey.user_key)) {
+          (ParseInternalKey(input_->key(), &next_ikey, allow_data_in_errors_)
+               .ok()) &&
+          cmp_->EqualWithoutTimestamp(ikey_.user_key, next_ikey.user_key)) {
         valid_ = true;
         at_next_ = true;
       }
@@ -613,8 +748,9 @@ void CompactionIterator::NextFromInput() {
       // have hit (A)
       // We encapsulate the merge related state machine in a different
       // object to minimize change to the existing flow.
-      Status s = merge_helper_->MergeUntil(input_, range_del_agg_,
-                                           prev_snapshot, bottommost_level_);
+      Status s =
+          merge_helper_->MergeUntil(input_, range_del_agg_, prev_snapshot,
+                                    bottommost_level_, allow_data_in_errors_);
       merge_out_iter_.SeekToFirst();
 
       if (!s.ok() && !s.IsMergeInProgress()) {
@@ -625,13 +761,13 @@ void CompactionIterator::NextFromInput() {
         //       These will be correctly set below.
         key_ = merge_out_iter_.key();
         value_ = merge_out_iter_.value();
-        pikStatus = ParseInternalKey(key_, &ikey_);
+        pik_status = ParseInternalKey(key_, &ikey_, allow_data_in_errors_);
         // MergeUntil stops when it encounters a corrupt key and does not
         // include them in the result, so we expect the keys here to valid.
-        assert(pikStatus.ok());
-        if (!pikStatus.ok()) {
-          ROCKS_LOG_FATAL(info_log_, "Invalid key (%s) in compaction",
-                          key_.ToString(true).c_str());
+        assert(pik_status.ok());
+        if (!pik_status.ok()) {
+          ROCKS_LOG_FATAL(info_log_, "Invalid key in compaction. %s",
+                          pik_status.getState());
         }
         // Keep current_key_ in sync.
         current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
@@ -677,40 +813,142 @@ void CompactionIterator::NextFromInput() {
   }
 }
 
+bool CompactionIterator::ExtractLargeValueIfNeededImpl() {
+  if (!blob_file_builder_) {
+    return false;
+  }
+
+  blob_index_.clear();
+  const Status s = blob_file_builder_->Add(user_key(), value_, &blob_index_);
+
+  if (!s.ok()) {
+    status_ = s;
+    valid_ = false;
+
+    return false;
+  }
+
+  if (blob_index_.empty()) {
+    return false;
+  }
+
+  value_ = blob_index_;
+
+  return true;
+}
+
+void CompactionIterator::ExtractLargeValueIfNeeded() {
+  assert(ikey_.type == kTypeValue);
+
+  if (!ExtractLargeValueIfNeededImpl()) {
+    return;
+  }
+
+  ikey_.type = kTypeBlobIndex;
+  current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
+}
+
+void CompactionIterator::GarbageCollectBlobIfNeeded() {
+  assert(ikey_.type == kTypeBlobIndex);
+
+  if (!compaction_) {
+    return;
+  }
+
+  // GC for integrated BlobDB
+  if (compaction_->enable_blob_garbage_collection()) {
+    BlobIndex blob_index;
+
+    {
+      const Status s = blob_index.DecodeFrom(value_);
+
+      if (!s.ok()) {
+        status_ = s;
+        valid_ = false;
+
+        return;
+      }
+    }
+
+    if (blob_index.IsInlined() || blob_index.HasTTL()) {
+      status_ = Status::Corruption("Unexpected TTL/inlined blob index");
+      valid_ = false;
+
+      return;
+    }
+
+    if (blob_index.file_number() >=
+        blob_garbage_collection_cutoff_file_number_) {
+      return;
+    }
+
+    const Version* const version = compaction_->input_version();
+    assert(version);
+
+    uint64_t bytes_read = 0;
+
+    {
+      const Status s = version->GetBlob(ReadOptions(), user_key(), blob_index,
+                                        &blob_value_, &bytes_read);
+
+      if (!s.ok()) {
+        status_ = s;
+        valid_ = false;
+
+        return;
+      }
+    }
+
+    ++iter_stats_.num_blobs_read;
+    iter_stats_.total_blob_bytes_read += bytes_read;
+
+    value_ = blob_value_;
+
+    if (ExtractLargeValueIfNeededImpl()) {
+      return;
+    }
+
+    ikey_.type = kTypeValue;
+    current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
+
+    return;
+  }
+
+  // GC for stacked BlobDB
+  if (compaction_filter_ &&
+      compaction_filter_->IsStackedBlobDbInternalCompactionFilter()) {
+    const auto blob_decision = compaction_filter_->PrepareBlobOutput(
+        user_key(), value_, &compaction_filter_value_);
+
+    if (blob_decision == CompactionFilter::BlobDecision::kCorruption) {
+      status_ =
+          Status::Corruption("Corrupted blob reference encountered during GC");
+      valid_ = false;
+
+      return;
+    }
+
+    if (blob_decision == CompactionFilter::BlobDecision::kIOError) {
+      status_ = Status::IOError("Could not relocate blob during GC");
+      valid_ = false;
+
+      return;
+    }
+
+    if (blob_decision == CompactionFilter::BlobDecision::kChangeValue) {
+      value_ = compaction_filter_value_;
+
+      return;
+    }
+  }
+}
+
 void CompactionIterator::PrepareOutput() {
   if (valid_) {
     if (ikey_.type == kTypeValue) {
-      if (blob_file_builder_) {
-        blob_index_.clear();
-        const Status s =
-            blob_file_builder_->Add(user_key(), value_, &blob_index_);
-
-        if (!s.ok()) {
-          status_ = s;
-          valid_ = false;
-        } else if (!blob_index_.empty()) {
-          value_ = blob_index_;
-          ikey_.type = kTypeBlobIndex;
-          current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
-        }
-      }
+      ExtractLargeValueIfNeeded();
     } else if (ikey_.type == kTypeBlobIndex) {
-      if (compaction_filter_) {
-        const auto blob_decision = compaction_filter_->PrepareBlobOutput(
-            user_key(), value_, &compaction_filter_value_);
-
-        if (blob_decision == CompactionFilter::BlobDecision::kCorruption) {
-          status_ = Status::Corruption(
-              "Corrupted blob reference encountered during GC");
-          valid_ = false;
-        } else if (blob_decision == CompactionFilter::BlobDecision::kIOError) {
-          status_ = Status::IOError("Could not relocate blob during GC");
-          valid_ = false;
-        } else if (blob_decision ==
-                   CompactionFilter::BlobDecision::kChangeValue) {
-          value_ = compaction_filter_value_;
-        }
-      }
+      GarbageCollectBlobIfNeeded();
     }
 
     // Zeroing out the sequence number leads to better compression.
@@ -735,7 +973,18 @@ void CompactionIterator::PrepareOutput() {
                         ikey_.type);
       }
       ikey_.sequence = 0;
-      current_key_.UpdateInternalKey(0, ikey_.type);
+      if (!timestamp_size_) {
+        current_key_.UpdateInternalKey(0, ikey_.type);
+      } else if (full_history_ts_low_ && cmp_with_history_ts_low_ < 0) {
+        // We can also zero out timestamp for better compression.
+        // For the same user key (excluding timestamp), the timestamp-based
+        // history can be collapsed to save some space if the timestamp is
+        // older than *full_history_ts_low_.
+        const std::string kTsMin(timestamp_size_, static_cast<char>(0));
+        const Slice ts_slice = kTsMin;
+        ikey_.SetTimestamp(ts_slice);
+        current_key_.UpdateInternalKey(0, ikey_.type, &ts_slice);
+      }
     }
   }
 }
@@ -827,4 +1076,30 @@ bool CompactionIterator::IsInEarliestSnapshot(SequenceNumber sequence) {
   return in_snapshot == SnapshotCheckerResult::kInSnapshot;
 }
 
+uint64_t CompactionIterator::ComputeBlobGarbageCollectionCutoffFileNumber(
+    const CompactionProxy* compaction) {
+  if (!compaction) {
+    return 0;
+  }
+
+  if (!compaction->enable_blob_garbage_collection()) {
+    return 0;
+  }
+
+  Version* const version = compaction->input_version();
+  assert(version);
+
+  const VersionStorageInfo* const storage_info = version->storage_info();
+  assert(storage_info);
+
+  const auto& blob_files = storage_info->GetBlobFiles();
+
+  auto it = blob_files.begin();
+  std::advance(
+      it, compaction->blob_garbage_collection_age_cutoff() * blob_files.size());
+
+  return it != blob_files.end() ? it->first
+                                : std::numeric_limits<uint64_t>::max();
+}
+
 }  // namespace ROCKSDB_NAMESPACE

db/compaction/compaction_iterator.h

@@ -5,6 +5,7 @@
 #pragma once
 
 #include <algorithm>
+#include <cinttypes>
 #include <deque>
 #include <string>
 #include <unordered_set>
@@ -29,33 +30,76 @@ class CompactionIterator {
   // CompactionIterator uses. Tests can override it.
   class CompactionProxy {
    public:
-    explicit CompactionProxy(const Compaction* compaction)
-        : compaction_(compaction) {}
-
     virtual ~CompactionProxy() = default;
-    virtual int level(size_t /*compaction_input_level*/ = 0) const {
-      return compaction_->level();
-    }
+
+    virtual int level() const = 0;
+
     virtual bool KeyNotExistsBeyondOutputLevel(
-        const Slice& user_key, std::vector<size_t>* level_ptrs) const {
+        const Slice& user_key, std::vector<size_t>* level_ptrs) const = 0;
+
+    virtual bool bottommost_level() const = 0;
+
+    virtual int number_levels() const = 0;
+
+    virtual Slice GetLargestUserKey() const = 0;
+
+    virtual bool allow_ingest_behind() const = 0;
+
+    virtual bool preserve_deletes() const = 0;
+
+    virtual bool enable_blob_garbage_collection() const = 0;
+
+    virtual double blob_garbage_collection_age_cutoff() const = 0;
+
+    virtual Version* input_version() const = 0;
+  };
+
+  class RealCompaction : public CompactionProxy {
+   public:
+    explicit RealCompaction(const Compaction* compaction)
+        : compaction_(compaction) {
+      assert(compaction_);
+      assert(compaction_->immutable_cf_options());
+      assert(compaction_->mutable_cf_options());
+    }
+
+    int level() const override { return compaction_->level(); }
+
+    bool KeyNotExistsBeyondOutputLevel(
+        const Slice& user_key, std::vector<size_t>* level_ptrs) const override {
       return compaction_->KeyNotExistsBeyondOutputLevel(user_key, level_ptrs);
     }
-    virtual bool bottommost_level() const {
+
+    bool bottommost_level() const override {
       return compaction_->bottommost_level();
     }
-    virtual int number_levels() const { return compaction_->number_levels(); }
-    virtual Slice GetLargestUserKey() const {
+
+    int number_levels() const override { return compaction_->number_levels(); }
+
+    Slice GetLargestUserKey() const override {
       return compaction_->GetLargestUserKey();
     }
-    virtual bool allow_ingest_behind() const {
+
+    bool allow_ingest_behind() const override {
       return compaction_->immutable_cf_options()->allow_ingest_behind;
     }
-    virtual bool preserve_deletes() const {
+
+    bool preserve_deletes() const override {
       return compaction_->immutable_cf_options()->preserve_deletes;
     }
 
-   protected:
-    CompactionProxy() = default;
+    bool enable_blob_garbage_collection() const override {
+      return compaction_->mutable_cf_options()->enable_blob_garbage_collection;
+    }
+
+    double blob_garbage_collection_age_cutoff() const override {
+      return compaction_->mutable_cf_options()
+          ->blob_garbage_collection_age_cutoff;
+    }
+
+    Version* input_version() const override {
+      return compaction_->input_version();
+    }
 
    private:
     const Compaction* compaction_;
@@ -75,7 +119,8 @@ class CompactionIterator {
                      const std::atomic<bool>* shutting_down = nullptr,
                      const SequenceNumber preserve_deletes_seqnum = 0,
                      const std::atomic<int>* manual_compaction_paused = nullptr,
-                     const std::shared_ptr<Logger> info_log = nullptr);
+                     const std::shared_ptr<Logger> info_log = nullptr,
+                     const std::string* full_history_ts_low = nullptr);
 
   // Constructor with custom CompactionProxy, used for tests.
   CompactionIterator(InternalIterator* input, const Comparator* cmp,
@@ -92,7 +137,8 @@ class CompactionIterator {
                      const std::atomic<bool>* shutting_down = nullptr,
                      const SequenceNumber preserve_deletes_seqnum = 0,
                      const std::atomic<int>* manual_compaction_paused = nullptr,
-                     const std::shared_ptr<Logger> info_log = nullptr);
+                     const std::shared_ptr<Logger> info_log = nullptr,
+                     const std::string* full_history_ts_low = nullptr);
 
   ~CompactionIterator();
 
@@ -121,11 +167,30 @@ class CompactionIterator {
   // Processes the input stream to find the next output
   void NextFromInput();
 
-  // Do last preparations before presenting the output to the callee. At this
-  // point this only zeroes out the sequence number if possible for better
-  // compression.
+  // Do final preparations before presenting the output to the callee.
   void PrepareOutput();
 
+  // Passes the output value to the blob file builder (if any), and replaces it
+  // with the corresponding blob reference if it has been actually written to a
+  // blob file (i.e. if it passed the value size check). Returns true if the
+  // value got extracted to a blob file, false otherwise.
+  bool ExtractLargeValueIfNeededImpl();
+
+  // Extracts large values as described above, and updates the internal key's
+  // type to kTypeBlobIndex if the value got extracted. Should only be called
+  // for regular values (kTypeValue).
+  void ExtractLargeValueIfNeeded();
+
+  // Relocates valid blobs residing in the oldest blob files if garbage
+  // collection is enabled. Relocated blobs are written to new blob files or
+  // inlined in the LSM tree depending on the current settings (i.e.
+  // enable_blob_files and min_blob_size). Should only be called for blob
+  // references (kTypeBlobIndex).
+  //
+  // Note: the stacked BlobDB implementation's compaction filter based GC
+  // algorithm is also called from here.
+  void GarbageCollectBlobIfNeeded();
+
   // Invoke compaction filter if needed.
   // Return true on success, false on failures (e.g.: kIOError).
   bool InvokeFilterIfNeeded(bool* need_skip, Slice* skip_until);
@@ -152,6 +217,23 @@ class CompactionIterator {
 
   bool IsInEarliestSnapshot(SequenceNumber sequence);
 
+  // Extract user-defined timestamp from user key if possible and compare it
+  // with *full_history_ts_low_ if applicable.
+  inline void UpdateTimestampAndCompareWithFullHistoryLow() {
+    if (!timestamp_size_) {
+      return;
+    }
+    Slice ts = ExtractTimestampFromUserKey(ikey_.user_key, timestamp_size_);
+    curr_ts_.assign(ts.data(), ts.size());
+    if (full_history_ts_low_) {
+      cmp_with_history_ts_low_ =
+          cmp_->CompareTimestamp(ts, *full_history_ts_low_);
+    }
+  }
+
+  static uint64_t ComputeBlobGarbageCollectionCutoffFileNumber(
+      const CompactionProxy* compaction);
+
   InternalIterator* input_;
   const Comparator* cmp_;
   MergeHelper* merge_helper_;
@@ -166,6 +248,7 @@ class CompactionIterator {
   const SequenceNumber earliest_write_conflict_snapshot_;
   const SnapshotChecker* const snapshot_checker_;
   Env* env_;
+  SystemClock* clock_;
   bool report_detailed_time_;
   bool expect_valid_internal_key_;
   CompactionRangeDelAggregator* range_del_agg_;
@@ -181,6 +264,20 @@ class CompactionIterator {
   SequenceNumber earliest_snapshot_;
   SequenceNumber latest_snapshot_;
 
+  std::shared_ptr<Logger> info_log_;
+
+  bool allow_data_in_errors_;
+
+  // Comes from comparator.
+  const size_t timestamp_size_;
+
+  // Lower bound timestamp to retain full history in terms of user-defined
+  // timestamp. If a key's timestamp is older than full_history_ts_low_, then
+  // the key *may* be eligible for garbage collection (GC). The skipping logic
+  // is in `NextFromInput()` and `PrepareOutput()`.
+  // If nullptr, NO GC will be performed and all history will be preserved.
+  const std::string* const full_history_ts_low_;
+
   // State
   //
   // Points to a copy of the current compaction iterator output (current_key_)
@@ -199,11 +296,13 @@ class CompactionIterator {
   // Stores whether ikey_.user_key is valid. If set to false, the user key is
   // not compared against the current key in the underlying iterator.
   bool has_current_user_key_ = false;
-  bool at_next_ = false;  // If false, the iterator
-  // Holds a copy of the current compaction iterator output (or current key in
-  // the underlying iterator during NextFromInput()).
+  // If false, the iterator holds a copy of the current compaction iterator
+  // output (or current key in the underlying iterator during NextFromInput()).
+  bool at_next_ = false;
+
   IterKey current_key_;
   Slice current_user_key_;
+  std::string curr_ts_;
   SequenceNumber current_user_key_sequence_;
   SequenceNumber current_user_key_snapshot_;
 
@@ -218,7 +317,11 @@ class CompactionIterator {
   // PinnedIteratorsManager used to pin input_ Iterator blocks while reading
   // merge operands and then releasing them after consuming them.
   PinnedIteratorsManager pinned_iters_mgr_;
+
+  uint64_t blob_garbage_collection_cutoff_file_number_;
+
   std::string blob_index_;
+  PinnableSlice blob_value_;
   std::string compaction_filter_value_;
   InternalKey compaction_filter_skip_until_;
   // "level_ptrs" holds indices that remember which file of an associated
@@ -233,9 +336,9 @@ class CompactionIterator {
   // Used to avoid purging uncommitted values. The application can specify
   // uncommitted values by providing a SnapshotChecker object.
   bool current_key_committed_;
-  std::shared_ptr<Logger> info_log_;
 
-  bool allow_data_in_errors_;
+  // Saved result of ucmp->CompareTimestamp(current_ts_, *full_history_ts_low_)
+  int cmp_with_history_ts_low_;
 
   bool IsShuttingDown() {
     // This is a best-effort facility, so memory_order_relaxed is sufficient.

db/compaction/compaction_iterator_test.cc

@@ -38,7 +38,7 @@ class NoMergingMergeOp : public MergeOperator {
 
 // Compaction filter that gets stuck when it sees a particular key,
 // then gets unstuck when told to.
-// Always returns Decition::kRemove.
+// Always returns Decision::kRemove.
 class StallingFilter : public CompactionFilter {
  public:
   Decision FilterV2(int /*level*/, const Slice& key, ValueType /*type*/,
@@ -156,23 +156,32 @@ class LoggingForwardVectorIterator : public InternalIterator {
 
 class FakeCompaction : public CompactionIterator::CompactionProxy {
  public:
-  FakeCompaction() = default;
+  int level() const override { return 0; }
 
-  int level(size_t /*compaction_input_level*/) const override { return 0; }
   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 {
     return "\xff\xff\xff\xff\xff\xff\xff\xff\xff";
   }
+
   bool allow_ingest_behind() const override { return is_allow_ingest_behind; }
 
   bool preserve_deletes() const override { return false; }
 
+  bool enable_blob_garbage_collection() const override { return false; }
+
+  double blob_garbage_collection_age_cutoff() const override { return 0.0; }
+
+  Version* input_version() const override { return nullptr; }
+
   bool key_not_exists_beyond_output_level = false;
 
   bool is_bottommost_level = false;
@@ -180,7 +189,7 @@ class FakeCompaction : public CompactionIterator::CompactionProxy {
   bool is_allow_ingest_behind = false;
 };
 
-// A simplifed snapshot checker which assumes each snapshot has a global
+// A simplified snapshot checker which assumes each snapshot has a global
 // last visible sequence.
 class TestSnapshotChecker : public SnapshotChecker {
  public:
@@ -216,6 +225,9 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
   CompactionIteratorTest()
       : cmp_(BytewiseComparator()), icmp_(cmp_), snapshots_({}) {}
 
+  explicit CompactionIteratorTest(const Comparator* ucmp)
+      : cmp_(ucmp), icmp_(cmp_), snapshots_({}) {}
+
   void InitIterators(
       const std::vector<std::string>& ks, const std::vector<std::string>& vs,
       const std::vector<std::string>& range_del_ks,
@@ -224,7 +236,9 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
       SequenceNumber last_committed_sequence = kMaxSequenceNumber,
       MergeOperator* merge_op = nullptr, CompactionFilter* filter = nullptr,
       bool bottommost_level = false,
-      SequenceNumber earliest_write_conflict_snapshot = kMaxSequenceNumber) {
+      SequenceNumber earliest_write_conflict_snapshot = kMaxSequenceNumber,
+      bool key_not_exists_beyond_output_level = false,
+      const std::string* full_history_ts_low = nullptr) {
     std::unique_ptr<InternalIterator> unfragmented_range_del_iter(
         new test::VectorIterator(range_del_ks, range_del_vs));
     auto tombstone_list = std::make_shared<FragmentedRangeTombstoneList>(
@@ -236,10 +250,12 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
     range_del_agg_->AddTombstones(std::move(range_del_iter));
 
     std::unique_ptr<CompactionIterator::CompactionProxy> compaction;
-    if (filter || bottommost_level) {
+    if (filter || bottommost_level || key_not_exists_beyond_output_level) {
       compaction_proxy_ = new FakeCompaction();
       compaction_proxy_->is_bottommost_level = bottommost_level;
       compaction_proxy_->is_allow_ingest_behind = AllowIngestBehind();
+      compaction_proxy_->key_not_exists_beyond_output_level =
+          key_not_exists_beyond_output_level;
       compaction.reset(compaction_proxy_);
     }
     bool use_snapshot_checker = UseSnapshotChecker() || GetParam();
@@ -252,6 +268,11 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
                         0 /*latest_snapshot*/, snapshot_checker_.get(),
                         0 /*level*/, nullptr /*statistics*/, &shutting_down_));
 
+    if (c_iter_) {
+      // Since iter_ is still used in ~CompactionIterator(), we call
+      // ~CompactionIterator() first.
+      c_iter_.reset();
+    }
     iter_.reset(new LoggingForwardVectorIterator(ks, vs));
     iter_->SeekToFirst();
     c_iter_.reset(new CompactionIterator(
@@ -259,8 +280,10 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
         earliest_write_conflict_snapshot, snapshot_checker_.get(),
         Env::Default(), false /* report_detailed_time */, false,
         range_del_agg_.get(), nullptr /* blob_file_builder */,
-        false /*allow_data_in_errors*/, std::move(compaction), filter,
-        &shutting_down_));
+        true /*allow_data_in_errors*/, std::move(compaction), filter,
+        &shutting_down_, /*preserve_deletes_seqnum=*/0,
+        /*manual_compaction_paused=*/nullptr, /*info_log=*/nullptr,
+        full_history_ts_low));
   }
 
   void AddSnapshot(SequenceNumber snapshot,
@@ -282,10 +305,13 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
       MergeOperator* merge_operator = nullptr,
       CompactionFilter* compaction_filter = nullptr,
       bool bottommost_level = false,
-      SequenceNumber earliest_write_conflict_snapshot = kMaxSequenceNumber) {
+      SequenceNumber earliest_write_conflict_snapshot = kMaxSequenceNumber,
+      bool key_not_exists_beyond_output_level = false,
+      const std::string* full_history_ts_low = nullptr) {
     InitIterators(input_keys, input_values, {}, {}, kMaxSequenceNumber,
                   last_committed_seq, merge_operator, compaction_filter,
-                  bottommost_level, earliest_write_conflict_snapshot);
+                  bottommost_level, earliest_write_conflict_snapshot,
+                  key_not_exists_beyond_output_level, full_history_ts_low);
     c_iter_->SeekToFirst();
     for (size_t i = 0; i < expected_keys.size(); i++) {
       std::string info = "i = " + ToString(i);
@@ -299,6 +325,11 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
     ASSERT_FALSE(c_iter_->Valid());
   }
 
+  void ClearSnapshots() {
+    snapshots_.clear();
+    snapshot_map_.clear();
+  }
+
   const Comparator* cmp_;
   const InternalKeyComparator icmp_;
   std::vector<SequenceNumber> snapshots_;
@@ -680,7 +711,7 @@ TEST_P(CompactionIteratorTest, ZeroOutSequenceAtBottomLevel) {
   RunTest({test::KeyStr("a", 1, kTypeValue), test::KeyStr("b", 2, kTypeValue)},
           {"v1", "v2"},
           {test::KeyStr("a", 0, kTypeValue), test::KeyStr("b", 2, kTypeValue)},
-          {"v1", "v2"}, kMaxSequenceNumber /*last_commited_seq*/,
+          {"v1", "v2"}, kMaxSequenceNumber /*last_committed_seq*/,
           nullptr /*merge_operator*/, nullptr /*compaction_filter*/,
           true /*bottommost_level*/);
 }
@@ -689,15 +720,14 @@ TEST_P(CompactionIteratorTest, ZeroOutSequenceAtBottomLevel) {
 // permanently.
 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)},
-          {"", ""},
-          kMaxSequenceNumber /*last_commited_seq*/, nullptr /*merge_operator*/,
-          nullptr /*compaction_filter*/, true /*bottommost_level*/);
+  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)},
+      {"", ""}, kMaxSequenceNumber /*last_committed_seq*/,
+      nullptr /*merge_operator*/, nullptr /*compaction_filter*/,
+      true /*bottommost_level*/);
 }
 
 // In bottommost level, single deletions earlier than earliest snapshot can be
@@ -707,7 +737,7 @@ TEST_P(CompactionIteratorTest, RemoveSingleDeletionAtBottomLevel) {
   RunTest({test::KeyStr("a", 1, kTypeSingleDeletion),
            test::KeyStr("b", 2, kTypeSingleDeletion)},
           {"", ""}, {test::KeyStr("b", 2, kTypeSingleDeletion)}, {""},
-          kMaxSequenceNumber /*last_commited_seq*/, nullptr /*merge_operator*/,
+          kMaxSequenceNumber /*last_committed_seq*/, nullptr /*merge_operator*/,
           nullptr /*compaction_filter*/, true /*bottommost_level*/);
 }
 
@@ -864,7 +894,7 @@ TEST_F(CompactionIteratorWithSnapshotCheckerTest,
           {"v1", "v2", "v3"},
           {test::KeyStr("a", 0, kTypeValue), test::KeyStr("b", 2, kTypeValue),
            test::KeyStr("c", 3, kTypeValue)},
-          {"v1", "v2", "v3"}, kMaxSequenceNumber /*last_commited_seq*/,
+          {"v1", "v2", "v3"}, kMaxSequenceNumber /*last_committed_seq*/,
           nullptr /*merge_operator*/, nullptr /*compaction_filter*/,
           true /*bottommost_level*/);
 }
@@ -875,9 +905,7 @@ TEST_F(CompactionIteratorWithSnapshotCheckerTest,
   RunTest(
       {test::KeyStr("a", 1, kTypeDeletion), test::KeyStr("b", 2, kTypeDeletion),
        test::KeyStr("c", 3, kTypeDeletion)},
-      {"", "", ""},
-      {},
-      {"", ""}, kMaxSequenceNumber /*last_commited_seq*/,
+      {"", "", ""}, {}, {"", ""}, kMaxSequenceNumber /*last_committed_seq*/,
       nullptr /*merge_operator*/, nullptr /*compaction_filter*/,
       true /*bottommost_level*/);
 }
@@ -885,15 +913,14 @@ TEST_F(CompactionIteratorWithSnapshotCheckerTest,
 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*/);
+  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_committed_seq*/,
+          nullptr /*merge_operator*/, nullptr /*compaction_filter*/,
+          true /*bottommost_level*/);
 }
 
 TEST_F(CompactionIteratorWithSnapshotCheckerTest,
@@ -905,7 +932,7 @@ TEST_F(CompactionIteratorWithSnapshotCheckerTest,
           {"", "", ""},
           {test::KeyStr("b", 2, kTypeSingleDeletion),
            test::KeyStr("c", 3, kTypeSingleDeletion)},
-          {"", ""}, kMaxSequenceNumber /*last_commited_seq*/,
+          {"", ""}, kMaxSequenceNumber /*last_committed_seq*/,
           nullptr /*merge_operator*/, nullptr /*compaction_filter*/,
           true /*bottommost_level*/);
 }
@@ -939,9 +966,24 @@ TEST_F(CompactionIteratorWithSnapshotCheckerTest,
           2 /*earliest_write_conflict_snapshot*/);
 }
 
+// Same as above but with a blob index. In addition to the value getting
+// trimmed, the type of the KV is changed to kTypeValue.
+TEST_F(CompactionIteratorWithSnapshotCheckerTest,
+       KeepSingleDeletionForWriteConflictChecking_BlobIndex) {
+  AddSnapshot(2, 0);
+  RunTest({test::KeyStr("a", 2, kTypeSingleDeletion),
+           test::KeyStr("a", 1, kTypeBlobIndex)},
+          {"", "fake_blob_index"},
+          {test::KeyStr("a", 2, kTypeSingleDeletion),
+           test::KeyStr("a", 1, kTypeValue)},
+          {"", ""}, 2 /*last_committed_seq*/, nullptr /*merge_operator*/,
+          nullptr /*compaction_filter*/, false /*bottommost_level*/,
+          2 /*earliest_write_conflict_snapshot*/);
+}
+
 // Compaction filter should keep uncommitted key as-is, and
-//   * Convert the latest velue to deletion, and/or
-//   * if latest value is a merge, apply filter to all suequent merges.
+//   * Convert the latest value to deletion, and/or
+//   * if latest value is a merge, apply filter to all subsequent merges.
 
 TEST_F(CompactionIteratorWithSnapshotCheckerTest, CompactionFilter_Value) {
   std::unique_ptr<CompactionFilter> compaction_filter(
@@ -1033,6 +1075,188 @@ INSTANTIATE_TEST_CASE_P(CompactionIteratorWithAllowIngestBehindTestInstance,
                         CompactionIteratorWithAllowIngestBehindTest,
                         testing::Values(true, false));
 
+class CompactionIteratorTsGcTest : public CompactionIteratorTest {
+ public:
+  CompactionIteratorTsGcTest()
+      : CompactionIteratorTest(test::ComparatorWithU64Ts()) {}
+};
+
+TEST_P(CompactionIteratorTsGcTest, NoKeyEligibleForGC) {
+  constexpr char user_key[][2] = {{'a', '\0'}, {'b', '\0'}};
+  const std::vector<std::string> input_keys = {
+      test::KeyStr(/*ts=*/103, user_key[0], /*seq=*/4, kTypeValue),
+      test::KeyStr(/*ts=*/102, user_key[0], /*seq=*/3,
+                   kTypeDeletionWithTimestamp),
+      test::KeyStr(/*ts=*/104, user_key[1], /*seq=*/5, kTypeValue)};
+  const std::vector<std::string> input_values = {"a3", "", "b2"};
+  std::string full_history_ts_low;
+  // All keys' timestamps are newer than or equal to 102, thus none of them
+  // will be eligible for GC.
+  PutFixed64(&full_history_ts_low, 102);
+  const std::vector<std::string>& expected_keys = input_keys;
+  const std::vector<std::string>& expected_values = input_values;
+  const std::vector<std::pair<bool, bool>> params = {
+      {false, false}, {false, true}, {true, true}};
+  for (const std::pair<bool, bool>& param : params) {
+    const bool bottommost_level = param.first;
+    const bool key_not_exists_beyond_output_level = param.second;
+    RunTest(input_keys, input_values, expected_keys, expected_values,
+            /*last_committed_seq=*/kMaxSequenceNumber,
+            /*merge_operator=*/nullptr, /*compaction_filter=*/nullptr,
+            bottommost_level,
+            /*earliest_write_conflict_snapshot=*/kMaxSequenceNumber,
+            key_not_exists_beyond_output_level, &full_history_ts_low);
+  }
+}
+
+TEST_P(CompactionIteratorTsGcTest, AllKeysOlderThanThreshold) {
+  constexpr char user_key[][2] = {{'a', '\0'}, {'b', '\0'}};
+  const std::vector<std::string> input_keys = {
+      test::KeyStr(/*ts=*/103, user_key[0], /*seq=*/4,
+                   kTypeDeletionWithTimestamp),
+      test::KeyStr(/*ts=*/102, user_key[0], /*seq=*/3, kTypeValue),
+      test::KeyStr(/*ts=*/101, user_key[0], /*seq=*/2, kTypeValue),
+      test::KeyStr(/*ts=*/104, user_key[1], /*seq=*/5, kTypeValue)};
+  const std::vector<std::string> input_values = {"", "a2", "a1", "b5"};
+  std::string full_history_ts_low;
+  PutFixed64(&full_history_ts_low, std::numeric_limits<uint64_t>::max());
+  {
+    // With a snapshot at seq 3, both the deletion marker and the key at 3 must
+    // be preserved.
+    AddSnapshot(3);
+    const std::vector<std::string> expected_keys = {
+        input_keys[0], input_keys[1], input_keys[3]};
+    const std::vector<std::string> expected_values = {"", "a2", "b5"};
+    RunTest(input_keys, input_values, expected_keys, expected_values,
+            /*last_committed_seq=*/kMaxSequenceNumber,
+            /*merge_operator=*/nullptr, /*compaction_filter=*/nullptr,
+            /*bottommost_level=*/false,
+            /*earliest_write_conflict_snapshot=*/kMaxSequenceNumber,
+            /*key_not_exists_beyond_output_level=*/false, &full_history_ts_low);
+    ClearSnapshots();
+  }
+  {
+    // No snapshot, the deletion marker should be preserved because the user
+    // key may appear beyond output level.
+    const std::vector<std::string> expected_keys = {input_keys[0],
+                                                    input_keys[3]};
+    const std::vector<std::string> expected_values = {"", "b5"};
+    RunTest(input_keys, input_values, expected_keys, expected_values,
+            /*last_committed_seq=*/kMaxSequenceNumber,
+            /*merge_operator=*/nullptr, /*compaction_filter=*/nullptr,
+            /*bottommost_level=*/false,
+            /*earliest_write_conflict_snapshot=*/kMaxSequenceNumber,
+            /*key_not_exists_beyond_output_level=*/false, &full_history_ts_low);
+  }
+  {
+    // No snapshot, the deletion marker can be dropped because the user key
+    // does not appear in higher levels.
+    const std::vector<std::string> expected_keys = {input_keys[3]};
+    const std::vector<std::string> expected_values = {"b5"};
+    RunTest(input_keys, input_values, expected_keys, expected_values,
+            /*last_committed_seq=*/kMaxSequenceNumber,
+            /*merge_operator=*/nullptr, /*compaction_filter=*/nullptr,
+            /*bottommost_level=*/false,
+            /*earliest_write_conflict_snapshot=*/kMaxSequenceNumber,
+            /*key_not_exists_beyond_output_level=*/true, &full_history_ts_low);
+  }
+}
+
+TEST_P(CompactionIteratorTsGcTest, NewHidesOldSameSnapshot) {
+  constexpr char user_key[] = "a";
+  const std::vector<std::string> input_keys = {
+      test::KeyStr(/*ts=*/103, user_key, /*seq=*/4, kTypeDeletionWithTimestamp),
+      test::KeyStr(/*ts=*/102, user_key, /*seq=*/3, kTypeValue),
+      test::KeyStr(/*ts=*/101, user_key, /*seq=*/2, kTypeValue),
+      test::KeyStr(/*ts=*/100, user_key, /*seq=*/1, kTypeValue)};
+  const std::vector<std::string> input_values = {"", "a2", "a1", "a0"};
+  {
+    std::string full_history_ts_low;
+    // Keys whose timestamps larger than or equal to 102 will be preserved.
+    PutFixed64(&full_history_ts_low, 102);
+    const std::vector<std::string> expected_keys = {input_keys[0],
+                                                    input_keys[1]};
+    const std::vector<std::string> expected_values = {"", "a2"};
+    RunTest(input_keys, input_values, expected_keys, expected_values,
+            /*last_committed_seq=*/kMaxSequenceNumber,
+            /*merge_operator=*/nullptr, /*compaction_filter=*/nullptr,
+            /*bottommost_level=*/false,
+            /*earliest_write_conflict_snapshot=*/kMaxSequenceNumber,
+            /*key_not_exists_beyond_output_level=*/false, &full_history_ts_low);
+  }
+}
+
+TEST_P(CompactionIteratorTsGcTest, DropTombstones) {
+  constexpr char user_key[] = "a";
+  const std::vector<std::string> input_keys = {
+      test::KeyStr(/*ts=*/103, user_key, /*seq=*/4, kTypeDeletionWithTimestamp),
+      test::KeyStr(/*ts=*/102, user_key, /*seq=*/3, kTypeValue),
+      test::KeyStr(/*ts=*/101, user_key, /*seq=*/2, kTypeDeletionWithTimestamp),
+      test::KeyStr(/*ts=*/100, user_key, /*seq=*/1, kTypeValue)};
+  const std::vector<std::string> input_values = {"", "a2", "", "a0"};
+  const std::vector<std::string> expected_keys = {input_keys[0], input_keys[1]};
+  const std::vector<std::string> expected_values = {"", "a2"};
+
+  // Take a snapshot at seq 2.
+  AddSnapshot(2);
+
+  {
+    // Non-bottommost level, but key does not exist beyond output level.
+    std::string full_history_ts_low;
+    PutFixed64(&full_history_ts_low, 102);
+    RunTest(input_keys, input_values, expected_keys, expected_values,
+            /*last_committed_sequence=*/kMaxSequenceNumber,
+            /*merge_op=*/nullptr, /*compaction_filter=*/nullptr,
+            /*bottommost_level=*/false,
+            /*earliest_write_conflict_snapshot=*/kMaxSequenceNumber,
+            /*key_not_exists_beyond_output_level=*/true, &full_history_ts_low);
+  }
+  {
+    // Bottommost level
+    std::string full_history_ts_low;
+    PutFixed64(&full_history_ts_low, 102);
+    RunTest(input_keys, input_values, expected_keys, expected_values,
+            /*last_committed_seq=*/kMaxSequenceNumber,
+            /*merge_operator=*/nullptr, /*compaction_filter=*/nullptr,
+            /*bottommost_level=*/true,
+            /*earliest_write_conflict_snapshot=*/kMaxSequenceNumber,
+            /*key_not_exists_beyond_output_level=*/false, &full_history_ts_low);
+  }
+}
+
+TEST_P(CompactionIteratorTsGcTest, RewriteTs) {
+  constexpr char user_key[] = "a";
+  const std::vector<std::string> input_keys = {
+      test::KeyStr(/*ts=*/103, user_key, /*seq=*/4, kTypeDeletionWithTimestamp),
+      test::KeyStr(/*ts=*/102, user_key, /*seq=*/3, kTypeValue),
+      test::KeyStr(/*ts=*/101, user_key, /*seq=*/2, kTypeDeletionWithTimestamp),
+      test::KeyStr(/*ts=*/100, user_key, /*seq=*/1, kTypeValue)};
+  const std::vector<std::string> input_values = {"", "a2", "", "a0"};
+  const std::vector<std::string> expected_keys = {
+      input_keys[0], input_keys[1], input_keys[2],
+      test::KeyStr(/*ts=*/0, user_key, /*seq=*/0, kTypeValue)};
+  const std::vector<std::string> expected_values = {"", "a2", "", "a0"};
+
+  AddSnapshot(1);
+  AddSnapshot(2);
+
+  {
+    // Bottommost level and need to rewrite both ts and seq.
+    std::string full_history_ts_low;
+    PutFixed64(&full_history_ts_low, 102);
+    RunTest(input_keys, input_values, expected_keys, expected_values,
+            /*last_committed_seq=*/kMaxSequenceNumber,
+            /*merge_operator=*/nullptr, /*compaction_filter=*/nullptr,
+            /*bottommost_level=*/true,
+            /*earliest_write_conflict_snapshot=*/kMaxSequenceNumber,
+            /*key_not_exists_beyond_output_level=*/true, &full_history_ts_low);
+  }
+}
+
+INSTANTIATE_TEST_CASE_P(CompactionIteratorTsGcTestInstance,
+                        CompactionIteratorTsGcTest,
+                        testing::Values(true, false));
+
 }  // namespace ROCKSDB_NAMESPACE
 
 int main(int argc, char** argv) {

db/compaction/compaction_job.cc

@@ -20,6 +20,8 @@
 #include <utility>
 #include <vector>
 
+#include "db/blob/blob_file_addition.h"
+#include "db/blob/blob_file_builder.h"
 #include "db/builder.h"
 #include "db/db_impl/db_impl.h"
 #include "db/db_iter.h"
@@ -44,6 +46,7 @@
 #include "monitoring/iostats_context_imp.h"
 #include "monitoring/perf_context_imp.h"
 #include "monitoring/thread_status_util.h"
+#include "options/options_helper.h"
 #include "port/port.h"
 #include "rocksdb/db.h"
 #include "rocksdb/env.h"
@@ -138,6 +141,7 @@ struct CompactionJob::SubcompactionState {
 
   // State kept for output being generated
   std::vector<Output> outputs;
+  std::vector<BlobFileAddition> blob_file_additions;
   std::unique_ptr<WritableFileWriter> outfile;
   std::unique_ptr<TableBuilder> builder;
 
@@ -146,7 +150,7 @@ struct CompactionJob::SubcompactionState {
       // This subcompaction's output could be empty if compaction was aborted
       // before this subcompaction had a chance to generate any output files.
       // When subcompactions are executed sequentially this is more likely and
-      // will be particulalry likely for the later subcompactions to be empty.
+      // will be particularly likely for the later subcompactions to be empty.
       // Once they are run in parallel however it should be much rarer.
       return nullptr;
     } else {
@@ -231,21 +235,13 @@ struct CompactionJob::CompactionState {
   std::vector<CompactionJob::SubcompactionState> sub_compact_states;
   Status status;
 
-  uint64_t total_bytes;
-  uint64_t num_output_records;
+  size_t num_output_files = 0;
+  uint64_t total_bytes = 0;
+  size_t num_blob_output_files = 0;
+  uint64_t total_blob_bytes = 0;
+  uint64_t num_output_records = 0;
 
-  explicit CompactionState(Compaction* c)
-      : compaction(c),
-        total_bytes(0),
-        num_output_records(0) {}
-
-  size_t NumOutputFiles() {
-    size_t total = 0;
-    for (auto& s : sub_compact_states) {
-      total += s.outputs.size();
-    }
-    return total;
-  }
+  explicit CompactionState(Compaction* c) : compaction(c) {}
 
   Slice SmallestUserKey() {
     for (const auto& sub_compact_state : sub_compact_states) {
@@ -272,11 +268,29 @@ struct CompactionJob::CompactionState {
 };
 
 void CompactionJob::AggregateStatistics() {
+  assert(compact_);
+
   for (SubcompactionState& sc : compact_->sub_compact_states) {
+    auto& outputs = sc.outputs;
+
+    if (!outputs.empty() && !outputs.back().meta.fd.file_size) {
+      // An error occurred, so ignore the last output.
+      outputs.pop_back();
+    }
+
+    compact_->num_output_files += outputs.size();
     compact_->total_bytes += sc.total_bytes;
+
+    const auto& blobs = sc.blob_file_additions;
+
+    compact_->num_blob_output_files += blobs.size();
+
+    for (const auto& blob : blobs) {
+      compact_->total_blob_bytes += blob.GetTotalBlobBytes();
+    }
+
     compact_->num_output_records += sc.num_output_records;
-  }
-  for (SubcompactionState& sc : compact_->sub_compact_states) {
+
     compaction_job_stats_->Add(sc.compaction_job_stats);
   }
 }
@@ -286,7 +300,8 @@ CompactionJob::CompactionJob(
     const FileOptions& file_options, VersionSet* versions,
     const std::atomic<bool>* shutting_down,
     const SequenceNumber preserve_deletes_seqnum, LogBuffer* log_buffer,
-    FSDirectory* db_directory, FSDirectory* output_directory, Statistics* stats,
+    FSDirectory* db_directory, FSDirectory* output_directory,
+    FSDirectory* blob_output_directory, Statistics* stats,
     InstrumentedMutex* db_mutex, ErrorHandler* db_error_handler,
     std::vector<SequenceNumber> existing_snapshots,
     SequenceNumber earliest_write_conflict_snapshot,
@@ -295,15 +310,21 @@ CompactionJob::CompactionJob(
     const std::string& dbname, CompactionJobStats* compaction_job_stats,
     Env::Priority thread_pri, const std::shared_ptr<IOTracer>& io_tracer,
     const std::atomic<int>* manual_compaction_paused, const std::string& db_id,
-    const std::string& db_session_id)
-    : job_id_(job_id),
-      compact_(new CompactionState(compaction)),
-      compaction_job_stats_(compaction_job_stats),
+    const std::string& db_session_id, std::string full_history_ts_low,
+    BlobFileCompletionCallback* blob_callback)
+    : compact_(new CompactionState(compaction)),
       compaction_stats_(compaction->compaction_reason(), 1),
+      db_options_(db_options),
+      log_buffer_(log_buffer),
+      output_directory_(output_directory),
+      stats_(stats),
+      bottommost_level_(false),
+      write_hint_(Env::WLTH_NOT_SET),
+      job_id_(job_id),
+      compaction_job_stats_(compaction_job_stats),
       dbname_(dbname),
       db_id_(db_id),
       db_session_id_(db_session_id),
-      db_options_(db_options),
       file_options_(file_options),
       env_(db_options.env),
       io_tracer_(io_tracer),
@@ -314,10 +335,8 @@ CompactionJob::CompactionJob(
       shutting_down_(shutting_down),
       manual_compaction_paused_(manual_compaction_paused),
       preserve_deletes_seqnum_(preserve_deletes_seqnum),
-      log_buffer_(log_buffer),
       db_directory_(db_directory),
-      output_directory_(output_directory),
-      stats_(stats),
+      blob_output_directory_(blob_output_directory),
       db_mutex_(db_mutex),
       db_error_handler_(db_error_handler),
       existing_snapshots_(std::move(existing_snapshots)),
@@ -325,11 +344,11 @@ CompactionJob::CompactionJob(
       snapshot_checker_(snapshot_checker),
       table_cache_(std::move(table_cache)),
       event_logger_(event_logger),
-      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) {
+      thread_pri_(thread_pri),
+      full_history_ts_low_(std::move(full_history_ts_low)),
+      blob_callback_(blob_callback) {
   assert(compaction_job_stats_ != nullptr);
   assert(log_buffer_ != nullptr);
   const auto* cfd = compact_->compaction->column_family_data();
@@ -391,7 +410,7 @@ void CompactionJob::Prepare() {
   AutoThreadOperationStageUpdater stage_updater(
       ThreadStatus::STAGE_COMPACTION_PREPARE);
 
-  // Generate file_levels_ for compaction berfore making Iterator
+  // Generate file_levels_ for compaction before making Iterator
   auto* c = compact_->compaction;
   assert(c->column_family_data() != nullptr);
   assert(c->column_family_data()->current()->storage_info()->NumLevelFiles(
@@ -403,7 +422,7 @@ void CompactionJob::Prepare() {
 
   if (c->ShouldFormSubcompactions()) {
     {
-      StopWatch sw(env_, stats_, SUBCOMPACTION_SETUP_TIME);
+      StopWatch sw(db_options_.clock, stats_, SUBCOMPACTION_SETUP_TIME);
       GenSubcompactionBoundaries();
     }
     assert(sizes_.size() == boundaries_.size() + 1);
@@ -569,7 +588,7 @@ Status CompactionJob::Run() {
 
   const size_t num_threads = compact_->sub_compact_states.size();
   assert(num_threads > 0);
-  const uint64_t start_micros = env_->NowMicros();
+  const uint64_t start_micros = db_options_.clock->NowMicros();
 
   // Launch a thread for each of subcompactions 1...num_threads-1
   std::vector<port::Thread> thread_pool;
@@ -588,7 +607,7 @@ Status CompactionJob::Run() {
     thread.join();
   }
 
-  compaction_stats_.micros = env_->NowMicros() - start_micros;
+  compaction_stats_.micros = db_options_.clock->NowMicros() - start_micros;
   compaction_stats_.cpu_micros = 0;
   for (size_t i = 0; i < compact_->sub_compact_states.size(); i++) {
     compaction_stats_.cpu_micros +=
@@ -604,18 +623,34 @@ Status CompactionJob::Run() {
   // Check if any thread encountered an error during execution
   Status status;
   IOStatus io_s;
+  bool wrote_new_blob_files = false;
+
   for (const auto& state : compact_->sub_compact_states) {
     if (!state.status.ok()) {
       status = state.status;
       io_s = state.io_status;
       break;
     }
+
+    if (!state.blob_file_additions.empty()) {
+      wrote_new_blob_files = true;
+    }
   }
+
   if (io_status_.ok()) {
     io_status_ = io_s;
   }
-  if (status.ok() && output_directory_) {
-    io_s = output_directory_->Fsync(IOOptions(), nullptr);
+  if (status.ok()) {
+    constexpr IODebugContext* dbg = nullptr;
+
+    if (output_directory_) {
+      io_s = output_directory_->Fsync(IOOptions(), dbg);
+    }
+
+    if (io_s.ok() && wrote_new_blob_files && blob_output_directory_ &&
+        blob_output_directory_ != output_directory_) {
+      io_s = blob_output_directory_->Fsync(IOOptions(), dbg);
+    }
   }
   if (io_status_.ok()) {
     io_status_ = io_s;
@@ -721,6 +756,7 @@ Status CompactionJob::Run() {
   // Finish up all book-keeping to unify the subcompaction results
   AggregateStatistics();
   UpdateCompactionStats();
+
   RecordCompactionIOStats();
   LogFlush(db_options_.info_log);
   TEST_SYNC_POINT("CompactionJob::Run():End");
@@ -730,11 +766,16 @@ Status CompactionJob::Run() {
 }
 
 Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options) {
+  assert(compact_);
+
   AutoThreadOperationStageUpdater stage_updater(
       ThreadStatus::STAGE_COMPACTION_INSTALL);
   db_mutex_->AssertHeld();
   Status status = compact_->status;
+
   ColumnFamilyData* cfd = compact_->compaction->column_family_data();
+  assert(cfd);
+
   cfd->internal_stats()->AddCompactionStats(
       compact_->compaction->output_level(), thread_pri_, compaction_stats_);
 
@@ -744,6 +785,7 @@ Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options) {
   if (!versions_->io_status().ok()) {
     io_status_ = versions_->io_status();
   }
+
   VersionStorageInfo::LevelSummaryStorage tmp;
   auto vstorage = cfd->current()->storage_info();
   const auto& stats = compaction_stats_;
@@ -753,40 +795,58 @@ Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options) {
   double bytes_read_per_sec = 0;
   double bytes_written_per_sec = 0;
 
-  if (stats.bytes_read_non_output_levels > 0) {
-    read_write_amp = (stats.bytes_written + stats.bytes_read_output_level +
-                      stats.bytes_read_non_output_levels) /
-                     static_cast<double>(stats.bytes_read_non_output_levels);
-    write_amp = stats.bytes_written /
-                static_cast<double>(stats.bytes_read_non_output_levels);
+  const uint64_t bytes_read_non_output_and_blob =
+      stats.bytes_read_non_output_levels + stats.bytes_read_blob;
+  const uint64_t bytes_read_all =
+      stats.bytes_read_output_level + bytes_read_non_output_and_blob;
+  const uint64_t bytes_written_all =
+      stats.bytes_written + stats.bytes_written_blob;
+
+  if (bytes_read_non_output_and_blob > 0) {
+    read_write_amp = (bytes_written_all + bytes_read_all) /
+                     static_cast<double>(bytes_read_non_output_and_blob);
+    write_amp =
+        bytes_written_all / static_cast<double>(bytes_read_non_output_and_blob);
   }
   if (stats.micros > 0) {
-    bytes_read_per_sec =
-        (stats.bytes_read_non_output_levels + stats.bytes_read_output_level) /
-        static_cast<double>(stats.micros);
+    bytes_read_per_sec = bytes_read_all / static_cast<double>(stats.micros);
     bytes_written_per_sec =
-        stats.bytes_written / static_cast<double>(stats.micros);
+        bytes_written_all / static_cast<double>(stats.micros);
   }
 
+  const std::string& column_family_name = cfd->GetName();
+
+  constexpr double kMB = 1048576.0;
+
   ROCKS_LOG_BUFFER(
       log_buffer_,
       "[%s] compacted to: %s, MB/sec: %.1f rd, %.1f wr, level %d, "
-      "files in(%d, %d) out(%d) "
-      "MB in(%.1f, %.1f) out(%.1f), read-write-amplify(%.1f) "
-      "write-amplify(%.1f) %s, records in: %" PRIu64
+      "files in(%d, %d) out(%d +%d blob) "
+      "MB in(%.1f, %.1f +%.1f blob) out(%.1f +%.1f blob), "
+      "read-write-amplify(%.1f) write-amplify(%.1f) %s, records in: %" PRIu64
       ", records dropped: %" PRIu64 " output_compression: %s\n",
-      cfd->GetName().c_str(), vstorage->LevelSummary(&tmp), bytes_read_per_sec,
-      bytes_written_per_sec, compact_->compaction->output_level(),
+      column_family_name.c_str(), vstorage->LevelSummary(&tmp),
+      bytes_read_per_sec, bytes_written_per_sec,
+      compact_->compaction->output_level(),
       stats.num_input_files_in_non_output_levels,
       stats.num_input_files_in_output_level, stats.num_output_files,
-      stats.bytes_read_non_output_levels / 1048576.0,
-      stats.bytes_read_output_level / 1048576.0,
-      stats.bytes_written / 1048576.0, read_write_amp, write_amp,
-      status.ToString().c_str(), stats.num_input_records,
+      stats.num_output_files_blob, stats.bytes_read_non_output_levels / kMB,
+      stats.bytes_read_output_level / kMB, stats.bytes_read_blob / kMB,
+      stats.bytes_written / kMB, stats.bytes_written_blob / kMB, read_write_amp,
+      write_amp, status.ToString().c_str(), stats.num_input_records,
       stats.num_dropped_records,
       CompressionTypeToString(compact_->compaction->output_compression())
           .c_str());
 
+  const auto& blob_files = vstorage->GetBlobFiles();
+  if (!blob_files.empty()) {
+    ROCKS_LOG_BUFFER(log_buffer_,
+                     "[%s] Blob file summary: head=%" PRIu64 ", tail=%" PRIu64
+                     "\n",
+                     column_family_name.c_str(), blob_files.begin()->first,
+                     blob_files.rbegin()->first);
+  }
+
   UpdateCompactionJobStats(stats);
 
   auto stream = event_logger_->LogToBuffer(log_buffer_);
@@ -795,11 +855,18 @@ Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options) {
          << "compaction_time_micros" << stats.micros
          << "compaction_time_cpu_micros" << stats.cpu_micros << "output_level"
          << compact_->compaction->output_level() << "num_output_files"
-         << compact_->NumOutputFiles() << "total_output_size"
-         << compact_->total_bytes << "num_input_records"
-         << stats.num_input_records << "num_output_records"
-         << compact_->num_output_records << "num_subcompactions"
-         << compact_->sub_compact_states.size() << "output_compression"
+         << compact_->num_output_files << "total_output_size"
+         << compact_->total_bytes;
+
+  if (compact_->num_blob_output_files > 0) {
+    stream << "num_blob_output_files" << compact_->num_blob_output_files
+           << "total_blob_output_size" << compact_->total_blob_bytes;
+  }
+
+  stream << "num_input_records" << stats.num_input_records
+         << "num_output_records" << compact_->num_output_records
+         << "num_subcompactions" << compact_->sub_compact_states.size()
+         << "output_compression"
          << CompressionTypeToString(compact_->compaction->output_compression());
 
   stream << "num_single_delete_mismatches"
@@ -823,14 +890,20 @@ Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options) {
   }
   stream.EndArray();
 
+  if (!blob_files.empty()) {
+    stream << "blob_file_head" << blob_files.begin()->first;
+    stream << "blob_file_tail" << blob_files.rbegin()->first;
+  }
+
   CleanupCompaction();
   return status;
 }
 
 void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
-  assert(sub_compact != nullptr);
+  assert(sub_compact);
+  assert(sub_compact->compaction);
 
-  uint64_t prev_cpu_micros = env_->NowCPUNanos() / 1000;
+  uint64_t prev_cpu_micros = db_options_.clock->CPUNanos() / 1000;
 
   ColumnFamilyData* cfd = sub_compact->compaction->column_family_data();
 
@@ -892,13 +965,30 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
   }
 
   MergeHelper merge(
-      env_, cfd->user_comparator(), cfd->ioptions()->merge_operator,
+      env_, cfd->user_comparator(), cfd->ioptions()->merge_operator.get(),
       compaction_filter, db_options_.info_log.get(),
       false /* internal key corruption is expected */,
       existing_snapshots_.empty() ? 0 : existing_snapshots_.back(),
       snapshot_checker_, compact_->compaction->level(),
       db_options_.statistics.get());
 
+  const MutableCFOptions* mutable_cf_options =
+      sub_compact->compaction->mutable_cf_options();
+  assert(mutable_cf_options);
+
+  std::vector<std::string> blob_file_paths;
+
+  std::unique_ptr<BlobFileBuilder> blob_file_builder(
+      mutable_cf_options->enable_blob_files
+          ? new BlobFileBuilder(versions_, fs_.get(),
+                                sub_compact->compaction->immutable_cf_options(),
+                                mutable_cf_options, &file_options_, job_id_,
+                                cfd->GetID(), cfd->GetName(),
+                                Env::IOPriority::IO_LOW, write_hint_,
+                                io_tracer_, blob_callback_, &blob_file_paths,
+                                &sub_compact->blob_file_additions)
+          : nullptr);
+
   TEST_SYNC_POINT("CompactionJob::Run():Inprogress");
   TEST_SYNC_POINT_CALLBACK(
       "CompactionJob::Run():PausingManualCompaction:1",
@@ -916,15 +1006,17 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
   }
 
   Status status;
+  const std::string* const full_history_ts_low =
+      full_history_ts_low_.empty() ? nullptr : &full_history_ts_low_;
   sub_compact->c_iter.reset(new CompactionIterator(
       input.get(), cfd->user_comparator(), &merge, versions_->LastSequence(),
       &existing_snapshots_, earliest_write_conflict_snapshot_,
       snapshot_checker_, env_, ShouldReportDetailedTime(env_, stats_),
       /*expect_valid_internal_key=*/true, &range_del_agg,
-      /* blob_file_builder */ nullptr, db_options_.allow_data_in_errors,
+      blob_file_builder.get(), db_options_.allow_data_in_errors,
       sub_compact->compaction, compaction_filter, shutting_down_,
-      preserve_deletes_seqnum_, manual_compaction_paused_,
-      db_options_.info_log));
+      preserve_deletes_seqnum_, manual_compaction_paused_, db_options_.info_log,
+      full_history_ts_low));
   auto c_iter = sub_compact->c_iter.get();
   c_iter->SeekToFirst();
   if (c_iter->Valid() && sub_compact->compaction->output_level() != 0) {
@@ -1037,6 +1129,10 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
     }
   }
 
+  sub_compact->compaction_job_stats.num_blobs_read =
+      c_iter_stats.num_blobs_read;
+  sub_compact->compaction_job_stats.total_blob_bytes_read =
+      c_iter_stats.total_blob_bytes_read;
   sub_compact->compaction_job_stats.num_input_deletion_records =
       c_iter_stats.num_input_deletion_records;
   sub_compact->compaction_job_stats.num_corrupt_keys =
@@ -1093,8 +1189,17 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
     RecordDroppedKeys(range_del_out_stats, &sub_compact->compaction_job_stats);
   }
 
+  if (blob_file_builder) {
+    if (status.ok()) {
+      status = blob_file_builder->Finish();
+    } else {
+      blob_file_builder->Abandon();
+    }
+    blob_file_builder.reset();
+  }
+
   sub_compact->compaction_job_stats.cpu_micros =
-      env_->NowCPUNanos() / 1000 - prev_cpu_micros;
+      db_options_.clock->CPUNanos() / 1000 - prev_cpu_micros;
 
   if (measure_io_stats_) {
     sub_compact->compaction_job_stats.file_write_nanos +=
@@ -1352,7 +1457,6 @@ Status CompactionJob::FinishCompactionOutputFile(
              ExtractInternalKeyFooter(meta->smallest.Encode()) !=
                  PackSequenceAndType(0, kTypeRangeDeletion));
     }
-    meta->marked_for_compaction = sub_compact->builder->NeedCompact();
   }
   const uint64_t current_entries = sub_compact->builder->NumEntries();
   if (s.ok()) {
@@ -1367,13 +1471,14 @@ Status CompactionJob::FinishCompactionOutputFile(
   const uint64_t current_bytes = sub_compact->builder->FileSize();
   if (s.ok()) {
     meta->fd.file_size = current_bytes;
+    meta->marked_for_compaction = sub_compact->builder->NeedCompact();
   }
   sub_compact->current_output()->finished = true;
   sub_compact->total_bytes += current_bytes;
 
