rocksdb/CMakeLists.txt

1062 lines
36 KiB
CMake
Raw Normal View History

# Prerequisites for Windows:
# This cmake build is for Windows 64-bit only.
#
# Prerequisites:
# You must have at least Visual Studio 2015 Update 3. Start the Developer Command Prompt window that is a part of Visual Studio installation.
# Run the build commands from within the Developer Command Prompt window to have paths to the compiler and runtime libraries set.
# You must have git.exe in your %PATH% environment variable.
#
# To build Rocksdb for Windows is as easy as 1-2-3-4-5:
#
# 1. Update paths to third-party libraries in thirdparty.inc file
# 2. Create a new directory for build artifacts
# mkdir build
# cd build
# 3. Run cmake to generate project files for Windows, add more options to enable required third-party libraries.
# See thirdparty.inc for more information.
# sample command: cmake -G "Visual Studio 15 Win64" -DWITH_GFLAGS=1 -DWITH_SNAPPY=1 -DWITH_JEMALLOC=1 -DWITH_JNI=1 ..
# 4. Then build the project in debug mode (you may want to add /m[:<N>] flag to run msbuild in <N> parallel threads
# or simply /m to use all avail cores)
# msbuild rocksdb.sln
#
# rocksdb.sln build features exclusions of test only code in Release. If you build ALL_BUILD then everything
# will be attempted but test only code does not build in Release mode.
#
# 5. And release mode (/m[:<N>] is also supported)
# msbuild rocksdb.sln /p:Configuration=Release
#
# Linux:
#
# 1. Install a recent toolchain such as devtoolset-3 if you're on a older distro. C++11 required.
# 2. mkdir build; cd build
# 3. cmake ..
# 4. make -j
cmake_minimum_required(VERSION 2.8.12)
project(rocksdb)
enable_language(CXX)
enable_language(C)
enable_language(ASM)
if(POLICY CMP0042)
cmake_policy(SET CMP0042 NEW)
endif()
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_LIST_DIR}/cmake/modules/")
option(WITH_JEMALLOC "build with JeMalloc" OFF)
option(WITH_SNAPPY "build with SNAPPY" OFF)
option(WITH_LZ4 "build with lz4" OFF)
option(WITH_ZLIB "build with zlib" OFF)
if(MSVC)
# Defaults currently different for GFLAGS.
# We will address find_package work a little later
option(WITH_GFLAGS "build with GFlags" OFF)
option(WITH_XPRESS "build with windows built in compression" OFF)
include(${CMAKE_CURRENT_SOURCE_DIR}/thirdparty.inc)
else()
if(CMAKE_SYSTEM_NAME MATCHES "FreeBSD")
# FreeBSD has jemalloc as default malloc
# but it does not have all the jemalloc files in include/...
set(WITH_JEMALLOC ON)
else()
if(WITH_JEMALLOC)
find_package(JeMalloc REQUIRED)
add_definitions(-DROCKSDB_JEMALLOC -DJEMALLOC_NO_DEMANGLE)
include_directories(${JEMALLOC_INCLUDE_DIR})
list(APPEND THIRDPARTY_LIBS ${JEMALLOC_LIBRARIES})
endif()
endif()
Fix race condition causing double deletion of ssts Summary: Possible interleaved execution of background compaction thread calling `FindObsoleteFiles (no full scan) / PurgeObsoleteFiles` and user thread calling `FindObsoleteFiles (full scan) / PurgeObsoleteFiles` can lead to race condition on which RocksDB attempts to delete a file twice. The second attempt will fail and return `IO error`. This may occur to other files, but this PR targets sst. Also add a unit test to verify that this PR fixes the issue. The newly added unit test `obsolete_files_test` has a test case for this scenario, implemented in `ObsoleteFilesTest#RaceForObsoleteFileDeletion`. `TestSyncPoint`s are used to coordinate the interleaving the `user_thread` and background compaction thread. They execute as follows ``` timeline user_thread background_compaction thread t1 | FindObsoleteFiles(full_scan=false) t2 | FindObsoleteFiles(full_scan=true) t3 | PurgeObsoleteFiles t4 | PurgeObsoleteFiles V ``` When `user_thread` invokes `FindObsoleteFiles` with full scan, it collects ALL files in RocksDB directory, including the ones that background compaction thread have collected in its job context. Then `user_thread` will see an IO error when trying to delete these files in `PurgeObsoleteFiles` because background compaction thread has already deleted the file in `PurgeObsoleteFiles`. To fix this, we make RocksDB remember which (SST) files have been found by threads after calling `FindObsoleteFiles` (see `DBImpl#files_grabbed_for_purge_`). Therefore, when another thread calls `FindObsoleteFiles` with full scan, it will not collect such files. ajkr could you take a look and comment? Thanks! Closes https://github.com/facebook/rocksdb/pull/3638 Differential Revision: D7384372 Pulled By: riversand963 fbshipit-source-id: 01489516d60012e722ee65a80e1449e589ce26d3
2018-03-28 19:23:31 +02:00
# No config file for this
option(WITH_GFLAGS "build with GFlags" ON)
if(WITH_GFLAGS)
