Fix a nuisance compiler warning from clang (#12144)

Summary:
Example:

```
cache/clock_cache.cc:56:7: error: fallthrough annotation in unreachable code [-Werror,-Wimplicit-fallthrough]
      FALLTHROUGH_INTENDED;
      ^
./port/lang.h:10:30: note: expanded from macro 'FALLTHROUGH_INTENDED'
                             ^
```

In clang < 14, this is annoyingly generated from -Wimplicit-fallthrough, but was changed to -Wunreachable-code-fallthrough (implied by -Wunreachable-code) in clang 14. See https://reviews.llvm.org/D107933 for how this nuisance pattern generated false positives similar to ours in the Linux kernel.

Just to underscore the ridiculousness of this warning, here an error is reported on the annotation, not the call to do_something(), depending on the constexpr value (https://godbolt.org/z/EvxqdPTdr):

```
#include <atomic>
void do_something();
void test(int v) {
    switch (v) {
        case 1:
            if constexpr (std::atomic<long>::is_always_lock_free) {
                return;
            } else {
                do_something();
                [[fallthrough]];
            }
        case 2:
            return;
    }
}
```

Pull Request resolved: https://github.com/facebook/rocksdb/pull/12144

Test Plan: Added the warning to our Makefile for USE_CLANG, which reproduced the warning-as-error as shown above, but is now fixed.

Reviewed By: jaykorean

Differential Revision: D52139615

Pulled By: pdillinger

fbshipit-source-id: ba967ae700c0916d1a478bc465cf917633e337d9

.circleci/config.yml

@@ -146,12 +146,9 @@ commands:
   install-maven:
     steps:
       - run:
-          name: Install Maven
+          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
+            sudo apt-get update -y && sudo apt-get install -y maven
 
   setup-folly:
     steps:
@@ -234,7 +231,6 @@ 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 +244,7 @@ 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 +543,7 @@ 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 +553,7 @@ 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
 
@@ -568,7 +564,7 @@ jobs:
       - pre-steps
       - run:
           name: "run crashtest"
-          command: ulimit -S -n `ulimit -H -n` && make V=1 -j32 CRASH_TEST_EXT_ARGS='--duration=10800' 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
 
@@ -627,7 +623,7 @@ jobs:
       - windows-build-steps
 
   build-linux-java:
-    executor: linux-java-docker
+    executor: linux-docker
     resource_class: large
     steps:
       - pre-steps
@@ -640,13 +636,17 @@ 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
+    machine:
+      image: ubuntu-2004:202111-02
     resource_class: large
+    environment:
+      JAVA_HOME: /usr/lib/jvm/java-8-openjdk-amd64
     steps:
+      - install-maven
       - pre-steps
       - run:
           name: "Set Java Environment"
@@ -655,10 +655,9 @@ jobs:
             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'
+          command: make V=1 J=8 -j8 jpmd
       - post-pmd-steps
 
   build-linux-java-static:
@@ -878,11 +877,68 @@ 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-avx2
+      - 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
+      - build-linux-java-pmd
+  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 +958,9 @@ 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

.github/actions/build-folly/action.yml

@@ -1,7 +0,0 @@
-name: build-folly
-runs:
-  using: composite
-  steps:
-  - name: Build folly and dependencies
-    run: make build_folly
-    shell: bash

.github/actions/build-for-benchmarks/action.yml

@@ -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

.github/actions/increase-max-open-files-on-macos/action.yml

@@ -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

.github/actions/install-gflags-on-macos/action.yml

@@ -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

.github/actions/install-gflags/action.yml

@@ -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

.github/actions/install-jdk8-on-macos/action.yml

@@ -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

.github/actions/install-maven/action.yml

@@ -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

.github/actions/perform-benchmarks/action.yml

@@ -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

.github/actions/post-benchmarks/action.yml

@@ -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

.github/actions/post-steps/action.yml

@@ -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

.github/actions/pre-steps-macos/action.yml

@@ -1,5 +0,0 @@
-name: pre-steps-macos
-runs:
-  using: composite
-  steps:
-  - uses: "./.github/actions/pre-steps"

.github/actions/pre-steps/action.yml

@@ -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

.github/actions/setup-folly/action.yml

@@ -1,7 +0,0 @@
-name: setup-folly
-runs:
-  using: composite
-  steps:
-  - name: Checkout folly sources
-    run: make checkout_folly
-    shell: bash

.github/actions/setup-upstream/action.yml

@@ -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

.github/actions/windows-build-steps/action.yml

@@ -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

.github/workflows/benchmark-linux.yml

@@ -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"

.github/workflows/nightly-candidate.yml

@@ -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"

.github/workflows/nightly.yml

@@ -1,108 +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"
-  build-linux-arm-test-full:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 4-core-ubuntu-arm
-    steps:
-      - uses: actions/checkout@v4.1.0
-      - uses: "./.github/actions/pre-steps"
-      - run: sudo apt-get update && sudo apt-get install -y build-essential libgflags-dev
-      - run: make V=1 J=4 -j4 check
-      - uses: "./.github/actions/post-steps"

.github/workflows/pr-jobs-candidate.yml

@@ -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"

.github/workflows/pr-jobs.yml

@@ -1,619 +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"
-  build-linux-arm:
-    if: ${{ github.repository_owner == 'facebook' }}
-    runs-on:
-      labels: 4-core-ubuntu-arm
-    steps:
-      - uses: actions/checkout@v4.1.0
-      - uses: "./.github/actions/pre-steps"
-      - run: sudo apt-get update && sudo apt-get install -y build-essential
-      - run: ROCKSDBTESTS_PLATFORM_DEPENDENT=only make V=1 J=4 -j4 all_but_some_tests check_some
-      - uses: "./.github/actions/post-steps"

.github/workflows/sanity_check.yml

@@ -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

CMakeLists.txt

@@ -44,29 +44,6 @@ project(rocksdb
   HOMEPAGE_URL https://rocksdb.org/
   LANGUAGES CXX C ASM)
 
-if(APPLE)
-  # On macOS Cmake, when cross-compiling, sometimes CMAKE_SYSTEM_PROCESSOR wrongfully stays
-  # the same as CMAKE_HOST_SYSTEM_PROCESSOR regardless the target CPU.
-  # The manual call to  set(CMAKE_SYSTEM_PROCESSOR) has to be set after the project() call.
-  # because project() might reset CMAKE_SYSTEM_PROCESSOR back to the value of CMAKE_HOST_SYSTEM_PROCESSOR.
-  # Check if CMAKE_SYSTEM_PROCESSOR is not equal to CMAKE_OSX_ARCHITECTURES
-  if(NOT CMAKE_OSX_ARCHITECTURES STREQUAL "")
-    if(NOT CMAKE_SYSTEM_PROCESSOR STREQUAL CMAKE_OSX_ARCHITECTURES)
-      # Split CMAKE_OSX_ARCHITECTURES into a list
-      string(REPLACE ";" " " ARCH_LIST ${CMAKE_OSX_ARCHITECTURES})
-      separate_arguments(ARCH_LIST UNIX_COMMAND ${ARCH_LIST})
-      # Count the number of architectures
-      list(LENGTH ARCH_LIST ARCH_COUNT)
-      # Ensure that exactly one architecture is specified
-      if(NOT ARCH_COUNT EQUAL 1)
-          message(FATAL_ERROR "CMAKE_OSX_ARCHITECTURES must have exactly one value. Current value: ${CMAKE_OSX_ARCHITECTURES}")
-      endif()
-      set(CMAKE_SYSTEM_PROCESSOR ${CMAKE_OSX_ARCHITECTURES})
-      message(STATUS "CMAKE_SYSTEM_PROCESSOR is manually set to ${CMAKE_SYSTEM_PROCESSOR}")
-    endif()
-  endif()
-endif()
-
 if(POLICY CMP0042)
   cmake_policy(SET CMP0042 NEW)
 endif()
@@ -99,6 +76,11 @@ if (WITH_WINDOWS_UTF8_FILENAMES)
 endif()
 option(ROCKSDB_BUILD_SHARED "Build shared versions of the RocksDB libraries" ON)
 
+if ($ENV{CIRCLECI})
+  message(STATUS "Build for CircieCI env, a few tests may be disabled")
+  add_definitions(-DCIRCLECI)
+endif()
+
 if( NOT DEFINED CMAKE_CXX_STANDARD )
   set(CMAKE_CXX_STANDARD 17)
 endif()
@@ -189,7 +171,7 @@ else()
   if(WITH_ZSTD)
     find_package(zstd REQUIRED)
     add_definitions(-DZSTD)
-    include_directories(${ZSTD_INCLUDE_DIRS})
+    include_directories(${ZSTD_INCLUDE_DIR})
     list(APPEND THIRDPARTY_LIBS zstd::zstd)
   endif()
 endif()
@@ -205,7 +187,7 @@ endif()
 
 if(MSVC)
   set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /Zi /nologo /EHsc /GS /Gd /GR /GF /fp:precise /Zc:wchar_t /Zc:forScope /errorReport:queue")
-  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /FC /d2Zi+ /W4 /wd4127 /wd4996 /wd4100 /wd4324")
+  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /FC /d2Zi+ /W4 /wd4127 /wd4800 /wd4996 /wd4351 /wd4100 /wd4204 /wd4324")
 else()
   set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -W -Wextra -Wall -pthread")
   set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wsign-compare -Wshadow -Wno-unused-parameter -Wno-unused-variable -Woverloaded-virtual -Wnon-virtual-dtor -Wno-missing-field-initializers -Wno-strict-aliasing -Wno-invalid-offsetof")
@@ -214,7 +196,6 @@ else()
   endif()
   if(MINGW)
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-format")
-    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wa,-mbig-obj")
     add_definitions(-D_POSIX_C_SOURCE=1)
   endif()
   if(NOT CMAKE_BUILD_TYPE STREQUAL "Debug")
@@ -395,7 +376,7 @@ option(WITH_NUMA "build with NUMA policy support" OFF)
 if(WITH_NUMA)
   find_package(NUMA REQUIRED)
   add_definitions(-DNUMA)
-  include_directories(${NUMA_INCLUDE_DIRS})
+  include_directories(${NUMA_INCLUDE_DIR})
   list(APPEND THIRDPARTY_LIBS NUMA::NUMA)
 endif()
 
@@ -653,7 +634,6 @@ set(SOURCES
         cache/sharded_cache.cc
         cache/tiered_secondary_cache.cc
         db/arena_wrapped_db_iter.cc
-        db/attribute_group_iterator_impl.cc
         db/blob/blob_contents.cc
         db/blob/blob_fetcher.cc
         db/blob/blob_file_addition.cc
@@ -670,7 +650,6 @@ set(SOURCES
         db/blob/prefetch_buffer_collection.cc
         db/builder.cc
         db/c.cc
-        db/coalescing_iterator.cc
         db/column_family.cc
         db/compaction/compaction.cc
         db/compaction/compaction_iterator.cc
@@ -955,7 +934,6 @@ set(SOURCES
         utilities/transactions/write_prepared_txn_db.cc
         utilities/transactions/write_unprepared_txn.cc
         utilities/transactions/write_unprepared_txn_db.cc
-        utilities/types_util.cc
         utilities/ttl/db_ttl_impl.cc
         utilities/wal_filter.cc
         utilities/write_batch_with_index/write_batch_with_index.cc
@@ -1037,10 +1015,8 @@ endif()
 
 else()
   list(APPEND SOURCES
-    db/db_impl/db_impl_follower.cc
     port/port_posix.cc
     env/env_posix.cc
-    env/fs_on_demand.cc
     env/fs_posix.cc
     env/io_posix.cc)
 endif()
@@ -1365,7 +1341,6 @@ if(WITH_TESTS)
         db/file_indexer_test.cc
         db/filename_test.cc
         db/flush_job_test.cc
-        db/db_follower_test.cc
         db/import_column_family_test.cc
         db/listener_test.cc
         db/log_test.cc
@@ -1373,7 +1348,6 @@ if(WITH_TESTS)
         db/memtable_list_test.cc
         db/merge_helper_test.cc
         db/merge_test.cc
-        db/multi_cf_iterator_test.cc
         db/options_file_test.cc
         db/perf_context_test.cc
         db/periodic_task_scheduler_test.cc
@@ -1454,7 +1428,6 @@ 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
@@ -1489,7 +1462,6 @@ if(WITH_TESTS)
         utilities/transactions/lock/range/range_locking_test.cc
         utilities/transactions/timestamped_snapshot_test.cc
         utilities/ttl/ttl_test.cc
-        utilities/types_util_test.cc
         utilities/util_merge_operators_test.cc
         utilities/write_batch_with_index/write_batch_with_index_test.cc
 	${PLUGIN_TESTS}
@@ -1505,7 +1477,7 @@ 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} ${FOLLY_LIBS})
@@ -1530,7 +1502,7 @@ if(WITH_TESTS)
       target_link_libraries(${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})
 
@@ -1550,7 +1522,7 @@ 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()
 

HISTORY.md

@@ -1,215 +1,14 @@
 # Rocksdb Change Log
 > NOTE: Entries for next release do not go here. Follow instructions in `unreleased_history/README.txt`
 
-## 9.3.2 (08/02/2024)
+## 8.9.2 (12/12/2023)
 ### Bug Fixes
-* *Make DestroyDB supports slow deletion when it's configured in `SstFileManager`. The slow deletion is subject to the configured `rate_bytes_per_sec`, but not subject to the `max_trash_db_ratio`.
-* Fix a race condition in pessimistic transactions that could allow multiple transactions with the same name to be registered simultaneously, resulting in a crash or other unpredictable behavior.
-
-## 9.3.1 (05/25/2024)
-### Bug Fixes
-* [internal only] Build script improvement
-
-## 9.3.0 (05/17/2024)
-### New Features
-* Optimistic transactions and pessimistic transactions with the WriteCommitted policy now support the `GetEntity` API.
-* Added new `Iterator` property, "rocksdb.iterator.is-value-pinned", for checking whether the `Slice` returned by `Iterator::value()` can be used until the `Iterator` is destroyed.
-* Optimistic transactions and WriteCommitted pessimistic transactions now support the `MultiGetEntity` API.
-* Optimistic transactions and pessimistic transactions with the WriteCommitted policy now support the `PutEntity` API. Support for read APIs and other write policies (WritePrepared, WriteUnprepared) will be added later.
-
-### Public API Changes
-* Exposed block based metadata cache options via C API
-* Exposed compaction pri via c api.
-* Add a kAdmPolicyAllowAll option to TieredAdmissionPolicy that admits all blocks evicted from the primary block cache into the compressed secondary cache.
-
-### Behavior Changes
-* CompactRange() with change_level=true on a CF with FIFO compaction will return Status::NotSupported().
-* External file ingestion with FIFO compaction will always ingest to L0.
-
-### Bug Fixes
-* Fixed a bug for databases using `DBOptions::allow_2pc == true` (all `TransactionDB`s except `OptimisticTransactionDB`) that have exactly one column family. Due to a missing WAL sync, attempting to open the DB could have returned a `Status::Corruption` with a message like "SST file is ahead of WALs".
-* Fix a bug in CreateColumnFamilyWithImport() where if multiple CFs are imported, we were not resetting files' epoch number and L0 files can have overlapping key range but the same epoch number.
-* Fixed race conditions when `ColumnFamilyOptions::inplace_update_support == true` between user overwrites and reads on the same key.
-* Fix a bug where `CompactFiles()` can compact files of range conflict with other ongoing compactions' when `preclude_last_level_data_seconds > 0` is used
-* Fixed a false positive `Status::Corruption` reported when reopening a DB that used `DBOptions::recycle_log_file_num > 0` and `DBOptions::wal_compression != kNoCompression`.
-* While WAL is locked with LockWAL(), some operations like Flush() and IngestExternalFile() are now blocked as they should have been.
-* Fixed a bug causing stale memory access when using the TieredSecondaryCache with an NVM secondary cache, and a file system that supports return an FS allocated buffer for MultiRead (FSSupportedOps::kFSBuffer is set).
-
-## 9.2.0 (05/01/2024)
-### New Features
-* Added two options `deadline` and `max_size_bytes` for CacheDumper to exit early
-* Added a new API `GetEntityFromBatchAndDB` to `WriteBatchWithIndex` that can be used for wide-column point lookups with read-your-own-writes consistency. Similarly to `GetFromBatchAndDB`, the API can combine data from the write batch with data from the underlying database if needed. See the API comments for more details.
-* [Experimental] Introduce two new cross-column-family iterators - CoalescingIterator and AttributeGroupIterator. The CoalescingIterator enables users to iterate over multiple column families and access their values and columns. During this iteration, if the same key exists in more than one column family, the keys in the later column family will overshadow the previous ones. The AttributeGroupIterator allows users to gather wide columns per Column Family and create attribute groups while iterating over keys across all CFs.
-* Added a new API `MultiGetEntityFromBatchAndDB` to `WriteBatchWithIndex` that can be used for batched wide-column point lookups with read-your-own-writes consistency. Similarly to `MultiGetFromBatchAndDB`, the API can combine data from the write batch with data from the underlying database if needed. See the API comments for more details.
-* *Adds a `SstFileReader::NewTableIterator` API to support programmatically read a SST file as a raw table file.
-* Add an option to `WaitForCompactOptions` - `wait_for_purge` to make `WaitForCompact()` API wait for background purge to complete
-
-### Public API Changes
-* DeleteRange() will return NotSupported() if row_cache is configured since they don't work together in some cases.
-* Deprecated `CompactionOptions::compression` since `CompactionOptions`'s API for configuring compression was incomplete, unsafe, and likely unnecessary
-* Using `OptionChangeMigration()` to migrate from non-FIFO to FIFO compaction
-with `Options::compaction_options_fifo.max_table_files_size` > 0 can cause
-the whole DB to be dropped right after migration if the migrated data is larger than
-`max_table_files_size`
-
-### Behavior Changes
-* Enabling `BlockBasedTableOptions::block_align` is now incompatible (i.e., APIs will return `Status::InvalidArgument`) with more ways of enabling compression: `CompactionOptions::compression`, `ColumnFamilyOptions::compression_per_level`, and `ColumnFamilyOptions::bottommost_compression`.
-* Changed the default value of `CompactionOptions::compression` to `kDisableCompressionOption`, which means the compression type is determined by the `ColumnFamilyOptions`.
-* `BlockBasedTableOptions::optimize_filters_for_memory` is now set to true by default. When `partition_filters=false`, this could lead to somewhat increased average RSS memory usage by the block cache, but this "extra" usage is within the allowed memory budget and should make memory usage more consistent (by minimizing internal fragmentation for more kinds of blocks).
-* Dump all keys for cache dumper impl if `SetDumpFilter()` is not called
-* `CompactRange()` with `CompactRangeOptions::change_level = true` and `CompactRangeOptions::target_level = 0` that ends up moving more than 1 file from non-L0 to L0 will return `Status::Aborted()`.
-* On distributed file systems that support file system level checksum verification and reconstruction reads, RocksDB will now retry a file read if the initial read fails RocksDB block level or record level checksum verification. This applies to MANIFEST file reads when the DB is opened, and to SST file reads at all times.
-
-### Bug Fixes
-* Fix a bug causing `VerifyFileChecksums()` to return false-positive corruption under `BlockBasedTableOptions::block_align=true`
-* Provide consistent view of the database across the column families for `NewIterators()` API.
-* Fixed feature interaction bug for `DeleteRange()` together with `ColumnFamilyOptions::memtable_insert_with_hint_prefix_extractor`. The impact of this bug would likely be corruption or crashing.
-* Fixed hang in `DisableManualCompactions()` where compactions waiting to be scheduled due to conflicts would not be canceled promptly
-* Fixed a regression when `ColumnFamilyOptions::max_successive_merges > 0` where the CPU overhead for deciding whether to merge could have increased unless the user had set the option `ColumnFamilyOptions::strict_max_successive_merges`
-* Fixed a bug in `MultiGet()` and `MultiGetEntity()` together with blob files (`ColumnFamilyOptions::enable_blob_files == true`). An error looking up one of the keys could cause the results to be wrong for other keys for which the statuses were `Status::OK`.
-* Fixed a bug where wrong padded bytes are used to generate file checksum and `DataVerificationInfo::checksum` upon file creation
-* Correctly implemented the move semantics of `PinnableWideColumns`.
-* Fixed a bug when the recycle_log_file_num in DBOptions is changed from 0 to non-zero when a DB is reopened. On a subsequent reopen, if a log file created when recycle_log_file_num==0 was reused previously, is alive and is empty, we could end up inserting stale WAL records into the memtable.
-* *Fix a bug where obsolete files' deletion during DB::Open are not rate limited with `SstFilemManager`'s slow deletion feature even if it's configured.
-
-## 9.1.0 (03/22/2024)
-### New Features
-* Added an option, `GetMergeOperandsOptions::continue_cb`, to give users the ability to end `GetMergeOperands()`'s lookup process before all merge operands were found.
-* *Add sanity checks for ingesting external files that currently checks if the user key comparator used to create the file is compatible with the column family's user key comparator.
-*Support ingesting external files for column family that has user-defined timestamps in memtable only enabled.
-* On file systems that support storage level data checksum and reconstruction, retry SST block reads for point lookups, scans, and flush and compaction if there's a checksum mismatch on the initial read.
-* Some enhancements and fixes to experimental Temperature handling features, including new `default_write_temperature` CF option and opening an `SstFileWriter` with a temperature.
-* `WriteBatchWithIndex` now supports wide-column point lookups via the `GetEntityFromBatch` API. See the API comments for more details.
-* *Implement experimental features: API `Iterator::GetProperty("rocksdb.iterator.write-time")` to allow users to get data's approximate write unix time and write data with a specific write time via `WriteBatch::TimedPut` API.
-
-### Public API Changes
-* Best-effort recovery (`best_efforts_recovery == true`) may now be used together with atomic flush (`atomic_flush == true`). The all-or-nothing recovery guarantee for atomically flushed data will be upheld.
-* Remove deprecated option `bottommost_temperature`, already replaced by `last_level_temperature`
-* Added new PerfContext counters for block cache bytes read - block_cache_index_read_byte, block_cache_filter_read_byte, block_cache_compression_dict_read_byte, and block_cache_read_byte.
-* Deprecate experimental Remote Compaction APIs - StartV2() and WaitForCompleteV2() and introduce Schedule() and Wait(). The new APIs essentially does the same thing as the old APIs. They allow taking externally generated unique id to wait for remote compaction to complete.
-* *For API `WriteCommittedTransaction::GetForUpdate`, if the column family enables user-defined timestamp, it was mandated that argument `do_validate` cannot be false, and UDT based validation has to be done with a user set read timestamp. It's updated to make the UDT based validation optional if user sets `do_validate` to false and does not set a read timestamp. With this, `GetForUpdate` skips UDT based validation and it's users' responsibility to enforce the UDT invariant. SO DO NOT skip this UDT-based validation if users do not have ways to enforce the UDT invariant. Ways to enforce the invariant on the users side include manage a monotonically increasing timestamp, commit transactions in a single thread etc.
-* Defined a new PerfLevel `kEnableWait` to measure time spent by user threads blocked in RocksDB other than mutex, such as a write thread waiting to be added to a write group, a write thread delayed or stalled etc.
-* `RateLimiter`'s API no longer requires the burst size to be the refill size. Users of `NewGenericRateLimiter()` can now provide burst size in `single_burst_bytes`. Implementors of `RateLimiter::SetSingleBurstBytes()` need to adapt their implementations to match the changed API doc.
-* Add `write_memtable_time` to the newly introduced PerfLevel `kEnableWait`.
-
-### Behavior Changes
-* `RateLimiter`s created by `NewGenericRateLimiter()` no longer modify the refill period when `SetSingleBurstBytes()` is called.
-* Merge writes will only keep merge operand count within `ColumnFamilyOptions::max_successive_merges` when the key's merge operands are all found in memory, unless `strict_max_successive_merges` is explicitly set.
-
-### Bug Fixes
-* Fixed `kBlockCacheTier` reads to return `Status::Incomplete` when I/O is needed to fetch a merge chain's base value from a blob file.
-* Fixed `kBlockCacheTier` reads to return `Status::Incomplete` on table cache miss rather than incorrectly returning an empty value.
-* Fixed a data race in WalManager that may affect how frequent PurgeObsoleteWALFiles() runs.
-* Re-enable the recycle_log_file_num option in DBOptions for kPointInTimeRecovery WAL recovery mode, which was previously disabled due to a bug in the recovery logic. This option is incompatible with WriteOptions::disableWAL. A Status::InvalidArgument() will be returned if disableWAL is specified.
-
-### Performance Improvements
-* Java API `multiGet()` variants now take advantage of the underlying batched `multiGet()` performance improvements.
-Before
-```
-Benchmark (columnFamilyTestType) (keyCount) (keySize) (multiGetSize) (valueSize) Mode Cnt Score Error Units
-MultiGetBenchmarks.multiGetList10 no_column_family 10000 16 100 64 thrpt 25 6315.541 ± 8.106 ops/s
-MultiGetBenchmarks.multiGetList10 no_column_family 10000 16 100 1024 thrpt 25 6975.468 ± 68.964 ops/s
-```
-After
-```
-Benchmark (columnFamilyTestType) (keyCount) (keySize) (multiGetSize) (valueSize) Mode Cnt Score Error Units
-MultiGetBenchmarks.multiGetList10 no_column_family 10000 16 100 64 thrpt 25 7046.739 ± 13.299 ops/s
-MultiGetBenchmarks.multiGetList10 no_column_family 10000 16 100 1024 thrpt 25 7654.521 ± 60.121 ops/s
-```
-
-## 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)
-
-### 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.
-
-### 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.
+* Sanitize the cache_opts->secondary_cache field in TieredCacheOptions. It can be mistakenly set by the user, which would conflict with nvm_sec_cache in TieredCacheOptions if not sanitized to nullptr.
 
-### 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
-
+## 8.9.1 (12/8/2023)
+### Bug Fixes
+* Avoid destroying the periodic task scheduler's default timer in order to prevent static destruction order issues.
 
 ## 8.9.0 (11/17/2023)
 ### New Features

INSTALL.md

@@ -126,6 +126,7 @@ most processors made since roughly 2013.
         * Update XCode:  run `xcode-select --install` (or install it from XCode App's settting).
         * Install via [homebrew](http://brew.sh/).
             * If you're first time developer in MacOS, you still need to run: `xcode-select --install` in your command line.
+            * run `brew tap homebrew/versions; brew install gcc7 --use-llvm` to install gcc 7 (or higher).
     * run `brew install rocksdb`
 
 * **FreeBSD** (11.01):
@@ -168,26 +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`.

