Fix the ZStd checksum (#12005 )

Summary: Somehow we had the wrong checksum when validating the ZStd 1.5.5 download for RocksJava in the previous Pull Request - https://github.com/facebook/rocksdb/pull/9304. This PR fixes that. Pull Request resolved: https://github.com/facebook/rocksdb/pull/12005 Reviewed By: jaykorean Differential Revision: D50840338 Pulled By: cbi42 fbshipit-source-id: 8a92779d3bef013d812eecb89aaaf33fc73991ec
Update HISTORY.md and version for 8.7.3
2026-11-06 11:02:36 -05:00 · 2023-10-31 13:22:47 -07:00 · 2023-10-30 14:07:10 -07:00 · 2023-10-30 11:43:56 -07:00 · 2023-10-25 21:20:16 -07:00 · 2023-10-25 20:48:44 -07:00
836 changed files with 16338 additions and 38053 deletions
@@ -54,7 +54,6 @@ commands:
          command: |
            echo "Installing CMake..."
            choco install cmake --installargs 'ADD_CMAKE_TO_PATH=System' -y
-            choco install liberica8jdk -y
            mkdir $Env:THIRDPARTY_HOME
            cd $Env:THIRDPARTY_HOME
            echo "Building Snappy dependency..."
@@ -68,10 +67,9 @@ commands:
      - run:
          name: "Build RocksDB"
          command: |
-            $env:Path = $env:JAVA_HOME + ";" + $env:Path
            mkdir build
            cd build
-            & $Env:CMAKE_BIN -G "$Env:CMAKE_GENERATOR" -DCMAKE_BUILD_TYPE=Debug -DOPTDBG=1 -DPORTABLE="$Env:CMAKE_PORTABLE" -DSNAPPY=1 -DJNI=1 ..
+            & $Env:CMAKE_BIN -G "$Env:CMAKE_GENERATOR" -DCMAKE_BUILD_TYPE=Debug -DOPTDBG=1 -DPORTABLE=1 -DSNAPPY=1 -DJNI=1 ..
            cd ..
            echo "Building with VS version: $Env:CMAKE_GENERATOR"
            msbuild.exe build/rocksdb.sln -maxCpuCount -property:Configuration=Debug -property:Platform=x64
@@ -80,11 +78,6 @@ commands:
          shell: powershell.exe
          command: |
            build_tools\run_ci_db_test.ps1 -SuiteRun arena_test,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
-      - run:
-          name: "Test RocksJava"
-          command: |
-            cd build\java
-            & $Env:CTEST_BIN -C Debug -j 16
  pre-steps-macos:
      steps:
        - pre-steps
@@ -112,15 +105,6 @@ commands:
          path: /tmp/core_dumps
          when: on_fail

-  post-pmd-steps:
-    steps:
-      - store_artifacts:
-          path: /home/circleci/project/java/target/pmd.xml
-          when: on_fail
-      - store_artifacts:
-          path: /home/circleci/project/java/target/site
-          when: on_fail
-
  upgrade-cmake:
    steps:
      - run:
@@ -143,16 +127,6 @@ commands:
          command: |
            HOMEBREW_NO_AUTO_UPDATE=1 brew install gflags

-  install-maven:
-    steps:
-      - run:
-          name: Install Maven
-          command: |
-            wget --no-check-certificate https://dlcdn.apache.org/maven/maven-3/3.9.6/binaries/apache-maven-3.9.6-bin.tar.gz
-            tar zxf apache-maven-3.9.6-bin.tar.gz
-            echo "export M2_HOME=$(pwd)/apache-maven-3.9.6" >> $BASH_ENV
-            echo 'export PATH=$M2_HOME/bin:$PATH' >> $BASH_ENV
-
  setup-folly:
    steps:
      - run:
@@ -234,7 +208,6 @@ executors:
      - image: zjay437/rocksdb:0.6
  linux-java-docker:
    docker:
-      # This is the Docker Image used for building RocksJava releases, see: https://github.com/evolvedbinary/docker-rocksjava
      - image: evolvedbinary/rocksjava:centos6_x64-be

 jobs:
@@ -248,7 +221,7 @@ jobs:
      - increase-max-open-files-on-macos
      - install-gflags-on-macos
      - pre-steps-macos
-      - run: ulimit -S -n `ulimit -H -n` && make V=1 J=16 -j16 all
+      - run: ulimit -S -n `ulimit -H -n` && OPT=-DCIRCLECI make V=1 J=16 -j16 all
      - post-steps

  build-macos-cmake:
@@ -547,7 +520,7 @@ jobs:
    resource_class: large
    steps:
      - pre-steps
-      - run: ulimit -S -n `ulimit -H -n` && make V=1 -j8 CRASH_TEST_EXT_ARGS='--duration=960 --max_key=2500000' blackbox_crash_test_with_atomic_flush
+      - run: ulimit -S -n `ulimit -H -n` && make V=1 -j8 CRASH_TEST_EXT_ARGS='--duration=960 --max_key=2500000 --use_io_uring=0' blackbox_crash_test_with_atomic_flush
      - post-steps

  build-linux-crashtest-tiered-storage-bb:
@@ -557,7 +530,7 @@ jobs:
      - pre-steps
      - run:
          name: "run crashtest"
-          command: ulimit -S -n `ulimit -H -n` && make V=1 -j32 CRASH_TEST_EXT_ARGS='--duration=10800' blackbox_crash_test_with_tiered_storage
+          command: ulimit -S -n `ulimit -H -n` && make V=1 -j32 CRASH_TEST_EXT_ARGS='--duration=10800 --use_io_uring=0' blackbox_crash_test_with_tiered_storage
          no_output_timeout: 100m
      - post-steps

@@ -568,28 +541,10 @@ jobs:
      - pre-steps
      - run:
          name: "run crashtest"
-          command: ulimit -S -n `ulimit -H -n` && make V=1 -j32 CRASH_TEST_EXT_ARGS='--duration=10800' whitebox_crash_test_with_tiered_storage
+          command: ulimit -S -n `ulimit -H -n` && make V=1 -j32 CRASH_TEST_EXT_ARGS='--duration=10800 --use_io_uring=0' whitebox_crash_test_with_tiered_storage
          no_output_timeout: 100m
      - post-steps

-  build-windows-vs2022-avx2:
-    executor:
-      name: win/server-2022
-      size: 2xlarge
-    environment:
-      THIRDPARTY_HOME: C:/Users/circleci/thirdparty
-      CMAKE_HOME: C:/Program Files/CMake
-      CMAKE_BIN: C:/Program Files/CMake/bin/cmake.exe
-      CTEST_BIN: C:/Program Files/CMake/bin/ctest.exe
-      JAVA_HOME: C:/Program Files/BellSoft/LibericaJDK-8
-      SNAPPY_HOME: C:/Users/circleci/thirdparty/snappy-1.1.8
-      SNAPPY_INCLUDE: C:/Users/circleci/thirdparty/snappy-1.1.8;C:/Users/circleci/thirdparty/snappy-1.1.8/build
-      SNAPPY_LIB_DEBUG: C:/Users/circleci/thirdparty/snappy-1.1.8/build/Debug/snappy.lib
-      CMAKE_GENERATOR: Visual Studio 17 2022
-      CMAKE_PORTABLE: AVX2
-    steps:
-      - windows-build-steps
-
  build-windows-vs2022:
    executor:
      name: win/server-2022
@@ -598,13 +553,10 @@ jobs:
      THIRDPARTY_HOME: C:/Users/circleci/thirdparty
      CMAKE_HOME: C:/Program Files/CMake
      CMAKE_BIN: C:/Program Files/CMake/bin/cmake.exe
-      CTEST_BIN: C:/Program Files/CMake/bin/ctest.exe
-      JAVA_HOME: C:/Program Files/BellSoft/LibericaJDK-8
      SNAPPY_HOME: C:/Users/circleci/thirdparty/snappy-1.1.8
      SNAPPY_INCLUDE: C:/Users/circleci/thirdparty/snappy-1.1.8;C:/Users/circleci/thirdparty/snappy-1.1.8/build
      SNAPPY_LIB_DEBUG: C:/Users/circleci/thirdparty/snappy-1.1.8/build/Debug/snappy.lib
      CMAKE_GENERATOR: Visual Studio 17 2022
-      CMAKE_PORTABLE: 1
    steps:
      - windows-build-steps

@@ -616,18 +568,15 @@ jobs:
      THIRDPARTY_HOME: C:/Users/circleci/thirdparty
      CMAKE_HOME: C:/Program Files/CMake
      CMAKE_BIN: C:/Program Files/CMake/bin/cmake.exe
-      CTEST_BIN: C:/Program Files/CMake/bin/ctest.exe
-      JAVA_HOME: C:/Program Files/BellSoft/LibericaJDK-8
      SNAPPY_HOME: C:/Users/circleci/thirdparty/snappy-1.1.8
      SNAPPY_INCLUDE: C:/Users/circleci/thirdparty/snappy-1.1.8;C:/Users/circleci/thirdparty/snappy-1.1.8/build
      SNAPPY_LIB_DEBUG: C:/Users/circleci/thirdparty/snappy-1.1.8/build/Debug/snappy.lib
      CMAKE_GENERATOR: Visual Studio 16 2019
-      CMAKE_PORTABLE: 1
    steps:
      - windows-build-steps

  build-linux-java:
-    executor: linux-java-docker
+    executor: linux-docker
    resource_class: large
    steps:
      - pre-steps
@@ -640,27 +589,9 @@ jobs:
            which javac && javac -version
      - run:
          name: "Test RocksDBJava"
-          command: scl enable devtoolset-7 'make V=1 J=8 -j8 jtest'
+          command: make V=1 J=8 -j8 jtest
      - post-steps

-  build-linux-java-pmd:
-    executor: linux-java-docker
-    resource_class: large
-    steps:
-      - pre-steps
-      - run:
-          name: "Set Java Environment"
-          command: |
-            echo "JAVA_HOME=${JAVA_HOME}"
-            echo 'export PATH=$JAVA_HOME/bin:$PATH' >> $BASH_ENV
-            which java && java -version
-            which javac && javac -version
-      - install-maven
-      - run:
-          name: "PMD RocksDBJava"
-          command: scl enable devtoolset-7 'make V=1 J=8 -j8 jpmd'
-      - post-pmd-steps
-
  build-linux-java-static:
    executor: linux-java-docker
    resource_class: large
@@ -878,11 +809,66 @@ jobs:
      - perform-benchmarks
      - post-benchmarks

-workflows: # Only jobs that haven't been successfully migrated to GitHub Actions
+workflows:
  version: 2
+  jobs-linux-run-tests:
+    jobs:
+      - build-linux
+      - build-linux-cmake-with-folly
+      - build-linux-cmake-with-folly-lite-no-test
+      - build-linux-gcc-7-with-folly
+      - build-linux-gcc-7-with-folly-lite-no-test
+      - build-linux-cmake-with-folly-coroutines
+      - build-linux-cmake-with-benchmark
+      - build-linux-encrypted_env-no_compression
+  jobs-linux-run-tests-san:
+    jobs:
+      - build-linux-clang10-asan
+      - build-linux-clang10-ubsan
+      - build-linux-clang10-mini-tsan
+      - build-linux-static_lib-alt_namespace-status_checked
+  jobs-linux-no-test-run:
+    jobs:
+      - build-linux-release
+      - build-linux-release-rtti
+      - build-examples
+      - build-fuzzers
+      - build-linux-clang-no_test_run
+      - build-linux-clang-13-no_test_run
+      - build-linux-gcc-8-no_test_run
+      - build-linux-gcc-10-cxx20-no_test_run
+      - build-linux-gcc-11-no_test_run
+      - build-linux-arm-cmake-no_test_run
+  jobs-linux-other-checks:
+    jobs:
+      - build-linux-clang10-clang-analyze
+      - build-linux-unity-and-headers
+      - build-linux-mini-crashtest
+  jobs-windows:
+    jobs:
+      - build-windows-vs2022
+      - build-windows-vs2019
+      - build-cmake-mingw
+  jobs-java:
+    jobs:
+      - build-linux-java
+      - build-linux-java-static
+      - build-macos-java
+      - build-macos-java-static
+      - build-macos-java-static-universal
+  jobs-macos:
+    jobs:
+      - build-macos
+      - build-macos-cmake:
+          run_even_tests: true
+      - build-macos-cmake:
+          run_even_tests: false
  jobs-linux-arm:
    jobs:
      - build-linux-arm
+  build-fuzzers:
+    jobs:
+      - build-fuzzers
  benchmark-linux:
    triggers:
      - schedule:
@@ -902,4 +888,9 @@ workflows: # Only jobs that haven't been successfully migrated to GitHub Actions
              only:
                - main
    jobs:
+      - build-format-compatible
      - build-linux-arm-test-full
+      - build-linux-run-microbench
+      - build-linux-non-shm
+      - build-linux-clang-13-asan-ubsan-with-folly
+      - build-linux-valgrind
@@ -1,7 +0,0 @@
-name: build-folly
-runs:
-  using: composite
-  steps:
-  - name: Build folly and dependencies
-    run: make build_folly
-    shell: bash
@@ -1,8 +0,0 @@
-name: build-for-benchmarks
-runs:
-  using: composite
-  steps:
-  - uses: "./.github/actions/pre-steps"
-  - name: Linux build for benchmarks
-    run: make V=1 J=8 -j8 release
-    shell: bash
@@ -1,10 +0,0 @@
-name: increase-max-open-files-on-macos
-runs:
-  using: composite
-  steps:
-  - name: Increase max open files
-    run: |-
-      sudo sysctl -w kern.maxfiles=1048576
-      sudo sysctl -w kern.maxfilesperproc=1048576
-      sudo launchctl limit maxfiles 1048576
-    shell: bash
@@ -1,7 +0,0 @@
-name: install-gflags-on-macos
-runs:
-  using: composite
-  steps:
-  - name: Install gflags on macos
-    run: HOMEBREW_NO_AUTO_UPDATE=1 brew install gflags
-    shell: bash
@@ -1,7 +0,0 @@
-name: install-gflags
-runs:
-  using: composite
-  steps:
-  - name: Install gflags
-    run: sudo apt-get update -y && sudo apt-get install -y libgflags-dev
-    shell: bash
@@ -1,9 +0,0 @@
-name: install-jdk8-on-macos
-runs:
-  using: composite
-  steps:
-  - name: Install JDK 8 on macos
-    run: |-
-      HOMEBREW_NO_AUTO_UPDATE=1 brew tap bell-sw/liberica
-      HOMEBREW_NO_AUTO_UPDATE=1 brew install --cask liberica-jdk8
-    shell: bash
@@ -1,11 +0,0 @@
-name: install-maven
-runs:
-  using: composite
-  steps:
-  - name: Install Maven
-    run: |
-      wget --no-check-certificate https://dlcdn.apache.org/maven/maven-3/3.9.6/binaries/apache-maven-3.9.6-bin.tar.gz
-      tar zxf apache-maven-3.9.6-bin.tar.gz
-      echo "export M2_HOME=$(pwd)/apache-maven-3.9.6" >> $GITHUB_ENV
-      echo "$(pwd)/apache-maven-3.9.6/bin" >> $GITHUB_PATH
-    shell: bash
@@ -1,22 +0,0 @@
-name: perform-benchmarks
-runs:
-  using: composite
-  steps:
-  - name: Test low-variance benchmarks
-    run: "./tools/benchmark_ci.py --db_dir ${{ runner.temp }}/rocksdb-benchmark-datadir --output_dir ${{ runner.temp }}/benchmark-results --num_keys 20000000"
-    env:
-      LD_LIBRARY_PATH: "/usr/local/lib"
-      DURATION_RO: 300
-      DURATION_RW: 500
-      NUM_THREADS: 1
-      MAX_BACKGROUND_JOBS: 4
-      CI_TESTS_ONLY: 'true'
-      WRITE_BUFFER_SIZE_MB: 16
-      TARGET_FILE_SIZE_BASE_MB: 16
-      MAX_BYTES_FOR_LEVEL_BASE_MB: 64
-      COMPRESSION_TYPE: none
-      CACHE_INDEX_AND_FILTER_BLOCKS: 1
-      MIN_LEVEL_TO_COMPRESS: 3
-      CACHE_SIZE_MB: 10240
-      MB_WRITE_PER_SEC: 2
-    shell: bash
@@ -1,17 +0,0 @@
-name: post-benchmarks
-runs:
-  using: composite
-  steps:
-  - name: Upload Benchmark Results artifact
-    uses: actions/upload-artifact@v4.0.0
-    with:
-      name: benchmark-results
-      path: "${{ runner.temp }}/benchmark-results/**"
-      if-no-files-found: error
-  - name: Send benchmark report to visualisation
-    run: |-
-      set +e
-      set +o pipefail
-      ./build_tools/benchmark_log_tool.py --tsvfile ${{ runner.temp }}/benchmark-results/report.tsv --esdocument https://search-rocksdb-bench-k2izhptfeap2hjfxteolsgsynm.us-west-2.es.amazonaws.com/bench_test3_rix/_doc
-      true
-    shell: bash
@@ -1,38 +0,0 @@
-name: post-steps
-description: Steps that are taken after a RocksDB job
-inputs:
-  artifact-prefix:
-    description: Prefix to append to the name of artifacts that are uploaded
-    required: true
-    default: "${{ github.job }}"
-runs:
-  using: composite
-  steps:
-  - name: Upload Test Results artifact
-    uses: actions/upload-artifact@v4.0.0
-    with:
-      name: "${{ inputs.artifact-prefix }}-test-results"
-      path: "${{ runner.temp }}/test-results/**"
-  - name: Upload DB LOG file artifact
-    uses: actions/upload-artifact@v4.0.0
-    with:
-      name: "${{ inputs.artifact-prefix }}-db-log-file"
-      path: LOG
-  - name: Copy Test Logs (on Failure)
-    if: ${{ failure() }}
-    run: |
-      mkdir -p ${{ runner.temp }}/failure-test-logs
-      cp -r t/* ${{ runner.temp }}/failure-test-logs
-    shell: bash
-  - name: Upload Test Logs (on Failure) artifact
-    uses: actions/upload-artifact@v4.0.0
-    with:
-      name: "${{ inputs.artifact-prefix }}-failure-test-logs"
-      path: ${{ runner.temp }}/failure-test-logs/**
-      if-no-files-found: ignore
-  - name: Upload Core Dumps artifact
-    uses: actions/upload-artifact@v4.0.0
-    with:
-      name: "${{ inputs.artifact-prefix }}-core-dumps"
-      path: "core.*"
-      if-no-files-found: ignore
@@ -1,5 +0,0 @@
-name: pre-steps-macos
-runs:
-  using: composite
-  steps:
-  - uses: "./.github/actions/pre-steps"
@@ -1,18 +0,0 @@
-name: pre-steps
-runs:
-  using: composite
-  steps:
-  - name: Setup Environment Variables
-    run: |-
-      echo "GTEST_THROW_ON_FAILURE=0" >> "$GITHUB_ENV"
-      echo "GTEST_OUTPUT=\"xml:${{ runner.temp }}/test-results/\"" >> "$GITHUB_ENV"
-      echo "SKIP_FORMAT_BUCK_CHECKS=1" >> "$GITHUB_ENV"
-      echo "GTEST_COLOR=1" >> "$GITHUB_ENV"
-      echo "CTEST_OUTPUT_ON_FAILURE=1" >> "$GITHUB_ENV"
-      echo "CTEST_TEST_TIMEOUT=300" >> "$GITHUB_ENV"
-      echo "ZLIB_DOWNLOAD_BASE=https://rocksdb-deps.s3.us-west-2.amazonaws.com/pkgs/zlib" >> "$GITHUB_ENV"
-      echo "BZIP2_DOWNLOAD_BASE=https://rocksdb-deps.s3.us-west-2.amazonaws.com/pkgs/bzip2" >> "$GITHUB_ENV"
-      echo "SNAPPY_DOWNLOAD_BASE=https://rocksdb-deps.s3.us-west-2.amazonaws.com/pkgs/snappy" >> "$GITHUB_ENV"
-      echo "LZ4_DOWNLOAD_BASE=https://rocksdb-deps.s3.us-west-2.amazonaws.com/pkgs/lz4" >> "$GITHUB_ENV"
-      echo "ZSTD_DOWNLOAD_BASE=https://rocksdb-deps.s3.us-west-2.amazonaws.com/pkgs/zstd" >> "$GITHUB_ENV"
-    shell: bash
@@ -1,7 +0,0 @@
-name: setup-folly
-runs:
-  using: composite
-  steps:
-  - name: Checkout folly sources
-    run: make checkout_folly
-    shell: bash
@@ -1,20 +0,0 @@
-name: build-folly
-runs:
-  using: composite
-  steps:
-  - name: Fix repo ownership
-    # Needed in some cases, as safe.directory setting doesn't take effect
-    # under env -i
-    run: chown `whoami` . || true
-    shell: bash
-  - name: Set upstream
-    run: git remote add upstream https://github.com/facebook/rocksdb.git
-    shell: bash
-  - name: Fetch upstream
-    run: git fetch upstream
-    shell: bash
-  - name: Git status
-    # NOTE: some old branch builds under check_format_compatible.sh invoke
-    # git under env -i
-    run: git status && git remote -v && env -i git branch
-    shell: bash
@@ -1,54 +0,0 @@
-name: windows-build-steps
-runs:
-  using: composite
-  steps:
-  - name: Add msbuild to PATH
-    uses: microsoft/setup-msbuild@v1.3.1
-  - name: Custom steps
-    env:
-      THIRDPARTY_HOME: ${{ github.workspace }}/thirdparty
-      CMAKE_HOME: C:/Program Files/CMake
-      CMAKE_BIN: C:/Program Files/CMake/bin/cmake.exe
-      CTEST_BIN: C:/Program Files/CMake/bin/ctest.exe
-      JAVA_HOME: C:/Program Files/BellSoft/LibericaJDK-8
-      SNAPPY_HOME: ${{ github.workspace }}/thirdparty/snappy-1.1.8
-      SNAPPY_INCLUDE: ${{ github.workspace }}/thirdparty/snappy-1.1.8;${{ github.workspace }}/thirdparty/snappy-1.1.8/build
-      SNAPPY_LIB_DEBUG: ${{ github.workspace }}/thirdparty/snappy-1.1.8/build/Debug/snappy.lib
-    run: |-
-      # NOTE: if ... Exit $LASTEXITCODE lines needed to exit and report failure
-      echo ===================== Install Dependencies =====================
-      choco install liberica8jdk -y
-      if(!$?) { Exit $LASTEXITCODE }
-      mkdir $Env:THIRDPARTY_HOME
-      cd $Env:THIRDPARTY_HOME
-      echo "Building Snappy dependency..."
-      curl -Lo snappy-1.1.8.zip https://github.com/google/snappy/archive/refs/tags/1.1.8.zip
-      if(!$?) { Exit $LASTEXITCODE }
-      unzip -q snappy-1.1.8.zip
-      if(!$?) { Exit $LASTEXITCODE }
-      cd snappy-1.1.8
-      mkdir build
-      cd build
-      & cmake -G "$Env:CMAKE_GENERATOR" ..
-      if(!$?) { Exit $LASTEXITCODE }
-      msbuild Snappy.sln -maxCpuCount -property:Configuration=Debug -property:Platform=x64
-      if(!$?) { Exit $LASTEXITCODE }
-      echo ======================== Build RocksDB =========================
-      cd ${{ github.workspace }}
-      $env:Path = $env:JAVA_HOME + ";" + $env:Path
-      mkdir build
-      cd build
-      & cmake -G "$Env:CMAKE_GENERATOR" -DCMAKE_BUILD_TYPE=Debug -DOPTDBG=1 -DPORTABLE="$Env:CMAKE_PORTABLE" -DSNAPPY=1 -DJNI=1 ..
-      if(!$?) { Exit $LASTEXITCODE }
-      cd ..
-      echo "Building with VS version: $Env:CMAKE_GENERATOR"
-      msbuild build/rocksdb.sln -maxCpuCount -property:Configuration=Debug -property:Platform=x64
-      if(!$?) { Exit $LASTEXITCODE }
-      echo ========================= Test RocksDB =========================
-      build_tools\run_ci_db_test.ps1 -SuiteRun arena_test,db_basic_test,db_test,db_test2,db_merge_operand_test,bloom_test,c_test,coding_test,crc32c_test,dynamic_bloom_test,env_basic_test,env_test,hash_test,random_test -Concurrency 16
-      if(!$?) { Exit $LASTEXITCODE }
-      echo ======================== Test RocksJava ========================
-      cd build\java
-      & ctest -C Debug -j 16
-      if(!$?) { Exit $LASTEXITCODE }
-    shell: pwsh
@@ -1,15 +0,0 @@
-name: facebook/rocksdb/benchmark-linux
-on: workflow_dispatch
-jobs:
-  # FIXME: when this job is fixed, it should be given a cron schedule like
-  # schedule:
-  # - cron: 0 * * * *
-  # workflow_dispatch:
-  benchmark-linux:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on: ubuntu-latest
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/build-for-benchmarks"
-    - uses: "./.github/actions/perform-benchmarks"
-    - uses: "./.github/actions/post-benchmarks"
@@ -1,18 +0,0 @@
-name: facebook/rocksdb/nightly
-on: workflow_dispatch
-jobs:
-  # These jobs would be in nightly but are failing or otherwise broken for
-  # some reason.
-  build-linux-arm-test-full:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: arm64large
-    container:
-      image: ubuntu-2004:202111-02
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - uses: "./.github/actions/install-gflags"
-    - run: make V=1 J=4 -j4 check
-    - uses: "./.github/actions/post-steps"
@@ -1,98 +0,0 @@
-name: facebook/rocksdb/nightly
-on:
-  schedule:
-  - cron: 0 9 * * *
-  workflow_dispatch:
-jobs:
-  build-format-compatible:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-      with:
-        fetch-depth: 0 # Need full repo history
-        fetch-tags: true
-    - uses: "./.github/actions/setup-upstream"
-    - uses: "./.github/actions/pre-steps"
-    - name: test
-      run: |-
-        export TEST_TMPDIR=/dev/shm/rocksdb
-        rm -rf /dev/shm/rocksdb
-        mkdir /dev/shm/rocksdb
-        git config --global --add safe.directory /__w/rocksdb/rocksdb
-        tools/check_format_compatible.sh
-    - uses: "./.github/actions/post-steps"
-  build-linux-run-microbench:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: DEBUG_LEVEL=0 make -j32 run_microbench
-    - uses: "./.github/actions/post-steps"
-  build-linux-non-shm:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    env:
-      TEST_TMPDIR: "/tmp/rocksdb_test_tmp"
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: make V=1 -j32 check
-    - uses: "./.github/actions/post-steps"
-  build-linux-clang-13-asan-ubsan-with-folly:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - uses: "./.github/actions/setup-folly"
-    - uses: "./.github/actions/build-folly"
-    - run: CC=clang-13 CXX=clang++-13 LIB_MODE=static USE_CLANG=1 USE_FOLLY=1 COMPILE_WITH_UBSAN=1 COMPILE_WITH_ASAN=1 make -j32 check
-    - uses: "./.github/actions/post-steps"
-  build-linux-valgrind:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: PORTABLE=1 make V=1 -j32 valgrind_test
-    - uses: "./.github/actions/post-steps"
-  build-windows-vs2022-avx2:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on: windows-2022
-    env:
-      CMAKE_GENERATOR: Visual Studio 17 2022
-      CMAKE_PORTABLE: AVX2
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/windows-build-steps"
-  build-windows-vs2022:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on: windows-2022
-    env:
-      CMAKE_GENERATOR: Visual Studio 17 2022
-      CMAKE_PORTABLE: 1
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/windows-build-steps"
@@ -1,46 +0,0 @@
-name: facebook/rocksdb/pr-jobs-candidate
-on: workflow_dispatch
-jobs:
-  # These jobs would be in pr-jobs but are failing or otherwise broken for
-  # some reason.
-  # =========================== ARM Jobs ============================ #
-  build-linux-arm:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: arm64large # GitHub hosted ARM runners do not yet exist
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - uses: "./.github/actions/install-gflags"
-    - run: ROCKSDBTESTS_PLATFORM_DEPENDENT=only make V=1 J=4 -j4 all_but_some_tests check_some
-    - uses: "./.github/actions/post-steps"
-  build-linux-arm-cmake-no_test_run:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: arm64large # GitHub hosted ARM runners do not yet exist
-    env:
-      JAVA_HOME: "/usr/lib/jvm/java-8-openjdk-arm64"
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - uses: "./.github/actions/install-gflags"
-    - name: Set Java Environment
-      run: |-
-        echo "JAVA_HOME=${JAVA_HOME}"
-        echo 'export PATH=$JAVA_HOME/bin:$PATH' >> $BASH_ENV
-        which java && java -version
-        which javac && javac -version
-    - name: Build with cmake
-      run: |-
-        mkdir build
-        cd build
-        cmake -DCMAKE_BUILD_TYPE=Release -DWITH_TESTS=0 -DWITH_GFLAGS=1 -DWITH_BENCHMARK_TOOLS=0 -DWITH_TOOLS=0 -DWITH_CORE_TOOLS=1 ..
-        make -j4
-    - name: Build Java with cmake
-      run: |-
-        rm -rf build
-        mkdir build
-        cd build
-        cmake -DJNI=1 -DCMAKE_BUILD_TYPE=Release -DWITH_GFLAGS=1 ..
-        make -j4 rocksdb rocksdbjni
-    - uses: "./.github/actions/post-steps"
@@ -1,609 +0,0 @@
-name: facebook/rocksdb/pr-jobs
-on: [push, pull_request]
-jobs:
-  # NOTE: multiple workflows would be recommended, but the current GHA UI in
-  # PRs doesn't make it clear when there's an overall error with a workflow,
-  # making it easy to overlook something broken. Grouping everything into one
-  # workflow minimizes the problem because it will be suspicious if there are
-  # no GHA results.
-  #
-  # The if: ${{ github.repository_owner == 'facebook' }} lines prevent the
-  # jobs from attempting to run on repo forks, because of a few problems:
-  # * runs-on labels are repository (owner) specific, so the job might wait
-  # for days waiting for a runner that simply isn't available.
-  # * Pushes to branches on forks for pull requests (the normal process) would
-  # run the workflow jobs twice: once in the pull-from fork and once for the PR
-  # destination repo. This is wasteful and dumb.
-  # * It is not known how to avoid copy-pasting the line to each job,
-  # increasing the risk of misconfiguration, especially on forks that might
-  # want to run with this GHA setup.
-  #
-  # DEBUGGING WITH SSH: Temporarily add this as a job step, either before the
-  # step of interest without the "if:" line or after the failing step with the
-  # "if:" line. Then use ssh command printed in CI output.
-  #  - name: Setup tmate session # TEMPORARY!
-  #    if: ${{ failure() }}
-  #    uses: mxschmitt/action-tmate@v3
-  #    with:
-  #      limit-access-to-actor: true
-
-  # ======================== Fast Initial Checks ====================== #
-  check-format-and-targets:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on: ubuntu-latest
-    steps:
-    - uses: actions/checkout@v4.1.0
-      with:
-        fetch-depth: 0 # Need full checkout to determine merge base
-        fetch-tags: true
-    - uses: "./.github/actions/setup-upstream"
-    - name: Setup Python
-      uses: actions/setup-python@v5
-    - name: Install Dependencies
-      run: python -m pip install --upgrade pip
-    - name: Install argparse
-      run: pip install argparse
-    - name: Download clang-format-diff.py
-      run: wget https://raw.githubusercontent.com/llvm/llvm-project/release/12.x/clang/tools/clang-format/clang-format-diff.py
-    - name: Check format
-      run: VERBOSE_CHECK=1 make check-format
-    - name: Compare buckify output
-      run: make check-buck-targets
-    - name: Simple source code checks
-      run: make check-sources
-  # ========================= Linux With Tests ======================== #
-  build-linux:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: make V=1 J=32 -j32 check
-    - uses: "./.github/actions/post-steps"
-  build-linux-cmake-mingw:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 4-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: update-alternatives --set x86_64-w64-mingw32-g++ /usr/bin/x86_64-w64-mingw32-g++-posix
-    - name: Build cmake-mingw
-      run: |-
-        export PATH=$JAVA_HOME/bin:$PATH
-        echo "JAVA_HOME=${JAVA_HOME}"
-        which java && java -version
-        which javac && javac -version
-        mkdir build && cd build && cmake -DJNI=1 -DWITH_GFLAGS=OFF .. -DCMAKE_C_COMPILER=x86_64-w64-mingw32-gcc -DCMAKE_CXX_COMPILER=x86_64-w64-mingw32-g++ -DCMAKE_SYSTEM_NAME=Windows && make -j4 rocksdb rocksdbjni
-    - uses: "./.github/actions/post-steps"
-  build-linux-cmake-with-folly:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - uses: "./.github/actions/setup-folly"
-    - uses: "./.github/actions/build-folly"
-    - run: "(mkdir build && cd build && cmake -DUSE_FOLLY=1 -DWITH_GFLAGS=1 -DROCKSDB_BUILD_SHARED=0 .. && make V=1 -j20 && ctest -j20)"
-    - uses: "./.github/actions/post-steps"
-  build-linux-cmake-with-folly-lite-no-test:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - uses: "./.github/actions/setup-folly"
-    - run: "(mkdir build && cd build && cmake -DUSE_FOLLY_LITE=1 -DWITH_GFLAGS=1 .. && make V=1 -j20)"
-    - uses: "./.github/actions/post-steps"
-  build-linux-gcc-7-with-folly:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - uses: "./.github/actions/setup-folly"
-    - uses: "./.github/actions/build-folly"
-    - run: USE_FOLLY=1 LIB_MODE=static CC=gcc-7 CXX=g++-7 V=1 make -j32 check
-    - uses: "./.github/actions/post-steps"
-  build-linux-gcc-7-with-folly-lite-no-test:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - uses: "./.github/actions/setup-folly"
-    - run: USE_FOLLY_LITE=1 CC=gcc-7 CXX=g++-7 V=1 make -j32 all
-    - uses: "./.github/actions/post-steps"
-  build-linux-cmake-with-folly-coroutines:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    env:
-      CC: gcc-10
-      CXX: g++-10
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - uses: "./.github/actions/setup-folly"
-    - uses: "./.github/actions/build-folly"
-    - run: "(mkdir build && cd build && cmake -DUSE_COROUTINES=1 -DWITH_GFLAGS=1 -DROCKSDB_BUILD_SHARED=0 .. && make V=1 -j20 && ctest -j20)"
-    - uses: "./.github/actions/post-steps"
-  build-linux-cmake-with-benchmark:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: mkdir build && cd build && cmake -DWITH_GFLAGS=1 -DWITH_BENCHMARK=1 .. && make V=1 -j20 && ctest -j20
-    - uses: "./.github/actions/post-steps"
-  build-linux-encrypted_env-no_compression:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: ENCRYPTED_ENV=1 ROCKSDB_DISABLE_SNAPPY=1 ROCKSDB_DISABLE_ZLIB=1 ROCKSDB_DISABLE_BZIP=1 ROCKSDB_DISABLE_LZ4=1 ROCKSDB_DISABLE_ZSTD=1 make V=1 J=32 -j32 check
-    - run: "./sst_dump --help | grep -E -q 'Supported compression types: kNoCompression$' # Verify no compiled in compression\n"
-    - uses: "./.github/actions/post-steps"
-  # ======================== Linux No Test Runs ======================= #
-  build-linux-release:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - run: make V=1 -j32 LIB_MODE=shared release
-    - run: ls librocksdb.so
-    - run: "./db_stress --version"
-    - run: make clean
-    - run: make V=1 -j32 release
-    - run: ls librocksdb.a
-    - run: "./db_stress --version"
-    - run: make clean
-    - run: apt-get remove -y libgflags-dev
-    - run: make V=1 -j32 LIB_MODE=shared release
-    - run: ls librocksdb.so
-    - run: if ./db_stress --version; then false; else true; fi
-    - run: make clean
-    - run: make V=1 -j32 release
-    - run: ls librocksdb.a
-    - run: if ./db_stress --version; then false; else true; fi
-    - uses: "./.github/actions/post-steps"
-  build-linux-release-rtti:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 8-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - run: USE_RTTI=1 DEBUG_LEVEL=0 make V=1 -j16 static_lib tools db_bench
-    - run: "./db_stress --version"
-    - run: make clean
-    - run: apt-get remove -y libgflags-dev
-    - run: USE_RTTI=1 DEBUG_LEVEL=0 make V=1 -j16 static_lib tools db_bench
-    - run: if ./db_stress --version; then false; else true; fi
-  build-examples:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 4-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - name: Build examples
-      run: make V=1 -j4 static_lib && cd examples && make V=1 -j4
-    - uses: "./.github/actions/post-steps"
-  build-fuzzers:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 4-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - name: Build rocksdb lib
-      run: CC=clang-13 CXX=clang++-13 USE_CLANG=1 make -j4 static_lib
-    - name: Build fuzzers
-      run: cd fuzz && make sst_file_writer_fuzzer db_fuzzer db_map_fuzzer
-    - uses: "./.github/actions/post-steps"
-  build-linux-clang-no_test_run:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 8-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - run: CC=clang CXX=clang++ USE_CLANG=1 PORTABLE=1 make V=1 -j16 all
-    - uses: "./.github/actions/post-steps"
-  build-linux-clang-13-no_test_run:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: CC=clang-13 CXX=clang++-13 USE_CLANG=1 make -j32 all microbench
-    - uses: "./.github/actions/post-steps"
-  build-linux-gcc-8-no_test_run:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: CC=gcc-8 CXX=g++-8 V=1 make -j32 all
-    - uses: "./.github/actions/post-steps"
-  build-linux-gcc-10-cxx20-no_test_run:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: CC=gcc-10 CXX=g++-10 V=1 ROCKSDB_CXX_STANDARD=c++20 make -j32 all
-    - uses: "./.github/actions/post-steps"
-  build-linux-gcc-11-no_test_run:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: LIB_MODE=static CC=gcc-11 CXX=g++-11 V=1 make -j32 all microbench
-    - uses: "./.github/actions/post-steps"
-  # ======================== Linux Other Checks ======================= #
-  build-linux-clang10-clang-analyze:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - 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
-    - uses: "./.github/actions/post-steps"
-    - name: compress test report
-      run: tar -cvzf scan_build_report.tar.gz scan_build_report
-      if: failure()
-    - uses: actions/upload-artifact@v4.0.0
-      with:
-        name: scan-build-report
-        path: scan_build_report.tar.gz
-  build-linux-unity-and-headers:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 4-core-ubuntu
-    container:
-      image: gcc:latest
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - run: apt-get update -y && apt-get install -y libgflags-dev
-    - name: Unity build
-      run: make V=1 -j8 unity_test
-    - run: make V=1 -j8 -k check-headers
-    - uses: "./.github/actions/post-steps"
-  build-linux-mini-crashtest:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 4-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: ulimit -S -n `ulimit -H -n` && make V=1 -j8 CRASH_TEST_EXT_ARGS='--duration=960 --max_key=2500000' blackbox_crash_test_with_atomic_flush
-    - uses: "./.github/actions/post-steps"
-  # ======================= Linux with Sanitizers ===================== #
-  build-linux-clang10-asan:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 32-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: COMPILE_WITH_ASAN=1 CC=clang-10 CXX=clang++-10 ROCKSDB_DISABLE_ALIGNED_NEW=1 USE_CLANG=1 make V=1 -j32 check
-    - uses: "./.github/actions/post-steps"
-  build-linux-clang10-ubsan:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - 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
-    - uses: "./.github/actions/post-steps"
-  build-linux-clang10-mini-tsan:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 32-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: COMPILE_WITH_TSAN=1 CC=clang-13 CXX=clang++-13 ROCKSDB_DISABLE_ALIGNED_NEW=1 USE_CLANG=1 make V=1 -j32 check
-    - uses: "./.github/actions/post-steps"
-  build-linux-static_lib-alt_namespace-status_checked:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 16-core-ubuntu
-    container:
-      image: zjay437/rocksdb:0.6
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/pre-steps"
-    - run: ASSERT_STATUS_CHECKED=1 TEST_UINT128_COMPAT=1 ROCKSDB_MODIFY_NPHASH=1 LIB_MODE=static OPT="-DROCKSDB_NAMESPACE=alternative_rocksdb_ns" make V=1 -j24 check
-    - uses: "./.github/actions/post-steps"
-  # ========================= MacOS build only ======================== #
-  build-macos:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on: macos-13
-    env:
-      ROCKSDB_DISABLE_JEMALLOC: 1
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: maxim-lobanov/setup-xcode@v1.6.0
-      with:
-        xcode-version: 14.3.1
-    - uses: "./.github/actions/increase-max-open-files-on-macos"
-    - uses: "./.github/actions/install-gflags-on-macos"
-    - uses: "./.github/actions/pre-steps-macos"
-    - name: Build
-      run: ulimit -S -n `ulimit -H -n` && make V=1 J=16 -j16 all
-    - uses: "./.github/actions/post-steps"
-  # ========================= MacOS with Tests ======================== #
-  build-macos-cmake:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on: macos-13
-    strategy:
-      matrix:
-        run_even_tests: [true, false]
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: maxim-lobanov/setup-xcode@v1.6.0
-      with:
-        xcode-version: 14.3.1
-    - uses: "./.github/actions/increase-max-open-files-on-macos"
-    - uses: "./.github/actions/install-gflags-on-macos"
-    - uses: "./.github/actions/pre-steps-macos"
-    - name: cmake generate project file
-      run: ulimit -S -n `ulimit -H -n` && mkdir build && cd build && cmake -DWITH_GFLAGS=1 ..
-    - name: Build tests
-      run: cd build && make V=1 -j16
-    - name: Run even tests
-      run: ulimit -S -n `ulimit -H -n` && cd build && ctest -j16 -I 0,,2
-      if: ${{ matrix.run_even_tests }}
-    - name: Run odd tests
-      run: ulimit -S -n `ulimit -H -n` && cd build && ctest -j16 -I 1,,2
-      if: ${{ ! matrix.run_even_tests  }}
-    - uses: "./.github/actions/post-steps"
-  # ======================== Windows with Tests ======================= #
-  # NOTE: some windows jobs are in "nightly" to save resources
-  build-windows-vs2019:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on: windows-2019
-    env:
-      CMAKE_GENERATOR: Visual Studio 16 2019
-      CMAKE_PORTABLE: 1
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/windows-build-steps"
-  # ============================ Java Jobs ============================ #
-  build-linux-java:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 4-core-ubuntu
-    container:
-      image: evolvedbinary/rocksjava:centos6_x64-be
-      options: --shm-size=16gb
-    steps:
-    # The docker image is intentionally based on an OS that has an older GLIBC version.
-    # That GLIBC is incompatibile with GitHub's actions/checkout. Thus we implement a manual checkout step.
-    - name: Checkout
-      env:
-        GH_TOKEN: ${{ github.token }}
-      run: |
-        chown `whoami` . || true
-        git clone --no-checkout https://oath2:$GH_TOKEN@github.com/${{ github.repository }}.git .
-        git -c protocol.version=2 fetch --update-head-ok --no-tags --prune --no-recurse-submodules --depth=1 origin +${{ github.sha }}:${{ github.ref }}
-        git checkout --progress --force ${{ github.ref }}
-        git log -1 --format='%H'
-    - uses: "./.github/actions/pre-steps"
-    - name: Set Java Environment
-      run: |-
-        echo "JAVA_HOME=${JAVA_HOME}"
-        which java && java -version
-        which javac && javac -version
-    - name: Test RocksDBJava
-      run: scl enable devtoolset-7 'make V=1 J=8 -j8 jtest'
-    # NOTE: post-steps skipped because of compatibility issues with docker image
-  build-linux-java-static:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 4-core-ubuntu
-    container:
-      image: evolvedbinary/rocksjava:centos6_x64-be
-      options: --shm-size=16gb
-    steps:
-    # The docker image is intentionally based on an OS that has an older GLIBC version.
-    # That GLIBC is incompatibile with GitHub's actions/checkout. Thus we implement a manual checkout step.
-    - name: Checkout
-      env:
-        GH_TOKEN: ${{ github.token }}
-      run: |
-        chown `whoami` . || true
-        git clone --no-checkout https://oath2:$GH_TOKEN@github.com/${{ github.repository }}.git .
-        git -c protocol.version=2 fetch --update-head-ok --no-tags --prune --no-recurse-submodules --depth=1 origin +${{ github.sha }}:${{ github.ref }}
-        git checkout --progress --force ${{ github.ref }}
-        git log -1 --format='%H'
-    - uses: "./.github/actions/pre-steps"
-    - name: Set Java Environment
-      run: |-
-        echo "JAVA_HOME=${JAVA_HOME}"
-        which java && java -version
-        which javac && javac -version
-    - name: Build RocksDBJava Static Library
-      run: scl enable devtoolset-7 'make V=1 J=8 -j8 rocksdbjavastatic'
-    # NOTE: post-steps skipped because of compatibility issues with docker image
-  build-macos-java:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on: macos-13
-    env:
-      JAVA_HOME: "/Library/Java/JavaVirtualMachines/liberica-jdk-8.jdk/Contents/Home"
-      ROCKSDB_DISABLE_JEMALLOC: 1
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: maxim-lobanov/setup-xcode@v1.6.0
-      with:
-        xcode-version: 14.3.1
-    - uses: "./.github/actions/increase-max-open-files-on-macos"
-    - uses: "./.github/actions/install-gflags-on-macos"
-    - uses: "./.github/actions/install-jdk8-on-macos"
-    - uses: "./.github/actions/pre-steps-macos"
-    - name: Set Java Environment
-      run: |-
-        echo "JAVA_HOME=${JAVA_HOME}"
-        which java && java -version
-        which javac && javac -version
-    - name: Test RocksDBJava
-      run: make V=1 J=16 -j16 jtest
-    - uses: "./.github/actions/post-steps"
-  build-macos-java-static:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on: macos-13
-    env:
-      JAVA_HOME: "/Library/Java/JavaVirtualMachines/liberica-jdk-8.jdk/Contents/Home"
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: maxim-lobanov/setup-xcode@v1.6.0
-      with:
-        xcode-version: 14.3.1
-    - uses: "./.github/actions/increase-max-open-files-on-macos"
-    - uses: "./.github/actions/install-gflags-on-macos"
-    - uses: "./.github/actions/install-jdk8-on-macos"
-    - uses: "./.github/actions/pre-steps-macos"
-    - name: Set Java Environment
-      run: |-
-        echo "JAVA_HOME=${JAVA_HOME}"
-        which java && java -version
-        which javac && javac -version
-    - name: Build RocksDBJava x86 and ARM Static Libraries
-      run: make V=1 J=16 -j16 rocksdbjavastaticosx
-    - uses: "./.github/actions/post-steps"
-  build-macos-java-static-universal:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on: macos-13
-    env:
-      JAVA_HOME: "/Library/Java/JavaVirtualMachines/liberica-jdk-8.jdk/Contents/Home"
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: maxim-lobanov/setup-xcode@v1.6.0
-      with:
-        xcode-version: 14.3.1
-    - uses: "./.github/actions/increase-max-open-files-on-macos"
-    - uses: "./.github/actions/install-gflags-on-macos"
-    - uses: "./.github/actions/install-jdk8-on-macos"
-    - uses: "./.github/actions/pre-steps-macos"
-    - name: Set Java Environment
-      run: |-
-        echo "JAVA_HOME=${JAVA_HOME}"
-        which java && java -version
-        which javac && javac -version
-    - name: Build RocksDBJava Universal Binary Static Library
-      run: make V=1 J=16 -j16 rocksdbjavastaticosx_ub
-    - uses: "./.github/actions/post-steps"
-  build-linux-java-pmd:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 4-core-ubuntu
-    container:
-      image: evolvedbinary/rocksjava:rockylinux8_x64-be
-      options: --shm-size=16gb
-    steps:
-    - uses: actions/checkout@v4.1.0
-    - uses: "./.github/actions/install-maven"
-    - uses: "./.github/actions/pre-steps"
-    - name: Set Java Environment
-      run: |-
-        echo "JAVA_HOME=${JAVA_HOME}"
-        which java && java -version
-        which javac && javac -version
-    - name: PMD RocksDBJava
-      run: make V=1 J=8 -j8 jpmd
-    - uses: actions/upload-artifact@v4.0.0
-      with:
-        name: pmd-report
-        path: "${{ github.workspace }}/java/target/pmd.xml"
-    - uses: actions/upload-artifact@v4.0.0
-      with:
-        name: maven-site
-        path: "${{ github.workspace }}/java/target/site"
@@ -0,0 +1,45 @@
+name: Check buck targets and code format
+on: [push, pull_request]
+permissions:
+  contents: read
+
+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
+      run: wget https://raw.githubusercontent.com/llvm/llvm-project/release/12.x/clang/tools/clang-format/clang-format-diff.py
+
+    - name: Check format
+      run: VERBOSE_CHECK=1 make check-format
+
+    - name: Compare buckify output
+      run: make check-buck-targets
+
+    - name: Simple source code checks
+      run: make check-sources
@@ -76,6 +76,11 @@ if (WITH_WINDOWS_UTF8_FILENAMES)
 endif()
 option(ROCKSDB_BUILD_SHARED "Build shared versions of the RocksDB libraries" ON)

+if ($ENV{CIRCLECI})
+  message(STATUS "Build for CircieCI env, a few tests may be disabled")
+  add_definitions(-DCIRCLECI)
+endif()
+
 if( NOT DEFINED CMAKE_CXX_STANDARD )
  set(CMAKE_CXX_STANDARD 17)
 endif()
@@ -166,7 +171,7 @@ else()
  if(WITH_ZSTD)
    find_package(zstd REQUIRED)
    add_definitions(-DZSTD)
-    include_directories(${ZSTD_INCLUDE_DIRS})
+    include_directories(${ZSTD_INCLUDE_DIR})
    list(APPEND THIRDPARTY_LIBS zstd::zstd)
  endif()
 endif()
@@ -371,7 +376,7 @@ option(WITH_NUMA "build with NUMA policy support" OFF)
 if(WITH_NUMA)
  find_package(NUMA REQUIRED)
  add_definitions(-DNUMA)
-  include_directories(${NUMA_INCLUDE_DIRS})
+  include_directories(${NUMA_INCLUDE_DIR})
  list(APPEND THIRDPARTY_LIBS NUMA::NUMA)
 endif()

@@ -767,7 +772,6 @@ set(SOURCES
        options/configurable.cc
        options/customizable.cc
        options/db_options.cc
-        options/offpeak_time_info.cc
        options/options.cc
        options/options_helper.cc
        options/options_parser.cc
@@ -1101,15 +1105,11 @@ set(BUILD_VERSION_CC ${CMAKE_BINARY_DIR}/build_version.cc)
 configure_file(util/build_version.cc.in ${BUILD_VERSION_CC} @ONLY)

 add_library(${ROCKSDB_STATIC_LIB} STATIC ${SOURCES} ${BUILD_VERSION_CC})
-target_include_directories(${ROCKSDB_STATIC_LIB} PUBLIC
-  $<BUILD_INTERFACE:${PROJECT_SOURCE_DIR}/include>)
 target_link_libraries(${ROCKSDB_STATIC_LIB} PRIVATE
  ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})

 if(ROCKSDB_BUILD_SHARED)
  add_library(${ROCKSDB_SHARED_LIB} SHARED ${SOURCES} ${BUILD_VERSION_CC})
-  target_include_directories(${ROCKSDB_SHARED_LIB} PUBLIC
-    $<BUILD_INTERFACE:${PROJECT_SOURCE_DIR}/include>)
  target_link_libraries(${ROCKSDB_SHARED_LIB} PRIVATE
    ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})

@@ -1423,7 +1423,6 @@ if(WITH_TESTS)
        util/ribbon_test.cc
        util/slice_test.cc
        util/slice_transform_test.cc
-        util/string_util_test.cc
        util/timer_queue_test.cc
        util/timer_test.cc
        util/thread_list_test.cc
@@ -1473,7 +1472,7 @@ if(WITH_TESTS)
      utilities/cassandra/test_utils.cc
  )
  enable_testing()
-  add_custom_target(rocksdb_check COMMAND ${CMAKE_CTEST_COMMAND})
+  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} ${FOLLY_LIBS})
@@ -1498,7 +1497,7 @@ if(WITH_TESTS)
      target_link_libraries(${exename}${ARTIFACT_SUFFIX} testutillib${ARTIFACT_SUFFIX} testharness gtest ${THIRDPARTY_LIBS} ${ROCKSDB_LIB})
      if(NOT "${exename}" MATCHES "db_sanity_test")
        gtest_discover_tests(${exename} DISCOVERY_TIMEOUT 120)
-        add_dependencies(rocksdb_check ${exename}${ARTIFACT_SUFFIX})
+        add_dependencies(check ${exename}${ARTIFACT_SUFFIX})
      endif()
  endforeach(sourcefile ${TESTS})

@@ -1518,7 +1517,7 @@ if(WITH_TESTS)
    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(rocksdb_check c_test)
+    add_dependencies(check c_test)
  endif()
 endif()

@@ -1597,3 +1596,6 @@ option(WITH_BENCHMARK "build benchmark tests" OFF)
 if(WITH_BENCHMARK)
  add_subdirectory(${PROJECT_SOURCE_DIR}/microbench/)
 endif()
+
+target_include_directories(${PROJECT_NAME} PUBLIC
+  $<BUILD_INTERFACE:${PROJECT_SOURCE_DIR}/include>)
@@ -1,152 +1,22 @@
 # Rocksdb Change Log
 > NOTE: Entries for next release do not go here. Follow instructions in `unreleased_history/README.txt`

-## 9.0.1 (04/11/2024)
-### Bug Fixes
-* Fixed CMake Javadoc and source jar builds
-* Fixed Java `SstFileMetaData` to prevent throwing `java.lang.NoSuchMethodError`
-
-## 9.0.0 (02/16/2024)
-### New Features
-* Provide support for FSBuffer for point lookups. Also added support for scans and compactions that don't go through prefetching.
-* *Make `SstFileWriter` create SST files without persisting user defined timestamps when the `Option.persist_user_defined_timestamps` flag is set to false.
-* Add support for user-defined timestamps in APIs `DeleteFilesInRanges` and `GetPropertiesOfTablesInRange`.
-* Mark wal\_compression feature as production-ready. Currently only compatible with ZSTD compression.
-
-### Public API Changes
-* Allow setting Stderr logger via C API
-* Declare one Get and one MultiGet variant as pure virtual, and make all the other variants non-overridable. The methods required to be implemented by derived classes of DB allow returning timestamps. It is up to the implementation to check and return an error if timestamps are not supported. The non-batched MultiGet APIs are reimplemented in terms of batched MultiGet, so callers might see a performance improvement.
-* Exposed mode option to Rate Limiter via c api.
-* Removed deprecated option `access_hint_on_compaction_start`
-* Removed deprecated option `ColumnFamilyOptions::check_flush_compaction_key_order`
-* *Remove the default `WritableFile::GetFileSize` and `FSWritableFile::GetFileSize` implementation that returns 0 and make it pure virtual, so that subclasses are enforced to explicitly provide an implementation.
-* Removed deprecated option `ColumnFamilyOptions::level_compaction_dynamic_file_size`
-* *Removed tickers with typos "rocksdb.error.handler.bg.errro.count", "rocksdb.error.handler.bg.io.errro.count", "rocksdb.error.handler.bg.retryable.io.errro.count".
-* Remove the force mode for `EnableFileDeletions` API because it is unsafe with no known legitimate use.
-* Removed deprecated option `ColumnFamilyOptions::ignore_max_compaction_bytes_for_input`
-* `sst_dump --command=check` now compares the number of records in a table with `num_entries` in table property, and reports corruption if there is a mismatch. API `SstFileDumper::ReadSequential()` is updated to optionally do this verification. (#12322)
-
+## 8.7.3 (10/30/2023)
 ### Behavior Changes
-* format\_version=6 is the new default setting in BlockBasedTableOptions, for more robust data integrity checking. DBs and SST files written with this setting cannot be read by RocksDB versions before 8.6.0.
-* Compactions can be scheduled in parallel in an additional scenario: multiple files are marked for compaction within a single column family
-* For leveled compaction, RocksDB will try to do intra-L0 compaction if the total L0 size is small compared to Lbase (#12214). Users with atomic_flush=true are more likely to see the impact of this change.
-
-### Bug Fixes
-* Fixed a data race in `DBImpl::RenameTempFileToOptionsFile`.
-* Fix some perf context statistics error in write steps. which include missing write_memtable_time in unordered_write. missing write_memtable_time in PipelineWrite when Writer stat is STATE_PARALLEL_MEMTABLE_WRITER. missing write_delay_time when calling DelayWrite in WriteImplWALOnly function.
-* Fixed a bug that can, under rare circumstances, cause MultiGet to return an incorrect result for a duplicate key in a MultiGet batch.
-* Fix a bug where older data of an ingested key can be returned for read when universal compaction is used
-
-## 8.11.0 (01/19/2024)
-### New Features
-* Add new statistics: `rocksdb.sst.write.micros` measures time of each write to SST file; `rocksdb.file.write.{flush|compaction|db.open}.micros` measure time of each write to SST table (currently only block-based table format) and blob file for flush, compaction and db open.
+* Deleting stale files upon recovery are delegated to SstFileManger if available so they can be rate limited.

