Integrated BlobDB for backup/restore support (#8129)

Summary:
Add support for blob files for backup/restore like table files.
    Since DB session ID is currently not supported for blob files (there is no place to store it in
    the header), so for blob files uses the
    kLegacyCrc32cAndFileSize naming scheme even if
    share_files_with_checksum_naming is set to kUseDbSessionId.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8129

Test Plan: Add new test units

Reviewed By: ltamasi

Differential Revision: D27408510

Pulled By: akankshamahajan15

fbshipit-source-id: b27434d189a639ef3e6ad165c61a143a2daaf06e

.circleci/cat_ignore_eagain

@@ -0,0 +1,54 @@
+#! /bin/bash
+
+# Work around issue with parallel make output causing random error, as in
+# make[1]: write error: stdout
+# Probably due to a kernel bug:
+#   https://bugs.launchpad.net/ubuntu/+source/linux-signed/+bug/1814393
+# Seems to affect image ubuntu-1604:201903-01 and ubuntu-1604:202004-01
+
+cd "$(dirname $0)"
+
+if [ ! -x cat_ignore_eagain.out ]; then
+    cc -x c -o cat_ignore_eagain.out - << EOF
+#include <unistd.h>
+#include <errno.h>
+#include <stdio.h>
+int main() {
+    int n, m, p;
+    char buf[1024];
+    for (;;) {
+        n = read(STDIN_FILENO, buf, 1024);
+        if (n > 0 && n <= 1024) {
+            for (m = 0; m < n;) {
+                p = write(STDOUT_FILENO, buf + m, n - m);
+                if (p < 0) {
+                    if (errno == EAGAIN) {
+                        // ignore but pause a bit
+                        usleep(100);
+                    } else {
+                        perror("write failed");
+                        return 42;
+                    }
+                } else {
+                    m += p;
+                }
+            }
+        } else if (n < 0) {
+            if (errno == EAGAIN) {
+                // ignore but pause a bit
+                usleep(100);
+            } else {
+                // Some non-ignorable error
+                perror("read failed");
+                return 43;
+            }
+        } else {
+            // EOF
+            return 0;
+        }
+    }
+}
+EOF
+fi
+
+exec ./cat_ignore_eagain.out

.circleci/config.yml

@@ -0,0 +1,703 @@
+version: 2.1
+
+orbs:
+  win: circleci/windows@2.4.0
+  slack: circleci/slack@3.4.2
+
+aliases:
+  - &notify-on-master-failure
+    fail_only: true
+    only_for_branches: master
+
+commands:
+  install-pyenv-on-macos:
+    steps:
+      - run:
+          name: Install pyenv on macos
+          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:
+          name: Increase max open files
+          command: |
+            sudo sysctl -w kern.maxfiles=1048576
+            sudo sysctl -w kern.maxfilesperproc=1048576
+            sudo launchctl limit maxfiles 1048576
+
+  pre-steps:
+    parameters:
+      python-version:
+        default: "3.5.9"
+        type: string
+    steps:
+      - checkout
+      - run: pyenv install --skip-existing <<parameters.python-version>>
+      - run: pyenv global <<parameters.python-version>>
+      - run:
+          name: Setup Environment Variables
+          command: |
+            echo "export GTEST_THROW_ON_FAILURE=0" >> $BASH_ENV
+            echo "export GTEST_OUTPUT=\"xml:/tmp/test-results/\"" >> $BASH_ENV
+            echo "export SKIP_FORMAT_BUCK_CHECKS=1" >> $BASH_ENV
+            echo "export PRINT_PARALLEL_OUTPUTS=1" >> $BASH_ENV
+
+  pre-steps-macos:
+      steps:
+        - pre-steps:
+            python-version: "3.6.0"
+
+  post-steps:
+    steps:
+      - slack/status: *notify-on-master-failure
+      - store_test_results: # store test result if there's any
+          path: /tmp/test-results
+      - store_artifacts: # store LOG for debugging if there's any
+          path: LOG
+      - run: # on fail, compress Test Logs for diagnosing the issue
+         name: Compress Test Logs
+         command: tar -cvzf t.tar.gz t
+         when: on_fail
+      - store_artifacts: # on fail, store Test Logs for diagnosing the issue
+          path: t.tar.gz
+          destination: test_logs
+          when: on_fail
+
+  install-clang-10:
+    steps:
+      - run:
+          name: Install Clang 10
+          command: |
+            echo "deb http://apt.llvm.org/xenial/ llvm-toolchain-xenial-10 main" | sudo tee -a /etc/apt/sources.list
+            echo "deb-src http://apt.llvm.org/xenial/ llvm-toolchain-xenial-10 main" | sudo tee -a /etc/apt/sources.list
+            echo "APT::Acquire::Retries \"10\";" | sudo tee -a /etc/apt/apt.conf.d/80-retries # llvm.org unreliable
+            sudo apt-get update -y && sudo apt-get install -y clang-10
+
+  install-gflags:
+    steps:
+      - run:
+          name: Install gflags
+          command: |
+            sudo apt-get update -y && sudo apt-get install -y libgflags-dev
+
+  install-gflags-on-macos:
+    steps:
+      - run:
+          name: Install gflags on macos
+          command: |
+            HOMEBREW_NO_AUTO_UPDATE=1 brew install gflags
+
+  install-gtest-parallel:
+    steps:
+      - run:
+          name: Install gtest-parallel
+          command: |
+            git clone --single-branch --branch master --depth 1 https://github.com/google/gtest-parallel.git ~/gtest-parallel
+            echo 'export PATH=$HOME/gtest-parallel:$PATH' >> $BASH_ENV
+
+  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:
+    machine:
+      image: 'windows-server-2019-vs2019:stable'
+      resource_class: windows.2xlarge
+      shell: bash.exe
+
+jobs:
+  build-macos:
+    macos:
+      xcode: 9.4.1
+    steps:
+      - increase-max-open-files-on-macos
+      - install-pyenv-on-macos
+      - 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
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - 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
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - 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: large
+    steps:
+      - checkout # check out the code in the project directory
+      - 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 -j8 release | .circleci/cat_ignore_eagain
+      - run: ./db_stress --version # ensure with gflags
+      - post-steps
+
+  build-linux-release-rtti:
+    machine:
+      image: ubuntu-1604:201903-01
+    resource_class: large
+    steps:
+      - checkout # check out the code in the project directory
+      - run: make clean
+      - 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 -j8 static_lib tools db_bench | .circleci/cat_ignore_eagain
+      - run: ./db_stress --version # ensure with gflags
+
+  build-linux-lite:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - run: LITE=1 make V=1 J=32 -j32 check | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-linux-lite-release:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: large
+    steps:
+      - checkout # check out the code in the project directory
+      - 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 -j8 release | .circleci/cat_ignore_eagain
+      - run: ./db_stress --version # ensure with gflags
+      - post-steps
+
+  build-linux-clang-no_test_run:
+    machine:
+      image: ubuntu-1604:202007-01
+    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 -j16 all | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-linux-clang10-asan:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - install-clang-10
+      - run: COMPILE_WITH_ASAN=1 CC=clang-10 CXX=clang++-10 ROCKSDB_DISABLE_ALIGNED_NEW=1 USE_CLANG=1 make V=1 -j32 check | .circleci/cat_ignore_eagain # aligned new doesn't work for reason we haven't figured out
+      - post-steps
+
+  build-linux-clang10-mini-tsan:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - install-clang-10
+      - run: COMPILE_WITH_TSAN=1 CC=clang-10 CXX=clang++-10 ROCKSDB_DISABLE_ALIGNED_NEW=1 USE_CLANG=1 make V=1 -j32 check | .circleci/cat_ignore_eagain # aligned new doesn't work for reason we haven't figured out.
+      - post-steps
+
+  build-linux-clang10-ubsan:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - install-clang-10
+      - run: COMPILE_WITH_UBSAN=1 OPT="-fsanitize-blacklist=.circleci/ubsan_suppression_list.txt" CC=clang-10 CXX=clang++-10 ROCKSDB_DISABLE_ALIGNED_NEW=1 USE_CLANG=1 make V=1 -j32 ubsan_check | .circleci/cat_ignore_eagain # aligned new doesn't work for reason we haven't figured out
+      - post-steps
+
+  build-linux-clang10-clang-analyze:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - install-clang-10
+      - run: sudo apt-get update -y && sudo apt-get install -y clang-tools-10
+      - 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-no_test_run:
+    machine:
+      image: ubuntu-1604:202007-01
+    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 -j8) | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-linux-unity:
+    docker: # executor type
+      - image: gcc:latest
+    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 -j8 unity_test | .circleci/cat_ignore_eagain
+      - post-steps
+
+  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 -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:
+    executor: windows-2xlarge
+    parameters:
+      extra_cmake_opt:
+        default: ""
+        type: string
+      vs_year:
+        default: "2019"
+        type: string
+      cmake_generator:
+        default: "Visual Studio 16 2019"
+        type: string
+    environment:
+      THIRDPARTY_HOME: C:/Users/circleci/thirdparty
+      CMAKE_HOME: C:/Users/circleci/thirdparty/cmake-3.16.4-win64-x64
+      CMAKE_BIN: C:/Users/circleci/thirdparty/cmake-3.16.4-win64-x64/bin/cmake.exe
+      SNAPPY_HOME: C:/Users/circleci/thirdparty/snappy-1.1.7
+      SNAPPY_INCLUDE: C:/Users/circleci/thirdparty/snappy-1.1.7;C:/Users/circleci/thirdparty/snappy-1.1.7/build
+      SNAPPY_LIB_DEBUG: C:/Users/circleci/thirdparty/snappy-1.1.7/build/Debug/snappy.lib
+      VS_YEAR: <<parameters.vs_year>>
+      CMAKE_GENERATOR: <<parameters.cmake_generator>>
+    steps:
+      - checkout
+      - run:
+          name: "Setup VS"
+          command: |
+            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:
+          name: "Install thirdparty dependencies"
+          command: |
+            mkdir ${THIRDPARTY_HOME}
+            cd ${THIRDPARTY_HOME}
+            echo "Installing CMake..."
+            curl --fail --silent --show-error --output cmake-3.16.4-win64-x64.zip --location https://github.com/Kitware/CMake/releases/download/v3.16.4/cmake-3.16.4-win64-x64.zip
+            unzip -q cmake-3.16.4-win64-x64.zip
+            echo "Building Snappy dependency..."
+            curl --fail --silent --show-error --output snappy-1.1.7.zip --location https://github.com/google/snappy/archive/1.1.7.zip
+            unzip -q snappy-1.1.7.zip
+            cd snappy-1.1.7
+            mkdir build
+            cd build
+            ${CMAKE_BIN} -G "${CMAKE_GENERATOR}" ..
+            msbuild.exe Snappy.sln -maxCpuCount -property:Configuration=Debug -property:Platform=x64
+      - run:
+          name: "Build RocksDB"
+          command: |
+            mkdir build
+            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: ${CMAKE_GENERATOR}"
+            msbuild.exe build/rocksdb.sln -maxCpuCount -property:Configuration=Debug -property:Platform=x64
+      - run:
+          name: "Test RocksDB"
+          shell: powershell.exe
+          command: |
+            build_tools\run_ci_db_test.ps1 -SuiteRun db_basic_test,db_test,db_test2,db_merge_operand_test,bloom_test,c_test,coding_test,crc32c_test,dynamic_bloom_test,env_basic_test,env_test,hash_test,random_test -Concurrency 16
+
+  build-linux-java:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: large
+    environment:
+      JAVA_HOME: /usr/lib/jvm/java-1.8.0-openjdk-amd64
+    steps:
+      - pre-steps
+      - install-gflags
+      - run:
+          name: "Set Java Environment"
+          command: |
+            echo "JAVA_HOME=${JAVA_HOME}"
+            echo 'export PATH=$JAVA_HOME/bin:$PATH' >> $BASH_ENV
+            which java && java -version
+            which javac && javac -version
+      - run:
+          name: "Build RocksDBJava Shared Library"
+          command: make V=1 J=8 -j8 rocksdbjava | .circleci/cat_ignore_eagain
+      - run:
+          name: "Test RocksDBJava"
+          command: make V=1 J=8 -j8 jtest | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-linux-java-static:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: large
+    environment:
+      JAVA_HOME: /usr/lib/jvm/java-1.8.0-openjdk-amd64
+    steps:
+      - pre-steps
+      - install-gflags
+      - run:
+          name: "Set Java Environment"
+          command: |
+            echo "JAVA_HOME=${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' >> $BASH_ENV
+            which java && java -version
+            which javac && javac -version
+      - run:
+          name: "Build RocksDBJava Static Library"
+          command: make V=1 J=8 -j8 rocksdbjavastatic | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-examples:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: large
+    steps:
+      - pre-steps
+      - install-gflags
+      - run:
+          name: "Build examples"
+          command: |
+            OPT=-DTRAVIS V=1 make -j4 static_lib && cd examples && make -j4 | ../.circleci/cat_ignore_eagain
+      - post-steps
+
+  build-cmake-mingw:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - run: sudo apt-get update -y && sudo apt-get install -y mingw-w64
+      - run: sudo update-alternatives --set x86_64-w64-mingw32-g++ /usr/bin/x86_64-w64-mingw32-g++-posix
+      - run:
+          name: "Build cmake-mingw"
+          command: |
+            sudo apt-get install snapd && sudo snap install cmake --beta --classic
+            export PATH=/snap/bin:$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
+    resource_class: 2xlarge
+    parameters:
+      start_test:
+        default: ""
+        type: string
+      end_test:
+        default: ""
+        type: string
+    steps:
+      - pre-steps
+      - install-gflags
+      - install-gtest-parallel
+      - run:
+          name: "Build unit tests"
+          command: |
+            echo "env: $(env)"
+            echo "** done env"
+            ROCKSDBTESTS_START=<<parameters.start_test>> ROCKSDBTESTS_END=<<parameters.end_test>> ROCKSDBTESTS_SUBSET_TESTS_TO_FILE=/tmp/test_list make V=1 -j32 --output-sync=target build_subset_tests
+      - run:
+          name: "Run unit tests in parallel"
+          command: |
+            sed -i 's/[[:space:]]*$//; s/ / \.\//g; s/.*/.\/&/' /tmp/test_list
+            cat /tmp/test_list
+            export TEST_TMPDIR=/tmp/rocksdb_test_tmp
+            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-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
+  build-linux-lite:
+    jobs:
+      - build-linux-lite
+  build-linux-release:
+    jobs:
+      - build-linux-release
+  build-linux-release-rtti:
+    jobs:
+      - build-linux-release-rtti
+  build-linux-lite-release:
+    jobs:
+      - build-linux-lite-release
+  build-linux-clang10-asan:
+    jobs:
+      - build-linux-clang10-asan
+  build-linux-clang10-mini-tsan:
+    jobs:
+      - build-linux-clang10-mini-tsan
+  build-linux-clang10-ubsan:
+    jobs:
+      - build-linux-clang10-ubsan
+  build-linux-clang10-clang-analyze:
+    jobs:
+      - build-linux-clang10-clang-analyze
+  build-linux-unity:
+    jobs:
+      - build-linux-unity
+  build-windows-vs2019:
+    jobs:
+      - 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-java:
+    jobs:
+      - build-linux-java
+      - build-linux-java-static
+      - build-macos-java
+      - build-macos-java-static
+  build-examples:
+    jobs:
+      - build-examples
+  build-linux-non-shm:
+    jobs:
+      - build-linux-non-shm:
+          start_test: ""
+          end_test: "db_options_test" # make sure unique in src.mk
+      - build-linux-non-shm:
+          start_test: "db_options_test" # make sure unique in src.mk
+          end_test: "filename_test" # make sure unique in src.mk
+      - build-linux-non-shm:
+          start_test: "filename_test" # make sure unique in src.mk
+          end_test: "statistics_test" # make sure unique in src.mk
+      - build-linux-non-shm:
+          start_test: "statistics_test" # make sure unique in src.mk
+          end_test: ""
+  build-linux-compilers-no_test_run:
+    jobs:
+      - 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-macos:
+    jobs:
+      - build-macos
+  build-macos-cmake:
+    jobs:
+      - build-macos-cmake
+  build-cmake-mingw:
+    jobs:
+      - build-cmake-mingw
+  nightly:
+    triggers:
+      - schedule:
+          cron: "0 0 * * *"
+          filters:
+            branches:
+              only:
+                - master
+    jobs:
+      - build-format-compatible

.circleci/ubsan_suppression_list.txt

@@ -0,0 +1,6 @@
+# Supress UBSAN warnings related to stl_tree.h, e.g.
+# UndefinedBehaviorSanitizer: undefined-behavior /usr/bin/../lib/gcc/x86_64-linux-gnu/5.4.0/../../../../include/c++/5.4.0/bits/stl_tree.h:1505:43 in 
+# /usr/bin/../lib/gcc/x86_64-linux-gnu/5.4.0/../../../../include/c++/5.4.0/bits/stl_tree.h:1505:43:
+# runtime error: upcast of address 0x000001fa8820 with insufficient space for an object of type
+# 'std::_Rb_tree_node<std::pair<const std::__cxx11::basic_string<char>, rocksdb::(anonymous namespace)::LockHoldingInfo> >'
+src:*bits/stl_tree.h

.circleci/vs2015_install.ps1

@@ -0,0 +1,24 @@
+$VS_DOWNLOAD_LINK = "https://go.microsoft.com/fwlink/?LinkId=691126"
+$COLLECT_DOWNLOAD_LINK = "https://aka.ms/vscollect.exe"
+curl.exe --retry 3 -kL $VS_DOWNLOAD_LINK --output vs_installer.exe
+if ($LASTEXITCODE -ne 0) {
+    echo "Download of the VS 2015 installer failed"
+    exit 1
+}
+$VS_INSTALL_ARGS = @("/Quiet", "/NoRestart")
+$process = Start-Process "${PWD}\vs_installer.exe" -ArgumentList $VS_INSTALL_ARGS -NoNewWindow -Wait -PassThru
+Remove-Item -Path vs_installer.exe -Force
+$exitCode = $process.ExitCode
+if (($exitCode -ne 0) -and ($exitCode -ne 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."
+        exit 1
+    }
+    Start-Process "${PWD}\Collect.exe" -NoNewWindow -Wait -PassThru
+    New-Item -Path "C:\w\build-results" -ItemType "directory" -Force
+    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

@@ -0,0 +1,35 @@
+$VS_DOWNLOAD_LINK = "https://aka.ms/vs/15/release/vs_buildtools.exe"
+$COLLECT_DOWNLOAD_LINK = "https://aka.ms/vscollect.exe"
+$VS_INSTALL_ARGS = @("--nocache","--quiet","--wait", "--add Microsoft.VisualStudio.Workload.VCTools",
+                                                     "--add Microsoft.VisualStudio.Component.VC.Tools.14.13",
+                                                     "--add Microsoft.Component.MSBuild",
+                                                     "--add Microsoft.VisualStudio.Component.Roslyn.Compiler",
+                                                     "--add Microsoft.VisualStudio.Component.TextTemplating",
+                                                     "--add Microsoft.VisualStudio.Component.VC.CoreIde",
+                                                     "--add Microsoft.VisualStudio.Component.VC.Redist.14.Latest",
+                                                     "--add Microsoft.VisualStudio.ComponentGroup.NativeDesktop.Core",
+                                                     "--add Microsoft.VisualStudio.Component.VC.Tools.x86.x64",
+                                                     "--add Microsoft.VisualStudio.ComponentGroup.NativeDesktop.Win81")
+
+curl.exe --retry 3 -kL $VS_DOWNLOAD_LINK --output vs_installer.exe
+if ($LASTEXITCODE -ne 0) {
+    echo "Download of the VS 2017 installer failed"
+    exit 1
+}
+
+$process = Start-Process "${PWD}\vs_installer.exe" -ArgumentList $VS_INSTALL_ARGS -NoNewWindow -Wait -PassThru
+Remove-Item -Path vs_installer.exe -Force
+$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]."
+    curl.exe --retry 3 -kL $COLLECT_DOWNLOAD_LINK --output Collect.exe
+    if ($LASTEXITCODE -ne 0) {
+        echo "Download of the VS Collect tool failed."
+        exit 1
+    }
+    Start-Process "${PWD}\Collect.exe" -NoNewWindow -Wait -PassThru
+    New-Item -Path "C:\w\build-results" -ItemType "directory" -Force
+    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

@@ -0,0 +1,41 @@
+name: Check buck targets and code format
+on: [push, pull_request]
+jobs:
+  check:
+    name: Check TARGETS file and code format
+    runs-on: ubuntu-latest
+    steps:
+    - name: Checkout feature branch
+      uses: actions/checkout@v2
+      with:
+        fetch-depth: 0
+
+    - name: Fetch from upstream
+      run: |
+        git remote add upstream https://github.com/facebook/rocksdb.git && git fetch upstream
+
+    - name: Where am I
+      run: |
+        echo git status && git status
+        echo "git remote -v" && git remote -v
+        echo git branch && git branch
+
+    - name: Setup Python
+      uses: actions/setup-python@v1
+
+    - name: Install Dependencies
+      run: python -m pip install --upgrade pip
+
+    - name: Install argparse
+      run: pip install argparse
+
+    - name: Download clang-format-diff.py
+      uses: wei/wget@v1
+      with:
+        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
+
+    - name: Compare buckify output
+      run: make check-buck-targets

.gitignore

@@ -1,4 +1,5 @@
 make_config.mk
+rocksdb.pc
 
 *.a
 *.arc
@@ -7,6 +8,7 @@ make_config.mk
 *.gcda
 *.gcno
 *.o
+*.o.tmp
 *.so
 *.so.*
 *_test
@@ -34,6 +36,7 @@ manifest_dump
 sst_dump
 blob_dump
 block_cache_trace_analyzer
+db_with_timestamp_basic_test
 tools/block_cache_analyzer/*.pyc
 column_aware_encoding_exp
 util/build_version.cc
@@ -51,6 +54,7 @@ db_test2
 trace_analyzer
 trace_analyzer_test
 block_cache_trace_analyzer
+io_tracer_parser
 .DS_Store
 .vs
 .vscode
@@ -82,3 +86,12 @@ fbcode/
 fbcode
 buckifier/*.pyc
 buckifier/__pycache__
+
+compile_commands.json
+clang-format-diff.py
+.py3/
+
+fuzz/proto/gen/
+fuzz/crash-*
+
+cmake-build-*

.travis.yml

@@ -2,18 +2,18 @@ dist: xenial
 language: cpp
 os:
   - linux
-  - osx
+arch:
+  - arm64
+  - ppc64le
 compiler:
   - clang
   - gcc
-osx_image: xcode9.4
-jdk:
-  - openjdk7
 cache:
   - ccache
 
 addons:
   apt:
+    update: true
     sources:
       - ubuntu-toolchain-r-test
     packages:
@@ -24,15 +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
@@ -48,45 +39,151 @@ env:
   - JOB_NAME=examples # 5-7 minutes
   - JOB_NAME=cmake # 3-5 minutes
   - JOB_NAME=cmake-gcc8 # 3-5 minutes
+  - JOB_NAME=cmake-gcc9 # 3-5 minutes
+  - JOB_NAME=cmake-gcc9-c++20 # 3-5 minutes
   - JOB_NAME=cmake-mingw # 3 minutes
+  - JOB_NAME=make-gcc4.8
+  - JOB_NAME=status_checked
 
 matrix:
   exclude:
-  - os: osx
-    env: TEST_GROUP=1
-  - os: osx
-    env: TEST_GROUP=2
-  - os: osx
-    env: TEST_GROUP=3
-  - os: osx
-    env: TEST_GROUP=4
-  - os: osx
-    env: JOB_NAME=cmake-gcc8
-  - os : osx
+  - os : linux
+    arch: arm64
     env: JOB_NAME=cmake-mingw
   - os : linux
+    arch: arm64
+    env: JOB_NAME=make-gcc4.8
+  - os: linux
+    arch: ppc64le
+    env: JOB_NAME=cmake-mingw
+  - os: linux
+    arch: ppc64le
+    env: JOB_NAME=make-gcc4.8
+  - os: linux
     compiler: clang
-  - os : osx
-    compiler: gcc
+  - 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 : 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/ AND commit_message !~ /java/
+    os : linux
+    arch: arm64
+    env: JOB_NAME=java_test
+  - if: type = pull_request AND commit_message !~ /FULL_CI/ AND commit_message !~ /java/
+    os: linux
+    arch: ppc64le
+    env: JOB_NAME=java_test
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: JOB_NAME=lite_build
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: JOB_NAME=lite_build
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: JOB_NAME=examples
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: JOB_NAME=examples
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: JOB_NAME=cmake-gcc8
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: JOB_NAME=cmake-gcc8
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: JOB_NAME=cmake-gcc9
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: JOB_NAME=cmake-gcc9
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: JOB_NAME=cmake-gcc9-c++20
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: JOB_NAME=cmake-gcc9-c++20
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: JOB_NAME=status_checked
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    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;
+      sudo apt-get install -y g++-8 || exit $?;
       CC=gcc-8 && CXX=g++-8;
     fi
+  - if [ "${JOB_NAME}" == cmake-gcc9 ] || [ "${JOB_NAME}" == cmake-gcc9-c++20 ]; then
+      sudo apt-get install -y g++-9 || exit $?;
+      CC=gcc-9 && CXX=g++-9;
+    fi
   - if [ "${JOB_NAME}" == cmake-mingw ]; then
-      sudo apt-get install -y mingw-w64 ;
+      sudo apt-get install -y mingw-w64 || exit $?;
     fi
-  - if [[ "${JOB_NAME}" == cmake* ]] && [ "${TRAVIS_OS_NAME}" == linux ]; then
-      CMAKE_DIST_URL="https://rocksdb-deps.s3-us-west-2.amazonaws.com/cmake/cmake-3.14.5-Linux-$(uname -m).tar.bz2";
-      TAR_OPT="--strip-components=1 -xj";
-      mkdir cmake-dist && curl --silent --fail --show-error --location "${CMAKE_DIST_URL}" | tar -C cmake-dist ${TAR_OPT} && export PATH=$PWD/cmake-dist/bin:$PATH;
+  - if [ "${JOB_NAME}" == make-gcc4.8 ]; then
+      sudo apt-get install -y g++-4.8 || exit $?;
+      CC=gcc-4.8 && CXX=g++-4.8;
     fi
-  - if [[ "${JOB_NAME}" == java_test ]]; then
-      java -version && echo "JAVA_HOME=${JAVA_HOME}";
+  - |
+    if [[ "${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
+      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
     fi
 
 before_script:
@@ -95,41 +192,53 @@ before_script:
   - ulimit -n 8192
 
 script:
-  - ${CXX} --version
+  - date; ${CXX} --version
   - if [ `command -v ccache` ]; then ccache -C; fi
   - case $TEST_GROUP in
     platform_dependent)
-      OPT=-DTRAVIS V=1 ROCKSDBTESTS_END=db_block_cache_test 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_START=db_block_cache_test ROCKSDBTESTS_END=db_iter_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_START=db_iter_test ROCKSDBTESTS_END=options_file_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 V=1 ROCKSDBTESTS_START=options_file_test ROCKSDBTESTS_END=write_prepared_transaction_test make -j4 check_some
+      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_START=write_prepared_transaction_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 static_lib tools
+      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;
       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
       ;;
     cmake*)
-      mkdir build && cd build && cmake -DJNI=1 .. -DCMAKE_BUILD_TYPE=Release && make -j4 rocksdb rocksdbjni
+      case $JOB_NAME in
+        *-c++20)
+          OPT=-DCMAKE_CXX_STANDARD=20
+          ;;
+      esac
+
+      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 LIB_MODE=shared V=1 SKIP_LINK=1 make -j4 all && [ "Linking broken because libgflags compiled with newer ABI" ]
+      ;;
+    status_checked)
+      OPT=-DTRAVIS LIB_MODE=shared V=1 ASSERT_STATUS_CHECKED=1 make -j4 check_some
       ;;
     esac
 notifications:

CMakeLists.txt

@@ -83,15 +83,18 @@ else()
   option(WITH_FOLLY_DISTRIBUTED_MUTEX "build with folly::DistributedMutex" OFF)
 endif()
 
+if( NOT DEFINED CMAKE_CXX_STANDARD )
+  set(CMAKE_CXX_STANDARD 11)
+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()
-  if(CMAKE_SYSTEM_NAME MATCHES "FreeBSD")
+  if(CMAKE_SYSTEM_NAME MATCHES "FreeBSD" AND NOT CMAKE_SYSTEM_NAME MATCHES "kFreeBSD")
     # FreeBSD has jemalloc as default malloc
     # but it does not have all the jemalloc files in include/...
     set(WITH_JEMALLOC ON)
@@ -103,18 +106,40 @@ else()
     endif()
   endif()
 
-  # No config file for this
+  if(MINGW)
+    option(WITH_GFLAGS "build with GFlags" OFF)
+  else()
+    option(WITH_GFLAGS "build with GFlags" ON)
+  endif()
+  set(GFLAGS_LIB)
   if(WITH_GFLAGS)
-    find_package(gflags REQUIRED)
+    # Config with namespace available since gflags 2.2.2
+    option(GFLAGS_USE_TARGET_NAMESPACE "Use gflags import target with namespace." ON)
+    find_package(gflags CONFIG)
+    if(gflags_FOUND)
+      if(TARGET ${GFLAGS_TARGET})
+        # Config with GFLAGS_TARGET available since gflags 2.2.0
+        set(GFLAGS_LIB ${GFLAGS_TARGET})
+      else()
+        # Config with GFLAGS_LIBRARIES available since gflags 2.1.0
+        set(GFLAGS_LIB ${GFLAGS_LIBRARIES})
+      endif()
+    else()
+      find_package(gflags REQUIRED)
+        set(GFLAGS_LIB gflags::gflags)
+    endif()
+    include_directories(${GFLAGS_INCLUDE_DIR})
+    list(APPEND THIRDPARTY_LIBS ${GFLAGS_LIB})
     add_definitions(-DGFLAGS=1)
-    include_directories(${gflags_INCLUDE_DIR})
-    list(APPEND THIRDPARTY_LIBS gflags::gflags)
   endif()
 
   if(WITH_SNAPPY)
-    find_package(snappy REQUIRED)
+    find_package(Snappy CONFIG)
+    if(NOT Snappy_FOUND)
+      find_package(Snappy REQUIRED)
+    endif()
     add_definitions(-DSNAPPY)
-    list(APPEND THIRDPARTY_LIBS snappy::snappy)
+    list(APPEND THIRDPARTY_LIBS Snappy::snappy)
   endif()
 
   if(WITH_ZLIB)
@@ -149,23 +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)
@@ -178,20 +205,20 @@ 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 "${CMAKE_CXX_FLAGS} -Wno-format -fno-asynchronous-unwind-tables")
     add_definitions(-D_POSIX_C_SOURCE=1)
   endif()
-  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++11")
   if(NOT CMAKE_BUILD_TYPE STREQUAL "Debug")
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fno-omit-frame-pointer")
     include(CheckCXXCompilerFlag)
@@ -203,39 +230,66 @@ else()
 endif()
 
 include(CheckCCompilerFlag)
-if(CMAKE_SYSTEM_PROCESSOR MATCHES "ppc64le")
+if(CMAKE_SYSTEM_PROCESSOR MATCHES "^(powerpc|ppc)64")
+  CHECK_C_COMPILER_FLAG("-mcpu=power9" HAS_POWER9)
+  if(HAS_POWER9)
+    set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mcpu=power9 -mtune=power9")
+    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mcpu=power9 -mtune=power9")
+  else()
+    CHECK_C_COMPILER_FLAG("-mcpu=power8" HAS_POWER8)
+    if(HAS_POWER8)
+      set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mcpu=power8 -mtune=power8")
+      set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mcpu=power8 -mtune=power8")
+    endif(HAS_POWER8)
+  endif(HAS_POWER9)
   CHECK_C_COMPILER_FLAG("-maltivec" HAS_ALTIVEC)
   if(HAS_ALTIVEC)
     message(STATUS " HAS_ALTIVEC yes")
     set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -maltivec")
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -maltivec")
-    set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mcpu=power8")
-    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mcpu=power8")
   endif(HAS_ALTIVEC)
-endif(CMAKE_SYSTEM_PROCESSOR MATCHES "ppc64le")
+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)
+option(FORCE_AVX "force building with AVX, even when PORTABLE=ON" OFF)
+option(FORCE_AVX2 "force building with AVX2, even when PORTABLE=ON" OFF)
 if(PORTABLE)
   # MSVC does not need a separate compiler flag to enable SSE4.2; if nmmintrin.h
   # is available, it is available by default.
   if(FORCE_SSE42 AND NOT MSVC)
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -msse4.2 -mpclmul")
   endif()
+  if(MSVC)
+    if(FORCE_AVX)
+      set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /arch:AVX")
+    endif()
+    # MSVC automatically enables BMI / lzcnt with AVX2.
+    if(FORCE_AVX2)
+      set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /arch:AVX2")
+    endif()
+  else()
+    if(FORCE_AVX)
+      set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mavx")
+    endif()
+    if(FORCE_AVX2)
+      set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mavx2 -mbmi -mlzcnt")
+    endif()
+  endif()
 else()
   if(MSVC)
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /arch:AVX2")
   else()
-    if(NOT HAVE_POWER8 AND NOT HAS_ARMV8_CRC)
+    if(NOT CMAKE_SYSTEM_PROCESSOR MATCHES "^(powerpc|ppc)64" AND NOT HAS_ARMV8_CRC)
       set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -march=native")
     endif()
   endif()
@@ -245,6 +299,7 @@ include(CheckCXXSourceCompiles)
 if(NOT MSVC)
   set(CMAKE_REQUIRED_FLAGS "-msse4.2 -mpclmul")
 endif()
+
 CHECK_CXX_SOURCE_COMPILES("
 #include <cstdint>
 #include <nmmintrin.h>
@@ -343,6 +398,12 @@ if(NOT WITH_DYNAMIC_EXTENSION)
   add_definitions(-DROCKSDB_NO_DYNAMIC_EXTENSION)
 endif()
 
+option(ASSERT_STATUS_CHECKED "build with assert status checked" OFF)
+if (ASSERT_STATUS_CHECKED)
+  message(STATUS "Build with assert status checked")
+  add_definitions(-DROCKSDB_ASSERT_STATUS_CHECKED)
+endif()
+
 if(DEFINED USE_RTTI)
   if(USE_RTTI)
     message(STATUS "Enabling RTTI")
@@ -377,7 +438,15 @@ if(MSVC)
     message(STATUS "Debug optimization is enabled")
     set(CMAKE_CXX_FLAGS_DEBUG "/Oxt")
   else()
-    set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} /Od /RTC1 /Gm")
+    set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} /Od /RTC1")
+
+    # Minimal Build is deprecated after MSVC 2015
+    if( MSVC_VERSION GREATER 1900 )
+      set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} /Gm-")
+    else()
+      set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} /Gm")
+    endif()
+
   endif()
   if(WITH_RUNTIME_DEBUG)
     set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} /${RUNTIME_LIBRARY}d")
@@ -404,15 +473,12 @@ 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")
   add_definitions(-DOS_SOLARIS)
+elseif(CMAKE_SYSTEM_NAME MATCHES "kFreeBSD")
+  add_definitions(-DOS_GNU_KFREEBSD)
 elseif(CMAKE_SYSTEM_NAME MATCHES "FreeBSD")
   add_definitions(-DOS_FREEBSD)
 elseif(CMAKE_SYSTEM_NAME MATCHES "NetBSD")
@@ -471,7 +537,11 @@ if(HAVE_PTHREAD_MUTEX_ADAPTIVE_NP)
 endif()
 
 include(CheckCXXSymbolExists)
-check_cxx_symbol_exists(malloc_usable_size malloc.h HAVE_MALLOC_USABLE_SIZE)
+if(CMAKE_SYSTEM_NAME MATCHES "^FreeBSD")
+  check_cxx_symbol_exists(malloc_usable_size malloc_np.h HAVE_MALLOC_USABLE_SIZE)
+else()
+  check_cxx_symbol_exists(malloc_usable_size malloc.h HAVE_MALLOC_USABLE_SIZE)
+endif()
 if(HAVE_MALLOC_USABLE_SIZE)
   add_definitions(-DROCKSDB_MALLOC_USABLE_SIZE)
 endif()
@@ -481,9 +551,13 @@ if(HAVE_SCHED_GETCPU)
   add_definitions(-DROCKSDB_SCHED_GETCPU_PRESENT)
 endif()
 
+check_cxx_symbol_exists(getauxval auvx.h HAVE_AUXV_GETAUXVAL)
+if(HAVE_AUXV_GETAUXVAL)
+  add_definitions(-DROCKSDB_AUXV_GETAUXVAL_PRESENT)
+endif()
+
 include_directories(${PROJECT_SOURCE_DIR})
 include_directories(${PROJECT_SOURCE_DIR}/include)
-include_directories(SYSTEM ${PROJECT_SOURCE_DIR}/third-party/gtest-1.8.1/fused-src)
 if(WITH_FOLLY_DISTRIBUTED_MUTEX)
   include_directories(${PROJECT_SOURCE_DIR}/third-party/folly)
 endif()
@@ -492,14 +566,23 @@ find_package(Threads REQUIRED)
 # Main library source code
 
 set(SOURCES
+        cache/cache.cc
         cache/clock_cache.cc
         cache/lru_cache.cc
         cache/sharded_cache.cc
         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
+        db/blob/blob_log_format.cc
+        db/blob/blob_log_sequential_reader.cc
+        db/blob/blob_log_writer.cc
         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
@@ -507,8 +590,10 @@ set(SOURCES
         db/compaction/compaction_picker_fifo.cc
         db/compaction/compaction_picker_level.cc
         db/compaction/compaction_picker_universal.cc
+        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
@@ -539,6 +624,8 @@ set(SOURCES
         db/memtable_list.cc
         db/merge_helper.cc
         db/merge_operator.cc
+        db/output_validator.cc
+        db/periodic_work_scheduler.cc
         db/range_del_aggregator.cc
         db/range_tombstone_fragmenter.cc
         db/repair.cc
@@ -549,22 +636,28 @@ set(SOURCES
         db/trim_history_scheduler.cc
         db/version_builder.cc
         db/version_edit.cc
+        db/version_edit_handler.cc
         db/version_set.cc
+        db/wal_edit.cc
         db/wal_manager.cc
         db/write_batch.cc
         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
@@ -577,6 +670,7 @@ set(SOURCES
         memory/arena.cc
         memory/concurrent_arena.cc
         memory/jemalloc_nodump_allocator.cc
+        memory/memkind_kmem_allocator.cc
         memtable/alloc_tracker.cc
         memtable/hash_linklist_rep.cc
         memtable/hash_skiplist_rep.cc
@@ -597,19 +691,23 @@ set(SOURCES
         monitoring/thread_status_util.cc
         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
         options/options_parser.cc
-        options/options_sanity_check.cc
         port/stack_trace.cc
         table/adaptive/adaptive_table_factory.cc
+        table/block_based/binary_search_index_reader.cc
         table/block_based/block.cc
         table/block_based/block_based_filter_block.cc
         table/block_based/block_based_table_builder.cc
         table/block_based/block_based_table_factory.cc
+        table/block_based/block_based_table_iterator.cc
         table/block_based/block_based_table_reader.cc
         table/block_based/block_builder.cc
+        table/block_based/block_prefetcher.cc
         table/block_based/block_prefix_index.cc
         table/block_based/data_block_hash_index.cc
         table/block_based/data_block_footer.cc
@@ -617,9 +715,14 @@ set(SOURCES
         table/block_based/filter_policy.cc
         table/block_based/flush_block_policy.cc
         table/block_based/full_filter_block.cc
+        table/block_based/hash_index_reader.cc
         table/block_based/index_builder.cc
+        table/block_based/index_reader_common.cc
         table/block_based/parsed_full_filter_block.cc
         table/block_based/partitioned_filter_block.cc
+        table/block_based/partitioned_index_iterator.cc
+        table/block_based/partitioned_index_reader.cc
+        table/block_based/reader_common.cc
         table/block_based/uncompression_dict_reader.cc
         table/block_fetcher.cc
         table/cuckoo/cuckoo_table_builder.cc
@@ -637,8 +740,10 @@ set(SOURCES
         table/plain/plain_table_index.cc
         table/plain/plain_table_key_coding.cc
         table/plain/plain_table_reader.cc
+        table/sst_file_dumper.cc
         table/sst_file_reader.cc
         table/sst_file_writer.cc
+        table/table_factory.cc
         table/table_properties.cc
         table/two_level_iterator.cc
         test_util/sync_point.cc
@@ -647,12 +752,14 @@ set(SOURCES
         test_util/transaction_test_util.cc
         tools/block_cache_analyzer/block_cache_trace_analyzer.cc
         tools/dump/db_dump_tool.cc
+        tools/io_tracer_parser_tool.cc
         tools/ldb_cmd.cc
         tools/ldb_tool.cc
         tools/sst_dump_tool.cc
         tools/trace_analyzer_tool.cc
         trace_replay/trace_replay.cc
         trace_replay/block_cache_tracer.cc
+        trace_replay/io_tracer.cc
         util/coding.cc
         util/compaction_job_stats_impl.cc
         util/comparator.cc
@@ -664,6 +771,7 @@ 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
@@ -678,9 +786,6 @@ set(SOURCES
         utilities/blob_db/blob_db_impl_filesnapshot.cc
         utilities/blob_db/blob_dump_tool.cc
         utilities/blob_db/blob_file.cc
-        utilities/blob_db/blob_log_reader.cc
-        utilities/blob_db/blob_log_writer.cc
-        utilities/blob_db/blob_log_format.cc
         utilities/cassandra/cassandra_compaction_filter.cc
         utilities/cassandra/format.cc
         utilities/cassandra/merge_operator.cc
@@ -689,6 +794,8 @@ set(SOURCES
         utilities/debug.cc
         utilities/env_mirror.cc
         utilities/env_timed.cc
+        utilities/fault_injection_env.cc
+        utilities/fault_injection_fs.cc
         utilities/leveldb_options/leveldb_options.cc
         utilities/memory/memory_util.cc
         utilities/merge_operators/bytesxor.cc
@@ -710,6 +817,11 @@ 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_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
@@ -717,7 +829,6 @@ 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
@@ -725,8 +836,21 @@ 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(
@@ -734,11 +858,11 @@ if(HAVE_SSE42 AND NOT MSVC)
     PROPERTIES COMPILE_FLAGS "-msse4.2 -mpclmul")
 endif()
 
-if(HAVE_POWER8)
+if(CMAKE_SYSTEM_PROCESSOR MATCHES "^(powerpc|ppc)64")
   list(APPEND SOURCES
     util/crc32c_ppc.c
     util/crc32c_ppc_asm.S)
-endif(HAVE_POWER8)
+endif(CMAKE_SYSTEM_PROCESSOR MATCHES "^(powerpc|ppc)64")
 
 if(HAS_ARMV8_CRC)
   list(APPEND SOURCES
@@ -751,9 +875,13 @@ if(WIN32)
     port/win/env_win.cc
     port/win/env_default.cc
     port/win/port_win.cc
-    port/win/win_logger.cc
-    port/win/win_thread.cc)
-
+    port/win/win_logger.cc)
+  if(NOT MINGW)
+    # Mingw only supports std::thread when using
+    # posix threads.
+    list(APPEND SOURCES
+      port/win/win_thread.cc)
+  endif()
 if(WITH_XPRESS)
   list(APPEND SOURCES
     port/win/xpress_win.cc)
@@ -799,13 +927,13 @@ else()
   set(SYSTEM_LIBS ${CMAKE_THREAD_LIBS_INIT})
 endif()
 
-add_library(${ROCKSDB_STATIC_LIB} STATIC ${SOURCES})
-target_link_libraries(${ROCKSDB_STATIC_LIB}
+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 ${SOURCES})
-  target_link_libraries(${ROCKSDB_SHARED_LIB}
+  add_library(${ROCKSDB_SHARED_LIB} SHARED ${SOURCES} ${BUILD_VERSION_CC})
+  target_link_libraries(${ROCKSDB_SHARED_LIB} PRIVATE
     ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
 
   if(WIN32)
@@ -822,8 +950,7 @@ if(ROCKSDB_BUILD_SHARED)
                           LINKER_LANGUAGE CXX
                           VERSION ${rocksdb_VERSION}
                           SOVERSION ${rocksdb_VERSION_MAJOR}
-                          CXX_STANDARD 11
-                          OUTPUT_NAME "rocksdb")
+                          OUTPUT_NAME "rocksdb${ARTIFACT_SUFFIX}")
   endif()
 endif()
 
@@ -834,6 +961,16 @@ else()
 endif()
 
 option(WITH_JNI "build with JNI" OFF)
+# Tests are excluded from Release builds
+CMAKE_DEPENDENT_OPTION(WITH_TESTS "build with tests" ON
+  "CMAKE_BUILD_TYPE STREQUAL Debug" OFF)
+option(WITH_BENCHMARK_TOOLS "build with benchmarks" ON)
+option(WITH_CORE_TOOLS "build with ldb and sst_dump" ON)
+option(WITH_TOOLS "build with tools" ON)
+
+if(WITH_TESTS OR WITH_BENCHMARK_TOOLS OR WITH_TOOLS OR WITH_JNI OR JNI)
+  include_directories(SYSTEM ${PROJECT_SOURCE_DIR}/third-party/gtest-1.8.1/fused-src)
+endif()
 if(WITH_JNI OR JNI)
   message(STATUS "JNI library is enabled")
   add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/java)
@@ -871,6 +1008,8 @@ if(NOT WIN32 OR ROCKSDB_INSTALL_ON_WINDOWS)
 
   install(DIRECTORY include/rocksdb COMPONENT devel DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}")
 
+  install(DIRECTORY "${PROJECT_SOURCE_DIR}/cmake/modules" COMPONENT devel DESTINATION ${package_config_destination})
+
   install(
     TARGETS ${ROCKSDB_STATIC_LIB}
     EXPORT RocksDBTargets
@@ -907,18 +1046,34 @@ if(NOT WIN32 OR ROCKSDB_INSTALL_ON_WINDOWS)
   )
 endif()
 
-# Tests are excluded from Release builds
-CMAKE_DEPENDENT_OPTION(WITH_TESTS "build with tests" ON
-  "CMAKE_BUILD_TYPE STREQUAL Debug" OFF)
-if(WITH_TESTS)
+option(WITH_ALL_TESTS "Build all test, rather than a small subset" ON)
+
+if(WITH_TESTS OR WITH_BENCHMARK_TOOLS)
   add_subdirectory(third-party/gtest-1.8.1/fused-src/gtest)
   add_library(testharness STATIC
+  test_util/mock_time_env.cc
   test_util/testharness.cc)
   target_link_libraries(testharness gtest)
+endif()
 