Makefile

@@ -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 -Wimplicit-fallthrough
 endif
 
 ifeq ($(PLATFORM), OS_OPENBSD)
@@ -1610,9 +1609,6 @@ object_registry_test: $(OBJ_DIR)/utilities/object_registry_test.o $(TEST_LIBRARY
 ttl_test: $(OBJ_DIR)/utilities/ttl/ttl_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
-types_util_test: $(OBJ_DIR)/utilities/types_util_test.o $(TEST_LIBRARY) $(LIBRARY)
-	$(AM_LINK)
-
 write_batch_with_index_test: $(OBJ_DIR)/utilities/write_batch_with_index/write_batch_with_index_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
@@ -1643,9 +1639,6 @@ wal_edit_test: $(OBJ_DIR)/db/wal_edit_test.o $(TEST_LIBRARY) $(LIBRARY)
 dbformat_test: $(OBJ_DIR)/db/dbformat_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
-multi_cf_iterator_test: $(OBJ_DIR)/db/multi_cf_iterator_test.o $(TEST_LIBRARY) $(LIBRARY)
-	$(AM_LINK)
-
 env_basic_test: $(OBJ_DIR)/env/env_basic_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
@@ -1775,9 +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)
 
@@ -1922,9 +1912,6 @@ sst_file_reader_test: $(OBJ_DIR)/table/sst_file_reader_test.o $(TEST_LIBRARY) $(
 db_secondary_test: $(OBJ_DIR)/db/db_secondary_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
-db_follower_test: $(OBJ_DIR)/db/db_follower_test.o $(TEST_LIBRARY) $(LIBRARY)
-	$(AM_LINK)
-
 block_cache_tracer_test: $(OBJ_DIR)/trace_replay/block_cache_tracer_test.o $(TEST_LIBRARY) $(LIBRARY)
 	$(AM_LINK)
 
@@ -2073,7 +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, $(MACHINE)))
 		ARCH := 64
 	else
 		ARCH := 32
@@ -2094,7 +2081,7 @@ ifneq ($(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)$(JNI_LIBC_POSTFIX).so
@@ -2107,8 +2094,8 @@ ROCKSDB_JAVADOCS_JAR = rocksdbjni-$(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
@@ -2341,47 +2328,43 @@ rocksdbjavastaticreleasedocker: rocksdbjavastaticosx rocksdbjavastaticdockerx86
 
 rocksdbjavastaticdockerx86:
 	mkdir -p java/target
-	docker run --rm --name rocksdb_linux_x86-be --platform linux/386 --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) --env J=$(J) evolvedbinary/rocksjava:centos6_x86-be /rocksdb-host/java/crossbuild/docker-build-linux.sh
+	docker run --rm --name rocksdb_linux_x86-be --platform linux/386 --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:centos6_x86-be /rocksdb-host/java/crossbuild/docker-build-linux-centos.sh
 
 rocksdbjavastaticdockerx86_64:
 	mkdir -p java/target
-	docker run --rm --name rocksdb_linux_x64-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) --env J=$(J) evolvedbinary/rocksjava:centos6_x64-be /rocksdb-host/java/crossbuild/docker-build-linux.sh
+	docker run --rm --name rocksdb_linux_x64-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:centos6_x64-be /rocksdb-host/java/crossbuild/docker-build-linux-centos.sh
 
 rocksdbjavastaticdockerppc64le:
 	mkdir -p java/target
-	docker run --rm --name rocksdb_linux_ppc64le-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) --env J=$(J) evolvedbinary/rocksjava:centos7_ppc64le-be /rocksdb-host/java/crossbuild/docker-build-linux.sh
+	docker run --rm --name rocksdb_linux_ppc64le-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:centos7_ppc64le-be /rocksdb-host/java/crossbuild/docker-build-linux-centos.sh
 
 rocksdbjavastaticdockerarm64v8:
 	mkdir -p java/target
-	docker run --rm --name rocksdb_linux_arm64v8-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) --env J=$(J) evolvedbinary/rocksjava:centos7_arm64v8-be /rocksdb-host/java/crossbuild/docker-build-linux.sh
+	docker run --rm --name rocksdb_linux_arm64v8-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:centos7_arm64v8-be /rocksdb-host/java/crossbuild/docker-build-linux-centos.sh
 
 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) --env J=$(J) evolvedbinary/rocksjava:ubuntu18_s390x-be /rocksdb-host/java/crossbuild/docker-build-linux.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) --env J=$(J) evolvedbinary/rocksjava:ubuntu20_riscv64-be /rocksdb-host/java/crossbuild/docker-build-linux.sh
+	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
 
 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) --env J=$(J) evolvedbinary/rocksjava:alpine3_x86-be /rocksdb-host/java/crossbuild/docker-build-linux.sh
+	docker run --rm --name rocksdb_linux_x86-musl-be --platform linux/386 --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:alpine3_x86-be /rocksdb-host/java/crossbuild/docker-build-linux-alpine.sh
 
 rocksdbjavastaticdockerx86_64musl:
 	mkdir -p java/target
-	docker run --rm --name rocksdb_linux_x64-musl-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) --env J=$(J) evolvedbinary/rocksjava:alpine3_x64-be /rocksdb-host/java/crossbuild/docker-build-linux.sh
+	docker run --rm --name rocksdb_linux_x64-musl-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:alpine3_x64-be /rocksdb-host/java/crossbuild/docker-build-linux-alpine.sh
 
 rocksdbjavastaticdockerppc64lemusl:
 	mkdir -p java/target
-	docker run --rm --name rocksdb_linux_ppc64le-musl-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) --env J=$(J) evolvedbinary/rocksjava:alpine3_ppc64le-be /rocksdb-host/java/crossbuild/docker-build-linux.sh
+	docker run --rm --name rocksdb_linux_ppc64le-musl-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:alpine3_ppc64le-be /rocksdb-host/java/crossbuild/docker-build-linux-alpine.sh
 
 rocksdbjavastaticdockerarm64v8musl:
 	mkdir -p java/target
-	docker run --rm --name rocksdb_linux_arm64v8-musl-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) --env J=$(J) evolvedbinary/rocksjava:alpine3_arm64v8-be /rocksdb-host/java/crossbuild/docker-build-linux.sh
+	docker run --rm --name rocksdb_linux_arm64v8-musl-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:alpine3_arm64v8-be /rocksdb-host/java/crossbuild/docker-build-linux-alpine.sh
 
 rocksdbjavastaticdockers390xmusl:
 	mkdir -p java/target
-	docker run --rm --name rocksdb_linux_s390x-musl-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) --env J=$(J) evolvedbinary/rocksjava:alpine3_s390x-be /rocksdb-host/java/crossbuild/docker-build-linux.sh
+	docker run --rm --name rocksdb_linux_s390x-musl-be --attach stdin --attach stdout --attach stderr --volume $(HOME)/.m2:/root/.m2:ro --volume `pwd`:/rocksdb-host:ro --volume /rocksdb-local-build --volume `pwd`/java/target:/rocksdb-java-target --env DEBUG_LEVEL=$(DEBUG_LEVEL) evolvedbinary/rocksjava:alpine3_s390x-be /rocksdb-host/java/crossbuild/docker-build-linux-alpine.sh
 
 rocksdbjavastaticpublish: rocksdbjavastaticrelease rocksdbjavastaticpublishcentral
 

TARGETS

@@ -4,7 +4,6 @@
 # This file is a Facebook-specific integration for buck builds, so can
 # only be validated by Facebook employees.
 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")
-load("@fbcode_macros//build_defs:export_files.bzl", "export_file")
 
 
 cpp_library_wrapper(name="rocksdb_lib", srcs=[
@@ -22,7 +21,6 @@ cpp_library_wrapper(name="rocksdb_lib", srcs=[
         "cache/sharded_cache.cc",
         "cache/tiered_secondary_cache.cc",
         "db/arena_wrapped_db_iter.cc",
-        "db/attribute_group_iterator_impl.cc",
         "db/blob/blob_contents.cc",
         "db/blob/blob_fetcher.cc",
         "db/blob/blob_file_addition.cc",
@@ -39,7 +37,6 @@ cpp_library_wrapper(name="rocksdb_lib", srcs=[
         "db/blob/prefetch_buffer_collection.cc",
         "db/builder.cc",
         "db/c.cc",
-        "db/coalescing_iterator.cc",
         "db/column_family.cc",
         "db/compaction/compaction.cc",
         "db/compaction/compaction_iterator.cc",
@@ -61,7 +58,6 @@ cpp_library_wrapper(name="rocksdb_lib", srcs=[
         "db/db_impl/db_impl_debug.cc",
         "db/db_impl/db_impl_experimental.cc",
         "db/db_impl/db_impl_files.cc",
-        "db/db_impl/db_impl_follower.cc",
         "db/db_impl/db_impl_open.cc",
         "db/db_impl/db_impl_readonly.cc",
         "db/db_impl/db_impl_secondary.cc",
@@ -119,7 +115,6 @@ cpp_library_wrapper(name="rocksdb_lib", srcs=[
         "env/env_posix.cc",
         "env/file_system.cc",
         "env/file_system_tracer.cc",
-        "env/fs_on_demand.cc",
         "env/fs_posix.cc",
         "env/fs_remap.cc",
         "env/io_posix.cc",
@@ -350,7 +345,6 @@ cpp_library_wrapper(name="rocksdb_lib", srcs=[
         "utilities/transactions/write_unprepared_txn.cc",
         "utilities/transactions/write_unprepared_txn_db.cc",
         "utilities/ttl/db_ttl_impl.cc",
-        "utilities/types_util.cc",
         "utilities/wal_filter.cc",
         "utilities/write_batch_with_index/write_batch_with_index.cc",
         "utilities/write_batch_with_index/write_batch_with_index_internal.cc",
@@ -4798,12 +4792,6 @@ cpp_unittest_wrapper(name="db_flush_test",
             extra_compiler_flags=[])
 
 
-cpp_unittest_wrapper(name="db_follower_test",
-            srcs=["db/db_follower_test.cc"],
-            deps=[":rocksdb_test_lib"],
-            extra_compiler_flags=[])
-
-
 cpp_unittest_wrapper(name="db_inplace_update_test",
             srcs=["db/db_inplace_update_test.cc"],
             deps=[":rocksdb_test_lib"],
@@ -5238,12 +5226,6 @@ cpp_unittest_wrapper(name="mock_env_test",
             extra_compiler_flags=[])
 
 
-cpp_unittest_wrapper(name="multi_cf_iterator_test",
-            srcs=["db/multi_cf_iterator_test.cc"],
-            deps=[":rocksdb_test_lib"],
-            extra_compiler_flags=[])
-
-
 cpp_unittest_wrapper(name="object_registry_test",
             srcs=["utilities/object_registry_test.cc"],
             deps=[":rocksdb_test_lib"],
@@ -5538,12 +5520,6 @@ cpp_unittest_wrapper(name="ttl_test",
             extra_compiler_flags=[])
 
 
-cpp_unittest_wrapper(name="types_util_test",
-            srcs=["utilities/types_util_test.cc"],
-            deps=[":rocksdb_test_lib"],
-            extra_compiler_flags=[])
-
-
 cpp_unittest_wrapper(name="udt_util_test",
             srcs=["util/udt_util_test.cc"],
             deps=[":rocksdb_test_lib"],

buckifier/targets_cfg.py

@@ -7,7 +7,6 @@ rocksdb_target_header_template = """# This file \100generated by:
 # This file is a Facebook-specific integration for buck builds, so can
 # only be validated by Facebook employees.
 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")
-load("@fbcode_macros//build_defs:export_files.bzl", "export_file")
 
 """
 

build_tools/build_detect_platform

@@ -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

build_tools/dependencies_platform010.sh

@@ -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

build_tools/fbcode_config.sh

@@ -113,7 +113,7 @@ CLANG_LIB="$CLANG_BASE/lib"
 CLANG_SRC="$CLANG_BASE/../../src"
 
 CLANG_ANALYZER="$CLANG_BIN/clang++"
-CLANG_SCAN_BUILD="$CLANG_BIN/scan-build
+CLANG_SCAN_BUILD="$CLANG_SRC/llvm/tools/clang/tools/scan-build/bin/scan-build"
 
 if [ -z "$USE_CLANG" ]; then
   # gcc

build_tools/fbcode_config_platform010.sh

@@ -110,7 +110,7 @@ CLANG_LIB="$CLANG_BASE/lib"
 CLANG_SRC="$CLANG_BASE/../../src"
 
 CLANG_ANALYZER="$CLANG_BIN/clang++"
-CLANG_SCAN_BUILD="$CLANG_BIN/scan-build"
+CLANG_SCAN_BUILD="$CLANG_SRC/llvm/clang/tools/scan-build/bin/scan-build"
 
 if [ -z "$USE_CLANG" ]; then
   # gcc

build_tools/format-diff.sh

@@ -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!"

build_tools/gnu_parallel

@@ -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 {

build_tools/update_dependencies.sh

@@ -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
 

cache/cache_bench_tool.cc

@@ -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.");
@@ -194,7 +190,7 @@ class SharedState {
       : cv_(&mu_),
         cache_bench_(cache_bench) {}
 
-  ~SharedState() = default;
+  ~SharedState() {}
 
   port::Mutex* GetMutex() { return &mu_; }
 
@@ -395,7 +391,6 @@ 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
@@ -425,7 +420,7 @@ class CacheBench {
     }
   }
 
-  ~CacheBench() = default;
+  ~CacheBench() {}
 
   void PopulateCache() {
     Random64 rnd(FLAGS_seed);

cache/cache_reservation_manager_test.cc

@@ -129,6 +129,7 @@ TEST_F(CacheReservationManagerTest,
 
 TEST(CacheReservationManagerIncreaseReservcationOnFullCacheTest,
      IncreaseCacheReservationOnFullCache) {
+  ;
   constexpr std::size_t kSizeDummyEntry =
       CacheReservationManagerImpl<CacheEntryRole::kMisc>::GetDummyEntrySize();
   constexpr std::size_t kSmallCacheCapacity = 4 * kSizeDummyEntry;

cache/cache_test.cc

@@ -106,7 +106,7 @@ class CacheTest : public testing::Test,
     type_ = GetParam();
   }
 
-  ~CacheTest() override = default;
+  ~CacheTest() override {}
 
   // These functions encode/decode keys in tests cases that use
   // int keys.
@@ -766,9 +766,7 @@ TEST_P(CacheTest, OverCapacity) {
     std::string key = EncodeKey(i + 1);
     auto h = cache.Lookup(key);
     ASSERT_TRUE(h != nullptr);
-    if (h) {
-      cache.Release(h);
-    }
+    if (h) cache.Release(h);
   }
 
   // the cache is over capacity since nothing could be evicted
@@ -779,7 +777,7 @@ TEST_P(CacheTest, OverCapacity) {
 
   if (IsHyperClock()) {
     // Make sure eviction is triggered.
-    ASSERT_OK(cache.Insert(EncodeKey(-1), nullptr, 1, handles.data()));
+    ASSERT_OK(cache.Insert(EncodeKey(-1), nullptr, 1, &handles[0]));
 
     // cache is under capacity now since elements were released
     ASSERT_GE(n, cache.get()->GetUsage());

cache/clock_cache.cc

@@ -15,7 +15,6 @@
 #include <cassert>
 #include <cinttypes>
 #include <cstddef>
-#include <cstdint>
 #include <exception>
 #include <functional>
 #include <numeric>
@@ -95,8 +94,7 @@ inline void Unref(const ClockHandle& h, uint64_t count = 1) {
   (void)old_meta;
 }
 
-inline bool ClockUpdate(ClockHandle& h, BaseClockTable::EvictionData* data,
-                        bool* purgeable = nullptr) {
+inline bool ClockUpdate(ClockHandle& h, bool* purgeable = nullptr) {
   uint64_t meta;
   if (purgeable) {
     assert(*purgeable == false);
@@ -128,7 +126,6 @@ inline bool ClockUpdate(ClockHandle& h, BaseClockTable::EvictionData* data,
       (meta >> ClockHandle::kReleaseCounterShift) & ClockHandle::kCounterMask;
   if (acquire_count != release_count) {
     // Only clock update entries with no outstanding refs
-    data->seen_pinned_count++;
     return false;
   }
   if ((meta >> ClockHandle::kStateShift == ClockHandle::kStateVisible) &&
@@ -152,8 +149,6 @@ inline bool ClockUpdate(ClockHandle& h, BaseClockTable::EvictionData* data,
                               << ClockHandle::kStateShift) |
                                  (meta & ClockHandle::kHitBitMask))) {
     // Took ownership.
-    data->freed_charge += h.GetTotalCharge();
-    data->freed_count += 1;
     return true;
   } else {
     // Compare-exchange failing probably
@@ -361,18 +356,6 @@ void ConstApplyToEntriesRange(const Func& func, const HandleImpl* begin,
   }
 }
 
-constexpr uint32_t kStrictCapacityLimitBit = 1u << 31;
-
-uint32_t SanitizeEncodeEecAndScl(int eviction_effort_cap,
-                                 bool strict_capacit_limit) {
-  eviction_effort_cap = std::max(int{1}, eviction_effort_cap);
-  eviction_effort_cap =
-      std::min(static_cast<int>(~kStrictCapacityLimitBit), eviction_effort_cap);
-  uint32_t eec_and_scl = static_cast<uint32_t>(eviction_effort_cap);
-  eec_and_scl |= strict_capacit_limit ? kStrictCapacityLimitBit : 0;
-  return eec_and_scl;
-}
-
 }  // namespace
 
 void ClockHandleBasicData::FreeData(MemoryAllocator* allocator) const {
@@ -402,20 +385,17 @@ HandleImpl* BaseClockTable::StandaloneInsert(
 
 template <class Table>
 typename Table::HandleImpl* BaseClockTable::CreateStandalone(
-    ClockHandleBasicData& proto, size_t capacity, uint32_t eec_and_scl,
+    ClockHandleBasicData& proto, size_t capacity, bool strict_capacity_limit,
     bool allow_uncharged) {
   Table& derived = static_cast<Table&>(*this);
   typename Table::InsertState state;
   derived.StartInsert(state);
 
   const size_t total_charge = proto.GetTotalCharge();
-  // NOTE: we can use eec_and_scl as eviction_effort_cap below because
-  // strict_capacity_limit=true is supposed to disable the limit on eviction
-  // effort, and a large value effectively does that.
-  if (eec_and_scl & kStrictCapacityLimitBit) {
+  if (strict_capacity_limit) {
     Status s = ChargeUsageMaybeEvictStrict<Table>(
         total_charge, capacity,
-        /*need_evict_for_occupancy=*/false, eec_and_scl, state);
+        /*need_evict_for_occupancy=*/false, state);
     if (!s.ok()) {
       if (allow_uncharged) {
         proto.total_charge = 0;
@@ -427,7 +407,7 @@ typename Table::HandleImpl* BaseClockTable::CreateStandalone(
     // Case strict_capacity_limit == false
     bool success = ChargeUsageMaybeEvictNonStrict<Table>(
         total_charge, capacity,
-        /*need_evict_for_occupancy=*/false, eec_and_scl, state);
+        /*need_evict_for_occupancy=*/false, state);
     if (!success) {
       // Force the issue
       usage_.FetchAddRelaxed(total_charge);
@@ -440,7 +420,7 @@ typename Table::HandleImpl* BaseClockTable::CreateStandalone(
 template <class Table>
 Status BaseClockTable::ChargeUsageMaybeEvictStrict(
     size_t total_charge, size_t capacity, bool need_evict_for_occupancy,
-    uint32_t eviction_effort_cap, typename Table::InsertState& state) {
+    typename Table::InsertState& state) {
   if (total_charge > capacity) {
     return Status::MemoryLimit(
         "Cache entry too large for a single cache shard: " +
@@ -465,8 +445,7 @@ Status BaseClockTable::ChargeUsageMaybeEvictStrict(
   }
   if (request_evict_charge > 0) {
     EvictionData data;
-    static_cast<Table*>(this)->Evict(request_evict_charge, state, &data,
-                                     eviction_effort_cap);
+    static_cast<Table*>(this)->Evict(request_evict_charge, state, &data);
     occupancy_.FetchSub(data.freed_count);
     if (LIKELY(data.freed_charge > need_evict_charge)) {
       assert(data.freed_count > 0);
@@ -496,7 +475,7 @@ Status BaseClockTable::ChargeUsageMaybeEvictStrict(
 template <class Table>
 inline bool BaseClockTable::ChargeUsageMaybeEvictNonStrict(
     size_t total_charge, size_t capacity, bool need_evict_for_occupancy,
-    uint32_t eviction_effort_cap, typename Table::InsertState& state) {
+    typename Table::InsertState& state) {
   // For simplicity, we consider that either the cache can accept the insert
   // with no evictions, or we must evict enough to make (at least) enough
   // space. It could lead to unnecessary failures or excessive evictions in
@@ -532,8 +511,7 @@ inline bool BaseClockTable::ChargeUsageMaybeEvictNonStrict(
   }
   EvictionData data;
   if (need_evict_charge > 0) {
-    static_cast<Table*>(this)->Evict(need_evict_charge, state, &data,
-                                     eviction_effort_cap);
+    static_cast<Table*>(this)->Evict(need_evict_charge, state, &data);
     // Deal with potential occupancy deficit
     if (UNLIKELY(need_evict_for_occupancy) && data.freed_count == 0) {
       assert(data.freed_charge == 0);
@@ -552,7 +530,11 @@ inline bool BaseClockTable::ChargeUsageMaybeEvictNonStrict(
   return true;
 }
 
-void BaseClockTable::TrackAndReleaseEvictedEntry(ClockHandle* h) {
+void BaseClockTable::TrackAndReleaseEvictedEntry(
+    ClockHandle* h, BaseClockTable::EvictionData* data) {
+  data->freed_charge += h->GetTotalCharge();
+  data->freed_count += 1;
+
   bool took_value_ownership = false;
   if (eviction_callback_) {
     // For key reconstructed from hash
@@ -560,7 +542,7 @@ void BaseClockTable::TrackAndReleaseEvictedEntry(ClockHandle* h) {
     took_value_ownership =
         eviction_callback_(ClockCacheShard<FixedHyperClockTable>::ReverseHash(
                                h->GetHash(), &unhashed, hash_seed_),
-                           static_cast<Cache::Handle*>(h),
+                           reinterpret_cast<Cache::Handle*>(h),
                            h->meta.LoadRelaxed() & ClockHandle::kHitBitMask);
   }
   if (!took_value_ownership) {
@@ -569,20 +551,11 @@ void BaseClockTable::TrackAndReleaseEvictedEntry(ClockHandle* h) {
   MarkEmpty(*h);
 }
 
-bool IsEvictionEffortExceeded(const BaseClockTable::EvictionData& data,
-                              uint32_t eviction_effort_cap) {
-  // Basically checks whether the ratio of useful effort to wasted effort is
-  // too low, with a start-up allowance for wasted effort before any useful
-  // effort.
-  return (data.freed_count + 1U) * uint64_t{eviction_effort_cap} <=
-         data.seen_pinned_count;
-}
-
 template <class Table>
 Status BaseClockTable::Insert(const ClockHandleBasicData& proto,
                               typename Table::HandleImpl** handle,
                               Cache::Priority priority, size_t capacity,
-                              uint32_t eec_and_scl) {
+                              bool strict_capacity_limit) {
   using HandleImpl = typename Table::HandleImpl;
   Table& derived = static_cast<Table&>(*this);
 
@@ -600,12 +573,9 @@ Status BaseClockTable::Insert(const ClockHandleBasicData& proto,
   // strict_capacity_limit, but mostly pessimistic.
   bool use_standalone_insert = false;
   const size_t total_charge = proto.GetTotalCharge();
-  // NOTE: we can use eec_and_scl as eviction_effort_cap below because
-  // strict_capacity_limit=true is supposed to disable the limit on eviction
-  // effort, and a large value effectively does that.
-  if (eec_and_scl & kStrictCapacityLimitBit) {
+  if (strict_capacity_limit) {
     Status s = ChargeUsageMaybeEvictStrict<Table>(
-        total_charge, capacity, need_evict_for_occupancy, eec_and_scl, state);
+        total_charge, capacity, need_evict_for_occupancy, state);
     if (!s.ok()) {
       // Revert occupancy
       occupancy_.FetchSubRelaxed(1);
@@ -614,7 +584,7 @@ Status BaseClockTable::Insert(const ClockHandleBasicData& proto,
   } else {
     // Case strict_capacity_limit == false
     bool success = ChargeUsageMaybeEvictNonStrict<Table>(
-        total_charge, capacity, need_evict_for_occupancy, eec_and_scl, state);
+        total_charge, capacity, need_evict_for_occupancy, state);
     if (!success) {
       // Revert occupancy
       occupancy_.FetchSubRelaxed(1);
@@ -718,7 +688,8 @@ void BaseClockTable::TEST_ReleaseNMinus1(ClockHandle* h, size_t n) {
 #endif
 
 FixedHyperClockTable::FixedHyperClockTable(
-    size_t capacity, CacheMetadataChargePolicy metadata_charge_policy,
+    size_t capacity, bool /*strict_capacity_limit*/,
+    CacheMetadataChargePolicy metadata_charge_policy,
     MemoryAllocator* allocator,
     const Cache::EvictionCallback* eviction_callback, const uint32_t* hash_seed,
     const Opts& opts)
@@ -1113,8 +1084,7 @@ inline void FixedHyperClockTable::ReclaimEntryUsage(size_t total_charge) {
 }
 
 inline void FixedHyperClockTable::Evict(size_t requested_charge, InsertState&,
-                                        EvictionData* data,
-                                        uint32_t eviction_effort_cap) {
+                                        EvictionData* data) {
   // precondition
   assert(requested_charge > 0);
 
@@ -1135,10 +1105,10 @@ inline void FixedHyperClockTable::Evict(size_t requested_charge, InsertState&,
   for (;;) {
     for (size_t i = 0; i < step_size; i++) {
       HandleImpl& h = array_[ModTableSize(Lower32of64(old_clock_pointer + i))];
-      bool evicting = ClockUpdate(h, data);
+      bool evicting = ClockUpdate(h);
       if (evicting) {
         Rollback(h.hashed_key, &h);
-        TrackAndReleaseEvictedEntry(&h);
+        TrackAndReleaseEvictedEntry(&h, data);
       }
     }
 
@@ -1149,10 +1119,6 @@ inline void FixedHyperClockTable::Evict(size_t requested_charge, InsertState&,
     if (old_clock_pointer >= max_clock_pointer) {
       return;
     }
-    if (IsEvictionEffortExceeded(*data, eviction_effort_cap)) {
-      eviction_effort_exceeded_count_.FetchAddRelaxed(1);
-      return;
-    }
 
     // Advance clock pointer (concurrently)
     old_clock_pointer = clock_pointer_.FetchAddRelaxed(step_size);
@@ -1167,11 +1133,10 @@ ClockCacheShard<Table>::ClockCacheShard(
     const Cache::EvictionCallback* eviction_callback, const uint32_t* hash_seed,
     const typename Table::Opts& opts)
     : CacheShardBase(metadata_charge_policy),
-      table_(capacity, metadata_charge_policy, allocator, eviction_callback,
-             hash_seed, opts),
+      table_(capacity, strict_capacity_limit, metadata_charge_policy, allocator,
+             eviction_callback, hash_seed, opts),
       capacity_(capacity),
-      eec_and_scl_(SanitizeEncodeEecAndScl(opts.eviction_effort_cap,
-                                           strict_capacity_limit)) {
+      strict_capacity_limit_(strict_capacity_limit) {
   // Initial charge metadata should not exceed capacity
   assert(table_.GetUsage() <= capacity_.LoadRelaxed() ||
          capacity_.LoadRelaxed() < sizeof(HandleImpl));
@@ -1247,11 +1212,7 @@ void ClockCacheShard<Table>::SetCapacity(size_t capacity) {
 template <class Table>
 void ClockCacheShard<Table>::SetStrictCapacityLimit(
     bool strict_capacity_limit) {
-  if (strict_capacity_limit) {
-    eec_and_scl_.FetchOrRelaxed(kStrictCapacityLimitBit);
-  } else {
-    eec_and_scl_.FetchAndRelaxed(~kStrictCapacityLimitBit);
-  }
+  strict_capacity_limit_.StoreRelaxed(strict_capacity_limit);
   // next Insert will take care of any necessary evictions
 }
 
