add missing release notes to HISTORY.md

Add tests for paranoid checks with range deletion (#7521 )
Summary: Added unit tests that have paranoid_check = true that perform range deletions. At the moment, the deleted ranges do not appear to be checked as part of the paranoid checks. Pull Request resolved: https://github.com/facebook/rocksdb/pull/7521 Reviewed By: zhichao-cao Differential Revision: D24262175 Pulled By: ajkr fbshipit-source-id: 1035e968f7ab8ccaa7af086b835a4e72c7e56743
2026-07-07 14:47:40 +08:00 · 2020-10-13 10:39:10 -07:00 · 2020-10-13 10:20:36 -07:00 · 2020-10-12 15:18:00 -07:00 · 2020-10-12 14:10:14 -07:00 · 2020-10-12 11:22:25 -07:00
1230 changed files with 124301 additions and 42716 deletions
@@ -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
@@ -0,0 +1,459 @@
+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
+
+  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:
+    steps:
+      - checkout
+      - run: pyenv install --skip-existing 3.5.9
+      - run: pyenv global 3.5.9
+      - 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" >> $BASH_ENV
+
+  post-steps:
+    steps:
+      - slack/status: *notify-on-master-failure
+      - store_test_results: # store test result if there's any
+          path: /tmp/test-results
+      - store_artifacts: # store LOG for debugging if there's any
+          path: LOG
+
+  install-clang-10:
+    steps:
+      - run:
+          name: Install Clang 10
+          command: |
+            echo "deb http://apt.llvm.org/xenial/ llvm-toolchain-xenial-10 main" | sudo tee -a /etc/apt/sources.list
+            echo "deb-src http://apt.llvm.org/xenial/ llvm-toolchain-xenial-10 main" | sudo tee -a /etc/apt/sources.list
+            echo "APT::Acquire::Retries \"10\";" | sudo tee -a /etc/apt/apt.conf.d/80-retries # llvm.org unreliable
+            sudo apt-get update -y && sudo apt-get install -y clang-10
+
+  install-gflags:
+    steps:
+      - run:
+          name: Install gflags
+          command: |
+            sudo apt-get update -y && sudo apt-get install -y libgflags-dev
+  
+  install-gflags-on-macos:
+    steps:
+      - run:
+          name: Install gflags on macos
+          command: |
+            HOMEBREW_NO_AUTO_UPDATE=1 brew install gflags
+
+  install-gtest-parallel:
+    steps:
+      - run:
+          name: Install gtest-parallel
+          command: |
+            git clone --single-branch --branch master --depth 1 https://github.com/google/gtest-parallel.git ~/gtest-parallel
+            echo "export PATH=$HOME/gtest-parallel:$PATH" >> $BASH_ENV
+
+executors:
+  windows-2xlarge:
+    machine:
+      image: 'windows-server-2019-vs2019:stable'
+      resource_class: windows.2xlarge
+      shell: bash.exe
+
+jobs:
+  build-macos:
+    macos:
+      xcode: 11.3.0
+    steps:
+      - increase-max-open-files-on-macos
+      - install-pyenv-on-macos
+      - pre-steps
+      - install-gflags-on-macos
+      - run: ulimit -S -n 1048576 && OPT=-DCIRCLECI make V=1 J=32 -j32 check | .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-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 LIB_MODE=shared OPT="-DROCKSDB_NAMESPACE=alternative_rocksdb_ns" make V=1 -j32 all check_some | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-linux-release:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - checkout # check out the code in the project directory
+      - run: make V=1 -j32 release | .circleci/cat_ignore_eagain
+      - run: if ./db_stress --version; then false; else true; fi # ensure without gflags
+      - install-gflags
+      - run: make V=1 -j32 release | .circleci/cat_ignore_eagain
+      - run: ./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 -j16 static_lib tools db_bench | .circleci/cat_ignore_eagain
+      - run: if ./db_stress --version; then false; else true; fi # ensure without gflags
+      - run: sudo apt-get update -y && sudo apt-get install -y libgflags-dev
+      - run: make clean
+      - run: USE_RTTI=1 DEBUG_LEVEL=0 make V=1 -j16 static_lib tools db_bench | .circleci/cat_ignore_eagain
+      - run: ./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 -j32 release | .circleci/cat_ignore_eagain
+      - run: if ./db_stress --version; then false; else true; fi # ensure without gflags
+      - install-gflags
+      - run: LITE=1 make V=1 -j32 release | .circleci/cat_ignore_eagain
+      - run: ./db_stress --version # ensure with gflags
+      - post-steps
+
+  build-linux-clang-no-test:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - checkout # check out the code in the project directory
+      - run: sudo apt-get update -y && sudo apt-get install -y clang libgflags-dev
+      - run: CC=clang CXX=clang++ USE_CLANG=1 PORTABLE=1 make V=1 -j32 all | .circleci/cat_ignore_eagain
+      - 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:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: 2xlarge
+    steps:
+      - checkout # check out the code in the project directory
+      - run: (mkdir build && cd build && cmake -DWITH_GFLAGS=0 .. && make V=1 -j32) | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-linux-unity:
+    docker: # executor type
+      - image: gcc:latest
+    resource_class: xlarge
+    steps:
+      - checkout # check out the code in the project directory
+      - run: apt-get update -y && apt-get install -y libgflags-dev
+      - run: TEST_TMPDIR=/dev/shm && make V=1 -j16 unity_test | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-linux-gcc-4-8:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: large
+    steps:
+      - pre-steps
+      - run: sudo apt-get update -y && sudo apt-get install gcc-4.8 g++-4.8 libgflags-dev
+      - run: CC=gcc-4.8 CXX=g++-4.8 V=1 SKIP_LINK=1 make -j4 all | .circleci/cat_ignore_eagain # Linking broken because libgflags compiled with newer ABI
+      - 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}" == "2017" ]]; then
+              powershell .circleci/vs2017_install.ps1
+            elif [[ "${VS_YEAR}" == "2015" ]]; then
+              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 ..
+            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: 2xlarge
+    steps:
+      - pre-steps
+      - install-gflags
+      - run:
+          name: "Build RocksDBJava"
+          command: |
+            export JAVA_HOME=/usr/lib/jvm/java-1.8.0-openjdk-amd64
+            export PATH=$JAVA_HOME/bin:$PATH
+            echo "JAVA_HOME=${JAVA_HOME}"
+            which java && java -version
+            which javac && javac -version
+            make V=1 J=32 -j32 rocksdbjava jtest | .circleci/cat_ignore_eagain
+      - post-steps
+
+  build-examples:
+    machine:
+      image: ubuntu-1604:202007-01
+    resource_class: medium
+    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-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
+            /usr/bin/python ../gtest-parallel/gtest-parallel $(</tmp/test_list) --output_dir=/tmp | cat  # pipe to cat to continuously output status on circleci UI. Otherwise, no status will be printed while the job is running.
+      - post-steps
+
+workflows:
+  build-linux:
+    jobs:
+      - build-linux
+  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-clang-no-test:
+    jobs:
+      - build-linux-clang-no-test
+  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-cmake:
+    jobs:
+      - build-linux-cmake
+  build-linux-unity:
+    jobs:
+      - build-linux-unity
+  build-windows:
+    jobs:
+      - build-windows
+  build-windows-vs2017:
+    jobs:
+      - build-windows:
+          vs_year: "2017"
+          cmake_generator: "Visual Studio 15 Win64"
+  build-windows-vs2015:
+    jobs:
+      - build-windows:
+          vs_year: "2015"
+          cmake_generator: "Visual Studio 14 Win64"
+  build-windows-cxx20:
+    jobs:
+      - build-windows:
+          extra_cmake_opt: -DCMAKE_CXX_STANDARD=20
+  build-java:
+    jobs:
+      - build-linux-java
+  build-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-gcc-4-8:
+    jobs:
+      - build-linux-gcc-4-8
+  build-macos:
+    jobs:
+      - build-macos
@@ -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
@@ -0,0 +1,23 @@
+$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 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 "${PWD}\Collect.exe" -NoNewWindow -Wait -PassThru
+    New-Item -Path "C:\w\build-results" -ItemType "directory" -Force
+    Copy-Item -Path "C:\Users\circleci\AppData\Local\Temp\vslogs.zip" -Destination "C:\w\build-results\"
+    exit 1
+}
@@ -0,0 +1,34 @@
+$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
+}
@@ -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-mirror/clang/master/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
@@ -1,4 +1,5 @@
 make_config.mk
+rocksdb.pc

 *.a
 *.arc
@@ -34,6 +35,7 @@ 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
@@ -51,7 +53,10 @@ db_test2
 trace_analyzer
 trace_analyzer_test
 block_cache_trace_analyzer
+io_tracer_parser
 .DS_Store
+.vs
+.vscode

 java/out
 java/target
@@ -79,3 +84,8 @@ tp2/
 fbcode/
 fbcode
 buckifier/*.pyc
+buckifier/__pycache__
+
+compile_commands.json
+clang-format-diff.py
+.py3/
@@ -3,28 +3,39 @@ language: cpp
 os:
  - linux
  - osx
+arch:
+  - amd64
+  - arm64
+  - ppc64le
 compiler:
  - clang
  - gcc
-osx_image: xcode8.3
-jdk:
-  - openjdk7
+osx_image: xcode9.4
 cache:
  - ccache
-  - apt

 addons:
  apt:
    sources:
      - ubuntu-toolchain-r-test
    packages:
-      - curl
-      - g++-8
-      - libbz2-dev
      - libgflags-dev
+      - libbz2-dev
+      - liblz4-dev
      - libsnappy-dev
-      - mingw-w64
+      - 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
  - TEST_GROUP=1 # 33-35 minutes
@@ -39,41 +50,213 @@ 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: osx
+    env: JOB_NAME=cmake-gcc9
+  - os: osx
+    env: JOB_NAME=cmake-gcc9-c++20
+  - os: osx
+    env: JOB_NAME=cmake-mingw
+  - os: osx
+    env: JOB_NAME=make-gcc4.8
+  - os: osx
+    arch: ppc64le
+  - os: osx
+    compiler: gcc
+  - os : linux
+    arch: arm64
    env: JOB_NAME=cmake-mingw
  - 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
+  # Exclude all but most unique cmake variants for pull requests, but build all in branches
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: amd64
+    env: JOB_NAME=cmake
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: amd64
+    env: JOB_NAME=cmake-gcc8
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: amd64
+    env: JOB_NAME=cmake-gcc9
+  # Exclude most osx, arm64 and ppc64le tests for pull requests, but build in branches
+  # Temporarily disable ppc64le cmake test while snapd is broken
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: JOB_NAME=cmake
+  # NB: the cmake build is a partial java test
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: osx
+    env: TEST_GROUP=1
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: TEST_GROUP=1
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: TEST_GROUP=1
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: osx
+    env: TEST_GROUP=2
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: TEST_GROUP=2
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: TEST_GROUP=2
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: osx
+    env: TEST_GROUP=3
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: TEST_GROUP=3
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: TEST_GROUP=3
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: osx
+    env: TEST_GROUP=4
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os : linux
+    arch: arm64
+    env: TEST_GROUP=4
+  - if: type = pull_request AND commit_message !~ /FULL_CI/
+    os: linux
+    arch: ppc64le
+    env: TEST_GROUP=4
+  - if: type = pull_request AND commit_message !~ /FULL_CI/ AND commit_message !~ /java/
+    os : osx
+    env: JOB_NAME=java_test
+  - if: type = pull_request AND commit_message !~ /FULL_CI/ AND commit_message !~ /java/
+    os : linux
+    arch: arm64
+    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 : osx
+    env: JOB_NAME=lite_build
+  - 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 : osx
+    env: JOB_NAME=examples
+  - 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 : osx
+    env: JOB_NAME=status_checked
+  - 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

-# https://docs.travis-ci.com/user/caching/#ccache-cache
 install:
  - if [ "${TRAVIS_OS_NAME}" == osx ]; then
-      brew install ccache zstd lz4 snappy xz;
      PATH=$PATH:/usr/local/opt/ccache/libexec;
    fi
  - if [ "${JOB_NAME}" == cmake-gcc8 ]; then
+      sudo apt-get install -y g++-8 || exit $?;
      CC=gcc-8 && CXX=g++-8;
    fi
-  - if [[ "${JOB_NAME}" == cmake* ]] && [ "${TRAVIS_OS_NAME}" == linux ]; then
-      mkdir cmake-dist && curl -sfSL https://github.com/Kitware/CMake/releases/download/v3.14.5/cmake-3.14.5-Linux-x86_64.tar.gz | tar --strip-components=1 -C cmake-dist -xz && export PATH=$PWD/cmake-dist/bin:$PATH;
+  - if [ "${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}" == java_test ]]; then
-      java -version && echo "JAVA_HOME=${JAVA_HOME}";
+  - if [ "${JOB_NAME}" == cmake-mingw ]; then
+      sudo apt-get install -y mingw-w64 || exit $?;
+    fi
+  - if [ "${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}" == cmake* ]] && [ "${TRAVIS_OS_NAME}" == linux ]; then
+      sudo apt-get install snapd && sudo snap install cmake --beta --classic || exit $?;
+      export PATH=/snap/bin:$PATH;
+    fi
+  - |
+    if [[ "${JOB_NAME}" == java_test || "${JOB_NAME}" == cmake* ]]; then
+      # Ensure JDK 8
+      if [ "${TRAVIS_OS_NAME}" == osx ]; then
+        brew tap AdoptOpenJDK/openjdk || exit $?
+        brew cask install adoptopenjdk8 || exit $?
+        export JAVA_HOME=$(/usr/libexec/java_home)
+      else
+        sudo apt-get install -y openjdk-8-jdk || exit $?
+        export PATH=/usr/lib/jvm/java-8-openjdk-$(dpkg --print-architecture)/bin:$PATH
+        export JAVA_HOME=/usr/lib/jvm/java-8-openjdk-$(dpkg --print-architecture)
+      fi
+      echo "JAVA_HOME=${JAVA_HOME}"
+      which java && java -version
+      which javac && javac -version
    fi

 before_script:
@@ -82,23 +265,23 @@ 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 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=full_filter_block_test make -j4 check_some
+      OPT=-DTRAVIS V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=exclude ROCKSDBTESTS_END=backupable_db_test make -j4 check_some
      ;;
    2)
-      OPT=-DTRAVIS V=1 make -j4 tools && OPT=-DTRAVIS V=1 ROCKSDBTESTS_START=full_filter_block_test ROCKSDBTESTS_END=write_batch_with_index_test make -j4 check_some
+      OPT="-DTRAVIS -DROCKSDB_NAMESPACE=alternative_rocksdb_ns" V=1 make -j4 tools && OPT="-DTRAVIS -DROCKSDB_NAMESPACE=alternative_rocksdb_ns" V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=exclude ROCKSDBTESTS_START=backupable_db_test ROCKSDBTESTS_END=db_universal_compaction_test make -j4 check_some
      ;;
    3)
-      OPT=-DTRAVIS V=1 ROCKSDBTESTS_START=write_batch_with_index_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 V=1 ROCKSDBTESTS_PLATFORM_DEPENDENT=exclude ROCKSDBTESTS_START=table_properties_collector_test make -j4 check_some
      ;;
    esac
  - case $JOB_NAME in
@@ -106,21 +289,31 @@ script:
      OPT=-DTRAVIS V=1 make rocksdbjava jtest
      ;;
    lite_build)
-      OPT='-DTRAVIS -DROCKSDB_LITE' V=1 make -j4 static_lib tools
+      OPT='-DTRAVIS -DROCKSDB_LITE' V=1 make -j4 all
      ;;
    examples)
      OPT=-DTRAVIS 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 .. -DCMAKE_C_COMPILER=x86_64-w64-mingw32-gcc -DCMAKE_CXX_COMPILER=x86_64-w64-mingw32-g++ -DCMAKE_SYSTEM_NAME=Windows && make -j4 rocksdb rocksdbjni
+      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 V=1 SKIP_LINK=1 make -j4 all && [ "Linking broken because libgflags compiled with newer ABI" ]
+      ;;
+    status_checked)
+      OPT=-DTRAVIS V=1 ASSERT_STATUS_CHECKED=1 make -j4 check_some
      ;;
    esac
 notifications:
    email:
      - leveldb@fb.com
-    webhooks:
-      - https://buildtimetrend.herokuapp.com/travis
@@ -14,7 +14,7 @@
 #        cd build
 # 3. Run cmake to generate project files for Windows, add more options to enable required third-party libraries.
 #    See thirdparty.inc for more information.
-#        sample command: cmake -G "Visual Studio 15 Win64" -DWITH_GFLAGS=1 -DWITH_SNAPPY=1 -DWITH_JEMALLOC=1 -DWITH_JNI=1 ..
+#        sample command: cmake -G "Visual Studio 15 Win64" -DCMAKE_BUILD_TYPE=Release -DWITH_GFLAGS=1 -DWITH_SNAPPY=1 -DWITH_JEMALLOC=1 -DWITH_JNI=1 ..
 # 4. Then build the project in debug mode (you may want to add /m[:<N>] flag to run msbuild in <N> parallel threads
 #                                          or simply /m to use all avail cores)
 #        msbuild rocksdb.sln
@@ -45,6 +45,16 @@ if(POLICY CMP0042)
  cmake_policy(SET CMP0042 NEW)
 endif()

+if(NOT CMAKE_BUILD_TYPE)
+  if(EXISTS "${CMAKE_SOURCE_DIR}/.git")
+    set(default_build_type "Debug")
+  else()
+    set(default_build_type "RelWithDebInfo")
+  endif()
+  set(CMAKE_BUILD_TYPE "${default_build_type}" CACHE STRING
+    "Default BUILD_TYPE is ${default_build_type}" FORCE)
+endif()
+
 find_program(CCACHE_FOUND ccache)
 if(CCACHE_FOUND)
  set_property(GLOBAL PROPERTY RULE_LAUNCH_COMPILE ccache)
@@ -62,19 +72,30 @@ if (WITH_WINDOWS_UTF8_FILENAMES)
 endif()
 # third-party/folly is only validated to work on Linux and Windows for now.
 # So only turn it on there by default.
-if(CMAKE_SYSTEM_NAME MATCHES "Linux" OR CMAKE_SYSTEM_NAME MATCHES "Windows")
-  option(WITH_FOLLY_DISTRIBUTED_MUTEX "build with folly::DistributedMutex" ON)
+if(CMAKE_SYSTEM_NAME MATCHES "Linux|Windows")
+  if(MSVC AND MSVC_VERSION LESS 1910)
+    # Folly does not compile with MSVC older than VS2017
+    option(WITH_FOLLY_DISTRIBUTED_MUTEX "build with folly::DistributedMutex" OFF)
+  else()
+    option(WITH_FOLLY_DISTRIBUTED_MUTEX "build with folly::DistributedMutex" ON)
+  endif()
 else()
  option(WITH_FOLLY_DISTRIBUTED_MUTEX "build with folly::DistributedMutex" OFF)
 endif()
+
+if( NOT DEFINED CMAKE_CXX_STANDARD )
+  set(CMAKE_CXX_STANDARD 11)
+endif()
+
+include(CMakeDependentOption)
+CMAKE_DEPENDENT_OPTION(WITH_GFLAGS "build with GFlags" ON
+  "NOT MSVC;NOT MINGW" OFF)
+
 if(MSVC)
-  # Defaults currently different for GFLAGS.
-  #  We will address find_package work a little later
-  option(WITH_GFLAGS "build with GFlags" OFF)
  option(WITH_XPRESS "build with windows built in compression" OFF)
  include(${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)
@@ -86,21 +107,35 @@ else()
    endif()
  endif()

-  # No config file for this
-  option(WITH_GFLAGS "build with GFlags" ON)
+  set(GFLAGS_LIB)
  if(WITH_GFLAGS)
-    find_package(gflags)
+    # 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)
-      add_definitions(-DGFLAGS=1)
-      include_directories(${gflags_INCLUDE_DIR})
-      list(APPEND THIRDPARTY_LIBS ${gflags_LIBRARIES})
+      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)
  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)
@@ -174,10 +209,9 @@ 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(MINGW)
-    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-format")
+    set(CMAKE_CXX_FLAGS "${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)
@@ -189,16 +223,25 @@ 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")
        CHECK_C_COMPILER_FLAG("-march=armv8-a+crc+crypto" HAS_ARMV8_CRC)
@@ -211,17 +254,35 @@ endif(CMAKE_SYSTEM_PROCESSOR MATCHES "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()
@@ -231,6 +292,7 @@ include(CheckCXXSourceCompiles)
 if(NOT MSVC)
  set(CMAKE_REQUIRED_FLAGS "-msse4.2 -mpclmul")
 endif()
+
 CHECK_CXX_SOURCE_COMPILES("
 #include <cstdint>
 #include <nmmintrin.h>
@@ -329,6 +391,12 @@ 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")
@@ -363,7 +431,15 @@ 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")
@@ -399,6 +475,8 @@ 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")
@@ -457,7 +535,11 @@ 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()
@@ -467,9 +549,13 @@ 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()
@@ -478,9 +564,19 @@ 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_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
@@ -492,6 +588,7 @@ 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/db_impl.cc
@@ -524,6 +621,8 @@ 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
@@ -534,7 +633,9 @@ 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
@@ -544,17 +645,26 @@ set(SOURCES
        env/env_chroot.cc
        env/env_encryption.cc
        env/env_hdfs.cc
+        env/file_system.cc
+        env/file_system_tracer.cc
        env/mock_env.cc
        file/delete_scheduler.cc
+        file/file_prefetch_buffer.cc
        file/file_util.cc
        file/filename.cc
+        file/random_access_file_reader.cc
+        file/read_write_util.cc
+        file/readahead_raf.cc
+        file/sequence_file_reader.cc
        file/sst_file_manager_impl.cc
+        file/writable_file_writer.cc
        logging/auto_roll_logger.cc
        logging/event_logger.cc
        logging/log_buffer.cc
        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
@@ -575,27 +685,37 @@ set(SOURCES
        monitoring/thread_status_util.cc
        monitoring/thread_status_util_debug.cc
        options/cf_options.cc
+        options/configurable.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
        table/block_based/filter_block_reader_common.cc
+        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
@@ -613,8 +733,10 @@ 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
@@ -622,15 +744,15 @@ set(SOURCES
        test_util/testutil.cc
        test_util/transaction_test_util.cc
        tools/block_cache_analyzer/block_cache_trace_analyzer.cc
-        tools/db_bench_tool.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
-        util/bloom.cc
+        trace_replay/io_tracer.cc
        util/coding.cc
        util/compaction_job_stats_impl.cc
        util/comparator.cc
@@ -638,13 +760,12 @@ set(SOURCES
        util/concurrent_task_limiter_impl.cc
        util/crc32c.cc
        util/dynamic_bloom.cc
-        util/file_reader_writer.cc
-        util/filter_policy.cc
        util/hash.cc
        util/murmurhash.cc
        util/random.cc
        util/rate_limiter.cc
        util/slice.cc
+        util/file_checksum_helper.cc
        util/status.cc
        util/string_util.cc
        util/thread_local.cc
@@ -657,9 +778,6 @@ 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
@@ -668,6 +786,8 @@ 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
@@ -689,6 +809,8 @@ set(SOURCES
        utilities/simulator_cache/sim_cache.cc
        utilities/table_properties_collectors/compact_on_deletion_collector.cc
        utilities/trace/file_trace_reader_writer.cc
+        utilities/transactions/lock/lock_tracker.cc
+        utilities/transactions/lock/point_lock_tracker.cc
        utilities/transactions/optimistic_transaction_db_impl.cc
        utilities/transactions/optimistic_transaction.cc
        utilities/transactions/pessimistic_transaction.cc
@@ -713,11 +835,11 @@ 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
@@ -730,9 +852,13 @@ 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)
@@ -747,6 +873,7 @@ else()
  list(APPEND SOURCES
    port/port_posix.cc
    env/env_posix.cc
+    env/fs_posix.cc
    env/io_posix.cc)
 endif()

@@ -761,7 +888,8 @@ endif()

 set(ROCKSDB_STATIC_LIB rocksdb${ARTIFACT_SUFFIX})
 set(ROCKSDB_SHARED_LIB rocksdb-shared${ARTIFACT_SUFFIX})
-set(ROCKSDB_IMPORT_LIB ${ROCKSDB_SHARED_LIB})
+
+option(ROCKSDB_BUILD_SHARED "Build shared versions of the RocksDB libraries" ON)

 option(WITH_LIBRADOS "Build with librados" OFF)
 if(WITH_LIBRADOS)
@@ -771,42 +899,55 @@ if(WITH_LIBRADOS)
 endif()

 if(WIN32)
-  set(SYSTEM_LIBS ${SYSTEM_LIBS} Shlwapi.lib Rpcrt4.lib)
-  set(LIBS ${ROCKSDB_STATIC_LIB} ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
+  set(SYSTEM_LIBS ${SYSTEM_LIBS} shlwapi.lib rpcrt4.lib)
 else()
  set(SYSTEM_LIBS ${CMAKE_THREAD_LIBS_INIT})
-  set(LIBS ${ROCKSDB_SHARED_LIB} ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
-
-  add_library(${ROCKSDB_SHARED_LIB} SHARED ${SOURCES})
-  target_link_libraries(${ROCKSDB_SHARED_LIB}
-    ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
-  set_target_properties(${ROCKSDB_SHARED_LIB} PROPERTIES
-                        LINKER_LANGUAGE CXX
-                        VERSION ${rocksdb_VERSION}
-                        SOVERSION ${rocksdb_VERSION_MAJOR}
-                        CXX_STANDARD 11
-                        OUTPUT_NAME "rocksdb")
 endif()

 add_library(${ROCKSDB_STATIC_LIB} STATIC ${SOURCES})
-target_link_libraries(${ROCKSDB_STATIC_LIB}
+target_link_libraries(${ROCKSDB_STATIC_LIB} PRIVATE
  ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})

-if(WIN32)
-  add_library(${ROCKSDB_IMPORT_LIB} SHARED ${SOURCES})
-  target_link_libraries(${ROCKSDB_IMPORT_LIB}
+if(ROCKSDB_BUILD_SHARED)
+  add_library(${ROCKSDB_SHARED_LIB} SHARED ${SOURCES})
+  target_link_libraries(${ROCKSDB_SHARED_LIB} PRIVATE
    ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
-  set_target_properties(${ROCKSDB_IMPORT_LIB} PROPERTIES
-    COMPILE_DEFINITIONS "ROCKSDB_DLL;ROCKSDB_LIBRARY_EXPORTS")
-  if(MSVC)
-    set_target_properties(${ROCKSDB_STATIC_LIB} PROPERTIES
-      COMPILE_FLAGS "/Fd${CMAKE_CFG_INTDIR}/${ROCKSDB_STATIC_LIB}.pdb")
-    set_target_properties(${ROCKSDB_IMPORT_LIB} PROPERTIES
-      COMPILE_FLAGS "/Fd${CMAKE_CFG_INTDIR}/${ROCKSDB_IMPORT_LIB}.pdb")
+
+  if(WIN32)
+    set_target_properties(${ROCKSDB_SHARED_LIB} PROPERTIES
+      COMPILE_DEFINITIONS "ROCKSDB_DLL;ROCKSDB_LIBRARY_EXPORTS")
+    if(MSVC)
+      set_target_properties(${ROCKSDB_STATIC_LIB} PROPERTIES
+        COMPILE_FLAGS "/Fd${CMAKE_CFG_INTDIR}/${ROCKSDB_STATIC_LIB}.pdb")
+      set_target_properties(${ROCKSDB_SHARED_LIB} PROPERTIES
+        COMPILE_FLAGS "/Fd${CMAKE_CFG_INTDIR}/${ROCKSDB_SHARED_LIB}.pdb")
+    endif()
+  else()
+    set_target_properties(${ROCKSDB_SHARED_LIB} PROPERTIES
+                          LINKER_LANGUAGE CXX
+                          VERSION ${rocksdb_VERSION}
+                          SOVERSION ${rocksdb_VERSION_MAJOR}
+                          OUTPUT_NAME "rocksdb")
  endif()
 endif()

+if(ROCKSDB_BUILD_SHARED AND NOT WIN32)
+  set(ROCKSDB_LIB ${ROCKSDB_SHARED_LIB})
+else()
+  set(ROCKSDB_LIB ${ROCKSDB_STATIC_LIB})
+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)
@@ -844,6 +985,8 @@ 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 ${ROCKSDB_STATIC_LIB}
    EXPORT RocksDBTargets
@@ -852,15 +995,17 @@ if(NOT WIN32 OR ROCKSDB_INSTALL_ON_WINDOWS)
    INCLUDES DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}"
  )

-  install(
-    TARGETS ${ROCKSDB_SHARED_LIB}
-    EXPORT RocksDBTargets
-    COMPONENT runtime
-    ARCHIVE DESTINATION "${CMAKE_INSTALL_LIBDIR}"
-    RUNTIME DESTINATION "${CMAKE_INSTALL_BINDIR}"
-    LIBRARY DESTINATION "${CMAKE_INSTALL_LIBDIR}"
-    INCLUDES DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}"
-  )
+  if(ROCKSDB_BUILD_SHARED)
+    install(
+      TARGETS ${ROCKSDB_SHARED_LIB}
+      EXPORT RocksDBTargets
+      COMPONENT runtime
+      ARCHIVE DESTINATION "${CMAKE_INSTALL_LIBDIR}"
+      RUNTIME DESTINATION "${CMAKE_INSTALL_BINDIR}"
+      LIBRARY DESTINATION "${CMAKE_INSTALL_LIBDIR}"
+      INCLUDES DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}"
+    )
+  endif()

  install(
    EXPORT RocksDBTargets
@@ -878,12 +1023,30 @@ if(NOT WIN32 OR ROCKSDB_INSTALL_ON_WINDOWS)
  )
 endif()

-option(WITH_TESTS "build with tests" ON)
-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_garbage_test.cc
+        db/blob/blob_file_reader_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
@@ -893,8 +1056,7 @@ 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
@@ -920,12 +1082,14 @@ if(WITH_TESTS)
        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
@@ -941,6 +1105,7 @@ 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
@@ -951,17 +1116,21 @@ 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
@@ -969,9 +1138,11 @@ if(WITH_TESTS)
        monitoring/iostats_context_test.cc
        monitoring/statistics_test.cc
        monitoring/stats_history_test.cc
+        options/configurable_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
@@ -982,7 +1153,11 @@ 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/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
@@ -991,17 +1166,22 @@ if(WITH_TESTS)
        util/bloom_test.cc
        util/coding_test.cc
        util/crc32c_test.cc
+        util/defer_test.cc
        util/dynamic_bloom_test.cc
        util/file_reader_writer_test.cc
        util/filelock_test.cc
        util/hash_test.cc
        util/heap_test.cc
+        util/random_test.cc
        util/rate_limiter_test.cc
        util/repeatable_thread_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
@@ -1021,11 +1201,13 @@ if(WITH_TESTS)
        utilities/table_properties_collectors/compact_on_deletion_collector_test.cc
        utilities/transactions/optimistic_transaction_test.cc
        utilities/transactions/transaction_test.cc
+        utilities/transactions/transaction_lock_mgr_test.cc
        utilities/transactions/write_prepared_transaction_test.cc
        utilities/transactions/write_unprepared_transaction_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()
@@ -1034,35 +1216,17 @@ if(WITH_TESTS)
    list(APPEND TESTS third-party/folly/folly/synchronization/test/DistributedMutexTest.cpp)
  endif()

-  set(BENCHMARKS
-    cache/cache_bench.cc
-    memtable/memtablerep_bench.cc
-    db/range_del_aggregator_bench.cc
-    tools/db_bench.cc
-    table/table_reader_bench.cc
-    utilities/persistent_cache/hash_table_bench.cc)
-  add_library(testharness OBJECT test_util/testharness.cc)
-  foreach(sourcefile ${BENCHMARKS})
-    get_filename_component(exename ${sourcefile} NAME_WE)
-    add_executable(${exename}${ARTIFACT_SUFFIX} ${sourcefile}
-      $<TARGET_OBJECTS:testharness>)
-    target_link_libraries(${exename}${ARTIFACT_SUFFIX} gtest ${LIBS})
-  endforeach(sourcefile ${BENCHMARKS})
-
-  # For test util library that is build only in DEBUG mode
-  # and linked to tests. Add test only code that is not #ifdefed for Release here.
  set(TESTUTIL_SOURCE
      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
  )
-  # test utilities are only build in debug
  enable_testing()
  add_custom_target(check COMMAND ${CMAKE_CTEST_COMMAND})
  set(TESTUTILLIB testutillib${ARTIFACT_SUFFIX})
  add_library(${TESTUTILLIB} STATIC ${TESTUTIL_SOURCE})
+  target_link_libraries(${TESTUTILLIB} ${ROCKSDB_LIB})
  if(MSVC)
    set_target_properties(${TESTUTILLIB} PROPERTIES COMPILE_FLAGS "/Fd${CMAKE_CFG_INTDIR}/testutillib${ARTIFACT_SUFFIX}.pdb")
  endif()
@@ -1070,48 +1234,99 @@ if(WITH_TESTS)
        PROPERTIES EXCLUDE_FROM_DEFAULT_BUILD_RELEASE 1
        EXCLUDE_FROM_DEFAULT_BUILD_MINRELEASE 1
        EXCLUDE_FROM_DEFAULT_BUILD_RELWITHDEBINFO 1
-        )
+  )

-  # Tests are excluded from Release builds
-  set(TEST_EXES ${TESTS})
-
-  foreach(sourcefile ${TEST_EXES})
+  foreach(sourcefile ${TESTS})
      get_filename_component(exename ${sourcefile} NAME_WE)
-      add_executable(${CMAKE_PROJECT_NAME}_${exename}${ARTIFACT_SUFFIX} ${sourcefile}
-        $<TARGET_OBJECTS:testharness>)
+      add_executable(${CMAKE_PROJECT_NAME}_${exename}${ARTIFACT_SUFFIX} ${sourcefile})
      set_target_properties(${CMAKE_PROJECT_NAME}_${exename}${ARTIFACT_SUFFIX}
        PROPERTIES EXCLUDE_FROM_DEFAULT_BUILD_RELEASE 1
        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} gtest ${LIBS})
+      )
+      target_link_libraries(${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()
-  endforeach(sourcefile ${TEST_EXES})
+      if("${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})

-  # C executables must link to a shared object
-  set(C_TESTS db/c_test.c)
-  set(C_TEST_EXES ${C_TESTS})
+  if(WIN32)
+    # C executables must link to a shared object
+    if(ROCKSDB_BUILD_SHARED)
+      set(ROCKSDB_LIB_FOR_C ${ROCKSDB_SHARED_LIB})
+    else()
+      set(ROCKSDB_LIB_FOR_C OFF)
+    endif()
+  else()
+    set(ROCKSDB_LIB_FOR_C ${ROCKSDB_LIB})
+  endif()