+if(WITH_TESTS)
   set(TESTS
+        db/db_basic_test.cc
+        env/env_basic_test.cc
+  )
+  if(WITH_ALL_TESTS)
+    list(APPEND TESTS
         cache/cache_test.cc
         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
         db/compaction/compaction_job_stats_test.cc
@@ -928,8 +1083,7 @@ if(WITH_TESTS)
         db/comparator_db_test.cc
         db/corruption_test.cc
         db/cuckoo_table_db_test.cc
-        db/db_basic_test.cc
-        db/db_blob_index_test.cc
+        db/db_with_timestamp_basic_test.cc
         db/db_block_cache_test.cc
         db/db_bloom_filter_test.cc
         db/db_compaction_filter_test.cc
@@ -941,6 +1095,7 @@ 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
@@ -948,19 +1103,21 @@ 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
         db/db_tailing_iter_test.cc
         db/db_test.cc
         db/db_test2.cc
+        db/db_logical_block_size_cache_test.cc
         db/db_universal_compaction_test.cc
         db/db_wal_test.cc
+        db/db_with_timestamp_compaction_test.cc
         db/db_write_test.cc
         db/dbformat_test.cc
         db/deletefile_test.cc
-        db/error_handler_test.cc
+        db/error_handler_fs_test.cc
         db/obsolete_files_test.cc
         db/external_sst_file_basic_test.cc
         db/external_sst_file_test.cc
@@ -976,6 +1133,7 @@ if(WITH_TESTS)
         db/merge_test.cc
         db/options_file_test.cc
         db/perf_context_test.cc
+        db/periodic_work_scheduler_test.cc
         db/plain_table_db_test.cc
         db/prefix_test.cc
         db/range_del_aggregator_test.cc
@@ -986,17 +1144,21 @@ if(WITH_TESTS)
         db/version_edit_test.cc
         db/version_set_test.cc
         db/wal_manager_test.cc
+        db/wal_edit_test.cc
         db/write_batch_test.cc
         db/write_callback_test.cc
         db/write_controller_test.cc
-        env/env_basic_test.cc
         env/env_test.cc
+        env/io_posix_test.cc
         env/mock_env_test.cc
         file/delete_scheduler_test.cc
+        file/prefetch_test.cc
+        file/random_access_file_reader_test.cc
         logging/auto_roll_logger_test.cc
         logging/env_logger_test.cc
         logging/event_logger_test.cc
         memory/arena_test.cc
+        memory/memkind_kmem_allocator_test.cc
         memtable/inlineskiplist_test.cc
         memtable/skiplist_test.cc
         memtable/write_buffer_manager_test.cc
@@ -1004,9 +1166,12 @@ if(WITH_TESTS)
         monitoring/iostats_context_test.cc
         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
+        table/block_based/block_based_table_reader_test.cc
         table/block_based/block_test.cc
         table/block_based/data_block_hash_index_test.cc
         table/block_based/full_filter_block_test.cc
@@ -1017,7 +1182,12 @@ if(WITH_TESTS)
         table/merger_test.cc
         table/sst_file_reader_test.cc
         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
         tools/ldb_cmd_test.cc
         tools/reduce_levels_test.cc
         tools/sst_dump_test.cc
@@ -1035,11 +1205,14 @@ 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
+        util/timer_test.cc
         util/thread_list_test.cc
         util/thread_local_test.cc
+        util/work_queue_test.cc
         utilities/backupable/backupable_db_test.cc
         utilities/blob_db/blob_db_test.cc
         utilities/cassandra/cassandra_functional_test.cc
@@ -1059,11 +1232,14 @@ 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/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
-  )
+    )
+  endif()
   if(WITH_LIBRADOS)
     list(APPEND TESTS utilities/env_librados_test.cc)
   endif()
@@ -1076,7 +1252,6 @@ if(WITH_TESTS)
       db/db_test_util.cc
       monitoring/thread_status_updater_debug.cc
       table/mock_table.cc
-      test_util/fault_injection_test_env.cc
       utilities/cassandra/test_utils.cc
   )
   enable_testing()
@@ -1091,7 +1266,7 @@ if(WITH_TESTS)
         PROPERTIES EXCLUDE_FROM_DEFAULT_BUILD_RELEASE 1
         EXCLUDE_FROM_DEFAULT_BUILD_MINRELEASE 1
         EXCLUDE_FROM_DEFAULT_BUILD_RELWITHDEBINFO 1
-        )
+  )
 
   foreach(sourcefile ${TESTS})
       get_filename_component(exename ${sourcefile} NAME_WE)
@@ -1101,12 +1276,16 @@ if(WITH_TESTS)
         EXCLUDE_FROM_DEFAULT_BUILD_MINRELEASE 1
         EXCLUDE_FROM_DEFAULT_BUILD_RELWITHDEBINFO 1
         OUTPUT_NAME ${exename}${ARTIFACT_SUFFIX}
-        )
-      target_link_libraries(${CMAKE_PROJECT_NAME}_${exename}${ARTIFACT_SUFFIX} testutillib${ARTIFACT_SUFFIX} testharness gtest ${ROCKSDB_LIB})
+      )
+      target_link_libraries(${CMAKE_PROJECT_NAME}_${exename}${ARTIFACT_SUFFIX} testutillib${ARTIFACT_SUFFIX} testharness gtest ${THIRDPARTY_LIBS} ${ROCKSDB_LIB})
       if(NOT "${exename}" MATCHES "db_sanity_test")
         add_test(NAME ${exename} COMMAND ${exename}${ARTIFACT_SUFFIX})
         add_dependencies(check ${CMAKE_PROJECT_NAME}_${exename}${ARTIFACT_SUFFIX})
       endif()
+      if("${exename}" MATCHES "env_librados_test")
+        # env_librados_test.cc uses librados directly
+        target_link_libraries(${CMAKE_PROJECT_NAME}_${exename}${ARTIFACT_SUFFIX} rados)
+      endif()
   endforeach(sourcefile ${TESTS})
 
   if(WIN32)
@@ -1122,55 +1301,51 @@ if(WITH_TESTS)
 
   if(ROCKSDB_LIB_FOR_C)
     set(C_TESTS db/c_test.c)
-    # C executables must link to a shared object
     add_executable(c_test db/c_test.c)
-    target_link_libraries(c_test ${ROCKSDB_SHARED_LIB} testharness)
+    target_link_libraries(c_test ${ROCKSDB_LIB_FOR_C} testharness)
     add_test(NAME c_test COMMAND c_test${ARTIFACT_SUFFIX})
     add_dependencies(check c_test)
   endif()
 endif()
 
-option(WITH_BENCHMARK_TOOLS "build with benchmarks" ON)
 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
-    ${ROCKSDB_LIB})
+  target_link_libraries(db_bench${ARTIFACT_SUFFIX}
+    ${ROCKSDB_LIB} ${THIRDPARTY_LIBS})
 
-  add_executable(cache_bench
+  add_executable(cache_bench${ARTIFACT_SUFFIX}
     cache/cache_bench.cc)
-  target_link_libraries(cache_bench
-    ${ROCKSDB_LIB})
+  target_link_libraries(cache_bench${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
-    ${ROCKSDB_LIB})
+  target_link_libraries(memtablerep_bench${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
-    ${ROCKSDB_LIB})
+  target_link_libraries(range_del_aggregator_bench${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
-    ${ROCKSDB_LIB} testharness)
+  target_link_libraries(table_reader_bench${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
-    ${ROCKSDB_LIB})
+  target_link_libraries(filter_bench${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
-    ${ROCKSDB_LIB})
+  target_link_libraries(hash_table_bench${ARTIFACT_SUFFIX}
+    ${ROCKSDB_LIB} ${GFLAGS_LIB})
 endif()
 
-option(WITH_CORE_TOOLS "build with ldb and sst_dump" ON)
-option(WITH_TOOLS "build with tools" ON)
 if(WITH_CORE_TOOLS OR WITH_TOOLS)
   add_subdirectory(tools)
   add_custom_target(core_tools
@@ -1182,3 +1357,8 @@ if(WITH_TOOLS)
   add_custom_target(tools
     DEPENDS ${tool_deps})
 endif()
+
+option(WITH_EXAMPLES "build with examples" OFF)
+if(WITH_EXAMPLES)
+  add_subdirectory(examples)
+endif()

HISTORY.md

@@ -1,11 +1,346 @@
 # Rocksdb Change Log
 ## Unreleased
+### 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().
+
 ### 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.
+
+### 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
+* For new integrated BlobDB, add support for blob files for backup/restore like table files. 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.
+
+### 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.
+
+## 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
+* Fixed a bug after a `CompactRange()` with `CompactRangeOptions::change_level` set fails due to a conflict in the level change step, which caused all subsequent calls to `CompactRange()` with `CompactRangeOptions::change_level` set to incorrectly fail with a `Status::NotSupported("another thread is refitting")` error.
+* Fixed a bug that the bottom most level compaction could still be a trivial move even if `BottommostLevelCompaction.kForce` or `kForceOptimized` is set.
+
+### Public API Change
+* The methods to create and manage EncrypedEnv have been changed.  The EncryptionProvider is now passed to NewEncryptedEnv as a shared pointer, rather than a raw pointer.  Comparably, the CTREncryptedProvider now takes a shared pointer, rather than a reference, to a BlockCipher.  CreateFromString methods have been added to BlockCipher and EncryptionProvider to provide a single API by which different ciphers and providers can be created, respectively.
+* The internal classes (CTREncryptionProvider, ROT13BlockCipher, CTRCipherStream) associated with the EncryptedEnv have been moved out of the public API.  To create a CTREncryptionProvider, one can either use EncryptionProvider::NewCTRProvider, or EncryptionProvider::CreateFromString("CTR").  To create a new ROT13BlockCipher, one can either use BlockCipher::NewROT13Cipher or BlockCipher::CreateFromString("ROT13").
+* The EncryptionProvider::AddCipher method has been added to allow keys to be added to an EncryptionProvider.  This API will allow future providers to support multiple cipher keys.
+* Add a new option "allow_data_in_errors". When this new option is set by users, it allows users to opt-in to get error messages containing corrupted keys/values. Corrupt keys, values will be logged in the messages, logs, status etc. that will help users with the useful information regarding affected data. By default value of this option is set false to prevent users data to be exposed in the messages so currently, data will be redacted from logs, messages, status by default.
+* AdvancedColumnFamilyOptions::force_consistency_checks is now true by default, for more proactive DB corruption detection at virtually no cost (estimated two extra CPU cycles per million on a major production workload). Corruptions reported by these checks now mention "force_consistency_checks" in case a false positive corruption report is suspected and the option needs to be disabled (unlikely). Since existing column families have a saved setting for force_consistency_checks, only new column families will pick up the new default.
+
+### General Improvements
+* 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
+* Fix a performance regression introduced in 6.4 that makes a upper bound check for every Next() even if keys are within a data block that is within the upper bound.
+* Fix a possible corruption to the LSM state (overlapping files within a level) when a `CompactRange()` for refitting levels (`CompactRangeOptions::change_level == true`) and another manual compaction are executed in parallel.
+* Sanitize `recycle_log_file_num` to zero when the user attempts to enable it in combination with `WALRecoveryMode::kTolerateCorruptedTailRecords`. Previously the two features were allowed together, which compromised the user's configured crash-recovery guarantees.
+* Fix a bug where a level refitting in CompactRange() might race with an automatic compaction that puts the data to the target level of the refitting. The bug has been there for years.
+* Fixed a bug in version 6.12 in which BackupEngine::CreateNewBackup could fail intermittently with non-OK status when backing up a read-write DB configured with a DBOptions::file_checksum_gen_factory.
+* Fix useless no-op compactions scheduled upon snapshot release when options.disable-auto-compactions = true.
+* Fix a bug when max_write_buffer_size_to_maintain is set, immutable flushed memtable destruction is delayed until the next super version is installed. A memtable is not added to delete list because of its reference hold by super version and super version doesn't switch because of empt delete list. So memory usage keeps on increasing beyond write_buffer_size + max_write_buffer_size_to_maintain.
+* Avoid converting MERGES to PUTS when allow_ingest_behind is true.
+* Fix compression dictionary sampling together with `SstFileWriter`. Previously, the dictionary would be trained/finalized immediately with zero samples. Now, the whole `SstFileWriter` file is buffered in memory and then sampled.
+* Fix a bug with `avoid_unnecessary_blocking_io=1` and creating backups (BackupEngine::CreateNewBackup) or checkpoints (Checkpoint::Create). With this setting and WAL enabled, these operations could randomly fail with non-OK status.
+* Fix a bug in which bottommost compaction continues to advance the underlying InternalIterator to skip tombstones even after shutdown.
+
+### New Features
+* A new field `std::string requested_checksum_func_name` is added to `FileChecksumGenContext`, which enables the checksum factory to create generators for a suite of different functions.
+* Added a new subcommand, `ldb unsafe_remove_sst_file`, which removes a lost or corrupt SST file from a DB's metadata. This command involves data loss and must not be used on a live DB.
+
+### Performance Improvements
+* Reduce thread number for multiple DB instances by re-using one global thread for statistics dumping and persisting.
+* Reduce write-amp in heavy write bursts in `kCompactionStyleLevel` compaction style with `level_compaction_dynamic_level_bytes` set.
+* BackupEngine incremental backups no longer read DB table files that are already saved to a shared part of the backup directory, unless `share_files_with_checksum` is used with `kLegacyCrc32cAndFileSize` naming (discouraged).
+  * For `share_files_with_checksum`, we are confident there is no regression (vs. pre-6.12) in detecting DB or backup corruption at backup creation time, mostly because the old design did not leverage this extra checksum computation for detecting inconsistencies at backup creation time.
+  * For `share_table_files` without "checksum" (not recommended), there is a regression in detecting fundamentally unsafe use of the option, greatly mitigated by file size checking (under "Behavior Changes"). Almost no reason to use `share_files_with_checksum=false` should remain.
+  * `DB::VerifyChecksum` and `BackupEngine::VerifyBackup` with checksum checking are still able to catch corruptions that `CreateNewBackup` does not.
+
+### Public API Change
+* Expose kTypeDeleteWithTimestamp in EntryType and update GetEntryType() accordingly.
+* Added file_checksum and file_checksum_func_name to TableFileCreationInfo, which can pass the table file checksum information through the OnTableFileCreated callback during flush and compaction.
+* A warning is added to `DB::DeleteFile()` API describing its known problems and deprecation plan.
+* Add a new stats level, i.e. StatsLevel::kExceptTickers (PR7329) to exclude tickers even if application passes a non-null Statistics object.
+* Added a new status code IOStatus::IOFenced() for the Env/FileSystem to indicate that writes from this instance are fenced off. Like any other background error, this error is returned to the user in Put/Merge/Delete/Flush calls and can be checked using Status::IsIOFenced().
+
+### Behavior Changes
+* File abstraction `FSRandomAccessFile.Prefetch()` default return status is changed from `OK` to `NotSupported`. If the user inherited file doesn't implement prefetch, RocksDB will create internal prefetch buffer to improve read performance.
+* When retryabel IO error happens during Flush (manifest write error is excluded) and WAL is disabled, originally it is mapped to kHardError. Now,it is mapped to soft error. So DB will not stall the writes unless the memtable is full. At the same time, when auto resume is triggered to recover the retryable IO error during Flush, SwitchMemtable is not called to avoid generating to many small immutable memtables. If WAL is enabled, no behavior changes.
+* When considering whether a table file is already backed up in a shared part of backup directory, BackupEngine would already query the sizes of source (DB) and pre-existing destination (backup) files. BackupEngine now uses these file sizes to detect corruption, as at least one of (a) old backup, (b) backup in progress, or (c) current DB is corrupt if there's a size mismatch.
+
+### Others
+* Error in prefetching partitioned index blocks will not be swallowed. It will fail the query and return the IOError users.
+
+## 6.12 (2020-07-28)
+### Public API Change
+* Encryption file classes now exposed for inheritance in env_encryption.h
+* File I/O listener is extended to cover more I/O operations. Now class `EventListener` in listener.h contains new callback functions: `OnFileFlushFinish()`, `OnFileSyncFinish()`, `OnFileRangeSyncFinish()`, `OnFileTruncateFinish()`, and ``OnFileCloseFinish()``.
+* `FileOperationInfo` now reports `duration` measured by `std::chrono::steady_clock` and `start_ts` measured by `std::chrono::system_clock` instead of start and finish timestamps measured by `system_clock`. Note that `system_clock` is called before `steady_clock` in program order at operation starts.
+* `DB::GetDbSessionId(std::string& session_id)` is added. `session_id` stores a unique identifier that gets reset every time the DB is opened. This DB session ID should be unique among all open DB instances on all hosts, and should be unique among re-openings of the same or other DBs. This identifier is recorded in the LOG file on the line starting with "DB Session ID:".
+* `DB::OpenForReadOnly()` now returns `Status::NotFound` when the specified DB directory does not exist. Previously the error returned depended on the underlying `Env`. This change is available in all 6.11 releases as well.
+* A parameter `verify_with_checksum` is added to `BackupEngine::VerifyBackup`, which is false by default. If it is ture, `BackupEngine::VerifyBackup` verifies checksums and file sizes of backup files. Pass `false` for `verify_with_checksum` to maintain the previous behavior and performance of `BackupEngine::VerifyBackup`, by only verifying sizes of backup files.
+
+### Behavior Changes
+* Best-efforts recovery ignores CURRENT file completely. If CURRENT file is missing during recovery, best-efforts recovery still proceeds with MANIFEST file(s).
+* In best-efforts recovery, an error that is not Corruption or IOError::kNotFound or IOError::kPathNotFound will be overwritten silently. Fix this by checking all non-ok cases and return early.
+* When `file_checksum_gen_factory` is set to `GetFileChecksumGenCrc32cFactory()`, BackupEngine will compare the crc32c checksums of table files computed when creating a backup to the expected checksums stored in the DB manifest, and will fail `CreateNewBackup()` on mismatch (corruption). If the `file_checksum_gen_factory` is not set or set to any other customized factory, there is no checksum verification to detect if SST files in a DB are corrupt when read, copied, and independently checksummed by BackupEngine.
+* When a DB sets `stats_dump_period_sec > 0`, either as the initial value for DB open or as a dynamic option change, the first stats dump is staggered in the following X seconds, where X is an integer in `[0, stats_dump_period_sec)`. Subsequent stats dumps are still spaced `stats_dump_period_sec` seconds apart.
+* When the paranoid_file_checks option is true, a hash is generated of all keys and values are generated when the SST file is written, and then the values are read back in to validate the file.  A corruption is signaled if the two hashes do not match.
+
+### Bug fixes
+* Compressed block cache was automatically disabled with read-only DBs by mistake. Now it is fixed: compressed block cache will be in effective with read-only DB too.
+* Fix a bug of wrong iterator result if another thread finishes an update and a DB flush between two statement.
+* Disable file deletion after MANIFEST write/sync failure until db re-open or Resume() so that subsequent re-open will not see MANIFEST referencing deleted SSTs.
+* Fix a bug when index_type == kTwoLevelIndexSearch in PartitionedIndexBuilder to update FlushPolicy to point to internal key partitioner when it changes from user-key mode to internal-key mode in index partition.
+* Make compaction report InternalKey corruption while iterating over the input.
+* Fix a bug which may cause MultiGet to be slow because it may read more data than requested, but this won't affect correctness. The bug was introduced in 6.10 release.
+* Fail recovery and report once hitting a physical log record checksum mismatch, while reading MANIFEST. RocksDB should not continue processing the MANIFEST any further.
+* Fixed a bug in size-amp-triggered and periodic-triggered universal compaction, where the compression settings for the first input level were used rather than the compression settings for the output (bottom) level.
+
+### New Features
+* DB identity (`db_id`) and DB session identity (`db_session_id`) are added to table properties and stored in SST files. SST files generated from SstFileWriter and Repairer have DB identity “SST Writer” and “DB Repairer”, respectively. Their DB session IDs are generated in the same way as `DB::GetDbSessionId`. The session ID for SstFileWriter (resp., Repairer) resets every time `SstFileWriter::Open` (resp., `Repairer::Run`) is called.
+* Added experimental option BlockBasedTableOptions::optimize_filters_for_memory for reducing allocated memory size of Bloom filters (~10% savings with Jemalloc) while preserving the same general accuracy. To have an effect, the option requires format_version=5 and malloc_usable_size. Enabling this option is forward and backward compatible with existing format_version=5.
+* `BackupableDBOptions::share_files_with_checksum_naming` is added with new default behavior for naming backup files with `share_files_with_checksum`, to address performance and backup integrity issues. See API comments for details.
+* Added auto resume function to automatically recover the DB from background Retryable IO Error. When retryable IOError happens during flush and WAL write, the error is mapped to Hard Error and DB will be in read mode. When retryable IO Error happens during compaction, the error will be mapped to Soft Error. DB is still in write/read mode. Autoresume function will create a thread for a DB to call DB->ResumeImpl() to try the recover for Retryable IO Error during flush and WAL write. Compaction will be rescheduled by itself if retryable IO Error happens. Auto resume may also cause other Retryable IO Error during the recovery, so the recovery will fail. Retry the auto resume may solve the issue, so we use max_bgerror_resume_count to decide how many resume cycles will be tried in total. If it is <=0, auto resume retryable IO Error is disabled. Default is INT_MAX, which will lead to a infinit auto resume. bgerror_resume_retry_interval decides the time interval between two auto resumes.
+* Option `max_subcompactions` can be set dynamically using DB::SetDBOptions().
+* Added experimental ColumnFamilyOptions::sst_partitioner_factory to define determine the partitioning of sst files. This helps compaction to split the files on interesting boundaries (key prefixes) to make propagation of sst files less write amplifying (covering the whole key space).
+
+### Performance Improvements
+* Eliminate key copies for internal comparisons while accessing ingested block-based tables.
+* Reduce key comparisons during random access in all block-based tables.
+* BackupEngine avoids unnecessary repeated checksum computation for backing up a table file to the `shared_checksum` directory when using `share_files_with_checksum_naming = kUseDbSessionId` (new default), except on SST files generated before this version of RocksDB, which fall back on using `kLegacyCrc32cAndFileSize`.
+
+## 6.11 (6/12/2020)
+### Bug Fixes
+* Fix consistency checking error swallowing in some cases when options.force_consistency_checks = true.
+* Fix possible false NotFound status from batched MultiGet using index type kHashSearch.
+* Fix corruption caused by enabling delete triggered compaction (NewCompactOnDeletionCollectorFactory) in universal compaction mode, along with parallel compactions. The bug can result in two parallel compactions picking the same input files, resulting in the DB resurrecting older and deleted versions of some keys.
+* Fix a use-after-free bug in best-efforts recovery. column_family_memtables_ needs to point to valid ColumnFamilySet.
+* Let best-efforts recovery ignore corrupted files during table loading.
+* Fix corrupt key read from ingested file when iterator direction switches from reverse to forward at a key that is a prefix of another key in the same file. It is only possible in files with a non-zero global seqno.
+* Fix abnormally large estimate from GetApproximateSizes when a range starts near the end of one SST file and near the beginning of another. Now GetApproximateSizes consistently and fairly includes the size of SST metadata in addition to data blocks, attributing metadata proportionally among the data blocks based on their size.
+* Fix potential file descriptor leakage in PosixEnv's IsDirectory() and NewRandomAccessFile().
+* Fix false negative from the VerifyChecksum() API when there is a checksum mismatch in an index partition block in a BlockBasedTable format table file (index_type is kTwoLevelIndexSearch).
+* Fix sst_dump to return non-zero exit code if the specified file is not a recognized SST file or fails requested checks.
+* Fix incorrect results from batched MultiGet for duplicate keys, when the duplicate key matches the largest key of an SST file and the value type for the key in the file is a merge value.
+
+### Public API Change
+* Flush(..., column_family) may return Status::ColumnFamilyDropped() instead of Status::InvalidArgument() if column_family is dropped while processing the flush request.
+* BlobDB now explicitly disallows using the default column family's storage directories as blob directory.
+* DeleteRange now returns `Status::InvalidArgument` if the range's end key comes before its start key according to the user comparator. Previously the behavior was undefined.
+* ldb now uses options.force_consistency_checks = true by default and "--disable_consistency_checks" is added to disable it.
+* DB::OpenForReadOnly no longer creates files or directories if the named DB does not exist, unless create_if_missing is set to true.
+* The consistency checks that validate LSM state changes (table file additions/deletions during flushes and compactions) are now stricter, more efficient, and no longer optional, i.e. they are performed even if `force_consistency_checks` is `false`.
+* Disable delete triggered compaction (NewCompactOnDeletionCollectorFactory) in universal compaction mode and num_levels = 1 in order to avoid a corruption bug.
+* `pin_l0_filter_and_index_blocks_in_cache` no longer applies to L0 files larger than `1.5 * write_buffer_size` to give more predictable memory usage. Such L0 files may exist due to intra-L0 compaction, external file ingestion, or user dynamically changing `write_buffer_size` (note, however, that files that are already pinned will continue being pinned, even after such a dynamic change).
+* In point-in-time wal recovery mode, fail database recovery in case of IOError while reading the WAL to avoid data loss.
+* A new method `Env::LowerThreadPoolCPUPriority(Priority, CpuPriority)` is added to `Env` to be able to lower to a specific priority such as `CpuPriority::kIdle`.
+
+### New Features
+* sst_dump to add a new --readahead_size argument. Users can specify read size when scanning the data. Sst_dump also tries to prefetch tail part of the SST files so usually some number of I/Os are saved there too.
+* Generate file checksum in SstFileWriter if Options.file_checksum_gen_factory is set. The checksum and checksum function name are stored in ExternalSstFileInfo after the sst file write is finished.
+* Add a value_size_soft_limit in read options which limits the cumulative value size of keys read in batches in MultiGet. Once the cumulative value size of found keys exceeds read_options.value_size_soft_limit, all the remaining keys are returned with status Abort without further finding their values. By default the value_size_soft_limit is std::numeric_limits<uint64_t>::max().
+* Enable SST file ingestion with file checksum information when calling IngestExternalFiles(const std::vector<IngestExternalFileArg>& args). Added files_checksums and files_checksum_func_names to IngestExternalFileArg such that user can ingest the sst files with their file checksum information. Added verify_file_checksum to IngestExternalFileOptions (default is True). To be backward compatible, if DB does not enable file checksum or user does not provide checksum information (vectors of files_checksums and files_checksum_func_names are both empty), verification of file checksum is always sucessful. If DB enables file checksum, DB will always generate the checksum for each ingested SST file during Prepare stage of ingestion and store the checksum in Manifest, unless verify_file_checksum is False and checksum information is provided by the application. In this case, we only verify the checksum function name and directly store the ingested checksum in Manifest. If verify_file_checksum is set to True, DB will verify the ingested checksum and function name with the genrated ones. Any mismatch will fail the ingestion. Note that, if IngestExternalFileOptions::write_global_seqno is True, the seqno will be changed in the ingested file. Therefore, the checksum of the file will be changed. In this case, a new checksum will be generated after the seqno is updated and be stored in the Manifest.
+
+### Performance Improvements
+* Eliminate redundant key comparisons during random access in block-based tables.
+
+## 6.10 (5/2/2020)
+### Bug Fixes
+* Fix wrong result being read from ingested file. May happen when a key in the file happen to be prefix of another key also in the file. The issue can further cause more data corruption. The issue exists with rocksdb >= 5.0.0 since DB::IngestExternalFile() was introduced.
+* Finish implementation of BlockBasedTableOptions::IndexType::kBinarySearchWithFirstKey. It's now ready for use. Significantly reduces read amplification in some setups, especially for iterator seeks.
+* Fix a bug by updating CURRENT file so that it points to the correct MANIFEST file after best-efforts recovery.
+* Fixed a bug where ColumnFamilyHandle objects were not cleaned up in case an error happened during BlobDB's open after the base DB had been opened.
+* Fix a potential undefined behavior caused by trying to dereference nullable pointer (timestamp argument) in DB::MultiGet.
+* Fix a bug caused by not including user timestamp in MultiGet LookupKey construction. This can lead to wrong query result since the trailing bytes of a user key, if not shorter than timestamp, will be mistaken for user timestamp.
+* Fix a bug caused by using wrong compare function when sorting the input keys of MultiGet with timestamps.
 * Upgraded version of bzip library (1.0.6 -> 1.0.8) used with RocksJava to address potential vulnerabilities if an attacker can manipulate compressed data saved and loaded by RocksDB (not normal). See issue #6703.
 
-## 6.8.1 (03/30/2020)
+### Public API Change
+* Add a ConfigOptions argument to the APIs dealing with converting options to and from strings and files.  The ConfigOptions is meant to replace some of the options (such as input_strings_escaped and ignore_unknown_options) and allow for more parameters to be passed in the future without changing the function signature.
+* Add NewFileChecksumGenCrc32cFactory to the file checksum public API, such that the builtin Crc32c based file checksum generator factory can be used by applications.
+* Add IsDirectory to Env and FS to indicate if a path is a directory.
+
+### New Features
+* Added support for pipelined & parallel compression optimization for `BlockBasedTableBuilder`. This optimization makes block building, block compression and block appending a pipeline, and uses multiple threads to accelerate block compression. Users can set `CompressionOptions::parallel_threads` greater than 1 to enable compression parallelism. This feature is experimental for now.
+* Provide an allocator for memkind to be used with block cache. This is to work with memory technologies (Intel DCPMM is one such technology currently available) that require different libraries for allocation and management (such as PMDK and memkind). The high capacities available make it possible to provision large caches (up to several TBs in size) beyond what is achievable with DRAM.
+* Option `max_background_flushes` can be set dynamically using DB::SetDBOptions().
+* Added functionality in sst_dump tool to check the compressed file size for different compression levels and print the time spent on compressing files with each compression type. Added arguments `--compression_level_from` and `--compression_level_to` to report size of all compression levels and one compression_type must be specified with it so that it will report compressed sizes of one compression type with different levels.
+* Added statistics for redundant insertions into block cache: rocksdb.block.cache.*add.redundant. (There is currently no coordination to ensure that only one thread loads a table block when many threads are trying to access that same table block.)
+
+### Bug Fixes
+* Fix a bug when making options.bottommost_compression, options.compression_opts and options.bottommost_compression_opts dynamically changeable: the modified values are not written to option files or returned back to users when being queried.
+* Fix a bug where index key comparisons were unaccounted in `PerfContext::user_key_comparison_count` for lookups in files written with `format_version >= 3`.
+* Fix many bloom.filter statistics not being updated in batch MultiGet.
+
+### Performance Improvements
+* Improve performance of batch MultiGet with partitioned filters, by sharing block cache lookups to applicable filter blocks.
+* Reduced memory copies when fetching and uncompressing compressed blocks from sst files.
+
+## 6.9.0 (03/29/2020)
 ### Behavior changes
-* Since RocksDB 6.8.0, ttl-based FIFO compaction can drop a file whose oldest key becomes older than options.ttl while others have not. This fix reverts this and makes ttl-based FIFO compaction use the file's flush time as the criterion. This fix also requires that max_open_files = -1 and compaction_options_fifo.allow_compaction = false to function properly.
+* Since RocksDB 6.8, ttl-based FIFO compaction can drop a file whose oldest key becomes older than options.ttl while others have not. This fix reverts this and makes ttl-based FIFO compaction use the file's flush time as the criterion. This fix also requires that max_open_files = -1 and compaction_options_fifo.allow_compaction = false to function properly.
+
+### Public API Change
+* Fix spelling so that API now has correctly spelled transaction state name `COMMITTED`, while the old misspelled `COMMITED` is still available as an alias.
+* Updated default format_version in BlockBasedTableOptions from 2 to 4. SST files generated with the new default can be read by RocksDB versions 5.16 and newer, and use more efficient encoding of keys in index blocks.
+* A new parameter `CreateBackupOptions` is added to both `BackupEngine::CreateNewBackup` and `BackupEngine::CreateNewBackupWithMetadata`, you can decrease CPU priority of `BackupEngine`'s background threads by setting `decrease_background_thread_cpu_priority` and `background_thread_cpu_priority` in `CreateBackupOptions`.
+* Updated the public API of SST file checksum. Introduce the FileChecksumGenFactory to create the FileChecksumGenerator for each SST file, such that the FileChecksumGenerator is not shared and it can be more general for checksum implementations. Changed the FileChecksumGenerator interface from Value, Extend, and GetChecksum to Update, Finalize, and GetChecksum. Finalize should be only called once after all data is processed to generate the final checksum. Temproal data should be maintained by the FileChecksumGenerator object itself and finally it can return the checksum string.
+
+### Bug Fixes
+* Fix a bug where range tombstone blocks in ingested files were cached incorrectly during ingestion. If range tombstones were read from those incorrectly cached blocks, the keys they covered would be exposed.
+* Fix a data race that might cause crash when calling DB::GetCreationTimeOfOldestFile() by a small chance. The bug was introduced in 6.6 Release.
+* Fix a bug where a boolean value optimize_filters_for_hits was for max threads when calling load table handles after a flush or compaction. The value is correct to 1. The bug should not cause user visible problems.
+* Fix a bug which might crash the service when write buffer manager fails to insert the dummy handle to the block cache.
+
+### Performance Improvements
+* In CompactRange, for levels starting from 0, if the level does not have any file with any key falling in the specified range, the level is skipped. So instead of always compacting from level 0, the compaction starts from the first level with keys in the specified range until the last such level.
+* Reduced memory copy when reading sst footer and blobdb in direct IO mode.
+* When restarting a database with large numbers of sst files, large amount of CPU time is spent on getting logical block size of the sst files, which slows down the starting progress, this inefficiency is optimized away with an internal cache for the logical block sizes.
+
+### New Features
+* Basic support for user timestamp in iterator. Seek/SeekToFirst/Next and lower/upper bounds are supported. Reverse iteration is not supported. Merge is not considered.
+* When file lock failure when the lock is held by the current process, return acquiring time and thread ID in the error message.
+* Added a new option, best_efforts_recovery (default: false), to allow database to open in a db dir with missing table files. During best efforts recovery, missing table files are ignored, and database recovers to the most recent state without missing table file. Cross-column-family consistency is not guaranteed even if WAL is enabled.
+* options.bottommost_compression, options.compression_opts and options.bottommost_compression_opts are now dynamically changeable.
 
 ## 6.8.0 (02/24/2020)
 ### Java API Changes

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:

LANGUAGE-BINDINGS.md

@@ -10,7 +10,9 @@ This is the list of all known third-party language bindings for RocksDB. If some
 * Ruby - http://rubygems.org/gems/rocksdb-ruby
 * Haskell - https://hackage.haskell.org/package/rocksdb-haskell
 * PHP - https://github.com/Photonios/rocksdb-php
-* C# - https://github.com/warrenfalk/rocksdb-sharp
+* C#
+    * https://github.com/warrenfalk/rocksdb-sharp
+    * https://github.com/curiosity-ai/rocksdb-sharp
 * Rust
     * https://github.com/pingcap/rust-rocksdb (used in production fork of https://github.com/spacejam/rust-rocksdb)
     * https://github.com/spacejam/rust-rocksdb

PLUGINS.md

@@ -0,0 +1,3 @@
+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

README.md

@@ -1,8 +1,9 @@
 ## RocksDB: A Persistent Key-Value Store for Flash and RAM Storage
 
-[![Linux/Mac Build Status](https://travis-ci.org/facebook/rocksdb.svg?branch=master)](https://travis-ci.org/facebook/rocksdb)
-[![Windows 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)
+[![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-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)
@@ -24,7 +25,7 @@ The public interface is in `include/`.  Callers should not include or
 rely on the details of any other header files in this package.  Those
 internal APIs may be changed without warning.
 
-Design discussions are conducted in https://www.facebook.com/groups/rocksdb.dev/
+Design discussions are conducted in https://www.facebook.com/groups/rocksdb.dev/ and https://rocksdb.slack.com/
 
 ## License
 

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.
@@ -96,13 +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.
+

appveyor.yml

@@ -1,6 +1,6 @@
 version: 1.0.{build}
 
-image: Visual Studio 2017
+image: Visual Studio 2019
 
 environment:
   JAVA_HOME: C:\Program Files\Java\jdk1.8.0
@@ -34,7 +34,8 @@ install:
   - cd snappy-1.1.7
   - mkdir build
   - cd build
-  - cmake -G "%CMAKE_GENERATOR%" ..
+  - if DEFINED CMAKE_PLATEFORM_NAME (set "PLATEFORM_OPT=-A %CMAKE_PLATEFORM_NAME%")
+  - cmake .. -G "%CMAKE_GENERATOR%" %PLATEFORM_OPT%
   - msbuild Snappy.sln /p:Configuration=Debug /p:Platform=x64
   - msbuild Snappy.sln /p:Configuration=Release /p:Platform=x64
   - echo "Building LZ4 dependency..."
@@ -57,7 +58,8 @@ install:
 before_build:
   - md %APPVEYOR_BUILD_FOLDER%\build
   - cd %APPVEYOR_BUILD_FOLDER%\build
-  - cmake -G "%CMAKE_GENERATOR%" -DCMAKE_BUILD_TYPE=Debug -DOPTDBG=1 -DPORTABLE=1 -DSNAPPY=1 -DLZ4=1 -DZSTD=1 -DXPRESS=1 -DJNI=1 ..
+  - if DEFINED CMAKE_PLATEFORM_NAME (set "PLATEFORM_OPT=-A %CMAKE_PLATEFORM_NAME%")
+  - cmake .. -G "%CMAKE_GENERATOR%" %PLATEFORM_OPT% %CMAKE_OPT% -DCMAKE_BUILD_TYPE=Debug -DOPTDBG=1 -DPORTABLE=1 -DSNAPPY=1 -DLZ4=1 -DZSTD=1 -DXPRESS=1 -DJNI=1 -DWITH_ALL_TESTS=0
   - cd ..
 
 build:
@@ -68,7 +70,7 @@ build:
 test:
 
 test_script:
-  - ps: build_tools\run_ci_db_test.ps1 -SuiteRun db_basic_test,db_test2,db_test,env_basic_test,env_test,db_merge_operand_test -Concurrency 8
+  - ps: build_tools\run_ci_db_test.ps1 -SuiteRun db_basic_test,env_basic_test -Concurrency 8
 
 on_failure:
   - cmd: 7z a build-failed.zip %APPVEYOR_BUILD_FOLDER%\build\ && appveyor PushArtifact build-failed.zip

buckifier/buckify_rocksdb.py

@@ -20,14 +20,14 @@ from util import ColorString
 # User can pass extra dependencies as a JSON object via command line, and this
 # script can include these dependencies in the generate TARGETS file.
 # Usage:
-# $python buckifier/buckify_rocksdb.py
+# $python3 buckifier/buckify_rocksdb.py
 # (This generates a TARGET file without user-specified dependency for unit
 # tests.)
-# $python buckifier/buckify_rocksdb.py \
-#        '{"fake": { \
-#                      "extra_deps": [":test_dep", "//fakes/module:mock1"],  \
-#                      "extra_compiler_flags": ["-DROCKSDB_LITE", "-Os"], \
-#                  } \
+# $python3 buckifier/buckify_rocksdb.py \
+#        '{"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
@@ -48,8 +48,8 @@ def parse_src_mk(repo_path):
         if '=' in line:
             current_src = line.split('=')[0].strip()
             src_files[current_src] = []
-        elif '.cc' in line:
-            src_path = line.split('.cc')[0].strip() + '.cc'
+        elif '.c' in line:
+            src_path = line.split('\\')[0].strip()
             src_files[current_src].append(src_path)
     return src_files
 
@@ -69,45 +69,28 @@ def get_cc_files(repo_path):
     return cc_files
 
 
-# Get tests from Makefile
-def get_tests(repo_path):
+# Get non_parallel tests from Makefile
+def get_non_parallel_tests(repo_path):
     Makefile = repo_path + "/Makefile"
 
-    # Dictionary TEST_NAME => IS_PARALLEL
-    tests = {}
+    s = set({})
 
-    found_tests = False
+    found_non_parallel_tests = False
     for line in open(Makefile):
         line = line.strip()
-        if line.startswith("TESTS ="):
-            found_tests = True
-        elif found_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()
-                tests[line] = False
+                s.add(line)
             else:
-                # we consumed all the tests
+                # we consumed all the non_parallel tests
                 break
 
-    found_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.endswith("\\"):
-                # remove the trailing \
-                line = line[:-1]
-                line = line.strip()
-                tests[line] = True
-            else:
-                # we consumed all the parallel tests
-                break
-
-    return tests
-
+    return s
 
 # Parse extra dependencies passed by user from command line
 def get_dependencies():
@@ -140,18 +123,38 @@ 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 tests from Makefile
-    tests = get_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 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,
+        extra_external_deps="",
+        link_whole=True)
     # rocksdb_test_lib
     TARGETS.add_library(
         "rocksdb_test_lib",
@@ -159,7 +162,10 @@ def generate_targets(repo_path, deps_map):
         src_mk.get("TEST_LIB_SOURCES", []) +
         src_mk.get("EXP_LIB_SOURCES", []) +
         src_mk.get("ANALYZER_LIB_SOURCES", []),
-        [":rocksdb_lib"])
+        [":rocksdb_lib"],
+        extra_external_deps=""" + [
+        ("googletest", None, "gtest"),
+    ]""")
     # rocksdb_tools_lib
     TARGETS.add_library(
         "rocksdb_tools_lib",
@@ -168,40 +174,50 @@ def generate_targets(repo_path, deps_map):
         ["test_util/testutil.cc"],
         [":rocksdb_lib"])
     # rocksdb_stress_lib
-    TARGETS.add_library(
+    TARGETS.add_rocksdb_library(
         "rocksdb_stress_lib",
         src_mk.get("ANALYZER_LIB_SOURCES", [])
         + src_mk.get('STRESS_LIB_SOURCES', [])
-        + ["test_util/testutil.cc"],
-        [":rocksdb_lib"])
+        + ["test_util/testutil.cc"])
+
+    print("Extra dependencies:\n{0}".format(json.dumps(deps_map)))
+
+    # Dictionary test executable name -> relative source file path
+    test_source_map = {}
+    print(src_mk)
+
+    # c_test.c is added through TARGETS.add_c_test(). If there
+    # are more than one .c test file, we need to extend
+    # TARGETS.add_c_test() to include other C tests too.
+    for test_src in src_mk.get("TEST_MAIN_SOURCES_C", []):
+        if test_src != 'db/c_test.c':
+            print("Don't know how to deal with " + test_src)
+            return False
+    TARGETS.add_c_test()
+
+    for test_src in src_mk.get("TEST_MAIN_SOURCES", []):
+        test = test_src.split('.c')[0].strip().split('/')[-1].strip()
+        test_source_map[test] = test_src
+        print("" + test + " " + test_src)
 
-    print("Extra dependencies:\n{0}".format(str(deps_map)))
-    # test for every test we found in the Makefile
     for target_alias, deps in deps_map.items():
-        for test in sorted(tests):
-            match_src = [src for src in cc_files if ("/%s.c" % test) in src]
-            if len(match_src) == 0:
-                print(ColorString.warning("Cannot find .cc file for %s" % test))
-                continue
-            elif len(match_src) > 1:
-                print(ColorString.warning("Found more than one .cc for %s" % test))
-                print(match_src)
+        for test, test_src in sorted(test_source_map.items()):
+            if len(test) == 0:
+                print(ColorString.warning("Failed to get test name for %s" % test_src))
                 continue
 
-            assert(len(match_src) == 1)
-            is_parallel = tests[test]
             test_target_name = \
                 test if not target_alias else test + "_" + target_alias
             TARGETS.register_test(
                 test_target_name,
-                match_src[0],
-                is_parallel,
-                deps['extra_deps'],
-                deps['extra_compiler_flags'])
+                test_src,
+                test not in non_parallel_tests,
+                json.dumps(deps['extra_deps']),
+                json.dumps(deps['extra_compiler_flags']))
 
             if test in _EXPORTED_TEST_LIBS:
                 test_library = "%s_lib" % test_target_name
-                TARGETS.add_library(test_library, match_src, [":rocksdb_test_lib"])
+                TARGETS.add_library(test_library, [test_src], [":rocksdb_test_lib"])
     TARGETS.flush_tests()
 
     print(ColorString.info("Generated TARGETS Summary:"))
@@ -220,6 +236,7 @@ def get_rocksdb_path():
 
     return rocksdb_path
 
+
 def exit_with_error(msg):
     print(ColorString.error(msg))
     sys.exit(1)

buckifier/check_buck_targets.sh

@@ -0,0 +1,32 @@
+#!/usr/bin/env bash
+# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
+# If clang_format_diff.py command is not specfied, we assume we are able to
+# access directly without any path.
+
+TGT_DIFF=`git diff TARGETS | head -n 1`
+
+if [ ! -z "$TGT_DIFF" ]
+then
+  echo "TARGETS file has uncommitted changes. Skip this check."
+  exit 0
+fi
+
+echo Backup original TARGETS file.
+
+cp TARGETS TARGETS.bkp
+
+${PYTHON:-python3} buckifier/buckify_rocksdb.py
+
+TGT_DIFF=`git diff TARGETS | head -n 1`
+
+if [ -z "$TGT_DIFF" ]
+then
+  mv TARGETS.bkp TARGETS
+  exit 0
+else
+  echo "Please run '${PYTHON:-python3} buckifier/buckify_rocksdb.py' to update TARGETS file."
+  echo "Do not manually update TARGETS file."
+  ${PYTHON:-python3} --version
+  mv TARGETS.bkp TARGETS
+  exit 1
+fi

buckifier/targets_builder.py

@@ -25,10 +25,12 @@ 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, 'w')
-        self.targets_file.write(targets_cfg.rocksdb_target_header)
+        self.targets_file = open(path, 'wb')
+        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
         self.total_test = 0
@@ -37,26 +39,68 @@ class TARGETSBuilder(object):
     def __del__(self):
         self.targets_file.close()
 
-    def add_library(self, name, srcs, deps=None, headers=None):
+    def add_library(self, name, srcs, deps=None, headers=None,
+                    extra_external_deps="", link_whole=False):
         headers_attr_prefix = ""
         if headers is None:
             headers_attr_prefix = "auto_"
             headers = "AutoHeaders.RECURSIVE_GLOB"
+        else:
+            headers = "[" + pretty_list(headers) + "]"
         self.targets_file.write(targets_cfg.library_template.format(
             name=name,
             srcs=pretty_list(srcs),
             headers_attr_prefix=headers_attr_prefix,
             headers=headers,
-            deps=pretty_list(deps)))
+            deps=pretty_list(deps),
+            extra_external_deps=extra_external_deps,
+            link_whole=link_whole).encode("utf-8"))
+        self.total_lib = self.total_lib + 1
+
+    def add_rocksdb_library(self, name, srcs, headers=None):
+        headers_attr_prefix = ""
+        if headers is None:
+            headers_attr_prefix = "auto_"
+            headers = "AutoHeaders.RECURSIVE_GLOB"
+        else:
+            headers = "[" + pretty_list(headers) + "]"
+        self.targets_file.write(targets_cfg.rocksdb_library_template.format(
+            name=name,
+            srcs=pretty_list(srcs),
+            headers_attr_prefix=headers_attr_prefix,
+            headers=headers).encode("utf-8"))
         self.total_lib = self.total_lib + 1
 
     def add_binary(self, name, srcs, deps=None):
-        self.targets_file.write(targets_cfg.binary_template % (
-            name,
-            pretty_list(srcs),
-            pretty_list(deps)))
+        self.targets_file.write(targets_cfg.binary_template.format(
+            name=name,
+            srcs=pretty_list(srcs),
+            deps=pretty_list(deps)).encode("utf-8"))
         self.total_bin = self.total_bin + 1
 
+    def add_c_test(self):
+        self.targets_file.write(b"""
+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
+
+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,
                       test_name,
                       src,
@@ -76,5 +120,5 @@ class TARGETSBuilder(object):
         self.total_test = self.total_test + 1
 
     def flush_tests(self):
-        self.targets_file.write(targets_cfg.unittests_template % self.tests_cfg)
+        self.targets_file.write(targets_cfg.unittests_template.format(tests=self.tests_cfg).encode("utf-8"))
         self.tests_cfg = ""

buckifier/targets_cfg.py

@@ -4,7 +4,9 @@ from __future__ import division
 from __future__ import print_function
 from __future__ import unicode_literals
 
-rocksdb_target_header = """# This file \100generated by `python buckifier/buckify_rocksdb.py`
+rocksdb_target_header_template = \
+    """# 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.
@@ -30,15 +32,17 @@ ROCKSDB_EXTERNAL_DEPS = [
     ("zlib", None, "z"),
     ("gflags", None, "gflags"),
     ("lz4", None, "lz4"),
-    ("zstd", None),
-    ("tbb", None),
-    ("googletest", None, "gtest"),
+    ("zstd", None, "zstd"),
 ]
 
 ROCKSDB_OS_DEPS = [
     (
         "linux",
-        ["third-party//numa:numa", "third-party//liburing:uring"],
+        ["third-party//numa:numa", "third-party//liburing:uring", "third-party//tbb:tbb"],
+    ),
+    (
+        "macos",
+        ["third-party//tbb:tbb"],
     ),
 ]
 
@@ -56,17 +60,27 @@ ROCKSDB_OS_PREPROCESSOR_FLAGS = [
             "-DHAVE_SSE42",
             "-DLIBURING",
             "-DNUMA",
+            "-DROCKSDB_PLATFORM_POSIX",
+            "-DROCKSDB_LIB_IO_POSIX",
+            "-DTBB",
         ],
     ),
     (
         "macos",
-        ["-DOS_MACOSX"],
+        [
+            "-DOS_MACOSX",
+            "-DROCKSDB_PLATFORM_POSIX",
+            "-DROCKSDB_LIB_IO_POSIX",
+            "-DTBB",
+        ],
+    ),
+    (
+        "windows",
+        [ "-DOS_WIN", "-DWIN32", "-D_MBCS", "-DWIN64", "-DNOMINMAX" ]
     ),
 ]
 
 ROCKSDB_PREPROCESSOR_FLAGS = [
-    "-DROCKSDB_PLATFORM_POSIX",
-    "-DROCKSDB_LIB_IO_POSIX",
     "-DROCKSDB_SUPPORT_THREAD_LOCAL",
 
     # Flags to enable libs we include
@@ -77,21 +91,22 @@ ROCKSDB_PREPROCESSOR_FLAGS = [
     "-DZSTD",
     "-DZSTD_STATIC_LINKING_ONLY",
     "-DGFLAGS=gflags",
-    "-DTBB",
 