+## 8.7.2 (10/25/2023)
 ### Public API Changes
-* Added another enumerator `kVerify` to enum class `FileOperationType` in listener.h. Update your `switch` statements as needed.
-* Add CompressionOptions to the CompressedSecondaryCacheOptions structure to allow users to specify library specific options when creating the compressed secondary cache.
-* Deprecated several options: `level_compaction_dynamic_file_size`, `ignore_max_compaction_bytes_for_input`, `check_flush_compaction_key_order`, `flush_verify_memtable_count`, `compaction_verify_record_count`, `fail_if_options_file_error`, and `enforce_single_del_contracts`
-* Exposed options ttl via c api.
-
-### Behavior Changes
-* `rocksdb.blobdb.blob.file.write.micros` expands to also measure time writing the header and footer. Therefore the COUNT may be higher and values may be smaller than before. For stacked BlobDB, it no longer measures the time of explictly flushing blob file.
-* Files will be compacted to the next level if the data age exceeds periodic_compaction_seconds except for the last level.
-* Reduced the compaction debt ratio trigger for scheduling parallel compactions
-* For leveled compaction with default compaction pri (kMinOverlappingRatio), files marked for compaction will be prioritized over files not marked when picking a file from a level for compaction.
-
-### Bug Fixes
-* Fix bug in auto_readahead_size that combined with IndexType::kBinarySearchWithFirstKey + fails or iterator lands at a wrong key
-* Fixed some cases in which DB file corruption was detected but ignored on creating a backup with BackupEngine.
-* Fix bugs where `rocksdb.blobdb.blob.file.synced` includes blob files failed to get synced and `rocksdb.blobdb.blob.file.bytes.written` includes blob bytes failed to get written.
-* Fixed a possible memory leak or crash on a failure (such as I/O error) in automatic atomic flush of multiple column families.
-* Fixed some cases of in-memory data corruption using mmap reads with `BackupEngine`, `sst_dump`, or `ldb`.
-* Fixed issues with experimental `preclude_last_level_data_seconds` option that could interfere with expected data tiering.
-* Fixed the handling of the edge case when all existing blob files become unreferenced. Such files are now correctly deleted.
-
-## 8.10.0 (12/15/2023)
-### New Features
-* Provide support for async_io to trim readahead_size by doing block cache lookup
-* Added initial wide-column support in `WriteBatchWithIndex`. This includes the `PutEntity` API and support for wide columns in the existing read APIs (`GetFromBatch`, `GetFromBatchAndDB`, `MultiGetFromBatchAndDB`, and `BaseDeltaIterator`).
-
-### Public API Changes
-* Custom implementations of `TablePropertiesCollectorFactory` may now return a `nullptr` collector to decline processing a file, reducing callback overheads in such cases.
-
-### Behavior Changes
-* Make ReadOptions.auto_readahead_size default true which does prefetching optimizations for forward scans if iterate_upper_bound and block_cache is also specified.
-* Compactions can be scheduled in parallel in an additional scenario: high compaction debt relative to the data size
-* HyperClockCache now has built-in protection against excessive CPU consumption under the extreme stress condition of no (or very few) evictable cache entries, which can slightly increase memory usage such conditions. New option `HyperClockCacheOptions::eviction_effort_cap` controls the space-time trade-off of the response. The default should be generally well-balanced, with no measurable affect on normal operation.
-
-### Bug Fixes
-* Fix a corner case with auto_readahead_size where Prev Operation returns NOT SUPPORTED error when scans direction is changed from forward to backward.
-* Avoid destroying the periodic task scheduler's default timer in order to prevent static destruction order issues.
-* Fix double counting of BYTES_WRITTEN ticker when doing writes with transactions.
-* Fix a WRITE_STALL counter that was reporting wrong value in few cases.
-* A lookup by MultiGet in a TieredCache that goes to the local flash cache and finishes with very low latency, i.e before the subsequent call to WaitAll, is ignored, resulting in a false negative and a memory leak.
-
-### Performance Improvements
-* Java API extensions to improve consistency and completeness of APIs
-1 Extended `RocksDB.get([ColumnFamilyHandle columnFamilyHandle,] ReadOptions opt, ByteBuffer key, ByteBuffer value)` which now accepts indirect buffer parameters as well as direct buffer parameters
-2 Extended `RocksDB.put( [ColumnFamilyHandle columnFamilyHandle,] WriteOptions writeOpts, final ByteBuffer key, final ByteBuffer value)` which now accepts indirect buffer parameters as well as direct buffer parameters
-3 Added `RocksDB.merge([ColumnFamilyHandle columnFamilyHandle,] WriteOptions writeOptions, ByteBuffer key, ByteBuffer value)` methods with the same parameter options as `put(...)` - direct and indirect buffers are supported
-4 Added `RocksIterator.key( byte[] key [, int offset, int len])` methods which retrieve the iterator key into the supplied buffer
-5 Added `RocksIterator.value( byte[] value [, int offset, int len])` methods which retrieve the iterator value into the supplied buffer
-6 Deprecated `get(final ColumnFamilyHandle columnFamilyHandle, final ReadOptions readOptions, byte[])` in favour of `get(final ReadOptions readOptions, final ColumnFamilyHandle columnFamilyHandle, byte[])` which has consistent parameter ordering with other methods in the same class
-7 Added `Transaction.get( ReadOptions opt, [ColumnFamilyHandle columnFamilyHandle, ] byte[] key, byte[] value)` methods which retrieve the requested value into the supplied buffer
-8 Added `Transaction.get( ReadOptions opt, [ColumnFamilyHandle columnFamilyHandle, ] ByteBuffer key, ByteBuffer value)` methods which retrieve the requested value into the supplied buffer
-9 Added `Transaction.getForUpdate( ReadOptions readOptions, [ColumnFamilyHandle columnFamilyHandle, ] byte[] key, byte[] value, boolean exclusive [, boolean doValidate])` methods which retrieve the requested value into the supplied buffer
-10 Added `Transaction.getForUpdate( ReadOptions readOptions, [ColumnFamilyHandle columnFamilyHandle, ] ByteBuffer key, ByteBuffer value, boolean exclusive [, boolean doValidate])` methods which retrieve the requested value into the supplied buffer
-11 Added `Transaction.getIterator()` method as a convenience which defaults the `ReadOptions` value supplied to existing `Transaction.iterator()` methods. This mirrors the existing `RocksDB.iterator()` method.
-12 Added `Transaction.put([ColumnFamilyHandle columnFamilyHandle, ]  ByteBuffer key, ByteBuffer value [, boolean assumeTracked])` methods which supply the key, and the value to be written in a `ByteBuffer` parameter
-13 Added `Transaction.merge([ColumnFamilyHandle columnFamilyHandle, ] ByteBuffer key, ByteBuffer value [,  boolean assumeTracked])` methods which supply the key, and the value to be written/merged in a `ByteBuffer` parameter
-14 Added `Transaction.mergeUntracked([ColumnFamilyHandle columnFamilyHandle, ] ByteBuffer key, ByteBuffer value)` methods which supply the key, and the value to be written/merged in a `ByteBuffer` parameter
-
-
-## 8.9.0 (11/17/2023)
-### New Features
-* Add GetEntity() and PutEntity() API implementation for Attribute Group support. Through the use of Column Families, AttributeGroup enables users to logically group wide-column entities.
-
-### Public API Changes
-* Added rocksdb_ratelimiter_create_auto_tuned API to create an auto-tuned GenericRateLimiter.
-* Added clipColumnFamily() to the Java API to clip the entries in the CF according to the range [begin_key, end_key).
-* Make the `EnableFileDeletion` API not default to force enabling. For users that rely on this default behavior and still
-want to continue to use force enabling, they need to explicitly pass a `true` to `EnableFileDeletion`.
 * Add new Cache APIs GetSecondaryCacheCapacity() and GetSecondaryCachePinnedUsage() to return the configured capacity, and cache reservation charged to the secondary cache.

-### Behavior Changes
-* During off-peak hours defined by `daily_offpeak_time_utc`, the compaction picker will select a larger number of files for periodic compaction. This selection will include files that are projected to expire by the next off-peak start time, ensuring that these files are not chosen for periodic compaction outside of off-peak hours.
-* If an error occurs when writing to a trace file after `DB::StartTrace()`, the subsequent trace writes are skipped to avoid writing to a file that has previously seen error. In this case, `DB::EndTrace()` will also return a non-ok status with info about the error occured previously in its status message.
-* Deleting stale files upon recovery are delegated to SstFileManger if available so they can be rate limited.
-* Make RocksDB only call `TablePropertiesCollector::Finish()` once.
-* When `WAL_ttl_seconds > 0`, we now process archived WALs for deletion at least every `WAL_ttl_seconds / 2` seconds. Previously it could be less frequent in case of small `WAL_ttl_seconds` values when size-based expiration (`WAL_size_limit_MB > 0 `) was simultaneously enabled.
-
 ### Bug Fixes
-* Fixed a crash or assertion failure bug in experimental new HyperClockCache variant, especially when running with a SecondaryCache.
-* Fix a race between flush error recovery and db destruction that can lead to db crashing.
-* Fixed some bugs in the index builder/reader path for user-defined timestamps in Memtable only feature.
-
-## 8.8.0 (10/23/2023)
-### New Features
-* Introduce AttributeGroup by adding the first AttributeGroup support API, MultiGetEntity(). Through the use of Column Families, AttributeGroup enables users to logically group wide-column entities. More APIs to support AttributeGroup will come soon, including GetEntity, PutEntity, and others.
-* Added new tickers `rocksdb.fifo.{max.size|ttl}.compactions` to count FIFO compactions that drop files for different reasons
-* Add an experimental offpeak duration awareness by setting `DBOptions::daily_offpeak_time_utc` in "HH:mm-HH:mm" format. This information will be used for resource optimization in the future
-* Users can now change the max bytes granted in a single refill period (i.e, burst) during runtime by `SetSingleBurstBytes()` for RocksDB rate limiter
-
-### Public API Changes
-* The default value of `DBOptions::fail_if_options_file_error` changed from `false` to `true`. Operations that set in-memory options (e.g., `DB::Open*()`, `DB::SetOptions()`, `DB::CreateColumnFamily*()`, and `DB::DropColumnFamily()`) but fail to persist the change will now return a non-OK `Status` by default.
-
-### Behavior Changes
-* For non direct IO, eliminate the file system prefetching attempt for compaction read when `Options::compaction_readahead_size` is 0
-* During a write stop, writes now block on in-progress recovery attempts
+* Fixed a possible underflow when computing the compressed secondary cache share of memory reservations while updating the compressed secondary to total block cache ratio.
+* Fix an assertion failure when UpdeteTieredCache() is called in an idempotent manner.

+## 8.7.1 (10/20/2023)
 ### Bug Fixes
 * Fix a bug in auto_readahead_size where first_internal_key of index blocks wasn't copied properly resulting in corruption error when first_internal_key was used for comparison.
-* Fixed a bug where compaction read under non direct IO still falls back to RocksDB internal prefetching after file system's prefetching returns non-OK status other than `Status::NotSupported()`
 * Add bounds check in WBWIIteratorImpl and make BaseDeltaIterator, WriteUnpreparedTxn and WritePreparedTxn respect the upper bound and lower bound in ReadOption. See 11680.
-* Fixed the handling of wide-column base values in the `max_successive_merges` logic.
-* Fixed a rare race bug involving a concurrent combination of Create/DropColumnFamily and/or Set(DB)Options that could lead to inconsistency between (a) the DB's reported options state, (b) the DB options in effect, and (c) the latest persisted OPTIONS file.
-* Fixed a possible underflow when computing the compressed secondary cache share of memory reservations while updating the compressed secondary to total block cache ratio.
-
-### Performance Improvements
-* Improved the I/O efficiency of DB::Open a new DB with `create_missing_column_families=true` and many column families.

 ## 8.7.0 (09/22/2023)
 ### New Features
@@ -171,6 +41,7 @@ want to continue to use force enabling, they need to explicitly pass a `true` to
 * Compaction read performance will regress when `Options::compaction_readahead_size` is explicitly set to 0
 * Universal size amp compaction will conditionally exclude some of the newest L0 files when selecting input with a small negative impact to size amp. This is to prevent a large number of L0 files from being locked by a size amp compaction, potentially leading to write stop with a few more flushes.
 * Change ldb scan command delimiter from ':' to '==>'.
+* For non direct IO, eliminate the file system prefetching attempt for compaction read when `Options::compaction_readahead_size` is 0

 ### Bug Fixes
 * Fix a bug where if there is an error reading from offset 0 of a file from L1+ and that the file is not the first file in the sorted run, data can be lost in compaction and read/scan can return incorrect results.
@@ -188,9 +59,10 @@ want to continue to use force enabling, they need to explicitly pass a `true` to
 * Fix an issue in sst dump tool to handle bounds specified for data with user-defined timestamps.
 * When auto_readahead_size is enabled, update readahead upper bound during readahead trimming when reseek changes iterate_upper_bound dynamically.
 * Fixed a bug where `rocksdb.file.read.verify.file.checksums.micros` is not populated
+* Fixed a bug where compaction read under non direct IO still falls back to RocksDB internal prefetching after file system's prefetching returns non-OK status other than `Status::NotSupported()`

 ### Performance Improvements
-* Added additional improvements in tuning readahead_size during Scans when auto_readahead_size is enabled. However it's not supported with Iterator::Prev operation and will return NotSupported error.
+* Added additional improvements in tuning readahead_size during Scans when auto_readahead_size is enabled. However it's not recommended for backward scans and might impact the performance. More details in options.h.
 * During async_io, the Seek happens in 2 phases. Phase 1 starts an asynchronous read on a block cache miss, and phase 2 waits for it to complete and finishes the seek. In both phases, it tries to lookup the block cache for the data block first before looking in the prefetch buffer. It's optimized by doing the block cache lookup only in the first phase that would save some CPU.

 ## 8.6.0 (08/18/2023)
@@ -17,7 +17,7 @@ There are few options when compiling RocksDB:
 * `make check` will compile and run all the unit tests. `make check` will compile RocksDB in debug mode.

 * `make all` will compile our static library, and all our tools and unit tests. Our tools
-depend on gflags 2.2.0 or newer. You will need to have gflags installed to run `make all`. This will compile RocksDB in debug mode. Don't
+depend on gflags. You will need to have gflags installed to run `make all`. This will compile RocksDB in debug mode. Don't
 use binaries compiled by `make all` in production.

 * By default the binary we produce is optimized for the CPU you're compiling on
@@ -77,7 +77,7 @@ most processors made since roughly 2013.

              git clone https://github.com/gflags/gflags.git
              cd gflags
-              git checkout v2.2.0
+              git checkout v2.0
              ./configure && make && sudo make install

      **Notice**: Once installed, please add the include path for gflags to your `CPATH` environment variable and the
@@ -169,26 +169,21 @@ most processors made since roughly 2013.

    * Install the dependencies for RocksDB:

-      `pkg_add gmake gflags snappy bzip2 lz4 zstd git bash findutils gnuwatch`
+        pkg_add gmake gflags snappy bzip2 lz4 zstd git jdk bash findutils gnuwatch

    * Build RocksDB from source:

-        ```bash
        cd ~
        git clone https://github.com/facebook/rocksdb.git
        cd rocksdb
        gmake static_lib
-        ```

    * Build RocksJava from source (optional):
-        * In OpenBSD, JDK depends on XWindows system, so please check that you installed OpenBSD with `xbase` package.
-        * Install dependencies : `pkg_add -v jdk%1.8`
-        ```bash
+
        cd rocksdb
        export JAVA_HOME=/usr/local/jdk-1.8.0
        export PATH=$PATH:/usr/local/jdk-1.8.0/bin
-        gmake rocksdbjava SHA256_CMD='sha256 -q'
-        ```
+        gmake rocksdbjava

 * **iOS**:
  * Run: `TARGET_OS=IOS make static_lib`. When building the project which uses rocksdb iOS library, make sure to define an important pre-processing macros: `IOS_CROSS_COMPILE`.
@@ -539,8 +539,7 @@ endif

 ifdef USE_CLANG
 	# Used by some teams in Facebook
-	WARNING_FLAGS += -Wshift-sign-overflow -Wambiguous-reversed-operator \
-	  -Wimplicit-fallthrough -Wreinterpret-base-class -Wundefined-reinterpret-cast
+	WARNING_FLAGS += -Wshift-sign-overflow -Wambiguous-reversed-operator
 endif

 ifeq ($(PLATFORM), OS_OPENBSD)
@@ -995,7 +994,7 @@ endif

 .PHONY: check_0
 check_0:
-	@printf '%s\n' ''						\
+	printf '%s\n' ''						\
 	  'To monitor subtest <duration,pass/fail,name>,'		\
 	  '  run "make watch-log" in a separate window' '';		\
 	{ \
@@ -1017,7 +1016,7 @@ valgrind-exclude-regexp = InlineSkipTest.ConcurrentInsert|TransactionStressTest.
 .PHONY: valgrind_check_0
 valgrind_check_0: test_log_prefix := valgrind_
 valgrind_check_0:
-	@printf '%s\n' ''						\
+	printf '%s\n' ''						\
 	  'To monitor subtest <duration,pass/fail,name>,'		\
 	  '  run "make watch-log" in a separate window' '';		\
 	{								\
@@ -1769,9 +1768,6 @@ cuckoo_table_db_test: $(OBJ_DIR)/db/cuckoo_table_db_test.o $(TEST_LIBRARY) $(LIB
 listener_test: $(OBJ_DIR)/db/listener_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)

-string_util_test: $(OBJ_DIR)/util/string_util_test.o $(TEST_LIBRARY) $(LIBRARY)
-	$(AM_LINK)
-
 thread_list_test: $(OBJ_DIR)/util/thread_list_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)

@@ -2064,7 +2060,7 @@ JAVA_INCLUDE = -I$(JAVA_HOME)/include/ -I$(JAVA_HOME)/include/linux
 ifeq ($(PLATFORM), OS_SOLARIS)
 	ARCH := $(shell isainfo -b)
 else ifeq ($(PLATFORM), OS_OPENBSD)
-	ifneq (,$(filter amd64 ppc64 ppc64le s390x arm64 aarch64 riscv64 sparc64 loongarch64, $(MACHINE)))
+	ifneq (,$(filter amd64 ppc64 ppc64le s390x arm64 aarch64 sparc64 loongarch64, $(MACHINE)))
 		ARCH := 64
 	else
 		ARCH := 32
@@ -2085,7 +2081,7 @@ ifneq ($(origin JNI_LIBC), undefined)
 endif

 ifeq (,$(ROCKSDBJNILIB))
-ifneq (,$(filter ppc% s390x arm64 aarch64 riscv64 sparc64 loongarch64, $(MACHINE)))
+ifneq (,$(filter ppc% s390x arm64 aarch64 sparc64 loongarch64, $(MACHINE)))
 	ROCKSDBJNILIB = librocksdbjni-linux-$(MACHINE)$(JNI_LIBC_POSTFIX).so
 else
 	ROCKSDBJNILIB = librocksdbjni-linux$(ARCH)$(JNI_LIBC_POSTFIX).so
@@ -2098,8 +2094,8 @@ ROCKSDB_JAVADOCS_JAR = rocksdbjni-$(ROCKSDB_JAVA_VERSION)-javadoc.jar
 ROCKSDB_SOURCES_JAR = rocksdbjni-$(ROCKSDB_JAVA_VERSION)-sources.jar
 SHA256_CMD = sha256sum

-ZLIB_VER ?= 1.3.1
-ZLIB_SHA256 ?= 9a93b2b7dfdac77ceba5a558a580e74667dd6fede4585b91eefb60f03b72df23
+ZLIB_VER ?= 1.3
+ZLIB_SHA256 ?= ff0ba4c292013dbc27530b3a81e1f9a813cd39de01ca5e0f8bf355702efa593e
 ZLIB_DOWNLOAD_BASE ?= http://zlib.net
 BZIP2_VER ?= 1.0.8
 BZIP2_SHA256 ?= ab5a03176ee106d3f0fa90e381da478ddae405918153cca248e682cd0c4a2269
@@ -2350,10 +2346,6 @@ rocksdbjavastaticdockers390x:
 	mkdir -p java/target
 	docker run --rm --name rocksdb_linux_s390x-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:ubuntu18_s390x-be /rocksdb-host/java/crossbuild/docker-build-linux-centos.sh

-rocksdbjavastaticdockerriscv64:
-	mkdir -p java/target
-	docker run --rm --name rocksdb_linux_riscv64-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:ubuntu20_riscv64-be /rocksdb-host/java/crossbuild/docker-build-linux-centos.sh
-
 rocksdbjavastaticdockerx86musl:
 	mkdir -p java/target
 	docker run --rm --name rocksdb_linux_x86-musl-be --platform linux/386 --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:alpine3_x86-be /rocksdb-host/java/crossbuild/docker-build-linux-alpine.sh
@@ -2427,9 +2419,6 @@ jtest_run:
 jtest: rocksdbjava
 	cd java;$(MAKE) sample test

-jpmd: rocksdbjava rocksdbjavageneratepom
-	cd java;$(MAKE) pmd
-
 jdb_bench:
 	cd java;$(MAKE) db_bench;

@@ -5,5 +5,4 @@ This is the list of all known third-party plugins for RocksDB. If something is m
 * [ZenFS](https://github.com/westerndigitalcorporation/zenfs): a file system for zoned block devices
 * [RADOS](https://github.com/riversand963/rocksdb-rados-env): an Env used for interacting with RADOS. Migrated from RocksDB main repo.
 * [PMEM](https://github.com/pmem/pmem-rocksdb-plugin): a collection of plugins to enable Persistent Memory on RocksDB.
-* [IPPCP](https://github.com/intel/ippcp-plugin-rocksdb): a plugin to enable encryption on RocksDB based on Intel optimized open source IPP-Crypto library.
-* [encfs](https://github.com/pegasus-kv/encfs): a plugin to enable encryption on RocksDB based on OpenSSL library.
+* [IPPCP](https://github.com/intel/ippcp-plugin-rocksdb): a plugin to enable encryption on RocksDB based on Intel optimized open source IPP-Crypto library.
@@ -3,6 +3,8 @@
 # --> DO NOT EDIT MANUALLY <--
 # This file is a Facebook-specific integration for buck builds, so can
 # only be validated by Facebook employees.
+#
+# @noautodeps @nocodemods
 load("//rocks/buckifier:defs.bzl", "cpp_library_wrapper","rocks_cpp_library_wrapper","cpp_binary_wrapper","cpp_unittest_wrapper","fancy_bench_wrapper","add_c_test_wrapper")


@@ -163,7 +165,6 @@ cpp_library_wrapper(name="rocksdb_lib", srcs=[
        "options/configurable.cc",
        "options/customizable.cc",
        "options/db_options.cc",
-        "options/offpeak_time_info.cc",
        "options/options.cc",
        "options/options_helper.cc",
        "options/options_parser.cc",
@@ -394,7 +395,6 @@ rocks_cpp_library_wrapper(name="rocksdb_stress_lib", srcs=[
        "db_stress_tool/db_stress_stat.cc",
        "db_stress_tool/db_stress_test_base.cc",
        "db_stress_tool/db_stress_tool.cc",
-        "db_stress_tool/db_stress_wide_merge_operator.cc",
        "db_stress_tool/expected_state.cc",
        "db_stress_tool/expected_value.cc",
        "db_stress_tool/multi_ops_txns_stress.cc",
@@ -407,8 +407,6 @@ rocks_cpp_library_wrapper(name="rocksdb_stress_lib", srcs=[

 cpp_binary_wrapper(name="db_stress", srcs=["db_stress_tool/db_stress.cc"], deps=[":rocksdb_stress_lib"], extra_preprocessor_flags=[], extra_bench_libs=False)

-cpp_binary_wrapper(name="cache_bench", srcs=["cache/cache_bench.cc"], deps=[":rocksdb_cache_bench_tools_lib"], extra_preprocessor_flags=[], extra_bench_libs=False)
-
 cpp_binary_wrapper(name="ribbon_bench", srcs=["microbench/ribbon_bench.cc"], deps=[], extra_preprocessor_flags=[], extra_bench_libs=True)

 cpp_binary_wrapper(name="db_basic_bench", srcs=["microbench/db_basic_bench.cc"], deps=[], extra_preprocessor_flags=[], extra_bench_libs=True)
@@ -5621,5 +5619,3 @@ cpp_unittest_wrapper(name="write_unprepared_transaction_test",
            deps=[":rocksdb_test_lib"],
            extra_compiler_flags=[])

-
-export_file(name = "tools/db_crashtest.py")
@@ -152,9 +152,6 @@ LzLabs is using RocksDB as a storage engine in their multi-database distributed
 ## ArangoDB
 [ArangoDB](https://www.arangodb.com/) is a native multi-model database with flexible data models for documents, graphs, and key-values, for building high performance applications using a convenient SQL-like query language or JavaScript extensions. It uses RocksDB as its storage engine.

-## Qdrant
-[Qdrant](https://qdrant.tech/) is an open source vector database, it [uses](https://qdrant.tech/documentation/concepts/storage/) RocksDB as its persistent storage.
-
 ## Milvus
 [Milvus](https://milvus.io/) is an open source vector database for unstructured data. It uses RocksDB not only as one of the supported kv storage engines, but also as a message queue.

@@ -197,10 +197,6 @@ def generate_targets(repo_path, deps_map):
    TARGETS.add_binary(
        "db_stress", ["db_stress_tool/db_stress.cc"], [":rocksdb_stress_lib"]
    )
-    # cache_bench binary
-    TARGETS.add_binary(
-        "cache_bench", ["cache/cache_bench.cc"], [":rocksdb_cache_bench_tools_lib"]
-    )
    # bench binaries
    for src in src_mk.get("MICROBENCH_SOURCES", []):
        name = src.rsplit("/", 1)[1].split(".")[0] if "/" in src else src.split(".")[0]
@@ -303,7 +299,6 @@ def generate_targets(repo_path, deps_map):
                    deps=json.dumps(deps["extra_deps"] + [":rocksdb_test_lib"]),
                    extra_compiler_flags=json.dumps(deps["extra_compiler_flags"]),
                )
-    TARGETS.export_file("tools/db_crashtest.py")

    print(ColorString.info("Generated TARGETS Summary:"))
    print(ColorString.info("- %d libs" % TARGETS.total_lib))
@@ -148,9 +148,3 @@ add_c_test_wrapper()
                ).encode("utf-8")
            )
        self.total_test = self.total_test + 1
-
-    def export_file(self, name):
-        with open(self.path, "a") as targets_file:
-            targets_file.write(
-                targets_cfg.export_file_template.format(name=name)
-            )
@@ -6,6 +6,8 @@ rocksdb_target_header_template = """# This file \100generated by:
 # --> DO NOT EDIT MANUALLY <--
 # This file is a Facebook-specific integration for buck builds, so can
 # only be validated by Facebook employees.
+#
+# @noautodeps @nocodemods
 load("//rocks/buckifier:defs.bzl", "cpp_library_wrapper","rocks_cpp_library_wrapper","cpp_binary_wrapper","cpp_unittest_wrapper","fancy_bench_wrapper","add_c_test_wrapper")

 """
@@ -37,7 +39,3 @@ fancy_bench_template = """
 fancy_bench_wrapper(suite_name="{name}", binary_to_bench_to_metric_list_map={bench_config}, slow={slow}, expected_runtime={expected_runtime}, sl_iterations={sl_iterations}, regression_threshold={regression_threshold})

 """
-
-export_file_template = """
-export_file(name = "{name}")
-"""
@@ -163,6 +163,24 @@ case "$TARGET_OS" in
            PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -latomic"
        fi
        PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lpthread -lrt -ldl"
+        if test -z "$ROCKSDB_USE_IO_URING"; then
+            ROCKSDB_USE_IO_URING=1
+        fi
+        if test "$ROCKSDB_USE_IO_URING" -ne 0; then
+            # check for liburing
+            $CXX $PLATFORM_CXXFLAGS -x c++ - -luring -o test.o 2>/dev/null  <<EOF
+              #include <liburing.h>
+              int main() {
+                struct io_uring ring;
+                io_uring_queue_init(1, &ring, 0);
+                return 0;
+              }
+EOF
+            if [ "$?" = 0 ]; then
+                PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -luring"
+                COMMON_FLAGS="$COMMON_FLAGS -DROCKSDB_IOURING_PRESENT"
+            fi
+        fi
        # PORT_FILES=port/linux/linux_specific.cc
        ;;
    SunOS)
@@ -596,29 +614,11 @@ EOF
        fi
    fi

-    if test -z "$ROCKSDB_USE_IO_URING"; then
-        ROCKSDB_USE_IO_URING=1
-    fi
-    if [ "$ROCKSDB_USE_IO_URING" -ne 0 -a "$PLATFORM" = OS_LINUX ]; then
-        # check for liburing
-        $CXX $PLATFORM_CXXFLAGS -x c++ - -luring -o test.o 2>/dev/null  <<EOF
-          #include <liburing.h>
-          int main() {
-            struct io_uring ring;
-            io_uring_queue_init(1, &ring, 0);
-            return 0;
-          }
-EOF
-        if [ "$?" = 0 ]; then
-            PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -luring"
-            COMMON_FLAGS="$COMMON_FLAGS -DROCKSDB_IOURING_PRESENT"
-        fi
-    fi
 fi

 # TODO(tec): Fix -Wshorten-64-to-32 errors on FreeBSD and enable the warning.
 # -Wshorten-64-to-32 breaks compilation on FreeBSD aarch64 and i386
-if ! { [ "$TARGET_OS" = FreeBSD -o "$TARGET_OS" = OpenBSD ] && [ "$TARGET_ARCHITECTURE" = arm64 -o "$TARGET_ARCHITECTURE" = i386 ]; }; then
+if ! { [ "$TARGET_OS" = FreeBSD ] && [ "$TARGET_ARCHITECTURE" = arm64 -o "$TARGET_ARCHITECTURE" = i386 ]; }; then
  # Test whether -Wshorten-64-to-32 is available
  $CXX $PLATFORM_CXXFLAGS -x c++ - -o test.o -Wshorten-64-to-32 2>/dev/null  <<EOF
    int main() {}
@@ -647,10 +647,8 @@ if [ "$PORTABLE" == "" ] || [ "$PORTABLE" == 0 ]; then
    fi
    COMMON_FLAGS="$COMMON_FLAGS"
  elif test -n "`echo $TARGET_ARCHITECTURE | grep ^riscv64`"; then
-    RISC_ISA=$(cat /proc/cpuinfo | grep -E '^isa\s*:' | head -1 | cut --delimiter=: -f 2 | cut -b 2-)
-    if [ -n "${RISCV_ISA}" ]; then
-      COMMON_FLAGS="$COMMON_FLAGS -march=${RISC_ISA}"
-    fi
+    RISC_ISA=$(cat /proc/cpuinfo | grep isa | head -1 | cut --delimiter=: -f 2 | cut -b 2-)
+    COMMON_FLAGS="$COMMON_FLAGS -march=${RISC_ISA}"
  elif [ "$TARGET_OS" == "IOS" ]; then
    COMMON_FLAGS="$COMMON_FLAGS"
  else
@@ -662,7 +660,8 @@ else
    if test -n "`echo $TARGET_ARCHITECTURE | grep ^s390x`"; then
      COMMON_FLAGS="$COMMON_FLAGS -march=z196 "
    elif test -n "`echo $TARGET_ARCHITECTURE | grep ^riscv64`"; then
-      COMMON_FLAGS="$COMMON_FLAGS -march=rv64gc"
+      RISC_ISA=$(cat /proc/cpuinfo | grep isa | head -1 | cut --delimiter=: -f 2 | cut -b 2-)
+      COMMON_FLAGS="$COMMON_FLAGS -march=${RISC_ISA}"
    elif test "$USE_SSE"; then
      # USE_SSE is DEPRECATED
      # This is a rough approximation of the old USE_SSE behavior
@@ -1,22 +1,22 @@
 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
 # The file is generated using update_dependencies.sh.
-GCC_BASE=/mnt/gvfs/third-party2/gcc/62de5a92e5f23c661c3d4b9f322e04eb14e7a5bd/11.x/centos8-native/886b5eb
-CLANG_BASE=/mnt/gvfs/third-party2/llvm-fb/1f6edd1ff15c99c861afc8f3cd69054cd974dd64/15/platform010/72a2ff8
-LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/d1129753c8361ac8e9453c0f4291337a4507ebe6/11.x/platform010/5684a5a
-GLIBC_BASE=/mnt/gvfs/third-party2/glibc/fed6e93d87571fb162734c86636119d45a398963/2.34/platform010/f259413
-SNAPPY_BASE=/mnt/gvfs/third-party2/snappy/31a346126a1f3b64812c362511cb04cc1bd40855/1.1.8/platform010/76ebdda
-ZLIB_BASE=/mnt/gvfs/third-party2/zlib/0c65c05468b5a38cef1a106a1f526463e120c8dd/1.2.8/platform010/76ebdda
+GCC_BASE=/mnt/gvfs/third-party2/gcc/e40bde78650fa91b8405a857e3f10bf336633fb0/11.x/centos7-native/886b5eb
+CLANG_BASE=/mnt/gvfs/third-party2/llvm-fb/2043340983c032915adbb6f78903dc855b65aee8/12/platform010/9520e0f
+LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/c00dcc6a3e4125c7e8b248e9a79c14b78ac9e0ca/11.x/platform010/5684a5a
+GLIBC_BASE=/mnt/gvfs/third-party2/glibc/0b9c8e4b060eda62f3bc1c6127bbe1256697569b/2.34/platform010/f259413
+SNAPPY_BASE=/mnt/gvfs/third-party2/snappy/bc9647f7912b131315827d65cb6189c21f381d05/1.1.3/platform010/76ebdda
+ZLIB_BASE=/mnt/gvfs/third-party2/zlib/a6f5f3f1d063d2d00cd02fc12f0f05fc3ab3a994/1.2.11/platform010/76ebdda
 BZIP2_BASE=/mnt/gvfs/third-party2/bzip2/09703139cfc376bd8a82642385a0e97726b28287/1.0.6/platform010/76ebdda
-LZ4_BASE=/mnt/gvfs/third-party2/lz4/ff23d17b932725cc1734a14896a8b67c518ba169/1.9.4/platform010/76ebdda
-ZSTD_BASE=/mnt/gvfs/third-party2/zstd/576397d8b1d9cea7306ad1e454d5e55caaa2ff1c/1.4.x/platform010/64091f4
-GFLAGS_BASE=/mnt/gvfs/third-party2/gflags/fecac07861cb829f5e60dbeff0503d3272db73c0/2.2.0/platform010/76ebdda
-JEMALLOC_BASE=/mnt/gvfs/third-party2/jemalloc/0bb3f5756788ce26e2e16a1cb2f2af2c59b51abe/master/platform010/f57cc4a
-NUMA_BASE=/mnt/gvfs/third-party2/numa/5b602edd46fda54cdd7ea45f77dbe4061206e174/2.0.11/platform010/76ebdda
-LIBUNWIND_BASE=/mnt/gvfs/third-party2/libunwind/97cac22a149c2e202917e05d44e87e516b68216f/1.4/platform010/5074a48
-TBB_BASE=/mnt/gvfs/third-party2/tbb/53953ebc4e3eda85ad6fc3e429ba146035e97b90/2018_U5/platform010/76ebdda
+LZ4_BASE=/mnt/gvfs/third-party2/lz4/60220d6a5bf7722b9cc239a1368c596619b12060/1.9.1/platform010/76ebdda
+ZSTD_BASE=/mnt/gvfs/third-party2/zstd/50eace8143eaaea9473deae1f3283e0049e05633/1.4.x/platform010/64091f4
+GFLAGS_BASE=/mnt/gvfs/third-party2/gflags/5d27e5919771603da06000a027b12f799e58a4f7/2.2.0/platform010/76ebdda
+JEMALLOC_BASE=/mnt/gvfs/third-party2/jemalloc/b62912d333ef33f9760efa6219dbe3fe6abb3b0e/master/platform010/f57cc4a
+NUMA_BASE=/mnt/gvfs/third-party2/numa/6b412770957aa3c8a87e5e0dcd8cc2f45f393bc0/2.0.11/platform010/76ebdda
+LIBUNWIND_BASE=/mnt/gvfs/third-party2/libunwind/52f69816e936e147664ad717eb71a1a0e9dc973a/1.4/platform010/5074a48
+TBB_BASE=/mnt/gvfs/third-party2/tbb/c9cc192099fa84c0dcd0ffeedd44a373ad6e4925/2018_U5/platform010/76ebdda
 LIBURING_BASE=/mnt/gvfs/third-party2/liburing/a98e2d137007e3ebf7f33bd6f99c2c56bdaf8488/20210212/platform010/76ebdda
 BENCHMARK_BASE=/mnt/gvfs/third-party2/benchmark/780c7a0f9cf0967961e69ad08e61cddd85d61821/trunk/platform010/76ebdda
-KERNEL_HEADERS_BASE=/mnt/gvfs/third-party2/kernel-headers/624a2f8f6c93c3c1df8aa4a6255d8202631a6c80/fb/platform010/da39a3e
-BINUTILS_BASE=/mnt/gvfs/third-party2/binutils/39579e8603b48b3540f8b0633f43adf29acccb8b/2.37/centos8-native/da39a3e
-VALGRIND_BASE=/mnt/gvfs/third-party2/valgrind/cd9cc656d49ecb53797ce4d055e49fde29fd57ff/3.19.0/platform010/76ebdda
+KERNEL_HEADERS_BASE=/mnt/gvfs/third-party2/kernel-headers/02d9f76aaaba580611cf75e741753c800c7fdc12/fb/platform010/da39a3e
+BINUTILS_BASE=/mnt/gvfs/third-party2/binutils/938dc3f064ef3a48c0446f5b11d788d50b3eb5ee/2.37/centos7-native/da39a3e
+VALGRIND_BASE=/mnt/gvfs/third-party2/valgrind/429a6b3203eb415f1599bd15183659153129188e/3.15.0/platform010/76ebdda
 LUA_BASE=/mnt/gvfs/third-party2/lua/363787fa5cac2a8aa20638909210443278fa138e/5.3.4/platform010/9079c97
@@ -137,11 +137,11 @@ then
  # 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" | $CLANG_FORMAT_DIFF -p 1) || true
+  diffs=$(git diff -U0 "$FORMAT_UPSTREAM_MERGE_BASE" | $CLANG_FORMAT_DIFF -p 1)
  echo "Checking format of changes not yet in $FORMAT_UPSTREAM..."
 else
  # Check the format of uncommitted lines,
-  diffs=$(git diff -U0 HEAD | $CLANG_FORMAT_DIFF -p 1) || true
+  diffs=$(git diff -U0 HEAD | $CLANG_FORMAT_DIFF -p 1)
  echo "Checking format of uncommitted changes..."
 fi

@@ -149,9 +149,6 @@ if [ -z "$diffs" ]
 then
  echo "Nothing needs to be reformatted!"
  exit 0
-elif [ $? -ne 1 ]; then
-  # CLANG_FORMAT_DIFF will exit on 1 while there is suggested changes.
-  exit $?
 elif [ $CHECK_ONLY ]
 then
  echo "Your change has unformatted code. Please run make format!"
@@ -1561,7 +1561,7 @@ sub save_stdin_stdout_stderr {
 	::die_bug("Can't dup STDERR: $!");
    open $Global::original_stdin, "<&", "STDIN" or
 	::die_bug("Can't dup STDIN: $!");
-    $Global::is_terminal = (-t $Global::original_stderr) && !$ENV{'CIRCLECI'}&& !$ENV{'GITHUB_ACTIONS'}  && !$ENV{'TRAVIS'};
+    $Global::is_terminal = (-t $Global::original_stderr) && !$ENV{'CIRCLECI'} && !$ENV{'TRAVIS'};
 }

 sub enough_file_handles {
@@ -75,8 +75,8 @@ touch "$OUTPUT"
 echo "Writing dependencies to $OUTPUT"

 # Compilers locations
-GCC_BASE=`readlink -f $TP2_LATEST/gcc/11.x/centos8-native/*/`
-CLANG_BASE=`readlink -f $TP2_LATEST/llvm-fb/15/platform010/*/`
+GCC_BASE=`readlink -f $TP2_LATEST/gcc/11.x/centos7-native/*/`
+CLANG_BASE=`readlink -f $TP2_LATEST/llvm-fb/12/platform010/*/`

 log_header
 log_variable GCC_BASE
@@ -86,7 +86,7 @@ log_variable CLANG_BASE
 get_lib_base libgcc     11.x    platform010
 get_lib_base glibc      2.34    platform010
 get_lib_base snappy     LATEST  platform010
-get_lib_base zlib       1.2.8   platform010
+get_lib_base zlib       LATEST  platform010
 get_lib_base bzip2      LATEST  platform010
 get_lib_base lz4        LATEST  platform010
 get_lib_base zstd       LATEST  platform010
@@ -99,7 +99,7 @@ get_lib_base liburing   LATEST  platform010
 get_lib_base benchmark  LATEST  platform010

 get_lib_base kernel-headers fb platform010
-get_lib_base binutils   LATEST centos8-native
+get_lib_base binutils   LATEST centos7-native
 get_lib_base valgrind   LATEST platform010
 get_lib_base lua        5.3.4  platform010

@@ -48,10 +48,6 @@ DEFINE_uint64(cache_size, 1 * GiB,
              "Number of bytes to use as a cache of uncompressed data.");
 DEFINE_int32(num_shard_bits, -1,
             "ShardedCacheOptions::shard_bits. Default = auto");
-DEFINE_int32(
-    eviction_effort_cap,
-    ROCKSDB_NAMESPACE::HyperClockCacheOptions(1, 1).eviction_effort_cap,
-    "HyperClockCacheOptions::eviction_effort_cap");

 DEFINE_double(resident_ratio, 0.25,
              "Ratio of keys fitting in cache to keyspace.");
@@ -67,22 +63,11 @@ DEFINE_int32(
 DEFINE_uint32(skew, 5, "Degree of skew in key selection. 0 = no skew");
 DEFINE_bool(populate_cache, true, "Populate cache before operations");

-DEFINE_double(pinned_ratio, 0.25,
-              "Keep roughly this portion of entries pinned in cache.");
-DEFINE_double(
-    vary_capacity_ratio, 0.0,
-    "If greater than 0.0, will periodically vary the capacity between this "
-    "ratio less than full size and full size. If vary_capacity_ratio + "
-    "pinned_ratio is close to or exceeds 1.0, the cache might thrash.");
-
-DEFINE_uint32(lookup_insert_percent, 82,
+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(blind_insert_percent, 5,
-              "Ratio of insert without keeping handle 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,
@@ -116,6 +101,7 @@ DEFINE_uint32(seed, 0, "Hashing/random seed to use. 0 = choose at random");

 DEFINE_string(secondary_cache_uri, "",
              "Full URI for creating a custom secondary cache object");
+static class std::shared_ptr<ROCKSDB_NAMESPACE::SecondaryCache> secondary_cache;

 DEFINE_string(cache_type, "lru_cache", "Type of block cache.");

@@ -214,19 +200,16 @@ class SharedState {

  bool Started() const { return start_; }

-  void AddLookupStats(uint64_t hits, uint64_t misses, size_t pinned_count) {
+  void AddLookupStats(uint64_t hits, uint64_t misses) {
    MutexLock l(&mu_);
    lookup_count_ += hits + misses;
    lookup_hits_ += hits;
-    pinned_count_ += pinned_count;
  }

  double GetLookupHitRatio() const {
    return 1.0 * lookup_hits_ / lookup_count_;
  }

-  size_t GetPinnedCount() const { return pinned_count_; }
-
 private:
  port::Mutex mu_;
  port::CondVar cv_;
@@ -238,7 +221,6 @@ class SharedState {
  uint64_t num_done_ = 0;
  uint64_t lookup_count_ = 0;
  uint64_t lookup_hits_ = 0;
-  size_t pinned_count_ = 0;
 };

 // Per-thread state for concurrent executions of the same benchmark.
@@ -331,28 +313,6 @@ Cache::CacheItemHelper helper2(CacheEntryRole::kIndexBlock, DeleteFn, SizeFn,
 Cache::CacheItemHelper helper3_wos(CacheEntryRole::kFilterBlock, DeleteFn);
 Cache::CacheItemHelper helper3(CacheEntryRole::kFilterBlock, DeleteFn, SizeFn,
                               SaveToFn, CreateFn, &helper3_wos);
-
-void ConfigureSecondaryCache(ShardedCacheOptions& opts) {
-  if (!FLAGS_secondary_cache_uri.empty()) {
-    std::shared_ptr<SecondaryCache> secondary_cache;
-    Status s = SecondaryCache::CreateFromString(
-        ConfigOptions(), FLAGS_secondary_cache_uri, &secondary_cache);
-    if (secondary_cache == nullptr) {
-      fprintf(stderr,
-              "No secondary cache registered matching string: %s status=%s\n",
-              FLAGS_secondary_cache_uri.c_str(), s.ToString().c_str());
-      exit(1);
-    }
-    opts.secondary_cache = secondary_cache;
-  }
-}
-
-ShardedCacheBase* AsShardedCache(Cache* c) {
-  if (!FLAGS_secondary_cache_uri.empty()) {
-    c = static_cast_with_check<CacheWrapper>(c)->GetTarget().get();
-  }
-  return static_cast_with_check<ShardedCacheBase>(c);
-}
 }  // namespace

 class CacheBench {
@@ -367,9 +327,7 @@ class CacheBench {
                                 FLAGS_lookup_insert_percent),
        insert_threshold_(lookup_insert_threshold_ +
                          kHundredthUint64 * FLAGS_insert_percent),
-        blind_insert_threshold_(insert_threshold_ +
-                                kHundredthUint64 * FLAGS_blind_insert_percent),
-        lookup_threshold_(blind_insert_threshold_ +
+        lookup_threshold_(insert_threshold_ +
                          kHundredthUint64 * FLAGS_lookup_percent),
        erase_threshold_(lookup_threshold_ +
                         kHundredthUint64 * FLAGS_erase_percent) {
@@ -395,7 +353,6 @@ class CacheBench {
          FLAGS_cache_size, /*estimated_entry_charge=*/0, FLAGS_num_shard_bits);
      opts.hash_seed = BitwiseAnd(FLAGS_seed, INT32_MAX);
      opts.memory_allocator = allocator;
-      opts.eviction_effort_cap = FLAGS_eviction_effort_cap;
      if (FLAGS_cache_type == "fixed_hyper_clock_cache" ||
          FLAGS_cache_type == "hyper_clock_cache") {
        opts.estimated_entry_charge = FLAGS_value_bytes_estimate > 0
@@ -409,7 +366,6 @@ class CacheBench {
        fprintf(stderr, "Cache type not supported.\n");
        exit(1);
      }
-      ConfigureSecondaryCache(opts);
      cache_ = opts.MakeSharedCache();
    } else if (FLAGS_cache_type == "lru_cache") {
      LRUCacheOptions opts(FLAGS_cache_size, FLAGS_num_shard_bits,
@@ -417,7 +373,19 @@ class CacheBench {
                           0.5 /* high_pri_pool_ratio */);
      opts.hash_seed = BitwiseAnd(FLAGS_seed, INT32_MAX);
      opts.memory_allocator = allocator;
-      ConfigureSecondaryCache(opts);
+      if (!FLAGS_secondary_cache_uri.empty()) {
+        Status s = SecondaryCache::CreateFromString(
+            ConfigOptions(), FLAGS_secondary_cache_uri, &secondary_cache);
+        if (secondary_cache == nullptr) {
+          fprintf(
+              stderr,
+              "No secondary cache registered matching string: %s status=%s\n",
+              FLAGS_secondary_cache_uri.c_str(), s.ToString().c_str());
+          exit(1);
+        }
+        opts.secondary_cache = secondary_cache;
+      }
+
      cache_ = NewLRUCache(opts);
    } else {
      fprintf(stderr, "Cache type not supported.\n");
@@ -537,8 +505,6 @@ class CacheBench {
    size_t slot = cache_->GetTableAddressCount();
    printf("Final load factor: %g (%zu / %zu)\n", 1.0 * occ / slot, occ, slot);