@@ -1273,7 +1234,7 @@ Status ClockCacheShard<Table>::Insert(const Slice& key,
   proto.total_charge = charge;
   return table_.template Insert<Table>(proto, handle, priority,
                                        capacity_.LoadRelaxed(),
-                                       eec_and_scl_.LoadRelaxed());
+                                       strict_capacity_limit_.LoadRelaxed());
 }
 
 template <class Table>
@@ -1288,9 +1249,9 @@ typename Table::HandleImpl* ClockCacheShard<Table>::CreateStandalone(
   proto.value = obj;
   proto.helper = helper;
   proto.total_charge = charge;
-  return table_.template CreateStandalone<Table>(proto, capacity_.LoadRelaxed(),
-                                                 eec_and_scl_.LoadRelaxed(),
-                                                 allow_uncharged);
+  return table_.template CreateStandalone<Table>(
+      proto, capacity_.LoadRelaxed(), strict_capacity_limit_.LoadRelaxed(),
+      allow_uncharged);
 }
 
 template <class Table>
@@ -1428,19 +1389,19 @@ BaseHyperClockCache<Table>::BaseHyperClockCache(
 
 template <class Table>
 Cache::ObjectPtr BaseHyperClockCache<Table>::Value(Handle* handle) {
-  return static_cast<const typename Table::HandleImpl*>(handle)->value;
+  return reinterpret_cast<const typename Table::HandleImpl*>(handle)->value;
 }
 
 template <class Table>
 size_t BaseHyperClockCache<Table>::GetCharge(Handle* handle) const {
-  return static_cast<const typename Table::HandleImpl*>(handle)
+  return reinterpret_cast<const typename Table::HandleImpl*>(handle)
       ->GetTotalCharge();
 }
 
 template <class Table>
 const Cache::CacheItemHelper* BaseHyperClockCache<Table>::GetCacheItemHelper(
     Handle* handle) const {
-  auto h = static_cast<const typename Table::HandleImpl*>(handle);
+  auto h = reinterpret_cast<const typename Table::HandleImpl*>(handle);
   return h->helper;
 }
 
@@ -1542,20 +1503,14 @@ void BaseHyperClockCache<Table>::ReportProblems(
     const std::shared_ptr<Logger>& info_log) const {
   if (info_log->GetInfoLogLevel() <= InfoLogLevel::DEBUG_LEVEL) {
     LoadVarianceStats slot_stats;
-    uint64_t eviction_effort_exceeded_count = 0;
     this->ForEachShard([&](const BaseHyperClockCache<Table>::Shard* shard) {
       size_t count = shard->GetTableAddressCount();
       for (size_t i = 0; i < count; ++i) {
         slot_stats.Add(IsSlotOccupied(*shard->GetTable().HandlePtr(i)));
       }
-      eviction_effort_exceeded_count +=
-          shard->GetTable().GetEvictionEffortExceededCount();
     });
     ROCKS_LOG_AT_LEVEL(info_log, InfoLogLevel::DEBUG_LEVEL,
                        "Slot occupancy stats: %s", slot_stats.Report().c_str());
-    ROCKS_LOG_AT_LEVEL(info_log, InfoLogLevel::DEBUG_LEVEL,
-                       "Eviction effort exceeded: %" PRIu64,
-                       eviction_effort_exceeded_count);
   }
 }
 
@@ -1953,7 +1908,8 @@ class AutoHyperClockTable::ChainRewriteLock {
 };
 
 AutoHyperClockTable::AutoHyperClockTable(
-    size_t capacity, CacheMetadataChargePolicy metadata_charge_policy,
+    size_t capacity, bool /*strict_capacity_limit*/,
+    CacheMetadataChargePolicy metadata_charge_policy,
     MemoryAllocator* allocator,
     const Cache::EvictionCallback* eviction_callback, const uint32_t* hash_seed,
     const Opts& opts)
@@ -2634,8 +2590,7 @@ using ClockUpdateChainLockedOpData =
 template <class OpData>
 void AutoHyperClockTable::PurgeImplLocked(OpData* op_data,
                                           ChainRewriteLock& rewrite_lock,
-                                          size_t home,
-                                          BaseClockTable::EvictionData* data) {
+                                          size_t home) {
   constexpr bool kIsPurge = std::is_same_v<OpData, PurgeLockedOpData>;
   constexpr bool kIsClockUpdateChain =
       std::is_same_v<OpData, ClockUpdateChainLockedOpData>;
@@ -2677,7 +2632,7 @@ void AutoHyperClockTable::PurgeImplLocked(OpData* op_data,
       assert(home == BottomNBits(h->hashed_key[1], home_shift));
       if constexpr (kIsClockUpdateChain) {
         // Clock update and/or check for purgeable (under (de)construction)
-        if (ClockUpdate(*h, data, &purgeable)) {
+        if (ClockUpdate(*h, &purgeable)) {
           // Remember for finishing eviction
           op_data->push_back(h);
           // Entries for eviction become purgeable
@@ -2687,7 +2642,6 @@ void AutoHyperClockTable::PurgeImplLocked(OpData* op_data,
         }
       } else {
         (void)op_data;
-        (void)data;
         purgeable = ((h->meta.Load() >> ClockHandle::kStateShift) &
                      ClockHandle::kStateShareableBit) == 0;
       }
@@ -2765,8 +2719,7 @@ using PurgeOpData = const UniqueId64x2;
 using ClockUpdateChainOpData = ClockUpdateChainLockedOpData;
 
 template <class OpData>
-void AutoHyperClockTable::PurgeImpl(OpData* op_data, size_t home,
-                                    BaseClockTable::EvictionData* data) {
+void AutoHyperClockTable::PurgeImpl(OpData* op_data, size_t home) {
   // Early efforts to make AutoHCC fully wait-free ran into too many problems
   // that needed obscure and potentially inefficient work-arounds to have a
   // chance at working.
@@ -2847,9 +2800,9 @@ void AutoHyperClockTable::PurgeImpl(OpData* op_data, size_t home,
   if (!rewrite_lock.IsEnd()) {
     if constexpr (kIsPurge) {
       PurgeLockedOpData* locked_op_data{};
-      PurgeImplLocked(locked_op_data, rewrite_lock, home, data);
+      PurgeImplLocked(locked_op_data, rewrite_lock, home);
     } else {
-      PurgeImplLocked(op_data, rewrite_lock, home, data);
+      PurgeImplLocked(op_data, rewrite_lock, home);
     }
   }
 }
@@ -3452,8 +3405,7 @@ void AutoHyperClockTable::EraseUnRefEntries() {
 }
 
 void AutoHyperClockTable::Evict(size_t requested_charge, InsertState& state,
-                                EvictionData* data,
-                                uint32_t eviction_effort_cap) {
+                                EvictionData* data) {
   // precondition
   assert(requested_charge > 0);
 
@@ -3511,12 +3463,12 @@ void AutoHyperClockTable::Evict(size_t requested_charge, InsertState& state,
         if (home >= used_length) {
           break;
         }
-        PurgeImpl(&to_finish_eviction, home, data);
+        PurgeImpl(&to_finish_eviction, home);
       }
     }
 
     for (HandleImpl* h : to_finish_eviction) {
-      TrackAndReleaseEvictedEntry(h);
+      TrackAndReleaseEvictedEntry(h, data);
       // NOTE: setting likely_empty_slot here can cause us to reduce the
       // portion of "at home" entries, probably because an evicted entry
       // is more likely to come back than a random new entry and would be
@@ -3544,11 +3496,6 @@ void AutoHyperClockTable::Evict(size_t requested_charge, InsertState& state,
     if (old_clock_pointer + step_size >= max_clock_pointer) {
       return;
     }
-
-    if (IsEvictionEffortExceeded(*data, eviction_effort_cap)) {
-      eviction_effort_exceeded_count_.FetchAddRelaxed(1);
-      return;
-    }
   }
 }
 

cache/clock_cache.h

@@ -11,7 +11,6 @@
 
 #include <array>
 #include <atomic>
-#include <climits>
 #include <cstddef>
 #include <cstdint>
 #include <memory>
@@ -298,7 +297,7 @@ 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
@@ -375,14 +374,6 @@ struct ClockHandle : public ClockHandleBasicData {
 
 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,
@@ -395,13 +386,13 @@ 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);
 
@@ -415,17 +406,12 @@ class BaseClockTable {
 
   uint64_t GetYieldCount() const { return yield_count_.LoadRelaxed(); }
 
-  uint64_t GetEvictionEffortExceededCount() const {
-    return eviction_effort_exceeded_count_.LoadRelaxed();
-  }
-
   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 +436,6 @@ 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 +449,6 @@ 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
@@ -477,15 +461,9 @@ class BaseClockTable {
   RelaxedAtomic<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_{};
-
   // TODO: is this separation needed if we don't do background evictions?
   ALIGN_AS(CACHE_LINE_SIZE)
   // Number of elements in the table.
@@ -539,19 +517,17 @@ 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 +549,7 @@ 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);
 
@@ -828,20 +803,18 @@ class AutoHyperClockTable : public BaseClockTable {
     }
   };  // 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 +841,7 @@ 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 +906,7 @@ 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 +916,7 @@ 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
@@ -1107,10 +1078,9 @@ class ALIGN_AS(CACHE_LINE_SIZE) ClockCacheShard final : public CacheShardBase {
   // (Relaxed: eventual consistency/update is OK)
   RelaxedAtomic<size_t> capacity_;
 
-  // Encodes eviction_effort_cap (bottom 31 bits) and strict_capacity_limit
-  // (top bit). See HyperClockCacheOptions::eviction_effort_cap etc.
+  // Whether to reject insertion if cache reaches its full capacity.
   // (Relaxed: eventual consistency/update is OK)
-  RelaxedAtomic<uint32_t> eec_and_scl_;
+  RelaxedAtomic<bool> strict_capacity_limit_;
 };  // class ClockCacheShard
 
 template <class Table>

cache/compressed_secondary_cache.cc

@@ -26,12 +26,12 @@ CompressedSecondaryCache::CompressedSecondaryCache(
               cache_))),
       disable_cache_(opts.capacity == 0) {}
 
-CompressedSecondaryCache::~CompressedSecondaryCache() = default;
+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.
@@ -51,7 +51,6 @@ 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 +137,6 @@ 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 +190,13 @@ 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 +289,6 @@ 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);
@@ -384,7 +377,7 @@ const Cache::CacheItemHelper* CompressedSecondaryCache::GetHelper(
             chunks_head = chunks_head->next;
             tmp_chunk->Free();
             obj = nullptr;
-          }
+          };
         }};
     return &kHelper;
   } else {

cache/compressed_secondary_cache.h

@@ -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; }
 

cache/compressed_secondary_cache_test.cc

@@ -33,7 +33,7 @@ const std::string key3 = "____    ____key3";
 class CompressedSecondaryCacheTestBase : public testing::Test,
                                          public WithCacheType {
  public:
-  CompressedSecondaryCacheTestBase() = default;
+  CompressedSecondaryCacheTestBase() {}
   ~CompressedSecondaryCacheTestBase() override = default;
 
  protected:
@@ -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
@@ -1235,7 +1233,7 @@ TEST_P(CompressedSecCacheTestWithTiered, DynamicUpdate) {
 
 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));
@@ -1329,7 +1327,7 @@ TEST_P(CompressedSecCacheTestWithTiered, DynamicUpdateWithReservation) {
 
 TEST_P(CompressedSecCacheTestWithTiered, ReservationOverCapacity) {
   CompressedSecondaryCache* sec_cache =
-      static_cast<CompressedSecondaryCache*>(GetSecondaryCache());
+      reinterpret_cast<CompressedSecondaryCache*>(GetSecondaryCache());
   std::shared_ptr<Cache> tiered_cache = GetTieredCache();
 
   ASSERT_OK(cache_res_mgr()->UpdateCacheReservation(110 << 20));

cache/lru_cache.cc

@@ -277,8 +277,8 @@ void LRUCacheShard::LRU_Insert(LRUHandle* e) {
     e->SetInHighPriPool(false);
     e->SetInLowPriPool(true);
     low_pri_pool_usage_ += e->total_charge;
-    lru_low_pri_ = e;
     MaintainPoolSize();
+    lru_low_pri_ = e;
   } else {
     // Insert "e" to the head of bottom-pri pool.
     e->next = lru_bottom_pri_->next;
@@ -301,7 +301,6 @@ void LRUCacheShard::MaintainPoolSize() {
     // Overflow last entry in high-pri pool to low-pri pool.
     lru_low_pri_ = lru_low_pri_->next;
     assert(lru_low_pri_ != &lru_);
-    assert(lru_low_pri_->InHighPriPool());
     lru_low_pri_->SetInHighPriPool(false);
     lru_low_pri_->SetInLowPriPool(true);
     assert(high_pri_pool_usage_ >= lru_low_pri_->total_charge);
@@ -313,7 +312,6 @@ void LRUCacheShard::MaintainPoolSize() {
     // Overflow last entry in low-pri pool to bottom-pri pool.
     lru_bottom_pri_ = lru_bottom_pri_->next;
     assert(lru_bottom_pri_ != &lru_);
-    assert(lru_bottom_pri_->InLowPriPool());
     lru_bottom_pri_->SetInHighPriPool(false);
     lru_bottom_pri_->SetInLowPriPool(false);
     assert(low_pri_pool_usage_ >= lru_bottom_pri_->total_charge);
@@ -341,7 +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);
@@ -507,7 +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);
@@ -662,18 +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;
 }
 

cache/lru_cache.h

@@ -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;

cache/lru_cache_test.cc

@@ -32,7 +32,7 @@ namespace ROCKSDB_NAMESPACE {
 
 class LRUCacheTest : public testing::Test {
  public:
-  LRUCacheTest() = default;
+  LRUCacheTest() {}
   ~LRUCacheTest() override { DeleteCache(); }
 
   void DeleteCache() {
@@ -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,
@@ -57,11 +57,10 @@ class LRUCacheTest : public testing::Test {
   }
 
   void Insert(const std::string& key,
-              Cache::Priority priority = Cache::Priority::LOW,
-              size_t charge = 1) {
+              Cache::Priority priority = Cache::Priority::LOW) {
     EXPECT_OK(cache_->Insert(key, 0 /*hash*/, nullptr /*value*/,
-                             &kNoopCacheItemHelper, charge, nullptr /*handle*/,
-                             priority));
+                             &kNoopCacheItemHelper, 1 /*charge*/,
+                             nullptr /*handle*/, priority));
   }
 
   void Insert(char key, Cache::Priority priority = Cache::Priority::LOW) {
@@ -145,10 +144,8 @@ class LRUCacheTest : public testing::Test {
     ASSERT_EQ(num_bottom_pri_pool_keys, bottom_pri_pool_keys);
   }
 
- protected:
-  LRUCacheShard* cache_ = nullptr;
-
  private:
+  LRUCacheShard* cache_ = nullptr;
   Cache::EvictionCallback eviction_callback_;
 };
 
@@ -381,7 +378,7 @@ class ClockCacheTest : public testing::Test {
   using Table = typename Shard::Table;
   using TableOpts = typename Table::Opts;
 
-  ClockCacheTest() = default;
+  ClockCacheTest() {}
   ~ClockCacheTest() override { DeleteShard(); }
 
   void DeleteShard() {
@@ -392,12 +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);
@@ -448,20 +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:
@@ -694,53 +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;
@@ -1093,8 +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);
 
@@ -1979,7 +1920,7 @@ TEST_P(BasicSecondaryCacheTest, BasicWaitAllTest) {
     ah.priority = Cache::Priority::LOW;
     cache->StartAsyncLookup(ah);
   }
-  cache->WaitAll(async_handles.data(), async_handles.size());
+  cache->WaitAll(&async_handles[0], async_handles.size());
   for (size_t i = 0; i < async_handles.size(); ++i) {
     SCOPED_TRACE("i = " + std::to_string(i));
     Cache::Handle* result = async_handles[i].Result();
@@ -2706,23 +2647,6 @@ TEST_P(DBSecondaryCacheTest, TestSecondaryCacheOptionTwoDB) {
   ASSERT_OK(DestroyDB(dbname2, options));
 }
 
-TEST_F(LRUCacheTest, InsertAfterReducingCapacity) {
-  // Fix a bug in LRU cache where it may try to remove a low pri entry's
-  // charge from high pri pool. It causes
-  // Assertion failed: (high_pri_pool_usage_ >= lru_low_pri_->total_charge),
-  // function MaintainPoolSize, file lru_cache.cc
-  NewCache(/*capacity=*/10, /*high_pri_pool_ratio=*/0.2,
-           /*low_pri_pool_ratio=*/0.8);
-  // high pri pool size and usage are both 2
-  Insert("x", Cache::Priority::HIGH);
-  Insert("y", Cache::Priority::HIGH);
-  cache_->SetCapacity(5);
-  // high_pri_pool_size is 1, the next time we try to maintain pool size,
-  // we will move entries from high pri pool to low pri pool
-  // The bug was deducting this entry's charge from high pri pool usage.
-  Insert("aaa", Cache::Priority::LOW, /*charge=*/3);
-}
-
 }  // namespace ROCKSDB_NAMESPACE
 
 int main(int argc, char** argv) {

cache/secondary_cache_adapter.cc

@@ -134,14 +134,12 @@ bool CacheWithSecondaryAdapter::EvictionHandler(const Slice& key,
     auto obj = target_->Value(handle);
     // Ignore dummy entry
     if (obj != kDummyObj) {
-      bool force = false;
+      bool hit = false;
       if (adm_policy_ == TieredAdmissionPolicy::kAdmPolicyAllowCacheHits) {
-        force = was_hit;
-      } else if (adm_policy_ == TieredAdmissionPolicy::kAdmPolicyAllowAll) {
-        force = true;
+        hit = was_hit;
       }
       // Spill into secondary cache.
-      secondary_cache_->Insert(key, obj, helper, force).PermitUncheckedError();
+      secondary_cache_->Insert(key, obj, helper, hit).PermitUncheckedError();
     }
   }
   // Never takes ownership of obj
@@ -296,8 +294,7 @@ 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);
@@ -351,10 +348,10 @@ 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();
@@ -663,7 +660,6 @@ std::shared_ptr<Cache> NewTieredCache(const TieredCacheOptions& _opts) {
         break;
       case TieredAdmissionPolicy::kAdmPolicyPlaceholder:
       case TieredAdmissionPolicy::kAdmPolicyAllowCacheHits:
-      case TieredAdmissionPolicy::kAdmPolicyAllowAll:
         if (opts.nvm_sec_cache) {
           valid_adm_policy = false;
         }

cache/sharded_cache.h

@@ -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(); });
   }
 

cache/tiered_secondary_cache.cc

@@ -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 {

cache/tiered_secondary_cache.h

@@ -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 {

cache/tiered_secondary_cache_test.cc

@@ -62,7 +62,7 @@ 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);
 
@@ -386,7 +386,7 @@ TEST_F(DBTieredSecondaryCacheTest, BasicMultiGetTest) {
   keys.push_back(Key(8));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 3u);
@@ -400,7 +400,7 @@ TEST_F(DBTieredSecondaryCacheTest, BasicMultiGetTest) {
   keys.push_back(Key(20));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
@@ -414,7 +414,7 @@ TEST_F(DBTieredSecondaryCacheTest, BasicMultiGetTest) {
   keys.push_back(Key(8));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
@@ -428,7 +428,7 @@ TEST_F(DBTieredSecondaryCacheTest, BasicMultiGetTest) {
   keys.push_back(Key(8));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
@@ -442,7 +442,7 @@ TEST_F(DBTieredSecondaryCacheTest, BasicMultiGetTest) {
   keys.push_back(Key(8));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
@@ -456,7 +456,7 @@ TEST_F(DBTieredSecondaryCacheTest, BasicMultiGetTest) {
   keys.push_back(Key(20));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
@@ -470,7 +470,7 @@ TEST_F(DBTieredSecondaryCacheTest, BasicMultiGetTest) {
   keys.push_back(Key(20));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
@@ -484,7 +484,7 @@ TEST_F(DBTieredSecondaryCacheTest, BasicMultiGetTest) {
   keys.push_back(Key(20));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
@@ -528,7 +528,7 @@ TEST_F(DBTieredSecondaryCacheTest, WaitAllTest) {
   keys.push_back(Key(8));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 3u);
@@ -542,7 +542,7 @@ TEST_F(DBTieredSecondaryCacheTest, WaitAllTest) {
   keys.push_back(Key(20));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
@@ -561,7 +561,7 @@ TEST_F(DBTieredSecondaryCacheTest, WaitAllTest) {
   keys.push_back(Key(36));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 10u);
@@ -582,7 +582,7 @@ TEST_F(DBTieredSecondaryCacheTest, WaitAllTest) {
   keys.push_back(Key(8));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 10u);
@@ -629,7 +629,7 @@ TEST_F(DBTieredSecondaryCacheTest, ReadyBeforeWaitAllTest) {
   keys.push_back(Key(8));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 3u);
@@ -644,7 +644,7 @@ TEST_F(DBTieredSecondaryCacheTest, ReadyBeforeWaitAllTest) {
   keys.push_back(Key(20));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
@@ -659,7 +659,7 @@ TEST_F(DBTieredSecondaryCacheTest, ReadyBeforeWaitAllTest) {
   keys.push_back(Key(8));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 6u);
@@ -676,7 +676,7 @@ TEST_F(DBTieredSecondaryCacheTest, ReadyBeforeWaitAllTest) {
   keys.push_back(Key(36));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 8u);
@@ -691,7 +691,7 @@ TEST_F(DBTieredSecondaryCacheTest, ReadyBeforeWaitAllTest) {
   keys.push_back(Key(36));
   values = MultiGet(keys, /*snapshot=*/nullptr, /*async=*/true);
   ASSERT_EQ(values.size(), keys.size());
-  for (const auto& value : values) {
+  for (auto value : values) {
     ASSERT_EQ(1007, value.size());
   }
   ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 8u);
@@ -816,174 +816,11 @@ TEST_P(DBTieredAdmPolicyTest, CompressedOnlyTest) {
   Destroy(options);
 }
 
-TEST_P(DBTieredAdmPolicyTest, CompressedCacheAdmission) {
-  if (!LZ4_Supported()) {
-    ROCKSDB_GTEST_SKIP("This test requires LZ4 support.");
-    return;
-  }
-
-  BlockBasedTableOptions table_options;
-  // We want a block cache of size 5KB, and a compressed secondary cache of
-  // size 5KB. However, we specify a block cache size of 256KB here in order
-  // to take into account the cache reservation in the block cache on
-  // behalf of the compressed cache. The unit of cache reservation is 256KB.
-  // The effective block cache capacity will be calculated as 256 + 5 = 261KB,
-  // and 256KB will be reserved for the compressed cache, leaving 10KB for
-  // the primary block cache. We only have to worry about this here because
-  // the cache size is so small.
-  table_options.block_cache = NewCache(256 * 1024, 5 * 1024, 0, GetParam());
-  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));
-
-  size_t comp_cache_usage = compressed_secondary_cache()->TEST_GetUsage();
-  // Disable paranoid_file_checks so that flush will not read back the newly
-  // written file
-  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());
-
-  // The second Get (for 5) will evict the data block loaded by the first
-  // Get, which will be admitted into the compressed secondary cache only
-  // for the kAdmPolicyAllowAll policy
-  std::string v = Get(Key(0));
-  ASSERT_EQ(1007, v.size());
-
-  v = Get(Key(5));
-  ASSERT_EQ(1007, v.size());
-
-  if (GetParam() == TieredAdmissionPolicy::kAdmPolicyAllowAll) {
-    ASSERT_GT(compressed_secondary_cache()->TEST_GetUsage(),
-              comp_cache_usage + 128);
-  } else {
-    ASSERT_LT(compressed_secondary_cache()->TEST_GetUsage(),
-              comp_cache_usage + 128);
-  }
-
-  Destroy(options);
-}
-
-TEST_F(DBTieredSecondaryCacheTest, FSBufferTest) {
-  class WrapFS : public FileSystemWrapper {
-   public:
-    explicit WrapFS(const std::shared_ptr<FileSystem>& _target)
-        : FileSystemWrapper(_target) {}
-    ~WrapFS() override {}
-    const char* Name() const override { return "WrapFS"; }
-
-    IOStatus NewRandomAccessFile(const std::string& fname,
-                                 const FileOptions& opts,
-                                 std::unique_ptr<FSRandomAccessFile>* result,
-                                 IODebugContext* dbg) override {
-      class WrappedRandomAccessFile : public FSRandomAccessFileOwnerWrapper {
-       public:
-        explicit WrappedRandomAccessFile(
-            std::unique_ptr<FSRandomAccessFile>& file)
-            : FSRandomAccessFileOwnerWrapper(std::move(file)) {}
-
-        IOStatus MultiRead(FSReadRequest* reqs, size_t num_reqs,
-                           const IOOptions& options,
-                           IODebugContext* dbg) override {
-          for (size_t i = 0; i < num_reqs; ++i) {
-            FSReadRequest& req = reqs[i];
-            FSAllocationPtr buffer(new char[req.len], [](void* ptr) {
-              delete[] static_cast<char*>(ptr);
-            });
-            req.fs_scratch = std::move(buffer);
-            req.status = Read(req.offset, req.len, options, &req.result,
-                              static_cast<char*>(req.fs_scratch.get()), dbg);
-          }
-          return IOStatus::OK();
-        }
-      };
-
-      std::unique_ptr<FSRandomAccessFile> file;
-      IOStatus s = target()->NewRandomAccessFile(fname, opts, &file, dbg);
-      EXPECT_OK(s);
-      result->reset(new WrappedRandomAccessFile(file));
-
-      return s;
-    }
-
-    void SupportedOps(int64_t& supported_ops) override {
-      supported_ops = 1 << FSSupportedOps::kAsyncIO;
-      supported_ops |= 1 << FSSupportedOps::kFSBuffer;
-    }
-  };
-
-  if (!LZ4_Supported()) {
-    ROCKSDB_GTEST_SKIP("This test requires LZ4 support.");
-    return;
-  }
-
-  std::shared_ptr<WrapFS> wrap_fs =
-      std::make_shared<WrapFS>(env_->GetFileSystem());
-  std::unique_ptr<Env> wrap_env(new CompositeEnvWrapper(env_, wrap_fs));
-  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.env = wrap_env.get();
-
-  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 (const 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);
-
-  std::string v = Get(Key(12));
-  ASSERT_EQ(1007, v.size());
-  ASSERT_EQ(nvm_sec_cache()->num_insert_saved(), 4u);
-  ASSERT_EQ(nvm_sec_cache()->num_misses(), 4u);
-  ASSERT_EQ(options.statistics->getTickerCount(BLOCK_CACHE_MISS), 4u);
-
-  Close();
-  Destroy(options);
-}
-
 INSTANTIATE_TEST_CASE_P(
     DBTieredAdmPolicyTest, DBTieredAdmPolicyTest,
     ::testing::Values(TieredAdmissionPolicy::kAdmPolicyAuto,
                       TieredAdmissionPolicy::kAdmPolicyPlaceholder,
-                      TieredAdmissionPolicy::kAdmPolicyAllowCacheHits,
-                      TieredAdmissionPolicy::kAdmPolicyAllowAll));
+                      TieredAdmissionPolicy::kAdmPolicyAllowCacheHits));
 
 }  // namespace ROCKSDB_NAMESPACE
 

cache/typed_cache.h

@@ -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,

cmake/modules/Findzstd.cmake

@@ -1,29 +1,29 @@
 # - Find zstd
 # Find the zstd compression library and includes
 #
-# ZSTD_INCLUDE_DIRS - where to find zstd.h, etc.
-# ZSTD_LIBRARIES - List of libraries when using zstd.
-# ZSTD_FOUND - True if zstd found.
+# zstd_INCLUDE_DIRS - where to find zstd.h, etc.
+# zstd_LIBRARIES - List of libraries when using zstd.
+# zstd_FOUND - True if zstd found.
 