     # 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")
 
@@ -114,6 +129,11 @@ ROCKSDB_OS_DEPS += ([(
     "linux",
     ["third-party//jemalloc:headers"],
 )] if sanitizer == "" else [])
+
+ROCKSDB_LIB_DEPS = [
+    ":rocksdb_lib",
+    ":rocksdb_test_lib",
+] if not is_opt_mode else [":rocksdb_lib"]
 """
 
 
@@ -127,19 +147,38 @@ 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},
+)
+"""
+
+rocksdb_library_template = """
+cpp_library(
+    name = "{name}",
+    srcs = [{srcs}],
+    {headers_attr_prefix}headers = {headers},
+    arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
+    compiler_flags = ROCKSDB_COMPILER_FLAGS,
+    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,
 )
 """
 
 binary_template = """
 cpp_binary(
-    name = "%s",
-    srcs = [%s],
+    name = "{name}",
+    srcs = [{srcs}],
     arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
     compiler_flags = ROCKSDB_COMPILER_FLAGS,
     preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
-    deps = [%s],
+    include_paths = ROCKSDB_INCLUDE_PATHS,
+    deps = [{deps}],
     external_deps = ROCKSDB_EXTERNAL_DEPS,
 )
 """
@@ -156,24 +195,24 @@ test_cfg_template = """    [
 unittests_template = """
 # [test_name, test_src, test_type, extra_deps, extra_compiler_flags]
 ROCKS_TESTS = [
-%s]
+{tests}]
 
 # Generate a test rule for each entry in ROCKS_TESTS
 # Do not build the tests in opt mode, since SyncPoint and other test code
 # will not be included.
 [
-    test_binary(
-        extra_compiler_flags = extra_compiler_flags,
-        extra_deps = extra_deps,
-        parallelism = parallelism,
-        rocksdb_arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
-        rocksdb_compiler_flags = ROCKSDB_COMPILER_FLAGS,
-        rocksdb_external_deps = ROCKSDB_EXTERNAL_DEPS,
-        rocksdb_os_deps = ROCKSDB_OS_DEPS,
-        rocksdb_os_preprocessor_flags = ROCKSDB_OS_PREPROCESSOR_FLAGS,
-        rocksdb_preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
-        test_cc = test_cc,
-        test_name = test_name,
+    cpp_unittest(
+        name = test_name,
+        srcs = [test_cc],
+        arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
+        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"),
+        ],
     )
     for test_name, test_cc, parallelism, extra_deps, extra_compiler_flags in ROCKS_TESTS
     if not is_opt_mode

build_tools/build_detect_platform

@@ -9,6 +9,7 @@
 #   PLATFORM_LDFLAGS            Linker flags
 #   JAVA_LDFLAGS                Linker flags for RocksDBJava
 #   JAVA_STATIC_LDFLAGS         Linker flags for RocksDBJava static build
+#   JAVAC_ARGS                  Arguments for javac
 #   PLATFORM_SHARED_EXT         Extension for shared libraries
 #   PLATFORM_SHARED_LDFLAGS     Flags for building shared library
 #   PLATFORM_SHARED_CFLAGS      Flags for compiling objects for shared library
@@ -27,6 +28,7 @@
 #       -DZSTD                      if the ZSTD library is present
 #       -DNUMA                      if the NUMA library is present
 #       -DTBB                       if the TBB library is present
+#       -DMEMKIND                   if the memkind library is present
 #
 # Using gflags in rocksdb:
 # Our project depends on gflags, which requires users to take some extra steps
@@ -43,8 +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"
 
@@ -58,8 +65,16 @@ if [ -z "$ROCKSDB_NO_FBCODE" -a -d /mnt/gvfs/third-party ]; then
       source "$PWD/build_tools/fbcode_config4.8.1.sh"
     elif [ -n "$ROCKSDB_FBCODE_BUILD_WITH_5xx" ]; then
       source "$PWD/build_tools/fbcode_config.sh"
-    else
+    elif [ -n "$ROCKSDB_FBCODE_BUILD_WITH_PLATFORM007" ]; then
       source "$PWD/build_tools/fbcode_config_platform007.sh"
+    elif [ -n "$ROCKSDB_FBCODE_BUILD_WITH_PLATFORM009" ]; then
+      source "$PWD/build_tools/fbcode_config_platform009.sh"
+    elif [ -z "$USE_CLANG" ]; then
+        # Still use platform007 for gcc by default for build break on
+        # some hosts.
+      source "$PWD/build_tools/fbcode_config_platform007.sh"
+    else
+      source "$PWD/build_tools/fbcode_config_platform009.sh"
     fi
 fi
 
@@ -87,6 +102,16 @@ if test -z "$CXX"; then
     fi
 fi
 
+if test -z "$AR"; then
+    if [ -x "$(command -v gcc-ar)" ]; then
+        AR=gcc-ar
+    elif [ -x "$(command -v llvm-ar)" ]; then
+        AR=llvm-ar
+    else
+        AR=ar
+    fi
+fi
+
 # Detect OS
 if test -z "$TARGET_OS"; then
     TARGET_OS=`uname -s`
@@ -149,10 +174,10 @@ case "$TARGET_OS" in
         else
             PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -latomic"
         fi
-        PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lpthread -lrt"
+        PLATFORM_LDFLAGS="$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 $PLATFORM_CXXFLAGS -x c++ - -luring -o /dev/null 2>/dev/null  <<EOF
               #include <liburing.h>
               int main() {
                 struct io_uring ring;
@@ -190,6 +215,17 @@ EOF
         PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lpthread"
         # PORT_FILES=port/freebsd/freebsd_specific.cc
         ;;
+    GNU/kFreeBSD)
+        PLATFORM=OS_GNU_KFREEBSD
+        COMMON_FLAGS="$COMMON_FLAGS -DOS_GNU_KFREEBSD"
+        if [ -z "$USE_CLANG" ]; then
+            COMMON_FLAGS="$COMMON_FLAGS -fno-builtin-memcmp"
+        else
+            PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -latomic"
+        fi
+        PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lpthread -lrt"
+        # PORT_FILES=port/gnu_kfreebsd/gnu_kfreebsd_specific.cc
+        ;;
     NetBSD)
         PLATFORM=OS_NETBSD
         COMMON_FLAGS="$COMMON_FLAGS -fno-builtin-memcmp -D_REENTRANT -DOS_NETBSD"
@@ -239,15 +275,20 @@ esac
 PLATFORM_CXXFLAGS="$PLATFORM_CXXFLAGS ${CXXFLAGS}"
 JAVA_LDFLAGS="$PLATFORM_LDFLAGS"
 JAVA_STATIC_LDFLAGS="$PLATFORM_LDFLAGS"
+JAVAC_ARGS="-source 7"
 
 if [ "$CROSS_COMPILE" = "true" -o "$FBCODE_BUILD" = "true" ]; then
     # Cross-compiling; do not try any compilation tests.
     # Also don't need any compilation tests if compiling on fbcode
+    if [ "$FBCODE_BUILD" = "true" ]; then
+      # Enable backtrace on fbcode since the necessary libraries are present
+      COMMON_FLAGS="$COMMON_FLAGS -DROCKSDB_BACKTRACE"
+    fi
     true
 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() {
@@ -263,7 +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
@@ -278,30 +319,38 @@ EOF
         # Test whether gflags library is installed
         # http://gflags.github.io/gflags/
         # check if the namespace is gflags
-        $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
+        if $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
           #include <gflags/gflags.h>
+          using namespace GFLAGS_NAMESPACE;
           int main() {}
 EOF
-        if [ "$?" = 0 ]; then
-            COMMON_FLAGS="$COMMON_FLAGS -DGFLAGS=1"
-            PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lgflags"
-        else
-          # check if namespace is google
-          $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
+        then
+          COMMON_FLAGS="$COMMON_FLAGS -DGFLAGS=1"
+          PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lgflags"
+        # check if namespace is gflags
+        elif $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
+            #include <gflags/gflags.h>
+            using namespace gflags;
+            int main() {}
+EOF
+        then
+          COMMON_FLAGS="$COMMON_FLAGS -DGFLAGS=1 -DGFLAGS_NAMESPACE=gflags"
+          PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lgflags"
+        # check if namespace is google
+        elif $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
             #include <gflags/gflags.h>
             using namespace google;
             int main() {}
 EOF
-          if [ "$?" = 0 ]; then
-              COMMON_FLAGS="$COMMON_FLAGS -DGFLAGS=google"
-              PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lgflags"
-          fi
+        then
+          COMMON_FLAGS="$COMMON_FLAGS -DGFLAGS=1 -DGFLAGS_NAMESPACE=google"
+          PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lgflags"
         fi
     fi
 
     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
@@ -314,7 +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
@@ -327,7 +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() {}
@@ -341,7 +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
@@ -354,7 +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() {}
@@ -368,7 +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
@@ -381,7 +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
@@ -402,7 +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"
@@ -411,7 +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);
@@ -424,9 +473,25 @@ EOF
         fi
     fi
 
+    if ! test $ROCKSDB_DISABLE_MEMKIND; then
+        # Test whether memkind library is installed
+        $CXX $PLATFORM_CXXFLAGS $COMMON_FLAGS -lmemkind -x c++ - -o /dev/null 2>/dev/null  <<EOF
+          #include <memkind.h>
+          int main() {
+            memkind_malloc(MEMKIND_DAX_KMEM, 1024);
+            return 0;
+          }
+EOF
+        if [ "$?" = 0 ]; then
+            COMMON_FLAGS="$COMMON_FLAGS -DMEMKIND"
+            PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lmemkind"
+            JAVA_LDFLAGS="$JAVA_LDFLAGS -lmemkind"
+        fi
+    fi
+
     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;
@@ -441,7 +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 $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
           #include <execinfo.h>
           int main() {
             void* frames[1];
@@ -453,7 +518,7 @@ EOF
             COMMON_FLAGS="$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];
@@ -470,7 +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;
           }
@@ -482,7 +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);
@@ -496,7 +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();
@@ -508,6 +573,20 @@ EOF
         fi
     fi
 
+    if ! test $ROCKSDB_DISABLE_AUXV_GETAUXVAL; then
+        # Test whether getauxval is supported
+        $CXX $PLATFORM_CXXFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+          #include <sys/auxv.h>
+          int main() {
+            uint64_t auxv = getauxval(AT_HWCAP);
+            (void)auxv;
+          }
+EOF
+        if [ "$?" = 0 ]; then
+            COMMON_FLAGS="$COMMON_FLAGS -DROCKSDB_AUXV_GETAUXVAL_PRESENT"
+        fi
+    fi
+
     if ! test $ROCKSDB_DISABLE_ALIGNED_NEW; then
         # Test whether c++17 aligned-new is supported
         $CXX $PLATFORM_CXXFLAGS -faligned-new -x c++ - -o /dev/null 2>/dev/null <<EOF
@@ -524,7 +603,7 @@ fi
 # -Wshorten-64-to-32 breaks compilation on FreeBSD i386
 if ! [ "$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
@@ -575,6 +654,35 @@ else
   if test "$USE_SSE"; then
     TRY_SSE_ETC="1"
   fi
+
+  if [[ "${PLATFORM}" == "OS_MACOSX" ]]; then
+    # For portability compile for macOS 10.12 (2016) or newer
+    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
 
 if test "$TRY_SSE_ETC"; then
@@ -584,11 +692,18 @@ if test "$TRY_SSE_ETC"; then
   # It doesn't even really check that your current CPU is compatible.
   #
   # SSE4.2 available since nehalem, ca. 2008-2010
+  # Includes POPCNT for BitsSetToOne, BitParity
   TRY_SSE42="-msse4.2"
   # PCLMUL available since westmere, ca. 2010-2011
   TRY_PCLMUL="-mpclmul"
   # AVX2 available since haswell, ca. 2013-2015
   TRY_AVX2="-mavx2"
+  # BMI available since haswell, ca. 2013-2015
+  # Primarily for TZCNT for CountTrailingZeroBits
+  TRY_BMI="-mbmi"
+  # LZCNT available since haswell, ca. 2013-2015
+  # For FloorLog2
+  TRY_LZCNT="-mlzcnt"
 fi
 
 $CXX $PLATFORM_CXXFLAGS $COMMON_FLAGS $TRY_SSE42 -x c++ - -o /dev/null 2>/dev/null <<EOF
@@ -637,6 +752,34 @@ elif test "$USE_SSE"; then
   echo "warning: USE_SSE specified but compiler could not use AVX2 intrinsics, disabling" >&2
 fi
 
+$CXX $PLATFORM_CXXFLAGS $COMMON_FLAGS $TRY_BMI -x c++ - -o /dev/null 2>/dev/null <<EOF
+  #include <cstdint>
+  #include <immintrin.h>
+  int main(int argc, char *argv[]) {
+    (void)argv;
+    return (int)_tzcnt_u64((uint64_t)argc);
+  }
+EOF
+if [ "$?" = 0 ]; then
+  COMMON_FLAGS="$COMMON_FLAGS $TRY_BMI -DHAVE_BMI"
+elif test "$USE_SSE"; then
+  echo "warning: USE_SSE specified but compiler could not use BMI intrinsics, disabling" >&2
+fi
+
+$CXX $PLATFORM_CXXFLAGS $COMMON_FLAGS $TRY_LZCNT -x c++ - -o /dev/null 2>/dev/null <<EOF
+  #include <cstdint>
+  #include <immintrin.h>
+  int main(int argc, char *argv[]) {
+    (void)argv;
+    return (int)_lzcnt_u64((uint64_t)argc);
+  }
+EOF
+if [ "$?" = 0 ]; then
+  COMMON_FLAGS="$COMMON_FLAGS $TRY_LZCNT -DHAVE_LZCNT"
+elif test "$USE_SSE"; then
+  echo "warning: USE_SSE specified but compiler could not use LZCNT intrinsics, disabling" >&2
+fi
+
 $CXX $PLATFORM_CXXFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null <<EOF
   #include <cstdint>
   int main() {
@@ -668,6 +811,7 @@ EOF
   fi
 fi
 
+
 if [ "$FBCODE_BUILD" != "true" -a "$PLATFORM" = OS_LINUX ]; then
   $CXX $COMMON_FLAGS $PLATFORM_SHARED_CFLAGS -x c++ -c - -o test_dl.o 2>/dev/null <<EOF
   void dummy_func() {}
@@ -692,10 +836,16 @@ ROCKSDB_PATCH=`build_tools/version.sh patch`
 
 echo "CC=$CC" >> "$OUTPUT"
 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"
 echo "PLATFORM_CXXFLAGS=$PLATFORM_CXXFLAGS" >> "$OUTPUT"
@@ -728,3 +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/dependencies_platform009.sh

@@ -0,0 +1,20 @@
+# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
+GCC_BASE=/mnt/gvfs/third-party2/gcc/1795efe5f06778c15a92c8f9a2aba5dc496d9d4d/9.x/centos7-native/3bed279
+CLANG_BASE=/mnt/gvfs/third-party2/llvm-fb/7318eaac22659b6ff2fe43918e4b69fd0772a8a7/9.0.0/platform009/651ee30
+LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/4959b39cfbe5965a37c861c4c327fa7c5c759b87/9.x/platform009/9202ce7
+GLIBC_BASE=/mnt/gvfs/third-party2/glibc/45ce3375cdc77ecb2520bbf8f0ecddd3f98efd7a/2.30/platform009/f259413
+SNAPPY_BASE=/mnt/gvfs/third-party2/snappy/be4de3205e029101b18aa8103daa696c2bef3b19/1.1.3/platform009/7f3b187
+ZLIB_BASE=/mnt/gvfs/third-party2/zlib/3c160ac5c67e257501e24c6c1d00ad5e01d73db6/1.2.8/platform009/7f3b187
+BZIP2_BASE=/mnt/gvfs/third-party2/bzip2/73a237ac5bc0a5f5d67b39b8d253cfebaab88684/1.0.6/platform009/7f3b187
+LZ4_BASE=/mnt/gvfs/third-party2/lz4/ec6573523b0ce55ef6373a4801189027cf07bb2c/1.9.1/platform009/7f3b187
+ZSTD_BASE=/mnt/gvfs/third-party2/zstd/64c58a207d2495e83abc57a500a956df09b79a7c/1.4.x/platform009/ba86d1f
+GFLAGS_BASE=/mnt/gvfs/third-party2/gflags/824d0a8a5abb5b121afd1b35fc3896407ea50092/2.2.0/platform009/7f3b187
+JEMALLOC_BASE=/mnt/gvfs/third-party2/jemalloc/d9aef9feb850b168a68736420f217b01cce11a89/master/platform009/c305944
+NUMA_BASE=/mnt/gvfs/third-party2/numa/0af65f71e23a67bf65dc91b11f95caa39325c432/2.0.11/platform009/7f3b187
+LIBUNWIND_BASE=/mnt/gvfs/third-party2/libunwind/02486dac347645d31dce116f44e1de3177315be2/1.4/platform009/5191652
+TBB_BASE=/mnt/gvfs/third-party2/tbb/2e0ec671e550bfca347300bf3f789d9c0fff24ad/2018_U5/platform009/7f3b187
+LIBURING_BASE=/mnt/gvfs/third-party2/liburing/70dbd9cfee63a25611417d09433a86d7711b3990/20200729/platform009/7f3b187
+KERNEL_HEADERS_BASE=/mnt/gvfs/third-party2/kernel-headers/32b8a2407b634df3f8f948ba373fc4acc6a18296/fb/platform009/da39a3e
+BINUTILS_BASE=/mnt/gvfs/third-party2/binutils/08634589372fa5f237bfd374e8c644a8364e78c1/2.32/platform009/ba86d1f/
+VALGRIND_BASE=/mnt/gvfs/third-party2/valgrind/6ae525939ad02e5e676855082fbbc7828dbafeac/3.15.0/platform009/7f3b187
+LUA_BASE=/mnt/gvfs/third-party2/lua/162efd9561a3d21f6869f4814011e9cf1b3ff4dc/5.3.4/platform009/a6271c4

build_tools/fbcode_config.sh

@@ -109,6 +109,7 @@ if [ -z "$USE_CLANG" ]; then
   # gcc
   CC="$GCC_BASE/bin/gcc"
   CXX="$GCC_BASE/bin/g++"
+  AR="$GCC_BASE/bin/gcc-ar"
 
   CFLAGS+=" -B$BINUTILS/gold"
   CFLAGS+=" -isystem $GLIBC_INCLUDE"
@@ -119,6 +120,7 @@ else
   CLANG_INCLUDE="$CLANG_LIB/clang/stable/include"
   CC="$CLANG_BIN/clang"
   CXX="$CLANG_BIN/clang++"
+  AR="$CLANG_BIN/llvm-ar"
 
   KERNEL_HEADERS_INCLUDE="$KERNEL_HEADERS_BASE/include"
 

build_tools/fbcode_config4.8.1.sh

@@ -69,6 +69,7 @@ if [ -z "$USE_CLANG" ]; then
   # gcc
   CC="$GCC_BASE/bin/gcc"
   CXX="$GCC_BASE/bin/g++"
+  CXX="$GCC_BASE/bin/gcc-ar"
 
   CFLAGS="-B$BINUTILS/gold -m64 -mtune=generic"
   CFLAGS+=" -isystem $GLIBC_INCLUDE"
@@ -81,6 +82,7 @@ else
   CLANG_INCLUDE="$CLANG_LIB/clang/*/include"
   CC="$CLANG_BIN/clang"
   CXX="$CLANG_BIN/clang++"
+  AR="$CLANG_BIN/llvm-ar"
 
   KERNEL_HEADERS_INCLUDE="$KERNEL_HEADERS_BASE/include/"
 

build_tools/fbcode_config_platform007.sh

@@ -118,6 +118,7 @@ if [ -z "$USE_CLANG" ]; then
   # gcc
   CC="$GCC_BASE/bin/gcc"
   CXX="$GCC_BASE/bin/g++"
+  AR="$GCC_BASE/bin/gcc-ar"
 
   CFLAGS+=" -B$BINUTILS/gold"
   CFLAGS+=" -isystem $LIBGCC_INCLUDE"
@@ -128,6 +129,7 @@ else
   CLANG_INCLUDE="$CLANG_LIB/clang/stable/include"
   CC="$CLANG_BIN/clang"
   CXX="$CLANG_BIN/clang++"
+  AR="$CLANG_BIN/llvm-ar"
 
   KERNEL_HEADERS_INCLUDE="$KERNEL_HEADERS_BASE/include"
 

build_tools/fbcode_config_platform009.sh

@@ -0,0 +1,172 @@
+#!/bin/sh
+# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
+#
+# Set environment variables so that we can compile rocksdb using
+# fbcode settings.  It uses the latest g++ and clang compilers and also
+# uses jemalloc
+# Environment variables that change the behavior of this script:
+# PIC_BUILD -- if true, it will only take pic versions of libraries from fbcode. libraries that don't have pic variant will not be included
+
+
+BASEDIR=`dirname $BASH_SOURCE`
+source "$BASEDIR/dependencies_platform009.sh"
+
+CFLAGS=""
+
+# libgcc
+LIBGCC_INCLUDE="$LIBGCC_BASE/include/c++/9.3.0"
+LIBGCC_LIBS=" -L $LIBGCC_BASE/lib"
+
+# glibc
+GLIBC_INCLUDE="$GLIBC_BASE/include"
+GLIBC_LIBS=" -L $GLIBC_BASE/lib"
+
+# snappy
+SNAPPY_INCLUDE=" -I $SNAPPY_BASE/include/"
+if test -z $PIC_BUILD; then
+  SNAPPY_LIBS=" $SNAPPY_BASE/lib/libsnappy.a"
+else
+  SNAPPY_LIBS=" $SNAPPY_BASE/lib/libsnappy_pic.a"
+fi
+CFLAGS+=" -DSNAPPY"
+
+if test -z $PIC_BUILD; then
+  # location of zlib headers and libraries
+  ZLIB_INCLUDE=" -I $ZLIB_BASE/include/"
+  ZLIB_LIBS=" $ZLIB_BASE/lib/libz.a"
+  CFLAGS+=" -DZLIB"
+
+  # location of bzip headers and libraries
+  BZIP_INCLUDE=" -I $BZIP2_BASE/include/"
+  BZIP_LIBS=" $BZIP2_BASE/lib/libbz2.a"
+  CFLAGS+=" -DBZIP2"
+
+  LZ4_INCLUDE=" -I $LZ4_BASE/include/"
+  LZ4_LIBS=" $LZ4_BASE/lib/liblz4.a"
+  CFLAGS+=" -DLZ4"
+fi
+
+ZSTD_INCLUDE=" -I $ZSTD_BASE/include/"
+if test -z $PIC_BUILD; then
+  ZSTD_LIBS=" $ZSTD_BASE/lib/libzstd.a"
+else
+  ZSTD_LIBS=" $ZSTD_BASE/lib/libzstd_pic.a"
+fi
+CFLAGS+=" -DZSTD"
+
+# location of gflags headers and libraries
+GFLAGS_INCLUDE=" -I $GFLAGS_BASE/include/"
+if test -z $PIC_BUILD; then
+  GFLAGS_LIBS=" $GFLAGS_BASE/lib/libgflags.a"
+else
+  GFLAGS_LIBS=" $GFLAGS_BASE/lib/libgflags_pic.a"
+fi
+CFLAGS+=" -DGFLAGS=gflags"
+
+# location of jemalloc
+JEMALLOC_INCLUDE=" -I $JEMALLOC_BASE/include/"
+JEMALLOC_LIB=" $JEMALLOC_BASE/lib/libjemalloc.a"
+
+if test -z $PIC_BUILD; then
+  # location of numa
+  NUMA_INCLUDE=" -I $NUMA_BASE/include/"
+  NUMA_LIB=" $NUMA_BASE/lib/libnuma.a"
+  CFLAGS+=" -DNUMA"
+
+  # location of libunwind
+  LIBUNWIND="$LIBUNWIND_BASE/lib/libunwind.a"
+fi
+
+# location of TBB
+TBB_INCLUDE=" -isystem $TBB_BASE/include/"
+if test -z $PIC_BUILD; then
+  TBB_LIBS="$TBB_BASE/lib/libtbb.a"
+else
+  TBB_LIBS="$TBB_BASE/lib/libtbb_pic.a"
+fi
+CFLAGS+=" -DTBB"
+
+# location of LIBURING
+LIBURING_INCLUDE=" -isystem $LIBURING_BASE/include/"
+if test -z $PIC_BUILD; then
+  LIBURING_LIBS="$LIBURING_BASE/lib/liburing.a"
+else
+  LIBURING_LIBS="$LIBURING_BASE/lib/liburing_pic.a"
+fi
+CFLAGS+=" -DLIBURING"
+
+test "$USE_SSE" || USE_SSE=1
+export USE_SSE
+test "$PORTABLE" || PORTABLE=1
+export PORTABLE
+
+BINUTILS="$BINUTILS_BASE/bin"
+AR="$BINUTILS/ar"
+
+DEPS_INCLUDE="$SNAPPY_INCLUDE $ZLIB_INCLUDE $BZIP_INCLUDE $LZ4_INCLUDE $ZSTD_INCLUDE $GFLAGS_INCLUDE $NUMA_INCLUDE $TBB_INCLUDE $LIBURING_INCLUDE"
+
+STDLIBS="-L $GCC_BASE/lib64"
+
+CLANG_BIN="$CLANG_BASE/bin"
+CLANG_LIB="$CLANG_BASE/lib"
+CLANG_SRC="$CLANG_BASE/../../src"
+
+CLANG_ANALYZER="$CLANG_BIN/clang++"
+CLANG_SCAN_BUILD="$CLANG_SRC/llvm/clang/tools/scan-build/bin/scan-build"
+
+if [ -z "$USE_CLANG" ]; then
+  # gcc
+  CC="$GCC_BASE/bin/gcc"
+  CXX="$GCC_BASE/bin/g++"
+  AR="$GCC_BASE/bin/gcc-ar"
+
+  CFLAGS+=" -B$BINUTILS/gold"
+  CFLAGS+=" -isystem $LIBGCC_INCLUDE"
+  CFLAGS+=" -isystem $GLIBC_INCLUDE"
+  JEMALLOC=1
+else
+  # clang
+  CLANG_INCLUDE="$CLANG_LIB/clang/stable/include"
+  CC="$CLANG_BIN/clang"
+  CXX="$CLANG_BIN/clang++"
+  AR="$CLANG_BIN/llvm-ar"
+
+  KERNEL_HEADERS_INCLUDE="$KERNEL_HEADERS_BASE/include"
+
+  CFLAGS+=" -B$BINUTILS/gold -nostdinc -nostdlib"
+  CFLAGS+=" -isystem $LIBGCC_BASE/include/c++/9.x "
+  CFLAGS+=" -isystem $LIBGCC_BASE/include/c++/9.x/x86_64-facebook-linux "
+  CFLAGS+=" -isystem $GLIBC_INCLUDE"
+  CFLAGS+=" -isystem $LIBGCC_INCLUDE"
+  CFLAGS+=" -isystem $CLANG_INCLUDE"
+  CFLAGS+=" -isystem $KERNEL_HEADERS_INCLUDE/linux "
+  CFLAGS+=" -isystem $KERNEL_HEADERS_INCLUDE "
+  CFLAGS+=" -Wno-expansion-to-defined "
+  CXXFLAGS="-nostdinc++"
+fi
+
+CFLAGS+=" $DEPS_INCLUDE"
+CFLAGS+=" -DROCKSDB_PLATFORM_POSIX -DROCKSDB_LIB_IO_POSIX -DROCKSDB_FALLOCATE_PRESENT -DROCKSDB_MALLOC_USABLE_SIZE -DROCKSDB_RANGESYNC_PRESENT -DROCKSDB_SCHED_GETCPU_PRESENT -DROCKSDB_SUPPORT_THREAD_LOCAL -DHAVE_SSE42 -DROCKSDB_IOURING_PRESENT"
+CXXFLAGS+=" $CFLAGS"
+
+EXEC_LDFLAGS=" $SNAPPY_LIBS $ZLIB_LIBS $BZIP_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS $NUMA_LIB $TBB_LIBS $LIBURING_LIBS"
+EXEC_LDFLAGS+=" -B$BINUTILS/gold"
+EXEC_LDFLAGS+=" -Wl,--dynamic-linker,/usr/local/fbcode/platform009/lib/ld.so"
+EXEC_LDFLAGS+=" $LIBUNWIND"
+EXEC_LDFLAGS+=" -Wl,-rpath=/usr/local/fbcode/platform009/lib"
+# required by libtbb
+EXEC_LDFLAGS+=" -ldl"
+
+PLATFORM_LDFLAGS="$LIBGCC_LIBS $GLIBC_LIBS $STDLIBS -lgcc -lstdc++"
+
+EXEC_LDFLAGS_SHARED="$SNAPPY_LIBS $ZLIB_LIBS $BZIP_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS $TBB_LIBS $LIBURING_LIBS"
+
+VALGRIND_VER="$VALGRIND_BASE/bin/"
+
+# lua not supported because it's on track for deprecation, I think
+LUA_PATH=
+LUA_LIB=
+
+USE_FOLLY_DISTRIBUTED_MUTEX=1
+
+export CC CXX AR CFLAGS CXXFLAGS EXEC_LDFLAGS EXEC_LDFLAGS_SHARED VALGRIND_VER JEMALLOC_LIB JEMALLOC_INCLUDE CLANG_ANALYZER CLANG_SCAN_BUILD LUA_PATH LUA_LIB

build_tools/format-diff.sh

@@ -2,41 +2,99 @@
 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 # If clang_format_diff.py command is not specfied, we assume we are able to
 # access directly without any path.
-if [ -z $CLANG_FORMAT_DIFF ]
-then
-CLANG_FORMAT_DIFF="clang-format-diff.py"
-fi
 
-# Check clang-format-diff.py
-if ! which $CLANG_FORMAT_DIFF &> /dev/null
-then
-  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 "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."
-  exit 128
-fi
+print_usage () {
+  echo "Usage:"
+  echo "format-diff.sh [OPTIONS]"
+  echo "-c: check only."
+  echo "-h: print this message."
+}
 
-# Check argparse, a library that clang-format-diff.py requires.
-python 2>/dev/null << EOF
-import argparse
-EOF
+while getopts ':ch' OPTION; do
+  case "$OPTION" in
+    c)
+      CHECK_ONLY=1
+      ;;
+    h)
+      print_usage
+      exit 1
+      ;;
+    ?)
+      print_usage
+      exit 1
+      ;;
+  esac
+done
 
-if [ "$?" != 0 ]
-then
-  echo "To run clang-format-diff.py, we'll need the library "argparse" to be"
-  echo "installed. You can try either of the follow ways to install it:"
-  echo "  1. Manually download argparse: https://pypi.python.org/pypi/argparse"
-  echo "  2. easy_install argparse (if you have easy_install)"
-  echo "  3. pip install argparse (if you have pip)"
-  exit 129
+REPO_ROOT="$(git rev-parse --show-toplevel)"
+
+if [ "$CLANG_FORMAT_DIFF" ]; then
+  echo "Note: CLANG_FORMAT_DIFF='$CLANG_FORMAT_DIFF'"
+  # Dry run to confirm dependencies like argparse
+  if $CLANG_FORMAT_DIFF --help >/dev/null < /dev/null; then
+    true #Good
+  else
+    exit 128
+  fi
+else
+  # First try directly executing the possibilities
+  if clang-format-diff.py --help &> /dev/null < /dev/null; then
+    CLANG_FORMAT_DIFF=clang-format-diff.py
+  elif $REPO_ROOT/clang-format-diff.py --help &> /dev/null < /dev/null; then
+    CLANG_FORMAT_DIFF=$REPO_ROOT/clang-format-diff.py
+  else
+    # This probably means we need to directly invoke the interpreter.
+    # But first find clang-format-diff.py
+    if [ -f "$REPO_ROOT/clang-format-diff.py" ]; then
+      CFD_PATH="$REPO_ROOT/clang-format-diff.py"
+    elif which clang-format-diff.py &> /dev/null; then
+      CFD_PATH="$(which clang-format-diff.py)"
+    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 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 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
+    if echo import argparse | ${PYTHON:-python3}; then
+      true # Good
+    else
+      echo "To run clang-format-diff.py, we'll need the library "argparse" to be"
+      echo "installed. You can try either of the follow ways to install it:"
+      echo "  1. Manually download argparse: https://pypi.python.org/pypi/argparse"
+      echo "  2. easy_install argparse (if you have easy_install)"
+      echo "  3. pip install argparse (if you have pip)"
+      exit 129
+    fi
+    # Unfortunately, some machines have a Python2 clang-format-diff.py
+    # installed but only a Python3 interpreter installed. Unfortunately,
+    # automatic 2to3 migration is insufficient, so suggest downloading latest.
+    if grep -q "print '" "$CFD_PATH" && \
+       ${PYTHON:-python3} --version | grep -q 'ython 3'; then
+      echo "You have clang-format-diff.py for Python 2 but are using a Python 3"
+      echo "interpreter (${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
+    if $CLANG_FORMAT_DIFF --help >/dev/null < /dev/null; then
+      true #Good
+    else
+      exit 128
+    fi
+  fi
 fi
 
 # TODO(kailiu) following work is not complete since we still need to figure
@@ -78,15 +136,25 @@ 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" ]
 then
   echo "Nothing needs to be reformatted!"
   exit 0
+elif [ $CHECK_ONLY ]
+then
+  echo "Your change has unformatted code. Please run make format!"
+  if [ $VERBOSE_CHECK ]; then
+    clang-format --version
+    echo "$diffs"
+  fi
+  exit 1
 fi
 
 # Highlight the insertion/deletion from the clang-format-diff.py's output
@@ -121,7 +189,7 @@ if [ -z "$uncommitted_code" ]
 then
   git diff -U0 "$FORMAT_UPSTREAM_MERGE_BASE" | $CLANG_FORMAT_DIFF -i -p 1
 else
-  git diff -U0 HEAD^ | $CLANG_FORMAT_DIFF -i -p 1
+  git diff -U0 HEAD | $CLANG_FORMAT_DIFF -i -p 1
 fi
 echo "Files reformatted!"
 

build_tools/gnu_parallel

@@ -5801,7 +5801,7 @@ sub workdir {
 		    . "-" . $self->seq();
 	    } else {
 		$workdir = $opt::workdir;
-		# Rsync treats /./ special. We dont want that
+		# Rsync treats /./ special. We don't want that
 		$workdir =~ s:/\./:/:g; # Remove /./
 		$workdir =~ s:/+$::; # Remove ending / if any
 		$workdir =~ s:^\./::g; # Remove starting ./ if any

build_tools/make_package.sh

@@ -103,31 +103,26 @@ function main() {
   gem_install fpm
 
   make static_lib
-  make install INSTALL_PATH=package
-
-  cd package
-
-  LIB_DIR=lib
-  if [[ -z "$ARCH" ]]; then
-      ARCH=$(getconf LONG_BIT)
-  fi
-  if [[ ("$FPM_OUTPUT" = "rpm") && ($ARCH -eq 64) ]]; then
-      mv lib lib64
-      LIB_DIR=lib64
+  LIBDIR=/usr/lib
+  if [[ $FPM_OUTPUT = "rpm" ]]; then
+      LIBDIR=$(rpm --eval '%_libdir')
   fi
 
+  rm -rf package
+  make install DESTDIR=package PREFIX=/usr LIBDIR=$LIBDIR
+
   fpm \
     -s dir \
     -t $FPM_OUTPUT \
+    -C package \
     -n rocksdb \
     -v $1 \
-    --prefix /usr \
     --url http://rocksdb.org/ \
     -m rocksdb@fb.com \
     --license BSD \
     --vendor Facebook \
     --description "RocksDB is an embeddable persistent key-value store for fast storage." \
-    include $LIB_DIR
+    usr
 }
 
 # shellcheck disable=SC2068

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 +411,74 @@ 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
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
+#
+# RocksDB blackbox stress/crash test
+#
+BLACKBOX_STRESS_CRASH_TEST_COMMANDS="[
+    {
+        'name':'Rocksdb Blackbox Stress and Crash Test',
+        'oncall':'$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 blackbox crash tests',
+                'timeout': 86400,
+                '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
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
+#
+# RocksDB whitebox stress/crash test
+#
+WHITEBOX_STRESS_CRASH_TEST_COMMANDS="[
+    {
+        'name':'Rocksdb Whitebox Stress and Crash Test',
+        'oncall':'$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 whitebox crash tests',
+                'timeout': 86400,
+                '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
             },
@@ -434,14 +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
             },
@@ -464,14 +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
             },
@@ -493,7 +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
             }
@@ -516,7 +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',
                 $PARSER
             }
@@ -539,7 +636,55 @@ ASAN_CRASH_TEST_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug asan_crash_test',
                 'timeout': 86400,
-                'shell':'$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
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
+#
+# RocksDB blackbox crash testing under address sanitizer
+#
+ASAN_BLACKBOX_CRASH_TEST_COMMANDS="[
+    {
+        'name':'Rocksdb blackbox crash test under ASAN',
+        'oncall':'$ONCALL',
+        'executeLocal': 'true',
+        'timeout': 86400,
+        'steps': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug blackbox asan_crash_test',
+                'timeout': 86400,
+                '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
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
+#
+# RocksDB whitebox crash testing under address sanitizer
+#
+ASAN_WHITEBOX_CRASH_TEST_COMMANDS="[
+    {
+        'name':'Rocksdb whitebox crash test under ASAN',
+        'oncall':'$ONCALL',
+        'executeLocal': 'true',
+        'timeout': 86400,
+        'steps': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug whitebox asan_crash_test',
+                'timeout': 86400,
+                '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',
                 $PARSER
             },
@@ -563,7 +708,7 @@ ASAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug asan_crash_test_with_atomic_flush',
                 'timeout': 86400,
-                'shell':'$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',
                 $PARSER
             },
@@ -587,7 +732,7 @@ ASAN_CRASH_TEST_WITH_TXN_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug asan_crash_test_with_txn',
                 'timeout': 86400,
-                'shell':'$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
             },
@@ -609,7 +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
             }
@@ -619,7 +764,7 @@ UBSAN_TEST_COMMANDS="[
 ]"
 
 #
-# RocksDB crash testing under udnefined behavior sanitizer
+# RocksDB crash testing under undefined behavior sanitizer
 #
 UBSAN_CRASH_TEST_COMMANDS="[
     {
@@ -632,7 +777,55 @@ 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
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
+#
+# RocksDB crash testing under undefined behavior sanitizer
+#
+UBSAN_BLACKBOX_CRASH_TEST_COMMANDS="[
+    {
+        'name':'Rocksdb blackbox crash test under UBSAN',
+        'oncall':'$ONCALL',
+        'executeLocal': 'true',
+        'timeout': 86400,
+        'steps': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug blackbox ubsan_crash_test',
+                'timeout': 86400,
+                '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
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
+#
+# RocksDB crash testing under undefined behavior sanitizer
+#
+UBSAN_WHITEBOX_CRASH_TEST_COMMANDS="[
+    {
+        'name':'Rocksdb whitebox crash test under UBSAN',
+        'oncall':'$ONCALL',
+        'executeLocal': 'true',
+        'timeout': 86400,
+        'steps': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug whitebox ubsan_crash_test',
+                'timeout': 86400,
+                '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',
                 $PARSER
             },
@@ -656,7 +849,7 @@ UBSAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug ubsan_crash_test_with_atomic_flush',
                 'timeout': 86400,
-                'shell':'$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',
                 $PARSER
             },
@@ -680,7 +873,7 @@ UBSAN_CRASH_TEST_WITH_TXN_COMMANDS="[
             {
                 'name':'Build and run RocksDB debug ubsan_crash_test_with_txn',
                 'timeout': 86400,
-                'shell':'$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',
                 $PARSER
             },
@@ -704,7 +897,7 @@ VALGRIND_TEST_COMMANDS="[
             {
                 'name':'Run RocksDB debug unit tests',
                 'timeout': 86400,
-                'shell':'$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
             },
@@ -727,7 +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
             },
@@ -750,7 +943,55 @@ 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
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
+#
+# RocksDB blackbox crash test under TSAN
+#
+TSAN_BLACKBOX_CRASH_TEST_COMMANDS="[
+    {
+        'name':'Rocksdb Blackbox Crash Test under TSAN',
+        'oncall':'$ONCALL',
+        'executeLocal': 'true',
+        'timeout': 86400,
+        'steps': [
+            $CLEANUP_ENV,
+            {
+                'name':'Compile and run',
+                'timeout': 86400,
+                '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
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
+#
+# RocksDB whitebox crash test under TSAN
+#
+TSAN_WHITEBOX_CRASH_TEST_COMMANDS="[
+    {
+        'name':'Rocksdb Whitebox Crash Test under TSAN',
+        'oncall':'$ONCALL',
+        'executeLocal': 'true',
+        'timeout': 86400,
+        'steps': [
+            $CLEANUP_ENV,
+            {
+                'name':'Compile and run',
+                'timeout': 86400,
+                '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
             },
@@ -774,7 +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
             },
@@ -798,7 +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
             },
@@ -818,6 +1059,8 @@ run_format_compatible()
   rm -rf /dev/shm/rocksdb
   mkdir /dev/shm/rocksdb
 
+  export https_proxy="fwdproxy:8080"
+  
   tools/check_format_compatible.sh
 }
 
@@ -830,7 +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
             },
@@ -852,6 +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
 }
 
@@ -864,7 +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
             },
@@ -895,6 +1139,7 @@ run_regression()
   strip librocksdb.a
   send_size_to_ods static_lib_stripped $(stat --printf="%s" librocksdb.a)
 
+  make clean
   make -j$(nproc) shared_lib
   send_size_to_ods shared_lib $(stat --printf="%s" `readlink -f librocksdb.so`)
   strip `readlink -f librocksdb.so`
@@ -907,6 +1152,7 @@ run_regression()
   strip librocksdb.a
   send_size_to_ods static_lib_lite_stripped $(stat --printf="%s" librocksdb.a)
 
+  make clean
   make LITE=1 -j$(nproc) shared_lib
   send_size_to_ods shared_lib_lite $(stat --printf="%s" `readlink -f librocksdb.so`)
   strip `readlink -f librocksdb.so`
@@ -921,7 +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
             },
@@ -942,7 +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
             },
@@ -992,12 +1238,21 @@ case $1 in
   stress_crash)
     echo $STRESS_CRASH_TEST_COMMANDS
     ;;
+  blackbox_stress_crash)
+    echo $BLACKBOX_STRESS_CRASH_TEST_COMMANDS
+    ;;
+  whitebox_stress_crash)
+    echo $WHITEBOX_STRESS_CRASH_TEST_COMMANDS
+    ;;
   stress_crash_with_atomic_flush)
     echo $STRESS_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS
     ;;
   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
     ;;
@@ -1007,6 +1262,12 @@ case $1 in
   asan_crash)
     echo $ASAN_CRASH_TEST_COMMANDS
     ;;
+  blackbox_asan_crash)
+    echo $ASAN_BLACKBOX_CRASH_TEST_COMMANDS
+    ;;
+  whitebox_asan_crash)
+    echo $ASAN_WHITEBOX_CRASH_TEST_COMMANDS
+    ;;
   asan_crash_with_atomic_flush)
     echo $ASAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS
     ;;
@@ -1019,6 +1280,12 @@ case $1 in
   ubsan_crash)
     echo $UBSAN_CRASH_TEST_COMMANDS
     ;;
+  blackbox_ubsan_crash)
+    echo $UBSAN_BLACKBOX_CRASH_TEST_COMMANDS
+    ;;
+  whitebox_ubsan_crash)
+    echo $UBSAN_WHITEBOX_CRASH_TEST_COMMANDS
+    ;;
   ubsan_crash_with_atomic_flush)
     echo $UBSAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS
     ;;
@@ -1034,6 +1301,12 @@ case $1 in
   tsan_crash)
     echo $TSAN_CRASH_TEST_COMMANDS
     ;;
+  blackbox_tsan_crash)
+    echo $TSAN_BLACKBOX_CRASH_TEST_COMMANDS
+    ;;
+  whitebox_tsan_crash)
+    echo $TSAN_WHITEBOX_CRASH_TEST_COMMANDS
+    ;;
   tsan_crash_with_atomic_flush)
     echo $TSAN_CRASH_TEST_WITH_ATOMIC_FLUSH_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
 
- 
+

build_tools/setup_centos7.sh

@@ -1,9 +1,9 @@
 #!/bin/bash
 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
-set -e
+set -ex
 
-ROCKSDB_VERSION="5.10.3"
-ZSTD_VERSION="1.1.3"
+ROCKSDB_VERSION="6.7.3"
+ZSTD_VERSION="1.4.4"
 
 echo "This script configures CentOS with everything needed to build and run RocksDB"
 
@@ -40,5 +40,6 @@ cd /usr/local/rocksdb
 chown -R vagrant:vagrant /usr/local/rocksdb/
 sudo -u vagrant make static_lib
 cd examples/
-sudo -u vagrant make all
-sudo -u vagrant ./c_simple_example
+sudo -u vagrant LD_LIBRARY_PATH=/usr/local/lib/ make all
+sudo -u vagrant LD_LIBRARY_PATH=/usr/local/lib/ ./c_simple_example
+

cache/cache.cc

@@ -0,0 +1,63 @@
+//  Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
+//  This source code is licensed under both the GPLv2 (found in the
+//  COPYING file in the root directory) and Apache 2.0 License
+//  (found in the LICENSE.Apache file in the root directory).
+//
+// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file. See the AUTHORS file for names of contributors.
+
+#include "rocksdb/cache.h"
+
+#include "cache/lru_cache.h"
+#include "rocksdb/utilities/options_type.h"
+#include "util/string_util.h"
+
+namespace ROCKSDB_NAMESPACE {
+#ifndef ROCKSDB_LITE
+static std::unordered_map<std::string, OptionTypeInfo>
+    lru_cache_options_type_info = {
+        {"capacity",
+         {offsetof(struct LRUCacheOptions, capacity), OptionType::kSizeT,
+          OptionVerificationType::kNormal, OptionTypeFlags::kMutable}},
+        {"num_shard_bits",
+         {offsetof(struct LRUCacheOptions, num_shard_bits), OptionType::kInt,
+          OptionVerificationType::kNormal, OptionTypeFlags::kMutable}},
+        {"strict_capacity_limit",
+         {offsetof(struct LRUCacheOptions, strict_capacity_limit),
+          OptionType::kBoolean, OptionVerificationType::kNormal,
+          OptionTypeFlags::kMutable}},
+        {"high_pri_pool_ratio",
+         {offsetof(struct LRUCacheOptions, high_pri_pool_ratio),
+          OptionType::kDouble, OptionVerificationType::kNormal,
+          OptionTypeFlags::kMutable}},
+};
+#endif  // ROCKSDB_LITE
+
+Status Cache::CreateFromString(const ConfigOptions& config_options,
+                               const std::string& value,
+                               std::shared_ptr<Cache>* result) {
+  Status status;
+  std::shared_ptr<Cache> cache;
+  if (value.find('=') == std::string::npos) {
+    cache = NewLRUCache(ParseSizeT(value));
+  } else {
+#ifndef ROCKSDB_LITE
+    LRUCacheOptions cache_opts;
+    status = OptionTypeInfo::ParseStruct(
+        config_options, "", &lru_cache_options_type_info, "", value,
+        reinterpret_cast<char*>(&cache_opts));
+    if (status.ok()) {
+      cache = NewLRUCache(cache_opts);
+    }
+#else
+    (void)config_options;
+    status = Status::NotSupported("Cannot load cache in LITE mode ", value);
+#endif  //! ROCKSDB_LITE
+  }
+  if (status.ok()) {
+    result->swap(cache);
+  }
+  return status;
+}
+}  // namespace ROCKSDB_NAMESPACE

cache/cache_bench.cc

@@ -11,37 +11,52 @@ int main() {
 }
 #else
 
-#include <stdio.h>
 #include <sys/types.h>
+
 #include <cinttypes>
+#include <cstdio>
+#include <limits>
 
 #include "port/port.h"
 #include "rocksdb/cache.h"
 #include "rocksdb/db.h"
 #include "rocksdb/env.h"
+#include "rocksdb/system_clock.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 const uint32_t KB = 1024;
+static constexpr uint32_t KiB = uint32_t{1} << 10;
+static constexpr uint32_t MiB = KiB << 10;
+static constexpr uint64_t GiB = MiB << 10;
 