-    printf("Final pinned count: %zu\n", shared.GetPinnedCount());
-
    if (FLAGS_histograms) {
      printf("\nOperation latency (ns):\n");
      HistogramImpl combined;
@@ -570,7 +536,6 @@ class CacheBench {
  // Cumulative thresholds in the space of a random uint64_t
  const uint64_t lookup_insert_threshold_;
  const uint64_t insert_threshold_;
-  const uint64_t blind_insert_threshold_;
  const uint64_t lookup_threshold_;
  const uint64_t erase_threshold_;

@@ -686,44 +651,29 @@ class CacheBench {
    uint64_t lookup_misses = 0;
    uint64_t lookup_hits = 0;
    // To hold handles for a non-trivial amount of time
-    std::deque<Cache::Handle*> pinned;
-    size_t total_pin_count = static_cast<size_t>(
-        (FLAGS_cache_size * FLAGS_pinned_ratio) / FLAGS_value_bytes + 0.999999);
-    // For this thread. Some round up, some round down, as appropriate
-    size_t pin_count = (total_pin_count + thread->tid) / FLAGS_threads;
-
+    Cache::Handle* handle = nullptr;
    KeyGen gen;
    const auto clock = SystemClock::Default().get();
    uint64_t start_time = clock->NowMicros();
    StopWatchNano timer(clock);
    auto system_clock = SystemClock::Default();
-    size_t steps_to_next_capacity_change = 0;

    for (uint64_t i = 0; i < FLAGS_ops_per_thread; i++) {
      Slice key = gen.GetRand(thread->rnd, max_key_, FLAGS_skew);
      uint64_t random_op = thread->rnd.Next();

-      if (FLAGS_vary_capacity_ratio > 0.0 && thread->tid == 0) {
-        if (steps_to_next_capacity_change == 0) {
-          double cut_ratio = static_cast<double>(thread->rnd.Next()) /
-                             static_cast<double>(UINT64_MAX) *
-                             FLAGS_vary_capacity_ratio;
-          cache_->SetCapacity(FLAGS_cache_size * (1.0 - cut_ratio));
-          steps_to_next_capacity_change =
-              static_cast<size_t>(FLAGS_ops_per_thread / 100);
-        } else {
-          --steps_to_next_capacity_change;
-        }
-      }
-
      if (FLAGS_histograms) {
        timer.Start();
      }

      if (random_op < lookup_insert_threshold_) {
+        if (handle) {
+          cache_->Release(handle);
+          handle = nullptr;
+        }
        // do lookup
-        auto handle = cache_->Lookup(key, &helper2, /*context*/ nullptr,
-                                     Cache::Priority::LOW);
+        handle = cache_->Lookup(key, &helper2, /*context*/ nullptr,
+                                Cache::Priority::LOW);
        if (handle) {
          ++lookup_hits;
          if (!FLAGS_lean) {
@@ -731,31 +681,32 @@ class CacheBench {
            result += NPHash64(static_cast<char*>(cache_->Value(handle)),
                               FLAGS_value_bytes);
          }
-          pinned.push_back(handle);
        } else {
          ++lookup_misses;
          // do insert
          Status s = cache_->Insert(
              key, createValue(thread->rnd, cache_->memory_allocator()),
-              &helper2, FLAGS_value_bytes, &pinned.emplace_back());
+              &helper2, FLAGS_value_bytes, &handle);
          assert(s.ok());
        }
      } else if (random_op < insert_threshold_) {
+        if (handle) {
+          cache_->Release(handle);
+          handle = nullptr;
+        }
        // do insert
        Status s = cache_->Insert(
            key, createValue(thread->rnd, cache_->memory_allocator()), &helper3,
-            FLAGS_value_bytes, &pinned.emplace_back());
-        assert(s.ok());
-      } else if (random_op < blind_insert_threshold_) {
-        // insert without keeping a handle
-        Status s = cache_->Insert(
-            key, createValue(thread->rnd, cache_->memory_allocator()), &helper3,
-            FLAGS_value_bytes);
+            FLAGS_value_bytes, &handle);
        assert(s.ok());
      } else if (random_op < lookup_threshold_) {
+        if (handle) {
+          cache_->Release(handle);
+          handle = nullptr;
+        }
        // do lookup
-        auto handle = cache_->Lookup(key, &helper2, /*context*/ nullptr,
-                                     Cache::Priority::LOW);
+        handle = cache_->Lookup(key, &helper2, /*context*/ nullptr,
+                                Cache::Priority::LOW);
        if (handle) {
          ++lookup_hits;
          if (!FLAGS_lean) {
@@ -763,7 +714,6 @@ class CacheBench {
            result += NPHash64(static_cast<char*>(cache_->Value(handle)),
                               FLAGS_value_bytes);
          }
-          pinned.push_back(handle);
        } else {
          ++lookup_misses;
        }
@@ -777,6 +727,7 @@ class CacheBench {
      if (FLAGS_histograms) {
        thread->latency_ns_hist.Add(timer.ElapsedNanos());
      }
+      thread->shared->AddLookupStats(lookup_hits, lookup_misses);
      if (FLAGS_usleep > 0) {
        unsigned us =
            static_cast<unsigned>(thread->rnd.Uniform(FLAGS_usleep + 1));
@@ -784,17 +735,12 @@ class CacheBench {
          system_clock->SleepForMicroseconds(us);
        }
      }
-      while (pinned.size() > pin_count) {
-        cache_->Release(pinned.front());
-        pinned.pop_front();
-      }
    }
    if (FLAGS_early_exit) {
      MutexLock l(thread->shared->GetMutex());
      exit(0);
    }
-    thread->shared->AddLookupStats(lookup_hits, lookup_misses, pinned.size());
-    for (auto handle : pinned) {
+    if (handle) {
      cache_->Release(handle);
      handle = nullptr;
    }
@@ -823,7 +769,8 @@ class CacheBench {
    printf("Cache size          : %s\n",
           BytesToHumanString(FLAGS_cache_size).c_str());
    printf("Num shard bits      : %d\n",
-           AsShardedCache(cache_.get())->GetNumShardBits());
+           static_cast_with_check<ShardedCacheBase>(cache_.get())
+               ->GetNumShardBits());
    printf("Max key             : %" PRIu64 "\n", max_key_);
    printf("Resident ratio      : %g\n", FLAGS_resident_ratio);
    printf("Skew degree         : %u\n", FLAGS_skew);
@@ -11,7 +11,6 @@

 #include <array>
 #include <atomic>
-#include <climits>
 #include <cstddef>
 #include <cstdint>
 #include <memory>
@@ -25,7 +24,6 @@
 #include "port/port.h"
 #include "rocksdb/cache.h"
 #include "rocksdb/secondary_cache.h"
-#include "util/atomic.h"
 #include "util/autovector.h"
 #include "util/math.h"

@@ -298,7 +296,7 @@ class ClockCacheTest;

 // ----------------------------------------------------------------------- //

-struct ClockHandleBasicData : public Cache::Handle {
+struct ClockHandleBasicData {
  Cache::ObjectPtr value = nullptr;
  const Cache::CacheItemHelper* helper = nullptr;
  // A lossless, reversible hash of the fixed-size (16 byte) cache key. This
@@ -370,19 +368,11 @@ struct ClockHandle : public ClockHandleBasicData {
  // TODO: make these coundown values tuning parameters for eviction?

  // See above. Mutable for read reference counting.
-  mutable AcqRelAtomic<uint64_t> meta{};
+  mutable std::atomic<uint64_t> meta{};
 };  // struct ClockHandle

 class BaseClockTable {
 public:
-  struct BaseOpts {
-    explicit BaseOpts(int _eviction_effort_cap)
-        : eviction_effort_cap(_eviction_effort_cap) {}
-    explicit BaseOpts(const HyperClockCacheOptions& opts)
-        : BaseOpts(opts.eviction_effort_cap) {}
-    int eviction_effort_cap;
-  };
-
  BaseClockTable(CacheMetadataChargePolicy metadata_charge_policy,
                 MemoryAllocator* allocator,
                 const Cache::EvictionCallback* eviction_callback,
@@ -395,37 +385,36 @@ class BaseClockTable {
  template <class Table>
  typename Table::HandleImpl* CreateStandalone(ClockHandleBasicData& proto,
                                               size_t capacity,
-                                               uint32_t eec_and_scl,
+                                               bool strict_capacity_limit,
                                               bool allow_uncharged);

  template <class Table>
  Status Insert(const ClockHandleBasicData& proto,
                typename Table::HandleImpl** handle, Cache::Priority priority,
-                size_t capacity, uint32_t eec_and_scl);
+                size_t capacity, bool strict_capacity_limit);

  void Ref(ClockHandle& handle);

-  size_t GetOccupancy() const { return occupancy_.LoadRelaxed(); }
+  size_t GetOccupancy() const {
+    return occupancy_.load(std::memory_order_relaxed);
+  }

-  size_t GetUsage() const { return usage_.LoadRelaxed(); }
+  size_t GetUsage() const { return usage_.load(std::memory_order_relaxed); }

-  size_t GetStandaloneUsage() const { return standalone_usage_.LoadRelaxed(); }
+  size_t GetStandaloneUsage() const {
+    return standalone_usage_.load(std::memory_order_relaxed);
+  }

  uint32_t GetHashSeed() const { return hash_seed_; }

-  uint64_t GetYieldCount() const { return yield_count_.LoadRelaxed(); }
-
-  uint64_t GetEvictionEffortExceededCount() const {
-    return eviction_effort_exceeded_count_.LoadRelaxed();
-  }
+  uint64_t GetYieldCount() const { return yield_count_.load(); }

  struct EvictionData {
    size_t freed_charge = 0;
    size_t freed_count = 0;
-    size_t seen_pinned_count = 0;
  };

-  void TrackAndReleaseEvictedEntry(ClockHandle* h);
+  void TrackAndReleaseEvictedEntry(ClockHandle* h, EvictionData* data);

 #ifndef NDEBUG
  // Acquire N references
@@ -450,7 +439,6 @@ class BaseClockTable {
  template <class Table>
  Status ChargeUsageMaybeEvictStrict(size_t total_charge, size_t capacity,
                                     bool need_evict_for_occupancy,
-                                     uint32_t eviction_effort_cap,
                                     typename Table::InsertState& state);

  // Helper for updating `usage_` for new entry with given `total_charge`
@@ -464,7 +452,6 @@ class BaseClockTable {
  template <class Table>
  bool ChargeUsageMaybeEvictNonStrict(size_t total_charge, size_t capacity,
                                      bool need_evict_for_occupancy,
-                                      uint32_t eviction_effort_cap,
                                      typename Table::InsertState& state);

 protected:  // data
@@ -473,29 +460,21 @@ class BaseClockTable {
  // operations in ClockCacheShard.

  // Clock algorithm sweep pointer.
-  // (Relaxed: only needs to be consistent with itself.)
-  RelaxedAtomic<uint64_t> clock_pointer_{};
+  std::atomic<uint64_t> clock_pointer_{};

  // Counter for number of times we yield to wait on another thread.
-  // It is normal for this to occur rarely in normal operation.
-  // (Relaxed: a simple stat counter.)
-  RelaxedAtomic<uint64_t> yield_count_{};
-
-  // Counter for number of times eviction effort cap is exceeded.
-  // It is normal for this to occur rarely in normal operation.
-  // (Relaxed: a simple stat counter.)
-  RelaxedAtomic<uint64_t> eviction_effort_exceeded_count_{};
+  std::atomic<uint64_t> yield_count_{};

  // TODO: is this separation needed if we don't do background evictions?
  ALIGN_AS(CACHE_LINE_SIZE)
  // Number of elements in the table.
-  AcqRelAtomic<size_t> occupancy_{};
+  std::atomic<size_t> occupancy_{};

  // Memory usage by entries tracked by the cache (including standalone)
-  AcqRelAtomic<size_t> usage_{};
+  std::atomic<size_t> usage_{};

  // Part of usage by standalone entries (not in table)
-  AcqRelAtomic<size_t> standalone_usage_{};
+  std::atomic<size_t> standalone_usage_{};

  ALIGN_AS(CACHE_LINE_SIZE)
  const CacheMetadataChargePolicy metadata_charge_policy_;
@@ -521,11 +500,7 @@ class FixedHyperClockTable : public BaseClockTable {
  struct ALIGN_AS(64U) HandleImpl : public ClockHandle {
    // The number of elements that hash to this slot or a lower one, but wind
    // up in this slot or a higher one.
-    // (Relaxed: within a Cache op, does not need consistency with entries
-    // inserted/removed during that op. For example, a Lookup() that
-    // happens-after an Insert() will see an appropriate displacements value
-    // for the entry to be in a published state.)
-    RelaxedAtomic<uint32_t> displacements{};
+    std::atomic<uint32_t> displacements{};

    // Whether this is a "deteched" handle that is independently allocated
    // with `new` (so must be deleted with `delete`).
@@ -539,19 +514,17 @@ class FixedHyperClockTable : public BaseClockTable {
    inline void SetStandalone() { standalone = true; }
  };  // struct HandleImpl

-  struct Opts : public BaseOpts {
-    explicit Opts(size_t _estimated_value_size, int _eviction_effort_cap)
-        : BaseOpts(_eviction_effort_cap),
-          estimated_value_size(_estimated_value_size) {}
-    explicit Opts(const HyperClockCacheOptions& opts)
-        : BaseOpts(opts.eviction_effort_cap) {
+  struct Opts {
+    explicit Opts(size_t _estimated_value_size)
+        : estimated_value_size(_estimated_value_size) {}
+    explicit Opts(const HyperClockCacheOptions& opts) {
      assert(opts.estimated_entry_charge > 0);
      estimated_value_size = opts.estimated_entry_charge;
    }
    size_t estimated_value_size;
  };

-  FixedHyperClockTable(size_t capacity,
+  FixedHyperClockTable(size_t capacity, bool strict_capacity_limit,
                       CacheMetadataChargePolicy metadata_charge_policy,
                       MemoryAllocator* allocator,
                       const Cache::EvictionCallback* eviction_callback,
@@ -573,8 +546,7 @@ class FixedHyperClockTable : public BaseClockTable {
  // Runs the clock eviction algorithm trying to reclaim at least
  // requested_charge. Returns how much is evicted, which could be less
  // if it appears impossible to evict the requested amount without blocking.
-  void Evict(size_t requested_charge, InsertState& state, EvictionData* data,
-             uint32_t eviction_effort_cap);
+  void Evict(size_t requested_charge, InsertState& state, EvictionData* data);

  HandleImpl* Lookup(const UniqueId64x2& hashed_key);

@@ -815,33 +787,32 @@ class AutoHyperClockTable : public BaseClockTable {

    // See above. The head pointer is logically independent of the rest of
    // the entry, including the chain next pointer.
-    AcqRelAtomic<uint64_t> head_next_with_shift{kUnusedMarker};
-    AcqRelAtomic<uint64_t> chain_next_with_shift{kUnusedMarker};
+    std::atomic<uint64_t> head_next_with_shift{kUnusedMarker};
+    std::atomic<uint64_t> chain_next_with_shift{kUnusedMarker};

    // For supporting CreateStandalone and some fallback cases.
    inline bool IsStandalone() const {
-      return head_next_with_shift.Load() == kStandaloneMarker;
+      return head_next_with_shift.load(std::memory_order_acquire) ==
+             kStandaloneMarker;
    }

    inline void SetStandalone() {
-      head_next_with_shift.Store(kStandaloneMarker);
+      head_next_with_shift.store(kStandaloneMarker, std::memory_order_release);
    }
  };  // struct HandleImpl

-  struct Opts : public BaseOpts {
-    explicit Opts(size_t _min_avg_value_size, int _eviction_effort_cap)
-        : BaseOpts(_eviction_effort_cap),
-          min_avg_value_size(_min_avg_value_size) {}
+  struct Opts {
+    explicit Opts(size_t _min_avg_value_size)
+        : min_avg_value_size(_min_avg_value_size) {}

-    explicit Opts(const HyperClockCacheOptions& opts)
-        : BaseOpts(opts.eviction_effort_cap) {
+    explicit Opts(const HyperClockCacheOptions& opts) {
      assert(opts.estimated_entry_charge == 0);
      min_avg_value_size = opts.min_avg_entry_charge;
    }
    size_t min_avg_value_size;
  };

-  AutoHyperClockTable(size_t capacity,
+  AutoHyperClockTable(size_t capacity, bool strict_capacity_limit,
                      CacheMetadataChargePolicy metadata_charge_policy,
                      MemoryAllocator* allocator,
                      const Cache::EvictionCallback* eviction_callback,
@@ -868,8 +839,7 @@ class AutoHyperClockTable : public BaseClockTable {
  // Runs the clock eviction algorithm trying to reclaim at least
  // requested_charge. Returns how much is evicted, which could be less
  // if it appears impossible to evict the requested amount without blocking.
-  void Evict(size_t requested_charge, InsertState& state, EvictionData* data,
-             uint32_t eviction_effort_cap);
+  void Evict(size_t requested_charge, InsertState& state, EvictionData* data);

  HandleImpl* Lookup(const UniqueId64x2& hashed_key);

@@ -934,8 +904,7 @@ class AutoHyperClockTable : public BaseClockTable {
  // with proper handling to ensure all existing data is seen even in the
  // presence of concurrent insertions, etc. (See implementation.)
  template <class OpData>
-  void PurgeImpl(OpData* op_data, size_t home = SIZE_MAX,
-                 EvictionData* data = nullptr);
+  void PurgeImpl(OpData* op_data, size_t home = SIZE_MAX);

  // An RAII wrapper for locking a chain of entries for removals. See
  // implementation.
@@ -945,7 +914,7 @@ class AutoHyperClockTable : public BaseClockTable {
  // implementation.
  template <class OpData>
  void PurgeImplLocked(OpData* op_data, ChainRewriteLock& rewrite_lock,
-                       size_t home, EvictionData* data);
+                       size_t home);

  // Update length_info_ as much as possible without waiting, given a known
  // usable (ready for inserts and lookups) grow_home. (Previous grow_homes
@@ -970,25 +939,22 @@ class AutoHyperClockTable : public BaseClockTable {
  // log time to find the correct chain, but normally this value enables
  // readers to find the correct chain on the first try.
  //
-  // To maximize parallelization of Grow() operations, this field is only
-  // updated opportunistically after Grow() operations and in DoInsert() where
-  // it is found to be out-of-date. See CatchUpLengthInfoNoWait().
-  AcqRelAtomic<uint64_t> length_info_;
+  // NOTES: length_info_ is only updated at the end of a Grow operation,
+  // so that waiting in Grow operations isn't done while entries are pinned
+  // for internal operation purposes. Thus, Lookup and Insert have to
+  // detect and support cases where length_info hasn't caught up to updated
+  // chains. Winning grow thread is the one that transitions
+  // head_next_with_shift from zeros. Grow threads can spin/yield wait for
+  // preconditions and postconditions to be met.
+  std::atomic<uint64_t> length_info_;

  // An already-computed version of the usable length times the max load
  // factor. Could be slightly out of date but GrowIfNeeded()/Grow() handle
  // that internally.
-  // (Relaxed: allowed to lag behind length_info_ by a little)
-  RelaxedAtomic<size_t> occupancy_limit_;
-
-  // The next index to use from array_ upon the next Grow(). Might be ahead of
-  // length_info_.
-  // (Relaxed: self-contained source of truth for next grow home)
-  RelaxedAtomic<size_t> grow_frontier_;
+  std::atomic<size_t> occupancy_limit_;

  // See explanation in AutoHyperClockTable::Evict
-  // (Relaxed: allowed to lag behind clock_pointer_ and length_info_ state)
-  RelaxedAtomic<size_t> clock_pointer_mask_;
+  std::atomic<size_t> clock_pointer_mask_;
 };  // class AutoHyperClockTable

 // A single shard of sharded cache.
@@ -1104,13 +1070,10 @@ class ALIGN_AS(CACHE_LINE_SIZE) ClockCacheShard final : public CacheShardBase {
  Table table_;

  // Maximum total charge of all elements stored in the table.
-  // (Relaxed: eventual consistency/update is OK)
-  RelaxedAtomic<size_t> capacity_;
+  std::atomic<size_t> capacity_;

-  // Encodes eviction_effort_cap (bottom 31 bits) and strict_capacity_limit
-  // (top bit). See HyperClockCacheOptions::eviction_effort_cap etc.
-  // (Relaxed: eventual consistency/update is OK)
-  RelaxedAtomic<uint32_t> eec_and_scl_;
+  // Whether to reject insertion if cache reaches its full capacity.
+  std::atomic<bool> strict_capacity_limit_;
 };  // class ClockCacheShard

 template <class Table>
@@ -31,15 +31,11 @@ CompressedSecondaryCache::~CompressedSecondaryCache() {}
 std::unique_ptr<SecondaryCacheResultHandle> CompressedSecondaryCache::Lookup(
    const Slice& key, const Cache::CacheItemHelper* helper,
    Cache::CreateContext* create_context, bool /*wait*/, bool advise_erase,
-    Statistics* stats, bool& kept_in_sec_cache) {
+    bool& kept_in_sec_cache) {
  assert(helper);
-  // This is a minor optimization. Its ok to skip it in TSAN in order to
-  // avoid a false positive.
-#ifndef __SANITIZE_THREAD__
  if (disable_cache_) {
    return nullptr;
  }
-#endif

  std::unique_ptr<SecondaryCacheResultHandle> handle;
  kept_in_sec_cache = false;
@@ -51,7 +47,6 @@ std::unique_ptr<SecondaryCacheResultHandle> CompressedSecondaryCache::Lookup(
  void* handle_value = cache_->Value(lru_handle);
  if (handle_value == nullptr) {
    cache_->Release(lru_handle, /*erase_if_last_ref=*/false);
-    RecordTick(stats, COMPRESSED_SECONDARY_CACHE_DUMMY_HITS);
    return nullptr;
  }

@@ -138,7 +133,6 @@ std::unique_ptr<SecondaryCacheResultHandle> CompressedSecondaryCache::Lookup(
    cache_->Release(lru_handle, /*erase_if_last_ref=*/false);
  }
  handle.reset(new CompressedSecondaryCacheResultHandle(value, charge));
-  RecordTick(stats, COMPRESSED_SECONDARY_CACHE_HITS);
  return handle;
 }

@@ -192,13 +186,13 @@ Status CompressedSecondaryCache::InsertInternal(
      type == kNoCompression &&
      !cache_options_.do_not_compress_roles.Contains(helper->role)) {
    PERF_COUNTER_ADD(compressed_sec_cache_uncompressed_bytes, data_size);
+    CompressionOptions compression_opts;
    CompressionContext compression_context(cache_options_.compression_type,
-                                           cache_options_.compression_opts);
+                                           compression_opts);
    uint64_t sample_for_compression{0};
    CompressionInfo compression_info(
-        cache_options_.compression_opts, compression_context,
-        CompressionDict::GetEmptyDict(), cache_options_.compression_type,
-        sample_for_compression);
+        compression_opts, compression_context, CompressionDict::GetEmptyDict(),
+        cache_options_.compression_type, sample_for_compression);

    bool success =
        CompressData(val, compression_info,
@@ -291,11 +285,6 @@ std::string CompressedSecondaryCache::GetPrintableOptions() const {
  snprintf(buffer, kBufferSize, "    compression_type : %s\n",
           CompressionTypeToString(cache_options_.compression_type).c_str());
  ret.append(buffer);
-  snprintf(buffer, kBufferSize, "    compression_opts : %s\n",
-           CompressionOptionsToString(
-               const_cast<CompressionOptions&>(cache_options_.compression_opts))
-               .c_str());
-  ret.append(buffer);
  snprintf(buffer, kBufferSize, "    compress_format_version : %d\n",
           cache_options_.compress_format_version);
  ret.append(buffer);
@@ -86,7 +86,7 @@ class CompressedSecondaryCache : public SecondaryCache {
  std::unique_ptr<SecondaryCacheResultHandle> Lookup(
      const Slice& key, const Cache::CacheItemHelper* helper,
      Cache::CreateContext* create_context, bool /*wait*/, bool advise_erase,
-      Statistics* stats, bool& kept_in_sec_cache) override;
+      bool& kept_in_sec_cache) override;

  bool SupportForceErase() const override { return true; }

@@ -44,7 +44,7 @@ class CompressedSecondaryCacheTestBase : public testing::Test,
    // Lookup an non-existent key.
    std::unique_ptr<SecondaryCacheResultHandle> handle0 =
        sec_cache->Lookup(key0, GetHelper(), this, true, /*advise_erase=*/true,
-                          /*stats=*/nullptr, kept_in_sec_cache);
+                          kept_in_sec_cache);
    ASSERT_EQ(handle0, nullptr);

    Random rnd(301);
@@ -59,7 +59,7 @@ class CompressedSecondaryCacheTestBase : public testing::Test,

    std::unique_ptr<SecondaryCacheResultHandle> handle1_1 =
        sec_cache->Lookup(key1, GetHelper(), this, true, /*advise_erase=*/false,
-                          /*stats=*/nullptr, kept_in_sec_cache);
+                          kept_in_sec_cache);
    ASSERT_EQ(handle1_1, nullptr);

    // Insert and Lookup the item k1 for the second time and advise erasing it.
@@ -68,7 +68,7 @@ class CompressedSecondaryCacheTestBase : public testing::Test,

    std::unique_ptr<SecondaryCacheResultHandle> handle1_2 =
        sec_cache->Lookup(key1, GetHelper(), this, true, /*advise_erase=*/true,
-                          /*stats=*/nullptr, kept_in_sec_cache);
+                          kept_in_sec_cache);
    ASSERT_NE(handle1_2, nullptr);
    ASSERT_FALSE(kept_in_sec_cache);
    if (sec_cache_is_compressed) {
@@ -89,7 +89,7 @@ class CompressedSecondaryCacheTestBase : public testing::Test,
    // Lookup the item k1 again.
    std::unique_ptr<SecondaryCacheResultHandle> handle1_3 =
        sec_cache->Lookup(key1, GetHelper(), this, true, /*advise_erase=*/true,
-                          /*stats=*/nullptr, kept_in_sec_cache);
+                          kept_in_sec_cache);
    ASSERT_EQ(handle1_3, nullptr);

    // Insert and Lookup the item k2.
@@ -99,7 +99,7 @@ class CompressedSecondaryCacheTestBase : public testing::Test,
    ASSERT_EQ(get_perf_context()->compressed_sec_cache_insert_dummy_count, 2);
    std::unique_ptr<SecondaryCacheResultHandle> handle2_1 =
        sec_cache->Lookup(key2, GetHelper(), this, true, /*advise_erase=*/false,
-                          /*stats=*/nullptr, kept_in_sec_cache);
+                          kept_in_sec_cache);
    ASSERT_EQ(handle2_1, nullptr);

    ASSERT_OK(sec_cache->Insert(key2, &item2, GetHelper(), false));
@@ -115,7 +115,7 @@ class CompressedSecondaryCacheTestBase : public testing::Test,
    }
    std::unique_ptr<SecondaryCacheResultHandle> handle2_2 =
        sec_cache->Lookup(key2, GetHelper(), this, true, /*advise_erase=*/false,
-                          /*stats=*/nullptr, kept_in_sec_cache);
+                          kept_in_sec_cache);
    ASSERT_NE(handle2_2, nullptr);
    std::unique_ptr<TestItem> val2 =
        std::unique_ptr<TestItem>(static_cast<TestItem*>(handle2_2->Value()));
@@ -196,14 +196,14 @@ class CompressedSecondaryCacheTestBase : public testing::Test,
    bool kept_in_sec_cache{false};
    std::unique_ptr<SecondaryCacheResultHandle> handle1 =
        sec_cache->Lookup(key1, GetHelper(), this, true, /*advise_erase=*/false,
-                          /*stats=*/nullptr, kept_in_sec_cache);
+                          kept_in_sec_cache);
    ASSERT_EQ(handle1, nullptr);

    // Insert k2 and k1 is evicted.
    ASSERT_OK(sec_cache->Insert(key2, &item2, GetHelper(), false));
    std::unique_ptr<SecondaryCacheResultHandle> handle2 =
        sec_cache->Lookup(key2, GetHelper(), this, true, /*advise_erase=*/false,
-                          /*stats=*/nullptr, kept_in_sec_cache);
+                          kept_in_sec_cache);
    ASSERT_NE(handle2, nullptr);
    std::unique_ptr<TestItem> val2 =
        std::unique_ptr<TestItem>(static_cast<TestItem*>(handle2->Value()));
@@ -215,14 +215,14 @@ class CompressedSecondaryCacheTestBase : public testing::Test,

    std::unique_ptr<SecondaryCacheResultHandle> handle1_1 =
        sec_cache->Lookup(key1, GetHelper(), this, true, /*advise_erase=*/false,
-                          /*stats=*/nullptr, kept_in_sec_cache);
+                          kept_in_sec_cache);
    ASSERT_EQ(handle1_1, nullptr);

    // Create Fails.
    SetFailCreate(true);
    std::unique_ptr<SecondaryCacheResultHandle> handle2_1 =
        sec_cache->Lookup(key2, GetHelper(), this, true, /*advise_erase=*/true,
-                          /*stats=*/nullptr, kept_in_sec_cache);
+                          kept_in_sec_cache);
    ASSERT_EQ(handle2_1, nullptr);

    // Save Fails.
@@ -912,9 +912,9 @@ TEST_P(CompressedSecondaryCacheTestWithCompressionParam, EntryRoles) {
    ASSERT_EQ(get_perf_context()->compressed_sec_cache_insert_real_count, 1U);

    bool kept_in_sec_cache{true};
-    std::unique_ptr<SecondaryCacheResultHandle> handle = sec_cache->Lookup(
-        ith_key, GetHelper(role), this, true,
-        /*advise_erase=*/true, /*stats=*/nullptr, kept_in_sec_cache);
+    std::unique_ptr<SecondaryCacheResultHandle> handle =
+        sec_cache->Lookup(ith_key, GetHelper(role), this, true,
+                          /*advise_erase=*/true, kept_in_sec_cache);
    ASSERT_NE(handle, nullptr);

    // Lookup returns the right data
@@ -992,8 +992,6 @@ class CompressedSecCacheTestWithTiered
        /*_capacity=*/0,
        /*_estimated_entry_charge=*/256 << 10,
        /*_num_shard_bits=*/0);
-    // eviction_effort_cap setting simply to avoid churn in existing test
-    hcc_opts.eviction_effort_cap = 100;
    TieredCacheOptions opts;
    lru_opts.capacity = 0;
    lru_opts.num_shard_bits = 0;
@@ -1062,7 +1060,7 @@ bool CacheUsageWithinBounds(size_t val1, size_t val2, size_t error) {

 TEST_P(CompressedSecCacheTestWithTiered, CacheReservationManager) {
  CompressedSecondaryCache* sec_cache =
-      static_cast<CompressedSecondaryCache*>(GetSecondaryCache());
+      reinterpret_cast<CompressedSecondaryCache*>(GetSecondaryCache());

  // Use EXPECT_PRED3 instead of EXPECT_NEAR to void too many size_t to
  // double explicit casts
@@ -1085,7 +1083,7 @@ TEST_P(CompressedSecCacheTestWithTiered, CacheReservationManager) {
 TEST_P(CompressedSecCacheTestWithTiered,
       CacheReservationManagerMultipleUpdate) {
  CompressedSecondaryCache* sec_cache =
-      static_cast<CompressedSecondaryCache*>(GetSecondaryCache());
+      reinterpret_cast<CompressedSecondaryCache*>(GetSecondaryCache());

  EXPECT_PRED3(CacheUsageWithinBounds, GetCache()->GetUsage(), (30 << 20),
               GetPercent(30 << 20, 1));
@@ -1171,7 +1169,7 @@ TEST_P(CompressedSecCacheTestWithTiered, AdmissionPolicy) {

 TEST_P(CompressedSecCacheTestWithTiered, DynamicUpdate) {
  CompressedSecondaryCache* sec_cache =
-      static_cast<CompressedSecondaryCache*>(GetSecondaryCache());
+      reinterpret_cast<CompressedSecondaryCache*>(GetSecondaryCache());
  std::shared_ptr<Cache> tiered_cache = GetTieredCache();

  // Use EXPECT_PRED3 instead of EXPECT_NEAR to void too many size_t to
@@ -1226,16 +1224,18 @@ TEST_P(CompressedSecCacheTestWithTiered, DynamicUpdate) {
  ASSERT_OK(sec_cache->GetCapacity(sec_capacity));
  ASSERT_EQ(sec_capacity, 0);

-  ASSERT_OK(UpdateTieredCache(tiered_cache, -1, 0.3));
-  EXPECT_PRED3(CacheUsageWithinBounds, GetCache()->GetUsage(), (30 << 20),
-               GetPercent(30 << 20, 1));
+  ASSERT_NOK(UpdateTieredCache(tiered_cache, -1, 0.3));
+  // Only check usage for LRU cache. HCC shows a 64KB usage for some reason
+  if (std::get<0>(GetParam()) == PrimaryCacheType::kCacheTypeLRU) {
+    ASSERT_EQ(GetCache()->GetUsage(), 0);
+  }
  ASSERT_OK(sec_cache->GetCapacity(sec_capacity));
-  ASSERT_EQ(sec_capacity, (30 << 20));
+  ASSERT_EQ(sec_capacity, 0);
 }

 TEST_P(CompressedSecCacheTestWithTiered, DynamicUpdateWithReservation) {
  CompressedSecondaryCache* sec_cache =
-      static_cast<CompressedSecondaryCache*>(GetSecondaryCache());
+      reinterpret_cast<CompressedSecondaryCache*>(GetSecondaryCache());
  std::shared_ptr<Cache> tiered_cache = GetTieredCache();

  ASSERT_OK(cache_res_mgr()->UpdateCacheReservation(10 << 20));
@@ -1316,50 +1316,28 @@ TEST_P(CompressedSecCacheTestWithTiered, DynamicUpdateWithReservation) {
  ASSERT_OK(sec_cache->GetCapacity(sec_capacity));
  ASSERT_EQ(sec_capacity, 0);

-  ASSERT_OK(UpdateTieredCache(tiered_cache, -1, 0.3));
-  EXPECT_PRED3(CacheUsageWithinBounds, GetCache()->GetUsage(), (37 << 20),
-               GetPercent(37 << 20, 1));
-  EXPECT_PRED3(CacheUsageWithinBounds, sec_cache->TEST_GetUsage(), (3 << 20),
-               GetPercent(3 << 20, 1));
-  ASSERT_OK(sec_cache->GetCapacity(sec_capacity));
-  ASSERT_EQ(sec_capacity, 30 << 20);
-
  ASSERT_OK(cache_res_mgr()->UpdateCacheReservation(0));
 }

-TEST_P(CompressedSecCacheTestWithTiered, ReservationOverCapacity) {
-  CompressedSecondaryCache* sec_cache =
-      static_cast<CompressedSecondaryCache*>(GetSecondaryCache());
+TEST_P(CompressedSecCacheTestWithTiered,
+       DynamicUpdateWithReservationUnderflow) {
  std::shared_ptr<Cache> tiered_cache = GetTieredCache();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
+      {{"CacheWithSecondaryAdapter::Release:ChargeSecCache1",
+        "CacheWithSecondaryAdapter::UpdateCacheReservationRatio:Begin"},
+       {"CacheWithSecondaryAdapter::UpdateCacheReservationRatio:End",
+        "CacheWithSecondaryAdapter::Release:ChargeSecCache2"}});
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();

-  ASSERT_OK(cache_res_mgr()->UpdateCacheReservation(110 << 20));
-  // Use EXPECT_PRED3 instead of EXPECT_NEAR to void too many size_t to
-  // double explicit casts
-  EXPECT_PRED3(CacheUsageWithinBounds, GetCache()->GetUsage(), (110 << 20),
-               GetPercent(110 << 20, 1));
-  EXPECT_PRED3(CacheUsageWithinBounds, sec_cache->TEST_GetUsage(), (30 << 20),
-               GetPercent(30 << 20, 1));
-  size_t sec_capacity;
-  ASSERT_OK(sec_cache->GetCapacity(sec_capacity));
-  ASSERT_EQ(sec_capacity, (30 << 20));
+  port::Thread reserve_release_thread([&]() {
+    EXPECT_EQ(cache_res_mgr()->UpdateCacheReservation(50), Status::OK());
+    EXPECT_EQ(cache_res_mgr()->UpdateCacheReservation(0), Status::OK());
+  });
+  ASSERT_OK(UpdateTieredCache(tiered_cache, 100 << 20, 0.01));
+  reserve_release_thread.join();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();

-  ASSERT_OK(UpdateTieredCache(tiered_cache, -1, 0.39));
-  EXPECT_PRED3(CacheUsageWithinBounds, GetCache()->GetUsage(), (110 << 20),
-               GetPercent(110 << 20, 1));
-  EXPECT_PRED3(CacheUsageWithinBounds, sec_cache->TEST_GetUsage(), (39 << 20),
-               GetPercent(39 << 20, 1));
-  ASSERT_OK(sec_cache->GetCapacity(sec_capacity));
-  ASSERT_EQ(sec_capacity, (39 << 20));
-
-  ASSERT_OK(cache_res_mgr()->UpdateCacheReservation(90 << 20));
-  EXPECT_PRED3(CacheUsageWithinBounds, GetCache()->GetUsage(), (94 << 20),
-               GetPercent(94 << 20, 1));
-  EXPECT_PRED3(CacheUsageWithinBounds, sec_cache->TEST_GetUsage(), (35 << 20),
-               GetPercent(35 << 20, 1));
-  ASSERT_OK(sec_cache->GetCapacity(sec_capacity));
-  ASSERT_EQ(sec_capacity, (39 << 20));
-
-  ASSERT_OK(cache_res_mgr()->UpdateCacheReservation(0));
+  ASSERT_OK(UpdateTieredCache(tiered_cache, 100 << 20, 0.3));
 }

 INSTANTIATE_TEST_CASE_P(
@@ -339,7 +339,8 @@ void LRUCacheShard::NotifyEvicted(
  MemoryAllocator* alloc = table_.GetAllocator();
  for (LRUHandle* entry : evicted_handles) {
    if (eviction_callback_ &&
-        eviction_callback_(entry->key(), static_cast<Cache::Handle*>(entry),
+        eviction_callback_(entry->key(),
+                           reinterpret_cast<Cache::Handle*>(entry),
                           entry->HasHit())) {
      // Callback took ownership of obj; just free handle
      free(entry);
@@ -505,7 +506,7 @@ bool LRUCacheShard::Release(LRUHandle* e, bool /*useful*/,
    // Only call eviction callback if we're sure no one requested erasure
    // FIXME: disabled because of test churn
    if (false && was_in_cache && !erase_if_last_ref && eviction_callback_ &&
-        eviction_callback_(e->key(), static_cast<Cache::Handle*>(e),
+        eviction_callback_(e->key(), reinterpret_cast<Cache::Handle*>(e),
                           e->HasHit())) {
      // Callback took ownership of obj; just free handle
      free(e);
@@ -660,18 +661,18 @@ LRUCache::LRUCache(const LRUCacheOptions& opts) : ShardedCache(opts) {
 }

 Cache::ObjectPtr LRUCache::Value(Handle* handle) {
-  auto h = static_cast<const LRUHandle*>(handle);
+  auto h = reinterpret_cast<const LRUHandle*>(handle);
  return h->value;
 }

 size_t LRUCache::GetCharge(Handle* handle) const {
-  return static_cast<const LRUHandle*>(handle)->GetCharge(
+  return reinterpret_cast<const LRUHandle*>(handle)->GetCharge(
      GetShard(0).metadata_charge_policy_);
 }

 const Cache::CacheItemHelper* LRUCache::GetCacheItemHelper(
    Handle* handle) const {
-  auto h = static_cast<const LRUHandle*>(handle);
+  auto h = reinterpret_cast<const LRUHandle*>(handle);
  return h->helper;
 }

@@ -47,7 +47,7 @@ namespace lru_cache {
 // LRUCacheShard::Lookup.
 // While refs > 0, public properties like value and deleter must not change.

-struct LRUHandle : public Cache::Handle {
+struct LRUHandle {
  Cache::ObjectPtr value;
  const Cache::CacheItemHelper* helper;
  LRUHandle* next_hash;
@@ -47,7 +47,7 @@ class LRUCacheTest : public testing::Test {
                double low_pri_pool_ratio = 1.0,
                bool use_adaptive_mutex = kDefaultToAdaptiveMutex) {
    DeleteCache();
-    cache_ = static_cast<LRUCacheShard*>(
+    cache_ = reinterpret_cast<LRUCacheShard*>(
        port::cacheline_aligned_alloc(sizeof(LRUCacheShard)));
    new (cache_) LRUCacheShard(capacity, /*strict_capacity_limit=*/false,
                               high_pri_pool_ratio, low_pri_pool_ratio,
@@ -389,12 +389,12 @@ class ClockCacheTest : public testing::Test {
    }
  }

-  void NewShard(size_t capacity, bool strict_capacity_limit = true,
-                int eviction_effort_cap = 30) {
+  void NewShard(size_t capacity, bool strict_capacity_limit = true) {
    DeleteShard();
-    shard_ = static_cast<Shard*>(port::cacheline_aligned_alloc(sizeof(Shard)));
+    shard_ =
+        reinterpret_cast<Shard*>(port::cacheline_aligned_alloc(sizeof(Shard)));

-    TableOpts opts{1 /*value_size*/, eviction_effort_cap};
+    TableOpts opts{1 /*value_size*/};
    new (shard_)
        Shard(capacity, strict_capacity_limit, kDontChargeCacheMetadata,
              /*allocator*/ nullptr, &eviction_callback_, &hash_seed_, opts);
@@ -445,20 +445,12 @@ class ClockCacheTest : public testing::Test {
    return Slice(reinterpret_cast<const char*>(&hashed_key), 16U);
  }

-  // A bad hash function for testing / stressing collision handling
  static inline UniqueId64x2 TestHashedKey(char key) {
    // For testing hash near-collision behavior, put the variance in
    // hashed_key in bits that are unlikely to be used as hash bits.
    return {(static_cast<uint64_t>(key) << 56) + 1234U, 5678U};
  }

-  // A reasonable hash function, for testing "typical behavior" etc.
-  template <typename T>
-  static inline UniqueId64x2 CheapHash(T i) {
-    return {static_cast<uint64_t>(i) * uint64_t{0x85EBCA77C2B2AE63},
-            static_cast<uint64_t>(i) * uint64_t{0xC2B2AE3D27D4EB4F}};
-  }
-
  Shard* shard_ = nullptr;

 private:
@@ -691,53 +683,6 @@ TYPED_TEST(ClockCacheTest, ClockEvictionTest) {
  }
 }

-TYPED_TEST(ClockCacheTest, ClockEvictionEffortCapTest) {
-  using HandleImpl = typename ClockCacheTest<TypeParam>::Shard::HandleImpl;
-  for (bool strict_capacity_limit : {true, false}) {
-    SCOPED_TRACE("strict_capacity_limit = " +
-                 std::to_string(strict_capacity_limit));
-    for (int eec : {-42, 0, 1, 10, 100, 1000}) {
-      SCOPED_TRACE("eviction_effort_cap = " + std::to_string(eec));
-      constexpr size_t kCapacity = 1000;
-      // Start with much larger capacity to ensure that we can go way over
-      // capacity without reaching table occupancy limit.
-      this->NewShard(3 * kCapacity, strict_capacity_limit, eec);
-      auto& shard = *this->shard_;
-      shard.SetCapacity(kCapacity);
-
-      // Nearly fill the cache with pinned entries, then add a bunch of
-      // non-pinned entries. eviction_effort_cap should affect how many
-      // evictable entries are present beyond the cache capacity, despite
-      // being evictable.
-      constexpr size_t kCount = kCapacity - 1;
-      std::unique_ptr<HandleImpl* []> ha { new HandleImpl* [kCount] {} };
-      for (size_t i = 0; i < 2 * kCount; ++i) {
-        UniqueId64x2 hkey = this->CheapHash(i);
-        ASSERT_OK(shard.Insert(
-            this->TestKey(hkey), hkey, nullptr /*value*/, &kNoopCacheItemHelper,
-            1 /*charge*/, i < kCount ? &ha[i] : nullptr, Cache::Priority::LOW));
-      }
-
-      if (strict_capacity_limit) {
-        // If strict_capacity_limit is enabled, the cache will never exceed its
-        // capacity
-        EXPECT_EQ(shard.GetOccupancyCount(), kCapacity);
-      } else {
-        // Rough inverse relationship between cap and possible memory
-        // explosion, which shows up as increased table occupancy count.
-        int effective_eec = std::max(int{1}, eec) + 1;
-        EXPECT_NEAR(shard.GetOccupancyCount() * 1.0,
-                    kCount * (1 + 1.4 / effective_eec),
-                    kCount * (0.6 / effective_eec) + 1.0);
-      }
-
-      for (size_t i = 0; i < kCount; ++i) {
-        shard.Release(ha[i]);
-      }
-    }
-  }
-}
-
 namespace {
 struct DeleteCounter {
  int deleted = 0;
@@ -1090,8 +1035,7 @@ class TestSecondaryCache : public SecondaryCache {
  std::unique_ptr<SecondaryCacheResultHandle> Lookup(
      const Slice& key, const Cache::CacheItemHelper* helper,
      Cache::CreateContext* create_context, bool /*wait*/,
-      bool /*advise_erase*/, Statistics* /*stats*/,
-      bool& kept_in_sec_cache) override {
+      bool /*advise_erase*/, bool& kept_in_sec_cache) override {
    std::string key_str = key.ToString();
    TEST_SYNC_POINT_CALLBACK("TestSecondaryCache::Lookup", &key_str);

@@ -83,10 +83,7 @@ CacheWithSecondaryAdapter::CacheWithSecondaryAdapter(
    : CacheWrapper(std::move(target)),
      secondary_cache_(std::move(secondary_cache)),
      adm_policy_(adm_policy),
-      distribute_cache_res_(distribute_cache_res),
-      placeholder_usage_(0),
-      reserved_usage_(0),
-      sec_reserved_(0) {
+      distribute_cache_res_(distribute_cache_res) {
  target_->SetEvictionCallback(
      [this](const Slice& key, Handle* handle, bool was_hit) {
        return EvictionHandler(key, handle, was_hit);
@@ -106,7 +103,8 @@ CacheWithSecondaryAdapter::CacheWithSecondaryAdapter(
    // secondary cache is freed from the reservation.
    s = pri_cache_res_->UpdateCacheReservation(sec_capacity);
    assert(s.ok());
-    sec_cache_res_ratio_ = (double)sec_capacity / target_->GetCapacity();
+    sec_cache_res_ratio_.store((double)sec_capacity / target_->GetCapacity(),
+                               std::memory_order_relaxed);
  }
 }

@@ -115,12 +113,10 @@ CacheWithSecondaryAdapter::~CacheWithSecondaryAdapter() {
  // use after free
  target_->SetEvictionCallback({});
 #ifndef NDEBUG
-  if (distribute_cache_res_) {
+  if (distribute_cache_res_ && !ratio_changed_) {
    size_t sec_capacity = 0;
    Status s = secondary_cache_->GetCapacity(sec_capacity);
    assert(s.ok());
-    assert(placeholder_usage_ == 0);
-    assert(reserved_usage_ == 0);
    assert(pri_cache_res_->GetTotalMemoryUsed() == sec_capacity);
  }
 #endif  // NDEBUG
@@ -240,31 +236,13 @@ Status CacheWithSecondaryAdapter::Insert(const Slice& key, ObjectPtr value,
                                         const Slice& compressed_value,
                                         CompressionType type) {
  Status s = target_->Insert(key, value, helper, charge, handle, priority);
-  if (s.ok() && value == nullptr && distribute_cache_res_ && handle) {
-    charge = target_->GetCharge(*handle);
-
-    MutexLock l(&cache_res_mutex_);
-    placeholder_usage_ += charge;
-    // Check if total placeholder reservation is more than the overall
-    // cache capacity. If it is, then we don't try to charge the
-    // secondary cache because we don't want to overcharge it (beyond
-    // its capacity).
-    // In order to make this a bit more lightweight, we also check if
-    // the difference between placeholder_usage_ and reserved_usage_ is
-    // atleast kReservationChunkSize and avoid any adjustments if not.
-    if ((placeholder_usage_ <= target_->GetCapacity()) &&
-        ((placeholder_usage_ - reserved_usage_) >= kReservationChunkSize)) {
-      reserved_usage_ = placeholder_usage_ & ~(kReservationChunkSize - 1);
-      size_t new_sec_reserved =
-          static_cast<size_t>(reserved_usage_ * sec_cache_res_ratio_);
-      size_t sec_charge = new_sec_reserved - sec_reserved_;
-      s = secondary_cache_->Deflate(sec_charge);
-      assert(s.ok());
-      s = pri_cache_res_->UpdateCacheReservation(sec_charge,
-                                                 /*increase=*/false);
-      assert(s.ok());
-      sec_reserved_ += sec_charge;
-    }
+  if (s.ok() && value == nullptr && distribute_cache_res_) {
+    size_t sec_charge = static_cast<size_t>(
+        charge * (sec_cache_res_ratio_.load(std::memory_order_relaxed)));
+    s = secondary_cache_->Deflate(sec_charge);
+    assert(s.ok());
+    s = pri_cache_res_->UpdateCacheReservation(sec_charge, /*increase=*/false);
+    assert(s.ok());
  }
  // Warm up the secondary cache with the compressed block. The secondary
  // cache may choose to ignore it based on the admission policy.
@@ -294,8 +272,7 @@ Cache::Handle* CacheWithSecondaryAdapter::Lookup(const Slice& key,
    bool kept_in_sec_cache = false;
    std::unique_ptr<SecondaryCacheResultHandle> secondary_handle =
        secondary_cache_->Lookup(key, helper, create_context, /*wait*/ true,
-                                 found_dummy_entry, stats,
-                                 /*out*/ kept_in_sec_cache);
+                                 found_dummy_entry, /*out*/ kept_in_sec_cache);
    if (secondary_handle) {
      result = Promote(std::move(secondary_handle), key, helper, priority,
                       stats, found_dummy_entry, kept_in_sec_cache);
@@ -310,27 +287,14 @@ bool CacheWithSecondaryAdapter::Release(Handle* handle,
    ObjectPtr v = target_->Value(handle);
    if (v == nullptr && distribute_cache_res_) {
      size_t charge = target_->GetCharge(handle);
-
-      MutexLock l(&cache_res_mutex_);
-      placeholder_usage_ -= charge;
-      // Check if total placeholder reservation is more than the overall
-      // cache capacity. If it is, then we do nothing as reserved_usage_ must
-      // be already maxed out
-      if ((placeholder_usage_ <= target_->GetCapacity()) &&
-          (placeholder_usage_ < reserved_usage_)) {
-        // Adjust reserved_usage_ in chunks of kReservationChunkSize, so
-        // we don't hit this slow path too often.
-        reserved_usage_ = placeholder_usage_ & ~(kReservationChunkSize - 1);
-        size_t new_sec_reserved =
-            static_cast<size_t>(reserved_usage_ * sec_cache_res_ratio_);
-        size_t sec_charge = sec_reserved_ - new_sec_reserved;
-        Status s = secondary_cache_->Inflate(sec_charge);
-        assert(s.ok());
-        s = pri_cache_res_->UpdateCacheReservation(sec_charge,
-                                                   /*increase=*/true);
-        assert(s.ok());
-        sec_reserved_ -= sec_charge;
-      }
+      size_t sec_charge = static_cast<size_t>(
+          charge * (sec_cache_res_ratio_.load(std::memory_order_relaxed)));
+      TEST_SYNC_POINT("CacheWithSecondaryAdapter::Release:ChargeSecCache1");
+      TEST_SYNC_POINT("CacheWithSecondaryAdapter::Release:ChargeSecCache2");
+      Status s = secondary_cache_->Inflate(sec_charge);
+      assert(s.ok());
+      s = pri_cache_res_->UpdateCacheReservation(sec_charge, /*increase=*/true);
+      assert(s.ok());
    }
  }
  return target_->Release(handle, erase_if_last_ref);
@@ -349,10 +313,10 @@ void CacheWithSecondaryAdapter::StartAsyncLookupOnMySecondary(
  assert(async_handle.result_handle == nullptr);

  std::unique_ptr<SecondaryCacheResultHandle> secondary_handle =
-      secondary_cache_->Lookup(
-          async_handle.key, async_handle.helper, async_handle.create_context,
-          /*wait*/ false, async_handle.found_dummy_entry, async_handle.stats,
-          /*out*/ async_handle.kept_in_sec_cache);
+      secondary_cache_->Lookup(async_handle.key, async_handle.helper,
+                               async_handle.create_context, /*wait*/ false,
+                               async_handle.found_dummy_entry,
+                               /*out*/ async_handle.kept_in_sec_cache);
  if (secondary_handle) {
    // TODO with stacked secondaries: Check & process if already ready?
    async_handle.pending_handle = secondary_handle.release();
@@ -477,11 +441,13 @@ const char* CacheWithSecondaryAdapter::Name() const {
 // where the new capacity < total cache reservations.
 void CacheWithSecondaryAdapter::SetCapacity(size_t capacity) {
  size_t sec_capacity = static_cast<size_t>(
-      capacity * (distribute_cache_res_ ? sec_cache_res_ratio_ : 0.0));
+      capacity * (distribute_cache_res_
+                      ? sec_cache_res_ratio_.load(std::memory_order_relaxed)
+                      : 0.0));
  size_t old_sec_capacity = 0;

  if (distribute_cache_res_) {
-    MutexLock m(&cache_res_mutex_);
+    MutexLock m(&mutex_);