-find_path(ZSTD_INCLUDE_DIRS
+find_path(zstd_INCLUDE_DIRS
   NAMES zstd.h
   HINTS ${zstd_ROOT_DIR}/include)
 
-find_library(ZSTD_LIBRARIES
+find_library(zstd_LIBRARIES
   NAMES zstd
   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
-  ZSTD_INCLUDE_DIRS)
+  zstd_LIBRARIES
+  zstd_INCLUDE_DIRS)
 
-if(ZSTD_FOUND AND NOT (TARGET zstd::zstd))
+if(zstd_FOUND AND NOT (TARGET zstd::zstd))
   add_library (zstd::zstd UNKNOWN IMPORTED)
   set_target_properties(zstd::zstd
     PROPERTIES
-      IMPORTED_LOCATION ${ZSTD_LIBRARIES}
-      INTERFACE_INCLUDE_DIRECTORIES ${ZSTD_INCLUDE_DIRS})
+      IMPORTED_LOCATION ${zstd_LIBRARIES}
+      INTERFACE_INCLUDE_DIRECTORIES ${zstd_INCLUDE_DIRS})
 endif()

db/arena_wrapped_db_iter.cc

@@ -43,13 +43,14 @@ void ArenaWrappedDBIter::Init(
     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_iteration,
-    uint64_t version_number, ReadCallback* read_callback,
-    ColumnFamilyHandleImpl* cfh, bool expose_blob_index, bool allow_refresh) {
+    uint64_t version_number, ReadCallback* read_callback, DBImpl* db_impl,
+    ColumnFamilyData* cfd, bool expose_blob_index, bool allow_refresh) {
   auto mem = arena_.AllocateAligned(sizeof(DBIter));
-  db_iter_ = new (mem) DBIter(
-      env, read_options, ioptions, mutable_cf_options, ioptions.user_comparator,
-      /* iter */ nullptr, version, sequence, true,
-      max_sequential_skip_in_iteration, read_callback, cfh, expose_blob_index);
+  db_iter_ =
+      new (mem) DBIter(env, read_options, ioptions, mutable_cf_options,
+                       ioptions.user_comparator, /* iter */ nullptr, version,
+                       sequence, true, max_sequential_skip_in_iteration,
+                       read_callback, db_impl, cfd, expose_blob_index);
   sv_number_ = version_number;
   read_options_ = read_options;
   allow_refresh_ = allow_refresh;
@@ -64,44 +65,40 @@ void ArenaWrappedDBIter::Init(
 Status ArenaWrappedDBIter::Refresh() { return Refresh(nullptr); }
 
 Status ArenaWrappedDBIter::Refresh(const Snapshot* snapshot) {
-  if (cfh_ == nullptr || !allow_refresh_) {
+  if (cfd_ == nullptr || db_impl_ == nullptr || !allow_refresh_) {
     return Status::NotSupported("Creating renew iterator is not allowed.");
   }
   assert(db_iter_ != nullptr);
-  auto cfd = cfh_->cfd();
-  auto db_impl = cfh_->db();
-
   // TODO(yiwu): For last_seq_same_as_publish_seq_==false, this is not the
   // correct behavior. Will be corrected automatically when we take a snapshot
   // here for the case of WritePreparedTxnDB.
-  uint64_t cur_sv_number = cfd->GetSuperVersionNumber();
+  uint64_t cur_sv_number = cfd_->GetSuperVersionNumber();
   // If we recreate a new internal iterator below (NewInternalIterator()),
   // we will pass in read_options_. We need to make sure it
   // has the right snapshot.
   read_options_.snapshot = snapshot;
   TEST_SYNC_POINT("ArenaWrappedDBIter::Refresh:1");
   TEST_SYNC_POINT("ArenaWrappedDBIter::Refresh:2");
-
   auto reinit_internal_iter = [&]() {
     Env* env = db_iter_->env();
     db_iter_->~DBIter();
     arena_.~Arena();
     new (&arena_) Arena();
 
-    SuperVersion* sv = cfd->GetReferencedSuperVersion(db_impl);
+    SuperVersion* sv = cfd_->GetReferencedSuperVersion(db_impl_);
     assert(sv->version_number >= cur_sv_number);
-    SequenceNumber read_seq = GetSeqNum(db_impl, snapshot);
+    SequenceNumber read_seq = GetSeqNum(db_impl_, snapshot);
     if (read_callback_) {
       read_callback_->Refresh(read_seq);
     }
-    Init(env, read_options_, *(cfd->ioptions()), sv->mutable_cf_options,
+    Init(env, read_options_, *(cfd_->ioptions()), sv->mutable_cf_options,
          sv->current, read_seq,
          sv->mutable_cf_options.max_sequential_skip_in_iterations,
-         sv->version_number, read_callback_, cfh_, expose_blob_index_,
+         sv->version_number, read_callback_, db_impl_, cfd_, expose_blob_index_,
          allow_refresh_);
 
-    InternalIterator* internal_iter = db_impl->NewInternalIterator(
-        read_options_, cfd, sv, &arena_, read_seq,
+    InternalIterator* internal_iter = db_impl_->NewInternalIterator(
+        read_options_, cfd_, sv, &arena_, read_seq,
         /* allow_unprepared_value */ true, /* db_iter */ this);
     SetIterUnderDBIter(internal_iter);
   };
@@ -110,10 +107,10 @@ Status ArenaWrappedDBIter::Refresh(const Snapshot* snapshot) {
       reinit_internal_iter();
       break;
     } else {
-      SequenceNumber read_seq = GetSeqNum(db_impl, snapshot);
+      SequenceNumber read_seq = GetSeqNum(db_impl_, snapshot);
       // Refresh range-tombstones in MemTable
       if (!read_options_.ignore_range_deletions) {
-        SuperVersion* sv = cfd->GetThreadLocalSuperVersion(db_impl);
+        SuperVersion* sv = cfd_->GetThreadLocalSuperVersion(db_impl_);
         TEST_SYNC_POINT_CALLBACK("ArenaWrappedDBIter::Refresh:SV", nullptr);
         auto t = sv->mem->NewRangeTombstoneIterator(
             read_options_, read_seq, false /* immutable_memtable */);
@@ -126,13 +123,13 @@ Status ArenaWrappedDBIter::Refresh(const Snapshot* snapshot) {
           // will be freed during db_iter destruction there.
           if (memtable_range_tombstone_iter_) {
             assert(!*memtable_range_tombstone_iter_ ||
-                   sv_number_ != cfd->GetSuperVersionNumber());
+                   sv_number_ != cfd_->GetSuperVersionNumber());
           }
           delete t;
         } else {  // current mutable memtable has range tombstones
           if (!memtable_range_tombstone_iter_) {
             delete t;
-            db_impl->ReturnAndCleanupSuperVersion(cfd, sv);
+            db_impl_->ReturnAndCleanupSuperVersion(cfd_, sv);
             // The memtable under DBIter did not have range tombstone before
             // refresh.
             reinit_internal_iter();
@@ -141,13 +138,13 @@ Status ArenaWrappedDBIter::Refresh(const Snapshot* snapshot) {
             delete *memtable_range_tombstone_iter_;
             *memtable_range_tombstone_iter_ = new TruncatedRangeDelIterator(
                 std::unique_ptr<FragmentedRangeTombstoneIterator>(t),
-                &cfd->internal_comparator(), nullptr, nullptr);
+                &cfd_->internal_comparator(), nullptr, nullptr);
           }
         }
-        db_impl->ReturnAndCleanupSuperVersion(cfd, sv);
+        db_impl_->ReturnAndCleanupSuperVersion(cfd_, sv);
       }
       // Check again if the latest super version number is changed
-      uint64_t latest_sv_number = cfd->GetSuperVersionNumber();
+      uint64_t latest_sv_number = cfd_->GetSuperVersionNumber();
       if (latest_sv_number != cur_sv_number) {
         // If the super version number is changed after refreshing,
         // fallback to Re-Init the InternalIterator
@@ -166,14 +163,14 @@ 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,
-    uint64_t version_number, ReadCallback* read_callback,
-    ColumnFamilyHandleImpl* cfh, bool expose_blob_index, bool allow_refresh) {
+    uint64_t version_number, ReadCallback* read_callback, DBImpl* db_impl,
+    ColumnFamilyData* cfd, bool expose_blob_index, bool allow_refresh) {
   ArenaWrappedDBIter* iter = new ArenaWrappedDBIter();
   iter->Init(env, read_options, ioptions, mutable_cf_options, version, sequence,
              max_sequential_skip_in_iterations, version_number, read_callback,
-             cfh, expose_blob_index, allow_refresh);
-  if (cfh != nullptr && allow_refresh) {
-    iter->StoreRefreshInfo(cfh, read_callback, expose_blob_index);
+             db_impl, cfd, expose_blob_index, allow_refresh);
+  if (db_impl != nullptr && cfd != nullptr && allow_refresh) {
+    iter->StoreRefreshInfo(db_impl, cfd, read_callback, expose_blob_index);
   }
 
   return iter;

db/arena_wrapped_db_iter.h

@@ -87,14 +87,15 @@ class ArenaWrappedDBIter : public Iterator {
             const MutableCFOptions& mutable_cf_options, const Version* version,
             const SequenceNumber& sequence,
             uint64_t max_sequential_skip_in_iterations, uint64_t version_number,
-            ReadCallback* read_callback, ColumnFamilyHandleImpl* cfh,
+            ReadCallback* read_callback, DBImpl* db_impl, ColumnFamilyData* cfd,
             bool expose_blob_index, bool allow_refresh);
 
   // Store some parameters so we can refresh the iterator at a later point
   // with these same params
-  void StoreRefreshInfo(ColumnFamilyHandleImpl* cfh,
+  void StoreRefreshInfo(DBImpl* db_impl, ColumnFamilyData* cfd,
                         ReadCallback* read_callback, bool expose_blob_index) {
-    cfh_ = cfh;
+    db_impl_ = db_impl;
+    cfd_ = cfd;
     read_callback_ = read_callback;
     expose_blob_index_ = expose_blob_index;
   }
@@ -103,7 +104,8 @@ class ArenaWrappedDBIter : public Iterator {
   DBIter* db_iter_ = nullptr;
   Arena arena_;
   uint64_t sv_number_;
-  ColumnFamilyHandleImpl* cfh_ = nullptr;
+  ColumnFamilyData* cfd_ = nullptr;
+  DBImpl* db_impl_ = nullptr;
   ReadOptions read_options_;
   ReadCallback* read_callback_;
   bool expose_blob_index_ = false;
@@ -114,13 +116,13 @@ class ArenaWrappedDBIter : public Iterator {
 };
 
 // Generate the arena wrapped iterator class.
-// `cfh` is used for reneweal. If left null, renewal will not
+// `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,
     uint64_t version_number, ReadCallback* read_callback,
-    ColumnFamilyHandleImpl* cfh = nullptr, bool expose_blob_index = false,
-    bool allow_refresh = true);
+    DBImpl* db_impl = nullptr, ColumnFamilyData* cfd = nullptr,
+    bool expose_blob_index = false, bool allow_refresh = true);
 }  // namespace ROCKSDB_NAMESPACE

db/attribute_group_iterator_impl.cc

@@ -1,20 +0,0 @@
-//  Copyright (c) Meta Platforms, Inc. and affiliates.
-//  This source code is licensed under both the GPLv2 (found in the
-//  COPYING file in the root directory) and Apache 2.0 License
-//  (found in the LICENSE.Apache file in the root directory).
-
-#include "db/attribute_group_iterator_impl.h"
-
-namespace ROCKSDB_NAMESPACE {
-
-const AttributeGroups kNoAttributeGroups;
-const IteratorAttributeGroups kNoIteratorAttributeGroups;
-
-void AttributeGroupIteratorImpl::AddToAttributeGroups(
-    const autovector<MultiCfIteratorInfo>& items) {
-  for (const auto& item : items) {
-    attribute_groups_.emplace_back(item.cfh, &item.iterator->columns());
-  }
-}
-
-}  // namespace ROCKSDB_NAMESPACE

db/attribute_group_iterator_impl.h

@@ -1,83 +0,0 @@
-//  Copyright (c) Meta Platforms, Inc. and affiliates.
-//  This source code is licensed under both the GPLv2 (found in the
-//  COPYING file in the root directory) and Apache 2.0 License
-//  (found in the LICENSE.Apache file in the root directory).
-
-#pragma once
-
-#include "db/multi_cf_iterator_impl.h"
-#include "rocksdb/attribute_groups.h"
-
-namespace ROCKSDB_NAMESPACE {
-
-class AttributeGroupIteratorImpl : public AttributeGroupIterator {
- public:
-  AttributeGroupIteratorImpl(
-      const Comparator* comparator,
-      const std::vector<ColumnFamilyHandle*>& column_families,
-      const std::vector<Iterator*>& child_iterators)
-      : impl_(
-            comparator, column_families, child_iterators, [this]() { Reset(); },
-            [this](const autovector<MultiCfIteratorInfo>& items) {
-              AddToAttributeGroups(items);
-            }) {}
-  ~AttributeGroupIteratorImpl() override {}
-
-  // No copy allowed
-  AttributeGroupIteratorImpl(const AttributeGroupIteratorImpl&) = delete;
-  AttributeGroupIteratorImpl& operator=(const AttributeGroupIteratorImpl&) =
-      delete;
-
-  bool Valid() const override { return impl_.Valid(); }
-  void SeekToFirst() override { impl_.SeekToFirst(); }
-  void SeekToLast() override { impl_.SeekToLast(); }
-  void Seek(const Slice& target) override { impl_.Seek(target); }
-  void SeekForPrev(const Slice& target) override { impl_.SeekForPrev(target); }
-  void Next() override { impl_.Next(); }
-  void Prev() override { impl_.Prev(); }
-  Slice key() const override { return impl_.key(); }
-  Status status() const override { return impl_.status(); }
-
-  const IteratorAttributeGroups& attribute_groups() const override {
-    assert(Valid());
-    return attribute_groups_;
-  }
-
-  void Reset() { attribute_groups_.clear(); }
-
- private:
-  MultiCfIteratorImpl impl_;
-  IteratorAttributeGroups attribute_groups_;
-  void AddToAttributeGroups(const autovector<MultiCfIteratorInfo>& items);
-};
-
-class EmptyAttributeGroupIterator : public AttributeGroupIterator {
- public:
-  explicit EmptyAttributeGroupIterator(const Status& s) : status_(s) {}
-  bool Valid() const override { return false; }
-  void Seek(const Slice& /*target*/) override {}
-  void SeekForPrev(const Slice& /*target*/) override {}
-  void SeekToFirst() override {}
-  void SeekToLast() override {}
-  void Next() override { assert(false); }
-  void Prev() override { assert(false); }
-  Slice key() const override {
-    assert(false);
-    return Slice();
-  }
-  Status status() const override { return status_; }
-
-  const IteratorAttributeGroups& attribute_groups() const override {
-    return kNoIteratorAttributeGroups;
-  }
-
- private:
-  Status status_;
-};
-
-inline std::unique_ptr<AttributeGroupIterator> NewAttributeGroupErrorIterator(
-    const Status& status) {
-  return std::make_unique<EmptyAttributeGroupIterator>(status);
-}
-
-}  // namespace ROCKSDB_NAMESPACE

db/blob/blob_file_builder.cc

@@ -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);
 

db/blob/blob_file_builder.h

@@ -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_;

db/blob/blob_file_builder_test.cc

@@ -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);
 

db/blob/blob_file_cache_test.cc

@@ -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

db/blob/blob_file_reader_test.cc

@@ -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
@@ -405,7 +404,7 @@ TEST_F(BlobFileReaderTest, CreateReaderAndGetBlob) {
 
     requests_buf[0] =
         BlobReadRequest(key_refs[0], blob_offsets[0], blob_sizes[0],
-                        kNoCompression, nullptr, statuses_buf.data());
+                        kNoCompression, nullptr, &statuses_buf[0]);
     requests_buf[1] =
         BlobReadRequest(key_refs[1], blob_offsets[1], blob_sizes[1] + 1,
                         kNoCompression, nullptr, &statuses_buf[1]);
@@ -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;

db/blob/blob_log_writer.cc

@@ -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;
 }
 

db/blob/blob_log_writer.h

@@ -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_;

db/blob/blob_source_test.cc

@@ -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
@@ -168,8 +167,8 @@ TEST_F(BlobSourceTest, GetBlobsFromCache) {
 
   uint64_t file_size = BlobLogHeader::kSize;
   for (size_t i = 0; i < num_blobs; ++i) {
-    keys.emplace_back(key_strs[i]);
-    blobs.emplace_back(blob_strs[i]);
+    keys.push_back({key_strs[i]});
+    blobs.push_back({blob_strs[i]});
     file_size += BlobLogRecord::kHeaderSize + keys[i].size() + blobs[i].size();
   }
   file_size += BlobLogFooter::kSize;
@@ -482,8 +481,8 @@ TEST_F(BlobSourceTest, GetCompressedBlobs) {
   std::vector<Slice> blobs;
 
   for (size_t i = 0; i < num_blobs; ++i) {
-    keys.emplace_back(key_strs[i]);
-    blobs.emplace_back(blob_strs[i]);
+    keys.push_back({key_strs[i]});
+    blobs.push_back({blob_strs[i]});
   }
 
   std::vector<uint64_t> blob_offsets(keys.size());
@@ -610,8 +609,8 @@ TEST_F(BlobSourceTest, MultiGetBlobsFromMultiFiles) {
   uint64_t file_size = BlobLogHeader::kSize;
   uint64_t blob_value_bytes = 0;
   for (size_t i = 0; i < num_blobs; ++i) {
-    keys.emplace_back(key_strs[i]);
-    blobs.emplace_back(blob_strs[i]);
+    keys.push_back({key_strs[i]});
+    blobs.push_back({blob_strs[i]});
     blob_value_bytes += blobs[i].size();
     file_size += BlobLogRecord::kHeaderSize + keys[i].size() + blobs[i].size();
   }
@@ -802,8 +801,8 @@ TEST_F(BlobSourceTest, MultiGetBlobsFromCache) {
 
   uint64_t file_size = BlobLogHeader::kSize;
   for (size_t i = 0; i < num_blobs; ++i) {
-    keys.emplace_back(key_strs[i]);
-    blobs.emplace_back(blob_strs[i]);
+    keys.push_back({key_strs[i]});
+    blobs.push_back({blob_strs[i]});
     file_size += BlobLogRecord::kHeaderSize + keys[i].size() + blobs[i].size();
   }
   file_size += BlobLogFooter::kSize;
@@ -1164,7 +1163,7 @@ TEST_F(BlobSecondaryCacheTest, GetBlobsFromSecondaryCache) {
     ASSERT_OK(blob_source.GetBlob(read_options, keys[0], file_number,
                                   blob_offsets[0], file_size, blob_sizes[0],
                                   kNoCompression, nullptr /* prefetch_buffer */,
-                                  values.data(), nullptr /* bytes_read */));
+                                  &values[0], nullptr /* bytes_read */));
     // Release cache handle
     values[0].Reset();
 
@@ -1183,7 +1182,7 @@ TEST_F(BlobSecondaryCacheTest, GetBlobsFromSecondaryCache) {
     ASSERT_OK(blob_source.GetBlob(read_options, keys[0], file_number,
                                   blob_offsets[0], file_size, blob_sizes[0],
                                   kNoCompression, nullptr /* prefetch_buffer */,
-                                  values.data(), nullptr /* bytes_read */));
+                                  &values[0], nullptr /* bytes_read */));
     ASSERT_EQ(values[0], blobs[0]);
     ASSERT_TRUE(
         blob_source.TEST_BlobInCache(file_number, file_size, blob_offsets[0]));
@@ -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);
 
@@ -1263,7 +1262,7 @@ TEST_F(BlobSecondaryCacheTest, GetBlobsFromSecondaryCache) {
       ASSERT_OK(blob_source.GetBlob(
           read_options, keys[0], file_number, blob_offsets[0], file_size,
           blob_sizes[0], kNoCompression, nullptr /* prefetch_buffer */,
-          values.data(), nullptr /* bytes_read */));
+          &values[0], nullptr /* bytes_read */));
       ASSERT_EQ(values[0], blobs[0]);
 
       // Release cache handle
@@ -1365,8 +1364,8 @@ class BlobSourceCacheReservationTest : public DBTestBase {
 
     blob_file_size_ = BlobLogHeader::kSize;
     for (size_t i = 0; i < kNumBlobs; ++i) {
-      keys_.emplace_back(key_strs_[i]);
-      blobs_.emplace_back(blob_strs_[i]);
+      keys_.push_back({key_strs_[i]});
+      blobs_.push_back({blob_strs_[i]});
       blob_file_size_ +=
           BlobLogRecord::kHeaderSize + keys_[i].size() + blobs_[i].size();
     }

db/blob/db_blob_basic_test.cc

@@ -418,8 +418,8 @@ TEST_F(DBBlobBasicTest, MultiGetBlobs) {
     std::array<PinnableSlice, num_keys> values;
     std::array<Status, num_keys> statuses;
 
-    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys, &keys[0],
+                  &values[0], &statuses[0]);
 
     ASSERT_OK(statuses[0]);
     ASSERT_EQ(values[0], first_value);
@@ -441,8 +441,8 @@ TEST_F(DBBlobBasicTest, MultiGetBlobs) {
     std::array<PinnableSlice, num_keys> values;
     std::array<Status, num_keys> statuses;
 
-    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys, &keys[0],
+                  &values[0], &statuses[0]);
 
     ASSERT_OK(statuses[0]);
     ASSERT_EQ(values[0], first_value);
@@ -512,8 +512,8 @@ TEST_F(DBBlobBasicTest, MultiGetBlobsFromCache) {
     std::array<PinnableSlice, num_keys> values;
     std::array<Status, num_keys> statuses;
 
-    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys, &keys[0],
+                  &values[0], &statuses[0]);
 
     ASSERT_OK(statuses[0]);
     ASSERT_EQ(values[0], first_value);
@@ -534,8 +534,8 @@ TEST_F(DBBlobBasicTest, MultiGetBlobsFromCache) {
     std::array<PinnableSlice, num_keys> values;
     std::array<Status, num_keys> statuses;
 
-    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys, &keys[0],
+                  &values[0], &statuses[0]);
 
     ASSERT_OK(statuses[0]);
     ASSERT_EQ(values[0], first_value);
@@ -553,8 +553,8 @@ TEST_F(DBBlobBasicTest, MultiGetBlobsFromCache) {
     std::array<PinnableSlice, num_keys> values;
     std::array<Status, num_keys> statuses;
 
-    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys, &keys[0],
+                  &values[0], &statuses[0]);
 
     ASSERT_OK(statuses[0]);
     ASSERT_EQ(values[0], first_value);
@@ -574,8 +574,8 @@ TEST_F(DBBlobBasicTest, MultiGetBlobsFromCache) {
     std::array<PinnableSlice, num_keys> values;
     std::array<Status, num_keys> statuses;
 
-    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys, &keys[0],
+                  &values[0], &statuses[0]);
 
     ASSERT_OK(statuses[0]);
     ASSERT_EQ(values[0], first_value);
@@ -758,8 +758,8 @@ TEST_F(DBBlobBasicTest, MultiGetWithDirectIO) {
     //
     // [offset=0, len=12288]
 
-    db_->MultiGet(ReadOptions(), db_->DefaultColumnFamily(), num_keys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(ReadOptions(), db_->DefaultColumnFamily(), num_keys, &keys[0],
+                  &values[0], &statuses[0]);
 
     SyncPoint::GetInstance()->DisableProcessing();
     SyncPoint::GetInstance()->ClearAllCallBacks();
@@ -829,8 +829,8 @@ TEST_F(DBBlobBasicTest, MultiGetBlobsFromMultipleFiles) {
   {
     std::array<PinnableSlice, kNumKeys> values;
     std::array<Status, kNumKeys> statuses;
-    db_->MultiGet(read_options, db_->DefaultColumnFamily(), kNumKeys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), kNumKeys, &keys[0],
+                  &values[0], &statuses[0]);
 
     for (size_t i = 0; i < kNumKeys; ++i) {
       ASSERT_OK(statuses[i]);
@@ -843,8 +843,8 @@ TEST_F(DBBlobBasicTest, MultiGetBlobsFromMultipleFiles) {
   {
     std::array<PinnableSlice, kNumKeys> values;
     std::array<Status, kNumKeys> statuses;
-    db_->MultiGet(read_options, db_->DefaultColumnFamily(), kNumKeys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), kNumKeys, &keys[0],
+                  &values[0], &statuses[0]);
 
     for (size_t i = 0; i < kNumKeys; ++i) {
       ASSERT_TRUE(statuses[i].IsIncomplete());
@@ -858,8 +858,8 @@ TEST_F(DBBlobBasicTest, MultiGetBlobsFromMultipleFiles) {
   {
     std::array<PinnableSlice, kNumKeys> values;
     std::array<Status, kNumKeys> statuses;
-    db_->MultiGet(read_options, db_->DefaultColumnFamily(), kNumKeys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), kNumKeys, &keys[0],
+                  &values[0], &statuses[0]);
 
     for (size_t i = 0; i < kNumKeys; ++i) {
       ASSERT_OK(statuses[i]);
@@ -872,8 +872,8 @@ TEST_F(DBBlobBasicTest, MultiGetBlobsFromMultipleFiles) {
   {
     std::array<PinnableSlice, kNumKeys> values;
     std::array<Status, kNumKeys> statuses;
-    db_->MultiGet(read_options, db_->DefaultColumnFamily(), kNumKeys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), kNumKeys, &keys[0],
+                  &values[0], &statuses[0]);
 
     for (size_t i = 0; i < kNumKeys; ++i) {
       ASSERT_OK(statuses[i]);
@@ -1182,30 +1182,6 @@ TEST_F(DBBlobBasicTest, GetMergeBlobWithPut) {
   ASSERT_EQ(Get("Key1"), "v1,v2,v3");
 }
 
-TEST_F(DBBlobBasicTest, GetMergeBlobFromMemoryTier) {
-  Options options = GetDefaultOptions();
-  options.merge_operator = MergeOperators::CreateStringAppendOperator();
-  options.enable_blob_files = true;
-  options.min_blob_size = 0;
-
-  Reopen(options);
-
-  ASSERT_OK(Put(Key(0), "v1"));
-  ASSERT_OK(Flush());
-  ASSERT_OK(Merge(Key(0), "v2"));
-  ASSERT_OK(Flush());
-
-  // Regular `Get()` loads data block to cache.
-  std::string value;
-  ASSERT_OK(db_->Get(ReadOptions(), Key(0), &value));
-  ASSERT_EQ("v1,v2", value);
-
-  // Base value blob is still uncached, so an in-memory read will fail.
-  ReadOptions read_options;
-  read_options.read_tier = kBlockCacheTier;
-  ASSERT_TRUE(db_->Get(read_options, Key(0), &value).IsIncomplete());
-}
-
 TEST_F(DBBlobBasicTest, MultiGetMergeBlobWithPut) {
   constexpr size_t num_keys = 3;
 