find_package(gflags)
if(gflags_FOUND)
add_definitions(-DGFLAGS=1)
include_directories(${gflags_INCLUDE_DIR})
list(APPEND THIRDPARTY_LIBS ${gflags_LIBRARIES})
endif()
endif()
if(WITH_SNAPPY)
find_package(snappy REQUIRED)
add_definitions(-DSNAPPY)
include_directories(${SNAPPY_INCLUDE_DIR})
list(APPEND THIRDPARTY_LIBS ${SNAPPY_LIBRARIES})
endif()
if(WITH_ZLIB)
find_package(zlib REQUIRED)
add_definitions(-DZLIB)
include_directories(${ZLIB_INCLUDE_DIR})
list(APPEND THIRDPARTY_LIBS ${ZLIB_LIBRARIES})
endif()
option(WITH_BZ2 "build with bzip2" OFF)
if(WITH_BZ2)
find_package(bzip2 REQUIRED)
add_definitions(-DBZIP2)
include_directories(${BZIP2_INCLUDE_DIR})
list(APPEND THIRDPARTY_LIBS ${BZIP2_LIBRARIES})
endif()
if(WITH_LZ4)
find_package(lz4 REQUIRED)
add_definitions(-DLZ4)
include_directories(${LZ4_INCLUDE_DIR})
list(APPEND THIRDPARTY_LIBS ${LZ4_LIBRARIES})
endif()
option(WITH_ZSTD "build with zstd" OFF)
if(WITH_ZSTD)
find_package(zstd REQUIRED)
add_definitions(-DZSTD)
include_directories(${ZSTD_INCLUDE_DIR})
list(APPEND THIRDPARTY_LIBS ${ZSTD_LIBRARIES})
endif()
endif()
string(TIMESTAMP GIT_DATE_TIME "%Y/%m/%d %H:%M:%S" UTC)
find_package(Git)
if(GIT_FOUND AND EXISTS "${CMAKE_CURRENT_SOURCE_DIR}/.git")
if(WIN32)
execute_process(COMMAND $ENV{COMSPEC} /C ${GIT_EXECUTABLE} -C ${CMAKE_CURRENT_SOURCE_DIR} rev-parse HEAD OUTPUT_VARIABLE GIT_SHA)
else()
execute_process(COMMAND ${GIT_EXECUTABLE} -C ${CMAKE_CURRENT_SOURCE_DIR} rev-parse HEAD OUTPUT_VARIABLE GIT_SHA)
endif()
else()
set(GIT_SHA 0)
endif()
string(REGEX REPLACE "[^0-9a-f]+" "" GIT_SHA "${GIT_SHA}")
# Read rocksdb version from version.h header file.
file(READ include/rocksdb/version.h version_header_file)
string(REGEX MATCH "#define ROCKSDB_MAJOR ([0-9]+)" _ ${version_header_file})
set(ROCKSDB_VERSION_MAJOR ${CMAKE_MATCH_1})
string(REGEX MATCH "#define ROCKSDB_MINOR ([0-9]+)" _ ${version_header_file})
set(ROCKSDB_VERSION_MINOR ${CMAKE_MATCH_1})
string(REGEX MATCH "#define ROCKSDB_PATCH ([0-9]+)" _ ${version_header_file})
set(ROCKSDB_VERSION_PATCH ${CMAKE_MATCH_1})
set(ROCKSDB_VERSION ${ROCKSDB_VERSION_MAJOR}.${ROCKSDB_VERSION_MINOR}.${ROCKSDB_VERSION_PATCH})
cross-platform compatibility improvements Summary: We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS. See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build. I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream. Closes https://github.com/facebook/rocksdb/pull/2199 Differential Revision: D5054042 Pulled By: yiwu-arbug fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
2017-05-15 23:42:32 +02:00
option(WITH_MD_LIBRARY "build with MD" ON)
if(WIN32 AND MSVC)
if(WITH_MD_LIBRARY)
set(RUNTIME_LIBRARY "MD")
else()
set(RUNTIME_LIBRARY "MT")
endif()
cross-platform compatibility improvements Summary: We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS. See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build. I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream. Closes https://github.com/facebook/rocksdb/pull/2199 Differential Revision: D5054042 Pulled By: yiwu-arbug fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
2017-05-15 23:42:32 +02:00
endif()
set(BUILD_VERSION_CC ${CMAKE_BINARY_DIR}/build_version.cc)
configure_file(util/build_version.cc.in ${BUILD_VERSION_CC} @ONLY)
add_library(build_version OBJECT ${BUILD_VERSION_CC})
target_include_directories(build_version PRIVATE
${CMAKE_CURRENT_SOURCE_DIR}/util)
if(MSVC)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /Zi /nologo /EHsc /GS /Gd /GR /GF /fp:precise /Zc:wchar_t /Zc:forScope /errorReport:queue")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /FC /d2Zi+ /W4 /wd4127 /wd4800 /wd4996 /wd4351 /wd4100 /wd4204 /wd4324")
else()
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -W -Wextra -Wall")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wsign-compare -Wshadow -Wno-unused-parameter -Wno-unused-variable -Woverloaded-virtual -Wnon-virtual-dtor -Wno-missing-field-initializers -Wno-strict-aliasing")
if(MINGW)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-format")
endif()
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++11")
if(NOT CMAKE_BUILD_TYPE STREQUAL "Debug")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -O2 -fno-omit-frame-pointer")
include(CheckCXXCompilerFlag)
CHECK_CXX_COMPILER_FLAG("-momit-leaf-frame-pointer" HAVE_OMIT_LEAF_FRAME_POINTER)
if(HAVE_OMIT_LEAF_FRAME_POINTER)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -momit-leaf-frame-pointer")
endif()
endif()
endif()
include(CheckCCompilerFlag)
if(CMAKE_SYSTEM_PROCESSOR MATCHES "ppc64le")
CHECK_C_COMPILER_FLAG("-maltivec" HAS_ALTIVEC)
if(HAS_ALTIVEC)