   // Finish and check for file errors
   if (s.ok()) {
-    StopWatch sw(env_, stats_, COMPACTION_OUTFILE_SYNC_MICROS);
+    StopWatch sw(db_options_.clock, stats_, COMPACTION_OUTFILE_SYNC_MICROS);
     io_s = sub_compact->outfile->Sync(db_options_.use_fsync);
   }
   if (s.ok() && io_s.ok()) {
@@ -1410,7 +1515,18 @@ Status CompactionJob::FinishCompactionOutputFile(
     std::string fname =
         TableFileName(sub_compact->compaction->immutable_cf_options()->cf_paths,
                       meta->fd.GetNumber(), meta->fd.GetPathId());
-    env_->DeleteFile(fname);
+
+    // TODO(AR) it is not clear if there are any larger implications if
+    // DeleteFile fails here
+    Status ds = env_->DeleteFile(fname);
+    if (!ds.ok()) {
+      ROCKS_LOG_WARN(
+          db_options_.info_log,
+          "[%s] [JOB %d] Unable to remove SST file for table #%" PRIu64
+          " at bottom level%s",
+          cfd->GetName().c_str(), job_id_, output_number,
+          meta->marked_for_compaction ? " (need compaction)" : "");
+    }
 
     // Also need to remove the file from outputs, or it will be added to the
     // VersionEdit.
@@ -1434,9 +1550,7 @@ Status CompactionJob::FinishCompactionOutputFile(
   FileDescriptor output_fd;
   uint64_t oldest_blob_file_number = kInvalidBlobFileNumber;
   if (meta != nullptr) {
-    fname =
-        TableFileName(sub_compact->compaction->immutable_cf_options()->cf_paths,
-                      meta->fd.GetNumber(), meta->fd.GetPathId());
+    fname = GetTableFileName(meta->fd.GetNumber());
     output_fd = meta->fd;
     oldest_blob_file_number = meta->oldest_blob_file_number;
   } else {
@@ -1465,9 +1579,7 @@ Status CompactionJob::FinishCompactionOutputFile(
           "CompactionJob::FinishCompactionOutputFile:"
           "MaxAllowedSpaceReached");
       InstrumentedMutexLock l(db_mutex_);
-      // Should handle return error?
-      db_error_handler_->SetBGError(s, BackgroundErrorReason::kCompaction)
-          .PermitUncheckedError();
+      db_error_handler_->SetBGError(s, BackgroundErrorReason::kCompaction);
     }
   }
 #endif
@@ -1479,9 +1591,13 @@ Status CompactionJob::FinishCompactionOutputFile(
 
 Status CompactionJob::InstallCompactionResults(
     const MutableCFOptions& mutable_cf_options) {
+  assert(compact_);
+
   db_mutex_->AssertHeld();
 
   auto* compaction = compact_->compaction;
+  assert(compaction);
+
   // paranoia: verify that the files that we started with
   // still exist in the current version and in the same original level.
   // This ensures that a concurrent compaction did not erroneously
@@ -1497,23 +1613,32 @@ Status CompactionJob::InstallCompactionResults(
 
   {
     Compaction::InputLevelSummaryBuffer inputs_summary;
-    ROCKS_LOG_INFO(
-        db_options_.info_log, "[%s] [JOB %d] Compacted %s => %" PRIu64 " bytes",
-        compaction->column_family_data()->GetName().c_str(), job_id_,
-        compaction->InputLevelSummary(&inputs_summary), compact_->total_bytes);
+    ROCKS_LOG_INFO(db_options_.info_log,
+                   "[%s] [JOB %d] Compacted %s => %" PRIu64 " bytes",
+                   compaction->column_family_data()->GetName().c_str(), job_id_,
+                   compaction->InputLevelSummary(&inputs_summary),
+                   compact_->total_bytes + compact_->total_blob_bytes);
   }
 
+  VersionEdit* const edit = compaction->edit();
+  assert(edit);
+
   // Add compaction inputs
-  compaction->AddInputDeletions(compact_->compaction->edit());
+  compaction->AddInputDeletions(edit);
 
   for (const auto& sub_compact : compact_->sub_compact_states) {
     for (const auto& out : sub_compact.outputs) {
-      compaction->edit()->AddFile(compaction->output_level(), out.meta);
+      edit->AddFile(compaction->output_level(), out.meta);
+    }
+
+    for (const auto& blob : sub_compact.blob_file_additions) {
+      edit->AddBlobFile(blob);
     }
   }
+
   return versions_->LogAndApply(compaction->column_family_data(),
-                                mutable_cf_options, compaction->edit(),
-                                db_mutex_, db_directory_);
+                                mutable_cf_options, edit, db_mutex_,
+                                db_directory_);
 }
 
 void CompactionJob::RecordCompactionIOStats() {
@@ -1545,9 +1670,7 @@ Status CompactionJob::OpenCompactionOutputFile(
   assert(sub_compact->builder == nullptr);
   // no need to lock because VersionSet::next_file_number_ is atomic
   uint64_t file_number = versions_->NewFileNumber();
-  std::string fname =
-      TableFileName(sub_compact->compaction->immutable_cf_options()->cf_paths,
-                    file_number, sub_compact->compaction->output_path_id());
+  std::string fname = GetTableFileName(file_number);
   // Fire events.
   ColumnFamilyData* cfd = sub_compact->compaction->column_family_data();
 #ifndef ROCKSDB_LITE
@@ -1590,7 +1713,7 @@ Status CompactionJob::OpenCompactionOutputFile(
 
   // Try to figure out the output file's oldest ancester time.
   int64_t temp_current_time = 0;
-  auto get_time_status = env_->GetCurrentTime(&temp_current_time);
+  auto get_time_status = db_options_.clock->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,
@@ -1621,14 +1744,16 @@ Status CompactionJob::OpenCompactionOutputFile(
 
   writable_file->SetIOPriority(Env::IOPriority::IO_LOW);
   writable_file->SetWriteLifeTimeHint(write_hint_);
+  FileTypeSet tmp_set = db_options_.checksum_handoff_file_types;
   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, file_options_, env_, io_tracer_,
-      db_options_.statistics.get(), listeners,
-      db_options_.file_checksum_gen_factory.get()));
+      std::move(writable_file), fname, file_options_, db_options_.clock,
+      io_tracer_, db_options_.statistics.get(), listeners,
+      db_options_.file_checksum_gen_factory.get(),
+      tmp_set.Contains(FileType::kTableFile)));
 
   // If the Column family flag is to only optimize filters for hits,
   // we can skip creating filters if this is the bottommost_level where
@@ -1641,7 +1766,6 @@ 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,
       oldest_ancester_time, 0 /* oldest_key_time */,
@@ -1689,6 +1813,8 @@ void CopyPrefix(const Slice& src, size_t prefix_length, std::string* dst) {
 #endif  // !ROCKSDB_LITE
 
 void CompactionJob::UpdateCompactionStats() {
+  assert(compact_);
+
   Compaction* compaction = compact_->compaction;
   compaction_stats_.num_input_files_in_non_output_levels = 0;
   compaction_stats_.num_input_files_in_output_level = 0;
@@ -1706,27 +1832,20 @@ void CompactionJob::UpdateCompactionStats() {
     }
   }
 
-  uint64_t num_output_records = 0;
+  assert(compaction_job_stats_);
+  compaction_stats_.bytes_read_blob =
+      compaction_job_stats_->total_blob_bytes_read;
 
-  for (const auto& sub_compact : compact_->sub_compact_states) {
-    size_t num_output_files = sub_compact.outputs.size();
-    if (sub_compact.builder != nullptr) {
-      // An error occurred so ignore the last output.
-      assert(num_output_files > 0);
-      --num_output_files;
-    }
-    compaction_stats_.num_output_files += static_cast<int>(num_output_files);
+  compaction_stats_.num_output_files =
+      static_cast<int>(compact_->num_output_files);
+  compaction_stats_.num_output_files_blob =
+      static_cast<int>(compact_->num_blob_output_files);
+  compaction_stats_.bytes_written = compact_->total_bytes;
+  compaction_stats_.bytes_written_blob = compact_->total_blob_bytes;
 
-    num_output_records += sub_compact.num_output_records;
-
-    for (const auto& out : sub_compact.outputs) {
-      compaction_stats_.bytes_written += out.meta.fd.file_size;
-    }
-  }
-
-  if (compaction_stats_.num_input_records > num_output_records) {
+  if (compaction_stats_.num_input_records > compact_->num_output_records) {
     compaction_stats_.num_dropped_records =
-        compaction_stats_.num_input_records - num_output_records;
+        compaction_stats_.num_input_records - compact_->num_output_records;
   }
 }
 
@@ -1762,10 +1881,12 @@ void CompactionJob::UpdateCompactionJobStats(
 
   // output information
   compaction_job_stats_->total_output_bytes = stats.bytes_written;
+  compaction_job_stats_->total_output_bytes_blob = stats.bytes_written_blob;
   compaction_job_stats_->num_output_records = compact_->num_output_records;
   compaction_job_stats_->num_output_files = stats.num_output_files;
+  compaction_job_stats_->num_output_files_blob = stats.num_output_files_blob;
 
-  if (compact_->NumOutputFiles() > 0U) {
+  if (stats.num_output_files > 0) {
     CopyPrefix(compact_->SmallestUserKey(),
                CompactionJobStats::kMaxPrefixLength,
                &compaction_job_stats_->smallest_output_key_prefix);
@@ -1812,4 +1933,132 @@ void CompactionJob::LogCompaction() {
   }
 }
 
+std::string CompactionJob::GetTableFileName(uint64_t file_number) {
+  return TableFileName(compact_->compaction->immutable_cf_options()->cf_paths,
+                       file_number, compact_->compaction->output_path_id());
+}
+
+std::string CompactionServiceCompactionJob::GetTableFileName(
+    uint64_t file_number) {
+  return MakeTableFileName(output_path_, file_number);
+}
+
+CompactionServiceCompactionJob::CompactionServiceCompactionJob(
+    int job_id, Compaction* compaction, const ImmutableDBOptions& db_options,
+    const FileOptions& file_options, VersionSet* versions,
+    const std::atomic<bool>* shutting_down, LogBuffer* log_buffer,
+    FSDirectory* output_directory, Statistics* stats,
+    InstrumentedMutex* db_mutex, ErrorHandler* db_error_handler,
+    std::vector<SequenceNumber> existing_snapshots,
+    std::shared_ptr<Cache> table_cache, EventLogger* event_logger,
+    const std::string& dbname, const std::shared_ptr<IOTracer>& io_tracer,
+    const std::string& db_id, const std::string& db_session_id,
+    const std::string& output_path,
+    const CompactionServiceInput& compaction_service_input,
+    CompactionServiceResult* compaction_service_result)
+    : CompactionJob(
+          job_id, compaction, db_options, file_options, versions, shutting_down,
+          0, log_buffer, nullptr, output_directory, nullptr, stats, db_mutex,
+          db_error_handler, existing_snapshots, kMaxSequenceNumber, nullptr,
+          table_cache, event_logger,
+          compaction->mutable_cf_options()->paranoid_file_checks,
+          compaction->mutable_cf_options()->report_bg_io_stats, dbname,
+          &(compaction_service_result->stats), Env::Priority::USER, io_tracer,
+          nullptr, db_id, db_session_id,
+          compaction->column_family_data()->GetFullHistoryTsLow()),
+      output_path_(output_path),
+      compaction_input_(compaction_service_input),
+      compaction_result_(compaction_service_result) {}
+
+Status CompactionServiceCompactionJob::Run() {
+  AutoThreadOperationStageUpdater stage_updater(
+      ThreadStatus::STAGE_COMPACTION_RUN);
+
+  auto* c = compact_->compaction;
+  assert(c->column_family_data() != nullptr);
+  assert(c->column_family_data()->current()->storage_info()->NumLevelFiles(
+             compact_->compaction->level()) > 0);
+
+  write_hint_ =
+      c->column_family_data()->CalculateSSTWriteHint(c->output_level());
+  bottommost_level_ = c->bottommost_level();
+
+  compact_->sub_compact_states.emplace_back(c, compaction_input_.begin,
+                                            compaction_input_.end,
+                                            compaction_input_.approx_size);
+
+  log_buffer_->FlushBufferToLog();
+  LogCompaction();
+  const uint64_t start_micros = db_options_.clock->NowMicros();
+  // Pick the only sub-compaction we should have
+  assert(compact_->sub_compact_states.size() == 1);
+  SubcompactionState* sub_compact = compact_->sub_compact_states.data();
+
+  ProcessKeyValueCompaction(sub_compact);
+
+  compaction_stats_.micros = db_options_.clock->NowMicros() - start_micros;
+  compaction_stats_.cpu_micros = sub_compact->compaction_job_stats.cpu_micros;
+
+  RecordTimeToHistogram(stats_, COMPACTION_TIME, compaction_stats_.micros);
+  RecordTimeToHistogram(stats_, COMPACTION_CPU_TIME,
+                        compaction_stats_.cpu_micros);
+
+  Status status = sub_compact->status;
+  IOStatus io_s = sub_compact->io_status;
+
+  if (io_status_.ok()) {
+    io_status_ = io_s;
+  }
+
+  if (status.ok()) {
+    constexpr IODebugContext* dbg = nullptr;
+
+    if (output_directory_) {
+      io_s = output_directory_->Fsync(IOOptions(), dbg);
+    }
+  }
+  if (io_status_.ok()) {
+    io_status_ = io_s;
+  }
+  if (status.ok()) {
+    status = io_s;
+  }
+  if (status.ok()) {
+    // TODO: Add verify_table() and VerifyCompactionFileConsistency()
+  }
+
+  // Finish up all book-keeping to unify the subcompaction results
+  AggregateStatistics();
+  UpdateCompactionStats();
+
+  compaction_result_->bytes_written = IOSTATS(bytes_written);
+  compaction_result_->bytes_read = IOSTATS(bytes_read);
+  RecordCompactionIOStats();
+
+  LogFlush(db_options_.info_log);
+  compact_->status = status;
+  compact_->status.PermitUncheckedError();
+
+  // Build compaction result
+  compaction_result_->output_level = compact_->compaction->output_level();
+  compaction_result_->output_path = output_path_;
+  for (const auto& output_file : sub_compact->outputs) {
+    auto& meta = output_file.meta;
+    compaction_result_->output_files.emplace_back(
+        MakeTableFileName(meta.fd.GetNumber()), meta.fd.smallest_seqno,
+        meta.fd.largest_seqno, meta.smallest.Encode().ToString(),
+        meta.largest.Encode().ToString(), meta.oldest_ancester_time,
+        meta.file_creation_time, output_file.validator.GetHash(),
+        meta.marked_for_compaction);
+  }
+  compaction_result_->num_output_records = sub_compact->num_output_records;
+  compaction_result_->total_bytes = sub_compact->total_bytes;
+
+  return status;
+}
+
+void CompactionServiceCompactionJob::CleanupCompaction() {
+  CompactionJob::CleanupCompaction();
+}
+
 }  // namespace ROCKSDB_NAMESPACE

db/compaction/compaction_job.h

@@ -17,6 +17,7 @@
 #include <utility>
 #include <vector>
 
+#include "db/blob/blob_file_completion_callback.h"
 #include "db/column_family.h"
 #include "db/compaction/compaction_iterator.h"
 #include "db/dbformat.h"
@@ -50,6 +51,7 @@ class Arena;
 class ErrorHandler;
 class MemTable;
 class SnapshotChecker;
+class SystemClock;
 class TableCache;
 class Version;
 class VersionEdit;
@@ -68,8 +70,8 @@ class CompactionJob {
       const std::atomic<bool>* shutting_down,
       const SequenceNumber preserve_deletes_seqnum, LogBuffer* log_buffer,
       FSDirectory* db_directory, FSDirectory* output_directory,
-      Statistics* stats, InstrumentedMutex* db_mutex,
-      ErrorHandler* db_error_handler,
+      FSDirectory* blob_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,
@@ -78,9 +80,11 @@ class CompactionJob {
       const std::string& dbname, CompactionJobStats* compaction_job_stats,
       Env::Priority thread_pri, const std::shared_ptr<IOTracer>& io_tracer,
       const std::atomic<int>* manual_compaction_paused = nullptr,
-      const std::string& db_id = "", const std::string& db_session_id = "");
+      const std::string& db_id = "", const std::string& db_session_id = "",
+      std::string full_history_ts_low = "",
+      BlobFileCompletionCallback* blob_callback = nullptr);
 
-  ~CompactionJob();
+  virtual ~CompactionJob();
 
   // no copy/move
   CompactionJob(CompactionJob&& job) = delete;
@@ -103,11 +107,35 @@ class CompactionJob {
   // Return the IO status
   IOStatus io_status() const { return io_status_; }
 
- private:
+ protected:
   struct SubcompactionState;
+  // CompactionJob state
+  struct CompactionState;
 
   void AggregateStatistics();
+  void UpdateCompactionStats();
+  void LogCompaction();
+  void RecordCompactionIOStats();
+  void CleanupCompaction();
 
+  // Call compaction filter. Then iterate through input and compact the
+  // kv-pairs
+  void ProcessKeyValueCompaction(SubcompactionState* sub_compact);
+
+  CompactionState* compact_;
+  InternalStats::CompactionStats compaction_stats_;
+  const ImmutableDBOptions& db_options_;
+  LogBuffer* log_buffer_;
+  FSDirectory* output_directory_;
+  Statistics* stats_;
+  // Is this compaction creating a file in the bottom most level?
+  bool bottommost_level_;
+
+  Env::WriteLifeTimeHint write_hint_;
+
+  IOStatus io_status_;
+
+ private:
   // Generates a histogram representing potential divisions of key ranges from
   // the input. It adds the starting and/or ending keys of certain input files
   // to the working set and then finds the approximate size of data in between
@@ -118,9 +146,6 @@ class CompactionJob {
   // update the thread status for starting a compaction.
   void ReportStartedCompaction(Compaction* compaction);
   void AllocateCompactionOutputFileNumbers();
-  // Call compaction filter. Then iterate through input and compact the
-  // kv-pairs
-  void ProcessKeyValueCompaction(SubcompactionState* sub_compact);
 
   Status FinishCompactionOutputFile(
       const Status& input_status, SubcompactionState* sub_compact,
@@ -128,33 +153,23 @@ class CompactionJob {
       CompactionIterationStats* range_del_out_stats,
       const Slice* next_table_min_key = nullptr);
   Status InstallCompactionResults(const MutableCFOptions& mutable_cf_options);
-  void RecordCompactionIOStats();
   Status OpenCompactionOutputFile(SubcompactionState* sub_compact);
-  void CleanupCompaction();
   void UpdateCompactionJobStats(
     const InternalStats::CompactionStats& stats) const;
   void RecordDroppedKeys(const CompactionIterationStats& c_iter_stats,
                          CompactionJobStats* compaction_job_stats = nullptr);
 
-  void UpdateCompactionStats();
   void UpdateCompactionInputStatsHelper(
       int* num_files, uint64_t* bytes_read, int input_level);
 
-  void LogCompaction();
-
   int job_id_;
 
-  // CompactionJob state
-  struct CompactionState;
-  CompactionState* compact_;
   CompactionJobStats* compaction_job_stats_;
-  InternalStats::CompactionStats compaction_stats_;
 
   // DBImpl state
   const std::string& dbname_;
   const std::string db_id_;
   const std::string db_session_id_;
-  const ImmutableDBOptions& db_options_;
   const FileOptions file_options_;
 
   Env* env_;
@@ -166,10 +181,8 @@ class CompactionJob {
   const std::atomic<bool>* shutting_down_;
   const std::atomic<int>* manual_compaction_paused_;
   const SequenceNumber preserve_deletes_seqnum_;
-  LogBuffer* log_buffer_;
   FSDirectory* db_directory_;
-  FSDirectory* output_directory_;
-  Statistics* stats_;
+  FSDirectory* blob_output_directory_;
   InstrumentedMutex* db_mutex_;
   ErrorHandler* db_error_handler_;
   // If there were two snapshots with seq numbers s1 and
@@ -189,17 +202,128 @@ class CompactionJob {
 
   EventLogger* event_logger_;
 
-  // Is this compaction creating a file in the bottom most level?
-  bool bottommost_level_;
   bool paranoid_file_checks_;
   bool measure_io_stats_;
   // Stores the Slices that designate the boundaries for each subcompaction
   std::vector<Slice> boundaries_;
   // 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_;
-  IOStatus io_status_;
+  std::string full_history_ts_low_;
+  BlobFileCompletionCallback* blob_callback_;
+
+  // Get table file name in where it's outputting to, which should also be in
+  // `output_directory_`.
+  virtual std::string GetTableFileName(uint64_t file_number);
+};
+
+// CompactionServiceInput is used the pass compaction information between two
+// db instances. It contains the information needed to do a compaction. It
+// doesn't contain the LSM tree information, which is passed though MANIFEST
+// file.
+struct CompactionServiceInput {
+  ColumnFamilyDescriptor column_family;
+
+  DBOptions db_options;
+
+  std::vector<SequenceNumber> snapshots;
+
+  // SST files for compaction, it should already be expended to include all the
+  // files needed for this compaction, for both input level files and output
+  // level files.
+  std::vector<std::string> input_files;
+  int output_level;
+
+  // information for subcompaction
+  Slice* begin = nullptr;
+  Slice* end = nullptr;
+  uint64_t approx_size = 0;
+};
+
+// CompactionServiceOutputFile is the metadata for the output SST file
+struct CompactionServiceOutputFile {
+  std::string file_name;
+  SequenceNumber smallest_seqno;
+  SequenceNumber largest_seqno;
+  std::string smallest_internal_key;
+  std::string largest_internal_key;
+  uint64_t oldest_ancester_time;
+  uint64_t file_creation_time;
+  uint64_t paranoid_hash;
+  bool marked_for_compaction;
+
+  CompactionServiceOutputFile() = default;
+  CompactionServiceOutputFile(
+      const std::string& name, SequenceNumber smallest, SequenceNumber largest,
+      std::string _smallest_internal_key, std::string _largest_internal_key,
+      uint64_t _oldest_ancester_time, uint64_t _file_creation_time,
+      uint64_t _paranoid_hash, bool _marked_for_compaction)
+      : file_name(name),
+        smallest_seqno(smallest),
+        largest_seqno(largest),
+        smallest_internal_key(std::move(_smallest_internal_key)),
+        largest_internal_key(std::move(_largest_internal_key)),
+        oldest_ancester_time(_oldest_ancester_time),
+        file_creation_time(_file_creation_time),
+        paranoid_hash(_paranoid_hash),
+        marked_for_compaction(_marked_for_compaction) {}
+};
+
+// CompactionServiceResult contains the compaction result from a different db
+// instance, with these information, the primary db instance with write
+// permission is able to install the result to the DB.
+struct CompactionServiceResult {
+  std::vector<CompactionServiceOutputFile> output_files;
+  int output_level;
+
+  // location of the output files
+  std::string output_path;
+
+  // some statistics about the compaction
+  uint64_t num_output_records;
+  uint64_t total_bytes;
+  uint64_t bytes_read;
+  uint64_t bytes_written;
+  CompactionJobStats stats;
+};
+
+// CompactionServiceCompactionJob is an read-only compaction job, it takes
+// input information from `compaction_service_input` and put result information
+// in `compaction_service_result`, the SST files are generated to `output_path`.
+class CompactionServiceCompactionJob : private CompactionJob {
+ public:
+  CompactionServiceCompactionJob(
+      int job_id, Compaction* compaction, const ImmutableDBOptions& db_options,
+      const FileOptions& file_options, VersionSet* versions,
+      const std::atomic<bool>* shutting_down, LogBuffer* log_buffer,
+      FSDirectory* output_directory, Statistics* stats,
+      InstrumentedMutex* db_mutex, ErrorHandler* db_error_handler,
+      std::vector<SequenceNumber> existing_snapshots,
+      std::shared_ptr<Cache> table_cache, EventLogger* event_logger,
+      const std::string& dbname, const std::shared_ptr<IOTracer>& io_tracer,
+      const std::string& db_id, const std::string& db_session_id,
+      const std::string& output_path,
+      const CompactionServiceInput& compaction_service_input,
+      CompactionServiceResult* compaction_service_result);
+
+  // Run the compaction in current thread and return the result
+  Status Run();
+
+  void CleanupCompaction();
+
+  IOStatus io_status() const { return CompactionJob::io_status(); }
+
+ private:
+  // Get table file name in output_path
+  std::string GetTableFileName(uint64_t file_number) override;
+  // Specific the compaction output path, otherwise it uses default DB path
+  const std::string output_path_;
+
+  // Compaction job input
+  const CompactionServiceInput& compaction_input_;
+
+  // Compaction job result
+  CompactionServiceResult* compaction_result_;
 };
 
 }  // namespace ROCKSDB_NAMESPACE

db/compaction/compaction_job_stats_test.cc

@@ -24,7 +24,6 @@
 #include "db/write_batch_internal.h"
 #include "env/mock_env.h"
 #include "file/filename.h"
-#include "logging/logging.h"
 #include "memtable/hash_linklist_rep.h"
 #include "monitoring/statistics.h"
 #include "monitoring/thread_status_util.h"
@@ -298,15 +297,14 @@ class CompactionJobStatsTest : public testing::Test,
     return result;
   }
 
-  uint64_t Size(const Slice& start, const Slice& limit, int cf = 0) {
+  Status Size(uint64_t* size, const Slice& start, const Slice& limit,
+              int cf = 0) {
     Range r(start, limit);
-    uint64_t size;
     if (cf == 0) {
-      db_->GetApproximateSizes(&r, 1, &size);
+      return db_->GetApproximateSizes(&r, 1, size);
     } else {
-      db_->GetApproximateSizes(handles_[1], &r, 1, &size);
+      return db_->GetApproximateSizes(handles_[1], &r, 1, size);
     }
-    return size;
   }
 
   void Compact(int cf, const Slice& start, const Slice& limit,

db/compaction/compaction_job_test.cc

@@ -5,6 +5,8 @@
 
 #ifndef ROCKSDB_LITE
 
+#include "db/compaction/compaction_job.h"
+
 #include <algorithm>
 #include <array>
 #include <cinttypes>
@@ -14,13 +16,13 @@
 
 #include "db/blob/blob_index.h"
 #include "db/column_family.h"
-#include "db/compaction/compaction_job.h"
 #include "db/db_impl/db_impl.h"
 #include "db/error_handler.h"
 #include "db/version_set.h"
 #include "file/writable_file_writer.h"
 #include "rocksdb/cache.h"
 #include "rocksdb/db.h"
+#include "rocksdb/file_system.h"
 #include "rocksdb/options.h"
 #include "rocksdb/write_buffer_manager.h"
 #include "table/mock_table.h"
@@ -67,13 +69,14 @@ void VerifyInitializationOfCompactionJobStats(
 
 }  // namespace
 
-// TODO(icanadi) Make it simpler once we mock out VersionSet
-class CompactionJobTest : public testing::Test {
- public:
-  CompactionJobTest()
+class CompactionJobTestBase : public testing::Test {
+ protected:
+  CompactionJobTestBase(std::string dbname, const Comparator* ucmp,
+                        std::function<std::string(uint64_t)> encode_u64_ts)
       : env_(Env::Default()),
-        fs_(std::make_shared<LegacyFileSystemWrapper>(env_)),
-        dbname_(test::PerThreadDBPath("compaction_job_test")),
+        fs_(env_->GetFileSystem()),
+        dbname_(std::move(dbname)),
+        ucmp_(ucmp),
         db_options_(),
         mutable_cf_options_(cf_options_),
         mutable_db_options_(),
@@ -86,12 +89,17 @@ 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_(nullptr, db_options_, &mutex_),
+        encode_u64_ts_(std::move(encode_u64_ts)) {}
+
+  void SetUp() override {
     EXPECT_OK(env_->CreateDirIfMissing(dbname_));
     db_options_.env = env_;
     db_options_.fs = fs_;
     db_options_.db_paths.emplace_back(dbname_,
                                       std::numeric_limits<uint64_t>::max());
+    cf_options_.comparator = ucmp_;
+    cf_options_.table_factory = mock_table_factory_;
   }
 
   std::string GenerateFileName(uint64_t file_number) {
@@ -102,9 +110,10 @@ class CompactionJobTest : public testing::Test {
     return TableFileName(db_paths, meta.fd.GetNumber(), meta.fd.GetPathId());
   }
 
-  static std::string KeyStr(const std::string& user_key,
-                            const SequenceNumber seq_num, const ValueType t) {
-    return InternalKey(user_key, seq_num, t).Encode().ToString();
+  std::string KeyStr(const std::string& user_key, const SequenceNumber seq_num,
+                     const ValueType t, uint64_t ts = 0) {
+    std::string user_key_with_ts = user_key + encode_u64_ts_(ts);
+    return InternalKey(user_key_with_ts, seq_num, t).Encode().ToString();
   }
 
   static std::string BlobStr(uint64_t blob_file_number, uint64_t offset,
@@ -144,7 +153,8 @@ class CompactionJobTest : public testing::Test {
       std::string skey;
       std::string value;
       std::tie(skey, value) = kv;
-      const Status pikStatus = ParseInternalKey(skey, &key);
+      const Status pik_status =
+          ParseInternalKey(skey, &key, true /* log_err_key */);
 
       smallest_seqno = std::min(smallest_seqno, key.sequence);
       largest_seqno = std::max(largest_seqno, key.sequence);
@@ -162,7 +172,7 @@ class CompactionJobTest : public testing::Test {
 
       first_key = false;
 
-      if (pikStatus.ok() && key.type == kTypeBlobIndex) {
+      if (pik_status.ok() && key.type == kTypeBlobIndex) {
         BlobIndex blob_index;
         const Status s = blob_index.DecodeFrom(value);
         if (!s.ok()) {
@@ -207,9 +217,9 @@ class CompactionJobTest : public testing::Test {
   // returns expected result after compaction
   mock::KVVector CreateTwoFiles(bool gen_corrupted_keys) {
     stl_wrappers::KVMap expected_results;
-    const int kKeysPerFile = 10000;
-    const int kCorruptKeysPerFile = 200;
-    const int kMatchingKeys = kKeysPerFile / 2;
+    constexpr int kKeysPerFile = 10000;
+    constexpr int kCorruptKeysPerFile = 200;
+    constexpr int kMatchingKeys = kKeysPerFile / 2;
     SequenceNumber sequence_number = 0;
 
     auto corrupt_id = [&](int id) {
@@ -238,7 +248,7 @@ class CompactionJobTest : public testing::Test {
               {bottommost_internal_key.Encode().ToString(), value});
         }
       }
-      mock::SortKVVector(&contents);
+      mock::SortKVVector(&contents, ucmp_);
 
       AddMockFile(contents);
     }
@@ -254,33 +264,28 @@ class CompactionJobTest : public testing::Test {
   }
 
   void NewDB() {
-    DestroyDB(dbname_, Options());
+    EXPECT_OK(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_,
                        /*block_cache_tracer=*/nullptr, /*io_tracer=*/nullptr));
     compaction_job_stats_.Reset();
-    SetIdentityFile(env_, dbname_);
+    ASSERT_OK(SetIdentityFile(env_, dbname_));
 
     VersionEdit new_db;
-    if (db_options_.write_dbid_to_manifest) {
-      DBImpl* impl = new DBImpl(DBOptions(), dbname_);
-      std::string db_id;
-      impl->GetDbIdentityFromIdentityFile(&db_id);
-      new_db.SetDBId(db_id);
-    }
     new_db.SetLogNumber(0);
     new_db.SetNextFile(2);
     new_db.SetLastSequence(0);
 
     const std::string manifest = DescriptorFileName(dbname_, 1);
-    std::unique_ptr<WritableFile> file;
-    Status s = env_->NewWritableFile(
-        manifest, &file, env_->OptimizeForManifestWrite(env_options_));
+    std::unique_ptr<WritableFileWriter> file_writer;
+    const auto& fs = env_->GetFileSystem();
+    Status s = WritableFileWriter::Create(
+        fs, manifest, fs->OptimizeForManifestWrite(env_options_), &file_writer,
+        nullptr);
+
     ASSERT_OK(s);
-    std::unique_ptr<WritableFileWriter> file_writer(new WritableFileWriter(
-        NewLegacyWritableFileWrapper(std::move(file)), manifest, env_options_));
     {
       log::Writer log(std::move(file_writer), 0, false);
       std::string record;
@@ -293,13 +298,12 @@ class CompactionJobTest : public testing::Test {
 
     ASSERT_OK(s);
 
-    std::vector<ColumnFamilyDescriptor> column_families;
-    cf_options_.table_factory = mock_table_factory_;
     cf_options_.merge_operator = merge_op_;
     cf_options_.compaction_filter = compaction_filter_.get();
+    std::vector<ColumnFamilyDescriptor> column_families;
     column_families.emplace_back(kDefaultColumnFamilyName, cf_options_);
 
-    EXPECT_OK(versions_->Recover(column_families, false));
+    ASSERT_OK(versions_->Recover(column_families, false));
     cfd_ = versions_->GetColumnFamilySet()->GetDefault();
   }
 
@@ -337,19 +341,22 @@ class CompactionJobTest : public testing::Test {
     EventLogger event_logger(db_options_.info_log.get());
     // TODO(yiwu) add a mock snapshot checker and add test for it.
     SnapshotChecker* snapshot_checker = nullptr;
+    ASSERT_TRUE(full_history_ts_low_.empty() ||
+                ucmp_->timestamp_size() == full_history_ts_low_.size());
     CompactionJob compaction_job(
         0, &compaction, db_options_, env_options_, versions_.get(),
         &shutting_down_, preserve_deletes_seqnum_, &log_buffer, nullptr,
-        nullptr, nullptr, &mutex_, &error_handler_, snapshots,
+        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, nullptr /* IOTracer */);
+        Env::Priority::USER, nullptr /* IOTracer */,
+        /*manual_compaction_paused=*/nullptr, /*db_id=*/"",
+        /*db_session_id=*/"", full_history_ts_low_);
     VerifyInitializationOfCompactionJobStats(compaction_job_stats_);
 
     compaction_job.Prepare();
     mutex_.Unlock();
-    Status s;
-    s = compaction_job.Run();
+    Status s = compaction_job.Run();
     ASSERT_OK(s);
     ASSERT_OK(compaction_job.io_status());
     mutex_.Lock();
@@ -379,6 +386,7 @@ class CompactionJobTest : public testing::Test {
   Env* env_;
   std::shared_ptr<FileSystem> fs_;
   std::string dbname_;
+  const Comparator* const ucmp_;
   EnvOptions env_options_;
   ImmutableDBOptions db_options_;
   ColumnFamilyOptions cf_options_;
@@ -397,6 +405,17 @@ class CompactionJobTest : public testing::Test {
   std::unique_ptr<CompactionFilter> compaction_filter_;
   std::shared_ptr<MergeOperator> merge_op_;
   ErrorHandler error_handler_;
+  std::string full_history_ts_low_;
+  const std::function<std::string(uint64_t)> encode_u64_ts_;
+};
+
+// TODO(icanadi) Make it simpler once we mock out VersionSet
+class CompactionJobTest : public CompactionJobTestBase {
+ public:
+  CompactionJobTest()
+      : CompactionJobTestBase(test::PerThreadDBPath("compaction_job_test"),
+                              BytewiseComparator(),
+                              [](uint64_t /*ts*/) { return ""; }) {}
 };
 
 TEST_F(CompactionJobTest, Simple) {
@@ -1077,6 +1096,118 @@ TEST_F(CompactionJobTest, OldestBlobFileNumber) {
                 /* expected_oldest_blob_file_number */ 19);
 }
 
+class CompactionJobTimestampTest : public CompactionJobTestBase {
+ public:
+  CompactionJobTimestampTest()
+      : CompactionJobTestBase(test::PerThreadDBPath("compaction_job_ts_test"),
+                              test::ComparatorWithU64Ts(), test::EncodeInt) {}
+};
+
+TEST_F(CompactionJobTimestampTest, GCDisabled) {
+  NewDB();
+
+  auto file1 =
+      mock::MakeMockFile({{KeyStr("a", 10, ValueType::kTypeValue, 100), "a10"},
+                          {KeyStr("a", 9, ValueType::kTypeValue, 99), "a9"},
+                          {KeyStr("b", 8, ValueType::kTypeValue, 98), "b8"}});
+  AddMockFile(file1);
+
+  auto file2 = mock::MakeMockFile(
+      {{KeyStr("b", 7, ValueType::kTypeDeletionWithTimestamp, 97), ""},
+       {KeyStr("c", 6, ValueType::kTypeDeletionWithTimestamp, 96), ""},
+       {KeyStr("c", 5, ValueType::kTypeValue, 95), "c5"}});
+  AddMockFile(file2);
+
+  SetLastSequence(10);
+
+  auto expected_results = mock::MakeMockFile(
+      {{KeyStr("a", 10, ValueType::kTypeValue, 100), "a10"},
+       {KeyStr("a", 9, ValueType::kTypeValue, 99), "a9"},
+       {KeyStr("b", 8, ValueType::kTypeValue, 98), "b8"},
+       {KeyStr("b", 7, ValueType::kTypeDeletionWithTimestamp, 97), ""},
+       {KeyStr("c", 6, ValueType::kTypeDeletionWithTimestamp, 96), ""},
+       {KeyStr("c", 5, ValueType::kTypeValue, 95), "c5"}});
+  const auto& files = cfd_->current()->storage_info()->LevelFiles(0);
+  RunCompaction({files}, expected_results);
+}
+
+TEST_F(CompactionJobTimestampTest, NoKeyExpired) {
+  NewDB();
+
+  auto file1 =
+      mock::MakeMockFile({{KeyStr("a", 6, ValueType::kTypeValue, 100), "a6"},
+                          {KeyStr("b", 7, ValueType::kTypeValue, 101), "b7"},
+                          {KeyStr("c", 5, ValueType::kTypeValue, 99), "c5"}});
+  AddMockFile(file1);
+
+  auto file2 =
+      mock::MakeMockFile({{KeyStr("a", 4, ValueType::kTypeValue, 98), "a4"},
+                          {KeyStr("c", 3, ValueType::kTypeValue, 97), "c3"}});
+  AddMockFile(file2);
+
+  SetLastSequence(101);
+
+  auto expected_results =
+      mock::MakeMockFile({{KeyStr("a", 6, ValueType::kTypeValue, 100), "a6"},
+                          {KeyStr("a", 4, ValueType::kTypeValue, 98), "a4"},
+                          {KeyStr("b", 7, ValueType::kTypeValue, 101), "b7"},
+                          {KeyStr("c", 5, ValueType::kTypeValue, 99), "c5"},
+                          {KeyStr("c", 3, ValueType::kTypeValue, 97), "c3"}});
+  const auto& files = cfd_->current()->storage_info()->LevelFiles(0);
+
+  full_history_ts_low_ = encode_u64_ts_(0);
+  RunCompaction({files}, expected_results);
+}
+
+TEST_F(CompactionJobTimestampTest, AllKeysExpired) {
+  NewDB();
+
+  auto file1 = mock::MakeMockFile(
+      {{KeyStr("a", 5, ValueType::kTypeDeletionWithTimestamp, 100), ""},
+       {KeyStr("b", 6, ValueType::kTypeValue, 99), "b6"}});
+  AddMockFile(file1);
+
+  auto file2 = mock::MakeMockFile(
+      {{KeyStr("a", 4, ValueType::kTypeValue, 98), "a4"},
+       {KeyStr("b", 3, ValueType::kTypeDeletionWithTimestamp, 97), ""},
+       {KeyStr("b", 2, ValueType::kTypeValue, 96), "b2"}});
+  AddMockFile(file2);
+
+  SetLastSequence(6);
+
+  auto expected_results =
+      mock::MakeMockFile({{KeyStr("b", 0, ValueType::kTypeValue, 0), "b6"}});
+  const auto& files = cfd_->current()->storage_info()->LevelFiles(0);
+
+  full_history_ts_low_ = encode_u64_ts_(std::numeric_limits<uint64_t>::max());
+  RunCompaction({files}, expected_results);
+}
+
+TEST_F(CompactionJobTimestampTest, SomeKeysExpired) {
+  NewDB();
+
+  auto file1 =
+      mock::MakeMockFile({{KeyStr("a", 5, ValueType::kTypeValue, 50), "a5"},
+                          {KeyStr("b", 6, ValueType::kTypeValue, 49), "b6"}});
+  AddMockFile(file1);
+
+  auto file2 = mock::MakeMockFile(
+      {{KeyStr("a", 3, ValueType::kTypeValue, 48), "a3"},
+       {KeyStr("a", 2, ValueType::kTypeValue, 46), "a2"},
+       {KeyStr("b", 4, ValueType::kTypeDeletionWithTimestamp, 47), ""}});
+  AddMockFile(file2);
+
+  SetLastSequence(6);
+
+  auto expected_results =
+      mock::MakeMockFile({{KeyStr("a", 5, ValueType::kTypeValue, 50), "a5"},
+                          {KeyStr("b", 6, ValueType::kTypeValue, 49), "b6"}});
+  const auto& files = cfd_->current()->storage_info()->LevelFiles(0);
+
+  full_history_ts_low_ = encode_u64_ts_(49);
+  RunCompaction({files}, expected_results);
+}
+
 }  // namespace ROCKSDB_NAMESPACE
 
 int main(int argc, char** argv) {

db/compaction/compaction_picker.cc

@@ -139,11 +139,9 @@ CompressionOptions GetCompressionOptions(const MutableCFOptions& cf_options,
   if (!enable_compression) {
     return cf_options.compression_opts;
   }
-  // If bottommost_compression is set and we are compacting to the
-  // bottommost level then we should use the specified compression options
-  // for the bottmomost_compression.
-  if (cf_options.bottommost_compression != kDisableCompressionOption &&
-      level >= (vstorage->num_non_empty_levels() - 1) &&
+  // If bottommost_compression_opts is enabled and we are compacting to the
+  // bottommost level then we should use the specified compression options.
+  if (level >= (vstorage->num_non_empty_levels() - 1) &&
       cf_options.bottommost_compression_opts.enabled) {
     return cf_options.bottommost_compression_opts;
   }
@@ -1006,6 +1004,7 @@ Status CompactionPicker::SanitizeCompactionInputFiles(
   // any currently-existing files.
   for (auto file_num : *input_files) {
     bool found = false;
+    int input_file_level = -1;
     for (const auto& level_meta : cf_meta.levels) {
       for (const auto& file_meta : level_meta.files) {
         if (file_num == TableFileNameToNumber(file_meta.name)) {
@@ -1015,6 +1014,7 @@ Status CompactionPicker::SanitizeCompactionInputFiles(
                                    " is already being compacted.");
           }
           found = true;
+          input_file_level = level_meta.level;
           break;
         }
       }
@@ -1027,6 +1027,13 @@ Status CompactionPicker::SanitizeCompactionInputFiles(
           "Specified compaction input file " + MakeTableFileName("", file_num) +
           " does not exist in column family " + cf_meta.name + ".");
     }
+    if (input_file_level > output_level) {
+      return Status::InvalidArgument(
+          "Cannot compact file to up level, input file: " +
+          MakeTableFileName("", file_num) + " level " +
+          ToString(input_file_level) + " > output level " +
+          ToString(output_level));
+    }
   }
 
   return Status::OK();
@@ -1045,6 +1052,8 @@ void CompactionPicker::RegisterCompaction(Compaction* c) {
     level0_compactions_in_progress_.insert(c);
   }
   compactions_in_progress_.insert(c);
+  TEST_SYNC_POINT_CALLBACK("CompactionPicker::RegisterCompaction:Registered",
+                           c);
 }
 
 void CompactionPicker::UnregisterCompaction(Compaction* c) {

db/compaction/compaction_picker_fifo.cc

@@ -45,7 +45,7 @@ Compaction* FIFOCompactionPicker::PickTTLCompaction(
   uint64_t total_size = GetTotalFilesSize(level_files);
 
   int64_t _current_time;
-  auto status = ioptions_.env->GetCurrentTime(&_current_time);
+  auto status = ioptions_.clock->GetCurrentTime(&_current_time);
   if (!status.ok()) {
     ROCKS_LOG_BUFFER(log_buffer,
                      "[%s] FIFO compaction: Couldn't get current time: %s. "

db/compaction/compaction_picker_test.cc

@@ -12,7 +12,6 @@
 #include "db/compaction/compaction_picker_level.h"
 #include "db/compaction/compaction_picker_universal.h"
 
-#include "logging/logging.h"
 #include "test_util/testharness.h"
 #include "test_util/testutil.h"
 #include "util/string_util.h"
@@ -142,7 +141,7 @@ class CompactionPickerTest : public testing::Test {
     if (temp_vstorage_) {
       VersionBuilder builder(FileOptions(), &ioptions_, nullptr,
                              vstorage_.get(), nullptr);
-      builder.SaveTo(temp_vstorage_.get());
+      ASSERT_OK(builder.SaveTo(temp_vstorage_.get()));
       vstorage_ = std::move(temp_vstorage_);
     }
     vstorage_->CalculateBaseBytes(ioptions_, mutable_cf_options_);
@@ -651,7 +650,7 @@ TEST_F(CompactionPickerTest, UniversalPeriodicCompaction3) {
 
 TEST_F(CompactionPickerTest, UniversalPeriodicCompaction4) {
   // The case where universal periodic compaction couldn't form
-  // a compaction that inlcudes any file marked for periodic compaction.
+  // a compaction that includes any file marked for periodic compaction.
   // Right now we form the compaction anyway if it is more than one
   // sorted run. Just put the case here to validate that it doesn't
   // crash.
@@ -801,7 +800,7 @@ TEST_F(CompactionPickerTest, CompactionPriMinOverlapping2) {
   Add(2, 6U, "150", "175",
       60000000U);  // Overlaps with file 26, 27, total size 521M
   Add(2, 7U, "176", "200", 60000000U);  // Overlaps with file 27, 28, total size
-                                        // 520M, the smalelst overlapping
+                                        // 520M, the smallest overlapping
   Add(2, 8U, "201", "300",
       60000000U);  // Overlaps with file 28, 29, total size 521M
 
@@ -1229,7 +1228,7 @@ TEST_F(CompactionPickerTest, NotScheduleL1IfL0WithHigherPri1) {
   Add(0, 32U, "001", "400", 1000000000U, 0, 0);
   Add(0, 33U, "001", "400", 1000000000U, 0, 0);
 
-  // L1 total size 2GB, score 2.2. If one file being comapcted, score 1.1.
+  // L1 total size 2GB, score 2.2. If one file being compacted, score 1.1.
   Add(1, 4U, "050", "300", 1000000000U, 0, 0);
   file_map_[4u].first->being_compacted = true;
   Add(1, 5U, "301", "350", 1000000000U, 0, 0);
@@ -1262,7 +1261,7 @@ TEST_F(CompactionPickerTest, NotScheduleL1IfL0WithHigherPri2) {
   Add(0, 32U, "001", "400", 1000000000U, 0, 0);
   Add(0, 33U, "001", "400", 1000000000U, 0, 0);
 
-  // L1 total size 2GB, score 2.2. If one file being comapcted, score 1.1.
+  // L1 total size 2GB, score 2.2. If one file being compacted, score 1.1.
   Add(1, 4U, "050", "300", 1000000000U, 0, 0);
   Add(1, 5U, "301", "350", 1000000000U, 0, 0);
 

db/compaction/compaction_picker_universal.cc

@@ -733,7 +733,7 @@ Compaction* UniversalCompactionBuilder::PickCompactionToReduceSortedRuns(
 }
 
 // Look at overall size amplification. If size amplification
-// exceeeds the configured value, then do a compaction
+// exceeds the configured value, then do a compaction
 // of the candidate files all the way upto the earliest
 // base file (overrides configured values of file-size ratios,
 // min_merge_width and max_merge_width).

db/comparator_db_test.cc

@@ -17,7 +17,6 @@
 #include "util/string_util.h"
 #include "utilities/merge_operators.h"
 
-using std::unique_ptr;
 
 namespace ROCKSDB_NAMESPACE {
 namespace {

db/corruption_test.cc

@@ -9,7 +9,6 @@
 
 #ifndef ROCKSDB_LITE
 
-#include <errno.h>
 #include <fcntl.h>
 #include <sys/stat.h>
 #include <sys/types.h>
@@ -20,8 +19,8 @@
 #include "db/db_test_util.h"
 #include "db/log_format.h"
 #include "db/version_set.h"
-#include "env/composite_env_wrapper.h"
 #include "file/filename.h"
+#include "port/stack_trace.h"
 #include "rocksdb/cache.h"
 #include "rocksdb/convenience.h"
 #include "rocksdb/db.h"
@@ -33,16 +32,18 @@
 #include "table/mock_table.h"
 #include "test_util/testharness.h"
 #include "test_util/testutil.h"
+#include "util/cast_util.h"
 #include "util/random.h"
 #include "util/string_util.h"
 
 namespace ROCKSDB_NAMESPACE {
 
-static const int kValueSize = 1000;
+static constexpr int kValueSize = 1000;
 
 class CorruptionTest : public testing::Test {
  public:
-  test::ErrorEnv env_;
+  std::shared_ptr<Env> env_guard_;
+  test::ErrorEnv* env_;
   std::string dbname_;
   std::shared_ptr<Cache> tiny_cache_;
   Options options_;
@@ -53,9 +54,19 @@ class CorruptionTest : public testing::Test {
     // set it to 0), test SequenceNumberRecovery will fail, likely because of a
     // bug in recovery code. Keep it 4 for now to make the test passes.
     tiny_cache_ = NewLRUCache(100, 4);
+    Env* base_env = Env::Default();
+#ifndef ROCKSDB_LITE
+    const char* test_env_uri = getenv("TEST_ENV_URI");
+    if (test_env_uri) {
+      Status s = Env::LoadEnv(test_env_uri, &base_env, &env_guard_);
+      EXPECT_OK(s);
+      EXPECT_NE(Env::Default(), base_env);
+    }
+#endif  //! ROCKSDB_LITE
+    env_ = new test::ErrorEnv(base_env);
     options_.wal_recovery_mode = WALRecoveryMode::kTolerateCorruptedTailRecords;
-    options_.env = &env_;
-    dbname_ = test::PerThreadDBPath("corruption_test");
+    options_.env = env_;
+    dbname_ = test::PerThreadDBPath(env_, "corruption_test");
     Status s = DestroyDB(dbname_, options_);
     EXPECT_OK(s);
 
@@ -69,9 +80,19 @@ class CorruptionTest : public testing::Test {
   }
 
   ~CorruptionTest() override {
+    SyncPoint::GetInstance()->DisableProcessing();
+    SyncPoint::GetInstance()->LoadDependency({});
+    SyncPoint::GetInstance()->ClearAllCallBacks();
     delete db_;
     db_ = nullptr;
-    DestroyDB(dbname_, Options());
+    if (getenv("KEEP_DB")) {
+      fprintf(stdout, "db is still at %s\n", dbname_.c_str());
+    } else {
+      Options opts;
+      opts.env = env_->target();
+      EXPECT_OK(DestroyDB(dbname_, opts));
+    }
+    delete env_;
   }
 
   void CloseDb() {
@@ -86,7 +107,7 @@ class CorruptionTest : public testing::Test {
     if (opt.env == Options().env) {
       // If env is not overridden, replace it with ErrorEnv.
       // Otherwise, the test already uses a non-default Env.
-      opt.env = &env_;
+      opt.env = env_;
     }
     opt.arena_block_size = 4096;
     BlockBasedTableOptions table_options;
@@ -169,7 +190,7 @@ class CorruptionTest : public testing::Test {
   void Corrupt(FileType filetype, int offset, int bytes_to_corrupt) {
     // Pick file to corrupt
     std::vector<std::string> filenames;
-    ASSERT_OK(env_.GetChildren(dbname_, &filenames));
+    ASSERT_OK(env_->GetChildren(dbname_, &filenames));
     uint64_t number;
     FileType type;
     std::string fname;
@@ -184,7 +205,7 @@ class CorruptionTest : public testing::Test {
     }
     ASSERT_TRUE(!fname.empty()) << filetype;
 
-    test::CorruptFile(fname, offset, bytes_to_corrupt);
+    ASSERT_OK(test::CorruptFile(env_, fname, offset, bytes_to_corrupt));
   }
 
   // corrupts exactly one file at level `level`. if no file found at level,
@@ -194,7 +215,8 @@ class CorruptionTest : public testing::Test {
     db_->GetLiveFilesMetaData(&metadata);
     for (const auto& m : metadata) {
       if (m.level == level) {
-        test::CorruptFile(dbname_ + "/" + m.name, offset, bytes_to_corrupt);
+        ASSERT_OK(test::CorruptFile(env_, dbname_ + "/" + m.name, offset,
+                                    bytes_to_corrupt));
         return;
       }
     }
@@ -249,8 +271,8 @@ TEST_F(CorruptionTest, Recovery) {
   // is not available for WAL though.
   CloseDb();
 #endif
-  Corrupt(kLogFile, 19, 1);      // WriteBatch tag for first record
-  Corrupt(kLogFile, log::kBlockSize + 1000, 1);  // Somewhere in second block
+  Corrupt(kWalFile, 19, 1);  // WriteBatch tag for first record
+  Corrupt(kWalFile, log::kBlockSize + 1000, 1);  // Somewhere in second block
   ASSERT_TRUE(!TryReopen().ok());
   options_.paranoid_checks = false;
   Reopen(&options_);
@@ -260,14 +282,14 @@ TEST_F(CorruptionTest, Recovery) {
 }
 
 TEST_F(CorruptionTest, RecoverWriteError) {
-  env_.writable_file_error_ = true;
+  env_->writable_file_error_ = true;
   Status s = TryReopen();
   ASSERT_TRUE(!s.ok());
 }
 
 TEST_F(CorruptionTest, NewFileErrorDuringWrite) {
   // Do enough writing to force minor compaction
-  env_.writable_file_error_ = true;
+  env_->writable_file_error_ = true;
   const int num =
       static_cast<int>(3 + (Options().write_buffer_size / kValueSize));
   std::string value_storage;
@@ -283,8 +305,8 @@ TEST_F(CorruptionTest, NewFileErrorDuringWrite) {
     ASSERT_TRUE(!failed || !s.ok());
   }
   ASSERT_TRUE(!s.ok());
-  ASSERT_GE(env_.num_writable_file_errors_, 1);
-  env_.writable_file_error_ = false;
+  ASSERT_GE(env_->num_writable_file_errors_, 1);
+  env_->writable_file_error_ = false;
   Reopen();
 }
 
@@ -302,7 +324,7 @@ TEST_F(CorruptionTest, TableFile) {
 
 TEST_F(CorruptionTest, VerifyChecksumReadahead) {
   Options options;
-  SpecialEnv senv(Env::Default());
+  SpecialEnv senv(env_->target());
   options.env = &senv;
   // Disable block cache as we are going to check checksum for
   // the same file twice and measure number of reads.
@@ -424,6 +446,7 @@ TEST_F(CorruptionTest, CorruptedDescriptor) {
 
 TEST_F(CorruptionTest, CompactionInputError) {
   Options options;
+  options.env = env_;
   Reopen(&options);
   Build(10);
   DBImpl* dbi = static_cast_with_check<DBImpl>(db_);
@@ -444,6 +467,7 @@ TEST_F(CorruptionTest, CompactionInputError) {
 
 TEST_F(CorruptionTest, CompactionInputErrorParanoid) {
   Options options;
+  options.env = env_;
   options.paranoid_checks = true;
   options.write_buffer_size = 131072;
   options.max_write_buffer_number = 2;
@@ -514,14 +538,15 @@ TEST_F(CorruptionTest, RangeDeletionCorrupted) {
   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(NewLegacyRandomAccessFileWrapper(file),
-                                 filename));
+  FileOptions file_opts;
+  const auto& fs = options_.env->GetFileSystem();
+  std::unique_ptr<RandomAccessFileReader> file_reader;
+  ASSERT_OK(RandomAccessFileReader::Create(fs, filename, file_opts,
+                                           &file_reader, nullptr));
 
   uint64_t file_size;
-  ASSERT_OK(options_.env->GetFileSize(filename, &file_size));
+  ASSERT_OK(
+      fs->GetFileSize(filename, file_opts.io_options, &file_size, nullptr));
 
   BlockHandle range_del_handle;
   ASSERT_OK(FindMetaBlock(
@@ -529,13 +554,15 @@ TEST_F(CorruptionTest, RangeDeletionCorrupted) {
       ImmutableCFOptions(options_), kRangeDelBlock, &range_del_handle));
 
   ASSERT_OK(TryReopen());
-  test::CorruptFile(filename, static_cast<int>(range_del_handle.offset()), 1);
+  ASSERT_OK(test::CorruptFile(env_, filename,
+                              static_cast<int>(range_del_handle.offset()), 1));
   ASSERT_TRUE(TryReopen().IsCorruption());
 }
 
 TEST_F(CorruptionTest, FileSystemStateCorrupted) {
   for (int iter = 0; iter < 2; ++iter) {
     Options options;
+    options.env = env_;
     options.paranoid_checks = true;
     options.create_if_missing = true;
     Reopen(&options);
@@ -544,7 +571,7 @@ TEST_F(CorruptionTest, FileSystemStateCorrupted) {
     DBImpl* dbi = static_cast_with_check<DBImpl>(db_);
     std::vector<LiveFileMetaData> metadata;
     dbi->GetLiveFilesMetaData(&metadata);
-    ASSERT_GT(metadata.size(), size_t(0));
+    ASSERT_GT(metadata.size(), 0);
     std::string filename = dbname_ + metadata[0].name;
 
     delete db_;
@@ -552,15 +579,15 @@ TEST_F(CorruptionTest, FileSystemStateCorrupted) {
 
     if (iter == 0) {  // corrupt file size
       std::unique_ptr<WritableFile> file;
-      env_.NewWritableFile(filename, &file, EnvOptions());
+      ASSERT_OK(env_->NewWritableFile(filename, &file, EnvOptions()));
       ASSERT_OK(file->Append(Slice("corrupted sst")));
       file.reset();
       Status x = TryReopen(&options);
       ASSERT_TRUE(x.IsCorruption());
     } else {  // delete the file
-      ASSERT_OK(env_.DeleteFile(filename));
+      ASSERT_OK(env_->DeleteFile(filename));
       Status x = TryReopen(&options);
-      ASSERT_TRUE(x.IsPathNotFound());
+      ASSERT_TRUE(x.IsCorruption());
     }
 
     ASSERT_OK(DestroyDB(dbname_, options_));
@@ -574,6 +601,7 @@ static const auto& corruption_modes = {
 
 TEST_F(CorruptionTest, ParanoidFileChecksOnFlush) {
   Options options;
+  options.env = env_;
   options.check_flush_compaction_key_order = false;
   options.paranoid_file_checks = true;
   options.create_if_missing = true;
@@ -588,7 +616,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksOnFlush) {
     options.table_factory = mock;
     mock->SetCorruptionMode(mode);
     ASSERT_OK(DB::Open(options, dbname_, &db_));
-    assert(db_ != nullptr);
+    assert(db_ != nullptr);  // suppress false clang-analyze report
     Build(10);
     s = db_->Flush(FlushOptions());
     if (mode == mock::MockTableFactory::kCorruptNone) {
@@ -601,6 +629,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksOnFlush) {
 
 TEST_F(CorruptionTest, ParanoidFileChecksOnCompact) {
   Options options;
+  options.env = env_;
   options.paranoid_file_checks = true;
   options.create_if_missing = true;
   options.check_flush_compaction_key_order = false;
@@ -613,7 +642,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksOnCompact) {
         std::make_shared<mock::MockTableFactory>();
     options.table_factory = mock;
     ASSERT_OK(DB::Open(options, dbname_, &db_));
-    assert(db_ != nullptr);
+    assert(db_ != nullptr);  // suppress false clang-analyze report
     Build(100, 2);
     // ASSERT_OK(db_->Flush(FlushOptions()));
     DBImpl* dbi = static_cast_with_check<DBImpl>(db_);
@@ -630,6 +659,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksOnCompact) {
 
 TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRangeFirst) {
   Options options;
+  options.env = env_;
   options.check_flush_compaction_key_order = false;
   options.paranoid_file_checks = true;
   options.create_if_missing = true;
@@ -639,7 +669,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRangeFirst) {
     ASSERT_OK(DestroyDB(dbname_, options));
     ASSERT_OK(DB::Open(options, dbname_, &db_));
     std::string start, end;
-    assert(db_ != nullptr);
+    assert(db_ != nullptr);  // suppress false clang-analyze report
     ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(),
                                Key(3, &start), Key(7, &end)));
     auto snap = db_->GetSnapshot();
@@ -662,6 +692,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRangeFirst) {
 
 TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRange) {
   Options options;
+  options.env = env_;
   options.check_flush_compaction_key_order = false;
   options.paranoid_file_checks = true;
   options.create_if_missing = true;
@@ -670,7 +701,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRange) {
     db_ = nullptr;
     ASSERT_OK(DestroyDB(dbname_, options));
     ASSERT_OK(DB::Open(options, dbname_, &db_));
-    assert(db_ != nullptr);
+    assert(db_ != nullptr);  // suppress false clang-analyze report
     Build(10, 0, 0);
     std::string start, end;
     ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(),
@@ -697,6 +728,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRange) {
 
 TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRangeLast) {
   Options options;
+  options.env = env_;
   options.check_flush_compaction_key_order = false;
   options.paranoid_file_checks = true;
   options.create_if_missing = true;
@@ -705,7 +737,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRangeLast) {
     db_ = nullptr;
     ASSERT_OK(DestroyDB(dbname_, options));
     ASSERT_OK(DB::Open(options, dbname_, &db_));
-    assert(db_ != nullptr);
+    assert(db_ != nullptr);  // suppress false clang-analyze report
     std::string start, end;
     Build(10);
     ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(),
@@ -729,6 +761,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRangeLast) {
 
 TEST_F(CorruptionTest, LogCorruptionErrorsInCompactionIterator) {
   Options options;
+  options.env = env_;
   options.create_if_missing = true;
   options.allow_data_in_errors = true;
   auto mode = mock::MockTableFactory::kCorruptKey;
@@ -742,7 +775,7 @@ TEST_F(CorruptionTest, LogCorruptionErrorsInCompactionIterator) {
   options.table_factory = mock;
 
   ASSERT_OK(DB::Open(options, dbname_, &db_));
-  assert(db_ != nullptr);
+  assert(db_ != nullptr);  // suppress false clang-analyze report
   Build(100, 2);
 
   DBImpl* dbi = static_cast_with_check<DBImpl>(db_);
@@ -754,6 +787,7 @@ TEST_F(CorruptionTest, LogCorruptionErrorsInCompactionIterator) {
 
 TEST_F(CorruptionTest, CompactionKeyOrderCheck) {
   Options options;
+  options.env = env_;
   options.paranoid_file_checks = false;
   options.create_if_missing = true;
   options.check_flush_compaction_key_order = false;
@@ -764,7 +798,7 @@ TEST_F(CorruptionTest, CompactionKeyOrderCheck) {
       std::make_shared<mock::MockTableFactory>();
   options.table_factory = mock;
   ASSERT_OK(DB::Open(options, dbname_, &db_));
-  assert(db_ != nullptr);
+  assert(db_ != nullptr);  // suppress false clang-analyze report
   mock->SetCorruptionMode(mock::MockTableFactory::kCorruptReorderKey);
   Build(100, 2);
   DBImpl* dbi = static_cast_with_check<DBImpl>(db_);
@@ -777,6 +811,7 @@ TEST_F(CorruptionTest, CompactionKeyOrderCheck) {
 
 TEST_F(CorruptionTest, FlushKeyOrderCheck) {
   Options options;
+  options.env = env_;
   options.paranoid_file_checks = false;
   options.create_if_missing = true;
   ASSERT_OK(db_->SetOptions({{"check_flush_compaction_key_order", "true"}}));
@@ -805,7 +840,6 @@ TEST_F(CorruptionTest, FlushKeyOrderCheck) {
 }
 
 TEST_F(CorruptionTest, DisableKeyOrderCheck) {
-  Options options;
   ASSERT_OK(db_->SetOptions({{"check_flush_compaction_key_order", "false"}}));
   DBImpl* dbi = static_cast_with_check<DBImpl>(db_);
 
@@ -824,10 +858,55 @@ TEST_F(CorruptionTest, DisableKeyOrderCheck) {
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
 }
 
+TEST_F(CorruptionTest, VerifyWholeTableChecksum) {
+  CloseDb();
+  Options options;
+  options.env = env_;
+  ASSERT_OK(DestroyDB(dbname_, options));
+  options.create_if_missing = true;
+  options.file_checksum_gen_factory =
+      ROCKSDB_NAMESPACE::GetFileChecksumGenCrc32cFactory();
+  Reopen(&options);
+
+  Build(10, 5);
+
+  ASSERT_OK(db_->VerifyFileChecksums(ReadOptions()));
+  CloseDb();
+
+  // Corrupt the first byte of each table file, this must be data block.
+  Corrupt(kTableFile, 0, 1);
+
+  ASSERT_OK(TryReopen(&options));
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+  int count{0};
+  SyncPoint::GetInstance()->SetCallBack(
+      "DBImpl::VerifyFullFileChecksum:mismatch", [&](void* arg) {
+        auto* s = reinterpret_cast<Status*>(arg);
+        ASSERT_NE(s, nullptr);
+        ++count;
+        ASSERT_NOK(*s);
+      });
+  SyncPoint::GetInstance()->EnableProcessing();
+  ASSERT_TRUE(db_->VerifyFileChecksums(ReadOptions()).IsCorruption());
+  ASSERT_EQ(1, count);
+}
+
 }  // namespace ROCKSDB_NAMESPACE
 
+#ifdef ROCKSDB_UNITTESTS_WITH_CUSTOM_OBJECTS_FROM_STATIC_LIBS
+extern "C" {
+void RegisterCustomObjects(int argc, char** argv);
+}
+#else
+void RegisterCustomObjects(int /*argc*/, char** /*argv*/) {}
+#endif  // !ROCKSDB_UNITTESTS_WITH_CUSTOM_OBJECTS_FROM_STATIC_LIBS
+
 int main(int argc, char** argv) {
+  ROCKSDB_NAMESPACE::port::InstallStackTraceHandler();
   ::testing::InitGoogleTest(&argc, argv);
+  RegisterCustomObjects(argc, argv);
   return RUN_ALL_TESTS();
 }
 

db/cuckoo_table_db_test.cc

@@ -63,6 +63,15 @@ class CuckooTableDBTest : public testing::Test {
     ASSERT_OK(DB::Open(opts, dbname_, &db_));
   }
 
+  void DestroyAndReopen(Options* options) {
+    assert(options);
+    ASSERT_OK(db_->Close());
+    delete db_;
+    db_ = nullptr;
+    ASSERT_OK(DestroyDB(dbname_, *options));
+    Reopen(options);
+  }
+
   Status Put(const Slice& k, const Slice& v) {
     return db_->Put(WriteOptions(), k, v);
   }
@@ -120,10 +129,10 @@ TEST_F(CuckooTableDBTest, Flush) {
   ASSERT_OK(Put("key1", "v1"));
   ASSERT_OK(Put("key2", "v2"));
   ASSERT_OK(Put("key3", "v3"));
-  dbfull()->TEST_FlushMemTable();
+  ASSERT_OK(dbfull()->TEST_FlushMemTable());
 
   TablePropertiesCollection ptc;
-  reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc);
+  ASSERT_OK(reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
   ASSERT_EQ(1U, ptc.size());
   ASSERT_EQ(3U, ptc.begin()->second->num_entries);
   ASSERT_EQ("1", FilesPerLevel());
@@ -137,9 +146,9 @@ TEST_F(CuckooTableDBTest, Flush) {
   ASSERT_OK(Put("key4", "v4"));
   ASSERT_OK(Put("key5", "v5"));
   ASSERT_OK(Put("key6", "v6"));
-  dbfull()->TEST_FlushMemTable();
+  ASSERT_OK(dbfull()->TEST_FlushMemTable());
 
-  reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc);
+  ASSERT_OK(reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
   ASSERT_EQ(2U, ptc.size());
   auto row = ptc.begin();
   ASSERT_EQ(3U, row->second->num_entries);
@@ -155,8 +164,8 @@ TEST_F(CuckooTableDBTest, Flush) {
   ASSERT_OK(Delete("key6"));
   ASSERT_OK(Delete("key5"));
   ASSERT_OK(Delete("key4"));
-  dbfull()->TEST_FlushMemTable();
-  reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc);
+  ASSERT_OK(dbfull()->TEST_FlushMemTable());
+  ASSERT_OK(reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
   ASSERT_EQ(3U, ptc.size());
   row = ptc.begin();
   ASSERT_EQ(3U, row->second->num_entries);
@@ -177,10 +186,10 @@ TEST_F(CuckooTableDBTest, FlushWithDuplicateKeys) {
   ASSERT_OK(Put("key1", "v1"));
   ASSERT_OK(Put("key2", "v2"));
   ASSERT_OK(Put("key1", "v3"));  // Duplicate
-  dbfull()->TEST_FlushMemTable();
+  ASSERT_OK(dbfull()->TEST_FlushMemTable());
 
   TablePropertiesCollection ptc;
-  reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc);
+  ASSERT_OK(reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
   ASSERT_EQ(1U, ptc.size());
   ASSERT_EQ(2U, ptc.begin()->second->num_entries);
   ASSERT_EQ("1", FilesPerLevel());
@@ -205,12 +214,12 @@ static std::string Uint64Key(uint64_t i) {
 TEST_F(CuckooTableDBTest, Uint64Comparator) {
   Options options = CurrentOptions();
   options.comparator = test::Uint64Comparator();
-  Reopen(&options);
+  DestroyAndReopen(&options);
 
   ASSERT_OK(Put(Uint64Key(1), "v1"));
   ASSERT_OK(Put(Uint64Key(2), "v2"));
   ASSERT_OK(Put(Uint64Key(3), "v3"));
-  dbfull()->TEST_FlushMemTable();
+  ASSERT_OK(dbfull()->TEST_FlushMemTable());
 
   ASSERT_EQ("v1", Get(Uint64Key(1)));
   ASSERT_EQ("v2", Get(Uint64Key(2)));
@@ -219,10 +228,10 @@ TEST_F(CuckooTableDBTest, Uint64Comparator) {
 
   // Add more keys.
   ASSERT_OK(Delete(Uint64Key(2)));  // Delete.
-  dbfull()->TEST_FlushMemTable();
+  ASSERT_OK(dbfull()->TEST_FlushMemTable());
   ASSERT_OK(Put(Uint64Key(3), "v0"));  // Update.
   ASSERT_OK(Put(Uint64Key(4), "v4"));
-  dbfull()->TEST_FlushMemTable();
+  ASSERT_OK(dbfull()->TEST_FlushMemTable());
   ASSERT_EQ("v1", Get(Uint64Key(1)));
   ASSERT_EQ("NOT_FOUND", Get(Uint64Key(2)));
   ASSERT_EQ("v0", Get(Uint64Key(3)));
@@ -242,11 +251,11 @@ TEST_F(CuckooTableDBTest, CompactionIntoMultipleFiles) {
   for (int idx = 0; idx < 28; ++idx) {
     ASSERT_OK(Put(Key(idx), std::string(10000, 'a' + char(idx))));
   }
-  dbfull()->TEST_WaitForFlushMemTable();
+  ASSERT_OK(dbfull()->TEST_WaitForFlushMemTable());
   ASSERT_EQ("1", FilesPerLevel());
 
-  dbfull()->TEST_CompactRange(0, nullptr, nullptr, nullptr,
-                              true /* disallow trivial move */);
+  ASSERT_OK(dbfull()->TEST_CompactRange(0, nullptr, nullptr, nullptr,
+                                        true /* disallow trivial move */));
   ASSERT_EQ("0,2", FilesPerLevel());
   for (int idx = 0; idx < 28; ++idx) {
     ASSERT_EQ(std::string(10000, 'a' + char(idx)), Get(Key(idx)));
@@ -265,15 +274,15 @@ TEST_F(CuckooTableDBTest, SameKeyInsertedInTwoDifferentFilesAndCompacted) {
   for (int idx = 0; idx < 11; ++idx) {
     ASSERT_OK(Put(Key(idx), std::string(10000, 'a')));
   }
-  dbfull()->TEST_WaitForFlushMemTable();
+  ASSERT_OK(dbfull()->TEST_WaitForFlushMemTable());
   ASSERT_EQ("1", FilesPerLevel());
 
   // Generate one more file in level-0, and should trigger level-0 compaction
   for (int idx = 0; idx < 11; ++idx) {
     ASSERT_OK(Put(Key(idx), std::string(10000, 'a' + char(idx))));
   }
-  dbfull()->TEST_WaitForFlushMemTable();
-  dbfull()->TEST_CompactRange(0, nullptr, nullptr);
+  ASSERT_OK(dbfull()->TEST_WaitForFlushMemTable());
+  ASSERT_OK(dbfull()->TEST_CompactRange(0, nullptr, nullptr));
 
   ASSERT_EQ("0,1", FilesPerLevel());
   for (int idx = 0; idx < 11; ++idx) {
@@ -294,7 +303,7 @@ TEST_F(CuckooTableDBTest, AdaptiveTable) {
   ASSERT_OK(Put("key1", "v1"));
   ASSERT_OK(Put("key2", "v2"));
   ASSERT_OK(Put("key3", "v3"));
-  dbfull()->TEST_FlushMemTable();
+  ASSERT_OK(dbfull()->TEST_FlushMemTable());
 
   // Write some keys using plain table.
   std::shared_ptr<TableFactory> block_based_factory(
@@ -310,7 +319,7 @@ TEST_F(CuckooTableDBTest, AdaptiveTable) {
   Reopen(&options);
   ASSERT_OK(Put("key4", "v4"));
   ASSERT_OK(Put("key1", "v5"));
-  dbfull()->TEST_FlushMemTable();
+  ASSERT_OK(dbfull()->TEST_FlushMemTable());
 
   // Write some keys using block based table.
   options.table_factory.reset(NewAdaptiveTableFactory(
@@ -319,7 +328,7 @@ TEST_F(CuckooTableDBTest, AdaptiveTable) {
   Reopen(&options);
   ASSERT_OK(Put("key5", "v6"));
   ASSERT_OK(Put("key2", "v7"));
-  dbfull()->TEST_FlushMemTable();
+  ASSERT_OK(dbfull()->TEST_FlushMemTable());
 
   ASSERT_EQ("v5", Get("key1"));
   ASSERT_EQ("v7", Get("key2"));

db/db_basic_test.cc

@@ -12,6 +12,7 @@
 
 #include "db/db_test_util.h"
 #include "port/stack_trace.h"
+#include "rocksdb/flush_block_policy.h"
 #include "rocksdb/merge_operator.h"
 #include "rocksdb/perf_context.h"
 #include "rocksdb/utilities/debug.h"
@@ -20,6 +21,7 @@
 #if !defined(ROCKSDB_LITE)
 #include "test_util/sync_point.h"
 #endif
+#include "util/file_checksum_helper.h"
 #include "util/random.h"
 #include "utilities/fault_injection_env.h"
 #include "utilities/merge_operators.h"
@@ -37,7 +39,10 @@ TEST_F(DBBasicTest, OpenWhenOpen) {
   options.env = env_;
   DB* db2 = nullptr;
   Status s = DB::Open(options, dbname_, &db2);
-  ASSERT_NOK(s);
+  ASSERT_NOK(s) << [db2]() {
+    delete db2;
+    return "db2 open: ok";
+  }();
   ASSERT_EQ(Status::Code::kIOError, s.code());
   ASSERT_EQ(Status::SubCode::kNone, s.subcode());
   ASSERT_TRUE(strstr(s.getState(), "lock ") != nullptr);
@@ -142,7 +147,7 @@ TEST_F(DBBasicTest, ReadOnlyDB) {
 
   // Reopen and flush memtable.
   Reopen(options);
-  Flush();
+  ASSERT_OK(Flush());
   Close();
   // Now check keys in read only mode.
   ASSERT_OK(ReadOnlyReopen(options));
@@ -178,7 +183,7 @@ TEST_F(DBBasicTest, ReadOnlyDBWithWriteDBIdToManifestSet) {
 
   // Reopen and flush memtable.
   Reopen(options);
-  Flush();
+  ASSERT_OK(Flush());
   Close();
   // Now check keys in read only mode.
   ASSERT_OK(ReadOnlyReopen(options));
@@ -201,7 +206,7 @@ TEST_F(DBBasicTest, CompactedDB) {
   Reopen(options);
   // 1 L0 file, use CompactedDB if max_open_files = -1
   ASSERT_OK(Put("aaa", DummyString(kFileSize / 2, '1')));
-  Flush();
+  ASSERT_OK(Flush());
   Close();
   ASSERT_OK(ReadOnlyReopen(options));
   Status s = Put("new", "value");
@@ -219,12 +224,12 @@ TEST_F(DBBasicTest, CompactedDB) {
   Reopen(options);
   // Add more L0 files
   ASSERT_OK(Put("bbb", DummyString(kFileSize / 2, '2')));
-  Flush();
+  ASSERT_OK(Flush());
   ASSERT_OK(Put("aaa", DummyString(kFileSize / 2, 'a')));
-  Flush();
+  ASSERT_OK(Flush());
   ASSERT_OK(Put("bbb", DummyString(kFileSize / 2, 'b')));
   ASSERT_OK(Put("eee", DummyString(kFileSize / 2, 'e')));
-  Flush();
+  ASSERT_OK(Flush());
   Close();
 
   ASSERT_OK(ReadOnlyReopen(options));
@@ -401,16 +406,18 @@ TEST_F(DBBasicTest, GetSnapshot) {
 
 TEST_F(DBBasicTest, CheckLock) {
   do {
-    DB* localdb;
+    DB* localdb = nullptr;
     Options options = CurrentOptions();
     ASSERT_OK(TryReopen(options));
 
     // second open should fail
     Status s = DB::Open(options, dbname_, &localdb);
-    ASSERT_NOK(s);
+    ASSERT_NOK(s) << [localdb]() {
+      delete localdb;
+      return "localdb open: ok";
+    }();
 #ifdef OS_LINUX
-    ASSERT_TRUE(s.ToString().find("lock hold by current process") !=
-                std::string::npos);
+    ASSERT_TRUE(s.ToString().find("lock ") != std::string::npos);
 #endif  // OS_LINUX
   } while (ChangeCompactOptions());
 }
@@ -1407,10 +1414,10 @@ TEST_F(DBBasicTest, MultiGetBatchedSortedMultiFile) {
     // mix with memtable
     ASSERT_OK(Put(1, "k1", "v1"));
     ASSERT_OK(Put(1, "k2", "v2"));
-    Flush(1);
+    ASSERT_OK(Flush(1));
     ASSERT_OK(Put(1, "k3", "v3"));
     ASSERT_OK(Put(1, "k4", "v4"));
-    Flush(1);
+    ASSERT_OK(Flush(1));
     ASSERT_OK(Delete(1, "k4"));
     ASSERT_OK(Put(1, "k5", "v5"));
     ASSERT_OK(Delete(1, "no_key"));
@@ -1453,19 +1460,19 @@ TEST_F(DBBasicTest, MultiGetBatchedDuplicateKeys) {
   // mix with memtable
   ASSERT_OK(Merge(1, "k1", "v1"));
   ASSERT_OK(Merge(1, "k2", "v2"));
-  Flush(1);
+  ASSERT_OK(Flush(1));
   MoveFilesToLevel(2, 1);
   ASSERT_OK(Merge(1, "k3", "v3"));
   ASSERT_OK(Merge(1, "k4", "v4"));
-  Flush(1);
+  ASSERT_OK(Flush(1));
   MoveFilesToLevel(2, 1);
   ASSERT_OK(Merge(1, "k4", "v4_2"));
   ASSERT_OK(Merge(1, "k6", "v6"));
-  Flush(1);
+  ASSERT_OK(Flush(1));
   MoveFilesToLevel(2, 1);
   ASSERT_OK(Merge(1, "k7", "v7"));
   ASSERT_OK(Merge(1, "k8", "v8"));
-  Flush(1);
+  ASSERT_OK(Flush(1));
   MoveFilesToLevel(2, 1);
 
   get_perf_context()->Reset();
@@ -1505,12 +1512,12 @@ TEST_F(DBBasicTest, MultiGetBatchedMultiLevel) {
     ASSERT_OK(Put("key_" + std::to_string(i), "val_l2_" + std::to_string(i)));
     num_keys++;
     if (num_keys == 8) {
-      Flush();
+      ASSERT_OK(Flush());
       num_keys = 0;
     }
   }
   if (num_keys > 0) {
-    Flush();
+    ASSERT_OK(Flush());
     num_keys = 0;
   }
   MoveFilesToLevel(2);
@@ -1519,12 +1526,12 @@ TEST_F(DBBasicTest, MultiGetBatchedMultiLevel) {
     ASSERT_OK(Put("key_" + std::to_string(i), "val_l1_" + std::to_string(i)));
     num_keys++;
     if (num_keys == 8) {
-      Flush();
+      ASSERT_OK(Flush());
       num_keys = 0;
     }
   }
   if (num_keys > 0) {
-    Flush();
+    ASSERT_OK(Flush());
     num_keys = 0;
   }
   MoveFilesToLevel(1);
@@ -1533,12 +1540,12 @@ TEST_F(DBBasicTest, MultiGetBatchedMultiLevel) {
     ASSERT_OK(Put("key_" + std::to_string(i), "val_l0_" + std::to_string(i)));
     num_keys++;
     if (num_keys == 8) {
-      Flush();
+      ASSERT_OK(Flush());
       num_keys = 0;
     }
   }
   if (num_keys > 0) {
-    Flush();
+    ASSERT_OK(Flush());
     num_keys = 0;
   }
   ASSERT_EQ(0, num_keys);
@@ -1584,12 +1591,12 @@ TEST_F(DBBasicTest, MultiGetBatchedMultiLevelMerge) {
     ASSERT_OK(Put("key_" + std::to_string(i), "val_l2_" + std::to_string(i)));
     num_keys++;
     if (num_keys == 8) {
-      Flush();
+      ASSERT_OK(Flush());
       num_keys = 0;
     }
   }
   if (num_keys > 0) {
-    Flush();
+    ASSERT_OK(Flush());
     num_keys = 0;
   }
   MoveFilesToLevel(2);
@@ -1598,12 +1605,12 @@ TEST_F(DBBasicTest, MultiGetBatchedMultiLevelMerge) {
     ASSERT_OK(Merge("key_" + std::to_string(i), "val_l1_" + std::to_string(i)));
     num_keys++;
     if (num_keys == 8) {
-      Flush();
+      ASSERT_OK(Flush());
       num_keys = 0;
     }
   }
   if (num_keys > 0) {
-    Flush();
+    ASSERT_OK(Flush());
     num_keys = 0;
   }
   MoveFilesToLevel(1);
@@ -1612,12 +1619,12 @@ TEST_F(DBBasicTest, MultiGetBatchedMultiLevelMerge) {
     ASSERT_OK(Merge("key_" + std::to_string(i), "val_l0_" + std::to_string(i)));
     num_keys++;
     if (num_keys == 8) {
-      Flush();
+      ASSERT_OK(Flush());
       num_keys = 0;
     }
   }
   if (num_keys > 0) {
-    Flush();
+    ASSERT_OK(Flush());
     num_keys = 0;
   }
   ASSERT_EQ(0, num_keys);
@@ -1699,7 +1706,7 @@ TEST_F(DBBasicTest, MultiGetBatchedValueSize) {
     ASSERT_OK(Put(1, "k7", "v7_"));
     ASSERT_OK(Put(1, "k3", "v3_"));
     ASSERT_OK(Put(1, "k4", "v4"));
-    Flush(1);
+    ASSERT_OK(Flush(1));
     ASSERT_OK(Delete(1, "k4"));
     ASSERT_OK(Put(1, "k11", "v11"));
     ASSERT_OK(Delete(1, "no_key"));
@@ -1709,7 +1716,7 @@ TEST_F(DBBasicTest, MultiGetBatchedValueSize) {
     ASSERT_OK(Put(1, "k15", "v15"));
     ASSERT_OK(Put(1, "k16", "v16"));
     ASSERT_OK(Put(1, "k17", "v17"));
-    Flush(1);
+    ASSERT_OK(Flush(1));
 
     ASSERT_OK(Put(1, "k1", "v1_"));
     ASSERT_OK(Put(1, "k2", "v2_"));
@@ -1779,12 +1786,12 @@ TEST_F(DBBasicTest, MultiGetBatchedValueSizeMultiLevelMerge) {
     ASSERT_OK(Put("key_" + std::to_string(i), "val_l2_" + std::to_string(i)));
     num_keys++;
     if (num_keys == 8) {
-      Flush();
+      ASSERT_OK(Flush());
       num_keys = 0;
     }
   }
   if (num_keys > 0) {
-    Flush();
+    ASSERT_OK(Flush());
     num_keys = 0;
   }
   MoveFilesToLevel(2);
@@ -1793,12 +1800,12 @@ TEST_F(DBBasicTest, MultiGetBatchedValueSizeMultiLevelMerge) {
     ASSERT_OK(Merge("key_" + std::to_string(i), "val_l1_" + std::to_string(i)));
     num_keys++;
     if (num_keys == 8) {
-      Flush();
+      ASSERT_OK(Flush());
       num_keys = 0;
     }
   }
   if (num_keys > 0) {
-    Flush();
+    ASSERT_OK(Flush());
     num_keys = 0;
   }
   MoveFilesToLevel(1);
@@ -1807,12 +1814,12 @@ TEST_F(DBBasicTest, MultiGetBatchedValueSizeMultiLevelMerge) {
     ASSERT_OK(Merge("key_" + std::to_string(i), "val_l0_" + std::to_string(i)));
     num_keys++;
     if (num_keys == 8) {
-      Flush();
+      ASSERT_OK(Flush());
       num_keys = 0;
     }
   }
   if (num_keys > 0) {
-    Flush();
+    ASSERT_OK(Flush());
     num_keys = 0;
   }
   ASSERT_EQ(0, num_keys);
@@ -1875,6 +1882,7 @@ TEST_F(DBBasicTest, MultiGetStats) {
   Options options;
   options.create_if_missing = true;
   options.disable_auto_compactions = true;
+  options.env = env_;
   options.statistics = ROCKSDB_NAMESPACE::CreateDBStatistics();
   BlockBasedTableOptions table_options;
   table_options.block_size = 1;
@@ -1884,7 +1892,7 @@ TEST_F(DBBasicTest, MultiGetStats) {
   table_options.no_block_cache = true;
   table_options.cache_index_and_filter_blocks = false;
   table_options.filter_policy.reset(NewBloomFilterPolicy(10, false));
-  options.table_factory.reset(new BlockBasedTableFactory(table_options));
+  options.table_factory.reset(NewBlockBasedTableFactory(table_options));
   CreateAndReopenWithCF({"pikachu"}, options);
 
   int total_keys = 2000;
@@ -1901,10 +1909,10 @@ TEST_F(DBBasicTest, MultiGetStats) {
     keys[i] = Slice(keys_str[i]);
     ASSERT_OK(Put(1, "k" + std::to_string(i), rnd.RandomString(1000)));
     if (i % 100 == 0) {
-      Flush(1);
+      ASSERT_OK(Flush(1));
     }
   }
-  Flush(1);
+  ASSERT_OK(Flush(1));
   MoveFilesToLevel(2, 1);
 
   for (int i = 501; i < 1000; ++i) {
@@ -1912,11 +1920,11 @@ TEST_F(DBBasicTest, MultiGetStats) {
     keys[i] = Slice(keys_str[i]);
     ASSERT_OK(Put(1, "k" + std::to_string(i), rnd.RandomString(1000)));
     if (i % 100 == 0) {
-      Flush(1);
+      ASSERT_OK(Flush(1));
     }
   }
 
-  Flush(1);
+  ASSERT_OK(Flush(1));
   MoveFilesToLevel(2, 1);
 
   for (int i = 1001; i < total_keys; ++i) {
@@ -1924,10 +1932,10 @@ TEST_F(DBBasicTest, MultiGetStats) {
     keys[i] = Slice(keys_str[i]);
     ASSERT_OK(Put(1, "k" + std::to_string(i), rnd.RandomString(1000)));
     if (i % 100 == 0) {
-      Flush(1);
+      ASSERT_OK(Flush(1));
     }
   }
-  Flush(1);
+  ASSERT_OK(Flush(1));
   Close();
 
   ReopenWithColumnFamilies({"default", "pikachu"}, options);
@@ -1954,7 +1962,7 @@ TEST_F(DBBasicTest, MultiGetStats) {
   ASSERT_GT(hist_sst.max, 0);
 
   // Minimun number of blocks read in a level.
-  ASSERT_EQ(hist_data_blocks.min, 0);
+  ASSERT_EQ(hist_data_blocks.min, 3);
   ASSERT_GT(hist_index_and_filter_blocks.min, 0);
   // Minimun number of sst files read in a level.
   ASSERT_GT(hist_sst.max, 0);
@@ -2035,11 +2043,11 @@ TEST_P(DBMultiGetRowCacheTest, MultiGetBatched) {
     ASSERT_OK(Put(1, "k2", "v2"));
     ASSERT_OK(Put(1, "k3", "v3"));
     ASSERT_OK(Put(1, "k4", "v4"));
-    Flush(1);
+    ASSERT_OK(Flush(1));
     ASSERT_OK(Put(1, "k5", "v5"));
     const Snapshot* snap1 = dbfull()->GetSnapshot();
     ASSERT_OK(Delete(1, "k4"));
-    Flush(1);
+    ASSERT_OK(Flush(1));
     const Snapshot* snap2 = dbfull()->GetSnapshot();
 
     get_perf_context()->Reset();
@@ -2168,7 +2176,7 @@ TEST_F(DBBasicTest, MultiGetIOBufferOverrun) {
   table_options.block_size = 16 * 1024;
   ASSERT_TRUE(table_options.block_size >
             BlockBasedTable::kMultiGetReadStackBufSize);
-  options.table_factory.reset(new BlockBasedTableFactory(table_options));
+  options.table_factory.reset(NewBlockBasedTableFactory(table_options));
   Reopen(options);
 
   std::string zero_str(128, '\0');
@@ -2178,7 +2186,7 @@ TEST_F(DBBasicTest, MultiGetIOBufferOverrun) {
     std::string value(rnd.RandomString(128) + zero_str);
     assert(Put(Key(i), value) == Status::OK());
   }
-  Flush();
+  ASSERT_OK(Flush());
 
   std::vector<std::string> key_data(10);
   std::vector<Slice> keys;
@@ -2278,6 +2286,43 @@ class TableFileListener : public EventListener {
 };
 }  // namespace
 
+TEST_F(DBBasicTest, LastSstFileNotInManifest) {
+  // If the last sst file is not tracked in MANIFEST,
+  // or the VersionEdit for the last sst file is not synced,
+  // on recovery, the last sst file should be deleted,
+  // and new sst files shouldn't reuse its file number.
+  Options options = CurrentOptions();
+  DestroyAndReopen(options);
+  Close();
+
+  // Manually add a sst file.
+  constexpr uint64_t kSstFileNumber = 100;
+  const std::string kSstFile = MakeTableFileName(dbname_, kSstFileNumber);
+  ASSERT_OK(WriteStringToFile(env_, /* data = */ "bad sst file content",
+                              /* fname = */ kSstFile,
+                              /* should_sync = */ true));
+  ASSERT_OK(env_->FileExists(kSstFile));
+
+  TableFileListener* listener = new TableFileListener();
+  options.listeners.emplace_back(listener);
+  Reopen(options);
+  // kSstFile should already be deleted.
+  ASSERT_TRUE(env_->FileExists(kSstFile).IsNotFound());
+
+  ASSERT_OK(Put("k", "v"));
+  ASSERT_OK(Flush());
+  // New sst file should have file number > kSstFileNumber.
+  std::vector<std::string>& files =
+      listener->GetFiles(kDefaultColumnFamilyName);
+  ASSERT_EQ(files.size(), 1);
+  const std::string fname = files[0].erase(0, (dbname_ + "/").size());
+  uint64_t number = 0;
+  FileType type = kTableFile;
+  ASSERT_TRUE(ParseFileName(fname, &number, &type));
+  ASSERT_EQ(type, kTableFile);
+  ASSERT_GT(number, kSstFileNumber);
+}
+
 TEST_F(DBBasicTest, RecoverWithMissingFiles) {
   Options options = CurrentOptions();
   DestroyAndReopen(options);
@@ -2405,7 +2450,7 @@ TEST_F(DBBasicTest, RecoverWithNoManifest) {
     ASSERT_OK(env_->GetChildren(dbname_, &files));
     for (const auto& file : files) {
       uint64_t number = 0;
-      FileType type = kLogFile;
+      FileType type = kWalFile;
       if (ParseFileName(file, &number, &type) && type == kDescriptorFile) {
         ASSERT_OK(env_->DeleteFile(dbname_ + "/" + file));
       }
@@ -2461,6 +2506,42 @@ TEST_F(DBBasicTest, SkipWALIfMissingTableFiles) {
   ASSERT_FALSE(iter->Valid());
   ASSERT_OK(iter->status());
 }
+
+TEST_F(DBBasicTest, DisableTrackWal) {
+  // If WAL tracking was enabled, and then disabled during reopen,
+  // the previously tracked WALs should be removed from MANIFEST.
+
+  Options options = CurrentOptions();
+  options.track_and_verify_wals_in_manifest = true;
+  // extremely small write buffer size,
+  // so that new WALs are created more frequently.
+  options.write_buffer_size = 100;
+  options.env = env_;
+  DestroyAndReopen(options);
+  for (int i = 0; i < 100; i++) {
+    ASSERT_OK(Put("foo" + std::to_string(i), "value" + std::to_string(i)));
+  }
+  ASSERT_OK(dbfull()->TEST_SwitchMemtable());
+  ASSERT_OK(db_->SyncWAL());
+  // Some WALs are tracked.
+  ASSERT_FALSE(dbfull()->TEST_GetVersionSet()->GetWalSet().GetWals().empty());
+  Close();
+
+  // Disable WAL tracking.
+  options.track_and_verify_wals_in_manifest = false;
+  options.create_if_missing = false;
+  ASSERT_OK(TryReopen(options));
+  // Previously tracked WALs are cleared.
+  ASSERT_TRUE(dbfull()->TEST_GetVersionSet()->GetWalSet().GetWals().empty());
+  Close();
+
+  // Re-enable WAL tracking again.
+  options.track_and_verify_wals_in_manifest = true;
+  options.create_if_missing = false;
+  ASSERT_OK(TryReopen(options));
+  ASSERT_TRUE(dbfull()->TEST_GetVersionSet()->GetWalSet().GetWals().empty());
+  Close();
+}
 #endif  // !ROCKSDB_LITE
 
 TEST_F(DBBasicTest, ManifestChecksumMismatch) {
@@ -2490,6 +2571,64 @@ TEST_F(DBBasicTest, ManifestChecksumMismatch) {
   ASSERT_TRUE(s.IsCorruption());
 }
 
+#ifndef ROCKSDB_LITE
+class DBBasicTestTrackWal : public DBTestBase,
+                            public testing::WithParamInterface<bool> {
+ public:
+  DBBasicTestTrackWal()
+      : DBTestBase("/db_basic_test_track_wal", /*env_do_fsync=*/false) {}
+
+  int CountWalFiles() {
+    VectorLogPtr log_files;
+    EXPECT_OK(dbfull()->GetSortedWalFiles(log_files));
+    return static_cast<int>(log_files.size());
+  };
+};
+
+TEST_P(DBBasicTestTrackWal, DoNotTrackObsoleteWal) {
+  // If a WAL becomes obsolete after flushing, but is not deleted from disk yet,
+  // then if SyncWAL is called afterwards, the obsolete WAL should not be
+  // tracked in MANIFEST.
+
+  Options options = CurrentOptions();
+  options.create_if_missing = true;
+  options.track_and_verify_wals_in_manifest = true;
+  options.atomic_flush = GetParam();
+
+  DestroyAndReopen(options);
+  CreateAndReopenWithCF({"cf"}, options);
+  ASSERT_EQ(handles_.size(), 2);  // default, cf
+  // Do not delete WALs.
+  ASSERT_OK(db_->DisableFileDeletions());
+  constexpr int n = 10;
+  std::vector<std::unique_ptr<LogFile>> wals(n);
+  for (size_t i = 0; i < n; i++) {
+    // Generate a new WAL for each key-value.
+    const int cf = i % 2;
+    ASSERT_OK(db_->GetCurrentWalFile(&wals[i]));
+    ASSERT_OK(Put(cf, "k" + std::to_string(i), "v" + std::to_string(i)));
+    ASSERT_OK(Flush({0, 1}));
+  }
+  ASSERT_EQ(CountWalFiles(), n);
+  // Since all WALs are obsolete, no WAL should be tracked in MANIFEST.
+  ASSERT_OK(db_->SyncWAL());
+
+  // Manually delete all WALs.
+  Close();
+  for (const auto& wal : wals) {
+    ASSERT_OK(env_->DeleteFile(LogFileName(dbname_, wal->LogNumber())));
+  }
+
+  // If SyncWAL tracks the obsolete WALs in MANIFEST,
+  // reopen will fail because the WALs are missing from disk.
+  ASSERT_OK(TryReopenWithColumnFamilies({"default", "cf"}, options));
+  Destroy(options);
+}
+
+INSTANTIATE_TEST_CASE_P(DBBasicTestTrackWal, DBBasicTestTrackWal,
+                        testing::Bool());
+#endif  // ROCKSDB_LITE
+
 class DBBasicTestMultiGet : public DBTestBase {
  public:
   DBBasicTestMultiGet(std::string test_dir, int num_cfs, bool compressed_cache,
@@ -2549,12 +2688,13 @@ class DBBasicTestMultiGet : public DBTestBase {
     table_options.block_cache_compressed = compressed_cache_;
     table_options.flush_block_policy_factory.reset(
         new MyFlushBlockPolicyFactory());
-    options.table_factory.reset(new BlockBasedTableFactory(table_options));
+    options.table_factory.reset(NewBlockBasedTableFactory(table_options));
     if (!compression_enabled_) {
       options.compression = kNoCompression;
     } else {
       options.compression_opts.parallel_threads = compression_parallel_threads;
     }
+    options_ = options;
     Reopen(options);
 
     if (num_cfs > 1) {
@@ -2575,9 +2715,9 @@ class DBBasicTestMultiGet : public DBTestBase {
                                : Put(cf, Key(i), values_[i])) == Status::OK());
       }
       if (num_cfs == 1) {
-        Flush();
+        EXPECT_OK(Flush());
       } else {
-        dbfull()->Flush(FlushOptions(), handles_[cf]);
+        EXPECT_OK(dbfull()->Flush(FlushOptions(), handles_[cf]));
       }
 
       for (int i = 0; i < 100; ++i) {
@@ -2589,9 +2729,9 @@ class DBBasicTestMultiGet : public DBTestBase {
                Status::OK());
       }
       if (num_cfs == 1) {
-        Flush();
+        EXPECT_OK(Flush());
       } else {
-        dbfull()->Flush(FlushOptions(), handles_[cf]);
+        EXPECT_OK(dbfull()->Flush(FlushOptions(), handles_[cf]));
       }
     }
   }
@@ -2624,6 +2764,7 @@ class DBBasicTestMultiGet : public DBTestBase {
   bool compression_enabled() { return compression_enabled_; }
   bool has_compressed_cache() { return compressed_cache_ != nullptr; }
   bool has_uncompressed_cache() { return uncompressed_cache_ != nullptr; }
+  Options get_options() { return options_; }
 
   static void SetUpTestCase() {}
   static void TearDownTestCase() {}
@@ -2679,6 +2820,7 @@ class DBBasicTestMultiGet : public DBTestBase {
 
     const char* Name() const override { return "MyBlockCache"; }
 
+    using Cache::Insert;
     Status Insert(const Slice& key, void* value, size_t charge,
                   void (*deleter)(const Slice& key, void* value),
                   Handle** handle = nullptr,
@@ -2687,6 +2829,7 @@ class DBBasicTestMultiGet : public DBTestBase {
       return target_->Insert(key, value, charge, deleter, handle, priority);
     }
 
+    using Cache::Lookup;
     Handle* Lookup(const Slice& key, Statistics* stats = nullptr) override {
       num_lookups_++;
       Handle* handle = target_->Lookup(key, stats);
@@ -2709,6 +2852,7 @@ class DBBasicTestMultiGet : public DBTestBase {
 
   std::shared_ptr<MyBlockCache> compressed_cache_;
   std::shared_ptr<MyBlockCache> uncompressed_cache_;
+  Options options_;
   bool compression_enabled_;
   std::vector<std::string> values_;
   std::vector<std::string> uncompressable_values_;
@@ -2851,6 +2995,123 @@ TEST_P(DBBasicTestWithParallelIO, MultiGet) {
   }
 }
 
+#ifndef ROCKSDB_LITE
+TEST_P(DBBasicTestWithParallelIO, MultiGetDirectIO) {
+  class FakeDirectIOEnv : public EnvWrapper {
+    class FakeDirectIOSequentialFile;
+    class FakeDirectIORandomAccessFile;
+
+   public:
+    FakeDirectIOEnv(Env* env) : EnvWrapper(env) {}
+
+    Status NewRandomAccessFile(const std::string& fname,
+                               std::unique_ptr<RandomAccessFile>* result,
+                               const EnvOptions& options) override {
+      std::unique_ptr<RandomAccessFile> file;
+      assert(options.use_direct_reads);
+      EnvOptions opts = options;
+      opts.use_direct_reads = false;
+      Status s = target()->NewRandomAccessFile(fname, &file, opts);
+      if (!s.ok()) {
+        return s;
+      }
+      result->reset(new FakeDirectIORandomAccessFile(std::move(file)));
+      return s;
+    }
+
+   private:
+    class FakeDirectIOSequentialFile : public SequentialFileWrapper {
+     public:
+      FakeDirectIOSequentialFile(std::unique_ptr<SequentialFile>&& file)
+          : SequentialFileWrapper(file.get()), file_(std::move(file)) {}
+      ~FakeDirectIOSequentialFile() {}
+
+      bool use_direct_io() const override { return true; }
+      size_t GetRequiredBufferAlignment() const override { return 1; }
+
+     private:
+      std::unique_ptr<SequentialFile> file_;
+    };
+
+    class FakeDirectIORandomAccessFile : public RandomAccessFileWrapper {
+     public:
+      FakeDirectIORandomAccessFile(std::unique_ptr<RandomAccessFile>&& file)
+          : RandomAccessFileWrapper(file.get()), file_(std::move(file)) {}
+      ~FakeDirectIORandomAccessFile() {}
+
+      bool use_direct_io() const override { return true; }
+      size_t GetRequiredBufferAlignment() const override { return 1; }
+
+     private:
+      std::unique_ptr<RandomAccessFile> file_;
+    };
+  };
+
+  std::unique_ptr<FakeDirectIOEnv> env(new FakeDirectIOEnv(env_));
+  Options opts = get_options();
+  opts.env = env.get();
+  opts.use_direct_reads = true;
+  Reopen(opts);
+
+  std::vector<std::string> key_data(10);
+  std::vector<Slice> keys;
+  // We cannot resize a PinnableSlice vector, so just set initial size to
+  // largest we think we will need
+  std::vector<PinnableSlice> values(10);
+  std::vector<Status> statuses;
+  ReadOptions ro;
+  ro.fill_cache = fill_cache();
+
+  // Warm up the cache first
+  key_data.emplace_back(Key(0));
+  keys.emplace_back(Slice(key_data.back()));
+  key_data.emplace_back(Key(50));
+  keys.emplace_back(Slice(key_data.back()));
+  statuses.resize(keys.size());
+
+  dbfull()->MultiGet(ro, dbfull()->DefaultColumnFamily(), keys.size(),
+                     keys.data(), values.data(), statuses.data(), true);
+  ASSERT_TRUE(CheckValue(0, values[0].ToString()));
+  ASSERT_TRUE(CheckValue(50, values[1].ToString()));
+
+  int random_reads = env_->random_read_counter_.Read();
+  key_data[0] = Key(1);
+  key_data[1] = Key(51);
+  keys[0] = Slice(key_data[0]);
+  keys[1] = Slice(key_data[1]);
+  values[0].Reset();
+  values[1].Reset();
+  if (uncompressed_cache_) {
+    uncompressed_cache_->SetCapacity(0);
+    uncompressed_cache_->SetCapacity(1048576);
+  }
+  dbfull()->MultiGet(ro, dbfull()->DefaultColumnFamily(), keys.size(),
+                     keys.data(), values.data(), statuses.data(), true);
+  ASSERT_TRUE(CheckValue(1, values[0].ToString()));
+  ASSERT_TRUE(CheckValue(51, values[1].ToString()));
+
+  bool read_from_cache = false;
+  if (fill_cache()) {
+    if (has_uncompressed_cache()) {
+      read_from_cache = true;
+    } else if (has_compressed_cache() && compression_enabled()) {
+      read_from_cache = true;
+    }
+  }
+
+  int expected_reads = random_reads;
+  if (!compression_enabled() || !has_compressed_cache()) {
+    expected_reads += 2;
+  } else {
+    expected_reads += (read_from_cache ? 0 : 2);
+  }
+  if (env_->random_read_counter_.Read() != expected_reads) {
+    ASSERT_EQ(env_->random_read_counter_.Read(), expected_reads);
+  }
+  Close();
+}
+#endif  // ROCKSDB_LITE
+
 TEST_P(DBBasicTestWithParallelIO, MultiGetWithChecksumMismatch) {
   std::vector<std::string> key_data(10);
   std::vector<Slice> keys;
@@ -2976,7 +3237,7 @@ class DeadlineFS : public FileSystemWrapper {
   // or to simply delay but return success anyway. The latter mimics the
   // behavior of PosixFileSystem, which does not enforce any timeout
   explicit DeadlineFS(SpecialEnv* env, bool error_on_delay)
-      : FileSystemWrapper(FileSystem::Default()),
+      : FileSystemWrapper(env->GetFileSystem()),
         deadline_(std::chrono::microseconds::zero()),
         io_timeout_(std::chrono::microseconds::zero()),
         env_(env),
@@ -3015,7 +3276,9 @@ class DeadlineFS : public FileSystemWrapper {
 
   // Increment the IO counter and return a delay in microseconds
   IOStatus ShouldDelay(const IOOptions& opts) {
-    if (!deadline_.count() && !io_timeout_.count()) {
+    if (timedout_) {
+      return IOStatus::TimedOut();
+    } else if (!deadline_.count() && !io_timeout_.count()) {
       return IOStatus::OK();
     }
     if (!ignore_deadline_ && delay_trigger_ == io_count_++) {
@@ -3151,7 +3414,7 @@ TEST_F(DBBasicTestMultiGetDeadline, MultiGetDeadlineExceeded) {
   std::shared_ptr<Cache> cache = NewLRUCache(1048576);
   BlockBasedTableOptions table_options;
   table_options.block_cache = cache;
-  options.table_factory.reset(new BlockBasedTableFactory(table_options));
+  options.table_factory.reset(NewBlockBasedTableFactory(table_options));
   options.env = env.get();
   SetTimeElapseOnlySleepOnReopen(&options);
   ReopenWithColumnFamilies(GetCFNames(), options);
@@ -3303,6 +3566,51 @@ TEST_F(DBBasicTest, ManifestWriteFailure) {
   Reopen(options);
 }
 
+#ifndef ROCKSDB_LITE
+TEST_F(DBBasicTest, VerifyFileChecksums) {
+  Options options = GetDefaultOptions();
+  options.create_if_missing = true;
+  options.env = env_;
+  DestroyAndReopen(options);
+  ASSERT_OK(Put("a", "value"));
+  ASSERT_OK(Flush());
+  ASSERT_TRUE(db_->VerifyFileChecksums(ReadOptions()).IsInvalidArgument());
+
+  options.file_checksum_gen_factory = GetFileChecksumGenCrc32cFactory();
+  Reopen(options);
+  ASSERT_OK(db_->VerifyFileChecksums(ReadOptions()));
+
+  // Write an L0 with checksum computed.
+  ASSERT_OK(Put("b", "value"));
+  ASSERT_OK(Flush());
+
+  ASSERT_OK(db_->VerifyFileChecksums(ReadOptions()));
+
+  // Does the right thing but with the wrong name -- using it should lead to an
+  // error.
+  class MisnamedFileChecksumGenerator : public FileChecksumGenCrc32c {
+   public:
+    MisnamedFileChecksumGenerator(const FileChecksumGenContext& context)
+        : FileChecksumGenCrc32c(context) {}
+
+    const char* Name() const override { return "sha1"; }
+  };
+
+  class MisnamedFileChecksumGenFactory : public FileChecksumGenCrc32cFactory {
+   public:
+    std::unique_ptr<FileChecksumGenerator> CreateFileChecksumGenerator(
+        const FileChecksumGenContext& context) override {
+      return std::unique_ptr<FileChecksumGenerator>(
+          new MisnamedFileChecksumGenerator(context));
+    }
+  };
+
+  options.file_checksum_gen_factory.reset(new MisnamedFileChecksumGenFactory());
+  Reopen(options);
+  ASSERT_TRUE(db_->VerifyFileChecksums(ReadOptions()).IsInvalidArgument());
+}
+#endif  // !ROCKSDB_LITE
+
 // A test class for intercepting random reads and injecting artificial
 // delays. Used for testing the deadline/timeout feature
 class DBBasicTestDeadline
@@ -3363,7 +3671,7 @@ TEST_P(DBBasicTestDeadline, PointLookupDeadline) {
       std::string key = "k" + ToString(i);
       ASSERT_OK(Put(key, rnd.RandomString(100)));
     }
-    Flush();
+    ASSERT_OK(Flush());
 
     bool timedout = true;
     // A timeout will be forced when the IO counter reaches this value

db/db_block_cache_test.cc

@@ -446,6 +446,7 @@ class MockCache : public LRUCache {
                  false /*strict_capacity_limit*/, 0.0 /*high_pri_pool_ratio*/) {
   }
 
+  using ShardedCache::Insert;
   Status Insert(const Slice& key, void* value, size_t charge,
                 void (*deleter)(const Slice& key, void* value), Handle** handle,
                 Priority priority) override {
@@ -533,6 +534,7 @@ class LookupLiarCache : public CacheWrapper {
   explicit LookupLiarCache(std::shared_ptr<Cache> target)
       : CacheWrapper(std::move(target)) {}
 
+  using Cache::Lookup;
   Handle* Lookup(const Slice& key, Statistics* stats) override {
     if (nth_lookup_not_found_ == 1) {
       nth_lookup_not_found_ = 0;
@@ -677,7 +679,7 @@ TEST_F(DBBlockCacheTest, ParanoidFileChecks) {
   // Create a new SST file. This will further trigger a compaction
   // and generate another file.
   ASSERT_OK(Flush(1));
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
   ASSERT_EQ(3, /* Totally 3 files created up to now */
             TestGetTickerCount(options, BLOCK_CACHE_ADD));
 
@@ -692,7 +694,7 @@ TEST_F(DBBlockCacheTest, ParanoidFileChecks) {
   ASSERT_OK(Put(1, "1_key4", "val4"));
   ASSERT_OK(Put(1, "9_key4", "val4"));
   ASSERT_OK(Flush(1));
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
   ASSERT_EQ(3, /* Totally 3 files created up to now */
             TestGetTickerCount(options, BLOCK_CACHE_ADD));
 }
@@ -837,8 +839,9 @@ TEST_F(DBBlockCacheTest, CacheCompressionDict) {
   Random rnd(301);
   for (auto compression_type : compression_types) {
     Options options = CurrentOptions();
-    options.compression = compression_type;
-    options.compression_opts.max_dict_bytes = 4096;
+    options.bottommost_compression = compression_type;
+    options.bottommost_compression_opts.max_dict_bytes = 4096;
+    options.bottommost_compression_opts.enabled = true;
     options.create_if_missing = true;
     options.num_levels = 2;
     options.statistics = ROCKSDB_NAMESPACE::CreateDBStatistics();
@@ -859,7 +862,7 @@ TEST_F(DBBlockCacheTest, CacheCompressionDict) {
       }
       ASSERT_OK(Flush());
     }
-    dbfull()->TEST_WaitForCompact();
+    ASSERT_OK(dbfull()->TEST_WaitForCompact());
     ASSERT_EQ(0, NumTableFilesAtLevel(0));
     ASSERT_EQ(kNumFiles, NumTableFilesAtLevel(1));
 
@@ -991,6 +994,9 @@ TEST_P(DBBlockCachePinningTest, TwoLevelDB) {
       ++expected_index_misses;
     }
   }
+  if (unpartitioned_pinning_ == PinningTier::kNone) {
+    ++expected_compression_dict_misses;
+  }
   ASSERT_EQ(expected_filter_misses,
             TestGetTickerCount(options, BLOCK_CACHE_FILTER_MISS));
   ASSERT_EQ(expected_index_misses,

db/db_bloom_filter_test.cc

@@ -7,6 +7,9 @@
 // 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 <iomanip>
+#include <sstream>
+
 #include "db/db_test_util.h"
 #include "options/options_helper.h"
 #include "port/stack_trace.h"
@@ -128,8 +131,8 @@ TEST_P(DBBloomFilterTestDefFormatVersion, KeyMayExist) {
     ASSERT_EQ(cache_added, TestGetTickerCount(options, BLOCK_CACHE_ADD));
 
     ASSERT_OK(Flush(1));
-    dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1],
-                                true /* disallow trivial move */);
+    ASSERT_OK(dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1],
+                                          true /* disallow trivial move */));
 
     numopen = TestGetTickerCount(options, NO_FILE_OPENS);
     cache_added = TestGetTickerCount(options, BLOCK_CACHE_ADD);
@@ -178,7 +181,7 @@ TEST_F(DBBloomFilterTest, GetFilterByPrefixBloomCustomPrefixExtractor) {
     ASSERT_OK(dbfull()->Put(wo, "barbarbar2", "foo2"));
     ASSERT_OK(dbfull()->Put(wo, "foofoofoo", "bar"));
 
-    dbfull()->Flush(fo);
+    ASSERT_OK(dbfull()->Flush(fo));
 
     ASSERT_EQ("foo", Get("barbarbar"));
     ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 0);
@@ -244,7 +247,7 @@ TEST_F(DBBloomFilterTest, GetFilterByPrefixBloom) {
     ASSERT_OK(dbfull()->Put(wo, "barbarbar2", "foo2"));
     ASSERT_OK(dbfull()->Put(wo, "foofoofoo", "bar"));
 
-    dbfull()->Flush(fo);
+    ASSERT_OK(dbfull()->Flush(fo));
 
     ASSERT_EQ("foo", Get("barbarbar"));
     ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 0);
@@ -297,7 +300,7 @@ TEST_F(DBBloomFilterTest, WholeKeyFilterProp) {
     // ranges.
     ASSERT_OK(dbfull()->Put(wo, "aaa", ""));
     ASSERT_OK(dbfull()->Put(wo, "zzz", ""));
-    dbfull()->Flush(fo);
+    ASSERT_OK(dbfull()->Flush(fo));
 
     Reopen(options);
     ASSERT_EQ("NOT_FOUND", Get("foo"));
@@ -328,7 +331,7 @@ TEST_F(DBBloomFilterTest, WholeKeyFilterProp) {
     // ranges.
     ASSERT_OK(dbfull()->Put(wo, "aaa", ""));
     ASSERT_OK(dbfull()->Put(wo, "zzz", ""));
-    db_->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+    ASSERT_OK(db_->CompactRange(CompactRangeOptions(), nullptr, nullptr));
 
     // Reopen with both of whole key off and prefix extractor enabled.
     // Still no bloom filter should be used.
@@ -351,7 +354,7 @@ TEST_F(DBBloomFilterTest, WholeKeyFilterProp) {
     // ranges.
     ASSERT_OK(dbfull()->Put(wo, "aaa", ""));
     ASSERT_OK(dbfull()->Put(wo, "zzz", ""));
-    db_->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+    ASSERT_OK(db_->CompactRange(CompactRangeOptions(), nullptr, nullptr));
 
     options.prefix_extractor.reset();
     bbto.whole_key_filtering = true;
@@ -364,7 +367,7 @@ TEST_F(DBBloomFilterTest, WholeKeyFilterProp) {
     // not filtered out by key ranges.
     ASSERT_OK(dbfull()->Put(wo, "aaa", ""));
     ASSERT_OK(dbfull()->Put(wo, "zzz", ""));
-    Flush();
+    ASSERT_OK(Flush());
 
     // Now we have two files:
     // File 1: An older file with prefix bloom.
@@ -467,7 +470,7 @@ TEST_P(DBBloomFilterTestWithParam, BloomFilter) {
     for (int i = 0; i < N; i += 100) {
       ASSERT_OK(Put(1, Key(i), Key(i)));
     }
-    Flush(1);
+    ASSERT_OK(Flush(1));
 
     // Prevent auto compactions triggered by seeks
     env_->delay_sstable_sync_.store(true, std::memory_order_release);
@@ -514,24 +517,24 @@ INSTANTIATE_TEST_CASE_P(
     ::testing::Values(
         std::make_tuple(BFP::kDeprecatedBlock, false,
                         test::kDefaultFormatVersion),
-        std::make_tuple(BFP::kAuto, true, test::kDefaultFormatVersion),
-        std::make_tuple(BFP::kAuto, false, test::kDefaultFormatVersion)));
+        std::make_tuple(BFP::kAutoBloom, true, test::kDefaultFormatVersion),
+        std::make_tuple(BFP::kAutoBloom, false, test::kDefaultFormatVersion)));
 
 INSTANTIATE_TEST_CASE_P(
     FormatDef, DBBloomFilterTestWithParam,
     ::testing::Values(
         std::make_tuple(BFP::kDeprecatedBlock, false,
                         test::kDefaultFormatVersion),
-        std::make_tuple(BFP::kAuto, true, test::kDefaultFormatVersion),
-        std::make_tuple(BFP::kAuto, false, test::kDefaultFormatVersion)));
+        std::make_tuple(BFP::kAutoBloom, true, test::kDefaultFormatVersion),
+        std::make_tuple(BFP::kAutoBloom, false, test::kDefaultFormatVersion)));
 
 INSTANTIATE_TEST_CASE_P(
     FormatLatest, DBBloomFilterTestWithParam,
     ::testing::Values(
         std::make_tuple(BFP::kDeprecatedBlock, false,
                         test::kLatestFormatVersion),
-        std::make_tuple(BFP::kAuto, true, test::kLatestFormatVersion),
-        std::make_tuple(BFP::kAuto, false, test::kLatestFormatVersion)));
+        std::make_tuple(BFP::kAutoBloom, true, test::kLatestFormatVersion),
+        std::make_tuple(BFP::kAutoBloom, false, test::kLatestFormatVersion)));
 #endif  // ROCKSDB_VALGRIND_RUN
 
 TEST_F(DBBloomFilterTest, BloomFilterRate) {
@@ -807,6 +810,7 @@ class TestingContextCustomFilterPolicy
 TEST_F(DBBloomFilterTest, ContextCustomFilterPolicy) {
   for (bool fifo : {true, false}) {
     Options options = CurrentOptions();
+    options.max_open_files = fifo ? -1 : options.max_open_files;
     options.statistics = ROCKSDB_NAMESPACE::CreateDBStatistics();
     options.compaction_style =
         fifo ? kCompactionStyleFIFO : kCompactionStyleLevel;
@@ -817,6 +821,7 @@ TEST_F(DBBloomFilterTest, ContextCustomFilterPolicy) {
     table_options.format_version = 5;
     options.table_factory.reset(NewBlockBasedTableFactory(table_options));
 
+    TryReopen(options);
     CreateAndReopenWithCF({fifo ? "abe" : "bob"}, options);
 
     const int maxKey = 10000;
@@ -880,7 +885,7 @@ TEST_F(DBBloomFilterTest, ContextCustomFilterPolicy) {
 
     // Destroy
     ASSERT_OK(dbfull()->DropColumnFamily(handles_[1]));
-    dbfull()->DestroyColumnFamilyHandle(handles_[1]);
+    ASSERT_OK(dbfull()->DestroyColumnFamilyHandle(handles_[1]));
     handles_[1] = nullptr;
   }
 }
@@ -1444,9 +1449,9 @@ void PrefixScanInit(DBBloomFilterTest* dbtest) {
   snprintf(buf, sizeof(buf), "%02d______:end", 10);
   keystr = std::string(buf);
   ASSERT_OK(dbtest->Put(keystr, keystr));
-  dbtest->Flush();
-  dbtest->dbfull()->CompactRange(CompactRangeOptions(), nullptr,
-                                 nullptr);  // move to level 1
+  ASSERT_OK(dbtest->Flush());
+  ASSERT_OK(dbtest->dbfull()->CompactRange(CompactRangeOptions(), nullptr,
+                                           nullptr));  // move to level 1
 