-  foreach(sourcefile ${C_TEST_EXES})
-      string(REPLACE ".c" "" exename ${sourcefile})
-      string(REGEX REPLACE "^((.+)/)+" "" exename ${exename})
-      add_executable(${exename}${ARTIFACT_SUFFIX} ${sourcefile})
-      set_target_properties(${exename}${ARTIFACT_SUFFIX}
-        PROPERTIES EXCLUDE_FROM_DEFAULT_BUILD_RELEASE 1
-        EXCLUDE_FROM_DEFAULT_BUILD_MINRELEASE 1
-        EXCLUDE_FROM_DEFAULT_BUILD_RELWITHDEBINFO 1
-        )
-      target_link_libraries(${exename}${ARTIFACT_SUFFIX} ${ROCKSDB_IMPORT_LIB} testutillib${ARTIFACT_SUFFIX})
-      add_test(NAME ${exename} COMMAND ${exename}${ARTIFACT_SUFFIX})
-      add_dependencies(check ${exename}${ARTIFACT_SUFFIX})
-  endforeach(sourcefile ${C_TEST_EXES})
+  if(ROCKSDB_LIB_FOR_C)
+    set(C_TESTS db/c_test.c)
+    add_executable(c_test db/c_test.c)
+    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_TOOLS "build with tools" ON)
-if(WITH_TOOLS)
+if(WITH_BENCHMARK_TOOLS)
+  add_executable(db_bench
+    tools/db_bench.cc
+    tools/db_bench_tool.cc)
+  target_link_libraries(db_bench
+    ${ROCKSDB_LIB} ${THIRDPARTY_LIBS})
+
+  add_executable(cache_bench
+    cache/cache_bench.cc)
+  target_link_libraries(cache_bench
+    ${ROCKSDB_LIB} ${GFLAGS_LIB})
+
+  add_executable(memtablerep_bench
+    memtable/memtablerep_bench.cc)
+  target_link_libraries(memtablerep_bench
+    ${ROCKSDB_LIB} ${GFLAGS_LIB})
+
+  add_executable(range_del_aggregator_bench
+    db/range_del_aggregator_bench.cc)
+  target_link_libraries(range_del_aggregator_bench
+    ${ROCKSDB_LIB} ${GFLAGS_LIB})
+
+  add_executable(table_reader_bench
+    table/table_reader_bench.cc)
+  target_link_libraries(table_reader_bench
+    ${ROCKSDB_LIB} testharness ${GFLAGS_LIB})
+
+  add_executable(filter_bench
+    util/filter_bench.cc)
+  target_link_libraries(filter_bench
+    ${ROCKSDB_LIB} ${GFLAGS_LIB})
+
+  add_executable(hash_table_bench
+    utilities/persistent_cache/hash_table_bench.cc)
+  target_link_libraries(hash_table_bench
+    ${ROCKSDB_LIB} ${GFLAGS_LIB})
+endif()
+
+if(WITH_CORE_TOOLS OR WITH_TOOLS)
  add_subdirectory(tools)
+  add_custom_target(core_tools
+    DEPENDS ${core_tool_deps})
+endif()
+
+if(WITH_TOOLS)
+  add_subdirectory(db_stress_tool)
+  add_custom_target(tools
+    DEPENDS ${tool_deps})
+endif()
+
+option(WITH_EXAMPLES "build with examples" OFF)
+if(WITH_EXAMPLES)
+  add_subdirectory(examples)
 endif()
@@ -1,25 +1,364 @@
 # Rocksdb Change Log
 ## Unreleased
 ### 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.
+
+### 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.
+
+## 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.
+
+## 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.
+
+### 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, 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
+* Major breaking changes to Java comparators, toward standardizing on ByteBuffer for performant, locale-neutral operations on keys (#6252).
+* Added overloads of common API methods using direct ByteBuffers for keys and values (#2283).
+
+### Bug Fixes
+* Fix incorrect results while block-based table uses kHashSearch, together with Prev()/SeekForPrev().
+* Fix a bug that prevents opening a DB after two consecutive crash with TransactionDB, where the first crash recovers from a corrupted WAL with kPointInTimeRecovery but the second cannot.
+* Fixed issue #6316 that can cause a corruption of the MANIFEST file in the middle when writing to it fails due to no disk space.
+* Add DBOptions::skip_checking_sst_file_sizes_on_db_open. It disables potentially expensive checking of all sst file sizes in DB::Open().
+* BlobDB now ignores trivially moved files when updating the mapping between blob files and SSTs. This should mitigate issue #6338 where out of order flush/compaction notifications could trigger an assertion with the earlier code.
+* Batched MultiGet() ignores IO errors while reading data blocks, causing it to potentially continue looking for a key and returning stale results.
+* `WriteBatchWithIndex::DeleteRange` returns `Status::NotSupported`. Previously it returned success even though reads on the batch did not account for range tombstones. The corresponding language bindings now cannot be used. In C, that includes `rocksdb_writebatch_wi_delete_range`, `rocksdb_writebatch_wi_delete_range_cf`, `rocksdb_writebatch_wi_delete_rangev`, and `rocksdb_writebatch_wi_delete_rangev_cf`. In Java, that includes `WriteBatchWithIndex::deleteRange`.
+* Assign new MANIFEST file number when caller tries to create a new MANIFEST by calling LogAndApply(..., new_descriptor_log=true). This bug can cause MANIFEST being overwritten during recovery if options.write_dbid_to_manifest = true and there are WAL file(s).
+
+### Performance Improvements
+* Perfom readahead when reading from option files. Inside DB, options.log_readahead_size will be used as the readahead size. In other cases, a default 512KB is used.
+
+### Public API Change
+* The BlobDB garbage collector now emits the statistics `BLOB_DB_GC_NUM_FILES` (number of blob files obsoleted during GC), `BLOB_DB_GC_NUM_NEW_FILES` (number of new blob files generated during GC), `BLOB_DB_GC_FAILURES` (number of failed GC passes), `BLOB_DB_GC_NUM_KEYS_RELOCATED` (number of blobs relocated during GC), and `BLOB_DB_GC_BYTES_RELOCATED` (total size of blobs relocated during GC). On the other hand, the following statistics, which are not relevant for the new GC implementation, are now deprecated: `BLOB_DB_GC_NUM_KEYS_OVERWRITTEN`, `BLOB_DB_GC_NUM_KEYS_EXPIRED`, `BLOB_DB_GC_BYTES_OVERWRITTEN`, `BLOB_DB_GC_BYTES_EXPIRED`, and `BLOB_DB_GC_MICROS`.
+* Disable recycle_log_file_num when an inconsistent recovery modes are requested: kPointInTimeRecovery and kAbsoluteConsistency
+
+### New Features
+* Added the checksum for each SST file generated by Flush or Compaction. Added sst_file_checksum_func to Options such that user can plugin their own SST file checksum function via override the FileChecksumFunc class. If user does not set the sst_file_checksum_func, SST file checksum calculation will not be enabled. The checksum information inlcuding uint32_t checksum value and a checksum function name (string). The checksum information is stored in FileMetadata in version store and also logged to MANIFEST. A new tool is added to LDB such that user can dump out a list of file checksum information from MANIFEST (stored in an unordered_map).
+* `db_bench` now supports `value_size_distribution_type`, `value_size_min`, `value_size_max` options for generating random variable sized value. Added `blob_db_compression_type` option for BlobDB to enable blob compression.
+* Replace RocksDB namespace "rocksdb" with flag "ROCKSDB_NAMESPACE" which if is not defined, defined as "rocksdb" in header file rocksdb_namespace.h.
+
+## 6.7.0 (01/21/2020)
+### Public API Change
+* Added a rocksdb::FileSystem class in include/rocksdb/file_system.h to encapsulate file creation/read/write operations, and an option DBOptions::file_system to allow a user to pass in an instance of rocksdb::FileSystem. If its a non-null value, this will take precendence over DBOptions::env for file operations. A new API rocksdb::FileSystem::Default() returns a platform default object. The DBOptions::env option and Env::Default() API will continue to be used for threading and other OS related functions, and where DBOptions::file_system is not specified, for file operations. For storage developers who are accustomed to rocksdb::Env, the interface in rocksdb::FileSystem is new and will probably undergo some changes as more storage systems are ported to it from rocksdb::Env. As of now, no env other than Posix has been ported to the new interface.
+* A new rocksdb::NewSstFileManager() API that allows the caller to pass in separate Env and FileSystem objects.
+* Changed Java API for RocksDB.keyMayExist functions to use Holder<byte[]> instead of StringBuilder, so that retrieved values need not decode to Strings.
+* A new `OptimisticTransactionDBOptions` Option that allows users to configure occ validation policy. The default policy changes from kValidateSerial to kValidateParallel to reduce mutex contention.
+
+### Bug Fixes
+* Fix a bug that can cause unnecessary bg thread to be scheduled(#6104).
+* Fix crash caused by concurrent CF iterations and drops(#6147).
+* Fix a race condition for cfd->log_number_ between manifest switch and memtable switch (PR 6249) when number of column families is greater than 1.
+* Fix a bug on fractional cascading index when multiple files at the same level contain the same smallest user key, and those user keys are for merge operands. In this case, Get() the exact key may miss some merge operands.
+* Delcare kHashSearch index type feature-incompatible with index_block_restart_interval larger than 1.
+* Fixed an issue where the thread pools were not resized upon setting `max_background_jobs` dynamically through the `SetDBOptions` interface.
+* Fix a bug that can cause write threads to hang when a slowdown/stall happens and there is a mix of writers with WriteOptions::no_slowdown set/unset.
+* Fixed an issue where an incorrect "number of input records" value was used to compute the "records dropped" statistics for compactions.
+* Fix a regression bug that causes segfault when hash is used, max_open_files != -1 and total order seek is used and switched back.
+
+### New Features
+* It is now possible to enable periodic compactions for the base DB when using BlobDB.
+* BlobDB now garbage collects non-TTL blobs when `enable_garbage_collection` is set to `true` in `BlobDBOptions`. Garbage collection is performed during compaction: any valid blobs located in the oldest N files (where N is the number of non-TTL blob files multiplied by the value of `BlobDBOptions::garbage_collection_cutoff`) encountered during compaction get relocated to new blob files, and old blob files are dropped once they are no longer needed. Note: we recommend enabling periodic compactions for the base DB when using this feature to deal with the case when some old blob files are kept alive by SSTs that otherwise do not get picked for compaction.
+* `db_bench` now supports the `garbage_collection_cutoff` option for BlobDB.
+* Introduce ReadOptions.auto_prefix_mode. When set to true, iterator will return the same result as total order seek, but may choose to use prefix seek internally based on seek key and iterator upper bound.
+* MultiGet() can use IO Uring to parallelize read from the same SST file. This featuer is by default disabled. It can be enabled with environment variable ROCKSDB_USE_IO_URING.
+
+## 6.6.2 (01/13/2020)
+### Bug Fixes
+* Fixed a bug where non-L0 compaction input files were not considered to compute the `creation_time` of new compaction outputs.
+
+## 6.6.1 (01/02/2020)
+### Bug Fixes
+* Fix a bug in WriteBatchWithIndex::MultiGetFromBatchAndDB, which is called by Transaction::MultiGet, that causes due to stale pointer access when the number of keys is > 32
+* Fixed two performance issues related to memtable history trimming. First, a new SuperVersion is now created only if some memtables were actually trimmed. Second, trimming is only scheduled if there is at least one flushed memtable that is kept in memory for the purposes of transaction conflict checking.
+* BlobDB no longer updates the SST to blob file mapping upon failed compactions.
+* Fix a bug in which a snapshot read through an iterator could be affected by a DeleteRange after the snapshot (#6062).
+* Fixed a bug where BlobDB was comparing the `ColumnFamilyHandle` pointers themselves instead of only the column family IDs when checking whether an API call uses the default column family or not.
+* Delete superversions in BackgroundCallPurge.
+* Fix use-after-free and double-deleting files in BackgroundCallPurge().
+
+## 6.6.0 (11/25/2019)
+### Bug Fixes
+* Fix data corruption caused by output of intra-L0 compaction on ingested file not being placed in correct order in L0.
+* Fix a data race between Version::GetColumnFamilyMetaData() and Compaction::MarkFilesBeingCompacted() for access to being_compacted (#6056). The current fix acquires the db mutex during Version::GetColumnFamilyMetaData(), which may cause regression.
+* Fix a bug in DBIter that is_blob_ state isn't updated when iterating backward using seek.
+* Fix a bug when format_version=3, partitioned filters, and prefix search are used in conjunction. The bug could result into Seek::(prefix) returning NotFound for an existing prefix.
+* Revert the feature "Merging iterator to avoid child iterator reseek for some cases (#5286)" since it might cause strong results when reseek happens with a different iterator upper bound.
+* Fix a bug causing a crash during ingest external file when background compaction cause severe error (file not found).
+* Fix a bug when partitioned filters and prefix search are used in conjunction, ::SeekForPrev could return invalid for an existing prefix. ::SeekForPrev might be called by the user, or internally on ::Prev, or within ::Seek if the return value involves Delete or a Merge operand.
+* Fix OnFlushCompleted fired before flush result persisted in MANIFEST when there's concurrent flush job. The bug exists since OnFlushCompleted was introduced in rocksdb 3.8.
+* Fixed an sst_dump crash on some plain table SST files.
+* Fixed a memory leak in some error cases of opening plain table SST files.
+* Fix a bug when a crash happens while calling WriteLevel0TableForRecovery for multiple column families, leading to a column family's log number greater than the first corrutped log number when the DB is being opened in PointInTime recovery mode during next recovery attempt (#5856).
+
+### New Features
+* Universal compaction to support options.periodic_compaction_seconds. A full compaction will be triggered if any file is over the threshold.
+* `GetLiveFilesMetaData` and `GetColumnFamilyMetaData` now expose the file number of SST files as well as the oldest blob file referenced by each SST.
+* A batched MultiGet API (DB::MultiGet()) that supports retrieving keys from multiple column families.
+* Full and partitioned filters in the block-based table use an improved Bloom filter implementation, enabled with format_version 5 (or above) because previous releases cannot read this filter. This replacement is faster and more accurate, especially for high bits per key or millions of keys in a single (full) filter. For example, the new Bloom filter has the same false positive rate at 9.55 bits per key as the old one at 10 bits per key, and a lower false positive rate at 16 bits per key than the old one at 100 bits per key.
+* Added AVX2 instructions to USE_SSE builds to accelerate the new Bloom filter and XXH3-based hash function on compatible x86_64 platforms (Haswell and later, ~2014).
+* Support options.ttl or options.periodic_compaction_seconds with options.max_open_files = -1. File's oldest ancester time and file creation time will be written to manifest. If it is availalbe, this information will be used instead of creation_time and file_creation_time in table properties.
+* Setting options.ttl for universal compaction now has the same meaning as setting periodic_compaction_seconds.
+* SstFileMetaData also returns file creation time and oldest ancester time.
+* The `sst_dump` command line tool `recompress` command now displays how many blocks were compressed and how many were not, in particular how many were not compressed because the compression ratio was not met (12.5% threshold for GoodCompressionRatio), as seen in the `number.block.not_compressed` counter stat since version 6.0.0.
+* The block cache usage is now takes into account the overhead of metadata per each entry. This results into more accurate management of memory. A side-effect of this feature is that less items are fit into the block cache of the same size, which would result to higher cache miss rates. This can be remedied by increasing the block cache size or passing kDontChargeCacheMetadata to its constuctor to restore the old behavior.
+* When using BlobDB, a mapping is maintained and persisted in the MANIFEST between each SST file and the oldest non-TTL blob file it references.
+* `db_bench` now supports and by default issues non-TTL Puts to BlobDB. TTL Puts can be enabled by specifying a non-zero value for the `blob_db_max_ttl_range` command line parameter explicitly.
+* `sst_dump` now supports printing BlobDB blob indexes in a human-readable format. This can be enabled by specifying the `decode_blob_index` flag on the command line.
+* A number of new information elements are now exposed through the EventListener interface. For flushes, the file numbers of the new SST file and the oldest blob file referenced by the SST are propagated. For compactions, the level, file number, and the oldest blob file referenced are passed to the client for each compaction input and output file.
+
+### Public API Change
+* RocksDB release 4.1 or older will not be able to open DB generated by the new release. 4.2 was released on Feb 23, 2016.
+* TTL Compactions in Level compaction style now initiate successive cascading compactions on a key range so that it reaches the bottom level quickly on TTL expiry. `creation_time` table property for compaction output files is now set to the minimum of the creation times of all compaction inputs.
+* With FIFO compaction style, options.periodic_compaction_seconds will have the same meaning as options.ttl. Whichever stricter will be used. With the default options.periodic_compaction_seconds value with options.ttl's default of 0, RocksDB will give a default of 30 days.
+* Added an API GetCreationTimeOfOldestFile(uint64_t* creation_time) to get the file_creation_time of the oldest SST file in the DB.
+* FilterPolicy now exposes additional API to make it possible to choose filter configurations based on context, such as table level and compaction style. See `LevelAndStyleCustomFilterPolicy` in db_bloom_filter_test.cc. While most existing custom implementations of FilterPolicy should continue to work as before, those wrapping the return of NewBloomFilterPolicy will require overriding new function `GetBuilderWithContext()`, because calling `GetFilterBitsBuilder()` on the FilterPolicy returned by NewBloomFilterPolicy is no longer supported.
+* An unlikely usage of FilterPolicy is no longer supported. Calling GetFilterBitsBuilder() on the FilterPolicy returned by NewBloomFilterPolicy will now cause an assertion violation in debug builds, because RocksDB has internally migrated to a more elaborate interface that is expected to evolve further. Custom implementations of FilterPolicy should work as before, except those wrapping the return of NewBloomFilterPolicy, which will require a new override of a protected function in FilterPolicy.
+* NewBloomFilterPolicy now takes bits_per_key as a double instead of an int. This permits finer control over the memory vs. accuracy trade-off in the new Bloom filter implementation and should not change source code compatibility.
+* The option BackupableDBOptions::max_valid_backups_to_open is now only used when opening BackupEngineReadOnly. When opening a read/write BackupEngine, anything but the default value logs a warning and is treated as the default. This change ensures that backup deletion has proper accounting of shared files to ensure they are deleted when no longer referenced by a backup.
+* Deprecate `snap_refresh_nanos` option.
+* Added DisableManualCompaction/EnableManualCompaction to stop and resume manual compaction.
+* Add TryCatchUpWithPrimary() to StackableDB in non-LITE mode.
+* Add a new Env::LoadEnv() overloaded function to return a shared_ptr to Env.
+* Flush sets file name to "(nil)" for OnTableFileCreationCompleted() if the flush does not produce any L0. This can happen if the file is empty thus delete by RocksDB.
+
+### Default Option Changes
+* Changed the default value of periodic_compaction_seconds to `UINT64_MAX - 1` which allows RocksDB to auto-tune periodic compaction scheduling. When using the default value, periodic compactions are now auto-enabled if a compaction filter is used. A value of `0` will turn off the feature completely.
+* Changed the default value of ttl to `UINT64_MAX - 1` which allows RocksDB to auto-tune ttl value. When using the default value, TTL will be auto-enabled to 30 days, when the feature is supported. To revert the old behavior, you can explicitly set it to 0.
+
+### Performance Improvements
+* For 64-bit hashing, RocksDB is standardizing on a slightly modified preview version of XXH3. This function is now used for many non-persisted hashes, along with fastrange64() in place of the modulus operator, and some benchmarks show a slight improvement.
+* Level iterator to invlidate the iterator more often in prefix seek and the level is filtered out by prefix bloom.
+
+## 6.5.2 (11/15/2019)
+### Bug Fixes
+* Fix a assertion failure in MultiGet() when BlockBasedTableOptions::no_block_cache is true and there is no compressed block cache
+* Fix a buffer overrun problem in BlockBasedTable::MultiGet() when compression is enabled and no compressed block cache is configured.
+* If a call to BackupEngine::PurgeOldBackups or BackupEngine::DeleteBackup suffered a crash, power failure, or I/O error, files could be left over from old backups that could only be purged with a call to GarbageCollect. Any call to PurgeOldBackups, DeleteBackup, or GarbageCollect should now suffice to purge such files.
+
+## 6.5.1 (10/16/2019)
+### Bug Fixes
+* Revert the feature "Merging iterator to avoid child iterator reseek for some cases (#5286)" since it might cause strange results when reseek happens with a different iterator upper bound.
+* Fix a bug in BlockBasedTableIterator that might return incorrect results when reseek happens with a different iterator upper bound.
+* Fix a bug when partitioned filters and prefix search are used in conjunction, ::SeekForPrev could return invalid for an existing prefix. ::SeekForPrev might be called by the user, or internally on ::Prev, or within ::Seek if the return value involves Delete or a Merge operand.
+
+## 6.5.0 (9/13/2019)
+### Bug Fixes
 * Fixed a number of data races in BlobDB.
 * Fix a bug where the compaction snapshot refresh feature is not disabled as advertised when `snap_refresh_nanos` is set to 0..
 * Fix bloom filter lookups by the MultiGet batching API when BlockBasedTableOptions::whole_key_filtering is false, by checking that a key is in the perfix_extractor domain and extracting the prefix before looking up.
 * Fix a bug in file ingestion caused by incorrect file number allocation when the number of column families involved in the ingestion exceeds 2.
+
 ### New Features
 * Introduced DBOptions::max_write_batch_group_size_bytes to configure maximum limit on number of bytes that are written in a single batch of WAL or memtable write. It is followed when the leader write size is larger than 1/8 of this limit.
 * VerifyChecksum() by default will issue readahead. Allow ReadOptions to be passed in to those functions to override the readhead size. For checksum verifying before external SST file ingestion, a new option IngestExternalFileOptions.verify_checksums_readahead_size, is added for this readahead setting.
 * When user uses options.force_consistency_check in RocksDb, instead of crashing the process, we now pass the error back to the users without killing the process.
 * Add an option `memtable_insert_hint_per_batch` to WriteOptions. If it is true, each WriteBatch will maintain its own insert hints for each memtable in concurrent write. See include/rocksdb/options.h for more details.
+
 ### Public API Change
 * Added max_write_buffer_size_to_maintain option to better control memory usage of immutable memtables.
 * Added a lightweight API GetCurrentWalFile() to get last live WAL filename and size. Meant to be used as a helper for backup/restore tooling in a larger ecosystem such as MySQL with a MyRocks storage engine.
 * The MemTable Bloom filter, when enabled, now always uses cache locality. Options::bloom_locality now only affects the PlainTable SST format.
+
 ### Performance Improvements
 * Improve the speed of the MemTable Bloom filter, reducing the write overhead of enabling it by 1/3 to 1/2, with similar benefit to read performance.

 ## 6.4.0 (7/30/2019)
 ### Default Option Change
-* LRUCacheOptions.high_pri_pool_ratio is set to 0.5 (previously 0.0) by default, which means that by default midpoint insertion is enabled. The same change is made for the default value of high_pri_pool_ratio argument in NewLRUCache(). When block cache is not explictly created, the small block cache created by BlockBasedTable will still has this option to be 0.0.
+* LRUCacheOptions.high_pri_pool_ratio is set to 0.5 (previously 0.0) by default, which means that by default midpoint insertion is enabled. The same change is made for the default value of high_pri_pool_ratio argument in NewLRUCache(). When block cache is not explicitly created, the small block cache created by BlockBasedTable will still has this option to be 0.0.
 * Change BlockBasedTableOptions.cache_index_and_filter_blocks_with_high_priority's default value from false to true.

 ### Public API Change
@@ -43,7 +382,7 @@
 * Support loading custom objects in unit tests. In the affected unit tests, RocksDB will create custom Env objects based on environment variable TEST_ENV_URI. Users need to make sure custom object types are properly registered. For example, a static library should expose a `RegisterCustomObjects` function. By linking the unit test binary with the static library, the unit test can execute this function.

 ### Performance Improvements
-* Reduce iterator key comparision for upper/lower bound check.
+* Reduce iterator key comparison for upper/lower bound check.
 * Improve performance of row_cache: make reads with newer snapshots than data in an SST file share the same cache key, except in some transaction cases.
 * The compression dictionary is no longer copied to a new object upon retrieval.

@@ -102,7 +441,6 @@
 * Fix a bug caused by secondary not skipping the beginning of new MANIFEST.
 * On DB open, delete WAL trash files left behind in wal_dir

-
 ## 6.2.0 (4/30/2019)
 ### New Features
 * Add an option `strict_bytes_per_sync` that causes a file-writing thread to block rather than exceed the limit on bytes pending writeback specified by `bytes_per_sync` or `wal_bytes_per_sync`.
@@ -140,7 +478,7 @@
 * Introduce two more stats levels, kExceptHistogramOrTimers and kExceptTimers.
 * Added a feature to perform data-block sampling for compressibility, and report stats to user.
 * Add support for trace filtering.
-* Add DBOptions.avoid_unnecessary_blocking_io. If true, we avoid file deletion when destorying ColumnFamilyHandle and Iterator. Instead, a job is scheduled to delete the files in background.
+* Add DBOptions.avoid_unnecessary_blocking_io. If true, we avoid file deletion when destroying ColumnFamilyHandle and Iterator. Instead, a job is scheduled to delete the files in background.

 ### Public API Change
 * Remove bundled fbson library.
@@ -519,7 +857,7 @@ if set to something > 0 user will see 2 changes in iterators behavior 1) only ke

 ## 5.2.0 (02/08/2017)
 ### Public API Change
-* NewLRUCache() will determine number of shard bits automatically based on capacity, if the user doesn't pass one. This also impacts the default block cache when the user doesn't explict provide one.
+* NewLRUCache() will determine number of shard bits automatically based on capacity, if the user doesn't pass one. This also impacts the default block cache when the user doesn't explicit provide one.
 * Change the default of delayed slowdown value to 16MB/s and further increase the L0 stop condition to 36 files.
 * Options::use_direct_writes and Options::use_direct_reads are now ready to use.
 * (Experimental) Two-level indexing that partition the index and creates a 2nd level index on the partitions. The feature can be enabled by setting kTwoLevelIndexSearch as IndexType and configuring index_per_partition.
@@ -590,7 +928,7 @@ if set to something > 0 user will see 2 changes in iterators behavior 1) only ke

 ### New Features
 * A tool to migrate DB after options change. See include/rocksdb/utilities/option_change_migration.h.
-* Add ReadOptions.background_purge_on_iterator_cleanup. If true, we avoid file deletion when destorying iterators.
+* Add ReadOptions.background_purge_on_iterator_cleanup. If true, we avoid file deletion when destroying iterators.

 ## 4.10.0 (7/5/2016)
 ### Public API Change
@@ -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
@@ -1,7 +1,8 @@
 ## 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)
+[![CircleCI Status](https://circleci.com/gh/facebook/rocksdb.svg?style=svg)](https://circleci.com/gh/facebook/rocksdb)
+[![TravisCI Status](https://travis-ci.org/facebook/rocksdb.svg?branch=master)](https://travis-ci.org/facebook/rocksdb)
+[![Appveyor Build status](https://ci.appveyor.com/api/projects/status/fbgfu0so3afcno78/branch/master?svg=true)](https://ci.appveyor.com/project/Facebook/rocksdb/branch/master)
 [![PPC64le Build Status](http://140.211.168.68:8080/buildStatus/icon?job=Rocksdb)](http://140.211.168.68:8080/job/Rocksdb)

 RocksDB is developed and maintained by Facebook Database Engineering Team.
@@ -9,11 +10,11 @@ It is built on earlier work on [LevelDB](https://github.com/google/leveldb) by S
 and Jeff Dean (jeff@google.com)

 This code is a library that forms the core building block for a fast
-key value server, especially suited for storing data on flash drives.
+key-value server, especially suited for storing data on flash drives.
 It has a Log-Structured-Merge-Database (LSM) design with flexible tradeoffs
 between Write-Amplification-Factor (WAF), Read-Amplification-Factor (RAF)
 and Space-Amplification-Factor (SAF). It has multi-threaded compactions,
-making it specially suitable for storing multiple terabytes of data in a
+making it especially suitable for storing multiple terabytes of data in a
 single database.

 Start with example usage here: https://github.com/facebook/rocksdb/tree/master/examples
@@ -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

@@ -1,3 +1,8 @@
+# This file @generated by `python3 buckifier/buckify_rocksdb.py`
+# --> DO NOT EDIT MANUALLY <--
+# This file is a Facebook-specific integration for buck builds, so can
+# only be validated by Facebook employees.
+#
 load("@fbcode_macros//build_defs:auto_headers.bzl", "AutoHeaders")
 load("@fbcode_macros//build_defs:cpp_library.bzl", "cpp_library")
 load(":defs.bzl", "test_binary")
@@ -20,14 +25,16 @@ ROCKSDB_EXTERNAL_DEPS = [
    ("gflags", None, "gflags"),
    ("lz4", None, "lz4"),
    ("zstd", None),
-    ("tbb", None),
-    ("googletest", None, "gtest"),
 ]

 ROCKSDB_OS_DEPS = [
    (
        "linux",
-        ["third-party//numa:numa"],
+        ["third-party//numa:numa", "third-party//liburing:uring", "third-party//tbb:tbb"],
+    ),
+    (
+        "macos",
+        ["third-party//tbb:tbb"],
    ),
 ]

@@ -41,19 +48,31 @@ ROCKSDB_OS_PREPROCESSOR_FLAGS = [
            "-DROCKSDB_PTHREAD_ADAPTIVE_MUTEX",
            "-DROCKSDB_RANGESYNC_PRESENT",
            "-DROCKSDB_SCHED_GETCPU_PRESENT",
+            "-DROCKSDB_IOURING_PRESENT",
            "-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
@@ -64,7 +83,6 @@ ROCKSDB_PREPROCESSOR_FLAGS = [
    "-DZSTD",
    "-DZSTD_STATIC_LINKING_ONLY",
    "-DGFLAGS=gflags",
-    "-DTBB",

    # Added missing flags from output of build_detect_platform
    "-DROCKSDB_BACKTRACE",
@@ -102,12 +120,27 @@ ROCKSDB_OS_DEPS += ([(
    ["third-party//jemalloc:headers"],
 )] if sanitizer == "" else [])

+ROCKSDB_LIB_DEPS = [
+    ":rocksdb_lib",
+    ":rocksdb_test_lib",
+] if not is_opt_mode else [":rocksdb_lib"]
+
 cpp_library(
    name = "rocksdb_lib",
    srcs = [
+        "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_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",
@@ -119,6 +152,7 @@ cpp_library(
        "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/db_impl.cc",
@@ -151,6 +185,8 @@ cpp_library(
        "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",
@@ -161,7 +197,9 @@ cpp_library(
        "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",
@@ -172,18 +210,28 @@ cpp_library(
        "env/env_encryption.cc",
        "env/env_hdfs.cc",
        "env/env_posix.cc",
+        "env/file_system.cc",
+        "env/file_system_tracer.cc",
+        "env/fs_posix.cc",
        "env/io_posix.cc",
        "env/mock_env.cc",
        "file/delete_scheduler.cc",
+        "file/file_prefetch_buffer.cc",
        "file/file_util.cc",
        "file/filename.cc",
+        "file/random_access_file_reader.cc",
+        "file/read_write_util.cc",
+        "file/readahead_raf.cc",
+        "file/sequence_file_reader.cc",
        "file/sst_file_manager_impl.cc",
+        "file/writable_file_writer.cc",
        "logging/auto_roll_logger.cc",
        "logging/event_logger.cc",
        "logging/log_buffer.cc",
        "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",
@@ -205,28 +253,44 @@ cpp_library(
        "monitoring/thread_status_util.cc",
        "monitoring/thread_status_util_debug.cc",
        "options/cf_options.cc",
+        "options/configurable.cc",
        "options/db_options.cc",
        "options/options.cc",
        "options/options_helper.cc",
        "options/options_parser.cc",
-        "options/options_sanity_check.cc",
        "port/port_posix.cc",
        "port/stack_trace.cc",
+        "port/win/env_default.cc",
+        "port/win/env_win.cc",
+        "port/win/io_win.cc",
+        "port/win/port_win.cc",
+        "port/win/win_logger.cc",
+        "port/win/win_thread.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_footer.cc",
        "table/block_based/data_block_hash_index.cc",
        "table/block_based/filter_block_reader_common.cc",
+        "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",
@@ -244,20 +308,23 @@ cpp_library(
        "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",
        "test_util/sync_point_impl.cc",
        "test_util/transaction_test_util.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",
        "trace_replay/block_cache_tracer.cc",
+        "trace_replay/io_tracer.cc",
        "trace_replay/trace_replay.cc",
-        "util/bloom.cc",
        "util/build_version.cc",
        "util/coding.cc",
        "util/compaction_job_stats_impl.cc",
@@ -266,8 +333,7 @@ cpp_library(
        "util/concurrent_task_limiter_impl.cc",
        "util/crc32c.cc",
        "util/dynamic_bloom.cc",
-        "util/file_reader_writer.cc",
-        "util/filter_policy.cc",
+        "util/file_checksum_helper.cc",
        "util/hash.cc",
        "util/murmurhash.cc",
        "util/random.cc",
@@ -285,9 +351,6 @@ cpp_library(
        "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_format.cc",
-        "utilities/blob_db/blob_log_reader.cc",
-        "utilities/blob_db/blob_log_writer.cc",
        "utilities/cassandra/cassandra_compaction_filter.cc",
        "utilities/cassandra/format.cc",
        "utilities/cassandra/merge_operator.cc",
@@ -297,6 +360,8 @@ cpp_library(
        "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",
@@ -318,6 +383,8 @@ cpp_library(
        "utilities/simulator_cache/sim_cache.cc",
        "utilities/table_properties_collectors/compact_on_deletion_collector.cc",
        "utilities/trace/file_trace_reader_writer.cc",
+        "utilities/transactions/lock/lock_tracker.cc",
+        "utilities/transactions/lock/point_lock_tracker.cc",
        "utilities/transactions/optimistic_transaction.cc",
        "utilities/transactions/optimistic_transaction_db_impl.cc",
        "utilities/transactions/pessimistic_transaction.cc",
@@ -343,6 +410,295 @@ cpp_library(
    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
    deps = [],
    external_deps = ROCKSDB_EXTERNAL_DEPS,
+    link_whole = False,
+)
+
+cpp_library(
+    name = "rocksdb_whole_archive_lib",
+    srcs = [
+        "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_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_job.cc",
+        "db/compaction/compaction_picker.cc",
+        "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/db_impl.cc",
+        "db/db_impl/db_impl_compaction_flush.cc",
+        "db/db_impl/db_impl_debug.cc",
+        "db/db_impl/db_impl_experimental.cc",
+        "db/db_impl/db_impl_files.cc",
+        "db/db_impl/db_impl_open.cc",
+        "db/db_impl/db_impl_readonly.cc",
+        "db/db_impl/db_impl_secondary.cc",
+        "db/db_impl/db_impl_write.cc",
+        "db/db_info_dumper.cc",
+        "db/db_iter.cc",
+        "db/dbformat.cc",
+        "db/error_handler.cc",
+        "db/event_helpers.cc",
+        "db/experimental.cc",
+        "db/external_sst_file_ingestion_job.cc",
+        "db/file_indexer.cc",
+        "db/flush_job.cc",
+        "db/flush_scheduler.cc",
+        "db/forward_iterator.cc",
+        "db/import_column_family_job.cc",
+        "db/internal_stats.cc",
+        "db/log_reader.cc",
+        "db/log_writer.cc",
+        "db/logs_with_prep_tracker.cc",
+        "db/malloc_stats.cc",
+        "db/memtable.cc",
+        "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",
+        "db/snapshot_impl.cc",
+        "db/table_cache.cc",
+        "db/table_properties_collector.cc",
+        "db/transaction_log_impl.cc",
+        "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/env.cc",
+        "env/env_chroot.cc",
+        "env/env_encryption.cc",
+        "env/env_hdfs.cc",
+        "env/env_posix.cc",
+        "env/file_system.cc",
+        "env/file_system_tracer.cc",
+        "env/fs_posix.cc",
+        "env/io_posix.cc",
+        "env/mock_env.cc",
+        "file/delete_scheduler.cc",
+        "file/file_prefetch_buffer.cc",
+        "file/file_util.cc",
+        "file/filename.cc",
+        "file/random_access_file_reader.cc",
+        "file/read_write_util.cc",
+        "file/readahead_raf.cc",
+        "file/sequence_file_reader.cc",
+        "file/sst_file_manager_impl.cc",
+        "file/writable_file_writer.cc",
+        "logging/auto_roll_logger.cc",
+        "logging/event_logger.cc",
+        "logging/log_buffer.cc",
+        "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",
+        "memtable/skiplistrep.cc",
+        "memtable/vectorrep.cc",
+        "memtable/write_buffer_manager.cc",
+        "monitoring/histogram.cc",
+        "monitoring/histogram_windowing.cc",
+        "monitoring/in_memory_stats_history.cc",
+        "monitoring/instrumented_mutex.cc",
+        "monitoring/iostats_context.cc",
+        "monitoring/perf_context.cc",
+        "monitoring/perf_level.cc",
+        "monitoring/persistent_stats_history.cc",
+        "monitoring/statistics.cc",
+        "monitoring/thread_status_impl.cc",
+        "monitoring/thread_status_updater.cc",
+        "monitoring/thread_status_updater_debug.cc",
+        "monitoring/thread_status_util.cc",
+        "monitoring/thread_status_util_debug.cc",
+        "options/cf_options.cc",
+        "options/configurable.cc",
+        "options/db_options.cc",
+        "options/options.cc",
+        "options/options_helper.cc",
+        "options/options_parser.cc",
+        "port/port_posix.cc",
+        "port/stack_trace.cc",
+        "port/win/env_default.cc",
+        "port/win/env_win.cc",
+        "port/win/io_win.cc",
+        "port/win/port_win.cc",
+        "port/win/win_logger.cc",
+        "port/win/win_thread.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_footer.cc",
+        "table/block_based/data_block_hash_index.cc",
+        "table/block_based/filter_block_reader_common.cc",
+        "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",
+        "table/cuckoo/cuckoo_table_factory.cc",
+        "table/cuckoo/cuckoo_table_reader.cc",
+        "table/format.cc",
+        "table/get_context.cc",
+        "table/iterator.cc",
+        "table/merging_iterator.cc",
+        "table/meta_blocks.cc",
+        "table/persistent_cache_helper.cc",
+        "table/plain/plain_table_bloom.cc",
+        "table/plain/plain_table_builder.cc",
+        "table/plain/plain_table_factory.cc",
+        "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",
+        "test_util/sync_point_impl.cc",
+        "test_util/transaction_test_util.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",
+        "trace_replay/block_cache_tracer.cc",
+        "trace_replay/io_tracer.cc",
+        "trace_replay/trace_replay.cc",
+        "util/build_version.cc",
+        "util/coding.cc",
+        "util/compaction_job_stats_impl.cc",
+        "util/comparator.cc",
+        "util/compression_context_cache.cc",
+        "util/concurrent_task_limiter_impl.cc",
+        "util/crc32c.cc",
+        "util/dynamic_bloom.cc",
+        "util/file_checksum_helper.cc",
+        "util/hash.cc",
+        "util/murmurhash.cc",
+        "util/random.cc",
+        "util/rate_limiter.cc",
+        "util/slice.cc",
+        "util/status.cc",
+        "util/string_util.cc",
+        "util/thread_local.cc",
+        "util/threadpool_imp.cc",
+        "util/xxhash.cc",
+        "utilities/backupable/backupable_db.cc",
+        "utilities/blob_db/blob_compaction_filter.cc",
+        "utilities/blob_db/blob_db.cc",
+        "utilities/blob_db/blob_db_impl.cc",
+        "utilities/blob_db/blob_db_impl_filesnapshot.cc",
+        "utilities/blob_db/blob_dump_tool.cc",
+        "utilities/blob_db/blob_file.cc",
+        "utilities/cassandra/cassandra_compaction_filter.cc",
+        "utilities/cassandra/format.cc",
+        "utilities/cassandra/merge_operator.cc",
+        "utilities/checkpoint/checkpoint_impl.cc",
+        "utilities/compaction_filters/remove_emptyvalue_compactionfilter.cc",
+        "utilities/convenience/info_log_finder.cc",
+        "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",
+        "utilities/merge_operators/max.cc",
+        "utilities/merge_operators/put.cc",
+        "utilities/merge_operators/sortlist.cc",
+        "utilities/merge_operators/string_append/stringappend.cc",
+        "utilities/merge_operators/string_append/stringappend2.cc",
+        "utilities/merge_operators/uint64add.cc",
+        "utilities/object_registry.cc",
+        "utilities/option_change_migration/option_change_migration.cc",
+        "utilities/options/options_util.cc",
+        "utilities/persistent_cache/block_cache_tier.cc",
+        "utilities/persistent_cache/block_cache_tier_file.cc",
+        "utilities/persistent_cache/block_cache_tier_metadata.cc",
+        "utilities/persistent_cache/persistent_cache_tier.cc",
+        "utilities/persistent_cache/volatile_tier_impl.cc",
+        "utilities/simulator_cache/cache_simulator.cc",
+        "utilities/simulator_cache/sim_cache.cc",
+        "utilities/table_properties_collectors/compact_on_deletion_collector.cc",
+        "utilities/trace/file_trace_reader_writer.cc",
+        "utilities/transactions/lock/lock_tracker.cc",
+        "utilities/transactions/lock/point_lock_tracker.cc",
+        "utilities/transactions/optimistic_transaction.cc",
+        "utilities/transactions/optimistic_transaction_db_impl.cc",
+        "utilities/transactions/pessimistic_transaction.cc",
+        "utilities/transactions/pessimistic_transaction_db.cc",
+        "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",
+        "utilities/transactions/write_unprepared_txn.cc",
+        "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",
+    ],
+    auto_headers = AutoHeaders.RECURSIVE_GLOB,
+    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,
+    deps = [],
+    external_deps = ROCKSDB_EXTERNAL_DEPS,
+    link_whole = True,
 )

 cpp_library(
@@ -350,7 +706,7 @@ cpp_library(
    srcs = [
        "db/db_test_util.cc",
        "table/mock_table.cc",
-        "test_util/fault_injection_test_env.cc",
+        "test_util/mock_time_env.cc",
        "test_util/testharness.cc",
        "test_util/testutil.cc",
        "tools/block_cache_analyzer/block_cache_trace_analyzer.cc",
@@ -364,7 +720,10 @@ cpp_library(
    os_preprocessor_flags = ROCKSDB_OS_PREPROCESSOR_FLAGS,
    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
    deps = [":rocksdb_lib"],
-    external_deps = ROCKSDB_EXTERNAL_DEPS,
+    external_deps = ROCKSDB_EXTERNAL_DEPS + [
+        ("googletest", None, "gtest"),
+    ],
+    link_whole = False,
 )

 cpp_library(
@@ -383,8 +742,56 @@ cpp_library(
    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
    deps = [":rocksdb_lib"],
    external_deps = ROCKSDB_EXTERNAL_DEPS,
+    link_whole = False,
 )

+cpp_library(
+    name = "rocksdb_stress_lib",
+    srcs = [
+        "db_stress_tool/batched_ops_stress.cc",
+        "db_stress_tool/cf_consistency_stress.cc",
+        "db_stress_tool/db_stress_common.cc",
+        "db_stress_tool/db_stress_driver.cc",
+        "db_stress_tool/db_stress_gflags.cc",
+        "db_stress_tool/db_stress_shared_state.cc",
+        "db_stress_tool/db_stress_test_base.cc",
+        "db_stress_tool/db_stress_tool.cc",
+        "db_stress_tool/no_batched_ops_stress.cc",
+        "test_util/testutil.cc",
+        "tools/block_cache_analyzer/block_cache_trace_analyzer.cc",
+        "tools/trace_analyzer_tool.cc",
+    ],
+    auto_headers = AutoHeaders.RECURSIVE_GLOB,
+    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,
+    deps = ROCKSDB_LIB_DEPS,
+    external_deps = ROCKSDB_EXTERNAL_DEPS,
+)
+
+if not is_opt_mode:
+    cpp_binary(
+        name = "c_test_bin",
+        srcs = ["db/c_test.c"],
+        arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
+        os_preprocessor_flags = ROCKSDB_OS_PREPROCESSOR_FLAGS,
+        compiler_flags = ROCKSDB_COMPILER_FLAGS,
+        preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+        deps = [":rocksdb_test_lib"],
+    )
+
+if not is_opt_mode:
+    custom_unittest(
+        "c_test",
+        command = [
+            native.package_name() + "/buckifier/rocks_test_runner.sh",
+            "$(location :{})".format("c_test_bin"),
+        ],
+        type = "simple",
+    )
+
 cpp_library(
    name = "env_basic_test_lib",
    srcs = ["env/env_basic_test.cc"],
@@ -396,6 +803,7 @@ cpp_library(
    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
    deps = [":rocksdb_test_lib"],
    external_deps = ROCKSDB_EXTERNAL_DEPS,
+    link_whole = False,
 )

 # [test_name, test_src, test_type, extra_deps, extra_compiler_flags]
@@ -435,6 +843,34 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "blob_file_addition_test",
+        "db/blob/blob_file_addition_test.cc",
+        "serial",
+        [],
+        [],
+    ],
+    [
+        "blob_file_builder_test",
+        "db/blob/blob_file_builder_test.cc",
+        "serial",
+        [],
+        [],
+    ],
+    [
+        "blob_file_garbage_test",
+        "db/blob/blob_file_garbage_test.cc",
+        "serial",
+        [],
+        [],
+    ],
+    [
+        "blob_file_reader_test",
+        "db/blob/blob_file_reader_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "block_based_filter_block_test",
        "table/block_based/block_based_filter_block_test.cc",
@@ -442,6 +878,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "block_based_table_reader_test",
+        "table/block_based/block_based_table_reader_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "block_cache_trace_analyzer_test",
        "tools/block_cache_analyzer/block_cache_trace_analyzer_test.cc",
@@ -456,6 +899,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "block_fetcher_test",
+        "table/block_fetcher_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "block_test",
        "table/block_based/block_test.cc",
@@ -470,13 +920,6 @@ ROCKS_TESTS = [
        [],
        [],
    ],
-    [
-        "c_test",
-        "db/c_test.c",
-        "serial",
-        [],
-        [],
-    ],
    [
        "cache_simulator_test",
        "utilities/simulator_cache/cache_simulator_test.cc",
@@ -543,7 +986,7 @@ ROCKS_TESTS = [
    [
        "column_family_test",
        "db/column_family_test.cc",
-        "serial",
+        "parallel",
        [],
        [],
    ],
@@ -596,6 +1039,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "configurable_test",
+        "options/configurable_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "corruption_test",
        "db/corruption_test.cc",
@@ -647,7 +1097,7 @@ ROCKS_TESTS = [
    ],
    [
        "db_blob_index_test",
-        "db/db_blob_index_test.cc",
+        "db/blob/db_blob_index_test.cc",
        "serial",
        [],
        [],
@@ -743,6 +1193,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "db_logical_block_size_cache_test",
+        "db/db_logical_block_size_cache_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "db_memtable_test",
        "db/db_memtable_test.cc",
@@ -751,15 +1208,15 @@ ROCKS_TESTS = [
        [],
    ],
    [
-        "db_merge_operator_test",
-        "db/db_merge_operator_test.cc",
-        "parallel",
+        "db_merge_operand_test",
+        "db/db_merge_operand_test.cc",
+        "serial",
        [],
        [],
    ],
    [
-        "db_merge_operand_test",
-        "db/db_merge_operand_test.cc",
+        "db_merge_operator_test",
+        "db/db_merge_operator_test.cc",
        "parallel",
        [],
        [],
@@ -830,7 +1287,7 @@ ROCKS_TESTS = [
    [
        "db_test2",
        "db/db_test2.cc",
-        "serial",
+        "parallel",
        [],
        [],
    ],
@@ -848,6 +1305,20 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "db_with_timestamp_basic_test",
+        "db/db_with_timestamp_basic_test.cc",
+        "serial",
+        [],
+        [],
+    ],
+    [
+        "db_with_timestamp_compaction_test",
+        "db/db_with_timestamp_compaction_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "db_write_test",
        "db/db_write_test.cc",
@@ -862,6 +1333,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "defer_test",
+        "util/defer_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "delete_scheduler_test",
        "file/delete_scheduler_test.cc",
@@ -912,8 +1390,8 @@ ROCKS_TESTS = [
        [],
    ],
    [
-        "error_handler_test",
-        "db/error_handler_test.cc",
+        "error_handler_fs_test",
+        "db/error_handler_fs_test.cc",
        "serial",
        [],
        [],
@@ -1030,6 +1508,27 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "io_posix_test",
+        "env/io_posix_test.cc",
+        "serial",
+        [],
+        [],
+    ],
+    [
+        "io_tracer_parser_test",
+        "tools/io_tracer_parser_test.cc",
+        "serial",
+        [],
+        [],
+    ],
+    [
+        "io_tracer_test",
+        "trace_replay/io_tracer_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "iostats_context_test",
        "monitoring/iostats_context_test.cc",
@@ -1072,6 +1571,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "memkind_kmem_allocator_test",
+        "memory/memkind_kmem_allocator_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "memory_test",
        "utilities/memory/memory_test.cc",
@@ -1184,6 +1690,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "periodic_work_scheduler_test",
+        "db/periodic_work_scheduler_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "persistent_cache_test",
        "utilities/persistent_cache/persistent_cache_test.cc",
@@ -1198,6 +1711,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "prefetch_test",
+        "file/prefetch_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "prefix_test",
        "db/prefix_test.cc",
@@ -1205,6 +1725,20 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "random_access_file_reader_test",
+        "file/random_access_file_reader_test.cc",
+        "serial",
+        [],
+        [],
+    ],
+    [
+        "random_test",
+        "util/random_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "range_del_aggregator_test",
        "db/range_del_aggregator_test.cc",
@@ -1261,6 +1795,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "slice_test",
+        "util/slice_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "slice_transform_test",
        "util/slice_transform_test.cc",
@@ -1317,6 +1858,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "testutil_test",
+        "test_util/testutil_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "thread_list_test",
        "util/thread_list_test.cc",
@@ -1338,6 +1886,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "timer_test",
+        "util/timer_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "trace_analyzer_test",
        "tools/trace_analyzer_test.cc",
@@ -1345,6 +1900,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "transaction_lock_mgr_test",
+        "utilities/transactions/transaction_lock_mgr_test.cc",
+        "parallel",
+        [],
+        [],
+    ],
    [
        "transaction_test",
        "utilities/transactions/transaction_test.cc",
@@ -1394,6 +1956,13 @@ ROCKS_TESTS = [
        [],
        [],
    ],
+    [
+        "work_queue_test",
+        "util/work_queue_test.cc",
+        "serial",
+        [],
+        [],
+    ],
    [
        "write_batch_test",
        "db/write_batch_test.cc",
@@ -1449,18 +2018,17 @@ 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,
+        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
@@ -97,5 +97,15 @@ 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/.
@@ -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
@@ -17,41 +17,51 @@ environment:
  ZSTD_INCLUDE: $(ZSTD_HOME)\lib;$(ZSTD_HOME)\lib\dictBuilder
  ZSTD_LIB_DEBUG: $(ZSTD_HOME)\build\VS2010\bin\x64_Debug\libzstd_static.lib
  ZSTD_LIB_RELEASE: $(ZSTD_HOME)\build\VS2010\bin\x64_Release\libzstd_static.lib
+  matrix:
+    - APPVEYOR_BUILD_WORKER_IMAGE: Visual Studio 2015
+      CMAKE_GENERATOR: Visual Studio 14 Win64
+      DEV_ENV: C:\Program Files (x86)\Microsoft Visual Studio 14.0\Common7\IDE\devenv.com
+    - APPVEYOR_BUILD_WORKER_IMAGE: Visual Studio 2017
+      CMAKE_GENERATOR: Visual Studio 15 Win64
+      DEV_ENV: C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\Common7\IDE\devenv.com

 install:
  - md %THIRDPARTY_HOME%
  - echo "Building Snappy dependency..."
  - cd %THIRDPARTY_HOME%
-  - curl -fsSL -o snappy-1.1.7.zip https://github.com/google/snappy/archive/1.1.7.zip
+  - curl --fail --silent --show-error --output snappy-1.1.7.zip --location https://github.com/google/snappy/archive/1.1.7.zip
  - unzip snappy-1.1.7.zip
  - cd snappy-1.1.7
  - mkdir build
  - cd build
-  - cmake -DCMAKE_GENERATOR_PLATFORM=x64 ..
+  - 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..."
  - cd %THIRDPARTY_HOME%
-  - curl -fsSL -o lz4-1.8.3.zip https://github.com/lz4/lz4/archive/v1.8.3.zip
+  - curl --fail --silent --show-error --output lz4-1.8.3.zip --location https://github.com/lz4/lz4/archive/v1.8.3.zip
  - unzip lz4-1.8.3.zip
  - cd lz4-1.8.3\visual\VS2010
-  - ps: $CMD="C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\Common7\IDE\devenv.com"; & $CMD lz4.sln /upgrade
+  - ps: $CMD="$Env:DEV_ENV"; & $CMD lz4.sln /upgrade
  - msbuild lz4.sln /p:Configuration=Debug /p:Platform=x64
  - msbuild lz4.sln /p:Configuration=Release /p:Platform=x64
  - echo "Building ZStd dependency..."
  - cd %THIRDPARTY_HOME%
-  - curl -fsSL -o zstd-1.4.0.zip https://github.com/facebook/zstd/archive/v1.4.0.zip
+  - curl --fail --silent --show-error --output zstd-1.4.0.zip --location https://github.com/facebook/zstd/archive/v1.4.0.zip
  - unzip zstd-1.4.0.zip
  - cd zstd-1.4.0\build\VS2010
-  - ps: $CMD="C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\Common7\IDE\devenv.com"; & $CMD zstd.sln /upgrade
+  - ps: $CMD="$Env:DEV_ENV"; & $CMD zstd.sln /upgrade
  - msbuild zstd.sln /p:Configuration=Debug /p:Platform=x64
  - msbuild zstd.sln /p:Configuration=Release /p:Platform=x64

 before_build:
  - md %APPVEYOR_BUILD_FOLDER%\build
  - cd %APPVEYOR_BUILD_FOLDER%\build
-  - cmake -G "Visual Studio 15 Win64" -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:
  project: build\rocksdb.sln
  parallel: true
@@ -60,7 +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
@@ -3,6 +3,10 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 from __future__ import unicode_literals
+try:
+    from builtins import str
+except ImportError:
+    from __builtin__ import str
 from targets_builder import TARGETSBuilder
 import json
 import os
@@ -16,10 +20,10 @@ 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 \
+# $python3 buckifier/buckify_rocksdb.py \
 #        '{"fake": { \
 #                      "extra_deps": [":test_dep", "//fakes/module:mock1"],  \
 #                      "extra_compiler_flags": ["-DROCKSDB_LITE", "-Os"], \
@@ -44,8 +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