message(STATUS " HAS_ALTIVEC yes")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -maltivec")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -maltivec")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -mcpu=power8")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mcpu=power8")
endif(HAS_ALTIVEC)
endif(CMAKE_SYSTEM_PROCESSOR MATCHES "ppc64le")
cross-platform compatibility improvements Summary: We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS. See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build. I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream. Closes https://github.com/facebook/rocksdb/pull/2199 Differential Revision: D5054042 Pulled By: yiwu-arbug fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
2017-05-15 23:42:32 +02:00
option(PORTABLE "build a portable binary" OFF)
option(FORCE_SSE42 "force building with SSE4.2, even when PORTABLE=ON" OFF)
if(PORTABLE)
# MSVC does not need a separate compiler flag to enable SSE4.2; if nmmintrin.h
# is available, it is available by default.
if(FORCE_SSE42 AND NOT MSVC)
Port 3 way SSE4.2 crc32c implementation from Folly Summary: **# Summary** RocksDB uses SSE crc32 intrinsics to calculate the crc32 values but it does it in single way fashion (not pipelined on single CPU core). Intel's whitepaper () published an algorithm that uses 3-way pipelining for the crc32 intrinsics, then use pclmulqdq intrinsic to combine the values. Because pclmulqdq has overhead on its own, this algorithm will show perf gains on buffers larger than 216 bytes, which makes RocksDB a perfect user, since most of the buffers RocksDB call crc32c on is over 4KB. Initial db_bench show tremendous CPU gain. This change uses the 3-way SSE algorithm by default. The old SSE algorithm is now behind a compiler tag NO_THREEWAY_CRC32C. If user compiles the code with NO_THREEWAY_CRC32C=1 then the old SSE Crc32c algorithm would be used. If the server does not have SSE4.2 at the run time the slow way (Non SSE) will be used. **# Performance Test Results** We ran the FillRandom and ReadRandom benchmarks in db_bench. ReadRandom is the point of interest here since it calculates the CRC32 for the in-mem buffers. We did 3 runs for each algorithm. Before this change the CRC32 value computation takes about 11.5% of total CPU cost, and with the new 3-way algorithm it reduced to around 4.5%. The overall throughput also improved from 25.53MB/s to 27.63MB/s. 1) ReadRandom in db_bench overall metrics PER RUN Algorithm | run | micros/op | ops/sec |Throughput (MB/s) 3-way | 1 | 4.143 | 241387 | 26.7 3-way | 2 | 3.775 | 264872 | 29.3 3-way | 3 | 4.116 | 242929 | 26.9 FastCrc32c|1 | 4.037 | 247727 | 27.4 FastCrc32c|2 | 4.648 | 215166 | 23.8 FastCrc32c|3 | 4.352 | 229799 | 25.4 AVG Algorithm | Average of micros/op | Average of ops/sec | Average of Throughput (MB/s) 3-way | 4.01 | 249,729 | 27.63 FastCrc32c | 4.35 | 230,897 | 25.53 2) Crc32c computation CPU cost (inclusive samples percentage) PER RUN Implementation | run |  TotalSamples | Crc32c percentage 3-way   | 1    |  4,572,250,000 | 4.37% 3-way   | 2    |  3,779,250,000 | 4.62% 3-way   | 3    |  4,129,500,000 | 4.48% FastCrc32c     | 1    |  4,663,500,000 | 11.24% FastCrc32c     | 2    |  4,047,500,000 | 12.34% FastCrc32c     | 3    |  4,366,750,000 | 11.68% **# Test Plan** make -j64 corruption_test && ./corruption_test By default it uses 3-way SSE algorithm NO_THREEWAY_CRC32C=1 make -j64 corruption_test && ./corruption_test make clean && DEBUG_LEVEL=0 make -j64 db_bench make clean && DEBUG_LEVEL=0 NO_THREEWAY_CRC32C=1 make -j64 db_bench Closes https://github.com/facebook/rocksdb/pull/3173 Differential Revision: D6330882 Pulled By: yingsu00 fbshipit-source-id: 8ec3d89719533b63b536a736663ca6f0dd4482e9
2017-12-20 03:20:50 +01:00
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -msse4.2 -mpclmul")
endif()
else()
cross-platform compatibility improvements Summary: We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS. See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build. I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream. Closes https://github.com/facebook/rocksdb/pull/2199 Differential Revision: D5054042 Pulled By: yiwu-arbug fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
2017-05-15 23:42:32 +02:00
if(MSVC)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /arch:AVX2")
else()
if(NOT HAVE_POWER8)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -march=native")
endif()
endif()
endif()