   // GROUP 1
   for (int i = 1; i <= small_range_sstfiles; i++) {
@@ -1563,21 +1568,21 @@ TEST_F(DBBloomFilterTest, OptimizeFiltersForHits) {
   for (int key : keys) {
     ASSERT_OK(Put(1, Key(key), "val"));
     if (++num_inserted % 1000 == 0) {
-      dbfull()->TEST_WaitForFlushMemTable();
-      dbfull()->TEST_WaitForCompact();
+      ASSERT_OK(dbfull()->TEST_WaitForFlushMemTable());
+      ASSERT_OK(dbfull()->TEST_WaitForCompact());
     }
   }
   ASSERT_OK(Put(1, Key(0), "val"));
   ASSERT_OK(Put(1, Key(numkeys), "val"));
   ASSERT_OK(Flush(1));
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
 
   if (NumTableFilesAtLevel(0, 1) == 0) {
     // No Level 0 file. Create one.
     ASSERT_OK(Put(1, Key(0), "val"));
     ASSERT_OK(Put(1, Key(numkeys), "val"));
     ASSERT_OK(Flush(1));
-    dbfull()->TEST_WaitForCompact();
+    ASSERT_OK(dbfull()->TEST_WaitForCompact());
   }
 
   for (int i = 1; i < numkeys; i += 2) {
@@ -1682,7 +1687,8 @@ TEST_F(DBBloomFilterTest, OptimizeFiltersForHits) {
       BottommostLevelCompaction::kSkip;
   compact_options.change_level = true;
   compact_options.target_level = 7;
-  db_->CompactRange(compact_options, handles_[1], nullptr, nullptr);
+  ASSERT_TRUE(db_->CompactRange(compact_options, handles_[1], nullptr, nullptr)
+                  .IsNotSupported());
 
   ASSERT_EQ(trivial_move, 1);
   ASSERT_EQ(non_trivial_move, 0);
@@ -1714,10 +1720,10 @@ TEST_F(DBBloomFilterTest, OptimizeFiltersForHits) {
 
 int CountIter(std::unique_ptr<Iterator>& iter, const Slice& key) {
   int count = 0;
-  for (iter->Seek(key); iter->Valid() && iter->status() == Status::OK();
-       iter->Next()) {
+  for (iter->Seek(key); iter->Valid(); iter->Next()) {
     count++;
   }
+  EXPECT_OK(iter->status());
   return count;
 }
 
@@ -1730,6 +1736,7 @@ TEST_F(DBBloomFilterTest, DynamicBloomFilterUpperBound) {
     int using_full_builder = bfp_impl != BFP::kDeprecatedBlock;
     Options options;
     options.create_if_missing = true;
+    options.env = CurrentOptions().env;
     options.prefix_extractor.reset(NewCappedPrefixTransform(4));
     options.disable_auto_compactions = true;
     options.statistics = CreateDBStatistics();
@@ -1746,7 +1753,7 @@ TEST_F(DBBloomFilterTest, DynamicBloomFilterUpperBound) {
     ASSERT_OK(Put("abcdxxx1", "val2"));
     ASSERT_OK(Put("abcdxxx2", "val3"));
     ASSERT_OK(Put("abcdxxx3", "val4"));
-    dbfull()->Flush(FlushOptions());
+    ASSERT_OK(dbfull()->Flush(FlushOptions()));
     {
       // prefix_extractor has not changed, BF will always be read
       Slice upper_bound("abce");
@@ -1860,6 +1867,7 @@ TEST_F(DBBloomFilterTest, DynamicBloomFilterMultipleSST) {
   for (auto bfp_impl : BFP::kAllFixedImpls) {
     int using_full_builder = bfp_impl != BFP::kDeprecatedBlock;
     Options options;
+    options.env = CurrentOptions().env;
     options.create_if_missing = true;
     options.prefix_extractor.reset(NewFixedPrefixTransform(1));
     options.disable_auto_compactions = true;
@@ -1903,7 +1911,7 @@ TEST_F(DBBloomFilterTest, DynamicBloomFilterMultipleSST) {
     ASSERT_OK(Put("foo4", "bar4"));
     ASSERT_OK(Put("foq5", "bar5"));
     ASSERT_OK(Put("fpb", "1"));
-    dbfull()->Flush(FlushOptions());
+    ASSERT_OK(dbfull()->Flush(FlushOptions()));
     {
       // BF is cappped:3 now
       std::unique_ptr<Iterator> iter_tmp(db_->NewIterator(read_options));
@@ -1927,7 +1935,7 @@ TEST_F(DBBloomFilterTest, DynamicBloomFilterMultipleSST) {
     ASSERT_OK(Put("foo7", "bar7"));
     ASSERT_OK(Put("foq8", "bar8"));
     ASSERT_OK(Put("fpc", "2"));
-    dbfull()->Flush(FlushOptions());
+    ASSERT_OK(dbfull()->Flush(FlushOptions()));
     {
       // BF is fixed:2 now
       std::unique_ptr<Iterator> iter_tmp(db_->NewIterator(read_options));
@@ -2038,10 +2046,10 @@ TEST_F(DBBloomFilterTest, DynamicBloomFilterNewColumnFamily) {
       ASSERT_EQ(TestGetTickerCount(options, BLOOM_FILTER_PREFIX_USEFUL), 0);
     }
     ASSERT_OK(dbfull()->DropColumnFamily(handles_[2]));
-    dbfull()->DestroyColumnFamilyHandle(handles_[2]);
+    ASSERT_OK(dbfull()->DestroyColumnFamilyHandle(handles_[2]));
     handles_[2] = nullptr;
     ASSERT_OK(dbfull()->DropColumnFamily(handles_[1]));
-    dbfull()->DestroyColumnFamilyHandle(handles_[1]);
+    ASSERT_OK(dbfull()->DestroyColumnFamilyHandle(handles_[1]));
     handles_[1] = nullptr;
     iteration++;
   }
@@ -2052,6 +2060,7 @@ TEST_F(DBBloomFilterTest, DynamicBloomFilterNewColumnFamily) {
 TEST_F(DBBloomFilterTest, DynamicBloomFilterOptions) {
   for (auto bfp_impl : BFP::kAllFixedImpls) {
     Options options;
+    options.env = CurrentOptions().env;
     options.create_if_missing = true;
     options.prefix_extractor.reset(NewFixedPrefixTransform(1));
     options.disable_auto_compactions = true;
@@ -2113,6 +2122,54 @@ TEST_F(DBBloomFilterTest, DynamicBloomFilterOptions) {
   }
 }
 
+TEST_F(DBBloomFilterTest, SeekForPrevWithPartitionedFilters) {
+  Options options = CurrentOptions();
+  constexpr size_t kNumKeys = 10000;
+  static_assert(kNumKeys <= 10000, "kNumKeys have to be <= 10000");
+  options.memtable_factory.reset(new SpecialSkipListFactory(kNumKeys + 10));
+  options.create_if_missing = true;
+  constexpr size_t kPrefixLength = 4;
+  options.prefix_extractor.reset(NewFixedPrefixTransform(kPrefixLength));
+  options.compression = kNoCompression;
+  BlockBasedTableOptions bbto;
+  bbto.filter_policy.reset(NewBloomFilterPolicy(50));
+  bbto.index_shortening =
+      BlockBasedTableOptions::IndexShorteningMode::kNoShortening;
+  bbto.block_size = 128;
+  bbto.metadata_block_size = 128;
+  bbto.partition_filters = true;
+  bbto.index_type = BlockBasedTableOptions::IndexType::kTwoLevelIndexSearch;
+  options.table_factory.reset(NewBlockBasedTableFactory(bbto));
+  DestroyAndReopen(options);
+
+  const std::string value(64, '\0');
+
+  WriteOptions write_opts;
+  write_opts.disableWAL = true;
+  for (size_t i = 0; i < kNumKeys; ++i) {
+    std::ostringstream oss;
+    oss << std::setfill('0') << std::setw(4) << std::fixed << i;
+    ASSERT_OK(db_->Put(write_opts, oss.str(), value));
+  }
+  ASSERT_OK(Flush());
+
+  ReadOptions read_opts;
+  // Use legacy, implicit prefix seek
+  read_opts.total_order_seek = false;
+  read_opts.auto_prefix_mode = false;
+  std::unique_ptr<Iterator> it(db_->NewIterator(read_opts));
+  for (size_t i = 0; i < kNumKeys; ++i) {
+    // Seek with a key after each one added but with same prefix. One will
+    // surely cross a partition boundary.
+    std::ostringstream oss;
+    oss << std::setfill('0') << std::setw(4) << std::fixed << i << "a";
+    it->SeekForPrev(oss.str());
+    ASSERT_OK(it->status());
+    ASSERT_TRUE(it->Valid());
+  }
+  it.reset();
+}
+
 #endif  // ROCKSDB_LITE
 
 }  // namespace ROCKSDB_NAMESPACE

db/db_compaction_filter_test.cc

@@ -42,7 +42,7 @@ class DBTestCompactionFilterWithCompactParam
         option_config_ == kUniversalSubcompactions) {
       assert(options.max_subcompactions > 1);
     }
-    TryReopen(options);
+    Reopen(options);
   }
 };
 
@@ -276,7 +276,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilter) {
   for (int i = 0; i < 100000; i++) {
     char key[100];
     snprintf(key, sizeof(key), "B%010d", i);
-    Put(1, key, value);
+    ASSERT_OK(Put(1, key, value));
   }
   ASSERT_OK(Flush(1));
 
@@ -284,10 +284,10 @@ TEST_F(DBTestCompactionFilter, CompactionFilter) {
   // the compaction is each level invokes the filter for
   // all the keys in that level.
   cfilter_count = 0;
-  dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1]);
+  ASSERT_OK(dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1]));
   ASSERT_EQ(cfilter_count, 100000);
   cfilter_count = 0;
-  dbfull()->TEST_CompactRange(1, nullptr, nullptr, handles_[1]);
+  ASSERT_OK(dbfull()->TEST_CompactRange(1, nullptr, nullptr, handles_[1]));
   ASSERT_EQ(cfilter_count, 100000);
 
   ASSERT_EQ(NumTableFilesAtLevel(0, 1), 0);
@@ -314,13 +314,14 @@ TEST_F(DBTestCompactionFilter, CompactionFilter) {
     ASSERT_OK(iter->status());
     while (iter->Valid()) {
       ParsedInternalKey ikey(Slice(), 0, kTypeValue);
-      ASSERT_OK(ParseInternalKey(iter->key(), &ikey));
+      ASSERT_OK(ParseInternalKey(iter->key(), &ikey, true /* log_err_key */));
       total++;
       if (ikey.sequence != 0) {
         count++;
       }
       iter->Next();
     }
+    ASSERT_OK(iter->status());
   }
   ASSERT_EQ(total, 100000);
   ASSERT_EQ(count, 0);
@@ -337,10 +338,10 @@ TEST_F(DBTestCompactionFilter, CompactionFilter) {
   // means that all keys should pass at least once
   // via the compaction filter
   cfilter_count = 0;
-  dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1]);
+  ASSERT_OK(dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1]));
   ASSERT_EQ(cfilter_count, 100000);
   cfilter_count = 0;
-  dbfull()->TEST_CompactRange(1, nullptr, nullptr, handles_[1]);
+  ASSERT_OK(dbfull()->TEST_CompactRange(1, nullptr, nullptr, handles_[1]));
   ASSERT_EQ(cfilter_count, 100000);
   ASSERT_EQ(NumTableFilesAtLevel(0, 1), 0);
   ASSERT_EQ(NumTableFilesAtLevel(1, 1), 0);
@@ -369,10 +370,10 @@ TEST_F(DBTestCompactionFilter, CompactionFilter) {
   // verify that at the end of the compaction process,
   // nothing is left.
   cfilter_count = 0;
-  dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1]);
+  ASSERT_OK(dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1]));
   ASSERT_EQ(cfilter_count, 100000);
   cfilter_count = 0;
-  dbfull()->TEST_CompactRange(1, nullptr, nullptr, handles_[1]);
+  ASSERT_OK(dbfull()->TEST_CompactRange(1, nullptr, nullptr, handles_[1]));
   ASSERT_EQ(cfilter_count, 0);
   ASSERT_EQ(NumTableFilesAtLevel(0, 1), 0);
   ASSERT_EQ(NumTableFilesAtLevel(1, 1), 0);
@@ -387,6 +388,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilter) {
       count++;
       iter->Next();
     }
+    ASSERT_OK(iter->status());
     ASSERT_EQ(count, 0);
   }
 
@@ -405,7 +407,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilter) {
     ASSERT_OK(iter->status());
     while (iter->Valid()) {
       ParsedInternalKey ikey(Slice(), 0, kTypeValue);
-      ASSERT_OK(ParseInternalKey(iter->key(), &ikey));
+      ASSERT_OK(ParseInternalKey(iter->key(), &ikey, true /* log_err_key */));
       ASSERT_NE(ikey.sequence, (unsigned)0);
       count++;
       iter->Next();
@@ -427,9 +429,9 @@ TEST_F(DBTestCompactionFilter, CompactionFilterDeletesAll) {
   // put some data
   for (int table = 0; table < 4; ++table) {
     for (int i = 0; i < 10 + table; ++i) {
-      Put(ToString(table * 100 + i), "val");
+      ASSERT_OK(Put(ToString(table * 100 + i), "val"));
     }
-    Flush();
+    ASSERT_OK(Flush());
   }
 
   // this will produce empty file (delete compaction filter)
@@ -440,6 +442,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilterDeletesAll) {
 
   Iterator* itr = db_->NewIterator(ReadOptions());
   itr->SeekToFirst();
+  ASSERT_OK(itr->status());
   // empty db
   ASSERT_TRUE(!itr->Valid());
 
@@ -463,25 +466,25 @@ TEST_P(DBTestCompactionFilterWithCompactParam,
   for (int i = 0; i < 100001; i++) {
     char key[100];
     snprintf(key, sizeof(key), "B%010d", i);
-    Put(1, key, value);
+    ASSERT_OK(Put(1, key, value));
   }
 
   // push all files to  lower levels
   ASSERT_OK(Flush(1));
   if (option_config_ != kUniversalCompactionMultiLevel &&
       option_config_ != kUniversalSubcompactions) {
-    dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1]);
-    dbfull()->TEST_CompactRange(1, nullptr, nullptr, handles_[1]);
+    ASSERT_OK(dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1]));
+    ASSERT_OK(dbfull()->TEST_CompactRange(1, nullptr, nullptr, handles_[1]));
   } else {
-    dbfull()->CompactRange(CompactRangeOptions(), handles_[1], nullptr,
-                           nullptr);
+    ASSERT_OK(dbfull()->CompactRange(CompactRangeOptions(), handles_[1],
+                                     nullptr, nullptr));
   }
 
   // re-write all data again
   for (int i = 0; i < 100001; i++) {
     char key[100];
     snprintf(key, sizeof(key), "B%010d", i);
-    Put(1, key, value);
+    ASSERT_OK(Put(1, key, value));
   }
 
   // push all files to  lower levels. This should
@@ -489,11 +492,11 @@ TEST_P(DBTestCompactionFilterWithCompactParam,
   ASSERT_OK(Flush(1));
   if (option_config_ != kUniversalCompactionMultiLevel &&
       option_config_ != kUniversalSubcompactions) {
-    dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1]);
-    dbfull()->TEST_CompactRange(1, nullptr, nullptr, handles_[1]);
+    ASSERT_OK(dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1]));
+    ASSERT_OK(dbfull()->TEST_CompactRange(1, nullptr, nullptr, handles_[1]));
   } else {
-    dbfull()->CompactRange(CompactRangeOptions(), handles_[1], nullptr,
-                           nullptr);
+    ASSERT_OK(dbfull()->CompactRange(CompactRangeOptions(), handles_[1],
+                                     nullptr, nullptr));
   }
 
   // verify that all keys now have the new value that
@@ -531,7 +534,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilterWithMergeOperator) {
   ASSERT_OK(Flush());
   std::string newvalue = Get("foo");
   ASSERT_EQ(newvalue, three);
-  dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+  ASSERT_OK(dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr));
   newvalue = Get("foo");
   ASSERT_EQ(newvalue, three);
 
@@ -539,12 +542,12 @@ TEST_F(DBTestCompactionFilter, CompactionFilterWithMergeOperator) {
   // merge keys.
   ASSERT_OK(db_->Put(WriteOptions(), "bar", two));
   ASSERT_OK(Flush());
-  dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+  ASSERT_OK(dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr));
   newvalue = Get("bar");
   ASSERT_EQ("NOT_FOUND", newvalue);
   ASSERT_OK(db_->Merge(WriteOptions(), "bar", two));
   ASSERT_OK(Flush());
-  dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+  ASSERT_OK(dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr));
   newvalue = Get("bar");
   ASSERT_EQ(two, two);
 
@@ -555,7 +558,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilterWithMergeOperator) {
   ASSERT_OK(Flush());
   newvalue = Get("foobar");
   ASSERT_EQ(newvalue, three);
-  dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+  ASSERT_OK(dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr));
   newvalue = Get("foobar");
   ASSERT_EQ(newvalue, three);
 
@@ -568,7 +571,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilterWithMergeOperator) {
   ASSERT_OK(Flush());
   newvalue = Get("barfoo");
   ASSERT_EQ(newvalue, four);
-  dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+  ASSERT_OK(dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr));
   newvalue = Get("barfoo");
   ASSERT_EQ(newvalue, four);
 }
@@ -590,21 +593,21 @@ TEST_F(DBTestCompactionFilter, CompactionFilterContextManual) {
     for (int i = 0; i < num_keys_per_file; i++) {
       char key[100];
       snprintf(key, sizeof(key), "B%08d%02d", i, j);
-      Put(key, value);
+      ASSERT_OK(Put(key, value));
     }
-    dbfull()->TEST_FlushMemTable();
+    ASSERT_OK(dbfull()->TEST_FlushMemTable());
     // Make sure next file is much smaller so automatic compaction will not
     // be triggered.
     num_keys_per_file /= 2;
   }
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
 
   // Force a manual compaction
   cfilter_count = 0;
   filter->expect_manual_compaction_.store(true);
   filter->expect_full_compaction_.store(true);
   filter->expect_cf_id_.store(0);
-  dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+  ASSERT_OK(dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr));
   ASSERT_EQ(cfilter_count, 700);
   ASSERT_EQ(NumSortedRuns(0), 1);
   ASSERT_TRUE(filter->compaction_filter_created());
@@ -624,7 +627,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilterContextManual) {
     ASSERT_OK(iter->status());
     while (iter->Valid()) {
       ParsedInternalKey ikey(Slice(), 0, kTypeValue);
-      ASSERT_OK(ParseInternalKey(iter->key(), &ikey));
+      ASSERT_OK(ParseInternalKey(iter->key(), &ikey, true /* log_err_key */));
       total++;
       if (ikey.sequence != 0) {
         count++;
@@ -654,14 +657,14 @@ TEST_F(DBTestCompactionFilter, CompactionFilterContextCfId) {
     for (int i = 0; i < num_keys_per_file; i++) {
       char key[100];
       snprintf(key, sizeof(key), "B%08d%02d", i, j);
-      Put(1, key, value);
+      ASSERT_OK(Put(1, key, value));
     }
-    Flush(1);
+    ASSERT_OK(Flush(1));
     // Make sure next file is much smaller so automatic compaction will not
     // be triggered.
     num_keys_per_file /= 2;
   }
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
 
   ASSERT_TRUE(filter->compaction_filter_created());
 }
@@ -680,9 +683,9 @@ TEST_F(DBTestCompactionFilter, CompactionFilterIgnoreSnapshot) {
   const Snapshot* snapshot = nullptr;
   for (int table = 0; table < 4; ++table) {
     for (int i = 0; i < 10; ++i) {
-      Put(ToString(table * 100 + i), "val");
+      ASSERT_OK(Put(ToString(table * 100 + i), "val"));
     }
-    Flush();
+    ASSERT_OK(Flush());
 
     if (table == 0) {
       snapshot = db_->GetSnapshot();
@@ -702,6 +705,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilterIgnoreSnapshot) {
     read_options.snapshot = snapshot;
     std::unique_ptr<Iterator> iter(db_->NewIterator(read_options));
     iter->SeekToFirst();
+    ASSERT_OK(iter->status());
     int count = 0;
     while (iter->Valid()) {
       count++;
@@ -710,6 +714,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilterIgnoreSnapshot) {
     ASSERT_EQ(count, 6);
     read_options.snapshot = nullptr;
     std::unique_ptr<Iterator> iter1(db_->NewIterator(read_options));
+    ASSERT_OK(iter1->status());
     iter1->SeekToFirst();
     count = 0;
     while (iter1->Valid()) {
@@ -740,9 +745,9 @@ TEST_F(DBTestCompactionFilter, SkipUntil) {
     for (int i = table * 6; i < 39 + table * 11; ++i) {
       char key[100];
       snprintf(key, sizeof(key), "%010d", table * 100 + i);
-      Put(key, std::to_string(table * 1000 + i));
+      ASSERT_OK(Put(key, std::to_string(table * 1000 + i)));
     }
-    Flush();
+    ASSERT_OK(Flush());
   }
 
   cfilter_skips = 0;
@@ -781,10 +786,10 @@ TEST_F(DBTestCompactionFilter, SkipUntilWithBloomFilter) {
   options.create_if_missing = true;
   DestroyAndReopen(options);
 
-  Put("0000000010", "v10");
-  Put("0000000020", "v20");  // skipped
-  Put("0000000050", "v50");
-  Flush();
+  ASSERT_OK(Put("0000000010", "v10"));
+  ASSERT_OK(Put("0000000020", "v20"));  // skipped
+  ASSERT_OK(Put("0000000050", "v50"));
+  ASSERT_OK(Flush());
 
   cfilter_skips = 0;
   EXPECT_OK(db_->CompactRange(CompactRangeOptions(), nullptr, nullptr));
@@ -822,13 +827,13 @@ TEST_F(DBTestCompactionFilter, IgnoreSnapshotsFalse) {
   options.compaction_filter = new TestNotSupportedFilter();
   DestroyAndReopen(options);
 
-  Put("a", "v10");
-  Put("z", "v20");
-  Flush();
+  ASSERT_OK(Put("a", "v10"));
+  ASSERT_OK(Put("z", "v20"));
+  ASSERT_OK(Flush());
 
-  Put("a", "v10");
-  Put("z", "v20");
-  Flush();
+  ASSERT_OK(Put("a", "v10"));
+  ASSERT_OK(Put("z", "v20"));
+  ASSERT_OK(Flush());
 
   // Comapction should fail because IgnoreSnapshots() = false
   EXPECT_TRUE(db_->CompactRange(CompactRangeOptions(), nullptr, nullptr)

db/db_dynamic_level_test.cc

@@ -13,6 +13,7 @@
 #if !defined(ROCKSDB_LITE)
 
 #include "db/db_test_util.h"
+#include "env/mock_env.h"
 #include "port/port.h"
 #include "port/stack_trace.h"
 #include "util/random.h"
@@ -102,7 +103,8 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase) {
       }
 
       // Test compact range works
-      dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+      ASSERT_OK(
+          dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr));
       // All data should be in the last level.
       ColumnFamilyMetaData cf_meta;
       db_->GetColumnFamilyMetaData(&cf_meta);
@@ -141,6 +143,7 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
   options.max_background_compactions = 2;
   options.num_levels = 5;
   options.max_compaction_bytes = 0;  // Force not expanding in compactions
+  options.db_host_id = "";  // Setting this messes up the file size calculation
   BlockBasedTableOptions table_options;
   table_options.block_size = 1024;
   options.table_factory.reset(NewBlockBasedTableFactory(table_options));
@@ -165,8 +168,8 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
   ASSERT_OK(dbfull()->SetOptions({
       {"disable_auto_compactions", "false"},
   }));
-  Flush();
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(Flush());
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
   ASSERT_TRUE(db_->GetIntProperty("rocksdb.base-level", &int_prop));
   ASSERT_EQ(4U, int_prop);
 
@@ -183,8 +186,8 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
   ASSERT_OK(dbfull()->SetOptions({
       {"disable_auto_compactions", "false"},
   }));
-  Flush();
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(Flush());
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
   ASSERT_TRUE(db_->GetIntProperty("rocksdb.base-level", &int_prop));
   ASSERT_EQ(3U, int_prop);
   ASSERT_TRUE(db_->GetProperty("rocksdb.num-files-at-level1", &str_prop));
@@ -204,8 +207,8 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
   ASSERT_OK(dbfull()->SetOptions({
       {"disable_auto_compactions", "false"},
   }));
-  Flush();
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(Flush());
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
   ASSERT_TRUE(db_->GetIntProperty("rocksdb.base-level", &int_prop));
   ASSERT_EQ(3U, int_prop);
 
@@ -233,8 +236,8 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
   }));
 
   TEST_SYNC_POINT("DynamicLevelMaxBytesBase2:0");
-  Flush();
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(Flush());
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
   ASSERT_TRUE(db_->GetIntProperty("rocksdb.base-level", &int_prop));
   ASSERT_EQ(2U, int_prop);
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
@@ -263,7 +266,7 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBase2) {
   }
   TEST_SYNC_POINT("DynamicLevelMaxBytesBase2:2");
 
-  Flush();
+  ASSERT_OK(Flush());
 
   thread.join();
 
@@ -301,7 +304,7 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesCompactRange) {
   DestroyAndReopen(options);
 
   // Compact against empty DB
-  dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+  ASSERT_OK(dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr));
 
   uint64_t int_prop;
   std::string str_prop;
@@ -315,13 +318,13 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesCompactRange) {
     ASSERT_OK(
         Put(Key(static_cast<int>(rnd.Uniform(kMaxKey))), rnd.RandomString(80)));
   }
-  Flush();
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(Flush());
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
   if (NumTableFilesAtLevel(0) == 0) {
     // Make sure level 0 is not empty
     ASSERT_OK(
         Put(Key(static_cast<int>(rnd.Uniform(kMaxKey))), rnd.RandomString(80)));
-    Flush();
+    ASSERT_OK(Flush());
   }
 
   ASSERT_TRUE(db_->GetIntProperty("rocksdb.base-level", &int_prop));
@@ -342,7 +345,7 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesCompactRange) {
       });
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
 
-  dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+  ASSERT_OK(dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr));
   ASSERT_EQ(output_levels.size(), 2);
   ASSERT_TRUE(output_levels.find(3) != output_levels.end());
   ASSERT_TRUE(output_levels.find(4) != output_levels.end());
@@ -388,8 +391,8 @@ TEST_F(DBTestDynamicLevel, DynamicLevelMaxBytesBaseInc) {
     PutFixed32(&value, static_cast<uint32_t>(i));
     ASSERT_OK(Put(Key(i), value));
   }
-  Flush();
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(Flush());
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
 
   ASSERT_EQ(non_trivial, 0);
@@ -448,7 +451,7 @@ TEST_F(DBTestDynamicLevel, DISABLED_MigrateToDynamicLevelMaxBytesBase) {
     ASSERT_OK(Delete(Key(i / 10)));
   }
   verify_func(total_keys, false);
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
 
   options.level_compaction_dynamic_level_bytes = true;
   options.disable_auto_compactions = true;
@@ -463,7 +466,7 @@ TEST_F(DBTestDynamicLevel, DISABLED_MigrateToDynamicLevelMaxBytesBase) {
     CompactRangeOptions compact_options;
     compact_options.change_level = true;
     compact_options.target_level = options.num_levels - 1;
-    dbfull()->CompactRange(compact_options, nullptr, nullptr);
+    ASSERT_OK(dbfull()->CompactRange(compact_options, nullptr, nullptr));
     compaction_finished.store(true);
   });
   do {
@@ -483,7 +486,7 @@ TEST_F(DBTestDynamicLevel, DISABLED_MigrateToDynamicLevelMaxBytesBase) {
   }
 
   verify_func(total_keys2, false);
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
   verify_func(total_keys2, false);
 
   // Base level is not level 1

db/db_encryption_test.cc

@@ -18,6 +18,13 @@ class DBEncryptionTest : public DBTestBase {
  public:
   DBEncryptionTest()
       : DBTestBase("/db_encryption_test", /*env_do_fsync=*/true) {}
+  Env* GetTargetEnv() {
+    if (encrypted_env_ != nullptr) {
+      return (static_cast<EnvWrapper*>(encrypted_env_))->target();
+    } else {
+      return env_;
+    }
+  }
 };
 
 #ifndef ROCKSDB_LITE
@@ -34,20 +41,20 @@ TEST_F(DBEncryptionTest, CheckEncrypted) {
   auto status = env_->GetChildren(dbname_, &fileNames);
   ASSERT_OK(status);
 
-  auto defaultEnv = Env::Default();
+  Env* target = GetTargetEnv();
   int hits = 0;
   for (auto it = fileNames.begin() ; it != fileNames.end(); ++it) {
-    if ((*it == "..") || (*it == ".")) {
+    if (*it == "LOCK") {
       continue;
     }
     auto filePath = dbname_ + "/" + *it;
     std::unique_ptr<SequentialFile> seqFile;
     auto envOptions = EnvOptions(CurrentOptions());
-    status = defaultEnv->NewSequentialFile(filePath, &seqFile, envOptions);
+    status = target->NewSequentialFile(filePath, &seqFile, envOptions);
     ASSERT_OK(status);
 
     uint64_t fileSize;
-    status = defaultEnv->GetFileSize(filePath, &fileSize);
+    status = target->GetFileSize(filePath, &fileSize);
     ASSERT_OK(status);
 
     std::string scratch;
@@ -85,7 +92,7 @@ TEST_F(DBEncryptionTest, CheckEncrypted) {
 }
 
 TEST_F(DBEncryptionTest, ReadEmptyFile) {
-  auto defaultEnv = Env::Default();
+  auto defaultEnv = GetTargetEnv();
 
   // create empty file for reading it back in later
   auto envOptions = EnvOptions(CurrentOptions());

db/db_flush_test.cc

@@ -11,13 +11,16 @@
 
 #include "db/db_impl/db_impl.h"
 #include "db/db_test_util.h"
+#include "env/mock_env.h"
 #include "file/filename.h"
 #include "port/port.h"
 #include "port/stack_trace.h"
+#include "rocksdb/utilities/transaction_db.h"
 #include "test_util/sync_point.h"
 #include "util/cast_util.h"
 #include "util/mutexlock.h"
 #include "utilities/fault_injection_env.h"
+#include "utilities/fault_injection_fs.h"
 
 namespace ROCKSDB_NAMESPACE {
 
@@ -79,29 +82,16 @@ 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);
   ASSERT_OK(Put("key", "value"));
-  auto* cfd =
-      static_cast_with_check<ColumnFamilyHandleImpl>(db_->DefaultColumnFamily())
-          ->cfd();
   FlushOptions flush_options;
   flush_options.wait = false;
   ASSERT_OK(dbfull()->Flush(flush_options));
   // 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");
@@ -112,9 +102,6 @@ TEST_F(DBFlushTest, SyncFail) {
 #ifndef ROCKSDB_LITE
   ASSERT_EQ("", FilesPerLevel());  // flush failed.
 #endif                             // ROCKSDB_LITE
-  // Backgroun flush job should release ref count to current version.
-  ASSERT_EQ(current_before, cfd->current());
-  ASSERT_EQ(refs_before, cfd->current()->TEST_refs());
   Destroy(options);
 }
 
@@ -179,6 +166,66 @@ TEST_F(DBFlushTest, FlushInLowPriThreadPool) {
   ASSERT_EQ(1, num_compactions);
 }
 
+// Test when flush job is submitted to low priority thread pool and when DB is
+// closed in the meanwhile, CloseHelper doesn't hang.
+TEST_F(DBFlushTest, CloseDBWhenFlushInLowPri) {
+  Options options = CurrentOptions();
+  options.max_background_flushes = 1;
+  options.max_total_wal_size = 8192;
+
+  DestroyAndReopen(options);
+  CreateColumnFamilies({"cf1", "cf2"}, options);
+
+  env_->SetBackgroundThreads(0, Env::HIGH);
+  env_->SetBackgroundThreads(1, Env::LOW);
+  test::SleepingBackgroundTask sleeping_task_low;
+  int num_flushes = 0;
+
+  SyncPoint::GetInstance()->SetCallBack("DBImpl::BGWorkFlush",
+                                        [&](void* /*arg*/) { ++num_flushes; });
+
+  int num_low_flush_unscheduled = 0;
+  SyncPoint::GetInstance()->SetCallBack(
+      "DBImpl::UnscheduleLowFlushCallback", [&](void* /*arg*/) {
+        num_low_flush_unscheduled++;
+        // There should be one flush job in low pool that needs to be
+        // unscheduled
+        ASSERT_EQ(num_low_flush_unscheduled, 1);
+      });
+
+  int num_high_flush_unscheduled = 0;
+  SyncPoint::GetInstance()->SetCallBack(
+      "DBImpl::UnscheduleHighFlushCallback", [&](void* /*arg*/) {
+        num_high_flush_unscheduled++;
+        // There should be no flush job in high pool
+        ASSERT_EQ(num_high_flush_unscheduled, 0);
+      });
+
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  ASSERT_OK(Put(0, "key1", DummyString(8192)));
+  // Block thread so that flush cannot be run and can be removed from the queue
+  // when called Unschedule.
+  env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask, &sleeping_task_low,
+                 Env::Priority::LOW);
+  sleeping_task_low.WaitUntilSleeping();
+
+  // Trigger flush and flush job will be scheduled to LOW priority thread.
+  ASSERT_OK(Put(0, "key2", DummyString(8192)));
+
+  // Close DB and flush job in low priority queue will be removed without
+  // running.
+  Close();
+  sleeping_task_low.WakeUp();
+  sleeping_task_low.WaitUntilDone();
+  ASSERT_EQ(0, num_flushes);
+
+  TryReopenWithColumnFamilies({"default", "cf1", "cf2"}, options);
+  ASSERT_OK(Put(0, "key3", DummyString(8192)));
+  ASSERT_OK(Flush(0));
+  ASSERT_EQ(1, num_flushes);
+}
+
 TEST_F(DBFlushTest, ManualFlushWithMinWriteBufferNumberToMerge) {
   Options options = CurrentOptions();
   options.write_buffer_size = 100;
@@ -453,6 +500,7 @@ TEST_F(DBFlushTest, FlushWithBlob) {
   options.enable_blob_files = true;
   options.min_blob_size = min_blob_size;
   options.disable_auto_compactions = true;
+  options.env = env_;
 
   Reopen(options);
 
@@ -470,9 +518,7 @@ TEST_F(DBFlushTest, FlushWithBlob) {
   ASSERT_OK(Flush());
 
   ASSERT_EQ(Get("key1"), short_value);
-
-  // TODO: enable once Get support is implemented for blobs
-  // ASSERT_EQ(Get("key2"), long_value);
+  ASSERT_EQ(Get("key2"), long_value);
 
   VersionSet* const versions = dbfull()->TEST_GetVersionSet();
   assert(versions);
@@ -511,26 +557,225 @@ TEST_F(DBFlushTest, FlushWithBlob) {
   const InternalStats* const internal_stats = cfd->internal_stats();
   assert(internal_stats);
 
-  const uint64_t expected_bytes =
-      table_file->fd.GetFileSize() + blob_file->GetTotalBlobBytes();
-
   const auto& compaction_stats = internal_stats->TEST_GetCompactionStats();
   ASSERT_FALSE(compaction_stats.empty());
-  ASSERT_EQ(compaction_stats[0].bytes_written, expected_bytes);
-  ASSERT_EQ(compaction_stats[0].num_output_files, 2);
+  ASSERT_EQ(compaction_stats[0].bytes_written, table_file->fd.GetFileSize());
+  ASSERT_EQ(compaction_stats[0].bytes_written_blob,
+            blob_file->GetTotalBlobBytes());
+  ASSERT_EQ(compaction_stats[0].num_output_files, 1);
+  ASSERT_EQ(compaction_stats[0].num_output_files_blob, 1);
 
   const uint64_t* const cf_stats_value = internal_stats->TEST_GetCFStatsValue();
-  ASSERT_EQ(cf_stats_value[InternalStats::BYTES_FLUSHED], expected_bytes);
+  ASSERT_EQ(cf_stats_value[InternalStats::BYTES_FLUSHED],
+            compaction_stats[0].bytes_written +
+                compaction_stats[0].bytes_written_blob);
 #endif  // ROCKSDB_LITE
 }
 
+TEST_F(DBFlushTest, FlushWithChecksumHandoff1) {
+  if (mem_env_ || encrypted_env_) {
+    ROCKSDB_GTEST_SKIP("Test requires non-mem or non-encrypted environment");
+    return;
+  }
+  std::shared_ptr<FaultInjectionTestFS> fault_fs(
+      new FaultInjectionTestFS(FileSystem::Default()));
+  std::unique_ptr<Env> fault_fs_env(NewCompositeEnv(fault_fs));
+  Options options = CurrentOptions();
+  options.write_buffer_size = 100;
+  options.max_write_buffer_number = 4;
+  options.min_write_buffer_number_to_merge = 3;
+  options.disable_auto_compactions = true;
+  options.env = fault_fs_env.get();
+  options.checksum_handoff_file_types.Add(FileType::kTableFile);
+  Reopen(options);
+
+  fault_fs->SetChecksumHandoffFuncType(ChecksumType::kCRC32c);
+  ASSERT_OK(Put("key1", "value1"));
+  ASSERT_OK(Put("key2", "value2"));
+  ASSERT_OK(dbfull()->TEST_SwitchMemtable());
+
+  // The hash does not match, write fails
+  // fault_fs->SetChecksumHandoffFuncType(ChecksumType::kxxHash);
+  // Since the file system returns IOStatus::Corruption, it is an
+  // unrecoverable error.
+  SyncPoint::GetInstance()->SetCallBack("FlushJob::Start", [&](void*) {
+    fault_fs->SetChecksumHandoffFuncType(ChecksumType::kxxHash);
+  });
+  ASSERT_OK(Put("key3", "value3"));
+  ASSERT_OK(Put("key4", "value4"));
+  SyncPoint::GetInstance()->EnableProcessing();
+  Status s = Flush();
+  ASSERT_EQ(s.severity(),
+            ROCKSDB_NAMESPACE::Status::Severity::kUnrecoverableError);
+  SyncPoint::GetInstance()->DisableProcessing();
+  Destroy(options);
+  Reopen(options);
+
+  // The file system does not support checksum handoff. The check
+  // will be ignored.
+  fault_fs->SetChecksumHandoffFuncType(ChecksumType::kNoChecksum);
+  ASSERT_OK(Put("key5", "value5"));
+  ASSERT_OK(Put("key6", "value6"));
+  ASSERT_OK(dbfull()->TEST_SwitchMemtable());
+
+  // Each write will be similated as corrupted.
+  // Since the file system returns IOStatus::Corruption, it is an
+  // unrecoverable error.
+  fault_fs->SetChecksumHandoffFuncType(ChecksumType::kCRC32c);
+  SyncPoint::GetInstance()->SetCallBack("FlushJob::Start", [&](void*) {
+    fault_fs->IngestDataCorruptionBeforeWrite();
+  });
+  ASSERT_OK(Put("key7", "value7"));
+  ASSERT_OK(Put("key8", "value8"));
+  SyncPoint::GetInstance()->EnableProcessing();
+  s = Flush();
+  ASSERT_EQ(s.severity(),
+            ROCKSDB_NAMESPACE::Status::Severity::kUnrecoverableError);
+  SyncPoint::GetInstance()->DisableProcessing();
+
+  Destroy(options);
+}
+
+TEST_F(DBFlushTest, FlushWithChecksumHandoff2) {
+  if (mem_env_ || encrypted_env_) {
+    ROCKSDB_GTEST_SKIP("Test requires non-mem or non-encrypted environment");
+    return;
+  }
+  std::shared_ptr<FaultInjectionTestFS> fault_fs(
+      new FaultInjectionTestFS(FileSystem::Default()));
+  std::unique_ptr<Env> fault_fs_env(NewCompositeEnv(fault_fs));
+  Options options = CurrentOptions();
+  options.write_buffer_size = 100;
+  options.max_write_buffer_number = 4;
+  options.min_write_buffer_number_to_merge = 3;
+  options.disable_auto_compactions = true;
+  options.env = fault_fs_env.get();
+  Reopen(options);
+
+  fault_fs->SetChecksumHandoffFuncType(ChecksumType::kCRC32c);
+  ASSERT_OK(Put("key1", "value1"));
+  ASSERT_OK(Put("key2", "value2"));
+  ASSERT_OK(Flush());
+
+  // options is not set, the checksum handoff will not be triggered
+  SyncPoint::GetInstance()->SetCallBack("FlushJob::Start", [&](void*) {
+    fault_fs->SetChecksumHandoffFuncType(ChecksumType::kxxHash);
+  });
+  ASSERT_OK(Put("key3", "value3"));
+  ASSERT_OK(Put("key4", "value4"));
+  SyncPoint::GetInstance()->EnableProcessing();
+  ASSERT_OK(Flush());
+  SyncPoint::GetInstance()->DisableProcessing();
+  Destroy(options);
+  Reopen(options);
+
+  // The file system does not support checksum handoff. The check
+  // will be ignored.
+  fault_fs->SetChecksumHandoffFuncType(ChecksumType::kNoChecksum);
+  ASSERT_OK(Put("key5", "value5"));
+  ASSERT_OK(Put("key6", "value6"));
+  ASSERT_OK(Flush());
+
+  // options is not set, the checksum handoff will not be triggered
+  fault_fs->SetChecksumHandoffFuncType(ChecksumType::kCRC32c);
+  SyncPoint::GetInstance()->SetCallBack("FlushJob::Start", [&](void*) {
+    fault_fs->IngestDataCorruptionBeforeWrite();
+  });
+  ASSERT_OK(Put("key7", "value7"));
+  ASSERT_OK(Put("key8", "value8"));
+  SyncPoint::GetInstance()->EnableProcessing();
+  ASSERT_OK(Flush());
+  SyncPoint::GetInstance()->DisableProcessing();
+
+  Destroy(options);
+}
+
+TEST_F(DBFlushTest, FlushWithChecksumHandoffManifest1) {
+  if (mem_env_ || encrypted_env_) {
+    ROCKSDB_GTEST_SKIP("Test requires non-mem or non-encrypted environment");
+    return;
+  }
+  std::shared_ptr<FaultInjectionTestFS> fault_fs(
+      new FaultInjectionTestFS(FileSystem::Default()));
+  std::unique_ptr<Env> fault_fs_env(NewCompositeEnv(fault_fs));
+  Options options = CurrentOptions();
+  options.write_buffer_size = 100;
+  options.max_write_buffer_number = 4;
+  options.min_write_buffer_number_to_merge = 3;
+  options.disable_auto_compactions = true;
+  options.env = fault_fs_env.get();
+  options.checksum_handoff_file_types.Add(FileType::kDescriptorFile);
+  fault_fs->SetChecksumHandoffFuncType(ChecksumType::kCRC32c);
+  Reopen(options);
+
+  ASSERT_OK(Put("key1", "value1"));
+  ASSERT_OK(Put("key2", "value2"));
+  ASSERT_OK(Flush());
+
+  // The hash does not match, write fails
+  // fault_fs->SetChecksumHandoffFuncType(ChecksumType::kxxHash);
+  // Since the file system returns IOStatus::Corruption, it is mapped to
+  // kFatalError error.
+  ASSERT_OK(Put("key3", "value3"));
+  SyncPoint::GetInstance()->SetCallBack(
+      "VersionSet::LogAndApply:WriteManifest", [&](void*) {
+        fault_fs->SetChecksumHandoffFuncType(ChecksumType::kxxHash);
+      });
+  ASSERT_OK(Put("key3", "value3"));
+  ASSERT_OK(Put("key4", "value4"));
+  SyncPoint::GetInstance()->EnableProcessing();
+  Status s = Flush();
+  ASSERT_EQ(s.severity(), ROCKSDB_NAMESPACE::Status::Severity::kFatalError);
+  SyncPoint::GetInstance()->DisableProcessing();
+  Destroy(options);
+}
+
+TEST_F(DBFlushTest, FlushWithChecksumHandoffManifest2) {
+  if (mem_env_ || encrypted_env_) {
+    ROCKSDB_GTEST_SKIP("Test requires non-mem or non-encrypted environment");
+    return;
+  }
+  std::shared_ptr<FaultInjectionTestFS> fault_fs(
+      new FaultInjectionTestFS(FileSystem::Default()));
+  std::unique_ptr<Env> fault_fs_env(NewCompositeEnv(fault_fs));
+  Options options = CurrentOptions();
+  options.write_buffer_size = 100;
+  options.max_write_buffer_number = 4;
+  options.min_write_buffer_number_to_merge = 3;
+  options.disable_auto_compactions = true;
+  options.env = fault_fs_env.get();
+  options.checksum_handoff_file_types.Add(FileType::kDescriptorFile);
+  fault_fs->SetChecksumHandoffFuncType(ChecksumType::kNoChecksum);
+  Reopen(options);
+  // The file system does not support checksum handoff. The check
+  // will be ignored.
+  ASSERT_OK(Put("key5", "value5"));
+  ASSERT_OK(Put("key6", "value6"));
+  ASSERT_OK(Flush());
+
+  // Each write will be similated as corrupted.
+  // Since the file system returns IOStatus::Corruption, it is mapped to
+  // kFatalError error.
+  fault_fs->SetChecksumHandoffFuncType(ChecksumType::kCRC32c);
+  SyncPoint::GetInstance()->SetCallBack(
+      "VersionSet::LogAndApply:WriteManifest",
+      [&](void*) { fault_fs->IngestDataCorruptionBeforeWrite(); });
+  ASSERT_OK(Put("key7", "value7"));
+  ASSERT_OK(Put("key8", "value8"));
+  SyncPoint::GetInstance()->EnableProcessing();
+  Status s = Flush();
+  ASSERT_EQ(s.severity(), ROCKSDB_NAMESPACE::Status::Severity::kFatalError);
+  SyncPoint::GetInstance()->DisableProcessing();
+
+  Destroy(options);
+}
+
 class DBFlushTestBlobError : public DBFlushTest,
                              public testing::WithParamInterface<std::string> {
  public:
-  DBFlushTestBlobError() : fault_injection_env_(env_) {}
-  ~DBFlushTestBlobError() { Close(); }
+  DBFlushTestBlobError() : sync_point_(GetParam()) {}
 
-  FaultInjectionTestEnv fault_injection_env_;
+  std::string sync_point_;
 };
 
 INSTANTIATE_TEST_CASE_P(DBFlushTestBlobError, DBFlushTestBlobError,
@@ -542,19 +787,18 @@ TEST_P(DBFlushTestBlobError, FlushError) {
   Options options;
   options.enable_blob_files = true;
   options.disable_auto_compactions = true;
-  options.env = &fault_injection_env_;
+  options.env = env_;
 
   Reopen(options);
 
   ASSERT_OK(Put("key", "blob"));
 
-  SyncPoint::GetInstance()->SetCallBack(GetParam(), [this](void* /* arg */) {
-    fault_injection_env_.SetFilesystemActive(false, Status::IOError());
+  SyncPoint::GetInstance()->SetCallBack(sync_point_, [this](void* arg) {
+    Status* const s = static_cast<Status*>(arg);
+    assert(s);
+
+    (*s) = Status::IOError(sync_point_);
   });
-  SyncPoint::GetInstance()->SetCallBack(
-      "BuildTable:BeforeFinishBuildTable", [this](void* /* arg */) {
-        fault_injection_env_.SetFilesystemActive(true);
-      });
   SyncPoint::GetInstance()->EnableProcessing();
 
   ASSERT_NOK(Flush());
@@ -601,14 +845,117 @@ TEST_P(DBFlushTestBlobError, FlushError) {
 
   const auto& compaction_stats = internal_stats->TEST_GetCompactionStats();
   ASSERT_FALSE(compaction_stats.empty());
-  ASSERT_EQ(compaction_stats[0].bytes_written, 0);
-  ASSERT_EQ(compaction_stats[0].num_output_files, 0);
+
+  if (sync_point_ == "BlobFileBuilder::WriteBlobToFile:AddRecord") {
+    ASSERT_EQ(compaction_stats[0].bytes_written, 0);
+    ASSERT_EQ(compaction_stats[0].bytes_written_blob, 0);
+    ASSERT_EQ(compaction_stats[0].num_output_files, 0);
+    ASSERT_EQ(compaction_stats[0].num_output_files_blob, 0);
+  } else {
+    // SST file writing succeeded; blob file writing failed (during Finish)
+    ASSERT_GT(compaction_stats[0].bytes_written, 0);
+    ASSERT_EQ(compaction_stats[0].bytes_written_blob, 0);
+    ASSERT_EQ(compaction_stats[0].num_output_files, 1);
+    ASSERT_EQ(compaction_stats[0].num_output_files_blob, 0);
+  }
 
   const uint64_t* const cf_stats_value = internal_stats->TEST_GetCFStatsValue();
-  ASSERT_EQ(cf_stats_value[InternalStats::BYTES_FLUSHED], 0);
+  ASSERT_EQ(cf_stats_value[InternalStats::BYTES_FLUSHED],
+            compaction_stats[0].bytes_written +
+                compaction_stats[0].bytes_written_blob);
 #endif  // ROCKSDB_LITE
 }
 
+#ifndef ROCKSDB_LITE
+TEST_P(DBAtomicFlushTest, ManualFlushUnder2PC) {
+  Options options = CurrentOptions();
+  options.create_if_missing = true;
+  options.allow_2pc = true;
+  options.atomic_flush = GetParam();
+  // 64MB so that memtable flush won't be trigger by the small writes.
+  options.write_buffer_size = (static_cast<size_t>(64) << 20);
+
+  // Destroy the DB to recreate as a TransactionDB.
+  Close();
+  Destroy(options, true);
+
+  // Create a TransactionDB.
+  TransactionDB* txn_db = nullptr;
+  TransactionDBOptions txn_db_opts;
+  txn_db_opts.write_policy = TxnDBWritePolicy::WRITE_COMMITTED;
+  ASSERT_OK(TransactionDB::Open(options, txn_db_opts, dbname_, &txn_db));
+  ASSERT_NE(txn_db, nullptr);
+  db_ = txn_db;
+
+  // Create two more columns other than default CF.
+  std::vector<std::string> cfs = {"puppy", "kitty"};
+  CreateColumnFamilies(cfs, options);
+  ASSERT_EQ(handles_.size(), 2);
+  ASSERT_EQ(handles_[0]->GetName(), cfs[0]);
+  ASSERT_EQ(handles_[1]->GetName(), cfs[1]);
+  const size_t kNumCfToFlush = options.atomic_flush ? 2 : 1;
+
+  WriteOptions wopts;
+  TransactionOptions txn_opts;
+  // txn1 only prepare, but does not commit.
+  // The WAL containing the prepared but uncommitted data must be kept.
+  Transaction* txn1 = txn_db->BeginTransaction(wopts, txn_opts, nullptr);
+  // txn2 not only prepare, but also commit.
+  Transaction* txn2 = txn_db->BeginTransaction(wopts, txn_opts, nullptr);
+  ASSERT_NE(txn1, nullptr);
+  ASSERT_NE(txn2, nullptr);
+  for (size_t i = 0; i < kNumCfToFlush; i++) {
+    ASSERT_OK(txn1->Put(handles_[i], "k1", "v1"));
+    ASSERT_OK(txn2->Put(handles_[i], "k2", "v2"));
+  }
+  // A txn must be named before prepare.
+  ASSERT_OK(txn1->SetName("txn1"));
+  ASSERT_OK(txn2->SetName("txn2"));
+  // Prepare writes to WAL, but not to memtable. (WriteCommitted)
+  ASSERT_OK(txn1->Prepare());
+  ASSERT_OK(txn2->Prepare());
+  // Commit writes to memtable.
+  ASSERT_OK(txn2->Commit());
+  delete txn1;
+  delete txn2;
+
+  // There are still data in memtable not flushed.
+  // But since data is small enough to reside in the active memtable,
+  // there are no immutable memtable.
+  for (size_t i = 0; i < kNumCfToFlush; i++) {
+    auto cfh = static_cast<ColumnFamilyHandleImpl*>(handles_[i]);
+    ASSERT_EQ(0, cfh->cfd()->imm()->NumNotFlushed());
+    ASSERT_FALSE(cfh->cfd()->mem()->IsEmpty());
+  }
+
+  // Atomic flush memtables,
+  // the min log with prepared data should be written to MANIFEST.
+  std::vector<ColumnFamilyHandle*> cfs_to_flush(kNumCfToFlush);
+  for (size_t i = 0; i < kNumCfToFlush; i++) {
+    cfs_to_flush[i] = handles_[i];
+  }
+  ASSERT_OK(txn_db->Flush(FlushOptions(), cfs_to_flush));
+
+  // There are no remaining data in memtable after flush.
+  for (size_t i = 0; i < kNumCfToFlush; i++) {
+    auto cfh = static_cast<ColumnFamilyHandleImpl*>(handles_[i]);
+    ASSERT_EQ(0, cfh->cfd()->imm()->NumNotFlushed());
+    ASSERT_TRUE(cfh->cfd()->mem()->IsEmpty());
+    ASSERT_EQ(cfh->cfd()->GetFlushReason(), FlushReason::kManualFlush);
+  }
+
+  // The recovered min log number with prepared data should be non-zero.
+  // In 2pc mode, MinLogNumberToKeep returns the
+  // VersionSet::min_log_number_to_keep_2pc recovered from MANIFEST, if it's 0,
+  // it means atomic flush didn't write the min_log_number_to_keep to MANIFEST.
+  cfs.push_back(kDefaultColumnFamilyName);
+  ASSERT_OK(TryReopenWithColumnFamilies(cfs, options));
+  DBImpl* db_impl = reinterpret_cast<DBImpl*>(db_);
+  ASSERT_TRUE(db_impl->allow_2pc());
+  ASSERT_NE(db_impl->MinLogNumberToKeep(), 0);
+}
+#endif  // ROCKSDB_LITE
+
 TEST_P(DBAtomicFlushTest, ManualAtomicFlush) {
   Options options = CurrentOptions();
   options.create_if_missing = true;
@@ -623,18 +970,84 @@ TEST_P(DBAtomicFlushTest, ManualAtomicFlush) {
   for (size_t i = 0; i != num_cfs; ++i) {
     ASSERT_OK(Put(static_cast<int>(i) /*cf*/, "key", "value", wopts));
   }
+
+  for (size_t i = 0; i != num_cfs; ++i) {
+    auto cfh = static_cast<ColumnFamilyHandleImpl*>(handles_[i]);
+    ASSERT_EQ(0, cfh->cfd()->imm()->NumNotFlushed());
+    ASSERT_FALSE(cfh->cfd()->mem()->IsEmpty());
+  }
+
   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(cfh->cfd()->GetFlushReason(), FlushReason::kManualFlush);
     ASSERT_EQ(0, cfh->cfd()->imm()->NumNotFlushed());
     ASSERT_TRUE(cfh->cfd()->mem()->IsEmpty());
   }
 }
 
+TEST_P(DBAtomicFlushTest, PrecomputeMinLogNumberToKeepNon2PC) {
+  Options options = CurrentOptions();
+  options.create_if_missing = true;
+  options.atomic_flush = GetParam();
+  options.write_buffer_size = (static_cast<size_t>(64) << 20);
+  CreateAndReopenWithCF({"pikachu"}, options);
+
+  const size_t num_cfs = handles_.size();
+  ASSERT_EQ(num_cfs, 2);
+  WriteOptions wopts;
+  for (size_t i = 0; i != num_cfs; ++i) {
+    ASSERT_OK(Put(static_cast<int>(i) /*cf*/, "key", "value", wopts));
+  }
+
+  {
+    // Flush the default CF only.
+    std::vector<int> cf_ids{0};
+    ASSERT_OK(Flush(cf_ids));
+
+    autovector<ColumnFamilyData*> flushed_cfds;
+    autovector<autovector<VersionEdit*>> flush_edits;
+    auto flushed_cfh = static_cast<ColumnFamilyHandleImpl*>(handles_[0]);
+    flushed_cfds.push_back(flushed_cfh->cfd());
+    flush_edits.push_back({});
+    auto unflushed_cfh = static_cast<ColumnFamilyHandleImpl*>(handles_[1]);
+
+    ASSERT_EQ(PrecomputeMinLogNumberToKeepNon2PC(dbfull()->TEST_GetVersionSet(),
+                                                 flushed_cfds, flush_edits),
+              unflushed_cfh->cfd()->GetLogNumber());
+  }
+
+  {
+    // Flush all CFs.
+    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));
+    uint64_t log_num_after_flush = dbfull()->TEST_GetCurrentLogNumber();
+
+    uint64_t min_log_number_to_keep = port::kMaxUint64;
+    autovector<ColumnFamilyData*> flushed_cfds;
+    autovector<autovector<VersionEdit*>> flush_edits;
+    for (size_t i = 0; i != num_cfs; ++i) {
+      auto cfh = static_cast<ColumnFamilyHandleImpl*>(handles_[i]);
+      flushed_cfds.push_back(cfh->cfd());
+      flush_edits.push_back({});
+      min_log_number_to_keep =
+          std::min(min_log_number_to_keep, cfh->cfd()->GetLogNumber());
+    }
+    ASSERT_EQ(min_log_number_to_keep, log_num_after_flush);
+    ASSERT_EQ(PrecomputeMinLogNumberToKeepNon2PC(dbfull()->TEST_GetVersionSet(),
+                                                 flushed_cfds, flush_edits),
+              min_log_number_to_keep);
+  }
+}
+
 TEST_P(DBAtomicFlushTest, AtomicFlushTriggeredByMemTableFull) {
   Options options = CurrentOptions();
   options.create_if_missing = true;

db/db_impl/compacted_db_impl.cc

@@ -4,7 +4,7 @@
 //  (found in the LICENSE.Apache file in the root directory).
 