@@ -65,27 +69,11 @@ def get_cc_files(repo_path):
    return cc_files


-# Get tests from Makefile
-def get_tests(repo_path):
+# Get parallel tests from Makefile
+def get_parallel_tests(repo_path):
    Makefile = repo_path + "/Makefile"

-    # Dictionary TEST_NAME => IS_PARALLEL
-    tests = {}
-
-    found_tests = False
-    for line in open(Makefile):
-        line = line.strip()
-        if line.startswith("TESTS ="):
-            found_tests = True
-        elif found_tests:
-            if line.endswith("\\"):
-                # remove the trailing \
-                line = line[:-1]
-                line = line.strip()
-                tests[line] = False
-            else:
-                # we consumed all the tests
-                break
+    s = set({})

    found_parallel_tests = False
    for line in open(Makefile):
@@ -97,20 +85,19 @@ def get_tests(repo_path):
                # remove the trailing \
                line = line[:-1]
                line = line.strip()
-                tests[line] = True
+                s.add(line)
            else:
                # we consumed all the parallel tests
                break

-    return tests
-
+    return s

 # Parse extra dependencies passed by user from command line
 def get_dependencies():
    deps_map = {
-        ''.encode('ascii'): {
-            'extra_deps'.encode('ascii'): [],
-            'extra_compiler_flags'.encode('ascii'): []
+        '': {
+            'extra_deps': [],
+            'extra_compiler_flags': []
        }
    }
    if len(sys.argv) < 2:
@@ -119,13 +106,7 @@ def get_dependencies():
    def encode_dict(data):
        rv = {}
        for k, v in data.items():
-            if isinstance(k, unicode):
-                k = k.encode('ascii')
-            if isinstance(v, unicode):
-                v = v.encode('ascii')
-            elif isinstance(v, list):
-                v = [x.encode('ascii') for x in v]
-            elif isinstance(v, dict):
+            if isinstance(v, dict):
                v = encode_dict(v)
            rv[k] = v
        return rv
@@ -142,18 +123,28 @@ 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 parallel tests from Makefile
+    parallel_tests = get_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 parallel_tests is None:
        return False

    TARGETS = TARGETSBuilder("%s/TARGETS" % repo_path)
+
    # rocksdb_lib
    TARGETS.add_library(
        "rocksdb_lib",
        src_mk["LIB_SOURCES"] +
        src_mk["TOOL_LIB_SOURCES"])
+    # rocksdb_whole_archive_lib
+    TARGETS.add_library(
+        "rocksdb_whole_archive_lib",
+        src_mk["LIB_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",
@@ -161,7 +152,10 @@ 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",
@@ -169,34 +163,51 @@ def generate_targets(repo_path, deps_map):
        src_mk.get("ANALYZER_LIB_SOURCES", []) +
        ["test_util/testutil.cc"],
        [":rocksdb_lib"])
+    # rocksdb_stress_lib
+    TARGETS.add_rocksdb_library(
+        "rocksdb_stress_lib",
+        src_mk.get("ANALYZER_LIB_SOURCES", [])
+        + src_mk.get('STRESS_LIB_SOURCES', [])
+        + ["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 in 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:"))
@@ -215,6 +226,7 @@ def get_rocksdb_path():

    return rocksdb_path

+
 def exit_with_error(msg):
    print(ColorString.error(msg))
    sys.exit(1)
@@ -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
@@ -3,6 +3,12 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 from __future__ import unicode_literals
+try:
+    from builtins import object
+    from builtins import str
+except ImportError:
+    from __builtin__ import object
+    from __builtin__ import str
 import targets_cfg

 def pretty_list(lst, indent=8):
@@ -18,11 +24,12 @@ def pretty_list(lst, indent=8):
    return res


-class TARGETSBuilder:
+class TARGETSBuilder(object):
    def __init__(self, path):
        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
+        self.targets_file.write(header.encode("utf-8"))
        self.total_lib = 0
        self.total_bin = 0
        self.total_test = 0
@@ -31,26 +38,69 @@ class TARGETSBuilder:
    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"""
+if not is_opt_mode:
+    cpp_binary(
+        name = "c_test_bin",
+        srcs = ["db/c_test.c"],
+        arch_preprocessor_flags = ROCKSDB_ARCH_PREPROCESSOR_FLAGS,
+        os_preprocessor_flags = ROCKSDB_OS_PREPROCESSOR_FLAGS,
+        compiler_flags = ROCKSDB_COMPILER_FLAGS,
+        preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
+        deps = [":rocksdb_test_lib"],
+    )
+
+if not is_opt_mode:
+    custom_unittest(
+        "c_test",
+        command = [
+            native.package_name() + "/buckifier/rocks_test_runner.sh",
+            "$(location :{})".format("c_test_bin"),
+        ],
+        type = "simple",
+    )
+""")
+
    def register_test(self,
                      test_name,
                      src,
@@ -70,5 +120,5 @@ class TARGETSBuilder:
        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 = ""
@@ -3,7 +3,14 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 from __future__ import unicode_literals
-rocksdb_target_header = """load("@fbcode_macros//build_defs:auto_headers.bzl", "AutoHeaders")
+
+rocksdb_target_header_template = \
+    """# This file \100generated by `python3 buckifier/buckify_rocksdb.py`
+# --> DO NOT EDIT MANUALLY <--
+# This file is a Facebook-specific integration for buck builds, so can
+# only be validated by Facebook employees.
+#
+load("@fbcode_macros//build_defs:auto_headers.bzl", "AutoHeaders")
 load("@fbcode_macros//build_defs:cpp_library.bzl", "cpp_library")
 load(":defs.bzl", "test_binary")

@@ -25,14 +32,16 @@ ROCKSDB_EXTERNAL_DEPS = [
    ("gflags", None, "gflags"),
    ("lz4", None, "lz4"),
    ("zstd", None),
-    ("tbb", None),
-    ("googletest", None, "gtest"),
 ]

 ROCKSDB_OS_DEPS = [
    (
        "linux",
-        ["third-party//numa:numa"],
+        ["third-party//numa:numa", "third-party//liburing:uring", "third-party//tbb:tbb"],
+    ),
+    (
+        "macos",
+        ["third-party//tbb:tbb"],
    ),
 ]

@@ -46,19 +55,31 @@ ROCKSDB_OS_PREPROCESSOR_FLAGS = [
            "-DROCKSDB_PTHREAD_ADAPTIVE_MUTEX",
            "-DROCKSDB_RANGESYNC_PRESENT",
            "-DROCKSDB_SCHED_GETCPU_PRESENT",
+            "-DROCKSDB_IOURING_PRESENT",
            "-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
@@ -69,7 +90,6 @@ ROCKSDB_PREPROCESSOR_FLAGS = [
    "-DZSTD",
    "-DZSTD_STATIC_LINKING_ONLY",
    "-DGFLAGS=gflags",
-    "-DTBB",

    # Added missing flags from output of build_detect_platform
    "-DROCKSDB_BACKTRACE",
@@ -106,6 +126,11 @@ 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"]
 """


@@ -120,18 +145,34 @@ cpp_library(
    os_preprocessor_flags = ROCKSDB_OS_PREPROCESSOR_FLAGS,
    preprocessor_flags = ROCKSDB_PREPROCESSOR_FLAGS,
    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,
+    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],
+    deps = [{deps}],
    external_deps = ROCKSDB_EXTERNAL_DEPS,
 )
 """
@@ -148,24 +189,23 @@ 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,
+        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
@@ -6,11 +6,16 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 from __future__ import unicode_literals
+try:
+    from builtins import object
+except ImportError:
+    from __builtin__ import object
 import subprocess
+import sys
 import os
 import time

-class ColorString:
+class ColorString(object):
    """ Generate colorful strings on terminal """
    HEADER = '\033[95m'
    BLUE = '\033[94m'
@@ -21,7 +26,13 @@ class ColorString:

    @staticmethod
    def _make_color_str(text, color):
-        return "".join([color, text.encode('utf-8'), ColorString.ENDC])
+        # In Python2, default encoding for unicode string is ASCII
+        if sys.version_info.major <= 2:
+            return "".join(
+                [color, text.encode('utf-8'), ColorString.ENDC])
+        # From Python3, default encoding for unicode string is UTF-8
+        return "".join(
+            [color, text, ColorString.ENDC])

    @staticmethod
    def ok(text):
@@ -1,377 +0,0 @@
-<?php
-// Copyright 2004-present Facebook. 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).
-
-// Name of the environment variables which need to be set by the entity which
-// triggers continuous runs so that code at the end of the file gets executed
-// and Sandcastle run starts.
-const ENV_POST_RECEIVE_HOOK = "POST_RECEIVE_HOOK";
-const ENV_HTTPS_APP_VALUE = "HTTPS_APP_VALUE";
-const ENV_HTTPS_TOKEN_VALUE = "HTTPS_TOKEN_VALUE";
-
-const PRIMARY_TOKEN_FILE = '/home/krad/.sandcastle';
-const CONT_RUN_ALIAS = "leveldb";
-
-//////////////////////////////////////////////////////////////////////
-/*  Run tests in sandcastle */
-function postURL($diffID, $url) {
-  assert(strlen($diffID) > 0);
-  assert(is_numeric($diffID));
-  assert(strlen($url) > 0);
-
-  $cmd_args = array(
-    'diff_id' => (int)$diffID,
-    'name' => sprintf(
-      'click here for sandcastle tests for D%d',
-      (int)$diffID
-    ),
-    'link' => $url
-  );
-  $cmd = 'echo ' . escapeshellarg(json_encode($cmd_args))
-         . ' | arc call-conduit differential.updateunitresults';
-
-  shell_exec($cmd);
-}
-
-function buildUpdateTestStatusCmd($diffID, $test, $status) {
-  assert(strlen($diffID) > 0);
-  assert(is_numeric($diffID));
-  assert(strlen($test) > 0);
-  assert(strlen($status) > 0);
-
-  $cmd_args = array(
-    'diff_id' => (int)$diffID,
-    'name' => $test,
-    'result' => $status
-  );
-
-  $cmd = 'echo ' . escapeshellarg(json_encode($cmd_args))
-         . ' | arc call-conduit differential.updateunitresults';
-
-  return $cmd;
-}
-
-function updateTestStatus($diffID, $test) {
-  assert(strlen($diffID) > 0);
-  assert(is_numeric($diffID));
-  assert(strlen($test) > 0);
-
-  shell_exec(buildUpdateTestStatusCmd($diffID, $test, "waiting"));
-}
-
-function getSteps($applyDiff, $diffID, $username, $test) {
-  assert(strlen($username) > 0);
-  assert(strlen($test) > 0);
-
-  if ($applyDiff) {
-    assert(strlen($diffID) > 0);
-    assert(is_numeric($diffID));
-
-    $arcrc_content = (PHP_OS == "Darwin" ?
-        exec("cat ~/.arcrc | gzip -f | base64") :
-            exec("cat ~/.arcrc | gzip -f | base64 -w0"));
-    assert(strlen($arcrc_content) > 0);
-
-    // Sandcastle machines don't have arc setup. We copy the user certificate
-    // and authenticate using that in Sandcastle.
-    $setup = array(
-      "name" => "Setup arcrc",
-      "shell" => "echo " . escapeshellarg($arcrc_content) . " | base64 --decode"
-                 . " | gzip -d > ~/.arcrc",
-      "user" => "root"
-    );
-
-    // arc demands certain permission on its config.
-    // also fix the sticky bit issue in sandcastle
-    $fix_permission = array(
-      "name" => "Fix environment",
-      "shell" => "chmod 600 ~/.arcrc && chmod +t /dev/shm",
-      "user" => "root"
-    );
-
-    // Construct the steps in the order of execution.
-    $steps[] = $setup;
-    $steps[] = $fix_permission;
-  }
-
-  // fbcode is a sub-repo. We cannot patch until we add it to ignore otherwise
-  // Git thinks it is an uncommitted change.
-  $fix_git_ignore = array(
-    "name" => "Fix git ignore",
-    "shell" => "echo fbcode >> .git/info/exclude",
-    "user" => "root"
-  );
-
-  // This fixes "FATAL: ThreadSanitizer can not mmap the shadow memory"
-  // Source:
-  // https://github.com/google/sanitizers/wiki/ThreadSanitizerCppManual#FAQ
-  $fix_kernel_issue = array(
-    "name" => "Fix kernel issue with tsan",
-    "shell" => "echo 2 >/proc/sys/kernel/randomize_va_space",
-    "user" => "root"
-  );
-
-  $steps[] = $fix_git_ignore;
-  $steps[] = $fix_kernel_issue;
-
-  // This will be the command used to execute particular type of tests.
-  $cmd = "";
-
-  if ($applyDiff) {
-    // Patch the code (keep your fingures crossed).
-    $patch = array(
-      "name" => "Patch " . $diffID,
-      "shell" => "arc --arcrc-file ~/.arcrc "
-                  . "patch --nocommit --diff " . escapeshellarg($diffID),
-      "user" => "root"
-    );
-
-    $steps[] = $patch;
-
-    updateTestStatus($diffID, $test);
-    $cmd = buildUpdateTestStatusCmd($diffID, $test, "running") . "; ";
-  }
-
-  // Run the actual command.
-  $cmd = $cmd . "J=$(nproc) ./build_tools/precommit_checker.py " .
-           escapeshellarg($test) . "; exit_code=$?; ";
-
-  if ($applyDiff) {
-    $cmd = $cmd . "([[ \$exit_code -eq 0 ]] &&"
-                . buildUpdateTestStatusCmd($diffID, $test, "pass") . ")"
-                . "||" . buildUpdateTestStatusCmd($diffID, $test, "fail")
-                . "; ";
-  }
-
-  // shell command to sort the tests based on exit code and print
-  // the output of the log files.
-  $cat_sorted_logs = "
-    while read code log_file;
-      do echo \"################ cat \$log_file [exit_code : \$code] ################\";
-      cat \$log_file;
-    done < <(tail -n +2 LOG | sort -k7,7n -k4,4gr | awk '{print \$7,\$NF}')";
-
-  // Shell command to cat all log files
-  $cat_all_logs = "for f in `ls t/!(run-*)`; do echo \$f;cat \$f; done";
-
-  // If LOG file exist use it to cat log files sorted by exit code, otherwise
-  // cat everything
-  $logs_cmd = "if [ -f LOG ]; then {$cat_sorted_logs}; else {$cat_all_logs}; fi";
-
-  $cmd = $cmd . " cat /tmp/precommit-check.log"
-              . "; shopt -s extglob; {$logs_cmd}"
-              . "; shopt -u extglob; [[ \$exit_code -eq 0 ]]";
-  assert(strlen($cmd) > 0);
-
-  $run_test = array(
-    "name" => "Run " . $test,
-    "shell" => $cmd,
-    "user" => "root",
-    "parser" => "python build_tools/error_filter.py " . escapeshellarg($test),
-  );
-
-  $steps[] = $run_test;
-
-  if ($applyDiff) {
-    // Clean up the user arc config we are using.
-    $cleanup = array(
-      "name" => "Arc cleanup",
-      "shell" => "rm -f ~/.arcrc",
-      "user" => "root"
-    );
-
-    $steps[] = $cleanup;
-  }
-
-  assert(count($steps) > 0);
-  return $steps;
-}
-
-function getSandcastleConfig() {
-  $sandcastle_config = array();
-
-  $cwd = getcwd();
-  $cwd_token_file = "{$cwd}/.sandcastle";
-  // This is a case when we're executed from a continuous run. Fetch the values
-  // from the environment.
-  if (getenv(ENV_POST_RECEIVE_HOOK)) {
-    $sandcastle_config[0] = getenv(ENV_HTTPS_APP_VALUE);
-    $sandcastle_config[1] = getenv(ENV_HTTPS_TOKEN_VALUE);
-  } else {
-    // This is a typical `[p]arc diff` case. Fetch the values from the specific
-    // configuration files.
-    for ($i = 0; $i < 50; $i++) {
-      if (file_exists(PRIMARY_TOKEN_FILE) ||
-          file_exists($cwd_token_file)) {
-        break;
-      }
-      // If we failed to fetch the tokens, sleep for 0.2 second and try again
-      usleep(200000);
-    }
-    assert(file_exists(PRIMARY_TOKEN_FILE) ||
-           file_exists($cwd_token_file));
-
-    // Try the primary location first, followed by a secondary.
-    if (file_exists(PRIMARY_TOKEN_FILE)) {
-      $cmd = 'cat ' . PRIMARY_TOKEN_FILE;
-    } else {
-      $cmd = 'cat ' . escapeshellarg($cwd_token_file);
-    }
-
-    assert(strlen($cmd) > 0);
-    $sandcastle_config = explode(':', rtrim(shell_exec($cmd)));
-  }
-
-  // In this case be very explicit about the implications.
-  if (count($sandcastle_config) != 2) {
-    echo "Sandcastle configuration files don't contain valid information " .
-         "or the necessary environment variables aren't defined. Unable " .
-         "to validate the code changes.";
-    exit(1);
-  }
-
-  assert(strlen($sandcastle_config[0]) > 0);
-  assert(strlen($sandcastle_config[1]) > 0);
-  assert(count($sandcastle_config) > 0);
-
-  return $sandcastle_config;
-}
-
-// This function can be called either from `[p]arc diff` command or during
-// the Git post-receive hook.
- function startTestsInSandcastle($applyDiff, $workflow, $diffID) {
-  // Default options don't terminate on failure, but that's what we want. In
-  // the current case we use assertions intentionally as "terminate on failure
-  // invariants".
-  assert_options(ASSERT_BAIL, true);
-
-  // In case of a diff we'll send notificatios to the author. Else it'll go to
-  // the entire team because failures indicate that build quality has regressed.
-  $username = $applyDiff ? exec("whoami") : CONT_RUN_ALIAS;
-  assert(strlen($username) > 0);
-
-  if ($applyDiff) {
-    assert($workflow);
-    assert(strlen($diffID) > 0);
-    assert(is_numeric($diffID));
-  }
-
-  // List of tests we want to run in Sandcastle.
-  $tests = array("unit", "unit_non_shm", "unit_481", "clang_unit", "tsan",
-                 "asan", "lite_test", "valgrind", "release", "release_481",
-                 "clang_release", "clang_analyze", "code_cov",
-                 "java_build", "no_compression", "unity", "ubsan");
-
-  $send_email_template = array(
-    'type' => 'email',
-    'triggers' => array('fail'),
-    'emails' => array($username . '@fb.com'),
-  );
-
-  // Construct a job definition for each test and add it to the master plan.
-  foreach ($tests as $test) {
-    $stepName = "RocksDB diff " . $diffID . " test " . $test;
-
-    if (!$applyDiff) {
-      $stepName = "RocksDB continuous integration test " . $test;
-    }
-
-    $arg[] = array(
-      "name" => $stepName,
-      "report" => array($send_email_template),
-      "steps" => getSteps($applyDiff, $diffID, $username, $test)
-    );
-  }
-
-  // We cannot submit the parallel execution master plan to Sandcastle and
-  // need supply the job plan as a determinator. So we construct a small job
-  // that will spit out the master job plan which Sandcastle will parse and
-  // execute. Why compress the job definitions? Otherwise we run over the max
-  // string size.
-  $cmd = "echo " . base64_encode(json_encode($arg))
-         . (PHP_OS == "Darwin" ?
-             " | gzip -f | base64" :
-                 " | gzip -f | base64 -w0");
-  assert(strlen($cmd) > 0);
-
-  $arg_encoded = shell_exec($cmd);
-  assert(strlen($arg_encoded) > 0);
-
-  $runName = "Run diff " . $diffID . "for user " . $username;
-
-  if (!$applyDiff) {
-    $runName = "RocksDB continuous integration build and test run";
-  }
-
-  $command = array(
-    "name" => $runName,
-    "steps" => array()
-  );
-
-  $command["steps"][] = array(
-    "name" => "Generate determinator",
-    "shell" => "echo " . $arg_encoded . " | base64 --decode | gzip -d"
-               . " | base64 --decode",
-    "determinator" => true,
-    "user" => "root"
-  );
-
-  // Submit to Sandcastle.
-  $url = 'https://interngraph.intern.facebook.com/sandcastle/create';
-
-  $job = array(
-    'command' => 'SandcastleUniversalCommand',
-    'args' => $command,
-    'capabilities' => array(
-      'vcs' => 'rocksdb-int-git',
-      'type' => 'lego',
-    ),
-    'hash' => 'origin/master',
-    'user' => $username,
-    'alias' => 'rocksdb-precommit',
-    'tags' => array('rocksdb'),
-    'description' => 'Rocksdb precommit job',
-  );
-
-  // Fetch the configuration necessary to submit a successful HTTPS request.
-  $sandcastle_config = getSandcastleConfig();
-
-  $app = $sandcastle_config[0];
-  $token = $sandcastle_config[1];
-
-  $cmd = 'curl -s -k '
-          . ' -F app=' . escapeshellarg($app)
-          . ' -F token=' . escapeshellarg($token)
-          . ' -F job=' . escapeshellarg(json_encode($job))
-          .' ' . escapeshellarg($url);
-
-  $output = shell_exec($cmd);
-  assert(strlen($output) > 0);
-
-  // Extract Sandcastle URL from the response.
-  preg_match('/url": "(.+)"/', $output, $sandcastle_url);
-
-  assert(count($sandcastle_url) > 0, "Unable to submit Sandcastle request.");
-  assert(strlen($sandcastle_url[1]) > 0, "Unable to extract Sandcastle URL.");
-
-  if ($applyDiff) {
-    echo "\nSandcastle URL: " . $sandcastle_url[1] . "\n";
-    // Ask Phabricator to display it on the diff UI.
-    postURL($diffID, $sandcastle_url[1]);
-  } else {
-    echo "Continuous integration started Sandcastle tests. You can look at ";
-    echo "the progress at:\n" . $sandcastle_url[1] . "\n";
-  }
-}
-
-// Continuous run cript will set the environment variable and based on that
-// we'll trigger the execution of tests in Sandcastle. In that case we don't
-// need to apply any diffs and there's no associated workflow either.
-if (getenv(ENV_POST_RECEIVE_HOOK)) {
-  startTestsInSandcastle(
-    false /* $applyDiff */,
-    NULL /* $workflow */,
-    NULL /* $diffID */);
-}
@@ -9,6 +9,7 @@
 #   PLATFORM_LDFLAGS            Linker flags
 #   JAVA_LDFLAGS                Linker flags for RocksDBJava
 #   JAVA_STATIC_LDFLAGS         Linker flags for RocksDBJava static build
+#   JAVAC_ARGS                  Arguments for javac
 #   PLATFORM_SHARED_EXT         Extension for shared libraries
 #   PLATFORM_SHARED_LDFLAGS     Flags for building shared library
 #   PLATFORM_SHARED_CFLAGS      Flags for compiling objects for shared library
@@ -27,6 +28,7 @@
 #       -DZSTD                      if the ZSTD library is present
 #       -DNUMA                      if the NUMA library is present
 #       -DTBB                       if the TBB library is present
+#       -DMEMKIND                   if the memkind library is present
 #
 # Using gflags in rocksdb:
 # Our project depends on gflags, which requires users to take some extra steps
@@ -56,10 +58,18 @@ if [ -z "$ROCKSDB_NO_FBCODE" -a -d /mnt/gvfs/third-party ]; then
    if [ -n "$ROCKSDB_FBCODE_BUILD_WITH_481" ]; then
      # we need this to build with MySQL. Don't use for other purposes.
      source "$PWD/build_tools/fbcode_config4.8.1.sh"
+    elif [ -n "$ROCKSDB_FBCODE_BUILD_WITH_5xx" ]; then
+      source "$PWD/build_tools/fbcode_config.sh"
    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 "$PWD/build_tools/fbcode_config.sh"
+      source "$PWD/build_tools/fbcode_config_platform009.sh"
    fi
 fi

@@ -87,6 +97,16 @@ 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`
@@ -149,7 +169,22 @@ 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 $CFLAGS -x c++ - -luring -o /dev/null 2>/dev/null  <<EOF
+              #include <liburing.h>
+              int main() {
+                struct io_uring ring;
+                io_uring_queue_init(1, &ring, 0);
+                return 0;
+              }
+EOF
+            if [ "$?" = 0 ]; then
+                PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -luring"
+                COMMON_FLAGS="$COMMON_FLAGS -DROCKSDB_IOURING_PRESENT"
+            fi
+        fi
        if test -z "$USE_FOLLY_DISTRIBUTED_MUTEX"; then
          USE_FOLLY_DISTRIBUTED_MUTEX=1
        fi
@@ -175,6 +210,17 @@ case "$TARGET_OS" in
        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="$COMMON_FLAGS -fno-builtin-memcmp -D_REENTRANT -DOS_NETBSD"
@@ -224,10 +270,15 @@ esac
 PLATFORM_CXXFLAGS="$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
@@ -263,24 +314,32 @@ 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 $CFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
          #include <gflags/gflags.h>
+          using namespace GFLAGS_NAMESPACE;
          int main() {}
 EOF
-        if [ "$?" = 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 $CFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
+            #include <gflags/gflags.h>
+            using namespace gflags;
+            int main() {}
+EOF
+        then
+          COMMON_FLAGS="$COMMON_FLAGS -DGFLAGS=1 -DGFLAGS_NAMESPACE=gflags"
+          PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lgflags"
+        # check if namespace is google
+        elif $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null << EOF
            #include <gflags/gflags.h>
            using namespace google;
            int main() {}
 EOF
-          if [ "$?" = 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

@@ -409,6 +468,22 @@ EOF
        fi
    fi

+    if ! test $ROCKSDB_DISABLE_MEMKIND; then
+        # Test whether memkind library is installed
+        $CXX $CFLAGS $COMMON_FLAGS -lmemkind -x c++ - -o /dev/null 2>/dev/null  <<EOF
+          #include <memkind.h>
+          int main() {
+            memkind_malloc(MEMKIND_DAX_KMEM, 1024);
+            return 0;
+          }
+EOF
+        if [ "$?" = 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
@@ -493,6 +568,20 @@ EOF
        fi
    fi

+    if ! test $ROCKSDB_DISABLE_AUXV_GETAUXVAL; then
+        # Test whether getauxval is supported
+        $CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null  <<EOF
+          #include <sys/auxv.h>
+          int main() {
+            uint64_t auxv = getauxval(AT_HWCAP);
+            (void)auxv;
+          }
+EOF
+        if [ "$?" = 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 $PLATFORM_CXXFLAGS -faligned-new -x c++ - -o /dev/null 2>/dev/null <<EOF
@@ -560,6 +649,13 @@ else
  if test "$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"
+  fi
 fi

 if test "$TRY_SSE_ETC"; then
@@ -569,9 +665,18 @@ if test "$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 $TRY_SSE42 -x c++ - -o /dev/null 2>/dev/null <<EOF
@@ -605,6 +710,62 @@ elif test "$USE_SSE"; then
  echo "warning: USE_SSE specified but compiler could not use PCLMUL intrinsics, disabling" >&2
 fi

+$CXX $PLATFORM_CXXFLAGS $COMMON_FLAGS $TRY_AVX2 -x c++ - -o /dev/null 2>/dev/null <<EOF
+  #include <cstdint>
+  #include <immintrin.h>
+  int main() {
+    const auto a = _mm256_setr_epi32(0, 1, 2, 3, 4, 7, 6, 5);
+    const auto b = _mm256_permutevar8x32_epi32(a, a);
+    (void)b;
+  }
+EOF
+if [ "$?" = 0 ]; then
+  COMMON_FLAGS="$COMMON_FLAGS $TRY_AVX2 -DHAVE_AVX2"
+elif test "$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 $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null <<EOF
+  #include <cstdint>
+  int main() {
+    uint64_t a = 0xffffFFFFffffFFFF;
+    __uint128_t b = __uint128_t(a) * a;
+    a = static_cast<uint64_t>(b >> 64);
+    (void)a;
+  }
+EOF
+if [ "$?" = 0 ]; then
+  COMMON_FLAGS="$COMMON_FLAGS -DHAVE_UINT128_EXTENSION"
+fi
+
 # iOS doesn't support thread-local storage, but this check would erroneously
 # succeed because the cross-compiler flags are added by the Makefile, not this
 # script.
@@ -623,6 +784,7 @@ 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() {}
@@ -647,10 +809,12 @@ 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 "JAVA_LDFLAGS=$JAVA_LDFLAGS" >> "$OUTPUT"
 echo "JAVA_STATIC_LDFLAGS=$JAVA_STATIC_LDFLAGS" >> "$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"
@@ -1,137 +0,0 @@
-#!/usr/bin/env bash
-#
-# Copyright (c) 2016, Facebook. All rights reserved.
-#
-# Overall wrapper script for RocksDB continuous builds. The implementation is a
-# trivial pulling scheme. We loop infinitely, check if any new changes have been
-# committed, if yes then trigger a Sandcastle run, and finally go to sleep again
-# for a certain interval.
-#
-
-SRC_GIT_REPO=/data/git/rocksdb-public
-error=0
-
-function log {
-  DATE=`date +%Y-%m-%d:%H:%M:%S`
-  # shellcheck disable=SC2068
-  echo $DATE $@
-}
-
-function log_err {
-  # shellcheck disable=SC2145
-  log "ERROR: $@ Error code: $error."
-}
-
-function update_repo_status {
-  # Update the parent first.
-  pushd $SRC_GIT_REPO
-
-  # This is a fatal error. Something in the environment isn't right and we will
-  # terminate the execution.
-  error=$?
-  if [ ! $error -eq 0 ]; then
-    log_err "Where is $SRC_GIT_REPO?"
-    exit $error
-  fi
-
-  HTTPS_PROXY=fwdproxy:8080 git fetch -f
-
-  error=$?
-  if [ ! $error -eq 0 ]; then
-    log_err "git fetch -f failed."
-    popd
-    return $error
-  fi
-
-  git update-ref refs/heads/master refs/remotes/origin/master
-
-  error=$?
-  if [ ! $error -eq 0 ]; then
-    log_err "git update-ref failed."
-    popd
-    return $error
-  fi
-
-  popd
-
-  # We're back in an instance-specific directory. Get the latest changes.
-  git pull --rebase
-
-  error=$?
-  if [ ! $error -eq 0 ]; then
-    log_err "git pull --rebase failed."
-    return $error
-  fi
-}
-
-#
-# Execution starts here.
-#
-
-# Path to the determinator from the root of the RocksDB repo.
-CONTRUN_DETERMINATOR=./build_tools/RocksDBCommonHelper.php
-
-# Value of the previous commit.
-PREV_COMMIT=
-
-log "Starting to monitor for new RocksDB changes ..."
-log "Running under `pwd` as `whoami`."
-
-# Paranoia. Make sure that we're using the right branch.
-git checkout master
-
-error=$?
-if [ ! $error -eq 0 ]; then
-  log_err "This is not good. Can't checkout master. Bye-bye!"
-  exit 1
-fi
-
-# We'll run forever and let the execution environment terminate us if we'll
-# exceed whatever timeout is set for the job.
-while true;
-do
-  # Get the latest changes committed.
-  update_repo_status
-
-  error=$?
-  if [  $error -eq 0 ]; then
-    LAST_COMMIT=`git log -1 | head -1 | grep commit | awk '{ print $2; }'`
-
-    log "Last commit is '$LAST_COMMIT', previous commit is '$PREV_COMMIT'."
-
-    if [ "$PREV_COMMIT" == "$LAST_COMMIT" ]; then
-      log "There were no changes since the last time I checked. Going to sleep."
-    else
-      if [ ! -z "$LAST_COMMIT" ]; then
-        log "New code has been committed or previous commit not known. " \
-            "Will trigger the tests."
-
-        PREV_COMMIT=$LAST_COMMIT
-        log "Updated previous commit to '$PREV_COMMIT'."
-
-        #
-        # This is where we'll trigger the Sandcastle run. The values for
-        # HTTPS_APP_VALUE and HTTPS_APP_VALUE will be set in the container we're
-        # running in.
-        #
-        POST_RECEIVE_HOOK=1 php $CONTRUN_DETERMINATOR
-
-        error=$?
-        if [ $error -eq 0 ]; then
-          log "Sandcastle run successfully triggered."
-        else
-          log_err "Failed to trigger Sandcastle run."
-        fi
-      else
-        log_err "Previous commit not updated. Don't know what the last one is."
-      fi
-    fi
-  else
-    log_err "Getting latest changes failed. Will skip running tests for now."
-  fi
-
-  # Always sleep, even if errors happens while trying to determine the latest
-  # commit. This will prevent us terminating in case of transient errors.
-  log "Will go to sleep for 5 minutes."
-  sleep 5m
-done
@@ -1,19 +1,19 @@
 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
-GCC_BASE=/mnt/gvfs/third-party2/gcc/112ec378fec7002ad3e09afde022e656049f7191/5.x/centos7-native/c447969
-CLANG_BASE=/mnt/gvfs/third-party2/llvm-fb/04999bdb3ce81a11073535dcb00b5e13dc1cbaf5/stable/centos7-native/c9f9104
-LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/92b0c8e5c8eecc71eb042594ce1ab3413799b385/5.x/gcc-5-glibc-2.23/339d858
-GLIBC_BASE=/mnt/gvfs/third-party2/glibc/3d8698d5973ba94f41620a80a67e4457fdf01e90/2.23/gcc-5-glibc-2.23/ca1d1c0
+GCC_BASE=/mnt/gvfs/third-party2/gcc/7331085db891a2ef4a88a48a751d834e8d68f4cb/5.x/centos7-native/c447969
+CLANG_BASE=/mnt/gvfs/third-party2/llvm-fb/1bd23f9917738974ad0ff305aa23eb5f93f18305/9.0.0/centos7-native/c9f9104
+LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/6ace84e956873d53638c738b6f65f3f469cca74c/5.x/gcc-5-glibc-2.23/339d858
+GLIBC_BASE=/mnt/gvfs/third-party2/glibc/192b0f42d63dcf6210d6ceae387b49af049e6e0c/2.23/gcc-5-glibc-2.23/ca1d1c0
 SNAPPY_BASE=/mnt/gvfs/third-party2/snappy/7f9bdaada18f59bc27ec2b0871eb8a6144343aef/1.1.3/gcc-5-glibc-2.23/9bc6787
-ZLIB_BASE=/mnt/gvfs/third-party2/zlib/22c2d65676fb7c23cfa797c4f6937f38b026f3cf/1.2.8/gcc-5-glibc-2.23/9bc6787
+ZLIB_BASE=/mnt/gvfs/third-party2/zlib/2d9f0b9a4274cc21f61272a9e89bdb859bce8f1f/1.2.8/gcc-5-glibc-2.23/9bc6787
 BZIP2_BASE=/mnt/gvfs/third-party2/bzip2/dc49a21c5fceec6456a7a28a94dcd16690af1337/1.0.6/gcc-5-glibc-2.23/9bc6787
-LZ4_BASE=/mnt/gvfs/third-party2/lz4/907b498203d297947f3bb70b9466f47e100f1873/r131/gcc-5-glibc-2.23/9bc6787
-ZSTD_BASE=/mnt/gvfs/third-party2/zstd/af6628a46758f1a15484a1760cd7294164bc5ba1/1.3.5/gcc-5-glibc-2.23/03859b5
+LZ4_BASE=/mnt/gvfs/third-party2/lz4/0f607f8fc442ea7d6b876931b1898bb573d5e5da/1.9.1/gcc-5-glibc-2.23/9bc6787
+ZSTD_BASE=/mnt/gvfs/third-party2/zstd/ca22bc441a4eb709e9e0b1f9fec9750fed7b31c5/1.4.x/gcc-5-glibc-2.23/03859b5
 GFLAGS_BASE=/mnt/gvfs/third-party2/gflags/0b9929d2588991c65a57168bf88aff2db87c5d48/2.2.0/gcc-5-glibc-2.23/9bc6787
-JEMALLOC_BASE=/mnt/gvfs/third-party2/jemalloc/b1a0e56c1e3e6929813a4331ade3a58ff083afbb/master/gcc-5-glibc-2.23/aa64d6b
-NUMA_BASE=/mnt/gvfs/third-party2/numa/9cbf2460284c669ed19c3ccb200a71f7dd7e53c7/2.0.11/gcc-5-glibc-2.23/9bc6787
-LIBUNWIND_BASE=/mnt/gvfs/third-party2/libunwind/bf3d7497fe4e6d007354f0adffa16ce3003f8338/1.3/gcc-5-glibc-2.23/b443de1
-TBB_BASE=/mnt/gvfs/third-party2/tbb/ff4e0b093534704d8abab678a4fd7f5ea7b094c7/2018_U5/gcc-5-glibc-2.23/9bc6787
-KERNEL_HEADERS_BASE=/mnt/gvfs/third-party2/kernel-headers/b5c4a61a5c483ba24722005ae07895971a2ac707/4.0.9-36_fbk5_2933_gd092e3f/gcc-5-glibc-2.23/da39a3e
-BINUTILS_BASE=/mnt/gvfs/third-party2/binutils/55031de95a2b46c82948743419a603b3d6aefe28/2.29.1/centos7-native/da39a3e
-VALGRIND_BASE=/mnt/gvfs/third-party2/valgrind/f3f697a28122e6bcd513273dd9c1ff23852fc59f/3.13.0/gcc-5-glibc-2.23/9bc6787
+JEMALLOC_BASE=/mnt/gvfs/third-party2/jemalloc/c26f08f47ac35fc31da2633b7da92d6b863246eb/master/gcc-5-glibc-2.23/0c8f76d
+NUMA_BASE=/mnt/gvfs/third-party2/numa/3f3fb57a5ccc5fd21c66416c0b83e0aa76a05376/2.0.11/gcc-5-glibc-2.23/9bc6787
+LIBUNWIND_BASE=/mnt/gvfs/third-party2/libunwind/40c73d874898b386a71847f1b99115d93822d11f/1.4/gcc-5-glibc-2.23/b443de1
+TBB_BASE=/mnt/gvfs/third-party2/tbb/4ce8e8dba77cdbd81b75d6f0c32fd7a1b76a11ec/2018_U5/gcc-5-glibc-2.23/9bc6787
+KERNEL_HEADERS_BASE=/mnt/gvfs/third-party2/kernel-headers/fb251ecd2f5ae16f8671f7014c246e52a748fe0b/4.0.9-36_fbk5_2933_gd092e3f/gcc-5-glibc-2.23/da39a3e
+BINUTILS_BASE=/mnt/gvfs/third-party2/binutils/2e3cb7d119b3cea5f1e738cc13a1ac69f49eb875/2.29.1/centos7-native/da39a3e
+VALGRIND_BASE=/mnt/gvfs/third-party2/valgrind/d42d152a15636529b0861ec493927200ebebca8e/3.15.0/gcc-5-glibc-2.23/9bc6787
 LUA_BASE=/mnt/gvfs/third-party2/lua/f0cd714433206d5139df61659eb7b28b1dea6683/5.2.3/gcc-5-glibc-2.23/65372bd
@@ -1,19 +1,20 @@
 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
-GCC_BASE=/mnt/gvfs/third-party2/gcc/6e8e715624fd15256a7970073387793dfcf79b46/7.x/centos7-native/b2ef2b6
-CLANG_BASE=/mnt/gvfs/third-party2/llvm-fb/ef37e1faa1c29782abfac1ae65a291b9b7966f6d/stable/centos7-native/c9f9104
-LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/c67031f0f739ac61575a061518d6ef5038f99f90/7.x/platform007/5620abc
-GLIBC_BASE=/mnt/gvfs/third-party2/glibc/60d6f124a78798b73944f5ba87c2306ae3460153/2.26/platform007/f259413
+GCC_BASE=/mnt/gvfs/third-party2/gcc/7331085db891a2ef4a88a48a751d834e8d68f4cb/7.x/centos7-native/b2ef2b6
+CLANG_BASE=/mnt/gvfs/third-party2/llvm-fb/963d9aeda70cc4779885b1277484fe7544a04e3e/9.0.0/platform007/9e92d53/
+LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/6ace84e956873d53638c738b6f65f3f469cca74c/7.x/platform007/5620abc
+GLIBC_BASE=/mnt/gvfs/third-party2/glibc/192b0f42d63dcf6210d6ceae387b49af049e6e0c/2.26/platform007/f259413
 SNAPPY_BASE=/mnt/gvfs/third-party2/snappy/7f9bdaada18f59bc27ec2b0871eb8a6144343aef/1.1.3/platform007/ca4da3d
-ZLIB_BASE=/mnt/gvfs/third-party2/zlib/22c2d65676fb7c23cfa797c4f6937f38b026f3cf/1.2.8/platform007/ca4da3d
+ZLIB_BASE=/mnt/gvfs/third-party2/zlib/2d9f0b9a4274cc21f61272a9e89bdb859bce8f1f/1.2.8/platform007/ca4da3d
 BZIP2_BASE=/mnt/gvfs/third-party2/bzip2/dc49a21c5fceec6456a7a28a94dcd16690af1337/1.0.6/platform007/ca4da3d
-LZ4_BASE=/mnt/gvfs/third-party2/lz4/907b498203d297947f3bb70b9466f47e100f1873/r131/platform007/ca4da3d
-ZSTD_BASE=/mnt/gvfs/third-party2/zstd/3ee276cbacfad3074e3f07bf826ac47f06970f4e/1.3.5/platform007/15a3614
+LZ4_BASE=/mnt/gvfs/third-party2/lz4/0f607f8fc442ea7d6b876931b1898bb573d5e5da/1.9.1/platform007/ca4da3d
+ZSTD_BASE=/mnt/gvfs/third-party2/zstd/ca22bc441a4eb709e9e0b1f9fec9750fed7b31c5/1.4.x/platform007/15a3614
 GFLAGS_BASE=/mnt/gvfs/third-party2/gflags/0b9929d2588991c65a57168bf88aff2db87c5d48/2.2.0/platform007/ca4da3d
-JEMALLOC_BASE=/mnt/gvfs/third-party2/jemalloc/9c910d36d6235cc40e8ff559358f1833452300ca/master/platform007/5b0f53e
-NUMA_BASE=/mnt/gvfs/third-party2/numa/9cbf2460284c669ed19c3ccb200a71f7dd7e53c7/2.0.11/platform007/ca4da3d
-LIBUNWIND_BASE=/mnt/gvfs/third-party2/libunwind/bf3d7497fe4e6d007354f0adffa16ce3003f8338/1.3/platform007/6f3e0a9
-TBB_BASE=/mnt/gvfs/third-party2/tbb/ff4e0b093534704d8abab678a4fd7f5ea7b094c7/2018_U5/platform007/ca4da3d
-KERNEL_HEADERS_BASE=/mnt/gvfs/third-party2/kernel-headers/b5c4a61a5c483ba24722005ae07895971a2ac707/fb/platform007/da39a3e
-BINUTILS_BASE=/mnt/gvfs/third-party2/binutils/92ff90349e2f43ea0a8246d8b1cf17b6869013e3/2.29.1/centos7-native/da39a3e
-VALGRIND_BASE=/mnt/gvfs/third-party2/valgrind/f3f697a28122e6bcd513273dd9c1ff23852fc59f/3.13.0/platform007/ca4da3d
+JEMALLOC_BASE=/mnt/gvfs/third-party2/jemalloc/c26f08f47ac35fc31da2633b7da92d6b863246eb/master/platform007/c26c002
+NUMA_BASE=/mnt/gvfs/third-party2/numa/3f3fb57a5ccc5fd21c66416c0b83e0aa76a05376/2.0.11/platform007/ca4da3d
+LIBUNWIND_BASE=/mnt/gvfs/third-party2/libunwind/40c73d874898b386a71847f1b99115d93822d11f/1.4/platform007/6f3e0a9
+TBB_BASE=/mnt/gvfs/third-party2/tbb/4ce8e8dba77cdbd81b75d6f0c32fd7a1b76a11ec/2018_U5/platform007/ca4da3d
+LIBURING_BASE=/mnt/gvfs/third-party2/liburing/79427253fd0d42677255aacfe6d13bfe63f752eb/20190828/platform007/ca4da3d
+KERNEL_HEADERS_BASE=/mnt/gvfs/third-party2/kernel-headers/fb251ecd2f5ae16f8671f7014c246e52a748fe0b/fb/platform007/da39a3e
+BINUTILS_BASE=/mnt/gvfs/third-party2/binutils/ab9f09bba370e7066cafd4eb59752db93f2e8312/2.29.1/platform007/15a3614
+VALGRIND_BASE=/mnt/gvfs/third-party2/valgrind/d42d152a15636529b0861ec493927200ebebca8e/3.15.0/platform007/ca4da3d
 LUA_BASE=/mnt/gvfs/third-party2/lua/f0cd714433206d5139df61659eb7b28b1dea6683/5.3.4/platform007/5007832
@@ -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
@@ -133,13 +133,16 @@ _TEST_NAME_TO_PARSERS = {
    'lite_test': [CompilerErrorParser, GTestErrorParser],
    'stress_crash': [CompilerErrorParser, DbCrashErrorParser],
    'stress_crash_with_atomic_flush': [CompilerErrorParser, DbCrashErrorParser],
+    'stress_crash_with_txn': [CompilerErrorParser, DbCrashErrorParser],
    'write_stress': [CompilerErrorParser, WriteStressErrorParser],
    'asan': [CompilerErrorParser, GTestErrorParser, AsanErrorParser],
    'asan_crash': [CompilerErrorParser, AsanErrorParser, DbCrashErrorParser],
    'asan_crash_with_atomic_flush': [CompilerErrorParser, AsanErrorParser, DbCrashErrorParser],
+    'asan_crash_with_txn': [CompilerErrorParser, AsanErrorParser, DbCrashErrorParser],
    'ubsan': [CompilerErrorParser, GTestErrorParser, UbsanErrorParser],
    'ubsan_crash': [CompilerErrorParser, UbsanErrorParser, DbCrashErrorParser],
    'ubsan_crash_with_atomic_flush': [CompilerErrorParser, UbsanErrorParser, DbCrashErrorParser],
+    'ubsan_crash_with_txn': [CompilerErrorParser, UbsanErrorParser, DbCrashErrorParser],
    'valgrind': [CompilerErrorParser, GTestErrorParser, ValgrindErrorParser],
    'tsan': [CompilerErrorParser, GTestErrorParser, TsanErrorParser],
    'format_compatible': [CompilerErrorParser, CompatErrorParser],
@@ -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"