-DEFINE_int32(threads, 16, "Number of concurrent threads to run.");
-DEFINE_int64(cache_size, 8 * KB * KB,
-             "Number of bytes to use as a cache of uncompressed data.");
-DEFINE_int32(num_shard_bits, 4, "shard_bits.");
+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_int64(max_key, 1 * KB * KB * KB, "Max number of key to place in cache");
-DEFINE_uint64(ops_per_thread, 1200000, "Number of operations per thread.");
+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_bool(populate_cache, false, "Populate cache before operations");
-DEFINE_int32(insert_percent, 40,
-             "Ratio of insert to total workload (expressed as a percentage)");
-DEFINE_int32(lookup_percent, 50,
-             "Ratio of lookup to total workload (expressed as a percentage)");
-DEFINE_int32(erase_percent, 10,
-             "Ratio of erase to total workload (expressed as a percentage)");
+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, "");
 
@@ -49,21 +64,15 @@ namespace ROCKSDB_NAMESPACE {
 
 class CacheBench;
 namespace {
-void deleter(const Slice& /*key*/, void* value) {
-    delete reinterpret_cast<char *>(value);
-}
-
 // State shared by all concurrent executions of the same benchmark.
 class SharedState {
  public:
   explicit SharedState(CacheBench* cache_bench)
       : cv_(&mu_),
-        num_threads_(FLAGS_threads),
         num_initialized_(0),
         start_(false),
         num_done_(0),
-        cache_bench_(cache_bench) {
-  }
+        cache_bench_(cache_bench) {}
 
   ~SharedState() {}
 
@@ -87,13 +96,9 @@ class SharedState {
     num_done_++;
   }
 
-  bool AllInitialized() const {
-    return num_initialized_ >= num_threads_;
-  }
+  bool AllInitialized() const { return num_initialized_ >= FLAGS_threads; }
 
-  bool AllDone() const {
-    return num_done_ >= num_threads_;
-  }
+  bool AllDone() const { return num_done_ >= FLAGS_threads; }
 
   void SetStart() {
     start_ = true;
@@ -107,7 +112,6 @@ class SharedState {
   port::Mutex mu_;
   port::CondVar cv_;
 
-  const uint64_t num_threads_;
   uint64_t num_initialized_;
   bool start_;
   uint64_t num_done_;
@@ -118,17 +122,69 @@ class SharedState {
 // Per-thread state for concurrent executions of the same benchmark.
 struct ThreadState {
   uint32_t tid;
-  Random rnd;
+  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() : num_threads_(FLAGS_threads) {
+  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_) {
@@ -138,30 +194,32 @@ class CacheBench {
     } 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() {
-    Random rnd(1);
-    for (int64_t i = 0; i < FLAGS_cache_size; i++) {
-      uint64_t rand_key = rnd.Next() % FLAGS_max_key;
-      // Cast uint64* to be char*, data would be copied to cache
-      Slice key(reinterpret_cast<char*>(&rand_key), 8);
-      // do insert
-      cache_->Insert(key, new char[10], 1, &deleter);
+    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();
+    const auto& clock = env->GetSystemClock();
 
     PrintEnv();
     SharedState shared(this);
-    std::vector<ThreadState*> threads(num_threads_);
-    for (uint32_t i = 0; i < num_threads_; i++) {
-      threads[i] = new ThreadState(i, &shared);
-      env->StartThread(ThreadBody, threads[i]);
+    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());
@@ -169,7 +227,7 @@ class CacheBench {
         shared.GetCondVar()->Wait();
       }
       // Record start time
-      uint64_t start_time = env->NowMicros();
+      uint64_t start_time = clock->NowMicros();
 
       // Start all threads
       shared.SetStart();
@@ -181,7 +239,7 @@ class CacheBench {
       }
 
       // Record end time
-      uint64_t end_time = env->NowMicros();
+      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);
@@ -192,10 +250,15 @@ class CacheBench {
 
  private:
   std::shared_ptr<Cache> cache_;
-  uint32_t num_threads_;
+  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 = reinterpret_cast<ThreadState*>(v);
+    ThreadState* thread = static_cast<ThreadState*>(v);
     SharedState* shared = thread->shared;
 
     {
@@ -220,40 +283,78 @@ class CacheBench {
   }
 
   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++) {
-      uint64_t rand_key = thread->rnd.Next() % FLAGS_max_key;
-      // Cast uint64* to be char*, data would be copied to cache
-      Slice key(reinterpret_cast<char*>(&rand_key), 8);
-      int32_t prob_op = thread->rnd.Uniform(100);
-      if (prob_op >= 0 && prob_op < FLAGS_insert_percent) {
-        // do insert
-        cache_->Insert(key, new char[10], 1, &deleter);
-      } else if (prob_op -= FLAGS_insert_percent &&
-                 prob_op < FLAGS_lookup_percent) {
-        // do lookup
-        auto handle = cache_->Lookup(key);
+      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;
         }
-      } else if (prob_op -= FLAGS_lookup_percent &&
-                 prob_op < FLAGS_erase_percent) {
+        // 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   : %d\n", FLAGS_threads);
+    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      : %d\n", FLAGS_num_shard_bits);
-    printf("Max key             : %" PRIu64 "\n", FLAGS_max_key);
-    printf("Populate cache      : %d\n", FLAGS_populate_cache);
-    printf("Insert percentage   : %d%%\n", FLAGS_insert_percent);
-    printf("Lookup percentage   : %d%%\n", FLAGS_lookup_percent);
-    printf("Erase percentage    : %d%%\n", FLAGS_erase_percent);
+    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");
   }
 };
@@ -270,6 +371,8 @@ int main(int argc, char** argv) {
   ROCKSDB_NAMESPACE::CacheBench bench;
   if (FLAGS_populate_cache) {
     bench.PopulateCache();
+    printf("Population complete\n");
+    printf("----------------------------\n");
   }
   if (bench.Run()) {
     return 0;

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/cache_test.cc

@@ -117,8 +117,8 @@ class CacheTest : public testing::TestWithParam<std::string> {
 
   void Insert(std::shared_ptr<Cache> cache, int key, int value,
               int charge = 1) {
-    cache->Insert(EncodeKey(key), EncodeValue(value), charge,
-                  &CacheTest::Deleter);
+    EXPECT_OK(cache->Insert(EncodeKey(key), EncodeValue(value), charge,
+                            &CacheTest::Deleter));
   }
 
   void Erase(std::shared_ptr<Cache> cache, int key) {
@@ -167,9 +167,10 @@ TEST_P(CacheTest, UsageTest) {
   for (int i = 1; i < 100; ++i) {
     std::string key(i, 'a');
     auto kv_size = key.size() + 5;
-    cache->Insert(key, reinterpret_cast<void*>(value), kv_size, dumbDeleter);
-    precise_cache->Insert(key, reinterpret_cast<void*>(value), kv_size,
-                          dumbDeleter);
+    ASSERT_OK(cache->Insert(key, reinterpret_cast<void*>(value), kv_size,
+                            dumbDeleter));
+    ASSERT_OK(precise_cache->Insert(key, reinterpret_cast<void*>(value),
+                                    kv_size, dumbDeleter));
     usage += kv_size;
     ASSERT_EQ(usage, cache->GetUsage());
     ASSERT_LT(usage, precise_cache->GetUsage());
@@ -183,10 +184,10 @@ TEST_P(CacheTest, UsageTest) {
   // make sure the cache will be overloaded
   for (uint64_t i = 1; i < kCapacity; ++i) {
     auto key = ToString(i);
-    cache->Insert(key, reinterpret_cast<void*>(value), key.size() + 5,
-                  dumbDeleter);
-    precise_cache->Insert(key, reinterpret_cast<void*>(value), key.size() + 5,
-                          dumbDeleter);
+    ASSERT_OK(cache->Insert(key, reinterpret_cast<void*>(value), key.size() + 5,
+                            dumbDeleter));
+    ASSERT_OK(precise_cache->Insert(key, reinterpret_cast<void*>(value),
+                                    key.size() + 5, dumbDeleter));
   }
 
   // the usage should be close to the capacity
@@ -215,11 +216,12 @@ TEST_P(CacheTest, PinnedUsageTest) {
     auto kv_size = key.size() + 5;
     Cache::Handle* handle;
     Cache::Handle* handle_in_precise_cache;
-    cache->Insert(key, reinterpret_cast<void*>(value), kv_size, dumbDeleter,
-                  &handle);
+    ASSERT_OK(cache->Insert(key, reinterpret_cast<void*>(value), kv_size,
+                            dumbDeleter, &handle));
     assert(handle);
-    precise_cache->Insert(key, reinterpret_cast<void*>(value), kv_size,
-                          dumbDeleter, &handle_in_precise_cache);
+    ASSERT_OK(precise_cache->Insert(key, reinterpret_cast<void*>(value),
+                                    kv_size, dumbDeleter,
+                                    &handle_in_precise_cache));
     assert(handle_in_precise_cache);
     pinned_usage += kv_size;
     ASSERT_EQ(pinned_usage, cache->GetPinnedUsage());
@@ -254,10 +256,10 @@ TEST_P(CacheTest, PinnedUsageTest) {
   // check that overloading the cache does not change the pinned usage
   for (uint64_t i = 1; i < 2 * kCapacity; ++i) {
     auto key = ToString(i);
-    cache->Insert(key, reinterpret_cast<void*>(value), key.size() + 5,
-                  dumbDeleter);
-    precise_cache->Insert(key, reinterpret_cast<void*>(value), key.size() + 5,
-                          dumbDeleter);
+    ASSERT_OK(cache->Insert(key, reinterpret_cast<void*>(value), key.size() + 5,
+                            dumbDeleter));
+    ASSERT_OK(precise_cache->Insert(key, reinterpret_cast<void*>(value),
+                                    key.size() + 5, dumbDeleter));
   }
   ASSERT_EQ(pinned_usage, cache->GetPinnedUsage());
   ASSERT_EQ(precise_cache_pinned_usage, precise_cache->GetPinnedUsage());

cache/clock_cache.cc

@@ -182,7 +182,7 @@ struct CacheHandle {
   void (*deleter)(const Slice&, void* value);
 
   // Flags and counters associated with the cache handle:
-  //   lowest bit: n-cache bit
+  //   lowest bit: in-cache bit
   //   second lowest bit: usage bit
   //   the rest bits: reference count
   // The handle is unused when flags equals to 0. The thread decreases the count
@@ -341,7 +341,8 @@ class ClockCacheShard final : public CacheShard {
   CacheHandle* Insert(const Slice& key, uint32_t hash, void* value,
                       size_t change,
                       void (*deleter)(const Slice& key, void* value),
-                      bool hold_reference, CleanupContext* context);
+                      bool hold_reference, CleanupContext* context,
+                      bool* overwritten);
 
   // Guards list_, head_, and recycle_. In addition, updating table_ also has
   // to hold the mutex, to avoid the cache being in inconsistent state.
@@ -564,7 +565,8 @@ void ClockCacheShard::SetStrictCapacityLimit(bool strict_capacity_limit) {
 CacheHandle* ClockCacheShard::Insert(
     const Slice& key, uint32_t hash, void* value, size_t charge,
     void (*deleter)(const Slice& key, void* value), bool hold_reference,
-    CleanupContext* context) {
+    CleanupContext* context, bool* overwritten) {
+  assert(overwritten != nullptr && *overwritten == false);
   size_t total_charge =
       CacheHandle::CalcTotalCharge(key, charge, metadata_charge_policy_);
   MutexLock l(&mutex_);
@@ -597,6 +599,7 @@ CacheHandle* ClockCacheShard::Insert(
   handle->flags.store(flags, std::memory_order_relaxed);
   HashTable::accessor accessor;
   if (table_.find(accessor, CacheKey(key, hash))) {
+    *overwritten = true;
     CacheHandle* existing_handle = accessor->second;
     table_.erase(accessor);
     UnsetInCache(existing_handle, context);
@@ -619,8 +622,9 @@ Status ClockCacheShard::Insert(const Slice& key, uint32_t hash, void* value,
   char* key_data = new char[key.size()];
   memcpy(key_data, key.data(), key.size());
   Slice key_copy(key_data, key.size());
+  bool overwritten = false;
   CacheHandle* handle = Insert(key_copy, hash, value, charge, deleter,
-                               out_handle != nullptr, &context);
+                               out_handle != nullptr, &context, &overwritten);
   Status s;
   if (out_handle != nullptr) {
     if (handle == nullptr) {
@@ -629,6 +633,10 @@ Status ClockCacheShard::Insert(const Slice& key, uint32_t hash, void* value,
       *out_handle = reinterpret_cast<Cache::Handle*>(handle);
     }
   }
+  if (overwritten) {
+    assert(s.ok());
+    s = Status::OkOverwritten();
+  }
   Cleanup(context);
   return s;
 }

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"
@@ -386,6 +385,7 @@ Status LRUCacheShard::Insert(const Slice& key, uint32_t hash, void* value,
       LRUHandle* old = table_.Insert(e);
       usage_ += total_charge;
       if (old != nullptr) {
+        s = Status::OkOverwritten();
         assert(old->InCache());
         old->SetInCache(false);
         if (!old->HasRefs()) {

cache/lru_cache.h

@@ -67,7 +67,7 @@ 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),
   };
 
@@ -130,7 +130,7 @@ struct LRUHandle {
     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;
@@ -239,7 +239,7 @@ 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:
@@ -328,7 +328,7 @@ class LRUCache
 
   //  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

@@ -30,15 +30,16 @@ class LRUCacheTest : public testing::Test {
     DeleteCache();
     cache_ = reinterpret_cast<LRUCacheShard*>(
         port::cacheline_aligned_alloc(sizeof(LRUCacheShard)));
-    new (cache_) LRUCacheShard(capacity, false /*strict_capcity_limit*/,
+    new (cache_) LRUCacheShard(capacity, false /*strict_capacity_limit*/,
                                high_pri_pool_ratio, use_adaptive_mutex,
                                kDontChargeCacheMetadata);
   }
 
   void Insert(const std::string& key,
               Cache::Priority priority = Cache::Priority::LOW) {
-    cache_->Insert(key, 0 /*hash*/, nullptr /*value*/, 1 /*charge*/,
-                   nullptr /*deleter*/, nullptr /*handle*/, priority);
+    EXPECT_OK(cache_->Insert(key, 0 /*hash*/, nullptr /*value*/, 1 /*charge*/,
+                             nullptr /*deleter*/, nullptr /*handle*/,
+                             priority));
   }
 
   void Insert(char key, Cache::Priority priority = Cache::Priority::LOW) {

cmake/RocksDBConfig.cmake.in

@@ -1,3 +1,54 @@
 @PACKAGE_INIT@
+
+list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_LIST_DIR}/modules")
+
+include(CMakeFindDependencyMacro)
+
+set(GFLAGS_USE_TARGET_NAMESPACE @GFLAGS_USE_TARGET_NAMESPACE@)
+
+if(@WITH_JEMALLOC@)
+  find_dependency(JeMalloc)
+endif()
+
+if(@WITH_GFLAGS@)
+  find_dependency(gflags CONFIG)
+  if(NOT gflags_FOUND)
+    find_dependency(gflags)
+  endif()
+endif()
+
+if(@WITH_SNAPPY@)
+  find_dependency(Snappy CONFIG)
+  if(NOT Snappy_FOUND)
+    find_dependency(Snappy)
+  endif()
+endif()
+
+if(@WITH_ZLIB@)
+  find_dependency(ZLIB)
+endif()
+
+if(@WITH_BZ2@)
+  find_dependency(BZip2)
+endif()
+
+if(@WITH_LZ4@)
+  find_dependency(lz4)
+endif()
+
+if(@WITH_ZSTD@)
+  find_dependency(zstd)
+endif()
+
+if(@WITH_NUMA@)
+  find_dependency(NUMA)
+endif()
+
+if(@WITH_TBB@)
+  find_dependency(TBB)
+endif()
+
+find_dependency(Threads)
+
 include("${CMAKE_CURRENT_LIST_DIR}/RocksDBTargets.cmake")
 check_required_components(RocksDB)

cmake/modules/CxxFlags.cmake

@@ -0,0 +1,7 @@
+macro(get_cxx_std_flags FLAGS_VARIABLE)
+  if( CMAKE_CXX_STANDARD_REQUIRED )
+    set(${FLAGS_VARIABLE} ${CMAKE_CXX${CMAKE_CXX_STANDARD}_STANDARD_COMPILE_OPTION})
+  else()
+    set(${FLAGS_VARIABLE} ${CMAKE_CXX${CMAKE_CXX_STANDARD}_EXTENSION_COMPILE_OPTION})
+  endif()
+endmacro()

cmake/modules/FindSnappy.cmake

@@ -0,0 +1,29 @@
+# - Find Snappy
+# Find the snappy compression library and includes
+#
+# Snappy_INCLUDE_DIRS - where to find snappy.h, etc.
+# Snappy_LIBRARIES - List of libraries when using snappy.
+# Snappy_FOUND - True if snappy found.
+
+find_path(Snappy_INCLUDE_DIRS
+  NAMES snappy.h
+  HINTS ${snappy_ROOT_DIR}/include)
+
+find_library(Snappy_LIBRARIES
+  NAMES snappy
+  HINTS ${snappy_ROOT_DIR}/lib)
+
+include(FindPackageHandleStandardArgs)
+find_package_handle_standard_args(Snappy DEFAULT_MSG Snappy_LIBRARIES Snappy_INCLUDE_DIRS)
+
+mark_as_advanced(
+  Snappy_LIBRARIES
+  Snappy_INCLUDE_DIRS)
+
+if(Snappy_FOUND AND NOT (TARGET Snappy::snappy))
+  add_library (Snappy::snappy UNKNOWN IMPORTED)
+  set_target_properties(Snappy::snappy
+    PROPERTIES
+      IMPORTED_LOCATION ${Snappy_LIBRARIES}
+      INTERFACE_INCLUDE_DIRECTORIES ${Snappy_INCLUDE_DIRS})
+endif()

cmake/modules/Findgflags.cmake

@@ -1,8 +1,8 @@
 # - Find gflags library
 # Find the gflags includes and library
 #
-# gflags_INCLUDE_DIR - where to find gflags.h.
-# gflags_LIBRARIES - List of libraries when using gflags.
+# GFLAGS_INCLUDE_DIR - where to find gflags.h.
+# GFLAGS_LIBRARIES - List of libraries when using gflags.
 # gflags_FOUND - True if gflags found.
 
 find_path(GFLAGS_INCLUDE_DIR

cmake/modules/Findsnappy.cmake

@@ -1,29 +0,0 @@
-# - Find Snappy
-# Find the snappy compression library and includes
-#
-# snappy_INCLUDE_DIRS - where to find snappy.h, etc.
-# snappy_LIBRARIES - List of libraries when using snappy.
-# snappy_FOUND - True if snappy found.
-
-find_path(snappy_INCLUDE_DIRS
-  NAMES snappy.h
-  HINTS ${snappy_ROOT_DIR}/include)
-
-find_library(snappy_LIBRARIES
-  NAMES snappy
-  HINTS ${snappy_ROOT_DIR}/lib)
-
-include(FindPackageHandleStandardArgs)
-find_package_handle_standard_args(snappy DEFAULT_MSG snappy_LIBRARIES snappy_INCLUDE_DIRS)
-
-mark_as_advanced(
-  snappy_LIBRARIES
-  snappy_INCLUDE_DIRS)
-
-if(snappy_FOUND AND NOT (TARGET snappy::snappy))
-  add_library (snappy::snappy UNKNOWN IMPORTED)
-  set_target_properties(snappy::snappy
-    PROPERTIES
-      IMPORTED_LOCATION ${snappy_LIBRARIES}
-      INTERFACE_INCLUDE_DIRECTORIES ${snappy_INCLUDE_DIRS})
-endif()

coverage/coverage_test.sh

@@ -12,21 +12,24 @@ fi
 ROOT=".."
 # Fetch right version of gcov
 if [ -d /mnt/gvfs/third-party -a -z "$CXX" ]; then
-  source $ROOT/build_tools/fbcode_config.sh
+  source $ROOT/build_tools/fbcode_config_platform007.sh
   GCOV=$GCC_BASE/bin/gcov
 else
   GCOV=$(which gcov)
 fi
+echo -e "Using $GCOV"
 
 COVERAGE_DIR="$PWD/COVERAGE_REPORT"
 mkdir -p $COVERAGE_DIR
 
 # Find all gcno files to generate the coverage report
 
+PYTHON=${1:-`which python`}
+echo -e "Using $PYTHON"
 GCNO_FILES=`find $ROOT -name "*.gcno"`
 $GCOV --preserve-paths --relative-only --no-output $GCNO_FILES 2>/dev/null |
   # Parse the raw gcov report to more human readable form.
-  python $ROOT/coverage/parse_gcov_output.py |
+  $PYTHON $ROOT/coverage/parse_gcov_output.py |
   # Write the output to both stdout and report file.
   tee $COVERAGE_DIR/coverage_report_all.txt &&
 echo -e "Generated coverage report for all files: $COVERAGE_DIR/coverage_report_all.txt\n"
@@ -41,7 +44,7 @@ RECENT_REPORT=$COVERAGE_DIR/coverage_report_recent.txt
 
 echo -e "Recently updated files: $LATEST_FILES\n" > $RECENT_REPORT
 $GCOV --preserve-paths --relative-only --no-output $GCNO_FILES 2>/dev/null |
-  python $ROOT/coverage/parse_gcov_output.py -interested-files $LATEST_FILES |
+  $PYTHON $ROOT/coverage/parse_gcov_output.py -interested-files $LATEST_FILES |
   tee -a $RECENT_REPORT &&
 echo -e "Generated coverage report for recently updated files: $RECENT_REPORT\n"
 

coverage/parse_gcov_output.py

@@ -1,10 +1,12 @@
-#!/usr/bin/env python2
+#!/usr/bin/env python
 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
+
+from __future__ import print_function
+
+import optparse
 import re
 import sys
 
-from optparse import OptionParser
-
 # the gcov report follows certain pattern. Each file will have two lines
 # of report, from which we can extract the file name, total lines and coverage
 # percentage.
@@ -48,7 +50,7 @@ def parse_gcov_report(gcov_input):
 def get_option_parser():
     usage = "Parse the gcov output and generate more human-readable code " +\
             "coverage report."
-    parser = OptionParser(usage)
+    parser = optparse.OptionParser(usage)
 
     parser.add_option(
         "--interested-files", "-i",
@@ -73,8 +75,8 @@ def display_file_coverage(per_file_coverage, total_coverage):
     header_template = \
         "%" + str(max_file_name_length) + "s\t%s\t%s"
     separator = "-" * (max_file_name_length + 10 + 20)
-    print header_template % ("Filename", "Coverage", "Lines")  # noqa: E999 T25377293 Grandfathered in
-    print separator
+    print(header_template % ("Filename", "Coverage", "Lines"))  # noqa: E999 T25377293 Grandfathered in
+    print(separator)
 
     # -- Print body
     # template for printing coverage report for each file.
@@ -82,12 +84,12 @@ def display_file_coverage(per_file_coverage, total_coverage):
 
     for fname, coverage_info in per_file_coverage.items():
         coverage, lines = coverage_info
-        print record_template % (fname, coverage, lines)
+        print(record_template % (fname, coverage, lines))
 
     # -- Print footer
     if total_coverage:
-        print separator
-        print record_template % ("Total", total_coverage[0], total_coverage[1])
+        print(separator)
+        print(record_template % ("Total", total_coverage[0], total_coverage[1]))
 
 def report_coverage():
     parser = get_option_parser()
@@ -111,7 +113,7 @@ def report_coverage():
         total_coverage = None
 
     if not len(per_file_coverage):
-        print >> sys.stderr, "Cannot find coverage info for the given files."
+        print("Cannot find coverage info for the given files.", file=sys.stderr)
         return
     display_file_coverage(per_file_coverage, total_coverage)
 

db/arena_wrapped_db_iter.cc

@@ -30,21 +30,21 @@ 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;
 }
 
@@ -56,8 +56,9 @@ Status ArenaWrappedDBIter::Refresh() {
   // TODO(yiwu): For last_seq_same_as_publish_seq_==false, this is not the
   // correct behavior. Will be corrected automatically when we take a snapshot
   // here for the case of WritePreparedTxnDB.
-  SequenceNumber latest_seq = db_impl_->GetLatestSequenceNumber();
   uint64_t cur_sv_number = cfd_->GetSuperVersionNumber();
+  TEST_SYNC_POINT("ArenaWrappedDBIter::Refresh:1");
+  TEST_SYNC_POINT("ArenaWrappedDBIter::Refresh:2");
   if (sv_number_ != cur_sv_number) {
     Env* env = db_iter_->env();
     db_iter_->~DBIter();
@@ -65,20 +66,22 @@ Status ArenaWrappedDBIter::Refresh() {
     new (&arena_) Arena();
 
     SuperVersion* sv = cfd_->GetReferencedSuperVersion(db_impl_);
+    SequenceNumber latest_seq = db_impl_->GetLatestSequenceNumber();
     if (read_callback_) {
       read_callback_->Refresh(latest_seq);
     }
     Init(env, read_options_, *(cfd_->ioptions()), sv->mutable_cf_options,
-         latest_seq, sv->mutable_cf_options.max_sequential_skip_in_iterations,
-         cur_sv_number, read_callback_, db_impl_, cfd_, allow_blob_,
+         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(
         read_options_, cfd_, sv, &arena_, db_iter_->GetRangeDelAggregator(),
-        latest_seq);
+        latest_seq, /* allow_unprepared_value */ true);
     SetIterUnderDBIter(internal_iter);
   } else {
-    db_iter_->set_sequence(latest_seq);
+    db_iter_->set_sequence(db_impl_->GetLatestSequenceNumber());
     db_iter_->set_valid(false);
   }
   return Status::OK();
@@ -87,17 +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(read_options, 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
@@ -41,6 +42,7 @@ class ArenaWrappedDBIter : public Iterator {
   virtual ReadRangeDelAggregator* GetRangeDelAggregator() {
     return db_iter_->GetRangeDelAggregator();
   }
+  const ReadOptions& GetReadOptions() { return read_options_; }
 
   // Set the internal iterator wrapped inside the DB Iterator. Usually it is
   // a merging iterator.
@@ -51,6 +53,8 @@ class ArenaWrappedDBIter : public Iterator {
   bool Valid() const override { return db_iter_->Valid(); }
   void SeekToFirst() override { db_iter_->SeekToFirst(); }
   void SeekToLast() override { db_iter_->SeekToLast(); }
+  // 'target' does not contain timestamp, even if user timestamp feature is
+  // enabled.
   void Seek(const Slice& target) override { db_iter_->Seek(target); }
   void SeekForPrev(const Slice& target) override {
     db_iter_->SeekForPrev(target);
@@ -60,6 +64,7 @@ class ArenaWrappedDBIter : public Iterator {
   Slice key() const override { return db_iter_->key(); }
   Slice value() const override { return db_iter_->value(); }
   Status status() const override { return db_iter_->status(); }
+  Slice timestamp() const override { return db_iter_->timestamp(); }
   bool IsBlob() const { return db_iter_->IsBlob(); }
 
   Status GetProperty(std::string prop_name, std::string* prop) override;
@@ -68,22 +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(const ReadOptions& read_options, DBImpl* db_impl,
-                        ColumnFamilyData* cfd, ReadCallback* read_callback,
-                        bool allow_blob) {
-    read_options_ = read_options;
+  void StoreRefreshInfo(DBImpl* db_impl, ColumnFamilyData* cfd,
+                        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:
@@ -94,7 +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;
 };
 
@@ -104,9 +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_constants.h

@@ -0,0 +1,16 @@
+//  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 <cstdint>
+
+#include "rocksdb/rocksdb_namespace.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+constexpr uint64_t kInvalidBlobFileNumber = 0;
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_addition.cc

@@ -0,0 +1,154 @@
+//  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_addition.h"
+
+#include <ostream>
+#include <sstream>
+
+#include "logging/event_logger.h"
+#include "rocksdb/slice.h"
+#include "rocksdb/status.h"
+#include "test_util/sync_point.h"
+#include "util/coding.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+// Tags for custom fields. Note that these get persisted in the manifest,
+// so existing tags should not be modified.
+enum BlobFileAddition::CustomFieldTags : uint32_t {
+  kEndMarker,
+
+  // Add forward compatible fields here
+
+  /////////////////////////////////////////////////////////////////////
+
+  kForwardIncompatibleMask = 1 << 6,
+
+  // Add forward incompatible fields here
+};
+
+void BlobFileAddition::EncodeTo(std::string* output) const {
+  PutVarint64(output, blob_file_number_);
+  PutVarint64(output, total_blob_count_);
+  PutVarint64(output, total_blob_bytes_);
+  PutLengthPrefixedSlice(output, checksum_method_);
+  PutLengthPrefixedSlice(output, checksum_value_);
+
+  // Encode any custom fields here. The format to use is a Varint32 tag (see
+  // CustomFieldTags above) followed by a length prefixed slice. Unknown custom
+  // fields will be ignored during decoding unless they're in the forward
+  // incompatible range.
+
+  TEST_SYNC_POINT_CALLBACK("BlobFileAddition::EncodeTo::CustomFields", output);
+
+  PutVarint32(output, kEndMarker);
+}
+
+Status BlobFileAddition::DecodeFrom(Slice* input) {
+  constexpr char class_name[] = "BlobFileAddition";
+
+  if (!GetVarint64(input, &blob_file_number_)) {
+    return Status::Corruption(class_name, "Error decoding blob file number");
+  }
+
+  if (!GetVarint64(input, &total_blob_count_)) {
+    return Status::Corruption(class_name, "Error decoding total blob count");
+  }
+
+  if (!GetVarint64(input, &total_blob_bytes_)) {
+    return Status::Corruption(class_name, "Error decoding total blob bytes");
+  }
+
+  Slice checksum_method;
+  if (!GetLengthPrefixedSlice(input, &checksum_method)) {
+    return Status::Corruption(class_name, "Error decoding checksum method");
+  }
+  checksum_method_ = checksum_method.ToString();
+
+  Slice checksum_value;
+  if (!GetLengthPrefixedSlice(input, &checksum_value)) {
+    return Status::Corruption(class_name, "Error decoding checksum value");
+  }
+  checksum_value_ = checksum_value.ToString();
+
+  while (true) {
+    uint32_t custom_field_tag = 0;
+    if (!GetVarint32(input, &custom_field_tag)) {
+      return Status::Corruption(class_name, "Error decoding custom field tag");
+    }
+
+    if (custom_field_tag == kEndMarker) {
+      break;
+    }
+
+    if (custom_field_tag & kForwardIncompatibleMask) {
+      return Status::Corruption(
+          class_name, "Forward incompatible custom field encountered");
+    }
+
+    Slice custom_field_value;
+    if (!GetLengthPrefixedSlice(input, &custom_field_value)) {
+      return Status::Corruption(class_name,
+                                "Error decoding custom field value");
+    }
+  }
+
+  return Status::OK();
+}
+
+std::string BlobFileAddition::DebugString() const {
+  std::ostringstream oss;
+
+  oss << *this;
+
+  return oss.str();
+}
+
+std::string BlobFileAddition::DebugJSON() const {
+  JSONWriter jw;
+
+  jw << *this;
+
+  jw.EndObject();
+
+  return jw.Get();
+}
+
+bool operator==(const BlobFileAddition& lhs, const BlobFileAddition& rhs) {
+  return lhs.GetBlobFileNumber() == rhs.GetBlobFileNumber() &&
+         lhs.GetTotalBlobCount() == rhs.GetTotalBlobCount() &&
+         lhs.GetTotalBlobBytes() == rhs.GetTotalBlobBytes() &&
+         lhs.GetChecksumMethod() == rhs.GetChecksumMethod() &&
+         lhs.GetChecksumValue() == rhs.GetChecksumValue();
+}
+
+bool operator!=(const BlobFileAddition& lhs, const BlobFileAddition& rhs) {
+  return !(lhs == rhs);
+}
+
+std::ostream& operator<<(std::ostream& os,
+                         const BlobFileAddition& blob_file_addition) {
+  os << "blob_file_number: " << blob_file_addition.GetBlobFileNumber()
+     << " total_blob_count: " << blob_file_addition.GetTotalBlobCount()
+     << " total_blob_bytes: " << blob_file_addition.GetTotalBlobBytes()
+     << " checksum_method: " << blob_file_addition.GetChecksumMethod()
+     << " checksum_value: " << blob_file_addition.GetChecksumValue();
+
+  return os;
+}
+
+JSONWriter& operator<<(JSONWriter& jw,
+                       const BlobFileAddition& blob_file_addition) {
+  jw << "BlobFileNumber" << blob_file_addition.GetBlobFileNumber()
+     << "TotalBlobCount" << blob_file_addition.GetTotalBlobCount()
+     << "TotalBlobBytes" << blob_file_addition.GetTotalBlobBytes()
+     << "ChecksumMethod" << blob_file_addition.GetChecksumMethod()
+     << "ChecksumValue" << blob_file_addition.GetChecksumValue();
+
+  return jw;
+}
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_addition.h

@@ -0,0 +1,67 @@
+//  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 <cstdint>
+#include <iosfwd>
+#include <string>
+
+#include "db/blob/blob_constants.h"
+#include "rocksdb/rocksdb_namespace.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class JSONWriter;
+class Slice;
+class Status;
+
+class BlobFileAddition {
+ public:
+  BlobFileAddition() = default;
+
+  BlobFileAddition(uint64_t blob_file_number, uint64_t total_blob_count,
+                   uint64_t total_blob_bytes, std::string checksum_method,
+                   std::string checksum_value)
+      : blob_file_number_(blob_file_number),
+        total_blob_count_(total_blob_count),
+        total_blob_bytes_(total_blob_bytes),
+        checksum_method_(std::move(checksum_method)),
+        checksum_value_(std::move(checksum_value)) {
+    assert(checksum_method_.empty() == checksum_value_.empty());
+  }
+
+  uint64_t GetBlobFileNumber() const { return blob_file_number_; }
+  uint64_t GetTotalBlobCount() const { return total_blob_count_; }
+  uint64_t GetTotalBlobBytes() const { return total_blob_bytes_; }
+  const std::string& GetChecksumMethod() const { return checksum_method_; }
+  const std::string& GetChecksumValue() const { return checksum_value_; }
+
+  void EncodeTo(std::string* output) const;
+  Status DecodeFrom(Slice* input);
+
+  std::string DebugString() const;
+  std::string DebugJSON() const;
+
+ private:
+  enum CustomFieldTags : uint32_t;
+
+  uint64_t blob_file_number_ = kInvalidBlobFileNumber;
+  uint64_t total_blob_count_ = 0;
+  uint64_t total_blob_bytes_ = 0;
+  std::string checksum_method_;
+  std::string checksum_value_;
+};
+
+bool operator==(const BlobFileAddition& lhs, const BlobFileAddition& rhs);
+bool operator!=(const BlobFileAddition& lhs, const BlobFileAddition& rhs);
+
+std::ostream& operator<<(std::ostream& os,
+                         const BlobFileAddition& blob_file_addition);
+JSONWriter& operator<<(JSONWriter& jw,
+                       const BlobFileAddition& blob_file_addition);
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_addition_test.cc

@@ -0,0 +1,206 @@
+//  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_addition.h"
+
+#include <cstdint>
+#include <cstring>
+#include <string>
+
+#include "test_util/sync_point.h"
+#include "test_util/testharness.h"
+#include "util/coding.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class BlobFileAdditionTest : public testing::Test {
+ public:
+  static void TestEncodeDecode(const BlobFileAddition& blob_file_addition) {
+    std::string encoded;
+    blob_file_addition.EncodeTo(&encoded);
+
+    BlobFileAddition decoded;
+    Slice input(encoded);
+    ASSERT_OK(decoded.DecodeFrom(&input));
+
+    ASSERT_EQ(blob_file_addition, decoded);
+  }
+};
+
+TEST_F(BlobFileAdditionTest, Empty) {
+  BlobFileAddition blob_file_addition;
+
+  ASSERT_EQ(blob_file_addition.GetBlobFileNumber(), kInvalidBlobFileNumber);
+  ASSERT_EQ(blob_file_addition.GetTotalBlobCount(), 0);
+  ASSERT_EQ(blob_file_addition.GetTotalBlobBytes(), 0);
+  ASSERT_TRUE(blob_file_addition.GetChecksumMethod().empty());
+  ASSERT_TRUE(blob_file_addition.GetChecksumValue().empty());
+
+  TestEncodeDecode(blob_file_addition);
+}
+
+TEST_F(BlobFileAdditionTest, NonEmpty) {
+  constexpr uint64_t blob_file_number = 123;
+  constexpr uint64_t total_blob_count = 2;
+  constexpr uint64_t total_blob_bytes = 123456;
+  const std::string checksum_method("SHA1");
+  const std::string checksum_value("bdb7f34a59dfa1592ce7f52e99f98c570c525cbd");
+
+  BlobFileAddition blob_file_addition(blob_file_number, total_blob_count,
+                                      total_blob_bytes, checksum_method,
+                                      checksum_value);
+
+  ASSERT_EQ(blob_file_addition.GetBlobFileNumber(), blob_file_number);
+  ASSERT_EQ(blob_file_addition.GetTotalBlobCount(), total_blob_count);
+  ASSERT_EQ(blob_file_addition.GetTotalBlobBytes(), total_blob_bytes);
+  ASSERT_EQ(blob_file_addition.GetChecksumMethod(), checksum_method);
+  ASSERT_EQ(blob_file_addition.GetChecksumValue(), checksum_value);
+
+  TestEncodeDecode(blob_file_addition);
+}
+
+TEST_F(BlobFileAdditionTest, DecodeErrors) {
+  std::string str;
+  Slice slice(str);
+
+  BlobFileAddition blob_file_addition;
+
+  {
+    const Status s = blob_file_addition.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "blob file number"));
+  }
+
+  constexpr uint64_t blob_file_number = 123;
+  PutVarint64(&str, blob_file_number);
+  slice = str;
+
+  {
+    const Status s = blob_file_addition.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "total blob count"));
+  }
+
+  constexpr uint64_t total_blob_count = 4567;
+  PutVarint64(&str, total_blob_count);
+  slice = str;
+
+  {
+    const Status s = blob_file_addition.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "total blob bytes"));
+  }
+
+  constexpr uint64_t total_blob_bytes = 12345678;
+  PutVarint64(&str, total_blob_bytes);
+  slice = str;
+
+  {
+    const Status s = blob_file_addition.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "checksum method"));
+  }
+
+  constexpr char checksum_method[] = "SHA1";
+  PutLengthPrefixedSlice(&str, checksum_method);
+  slice = str;
+
+  {
+    const Status s = blob_file_addition.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "checksum value"));
+  }
+
+  constexpr char checksum_value[] = "bdb7f34a59dfa1592ce7f52e99f98c570c525cbd";
+  PutLengthPrefixedSlice(&str, checksum_value);
+  slice = str;
+
+  {
+    const Status s = blob_file_addition.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "custom field tag"));
+  }
+
+  constexpr uint32_t custom_tag = 2;
+  PutVarint32(&str, custom_tag);
+  slice = str;
+
+  {
+    const Status s = blob_file_addition.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "custom field value"));
+  }
+}
+
+TEST_F(BlobFileAdditionTest, ForwardCompatibleCustomField) {
+  SyncPoint::GetInstance()->SetCallBack(
+      "BlobFileAddition::EncodeTo::CustomFields", [&](void* arg) {
+        std::string* output = static_cast<std::string*>(arg);
+
+        constexpr uint32_t forward_compatible_tag = 2;
+        PutVarint32(output, forward_compatible_tag);
+
+        PutLengthPrefixedSlice(output, "deadbeef");
+      });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  constexpr uint64_t blob_file_number = 678;
+  constexpr uint64_t total_blob_count = 9999;
+  constexpr uint64_t total_blob_bytes = 100000000;
+  const std::string checksum_method("CRC32");
+  const std::string checksum_value("3d87ff57");
+
+  BlobFileAddition blob_file_addition(blob_file_number, total_blob_count,
+                                      total_blob_bytes, checksum_method,
+                                      checksum_value);
+
+  TestEncodeDecode(blob_file_addition);
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+TEST_F(BlobFileAdditionTest, ForwardIncompatibleCustomField) {
+  SyncPoint::GetInstance()->SetCallBack(
+      "BlobFileAddition::EncodeTo::CustomFields", [&](void* arg) {
+        std::string* output = static_cast<std::string*>(arg);
+
+        constexpr uint32_t forward_incompatible_tag = (1 << 6) + 1;
+        PutVarint32(output, forward_incompatible_tag);
+
+        PutLengthPrefixedSlice(output, "foobar");
+      });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  constexpr uint64_t blob_file_number = 456;
+  constexpr uint64_t total_blob_count = 100;
+  constexpr uint64_t total_blob_bytes = 2000000;
+  const std::string checksum_method("CRC32B");
+  const std::string checksum_value("6dbdf23a");
+
+  BlobFileAddition blob_file_addition(blob_file_number, total_blob_count,
+                                      total_blob_bytes, checksum_method,
+                                      checksum_value);
+
+  std::string encoded;
+  blob_file_addition.EncodeTo(&encoded);
+
+  BlobFileAddition decoded_blob_file_addition;
+  Slice input(encoded);
+  const Status s = decoded_blob_file_addition.DecodeFrom(&input);
+
+  ASSERT_TRUE(s.IsCorruption());
+  ASSERT_TRUE(std::strstr(s.getState(), "Forward incompatible"));
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+}  // namespace ROCKSDB_NAMESPACE
+
+int main(int argc, char** argv) {
+  ::testing::InitGoogleTest(&argc, argv);
+  return RUN_ALL_TESTS();
+}

db/blob/blob_file_builder.cc

@@ -0,0 +1,350 @@
+//  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_builder.h"
+
+#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"
+#include "db/version_set.h"
+#include "file/filename.h"
+#include "file/read_write_util.h"
+#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, 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(); }, 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, 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)),
+      fs_(fs),
+      immutable_cf_options_(immutable_cf_options),
+      min_blob_size_(mutable_cf_options->min_blob_size),
+      blob_file_size_(mutable_cf_options->blob_file_size),
+      blob_compression_type_(mutable_cf_options->blob_compression_type),
+      file_options_(file_options),
+      job_id_(job_id),
+      column_family_id_(column_family_id),
+      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(fs_);
+  assert(immutable_cf_options_);
+  assert(file_options_);
+  assert(blob_file_paths_);
+  assert(blob_file_paths_->empty());
+  assert(blob_file_additions_);
+  assert(blob_file_additions_->empty());
+}
+
+BlobFileBuilder::~BlobFileBuilder() = default;
+
+Status BlobFileBuilder::Add(const Slice& key, const Slice& value,
+                            std::string* blob_index) {
+  assert(blob_index);
+  assert(blob_index->empty());
+
+  if (value.size() < min_blob_size_) {
+    return Status::OK();
+  }
+
+  {
+    const Status s = OpenBlobFileIfNeeded();
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  Slice blob = value;
+  std::string compressed_blob;
+
+  {
+    const Status s = CompressBlobIfNeeded(&blob, &compressed_blob);
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  uint64_t blob_file_number = 0;
+  uint64_t blob_offset = 0;
+
+  {
+    const Status s =
+        WriteBlobToFile(key, blob, &blob_file_number, &blob_offset);
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  {
+    const Status s = CloseBlobFileIfNeeded();
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  BlobIndex::EncodeBlob(blob_index, blob_file_number, blob_offset, blob.size(),
+                        blob_compression_type_);
+
+  return Status::OK();
+}
+
+Status BlobFileBuilder::Finish() {
+  if (!IsBlobFileOpen()) {
+    return Status::OK();
+  }
+
+  return CloseBlobFile();
+}
+
+bool BlobFileBuilder::IsBlobFileOpen() const { return !!writer_; }
+
+Status BlobFileBuilder::OpenBlobFileIfNeeded() {
+  if (IsBlobFileOpen()) {
+    return Status::OK();
+  }
+
+  assert(!blob_count_);
+  assert(!blob_bytes_);
+
+  assert(file_number_generator_);
+  const uint64_t blob_file_number = file_number_generator_();
+
+  assert(immutable_cf_options_);
+  assert(!immutable_cf_options_->cf_paths.empty());
+  std::string blob_file_path = BlobFileName(
+      immutable_cf_options_->cf_paths.front().path, blob_file_number);
+
+  std::unique_ptr<FSWritableFile> file;
+
+  {
+    assert(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;
+    }
+  }
+
+  // Note: files get added to blob_file_paths_ right after the open, so they
+  // can be cleaned up upon failure. Contrast this with blob_file_additions_,
+  // which only contains successfully written files.
+  assert(blob_file_paths_);
+  blob_file_paths_->emplace_back(std::move(blob_file_path));
+
+  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_,
+      immutable_cf_options_->clock, io_tracer_, statistics,
+      immutable_cf_options_->listeners,
+      immutable_cf_options_->file_checksum_gen_factory,
+      tmp_set.Contains(FileType::kBlobFile)));
+
+  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;
+
+  BlobLogHeader header(column_family_id_, blob_compression_type_, has_ttl,
+                       expiration_range);
+
+  {
+    Status s = blob_log_writer->WriteHeader(header);
+
+    TEST_SYNC_POINT_CALLBACK(
+        "BlobFileBuilder::OpenBlobFileIfNeeded:WriteHeader", &s);
+
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  writer_ = std::move(blob_log_writer);
+
+  assert(IsBlobFileOpen());
+
+  return Status::OK();
+}
+
+Status BlobFileBuilder::CompressBlobIfNeeded(
+    Slice* blob, std::string* compressed_blob) const {
+  assert(blob);
+  assert(compressed_blob);
+  assert(compressed_blob->empty());
+
+  if (blob_compression_type_ == kNoCompression) {
+    return Status::OK();
+  }
+
+  CompressionOptions opts;
+  CompressionContext context(blob_compression_type_);
+  constexpr uint64_t sample_for_compression = 0;
+
+  CompressionInfo info(opts, context, CompressionDict::GetEmptyDict(),
+                       blob_compression_type_, sample_for_compression);
+
+  constexpr uint32_t compression_format_version = 2;
+
+  if (!CompressData(*blob, info, compression_format_version, compressed_blob)) {
+    return Status::Corruption("Error compressing blob");
+  }
+
+  *blob = Slice(*compressed_blob);
+
+  return Status::OK();
+}
+
+Status BlobFileBuilder::WriteBlobToFile(const Slice& key, const Slice& blob,
+                                        uint64_t* blob_file_number,
+                                        uint64_t* blob_offset) {
+  assert(IsBlobFileOpen());
+  assert(blob_file_number);
+  assert(blob_offset);
+
+  uint64_t key_offset = 0;
+
+  Status s = writer_->AddRecord(key, blob, &key_offset, blob_offset);
+
+  TEST_SYNC_POINT_CALLBACK("BlobFileBuilder::WriteBlobToFile:AddRecord", &s);
+
+  if (!s.ok()) {
+    return s;
+  }
+
+  *blob_file_number = writer_->get_log_number();
+
+  ++blob_count_;
+  blob_bytes_ += BlobLogRecord::kHeaderSize + key.size() + blob.size();
+
+  return Status::OK();
+}
+
+Status BlobFileBuilder::CloseBlobFile() {
+  assert(IsBlobFileOpen());
+
+  BlobLogFooter footer;
+  footer.blob_count = blob_count_;
+
+  std::string checksum_method;
+  std::string checksum_value;
+
+  Status s = writer_->AppendFooter(footer, &checksum_method, &checksum_value);
+
+  TEST_SYNC_POINT_CALLBACK("BlobFileBuilder::WriteBlobToFile:AppendFooter", &s);
+
+  if (!s.ok()) {
+    return s;
+  }
+
+  const uint64_t blob_file_number = writer_->get_log_number();
+
+  assert(blob_file_additions_);
+  blob_file_additions_->emplace_back(blob_file_number, blob_count_, blob_bytes_,
+                                     std::move(checksum_method),
+                                     std::move(checksum_value));
+
+  assert(immutable_cf_options_);
+  ROCKS_LOG_INFO(immutable_cf_options_->info_log,
+                 "[%s] [JOB %d] Generated blob file #%" PRIu64 ": %" PRIu64
+                 " 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 s;
+}
+
+Status BlobFileBuilder::CloseBlobFileIfNeeded() {
+  assert(IsBlobFileOpen());
+
+  const WritableFileWriter* const file_writer = writer_->file();
+  assert(file_writer);
+
+  if (file_writer->GetFileSize() < blob_file_size_) {
+    return Status::OK();
+  }
+
+  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