 #ifndef ROCKSDB_LITE
-#include "db/compacted_db_impl.h"
+#include "db/db_impl/compacted_db_impl.h"
 
 #include "db/db_impl/db_impl.h"
 #include "db/version_set.h"
@@ -17,11 +17,13 @@ extern void MarkKeyMayExist(void* arg);
 extern bool SaveValue(void* arg, const ParsedInternalKey& parsed_key,
                       const Slice& v, bool hit_and_return);
 
-CompactedDBImpl::CompactedDBImpl(
-  const DBOptions& options, const std::string& dbname)
-  : DBImpl(options, dbname), cfd_(nullptr), version_(nullptr),
-    user_comparator_(nullptr) {
-}
+CompactedDBImpl::CompactedDBImpl(const DBOptions& options,
+                                 const std::string& dbname)
+    : DBImpl(options, dbname, /*seq_per_batch*/ false, +/*batch_per_txn*/ true,
+             /*read_only*/ true),
+      cfd_(nullptr),
+      version_(nullptr),
+      user_comparator_(nullptr) {}
 
 CompactedDBImpl::~CompactedDBImpl() {
 }
@@ -78,6 +80,7 @@ std::vector<Status> CompactedDBImpl::MultiGet(const ReadOptions& options,
                              nullptr, nullptr, nullptr, true, nullptr, nullptr);
       LookupKey lkey(keys[idx], kMaxSequenceNumber);
       Status s = r->Get(options, lkey.internal_key(), &get_context, nullptr);
+      assert(static_cast<size_t>(idx) < statuses.size());
       if (!s.ok() && !s.IsNotFound()) {
         statuses[idx] = s;
       } else {

db/db_impl/compacted_db_impl.h

@@ -18,7 +18,7 @@ class CompactedDBImpl : public DBImpl {
   CompactedDBImpl(const CompactedDBImpl&) = delete;
   void operator=(const CompactedDBImpl&) = delete;
 
-  virtual ~CompactedDBImpl();
+  ~CompactedDBImpl() override;
 
   static Status Open(const Options& options, const std::string& dbname,
                      DB** dbptr);
@@ -82,6 +82,11 @@ class CompactedDBImpl : public DBImpl {
                        ColumnFamilyHandle* /*column_family*/) override {
     return Status::NotSupported("Not supported in compacted db mode.");
   }
+
+  virtual Status SyncWAL() override {
+    return Status::NotSupported("Not supported in compacted db mode.");
+  }
+
   using DB::IngestExternalFile;
   virtual Status IngestExternalFile(
       ColumnFamilyHandle* /*column_family*/,

db/db_impl/db_impl.cc

@@ -18,6 +18,7 @@
 #include <cstdio>
 #include <map>
 #include <set>
+#include <sstream>
 #include <stdexcept>
 #include <string>
 #include <unordered_map>
@@ -52,7 +53,6 @@
 #include "db/version_set.h"
 #include "db/write_batch_internal.h"
 #include "db/write_callback.h"
-#include "env/composite_env_wrapper.h"
 #include "file/file_util.h"
 #include "file/filename.h"
 #include "file/random_access_file_reader.h"
@@ -82,6 +82,7 @@
 #include "rocksdb/stats_history.h"
 #include "rocksdb/status.h"
 #include "rocksdb/table.h"
+#include "rocksdb/version.h"
 #include "rocksdb/write_buffer_manager.h"
 #include "table/block_based/block.h"
 #include "table/block_based/block_based_table_factory.h"
@@ -93,7 +94,6 @@
 #include "table/two_level_iterator.h"
 #include "test_util/sync_point.h"
 #include "util/autovector.h"
-#include "util/build_version.h"
 #include "util/cast_util.h"
 #include "util/coding.h"
 #include "util/compression.h"
@@ -146,27 +146,31 @@ void DumpSupportInfo(Logger* logger) {
 }  // namespace
 
 DBImpl::DBImpl(const DBOptions& options, const std::string& dbname,
-               const bool seq_per_batch, const bool batch_per_txn)
+               const bool seq_per_batch, const bool batch_per_txn,
+               bool read_only)
     : dbname_(dbname),
       own_info_log_(options.info_log == nullptr),
-      initial_db_options_(SanitizeOptions(dbname, options)),
+      initial_db_options_(SanitizeOptions(dbname, options, read_only)),
       env_(initial_db_options_.env),
       io_tracer_(std::make_shared<IOTracer>()),
       immutable_db_options_(initial_db_options_),
       fs_(immutable_db_options_.fs, io_tracer_),
       mutable_db_options_(initial_db_options_),
       stats_(immutable_db_options_.statistics.get()),
-      mutex_(stats_, env_, DB_MUTEX_WAIT_MICROS,
+      mutex_(stats_, immutable_db_options_.clock, DB_MUTEX_WAIT_MICROS,
              immutable_db_options_.use_adaptive_mutex),
       default_cf_handle_(nullptr),
+      error_handler_(this, immutable_db_options_, &mutex_),
+      event_logger_(immutable_db_options_.info_log.get()),
       max_total_in_memory_state_(0),
       file_options_(BuildDBOptions(immutable_db_options_, mutable_db_options_)),
       file_options_for_compaction_(fs_->OptimizeForCompactionTableWrite(
           file_options_, immutable_db_options_)),
       seq_per_batch_(seq_per_batch),
       batch_per_txn_(batch_per_txn),
-      db_lock_(nullptr),
+      next_job_id_(1),
       shutting_down_(false),
+      db_lock_(nullptr),
       manual_compaction_paused_(false),
       bg_cv_(&mutex_),
       logfile_number_(0),
@@ -191,9 +195,8 @@ DBImpl::DBImpl(const DBOptions& options, const std::string& dbname,
       bg_purge_scheduled_(0),
       disable_delete_obsolete_files_(0),
       pending_purge_obsolete_files_(0),
-      delete_obsolete_files_last_run_(env_->NowMicros()),
+      delete_obsolete_files_last_run_(immutable_db_options_.clock->NowMicros()),
       last_stats_dump_time_microsec_(0),
-      next_job_id_(1),
       has_unpersisted_data_(false),
       unable_to_release_oldest_log_(false),
       num_running_ingest_file_(0),
@@ -201,7 +204,6 @@ DBImpl::DBImpl(const DBOptions& options, const std::string& dbname,
       wal_manager_(immutable_db_options_, file_options_, io_tracer_,
                    seq_per_batch),
 #endif  // ROCKSDB_LITE
-      event_logger_(immutable_db_options_.info_log.get()),
       bg_work_paused_(0),
       bg_compaction_paused_(0),
       refitting_level_(false),
@@ -230,8 +232,9 @@ DBImpl::DBImpl(const DBOptions& options, const std::string& dbname,
       own_sfm_(options.sst_file_manager == nullptr),
       preserve_deletes_(options.preserve_deletes),
       closed_(false),
-      error_handler_(this, immutable_db_options_, &mutex_),
-      atomic_flush_install_cv_(&mutex_) {
+      atomic_flush_install_cv_(&mutex_),
+      blob_callback_(immutable_db_options_.sst_file_manager.get(), &mutex_,
+                     &error_handler_) {
   // !batch_per_trx_ implies seq_per_batch_ because it is only unset for
   // WriteUnprepared, which should use seq_per_batch_.
   assert(batch_per_txn_ || seq_per_batch_);
@@ -267,6 +270,10 @@ DBImpl::DBImpl(const DBOptions& options, const std::string& dbname,
   // we won't drop any deletion markers until SetPreserveDeletesSequenceNumber()
   // is called by client and this seqnum is advanced.
   preserve_deletes_seqnum_.store(0);
+
+  if (write_buffer_manager_) {
+    wbm_stall_.reset(new WBMStallInterface());
+  }
 }
 
 Status DBImpl::Resume() {
@@ -306,18 +313,21 @@ Status DBImpl::ResumeImpl(DBRecoverContext context) {
   mutex_.AssertHeld();
   WaitForBackgroundWork();
 
-  Status bg_error = error_handler_.GetBGError();
   Status s;
   if (shutdown_initiated_) {
     // Returning shutdown status to SFM during auto recovery will cause it
     // to abort the recovery and allow the shutdown to progress
     s = Status::ShutdownInProgress();
   }
-  if (s.ok() && bg_error.severity() > Status::Severity::kHardError) {
-    ROCKS_LOG_INFO(
-        immutable_db_options_.info_log,
-        "DB resume requested but failed due to Fatal/Unrecoverable error");
-    s = bg_error;
+
+  if (s.ok()) {
+    Status bg_error = error_handler_.GetBGError();
+    if (bg_error.severity() > Status::Severity::kHardError) {
+      ROCKS_LOG_INFO(
+          immutable_db_options_.info_log,
+          "DB resume requested but failed due to Fatal/Unrecoverable error");
+      s = bg_error;
+    }
   }
 
   // Make sure the IO Status stored in version set is set to OK.
@@ -392,6 +402,11 @@ Status DBImpl::ResumeImpl(DBRecoverContext context) {
   FindObsoleteFiles(&job_context, true);
   if (s.ok()) {
     s = error_handler_.ClearBGError();
+  } else {
+    // NOTE: this is needed to pass ASSERT_STATUS_CHECKED
+    // in the DBSSTTest.DBWithMaxSpaceAllowedRandomized test.
+    // See https://github.com/facebook/rocksdb/pull/7715#issuecomment-754947952
+    error_handler_.GetRecoveryError().PermitUncheckedError();
   }
   mutex_.Unlock();
 
@@ -408,6 +423,12 @@ Status DBImpl::ResumeImpl(DBRecoverContext context) {
     if (file_deletion_disabled) {
       // Always return ok
       s = EnableFileDeletions(/*force=*/true);
+      if (!s.ok()) {
+        ROCKS_LOG_INFO(
+            immutable_db_options_.info_log,
+            "DB resume requested but could not enable file deletions [%s]",
+            s.ToString().c_str());
+      }
     }
     ROCKS_LOG_INFO(immutable_db_options_.info_log, "Successfully resumed DB");
   }
@@ -497,19 +518,20 @@ Status DBImpl::CloseHelper() {
   }
   mutex_.Unlock();
 
+  // Below check is added as recovery_error_ is not checked and it causes crash
+  // in DBSSTTest.DBWithMaxSpaceAllowedWithBlobFiles when space limit is
+  // reached.
+  error_handler_.GetRecoveryError().PermitUncheckedError();
+
   // CancelAllBackgroundWork called with false means we just set the shutdown
   // marker. After this we do a variant of the waiting and unschedule work
   // (to consider: moving all the waiting into CancelAllBackgroundWork(true))
   CancelAllBackgroundWork(false);
-  int bottom_compactions_unscheduled =
-      env_->UnSchedule(this, Env::Priority::BOTTOM);
-  int compactions_unscheduled = env_->UnSchedule(this, Env::Priority::LOW);
-  int flushes_unscheduled = env_->UnSchedule(this, Env::Priority::HIGH);
-  Status ret = Status::OK();
   mutex_.Lock();
-  bg_bottom_compaction_scheduled_ -= bottom_compactions_unscheduled;
-  bg_compaction_scheduled_ -= compactions_unscheduled;
-  bg_flush_scheduled_ -= flushes_unscheduled;
+  env_->UnSchedule(this, Env::Priority::BOTTOM);
+  env_->UnSchedule(this, Env::Priority::LOW);
+  env_->UnSchedule(this, Env::Priority::HIGH);
+  Status ret = Status::OK();
 
   // Wait for background work to finish
   while (bg_bottom_compaction_scheduled_ || bg_compaction_scheduled_ ||
@@ -637,11 +659,15 @@ Status DBImpl::CloseHelper() {
 
   if (immutable_db_options_.info_log && own_info_log_) {
     Status s = immutable_db_options_.info_log->Close();
-    if (!s.ok() && ret.ok()) {
+    if (!s.ok() && !s.IsNotSupported() && ret.ok()) {
       ret = s;
     }
   }
 
+  if (write_buffer_manager_ && wbm_stall_) {
+    write_buffer_manager_->RemoveDBFromQueue(wbm_stall_.get());
+  }
+
   if (ret.IsAborted()) {
     // Reserve IsAborted() error for those where users didn't release
     // certain resource and they can release them and come back and
@@ -671,8 +697,8 @@ void DBImpl::MaybeIgnoreError(Status* s) const {
 }
 
 const Status DBImpl::CreateArchivalDirectory() {
-  if (immutable_db_options_.wal_ttl_seconds > 0 ||
-      immutable_db_options_.wal_size_limit_mb > 0) {
+  if (immutable_db_options_.WAL_ttl_seconds > 0 ||
+      immutable_db_options_.WAL_size_limit_MB > 0) {
     std::string archivalPath = ArchivalDirectory(immutable_db_options_.wal_dir);
     return env_->CreateDirIfMissing(archivalPath);
   }
@@ -689,6 +715,18 @@ void DBImpl::PrintStatistics() {
 
 void DBImpl::StartPeriodicWorkScheduler() {
 #ifndef ROCKSDB_LITE
+
+#ifndef NDEBUG
+  // It only used by test to disable scheduler
+  bool disable_scheduler = false;
+  TEST_SYNC_POINT_CALLBACK(
+      "DBImpl::StartPeriodicWorkScheduler:DisableScheduler",
+      &disable_scheduler);
+  if (disable_scheduler) {
+    return;
+  }
+#endif  // !NDEBUG
+
   {
     InstrumentedMutexLock l(&mutex_);
     periodic_work_scheduler_ = PeriodicWorkScheduler::Default();
@@ -726,7 +764,8 @@ void DBImpl::PersistStats() {
     return;
   }
   TEST_SYNC_POINT("DBImpl::PersistStats:StartRunning");
-  uint64_t now_seconds = env_->NowMicros() / kMicrosInSecond;
+  uint64_t now_seconds =
+      immutable_db_options_.clock->NowMicros() / kMicrosInSecond;
 
   Statistics* statistics = immutable_db_options_.statistics.get();
   if (!statistics) {
@@ -976,6 +1015,7 @@ Status DBImpl::SetOptions(
   MutableCFOptions new_options;
   Status s;
   Status persist_options_status;
+  persist_options_status.PermitUncheckedError();  // Allow uninitialized access
   SuperVersionContext sv_context(/* create_superversion */ true);
   {
     auto db_options = GetDBOptions();
@@ -1281,7 +1321,11 @@ Status DBImpl::SyncWAL() {
   TEST_SYNC_POINT("DBImpl::SyncWAL:BeforeMarkLogsSynced:1");
   {
     InstrumentedMutexLock l(&mutex_);
-    MarkLogsSynced(current_log_number, need_log_dir_sync, status);
+    if (status.ok()) {
+      status = MarkLogsSynced(current_log_number, need_log_dir_sync);
+    } else {
+      MarkLogsNotSynced(current_log_number);
+    }
   }
   TEST_SYNC_POINT("DBImpl::SyncWAL:BeforeMarkLogsSynced:2");
 
@@ -1307,27 +1351,54 @@ Status DBImpl::UnlockWAL() {
   return Status::OK();
 }
 
-void DBImpl::MarkLogsSynced(uint64_t up_to, bool synced_dir,
-                            const Status& status) {
+Status DBImpl::MarkLogsSynced(uint64_t up_to, bool synced_dir) {
   mutex_.AssertHeld();
-  if (synced_dir && logfile_number_ == up_to && status.ok()) {
+  if (synced_dir && logfile_number_ == up_to) {
     log_dir_synced_ = true;
   }
+  VersionEdit synced_wals;
   for (auto it = logs_.begin(); it != logs_.end() && it->number <= up_to;) {
-    auto& log = *it;
-    assert(log.getting_synced);
-    if (status.ok() && logs_.size() > 1) {
-      logs_to_free_.push_back(log.ReleaseWriter());
+    auto& wal = *it;
+    assert(wal.getting_synced);
+    if (logs_.size() > 1) {
+      if (immutable_db_options_.track_and_verify_wals_in_manifest &&
+          wal.writer->file()->GetFileSize() > 0) {
+        synced_wals.AddWal(wal.number,
+                           WalMetadata(wal.writer->file()->GetFileSize()));
+      }
+      logs_to_free_.push_back(wal.ReleaseWriter());
       // To modify logs_ both mutex_ and log_write_mutex_ must be held
       InstrumentedMutexLock l(&log_write_mutex_);
       it = logs_.erase(it);
     } else {
-      log.getting_synced = false;
+      wal.getting_synced = false;
       ++it;
     }
   }
-  assert(!status.ok() || logs_.empty() || logs_[0].number > up_to ||
+  assert(logs_.empty() || logs_[0].number > up_to ||
          (logs_.size() == 1 && !logs_[0].getting_synced));
+
+  Status s;
+  if (synced_wals.IsWalAddition()) {
+    // not empty, write to MANIFEST.
+    s = versions_->LogAndApplyToDefaultColumnFamily(&synced_wals, &mutex_);
+    if (!s.ok() && versions_->io_status().IsIOError()) {
+      s = error_handler_.SetBGError(versions_->io_status(),
+                                    BackgroundErrorReason::kManifestWrite);
+    }
+  }
+  log_sync_cv_.SignalAll();
+  return s;
+}
+
+void DBImpl::MarkLogsNotSynced(uint64_t up_to) {
+  mutex_.AssertHeld();
+  for (auto it = logs_.begin(); it != logs_.end() && it->number <= up_to;
+       ++it) {
+    auto& wal = *it;
+    assert(wal.getting_synced);
+    wal.getting_synced = false;
+  }
   log_sync_cv_.SignalAll();
 }
 
@@ -1595,8 +1666,8 @@ Status DBImpl::GetImpl(const ReadOptions& read_options, const Slice& key,
   }
 #endif  // NDEBUG
 
-  PERF_CPU_TIMER_GUARD(get_cpu_nanos, env_);
-  StopWatch sw(env_, stats_, DB_GET);
+  PERF_CPU_TIMER_GUARD(get_cpu_nanos, immutable_db_options_.clock);
+  StopWatch sw(immutable_db_options_.clock, stats_, DB_GET);
   PERF_TIMER_GUARD(get_snapshot_time);
 
   auto cfh = static_cast_with_check<ColumnFamilyHandleImpl>(
@@ -1660,7 +1731,9 @@ Status DBImpl::GetImpl(const ReadOptions& read_options, const Slice& key,
   }
   // If timestamp is used, we use read callback to ensure <key,t,s> is returned
   // only if t <= read_opts.timestamp and s <= snapshot.
-  if (ts_sz > 0 && !get_impl_options.callback) {
+  if (ts_sz > 0) {
+    assert(!get_impl_options
+                .callback);  // timestamp with callback is not supported
     read_cb.Refresh(snapshot);
     get_impl_options.callback = &read_cb;
   }
@@ -1784,8 +1857,8 @@ std::vector<Status> DBImpl::MultiGet(
     const std::vector<ColumnFamilyHandle*>& column_family,
     const std::vector<Slice>& keys, std::vector<std::string>* values,
     std::vector<std::string>* timestamps) {
-  PERF_CPU_TIMER_GUARD(get_cpu_nanos, env_);
-  StopWatch sw(env_, stats_, DB_MULTIGET);
+  PERF_CPU_TIMER_GUARD(get_cpu_nanos, immutable_db_options_.clock);
+  StopWatch sw(immutable_db_options_.clock, stats_, DB_MULTIGET);
   PERF_TIMER_GUARD(get_snapshot_time);
 
 #ifndef NDEBUG
@@ -1914,9 +1987,8 @@ std::vector<Status> DBImpl::MultiGet(
         break;
       }
     }
-
     if (read_options.deadline.count() &&
-        env_->NowMicros() >
+        immutable_db_options_.clock->NowMicros() >
             static_cast<uint64_t>(read_options.deadline.count())) {
       break;
     }
@@ -1925,8 +1997,8 @@ std::vector<Status> DBImpl::MultiGet(
   if (keys_read < num_keys) {
     // The only reason to break out of the loop is when the deadline is
     // exceeded
-    assert(env_->NowMicros() >
-        static_cast<uint64_t>(read_options.deadline.count()));
+    assert(immutable_db_options_.clock->NowMicros() >
+           static_cast<uint64_t>(read_options.deadline.count()));
     for (++keys_read; keys_read < num_keys; ++keys_read) {
       stat_list[keys_read] = Status::TimedOut();
     }
@@ -2172,7 +2244,7 @@ void DBImpl::MultiGet(const ReadOptions& read_options, const size_t num_keys,
   for (; cf_iter != multiget_cf_data.end(); ++cf_iter) {
     s = MultiGetImpl(read_options, cf_iter->start, cf_iter->num_keys,
                      &sorted_keys, cf_iter->super_version, consistent_seqnum,
-                     read_callback, nullptr);
+                     read_callback);
     if (!s.ok()) {
       break;
     }
@@ -2336,15 +2408,16 @@ void DBImpl::MultiGetWithCallback(
   }
 
   GetWithTimestampReadCallback timestamp_read_callback(0);
-  ReadCallback* read_callback = nullptr;
+  ReadCallback* read_callback = callback;
   if (read_options.timestamp && read_options.timestamp->size() > 0) {
+    assert(!read_callback);  // timestamp with callback is not supported
     timestamp_read_callback.Refresh(consistent_seqnum);
     read_callback = &timestamp_read_callback;
   }
 
   Status s = MultiGetImpl(read_options, 0, num_keys, sorted_keys,
                           multiget_cf_data[0].super_version, consistent_seqnum,
-                          read_callback, nullptr);
+                          read_callback);
   assert(s.ok() || s.IsTimedOut() || s.IsAborted());
   ReturnAndCleanupSuperVersion(multiget_cf_data[0].cfd,
                                multiget_cf_data[0].super_version);
@@ -2363,9 +2436,9 @@ Status DBImpl::MultiGetImpl(
     const ReadOptions& read_options, size_t start_key, size_t num_keys,
     autovector<KeyContext*, MultiGetContext::MAX_BATCH_SIZE>* sorted_keys,
     SuperVersion* super_version, SequenceNumber snapshot,
-    ReadCallback* callback, bool* is_blob_index) {
-  PERF_CPU_TIMER_GUARD(get_cpu_nanos, env_);
-  StopWatch sw(env_, stats_, DB_MULTIGET);
+    ReadCallback* callback) {
+  PERF_CPU_TIMER_GUARD(get_cpu_nanos, immutable_db_options_.clock);
+  StopWatch sw(immutable_db_options_.clock, stats_, DB_MULTIGET);
 
   // For each of the given keys, apply the entire "get" process as follows:
   // First look in the memtable, then in the immutable memtable (if any).
@@ -2376,7 +2449,7 @@ Status DBImpl::MultiGetImpl(
   uint64_t curr_value_size = 0;
   while (keys_left) {
     if (read_options.deadline.count() &&
-        env_->NowMicros() >
+        immutable_db_options_.clock->NowMicros() >
             static_cast<uint64_t>(read_options.deadline.count())) {
       s = Status::TimedOut();
       break;
@@ -2402,11 +2475,9 @@ Status DBImpl::MultiGetImpl(
         (read_options.read_tier == kPersistedTier &&
          has_unpersisted_data_.load(std::memory_order_relaxed));
     if (!skip_memtable) {
-      super_version->mem->MultiGet(read_options, &range, callback,
-                                   is_blob_index);
+      super_version->mem->MultiGet(read_options, &range, callback);
       if (!range.empty()) {
-        super_version->imm->MultiGet(read_options, &range, callback,
-                                     is_blob_index);
+        super_version->imm->MultiGet(read_options, &range, callback);
       }
       if (!range.empty()) {
         lookup_current = true;
@@ -2416,8 +2487,7 @@ Status DBImpl::MultiGetImpl(
     }
     if (lookup_current) {
       PERF_TIMER_GUARD(get_from_output_files_time);
-      super_version->current->MultiGet(read_options, &range, callback,
-                                       is_blob_index);
+      super_version->current->MultiGet(read_options, &range, callback);
     }
     curr_value_size = range.GetValueSize();
     if (curr_value_size > read_options.value_size_soft_limit) {
@@ -2771,7 +2841,7 @@ Iterator* DBImpl::NewIterator(const ReadOptions& read_options,
                                     /* allow_unprepared_value */ true);
     result = NewDBIterator(
         env_, read_options, *cfd->ioptions(), sv->mutable_cf_options,
-        cfd->user_comparator(), iter, kMaxSequenceNumber,
+        cfd->user_comparator(), iter, sv->current, kMaxSequenceNumber,
         sv->mutable_cf_options.max_sequential_skip_in_iterations, read_callback,
         this, cfd);
 #endif
@@ -2792,7 +2862,7 @@ ArenaWrappedDBIter* DBImpl::NewIteratorImpl(const ReadOptions& read_options,
                                             ColumnFamilyData* cfd,
                                             SequenceNumber snapshot,
                                             ReadCallback* read_callback,
-                                            bool allow_blob,
+                                            bool expose_blob_index,
                                             bool allow_refresh) {
   SuperVersion* sv = cfd->GetReferencedSuperVersion(this);
 
@@ -2857,9 +2927,9 @@ ArenaWrappedDBIter* DBImpl::NewIteratorImpl(const ReadOptions& read_options,
   // likely that any iterator pointer is close to the iterator it points to so
   // that they are likely to be in the same cache line and/or page.
   ArenaWrappedDBIter* db_iter = NewArenaWrappedDbIterator(
-      env_, read_options, *cfd->ioptions(), sv->mutable_cf_options, snapshot,
-      sv->mutable_cf_options.max_sequential_skip_in_iterations,
-      sv->version_number, read_callback, this, cfd, allow_blob,
+      env_, read_options, *cfd->ioptions(), sv->mutable_cf_options, sv->current,
+      snapshot, sv->mutable_cf_options.max_sequential_skip_in_iterations,
+      sv->version_number, read_callback, this, cfd, expose_blob_index,
       read_options.snapshot != nullptr ? false : allow_refresh);
 
   InternalIterator* internal_iter = NewInternalIterator(
@@ -2897,7 +2967,7 @@ Status DBImpl::NewIterators(
                                       /* allow_unprepared_value */ true);
       iterators->push_back(NewDBIterator(
           env_, read_options, *cfd->ioptions(), sv->mutable_cf_options,
-          cfd->user_comparator(), iter, kMaxSequenceNumber,
+          cfd->user_comparator(), iter, sv->current, kMaxSequenceNumber,
           sv->mutable_cf_options.max_sequential_skip_in_iterations,
           read_callback, this, cfd));
     }
@@ -2932,7 +3002,8 @@ const Snapshot* DBImpl::GetSnapshotForWriteConflictBoundary() {
 SnapshotImpl* DBImpl::GetSnapshotImpl(bool is_write_conflict_boundary,
                                       bool lock) {
   int64_t unix_time = 0;
-  env_->GetCurrentTime(&unix_time).PermitUncheckedError();  // Ignore error
+  immutable_db_options_.clock->GetCurrentTime(&unix_time)
+      .PermitUncheckedError();  // Ignore error
   SnapshotImpl* s = new SnapshotImpl;
 
   if (lock) {
@@ -3069,20 +3140,25 @@ const std::string& DBImpl::GetName() const { return dbname_; }
 Env* DBImpl::GetEnv() const { return env_; }
 
 FileSystem* DB::GetFileSystem() const {
-  static LegacyFileSystemWrapper fs_wrap(GetEnv());
-  return &fs_wrap;
+  const auto& fs = GetEnv()->GetFileSystem();
+  return fs.get();
 }
 
 FileSystem* DBImpl::GetFileSystem() const {
   return immutable_db_options_.fs.get();
 }
 
+SystemClock* DBImpl::GetSystemClock() const {
+  return immutable_db_options_.clock;
+}
+
 #ifndef ROCKSDB_LITE
 
-Status DBImpl::StartIOTrace(Env* env, const TraceOptions& trace_options,
+Status DBImpl::StartIOTrace(const TraceOptions& trace_options,
                             std::unique_ptr<TraceWriter>&& trace_writer) {
   assert(trace_writer != nullptr);
-  return io_tracer_->StartIOTrace(env, trace_options, std::move(trace_writer));
+  return io_tracer_->StartIOTrace(GetSystemClock(), trace_options,
+                                  std::move(trace_writer));
 }
 
 Status DBImpl::EndIOTrace() {
@@ -3368,6 +3444,10 @@ Status DBImpl::GetApproximateSizes(const SizeApproximationOptions& options,
     return Status::InvalidArgument("Invalid options");
   }
 
+  const Comparator* const ucmp = column_family->GetComparator();
+  assert(ucmp);
+  size_t ts_sz = ucmp->timestamp_size();
+
   Version* v;
   auto cfh = static_cast_with_check<ColumnFamilyHandleImpl>(column_family);
   auto cfd = cfh->cfd();
@@ -3375,9 +3455,23 @@ Status DBImpl::GetApproximateSizes(const SizeApproximationOptions& options,
   v = sv->current;
 
   for (int i = 0; i < n; i++) {
+    Slice start = range[i].start;
+    Slice limit = range[i].limit;
+
+    // Add timestamp if needed
+    std::string start_with_ts, limit_with_ts;
+    if (ts_sz > 0) {
+      // Maximum timestamp means including all key with any timestamp
+      AppendKeyWithMaxTimestamp(&start_with_ts, start, ts_sz);
+      // Append a maximum timestamp as the range limit is exclusive:
+      // [start, limit)
+      AppendKeyWithMaxTimestamp(&limit_with_ts, limit, ts_sz);
+      start = start_with_ts;
+      limit = limit_with_ts;
+    }
     // Convert user_key into a corresponding internal key.
-    InternalKey k1(range[i].start, kMaxSequenceNumber, kValueTypeForSeek);
-    InternalKey k2(range[i].limit, kMaxSequenceNumber, kValueTypeForSeek);
+    InternalKey k1(start, kMaxSequenceNumber, kValueTypeForSeek);
+    InternalKey k2(limit, kMaxSequenceNumber, kValueTypeForSeek);
     sizes[i] = 0;
     if (options.include_files) {
       sizes[i] += versions_->ApproximateSize(
@@ -3429,14 +3523,13 @@ Status DBImpl::DeleteFile(std::string name) {
   FileType type;
   WalFileType log_type;
   if (!ParseFileName(name, &number, &type, &log_type) ||
-      (type != kTableFile && type != kLogFile)) {
+      (type != kTableFile && type != kWalFile)) {
     ROCKS_LOG_ERROR(immutable_db_options_.info_log, "DeleteFile %s failed.\n",
                     name.c_str());
     return Status::InvalidArgument("Invalid file name");
   }
 
-  Status status;
-  if (type == kLogFile) {
+  if (type == kWalFile) {
     // Only allow deleting archived log files
     if (log_type != kArchivedLogFile) {
       ROCKS_LOG_ERROR(immutable_db_options_.info_log,
@@ -3444,7 +3537,7 @@ Status DBImpl::DeleteFile(std::string name) {
                       name.c_str());
       return Status::NotSupported("Delete only supported for archived logs");
     }
-    status = wal_manager_.DeleteFile(name, number);
+    Status status = wal_manager_.DeleteFile(name, number);
     if (!status.ok()) {
       ROCKS_LOG_ERROR(immutable_db_options_.info_log,
                       "DeleteFile %s failed -- %s.\n", name.c_str(),
@@ -3453,6 +3546,7 @@ Status DBImpl::DeleteFile(std::string name) {
     return status;
   }
 
+  Status status;
   int level;
   FileMetaData* metadata;
   ColumnFamilyData* cfd;
@@ -3526,7 +3620,7 @@ Status DBImpl::DeleteFile(std::string name) {
 Status DBImpl::DeleteFilesInRanges(ColumnFamilyHandle* column_family,
                                    const RangePtr* ranges, size_t n,
                                    bool include_end) {
-  Status status;
+  Status status = Status::OK();
   auto cfh = static_cast_with_check<ColumnFamilyHandleImpl>(column_family);
   ColumnFamilyData* cfd = cfh->cfd();
   VersionEdit edit;
@@ -3585,7 +3679,7 @@ Status DBImpl::DeleteFilesInRanges(ColumnFamilyHandle* column_family,
     }
     if (edit.GetDeletedFiles().empty()) {
       job_context.Clean();
-      return Status::OK();
+      return status;
     }
     input_version->Ref();
     status = versions_->LogAndApply(cfd, *cfd->GetLatestMutableCFOptions(),
@@ -3861,7 +3955,8 @@ Status DestroyDB(const std::string& dbname, const Options& options,
         std::string path_to_delete = dbname + "/" + fname;
         if (type == kMetaDatabase) {
           del = DestroyDB(path_to_delete, options);
-        } else if (type == kTableFile || type == kLogFile) {
+        } else if (type == kTableFile || type == kWalFile ||
+                   type == kBlobFile) {
           del = DeleteDBFile(&soptions, path_to_delete, dbname,
                              /*force_bg=*/false, /*force_fg=*/!wal_in_db_path);
         } else {
@@ -3886,9 +3981,10 @@ Status DestroyDB(const std::string& dbname, const Options& options,
       if (env->GetChildren(path, &filenames).ok()) {
         for (const auto& fname : filenames) {
           if (ParseFileName(fname, &number, &type) &&
-              type == kTableFile) {  // Lock file will be deleted at end
-            std::string table_path = path + "/" + fname;
-            Status del = DeleteDBFile(&soptions, table_path, dbname,
+              (type == kTableFile ||
+               type == kBlobFile)) {  // Lock file will be deleted at end
+            std::string file_path = path + "/" + fname;
+            Status del = DeleteDBFile(&soptions, file_path, dbname,
                                       /*force_bg=*/false, /*force_fg=*/false);
             if (!del.ok() && result.ok()) {
               result = del;
@@ -3915,7 +4011,7 @@ Status DestroyDB(const std::string& dbname, const Options& options,
     if (env->GetChildren(archivedir, &archiveFiles).ok()) {
       // Delete archival files.
       for (const auto& file : archiveFiles) {
-        if (ParseFileName(file, &number, &type) && type == kLogFile) {
+        if (ParseFileName(file, &number, &type) && type == kWalFile) {
           Status del =
               DeleteDBFile(&soptions, archivedir + "/" + file, archivedir,
                            /*force_bg=*/false, /*force_fg=*/!wal_in_db_path);
@@ -3931,7 +4027,7 @@ Status DestroyDB(const std::string& dbname, const Options& options,
     // Delete log files in the WAL dir
     if (wal_dir_exists) {
       for (const auto& file : walDirFiles) {
-        if (ParseFileName(file, &number, &type) && type == kLogFile) {
+        if (ParseFileName(file, &number, &type) && type == kWalFile) {
           Status del =
               DeleteDBFile(&soptions, LogFileName(soptions.wal_dir, number),
                            soptions.wal_dir, /*force_bg=*/false,
@@ -4132,16 +4228,17 @@ void DBImpl::EraseThreadStatusDbInfo() const {}
 //
 // A global method that can dump out the build version
 void DumpRocksDBBuildVersion(Logger* log) {
-#if !defined(IOS_CROSS_COMPILE)
-  // if we compile with Xcode, we don't run build_detect_version, so we don't
-  // generate util/build_version.cc
-  ROCKS_LOG_HEADER(log, "RocksDB version: %d.%d.%d\n", ROCKSDB_MAJOR,
-                   ROCKSDB_MINOR, ROCKSDB_PATCH);
-  ROCKS_LOG_HEADER(log, "Git sha %s", rocksdb_build_git_sha);
-  ROCKS_LOG_HEADER(log, "Compile date %s", rocksdb_build_compile_date);
-#else
-  (void)log;  // ignore "-Wunused-parameter"
-#endif
+  ROCKS_LOG_HEADER(log, "RocksDB version: %s\n",
+                   GetRocksVersionAsString().c_str());
+  const auto& props = GetRocksBuildProperties();
+  const auto& sha = props.find("rocksdb_build_git_sha");
+  if (sha != props.end()) {
+    ROCKS_LOG_HEADER(log, "Git sha %s", sha->second.c_str());
+  }
+  const auto date = props.find("rocksdb_build_date");
+  if (date != props.end()) {
+    ROCKS_LOG_HEADER(log, "Compile date %s", date->second.c_str());
+  }
 }
 
 #ifndef ROCKSDB_LITE
@@ -4306,7 +4403,7 @@ Status DBImpl::IngestExternalFiles(
     }
   }
   // Ingest multiple external SST files atomically.
-  size_t num_cfs = args.size();
+  const size_t num_cfs = args.size();
   for (size_t i = 0; i != num_cfs; ++i) {
     if (args[i].external_files.empty()) {
       char err_msg[128] = {0};
@@ -4343,14 +4440,11 @@ Status DBImpl::IngestExternalFiles(
   std::vector<ExternalSstFileIngestionJob> ingestion_jobs;
   for (const auto& arg : args) {
     auto* cfd = static_cast<ColumnFamilyHandleImpl*>(arg.column_family)->cfd();
-    ingestion_jobs.emplace_back(
-        env_, versions_.get(), cfd, immutable_db_options_, file_options_,
-        &snapshots_, arg.options, &directories_, &event_logger_, io_tracer_);
-  }
-  std::vector<std::pair<bool, Status>> exec_results;
-  for (size_t i = 0; i != num_cfs; ++i) {
-    exec_results.emplace_back(false, Status::OK());
+    ingestion_jobs.emplace_back(versions_.get(), cfd, immutable_db_options_,
+                                file_options_, &snapshots_, arg.options,
+                                &directories_, &event_logger_, io_tracer_);
   }
+
   // TODO(yanqin) maybe make jobs run in parallel
   uint64_t start_file_number = next_file_number;
   for (size_t i = 1; i != num_cfs; ++i) {
@@ -4358,10 +4452,13 @@ Status DBImpl::IngestExternalFiles(
     auto* cfd =
         static_cast<ColumnFamilyHandleImpl*>(args[i].column_family)->cfd();
     SuperVersion* super_version = cfd->GetReferencedSuperVersion(this);
-    exec_results[i].second = ingestion_jobs[i].Prepare(
+    Status es = ingestion_jobs[i].Prepare(
         args[i].external_files, args[i].files_checksums,
         args[i].files_checksum_func_names, start_file_number, super_version);
-    exec_results[i].first = true;
+    // capture first error only
+    if (!es.ok() && status.ok()) {
+      status = es;
+    }
     CleanupSuperVersion(super_version);
   }
   TEST_SYNC_POINT("DBImpl::IngestExternalFiles:BeforeLastJobPrepare:0");
@@ -4370,23 +4467,17 @@ Status DBImpl::IngestExternalFiles(
     auto* cfd =
         static_cast<ColumnFamilyHandleImpl*>(args[0].column_family)->cfd();
     SuperVersion* super_version = cfd->GetReferencedSuperVersion(this);
-    exec_results[0].second = ingestion_jobs[0].Prepare(
+    Status es = ingestion_jobs[0].Prepare(
         args[0].external_files, args[0].files_checksums,
         args[0].files_checksum_func_names, next_file_number, super_version);
-    exec_results[0].first = true;
-    CleanupSuperVersion(super_version);
-  }
-  for (const auto& exec_result : exec_results) {
-    if (!exec_result.second.ok()) {
-      status = exec_result.second;
-      break;
+    if (!es.ok()) {
+      status = es;
     }
+    CleanupSuperVersion(super_version);
   }
   if (!status.ok()) {
     for (size_t i = 0; i != num_cfs; ++i) {
-      if (exec_results[i].first) {
-        ingestion_jobs[i].Cleanup(status);
-      }
+      ingestion_jobs[i].Cleanup(status);
     }
     InstrumentedMutexLock l(&mutex_);
     ReleaseFileNumberFromPendingOutputs(pending_output_elem);
@@ -4556,8 +4647,7 @@ Status DBImpl::IngestExternalFiles(
       // TODO: distinguish between MANIFEST write and CURRENT renaming
       const IOStatus& io_s = versions_->io_status();
       // Should handle return error?
-      error_handler_.SetBGError(io_s, BackgroundErrorReason::kManifestWrite)
-          .PermitUncheckedError();
+      error_handler_.SetBGError(io_s, BackgroundErrorReason::kManifestWrite);
     }
 
     // Resume writes to the DB
@@ -4619,9 +4709,9 @@ Status DBImpl::CreateColumnFamilyWithImport(
   // Import sst files from metadata.
   auto cfh = static_cast_with_check<ColumnFamilyHandleImpl>(*handle);
   auto cfd = cfh->cfd();
-  ImportColumnFamilyJob import_job(env_, versions_.get(), cfd,
-                                   immutable_db_options_, file_options_,
-                                   import_options, metadata.files, io_tracer_);
+  ImportColumnFamilyJob import_job(versions_.get(), cfd, immutable_db_options_,
+                                   file_options_, import_options,
+                                   metadata.files, io_tracer_);
 
   SuperVersionContext dummy_sv_ctx(/* create_superversion */ true);
   VersionEdit dummy_edit;
@@ -4719,14 +4809,36 @@ Status DBImpl::CreateColumnFamilyWithImport(
                       temp_s.ToString().c_str());
     }
     // Always returns Status::OK()
-    assert(DestroyColumnFamilyHandle(*handle).ok());
+    temp_s = DestroyColumnFamilyHandle(*handle);
+    assert(temp_s.ok());
     *handle = nullptr;
   }
   return status;
 }
 
+Status DBImpl::VerifyFileChecksums(const ReadOptions& read_options) {
+  return VerifyChecksumInternal(read_options, /*use_file_checksum=*/true);
+}
+
 Status DBImpl::VerifyChecksum(const ReadOptions& read_options) {
+  return VerifyChecksumInternal(read_options, /*use_file_checksum=*/false);
+}
+
+Status DBImpl::VerifyChecksumInternal(const ReadOptions& read_options,
+                                      bool use_file_checksum) {
   Status s;
+
+  if (use_file_checksum) {
+    FileChecksumGenFactory* const file_checksum_gen_factory =
+        immutable_db_options_.file_checksum_gen_factory.get();
+    if (!file_checksum_gen_factory) {
+      s = Status::InvalidArgument(
+          "Cannot verify file checksum if options.file_checksum_gen_factory is "
+          "null");
+      return s;
+    }
+  }
+
   std::vector<ColumnFamilyData*> cfd_list;
   {
     InstrumentedMutexLock l(&mutex_);
@@ -4741,11 +4853,12 @@ Status DBImpl::VerifyChecksum(const ReadOptions& read_options) {
   for (auto cfd : cfd_list) {
     sv_list.push_back(cfd->GetReferencedSuperVersion(this));
   }
+
   for (auto& sv : sv_list) {
     VersionStorageInfo* vstorage = sv->current->storage_info();
     ColumnFamilyData* cfd = sv->current->cfd();
     Options opts;
-    {
+    if (!use_file_checksum) {
       InstrumentedMutexLock l(&mutex_);
       opts = Options(BuildDBOptions(immutable_db_options_, mutable_db_options_),
                      cfd->GetLatestCFOptions());
@@ -4753,17 +4866,44 @@ Status DBImpl::VerifyChecksum(const ReadOptions& read_options) {
     for (int i = 0; i < vstorage->num_non_empty_levels() && s.ok(); i++) {
       for (size_t j = 0; j < vstorage->LevelFilesBrief(i).num_files && s.ok();
            j++) {
-        const auto& fd = vstorage->LevelFilesBrief(i).files[j].fd;
+        const auto& fd_with_krange = vstorage->LevelFilesBrief(i).files[j];
+        const auto& fd = fd_with_krange.fd;
+        const FileMetaData* fmeta = fd_with_krange.file_metadata;
+        assert(fmeta);
         std::string fname = TableFileName(cfd->ioptions()->cf_paths,
                                           fd.GetNumber(), fd.GetPathId());
-        s = ROCKSDB_NAMESPACE::VerifySstFileChecksum(opts, file_options_,
-                                                     read_options, fname);
+        if (use_file_checksum) {
+          s = VerifyFullFileChecksum(fmeta->file_checksum,
+                                     fmeta->file_checksum_func_name, fname,
+                                     read_options);
+        } else {
+          s = ROCKSDB_NAMESPACE::VerifySstFileChecksum(opts, file_options_,
+                                                       read_options, fname);
+        }
+      }
+    }
+
+    if (s.ok() && use_file_checksum) {
+      const auto& blob_files = vstorage->GetBlobFiles();
+      for (const auto& pair : blob_files) {
+        const uint64_t blob_file_number = pair.first;
+        const auto& meta = pair.second;
+        assert(meta);
+        const std::string blob_file_name = BlobFileName(
+            cfd->ioptions()->cf_paths.front().path, blob_file_number);
+        s = VerifyFullFileChecksum(meta->GetChecksumValue(),
+                                   meta->GetChecksumMethod(), blob_file_name,
+                                   read_options);
+        if (!s.ok()) {
+          break;
+        }
       }
     }
     if (!s.ok()) {
       break;
     }
   }
+
   bool defer_purge =
           immutable_db_options().avoid_unnecessary_blocking_io;
   {
@@ -4788,6 +4928,36 @@ Status DBImpl::VerifyChecksum(const ReadOptions& read_options) {
   return s;
 }
 
+Status DBImpl::VerifyFullFileChecksum(const std::string& file_checksum_expected,
+                                      const std::string& func_name_expected,
+                                      const std::string& fname,
+                                      const ReadOptions& read_options) {
+  Status s;
+  if (file_checksum_expected == kUnknownFileChecksum) {
+    return s;
+  }
+  std::string file_checksum;
+  std::string func_name;
+  s = ROCKSDB_NAMESPACE::GenerateOneFileChecksum(
+      fs_.get(), fname, immutable_db_options_.file_checksum_gen_factory.get(),
+      func_name_expected, &file_checksum, &func_name,
+      read_options.readahead_size, immutable_db_options_.allow_mmap_reads,
+      io_tracer_, immutable_db_options_.rate_limiter.get());
+  if (s.ok()) {
+    assert(func_name_expected == func_name);
+    if (file_checksum != file_checksum_expected) {
+      std::ostringstream oss;
+      oss << fname << " file checksum mismatch, ";
+      oss << "expecting "
+          << Slice(file_checksum_expected).ToString(/*hex=*/true);
+      oss << ", but actual " << Slice(file_checksum).ToString(/*hex=*/true);
+      s = Status::Corruption(oss.str());
+      TEST_SYNC_POINT_CALLBACK("DBImpl::VerifyFullFileChecksum:mismatch", &s);
+    }
+  }
+  return s;
+}
+
 void DBImpl::NotifyOnExternalFileIngested(
     ColumnFamilyData* cfd, const ExternalSstFileIngestionJob& ingestion_job) {
   if (immutable_db_options_.listeners.empty()) {
@@ -4817,7 +4987,8 @@ void DBImpl::WaitForIngestFile() {
 Status DBImpl::StartTrace(const TraceOptions& trace_options,
                           std::unique_ptr<TraceWriter>&& trace_writer) {
   InstrumentedMutexLock lock(&trace_mutex_);
-  tracer_.reset(new Tracer(env_, trace_options, std::move(trace_writer)));
+  tracer_.reset(new Tracer(immutable_db_options_.clock, trace_options,
+                           std::move(trace_writer)));
   return Status::OK();
 }
 
@@ -4836,8 +5007,8 @@ Status DBImpl::EndTrace() {
 Status DBImpl::StartBlockCacheTrace(
     const TraceOptions& trace_options,
     std::unique_ptr<TraceWriter>&& trace_writer) {
-  return block_cache_tracer_.StartTrace(env_, trace_options,
-                                        std::move(trace_writer));
+  return block_cache_tracer_.StartTrace(immutable_db_options_.clock,
+                                        trace_options, std::move(trace_writer));
 }
 
 Status DBImpl::EndBlockCacheTrace() {
@@ -4845,24 +5016,27 @@ Status DBImpl::EndBlockCacheTrace() {
   return Status::OK();
 }
 
-Status DBImpl::TraceIteratorSeek(const uint32_t& cf_id, const Slice& key) {
+Status DBImpl::TraceIteratorSeek(const uint32_t& cf_id, const Slice& key,
+                                 const Slice& lower_bound,
+                                 const Slice upper_bound) {
   Status s;
   if (tracer_) {
     InstrumentedMutexLock lock(&trace_mutex_);
     if (tracer_) {
-      s = tracer_->IteratorSeek(cf_id, key);
+      s = tracer_->IteratorSeek(cf_id, key, lower_bound, upper_bound);
     }
   }
   return s;
 }
 
-Status DBImpl::TraceIteratorSeekForPrev(const uint32_t& cf_id,
-                                        const Slice& key) {
+Status DBImpl::TraceIteratorSeekForPrev(const uint32_t& cf_id, const Slice& key,
+                                        const Slice& lower_bound,
+                                        const Slice upper_bound) {
   Status s;
   if (tracer_) {
     InstrumentedMutexLock lock(&trace_mutex_);
     if (tracer_) {
-      s = tracer_->IteratorSeekForPrev(cf_id, key);
+      s = tracer_->IteratorSeekForPrev(cf_id, key, lower_bound, upper_bound);
     }
   }
   return s;

db/db_impl/db_impl.h

@@ -54,9 +54,6 @@
 #include "rocksdb/transaction_log.h"
 #include "rocksdb/write_buffer_manager.h"
 #include "table/scoped_arena_iterator.h"
-#include "trace_replay/block_cache_tracer.h"
-#include "trace_replay/io_tracer.h"
-#include "trace_replay/trace_replay.h"
 #include "util/autovector.h"
 #include "util/hash.h"
 #include "util/repeatable_thread.h"
@@ -132,7 +129,8 @@ class Directories {
 class DBImpl : public DB {
  public:
   DBImpl(const DBOptions& options, const std::string& dbname,
-         const bool seq_per_batch = false, const bool batch_per_txn = true);
+         const bool seq_per_batch = false, const bool batch_per_txn = true,
+         bool read_only = false);
   // No copying allowed
   DBImpl(const DBImpl&) = delete;
   void operator=(const DBImpl&) = delete;
@@ -431,8 +429,29 @@ class DBImpl : public DB {
       const ExportImportFilesMetaData& metadata,
       ColumnFamilyHandle** handle) override;
 
+  using DB::VerifyFileChecksums;
+  Status VerifyFileChecksums(const ReadOptions& read_options) override;
+
   using DB::VerifyChecksum;
   virtual Status VerifyChecksum(const ReadOptions& /*read_options*/) override;
+  // Verify the checksums of files in db. Currently only tables are checked.
+  //
+  // read_options: controls file I/O behavior, e.g. read ahead size while
+  //               reading all the live table files.
+  //
+  // use_file_checksum: if false, verify the block checksums of all live table
+  //                    in db. Otherwise, obtain the file checksums and compare
+  //                    with the MANIFEST. Currently, file checksums are
+  //                    recomputed by reading all table files.
+  //
+  // Returns: OK if there is no file whose file or block checksum mismatches.
+  Status VerifyChecksumInternal(const ReadOptions& read_options,
+                                bool use_file_checksum);
+
+  Status VerifyFullFileChecksum(const std::string& file_checksum_expected,
+                                const std::string& func_name_expected,
+                                const std::string& fpath,
+                                const ReadOptions& read_options);
 
   using DB::StartTrace;
   virtual Status StartTrace(
@@ -451,7 +470,7 @@ class DBImpl : public DB {
   Status EndBlockCacheTrace() override;
 
   using DB::StartIOTrace;
-  Status StartIOTrace(Env* env, const TraceOptions& options,
+  Status StartIOTrace(const TraceOptions& options,
                       std::unique_ptr<TraceWriter>&& trace_writer) override;
 
   using DB::EndIOTrace;
@@ -468,6 +487,7 @@ class DBImpl : public DB {
 #endif  // ROCKSDB_LITE
 
   // ---- End of implementations of the DB interface ----
+  SystemClock* GetSystemClock() const;
 
   struct GetImplOptions {
     ColumnFamilyHandle* column_family = nullptr;
@@ -502,7 +522,7 @@ class DBImpl : public DB {
                                       ColumnFamilyData* cfd,
                                       SequenceNumber snapshot,
                                       ReadCallback* read_callback,
-                                      bool allow_blob = false,
+                                      bool expose_blob_index = false,
                                       bool allow_refresh = true);
 
   virtual SequenceNumber GetLastPublishedSequence() const {
@@ -575,8 +595,11 @@ class DBImpl : public DB {
                                  bool* found_record_for_key,
                                  bool* is_blob_index = nullptr);
 
-  Status TraceIteratorSeek(const uint32_t& cf_id, const Slice& key);
-  Status TraceIteratorSeekForPrev(const uint32_t& cf_id, const Slice& key);
+  Status TraceIteratorSeek(const uint32_t& cf_id, const Slice& key,
+                           const Slice& lower_bound, const Slice upper_bound);
+  Status TraceIteratorSeekForPrev(const uint32_t& cf_id, const Slice& key,
+                                  const Slice& lower_bound,
+                                  const Slice upper_bound);
 #endif  // ROCKSDB_LITE
 
   // Similar to GetSnapshot(), but also lets the db know that this snapshot
@@ -902,7 +925,7 @@ class DBImpl : public DB {
                            ColumnFamilyHandle* column_family = nullptr,
                            bool disallow_trivial_move = false);
 
-  void TEST_SwitchWAL();
+  Status TEST_SwitchWAL();
 
   bool TEST_UnableToReleaseOldestLog() { return unable_to_release_oldest_log_; }
 
@@ -934,6 +957,9 @@ class DBImpl : public DB {
   // is only for the special test of CancelledCompactions
   Status TEST_WaitForCompact(bool waitUnscheduled = false);
 
+  // Get the background error status
+  Status TEST_GetBGError();
+
   // Return the maximum overlapping data (in bytes) at next level for any
   // file at a level >= 1.
   int64_t TEST_MaxNextLevelOverlappingBytes(
@@ -997,6 +1023,12 @@ class DBImpl : public DB {
 
   VersionSet* TEST_GetVersionSet() const { return versions_.get(); }
 
+  uint64_t TEST_GetCurrentLogNumber() const {
+    InstrumentedMutexLock l(mutex());
+    assert(!logs_.empty());
+    return logs_.back().number;
+  }
+
   const std::unordered_set<uint64_t>& TEST_GetFilesGrabbedForPurge() const {
     return files_grabbed_for_purge_;
   }
@@ -1016,6 +1048,56 @@ class DBImpl : public DB {
   // flush LOG out of application buffer
   void FlushInfoLog();
 
+  // Interface to block and signal the DB in case of stalling writes by
+  // WriteBufferManager. Each DBImpl object contains ptr to WBMStallInterface.
+  // When DB needs to be blocked or signalled by WriteBufferManager,
+  // state_ is changed accordingly.
+  class WBMStallInterface : public StallInterface {
+   public:
+    enum State {
+      BLOCKED = 0,
+      RUNNING,
+    };
+
+    WBMStallInterface() : state_cv_(&state_mutex_) {
+      MutexLock lock(&state_mutex_);
+      state_ = State::RUNNING;
+    }
+
+    void SetState(State state) {
+      MutexLock lock(&state_mutex_);
+      state_ = state;
+    }
+
+    // Change the state_ to State::BLOCKED and wait until its state is
+    // changed by WriteBufferManager. When stall is cleared, Signal() is
+    // called to change the state and unblock the DB.
+    void Block() override {
+      MutexLock lock(&state_mutex_);
+      while (state_ == State::BLOCKED) {
+        TEST_SYNC_POINT("WBMStallInterface::BlockDB");
+        state_cv_.Wait();
+      }
+    }
+
+    // Called from WriteBufferManager. This function changes the state_
+    // to State::RUNNING indicating the stall is cleared and DB can proceed.
+    void Signal() override {
+      MutexLock lock(&state_mutex_);
+      state_ = State::RUNNING;
+      state_cv_.Signal();
+    }
+
+   private:
+    // Conditional variable and mutex to block and
+    // signal the DB during stalling process.
+    port::Mutex state_mutex_;
+    port::CondVar state_cv_;
+    // state represting whether DB is running or blocked because of stall by
+    // WriteBufferManager.
+    State state_;
+  };
+
  protected:
   const std::string dbname_;
   std::string db_id_;
@@ -1048,6 +1130,14 @@ class DBImpl : public DB {
   ColumnFamilyHandleImpl* default_cf_handle_;
   InternalStats* default_cf_internal_stats_;
 
+  // table_cache_ provides its own synchronization
+  std::shared_ptr<Cache> table_cache_;
+
+  ErrorHandler error_handler_;
+
+  // Unified interface for logging events
+  EventLogger event_logger_;
+
   // 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_;
@@ -1078,12 +1168,27 @@ class DBImpl : public DB {
   // Default: true
   const bool batch_per_txn_;
 
+  // Each flush or compaction gets its own job id. this counter makes sure
+  // they're unique
+  std::atomic<int> next_job_id_;
+
+  std::atomic<bool> shutting_down_;
+
   // 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.
   // If need_enter_write_thread = false, the method will enter write thread.
   Status WriteOptionsFile(bool need_mutex_lock, bool need_enter_write_thread);
 
+  Status CompactRangeInternal(const CompactRangeOptions& options,
+                              ColumnFamilyHandle* column_family,
+                              const Slice* begin, const Slice* end);
+
+  Status GetApproximateSizesInternal(const SizeApproximationOptions& options,
+                                     ColumnFamilyHandle* column_family,
+                                     const Range* range, int n,
+                                     uint64_t* sizes);
+
   // The following two functions can only be called when:
   // 1. WriteThread::Writer::EnterUnbatched() is used.
   // 2. db_mutex is NOT held
@@ -1195,14 +1300,22 @@ class DBImpl : public DB {
 
   virtual bool OwnTablesAndLogs() const { return true; }
 
+  // Set DB identity file, and write DB ID to manifest if necessary.
+  Status SetDBId(bool read_only);
+
   // REQUIRES: db mutex held when calling this function, but the db mutex can
   // be released and re-acquired. Db mutex will be held when the function
   // returns.
-  // After best-efforts recovery, there may be SST files in db/cf paths that are
-  // not referenced in the MANIFEST. We delete these SST files. In the
+  // After recovery, there may be SST files in db/cf paths that are
+  // not referenced in the MANIFEST (e.g.
+  // 1. It's best effort recovery;
+  // 2. The VersionEdits referencing the SST files are appended to
+  // MANIFEST, DB crashes when syncing the MANIFEST, the VersionEdits are
+  // still not synced to MANIFEST during recovery.)
+  // We delete these SST files. In the
   // meantime, we find out the largest file number present in the paths, and
   // bump up the version set's next_file_number_ to be 1 + largest_file_number.
-  Status FinishBestEffortsRecovery();
+  Status DeleteUnreferencedSstFiles();
 
   // SetDbSessionId() should be called in the constuctor DBImpl()
   // to ensure that db_session_id_ gets updated every time the DB is opened
@@ -1260,6 +1373,7 @@ class DBImpl : public DB {
 
   struct LogFileNumberSize {
     explicit LogFileNumberSize(uint64_t _number) : number(_number) {}
+    LogFileNumberSize() {}
     void AddSize(uint64_t new_size) { size += new_size; }
     uint64_t number;
     uint64_t size = 0;
@@ -1365,6 +1479,7 @@ class DBImpl : public DB {
     DBImpl* db;
     // background compaction takes ownership of `prepicked_compaction`.
     PrepickedCompaction* prepicked_compaction;
+    Env::Priority compaction_pri_;
   };
 
   // Initialize the built-in column family for persistent stats. Depending on
@@ -1459,6 +1574,12 @@ class DBImpl : public DB {
   Status WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
                                      MemTable* mem, VersionEdit* edit);
 
+  // Get the size of a log file and, if truncate is true, truncate the
+  // log file to its actual size, thereby freeing preallocated space.
+  // Return success even if truncate fails
+  Status GetLogSizeAndMaybeTruncate(uint64_t wal_number, bool truncate,
+                                    LogFileNumberSize* log);
+
   // 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
@@ -1469,6 +1590,10 @@ class DBImpl : public DB {
   //            `num_bytes` going through.
   Status DelayWrite(uint64_t num_bytes, const WriteOptions& write_options);
 
+  // Begin stalling of writes when memory usage increases beyond a certain
+  // threshold.
+  void WriteBufferManagerStallWrites();
+
   Status ThrottleLowPriWritesIfNeeded(const WriteOptions& write_options,
                                       WriteBatch* my_batch);
 
@@ -1548,7 +1673,7 @@ class DBImpl : public DB {
   Status SwitchWAL(WriteContext* write_context);
 
   // REQUIRES: mutex locked and in write thread.
-  Status HandleWriteBufferFull(WriteContext* write_context);
+  Status HandleWriteBufferManagerFlush(WriteContext* write_context);
 
   // REQUIRES: mutex locked
   Status PreprocessWrite(const WriteOptions& write_options, bool* need_log_sync,
@@ -1682,7 +1807,9 @@ class DBImpl : public DB {
       std::unique_ptr<TaskLimiterToken>* token, LogBuffer* log_buffer);
 
   // helper function to call after some of the logs_ were synced
-  void MarkLogsSynced(uint64_t up_to, bool synced_dir, const Status& status);
+  Status MarkLogsSynced(uint64_t up_to, bool synced_dir);
+  // WALs with log number up to up_to are not synced successfully.
+  void MarkLogsNotSynced(uint64_t up_to);
 
   SnapshotImpl* GetSnapshotImpl(bool is_write_conflict_boundary,
                                 bool lock = true);
@@ -1821,13 +1948,11 @@ class DBImpl : public DB {
   Status MultiGetImpl(
       const ReadOptions& read_options, size_t start_key, size_t num_keys,
       autovector<KeyContext*, MultiGetContext::MAX_BATCH_SIZE>* sorted_keys,
-      SuperVersion* sv, SequenceNumber snap_seqnum, ReadCallback* callback,
-      bool* is_blob_index);
+      SuperVersion* sv, SequenceNumber snap_seqnum, ReadCallback* callback);
 
   Status DisableFileDeletionsWithLock();
 
-  // table_cache_ provides its own synchronization
-  std::shared_ptr<Cache> table_cache_;
+  Status IncreaseFullHistoryTsLow(ColumnFamilyData* cfd, std::string ts_low);
 
   // Lock over the persistent DB state.  Non-nullptr iff successfully acquired.
   FileLock* db_lock_;
@@ -1842,8 +1967,6 @@ class DBImpl : public DB {
   // mutex_, the order should be first mutex_ and then log_write_mutex_.
   InstrumentedMutex log_write_mutex_;
 
-  std::atomic<bool> shutting_down_;
-
   // If zero, manual compactions are allowed to proceed. If non-zero, manual
   // compactions may still be running, but will quickly fail with
   // `Status::Incomplete`. The value indicates how many threads have paused
@@ -2052,10 +2175,6 @@ class DBImpl : public DB {
   // Number of threads intending to write to memtable
   std::atomic<size_t> pending_memtable_writes_ = {};
 
-  // Each flush or compaction gets its own job id. this counter makes sure
-  // they're unique
-  std::atomic<int> next_job_id_;
-
   // A flag indicating whether the current rocksdb database has any
   // data that is not yet persisted into either WAL or SST file.
   // Used when disableWAL is true.
@@ -2084,9 +2203,6 @@ class DBImpl : public DB {
   WalManager wal_manager_;
 #endif  // ROCKSDB_LITE
 
-  // Unified interface for logging events
-  EventLogger event_logger_;
-
   // A value of > 0 temporarily disables scheduling of background work
   int bg_work_paused_;
 
@@ -2154,8 +2270,6 @@ class DBImpl : public DB {
   // Flag to check whether Close() has been called on this 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
@@ -2168,11 +2282,18 @@ class DBImpl : public DB {
   InstrumentedCondVar atomic_flush_install_cv_;
 
   bool wal_in_db_path_;
+
+  BlobFileCompletionCallback blob_callback_;
+
+  // Pointer to WriteBufferManager stalling interface.
+  std::unique_ptr<StallInterface> wbm_stall_;
 };
 
-extern Options SanitizeOptions(const std::string& db, const Options& src);
+extern Options SanitizeOptions(const std::string& db, const Options& src,
+                               bool read_only = false);
 