@@ -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/"

@@ -86,6 +86,15 @@ else
 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 "$PORTABLE" || PORTABLE=1
@@ -94,7 +103,7 @@ export PORTABLE
 BINUTILS="$BINUTILS_BASE/bin"
 AR="$BINUTILS/ar"

-DEPS_INCLUDE="$SNAPPY_INCLUDE $ZLIB_INCLUDE $BZIP_INCLUDE $LZ4_INCLUDE $ZSTD_INCLUDE $GFLAGS_INCLUDE $NUMA_INCLUDE $TBB_INCLUDE"
+DEPS_INCLUDE="$SNAPPY_INCLUDE $ZLIB_INCLUDE $BZIP_INCLUDE $LZ4_INCLUDE $ZSTD_INCLUDE $GFLAGS_INCLUDE $NUMA_INCLUDE $TBB_INCLUDE $LIBURING_INCLUDE"

 STDLIBS="-L $GCC_BASE/lib64"

@@ -109,6 +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 $LIBGCC_INCLUDE"
@@ -119,6 +129,7 @@ else
  CLANG_INCLUDE="$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"

@@ -135,10 +146,10 @@ else
 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"
+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"
+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/platform007/lib/ld.so"
 EXEC_LDFLAGS+=" $LIBUNWIND"
@@ -148,7 +159,7 @@ 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 $TBB_LIBS"
+EXEC_LDFLAGS_SHARED="$SNAPPY_LIBS $ZLIB_LIBS $BZIP_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS $TBB_LIBS $LIBURING_LIBS"

 VALGRIND_VER="$VALGRIND_BASE/bin/"

@@ -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
@@ -2,36 +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 "Then, move both files (i.e. ${CLANG_FORMAT_DIFF} and clang-format) to some directory within PATH=${PATH}"
-  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 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
+    # Check argparse pre-req on interpreter, or it will fail
+    if echo import argparse | ${PYTHON:-python3}; then
+      true # Good
+    else
+      echo "To run clang-format-diff.py, we'll need the library "argparse" to be"
+      echo "installed. You can try either of the follow ways to install it:"
+      echo "  1. Manually download argparse: https://pypi.python.org/pypi/argparse"
+      echo "  2. easy_install argparse (if you have easy_install)"
+      echo "  3. pip install argparse (if you have pip)"
+      exit 129
+    fi
+    # Unfortunately, some machines have a Python2 clang-format-diff.py
+    # installed but only a Python3 interpreter installed. Rather than trying
+    # different Python versions that might be installed, we can try migrating
+    # the code to Python3 if it looks like Python2
+    if grep -q "print '" "$CFD_PATH" && \
+       ${PYTHON:-python3} --version | grep -q 'ython 3'; then
+      if [ ! -f "$REPO_ROOT/.py3/clang-format-diff.py" ]; then
+        echo "Migrating $CFD_PATH to Python3 in a hidden file"
+        mkdir -p "$REPO_ROOT/.py3"
+        ${PYTHON:-python3} -m lib2to3 -w -n -o "$REPO_ROOT/.py3" "$CFD_PATH" > /dev/null || exit 128
+      fi
+      CFD_PATH="$REPO_ROOT/.py3/clang-format-diff.py"
+    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
@@ -54,15 +117,25 @@ fi
 set -e

 uncommitted_code=`git diff HEAD`
-LAST_MASTER=`git merge-base master HEAD`

 # If there's no uncommitted changes, we assume user are doing post-commit
-# format check, in which case we'll check the modified lines since last commit
-# from master. Otherwise, we'll check format of the uncommitted code only.
+# format check, in which case we'll try to check the modified lines vs. the
+# facebook/rocksdb.git master branch. Otherwise, we'll check format of the
+# uncommitted code only.
 if [ -z "$uncommitted_code" ]
 then
-  # Check the format of last commit
-  diffs=$(git diff -U0 $LAST_MASTER^ | $CLANG_FORMAT_DIFF -p 1)
+  # Attempt to get name of facebook/rocksdb.git remote.
+  [ "$FORMAT_REMOTE" ] || FORMAT_REMOTE="$(git remote -v | grep 'facebook/rocksdb.git' | head -n 1 | cut -f 1)"
+  # Fall back on 'origin' if that fails
+  [ "$FORMAT_REMOTE" ] || FORMAT_REMOTE=origin
+  # Use master branch from that remote
+  [ "$FORMAT_UPSTREAM" ] || FORMAT_UPSTREAM="$FORMAT_REMOTE/master"
+  # Get the common ancestor with that remote branch. Everything after that
+  # common ancestor would be considered the contents of a pull request, so
+  # should be relevant for formatting fixes.
+  FORMAT_UPSTREAM_MERGE_BASE="$(git merge-base "$FORMAT_UPSTREAM" HEAD)"
+  # Get the differences
+  diffs=$(git diff -U0 "$FORMAT_UPSTREAM_MERGE_BASE" | $CLANG_FORMAT_DIFF -p 1)
 else
  # Check the format of uncommitted lines,
  diffs=$(git diff -U0 HEAD | $CLANG_FORMAT_DIFF -p 1)
@@ -72,16 +145,24 @@ if [ -z "$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
 COLOR_END="\033[0m"
-COLOR_RED="\033[0;31m" 
-COLOR_GREEN="\033[0;32m" 
+COLOR_RED="\033[0;31m"
+COLOR_GREEN="\033[0;32m"

 echo -e "Detect lines that doesn't follow the format rules:\r"
 # Add the color to the diff. lines added will be green; lines removed will be red.
-echo "$diffs" | 
+echo "$diffs" |
  sed -e "s/\(^-.*$\)/`echo -e \"$COLOR_RED\1$COLOR_END\"`/" |
  sed -e "s/\(^+.*$\)/`echo -e \"$COLOR_GREEN\1$COLOR_END\"`/"

@@ -104,9 +185,9 @@ fi
 # Do in-place format adjustment.
 if [ -z "$uncommitted_code" ]
 then
-  git diff -U0 $LAST_MASTER^ | $CLANG_FORMAT_DIFF -i -p 1
+  git diff -U0 "$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!"

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

 from __future__ import absolute_import
@@ -378,7 +378,7 @@ function send_to_ods {
    echo >&2 "ERROR: Key $key doesn't have a value."
    return
  fi
-  curl -s "https://www.intern.facebook.com/intern/agent/ods_set.php?entity=rocksdb_build$git_br&key=$key&value=$value" \
+  curl --silent "https://www.intern.facebook.com/intern/agent/ods_set.php?entity=rocksdb_build$git_br&key=$key&value=$value" \
    --connect-timeout 60
 }

@@ -414,7 +414,68 @@ STRESS_CRASH_TEST_COMMANDS="[
                'shell':'$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': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug stress tests',
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
+                'user':'root',
+                $PARSER
+            },
+            {
+                'name':'Build and run RocksDB debug blackbox crash tests',
+                'timeout': 86400,
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 blackbox_crash_test || $CONTRUN_NAME=blackbox_crash_test $TASK_CREATION_TOOL',
+                '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': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug stress tests',
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
+                'user':'root',
+                $PARSER
+            },
+            {
+                'name':'Build and run RocksDB debug whitebox crash tests',
+                'timeout': 86400,
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 whitebox_crash_test || $CONTRUN_NAME=whitebox_crash_test $TASK_CREATION_TOOL',
+                'user':'root',
+                $PARSER
+            },
+            $UPLOAD_DB_DIR,
        ],
        $REPORT
    }
@@ -450,6 +511,36 @@ STRESS_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
    }
 ]"

+#
+# RocksDB stress/crash test with txn
+#
+STRESS_CRASH_TEST_WITH_TXN_COMMANDS="[
+    {
+        'name':'Rocksdb Stress and Crash Test with txn',
+        'oncall':'$ONCALL',
+        'executeLocal': 'true',
+        'timeout': 86400,
+        'steps': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug stress tests',
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 db_stress || $CONTRUN_NAME=db_stress $TASK_CREATION_TOOL',
+                'user':'root',
+                $PARSER
+            },
+            {
+                'name':'Build and run RocksDB debug crash tests with txn',
+                'timeout': 86400,
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 crash_test_with_txn || $CONTRUN_NAME=crash_test_with_txn $TASK_CREATION_TOOL',
+                'user':'root',
+                $PARSER
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
 # RocksDB write stress test.
 # We run on disk device on purpose (i.e. no $SHM)
 # because we want to add some randomness to fsync commands
@@ -512,6 +603,55 @@ ASAN_CRASH_TEST_COMMANDS="[
                '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': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug blackbox asan_crash_test',
+                'timeout': 86400,
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH 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': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug whitebox asan_crash_test',
+                'timeout': 86400,
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 whitebox_asan_crash_test || $CONTRUN_NAME=whitebox_asan_crash_test $TASK_CREATION_TOOL',
+                'user':'root',
+                $PARSER
+            },
+            $UPLOAD_DB_DIR,
        ],
        $REPORT
    }
@@ -541,6 +681,30 @@ ASAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
    }
 ]"

+#
+# RocksDB crash testing with txn under address sanitizer
+#
+ASAN_CRASH_TEST_WITH_TXN_COMMANDS="[
+    {
+        'name':'Rocksdb crash test with txn under ASAN',
+        'oncall':'$ONCALL',
+        'executeLocal': 'true',
+        'timeout': 86400,
+        'steps': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug asan_crash_test_with_txn',
+                'timeout': 86400,
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 asan_crash_test_with_txn || $CONTRUN_NAME=asan_crash_test_with_txn $TASK_CREATION_TOOL',
+                'user':'root',
+                $PARSER
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
 #
 # RocksDB test under undefined behavior sanitizer
 #
@@ -553,7 +717,7 @@ UBSAN_TEST_COMMANDS="[
            $CLEANUP_ENV,
            {
                'name':'Test RocksDB debug under UBSAN',
-                'shell':'set -o pipefail && $SHM $UBSAN $DEBUG make $PARALLELISM ubsan_check || $CONTRUN_NAME=ubsan_check $TASK_CREATION_TOOL',
+                'shell':'set -o pipefail && $SHM $UBSAN $CLANG $DEBUG make $PARALLELISM ubsan_check || $CONTRUN_NAME=ubsan_check $TASK_CREATION_TOOL',
                'user':'root',
                $PARSER
            }
@@ -563,7 +727,7 @@ UBSAN_TEST_COMMANDS="[
 ]"

 #
-# RocksDB crash testing under udnefined behavior sanitizer
+# RocksDB crash testing under undefined behavior sanitizer
 #
 UBSAN_CRASH_TEST_COMMANDS="[
    {
@@ -576,10 +740,59 @@ UBSAN_CRASH_TEST_COMMANDS="[
            {
                'name':'Build and run RocksDB debug ubsan_crash_test',
                'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 ubsan_crash_test || $CONTRUN_NAME=ubsan_crash_test $TASK_CREATION_TOOL',
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH $CLANG 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': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug blackbox ubsan_crash_test',
+                'timeout': 86400,
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH $CLANG 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': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug whitebox ubsan_crash_test',
+                'timeout': 86400,
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH $CLANG make J=1 whitebox_ubsan_crash_test || $CONTRUN_NAME=whitebox_ubsan_crash_test $TASK_CREATION_TOOL',
+                'user':'root',
+                $PARSER
+            },
+            $UPLOAD_DB_DIR,
        ],
        $REPORT
    }
@@ -599,7 +812,31 @@ UBSAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
            {
                'name':'Build and run RocksDB debug ubsan_crash_test_with_atomic_flush',
                'timeout': 86400,
-                'shell':'$SHM $DEBUG $NON_TSAN_CRASH make J=1 ubsan_crash_test_with_atomic_flush || $CONTRUN_NAME=ubsan_crash_test_with_atomic_flush $TASK_CREATION_TOOL',
+                'shell':'$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',
+                'user':'root',
+                $PARSER
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
+#
+# RocksDB crash testing with txn under undefined behavior sanitizer
+#
+UBSAN_CRASH_TEST_WITH_TXN_COMMANDS="[
+    {
+        'name':'Rocksdb crash test with txn under UBSAN',
+        'oncall':'$ONCALL',
+        'executeLocal': 'true',
+        'timeout': 86400,
+        'steps': [
+            $CLEANUP_ENV,
+            {
+                'name':'Build and run RocksDB debug ubsan_crash_test_with_txn',
+                'timeout': 86400,
+                'shell':'$SHM $DEBUG $NON_TSAN_CRASH $CLANG make J=1 ubsan_crash_test_with_txn || $CONTRUN_NAME=ubsan_crash_test_with_txn $TASK_CREATION_TOOL',
                'user':'root',
                $PARSER
            },
@@ -673,6 +910,55 @@ TSAN_CRASH_TEST_COMMANDS="[
                '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':'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':'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
+            },
+            $UPLOAD_DB_DIR,
        ],
        $REPORT
    }
@@ -702,6 +988,30 @@ TSAN_CRASH_TEST_WITH_ATOMIC_FLUSH_COMMANDS="[
    }
 ]"

+#
+# RocksDB crash test with txn under TSAN
+#
+TSAN_CRASH_TEST_WITH_TXN_COMMANDS="[
+    {
+        'name':'Rocksdb Crash Test with txn under TSAN',
+        'oncall':'$ONCALL',
+        'executeLocal': 'true',
+        'timeout': 86400,
+        'steps': [
+            $CLEANUP_ENV,
+            {
+                'name':'Compile and run',
+                'timeout': 86400,
+                'shell':'set -o pipefail && $SHM $DEBUG $TSAN $TSAN_CRASH CRASH_TEST_KILL_ODD=1887 make J=1 crash_test_with_txn || $CONTRUN_NAME=tsan_crash_test_with_txn $TASK_CREATION_TOOL',
+                'user':'root',
+                $PARSER
+            },
+            $UPLOAD_DB_DIR,
+        ],
+        $REPORT
+    }
+]"
+
 #
 # RocksDB format compatible
 #
@@ -712,6 +1022,8 @@ run_format_compatible()
  rm -rf /dev/shm/rocksdb
  mkdir /dev/shm/rocksdb

+  export https_proxy="fwdproxy:8080"
+  
  tools/check_format_compatible.sh
 }

@@ -778,7 +1090,7 @@ run_regression()

  # parameters: $1 -- key, $2 -- value
  function send_size_to_ods {
-    curl -s "https://www.intern.facebook.com/intern/agent/ods_set.php?entity=rocksdb_build&key=rocksdb.build_size.$1&value=$2" \
+    curl --silent "https://www.intern.facebook.com/intern/agent/ods_set.php?entity=rocksdb_build&key=rocksdb.build_size.$1&value=$2" \
      --connect-timeout 60
  }

@@ -789,6 +1101,7 @@ 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`
@@ -801,6 +1114,7 @@ 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`
@@ -886,9 +1200,18 @@ 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
+    ;;
  write_stress)
    echo $WRITE_STRESS_COMMANDS
    ;;
@@ -898,18 +1221,36 @@ 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
    ;;
+  asan_crash_with_txn)
+    echo $ASAN_CRASH_TEST_WITH_TXN_COMMANDS
+    ;;
  ubsan)
    echo $UBSAN_TEST_COMMANDS
    ;;
  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
    ;;
+  ubsan_crash_with_txn)
+    echo $UBSAN_CRASH_TEST_WITH_TXN_COMMANDS
+    ;;
  valgrind)
    echo $VALGRIND_TEST_COMMANDS
    ;;
@@ -919,9 +1260,18 @@ 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
    ;;
+  tsan_crash_with_txn)
+    echo $TSAN_CRASH_TEST_WITH_TXN_COMMANDS
+    ;;
  format_compatible)
    echo $FORMAT_COMPATIBLE_COMMANDS
    ;;
@@ -945,5 +1295,6 @@ case $1 in
    ;;
  *)
    echo "Invalid determinator command"
+    exit 1
    ;;
 esac
@@ -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
+
@@ -6,6 +6,12 @@
 BASEDIR=$(dirname $0)
 OUTPUT=""

+function log_header()
+{
+  echo "# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved." >> "$OUTPUT"
+}
+
+
 function log_variable()
 {
  echo "$1=${!1}" >> "$OUTPUT"
@@ -69,6 +75,7 @@ echo "Writing dependencies to $OUTPUT"
 GCC_BASE=`readlink -f $TP2_LATEST/gcc/7.x/centos7-native/*/`
 CLANG_BASE=`readlink -f $TP2_LATEST/llvm-fb/stable/centos7-native/*/`

+log_header
 log_variable GCC_BASE
 log_variable CLANG_BASE

@@ -85,6 +92,7 @@ get_lib_base jemalloc   LATEST  platform007
 get_lib_base numa       LATEST  platform007
 get_lib_base libunwind  LATEST  platform007
 get_lib_base tbb        LATEST  platform007
+get_lib_base liburing   LATEST  platform007

 get_lib_base kernel-headers fb platform007
 get_lib_base binutils   LATEST centos7-native
@@ -108,6 +116,7 @@ echo "Writing dependencies to $OUTPUT"
 GCC_BASE=`readlink -f $TP2_LATEST/gcc/5.x/centos7-native/*/`
 CLANG_BASE=`readlink -f $TP2_LATEST/llvm-fb/stable/centos7-native/*/`

+log_header
 log_variable GCC_BASE
 log_variable CLANG_BASE

@@ -147,6 +156,7 @@ echo "Writing 4.8.1 dependencies to $OUTPUT"
 GCC_BASE=`readlink -f $TP2_LATEST/gcc/4.8.1/centos6-native/*/`
 CLANG_BASE=`readlink -f $TP2_LATEST/llvm-fb/stable/centos6-native/*/`

+log_header
 log_variable GCC_BASE
 log_variable CLANG_BASE

@@ -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
@@ -11,59 +11,66 @@ int main() {
 }
 #else

-#include <cinttypes>
-#include <sys/types.h>
 #include <stdio.h>
+#include <sys/types.h>
+#include <cinttypes>
+#include <limits>

 #include "port/port.h"
 #include "rocksdb/cache.h"
 #include "rocksdb/db.h"
 #include "rocksdb/env.h"
+#include "util/coding.h"
 #include "util/gflags_compat.h"
+#include "util/hash.h"
 #include "util/mutexlock.h"
 #include "util/random.h"

 using GFLAGS_NAMESPACE::ParseCommandLineFlags;

-static 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, "");

-namespace rocksdb {
+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 +94,9 @@ 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 +110,6 @@ 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 +120,69 @@ 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 +192,31 @@ 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::Env* env = rocksdb::Env::Default();
+    ROCKSDB_NAMESPACE::Env* env = ROCKSDB_NAMESPACE::Env::Default();

    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());
@@ -192,10 +247,15 @@ 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;

    {
@@ -220,44 +280,82 @@ 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");
  }
 };
-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE

 int main(int argc, char** argv) {
  ParseCommandLineFlags(&argc, &argv, true);
@@ -267,9 +365,11 @@ int main(int argc, char** argv) {
    exit(1);
  }

-  rocksdb::CacheBench bench;
+  ROCKSDB_NAMESPACE::CacheBench bench;
  if (FLAGS_populate_cache) {
    bench.PopulateCache();
+    printf("Population complete\n");
+    printf("----------------------------\n");
  }
  if (bench.Run()) {
    return 0;
@@ -20,7 +20,7 @@
 #include "util/coding.h"
 #include "util/string_util.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 // Conversions between numeric keys/values and the types expected by Cache.
 static std::string EncodeKey(int k) {
@@ -86,14 +86,22 @@ class CacheTest : public testing::TestWithParam<std::string> {
    return nullptr;
  }

-  std::shared_ptr<Cache> NewCache(size_t capacity, int num_shard_bits,
-                                  bool strict_capacity_limit) {
+  std::shared_ptr<Cache> NewCache(
+      size_t capacity, int num_shard_bits, bool strict_capacity_limit,
+      CacheMetadataChargePolicy charge_policy = kDontChargeCacheMetadata) {
    auto type = GetParam();
    if (type == kLRU) {
-      return NewLRUCache(capacity, num_shard_bits, strict_capacity_limit, 0.0);
+      LRUCacheOptions co;
+      co.capacity = capacity;
+      co.num_shard_bits = num_shard_bits;
+      co.strict_capacity_limit = strict_capacity_limit;
+      co.high_pri_pool_ratio = 0;
+      co.metadata_charge_policy = charge_policy;
+      return NewLRUCache(co);
    }
    if (type == kClock) {
-      return NewClockCache(capacity, num_shard_bits, strict_capacity_limit);
+      return NewClockCache(capacity, num_shard_bits, strict_capacity_limit,
+                           charge_policy);
    }
    return nullptr;
  }
@@ -109,8 +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) {
@@ -143,10 +151,15 @@ class CacheTest : public testing::TestWithParam<std::string> {
 };
 CacheTest* CacheTest::current_;

+class LRUCacheTest : public CacheTest {};
+
 TEST_P(CacheTest, UsageTest) {
  // cache is std::shared_ptr and will be automatically cleaned up.
  const uint64_t kCapacity = 100000;
-  auto cache = NewCache(kCapacity, 8, false);
+  auto cache = NewCache(kCapacity, 8, false, kDontChargeCacheMetadata);
+  auto precise_cache = NewCache(kCapacity, 0, false, kFullChargeCacheMetadata);
+  ASSERT_EQ(0, cache->GetUsage());
+  ASSERT_EQ(0, precise_cache->GetUsage());

  size_t usage = 0;
  char value[10] = "abcdef";
@@ -154,32 +167,47 @@ 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);
+    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());
  }

+  cache->EraseUnRefEntries();
+  precise_cache->EraseUnRefEntries();
+  ASSERT_EQ(0, cache->GetUsage());
+  ASSERT_EQ(0, precise_cache->GetUsage());
+
  // 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);
+    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
  ASSERT_GT(kCapacity, cache->GetUsage());
+  ASSERT_GT(kCapacity, precise_cache->GetUsage());
  ASSERT_LT(kCapacity * 0.95, cache->GetUsage());
+  ASSERT_LT(kCapacity * 0.95, precise_cache->GetUsage());
 }

 TEST_P(CacheTest, PinnedUsageTest) {
  // cache is std::shared_ptr and will be automatically cleaned up.
-  const uint64_t kCapacity = 100000;
-  auto cache = NewCache(kCapacity, 8, false);
+  const uint64_t kCapacity = 200000;
+  auto cache = NewCache(kCapacity, 8, false, kDontChargeCacheMetadata);
+  auto precise_cache = NewCache(kCapacity, 8, false, kFullChargeCacheMetadata);

  size_t pinned_usage = 0;
  char value[10] = "abcdef";

  std::forward_list<Cache::Handle*> unreleased_handles;
+  std::forward_list<Cache::Handle*> unreleased_handles_in_precise_cache;

  // Add entries. Unpin some of them after insertion. Then, pin some of them
  // again. Check GetPinnedUsage().
@@ -187,40 +215,73 @@ TEST_P(CacheTest, PinnedUsageTest) {
    std::string key(i, 'a');
    auto kv_size = key.size() + 5;
    Cache::Handle* handle;
-    cache->Insert(key, reinterpret_cast<void*>(value), kv_size, dumbDeleter,
-                  &handle);
+    Cache::Handle* handle_in_precise_cache;
+    ASSERT_OK(cache->Insert(key, reinterpret_cast<void*>(value), kv_size,
+                            dumbDeleter, &handle));
+    assert(handle);
+    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());
+    ASSERT_LT(pinned_usage, precise_cache->GetPinnedUsage());
    if (i % 2 == 0) {
      cache->Release(handle);
+      precise_cache->Release(handle_in_precise_cache);
      pinned_usage -= kv_size;
      ASSERT_EQ(pinned_usage, cache->GetPinnedUsage());
+      ASSERT_LT(pinned_usage, precise_cache->GetPinnedUsage());
    } else {
      unreleased_handles.push_front(handle);
+      unreleased_handles_in_precise_cache.push_front(handle_in_precise_cache);
    }
    if (i % 3 == 0) {
      unreleased_handles.push_front(cache->Lookup(key));
+      auto x = precise_cache->Lookup(key);
+      assert(x);
+      unreleased_handles_in_precise_cache.push_front(x);
      // If i % 2 == 0, then the entry was unpinned before Lookup, so pinned
      // usage increased
      if (i % 2 == 0) {
        pinned_usage += kv_size;
      }
      ASSERT_EQ(pinned_usage, cache->GetPinnedUsage());
+      ASSERT_LT(pinned_usage, precise_cache->GetPinnedUsage());
    }
  }
+  auto precise_cache_pinned_usage = precise_cache->GetPinnedUsage();
+  ASSERT_LT(pinned_usage, precise_cache_pinned_usage);

  // 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);
+    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->EraseUnRefEntries();
+  precise_cache->EraseUnRefEntries();
+  ASSERT_EQ(pinned_usage, cache->GetPinnedUsage());
+  ASSERT_EQ(precise_cache_pinned_usage, precise_cache->GetPinnedUsage());

  // release handles for pinned entries to prevent memory leaks
  for (auto handle : unreleased_handles) {
    cache->Release(handle);
  }
+  for (auto handle : unreleased_handles_in_precise_cache) {
+    precise_cache->Release(handle);
+  }
+  ASSERT_EQ(0, cache->GetPinnedUsage());
+  ASSERT_EQ(0, precise_cache->GetPinnedUsage());
+  cache->EraseUnRefEntries();
+  precise_cache->EraseUnRefEntries();
+  ASSERT_EQ(0, cache->GetUsage());
+  ASSERT_EQ(0, precise_cache->GetUsage());
 }

 TEST_P(CacheTest, HitAndMiss) {
@@ -306,7 +367,7 @@ TEST_P(CacheTest, EvictionPolicy) {
  Insert(200, 201);

  // Frequently used entry must be kept around
-  for (int i = 0; i < kCacheSize + 200; i++) {
+  for (int i = 0; i < kCacheSize * 2; i++) {
    Insert(1000+i, 2000+i);
    ASSERT_EQ(101, Lookup(100));
  }
@@ -359,7 +420,7 @@ TEST_P(CacheTest, EvictionPolicyRef) {
  Insert(303, 104);

  // Insert entries much more than Cache capacity
-  for (int i = 0; i < kCacheSize + 200; i++) {
+  for (int i = 0; i < kCacheSize * 2; i++) {
    Insert(1000 + i, 2000 + i);
  }

@@ -550,10 +611,10 @@ TEST_P(CacheTest, SetCapacity) {
  }
 }

-TEST_P(CacheTest, SetStrictCapacityLimit) {
+TEST_P(LRUCacheTest, SetStrictCapacityLimit) {
  // test1: set the flag to false. Insert more keys than capacity. See if they
  // all go through.
-  std::shared_ptr<Cache> cache = NewLRUCache(5, 0, false);
+  std::shared_ptr<Cache> cache = NewCache(5, 0, false);
  std::vector<Cache::Handle*> handles(10);
  Status s;
  for (size_t i = 0; i < 10; i++) {
@@ -579,7 +640,7 @@ TEST_P(CacheTest, SetStrictCapacityLimit) {
  }

  // test3: init with flag being true.
-  std::shared_ptr<Cache> cache2 = NewLRUCache(5, 0, true);
+  std::shared_ptr<Cache> cache2 = NewCache(5, 0, true);
  for (size_t i = 0; i < 5; i++) {
    std::string key = ToString(i + 1);
    s = cache2->Insert(key, new Value(i + 1), 1, &deleter, &handles[i]);
@@ -697,15 +758,16 @@ TEST_P(CacheTest, GetCharge) {
 }

 #ifdef SUPPORT_CLOCK_CACHE
-std::shared_ptr<Cache> (*new_clock_cache_func)(size_t, int,
-                                               bool) = NewClockCache;
+std::shared_ptr<Cache> (*new_clock_cache_func)(
+    size_t, int, bool, CacheMetadataChargePolicy) = NewClockCache;
 INSTANTIATE_TEST_CASE_P(CacheTestInstance, CacheTest,
                        testing::Values(kLRU, kClock));
 #else
 INSTANTIATE_TEST_CASE_P(CacheTestInstance, CacheTest, testing::Values(kLRU));
 #endif  // SUPPORT_CLOCK_CACHE
+INSTANTIATE_TEST_CASE_P(CacheTestInstance, LRUCacheTest, testing::Values(kLRU));

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE

 int main(int argc, char** argv) {
  ::testing::InitGoogleTest(&argc, argv);
@@ -11,15 +11,16 @@

 #ifndef SUPPORT_CLOCK_CACHE

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

-std::shared_ptr<Cache> NewClockCache(size_t /*capacity*/, int /*num_shard_bits*/,
-                                     bool /*strict_capacity_limit*/) {
+std::shared_ptr<Cache> NewClockCache(
+    size_t /*capacity*/, int /*num_shard_bits*/, bool /*strict_capacity_limit*/,
+    CacheMetadataChargePolicy /*metadata_charge_policy*/) {
  // Clock cache not supported.
  return nullptr;
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE

 #else

@@ -35,11 +36,12 @@ std::shared_ptr<Cache> NewClockCache(size_t /*capacity*/, int /*num_shard_bits*/
 #include "tbb/concurrent_hash_map.h"

 #include "cache/sharded_cache.h"
+#include "port/malloc.h"
 #include "port/port.h"
 #include "util/autovector.h"
 #include "util/mutexlock.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 namespace {

@@ -180,7 +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
@@ -202,6 +204,27 @@ struct CacheHandle {
    deleter = a.deleter;
    return *this;
  }
+
+  inline static size_t CalcTotalCharge(
+      Slice key, size_t charge,
+      CacheMetadataChargePolicy metadata_charge_policy) {
+    size_t meta_charge = 0;
+    if (metadata_charge_policy == kFullChargeCacheMetadata) {
+      meta_charge += sizeof(CacheHandle);
+#ifdef ROCKSDB_MALLOC_USABLE_SIZE
+      meta_charge +=
+          malloc_usable_size(static_cast<void*>(const_cast<char*>(key.data())));
+#else
+      meta_charge += key.size();
+#endif
+    }
+    return charge + meta_charge;
+  }
+
+  inline size_t CalcTotalCharge(
+      CacheMetadataChargePolicy metadata_charge_policy) {
+    return CalcTotalCharge(key, charge, metadata_charge_policy);
+  }
 };

 // Key of hash map. We store hash value with the key for convenience.
@@ -318,7 +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.
@@ -404,11 +428,12 @@ void ClockCacheShard::RecycleHandle(CacheHandle* handle,
  assert(!InCache(handle->flags) && CountRefs(handle->flags) == 0);
  context->to_delete_key.push_back(handle->key.data());
  context->to_delete_value.emplace_back(*handle);
+  size_t total_charge = handle->CalcTotalCharge(metadata_charge_policy_);
  handle->key.clear();
  handle->value = nullptr;
  handle->deleter = nullptr;
  recycle_.push_back(handle);
-  usage_.fetch_sub(handle->charge, std::memory_order_relaxed);
+  usage_.fetch_sub(total_charge, std::memory_order_relaxed);
 }

 void ClockCacheShard::Cleanup(const CleanupContext& context) {
@@ -434,7 +459,8 @@ bool ClockCacheShard::Ref(Cache::Handle* h) {
                                            std::memory_order_relaxed)) {
      if (CountRefs(flags) == 0) {
        // No reference count before the operation.
-        pinned_usage_.fetch_add(handle->charge, std::memory_order_relaxed);
+        size_t total_charge = handle->CalcTotalCharge(metadata_charge_policy_);
+        pinned_usage_.fetch_add(total_charge, std::memory_order_relaxed);
      }
      return true;
    }
@@ -454,7 +480,8 @@ bool ClockCacheShard::Unref(CacheHandle* handle, bool set_usage,
  assert(CountRefs(flags) > 0);
  if (CountRefs(flags) == 1) {
    // this is the last reference.
-    pinned_usage_.fetch_sub(handle->charge, std::memory_order_relaxed);
+    size_t total_charge = handle->CalcTotalCharge(metadata_charge_policy_);
+    pinned_usage_.fetch_sub(total_charge, std::memory_order_relaxed);
    // Cleanup if it is the last reference.
    if (!InCache(flags)) {
      MutexLock l(&mutex_);
@@ -538,9 +565,12 @@ 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_);
-  bool success = EvictFromCache(charge, context);
+  bool success = EvictFromCache(total_charge, context);
  bool strict = strict_capacity_limit_.load(std::memory_order_relaxed);
  if (!success && (strict || !hold_reference)) {
    context->to_delete_key.push_back(key.data());
@@ -569,15 +599,16 @@ 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);
  }
  table_.insert(HashTable::value_type(CacheKey(key, hash), handle));
  if (hold_reference) {
-    pinned_usage_.fetch_add(charge, std::memory_order_relaxed);
+    pinned_usage_.fetch_add(total_charge, std::memory_order_relaxed);
  }
-  usage_.fetch_add(charge, std::memory_order_relaxed);
+  usage_.fetch_add(total_charge, std::memory_order_relaxed);
  return handle;
 }

@@ -591,8 +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) {
@@ -601,6 +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;
 }
@@ -674,10 +710,14 @@ void ClockCacheShard::EraseUnRefEntries() {

 class ClockCache final : public ShardedCache {
 public:
-  ClockCache(size_t capacity, int num_shard_bits, bool strict_capacity_limit)
+  ClockCache(size_t capacity, int num_shard_bits, bool strict_capacity_limit,
+             CacheMetadataChargePolicy metadata_charge_policy)
      : ShardedCache(capacity, num_shard_bits, strict_capacity_limit) {
    int num_shards = 1 << num_shard_bits;
    shards_ = new ClockCacheShard[num_shards];
+    for (int i = 0; i < num_shards; i++) {
+      shards_[i].set_metadata_charge_policy(metadata_charge_policy);
+    }
    SetCapacity(capacity);
    SetStrictCapacityLimit(strict_capacity_limit);
  }
@@ -714,15 +754,16 @@ class ClockCache final : public ShardedCache {

 }  // end anonymous namespace

-std::shared_ptr<Cache> NewClockCache(size_t capacity, int num_shard_bits,
-                                     bool strict_capacity_limit) {
+std::shared_ptr<Cache> NewClockCache(
+    size_t capacity, int num_shard_bits, bool strict_capacity_limit,
+    CacheMetadataChargePolicy metadata_charge_policy) {
  if (num_shard_bits < 0) {
    num_shard_bits = GetDefaultCacheShardBits(capacity);
  }
-  return std::make_shared<ClockCache>(capacity, num_shard_bits,
-                                      strict_capacity_limit);
+  return std::make_shared<ClockCache>(
+      capacity, num_shard_bits, strict_capacity_limit, metadata_charge_policy);
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE

 #endif  // SUPPORT_CLOCK_CACHE
@@ -16,7 +16,7 @@

 #include "util/mutexlock.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 LRUHandleTable::LRUHandleTable() : list_(nullptr), length_(0), elems_(0) {
  Resize();
@@ -97,7 +97,8 @@ void LRUHandleTable::Resize() {

 LRUCacheShard::LRUCacheShard(size_t capacity, bool strict_capacity_limit,
                             double high_pri_pool_ratio,
-                             bool use_adaptive_mutex)
+                             bool use_adaptive_mutex,
+                             CacheMetadataChargePolicy metadata_charge_policy)
    : capacity_(0),
      high_pri_pool_usage_(0),
      strict_capacity_limit_(strict_capacity_limit),
@@ -106,6 +107,7 @@ LRUCacheShard::LRUCacheShard(size_t capacity, bool strict_capacity_limit,
      usage_(0),
      lru_usage_(0),
      mutex_(use_adaptive_mutex) {
+  set_metadata_charge_policy(metadata_charge_policy);
  // Make empty circular linked list
  lru_.next = &lru_;
  lru_.prev = &lru_;
@@ -124,7 +126,9 @@ void LRUCacheShard::EraseUnRefEntries() {
      LRU_Remove(old);
      table_.Remove(old->key(), old->hash);
      old->SetInCache(false);
-      usage_ -= old->charge;
+      size_t total_charge = old->CalcTotalCharge(metadata_charge_policy_);
+      assert(usage_ >= total_charge);
+      usage_ -= total_charge;
      last_reference_list.push_back(old);
    }
  }
@@ -180,16 +184,19 @@ void LRUCacheShard::LRU_Remove(LRUHandle* e) {
  e->next->prev = e->prev;
  e->prev->next = e->next;
  e->prev = e->next = nullptr;
-  lru_usage_ -= e->charge;
+  size_t total_charge = e->CalcTotalCharge(metadata_charge_policy_);
+  assert(lru_usage_ >= total_charge);
+  lru_usage_ -= total_charge;
  if (e->InHighPriPool()) {
-    assert(high_pri_pool_usage_ >= e->charge);
-    high_pri_pool_usage_ -= e->charge;
+    assert(high_pri_pool_usage_ >= total_charge);
+    high_pri_pool_usage_ -= total_charge;
  }
 }

 void LRUCacheShard::LRU_Insert(LRUHandle* e) {
  assert(e->next == nullptr);
  assert(e->prev == nullptr);
+  size_t total_charge = e->CalcTotalCharge(metadata_charge_policy_);
  if (high_pri_pool_ratio_ > 0 && (e->IsHighPri() || e->HasHit())) {
    // Inset "e" to head of LRU list.
    e->next = &lru_;
@@ -197,7 +204,7 @@ void LRUCacheShard::LRU_Insert(LRUHandle* e) {
    e->prev->next = e;
    e->next->prev = e;
    e->SetInHighPriPool(true);
-    high_pri_pool_usage_ += e->charge;
+    high_pri_pool_usage_ += total_charge;
    MaintainPoolSize();
  } else {
    // Insert "e" to the head of low-pri pool. Note that when
@@ -209,7 +216,7 @@ void LRUCacheShard::LRU_Insert(LRUHandle* e) {
    e->SetInHighPriPool(false);
    lru_low_pri_ = e;
  }
-  lru_usage_ += e->charge;
+  lru_usage_ += total_charge;
 }

 void LRUCacheShard::MaintainPoolSize() {
@@ -218,7 +225,10 @@ void LRUCacheShard::MaintainPoolSize() {
    lru_low_pri_ = lru_low_pri_->next;
    assert(lru_low_pri_ != &lru_);
    lru_low_pri_->SetInHighPriPool(false);
-    high_pri_pool_usage_ -= lru_low_pri_->charge;
+    size_t total_charge =
+        lru_low_pri_->CalcTotalCharge(metadata_charge_policy_);
+    assert(high_pri_pool_usage_ >= total_charge);
+    high_pri_pool_usage_ -= total_charge;
  }
 }

@@ -231,7 +241,9 @@ void LRUCacheShard::EvictFromLRU(size_t charge,
    LRU_Remove(old);
    table_.Remove(old->key(), old->hash);
    old->SetInCache(false);
-    usage_ -= old->charge;
+    size_t old_total_charge = old->CalcTotalCharge(metadata_charge_policy_);
+    assert(usage_ >= old_total_charge);
+    usage_ -= old_total_charge;
    deleted->push_back(old);
  }
 }
@@ -311,7 +323,9 @@ bool LRUCacheShard::Release(Cache::Handle* handle, bool force_erase) {
      }
    }
    if (last_reference) {
-      usage_ -= e->charge;
+      size_t total_charge = e->CalcTotalCharge(metadata_charge_policy_);
+      assert(usage_ >= total_charge);
+      usage_ -= total_charge;
    }
  }

@@ -345,15 +359,16 @@ Status LRUCacheShard::Insert(const Slice& key, uint32_t hash, void* value,
  e->SetInCache(true);
  e->SetPriority(priority);
  memcpy(e->key_data, key.data(), key.size());
+  size_t total_charge = e->CalcTotalCharge(metadata_charge_policy_);

  {
    MutexLock l(&mutex_);

    // Free the space following strict LRU policy until enough space
    // is freed or the lru list is empty
-    EvictFromLRU(charge, &last_reference_list);
+    EvictFromLRU(total_charge, &last_reference_list);

-    if ((usage_ + charge) > capacity_ &&
+    if ((usage_ + total_charge) > capacity_ &&
        (strict_capacity_limit_ || handle == nullptr)) {
      if (handle == nullptr) {
        // Don't insert the entry but still return ok, as if the entry inserted
@@ -369,14 +384,18 @@ Status LRUCacheShard::Insert(const Slice& key, uint32_t hash, void* value,
      // Insert into the cache. Note that the cache might get larger than its
      // capacity if not enough space was freed up.
      LRUHandle* old = table_.Insert(e);
-      usage_ += e->charge;
+      usage_ += total_charge;
      if (old != nullptr) {
+        s = Status::OkOverwritten();
        assert(old->InCache());
        old->SetInCache(false);
        if (!old->HasRefs()) {
          // old is on LRU because it's in cache and its reference count is 0
          LRU_Remove(old);
-          usage_ -= old->charge;
+          size_t old_total_charge =
+              old->CalcTotalCharge(metadata_charge_policy_);
+          assert(usage_ >= old_total_charge);
+          usage_ -= old_total_charge;
          last_reference_list.push_back(old);
        }
      }
@@ -409,7 +428,9 @@ void LRUCacheShard::Erase(const Slice& key, uint32_t hash) {
      if (!e->HasRefs()) {
        // The entry is in LRU since it's in hash and has no external references
        LRU_Remove(e);
-        usage_ -= e->charge;
+        size_t total_charge = e->CalcTotalCharge(metadata_charge_policy_);
+        assert(usage_ >= total_charge);
+        usage_ -= total_charge;
        last_reference = true;
      }
    }
@@ -447,7 +468,8 @@ std::string LRUCacheShard::GetPrintableOptions() const {
 LRUCache::LRUCache(size_t capacity, int num_shard_bits,
                   bool strict_capacity_limit, double high_pri_pool_ratio,
                   std::shared_ptr<MemoryAllocator> allocator,
-                   bool use_adaptive_mutex)
+                   bool use_adaptive_mutex,
+                   CacheMetadataChargePolicy metadata_charge_policy)
    : ShardedCache(capacity, num_shard_bits, strict_capacity_limit,
                   std::move(allocator)) {
  num_shards_ = 1 << num_shard_bits;
@@ -457,7 +479,7 @@ LRUCache::LRUCache(size_t capacity, int num_shard_bits,
  for (int i = 0; i < num_shards_; i++) {
    new (&shards_[i])
        LRUCacheShard(per_shard, strict_capacity_limit, high_pri_pool_ratio,
-            use_adaptive_mutex);
+                      use_adaptive_mutex, metadata_charge_policy);
  }
 }

@@ -526,15 +548,15 @@ std::shared_ptr<Cache> NewLRUCache(const LRUCacheOptions& cache_opts) {
  return NewLRUCache(cache_opts.capacity, cache_opts.num_shard_bits,
                     cache_opts.strict_capacity_limit,
                     cache_opts.high_pri_pool_ratio,
-                     cache_opts.memory_allocator,
-                     cache_opts.use_adaptive_mutex);
+                     cache_opts.memory_allocator, cache_opts.use_adaptive_mutex,
+                     cache_opts.metadata_charge_policy);
 }

 std::shared_ptr<Cache> NewLRUCache(
    size_t capacity, int num_shard_bits, bool strict_capacity_limit,
    double high_pri_pool_ratio,
-    std::shared_ptr<MemoryAllocator> memory_allocator,
-    bool use_adaptive_mutex) {
+    std::shared_ptr<MemoryAllocator> memory_allocator, bool use_adaptive_mutex,
+    CacheMetadataChargePolicy metadata_charge_policy) {
  if (num_shard_bits >= 20) {
    return nullptr;  // the cache cannot be sharded into too many fine pieces
  }
@@ -545,10 +567,9 @@ std::shared_ptr<Cache> NewLRUCache(
  if (num_shard_bits < 0) {
    num_shard_bits = GetDefaultCacheShardBits(capacity);
  }
-  return std::make_shared<LRUCache>(capacity, num_shard_bits,
-                                    strict_capacity_limit, high_pri_pool_ratio,
-                                    std::move(memory_allocator),
-                                    use_adaptive_mutex);
+  return std::make_shared<LRUCache>(
+      capacity, num_shard_bits, strict_capacity_limit, high_pri_pool_ratio,
+      std::move(memory_allocator), use_adaptive_mutex, metadata_charge_policy);
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -12,10 +12,11 @@

 #include "cache/sharded_cache.h"

+#include "port/malloc.h"
 #include "port/port.h"
 #include "util/autovector.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 // LRU cache implementation. This class is not thread-safe.