@@ -0,0 +1,100 @@
+//  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 <functional>
+#include <memory>
+#include <string>
+#include <vector>
+
+#include "rocksdb/compression_type.h"
+#include "rocksdb/env.h"
+#include "rocksdb/rocksdb_namespace.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class VersionSet;
+class FileSystem;
+class SystemClock;
+struct ImmutableCFOptions;
+struct MutableCFOptions;
+struct FileOptions;
+class BlobFileAddition;
+class Status;
+class Slice;
+class BlobLogWriter;
+class IOTracer;
+class BlobFileCompletionCallback;
+
+class BlobFileBuilder {
+ public:
+  BlobFileBuilder(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(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);
+
+  BlobFileBuilder(const BlobFileBuilder&) = delete;
+  BlobFileBuilder& operator=(const BlobFileBuilder&) = delete;
+
+  ~BlobFileBuilder();
+
+  Status Add(const Slice& key, const Slice& value, std::string* blob_index);
+  Status Finish();
+  void Abandon();
+
+ private:
+  bool IsBlobFileOpen() const;
+  Status OpenBlobFileIfNeeded();
+  Status CompressBlobIfNeeded(Slice* blob, std::string* compressed_blob) const;
+  Status WriteBlobToFile(const Slice& key, const Slice& blob,
+                         uint64_t* blob_file_number, uint64_t* blob_offset);
+  Status CloseBlobFile();
+  Status CloseBlobFileIfNeeded();
+
+  std::function<uint64_t()> file_number_generator_;
+  FileSystem* fs_;
+  const ImmutableCFOptions* immutable_cf_options_;
+  uint64_t min_blob_size_;
+  uint64_t blob_file_size_;
+  CompressionType blob_compression_type_;
+  const FileOptions* file_options_;
+  int job_id_;
+  uint32_t column_family_id_;
+  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_;
+  uint64_t blob_count_;
+  uint64_t blob_bytes_;
+};
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_builder_test.cc

@@ -0,0 +1,673 @@
+//  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_builder.h"
+
+#include <cassert>
+#include <cinttypes>
+#include <string>
+#include <utility>
+#include <vector>
+
+#include "db/blob/blob_file_addition.h"
+#include "db/blob/blob_index.h"
+#include "db/blob/blob_log_format.h"
+#include "db/blob/blob_log_sequential_reader.h"
+#include "env/mock_env.h"
+#include "file/filename.h"
+#include "file/random_access_file_reader.h"
+#include "options/cf_options.h"
+#include "rocksdb/env.h"
+#include "rocksdb/file_checksum.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"
+
+namespace ROCKSDB_NAMESPACE {
+
+class TestFileNumberGenerator {
+ public:
+  uint64_t operator()() { return ++next_file_number_; }
+
+ private:
+  uint64_t next_file_number_ = 1;
+};
+
+class BlobFileBuilderTest : public testing::Test {
+ protected:
+  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,
+                      uint32_t column_family_id,
+                      CompressionType blob_compression_type,
+                      const std::vector<std::pair<std::string, std::string>>&
+                          expected_key_value_pairs,
+                      const std::vector<std::string>& blob_indexes) {
+    assert(expected_key_value_pairs.size() == blob_indexes.size());
+
+    std::unique_ptr<FSRandomAccessFile> file;
+    constexpr IODebugContext* dbg = nullptr;
+    ASSERT_OK(
+        fs_->NewRandomAccessFile(blob_file_path, file_options_, &file, dbg));
+
+    std::unique_ptr<RandomAccessFileReader> file_reader(
+        new RandomAccessFileReader(std::move(file), blob_file_path, clock_));
+
+    constexpr Statistics* statistics = nullptr;
+    BlobLogSequentialReader blob_log_reader(std::move(file_reader), clock_,
+                                            statistics);
+
+    BlobLogHeader header;
+    ASSERT_OK(blob_log_reader.ReadHeader(&header));
+    ASSERT_EQ(header.version, kVersion1);
+    ASSERT_EQ(header.column_family_id, column_family_id);
+    ASSERT_EQ(header.compression, blob_compression_type);
+    ASSERT_FALSE(header.has_ttl);
+    ASSERT_EQ(header.expiration_range, ExpirationRange());
+
+    for (size_t i = 0; i < expected_key_value_pairs.size(); ++i) {
+      BlobLogRecord record;
+      uint64_t blob_offset = 0;
+
+      ASSERT_OK(blob_log_reader.ReadRecord(
+          &record, BlobLogSequentialReader::kReadHeaderKeyBlob, &blob_offset));
+
+      // Check the contents of the blob file
+      const auto& expected_key_value = expected_key_value_pairs[i];
+      const auto& key = expected_key_value.first;
+      const auto& value = expected_key_value.second;
+
+      ASSERT_EQ(record.key_size, key.size());
+      ASSERT_EQ(record.value_size, value.size());
+      ASSERT_EQ(record.expiration, 0);
+      ASSERT_EQ(record.key, key);
+      ASSERT_EQ(record.value, value);
+
+      // Make sure the blob reference returned by the builder points to the
+      // right place
+      BlobIndex blob_index;
+      ASSERT_OK(blob_index.DecodeFrom(blob_indexes[i]));
+      ASSERT_FALSE(blob_index.IsInlined());
+      ASSERT_FALSE(blob_index.HasTTL());
+      ASSERT_EQ(blob_index.file_number(), blob_file_number);
+      ASSERT_EQ(blob_index.offset(), blob_offset);
+      ASSERT_EQ(blob_index.size(), value.size());
+    }
+
+    BlobLogFooter footer;
+    ASSERT_OK(blob_log_reader.ReadFooter(&footer));
+    ASSERT_EQ(footer.blob_count, expected_key_value_pairs.size());
+    ASSERT_EQ(footer.expiration_range, ExpirationRange());
+  }
+
+  MockEnv mock_env_;
+  FileSystem* fs_;
+  SystemClock* clock_;
+  FileOptions file_options_;
+};
+
+TEST_F(BlobFileBuilderTest, BuildAndCheckOneFile) {
+  // Build a single blob file
+  constexpr size_t number_of_blobs = 10;
+  constexpr size_t key_size = 1;
+  constexpr size_t value_size = 4;
+  constexpr size_t value_offset = 1234;
+
+  Options options;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_,
+                            "BlobFileBuilderTest_BuildAndCheckOneFile"),
+      0);
+  options.enable_blob_files = true;
+  options.env = &mock_env_;
+
+  ImmutableCFOptions immutable_cf_options(options);
+  MutableCFOptions mutable_cf_options(options);
+
+  constexpr int job_id = 1;
+  constexpr uint32_t column_family_id = 123;
+  constexpr char column_family_name[] = "foobar";
+  constexpr Env::IOPriority io_priority = Env::IO_HIGH;
+  constexpr Env::WriteLifeTimeHint write_hint = Env::WLTH_MEDIUM;
+
+  std::vector<std::string> blob_file_paths;
+  std::vector<BlobFileAddition> blob_file_additions;
+
+  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(
+      number_of_blobs);
+  std::vector<std::string> blob_indexes(number_of_blobs);
+
+  for (size_t i = 0; i < number_of_blobs; ++i) {
+    auto& expected_key_value = expected_key_value_pairs[i];
+
+    auto& key = expected_key_value.first;
+    key = std::to_string(i);
+    assert(key.size() == key_size);
+
+    auto& value = expected_key_value.second;
+    value = std::to_string(i + value_offset);
+    assert(value.size() == value_size);
+
+    auto& blob_index = blob_indexes[i];
+
+    ASSERT_OK(builder.Add(key, value, &blob_index));
+    ASSERT_FALSE(blob_index.empty());
+  }
+
+  ASSERT_OK(builder.Finish());
+
+  // Check the metadata generated
+  constexpr uint64_t blob_file_number = 2;
+
+  ASSERT_EQ(blob_file_paths.size(), 1);
+
+  const std::string& blob_file_path = blob_file_paths[0];
+
+  ASSERT_EQ(blob_file_path,
+            BlobFileName(immutable_cf_options.cf_paths.front().path,
+                         blob_file_number));
+
+  ASSERT_EQ(blob_file_additions.size(), 1);
+
+  const auto& blob_file_addition = blob_file_additions[0];
+
+  ASSERT_EQ(blob_file_addition.GetBlobFileNumber(), blob_file_number);
+  ASSERT_EQ(blob_file_addition.GetTotalBlobCount(), number_of_blobs);
+  ASSERT_EQ(
+      blob_file_addition.GetTotalBlobBytes(),
+      number_of_blobs * (BlobLogRecord::kHeaderSize + key_size + value_size));
+
+  // Verify the contents of the new blob file as well as the blob references
+  VerifyBlobFile(blob_file_number, blob_file_path, column_family_id,
+                 kNoCompression, expected_key_value_pairs, blob_indexes);
+}
+
+TEST_F(BlobFileBuilderTest, BuildAndCheckMultipleFiles) {
+  // Build multiple blob files: file size limit is set to the size of a single
+  // value, so each blob ends up in a file of its own
+  constexpr size_t number_of_blobs = 10;
+  constexpr size_t key_size = 1;
+  constexpr size_t value_size = 10;
+  constexpr size_t value_offset = 1234567890;
+
+  Options options;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_,
+                            "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);
+
+  constexpr int job_id = 1;
+  constexpr uint32_t column_family_id = 123;
+  constexpr char column_family_name[] = "foobar";
+  constexpr Env::IOPriority io_priority = Env::IO_HIGH;
+  constexpr Env::WriteLifeTimeHint write_hint = Env::WLTH_MEDIUM;
+
+  std::vector<std::string> blob_file_paths;
+  std::vector<BlobFileAddition> blob_file_additions;
+
+  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(
+      number_of_blobs);
+  std::vector<std::string> blob_indexes(number_of_blobs);
+
+  for (size_t i = 0; i < number_of_blobs; ++i) {
+    auto& expected_key_value = expected_key_value_pairs[i];
+
+    auto& key = expected_key_value.first;
+    key = std::to_string(i);
+    assert(key.size() == key_size);
+
+    auto& value = expected_key_value.second;
+    value = std::to_string(i + value_offset);
+    assert(value.size() == value_size);
+
+    auto& blob_index = blob_indexes[i];
+
+    ASSERT_OK(builder.Add(key, value, &blob_index));
+    ASSERT_FALSE(blob_index.empty());
+  }
+
+  ASSERT_OK(builder.Finish());
+
+  // Check the metadata generated
+  ASSERT_EQ(blob_file_paths.size(), number_of_blobs);
+  ASSERT_EQ(blob_file_additions.size(), number_of_blobs);
+
+  for (size_t i = 0; i < number_of_blobs; ++i) {
+    const uint64_t blob_file_number = i + 2;
+
+    ASSERT_EQ(blob_file_paths[i],
+              BlobFileName(immutable_cf_options.cf_paths.front().path,
+                           blob_file_number));
+
+    const auto& blob_file_addition = blob_file_additions[i];
+
+    ASSERT_EQ(blob_file_addition.GetBlobFileNumber(), blob_file_number);
+    ASSERT_EQ(blob_file_addition.GetTotalBlobCount(), 1);
+    ASSERT_EQ(blob_file_addition.GetTotalBlobBytes(),
+              BlobLogRecord::kHeaderSize + key_size + value_size);
+  }
+
+  // Verify the contents of the new blob files as well as the blob references
+  for (size_t i = 0; i < number_of_blobs; ++i) {
+    std::vector<std::pair<std::string, std::string>> expected_key_value_pair{
+        expected_key_value_pairs[i]};
+    std::vector<std::string> blob_index{blob_indexes[i]};
+
+    VerifyBlobFile(i + 2, blob_file_paths[i], column_family_id, kNoCompression,
+                   expected_key_value_pair, blob_index);
+  }
+}
+
+TEST_F(BlobFileBuilderTest, InlinedValues) {
+  // All values are below the min_blob_size threshold; no blob files get written
+  constexpr size_t number_of_blobs = 10;
+  constexpr size_t key_size = 1;
+  constexpr size_t value_size = 10;
+  constexpr size_t value_offset = 1234567890;
+
+  Options options;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_, "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);
+
+  constexpr int job_id = 1;
+  constexpr uint32_t column_family_id = 123;
+  constexpr char column_family_name[] = "foobar";
+  constexpr Env::IOPriority io_priority = Env::IO_HIGH;
+  constexpr Env::WriteLifeTimeHint write_hint = Env::WLTH_MEDIUM;
+
+  std::vector<std::string> blob_file_paths;
+  std::vector<BlobFileAddition> blob_file_additions;
+
+  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) {
+    const std::string key = std::to_string(i);
+    assert(key.size() == key_size);
+
+    const std::string value = std::to_string(i + value_offset);
+    assert(value.size() == value_size);
+
+    std::string blob_index;
+    ASSERT_OK(builder.Add(key, value, &blob_index));
+    ASSERT_TRUE(blob_index.empty());
+  }
+
+  ASSERT_OK(builder.Finish());
+
+  // Check the metadata generated
+  ASSERT_TRUE(blob_file_paths.empty());
+  ASSERT_TRUE(blob_file_additions.empty());
+}
+
+TEST_F(BlobFileBuilderTest, Compression) {
+  // Build a blob file with a compressed blob
+  if (!Snappy_Supported()) {
+    return;
+  }
+
+  constexpr size_t key_size = 1;
+  constexpr size_t value_size = 100;
+
+  Options options;
+  options.cf_paths.emplace_back(
+      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);
+
+  constexpr int job_id = 1;
+  constexpr uint32_t column_family_id = 123;
+  constexpr char column_family_name[] = "foobar";
+  constexpr Env::IOPriority io_priority = Env::IO_HIGH;
+  constexpr Env::WriteLifeTimeHint write_hint = Env::WLTH_MEDIUM;
+
+  std::vector<std::string> blob_file_paths;
+  std::vector<BlobFileAddition> blob_file_additions;
+
+  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");
+  const std::string uncompressed_value(value_size, 'x');
+
+  std::string blob_index;
+
+  ASSERT_OK(builder.Add(key, uncompressed_value, &blob_index));
+  ASSERT_FALSE(blob_index.empty());
+
+  ASSERT_OK(builder.Finish());
+
+  // Check the metadata generated
+  constexpr uint64_t blob_file_number = 2;
+
+  ASSERT_EQ(blob_file_paths.size(), 1);
+
+  const std::string& blob_file_path = blob_file_paths[0];
+
+  ASSERT_EQ(blob_file_path,
+            BlobFileName(immutable_cf_options.cf_paths.front().path,
+                         blob_file_number));
+
+  ASSERT_EQ(blob_file_additions.size(), 1);
+
+  const auto& blob_file_addition = blob_file_additions[0];
+
+  ASSERT_EQ(blob_file_addition.GetBlobFileNumber(), blob_file_number);
+  ASSERT_EQ(blob_file_addition.GetTotalBlobCount(), 1);
+
+  CompressionOptions opts;
+  CompressionContext context(kSnappyCompression);
+  constexpr uint64_t sample_for_compression = 0;
+
+  CompressionInfo info(opts, context, CompressionDict::GetEmptyDict(),
+                       kSnappyCompression, sample_for_compression);
+
+  std::string compressed_value;
+  ASSERT_TRUE(Snappy_Compress(info, uncompressed_value.data(),
+                              uncompressed_value.size(), &compressed_value));
+
+  ASSERT_EQ(blob_file_addition.GetTotalBlobBytes(),
+            BlobLogRecord::kHeaderSize + key_size + compressed_value.size());
+
+  // Verify the contents of the new blob file as well as the blob reference
+  std::vector<std::pair<std::string, std::string>> expected_key_value_pairs{
+      {key, compressed_value}};
+  std::vector<std::string> blob_indexes{blob_index};
+
+  VerifyBlobFile(blob_file_number, blob_file_path, column_family_id,
+                 kSnappyCompression, expected_key_value_pairs, blob_indexes);
+}
+
+TEST_F(BlobFileBuilderTest, CompressionError) {
+  // Simulate an error during compression
+  if (!Snappy_Supported()) {
+    return;
+  }
+
+  Options options;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_, "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);
+
+  constexpr int job_id = 1;
+  constexpr uint32_t column_family_id = 123;
+  constexpr char column_family_name[] = "foobar";
+  constexpr Env::IOPriority io_priority = Env::IO_HIGH;
+  constexpr Env::WriteLifeTimeHint write_hint = Env::WLTH_MEDIUM;
+
+  std::vector<std::string> blob_file_paths;
+  std::vector<BlobFileAddition> blob_file_additions;
+
+  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",
+                                        [](void* arg) {
+                                          bool* ret = static_cast<bool*>(arg);
+                                          *ret = false;
+                                        });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  constexpr char key[] = "1";
+  constexpr char value[] = "deadbeef";
+
+  std::string blob_index;
+
+  ASSERT_TRUE(builder.Add(key, value, &blob_index).IsCorruption());
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+
+  constexpr uint64_t blob_file_number = 2;
+
+  ASSERT_EQ(blob_file_paths.size(), 1);
+  ASSERT_EQ(blob_file_paths[0],
+            BlobFileName(immutable_cf_options.cf_paths.front().path,
+                         blob_file_number));
+
+  ASSERT_TRUE(blob_file_additions.empty());
+}
+
+TEST_F(BlobFileBuilderTest, Checksum) {
+  // Build a blob file with checksum
+
+  class DummyFileChecksumGenerator : public FileChecksumGenerator {
+   public:
+    void Update(const char* /* data */, size_t /* n */) override {}
+
+    void Finalize() override {}
+
+    std::string GetChecksum() const override { return std::string("dummy"); }
+
+    const char* Name() const override { return "DummyFileChecksum"; }
+  };
+
+  class DummyFileChecksumGenFactory : public FileChecksumGenFactory {
+   public:
+    std::unique_ptr<FileChecksumGenerator> CreateFileChecksumGenerator(
+        const FileChecksumGenContext& /* context */) override {
+      return std::unique_ptr<FileChecksumGenerator>(
+          new DummyFileChecksumGenerator);
+    }
+
+    const char* Name() const override { return "DummyFileChecksumGenFactory"; }
+  };
+
+  Options options;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_, "BlobFileBuilderTest_Checksum"), 0);
+  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);
+
+  constexpr int job_id = 1;
+  constexpr uint32_t column_family_id = 123;
+  constexpr char column_family_name[] = "foobar";
+  constexpr Env::IOPriority io_priority = Env::IO_HIGH;
+  constexpr Env::WriteLifeTimeHint write_hint = Env::WLTH_MEDIUM;
+
+  std::vector<std::string> blob_file_paths;
+  std::vector<BlobFileAddition> blob_file_additions;
+
+  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");
+  const std::string value("deadbeef");
+
+  std::string blob_index;
+
+  ASSERT_OK(builder.Add(key, value, &blob_index));
+  ASSERT_FALSE(blob_index.empty());
+
+  ASSERT_OK(builder.Finish());
+
+  // Check the metadata generated
+  constexpr uint64_t blob_file_number = 2;
+
+  ASSERT_EQ(blob_file_paths.size(), 1);
+
+  const std::string& blob_file_path = blob_file_paths[0];
+
+  ASSERT_EQ(blob_file_path,
+            BlobFileName(immutable_cf_options.cf_paths.front().path,
+                         blob_file_number));
+
+  ASSERT_EQ(blob_file_additions.size(), 1);
+
+  const auto& blob_file_addition = blob_file_additions[0];
+
+  ASSERT_EQ(blob_file_addition.GetBlobFileNumber(), blob_file_number);
+  ASSERT_EQ(blob_file_addition.GetTotalBlobCount(), 1);
+  ASSERT_EQ(blob_file_addition.GetTotalBlobBytes(),
+            BlobLogRecord::kHeaderSize + key.size() + value.size());
+  ASSERT_EQ(blob_file_addition.GetChecksumMethod(), "DummyFileChecksum");
+  ASSERT_EQ(blob_file_addition.GetChecksumValue(), "dummy");
+
+  // Verify the contents of the new blob file as well as the blob reference
+  std::vector<std::pair<std::string, std::string>> expected_key_value_pairs{
+      {key, value}};
+  std::vector<std::string> blob_indexes{blob_index};
+
+  VerifyBlobFile(blob_file_number, blob_file_path, column_family_id,
+                 kNoCompression, expected_key_value_pairs, blob_indexes);
+}
+
+class BlobFileBuilderIOErrorTest
+    : public testing::Test,
+      public testing::WithParamInterface<std::string> {
+ protected:
+  BlobFileBuilderIOErrorTest()
+      : mock_env_(Env::Default()),
+        fs_(mock_env_.GetFileSystem().get()),
+        sync_point_(GetParam()) {}
+
+  MockEnv mock_env_;
+  FileSystem* fs_;
+  FileOptions file_options_;
+  std::string sync_point_;
+};
+
+INSTANTIATE_TEST_CASE_P(
+    BlobFileBuilderTest, BlobFileBuilderIOErrorTest,
+    ::testing::ValuesIn(std::vector<std::string>{
+        "BlobFileBuilder::OpenBlobFileIfNeeded:NewWritableFile",
+        "BlobFileBuilder::OpenBlobFileIfNeeded:WriteHeader",
+        "BlobFileBuilder::WriteBlobToFile:AddRecord",
+        "BlobFileBuilder::WriteBlobToFile:AppendFooter"}));
+
+TEST_P(BlobFileBuilderIOErrorTest, IOError) {
+  // Simulate an I/O error during the specified step of Add()
+  // Note: blob_file_size will be set to value_size in order for the first blob
+  // to trigger close
+  constexpr size_t value_size = 8;
+
+  Options options;
+  options.cf_paths.emplace_back(
+      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);
+
+  constexpr int job_id = 1;
+  constexpr uint32_t column_family_id = 123;
+  constexpr char column_family_name[] = "foobar";
+  constexpr Env::IOPriority io_priority = Env::IO_HIGH;
+  constexpr Env::WriteLifeTimeHint write_hint = Env::WLTH_MEDIUM;
+
+  std::vector<std::string> blob_file_paths;
+  std::vector<BlobFileAddition> blob_file_additions;
+
+  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) {
+    Status* const s = static_cast<Status*>(arg);
+    assert(s);
+
+    (*s) = Status::IOError(sync_point_);
+  });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  constexpr char key[] = "1";
+  constexpr char value[] = "deadbeef";
+
+  std::string blob_index;
+
+  ASSERT_TRUE(builder.Add(key, value, &blob_index).IsIOError());
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+
+  if (sync_point_ == "BlobFileBuilder::OpenBlobFileIfNeeded:NewWritableFile") {
+    ASSERT_TRUE(blob_file_paths.empty());
+  } else {
+    constexpr uint64_t blob_file_number = 2;
+
+    ASSERT_EQ(blob_file_paths.size(), 1);
+    ASSERT_EQ(blob_file_paths[0],
+              BlobFileName(immutable_cf_options.cf_paths.front().path,
+                           blob_file_number));
+  }
+
+  ASSERT_TRUE(blob_file_additions.empty());
+}
+
+}  // namespace ROCKSDB_NAMESPACE
+
+int main(int argc, char** argv) {
+  ::testing::InitGoogleTest(&argc, argv);
+  return RUN_ALL_TESTS();
+}

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 +1,53 @@
+//  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:
+  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;
+
+#ifndef ROCKSDB_LITE
+    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);
+      }
+    }
+#else
+    (void)file_name;
+#endif  // ROCKSDB_LITE
+    return s;
+  }
+
+ private:
+  SstFileManager* sst_file_manager_;
+  InstrumentedMutex* mutex_;
+  ErrorHandler* error_handler_;
+};
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_garbage.cc

@@ -0,0 +1,134 @@
+//  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_garbage.h"
+
+#include <ostream>
+#include <sstream>
+
+#include "logging/event_logger.h"
+#include "rocksdb/slice.h"
+#include "rocksdb/status.h"
+#include "test_util/sync_point.h"
+#include "util/coding.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+// Tags for custom fields. Note that these get persisted in the manifest,
+// so existing tags should not be modified.
+enum BlobFileGarbage::CustomFieldTags : uint32_t {
+  kEndMarker,
+
+  // Add forward compatible fields here
+
+  /////////////////////////////////////////////////////////////////////
+
+  kForwardIncompatibleMask = 1 << 6,
+
+  // Add forward incompatible fields here
+};
+
+void BlobFileGarbage::EncodeTo(std::string* output) const {
+  PutVarint64(output, blob_file_number_);
+  PutVarint64(output, garbage_blob_count_);
+  PutVarint64(output, garbage_blob_bytes_);
+
+  // Encode any custom fields here. The format to use is a Varint32 tag (see
+  // CustomFieldTags above) followed by a length prefixed slice. Unknown custom
+  // fields will be ignored during decoding unless they're in the forward
+  // incompatible range.
+
+  TEST_SYNC_POINT_CALLBACK("BlobFileGarbage::EncodeTo::CustomFields", output);
+
+  PutVarint32(output, kEndMarker);
+}
+
+Status BlobFileGarbage::DecodeFrom(Slice* input) {
+  constexpr char class_name[] = "BlobFileGarbage";
+
+  if (!GetVarint64(input, &blob_file_number_)) {
+    return Status::Corruption(class_name, "Error decoding blob file number");
+  }
+
+  if (!GetVarint64(input, &garbage_blob_count_)) {
+    return Status::Corruption(class_name, "Error decoding garbage blob count");
+  }
+
+  if (!GetVarint64(input, &garbage_blob_bytes_)) {
+    return Status::Corruption(class_name, "Error decoding garbage blob bytes");
+  }
+
+  while (true) {
+    uint32_t custom_field_tag = 0;
+    if (!GetVarint32(input, &custom_field_tag)) {
+      return Status::Corruption(class_name, "Error decoding custom field tag");
+    }
+
+    if (custom_field_tag == kEndMarker) {
+      break;
+    }
+
+    if (custom_field_tag & kForwardIncompatibleMask) {
+      return Status::Corruption(
+          class_name, "Forward incompatible custom field encountered");
+    }
+
+    Slice custom_field_value;
+    if (!GetLengthPrefixedSlice(input, &custom_field_value)) {
+      return Status::Corruption(class_name,
+                                "Error decoding custom field value");
+    }
+  }
+
+  return Status::OK();
+}
+
+std::string BlobFileGarbage::DebugString() const {
+  std::ostringstream oss;
+
+  oss << *this;
+
+  return oss.str();
+}
+
+std::string BlobFileGarbage::DebugJSON() const {
+  JSONWriter jw;
+
+  jw << *this;
+
+  jw.EndObject();
+
+  return jw.Get();
+}
+
+bool operator==(const BlobFileGarbage& lhs, const BlobFileGarbage& rhs) {
+  return lhs.GetBlobFileNumber() == rhs.GetBlobFileNumber() &&
+         lhs.GetGarbageBlobCount() == rhs.GetGarbageBlobCount() &&
+         lhs.GetGarbageBlobBytes() == rhs.GetGarbageBlobBytes();
+}
+
+bool operator!=(const BlobFileGarbage& lhs, const BlobFileGarbage& rhs) {
+  return !(lhs == rhs);
+}
+
+std::ostream& operator<<(std::ostream& os,
+                         const BlobFileGarbage& blob_file_garbage) {
+  os << "blob_file_number: " << blob_file_garbage.GetBlobFileNumber()
+     << " garbage_blob_count: " << blob_file_garbage.GetGarbageBlobCount()
+     << " garbage_blob_bytes: " << blob_file_garbage.GetGarbageBlobBytes();
+
+  return os;
+}
+
+JSONWriter& operator<<(JSONWriter& jw,
+                       const BlobFileGarbage& blob_file_garbage) {
+  jw << "BlobFileNumber" << blob_file_garbage.GetBlobFileNumber()
+     << "GarbageBlobCount" << blob_file_garbage.GetGarbageBlobCount()
+     << "GarbageBlobBytes" << blob_file_garbage.GetGarbageBlobBytes();
+
+  return jw;
+}
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_garbage.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).
+
+#pragma once
+
+#include <cstdint>
+#include <iosfwd>
+#include <string>
+
+#include "db/blob/blob_constants.h"
+#include "rocksdb/rocksdb_namespace.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class JSONWriter;
+class Slice;
+class Status;
+
+class BlobFileGarbage {
+ public:
+  BlobFileGarbage() = default;
+
+  BlobFileGarbage(uint64_t blob_file_number, uint64_t garbage_blob_count,
+                  uint64_t garbage_blob_bytes)
+      : blob_file_number_(blob_file_number),
+        garbage_blob_count_(garbage_blob_count),
+        garbage_blob_bytes_(garbage_blob_bytes) {}
+
+  uint64_t GetBlobFileNumber() const { return blob_file_number_; }
+  uint64_t GetGarbageBlobCount() const { return garbage_blob_count_; }
+  uint64_t GetGarbageBlobBytes() const { return garbage_blob_bytes_; }
+
+  void EncodeTo(std::string* output) const;
+  Status DecodeFrom(Slice* input);
+
+  std::string DebugString() const;
+  std::string DebugJSON() const;
+
+ private:
+  enum CustomFieldTags : uint32_t;
+
+  uint64_t blob_file_number_ = kInvalidBlobFileNumber;
+  uint64_t garbage_blob_count_ = 0;
+  uint64_t garbage_blob_bytes_ = 0;
+};
+
+bool operator==(const BlobFileGarbage& lhs, const BlobFileGarbage& rhs);
+bool operator!=(const BlobFileGarbage& lhs, const BlobFileGarbage& rhs);
+
+std::ostream& operator<<(std::ostream& os,
+                         const BlobFileGarbage& blob_file_garbage);
+JSONWriter& operator<<(JSONWriter& jw,
+                       const BlobFileGarbage& blob_file_garbage);
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_garbage_test.cc

@@ -0,0 +1,173 @@
+//  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_garbage.h"
+
+#include <cstdint>
+#include <cstring>
+#include <string>
+
+#include "test_util/sync_point.h"
+#include "test_util/testharness.h"
+#include "util/coding.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class BlobFileGarbageTest : public testing::Test {
+ public:
+  static void TestEncodeDecode(const BlobFileGarbage& blob_file_garbage) {
+    std::string encoded;
+    blob_file_garbage.EncodeTo(&encoded);
+
+    BlobFileGarbage decoded;
+    Slice input(encoded);
+    ASSERT_OK(decoded.DecodeFrom(&input));
+
+    ASSERT_EQ(blob_file_garbage, decoded);
+  }
+};
+
+TEST_F(BlobFileGarbageTest, Empty) {
+  BlobFileGarbage blob_file_garbage;
+
+  ASSERT_EQ(blob_file_garbage.GetBlobFileNumber(), kInvalidBlobFileNumber);
+  ASSERT_EQ(blob_file_garbage.GetGarbageBlobCount(), 0);
+  ASSERT_EQ(blob_file_garbage.GetGarbageBlobBytes(), 0);
+
+  TestEncodeDecode(blob_file_garbage);
+}
+
+TEST_F(BlobFileGarbageTest, NonEmpty) {
+  constexpr uint64_t blob_file_number = 123;
+  constexpr uint64_t garbage_blob_count = 1;
+  constexpr uint64_t garbage_blob_bytes = 9876;
+
+  BlobFileGarbage blob_file_garbage(blob_file_number, garbage_blob_count,
+                                    garbage_blob_bytes);
+
+  ASSERT_EQ(blob_file_garbage.GetBlobFileNumber(), blob_file_number);
+  ASSERT_EQ(blob_file_garbage.GetGarbageBlobCount(), garbage_blob_count);
+  ASSERT_EQ(blob_file_garbage.GetGarbageBlobBytes(), garbage_blob_bytes);
+
+  TestEncodeDecode(blob_file_garbage);
+}
+
+TEST_F(BlobFileGarbageTest, DecodeErrors) {
+  std::string str;
+  Slice slice(str);
+
+  BlobFileGarbage blob_file_garbage;
+
+  {
+    const Status s = blob_file_garbage.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "blob file number"));
+  }
+
+  constexpr uint64_t blob_file_number = 123;
+  PutVarint64(&str, blob_file_number);
+  slice = str;
+
+  {
+    const Status s = blob_file_garbage.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "garbage blob count"));
+  }
+
+  constexpr uint64_t garbage_blob_count = 4567;
+  PutVarint64(&str, garbage_blob_count);
+  slice = str;
+
+  {
+    const Status s = blob_file_garbage.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "garbage blob bytes"));
+  }
+
+  constexpr uint64_t garbage_blob_bytes = 12345678;
+  PutVarint64(&str, garbage_blob_bytes);
+  slice = str;
+
+  {
+    const Status s = blob_file_garbage.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "custom field tag"));
+  }
+
+  constexpr uint32_t custom_tag = 2;
+  PutVarint32(&str, custom_tag);
+  slice = str;
+
+  {
+    const Status s = blob_file_garbage.DecodeFrom(&slice);
+    ASSERT_TRUE(s.IsCorruption());
+    ASSERT_TRUE(std::strstr(s.getState(), "custom field value"));
+  }
+}
+
+TEST_F(BlobFileGarbageTest, ForwardCompatibleCustomField) {
+  SyncPoint::GetInstance()->SetCallBack(
+      "BlobFileGarbage::EncodeTo::CustomFields", [&](void* arg) {
+        std::string* output = static_cast<std::string*>(arg);
+
+        constexpr uint32_t forward_compatible_tag = 2;
+        PutVarint32(output, forward_compatible_tag);
+
+        PutLengthPrefixedSlice(output, "deadbeef");
+      });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  constexpr uint64_t blob_file_number = 678;
+  constexpr uint64_t garbage_blob_count = 9999;
+  constexpr uint64_t garbage_blob_bytes = 100000000;
+
+  BlobFileGarbage blob_file_garbage(blob_file_number, garbage_blob_count,
+                                    garbage_blob_bytes);
+
+  TestEncodeDecode(blob_file_garbage);
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+TEST_F(BlobFileGarbageTest, ForwardIncompatibleCustomField) {
+  SyncPoint::GetInstance()->SetCallBack(
+      "BlobFileGarbage::EncodeTo::CustomFields", [&](void* arg) {
+        std::string* output = static_cast<std::string*>(arg);
+
+        constexpr uint32_t forward_incompatible_tag = (1 << 6) + 1;
+        PutVarint32(output, forward_incompatible_tag);
+
+        PutLengthPrefixedSlice(output, "foobar");
+      });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  constexpr uint64_t blob_file_number = 456;
+  constexpr uint64_t garbage_blob_count = 100;
+  constexpr uint64_t garbage_blob_bytes = 2000000;
+
+  BlobFileGarbage blob_file_garbage(blob_file_number, garbage_blob_count,
+                                    garbage_blob_bytes);
+
+  std::string encoded;
+  blob_file_garbage.EncodeTo(&encoded);
+
+  BlobFileGarbage decoded_blob_file_addition;
+  Slice input(encoded);
+  const Status s = decoded_blob_file_addition.DecodeFrom(&input);
+
+  ASSERT_TRUE(s.IsCorruption());
+  ASSERT_TRUE(std::strstr(s.getState(), "Forward incompatible"));
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+}  // namespace ROCKSDB_NAMESPACE
+
+int main(int argc, char** argv) {
+  ::testing::InitGoogleTest(&argc, argv);
+  return RUN_ALL_TESTS();
+}

db/blob/blob_file_meta.cc

@@ -0,0 +1,55 @@
+//  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_meta.h"
+
+#include <ostream>
+#include <sstream>
+
+namespace ROCKSDB_NAMESPACE {
+
+std::string SharedBlobFileMetaData::DebugString() const {
+  std::ostringstream oss;
+  oss << (*this);
+
+  return oss.str();
+}
+
+std::ostream& operator<<(std::ostream& os,
+                         const SharedBlobFileMetaData& shared_meta) {
+  os << "blob_file_number: " << shared_meta.GetBlobFileNumber()
+     << " total_blob_count: " << shared_meta.GetTotalBlobCount()
+     << " total_blob_bytes: " << shared_meta.GetTotalBlobBytes()
+     << " checksum_method: " << shared_meta.GetChecksumMethod()
+     << " checksum_value: " << shared_meta.GetChecksumValue();
+
+  return os;
+}
+
+std::string BlobFileMetaData::DebugString() const {
+  std::ostringstream oss;
+  oss << (*this);
+
+  return oss.str();
+}
+
+std::ostream& operator<<(std::ostream& os, const BlobFileMetaData& meta) {
+  const auto& shared_meta = meta.GetSharedMeta();
+  assert(shared_meta);
+  os << (*shared_meta);
+
+  os << " linked_ssts: {";
+  for (uint64_t file_number : meta.GetLinkedSsts()) {
+    os << ' ' << file_number;
+  }
+  os << " }";
+
+  os << " garbage_blob_count: " << meta.GetGarbageBlobCount()
+     << " garbage_blob_bytes: " << meta.GetGarbageBlobBytes();
+
+  return os;
+}
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_meta.h

@@ -0,0 +1,164 @@
+//  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 <iosfwd>
+#include <memory>
+#include <string>
+#include <unordered_set>
+
+#include "rocksdb/rocksdb_namespace.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+// SharedBlobFileMetaData represents the immutable part of blob files' metadata,
+// like the blob file number, total number and size of blobs, or checksum
+// method and value. There is supposed to be one object of this class per blob
+// file (shared across all versions that include the blob file in question);
+// hence, the type is neither copyable nor movable. A blob file can be marked
+// obsolete when the corresponding SharedBlobFileMetaData object is destroyed.
+
+class SharedBlobFileMetaData {
+ public:
+  static std::shared_ptr<SharedBlobFileMetaData> Create(
+      uint64_t blob_file_number, uint64_t total_blob_count,
+      uint64_t total_blob_bytes, std::string checksum_method,
+      std::string checksum_value) {
+    return std::shared_ptr<SharedBlobFileMetaData>(new SharedBlobFileMetaData(
+        blob_file_number, total_blob_count, total_blob_bytes,
+        std::move(checksum_method), std::move(checksum_value)));
+  }
+
+  template <typename Deleter>
+  static std::shared_ptr<SharedBlobFileMetaData> Create(
+      uint64_t blob_file_number, uint64_t total_blob_count,
+      uint64_t total_blob_bytes, std::string checksum_method,
+      std::string checksum_value, Deleter deleter) {
+    return std::shared_ptr<SharedBlobFileMetaData>(
+        new SharedBlobFileMetaData(blob_file_number, total_blob_count,
+                                   total_blob_bytes, std::move(checksum_method),
+                                   std::move(checksum_value)),
+        deleter);
+  }
+
+  SharedBlobFileMetaData(const SharedBlobFileMetaData&) = delete;
+  SharedBlobFileMetaData& operator=(const SharedBlobFileMetaData&) = delete;
+
+  SharedBlobFileMetaData(SharedBlobFileMetaData&&) = delete;
+  SharedBlobFileMetaData& operator=(SharedBlobFileMetaData&&) = delete;
+
+  uint64_t GetBlobFileNumber() const { return blob_file_number_; }
+  uint64_t GetTotalBlobCount() const { return total_blob_count_; }
+  uint64_t GetTotalBlobBytes() const { return total_blob_bytes_; }
+  const std::string& GetChecksumMethod() const { return checksum_method_; }
+  const std::string& GetChecksumValue() const { return checksum_value_; }
+
+  std::string DebugString() const;
+
+ private:
+  SharedBlobFileMetaData(uint64_t blob_file_number, uint64_t total_blob_count,
+                         uint64_t total_blob_bytes, std::string checksum_method,
+                         std::string checksum_value)
+      : blob_file_number_(blob_file_number),
+        total_blob_count_(total_blob_count),
+        total_blob_bytes_(total_blob_bytes),
+        checksum_method_(std::move(checksum_method)),
+        checksum_value_(std::move(checksum_value)) {
+    assert(checksum_method_.empty() == checksum_value_.empty());
+  }
+
+  uint64_t blob_file_number_;
+  uint64_t total_blob_count_;
+  uint64_t total_blob_bytes_;
+  std::string checksum_method_;
+  std::string checksum_value_;
+};
+
+std::ostream& operator<<(std::ostream& os,
+                         const SharedBlobFileMetaData& shared_meta);
+
+// BlobFileMetaData contains the part of the metadata for blob files that can
+// vary across versions, like the amount of garbage in the blob file. In
+// addition, BlobFileMetaData objects point to and share the ownership of the
+// SharedBlobFileMetaData object for the corresponding blob file. Similarly to
+// SharedBlobFileMetaData, BlobFileMetaData are not copyable or movable. They
+// are meant to be jointly owned by the versions in which the blob file has the
+// same (immutable *and* mutable) state.
+
+class BlobFileMetaData {
+ public:
+  using LinkedSsts = std::unordered_set<uint64_t>;
+
+  static std::shared_ptr<BlobFileMetaData> Create(
+      std::shared_ptr<SharedBlobFileMetaData> shared_meta,
+      LinkedSsts linked_ssts, uint64_t garbage_blob_count,
+      uint64_t garbage_blob_bytes) {
+    return std::shared_ptr<BlobFileMetaData>(
+        new BlobFileMetaData(std::move(shared_meta), std::move(linked_ssts),
+                             garbage_blob_count, garbage_blob_bytes));
+  }
+
+  BlobFileMetaData(const BlobFileMetaData&) = delete;
+  BlobFileMetaData& operator=(const BlobFileMetaData&) = delete;
+
+  BlobFileMetaData(BlobFileMetaData&&) = delete;
+  BlobFileMetaData& operator=(BlobFileMetaData&&) = delete;
+
+  const std::shared_ptr<SharedBlobFileMetaData>& GetSharedMeta() const {
+    return shared_meta_;
+  }
+
+  uint64_t GetBlobFileNumber() const {
+    assert(shared_meta_);
+    return shared_meta_->GetBlobFileNumber();
+  }
+  uint64_t GetTotalBlobCount() const {
+    assert(shared_meta_);
+    return shared_meta_->GetTotalBlobCount();
+  }
+  uint64_t GetTotalBlobBytes() const {
+    assert(shared_meta_);
+    return shared_meta_->GetTotalBlobBytes();
+  }
+  const std::string& GetChecksumMethod() const {
+    assert(shared_meta_);
+    return shared_meta_->GetChecksumMethod();
+  }
+  const std::string& GetChecksumValue() const {
+    assert(shared_meta_);
+    return shared_meta_->GetChecksumValue();
+  }
+
+  const LinkedSsts& GetLinkedSsts() const { return linked_ssts_; }
+
+  uint64_t GetGarbageBlobCount() const { return garbage_blob_count_; }
+  uint64_t GetGarbageBlobBytes() const { return garbage_blob_bytes_; }
+
+  std::string DebugString() const;
+
+ private:
+  BlobFileMetaData(std::shared_ptr<SharedBlobFileMetaData> shared_meta,
+                   LinkedSsts linked_ssts, uint64_t garbage_blob_count,
+                   uint64_t garbage_blob_bytes)
+      : shared_meta_(std::move(shared_meta)),
+        linked_ssts_(std::move(linked_ssts)),
+        garbage_blob_count_(garbage_blob_count),
+        garbage_blob_bytes_(garbage_blob_bytes) {
+    assert(shared_meta_);
+    assert(garbage_blob_count_ <= shared_meta_->GetTotalBlobCount());
+    assert(garbage_blob_bytes_ <= shared_meta_->GetTotalBlobBytes());
+  }
+
+  std::shared_ptr<SharedBlobFileMetaData> shared_meta_;
+  LinkedSsts linked_ssts_;
+  uint64_t garbage_blob_count_;
+  uint64_t garbage_blob_bytes_;
+};
+
+std::ostream& operator<<(std::ostream& os, const BlobFileMetaData& meta);
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_reader.cc