-extern DBOptions SanitizeOptions(const std::string& db, const DBOptions& src);
+extern DBOptions SanitizeOptions(const std::string& db, const DBOptions& src,
+                                 bool read_only = false);
 
 extern CompressionType GetCompressionFlush(
     const ImmutableCFOptions& ioptions,
@@ -2184,11 +2305,27 @@ extern CompressionType GetCompressionFlush(
 // `memtables_to_flush`) will be flushed and thus will not depend on any WAL
 // file.
 // The function is only applicable to 2pc mode.
-extern uint64_t PrecomputeMinLogNumberToKeep(
+extern uint64_t PrecomputeMinLogNumberToKeep2PC(
     VersionSet* vset, const ColumnFamilyData& cfd_to_flush,
-    autovector<VersionEdit*> edit_list,
+    const autovector<VersionEdit*>& edit_list,
     const autovector<MemTable*>& memtables_to_flush,
     LogsWithPrepTracker* prep_tracker);
+// For atomic flush.
+extern uint64_t PrecomputeMinLogNumberToKeep2PC(
+    VersionSet* vset, const autovector<ColumnFamilyData*>& cfds_to_flush,
+    const autovector<autovector<VersionEdit*>>& edit_lists,
+    const autovector<const autovector<MemTable*>*>& memtables_to_flush,
+    LogsWithPrepTracker* prep_tracker);
+
+// In non-2PC mode, WALs with log number < the returned number can be
+// deleted after the cfd_to_flush column family is flushed successfully.
+extern uint64_t PrecomputeMinLogNumberToKeepNon2PC(
+    VersionSet* vset, const ColumnFamilyData& cfd_to_flush,
+    const autovector<VersionEdit*>& edit_list);
+// For atomic flush.
+extern uint64_t PrecomputeMinLogNumberToKeepNon2PC(
+    VersionSet* vset, const autovector<ColumnFamilyData*>& cfds_to_flush,
+    const autovector<autovector<VersionEdit*>>& edit_lists);
 
 // `cfd_to_flush` is the column family whose memtable will be flushed and thus
 // will not depend on any WAL file. nullptr means no memtable is being flushed.
@@ -2196,6 +2333,10 @@ extern uint64_t PrecomputeMinLogNumberToKeep(
 extern uint64_t FindMinPrepLogReferencedByMemTable(
     VersionSet* vset, const ColumnFamilyData* cfd_to_flush,
     const autovector<MemTable*>& memtables_to_flush);
+// For atomic flush.
+extern uint64_t FindMinPrepLogReferencedByMemTable(
+    VersionSet* vset, const autovector<ColumnFamilyData*>& cfds_to_flush,
+    const autovector<const autovector<MemTable*>*>& memtables_to_flush);
 
 // Fix user-supplied options to be reasonable
 template <class T, class V>

db/db_impl/db_impl_compaction_flush.cc

@@ -35,8 +35,10 @@ bool DBImpl::EnoughRoomForCompaction(
     // Pass the current bg_error_ to SFM so it can decide what checks to
     // perform. If this DB instance hasn't seen any error yet, the SFM can be
     // optimistic and not do disk space checks
-    enough_room =
-        sfm->EnoughRoomForCompaction(cfd, inputs, error_handler_.GetBGError());
+    Status bg_error = error_handler_.GetBGError();
+    enough_room = sfm->EnoughRoomForCompaction(cfd, inputs, bg_error);
+    bg_error.PermitUncheckedError();  // bg_error is just a copy of the Status
+                                      // from the error_handler_
     if (enough_room) {
       *sfm_reserved_compact_space = true;
     }
@@ -123,20 +125,17 @@ IOStatus DBImpl::SyncClosedLogs(JobContext* job_context) {
 
     // "number <= current_log_number - 1" is equivalent to
     // "number < current_log_number".
-    MarkLogsSynced(current_log_number - 1, true, io_s);
+    if (io_s.ok()) {
+      io_s = status_to_io_status(MarkLogsSynced(current_log_number - 1, true));
+    } else {
+      MarkLogsNotSynced(current_log_number - 1);
+    }
     if (!io_s.ok()) {
-      if (total_log_size_ > 0) {
-        error_handler_.SetBGError(io_s, BackgroundErrorReason::kFlush)
-            .PermitUncheckedError();
-      } else {
-        // If the WAL is empty, we use different error reason
-        error_handler_.SetBGError(io_s, BackgroundErrorReason::kFlushNoWAL)
-            .PermitUncheckedError();
-      }
       TEST_SYNC_POINT("DBImpl::SyncClosedLogs:Failed");
       return io_s;
     }
   }
+  TEST_SYNC_POINT("DBImpl::SyncClosedLogs:end");
   return io_s;
 }
 
@@ -155,27 +154,25 @@ Status DBImpl::FlushMemTableToOutputFile(
 
   FlushJob flush_job(
       dbname_, cfd, immutable_db_options_, mutable_cf_options,
-      nullptr /* memtable_id */, file_options_for_compaction_, versions_.get(),
-      &mutex_, &shutting_down_, snapshot_seqs, earliest_write_conflict_snapshot,
-      snapshot_checker, job_context, log_buffer, directories_.GetDbDir(),
-      GetDataDir(cfd, 0U),
+      port::kMaxUint64 /* memtable_id */, file_options_for_compaction_,
+      versions_.get(), &mutex_, &shutting_down_, snapshot_seqs,
+      earliest_write_conflict_snapshot, snapshot_checker, job_context,
+      log_buffer, directories_.GetDbDir(), GetDataDir(cfd, 0U),
       GetCompressionFlush(*cfd->ioptions(), mutable_cf_options), stats_,
       &event_logger_, mutable_cf_options.report_bg_io_stats,
       true /* sync_output_directory */, true /* write_manifest */, thread_pri,
-      io_tracer_, db_id_, db_session_id_);
+      io_tracer_, db_id_, db_session_id_, cfd->GetFullHistoryTsLow(),
+      &blob_callback_);
   FileMetaData file_meta;
 
-  TEST_SYNC_POINT("DBImpl::FlushMemTableToOutputFile:BeforePickMemtables");
-  flush_job.PickMemTable();
-  TEST_SYNC_POINT("DBImpl::FlushMemTableToOutputFile:AfterPickMemtables");
-
 #ifndef ROCKSDB_LITE
   // may temporarily unlock and lock the mutex.
   NotifyOnFlushBegin(cfd, &file_meta, mutable_cf_options, job_context->job_id);
 #endif  // ROCKSDB_LITE
 
   Status s;
-  IOStatus io_s = IOStatus::OK();
+  bool need_cancel = false;
+  IOStatus log_io_s = IOStatus::OK();
   if (logfile_number_ > 0 &&
       versions_->GetColumnFamilySet()->NumberOfColumnFamilies() > 1) {
     // If there are more than one column families, we need to make sure that
@@ -184,16 +181,24 @@ Status DBImpl::FlushMemTableToOutputFile(
     // flushed SST may contain data from write batches whose updates to
     // other column families are missing.
     // SyncClosedLogs() may unlock and re-lock the db_mutex.
-    io_s = SyncClosedLogs(job_context);
-    s = io_s;
-    if (!io_s.ok() && !io_s.IsShutdownInProgress() &&
-        !io_s.IsColumnFamilyDropped()) {
-      error_handler_.SetBGError(io_s, BackgroundErrorReason::kFlush)
-          .PermitUncheckedError();
+    log_io_s = SyncClosedLogs(job_context);
+    if (!log_io_s.ok() && !log_io_s.IsShutdownInProgress() &&
+        !log_io_s.IsColumnFamilyDropped()) {
+      error_handler_.SetBGError(log_io_s, BackgroundErrorReason::kFlush);
     }
   } else {
     TEST_SYNC_POINT("DBImpl::SyncClosedLogs:Skip");
   }
+  s = log_io_s;
+
+  // If the log sync failed, we do not need to pick memtable. Otherwise,
+  // num_flush_not_started_ needs to be rollback.
+  TEST_SYNC_POINT("DBImpl::FlushMemTableToOutputFile:BeforePickMemtables");
+  if (s.ok()) {
+    flush_job.PickMemTable();
+    need_cancel = true;
+  }
+  TEST_SYNC_POINT("DBImpl::FlushMemTableToOutputFile:AfterPickMemtables");
 
   // Within flush_job.Run, rocksdb may call event listener to notify
   // file creation and deletion.
@@ -203,11 +208,16 @@ Status DBImpl::FlushMemTableToOutputFile(
   // is unlocked by the current thread.
   if (s.ok()) {
     s = flush_job.Run(&logs_with_prep_tracker_, &file_meta);
-  } else {
+    need_cancel = false;
+  }
+
+  if (!s.ok() && need_cancel) {
     flush_job.Cancel();
   }
-  if (io_s.ok()) {
-    io_s = flush_job.io_status();
+  IOStatus io_s = IOStatus::OK();
+  io_s = flush_job.io_status();
+  if (s.ok()) {
+    s = io_s;
   }
 
   if (s.ok()) {
@@ -243,30 +253,30 @@ Status DBImpl::FlushMemTableToOutputFile(
   if (!s.ok() && !s.IsShutdownInProgress() && !s.IsColumnFamilyDropped()) {
     if (!io_s.ok() && !io_s.IsShutdownInProgress() &&
         !io_s.IsColumnFamilyDropped()) {
+      assert(log_io_s.ok());
       // Error while writing to MANIFEST.
       // In fact, versions_->io_status() can also be the result of renaming
       // CURRENT file. With current code, it's just difficult to tell. So just
       // be pessimistic and try write to a new MANIFEST.
       // TODO: distinguish between MANIFEST write and CURRENT renaming
       if (!versions_->io_status().ok()) {
-        // Should handle return error?
-        error_handler_.SetBGError(io_s, BackgroundErrorReason::kManifestWrite)
-            .PermitUncheckedError();
-      } else if (total_log_size_ > 0) {
-        // Should handle return error?
-        error_handler_.SetBGError(io_s, BackgroundErrorReason::kFlush)
-            .PermitUncheckedError();
+        // If WAL sync is successful (either WAL size is 0 or there is no IO
+        // error), all the Manifest write will be map to soft error.
+        // TODO: kManifestWriteNoWAL and kFlushNoWAL are misleading. Refactor is
+        // needed.
+        error_handler_.SetBGError(io_s,
+                                  BackgroundErrorReason::kManifestWriteNoWAL);
       } else {
-        // If the WAL is empty, we use different error reason
-        // Should handle return error?
-        error_handler_.SetBGError(io_s, BackgroundErrorReason::kFlushNoWAL)
-            .PermitUncheckedError();
+        // If WAL sync is successful (either WAL size is 0 or there is no IO
+        // error), all the other SST file write errors will be set as
+        // kFlushNoWAL.
+        error_handler_.SetBGError(io_s, BackgroundErrorReason::kFlushNoWAL);
       }
     } else {
-      Status new_bg_error = s;
-      // Should handle return error?
-      error_handler_.SetBGError(new_bg_error, BackgroundErrorReason::kFlush)
-          .PermitUncheckedError();
+      if (log_io_s.ok()) {
+        Status new_bg_error = s;
+        error_handler_.SetBGError(new_bg_error, BackgroundErrorReason::kFlush);
+      }
     }
   } else {
     // If we got here, then we decided not to care about the i_os status (either
@@ -284,16 +294,17 @@ Status DBImpl::FlushMemTableToOutputFile(
       // Notify sst_file_manager that a new file was added
       std::string file_path = MakeTableFileName(
           cfd->ioptions()->cf_paths[0].path, file_meta.fd.GetNumber());
-      sfm->OnAddFile(file_path);
+      // TODO (PR7798).  We should only add the file to the FileManager if it
+      // exists. Otherwise, some tests may fail.  Ignore the error in the
+      // interim.
+      sfm->OnAddFile(file_path).PermitUncheckedError();
       if (sfm->IsMaxAllowedSpaceReached()) {
         Status new_bg_error =
             Status::SpaceLimit("Max allowed space was reached");
         TEST_SYNC_POINT_CALLBACK(
             "DBImpl::FlushMemTableToOutputFile:MaxAllowedSpaceReached",
             &new_bg_error);
-        // Should handle this error?
-        error_handler_.SetBGError(new_bg_error, BackgroundErrorReason::kFlush)
-            .PermitUncheckedError();
+        error_handler_.SetBGError(new_bg_error, BackgroundErrorReason::kFlush);
       }
     }
 #endif  // ROCKSDB_LITE
@@ -309,30 +320,22 @@ Status DBImpl::FlushMemTablesToOutputFiles(
     return AtomicFlushMemTablesToOutputFiles(
         bg_flush_args, made_progress, job_context, log_buffer, thread_pri);
   }
+  assert(bg_flush_args.size() == 1);
   std::vector<SequenceNumber> snapshot_seqs;
   SequenceNumber earliest_write_conflict_snapshot;
   SnapshotChecker* snapshot_checker;
   GetSnapshotContext(job_context, &snapshot_seqs,
                      &earliest_write_conflict_snapshot, &snapshot_checker);
-  Status status;
-  for (auto& arg : bg_flush_args) {
-    ColumnFamilyData* cfd = arg.cfd_;
-    MutableCFOptions mutable_cf_options = *cfd->GetLatestMutableCFOptions();
-    SuperVersionContext* superversion_context = arg.superversion_context_;
-    Status s = FlushMemTableToOutputFile(
-        cfd, mutable_cf_options, made_progress, job_context,
-        superversion_context, snapshot_seqs, earliest_write_conflict_snapshot,
-        snapshot_checker, log_buffer, thread_pri);
-    if (!s.ok()) {
-      status = s;
-      if (!s.IsShutdownInProgress() && !s.IsColumnFamilyDropped()) {
-        // At this point, DB is not shutting down, nor is cfd dropped.
-        // Something is wrong, thus we break out of the loop.
-        break;
-      }
-    }
-  }
-  return status;
+  const auto& bg_flush_arg = bg_flush_args[0];
+  ColumnFamilyData* cfd = bg_flush_arg.cfd_;
+  MutableCFOptions mutable_cf_options = *cfd->GetLatestMutableCFOptions();
+  SuperVersionContext* superversion_context =
+      bg_flush_arg.superversion_context_;
+  Status s = FlushMemTableToOutputFile(
+      cfd, mutable_cf_options, made_progress, job_context, superversion_context,
+      snapshot_seqs, earliest_write_conflict_snapshot, snapshot_checker,
+      log_buffer, thread_pri);
+  return s;
 }
 
 /*
@@ -395,7 +398,7 @@ Status DBImpl::AtomicFlushMemTablesToOutputFiles(
 
     all_mutable_cf_options.emplace_back(*cfd->GetLatestMutableCFOptions());
     const MutableCFOptions& mutable_cf_options = all_mutable_cf_options.back();
-    const uint64_t* max_memtable_id = &(bg_flush_args[i].max_memtable_id_);
+    uint64_t max_memtable_id = bg_flush_args[i].max_memtable_id_;
     jobs.emplace_back(new FlushJob(
         dbname_, cfd, immutable_db_options_, mutable_cf_options,
         max_memtable_id, file_options_for_compaction_, versions_.get(), &mutex_,
@@ -404,13 +407,13 @@ Status DBImpl::AtomicFlushMemTablesToOutputFiles(
         data_dir, GetCompressionFlush(*cfd->ioptions(), mutable_cf_options),
         stats_, &event_logger_, mutable_cf_options.report_bg_io_stats,
         false /* sync_output_directory */, false /* write_manifest */,
-        thread_pri, io_tracer_, db_id_, db_session_id_));
-    jobs.back()->PickMemTable();
+        thread_pri, io_tracer_, db_id_, db_session_id_,
+        cfd->GetFullHistoryTsLow()));
   }
 
   std::vector<FileMetaData> file_meta(num_cfs);
   Status s;
-  IOStatus io_s;
+  IOStatus log_io_s = IOStatus::OK();
   assert(num_cfs == static_cast<int>(jobs.size()));
 
 #ifndef ROCKSDB_LITE
@@ -425,18 +428,36 @@ Status DBImpl::AtomicFlushMemTablesToOutputFiles(
   if (logfile_number_ > 0) {
     // TODO (yanqin) investigate whether we should sync the closed logs for
     // single column family case.
-    io_s = SyncClosedLogs(job_context);
-    s = io_s;
+    log_io_s = SyncClosedLogs(job_context);
+    if (!log_io_s.ok() && !log_io_s.IsShutdownInProgress() &&
+        !log_io_s.IsColumnFamilyDropped()) {
+      if (total_log_size_ > 0) {
+        error_handler_.SetBGError(log_io_s, BackgroundErrorReason::kFlush);
+      } else {
+        // If the WAL is empty, we use different error reason
+        error_handler_.SetBGError(log_io_s, BackgroundErrorReason::kFlushNoWAL);
+      }
+    }
   }
+  s = log_io_s;
 
   // exec_status stores the execution status of flush_jobs as
   // <bool /* executed */, Status /* status code */>
   autovector<std::pair<bool, Status>> exec_status;
   autovector<IOStatus> io_status;
+  std::vector<bool> pick_status;
   for (int i = 0; i != num_cfs; ++i) {
     // Initially all jobs are not executed, with status OK.
     exec_status.emplace_back(false, Status::OK());
     io_status.emplace_back(IOStatus::OK());
+    pick_status.push_back(false);
+  }
+
+  if (s.ok()) {
+    for (int i = 0; i != num_cfs; ++i) {
+      jobs[i]->PickMemTable();
+      pick_status[i] = true;
+    }
   }
 
   if (s.ok()) {
@@ -477,6 +498,7 @@ Status DBImpl::AtomicFlushMemTablesToOutputFiles(
     s = error_status.ok() ? s : error_status;
   }
 
+  IOStatus io_s = IOStatus::OK();
   if (io_s.ok()) {
     IOStatus io_error = IOStatus::OK();
     for (int i = 0; i != static_cast<int>(io_status.size()); i++) {
@@ -512,12 +534,12 @@ Status DBImpl::AtomicFlushMemTablesToOutputFiles(
     // Have to cancel the flush jobs that have NOT executed because we need to
     // unref the versions.
     for (int i = 0; i != num_cfs; ++i) {
-      if (!exec_status[i].first) {
+      if (pick_status[i] && !exec_status[i].first) {
         jobs[i]->Cancel();
       }
     }
     for (int i = 0; i != num_cfs; ++i) {
-      if (exec_status[i].first && exec_status[i].second.ok()) {
+      if (exec_status[i].second.ok() && exec_status[i].first) {
         auto& mems = jobs[i]->GetMemTables();
         cfds[i]->imm()->RollbackMemtableFlush(mems,
                                               file_meta[i].fd.GetNumber());
@@ -581,7 +603,7 @@ Status DBImpl::AtomicFlushMemTablesToOutputFiles(
 
     s = InstallMemtableAtomicFlushResults(
         nullptr /* imm_lists */, tmp_cfds, mutable_cf_options_list, mems_list,
-        versions_.get(), &mutex_, tmp_file_meta,
+        versions_.get(), &logs_with_prep_tracker_, &mutex_, tmp_file_meta,
         &job_context->memtables_to_free, directories_.GetDbDir(), log_buffer);
   }
 
@@ -627,7 +649,7 @@ Status DBImpl::AtomicFlushMemTablesToOutputFiles(
     auto sfm = static_cast<SstFileManagerImpl*>(
         immutable_db_options_.sst_file_manager.get());
     assert(all_mutable_cf_options.size() == static_cast<size_t>(num_cfs));
-    for (int i = 0; i != num_cfs; ++i) {
+    for (int i = 0; s.ok() && i != num_cfs; ++i) {
       if (cfds[i]->IsDropped()) {
         continue;
       }
@@ -636,14 +658,16 @@ Status DBImpl::AtomicFlushMemTablesToOutputFiles(
       if (sfm) {
         std::string file_path = MakeTableFileName(
             cfds[i]->ioptions()->cf_paths[0].path, file_meta[i].fd.GetNumber());
-        sfm->OnAddFile(file_path);
+        // TODO (PR7798).  We should only add the file to the FileManager if it
+        // exists. Otherwise, some tests may fail.  Ignore the error in the
+        // interim.
+        sfm->OnAddFile(file_path).PermitUncheckedError();
         if (sfm->IsMaxAllowedSpaceReached() &&
             error_handler_.GetBGError().ok()) {
           Status new_bg_error =
               Status::SpaceLimit("Max allowed space was reached");
-          // Should Handle this error?
-          error_handler_.SetBGError(new_bg_error, BackgroundErrorReason::kFlush)
-              .PermitUncheckedError();
+          error_handler_.SetBGError(new_bg_error,
+                                    BackgroundErrorReason::kFlush);
         }
       }
     }
@@ -654,30 +678,30 @@ Status DBImpl::AtomicFlushMemTablesToOutputFiles(
   // it is not because of CF drop.
   if (!s.ok() && !s.IsColumnFamilyDropped()) {
     if (!io_s.ok() && !io_s.IsColumnFamilyDropped()) {
+      assert(log_io_s.ok());
       // Error while writing to MANIFEST.
       // In fact, versions_->io_status() can also be the result of renaming
       // CURRENT file. With current code, it's just difficult to tell. So just
       // be pessimistic and try write to a new MANIFEST.
       // TODO: distinguish between MANIFEST write and CURRENT renaming
       if (!versions_->io_status().ok()) {
-        // Should Handle this error?
-        error_handler_.SetBGError(io_s, BackgroundErrorReason::kManifestWrite)
-            .PermitUncheckedError();
-      } else if (total_log_size_ > 0) {
-        // Should Handle this error?
-        error_handler_.SetBGError(io_s, BackgroundErrorReason::kFlush)
-            .PermitUncheckedError();
+        // If WAL sync is successful (either WAL size is 0 or there is no IO
+        // error), all the Manifest write will be map to soft error.
+        // TODO: kManifestWriteNoWAL and kFlushNoWAL are misleading. Refactor
+        // is needed.
+        error_handler_.SetBGError(io_s,
+                                  BackgroundErrorReason::kManifestWriteNoWAL);
       } else {
-        // If the WAL is empty, we use different error reason
-        // Should Handle this error?
-        error_handler_.SetBGError(io_s, BackgroundErrorReason::kFlushNoWAL)
-            .PermitUncheckedError();
+        // If WAL sync is successful (either WAL size is 0 or there is no IO
+        // error), all the other SST file write errors will be set as
+        // kFlushNoWAL.
+        error_handler_.SetBGError(io_s, BackgroundErrorReason::kFlushNoWAL);
       }
     } else {
-      Status new_bg_error = s;
-      // Should Handle this error?
-      error_handler_.SetBGError(new_bg_error, BackgroundErrorReason::kFlush)
-          .PermitUncheckedError();
+      if (log_io_s.ok()) {
+        Status new_bg_error = s;
+        error_handler_.SetBGError(new_bg_error, BackgroundErrorReason::kFlush);
+      }
     }
   }
 
@@ -777,7 +801,64 @@ void DBImpl::NotifyOnFlushCompleted(
 
 Status DBImpl::CompactRange(const CompactRangeOptions& options,
                             ColumnFamilyHandle* column_family,
-                            const Slice* begin, const Slice* end) {
+                            const Slice* begin_without_ts,
+                            const Slice* end_without_ts) {
+  if (manual_compaction_paused_.load(std::memory_order_acquire) > 0) {
+    return Status::Incomplete(Status::SubCode::kManualCompactionPaused);
+  }
+
+  const Comparator* const ucmp = column_family->GetComparator();
+  assert(ucmp);
+  size_t ts_sz = ucmp->timestamp_size();
+  if (ts_sz == 0) {
+    return CompactRangeInternal(options, column_family, begin_without_ts,
+                                end_without_ts);
+  }
+
+  std::string begin_str;
+  std::string end_str;
+
+  // CompactRange compact all keys: [begin, end] inclusively. Add maximum
+  // timestamp to include all `begin` keys, and add minimal timestamp to include
+  // all `end` keys.
+  if (begin_without_ts != nullptr) {
+    AppendKeyWithMaxTimestamp(&begin_str, *begin_without_ts, ts_sz);
+  }
+  if (end_without_ts != nullptr) {
+    AppendKeyWithMinTimestamp(&end_str, *end_without_ts, ts_sz);
+  }
+  Slice begin(begin_str);
+  Slice end(end_str);
+
+  Slice* begin_with_ts = begin_without_ts ? &begin : nullptr;
+  Slice* end_with_ts = end_without_ts ? &end : nullptr;
+
+  return CompactRangeInternal(options, column_family, begin_with_ts,
+                              end_with_ts);
+}
+
+Status DBImpl::IncreaseFullHistoryTsLow(ColumnFamilyData* cfd,
+                                        std::string ts_low) {
+  VersionEdit edit;
+  edit.SetColumnFamily(cfd->GetID());
+  edit.SetFullHistoryTsLow(ts_low);
+
+  InstrumentedMutexLock l(&mutex_);
+  std::string current_ts_low = cfd->GetFullHistoryTsLow();
+  const Comparator* ucmp = cfd->user_comparator();
+  if (!current_ts_low.empty() &&
+      ucmp->CompareTimestamp(ts_low, current_ts_low) < 0) {
+    return Status::InvalidArgument(
+        "Cannot decrease full_history_timestamp_low");
+  }
+
+  return versions_->LogAndApply(cfd, *cfd->GetLatestMutableCFOptions(), &edit,
+                                &mutex_);
+}
+
+Status DBImpl::CompactRangeInternal(const CompactRangeOptions& options,
+                                    ColumnFamilyHandle* column_family,
+                                    const Slice* begin, const Slice* end) {
   auto cfh = static_cast_with_check<ColumnFamilyHandleImpl>(column_family);
   auto cfd = cfh->cfd();
 
@@ -786,18 +867,36 @@ Status DBImpl::CompactRange(const CompactRangeOptions& options,
   }
 
   bool flush_needed = true;
+
+  // Update full_history_ts_low if it's set
+  if (options.full_history_ts_low != nullptr &&
+      !options.full_history_ts_low->empty()) {
+    std::string ts_low = options.full_history_ts_low->ToString();
+    if (begin != nullptr || end != nullptr) {
+      return Status::InvalidArgument(
+          "Cannot specify compaction range with full_history_ts_low");
+    }
+    Status s = IncreaseFullHistoryTsLow(cfd, ts_low);
+    if (!s.ok()) {
+      LogFlush(immutable_db_options_.info_log);
+      return s;
+    }
+  }
+
+  Status s;
   if (begin != nullptr && end != nullptr) {
     // TODO(ajkr): We could also optimize away the flush in certain cases where
     // one/both sides of the interval are unbounded. But it requires more
     // changes to RangesOverlapWithMemtables.
     Range range(*begin, *end);
     SuperVersion* super_version = cfd->GetReferencedSuperVersion(this);
-    cfd->RangesOverlapWithMemtables({range}, super_version, &flush_needed);
+    s = cfd->RangesOverlapWithMemtables(
+        {range}, super_version, immutable_db_options_.allow_data_in_errors,
+        &flush_needed);
     CleanupSuperVersion(super_version);
   }
 
-  Status s;
-  if (flush_needed) {
+  if (s.ok() && flush_needed) {
     FlushOptions fo;
     fo.allow_write_stall = options.allow_write_stall;
     if (immutable_db_options_.atomic_flush) {
@@ -950,7 +1049,8 @@ Status DBImpl::CompactRange(const CompactRangeOptions& options,
       s = ReFitLevel(cfd, final_output_level, options.target_level);
       TEST_SYNC_POINT("DBImpl::CompactRange:PostRefitLevel");
       // ContinueBackgroundWork always return Status::OK().
-      assert(ContinueBackgroundWork().ok());
+      Status temp_s = ContinueBackgroundWork();
+      assert(temp_s.ok());
     }
     EnableManualCompaction();
   }
@@ -1146,13 +1246,15 @@ Status DBImpl::CompactFilesImpl(
       job_context->job_id, c.get(), immutable_db_options_,
       file_options_for_compaction_, versions_.get(), &shutting_down_,
       preserve_deletes_seqnum_.load(), log_buffer, directories_.GetDbDir(),
-      GetDataDir(c->column_family_data(), c->output_path_id()), stats_, &mutex_,
-      &error_handler_, snapshot_seqs, earliest_write_conflict_snapshot,
-      snapshot_checker, table_cache_, &event_logger_,
+      GetDataDir(c->column_family_data(), c->output_path_id()),
+      GetDataDir(c->column_family_data(), 0), stats_, &mutex_, &error_handler_,
+      snapshot_seqs, earliest_write_conflict_snapshot, snapshot_checker,
+      table_cache_, &event_logger_,
       c->mutable_cf_options()->paranoid_file_checks,
       c->mutable_cf_options()->report_bg_io_stats, dbname_,
       &compaction_job_stats, Env::Priority::USER, io_tracer_,
-      &manual_compaction_paused_, db_id_, db_session_id_);
+      &manual_compaction_paused_, db_id_, db_session_id_,
+      c->column_family_data()->GetFullHistoryTsLow());
 
   // Creating a compaction influences the compaction score because the score
   // takes running compactions into account (by skipping files that are already
@@ -1166,7 +1268,8 @@ Status DBImpl::CompactFilesImpl(
   mutex_.Unlock();
   TEST_SYNC_POINT("CompactFilesImpl:0");
   TEST_SYNC_POINT("CompactFilesImpl:1");
-  compaction_job.Run();
+  // Ignore the status here, as it will be checked in the Install down below...
+  compaction_job.Run().PermitUncheckedError();
   TEST_SYNC_POINT("CompactFilesImpl:2");
   TEST_SYNC_POINT("CompactFilesImpl:3");
   mutex_.Lock();
@@ -1216,11 +1319,9 @@ Status DBImpl::CompactFilesImpl(
                    job_context->job_id, status.ToString().c_str());
     IOStatus io_s = compaction_job.io_status();
     if (!io_s.ok()) {
-      error_handler_.SetBGError(io_s, BackgroundErrorReason::kCompaction)
-          .PermitUncheckedError();
+      error_handler_.SetBGError(io_s, BackgroundErrorReason::kCompaction);
     } else {
-      error_handler_.SetBGError(status, BackgroundErrorReason::kCompaction)
-          .PermitUncheckedError();
+      error_handler_.SetBGError(status, BackgroundErrorReason::kCompaction);
     }
   }
 
@@ -1365,8 +1466,6 @@ Status DBImpl::ReFitLevel(ColumnFamilyData* cfd, int level, int target_level) {
 
   SuperVersionContext sv_context(/* create_superversion */ true);
 
-  Status status;
-
   InstrumentedMutexLock guard_lock(&mutex_);
 
   // only allow one thread refitting
@@ -1430,8 +1529,9 @@ Status DBImpl::ReFitLevel(ColumnFamilyData* cfd, int level, int target_level) {
                     "[%s] Apply version edit:\n%s", cfd->GetName().c_str(),
                     edit.DebugString().data());
 
-    status = versions_->LogAndApply(cfd, mutable_cf_options, &edit, &mutex_,
-                                    directories_.GetDbDir());
+    Status status = versions_->LogAndApply(cfd, mutable_cf_options, &edit,
+                                           &mutex_, directories_.GetDbDir());
+
     InstallSuperVersionAndScheduleWork(cfd, &sv_context, mutable_cf_options);
 
     ROCKS_LOG_DEBUG(immutable_db_options_.info_log, "[%s] LogAndApply: %s\n",
@@ -1442,12 +1542,14 @@ Status DBImpl::ReFitLevel(ColumnFamilyData* cfd, int level, int target_level) {
                       "[%s] After refitting:\n%s", cfd->GetName().c_str(),
                       cfd->current()->DebugString().data());
     }
+    sv_context.Clean();
+    refitting_level_ = false;
+
+    return status;
   }
 
-  sv_context.Clean();
   refitting_level_ = false;
-
-  return status;
+  return Status::OK();
 }
 
 int DBImpl::NumberLevels(ColumnFamilyHandle* column_family) {
@@ -1643,6 +1745,7 @@ Status DBImpl::RunManualCompaction(
       }
       ca = new CompactionArg;
       ca->db = this;
+      ca->compaction_pri_ = Env::Priority::LOW;
       ca->prepicked_compaction = new PrepickedCompaction;
       ca->prepicked_compaction->manual_compaction_state = &manual;
       ca->prepicked_compaction->compaction = compaction;
@@ -1689,8 +1792,9 @@ void DBImpl::GenerateFlushRequest(const autovector<ColumnFamilyData*>& cfds,
 Status DBImpl::FlushMemTable(ColumnFamilyData* cfd,
                              const FlushOptions& flush_options,
                              FlushReason flush_reason, bool writes_stopped) {
+  // This method should not be called if atomic_flush is true.
+  assert(!immutable_db_options_.atomic_flush);
   Status s;
-  uint64_t flush_memtable_id = 0;
   if (!flush_options.allow_write_stall) {
     bool flush_needed = true;
     s = WaitUntilFlushWouldNotStallWrites(cfd, &flush_needed);
@@ -1700,7 +1804,8 @@ Status DBImpl::FlushMemTable(ColumnFamilyData* cfd,
     }
   }
 
-  FlushRequest flush_req;
+  autovector<FlushRequest> flush_reqs;
+  autovector<uint64_t> memtable_ids_to_wait;
   {
     WriteContext context;
     InstrumentedMutexLock guard_lock(&mutex_);
@@ -1715,18 +1820,24 @@ Status DBImpl::FlushMemTable(ColumnFamilyData* cfd,
     }
     WaitForPendingWrites();
 
-    if (!cfd->mem()->IsEmpty() || !cached_recoverable_state_empty_.load()) {
-      if (flush_reason != FlushReason::kErrorRecoveryRetryFlush) {
-        s = SwitchMemtable(cfd, &context);
-      } else {
-        assert(cfd->imm()->NumNotFlushed() > 0);
-      }
+    if (flush_reason != FlushReason::kErrorRecoveryRetryFlush &&
+        (!cfd->mem()->IsEmpty() || !cached_recoverable_state_empty_.load())) {
+      // Note that, when flush reason is kErrorRecoveryRetryFlush, during the
+      // auto retry resume, we want to avoid creating new small memtables.
+      // Therefore, SwitchMemtable will not be called. Also, since ResumeImpl
+      // will iterate through all the CFs and call FlushMemtable during auto
+      // retry resume, it is possible that in some CFs,
+      // cfd->imm()->NumNotFlushed() = 0. In this case, so no flush request will
+      // be created and scheduled, status::OK() will be returned.
+      s = SwitchMemtable(cfd, &context);
     }
+    const uint64_t flush_memtable_id = port::kMaxUint64;
     if (s.ok()) {
       if (cfd->imm()->NumNotFlushed() != 0 || !cfd->mem()->IsEmpty() ||
           !cached_recoverable_state_empty_.load()) {
-        flush_memtable_id = cfd->imm()->GetLatestMemTableID();
-        flush_req.emplace_back(cfd, flush_memtable_id);
+        FlushRequest req{{cfd, flush_memtable_id}};
+        flush_reqs.emplace_back(std::move(req));
+        memtable_ids_to_wait.emplace_back(cfd->imm()->GetLatestMemTableID());
       }
       if (immutable_db_options_.persist_stats_to_disk &&
           flush_reason != FlushReason::kErrorRecoveryRetryFlush) {
@@ -1752,15 +1863,19 @@ Status DBImpl::FlushMemTable(ColumnFamilyData* cfd,
                            "to avoid holding old logs",
                            cfd->GetName().c_str());
             s = SwitchMemtable(cfd_stats, &context);
-            flush_memtable_id = cfd_stats->imm()->GetLatestMemTableID();
-            flush_req.emplace_back(cfd_stats, flush_memtable_id);
+            FlushRequest req{{cfd_stats, flush_memtable_id}};
+            flush_reqs.emplace_back(std::move(req));
+            memtable_ids_to_wait.emplace_back(
+                cfd->imm()->GetLatestMemTableID());
           }
         }
       }
     }
-    if (s.ok() && !flush_req.empty()) {
-      for (auto& elem : flush_req) {
-        ColumnFamilyData* loop_cfd = elem.first;
+
+    if (s.ok() && !flush_reqs.empty()) {
+      for (const auto& req : flush_reqs) {
+        assert(req.size() == 1);
+        ColumnFamilyData* loop_cfd = req[0].first;
         loop_cfd->imm()->FlushRequested();
       }
       // If the caller wants to wait for this flush to complete, it indicates
@@ -1768,12 +1883,15 @@ Status DBImpl::FlushMemTable(ColumnFamilyData* cfd,
       // other threads which may drop the column family concurrently.
       // Therefore, we increase the cfd's ref count.
       if (flush_options.wait) {
-        for (auto& elem : flush_req) {
-          ColumnFamilyData* loop_cfd = elem.first;
+        for (const auto& req : flush_reqs) {
+          assert(req.size() == 1);
+          ColumnFamilyData* loop_cfd = req[0].first;
           loop_cfd->Ref();
         }
       }
-      SchedulePendingFlush(flush_req, flush_reason);
+      for (const auto& req : flush_reqs) {
+        SchedulePendingFlush(req, flush_reason);
+      }
       MaybeScheduleFlushOrCompaction();
     }
 
@@ -1789,9 +1907,11 @@ Status DBImpl::FlushMemTable(ColumnFamilyData* cfd,
   if (s.ok() && flush_options.wait) {
     autovector<ColumnFamilyData*> cfds;
     autovector<const uint64_t*> flush_memtable_ids;
-    for (auto& iter : flush_req) {
-      cfds.push_back(iter.first);
-      flush_memtable_ids.push_back(&(iter.second));
+    assert(flush_reqs.size() == memtable_ids_to_wait.size());
+    for (size_t i = 0; i < flush_reqs.size(); ++i) {
+      assert(flush_reqs[i].size() == 1);
+      cfds.push_back(flush_reqs[i][0].first);
+      flush_memtable_ids.push_back(&(memtable_ids_to_wait[i]));
     }
     s = WaitForFlushMemTables(
         cfds, flush_memtable_ids,
@@ -1973,12 +2093,12 @@ Status DBImpl::WaitUntilFlushWouldNotStallWrites(ColumnFamilyData* cfd,
       // check whether one extra immutable memtable or an extra L0 file would
       // cause write stalling mode to be entered. It could still enter stall
       // mode due to pending compaction bytes, but that's less common
-      write_stall_condition =
-          ColumnFamilyData::GetWriteStallConditionAndCause(
-              cfd->imm()->NumNotFlushed() + 1,
-              vstorage->l0_delay_trigger_count() + 1,
-              vstorage->estimated_compaction_needed_bytes(), mutable_cf_options)
-              .first;
+      write_stall_condition = ColumnFamilyData::GetWriteStallConditionAndCause(
+                                  cfd->imm()->NumNotFlushed() + 1,
+                                  vstorage->l0_delay_trigger_count() + 1,
+                                  vstorage->estimated_compaction_needed_bytes(),
+                                  mutable_cf_options, *cfd->ioptions())
+                                  .first;
     } while (write_stall_condition != WriteStallCondition::kNormal);
   }
   return Status::OK();
@@ -2157,6 +2277,7 @@ void DBImpl::MaybeScheduleFlushOrCompaction() {
          unscheduled_compactions_ > 0) {
     CompactionArg* ca = new CompactionArg;
     ca->db = this;
+    ca->compaction_pri_ = Env::Priority::LOW;
     ca->prepicked_compaction = nullptr;
     bg_compaction_scheduled_++;
     unscheduled_compactions_--;
@@ -2216,6 +2337,17 @@ DBImpl::FlushRequest DBImpl::PopFirstFromFlushQueue() {
   assert(!flush_queue_.empty());
   FlushRequest flush_req = flush_queue_.front();
   flush_queue_.pop_front();
+  if (!immutable_db_options_.atomic_flush) {
+    assert(flush_req.size() == 1);
+  }
+  for (const auto& elem : flush_req) {
+    if (!immutable_db_options_.atomic_flush) {
+      ColumnFamilyData* cfd = elem.first;
+      assert(cfd);
+      assert(cfd->queued_for_flush());
+      cfd->set_queued_for_flush(false);
+    }
+  }
   // TODO: need to unset flush reason?
   return flush_req;
 }
@@ -2248,19 +2380,36 @@ ColumnFamilyData* DBImpl::PickCompactionFromQueue(
 
 void DBImpl::SchedulePendingFlush(const FlushRequest& flush_req,
                                   FlushReason flush_reason) {
+  mutex_.AssertHeld();
   if (flush_req.empty()) {
     return;
   }
-  for (auto& iter : flush_req) {
-    ColumnFamilyData* cfd = iter.first;
-    cfd->Ref();
-    cfd->SetFlushReason(flush_reason);
+  if (!immutable_db_options_.atomic_flush) {
+    // For the non-atomic flush case, we never schedule multiple column
+    // families in the same flush request.
+    assert(flush_req.size() == 1);
+    ColumnFamilyData* cfd = flush_req[0].first;
+    assert(cfd);
+    if (!cfd->queued_for_flush() && cfd->imm()->IsFlushPending()) {
+      cfd->Ref();
+      cfd->set_queued_for_flush(true);
+      cfd->SetFlushReason(flush_reason);
+      ++unscheduled_flushes_;
+      flush_queue_.push_back(flush_req);
+    }
+  } else {
+    for (auto& iter : flush_req) {
+      ColumnFamilyData* cfd = iter.first;
+      cfd->Ref();
+      cfd->SetFlushReason(flush_reason);
+    }
+    ++unscheduled_flushes_;
+    flush_queue_.push_back(flush_req);
   }
-  ++unscheduled_flushes_;
-  flush_queue_.push_back(flush_req);
 }
 
 void DBImpl::SchedulePendingCompaction(ColumnFamilyData* cfd) {
+  mutex_.AssertHeld();
   if (!cfd->queued_for_compaction() && cfd->NeedsCompaction()) {
     AddToCompactionQueue(cfd);
     ++unscheduled_compactions_;
@@ -2316,7 +2465,16 @@ void DBImpl::BGWorkPurge(void* db) {
 }
 
 void DBImpl::UnscheduleCompactionCallback(void* arg) {
-  CompactionArg ca = *(reinterpret_cast<CompactionArg*>(arg));
+  CompactionArg* ca_ptr = reinterpret_cast<CompactionArg*>(arg);
+  Env::Priority compaction_pri = ca_ptr->compaction_pri_;
+  if (Env::Priority::BOTTOM == compaction_pri) {
+    // Decrement bg_bottom_compaction_scheduled_ if priority is BOTTOM
+    ca_ptr->db->bg_bottom_compaction_scheduled_--;
+  } else if (Env::Priority::LOW == compaction_pri) {
+    // Decrement bg_compaction_scheduled_ if priority is LOW
+    ca_ptr->db->bg_compaction_scheduled_--;
+  }
+  CompactionArg ca = *(ca_ptr);
   delete reinterpret_cast<CompactionArg*>(arg);
   if (ca.prepicked_compaction != nullptr) {
     if (ca.prepicked_compaction->compaction != nullptr) {
@@ -2328,6 +2486,14 @@ void DBImpl::UnscheduleCompactionCallback(void* arg) {
 }
 
 void DBImpl::UnscheduleFlushCallback(void* arg) {
+  // Decrement bg_flush_scheduled_ in flush callback
+  reinterpret_cast<FlushThreadArg*>(arg)->db_->bg_flush_scheduled_--;
+  Env::Priority flush_pri = reinterpret_cast<FlushThreadArg*>(arg)->thread_pri_;
+  if (Env::Priority::LOW == flush_pri) {
+    TEST_SYNC_POINT("DBImpl::UnscheduleLowFlushCallback");
+  } else if (Env::Priority::HIGH == flush_pri) {
+    TEST_SYNC_POINT("DBImpl::UnscheduleHighFlushCallback");
+  }
   delete reinterpret_cast<FlushThreadArg*>(arg);
   TEST_SYNC_POINT("DBImpl::UnscheduleFlushCallback");
 }
@@ -2420,6 +2586,8 @@ void DBImpl::BackgroundCallFlush(Env::Priority thread_pri) {
 
   LogBuffer log_buffer(InfoLogLevel::INFO_LEVEL,
                        immutable_db_options_.info_log.get());
+  TEST_SYNC_POINT("DBImpl::BackgroundCallFlush:Start:1");
+  TEST_SYNC_POINT("DBImpl::BackgroundCallFlush:Start:2");
   {
     InstrumentedMutexLock l(&mutex_);
     assert(bg_flush_scheduled_);
@@ -2448,7 +2616,7 @@ void DBImpl::BackgroundCallFlush(Env::Priority thread_pri) {
                       s.ToString().c_str(), error_cnt);
       log_buffer.FlushBufferToLog();
       LogFlush(immutable_db_options_.info_log);
-      env_->SleepForMicroseconds(1000000);
+      immutable_db_options_.clock->SleepForMicroseconds(1000000);
       mutex_.Lock();
     }
 
@@ -2521,7 +2689,8 @@ void DBImpl::BackgroundCallCompaction(PrepickedCompaction* prepicked_compaction,
     if (s.IsBusy()) {
       bg_cv_.SignalAll();  // In case a waiter can proceed despite the error
       mutex_.Unlock();
-      env_->SleepForMicroseconds(10000);  // prevent hot loop
+      immutable_db_options_.clock->SleepForMicroseconds(
+          10000);  // prevent hot loop
       mutex_.Lock();
     } else if (!s.ok() && !s.IsShutdownInProgress() &&
                !s.IsManualCompactionPaused() && !s.IsColumnFamilyDropped()) {
@@ -2539,7 +2708,7 @@ void DBImpl::BackgroundCallCompaction(PrepickedCompaction* prepicked_compaction,
                       "Accumulated background error counts: %" PRIu64,
                       s.ToString().c_str(), error_cnt);
       LogFlush(immutable_db_options_.info_log);
-      env_->SleepForMicroseconds(1000000);
+      immutable_db_options_.clock->SleepForMicroseconds(1000000);
       mutex_.Lock();
     } else if (s.IsManualCompactionPaused()) {
       ManualCompactionState* m = prepicked_compaction->manual_compaction_state;
@@ -2555,7 +2724,8 @@ void DBImpl::BackgroundCallCompaction(PrepickedCompaction* prepicked_compaction,
     // failure). Thus, we force full scan in FindObsoleteFiles()
     FindObsoleteFiles(&job_context, !s.ok() && !s.IsShutdownInProgress() &&
                                         !s.IsManualCompactionPaused() &&
-                                        !s.IsColumnFamilyDropped());
+                                        !s.IsColumnFamilyDropped() &&
+                                        !s.IsBusy());
     TEST_SYNC_POINT("DBImpl::BackgroundCallCompaction:FoundObsoleteFiles");
 
     // delete unnecessary files if any, this is done outside the mutex
@@ -2928,6 +3098,7 @@ Status DBImpl::BackgroundCompaction(bool* made_progress,
     TEST_SYNC_POINT("DBImpl::BackgroundCompaction:ForwardToBottomPriPool");
     CompactionArg* ca = new CompactionArg;
     ca->db = this;
+    ca->compaction_pri_ = Env::Priority::BOTTOM;
     ca->prepicked_compaction = new PrepickedCompaction;
     ca->prepicked_compaction->compaction = c.release();
     ca->prepicked_compaction->manual_compaction_state = nullptr;
@@ -2953,14 +3124,15 @@ Status DBImpl::BackgroundCompaction(bool* made_progress,
         job_context->job_id, c.get(), immutable_db_options_,
         file_options_for_compaction_, versions_.get(), &shutting_down_,
         preserve_deletes_seqnum_.load(), log_buffer, directories_.GetDbDir(),
-        GetDataDir(c->column_family_data(), c->output_path_id()), stats_,
-        &mutex_, &error_handler_, snapshot_seqs,
-        earliest_write_conflict_snapshot, snapshot_checker, table_cache_,
-        &event_logger_, c->mutable_cf_options()->paranoid_file_checks,
+        GetDataDir(c->column_family_data(), c->output_path_id()),
+        GetDataDir(c->column_family_data(), 0), stats_, &mutex_,
+        &error_handler_, snapshot_seqs, earliest_write_conflict_snapshot,
+        snapshot_checker, table_cache_, &event_logger_,
+        c->mutable_cf_options()->paranoid_file_checks,
         c->mutable_cf_options()->report_bg_io_stats, dbname_,
         &compaction_job_stats, thread_pri, io_tracer_,
         is_manual ? &manual_compaction_paused_ : nullptr, db_id_,
-        db_session_id_);
+        db_session_id_, c->column_family_data()->GetFullHistoryTsLow());
     compaction_job.Prepare();
 
     NotifyOnCompactionBegin(c->column_family_data(), c.get(), status,
@@ -3025,10 +3197,9 @@ Status DBImpl::BackgroundCompaction(bool* made_progress,
       auto err_reason = versions_->io_status().ok()
                             ? BackgroundErrorReason::kCompaction
                             : BackgroundErrorReason::kManifestWrite;
-      error_handler_.SetBGError(io_s, err_reason).PermitUncheckedError();
+      error_handler_.SetBGError(io_s, err_reason);
     } else {
-      error_handler_.SetBGError(status, BackgroundErrorReason::kCompaction)
-          .PermitUncheckedError();
+      error_handler_.SetBGError(status, BackgroundErrorReason::kCompaction);
     }
     if (c != nullptr && !is_manual && !error_handler_.IsBGWorkStopped()) {
       // Put this cfd back in the compaction queue so we can retry after some

db/db_impl/db_impl_debug.cc

@@ -22,12 +22,13 @@ uint64_t DBImpl::TEST_GetLevel0TotalSize() {
   return default_cf_handle_->cfd()->current()->storage_info()->NumLevelBytes(0);
 }
 
-void DBImpl::TEST_SwitchWAL() {
+Status DBImpl::TEST_SwitchWAL() {
   WriteContext write_context;
   InstrumentedMutexLock l(&mutex_);
   void* writer = TEST_BeginWrite();
-  SwitchWAL(&write_context);
+  auto s = SwitchWAL(&write_context);
   TEST_EndWrite(writer);
+  return s;
 }
 
 bool DBImpl::TEST_WALBufferIsEmpty(bool lock) {
@@ -170,12 +171,17 @@ 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_.GetBGError().ok())) {
     bg_cv_.Wait();
   }
   return error_handler_.GetBGError();
 }
 
+Status DBImpl::TEST_GetBGError() {
+  InstrumentedMutexLock l(&mutex_);
+  return error_handler_.GetBGError();
+}
+
 void DBImpl::TEST_LockMutex() { mutex_.Lock(); }
 
 void DBImpl::TEST_UnlockMutex() { mutex_.Unlock(); }

db/db_impl/db_impl_files.cc

@@ -6,16 +6,17 @@
 // 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/db_impl/db_impl.h"
-
 #include <cinttypes>
 #include <set>
 #include <unordered_set>
+
+#include "db/db_impl/db_impl.h"
 #include "db/event_helpers.h"
 #include "db/memtable_list.h"
 #include "file/file_util.h"
 #include "file/filename.h"
 #include "file/sst_file_manager_impl.h"
+#include "port/port.h"
 #include "util/autovector.h"
 
 namespace ROCKSDB_NAMESPACE {
@@ -119,7 +120,7 @@ void DBImpl::FindObsoleteFiles(JobContext* job_context, bool force,
              mutable_db_options_.delete_obsolete_files_period_micros == 0) {
     doing_the_full_scan = true;
   } else {
-    const uint64_t now_micros = env_->NowMicros();
+    const uint64_t now_micros = immutable_db_options_.clock->NowMicros();
     if ((delete_obsolete_files_last_run_ +
          mutable_db_options_.delete_obsolete_files_period_micros) <
         now_micros) {
@@ -190,7 +191,8 @@ void DBImpl::FindObsoleteFiles(JobContext* job_context, bool force,
       // set of all files in the directory. We'll exclude files that are still
       // alive in the subsequent processings.
       std::vector<std::string> files;
-      env_->GetChildren(path, &files).PermitUncheckedError();  // Ignore errors
+      Status s = env_->GetChildren(path, &files);
+      s.PermitUncheckedError();  // TODO: What should we do on error?
       for (const std::string& file : files) {
         uint64_t number;
         FileType type;
@@ -206,7 +208,8 @@ void DBImpl::FindObsoleteFiles(JobContext* job_context, bool force,
           continue;
         }
 
-        // TODO(icanadi) clean up this mess to avoid having one-off "/" prefixes
+        // TODO(icanadi) clean up this mess to avoid having one-off "/"
+        // prefixes
         job_context->full_scan_candidate_files.emplace_back("/" + file, path);
       }
     }
@@ -214,9 +217,8 @@ void DBImpl::FindObsoleteFiles(JobContext* job_context, bool force,
     // Add log files in wal_dir
     if (immutable_db_options_.wal_dir != dbname_) {
       std::vector<std::string> log_files;
-      env_->GetChildren(immutable_db_options_.wal_dir,
-                        &log_files)
-          .PermitUncheckedError();  // Ignore errors
+      Status s = env_->GetChildren(immutable_db_options_.wal_dir, &log_files);
+      s.PermitUncheckedError();  // TODO: What should we do on error?
       for (const std::string& log_file : log_files) {
         job_context->full_scan_candidate_files.emplace_back(
             log_file, immutable_db_options_.wal_dir);
@@ -226,9 +228,9 @@ void DBImpl::FindObsoleteFiles(JobContext* job_context, bool force,
     if (!immutable_db_options_.db_log_dir.empty() &&
         immutable_db_options_.db_log_dir != dbname_) {
       std::vector<std::string> info_log_files;
-      // Ignore errors
-      env_->GetChildren(immutable_db_options_.db_log_dir, &info_log_files)
-          .PermitUncheckedError();
+      Status s =
+          env_->GetChildren(immutable_db_options_.db_log_dir, &info_log_files);
+      s.PermitUncheckedError();  // TODO: What should we do on error?
       for (std::string& log_file : info_log_files) {
         job_context->full_scan_candidate_files.emplace_back(
             log_file, immutable_db_options_.db_log_dir);
@@ -319,7 +321,7 @@ void DBImpl::DeleteObsoleteFileImpl(int job_id, const std::string& fname,
                            const_cast<std::string*>(&fname));
 
   Status file_deletion_status;
-  if (type == kTableFile || type == kBlobFile || type == kLogFile) {
+  if (type == kTableFile || type == kBlobFile || type == kWalFile) {
     file_deletion_status =
         DeleteDBFile(&immutable_db_options_, fname, path_to_sync,
                      /*force_bg=*/false, /*force_fg=*/!wal_in_db_path_);
@@ -466,7 +468,7 @@ void DBImpl::PurgeObsoleteFiles(JobContext& state, bool schedule_only) {
 
     bool keep = true;
     switch (type) {
-      case kLogFile:
+      case kWalFile:
         keep = ((number >= state.log_number) ||
                 (number == state.prev_log_number) ||
                 (log_recycle_files_set.find(number) !=
@@ -546,7 +548,7 @@ void DBImpl::PurgeObsoleteFiles(JobContext& state, bool schedule_only) {
       dir_to_sync = candidate_file.file_path;
     } else {
       dir_to_sync =
-          (type == kLogFile) ? immutable_db_options_.wal_dir : dbname_;
+          (type == kWalFile) ? immutable_db_options_.wal_dir : dbname_;
       fname = dir_to_sync +
               ((!dir_to_sync.empty() && dir_to_sync.back() == '/') ||
                        (!to_delete.empty() && to_delete.front() == '/')
@@ -556,8 +558,8 @@ void DBImpl::PurgeObsoleteFiles(JobContext& state, bool schedule_only) {
     }
 
 #ifndef ROCKSDB_LITE
-    if (type == kLogFile && (immutable_db_options_.wal_ttl_seconds > 0 ||
-                             immutable_db_options_.wal_size_limit_mb > 0)) {
+    if (type == kWalFile && (immutable_db_options_.WAL_ttl_seconds > 0 ||
+                             immutable_db_options_.WAL_size_limit_MB > 0)) {
       wal_manager_.ArchiveWALFile(fname, number);
       continue;
     }
@@ -658,13 +660,15 @@ uint64_t FindMinPrepLogReferencedByMemTable(
 
   // we must look through the memtables for two phase transactions
   // that have been committed but not yet flushed
+  std::unordered_set<MemTable*> memtables_to_flush_set(
+      memtables_to_flush.begin(), memtables_to_flush.end());
   for (auto loop_cfd : *vset->GetColumnFamilySet()) {
     if (loop_cfd->IsDropped() || loop_cfd == cfd_to_flush) {
       continue;
     }
 
     auto log = loop_cfd->imm()->PrecomputeMinLogContainingPrepSection(
-        memtables_to_flush);
+        &memtables_to_flush_set);
 
     if (log > 0 && (min_log == 0 || log < min_log)) {
       min_log = log;
@@ -680,16 +684,41 @@ uint64_t FindMinPrepLogReferencedByMemTable(
   return min_log;
 }
 
-uint64_t PrecomputeMinLogNumberToKeep(
+uint64_t FindMinPrepLogReferencedByMemTable(
+    VersionSet* vset, const autovector<ColumnFamilyData*>& cfds_to_flush,
+    const autovector<const autovector<MemTable*>*>& memtables_to_flush) {
+  uint64_t min_log = 0;
+
+  std::unordered_set<ColumnFamilyData*> cfds_to_flush_set(cfds_to_flush.begin(),
+                                                          cfds_to_flush.end());
+  std::unordered_set<MemTable*> memtables_to_flush_set;
+  for (const autovector<MemTable*>* memtables : memtables_to_flush) {
+    memtables_to_flush_set.insert(memtables->begin(), memtables->end());
+  }
+  for (auto loop_cfd : *vset->GetColumnFamilySet()) {
+    if (loop_cfd->IsDropped() || cfds_to_flush_set.count(loop_cfd)) {
+      continue;
+    }
+
+    auto log = loop_cfd->imm()->PrecomputeMinLogContainingPrepSection(
+        &memtables_to_flush_set);
+    if (log > 0 && (min_log == 0 || log < min_log)) {
+      min_log = log;
+    }
+
+    log = loop_cfd->mem()->GetMinLogContainingPrepSection();
+    if (log > 0 && (min_log == 0 || log < min_log)) {
+      min_log = log;
+    }
+  }
+
+  return min_log;
+}
+
+uint64_t PrecomputeMinLogNumberToKeepNon2PC(
     VersionSet* vset, const ColumnFamilyData& cfd_to_flush,
-    autovector<VersionEdit*> edit_list,
-    const autovector<MemTable*>& memtables_to_flush,
-    LogsWithPrepTracker* prep_tracker) {
+    const autovector<VersionEdit*>& edit_list) {
   assert(vset != nullptr);
-  assert(prep_tracker != nullptr);
-  // Calculate updated min_log_number_to_keep
-  // Since the function should only be called in 2pc mode, log number in
-  // the version edit should be sufficient.
 