@@ -1230,8 +1206,8 @@ TEST_F(DBBlobBasicTest, MultiGetMergeBlobWithPut) {
   std::array<PinnableSlice, num_keys> values;
   std::array<Status, num_keys> statuses;
 
-  db_->MultiGet(ReadOptions(), db_->DefaultColumnFamily(), num_keys,
-                keys.data(), values.data(), statuses.data());
+  db_->MultiGet(ReadOptions(), db_->DefaultColumnFamily(), num_keys, &keys[0],
+                &values[0], &statuses[0]);
 
   ASSERT_OK(statuses[0]);
   ASSERT_EQ(values[0], "v0_0,v0_1,v0_2");
@@ -1494,8 +1470,8 @@ TEST_P(DBBlobBasicIOErrorMultiGetTest, MultiGetBlobs_IOError) {
   });
   SyncPoint::GetInstance()->EnableProcessing();
 
-  db_->MultiGet(ReadOptions(), db_->DefaultColumnFamily(), num_keys,
-                keys.data(), values.data(), statuses.data());
+  db_->MultiGet(ReadOptions(), db_->DefaultColumnFamily(), num_keys, &keys[0],
+                &values[0], &statuses[0]);
 
   SyncPoint::GetInstance()->DisableProcessing();
   SyncPoint::GetInstance()->ClearAllCallBacks();
@@ -1551,57 +1527,6 @@ TEST_P(DBBlobBasicIOErrorMultiGetTest, MultipleBlobFiles) {
   ASSERT_TRUE(statuses[1].IsIOError());
 }
 
-TEST_F(DBBlobBasicTest, MultiGetFindTable_IOError) {
-  // Repro test for a specific bug where `MultiGet()` would fail to open a table
-  // in `FindTable()` and then proceed to return raw blob handles for the other
-  // keys.
-  Options options = GetDefaultOptions();
-  options.enable_blob_files = true;
-  options.min_blob_size = 0;
-
-  Reopen(options);
-
-  // Force no table cache so every read will preload the SST file.
-  dbfull()->TEST_table_cache()->SetCapacity(0);
-
-  constexpr size_t num_keys = 2;
-
-  constexpr char key1[] = "key1";
-  constexpr char value1[] = "blob1";
-
-  ASSERT_OK(Put(key1, value1));
-  ASSERT_OK(Flush());
-
-  constexpr char key2[] = "key2";
-  constexpr char value2[] = "blob2";
-
-  ASSERT_OK(Put(key2, value2));
-  ASSERT_OK(Flush());
-
-  std::atomic<int> num_files_opened = 0;
-  // This test would be more realistic if we injected an `IOError` from the
-  // `FileSystem`
-  SyncPoint::GetInstance()->SetCallBack(
-      "TableCache::MultiGet:FindTable", [&](void* status) {
-        num_files_opened++;
-        if (num_files_opened == 2) {
-          Status* s = static_cast<Status*>(status);
-          *s = Status::IOError();
-        }
-      });
-  SyncPoint::GetInstance()->EnableProcessing();
-
-  std::array<Slice, num_keys> keys{{key1, key2}};
-  std::array<PinnableSlice, num_keys> values;
-  std::array<Status, num_keys> statuses;
-  db_->MultiGet(ReadOptions(), db_->DefaultColumnFamily(), num_keys,
-                keys.data(), values.data(), statuses.data());
-
-  ASSERT_TRUE(statuses[0].IsIOError());
-  ASSERT_OK(statuses[1]);
-  ASSERT_EQ(value2, values[1]);
-}
-
 namespace {
 
 class ReadBlobCompactionFilter : public CompactionFilter {
@@ -1895,7 +1820,7 @@ TEST_F(DBBlobBasicTest, GetEntityBlob) {
     std::array<Status, num_keys> statuses;
 
     db_->MultiGetEntity(ReadOptions(), db_->DefaultColumnFamily(), num_keys,
-                        keys.data(), results.data(), statuses.data());
+                        &keys[0], &results[0], &statuses[0]);
 
     ASSERT_OK(statuses[0]);
     ASSERT_EQ(results[0].columns(), expected_columns);
@@ -1992,8 +1917,8 @@ TEST_F(DBBlobWithTimestampTest, MultiGetBlobs) {
     std::array<PinnableSlice, num_keys> values;
     std::array<Status, num_keys> statuses;
 
-    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys,
-                  keys.data(), values.data(), statuses.data());
+    db_->MultiGet(read_options, db_->DefaultColumnFamily(), num_keys, &keys[0],
+                  &values[0], &statuses[0]);
 
     ASSERT_OK(statuses[0]);
     ASSERT_EQ(values[0], first_value);
@@ -2076,8 +2001,8 @@ TEST_F(DBBlobWithTimestampTest, MultiGetMergeBlobWithPut) {
   std::array<PinnableSlice, num_keys> values;
   std::array<Status, num_keys> statuses;
 
-  db_->MultiGet(read_opts, db_->DefaultColumnFamily(), num_keys, keys.data(),
-                values.data(), statuses.data());
+  db_->MultiGet(read_opts, db_->DefaultColumnFamily(), num_keys, &keys[0],
+                &values[0], &statuses[0]);
 
   ASSERT_OK(statuses[0]);
   ASSERT_EQ(values[0], "v0_0,v0_1,v0_2");

db/blob/db_blob_index_test.cc

@@ -45,10 +45,10 @@ class DBBlobIndexTest : public DBTestBase {
   DBBlobIndexTest() : DBTestBase("db_blob_index_test", /*env_do_fsync=*/true) {}
 
   ColumnFamilyHandle* cfh() { return dbfull()->DefaultColumnFamily(); }
-  ColumnFamilyHandleImpl* cfh_impl() {
-    return static_cast_with_check<ColumnFamilyHandleImpl>(cfh());
+
+  ColumnFamilyData* cfd() {
+    return static_cast_with_check<ColumnFamilyHandleImpl>(cfh())->cfd();
   }
-  ColumnFamilyData* cfd() { return cfh_impl()->cfd(); }
 
   Status PutBlobIndex(WriteBatch* batch, const Slice& key,
                       const Slice& blob_index) {
@@ -96,9 +96,11 @@ class DBBlobIndexTest : public DBTestBase {
   }
 
   ArenaWrappedDBIter* GetBlobIterator() {
+    ColumnFamilyData* column_family = cfd();
     DBImpl* db_impl = dbfull();
     return db_impl->NewIteratorImpl(
-        ReadOptions(), cfh_impl(), cfd()->GetReferencedSuperVersion(db_impl),
+        ReadOptions(), column_family,
+        column_family->GetReferencedSuperVersion(db_impl),
         db_impl->GetLatestSequenceNumber(), nullptr /*read_callback*/,
         true /*expose_blob_index*/);
   }
@@ -323,7 +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());
     };
   };
 

db/blob/prefetch_buffer_collection.cc

@@ -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();

db/builder.cc

@@ -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,
@@ -70,9 +68,10 @@ Status BuildTable(
     bool paranoid_file_checks, InternalStats* internal_stats,
     IOStatus* io_status, const std::shared_ptr<IOTracer>& io_tracer,
     BlobFileCreationReason blob_creation_reason,
-    UnownedPtr<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,
+    const SeqnoToTimeMapping& seqno_to_time_mapping, EventLogger* event_logger,
+    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};
@@ -207,74 +208,43 @@ 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;
 
-    SequenceNumber smallest_preferred_seqno = kMaxSequenceNumber;
     std::string key_after_flush_buf;
-    std::string value_buf;
     c_iter.SeekToFirst();
     for (; c_iter.Valid(); c_iter.Next()) {
       const Slice& key = c_iter.key();
       const Slice& value = c_iter.value();
-      ParsedInternalKey ikey = c_iter.ikey();
-      key_after_flush_buf.assign(key.data(), key.size());
-      Slice key_after_flush = key_after_flush_buf;
-      Slice value_after_flush = value;
-
+      const ParsedInternalKey& ikey = c_iter.ikey();
+      Slice key_after_flush = key;
       // If user defined timestamps will be stripped from user key after flush,
       // 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;
       }
 
-      if (ikey.type == kTypeValuePreferredSeqno) {
-        auto [unpacked_value, unix_write_time] =
-            ParsePackedValueWithWriteTime(value);
-        SequenceNumber preferred_seqno =
-            seqno_to_time_mapping
-                ? seqno_to_time_mapping->GetProximalSeqnoBeforeTime(
-                      unix_write_time)
-                : kMaxSequenceNumber;
-        if (preferred_seqno < ikey.sequence) {
-          value_after_flush =
-              PackValueAndSeqno(unpacked_value, preferred_seqno, &value_buf);
-          smallest_preferred_seqno =
-              std::min(smallest_preferred_seqno, preferred_seqno);
-        } else {
-          // Cannot get a useful preferred seqno, convert it to a kTypeValue.
-          UpdateInternalKey(&key_after_flush_buf, ikey.sequence, kTypeValue);
-          ikey = ParsedInternalKey(ikey.user_key, ikey.sequence, kTypeValue);
-          key_after_flush = key_after_flush_buf;
-          value_after_flush = ParsePackedValueForValue(value);
-        }
-      }
-
       //  Generate a rolling 64-bit hash of the key and values
       //  Note :
       //  Here "key" integrates 'sequence_number'+'kType'+'user key'.
-      s = output_validator.Add(key_after_flush, value_after_flush);
+      s = output_validator.Add(key_after_flush, value);
       if (!s.ok()) {
         break;
       }
-      builder->Add(key_after_flush, value_after_flush);
+      builder->Add(key_after_flush, value);
 
-      s = meta->UpdateBoundaries(key_after_flush, value_after_flush,
-                                 ikey.sequence, ikey.type);
+      s = meta->UpdateBoundaries(key_after_flush, value, ikey.sequence,
+                                 ikey.type);
       if (!s.ok()) {
         break;
       }
 
       // 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));
@@ -291,12 +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_,
@@ -308,7 +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);
           }
@@ -327,17 +294,12 @@ Status BuildTable(
     if (!s.ok() || empty) {
       builder->Abandon();
     } else {
-      SeqnoToTimeMapping relevant_mapping;
-      if (seqno_to_time_mapping) {
-        relevant_mapping.CopyFromSeqnoRange(
-            *seqno_to_time_mapping,
-            std::min(meta->fd.smallest_seqno, smallest_preferred_seqno),
-            meta->fd.largest_seqno);
-        relevant_mapping.SetCapacity(kMaxSeqnoTimePairsPerSST);
-        relevant_mapping.Enforce(tboptions.file_creation_time);
-      }
+      std::string seqno_to_time_mapping_str;
+      seqno_to_time_mapping.Encode(
+          seqno_to_time_mapping_str, meta->fd.smallest_seqno,
+          meta->fd.largest_seqno, meta->file_creation_time);
       builder->SetSeqnoTimeTableProperties(
-          relevant_mapping,
+          seqno_to_time_mapping_str,
           ioptions.compaction_style == CompactionStyle::kCompactionStyleFIFO
               ? meta->file_creation_time
               : meta->oldest_ancester_time);
@@ -384,16 +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.
@@ -437,9 +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,
@@ -452,6 +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
@@ -476,13 +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());

db/builder.h

@@ -23,6 +23,7 @@
 #include "rocksdb/status.h"
 #include "rocksdb/table_properties.h"
 #include "rocksdb/types.h"
+#include "table/scoped_arena_iterator.h"
 
 namespace ROCKSDB_NAMESPACE {
 
@@ -49,11 +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,
@@ -63,8 +64,9 @@ Status BuildTable(
     bool paranoid_file_checks, InternalStats* internal_stats,
     IOStatus* io_status, const std::shared_ptr<IOTracer>& io_tracer,
     BlobFileCreationReason blob_creation_reason,
-    UnownedPtr<const SeqnoToTimeMapping> seqno_to_time_mapping,
+    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,

db/c.cc

@@ -46,7 +46,6 @@
 #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 +115,6 @@ 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;
@@ -446,7 +444,7 @@ struct rocksdb_mergeoperator_t : public MergeOperator {
     size_t new_value_len;
     char* tmp_new_value = (*full_merge_)(
         state_, merge_in.key.data(), merge_in.key.size(), existing_value_data,
-        existing_value_len, operand_pointers.data(), operand_sizes.data(),
+        existing_value_len, &operand_pointers[0], &operand_sizes[0],
         static_cast<int>(n), &success, &new_value_len);
     merge_out->new_value.assign(tmp_new_value, new_value_len);
 
@@ -475,9 +473,8 @@ struct rocksdb_mergeoperator_t : public MergeOperator {
     unsigned char success;
     size_t new_value_len;
     char* tmp_new_value = (*partial_merge_)(
-        state_, key.data(), key.size(), operand_pointers.data(),
-        operand_sizes.data(), static_cast<int>(operand_count), &success,
-        &new_value_len);
+        state_, key.data(), key.size(), &operand_pointers[0], &operand_sizes[0],
+        static_cast<int>(operand_count), &success, &new_value_len);
     new_value->assign(tmp_new_value, new_value_len);
 
     if (delete_value_ != nullptr) {
@@ -887,9 +884,9 @@ rocksdb_t* rocksdb_open_and_trim_history(
     size_t trim_tslen, char** errptr) {
   std::vector<ColumnFamilyDescriptor> column_families;
   for (int i = 0; i < num_column_families; i++) {
-    column_families.emplace_back(
+    column_families.push_back(ColumnFamilyDescriptor(
         std::string(column_family_names[i]),
-        ColumnFamilyOptions(column_family_options[i]->rep));
+        ColumnFamilyOptions(column_family_options[i]->rep)));
   }
 
   std::string trim_ts_(trim_ts, trim_tslen);
@@ -920,9 +917,9 @@ rocksdb_t* rocksdb_open_column_families(
     rocksdb_column_family_handle_t** column_family_handles, char** errptr) {
   std::vector<ColumnFamilyDescriptor> column_families;
   for (int i = 0; i < num_column_families; i++) {
-    column_families.emplace_back(
+    column_families.push_back(ColumnFamilyDescriptor(
         std::string(column_family_names[i]),
-        ColumnFamilyOptions(column_family_options[i]->rep));
+        ColumnFamilyOptions(column_family_options[i]->rep)));
   }
 
   DB* db;
@@ -954,9 +951,9 @@ rocksdb_t* rocksdb_open_column_families_with_ttl(
   for (int i = 0; i < num_column_families; i++) {
     ttls_vec.push_back(ttls[i]);
 
-    column_families.emplace_back(
+    column_families.push_back(ColumnFamilyDescriptor(
         std::string(column_family_names[i]),
-        ColumnFamilyOptions(column_family_options[i]->rep));
+        ColumnFamilyOptions(column_family_options[i]->rep)));
   }
 
   ROCKSDB_NAMESPACE::DBWithTTL* db;
@@ -986,9 +983,9 @@ rocksdb_t* rocksdb_open_for_read_only_column_families(
     unsigned char error_if_wal_file_exists, char** errptr) {
   std::vector<ColumnFamilyDescriptor> column_families;
   for (int i = 0; i < num_column_families; i++) {
-    column_families.emplace_back(
+    column_families.push_back(ColumnFamilyDescriptor(
         std::string(column_family_names[i]),
-        ColumnFamilyOptions(column_family_options[i]->rep));
+        ColumnFamilyOptions(column_family_options[i]->rep)));
   }
 
   DB* db;
@@ -1082,7 +1079,7 @@ rocksdb_column_family_handle_t** rocksdb_create_column_families(
   std::vector<ColumnFamilyHandle*> handles;
   std::vector<std::string> names;
   for (int i = 0; i != num_column_families; ++i) {
-    names.emplace_back(column_family_names[i]);
+    names.push_back(std::string(column_family_names[i]));
   }
   SaveError(errptr, db->rep->CreateColumnFamilies(
                         ColumnFamilyOptions(column_family_options->rep), names,
@@ -1688,11 +1685,6 @@ 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)) {
@@ -1880,8 +1872,9 @@ 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,
@@ -2745,24 +2738,6 @@ void rocksdb_options_set_block_based_table_factory(
   }
 }
 
-void rocksdb_block_based_options_set_top_level_index_pinning_tier(
-    rocksdb_block_based_table_options_t* options, int v) {
-  options->rep.metadata_cache_options.top_level_index_pinning =
-      static_cast<ROCKSDB_NAMESPACE::PinningTier>(v);
-}
-
-void rocksdb_block_based_options_set_partition_pinning_tier(
-    rocksdb_block_based_table_options_t* options, int v) {
-  options->rep.metadata_cache_options.partition_pinning =
-      static_cast<ROCKSDB_NAMESPACE::PinningTier>(v);
-}
-
-void rocksdb_block_based_options_set_unpartitioned_pinning_tier(
-    rocksdb_block_based_table_options_t* options, int v) {
-  options->rep.metadata_cache_options.unpartitioned_pinning =
-      static_cast<ROCKSDB_NAMESPACE::PinningTier>(v);
-}
-
 rocksdb_cuckoo_table_options_t* rocksdb_cuckoo_options_create() {
   return new rocksdb_cuckoo_table_options_t;
 }
@@ -2807,9 +2782,7 @@ void rocksdb_options_set_cuckoo_table_factory(
 void rocksdb_set_options(rocksdb_t* db, int count, const char* const keys[],
                          const char* const values[], char** errptr) {
   std::unordered_map<std::string, std::string> options_map;
-  for (int i = 0; i < count; i++) {
-    options_map[keys[i]] = values[i];
-  }
+  for (int i = 0; i < count; i++) options_map[keys[i]] = values[i];
   SaveError(errptr, db->rep->SetOptions(options_map));
 }
 
@@ -2818,9 +2791,7 @@ void rocksdb_set_options_cf(rocksdb_t* db,
                             const char* const keys[],
                             const char* const values[], char** errptr) {
   std::unordered_map<std::string, std::string> options_map;
-  for (int i = 0; i < count; i++) {
-    options_map[keys[i]] = values[i];
-  }
+  for (int i = 0; i < count; i++) options_map[keys[i]] = values[i];
   SaveError(errptr, db->rep->SetOptions(handle->rep, options_map));
 }
 
@@ -2948,23 +2919,6 @@ void rocksdb_options_set_cf_paths(rocksdb_options_t* opt,
   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);
 }
@@ -2975,12 +2929,6 @@ 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);
 }
@@ -3101,14 +3049,6 @@ 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;
@@ -3171,7 +3111,7 @@ 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;
 }
@@ -3543,6 +3483,35 @@ 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;
@@ -3948,14 +3917,6 @@ void rocksdb_options_set_fifo_compaction_options(
   opt->rep.compaction_options_fifo = fifo->rep;
 }
 
-void rocksdb_options_set_compaction_pri(rocksdb_options_t* opt, int pri) {
-  opt->rep.compaction_pri = static_cast<ROCKSDB_NAMESPACE::CompactionPri>(pri);
-}
-
-int rocksdb_options_get_compaction_pri(rocksdb_options_t* opt) {
-  return opt->rep.compaction_pri;
-}
-
 char* rocksdb_options_statistics_get_string(rocksdb_options_t* opt) {
   ROCKSDB_NAMESPACE::Statistics* statistics = opt->rep.statistics.get();
   if (statistics) {
@@ -4036,16 +3997,6 @@ rocksdb_ratelimiter_t* rocksdb_ratelimiter_create_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;
 }
@@ -4642,7 +4593,7 @@ 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();
 }
@@ -5091,9 +5042,7 @@ void rocksdb_env_lower_high_priority_thread_pool_cpu_priority(
 }
 
 void rocksdb_env_destroy(rocksdb_env_t* env) {
-  if (!env->is_default) {
-    delete env->rep;
-  }
+  if (!env->is_default) delete env->rep;
   delete env;
 }
 
@@ -5477,7 +5426,9 @@ 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,
@@ -5557,7 +5508,7 @@ size_t rocksdb_column_family_metadata_get_level_count(
 rocksdb_level_metadata_t* rocksdb_column_family_metadata_get_level_metadata(
     rocksdb_column_family_metadata_t* cf_meta, size_t i) {
   if (i >= cf_meta->rep.levels.size()) {
-    return nullptr;
+    return NULL;
   }
   rocksdb_level_metadata_t* level_meta =
       (rocksdb_level_metadata_t*)malloc(sizeof(rocksdb_level_metadata_t));
@@ -5772,9 +5723,9 @@ rocksdb_transactiondb_t* rocksdb_transactiondb_open_column_families(
     rocksdb_column_family_handle_t** column_family_handles, char** errptr) {
   std::vector<ColumnFamilyDescriptor> column_families;
   for (int i = 0; i < num_column_families; i++) {
-    column_families.emplace_back(
+    column_families.push_back(ColumnFamilyDescriptor(
         std::string(column_family_names[i]),
-        ColumnFamilyOptions(column_family_options[i]->rep));
+        ColumnFamilyOptions(column_family_options[i]->rep)));
   }
 
   TransactionDB* txn_db;
@@ -6566,9 +6517,9 @@ rocksdb_optimistictransactiondb_open_column_families(
     rocksdb_column_family_handle_t** column_family_handles, char** errptr) {
   std::vector<ColumnFamilyDescriptor> column_families;
   for (int i = 0; i < num_column_families; i++) {
-    column_families.emplace_back(
+    column_families.push_back(ColumnFamilyDescriptor(
         std::string(column_family_names[i]),
-        ColumnFamilyOptions(column_family_options[i]->rep));
+        ColumnFamilyOptions(column_family_options[i]->rep)));
   }
 
   OptimisticTransactionDB* otxn_db;

db/c_test.c

@@ -50,13 +50,12 @@ static void StartPhase(const char* name) {
 #endif
 static const char* GetTempDir(void) {
   const char* ret = getenv("TEST_TMPDIR");
-  if (ret == NULL || ret[0] == '\0') {
+  if (ret == NULL || ret[0] == '\0')
 #ifdef OS_WIN
     ret = getenv("TEMP");
 #else
     ret = "/tmp";
 #endif
-  }
   return ret;
 }
 #ifdef _MSC_VER
@@ -207,11 +206,10 @@ static int CmpCompare(void* arg, const char* a, size_t alen, const char* b,
   size_t n = (alen < blen) ? alen : blen;
   int r = memcmp(a, b, n);
   if (r == 0) {
-    if (alen < blen) {
+    if (alen < blen)
       r = -1;
-    } else if (alen > blen) {
+    else if (alen > blen)
       r = +1;
-    }
   }
   return r;
 }
@@ -704,10 +702,6 @@ int main(int argc, char** argv) {
   rocksdb_block_based_options_set_block_cache(table_options, cache);
   rocksdb_block_based_options_set_data_block_index_type(table_options, 1);
   rocksdb_block_based_options_set_data_block_hash_ratio(table_options, 0.75);
-  rocksdb_block_based_options_set_top_level_index_pinning_tier(table_options,
-                                                               1);
-  rocksdb_block_based_options_set_partition_pinning_tier(table_options, 2);
-  rocksdb_block_based_options_set_unpartitioned_pinning_tier(table_options, 3);
   rocksdb_options_set_block_based_table_factory(options, table_options);
 
   rocksdb_options_set_compression(options, rocksdb_no_compression);
@@ -724,11 +718,6 @@ int main(int argc, char** argv) {
   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);
@@ -889,8 +878,9 @@ int main(int argc, char** argv) {
   StartPhase("addfile");
   {
     rocksdb_envoptions_t* env_opt = rocksdb_envoptions_create();
+    rocksdb_options_t* io_options = rocksdb_options_create();
     rocksdb_sstfilewriter_t* writer =
-        rocksdb_sstfilewriter_create(env_opt, options);
+        rocksdb_sstfilewriter_create(env_opt, io_options);
 
     remove(sstfilename);
     rocksdb_sstfilewriter_open(writer, sstfilename, &err);
@@ -949,6 +939,7 @@ int main(int argc, char** argv) {
 
     rocksdb_ingestexternalfileoptions_destroy(ing_opt);
     rocksdb_sstfilewriter_destroy(writer);
+    rocksdb_options_destroy(io_options);
     rocksdb_envoptions_destroy(env_opt);
 
     // Delete all keys we just ingested
@@ -1322,8 +1313,6 @@ int main(int argc, char** argv) {
       policy = rocksdb_filterpolicy_create_ribbon_hybrid(8.0, 1);
     }
     rocksdb_block_based_options_set_filter_policy(table_options, policy);
-    rocksdb_block_based_options_set_optimize_filters_for_memory(table_options,
-                                                                0);
 
     // Create new database
     rocksdb_close(db);
@@ -1936,9 +1925,6 @@ int main(int argc, char** argv) {
     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));
 
@@ -2045,6 +2031,9 @@ 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));
 
@@ -2115,9 +2104,6 @@ int main(int argc, char** argv) {
     rocksdb_options_set_compaction_style(o, 2);
     CheckCondition(2 == rocksdb_options_get_compaction_style(o));
 
-    rocksdb_options_set_compaction_pri(o, 4);
-    CheckCondition(4 == rocksdb_options_get_compaction_pri(o));
-
     rocksdb_options_set_atomic_flush(o, 1);
     CheckCondition(1 == rocksdb_options_get_atomic_flush(o));
 
@@ -2242,6 +2228,8 @@ 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));
@@ -2373,10 +2361,6 @@ int main(int argc, char** argv) {
     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));
@@ -2524,6 +2508,11 @@ 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));
@@ -2622,10 +2611,6 @@ int main(int argc, char** argv) {
     CheckCondition(1 == rocksdb_options_get_compaction_style(copy));
     CheckCondition(2 == rocksdb_options_get_compaction_style(o));
 