@@ -128,6 +129,21 @@ struct LRUHandle {
    }
    delete[] reinterpret_cast<char*>(this);
  }
+
+  // Caclculate the memory usage by metadata
+  inline size_t CalcTotalCharge(
+      CacheMetadataChargePolicy metadata_charge_policy) {
+    size_t meta_charge = 0;
+    if (metadata_charge_policy == kFullChargeCacheMetadata) {
+#ifdef ROCKSDB_MALLOC_USABLE_SIZE
+      meta_charge += malloc_usable_size(static_cast<void*>(this));
+#else
+      // This is the size that is used when a new handle is created
+      meta_charge += sizeof(LRUHandle) - 1 + key_length;
+#endif
+    }
+    return charge + meta_charge;
+  }
 };

 // We provide our own simple hash table since it removes a whole bunch
@@ -176,7 +192,8 @@ class LRUHandleTable {
 class ALIGN_AS(CACHE_LINE_SIZE) LRUCacheShard final : public CacheShard {
 public:
  LRUCacheShard(size_t capacity, bool strict_capacity_limit,
-                double high_pri_pool_ratio, bool use_adaptive_mutex);
+                double high_pri_pool_ratio, bool use_adaptive_mutex,
+                CacheMetadataChargePolicy metadata_charge_policy);
  virtual ~LRUCacheShard() override = default;

  // Separate from constructor so caller can easily make an array of LRUCache
@@ -297,7 +314,9 @@ class LRUCache
  LRUCache(size_t capacity, int num_shard_bits, bool strict_capacity_limit,
           double high_pri_pool_ratio,
           std::shared_ptr<MemoryAllocator> memory_allocator = nullptr,
-           bool use_adaptive_mutex = kDefaultToAdaptiveMutex);
+           bool use_adaptive_mutex = kDefaultToAdaptiveMutex,
+           CacheMetadataChargePolicy metadata_charge_policy =
+               kDontChargeCacheMetadata);
  virtual ~LRUCache();
  virtual const char* Name() const override { return "LRUCache"; }
  virtual CacheShard* GetShard(int shard) override;
@@ -317,4 +336,4 @@ class LRUCache
  int num_shards_ = 0;
 };

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -10,7 +10,7 @@
 #include "port/port.h"
 #include "test_util/testharness.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 class LRUCacheTest : public testing::Test {
 public:
@@ -31,13 +31,15 @@ class LRUCacheTest : public testing::Test {
    cache_ = reinterpret_cast<LRUCacheShard*>(
        port::cacheline_aligned_alloc(sizeof(LRUCacheShard)));
    new (cache_) LRUCacheShard(capacity, false /*strict_capcity_limit*/,
-                               high_pri_pool_ratio, use_adaptive_mutex);
+                               high_pri_pool_ratio, 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) {
@@ -189,7 +191,7 @@ TEST_F(LRUCacheTest, EntriesWithPriority) {
  ValidateLRUList({"e", "f", "g", "Z", "d"}, 2);
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE

 int main(int argc, char** argv) {
  ::testing::InitGoogleTest(&argc, argv);
@@ -13,7 +13,7 @@

 #include "util/mutexlock.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 ShardedCache::ShardedCache(size_t capacity, int num_shard_bits,
                           bool strict_capacity_limit,
@@ -159,4 +159,4 @@ int GetDefaultCacheShardBits(size_t capacity) {
  return num_shard_bits;
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -16,7 +16,7 @@
 #include "rocksdb/cache.h"
 #include "util/hash.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 // Single cache shard interface.
 class CacheShard {
@@ -40,6 +40,13 @@ class CacheShard {
                                      bool thread_safe) = 0;
  virtual void EraseUnRefEntries() = 0;
  virtual std::string GetPrintableOptions() const { return ""; }
+  void set_metadata_charge_policy(
+      CacheMetadataChargePolicy metadata_charge_policy) {
+    metadata_charge_policy_ = metadata_charge_policy;
+  }
+
+ protected:
+  CacheMetadataChargePolicy metadata_charge_policy_ = kDontChargeCacheMetadata;
 };

 // Generic cache interface which shards cache by hash of keys. 2^num_shard_bits
@@ -101,4 +108,4 @@ class ShardedCache : public Cache {

 extern int GetDefaultCacheShardBits(size_t capacity);

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -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)
@@ -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()
@@ -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()
@@ -0,0 +1,29 @@
+# - 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_FOUND - True if gflags found.
+
+find_path(GFLAGS_INCLUDE_DIR
+  NAMES gflags/gflags.h)
+
+find_library(GFLAGS_LIBRARIES
+  NAMES gflags)
+
+include(FindPackageHandleStandardArgs)
+find_package_handle_standard_args(gflags
+  DEFAULT_MSG GFLAGS_LIBRARIES GFLAGS_INCLUDE_DIR)
+
+mark_as_advanced(
+  GFLAGS_LIBRARIES
+  GFLAGS_INCLUDE_DIR)
+
+if(gflags_FOUND AND NOT (TARGET gflags::gflags))
+  add_library(gflags::gflags UNKNOWN IMPORTED)
+  set_target_properties(gflags::gflags
+    PROPERTIES
+      IMPORTED_LOCATION ${GFLAGS_LIBRARIES}
+      INTERFACE_INCLUDE_DIRECTORIES ${GFLAGS_INCLUDE_DIR}
+      IMPORTED_LINK_INTERFACE_LANGUAGES "CXX")
+endif()
@@ -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()
@@ -1,7 +1,7 @@
 # Read rocksdb version from version.h header file.

 function(get_rocksdb_version version_var)
-  file(READ "${CMAKE_SOURCE_DIR}/include/rocksdb/version.h" version_header_file)
+  file(READ "${CMAKE_CURRENT_SOURCE_DIR}/include/rocksdb/version.h" version_header_file)
  foreach(component MAJOR MINOR PATCH)
    string(REGEX MATCH "#define ROCKSDB_${component} ([0-9]+)" _ ${version_header_file})
    set(ROCKSDB_VERSION_${component} ${CMAKE_MATCH_1})
@@ -12,21 +12,24 @@ fi
 ROOT=".."
 # Fetch right version of gcov
 if [ -d /mnt/gvfs/third-party -a -z "$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"

@@ -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)

@@ -0,0 +1,108 @@
+//  Copyright (c) 2011-present, Facebook, Inc.  All rights reserved.
+//  This source code is licensed under both the GPLv2 (found in the
+//  COPYING file in the root directory) and Apache 2.0 License
+//  (found in the LICENSE.Apache file in the root directory).
+//
+// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file. See the AUTHORS file for names of contributors.
+
+#include "db/arena_wrapped_db_iter.h"
+#include "memory/arena.h"
+#include "rocksdb/env.h"
+#include "rocksdb/iterator.h"
+#include "rocksdb/options.h"
+#include "table/internal_iterator.h"
+#include "table/iterator_wrapper.h"
+#include "util/user_comparator_wrapper.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+Status ArenaWrappedDBIter::GetProperty(std::string prop_name,
+                                       std::string* prop) {
+  if (prop_name == "rocksdb.iterator.super-version-number") {
+    // First try to pass the value returned from inner iterator.
+    if (!db_iter_->GetProperty(prop_name, prop).ok()) {
+      *prop = ToString(sv_number_);
+    }
+    return Status::OK();
+  }
+  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) {
+  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);
+  sv_number_ = version_number;
+  read_options_ = read_options;
+  allow_refresh_ = allow_refresh;
+}
+
+Status ArenaWrappedDBIter::Refresh() {
+  if (cfd_ == nullptr || db_impl_ == nullptr || !allow_refresh_) {
+    return Status::NotSupported("Creating renew iterator is not allowed.");
+  }
+  assert(db_iter_ != nullptr);
+  // 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.
+  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();
+    arena_.~Arena();
+    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_,
+         allow_refresh_);
+
+    InternalIterator* internal_iter = db_impl_->NewInternalIterator(
+        read_options_, cfd_, sv, &arena_, db_iter_->GetRangeDelAggregator(),
+        latest_seq, /* allow_unprepared_value */ true);
+    SetIterUnderDBIter(internal_iter);
+  } else {
+    db_iter_->set_sequence(db_impl_->GetLatestSequenceNumber());
+    db_iter_->set_valid(false);
+  }
+  return Status::OK();
+}
+
+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) {
+  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);
+  if (db_impl != nullptr && cfd != nullptr && allow_refresh) {
+    iter->StoreRefreshInfo(db_impl, cfd, read_callback, allow_blob);
+  }
+
+  return iter;
+}
+
+}  // namespace ROCKSDB_NAMESPACE
@@ -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).
+//
+// 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 <stdint.h>
+#include <string>
+#include "db/db_impl/db_impl.h"
+#include "db/db_iter.h"
+#include "db/dbformat.h"
+#include "db/range_del_aggregator.h"
+#include "memory/arena.h"
+#include "options/cf_options.h"
+#include "rocksdb/db.h"
+#include "rocksdb/iterator.h"
+#include "util/autovector.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+class Arena;
+
+// 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
+// a iterator hierarchy whose memory can be allocated inline. In that way,
+// accessing the iterator tree can be more cache friendly. It is also faster
+// to allocate.
+// When using the class's Iterator interface, the behavior is exactly
+// the same as the inner DBIter.
+class ArenaWrappedDBIter : public Iterator {
+ public:
+  virtual ~ArenaWrappedDBIter() { db_iter_->~DBIter(); }
+
+  // Get the arena to be used to allocate memory for DBIter to be wrapped,
+  // as well as child iterators in it.
+  virtual Arena* GetArena() { return &arena_; }
+  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.
+  virtual void SetIterUnderDBIter(InternalIterator* iter) {
+    db_iter_->SetIter(iter);
+  }
+
+  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);
+  }
+  void Next() override { db_iter_->Next(); }
+  void Prev() override { db_iter_->Prev(); }
+  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;
+
+  Status Refresh() override;
+
+  void 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_iterations, uint64_t version_number,
+            ReadCallback* read_callback, DBImpl* db_impl, ColumnFamilyData* cfd,
+            bool allow_blob, bool allow_refresh);
+
+  // Store some parameters so we can refresh the iterator at a later point
+  // with these same params
+  void StoreRefreshInfo(DBImpl* db_impl, ColumnFamilyData* cfd,
+                        ReadCallback* read_callback, bool allow_blob) {
+    db_impl_ = db_impl;
+    cfd_ = cfd;
+    read_callback_ = read_callback;
+    allow_blob_ = allow_blob;
+  }
+
+ private:
+  DBIter* db_iter_;
+  Arena arena_;
+  uint64_t sv_number_;
+  ColumnFamilyData* cfd_ = nullptr;
+  DBImpl* db_impl_ = nullptr;
+  ReadOptions read_options_;
+  ReadCallback* read_callback_;
+  bool allow_blob_ = false;
+  bool allow_refresh_ = true;
+};
+
+// Generate the arena wrapped iterator class.
+// `db_impl` and `cfd` are used for reneweal. If left null, renewal will not
+// be supported.
+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);
+}  // namespace ROCKSDB_NAMESPACE
@@ -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
@@ -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
@@ -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
@@ -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();
+}
@@ -0,0 +1,316 @@
+//  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_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 "rocksdb/slice.h"
+#include "rocksdb/status.h"
+#include "test_util/sync_point.h"
+#include "util/compression.h"
+
+namespace ROCKSDB_NAMESPACE {
+
+BlobFileBuilder::BlobFileBuilder(
+    VersionSet* versions, Env* env, 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,
+    std::vector<std::string>* blob_file_paths,
+    std::vector<BlobFileAddition>* blob_file_additions)
+    : BlobFileBuilder([versions]() { return versions->NewFileNumber(); }, env,
+                      fs, immutable_cf_options, mutable_cf_options,
+                      file_options, job_id, column_family_id,
+                      column_family_name, io_priority, write_hint,
+                      blob_file_paths, blob_file_additions) {}
+
+BlobFileBuilder::BlobFileBuilder(
+    std::function<uint64_t()> file_number_generator, Env* env, 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,
+    std::vector<std::string>* blob_file_paths,
+    std::vector<BlobFileAddition>* blob_file_additions)
+    : file_number_generator_(std::move(file_number_generator)),
+      env_(env),
+      fs_(fs),
+      immutable_cf_options_(immutable_cf_options),
+      min_blob_size_(mutable_cf_options->min_blob_size),
+      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),
+      blob_file_paths_(blob_file_paths),
+      blob_file_additions_(blob_file_additions),
+      blob_count_(0),
+      blob_bytes_(0) {
+  assert(file_number_generator_);
+  assert(env_);
+  assert(fs_);
+  assert(immutable_cf_options_);
+  assert(file_options_);
+  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;
+
+  {
+    TEST_SYNC_POINT("BlobFileBuilder::OpenBlobFileIfNeeded:NewWritableFile");
+
+    assert(file_options_);
+    const Status s =
+        NewWritableFile(fs_, blob_file_path, &file, *file_options_);
+    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_);
+
+  Statistics* const statistics = immutable_cf_options_->statistics;
+
+  std::unique_ptr<WritableFileWriter> file_writer(new WritableFileWriter(
+      std::move(file), blob_file_paths_->back(), *file_options_, env_,
+      nullptr /*IOTracer*/, statistics, immutable_cf_options_->listeners,
+      immutable_cf_options_->file_checksum_gen_factory));
+
+  std::unique_ptr<BlobLogWriter> blob_log_writer(
+      new BlobLogWriter(std::move(file_writer), env_, statistics,
+                        blob_file_number, immutable_cf_options_->use_fsync));
+
+  constexpr bool has_ttl = false;
+  constexpr ExpirationRange expiration_range;
+
+  BlobLogHeader header(column_family_id_, blob_compression_type_, has_ttl,
+                       expiration_range);
+
+  {
+    TEST_SYNC_POINT("BlobFileBuilder::OpenBlobFileIfNeeded:WriteHeader");
+
+    const Status s = blob_log_writer->WriteHeader(header);
+    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;
+
+  TEST_SYNC_POINT("BlobFileBuilder::WriteBlobToFile:AddRecord");
+
+  const Status s = writer_->AddRecord(key, blob, &key_offset, blob_offset);
+  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;
+
+  TEST_SYNC_POINT("BlobFileBuilder::WriteBlobToFile:AppendFooter");
+
+  const Status s =
+      writer_->AppendFooter(footer, &checksum_method, &checksum_value);
+  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_);
+
+  writer_.reset();
+  blob_count_ = 0;
+  blob_bytes_ = 0;
+
+  return Status::OK();
+}
+
+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();
+}
+
+}  // namespace ROCKSDB_NAMESPACE
@@ -0,0 +1,91 @@
+//  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;
+struct ImmutableCFOptions;
+struct MutableCFOptions;
+struct FileOptions;
+class BlobFileAddition;
+class Status;
+class Slice;
+class BlobLogWriter;
+
+class BlobFileBuilder {
+ public:
+  BlobFileBuilder(VersionSet* versions, Env* env, 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,
+                  std::vector<std::string>* blob_file_paths,
+                  std::vector<BlobFileAddition>* blob_file_additions);
+
+  BlobFileBuilder(std::function<uint64_t()> file_number_generator, Env* env,
+                  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,
+                  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();
+
+ 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_;
+  Env* env_;
+  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::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
@@ -0,0 +1,659 @@
+//  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/composite_env_wrapper.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_) {}
+
+  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,
+                                   &mock_env_));
+
+    constexpr Statistics* statistics = nullptr;
+    BlobLogSequentialReader blob_log_reader(std::move(file_reader), &mock_env_,
+                                            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_;
+  LegacyFileSystemWrapper fs_;
+  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;
+
+  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(), &mock_env_, &fs_,
+                          &immutable_cf_options, &mutable_cf_options,
+                          &file_options_, job_id, column_family_id,
+                          column_family_name, io_priority, write_hint,
+                          &blob_file_paths, &blob_file_additions);
+
+  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;
+
+  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(), &mock_env_, &fs_,
+                          &immutable_cf_options, &mutable_cf_options,
+                          &file_options_, job_id, column_family_id,
+                          column_family_name, io_priority, write_hint,
+                          &blob_file_paths, &blob_file_additions);
+
+  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;
+
+  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(), &mock_env_, &fs_,
+                          &immutable_cf_options, &mutable_cf_options,
+                          &file_options_, job_id, column_family_id,
+                          column_family_name, io_priority, write_hint,
+                          &blob_file_paths, &blob_file_additions);
+
+  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;
+
+  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(), &mock_env_, &fs_,
+                          &immutable_cf_options, &mutable_cf_options,
+                          &file_options_, job_id, column_family_id,
+                          column_family_name, io_priority, write_hint,
+                          &blob_file_paths, &blob_file_additions);
+
+  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;
+
+  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(), &mock_env_, &fs_,
+                          &immutable_cf_options, &mutable_cf_options,
+                          &file_options_, job_id, column_family_id,
+                          column_family_name, io_priority, write_hint,
+                          &blob_file_paths, &blob_file_additions);
+
+  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>();
+
+  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(), &mock_env_, &fs_,
+                          &immutable_cf_options, &mutable_cf_options,
+                          &file_options_, job_id, column_family_id,
+                          column_family_name, io_priority, write_hint,
+                          &blob_file_paths, &blob_file_additions);
+
+  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()),
+        fault_injection_env_(&mock_env_),
+        fs_(&fault_injection_env_),
+        sync_point_(GetParam()) {}
+
+  MockEnv mock_env_;
+  FaultInjectionTestEnv fault_injection_env_;
+  LegacyFileSystemWrapper 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(&fault_injection_env_,
+                            "BlobFileBuilderIOErrorTest_IOError"),
+      0);
+  options.enable_blob_files = true;
+  options.blob_file_size = value_size;
+
+  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(), &fault_injection_env_,
+                          &fs_, &immutable_cf_options, &mutable_cf_options,
+                          &file_options_, job_id, column_family_id,
+                          column_family_name, io_priority, write_hint,
+                          &blob_file_paths, &blob_file_additions);
+
+  SyncPoint::GetInstance()->SetCallBack(sync_point_, [this](void* /* arg */) {
+    fault_injection_env_.SetFilesystemActive(false,
+                                             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();
+}
@@ -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
@@ -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
@@ -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();
+}
@@ -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
@@ -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
@@ -0,0 +1,423 @@
+//  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,
+    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, &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, 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.env,
+      std::shared_ptr<IOTracer>(), immutable_cf_options.statistics,
+      BLOB_DB_BLOB_FILE_READ_MICROS, blob_file_read_hist,
+      immutable_cf_options.rate_limiter, immutable_cf_options.listeners));
+
+  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) 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);
+
+  Slice record_slice;
+  Buffer buf;
+  AlignedBuf aligned_buf;
+
+  {
+    TEST_SYNC_POINT("BlobFileReader::GetBlob:ReadFromFile");
+
+    const uint64_t record_offset = offset - adjustment;
+    const uint64_t record_size = value_size + adjustment;
+
+    const Status s = ReadFromFile(file_reader_.get(), record_offset,
+                                  static_cast<size_t>(record_size),
+                                  &record_slice, &buf, &aligned_buf);
+    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;
+    }
+  }
+
+  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
@@ -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).
+
+#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,
+                       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) 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, 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
@@ -0,0 +1,770 @@
+//  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/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.env));
+
+  constexpr Statistics* statistics = nullptr;
+  constexpr bool use_fsync = false;
+
+  BlobLogWriter blob_log_writer(std::move(file_writer),
+                                immutable_cf_options.env, statistics,
+                                blob_file_number, use_fsync);
+
+  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, &reader));
+
+  // Make sure the blob can be retrieved with and without checksum verification
+  ReadOptions read_options;
+  read_options.verify_checksums = false;
+
+  {
+    PinnableSlice value;
+
+    ASSERT_OK(reader->GetBlob(read_options, key, blob_offset, blob_size,
+                              kNoCompression, &value));
+    ASSERT_EQ(value, blob);
+  }
+
+  read_options.verify_checksums = true;
+
+  {
+    PinnableSlice value;
+
+    ASSERT_OK(reader->GetBlob(read_options, key, blob_offset, blob_size,
+                              kNoCompression, &value));
+    ASSERT_EQ(value, blob);
+  }
+
+  // Invalid offset (too close to start of file)
+  {
+    PinnableSlice value;
+
+    ASSERT_TRUE(reader
+                    ->GetBlob(read_options, key, blob_offset - 1, blob_size,
+                              kNoCompression, &value)
+                    .IsCorruption());
+  }
+
+  // Invalid offset (too close to end of file)
+  {
+    PinnableSlice value;
+
+    ASSERT_TRUE(reader
+                    ->GetBlob(read_options, key, blob_offset + 1, blob_size,
+                              kNoCompression, &value)
+                    .IsCorruption());
+  }
+
+  // Incorrect compression type
+  {
+    PinnableSlice value;
+
+    ASSERT_TRUE(
+        reader
+            ->GetBlob(read_options, key, blob_offset, blob_size, kZSTD, &value)
+            .IsCorruption());
+  }
+
+  // Incorrect key size
+  {
+    constexpr char shorter_key[] = "k";
+    PinnableSlice value;
+
+    ASSERT_TRUE(reader
+                    ->GetBlob(read_options, shorter_key,
+                              blob_offset - (sizeof(key) - sizeof(shorter_key)),
+                              blob_size, kNoCompression, &value)
+                    .IsCorruption());
+  }
+
+  // Incorrect key
+  {
+    constexpr char incorrect_key[] = "foo";
+    PinnableSlice value;
+
+    ASSERT_TRUE(reader
+                    ->GetBlob(read_options, incorrect_key, blob_offset,
+                              blob_size, kNoCompression, &value)
+                    .IsCorruption());
+  }
+
+  // Incorrect value size
+  {
+    PinnableSlice value;
+
+    ASSERT_TRUE(reader
+                    ->GetBlob(read_options, key, blob_offset, blob_size + 1,
+                              kNoCompression, &value)
+                    .IsCorruption());
+  }
+}
+
+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.env));
+
+    constexpr Statistics* statistics = nullptr;
+    constexpr bool use_fsync = false;
+
+    BlobLogWriter blob_log_writer(std::move(file_writer),
+                                  immutable_cf_options.env, statistics,
+                                  blob_file_number, use_fsync);
+
+    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, &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, &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, &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, &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,
+                                     &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, &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;
+
+  ASSERT_TRUE(reader
+                  ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
+                            kNoCompression, &value)
+                  .IsCorruption());
+
+  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, &reader));
+
+  // Make sure the blob can be retrieved with and without checksum verification
+  ReadOptions read_options;
+  read_options.verify_checksums = false;
+
+  {
+    PinnableSlice value;
+
+    ASSERT_OK(reader->GetBlob(read_options, key, blob_offset, blob_size,
+                              kSnappyCompression, &value));
+    ASSERT_EQ(value, blob);
+  }
+
+  read_options.verify_checksums = true;
+
+  {
+    PinnableSlice value;
+
+    ASSERT_OK(reader->GetBlob(read_options, key, blob_offset, blob_size,
+                              kSnappyCompression, &value));
+    ASSERT_EQ(value, blob);
+  }
+}
+
+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, &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;
+
+  ASSERT_TRUE(reader
+                  ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
+                            kSnappyCompression, &value)
+                  .IsCorruption());
+
+  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, &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;
+
+    ASSERT_TRUE(reader
+                    ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
+                              kNoCompression, &value)
+                    .IsIOError());
+  }
+
+  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, &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;
+
+    ASSERT_TRUE(reader
+                    ->GetBlob(ReadOptions(), key, blob_offset, blob_size,
+                              kNoCompression, &value)
+                    .IsCorruption());
+  }
+
+  SyncPoint::GetInstance()->DisableProcessing();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
+}  // namespace ROCKSDB_NAMESPACE
+
+int main(int argc, char** argv) {
+  ::testing::InitGoogleTest(&argc, argv);
+  return RUN_ALL_TESTS();
+}
@@ -3,14 +3,16 @@
 //  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 "rocksdb/options.h"
+#include <sstream>
+#include <string>
+
+#include "rocksdb/compression_type.h"
 #include "util/coding.h"
+#include "util/compression.h"
 #include "util/string_util.h"

-namespace rocksdb {
-namespace blob_db {
+namespace ROCKSDB_NAMESPACE {

 // BlobIndex is a pointer to the blob and metadata of the blob. The index is
 // stored in base DB as ValueType::kTypeBlobIndex.
@@ -81,6 +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);
@@ -109,6 +116,24 @@ class BlobIndex {
    return Status::OK();
  }

+  std::string DebugString(bool output_hex) const {
+    std::ostringstream oss;
+
+    if (IsInlined()) {
+      oss << "[inlined blob] value:" << value_.ToString(output_hex);
+    } else {
+      oss << "[blob ref] file:" << file_number_ << " offset:" << offset_
+          << " size:" << size_
+          << " compression: " << CompressionTypeToString(compression_);
+    }
+
+    if (HasTTL()) {
+      oss << " exp:" << expiration_;
+    }
+
+    return oss.str();
+  }
+
  static void EncodeInlinedTTL(std::string* dst, uint64_t expiration,
                               const Slice& value) {
    assert(dst != nullptr);
@@ -156,6 +181,4 @@ class BlobIndex {
  CompressionType compression_ = kNoCompression;
 };

-}  // namespace blob_db
-}  // namespace rocksdb
-#endif  // ROCKSDB_LITE
+}  // namespace ROCKSDB_NAMESPACE
@@ -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 blob_db {
+namespace ROCKSDB_NAMESPACE {

 void BlobLogHeader::EncodeTo(std::string* dst) {
  assert(dst != nullptr);
@@ -97,6 +95,10 @@ Status BlobLogFooter::DecodeFrom(Slice src) {
  return Status::OK();
 }

+uint64_t BlobLogRecord::CalculateAdjustmentForRecordHeader(uint64_t key_size) {
+  return key_size + kHeaderSize;
+}
+
 void BlobLogRecord::EncodeHeaderTo(std::string* dst) {
  assert(dst != nullptr);
  dst->clear();
@@ -144,6 +146,4 @@ Status BlobLogRecord::CheckBlobCRC() const {
  return Status::OK();
 }

-}  // namespace blob_db
-}  // namespace rocksdb
-#endif  // ROCKSDB_LITE
+}  // namespace ROCKSDB_NAMESPACE
@@ -7,9 +7,6 @@

 #pragma once

-#ifndef ROCKSDB_LITE
-
-#include <limits>
 #include <memory>
 #include <utility>

@@ -18,12 +15,10 @@
 #include "rocksdb/status.h"
 #include "rocksdb/types.h"

-namespace rocksdb {
-namespace blob_db {
+namespace ROCKSDB_NAMESPACE {

 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>;

@@ -43,11 +38,19 @@ using ExpirationRange = std::pair<uint64_t, uint64_t>;
 struct BlobLogHeader {
  static constexpr size_t kSize = 30;

+  BlobLogHeader() = default;
+  BlobLogHeader(uint32_t _column_family_id, CompressionType _compression,
+                bool _has_ttl, const ExpirationRange& _expiration_range)
+      : column_family_id(_column_family_id),
+        compression(_compression),
+        has_ttl(_has_ttl),
+        expiration_range(_expiration_range) {}
+
  uint32_t version = kVersion1;
  uint32_t column_family_id = 0;
  CompressionType compression = kNoCompression;
  bool has_ttl = false;
-  ExpirationRange expiration_range = std::make_pair(0, 0);
+  ExpirationRange expiration_range;

  void EncodeTo(std::string* dst);

@@ -101,6 +104,11 @@ 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);
+
  uint64_t key_size = 0;
  uint64_t value_size = 0;
  uint64_t expiration = 0;
@@ -120,6 +128,19 @@ struct BlobLogRecord {
  Status CheckBlobCRC() const;
 };

-}  // namespace blob_db
-}  // namespace rocksdb
-#endif  // ROCKSDB_LITE
+// 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
@@ -3,30 +3,35 @@
 //  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 "db/blob/blob_log_sequential_reader.h"

 #include <algorithm>

+#include "file/random_access_file_reader.h"
 #include "monitoring/statistics.h"
-#include "util/file_reader_writer.h"
 #include "util/stop_watch.h"

-namespace rocksdb {
-namespace blob_db {
+namespace ROCKSDB_NAMESPACE {

-Reader::Reader(std::unique_ptr<RandomAccessFileReader>&& file_reader, Env* env,
-               Statistics* statistics)
+BlobLogSequentialReader::BlobLogSequentialReader(
+    std::unique_ptr<RandomAccessFileReader>&& file_reader, Env* env,
+    Statistics* statistics)
    : file_(std::move(file_reader)),
      env_(env),
      statistics_(statistics),
-      buffer_(),
      next_byte_(0) {}

-Status Reader::ReadSlice(uint64_t size, Slice* slice, char* buf) {
+BlobLogSequentialReader::~BlobLogSequentialReader() = default;
+
+Status BlobLogSequentialReader::ReadSlice(uint64_t size, Slice* slice,
+                                          char* buf) {
+  assert(slice);
+  assert(file_);
+
  StopWatch read_sw(env_, statistics_, BLOB_DB_BLOB_FILE_READ_MICROS);
-  Status s = file_->Read(next_byte_, static_cast<size_t>(size), slice, buf);
+  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 +43,13 @@ 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 +62,13 @@ 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 +114,21 @@ Status Reader::ReadRecord(BlobLogRecord* record, ReadLevel level,
  return s;
 }

-}  // namespace blob_db
-}  // namespace rocksdb
-#endif  // ROCKSDB_LITE
+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
@@ -0,0 +1,76 @@
+//  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;
+
+/**
+ * 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,
+                          Env* env, 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_;
+  Env* env_;
+  Statistics* statistics_;
+
+  Slice buffer_;
+  char header_buf_[BlobLogRecord::kHeaderSize];
+
+  // which byte to read next
+  uint64_t next_byte_;
+};
+
+}  // namespace ROCKSDB_NAMESPACE
@@ -2,44 +2,45 @@
 //  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 "test_util/sync_point.h"
 #include "util/coding.h"
-#include "util/file_reader_writer.h"
 #include "util/stop_watch.h"
-#include "utilities/blob_db/blob_log_format.h"

-namespace rocksdb {
-namespace blob_db {
+namespace ROCKSDB_NAMESPACE {

-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,
+                             Env* env, Statistics* statistics,
+                             uint64_t log_number, bool use_fs, uint64_t boffset)
    : dest_(std::move(dest)),
      env_(env),
      statistics_(statistics),
      log_number_(log_number),
      block_offset_(boffset),
-      bytes_per_sync_(bpsync),
-      next_sync_offset_(0),
      use_fsync_(use_fs),
      last_elem_type_(kEtNone) {}

-Status Writer::Sync() {
+BlobLogWriter::~BlobLogWriter() = default;
+
+Status BlobLogWriter::Sync() {
+  TEST_SYNC_POINT("BlobLogWriter::Sync");
+
  StopWatch sync_sw(env_, 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;
@@ -56,7 +57,9 @@ 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 +69,34 @@ 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 +106,9 @@ 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 +119,8 @@ 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 +131,8 @@ 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;
@@ -110,9 +140,10 @@ 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) {
+Status BlobLogWriter::EmitPhysicalRecord(const std::string& headerbuf,
+                                         const Slice& key, const Slice& val,
+                                         uint64_t* key_offset,
+                                         uint64_t* blob_offset) {
  StopWatch write_sw(env_, statistics_, BLOB_DB_BLOB_FILE_WRITE_MICROS);
  Status s = dest_->Append(Slice(headerbuf));
  if (s.ok()) {
@@ -134,6 +165,4 @@ Status Writer::EmitPhysicalRecord(const std::string& headerbuf,
  return s;
 }

-}  // namespace blob_db
-}  // namespace rocksdb
-#endif  // ROCKSDB_LITE
+}  // namespace ROCKSDB_NAMESPACE
@@ -4,46 +4,42 @@
 //  (found in the LICENSE.Apache file in the root directory).
 #pragma once

-#ifndef ROCKSDB_LITE
-
 #include <cstdint>
 #include <memory>
 #include <string>

+#include "db/blob/blob_log_format.h"
 #include "rocksdb/env.h"
 #include "rocksdb/slice.h"
 #include "rocksdb/statistics.h"
 #include "rocksdb/status.h"
 #include "rocksdb/types.h"
-#include "utilities/blob_db/blob_log_format.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 class WritableFileWriter;

-namespace blob_db {
-
 /**
- * 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, Env* env,
+                Statistics* statistics, uint64_t log_number, bool use_fsync,
+                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 +54,8 @@ 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);

@@ -68,20 +65,14 @@ 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_;
  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_;

 public:
@@ -89,6 +80,4 @@ class Writer {
  ElemType last_elem_type_;
 };

-}  // namespace blob_db
-}  // namespace rocksdb
-#endif  // ROCKSDB_LITE
+}  // namespace ROCKSDB_NAMESPACE
@@ -12,6 +12,7 @@
 #include <utility>
 #include <vector>

+#include "db/arena_wrapped_db_iter.h"
 #include "db/column_family.h"
 #include "db/db_iter.h"
 #include "db/db_test_util.h"
@@ -22,7 +23,7 @@
 #include "util/string_util.h"
 #include "utilities/merge_operators.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 // kTypeBlobIndex is a value type used by BlobDB only. The base rocksdb
 // should accept the value type on write, and report not supported value
@@ -40,12 +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,
@@ -397,16 +399,39 @@ TEST_F(DBBlobIndexTest, Iterate) {
    verify(15, Status::kOk, get_value(16, 0), get_value(14, 0),
           create_blob_iterator, check_is_blob(false));

+#ifndef ROCKSDB_LITE
+    // Iterator with blob support and using seek.
+    ASSERT_OK(dbfull()->SetOptions(
+        cfh(), {{"max_sequential_skip_in_iterations", "0"}}));
+    verify(1, Status::kOk, get_value(1, 0), get_value(1, 0),
+           create_blob_iterator, check_is_blob(true));
+    verify(3, Status::kOk, get_value(3, 0), get_value(3, 0),
+           create_blob_iterator, check_is_blob(true));
+    verify(5, Status::kOk, get_value(5, 0), get_value(5, 0),
+           create_blob_iterator, check_is_blob(false));
+    verify(7, Status::kOk, get_value(8, 0), get_value(6, 0),
+           create_blob_iterator, check_is_blob(false));
+    verify(9, Status::kOk, get_value(10, 0), get_value(8, 0),
+           create_blob_iterator, check_is_blob(false));
+    verify(11, Status::kNotSupported, "", "", create_blob_iterator);
+    verify(13, Status::kOk,
+           get_value(13, 2) + "," + get_value(13, 1) + "," + get_value(13, 0),
+           get_value(13, 2) + "," + get_value(13, 1) + "," + get_value(13, 0),
+           create_blob_iterator, check_is_blob(false));
+    verify(15, Status::kOk, get_value(16, 0), get_value(14, 0),
+           create_blob_iterator, check_is_blob(false));
+#endif  // !ROCKSDB_LITE
+
    for (auto* snapshot : snapshots) {
      dbfull()->ReleaseSnapshot(snapshot);
    }
  }
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE

 int main(int argc, char** argv) {
-  rocksdb::port::InstallStackTraceHandler();
+  ROCKSDB_NAMESPACE::port::InstallStackTraceHandler();
  ::testing::InitGoogleTest(&argc, argv);
  return RUN_ALL_TESTS();
 }
@@ -13,15 +13,19 @@
 #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/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 "rocksdb/db.h"
@@ -33,10 +37,9 @@
 #include "table/format.h"
 #include "table/internal_iterator.h"
 #include "test_util/sync_point.h"
-#include "util/file_reader_writer.h"
 #include "util/stop_watch.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 class TableFactory;

@@ -50,7 +53,8 @@ TableBuilder* NewTableBuilder(
    uint64_t sample_for_compression, const CompressionOptions& compression_opts,
    int level, const bool skip_filters, const uint64_t creation_time,
    const uint64_t oldest_key_time, const uint64_t target_file_size,
-    const uint64_t file_creation_time) {
+    const 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());
@@ -60,17 +64,19 @@ TableBuilder* NewTableBuilder(
                          sample_for_compression, compression_opts,
                          skip_filters, column_family_name, level,
                          creation_time, oldest_key_time, target_file_size,
-                          file_creation_time),
+                          file_creation_time, db_id, db_session_id),
      column_family_id, file);
 }

 Status BuildTable(
-    const std::string& dbname, Env* env, const ImmutableCFOptions& ioptions,
-    const MutableCFOptions& mutable_cf_options, const EnvOptions& env_options,
+    const std::string& dbname, VersionSet* versions, Env* env, FileSystem* fs,
+    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,
@@ -79,15 +85,23 @@ Status BuildTable(
    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) {
+    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) {
  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();
@@ -99,12 +113,14 @@ 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;
-
  if (iter->Valid() || !range_del_agg->IsEmpty()) {
    TableBuilder* builder;
    std::unique_ptr<WritableFileWriter> file_writer;
@@ -114,31 +130,39 @@ Status BuildTable(
    compression_opts_for_flush.max_dict_bytes = 0;
    compression_opts_for_flush.zstd_max_train_bytes = 0;
    {
-      std::unique_ptr<WritableFile> file;
+      std::unique_ptr<FSWritableFile> file;
 #ifndef NDEBUG
-      bool use_direct_writes = env_options.use_direct_writes;
+      bool use_direct_writes = file_options.use_direct_writes;
      TEST_SYNC_POINT_CALLBACK("BuildTable:create_file", &use_direct_writes);
 #endif  // !NDEBUG
-      s = NewWritableFile(env, fname, &file, env_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, tp, reason, s);
+            job_id, meta->fd, kInvalidBlobFileNumber, tp, reason, s,
+            file_checksum, file_checksum_func_name);
        return s;
      }
      file->SetIOPriority(io_priority);
      file->SetWriteLifeTimeHint(write_hint);

-      file_writer.reset(
-          new WritableFileWriter(std::move(file), fname, env_options, env,
-                                 ioptions.statistics, ioptions.listeners));
+      file_writer.reset(new WritableFileWriter(
+          std::move(file), fname, file_options, env, io_tracer,
+          ioptions.statistics, ioptions.listeners,
+          ioptions.file_checksum_gen_factory));
+
      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);
+          0 /*target_file_size*/, file_creation_time, db_id, db_session_id);
    }

    MergeHelper merge(env, internal_comparator.user_comparator(),
@@ -147,17 +171,34 @@ 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, env, fs, &ioptions,
+                                  &mutable_cf_options, &file_options, job_id,
+                                  column_family_id, column_family_name,
+                                  io_priority, write_hint, &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);
+
    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, c_iter.ikey().sequence);
+      meta->UpdateBoundaries(key, value, ikey.sequence, ikey.type);

      // TODO(noetzli): Update stats after flush, too.
      if (io_priority == Env::IO_HIGH &&
@@ -166,26 +207,37 @@ Status BuildTable(
            ThreadStatus::FLUSH_BYTES_WRITTEN, IOSTATS(bytes_written));
      }
    }
+    if (!s.ok()) {
+      c_iter.status().PermitUncheckedError();
+    } else if (!c_iter.status().ok()) {
+      s = c_iter.status();
+    }
+    if (s.ok()) {
+      auto range_del_it = range_del_agg->NewIterator();
+      for (range_del_it->SeekToFirst(); range_del_it->Valid();
+           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);
+      }

-    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 (blob_file_builder) {
+        s = blob_file_builder->Finish();
+      }
    }

-    // Finish and check for builder errors
-    tp = builder->GetTableProperties();
-    bool empty = builder->NumEntries() == 0 && tp.num_range_deletions == 0;
-    s = c_iter.status();
+    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();
@@ -200,13 +252,28 @@ Status BuildTable(
    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);
+      *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;
+    }
+
+    // TODO Also check the IO status when create the Iterator.

    if (s.ok() && !empty) {
      // Verify that the table is usable
@@ -215,20 +282,32 @@ 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(), env_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");
+        }
      }
    }
  }
@@ -239,15 +318,33 @@ Status BuildTable(
  }

  if (!s.ok() || meta->fd.GetFileSize() == 0) {
-    env->DeleteFile(fname);
+    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 = fs->DeleteFile(blob_file_path, IOOptions(), dbg);
+        ignored.PermitUncheckedError();
+      }
+
+      blob_file_additions->clear();
+    }
  }

+  if (meta->fd.GetFileSize() == 0) {
+    fname = "(nil)";
+  }
  // Output to event logger and fire events.
  EventHelpers::LogAndNotifyTableFileCreationFinished(
      event_logger, ioptions.listeners, dbname, column_family_name, fname,
-      job_id, meta->fd, tp, reason, s);
+      job_id, meta->fd, meta->oldest_blob_file_number, tp, reason, s,
+      file_checksum, file_checksum_func_name);

  return s;
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -22,13 +22,15 @@
 #include "rocksdb/types.h"
 #include "table/scoped_arena_iterator.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 struct Options;
 struct FileMetaData;

+class VersionSet;
 class Env;
 struct EnvOptions;
+class BlobFileAddition;
 class Iterator;
 class SnapshotChecker;
 class TableCache;
@@ -51,7 +53,8 @@ TableBuilder* NewTableBuilder(
    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
@@ -62,12 +65,14 @@ 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, const ImmutableCFOptions& options,
-    const MutableCFOptions& mutable_cf_options, const EnvOptions& env_options,
+    const std::string& dbname, VersionSet* versions, Env* env, FileSystem* fs,
+    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,
@@ -77,11 +82,13 @@ extern Status BuildTable(
    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 = "");