   // Precompute the min log number containing unflushed data for the column
   // family being flushed (`cfd_to_flush`).
@@ -713,6 +742,58 @@ uint64_t PrecomputeMinLogNumberToKeep(
     min_log_number_to_keep =
         std::min(cf_min_log_number_to_keep, min_log_number_to_keep);
   }
+  return min_log_number_to_keep;
+}
+
+uint64_t PrecomputeMinLogNumberToKeepNon2PC(
+    VersionSet* vset, const autovector<ColumnFamilyData*>& cfds_to_flush,
+    const autovector<autovector<VersionEdit*>>& edit_lists) {
+  assert(vset != nullptr);
+  assert(!cfds_to_flush.empty());
+  assert(cfds_to_flush.size() == edit_lists.size());
+
+  uint64_t min_log_number_to_keep = port::kMaxUint64;
+  for (const auto& edit_list : edit_lists) {
+    uint64_t log = 0;
+    for (const auto& e : edit_list) {
+      if (e->HasLogNumber()) {
+        log = std::max(log, e->GetLogNumber());
+      }
+    }
+    if (log != 0) {
+      min_log_number_to_keep = std::min(min_log_number_to_keep, log);
+    }
+  }
+  if (min_log_number_to_keep == port::kMaxUint64) {
+    min_log_number_to_keep = cfds_to_flush[0]->GetLogNumber();
+    for (size_t i = 1; i < cfds_to_flush.size(); i++) {
+      min_log_number_to_keep =
+          std::min(min_log_number_to_keep, cfds_to_flush[i]->GetLogNumber());
+    }
+  }
+
+  std::unordered_set<const ColumnFamilyData*> flushed_cfds(
+      cfds_to_flush.begin(), cfds_to_flush.end());
+  min_log_number_to_keep =
+      std::min(min_log_number_to_keep,
+               vset->PreComputeMinLogNumberWithUnflushedData(flushed_cfds));
+
+  return min_log_number_to_keep;
+}
+
+uint64_t PrecomputeMinLogNumberToKeep2PC(
+    VersionSet* vset, const ColumnFamilyData& cfd_to_flush,
+    const autovector<VersionEdit*>& edit_list,
+    const autovector<MemTable*>& memtables_to_flush,
+    LogsWithPrepTracker* prep_tracker) {
+  assert(vset != nullptr);
+  assert(prep_tracker != nullptr);
+  // Calculate updated min_log_number_to_keep
+  // Since the function should only be called in 2pc mode, log number in
+  // the version edit should be sufficient.
+
+  uint64_t min_log_number_to_keep =
+      PrecomputeMinLogNumberToKeepNon2PC(vset, cfd_to_flush, edit_list);
 
   // if are 2pc we must consider logs containing prepared
   // sections of outstanding transactions.
@@ -741,7 +822,81 @@ uint64_t PrecomputeMinLogNumberToKeep(
   return min_log_number_to_keep;
 }
 
-Status DBImpl::FinishBestEffortsRecovery() {
+uint64_t PrecomputeMinLogNumberToKeep2PC(
+    VersionSet* vset, const autovector<ColumnFamilyData*>& cfds_to_flush,
+    const autovector<autovector<VersionEdit*>>& edit_lists,
+    const autovector<const autovector<MemTable*>*>& memtables_to_flush,
+    LogsWithPrepTracker* prep_tracker) {
+  assert(vset != nullptr);
+  assert(prep_tracker != nullptr);
+  assert(cfds_to_flush.size() == edit_lists.size());
+  assert(cfds_to_flush.size() == memtables_to_flush.size());
+
+  uint64_t min_log_number_to_keep =
+      PrecomputeMinLogNumberToKeepNon2PC(vset, cfds_to_flush, edit_lists);
+
+  uint64_t min_log_in_prep_heap =
+      prep_tracker->FindMinLogContainingOutstandingPrep();
+
+  if (min_log_in_prep_heap != 0 &&
+      min_log_in_prep_heap < min_log_number_to_keep) {
+    min_log_number_to_keep = min_log_in_prep_heap;
+  }
+
+  uint64_t min_log_refed_by_mem = FindMinPrepLogReferencedByMemTable(
+      vset, cfds_to_flush, memtables_to_flush);
+
+  if (min_log_refed_by_mem != 0 &&
+      min_log_refed_by_mem < min_log_number_to_keep) {
+    min_log_number_to_keep = min_log_refed_by_mem;
+  }
+
+  return min_log_number_to_keep;
+}
+
+Status DBImpl::SetDBId(bool read_only) {
+  Status s;
+  // Happens when immutable_db_options_.write_dbid_to_manifest is set to true
+  // the very first time.
+  if (db_id_.empty()) {
+    // Check for the IDENTITY file and create it if not there.
+    s = fs_->FileExists(IdentityFileName(dbname_), IOOptions(), nullptr);
+    // Typically Identity file is created in NewDB() and for some reason if
+    // it is no longer available then at this point DB ID is not in Identity
+    // file or Manifest.
+    if (s.IsNotFound()) {
+      // Create a new DB ID, saving to file only if allowed
+      if (read_only) {
+        db_id_ = env_->GenerateUniqueId();
+        return Status::OK();
+      } else {
+        s = SetIdentityFile(env_, dbname_);
+        if (!s.ok()) {
+          return s;
+        }
+      }
+    } else if (!s.ok()) {
+      assert(s.IsIOError());
+      return s;
+    }
+    s = GetDbIdentityFromIdentityFile(&db_id_);
+    if (immutable_db_options_.write_dbid_to_manifest && s.ok()) {
+      VersionEdit edit;
+      edit.SetDBId(db_id_);
+      Options options;
+      MutableCFOptions mutable_cf_options(options);
+      versions_->db_id_ = db_id_;
+      s = versions_->LogAndApply(versions_->GetColumnFamilySet()->GetDefault(),
+                                 mutable_cf_options, &edit, &mutex_, nullptr,
+                                 /* new_descriptor_log */ false);
+    }
+  } else if (!read_only) {
+    s = SetIdentityFile(env_, dbname_, db_id_);
+  }
+  return s;
+}
+
+Status DBImpl::DeleteUnreferencedSstFiles() {
   mutex_.AssertHeld();
   std::vector<std::string> paths;
   paths.push_back(NormalizePath(dbname_ + std::string(1, kFilePathSeparator)));
@@ -788,7 +943,7 @@ Status DBImpl::FinishBestEffortsRecovery() {
     return s;
   }
 
-  if (largest_file_number > next_file_number) {
+  if (largest_file_number >= next_file_number) {
     versions_->next_file_number_.store(largest_file_number + 1);
   }
 
@@ -797,8 +952,6 @@ Status DBImpl::FinishBestEffortsRecovery() {
   assert(versions_->GetColumnFamilySet());
   ColumnFamilyData* default_cfd = versions_->GetColumnFamilySet()->GetDefault();
   assert(default_cfd);
-  // Even if new_descriptor_log is false, we will still switch to a new
-  // MANIFEST and update CURRENT file, since this is in recovery.
   s = versions_->LogAndApply(
       default_cfd, *default_cfd->GetLatestMutableCFOptions(), &edit, &mutex_,
       directories_.GetDbDir(), /*new_descriptor_log*/ false);

db/db_impl/db_impl_open.cc

@@ -24,15 +24,17 @@
 #include "util/rate_limiter.h"
 
 namespace ROCKSDB_NAMESPACE {
-Options SanitizeOptions(const std::string& dbname, const Options& src) {
-  auto db_options = SanitizeOptions(dbname, DBOptions(src));
+Options SanitizeOptions(const std::string& dbname, const Options& src,
+                        bool read_only) {
+  auto db_options = SanitizeOptions(dbname, DBOptions(src), read_only);
   ImmutableDBOptions immutable_db_options(db_options);
   auto cf_options =
       SanitizeOptions(immutable_db_options, ColumnFamilyOptions(src));
   return Options(db_options, cf_options);
 }
 
-DBOptions SanitizeOptions(const std::string& dbname, const DBOptions& src) {
+DBOptions SanitizeOptions(const std::string& dbname, const DBOptions& src,
+                          bool read_only) {
   DBOptions result(src);
 
   if (result.env == nullptr) {
@@ -50,7 +52,7 @@ DBOptions SanitizeOptions(const std::string& dbname, const DBOptions& src) {
                              &result.max_open_files);
   }
 
-  if (result.info_log == nullptr) {
+  if (result.info_log == nullptr && !read_only) {
     Status s = CreateLoggerFromOptions(dbname, result, &result.info_log);
     if (!s.ok()) {
       // No place suitable for logging
@@ -147,7 +149,8 @@ DBOptions SanitizeOptions(const std::string& dbname, const DBOptions& src) {
     // DeleteScheduler::CleanupDirectory on the same dir later, it will be
     // safe
     std::vector<std::string> filenames;
-    result.env->GetChildren(result.wal_dir, &filenames).PermitUncheckedError();
+    Status s = result.env->GetChildren(result.wal_dir, &filenames);
+    s.PermitUncheckedError();  //**TODO: What to do on error?
     for (std::string& filename : filenames) {
       if (filename.find(".log.trash", filename.length() -
                                           std::string(".log.trash").length()) !=
@@ -163,7 +166,8 @@ DBOptions SanitizeOptions(const std::string& dbname, const DBOptions& src) {
   // was not used)
   auto sfm = static_cast<SstFileManagerImpl*>(result.sst_file_manager.get());
   for (size_t i = 0; i < result.db_paths.size(); i++) {
-    DeleteScheduler::CleanupDirectory(result.env, sfm, result.db_paths[i].path);
+    DeleteScheduler::CleanupDirectory(result.env, sfm, result.db_paths[i].path)
+        .PermitUncheckedError();
   }
 
   // Create a default SstFileManager for purposes of tracking compaction size
@@ -173,7 +177,7 @@ DBOptions SanitizeOptions(const std::string& dbname, const DBOptions& src) {
         NewSstFileManager(result.env, result.info_log));
     result.sst_file_manager = sst_file_manager;
   }
-#endif
+#endif  // !ROCKSDB_LITE
 
   if (!result.paranoid_checks) {
     result.skip_checking_sst_file_sizes_on_db_open = true;
@@ -281,23 +285,28 @@ Status DBImpl::NewDB(std::vector<std::string>* new_filenames) {
   ROCKS_LOG_INFO(immutable_db_options_.info_log, "Creating manifest 1 \n");
   const std::string manifest = DescriptorFileName(dbname_, 1);
   {
+    if (fs_->FileExists(manifest, IOOptions(), nullptr).ok()) {
+      fs_->DeleteFile(manifest, IOOptions(), nullptr).PermitUncheckedError();
+    }
     std::unique_ptr<FSWritableFile> file;
     FileOptions file_options = fs_->OptimizeForManifestWrite(file_options_);
     s = NewWritableFile(fs_.get(), manifest, &file, file_options);
     if (!s.ok()) {
       return s;
     }
+    FileTypeSet tmp_set = immutable_db_options_.checksum_handoff_file_types;
     file->SetPreallocationBlockSize(
         immutable_db_options_.manifest_preallocation_size);
     std::unique_ptr<WritableFileWriter> file_writer(new WritableFileWriter(
-        std::move(file), manifest, file_options, env_, io_tracer_,
-        nullptr /* stats */, immutable_db_options_.listeners));
+        std::move(file), manifest, file_options, immutable_db_options_.clock,
+        io_tracer_, nullptr /* stats */, immutable_db_options_.listeners,
+        nullptr, tmp_set.Contains(FileType::kDescriptorFile)));
     log::Writer log(std::move(file_writer), 0, false);
     std::string record;
     new_db.EncodeTo(&record);
     s = log.AddRecord(record);
     if (s.ok()) {
-      s = SyncManifest(env_, &immutable_db_options_, log.file());
+      s = SyncManifest(&immutable_db_options_, log.file());
     }
   }
   if (s.ok()) {
@@ -308,7 +317,7 @@ Status DBImpl::NewDB(std::vector<std::string>* new_filenames) {
           manifest.substr(manifest.find_last_of("/\\") + 1));
     }
   } else {
-    fs_->DeleteFile(manifest, IOOptions(), nullptr);
+    fs_->DeleteFile(manifest, IOOptions(), nullptr).PermitUncheckedError();
   }
   return s;
 }
@@ -401,7 +410,7 @@ Status DBImpl::Recover(
       }
       for (const std::string& file : files_in_dbname) {
         uint64_t number = 0;
-        FileType type = kLogFile;  // initialize
+        FileType type = kWalFile;  // initialize
         if (ParseFileName(file, &number, &type) && type == kDescriptorFile) {
           // Found MANIFEST (descriptor log), thus best-efforts recovery does
           // not have to treat the db as empty.
@@ -479,42 +488,14 @@ Status DBImpl::Recover(
       // TryRecover may delete previous column_family_set_.
       column_family_memtables_.reset(
           new ColumnFamilyMemTablesImpl(versions_->GetColumnFamilySet()));
-      s = FinishBestEffortsRecovery();
     }
   }
   if (!s.ok()) {
     return s;
   }
-  // Happens when immutable_db_options_.write_dbid_to_manifest is set to true
-  // the very first time.
-  if (db_id_.empty()) {
-    // Check for the IDENTITY file and create it if not there.
-    s = fs_->FileExists(IdentityFileName(dbname_), IOOptions(), nullptr);
-    // Typically Identity file is created in NewDB() and for some reason if
-    // it is no longer available then at this point DB ID is not in Identity
-    // file or Manifest.
-    if (s.IsNotFound()) {
-      s = SetIdentityFile(env_, dbname_);
-      if (!s.ok()) {
-        return s;
-      }
-    } else if (!s.ok()) {
-      assert(s.IsIOError());
-      return s;
-    }
-    s = GetDbIdentityFromIdentityFile(&db_id_);
-    if (immutable_db_options_.write_dbid_to_manifest && s.ok()) {
-      VersionEdit edit;
-      edit.SetDBId(db_id_);
-      Options options;
-      MutableCFOptions mutable_cf_options(options);
-      versions_->db_id_ = db_id_;
-      s = versions_->LogAndApply(versions_->GetColumnFamilySet()->GetDefault(),
-                             mutable_cf_options, &edit, &mutex_, nullptr,
-                             false);
-    }
-  } else {
-    s = SetIdentityFile(env_, dbname_, db_id_);
+  s = SetDBId(read_only);
+  if (s.ok() && !read_only) {
+    s = DeleteUnreferencedSstFiles();
   }
 
   if (immutable_db_options_.paranoid_checks && s.ok()) {
@@ -536,7 +517,7 @@ Status DBImpl::Recover(
 
   std::vector<std::string> files_in_wal_dir;
   if (s.ok()) {
-    // Initial max_total_in_memory_state_ before recovery logs. Log recovery
+    // Initial max_total_in_memory_state_ before recovery wals. Log recovery
     // may check this value to decide whether to flush.
     max_total_in_memory_state_ = 0;
     for (auto cfd : *versions_->GetColumnFamilySet()) {
@@ -571,32 +552,53 @@ Status DBImpl::Recover(
       return s;
     }
 
-    std::vector<uint64_t> logs;
+    std::unordered_map<uint64_t, std::string> wal_files;
     for (const auto& file : files_in_wal_dir) {
       uint64_t number;
       FileType type;
-      if (ParseFileName(file, &number, &type) && type == kLogFile) {
+      if (ParseFileName(file, &number, &type) && type == kWalFile) {
         if (is_new_db) {
           return Status::Corruption(
               "While creating a new Db, wal_dir contains "
               "existing log file: ",
               file);
         } else {
-          logs.push_back(number);
+          wal_files[number] =
+              LogFileName(immutable_db_options_.wal_dir, number);
         }
       }
     }
 
-    if (logs.size() > 0) {
+    if (immutable_db_options_.track_and_verify_wals_in_manifest) {
+      if (!immutable_db_options_.best_efforts_recovery) {
+        // Verify WALs in MANIFEST.
+        s = versions_->GetWalSet().CheckWals(env_, wal_files);
+      }  // else since best effort recovery does not recover from WALs, no need
+         // to check WALs.
+    } else if (!versions_->GetWalSet().GetWals().empty()) {
+      // Tracking is disabled, clear previously tracked WALs from MANIFEST,
+      // otherwise, in the future, if WAL tracking is enabled again,
+      // since the WALs deleted when WAL tracking is disabled are not persisted
+      // into MANIFEST, WAL check may fail.
+      VersionEdit edit;
+      WalNumber max_wal_number =
+          versions_->GetWalSet().GetWals().rbegin()->first;
+      edit.DeleteWalsBefore(max_wal_number + 1);
+      s = versions_->LogAndApplyToDefaultColumnFamily(&edit, &mutex_);
+    }
+    if (!s.ok()) {
+      return s;
+    }
+
+    if (!wal_files.empty()) {
       if (error_if_wal_file_exists) {
         return Status::Corruption(
             "The db was opened in readonly mode with error_if_wal_file_exists"
             "flag but a WAL file already exists");
       } else if (error_if_data_exists_in_wals) {
-        for (auto& log : logs) {
-          std::string fname = LogFileName(immutable_db_options_.wal_dir, log);
+        for (auto& wal_file : wal_files) {
           uint64_t bytes;
-          s = env_->GetFileSize(fname, &bytes);
+          s = env_->GetFileSize(wal_file.second, &bytes);
           if (s.ok()) {
             if (bytes > 0) {
               return Status::Corruption(
@@ -608,13 +610,19 @@ Status DBImpl::Recover(
       }
     }
 
-    if (!logs.empty()) {
-      // Recover in the order in which the logs were generated
-      std::sort(logs.begin(), logs.end());
-      bool corrupted_log_found = false;
-      s = RecoverLogFiles(logs, &next_sequence, read_only,
-                          &corrupted_log_found);
-      if (corrupted_log_found && recovered_seq != nullptr) {
+    if (!wal_files.empty()) {
+      // Recover in the order in which the wals were generated
+      std::vector<uint64_t> wals;
+      wals.reserve(wal_files.size());
+      for (const auto& wal_file : wal_files) {
+        wals.push_back(wal_file.first);
+      }
+      std::sort(wals.begin(), wals.end());
+
+      bool corrupted_wal_found = false;
+      s = RecoverLogFiles(wals, &next_sequence, read_only,
+                          &corrupted_wal_found);
+      if (corrupted_wal_found && recovered_seq != nullptr) {
         *recovered_seq = next_sequence;
       }
       if (!s.ok()) {
@@ -767,10 +775,10 @@ Status DBImpl::InitPersistStatsColumnFamily() {
   return s;
 }
 
-// REQUIRES: log_numbers are sorted in ascending order
-Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
+// REQUIRES: wal_numbers are sorted in ascending order
+Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& wal_numbers,
                                SequenceNumber* next_sequence, bool read_only,
-                               bool* corrupted_log_found) {
+                               bool* corrupted_wal_found) {
   struct LogReporter : public log::Reader::Reporter {
     Env* env;
     Logger* info_log;
@@ -800,10 +808,10 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
     auto stream = event_logger_.Log();
     stream << "job" << job_id << "event"
            << "recovery_started";
-    stream << "log_files";
+    stream << "wal_files";
     stream.StartArray();
-    for (auto log_number : log_numbers) {
-      stream << log_number;
+    for (auto wal_number : wal_numbers) {
+      stream << wal_number;
     }
     stream.EndArray();
   }
@@ -826,25 +834,25 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
   bool stop_replay_by_wal_filter = false;
   bool stop_replay_for_corruption = false;
   bool flushed = false;
-  uint64_t corrupted_log_number = kMaxSequenceNumber;
-  uint64_t min_log_number = MinLogNumberToKeep();
-  for (auto log_number : log_numbers) {
-    if (log_number < min_log_number) {
+  uint64_t corrupted_wal_number = kMaxSequenceNumber;
+  uint64_t min_wal_number = MinLogNumberToKeep();
+  for (auto wal_number : wal_numbers) {
+    if (wal_number < min_wal_number) {
       ROCKS_LOG_INFO(immutable_db_options_.info_log,
                      "Skipping log #%" PRIu64
                      " since it is older than min log to keep #%" PRIu64,
-                     log_number, min_log_number);
+                     wal_number, min_wal_number);
       continue;
     }
     // The previous incarnation may not have written any MANIFEST
     // records after allocating this log number.  So we manually
     // update the file number allocation counter in VersionSet.
-    versions_->MarkFileNumberUsed(log_number);
+    versions_->MarkFileNumberUsed(wal_number);
     // Open the log file
-    std::string fname = LogFileName(immutable_db_options_.wal_dir, log_number);
+    std::string fname = LogFileName(immutable_db_options_.wal_dir, wal_number);
 
     ROCKS_LOG_INFO(immutable_db_options_.info_log,
-                   "Recovering log #%" PRIu64 " mode %d", log_number,
+                   "Recovering log #%" PRIu64 " mode %d", wal_number,
                    static_cast<int>(immutable_db_options_.wal_recovery_mode));
     auto logFileDropped = [this, &fname]() {
       uint64_t bytes;
@@ -897,7 +905,7 @@ 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*/, wal_number);
 
     // Determine if we should tolerate incomplete records at the tail end of the
     // Read all the records and add to a memtable
@@ -945,7 +953,7 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
 
         WalFilter::WalProcessingOption wal_processing_option =
             immutable_db_options_.wal_filter->LogRecordFound(
-                log_number, fname, batch, &new_batch, &batch_changed);
+                wal_number, fname, batch, &new_batch, &batch_changed);
 
         switch (wal_processing_option) {
           case WalFilter::WalProcessingOption::kContinueProcessing:
@@ -997,7 +1005,7 @@ 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,
+                wal_number,
                 static_cast<int>(immutable_db_options_.wal_recovery_mode),
                 immutable_db_options_.wal_filter->Name(), new_count,
                 original_count);
@@ -1024,7 +1032,7 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
       bool has_valid_writes = false;
       status = WriteBatchInternal::InsertInto(
           &batch, column_family_memtables_.get(), &flush_scheduler_,
-          &trim_history_scheduler_, true, log_number, this,
+          &trim_history_scheduler_, true, wal_number, this,
           false /* concurrent_memtable_writes */, next_sequence,
           &has_valid_writes, seq_per_batch_, batch_per_txn_);
       MaybeIgnoreError(&status);
@@ -1044,7 +1052,7 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
           cfd->UnrefAndTryDelete();
           // If this asserts, it means that InsertInto failed in
           // filtering updates to already-flushed column families
-          assert(cfd->GetLogNumber() <= log_number);
+          assert(cfd->GetLogNumber() <= wal_number);
           auto iter = version_edits.find(cfd->GetID());
           assert(iter != version_edits.end());
           VersionEdit* edit = &iter->second;
@@ -1081,21 +1089,21 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
                           " seq #%" PRIu64
                           ". %s. This likely mean loss of synced WAL, "
                           "thus recovery fails.",
-                          log_number, *next_sequence,
+                          wal_number, *next_sequence,
                           status.ToString().c_str());
           return status;
         }
         // We should ignore the error but not continue replaying
         status = Status::OK();
         stop_replay_for_corruption = true;
-        corrupted_log_number = log_number;
-        if (corrupted_log_found != nullptr) {
-          *corrupted_log_found = true;
+        corrupted_wal_number = wal_number;
+        if (corrupted_wal_found != nullptr) {
+          *corrupted_wal_found = true;
         }
         ROCKS_LOG_INFO(immutable_db_options_.info_log,
                        "Point in time recovered to log #%" PRIu64
                        " seq #%" PRIu64,
-                       log_number, *next_sequence);
+                       wal_number, *next_sequence);
       } else {
         assert(immutable_db_options_.wal_recovery_mode ==
                    WALRecoveryMode::kTolerateCorruptedTailRecords ||
@@ -1121,7 +1129,7 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
   // corruption. This could during PIT recovery when the WAL is corrupted and
   // some (but not all) CFs are flushed
   // Exclude the PIT case where no log is dropped after the corruption point.
-  // This is to cover the case for empty logs after corrupted log, in which we
+  // This is to cover the case for empty wals after corrupted log, in which we
   // don't reset stop_replay_for_corruption.
   if (stop_replay_for_corruption == true &&
       (immutable_db_options_.wal_recovery_mode ==
@@ -1129,11 +1137,29 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
        immutable_db_options_.wal_recovery_mode ==
            WALRecoveryMode::kTolerateCorruptedTailRecords)) {
     for (auto cfd : *versions_->GetColumnFamilySet()) {
-      if (cfd->GetLogNumber() > corrupted_log_number) {
+      // One special case cause cfd->GetLogNumber() > corrupted_wal_number but
+      // the CF is still consistent: If a new column family is created during
+      // the flush and the WAL sync fails at the same time, the new CF points to
+      // the new WAL but the old WAL is curropted. Since the new CF is empty, it
+      // is still consistent. We add the check of CF sst file size to avoid the
+      // false positive alert.
+
+      // Note that, the check of (cfd->GetLiveSstFilesSize() > 0) may leads to
+      // the ignorance of a very rare inconsistency case caused in data
+      // canclation. One CF is empty due to KV deletion. But those operations
+      // are in the WAL. If the WAL is corrupted, the status of this CF might
+      // not be consistent with others. However, the consistency check will be
+      // bypassed due to empty CF.
+      // TODO: a better and complete implementation is needed to ensure strict
+      // consistency check in WAL recovery including hanlding the tailing
+      // issues.
+      if (cfd->GetLogNumber() > corrupted_wal_number &&
+          cfd->GetLiveSstFilesSize() > 0) {
         ROCKS_LOG_ERROR(immutable_db_options_.info_log,
                         "Column family inconsistency: SST file contains data"
                         " beyond the point of corruption.");
-        return Status::Corruption("SST file is ahead of WALs");
+        return Status::Corruption("SST file is ahead of WALs in CF " +
+                                  cfd->GetName());
       }
     }
   }
@@ -1144,16 +1170,16 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
   if (!read_only) {
     // no need to refcount since client still doesn't have access
     // to the DB and can not drop column families while we iterate
-    auto max_log_number = log_numbers.back();
+    const WalNumber max_wal_number = wal_numbers.back();
     for (auto cfd : *versions_->GetColumnFamilySet()) {
       auto iter = version_edits.find(cfd->GetID());
       assert(iter != version_edits.end());
       VersionEdit* edit = &iter->second;
 
-      if (cfd->GetLogNumber() > max_log_number) {
+      if (cfd->GetLogNumber() > max_wal_number) {
         // Column family cfd has already flushed the data
-        // from all logs. Memtable has to be empty because
-        // we filter the updates based on log_number
+        // from all wals. Memtable has to be empty because
+        // we filter the updates based on wal_number
         // (in WriteBatch::InsertInto)
         assert(cfd->mem()->GetFirstSequenceNumber() == 0);
         assert(edit->NumEntries() == 0);
@@ -1185,13 +1211,13 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
       // Update the log number info in the version edit corresponding to this
       // column family. Note that the version edits will be written to MANIFEST
       // together later.
-      // writing log_number in the manifest means that any log file
-      // with number strongly less than (log_number + 1) is already
+      // writing wal_number in the manifest means that any log file
+      // with number strongly less than (wal_number + 1) is already
       // recovered and should be ignored on next reincarnation.
-      // Since we already recovered max_log_number, we want all logs
-      // with numbers `<= max_log_number` (includes this one) to be ignored
+      // Since we already recovered max_wal_number, we want all wals
+      // with numbers `<= max_wal_number` (includes this one) to be ignored
       if (flushed || cfd->mem()->GetFirstSequenceNumber() == 0) {
-        edit->SetLogNumber(max_log_number + 1);
+        edit->SetLogNumber(max_wal_number + 1);
       }
     }
     if (status.ok()) {
@@ -1199,7 +1225,7 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
       // not actually used. that is because VersionSet assumes
       // VersionSet::next_file_number_ always to be strictly greater than any
       // log number
-      versions_->MarkFileNumberUsed(max_log_number + 1);
+      versions_->MarkFileNumberUsed(max_wal_number + 1);
 
       autovector<ColumnFamilyData*> cfds;
       autovector<const MutableCFOptions*> cf_opts;
@@ -1211,6 +1237,14 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
         assert(iter != version_edits.end());
         edit_lists.push_back({&iter->second});
       }
+
+      std::unique_ptr<VersionEdit> wal_deletion;
+      if (immutable_db_options_.track_and_verify_wals_in_manifest) {
+        wal_deletion.reset(new VersionEdit);
+        wal_deletion->DeleteWalsBefore(max_wal_number + 1);
+        edit_lists.back().push_back(wal_deletion.get());
+      }
+
       // write MANIFEST with update
       status = versions_->LogAndApply(cfds, cf_opts, edit_lists, &mutex_,
                                       directories_.GetDbDir(),
@@ -1218,8 +1252,16 @@ Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
     }
   }
 
-  if (status.ok() && data_seen && !flushed) {
-    status = RestoreAliveLogFiles(log_numbers);
+  if (status.ok()) {
+    if (data_seen && !flushed) {
+      status = RestoreAliveLogFiles(wal_numbers);
+    } else {
+      // If there's no data in the WAL, or we flushed all the data, still
+      // truncate the log file. If the process goes into a crash loop before
+      // the file is deleted, the preallocated space will never get freed.
+      GetLogSizeAndMaybeTruncate(wal_numbers.back(), true, nullptr)
+          .PermitUncheckedError();
+    }
   }
 
   event_logger_.Log() << "job" << job_id << "event"
@@ -1228,8 +1270,42 @@ 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()) {
+Status DBImpl::GetLogSizeAndMaybeTruncate(uint64_t wal_number, bool truncate,
+                                          LogFileNumberSize* log_ptr) {
+  LogFileNumberSize log(wal_number);
+  std::string fname = LogFileName(immutable_db_options_.wal_dir, wal_number);
+  Status s;
+  // This gets the appear size of the wals, not including preallocated space.
+  s = env_->GetFileSize(fname, &log.size);
+  if (s.ok() && truncate) {
+    std::unique_ptr<FSWritableFile> last_log;
+    Status truncate_status = fs_->ReopenWritableFile(
+        fname,
+        fs_->OptimizeForLogWrite(
+            file_options_,
+            BuildDBOptions(immutable_db_options_, mutable_db_options_)),
+        &last_log, nullptr);
+    if (truncate_status.ok()) {
+      truncate_status = last_log->Truncate(log.size, IOOptions(), nullptr);
+    }
+    if (truncate_status.ok()) {
+      truncate_status = last_log->Close(IOOptions(), nullptr);
+    }
+    // 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", wal_number,
+                     truncate_status.ToString().c_str());
+    }
+  }
+  if (log_ptr) {
+    *log_ptr = log;
+  }
+  return s;
+}
+
+Status DBImpl::RestoreAliveLogFiles(const std::vector<uint64_t>& wal_numbers) {
+  if (wal_numbers.empty()) {
     return Status::OK();
   }
   Status s;
@@ -1242,40 +1318,18 @@ Status DBImpl::RestoreAliveLogFiles(const std::vector<uint64_t>& log_numbers) {
   // 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);
+  for (auto wal_number : wal_numbers) {
+    // We preallocate space for wals, 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.
+    LogFileNumberSize log;
+    s = GetLogSizeAndMaybeTruncate(
+        wal_number, /*truncate=*/(wal_number == wal_numbers.back()), &log);
     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<FSWritableFile> last_log;
-      Status truncate_status = fs_->ReopenWritableFile(
-          fname,
-          fs_->OptimizeForLogWrite(
-              file_options_,
-              BuildDBOptions(immutable_db_options_, mutable_db_options_)),
-          &last_log, nullptr);
-      if (truncate_status.ok()) {
-        truncate_status = last_log->Truncate(log.size, IOOptions(), nullptr);
-      }
-      if (truncate_status.ok()) {
-        truncate_status = last_log->Close(IOOptions(), nullptr);
-      }
-      // 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();
@@ -1286,7 +1340,7 @@ Status DBImpl::RestoreAliveLogFiles(const std::vector<uint64_t>& log_numbers) {
 Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
                                            MemTable* mem, VersionEdit* edit) {
   mutex_.AssertHeld();
-  const uint64_t start_micros = env_->NowMicros();
+  const uint64_t start_micros = immutable_db_options_.clock->NowMicros();
 
   FileMetaData meta;
   std::vector<BlobFileAddition> blob_file_additions;
@@ -1314,7 +1368,7 @@ Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
         cfd->GetLatestMutableCFOptions()->paranoid_file_checks;
 
     int64_t _current_time = 0;
-    env_->GetCurrentTime(&_current_time)
+    immutable_db_options_.clock->GetCurrentTime(&_current_time)
         .PermitUncheckedError();  // ignore error
     const uint64_t current_time = static_cast<uint64_t>(_current_time);
     meta.oldest_ancester_time = current_time;
@@ -1340,20 +1394,20 @@ Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
 
       IOStatus io_s;
       s = BuildTable(
-          dbname_, versions_.get(), env_, fs_.get(), *cfd->ioptions(),
+          dbname_, versions_.get(), immutable_db_options_, *cfd->ioptions(),
           mutable_cf_options, file_options_for_compaction_, cfd->table_cache(),
           iter.get(), std::move(range_del_iters), &meta, &blob_file_additions,
           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,
           mutable_cf_options.compression_opts, paranoid_file_checks,
           cfd->internal_stats(), TableFileCreationReason::kRecovery, &io_s,
           io_tracer_, &event_logger_, job_id, Env::IO_HIGH,
-          nullptr /* table_properties */, -1 /* level */, current_time,
+          nullptr /* table_properties */, 0 /* level */, current_time,
           0 /* oldest_key_time */, write_hint, 0 /* file_creation_time */,
-          db_id_, db_session_id_);
+          db_id_, db_session_id_, nullptr /*full_history_ts_low*/,
+          &blob_callback_);
       LogFlush(immutable_db_options_.info_log);
       ROCKS_LOG_DEBUG(immutable_db_options_.info_log,
                       "[%s] [WriteLevel0TableForRecovery]"
@@ -1361,6 +1415,9 @@ Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
                       cfd->GetName().c_str(), meta.fd.GetNumber(),
                       meta.fd.GetFileSize(), s.ToString().c_str());
       mutex_.Lock();
+
+      io_s.PermitUncheckedError();  // TODO(AR) is this correct, or should we
+                                    // return io_s if not ok()?
     }
   }
   ReleaseFileNumberFromPendingOutputs(pending_outputs_inserted_elem);
@@ -1368,7 +1425,6 @@ Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
   // Note that if file_size is zero, the file has been deleted and
   // should not be added to the manifest.
   const bool has_output = meta.fd.GetFileSize() > 0;
-  assert(has_output || blob_file_additions.empty());
 
   constexpr int level = 0;
 
@@ -1380,26 +1436,30 @@ Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
                   meta.oldest_ancester_time, meta.file_creation_time,
                   meta.file_checksum, meta.file_checksum_func_name);
 
-    edit->SetBlobFileAdditions(std::move(blob_file_additions));
+    for (const auto& blob : blob_file_additions) {
+      edit->AddBlobFile(blob);
+    }
   }
 
   InternalStats::CompactionStats stats(CompactionReason::kFlush, 1);
-  stats.micros = env_->NowMicros() - start_micros;
+  stats.micros = immutable_db_options_.clock->NowMicros() - start_micros;
 
   if (has_output) {
     stats.bytes_written = meta.fd.GetFileSize();
-
-    const auto& blobs = edit->GetBlobFileAdditions();
-    for (const auto& blob : blobs) {
-      stats.bytes_written += blob.GetTotalBlobBytes();
-    }
-
-    stats.num_output_files = static_cast<int>(blobs.size()) + 1;
+    stats.num_output_files = 1;
   }
 
+  const auto& blobs = edit->GetBlobFileAdditions();
+  for (const auto& blob : blobs) {
+    stats.bytes_written_blob += blob.GetTotalBlobBytes();
+  }
+
+  stats.num_output_files_blob = static_cast<int>(blobs.size());
+
   cfd->internal_stats()->AddCompactionStats(level, Env::Priority::USER, stats);
-  cfd->internal_stats()->AddCFStats(InternalStats::BYTES_FLUSHED,
-                                    stats.bytes_written);
+  cfd->internal_stats()->AddCFStats(
+      InternalStats::BYTES_FLUSHED,
+      stats.bytes_written + stats.bytes_written_blob);
   RecordTick(stats_, COMPACT_WRITE_BYTES, meta.fd.GetFileSize());
   return s;
 }
@@ -1473,9 +1533,11 @@ IOStatus DBImpl::CreateWAL(uint64_t log_file_num, uint64_t recycle_log_number,
     lfile->SetPreallocationBlockSize(preallocate_block_size);
 
     const auto& listeners = immutable_db_options_.listeners;
+    FileTypeSet tmp_set = immutable_db_options_.checksum_handoff_file_types;
     std::unique_ptr<WritableFileWriter> file_writer(new WritableFileWriter(
-        std::move(lfile), log_fname, opt_file_options, env_, io_tracer_,
-        nullptr /* stats */, listeners));
+        std::move(lfile), log_fname, opt_file_options,
+        immutable_db_options_.clock, io_tracer_, nullptr /* stats */, listeners,
+        nullptr, tmp_set.Contains(FileType::kWalFile)));
     *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);
@@ -1556,6 +1618,7 @@ Status DBImpl::Open(const DBOptions& db_options, const std::string& dbname,
       InstrumentedMutexLock wl(&impl->log_write_mutex_);
       impl->logfile_number_ = new_log_number;
       assert(new_log != nullptr);
+      assert(impl->logs_.empty());
       impl->logs_.emplace_back(new_log_number, new_log);
     }
 
@@ -1610,7 +1673,7 @@ Status DBImpl::Open(const DBOptions& db_options, const std::string& dbname,
       // In WritePrepared there could be gap in sequence numbers. This breaks
       // the trick we use in kPointInTimeRecovery which assumes the first seq in
       // the log right after the corrupted log is one larger than the last seq
-      // we read from the logs. To let this trick keep working, we add a dummy
+      // we read from the wals. To let this trick keep working, we add a dummy
       // entry with the expected sequence to the first log right after recovery.
       // In non-WritePrepared case also the new log after recovery could be
       // empty, and thus missing the consecutive seq hint to distinguish
@@ -1703,6 +1766,8 @@ Status DBImpl::Open(const DBOptions& db_options, const std::string& dbname,
 
     std::vector<LiveFileMetaData> metadata;
 
+    // TODO: Once GetLiveFilesMetaData supports blob files, update the logic
+    // below to get known_file_sizes for blob files.
     impl->mutex_.Lock();
     impl->versions_->GetLiveFilesMetaData(&metadata);
     impl->mutex_.Unlock();
@@ -1728,21 +1793,19 @@ Status DBImpl::Open(const DBOptions& db_options, const std::string& dbname,
     paths.erase(std::unique(paths.begin(), paths.end()), paths.end());
     for (auto& path : paths) {
       std::vector<std::string> existing_files;
-      // TODO: Check for errors here?
       impl->immutable_db_options_.env->GetChildren(path, &existing_files)
-          .PermitUncheckedError();
+          .PermitUncheckedError();  //**TODO: What do to on error?
       for (auto& file_name : existing_files) {
         uint64_t file_number;
         FileType file_type;
         std::string file_path = path + "/" + file_name;
         if (ParseFileName(file_name, &file_number, &file_type) &&
-            file_type == kTableFile) {
+            (file_type == kTableFile || file_type == kBlobFile)) {
           // TODO: Check for errors from OnAddFile?
           if (known_file_sizes.count(file_name)) {
             // We're assuming that each sst file name exists in at most one of
             // the paths.
-            sfm->OnAddFile(file_path, known_file_sizes.at(file_name),
-                           /* compaction */ false)
+            sfm->OnAddFile(file_path, known_file_sizes.at(file_name))
                 .PermitUncheckedError();
           } else {
             sfm->OnAddFile(file_path).PermitUncheckedError();

db/db_impl/db_impl_readonly.cc

@@ -6,7 +6,7 @@
 #include "db/db_impl/db_impl_readonly.h"
 
 #include "db/arena_wrapped_db_iter.h"
-#include "db/compacted_db_impl.h"
+#include "db/db_impl/compacted_db_impl.h"
 #include "db/db_impl/db_impl.h"
 #include "db/db_iter.h"
 #include "db/merge_context.h"
@@ -19,7 +19,8 @@ namespace ROCKSDB_NAMESPACE {
 
 DBImplReadOnly::DBImplReadOnly(const DBOptions& db_options,
                                const std::string& dbname)
-    : DBImpl(db_options, dbname) {
+    : DBImpl(db_options, dbname, /*seq_per_batch*/ false,
+             /*batch_per_txn*/ true, /*read_only*/ true) {
   ROCKS_LOG_INFO(immutable_db_options_.info_log,
                  "Opening the db in read only mode");
   LogFlush(immutable_db_options_.info_log);
@@ -83,7 +84,7 @@ Iterator* DBImplReadOnly::NewIterator(const ReadOptions& read_options,
   ReadCallback* read_callback = nullptr;  // No read callback provided.
   auto db_iter = NewArenaWrappedDbIterator(
       env_, read_options, *cfd->ioptions(), super_version->mutable_cf_options,
-      read_seq,
+      super_version->current, read_seq,
       super_version->mutable_cf_options.max_sequential_skip_in_iterations,
       super_version->version_number, read_callback);
   auto internal_iter = NewInternalIterator(
@@ -115,7 +116,8 @@ Status DBImplReadOnly::NewIterators(
     auto* cfd = static_cast_with_check<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,
+        sv->current, read_seq,
         sv->mutable_cf_options.max_sequential_skip_in_iterations,
         sv->version_number, read_callback);
     auto* internal_iter = NewInternalIterator(
@@ -130,8 +132,8 @@ Status DBImplReadOnly::NewIterators(
 }
 
 namespace {
-// Return OK if dbname exists in the file system
-// or create_if_missing is false
+// Return OK if dbname exists in the file system or create it if
+// create_if_missing
 Status OpenForReadOnlyCheckExistence(const DBOptions& db_options,
                                      const std::string& dbname) {
   Status s;
@@ -142,9 +144,9 @@ Status OpenForReadOnlyCheckExistence(const DBOptions& db_options,
     uint64_t manifest_file_number;
     s = VersionSet::GetCurrentManifestPath(dbname, fs.get(), &manifest_path,
                                            &manifest_file_number);
-    if (!s.ok()) {
-      return Status::NotFound(CurrentFileName(dbname), "does not exist");
-    }
+  } else {
+    // Historic behavior that doesn't necessarily make sense
+    s = db_options.env->CreateDirIfMissing(dbname);
   }
   return s;
 }
@@ -152,7 +154,6 @@ Status OpenForReadOnlyCheckExistence(const DBOptions& db_options,
 
 Status DB::OpenForReadOnly(const Options& options, const std::string& dbname,
                            DB** dbptr, bool /*error_if_wal_file_exists*/) {
-  // If dbname does not exist in the file system, should not do anything
   Status s = OpenForReadOnlyCheckExistence(options, dbname);
   if (!s.ok()) {
     return s;

db/db_impl/db_impl_secondary.cc

@@ -17,8 +17,10 @@ namespace ROCKSDB_NAMESPACE {
 
 #ifndef ROCKSDB_LITE
 DBImplSecondary::DBImplSecondary(const DBOptions& db_options,
-                                 const std::string& dbname)
-    : DBImpl(db_options, dbname) {
+                                 const std::string& dbname,
+                                 std::string secondary_path)
+    : DBImpl(db_options, dbname, false, true, true),
+      secondary_path_(std::move(secondary_path)) {
   ROCKS_LOG_INFO(immutable_db_options_.info_log,
                  "Opening the db in secondary mode");
   LogFlush(immutable_db_options_.info_log);
@@ -112,7 +114,7 @@ Status DBImplSecondary::FindNewLogNumbers(std::vector<uint64_t>* logs) {
   for (size_t i = 0; i < filenames.size(); i++) {
     uint64_t number;
     FileType type;
-    if (ParseFileName(filenames[i], &number, &type) && type == kLogFile &&
+    if (ParseFileName(filenames[i], &number, &type) && type == kWalFile &&
         number >= log_number_min) {
       logs->push_back(number);
     }
@@ -327,8 +329,8 @@ 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_CPU_TIMER_GUARD(get_cpu_nanos, immutable_db_options_.clock);
+  StopWatch sw(immutable_db_options_.clock, stats_, DB_GET);
   PERF_TIMER_GUARD(get_snapshot_time);
 
   auto cfh = static_cast<ColumnFamilyHandleImpl*>(column_family);
@@ -421,7 +423,7 @@ ArenaWrappedDBIter* DBImplSecondary::NewIteratorImpl(
   SuperVersion* super_version = cfd->GetReferencedSuperVersion(this);
   auto db_iter = NewArenaWrappedDbIterator(
       env_, read_options, *cfd->ioptions(), super_version->mutable_cf_options,
-      snapshot,
+      super_version->current, snapshot,
       super_version->mutable_cf_options.max_sequential_skip_in_iterations,
       super_version->version_number, read_callback);
   auto internal_iter = NewInternalIterator(
@@ -519,7 +521,8 @@ Status DBImplSecondary::TryCatchUpWithPrimary() {
   {
     InstrumentedMutexLock lock_guard(&mutex_);
     s = static_cast_with_check<ReactiveVersionSet>(versions_.get())
-            ->ReadAndApply(&mutex_, &manifest_reader_, &cfds_changed);
+            ->ReadAndApply(&mutex_, &manifest_reader_,
+                           manifest_reader_status_.get(), &cfds_changed);
 
     ROCKS_LOG_INFO(immutable_db_options_.info_log, "Last sequence is %" PRIu64,
                    static_cast<uint64_t>(versions_->LastSequence()));
@@ -616,7 +619,7 @@ Status DB::OpenAsSecondary(
   }
 
   handles->clear();
-  DBImplSecondary* impl = new DBImplSecondary(tmp_opts, dbname);
+  DBImplSecondary* impl = new DBImplSecondary(tmp_opts, dbname, secondary_path);
   impl->versions_.reset(new ReactiveVersionSet(
       dbname, &impl->immutable_db_options_, impl->file_options_,
       impl->table_cache_.get(), impl->write_buffer_manager_,
@@ -662,6 +665,86 @@ Status DB::OpenAsSecondary(
   }
   return s;
 }
+
+Status DBImplSecondary::CompactWithoutInstallation(
+    ColumnFamilyHandle* cfh, const CompactionServiceInput& input,
+    CompactionServiceResult* result) {
+  InstrumentedMutexLock l(&mutex_);
+  auto cfd = static_cast_with_check<ColumnFamilyHandleImpl>(cfh)->cfd();
+  if (!cfd) {
+    return Status::InvalidArgument("Cannot find column family" +
+                                   cfh->GetName());
+  }
+
+  std::unordered_set<uint64_t> input_set;
+  for (const auto& file_name : input.input_files) {
+    input_set.insert(TableFileNameToNumber(file_name));
+  }
+
+  auto* version = cfd->current();
+
+  ColumnFamilyMetaData cf_meta;
+  version->GetColumnFamilyMetaData(&cf_meta);
+
+  const MutableCFOptions* mutable_cf_options = cfd->GetLatestMutableCFOptions();
+  ColumnFamilyOptions cf_options = cfd->GetLatestCFOptions();
+  VersionStorageInfo* vstorage = version->storage_info();
+
+  // Use comp_options to reuse some CompactFiles functions
+  CompactionOptions comp_options;
+  comp_options.compression = kDisableCompressionOption;
+  comp_options.output_file_size_limit = MaxFileSizeForLevel(
+      *mutable_cf_options, input.output_level, cf_options.compaction_style,
+      vstorage->base_level(), cf_options.level_compaction_dynamic_level_bytes);
+
+  std::vector<CompactionInputFiles> input_files;
+  Status s = cfd->compaction_picker()->GetCompactionInputsFromFileNumbers(
+      &input_files, &input_set, vstorage, comp_options);
+  if (!s.ok()) {
+    return s;
+  }
+
+  std::unique_ptr<Compaction> c;
+  assert(cfd->compaction_picker());
+  c.reset(cfd->compaction_picker()->CompactFiles(
+      comp_options, input_files, input.output_level, vstorage,
+      *mutable_cf_options, mutable_db_options_, 0));
+  assert(c != nullptr);
+
+  c->SetInputVersion(version);
+
+  // Create output directory if it's not existed yet
+  std::unique_ptr<FSDirectory> output_dir;
+  s = CreateAndNewDirectory(fs_.get(), secondary_path_, &output_dir);
+  if (!s.ok()) {
+    return s;
+  }
+
+  LogBuffer log_buffer(InfoLogLevel::INFO_LEVEL,
+                       immutable_db_options_.info_log.get());
+
+  const int job_id = next_job_id_.fetch_add(1);
+
+  CompactionServiceCompactionJob compaction_job(
+      job_id, c.get(), immutable_db_options_, file_options_for_compaction_,
+      versions_.get(), &shutting_down_, &log_buffer, output_dir.get(), stats_,
+      &mutex_, &error_handler_, input.snapshots, table_cache_, &event_logger_,
+      dbname_, io_tracer_, db_id_, db_session_id_, secondary_path_, input,
+      result);
+
+  mutex_.Unlock();
+  s = compaction_job.Run();
+  mutex_.Lock();
+
+  // clean up
+  compaction_job.io_status().PermitUncheckedError();
+  compaction_job.CleanupCompaction();
+  c->ReleaseCompactionFiles(s);
+  c.reset();
+
+  return s;
+}
+
 #else   // !ROCKSDB_LITE
 
 Status DB::OpenAsSecondary(const Options& /*options*/,

db/db_impl/db_impl_secondary.h

@@ -71,7 +71,8 @@ class LogReaderContainer {
 // effort attempts to catch up with the primary.
 class DBImplSecondary : public DBImpl {
  public:
-  DBImplSecondary(const DBOptions& options, const std::string& dbname);
+  DBImplSecondary(const DBOptions& options, const std::string& dbname,
+                  std::string secondary_path);
   ~DBImplSecondary() override;
 
   // Recover by replaying MANIFEST and WAL. Also initialize manifest_reader_
@@ -222,6 +223,14 @@ class DBImplSecondary : public DBImpl {
   // not flag the missing file as inconsistency.
   Status CheckConsistency() override;
 
+#ifndef NDEBUG
+  Status TEST_CompactWithoutInstallation(ColumnFamilyHandle* cfh,
+                                         const CompactionServiceInput& input,
+                                         CompactionServiceResult* result) {
+    return CompactWithoutInstallation(cfh, input, result);
+  }
+#endif  // NDEBUG
+
  protected:
   // ColumnFamilyCollector is a write batch handler which does nothing
   // except recording unique column family IDs
@@ -316,6 +325,13 @@ class DBImplSecondary : public DBImpl {
                          std::unordered_set<ColumnFamilyData*>* cfds_changed,
                          JobContext* job_context);
 
+  // Run compaction without installation, the output files will be placed in the
+  // secondary DB path. The LSM tree won't be changed, the secondary DB is still
+  // in read-only mode.
+  Status CompactWithoutInstallation(ColumnFamilyHandle* cfh,
+                                    const CompactionServiceInput& input,
+                                    CompactionServiceResult* result);
+
   std::unique_ptr<log::FragmentBufferedReader> manifest_reader_;
   std::unique_ptr<log::Reader::Reporter> manifest_reporter_;
   std::unique_ptr<Status> manifest_reader_status_;
@@ -326,6 +342,8 @@ class DBImplSecondary : public DBImpl {
 
   // Current WAL number replayed for each column family.
   std::unordered_map<ColumnFamilyData*, uint64_t> cfd_to_current_log_;
+
+  const std::string secondary_path_;
 };
 
 }  // namespace ROCKSDB_NAMESPACE

db/db_impl/db_impl_write.cc

@@ -160,10 +160,10 @@ Status DBImpl::WriteImpl(const WriteOptions& write_options,
     RecordTick(stats_, WRITE_WITH_WAL);
   }
 
-  StopWatch write_sw(env_, immutable_db_options_.statistics.get(), DB_WRITE);
+  StopWatch write_sw(immutable_db_options_.clock,
+                     immutable_db_options_.statistics.get(), DB_WRITE);
 
   write_thread_.JoinBatchGroup(&w);
-  Status status;
   if (w.state == WriteThread::STATE_PARALLEL_MEMTABLE_WRITER) {
     // we are a non-leader in a parallel group
 
@@ -193,8 +193,6 @@ Status DBImpl::WriteImpl(const WriteOptions& write_options,
     }
     assert(w.state == WriteThread::STATE_COMPLETED);
     // STATE_COMPLETED conditional below handles exit
-
-    status = w.FinalStatus();
   }
   if (w.state == WriteThread::STATE_COMPLETED) {
     if (log_used != nullptr) {
@@ -204,13 +202,11 @@ Status DBImpl::WriteImpl(const WriteOptions& write_options,
       *seq_used = w.sequence;
     }
     // write is complete and leader has updated sequence
-    // Should we handle it?
-    status.PermitUncheckedError();
     return w.FinalStatus();
   }
   // else we are the leader of the write batch group
   assert(w.state == WriteThread::STATE_GROUP_LEADER);
-
+  Status status;
   // Once reaches this point, the current writer "w" will try to do its write
   // job.  It may also pick up some of the remaining writers in the "writers_"
   // when it finds suitable, and finish them in the same write batch.
@@ -426,7 +422,11 @@ Status DBImpl::WriteImpl(const WriteOptions& write_options,
 
   if (need_log_sync) {
     mutex_.Lock();
-    MarkLogsSynced(logfile_number_, need_log_dir_sync, status);
+    if (status.ok()) {
+      status = MarkLogsSynced(logfile_number_, need_log_dir_sync);
+    } else {
+      MarkLogsNotSynced(logfile_number_);
+    }
     mutex_.Unlock();
     // Requesting sync with two_write_queues_ is expected to be very rare. We
     // hence provide a simple implementation that is not necessarily efficient.
@@ -466,7 +466,8 @@ Status DBImpl::PipelinedWriteImpl(const WriteOptions& write_options,
                                   uint64_t* log_used, uint64_t log_ref,
                                   bool disable_memtable, uint64_t* seq_used) {
   PERF_TIMER_GUARD(write_pre_and_post_process_time);
-  StopWatch write_sw(env_, immutable_db_options_.statistics.get(), DB_WRITE);
+  StopWatch write_sw(immutable_db_options_.clock,
+                     immutable_db_options_.statistics.get(), DB_WRITE);
 
   WriteContext write_context;
 
@@ -527,6 +528,8 @@ Status DBImpl::PipelinedWriteImpl(const WriteOptions& write_options,
     PERF_TIMER_STOP(write_pre_and_post_process_time);
 
     IOStatus io_s;
+    io_s.PermitUncheckedError();  // Allow io_s to be uninitialized
+
     if (w.status.ok() && !write_options.disableWAL) {
       PERF_TIMER_GUARD(write_wal_time);
       stats->AddDBStats(InternalStats::kIntStatsWriteDoneBySelf, 1);
@@ -551,14 +554,23 @@ Status DBImpl::PipelinedWriteImpl(const WriteOptions& write_options,
 
     if (need_log_sync) {
       mutex_.Lock();
-      MarkLogsSynced(logfile_number_, need_log_dir_sync, w.status);
+      if (w.status.ok()) {
+        w.status = MarkLogsSynced(logfile_number_, need_log_dir_sync);
+      } else {
+        MarkLogsNotSynced(logfile_number_);
+      }
       mutex_.Unlock();
     }
 
     write_thread_.ExitAsBatchGroupLeader(wal_write_group, w.status);
   }
 
+  // NOTE: the memtable_write_group is declared before the following
+  // `if` statement because its lifetime needs to be longer
+  // that the inner context  of the `if` as a reference to it
+  // may be used further below within the outer _write_thread
   WriteThread::WriteGroup memtable_write_group;
+
   if (w.state == WriteThread::STATE_MEMTABLE_WRITER_LEADER) {
     PERF_TIMER_GUARD(write_memtable_time);
     assert(w.ShouldWriteToMemtable());
@@ -575,6 +587,10 @@ Status DBImpl::PipelinedWriteImpl(const WriteOptions& write_options,
       versions_->SetLastSequence(memtable_write_group.last_sequence);
       write_thread_.ExitAsMemTableWriter(&w, memtable_write_group);
     }
+  } else {
+    // NOTE: the memtable_write_group is never really used,
+    // so we need to set its status to pass ASSERT_STATUS_CHECKED
+    memtable_write_group.status.PermitUncheckedError();
   }
 
   if (w.state == WriteThread::STATE_PARALLEL_MEMTABLE_WRITER) {
@@ -607,7 +623,8 @@ Status DBImpl::UnorderedWriteMemtable(const WriteOptions& write_options,
                                       SequenceNumber seq,
                                       const size_t sub_batch_cnt) {
   PERF_TIMER_GUARD(write_pre_and_post_process_time);
-  StopWatch write_sw(env_, immutable_db_options_.statistics.get(), DB_WRITE);
+  StopWatch write_sw(immutable_db_options_.clock,
+                     immutable_db_options_.statistics.get(), DB_WRITE);
 
   WriteThread::Writer w(write_options, my_batch, callback, log_ref,
                         false /*disable_memtable*/);
@@ -658,12 +675,12 @@ Status DBImpl::WriteImplWALOnly(
     const uint64_t log_ref, uint64_t* seq_used, const size_t sub_batch_cnt,
     PreReleaseCallback* pre_release_callback, const AssignOrder assign_order,
     const PublishLastSeq publish_last_seq, const bool disable_memtable) {
-  Status status;
   PERF_TIMER_GUARD(write_pre_and_post_process_time);
   WriteThread::Writer w(write_options, my_batch, callback, log_ref,
                         disable_memtable, sub_batch_cnt, pre_release_callback);
   RecordTick(stats_, WRITE_WITH_WAL);
-  StopWatch write_sw(env_, immutable_db_options_.statistics.get(), DB_WRITE);
+  StopWatch write_sw(immutable_db_options_.clock,
+                     immutable_db_options_.statistics.get(), DB_WRITE);
 
   write_thread->JoinBatchGroup(&w);
   assert(w.state != WriteThread::STATE_PARALLEL_MEMTABLE_WRITER);
@@ -680,6 +697,8 @@ Status DBImpl::WriteImplWALOnly(
   assert(w.state == WriteThread::STATE_GROUP_LEADER);
 
   if (publish_last_seq == kDoPublishLastSeq) {
+    Status status;
+
     // Currently we only use kDoPublishLastSeq in unordered_write
     assert(immutable_db_options_.unordered_write);
     WriteContext write_context;
@@ -756,7 +775,9 @@ Status DBImpl::WriteImplWALOnly(
     }
     seq_inc = total_batch_cnt;
   }
+  Status status;
   IOStatus io_s;
+  io_s.PermitUncheckedError();  // Allow io_s to be uninitialized
   if (!write_options.disableWAL) {
     io_s = ConcurrentWriteToWAL(write_group, log_used, &last_sequence, seq_inc);
     status = io_s;
@@ -842,8 +863,7 @@ void DBImpl::WriteStatusCheckOnLocked(const Status& status) {
   if (immutable_db_options_.paranoid_checks && !status.ok() &&
       !status.IsBusy() && !status.IsIncomplete()) {
     // Maybe change the return status to void?
-    error_handler_.SetBGError(status, BackgroundErrorReason::kWriteCallback)
-        .PermitUncheckedError();
+    error_handler_.SetBGError(status, BackgroundErrorReason::kWriteCallback);
   }
 }
 
@@ -855,8 +875,7 @@ void DBImpl::WriteStatusCheck(const Status& status) {
       !status.IsBusy() && !status.IsIncomplete()) {
     mutex_.Lock();
     // Maybe change the return status to void?
-    error_handler_.SetBGError(status, BackgroundErrorReason::kWriteCallback)
-        .PermitUncheckedError();
+    error_handler_.SetBGError(status, BackgroundErrorReason::kWriteCallback);
     mutex_.Unlock();
   }
 }
@@ -869,8 +888,7 @@ void DBImpl::IOStatusCheck(const IOStatus& io_status) {
       io_status.IsIOFenced()) {
     mutex_.Lock();
     // Maybe change the return status to void?
-    error_handler_.SetBGError(io_status, BackgroundErrorReason::kWriteCallback)
-        .PermitUncheckedError();
+    error_handler_.SetBGError(io_status, BackgroundErrorReason::kWriteCallback);
     mutex_.Unlock();
   }
 }
@@ -919,7 +937,7 @@ Status DBImpl::PreprocessWrite(const WriteOptions& write_options,
     // be flushed. We may end up with flushing much more DBs than needed. It's
     // suboptimal but still correct.
     WaitForPendingWrites();
-    status = HandleWriteBufferFull(write_context);
+    status = HandleWriteBufferManagerFlush(write_context);
   }
 
   if (UNLIKELY(status.ok() && !trim_history_scheduler_.Empty())) {
@@ -946,6 +964,20 @@ Status DBImpl::PreprocessWrite(const WriteOptions& write_options,
     PERF_TIMER_START(write_pre_and_post_process_time);
   }
 
+  // If memory usage exceeded beyond a certain threshold,
+  // write_buffer_manager_->ShouldStall() returns true to all threads writing to
+  // all DBs and writers will be stalled.
+  // It does soft checking because WriteBufferManager::buffer_limit_ has already
+  // exceeded at this point so no new write (including current one) will go
+  // through until memory usage is decreased.
+  if (UNLIKELY(status.ok() && write_buffer_manager_->ShouldStall())) {
+    if (write_options.no_slowdown) {
+      status = Status::Incomplete("Write stall");
+    } else {
+      WriteBufferManagerStallWrites();
+    }
+  }
+
   if (status.ok() && *need_log_sync) {
     // Wait until the parallel syncs are finished. Any sync process has to sync
     // the front log too so it is enough to check the status of front()
@@ -1079,7 +1111,7 @@ IOStatus DBImpl::WriteToWAL(const WriteThread::WriteGroup& write_group,
   }
 
   if (io_s.ok() && need_log_sync) {
-    StopWatch sw(env_, stats_, WAL_FILE_SYNC_MICROS);
+    StopWatch sw(immutable_db_options_.clock, stats_, WAL_FILE_SYNC_MICROS);
     // It's safe to access logs_ with unlocked mutex_ here because:
     //  - we've set getting_synced=true for all logs,
     //    so other threads won't pop from logs_ while we're here,
@@ -1327,16 +1359,23 @@ Status DBImpl::SwitchWAL(WriteContext* write_context) {
     }
     for (auto cfd : cfds) {
       cfd->imm()->FlushRequested();
+      if (!immutable_db_options_.atomic_flush) {
+        FlushRequest flush_req;
+        GenerateFlushRequest({cfd}, &flush_req);
+        SchedulePendingFlush(flush_req, FlushReason::kWalFull);
+      }
+    }
+    if (immutable_db_options_.atomic_flush) {
+      FlushRequest flush_req;
+      GenerateFlushRequest(cfds, &flush_req);
+      SchedulePendingFlush(flush_req, FlushReason::kWalFull);
     }
-    FlushRequest flush_req;
-    GenerateFlushRequest(cfds, &flush_req);
-    SchedulePendingFlush(flush_req, FlushReason::kWriteBufferManager);
     MaybeScheduleFlushOrCompaction();
   }
   return status;
 }
 