    Status s = secondary_cache_->GetCapacity(old_sec_capacity);
    if (!s.ok()) {
@@ -496,17 +462,9 @@ void CacheWithSecondaryAdapter::SetCapacity(size_t capacity) {
      // 3. Decrease the primary cache capacity to the total budget
      s = secondary_cache_->SetCapacity(sec_capacity);
      if (s.ok()) {
-        if (placeholder_usage_ > capacity) {
-          // Adjust reserved_usage_ down
-          reserved_usage_ = capacity & ~(kReservationChunkSize - 1);
-        }
-        size_t new_sec_reserved =
-            static_cast<size_t>(reserved_usage_ * sec_cache_res_ratio_);
        s = pri_cache_res_->UpdateCacheReservation(
-            (old_sec_capacity - sec_capacity) -
-                (sec_reserved_ - new_sec_reserved),
+            old_sec_capacity - sec_capacity,
            /*increase=*/false);
-        sec_reserved_ = new_sec_reserved;
        assert(s.ok());
        target_->SetCapacity(capacity);
      }
@@ -540,7 +498,7 @@ Status CacheWithSecondaryAdapter::GetSecondaryCachePinnedUsage(
    size_t& size) const {
  Status s;
  if (distribute_cache_res_) {
-    MutexLock m(&cache_res_mutex_);
+    MutexLock m(&mutex_);
    size_t capacity = 0;
    s = secondary_cache_->GetCapacity(capacity);
    if (s.ok()) {
@@ -568,11 +526,12 @@ Status CacheWithSecondaryAdapter::GetSecondaryCachePinnedUsage(
 // in the future.
 Status CacheWithSecondaryAdapter::UpdateCacheReservationRatio(
    double compressed_secondary_ratio) {
-  if (!distribute_cache_res_) {
+  if (!distribute_cache_res_ ||
+      sec_cache_res_ratio_.load(std::memory_order_relaxed) == 0.0) {
    return Status::NotSupported();
  }

-  MutexLock m(&cache_res_mutex_);
+  MutexLock m(&mutex_);
  size_t pri_capacity = target_->GetCapacity();
  size_t sec_capacity =
      static_cast<size_t>(pri_capacity * compressed_secondary_ratio);
@@ -582,17 +541,38 @@ Status CacheWithSecondaryAdapter::UpdateCacheReservationRatio(
    return s;
  }

+  TEST_SYNC_POINT(
+      "CacheWithSecondaryAdapter::UpdateCacheReservationRatio:Begin");
+
+  // There's a possible race condition here. Since the read of pri_cache_res_
+  // memory used (secondary cache usage charged to primary cache), and the
+  // change to sec_cache_res_ratio_ are not guarded by a mutex, its possible
+  // that an Insert/Release in another thread might decrease/increase the
+  // pri_cache_res_ reservation by the wrong amount. This should not be a
+  // problem because updating the sec/pri ratio is a rare operation, and
+  // the worst that can happen is we may over/under charge the secondary
+  // cache usage by a little bit. But we do need to protect against
+  // underflow of old_sec_reserved.
+  // TODO: Make the accounting more accurate by tracking the total memory
+  // reservation on the primary cache. This will also allow us to remove
+  // the restriction of not being able to change the sec/pri ratio from
+  // 0.0 to higher.
+  size_t sec_charge_to_pri = pri_cache_res_->GetTotalMemoryUsed();
+  size_t old_sec_reserved = (old_sec_capacity > sec_charge_to_pri)
+                                ? (old_sec_capacity - sec_charge_to_pri)
+                                : 0;
  // Calculate the new secondary cache reservation
-  // reserved_usage_ will never be > the cache capacity, so we don't
-  // have to worry about adjusting it here.
-  sec_cache_res_ratio_ = compressed_secondary_ratio;
-  size_t new_sec_reserved =
-      static_cast<size_t>(reserved_usage_ * sec_cache_res_ratio_);
+  size_t sec_reserved = static_cast<size_t>(
+      old_sec_reserved *
+      (double)(compressed_secondary_ratio /
+               sec_cache_res_ratio_.load(std::memory_order_relaxed)));
+  sec_cache_res_ratio_.store(compressed_secondary_ratio,
+                             std::memory_order_relaxed);
  if (sec_capacity > old_sec_capacity) {
    // We're increasing the ratio, thus ending up with a larger secondary
    // cache and a smaller usable primary cache capacity. Similar to
    // SetCapacity(), we try to avoid a temporary increase in total usage
-    // beyond the configured capacity -
+    // beyond teh configured capacity -
    // 1. A higher secondary cache ratio means it gets a higher share of
    //    cache reservations. So first account for that by deflating the
    //    secondary cache
@@ -600,13 +580,12 @@ Status CacheWithSecondaryAdapter::UpdateCacheReservationRatio(
    //    cache utilization (increase in capacity - increase in share of cache
    //    reservation)
    // 3. Increase secondary cache capacity
-    s = secondary_cache_->Deflate(new_sec_reserved - sec_reserved_);
+    s = secondary_cache_->Deflate(sec_reserved - old_sec_reserved);
    assert(s.ok());
    s = pri_cache_res_->UpdateCacheReservation(
-        (sec_capacity - old_sec_capacity) - (new_sec_reserved - sec_reserved_),
+        (sec_capacity - old_sec_capacity) - (sec_reserved - old_sec_reserved),
        /*increase=*/true);
    assert(s.ok());
-    sec_reserved_ = new_sec_reserved;
    s = secondary_cache_->SetCapacity(sec_capacity);
    assert(s.ok());
  } else {
@@ -620,16 +599,21 @@ Status CacheWithSecondaryAdapter::UpdateCacheReservationRatio(
    s = secondary_cache_->SetCapacity(sec_capacity);
    if (s.ok()) {
      s = pri_cache_res_->UpdateCacheReservation(
-          (old_sec_capacity - sec_capacity) -
-              (sec_reserved_ - new_sec_reserved),
+          (old_sec_capacity - sec_capacity) - (old_sec_reserved - sec_reserved),
          /*increase=*/false);
      assert(s.ok());
-      s = secondary_cache_->Inflate(sec_reserved_ - new_sec_reserved);
+      s = secondary_cache_->Inflate(old_sec_reserved - sec_reserved);
      assert(s.ok());
-      sec_reserved_ = new_sec_reserved;
    }
  }

+  TEST_SYNC_POINT("CacheWithSecondaryAdapter::UpdateCacheReservationRatio:End");
+#ifndef NDEBUG
+  // As mentioned in the function comments, we may accumulate some erros when
+  // the ratio is changed. We set a flag here which disables some assertions
+  // in the destructor
+  ratio_changed_ = true;
+#endif
  return s;
 }

@@ -684,14 +668,12 @@ std::shared_ptr<Cache> NewTieredCache(const TieredCacheOptions& _opts) {
        *(static_cast_with_check<LRUCacheOptions, ShardedCacheOptions>(
            opts.cache_opts));
    cache_opts.capacity = opts.total_capacity;
-    cache_opts.secondary_cache = nullptr;
    cache = cache_opts.MakeSharedCache();
  } else if (opts.cache_type == PrimaryCacheType::kCacheTypeHCC) {
    HyperClockCacheOptions cache_opts =
        *(static_cast_with_check<HyperClockCacheOptions, ShardedCacheOptions>(
            opts.cache_opts));
    cache_opts.capacity = opts.total_capacity;
-    cache_opts.secondary_cache = nullptr;
    cache = cache_opts.MakeSharedCache();
  } else {
    return nullptr;
@@ -60,8 +60,6 @@ class CacheWithSecondaryAdapter : public CacheWrapper {
  SecondaryCache* TEST_GetSecondaryCache() { return secondary_cache_.get(); }

 private:
-  static constexpr size_t kReservationChunkSize = 1 << 20;
-
  bool EvictionHandler(const Slice& key, Handle* handle, bool was_hit);

  void StartAsyncLookupOnMySecondary(AsyncLookupHandle& async_handle);
@@ -86,18 +84,11 @@ class CacheWithSecondaryAdapter : public CacheWrapper {
  std::shared_ptr<ConcurrentCacheReservationManager> pri_cache_res_;
  // Fraction of a cache memory reservation to be assigned to the secondary
  // cache
-  double sec_cache_res_ratio_;
-  // Mutex for use when managing cache memory reservations. Should not be used
-  // for other purposes, as it may risk causing deadlocks.
-  mutable port::Mutex cache_res_mutex_;
-  // Total memory reserved by placeholder entriesin the cache
-  size_t placeholder_usage_;
-  // Total placeholoder memory charged to both the primary and secondary
-  // caches. Will be <= placeholder_usage_.
-  size_t reserved_usage_;
-  // Amount of memory reserved in the secondary cache. This should be
-  // reserved_usage_ * sec_cache_res_ratio_ in steady state.
-  size_t sec_reserved_;
+  std::atomic<double> sec_cache_res_ratio_;
+  mutable port::Mutex mutex_;
+#ifndef NDEBUG
+  bool ratio_changed_ = false;
+#endif
 };

 }  // namespace ROCKSDB_NAMESPACE
@@ -139,7 +139,7 @@ class ShardedCache : public ShardedCacheBase {

  explicit ShardedCache(const ShardedCacheOptions& opts)
      : ShardedCacheBase(opts),
-        shards_(static_cast<CacheShard*>(port::cacheline_aligned_alloc(
+        shards_(reinterpret_cast<CacheShard*>(port::cacheline_aligned_alloc(
            sizeof(CacheShard) * GetNumShards()))),
        destroy_shards_in_dtor_(false) {}

@@ -192,7 +192,7 @@ class ShardedCache : public ShardedCacheBase {
    HashVal hash = CacheShard::ComputeHash(key, hash_seed_);
    HandleImpl* result = GetShard(hash).CreateStandalone(
        key, hash, obj, helper, charge, allow_uncharged);
-    return static_cast<Handle*>(result);
+    return reinterpret_cast<Handle*>(result);
  }

  Handle* Lookup(const Slice& key, const CacheItemHelper* helper = nullptr,
@@ -202,7 +202,7 @@ class ShardedCache : public ShardedCacheBase {
    HashVal hash = CacheShard::ComputeHash(key, hash_seed_);
    HandleImpl* result = GetShard(hash).Lookup(key, hash, helper,
                                               create_context, priority, stats);
-    return static_cast<Handle*>(result);
+    return reinterpret_cast<Handle*>(result);
  }

  void Erase(const Slice& key) override {
@@ -212,11 +212,11 @@ class ShardedCache : public ShardedCacheBase {

  bool Release(Handle* handle, bool useful,
               bool erase_if_last_ref = false) override {
-    auto h = static_cast<HandleImpl*>(handle);
+    auto h = reinterpret_cast<HandleImpl*>(handle);
    return GetShard(h->GetHash()).Release(h, useful, erase_if_last_ref);
  }
  bool Ref(Handle* handle) override {
-    auto h = static_cast<HandleImpl*>(handle);
+    auto h = reinterpret_cast<HandleImpl*>(handle);
    return GetShard(h->GetHash()).Ref(h);
  }
  bool Release(Handle* handle, bool erase_if_last_ref = false) override {
@@ -259,7 +259,7 @@ class ShardedCache : public ShardedCacheBase {
    } while (remaining_work);
  }

-  void EraseUnRefEntries() override {
+  virtual void EraseUnRefEntries() override {
    ForEachShard([](CacheShard* cs) { cs->EraseUnRefEntries(); });
  }

@@ -5,8 +5,6 @@

 #include "cache/tiered_secondary_cache.h"

-#include "monitoring/statistics_impl.h"
-
 namespace ROCKSDB_NAMESPACE {

 // Creation callback for use in the lookup path. It calls the upper layer
@@ -31,9 +29,6 @@ Status TieredSecondaryCache::MaybeInsertAndCreate(
    // TODO: Don't hardcode the source
    context->comp_sec_cache->InsertSaved(*context->key, data, type, source)
        .PermitUncheckedError();
-    RecordTick(context->stats, COMPRESSED_SECONDARY_CACHE_PROMOTIONS);
-  } else {
-    RecordTick(context->stats, COMPRESSED_SECONDARY_CACHE_PROMOTION_SKIPS);
  }
  // Primary cache will accept the object, so call its helper to create
  // the object
@@ -48,10 +43,10 @@ Status TieredSecondaryCache::MaybeInsertAndCreate(
 std::unique_ptr<SecondaryCacheResultHandle> TieredSecondaryCache::Lookup(
    const Slice& key, const Cache::CacheItemHelper* helper,
    Cache::CreateContext* create_context, bool wait, bool advise_erase,
-    Statistics* stats, bool& kept_in_sec_cache) {
+    bool& kept_in_sec_cache) {
  bool dummy = false;
  std::unique_ptr<SecondaryCacheResultHandle> result =
-      target()->Lookup(key, helper, create_context, wait, advise_erase, stats,
+      target()->Lookup(key, helper, create_context, wait, advise_erase,
                       /*kept_in_sec_cache=*/dummy);
  // We never want the item to spill back into the secondary cache
  kept_in_sec_cache = true;
@@ -71,10 +66,9 @@ std::unique_ptr<SecondaryCacheResultHandle> TieredSecondaryCache::Lookup(
    ctx.helper = helper;
    ctx.inner_ctx = create_context;
    ctx.comp_sec_cache = target();
-    ctx.stats = stats;

    return nvm_sec_cache_->Lookup(key, outer_helper, &ctx, wait, advise_erase,
-                                  stats, kept_in_sec_cache);
+                                  kept_in_sec_cache);
  }

  // If wait is false, i.e its an async lookup, we have to allocate a result
@@ -86,10 +80,8 @@ std::unique_ptr<SecondaryCacheResultHandle> TieredSecondaryCache::Lookup(
  handle->ctx()->helper = helper;
  handle->ctx()->inner_ctx = create_context;
  handle->ctx()->comp_sec_cache = target();
-  handle->ctx()->stats = stats;
-  handle->SetInnerHandle(
-      nvm_sec_cache_->Lookup(key, outer_helper, handle->ctx(), wait,
-                             advise_erase, stats, kept_in_sec_cache));
+  handle->SetInnerHandle(nvm_sec_cache_->Lookup(
+      key, outer_helper, handle->ctx(), wait, advise_erase, kept_in_sec_cache));
  if (!handle->inner_handle()) {
    handle.reset();
  } else {
@@ -117,8 +109,10 @@ void TieredSecondaryCache::WaitAll(
  }
  nvm_sec_cache_->WaitAll(nvm_handles);
  for (auto handle : my_handles) {
-    assert(handle->inner_handle()->IsReady());
-    handle->Complete();
+    assert(handle->IsReady());
+    auto nvm_handle = handle->inner_handle();
+    handle->SetSize(nvm_handle->Size());
+    handle->SetValue(nvm_handle->Value());
  }
 }

@@ -42,25 +42,27 @@ class TieredSecondaryCache : public SecondaryCacheWrapper {

  // This is a no-op as we currently don't allow demotion (i.e
  // insertion by the upper layer) of evicted blocks.
-  Status Insert(const Slice& /*key*/, Cache::ObjectPtr /*obj*/,
-                const Cache::CacheItemHelper* /*helper*/,
-                bool /*force_insert*/) override {
+  virtual Status Insert(const Slice& /*key*/, Cache::ObjectPtr /*obj*/,
+                        const Cache::CacheItemHelper* /*helper*/,
+                        bool /*force_insert*/) override {
    return Status::OK();
  }

  // Warm up the nvm tier directly
-  Status InsertSaved(const Slice& key, const Slice& saved,
-                     CompressionType type = CompressionType::kNoCompression,
-                     CacheTier source = CacheTier::kVolatileTier) override {
+  virtual Status InsertSaved(
+      const Slice& key, const Slice& saved,
+      CompressionType type = CompressionType::kNoCompression,
+      CacheTier source = CacheTier::kVolatileTier) override {
    return nvm_sec_cache_->InsertSaved(key, saved, type, source);
  }

-  std::unique_ptr<SecondaryCacheResultHandle> Lookup(
+  virtual std::unique_ptr<SecondaryCacheResultHandle> Lookup(
      const Slice& key, const Cache::CacheItemHelper* helper,
      Cache::CreateContext* create_context, bool wait, bool advise_erase,
-      Statistics* stats, bool& kept_in_sec_cache) override;
+      bool& kept_in_sec_cache) override;

-  void WaitAll(std::vector<SecondaryCacheResultHandle*> handles) override;
+  virtual void WaitAll(
+      std::vector<SecondaryCacheResultHandle*> handles) override;

 private:
  struct CreateContext : public Cache::CreateContext {
@@ -70,7 +72,6 @@ class TieredSecondaryCache : public SecondaryCacheWrapper {
    Cache::CreateContext* inner_ctx;
    std::shared_ptr<SecondaryCacheResultHandle> inner_handle;
    SecondaryCache* comp_sec_cache;
-    Statistics* stats;
  };

  class ResultHandle : public SecondaryCacheResultHandle {
@@ -78,10 +79,7 @@ class TieredSecondaryCache : public SecondaryCacheWrapper {
    ~ResultHandle() override {}

    bool IsReady() override {
-      if (inner_handle_ && inner_handle_->IsReady()) {
-        Complete();
-      }
-      return ready_;
+      return !inner_handle_ || inner_handle_->IsReady();
    }

    void Wait() override {
@@ -94,10 +92,10 @@ class TieredSecondaryCache : public SecondaryCacheWrapper {
    Cache::ObjectPtr Value() override { return value_; }

    void Complete() {
+      assert(IsReady());
      size_ = inner_handle_->Size();
      value_ = inner_handle_->Value();
      inner_handle_.reset();
-      ready_ = true;
    }

    void SetInnerHandle(std::unique_ptr<SecondaryCacheResultHandle>&& handle) {
@@ -117,7 +115,6 @@ class TieredSecondaryCache : public SecondaryCacheWrapper {
    CreateContext ctx_;
    size_t size_;
    Cache::ObjectPtr value_;
-    bool ready_ = false;
  };

  static void NoopDelete(Cache::ObjectPtr /*obj*/,
@@ -15,11 +15,10 @@ namespace ROCKSDB_NAMESPACE {

 class TestSecondaryCache : public SecondaryCache {
 public:
-  explicit TestSecondaryCache(size_t capacity, bool ready_before_wait)
+  explicit TestSecondaryCache(size_t capacity)
      : cache_(NewLRUCache(capacity, 0, false, 0.5 /* high_pri_pool_ratio */,
                           nullptr, kDefaultToAdaptiveMutex,
                           kDontChargeCacheMetadata)),
-        ready_before_wait_(ready_before_wait),
        num_insert_saved_(0),
        num_hits_(0),
        num_misses_(0) {}
@@ -62,7 +61,7 @@ class TestSecondaryCache : public SecondaryCache {
  std::unique_ptr<SecondaryCacheResultHandle> Lookup(
      const Slice& key, const Cache::CacheItemHelper* helper,
      Cache::CreateContext* create_context, bool wait, bool /*advise_erase*/,
-      Statistics* /*stats*/, bool& kept_in_sec_cache) override {
+      bool& kept_in_sec_cache) override {
    std::string key_str = key.ToString();
    TEST_SYNC_POINT_CALLBACK("TestSecondaryCache::Lookup", &key_str);

@@ -89,8 +88,7 @@ class TestSecondaryCache : public SecondaryCache {
                            /*alloc*/ nullptr, &value, &charge);
      if (s.ok()) {
        secondary_handle.reset(new TestSecondaryCacheResultHandle(
-            cache_.get(), handle, value, charge,
-            /*ready=*/wait || ready_before_wait_));
+            cache_.get(), handle, value, charge, /*ready=*/wait));
        kept_in_sec_cache = true;
      } else {
        cache_.Release(handle);
@@ -170,7 +168,6 @@ class TestSecondaryCache : public SecondaryCache {
      BasicTypedSharedCacheInterface<char[], CacheEntryRole::kMisc>;
  using TypedHandle = SharedCache::TypedHandle;
  SharedCache cache_;
-  bool ready_before_wait_;
  uint32_t num_insert_saved_;
  uint32_t num_hits_;
  uint32_t num_misses_;
@@ -182,10 +179,11 @@ class DBTieredSecondaryCacheTest : public DBTestBase {
  DBTieredSecondaryCacheTest()
      : DBTestBase("db_tiered_secondary_cache_test", /*env_do_fsync=*/true) {}

-  std::shared_ptr<Cache> NewCache(
-      size_t pri_capacity, size_t compressed_capacity, size_t nvm_capacity,
-      TieredAdmissionPolicy adm_policy = TieredAdmissionPolicy::kAdmPolicyAuto,
-      bool ready_before_wait = false) {
+  std::shared_ptr<Cache> NewCache(size_t pri_capacity,
+                                  size_t compressed_capacity,
+                                  size_t nvm_capacity,
+                                  TieredAdmissionPolicy adm_policy =
+                                      TieredAdmissionPolicy::kAdmPolicyAuto) {
    LRUCacheOptions lru_opts;
    TieredCacheOptions opts;
    lru_opts.capacity = 0;
@@ -196,11 +194,10 @@ class DBTieredSecondaryCacheTest : public DBTestBase {
    opts.comp_cache_opts.capacity = 0;
    opts.comp_cache_opts.num_shard_bits = 0;
    opts.total_capacity = pri_capacity + compressed_capacity;
-    opts.compressed_secondary_ratio = compressed_secondary_ratio_ =
+    opts.compressed_secondary_ratio =
        (double)compressed_capacity / opts.total_capacity;
    if (nvm_capacity > 0) {
-      nvm_sec_cache_.reset(
-          new TestSecondaryCache(nvm_capacity, ready_before_wait));
+      nvm_sec_cache_.reset(new TestSecondaryCache(nvm_capacity));
      opts.nvm_sec_cache = nvm_sec_cache_;
    }
    opts.adm_policy = adm_policy;
@@ -210,12 +207,6 @@ class DBTieredSecondaryCacheTest : public DBTestBase {
    return cache_;
  }

-  void ClearPrimaryCache() {
-    ASSERT_EQ(UpdateTieredCache(cache_, -1, 1.0), Status::OK());
-    ASSERT_EQ(UpdateTieredCache(cache_, -1, compressed_secondary_ratio_),
-              Status::OK());
-  }
-
  TestSecondaryCache* nvm_sec_cache() { return nvm_sec_cache_.get(); }

  CompressedSecondaryCache* compressed_secondary_cache() {
@@ -227,7 +218,6 @@ class DBTieredSecondaryCacheTest : public DBTestBase {
 private:
  std::shared_ptr<Cache> cache_;
  std::shared_ptr<TestSecondaryCache> nvm_sec_cache_;
-  double compressed_secondary_ratio_;
 };

 // In this test, the block size is set to 4096. Each value is 1007 bytes, so
@@ -592,116 +582,6 @@ TEST_F(DBTieredSecondaryCacheTest, WaitAllTest) {
  Destroy(options);
 }

-TEST_F(DBTieredSecondaryCacheTest, ReadyBeforeWaitAllTest) {
-  if (!LZ4_Supported()) {
-    ROCKSDB_GTEST_SKIP("This test requires LZ4 support.");
-    return;
-  }
-
-  BlockBasedTableOptions table_options;
-  table_options.block_cache = NewCache(250 * 1024, 20 * 1024, 256 * 1024,
-                                       TieredAdmissionPolicy::kAdmPolicyAuto,
-                                       /*ready_before_wait=*/true);
-  table_options.block_size = 4 * 1024;
-  table_options.cache_index_and_filter_blocks = false;
-  Options options = GetDefaultOptions();
-  options.create_if_missing = true;
-  options.table_factory.reset(NewBlockBasedTableFactory(table_options));
-  options.statistics = CreateDBStatistics();
-
-  options.paranoid_file_checks = false;
-  DestroyAndReopen(options);
-  Random rnd(301);
-  const int N = 256;
-  for (int i = 0; i < N; i++) {
-    std::string p_v;
-    test::CompressibleString(&rnd, 0.5, 1007, &p_v);
-    ASSERT_OK(Put(Key(i), p_v));
-  }
-
-  ASSERT_OK(Flush());
-
-  std::vector<std::string> keys;
-  std::vector<std::string> values;
-
-  keys.push_back(Key(0));
-  keys.push_back(Key(4));
-  keys.push_back(Key(8));
-  values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
-  ASSERT_EQ(values.size(), keys.size());
-  for (auto value : values) {
-    ASSERT_EQ(1007, value.size());
-  }
-  ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 3u);
-  ASSERT_EQ(nvm_sec_cache()->num_misses(), 3u);
-  ASSERT_EQ(nvm_sec_cache()->num_hits(), 0u);
-  ASSERT_EQ(options.statistics->getTickerCount(BLOCK_CACHE_MISS), 3u);
-
-  keys.clear();
-  values.clear();
-  keys.push_back(Key(12));
-  keys.push_back(Key(16));
-  keys.push_back(Key(20));
-  values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
-  ASSERT_EQ(values.size(), keys.size());
-  for (auto value : values) {
-    ASSERT_EQ(1007, value.size());
-  }
-  ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
-  ASSERT_EQ(nvm_sec_cache()->num_misses(), 6u);
-  ASSERT_EQ(nvm_sec_cache()->num_hits(), 0u);
-  ASSERT_EQ(options.statistics->getTickerCount(BLOCK_CACHE_MISS), 6u);
-
-  keys.clear();
-  values.clear();
-  keys.push_back(Key(0));
-  keys.push_back(Key(4));
-  keys.push_back(Key(8));
-  values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
-  ASSERT_EQ(values.size(), keys.size());
-  for (auto value : values) {
-    ASSERT_EQ(1007, value.size());
-  }
-  ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
-  ASSERT_EQ(nvm_sec_cache()->num_misses(), 6u);
-  ASSERT_EQ(nvm_sec_cache()->num_hits(), 3u);
-  ASSERT_EQ(options.statistics->getTickerCount(BLOCK_CACHE_MISS), 6u);
-
-  ClearPrimaryCache();
-
-  keys.clear();
-  values.clear();
-  keys.push_back(Key(0));
-  keys.push_back(Key(32));
-  keys.push_back(Key(36));
-  values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
-  ASSERT_EQ(values.size(), keys.size());
-  for (auto value : values) {
-    ASSERT_EQ(1007, value.size());
-  }
-  ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 8u);
-  ASSERT_EQ(nvm_sec_cache()->num_misses(), 8u);
-  ASSERT_EQ(nvm_sec_cache()->num_hits(), 4u);
-  ASSERT_EQ(options.statistics->getTickerCount(BLOCK_CACHE_MISS), 8u);
-
-  keys.clear();
-  values.clear();
-  keys.push_back(Key(0));
-  keys.push_back(Key(32));
-  keys.push_back(Key(36));
-  values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
-  ASSERT_EQ(values.size(), keys.size());
-  for (auto value : values) {
-    ASSERT_EQ(1007, value.size());
-  }
-  ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 8u);
-  ASSERT_EQ(nvm_sec_cache()->num_misses(), 8u);
-  ASSERT_EQ(nvm_sec_cache()->num_hits(), 4u);
-  ASSERT_EQ(options.statistics->getTickerCount(BLOCK_CACHE_MISS), 8u);
-
-  Destroy(options);
-}
-
 // This test is for iteration. It iterates through a set of keys in two
 // passes. First pass loads the compressed blocks into the nvm tier, and
 // the second pass should hit all of those blocks.
@@ -155,7 +155,7 @@ class BasicTypedCacheInterface : public BaseCacheInterface<CachePtr>,
  using BaseCacheInterface<CachePtr>::BaseCacheInterface;
  struct TypedAsyncLookupHandle : public Cache::AsyncLookupHandle {
    TypedHandle* Result() {
-      return static_cast<TypedHandle*>(Cache::AsyncLookupHandle::Result());
+      return reinterpret_cast<TypedHandle*>(Cache::AsyncLookupHandle::Result());
    }
  };

@@ -169,7 +169,8 @@ class BasicTypedCacheInterface : public BaseCacheInterface<CachePtr>,
  }

  inline TypedHandle* Lookup(const Slice& key, Statistics* stats = nullptr) {
-    return static_cast<TypedHandle*>(this->cache_->BasicLookup(key, stats));
+    return reinterpret_cast<TypedHandle*>(
+        this->cache_->BasicLookup(key, stats));
  }

  inline void StartAsyncLookup(TypedAsyncLookupHandle& async_handle) {
@@ -346,7 +347,7 @@ class FullTypedCacheInterface
      Priority priority = Priority::LOW, Statistics* stats = nullptr,
      CacheTier lowest_used_cache_tier = CacheTier::kNonVolatileBlockTier) {
    if (lowest_used_cache_tier > CacheTier::kVolatileTier) {
-      return static_cast<TypedHandle*>(this->cache_->Lookup(
+      return reinterpret_cast<TypedHandle*>(this->cache_->Lookup(
          key, GetFullHelper(), create_context, priority, stats));
    } else {
      return BasicTypedCacheInterface<TValue, kRole, CachePtr>::Lookup(key,
@@ -14,7 +14,7 @@ find_library(zstd_LIBRARIES
  HINTS ${zstd_ROOT_DIR}/lib)

 include(FindPackageHandleStandardArgs)
-find_package_handle_standard_args(zstd DEFAULT_MSG zstd_LIBRARIES ZSTD_INCLUDE_DIRS)
+find_package_handle_standard_args(zstd DEFAULT_MSG zstd_LIBRARIES zstd_INCLUDE_DIRS)

 mark_as_advanced(
  zstd_LIBRARIES
@@ -118,7 +118,7 @@ class ArenaWrappedDBIter : public Iterator {
 // Generate the arena wrapped iterator class.
 // `db_impl` and `cfd` are used for reneweal. If left null, renewal will not
 // be supported.
-ArenaWrappedDBIter* NewArenaWrappedDbIterator(
+extern ArenaWrappedDBIter* NewArenaWrappedDbIterator(
    Env* env, const ReadOptions& read_options, const ImmutableOptions& ioptions,
    const MutableCFOptions& mutable_cf_options, const Version* version,
    const SequenceNumber& sequence, uint64_t max_sequential_skip_in_iterations,
@@ -34,9 +34,9 @@ BlobFileBuilder::BlobFileBuilder(
    VersionSet* versions, FileSystem* fs,
    const ImmutableOptions* immutable_options,
    const MutableCFOptions* mutable_cf_options, const FileOptions* file_options,
-    const WriteOptions* write_options, std::string db_id,
-    std::string db_session_id, int job_id, uint32_t column_family_id,
-    const std::string& column_family_name, Env::WriteLifeTimeHint write_hint,
+    std::string db_id, std::string db_session_id, int job_id,
+    uint32_t column_family_id, const std::string& column_family_name,
+    Env::IOPriority io_priority, Env::WriteLifeTimeHint write_hint,
    const std::shared_ptr<IOTracer>& io_tracer,
    BlobFileCompletionCallback* blob_callback,
    BlobFileCreationReason creation_reason,
@@ -44,18 +44,18 @@ BlobFileBuilder::BlobFileBuilder(
    std::vector<BlobFileAddition>* blob_file_additions)
    : BlobFileBuilder([versions]() { return versions->NewFileNumber(); }, fs,
                      immutable_options, mutable_cf_options, file_options,
-                      write_options, db_id, db_session_id, job_id,
-                      column_family_id, column_family_name, write_hint,
-                      io_tracer, blob_callback, creation_reason,
-                      blob_file_paths, blob_file_additions) {}
+                      db_id, db_session_id, job_id, column_family_id,
+                      column_family_name, io_priority, write_hint, io_tracer,
+                      blob_callback, creation_reason, blob_file_paths,
+                      blob_file_additions) {}

 BlobFileBuilder::BlobFileBuilder(
    std::function<uint64_t()> file_number_generator, FileSystem* fs,
    const ImmutableOptions* immutable_options,
    const MutableCFOptions* mutable_cf_options, const FileOptions* file_options,
-    const WriteOptions* write_options, std::string db_id,
-    std::string db_session_id, int job_id, uint32_t column_family_id,
-    const std::string& column_family_name, Env::WriteLifeTimeHint write_hint,
+    std::string db_id, std::string db_session_id, int job_id,
+    uint32_t column_family_id, const std::string& column_family_name,
+    Env::IOPriority io_priority, Env::WriteLifeTimeHint write_hint,
    const std::shared_ptr<IOTracer>& io_tracer,
    BlobFileCompletionCallback* blob_callback,
    BlobFileCreationReason creation_reason,
@@ -69,12 +69,12 @@ BlobFileBuilder::BlobFileBuilder(
      blob_compression_type_(mutable_cf_options->blob_compression_type),
      prepopulate_blob_cache_(mutable_cf_options->prepopulate_blob_cache),
      file_options_(file_options),
-      write_options_(write_options),
      db_id_(std::move(db_id)),
      db_session_id_(std::move(db_session_id)),
      job_id_(job_id),
      column_family_id_(column_family_id),
      column_family_name_(column_family_name),
+      io_priority_(io_priority),
      write_hint_(write_hint),
      io_tracer_(io_tracer),
      blob_callback_(blob_callback),
@@ -87,7 +87,6 @@ BlobFileBuilder::BlobFileBuilder(
  assert(fs_);
  assert(immutable_options_);
  assert(file_options_);
-  assert(write_options_);
  assert(blob_file_paths_);
  assert(blob_file_paths_->empty());
  assert(blob_file_additions_);
@@ -208,14 +207,14 @@ Status BlobFileBuilder::OpenBlobFileIfNeeded() {
  blob_file_paths_->emplace_back(std::move(blob_file_path));

  assert(file);
-  file->SetIOPriority(write_options_->rate_limiter_priority);
+  file->SetIOPriority(io_priority_);
  file->SetWriteLifeTimeHint(write_hint_);
  FileTypeSet tmp_set = immutable_options_->checksum_handoff_file_types;
  Statistics* const statistics = immutable_options_->stats;
  std::unique_ptr<WritableFileWriter> file_writer(new WritableFileWriter(
      std::move(file), blob_file_paths_->back(), *file_options_,
      immutable_options_->clock, io_tracer_, statistics,
-      Histograms::BLOB_DB_BLOB_FILE_WRITE_MICROS, immutable_options_->listeners,
+      immutable_options_->listeners,
      immutable_options_->file_checksum_gen_factory.get(),
      tmp_set.Contains(FileType::kBlobFile), false));

@@ -232,7 +231,7 @@ Status BlobFileBuilder::OpenBlobFileIfNeeded() {
                       expiration_range);

  {
-    Status s = blob_log_writer->WriteHeader(*write_options_, header);
+    Status s = blob_log_writer->WriteHeader(header);

    TEST_SYNC_POINT_CALLBACK(
        "BlobFileBuilder::OpenBlobFileIfNeeded:WriteHeader", &s);
@@ -297,8 +296,7 @@ Status BlobFileBuilder::WriteBlobToFile(const Slice& key, const Slice& blob,

  uint64_t key_offset = 0;

-  Status s =
-      writer_->AddRecord(*write_options_, key, blob, &key_offset, blob_offset);
+  Status s = writer_->AddRecord(key, blob, &key_offset, blob_offset);

  TEST_SYNC_POINT_CALLBACK("BlobFileBuilder::WriteBlobToFile:AddRecord", &s);

@@ -323,8 +321,7 @@ Status BlobFileBuilder::CloseBlobFile() {
  std::string checksum_method;
  std::string checksum_value;

-  Status s = writer_->AppendFooter(*write_options_, footer, &checksum_method,
-                                   &checksum_value);
+  Status s = writer_->AppendFooter(footer, &checksum_method, &checksum_value);

  TEST_SYNC_POINT_CALLBACK("BlobFileBuilder::WriteBlobToFile:AppendFooter", &s);

@@ -13,7 +13,6 @@
 #include "rocksdb/advanced_options.h"
 #include "rocksdb/compression_type.h"
 #include "rocksdb/env.h"
-#include "rocksdb/options.h"
 #include "rocksdb/rocksdb_namespace.h"
 #include "rocksdb/types.h"

@@ -37,11 +36,11 @@ class BlobFileBuilder {
  BlobFileBuilder(VersionSet* versions, FileSystem* fs,
                  const ImmutableOptions* immutable_options,
                  const MutableCFOptions* mutable_cf_options,
-                  const FileOptions* file_options,
-                  const WriteOptions* write_options, std::string db_id,
+                  const FileOptions* file_options, std::string db_id,
                  std::string db_session_id, int job_id,
                  uint32_t column_family_id,
                  const std::string& column_family_name,
+                  Env::IOPriority io_priority,
                  Env::WriteLifeTimeHint write_hint,
                  const std::shared_ptr<IOTracer>& io_tracer,
                  BlobFileCompletionCallback* blob_callback,
@@ -52,11 +51,11 @@ class BlobFileBuilder {
  BlobFileBuilder(std::function<uint64_t()> file_number_generator,
                  FileSystem* fs, const ImmutableOptions* immutable_options,
                  const MutableCFOptions* mutable_cf_options,
-                  const FileOptions* file_options,
-                  const WriteOptions* write_options, std::string db_id,
+                  const FileOptions* file_options, std::string db_id,
                  std::string db_session_id, int job_id,
                  uint32_t column_family_id,
                  const std::string& column_family_name,
+                  Env::IOPriority io_priority,
                  Env::WriteLifeTimeHint write_hint,
                  const std::shared_ptr<IOTracer>& io_tracer,
                  BlobFileCompletionCallback* blob_callback,
@@ -93,12 +92,12 @@ class BlobFileBuilder {
  CompressionType blob_compression_type_;
  PrepopulateBlobCache prepopulate_blob_cache_;
  const FileOptions* file_options_;
-  const WriteOptions* write_options_;
  const std::string db_id_;
  const std::string db_session_id_;
  int job_id_;
  uint32_t column_family_id_;
  std::string column_family_name_;
+  Env::IOPriority io_priority_;
  Env::WriteLifeTimeHint write_hint_;
  std::shared_ptr<IOTracer> io_tracer_;
  BlobFileCompletionCallback* blob_callback_;
@@ -43,7 +43,6 @@ class BlobFileBuilderTest : public testing::Test {
    mock_env_.reset(MockEnv::Create(Env::Default()));
    fs_ = mock_env_->GetFileSystem().get();
    clock_ = mock_env_->GetSystemClock().get();
-    write_options_.rate_limiter_priority = Env::IO_HIGH;
  }

  void VerifyBlobFile(uint64_t blob_file_number,
@@ -114,7 +113,6 @@ class BlobFileBuilderTest : public testing::Test {
  FileSystem* fs_;
  SystemClock* clock_;
  FileOptions file_options_;
-  WriteOptions write_options_;
 };

 TEST_F(BlobFileBuilderTest, BuildAndCheckOneFile) {
@@ -138,6 +136,7 @@ TEST_F(BlobFileBuilderTest, BuildAndCheckOneFile) {
  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;
@@ -145,8 +144,8 @@ TEST_F(BlobFileBuilderTest, BuildAndCheckOneFile) {

  BlobFileBuilder builder(
      TestFileNumberGenerator(), fs_, &immutable_options, &mutable_cf_options,
-      &file_options_, &write_options_, "" /*db_id*/, "" /*db_session_id*/,
-      job_id, column_family_id, column_family_name, write_hint,
+      &file_options_, "" /*db_id*/, "" /*db_session_id*/, job_id,
+      column_family_id, column_family_name, io_priority, write_hint,
      nullptr /*IOTracer*/, nullptr /*BlobFileCompletionCallback*/,
      BlobFileCreationReason::kFlush, &blob_file_paths, &blob_file_additions);

@@ -222,6 +221,7 @@ TEST_F(BlobFileBuilderTest, BuildAndCheckMultipleFiles) {
  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;
@@ -229,8 +229,8 @@ TEST_F(BlobFileBuilderTest, BuildAndCheckMultipleFiles) {

  BlobFileBuilder builder(
      TestFileNumberGenerator(), fs_, &immutable_options, &mutable_cf_options,
-      &file_options_, &write_options_, "" /*db_id*/, "" /*db_session_id*/,
-      job_id, column_family_id, column_family_name, write_hint,
+      &file_options_, "" /*db_id*/, "" /*db_session_id*/, job_id,
+      column_family_id, column_family_name, io_priority, write_hint,
      nullptr /*IOTracer*/, nullptr /*BlobFileCompletionCallback*/,
      BlobFileCreationReason::kFlush, &blob_file_paths, &blob_file_additions);

@@ -309,6 +309,7 @@ TEST_F(BlobFileBuilderTest, InlinedValues) {
  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;
@@ -316,8 +317,8 @@ TEST_F(BlobFileBuilderTest, InlinedValues) {

  BlobFileBuilder builder(
      TestFileNumberGenerator(), fs_, &immutable_options, &mutable_cf_options,
-      &file_options_, &write_options_, "" /*db_id*/, "" /*db_session_id*/,
-      job_id, column_family_id, column_family_name, write_hint,
+      &file_options_, "" /*db_id*/, "" /*db_session_id*/, job_id,
+      column_family_id, column_family_name, io_priority, write_hint,
      nullptr /*IOTracer*/, nullptr /*BlobFileCompletionCallback*/,
      BlobFileCreationReason::kFlush, &blob_file_paths, &blob_file_additions);

@@ -363,6 +364,7 @@ TEST_F(BlobFileBuilderTest, Compression) {
  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;
@@ -370,8 +372,8 @@ TEST_F(BlobFileBuilderTest, Compression) {

  BlobFileBuilder builder(
      TestFileNumberGenerator(), fs_, &immutable_options, &mutable_cf_options,
-      &file_options_, &write_options_, "" /*db_id*/, "" /*db_session_id*/,
-      job_id, column_family_id, column_family_name, write_hint,
+      &file_options_, "" /*db_id*/, "" /*db_session_id*/, job_id,
+      column_family_id, column_family_name, io_priority, write_hint,
      nullptr /*IOTracer*/, nullptr /*BlobFileCompletionCallback*/,
      BlobFileCreationReason::kFlush, &blob_file_paths, &blob_file_additions);

@@ -446,6 +448,7 @@ TEST_F(BlobFileBuilderTest, CompressionError) {
  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;
@@ -453,8 +456,8 @@ TEST_F(BlobFileBuilderTest, CompressionError) {

  BlobFileBuilder builder(
      TestFileNumberGenerator(), fs_, &immutable_options, &mutable_cf_options,
-      &file_options_, &write_options_, "" /*db_id*/, "" /*db_session_id*/,
-      job_id, column_family_id, column_family_name, write_hint,
+      &file_options_, "" /*db_id*/, "" /*db_session_id*/, job_id,
+      column_family_id, column_family_name, io_priority, write_hint,
      nullptr /*IOTracer*/, nullptr /*BlobFileCompletionCallback*/,
      BlobFileCreationReason::kFlush, &blob_file_paths, &blob_file_additions);

@@ -525,6 +528,7 @@ TEST_F(BlobFileBuilderTest, Checksum) {
  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;
@@ -532,8 +536,8 @@ TEST_F(BlobFileBuilderTest, Checksum) {

  BlobFileBuilder builder(
      TestFileNumberGenerator(), fs_, &immutable_options, &mutable_cf_options,
-      &file_options_, &write_options_, "" /*db_id*/, "" /*db_session_id*/,
-      job_id, column_family_id, column_family_name, write_hint,
+      &file_options_, "" /*db_id*/, "" /*db_session_id*/, job_id,
+      column_family_id, column_family_name, io_priority, write_hint,
      nullptr /*IOTracer*/, nullptr /*BlobFileCompletionCallback*/,
      BlobFileCreationReason::kFlush, &blob_file_paths, &blob_file_additions);

@@ -585,13 +589,11 @@ class BlobFileBuilderIOErrorTest
  BlobFileBuilderIOErrorTest() : sync_point_(GetParam()) {
    mock_env_.reset(MockEnv::Create(Env::Default()));
    fs_ = mock_env_->GetFileSystem().get();
-    write_options_.rate_limiter_priority = Env::IO_HIGH;
  }

  std::unique_ptr<Env> mock_env_;
  FileSystem* fs_;
  FileOptions file_options_;
-  WriteOptions write_options_;
  std::string sync_point_;
 };

@@ -624,6 +626,7 @@ TEST_P(BlobFileBuilderIOErrorTest, IOError) {
  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;
@@ -631,8 +634,8 @@ TEST_P(BlobFileBuilderIOErrorTest, IOError) {

  BlobFileBuilder builder(
      TestFileNumberGenerator(), fs_, &immutable_options, &mutable_cf_options,
-      &file_options_, &write_options_, "" /*db_id*/, "" /*db_session_id*/,
-      job_id, column_family_id, column_family_name, write_hint,
+      &file_options_, "" /*db_id*/, "" /*db_session_id*/, job_id,
+      column_family_id, column_family_name, io_priority, write_hint,
      nullptr /*IOTracer*/, nullptr /*BlobFileCompletionCallback*/,
      BlobFileCreationReason::kFlush, &blob_file_paths, &blob_file_additions);

@@ -57,7 +57,7 @@ void WriteBlobFile(uint32_t column_family_id,
  BlobLogHeader header(column_family_id, kNoCompression, has_ttl,
                       expiration_range);

-  ASSERT_OK(blob_log_writer.WriteHeader(WriteOptions(), header));
+  ASSERT_OK(blob_log_writer.WriteHeader(header));

  constexpr char key[] = "key";
  constexpr char blob[] = "blob";
@@ -67,8 +67,7 @@ void WriteBlobFile(uint32_t column_family_id,
  uint64_t key_offset = 0;
  uint64_t blob_offset = 0;

-  ASSERT_OK(blob_log_writer.AddRecord(WriteOptions(), key, blob, &key_offset,
-                                      &blob_offset));
+  ASSERT_OK(blob_log_writer.AddRecord(key, blob, &key_offset, &blob_offset));

  BlobLogFooter footer;
  footer.blob_count = 1;
@@ -77,8 +76,8 @@ void WriteBlobFile(uint32_t column_family_id,
  std::string checksum_method;
  std::string checksum_value;

-  ASSERT_OK(blob_log_writer.AppendFooter(WriteOptions(), footer,
-                                         &checksum_method, &checksum_value));
+  ASSERT_OK(
+      blob_log_writer.AppendFooter(footer, &checksum_method, &checksum_value));
 }

 }  // anonymous namespace
@@ -63,7 +63,7 @@ void WriteBlobFile(const ImmutableOptions& immutable_options,
  BlobLogHeader header(column_family_id, compression, has_ttl,
                       expiration_range_header);

-  ASSERT_OK(blob_log_writer.WriteHeader(WriteOptions(), header));
+  ASSERT_OK(blob_log_writer.WriteHeader(header));

  std::vector<std::string> compressed_blobs(num);
  std::vector<Slice> blobs_to_write(num);
@@ -91,8 +91,7 @@ void WriteBlobFile(const ImmutableOptions& immutable_options,

  for (size_t i = 0; i < num; ++i) {
    uint64_t key_offset = 0;
-    ASSERT_OK(blob_log_writer.AddRecord(WriteOptions(), keys[i],
-                                        blobs_to_write[i], &key_offset,
+    ASSERT_OK(blob_log_writer.AddRecord(keys[i], blobs_to_write[i], &key_offset,
                                        &blob_offsets[i]));
  }

@@ -102,8 +101,8 @@ void WriteBlobFile(const ImmutableOptions& immutable_options,

  std::string checksum_method;
  std::string checksum_value;
-  ASSERT_OK(blob_log_writer.AppendFooter(WriteOptions(), footer,
-                                         &checksum_method, &checksum_value));
+  ASSERT_OK(
+      blob_log_writer.AppendFooter(footer, &checksum_method, &checksum_value));
 }

 // Creates a test blob file with a single blob in it. Note: this method
@@ -474,7 +473,7 @@ TEST_F(BlobFileReaderTest, Malformed) {
    BlobLogHeader header(column_family_id, kNoCompression, has_ttl,
                         expiration_range);

-    ASSERT_OK(blob_log_writer.WriteHeader(WriteOptions(), header));
+    ASSERT_OK(blob_log_writer.WriteHeader(header));
  }

  constexpr HistogramImpl* blob_file_read_hist = nullptr;
@@ -33,49 +33,35 @@ BlobLogWriter::BlobLogWriter(std::unique_ptr<WritableFileWriter>&& dest,

 BlobLogWriter::~BlobLogWriter() = default;

-Status BlobLogWriter::Sync(const WriteOptions& write_options) {
+Status BlobLogWriter::Sync() {
  TEST_SYNC_POINT("BlobLogWriter::Sync");