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -9,11 +9,12 @@

 #include "db/column_family.h"

-#include <cinttypes>
-#include <vector>
-#include <string>
 #include <algorithm>
+#include <cinttypes>
 #include <limits>
+#include <sstream>
+#include <string>
+#include <vector>

 #include "db/compaction/compaction_picker.h"
 #include "db/compaction/compaction_picker_fifo.h"
@@ -30,12 +31,14 @@
 #include "memtable/hash_skiplist_rep.h"
 #include "monitoring/thread_status_util.h"
 #include "options/options_helper.h"
-#include "table/block_based/block_based_table_factory.h"
+#include "port/port.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 ROCKSDB_NAMESPACE {

 ColumnFamilyHandleImpl::ColumnFamilyHandleImpl(
    ColumnFamilyData* column_family_data, DBImpl* db, InstrumentedMutex* mutex)
@@ -59,11 +62,8 @@ ColumnFamilyHandleImpl::~ColumnFamilyHandleImpl() {
    ColumnFamilyOptions initial_cf_options_copy = cfd_->initial_cf_options();
    JobContext job_context(0);
    mutex_->Lock();
-    if (cfd_->Unref()) {
-      bool dropped = cfd_->IsDropped();
-
-      delete cfd_;
-
+    bool dropped = cfd_->IsDropped();
+    if (cfd_->UnrefAndTryDelete()) {
      if (dropped) {
        db_->FindObsoleteFiles(&job_context, false, true);
      }
@@ -149,6 +149,16 @@ 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();
 }

@@ -187,6 +197,11 @@ Status CheckCFPathsSupported(const DBOptions& db_options,
  return Status::OK();
 }

+namespace {
+const uint64_t kDefaultTtl = 0xfffffffffffffffe;
+const uint64_t kDefaultPeriodicCompSecs = 0xfffffffffffffffe;
+};  // namespace
+
 ColumnFamilyOptions SanitizeOptions(const ImmutableDBOptions& db_options,
                                    const ColumnFamilyOptions& src) {
  ColumnFamilyOptions result = src;
@@ -342,6 +357,61 @@ ColumnFamilyOptions SanitizeOptions(const ImmutableDBOptions& db_options,
    result.max_compaction_bytes = result.target_file_size_base * 25;
  }

+  bool is_block_based_table = (result.table_factory->IsInstanceOf(
+      TableFactory::kBlockBasedTableName()));
+
+  const uint64_t kAdjustedTtl = 30 * 24 * 60 * 60;
+  if (result.ttl == kDefaultTtl) {
+    if (is_block_based_table &&
+        result.compaction_style != kCompactionStyleFIFO) {
+      result.ttl = kAdjustedTtl;
+    } else {
+      result.ttl = 0;
+    }
+  }
+
+  const uint64_t kAdjustedPeriodicCompSecs = 30 * 24 * 60 * 60;
+
+  // Turn on periodic compactions and set them to occur once every 30 days if
+  // compaction filters are used and periodic_compaction_seconds is set to the
+  // default value.
+  if (result.compaction_style != kCompactionStyleFIFO) {
+    if ((result.compaction_filter != nullptr ||
+         result.compaction_filter_factory != nullptr) &&
+        result.periodic_compaction_seconds == kDefaultPeriodicCompSecs &&
+        is_block_based_table) {
+      result.periodic_compaction_seconds = kAdjustedPeriodicCompSecs;
+    }
+  } else {
+    // result.compaction_style == kCompactionStyleFIFO
+    if (result.ttl == 0) {
+      if (is_block_based_table) {
+        if (result.periodic_compaction_seconds == kDefaultPeriodicCompSecs) {
+          result.periodic_compaction_seconds = kAdjustedPeriodicCompSecs;
+        }
+        result.ttl = result.periodic_compaction_seconds;
+      }
+    } else if (result.periodic_compaction_seconds != 0) {
+      result.ttl = std::min(result.ttl, result.periodic_compaction_seconds);
+    }
+  }
+
+  // TTL compactions would work similar to Periodic Compactions in Universal in
+  // most of the cases. So, if ttl is set, execute the periodic compaction
+  // codepath.
+  if (result.compaction_style == kCompactionStyleUniversal && result.ttl != 0) {
+    if (result.periodic_compaction_seconds != 0) {
+      result.periodic_compaction_seconds =
+          std::min(result.ttl, result.periodic_compaction_seconds);
+    } else {
+      result.periodic_compaction_seconds = result.ttl;
+    }
+  }
+
+  if (result.periodic_compaction_seconds == kDefaultPeriodicCompSecs) {
+    result.periodic_compaction_seconds = 0;
+  }
+
  return result;
 }

@@ -378,13 +448,18 @@ void SuperVersion::Cleanup() {
    to_delete.push_back(m);
  }
  current->Unref();
+  if (cfd->Unref()) {
+    delete cfd;
+  }
 }

-void SuperVersion::Init(MemTable* new_mem, MemTableListVersion* new_imm,
-                        Version* new_current) {
+void SuperVersion::Init(ColumnFamilyData* new_cfd, MemTable* new_mem,
+                        MemTableListVersion* new_imm, Version* new_current) {
+  cfd = new_cfd;
  mem = new_mem;
  imm = new_imm;
  current = new_current;
+  cfd->Ref();
  mem->Ref();
  imm->Ref();
  current->Ref();
@@ -408,12 +483,24 @@ 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 EnvOptions& env_options, ColumnFamilySet* column_family_set,
-    BlockCacheTracer* const block_cache_tracer)
+    const FileOptions& file_options, ColumnFamilySet* column_family_set,
+    BlockCacheTracer* const block_cache_tracer,
+    const std::shared_ptr<IOTracer>& io_tracer)
    : id_(id),
      name_(name),
      dummy_versions_(_dummy_versions),
@@ -444,7 +531,23 @@ 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.
@@ -454,8 +557,8 @@ ColumnFamilyData::ColumnFamilyData(
  if (_dummy_versions != nullptr) {
    internal_stats_.reset(
        new InternalStats(ioptions_.num_levels, db_options.env, this));
-    table_cache_.reset(new TableCache(ioptions_, env_options, _table_cache,
-                                      block_cache_tracer));
+    table_cache_.reset(new TableCache(ioptions_, file_options, _table_cache,
+                                      block_cache_tracer, io_tracer));
    if (ioptions_.compaction_style == kCompactionStyleLevel) {
      compaction_picker_.reset(
          new LevelCompactionPicker(ioptions_, &internal_comparator_));
@@ -520,21 +623,7 @@ ColumnFamilyData::~ColumnFamilyData() {
  // compaction_queue_ and we destroyed it
  assert(!queued_for_flush_);
  assert(!queued_for_compaction_);
-
-  if (super_version_ != nullptr) {
-    // Release SuperVersion reference kept in ThreadLocalPtr.
-    // This must be done outside of mutex_ since unref handler can lock mutex.
-    super_version_->db_mutex->Unlock();
-    local_sv_.reset();
-    super_version_->db_mutex->Lock();
-
-    bool is_last_reference __attribute__((__unused__));
-    is_last_reference = super_version_->Unref();
-    assert(is_last_reference);
-    super_version_->Cleanup();
-    delete super_version_;
-    super_version_ = nullptr;
-  }
+  assert(super_version_ == nullptr);

  if (dummy_versions_ != nullptr) {
    // List must be empty
@@ -552,6 +641,48 @@ 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() {
+  int old_refs = refs_.fetch_sub(1);
+  assert(old_refs > 0);
+
+  if (old_refs == 1) {
+    assert(super_version_ == nullptr);
+    delete this;
+    return true;
+  }
+
+  if (old_refs == 2 && super_version_ != nullptr) {
+    // Only the super_version_ holds me
+    SuperVersion* sv = super_version_;
+    super_version_ = nullptr;
+    // Release SuperVersion reference kept in ThreadLocalPtr.
+    // This must be done outside of mutex_ since unref handler can lock mutex.
+    sv->db_mutex->Unlock();
+    local_sv_.reset();
+    sv->db_mutex->Lock();
+
+    if (sv->Unref()) {
+      // May delete this ColumnFamilyData after calling Cleanup()
+      sv->Cleanup();
+      delete sv;
+      return true;
+    }
+  }
+  return false;
 }

 void ColumnFamilyData::SetDropped() {
@@ -888,8 +1019,8 @@ WriteStallCondition ColumnFamilyData::RecalculateWriteStallConditions(
  return write_stall_condition;
 }

-const EnvOptions* ColumnFamilyData::soptions() const {
-  return &(column_family_set_->env_options_);
+const FileOptions* ColumnFamilyData::soptions() const {
+  return &(column_family_set_->file_options_);
 }

 void ColumnFamilyData::SetCurrent(Version* current_version) {
@@ -924,13 +1055,19 @@ 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);
+      GetName(), mutable_options, mutable_db_options, current_->storage_info(),
+      log_buffer, earliest_mem_seqno);
  if (result != nullptr) {
    result->SetInputVersion(current_);
  }
@@ -965,10 +1102,12 @@ 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();
@@ -979,7 +1118,8 @@ Status ColumnFamilyData::RangesOverlapWithMemtables(
    ParsedInternalKey seek_result;
    if (status.ok()) {
      if (memtable_iter->Valid() &&
-          !ParseInternalKey(memtable_iter->key(), &seek_result)) {
+          ParseInternalKey(memtable_iter->key(), &seek_result) !=
+              Status::OK()) {
        status = Status::Corruption("DB have corrupted keys");
      }
    }
@@ -1000,14 +1140,16 @@ 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_);
@@ -1015,9 +1157,8 @@ Compaction* ColumnFamilyData::CompactRange(
  return result;
 }

-SuperVersion* ColumnFamilyData::GetReferencedSuperVersion(
-    InstrumentedMutex* db_mutex) {
-  SuperVersion* sv = GetThreadLocalSuperVersion(db_mutex);
+SuperVersion* ColumnFamilyData::GetReferencedSuperVersion(DBImpl* db) {
+  SuperVersion* sv = GetThreadLocalSuperVersion(db);
  sv->Ref();
  if (!ReturnThreadLocalSuperVersion(sv)) {
    // This Unref() corresponds to the Ref() in GetThreadLocalSuperVersion()
@@ -1029,8 +1170,7 @@ SuperVersion* ColumnFamilyData::GetReferencedSuperVersion(
  return sv;
 }

-SuperVersion* ColumnFamilyData::GetThreadLocalSuperVersion(
-    InstrumentedMutex* db_mutex) {
+SuperVersion* ColumnFamilyData::GetThreadLocalSuperVersion(DBImpl* db) {
  // The SuperVersion is cached in thread local storage to avoid acquiring
  // mutex when SuperVersion does not change since the last use. When a new
  // SuperVersion is installed, the compaction or flush thread cleans up
@@ -1057,16 +1197,21 @@ SuperVersion* ColumnFamilyData::GetThreadLocalSuperVersion(

    if (sv && sv->Unref()) {
      RecordTick(ioptions_.statistics, NUMBER_SUPERVERSION_CLEANUPS);
-      db_mutex->Lock();
+      db->mutex()->Lock();
      // NOTE: underlying resources held by superversion (sst files) might
      // not be released until the next background job.
      sv->Cleanup();
-      sv_to_delete = sv;
+      if (db->immutable_db_options().avoid_unnecessary_blocking_io) {
+        db->AddSuperVersionsToFreeQueue(sv);
+        db->SchedulePurge();
+      } else {
+        sv_to_delete = sv;
+      }
    } else {
-      db_mutex->Lock();
+      db->mutex()->Lock();
    }
    sv = super_version_->Ref();
-    db_mutex->Unlock();
+    db->mutex()->Unlock();

    delete sv_to_delete;
  }
@@ -1104,7 +1249,7 @@ void ColumnFamilyData::InstallSuperVersion(
  SuperVersion* new_superversion = sv_context->new_superversion.release();
  new_superversion->db_mutex = db_mutex;
  new_superversion->mutable_cf_options = mutable_cf_options;
-  new_superversion->Init(mem_, imm_.current(), current_);
+  new_superversion->Init(this, mem_, imm_.current(), current_);
  SuperVersion* old_superversion = super_version_;
  super_version_ = new_superversion;
  ++super_version_number_;
@@ -1161,6 +1306,11 @@ Status ColumnFamilyData::ValidateOptions(
  if (s.ok() && db_options.allow_concurrent_memtable_write) {
    s = CheckConcurrentWritesSupported(cf_options);
  }
+  if (s.ok() && db_options.unordered_write &&
+      cf_options.max_successive_merges != 0) {
+    s = Status::InvalidArgument(
+        "max_successive_merges > 0 is incompatible with unordered_write");
+  }
  if (s.ok()) {
    s = CheckCFPathsSupported(db_options, cf_options);
  }
@@ -1168,25 +1318,18 @@ Status ColumnFamilyData::ValidateOptions(
    return s;
  }

-  if (cf_options.ttl > 0) {
-    if (db_options.max_open_files != -1) {
-      return Status::NotSupported(
-          "TTL is only supported when files are always "
-          "kept open (set max_open_files = -1). ");
-    }
-    if (cf_options.table_factory->Name() != BlockBasedTableFactory().Name()) {
+  if (cf_options.ttl > 0 && cf_options.ttl != kDefaultTtl) {
+    if (!cf_options.table_factory->IsInstanceOf(
+            TableFactory::kBlockBasedTableName())) {
      return Status::NotSupported(
          "TTL is only supported in Block-Based Table format. ");
    }
  }

-  if (cf_options.periodic_compaction_seconds > 0) {
-    if (db_options.max_open_files != -1) {
-      return Status::NotSupported(
-          "Periodic Compaction is only supported when files are always "
-          "kept open (set max_open_files = -1). ");
-    }
-    if (cf_options.table_factory->Name() != BlockBasedTableFactory().Name()) {
+  if (cf_options.periodic_compaction_seconds > 0 &&
+      cf_options.periodic_compaction_seconds != kDefaultPeriodicCompSecs) {
+    if (!cf_options.table_factory->IsInstanceOf(
+            TableFactory::kBlockBasedTableName())) {
      return Status::NotSupported(
          "Periodic Compaction is only supported in "
          "Block-Based Table format. ");
@@ -1229,28 +1372,41 @@ Env::WriteLifeTimeHint ColumnFamilyData::CalculateSSTWriteHint(int level) {
  // L1: medium, L2: long, ...
  if (level - base_level >= 2) {
    return Env::WLTH_EXTREME;
+  } else if (level < base_level) {
+    // There is no restriction which prevents level passed in to be smaller
+    // than base_level.
+    return Env::WLTH_MEDIUM;
  }
  return static_cast<Env::WriteLifeTimeHint>(level - base_level +
                            static_cast<int>(Env::WLTH_MEDIUM));
 }

-Status ColumnFamilyData::AddDirectories() {
+Status ColumnFamilyData::AddDirectories(
+    std::map<std::string, std::shared_ptr<FSDirectory>>* created_dirs) {
  Status s;
+  assert(created_dirs != nullptr);
  assert(data_dirs_.empty());
  for (auto& p : ioptions_.cf_paths) {
-    std::unique_ptr<Directory> path_directory;
-    s = DBImpl::CreateAndNewDirectory(ioptions_.env, p.path, &path_directory);
-    if (!s.ok()) {
-      return s;
+    auto existing_dir = created_dirs->find(p.path);
+
+    if (existing_dir == created_dirs->end()) {
+      std::unique_ptr<FSDirectory> path_directory;
+      s = DBImpl::CreateAndNewDirectory(ioptions_.fs, p.path, &path_directory);
+      if (!s.ok()) {
+        return s;
+      }
+      assert(path_directory != nullptr);
+      data_dirs_.emplace_back(path_directory.release());
+      (*created_dirs)[p.path] = data_dirs_.back();
+    } else {
+      data_dirs_.emplace_back(existing_dir->second);
    }
-    assert(path_directory != nullptr);
-    data_dirs_.emplace_back(path_directory.release());
  }
  assert(data_dirs_.size() == ioptions_.cf_paths.size());
  return s;
 }

-Directory* ColumnFamilyData::GetDataDir(size_t path_id) const {
+FSDirectory* ColumnFamilyData::GetDataDir(size_t path_id) const {
  if (data_dirs_.empty()) {
    return nullptr;
  }
@@ -1261,23 +1417,26 @@ Directory* ColumnFamilyData::GetDataDir(size_t path_id) const {

 ColumnFamilySet::ColumnFamilySet(const std::string& dbname,
                                 const ImmutableDBOptions* db_options,
-                                 const EnvOptions& env_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,
-          env_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),
-      env_options_(env_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_;
@@ -1288,14 +1447,12 @@ ColumnFamilySet::~ColumnFamilySet() {
    // cfd destructor will delete itself from column_family_data_
    auto cfd = column_family_data_.begin()->second;
    bool last_ref __attribute__((__unused__));
-    last_ref = cfd->Unref();
+    last_ref = cfd->UnrefAndTryDelete();
    assert(last_ref);
-    delete cfd;
  }
  bool dummy_last_ref __attribute__((__unused__));
-  dummy_last_ref = dummy_cfd_->Unref();
+  dummy_last_ref = dummy_cfd_->UnrefAndTryDelete();
  assert(dummy_last_ref);
-  delete dummy_cfd_;
 }

 ColumnFamilyData* ColumnFamilySet::GetDefault() const {
@@ -1345,7 +1502,7 @@ 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_, env_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);
@@ -1413,7 +1570,7 @@ 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;
@@ -1427,4 +1584,4 @@ const Comparator* GetColumnFamilyUserComparator(
  return nullptr;
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -27,7 +27,7 @@
 #include "trace_replay/block_cache_tracer.h"
 #include "util/thread_local.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 class Version;
 class VersionSet;
@@ -198,6 +198,7 @@ class ColumnFamilyHandleInternal : public ColumnFamilyHandleImpl {
 struct SuperVersion {
  // Accessing members of this class is not thread-safe and requires external
  // synchronization (ie db mutex held or on write thread).
+  ColumnFamilyData* cfd;
  MemTable* mem;
  MemTableListVersion* imm;
  Version* current;
@@ -221,8 +222,8 @@ struct SuperVersion {
  // that needs to be deleted in to_delete vector. Unrefing those
  // objects needs to be done in the mutex
  void Cleanup();
-  void Init(MemTable* new_mem, MemTableListVersion* new_imm,
-            Version* new_current);
+  void Init(ColumnFamilyData* new_cfd, MemTable* new_mem,
+            MemTableListVersion* new_imm, Version* new_current);

  // The value of dummy is not actually used. kSVInUse takes its address as a
  // mark in the thread local storage to indicate the SuperVersion is in use
@@ -288,6 +289,11 @@ class ColumnFamilyData {
    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();
+
  // SetDropped() can only be called under following conditions:
  // 1) Holding a DB mutex,
  // 2) from single-threaded write thread, AND
@@ -318,7 +324,7 @@ class ColumnFamilyData {
  }
  FlushReason GetFlushReason() const { return flush_reason_; }
  // thread-safe
-  const EnvOptions* soptions() const;
+  const FileOptions* soptions() const;
  const ImmutableCFOptions* ioptions() const { return &ioptions_; }
  // REQUIRES: DB mutex held
  // This returns the MutableCFOptions used by current SuperVersion
@@ -381,6 +387,7 @@ class ColumnFamilyData {
  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.
@@ -406,6 +413,7 @@ 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,
@@ -430,11 +438,11 @@ class ColumnFamilyData {
  SuperVersion* GetSuperVersion() { return super_version_; }
  // thread-safe
  // Return a already referenced SuperVersion to be used safely.
-  SuperVersion* GetReferencedSuperVersion(InstrumentedMutex* db_mutex);
+  SuperVersion* GetReferencedSuperVersion(DBImpl* db);
  // thread-safe
  // Get SuperVersion stored in thread local storage. If it does not exist,
  // get a reference from a current SuperVersion.
-  SuperVersion* GetThreadLocalSuperVersion(InstrumentedMutex* db_mutex);
+  SuperVersion* GetThreadLocalSuperVersion(DBImpl* db);
  // Try to return SuperVersion back to thread local storage. Retrun true on
  // success and false on failure. It fails when the thread local storage
  // contains anything other than SuperVersion::kSVInUse flag.
@@ -491,22 +499,29 @@ class ColumnFamilyData {

  Env::WriteLifeTimeHint CalculateSSTWriteHint(int level);

-  Status AddDirectories();
+  // 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<FSDirectory>>* created_dirs);

-  Directory* GetDataDir(size_t path_id) const;
+  FSDirectory* GetDataDir(size_t path_id) const;

  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,
                   const ColumnFamilyOptions& options,
                   const ImmutableDBOptions& db_options,
-                   const EnvOptions& env_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_;
@@ -583,7 +598,9 @@ class ColumnFamilyData {
  std::atomic<uint64_t> last_memtable_id_;

  // Directories corresponding to cf_paths.
-  std::vector<std::unique_ptr<Directory>> data_dirs_;
+  std::vector<std::shared_ptr<FSDirectory>> data_dirs_;
+
+  bool db_paths_registered_;
 };

 // ColumnFamilySet has interesting thread-safety requirements
@@ -632,10 +649,11 @@ class ColumnFamilySet {

  ColumnFamilySet(const std::string& dbname,
                  const ImmutableDBOptions* db_options,
-                  const EnvOptions& env_options, Cache* table_cache,
-                  WriteBufferManager* write_buffer_manager,
-                  WriteController* write_controller,
-                  BlockCacheTracer* const block_cache_tracer);
+                  const FileOptions& file_options, Cache* table_cache,
+                  WriteBufferManager* _write_buffer_manager,
+                  WriteController* _write_controller,
+                  BlockCacheTracer* const block_cache_tracer,
+                  const std::shared_ptr<IOTracer>& io_tracer);
  ~ColumnFamilySet();

  ColumnFamilyData* GetDefault() const;
@@ -664,6 +682,10 @@ 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
@@ -690,11 +712,12 @@ class ColumnFamilySet {

  const std::string db_name_;
  const ImmutableDBOptions* const db_options_;
-  const EnvOptions env_options_;
+  const FileOptions file_options_;
  Cache* table_cache_;
  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
@@ -745,4 +768,4 @@ extern uint32_t GetColumnFamilyID(ColumnFamilyHandle* column_family);
 extern const Comparator* GetColumnFamilyUserComparator(
    ColumnFamilyHandle* column_family);

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -8,40 +8,32 @@
 // 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 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 {
 public:
@@ -67,9 +59,9 @@ class ColumnFamilyTestBase : public testing::Test {
 #ifndef ROCKSDB_LITE
    const char* test_env_uri = getenv("TEST_ENV_URI");
    if (test_env_uri) {
-      Status s = ObjectRegistry::NewInstance()->NewSharedObject(test_env_uri,
-                                                                &env_guard_);
-      base_env = env_guard_.get();
+      Env* test_env = nullptr;
+      Status s = Env::LoadEnv(test_env_uri, &test_env, &env_guard_);
+      base_env = test_env;
      EXPECT_OK(s);
      EXPECT_NE(Env::Default(), base_env);
    }
@@ -87,11 +79,16 @@ class ColumnFamilyTestBase : public testing::Test {
    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::SyncPoint::GetInstance()->DisableProcessing();
+    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();
@@ -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();
@@ -595,14 +596,15 @@ TEST_P(ColumnFamilyTest, DontReuseColumnFamilyID) {
 TEST_P(ColumnFamilyTest, CreateCFRaceWithGetAggProperty) {
  Open();

-  rocksdb::SyncPoint::GetInstance()->LoadDependency(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
      {{"DBImpl::WriteOptionsFile:1",
        "ColumnFamilyTest.CreateCFRaceWithGetAggProperty:1"},
       {"ColumnFamilyTest.CreateCFRaceWithGetAggProperty:2",
        "DBImpl::WriteOptionsFile:2"}});
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();

-  rocksdb::port::Thread thread([&] { CreateColumnFamilies({"one"}); });
+  ROCKSDB_NAMESPACE::port::Thread thread(
+      [&] { CreateColumnFamilies({"one"}); });

  TEST_SYNC_POINT("ColumnFamilyTest.CreateCFRaceWithGetAggProperty:1");
  uint64_t pv;
@@ -611,7 +613,7 @@ TEST_P(ColumnFamilyTest, CreateCFRaceWithGetAggProperty) {

  thread.join();

-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
 }
 #endif  // !ROCKSDB_LITE

@@ -655,7 +657,7 @@ TEST_P(FlushEmptyCFTestWithParam, FlushEmptyCFTest) {
  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);
@@ -718,7 +720,7 @@ TEST_P(FlushEmptyCFTestWithParam, FlushEmptyCFTest2) {
  // 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;
@@ -820,7 +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"});
@@ -847,13 +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);
@@ -892,10 +896,10 @@ 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));
    }
  }

@@ -923,7 +927,7 @@ 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);
@@ -988,6 +992,7 @@ TEST_P(ColumnFamilyTest, FlushTest) {
    ASSERT_OK(Put(0, "foofoo", "bar"));

    for (auto* it : iterators) {
+      ASSERT_OK(it->status());
      delete it;
    }
  }
@@ -1085,8 +1090,8 @@ 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);
  fault_env->SetFilesystemActive(false);
@@ -1253,13 +1258,14 @@ TEST_P(ColumnFamilyTest, DifferentWriteBufferSizes) {
 // #endif  // !ROCKSDB_LITE

 class TestComparator : public Comparator {
-  int Compare(const rocksdb::Slice& /*a*/,
-              const rocksdb::Slice& /*b*/) const override {
+  int Compare(const ROCKSDB_NAMESPACE::Slice& /*a*/,
+              const ROCKSDB_NAMESPACE::Slice& /*b*/) const override {
    return 0;
  }
  const char* Name() const override { return "Test"; }
-  void FindShortestSeparator(std::string* /*start*/,
-                             const rocksdb::Slice& /*limit*/) const override {}
+  void FindShortestSeparator(
+      std::string* /*start*/,
+      const ROCKSDB_NAMESPACE::Slice& /*limit*/) const override {}
  void FindShortSuccessor(std::string* /*key*/) const override {}
 };

@@ -1427,11 +1433,11 @@ TEST_P(ColumnFamilyTest, MultipleManualCompactions) {
    AssertFilesPerLevel(ToString(i + 1), 1);
  }
  bool cf_1_1 = true;
-  rocksdb::SyncPoint::GetInstance()->LoadDependency(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
      {{"ColumnFamilyTest::MultiManual:4", "ColumnFamilyTest::MultiManual:1"},
       {"ColumnFamilyTest::MultiManual:2", "ColumnFamilyTest::MultiManual:5"},
       {"ColumnFamilyTest::MultiManual:2", "ColumnFamilyTest::MultiManual:3"}});
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
      "DBImpl::BackgroundCompaction:NonTrivial:AfterRun", [&](void* /*arg*/) {
        if (cf_1_1) {
          TEST_SYNC_POINT("ColumnFamilyTest::MultiManual:4");
@@ -1440,7 +1446,7 @@ TEST_P(ColumnFamilyTest, MultipleManualCompactions) {
        }
      });

-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
  std::vector<port::Thread> threads;
  threads.emplace_back([&] {
    CompactRangeOptions compact_options;
@@ -1483,8 +1489,8 @@ TEST_P(ColumnFamilyTest, MultipleManualCompactions) {
    ASSERT_NE("NOT_FOUND", Get(1, *key_iter));
    key_iter++;
  }
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-  rocksdb::SyncPoint::GetInstance()->ClearAllCallBacks();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
  Close();
 }

@@ -1523,11 +1529,11 @@ TEST_P(ColumnFamilyTest, AutomaticAndManualCompactions) {
      dbfull()->TEST_write_controler().GetCompactionPressureToken();

  bool cf_1_1 = true;
-  rocksdb::SyncPoint::GetInstance()->LoadDependency(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
      {{"ColumnFamilyTest::AutoManual:4", "ColumnFamilyTest::AutoManual:1"},
       {"ColumnFamilyTest::AutoManual:2", "ColumnFamilyTest::AutoManual:5"},
       {"ColumnFamilyTest::AutoManual:2", "ColumnFamilyTest::AutoManual:3"}});
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
      "DBImpl::BackgroundCompaction:NonTrivial:AfterRun", [&](void* /*arg*/) {
        if (cf_1_1) {
          cf_1_1 = false;
@@ -1535,7 +1541,7 @@ TEST_P(ColumnFamilyTest, AutomaticAndManualCompactions) {
          TEST_SYNC_POINT("ColumnFamilyTest::AutoManual:3");
        }
      });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
  // SETUP column family "one" -- universal style
  for (int i = 0; i < one.level0_file_num_compaction_trigger; ++i) {
    PutRandomData(1, 10, 12000, true);
@@ -1553,7 +1559,7 @@ TEST_P(ColumnFamilyTest, AutomaticAndManualCompactions) {
    WaitForFlush(2);
    AssertFilesPerLevel(ToString(i + 1), 2);
  }
-  rocksdb::port::Thread threads([&] {
+  ROCKSDB_NAMESPACE::port::Thread threads([&] {
    CompactRangeOptions compact_options;
    compact_options.exclusive_manual_compaction = false;
    ASSERT_OK(
@@ -1580,8 +1586,8 @@ TEST_P(ColumnFamilyTest, AutomaticAndManualCompactions) {
    ASSERT_NE("NOT_FOUND", Get(1, *key_iter));
    key_iter++;
  }
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-  rocksdb::SyncPoint::GetInstance()->ClearAllCallBacks();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
 }

 TEST_P(ColumnFamilyTest, ManualAndAutomaticCompactions) {
@@ -1627,11 +1633,11 @@ TEST_P(ColumnFamilyTest, ManualAndAutomaticCompactions) {
  }
  bool cf_1_1 = true;
  bool cf_1_2 = true;
-  rocksdb::SyncPoint::GetInstance()->LoadDependency(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
      {{"ColumnFamilyTest::ManualAuto:4", "ColumnFamilyTest::ManualAuto:1"},
       {"ColumnFamilyTest::ManualAuto:5", "ColumnFamilyTest::ManualAuto:2"},
       {"ColumnFamilyTest::ManualAuto:2", "ColumnFamilyTest::ManualAuto:3"}});
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
      "DBImpl::BackgroundCompaction:NonTrivial:AfterRun", [&](void* /*arg*/) {
        if (cf_1_1) {
          TEST_SYNC_POINT("ColumnFamilyTest::ManualAuto:4");
@@ -1643,8 +1649,8 @@ TEST_P(ColumnFamilyTest, ManualAndAutomaticCompactions) {
        }
      });

-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-  rocksdb::port::Thread threads([&] {
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::port::Thread threads([&] {
    CompactRangeOptions compact_options;
    compact_options.exclusive_manual_compaction = false;
    ASSERT_OK(
@@ -1679,8 +1685,8 @@ TEST_P(ColumnFamilyTest, ManualAndAutomaticCompactions) {
    ASSERT_NE("NOT_FOUND", Get(1, *key_iter));
    key_iter++;
  }
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-  rocksdb::SyncPoint::GetInstance()->ClearAllCallBacks();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
 }

 TEST_P(ColumnFamilyTest, SameCFManualManualCompactions) {
@@ -1721,13 +1727,13 @@ TEST_P(ColumnFamilyTest, SameCFManualManualCompactions) {
  }
  bool cf_1_1 = true;
  bool cf_1_2 = true;
-  rocksdb::SyncPoint::GetInstance()->LoadDependency(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
      {{"ColumnFamilyTest::ManualManual:4", "ColumnFamilyTest::ManualManual:2"},
       {"ColumnFamilyTest::ManualManual:4", "ColumnFamilyTest::ManualManual:5"},
       {"ColumnFamilyTest::ManualManual:1", "ColumnFamilyTest::ManualManual:2"},
       {"ColumnFamilyTest::ManualManual:1",
        "ColumnFamilyTest::ManualManual:3"}});
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
      "DBImpl::BackgroundCompaction:NonTrivial:AfterRun", [&](void* /*arg*/) {
        if (cf_1_1) {
          TEST_SYNC_POINT("ColumnFamilyTest::ManualManual:4");
@@ -1739,8 +1745,8 @@ TEST_P(ColumnFamilyTest, SameCFManualManualCompactions) {
        }
      });

-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-  rocksdb::port::Thread threads([&] {
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::port::Thread threads([&] {
    CompactRangeOptions compact_options;
    compact_options.exclusive_manual_compaction = true;
    ASSERT_OK(
@@ -1760,7 +1766,7 @@ TEST_P(ColumnFamilyTest, SameCFManualManualCompactions) {
                        1);
  }

-  rocksdb::port::Thread threads1([&] {
+  ROCKSDB_NAMESPACE::port::Thread threads1([&] {
    CompactRangeOptions compact_options;
    compact_options.exclusive_manual_compaction = false;
    ASSERT_OK(
@@ -1781,8 +1787,8 @@ TEST_P(ColumnFamilyTest, SameCFManualManualCompactions) {
    ASSERT_NE("NOT_FOUND", Get(1, *key_iter));
    key_iter++;
  }
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-  rocksdb::SyncPoint::GetInstance()->ClearAllCallBacks();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
 }

 TEST_P(ColumnFamilyTest, SameCFManualAutomaticCompactions) {
@@ -1823,12 +1829,12 @@ TEST_P(ColumnFamilyTest, SameCFManualAutomaticCompactions) {
  }
  bool cf_1_1 = true;
  bool cf_1_2 = true;
-  rocksdb::SyncPoint::GetInstance()->LoadDependency(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
      {{"ColumnFamilyTest::ManualAuto:4", "ColumnFamilyTest::ManualAuto:2"},
       {"ColumnFamilyTest::ManualAuto:4", "ColumnFamilyTest::ManualAuto:5"},
       {"ColumnFamilyTest::ManualAuto:1", "ColumnFamilyTest::ManualAuto:2"},
       {"ColumnFamilyTest::ManualAuto:1", "ColumnFamilyTest::ManualAuto:3"}});
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
      "DBImpl::BackgroundCompaction:NonTrivial:AfterRun", [&](void* /*arg*/) {
        if (cf_1_1) {
          TEST_SYNC_POINT("ColumnFamilyTest::ManualAuto:4");
@@ -1840,8 +1846,8 @@ TEST_P(ColumnFamilyTest, SameCFManualAutomaticCompactions) {
        }
      });

-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-  rocksdb::port::Thread threads([&] {
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::port::Thread threads([&] {
    CompactRangeOptions compact_options;
    compact_options.exclusive_manual_compaction = false;
    ASSERT_OK(
@@ -1874,8 +1880,8 @@ TEST_P(ColumnFamilyTest, SameCFManualAutomaticCompactions) {
    ASSERT_NE("NOT_FOUND", Get(1, *key_iter));
    key_iter++;
  }
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-  rocksdb::SyncPoint::GetInstance()->ClearAllCallBacks();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
 }

 TEST_P(ColumnFamilyTest, SameCFManualAutomaticCompactionsLevel) {
@@ -1916,14 +1922,14 @@ TEST_P(ColumnFamilyTest, SameCFManualAutomaticCompactionsLevel) {
  }
  bool cf_1_1 = true;
  bool cf_1_2 = true;
-  rocksdb::SyncPoint::GetInstance()->LoadDependency(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
      {{"ColumnFamilyTest::ManualAuto:4", "ColumnFamilyTest::ManualAuto:2"},
       {"ColumnFamilyTest::ManualAuto:4", "ColumnFamilyTest::ManualAuto:5"},
       {"ColumnFamilyTest::ManualAuto:3", "ColumnFamilyTest::ManualAuto:2"},
       {"LevelCompactionPicker::PickCompactionBySize:0",
        "ColumnFamilyTest::ManualAuto:3"},
       {"ColumnFamilyTest::ManualAuto:1", "ColumnFamilyTest::ManualAuto:3"}});
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
      "DBImpl::BackgroundCompaction:NonTrivial:AfterRun", [&](void* /*arg*/) {
        if (cf_1_1) {
          TEST_SYNC_POINT("ColumnFamilyTest::ManualAuto:4");
@@ -1935,8 +1941,8 @@ TEST_P(ColumnFamilyTest, SameCFManualAutomaticCompactionsLevel) {
        }
      });

-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
-  rocksdb::port::Thread threads([&] {
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::port::Thread threads([&] {
    CompactRangeOptions compact_options;
    compact_options.exclusive_manual_compaction = false;
    ASSERT_OK(
@@ -1967,8 +1973,8 @@ TEST_P(ColumnFamilyTest, SameCFManualAutomaticCompactionsLevel) {
    ASSERT_NE("NOT_FOUND", Get(1, *key_iter));
    key_iter++;
  }
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-  rocksdb::SyncPoint::GetInstance()->ClearAllCallBacks();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
 }

 // In this test, we generate enough files to trigger automatic compactions.
@@ -2008,12 +2014,12 @@ TEST_P(ColumnFamilyTest, SameCFAutomaticManualCompactions) {

  bool cf_1_1 = true;
  bool cf_1_2 = true;
-  rocksdb::SyncPoint::GetInstance()->LoadDependency(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
      {{"ColumnFamilyTest::AutoManual:4", "ColumnFamilyTest::AutoManual:2"},
       {"ColumnFamilyTest::AutoManual:4", "ColumnFamilyTest::AutoManual:5"},
       {"CompactionPicker::CompactRange:Conflict",
        "ColumnFamilyTest::AutoManual:3"}});
-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
      "DBImpl::BackgroundCompaction:NonTrivial:AfterRun", [&](void* /*arg*/) {
        if (cf_1_1) {
          TEST_SYNC_POINT("ColumnFamilyTest::AutoManual:4");
@@ -2025,7 +2031,7 @@ TEST_P(ColumnFamilyTest, SameCFAutomaticManualCompactions) {
        }
      });

-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();

  // SETUP column family "one" -- universal style
  for (int i = 0; i < one.level0_file_num_compaction_trigger; ++i) {
@@ -2059,8 +2065,8 @@ TEST_P(ColumnFamilyTest, SameCFAutomaticManualCompactions) {
    ASSERT_NE("NOT_FOUND", Get(1, *key_iter));
    key_iter++;
  }
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-  rocksdb::SyncPoint::GetInstance()->ClearAllCallBacks();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
 }
 #endif  // !ROCKSDB_LITE

@@ -2071,6 +2077,7 @@ std::string IterStatus(Iterator* iter) {
  if (iter->Valid()) {
    result = iter->key().ToString() + "->" + iter->value().ToString();
  } else {
+    EXPECT_OK(iter->status());
    result = "(invalid)";
  }
  return result;
@@ -2325,7 +2332,7 @@ 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.
@@ -2365,6 +2372,44 @@ TEST_P(ColumnFamilyTest, ReadDroppedColumnFamily) {
  }
 }

+TEST_P(ColumnFamilyTest, LiveIteratorWithDroppedColumnFamily) {
+  db_options_.create_missing_column_families = true;
+  db_options_.max_open_files = 20;
+  // delete obsolete files always
+  db_options_.delete_obsolete_files_period_micros = 0;
+  Open({"default", "one", "two"});
+  ColumnFamilyOptions options;
+  options.level0_file_num_compaction_trigger = 100;
+  options.level0_slowdown_writes_trigger = 200;
+  options.level0_stop_writes_trigger = 200;
+  options.write_buffer_size = 100000;  // small write buffer size
+  Reopen({options, options, options});
+
+  // 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]));
+    iterator->SeekToFirst();
+
+    DropColumnFamilies({1});
+
+    // Make sure iterator created can still be used.
+    int count = 0;
+    for (; iterator->Valid(); iterator->Next()) {
+      ASSERT_OK(iterator->status());
+      ++count;
+    }
+    ASSERT_OK(iterator->status());
+    ASSERT_EQ(count, kKeysNum);
+  }
+
+  Reopen();
+  Close();
+  Destroy();
+}
+
 TEST_P(ColumnFamilyTest, FlushAndDropRaceCondition) {
  db_options_.create_missing_column_families = true;
  Open({"default", "one"});
@@ -2376,7 +2421,7 @@ TEST_P(ColumnFamilyTest, FlushAndDropRaceCondition) {
  options.write_buffer_size = 100000;  // small write buffer size
  Reopen({options, options});

-  rocksdb::SyncPoint::GetInstance()->LoadDependency(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
      {{"VersionSet::LogAndApply::ColumnFamilyDrop:0",
        "FlushJob::WriteLevel0Table"},
       {"VersionSet::LogAndApply::ColumnFamilyDrop:1",
@@ -2384,11 +2429,14 @@ TEST_P(ColumnFamilyTest, FlushAndDropRaceCondition) {
       {"FlushJob::InstallResults",
        "VersionSet::LogAndApply::ColumnFamilyDrop:2"}});

-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
  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();

  // 1MB should create ~10 files for each CF
  int kKeysNum = 10000;
@@ -2403,6 +2451,9 @@ 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();

@@ -2463,21 +2514,21 @@ TEST_P(ColumnFamilyTest, CreateAndDropRace) {

  auto main_thread_id = std::this_thread::get_id();

-  rocksdb::SyncPoint::GetInstance()->SetCallBack("PersistRocksDBOptions:start",
-                                                 [&](void* /*arg*/) {
-    auto current_thread_id = std::this_thread::get_id();
-    // If it's the main thread hitting this sync-point, then it
-    // will be blocked until some other thread update the test_stage.
-    if (main_thread_id == current_thread_id) {
-      test_stage = kMainThreadStartPersistingOptionsFile;
-      while (test_stage < kChildThreadFinishDroppingColumnFamily &&
-             !ordered_by_writethread) {
-        Env::Default()->SleepForMicroseconds(100);
-      }
-    }
-  });
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
+      "PersistRocksDBOptions:start", [&](void* /*arg*/) {
+        auto current_thread_id = std::this_thread::get_id();
+        // If it's the main thread hitting this sync-point, then it
+        // will be blocked until some other thread update the test_stage.
+        if (main_thread_id == current_thread_id) {
+          test_stage = kMainThreadStartPersistingOptionsFile;
+          while (test_stage < kChildThreadFinishDroppingColumnFamily &&
+                 !ordered_by_writethread) {
+            Env::Default()->SleepForMicroseconds(100);
+          }
+        }
+      });

-  rocksdb::SyncPoint::GetInstance()->SetCallBack(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
      "WriteThread::EnterUnbatched:Wait", [&](void* /*arg*/) {
        // This means a thread doing DropColumnFamily() is waiting for
        // other thread to finish persisting options.
@@ -2489,12 +2540,12 @@ TEST_P(ColumnFamilyTest, CreateAndDropRace) {
  Open({"default", "one", "two", "three"},
       {cf_opts[0], cf_opts[1], cf_opts[2], cf_opts[3]});

-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();

  // Start a thread that will drop the first column family
  // and its comparator
-  rocksdb::port::Thread drop_cf_thread(DropSingleColumnFamily, this, 1,
-                                       &comparators);
+  ROCKSDB_NAMESPACE::port::Thread drop_cf_thread(DropSingleColumnFamily, this,
+                                                 1, &comparators);

  DropColumnFamilies({2});

@@ -2507,8 +2558,8 @@ TEST_P(ColumnFamilyTest, CreateAndDropRace) {
    }
  }

-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
-  rocksdb::SyncPoint::GetInstance()->ClearAllCallBacks();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
 }
 #endif  // !ROCKSDB_LITE

@@ -2943,15 +2994,18 @@ TEST_P(ColumnFamilyTest, FlushCloseWALFiles) {
  ASSERT_OK(Put(0, "fodor", "mirko"));
  ASSERT_OK(Put(1, "fodor", "mirko"));

-  rocksdb::SyncPoint::GetInstance()->LoadDependency({
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency({
      {"DBImpl::BGWorkFlush:done", "FlushCloseWALFiles:0"},
  });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();

  // Block flush jobs from running
  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;
@@ -2962,7 +3016,7 @@ TEST_P(ColumnFamilyTest, FlushCloseWALFiles) {
  sleeping_task.WakeUp();
  sleeping_task.WaitUntilDone();
  TEST_SYNC_POINT("FlushCloseWALFiles:0");
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
  ASSERT_EQ(1, env.num_open_wal_file_.load());

  Reopen();
@@ -2984,6 +3038,7 @@ 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);

@@ -2997,6 +3052,9 @@ 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;
@@ -3033,6 +3091,7 @@ 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);