cross-platform compatibility improvements Summary: We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS. See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build. I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream. Closes https://github.com/facebook/rocksdb/pull/2199 Differential Revision: D5054042 Pulled By: yiwu-arbug fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
2017-05-15 23:42:32 +02:00
include(CheckCXXSourceCompiles)
if(NOT MSVC)
Port 3 way SSE4.2 crc32c implementation from Folly Summary: **# Summary** RocksDB uses SSE crc32 intrinsics to calculate the crc32 values but it does it in single way fashion (not pipelined on single CPU core). Intel's whitepaper () published an algorithm that uses 3-way pipelining for the crc32 intrinsics, then use pclmulqdq intrinsic to combine the values. Because pclmulqdq has overhead on its own, this algorithm will show perf gains on buffers larger than 216 bytes, which makes RocksDB a perfect user, since most of the buffers RocksDB call crc32c on is over 4KB. Initial db_bench show tremendous CPU gain. This change uses the 3-way SSE algorithm by default. The old SSE algorithm is now behind a compiler tag NO_THREEWAY_CRC32C. If user compiles the code with NO_THREEWAY_CRC32C=1 then the old SSE Crc32c algorithm would be used. If the server does not have SSE4.2 at the run time the slow way (Non SSE) will be used. **# Performance Test Results** We ran the FillRandom and ReadRandom benchmarks in db_bench. ReadRandom is the point of interest here since it calculates the CRC32 for the in-mem buffers. We did 3 runs for each algorithm. Before this change the CRC32 value computation takes about 11.5% of total CPU cost, and with the new 3-way algorithm it reduced to around 4.5%. The overall throughput also improved from 25.53MB/s to 27.63MB/s. 1) ReadRandom in db_bench overall metrics PER RUN Algorithm | run | micros/op | ops/sec |Throughput (MB/s) 3-way | 1 | 4.143 | 241387 | 26.7 3-way | 2 | 3.775 | 264872 | 29.3 3-way | 3 | 4.116 | 242929 | 26.9 FastCrc32c|1 | 4.037 | 247727 | 27.4 FastCrc32c|2 | 4.648 | 215166 | 23.8 FastCrc32c|3 | 4.352 | 229799 | 25.4 AVG Algorithm | Average of micros/op | Average of ops/sec | Average of Throughput (MB/s) 3-way | 4.01 | 249,729 | 27.63 FastCrc32c | 4.35 | 230,897 | 25.53 2) Crc32c computation CPU cost (inclusive samples percentage) PER RUN Implementation | run |  TotalSamples | Crc32c percentage 3-way   | 1    |  4,572,250,000 | 4.37% 3-way   | 2    |  3,779,250,000 | 4.62% 3-way   | 3    |  4,129,500,000 | 4.48% FastCrc32c     | 1    |  4,663,500,000 | 11.24% FastCrc32c     | 2    |  4,047,500,000 | 12.34% FastCrc32c     | 3    |  4,366,750,000 | 11.68% **# Test Plan** make -j64 corruption_test && ./corruption_test By default it uses 3-way SSE algorithm NO_THREEWAY_CRC32C=1 make -j64 corruption_test && ./corruption_test make clean && DEBUG_LEVEL=0 make -j64 db_bench make clean && DEBUG_LEVEL=0 NO_THREEWAY_CRC32C=1 make -j64 db_bench Closes https://github.com/facebook/rocksdb/pull/3173 Differential Revision: D6330882 Pulled By: yingsu00 fbshipit-source-id: 8ec3d89719533b63b536a736663ca6f0dd4482e9
2017-12-20 03:20:50 +01:00
set(CMAKE_REQUIRED_FLAGS "-msse4.2 -mpclmul")
endif()
cross-platform compatibility improvements Summary: We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS. See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build. I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream. Closes https://github.com/facebook/rocksdb/pull/2199 Differential Revision: D5054042 Pulled By: yiwu-arbug fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
2017-05-15 23:42:32 +02:00
CHECK_CXX_SOURCE_COMPILES("
#include <cstdint>
#include <nmmintrin.h>
Port 3 way SSE4.2 crc32c implementation from Folly Summary: **# Summary** RocksDB uses SSE crc32 intrinsics to calculate the crc32 values but it does it in single way fashion (not pipelined on single CPU core). Intel's whitepaper () published an algorithm that uses 3-way pipelining for the crc32 intrinsics, then use pclmulqdq intrinsic to combine the values. Because pclmulqdq has overhead on its own, this algorithm will show perf gains on buffers larger than 216 bytes, which makes RocksDB a perfect user, since most of the buffers RocksDB call crc32c on is over 4KB. Initial db_bench show tremendous CPU gain. This change uses the 3-way SSE algorithm by default. The old SSE algorithm is now behind a compiler tag NO_THREEWAY_CRC32C. If user compiles the code with NO_THREEWAY_CRC32C=1 then the old SSE Crc32c algorithm would be used. If the server does not have SSE4.2 at the run time the slow way (Non SSE) will be used. **# Performance Test Results** We ran the FillRandom and ReadRandom benchmarks in db_bench. ReadRandom is the point of interest here since it calculates the CRC32 for the in-mem