-    rocksdb_options_set_compaction_pri(copy, 1);
-    CheckCondition(1 == rocksdb_options_get_compaction_pri(copy));
-    CheckCondition(4 == rocksdb_options_get_compaction_pri(o));
-
     rocksdb_options_set_atomic_flush(copy, 0);
     CheckCondition(0 == rocksdb_options_get_atomic_flush(copy));
     CheckCondition(1 == rocksdb_options_get_atomic_flush(o));
@@ -2796,30 +2781,6 @@ 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;
@@ -3726,7 +3687,7 @@ 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 =

db/coalescing_iterator.cc

@@ -1,47 +0,0 @@
-//  Copyright (c) Meta Platforms, Inc. and affiliates.
-//  This source code is licensed under both the GPLv2 (found in the
-//  COPYING file in the root directory) and Apache 2.0 License
-//  (found in the LICENSE.Apache file in the root directory).
-
-#include "db/coalescing_iterator.h"
-
-#include "db/wide/wide_columns_helper.h"
-
-namespace ROCKSDB_NAMESPACE {
-
-void CoalescingIterator::Coalesce(
-    const autovector<MultiCfIteratorInfo>& items) {
-  assert(wide_columns_.empty());
-  MinHeap heap;
-  for (const auto& item : items) {
-    assert(item.iterator);
-    for (auto& column : item.iterator->columns()) {
-      heap.push(WideColumnWithOrder{&column, item.order});
-    }
-  }
-  if (heap.empty()) {
-    return;
-  }
-  wide_columns_.reserve(heap.size());
-  auto current = heap.top();
-  heap.pop();
-  while (!heap.empty()) {
-    int comparison = current.column->name().compare(heap.top().column->name());
-    if (comparison < 0) {
-      wide_columns_.push_back(*current.column);
-    } else if (comparison > 0) {
-      // Shouldn't reach here.
-      // Current item in the heap is greater than the top item in the min heap
-      assert(false);
-    }
-    current = heap.top();
-    heap.pop();
-  }
-  wide_columns_.push_back(*current.column);
-
-  if (WideColumnsHelper::HasDefaultColumn(wide_columns_)) {
-    value_ = WideColumnsHelper::GetDefaultColumn(wide_columns_);
-  }
-}
-
-}  // namespace ROCKSDB_NAMESPACE

db/coalescing_iterator.h

@@ -1,79 +0,0 @@
-//  Copyright (c) Meta Platforms, Inc. and affiliates.
-//  This source code is licensed under both the GPLv2 (found in the
-//  COPYING file in the root directory) and Apache 2.0 License
-//  (found in the LICENSE.Apache file in the root directory).
-
-#pragma once
-
-#include "db/multi_cf_iterator_impl.h"
-
-namespace ROCKSDB_NAMESPACE {
-
-// EXPERIMENTAL
-class CoalescingIterator : public Iterator {
- public:
-  CoalescingIterator(const Comparator* comparator,
-                     const std::vector<ColumnFamilyHandle*>& column_families,
-                     const std::vector<Iterator*>& child_iterators)
-      : impl_(
-            comparator, column_families, child_iterators, [this]() { Reset(); },
-            [this](const autovector<MultiCfIteratorInfo>& items) {
-              Coalesce(items);
-            }) {}
-  ~CoalescingIterator() override {}
-
-  // No copy allowed
-  CoalescingIterator(const CoalescingIterator&) = delete;
-  CoalescingIterator& operator=(const CoalescingIterator&) = delete;
-
-  bool Valid() const override { return impl_.Valid(); }
-  void SeekToFirst() override { impl_.SeekToFirst(); }
-  void SeekToLast() override { impl_.SeekToLast(); }
-  void Seek(const Slice& target) override { impl_.Seek(target); }
-  void SeekForPrev(const Slice& target) override { impl_.SeekForPrev(target); }
-  void Next() override { impl_.Next(); }
-  void Prev() override { impl_.Prev(); }
-  Slice key() const override { return impl_.key(); }
-  Status status() const override { return impl_.status(); }
-
-  Slice value() const override {
-    assert(Valid());
-    return value_;
-  }
-  const WideColumns& columns() const override {
-    assert(Valid());
-    return wide_columns_;
-  }
-
-  void Reset() {
-    value_.clear();
-    wide_columns_.clear();
-  }
-
- private:
-  MultiCfIteratorImpl impl_;
-  Slice value_;
-  WideColumns wide_columns_;
-
-  struct WideColumnWithOrder {
-    const WideColumn* column;
-    int order;
-  };
-
-  class WideColumnWithOrderComparator {
-   public:
-    explicit WideColumnWithOrderComparator() {}
-    bool operator()(const WideColumnWithOrder& a,
-                    const WideColumnWithOrder& b) const {
-      int c = a.column->name().compare(b.column->name());
-      return c == 0 ? a.order - b.order > 0 : c > 0;
-    }
-  };
-
-  using MinHeap =
-      BinaryHeap<WideColumnWithOrder, WideColumnWithOrderComparator>;
-
-  void Coalesce(const autovector<MultiCfIteratorInfo>& items);
-};
-
-}  // namespace ROCKSDB_NAMESPACE

db/column_family.cc

@@ -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;
     }
@@ -411,13 +411,6 @@ ColumnFamilyOptions SanitizeOptions(const ImmutableDBOptions& db_options,
           "periodic_compaction_seconds does not support FIFO compaction. You"
           "may want to set option TTL instead.");
     }
-    if (result.last_level_temperature != Temperature::kUnknown) {
-      ROCKS_LOG_WARN(
-          db_options.info_log.get(),
-          "last_level_temperature is ignored with FIFO compaction. Consider "
-          "CompactionOptionsFIFO::file_temperature_age_thresholds.");
-      result.last_level_temperature = Temperature::kUnknown;
-    }
   }
 
   // For universal compaction, `ttl` and `periodic_compaction_seconds` mean the
@@ -477,16 +470,13 @@ void SuperVersion::Cleanup() {
   cfd->UnrefAndTryDelete();
 }
 
-void SuperVersion::Init(
-    ColumnFamilyData* new_cfd, MemTable* new_mem, MemTableListVersion* new_imm,
-    Version* new_current,
-    std::shared_ptr<const SeqnoToTimeMapping> new_seqno_to_time_mapping) {
+void SuperVersion::Init(ColumnFamilyData* new_cfd, MemTable* new_mem,
+                        MemTableListVersion* new_imm, Version* new_current) {
   cfd = new_cfd;
   mem = new_mem;
   imm = new_imm;
   current = new_current;
   full_history_ts_low = cfd->GetFullHistoryTsLow();
-  seqno_to_time_mapping = std::move(new_seqno_to_time_mapping);
   cfd->Ref();
   mem->Ref();
   imm->Ref();
@@ -582,7 +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) {
@@ -843,8 +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);
 
@@ -874,43 +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>
@@ -1066,7 +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_ =
@@ -1076,32 +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();
       }
@@ -1189,22 +1132,21 @@ 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);
-  merge_iter_builder.AddIterator(super_version->mem->NewIterator(
-      read_opts, /*seqno_to_time_mapping=*/nullptr, &arena));
-  super_version->imm->AddIterators(read_opts, /*seqno_to_time_mapping=*/nullptr,
-                                   &merge_iter_builder,
+  merge_iter_builder.AddIterator(
+      super_version->mem->NewIterator(read_opts, &arena));
+  super_version->imm->AddIterators(read_opts, &merge_iter_builder,
                                    false /* add_range_tombstone_iter */);
-  ScopedArenaPtr<InternalIterator> memtable_iter(merge_iter_builder.Finish());
+  ScopedArenaIterator memtable_iter(merge_iter_builder.Finish());
 
   auto read_seq = super_version->current->version_set()->LastSequence();
   ReadRangeDelAggregator range_del_agg(&internal_comparator_, read_seq);
@@ -1234,8 +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)) {
@@ -1340,12 +1281,7 @@ void ColumnFamilyData::InstallSuperVersion(
     const MutableCFOptions& mutable_cf_options) {
   SuperVersion* new_superversion = sv_context->new_superversion.release();
   new_superversion->mutable_cf_options = mutable_cf_options;
-  new_superversion->Init(this, mem_, imm_.current(), current_,
-                         sv_context->new_seqno_to_time_mapping
-                             ? std::move(sv_context->new_seqno_to_time_mapping)
-                         : super_version_
-                             ? super_version_->ShareSeqnoToTimeMapping()
-                             : nullptr);
+  new_superversion->Init(this, mem_, imm_.current(), current_);
   SuperVersion* old_superversion = super_version_;
   super_version_ = new_superversion;
   if (old_superversion == nullptr || old_superversion->current != current() ||

db/column_family.h

@@ -26,7 +26,6 @@
 #include "rocksdb/env.h"
 #include "rocksdb/options.h"
 #include "trace_replay/block_cache_tracer.h"
-#include "util/cast_util.h"
 #include "util/hash_containers.h"
 #include "util/thread_local.h"
 
@@ -169,12 +168,11 @@ class ColumnFamilyHandleImpl : public ColumnFamilyHandle {
   // destroy without mutex
   virtual ~ColumnFamilyHandleImpl();
   virtual ColumnFamilyData* cfd() const { return cfd_; }
-  virtual DBImpl* db() const { return db_; }
 
-  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_;
@@ -195,7 +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_;
@@ -220,9 +218,6 @@ struct SuperVersion {
   // enable UDT feature, this is an empty string.
   std::string full_history_ts_low;
 
-  // A shared copy of the DB's seqno to time mapping.
-  std::shared_ptr<const SeqnoToTimeMapping> seqno_to_time_mapping{nullptr};
-
   // should be called outside the mutex
   SuperVersion() = default;
   ~SuperVersion();
@@ -236,23 +231,8 @@ struct SuperVersion {
   // that needs to be deleted in to_delete vector. Unrefing those
   // objects needs to be done in the mutex
   void Cleanup();
-  void Init(
-      ColumnFamilyData* new_cfd, MemTable* new_mem,
-      MemTableListVersion* new_imm, Version* new_current,
-      std::shared_ptr<const SeqnoToTimeMapping> new_seqno_to_time_mapping);
-
-  // Share the ownership of the seqno to time mapping object referred to in this
-  // SuperVersion. To be used by the new SuperVersion to be installed after this
-  // one if seqno to time mapping does not change in between these two
-  // SuperVersions. Or to share the ownership of the mapping with a FlushJob.
-  std::shared_ptr<const SeqnoToTimeMapping> ShareSeqnoToTimeMapping() {
-    return seqno_to_time_mapping;
-  }
-
-  // Access the seqno to time mapping object in this SuperVersion.
-  UnownedPtr<const SeqnoToTimeMapping> GetSeqnoToTimeMapping() const {
-    return seqno_to_time_mapping.get();
-  }
+  void Init(ColumnFamilyData* new_cfd, MemTable* new_mem,
+            MemTableListVersion* new_imm, Version* new_current);
 
   // The value of dummy is not actually used. kSVInUse takes its address as a
   // mark in the thread local storage to indicate the SuperVersion is in use
@@ -271,21 +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;
 
@@ -420,7 +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);
 
@@ -449,8 +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_; }
@@ -593,7 +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_;
@@ -873,17 +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_;
@@ -891,11 +872,12 @@ 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);
+extern const ImmutableOptions& GetImmutableOptions(
+    ColumnFamilyHandle* column_family);
 
 }  // namespace ROCKSDB_NAMESPACE

db/column_family_test.cc

@@ -270,7 +270,7 @@ class ColumnFamilyTestBase : public testing::Test {
 
   void Reopen(const std::vector<ColumnFamilyOptions> options = {}) {
     std::vector<std::string> names;
-    for (const auto& name : names_) {
+    for (auto name : names_) {
       if (name != "") {
         names.push_back(name);
       }
@@ -607,7 +607,7 @@ TEST_P(FlushEmptyCFTestWithParam, FlushEmptyCFTest) {
   // Preserve file system state up to here to simulate a crash condition.
   fault_env->SetFilesystemActive(false);
   std::vector<std::string> names;
-  for (const auto& name : names_) {
+  for (auto name : names_) {
     if (name != "") {
       names.push_back(name);
     }
@@ -669,7 +669,7 @@ TEST_P(FlushEmptyCFTestWithParam, FlushEmptyCFTest2) {
   // Preserve file system state up to here to simulate a crash condition.
   fault_env->SetFilesystemActive(false);
   std::vector<std::string> names;
-  for (const auto& name : names_) {
+  for (auto name : names_) {
     if (name != "") {
       names.push_back(name);
     }
@@ -1034,7 +1034,7 @@ TEST_P(ColumnFamilyTest, CrashAfterFlush) {
   fault_env->SetFilesystemActive(false);
 
   std::vector<std::string> names;
-  for (const auto& name : names_) {
+  for (auto name : names_) {
     if (name != "") {
       names.push_back(name);
     }
@@ -2604,92 +2604,90 @@ 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 +2706,7 @@ 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 +2718,14 @@ 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 +2744,7 @@ 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 +2753,14 @@ 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 +2785,19 @@ 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 +2853,53 @@ 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 +2932,29 @@ 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 +2971,6 @@ 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;
@@ -3407,13 +3289,9 @@ TEST_P(ColumnFamilyTest, DISABLED_LogTruncationTest) {
   for (size_t i = 0; i < filenames.size(); i++) {
     uint64_t number;
     FileType type;
-    if (!(ParseFileName(filenames[i], &number, &type))) {
-      continue;
-    }
+    if (!(ParseFileName(filenames[i], &number, &type))) continue;
 
-    if (type != kWalFile) {
-      continue;
-    }
+    if (type != kWalFile) continue;
 
     logfs.push_back(filenames[i]);
   }

db/compact_files_test.cc

@@ -34,8 +34,8 @@ class CompactFilesTest : public testing::Test {
 // A class which remembers the name of each flushed file.
 class FlushedFileCollector : public EventListener {
  public:
-  FlushedFileCollector() = default;
-  ~FlushedFileCollector() override = default;
+  FlushedFileCollector() {}
+  ~FlushedFileCollector() override {}
 
   void OnFlushCompleted(DB* /*db*/, const FlushJobInfo& info) override {
     std::lock_guard<std::mutex> lock(mutex_);
@@ -45,7 +45,7 @@ class FlushedFileCollector : public EventListener {
   std::vector<std::string> GetFlushedFiles() {
     std::lock_guard<std::mutex> lock(mutex_);
     std::vector<std::string> result;
-    for (const auto& fname : flushed_files_) {
+    for (auto fname : flushed_files_) {
       result.push_back(fname);
     }
     return result;
@@ -159,9 +159,7 @@ TEST_F(CompactFilesTest, MultipleLevel) {
   // Compact files except the file in L3
   std::vector<std::string> files;
   for (int i = 0; i < 6; ++i) {
-    if (i == 3) {
-      continue;
-    }
+    if (i == 3) continue;
     for (auto& file : meta.levels[i].files) {
       files.push_back(file.db_path + "/" + file.name);
     }
@@ -230,7 +228,7 @@ TEST_F(CompactFilesTest, ObsoleteFiles) {
   ASSERT_OK(static_cast_with_check<DBImpl>(db)->TEST_WaitForCompact());
 
   // verify all compaction input files are deleted
-  for (const auto& fname : l0_files) {
+  for (auto fname : l0_files) {
     ASSERT_EQ(Status::NotFound(), env_->FileExists(fname));
   }
   delete db;
@@ -441,50 +439,6 @@ TEST_F(CompactFilesTest, SentinelCompressionType) {
   }
 }
 
-TEST_F(CompactFilesTest, CompressionWithBlockAlign) {
-  Options options;
-  options.compression = CompressionType::kNoCompression;
-  options.create_if_missing = true;
-  options.disable_auto_compactions = true;
-
-  std::shared_ptr<FlushedFileCollector> collector =
-      std::make_shared<FlushedFileCollector>();
-  options.listeners.push_back(collector);
-
-  {
-    BlockBasedTableOptions bbto;
-    bbto.block_align = true;
-    options.table_factory.reset(NewBlockBasedTableFactory(bbto));
-  }
-
-  std::unique_ptr<DB> db;
-  {
-    DB* _db = nullptr;
-    ASSERT_OK(DB::Open(options, db_name_, &_db));
-    db.reset(_db);
-  }
-
-  ASSERT_OK(db->Put(WriteOptions(), "key", "val"));
-  ASSERT_OK(db->Flush(FlushOptions()));
-
-  // Ensure background work is fully finished including listener callbacks
-  // before accessing listener state.
-  ASSERT_OK(
-      static_cast_with_check<DBImpl>(db.get())->TEST_WaitForBackgroundWork());
-  auto l0_files = collector->GetFlushedFiles();
-  ASSERT_EQ(1, l0_files.size());
-
-  // We can run this test even without Snappy support because we expect the
-  // `CompactFiles()` to fail before actually invoking Snappy compression.
-  CompactionOptions compaction_opts;
-  compaction_opts.compression = CompressionType::kSnappyCompression;
-  ASSERT_TRUE(db->CompactFiles(compaction_opts, l0_files, 1 /* output_level */)
-                  .IsInvalidArgument());
-
-  compaction_opts.compression = CompressionType::kDisableCompressionOption;
-  ASSERT_OK(db->CompactFiles(compaction_opts, l0_files, 1 /* output_level */));
-}
-
 TEST_F(CompactFilesTest, GetCompactionJobInfo) {
   Options options;
   options.create_if_missing = true;
@@ -538,3 +492,4 @@ int main(int argc, char** argv) {
   ::testing::InitGoogleTest(&argc, argv);
   return RUN_ALL_TESTS();
 }
+

db/compaction/compaction.cc

@@ -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) {
@@ -160,9 +162,7 @@ std::vector<CompactionInputFiles> Compaction::PopulateWithAtomicBoundaries(
     AtomicCompactionUnitBoundary cur_boundary;
     size_t first_atomic_idx = 0;
     auto add_unit_boundary = [&](size_t to) {
-      if (first_atomic_idx == to) {
-        return;
-      }
+      if (first_atomic_idx == to) return;
       for (size_t k = first_atomic_idx; k < to; k++) {
         inputs[i].atomic_compaction_unit_boundaries.push_back(cur_boundary);
       }
@@ -352,9 +352,11 @@ 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) {
@@ -755,9 +757,7 @@ int InputSummary(const std::vector<FileMetaData*>& files, char* output,
     AppendHumanBytes(files.at(i)->fd.GetFileSize(), sztxt, 16);
     ret = snprintf(output + write, sz, "%" PRIu64 "(%s) ",
                    files.at(i)->fd.GetNumber(), sztxt);
-    if (ret < 0 || ret >= sz) {
-      break;
-    }
+    if (ret < 0 || ret >= sz) break;
     write += ret;
   }
   // if files.size() is non-zero, overwrite the last space

db/compaction/compaction.h

@@ -595,6 +595,6 @@ 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

db/compaction/compaction_iteration_stats.h

@@ -44,13 +44,6 @@ struct CompactionIterationStats {
   uint64_t total_blob_bytes_read = 0;
   uint64_t num_blobs_relocated = 0;
   uint64_t total_blob_bytes_relocated = 0;
-
-  // TimedPut diagnostics
-  // Total number of kTypeValuePreferredSeqno records encountered.
-  uint64_t num_input_timed_put_records = 0;
-  // Number of kTypeValuePreferredSeqno records we ended up swapping in
-  // preferred seqno.
-  uint64_t num_timed_put_swap_preferred_seqno = 0;
 };
 
 }  // namespace ROCKSDB_NAMESPACE

db/compaction/compaction_iterator.cc

@@ -491,8 +491,6 @@ void CompactionIterator::NextFromInput() {
     if (ikey_.type == kTypeDeletion || ikey_.type == kTypeSingleDeletion ||
         ikey_.type == kTypeDeletionWithTimestamp) {
       iter_stats_.num_input_deletion_records++;
-    } else if (ikey_.type == kTypeValuePreferredSeqno) {
-      iter_stats_.num_input_timed_put_records++;
     }
     iter_stats_.total_input_raw_key_bytes += key_.size();
     iter_stats_.total_input_raw_value_bytes += value_.size();
@@ -620,8 +618,7 @@ void CompactionIterator::NextFromInput() {
       // not compact out.  We will keep this Put, but can drop it's data.
       // (See Optimization 3, below.)
       if (ikey_.type != kTypeValue && ikey_.type != kTypeBlobIndex &&
-          ikey_.type != kTypeWideColumnEntity &&
-          ikey_.type != kTypeValuePreferredSeqno) {
+          ikey_.type != kTypeWideColumnEntity) {
         ROCKS_LOG_FATAL(info_log_, "Unexpected key %s for compaction output",
                         ikey_.DebugString(allow_data_in_errors_, true).c_str());
         assert(false);
@@ -635,8 +632,7 @@ void CompactionIterator::NextFromInput() {
         assert(false);
       }
 
-      if (ikey_.type == kTypeBlobIndex || ikey_.type == kTypeWideColumnEntity ||
-          ikey_.type == kTypeValuePreferredSeqno) {
+      if (ikey_.type == kTypeBlobIndex || ikey_.type == kTypeWideColumnEntity) {
         ikey_.type = kTypeValue;
         current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
       }
@@ -802,8 +798,7 @@ void CompactionIterator::NextFromInput() {
             // happened
             if (next_ikey.type != kTypeValue &&
                 next_ikey.type != kTypeBlobIndex &&
-                next_ikey.type != kTypeWideColumnEntity &&
-                next_ikey.type != kTypeValuePreferredSeqno) {
+                next_ikey.type != kTypeWideColumnEntity) {
               ++iter_stats_.num_single_del_mismatch;
             }
 
@@ -973,62 +968,6 @@ void CompactionIterator::NextFromInput() {
         validity_info_.SetValid(ValidContext::kKeepDel);
         at_next_ = true;
       }
-    } else if (ikey_.type == kTypeValuePreferredSeqno &&
-               DefinitelyInSnapshot(ikey_.sequence, earliest_snapshot_) &&
-               (bottommost_level_ ||
-                (compaction_ != nullptr &&
-                 compaction_->KeyNotExistsBeyondOutputLevel(ikey_.user_key,
-                                                            &level_ptrs_)))) {
-      // This section that attempts to swap preferred sequence number will not
-      // be invoked if this is a CompactionIterator created for flush, since
-      // `compaction_` will be nullptr and it's not bottommost either.
-      //
-      // The entries with the same user key and smaller sequence numbers are
-      // all in this earliest snapshot range to be iterated. Since those entries
-      // will be hidden by this entry [rule A], it's safe to swap in the
-      // preferred seqno now.
-      //
-      // It's otherwise not safe to swap in the preferred seqno since it's
-      // possible for entries in earlier snapshots to have sequence number that
-      // is smaller than this entry's sequence number but bigger than this
-      // entry's preferred sequence number. Swapping in the preferred sequence
-      // number will break the internal key ordering invariant for this key.
-      //
-      // A special case involving range deletion is handled separately below.
-      auto [unpacked_value, preferred_seqno] =
-          ParsePackedValueWithSeqno(value_);
-      assert(preferred_seqno < ikey_.sequence || ikey_.sequence == 0);
-      if (range_del_agg_->ShouldDelete(
-              key_, RangeDelPositioningMode::kForwardTraversal)) {
-        ++iter_stats_.num_record_drop_hidden;
-        ++iter_stats_.num_record_drop_range_del;
-        AdvanceInputIter();
-      } else {
-        InternalKey ikey_after_swap(ikey_.user_key,
-                                    std::min(preferred_seqno, ikey_.sequence),
-                                    kTypeValue);
-        Slice ikey_after_swap_slice(*ikey_after_swap.rep());
-        if (range_del_agg_->ShouldDelete(
-                ikey_after_swap_slice,
-                RangeDelPositioningMode::kForwardTraversal)) {
-          // A range tombstone that doesn't cover this kTypeValuePreferredSeqno
-          // entry will end up covering the entry, so it's not safe to swap
-          // preferred sequence number. In this case, we output the entry as is.
-          validity_info_.SetValid(ValidContext::kNewUserKey);
-        } else {
-          if (ikey_.sequence != 0) {
-            iter_stats_.num_timed_put_swap_preferred_seqno++;
-            saved_seq_for_penul_check_ = ikey_.sequence;
-            ikey_.sequence = preferred_seqno;
-          }
-          ikey_.type = kTypeValue;
-          current_key_.UpdateInternalKey(ikey_.sequence, ikey_.type);
-          key_ = current_key_.GetInternalKey();
-          ikey_.user_key = current_key_.GetUserKey();
-          value_ = unpacked_value;
-          validity_info_.SetValid(ValidContext::kSwapPreferredSeqno);
-        }
-      }
     } else if (ikey_.type == kTypeMerge) {
       if (!merge_helper_->HasOperator()) {
         status_ = Status::InvalidArgument(
@@ -1280,21 +1219,6 @@ void CompactionIterator::DecideOutputLevel() {
   }
 #endif  // NDEBUG
 
-  // saved_seq_for_penul_check_ is populated in `NextFromInput` when the
-  // entry's sequence number is non zero and validity context for output this
-  // entry is kSwapPreferredSeqno for use in `DecideOutputLevel`. It should be
-  // cleared out here unconditionally. Otherwise, it may end up getting consumed
-  // incorrectly by a different entry.
-  SequenceNumber seq_for_range_check =
-      (saved_seq_for_penul_check_.has_value() &&
-       saved_seq_for_penul_check_.value() != kMaxSequenceNumber)
-          ? saved_seq_for_penul_check_.value()
-          : ikey_.sequence;
-  saved_seq_for_penul_check_ = std::nullopt;
-  ParsedInternalKey ikey_for_range_check = ikey_;
-  if (seq_for_range_check != ikey_.sequence) {
-    ikey_for_range_check.sequence = seq_for_range_check;
-  }
   if (output_to_penultimate_level_) {
     // If it's decided to output to the penultimate level, but unsafe to do so,
     // still output to the last level. For example, moving the data from a lower
@@ -1303,7 +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_for_range_check);
+        compaction_->WithinPenultimateLevelOutputRange(ikey_);
     if (!safe_to_penultimate_level) {
       output_to_penultimate_level_ = false;
       // It could happen when disable/enable `last_level_temperature` while
@@ -1315,7 +1239,7 @@ void CompactionIterator::DecideOutputLevel() {
       // snapshot is released before enabling `last_level_temperature` feature
       // We will migrate the feature to `last_level_temperature` and maybe make
       // it not dynamically changeable.
-      if (seq_for_range_check > earliest_snapshot_) {
+      if (ikey_.sequence > earliest_snapshot_) {
         status_ = Status::Corruption(
             "Unsafe to store Seq later than snapshot in the last level if "
             "per_key_placement is enabled");

db/compaction/compaction_iterator.h

@@ -410,7 +410,6 @@ class CompactionIterator {
     kKeepDel = 9,
     kNewUserKey = 10,
     kRangeDeletion = 11,
-    kSwapPreferredSeqno = 12,
   };
 
   struct ValidityInfo {
@@ -437,17 +436,6 @@ class CompactionIterator {
   // iterator output (or current key in the underlying iterator during
   // NextFromInput()).
   ParsedInternalKey ikey_;
-
-  // When a kTypeValuePreferredSeqno entry's preferred seqno is safely swapped
-  // in in this compaction, this field saves its original sequence number for
-  // range checking whether it's safe to be placed on the penultimate level.
-  // This is to ensure when such an entry happens to be the right boundary of
-  // penultimate safe range, it won't get excluded because with the preferred
-  // seqno swapped in, it's now larger than the right boundary (itself before
-  // the swap). This is safe to do, because preferred seqno is swapped in only
-  // when no entries with the same user key exist on lower levels and this entry
-  // is already visible in the earliest snapshot.
-  std::optional<SequenceNumber> saved_seq_for_penul_check_ = kMaxSequenceNumber;
   // Stores whether ikey_.user_key is valid. If set to false, the user key is
   // not compared against the current key in the underlying iterator.
   bool has_current_user_key_ = false;

db/compaction/compaction_iterator_test.cc

@@ -17,14 +17,6 @@
 #include "utilities/merge_operators.h"
 
 namespace ROCKSDB_NAMESPACE {
-namespace {
-std::string ValueWithPreferredSeqno(std::string val,
-                                    SequenceNumber preferred_seqno = 0) {
-  std::string result = val;
-  PutFixed64(&result, preferred_seqno);
-  return result;
-}
-}  // namespace
 
 // Expects no merging attempts.
 class NoMergingMergeOp : public MergeOperator {
@@ -400,17 +392,6 @@ TEST_P(CompactionIteratorTest, CorruptionAfterSingleDeletion) {
   ASSERT_FALSE(c_iter_->Valid());
 }
 