@@ -0,0 +1,429 @@
+//  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_reader.h"
+
+#include <cassert>
+#include <string>
+
+#include "db/blob/blob_log_format.h"
+#include "file/filename.h"
+#include "options/cf_options.h"
+#include "rocksdb/file_system.h"
+#include "rocksdb/slice.h"
+#include "rocksdb/status.h"
+#include "test_util/sync_point.h"
+#include "util/compression.h"
+#include "util/crc32c.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+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);
+
+  uint64_t file_size = 0;
+  std::unique_ptr<RandomAccessFileReader> file_reader;
+
+  {
+    const Status s =
+        OpenFile(immutable_cf_options, file_options, blob_file_read_hist,
+                 blob_file_number, io_tracer, &file_size, &file_reader);
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  assert(file_reader);
+
+  CompressionType compression_type = kNoCompression;
+
+  {
+    const Status s =
+        ReadHeader(file_reader.get(), column_family_id, &compression_type);
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  {
+    const Status s = ReadFooter(file_size, file_reader.get());
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  blob_file_reader->reset(
+      new BlobFileReader(std::move(file_reader), file_size, compression_type));
+
+  return Status::OK();
+}
+
+Status BlobFileReader::OpenFile(
+    const ImmutableCFOptions& immutable_cf_options,
+    const FileOptions& file_opts, HistogramImpl* blob_file_read_hist,
+    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);
+
+  const auto& cf_paths = immutable_cf_options.cf_paths;
+  assert(!cf_paths.empty());
+
+  const std::string blob_file_path =
+      BlobFileName(cf_paths.front().path, blob_file_number);
+
+  FileSystem* const fs = immutable_cf_options.fs;
+  assert(fs);
+
+  constexpr IODebugContext* dbg = nullptr;
+
+  {
+    TEST_SYNC_POINT("BlobFileReader::OpenFile:GetFileSize");
+
+    const Status s =
+        fs->GetFileSize(blob_file_path, IOOptions(), file_size, dbg);
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  if (*file_size < BlobLogHeader::kSize + BlobLogFooter::kSize) {
+    return Status::Corruption("Malformed blob file");
+  }
+
+  std::unique_ptr<FSRandomAccessFile> file;
+
+  {
+    TEST_SYNC_POINT("BlobFileReader::OpenFile:NewRandomAccessFile");
+
+    const Status s =
+        fs->NewRandomAccessFile(blob_file_path, file_opts, &file, dbg);
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  assert(file);
+
+  if (immutable_cf_options.advise_random_on_open) {
+    file->Hint(FSRandomAccessFile::kRandom);
+  }
+
+  file_reader->reset(new RandomAccessFileReader(
+      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,
+      immutable_cf_options.listeners));
+
+  return Status::OK();
+}
+
+Status BlobFileReader::ReadHeader(const RandomAccessFileReader* file_reader,
+                                  uint32_t column_family_id,
+                                  CompressionType* compression_type) {
+  assert(file_reader);
+  assert(compression_type);
+
+  Slice header_slice;
+  Buffer buf;
+  AlignedBuf aligned_buf;
+
+  {
+    TEST_SYNC_POINT("BlobFileReader::ReadHeader:ReadFromFile");
+
+    constexpr uint64_t read_offset = 0;
+    constexpr size_t read_size = BlobLogHeader::kSize;
+
+    const Status s = ReadFromFile(file_reader, read_offset, read_size,
+                                  &header_slice, &buf, &aligned_buf);
+    if (!s.ok()) {
+      return s;
+    }
+
+    TEST_SYNC_POINT_CALLBACK("BlobFileReader::ReadHeader:TamperWithResult",
+                             &header_slice);
+  }
+
+  BlobLogHeader header;
+
+  {
+    const Status s = header.DecodeFrom(header_slice);
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  constexpr ExpirationRange no_expiration_range;
+
+  if (header.has_ttl || header.expiration_range != no_expiration_range) {
+    return Status::Corruption("Unexpected TTL blob file");
+  }
+
+  if (header.column_family_id != column_family_id) {
+    return Status::Corruption("Column family ID mismatch");
+  }
+
+  *compression_type = header.compression;
+
+  return Status::OK();
+}
+
+Status BlobFileReader::ReadFooter(uint64_t file_size,
+                                  const RandomAccessFileReader* file_reader) {
+  assert(file_size >= BlobLogHeader::kSize + BlobLogFooter::kSize);
+  assert(file_reader);
+
+  Slice footer_slice;
+  Buffer buf;
+  AlignedBuf aligned_buf;
+
+  {
+    TEST_SYNC_POINT("BlobFileReader::ReadFooter:ReadFromFile");
+
+    const uint64_t read_offset = file_size - BlobLogFooter::kSize;
+    constexpr size_t read_size = BlobLogFooter::kSize;
+
+    const Status s = ReadFromFile(file_reader, read_offset, read_size,
+                                  &footer_slice, &buf, &aligned_buf);
+    if (!s.ok()) {
+      return s;
+    }
+
+    TEST_SYNC_POINT_CALLBACK("BlobFileReader::ReadFooter:TamperWithResult",
+                             &footer_slice);
+  }
+
+  BlobLogFooter footer;
+
+  {
+    const Status s = footer.DecodeFrom(footer_slice);
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  constexpr ExpirationRange no_expiration_range;
+
+  if (footer.expiration_range != no_expiration_range) {
+    return Status::Corruption("Unexpected TTL blob file");
+  }
+
+  return Status::OK();
+}
+
+Status BlobFileReader::ReadFromFile(const RandomAccessFileReader* file_reader,
+                                    uint64_t read_offset, size_t read_size,
+                                    Slice* slice, Buffer* buf,
+                                    AlignedBuf* aligned_buf) {
+  assert(slice);
+  assert(buf);
+  assert(aligned_buf);
+
+  assert(file_reader);
+
+  Status s;
+
+  if (file_reader->use_direct_io()) {
+    constexpr char* scratch = nullptr;
+
+    s = file_reader->Read(IOOptions(), read_offset, read_size, slice, scratch,
+                          aligned_buf);
+  } else {
+    buf->reset(new char[read_size]);
+    constexpr AlignedBuf* aligned_scratch = nullptr;
+
+    s = file_reader->Read(IOOptions(), read_offset, read_size, slice,
+                          buf->get(), aligned_scratch);
+  }
+
+  if (!s.ok()) {
+    return s;
+  }
+
+  if (slice->size() != read_size) {
+    return Status::Corruption("Failed to read data from blob file");
+  }
+
+  return Status::OK();
+}
+
+BlobFileReader::BlobFileReader(
+    std::unique_ptr<RandomAccessFileReader>&& file_reader, uint64_t file_size,
+    CompressionType compression_type)
+    : file_reader_(std::move(file_reader)),
+      file_size_(file_size),
+      compression_type_(compression_type) {
+  assert(file_reader_);
+}
+
+BlobFileReader::~BlobFileReader() = default;
+
+Status BlobFileReader::GetBlob(const ReadOptions& read_options,
+                               const Slice& user_key, uint64_t offset,
+                               uint64_t value_size,
+                               CompressionType compression_type,
+                               PinnableSlice* value,
+                               uint64_t* bytes_read) const {
+  assert(value);
+
+  const uint64_t key_size = user_key.size();
+
+  if (!IsValidBlobOffset(offset, key_size, value_size, file_size_)) {
+    return Status::Corruption("Invalid blob offset");
+  }
+
+  if (compression_type != compression_type_) {
+    return Status::Corruption("Compression type mismatch when reading blob");
+  }
+
+  // Note: if verify_checksum is set, we read the entire blob record to be able
+  // to perform the verification; otherwise, we just read the blob itself. Since
+  // the offset in BlobIndex actually points to the blob value, we need to make
+  // an adjustment in the former case.
+  const uint64_t adjustment =
+      read_options.verify_checksums
+          ? BlobLogRecord::CalculateAdjustmentForRecordHeader(key_size)
+          : 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;
+
+  {
+    TEST_SYNC_POINT("BlobFileReader::GetBlob:ReadFromFile");
+
+    const Status s = ReadFromFile(file_reader_.get(), record_offset,
+                                  static_cast<size_t>(record_size),
+                                  &record_slice, &buf, &aligned_buf);
+    if (!s.ok()) {
+      return s;
+    }
+
+    TEST_SYNC_POINT_CALLBACK("BlobFileReader::GetBlob:TamperWithResult",
+                             &record_slice);
+  }
+
+  if (read_options.verify_checksums) {
+    const Status s = VerifyBlob(record_slice, user_key, value_size);
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  const Slice value_slice(record_slice.data() + adjustment, value_size);
+
+  {
+    const Status s =
+        UncompressBlobIfNeeded(value_slice, compression_type, value);
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  if (bytes_read) {
+    *bytes_read = record_size;
+  }
+
+  return Status::OK();
+}
+
+Status BlobFileReader::VerifyBlob(const Slice& record_slice,
+                                  const Slice& user_key, uint64_t value_size) {
+  BlobLogRecord record;
+
+  const Slice header_slice(record_slice.data(), BlobLogRecord::kHeaderSize);
+
+  {
+    const Status s = record.DecodeHeaderFrom(header_slice);
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  if (record.key_size != user_key.size()) {
+    return Status::Corruption("Key size mismatch when reading blob");
+  }
+
+  if (record.value_size != value_size) {
+    return Status::Corruption("Value size mismatch when reading blob");
+  }
+
+  record.key =
+      Slice(record_slice.data() + BlobLogRecord::kHeaderSize, record.key_size);
+  if (record.key != user_key) {
+    return Status::Corruption("Key mismatch when reading blob");
+  }
+
+  record.value = Slice(record.key.data() + record.key_size, value_size);
+
+  {
+    TEST_SYNC_POINT_CALLBACK("BlobFileReader::VerifyBlob:CheckBlobCRC",
+                             &record);
+
+    const Status s = record.CheckBlobCRC();
+    if (!s.ok()) {
+      return s;
+    }
+  }
+
+  return Status::OK();
+}
+
+Status BlobFileReader::UncompressBlobIfNeeded(const Slice& value_slice,
+                                              CompressionType compression_type,
+                                              PinnableSlice* value) {
+  assert(value);
+
+  if (compression_type == kNoCompression) {
+    SaveValue(value_slice, value);
+
+    return Status::OK();
+  }
+
+  UncompressionContext context(compression_type);
+  UncompressionInfo info(context, UncompressionDict::GetEmptyDict(),
+                         compression_type);
+
+  size_t uncompressed_size = 0;
+  constexpr uint32_t compression_format_version = 2;
+  constexpr MemoryAllocator* allocator = nullptr;
+
+  CacheAllocationPtr output =
+      UncompressData(info, value_slice.data(), value_slice.size(),
+                     &uncompressed_size, compression_format_version, allocator);
+
+  TEST_SYNC_POINT_CALLBACK(
+      "BlobFileReader::UncompressBlobIfNeeded:TamperWithResult", &output);
+
+  if (!output) {
+    return Status::Corruption("Unable to uncompress blob");
+  }
+
+  SaveValue(Slice(output.get(), uncompressed_size), value);
+
+  return Status::OK();
+}
+
+void BlobFileReader::SaveValue(const Slice& src, PinnableSlice* dst) {
+  assert(dst);
+
+  if (dst->IsPinned()) {
+    dst->Reset();
+  }
+
+  dst->PinSelf(src);
+}
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_reader.h

@@ -0,0 +1,85 @@
+//  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 <memory>
+
+#include "file/random_access_file_reader.h"
+#include "rocksdb/compression_type.h"
+#include "rocksdb/rocksdb_namespace.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class Status;
+struct ImmutableCFOptions;
+struct FileOptions;
+class HistogramImpl;
+struct ReadOptions;
+class Slice;
+class PinnableSlice;
+
+class BlobFileReader {
+ public:
+  static Status 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>* reader);
+
+  BlobFileReader(const BlobFileReader&) = delete;
+  BlobFileReader& operator=(const BlobFileReader&) = delete;
+
+  ~BlobFileReader();
+
+  Status GetBlob(const ReadOptions& read_options, const Slice& user_key,
+                 uint64_t offset, uint64_t value_size,
+                 CompressionType compression_type, PinnableSlice* value,
+                 uint64_t* bytes_read) const;
+
+ private:
+  BlobFileReader(std::unique_ptr<RandomAccessFileReader>&& file_reader,
+                 uint64_t file_size, CompressionType compression_type);
+
+  static Status OpenFile(const ImmutableCFOptions& immutable_cf_options,
+                         const FileOptions& file_opts,
+                         HistogramImpl* blob_file_read_hist,
+                         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,
+                           uint32_t column_family_id,
+                           CompressionType* compression_type);
+
+  static Status ReadFooter(uint64_t file_size,
+                           const RandomAccessFileReader* file_reader);
+
+  using Buffer = std::unique_ptr<char[]>;
+
+  static Status ReadFromFile(const RandomAccessFileReader* file_reader,
+                             uint64_t read_offset, size_t read_size,
+                             Slice* slice, Buffer* buf,
+                             AlignedBuf* aligned_buf);
+
+  static Status VerifyBlob(const Slice& record_slice, const Slice& user_key,
+                           uint64_t value_size);
+
+  static Status UncompressBlobIfNeeded(const Slice& value_slice,
+                                       CompressionType compression_type,
+                                       PinnableSlice* value);
+
+  static void SaveValue(const Slice& src, PinnableSlice* dst);
+
+  std::unique_ptr<RandomAccessFileReader> file_reader_;
+  uint64_t file_size_;
+  CompressionType compression_type_;
+};
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_reader_test.cc

@@ -0,0 +1,813 @@
+//  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_reader.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/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"
+
+namespace ROCKSDB_NAMESPACE {
+
+namespace {
+
+// Creates a test blob file with a single blob in it. Note: this method
+// makes it possible to test various corner cases by allowing the caller
+// to specify the contents of various blob file header/footer fields.
+void WriteBlobFile(const ImmutableCFOptions& immutable_cf_options,
+                   uint32_t column_family_id, bool has_ttl,
+                   const ExpirationRange& expiration_range_header,
+                   const ExpirationRange& expiration_range_footer,
+                   uint64_t blob_file_number, const Slice& key,
+                   const Slice& blob, CompressionType compression_type,
+                   uint64_t* blob_offset, uint64_t* blob_size) {
+  assert(!immutable_cf_options.cf_paths.empty());
+  assert(blob_offset);
+  assert(blob_size);
+
+  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);
+
+  BlobLogHeader header(column_family_id, compression_type, has_ttl,
+                       expiration_range_header);
+
+  ASSERT_OK(blob_log_writer.WriteHeader(header));
+
+  std::string compressed_blob;
+  Slice blob_to_write;
+
+  if (compression_type == kNoCompression) {
+    blob_to_write = blob;
+    *blob_size = blob.size();
+  } else {
+    CompressionOptions opts;
+    CompressionContext context(compression_type);
+    constexpr uint64_t sample_for_compression = 0;
+
+    CompressionInfo info(opts, context, CompressionDict::GetEmptyDict(),
+                         compression_type, sample_for_compression);
+
+    constexpr uint32_t compression_format_version = 2;
+
+    ASSERT_TRUE(
+        CompressData(blob, info, compression_format_version, &compressed_blob));
+
+    blob_to_write = compressed_blob;
+    *blob_size = compressed_blob.size();
+  }
+
+  uint64_t key_offset = 0;
+
+  ASSERT_OK(
+      blob_log_writer.AddRecord(key, blob_to_write, &key_offset, blob_offset));
+
+  BlobLogFooter footer;
+  footer.blob_count = 1;
+  footer.expiration_range = expiration_range_footer;
+
+  std::string checksum_method;
+  std::string checksum_value;
+
+  ASSERT_OK(
+      blob_log_writer.AppendFooter(footer, &checksum_method, &checksum_value));
+}
+
+}  // anonymous namespace
+
+class BlobFileReaderTest : public testing::Test {
+ protected:
+  BlobFileReaderTest() : mock_env_(Env::Default()) {}
+
+  MockEnv mock_env_;
+};
+
+TEST_F(BlobFileReaderTest, CreateReaderAndGetBlob) {
+  Options options;
+  options.env = &mock_env_;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_,
+                            "BlobFileReaderTest_CreateReaderAndGetBlob"),
+      0);
+  options.enable_blob_files = true;
+
+  ImmutableCFOptions immutable_cf_options(options);
+
+  constexpr uint32_t column_family_id = 1;
+  constexpr bool has_ttl = false;
+  constexpr ExpirationRange expiration_range;
+  constexpr uint64_t blob_file_number = 1;
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+
+  uint64_t blob_offset = 0;
+  uint64_t blob_size = 0;
+
+  WriteBlobFile(immutable_cf_options, column_family_id, has_ttl,
+                expiration_range, expiration_range, blob_file_number, key, blob,
+                kNoCompression, &blob_offset, &blob_size);
+
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  std::unique_ptr<BlobFileReader> 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;
+  read_options.verify_checksums = false;
+
+  {
+    PinnableSlice value;
+    uint64_t bytes_read = 0;
+
+    ASSERT_OK(reader->GetBlob(read_options, key, blob_offset, blob_size,
+                              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, &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, &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, &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, &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, &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, &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, &bytes_read)
+                    .IsCorruption());
+    ASSERT_EQ(bytes_read, 0);
+  }
+}
+
+TEST_F(BlobFileReaderTest, Malformed) {
+  // Write a blob file consisting of nothing but a header, and make sure we
+  // detect the error when we open it for reading
+
+  Options options;
+  options.env = &mock_env_;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_, "BlobFileReaderTest_Malformed"), 0);
+  options.enable_blob_files = true;
+
+  ImmutableCFOptions immutable_cf_options(options);
+
+  constexpr uint32_t column_family_id = 1;
+  constexpr uint64_t blob_file_number = 1;
+
+  {
+    constexpr bool has_ttl = false;
+    constexpr ExpirationRange expiration_range;
+
+    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);
+
+    BlobLogHeader header(column_family_id, kNoCompression, has_ttl,
+                         expiration_range);
+
+    ASSERT_OK(blob_log_writer.WriteHeader(header));
+  }
+
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  std::unique_ptr<BlobFileReader> reader;
+
+  ASSERT_TRUE(BlobFileReader::Create(immutable_cf_options, FileOptions(),
+                                     column_family_id, blob_file_read_hist,
+                                     blob_file_number, nullptr /*IOTracer*/,
+                                     &reader)
+                  .IsCorruption());
+}
+
+TEST_F(BlobFileReaderTest, TTL) {
+  Options options;
+  options.env = &mock_env_;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_, "BlobFileReaderTest_TTL"), 0);
+  options.enable_blob_files = true;
+
+  ImmutableCFOptions immutable_cf_options(options);
+
+  constexpr uint32_t column_family_id = 1;
+  constexpr bool has_ttl = true;
+  constexpr ExpirationRange expiration_range;
+  constexpr uint64_t blob_file_number = 1;
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+
+  uint64_t blob_offset = 0;
+  uint64_t blob_size = 0;
+
+  WriteBlobFile(immutable_cf_options, column_family_id, has_ttl,
+                expiration_range, expiration_range, blob_file_number, key, blob,
+                kNoCompression, &blob_offset, &blob_size);
+
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  std::unique_ptr<BlobFileReader> reader;
+
+  ASSERT_TRUE(BlobFileReader::Create(immutable_cf_options, FileOptions(),
+                                     column_family_id, blob_file_read_hist,
+                                     blob_file_number, nullptr /*IOTracer*/,
+                                     &reader)
+                  .IsCorruption());
+}
+
+TEST_F(BlobFileReaderTest, ExpirationRangeInHeader) {
+  Options options;
+  options.env = &mock_env_;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_,
+                            "BlobFileReaderTest_ExpirationRangeInHeader"),
+      0);
+  options.enable_blob_files = true;
+
+  ImmutableCFOptions immutable_cf_options(options);
+
+  constexpr uint32_t column_family_id = 1;
+  constexpr bool has_ttl = false;
+  const ExpirationRange expiration_range_header(
+      1, 2);  // can be made constexpr when we adopt C++14
+  constexpr ExpirationRange expiration_range_footer;
+  constexpr uint64_t blob_file_number = 1;
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+
+  uint64_t blob_offset = 0;
+  uint64_t blob_size = 0;
+
+  WriteBlobFile(immutable_cf_options, column_family_id, has_ttl,
+                expiration_range_header, expiration_range_footer,
+                blob_file_number, key, blob, kNoCompression, &blob_offset,
+                &blob_size);
+
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  std::unique_ptr<BlobFileReader> reader;
+
+  ASSERT_TRUE(BlobFileReader::Create(immutable_cf_options, FileOptions(),
+                                     column_family_id, blob_file_read_hist,
+                                     blob_file_number, nullptr /*IOTracer*/,
+                                     &reader)
+                  .IsCorruption());
+}
+
+TEST_F(BlobFileReaderTest, ExpirationRangeInFooter) {
+  Options options;
+  options.env = &mock_env_;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_,
+                            "BlobFileReaderTest_ExpirationRangeInFooter"),
+      0);
+  options.enable_blob_files = true;
+
+  ImmutableCFOptions immutable_cf_options(options);
+
+  constexpr uint32_t column_family_id = 1;
+  constexpr bool has_ttl = false;
+  constexpr ExpirationRange expiration_range_header;
+  const ExpirationRange expiration_range_footer(
+      1, 2);  // can be made constexpr when we adopt C++14
+  constexpr uint64_t blob_file_number = 1;
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+
+  uint64_t blob_offset = 0;
+  uint64_t blob_size = 0;
+
+  WriteBlobFile(immutable_cf_options, column_family_id, has_ttl,
+                expiration_range_header, expiration_range_footer,
+                blob_file_number, key, blob, kNoCompression, &blob_offset,
+                &blob_size);
+
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  std::unique_ptr<BlobFileReader> reader;
+
+  ASSERT_TRUE(BlobFileReader::Create(immutable_cf_options, FileOptions(),
+                                     column_family_id, blob_file_read_hist,
+                                     blob_file_number, nullptr /*IOTracer*/,
+                                     &reader)
+                  .IsCorruption());
+}
+
+TEST_F(BlobFileReaderTest, IncorrectColumnFamily) {
+  Options options;
+  options.env = &mock_env_;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_,
+                            "BlobFileReaderTest_IncorrectColumnFamily"),
+      0);
+  options.enable_blob_files = true;
+
+  ImmutableCFOptions immutable_cf_options(options);
+
+  constexpr uint32_t column_family_id = 1;
+  constexpr bool has_ttl = false;
+  constexpr ExpirationRange expiration_range;
+  constexpr uint64_t blob_file_number = 1;
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+
+  uint64_t blob_offset = 0;
+  uint64_t blob_size = 0;
+
+  WriteBlobFile(immutable_cf_options, column_family_id, has_ttl,
+                expiration_range, expiration_range, blob_file_number, key, blob,
+                kNoCompression, &blob_offset, &blob_size);
+
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  std::unique_ptr<BlobFileReader> reader;
+
+  constexpr uint32_t incorrect_column_family_id = 2;
+
+  ASSERT_TRUE(BlobFileReader::Create(immutable_cf_options, FileOptions(),
+                                     incorrect_column_family_id,
+                                     blob_file_read_hist, blob_file_number,
+                                     nullptr /*IOTracer*/, &reader)
+                  .IsCorruption());
+}
+
+TEST_F(BlobFileReaderTest, BlobCRCError) {
+  Options options;
+  options.env = &mock_env_;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_, "BlobFileReaderTest_BlobCRCError"), 0);
+  options.enable_blob_files = true;
+
+  ImmutableCFOptions immutable_cf_options(options);
+
+  constexpr uint32_t column_family_id = 1;
+  constexpr bool has_ttl = false;
+  constexpr ExpirationRange expiration_range;
+  constexpr uint64_t blob_file_number = 1;
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+
+  uint64_t blob_offset = 0;
+  uint64_t blob_size = 0;
+
+  WriteBlobFile(immutable_cf_options, column_family_id, has_ttl,
+                expiration_range, expiration_range, blob_file_number, key, blob,
+                kNoCompression, &blob_offset, &blob_size);
+
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  std::unique_ptr<BlobFileReader> 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) {
+        BlobLogRecord* const record = static_cast<BlobLogRecord*>(arg);
+        assert(record);
+
+        record->blob_crc = 0xfaceb00c;
+      });
+
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  PinnableSlice value;
+  uint64_t bytes_read = 0;
+
+  ASSERT_TRUE(reader
+                  ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
+                            kNoCompression, &value, &bytes_read)
+                  .IsCorruption());
+  ASSERT_EQ(bytes_read, 0);
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+TEST_F(BlobFileReaderTest, Compression) {
+  if (!Snappy_Supported()) {
+    return;
+  }
+
+  Options options;
+  options.env = &mock_env_;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_, "BlobFileReaderTest_Compression"), 0);
+  options.enable_blob_files = true;
+
+  ImmutableCFOptions immutable_cf_options(options);
+
+  constexpr uint32_t column_family_id = 1;
+  constexpr bool has_ttl = false;
+  constexpr ExpirationRange expiration_range;
+  constexpr uint64_t blob_file_number = 1;
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+
+  uint64_t blob_offset = 0;
+  uint64_t blob_size = 0;
+
+  WriteBlobFile(immutable_cf_options, column_family_id, has_ttl,
+                expiration_range, expiration_range, blob_file_number, key, blob,
+                kSnappyCompression, &blob_offset, &blob_size);
+
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  std::unique_ptr<BlobFileReader> 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;
+  read_options.verify_checksums = false;
+
+  {
+    PinnableSlice value;
+    uint64_t bytes_read = 0;
+
+    ASSERT_OK(reader->GetBlob(read_options, key, blob_offset, blob_size,
+                              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, &bytes_read));
+    ASSERT_EQ(value, blob);
+
+    constexpr uint64_t key_size = sizeof(key) - 1;
+    ASSERT_EQ(bytes_read,
+              BlobLogRecord::CalculateAdjustmentForRecordHeader(key_size) +
+                  blob_size);
+  }
+}
+
+TEST_F(BlobFileReaderTest, UncompressionError) {
+  if (!Snappy_Supported()) {
+    return;
+  }
+
+  Options options;
+  options.env = &mock_env_;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_,
+                            "BlobFileReaderTest_UncompressionError"),
+      0);
+  options.enable_blob_files = true;
+
+  ImmutableCFOptions immutable_cf_options(options);
+
+  constexpr uint32_t column_family_id = 1;
+  constexpr bool has_ttl = false;
+  constexpr ExpirationRange expiration_range;
+  constexpr uint64_t blob_file_number = 1;
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+
+  uint64_t blob_offset = 0;
+  uint64_t blob_size = 0;
+
+  WriteBlobFile(immutable_cf_options, column_family_id, has_ttl,
+                expiration_range, expiration_range, blob_file_number, key, blob,
+                kSnappyCompression, &blob_offset, &blob_size);
+
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  std::unique_ptr<BlobFileReader> 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) {
+        CacheAllocationPtr* const output =
+            static_cast<CacheAllocationPtr*>(arg);
+        assert(output);
+
+        output->reset();
+      });
+
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  PinnableSlice value;
+  uint64_t bytes_read = 0;
+
+  ASSERT_TRUE(reader
+                  ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
+                            kSnappyCompression, &value, &bytes_read)
+                  .IsCorruption());
+  ASSERT_EQ(bytes_read, 0);
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+class BlobFileReaderIOErrorTest
+    : public testing::Test,
+      public testing::WithParamInterface<std::string> {
+ protected:
+  BlobFileReaderIOErrorTest()
+      : mock_env_(Env::Default()),
+        fault_injection_env_(&mock_env_),
+        sync_point_(GetParam()) {}
+
+  MockEnv mock_env_;
+  FaultInjectionTestEnv fault_injection_env_;
+  std::string sync_point_;
+};
+
+INSTANTIATE_TEST_CASE_P(BlobFileReaderTest, BlobFileReaderIOErrorTest,
+                        ::testing::ValuesIn(std::vector<std::string>{
+                            "BlobFileReader::OpenFile:GetFileSize",
+                            "BlobFileReader::OpenFile:NewRandomAccessFile",
+                            "BlobFileReader::ReadHeader:ReadFromFile",
+                            "BlobFileReader::ReadFooter:ReadFromFile",
+                            "BlobFileReader::GetBlob:ReadFromFile"}));
+
+TEST_P(BlobFileReaderIOErrorTest, IOError) {
+  // Simulates an I/O error during the specified step
+
+  Options options;
+  options.env = &fault_injection_env_;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&fault_injection_env_,
+                            "BlobFileReaderIOErrorTest_IOError"),
+      0);
+  options.enable_blob_files = true;
+
+  ImmutableCFOptions immutable_cf_options(options);
+
+  constexpr uint32_t column_family_id = 1;
+  constexpr bool has_ttl = false;
+  constexpr ExpirationRange expiration_range;
+  constexpr uint64_t blob_file_number = 1;
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+
+  uint64_t blob_offset = 0;
+  uint64_t blob_size = 0;
+
+  WriteBlobFile(immutable_cf_options, column_family_id, has_ttl,
+                expiration_range, expiration_range, blob_file_number, key, blob,
+                kNoCompression, &blob_offset, &blob_size);
+
+  SyncPoint::GetInstance()->SetCallBack(sync_point_, [this](void* /* arg */) {
+    fault_injection_env_.SetFilesystemActive(false,
+                                             Status::IOError(sync_point_));
+  });
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  std::unique_ptr<BlobFileReader> 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");
+
+  if (fail_during_create) {
+    ASSERT_TRUE(s.IsIOError());
+  } else {
+    ASSERT_OK(s);
+
+    PinnableSlice value;
+    uint64_t bytes_read = 0;
+
+    ASSERT_TRUE(reader
+                    ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
+                              kNoCompression, &value, &bytes_read)
+                    .IsIOError());
+    ASSERT_EQ(bytes_read, 0);
+  }
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+class BlobFileReaderDecodingErrorTest
+    : public testing::Test,
+      public testing::WithParamInterface<std::string> {
+ protected:
+  BlobFileReaderDecodingErrorTest()
+      : mock_env_(Env::Default()), sync_point_(GetParam()) {}
+
+  MockEnv mock_env_;
+  std::string sync_point_;
+};
+
+INSTANTIATE_TEST_CASE_P(BlobFileReaderTest, BlobFileReaderDecodingErrorTest,
+                        ::testing::ValuesIn(std::vector<std::string>{
+                            "BlobFileReader::ReadHeader:TamperWithResult",
+                            "BlobFileReader::ReadFooter:TamperWithResult",
+                            "BlobFileReader::GetBlob:TamperWithResult"}));
+
+TEST_P(BlobFileReaderDecodingErrorTest, DecodingError) {
+  Options options;
+  options.env = &mock_env_;
+  options.cf_paths.emplace_back(
+      test::PerThreadDBPath(&mock_env_,
+                            "BlobFileReaderDecodingErrorTest_DecodingError"),
+      0);
+  options.enable_blob_files = true;
+
+  ImmutableCFOptions immutable_cf_options(options);
+
+  constexpr uint32_t column_family_id = 1;
+  constexpr bool has_ttl = false;
+  constexpr ExpirationRange expiration_range;
+  constexpr uint64_t blob_file_number = 1;
+  constexpr char key[] = "key";
+  constexpr char blob[] = "blob";
+
+  uint64_t blob_offset = 0;
+  uint64_t blob_size = 0;
+
+  WriteBlobFile(immutable_cf_options, column_family_id, has_ttl,
+                expiration_range, expiration_range, blob_file_number, key, blob,
+                kNoCompression, &blob_offset, &blob_size);
+
+  SyncPoint::GetInstance()->SetCallBack(sync_point_, [](void* arg) {
+    Slice* const slice = static_cast<Slice*>(arg);
+    assert(slice);
+    assert(!slice->empty());
+
+    slice->remove_prefix(1);
+  });
+
+  SyncPoint::GetInstance()->EnableProcessing();
+
+  constexpr HistogramImpl* blob_file_read_hist = nullptr;
+
+  std::unique_ptr<BlobFileReader> 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";
+
+  if (fail_during_create) {
+    ASSERT_TRUE(s.IsCorruption());
+  } else {
+    ASSERT_OK(s);
+
+    PinnableSlice value;
+    uint64_t bytes_read = 0;
+
+    ASSERT_TRUE(reader
+                    ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
+                              kNoCompression, &value, &bytes_read)
+                    .IsCorruption());
+    ASSERT_EQ(bytes_read, 0);
+  }
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+}  // namespace ROCKSDB_NAMESPACE
+
+int main(int argc, char** argv) {
+  ::testing::InitGoogleTest(&argc, argv);
+  return RUN_ALL_TESTS();
+}

db/blob/blob_index.h

@@ -3,13 +3,13 @@
 //  COPYING file in the root directory) and Apache 2.0 License
 //  (found in the LICENSE.Apache file in the root directory).
 #pragma once
-#ifndef ROCKSDB_LITE
 
 #include <sstream>
 #include <string>
 
-#include "rocksdb/options.h"
+#include "rocksdb/compression_type.h"
 #include "util/coding.h"
+#include "util/compression.h"
 #include "util/string_util.h"
 
 namespace ROCKSDB_NAMESPACE {
@@ -83,6 +83,11 @@ class BlobIndex {
     return size_;
   }
 
+  CompressionType compression() const {
+    assert(!IsInlined());
+    return compression_;
+  }
+
   Status DecodeFrom(Slice slice) {
     static const std::string kErrorMessage = "Error while decoding blob index";
     assert(slice.size() > 0);
@@ -118,7 +123,8 @@ class BlobIndex {
       oss << "[inlined blob] value:" << value_.ToString(output_hex);
     } else {
       oss << "[blob ref] file:" << file_number_ << " offset:" << offset_
-          << " size:" << size_;
+          << " size:" << size_
+          << " compression: " << CompressionTypeToString(compression_);
     }
 
     if (HasTTL()) {
@@ -176,4 +182,3 @@ class BlobIndex {
 };
 
 }  // namespace ROCKSDB_NAMESPACE
-#endif  // ROCKSDB_LITE

db/blob/blob_log_format.cc

@@ -3,15 +3,13 @@
 //  COPYING file in the root directory) and Apache 2.0 License
 //  (found in the LICENSE.Apache file in the root directory).
 //
-#ifndef ROCKSDB_LITE
 
-#include "utilities/blob_db/blob_log_format.h"
+#include "db/blob/blob_log_format.h"
 
 #include "util/coding.h"
 #include "util/crc32c.h"
 
 namespace ROCKSDB_NAMESPACE {
-namespace blob_db {
 
 void BlobLogHeader::EncodeTo(std::string* dst) {
   assert(dst != nullptr);
@@ -144,6 +142,4 @@ Status BlobLogRecord::CheckBlobCRC() const {
   return Status::OK();
 }
 
-}  // namespace blob_db
 }  // namespace ROCKSDB_NAMESPACE
-#endif  // ROCKSDB_LITE

db/blob/blob_log_format.h

@@ -7,9 +7,6 @@
 
 #pragma once
 
-#ifndef ROCKSDB_LITE
-
-#include <limits>
 #include <memory>
 #include <utility>
 
@@ -19,11 +16,9 @@
 #include "rocksdb/types.h"
 
 namespace ROCKSDB_NAMESPACE {
-namespace blob_db {
 
 constexpr uint32_t kMagicNumber = 2395959;  // 0x00248f37
 constexpr uint32_t kVersion1 = 1;
-constexpr uint64_t kNoExpiration = std::numeric_limits<uint64_t>::max();
 
 using ExpirationRange = std::pair<uint64_t, uint64_t>;
 
@@ -109,6 +104,13 @@ struct BlobLogRecord {
   // header include fields up to blob CRC
   static constexpr size_t kHeaderSize = 32;
 
+  // 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) {
+    return key_size + kHeaderSize;
+  }
+
   uint64_t key_size = 0;
   uint64_t value_size = 0;
   uint64_t expiration = 0;
@@ -128,6 +130,19 @@ struct BlobLogRecord {
   Status CheckBlobCRC() const;
 };
 
-}  // namespace blob_db
+// Checks whether a blob offset is potentially valid or not.
+inline bool IsValidBlobOffset(uint64_t value_offset, uint64_t key_size,
+                              uint64_t value_size, uint64_t file_size) {
+  if (value_offset <
+      BlobLogHeader::kSize + BlobLogRecord::kHeaderSize + key_size) {
+    return false;
+  }
+
+  if (value_offset + value_size + BlobLogFooter::kSize > file_size) {
+    return false;
+  }
+
+  return true;
+}
+
 }  // namespace ROCKSDB_NAMESPACE
-#endif  // ROCKSDB_LITE

db/blob/blob_log_sequential_reader.cc

@@ -3,30 +3,33 @@
 //  COPYING file in the root directory) and Apache 2.0 License
 //  (found in the LICENSE.Apache file in the root directory).
 //
-#ifndef ROCKSDB_LITE
 
-#include "utilities/blob_db/blob_log_reader.h"
-
-#include <algorithm>
+#include "db/blob/blob_log_sequential_reader.h"
 
 #include "file/random_access_file_reader.h"
 #include "monitoring/statistics.h"
 #include "util/stop_watch.h"
 
 namespace ROCKSDB_NAMESPACE {
-namespace blob_db {
 
-Reader::Reader(std::unique_ptr<RandomAccessFileReader>&& file_reader, Env* env,
-               Statistics* statistics)
+BlobLogSequentialReader::BlobLogSequentialReader(
+    std::unique_ptr<RandomAccessFileReader>&& file_reader, SystemClock* clock,
+    Statistics* statistics)
     : file_(std::move(file_reader)),
-      env_(env),
+      clock_(clock),
       statistics_(statistics),
-      buffer_(),
       next_byte_(0) {}
 
-Status Reader::ReadSlice(uint64_t size, Slice* slice, char* buf) {
-  StopWatch read_sw(env_, statistics_, BLOB_DB_BLOB_FILE_READ_MICROS);
-  Status s = file_->Read(next_byte_, static_cast<size_t>(size), slice, buf);
+BlobLogSequentialReader::~BlobLogSequentialReader() = default;
+
+Status BlobLogSequentialReader::ReadSlice(uint64_t size, Slice* slice,
+                                          char* buf) {
+  assert(slice);
+  assert(file_);
+
+  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;
   if (!s.ok()) {
     return s;
@@ -38,9 +41,13 @@ Status Reader::ReadSlice(uint64_t size, Slice* slice, char* buf) {
   return s;
 }
 
-Status Reader::ReadHeader(BlobLogHeader* header) {
-  assert(file_.get() != nullptr);
+Status BlobLogSequentialReader::ReadHeader(BlobLogHeader* header) {
+  assert(header);
   assert(next_byte_ == 0);
+
+  static_assert(BlobLogHeader::kSize <= sizeof(header_buf_),
+                "Buffer is smaller than BlobLogHeader::kSize");
+
   Status s = ReadSlice(BlobLogHeader::kSize, &buffer_, header_buf_);
   if (!s.ok()) {
     return s;
@@ -53,8 +60,13 @@ Status Reader::ReadHeader(BlobLogHeader* header) {
   return header->DecodeFrom(buffer_);
 }
 
-Status Reader::ReadRecord(BlobLogRecord* record, ReadLevel level,
-                          uint64_t* blob_offset) {
+Status BlobLogSequentialReader::ReadRecord(BlobLogRecord* record,
+                                           ReadLevel level,
+                                           uint64_t* blob_offset) {
+  assert(record);
+  static_assert(BlobLogRecord::kHeaderSize <= sizeof(header_buf_),
+                "Buffer is smaller than BlobLogRecord::kHeaderSize");
+
   Status s = ReadSlice(BlobLogRecord::kHeaderSize, &buffer_, header_buf_);
   if (!s.ok()) {
     return s;
@@ -100,6 +112,21 @@ Status Reader::ReadRecord(BlobLogRecord* record, ReadLevel level,
   return s;
 }
 
-}  // namespace blob_db
+Status BlobLogSequentialReader::ReadFooter(BlobLogFooter* footer) {
+  assert(footer);
+  static_assert(BlobLogFooter::kSize <= sizeof(header_buf_),
+                "Buffer is smaller than BlobLogFooter::kSize");
+
+  Status s = ReadSlice(BlobLogFooter::kSize, &buffer_, header_buf_);
+  if (!s.ok()) {
+    return s;
+  }
+
+  if (buffer_.size() != BlobLogFooter::kSize) {
+    return Status::Corruption("EOF reached before file footer");
+  }
+
+  return footer->DecodeFrom(buffer_);
+}
+
 }  // namespace ROCKSDB_NAMESPACE
-#endif  // ROCKSDB_LITE

db/blob/blob_log_sequential_reader.h

@@ -0,0 +1,78 @@
+//  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 <memory>
+
+#include "db/blob/blob_log_format.h"
+#include "rocksdb/slice.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class RandomAccessFileReader;
+class Env;
+class Statistics;
+class Status;
+class SystemClock;
+
+/**
+ * BlobLogSequentialReader is a general purpose log stream reader
+ * implementation. The actual job of reading from the device is implemented by
+ * the RandomAccessFileReader interface.
+ *
+ * Please see BlobLogWriter for details on the file and record layout.
+ */
+
+class BlobLogSequentialReader {
+ public:
+  enum ReadLevel {
+    kReadHeader,
+    kReadHeaderKey,
+    kReadHeaderKeyBlob,
+  };
+
+  // Create a reader that will return log records from "*file_reader".
+  BlobLogSequentialReader(std::unique_ptr<RandomAccessFileReader>&& file_reader,
+                          SystemClock* clock, Statistics* statistics);
+
+  // No copying allowed
+  BlobLogSequentialReader(const BlobLogSequentialReader&) = delete;
+  BlobLogSequentialReader& operator=(const BlobLogSequentialReader&) = delete;
+
+  ~BlobLogSequentialReader();
+
+  Status ReadHeader(BlobLogHeader* header);
+
+  // Read the next record into *record.  Returns true if read
+  // successfully, false if we hit end of the input. The contents filled in
+  // *record will only be valid until the next mutating operation on this
+  // reader.
+  // If blob_offset is non-null, return offset of the blob through it.
+  Status ReadRecord(BlobLogRecord* record, ReadLevel level = kReadHeader,
+                    uint64_t* blob_offset = nullptr);
+
+  Status ReadFooter(BlobLogFooter* footer);
+
+  void ResetNextByte() { next_byte_ = 0; }
+
+  uint64_t GetNextByte() const { return next_byte_; }
+
+ private:
+  Status ReadSlice(uint64_t size, Slice* slice, char* buf);
+
+  const std::unique_ptr<RandomAccessFileReader> file_;
+  SystemClock* clock_;
+
+  Statistics* statistics_;
+
+  Slice buffer_;
+  char header_buf_[BlobLogRecord::kHeaderSize];
+
+  // which byte to read next
+  uint64_t next_byte_;
+};
+
+}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_log_writer.cc

@@ -2,44 +2,47 @@
 //  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).
-#ifndef ROCKSDB_LITE
 
-#include "utilities/blob_db/blob_log_writer.h"
+#include "db/blob/blob_log_writer.h"
 
 #include <cstdint>
 #include <string>
 
+#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"
-#include "utilities/blob_db/blob_log_format.h"
 
 namespace ROCKSDB_NAMESPACE {
-namespace blob_db {
 
-Writer::Writer(std::unique_ptr<WritableFileWriter>&& dest, Env* env,
-               Statistics* statistics, uint64_t log_number, uint64_t bpsync,
-               bool use_fs, uint64_t boffset)
+BlobLogWriter::BlobLogWriter(std::unique_ptr<WritableFileWriter>&& dest,
+                             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),
-      bytes_per_sync_(bpsync),
-      next_sync_offset_(0),
       use_fsync_(use_fs),
+      do_flush_(do_flush),
       last_elem_type_(kEtNone) {}
 
-Status Writer::Sync() {
-  StopWatch sync_sw(env_, statistics_, BLOB_DB_BLOB_FILE_SYNC_MICROS);
+BlobLogWriter::~BlobLogWriter() = default;
+
+Status BlobLogWriter::Sync() {
+  TEST_SYNC_POINT("BlobLogWriter::Sync");
+
+  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;
 }
 
-Status Writer::WriteHeader(BlobLogHeader& header) {
+Status BlobLogWriter::WriteHeader(BlobLogHeader& header) {
   assert(block_offset_ == 0);
   assert(last_elem_type_ == kEtNone);
   std::string str;
@@ -48,7 +51,9 @@ Status Writer::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,
@@ -56,7 +61,9 @@ Status Writer::WriteHeader(BlobLogHeader& header) {
   return s;
 }
 
-Status Writer::AppendFooter(BlobLogFooter& footer) {
+Status BlobLogWriter::AppendFooter(BlobLogFooter& footer,
+                                   std::string* checksum_method,
+                                   std::string* checksum_value) {
   assert(block_offset_ != 0);
   assert(last_elem_type_ == kEtFileHdr || last_elem_type_ == kEtRecord);
 
@@ -66,7 +73,34 @@ Status Writer::AppendFooter(BlobLogFooter& footer) {
   Status s = dest_->Append(Slice(str));
   if (s.ok()) {
     block_offset_ += str.size();
-    s = dest_->Close();
+
+    s = Sync();
+
+    if (s.ok()) {
+      s = dest_->Close();
+
+      if (s.ok()) {
+        assert(!!checksum_method == !!checksum_value);
+
+        if (checksum_method) {
+          assert(checksum_method->empty());
+
+          std::string method = dest_->GetFileChecksumFuncName();
+          if (method != kUnknownFileChecksumFuncName) {
+            *checksum_method = std::move(method);
+          }
+        }
+        if (checksum_value) {
+          assert(checksum_value->empty());
+
+          std::string value = dest_->GetFileChecksum();
+          if (value != kUnknownFileChecksum) {
+            *checksum_value = std::move(value);
+          }
+        }
+      }
+    }
+
     dest_.reset();
   }
 
@@ -76,9 +110,9 @@ Status Writer::AppendFooter(BlobLogFooter& footer) {
   return s;
 }
 
-Status Writer::AddRecord(const Slice& key, const Slice& val,
-                         uint64_t expiration, uint64_t* key_offset,
-                         uint64_t* blob_offset) {
+Status BlobLogWriter::AddRecord(const Slice& key, const Slice& val,
+                                uint64_t expiration, uint64_t* key_offset,
+                                uint64_t* blob_offset) {
   assert(block_offset_ != 0);
   assert(last_elem_type_ == kEtFileHdr || last_elem_type_ == kEtRecord);
 
@@ -89,8 +123,8 @@ Status Writer::AddRecord(const Slice& key, const Slice& val,
   return s;
 }
 
-Status Writer::AddRecord(const Slice& key, const Slice& val,
-                         uint64_t* key_offset, uint64_t* blob_offset) {
+Status BlobLogWriter::AddRecord(const Slice& key, const Slice& val,
+                                uint64_t* key_offset, uint64_t* blob_offset) {
   assert(block_offset_ != 0);
   assert(last_elem_type_ == kEtFileHdr || last_elem_type_ == kEtRecord);
 
@@ -101,8 +135,8 @@ Status Writer::AddRecord(const Slice& key, const Slice& val,
   return s;
 }
 
-void Writer::ConstructBlobHeader(std::string* buf, const Slice& key,
-                                 const Slice& val, uint64_t expiration) {
+void BlobLogWriter::ConstructBlobHeader(std::string* buf, const Slice& key,
+                                        const Slice& val, uint64_t expiration) {
   BlobLogRecord record;
   record.key = key;
   record.value = val;
@@ -110,10 +144,11 @@ void Writer::ConstructBlobHeader(std::string* buf, const Slice& key,
   record.EncodeHeaderTo(buf);
 }
 
-Status Writer::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);
+Status BlobLogWriter::EmitPhysicalRecord(const std::string& headerbuf,
+                                         const Slice& key, const Slice& val,
+                                         uint64_t* key_offset,
+                                         uint64_t* blob_offset) {
+  StopWatch write_sw(clock_, statistics_, BLOB_DB_BLOB_FILE_WRITE_MICROS);
   Status s = dest_->Append(Slice(headerbuf));
   if (s.ok()) {
     s = dest_->Append(key);
@@ -121,7 +156,7 @@ Status Writer::EmitPhysicalRecord(const std::string& headerbuf,
   if (s.ok()) {
     s = dest_->Append(val);
   }
-  if (s.ok()) {
+  if (do_flush_ && s.ok()) {
     s = dest_->Flush();
   }
 
@@ -134,6 +169,4 @@ Status Writer::EmitPhysicalRecord(const std::string& headerbuf,
   return s;
 }
 
-}  // namespace blob_db
 }  // namespace ROCKSDB_NAMESPACE
-#endif  // ROCKSDB_LITE

db/blob/blob_log_writer.h

@@ -4,46 +4,41 @@
 //  (found in the LICENSE.Apache file in the root directory).
 #pragma once
 
-#ifndef ROCKSDB_LITE
-
 #include <cstdint>
 #include <memory>
 #include <string>
 
-#include "rocksdb/env.h"
+#include "db/blob/blob_log_format.h"
 #include "rocksdb/slice.h"
 #include "rocksdb/statistics.h"
 #include "rocksdb/status.h"
 #include "rocksdb/types.h"
-#include "utilities/blob_db/blob_log_format.h"
 
 namespace ROCKSDB_NAMESPACE {
 
 class WritableFileWriter;
-
-namespace blob_db {
-
+class SystemClock;
 /**
- * Writer is the blob log stream writer. It provides an append-only
+ * BlobLogWriter is the blob log stream writer. It provides an append-only
  * abstraction for writing blob data.
  *
  *
  * Look at blob_db_format.h to see the details of the record formats.
  */
 
-class Writer {
+class BlobLogWriter {
  public:
   // Create a writer that will append data to "*dest".
   // "*dest" must be initially empty.
-  // "*dest" must remain live while this Writer is in use.
-  Writer(std::unique_ptr<WritableFileWriter>&& dest, Env* env,
-         Statistics* statistics, uint64_t log_number, uint64_t bpsync,
-         bool use_fsync, uint64_t boffset = 0);
+  // "*dest" must remain live while this BlobLogWriter is in use.
+  BlobLogWriter(std::unique_ptr<WritableFileWriter>&& dest, SystemClock* clock,
+                Statistics* statistics, uint64_t log_number, bool use_fsync,
+                bool do_flush, uint64_t boffset = 0);
   // No copying allowed
-  Writer(const Writer&) = delete;
-  Writer& operator=(const Writer&) = delete;
+  BlobLogWriter(const BlobLogWriter&) = delete;
+  BlobLogWriter& operator=(const BlobLogWriter&) = delete;
 
-  ~Writer() = default;
+  ~BlobLogWriter();
 
   static void ConstructBlobHeader(std::string* buf, const Slice& key,
                                   const Slice& val, uint64_t expiration);
@@ -58,7 +53,8 @@ class Writer {
                             const Slice& val, uint64_t* key_offset,
                             uint64_t* blob_offset);
 
-  Status AppendFooter(BlobLogFooter& footer);
+  Status AppendFooter(BlobLogFooter& footer, std::string* checksum_method,
+                      std::string* checksum_value);
 
   Status WriteHeader(BlobLogHeader& header);
 
@@ -68,27 +64,20 @@ class Writer {
 
   uint64_t get_log_number() const { return log_number_; }
 
-  bool ShouldSync() const { return block_offset_ > next_sync_offset_; }
-
   Status Sync();
 
-  void ResetSyncPointer() { next_sync_offset_ += bytes_per_sync_; }
-
  private:
   std::unique_ptr<WritableFileWriter> dest_;
-  Env* env_;
+  SystemClock* clock_;
   Statistics* statistics_;
   uint64_t log_number_;
   uint64_t block_offset_;  // Current offset in block
-  uint64_t bytes_per_sync_;
-  uint64_t next_sync_offset_;
   bool use_fsync_;
+  bool do_flush_;
 
  public:
   enum ElemType { kEtNone, kEtFileHdr, kEtRecord, kEtFileFooter };
   ElemType last_elem_type_;
 };
 
-}  // namespace blob_db
 }  // namespace ROCKSDB_NAMESPACE
-#endif  // ROCKSDB_LITE

db/blob/db_blob_basic_test.cc

@@ -0,0 +1,417 @@
+//  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
+
+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

@@ -41,12 +41,13 @@ class DBBlobIndexTest : public DBTestBase {
                                        Tier::kImmutableMemtables,
                                        Tier::kL0SstFile, Tier::kLnSstFile};
 
-  DBBlobIndexTest() : DBTestBase("/db_blob_index_test") {}
+  DBBlobIndexTest()
+      : DBTestBase("/db_blob_index_test", /*env_do_fsync=*/true) {}
 
   ColumnFamilyHandle* cfh() { return dbfull()->DefaultColumnFamily(); }
 
   ColumnFamilyData* cfd() {
-    return reinterpret_cast<ColumnFamilyHandleImpl*>(cfh())->cfd();
+    return static_cast_with_check<ColumnFamilyHandleImpl>(cfh())->cfd();
   }
 
   Status PutBlobIndex(WriteBatch* batch, const Slice& key,
@@ -72,6 +73,9 @@ class DBBlobIndexTest : public DBTestBase {
     if (s.IsNotFound()) {
       return "NOT_FOUND";
     }
+    if (s.IsCorruption()) {
+      return "CORRUPTION";
+    }
     if (s.IsNotSupported()) {
       return "NOT_SUPPORTED";
     }
@@ -94,11 +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;
@@ -152,8 +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());
@@ -170,15 +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());
@@ -205,7 +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"));
@@ -217,8 +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},
@@ -281,6 +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);
@@ -294,6 +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) {
@@ -303,6 +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);
@@ -315,6 +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) {
@@ -352,7 +380,7 @@ TEST_F(DBBlobIndexTest, Iterate) {
             ASSERT_OK(Write(&batch));
             break;
           default:
-            assert(false);
+            FAIL();
         };
       }
       snapshots.push_back(dbfull()->GetSnapshot());
@@ -363,8 +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