buffers. We did 3 runs for each algorithm. Before this change the CRC32 value computation takes about 11.5% of total CPU cost, and with the new 3-way algorithm it reduced to around 4.5%. The overall throughput also improved from 25.53MB/s to 27.63MB/s. 1) ReadRandom in db_bench overall metrics PER RUN Algorithm | run | micros/op | ops/sec |Throughput (MB/s) 3-way | 1 | 4.143 | 241387 | 26.7 3-way | 2 | 3.775 | 264872 | 29.3 3-way | 3 | 4.116 | 242929 | 26.9 FastCrc32c|1 | 4.037 | 247727 | 27.4 FastCrc32c|2 | 4.648 | 215166 | 23.8 FastCrc32c|3 | 4.352 | 229799 | 25.4 AVG Algorithm | Average of micros/op | Average of ops/sec | Average of Throughput (MB/s) 3-way | 4.01 | 249,729 | 27.63 FastCrc32c | 4.35 | 230,897 | 25.53 2) Crc32c computation CPU cost (inclusive samples percentage) PER RUN Implementation | run |  TotalSamples | Crc32c percentage 3-way   | 1    |  4,572,250,000 | 4.37% 3-way   | 2    |  3,779,250,000 | 4.62% 3-way   | 3    |  4,129,500,000 | 4.48% FastCrc32c     | 1    |  4,663,500,000 | 11.24% FastCrc32c     | 2    |  4,047,500,000 | 12.34% FastCrc32c     | 3    |  4,366,750,000 | 11.68% **# Test Plan** make -j64 corruption_test && ./corruption_test By default it uses 3-way SSE algorithm NO_THREEWAY_CRC32C=1 make -j64 corruption_test && ./corruption_test make clean && DEBUG_LEVEL=0 make -j64 db_bench make clean && DEBUG_LEVEL=0 NO_THREEWAY_CRC32C=1 make -j64 db_bench Closes https://github.com/facebook/rocksdb/pull/3173 Differential Revision: D6330882 Pulled By: yingsu00 fbshipit-source-id: 8ec3d89719533b63b536a736663ca6f0dd4482e9
2017-12-20 03:20:50 +01:00
#include <wmmintrin.h>
cross-platform compatibility improvements Summary: We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS. See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build. I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream. Closes https://github.com/facebook/rocksdb/pull/2199 Differential Revision: D5054042 Pulled By: yiwu-arbug fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
2017-05-15 23:42:32 +02:00
int main() {
volatile uint32_t x = _mm_crc32_u32(0, 0);
Port 3 way SSE4.2 crc32c implementation from Folly Summary: **# Summary** RocksDB uses SSE crc32 intrinsics to calculate the crc32 values but it does it in single way fashion (not pipelined on single CPU core). Intel's whitepaper () published an algorithm that uses 3-way pipelining for the crc32 intrinsics, then use pclmulqdq intrinsic to combine the values. Because pclmulqdq has overhead on its own, this algorithm will show perf gains on buffers larger than 216 bytes, which makes RocksDB a perfect user, since most of the buffers RocksDB call crc32c on is over 4KB. Initial db_bench show tremendous CPU gain. This change uses the 3-way SSE algorithm by default. The old SSE algorithm is now behind a compiler tag NO_THREEWAY_CRC32C. If user compiles the code with NO_THREEWAY_CRC32C=1 then the old SSE Crc32c algorithm would be used. If the server does not have SSE4.2 at the run time the slow way (Non SSE) will be used. **# Performance Test Results** We ran the FillRandom and ReadRandom benchmarks in db_bench. ReadRandom is the point of interest here since it calculates the CRC32 for the in-mem buffers. We did 3 runs for each algorithm. Before this change the CRC32 value computation takes about 11.5% of total CPU cost, and with the new 3-way algorithm it reduced to around 4.5%. The overall throughput also improved from 25.53MB/s to 27.63MB/s. 1) ReadRandom in db_bench overall metrics PER RUN Algorithm | run | micros/op | ops/sec |Throughput (MB/s) 3-way | 1 | 4.143 | 241387 | 26.7 3-way | 2 | 3.775 | 264872 | 29.3 3-way | 3 | 4.116 | 242929 | 26.9 FastCrc32c|1 | 4.037 | 247727 | 27.4 FastCrc32c|2 | 4.648 | 215166 | 23.8 FastCrc32c|3 | 4.352 | 229799 | 25.4 AVG Algorithm | Average of micros/op | Average of ops/sec | Average of Throughput (MB/s) 3-way | 4.01 | 249,729 | 27.63 FastCrc32c | 4.35 | 230,897 | 25.53 2) Crc32c computation CPU cost (inclusive samples percentage) PER RUN Implementation | run |  TotalSamples | Crc32c percentage 3-way   | 1    |  4,572,250,000 | 4.37% 3-way   | 2    |  3,779,250,000 | 4.62% 3-way   | 3    |  4,129,500,000 | 4.48% FastCrc32c     | 1    |  4,663,500,000 | 11.24% FastCrc32c     | 2    |  4,047,500,000 | 12.34% FastCrc32c     | 3    |  4,366,750,000 | 11.68% **# Test Plan** make -j64 corruption_test && ./corruption_test By default it uses 3-way SSE algorithm NO_THREEWAY_CRC32C=1 make -j64 corruption_test && ./corruption_test make clean && DEBUG_LEVEL=0 make -j64 db_bench make clean && DEBUG_LEVEL=0 NO_THREEWAY_CRC32C=1 make -j64 db_bench Closes https://github.com/facebook/rocksdb/pull/3173 Differential Revision: D6330882 Pulled By: yingsu00 fbshipit-source-id: 8ec3d89719533b63b536a736663ca6f0dd4482e9