-// Tests compatibility of TimedPut and SingleDelete. TimedPut should act as if
-// it's a Put.
-TEST_P(CompactionIteratorTest, TimedPutAndSingleDelete) {
-  InitIterators({test::KeyStr("a", 5, kTypeSingleDeletion),
-                 test::KeyStr("a", 3, kTypeValuePreferredSeqno)},
-                {"", "val"}, {}, {}, 5);
-  c_iter_->SeekToFirst();
-  ASSERT_OK(c_iter_->status());
-  ASSERT_FALSE(c_iter_->Valid());
-}
-
 TEST_P(CompactionIteratorTest, SimpleRangeDeletion) {
   InitIterators({test::KeyStr("morning", 5, kTypeValue),
                  test::KeyStr("morning", 2, kTypeValue),
@@ -450,31 +431,6 @@ TEST_P(CompactionIteratorTest, RangeDeletionWithSnapshots) {
   ASSERT_FALSE(c_iter_->Valid());
 }
 
-// Tests compatibility of TimedPut and Range delete. TimedPut should act as if
-// it's a Put.
-TEST_P(CompactionIteratorTest, TimedPutAndRangeDeletion) {
-  InitIterators(
-      {test::KeyStr("morning", 5, kTypeValuePreferredSeqno),
-       test::KeyStr("morning", 2, kTypeValuePreferredSeqno),
-       test::KeyStr("night", 3, kTypeValuePreferredSeqno)},
-      {ValueWithPreferredSeqno("zao5"), ValueWithPreferredSeqno("zao2"),
-       ValueWithPreferredSeqno("wan")},
-      {test::KeyStr("ma", 4, kTypeRangeDeletion)}, {"mz"}, 5);
-  c_iter_->SeekToFirst();
-  ASSERT_TRUE(c_iter_->Valid());
-  ASSERT_EQ(test::KeyStr("morning", 5, kTypeValuePreferredSeqno),
-            c_iter_->key().ToString());
-  ASSERT_EQ(ValueWithPreferredSeqno("zao5"), c_iter_->value().ToString());
-  c_iter_->Next();
-  ASSERT_TRUE(c_iter_->Valid());
-  ASSERT_EQ(test::KeyStr("night", 3, kTypeValuePreferredSeqno),
-            c_iter_->key().ToString());
-  ASSERT_EQ(ValueWithPreferredSeqno("wan"), c_iter_->value().ToString());
-  c_iter_->Next();
-  ASSERT_OK(c_iter_->status());
-  ASSERT_FALSE(c_iter_->Valid());
-}
-
 TEST_P(CompactionIteratorTest, CompactionFilterSkipUntil) {
   class Filter : public CompactionFilter {
     Decision FilterV2(int /*level*/, const Slice& key, ValueType t,
@@ -546,11 +502,9 @@ TEST_P(CompactionIteratorTest, CompactionFilterSkipUntil) {
        test::KeyStr("f", 25, kTypeValue), test::KeyStr("g", 90, kTypeValue),
        test::KeyStr("h", 91, kTypeValue),  // keep
        test::KeyStr("i", 95, kTypeMerge),  // skip to "z"
-       test::KeyStr("j", 99, kTypeValue),
-       test::KeyStr("k", 100, kTypeValuePreferredSeqno)},
+       test::KeyStr("j", 99, kTypeValue)},
       {"av50", "am45", "bv60", "bv40", "cv35", "dm70", "em71", "fm65", "fm30",
-       "fv25", "gv90", "hv91", "im95", "jv99",
-       ValueWithPreferredSeqno("kv100")},
+       "fv25", "gv90", "hv91", "im95", "jv99"},
       {}, {}, kMaxSequenceNumber, kMaxSequenceNumber, &merge_op, &filter);
 
   // Compaction should output just "a", "e" and "h" keys.
@@ -660,87 +614,87 @@ TEST_P(CompactionIteratorTest, ShuttingDownInMerge) {
   EXPECT_EQ(2, filter.last_seen.load());
 }
 
-class Filter : public CompactionFilter {
-  Decision FilterV2(int /*level*/, const Slice& key, ValueType t,
-                    const Slice& existing_value, std::string* /*new_value*/,
-                    std::string* /*skip_until*/) const override {
-    std::string k = key.ToString();
-    std::string v = existing_value.ToString();
-
-    // See InitIterators() call below for the sequence of keys and their
-    // filtering decisions. Here we closely assert that compaction filter is
-    // called with the expected keys and only them, and with the right values.
-    if (k == "a") {
-      EXPECT_EQ(ValueType::kMergeOperand, t);
-      EXPECT_EQ("av1", v);
-      return Decision::kKeep;
-    } else if (k == "b") {
-      EXPECT_EQ(ValueType::kMergeOperand, t);
-      return Decision::kKeep;
-    } else if (k == "c") {
-      return Decision::kKeep;
-    }
-
-    ADD_FAILURE();
-    return Decision::kKeep;
-  }
-
-  const char* Name() const override {
-    return "CompactionIteratorTest.SingleMergeOperand::Filter";
-  }
-};
-
-class SingleMergeOp : public MergeOperator {
- public:
-  bool FullMergeV2(const MergeOperationInput& merge_in,
-                   MergeOperationOutput* merge_out) const override {
-    // See InitIterators() call below for why "c" is the only key for which
-    // FullMergeV2 should be called.
-    EXPECT_EQ("c", merge_in.key.ToString());
-
-    std::string temp_value;
-    if (merge_in.existing_value != nullptr) {
-      temp_value = merge_in.existing_value->ToString();
-    }
-
-    for (auto& operand : merge_in.operand_list) {
-      temp_value.append(operand.ToString());
-    }
-    merge_out->new_value = temp_value;
-
-    return true;
-  }
-
-  bool PartialMergeMulti(const Slice& key,
-                         const std::deque<Slice>& operand_list,
-                         std::string* new_value,
-                         Logger* /*logger*/) const override {
-    std::string string_key = key.ToString();
-    EXPECT_TRUE(string_key == "a" || string_key == "b");
-
-    if (string_key == "a") {
-      EXPECT_EQ(1, operand_list.size());
-    } else if (string_key == "b") {
-      EXPECT_EQ(2, operand_list.size());
-    }
-
-    std::string temp_value;
-    for (auto& operand : operand_list) {
-      temp_value.append(operand.ToString());
-    }
-    swap(temp_value, *new_value);
-
-    return true;
-  }
-
-  const char* Name() const override {
-    return "CompactionIteratorTest SingleMergeOp";
-  }
-
-  bool AllowSingleOperand() const override { return true; }
-};
-
 TEST_P(CompactionIteratorTest, SingleMergeOperand) {
+  class Filter : public CompactionFilter {
+    Decision FilterV2(int /*level*/, const Slice& key, ValueType t,
+                      const Slice& existing_value, std::string* /*new_value*/,
+                      std::string* /*skip_until*/) const override {
+      std::string k = key.ToString();
+      std::string v = existing_value.ToString();
+
+      // See InitIterators() call below for the sequence of keys and their
+      // filtering decisions. Here we closely assert that compaction filter is
+      // called with the expected keys and only them, and with the right values.
+      if (k == "a") {
+        EXPECT_EQ(ValueType::kMergeOperand, t);
+        EXPECT_EQ("av1", v);
+        return Decision::kKeep;
+      } else if (k == "b") {
+        EXPECT_EQ(ValueType::kMergeOperand, t);
+        return Decision::kKeep;
+      } else if (k == "c") {
+        return Decision::kKeep;
+      }
+
+      ADD_FAILURE();
+      return Decision::kKeep;
+    }
+
+    const char* Name() const override {
+      return "CompactionIteratorTest.SingleMergeOperand::Filter";
+    }
+  };
+
+  class SingleMergeOp : public MergeOperator {
+   public:
+    bool FullMergeV2(const MergeOperationInput& merge_in,
+                     MergeOperationOutput* merge_out) const override {
+      // See InitIterators() call below for why "c" is the only key for which
+      // FullMergeV2 should be called.
+      EXPECT_EQ("c", merge_in.key.ToString());
+
+      std::string temp_value;
+      if (merge_in.existing_value != nullptr) {
+        temp_value = merge_in.existing_value->ToString();
+      }
+
+      for (auto& operand : merge_in.operand_list) {
+        temp_value.append(operand.ToString());
+      }
+      merge_out->new_value = temp_value;
+
+      return true;
+    }
+
+    bool PartialMergeMulti(const Slice& key,
+                           const std::deque<Slice>& operand_list,
+                           std::string* new_value,
+                           Logger* /*logger*/) const override {
+      std::string string_key = key.ToString();
+      EXPECT_TRUE(string_key == "a" || string_key == "b");
+
+      if (string_key == "a") {
+        EXPECT_EQ(1, operand_list.size());
+      } else if (string_key == "b") {
+        EXPECT_EQ(2, operand_list.size());
+      }
+
+      std::string temp_value;
+      for (auto& operand : operand_list) {
+        temp_value.append(operand.ToString());
+      }
+      swap(temp_value, *new_value);
+
+      return true;
+    }
+
+    const char* Name() const override {
+      return "CompactionIteratorTest SingleMergeOp";
+    }
+
+    bool AllowSingleOperand() const override { return true; }
+  };
+
   SingleMergeOp merge_op;
   Filter filter;
   InitIterators(
@@ -765,24 +719,6 @@ TEST_P(CompactionIteratorTest, SingleMergeOperand) {
   ASSERT_EQ("cv1cv2", c_iter_->value().ToString());
 }
 
-// Tests compatibility of TimedPut and Merge operation. When a TimedPut is
-// merged with some merge operand in compaction, it will become a regular Put
-// and lose its preferred sequence number.
-TEST_P(CompactionIteratorTest, TimedPutAndMerge) {
-  SingleMergeOp merge_op;
-  Filter filter;
-  InitIterators({test::KeyStr("c", 90, kTypeMerge),
-                 test::KeyStr("c", 80, kTypeValuePreferredSeqno)},
-                {"cv2", ValueWithPreferredSeqno("cv1")}, {}, {},
-                kMaxSequenceNumber, kMaxSequenceNumber, &merge_op, &filter);
-
-  c_iter_->SeekToFirst();
-  ASSERT_TRUE(c_iter_->Valid());
-  ASSERT_EQ(test::KeyStr("c", 90, kTypeValue), c_iter_->key().ToString());
-  ASSERT_OK(c_iter_->status());
-  ASSERT_EQ("cv1cv2", c_iter_->value().ToString());
-}
-
 // In bottommost level, values earlier than earliest snapshot can be output
 // with sequence = 0.
 TEST_P(CompactionIteratorTest, ZeroOutSequenceAtBottomLevel) {
@@ -1027,22 +963,6 @@ TEST_F(CompactionIteratorWithSnapshotCheckerTest, DedupSameSnapshot_Value) {
       {"v4", "v3", "v1"}, 3 /*last_committed_seq*/);
 }
 
-TEST_F(CompactionIteratorWithSnapshotCheckerTest, DedupSameSnapshot_TimedPut) {
-  AddSnapshot(2, 1);
-  RunTest({test::KeyStr("foo", 4, kTypeValuePreferredSeqno),
-           test::KeyStr("foo", 3, kTypeValuePreferredSeqno),
-           test::KeyStr("foo", 2, kTypeValuePreferredSeqno),
-           test::KeyStr("foo", 1, kTypeValuePreferredSeqno)},
-          {ValueWithPreferredSeqno("v4"), ValueWithPreferredSeqno("v3"),
-           ValueWithPreferredSeqno("v2"), ValueWithPreferredSeqno("v1")},
-          {test::KeyStr("foo", 4, kTypeValuePreferredSeqno),
-           test::KeyStr("foo", 3, kTypeValuePreferredSeqno),
-           test::KeyStr("foo", 1, kTypeValuePreferredSeqno)},
-          {ValueWithPreferredSeqno("v4"), ValueWithPreferredSeqno("v3"),
-           ValueWithPreferredSeqno("v1")},
-          3 /*last_committed_seq*/);
-}
-
 TEST_F(CompactionIteratorWithSnapshotCheckerTest, DedupSameSnapshot_Deletion) {
   AddSnapshot(2, 1);
   RunTest(
@@ -1208,149 +1128,6 @@ TEST_F(CompactionIteratorWithSnapshotCheckerTest,
           2 /* earliest_write_conflict_snapshot */);
 }
 
-// Same as above but with a value with preferred seqno entry. In addition to the
-// value getting trimmed, the type of the KV is changed to kTypeValue.
-TEST_F(CompactionIteratorWithSnapshotCheckerTest,
-       KeepSingleDeletionForWriteConflictChecking_TimedPut) {
-  AddSnapshot(2, 0);
-  RunTest({test::KeyStr("a", 2, kTypeSingleDeletion),
-           test::KeyStr("a", 1, kTypeValuePreferredSeqno)},
-          {"", ValueWithPreferredSeqno("v1")},
-          {test::KeyStr("a", 2, kTypeSingleDeletion),
-           test::KeyStr("a", 1, kTypeValue)},
-          {"", ""}, 2 /* last_committed_seq */, nullptr /* merge_operator */,
-          nullptr /* compaction_filter */, false /* bottommost_level */,
-          2 /* earliest_write_conflict_snapshot */);
-}
-
-// Tests when a kTypeValuePreferredSeqno entry can have its preferred sequence
-// number swapped in. The required and sufficient conditions for an entry's
-// preferred sequence number to get swapped in are:
-// 1) The entry is visible to the earliest snapshot, AND
-// 2) No more entries with the same user key on lower levels, AND
-//    This is either because:
-//    2a) This is a compaction to the bottommost level, OR
-//    2b) Keys do not exist beyond output level
-// 3) The entry will not resurface a range deletion entry after swapping in the
-//    preferred sequence number.
-TEST_F(CompactionIteratorWithSnapshotCheckerTest,
-       TimedPut_NotVisibleToEarliestSnapshot_NoSwapPreferredSeqno) {
-  AddSnapshot(3);
-  RunTest({test::KeyStr("bar", 5, kTypeValuePreferredSeqno)},
-          {ValueWithPreferredSeqno("bv2", 2)},
-          {test::KeyStr("bar", 5, kTypeValuePreferredSeqno)},
-          {ValueWithPreferredSeqno("bv2", 2), "bv1"}, 5 /*last_committed_seq*/,
-          nullptr /*merge_operator*/, nullptr /*compaction_filter*/,
-          true /*bottommost_level*/,
-          kMaxSequenceNumber /*earliest_write_conflict_snapshot*/,
-          true /*key_not_exists_beyond_output_level*/);
-}
-
-TEST_F(CompactionIteratorWithSnapshotCheckerTest,
-       TimedPut_MoreEntriesInLowerLevels_NoSwapPreferredSeqno) {
-  // This tests mimics more entries in lower levels with `bottommost_level` and
-  // `key_not_exists_beyond_output_level` set to false.
-  RunTest({test::KeyStr("bar", 5, kTypeValuePreferredSeqno)},
-          {ValueWithPreferredSeqno("bv2", 2)},
-          {test::KeyStr("bar", 5, kTypeValuePreferredSeqno)},
-          {ValueWithPreferredSeqno("bv2", 2)}, 5 /*last_committed_seq*/,
-          nullptr /*merge_operator*/, nullptr /*compaction_filter*/,
-          false /*bottommost_level*/,
-          kMaxSequenceNumber /*earliest_write_conflict_snapshot*/,
-          false /*key_not_exists_beyond_output_level*/);
-}
-
-TEST_F(CompactionIteratorWithSnapshotCheckerTest,
-       TimedPut_ShouldBeCoverredByRangeDeletionBeforeSwap_NoOutput) {
-  InitIterators({test::KeyStr("morning", 5, kTypeValuePreferredSeqno),
-                 test::KeyStr("morning", 2, kTypeValuePreferredSeqno),
-                 test::KeyStr("night", 6, kTypeValue)},
-                {ValueWithPreferredSeqno("zao", 3),
-                 ValueWithPreferredSeqno("zao", 1), "wan"},
-                {test::KeyStr("ma", 6, kTypeRangeDeletion)}, {"mz"}, 6,
-                kMaxSequenceNumber /*last_committed_sequence*/,
-                nullptr /*merge_op*/, nullptr /*filter*/,
-                false /*bottommost_level*/,
-                kMaxSequenceNumber /*earliest_write_conflict_snapshot*/,
-                true /*key_not_exists_beyond_output_level*/);
-  c_iter_->SeekToFirst();
-  ASSERT_TRUE(c_iter_->Valid());
-  ASSERT_EQ(test::KeyStr("night", 6, kTypeValue), c_iter_->key().ToString());
-  ASSERT_EQ("wan", c_iter_->value().ToString());
-  c_iter_->Next();
-  ASSERT_FALSE(c_iter_->Valid());
-  ASSERT_OK(c_iter_->status());
-}
-
-TEST_F(CompactionIteratorWithSnapshotCheckerTest,
-       TimedPut_WillBeHiddenByRangeDeletionAfterSwap_NoSwap) {
-  InitIterators({test::KeyStr("morning", 5, kTypeValuePreferredSeqno),
-                 test::KeyStr("night", 6, kTypeValue)},
-                {ValueWithPreferredSeqno("zao", 3), "wan"},
-                {test::KeyStr("ma", 4, kTypeRangeDeletion)}, {"mz"}, 6,
-                kMaxSequenceNumber /*last_committed_sequence*/,
-                nullptr /*merge_op*/, nullptr /*filter*/,
-                false /*bottommost_level*/,
-                kMaxSequenceNumber /*earliest_write_conflict_snapshot*/,
-                true /*key_not_exists_beyond_output_level*/);
-  c_iter_->SeekToFirst();
-  ASSERT_TRUE(c_iter_->Valid());
-  ASSERT_EQ(test::KeyStr("morning", 5, kTypeValuePreferredSeqno),
-            c_iter_->key().ToString());
-  ASSERT_EQ(ValueWithPreferredSeqno("zao", 3), c_iter_->value().ToString());
-  c_iter_->Next();
-  ASSERT_TRUE(c_iter_->Valid());
-  ASSERT_EQ(test::KeyStr("night", 6, kTypeValue), c_iter_->key().ToString());
-  ASSERT_EQ("wan", c_iter_->value().ToString());
-  c_iter_->Next();
-  ASSERT_FALSE(c_iter_->Valid());
-  ASSERT_OK(c_iter_->status());
-}
-
-TEST_F(CompactionIteratorWithSnapshotCheckerTest,
-       TimedPut_BottomMostLevelVisibleToEarliestSnapshot_SwapPreferredSeqno) {
-  // Preferred seqno got swapped in and also zeroed out as a bottommost level
-  // optimization.
-  RunTest(
-      {test::KeyStr("bar", 5, kTypeValuePreferredSeqno),
-       test::KeyStr("bar", 4, kTypeValuePreferredSeqno),
-       test::KeyStr("foo", 6, kTypeValue)},
-      {ValueWithPreferredSeqno("bv2", 2), ValueWithPreferredSeqno("bv1", 1),
-       "fv1"},
-      {test::KeyStr("bar", 0, kTypeValue), test::KeyStr("foo", 0, kTypeValue)},
-      {"bv2", "fv1"}, 6 /*last_committed_seq*/, nullptr /*merge_operator*/,
-      nullptr /*compaction_filter*/, true /*bottommost_level*/);
-}
-
-TEST_F(
-    CompactionIteratorWithSnapshotCheckerTest,
-    TimedPut_NonBottomMostLevelVisibleToEarliestSnapshot_SwapPreferredSeqno) {
-  RunTest(
-      {test::KeyStr("bar", 5, kTypeValuePreferredSeqno),
-       test::KeyStr("bar", 4, kTypeValuePreferredSeqno),
-       test::KeyStr("foo", 6, kTypeValue)},
-      {ValueWithPreferredSeqno("bv2", 2), ValueWithPreferredSeqno("bv1", 1),
-       "fv1"},
-      {test::KeyStr("bar", 2, kTypeValue), test::KeyStr("foo", 6, kTypeValue)},
-      {"bv2", "fv1"}, 6 /*last_committed_seq*/, nullptr /*merge_operator*/,
-      nullptr /*compaction_filter*/, false /*bottommost_level*/,
-      kMaxSequenceNumber /*earliest_write_conflict_snapshot*/,
-      true /*key_not_exists_beyond_output_level*/);
-}
-
-TEST_F(CompactionIteratorWithSnapshotCheckerTest,
-       TimedPut_SequenceNumberAlreadyZeroedOut_ChangeType) {
-  RunTest(
-      {test::KeyStr("bar", 0, kTypeValuePreferredSeqno),
-       test::KeyStr("bar", 0, kTypeValuePreferredSeqno),
-       test::KeyStr("foo", 0, kTypeValue)},
-      {ValueWithPreferredSeqno("bv2", 2), ValueWithPreferredSeqno("bv1", 1),
-       "fv1"},
-      {test::KeyStr("bar", 0, kTypeValue), test::KeyStr("foo", 0, kTypeValue)},
-      {"bv2", "fv1"}, 6 /*last_committed_seq*/, nullptr /*merge_operator*/,
-      nullptr /*compaction_filter*/, true /*bottommost_level*/);
-}
-
 // Compaction filter should keep uncommitted key as-is, and
 //   * Convert the latest value to deletion, and/or
 //   * if latest value is a merge, apply filter to all subsequent merges.
@@ -1368,22 +1145,6 @@ TEST_F(CompactionIteratorWithSnapshotCheckerTest, CompactionFilter_Value) {
       nullptr /*merge_operator*/, compaction_filter.get());
 }
 
-TEST_F(CompactionIteratorWithSnapshotCheckerTest, CompactionFilter_TimedPut) {
-  // TODO(yuzhangyu): Add support for this type in compaction filter.
-  // Type kTypeValuePreferredSeqno is not explicitly exposed in the compaction
-  // filter API, so users can not operate on it through compaction filter API
-  // to remove/purge/change value etc. But this type of entry can be impacted by
-  // other entries' filter result, currently only kRemoveAndSkip type of result
-  // can affect it.
-  std::unique_ptr<CompactionFilter> compaction_filter(
-      new FilterAllKeysCompactionFilter());
-  RunTest({test::KeyStr("a", 2, kTypeValuePreferredSeqno)},
-          {ValueWithPreferredSeqno("v1")},
-          {test::KeyStr("a", 2, kTypeValuePreferredSeqno)},
-          {ValueWithPreferredSeqno("v1")}, 2 /*last_committed_seq*/,
-          nullptr /*merge_operator*/, compaction_filter.get());
-}
-
 TEST_F(CompactionIteratorWithSnapshotCheckerTest, CompactionFilter_Deletion) {
   std::unique_ptr<CompactionFilter> compaction_filter(
       new FilterAllKeysCompactionFilter());

db/compaction/compaction_job.cc

@@ -288,37 +288,39 @@ 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_to_time_mapping_
+    seqno_to_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_to_time_mapping_.Add(tp->seqno_to_time_mapping)
+              .PermitUncheckedError();
+          seqno_to_time_mapping_.Add(fmd->fd.smallest_seqno,
+                                     fmd->oldest_ancester_time);
         }
       }
     }
 
+    auto status = seqno_to_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_to_time_mapping_.TruncateOldEntries(_current_time);
       uint64_t preserve_time =
           static_cast<uint64_t>(_current_time) > preserve_time_duration
               ? _current_time - preserve_time_duration
@@ -342,16 +344,6 @@ void CompactionJob::Prepare() {
             1;
       }
     }
-    // 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);
   }
 }
 
@@ -404,9 +396,7 @@ void CompactionJob::AcquireSubcompactionResources(
 
 void CompactionJob::ShrinkSubcompactionResources(uint64_t num_extra_resources) {
   // Do nothing when we have zero resources to shrink
-  if (num_extra_resources == 0) {
-    return;
-  }
+  if (num_extra_resources == 0) return;
   db_mutex_->Lock();
   // We cannot release threads more than what we reserved before
   int extra_num_subcompaction_threads_released = env_->ReleaseThreads(
@@ -452,6 +442,14 @@ 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.
@@ -478,8 +476,7 @@ 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 &&
@@ -586,9 +583,7 @@ void CompactionJob::GenSubcompactionBoundaries() {
 
   TEST_SYNC_POINT_CALLBACK("CompactionJob::GenSubcompactionBoundaries:0",
                            &num_planned_subcompactions);
-  if (num_planned_subcompactions == 1) {
-    return;
-  }
+  if (num_planned_subcompactions == 1) return;
 
   // Group the ranges into subcompactions
   uint64_t target_range_size = std::max(
@@ -645,7 +640,7 @@ Status CompactionJob::Run() {
 
   // Always schedule the first subcompaction (whether or not there are also
   // others) in the current thread to be efficient with resources
-  ProcessKeyValueCompaction(compact_->sub_compact_states.data());
+  ProcessKeyValueCompaction(&compact_->sub_compact_states[0]);
 
   // Wait for all other threads (if there are any) to finish execution
   for (auto& thread : thread_pool) {
@@ -733,9 +728,8 @@ 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,
@@ -756,6 +750,7 @@ 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());
@@ -826,27 +821,24 @@ 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.");
       }
     }
   }
@@ -1138,9 +1130,6 @@ 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();
@@ -1275,17 +1264,18 @@ 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_;
@@ -1333,7 +1323,8 @@ 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()
@@ -1364,9 +1355,10 @@ 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;
@@ -1718,8 +1710,6 @@ 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);
 
@@ -1802,9 +1792,8 @@ 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);
 }
 
@@ -1854,14 +1843,13 @@ Status CompactionJob::OpenCompactionOutputFile(SubcompactionState* sub_compact,
   // Pass temperature of the last level files to FileSystem.
   FileOptions fo_copy = file_options_;
   Temperature temperature = sub_compact->compaction->output_temperature();
-  Temperature last_level_temp =
-      sub_compact->compaction->mutable_cf_options()->last_level_temperature;
-  // Here last_level_temperature supersedes default_write_temperature, when
-  // enabled and applicable
-  if (last_level_temp != Temperature::kUnknown &&
+  // only set for the last level compaction and also it's not output to
+  // penultimate level (when preclude_last_level feature is enabled)
+  if (temperature == Temperature::kUnknown &&
       sub_compact->compaction->is_last_level() &&
       !sub_compact->IsCurrentPenultimateLevel()) {
-    temperature = last_level_temp;
+    temperature =
+        sub_compact->compaction->mutable_cf_options()->last_level_temperature;
   }
   fo_copy.temperature = temperature;
 
@@ -1941,6 +1929,8 @@ 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_);
   }
 