-Status DBImpl::HandleWriteBufferFull(WriteContext* write_context) {
+Status DBImpl::HandleWriteBufferManagerFlush(WriteContext* write_context) {
   mutex_.AssertHeld();
   assert(write_context != nullptr);
   Status status;
@@ -1348,7 +1387,7 @@ Status DBImpl::HandleWriteBufferFull(WriteContext* write_context) {
   // suboptimal but still correct.
   ROCKS_LOG_INFO(
       immutable_db_options_.info_log,
-      "Flushing column family with oldest memtable entry. Write buffer is "
+      "Flushing column family with oldest memtable entry. Write buffers are "
       "using %" ROCKSDB_PRIszt " bytes out of a total of %" ROCKSDB_PRIszt ".",
       write_buffer_manager_->memory_usage(),
       write_buffer_manager_->buffer_size());
@@ -1406,10 +1445,17 @@ Status DBImpl::HandleWriteBufferFull(WriteContext* write_context) {
     }
     for (const auto cfd : cfds) {
       cfd->imm()->FlushRequested();
+      if (!immutable_db_options_.atomic_flush) {
+        FlushRequest flush_req;
+        GenerateFlushRequest({cfd}, &flush_req);
+        SchedulePendingFlush(flush_req, FlushReason::kWriteBufferManager);
+      }
+    }
+    if (immutable_db_options_.atomic_flush) {
+      FlushRequest flush_req;
+      GenerateFlushRequest(cfds, &flush_req);
+      SchedulePendingFlush(flush_req, FlushReason::kWriteBufferManager);
     }
-    FlushRequest flush_req;
-    GenerateFlushRequest(cfds, &flush_req);
-    SchedulePendingFlush(flush_req, FlushReason::kWriteBufferFull);
     MaybeScheduleFlushOrCompaction();
   }
   return status;
@@ -1429,8 +1475,10 @@ Status DBImpl::DelayWrite(uint64_t num_bytes,
   uint64_t time_delayed = 0;
   bool delayed = false;
   {
-    StopWatch sw(env_, stats_, WRITE_STALL, &time_delayed);
-    uint64_t delay = write_controller_.GetDelay(env_, num_bytes);
+    StopWatch sw(immutable_db_options_.clock, stats_, WRITE_STALL,
+                 &time_delayed);
+    uint64_t delay =
+        write_controller_.GetDelay(immutable_db_options_.clock, num_bytes);
     if (delay > 0) {
       if (write_options.no_slowdown) {
         return Status::Incomplete("Write stall");
@@ -1442,19 +1490,21 @@ Status DBImpl::DelayWrite(uint64_t num_bytes,
       write_thread_.BeginWriteStall();
       TEST_SYNC_POINT("DBImpl::DelayWrite:BeginWriteStallDone");
       mutex_.Unlock();
-      // We will delay the write until we have slept for delay ms or
-      // we don't need a delay anymore
-      const uint64_t kDelayInterval = 1000;
+      // We will delay the write until we have slept for `delay` microseconds
+      // or we don't need a delay anymore. We check for cancellation every 1ms
+      // (slightly longer because WriteController minimum delay is 1ms, in
+      // case of sleep imprecision, rounding, etc.)
+      const uint64_t kDelayInterval = 1001;
       uint64_t stall_end = sw.start_time() + delay;
       while (write_controller_.NeedsDelay()) {
-        if (env_->NowMicros() >= stall_end) {
+        if (immutable_db_options_.clock->NowMicros() >= stall_end) {
           // We already delayed this write `delay` microseconds
           break;
         }
 
         delayed = true;
         // Sleep for 0.001 seconds
-        env_->SleepForMicroseconds(kDelayInterval);
+        immutable_db_options_.clock->SleepForMicroseconds(kDelayInterval);
       }
       mutex_.Lock();
       write_thread_.EndWriteStall();
@@ -1500,6 +1550,29 @@ Status DBImpl::DelayWrite(uint64_t num_bytes,
   return s;
 }
 
+// REQUIRES: mutex_ is held
+// REQUIRES: this thread is currently at the front of the writer queue
+void DBImpl::WriteBufferManagerStallWrites() {
+  mutex_.AssertHeld();
+  // First block future writer threads who want to add themselves to the queue
+  // of WriteThread.
+  write_thread_.BeginWriteStall();
+  mutex_.Unlock();
+
+  // Change the state to State::Blocked.
+  static_cast<WBMStallInterface*>(wbm_stall_.get())
+      ->SetState(WBMStallInterface::State::BLOCKED);
+  // Then WriteBufferManager will add DB instance to its queue
+  // and block this thread by calling WBMStallInterface::Block().
+  write_buffer_manager_->BeginWriteStall(wbm_stall_.get());
+  wbm_stall_->Block();
+
+  mutex_.Lock();
+  // Stall has ended. Signal writer threads so that they can add
+  // themselves to the WriteThread queue for writes.
+  write_thread_.EndWriteStall();
+}
+
 Status DBImpl::ThrottleLowPriWritesIfNeeded(const WriteOptions& write_options,
                                             WriteBatch* my_batch) {
   assert(write_options.low_pri);
@@ -1633,10 +1706,16 @@ Status DBImpl::ScheduleFlushes(WriteContext* context) {
   if (status.ok()) {
     if (immutable_db_options_.atomic_flush) {
       AssignAtomicFlushSeq(cfds);
+      FlushRequest flush_req;
+      GenerateFlushRequest(cfds, &flush_req);
+      SchedulePendingFlush(flush_req, FlushReason::kWriteBufferFull);
+    } else {
+      for (auto* cfd : cfds) {
+        FlushRequest flush_req;
+        GenerateFlushRequest({cfd}, &flush_req);
+        SchedulePendingFlush(flush_req, FlushReason::kWriteBufferFull);
+      }
     }
-    FlushRequest flush_req;
-    GenerateFlushRequest(cfds, &flush_req);
-    SchedulePendingFlush(flush_req, FlushReason::kWriteBufferFull);
     MaybeScheduleFlushOrCompaction();
   }
   return status;
@@ -1782,32 +1861,74 @@ Status DBImpl::SwitchMemtable(ColumnFamilyData* cfd, WriteContext* context) {
     }
     // 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
-    // Should handle return error?
     if (!io_s.ok()) {
-      // Should handle return error?
-      error_handler_.SetBGError(io_s, BackgroundErrorReason::kMemTable)
-          .PermitUncheckedError();
+      error_handler_.SetBGError(io_s, BackgroundErrorReason::kMemTable);
     } else {
-      // Should handle return error?
-      error_handler_.SetBGError(s, BackgroundErrorReason::kMemTable)
-          .PermitUncheckedError();
+      error_handler_.SetBGError(s, BackgroundErrorReason::kMemTable);
     }
     // Read back bg_error in order to get the right severity
     s = error_handler_.GetBGError();
     return s;
   }
 
-  for (auto loop_cfd : *versions_->GetColumnFamilySet()) {
-    // all this is just optimization to delete logs that
-    // are no longer needed -- if CF is empty, that means it
-    // doesn't need that particular log to stay alive, so we just
-    // advance the log number. no need to persist this in the manifest
-    if (loop_cfd->mem()->GetFirstSequenceNumber() == 0 &&
-        loop_cfd->imm()->NumNotFlushed() == 0) {
-      if (creating_new_log) {
-        loop_cfd->SetLogNumber(logfile_number_);
+  bool empty_cf_updated = false;
+  if (immutable_db_options_.track_and_verify_wals_in_manifest &&
+      !immutable_db_options_.allow_2pc && creating_new_log) {
+    // In non-2pc mode, WALs become obsolete if they do not contain unflushed
+    // data. Updating the empty CF's log number might cause some WALs to become
+    // obsolete. So we should track the WAL obsoletion event before actually
+    // updating the empty CF's log number.
+    uint64_t min_wal_number_to_keep =
+        versions_->PreComputeMinLogNumberWithUnflushedData(logfile_number_);
+    if (min_wal_number_to_keep >
+        versions_->GetWalSet().GetMinWalNumberToKeep()) {
+      // Get a snapshot of the empty column families.
+      // LogAndApply may release and reacquire db
+      // mutex, during that period, column family may become empty (e.g. its
+      // flush succeeds), then it affects the computed min_log_number_to_keep,
+      // so we take a snapshot for consistency of column family data
+      // status. If a column family becomes non-empty afterwards, its active log
+      // should still be the created new log, so the min_log_number_to_keep is
+      // not affected.
+      autovector<ColumnFamilyData*> empty_cfs;
+      for (auto cf : *versions_->GetColumnFamilySet()) {
+        if (cf->IsEmpty()) {
+          empty_cfs.push_back(cf);
+        }
+      }
+
+      VersionEdit wal_deletion;
+      wal_deletion.DeleteWalsBefore(min_wal_number_to_keep);
+      s = versions_->LogAndApplyToDefaultColumnFamily(&wal_deletion, &mutex_);
+      if (!s.ok() && versions_->io_status().IsIOError()) {
+        s = error_handler_.SetBGError(versions_->io_status(),
+                                      BackgroundErrorReason::kManifestWrite);
+      }
+      if (!s.ok()) {
+        return s;
+      }
+
+      for (auto cf : empty_cfs) {
+        if (cf->IsEmpty()) {
+          cf->SetLogNumber(logfile_number_);
+          cf->mem()->SetCreationSeq(versions_->LastSequence());
+        }  // cf may become non-empty.
+      }
+      empty_cf_updated = true;
+    }
+  }
+  if (!empty_cf_updated) {
+    for (auto cf : *versions_->GetColumnFamilySet()) {
+      // all this is just optimization to delete logs that
+      // are no longer needed -- if CF is empty, that means it
+      // doesn't need that particular log to stay alive, so we just
+      // advance the log number. no need to persist this in the manifest
+      if (cf->IsEmpty()) {
+        if (creating_new_log) {
+          cf->SetLogNumber(logfile_number_);
+        }
+        cf->mem()->SetCreationSeq(versions_->LastSequence());
       }
-      loop_cfd->mem()->SetCreationSeq(versions_->LastSequence());
     }
   }
 

db/db_info_dumper.cc

@@ -62,11 +62,12 @@ void DumpDBFileSummary(const ImmutableDBOptions& options,
                 dbname.c_str(), file.c_str());
         }
         break;
-      case kLogFile:
+      case kWalFile:
         if (env->GetFileSize(dbname + "/" + file, &file_size).ok()) {
-          char str[16];
-          snprintf(str, sizeof(str), "%" PRIu64, file_size);
-          wal_info.append(file).append(" size: ").append(str).append(" ; ");
+          wal_info.append(file)
+              .append(" size: ")
+              .append(std::to_string(file_size))
+              .append(" ; ");
         } else {
           Error(options.info_log, "Error when reading LOG file: %s/%s\n",
                 dbname.c_str(), file.c_str());
@@ -118,11 +119,12 @@ void DumpDBFileSummary(const ImmutableDBOptions& options,
     wal_info.clear();
     for (const std::string& file : files) {
       if (ParseFileName(file, &number, &type)) {
-        if (type == kLogFile) {
+        if (type == kWalFile) {
           if (env->GetFileSize(options.wal_dir + "/" + file, &file_size).ok()) {
-            char str[16];
-            snprintf(str, sizeof(str), "%" PRIu64, file_size);
-            wal_info.append(file).append(" size: ").append(str).append(" ; ");
+            wal_info.append(file)
+                .append(" size: ")
+                .append(std::to_string(file_size))
+                .append(" ; ");
           } else {
             Error(options.info_log, "Error when reading LOG file %s/%s\n",
                   options.wal_dir.c_str(), file.c_str());

db/db_io_failure_test.cc

@@ -43,11 +43,15 @@ TEST_F(DBIOFailureTest, DropWrites) {
           if (level > 0 && level == dbfull()->NumberLevels() - 1) {
             break;
           }
-          dbfull()->TEST_CompactRange(level, nullptr, nullptr, nullptr,
-                                      true /* disallow trivial move */);
+          Status s =
+              dbfull()->TEST_CompactRange(level, nullptr, nullptr, nullptr,
+                                          true /* disallow trivial move */);
+          ASSERT_TRUE(s.ok() || s.IsCorruption());
         }
       } else {
-        dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+        Status s =
+            dbfull()->CompactRange(CompactRangeOptions(), nullptr, nullptr);
+        ASSERT_TRUE(s.ok() || s.IsCorruption());
       }
     }
 
@@ -56,7 +60,8 @@ TEST_F(DBIOFailureTest, DropWrites) {
     ASSERT_EQ("5", property_value);
 
     env_->drop_writes_.store(false, std::memory_order_release);
-    ASSERT_LT(CountFiles(), num_files + 3);
+    const size_t count = CountFiles();
+    ASSERT_LT(count, num_files + 3);
 
     // Check that compaction attempts slept after errors
     // TODO @krad: Figure out why ASSERT_EQ 5 keeps failing in certain compiler
@@ -82,7 +87,8 @@ TEST_F(DBIOFailureTest, DropWritesFlush) {
     ASSERT_TRUE(db_->GetProperty("rocksdb.background-errors", &property_value));
     ASSERT_EQ("0", property_value);
 
-    dbfull()->TEST_FlushMemTable(true);
+    // ASSERT file is too short
+    ASSERT_TRUE(dbfull()->TEST_FlushMemTable(true).IsCorruption());
 
     ASSERT_TRUE(db_->GetProperty("rocksdb.background-errors", &property_value));
     ASSERT_EQ("1", property_value);
@@ -166,7 +172,7 @@ TEST_F(DBIOFailureTest, ManifestWriteError) {
     ASSERT_EQ("bar", Get("foo"));
 
     // Memtable compaction (will succeed)
-    Flush();
+    ASSERT_OK(Flush());
     ASSERT_EQ("bar", Get("foo"));
     const int last = 2;
     MoveFilesToLevel(2);
@@ -174,7 +180,8 @@ TEST_F(DBIOFailureTest, ManifestWriteError) {
 
     // Merging compaction (will fail)
     error_type->store(true, std::memory_order_release);
-    dbfull()->TEST_CompactRange(last, nullptr, nullptr);  // Should fail
+    ASSERT_NOK(
+        dbfull()->TEST_CompactRange(last, nullptr, nullptr));  // Should fail
     ASSERT_EQ("bar", Get("foo"));
 
     error_type->store(false, std::memory_order_release);
@@ -192,7 +199,13 @@ TEST_F(DBIOFailureTest, ManifestWriteError) {
 
     // Merging compaction (will fail)
     error_type->store(true, std::memory_order_release);
-    dbfull()->TEST_CompactRange(last, nullptr, nullptr);  // Should fail
+    Status s =
+        dbfull()->TEST_CompactRange(last, nullptr, nullptr);  // Should fail
+    if (iter == 0) {
+      ASSERT_OK(s);
+    } else {
+      ASSERT_TRUE(s.IsIOError());
+    }
     ASSERT_EQ("bar", Get("foo"));
 
     // Recovery: should not lose data
@@ -220,18 +233,15 @@ TEST_F(DBIOFailureTest, PutFailsParanoid) {
   options.paranoid_checks = true;
   DestroyAndReopen(options);
   CreateAndReopenWithCF({"pikachu"}, options);
-  Status s;
 
   ASSERT_OK(Put(1, "foo", "bar"));
   ASSERT_OK(Put(1, "foo1", "bar1"));
   // simulate error
   env_->log_write_error_.store(true, std::memory_order_release);
-  s = Put(1, "foo2", "bar2");
-  ASSERT_TRUE(!s.ok());
+  ASSERT_NOK(Put(1, "foo2", "bar2"));
   env_->log_write_error_.store(false, std::memory_order_release);
-  s = Put(1, "foo3", "bar3");
   // the next put should fail, too
-  ASSERT_TRUE(!s.ok());
+  ASSERT_NOK(Put(1, "foo3", "bar3"));
   // but we're still able to read
   ASSERT_EQ("bar", Get(1, "foo"));
 
@@ -244,12 +254,10 @@ TEST_F(DBIOFailureTest, PutFailsParanoid) {
   ASSERT_OK(Put(1, "foo1", "bar1"));
   // simulate error
   env_->log_write_error_.store(true, std::memory_order_release);
-  s = Put(1, "foo2", "bar2");
-  ASSERT_TRUE(!s.ok());
+  ASSERT_NOK(Put(1, "foo2", "bar2"));
   env_->log_write_error_.store(false, std::memory_order_release);
-  s = Put(1, "foo3", "bar3");
   // the next put should NOT fail
-  ASSERT_TRUE(s.ok());
+  ASSERT_OK(Put(1, "foo3", "bar3"));
 }
 #if !(defined NDEBUG) || !defined(OS_WIN)
 TEST_F(DBIOFailureTest, FlushSstRangeSyncError) {
@@ -269,14 +277,14 @@ TEST_F(DBIOFailureTest, FlushSstRangeSyncError) {
 
   DestroyAndReopen(options);
   CreateAndReopenWithCF({"pikachu"}, options);
-  Status s;
 
+  const char* io_error_msg = "range sync dummy error";
   std::atomic<int> range_sync_called(0);
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
       "SpecialEnv::SStableFile::RangeSync", [&](void* arg) {
         if (range_sync_called.fetch_add(1) == 0) {
           Status* st = static_cast<Status*>(arg);
-          *st = Status::IOError("range sync dummy error");
+          *st = Status::IOError(io_error_msg);
         }
       });
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
@@ -298,7 +306,9 @@ TEST_F(DBIOFailureTest, FlushSstRangeSyncError) {
   ASSERT_OK(Put(1, "foo3_2", rnd_str));
   ASSERT_OK(Put(1, "foo3_3", rnd_str));
   ASSERT_OK(Put(1, "foo4", "bar"));
-  dbfull()->TEST_WaitForFlushMemTable(handles_[1]);
+  Status s = dbfull()->TEST_WaitForFlushMemTable(handles_[1]);
+  ASSERT_TRUE(s.IsIOError());
+  ASSERT_STREQ(s.getState(), io_error_msg);
 
   // Following writes should fail as flush failed.
   ASSERT_NOK(Put(1, "foo2", "bar3"));
@@ -328,7 +338,6 @@ TEST_F(DBIOFailureTest, CompactSstRangeSyncError) {
   options.table_factory.reset(NewBlockBasedTableFactory(table_options));
   DestroyAndReopen(options);
   CreateAndReopenWithCF({"pikachu"}, options);
-  Status s;
 
   Random rnd(301);
   std::string rnd_str =
@@ -342,21 +351,22 @@ TEST_F(DBIOFailureTest, CompactSstRangeSyncError) {
   ASSERT_OK(Put(1, "foo1_1", rnd_str));
   ASSERT_OK(Put(1, "foo1_2", rnd_str));
   ASSERT_OK(Put(1, "foo1_3", rnd_str));
-  Flush(1);
+  ASSERT_OK(Flush(1));
   ASSERT_OK(Put(1, "foo", "bar"));
   ASSERT_OK(Put(1, "foo3_1", rnd_str));
   ASSERT_OK(Put(1, "foo3_2", rnd_str));
   ASSERT_OK(Put(1, "foo3_3", rnd_str));
   ASSERT_OK(Put(1, "foo4", "bar"));
-  Flush(1);
-  dbfull()->TEST_WaitForFlushMemTable(handles_[1]);
+  ASSERT_OK(Flush(1));
+  ASSERT_OK(dbfull()->TEST_WaitForFlushMemTable(handles_[1]));
 
+  const char* io_error_msg = "range sync dummy error";
   std::atomic<int> range_sync_called(0);
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
       "SpecialEnv::SStableFile::RangeSync", [&](void* arg) {
         if (range_sync_called.fetch_add(1) == 0) {
           Status* st = static_cast<Status*>(arg);
-          *st = Status::IOError("range sync dummy error");
+          *st = Status::IOError(io_error_msg);
         }
       });
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
@@ -365,7 +375,9 @@ TEST_F(DBIOFailureTest, CompactSstRangeSyncError) {
                                  {
                                      {"disable_auto_compactions", "false"},
                                  }));
-  dbfull()->TEST_WaitForCompact();
+  Status s = dbfull()->TEST_WaitForCompact();
+  ASSERT_TRUE(s.IsIOError());
+  ASSERT_STREQ(s.getState(), io_error_msg);
 
   // Following writes should fail as flush failed.
   ASSERT_NOK(Put(1, "foo2", "bar3"));
@@ -389,13 +401,14 @@ TEST_F(DBIOFailureTest, FlushSstCloseError) {
 
   DestroyAndReopen(options);
   CreateAndReopenWithCF({"pikachu"}, options);
-  Status s;
+
+  const char* io_error_msg = "close dummy error";
   std::atomic<int> close_called(0);
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
       "SpecialEnv::SStableFile::Close", [&](void* arg) {
         if (close_called.fetch_add(1) == 0) {
           Status* st = static_cast<Status*>(arg);
-          *st = Status::IOError("close dummy error");
+          *st = Status::IOError(io_error_msg);
         }
       });
 
@@ -404,7 +417,9 @@ TEST_F(DBIOFailureTest, FlushSstCloseError) {
   ASSERT_OK(Put(1, "foo", "bar"));
   ASSERT_OK(Put(1, "foo1", "bar1"));
   ASSERT_OK(Put(1, "foo", "bar2"));
-  dbfull()->TEST_WaitForFlushMemTable(handles_[1]);
+  Status s = dbfull()->TEST_WaitForFlushMemTable(handles_[1]);
+  ASSERT_TRUE(s.IsIOError());
+  ASSERT_STREQ(s.getState(), io_error_msg);
 
   // Following writes should fail as flush failed.
   ASSERT_NOK(Put(1, "foo2", "bar3"));
@@ -429,25 +444,25 @@ TEST_F(DBIOFailureTest, CompactionSstCloseError) {
 
   DestroyAndReopen(options);
   CreateAndReopenWithCF({"pikachu"}, options);
-  Status s;
 
   ASSERT_OK(Put(1, "foo", "bar"));
   ASSERT_OK(Put(1, "foo2", "bar"));
-  Flush(1);
+  ASSERT_OK(Flush(1));
   ASSERT_OK(Put(1, "foo", "bar2"));
   ASSERT_OK(Put(1, "foo2", "bar"));
-  Flush(1);
+  ASSERT_OK(Flush(1));
   ASSERT_OK(Put(1, "foo", "bar3"));
   ASSERT_OK(Put(1, "foo2", "bar"));
-  Flush(1);
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(Flush(1));
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
 
+  const char* io_error_msg = "close dummy error";
   std::atomic<int> close_called(0);
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
       "SpecialEnv::SStableFile::Close", [&](void* arg) {
         if (close_called.fetch_add(1) == 0) {
           Status* st = static_cast<Status*>(arg);
-          *st = Status::IOError("close dummy error");
+          *st = Status::IOError(io_error_msg);
         }
       });
 
@@ -456,7 +471,9 @@ TEST_F(DBIOFailureTest, CompactionSstCloseError) {
                                  {
                                      {"disable_auto_compactions", "false"},
                                  }));
-  dbfull()->TEST_WaitForCompact();
+  Status s = dbfull()->TEST_WaitForCompact();
+  ASSERT_TRUE(s.IsIOError());
+  ASSERT_STREQ(s.getState(), io_error_msg);
 
   // Following writes should fail as compaction failed.
   ASSERT_NOK(Put(1, "foo2", "bar3"));
@@ -480,13 +497,14 @@ TEST_F(DBIOFailureTest, FlushSstSyncError) {
 
   DestroyAndReopen(options);
   CreateAndReopenWithCF({"pikachu"}, options);
-  Status s;
+
+  const char* io_error_msg = "sync dummy error";
   std::atomic<int> sync_called(0);
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
       "SpecialEnv::SStableFile::Sync", [&](void* arg) {
         if (sync_called.fetch_add(1) == 0) {
           Status* st = static_cast<Status*>(arg);
-          *st = Status::IOError("sync dummy error");
+          *st = Status::IOError(io_error_msg);
         }
       });
 
@@ -495,7 +513,9 @@ TEST_F(DBIOFailureTest, FlushSstSyncError) {
   ASSERT_OK(Put(1, "foo", "bar"));
   ASSERT_OK(Put(1, "foo1", "bar1"));
   ASSERT_OK(Put(1, "foo", "bar2"));
-  dbfull()->TEST_WaitForFlushMemTable(handles_[1]);
+  Status s = dbfull()->TEST_WaitForFlushMemTable(handles_[1]);
+  ASSERT_TRUE(s.IsIOError());
+  ASSERT_STREQ(s.getState(), io_error_msg);
 
   // Following writes should fail as flush failed.
   ASSERT_NOK(Put(1, "foo2", "bar3"));
@@ -521,25 +541,25 @@ TEST_F(DBIOFailureTest, CompactionSstSyncError) {
 
   DestroyAndReopen(options);
   CreateAndReopenWithCF({"pikachu"}, options);
-  Status s;
 
   ASSERT_OK(Put(1, "foo", "bar"));
   ASSERT_OK(Put(1, "foo2", "bar"));
-  Flush(1);
+  ASSERT_OK(Flush(1));
   ASSERT_OK(Put(1, "foo", "bar2"));
   ASSERT_OK(Put(1, "foo2", "bar"));
-  Flush(1);
+  ASSERT_OK(Flush(1));
   ASSERT_OK(Put(1, "foo", "bar3"));
   ASSERT_OK(Put(1, "foo2", "bar"));
-  Flush(1);
-  dbfull()->TEST_WaitForCompact();
+  ASSERT_OK(Flush(1));
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
 
+  const char* io_error_msg = "sync dummy error";
   std::atomic<int> sync_called(0);
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
       "SpecialEnv::SStableFile::Sync", [&](void* arg) {
         if (sync_called.fetch_add(1) == 0) {
           Status* st = static_cast<Status*>(arg);
-          *st = Status::IOError("close dummy error");
+          *st = Status::IOError(io_error_msg);
         }
       });
 
@@ -548,7 +568,9 @@ TEST_F(DBIOFailureTest, CompactionSstSyncError) {
                                  {
                                      {"disable_auto_compactions", "false"},
                                  }));
-  dbfull()->TEST_WaitForCompact();
+  Status s = dbfull()->TEST_WaitForCompact();
+  ASSERT_TRUE(s.IsIOError());
+  ASSERT_STREQ(s.getState(), io_error_msg);
 
   // Following writes should fail as compaction failed.
   ASSERT_NOK(Put(1, "foo2", "bar3"));

db/db_iter.cc

@@ -8,9 +8,10 @@
 // found in the LICENSE file. See the AUTHORS file for names of contributors.
 
 #include "db/db_iter.h"
-#include <string>
+
 #include <iostream>
 #include <limits>
+#include <string>
 
 #include "db/dbformat.h"
 #include "db/merge_context.h"
@@ -24,6 +25,7 @@
 #include "rocksdb/iterator.h"
 #include "rocksdb/merge_operator.h"
 #include "rocksdb/options.h"
+#include "rocksdb/system_clock.h"
 #include "table/internal_iterator.h"
 #include "table/iterator_wrapper.h"
 #include "trace_replay/trace_replay.h"
@@ -36,16 +38,19 @@ namespace ROCKSDB_NAMESPACE {
 DBIter::DBIter(Env* _env, const ReadOptions& read_options,
                const ImmutableCFOptions& cf_options,
                const MutableCFOptions& mutable_cf_options,
-               const Comparator* cmp, InternalIterator* iter, SequenceNumber s,
-               bool arena_mode, uint64_t max_sequential_skip_in_iterations,
+               const Comparator* cmp, InternalIterator* iter,
+               const Version* version, SequenceNumber s, bool arena_mode,
+               uint64_t max_sequential_skip_in_iterations,
                ReadCallback* read_callback, DBImpl* db_impl,
-               ColumnFamilyData* cfd, bool allow_blob)
+               ColumnFamilyData* cfd, bool expose_blob_index)
     : prefix_extractor_(mutable_cf_options.prefix_extractor.get()),
       env_(_env),
+      clock_(cf_options.clock),
       logger_(cf_options.info_log),
       user_comparator_(cmp),
-      merge_operator_(cf_options.merge_operator),
+      merge_operator_(cf_options.merge_operator.get()),
       iter_(iter),
+      version_(version),
       read_callback_(read_callback),
       sequence_(s),
       statistics_(cf_options.statistics),
@@ -65,7 +70,9 @@ DBIter::DBIter(Env* _env, const ReadOptions& read_options,
       expect_total_order_inner_iter_(prefix_extractor_ == nullptr ||
                                      read_options.total_order_seek ||
                                      read_options.auto_prefix_mode),
-      allow_blob_(allow_blob),
+      read_tier_(read_options.read_tier),
+      verify_checksums_(read_options.verify_checksums),
+      expose_blob_index_(expose_blob_index),
       is_blob_(false),
       arena_mode_(arena_mode),
       range_del_agg_(&cf_options.internal_comparator, s),
@@ -107,11 +114,11 @@ Status DBIter::GetProperty(std::string prop_name, std::string* prop) {
 }
 
 bool DBIter::ParseKey(ParsedInternalKey* ikey) {
-  if (ParseInternalKey(iter_.key(), ikey) != Status::OK()) {
-    status_ = Status::Corruption("corrupted internal key in DBIter");
+  Status s = ParseInternalKey(iter_.key(), ikey, false /* log_err_key */);
+  if (!s.ok()) {
+    status_ = Status::Corruption("In DBIter: ", s.getState());
     valid_ = false;
-    ROCKS_LOG_ERROR(logger_, "corrupted internal key in DBIter: %s",
-                    iter_.key().ToString(true).c_str());
+    ROCKS_LOG_ERROR(logger_, "In DBIter: %s", status_.getState());
     return false;
   } else {
     return true;
@@ -122,7 +129,7 @@ void DBIter::Next() {
   assert(valid_);
   assert(status_.ok());
 
-  PERF_CPU_TIMER_GUARD(iter_next_cpu_nanos, env_);
+  PERF_CPU_TIMER_GUARD(iter_next_cpu_nanos, clock_);
   // Release temporarily pinned blocks from last operation
   ReleaseTempPinnedData();
   local_stats_.skip_count_ += num_internal_keys_skipped_;
@@ -164,6 +171,42 @@ void DBIter::Next() {
   }
 }
 
+bool DBIter::SetBlobValueIfNeeded(const Slice& user_key,
+                                  const Slice& blob_index) {
+  assert(!is_blob_);
+
+  if (expose_blob_index_) {  // Stacked BlobDB implementation
+    is_blob_ = true;
+    return true;
+  }
+
+  if (!version_) {
+    status_ = Status::Corruption("Encountered unexpected blob index.");
+    valid_ = false;
+    return false;
+  }
+
+  // TODO: consider moving ReadOptions from ArenaWrappedDBIter to DBIter to
+  // avoid having to copy options back and forth.
+  ReadOptions read_options;
+  read_options.read_tier = read_tier_;
+  read_options.verify_checksums = verify_checksums_;
+
+  constexpr uint64_t* bytes_read = nullptr;
+
+  const Status s = version_->GetBlob(read_options, user_key, blob_index,
+                                     &blob_value_, bytes_read);
+
+  if (!s.ok()) {
+    status_ = s;
+    valid_ = false;
+    return false;
+  }
+
+  is_blob_ = true;
+  return true;
+}
+
 // PRE: saved_key_ has the current user key if skipping_saved_key
 // POST: saved_key_ should have the next user key if valid_,
 //       if the current entry is a result of merge
@@ -220,25 +263,28 @@ bool DBIter::FindNextUserEntryInternal(bool skipping_saved_key,
       is_key_seqnum_zero_ = false;
       return false;
     }
+    Slice user_key_without_ts =
+        StripTimestampFromUserKey(ikey_.user_key, timestamp_size_);
 
     is_key_seqnum_zero_ = (ikey_.sequence == 0);
 
     assert(iterate_upper_bound_ == nullptr ||
            iter_.UpperBoundCheckResult() != IterBoundCheck::kInbound ||
            user_comparator_.CompareWithoutTimestamp(
-               ikey_.user_key, /*a_has_ts=*/true, *iterate_upper_bound_,
+               user_key_without_ts, /*a_has_ts=*/false, *iterate_upper_bound_,
                /*b_has_ts=*/false) < 0);
     if (iterate_upper_bound_ != nullptr &&
         iter_.UpperBoundCheckResult() != IterBoundCheck::kInbound &&
         user_comparator_.CompareWithoutTimestamp(
-            ikey_.user_key, /*a_has_ts=*/true, *iterate_upper_bound_,
+            user_key_without_ts, /*a_has_ts=*/false, *iterate_upper_bound_,
             /*b_has_ts=*/false) >= 0) {
       break;
     }
 
     assert(prefix == nullptr || prefix_extractor_ != nullptr);
     if (prefix != nullptr &&
-        prefix_extractor_->Transform(ikey_.user_key).compare(*prefix) != 0) {
+        prefix_extractor_->Transform(user_key_without_ts).compare(*prefix) !=
+            0) {
       assert(prefix_same_as_start_);
       break;
     }
@@ -315,8 +361,14 @@ bool DBIter::FindNextUserEntryInternal(bool skipping_saved_key,
           case kTypeBlobIndex:
             if (start_seqnum_ > 0) {
               if (ikey_.sequence >= start_seqnum_) {
-                assert(ikey_.type != kTypeBlobIndex);
                 saved_key_.SetInternalKey(ikey_);
+
+                if (ikey_.type == kTypeBlobIndex) {
+                  if (!SetBlobValueIfNeeded(ikey_.user_key, iter_.value())) {
+                    return false;
+                  }
+                }
+
                 valid_ = true;
                 return true;
               } else {
@@ -330,6 +382,13 @@ bool DBIter::FindNextUserEntryInternal(bool skipping_saved_key,
               }
             } else if (timestamp_lb_) {
               saved_key_.SetInternalKey(ikey_);
+
+              if (ikey_.type == kTypeBlobIndex) {
+                if (!SetBlobValueIfNeeded(ikey_.user_key, iter_.value())) {
+                  return false;
+                }
+              }
+
               valid_ = true;
               return true;
             } else {
@@ -344,20 +403,13 @@ bool DBIter::FindNextUserEntryInternal(bool skipping_saved_key,
                 num_skipped = 0;
                 reseek_done = false;
                 PERF_COUNTER_ADD(internal_delete_skipped_count, 1);
-              } else if (ikey_.type == kTypeBlobIndex) {
-                if (!allow_blob_) {
-                  ROCKS_LOG_ERROR(logger_, "Encounter unexpected blob index.");
-                  status_ = Status::NotSupported(
-                      "Encounter unexpected blob index. Please open DB with "
-                      "ROCKSDB_NAMESPACE::blob_db::BlobDB instead.");
-                  valid_ = false;
-                  return false;
+              } else {
+                if (ikey_.type == kTypeBlobIndex) {
+                  if (!SetBlobValueIfNeeded(ikey_.user_key, iter_.value())) {
+                    return false;
+                  }
                 }
 
-                is_blob_ = true;
-                valid_ = true;
-                return true;
-              } else {
                 valid_ = true;
                 return true;
               }
@@ -436,11 +488,11 @@ bool DBIter::FindNextUserEntryInternal(bool skipping_saved_key,
               &last_key,
               ParsedInternalKey(saved_key_.GetUserKey(), 0, kTypeDeletion));
         } else {
-          std::string min_ts(timestamp_size_, static_cast<char>(0));
+          const std::string kTsMin(timestamp_size_, '\0');
           AppendInternalKeyWithDifferentTimestamp(
               &last_key,
               ParsedInternalKey(saved_key_.GetUserKey(), 0, kTypeDeletion),
-              min_ts);
+              kTsMin);
         }
         // Don't set skipping_saved_key = false because we may still see more
         // user-keys equal to saved_key_.
@@ -497,7 +549,6 @@ bool DBIter::MergeValuesNewToOld() {
   TEST_SYNC_POINT("DBIter::MergeValuesNewToOld:PushedFirstOperand");
 
   ParsedInternalKey ikey;
-  Status s;
   for (iter_.Next(); iter_.Valid(); iter_.Next()) {
     TEST_SYNC_POINT("DBIter::MergeValuesNewToOld:SteppedToNextOperand");
     if (!ParseKey(&ikey)) {
@@ -525,9 +576,9 @@ bool DBIter::MergeValuesNewToOld() {
       // hit a put, merge the put value with operands and store the
       // final result in saved_value_. We are done!
       const Slice val = iter_.value();
-      s = MergeHelper::TimedFullMerge(
+      Status s = MergeHelper::TimedFullMerge(
           merge_operator_, ikey.user_key, &val, merge_context_.GetOperands(),
-          &saved_value_, logger_, statistics_, env_, &pinned_value_, true);
+          &saved_value_, logger_, statistics_, clock_, &pinned_value_, true);
       if (!s.ok()) {
         valid_ = false;
         status_ = s;
@@ -547,15 +598,7 @@ bool DBIter::MergeValuesNewToOld() {
           iter_.value(), iter_.iter()->IsValuePinned() /* operand_pinned */);
       PERF_COUNTER_ADD(internal_merge_count, 1);
     } else if (kTypeBlobIndex == ikey.type) {
-      if (!allow_blob_) {
-        ROCKS_LOG_ERROR(logger_, "Encounter unexpected blob index.");
-        status_ = Status::NotSupported(
-            "Encounter unexpected blob index. Please open DB with "
-            "ROCKSDB_NAMESPACE::blob_db::BlobDB instead.");
-      } else {
-        status_ =
-            Status::NotSupported("Blob DB does not support merge operator.");
-      }
+      status_ = Status::NotSupported("BlobDB does not support merge operator.");
       valid_ = false;
       return false;
     } else {
@@ -576,10 +619,10 @@ bool DBIter::MergeValuesNewToOld() {
   // a deletion marker.
   // feed null as the existing value to the merge operator, such that
   // client can differentiate this scenario and do things accordingly.
-  s = MergeHelper::TimedFullMerge(merge_operator_, saved_key_.GetUserKey(),
-                                  nullptr, merge_context_.GetOperands(),
-                                  &saved_value_, logger_, statistics_, env_,
-                                  &pinned_value_, true);
+  Status s = MergeHelper::TimedFullMerge(
+      merge_operator_, saved_key_.GetUserKey(), nullptr,
+      merge_context_.GetOperands(), &saved_value_, logger_, statistics_, clock_,
+      &pinned_value_, true);
   if (!s.ok()) {
     valid_ = false;
     status_ = s;
@@ -591,17 +634,10 @@ bool DBIter::MergeValuesNewToOld() {
 }
 
 void DBIter::Prev() {
-  if (timestamp_size_ > 0) {
-    valid_ = false;
-    status_ = Status::NotSupported(
-        "SeekToLast/SeekForPrev/Prev currently not supported with timestamp.");
-    return;
-  }
-
   assert(valid_);
   assert(status_.ok());
 
-  PERF_CPU_TIMER_GUARD(iter_prev_cpu_nanos, env_);
+  PERF_CPU_TIMER_GUARD(iter_prev_cpu_nanos, clock_);
   ReleaseTempPinnedData();
   ResetInternalKeysSkippedCounter();
   bool ok = true;
@@ -636,9 +672,16 @@ bool DBIter::ReverseToForward() {
   // If that's the case, seek iter_ to current key.
   if (!expect_total_order_inner_iter() || !iter_.Valid()) {
     IterKey last_key;
-    last_key.SetInternalKey(ParsedInternalKey(
-        saved_key_.GetUserKey(), kMaxSequenceNumber, kValueTypeForSeek));
+    ParsedInternalKey pikey(saved_key_.GetUserKey(), kMaxSequenceNumber,
+                            kValueTypeForSeek);
+    if (timestamp_size_ > 0) {
+      // TODO: pre-create kTsMax.
+      const std::string kTsMax(timestamp_size_, '\xff');
+      pikey.SetTimestamp(kTsMax);
+    }
+    last_key.SetInternalKey(pikey);
     iter_.Seek(last_key.GetInternalKey());
+    RecordTick(statistics_, NUMBER_OF_RESEEKS_IN_ITERATION);
   }
 
   direction_ = kForward;
@@ -689,6 +732,7 @@ bool DBIter::ReverseToBackward() {
         iter_.SeekToLast();
       }
     }
+    RecordTick(statistics_, NUMBER_OF_RESEEKS_IN_ITERATION);
   }
 
   direction_ = kReverse;
@@ -703,7 +747,9 @@ void DBIter::PrevInternal(const Slice* prefix) {
 
     assert(prefix == nullptr || prefix_extractor_ != nullptr);
     if (prefix != nullptr &&
-        prefix_extractor_->Transform(saved_key_.GetUserKey())
+        prefix_extractor_
+                ->Transform(StripTimestampFromUserKey(saved_key_.GetUserKey(),
+                                                      timestamp_size_))
                 .compare(*prefix) != 0) {
       assert(prefix_same_as_start_);
       // Current key does not have the same prefix as start
@@ -712,11 +758,13 @@ void DBIter::PrevInternal(const Slice* prefix) {
     }
 
     assert(iterate_lower_bound_ == nullptr || iter_.MayBeOutOfLowerBound() ||
-           user_comparator_.Compare(saved_key_.GetUserKey(),
-                                    *iterate_lower_bound_) >= 0);
+           user_comparator_.CompareWithoutTimestamp(
+               saved_key_.GetUserKey(), /*a_has_ts=*/true,
+               *iterate_lower_bound_, /*b_has_ts=*/false) >= 0);
     if (iterate_lower_bound_ != nullptr && iter_.MayBeOutOfLowerBound() &&
-        user_comparator_.Compare(saved_key_.GetUserKey(),
-                                 *iterate_lower_bound_) < 0) {
+        user_comparator_.CompareWithoutTimestamp(
+            saved_key_.GetUserKey(), /*a_has_ts=*/true, *iterate_lower_bound_,
+            /*b_has_ts=*/false) < 0) {
       // We've iterated earlier than the user-specified lower bound.
       valid_ = false;
       return;
@@ -761,8 +809,8 @@ bool DBIter::FindValueForCurrentKey() {
   assert(iter_.Valid());
   merge_context_.Clear();
   current_entry_is_merged_ = false;
-  // last entry before merge (could be kTypeDeletion, kTypeSingleDeletion or
-  // kTypeValue)
+  // last entry before merge (could be kTypeDeletion,
+  // kTypeDeletionWithTimestamp, kTypeSingleDeletion or kTypeValue)
   ValueType last_not_merge_type = kTypeDeletion;
   ValueType last_key_entry_type = kTypeDeletion;
 
@@ -783,9 +831,13 @@ bool DBIter::FindValueForCurrentKey() {
                  timestamp_size_);
     }
     if (!IsVisible(ikey.sequence, ts) ||
-        !user_comparator_.Equal(ikey.user_key, saved_key_.GetUserKey())) {
+        !user_comparator_.EqualWithoutTimestamp(ikey.user_key,
+                                                saved_key_.GetUserKey())) {
       break;
     }
+    if (!ts.empty()) {
+      saved_timestamp_.assign(ts.data(), ts.size());
+    }
     if (TooManyInternalKeysSkipped()) {
       return false;
     }
@@ -810,14 +862,22 @@ bool DBIter::FindValueForCurrentKey() {
                 ikey, RangeDelPositioningMode::kBackwardTraversal)) {
           last_key_entry_type = kTypeRangeDeletion;
           PERF_COUNTER_ADD(internal_delete_skipped_count, 1);
-        } else {
-          assert(iter_.iter()->IsValuePinned());
+        } else if (iter_.iter()->IsValuePinned()) {
           pinned_value_ = iter_.value();
+        } else {
+          valid_ = false;
+          status_ = Status::NotSupported(
+              "Backward iteration not supported if underlying iterator's value "
+              "cannot be pinned.");
         }
         merge_context_.Clear();
         last_not_merge_type = last_key_entry_type;
+        if (!status_.ok()) {
+          return false;
+        }
         break;
       case kTypeDeletion:
+      case kTypeDeletionWithTimestamp:
       case kTypeSingleDeletion:
         merge_context_.Clear();
         last_not_merge_type = last_key_entry_type;
@@ -861,6 +921,7 @@ bool DBIter::FindValueForCurrentKey() {
   is_blob_ = false;
   switch (last_key_entry_type) {
     case kTypeDeletion:
+    case kTypeDeletionWithTimestamp:
     case kTypeSingleDeletion:
     case kTypeRangeDeletion:
       valid_ = false;
@@ -873,17 +934,10 @@ bool DBIter::FindValueForCurrentKey() {
         s = MergeHelper::TimedFullMerge(
             merge_operator_, saved_key_.GetUserKey(), nullptr,
             merge_context_.GetOperands(), &saved_value_, logger_, statistics_,
-            env_, &pinned_value_, true);
+            clock_, &pinned_value_, true);
       } else if (last_not_merge_type == kTypeBlobIndex) {
-        if (!allow_blob_) {
-          ROCKS_LOG_ERROR(logger_, "Encounter unexpected blob index.");
-          status_ = Status::NotSupported(
-              "Encounter unexpected blob index. Please open DB with "
-              "ROCKSDB_NAMESPACE::blob_db::BlobDB instead.");
-        } else {
-          status_ =
-              Status::NotSupported("Blob DB does not support merge operator.");
-        }
+        status_ =
+            Status::NotSupported("BlobDB does not support merge operator.");
         valid_ = false;
         return false;
       } else {
@@ -891,22 +945,17 @@ bool DBIter::FindValueForCurrentKey() {
         s = MergeHelper::TimedFullMerge(
             merge_operator_, saved_key_.GetUserKey(), &pinned_value_,
             merge_context_.GetOperands(), &saved_value_, logger_, statistics_,
-            env_, &pinned_value_, true);
+            clock_, &pinned_value_, true);
       }
       break;
     case kTypeValue:
       // do nothing - we've already has value in pinned_value_
       break;
     case kTypeBlobIndex:
-      if (!allow_blob_) {
-        ROCKS_LOG_ERROR(logger_, "Encounter unexpected blob index.");
-        status_ = Status::NotSupported(
-            "Encounter unexpected blob index. Please open DB with "
-            "ROCKSDB_NAMESPACE::blob_db::BlobDB instead.");
-        valid_ = false;
+      if (!SetBlobValueIfNeeded(saved_key_.GetUserKey(), pinned_value_)) {
         return false;
       }
-      is_blob_ = true;
+
       break;
     default:
       valid_ = false;
@@ -933,8 +982,17 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
   // FindValueForCurrentKeyUsingSeek()
   assert(pinned_iters_mgr_.PinningEnabled());
   std::string last_key;
-  AppendInternalKey(&last_key, ParsedInternalKey(saved_key_.GetUserKey(),
-                                                 sequence_, kValueTypeForSeek));
+  if (0 == timestamp_size_) {
+    AppendInternalKey(&last_key,
+                      ParsedInternalKey(saved_key_.GetUserKey(), sequence_,
+                                        kValueTypeForSeek));
+  } else {
+    AppendInternalKeyWithDifferentTimestamp(
+        &last_key,
+        ParsedInternalKey(saved_key_.GetUserKey(), sequence_,
+                          kValueTypeForSeek),
+        *timestamp_ub_);
+  }
   iter_.Seek(last_key);
   RecordTick(statistics_, NUMBER_OF_RESEEKS_IN_ITERATION);
 
@@ -958,7 +1016,8 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
                  timestamp_size_);
     }
 
-    if (!user_comparator_.Equal(ikey.user_key, saved_key_.GetUserKey())) {
+    if (!user_comparator_.EqualWithoutTimestamp(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.
@@ -975,26 +1034,28 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
 
   if (ikey.type == kTypeDeletion || ikey.type == kTypeSingleDeletion ||
       range_del_agg_.ShouldDelete(
-          ikey, RangeDelPositioningMode::kBackwardTraversal)) {
+          ikey, RangeDelPositioningMode::kBackwardTraversal) ||
+      kTypeDeletionWithTimestamp == ikey.type) {
     valid_ = false;
     return true;
   }
-  if (ikey.type == kTypeBlobIndex && !allow_blob_) {
-    ROCKS_LOG_ERROR(logger_, "Encounter unexpected blob index.");
-    status_ = Status::NotSupported(
-        "Encounter unexpected blob index. Please open DB with "
-        "ROCKSDB_NAMESPACE::blob_db::BlobDB instead.");
-    valid_ = false;
-    return false;
-  }
   if (!iter_.PrepareValue()) {
     valid_ = false;
     return false;
   }
+  if (timestamp_size_ > 0) {
+    Slice ts = ExtractTimestampFromUserKey(ikey.user_key, timestamp_size_);
+    saved_timestamp_.assign(ts.data(), ts.size());
+  }
   if (ikey.type == kTypeValue || ikey.type == kTypeBlobIndex) {
     assert(iter_.iter()->IsValuePinned());
     pinned_value_ = iter_.value();
-    is_blob_ = (ikey.type == kTypeBlobIndex);
+    if (ikey.type == kTypeBlobIndex) {
+      if (!SetBlobValueIfNeeded(ikey.user_key, pinned_value_)) {
+        return false;
+      }
+    }
+
     valid_ = true;
     return true;
   }
@@ -1037,7 +1098,7 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
       Status s = MergeHelper::TimedFullMerge(
           merge_operator_, saved_key_.GetUserKey(), &val,
           merge_context_.GetOperands(), &saved_value_, logger_, statistics_,
-          env_, &pinned_value_, true);
+          clock_, &pinned_value_, true);
       if (!s.ok()) {
         valid_ = false;
         status_ = s;
@@ -1050,15 +1111,7 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
           iter_.value(), iter_.iter()->IsValuePinned() /* operand_pinned */);
       PERF_COUNTER_ADD(internal_merge_count, 1);
     } else if (ikey.type == kTypeBlobIndex) {
-      if (!allow_blob_) {
-        ROCKS_LOG_ERROR(logger_, "Encounter unexpected blob index.");
-        status_ = Status::NotSupported(
-            "Encounter unexpected blob index. Please open DB with "
-            "ROCKSDB_NAMESPACE::blob_db::BlobDB instead.");
-      } else {
-        status_ =
-            Status::NotSupported("Blob DB does not support merge operator.");
-      }
+      status_ = Status::NotSupported("BlobDB does not support merge operator.");
       valid_ = false;
       return false;
     } else {
@@ -1072,7 +1125,7 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
 
   Status s = MergeHelper::TimedFullMerge(
       merge_operator_, saved_key_.GetUserKey(), nullptr,
-      merge_context_.GetOperands(), &saved_value_, logger_, statistics_, env_,
+      merge_context_.GetOperands(), &saved_value_, logger_, statistics_, clock_,
       &pinned_value_, true);
   if (!s.ok()) {
     valid_ = false;
@@ -1110,7 +1163,8 @@ bool DBIter::FindUserKeyBeforeSavedKey() {
       return false;
     }
 
-    if (user_comparator_.Compare(ikey.user_key, saved_key_.GetUserKey()) < 0) {
+    if (user_comparator_.CompareWithoutTimestamp(ikey.user_key,
+                                                 saved_key_.GetUserKey()) < 0) {
       return true;
     }
 
@@ -1134,8 +1188,14 @@ bool DBIter::FindUserKeyBeforeSavedKey() {
     if (num_skipped >= max_skip_) {
       num_skipped = 0;
       IterKey last_key;
-      last_key.SetInternalKey(ParsedInternalKey(
-          saved_key_.GetUserKey(), kMaxSequenceNumber, kValueTypeForSeek));
+      ParsedInternalKey pikey(saved_key_.GetUserKey(), kMaxSequenceNumber,
+                              kValueTypeForSeek);
+      if (timestamp_size_ > 0) {
+        // TODO: pre-create kTsMax.
+        const std::string kTsMax(timestamp_size_, '\xff');
+        pikey.SetTimestamp(kTsMax);
+      }
+      last_key.SetInternalKey(pikey);
       // It would be more efficient to use SeekForPrev() here, but some
       // iterators may not support it.
       iter_.Seek(last_key.GetInternalKey());
@@ -1212,24 +1272,50 @@ void DBIter::SetSavedKeyToSeekForPrevTarget(const Slice& target) {
   saved_key_.Clear();
   // now saved_key is used to store internal key.
   saved_key_.SetInternalKey(target, 0 /* sequence_number */,
-                            kValueTypeForSeekForPrev);
+                            kValueTypeForSeekForPrev, timestamp_ub_);
+
+  if (timestamp_size_ > 0) {
+    const std::string kTsMin(timestamp_size_, '\0');
+    Slice ts = kTsMin;
+    saved_key_.UpdateInternalKey(/*seq=*/0, kValueTypeForSeekForPrev, &ts);
+  }
 
   if (iterate_upper_bound_ != nullptr &&
-      user_comparator_.Compare(saved_key_.GetUserKey(),
-                               *iterate_upper_bound_) >= 0) {
+      user_comparator_.CompareWithoutTimestamp(
+          saved_key_.GetUserKey(), /*a_has_ts=*/true, *iterate_upper_bound_,
+          /*b_has_ts=*/false) >= 0) {
     saved_key_.Clear();
-    saved_key_.SetInternalKey(*iterate_upper_bound_, kMaxSequenceNumber);
+    saved_key_.SetInternalKey(*iterate_upper_bound_, kMaxSequenceNumber,
+                              kValueTypeForSeekForPrev, timestamp_ub_);
+    if (timestamp_size_ > 0) {
+      const std::string kTsMax(timestamp_size_, '\xff');
+      Slice ts = kTsMax;
+      saved_key_.UpdateInternalKey(kMaxSequenceNumber, kValueTypeForSeekForPrev,
+                                   &ts);
+    }
   }
 }
 
 void DBIter::Seek(const Slice& target) {
-  PERF_CPU_TIMER_GUARD(iter_seek_cpu_nanos, env_);
-  StopWatch sw(env_, statistics_, DB_SEEK);
+  PERF_CPU_TIMER_GUARD(iter_seek_cpu_nanos, clock_);
+  StopWatch sw(clock_, statistics_, DB_SEEK);
 
 #ifndef ROCKSDB_LITE
   if (db_impl_ != nullptr && cfd_ != nullptr) {
     // TODO: What do we do if this returns an error?
-    db_impl_->TraceIteratorSeek(cfd_->GetID(), target).PermitUncheckedError();
+    Slice lower_bound, upper_bound;
+    if (iterate_lower_bound_ != nullptr) {
+      lower_bound = *iterate_lower_bound_;
+    } else {
+      lower_bound = Slice("");
+    }
+    if (iterate_upper_bound_ != nullptr) {
+      upper_bound = *iterate_upper_bound_;
+    } else {
+      upper_bound = Slice("");
+    }
+    db_impl_->TraceIteratorSeek(cfd_->GetID(), target, lower_bound, upper_bound)
+        .PermitUncheckedError();
   }
 #endif  // ROCKSDB_LITE
 
@@ -1257,7 +1343,7 @@ void DBIter::Seek(const Slice& target) {
   // we need to find out the next key that is visible to the user.
   ClearSavedValue();
   if (prefix_same_as_start_) {
-    // The case where the iterator needs to be invalidated if it has exausted
+    // The case where the iterator needs to be invalidated if it has exhausted
     // keys within the same prefix of the seek key.
     assert(prefix_extractor_ != nullptr);
     Slice target_prefix = prefix_extractor_->Transform(target);
@@ -1285,24 +1371,30 @@ 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);
+  PERF_CPU_TIMER_GUARD(iter_seek_cpu_nanos, clock_);
+  StopWatch sw(clock_, statistics_, DB_SEEK);
 
 #ifndef ROCKSDB_LITE
   if (db_impl_ != nullptr && cfd_ != nullptr) {
     // TODO: What do we do if this returns an error?
-    db_impl_->TraceIteratorSeekForPrev(cfd_->GetID(), target)
+    Slice lower_bound, upper_bound;
+    if (iterate_lower_bound_ != nullptr) {
+      lower_bound = *iterate_lower_bound_;
+    } else {
+      lower_bound = Slice("");
+    }
+    if (iterate_upper_bound_ != nullptr) {
+      upper_bound = *iterate_upper_bound_;
+    } else {
+      upper_bound = Slice("");
+    }
+    db_impl_
+        ->TraceIteratorSeekForPrev(cfd_->GetID(), target, lower_bound,
+                                   upper_bound)
         .PermitUncheckedError();
   }
 #endif  // ROCKSDB_LITE
 
-  if (timestamp_size_ > 0) {
-    valid_ = false;
-    status_ = Status::NotSupported(
-        "SeekToLast/SeekForPrev/Prev currently not supported with timestamp.");
-    return;
-  }
-
   status_ = Status::OK();
   ReleaseTempPinnedData();
   ResetInternalKeysSkippedCounter();
@@ -1326,7 +1418,7 @@ void DBIter::SeekForPrev(const Slice& target) {
   // backward direction.
   ClearSavedValue();
   if (prefix_same_as_start_) {
-    // The case where the iterator needs to be invalidated if it has exausted
+    // The case where the iterator needs to be invalidated if it has exhausted
     // keys within the same prefix of the seek key.
     assert(prefix_extractor_ != nullptr);
     Slice target_prefix = prefix_extractor_->Transform(target);
@@ -1353,7 +1445,7 @@ void DBIter::SeekToFirst() {
     Seek(*iterate_lower_bound_);
     return;
   }
-  PERF_CPU_TIMER_GUARD(iter_seek_cpu_nanos, env_);
+  PERF_CPU_TIMER_GUARD(iter_seek_cpu_nanos, clock_);
   // Don't use iter_::Seek() if we set a prefix extractor
   // because prefix seek will be used.
   if (!expect_total_order_inner_iter()) {
@@ -1391,29 +1483,25 @@ void DBIter::SeekToFirst() {
   }
   if (valid_ && prefix_same_as_start_) {
     assert(prefix_extractor_ != nullptr);
-    prefix_.SetUserKey(prefix_extractor_->Transform(saved_key_.GetUserKey()));
+    prefix_.SetUserKey(prefix_extractor_->Transform(
+        StripTimestampFromUserKey(saved_key_.GetUserKey(), timestamp_size_)));
   }
 }
 
 void DBIter::SeekToLast() {
-  if (timestamp_size_ > 0) {
-    valid_ = false;
-    status_ = Status::NotSupported(
-        "SeekToLast/SeekForPrev/Prev currently not supported with timestamp.");
-    return;
-  }
-
   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() && 0 == user_comparator_.CompareWithoutTimestamp(
+                            *iterate_upper_bound_, /*a_has_ts=*/false, key(),
+                            /*b_has_ts=*/false)) {
       ReleaseTempPinnedData();
       PrevInternal(nullptr);
     }
     return;
   }
 
-  PERF_CPU_TIMER_GUARD(iter_seek_cpu_nanos, env_);
+  PERF_CPU_TIMER_GUARD(iter_seek_cpu_nanos, clock_);
   // Don't use iter_::Seek() if we set a prefix extractor
   // because prefix seek will be used.
   if (!expect_total_order_inner_iter()) {
@@ -1442,7 +1530,8 @@ void DBIter::SeekToLast() {
   }
   if (valid_ && prefix_same_as_start_) {
     assert(prefix_extractor_ != nullptr);
-    prefix_.SetUserKey(prefix_extractor_->Transform(saved_key_.GetUserKey()));
+    prefix_.SetUserKey(prefix_extractor_->Transform(
+        StripTimestampFromUserKey(saved_key_.GetUserKey(), timestamp_size_)));
   }
 }
 
@@ -1450,15 +1539,16 @@ Iterator* NewDBIterator(Env* env, const ReadOptions& read_options,
                         const ImmutableCFOptions& cf_options,
                         const MutableCFOptions& mutable_cf_options,
                         const Comparator* user_key_comparator,
-                        InternalIterator* internal_iter,
+                        InternalIterator* internal_iter, const Version* version,
                         const SequenceNumber& sequence,
                         uint64_t max_sequential_skip_in_iterations,
                         ReadCallback* read_callback, DBImpl* db_impl,
-                        ColumnFamilyData* cfd, bool allow_blob) {
-  DBIter* db_iter = new DBIter(
-      env, read_options, cf_options, mutable_cf_options, user_key_comparator,
-      internal_iter, sequence, false, max_sequential_skip_in_iterations,
-      read_callback, db_impl, cfd, allow_blob);
+                        ColumnFamilyData* cfd, bool expose_blob_index) {
+  DBIter* db_iter =
+      new DBIter(env, read_options, cf_options, mutable_cf_options,
+                 user_key_comparator, internal_iter, version, sequence, false,
+                 max_sequential_skip_in_iterations, read_callback, db_impl, cfd,
+                 expose_blob_index);
   return db_iter;
 }