@@ -3040,14 +3099,14 @@ TEST_P(ColumnFamilyTest, IteratorCloseWALFile2) {
  ASSERT_OK(Put(0, "fodor", "mirko"));
  ASSERT_OK(Put(1, "fodor", "mirko"));

-  rocksdb::SyncPoint::GetInstance()->LoadDependency({
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency({
      {"ColumnFamilyTest::IteratorCloseWALFile2:0",
       "DBImpl::BGWorkPurge:start"},
      {"ColumnFamilyTest::IteratorCloseWALFile2:2",
       "DBImpl::BackgroundCallFlush:start"},
      {"DBImpl::BGWorkPurge:end", "ColumnFamilyTest::IteratorCloseWALFile2:1"},
  });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();

  WriteOptions wo;
  wo.sync = true;
@@ -3065,7 +3124,7 @@ TEST_P(ColumnFamilyTest, IteratorCloseWALFile2) {
  TEST_SYNC_POINT("ColumnFamilyTest::IteratorCloseWALFile2:2");
  WaitForFlush(1);
  ASSERT_EQ(1, env.num_open_wal_file_.load());
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();

  Reopen();
  ASSERT_EQ("mirko", Get(0, "fodor"));
@@ -3109,14 +3168,14 @@ TEST_P(ColumnFamilyTest, ForwardIteratorCloseWALFile) {
  ASSERT_OK(Put(0, "fodor", "mirko"));
  ASSERT_OK(Put(1, "fodor", "mirko"));

-  rocksdb::SyncPoint::GetInstance()->LoadDependency({
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency({
      {"ColumnFamilyTest::IteratorCloseWALFile2:0",
       "DBImpl::BGWorkPurge:start"},
      {"ColumnFamilyTest::IteratorCloseWALFile2:2",
       "DBImpl::BackgroundCallFlush:start"},
      {"DBImpl::BGWorkPurge:end", "ColumnFamilyTest::IteratorCloseWALFile2:1"},
  });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();

  WriteOptions wo;
  wo.sync = true;
@@ -3127,6 +3186,8 @@ 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());

@@ -3140,7 +3201,7 @@ TEST_P(ColumnFamilyTest, ForwardIteratorCloseWALFile) {
  ASSERT_EQ(1, env.delete_count_.load());

  delete it;
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();

  Reopen();
  ASSERT_EQ("mirko", Get(0, "fodor"));
@@ -3157,29 +3218,29 @@ 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::SyncPoint::GetInstance()->LoadDependency(
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
      {{"DBImpl::SyncWAL:BeforeMarkLogsSynced:1",
        "ColumnFamilyTest::LogSyncConflictFlush:1"},
       {"ColumnFamilyTest::LogSyncConflictFlush:2",
        "DBImpl::SyncWAL:BeforeMarkLogsSynced:2"}});

-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();

-  rocksdb::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");
+  ASSERT_OK(Put(1, "foo", "bar"));
  Flush(1);

  TEST_SYNC_POINT("ColumnFamilyTest::LogSyncConflictFlush:2");

  thread.join();

-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
  Close();
 }
 #endif
@@ -3255,7 +3316,7 @@ TEST_P(ColumnFamilyTest, DISABLED_LogTruncationTest) {
  Close();

  // cleanup
-  env_->DeleteDir(backup_logs);
+  ASSERT_OK(env_->DeleteDir(backup_logs));
 }

 TEST_P(ColumnFamilyTest, DefaultCfPathsTest) {
@@ -3311,15 +3372,17 @@ TEST_P(ColumnFamilyTest, MultipleCFPathsTest) {

  // 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]) {
@@ -3328,7 +3391,7 @@ TEST_P(ColumnFamilyTest, MultipleCFPathsTest) {
  }
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE

 #ifdef ROCKSDB_UNITTESTS_WITH_CUSTOM_OBJECTS_FROM_STATIC_LIBS
 extern "C" {
@@ -3339,7 +3402,7 @@ void RegisterCustomObjects(int /*argc*/, char** /*argv*/) {}
 #endif  // !ROCKSDB_UNITTESTS_WITH_CUSTOM_OBJECTS_FROM_STATIC_LIBS

 int main(int argc, char** argv) {
-  rocksdb::port::InstallStackTraceHandler();
+  ROCKSDB_NAMESPACE::port::InstallStackTraceHandler();
  ::testing::InitGoogleTest(&argc, argv);
  RegisterCustomObjects(argc, argv);
  return RUN_ALL_TESTS();
@@ -16,9 +16,10 @@
 #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 ROCKSDB_NAMESPACE {

 class CompactFilesTest : public testing::Test {
 public:
@@ -81,11 +82,11 @@ TEST_F(CompactFilesTest, L0ConflictsFiles) {
  assert(s.ok());
  assert(db);

-  rocksdb::SyncPoint::GetInstance()->LoadDependency({
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency({
      {"CompactFilesImpl:0", "BackgroundCallCompaction:0"},
      {"BackgroundCallCompaction:1", "CompactFilesImpl:1"},
  });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();

  // create couple files
  // Background compaction starts and waits in BackgroundCallCompaction:0
@@ -95,7 +96,7 @@ TEST_F(CompactFilesTest, L0ConflictsFiles) {
    db->Flush(FlushOptions());
  }

-  rocksdb::ColumnFamilyMetaData meta;
+  ROCKSDB_NAMESPACE::ColumnFamilyMetaData meta;
  db->GetColumnFamilyMetaData(&meta);
  std::string file1;
  for (auto& file : meta.levels[0].files) {
@@ -108,12 +109,12 @@ TEST_F(CompactFilesTest, L0ConflictsFiles) {
      // The background compaction then notices that there is an L0 compaction
      // already in progress and doesn't do an L0 compaction
      // Once the background compaction finishes, the compact files finishes
-      ASSERT_OK(
-          db->CompactFiles(rocksdb::CompactionOptions(), {file1, file2}, 0));
+      ASSERT_OK(db->CompactFiles(ROCKSDB_NAMESPACE::CompactionOptions(),
+                                 {file1, file2}, 0));
      break;
    }
  }
-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
  delete db;
 }

@@ -148,7 +149,7 @@ TEST_F(CompactFilesTest, ObsoleteFiles) {

  auto l0_files = collector->GetFlushedFiles();
  ASSERT_OK(db->CompactFiles(CompactionOptions(), l0_files, 1));
-  reinterpret_cast<DBImpl*>(db)->TEST_WaitForCompact();
+  static_cast_with_check<DBImpl>(db)->TEST_WaitForCompact();

  // verify all compaction input files are deleted
  for (auto fname : l0_files) {
@@ -183,13 +184,13 @@ TEST_F(CompactFilesTest, NotCutOutputOnLevel0) {
  for (int i = 0; i < 500; ++i) {
    db->Put(WriteOptions(), ToString(i), std::string(1000, 'a' + (i % 26)));
  }
-  reinterpret_cast<DBImpl*>(db)->TEST_WaitForFlushMemTable();
+  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)));
  }
-  reinterpret_cast<DBImpl*>(db)->TEST_WaitForFlushMemTable();
+  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));
@@ -224,14 +225,14 @@ TEST_F(CompactFilesTest, CapturingPendingFiles) {
  auto l0_files = collector->GetFlushedFiles();
  EXPECT_EQ(5, l0_files.size());

-  rocksdb::SyncPoint::GetInstance()->LoadDependency({
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency({
      {"CompactFilesImpl:2", "CompactFilesTest.CapturingPendingFiles:0"},
      {"CompactFilesTest.CapturingPendingFiles:1", "CompactFilesImpl:3"},
  });
-  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();

  // Start compacting files.
-  rocksdb::port::Thread compaction_thread(
+  ROCKSDB_NAMESPACE::port::Thread compaction_thread(
      [&] { EXPECT_OK(db->CompactFiles(CompactionOptions(), l0_files, 1)); });

  // In the meantime flush another file.
@@ -242,7 +243,7 @@ TEST_F(CompactFilesTest, CapturingPendingFiles) {

  compaction_thread.join();

-  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();

  delete db;

@@ -296,12 +297,12 @@ TEST_F(CompactFilesTest, CompactionFilterWithGetSv) {
  db->Flush(FlushOptions());

  // Compact all L0 files using CompactFiles
-  rocksdb::ColumnFamilyMetaData meta;
+  ROCKSDB_NAMESPACE::ColumnFamilyMetaData meta;
  db->GetColumnFamilyMetaData(&meta);
  for (auto& file : meta.levels[0].files) {
    std::string fname = file.db_path + "/" + file.name;
    ASSERT_OK(
-        db->CompactFiles(rocksdb::CompactionOptions(), {fname}, 0));
+        db->CompactFiles(ROCKSDB_NAMESPACE::CompactionOptions(), {fname}, 0));
  }


@@ -347,7 +348,7 @@ TEST_F(CompactFilesTest, SentinelCompressionType) {
    compaction_opts.compression = CompressionType::kDisableCompressionOption;
    ASSERT_OK(db->CompactFiles(compaction_opts, l0_files, 1));

-    rocksdb::TablePropertiesCollection all_tables_props;
+    ROCKSDB_NAMESPACE::TablePropertiesCollection all_tables_props;
    ASSERT_OK(db->GetPropertiesOfAllTables(&all_tables_props));
    for (const auto& name_and_table_props : all_tables_props) {
      ASSERT_EQ(CompressionTypeToString(CompressionType::kZlibCompression),
@@ -383,11 +384,11 @@ TEST_F(CompactFilesTest, GetCompactionJobInfo) {
  for (int i = 0; i < 500; ++i) {
    db->Put(WriteOptions(), ToString(i), std::string(1000, 'a' + (i % 26)));
  }
-  reinterpret_cast<DBImpl*>(db)->TEST_WaitForFlushMemTable();
+  static_cast_with_check<DBImpl>(db)->TEST_WaitForFlushMemTable();
  auto l0_files_1 = collector->GetFlushedFiles();
  CompactionOptions co;
  co.compression = CompressionType::kLZ4Compression;
-  CompactionJobInfo compaction_job_info;
+  CompactionJobInfo compaction_job_info{};
  ASSERT_OK(
      db->CompactFiles(co, l0_files_1, 0, -1, nullptr, &compaction_job_info));
  ASSERT_EQ(compaction_job_info.base_input_level, 0);
@@ -402,7 +403,7 @@ TEST_F(CompactFilesTest, GetCompactionJobInfo) {
  delete db;
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE

 int main(int argc, char** argv) {
  ::testing::InitGoogleTest(&argc, argv);
@@ -5,11 +5,13 @@

 #ifndef ROCKSDB_LITE
 #include "db/compacted_db_impl.h"
+
 #include "db/db_impl/db_impl.h"
 #include "db/version_set.h"
 #include "table/get_context.h"
+#include "util/cast_util.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 extern void MarkKeyMayExist(void* arg);
 extern bool SaveValue(void* arg, const ParsedInternalKey& parsed_key,
@@ -37,10 +39,13 @@ Status CompactedDBImpl::Get(const ReadOptions& options, ColumnFamilyHandle*,
                            const Slice& key, PinnableSlice* value) {
  GetContext get_context(user_comparator_, nullptr, nullptr, nullptr,
                         GetContext::kNotFound, key, value, nullptr, nullptr,
-                         true, nullptr, nullptr);
+                         nullptr, true, nullptr, nullptr);
  LookupKey lkey(key, kMaxSequenceNumber);
-  files_.files[FindFile(key)].fd.table_reader->Get(options, lkey.internal_key(),
-                                                   &get_context, nullptr);
+  Status s = files_.files[FindFile(key)].fd.table_reader->Get(
+      options, lkey.internal_key(), &get_context, nullptr);
+  if (!s.ok() && !s.IsNotFound()) {
+    return s;
+  }
  if (get_context.State() == GetContext::kFound) {
    return Status::OK();
  }
@@ -70,12 +75,16 @@ std::vector<Status> CompactedDBImpl::MultiGet(const ReadOptions& options,
      std::string& value = (*values)[idx];
      GetContext get_context(user_comparator_, nullptr, nullptr, nullptr,
                             GetContext::kNotFound, keys[idx], &pinnable_val,
-                             nullptr, nullptr, true, nullptr, nullptr);
+                             nullptr, nullptr, nullptr, true, nullptr, nullptr);
      LookupKey lkey(keys[idx], kMaxSequenceNumber);
-      r->Get(options, lkey.internal_key(), &get_context, nullptr);
-      value.assign(pinnable_val.data(), pinnable_val.size());
-      if (get_context.State() == GetContext::kFound) {
-        statuses[idx] = Status::OK();
+      Status s = r->Get(options, lkey.internal_key(), &get_context, nullptr);
+      if (!s.ok() && !s.IsNotFound()) {
+        statuses[idx] = s;
+      } else {
+        value.assign(pinnable_val.data(), pinnable_val.size());
+        if (get_context.State() == GetContext::kFound) {
+          statuses[idx] = Status::OK();
+        }
      }
    }
    ++idx;
@@ -90,8 +99,8 @@ Status CompactedDBImpl::Init(const Options& options) {
                            ColumnFamilyOptions(options));
  Status s = Recover({cf}, true /* read only */, false, true);
  if (s.ok()) {
-    cfd_ = reinterpret_cast<ColumnFamilyHandleImpl*>(
-              DefaultColumnFamily())->cfd();
+    cfd_ = static_cast_with_check<ColumnFamilyHandleImpl>(DefaultColumnFamily())
+               ->cfd();
    cfd_->InstallSuperVersion(&sv_context, &mutex_);
  }
  mutex_.Unlock();
@@ -147,7 +156,7 @@ Status CompactedDBImpl::Open(const Options& options,
  std::unique_ptr<CompactedDBImpl> db(new CompactedDBImpl(db_options, dbname));
  Status s = db->Init(options);
  if (s.ok()) {
-    db->StartTimedTasks();
+    db->StartPeriodicWorkScheduler();
    ROCKS_LOG_INFO(db->immutable_db_options_.info_log,
                   "Opened the db as fully compacted mode");
    LogFlush(db->immutable_db_options_.info_log);
@@ -156,5 +165,5 @@ Status CompactedDBImpl::Open(const Options& options,
  return s;
 }

-}   // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
 #endif  // ROCKSDB_LITE
@@ -9,7 +9,7 @@
 #include <vector>
 #include "db/db_impl/db_impl.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 class CompactedDBImpl : public DBImpl {
 public:
@@ -109,5 +109,5 @@ class CompactedDBImpl : public DBImpl {
  const Comparator* user_comparator_;
  LevelFilesBrief files_;
 };
-}
+}  // namespace ROCKSDB_NAMESPACE
 #endif  // ROCKSDB_LITE
@@ -7,23 +7,25 @@
 // 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"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 const uint64_t kRangeTombstoneSentinel =
    PackSequenceAndType(kMaxSequenceNumber, kTypeRangeDeletion);

 int sstableKeyCompare(const Comparator* user_cmp, const InternalKey& a,
                      const InternalKey& b) {
-  auto c = user_cmp->Compare(a.user_key(), b.user_key());
+  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,
@@ -243,7 +246,7 @@ Compaction::Compaction(VersionStorageInfo* vstorage,
    compaction_reason_ = CompactionReason::kManualCompaction;
  }
  if (max_subcompactions_ == 0) {
-    max_subcompactions_ = immutable_cf_options_.max_subcompactions;
+    max_subcompactions_ = _mutable_db_options.max_subcompactions;
  }
  if (!bottommost_level_) {
    // Currently we only enable dictionary compression during compaction to the
@@ -275,9 +278,7 @@ Compaction::~Compaction() {
    input_version_->Unref();
  }
  if (cfd_ != nullptr) {
-    if (cfd_->Unref()) {
-      delete cfd_;
-    }
+    cfd_->UnrefAndTryDelete();
  }
 }

@@ -330,6 +331,8 @@ 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_) {
@@ -342,6 +345,13 @@ 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;
@@ -527,6 +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();
 }
@@ -545,21 +570,22 @@ bool Compaction::ShouldFormSubcompactions() const {
  }
 }

-uint64_t Compaction::MaxInputFileCreationTime() const {
-  uint64_t max_creation_time = 0;
-  for (const auto& file : inputs_[0].files) {
-    if (file->fd.table_reader != nullptr &&
-        file->fd.table_reader->GetTableProperties() != nullptr) {
-      uint64_t creation_time =
-          file->fd.table_reader->GetTableProperties()->creation_time;
-      max_creation_time = std::max(max_creation_time, creation_time);
+uint64_t Compaction::MinInputFileOldestAncesterTime() const {
+  uint64_t min_oldest_ancester_time = port::kMaxUint64;
+  for (const auto& level_files : inputs_) {
+    for (const auto& file : level_files.files) {
+      uint64_t oldest_ancester_time = file->TryGetOldestAncesterTime();
+      if (oldest_ancester_time != 0) {
+        min_oldest_ancester_time =
+            std::min(min_oldest_ancester_time, oldest_ancester_time);
+      }
    }
  }
-  return max_creation_time;
+  return min_oldest_ancester_time;
 }

 int Compaction::GetInputBaseLevel() const {
  return input_vstorage_->base_level();
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -11,9 +11,10 @@
 #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 ROCKSDB_NAMESPACE {
 // The file contains class Compaction, as well as some helper functions
 // and data structures used by the class.

@@ -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;

@@ -291,7 +296,7 @@ class Compaction {

  uint32_t max_subcompactions() const { return max_subcompactions_; }

-  uint64_t MaxInputFileCreationTime() const;
+  uint64_t MinInputFileOldestAncesterTime() const;

 private:
  // mark (or clear) all files that are being compacted
@@ -381,4 +386,4 @@ class Compaction {
 // Return sum of sizes of all files in `files`.
 extern uint64_t TotalFileSize(const std::vector<FileMetaData*>& files);

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -5,6 +5,8 @@

 #pragma once

+#include "rocksdb/rocksdb_namespace.h"
+
 struct CompactionIterationStats {
  // Compaction statistics

@@ -4,6 +4,10 @@
 //  (found in the LICENSE.Apache file in the root directory).

 #include "db/compaction/compaction_iterator.h"
+
+#include <cinttypes>
+
+#include "db/blob/blob_file_builder.h"
 #include "db/snapshot_checker.h"
 #include "port/likely.h"
 #include "rocksdb/listener.h"
@@ -26,7 +30,7 @@
  ((seq) <= earliest_snapshot_ && \
   (snapshot_checker_ == nullptr || LIKELY(IsInEarliestSnapshot(seq))))

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 CompactionIterator::CompactionIterator(
    InternalIterator* input, const Comparator* cmp, MergeHelper* merge_helper,
@@ -34,19 +38,22 @@ 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,
-    SnapshotListFetchCallback* snap_list_callback)
+    const std::atomic<int>* manual_compaction_paused,
+    const std::shared_ptr<Logger> info_log)
    : 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_filter, shutting_down, preserve_deletes_seqnum,
-          snap_list_callback) {}
+          manual_compaction_paused, info_log) {}

 CompactionIterator::CompactionIterator(
    InternalIterator* input, const Comparator* cmp, MergeHelper* merge_helper,
@@ -55,11 +62,13 @@ 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,
-    SnapshotListFetchCallback* snap_list_callback)
+    const std::atomic<int>* manual_compaction_paused,
+    const std::shared_ptr<Logger> info_log)
    : input_(input),
      cmp_(cmp),
      merge_helper_(merge_helper),
@@ -70,23 +79,45 @@ CompactionIterator::CompactionIterator(
      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),
      current_user_key_sequence_(0),
      current_user_key_snapshot_(0),
      merge_out_iter_(merge_helper_),
      current_key_committed_(false),
-      snap_list_callback_(snap_list_callback) {
+      info_log_(info_log),
+      allow_data_in_errors_(allow_data_in_errors) {
  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);
  }
-  ProcessSnapshotList();
+  if (snapshots_->size() == 0) {
+    // optimize for fast path if there are no snapshots
+    visible_at_tip_ = true;
+    earliest_snapshot_iter_ = snapshots_->end();
+    earliest_snapshot_ = kMaxSequenceNumber;
+    latest_snapshot_ = 0;
+  } else {
+    visible_at_tip_ = false;
+    earliest_snapshot_iter_ = snapshots_->begin();
+    earliest_snapshot_ = snapshots_->at(0);
+    latest_snapshot_ = snapshots_->back();
+  }
+#ifndef NDEBUG
+  // findEarliestVisibleSnapshot assumes this ordering.
+  for (size_t i = 1; i < snapshots_->size(); ++i) {
+    assert(snapshots_->at(i - 1) < snapshots_->at(i));
+  }
+#endif
  input_->SetPinnedItersMgr(&pinned_iters_mgr_);
  TEST_SYNC_POINT_CALLBACK("CompactionIterator:AfterInit", compaction_.get());
 }
@@ -120,11 +151,15 @@ 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_);
      // 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);
+      assert(s.ok());
+      if (!s.ok()) {
+        ROCKS_LOG_FATAL(info_log_, "Invalid key (%s) in compaction",
+                        key_.ToString(true).c_str());
+      }
+
      // Keep current_key_ in sync.
      current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
      key_ = current_key_.GetInternalKey();
@@ -155,7 +190,7 @@ 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)) {
@@ -198,61 +233,65 @@ void CompactionIterator::InvokeFilterIfNeeded(bool* need_skip,
      value_.clear();
      iter_stats_.num_record_drop_user++;
    } else if (filter == CompactionFilter::Decision::kChangeValue) {
+      if (ikey_.type == kTypeBlobIndex) {
+        // value transfer from blob file to inlined data
+        ikey_.type = kTypeValue;
+        current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
+      }
      value_ = compaction_filter_value_;
    } else if (filter == CompactionFilter::Decision::kRemoveAndSkipUntil) {
      *need_skip = true;
      compaction_filter_skip_until_.ConvertFromUserKey(kMaxSequenceNumber,
                                                       kValueTypeForSeek);
      *skip_until = compaction_filter_skip_until_.Encode();
+    } else if (filter == CompactionFilter::Decision::kChangeBlobIndex) {
+      if (ikey_.type == kTypeValue) {
+        // value transfer from inlined data to blob file
+        ikey_.type = kTypeBlobIndex;
+        current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
+      }
+      value_ = compaction_filter_value_;
+    } else if (filter == CompactionFilter::Decision::kIOError) {
+      status_ =
+          Status::IOError("Failed to access blob during compaction filter");
+      return false;
    }
  }
-}
-
-void CompactionIterator::ProcessSnapshotList() {
-#ifndef NDEBUG
-  // findEarliestVisibleSnapshot assumes this ordering.
-  for (size_t i = 1; i < snapshots_->size(); ++i) {
-    assert(snapshots_->at(i - 1) < snapshots_->at(i));
-  }
-#endif
-  if (snapshots_->size() == 0) {
-    // optimize for fast path if there are no snapshots
-    visible_at_tip_ = true;
-    earliest_snapshot_iter_ = snapshots_->end();
-    earliest_snapshot_ = kMaxSequenceNumber;
-    latest_snapshot_ = 0;
-  } else {
-    visible_at_tip_ = false;
-    earliest_snapshot_iter_ = snapshots_->begin();
-    earliest_snapshot_ = snapshots_->at(0);
-    latest_snapshot_ = snapshots_->back();
-  }
-  released_snapshots_.clear();
+  return true;
 }

 void CompactionIterator::NextFromInput() {
  at_next_ = false;
  valid_ = false;

-  while (!valid_ && input_->Valid() && !IsShuttingDown()) {
+  while (!valid_ && input_->Valid() && !IsPausingManualCompaction() &&
+         !IsShuttingDown()) {
    key_ = input_->key();
    value_ = input_->value();
    iter_stats_.num_input_records++;

-    if (!ParseInternalKey(key_, &ikey_)) {
+    Status pikStatus = ParseInternalKey(key_, &ikey_);
+    if (!pikStatus.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;
+        std::string msg("Corrupted internal key not expected.");
+        if (allow_data_in_errors_) {
+          msg.append(" Corrupt key: " + ikey_.user_key.ToString(/*hex=*/true) +
+                     ". ");
+          msg.append("key type: " + std::to_string(ikey_.type) + ".");
+          msg.append("seq: " + std::to_string(ikey_.sequence) + ".");
+        }
+        status_ = Status::Corruption(msg.c_str());
+        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;
    }
@@ -277,13 +316,6 @@ void CompactionIterator::NextFromInput() {
    // compaction filter). ikey_.user_key is pointing to the copy.
    if (!has_current_user_key_ ||
        !cmp_->Equal(ikey_.user_key, current_user_key_)) {
-      num_keys_++;
-      // Use num_keys_ to reduce the overhead of reading current time
-      if (snap_list_callback_ && snapshots_->size() &&
-          snap_list_callback_->TimeToRefresh(num_keys_)) {
-        snap_list_callback_->Refresh(snapshots_, latest_snapshot_);
-        ProcessSnapshotList();
-      }
      // First occurrence of this user key
      // Copy key for output
      key_ = current_key_.SetInternalKey(key_, &ikey_);
@@ -296,8 +328,9 @@ void CompactionIterator::NextFromInput() {

      // 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
@@ -317,8 +350,9 @@ 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;
        }
      }
    }
@@ -332,8 +366,7 @@ 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
@@ -349,7 +382,18 @@ void CompactionIterator::NextFromInput() {
      // 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) {
+        ROCKS_LOG_FATAL(info_log_,
+                        "Unexpected key type %d for compaction output",
+                        ikey_.type);
+      }
      assert(current_user_key_snapshot_ == last_snapshot);
+      if (current_user_key_snapshot_ != last_snapshot) {
+        ROCKS_LOG_FATAL(info_log_,
+                        "current_user_key_snapshot_ (%" PRIu64
+                        ") != last_snapshot (%" PRIu64 ")",
+                        current_user_key_snapshot_, last_snapshot);
+      }

      value_.clear();
      valid_ = true;
@@ -394,7 +438,8 @@ void CompactionIterator::NextFromInput() {

      // Check whether the next key exists, is not corrupt, and is the same key
      // as the single delete.
-      if (input_->Valid() && ParseInternalKey(input_->key(), &next_ikey) &&
+      if (input_->Valid() &&
+          ParseInternalKey(input_->key(), &next_ikey) == Status::OK() &&
          cmp_->Equal(ikey_.user_key, next_ikey.user_key)) {
        // Check whether the next key belongs to the same snapshot as the
        // SingleDelete.
@@ -491,6 +536,12 @@ void CompactionIterator::NextFromInput() {
      // checking since there has already been a record returned for this key
      // in this snapshot.
      assert(last_sequence >= current_user_key_sequence_);
+      if (last_sequence < current_user_key_sequence_) {
+        ROCKS_LOG_FATAL(info_log_,
+                        "last_sequence (%" PRIu64
+                        ") < current_user_key_sequence_ (%" PRIu64 ")",
+                        last_sequence, current_user_key_sequence_);
+      }

      ++iter_stats_.num_record_drop_hidden;  // (A)
      input_->Next();
@@ -528,11 +579,15 @@ void CompactionIterator::NextFromInput() {
      // 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) &&
+      while (!IsPausingManualCompaction() && !IsShuttingDown() &&
+             input_->Valid() &&
+             (ParseInternalKey(input_->key(), &next_ikey) == Status::OK()) &&
             cmp_->Equal(ikey_.user_key, next_ikey.user_key) &&
             (prev_snapshot == 0 ||
              DEFINITELY_NOT_IN_SNAPSHOT(next_ikey.sequence, prev_snapshot))) {
@@ -540,7 +595,8 @@ void CompactionIterator::NextFromInput() {
      }
      // If you find you still need to output a row with this key, we need to output the
      // delete too
-      if (input_->Valid() && ParseInternalKey(input_->key(), &next_ikey) &&
+      if (input_->Valid() &&
+          (ParseInternalKey(input_->key(), &next_ikey) == Status::OK()) &&
          cmp_->Equal(ikey_.user_key, next_ikey.user_key)) {
        valid_ = true;
        at_next_ = true;
@@ -569,11 +625,14 @@ 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_);
+        pikStatus = ParseInternalKey(key_, &ikey_);
        // 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);
+        assert(pikStatus.ok());
+        if (!pikStatus.ok()) {
+          ROCKS_LOG_FATAL(info_log_, "Invalid key (%s) in compaction",
+                          key_.ToString(true).c_str());
+        }
        // Keep current_key_ in sync.
        current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
        key_ = current_key_.GetInternalKey();
@@ -612,32 +671,82 @@ void CompactionIterator::NextFromInput() {
  if (!valid_ && IsShuttingDown()) {
    status_ = Status::ShutdownInProgress();
  }
+
+  if (IsPausingManualCompaction()) {
+    status_ = Status::Incomplete(Status::SubCode::kManualCompactionPaused);
+  }
 }

 void CompactionIterator::PrepareOutput() {
-  // Zeroing out the sequence number leads to better compression.
-  // If this is the bottommost level (no files in lower levels)
-  // and the earliest snapshot is larger than this seqno
-  // and the userkey differs from the last userkey in compaction
-  // then we can squash the seqno to zero.
-  //
-  // This is safe for TransactionDB write-conflict checking since transactions
-  // only care about sequence number larger than any active snapshots.
-  //
-  // Can we do the same for levels above bottom level as long as
-  // KeyNotExistsBeyondOutputLevel() return true?
-  if ((compaction_ != nullptr && !compaction_->allow_ingest_behind()) &&
-      ikeyNotNeededForIncrementalSnapshot() && bottommost_level_ && valid_ &&
-      IN_EARLIEST_SNAPSHOT(ikey_.sequence) && ikey_.type != kTypeMerge) {
-    assert(ikey_.type != kTypeDeletion && ikey_.type != kTypeSingleDeletion);
-    ikey_.sequence = 0;
-    current_key_.UpdateInternalKey(0, ikey_.type);
+  if (valid_) {
+    if (ikey_.type == kTypeValue) {
+      if (blob_file_builder_) {
+        blob_index_.clear();
+        const Status s =
+            blob_file_builder_->Add(user_key(), value_, &blob_index_);
+
+        if (!s.ok()) {
+          status_ = s;
+          valid_ = false;
+        } else if (!blob_index_.empty()) {
+          value_ = blob_index_;
+          ikey_.type = kTypeBlobIndex;
+          current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
+        }
+      }
+    } else if (ikey_.type == kTypeBlobIndex) {
+      if (compaction_filter_) {
+        const auto blob_decision = compaction_filter_->PrepareBlobOutput(
+            user_key(), value_, &compaction_filter_value_);
+
+        if (blob_decision == CompactionFilter::BlobDecision::kCorruption) {
+          status_ = Status::Corruption(
+              "Corrupted blob reference encountered during GC");
+          valid_ = false;
+        } else if (blob_decision == CompactionFilter::BlobDecision::kIOError) {
+          status_ = Status::IOError("Could not relocate blob during GC");
+          valid_ = false;
+        } else if (blob_decision ==
+                   CompactionFilter::BlobDecision::kChangeValue) {
+          value_ = compaction_filter_value_;
+        }
+      }
+    }
+
+    // Zeroing out the sequence number leads to better compression.
+    // If this is the bottommost level (no files in lower levels)
+    // and the earliest snapshot is larger than this seqno
+    // and the userkey differs from the last userkey in compaction
+    // then we can squash the seqno to zero.
+    //
+    // This is safe for TransactionDB write-conflict checking since transactions
+    // only care about sequence number larger than any active snapshots.
+    //
+    // Can we do the same for levels above bottom level as long as
+    // KeyNotExistsBeyondOutputLevel() return true?
+    if (valid_ && compaction_ != nullptr &&
+        !compaction_->allow_ingest_behind() &&
+        ikeyNotNeededForIncrementalSnapshot() && bottommost_level_ &&
+        IN_EARLIEST_SNAPSHOT(ikey_.sequence) && ikey_.type != kTypeMerge) {
+      assert(ikey_.type != kTypeDeletion && ikey_.type != kTypeSingleDeletion);
+      if (ikey_.type == kTypeDeletion || ikey_.type == kTypeSingleDeletion) {
+        ROCKS_LOG_FATAL(info_log_,
+                        "Unexpected key type %d for seq-zero optimization",
+                        ikey_.type);
+      }
+      ikey_.sequence = 0;
+      current_key_.UpdateInternalKey(0, ikey_.type);
+    }
  }
 }

 inline SequenceNumber CompactionIterator::findEarliestVisibleSnapshot(
    SequenceNumber in, SequenceNumber* prev_snapshot) {
  assert(snapshots_->size());
+  if (snapshots_->size() == 0) {
+    ROCKS_LOG_FATAL(info_log_,
+                    "No snapshot left in findEarliestVisibleSnapshot");
+  }
  auto snapshots_iter = std::lower_bound(
      snapshots_->begin(), snapshots_->end(), in);
  if (snapshots_iter == snapshots_->begin()) {
@@ -645,6 +754,10 @@ inline SequenceNumber CompactionIterator::findEarliestVisibleSnapshot(
  } else {
    *prev_snapshot = *std::prev(snapshots_iter);
    assert(*prev_snapshot < in);
+    if (*prev_snapshot >= in) {
+      ROCKS_LOG_FATAL(info_log_,
+                      "*prev_snapshot >= in in findEarliestVisibleSnapshot");
+    }
  }
  if (snapshot_checker_ == nullptr) {
    return snapshots_iter != snapshots_->end()
@@ -654,6 +767,9 @@ inline SequenceNumber CompactionIterator::findEarliestVisibleSnapshot(
  for (; snapshots_iter != snapshots_->end(); ++snapshots_iter) {
    auto cur = *snapshots_iter;
    assert(in <= cur);
+    if (in > cur) {
+      ROCKS_LOG_FATAL(info_log_, "in > cur in findEarliestVisibleSnapshot");
+    }
    // Skip if cur is in released_snapshots.
    if (has_released_snapshot && released_snapshots_.count(cur) > 0) {
      continue;
@@ -678,9 +794,14 @@ inline bool CompactionIterator::ikeyNotNeededForIncrementalSnapshot() {

 bool CompactionIterator::IsInEarliestSnapshot(SequenceNumber sequence) {
  assert(snapshot_checker_ != nullptr);
-  assert(earliest_snapshot_ == kMaxSequenceNumber ||
-         (earliest_snapshot_iter_ != snapshots_->end() &&
-          *earliest_snapshot_iter_ == earliest_snapshot_));
+  bool pre_condition = (earliest_snapshot_ == kMaxSequenceNumber ||
+                        (earliest_snapshot_iter_ != snapshots_->end() &&
+                         *earliest_snapshot_iter_ == earliest_snapshot_));
+  assert(pre_condition);
+  if (!pre_condition) {
+    ROCKS_LOG_FATAL(info_log_,
+                    "Pre-Condition is not hold in IsInEarliestSnapshot");
+  }
  auto in_snapshot =
      snapshot_checker_->CheckInSnapshot(sequence, earliest_snapshot_);
  while (UNLIKELY(in_snapshot == SnapshotCheckerResult::kSnapshotReleased)) {
@@ -699,7 +820,11 @@ bool CompactionIterator::IsInEarliestSnapshot(SequenceNumber sequence) {
        snapshot_checker_->CheckInSnapshot(sequence, earliest_snapshot_);
  }
  assert(in_snapshot != SnapshotCheckerResult::kSnapshotReleased);
+  if (in_snapshot == SnapshotCheckerResult::kSnapshotReleased) {
+    ROCKS_LOG_FATAL(info_log_,
+                    "Unexpected released snapshot in IsInEarliestSnapshot");
+  }
  return in_snapshot == SnapshotCheckerResult::kInSnapshot;
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -19,55 +19,9 @@
 #include "options/cf_options.h"
 #include "rocksdb/compaction_filter.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

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

 class CompactionIterator {
 public:
@@ -114,11 +68,14 @@ class 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,
                     const Compaction* compaction = nullptr,
                     const CompactionFilter* compaction_filter = nullptr,
                     const std::atomic<bool>* shutting_down = nullptr,
                     const SequenceNumber preserve_deletes_seqnum = 0,
-                     SnapshotListFetchCallback* snap_list_callback = nullptr);
+                     const std::atomic<int>* manual_compaction_paused = nullptr,
+                     const std::shared_ptr<Logger> info_log = nullptr);

  // Constructor with custom CompactionProxy, used for tests.
  CompactionIterator(InternalIterator* input, const Comparator* cmp,
@@ -128,11 +85,14 @@ class 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 = nullptr,
                     const std::atomic<bool>* shutting_down = nullptr,
                     const SequenceNumber preserve_deletes_seqnum = 0,
-                     SnapshotListFetchCallback* snap_list_callback = nullptr);
+                     const std::atomic<int>* manual_compaction_paused = nullptr,
+                     const std::shared_ptr<Logger> info_log = nullptr);

  ~CompactionIterator();

@@ -160,8 +120,6 @@ class CompactionIterator {
 private:
  // Processes the input stream to find the next output
  void NextFromInput();
-  // Process snapshots_ and assign related variables
-  void ProcessSnapshotList();

  // Do last preparations before presenting the output to the callee. At this
  // point this only zeroes out the sequence number if possible for better
@@ -169,7 +127,8 @@ class CompactionIterator {
  void PrepareOutput();

  // 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.
@@ -196,7 +155,7 @@ class CompactionIterator {
  InternalIterator* input_;
  const Comparator* cmp_;
  MergeHelper* merge_helper_;
-  std::vector<SequenceNumber>* snapshots_;
+  const std::vector<SequenceNumber>* snapshots_;
  // List of snapshots released during compaction.
  // findEarliestVisibleSnapshot() find them out from return of
  // snapshot_checker, and make sure they will not be returned as
@@ -210,9 +169,11 @@ class CompactionIterator {
  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<int>* manual_compaction_paused_;
  const SequenceNumber preserve_deletes_seqnum_;
  bool bottommost_level_;
  bool valid_ = false;
@@ -257,6 +218,7 @@ class CompactionIterator {
  // PinnedIteratorsManager used to pin input_ Iterator blocks while reading
  // merge operands and then releasing them after consuming them.
  PinnedIteratorsManager pinned_iters_mgr_;
+  std::string blob_index_;
  std::string compaction_filter_value_;
  InternalKey compaction_filter_skip_until_;
  // "level_ptrs" holds indices that remember which file of an associated
@@ -271,13 +233,19 @@ class CompactionIterator {
  // Used to avoid purging uncommitted values. The application can specify
  // uncommitted values by providing a SnapshotChecker object.
  bool current_key_committed_;
-  SnapshotListFetchCallback* snap_list_callback_;
-  // number of distinct keys processed
-  size_t num_keys_ = 0;
+  std::shared_ptr<Logger> info_log_;
+
+  bool allow_data_in_errors_;

  bool IsShuttingDown() {
    // This is a best-effort facility, so memory_order_relaxed is sufficient.
    return shutting_down_ && shutting_down_->load(std::memory_order_relaxed);
  }
+
+  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) > 0;
+  }
 };
-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -14,7 +14,7 @@
 #include "util/string_util.h"
 #include "utilities/merge_operators.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 // Expects no merging attempts.
 class NoMergingMergeOp : public MergeOperator {
@@ -169,13 +169,15 @@ class FakeCompaction : public CompactionIterator::CompactionProxy {
  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 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
@@ -184,7 +186,7 @@ class TestSnapshotChecker : public SnapshotChecker {
 public:
  explicit TestSnapshotChecker(
      SequenceNumber last_committed_sequence,
-      const std::unordered_map<SequenceNumber, SequenceNumber>& snapshots = {})
+      const std::unordered_map<SequenceNumber, SequenceNumber>& snapshots = {{}})
      : last_committed_sequence_(last_committed_sequence),
        snapshots_(snapshots) {}

@@ -237,6 +239,7 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
    if (filter || bottommost_level) {
      compaction_proxy_ = new FakeCompaction();
      compaction_proxy_->is_bottommost_level = bottommost_level;
+      compaction_proxy_->is_allow_ingest_behind = AllowIngestBehind();
      compaction.reset(compaction_proxy_);
    }
    bool use_snapshot_checker = UseSnapshotChecker() || GetParam();
@@ -255,7 +258,9 @@ class CompactionIteratorTest : public testing::TestWithParam<bool> {
        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 */,
+        false /*allow_data_in_errors*/, std::move(compaction), filter,
+        &shutting_down_));
  }

  void AddSnapshot(SequenceNumber snapshot,
@@ -266,6 +271,8 @@ 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,
@@ -288,6 +295,7 @@ 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());
  }

@@ -312,6 +320,7 @@ 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 +342,7 @@ 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 +359,7 @@ 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 +381,7 @@ 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 +475,7 @@ 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
@@ -494,7 +507,7 @@ TEST_P(CompactionIteratorTest, ShuttingDownInFilter) {
  compaction_proxy_->key_not_exists_beyond_output_level = true;

  std::atomic<bool> seek_done{false};
-  rocksdb::port::Thread compaction_thread([&] {
+  ROCKSDB_NAMESPACE::port::Thread compaction_thread([&] {
    c_iter_->SeekToFirst();
    EXPECT_FALSE(c_iter_->Valid());
    EXPECT_TRUE(c_iter_->status().IsShutdownInProgress());
@@ -531,7 +544,7 @@ TEST_P(CompactionIteratorTest, ShuttingDownInMerge) {
  compaction_proxy_->key_not_exists_beyond_output_level = true;

  std::atomic<bool> seek_done{false};
-  rocksdb::port::Thread compaction_thread([&] {
+  ROCKSDB_NAMESPACE::port::Thread compaction_thread([&] {
    c_iter_->SeekToFirst();
    ASSERT_FALSE(c_iter_->Valid());
    ASSERT_TRUE(c_iter_->status().IsShutdownInProgress());
@@ -656,6 +669,7 @@ 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());
 }

@@ -697,6 +711,18 @@ TEST_P(CompactionIteratorTest, RemoveSingleDeletionAtBottomLevel) {
          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));

@@ -968,7 +994,46 @@ TEST_F(CompactionIteratorWithSnapshotCheckerTest, CompactionFilter_FullMerge) {
      compaction_filter.get());
 }

-}  // namespace rocksdb
+// 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));
+
+}  // namespace ROCKSDB_NAMESPACE

 int main(int argc, char** argv) {
  ::testing::InitGoogleTest(&argc, argv);
@@ -7,8 +7,10 @@
 // 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 <cinttypes>
+#include "db/compaction/compaction_job.h"
+
 #include <algorithm>
+#include <cinttypes>
 #include <functional>
 #include <list>
 #include <memory>
@@ -19,7 +21,6 @@
 #include <vector>

 #include "db/builder.h"
-#include "db/compaction/compaction_job.h"
 #include "db/db_impl/db_impl.h"
 #include "db/db_iter.h"
 #include "db/dbformat.h"
@@ -31,10 +32,13 @@
 #include "db/memtable_list.h"
 #include "db/merge_context.h"
 #include "db/merge_helper.h"
+#include "db/output_validator.h"
 #include "db/range_del_aggregator.h"
 #include "db/version_set.h"
 #include "file/filename.h"
+#include "file/read_write_util.h"
 #include "file/sst_file_manager_impl.h"
+#include "file/writable_file_writer.h"
 #include "logging/log_buffer.h"
 #include "logging/logging.h"
 #include "monitoring/iostats_context_imp.h"
@@ -43,6 +47,7 @@
 #include "port/port.h"
 #include "rocksdb/db.h"
 #include "rocksdb/env.h"
+#include "rocksdb/sst_partitioner.h"
 #include "rocksdb/statistics.h"
 #include "rocksdb/status.h"
 #include "rocksdb/table.h"
@@ -52,13 +57,13 @@
 #include "table/table_builder.h"
 #include "test_util/sync_point.h"
 #include "util/coding.h"
-#include "util/file_reader_writer.h"
+#include "util/hash.h"
 #include "util/mutexlock.h"
 #include "util/random.h"
 #include "util/stop_watch.h"
 #include "util/string_util.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 const char* GetCompactionReasonString(CompactionReason compaction_reason) {
  switch (compaction_reason) {
@@ -115,9 +120,18 @@ struct CompactionJob::SubcompactionState {
  // The return status of this subcompaction
  Status status;