  StopWatch sync_sw(clock_, statistics_, BLOB_DB_BLOB_FILE_SYNC_MICROS);
-  IOOptions opts;
-  Status s = WritableFileWriter::PrepareIOOptions(write_options, opts);
-  if (s.ok()) {
-    s = dest_->Sync(opts, use_fsync_);
-  }
-  if (s.ok()) {
-    RecordTick(statistics_, BLOB_DB_BLOB_FILE_SYNCED);
-  }
+  Status s = dest_->Sync(use_fsync_);
+  RecordTick(statistics_, BLOB_DB_BLOB_FILE_SYNCED);
  return s;
 }

-Status BlobLogWriter::WriteHeader(const WriteOptions& write_options,
-                                  BlobLogHeader& header) {
+Status BlobLogWriter::WriteHeader(BlobLogHeader& header) {
  assert(block_offset_ == 0);
  assert(last_elem_type_ == kEtNone);
  std::string str;
  header.EncodeTo(&str);

-  IOOptions opts;
-  Status s = WritableFileWriter::PrepareIOOptions(write_options, opts);
-  if (s.ok()) {
-    s = dest_->Append(opts, Slice(str));
-  }
+  Status s = dest_->Append(Slice(str));
  if (s.ok()) {
    block_offset_ += str.size();
    if (do_flush_) {
-      s = dest_->Flush(opts);
+      s = dest_->Flush();
    }
  }
  last_elem_type_ = kEtFileHdr;
-  if (s.ok()) {
-    RecordTick(statistics_, BLOB_DB_BLOB_FILE_BYTES_WRITTEN,
-               BlobLogHeader::kSize);
-  }
+  RecordTick(statistics_, BLOB_DB_BLOB_FILE_BYTES_WRITTEN,
+             BlobLogHeader::kSize);
  return s;
 }

-Status BlobLogWriter::AppendFooter(const WriteOptions& write_options,
-                                   BlobLogFooter& footer,
+Status BlobLogWriter::AppendFooter(BlobLogFooter& footer,
                                   std::string* checksum_method,
                                   std::string* checksum_value) {
  assert(block_offset_ != 0);
@@ -89,17 +75,14 @@ Status BlobLogWriter::AppendFooter(const WriteOptions& write_options,
    s.PermitUncheckedError();
    return Status::IOError("Seen Error. Skip closing.");
  } else {
-    IOOptions opts;
-    s = WritableFileWriter::PrepareIOOptions(write_options, opts);
-    if (s.ok()) {
-      s = dest_->Append(opts, Slice(str));
-    }
+    s = dest_->Append(Slice(str));
    if (s.ok()) {
      block_offset_ += str.size();
-      s = Sync(write_options);
+
+      s = Sync();

      if (s.ok()) {
-        s = dest_->Close(opts);
+        s = dest_->Close();

        if (s.ok()) {
          assert(!!checksum_method == !!checksum_value);
@@ -128,15 +111,12 @@ Status BlobLogWriter::AppendFooter(const WriteOptions& write_options,
  }

  last_elem_type_ = kEtFileFooter;
-  if (s.ok()) {
-    RecordTick(statistics_, BLOB_DB_BLOB_FILE_BYTES_WRITTEN,
-               BlobLogFooter::kSize);
-  }
+  RecordTick(statistics_, BLOB_DB_BLOB_FILE_BYTES_WRITTEN,
+             BlobLogFooter::kSize);
  return s;
 }

-Status BlobLogWriter::AddRecord(const WriteOptions& write_options,
-                                const Slice& key, const Slice& val,
+Status BlobLogWriter::AddRecord(const Slice& key, const Slice& val,
                                uint64_t expiration, uint64_t* key_offset,
                                uint64_t* blob_offset) {
  assert(block_offset_ != 0);
@@ -145,13 +125,11 @@ Status BlobLogWriter::AddRecord(const WriteOptions& write_options,
  std::string buf;
  ConstructBlobHeader(&buf, key, val, expiration);

-  Status s =
-      EmitPhysicalRecord(write_options, buf, key, val, key_offset, blob_offset);
+  Status s = EmitPhysicalRecord(buf, key, val, key_offset, blob_offset);
  return s;
 }

-Status BlobLogWriter::AddRecord(const WriteOptions& write_options,
-                                const Slice& key, const Slice& val,
+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);
@@ -159,8 +137,7 @@ Status BlobLogWriter::AddRecord(const WriteOptions& write_options,
  std::string buf;
  ConstructBlobHeader(&buf, key, val, 0);

-  Status s =
-      EmitPhysicalRecord(write_options, buf, key, val, key_offset, blob_offset);
+  Status s = EmitPhysicalRecord(buf, key, val, key_offset, blob_offset);
  return s;
 }

@@ -173,34 +150,28 @@ void BlobLogWriter::ConstructBlobHeader(std::string* buf, const Slice& key,
  record.EncodeHeaderTo(buf);
 }

-Status BlobLogWriter::EmitPhysicalRecord(const WriteOptions& write_options,
-                                         const std::string& headerbuf,
+Status BlobLogWriter::EmitPhysicalRecord(const std::string& headerbuf,
                                         const Slice& key, const Slice& val,
                                         uint64_t* key_offset,
                                         uint64_t* blob_offset) {
-  IOOptions opts;
-  Status s = WritableFileWriter::PrepareIOOptions(write_options, opts);
+  StopWatch write_sw(clock_, statistics_, BLOB_DB_BLOB_FILE_WRITE_MICROS);
+  Status s = dest_->Append(Slice(headerbuf));
  if (s.ok()) {
-    s = dest_->Append(opts, Slice(headerbuf));
+    s = dest_->Append(key);
  }
  if (s.ok()) {
-    s = dest_->Append(opts, key);
-  }
-  if (s.ok()) {
-    s = dest_->Append(opts, val);
+    s = dest_->Append(val);
  }
  if (do_flush_ && s.ok()) {
-    s = dest_->Flush(opts);
+    s = dest_->Flush();
  }

  *key_offset = block_offset_ + BlobLogRecord::kHeaderSize;
  *blob_offset = *key_offset + key.size();
  block_offset_ = *blob_offset + val.size();
  last_elem_type_ = kEtRecord;
-  if (s.ok()) {
-    RecordTick(statistics_, BLOB_DB_BLOB_FILE_BYTES_WRITTEN,
-               BlobLogRecord::kHeaderSize + key.size() + val.size());
-  }
+  RecordTick(statistics_, BLOB_DB_BLOB_FILE_BYTES_WRITTEN,
+             BlobLogRecord::kHeaderSize + key.size() + val.size());
  return s;
 }

@@ -43,24 +43,20 @@ class BlobLogWriter {
  static void ConstructBlobHeader(std::string* buf, const Slice& key,
                                  const Slice& val, uint64_t expiration);

-  Status AddRecord(const WriteOptions& write_options, const Slice& key,
-                   const Slice& val, uint64_t* key_offset,
+  Status AddRecord(const Slice& key, const Slice& val, uint64_t* key_offset,
                   uint64_t* blob_offset);

-  Status AddRecord(const WriteOptions& write_options, const Slice& key,
-                   const Slice& val, uint64_t expiration, uint64_t* key_offset,
-                   uint64_t* blob_offset);
+  Status AddRecord(const Slice& key, const Slice& val, uint64_t expiration,
+                   uint64_t* key_offset, uint64_t* blob_offset);

-  Status EmitPhysicalRecord(const WriteOptions& write_options,
-                            const std::string& headerbuf, const Slice& key,
+  Status EmitPhysicalRecord(const std::string& headerbuf, const Slice& key,
                            const Slice& val, uint64_t* key_offset,
                            uint64_t* blob_offset);

-  Status AppendFooter(const WriteOptions& write_options, BlobLogFooter& footer,
-                      std::string* checksum_method,
+  Status AppendFooter(BlobLogFooter& footer, std::string* checksum_method,
                      std::string* checksum_value);

-  Status WriteHeader(const WriteOptions& write_options, BlobLogHeader& header);
+  Status WriteHeader(BlobLogHeader& header);

  WritableFileWriter* file() { return dest_.get(); }

@@ -68,7 +64,7 @@ class BlobLogWriter {

  uint64_t get_log_number() const { return log_number_; }

-  Status Sync(const WriteOptions& write_options);
+  Status Sync();

 private:
  std::unique_ptr<WritableFileWriter> dest_;
@@ -65,7 +65,7 @@ void WriteBlobFile(const ImmutableOptions& immutable_options,
  BlobLogHeader header(column_family_id, compression, has_ttl,
                       expiration_range_header);

-  ASSERT_OK(blob_log_writer.WriteHeader(WriteOptions(), header));
+  ASSERT_OK(blob_log_writer.WriteHeader(header));

  std::vector<std::string> compressed_blobs(num);
  std::vector<Slice> blobs_to_write(num);
@@ -93,8 +93,7 @@ void WriteBlobFile(const ImmutableOptions& immutable_options,

  for (size_t i = 0; i < num; ++i) {
    uint64_t key_offset = 0;
-    ASSERT_OK(blob_log_writer.AddRecord(WriteOptions(), keys[i],
-                                        blobs_to_write[i], &key_offset,
+    ASSERT_OK(blob_log_writer.AddRecord(keys[i], blobs_to_write[i], &key_offset,
                                        &blob_offsets[i]));
  }

@@ -104,8 +103,8 @@ void WriteBlobFile(const ImmutableOptions& immutable_options,

  std::string checksum_method;
  std::string checksum_value;
-  ASSERT_OK(blob_log_writer.AppendFooter(WriteOptions(), footer,
-                                         &checksum_method, &checksum_value));
+  ASSERT_OK(
+      blob_log_writer.AppendFooter(footer, &checksum_method, &checksum_value));
 }

 }  // anonymous namespace
@@ -1221,7 +1220,7 @@ TEST_F(BlobSecondaryCacheTest, GetBlobsFromSecondaryCache) {
      auto sec_handle0 = secondary_cache->Lookup(
          key0, BlobSource::SharedCacheInterface::GetFullHelper(),
          /*context*/ nullptr, true,
-          /*advise_erase=*/true, /*stats=*/nullptr, kept_in_sec_cache);
+          /*advise_erase=*/true, kept_in_sec_cache);
      ASSERT_FALSE(kept_in_sec_cache);
      ASSERT_NE(sec_handle0, nullptr);
      ASSERT_TRUE(sec_handle0->IsReady());
@@ -1249,7 +1248,7 @@ TEST_F(BlobSecondaryCacheTest, GetBlobsFromSecondaryCache) {
      auto sec_handle1 = secondary_cache->Lookup(
          key1, BlobSource::SharedCacheInterface::GetFullHelper(),
          /*context*/ nullptr, true,
-          /*advise_erase=*/true, /*stats=*/nullptr, kept_in_sec_cache);
+          /*advise_erase=*/true, kept_in_sec_cache);
      ASSERT_FALSE(kept_in_sec_cache);
      ASSERT_EQ(sec_handle1, nullptr);

@@ -168,7 +168,6 @@ TEST_F(DBBlobBasicTest, IterateBlobsFromCache) {
      ASSERT_EQ(iter->value().ToString(), blobs[i]);
      ++i;
    }
-    ASSERT_OK(iter->status());
    ASSERT_EQ(i, num_blobs);
    ASSERT_EQ(options.statistics->getAndResetTickerCount(BLOB_DB_CACHE_ADD), 0);
  }
@@ -204,7 +203,6 @@ TEST_F(DBBlobBasicTest, IterateBlobsFromCache) {
      ASSERT_EQ(iter->value().ToString(), blobs[i]);
      ++i;
    }
-    ASSERT_OK(iter->status());
    ASSERT_EQ(i, num_blobs);
    ASSERT_EQ(options.statistics->getAndResetTickerCount(BLOB_DB_CACHE_ADD),
              num_blobs);
@@ -226,7 +224,6 @@ TEST_F(DBBlobBasicTest, IterateBlobsFromCache) {
      ASSERT_EQ(iter->value().ToString(), blobs[i]);
      ++i;
    }
-    ASSERT_OK(iter->status());
    ASSERT_EQ(i, num_blobs);
    ASSERT_EQ(options.statistics->getAndResetTickerCount(BLOB_DB_CACHE_ADD), 0);
  }
@@ -2126,7 +2123,6 @@ TEST_F(DBBlobWithTimestampTest, IterateBlobs) {
                       /*key_is_internal*/ false);
      iter->Prev();
    }
-    ASSERT_OK(iter->status());
  }

  // Backward iteration, then reverse to forward.
@@ -2173,7 +2169,6 @@ TEST_F(DBBlobWithTimestampTest, IterateBlobs) {
        iter->Next();
      }
    }
-    ASSERT_OK(iter->status());
  }

  // Backward iterating multiple versions of the same key, get in this order:
@@ -2192,7 +2187,6 @@ TEST_F(DBBlobWithTimestampTest, IterateBlobs) {
        iter->Prev();
      }
    }
-    ASSERT_OK(iter->status());
  }

  int upper_bound_idx = num_blobs - 2;
@@ -2215,7 +2209,6 @@ TEST_F(DBBlobWithTimestampTest, IterateBlobs) {
        iter->Next();
      }
    }
-    ASSERT_OK(iter->status());
  }

  // Backward iteration with upper and lower bound.
@@ -2231,7 +2224,6 @@ TEST_F(DBBlobWithTimestampTest, IterateBlobs) {
        iter->Prev();
      }
    }
-    ASSERT_OK(iter->status());
  }
 }

@@ -325,7 +325,8 @@ TEST_F(DBBlobIndexTest, Iterate) {

  auto check_is_blob = [&](bool is_blob) {
    return [is_blob](Iterator* iterator) {
-      ASSERT_EQ(is_blob, static_cast<ArenaWrappedDBIter*>(iterator)->IsBlob());
+      ASSERT_EQ(is_blob,
+                reinterpret_cast<ArenaWrappedDBIter*>(iterator)->IsBlob());
    };
  };

@@ -11,10 +11,8 @@ FilePrefetchBuffer* PrefetchBufferCollection::GetOrCreatePrefetchBuffer(
    uint64_t file_number) {
  auto& prefetch_buffer = prefetch_buffers_[file_number];
  if (!prefetch_buffer) {
-    ReadaheadParams readahead_params;
-    readahead_params.initial_readahead_size = readahead_size_;
-    readahead_params.max_readahead_size = readahead_size_;
-    prefetch_buffer.reset(new FilePrefetchBuffer(readahead_params));
+    prefetch_buffer.reset(
+        new FilePrefetchBuffer(readahead_size_, readahead_size_));
  }

  return prefetch_buffer.get();
@@ -32,11 +32,9 @@
 #include "options/options_helper.h"
 #include "rocksdb/db.h"
 #include "rocksdb/env.h"
-#include "rocksdb/file_system.h"
 #include "rocksdb/iterator.h"
 #include "rocksdb/options.h"
 #include "rocksdb/table.h"
-#include "seqno_to_time_mapping.h"
 #include "table/block_based/block_based_table_builder.h"
 #include "table/format.h"
 #include "table/internal_iterator.h"
@@ -59,8 +57,8 @@ TableBuilder* NewTableBuilder(const TableBuilderOptions& tboptions,
 Status BuildTable(
    const std::string& dbname, VersionSet* versions,
    const ImmutableDBOptions& db_options, const TableBuilderOptions& tboptions,
-    const FileOptions& file_options, TableCache* table_cache,
-    InternalIterator* iter,
+    const FileOptions& file_options, const ReadOptions& read_options,
+    TableCache* table_cache, InternalIterator* iter,
    std::vector<std::unique_ptr<FragmentedRangeTombstoneIterator>>
        range_del_iters,
    FileMetaData* meta, std::vector<BlobFileAddition>* blob_file_additions,
@@ -71,8 +69,9 @@ Status BuildTable(
    IOStatus* io_status, const std::shared_ptr<IOTracer>& io_tracer,
    BlobFileCreationReason blob_creation_reason,
    const SeqnoToTimeMapping& seqno_to_time_mapping, EventLogger* event_logger,
-    int job_id, TableProperties* table_properties,
-    Env::WriteLifeTimeHint write_hint, const std::string* full_history_ts_low,
+    int job_id, const Env::IOPriority io_priority,
+    TableProperties* table_properties, Env::WriteLifeTimeHint write_hint,
+    const std::string* full_history_ts_low,
    BlobFileCompletionCallback* blob_callback, Version* version,
    uint64_t* num_input_entries, uint64_t* memtable_payload_bytes,
    uint64_t* memtable_garbage_bytes) {
@@ -83,8 +82,11 @@ Status BuildTable(
  auto& ioptions = tboptions.ioptions;
  // Reports the IOStats for flush for every following bytes.
  const size_t kReportFlushIOStatsEvery = 1048576;
-  OutputValidator output_validator(tboptions.internal_comparator,
-                                   /*enable_hash=*/paranoid_file_checks);
+  OutputValidator output_validator(
+      tboptions.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();
@@ -162,11 +164,11 @@ Status BuildTable(

      table_file_created = true;
      FileTypeSet tmp_set = ioptions.checksum_handoff_file_types;
-      file->SetIOPriority(tboptions.write_options.rate_limiter_priority);
+      file->SetIOPriority(io_priority);
      file->SetWriteLifeTimeHint(write_hint);
      file_writer.reset(new WritableFileWriter(
          std::move(file), fname, file_options, ioptions.clock, io_tracer,
-          ioptions.stats, Histograms::SST_WRITE_MICROS, ioptions.listeners,
+          ioptions.stats, ioptions.listeners,
          ioptions.file_checksum_gen_factory.get(),
          tmp_set.Contains(FileType::kTableFile), false));

@@ -186,11 +188,10 @@ Status BuildTable(
         blob_file_additions)
            ? new BlobFileBuilder(
                  versions, fs, &ioptions, &mutable_cf_options, &file_options,
-                  &(tboptions.write_options), tboptions.db_id,
-                  tboptions.db_session_id, job_id, tboptions.column_family_id,
-                  tboptions.column_family_name, write_hint, io_tracer,
-                  blob_callback, blob_creation_reason, &blob_file_paths,
-                  blob_file_additions)
+                  tboptions.db_id, tboptions.db_session_id, job_id,
+                  tboptions.column_family_id, tboptions.column_family_name,
+                  io_priority, write_hint, io_tracer, blob_callback,
+                  blob_creation_reason, &blob_file_paths, blob_file_additions)
            : nullptr);

    const std::atomic<bool> kManualCompactionCanceledFalse{false};
@@ -207,7 +208,7 @@ Status BuildTable(
        /*shutting_down=*/nullptr, db_options.info_log, full_history_ts_low);

    const size_t ts_sz = ucmp->timestamp_size();
-    const bool logical_strip_timestamp =
+    const bool strip_timestamp =
        ts_sz > 0 && !ioptions.persist_user_defined_timestamps;

    std::string key_after_flush_buf;
@@ -221,7 +222,7 @@ Status BuildTable(
      // the in memory version of the key act logically the same as one with a
      // minimum timestamp. We update the timestamp here so file boundary and
      // output validator, block builder all see the effect of the stripping.
-      if (logical_strip_timestamp) {
+      if (strip_timestamp) {
        key_after_flush_buf.clear();
        ReplaceInternalKeyWithMinTimestamp(&key_after_flush_buf, key, ts_sz);
        key_after_flush = key_after_flush_buf;
@@ -243,11 +244,7 @@ Status BuildTable(
      }

      // TODO(noetzli): Update stats after flush, too.
-      // TODO(hx235): Replace `rate_limiter_priority` with `io_activity` for
-      // flush IO in repair when we have an `Env::IOActivity` enum for it
-      if ((tboptions.write_options.io_activity == Env::IOActivity::kFlush ||
-           tboptions.write_options.io_activity == Env::IOActivity::kDBOpen ||
-           tboptions.write_options.rate_limiter_priority == Env::IO_HIGH) &&
+      if (io_priority == Env::IO_HIGH &&
          IOSTATS(bytes_written) >= kReportFlushIOStatsEvery) {
        ThreadStatusUtil::SetThreadOperationProperty(
            ThreadStatus::FLUSH_BYTES_WRITTEN, IOSTATS(bytes_written));
@@ -264,12 +261,9 @@ Status BuildTable(
      Slice last_tombstone_start_user_key{};
      for (range_del_it->SeekToFirst(); range_del_it->Valid();
           range_del_it->Next()) {
-        // When user timestamp should not be persisted, we logically strip a
-        // range tombstone's start and end key's timestamp (replace it with min
-        // timestamp) before passing them along to table builder and to update
-        // file boundaries.
-        auto tombstone = range_del_it->Tombstone(logical_strip_timestamp);
-        std::pair<InternalKey, Slice> kv = tombstone.Serialize();
+        auto tombstone = range_del_it->Tombstone();
+        auto kv = tombstone.Serialize();
+        // TODO(yuzhangyu): handle range deletion for UDT in memtables only.
        builder->Add(kv.first.Encode(), kv.second);
        InternalKey tombstone_end = tombstone.SerializeEndKey();
        meta->UpdateBoundariesForRange(kv.first, tombstone_end, tombstone.seq_,
@@ -281,7 +275,7 @@ Status BuildTable(
            SizeApproximationOptions approx_opts;
            approx_opts.files_size_error_margin = 0.1;
            meta->compensated_range_deletion_size += versions->ApproximateSize(
-                approx_opts, tboptions.read_options, version, kv.first.Encode(),
+                approx_opts, read_options, version, kv.first.Encode(),
                tombstone_end.Encode(), 0 /* start_level */, -1 /* end_level */,
                TableReaderCaller::kFlush);
          }
@@ -300,14 +294,12 @@ Status BuildTable(
    if (!s.ok() || empty) {
      builder->Abandon();
    } else {
-      SeqnoToTimeMapping relevant_mapping;
-      relevant_mapping.CopyFromSeqnoRange(seqno_to_time_mapping,
-                                          meta->fd.smallest_seqno,
-                                          meta->fd.largest_seqno);
-      relevant_mapping.SetCapacity(kMaxSeqnoTimePairsPerSST);
-      relevant_mapping.Enforce(tboptions.file_creation_time);
+      std::string seqno_time_mapping_str;
+      seqno_to_time_mapping.Encode(
+          seqno_time_mapping_str, meta->fd.smallest_seqno,
+          meta->fd.largest_seqno, meta->file_creation_time);
      builder->SetSeqnoTimeTableProperties(
-          relevant_mapping,
+          seqno_time_mapping_str,
          ioptions.compaction_style == CompactionStyle::kCompactionStyleFIFO
              ? meta->file_creation_time
              : meta->oldest_ancester_time);
@@ -354,16 +346,13 @@ Status BuildTable(

    // Finish and check for file errors
    TEST_SYNC_POINT("BuildTable:BeforeSyncTable");
-    IOOptions opts;
-    *io_status =
-        WritableFileWriter::PrepareIOOptions(tboptions.write_options, opts);
-    if (s.ok() && io_status->ok() && !empty) {
+    if (s.ok() && !empty) {
      StopWatch sw(ioptions.clock, ioptions.stats, TABLE_SYNC_MICROS);
-      *io_status = file_writer->Sync(opts, ioptions.use_fsync);
+      *io_status = file_writer->Sync(ioptions.use_fsync);
    }
    TEST_SYNC_POINT("BuildTable:BeforeCloseTableFile");
    if (s.ok() && io_status->ok() && !empty) {
-      *io_status = file_writer->Close(opts);
+      *io_status = file_writer->Close();
    }
    if (s.ok() && io_status->ok() && !empty) {
      // Add the checksum information to file metadata.
@@ -407,9 +396,9 @@ Status BuildTable(
      // No matter whether use_direct_io_for_flush_and_compaction is true,
      // the goal is to cache it here for further user reads.
      std::unique_ptr<InternalIterator> it(table_cache->NewIterator(
-          tboptions.read_options, file_options, tboptions.internal_comparator,
-          *meta, nullptr /* range_del_agg */,
-          mutable_cf_options.prefix_extractor, nullptr,
+          read_options, file_options, tboptions.internal_comparator, *meta,
+          nullptr /* range_del_agg */, mutable_cf_options.prefix_extractor,
+          nullptr,
          (internal_stats == nullptr) ? nullptr
                                      : internal_stats->GetFileReadHist(0),
          TableReaderCaller::kFlush, /*arena=*/nullptr,
@@ -422,6 +411,7 @@ Status BuildTable(
      s = it->status();
      if (s.ok() && paranoid_file_checks) {
        OutputValidator file_validator(tboptions.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
@@ -446,13 +436,8 @@ Status BuildTable(
    constexpr IODebugContext* dbg = nullptr;

    if (table_file_created) {
-      IOOptions opts;
-      Status prepare =
-          WritableFileWriter::PrepareIOOptions(tboptions.write_options, opts);
-      if (prepare.ok()) {
-        Status ignored = fs->DeleteFile(fname, opts, dbg);
-        ignored.PermitUncheckedError();
-      }
+      Status ignored = fs->DeleteFile(fname, IOOptions(), dbg);
+      ignored.PermitUncheckedError();
    }

    assert(blob_file_additions || blob_file_paths.empty());
@@ -50,11 +50,11 @@ TableBuilder* NewTableBuilder(const TableBuilderOptions& tboptions,
 //
 // @param column_family_name Name of the column family that is also identified
 //    by column_family_id, or empty string if unknown.
-Status BuildTable(
+extern Status BuildTable(
    const std::string& dbname, VersionSet* versions,
    const ImmutableDBOptions& db_options, const TableBuilderOptions& tboptions,
-    const FileOptions& file_options, TableCache* table_cache,
-    InternalIterator* iter,
+    const FileOptions& file_options, const ReadOptions& read_options,
+    TableCache* table_cache, InternalIterator* iter,
    std::vector<std::unique_ptr<FragmentedRangeTombstoneIterator>>
        range_del_iters,
    FileMetaData* meta, std::vector<BlobFileAddition>* blob_file_additions,
@@ -66,6 +66,7 @@ Status BuildTable(
    BlobFileCreationReason blob_creation_reason,
    const SeqnoToTimeMapping& seqno_to_time_mapping,
    EventLogger* event_logger = nullptr, int job_id = 0,
+    const Env::IOPriority io_priority = Env::IO_HIGH,
    TableProperties* table_properties = nullptr,
    Env::WriteLifeTimeHint write_hint = Env::WLTH_NOT_SET,
    const std::string* full_history_ts_low = nullptr,
@@ -45,8 +45,6 @@
 #include "rocksdb/utilities/transaction_db.h"
 #include "rocksdb/utilities/write_batch_with_index.h"
 #include "rocksdb/write_batch.h"
-#include "rocksdb/write_buffer_manager.h"
-#include "util/stderr_logger.h"
 #include "utilities/merge_operators.h"

 using ROCKSDB_NAMESPACE::BackupEngine;
@@ -116,7 +114,6 @@ using ROCKSDB_NAMESPACE::Snapshot;
 using ROCKSDB_NAMESPACE::SstFileMetaData;
 using ROCKSDB_NAMESPACE::SstFileWriter;
 using ROCKSDB_NAMESPACE::Status;
-using ROCKSDB_NAMESPACE::StderrLogger;
 using ROCKSDB_NAMESPACE::TablePropertiesCollectorFactory;
 using ROCKSDB_NAMESPACE::Transaction;
 using ROCKSDB_NAMESPACE::TransactionDB;
@@ -128,7 +125,6 @@ using ROCKSDB_NAMESPACE::WALRecoveryMode;
 using ROCKSDB_NAMESPACE::WritableFile;
 using ROCKSDB_NAMESPACE::WriteBatch;
 using ROCKSDB_NAMESPACE::WriteBatchWithIndex;
-using ROCKSDB_NAMESPACE::WriteBufferManager;
 using ROCKSDB_NAMESPACE::WriteOptions;

 using std::unordered_set;
@@ -223,9 +219,6 @@ struct rocksdb_memory_allocator_t {
 struct rocksdb_cache_t {
  std::shared_ptr<Cache> rep;
 };
-struct rocksdb_write_buffer_manager_t {
-  std::shared_ptr<WriteBufferManager> rep;
-};
 struct rocksdb_livefiles_t {
  std::vector<LiveFileMetaData> rep;
 };
@@ -1687,11 +1680,6 @@ void rocksdb_release_snapshot(rocksdb_t* db,
  delete snapshot;
 }

-uint64_t rocksdb_snapshot_get_sequence_number(
-    const rocksdb_snapshot_t* snapshot) {
-  return snapshot->rep->GetSequenceNumber();
-}
-
 char* rocksdb_property_value(rocksdb_t* db, const char* propname) {
  std::string tmp;
  if (db->rep->GetProperty(Slice(propname), &tmp)) {
@@ -1879,8 +1867,9 @@ void rocksdb_disable_file_deletions(rocksdb_t* db, char** errptr) {
  SaveError(errptr, db->rep->DisableFileDeletions());
 }

-void rocksdb_enable_file_deletions(rocksdb_t* db, char** errptr) {
-  SaveError(errptr, db->rep->EnableFileDeletions());
+void rocksdb_enable_file_deletions(rocksdb_t* db, unsigned char force,
+                                   char** errptr) {
+  SaveError(errptr, db->rep->EnableFileDeletions(force));
 }

 void rocksdb_destroy_db(const rocksdb_options_t* options, const char* name,
@@ -2915,33 +2904,6 @@ void rocksdb_options_set_db_paths(rocksdb_options_t* opt,
  opt->rep.db_paths = db_paths;
 }

-void rocksdb_options_set_cf_paths(rocksdb_options_t* opt,
-                                  const rocksdb_dbpath_t** dbpath_values,
-                                  size_t num_paths) {
-  std::vector<DbPath> cf_paths(num_paths);
-  for (size_t i = 0; i < num_paths; ++i) {
-    cf_paths[i] = dbpath_values[i]->rep;
-  }
-  opt->rep.cf_paths = cf_paths;
-}
-
-rocksdb_logger_t* rocksdb_logger_create_stderr_logger(int log_level,
-                                                      const char* prefix) {
-  rocksdb_logger_t* logger = new rocksdb_logger_t;
-
-  if (prefix) {
-    logger->rep = std::make_shared<StderrLogger>(
-        static_cast<InfoLogLevel>(log_level), prefix);
-  } else {
-    logger->rep =
-        std::make_shared<StderrLogger>(static_cast<InfoLogLevel>(log_level));
-  }
-
-  return logger;
-}
-
-void rocksdb_logger_destroy(rocksdb_logger_t* logger) { delete logger; }
-
 void rocksdb_options_set_env(rocksdb_options_t* opt, rocksdb_env_t* env) {
  opt->rep.env = (env ? env->rep : nullptr);
 }
@@ -2952,12 +2914,6 @@ void rocksdb_options_set_info_log(rocksdb_options_t* opt, rocksdb_logger_t* l) {
  }
 }

-rocksdb_logger_t* rocksdb_options_get_info_log(rocksdb_options_t* opt) {
-  rocksdb_logger_t* info_log = new rocksdb_logger_t;
-  info_log->rep = opt->rep.info_log;
-  return info_log;
-}
-
 void rocksdb_options_set_info_log_level(rocksdb_options_t* opt, int v) {
  opt->rep.info_log_level = static_cast<InfoLogLevel>(v);
 }
@@ -2979,11 +2935,6 @@ void rocksdb_options_set_write_buffer_size(rocksdb_options_t* opt, size_t s) {
  opt->rep.write_buffer_size = s;
 }

-void rocksdb_options_set_write_buffer_manager(
-    rocksdb_options_t* opt, rocksdb_write_buffer_manager_t* wbm) {
-  opt->rep.write_buffer_manager = wbm->rep;
-}
-
 size_t rocksdb_options_get_write_buffer_size(rocksdb_options_t* opt) {
  return opt->rep.write_buffer_size;
 }
@@ -3078,24 +3029,6 @@ void rocksdb_options_set_max_bytes_for_level_multiplier_additional(
  }
 }

-void rocksdb_options_set_ttl(rocksdb_options_t* opt, uint64_t seconds) {
-  opt->rep.ttl = seconds;
-}
-
-uint64_t rocksdb_options_get_ttl(rocksdb_options_t* opt) {
-  return opt->rep.ttl;
-}
-
-void rocksdb_options_set_periodic_compaction_seconds(rocksdb_options_t* opt,
-                                                     uint64_t seconds) {
-  opt->rep.periodic_compaction_seconds = seconds;
-}
-
-uint64_t rocksdb_options_get_periodic_compaction_seconds(
-    rocksdb_options_t* opt) {
-  return opt->rep.periodic_compaction_seconds;
-}
-
 void rocksdb_options_enable_statistics(rocksdb_options_t* opt) {
  opt->rep.statistics = ROCKSDB_NAMESPACE::CreateDBStatistics();
 }
@@ -3148,7 +3081,7 @@ void rocksdb_options_set_enable_blob_files(rocksdb_options_t* opt,
                                           unsigned char val) {
  opt->rep.enable_blob_files = val;
 }
-ROCKSDB_LIBRARY_API unsigned char rocksdb_options_get_enable_blob_files(
+extern ROCKSDB_LIBRARY_API unsigned char rocksdb_options_get_enable_blob_files(
    rocksdb_options_t* opt) {
  return opt->rep.enable_blob_files;
 }
@@ -3520,6 +3453,35 @@ unsigned char rocksdb_options_get_advise_random_on_open(
  return opt->rep.advise_random_on_open;
 }

+void rocksdb_options_set_access_hint_on_compaction_start(rocksdb_options_t* opt,
+                                                         int v) {
+  switch (v) {
+    case 0:
+      opt->rep.access_hint_on_compaction_start =
+          ROCKSDB_NAMESPACE::Options::NONE;
+      break;
+    case 1:
+      opt->rep.access_hint_on_compaction_start =
+          ROCKSDB_NAMESPACE::Options::NORMAL;
+      break;
+    case 2:
+      opt->rep.access_hint_on_compaction_start =
+          ROCKSDB_NAMESPACE::Options::SEQUENTIAL;
+      break;
+    case 3:
+      opt->rep.access_hint_on_compaction_start =
+          ROCKSDB_NAMESPACE::Options::WILLNEED;
+      break;
+    default:
+      assert(0);
+  }
+}
+
+int rocksdb_options_get_access_hint_on_compaction_start(
+    rocksdb_options_t* opt) {
+  return opt->rep.access_hint_on_compaction_start;
+}
+
 void rocksdb_options_set_use_adaptive_mutex(rocksdb_options_t* opt,
                                            unsigned char v) {
  opt->rep.use_adaptive_mutex = v;
@@ -3995,26 +3957,6 @@ rocksdb_ratelimiter_t* rocksdb_ratelimiter_create(int64_t rate_bytes_per_sec,
  return rate_limiter;
 }

-rocksdb_ratelimiter_t* rocksdb_ratelimiter_create_auto_tuned(
-    int64_t rate_bytes_per_sec, int64_t refill_period_us, int32_t fairness) {
-  rocksdb_ratelimiter_t* rate_limiter = new rocksdb_ratelimiter_t;
-  rate_limiter->rep.reset(NewGenericRateLimiter(rate_bytes_per_sec,
-                                                refill_period_us, fairness,
-                                                RateLimiter::Mode::kWritesOnly,
-                                                true));  // auto_tuned
-  return rate_limiter;
-}
-
-rocksdb_ratelimiter_t* rocksdb_ratelimiter_create_with_mode(
-    int64_t rate_bytes_per_sec, int64_t refill_period_us, int32_t fairness,
-    int mode, bool auto_tuned) {
-  rocksdb_ratelimiter_t* rate_limiter = new rocksdb_ratelimiter_t;
-  rate_limiter->rep.reset(
-      NewGenericRateLimiter(rate_bytes_per_sec, refill_period_us, fairness,
-                            static_cast<RateLimiter::Mode>(mode), auto_tuned));
-  return rate_limiter;
-}
-
 void rocksdb_ratelimiter_destroy(rocksdb_ratelimiter_t* limiter) {
  delete limiter;
 }
@@ -4611,7 +4553,7 @@ void rocksdb_readoptions_set_io_timeout(rocksdb_readoptions_t* opt,
  opt->rep.io_timeout = std::chrono::microseconds(microseconds);
 }

-ROCKSDB_LIBRARY_API uint64_t
+extern ROCKSDB_LIBRARY_API uint64_t
 rocksdb_readoptions_get_io_timeout(rocksdb_readoptions_t* opt) {
  return opt->rep.io_timeout.count();
 }
@@ -4924,60 +4866,6 @@ size_t rocksdb_cache_get_occupancy_count(const rocksdb_cache_t* cache) {
  return cache->rep->GetOccupancyCount();
 }

-rocksdb_write_buffer_manager_t* rocksdb_write_buffer_manager_create(
-    size_t buffer_size, bool allow_stall) {
-  rocksdb_write_buffer_manager_t* wbm = new rocksdb_write_buffer_manager_t;
-  wbm->rep.reset(new WriteBufferManager(buffer_size, {}, allow_stall));
-  return wbm;
-}
-
-rocksdb_write_buffer_manager_t* rocksdb_write_buffer_manager_create_with_cache(
-    size_t buffer_size, const rocksdb_cache_t* cache, bool allow_stall) {
-  rocksdb_write_buffer_manager_t* wbm = new rocksdb_write_buffer_manager_t;
-  wbm->rep.reset(new WriteBufferManager(buffer_size, cache->rep, allow_stall));
-  return wbm;
-}
-
-void rocksdb_write_buffer_manager_destroy(rocksdb_write_buffer_manager_t* wbm) {
-  delete wbm;
-}
-
-bool rocksdb_write_buffer_manager_enabled(rocksdb_write_buffer_manager_t* wbm) {
-  return wbm->rep->enabled();
-}
-
-bool rocksdb_write_buffer_manager_cost_to_cache(
-    rocksdb_write_buffer_manager_t* wbm) {
-  return wbm->rep->cost_to_cache();
-}
-
-size_t rocksdb_write_buffer_manager_memory_usage(
-    rocksdb_write_buffer_manager_t* wbm) {
-  return wbm->rep->memory_usage();
-}
-
-size_t rocksdb_write_buffer_manager_mutable_memtable_memory_usage(
-    rocksdb_write_buffer_manager_t* wbm) {
-  return wbm->rep->mutable_memtable_memory_usage();
-}
-
-size_t rocksdb_write_buffer_manager_dummy_entries_in_cache_usage(
-    rocksdb_write_buffer_manager_t* wbm) {
-  return wbm->rep->dummy_entries_in_cache_usage();
-}
-size_t rocksdb_write_buffer_manager_buffer_size(
-    rocksdb_write_buffer_manager_t* wbm) {
-  return wbm->rep->buffer_size();
-}
-void rocksdb_write_buffer_manager_set_buffer_size(
-    rocksdb_write_buffer_manager_t* wbm, size_t new_size) {
-  wbm->rep->SetBufferSize(new_size);
-}
-ROCKSDB_LIBRARY_API void rocksdb_write_buffer_manager_set_allow_stall(
-    rocksdb_write_buffer_manager_t* wbm, bool new_allow_stall) {
-  wbm->rep->SetAllowStall(new_allow_stall);
-}
-
 rocksdb_dbpath_t* rocksdb_dbpath_create(const char* path,
                                        uint64_t target_size) {
  rocksdb_dbpath_t* result = new rocksdb_dbpath_t;
@@ -5444,7 +5332,9 @@ uint64_t rocksdb_livefiles_deletions(const rocksdb_livefiles_t* lf, int index) {
  return lf->rep[index].num_deletions;
 }

-void rocksdb_livefiles_destroy(const rocksdb_livefiles_t* lf) { delete lf; }
+extern void rocksdb_livefiles_destroy(const rocksdb_livefiles_t* lf) {
+  delete lf;
+}

 void rocksdb_get_options_from_string(const rocksdb_options_t* base_options,
                                     const char* opts_str,
@@ -5574,11 +5464,6 @@ char* rocksdb_sst_file_metadata_get_relative_filename(
  return strdup(file_meta->rep->relative_filename.c_str());
 }

-char* rocksdb_sst_file_metadata_get_directory(
-    rocksdb_sst_file_metadata_t* file_meta) {
-  return strdup(file_meta->rep->directory.c_str());
-}
-
 uint64_t rocksdb_sst_file_metadata_get_size(
    rocksdb_sst_file_metadata_t* file_meta) {
  return file_meta->rep->size;
@@ -713,16 +713,6 @@ int main(int argc, char** argv) {
  rocksdb_options_set_ratelimiter(options, rate_limiter);
  rocksdb_ratelimiter_destroy(rate_limiter);

-  rate_limiter =
-      rocksdb_ratelimiter_create_auto_tuned(1000 * 1024 * 1024, 100 * 1000, 10);
-  rocksdb_options_set_ratelimiter(options, rate_limiter);
-  rocksdb_ratelimiter_destroy(rate_limiter);
-
-  rate_limiter = rocksdb_ratelimiter_create_with_mode(1000 * 1024 * 1024,
-                                                      100 * 1000, 10, 0, true);
-  rocksdb_options_set_ratelimiter(options, rate_limiter);
-  rocksdb_ratelimiter_destroy(rate_limiter);
-
  roptions = rocksdb_readoptions_create();
  rocksdb_readoptions_set_verify_checksums(roptions, 1);
  rocksdb_readoptions_set_fill_cache(roptions, 1);
@@ -1484,20 +1474,10 @@ int main(int argc, char** argv) {
    CheckCondition(cflen == 2);
    rocksdb_list_column_families_destroy(column_fams, cflen);

-    rocksdb_options_t* cf_options_1 = rocksdb_options_create();
-    rocksdb_options_t* cf_options_2 = rocksdb_options_create();
-
-    // use dbpathname2 as the cf_path for "cf1"
-    rocksdb_dbpath_t* dbpath2;
-    char dbpathname2[200];
-    snprintf(dbpathname2, sizeof(dbpathname2), "%s/rocksdb_c_test-%d-dbpath2",
-             GetTempDir(), ((int)geteuid()));
-    dbpath2 = rocksdb_dbpath_create(dbpathname2, 1024 * 1024);
-    const rocksdb_dbpath_t* cf_paths[1] = {dbpath2};
-    rocksdb_options_set_cf_paths(cf_options_2, cf_paths, 1);
+    rocksdb_options_t* cf_options = rocksdb_options_create();

    const char* cf_names[2] = {"default", "cf1"};
-    const rocksdb_options_t* cf_opts[2] = {cf_options_1, cf_options_2};
+    const rocksdb_options_t* cf_opts[2] = {cf_options, cf_options};
    rocksdb_column_family_handle_t* handles[2];

    LoadAndCheckLatestOptions(dbname, env, false, cache, NULL, 2, cf_names,
@@ -1525,37 +1505,6 @@ int main(int argc, char** argv) {
    rocksdb_flushoptions_t* flush_options = rocksdb_flushoptions_create();
    rocksdb_flushoptions_set_wait(flush_options, 1);
    rocksdb_flush_cf(db, flush_options, handles[1], &err);
-
-    // make sure all files in "cf1" are under the specified cf path
-    {
-      rocksdb_column_family_metadata_t* cf_meta =
-          rocksdb_get_column_family_metadata_cf(db, handles[1]);
-      size_t cf_file_count = rocksdb_column_family_metadata_get_size(cf_meta);
-      assert(cf_file_count > 0);
-      size_t level_count =
-          rocksdb_column_family_metadata_get_level_count(cf_meta);
-      assert(level_count > 0);
-      for (size_t l = 0; l < level_count; ++l) {
-        rocksdb_level_metadata_t* level_meta =
-            rocksdb_column_family_metadata_get_level_metadata(cf_meta, l);
-        assert(level_meta);
-
-        size_t file_count = rocksdb_level_metadata_get_file_count(level_meta);
-        for (size_t f = 0; f < file_count; ++f) {
-          rocksdb_sst_file_metadata_t* file_meta =
-              rocksdb_level_metadata_get_sst_file_metadata(level_meta, f);
-          assert(file_meta);
-          char* file_path = rocksdb_sst_file_metadata_get_directory(file_meta);
-          assert(strcmp(file_path, dbpathname2) == 0);
-          Free(&file_path);
-          rocksdb_sst_file_metadata_destroy(file_meta);
-        }
-        rocksdb_level_metadata_destroy(level_meta);
-      }
-
-      rocksdb_column_family_metadata_destroy(cf_meta);
-    }
-
    CheckNoError(err) rocksdb_flushoptions_destroy(flush_options);

    CheckGetCF(db, roptions, handles[1], "foo", "hello");
@@ -1719,9 +1668,7 @@ int main(int argc, char** argv) {
    }
    rocksdb_destroy_db(options, dbname, &err);
    rocksdb_options_destroy(db_options);
-    rocksdb_options_destroy(cf_options_1);
-    rocksdb_options_destroy(cf_options_2);
-    rocksdb_dbpath_destroy(dbpath2);
+    rocksdb_options_destroy(cf_options);
  }

  StartPhase("prefix");
@@ -1926,13 +1873,6 @@ int main(int argc, char** argv) {
    CheckCondition(2.0 ==
                   rocksdb_options_get_max_bytes_for_level_multiplier(o));

-    rocksdb_options_set_periodic_compaction_seconds(o, 100000);
-    CheckCondition(100000 ==
-                   rocksdb_options_get_periodic_compaction_seconds(o));
-
-    rocksdb_options_set_ttl(o, 5000);
-    CheckCondition(5000 == rocksdb_options_get_ttl(o));
-
    rocksdb_options_set_skip_stats_update_on_db_open(o, 1);
    CheckCondition(1 == rocksdb_options_get_skip_stats_update_on_db_open(o));

@@ -2039,6 +1979,8 @@ int main(int argc, char** argv) {
    rocksdb_options_set_advise_random_on_open(o, 1);
    CheckCondition(1 == rocksdb_options_get_advise_random_on_open(o));

+    rocksdb_options_set_access_hint_on_compaction_start(o, 3);
+    CheckCondition(3 == rocksdb_options_get_access_hint_on_compaction_start(o));

    rocksdb_options_set_use_adaptive_mutex(o, 1);
    CheckCondition(1 == rocksdb_options_get_use_adaptive_mutex(o));
@@ -2234,6 +2176,8 @@ int main(int argc, char** argv) {
    CheckCondition(18 == rocksdb_options_get_stats_dump_period_sec(copy));
    CheckCondition(5 == rocksdb_options_get_stats_persist_period_sec(copy));
    CheckCondition(1 == rocksdb_options_get_advise_random_on_open(copy));
+    CheckCondition(3 ==
+                   rocksdb_options_get_access_hint_on_compaction_start(copy));
    CheckCondition(1 == rocksdb_options_get_use_adaptive_mutex(copy));
    CheckCondition(19 == rocksdb_options_get_bytes_per_sync(copy));
    CheckCondition(20 == rocksdb_options_get_wal_bytes_per_sync(copy));
@@ -2359,16 +2303,6 @@ int main(int argc, char** argv) {
    CheckCondition(2.0 ==
                   rocksdb_options_get_max_bytes_for_level_multiplier(o));

-    rocksdb_options_set_periodic_compaction_seconds(copy, 8000);
-    CheckCondition(8000 ==
-                   rocksdb_options_get_periodic_compaction_seconds(copy));
-    CheckCondition(100000 ==
-                   rocksdb_options_get_periodic_compaction_seconds(o));
-
-    rocksdb_options_set_ttl(copy, 8000);
-    CheckCondition(8000 == rocksdb_options_get_ttl(copy));
-    CheckCondition(5000 == rocksdb_options_get_ttl(o));
-
    rocksdb_options_set_skip_stats_update_on_db_open(copy, 0);
    CheckCondition(0 == rocksdb_options_get_skip_stats_update_on_db_open(copy));
    CheckCondition(1 == rocksdb_options_get_skip_stats_update_on_db_open(o));
@@ -2516,6 +2450,11 @@ int main(int argc, char** argv) {
    CheckCondition(0 == rocksdb_options_get_advise_random_on_open(copy));
    CheckCondition(1 == rocksdb_options_get_advise_random_on_open(o));

+    rocksdb_options_set_access_hint_on_compaction_start(copy, 2);
+    CheckCondition(2 ==
+                   rocksdb_options_get_access_hint_on_compaction_start(copy));
+    CheckCondition(3 == rocksdb_options_get_access_hint_on_compaction_start(o));
+
    rocksdb_options_set_use_adaptive_mutex(copy, 0);
    CheckCondition(0 == rocksdb_options_get_use_adaptive_mutex(copy));
    CheckCondition(1 == rocksdb_options_get_use_adaptive_mutex(o));
@@ -2784,30 +2723,6 @@ int main(int argc, char** argv) {
    rocksdb_memory_allocator_destroy(allocator);
  }

-  StartPhase("stderr_logger");
-  {
-    rocksdb_options_t* o_no_prefix = rocksdb_options_create();
-    rocksdb_logger_t* no_prefix_logger =
-        rocksdb_logger_create_stderr_logger(3, NULL);
-    rocksdb_options_set_info_log(o_no_prefix, no_prefix_logger);
-    rocksdb_logger_t* no_prefix_info_log =
-        rocksdb_options_get_info_log(o_no_prefix);
-    CheckCondition(no_prefix_info_log != NULL);
-    rocksdb_logger_destroy(no_prefix_logger);
-    rocksdb_logger_destroy(no_prefix_info_log);
-    rocksdb_options_destroy(o_no_prefix);
-
-    rocksdb_options_t* o_prefix = rocksdb_options_create();
-    rocksdb_logger_t* prefix_logger =
-        rocksdb_logger_create_stderr_logger(3, "some prefix");
-    rocksdb_options_set_info_log(o_prefix, prefix_logger);
-    rocksdb_logger_t* prefix_info_log = rocksdb_options_get_info_log(o_prefix);
-    CheckCondition(prefix_info_log != NULL);
-    rocksdb_logger_destroy(prefix_logger);
-    rocksdb_logger_destroy(prefix_info_log);
-    rocksdb_options_destroy(o_prefix);
-  }
-
  StartPhase("env");
  {
    rocksdb_env_t* e;
@@ -3714,7 +3629,7 @@ int main(int argc, char** argv) {

  StartPhase("statistics");
  {
-    const uint32_t BYTES_WRITTEN_TICKER = 60;
+    const uint32_t BYTES_WRITTEN_TICKER = 40;
    const uint32_t DB_WRITE_HIST = 1;

    rocksdb_statistics_histogram_data_t* hist =
@@ -3808,30 +3723,6 @@ int main(int argc, char** argv) {
    rocksdb_wait_for_compact_options_destroy(wco);
  }

-  StartPhase("write_buffer_manager");
-  {
-    rocksdb_cache_t* lru;
-    lru = rocksdb_cache_create_lru(100);
-
-    rocksdb_write_buffer_manager_t* write_buffer_manager;
-    write_buffer_manager =
-        rocksdb_write_buffer_manager_create_with_cache(200, lru, false);
-
-    CheckCondition(true ==
-                   rocksdb_write_buffer_manager_enabled(write_buffer_manager));
-    CheckCondition(true == rocksdb_write_buffer_manager_cost_to_cache(
-                               write_buffer_manager));
-    CheckCondition(
-        200 == rocksdb_write_buffer_manager_buffer_size(write_buffer_manager));
-
-    rocksdb_write_buffer_manager_set_buffer_size(write_buffer_manager, 300);
-    CheckCondition(
-        300 == rocksdb_write_buffer_manager_buffer_size(write_buffer_manager));
-
-    rocksdb_write_buffer_manager_destroy(write_buffer_manager);
-    rocksdb_cache_destroy(lru);
-  }
-
  StartPhase("cancel_all_background_work");
  rocksdb_cancel_all_background_work(db, 1);

@@ -96,16 +96,16 @@ const Comparator* ColumnFamilyHandleImpl::GetComparator() const {
  return cfd()->user_comparator();
 }