2017-12-20 03:20:50 +01:00
const auto a = _mm_set_epi64x(0, 0);
const auto b = _mm_set_epi64x(0, 0);
const auto c = _mm_clmulepi64_si128(a, b, 0x00);
auto d = _mm_cvtsi128_si64(c);
cross-platform compatibility improvements Summary: We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS. See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build. I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream. Closes https://github.com/facebook/rocksdb/pull/2199 Differential Revision: D5054042 Pulled By: yiwu-arbug fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
2017-05-15 23:42:32 +02:00
}
" HAVE_SSE42)
unset(CMAKE_REQUIRED_FLAGS)
cross-platform compatibility improvements Summary: We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS. See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build. I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream. Closes https://github.com/facebook/rocksdb/pull/2199 Differential Revision: D5054042 Pulled By: yiwu-arbug fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
2017-05-15 23:42:32 +02:00
if(HAVE_SSE42)
add_definitions(-DHAVE_SSE42)
add_definitions(-DHAVE_PCLMUL)
cross-platform compatibility improvements Summary: We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS. See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build. I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream. Closes https://github.com/facebook/rocksdb/pull/2199 Differential Revision: D5054042 Pulled By: yiwu-arbug fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
2017-05-15 23:42:32 +02:00
elseif(FORCE_SSE42)
message(FATAL_ERROR "FORCE_SSE42=ON but unable to compile with SSE4.2 enabled")
endif()
CHECK_CXX_SOURCE_COMPILES("
#if defined(_MSC_VER) && !defined(__thread)
#define __thread __declspec(thread)
#endif
int main() {
static __thread int tls;
}
" HAVE_THREAD_LOCAL)
if(HAVE_THREAD_LOCAL)
add_definitions(-DROCKSDB_SUPPORT_THREAD_LOCAL)
endif()
option(FAIL_ON_WARNINGS "Treat compile warnings as errors" ON)
if(FAIL_ON_WARNINGS)
if(MSVC)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /WX")
else() # assume GCC
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Werror")
endif()
endif()
option(WITH_ASAN "build with ASAN" OFF)
if(WITH_ASAN)
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -fsanitize=address")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=address")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fsanitize=address")
if(WITH_JEMALLOC)
message(FATAL "ASAN does not work well with JeMalloc")
endif()
endif()
option(WITH_TSAN "build with TSAN" OFF)
if(WITH_TSAN)
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -fsanitize=thread -pie")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=thread -fPIC")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fsanitize=thread -fPIC")
if(WITH_JEMALLOC)
message(FATAL "TSAN does not work well with JeMalloc")
endif()
endif()
option(WITH_UBSAN "build with UBSAN" OFF)
if(WITH_UBSAN)
add_definitions(-DROCKSDB_UBSAN_RUN)
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -fsanitize=undefined")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=undefined")
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fsanitize=undefined")
if(WITH_JEMALLOC)
message(FATAL "UBSAN does not work well with JeMalloc")
endif()
endif()
option(WITH_NUMA "build with NUMA policy support" OFF)
if(WITH_NUMA)
find_package(NUMA REQUIRED)
add_definitions(-DNUMA)
include_directories(${NUMA_INCLUDE_DIR})
list(APPEND THIRDPARTY_LIBS ${NUMA_LIBRARIES})
endif()
option(WITH_TBB "build with Threading Building Blocks (TBB)" OFF)
if(WITH_TBB)
find_package(TBB REQUIRED)
add_definitions(-DTBB)
include_directories(${TBB_INCLUDE_DIR})
list(APPEND THIRDPARTY_LIBS ${TBB_LIBRARIES})
endif()
# Stall notifications eat some performance from inserts
option(DISABLE_STALL_NOTIF "Build with stall notifications" OFF)
if(DISABLE_STALL_NOTIF)
add_definitions(-DROCKSDB_DISABLE_STALL_NOTIFICATION)
endif()
# Used to run CI build and tests so we can run faster
set(OPTIMIZE_DEBUG_DEFAULT 0) # Debug build is unoptimized by default use -DOPTDBG=1 to optimize
if(DEFINED OPTDBG)
set(OPTIMIZE_DEBUG ${OPTDBG})
else()
set(OPTIMIZE_DEBUG ${OPTIMIZE_DEBUG_DEFAULT})
endif()
if(DEFINED USE_RTTI)
if(USE_RTTI)
message(STATUS "Enabling RTTI")
set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -DROCKSDB_USE_RTTI")
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -DROCKSDB_USE_RTTI")
else()
if(MSVC)
message(STATUS "Disabling RTTI in Release builds. Always on in Debug.")