@@ -13,19 +13,23 @@
 #include <deque>
 #include <vector>
 
+#include "db/blob/blob_file_builder.h"
 #include "db/compaction/compaction_iterator.h"
 #include "db/dbformat.h"
 #include "db/event_helpers.h"
 #include "db/internal_stats.h"
 #include "db/merge_helper.h"
+#include "db/output_validator.h"
 #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"
@@ -48,48 +52,58 @@ 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 uint64_t file_creation_time, const std::string& db_id,
+    const std::string& db_session_id) {
   assert((column_family_id ==
           TablePropertiesCollectorFactory::Context::kUnknownColumnFamily) ==
          column_family_name.empty());
   return ioptions.table_factory->NewTableBuilder(
       TableBuilderOptions(ioptions, moptions, internal_comparator,
                           int_tbl_prop_collector_factories, compression_type,
-                          sample_for_compression, compression_opts,
-                          skip_filters, column_family_name, level,
-                          creation_time, oldest_key_time, target_file_size,
-                          file_creation_time),
+                          compression_opts, 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, 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>>
         range_del_iters,
-    FileMetaData* meta, const InternalKeyComparator& internal_comparator,
+    FileMetaData* meta, std::vector<BlobFileAddition>* blob_file_additions,
+    const InternalKeyComparator& internal_comparator,
     const std::vector<std::unique_ptr<IntTblPropCollectorFactory>>*
         int_tbl_prop_collector_factories,
     uint32_t column_family_id, const std::string& column_family_name,
     std::vector<SequenceNumber> snapshots,
     SequenceNumber earliest_write_conflict_snapshot,
     SnapshotChecker* snapshot_checker, const CompressionType compression,
-    uint64_t sample_for_compression, const CompressionOptions& compression_opts,
-    bool paranoid_file_checks, InternalStats* internal_stats,
-    TableFileCreationReason reason, EventLogger* event_logger, int job_id,
-    const Env::IOPriority io_priority, TableProperties* table_properties,
-    int level, const uint64_t creation_time, const uint64_t oldest_key_time,
-    Env::WriteLifeTimeHint write_hint, const uint64_t file_creation_time) {
+    const CompressionOptions& compression_opts, bool paranoid_file_checks,
+    InternalStats* internal_stats, TableFileCreationReason reason,
+    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* full_history_ts_low,
+    BlobFileCompletionCallback* blob_callback) {
   assert((column_family_id ==
           TablePropertiesCollectorFactory::Context::kUnknownColumnFamily) ==
          column_family_name.empty());
   // Reports the IOStats for flush for every following bytes.
   const size_t kReportFlushIOStatsEvery = 1048576;
+  OutputValidator output_validator(
+      internal_comparator,
+      /*enable_order_check=*/
+      mutable_cf_options.check_flush_compaction_key_order,
+      /*enable_hash=*/paranoid_file_checks);
   Status s;
   meta->fd.file_size = 0;
   iter->SeekToFirst();
@@ -101,47 +115,56 @@ Status BuildTable(
 
   std::string fname = TableFileName(ioptions.cf_paths, meta->fd.GetNumber(),
                                     meta->fd.GetPathId());
+  std::vector<std::string> blob_file_paths;
+  std::string file_checksum = kUnknownFileChecksum;
+  std::string file_checksum_func_name = kUnknownFileChecksumFuncName;
 #ifndef ROCKSDB_LITE
   EventHelpers::NotifyTableFileCreationStarted(
       ioptions.listeners, dbname, column_family_name, fname, job_id, reason);
 #endif  // !ROCKSDB_LITE
-  TableProperties tp;
+  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
       bool use_direct_writes = file_options.use_direct_writes;
       TEST_SYNC_POINT_CALLBACK("BuildTable:create_file", &use_direct_writes);
 #endif  // !NDEBUG
-      s = NewWritableFile(fs, fname, &file, file_options);
+      IOStatus io_s = NewWritableFile(fs, fname, &file, file_options);
+      assert(s.ok());
+      s = io_s;
+      if (io_status->ok()) {
+        *io_status = io_s;
+      }
       if (!s.ok()) {
         EventHelpers::LogAndNotifyTableFileCreationFinished(
             event_logger, ioptions.listeners, dbname, column_family_name, fname,
-            job_id, meta->fd, kInvalidBlobFileNumber, tp, reason, s);
+            job_id, meta->fd, kInvalidBlobFileNumber, tp, reason, s,
+            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, ioptions.statistics,
-          ioptions.listeners, ioptions.sst_file_checksum_func));
+          std::move(file), fname, file_options, ioptions.clock, io_tracer,
+          ioptions.statistics, ioptions.listeners,
+          ioptions.file_checksum_gen_factory,
+          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,
-          0 /*target_file_size*/, file_creation_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(),
@@ -150,16 +173,36 @@ Status BuildTable(
                       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, 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);
+
     CompactionIterator c_iter(
         iter, internal_comparator.user_comparator(), &merge, kMaxSequenceNumber,
         &snapshots, earliest_write_conflict_snapshot, snapshot_checker, env,
         ShouldReportDetailedTime(env, ioptions.statistics),
-        true /* internal key corruption is not ok */, range_del_agg.get());
+        true /* internal key corruption is not ok */, range_del_agg.get(),
+        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()) {
       const Slice& key = c_iter.key();
       const Slice& value = c_iter.value();
       const ParsedInternalKey& ikey = c_iter.ikey();
+      // Generate a rolling 64-bit hash of the key and values
+      s = output_validator.Add(key, value);
+      if (!s.ok()) {
+        break;
+      }
       builder->Add(key, value);
       meta->UpdateBoundaries(key, value, ikey.sequence, ikey.type);
 
@@ -170,26 +213,34 @@ Status BuildTable(
             ThreadStatus::FLUSH_BYTES_WRITTEN, IOSTATS(bytes_written));
       }
     }
-
-    auto range_del_it = range_del_agg->NewIterator();
-    for (range_del_it->SeekToFirst(); range_del_it->Valid();
-         range_del_it->Next()) {
-      auto tombstone = range_del_it->Tombstone();
-      auto kv = tombstone.Serialize();
-      builder->Add(kv.first.Encode(), kv.second);
-      meta->UpdateBoundariesForRange(kv.first, tombstone.SerializeEndKey(),
-                                     tombstone.seq_, internal_comparator);
+    if (!s.ok()) {
+      c_iter.status().PermitUncheckedError();
+    } else if (!c_iter.status().ok()) {
+      s = c_iter.status();
     }
 
-    // Finish and check for builder errors
-    tp = builder->GetTableProperties();
-    bool empty = builder->NumEntries() == 0 && tp.num_range_deletions == 0;
-    s = c_iter.status();
+    if (s.ok()) {
+      auto range_del_it = range_del_agg->NewIterator();
+      for (range_del_it->SeekToFirst(); range_del_it->Valid();
+           range_del_it->Next()) {
+        auto tombstone = range_del_it->Tombstone();
+        auto kv = tombstone.Serialize();
+        builder->Add(kv.first.Encode(), kv.second);
+        meta->UpdateBoundariesForRange(kv.first, tombstone.SerializeEndKey(),
+                                       tombstone.seq_, internal_comparator);
+      }
+    }
+
+    TEST_SYNC_POINT("BuildTable:BeforeFinishBuildTable");
+    const bool empty = builder->IsEmpty();
     if (!s.ok() || empty) {
       builder->Abandon();
     } else {
       s = builder->Finish();
     }
+    if (io_status->ok()) {
+      *io_status = builder->io_status();
+    }
 
     if (s.ok() && !empty) {
       uint64_t file_size = builder->FileSize();
@@ -200,20 +251,41 @@ Status BuildTable(
       if (table_properties) {
         *table_properties = tp;
       }
-      // Add the checksum information to file metadata.
-      meta->file_checksum = builder->GetFileChecksum();
-      meta->file_checksum_func_name = builder->GetFileChecksumFuncName();
     }
     delete builder;
 
     // Finish and check for file errors
+    TEST_SYNC_POINT("BuildTable:BeforeSyncTable");
     if (s.ok() && !empty) {
-      StopWatch sw(env, ioptions.statistics, TABLE_SYNC_MICROS);
-      s = file_writer->Sync(ioptions.use_fsync);
+      StopWatch sw(ioptions.clock, ioptions.statistics, TABLE_SYNC_MICROS);
+      *io_status = file_writer->Sync(ioptions.use_fsync);
     }
-    if (s.ok() && !empty) {
-      s = file_writer->Close();
+    TEST_SYNC_POINT("BuildTable:BeforeCloseTableFile");
+    if (s.ok() && io_status->ok() && !empty) {
+      *io_status = file_writer->Close();
     }
+    if (s.ok() && io_status->ok() && !empty) {
+      // Add the checksum information to file metadata.
+      meta->file_checksum = file_writer->GetFileChecksum();
+      meta->file_checksum_func_name = file_writer->GetFileChecksumFuncName();
+      file_checksum = meta->file_checksum;
+      file_checksum_func_name = meta->file_checksum_func_name;
+    }
+
+    if (s.ok()) {
+      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) {
       // Verify that the table is usable
@@ -222,20 +294,32 @@ Status BuildTable(
       // No matter whether use_direct_io_for_flush_and_compaction is true,
       // we will regrad this verification as user reads since the goal is
       // to cache it here for further user reads
+      ReadOptions read_options;
       std::unique_ptr<InternalIterator> it(table_cache->NewIterator(
-          ReadOptions(), file_options, internal_comparator, *meta,
+          read_options, file_options, internal_comparator, *meta,
           nullptr /* range_del_agg */,
           mutable_cf_options.prefix_extractor.get(), nullptr,
           (internal_stats == nullptr) ? nullptr
                                       : internal_stats->GetFileReadHist(0),
           TableReaderCaller::kFlush, /*arena=*/nullptr,
-          /*skip_filter=*/false, level, /*smallest_compaction_key=*/nullptr,
-          /*largest_compaction_key*/ nullptr));
+          /*skip_filter=*/false, level,
+          MaxFileSizeForL0MetaPin(mutable_cf_options),
+          /*smallest_compaction_key=*/nullptr,
+          /*largest_compaction_key*/ nullptr,
+          /*allow_unprepared_value*/ false));
       s = it->status();
       if (s.ok() && paranoid_file_checks) {
+        OutputValidator file_validator(internal_comparator,
+                                       /*enable_order_check=*/true,
+                                       /*enable_hash=*/true);
         for (it->SeekToFirst(); it->Valid(); it->Next()) {
+          // Generate a rolling 64-bit hash of the key and values
+          file_validator.Add(it->key(), it->value()).PermitUncheckedError();
         }
         s = it->status();
+        if (s.ok() && !output_validator.CompareValidator(file_validator)) {
+          s = Status::Corruption("Paranoid checksums do not match");
+        }
       }
     }
   }
@@ -246,7 +330,23 @@ Status BuildTable(
   }
 
   if (!s.ok() || meta->fd.GetFileSize() == 0) {
-    fs->DeleteFile(fname, IOOptions(), nullptr);
+    TEST_SYNC_POINT("BuildTable:BeforeDeleteFile");
+
+    constexpr IODebugContext* dbg = nullptr;
+
+    Status ignored = fs->DeleteFile(fname, IOOptions(), dbg);
+    ignored.PermitUncheckedError();
+
+    assert(blob_file_additions || blob_file_paths.empty());
+
+    if (blob_file_additions) {
+      for (const std::string& blob_file_path : blob_file_paths) {
+        ignored = DeleteDBFile(&db_options, blob_file_path, dbname,
+                               /*force_bg=*/false, /*force_fg=*/false);
+        ignored.PermitUncheckedError();
+        TEST_SYNC_POINT("BuildTable::AfterDeleteFile");
+      }
+    }
   }
 
   if (meta->fd.GetFileSize() == 0) {
@@ -255,7 +355,8 @@ Status BuildTable(
   // Output to event logger and fire events.
   EventHelpers::LogAndNotifyTableFileCreationFinished(
       event_logger, ioptions.listeners, dbname, column_family_name, fname,
-      job_id, meta->fd, meta->oldest_blob_file_number, tp, reason, s);
+      job_id, meta->fd, meta->oldest_blob_file_number, tp, reason, s,
+      file_checksum, file_checksum_func_name);
 
   return s;
 }

db/builder.h

@@ -27,8 +27,10 @@ namespace ROCKSDB_NAMESPACE {
 struct Options;
 struct FileMetaData;
 
+class VersionSet;
 class Env;
 struct EnvOptions;
+class BlobFileAddition;
 class Iterator;
 class SnapshotChecker;
 class TableCache;
@@ -36,6 +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
@@ -47,11 +50,11 @@ 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,
-    const uint64_t file_creation_time = 0);
+    const uint64_t file_creation_time = 0, const std::string& db_id = "",
+    const std::string& db_session_id = "");
 
 // Build a Table file from the contents of *iter.  The generated file
 // will be named according to number specified in meta. On success, the rest of
@@ -62,27 +65,31 @@ 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, 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>>
         range_del_iters,
-    FileMetaData* meta, const InternalKeyComparator& internal_comparator,
+    FileMetaData* meta, std::vector<BlobFileAddition>* blob_file_additions,
+    const InternalKeyComparator& internal_comparator,
     const std::vector<std::unique_ptr<IntTblPropCollectorFactory>>*
         int_tbl_prop_collector_factories,
     uint32_t column_family_id, const std::string& column_family_name,
     std::vector<SequenceNumber> snapshots,
     SequenceNumber earliest_write_conflict_snapshot,
     SnapshotChecker* snapshot_checker, const CompressionType compression,
-    const uint64_t sample_for_compression,
     const CompressionOptions& compression_opts, bool paranoid_file_checks,
     InternalStats* internal_stats, TableFileCreationReason reason,
+    IOStatus* io_status, const std::shared_ptr<IOTracer>& io_tracer,
     EventLogger* event_logger = nullptr, int job_id = 0,
     const Env::IOPriority io_priority = Env::IO_HIGH,
     TableProperties* table_properties = nullptr, int level = -1,
     const uint64_t creation_time = 0, const uint64_t oldest_key_time = 0,
     Env::WriteLifeTimeHint write_hint = Env::WLTH_NOT_SET,
-    const uint64_t file_creation_time = 0);
+    const uint64_t file_creation_time = 0, const std::string& db_id = "",
+    const std::string& db_session_id = "",
+    const std::string* full_history_ts_low = nullptr,
+    BlobFileCompletionCallback* blob_callback = nullptr);
 
 }  // namespace ROCKSDB_NAMESPACE

db/column_family.cc

@@ -12,9 +12,11 @@
 #include <algorithm>
 #include <cinttypes>
 #include <limits>
+#include <sstream>
 #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"
@@ -31,9 +33,11 @@
 #include "monitoring/thread_status_util.h"
 #include "options/options_helper.h"
 #include "port/port.h"
-#include "table/block_based/block_based_table_factory.h"
+#include "rocksdb/convenience.h"
+#include "rocksdb/table.h"
 #include "table/merging_iterator.h"
 #include "util/autovector.h"
+#include "util/cast_util.h"
 #include "util/compression.h"
 
 namespace ROCKSDB_NAMESPACE {
@@ -147,6 +151,16 @@ Status CheckCompressionSupported(const ColumnFamilyOptions& cf_options) {
           "should be nonzero if we're using zstd's dictionary generator.");
     }
   }
+
+  if (!CompressionTypeSupported(cf_options.blob_compression_type)) {
+    std::ostringstream oss;
+    oss << "The specified blob compression type "
+        << CompressionTypeToString(cf_options.blob_compression_type)
+        << " is not available.";
+
+    return Status::InvalidArgument(oss.str());
+  }
+
   return Status::OK();
 }
 
@@ -322,7 +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
 
@@ -345,8 +361,8 @@ ColumnFamilyOptions SanitizeOptions(const ImmutableDBOptions& db_options,
     result.max_compaction_bytes = result.target_file_size_base * 25;
   }
 
-  bool is_block_based_table =
-      (result.table_factory->Name() == BlockBasedTableFactory().Name());
+  bool is_block_based_table = (result.table_factory->IsInstanceOf(
+      TableFactory::kBlockBasedTableName()));
 
   const uint64_t kAdjustedTtl = 30 * 24 * 60 * 60;
   if (result.ttl == kDefaultTtl) {
@@ -436,9 +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,
@@ -471,12 +485,24 @@ void SuperVersionUnrefHandle(void* ptr) {
 }
 }  // anonymous namespace
 
+std::vector<std::string> ColumnFamilyData::GetDbPaths() const {
+  std::vector<std::string> paths;
+  paths.reserve(ioptions_.cf_paths.size());
+  for (const DbPath& db_path : ioptions_.cf_paths) {
+    paths.emplace_back(db_path.path);
+  }
+  return paths;
+}
+
+const uint32_t ColumnFamilyData::kDummyColumnFamilyDataId = port::kMaxUint32;
+
 ColumnFamilyData::ColumnFamilyData(
     uint32_t id, const std::string& name, Version* _dummy_versions,
     Cache* _table_cache, WriteBufferManager* write_buffer_manager,
     const ColumnFamilyOptions& cf_options, const ImmutableDBOptions& db_options,
     const FileOptions& file_options, ColumnFamilySet* column_family_set,
-    BlockCacheTracer* const block_cache_tracer)
+    BlockCacheTracer* const block_cache_tracer,
+    const std::shared_ptr<IOTracer>& io_tracer)
     : id_(id),
       name_(name),
       dummy_versions_(_dummy_versions),
@@ -507,7 +533,23 @@ ColumnFamilyData::ColumnFamilyData(
       queued_for_compaction_(false),
       prev_compaction_needed_bytes_(0),
       allow_2pc_(db_options.allow_2pc),
-      last_memtable_id_(0) {
+      last_memtable_id_(0),
+      db_paths_registered_(false) {
+  if (id_ != kDummyColumnFamilyDataId) {
+    // TODO(cc): RegisterDbPaths can be expensive, considering moving it
+    // outside of this constructor which might be called with db mutex held.
+    // TODO(cc): considering using ioptions_.fs, currently some tests rely on
+    // EnvWrapper, that's the main reason why we use env here.
+    Status s = ioptions_.env->RegisterDbPaths(GetDbPaths());
+    if (s.ok()) {
+      db_paths_registered_ = true;
+    } else {
+      ROCKS_LOG_ERROR(
+          ioptions_.info_log,
+          "Failed to register data paths of column family (id: %d, name: %s)",
+          id_, name_.c_str());
+    }
+  }
   Ref();
 
   // Convert user defined table properties collector factories to internal ones.
@@ -516,9 +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));
+                                      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_));
@@ -587,7 +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);
@@ -601,9 +647,21 @@ ColumnFamilyData::~ColumnFamilyData() {
   for (MemTable* m : to_delete) {
     delete m;
   }
+
+  if (db_paths_registered_) {
+    // TODO(cc): considering using ioptions_.fs, currently some tests rely on
+    // EnvWrapper, that's the main reason why we use env here.
+    Status s = ioptions_.env->UnregisterDbPaths(GetDbPaths());
+    if (!s.ok()) {
+      ROCKS_LOG_ERROR(
+          ioptions_.info_log,
+          "Failed to unregister data paths of column family (id: %d, name: %s)",
+          id_, name_.c_str());
+    }
+  }
 }
 
-bool ColumnFamilyData::UnrefAndTryDelete() {
+bool ColumnFamilyData::UnrefAndTryDelete(SuperVersion* sv_under_cleanup) {
   int old_refs = refs_.fetch_sub(1);
   assert(old_refs > 0);
 
@@ -613,7 +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;
@@ -651,9 +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) {
@@ -775,7 +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 &&
@@ -789,7 +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 &&
@@ -817,7 +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;
 
@@ -1003,17 +1067,19 @@ void ColumnFamilyData::CreateNewMemtable(
 }
 
 bool ColumnFamilyData::NeedsCompaction() const {
-  return compaction_picker_->NeedsCompaction(current_->storage_info());
+  return !mutable_cf_options_.disable_auto_compactions &&
+         compaction_picker_->NeedsCompaction(current_->storage_info());
 }
 
 Compaction* ColumnFamilyData::PickCompaction(
-    const MutableCFOptions& mutable_options, LogBuffer* log_buffer) {
+    const MutableCFOptions& mutable_options,
+    const MutableDBOptions& mutable_db_options, LogBuffer* log_buffer) {
   SequenceNumber earliest_mem_seqno =
       std::min(mem_->GetEarliestSequenceNumber(),
                imm_.current()->GetEarliestSequenceNumber(false));
   auto* result = compaction_picker_->PickCompaction(
-      GetName(), mutable_options, current_->storage_info(), log_buffer,
-      earliest_mem_seqno);
+      GetName(), mutable_options, mutable_db_options, current_->storage_info(),
+      log_buffer, earliest_mem_seqno);
   if (result != nullptr) {
     result->SetInputVersion(current_);
   }
@@ -1029,7 +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
@@ -1048,10 +1114,12 @@ Status ColumnFamilyData::RangesOverlapWithMemtables(
       super_version->mem->NewRangeTombstoneIterator(read_opts, read_seq);
   range_del_agg.AddTombstones(
       std::unique_ptr<FragmentedRangeTombstoneIterator>(active_range_del_iter));
-  super_version->imm->AddRangeTombstoneIterators(read_opts, nullptr /* arena */,
-                                                 &range_del_agg);
-
   Status status;
+  status = super_version->imm->AddRangeTombstoneIterators(
+      read_opts, nullptr /* arena */, &range_del_agg);
+  // AddRangeTombstoneIterators always return Status::OK.
+  assert(status.ok());
+
   for (size_t i = 0; i < ranges.size() && status.ok() && !*overlap; ++i) {
     auto* vstorage = super_version->current->storage_info();
     auto* ucmp = vstorage->InternalComparator()->user_comparator();
@@ -1060,12 +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 = 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) {
@@ -1083,14 +1151,16 @@ const int ColumnFamilyData::kCompactAllLevels = -1;
 const int ColumnFamilyData::kCompactToBaseLevel = -2;
 
 Compaction* ColumnFamilyData::CompactRange(
-    const MutableCFOptions& mutable_cf_options, int input_level,
+    const MutableCFOptions& mutable_cf_options,
+    const MutableDBOptions& mutable_db_options, int input_level,
     int output_level, const CompactRangeOptions& compact_range_options,
     const InternalKey* begin, const InternalKey* end,
     InternalKey** compaction_end, bool* conflict,
     uint64_t max_file_num_to_ignore) {
   auto* result = compaction_picker_->CompactRange(
-      GetName(), mutable_cf_options, current_->storage_info(), input_level,
-      output_level, compact_range_options, begin, end, compaction_end, conflict,
+      GetName(), mutable_cf_options, mutable_db_options,
+      current_->storage_info(), input_level, output_level,
+      compact_range_options, begin, end, compaction_end, conflict,
       max_file_num_to_ignore);
   if (result != nullptr) {
     result->SetInputVersion(current_);
@@ -1260,7 +1330,8 @@ Status ColumnFamilyData::ValidateOptions(
   }
 
   if (cf_options.ttl > 0 && cf_options.ttl != kDefaultTtl) {
-    if (cf_options.table_factory->Name() != BlockBasedTableFactory().Name()) {
+    if (!cf_options.table_factory->IsInstanceOf(
+            TableFactory::kBlockBasedTableName())) {
       return Status::NotSupported(
           "TTL is only supported in Block-Based Table format. ");
     }
@@ -1268,30 +1339,40 @@ Status ColumnFamilyData::ValidateOptions(
 
   if (cf_options.periodic_compaction_seconds > 0 &&
       cf_options.periodic_compaction_seconds != kDefaultPeriodicCompSecs) {
-    if (cf_options.table_factory->Name() != BlockBasedTableFactory().Name()) {
+    if (!cf_options.table_factory->IsInstanceOf(
+            TableFactory::kBlockBasedTableName())) {
       return Status::NotSupported(
           "Periodic Compaction is only supported in "
           "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].");
+  }
+
   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;
@@ -1321,7 +1402,7 @@ Env::WriteLifeTimeHint ColumnFamilyData::CalculateSSTWriteHint(int level) {
 }
 
 Status ColumnFamilyData::AddDirectories(
-    std::map<std::string, std::shared_ptr<Directory>>* created_dirs) {
+    std::map<std::string, std::shared_ptr<FSDirectory>>* created_dirs) {
   Status s;
   assert(created_dirs != nullptr);
   assert(data_dirs_.empty());
@@ -1329,8 +1410,8 @@ Status ColumnFamilyData::AddDirectories(
     auto existing_dir = created_dirs->find(p.path);
 
     if (existing_dir == created_dirs->end()) {
-      std::unique_ptr<Directory> path_directory;
-      s = DBImpl::CreateAndNewDirectory(ioptions_.env, p.path, &path_directory);
+      std::unique_ptr<FSDirectory> path_directory;
+      s = DBImpl::CreateAndNewDirectory(ioptions_.fs, p.path, &path_directory);
       if (!s.ok()) {
         return s;
       }
@@ -1345,7 +1426,7 @@ Status ColumnFamilyData::AddDirectories(
   return s;
 }
 
-Directory* ColumnFamilyData::GetDataDir(size_t path_id) const {
+FSDirectory* ColumnFamilyData::GetDataDir(size_t path_id) const {
   if (data_dirs_.empty()) {
     return nullptr;
   }
@@ -1358,21 +1439,24 @@ ColumnFamilySet::ColumnFamilySet(const std::string& dbname,
                                  const ImmutableDBOptions* db_options,
                                  const FileOptions& file_options,
                                  Cache* table_cache,
-                                 WriteBufferManager* write_buffer_manager,
-                                 WriteController* write_controller,
-                                 BlockCacheTracer* const block_cache_tracer)
+                                 WriteBufferManager* _write_buffer_manager,
+                                 WriteController* _write_controller,
+                                 BlockCacheTracer* const block_cache_tracer,
+                                 const std::shared_ptr<IOTracer>& io_tracer)
     : max_column_family_(0),
       dummy_cfd_(new ColumnFamilyData(
-          0, "", nullptr, nullptr, nullptr, ColumnFamilyOptions(), *db_options,
-          file_options, nullptr, block_cache_tracer)),
+          ColumnFamilyData::kDummyColumnFamilyDataId, "", nullptr, nullptr,
+          nullptr, ColumnFamilyOptions(), *db_options, file_options, nullptr,
+          block_cache_tracer, io_tracer)),
       default_cfd_cache_(nullptr),
       db_name_(dbname),
       db_options_(db_options),
       file_options_(file_options),
       table_cache_(table_cache),
-      write_buffer_manager_(write_buffer_manager),
-      write_controller_(write_controller),
-      block_cache_tracer_(block_cache_tracer) {
+      write_buffer_manager_(_write_buffer_manager),
+      write_controller_(_write_controller),
+      block_cache_tracer_(block_cache_tracer),
+      io_tracer_(io_tracer) {
   // initialize linked list
   dummy_cfd_->prev_ = dummy_cfd_;
   dummy_cfd_->next_ = dummy_cfd_;
@@ -1438,7 +1522,7 @@ ColumnFamilyData* ColumnFamilySet::CreateColumnFamily(
   assert(column_families_.find(name) == column_families_.end());
   ColumnFamilyData* new_cfd = new ColumnFamilyData(
       id, name, dummy_versions, table_cache_, write_buffer_manager_, options,
-      *db_options_, file_options_, this, block_cache_tracer_);
+      *db_options_, file_options_, this, block_cache_tracer_, io_tracer_);
   column_families_.insert({name, id});
   column_family_data_.insert({id, new_cfd});
   max_column_family_ = std::max(max_column_family_, id);
@@ -1506,7 +1590,7 @@ ColumnFamilyHandle* ColumnFamilyMemTablesImpl::GetColumnFamilyHandle() {
 uint32_t GetColumnFamilyID(ColumnFamilyHandle* column_family) {
   uint32_t column_family_id = 0;
   if (column_family != nullptr) {
-    auto cfh = reinterpret_cast<ColumnFamilyHandleImpl*>(column_family);
+    auto cfh = static_cast_with_check<ColumnFamilyHandleImpl>(column_family);
     column_family_id = cfh->GetID();
   }
   return column_family_id;

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);
@@ -381,12 +377,14 @@ 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
   bool NeedsCompaction() const;
   // REQUIRES: DB mutex held
   Compaction* PickCompaction(const MutableCFOptions& mutable_options,
+                             const MutableDBOptions& mutable_db_options,
                              LogBuffer* log_buffer);
 
   // Check if the passed range overlap with any running compactions.
@@ -403,7 +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.
@@ -412,6 +411,7 @@ class ColumnFamilyData {
   static const int kCompactToBaseLevel;
   // REQUIRES: DB mutex held
   Compaction* CompactRange(const MutableCFOptions& mutable_cf_options,
+                           const MutableDBOptions& mutable_db_options,
                            int input_level, int output_level,
                            const CompactRangeOptions& compact_range_options,
                            const InternalKey* begin, const InternalKey* end,
@@ -441,7 +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);
@@ -475,9 +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
@@ -500,14 +502,30 @@ class ColumnFamilyData {
   // created_dirs remembers directory created, so that we don't need to call
   // the same data creation operation again.
   Status AddDirectories(
-      std::map<std::string, std::shared_ptr<Directory>>* created_dirs);
+      std::map<std::string, std::shared_ptr<FSDirectory>>* created_dirs);
 
-  Directory* GetDataDir(size_t path_id) const;
+  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:
   friend class ColumnFamilySet;
+  static const uint32_t kDummyColumnFamilyDataId;
   ColumnFamilyData(uint32_t id, const std::string& name,
                    Version* dummy_versions, Cache* table_cache,
                    WriteBufferManager* write_buffer_manager,
@@ -515,7 +533,10 @@ class ColumnFamilyData {
                    const ImmutableDBOptions& db_options,
                    const FileOptions& file_options,
                    ColumnFamilySet* column_family_set,
-                   BlockCacheTracer* const block_cache_tracer);
+                   BlockCacheTracer* const block_cache_tracer,
+                   const std::shared_ptr<IOTracer>& io_tracer);
+
+  std::vector<std::string> GetDbPaths() const;
 
   uint32_t id_;
   const std::string name_;
@@ -537,6 +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_;
 
@@ -592,7 +614,11 @@ class ColumnFamilyData {
   std::atomic<uint64_t> last_memtable_id_;
 
   // Directories corresponding to cf_paths.
-  std::vector<std::shared_ptr<Directory>> data_dirs_;
+  std::vector<std::shared_ptr<FSDirectory>> data_dirs_;
+
+  bool db_paths_registered_;
+
+  std::string full_history_ts_low_;
 };
 
 // ColumnFamilySet has interesting thread-safety requirements
@@ -642,9 +668,10 @@ class ColumnFamilySet {
   ColumnFamilySet(const std::string& dbname,
                   const ImmutableDBOptions* db_options,
                   const FileOptions& file_options, Cache* table_cache,
-                  WriteBufferManager* write_buffer_manager,
-                  WriteController* write_controller,
-                  BlockCacheTracer* const block_cache_tracer);
+                  WriteBufferManager* _write_buffer_manager,
+                  WriteController* _write_controller,
+                  BlockCacheTracer* const block_cache_tracer,
+                  const std::shared_ptr<IOTracer>& io_tracer);
   ~ColumnFamilySet();
 
   ColumnFamilyData* GetDefault() const;
@@ -673,6 +700,10 @@ class ColumnFamilySet {
 
   Cache* get_table_cache() { return table_cache_; }
 
+  WriteBufferManager* write_buffer_manager() { return write_buffer_manager_; }
+
+  WriteController* write_controller() { return write_controller_; }
+
  private:
   friend class ColumnFamilyData;
   // helper function that gets called from cfd destructor
@@ -704,6 +735,7 @@ class ColumnFamilySet {
   WriteBufferManager* write_buffer_manager_;
   WriteController* write_controller_;
   BlockCacheTracer* const block_cache_tracer_;
+  std::shared_ptr<IOTracer> io_tracer_;
 };
 
 // We use ColumnFamilyMemTablesImpl to provide WriteBatch a way to access

db/column_family_test.cc

@@ -8,45 +8,37 @@
 // found in the LICENSE file. See the AUTHORS file for names of contributors.
 
 #include <algorithm>
-#include <vector>
 #include <string>
 #include <thread>
+#include <vector>
 
 #include "db/db_impl/db_impl.h"
 #include "db/db_test_util.h"
-#include "memtable/hash_skiplist_rep.h"
 #include "options/options_parser.h"
 #include "port/port.h"
 #include "port/stack_trace.h"
+#include "rocksdb/convenience.h"
 #include "rocksdb/db.h"
 #include "rocksdb/env.h"
 #include "rocksdb/iterator.h"
 #include "rocksdb/utilities/object_registry.h"
-#include "test_util/fault_injection_test_env.h"
 #include "test_util/sync_point.h"
 #include "test_util/testharness.h"
 #include "test_util/testutil.h"
 #include "util/coding.h"
 #include "util/string_util.h"
+#include "utilities/fault_injection_env.h"
 #include "utilities/merge_operators.h"
 
 namespace ROCKSDB_NAMESPACE {
 
 static const int kValueSize = 1000;
 
-namespace {
-std::string RandomString(Random* rnd, int len) {
-  std::string r;
-  test::RandomString(rnd, len, &r);
-  return r;
-}
-}  // anonymous namespace
-
 // 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_;
   }
@@ -76,21 +68,26 @@ 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 {
     std::vector<ColumnFamilyDescriptor> column_families;
     for (auto h : handles_) {
       ColumnFamilyDescriptor cfdescriptor;
-      h->GetDescriptor(&cfdescriptor);
+      Status s = h->GetDescriptor(&cfdescriptor);
+#ifdef ROCKSDB_LITE
+      EXPECT_TRUE(s.IsNotSupported());
+#else
+      EXPECT_OK(s);
+#endif  // ROCKSDB_LITE
       column_families.push_back(cfdescriptor);
     }
-    Close();
     ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
     Destroy(column_families);
     delete env_;
@@ -109,11 +106,11 @@ class ColumnFamilyTestBase : public testing::Test {
       // preserves the implementation that was in place when all of the
       // magic values in this file were picked.
       *storage = std::string(kValueSize, ' ');
-      return Slice(*storage);
     } else {
       Random r(k);
-      return test::RandomString(&r, kValueSize, storage);
+      *storage = r.RandomString(kValueSize);
     }
+    return Slice(*storage);
   }
 
   void Build(int base, int n, int flush_every = 0) {
@@ -122,7 +119,7 @@ class ColumnFamilyTestBase : public testing::Test {
 
     for (int i = 0; i < n; i++) {
       if (flush_every != 0 && i != 0 && i % flush_every == 0) {
-        DBImpl* dbi = reinterpret_cast<DBImpl*>(db_);
+        DBImpl* dbi = static_cast_with_check<DBImpl>(db_);
         dbi->TEST_FlushMemTable();
       }
 
@@ -176,7 +173,7 @@ class ColumnFamilyTestBase : public testing::Test {
   void Close() {
     for (auto h : handles_) {
       if (h) {
-        db_->DestroyColumnFamilyHandle(h);
+        ASSERT_OK(db_->DestroyColumnFamilyHandle(h));
       }
     }
     handles_.clear();
@@ -190,8 +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_);
@@ -202,8 +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_,
@@ -227,7 +224,7 @@ class ColumnFamilyTestBase : public testing::Test {
     Open({"default"});
   }
 
-  DBImpl* dbfull() { return reinterpret_cast<DBImpl*>(db_); }
+  DBImpl* dbfull() { return static_cast_with_check<DBImpl>(db_); }
 
   int GetProperty(int cf, std::string property) {
     std::string value;
@@ -287,7 +284,11 @@ class ColumnFamilyTestBase : public testing::Test {
       // Verify the CF options of the returned CF handle.
       ColumnFamilyDescriptor desc;
       ASSERT_OK(handles_[cfi]->GetDescriptor(&desc));
-      RocksDBOptionsParser::VerifyCFOptions(desc.options, current_cf_opt);
+      // Need to sanitize the default column family options before comparing
+      // them.
+      ASSERT_OK(RocksDBOptionsParser::VerifyCFOptions(
+          ConfigOptions(), desc.options,
+          SanitizeOptions(dbfull()->immutable_db_options(), current_cf_opt)));
 #endif  // !ROCKSDB_LITE
       cfi++;
     }
@@ -313,7 +314,7 @@ class ColumnFamilyTestBase : public testing::Test {
   void DropColumnFamilies(const std::vector<int>& cfs) {
     for (auto cf : cfs) {
       ASSERT_OK(db_->DropColumnFamily(handles_[cf]));
-      db_->DestroyColumnFamilyHandle(handles_[cf]);
+      ASSERT_OK(db_->DestroyColumnFamilyHandle(handles_[cf]));
       handles_[cf] = nullptr;
       names_[cf] = "";
     }
@@ -327,14 +328,14 @@ class ColumnFamilyTestBase : public testing::Test {
       // 10 bytes for key, rest is value
       if (!save) {
         ASSERT_OK(Put(cf, test::RandomKey(&rnd_, 11),
-                      RandomString(&rnd_, key_value_size - 10)));
+                      rnd_.RandomString(key_value_size - 10)));
       } else {
         std::string key = test::RandomKey(&rnd_, 11);
         keys_[cf].insert(key);
-        ASSERT_OK(Put(cf, key, RandomString(&rnd_, key_value_size - 10)));
+        ASSERT_OK(Put(cf, key, rnd_.RandomString(key_value_size - 10)));
       }
     }
-    db_->FlushWAL(false);
+    ASSERT_OK(db_->FlushWAL(/*sync=*/false));
   }
 
 #ifndef ROCKSDB_LITE  // TEST functions in DB are not supported in lite
@@ -568,7 +569,7 @@ TEST_P(ColumnFamilyTest, DontReuseColumnFamilyID) {
     Open();
     CreateColumnFamilies({"one", "two", "three"});
     for (size_t i = 0; i < handles_.size(); ++i) {
-      auto cfh = reinterpret_cast<ColumnFamilyHandleImpl*>(handles_[i]);
+      auto cfh = static_cast_with_check<ColumnFamilyHandleImpl>(handles_[i]);
       ASSERT_EQ(i, cfh->GetID());
     }
     if (iter == 1) {
@@ -584,7 +585,7 @@ TEST_P(ColumnFamilyTest, DontReuseColumnFamilyID) {
     CreateColumnFamilies({"three2"});
     // ID 3 that was used for dropped column family "three" should not be
     // reused
-    auto cfh3 = reinterpret_cast<ColumnFamilyHandleImpl*>(handles_[3]);
+    auto cfh3 = static_cast_with_check<ColumnFamilyHandleImpl>(handles_[3]);
     ASSERT_EQ(4U, cfh3->GetID());
     Close();
     Destroy();
@@ -652,11 +653,11 @@ 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
-  db_->FlushWAL(false);
+  ASSERT_OK(db_->FlushWAL(/*sync=*/false));
 
   // Preserve file system state up to here to simulate a crash condition.
   fault_env->SetFilesystemActive(false);
@@ -707,19 +708,19 @@ 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
-  db_->FlushWAL(false);
+  ASSERT_OK(db_->FlushWAL(/*sync=*/false));
   // Preserve file system state up to here to simulate a crash condition.
   fault_env->SetFilesystemActive(false);
   std::vector<std::string> names;
@@ -821,7 +822,7 @@ TEST_P(ColumnFamilyTest, BulkAddDrop) {
 }
 
 TEST_P(ColumnFamilyTest, DropTest) {
-  // first iteration - dont reopen DB before dropping
+  // first iteration - don't reopen DB before dropping
   // second iteration - reopen DB before dropping
   for (int iter = 0; iter < 2; ++iter) {
     Open({"default"});
@@ -848,13 +849,15 @@ TEST_P(ColumnFamilyTest, WriteBatchFailure) {
   Open();
   CreateColumnFamiliesAndReopen({"one", "two"});
   WriteBatch batch;
-  batch.Put(handles_[0], Slice("existing"), Slice("column-family"));
-  batch.Put(handles_[1], Slice("non-existing"), Slice("column-family"));
+  ASSERT_OK(batch.Put(handles_[0], Slice("existing"), Slice("column-family")));
+  ASSERT_OK(
+      batch.Put(handles_[1], Slice("non-existing"), Slice("column-family")));
   ASSERT_OK(db_->Write(WriteOptions(), &batch));
   DropColumnFamilies({1});
   WriteOptions woptions_ignore_missing_cf;
   woptions_ignore_missing_cf.ignore_missing_column_families = true;
-  batch.Put(handles_[0], Slice("still here"), Slice("column-family"));
+  ASSERT_OK(
+      batch.Put(handles_[0], Slice("still here"), Slice("column-family")));
   ASSERT_OK(db_->Write(woptions_ignore_missing_cf, &batch));
   ASSERT_EQ("column-family", Get(0, "still here"));
   Status s = db_->Write(WriteOptions(), &batch);
@@ -893,11 +896,9 @@ TEST_P(ColumnFamilyTest, IgnoreRecoveredLog) {
   ASSERT_OK(env_->CreateDirIfMissing(dbname_));
   ASSERT_OK(env_->CreateDirIfMissing(backup_logs));
   std::vector<std::string> old_files;
-  env_->GetChildren(backup_logs, &old_files);
+  ASSERT_OK(env_->GetChildren(backup_logs, &old_files));
   for (auto& file : old_files) {
-    if (file != "." && file != "..") {
-      env_->DeleteFile(backup_logs + "/" + file);
-    }
+    ASSERT_OK(env_->DeleteFile(backup_logs + "/" + file));
   }
 
   column_family_options_.merge_operator =
@@ -924,11 +925,9 @@ TEST_P(ColumnFamilyTest, IgnoreRecoveredLog) {
 
   // copy the logs to backup
   std::vector<std::string> logs;
-  env_->GetChildren(db_options_.wal_dir, &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
@@ -953,9 +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);
       }
     }
   }
@@ -982,13 +979,14 @@ 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"));
     ASSERT_OK(Put(0, "foofoo", "bar"));
 
     for (auto* it : iterators) {
+      ASSERT_OK(it->status());
       delete it;
     }
   }
@@ -1086,10 +1084,10 @@ TEST_P(ColumnFamilyTest, CrashAfterFlush) {
   CreateColumnFamilies({"one"});
 
   WriteBatch batch;
-  batch.Put(handles_[0], Slice("foo"), Slice("bar"));
-  batch.Put(handles_[1], Slice("foo"), Slice("bar"));
+  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;
@@ -1099,7 +1097,7 @@ TEST_P(ColumnFamilyTest, CrashAfterFlush) {
     }
   }
   Close();
-  fault_env->DropUnsyncedFileData();
+  ASSERT_OK(fault_env->DropUnsyncedFileData());
   fault_env->ResetState();
   Open(names, {});
 
@@ -2073,6 +2071,7 @@ std::string IterStatus(Iterator* iter) {
   if (iter->Valid()) {
     result = iter->key().ToString() + "->" + iter->value().ToString();
   } else {
+    EXPECT_OK(iter->status());
     result = "(invalid)";
   }
   return result;
@@ -2231,7 +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();
 }
@@ -2327,7 +2326,7 @@ TEST_P(ColumnFamilyTest, ReadDroppedColumnFamily) {
         ASSERT_OK(db_->DropColumnFamily(handles_[2]));
       } else {
         // delete CF two
-        db_->DestroyColumnFamilyHandle(handles_[2]);
+        ASSERT_OK(db_->DestroyColumnFamilyHandle(handles_[2]));
         handles_[2] = nullptr;
       }
       // Make sure iterator created can still be used.
@@ -2383,7 +2382,6 @@ TEST_P(ColumnFamilyTest, LiveIteratorWithDroppedColumnFamily) {
   // 1MB should create ~10 files for each CF
   int kKeysNum = 10000;
   PutRandomData(1, kKeysNum, 100);
-
   {
     std::unique_ptr<Iterator> iterator(
         db_->NewIterator(ReadOptions(), handles_[1]));
@@ -2430,6 +2428,9 @@ TEST_P(ColumnFamilyTest, FlushAndDropRaceCondition) {
 
   env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask, &sleeping_task,
                  Env::Priority::HIGH);
+  // Make sure the task is sleeping. Otherwise, it might start to execute
+  // after sleeping_task.WaitUntilDone() and cause TSAN warning.
+  sleeping_task.WaitUntilSleeping();
 
   // 1MB should create ~10 files for each CF
   int kKeysNum = 10000;
@@ -2444,6 +2445,9 @@ TEST_P(ColumnFamilyTest, FlushAndDropRaceCondition) {
   // now we sleep again. this is just so we're certain that flush job finished
   env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask, &sleeping_task,
                  Env::Priority::HIGH);
+  // Make sure the task is sleeping. Otherwise, it might start to execute
+  // after sleeping_task.WaitUntilDone() and cause TSAN warning.
+  sleeping_task.WaitUntilSleeping();
   sleeping_task.WakeUp();
   sleeping_task.WaitUntilDone();
 
@@ -2993,6 +2997,9 @@ TEST_P(ColumnFamilyTest, FlushCloseWALFiles) {
   test::SleepingBackgroundTask sleeping_task;
   env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask, &sleeping_task,
                  Env::Priority::HIGH);
+  // Make sure the task is sleeping. Otherwise, it might start to execute
+  // after sleeping_task.WaitUntilDone() and cause TSAN warning.
+  sleeping_task.WaitUntilSleeping();
 
   WriteOptions wo;
   wo.sync = true;
@@ -3025,8 +3032,9 @@ TEST_P(ColumnFamilyTest, IteratorCloseWALFile1) {
   ASSERT_OK(Put(1, "fodor", "mirko"));
   // Create an iterator holding the current super version.
   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"));
@@ -3038,6 +3046,9 @@ TEST_P(ColumnFamilyTest, IteratorCloseWALFile1) {
   test::SleepingBackgroundTask sleeping_task;
   env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask, &sleeping_task,
                  Env::Priority::HIGH);
+  // Make sure the task is sleeping. Otherwise, it might start to execute
+  // after sleeping_task.WaitUntilDone() and cause TSAN warning.
+  sleeping_task.WaitUntilSleeping();
 
   WriteOptions wo;
   wo.sync = true;
@@ -3074,8 +3085,9 @@ TEST_P(ColumnFamilyTest, IteratorCloseWALFile2) {
   ReadOptions ro;
   ro.background_purge_on_iterator_cleanup = true;
   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"));
@@ -3129,7 +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.
@@ -3141,7 +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();
 
@@ -3168,6 +3180,8 @@ TEST_P(ColumnFamilyTest, ForwardIteratorCloseWALFile) {
   // Deleting the iterator will clear its super version, triggering
   // closing all files
   it->Seek("");
+  ASSERT_OK(it->status());
+
   ASSERT_EQ(2, env.num_open_wal_file_.load());
   ASSERT_EQ(0, env.delete_count_.load());
 
@@ -3198,8 +3212,8 @@ TEST_P(ColumnFamilyTest, LogSyncConflictFlush) {
   Open();
   CreateColumnFamiliesAndReopen({"one", "two"});
 
-  Put(0, "", "");
-  Put(1, "foo", "bar");
+  ASSERT_OK(Put(0, "", ""));
+  ASSERT_OK(Put(1, "foo", "bar"));
 
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
       {{"DBImpl::SyncWAL:BeforeMarkLogsSynced:1",
@@ -3209,12 +3223,12 @@ TEST_P(ColumnFamilyTest, LogSyncConflictFlush) {
 
   ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
 
-  ROCKSDB_NAMESPACE::port::Thread thread([&] { db_->SyncWAL(); });
+  ROCKSDB_NAMESPACE::port::Thread thread([&] { ASSERT_OK(db_->SyncWAL()); });
 
   TEST_SYNC_POINT("ColumnFamilyTest::LogSyncConflictFlush:1");
-  Flush(1);
-  Put(1, "foo", "bar");
-  Flush(1);
+  ASSERT_OK(Flush(1));
+  ASSERT_OK(Put(1, "foo", "bar"));
+  ASSERT_OK(Flush(1));
 
   TEST_SYNC_POINT("ColumnFamilyTest::LogSyncConflictFlush:2");
 
@@ -3236,7 +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);
@@ -3251,7 +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]);
   }
@@ -3296,7 +3310,7 @@ TEST_P(ColumnFamilyTest, DISABLED_LogTruncationTest) {
   Close();
 
   // cleanup
-  env_->DeleteDir(backup_logs);
+  ASSERT_OK(env_->DeleteDir(backup_logs));
 }
 
 TEST_P(ColumnFamilyTest, DefaultCfPathsTest) {
@@ -3312,14 +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.
@@ -3338,29 +3352,31 @@ 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_));
 
   // Re-open and verify the keys.
   Reopen({ColumnFamilyOptions(), cf_opt1, cf_opt2});
-  DBImpl* dbi = reinterpret_cast<DBImpl*>(db_);
+  DBImpl* dbi = static_cast_with_check<DBImpl>(db_);
   for (int cf = 1; cf != 3; ++cf) {
     ReadOptions read_options;
     read_options.readahead_size = 0;
     auto it = dbi->NewIterator(read_options, handles_[cf]);
     for (it->SeekToFirst(); it->Valid(); it->Next()) {
+      ASSERT_OK(it->status());
       Slice key(it->key());
       ASSERT_NE(keys_[cf].end(), keys_[cf].find(key.ToString()));
     }
+    ASSERT_OK(it->status());
     delete it;
 
     for (const auto& key : keys_[cf]) {
@@ -3369,6 +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

@@ -16,6 +16,7 @@
 #include "rocksdb/env.h"
 #include "test_util/sync_point.h"
 #include "test_util/testharness.h"
+#include "util/cast_util.h"
 #include "util/string_util.h"
 
 namespace ROCKSDB_NAMESPACE {
@@ -90,9 +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;
@@ -117,6 +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
@@ -137,18 +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));
-  reinterpret_cast<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) {
@@ -181,15 +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))));
   }
-  reinterpret_cast<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))));
   }
-  reinterpret_cast<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));
@@ -212,13 +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();
@@ -236,8 +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();
@@ -248,7 +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;
 }
@@ -292,8 +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;
@@ -336,8 +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());
@@ -376,14 +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))));
   }
-  reinterpret_cast<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

@@ -7,12 +7,14 @@
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file. See the AUTHORS file for names of contributors.
 