-void GetInternalTblPropCollFactory(
+void GetIntTblPropCollectorFactory(
    const ImmutableCFOptions& ioptions,
-    InternalTblPropCollFactories* internal_tbl_prop_coll_factories) {
-  assert(internal_tbl_prop_coll_factories);
+    IntTblPropCollectorFactories* int_tbl_prop_collector_factories) {
+  assert(int_tbl_prop_collector_factories);

  auto& collector_factories = ioptions.table_properties_collector_factories;
  for (size_t i = 0; i < ioptions.table_properties_collector_factories.size();
       ++i) {
    assert(collector_factories[i]);
-    internal_tbl_prop_coll_factories->emplace_back(
+    int_tbl_prop_collector_factories->emplace_back(
        new UserKeyTablePropertiesCollectorFactory(collector_factories[i]));
  }
 }
@@ -322,8 +322,8 @@ ColumnFamilyOptions SanitizeOptions(const ImmutableDBOptions& db_options,
    }
    ROCKS_LOG_WARN(db_options.logger,
                   "Adjust the value to "
-                   "level0_stop_writes_trigger(%d) "
-                   "level0_slowdown_writes_trigger(%d) "
+                   "level0_stop_writes_trigger(%d)"
+                   "level0_slowdown_writes_trigger(%d)"
                   "level0_file_num_compaction_trigger(%d)",
                   result.level0_stop_writes_trigger,
                   result.level0_slowdown_writes_trigger,
@@ -358,16 +358,16 @@ ColumnFamilyOptions SanitizeOptions(const ImmutableDBOptions& db_options,

  if (result.level_compaction_dynamic_level_bytes) {
    if (result.compaction_style != kCompactionStyleLevel) {
-      ROCKS_LOG_INFO(db_options.info_log.get(),
-                     "level_compaction_dynamic_level_bytes only makes sense "
+      ROCKS_LOG_WARN(db_options.info_log.get(),
+                     "level_compaction_dynamic_level_bytes only makes sense"
                     "for level-based compaction");
      result.level_compaction_dynamic_level_bytes = false;
    } else if (result.cf_paths.size() > 1U) {
      // we don't yet know how to make both of this feature and multiple
      // DB path work.
      ROCKS_LOG_WARN(db_options.info_log.get(),
-                     "multiple cf_paths/db_paths and "
-                     "level_compaction_dynamic_level_bytes "
+                     "multiple cf_paths/db_paths and"
+                     "level_compaction_dynamic_level_bytes"
                     "can't be used together");
      result.level_compaction_dynamic_level_bytes = false;
    }
@@ -572,7 +572,7 @@ ColumnFamilyData::ColumnFamilyData(
  Ref();

  // Convert user defined table properties collector factories to internal ones.
-  GetInternalTblPropCollFactory(ioptions_, &internal_tbl_prop_coll_factories_);
+  GetIntTblPropCollectorFactory(ioptions_, &int_tbl_prop_collector_factories_);

  // if _dummy_versions is nullptr, then this is a dummy column family.
  if (_dummy_versions != nullptr) {
@@ -833,8 +833,8 @@ std::unique_ptr<WriteControllerToken> SetupDelay(
  return write_controller->GetDelayToken(write_rate);
 }

-int GetL0FileCountForCompactionSpeedup(int level0_file_num_compaction_trigger,
-                                       int level0_slowdown_writes_trigger) {
+int GetL0ThresholdSpeedupCompaction(int level0_file_num_compaction_trigger,
+                                    int level0_slowdown_writes_trigger) {
  // SanitizeOptions() ensures it.
  assert(level0_file_num_compaction_trigger <= level0_slowdown_writes_trigger);

@@ -864,43 +864,6 @@ int GetL0FileCountForCompactionSpeedup(int level0_file_num_compaction_trigger,
    return static_cast<int>(res);
  }
 }
-
-uint64_t GetPendingCompactionBytesForCompactionSpeedup(
-    const MutableCFOptions& mutable_cf_options,
-    const VersionStorageInfo* vstorage) {
-  // Compaction debt relatively large compared to the stable (bottommost) data
-  // size indicates compaction fell behind.
-  const uint64_t kBottommostSizeDivisor = 8;
-  // Meaningful progress toward the slowdown trigger is another good indication.
-  const uint64_t kSlowdownTriggerDivisor = 4;
-
-  uint64_t bottommost_files_size = 0;
-  for (const auto& level_and_file : vstorage->BottommostFiles()) {
-    bottommost_files_size += level_and_file.second->fd.GetFileSize();
-  }
-
-  // Slowdown trigger might be zero but that means compaction speedup should
-  // always happen (undocumented/historical), so no special treatment is needed.
-  uint64_t slowdown_threshold =
-      mutable_cf_options.soft_pending_compaction_bytes_limit /
-      kSlowdownTriggerDivisor;
-
-  // Size of zero, however, should not be used to decide to speedup compaction.
-  if (bottommost_files_size == 0) {
-    return slowdown_threshold;
-  }
-
-  uint64_t size_threshold = bottommost_files_size / kBottommostSizeDivisor;
-  return std::min(size_threshold, slowdown_threshold);
-}
-
-uint64_t GetMarkedFileCountForCompactionSpeedup() {
-  // When just one file is marked, it is not clear that parallel compaction will
-  // help the compaction that the user nicely requested to happen sooner. When
-  // multiple files are marked, however, it is pretty clearly helpful, except
-  // for the rare case in which a single compaction grabs all the marked files.
-  return 2;
-}
 }  // anonymous namespace

 std::pair<WriteStallCondition, WriteStallCause>
@@ -1056,7 +1019,7 @@ WriteStallCondition ColumnFamilyData::RecalculateWriteStallConditions(
    } else {
      assert(write_stall_condition == WriteStallCondition::kNormal);
      if (vstorage->l0_delay_trigger_count() >=
-          GetL0FileCountForCompactionSpeedup(
+          GetL0ThresholdSpeedupCompaction(
              mutable_cf_options.level0_file_num_compaction_trigger,
              mutable_cf_options.level0_slowdown_writes_trigger)) {
        write_controller_token_ =
@@ -1066,32 +1029,22 @@ WriteStallCondition ColumnFamilyData::RecalculateWriteStallConditions(
            "[%s] Increasing compaction threads because we have %d level-0 "
            "files ",
            name_.c_str(), vstorage->l0_delay_trigger_count());
-      } else if (mutable_cf_options.soft_pending_compaction_bytes_limit == 0) {
+      } else if (vstorage->estimated_compaction_needed_bytes() >=
+                 mutable_cf_options.soft_pending_compaction_bytes_limit / 4) {
+        // Increase compaction threads if bytes needed for compaction exceeds
+        // 1/4 of threshold for slowing down.
        // If soft pending compaction byte limit is not set, always speed up
        // compaction.
        write_controller_token_ =
            write_controller->GetCompactionPressureToken();
-      } else if (vstorage->estimated_compaction_needed_bytes() >=
-                 GetPendingCompactionBytesForCompactionSpeedup(
-                     mutable_cf_options, vstorage)) {
-        write_controller_token_ =
-            write_controller->GetCompactionPressureToken();
-        ROCKS_LOG_INFO(
-            ioptions_.logger,
-            "[%s] Increasing compaction threads because of estimated pending "
-            "compaction "
-            "bytes %" PRIu64,
-            name_.c_str(), vstorage->estimated_compaction_needed_bytes());
-      } else if (uint64_t(vstorage->FilesMarkedForCompaction().size()) >=
-                 GetMarkedFileCountForCompactionSpeedup()) {
-        write_controller_token_ =
-            write_controller->GetCompactionPressureToken();
-        ROCKS_LOG_INFO(
-            ioptions_.logger,
-            "[%s] Increasing compaction threads because we have %" PRIu64
-            " files marked for compaction",
-            name_.c_str(),
-            uint64_t(vstorage->FilesMarkedForCompaction().size()));
+        if (mutable_cf_options.soft_pending_compaction_bytes_limit > 0) {
+          ROCKS_LOG_INFO(
+              ioptions_.logger,
+              "[%s] Increasing compaction threads because of estimated pending "
+              "compaction "
+              "bytes %" PRIu64,
+              name_.c_str(), vstorage->estimated_compaction_needed_bytes());
+        }
      } else {
        write_controller_token_.reset();
      }
@@ -1179,13 +1132,13 @@ bool ColumnFamilyData::RangeOverlapWithCompaction(
 }

 Status ColumnFamilyData::RangesOverlapWithMemtables(
-    const autovector<UserKeyRange>& ranges, SuperVersion* super_version,
+    const autovector<Range>& ranges, SuperVersion* super_version,
    bool allow_data_in_errors, bool* overlap) {
  assert(overlap != nullptr);
  *overlap = false;
  // Create an InternalIterator over all unflushed memtables
  Arena arena;
-  // TODO: plumb Env::IOActivity, Env::IOPriority
+  // TODO: plumb Env::IOActivity
  ReadOptions read_opts;
  read_opts.total_order_seek = true;
  MergeIteratorBuilder merge_iter_builder(&internal_comparator_, &arena);
@@ -1223,8 +1176,7 @@ Status ColumnFamilyData::RangesOverlapWithMemtables(

    if (status.ok()) {
      if (memtable_iter->Valid() &&
-          ucmp->CompareWithoutTimestamp(seek_result.user_key,
-                                        ranges[i].limit) <= 0) {
+          ucmp->Compare(seek_result.user_key, ranges[i].limit) <= 0) {
        *overlap = true;
      } else if (range_del_agg.IsRangeOverlapped(ranges[i].start,
                                                 ranges[i].limit)) {
@@ -1796,20 +1748,4 @@ const Comparator* GetColumnFamilyUserComparator(
  return nullptr;
 }

-const ImmutableOptions& GetImmutableOptions(ColumnFamilyHandle* column_family) {
-  assert(column_family);
-
-  ColumnFamilyHandleImpl* const handle =
-      static_cast_with_check<ColumnFamilyHandleImpl>(column_family);
-  assert(handle);
-
-  const ColumnFamilyData* const cfd = handle->cfd();
-  assert(cfd);
-
-  const ImmutableOptions* ioptions = cfd->ioptions();
-  assert(ioptions);
-
-  return *ioptions;
-}
-
 }  // namespace ROCKSDB_NAMESPACE
@@ -169,10 +169,10 @@ class ColumnFamilyHandleImpl : public ColumnFamilyHandle {
  virtual ~ColumnFamilyHandleImpl();
  virtual ColumnFamilyData* cfd() const { return cfd_; }

-  uint32_t GetID() const override;
-  const std::string& GetName() const override;
-  Status GetDescriptor(ColumnFamilyDescriptor* desc) override;
-  const Comparator* GetComparator() const override;
+  virtual uint32_t GetID() const override;
+  virtual const std::string& GetName() const override;
+  virtual Status GetDescriptor(ColumnFamilyDescriptor* desc) override;
+  virtual const Comparator* GetComparator() const override;

 private:
  ColumnFamilyData* cfd_;
@@ -193,7 +193,7 @@ class ColumnFamilyHandleInternal : public ColumnFamilyHandleImpl {
        internal_cfd_(nullptr) {}

  void SetCFD(ColumnFamilyData* _cfd) { internal_cfd_ = _cfd; }
-  ColumnFamilyData* cfd() const override { return internal_cfd_; }
+  virtual ColumnFamilyData* cfd() const override { return internal_cfd_; }

 private:
  ColumnFamilyData* internal_cfd_;
@@ -251,21 +251,22 @@ struct SuperVersion {
  autovector<MemTable*> to_delete;
 };

-Status CheckCompressionSupported(const ColumnFamilyOptions& cf_options);
+extern Status CheckCompressionSupported(const ColumnFamilyOptions& cf_options);

-Status CheckConcurrentWritesSupported(const ColumnFamilyOptions& cf_options);
+extern Status CheckConcurrentWritesSupported(
+    const ColumnFamilyOptions& cf_options);

-Status CheckCFPathsSupported(const DBOptions& db_options,
-                             const ColumnFamilyOptions& cf_options);
+extern Status CheckCFPathsSupported(const DBOptions& db_options,
+                                    const ColumnFamilyOptions& cf_options);

-ColumnFamilyOptions SanitizeOptions(const ImmutableDBOptions& db_options,
-                                    const ColumnFamilyOptions& src);
+extern ColumnFamilyOptions SanitizeOptions(const ImmutableDBOptions& db_options,
+                                           const ColumnFamilyOptions& src);
 // Wrap user defined table properties collector factories `from cf_options`
-// into internal ones in internal_tbl_prop_coll_factories. Add a system internal
+// into internal ones in int_tbl_prop_collector_factories. Add a system internal
 // one too.
-void GetInternalTblPropCollFactory(
+extern void GetIntTblPropCollectorFactory(
    const ImmutableCFOptions& ioptions,
-    InternalTblPropCollFactories* internal_tbl_prop_coll_factories);
+    IntTblPropCollectorFactories* int_tbl_prop_collector_factories);

 class ColumnFamilySet;

@@ -400,7 +401,7 @@ class ColumnFamilyData {
  //    duration of this function.
  //
  // Thread-safe
-  Status RangesOverlapWithMemtables(const autovector<UserKeyRange>& ranges,
+  Status RangesOverlapWithMemtables(const autovector<Range>& ranges,
                                    SuperVersion* super_version,
                                    bool allow_data_in_errors, bool* overlap);

@@ -429,8 +430,8 @@ class ColumnFamilyData {
    return internal_comparator_;
  }

-  const InternalTblPropCollFactories* internal_tbl_prop_coll_factories() const {
-    return &internal_tbl_prop_coll_factories_;
+  const IntTblPropCollectorFactories* int_tbl_prop_collector_factories() const {
+    return &int_tbl_prop_collector_factories_;
  }

  SuperVersion* GetSuperVersion() { return super_version_; }
@@ -573,7 +574,7 @@ class ColumnFamilyData {
  std::atomic<bool> dropped_;  // true if client dropped it

  const InternalKeyComparator internal_comparator_;
-  InternalTblPropCollFactories internal_tbl_prop_coll_factories_;
+  IntTblPropCollectorFactories int_tbl_prop_collector_factories_;

  const ColumnFamilyOptions initial_cf_options_;
  const ImmutableOptions ioptions_;
@@ -853,17 +854,17 @@ class ColumnFamilyMemTablesImpl : public ColumnFamilyMemTables {
  // REQUIRES: Seek() called first
  // REQUIRES: use this function of DBImpl::column_family_memtables_ should be
  //           under a DB mutex OR from a write thread
-  MemTable* GetMemTable() const override;
+  virtual MemTable* GetMemTable() const override;

  // Returns column family handle for the selected column family
  // REQUIRES: use this function of DBImpl::column_family_memtables_ should be
  //           under a DB mutex OR from a write thread
-  ColumnFamilyHandle* GetColumnFamilyHandle() override;
+  virtual ColumnFamilyHandle* GetColumnFamilyHandle() override;

  // Cannot be called while another thread is calling Seek().
  // REQUIRES: use this function of DBImpl::column_family_memtables_ should be
  //           under a DB mutex OR from a write thread
-  ColumnFamilyData* current() override { return current_; }
+  virtual ColumnFamilyData* current() override { return current_; }

 private:
  ColumnFamilySet* column_family_set_;
@@ -871,11 +872,9 @@ class ColumnFamilyMemTablesImpl : public ColumnFamilyMemTables {
  ColumnFamilyHandleInternal handle_;
 };

-uint32_t GetColumnFamilyID(ColumnFamilyHandle* column_family);
+extern uint32_t GetColumnFamilyID(ColumnFamilyHandle* column_family);

-const Comparator* GetColumnFamilyUserComparator(
+extern const Comparator* GetColumnFamilyUserComparator(
    ColumnFamilyHandle* column_family);

-const ImmutableOptions& GetImmutableOptions(ColumnFamilyHandle* column_family);
-
 }  // namespace ROCKSDB_NAMESPACE
@@ -8,7 +8,6 @@
 // found in the LICENSE file. See the AUTHORS file for names of contributors.

 #include <algorithm>
-#include <atomic>
 #include <string>
 #include <thread>
 #include <vector>
@@ -28,7 +27,6 @@
 #include "test_util/testharness.h"
 #include "test_util/testutil.h"
 #include "util/coding.h"
-#include "util/defer.h"
 #include "util/string_util.h"
 #include "utilities/fault_injection_env.h"
 #include "utilities/merge_operators.h"
@@ -2171,57 +2169,13 @@ TEST_P(ColumnFamilyTest, FlushStaleColumnFamilies) {
  Close();
 }

-namespace {
-struct CountOptionsFilesFs : public FileSystemWrapper {
-  explicit CountOptionsFilesFs(const std::shared_ptr<FileSystem>& t)
-      : FileSystemWrapper(t) {}
-  const char* Name() const override { return "CountOptionsFilesFs"; }
-
-  IOStatus NewWritableFile(const std::string& f, const FileOptions& file_opts,
-                           std::unique_ptr<FSWritableFile>* r,
-                           IODebugContext* dbg) override {
-    if (f.find("OPTIONS-") != std::string::npos) {
-      options_files_created.fetch_add(1, std::memory_order_relaxed);
-    }
-    return FileSystemWrapper::NewWritableFile(f, file_opts, r, dbg);
-  }
-
-  std::atomic<int> options_files_created{};
-};
-}  // namespace
-
 TEST_P(ColumnFamilyTest, CreateMissingColumnFamilies) {
-  // Can't accidentally add CFs to an existing DB
-  Open();
-  Close();
-  ASSERT_FALSE(db_options_.create_missing_column_families);
-  ASSERT_NOK(TryOpen({"one", "two"}));
-
-  // Nor accidentally create in a new DB
-  Destroy();
-  db_options_.create_if_missing = true;
-  ASSERT_NOK(TryOpen({"one", "two"}));
-
-  // Only with the option (new DB case)
+  Status s = TryOpen({"one", "two"});
+  ASSERT_TRUE(!s.ok());
  db_options_.create_missing_column_families = true;
-  // Also setup to count number of options files created (see check below)
-  auto my_fs =
-      std::make_shared<CountOptionsFilesFs>(db_options_.env->GetFileSystem());
-  auto my_env = std::make_unique<CompositeEnvWrapper>(db_options_.env, my_fs);
-  SaveAndRestore<Env*> save_restore_env(&db_options_.env, my_env.get());
-
-  ASSERT_OK(TryOpen({"default", "one", "two"}));
+  s = TryOpen({"default", "one", "two"});
+  ASSERT_TRUE(s.ok());
  Close();
-
-  // An older version would write an updated options file for each column
-  // family created under create_missing_column_families, which would be
-  // quadratic I/O in the number of column families.
-  ASSERT_EQ(my_fs->options_files_created.load(), 1);
-
-  // Add to existing DB case
-  ASSERT_OK(TryOpen({"default", "one", "two", "three", "four"}));
-  Close();
-  ASSERT_EQ(my_fs->options_files_created.load(), 2);
 }

 TEST_P(ColumnFamilyTest, SanitizeOptions) {
@@ -2473,10 +2427,7 @@ void DropSingleColumnFamily(ColumnFamilyTest* cf_test, int cf_id,
 }
 }  // anonymous namespace

-// This test attempts to set up a race condition in a way that is no longer
-// possible, causing the test to hang. If DBImpl::options_mutex_ is removed
-// in the future, this test might become relevant again.
-TEST_P(ColumnFamilyTest, DISABLED_CreateAndDropRace) {
+TEST_P(ColumnFamilyTest, CreateAndDropRace) {
  const int kCfCount = 5;
  std::vector<ColumnFamilyOptions> cf_opts;
  std::vector<Comparator*> comparators;
@@ -2538,53 +2489,6 @@ TEST_P(ColumnFamilyTest, DISABLED_CreateAndDropRace) {
  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
 }

-TEST_P(ColumnFamilyTest, CreateAndDropPeriodicRace) {
-  // This is a mini-stress test looking for inconsistency between the set of
-  // CFs in the DB, particularly whether any use preserve_internal_time_seconds,
-  // and whether that is accurately reflected in the periodic task setup.
-  constexpr size_t kNumThreads = 12;
-  std::vector<std::thread> threads;
-  bool last_cf_on = Random::GetTLSInstance()->OneIn(2);
-
-  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
-      "DBImpl::RegisterRecordSeqnoTimeWorker:BeforePeriodicTaskType",
-      [&](void* /*arg*/) { std::this_thread::yield(); });
-  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
-
-  ASSERT_EQ(column_family_options_.preserve_internal_time_seconds, 0U);
-  ColumnFamilyOptions other_opts = column_family_options_;
-  ColumnFamilyOptions last_opts = column_family_options_;
-  (last_cf_on ? last_opts : other_opts).preserve_internal_time_seconds =
-      1000000;
-  Open();
-
-  for (size_t i = 0; i < kNumThreads; i++) {
-    threads.emplace_back([this, &other_opts, i]() {
-      ColumnFamilyHandle* cfh;
-      ASSERT_OK(db_->CreateColumnFamily(other_opts, std::to_string(i), &cfh));
-      ASSERT_OK(db_->DropColumnFamily(cfh));
-      ASSERT_OK(db_->DestroyColumnFamilyHandle(cfh));
-    });
-  }
-
-  ColumnFamilyHandle* last_cfh;
-  ASSERT_OK(db_->CreateColumnFamily(last_opts, "last", &last_cfh));
-
-  for (auto& t : threads) {
-    t.join();
-  }
-
-  bool task_enabled = dbfull()->TEST_GetPeriodicTaskScheduler().TEST_HasTask(
-      PeriodicTaskType::kRecordSeqnoTime);
-  ASSERT_EQ(last_cf_on, task_enabled);
-
-  ASSERT_OK(db_->DropColumnFamily(last_cfh));
-  ASSERT_OK(db_->DestroyColumnFamilyHandle(last_cfh));
-
-  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
-  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
-}
-
 TEST_P(ColumnFamilyTest, WriteStallSingleColumnFamily) {
  const uint64_t kBaseRate = 800000u;
  db_options_.delayed_write_rate = kBaseRate;
@@ -2604,92 +2508,90 @@ TEST_P(ColumnFamilyTest, WriteStallSingleColumnFamily) {
  mutable_cf_options.hard_pending_compaction_bytes_limit = 2000;
  mutable_cf_options.disable_auto_compactions = false;

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

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

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

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

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

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

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

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

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

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

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

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

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

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

-  vstorage->TEST_set_estimated_compaction_needed_bytes(100, dbmu);
+  vstorage->TEST_set_estimated_compaction_needed_bytes(100);
  RecalculateWriteStallConditions(cfd, mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());
@@ -2708,7 +2610,7 @@ TEST_P(ColumnFamilyTest, WriteStallSingleColumnFamily) {
  ASSERT_EQ(kBaseRate / 1.25, GetDbDelayedWriteRate());

  vstorage->set_l0_delay_trigger_count(0);
-  vstorage->TEST_set_estimated_compaction_needed_bytes(300, dbmu);
+  vstorage->TEST_set_estimated_compaction_needed_bytes(300);
  RecalculateWriteStallConditions(cfd, mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
@@ -2720,14 +2622,14 @@ TEST_P(ColumnFamilyTest, WriteStallSingleColumnFamily) {
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate / 1.25 / 1.25 / 1.25, GetDbDelayedWriteRate());

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

  vstorage->set_l0_delay_trigger_count(0);
-  vstorage->TEST_set_estimated_compaction_needed_bytes(0, dbmu);
+  vstorage->TEST_set_estimated_compaction_needed_bytes(0);
  RecalculateWriteStallConditions(cfd, mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());
@@ -2746,7 +2648,7 @@ TEST_P(ColumnFamilyTest, WriteStallSingleColumnFamily) {
  ASSERT_EQ(kBaseRate, dbfull()->TEST_write_controler().delayed_write_rate());

  vstorage->set_l0_delay_trigger_count(60);
-  vstorage->TEST_set_estimated_compaction_needed_bytes(300, dbmu);
+  vstorage->TEST_set_estimated_compaction_needed_bytes(300);
  RecalculateWriteStallConditions(cfd, mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());
@@ -2755,14 +2657,14 @@ TEST_P(ColumnFamilyTest, WriteStallSingleColumnFamily) {

  mutable_cf_options.disable_auto_compactions = false;
  vstorage->set_l0_delay_trigger_count(70);
-  vstorage->TEST_set_estimated_compaction_needed_bytes(500, dbmu);
+  vstorage->TEST_set_estimated_compaction_needed_bytes(500);
  RecalculateWriteStallConditions(cfd, mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
  ASSERT_EQ(kBaseRate, GetDbDelayedWriteRate());

  vstorage->set_l0_delay_trigger_count(71);
-  vstorage->TEST_set_estimated_compaction_needed_bytes(501, dbmu);
+  vstorage->TEST_set_estimated_compaction_needed_bytes(501);
  RecalculateWriteStallConditions(cfd, mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
@@ -2787,21 +2689,19 @@ TEST_P(ColumnFamilyTest, CompactionSpeedupSingleColumnFamily) {
  mutable_cf_options.soft_pending_compaction_bytes_limit = 200;
  mutable_cf_options.hard_pending_compaction_bytes_limit = 2000;

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

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

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

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

@@ -2857,55 +2757,53 @@ TEST_P(ColumnFamilyTest, WriteStallTwoColumnFamilies) {
  MutableCFOptions mutable_cf_options1 = mutable_cf_options;
  mutable_cf_options1.soft_pending_compaction_bytes_limit = 500;

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

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

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

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

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

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

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

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

-  vstorage1->TEST_set_estimated_compaction_needed_bytes(600, dbmu);
+  vstorage1->TEST_set_estimated_compaction_needed_bytes(600);
  RecalculateWriteStallConditions(cfd1, mutable_cf_options);
  ASSERT_TRUE(!IsDbWriteStopped());
  ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
@@ -2938,31 +2836,29 @@ TEST_P(ColumnFamilyTest, CompactionSpeedupTwoColumnFamilies) {
  MutableCFOptions mutable_cf_options1 = mutable_cf_options;
  mutable_cf_options1.level0_slowdown_writes_trigger = 16;

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

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

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

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

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

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

@@ -2979,116 +2875,6 @@ TEST_P(ColumnFamilyTest, CompactionSpeedupTwoColumnFamilies) {
  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());
 }

-TEST_P(ColumnFamilyTest, CompactionSpeedupForCompactionDebt) {
-  db_options_.max_background_compactions = 6;
-  Open();
-  ColumnFamilyData* cfd =
-      static_cast<ColumnFamilyHandleImpl*>(db_->DefaultColumnFamily())->cfd();
-  MutableCFOptions mutable_cf_options(column_family_options_);
-  mutable_cf_options.soft_pending_compaction_bytes_limit =
-      std::numeric_limits<uint64_t>::max();
-
-  auto dbmu = dbfull()->TEST_Mutex();
-
-  {
-    // No bottommost data, so debt ratio cannot trigger speedup.
-    VersionStorageInfo* vstorage = cfd->current()->storage_info();
-    vstorage->TEST_set_estimated_compaction_needed_bytes(1048576 /* 1MB */,
-                                                         dbmu);
-    RecalculateWriteStallConditions(cfd, mutable_cf_options);
-    ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());
-  }
-
-  // Add a tiny amount of bottommost data.
-  ASSERT_OK(db_->Put(WriteOptions(), "foo", "bar"));
-  ASSERT_OK(db_->Flush(FlushOptions()));
-
-  {
-    // 1MB debt is way bigger than bottommost data so definitely triggers
-    // speedup.
-    VersionStorageInfo* vstorage = cfd->current()->storage_info();
-    vstorage->TEST_set_estimated_compaction_needed_bytes(1048576 /* 1MB */,
-                                                         dbmu);
-    RecalculateWriteStallConditions(cfd, mutable_cf_options);
-    ASSERT_EQ(6, dbfull()->TEST_BGCompactionsAllowed());
-
-    // Eight bytes is way smaller than bottommost data so definitely does not
-    // trigger speedup.
-    vstorage->TEST_set_estimated_compaction_needed_bytes(8, dbmu);
-    RecalculateWriteStallConditions(cfd, mutable_cf_options);
-    ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());
-  }
-}
-
-TEST_P(ColumnFamilyTest, CompactionSpeedupForMarkedFiles) {
-  const int kParallelismLimit = 3;
-  class AlwaysCompactTpc : public TablePropertiesCollector {
-   public:
-    Status Finish(UserCollectedProperties* /* properties */) override {
-      return Status::OK();
-    }
-
-    UserCollectedProperties GetReadableProperties() const override {
-      return UserCollectedProperties{};
-    }
-
-    const char* Name() const override { return "AlwaysCompactTpc"; }
-
-    bool NeedCompact() const override { return true; }
-  };
-
-  class AlwaysCompactTpcf : public TablePropertiesCollectorFactory {
-   public:
-    TablePropertiesCollector* CreateTablePropertiesCollector(
-        TablePropertiesCollectorFactory::Context /* context */) override {
-      return new AlwaysCompactTpc();
-    }
-
-    const char* Name() const override { return "AlwaysCompactTpcf"; }
-  };
-
-  column_family_options_.num_levels = 2;
-  column_family_options_.table_properties_collector_factories.emplace_back(
-      std::make_shared<AlwaysCompactTpcf>());
-  db_options_.max_background_compactions = kParallelismLimit;
-  Open();
-
-  // Make a nonempty last level. Only marked files in upper levels count.
-  ASSERT_OK(db_->Put(WriteOptions(), "foo", "bar"));
-  ASSERT_OK(db_->Flush(FlushOptions()));
-  WaitForCompaction();
-  AssertFilesPerLevel("0,1", 0 /* cf */);
-
-  // Block the compaction thread pool so marked files accumulate in L0.
-  test::SleepingBackgroundTask sleeping_tasks[kParallelismLimit];
-  for (int i = 0; i < kParallelismLimit; i++) {
-    env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask,
-                   &sleeping_tasks[i], Env::Priority::LOW);
-    sleeping_tasks[i].WaitUntilSleeping();
-  }
-
-  // Zero marked upper-level files. No speedup.
-  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());
-  AssertFilesPerLevel("0,1", 0 /* cf */);
-
-  // One marked upper-level file. No speedup.
-  ASSERT_OK(db_->Put(WriteOptions(), "foo", "bar"));
-  ASSERT_OK(db_->Flush(FlushOptions()));
-  ASSERT_EQ(1, dbfull()->TEST_BGCompactionsAllowed());
-  AssertFilesPerLevel("1,1", 0 /* cf */);
-
-  // Two marked upper-level files. Speedup.
-  ASSERT_OK(db_->Put(WriteOptions(), "foo", "bar"));
-  ASSERT_OK(db_->Flush(FlushOptions()));
-  ASSERT_EQ(kParallelismLimit, dbfull()->TEST_BGCompactionsAllowed());
-  AssertFilesPerLevel("2,1", 0 /* cf */);
-
-  for (int i = 0; i < kParallelismLimit; i++) {
-    sleeping_tasks[i].WakeUp();
-    sleeping_tasks[i].WaitUntilDone();
-  }
-}
-
 TEST_P(ColumnFamilyTest, CreateAndDestroyOptions) {
  std::unique_ptr<ColumnFamilyOptions> cfo(new ColumnFamilyOptions());
  ColumnFamilyHandle* cfh;
@@ -13,7 +13,6 @@
 #include <vector>

 #include "db/column_family.h"
-#include "db/dbformat.h"
 #include "logging/logging.h"
 #include "rocksdb/compaction_filter.h"
 #include "rocksdb/sst_partitioner.h"
@@ -22,6 +21,9 @@

 namespace ROCKSDB_NAMESPACE {

+const uint64_t kRangeTombstoneSentinel =
+    PackSequenceAndType(kMaxSequenceNumber, kTypeRangeDeletion);
+
 int sstableKeyCompare(const Comparator* uc, const Slice& a, const Slice& b) {
  auto c = uc->CompareWithoutTimestamp(ExtractUserKey(a), ExtractUserKey(b));
  if (c != 0) {
@@ -113,42 +115,6 @@ void Compaction::GetBoundaryKeys(
  }
 }

-void Compaction::GetBoundaryInternalKeys(
-    VersionStorageInfo* vstorage,
-    const std::vector<CompactionInputFiles>& inputs, InternalKey* smallest_key,
-    InternalKey* largest_key, int exclude_level) {
-  bool initialized = false;
-  const InternalKeyComparator* icmp = vstorage->InternalComparator();
-  for (size_t i = 0; i < inputs.size(); ++i) {
-    if (inputs[i].files.empty() || inputs[i].level == exclude_level) {
-      continue;
-    }
-    if (inputs[i].level == 0) {
-      // we need to consider all files on level 0
-      for (const auto* f : inputs[i].files) {
-        if (!initialized || icmp->Compare(f->smallest, *smallest_key) < 0) {
-          *smallest_key = f->smallest;
-        }
-        if (!initialized || icmp->Compare(f->largest, *largest_key) > 0) {
-          *largest_key = f->largest;
-        }
-        initialized = true;
-      }
-    } else {
-      // we only need to consider the first and last file
-      if (!initialized ||
-          icmp->Compare(inputs[i].files[0]->smallest, *smallest_key) < 0) {
-        *smallest_key = inputs[i].files[0]->smallest;
-      }
-      if (!initialized ||
-          icmp->Compare(inputs[i].files.back()->largest, *largest_key) > 0) {
-        *largest_key = inputs[i].files.back()->largest;
-      }
-      initialized = true;
-    }
-  }
-}
-
 std::vector<CompactionInputFiles> Compaction::PopulateWithAtomicBoundaries(
    VersionStorageInfo* vstorage, std::vector<CompactionInputFiles> inputs) {
  const Comparator* ucmp = vstorage->InternalComparator()->user_comparator();
@@ -350,9 +316,11 @@ Compaction::Compaction(

  // for the non-bottommost levels, it tries to build files match the target
  // file size, but not guaranteed. It could be 2x the size of the target size.
-  max_output_file_size_ = bottommost_level_ || grandparents_.empty()
-                              ? target_output_file_size_
-                              : 2 * target_output_file_size_;
+  max_output_file_size_ =
+      bottommost_level_ || grandparents_.empty() ||
+              !_immutable_options.level_compaction_dynamic_file_size
+          ? target_output_file_size_
+          : 2 * target_output_file_size_;

 #ifndef NDEBUG
  for (size_t i = 1; i < inputs_.size(); ++i) {
@@ -431,14 +399,9 @@ void Compaction::PopulatePenultimateLevelOutputRange() {
    }
  }

-  // FIXME: should make use of `penultimate_output_range_type_`.
-  // FIXME: when last level's input range does not overlap with
-  //  penultimate level, and penultimate level input is empty,
-  //  this call will not set penultimate_level_smallest_ or
-  //  penultimate_level_largest_. No keys will be compacted up.
-  GetBoundaryInternalKeys(input_vstorage_, inputs_,
-                          &penultimate_level_smallest_,
-                          &penultimate_level_largest_, exclude_level);
+  GetBoundaryKeys(input_vstorage_, inputs_,
+                  &penultimate_level_smallest_user_key_,
+                  &penultimate_level_largest_user_key_, exclude_level);
 }

 Compaction::~Compaction() {
@@ -463,11 +426,23 @@ bool Compaction::OverlapPenultimateLevelOutputRange(
  if (!SupportsPerKeyPlacement()) {
    return false;
  }
+  const Comparator* ucmp =
+      input_vstorage_->InternalComparator()->user_comparator();

-  // See FIXME in Compaction::PopulatePenultimateLevelOutputRange().
-  // We do not compact any key up in this case.
-  if (penultimate_level_smallest_.size() == 0 ||
-      penultimate_level_largest_.size() == 0) {
+  return ucmp->CompareWithoutTimestamp(
+             smallest_key, penultimate_level_largest_user_key_) <= 0 &&
+         ucmp->CompareWithoutTimestamp(
+             largest_key, penultimate_level_smallest_user_key_) >= 0;
+}
+
+// key includes timestamp if user-defined timestamp is enabled.
+bool Compaction::WithinPenultimateLevelOutputRange(const Slice& key) const {
+  if (!SupportsPerKeyPlacement()) {
+    return false;
+  }
+
+  if (penultimate_level_smallest_user_key_.empty() ||
+      penultimate_level_largest_user_key_.empty()) {
    return false;
  }

@@ -475,28 +450,9 @@ bool Compaction::OverlapPenultimateLevelOutputRange(
      input_vstorage_->InternalComparator()->user_comparator();

  return ucmp->CompareWithoutTimestamp(
-             smallest_key, penultimate_level_largest_.user_key()) <= 0 &&
+             key, penultimate_level_smallest_user_key_) >= 0 &&
         ucmp->CompareWithoutTimestamp(
-             largest_key, penultimate_level_smallest_.user_key()) >= 0;
-}
-
-// key includes timestamp if user-defined timestamp is enabled.
-bool Compaction::WithinPenultimateLevelOutputRange(
-    const ParsedInternalKey& ikey) const {
-  if (!SupportsPerKeyPlacement()) {
-    return false;
-  }
-
-  if (penultimate_level_smallest_.size() == 0 ||
-      penultimate_level_largest_.size() == 0) {
-    return false;
-  }
-
-  const InternalKeyComparator* icmp = input_vstorage_->InternalComparator();
-
-  // op_type of a key can change during compaction, e.g. Merge -> Put.
-  return icmp->CompareKeySeq(ikey, penultimate_level_smallest_.Encode()) >= 0 &&
-         icmp->CompareKeySeq(ikey, penultimate_level_largest_.Encode()) <= 0;
+             key, penultimate_level_largest_user_key_) <= 0;
 }

 bool Compaction::InputCompressionMatchesOutput() const {
@@ -353,6 +353,14 @@ class Compaction {

  Slice GetLargestUserKey() const { return largest_user_key_; }

+  Slice GetPenultimateLevelSmallestUserKey() const {
+    return penultimate_level_smallest_user_key_;
+  }
+
+  Slice GetPenultimateLevelLargestUserKey() const {
+    return penultimate_level_largest_user_key_;
+  }
+
  PenultimateOutputRangeType GetPenultimateOutputRangeType() const {
    return penultimate_output_range_type_;
  }
@@ -375,8 +383,10 @@ class Compaction {
  // per_key_placement feature, which is safe to place the key to the
  // penultimate level. different compaction strategy has different rules.
  // If per_key_placement is not supported, always return false.
+  // TODO: currently it doesn't support moving data from the last level to the
+  //  penultimate level
  //  key includes timestamp if user-defined timestamp is enabled.
-  bool WithinPenultimateLevelOutputRange(const ParsedInternalKey& ikey) const;
+  bool WithinPenultimateLevelOutputRange(const Slice& key) const;

  CompactionReason compaction_reason() const { return compaction_reason_; }

@@ -446,13 +456,6 @@ class Compaction {
                              Slice* smallest_key, Slice* largest_key,
                              int exclude_level = -1);

-  // get the smallest and largest internal key present in files to be compacted
-  static void GetBoundaryInternalKeys(
-      VersionStorageInfo* vstorage,
-      const std::vector<CompactionInputFiles>& inputs,
-      InternalKey* smallest_key, InternalKey* largest_key,
-      int exclude_level = -1);
-
  // populate penultimate level output range, which will be used to determine if
  // a key is safe to output to the penultimate level (details see
  // `Compaction::WithinPenultimateLevelOutputRange()`.
@@ -565,8 +568,8 @@ class Compaction {
  // Key range for penultimate level output
  // includes timestamp if user-defined timestamp is enabled.
  // penultimate_output_range_type_ shows the range type
-  InternalKey penultimate_level_smallest_;
-  InternalKey penultimate_level_largest_;
+  Slice penultimate_level_smallest_user_key_;
+  Slice penultimate_level_largest_user_key_;
  PenultimateOutputRangeType penultimate_output_range_type_ =
      PenultimateOutputRangeType::kNotSupported;
 };
@@ -595,6 +598,6 @@ struct PerKeyPlacementContext {
 #endif /* !NDEBUG */

 // Return sum of sizes of all files in `files`.
-uint64_t TotalFileSize(const std::vector<FileMetaData*>& files);
+extern uint64_t TotalFileSize(const std::vector<FileMetaData*>& files);

 }  // namespace ROCKSDB_NAMESPACE
@@ -1227,7 +1227,7 @@ void CompactionIterator::DecideOutputLevel() {
    // not from this compaction.
    // TODO: add statistic for declined output_to_penultimate_level
    bool safe_to_penultimate_level =
-        compaction_->WithinPenultimateLevelOutputRange(ikey_);
+        compaction_->WithinPenultimateLevelOutputRange(ikey_.user_key);
    if (!safe_to_penultimate_level) {
      output_to_penultimate_level_ = false;
      // It could happen when disable/enable `last_level_temperature` while
@@ -1256,13 +1256,10 @@ void CompactionIterator::PrepareOutput() {
      } else if (ikey_.type == kTypeBlobIndex) {
        GarbageCollectBlobIfNeeded();
      }
+    }

-      // For range del sentinel, we don't use it to cut files for bottommost
-      // compaction. So it should not make a difference which output level we
-      // decide.
-      if (compaction_ != nullptr && compaction_->SupportsPerKeyPlacement()) {
-        DecideOutputLevel();
-      }
+    if (compaction_ != nullptr && compaction_->SupportsPerKeyPlacement()) {
+      DecideOutputLevel();
    }

    // Zeroing out the sequence number leads to better compression.
@@ -119,8 +119,7 @@ class CompactionIterator {
    virtual bool SupportsPerKeyPlacement() const = 0;

    // `key` includes timestamp if user-defined timestamp is enabled.
-    virtual bool WithinPenultimateLevelOutputRange(
-        const ParsedInternalKey&) const = 0;
+    virtual bool WithinPenultimateLevelOutputRange(const Slice& key) const = 0;
  };

  class RealCompaction : public CompactionProxy {
@@ -187,9 +186,8 @@ class CompactionIterator {
    // Check if key is within penultimate level output range, to see if it's
    // safe to output to the penultimate level for per_key_placement feature.
    // `key` includes timestamp if user-defined timestamp is enabled.
-    bool WithinPenultimateLevelOutputRange(
-        const ParsedInternalKey& ikey) const override {
-      return compaction_->WithinPenultimateLevelOutputRange(ikey);
+    bool WithinPenultimateLevelOutputRange(const Slice& key) const override {
+      return compaction_->WithinPenultimateLevelOutputRange(key);
    }

   private:
@@ -184,9 +184,8 @@ class FakeCompaction : public CompactionIterator::CompactionProxy {
    return supports_per_key_placement;
  }

-  bool WithinPenultimateLevelOutputRange(
-      const ParsedInternalKey& key) const override {
-    return (!key.user_key.starts_with("unsafe_pb"));
+  bool WithinPenultimateLevelOutputRange(const Slice& key) const override {
+    return (!key.starts_with("unsafe_pb"));
  }

  bool key_not_exists_beyond_output_level = false;
@@ -288,47 +288,45 @@ void CompactionJob::Prepare() {

  if (preserve_time_duration > 0) {
    const ReadOptions read_options(Env::IOActivity::kCompaction);
-    // Setup seqno_to_time_mapping_ with relevant time range.
-    seqno_to_time_mapping_.SetMaxTimeSpan(preserve_time_duration);
+    // setup seqno_time_mapping_
+    seqno_time_mapping_.SetMaxTimeDuration(preserve_time_duration);
    for (const auto& each_level : *c->inputs()) {
      for (const auto& fmd : each_level.files) {
        std::shared_ptr<const TableProperties> tp;
        Status s =
            cfd->current()->GetTableProperties(read_options, &tp, fmd, nullptr);
        if (s.ok()) {
-          s = seqno_to_time_mapping_.DecodeFrom(tp->seqno_to_time_mapping);
-        }
-        if (!s.ok()) {
-          ROCKS_LOG_WARN(
-              db_options_.info_log,
-              "Problem reading or processing seqno-to-time mapping: %s",
-              s.ToString().c_str());
+          seqno_time_mapping_.Add(tp->seqno_to_time_mapping)
+              .PermitUncheckedError();
+          seqno_time_mapping_.Add(fmd->fd.smallest_seqno,
+                                  fmd->oldest_ancester_time);
        }
      }
    }

+    auto status = seqno_time_mapping_.Sort();
+    if (!status.ok()) {
+      ROCKS_LOG_WARN(db_options_.info_log,
+                     "Invalid sequence number to time mapping: Status: %s",
+                     status.ToString().c_str());
+    }
    int64_t _current_time = 0;
-    Status s = db_options_.clock->GetCurrentTime(&_current_time);
-    if (!s.ok()) {
+    status = db_options_.clock->GetCurrentTime(&_current_time);
+    if (!status.ok()) {
      ROCKS_LOG_WARN(db_options_.info_log,
                     "Failed to get current time in compaction: Status: %s",
-                     s.ToString().c_str());
+                     status.ToString().c_str());
      // preserve all time information
      preserve_time_min_seqno_ = 0;
      preclude_last_level_min_seqno_ = 0;
-      seqno_to_time_mapping_.Enforce();
    } else {
-      seqno_to_time_mapping_.Enforce(_current_time);
+      seqno_time_mapping_.TruncateOldEntries(_current_time);
      uint64_t preserve_time =
          static_cast<uint64_t>(_current_time) > preserve_time_duration
              ? _current_time - preserve_time_duration
              : 0;
-      // GetProximalSeqnoBeforeTime tells us the last seqno known to have been
-      // written at or before the given time. + 1 to get the minimum we should
-      // preserve without excluding anything that might have been written on or
-      // after the given time.
      preserve_time_min_seqno_ =
-          seqno_to_time_mapping_.GetProximalSeqnoBeforeTime(preserve_time) + 1;
+          seqno_time_mapping_.GetOldestSequenceNum(preserve_time);
      if (c->immutable_options()->preclude_last_level_data_seconds > 0) {
        uint64_t preclude_last_level_time =
            static_cast<uint64_t>(_current_time) >
@@ -337,21 +335,9 @@ void CompactionJob::Prepare() {
                      c->immutable_options()->preclude_last_level_data_seconds
                : 0;
        preclude_last_level_min_seqno_ =
-            seqno_to_time_mapping_.GetProximalSeqnoBeforeTime(
-                preclude_last_level_time) +
-            1;
+            seqno_time_mapping_.GetOldestSequenceNum(preclude_last_level_time);
      }
    }
-    // For accuracy of the GetProximalSeqnoBeforeTime queries above, we only
-    // limit the capacity after them.
-    // Here If we set capacity to the per-SST limit, we could be throwing away
-    // fidelity when a compaction output file has a narrower seqno range than
-    // all the inputs. If we only limit capacity for each compaction output, we
-    // could be doing a lot of unnecessary recomputation in a large compaction
-    // (up to quadratic in number of files). Thus, we do soemthing in the
-    // middle: enforce a resonably large constant size limit substantially
-    // larger than kMaxSeqnoTimePairsPerSST.
-    seqno_to_time_mapping_.SetCapacity(kMaxSeqnoToTimeEntries);
  }
 }

@@ -450,6 +436,14 @@ void CompactionJob::ReleaseSubcompactionResources() {
  ShrinkSubcompactionResources(extra_num_subcompaction_threads_reserved_);
 }

+struct RangeWithSize {
+  Range range;
+  uint64_t size;
+
+  RangeWithSize(const Slice& a, const Slice& b, uint64_t s = 0)
+      : range(a, b), size(s) {}
+};
+
 void CompactionJob::GenSubcompactionBoundaries() {
  // The goal is to find some boundary keys so that we can evenly partition
  // the compaction input data into max_subcompactions ranges.
@@ -476,8 +470,7 @@ void CompactionJob::GenSubcompactionBoundaries() {
  // overlap with N-1 other ranges. Since we requested a relatively large number
  // (128) of ranges from each input files, even N range overlapping would
  // cause relatively small inaccuracy.
-  ReadOptions read_options(Env::IOActivity::kCompaction);
-  read_options.rate_limiter_priority = GetRateLimiterPriority();
+  const ReadOptions read_options(Env::IOActivity::kCompaction);
  auto* c = compact_->compaction;
  if (c->max_subcompactions() <= 1 &&
      !(c->immutable_options()->compaction_pri == kRoundRobin &&
@@ -729,9 +722,8 @@ Status CompactionJob::Run() {
        // 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 verify_table_read_options(Env::IOActivity::kCompaction);
-        verify_table_read_options.rate_limiter_priority =
-            GetRateLimiterPriority();
+        const ReadOptions verify_table_read_options(
+            Env::IOActivity::kCompaction);
        InternalIterator* iter = cfd->table_cache()->NewIterator(
            verify_table_read_options, file_options_,
            cfd->internal_comparator(), files_output[file_idx]->meta,
@@ -752,6 +744,7 @@ Status CompactionJob::Run() {

        if (s.ok() && paranoid_file_checks_) {
          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());
@@ -822,27 +815,24 @@ Status CompactionJob::Run() {
    // input keys. So the number of keys it processed is not suitable for
    // verification here.
    // TODO: support verification when trim_ts_ is non-empty.
-    if (!(ts_sz > 0 && !trim_ts_.empty())) {
+    if (!(ts_sz > 0 && !trim_ts_.empty()) &&
+        db_options_.compaction_verify_record_count) {
      assert(compaction_stats_.stats.num_input_records > 0);
      // TODO: verify the number of range deletion entries.
      uint64_t expected =
          compaction_stats_.stats.num_input_records - num_input_range_del;
      uint64_t actual = compaction_job_stats_->num_input_records;
      if (expected != actual) {
-        char scratch[2345];
-        compact_->compaction->Summary(scratch, sizeof(scratch));
        std::string msg =
-            "Compaction number of input keys does not match "
-            "number of keys processed. Expected " +
-            std::to_string(expected) + " but processed " +
-            std::to_string(actual) + ". Compaction summary: " + scratch;
+            "Total number of input records: " + std::to_string(expected) +
+            ", but processed " + std::to_string(actual) + " records.";
        ROCKS_LOG_WARN(
-            db_options_.info_log, "[%s] [JOB %d] Compaction with status: %s",
+            db_options_.info_log, "[%s] [JOB %d] Compaction %s",
            compact_->compaction->column_family_data()->GetName().c_str(),
            job_context_->job_id, msg.c_str());
-        if (db_options_.compaction_verify_record_count) {
-          status = Status::Corruption(msg);
-        }
+        status = Status::Corruption(
+            "Compaction number of input keys does not match number of keys "
+            "processed.");
      }
    }
  }
@@ -1134,9 +1124,6 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
  // (b) CompactionFilter::Decision::kRemoveAndSkipUntil.
  read_options.total_order_seek = true;

-  const WriteOptions write_options(Env::IOPriority::IO_LOW,
-                                   Env::IOActivity::kCompaction);
-
  // Remove the timestamps from boundaries because boundaries created in
  // GenSubcompactionBoundaries doesn't strip away the timestamp.
  size_t ts_sz = cfd->user_comparator()->timestamp_size();
@@ -1271,17 +1258,18 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
          ? new BlobFileBuilder(
                versions_, fs_.get(),
                sub_compact->compaction->immutable_options(),
-                mutable_cf_options, &file_options_, &write_options, db_id_,
-                db_session_id_, job_id_, cfd->GetID(), cfd->GetName(),
+                mutable_cf_options, &file_options_, db_id_, db_session_id_,
+                job_id_, cfd->GetID(), cfd->GetName(), Env::IOPriority::IO_LOW,
                write_hint_, io_tracer_, blob_callback_,
                BlobFileCreationReason::kCompaction, &blob_file_paths,
                sub_compact->Current().GetBlobFileAdditionsPtr())
          : nullptr);

  TEST_SYNC_POINT("CompactionJob::Run():Inprogress");
-  TEST_SYNC_POINT_CALLBACK("CompactionJob::Run():PausingManualCompaction:1",
-                           static_cast<void*>(const_cast<std::atomic<bool>*>(
-                               &manual_compaction_canceled_)));
+  TEST_SYNC_POINT_CALLBACK(
+      "CompactionJob::Run():PausingManualCompaction:1",
+      reinterpret_cast<void*>(
+          const_cast<std::atomic<bool>*>(&manual_compaction_canceled_)));

  const std::string* const full_history_ts_low =
      full_history_ts_low_.empty() ? nullptr : &full_history_ts_low_;
@@ -1329,7 +1317,8 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
  Status status;
  TEST_SYNC_POINT_CALLBACK(
      "CompactionJob::ProcessKeyValueCompaction()::Processing",
-      static_cast<void*>(const_cast<Compaction*>(sub_compact->compaction)));
+      reinterpret_cast<void*>(
+          const_cast<Compaction*>(sub_compact->compaction)));
  uint64_t last_cpu_micros = prev_cpu_micros;
  while (status.ok() && !cfd->IsDropped() && c_iter->Valid()) {
    // Invariant: c_iter.status() is guaranteed to be OK if c_iter->Valid()
@@ -1360,9 +1349,10 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
      break;
    }