set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -DROCKSDB_USE_RTTI")
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} /GR-")
else()
message(STATUS "Disabling RTTI in Release builds")
set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -fno-rtti")
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -fno-rtti")
endif()
endif()
else()
message(STATUS "Enabling RTTI in Debug builds only (default)")
set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -DROCKSDB_USE_RTTI")
if(MSVC)
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} /GR-")
else()
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -fno-rtti")
endif()
endif()
if(MSVC)
if((${OPTIMIZE_DEBUG} EQUAL 1))
message(STATUS "Debug optimization is enabled")
set(CMAKE_CXX_FLAGS_DEBUG "/Oxt /${RUNTIME_LIBRARY}d")
else()
set(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} /Od /RTC1 /Gm /${RUNTIME_LIBRARY}d")
endif()
set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} /Oxt /Zp8 /Gm- /Gy /${RUNTIME_LIBRARY}")
set(CMAKE_SHARED_LINKER_FLAGS "${CMAKE_SHARED_LINKER_FLAGS} /DEBUG")
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} /DEBUG")
endif()
if(CMAKE_COMPILER_IS_GNUCXX)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fno-builtin-memcmp")
endif()
option(ROCKSDB_LITE "Build RocksDBLite version" OFF)
if(ROCKSDB_LITE)
add_definitions(-DROCKSDB_LITE)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fno-exceptions")
endif()
if(CMAKE_SYSTEM_NAME MATCHES "Cygwin")
add_definitions(-fno-builtin-memcmp -DCYGWIN)
elseif(CMAKE_SYSTEM_NAME MATCHES "Darwin")
add_definitions(-DOS_MACOSX)
if(CMAKE_SYSTEM_PROCESSOR MATCHES arm)
add_definitions(-DIOS_CROSS_COMPILE -DROCKSDB_LITE)
# no debug info for IOS, that will make our library big
add_definitions(-DNDEBUG)
endif()
elseif(CMAKE_SYSTEM_NAME MATCHES "Linux")
add_definitions(-DOS_LINUX)
elseif(CMAKE_SYSTEM_NAME MATCHES "SunOS")
add_definitions(-DOS_SOLARIS)
elseif(CMAKE_SYSTEM_NAME MATCHES "FreeBSD")
add_definitions(-DOS_FREEBSD)
elseif(CMAKE_SYSTEM_NAME MATCHES "NetBSD")
add_definitions(-DOS_NETBSD)
elseif(CMAKE_SYSTEM_NAME MATCHES "OpenBSD")
add_definitions(-DOS_OPENBSD)
elseif(CMAKE_SYSTEM_NAME MATCHES "DragonFly")
add_definitions(-DOS_DRAGONFLYBSD)
elseif(CMAKE_SYSTEM_NAME MATCHES "Android")
add_definitions(-DOS_ANDROID)
elseif(CMAKE_SYSTEM_NAME MATCHES "Windows")
add_definitions(-DWIN32 -DOS_WIN -D_MBCS -DWIN64 -DNOMINMAX)
if(MINGW)
add_definitions(-D_WIN32_WINNT=_WIN32_WINNT_VISTA)
endif()
endif()
if(NOT WIN32)
add_definitions(-DROCKSDB_PLATFORM_POSIX -DROCKSDB_LIB_IO_POSIX)
endif()
option(WITH_FALLOCATE "build with fallocate" ON)
if(WITH_FALLOCATE)
CHECK_CXX_SOURCE_COMPILES("
#include <fcntl.h>
#include <linux/falloc.h>
int main() {
int fd = open(\"/dev/null\", 0);
fallocate(fd, FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE, 0, 1024);
}
" HAVE_FALLOCATE)
if(HAVE_FALLOCATE)
add_definitions(-DROCKSDB_FALLOCATE_PRESENT)
endif()
endif()
CHECK_CXX_SOURCE_COMPILES("
#include <fcntl.h>
int main() {
int fd = open(\"/dev/null\", 0);
sync_file_range(fd, 0, 1024, SYNC_FILE_RANGE_WRITE);
}
" HAVE_SYNC_FILE_RANGE_WRITE)
if(HAVE_SYNC_FILE_RANGE_WRITE)
add_definitions(-DROCKSDB_RANGESYNC_PRESENT)
endif()
CHECK_CXX_SOURCE_COMPILES("
#include <pthread.h>
int main() {
(void) PTHREAD_MUTEX_ADAPTIVE_NP;
}
" HAVE_PTHREAD_MUTEX_ADAPTIVE_NP)
if(HAVE_PTHREAD_MUTEX_ADAPTIVE_NP)
add_definitions(-DROCKSDB_PTHREAD_ADAPTIVE_MUTEX)
endif()
include(CheckCXXSymbolExists)
check_cxx_symbol_exists(malloc_usable_size malloc.h HAVE_MALLOC_USABLE_SIZE)
if(HAVE_MALLOC_USABLE_SIZE)
add_definitions(-DROCKSDB_MALLOC_USABLE_SIZE)
endif()
check_cxx_symbol_exists(sched_getcpu sched.h HAVE_SCHED_GETCPU)