+  // The return IO Status of this subcompaction
+  IOStatus io_status;
+
  // Files produced by this subcompaction
  struct Output {
+    Output(FileMetaData&& _meta, const InternalKeyComparator& _icmp,
+           bool _enable_order_check, bool _enable_hash)
+        : meta(std::move(_meta)),
+          validator(_icmp, _enable_order_check, _enable_hash),
+          finished(false) {}
    FileMetaData meta;
+    OutputValidator validator;
    bool finished;
    std::shared_ptr<const TableProperties> table_properties;
  };
@@ -126,9 +140,10 @@ struct CompactionJob::SubcompactionState {
  std::vector<Output> outputs;
  std::unique_ptr<WritableFileWriter> outfile;
  std::unique_ptr<TableBuilder> builder;
+
  Output* current_output() {
    if (outputs.empty()) {
-      // This subcompaction's outptut could be empty if compaction was aborted
+      // This subcompaction's output could be empty if compaction was aborted
      // before this subcompaction had a chance to generate any output files.
      // When subcompactions are executed sequentially this is more likely and
      // will be particulalry likely for the later subcompactions to be empty.
@@ -139,14 +154,13 @@ struct CompactionJob::SubcompactionState {
    }
  }

-  uint64_t current_output_file_size;
+  uint64_t current_output_file_size = 0;

  // State during the subcompaction
-  uint64_t total_bytes;
-  uint64_t num_input_records;
-  uint64_t num_output_records;
+  uint64_t total_bytes = 0;
+  uint64_t num_output_records = 0;
  CompactionJobStats compaction_job_stats;
-  uint64_t approx_size;
+  uint64_t approx_size = 0;
  // An index that used to speed up ShouldStopBefore().
  size_t grandparent_index = 0;
  // The number of bytes overlapping between the current output and
@@ -155,51 +169,25 @@ struct CompactionJob::SubcompactionState {
  // A flag determine whether the key has been seen in ShouldStopBefore()
  bool seen_key = false;

-  SubcompactionState(Compaction* c, Slice* _start, Slice* _end,
-                     uint64_t size = 0)
-      : compaction(c),
-        start(_start),
-        end(_end),
-        outfile(nullptr),
-        builder(nullptr),
-        current_output_file_size(0),
-        total_bytes(0),
-        num_input_records(0),
-        num_output_records(0),
-        approx_size(size),
-        grandparent_index(0),
-        overlapped_bytes(0),
-        seen_key(false) {
+  SubcompactionState(Compaction* c, Slice* _start, Slice* _end, uint64_t size)
+      : compaction(c), start(_start), end(_end), approx_size(size) {
    assert(compaction != nullptr);
  }

-  SubcompactionState(SubcompactionState&& o) { *this = std::move(o); }
-
-  SubcompactionState& operator=(SubcompactionState&& o) {
-    compaction = std::move(o.compaction);
-    start = std::move(o.start);
-    end = std::move(o.end);
-    status = std::move(o.status);
-    outputs = std::move(o.outputs);
-    outfile = std::move(o.outfile);
-    builder = std::move(o.builder);
-    current_output_file_size = std::move(o.current_output_file_size);
-    total_bytes = std::move(o.total_bytes);
-    num_input_records = std::move(o.num_input_records);
-    num_output_records = std::move(o.num_output_records);
-    compaction_job_stats = std::move(o.compaction_job_stats);
-    approx_size = std::move(o.approx_size);
-    grandparent_index = std::move(o.grandparent_index);
-    overlapped_bytes = std::move(o.overlapped_bytes);
-    seen_key = std::move(o.seen_key);
-    return *this;
+  // Adds the key and value to the builder
+  // If paranoid is true, adds the key-value to the paranoid hash
+  Status AddToBuilder(const Slice& key, const Slice& value) {
+    auto curr = current_output();
+    assert(builder != nullptr);
+    assert(curr != nullptr);
+    Status s = curr->validator.Add(key, value);
+    if (!s.ok()) {
+      return s;
+    }
+    builder->Add(key, value);
+    return Status::OK();
  }

-  // Because member std::unique_ptrs do not have these.
-  SubcompactionState(const SubcompactionState&) = delete;
-
-  SubcompactionState& operator=(const SubcompactionState&) = delete;
-
  // Returns true iff we should stop building the current output
  // before processing "internal_key".
  bool ShouldStopBefore(const Slice& internal_key, uint64_t curr_file_size) {
@@ -244,13 +232,11 @@ struct CompactionJob::CompactionState {
  Status status;

  uint64_t total_bytes;
-  uint64_t num_input_records;
  uint64_t num_output_records;

  explicit CompactionState(Compaction* c)
      : compaction(c),
        total_bytes(0),
-        num_input_records(0),
        num_output_records(0) {}

  size_t NumOutputFiles() {
@@ -288,41 +274,45 @@ struct CompactionJob::CompactionState {
 void CompactionJob::AggregateStatistics() {
  for (SubcompactionState& sc : compact_->sub_compact_states) {
    compact_->total_bytes += sc.total_bytes;
-    compact_->num_input_records += sc.num_input_records;
    compact_->num_output_records += sc.num_output_records;
  }
-  if (compaction_job_stats_) {
-    for (SubcompactionState& sc : compact_->sub_compact_states) {
-      compaction_job_stats_->Add(sc.compaction_job_stats);
-    }
+  for (SubcompactionState& sc : compact_->sub_compact_states) {
+    compaction_job_stats_->Add(sc.compaction_job_stats);
  }
 }

 CompactionJob::CompactionJob(
    int job_id, Compaction* compaction, const ImmutableDBOptions& db_options,
-    const EnvOptions env_options, VersionSet* versions,
+    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,
+    FSDirectory* db_directory, FSDirectory* output_directory, Statistics* stats,
    InstrumentedMutex* db_mutex, ErrorHandler* db_error_handler,
    std::vector<SequenceNumber> existing_snapshots,
    SequenceNumber earliest_write_conflict_snapshot,
    const SnapshotChecker* snapshot_checker, std::shared_ptr<Cache> table_cache,
    EventLogger* event_logger, bool paranoid_file_checks, bool measure_io_stats,
    const std::string& dbname, CompactionJobStats* compaction_job_stats,
-    Env::Priority thread_pri, SnapshotListFetchCallback* snap_list_callback)
+    Env::Priority thread_pri, const std::shared_ptr<IOTracer>& io_tracer,
+    const std::atomic<int>* manual_compaction_paused, const std::string& db_id,
+    const std::string& db_session_id)
    : job_id_(job_id),
      compact_(new CompactionState(compaction)),
      compaction_job_stats_(compaction_job_stats),
      compaction_stats_(compaction->compaction_reason(), 1),
      dbname_(dbname),
+      db_id_(db_id),
+      db_session_id_(db_session_id),
      db_options_(db_options),
-      env_options_(env_options),
+      file_options_(file_options),
      env_(db_options.env),
-      env_options_for_read_(
-          env_->OptimizeForCompactionTableRead(env_options, db_options_)),
+      io_tracer_(io_tracer),
+      fs_(db_options.fs, io_tracer),
+      file_options_for_read_(
+          fs_->OptimizeForCompactionTableRead(file_options, db_options_)),
      versions_(versions),
      shutting_down_(shutting_down),
+      manual_compaction_paused_(manual_compaction_paused),
      preserve_deletes_seqnum_(preserve_deletes_seqnum),
      log_buffer_(log_buffer),
      db_directory_(db_directory),
@@ -331,7 +321,6 @@ CompactionJob::CompactionJob(
      db_mutex_(db_mutex),
      db_error_handler_(db_error_handler),
      existing_snapshots_(std::move(existing_snapshots)),
-      snap_list_callback_(snap_list_callback),
      earliest_write_conflict_snapshot_(earliest_write_conflict_snapshot),
      snapshot_checker_(snapshot_checker),
      table_cache_(std::move(table_cache)),
@@ -341,6 +330,7 @@ CompactionJob::CompactionJob(
      measure_io_stats_(measure_io_stats),
      write_hint_(Env::WLTH_NOT_SET),
      thread_pri_(thread_pri) {
+  assert(compaction_job_stats_ != nullptr);
  assert(log_buffer_ != nullptr);
  const auto* cfd = compact_->compaction->column_family_data();
  ThreadStatusUtil::SetColumnFamily(cfd, cfd->ioptions()->env,
@@ -392,10 +382,9 @@ void CompactionJob::ReportStartedCompaction(Compaction* compaction) {
  // to ensure GetThreadList() can always show them all together.
  ThreadStatusUtil::SetThreadOperation(ThreadStatus::OP_COMPACTION);

-  if (compaction_job_stats_) {
-    compaction_job_stats_->is_manual_compaction =
-        compaction->is_manual_compaction();
-  }
+  compaction_job_stats_->is_manual_compaction =
+      compaction->is_manual_compaction();
+  compaction_job_stats_->is_full_compaction = compaction->is_full_compaction();
 }

 void CompactionJob::Prepare() {
@@ -427,7 +416,11 @@ void CompactionJob::Prepare() {
    RecordInHistogram(stats_, NUM_SUBCOMPACTIONS_SCHEDULED,
                      compact_->sub_compact_states.size());
  } else {
-    compact_->sub_compact_states.emplace_back(c, nullptr, nullptr);
+    constexpr Slice* start = nullptr;
+    constexpr Slice* end = nullptr;
+    constexpr uint64_t size = 0;
+
+    compact_->sub_compact_states.emplace_back(c, start, end, size);
  }
 }

@@ -546,7 +539,7 @@ void CompactionJob::GenSubcompactionBoundaries() {
    // Greedily add ranges to the subcompaction until the sum of the ranges'
    // sizes becomes >= the expected mean size of a subcompaction
    sum = 0;
-    for (size_t i = 0; i < ranges.size() - 1; i++) {
+    for (size_t i = 0; i + 1 < ranges.size(); i++) {
      sum += ranges[i].size;
      if (subcompactions == 1) {
        // If there's only one left to schedule then it goes to the end so no
@@ -610,33 +603,42 @@ Status CompactionJob::Run() {

  // Check if any thread encountered an error during execution
  Status status;
+  IOStatus io_s;
  for (const auto& state : compact_->sub_compact_states) {
    if (!state.status.ok()) {
      status = state.status;
+      io_s = state.io_status;
      break;
    }
  }
-
-  if (status.ok() && output_directory_) {
-    status = output_directory_->Fsync();
+  if (io_status_.ok()) {
+    io_status_ = io_s;
+  }
+  if (status.ok() && output_directory_) {
+    io_s = output_directory_->Fsync(IOOptions(), nullptr);
+  }
+  if (io_status_.ok()) {
+    io_status_ = io_s;
+  }
+  if (status.ok()) {
+    status = io_s;
  }
-
  if (status.ok()) {
    thread_pool.clear();
-    std::vector<const FileMetaData*> files_meta;
+    std::vector<const CompactionJob::SubcompactionState::Output*> files_output;
    for (const auto& state : compact_->sub_compact_states) {
      for (const auto& output : state.outputs) {
-        files_meta.emplace_back(&output.meta);
+        files_output.emplace_back(&output);
      }
    }
    ColumnFamilyData* cfd = compact_->compaction->column_family_data();
    auto prefix_extractor =
        compact_->compaction->mutable_cf_options()->prefix_extractor.get();
-    std::atomic<size_t> next_file_meta_idx(0);
+    std::atomic<size_t> next_file_idx(0);
    auto verify_table = [&](Status& output_status) {
      while (true) {
-        size_t file_idx = next_file_meta_idx.fetch_add(1);
-        if (file_idx >= files_meta.size()) {
+        size_t file_idx = next_file_idx.fetch_add(1);
+        if (file_idx >= files_output.size()) {
          break;
        }
        // Verify that the table is usable
@@ -645,21 +647,40 @@ Status CompactionJob::Run() {
        // No matter whether use_direct_io_for_flush_and_compaction is true,
        // we will regard this verification as user reads since the goal is
        // to cache it here for further user reads
+        ReadOptions read_options;
        InternalIterator* iter = cfd->table_cache()->NewIterator(
-            ReadOptions(), env_options_, cfd->internal_comparator(),
-            *files_meta[file_idx], /*range_del_agg=*/nullptr, prefix_extractor,
+            read_options, file_options_, cfd->internal_comparator(),
+            files_output[file_idx]->meta, /*range_del_agg=*/nullptr,
+            prefix_extractor,
            /*table_reader_ptr=*/nullptr,
            cfd->internal_stats()->GetFileReadHist(
                compact_->compaction->output_level()),
            TableReaderCaller::kCompactionRefill, /*arena=*/nullptr,
            /*skip_filters=*/false, compact_->compaction->output_level(),
+            MaxFileSizeForL0MetaPin(
+                *compact_->compaction->mutable_cf_options()),
            /*smallest_compaction_key=*/nullptr,
-            /*largest_compaction_key=*/nullptr);
+            /*largest_compaction_key=*/nullptr,
+            /*allow_unprepared_value=*/false);
        auto s = iter->status();

        if (s.ok() && paranoid_file_checks_) {
-          for (iter->SeekToFirst(); iter->Valid(); iter->Next()) {}
-          s = iter->status();
+          OutputValidator validator(cfd->internal_comparator(),
+                                    /*_enable_order_check=*/true,
+                                    /*_enable_hash=*/true);
+          for (iter->SeekToFirst(); iter->Valid(); iter->Next()) {
+            s = validator.Add(iter->key(), iter->value());
+            if (!s.ok()) {
+              break;
+            }
+          }
+          if (s.ok()) {
+            s = iter->status();
+          }
+          if (s.ok() &&
+              !validator.CompareValidator(files_output[file_idx]->validator)) {
+            s = Status::Corruption("Paranoid checksums do not match");
+          }
        }

        delete iter;
@@ -720,6 +741,9 @@ Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options) {
  if (status.ok()) {
    status = InstallCompactionResults(mutable_cf_options);
  }
+  if (!versions_->io_status().ok()) {
+    io_status_ = versions_->io_status();
+  }
  VersionStorageInfo::LevelSummaryStorage tmp;
  auto vstorage = cfd->current()->storage_info();
  const auto& stats = compaction_stats_;
@@ -768,24 +792,22 @@ Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options) {
  auto stream = event_logger_->LogToBuffer(log_buffer_);
  stream << "job" << job_id_ << "event"
         << "compaction_finished"
-         << "compaction_time_micros" << compaction_stats_.micros
-         << "compaction_time_cpu_micros" << compaction_stats_.cpu_micros
-         << "output_level" << compact_->compaction->output_level()
-         << "num_output_files" << compact_->NumOutputFiles()
-         << "total_output_size" << compact_->total_bytes << "num_input_records"
-         << compact_->num_input_records << "num_output_records"
+         << "compaction_time_micros" << stats.micros
+         << "compaction_time_cpu_micros" << stats.cpu_micros << "output_level"
+         << compact_->compaction->output_level() << "num_output_files"
+         << compact_->NumOutputFiles() << "total_output_size"
+         << compact_->total_bytes << "num_input_records"
+         << stats.num_input_records << "num_output_records"
         << compact_->num_output_records << "num_subcompactions"
         << compact_->sub_compact_states.size() << "output_compression"
         << CompressionTypeToString(compact_->compaction->output_compression());

-  if (compaction_job_stats_ != nullptr) {
-    stream << "num_single_delete_mismatches"
-           << compaction_job_stats_->num_single_del_mismatch;
-    stream << "num_single_delete_fallthrough"
-           << compaction_job_stats_->num_single_del_fallthru;
-  }
+  stream << "num_single_delete_mismatches"
+         << compaction_job_stats_->num_single_del_mismatch;
+  stream << "num_single_delete_fallthrough"
+         << compaction_job_stats_->num_single_del_fallthru;

-  if (measure_io_stats_ && compaction_job_stats_ != nullptr) {
+  if (measure_io_stats_) {
    stream << "file_write_nanos" << compaction_job_stats_->file_write_nanos;
    stream << "file_range_sync_nanos"
           << compaction_job_stats_->file_range_sync_nanos;
@@ -831,11 +853,20 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {

  CompactionRangeDelAggregator range_del_agg(&cfd->internal_comparator(),
                                             existing_snapshots_);
+  ReadOptions read_options;
+  read_options.verify_checksums = true;
+  read_options.fill_cache = false;
+  // Compaction iterators shouldn't be confined to a single prefix.
+  // Compactions use Seek() for
+  // (a) concurrent compactions,
+  // (b) CompactionFilter::Decision::kRemoveAndSkipUntil.
+  read_options.total_order_seek = true;

  // Although the v2 aggregator is what the level iterator(s) know about,
  // the AddTombstones calls will be propagated down to the v1 aggregator.
-  std::unique_ptr<InternalIterator> input(versions_->MakeInputIterator(
-      sub_compact->compaction, &range_del_agg, env_options_for_read_));
+  std::unique_ptr<InternalIterator> input(
+      versions_->MakeInputIterator(read_options, sub_compact->compaction,
+                                   &range_del_agg, file_options_for_read_));

  AutoThreadOperationStageUpdater stage_updater(
      ThreadStatus::STAGE_COMPACTION_PROCESS_KV);
@@ -866,9 +897,13 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
      false /* internal key corruption is expected */,
      existing_snapshots_.empty() ? 0 : existing_snapshots_.back(),
      snapshot_checker_, compact_->compaction->level(),
-      db_options_.statistics.get(), shutting_down_);
+      db_options_.statistics.get());

  TEST_SYNC_POINT("CompactionJob::Run():Inprogress");
+  TEST_SYNC_POINT_CALLBACK(
+      "CompactionJob::Run():PausingManualCompaction:1",
+      reinterpret_cast<void*>(
+          const_cast<std::atomic<int>*>(manual_compaction_paused_)));

  Slice* start = sub_compact->start;
  Slice* end = sub_compact->end;
@@ -884,11 +919,12 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
  sub_compact->c_iter.reset(new CompactionIterator(
      input.get(), cfd->user_comparator(), &merge, versions_->LastSequence(),
      &existing_snapshots_, earliest_write_conflict_snapshot_,
-      snapshot_checker_, env_, ShouldReportDetailedTime(env_, stats_), false,
-      &range_del_agg, sub_compact->compaction, compaction_filter,
-      shutting_down_, preserve_deletes_seqnum_,
-      // Currently range_del_agg is incompatible with snapshot refresh feature.
-      range_del_agg.IsEmpty() ? snap_list_callback_ : nullptr));
+      snapshot_checker_, env_, ShouldReportDetailedTime(env_, stats_),
+      /*expect_valid_internal_key=*/true, &range_del_agg,
+      /* blob_file_builder */ nullptr, db_options_.allow_data_in_errors,
+      sub_compact->compaction, compaction_filter, shutting_down_,
+      preserve_deletes_seqnum_, manual_compaction_paused_,
+      db_options_.info_log));
  auto c_iter = sub_compact->c_iter.get();
  c_iter->SeekToFirst();
  if (c_iter->Valid() && sub_compact->compaction->output_level() != 0) {
@@ -900,6 +936,12 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
  }
  const auto& c_iter_stats = c_iter->iter_stats();

+  std::unique_ptr<SstPartitioner> partitioner =
+      sub_compact->compaction->output_level() == 0
+          ? nullptr
+          : sub_compact->compaction->CreateSstPartitioner();
+  std::string last_key_for_partitioner;
+
  while (status.ok() && !cfd->IsDropped() && c_iter->Valid()) {
    // Invariant: c_iter.status() is guaranteed to be OK if c_iter->Valid()
    // returns true.
@@ -926,12 +968,16 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
        break;
      }
    }
-    assert(sub_compact->builder != nullptr);
-    assert(sub_compact->current_output() != nullptr);
-    sub_compact->builder->Add(key, value);
-    sub_compact->current_output_file_size = sub_compact->builder->FileSize();
+    status = sub_compact->AddToBuilder(key, value);
+    if (!status.ok()) {
+      break;
+    }
+
+    sub_compact->current_output_file_size =
+        sub_compact->builder->EstimatedFileSize();
+    const ParsedInternalKey& ikey = c_iter->ikey();
    sub_compact->current_output()->meta.UpdateBoundaries(
-        key, c_iter->ikey().sequence);
+        key, value, ikey.sequence, ikey.type);
    sub_compact->num_output_records++;

    // Close output file if it is big enough. Two possibilities determine it's
@@ -943,26 +989,39 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
    // going to be 1.2MB and max_output_file_size = 1MB, prefer to have 0.6MB
    // and 0.6MB instead of 1MB and 0.2MB)
    bool output_file_ended = false;
-    Status input_status;
    if (sub_compact->compaction->output_level() != 0 &&
        sub_compact->current_output_file_size >=
            sub_compact->compaction->max_output_file_size()) {
      // (1) this key terminates the file. For historical reasons, the iterator
      // status before advancing will be given to FinishCompactionOutputFile().
-      input_status = input->status();
      output_file_ended = true;
    }
+    TEST_SYNC_POINT_CALLBACK(
+        "CompactionJob::Run():PausingManualCompaction:2",
+        reinterpret_cast<void*>(
+            const_cast<std::atomic<int>*>(manual_compaction_paused_)));
+    if (partitioner.get()) {
+      last_key_for_partitioner.assign(c_iter->user_key().data_,
+                                      c_iter->user_key().size_);
+    }
    c_iter->Next();
-    if (!output_file_ended && c_iter->Valid() &&
-        sub_compact->compaction->output_level() != 0 &&
-        sub_compact->ShouldStopBefore(c_iter->key(),
-                                      sub_compact->current_output_file_size) &&
-        sub_compact->builder != nullptr) {
-      // (2) this key belongs to the next file. For historical reasons, the
-      // iterator status after advancing will be given to
-      // FinishCompactionOutputFile().
-      input_status = input->status();
-      output_file_ended = true;
+    if (c_iter->status().IsManualCompactionPaused()) {
+      break;
+    }
+    if (!output_file_ended && c_iter->Valid()) {
+      if (((partitioner.get() &&
+            partitioner->ShouldPartition(PartitionerRequest(
+                last_key_for_partitioner, c_iter->user_key(),
+                sub_compact->current_output_file_size)) == kRequired) ||
+           (sub_compact->compaction->output_level() != 0 &&
+            sub_compact->ShouldStopBefore(
+                c_iter->key(), sub_compact->current_output_file_size))) &&
+          sub_compact->builder != nullptr) {
+        // (2) this key belongs to the next file. For historical reasons, the
+        // iterator status after advancing will be given to
+        // FinishCompactionOutputFile().
+        output_file_ended = true;
+      }
    }
    if (output_file_ended) {
      const Slice* next_key = nullptr;
@@ -970,15 +1029,14 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
        next_key = &c_iter->key();
      }
      CompactionIterationStats range_del_out_stats;
-      status =
-          FinishCompactionOutputFile(input_status, sub_compact, &range_del_agg,
-                                     &range_del_out_stats, next_key);
+      status = FinishCompactionOutputFile(input->status(), sub_compact,
+                                          &range_del_agg, &range_del_out_stats,
+                                          next_key);
      RecordDroppedKeys(range_del_out_stats,
                        &sub_compact->compaction_job_stats);
    }
  }

-  sub_compact->num_input_records = c_iter_stats.num_input_records;
  sub_compact->compaction_job_stats.num_input_deletion_records =
      c_iter_stats.num_input_deletion_records;
  sub_compact->compaction_job_stats.num_corrupt_keys =
@@ -1005,6 +1063,11 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
      shutting_down_->load(std::memory_order_relaxed)) {
    status = Status::ShutdownInProgress("Database shutdown");
  }
+  if ((status.ok() || status.IsColumnFamilyDropped()) &&
+      (manual_compaction_paused_ &&
+       manual_compaction_paused_->load(std::memory_order_relaxed) > 0)) {
+    status = Status::Incomplete(Status::SubCode::kManualCompactionPaused);
+  }
  if (status.ok()) {
    status = input->status();
  }
@@ -1024,7 +1087,7 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
    CompactionIterationStats range_del_out_stats;
    Status s = FinishCompactionOutputFile(status, sub_compact, &range_del_agg,
                                          &range_del_out_stats);
-    if (status.ok()) {
+    if (!s.ok() && status.ok()) {
      status = s;
    }
    RecordDroppedKeys(range_del_out_stats, &sub_compact->compaction_job_stats);
@@ -1050,6 +1113,16 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
      SetPerfLevel(prev_perf_level);
    }
  }
+#ifdef ROCKSDB_ASSERT_STATUS_CHECKED
+  if (!status.ok()) {
+    if (sub_compact->c_iter) {
+      sub_compact->c_iter->status().PermitUncheckedError();
+    }
+    if (input) {
+      input->status().PermitUncheckedError();
+    }
+  }
+#endif  // ROCKSDB_ASSERT_STATUS_CHECKED

  sub_compact->c_iter.reset();
  input.reset();
@@ -1110,6 +1183,8 @@ Status CompactionJob::FinishCompactionOutputFile(

  ColumnFamilyData* cfd = sub_compact->compaction->column_family_data();
  const Comparator* ucmp = cfd->user_comparator();
+  std::string file_checksum = kUnknownFileChecksum;
+  std::string file_checksum_func_name = kUnknownFileChecksumFuncName;

  // Check for iterator errors
  Status s = input_status;
@@ -1183,6 +1258,7 @@ Status CompactionJob::FinishCompactionOutputFile(
    } else {
      it->SeekToFirst();
    }
+    TEST_SYNC_POINT("CompactionJob::FinishCompactionOutputFile1");
    for (; it->Valid(); it->Next()) {
      auto tombstone = it->Tombstone();
      if (upper_bound != nullptr) {
@@ -1210,6 +1286,7 @@ Status CompactionJob::FinishCompactionOutputFile(
      auto kv = tombstone.Serialize();
      assert(lower_bound == nullptr ||
             ucmp->Compare(*lower_bound, kv.second) < 0);
+      // Range tombstone is not supported by output validator yet.
      sub_compact->builder->Add(kv.first.Encode(), kv.second);
      InternalKey smallest_candidate = std::move(kv.first);
      if (lower_bound != nullptr &&
@@ -1283,6 +1360,10 @@ Status CompactionJob::FinishCompactionOutputFile(
  } else {
    sub_compact->builder->Abandon();
  }
+  IOStatus io_s = sub_compact->builder->io_status();
+  if (s.ok()) {
+    s = io_s;
+  }
  const uint64_t current_bytes = sub_compact->builder->FileSize();
  if (s.ok()) {
    meta->fd.file_size = current_bytes;
@@ -1293,10 +1374,27 @@ Status CompactionJob::FinishCompactionOutputFile(
  // Finish and check for file errors
  if (s.ok()) {
    StopWatch sw(env_, stats_, COMPACTION_OUTFILE_SYNC_MICROS);
-    s = sub_compact->outfile->Sync(db_options_.use_fsync);
+    io_s = sub_compact->outfile->Sync(db_options_.use_fsync);
+  }
+  if (s.ok() && io_s.ok()) {
+    io_s = sub_compact->outfile->Close();
+  }
+  if (s.ok() && io_s.ok()) {
+    // Add the checksum information to file metadata.
+    meta->file_checksum = sub_compact->outfile->GetFileChecksum();
+    meta->file_checksum_func_name =
+        sub_compact->outfile->GetFileChecksumFuncName();
+    file_checksum = meta->file_checksum;
+    file_checksum_func_name = meta->file_checksum_func_name;
  }
  if (s.ok()) {
-    s = sub_compact->outfile->Close();
+    s = io_s;
+  }
+  if (sub_compact->io_status.ok()) {
+    sub_compact->io_status = io_s;
+    // Since this error is really a copy of the
+    // "normal" status, it does not also need to be checked
+    sub_compact->io_status.PermitUncheckedError();
  }
  sub_compact->outfile.reset();

@@ -1334,24 +1432,31 @@ Status CompactionJob::FinishCompactionOutputFile(
  }
  std::string fname;
  FileDescriptor output_fd;
+  uint64_t oldest_blob_file_number = kInvalidBlobFileNumber;
  if (meta != nullptr) {
    fname =
        TableFileName(sub_compact->compaction->immutable_cf_options()->cf_paths,
                      meta->fd.GetNumber(), meta->fd.GetPathId());
    output_fd = meta->fd;
+    oldest_blob_file_number = meta->oldest_blob_file_number;
  } else {
    fname = "(nil)";
  }
  EventHelpers::LogAndNotifyTableFileCreationFinished(
      event_logger_, cfd->ioptions()->listeners, dbname_, cfd->GetName(), fname,
-      job_id_, output_fd, tp, TableFileCreationReason::kCompaction, s);
+      job_id_, output_fd, oldest_blob_file_number, tp,
+      TableFileCreationReason::kCompaction, s, file_checksum,
+      file_checksum_func_name);

 #ifndef ROCKSDB_LITE
  // Report new file to SstFileManagerImpl
  auto sfm =
      static_cast<SstFileManagerImpl*>(db_options_.sst_file_manager.get());
  if (sfm && meta != nullptr && meta->fd.GetPathId() == 0) {
-    sfm->OnAddFile(fname);
+    Status add_s = sfm->OnAddFile(fname);
+    if (!add_s.ok() && s.ok()) {
+      s = add_s;
+    }
    if (sfm->IsMaxAllowedSpaceReached()) {
      // TODO(ajkr): should we return OK() if max space was reached by the final
      // compaction output file (similarly to how flush works when full)?
@@ -1360,7 +1465,9 @@ Status CompactionJob::FinishCompactionOutputFile(
          "CompactionJob::FinishCompactionOutputFile:"
          "MaxAllowedSpaceReached");
      InstrumentedMutexLock l(db_mutex_);
-      db_error_handler_->SetBGError(s, BackgroundErrorReason::kCompaction);
+      // Should handle return error?
+      db_error_handler_->SetBGError(s, BackgroundErrorReason::kCompaction)
+          .PermitUncheckedError();
    }
  }
 #endif
@@ -1411,10 +1518,22 @@ Status CompactionJob::InstallCompactionResults(

 void CompactionJob::RecordCompactionIOStats() {
  RecordTick(stats_, COMPACT_READ_BYTES, IOSTATS(bytes_read));
+  RecordTick(stats_, COMPACT_WRITE_BYTES, IOSTATS(bytes_written));
+  CompactionReason compaction_reason =
+      compact_->compaction->compaction_reason();
+  if (compaction_reason == CompactionReason::kFilesMarkedForCompaction) {
+    RecordTick(stats_, COMPACT_READ_BYTES_MARKED, IOSTATS(bytes_read));
+    RecordTick(stats_, COMPACT_WRITE_BYTES_MARKED, IOSTATS(bytes_written));
+  } else if (compaction_reason == CompactionReason::kPeriodicCompaction) {
+    RecordTick(stats_, COMPACT_READ_BYTES_PERIODIC, IOSTATS(bytes_read));
+    RecordTick(stats_, COMPACT_WRITE_BYTES_PERIODIC, IOSTATS(bytes_written));
+  } else if (compaction_reason == CompactionReason::kTtl) {
+    RecordTick(stats_, COMPACT_READ_BYTES_TTL, IOSTATS(bytes_read));
+    RecordTick(stats_, COMPACT_WRITE_BYTES_TTL, IOSTATS(bytes_written));
+  }
  ThreadStatusUtil::IncreaseThreadOperationProperty(
      ThreadStatus::COMPACTION_BYTES_READ, IOSTATS(bytes_read));
  IOSTATS_RESET(bytes_read);
-  RecordTick(stats_, COMPACT_WRITE_BYTES, IOSTATS(bytes_written));
  ThreadStatusUtil::IncreaseThreadOperationProperty(
      ThreadStatus::COMPACTION_BYTES_WRITTEN, IOSTATS(bytes_written));
  IOSTATS_RESET(bytes_written);
@@ -1437,13 +1556,22 @@ Status CompactionJob::OpenCompactionOutputFile(
      TableFileCreationReason::kCompaction);
 #endif  // !ROCKSDB_LITE
  // Make the output file
-  std::unique_ptr<WritableFile> writable_file;
+  std::unique_ptr<FSWritableFile> writable_file;
 #ifndef NDEBUG
-  bool syncpoint_arg = env_options_.use_direct_writes;
+  bool syncpoint_arg = file_options_.use_direct_writes;
  TEST_SYNC_POINT_CALLBACK("CompactionJob::OpenCompactionOutputFile",
                           &syncpoint_arg);
 #endif
-  Status s = NewWritableFile(env_, fname, &writable_file, env_options_);
+  Status s;
+  IOStatus io_s =
+      NewWritableFile(fs_.get(), fname, &writable_file, file_options_);
+  s = io_s;
+  if (sub_compact->io_status.ok()) {
+    sub_compact->io_status = io_s;
+    // Since this error is really a copy of the io_s that is checked below as s,
+    // it does not also need to be checked.
+    sub_compact->io_status.PermitUncheckedError();
+  }
  if (!s.ok()) {
    ROCKS_LOG_ERROR(
        db_options_.info_log,
@@ -1454,33 +1582,13 @@ Status CompactionJob::OpenCompactionOutputFile(
    LogFlush(db_options_.info_log);
    EventHelpers::LogAndNotifyTableFileCreationFinished(
        event_logger_, cfd->ioptions()->listeners, dbname_, cfd->GetName(),
-        fname, job_id_, FileDescriptor(), TableProperties(),
-        TableFileCreationReason::kCompaction, s);
+        fname, job_id_, FileDescriptor(), kInvalidBlobFileNumber,
+        TableProperties(), TableFileCreationReason::kCompaction, s,
+        kUnknownFileChecksum, kUnknownFileChecksumFuncName);
    return s;
  }

-  SubcompactionState::Output out;
-  out.meta.fd =
-      FileDescriptor(file_number, sub_compact->compaction->output_path_id(), 0);
-  out.finished = false;
-
-  sub_compact->outputs.push_back(out);
-  writable_file->SetIOPriority(Env::IO_LOW);
-  writable_file->SetWriteLifeTimeHint(write_hint_);
-  writable_file->SetPreallocationBlockSize(static_cast<size_t>(
-      sub_compact->compaction->OutputFilePreallocationSize()));
-  const auto& listeners =
-      sub_compact->compaction->immutable_cf_options()->listeners;
-  sub_compact->outfile.reset(
-      new WritableFileWriter(std::move(writable_file), fname, env_options_,
-                             env_, db_options_.statistics.get(), listeners));
-
-  // If the Column family flag is to only optimize filters for hits,
-  // we can skip creating filters if this is the bottommost_level where
-  // data is going to be found
-  bool skip_filters =
-      cfd->ioptions()->optimize_filters_for_hits && bottommost_level_;
-
+  // Try to figure out the output file's oldest ancester time.
  int64_t temp_current_time = 0;
  auto get_time_status = env_->GetCurrentTime(&temp_current_time);
  // Safe to proceed even if GetCurrentTime fails. So, log and proceed.
@@ -1490,13 +1598,44 @@ Status CompactionJob::OpenCompactionOutputFile(
                   get_time_status.ToString().c_str());
  }
  uint64_t current_time = static_cast<uint64_t>(temp_current_time);
-
-  uint64_t latest_key_time =
-      sub_compact->compaction->MaxInputFileCreationTime();
-  if (latest_key_time == 0) {
-    latest_key_time = current_time;
+  uint64_t oldest_ancester_time =
+      sub_compact->compaction->MinInputFileOldestAncesterTime();
+  if (oldest_ancester_time == port::kMaxUint64) {
+    oldest_ancester_time = current_time;
  }

+  // Initialize a SubcompactionState::Output and add it to sub_compact->outputs
+  {
+    FileMetaData meta;
+    meta.fd = FileDescriptor(file_number,
+                             sub_compact->compaction->output_path_id(), 0);
+    meta.oldest_ancester_time = oldest_ancester_time;
+    meta.file_creation_time = current_time;
+    sub_compact->outputs.emplace_back(
+        std::move(meta), cfd->internal_comparator(),
+        /*enable_order_check=*/
+        sub_compact->compaction->mutable_cf_options()
+            ->check_flush_compaction_key_order,
+        /*enable_hash=*/paranoid_file_checks_);
+  }
+
+  writable_file->SetIOPriority(Env::IOPriority::IO_LOW);
+  writable_file->SetWriteLifeTimeHint(write_hint_);
+  writable_file->SetPreallocationBlockSize(static_cast<size_t>(
+      sub_compact->compaction->OutputFilePreallocationSize()));
+  const auto& listeners =
+      sub_compact->compaction->immutable_cf_options()->listeners;
+  sub_compact->outfile.reset(new WritableFileWriter(
+      std::move(writable_file), fname, file_options_, env_, io_tracer_,
+      db_options_.statistics.get(), listeners,
+      db_options_.file_checksum_gen_factory.get()));
+
+  // If the Column family flag is to only optimize filters for hits,
+  // we can skip creating filters if this is the bottommost_level where
+  // data is going to be found
+  bool skip_filters =
+      cfd->ioptions()->optimize_filters_for_hits && bottommost_level_;
+
  sub_compact->builder.reset(NewTableBuilder(
      *cfd->ioptions(), *(sub_compact->compaction->mutable_cf_options()),
      cfd->internal_comparator(), cfd->int_tbl_prop_collector_factories(),
@@ -1504,9 +1643,10 @@ Status CompactionJob::OpenCompactionOutputFile(
      sub_compact->compaction->output_compression(),
      0 /*sample_for_compression */,
      sub_compact->compaction->output_compression_opts(),
-      sub_compact->compaction->output_level(), skip_filters, latest_key_time,
-      0 /* oldest_key_time */, sub_compact->compaction->max_output_file_size(),
-      current_time));
+      sub_compact->compaction->output_level(), skip_filters,
+      oldest_ancester_time, 0 /* oldest_key_time */,
+      sub_compact->compaction->max_output_file_size(), current_time, db_id_,
+      db_session_id_));
  LogFlush(db_options_.info_log);
  return s;
 }
@@ -1529,6 +1669,9 @@ void CompactionJob::CleanupCompaction() {
        TableCache::Evict(table_cache_.get(), out.meta.fd.GetNumber());
      }
    }
+    // TODO: sub_compact.io_status is not checked like status. Not sure if thats
+    // intentional. So ignoring the io_status as of now.
+    sub_compact.io_status.PermitUncheckedError();
  }
  delete compact_;
  compact_ = nullptr;
@@ -1563,6 +1706,8 @@ void CompactionJob::UpdateCompactionStats() {
    }
  }

+  uint64_t num_output_records = 0;
+
  for (const auto& sub_compact : compact_->sub_compact_states) {
    size_t num_output_files = sub_compact.outputs.size();
    if (sub_compact.builder != nullptr) {
@@ -1572,13 +1717,16 @@ void CompactionJob::UpdateCompactionStats() {
    }
    compaction_stats_.num_output_files += static_cast<int>(num_output_files);

+    num_output_records += sub_compact.num_output_records;
+
    for (const auto& out : sub_compact.outputs) {
      compaction_stats_.bytes_written += out.meta.fd.file_size;
    }
-    if (sub_compact.num_input_records > sub_compact.num_output_records) {
-      compaction_stats_.num_dropped_records +=
-          sub_compact.num_input_records - sub_compact.num_output_records;
-    }
+  }
+
+  if (compaction_stats_.num_input_records > num_output_records) {
+    compaction_stats_.num_dropped_records =
+        compaction_stats_.num_input_records - num_output_records;
  }
 }

@@ -1600,32 +1748,29 @@ void CompactionJob::UpdateCompactionInputStatsHelper(int* num_files,
 void CompactionJob::UpdateCompactionJobStats(
    const InternalStats::CompactionStats& stats) const {
 #ifndef ROCKSDB_LITE
-  if (compaction_job_stats_) {
-    compaction_job_stats_->elapsed_micros = stats.micros;
+  compaction_job_stats_->elapsed_micros = stats.micros;

-    // input information
-    compaction_job_stats_->total_input_bytes =
-        stats.bytes_read_non_output_levels + stats.bytes_read_output_level;
-    compaction_job_stats_->num_input_records = compact_->num_input_records;
-    compaction_job_stats_->num_input_files =
-        stats.num_input_files_in_non_output_levels +
-        stats.num_input_files_in_output_level;
-    compaction_job_stats_->num_input_files_at_output_level =
-        stats.num_input_files_in_output_level;
+  // input information
+  compaction_job_stats_->total_input_bytes =
+      stats.bytes_read_non_output_levels + stats.bytes_read_output_level;
+  compaction_job_stats_->num_input_records = stats.num_input_records;
+  compaction_job_stats_->num_input_files =
+      stats.num_input_files_in_non_output_levels +
+      stats.num_input_files_in_output_level;
+  compaction_job_stats_->num_input_files_at_output_level =
+      stats.num_input_files_in_output_level;

-    // output information
-    compaction_job_stats_->total_output_bytes = stats.bytes_written;
-    compaction_job_stats_->num_output_records = compact_->num_output_records;
-    compaction_job_stats_->num_output_files = stats.num_output_files;
+  // output information
+  compaction_job_stats_->total_output_bytes = stats.bytes_written;
+  compaction_job_stats_->num_output_records = compact_->num_output_records;
+  compaction_job_stats_->num_output_files = stats.num_output_files;

-    if (compact_->NumOutputFiles() > 0U) {
-      CopyPrefix(compact_->SmallestUserKey(),
-                 CompactionJobStats::kMaxPrefixLength,
-                 &compaction_job_stats_->smallest_output_key_prefix);
-      CopyPrefix(compact_->LargestUserKey(),
-                 CompactionJobStats::kMaxPrefixLength,
-                 &compaction_job_stats_->largest_output_key_prefix);
-    }
+  if (compact_->NumOutputFiles() > 0U) {
+    CopyPrefix(compact_->SmallestUserKey(),
+               CompactionJobStats::kMaxPrefixLength,
+               &compaction_job_stats_->smallest_output_key_prefix);
+    CopyPrefix(compact_->LargestUserKey(), CompactionJobStats::kMaxPrefixLength,
+               &compaction_job_stats_->largest_output_key_prefix);
  }
 #else
  (void)stats;
@@ -1667,4 +1812,4 @@ void CompactionJob::LogCompaction() {
  }
 }

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
@@ -44,7 +44,7 @@
 #include "util/stop_watch.h"
 #include "util/thread_local.h"

-namespace rocksdb {
+namespace ROCKSDB_NAMESPACE {

 class Arena;
 class ErrorHandler;
@@ -64,18 +64,21 @@ class CompactionJob {
 public:
  CompactionJob(
      int job_id, Compaction* compaction, const ImmutableDBOptions& db_options,
-      const EnvOptions env_options, VersionSet* versions,
+      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,
+      FSDirectory* db_directory, FSDirectory* output_directory,
+      Statistics* stats, InstrumentedMutex* db_mutex,
+      ErrorHandler* db_error_handler,
      std::vector<SequenceNumber> existing_snapshots,
      SequenceNumber earliest_write_conflict_snapshot,
      const SnapshotChecker* snapshot_checker,
      std::shared_ptr<Cache> table_cache, EventLogger* event_logger,
      bool paranoid_file_checks, bool measure_io_stats,
      const std::string& dbname, CompactionJobStats* compaction_job_stats,
-      Env::Priority thread_pri, SnapshotListFetchCallback* snap_list_callback);
+      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 = "");

  ~CompactionJob();

@@ -97,6 +100,9 @@ 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;

@@ -146,18 +152,23 @@ class CompactionJob {

  // DBImpl state
  const std::string& dbname_;
+  const std::string db_id_;
+  const std::string db_session_id_;
  const ImmutableDBOptions& db_options_;
-  const EnvOptions env_options_;
+  const FileOptions file_options_;

  Env* env_;
+  std::shared_ptr<IOTracer> io_tracer_;
+  FileSystemPtr fs_;
  // env_option optimized for compaction table reads
-  EnvOptions env_options_for_read_;
+  FileOptions file_options_for_read_;
  VersionSet* versions_;
  const std::atomic<bool>* shutting_down_;
+  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_;
  Statistics* stats_;
  InstrumentedMutex* db_mutex_;
  ErrorHandler* db_error_handler_;
@@ -166,7 +177,6 @@ class CompactionJob {
  // entirely within s1 and s2, then the earlier version of k1 can be safely
  // deleted because that version is not visible in any snapshot.
  std::vector<SequenceNumber> existing_snapshots_;
-  SnapshotListFetchCallback* snap_list_callback_;

  // This is the earliest snapshot that could be used for write-conflict
  // checking by a transaction.  For any user-key newer than this snapshot, we
@@ -189,6 +199,7 @@ class CompactionJob {
  std::vector<uint64_t> sizes_;
  Env::WriteLifeTimeHint write_hint_;
  Env::Priority thread_pri_;
+  IOStatus io_status_;
 };

-}  // namespace rocksdb
+}  // namespace ROCKSDB_NAMESPACE
--- a/Show More
+++ b/Show More