-    TEST_SYNC_POINT_CALLBACK("CompactionJob::Run():PausingManualCompaction:2",
-                             static_cast<void*>(const_cast<std::atomic<bool>*>(
-                                 &manual_compaction_canceled_)));
+    TEST_SYNC_POINT_CALLBACK(
+        "CompactionJob::Run():PausingManualCompaction:2",
+        reinterpret_cast<void*>(
+            const_cast<std::atomic<bool>*>(&manual_compaction_canceled_)));
    c_iter->Next();
    if (c_iter->status().IsManualCompactionPaused()) {
      break;
@@ -1580,7 +1570,7 @@ Status CompactionJob::FinishCompactionOutputFile(

  const uint64_t current_entries = outputs.NumEntries();

-  s = outputs.Finish(s, seqno_to_time_mapping_);
+  s = outputs.Finish(s, seqno_time_mapping_);

  if (s.ok()) {
    // With accurate smallest and largest key, we can get a slightly more
@@ -1714,8 +1704,6 @@ Status CompactionJob::InstallCompactionResults(
  db_mutex_->AssertHeld();

  const ReadOptions read_options(Env::IOActivity::kCompaction);
-  const WriteOptions write_options(Env::IOActivity::kCompaction);
-
  auto* compaction = compact_->compaction;
  assert(compaction);

@@ -1798,9 +1786,8 @@ Status CompactionJob::InstallCompactionResults(
  };

  return versions_->LogAndApply(
-      compaction->column_family_data(), mutable_cf_options, read_options,
-      write_options, edit, db_mutex_, db_directory_,
-      /*new_descriptor_log=*/false,
+      compaction->column_family_data(), mutable_cf_options, read_options, edit,
+      db_mutex_, db_directory_, /*new_descriptor_log=*/false,
      /*column_family_options=*/nullptr, manifest_wcb);
 }

@@ -1907,8 +1894,6 @@ Status CompactionJob::OpenCompactionOutputFile(SubcompactionState* sub_compact,
          sub_compact->start.has_value() ? &tmp_start : nullptr,
          sub_compact->end.has_value() ? &tmp_end : nullptr);
  if (oldest_ancester_time == std::numeric_limits<uint64_t>::max()) {
-    // TODO: fix DBSSTTest.GetTotalSstFilesSize and use
-    //  kUnknownOldestAncesterTime
    oldest_ancester_time = current_time;
  }

@@ -1936,6 +1921,8 @@ Status CompactionJob::OpenCompactionOutputFile(SubcompactionState* sub_compact,
    }

    outputs.AddOutput(std::move(meta), cfd->internal_comparator(),
+                      sub_compact->compaction->mutable_cf_options()
+                          ->check_flush_compaction_key_order,
                      paranoid_file_checks_);
  }

@@ -1948,17 +1935,13 @@ Status CompactionJob::OpenCompactionOutputFile(SubcompactionState* sub_compact,
      sub_compact->compaction->immutable_options()->listeners;
  outputs.AssignFileWriter(new WritableFileWriter(
      std::move(writable_file), fname, fo_copy, db_options_.clock, io_tracer_,
-      db_options_.stats, Histograms::SST_WRITE_MICROS, listeners,
-      db_options_.file_checksum_gen_factory.get(),
+      db_options_.stats, listeners, db_options_.file_checksum_gen_factory.get(),
      tmp_set.Contains(FileType::kTableFile), false));

  // TODO(hx235): pass in the correct `oldest_key_time` instead of `0`
-  const ReadOptions read_options(Env::IOActivity::kCompaction);
-  const WriteOptions write_options(Env::IOActivity::kCompaction);
  TableBuilderOptions tboptions(
      *cfd->ioptions(), *(sub_compact->compaction->mutable_cf_options()),
-      read_options, write_options, cfd->internal_comparator(),
-      cfd->internal_tbl_prop_coll_factories(),
+      cfd->internal_comparator(), cfd->int_tbl_prop_collector_factories(),
      sub_compact->compaction->output_compression(),
      sub_compact->compaction->output_compression_opts(), cfd->GetID(),
      cfd->GetName(), sub_compact->compaction->output_level(),
@@ -350,7 +350,7 @@ class CompactionJob {

  // Stores the sequence number to time mapping gathered from all input files
  // it also collects the smallest_seqno -> oldest_ancester_time from the SST.
-  SeqnoToTimeMapping seqno_to_time_mapping_;
+  SeqnoToTimeMapping seqno_time_mapping_;

  // Minimal sequence number for preserving the time information. The time info
  // older than this sequence number won't be preserved after the compaction and
@@ -215,9 +215,7 @@ class CompactionJobTestBase : public testing::Test {
            dbname_, &db_options_, env_options_, table_cache_.get(),
            &write_buffer_manager_, &write_controller_,
            /*block_cache_tracer=*/nullptr,
-            /*io_tracer=*/nullptr, /*db_id=*/"", /*db_session_id=*/"",
-            /*daily_offpeak_time_utc=*/"",
-            /*error_handler=*/nullptr, /*read_only=*/false)),
+            /*io_tracer=*/nullptr, /*db_id*/ "", /*db_session_id*/ "")),
        shutting_down_(false),
        mock_table_factory_(new mock::MockTableFactory()),
        error_handler_(nullptr, db_options_, &mutex_),
@@ -295,14 +293,11 @@ class CompactionJobTestBase : public testing::Test {
    Status s = WritableFileWriter::Create(fs_, table_name, FileOptions(),
                                          &file_writer, nullptr);
    ASSERT_OK(s);
-    const ReadOptions read_options;
-    const WriteOptions write_options;
    std::unique_ptr<TableBuilder> table_builder(
        cf_options_.table_factory->NewTableBuilder(
            TableBuilderOptions(*cfd_->ioptions(), mutable_cf_options_,
-                                read_options, write_options,
                                cfd_->internal_comparator(),
-                                cfd_->internal_tbl_prop_coll_factories(),
+                                cfd_->int_tbl_prop_collector_factories(),
                                CompressionType::kNoCompression,
                                CompressionOptions(), 0 /* column_family_id */,
                                kDefaultColumnFamilyName, -1 /* level */),
@@ -397,7 +392,7 @@ class CompactionJobTestBase : public testing::Test {
    mutex_.Lock();
    EXPECT_OK(versions_->LogAndApply(
        versions_->GetColumnFamilySet()->GetDefault(), mutable_cf_options_,
-        read_options_, write_options_, &edit, &mutex_, nullptr));
+        read_options_, &edit, &mutex_, nullptr));
    mutex_.Unlock();
  }

@@ -545,14 +540,13 @@ class CompactionJobTestBase : public testing::Test {
    ASSERT_OK(s);
    db_options_.info_log = info_log;

-    versions_.reset(new VersionSet(
-        dbname_, &db_options_, env_options_, table_cache_.get(),
-        &write_buffer_manager_, &write_controller_,
-        /*block_cache_tracer=*/nullptr, /*io_tracer=*/nullptr,
-        /*db_id=*/"", /*db_session_id=*/"", /*daily_offpeak_time_utc=*/"",
-        /*error_handler=*/nullptr, /*read_only=*/false));
+    versions_.reset(
+        new VersionSet(dbname_, &db_options_, env_options_, table_cache_.get(),
+                       &write_buffer_manager_, &write_controller_,
+                       /*block_cache_tracer=*/nullptr, /*io_tracer=*/nullptr,
+                       /*db_id*/ "", /*db_session_id*/ ""));
    compaction_job_stats_.Reset();
-    ASSERT_OK(SetIdentityFile(WriteOptions(), env_, dbname_));
+    ASSERT_OK(SetIdentityFile(env_, dbname_));

    VersionEdit new_db;
    new_db.SetLogNumber(0);
@@ -571,11 +565,11 @@ class CompactionJobTestBase : public testing::Test {
      log::Writer log(std::move(file_writer), 0, false);
      std::string record;
      new_db.EncodeTo(&record);
-      s = log.AddRecord(WriteOptions(), record);
+      s = log.AddRecord(record);
    }
    ASSERT_OK(s);
    // Make "CURRENT" file that points to the new manifest file.
-    s = SetCurrentFile(WriteOptions(), fs_.get(), dbname_, 1, nullptr);
+    s = SetCurrentFile(fs_.get(), dbname_, 1, nullptr);

    ASSERT_OK(s);

@@ -739,7 +733,6 @@ class CompactionJobTestBase : public testing::Test {
  MutableCFOptions mutable_cf_options_;
  MutableDBOptions mutable_db_options_;
  const ReadOptions read_options_;
-  const WriteOptions write_options_;
  std::shared_ptr<Cache> table_cache_;
  WriteController write_controller_;
  WriteBufferManager write_buffer_manager_;
@@ -1531,15 +1524,13 @@ TEST_F(CompactionJobTest, VerifyPenultimateLevelOutput) {
      {files0, files1, files2, files3}, input_levels,
      /*verify_func=*/[&](Compaction& comp) {
        for (char c = 'a'; c <= 'z'; c++) {
+          std::string c_str;
+          c_str = c;
+          const Slice key(c_str);
          if (c == 'a') {
-            ParsedInternalKey pik("a", 0U, kTypeValue);
-            ASSERT_FALSE(comp.WithinPenultimateLevelOutputRange(pik));
+            ASSERT_FALSE(comp.WithinPenultimateLevelOutputRange(key));
          } else {
-            std::string c_str{c};
-            // WithinPenultimateLevelOutputRange checks internal key range.
-            // 'z' is the last key, so set seqno properly.
-            ParsedInternalKey pik(c_str, c == 'z' ? 12U : 0U, kTypeValue);
-            ASSERT_TRUE(comp.WithinPenultimateLevelOutputRange(pik));
+            ASSERT_TRUE(comp.WithinPenultimateLevelOutputRange(key));
          }
        }
      });
@@ -1752,9 +1743,23 @@ TEST_F(CompactionJobTest, ResultSerialization) {
  }
 }

-TEST_F(CompactionJobTest, CutForMaxCompactionBytes) {
+class CompactionJobDynamicFileSizeTest
+    : public CompactionJobTestBase,
+      public ::testing::WithParamInterface<bool> {
+ public:
+  CompactionJobDynamicFileSizeTest()
+      : CompactionJobTestBase(
+            test::PerThreadDBPath("compaction_job_dynamic_file_size_test"),
+            BytewiseComparator(), [](uint64_t /*ts*/) { return ""; },
+            /*test_io_priority=*/false, TableTypeForTest::kMockTable) {}
+};
+
+TEST_P(CompactionJobDynamicFileSizeTest, CutForMaxCompactionBytes) {
  // dynamic_file_size option should have no impact on cutting for max
  // compaction bytes.
+  bool enable_dyanmic_file_size = GetParam();
+  cf_options_.level_compaction_dynamic_file_size = enable_dyanmic_file_size;
+
  NewDB();
  mutable_cf_options_.target_file_size_base = 80;
  mutable_cf_options_.max_compaction_bytes = 21;
@@ -1828,7 +1833,10 @@ TEST_F(CompactionJobTest, CutForMaxCompactionBytes) {
                {expected_file1, expected_file2});
 }

-TEST_F(CompactionJobTest, CutToSkipGrandparentFile) {
+TEST_P(CompactionJobDynamicFileSizeTest, CutToSkipGrandparentFile) {
+  bool enable_dyanmic_file_size = GetParam();
+  cf_options_.level_compaction_dynamic_file_size = enable_dyanmic_file_size;
+
  NewDB();
  // Make sure the grandparent level file size (10) qualifies skipping.
  // Currently, it has to be > 1/8 of target file size.
@@ -1863,15 +1871,28 @@ TEST_F(CompactionJobTest, CutToSkipGrandparentFile) {
      mock::MakeMockFile({{KeyStr("x", 4U, kTypeValue), "val"},
                          {KeyStr("z", 6U, kTypeValue), "val3"}});

+  auto expected_file_disable_dynamic_file_size =
+      mock::MakeMockFile({{KeyStr("a", 5U, kTypeValue), "val2"},
+                          {KeyStr("c", 3U, kTypeValue), "val"},
+                          {KeyStr("x", 4U, kTypeValue), "val"},
+                          {KeyStr("z", 6U, kTypeValue), "val3"}});
+
  SetLastSequence(6U);
  const std::vector<int> input_levels = {0, 1};
  auto lvl0_files = cfd_->current()->storage_info()->LevelFiles(0);
  auto lvl1_files = cfd_->current()->storage_info()->LevelFiles(1);
+  if (enable_dyanmic_file_size) {
    RunCompaction({lvl0_files, lvl1_files}, input_levels,
                  {expected_file1, expected_file2});
+  } else {
+    RunCompaction({lvl0_files, lvl1_files}, input_levels,
+                  {expected_file_disable_dynamic_file_size});
+  }
 }

-TEST_F(CompactionJobTest, CutToAlignGrandparentBoundary) {
+TEST_P(CompactionJobDynamicFileSizeTest, CutToAlignGrandparentBoundary) {
+  bool enable_dyanmic_file_size = GetParam();
+  cf_options_.level_compaction_dynamic_file_size = enable_dyanmic_file_size;
  NewDB();

  // MockTable has 1 byte per entry by default and each file is 10 bytes.
@@ -1938,15 +1959,40 @@ TEST_F(CompactionJobTest, CutToAlignGrandparentBoundary) {
  }
  expected_file2.emplace_back(KeyStr("s", 4U, kTypeValue), "val");

+  mock::KVVector expected_file_disable_dynamic_file_size1;
+  for (char i = 0; i < 10; i++) {
+    expected_file_disable_dynamic_file_size1.emplace_back(
+        KeyStr(std::string(1, ch + i), i + 10, kTypeValue),
+        "val" + std::to_string(i));
+  }
+
+  mock::KVVector expected_file_disable_dynamic_file_size2;
+  for (char i = 10; i < 12; i++) {
+    expected_file_disable_dynamic_file_size2.emplace_back(
+        KeyStr(std::string(1, ch + i), i + 10, kTypeValue),
+        "val" + std::to_string(i));
+  }
+
+  expected_file_disable_dynamic_file_size2.emplace_back(
+      KeyStr("s", 4U, kTypeValue), "val");
+
  SetLastSequence(22U);
  const std::vector<int> input_levels = {0, 1};
  auto lvl0_files = cfd_->current()->storage_info()->LevelFiles(0);
  auto lvl1_files = cfd_->current()->storage_info()->LevelFiles(1);
+  if (enable_dyanmic_file_size) {
    RunCompaction({lvl0_files, lvl1_files}, input_levels,
                  {expected_file1, expected_file2});
+  } else {
+    RunCompaction({lvl0_files, lvl1_files}, input_levels,
+                  {expected_file_disable_dynamic_file_size1,
+                   expected_file_disable_dynamic_file_size2});
+  }
 }

-TEST_F(CompactionJobTest, CutToAlignGrandparentBoundarySameKey) {
+TEST_P(CompactionJobDynamicFileSizeTest, CutToAlignGrandparentBoundarySameKey) {
+  bool enable_dyanmic_file_size = GetParam();
+  cf_options_.level_compaction_dynamic_file_size = enable_dyanmic_file_size;
  NewDB();

  // MockTable has 1 byte per entry by default and each file is 10 bytes.
@@ -1983,9 +2029,13 @@ TEST_F(CompactionJobTest, CutToAlignGrandparentBoundarySameKey) {
  AddMockFile(file5, 2);

  mock::KVVector expected_file1;
+  mock::KVVector expected_file_disable_dynamic_file_size;
+
  for (int i = 0; i < 8; i++) {
    expected_file1.emplace_back(KeyStr("a", 100 - i, kTypeValue),
                                "val" + std::to_string(100 - i));
+    expected_file_disable_dynamic_file_size.emplace_back(
+        KeyStr("a", 100 - i, kTypeValue), "val" + std::to_string(100 - i));
  }

  // make sure `b` is cut in a separated file (so internally it's not using
@@ -1994,6 +2044,9 @@ TEST_F(CompactionJobTest, CutToAlignGrandparentBoundarySameKey) {
  auto expected_file2 =
      mock::MakeMockFile({{KeyStr("b", 90U, kTypeValue), "valb"}});

+  expected_file_disable_dynamic_file_size.emplace_back(
+      KeyStr("b", 90U, kTypeValue), "valb");
+
  SetLastSequence(122U);
  const std::vector<int> input_levels = {0, 1};
  auto lvl0_files = cfd_->current()->storage_info()->LevelFiles(0);
@@ -2004,13 +2057,20 @@ TEST_F(CompactionJobTest, CutToAlignGrandparentBoundarySameKey) {
  for (int i = 80; i <= 100; i++) {
    snapshots.emplace_back(i);
  }
+  if (enable_dyanmic_file_size) {
    RunCompaction({lvl0_files, lvl1_files}, input_levels,
                  {expected_file1, expected_file2}, snapshots);
+  } else {
+    RunCompaction({lvl0_files, lvl1_files}, input_levels,
+                  {expected_file_disable_dynamic_file_size}, snapshots);
+  }
 }

-TEST_F(CompactionJobTest, CutForMaxCompactionBytesSameKey) {
+TEST_P(CompactionJobDynamicFileSizeTest, CutForMaxCompactionBytesSameKey) {
  // dynamic_file_size option should have no impact on cutting for max
  // compaction bytes.
+  bool enable_dyanmic_file_size = GetParam();
+  cf_options_.level_compaction_dynamic_file_size = enable_dyanmic_file_size;

  NewDB();
  mutable_cf_options_.target_file_size_base = 80;
@@ -2067,6 +2127,9 @@ TEST_F(CompactionJobTest, CutForMaxCompactionBytesSameKey) {
                {expected_file1, expected_file2, expected_file3}, snapshots);
 }

+INSTANTIATE_TEST_CASE_P(CompactionJobDynamicFileSizeTest,
+                        CompactionJobDynamicFileSizeTest, testing::Bool());
+
 class CompactionJobTimestampTest : public CompactionJobTestBase {
 public:
  CompactionJobTimestampTest()
@@ -18,18 +18,16 @@ void CompactionOutputs::NewBuilder(const TableBuilderOptions& tboptions) {
  builder_.reset(NewTableBuilder(tboptions, file_writer_.get()));
 }

-Status CompactionOutputs::Finish(
-    const Status& intput_status,
-    const SeqnoToTimeMapping& seqno_to_time_mapping) {
+Status CompactionOutputs::Finish(const Status& intput_status,
+                                 const SeqnoToTimeMapping& seqno_time_mapping) {
  FileMetaData* meta = GetMetaData();
  assert(meta != nullptr);
  Status s = intput_status;
  if (s.ok()) {
-    SeqnoToTimeMapping relevant_mapping;
-    relevant_mapping.CopyFromSeqnoRange(
-        seqno_to_time_mapping, meta->fd.smallest_seqno, meta->fd.largest_seqno);
-    relevant_mapping.SetCapacity(kMaxSeqnoTimePairsPerSST);
-    builder_->SetSeqnoTimeTableProperties(relevant_mapping,
+    std::string seqno_time_mapping_str;
+    seqno_time_mapping.Encode(seqno_time_mapping_str, meta->fd.smallest_seqno,
+                              meta->fd.largest_seqno, meta->file_creation_time);
+    builder_->SetSeqnoTimeTableProperties(seqno_time_mapping_str,
                                          meta->oldest_ancester_time);
    s = builder_->Finish();

@@ -62,15 +60,12 @@ IOStatus CompactionOutputs::WriterSyncClose(const Status& input_status,
                                            Statistics* statistics,
                                            bool use_fsync) {
  IOStatus io_s;
-  IOOptions opts;
-  io_s = WritableFileWriter::PrepareIOOptions(
-      WriteOptions(Env::IOActivity::kCompaction), opts);
-  if (input_status.ok() && io_s.ok()) {
+  if (input_status.ok()) {
    StopWatch sw(clock, statistics, COMPACTION_OUTFILE_SYNC_MICROS);
-    io_s = file_writer_->Sync(opts, use_fsync);
+    io_s = file_writer_->Sync(use_fsync);
  }
  if (input_status.ok() && io_s.ok()) {
-    io_s = file_writer_->Close(opts);
+    io_s = file_writer_->Close();
  }

  if (input_status.ok() && io_s.ok()) {
@@ -320,6 +315,7 @@ bool CompactionOutputs::ShouldStopBefore(const CompactionIterator& c_iter) {
        being_grandparent_gap_ ? 2 : 3;
    if (compaction_->immutable_options()->compaction_style ==
            kCompactionStyleLevel &&
+        compaction_->immutable_options()->level_compaction_dynamic_file_size &&
        num_grandparent_boundaries_crossed >=
            num_skippable_boundaries_crossed &&
        grandparent_overlapped_bytes_ - previous_overlapped_bytes >
@@ -341,6 +337,7 @@ bool CompactionOutputs::ShouldStopBefore(const CompactionIterator& c_iter) {
    // improvement.
    if (compaction_->immutable_options()->compaction_style ==
            kCompactionStyleLevel &&
+        compaction_->immutable_options()->level_compaction_dynamic_file_size &&
        current_output_file_size_ >=
            ((compaction_->target_output_file_size() + 99) / 100) *
                (50 + std::min(grandparent_boundary_switched_num_ * 5,
@@ -30,9 +30,11 @@ class CompactionOutputs {
  // compaction output file
  struct Output {
    Output(FileMetaData&& _meta, const InternalKeyComparator& _icmp,
-           bool _enable_hash, bool _finished, uint64_t precalculated_hash)
+           bool _enable_order_check, bool _enable_hash, bool _finished,
+           uint64_t precalculated_hash)
        : meta(std::move(_meta)),
-          validator(_icmp, _enable_hash, precalculated_hash),
+          validator(_icmp, _enable_order_check, _enable_hash,
+                    precalculated_hash),
          finished(_finished) {}
    FileMetaData meta;
    OutputValidator validator;
@@ -47,10 +49,10 @@ class CompactionOutputs {

  // Add generated output to the list
  void AddOutput(FileMetaData&& meta, const InternalKeyComparator& icmp,
-                 bool enable_hash, bool finished = false,
-                 uint64_t precalculated_hash = 0) {
-    outputs_.emplace_back(std::move(meta), icmp, enable_hash, finished,
-                          precalculated_hash);
+                 bool enable_order_check, bool enable_hash,
+                 bool finished = false, uint64_t precalculated_hash = 0) {
+    outputs_.emplace_back(std::move(meta), icmp, enable_order_check,
+                          enable_hash, finished, precalculated_hash);
  }

  // Set new table builder for the current output
@@ -105,7 +107,7 @@ class CompactionOutputs {

  // Finish the current output file
  Status Finish(const Status& intput_status,
-                const SeqnoToTimeMapping& seqno_to_time_mapping);
+                const SeqnoToTimeMapping& seqno_time_mapping);

  // Update output table properties from table builder
  void UpdateTableProperties() {
@@ -456,9 +456,10 @@ bool CompactionPicker::IsRangeInCompaction(VersionStorageInfo* vstorage,
 // Returns false if files on parent level are currently in compaction, which
 // means that we can't compact them
 bool CompactionPicker::SetupOtherInputs(
-    const std::string& cf_name, VersionStorageInfo* vstorage,
-    CompactionInputFiles* inputs, CompactionInputFiles* output_level_inputs,
-    int* parent_index, int base_index, bool only_expand_towards_right) {
+    const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
+    VersionStorageInfo* vstorage, CompactionInputFiles* inputs,
+    CompactionInputFiles* output_level_inputs, int* parent_index,
+    int base_index, bool only_expand_towards_right) {
  assert(!inputs->empty());
  assert(output_level_inputs->empty());
  const int input_level = inputs->level;
@@ -499,6 +500,7 @@ bool CompactionPicker::SetupOtherInputs(
  // user key, while excluding other entries for the same user key. This
  // can happen when one user key spans multiple files.
  if (!output_level_inputs->empty()) {
+    const uint64_t limit = mutable_cf_options.max_compaction_bytes;
    const uint64_t output_level_inputs_size =
        TotalFileSize(output_level_inputs->files);
    const uint64_t inputs_size = TotalFileSize(inputs->files);
@@ -525,6 +527,8 @@ bool CompactionPicker::SetupOtherInputs(
      try_overlapping_inputs = false;
    }
    if (try_overlapping_inputs && expanded_inputs.size() > inputs->size() &&
+        (mutable_cf_options.ignore_max_compaction_bytes_for_input ||
+         output_level_inputs_size + expanded_inputs_size < limit) &&
        !AreFilesInCompaction(expanded_inputs.files)) {
      InternalKey new_start, new_limit;
      GetRange(expanded_inputs, &new_start, &new_limit);
@@ -547,6 +551,8 @@ bool CompactionPicker::SetupOtherInputs(
                                             base_index, nullptr);
      expanded_inputs_size = TotalFileSize(expanded_inputs.files);
      if (expanded_inputs.size() > inputs->size() &&
+          (mutable_cf_options.ignore_max_compaction_bytes_for_input ||
+           output_level_inputs_size + expanded_inputs_size < limit) &&
          !AreFilesInCompaction(expanded_inputs.files)) {
        expand_inputs = true;
      }
@@ -806,8 +812,8 @@ Compaction* CompactionPicker::CompactRange(
  output_level_inputs.level = output_level;
  if (input_level != output_level) {
    int parent_index = -1;
-    if (!SetupOtherInputs(cf_name, vstorage, &inputs, &output_level_inputs,
-                          &parent_index, -1)) {
+    if (!SetupOtherInputs(cf_name, mutable_cf_options, vstorage, &inputs,
+                          &output_level_inputs, &parent_index, -1)) {
      // manual compaction is now multi-threaded, so it can
      // happen that SetupOtherInputs fails
      // we handle it higher in RunManualCompaction
@@ -186,6 +186,7 @@ class CompactionPicker {
      int penultimate_level) const;

  bool SetupOtherInputs(const std::string& cf_name,
+                        const MutableCFOptions& mutable_cf_options,
                        VersionStorageInfo* vstorage,
                        CompactionInputFiles* inputs,
                        CompactionInputFiles* output_level_inputs,
@@ -276,7 +277,8 @@ class NullCompactionPicker : public CompactionPicker {
  }

  // Always returns false.
-  bool NeedsCompaction(const VersionStorageInfo* /*vstorage*/) const override {
+  virtual bool NeedsCompaction(
+      const VersionStorageInfo* /*vstorage*/) const override {
    return false;
  }
 };
@@ -17,9 +17,6 @@
 #include "logging/log_buffer.h"
 #include "logging/logging.h"
 #include "options/options_helper.h"
-#include "rocksdb/listener.h"
-#include "rocksdb/statistics.h"
-#include "rocksdb/status.h"
 #include "util/string_util.h"

 namespace ROCKSDB_NAMESPACE {
@@ -18,27 +18,26 @@ class FIFOCompactionPicker : public CompactionPicker {
                       const InternalKeyComparator* icmp)
      : CompactionPicker(ioptions, icmp) {}

-  Compaction* PickCompaction(const std::string& cf_name,
-                             const MutableCFOptions& mutable_cf_options,
-                             const MutableDBOptions& mutable_db_options,
-                             VersionStorageInfo* version,
-                             LogBuffer* log_buffer) override;
+  virtual Compaction* PickCompaction(const std::string& cf_name,
+                                     const MutableCFOptions& mutable_cf_options,
+                                     const MutableDBOptions& mutable_db_options,
+                                     VersionStorageInfo* version,
+                                     LogBuffer* log_buffer) override;

-  Compaction* CompactRange(const std::string& cf_name,
-                           const MutableCFOptions& mutable_cf_options,
-                           const MutableDBOptions& mutable_db_options,
-                           VersionStorageInfo* vstorage, int input_level,
-                           int output_level,
-                           const CompactRangeOptions& compact_range_options,
-                           const InternalKey* begin, const InternalKey* end,
-                           InternalKey** compaction_end, bool* manual_conflict,
-                           uint64_t max_file_num_to_ignore,
-                           const std::string& trim_ts) override;
+  virtual Compaction* CompactRange(
+      const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
+      const MutableDBOptions& mutable_db_options, VersionStorageInfo* vstorage,
+      int input_level, int output_level,
+      const CompactRangeOptions& compact_range_options,
+      const InternalKey* begin, const InternalKey* end,
+      InternalKey** compaction_end, bool* manual_conflict,
+      uint64_t max_file_num_to_ignore, const std::string& trim_ts) override;

  // The maximum allowed output level.  Always returns 0.
-  int MaxOutputLevel() const override { return 0; }
+  virtual int MaxOutputLevel() const override { return 0; }

-  bool NeedsCompaction(const VersionStorageInfo* vstorage) const override;
+  virtual bool NeedsCompaction(
+      const VersionStorageInfo* vstorage) const override;

 private:
  Compaction* PickTTLCompaction(const std::string& cf_name,
@@ -45,13 +45,6 @@ bool LevelCompactionPicker::NeedsCompaction(
 }

 namespace {
-
-enum class CompactToNextLevel {
-  kNo,   // compact to the same level as the input file
-  kYes,  // compact to the next level except the last level to the same level
-  kSkipLastLevel,  // compact to the next level but skip the last level
-};
-
 // A class to build a leveled compaction step-by-step.
 class LevelCompactionBuilder {
 public:
@@ -113,15 +106,6 @@ class LevelCompactionBuilder {
  // otherwise, returns false.
  bool PickIntraL0Compaction();

-  // When total L0 size is small compared to Lbase, try to pick intra-L0
-  // compaction starting from the newest L0 file. This helps to prevent
-  // L0->Lbase compaction with large write-amp.
-  //
-  // Returns true iff an intra-L0 compaction is picked.
-  // `start_level_inputs_` and `output_level_` will be updated accordingly if
-  // a compaction is picked.
-  bool PickSizeBasedIntraL0Compaction();
-
  // Return true if TrivialMove is extended. `start_index` is the index of
  // the initial file picked, which should already be in `start_level_inputs_`.
  bool TryExtendNonL0TrivialMove(int start_index,
@@ -131,10 +115,9 @@ class LevelCompactionBuilder {
  // level_files is a vector of (level, file metadata) in ascending order of
  // level. If compact_to_next_level is true, compact the file to the next
  // level, otherwise, compact to the same level as the input file.
-  // If skip_last_level is true, skip the last level.
  void PickFileToCompact(
      const autovector<std::pair<int, FileMetaData*>>& level_files,
-      CompactToNextLevel compact_to_next_level);
+      bool compact_to_next_level);

  const std::string& cf_name_;
  VersionStorageInfo* vstorage_;
@@ -166,24 +149,20 @@ class LevelCompactionBuilder {

 void LevelCompactionBuilder::PickFileToCompact(
    const autovector<std::pair<int, FileMetaData*>>& level_files,
-    CompactToNextLevel compact_to_next_level) {
+    bool compact_to_next_level) {
  for (auto& level_file : level_files) {
    // If it's being compacted it has nothing to do here.
    // If this assert() fails that means that some function marked some
    // files as being_compacted, but didn't call ComputeCompactionScore()
    assert(!level_file.second->being_compacted);
    start_level_ = level_file.first;
-    if ((compact_to_next_level == CompactToNextLevel::kSkipLastLevel &&
+    if ((compact_to_next_level &&
         start_level_ == vstorage_->num_non_empty_levels() - 1) ||
        (start_level_ == 0 &&
         !compaction_picker_->level0_compactions_in_progress()->empty())) {
      continue;
    }
-
-    // Compact to the next level only if the file is not in the last level and
-    // compact_to_next_level is kYes or kSkipLastLevel.
-    if (compact_to_next_level != CompactToNextLevel::kNo &&
-        (start_level_ < vstorage_->num_non_empty_levels() - 1)) {
+    if (compact_to_next_level) {
      output_level_ =
          (start_level_ == 0) ? vstorage_->base_level() : start_level_ + 1;
    } else {
@@ -269,8 +248,7 @@ void LevelCompactionBuilder::SetupInitialFiles() {
  }

  // Bottommost Files Compaction on deleting tombstones
-  PickFileToCompact(vstorage_->BottommostFilesMarkedForCompaction(),
-                    CompactToNextLevel::kNo);
+  PickFileToCompact(vstorage_->BottommostFilesMarkedForCompaction(), false);
  if (!start_level_inputs_.empty()) {
    compaction_reason_ = CompactionReason::kBottommostFiles;
    return;
@@ -296,26 +274,21 @@ void LevelCompactionBuilder::SetupInitialFiles() {
    }
  }

-  PickFileToCompact(vstorage_->ExpiredTtlFiles(),
-                    CompactToNextLevel::kSkipLastLevel);
+  PickFileToCompact(vstorage_->ExpiredTtlFiles(), true);
  if (!start_level_inputs_.empty()) {
    compaction_reason_ = CompactionReason::kTtl;
    return;
  }

  // Periodic Compaction
-  PickFileToCompact(vstorage_->FilesMarkedForPeriodicCompaction(),
-                    ioptions_.level_compaction_dynamic_level_bytes
-                        ? CompactToNextLevel::kYes
-                        : CompactToNextLevel::kNo);
+  PickFileToCompact(vstorage_->FilesMarkedForPeriodicCompaction(), false);
  if (!start_level_inputs_.empty()) {
    compaction_reason_ = CompactionReason::kPeriodicCompaction;
    return;
  }

  // Forced blob garbage collection
-  PickFileToCompact(vstorage_->FilesMarkedForForcedBlobGC(),
-                    CompactToNextLevel::kNo);
+  PickFileToCompact(vstorage_->FilesMarkedForForcedBlobGC(), false);
  if (!start_level_inputs_.empty()) {
    compaction_reason_ = CompactionReason::kForcedBlobGC;
    return;
@@ -465,8 +438,9 @@ bool LevelCompactionBuilder::SetupOtherInputsIfNeeded() {
    }
    if (!is_l0_trivial_move_ &&
        !compaction_picker_->SetupOtherInputs(
-            cf_name_, vstorage_, &start_level_inputs_, &output_level_inputs_,
-            &parent_index_, base_index_, round_robin_expanding)) {
+            cf_name_, mutable_cf_options_, vstorage_, &start_level_inputs_,
+            &output_level_inputs_, &parent_index_, base_index_,
+            round_robin_expanding)) {
      return false;
    }

@@ -784,9 +758,6 @@ bool LevelCompactionBuilder::PickFileToCompact() {
  // being compacted at level 0.
  if (start_level_ == 0 &&
      !compaction_picker_->level0_compactions_in_progress()->empty()) {
-    if (PickSizeBasedIntraL0Compaction()) {
-      return true;
-    }
    TEST_SYNC_POINT("LevelCompactionPicker::PickCompactionBySize:0");
    return false;
  }
@@ -799,9 +770,6 @@ bool LevelCompactionBuilder::PickFileToCompact() {
  if (TryPickL0TrivialMove()) {
    return true;
  }
-  if (start_level_ == 0 && PickSizeBasedIntraL0Compaction()) {
-    return true;
-  }

  const std::vector<FileMetaData*>& level_files =
      vstorage_->LevelFiles(start_level_);
@@ -906,56 +874,6 @@ bool LevelCompactionBuilder::PickIntraL0Compaction() {
                               mutable_cf_options_.max_compaction_bytes,
                               &start_level_inputs_);
 }
-
-bool LevelCompactionBuilder::PickSizeBasedIntraL0Compaction() {
-  assert(start_level_ == 0);
-  int base_level = vstorage_->base_level();
-  if (base_level <= 0) {
-    return false;
-  }
-  const std::vector<FileMetaData*>& l0_files =
-      vstorage_->LevelFiles(/*level=*/0);
-  size_t min_num_file =
-      std::max(2, mutable_cf_options_.level0_file_num_compaction_trigger);
-  if (l0_files.size() < min_num_file) {
-    return false;
-  }
-  uint64_t l0_size = 0;
-  for (const auto& file : l0_files) {
-    l0_size += file->fd.GetFileSize();
-  }
-  const uint64_t min_lbase_size =
-      l0_size * static_cast<uint64_t>(std::max(
-                    10.0, mutable_cf_options_.max_bytes_for_level_multiplier));
-  assert(min_lbase_size >= l0_size);
-  const std::vector<FileMetaData*>& lbase_files =
-      vstorage_->LevelFiles(/*level=*/base_level);
-  uint64_t lbase_size = 0;
-  for (const auto& file : lbase_files) {
-    lbase_size += file->fd.GetFileSize();
-    if (lbase_size > min_lbase_size) {
-      break;
-    }
-  }
-  if (lbase_size <= min_lbase_size) {
-    return false;
-  }
-
-  start_level_inputs_.clear();
-  start_level_inputs_.level = 0;
-  for (const auto& file : l0_files) {
-    if (file->being_compacted) {
-      break;
-    }
-    start_level_inputs_.files.push_back(file);
-  }
-  if (start_level_inputs_.files.size() < min_num_file) {
-    start_level_inputs_.clear();
-    return false;
-  }
-  output_level_ = 0;
-  return true /* picked an intra-L0 compaction */;
-}
 }  // namespace

 Compaction* LevelCompactionPicker::PickCompaction(
@@ -20,13 +20,14 @@ class LevelCompactionPicker : public CompactionPicker {
  LevelCompactionPicker(const ImmutableOptions& ioptions,
                        const InternalKeyComparator* icmp)
      : CompactionPicker(ioptions, icmp) {}
-  Compaction* PickCompaction(const std::string& cf_name,
-                             const MutableCFOptions& mutable_cf_options,
-                             const MutableDBOptions& mutable_db_options,
-                             VersionStorageInfo* vstorage,
-                             LogBuffer* log_buffer) override;
+  virtual Compaction* PickCompaction(const std::string& cf_name,
+                                     const MutableCFOptions& mutable_cf_options,
+                                     const MutableDBOptions& mutable_db_options,
+                                     VersionStorageInfo* vstorage,
+                                     LogBuffer* log_buffer) override;

-  bool NeedsCompaction(const VersionStorageInfo* vstorage) const override;
+  virtual bool NeedsCompaction(
+      const VersionStorageInfo* vstorage) const override;
 };

 }  // namespace ROCKSDB_NAMESPACE
@@ -85,8 +85,7 @@ class CompactionPickerTestBase : public testing::Test {
    vstorage_.reset(new VersionStorageInfo(
        &icmp_, ucmp_, options_.num_levels, style, nullptr, false,
        EpochNumberRequirement::kMustPresent, ioptions_.clock,
-        options_.bottommost_file_compaction_delay,
-        OffpeakTimeOption(mutable_db_options_.daily_offpeak_time_utc)));
+        options_.bottommost_file_compaction_delay));
    vstorage_->PrepareForVersionAppend(ioptions_, mutable_cf_options_);
  }

@@ -96,8 +95,7 @@ class CompactionPickerTestBase : public testing::Test {
    temp_vstorage_.reset(new VersionStorageInfo(
        &icmp_, ucmp_, options_.num_levels, ioptions_.compaction_style,
        vstorage_.get(), false, EpochNumberRequirement::kMustPresent,
-        ioptions_.clock, options_.bottommost_file_compaction_delay,
-        OffpeakTimeOption(mutable_db_options_.daily_offpeak_time_utc)));
+        ioptions_.clock, options_.bottommost_file_compaction_delay));
  }

  void DeleteVersionStorage() {
@@ -1443,12 +1441,15 @@ TEST_F(CompactionPickerTest, CompactionPriMinOverlapping4) {
  ioptions_.compaction_pri = kMinOverlappingRatio;
  mutable_cf_options_.max_bytes_for_level_base = 10000000;
  mutable_cf_options_.max_bytes_for_level_multiplier = 10;
+  mutable_cf_options_.ignore_max_compaction_bytes_for_input = false;

-  // Overlaps with file 26, 27, ratio is
-  // (60000000U + 60000000U / 180000000U) = 0.67
+  // file 7 and 8 over lap with the same file, but file 8 is smaller so
+  // it will be picked.
+  // Overlaps with file 26, 27. And the file is compensated so will be
+  // picked up.
  Add(2, 6U, "150", "167", 60000000U, 0, 100, 100, 180000000U);
-  Add(2, 7U, "178", "189", 60000000U);  // Overlaps with file 28
-  Add(2, 8U, "401", "500", 61000000U);  // Overlaps with file 29
+  Add(2, 7U, "168", "169", 60000000U);  // Overlaps with file 27
+  Add(2, 8U, "201", "300", 61000000U);  // Overlaps with file 28

  Add(3, 26U, "160", "165", 60000000U);
  // Boosted file size in output level is not considered.
@@ -1462,7 +1463,7 @@ TEST_F(CompactionPickerTest, CompactionPriMinOverlapping4) {
      &log_buffer_));
  ASSERT_TRUE(compaction.get() != nullptr);
  ASSERT_EQ(1U, compaction->num_input_files(0));
-  // Picking file 6 because overlapping ratio is the biggest.
+  // Picking file 8 because overlapping ratio is the biggest.
  ASSERT_EQ(6U, compaction->input(0, 0)->fd.GetNumber());
 }

@@ -2465,23 +2466,20 @@ TEST_F(CompactionPickerTest, IsBottommostLevelTest) {
  DeleteVersionStorage();
 }

-TEST_F(CompactionPickerTest, IgnoreCompactionLimitWhenAddFileFromInputLevel) {
+TEST_F(CompactionPickerTest, MaxCompactionBytesHit) {
  mutable_cf_options_.max_bytes_for_level_base = 1000000u;
  mutable_cf_options_.max_compaction_bytes = 800000u;
+  mutable_cf_options_.ignore_max_compaction_bytes_for_input = false;
  ioptions_.level_compaction_dynamic_level_bytes = false;
  NewVersionStorage(6, kCompactionStyleLevel);
  // A compaction should be triggered and pick file 2 and 5.
-  // It pulls in other compaction input file from the input level L1
-  // without pulling in more output level files.
-  // Files 1, 3, 4 will be included in the compaction.
-  // File 6 is excluded since it overlaps with file 7.
+  // It can expand because adding file 1 and 3, the compaction size will
+  // exceed mutable_cf_options_.max_bytes_for_level_base.
  Add(1, 1U, "100", "150", 300000U);
  Add(1, 2U, "151", "200", 300001U, 0, 0);
  Add(1, 3U, "201", "250", 300000U, 0, 0);
  Add(1, 4U, "251", "300", 300000U, 0, 0);
-  Add(1, 6U, "325", "400", 300000U, 0, 0);
-  Add(2, 5U, "100", "350", 1U);
-  Add(2, 7U, "375", "425", 1U);
+  Add(2, 5U, "100", "256", 1U);
  UpdateVersionStorageInfo();

  std::unique_ptr<Compaction> compaction(level_compaction_picker.PickCompaction(
@@ -2489,12 +2487,37 @@ TEST_F(CompactionPickerTest, IgnoreCompactionLimitWhenAddFileFromInputLevel) {
      &log_buffer_));
  ASSERT_TRUE(compaction.get() != nullptr);
  ASSERT_EQ(2U, compaction->num_input_levels());
-  ASSERT_EQ(4U, compaction->num_input_files(0));
+  ASSERT_EQ(1U, compaction->num_input_files(0));
+  ASSERT_EQ(1U, compaction->num_input_files(1));
+  ASSERT_EQ(2U, compaction->input(0, 0)->fd.GetNumber());
+  ASSERT_EQ(5U, compaction->input(1, 0)->fd.GetNumber());
+}
+
+TEST_F(CompactionPickerTest, MaxCompactionBytesNotHit) {
+  mutable_cf_options_.max_bytes_for_level_base = 800000u;
+  mutable_cf_options_.max_compaction_bytes = 1000000u;
+  mutable_cf_options_.ignore_max_compaction_bytes_for_input = false;
+  ioptions_.level_compaction_dynamic_level_bytes = false;
+  NewVersionStorage(6, kCompactionStyleLevel);
+  // A compaction should be triggered and pick file 2 and 5.
+  // and it expands to file 1 and 3 too.
+  Add(1, 1U, "100", "150", 300000U);
+  Add(1, 2U, "151", "200", 300001U, 0, 0);
+  Add(1, 3U, "201", "250", 300000U, 0, 0);
+  Add(1, 4U, "251", "300", 300000U, 0, 0);
+  Add(2, 5U, "000", "251", 1U);
+  UpdateVersionStorageInfo();
+
+  std::unique_ptr<Compaction> compaction(level_compaction_picker.PickCompaction(
+      cf_name_, mutable_cf_options_, mutable_db_options_, vstorage_.get(),
+      &log_buffer_));
+  ASSERT_TRUE(compaction.get() != nullptr);
+  ASSERT_EQ(2U, compaction->num_input_levels());
+  ASSERT_EQ(3U, compaction->num_input_files(0));
  ASSERT_EQ(1U, compaction->num_input_files(1));
  ASSERT_EQ(1U, compaction->input(0, 0)->fd.GetNumber());
  ASSERT_EQ(2U, compaction->input(0, 1)->fd.GetNumber());
  ASSERT_EQ(3U, compaction->input(0, 2)->fd.GetNumber());
-  ASSERT_EQ(4U, compaction->input(0, 3)->fd.GetNumber());
  ASSERT_EQ(5U, compaction->input(1, 0)->fd.GetNumber());
 }

@@ -4124,177 +4147,9 @@ TEST_P(PerKeyPlacementCompactionPickerTest,
  }
 }

-TEST_F(CompactionPickerTest,
-       LevelCompactionPrioritizeFilesMarkedForCompaction1) {
-  int num_levels = ioptions_.num_levels;
-  ioptions_.level_compaction_dynamic_level_bytes = true;
-  ioptions_.compaction_pri = kMinOverlappingRatio;
-  mutable_cf_options_.max_bytes_for_level_base = 200;
-  mutable_cf_options_.max_bytes_for_level_multiplier = 10;
-  NewVersionStorage(num_levels, kCompactionStyleLevel);
-  // L5
-  // F4 [100, 200] size:100, F5 [300, 400] size:100,  F6 [500, 600] size:200
-  // F5 is marked for compaction
-  // L6
-  // F1 [100, 200] size:100, F2 [300, 400] size:1000, F3 [500, 600] size:100
-  //
-  // First compaction should pick F5 since it's marked for compaction.
-  // Second compaction should pick F6 since it has min overlap ratio.
-  Add(/*level=*/num_levels - 1, /*file_number=*/1U, /*smallest=*/"100",
-      /*largest=*/"200",
-      /*file_size=*/100, /*path_id=*/0, /*smallest_seq=*/0,
-      /*largest_seq=*/0, /*compensated_file_size=*/100,
-      /*marked_for_compact=*/false);
-  Add(/*level=*/num_levels - 1, /*file_number=*/2U, /*smallest=*/"300",
-      /*largest=*/"400",
-      /*file_size=*/1000, /*path_id=*/0, /*smallest_seq=*/0,
-      /*largest_seq=*/0, /*compensated_file_size=*/1000,
-      /*marked_for_compact=*/false);
-  Add(/*level=*/num_levels - 1, /*file_number=*/3U, /*smallest=*/"500",
-      /*largest=*/"600",
-      /*file_size=*/100, /*path_id=*/0, /*smallest_seq=*/0,
-      /*largest_seq=*/0, /*compensated_file_size=*/100,
-      /*marked_for_compact=*/false);
-
-  Add(/*level=*/num_levels - 2, /*file_number=*/4U, /*smallest=*/"100",
-      /*largest=*/"200",
-      /*file_size=*/100, /*path_id=*/0, /*smallest_seq=*/100,
-      /*largest_seq=*/200, /*compensated_file_size=*/100,
-      /*marked_for_compact=*/false);
-  // Marked for compaction, but with a larger overlap ratio.
-  Add(/*level=*/num_levels - 2, /*file_number=*/5U, /*smallest=*/"300",
-      /*largest=*/"400",
-      /*file_size=*/100, /*path_id=*/0, /*smallest_seq=*/300,
-      /*largest_seq=*/400, /*compensated_file_size=*/100,
-      /*marked_for_compact=*/true);
-  Add(/*level=*/num_levels - 2, /*file_number=*/6U, /*smallest=*/"500",
-      /*largest=*/"600",
-      /*file_size=*/200, /*path_id=*/0, /*smallest_seq=*/400,
-      /*largest_seq=*/500, /*compensated_file_size=*/200,
-      /*marked_for_compact=*/false);
-  UpdateVersionStorageInfo();
-
-  std::unique_ptr<Compaction> compaction(level_compaction_picker.PickCompaction(
-      cf_name_, mutable_cf_options_, mutable_db_options_, vstorage_.get(),
-      &log_buffer_));
-  ASSERT_TRUE(compaction);
-  ASSERT_EQ(num_levels - 2, compaction->start_level());
-  ASSERT_EQ(num_levels - 1, compaction->output_level());
-  ASSERT_EQ(1U, compaction->num_input_files(0));
-  ASSERT_EQ(5U, compaction->input(0, 0)->fd.GetNumber());
-
-  std::unique_ptr<Compaction> second_compaction(
-      level_compaction_picker.PickCompaction(cf_name_, mutable_cf_options_,
-                                             mutable_db_options_,
-                                             vstorage_.get(), &log_buffer_));
-  ASSERT_TRUE(second_compaction);
-  ASSERT_EQ(num_levels - 1, compaction->output_level());
-  ASSERT_EQ(num_levels - 2, compaction->start_level());
-  ASSERT_EQ(1U, compaction->num_input_files(0));
-  ASSERT_EQ(6U, second_compaction->input(0, 0)->fd.GetNumber());
-}
-
-TEST_F(CompactionPickerTest,
-       LevelCompactionPrioritizeFilesMarkedForCompaction2) {
-  int num_levels = ioptions_.num_levels;
-  ioptions_.level_compaction_dynamic_level_bytes = true;
-  ioptions_.compaction_pri = kMinOverlappingRatio;
-  mutable_cf_options_.max_bytes_for_level_base = 200;
-  mutable_cf_options_.max_bytes_for_level_multiplier = 10;
-  // L4
-  // F3 [100, 200] size:2000
-  // L5
-  // F2 [100, 200] size:2000, marked for compaction
-  // L6
-  // F1 [100, 200] size: 20000
-  //
-  // L4 should be prioritized over L5 since L4 has a higher compaction score.
-  // Files marked for compaction do not affect level picking order.
-  NewVersionStorage(num_levels, kCompactionStyleLevel);
-  Add(/*level=*/num_levels - 1, /*file_number=*/1U, /*smallest=*/"100",
-      /*largest=*/"200",
-      /*file_size=*/20000, /*path_id=*/0, /*smallest_seq=*/0,
-      /*largest_seq=*/0, /*compensated_file_size=*/100,
-      /*marked_for_compact=*/false);
-  // Level score should be 1.
-  Add(/*level=*/num_levels - 2, /*file_number=*/2U, /*smallest=*/"100",
-      /*largest=*/"200",
-      /*file_size=*/2000, /*path_id=*/0, /*smallest_seq=*/100,
-      /*largest_seq=*/200, /*compensated_file_size=*/2000,
-      /*marked_for_compact=*/true);
-  // Level score should be larger than L5.
-  Add(/*level=*/num_levels - 3, /*file_number=*/3U, /*smallest=*/"100",
-      /*largest=*/"200",
-      /*file_size=*/2000, /*path_id=*/0, /*smallest_seq=*/300,
-      /*largest_seq=*/400, /*compensated_file_size=*/2000,
-      /*marked_for_compact=*/false);
-  UpdateVersionStorageInfo();
-
-  std::unique_ptr<Compaction> compaction(level_compaction_picker.PickCompaction(
-      cf_name_, mutable_cf_options_, mutable_db_options_, vstorage_.get(),
-      &log_buffer_));
-  ASSERT_TRUE(compaction);
-  ASSERT_EQ(num_levels - 3, compaction->start_level());
-  ASSERT_EQ(num_levels - 2, compaction->output_level());
-  ASSERT_EQ(1U, compaction->num_input_files(0));
-  ASSERT_EQ(3U, compaction->input(0, 0)->fd.GetNumber());
-}
-
 INSTANTIATE_TEST_CASE_P(PerKeyPlacementCompactionPickerTest,
                        PerKeyPlacementCompactionPickerTest, ::testing::Bool());