@@ -1953,17 +1943,13 @@ 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(),

db/compaction/compaction_job.h

@@ -41,6 +41,7 @@
 #include "rocksdb/env.h"
 #include "rocksdb/memtablerep.h"
 #include "rocksdb/transaction_log.h"
+#include "table/scoped_arena_iterator.h"
 #include "util/autovector.h"
 #include "util/stop_watch.h"
 #include "util/thread_local.h"

db/compaction/compaction_job_stats_test.cc

@@ -46,6 +46,7 @@
 #include "table/block_based/block_based_table_factory.h"
 #include "table/mock_table.h"
 #include "table/plain/plain_table_factory.h"
+#include "table/scoped_arena_iterator.h"
 #include "test_util/sync_point.h"
 #include "test_util/testharness.h"
 #include "test_util/testutil.h"
@@ -130,7 +131,7 @@ class CompactionJobStatsTest : public testing::Test,
     ColumnFamilyOptions cf_opts(options);
     size_t cfi = handles_.size();
     handles_.resize(cfi + cfs.size());
-    for (const auto& cf : cfs) {
+    for (auto cf : cfs) {
       ASSERT_OK(db_->CreateColumnFamily(cf_opts, cf, &handles_[cfi++]));
     }
   }
@@ -159,7 +160,7 @@ class CompactionJobStatsTest : public testing::Test,
     EXPECT_EQ(cfs.size(), options.size());
     std::vector<ColumnFamilyDescriptor> column_families;
     for (size_t i = 0; i < cfs.size(); ++i) {
-      column_families.emplace_back(cfs[i], options[i]);
+      column_families.push_back(ColumnFamilyDescriptor(cfs[i], options[i]));
     }
     DBOptions db_opts = DBOptions(options[0]);
     return DB::Open(db_opts, dbname_, column_families, &handles_, &db_);

db/compaction/compaction_job_test.cc

@@ -217,7 +217,7 @@ class CompactionJobTestBase : public testing::Test {
             /*block_cache_tracer=*/nullptr,
             /*io_tracer=*/nullptr, /*db_id=*/"", /*db_session_id=*/"",
             /*daily_offpeak_time_utc=*/"",
-            /*error_handler=*/nullptr, /*read_only=*/false)),
+            /*error_handler=*/nullptr)),
         shutting_down_(false),
         mock_table_factory_(new mock::MockTableFactory()),
         error_handler_(nullptr, db_options_, &mutex_),
@@ -295,20 +295,17 @@ 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 */),
             file_writer.get()));
     // Build table.
-    for (const auto& kv : contents) {
+    for (auto kv : contents) {
       std::string key;
       std::string value;
       std::tie(key, value) = kv;
@@ -327,7 +324,7 @@ class CompactionJobTestBase : public testing::Test {
     SequenceNumber smallest_seqno = kMaxSequenceNumber;
     SequenceNumber largest_seqno = 0;
     uint64_t oldest_blob_file_number = kInvalidBlobFileNumber;
-    for (const auto& kv : contents) {
+    for (auto kv : contents) {
       ParsedInternalKey key;
       std::string skey;
       std::string value;
@@ -397,7 +394,7 @@ 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();
   }
 
@@ -550,9 +547,9 @@ class CompactionJobTestBase : public testing::Test {
         &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));
+        /*error_handler=*/nullptr));
     compaction_job_stats_.Reset();
-    ASSERT_OK(SetIdentityFile(WriteOptions(), env_, dbname_));
+    ASSERT_OK(SetIdentityFile(env_, dbname_));
 
     VersionEdit new_db;
     new_db.SetLogNumber(0);
@@ -571,11 +568,11 @@ 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 +736,6 @@ 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_;
@@ -1473,7 +1469,7 @@ TEST_F(CompactionJobTest, OldestBlobFileNumber) {
 }
 
 TEST_F(CompactionJobTest, VerifyPenultimateLevelOutput) {
-  cf_options_.last_level_temperature = Temperature::kCold;
+  cf_options_.bottommost_temperature = Temperature::kCold;
   SyncPoint::GetInstance()->SetCallBack(
       "Compaction::SupportsPerKeyPlacement:Enabled", [&](void* arg) {
         auto supports_per_key_placement = static_cast<bool*>(arg);
@@ -1752,9 +1748,23 @@ 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 +1838,10 @@ 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 +1876,28 @@ 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 +1964,40 @@ 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 +2034,13 @@ 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 +2049,9 @@ 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 +2062,20 @@ 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 +2132,9 @@ 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()

db/compaction/compaction_outputs.cc

@@ -25,13 +25,11 @@ Status CompactionOutputs::Finish(
   assert(meta != nullptr);
   Status s = intput_status;
   if (s.ok()) {
-    SeqnoToTimeMapping relevant_mapping;
-    relevant_mapping.CopyFromSeqnoRange(
-        seqno_to_time_mapping,
-        std::min(smallest_preferred_seqno_, meta->fd.smallest_seqno),
-        meta->fd.largest_seqno);
-    relevant_mapping.SetCapacity(kMaxSeqnoTimePairsPerSST);
-    builder_->SetSeqnoTimeTableProperties(relevant_mapping,
+    std::string seqno_to_time_mapping_str;
+    seqno_to_time_mapping.Encode(
+        seqno_to_time_mapping_str, meta->fd.smallest_seqno,
+        meta->fd.largest_seqno, meta->file_creation_time);
+    builder_->SetSeqnoTimeTableProperties(seqno_to_time_mapping_str,
                                           meta->oldest_ancester_time);
     s = builder_->Finish();
 
@@ -64,15 +62,12 @@ 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()) {
@@ -322,6 +317,7 @@ 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 >
@@ -343,6 +339,7 @@ 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,
@@ -424,11 +421,6 @@ Status CompactionOutputs::AddToOutput(
   }
 
   const ParsedInternalKey& ikey = c_iter.ikey();
-  if (ikey.type == kTypeValuePreferredSeqno) {
-    SequenceNumber preferred_seqno = ParsePackedValueForSeqno(value);
-    smallest_preferred_seqno_ =
-        std::min(smallest_preferred_seqno_, preferred_seqno);
-  }
   s = current_output().meta.UpdateBoundaries(key, value, ikey.sequence,
                                              ikey.type);
 

db/compaction/compaction_outputs.h

@@ -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
@@ -297,7 +299,6 @@ class CompactionOutputs {
   std::unique_ptr<TableBuilder> builder_;
   std::unique_ptr<WritableFileWriter> file_writer_;
   uint64_t current_output_file_size_ = 0;
-  SequenceNumber smallest_preferred_seqno_ = kMaxSequenceNumber;
 
   // all the compaction outputs so far
   std::vector<Output> outputs_;

db/compaction/compaction_picker.cc

@@ -130,7 +130,7 @@ CompactionPicker::CompactionPicker(const ImmutableOptions& ioptions,
                                    const InternalKeyComparator* icmp)
     : ioptions_(ioptions), icmp_(icmp) {}
 
-CompactionPicker::~CompactionPicker() = default;
+CompactionPicker::~CompactionPicker() {}
 
 // Delete this compaction from the list of running compactions.
 void CompactionPicker::ReleaseCompactionFiles(Compaction* c, Status status) {
@@ -340,8 +340,8 @@ Compaction* CompactionPicker::CompactFiles(
 #ifndef NDEBUG
   assert(input_files.size());
   // This compaction output should not overlap with a running compaction as
-  // `SanitizeAndConvertCompactionInputFiles` should've checked earlier and db
-  // mutex shouldn't have been released since.
+  // `SanitizeCompactionInputFiles` should've checked earlier and db mutex
+  // shouldn't have been released since.
   int start_level = Compaction::kInvalidLevel;
   for (const auto& in : input_files) {
     // input_files should already be sorted by level
@@ -377,8 +377,7 @@ Compaction* CompactionPicker::CompactFiles(
       output_level, compact_options.output_file_size_limit,
       mutable_cf_options.max_compaction_bytes, output_path_id, compression_type,
       GetCompressionOptions(mutable_cf_options, vstorage, output_level),
-      mutable_cf_options.default_write_temperature,
-      compact_options.max_subcompactions,
+      Temperature::kUnknown, compact_options.max_subcompactions,
       /* grandparents */ {}, true);
   RegisterCompaction(c);
   return c;
@@ -501,6 +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);
@@ -526,15 +526,10 @@ bool CompactionPicker::SetupOtherInputs(
     if (!ExpandInputsToCleanCut(cf_name, vstorage, &expanded_inputs)) {
       try_overlapping_inputs = false;
     }
-    // It helps to reduce write amp and avoid a further separate compaction
-    // to include more input level files without expanding output level files.
-    // So we apply a softer limit. We still need a limit to avoid overly large
-    // compactions and potential high space amp spikes.
-    const uint64_t limit =
-        MultiplyCheckOverflow(mutable_cf_options.max_compaction_bytes, 2.0);
     if (try_overlapping_inputs && expanded_inputs.size() > inputs->size() &&
-        !AreFilesInCompaction(expanded_inputs.files) &&
-        output_level_inputs_size + expanded_inputs_size < limit) {
+        (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);
       CompactionInputFiles expanded_output_level_inputs;
@@ -556,8 +551,9 @@ bool CompactionPicker::SetupOtherInputs(
                                              base_index, nullptr);
       expanded_inputs_size = TotalFileSize(expanded_inputs.files);
       if (expanded_inputs.size() > inputs->size() &&
-          !AreFilesInCompaction(expanded_inputs.files) &&
-          (output_level_inputs_size + expanded_inputs_size) < limit) {
+          (mutable_cf_options.ignore_max_compaction_bytes_for_input ||
+           output_level_inputs_size + expanded_inputs_size < limit) &&
+          !AreFilesInCompaction(expanded_inputs.files)) {
         expand_inputs = true;
       }
     }
@@ -674,8 +670,7 @@ Compaction* CompactionPicker::CompactRange(
         compact_range_options.target_path_id,
         GetCompressionType(vstorage, mutable_cf_options, output_level, 1),
         GetCompressionOptions(mutable_cf_options, vstorage, output_level),
-        mutable_cf_options.default_write_temperature,
-        compact_range_options.max_subcompactions,
+        Temperature::kUnknown, compact_range_options.max_subcompactions,
         /* grandparents */ {}, /* is manual */ true, trim_ts, /* score */ -1,
         /* deletion_compaction */ false, /* l0_files_might_overlap */ true,
         CompactionReason::kUnknown,
@@ -863,9 +858,8 @@ Compaction* CompactionPicker::CompactRange(
       GetCompressionType(vstorage, mutable_cf_options, output_level,
                          vstorage->base_level()),
       GetCompressionOptions(mutable_cf_options, vstorage, output_level),
-      mutable_cf_options.default_write_temperature,
-      compact_range_options.max_subcompactions, std::move(grandparents),
-      /* is manual */ true, trim_ts, /* score */ -1,
+      Temperature::kUnknown, compact_range_options.max_subcompactions,
+      std::move(grandparents), /* is manual */ true, trim_ts, /* score */ -1,
       /* deletion_compaction */ false, /* l0_files_might_overlap */ true,
       CompactionReason::kUnknown,
       compact_range_options.blob_garbage_collection_policy,
@@ -1040,18 +1034,19 @@ Status CompactionPicker::SanitizeCompactionInputFilesForAllLevels(
       }
     }
   }
+  if (RangeOverlapWithCompaction(smallestkey, largestkey, output_level)) {
+    return Status::Aborted(
+        "A running compaction is writing to the same output level in an "
+        "overlapping key range");
+  }
   return Status::OK();
 }
 
-Status CompactionPicker::SanitizeAndConvertCompactionInputFiles(
+Status CompactionPicker::SanitizeCompactionInputFiles(
     std::unordered_set<uint64_t>* input_files,
-    const ColumnFamilyMetaData& cf_meta, const int output_level,
-    const VersionStorageInfo* vstorage,
-    std::vector<CompactionInputFiles>* converted_input_files) const {
+    const ColumnFamilyMetaData& cf_meta, const int output_level) const {
   assert(static_cast<int>(cf_meta.levels.size()) - 1 ==
          cf_meta.levels[cf_meta.levels.size() - 1].level);
-  assert(converted_input_files);
-
   if (output_level >= static_cast<int>(cf_meta.levels.size())) {
     return Status::InvalidArgument(
         "Output level for column family " + cf_meta.name +
@@ -1077,6 +1072,7 @@ Status CompactionPicker::SanitizeAndConvertCompactionInputFiles(
 
   Status s = SanitizeCompactionInputFilesForAllLevels(input_files, cf_meta,
                                                       output_level);
+
   if (!s.ok()) {
     return s;
   }
@@ -1117,22 +1113,6 @@ Status CompactionPicker::SanitizeAndConvertCompactionInputFiles(
     }
   }
 
-  s = GetCompactionInputsFromFileNumbers(converted_input_files, input_files,
-                                         vstorage, CompactionOptions());
-  if (!s.ok()) {
-    return s;
-  }
-  assert(converted_input_files->size() > 0);
-  if (output_level != 0 &&
-      FilesRangeOverlapWithCompaction(
-          *converted_input_files, output_level,
-          Compaction::EvaluatePenultimateLevel(
-              vstorage, ioptions_, (*converted_input_files)[0].level,
-              output_level))) {
-    return Status::Aborted(
-        "A running compaction is writing to the same output level(s) in an "
-        "overlapping key range");
-  }
   return Status::OK();
 }
 

db/compaction/compaction_picker.h

@@ -86,20 +86,16 @@ class CompactionPicker {
 
   virtual bool NeedsCompaction(const VersionStorageInfo* vstorage) const = 0;
 
-  // Sanitize the input set of compaction input files and convert it to
-  // `std::vector<CompactionInputFiles>` in the output parameter
-  // `converted_input_files`.
-  // When the input parameters do not describe a valid
-  // compaction, the function will try to fix the input_files by adding
-  // necessary files.  If it's not possible to convert an invalid input_files
-  // into a valid one by adding more files, the function will return a
-  // non-ok status with specific reason.
-  //
-  Status SanitizeAndConvertCompactionInputFiles(
-      std::unordered_set<uint64_t>* input_files,
-      const ColumnFamilyMetaData& cf_meta, const int output_level,
-      const VersionStorageInfo* vstorage,
-      std::vector<CompactionInputFiles>* converted_input_files) const;
+// Sanitize the input set of compaction input files.
+// When the input parameters do not describe a valid compaction, the
+// function will try to fix the input_files by adding necessary
+// files.  If it's not possible to conver an invalid input_files
+// into a valid one by adding more files, the function will return a
+// non-ok status with specific reason.
+//
+  Status SanitizeCompactionInputFiles(std::unordered_set<uint64_t>* input_files,
+                                      const ColumnFamilyMetaData& cf_meta,
+                                      const int output_level) const;
 
   // Free up the files that participated in a compaction
   //
@@ -113,8 +109,8 @@ class CompactionPicker {
   // object.
   //
   // Caller must provide a set of input files that has been passed through
-  // `SanitizeAndConvertCompactionInputFiles` earlier. The lock should not be
-  // released between that call and this one.
+  // `SanitizeCompactionInputFiles` earlier. The lock should not be released
+  // between that call and this one.
   Compaction* CompactFiles(const CompactionOptions& compact_options,
                            const std::vector<CompactionInputFiles>& input_files,
                            int output_level, VersionStorageInfo* vstorage,
@@ -124,7 +120,6 @@ class CompactionPicker {
 
   // Converts a set of compaction input file numbers into
   // a list of CompactionInputFiles.
-  // TODO(hx235): remove the unused paramter `compact_options`
   Status GetCompactionInputsFromFileNumbers(
       std::vector<CompactionInputFiles>* input_files,
       std::unordered_set<uint64_t>* input_set,
@@ -230,8 +225,8 @@ class CompactionPicker {
  protected:
   const ImmutableOptions& ioptions_;
 
-  // A helper function to SanitizeAndConvertCompactionInputFiles() that
-  // sanitizes "input_files" by adding necessary files.
+// A helper function to SanitizeCompactionInputFiles() that
+// sanitizes "input_files" by adding necessary files.
   virtual Status SanitizeCompactionInputFilesForAllLevels(
       std::unordered_set<uint64_t>* input_files,
       const ColumnFamilyMetaData& cf_meta, const int output_level) const;
@@ -282,7 +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;
   }
 };

db/compaction/compaction_picker_fifo.cc

@@ -116,8 +116,7 @@ Compaction* FIFOCompactionPicker::PickTTLCompaction(
   Compaction* c = new Compaction(
       vstorage, ioptions_, mutable_cf_options, mutable_db_options,
       std::move(inputs), 0, 0, 0, 0, kNoCompression,
-      mutable_cf_options.compression_opts,
-      mutable_cf_options.default_write_temperature,
+      mutable_cf_options.compression_opts, Temperature::kUnknown,
       /* max_subcompactions */ 0, {}, /* is manual */ false,
       /* trim_ts */ "", vstorage->CompactionScore(0),
       /* is deletion compaction */ true, /* l0_files_might_overlap */ true,
@@ -186,8 +185,7 @@ Compaction* FIFOCompactionPicker::PickSizeCompaction(
             {comp_inputs}, 0, 16 * 1024 * 1024 /* output file size limit */,
             0 /* max compaction bytes, not applicable */,
             0 /* output path ID */, mutable_cf_options.compression,
-            mutable_cf_options.compression_opts,
-            mutable_cf_options.default_write_temperature,
+            mutable_cf_options.compression_opts, Temperature::kUnknown,
             0 /* max_subcompactions */, {}, /* is manual */ false,
             /* trim_ts */ "", vstorage->CompactionScore(0),
             /* is deletion compaction */ false,
@@ -282,8 +280,7 @@ Compaction* FIFOCompactionPicker::PickSizeCompaction(
       /* target_file_size */ 0,
       /* max_compaction_bytes */ 0,
       /* output_path_id */ 0, kNoCompression,
-      mutable_cf_options.compression_opts,
-      mutable_cf_options.default_write_temperature,
+      mutable_cf_options.compression_opts, Temperature::kUnknown,
       /* max_subcompactions */ 0, {}, /* is manual */ false,
       /* trim_ts */ "", vstorage->CompactionScore(0),
       /* is deletion compaction */ true,
@@ -417,7 +414,6 @@ Compaction* FIFOCompactionPicker::PickTemperatureChangeCompaction(
   if (inputs[0].files.empty()) {
     return nullptr;
   }
-  assert(compaction_target_temp != Temperature::kLastTemperature);
 
   Compaction* c = new Compaction(
       vstorage, ioptions_, mutable_cf_options, mutable_db_options,

db/compaction/compaction_picker_fifo.h

@@ -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,

db/compaction/compaction_picker_level.cc

@@ -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;
@@ -355,9 +328,7 @@ void LevelCompactionBuilder::SetupOtherFilesWithRoundRobinExpansion() {
   TEST_SYNC_POINT("LevelCompactionPicker::RoundRobin");
 
   // Only expand the inputs when we have selected a file in start_level_inputs_
-  if (start_level_inputs_.size() == 0) {
-    return;
-  }
+  if (start_level_inputs_.size() == 0) return;
 
   uint64_t start_lvl_bytes_no_compacting = 0;
   uint64_t curr_bytes_to_compact = 0;
@@ -553,7 +524,7 @@ Compaction* LevelCompactionBuilder::GetCompaction() {
       GetCompressionType(vstorage_, mutable_cf_options_, output_level_,
                          vstorage_->base_level()),
       GetCompressionOptions(mutable_cf_options_, vstorage_, output_level_),
-      mutable_cf_options_.default_write_temperature,
+      Temperature::kUnknown,
       /* max_subcompactions */ 0, std::move(grandparents_), is_manual_,
       /* trim_ts */ "", start_level_score_, false /* deletion_compaction */,
       l0_files_might_overlap, compaction_reason_);
@@ -787,9 +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;
   }
@@ -802,9 +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_);
@@ -909,56 +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(

db/compaction/compaction_picker_level.h

@@ -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

db/compaction/compaction_picker_test.cc

@@ -77,7 +77,7 @@ class CompactionPickerTestBase : public testing::Test {
     ioptions_.level_compaction_dynamic_level_bytes = false;
   }
 
-  ~CompactionPickerTestBase() override = default;
+  ~CompactionPickerTestBase() override {}
 
   void NewVersionStorage(int num_levels, CompactionStyle style) {
     DeleteVersionStorage();
@@ -214,7 +214,7 @@ class CompactionPickerTest : public CompactionPickerTestBase {
   explicit CompactionPickerTest()
       : CompactionPickerTestBase(BytewiseComparator()) {}
 
-  ~CompactionPickerTest() override = default;
+  ~CompactionPickerTest() override {}
 };
 
 class CompactionPickerU64TsTest : public CompactionPickerTestBase {
@@ -222,7 +222,7 @@ class CompactionPickerU64TsTest : public CompactionPickerTestBase {
   explicit CompactionPickerU64TsTest()
       : CompactionPickerTestBase(test::BytewiseComparatorWithU64TsWrapper()) {}
 
-  ~CompactionPickerU64TsTest() override = default;
+  ~CompactionPickerU64TsTest() override {}
 };
 
 TEST_F(CompactionPickerTest, Empty) {
@@ -1443,12 +1443,15 @@ 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 +1465,7 @@ 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());
 }
 
@@ -2465,60 +2468,20 @@ TEST_F(CompactionPickerTest, IsBottommostLevelTest) {
   DeleteVersionStorage();
 }
 
-TEST_F(CompactionPickerTest, CompactionLimitWhenAddFileFromInputLevel) {
+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 are eligible.
-  // File 6 is excluded since it overlaps with file 7.
-  // It can expand input level since in this case, the limit on compaction size
-  // is 2 * max_compaction_bytes.
+  // 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);
-  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(4U, 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());
-}
-
-TEST_F(CompactionPickerTest, HitCompactionLimitWhenAddFileFromInputLevel) {
-  mutable_cf_options_.max_bytes_for_level_base = 1000000u;
-  mutable_cf_options_.max_compaction_bytes = 800000u;
-  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 are eligible.
-  // File 6 is excluded since it overlaps with file 7.
-  // It can not expand input level since total compaction size hit the limit
-  // 2 * max_compaction_bytes.
-  Add(1, 1U, "100", "150", 400000U);
-  Add(1, 2U, "151", "200", 400001U, 0, 0);
-  Add(1, 3U, "201", "250", 400000U, 0, 0);
-  Add(1, 4U, "251", "300", 400000U, 0, 0);
-  Add(1, 6U, "325", "400", 400000U, 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(
@@ -2532,6 +2495,34 @@ TEST_F(CompactionPickerTest, HitCompactionLimitWhenAddFileFromInputLevel) {
   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(5U, compaction->input(1, 0)->fd.GetNumber());
+}
+
 TEST_F(CompactionPickerTest, IsTrivialMoveOn) {
   mutable_cf_options_.max_bytes_for_level_base = 10000u;
   mutable_cf_options_.max_compaction_bytes = 10001u;
@@ -4158,177 +4149,9 @@ 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
 

db/compaction/compaction_picker_universal.cc

@@ -870,7 +870,7 @@ Compaction* UniversalCompactionBuilder::PickCompactionToReduceSortedRuns(
                                            output_level, 1, enable_compression),
                         GetCompressionOptions(mutable_cf_options_, vstorage_,
                                               output_level, enable_compression),
-                        mutable_cf_options_.default_write_temperature,
+                        Temperature::kUnknown,
                         /* max_subcompactions */ 0, grandparents,
                         /* is manual */ false, /* trim_ts */ "", score_,
                         false /* deletion_compaction */,
@@ -1205,7 +1205,7 @@ Compaction* UniversalCompactionBuilder::PickIncrementalForReduceSizeAmp(
                          true /* enable_compression */),
       GetCompressionOptions(mutable_cf_options_, vstorage_, output_level,
                             true /* enable_compression */),
-      mutable_cf_options_.default_write_temperature,
+      Temperature::kUnknown,
       /* max_subcompactions */ 0, /* grandparents */ {}, /* is manual */ false,
       /* trim_ts */ "", score_, false /* deletion_compaction */,
       /* l0_files_might_overlap */ true,
@@ -1349,7 +1349,7 @@ Compaction* UniversalCompactionBuilder::PickDeleteTriggeredCompaction() {
       /* max_grandparent_overlap_bytes */ GetMaxOverlappingBytes(), path_id,
       GetCompressionType(vstorage_, mutable_cf_options_, output_level, 1),
       GetCompressionOptions(mutable_cf_options_, vstorage_, output_level),
-      mutable_cf_options_.default_write_temperature,
+      Temperature::kUnknown,
       /* max_subcompactions */ 0, grandparents, /* is manual */ false,
       /* trim_ts */ "", score_, false /* deletion_compaction */,
       /* l0_files_might_overlap */ true,
@@ -1442,7 +1442,7 @@ Compaction* UniversalCompactionBuilder::PickCompactionWithSortedRunRange(
                          true /* enable_compression */),
       GetCompressionOptions(mutable_cf_options_, vstorage_, output_level,
                             true /* enable_compression */),
-      mutable_cf_options_.default_write_temperature,
+      Temperature::kUnknown,
       /* max_subcompactions */ 0, /* grandparents */ {}, /* is manual */ false,
       /* trim_ts */ "", score_, false /* deletion_compaction */,
       /* l0_files_might_overlap */ true, compaction_reason);

db/compaction/compaction_picker_universal.h

@@ -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

db/compaction/compaction_service_job.cc

@@ -74,24 +74,24 @@ CompactionJob::ProcessKeyValueCompactionWithCompactionService(
       compaction_input.output_level, input_files_oss.str().c_str());
   CompactionServiceJobInfo info(dbname_, db_id_, db_session_id_,
                                 GetCompactionId(sub_compact), thread_pri_);
-  CompactionServiceScheduleResponse response =
-      db_options_.compaction_service->Schedule(info, compaction_input_binary);
-  switch (response.status) {
+  CompactionServiceJobStatus compaction_status =
+      db_options_.compaction_service->StartV2(info, compaction_input_binary);
+  switch (compaction_status) {
     case CompactionServiceJobStatus::kSuccess:
       break;
     case CompactionServiceJobStatus::kFailure:
       sub_compact->status = Status::Incomplete(
-          "CompactionService failed to schedule a remote compaction job.");
+          "CompactionService failed to start compaction job.");
       ROCKS_LOG_WARN(db_options_.info_log,
                      "[%s] [JOB %d] Remote compaction failed to start.",
                      compaction_input.column_family.name.c_str(), job_id_);
-      return response.status;
+      return compaction_status;
     case CompactionServiceJobStatus::kUseLocal:
       ROCKS_LOG_INFO(
           db_options_.info_log,
-          "[%s] [JOB %d] Remote compaction fallback to local by API (Schedule)",
+          "[%s] [JOB %d] Remote compaction fallback to local by API Start.",
           compaction_input.column_family.name.c_str(), job_id_);
-      return response.status;
+      return compaction_status;
     default:
       assert(false);  // unknown status
       break;
@@ -101,15 +101,14 @@ CompactionJob::ProcessKeyValueCompactionWithCompactionService(
                  "[%s] [JOB %d] Waiting for remote compaction...",
                  compaction_input.column_family.name.c_str(), job_id_);
   std::string compaction_result_binary;
-  CompactionServiceJobStatus compaction_status =
-      db_options_.compaction_service->Wait(response.scheduled_job_id,
-                                           &compaction_result_binary);
+  compaction_status = db_options_.compaction_service->WaitForCompleteV2(
+      info, &compaction_result_binary);
 
   if (compaction_status == CompactionServiceJobStatus::kUseLocal) {
-    ROCKS_LOG_INFO(
-        db_options_.info_log,
-        "[%s] [JOB %d] Remote compaction fallback to local by API (Wait)",
-        compaction_input.column_family.name.c_str(), job_id_);
+    ROCKS_LOG_INFO(db_options_.info_log,
+                   "[%s] [JOB %d] Remote compaction fallback to local by API "
+                   "WaitForComplete.",
+                   compaction_input.column_family.name.c_str(), job_id_);
     return compaction_status;
   }
 
@@ -196,8 +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(
@@ -831,3 +830,4 @@ bool CompactionServiceInput::TEST_Equals(CompactionServiceInput* other,
 }
 #endif  // NDEBUG
 }  // namespace ROCKSDB_NAMESPACE
+