+#include "db/compaction/compaction.h"
+
 #include <cinttypes>
 #include <vector>
 
 #include "db/column_family.h"
-#include "db/compaction/compaction.h"
 #include "rocksdb/compaction_filter.h"
+#include "rocksdb/sst_partitioner.h"
 #include "test_util/sync_point.h"
 #include "util/string_util.h"
 
@@ -23,7 +25,7 @@ const uint64_t kRangeTombstoneSentinel =
 
 int sstableKeyCompare(const Comparator* user_cmp, const InternalKey& a,
                       const InternalKey& b) {
-  auto c = user_cmp->Compare(a.user_key(), b.user_key());
+  auto c = user_cmp->CompareWithoutTimestamp(a.user_key(), b.user_key());
   if (c != 0) {
     return c;
   }
@@ -205,6 +207,7 @@ bool Compaction::IsFullCompaction(
 Compaction::Compaction(VersionStorageInfo* vstorage,
                        const ImmutableCFOptions& _immutable_cf_options,
                        const MutableCFOptions& _mutable_cf_options,
+                       const MutableDBOptions& _mutable_db_options,
                        std::vector<CompactionInputFiles> _inputs,
                        int _output_level, uint64_t _target_file_size,
                        uint64_t _max_compaction_bytes, uint32_t _output_path_id,
@@ -243,13 +246,7 @@ Compaction::Compaction(VersionStorageInfo* vstorage,
     compaction_reason_ = CompactionReason::kManualCompaction;
   }
   if (max_subcompactions_ == 0) {
-    max_subcompactions_ = immutable_cf_options_.max_subcompactions;
-  }
-  if (!bottommost_level_) {
-    // Currently we only enable dictionary compression during compaction to the
-    // bottommost level.
-    output_compression_opts_.max_dict_bytes = 0;
-    output_compression_opts_.zstd_max_train_bytes = 0;
+    max_subcompactions_ = _mutable_db_options.max_subcompactions;
   }
 
 #ifndef NDEBUG
@@ -328,6 +325,8 @@ bool Compaction::IsTrivialMove() const {
 
   // assert inputs_.size() == 1
 
+  std::unique_ptr<SstPartitioner> partitioner = CreateSstPartitioner();
+
   for (const auto& file : inputs_.front().files) {
     std::vector<FileMetaData*> file_grand_parents;
     if (output_level_ + 1 >= number_levels_) {
@@ -340,6 +339,13 @@ bool Compaction::IsTrivialMove() const {
     if (compaction_size > max_compaction_bytes_) {
       return false;
     }
+
+    if (partitioner.get() != nullptr) {
+      if (!partitioner->CanDoTrivialMove(file->smallest.user_key(),
+                                         file->largest.user_key())) {
+        return false;
+      }
+    }
   }
 
   return true;
@@ -371,7 +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;
@@ -507,7 +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));
 }
@@ -525,6 +537,21 @@ std::unique_ptr<CompactionFilter> Compaction::CreateCompactionFilter() const {
       context);
 }
 
+std::unique_ptr<SstPartitioner> Compaction::CreateSstPartitioner() const {
+  if (!immutable_cf_options_.sst_partitioner_factory) {
+    return nullptr;
+  }
+
+  SstPartitioner::Context context;
+  context.is_full_compaction = is_full_compaction_;
+  context.is_manual_compaction = is_manual_compaction_;
+  context.output_level = output_level_;
+  context.smallest_user_key = smallest_user_key_;
+  context.largest_user_key = largest_user_key_;
+  return immutable_cf_options_.sst_partitioner_factory->CreatePartitioner(
+      context);
+}
+
 bool Compaction::IsOutputLevelEmpty() const {
   return inputs_.back().level != output_level_ || inputs_.back().empty();
 }

db/compaction/compaction.h

@@ -11,6 +11,7 @@
 #include "db/version_set.h"
 #include "memory/arena.h"
 #include "options/cf_options.h"
+#include "rocksdb/sst_partitioner.h"
 #include "util/autovector.h"
 
 namespace ROCKSDB_NAMESPACE {
@@ -71,6 +72,7 @@ class Compaction {
   Compaction(VersionStorageInfo* input_version,
              const ImmutableCFOptions& immutable_cf_options,
              const MutableCFOptions& mutable_cf_options,
+             const MutableDBOptions& mutable_db_options,
              std::vector<CompactionInputFiles> inputs, int output_level,
              uint64_t target_file_size, uint64_t max_compaction_bytes,
              uint32_t output_path_id, CompressionType compression,
@@ -255,6 +257,9 @@ class Compaction {
   // Create a CompactionFilter from compaction_filter_factory
   std::unique_ptr<CompactionFilter> CreateCompactionFilter() const;
 
+  // Create a SstPartitioner from sst_partitioner_factory
+  std::unique_ptr<SstPartitioner> CreateSstPartitioner() const;
+
   // Is the input level corresponding to output_level_ empty?
   bool IsOutputLevelEmpty() const;
 
@@ -336,7 +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

@@ -3,9 +3,13 @@
 //  COPYING file in the root directory) and Apache 2.0 License
 //  (found in the LICENSE.Apache file in the root directory).
 
-#include <cinttypes>
-
 #include "db/compaction/compaction_iterator.h"
+
+#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"
@@ -36,20 +40,23 @@ CompactionIterator::CompactionIterator(
     SequenceNumber earliest_write_conflict_snapshot,
     const SnapshotChecker* snapshot_checker, Env* env,
     bool report_detailed_time, bool expect_valid_internal_key,
-    CompactionRangeDelAggregator* range_del_agg, const Compaction* compaction,
-    const CompactionFilter* compaction_filter,
+    CompactionRangeDelAggregator* range_del_agg,
+    BlobFileBuilder* blob_file_builder, bool allow_data_in_errors,
+    const Compaction* compaction, const CompactionFilter* compaction_filter,
     const std::atomic<bool>* shutting_down,
     const SequenceNumber preserve_deletes_seqnum,
-    const std::atomic<bool>* manual_compaction_paused,
-    const std::shared_ptr<Logger> info_log)
+    const std::atomic<int>* manual_compaction_paused,
+    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,
@@ -58,12 +65,14 @@ CompactionIterator::CompactionIterator(
     const SnapshotChecker* snapshot_checker, Env* env,
     bool report_detailed_time, bool expect_valid_internal_key,
     CompactionRangeDelAggregator* range_del_agg,
+    BlobFileBuilder* blob_file_builder, bool allow_data_in_errors,
     std::unique_ptr<CompactionProxy> compaction,
     const CompactionFilter* compaction_filter,
     const std::atomic<bool>* shutting_down,
     const SequenceNumber preserve_deletes_seqnum,
-    const std::atomic<bool>* manual_compaction_paused,
-    const std::shared_ptr<Logger> info_log)
+    const std::atomic<int>* manual_compaction_paused,
+    const std::shared_ptr<Logger> info_log,
+    const std::string* full_history_ts_low)
     : input_(input),
       cmp_(cmp),
       merge_helper_(merge_helper),
@@ -71,23 +80,33 @@ 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),
+      blob_file_builder_(blob_file_builder),
       compaction_(std::move(compaction)),
       compaction_filter_(compaction_filter),
       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) {
+      cmp_with_history_ts_low_(0) {
   assert(compaction_filter_ == nullptr || compaction_ != nullptr);
   assert(snapshots_ != nullptr);
-  bottommost_level_ =
-      compaction_ == nullptr ? false : compaction_->bottommost_level();
+  bottommost_level_ = compaction_ == nullptr
+                          ? false
+                          : compaction_->bottommost_level() &&
+                                !compaction_->allow_ingest_behind();
   if (compaction_ != nullptr) {
     level_ptrs_ = std::vector<size_t>(compaction_->number_levels(), 0);
   }
@@ -108,13 +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);
 }
 
@@ -142,14 +163,13 @@ void CompactionIterator::Next() {
     if (merge_out_iter_.Valid()) {
       key_ = merge_out_iter_.key();
       value_ = merge_out_iter_.value();
-      bool valid_key __attribute__((__unused__));
-      valid_key =  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(valid_key);
-      if (!valid_key) {
-        ROCKS_LOG_FATAL(info_log_, "Invalid key (%s) in compaction",
-                        key_.ToString(true).c_str());
+      assert(s.ok());
+      if (!s.ok()) {
+        ROCKS_LOG_FATAL(info_log_, "Invalid key in compaction. %s",
+                        s.getState());
       }
 
       // Keep current_key_ in sync.
@@ -182,57 +202,148 @@ void CompactionIterator::Next() {
   PrepareOutput();
 }
 
-void CompactionIterator::InvokeFilterIfNeeded(bool* need_skip,
+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_);
-      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;
-    }
-
-    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) {
-      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();
-    }
+  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,
+          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 (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);
+    }
+    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 !error;
 }
 
 void CompactionIterator::NextFromInput() {
@@ -245,27 +356,28 @@ void CompactionIterator::NextFromInput() {
     value_ = input_->value();
     iter_stats_.num_input_records++;
 
-    if (!ParseInternalKey(key_, &ikey_)) {
+    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_) {
-        assert(!"Corrupted internal key not expected.");
-        status_ = Status::Corruption("Corrupted internal key not expected.");
-        break;
+        status_ = pik_status;
+        return;
       }
       key_ = current_key_.SetInternalKey(key_);
       has_current_user_key_ = false;
       current_user_key_sequence_ = kMaxSequenceNumber;
       current_user_key_snapshot_ = 0;
-      iter_stats_.num_input_corrupt_records++;
       valid_ = true;
       break;
     }
     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();
@@ -278,25 +390,61 @@ 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
       // key.
-      if (current_key_committed_) {
-        InvokeFilterIfNeeded(&need_skip, &skip_until);
+      if (current_key_committed_ &&
+          !InvokeFilterIfNeeded(&need_skip, &skip_until)) {
+        break;
       }
     } else {
       // Update the current key to reflect the new sequence number/type without
@@ -316,8 +464,9 @@ void CompactionIterator::NextFromInput() {
         current_key_committed_ = KeyCommitted(ikey_.sequence);
         // Apply the compaction filter to the first committed version of the
         // user key.
-        if (current_key_committed_) {
-          InvokeFilterIfNeeded(&need_skip, &skip_until);
+        if (current_key_committed_ &&
+            !InvokeFilterIfNeeded(&need_skip, &skip_until)) {
+          break;
         }
       }
     }
@@ -331,8 +480,7 @@ void CompactionIterator::NextFromInput() {
     // If there are no snapshots, then this kv affect visibility at tip.
     // Otherwise, search though all existing snapshots to find the earliest
     // snapshot that is affected by this kv.
-    SequenceNumber last_sequence __attribute__((__unused__));
-    last_sequence = current_user_key_sequence_;
+    SequenceNumber last_sequence = current_user_key_sequence_;
     current_user_key_sequence_ = ikey_.sequence;
     SequenceNumber last_snapshot = current_user_key_snapshot_;
     SequenceNumber prev_snapshot = 0;  // 0 means no previous snapshot
@@ -347,8 +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);
@@ -361,6 +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;
@@ -404,7 +557,9 @@ 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) &&
+      if (input_->Valid() &&
+          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.
@@ -508,9 +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,
@@ -534,30 +692,47 @@ 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
+      assert(!compaction_ || compaction_->KeyNotExistsBeyondOutputLevel(
+                                 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
-      while (input_->Valid() && ParseInternalKey(input_->key(), &next_ikey) &&
-             cmp_->Equal(ikey_.user_key, next_ikey.user_key) &&
+      // 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, 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();
       }
       // If you find you still need to output a row with this key, we need to output the
       // delete too
-      if (input_->Valid() && ParseInternalKey(input_->key(), &next_ikey) &&
-          cmp_->Equal(ikey_.user_key, next_ikey.user_key)) {
+      if (input_->Valid() &&
+          (ParseInternalKey(input_->key(), &next_ikey, allow_data_in_errors_)
+               .ok()) &&
+          cmp_->EqualWithoutTimestamp(ikey_.user_key, next_ikey.user_key)) {
         valid_ = true;
         at_next_ = true;
       }
@@ -573,8 +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()) {
@@ -585,14 +761,13 @@ void CompactionIterator::NextFromInput() {
         //       These will be correctly set below.
         key_ = merge_out_iter_.key();
         value_ = merge_out_iter_.value();
-        bool valid_key __attribute__((__unused__));
-        valid_key = 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(valid_key);
-        if (!valid_key) {
-          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);
@@ -638,23 +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 (compaction_filter_ && ikey_.type == kTypeBlobIndex) {
-      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_;
-      }
+    if (ikey_.type == kTypeValue) {
+      ExtractLargeValueIfNeeded();
+    } else if (ikey_.type == kTypeBlobIndex) {
+      GarbageCollectBlobIfNeeded();
     }
 
     // Zeroing out the sequence number leads to better compression.
@@ -679,7 +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);
+      }
     }
   }
 }
@@ -771,4 +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>
@@ -21,72 +22,123 @@
 
 namespace ROCKSDB_NAMESPACE {
 
+class BlobFileBuilder;
+
 class CompactionIterator {
  public:
   // A wrapper around Compaction. Has a much smaller interface, only what
   // 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_;
   };
 
-  CompactionIterator(
-      InternalIterator* input, const Comparator* cmp, MergeHelper* merge_helper,
-      SequenceNumber last_sequence, std::vector<SequenceNumber>* snapshots,
-      SequenceNumber earliest_write_conflict_snapshot,
-      const SnapshotChecker* snapshot_checker, Env* env,
-      bool report_detailed_time, bool expect_valid_internal_key,
-      CompactionRangeDelAggregator* range_del_agg,
-      const Compaction* compaction = nullptr,
-      const CompactionFilter* compaction_filter = nullptr,
-      const std::atomic<bool>* shutting_down = nullptr,
-      const SequenceNumber preserve_deletes_seqnum = 0,
-      const std::atomic<bool>* manual_compaction_paused = nullptr,
-      const std::shared_ptr<Logger> info_log = nullptr);
+  CompactionIterator(InternalIterator* input, const Comparator* cmp,
+                     MergeHelper* merge_helper, SequenceNumber last_sequence,
+                     std::vector<SequenceNumber>* snapshots,
+                     SequenceNumber earliest_write_conflict_snapshot,
+                     const SnapshotChecker* snapshot_checker, Env* env,
+                     bool report_detailed_time, bool expect_valid_internal_key,
+                     CompactionRangeDelAggregator* range_del_agg,
+                     BlobFileBuilder* blob_file_builder,
+                     bool allow_data_in_errors,
+                     const Compaction* compaction = nullptr,
+                     const CompactionFilter* compaction_filter = nullptr,
+                     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::string* full_history_ts_low = nullptr);
 
   // Constructor with custom CompactionProxy, used for tests.
-  CompactionIterator(
-      InternalIterator* input, const Comparator* cmp, MergeHelper* merge_helper,
-      SequenceNumber last_sequence, std::vector<SequenceNumber>* snapshots,
-      SequenceNumber earliest_write_conflict_snapshot,
-      const SnapshotChecker* snapshot_checker, Env* env,
-      bool report_detailed_time, bool expect_valid_internal_key,
-      CompactionRangeDelAggregator* range_del_agg,
-      std::unique_ptr<CompactionProxy> compaction,
-      const CompactionFilter* compaction_filter = nullptr,
-      const std::atomic<bool>* shutting_down = nullptr,
-      const SequenceNumber preserve_deletes_seqnum = 0,
-      const std::atomic<bool>* manual_compaction_paused = nullptr,
-      const std::shared_ptr<Logger> info_log = nullptr);
+  CompactionIterator(InternalIterator* input, const Comparator* cmp,
+                     MergeHelper* merge_helper, SequenceNumber last_sequence,
+                     std::vector<SequenceNumber>* snapshots,
+                     SequenceNumber earliest_write_conflict_snapshot,
+                     const SnapshotChecker* snapshot_checker, Env* env,
+                     bool report_detailed_time, bool expect_valid_internal_key,
+                     CompactionRangeDelAggregator* range_del_agg,
+                     BlobFileBuilder* blob_file_builder,
+                     bool allow_data_in_errors,
+                     std::unique_ptr<CompactionProxy> compaction,
+                     const CompactionFilter* compaction_filter = nullptr,
+                     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::string* full_history_ts_low = nullptr);
 
   ~CompactionIterator();
 
@@ -115,13 +167,33 @@ 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.
-  void InvokeFilterIfNeeded(bool* need_skip, Slice* skip_until);
+  // Return true on success, false on failures (e.g.: kIOError).
+  bool InvokeFilterIfNeeded(bool* need_skip, Slice* skip_until);
 
   // Given a sequence number, return the sequence number of the
   // earliest snapshot that this sequence number is visible in.
@@ -145,6 +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_;
@@ -159,13 +248,15 @@ 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_;
+  BlobFileBuilder* blob_file_builder_;
   std::unique_ptr<CompactionProxy> compaction_;
   const CompactionFilter* compaction_filter_;
   const std::atomic<bool>* shutting_down_;
-  const std::atomic<bool>* manual_compaction_paused_;
+  const std::atomic<int>* manual_compaction_paused_;
   const SequenceNumber preserve_deletes_seqnum_;
   bool bottommost_level_;
   bool valid_ = false;
@@ -173,6 +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_)
@@ -191,11 +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_;
 
@@ -210,6 +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
@@ -224,7 +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_;
+
+  // 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.
@@ -234,7 +348,7 @@ class CompactionIterator {
   bool IsPausingManualCompaction() {
     // This is a best-effort facility, so memory_order_relaxed is sufficient.
     return manual_compaction_paused_ &&
-           manual_compaction_paused_->load(std::memory_order_relaxed);
+           manual_compaction_paused_->load(std::memory_order_relaxed) > 0;
   }
 };
 }  // namespace ROCKSDB_NAMESPACE

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*/,
@@ -156,29 +156,40 @@ 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 false; }
+
+  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;
+
+  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:
@@ -214,6 +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,
@@ -222,7 +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>(
@@ -234,9 +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();
@@ -249,13 +268,22 @@ 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(
         iter_.get(), cmp_, merge_helper_.get(), last_sequence, &snapshots_,
         earliest_write_conflict_snapshot, snapshot_checker_.get(),
         Env::Default(), false /* report_detailed_time */, false,
-        range_del_agg_.get(), std::move(compaction), filter, &shutting_down_));
+        range_del_agg_.get(), nullptr /* blob_file_builder */,
+        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,
@@ -266,6 +294,8 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
 
   virtual bool UseSnapshotChecker() const { return false; }
 
+  virtual bool AllowIngestBehind() const { return false; }
+
   void RunTest(
       const std::vector<std::string>& input_keys,
       const std::vector<std::string>& input_values,
@@ -275,10 +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);
@@ -288,9 +321,15 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
       ASSERT_EQ(expected_values[i], c_iter_->value().ToString()) << info;
       c_iter_->Next();
     }
+    ASSERT_OK(c_iter_->status());
     ASSERT_FALSE(c_iter_->Valid());
   }
 
+  void ClearSnapshots() {
+    snapshots_.clear();
+    snapshot_map_.clear();
+  }
+
   const Comparator* cmp_;
   const InternalKeyComparator icmp_;
   std::vector<SequenceNumber> snapshots_;
@@ -312,6 +351,7 @@ TEST_P(CompactionIteratorTest, EmptyResult) {
                  test::KeyStr("a", 3, kTypeValue)},
                 {"", "val"}, {}, {}, 5);
   c_iter_->SeekToFirst();
+  ASSERT_OK(c_iter_->status());
   ASSERT_FALSE(c_iter_->Valid());
 }
 
@@ -333,6 +373,7 @@ TEST_P(CompactionIteratorTest, CorruptionAfterSingleDeletion) {
   ASSERT_TRUE(c_iter_->Valid());
   ASSERT_EQ(test::KeyStr("b", 10, kTypeValue), c_iter_->key().ToString());
   c_iter_->Next();
+  ASSERT_OK(c_iter_->status());
   ASSERT_FALSE(c_iter_->Valid());
 }
 
@@ -349,6 +390,7 @@ TEST_P(CompactionIteratorTest, SimpleRangeDeletion) {
   ASSERT_TRUE(c_iter_->Valid());
   ASSERT_EQ(test::KeyStr("night", 3, kTypeValue), c_iter_->key().ToString());
   c_iter_->Next();
+  ASSERT_OK(c_iter_->status());
   ASSERT_FALSE(c_iter_->Valid());
 }
 
@@ -370,6 +412,7 @@ TEST_P(CompactionIteratorTest, RangeDeletionWithSnapshots) {
   ASSERT_TRUE(c_iter_->Valid());
   ASSERT_EQ(test::KeyStr("night", 40, kTypeValue), c_iter_->key().ToString());
   c_iter_->Next();
+  ASSERT_OK(c_iter_->status());
   ASSERT_FALSE(c_iter_->Valid());
 }
 
@@ -463,6 +506,7 @@ TEST_P(CompactionIteratorTest, CompactionFilterSkipUntil) {
   ASSERT_EQ(test::KeyStr("h", 91, kTypeValue), c_iter_->key().ToString());
   ASSERT_EQ("hv91", c_iter_->value().ToString());
   c_iter_->Next();
+  ASSERT_OK(c_iter_->status());
   ASSERT_FALSE(c_iter_->Valid());
 
   // Check that the compaction iterator did the correct sequence of calls on
@@ -656,6 +700,7 @@ TEST_P(CompactionIteratorTest, SingleMergeOperand) {
   ASSERT_TRUE(c_iter_->Valid());
   ASSERT_EQ("bv1bv2", c_iter_->value().ToString());
   c_iter_->Next();
+  ASSERT_OK(c_iter_->status());
   ASSERT_EQ("cv1cv2", c_iter_->value().ToString());
 }
 
@@ -666,7 +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*/);
 }
@@ -675,15 +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
@@ -693,10 +737,22 @@ 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*/);
 }
 
+TEST_P(CompactionIteratorTest, ConvertToPutAtBottom) {
+  std::shared_ptr<MergeOperator> merge_op =
+      MergeOperators::CreateStringAppendOperator();
+  RunTest({test::KeyStr("a", 4, kTypeMerge), test::KeyStr("a", 3, kTypeMerge),
+           test::KeyStr("a", 2, kTypeMerge), test::KeyStr("b", 1, kTypeValue)},
+          {"a4", "a3", "a2", "b1"},
+          {test::KeyStr("a", 0, kTypeValue), test::KeyStr("b", 0, kTypeValue)},
+          {"a2,a3,a4", "b1"}, kMaxSequenceNumber /*last_committed_seq*/,
+          merge_op.get(), nullptr /*compaction_filter*/,
+          true /*bottomost_level*/);
+}
+
 INSTANTIATE_TEST_CASE_P(CompactionIteratorTestInstance, CompactionIteratorTest,
                         testing::Values(true, false));
 
@@ -838,7 +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*/);
 }
@@ -849,9 +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*/);
 }
@@ -859,15 +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,
@@ -879,7 +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*/);
 }
@@ -913,9 +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(
@@ -968,6 +1036,227 @@ TEST_F(CompactionIteratorWithSnapshotCheckerTest, CompactionFilter_FullMerge) {
       compaction_filter.get());
 }
 
+// Tests how CompactionIterator work together with AllowIngestBehind.
+class CompactionIteratorWithAllowIngestBehindTest
+    : public CompactionIteratorTest {
+ public:
+  bool AllowIngestBehind() const override { return true; }
+};
+
+// When allow_ingest_behind is set, compaction iterator is not targeting
+// the bottommost level since there is no guarantee there won't be further
+// data ingested under the compaction output in future.
+TEST_P(CompactionIteratorWithAllowIngestBehindTest, NoConvertToPutAtBottom) {
+  std::shared_ptr<MergeOperator> merge_op =
+      MergeOperators::CreateStringAppendOperator();
+  RunTest({test::KeyStr("a", 4, kTypeMerge), test::KeyStr("a", 3, kTypeMerge),
+           test::KeyStr("a", 2, kTypeMerge), test::KeyStr("b", 1, kTypeValue)},
+          {"a4", "a3", "a2", "b1"},
+          {test::KeyStr("a", 4, kTypeMerge), test::KeyStr("b", 1, kTypeValue)},
+          {"a2,a3,a4", "b1"}, kMaxSequenceNumber /*last_committed_seq*/,
+          merge_op.get(), nullptr /*compaction_filter*/,
+          true /*bottomost_level*/);
+}
+
+TEST_P(CompactionIteratorWithAllowIngestBehindTest,
+       MergeToPutIfEncounteredPutAtBottom) {
+  std::shared_ptr<MergeOperator> merge_op =
+      MergeOperators::CreateStringAppendOperator();
+  RunTest({test::KeyStr("a", 4, kTypeMerge), test::KeyStr("a", 3, kTypeMerge),
+           test::KeyStr("a", 2, kTypeValue), test::KeyStr("b", 1, kTypeValue)},
+          {"a4", "a3", "a2", "b1"},
+          {test::KeyStr("a", 4, kTypeValue), test::KeyStr("b", 1, kTypeValue)},
+          {"a2,a3,a4", "b1"}, kMaxSequenceNumber /*last_committed_seq*/,
+          merge_op.get(), nullptr /*compaction_filter*/,
+          true /*bottomost_level*/);
+}
+
+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.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;
@@ -62,23 +64,25 @@ class VersionSet;
 // if needed.
 class CompactionJob {
  public:
-  CompactionJob(int job_id, Compaction* compaction,
-                const ImmutableDBOptions& db_options,
-                const FileOptions& file_options, VersionSet* versions,
-                const std::atomic<bool>* shutting_down,
-                const SequenceNumber preserve_deletes_seqnum,
-                LogBuffer* log_buffer, Directory* db_directory,
-                Directory* output_directory, Statistics* stats,
-                InstrumentedMutex* db_mutex, ErrorHandler* db_error_handler,
-                std::vector<SequenceNumber> existing_snapshots,
-                SequenceNumber earliest_write_conflict_snapshot,
-                const SnapshotChecker* snapshot_checker,
-                std::shared_ptr<Cache> table_cache, EventLogger* event_logger,
-                bool paranoid_file_checks, bool measure_io_stats,
-                const std::string& dbname,
-                CompactionJobStats* compaction_job_stats,
-                Env::Priority thread_pri,
-                const std::atomic<bool>* manual_compaction_paused = nullptr);
+  CompactionJob(
+      int job_id, Compaction* compaction, const ImmutableDBOptions& db_options,
+      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,
+      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,
+      std::shared_ptr<Cache> table_cache, EventLogger* event_logger,
+      bool paranoid_file_checks, bool measure_io_stats,
+      const std::string& dbname, CompactionJobStats* compaction_job_stats,
+      Env::Priority thread_pri, 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 = "",
+      std::string full_history_ts_low = "",
+      BlobFileCompletionCallback* blob_callback = nullptr);
 
   ~CompactionJob();
 
@@ -100,6 +104,9 @@ class CompactionJob {
   // Add compaction input/output to the current version
   Status Install(const MutableCFOptions& mutable_cf_options);
 
+  // Return the IO status
+  IOStatus io_status() const { return io_status_; }
+
  private:
   struct SubcompactionState;
 
@@ -149,20 +156,24 @@ class CompactionJob {
 
   // 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_;
-  FileSystem* fs_;
+  std::shared_ptr<IOTracer> io_tracer_;
+  FileSystemPtr fs_;
   // env_option optimized for compaction table reads
   FileOptions file_options_for_read_;
   VersionSet* versions_;
   const std::atomic<bool>* shutting_down_;
-  const std::atomic<bool>* manual_compaction_paused_;
+  const std::atomic<int>* manual_compaction_paused_;
   const SequenceNumber preserve_deletes_seqnum_;
   LogBuffer* log_buffer_;
-  Directory* db_directory_;
-  Directory* output_directory_;
+  FSDirectory* db_directory_;
+  FSDirectory* output_directory_;
+  FSDirectory* blob_output_directory_;
   Statistics* stats_;
   InstrumentedMutex* db_mutex_;
   ErrorHandler* db_error_handler_;
@@ -193,6 +204,9 @@ class CompactionJob {
   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_;
 };
 
 }  // 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"
@@ -52,6 +51,7 @@
 #include "test_util/sync_point.h"
 #include "test_util/testharness.h"
 #include "test_util/testutil.h"
+#include "util/cast_util.h"
 #include "util/compression.h"
 #include "util/hash.h"
 #include "util/mutexlock.h"
@@ -126,9 +126,7 @@ class CompactionJobStatsTest : public testing::Test,
   static void SetUpTestCase() {}
   static void TearDownTestCase() {}
 
-  DBImpl* dbfull() {
-    return reinterpret_cast<DBImpl*>(db_);
-  }
+  DBImpl* dbfull() { return static_cast_with_check<DBImpl>(db_); }
 
   void CreateColumnFamilies(const std::vector<std::string>& cfs,
                             const Options& options) {
@@ -299,15 +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,
@@ -460,6 +457,7 @@ class CompactionJobStatsChecker : public EventListener {
     ASSERT_EQ(current_stats.num_output_files,
         stats.num_output_files);
 
+    ASSERT_EQ(current_stats.is_full_compaction, stats.is_full_compaction);
     ASSERT_EQ(current_stats.is_manual_compaction,
         stats.is_manual_compaction);
 
@@ -572,7 +570,7 @@ CompactionJobStats NewManualCompactionJobStats(
     uint64_t num_input_records, size_t key_size, size_t value_size,
     size_t num_output_files, uint64_t num_output_records,
     double compression_ratio, uint64_t num_records_replaced,
-    bool is_manual = true) {
+    bool is_full = false, bool is_manual = true) {
   CompactionJobStats stats;
   stats.Reset();
 
@@ -596,6 +594,7 @@ CompactionJobStats NewManualCompactionJobStats(
   stats.total_input_raw_value_bytes =
       num_input_records * value_size;
 
+  stats.is_full_compaction = is_full;
   stats.is_manual_compaction = is_manual;
 
   stats.num_records_replaced = num_records_replaced;
@@ -797,7 +796,7 @@ TEST_P(CompactionJobStatsTest, CompactionJobStatsTest) {
     }
 
     ASSERT_OK(Flush(1));
-    reinterpret_cast<DBImpl*>(db_)->TEST_WaitForCompact();
+    ASSERT_OK(static_cast_with_check<DBImpl>(db_)->TEST_WaitForCompact());
 
     stats_checker->set_verify_next_comp_io_stats(true);
     std::atomic<bool> first_prepare_write(true);
@@ -895,7 +894,7 @@ TEST_P(CompactionJobStatsTest, DeletionStatsTest) {
   CompactRangeOptions cr_options;
   cr_options.change_level = true;
   cr_options.target_level = 2;
-  db_->CompactRange(cr_options, handles_[1], nullptr, nullptr);
+  ASSERT_OK(db_->CompactRange(cr_options, handles_[1], nullptr, nullptr));
   ASSERT_GT(NumTableFilesAtLevel(2, 1), 0);
 
   // Stage 2: Generate files including keys from the entire key range
@@ -982,26 +981,21 @@ TEST_P(CompactionJobStatsTest, UniversalCompactionTest) {
     if (num_input_units == 0) {
       continue;
     }
+    // A full compaction only happens when the number of flushes equals to
+    // the number of compaction input runs.
+    bool is_full = num_flushes == num_input_units;
     // The following statement determines the expected smallest key
-    // based on whether it is a full compaction.  A full compaction only
-    // happens when the number of flushes equals to the number of compaction
-    // input runs.
-    uint64_t smallest_key =
-        (num_flushes == num_input_units) ?
-            key_base : key_base * (num_flushes - 1);
+    // based on whether it is a full compaction.
+    uint64_t smallest_key = is_full ? key_base : key_base * (num_flushes - 1);
 
-    stats_checker->AddExpectedStats(
-        NewManualCompactionJobStats(
-            Key(smallest_key, 10),
-            Key(smallest_key + key_base * num_input_units - key_interval, 10),
-            num_input_units,
-            num_input_units > 2 ? num_input_units / 2 : 0,
-            num_keys_per_table * num_input_units,
-            kKeySize, kValueSize,
-            num_input_units,
-            num_keys_per_table * num_input_units,
-            1.0, 0, false));
-    dbfull()->TEST_WaitForCompact();
+    stats_checker->AddExpectedStats(NewManualCompactionJobStats(
+        Key(smallest_key, 10),
+        Key(smallest_key + key_base * num_input_units - key_interval, 10),
+        num_input_units, num_input_units > 2 ? num_input_units / 2 : 0,
+        num_keys_per_table * num_input_units, kKeySize, kValueSize,
+        num_input_units, num_keys_per_table * num_input_units, 1.0, 0, is_full,
+        false));
+    ASSERT_OK(dbfull()->TEST_WaitForCompact());
   }
   ASSERT_EQ(stats_checker->NumberOfUnverifiedStats(), 3U);
 
@@ -1012,7 +1006,7 @@ TEST_P(CompactionJobStatsTest, UniversalCompactionTest) {
         &rnd, start_key, start_key + key_base - 1,
         kKeySize, kValueSize, key_interval,
         compression_ratio, 1);
-    reinterpret_cast<DBImpl*>(db_)->TEST_WaitForCompact();
+    ASSERT_OK(static_cast_with_check<DBImpl>(db_)->TEST_WaitForCompact());
   }
   ASSERT_EQ(stats_checker->NumberOfUnverifiedStats(), 0U);
 }

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>
@@ -12,15 +14,15 @@
 #include <string>
 #include <tuple>
 
-#include "db/blob_index.h"
+#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"
@@ -67,30 +69,37 @@ 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_(),
         table_cache_(NewLRUCache(50000, 16)),
         write_buffer_manager_(db_options_.db_write_buffer_size),
-        versions_(new VersionSet(dbname_, &db_options_, env_options_,
-                                 table_cache_.get(), &write_buffer_manager_,
-                                 &write_controller_,
-                                 /*block_cache_tracer=*/nullptr)),
+        versions_(new VersionSet(
+            dbname_, &db_options_, env_options_, table_cache_.get(),
+            &write_buffer_manager_, &write_controller_,
+            /*block_cache_tracer=*/nullptr, /*io_tracer=*/nullptr)),
         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) {
@@ -101,9 +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,
@@ -129,7 +139,7 @@ class CompactionJobTest : public testing::Test {
     return blob_index;
   }
 
-  void AddMockFile(const stl_wrappers::KVMap& contents, int level = 0) {
+  void AddMockFile(const mock::KVVector& contents, int level = 0) {
     assert(contents.size() > 0);
 
     bool first_key = true;
@@ -143,7 +153,8 @@ class CompactionJobTest : public testing::Test {
       std::string skey;
       std::string value;
       std::tie(skey, value) = kv;
-      bool parsed = 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);
@@ -161,7 +172,7 @@ class CompactionJobTest : public testing::Test {
 
       first_key = false;
 
-      if (parsed && key.type == kTypeBlobIndex) {
+      if (pik_status.ok() && key.type == kTypeBlobIndex) {
         BlobIndex blob_index;
         const Status s = blob_index.DecodeFrom(value);
         if (!s.ok()) {
@@ -191,8 +202,9 @@ class CompactionJobTest : public testing::Test {
                  kUnknownFileChecksum, kUnknownFileChecksumFuncName);
 
     mutex_.Lock();
-    versions_->LogAndApply(versions_->GetColumnFamilySet()->GetDefault(),
-                           mutable_cf_options_, &edit, &mutex_);
+    EXPECT_OK(
+        versions_->LogAndApply(versions_->GetColumnFamilySet()->GetDefault(),
+                               mutable_cf_options_, &edit, &mutex_));
     mutex_.Unlock();
   }
 
@@ -203,11 +215,11 @@ class CompactionJobTest : public testing::Test {
   }
 
   // returns expected result after compaction
-  stl_wrappers::KVMap CreateTwoFiles(bool gen_corrupted_keys) {
-    auto expected_results = mock::MakeMockFile();
-    const int kKeysPerFile = 10000;
-    const int kCorruptKeysPerFile = 200;
-    const int kMatchingKeys = kKeysPerFile / 2;
+  mock::KVVector CreateTwoFiles(bool gen_corrupted_keys) {
+    stl_wrappers::KVMap expected_results;
+    constexpr int kKeysPerFile = 10000;
+    constexpr int kCorruptKeysPerFile = 200;
+    constexpr int kMatchingKeys = kKeysPerFile / 2;
     SequenceNumber sequence_number = 0;
 
     auto corrupt_id = [&](int id) {
@@ -230,49 +242,50 @@ class CompactionJobTest : public testing::Test {
           test::CorruptKeyType(&internal_key);
           test::CorruptKeyType(&bottommost_internal_key);
         }
-        contents.insert({ internal_key.Encode().ToString(), value });
+        contents.push_back({internal_key.Encode().ToString(), value});
         if (i == 1 || k < kMatchingKeys || corrupt_id(k - kMatchingKeys)) {
           expected_results.insert(
-              { bottommost_internal_key.Encode().ToString(), value });
+              {bottommost_internal_key.Encode().ToString(), value});
         }
       }
+      mock::SortKVVector(&contents, ucmp_);
 
       AddMockFile(contents);
     }
 
     SetLastSequence(sequence_number);
 
-    return expected_results;
+    mock::KVVector expected_results_kvvector;
+    for (auto& kv : expected_results) {
+      expected_results_kvvector.push_back({kv.first, kv.second});
+    }
+
+    return expected_results_kvvector;
   }
 
   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));
+    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;
@@ -281,21 +294,22 @@ class CompactionJobTest : public testing::Test {
     }
     ASSERT_OK(s);
     // Make "CURRENT" file that points to the new manifest file.
-    s = SetCurrentFile(env_, dbname_, 1, nullptr);
+    s = SetCurrentFile(fs_.get(), dbname_, 1, nullptr);
+
+    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();
   }
 
   void RunCompaction(
       const std::vector<std::vector<FileMetaData*>>& input_files,
-      const stl_wrappers::KVMap& expected_results,
+      const mock::KVVector& expected_results,
       const std::vector<SequenceNumber>& snapshots = {},
       SequenceNumber earliest_write_conflict_snapshot = kMaxSequenceNumber,
       int output_level = 1, bool verify = true,
@@ -314,11 +328,12 @@ class CompactionJobTest : public testing::Test {
       num_input_files += level_files.size();
     }
 
-    Compaction compaction(cfd->current()->storage_info(), *cfd->ioptions(),
-                          *cfd->GetLatestMutableCFOptions(),
-                          compaction_input_files, output_level, 1024 * 1024,
-                          10 * 1024 * 1024, 0, kNoCompression,
-                          cfd->ioptions()->compression_opts, 0, {}, true);
+    Compaction compaction(
+        cfd->current()->storage_info(), *cfd->ioptions(),
+        *cfd->GetLatestMutableCFOptions(), mutable_db_options_,
+        compaction_input_files, output_level, 1024 * 1024, 10 * 1024 * 1024, 0,
+        kNoCompression, cfd->GetLatestMutableCFOptions()->compression_opts, 0,
+        {}, true);
     compaction.SetInputVersion(cfd->current());
 
     LogBuffer log_buffer(InfoLogLevel::INFO_LEVEL, db_options_.info_log.get());
@@ -326,22 +341,27 @@ 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);
+        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();
     ASSERT_OK(compaction_job.Install(*cfd->GetLatestMutableCFOptions()));
+    ASSERT_OK(compaction_job.io_status());
     mutex_.Unlock();
 
     if (verify) {
@@ -366,10 +386,12 @@ 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_;
   MutableCFOptions mutable_cf_options_;
+  MutableDBOptions mutable_db_options_;
   std::shared_ptr<Cache> table_cache_;
   WriteController write_controller_;
   WriteBufferManager write_buffer_manager_;
@@ -383,6 +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) {
@@ -395,7 +428,7 @@ TEST_F(CompactionJobTest, Simple) {
   RunCompaction({ files }, expected_results);
 }
 
-TEST_F(CompactionJobTest, SimpleCorrupted) {
+TEST_F(CompactionJobTest, DISABLED_SimpleCorrupted) {
   NewDB();
 
   auto expected_results = CreateTwoFiles(true);
@@ -636,7 +669,7 @@ TEST_F(CompactionJobTest, FilterAllMergeOperands) {
   SetLastSequence(11U);
   auto files = cfd_->current()->storage_info()->LevelFiles(0);
 
-  stl_wrappers::KVMap empty_map;
+  mock::KVVector empty_map;
   RunCompaction({files}, empty_map);
 }
 
@@ -989,7 +1022,7 @@ TEST_F(CompactionJobTest, MultiSingleDelete) {
 // single deletion and the (single) deletion gets removed while the corrupt key
 // gets written out. TODO(noetzli): We probably want a better way to treat
 // corrupt keys.
-TEST_F(CompactionJobTest, CorruptionAfterDeletion) {
+TEST_F(CompactionJobTest, DISABLED_CorruptionAfterDeletion) {
   NewDB();
 
   auto file1 =
@@ -1063,6 +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

@@ -110,9 +110,9 @@ CompressionType GetCompressionType(const ImmutableCFOptions& ioptions,
 
   // If bottommost_compression is set and we are compacting to the
   // bottommost level then we should use it.
-  if (ioptions.bottommost_compression != kDisableCompressionOption &&
+  if (mutable_cf_options.bottommost_compression != kDisableCompressionOption &&
       level >= (vstorage->num_non_empty_levels() - 1)) {
-    return ioptions.bottommost_compression;
+    return mutable_cf_options.bottommost_compression;
   }
   // If the user has specified a different compression level for each level,
   // then pick the compression for that level.
@@ -132,22 +132,20 @@ CompressionType GetCompressionType(const ImmutableCFOptions& ioptions,
   }
 }
 
-CompressionOptions GetCompressionOptions(const ImmutableCFOptions& ioptions,
+CompressionOptions GetCompressionOptions(const MutableCFOptions& cf_options,
                                          const VersionStorageInfo* vstorage,
                                          int level,
                                          const bool enable_compression) {
   if (!enable_compression) {
-    return ioptions.compression_opts;
+    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 (ioptions.bottommost_compression != kDisableCompressionOption &&
-      level >= (vstorage->num_non_empty_levels() - 1) &&
-      ioptions.bottommost_compression_opts.enabled) {
-    return ioptions.bottommost_compression_opts;
+  // 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;
   }
-  return ioptions.compression_opts;
+  return cf_options.compression_opts;
 }
 
 CompactionPicker::CompactionPicker(const ImmutableCFOptions& ioptions,
@@ -332,7 +330,7 @@ Compaction* CompactionPicker::CompactFiles(
     const CompactionOptions& compact_options,
     const std::vector<CompactionInputFiles>& input_files, int output_level,
     VersionStorageInfo* vstorage, const MutableCFOptions& mutable_cf_options,
-    uint32_t output_path_id) {
+    const MutableDBOptions& mutable_db_options, uint32_t output_path_id) {
   assert(input_files.size());
   // This compaction output should not overlap with a running compaction as
   // `SanitizeCompactionInputFiles` should've checked earlier and db mutex
@@ -356,10 +354,10 @@ Compaction* CompactionPicker::CompactFiles(
     compression_type = compact_options.compression;
   }
   auto c = new Compaction(
-      vstorage, ioptions_, mutable_cf_options, input_files, output_level,
-      compact_options.output_file_size_limit,
+      vstorage, ioptions_, mutable_cf_options, mutable_db_options, input_files,
+      output_level, compact_options.output_file_size_limit,
       mutable_cf_options.max_compaction_bytes, output_path_id, compression_type,
-      GetCompressionOptions(ioptions_, vstorage, output_level),
+      GetCompressionOptions(mutable_cf_options, vstorage, output_level),
       compact_options.max_subcompactions,
       /* grandparents */ {}, true);
   RegisterCompaction(c);
@@ -563,7 +561,8 @@ void CompactionPicker::GetGrandparents(
 
 Compaction* CompactionPicker::CompactRange(
     const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
-    VersionStorageInfo* vstorage, int input_level, int output_level,
+    const MutableDBOptions& mutable_db_options, VersionStorageInfo* vstorage,
+    int input_level, int output_level,
     const CompactRangeOptions& compact_range_options, const InternalKey* begin,
     const InternalKey* end, InternalKey** compaction_end, bool* manual_conflict,
     uint64_t max_file_num_to_ignore) {
@@ -626,18 +625,19 @@ Compaction* CompactionPicker::CompactRange(
     }
 
     Compaction* c = new Compaction(
-        vstorage, ioptions_, mutable_cf_options, std::move(inputs),
-        output_level,
+        vstorage, ioptions_, mutable_cf_options, mutable_db_options,
+        std::move(inputs), output_level,
         MaxFileSizeForLevel(mutable_cf_options, output_level,
                             ioptions_.compaction_style),
         /* max_compaction_bytes */ LLONG_MAX,
         compact_range_options.target_path_id,
         GetCompressionType(ioptions_, vstorage, mutable_cf_options,
                            output_level, 1),
-        GetCompressionOptions(ioptions_, vstorage, output_level),
+        GetCompressionOptions(mutable_cf_options, vstorage, output_level),
         compact_range_options.max_subcompactions, /* grandparents */ {},
         /* is manual */ true);
     RegisterCompaction(c);
+    vstorage->ComputeCompactionScore(ioptions_, mutable_cf_options);
     return c;
   }
 
@@ -778,8 +778,8 @@ Compaction* CompactionPicker::CompactRange(
   std::vector<FileMetaData*> grandparents;
   GetGrandparents(vstorage, inputs, output_level_inputs, &grandparents);
   Compaction* compaction = new Compaction(
-      vstorage, ioptions_, mutable_cf_options, std::move(compaction_inputs),
-      output_level,
+      vstorage, ioptions_, mutable_cf_options, mutable_db_options,
+      std::move(compaction_inputs), output_level,
       MaxFileSizeForLevel(mutable_cf_options, output_level,
                           ioptions_.compaction_style, vstorage->base_level(),
                           ioptions_.level_compaction_dynamic_level_bytes),
@@ -787,7 +787,7 @@ Compaction* CompactionPicker::CompactRange(
       compact_range_options.target_path_id,
       GetCompressionType(ioptions_, vstorage, mutable_cf_options, output_level,
                          vstorage->base_level()),
-      GetCompressionOptions(ioptions_, vstorage, output_level),
+      GetCompressionOptions(mutable_cf_options, vstorage, output_level),
       compact_range_options.max_subcompactions, std::move(grandparents),
       /* is manual compaction */ true);
 
@@ -1004,6 +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)) {
@@ -1013,6 +1014,7 @@ Status CompactionPicker::SanitizeCompactionInputFiles(
                                    " is already being compacted.");
           }
           found = true;
+          input_file_level = level_meta.level;
           break;
         }
       }
@@ -1025,6 +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();
@@ -1043,6 +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) {
@@ -1085,6 +1096,8 @@ void CompactionPicker::PickFilesMarkedForCompaction(
   Random64 rnd(/* seed */ reinterpret_cast<uint64_t>(vstorage));
   size_t random_file_index = static_cast<size_t>(rnd.Uniform(
       static_cast<uint64_t>(vstorage->FilesMarkedForCompaction().size())));
+  TEST_SYNC_POINT_CALLBACK("CompactionPicker::PickFilesMarkedForCompaction",
+                           &random_file_index);
 
   if (continuation(vstorage->FilesMarkedForCompaction()[random_file_index])) {
     // found the compaction!