-TEST_F(CompactionPickerTest, IntraL0WhenL0IsSmall) {
-  mutable_cf_options_.level0_file_num_compaction_trigger = 4;
-  mutable_cf_options_.max_bytes_for_level_multiplier = 10;
-  for (const uint64_t lbase_size_multiplier : {1, 10, 11, 40}) {
-    SCOPED_TRACE("lbase_size_multiplier=" +
-                 std::to_string(lbase_size_multiplier));
-    NewVersionStorage(6, kCompactionStyleLevel);
-    // When L0 size is <= Lbase size / max_bytes_for_level_multiplier,
-    // intra-L0 compaction is picked. Otherwise, L0->L1
-    // compaction is picked.
-    Add(/*level=*/0, /*file_number=*/1U, /*smallest=*/"100",
-        /*largest=*/"200", /*file_size=*/1000, /*path_id=*/0,
-        /*smallest_seq=*/10, /*largest_seq=*/11,
-        /*compensated_file_size=*/1000);
-    Add(/*level=*/0, /*file_number=*/2U, /*smallest=*/"100",
-        /*largest=*/"100", /*file_size=*/1000, /*path_id=*/0,
-        /*smallest_seq=*/20, /*largest_seq=*/21,
-        /*compensated_file_size=*/1000);
-    Add(/*level=*/0, /*file_number=*/3U, /*smallest=*/"100",
-        /*largest=*/"200", /*file_size=*/1000, /*path_id=*/0,
-        /*smallest_seq=*/30, /*largest_seq=*/31,
-        /*compensated_file_size=*/1000);
-    Add(/*level=*/0, /*file_number=*/4U, /*smallest=*/"100",
-        /*largest=*/"200", /*file_size=*/1000, /*path_id=*/0,
-        /*smallest_seq=*/40, /*largest_seq=*/41,
-        /*compensated_file_size=*/1000);
-    const uint64_t l0_size = 4000;
-    const uint64_t lbase_size = l0_size * lbase_size_multiplier;
-    Add(/*level=*/1, /*file_number=*/5U, /*smallest=*/"100",
-        /*largest=*/"200", /*file_size=*/lbase_size, /*path_id=*/0,
-        /*smallest_seq=*/0, /*largest_seq=*/0,
-        /*compensated_file_size=*/lbase_size);
-    UpdateVersionStorageInfo();
-
-    LevelCompactionPicker compaction_picker(ioptions_, &icmp_);
-    std::unique_ptr<Compaction> compaction(compaction_picker.PickCompaction(
-        cf_name_, mutable_cf_options_, mutable_db_options_, vstorage_.get(),
-        &log_buffer_));
-    ASSERT_TRUE(compaction.get() != nullptr);
-    ASSERT_EQ(CompactionReason::kLevelL0FilesNum,
-              compaction->compaction_reason());
-    ASSERT_EQ(4U, compaction->num_input_files(0));
-    if (lbase_size_multiplier >
-        mutable_cf_options_.max_bytes_for_level_multiplier) {
-      ASSERT_EQ(1U, compaction->num_input_levels());
-      ASSERT_EQ(0, compaction->output_level());
-    } else {
-      ASSERT_EQ(2U, compaction->num_input_levels());
-      ASSERT_EQ(1, compaction->output_level());
-    }
-  }
-}

 }  // namespace ROCKSDB_NAMESPACE

@@ -1139,7 +1139,8 @@ Compaction* UniversalCompactionBuilder::PickIncrementalForReduceSizeAmp(
  // from bottom_start_idx and bottom_end_idx, but for now, we use
  // SetupOtherInputs() for simplicity.
  int parent_index = -1;  // Create and use bottom_start_idx?
-  if (!picker_->SetupOtherInputs(cf_name_, vstorage_, &second_last_level_inputs,
+  if (!picker_->SetupOtherInputs(cf_name_, mutable_cf_options_, vstorage_,
+                                 &second_last_level_inputs,
                                 &bottom_level_inputs, &parent_index,
                                 /*base_index=*/-1)) {
    return nullptr;
@@ -1310,8 +1311,9 @@ Compaction* UniversalCompactionBuilder::PickDeleteTriggeredCompaction() {
      int parent_index = -1;

      output_level_inputs.level = output_level;
-      if (!picker_->SetupOtherInputs(cf_name_, vstorage_, &start_level_inputs,
-                                     &output_level_inputs, &parent_index, -1)) {
+      if (!picker_->SetupOtherInputs(cf_name_, mutable_cf_options_, vstorage_,
+                                     &start_level_inputs, &output_level_inputs,
+                                     &parent_index, -1)) {
        return nullptr;
      }
      inputs.push_back(start_level_inputs);
@@ -17,13 +17,14 @@ class UniversalCompactionPicker : public CompactionPicker {
  UniversalCompactionPicker(const ImmutableOptions& ioptions,
                            const InternalKeyComparator* icmp)
      : CompactionPicker(ioptions, icmp) {}
-  Compaction* PickCompaction(const std::string& cf_name,
-                             const MutableCFOptions& mutable_cf_options,
-                             const MutableDBOptions& mutable_db_options,
-                             VersionStorageInfo* vstorage,
-                             LogBuffer* log_buffer) override;
-  int MaxOutputLevel() const override { return NumberLevels() - 1; }
+  virtual Compaction* PickCompaction(const std::string& cf_name,
+                                     const MutableCFOptions& mutable_cf_options,
+                                     const MutableDBOptions& mutable_db_options,
+                                     VersionStorageInfo* vstorage,
+                                     LogBuffer* log_buffer) override;
+  virtual int MaxOutputLevel() const override { return NumberLevels() - 1; }

-  bool NeedsCompaction(const VersionStorageInfo* vstorage) const override;
+  virtual bool NeedsCompaction(
+      const VersionStorageInfo* vstorage) const override;
 };
 }  // namespace ROCKSDB_NAMESPACE
@@ -195,8 +195,8 @@ CompactionJob::ProcessKeyValueCompactionWithCompactionService(

    auto cfd = compaction->column_family_data();
    sub_compact->Current().AddOutput(std::move(meta),
-                                     cfd->internal_comparator(), false, true,
-                                     file.paranoid_hash);
+                                     cfd->internal_comparator(), false, false,
+                                     true, file.paranoid_hash);
  }
  sub_compact->compaction_job_stats = compaction_result.stats;
  sub_compact->Current().SetNumOutputRecords(
@@ -1202,126 +1202,20 @@ TEST_P(TieredCompactionTest, RangeBasedTieredStorageLevel) {
  ASSERT_EQ("0,0,0,0,0,0,1", FilesPerLevel());
  ASSERT_EQ(GetSstSizeHelper(Temperature::kUnknown), 0);
  ASSERT_GT(GetSstSizeHelper(Temperature::kCold), 0);
+
  ASSERT_EQ(
      options.statistics->getTickerCount(COMPACTION_RANGE_DEL_DROP_OBSOLETE),
      1);
-
-  // Tests that we only compact keys up to penultimate level
-  // that are within penultimate level input's internal key range.
-  {
-    MutexLock l(&mutex);
-    hot_start = Key(0);
-    hot_end = Key(100);
-  }
-  const Snapshot* temp_snap = db_->GetSnapshot();
-  // Key(0) and Key(1) here are inserted with higher sequence number
-  // than Key(0) and Key(1) inserted above.
-  // Only Key(0) in last level will be compacted up, not Key(1).
-  ASSERT_OK(Put(Key(0), "value" + std::to_string(0)));
-  ASSERT_OK(Put(Key(1), "value" + std::to_string(100)));
-  ASSERT_OK(db_->CompactRange(cro, nullptr, nullptr));
-  ASSERT_EQ("0,0,0,0,0,1,1", FilesPerLevel());
-  {
-    std::vector<LiveFileMetaData> metas;
-    db_->GetLiveFilesMetaData(&metas);
-    for (const auto& f : metas) {
-      if (f.temperature == Temperature::kUnknown) {
-        // Expect Key(0), Key(0), Key(1)
-        ASSERT_EQ(f.num_entries, 3);
-        ASSERT_EQ(f.smallestkey, Key(0));
-        ASSERT_EQ(f.largestkey, Key(1));
-      } else {
-        ASSERT_EQ(f.temperature, Temperature::kCold);
-        // Key(2)-Key(49) and Key(100).
-        ASSERT_EQ(f.num_entries, 50);
-      }
-    }
-  }
-  db_->ReleaseSnapshot(temp_snap);
 }

 INSTANTIATE_TEST_CASE_P(TieredCompactionTest, TieredCompactionTest,
                        testing::Bool());

-TEST_P(TieredCompactionTest, CheckInternalKeyRange) {
-  // When compacting keys from the last level to penultimate level,
-  // output to penultimate level should be within internal key range
-  // of input files from penultimate level.
-  // Set up:
-  // L5:
-  //  File 1: DeleteRange[1, 3)@4, File 2: [3@5, 100@6]
-  // L6:
-  // File 3: [2@1, 3@2], File 4: [50@3]
-  //
-  // When File 1 and File 3 are being compacted,
-  // Key(3) cannot be compacted up, otherwise it causes
-  // inconsistency where File 3's Key(3) has a lower sequence number
-  // than File 2's Key(3).
-  const int kNumLevels = 7;
-  auto options = CurrentOptions();
-  SetColdTemperature(options);
-  options.level_compaction_dynamic_level_bytes = true;
-  options.num_levels = kNumLevels;
-  options.statistics = CreateDBStatistics();
-  options.max_subcompactions = 10;
-  options.preclude_last_level_data_seconds = 10000;
-  DestroyAndReopen(options);
-  auto cmp = options.comparator;
-
-  std::string hot_start = Key(0);
-  std::string hot_end = Key(0);
-  SyncPoint::GetInstance()->SetCallBack(
-      "CompactionIterator::PrepareOutput.context", [&](void* arg) {
-        auto context = static_cast<PerKeyPlacementContext*>(arg);
-        context->output_to_penultimate_level =
-            cmp->Compare(context->key, hot_start) >= 0 &&
-            cmp->Compare(context->key, hot_end) < 0;
-      });
-  SyncPoint::GetInstance()->EnableProcessing();
-  // File 1
-  ASSERT_OK(Put(Key(2), "val2"));
-  ASSERT_OK(Put(Key(3), "val3"));
-  ASSERT_OK(Flush());
-  MoveFilesToLevel(6);
-  // File 2
-  ASSERT_OK(Put(Key(50), "val50"));
-  ASSERT_OK(Flush());
-  MoveFilesToLevel(6);
-
-  const Snapshot* snapshot = db_->GetSnapshot();
-  hot_end = Key(100);
-  std::string start = Key(1);
-  std::string end = Key(3);
-  ASSERT_OK(
-      db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(), start, end));
-  ASSERT_OK(Flush());
-  MoveFilesToLevel(5);
-  // File 3
-  ASSERT_OK(Put(Key(3), "vall"));
-  ASSERT_OK(Put(Key(100), "val100"));
-  ASSERT_OK(Flush());
-  MoveFilesToLevel(5);
-  // Try to compact keys up
-  CompactRangeOptions cro;
-  cro.bottommost_level_compaction = BottommostLevelCompaction::kForce;
-  start = Key(1);
-  end = Key(2);
-  Slice begin_slice(start);
-  Slice end_slice(end);
-  ASSERT_OK(db_->CompactRange(cro, &begin_slice, &end_slice));
-  // Without internal key range checking, we get the following error:
-  // Corruption: force_consistency_checks(DEBUG): VersionBuilder: L5 has
-  // overlapping ranges: file #18 largest key: '6B6579303030303033' seq:102,
-  // type:1 vs. file #15 smallest key: '6B6579303030303033' seq:104, type:1
-  db_->ReleaseSnapshot(snapshot);
-}
-
 class PrecludeLastLevelTest : public DBTestBase {
 public:
  PrecludeLastLevelTest()
      : DBTestBase("preclude_last_level_test", /*env_do_fsync=*/false) {
    mock_clock_ = std::make_shared<MockSystemClock>(env_->GetSystemClock());
-    mock_clock_->SetCurrentTime(kMockStartTime);
    mock_env_ = std::make_unique<CompositeEnvWrapper>(env_, mock_clock_);
  }

@@ -1329,19 +1223,15 @@ class PrecludeLastLevelTest : public DBTestBase {
  std::unique_ptr<Env> mock_env_;
  std::shared_ptr<MockSystemClock> mock_clock_;

-  // Sufficient starting time that preserve time doesn't under-flow into
-  // pre-history
-  static constexpr uint32_t kMockStartTime = 10000000;
-
  void SetUp() override {
    mock_clock_->InstallTimedWaitFixCallback();
    SyncPoint::GetInstance()->SetCallBack(
        "DBImpl::StartPeriodicTaskScheduler:Init", [&](void* arg) {
          auto periodic_task_scheduler_ptr =
-              static_cast<PeriodicTaskScheduler*>(arg);
+              reinterpret_cast<PeriodicTaskScheduler*>(arg);
          periodic_task_scheduler_ptr->TEST_OverrideTimer(mock_clock_.get());
        });
-    mock_clock_->SetCurrentTime(kMockStartTime);
+    mock_clock_->SetCurrentTime(0);
  }
 };

@@ -1359,6 +1249,11 @@ TEST_F(PrecludeLastLevelTest, MigrationFromPreserveTimeManualCompaction) {
  options.num_levels = kNumLevels;
  DestroyAndReopen(options);

+  // pass some time first, otherwise the first a few keys write time are going
+  // to be zero, and internally zero has special meaning: kUnknownSeqnoTime
+  dbfull()->TEST_WaitForPeriodicTaskRun(
+      [&] { mock_clock_->MockSleepForSeconds(static_cast<int>(kKeyPerSec)); });
+
  int sst_num = 0;
  // Write files that are overlap and enough to trigger compaction
  for (; sst_num < kNumTrigger; sst_num++) {
@@ -1416,6 +1311,11 @@ TEST_F(PrecludeLastLevelTest, MigrationFromPreserveTimeAutoCompaction) {
  options.num_levels = kNumLevels;
  DestroyAndReopen(options);

+  // pass some time first, otherwise the first a few keys write time are going
+  // to be zero, and internally zero has special meaning: kUnknownSeqnoTime
+  dbfull()->TEST_WaitForPeriodicTaskRun(
+      [&] { mock_clock_->MockSleepForSeconds(static_cast<int>(kKeyPerSec)); });
+
  int sst_num = 0;
  // Write files that are overlap and enough to trigger compaction
  for (; sst_num < kNumTrigger; sst_num++) {
@@ -1487,6 +1387,11 @@ TEST_F(PrecludeLastLevelTest, MigrationFromPreserveTimePartial) {
  options.num_levels = kNumLevels;
  DestroyAndReopen(options);

+  // pass some time first, otherwise the first a few keys write time are going
+  // to be zero, and internally zero has special meaning: kUnknownSeqnoTime
+  dbfull()->TEST_WaitForPeriodicTaskRun(
+      [&] { mock_clock_->MockSleepForSeconds(static_cast<int>(kKeyPerSec)); });
+
  int sst_num = 0;
  // Write files that are overlap and enough to trigger compaction
  for (; sst_num < kNumTrigger; sst_num++) {
@@ -1575,8 +1480,9 @@ TEST_F(PrecludeLastLevelTest, SmallPrecludeTime) {
  ASSERT_EQ(tables_props.size(), 1);
  ASSERT_FALSE(tables_props.begin()->second->seqno_to_time_mapping.empty());
  SeqnoToTimeMapping tp_mapping;
-  ASSERT_OK(tp_mapping.DecodeFrom(
-      tables_props.begin()->second->seqno_to_time_mapping));
+  ASSERT_OK(
+      tp_mapping.Add(tables_props.begin()->second->seqno_to_time_mapping));
+  ASSERT_OK(tp_mapping.Sort());
  ASSERT_FALSE(tp_mapping.Empty());
  auto seqs = tp_mapping.TEST_GetInternalMapping();
  ASSERT_FALSE(seqs.empty());
@@ -1608,6 +1514,11 @@ TEST_F(PrecludeLastLevelTest, LastLevelOnlyCompactionPartial) {
  options.num_levels = kNumLevels;
  DestroyAndReopen(options);

+  // pass some time first, otherwise the first a few keys write time are going
+  // to be zero, and internally zero has special meaning: kUnknownSeqnoTime
+  dbfull()->TEST_WaitForPeriodicTaskRun(
+      [&] { mock_clock_->MockSleepForSeconds(static_cast<int>(kKeyPerSec)); });
+
  int sst_num = 0;
  // Write files that are overlap and enough to trigger compaction
  for (; sst_num < kNumTrigger; sst_num++) {
@@ -1681,6 +1592,11 @@ TEST_P(PrecludeLastLevelTestWithParms, LastLevelOnlyCompactionNoPreclude) {
  options.num_levels = kNumLevels;
  DestroyAndReopen(options);

+  // pass some time first, otherwise the first a few keys write time are going
+  // to be zero, and internally zero has special meaning: kUnknownSeqnoTime
+  dbfull()->TEST_WaitForPeriodicTaskRun(
+      [&] { mock_clock_->MockSleepForSeconds(static_cast<int>(kKeyPerSec)); });
+
  Random rnd(301);
  int sst_num = 0;
  // Write files that are overlap and enough to trigger compaction
@@ -1823,6 +1739,7 @@ TEST_P(PrecludeLastLevelTestWithParms, PeriodicCompactionToPenultimateLevel) {
  options.env = mock_env_.get();
  options.level0_file_num_compaction_trigger = kNumTrigger;
  options.num_levels = kNumLevels;
+  options.ignore_max_compaction_bytes_for_input = false;
  options.periodic_compaction_seconds = 10000;
  DestroyAndReopen(options);

@@ -1989,6 +1906,11 @@ TEST_F(PrecludeLastLevelTest, PartialPenultimateLevelCompaction) {
  options.num_levels = kNumLevels;
  DestroyAndReopen(options);

+  // pass some time first, otherwise the first a few keys write time are going
+  // to be zero, and internally zero has special meaning: kUnknownSeqnoTime
+  dbfull()->TEST_WaitForPeriodicTaskRun(
+      [&] { mock_clock_->MockSleepForSeconds(static_cast<int>(10)); });
+
  Random rnd(301);

  for (int i = 0; i < 300; i++) {
@@ -2095,6 +2017,12 @@ TEST_F(PrecludeLastLevelTest, RangeDelsCauseFileEndpointsToOverlap) {
  options.target_file_size_base = kFileBytes;
  DestroyAndReopen(options);

+  // pass some time first, otherwise the first a few keys write time are going
+  // to be zero, and internally zero has special meaning: kUnknownSeqnoTime
+  dbfull()->TEST_WaitForPeriodicTaskRun([&] {
+    mock_clock_->MockSleepForSeconds(static_cast<int>(kSecondsPerKey));
+  });
+
  // Flush an L0 file with the following contents (new to old):
  //
  // Range deletions [4, 6) [7, 8) [9, 11)
@@ -2211,6 +2139,7 @@ TEST_F(PrecludeLastLevelTest, RangeDelsCauseFileEndpointsToOverlap) {
  Close();
 }

+
 }  // namespace ROCKSDB_NAMESPACE

 int main(int argc, char** argv) {
@@ -165,7 +165,6 @@ void DoRandomIteraratorTest(DB* db, std::vector<std::string> source_strings,
    AssertItersEqual(iter.get(), result_iter.get());
    is_valid = iter->Valid();
  }
-  ASSERT_OK(iter->status());
 }

 class DoubleComparator : public Comparator {
@@ -34,7 +34,7 @@ Status DeleteFilesInRanges(DB* db, ColumnFamilyHandle* column_family,
 Status VerifySstFileChecksum(const Options& options,
                             const EnvOptions& env_options,
                             const std::string& file_path) {
-  // TODO: plumb Env::IOActivity, Env::IOPriority
+  // TODO: plumb Env::IOActivity
  const ReadOptions read_options;
  return VerifySstFileChecksum(options, env_options, read_options, file_path);
 }
@@ -53,9 +53,10 @@ class ErrorFS : public FileSystemWrapper {
        num_writable_file_errors_(0) {}
  const char* Name() const override { return "ErrorEnv"; }

-  IOStatus NewWritableFile(const std::string& fname, const FileOptions& opts,
-                           std::unique_ptr<FSWritableFile>* result,
-                           IODebugContext* dbg) override {
+  virtual IOStatus NewWritableFile(const std::string& fname,
+                                   const FileOptions& opts,
+                                   std::unique_ptr<FSWritableFile>* result,
+                                   IODebugContext* dbg) override {
    result->reset();
    if (writable_file_error_) {
      ++num_writable_file_errors_;
@@ -822,6 +823,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksOnFlush) {
  Options options;
  options.level_compaction_dynamic_level_bytes = false;
  options.env = env_.get();
+  options.check_flush_compaction_key_order = false;
  options.paranoid_file_checks = true;
  options.create_if_missing = true;
  Status s;
@@ -852,6 +854,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksOnCompact) {
  options.env = env_.get();
  options.paranoid_file_checks = true;
  options.create_if_missing = true;
+  options.check_flush_compaction_key_order = false;
  Status s;
  for (const auto& mode : corruption_modes) {
    delete db_;
@@ -883,6 +886,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRangeFirst) {
  Options options;
  options.level_compaction_dynamic_level_bytes = false;
  options.env = env_.get();
+  options.check_flush_compaction_key_order = false;
  options.paranoid_file_checks = true;
  options.create_if_missing = true;
  for (bool do_flush : {true, false}) {
@@ -919,6 +923,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRange) {
  Options options;
  options.level_compaction_dynamic_level_bytes = false;
  options.env = env_.get();
+  options.check_flush_compaction_key_order = false;
  options.paranoid_file_checks = true;
  options.create_if_missing = true;
  for (bool do_flush : {true, false}) {
@@ -958,6 +963,7 @@ TEST_F(CorruptionTest, ParanoidFileChecksWithDeleteRangeLast) {
  Options options;
  options.level_compaction_dynamic_level_bytes = false;
  options.env = env_.get();
+  options.check_flush_compaction_key_order = false;
  options.paranoid_file_checks = true;
  options.create_if_missing = true;
  for (bool do_flush : {true, false}) {
@@ -1026,6 +1032,7 @@ TEST_F(CorruptionTest, CompactionKeyOrderCheck) {
  options.env = env_.get();
  options.paranoid_file_checks = false;
  options.create_if_missing = true;
+  options.check_flush_compaction_key_order = false;
  delete db_;
  db_ = nullptr;
  ASSERT_OK(DestroyDB(dbname_, options));
@@ -1040,6 +1047,7 @@ TEST_F(CorruptionTest, CompactionKeyOrderCheck) {
  ASSERT_OK(dbi->TEST_FlushMemTable());

  mock->SetCorruptionMode(mock::MockTableFactory::kCorruptNone);
+  ASSERT_OK(db_->SetOptions({{"check_flush_compaction_key_order", "true"}}));
  CompactRangeOptions cro;
  cro.bottommost_level_compaction = BottommostLevelCompaction::kForce;
  ASSERT_NOK(
@@ -1052,6 +1060,7 @@ TEST_F(CorruptionTest, FlushKeyOrderCheck) {
  options.env = env_.get();
  options.paranoid_file_checks = false;
  options.create_if_missing = true;
+  ASSERT_OK(db_->SetOptions({{"check_flush_compaction_key_order", "true"}}));

  ASSERT_OK(db_->Put(WriteOptions(), "foo1", "v1"));
  ASSERT_OK(db_->Put(WriteOptions(), "foo2", "v1"));
@@ -1076,6 +1085,28 @@ TEST_F(CorruptionTest, FlushKeyOrderCheck) {
  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
 }

+TEST_F(CorruptionTest, DisableKeyOrderCheck) {
+  ASSERT_OK(db_->SetOptions({{"check_flush_compaction_key_order", "false"}}));
+  DBImpl* dbi = static_cast_with_check<DBImpl>(db_);
+
+  SyncPoint::GetInstance()->SetCallBack(
+      "OutputValidator::Add:order_check",
+      [&](void* /*arg*/) { ASSERT_TRUE(false); });
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
+  ASSERT_OK(db_->Put(WriteOptions(), "foo1", "v1"));
+  ASSERT_OK(db_->Put(WriteOptions(), "foo3", "v1"));
+  ASSERT_OK(dbi->TEST_FlushMemTable());
+  ASSERT_OK(db_->Put(WriteOptions(), "foo2", "v1"));
+  ASSERT_OK(db_->Put(WriteOptions(), "foo4", "v1"));
+  ASSERT_OK(dbi->TEST_FlushMemTable());
+  CompactRangeOptions cro;
+  cro.bottommost_level_compaction = BottommostLevelCompaction::kForce;
+  ASSERT_OK(
+      dbi->CompactRange(cro, dbi->DefaultColumnFamily(), nullptr, nullptr));
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
+  ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
+}
+
 TEST_F(CorruptionTest, VerifyWholeTableChecksum) {
  CloseDb();
  Options options;
@@ -1102,7 +1133,7 @@ TEST_F(CorruptionTest, VerifyWholeTableChecksum) {
  int count{0};
  SyncPoint::GetInstance()->SetCallBack(
      "DBImpl::VerifyFullFileChecksum:mismatch", [&](void* arg) {
-        auto* s = static_cast<Status*>(arg);
+        auto* s = reinterpret_cast<Status*>(arg);
        ASSERT_NE(s, nullptr);
        ++count;
        ASSERT_NOK(*s);
@@ -1247,7 +1278,7 @@ TEST_P(CrashDuringRecoveryWithCorruptionTest, CrashDuringRecovery) {
    SyncPoint::GetInstance()->ClearAllCallBacks();
    SyncPoint::GetInstance()->SetCallBack(
        "DBImpl::GetLogSizeAndMaybeTruncate:0", [&](void* arg) {
-          auto* tmp_s = static_cast<Status*>(arg);
+          auto* tmp_s = reinterpret_cast<Status*>(arg);
          assert(tmp_s);
          *tmp_s = Status::IOError("Injected");
        });
@@ -1429,7 +1460,7 @@ TEST_P(CrashDuringRecoveryWithCorruptionTest, TxnDbCrashDuringRecovery) {
    SyncPoint::GetInstance()->ClearAllCallBacks();
    SyncPoint::GetInstance()->SetCallBack(
        "DBImpl::Open::BeforeSyncWAL", [&](void* arg) {
-          auto* tmp_s = static_cast<Status*>(arg);
+          auto* tmp_s = reinterpret_cast<Status*>(arg);
          assert(tmp_s);
          *tmp_s = Status::IOError("Injected");
        });
@@ -1597,7 +1628,7 @@ TEST_P(CrashDuringRecoveryWithCorruptionTest, CrashDuringRecoveryWithFlush) {
    SyncPoint::GetInstance()->ClearAllCallBacks();
    SyncPoint::GetInstance()->SetCallBack(
        "DBImpl::GetLogSizeAndMaybeTruncate:0", [&](void* arg) {
-          auto* tmp_s = static_cast<Status*>(arg);
+          auto* tmp_s = reinterpret_cast<Status*>(arg);
          assert(tmp_s);
          *tmp_s = Status::IOError("Injected");
        });
@@ -131,7 +131,7 @@ TEST_F(CuckooTableDBTest, Flush) {
  ASSERT_OK(dbfull()->TEST_FlushMemTable());

  TablePropertiesCollection ptc;
-  ASSERT_OK(static_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
+  ASSERT_OK(reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
  VerifySstUniqueIds(ptc);
  ASSERT_EQ(1U, ptc.size());
  ASSERT_EQ(3U, ptc.begin()->second->num_entries);
@@ -148,7 +148,7 @@ TEST_F(CuckooTableDBTest, Flush) {
  ASSERT_OK(Put("key6", "v6"));
  ASSERT_OK(dbfull()->TEST_FlushMemTable());

-  ASSERT_OK(static_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
+  ASSERT_OK(reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
  VerifySstUniqueIds(ptc);
  ASSERT_EQ(2U, ptc.size());
  auto row = ptc.begin();
@@ -166,7 +166,7 @@ TEST_F(CuckooTableDBTest, Flush) {
  ASSERT_OK(Delete("key5"));
  ASSERT_OK(Delete("key4"));
  ASSERT_OK(dbfull()->TEST_FlushMemTable());
-  ASSERT_OK(static_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
+  ASSERT_OK(reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
  VerifySstUniqueIds(ptc);
  ASSERT_EQ(3U, ptc.size());
  row = ptc.begin();
@@ -191,7 +191,7 @@ TEST_F(CuckooTableDBTest, FlushWithDuplicateKeys) {
  ASSERT_OK(dbfull()->TEST_FlushMemTable());

  TablePropertiesCollection ptc;
-  ASSERT_OK(static_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
+  ASSERT_OK(reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc));
  VerifySstUniqueIds(ptc);
  ASSERT_EQ(1U, ptc.size());
  ASSERT_EQ(2U, ptc.begin()->second->num_entries);
@@ -25,60 +25,10 @@
 #include "util/random.h"
 #include "utilities/counted_fs.h"
 #include "utilities/fault_injection_env.h"
-#include "utilities/fault_injection_fs.h"
 #include "utilities/merge_operators.h"
 #include "utilities/merge_operators/string_append/stringappend.h"

 namespace ROCKSDB_NAMESPACE {
-namespace {
-class MyFlushBlockPolicy : public FlushBlockPolicy {
- public:
-  explicit MyFlushBlockPolicy(const int num_keys_in_block,
-                              const BlockBuilder& data_block_builder)
-      : num_keys_in_block_(num_keys_in_block),
-        num_keys_(0),
-        data_block_builder_(data_block_builder) {}
-
-  bool Update(const Slice& /*key*/, const Slice& /*value*/) override {
-    if (data_block_builder_.empty()) {
-      // First key in this block
-      num_keys_ = 1;
-      return false;
-    }
-    // Flush every 10 keys
-    if (num_keys_ == num_keys_in_block_) {
-      num_keys_ = 1;
-      return true;
-    }
-    num_keys_++;
-    return false;
-  }
-
- private:
-  const int num_keys_in_block_;
-  int num_keys_;
-  const BlockBuilder& data_block_builder_;
-};
-
-class MyFlushBlockPolicyFactory : public FlushBlockPolicyFactory {
- public:
-  explicit MyFlushBlockPolicyFactory(const int num_keys_in_block)
-      : num_keys_in_block_(num_keys_in_block) {}
-
-  virtual const char* Name() const override {
-    return "MyFlushBlockPolicyFactory";
-  }
-
-  virtual FlushBlockPolicy* NewFlushBlockPolicy(
-      const BlockBasedTableOptions& /*table_options*/,
-      const BlockBuilder& data_block_builder) const override {
-    return new MyFlushBlockPolicy(num_keys_in_block_, data_block_builder);
-  }
-
- private:
-  const int num_keys_in_block_;
-};
-}  // namespace

 static bool enable_io_uring = true;
 extern "C" bool RocksDbIOUringEnable() { return enable_io_uring; }
@@ -188,7 +138,6 @@ TEST_F(DBBasicTest, ReadOnlyDB) {
      ASSERT_OK(iter->status());
      ++count;
    }
-    ASSERT_OK(iter->status());
    // Always expect two keys: "foo" and "bar"
    ASSERT_EQ(count, 2);
  };
@@ -1903,214 +1852,6 @@ TEST_P(DBMultiGetTestWithParam, MultiGetBatchedMultiLevel) {
  }
 }

-TEST_P(DBMultiGetTestWithParam, MultiGetDuplicatesEmptyLevel) {
-#ifndef USE_COROUTINES
-  if (std::get<1>(GetParam())) {
-    ROCKSDB_GTEST_BYPASS("This test requires coroutine support");
-    return;
-  }
-#endif  // USE_COROUTINES
-  // Skip for unbatched MultiGet
-  if (!std::get<0>(GetParam())) {
-    ROCKSDB_GTEST_BYPASS("This test is only for batched MultiGet");
-    return;
-  }
-  std::shared_ptr<FaultInjectionTestFS> fault_fs(
-      new FaultInjectionTestFS(env_->GetFileSystem()));
-  std::unique_ptr<Env> env(new CompositeEnvWrapper(env_, fault_fs));
-  Options options = CurrentOptions();
-  options.env = env.get();
-  options.disable_auto_compactions = true;
-  options.merge_operator = MergeOperators::CreateStringAppendOperator();
-
-  LRUCacheOptions cache_opts;
-  cache_opts.capacity = 1 << 20;
-
-  BlockBasedTableOptions table_opts;
-  table_opts.metadata_cache_options.top_level_index_pinning = PinningTier::kAll;
-  table_opts.metadata_cache_options.partition_pinning = PinningTier::kNone;
-  table_opts.index_type =
-      BlockBasedTableOptions::IndexType::kTwoLevelIndexSearch;
-  table_opts.cache_index_and_filter_blocks = true;
-  table_opts.block_cache = cache_opts.MakeSharedCache();
-  table_opts.flush_block_policy_factory.reset(new MyFlushBlockPolicyFactory(1));
-
-  options.table_factory.reset(new BlockBasedTableFactory(table_opts));
-  Reopen(options);
-  int key;
-
-  // Setup the LSM so that the following search bounds are generated for
-  // key 9 for each level -
-  // Level 1 - lb = 0, rb = max
-  // Level 2 - lb = 0, rb = 0
-  // Level 3 - lb = 0, rb = -1
-  // Level 4 - lb = 0, rb = 0
-
-  key = 9;
-  ASSERT_OK(Put("key_" + std::to_string(key), "val_l2_" + std::to_string(key)));
-  ASSERT_OK(Flush());
-  MoveFilesToLevel(4);
-
-  key = 5;
-  ASSERT_OK(Put("key_" + std::to_string(key), "val_l2_" + std::to_string(key)));
-  key = 9;
-  ASSERT_OK(
-      Merge("key_" + std::to_string(key), "val_l2_" + std::to_string(key)));
-  const Snapshot* snap = dbfull()->GetSnapshot();
-  ASSERT_OK(
-      Merge("key_" + std::to_string(key), "val_l2_ext_" + std::to_string(key)));
-  ASSERT_OK(Flush());
-  // Leave level 3 empty
-  MoveFilesToLevel(2);
-
-  key = 2;
-  ASSERT_OK(Put("key_" + std::to_string(key), "val_l2_" + std::to_string(key)));
-  key = 6;
-  ASSERT_OK(
-      Merge("key_" + std::to_string(key), "val_l2_" + std::to_string(key)));
-  ASSERT_OK(Flush());
-  MoveFilesToLevel(1);
-
-  std::vector<std::string> keys;
-  std::vector<std::string> values;
-
-  keys.push_back("key_" + std::to_string(9));
-  keys.push_back("key_" + std::to_string(9));
-
-  int num_reads = 0;
-  SyncPoint::GetInstance()->SetCallBack(
-      "FaultInjectionTestFS::RandomRead", [&](void*) {
-        ++num_reads;
-        // Fail on the 2nd read. First read is index partition,
-        // second read is data block in level 1
-        if (num_reads == 2) {
-          fault_fs->SetFilesystemActive(false);
-        } else {
-          fault_fs->SetFilesystemActive(true);
-        }
-      });
-  SyncPoint::GetInstance()->EnableProcessing();
-  size_t capacity = table_opts.block_cache->GetCapacity();
-  table_opts.block_cache->SetCapacity(0);
-  table_opts.block_cache->SetCapacity(capacity);
-
-  values = MultiGet(keys, nullptr, std::get<1>(GetParam()));
-  ASSERT_EQ(values.size(), 2);
-
-  SyncPoint::GetInstance()->DisableProcessing();
-  dbfull()->ReleaseSnapshot(snap);
-  Destroy(options);
-}
-
-TEST_P(DBMultiGetTestWithParam, MultiGetDuplicatesNonEmptyLevel) {
-#ifndef USE_COROUTINES
-  if (std::get<1>(GetParam())) {
-    ROCKSDB_GTEST_BYPASS("This test requires coroutine support");
-    return;
-  }
-#endif  // USE_COROUTINES
-  // Skip for unbatched MultiGet
-  if (!std::get<0>(GetParam())) {
-    ROCKSDB_GTEST_BYPASS("This test is only for batched MultiGet");
-    return;
-  }
-  std::shared_ptr<FaultInjectionTestFS> fault_fs(
-      new FaultInjectionTestFS(env_->GetFileSystem()));
-  std::unique_ptr<Env> env(new CompositeEnvWrapper(env_, fault_fs));
-  Options options = CurrentOptions();
-  options.env = env.get();
-  options.disable_auto_compactions = true;
-  options.merge_operator = MergeOperators::CreateStringAppendOperator();
-
-  LRUCacheOptions cache_opts;
-  cache_opts.capacity = 1 << 20;
-
-  BlockBasedTableOptions table_opts;
-  table_opts.metadata_cache_options.top_level_index_pinning = PinningTier::kAll;
-  table_opts.metadata_cache_options.partition_pinning = PinningTier::kNone;
-  table_opts.index_type =
-      BlockBasedTableOptions::IndexType::kTwoLevelIndexSearch;
-  table_opts.cache_index_and_filter_blocks = true;
-  table_opts.block_cache = cache_opts.MakeSharedCache();
-  table_opts.flush_block_policy_factory.reset(new MyFlushBlockPolicyFactory(1));
-
-  options.table_factory.reset(new BlockBasedTableFactory(table_opts));
-  Reopen(options);
-  int key;
-
-  // Setup the LSM so that the following search bounds are generated for
-  // key 9 for each level -
-  // Level 1 - lb = 0, rb = max
-  // Level 2 - lb = 0, rb = 0
-  // Level 3 - lb = 0, rb = 1
-  // Level 4 - N/A
-
-  key = 8;
-  ASSERT_OK(Put("key_" + std::to_string(key), "val_l2_" + std::to_string(key)));
-  ASSERT_OK(Flush());
-  MoveFilesToLevel(4);
-
-  key = 7;
-  ASSERT_OK(Put("key_" + std::to_string(key), "val_l2_" + std::to_string(key)));
-  ASSERT_OK(Flush());
-
-  key = 9;
-  ASSERT_OK(Put("key_" + std::to_string(key), "val_l2_" + std::to_string(key)));
-  ASSERT_OK(Flush());
-  MoveFilesToLevel(3);
-
-  key = 5;
-  ASSERT_OK(Put("key_" + std::to_string(key), "val_l2_" + std::to_string(key)));
-  key = 9;
-  ASSERT_OK(
-      Merge("key_" + std::to_string(key), "merge1_l2_" + std::to_string(key)));
-  const Snapshot* snap = dbfull()->GetSnapshot();
-  ASSERT_OK(
-      Merge("key_" + std::to_string(key), "merge2_l2_" + std::to_string(key)));
-  ASSERT_OK(Flush());
-  MoveFilesToLevel(2);
-
-  key = 2;
-  ASSERT_OK(Put("key_" + std::to_string(key), "val_l2_" + std::to_string(key)));
-  key = 6;
-  ASSERT_OK(
-      Merge("key_" + std::to_string(key), "val_l2_" + std::to_string(key)));
-  ASSERT_OK(Flush());
-  MoveFilesToLevel(1);
-
-  std::vector<std::string> keys;
-  std::vector<std::string> values;
-
-  keys.push_back("key_" + std::to_string(9));
-  keys.push_back("key_" + std::to_string(9));
-
-  int num_reads = 0;
-  SyncPoint::GetInstance()->SetCallBack(
-      "FaultInjectionTestFS::RandomRead", [&](void*) {
-        ++num_reads;
-        // Fail on the 2nd read. First read is index partition,
-        // second read is data block in level 1
-        if (num_reads == 2) {
-          fault_fs->SetFilesystemActive(false);
-        } else {
-          fault_fs->SetFilesystemActive(true);
-        }
-      });
-  SyncPoint::GetInstance()->EnableProcessing();
-  size_t capacity = table_opts.block_cache->GetCapacity();
-  table_opts.block_cache->SetCapacity(0);
-  table_opts.block_cache->SetCapacity(capacity);
-
-  values = MultiGet(keys, nullptr, std::get<1>(GetParam()));
-  ASSERT_EQ(values.size(), 2);
-  ASSERT_EQ(values[0], "Corruption: Not active");
-  ASSERT_EQ(values[1], "val_l2_9,merge1_l2_9,merge2_l2_9");
-
-  SyncPoint::GetInstance()->DisableProcessing();
-  dbfull()->ReleaseSnapshot(snap);
-  Destroy(options);
-}
-
 TEST_P(DBMultiGetTestWithParam, MultiGetBatchedMultiLevelMerge) {
 #ifndef USE_COROUTINES
  if (std::get<1>(GetParam())) {
@@ -3342,7 +3083,8 @@ TEST_F(DBBasicTest, BestEffortsRecoveryWithVersionBuildingFailure) {
  SyncPoint::GetInstance()->SetCallBack(
      "VersionBuilder::CheckConsistencyBeforeReturn", [&](void* arg) {
        ASSERT_NE(nullptr, arg);
-        *(static_cast<Status*>(arg)) = Status::Corruption("Inject corruption");
+        *(reinterpret_cast<Status*>(arg)) =
+            Status::Corruption("Inject corruption");
      });
  SyncPoint::GetInstance()->EnableProcessing();

@@ -3383,8 +3125,7 @@ TEST_F(DBBasicTest, LastSstFileNotInManifest) {
  // Manually add a sst file.
  constexpr uint64_t kSstFileNumber = 100;
  const std::string kSstFile = MakeTableFileName(dbname_, kSstFileNumber);
-  ASSERT_OK(WriteStringToFile(env_,
-                              /* data = */ "bad sst file content",
+  ASSERT_OK(WriteStringToFile(env_, /* data = */ "bad sst file content",
                              /* fname = */ kSstFile,
                              /* should_sync = */ true));
  ASSERT_OK(env_->FileExists(kSstFile));
@@ -3773,7 +3514,7 @@ class DBBasicTestMultiGet : public DBTestBase {
      table_options.pin_l0_filter_and_index_blocks_in_cache = true;
    }
    table_options.flush_block_policy_factory.reset(
-        new MyFlushBlockPolicyFactory(10));
+        new MyFlushBlockPolicyFactory());
    options.table_factory.reset(NewBlockBasedTableFactory(table_options));
    if (!compression_enabled_) {
      options.compression = kNoCompression;
@@ -3861,6 +3602,46 @@ class DBBasicTestMultiGet : public DBTestBase {
  static void TearDownTestCase() {}

 protected:
+  class MyFlushBlockPolicyFactory : public FlushBlockPolicyFactory {
+   public:
+    MyFlushBlockPolicyFactory() {}
+
+    virtual const char* Name() const override {
+      return "MyFlushBlockPolicyFactory";
+    }
+
+    virtual FlushBlockPolicy* NewFlushBlockPolicy(
+        const BlockBasedTableOptions& /*table_options*/,
+        const BlockBuilder& data_block_builder) const override {
+      return new MyFlushBlockPolicy(data_block_builder);
+    }
+  };
+
+  class MyFlushBlockPolicy : public FlushBlockPolicy {
+   public:
+    explicit MyFlushBlockPolicy(const BlockBuilder& data_block_builder)
+        : num_keys_(0), data_block_builder_(data_block_builder) {}
+
+    bool Update(const Slice& /*key*/, const Slice& /*value*/) override {
+      if (data_block_builder_.empty()) {
+        // First key in this block
+        num_keys_ = 1;
+        return false;
+      }
+      // Flush every 10 keys
+      if (num_keys_ == 10) {
+        num_keys_ = 1;
+        return true;
+      }
+      num_keys_++;
+      return false;
+    }
+
+   private:
+    int num_keys_;
+    const BlockBuilder& data_block_builder_;
+  };
+
  class MyBlockCache : public CacheWrapper {
   public:
    explicit MyBlockCache(std::shared_ptr<Cache> target)
@@ -4277,7 +4058,7 @@ class DeadlineRandomAccessFile : public FSRandomAccessFileOwnerWrapper {
                     const IOOptions& options, IODebugContext* dbg) override;

  IOStatus ReadAsync(FSReadRequest& req, const IOOptions& opts,
-                     std::function<void(FSReadRequest&, void*)> cb,
+                     std::function<void(const FSReadRequest&, void*)> cb,
                     void* cb_arg, void** io_handle, IOHandleDeleter* del_fn,
                     IODebugContext* dbg) override;

@@ -4423,7 +4204,7 @@ IOStatus DeadlineRandomAccessFile::Read(uint64_t offset, size_t len,

 IOStatus DeadlineRandomAccessFile::ReadAsync(
    FSReadRequest& req, const IOOptions& opts,
-    std::function<void(FSReadRequest&, void*)> cb, void* cb_arg,
+    std::function<void(const FSReadRequest&, void*)> cb, void* cb_arg,
    void** io_handle, IOHandleDeleter* del_fn, IODebugContext* dbg) {
  const std::chrono::microseconds deadline = fs_.GetDeadline();
  const std::chrono::microseconds io_timeout = fs_.GetIOTimeout();
@@ -4504,35 +4285,68 @@ TEST_P(DBBasicTestMultiGetDeadline, MultiGetDeadlineExceeded) {
  SetTimeElapseOnlySleepOnReopen(&options);
  ReopenWithColumnFamilies(GetCFNames(), options);

-  // Test batched MultiGet with an IO delay in the first data block read.
-  // Both keys in the first CF should succeed as they're in the same data
-  // block and would form one batch, and we check for deadline between
-  // batches.
+  // Test the non-batched version of MultiGet with multiple column
+  // families
  std::vector<std::string> key_str;
  size_t i;
-  for (i = 0; i < 10; ++i) {
+  for (i = 0; i < 5; ++i) {
    key_str.emplace_back(Key(static_cast<int>(i)));
  }
  std::vector<ColumnFamilyHandle*> cfs(key_str.size());
+  ;
  std::vector<Slice> keys(key_str.size());
-  std::vector<PinnableSlice> pin_values(keys.size());
-
+  std::vector<std::string> values(key_str.size());
  for (i = 0; i < key_str.size(); ++i) {
-    // 2 keys per CF
-    cfs[i] = handles_[i / 2];
+    cfs[i] = handles_[i];
    keys[i] = Slice(key_str[i].data(), key_str[i].size());
  }
+
  ReadOptions ro;
  ro.deadline = std::chrono::microseconds{env->NowMicros() + 10000};
  ro.async_io = GetParam();
  // Delay the first IO
  fs->SetDelayTrigger(ro.deadline, ro.io_timeout, 0);

-  std::vector<Status> statuses(key_str.size());
-  dbfull()->MultiGet(ro, keys.size(), cfs.data(), keys.data(),
-                     pin_values.data(), statuses.data());
+  std::vector<Status> statuses = dbfull()->MultiGet(ro, cfs, keys, &values);
  // The first key is successful because we check after the lookup, but
  // subsequent keys fail due to deadline exceeded
+  CheckStatus(statuses, 1);
+
+  // Clear the cache
+  cache->SetCapacity(0);
+  cache->SetCapacity(1048576);
+  // Test non-batched Multiget with multiple column families and
+  // introducing an IO delay in one of the middle CFs
+  key_str.clear();
+  for (i = 0; i < 10; ++i) {
+    key_str.emplace_back(Key(static_cast<int>(i)));
+  }
+  cfs.resize(key_str.size());
+  keys.resize(key_str.size());
+  values.resize(key_str.size());
+  for (i = 0; i < key_str.size(); ++i) {
+    // 2 keys per CF
+    cfs[i] = handles_[i / 2];
+    keys[i] = Slice(key_str[i].data(), key_str[i].size());
+  }
+  ro.deadline = std::chrono::microseconds{env->NowMicros() + 10000};
+  fs->SetDelayTrigger(ro.deadline, ro.io_timeout, 1);
+  statuses = dbfull()->MultiGet(ro, cfs, keys, &values);
+  CheckStatus(statuses, 3);
+
+  // Test batched MultiGet with an IO delay in the first data block read.
+  // Both keys in the first CF should succeed as they're in the same data
+  // block and would form one batch, and we check for deadline between
+  // batches.
+  std::vector<PinnableSlice> pin_values(keys.size());
+  cache->SetCapacity(0);
+  cache->SetCapacity(1048576);
+  statuses.clear();
+  statuses.resize(keys.size());
+  ro.deadline = std::chrono::microseconds{env->NowMicros() + 10000};
+  fs->SetDelayTrigger(ro.deadline, ro.io_timeout, 0);
+  dbfull()->MultiGet(ro, keys.size(), cfs.data(), keys.data(),
+                     pin_values.data(), statuses.data());
  CheckStatus(statuses, 2);

  // Similar to the previous one, but an IO delay in the third CF data block
@@ -4600,8 +4414,6 @@ TEST_F(DBBasicTest, ManifestWriteFailure) {
  options.create_if_missing = true;
  options.disable_auto_compactions = true;
  options.env = env_;
-  options.enable_blob_files = true;
-  options.blob_file_size = 0;
  DestroyAndReopen(options);
  ASSERT_OK(Put("foo", "bar"));
  ASSERT_OK(Flush());
@@ -4610,7 +4422,7 @@ TEST_F(DBBasicTest, ManifestWriteFailure) {
  SyncPoint::GetInstance()->SetCallBack(
      "VersionSet::ProcessManifestWrites:AfterSyncManifest", [&](void* arg) {
        ASSERT_NE(nullptr, arg);
-        auto* s = static_cast<Status*>(arg);
+        auto* s = reinterpret_cast<Status*>(arg);
        ASSERT_OK(*s);
        // Manually overwrite return status
        *s = Status::IOError();
@@ -4622,11 +4434,6 @@ TEST_F(DBBasicTest, ManifestWriteFailure) {
  SyncPoint::GetInstance()->ClearAllCallBacks();
  SyncPoint::GetInstance()->EnableProcessing();
  Reopen(options);
-  // The IO error was a mocked one from the `AfterSyncManifest` callback. The
-  // Flush's VersionEdit actually made it into the Manifest. So these keys can
-  // be read back. Read them to check all live sst files and blob files.
-  ASSERT_EQ("bar", Get("foo"));
-  ASSERT_EQ("value", Get("key"));
 }

 TEST_F(DBBasicTest, DestroyDefaultCfHandle) {
@@ -4665,7 +4472,7 @@ TEST_F(DBBasicTest, FailOpenIfLoggerCreationFail) {
  SyncPoint::GetInstance()->ClearAllCallBacks();
  SyncPoint::GetInstance()->SetCallBack(
      "rocksdb::CreateLoggerFromOptions:AfterGetPath", [&](void* arg) {
-        auto* s = static_cast<Status*>(arg);
+        auto* s = reinterpret_cast<Status*>(arg);
        assert(s);
        *s = Status::IOError("Injected");
      });
@@ -389,7 +389,6 @@ TEST_F(DBBlockCacheTest, FillCacheAndIterateDB) {
  while (iter->Valid()) {
    iter->Next();
  }
-  ASSERT_OK(iter->status());
  delete iter;
  iter = nullptr;
 }
@@ -1373,7 +1372,7 @@ class StableCacheKeyTestFS : public FaultInjectionTestFS {
    SetFailGetUniqueId(true);
  }

-  ~StableCacheKeyTestFS() override {}
+  virtual ~StableCacheKeyTestFS() override {}

  IOStatus LinkFile(const std::string&, const std::string&, const IOOptions&,
                    IODebugContext*) override {
@@ -1424,7 +1423,7 @@ TEST_P(DBBlockCacheKeyTest, StableCacheKeys) {
    ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
        "BlockBasedTableBuilder::BlockBasedTableBuilder:PreSetupBaseCacheKey",
        [&](void* arg) {
-          TableProperties* props = static_cast<TableProperties*>(arg);
+          TableProperties* props = reinterpret_cast<TableProperties*>(arg);
          props->orig_file_number = 0;
        });
    ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();