if(HAVE_SCHED_GETCPU)
add_definitions(-DROCKSDB_SCHED_GETCPU_PRESENT)
endif()
include_directories(${PROJECT_SOURCE_DIR})
include_directories(${PROJECT_SOURCE_DIR}/include)
include_directories(SYSTEM ${PROJECT_SOURCE_DIR}/third-party/gtest-1.7.0/fused-src)
find_package(Threads REQUIRED)
add_subdirectory(third-party/gtest-1.7.0/fused-src/gtest)
# Main library source code
set(SOURCES
cache/clock_cache.cc
cache/lru_cache.cc
cache/sharded_cache.cc
db/builder.cc
db/c.cc
db/column_family.cc
db/compacted_db_impl.cc
db/compaction.cc
db/compaction_iterator.cc
db/compaction_job.cc
db/compaction_picker.cc
db/compaction_picker_universal.cc
db/convenience.cc
db/db_filesnapshot.cc
db/db_impl.cc
db/db_impl_write.cc
db/db_impl_compaction_flush.cc
db/db_impl_files.cc
db/db_impl_open.cc
db/db_impl_debug.cc
db/db_impl_experimental.cc
db/db_impl_readonly.cc
db/db_info_dumper.cc
db/db_iter.cc
db/dbformat.cc
db/event_helpers.cc
db/experimental.cc
db/external_sst_file_ingestion_job.cc
db/file_indexer.cc
db/flush_job.cc
db/flush_scheduler.cc
db/forward_iterator.cc
db/internal_stats.cc
Skip deleted WALs during recovery Summary: This patch record min log number to keep to the manifest while flushing SST files to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic. Before the commit, for 2PC case, we determined which log number to keep in FindObsoleteFiles(). We looked at the earliest logs with outstanding prepare entries, or prepare entries whose respective commit or abort are in memtable. With the commit, the same calculation is done while we apply the SST flush. Just before installing the flush file, we precompute the earliest log file to keep after the flush finishes using the same logic (but skipping the memtables just flushed), record this information to the manifest entry for this new flushed SST file. This pre-computed value is also remembered in memory, and will later be used to determine whether a log file can be deleted. This value is unlikely to change until next flush because the commit entry will stay in memtable. (In WritePrepared, we could have removed the older log files as soon as all prepared entries are committed. It's not yet done anyway. Even if we do it, the only thing we loss with this new approach is earlier log deletion between two flushes, which does not guarantee to happen anyway because the obsolete file clean-up function is only executed after flush or compaction) This min log number to keep is stored in the manifest using the safely-ignore customized field of AddFile entry, in order to guarantee that the DB generated using newer release can be opened by previous releases no older than 4.2. Closes https://github.com/facebook/rocksdb/pull/3765 Differential Revision: D7747618 Pulled By: siying fbshipit-source-id: d00c92105b4f83852e9754a1b70d6b64cb590729
2018-05-04 00:35:11 +02:00
db/logs_with_prep_tracker.cc
db/log_reader.cc
db/log_writer.cc
db: avoid `#include`ing malloc and jemalloc simultaneously Summary: This fixes a compilation failure on Linux when the system libc is not glibc. jemalloc's configure script incorrectly assumes that glibc is always used on Linux systems, producing glibc-style signatures; when the system libc is e.g. musl, the following error is observed: ``` [ 0%] Building CXX object CMakeFiles/rocksdb.dir/db/db_impl.cc.o In file included from /go/src/github.com/cockroachdb/cockroach/c-deps/rocksdb.src/table/block.h:19:0, from /go/src/github.com/cockroachdb/cockroach/c-deps/rocksdb.src/db/db_impl.cc:77: /x-tools/x86_64-unknown-linux-musl/x86_64-unknown-linux-musl/sysroot/usr/include/malloc.h:19:8: error: declaration of 'size_t malloc_usable_size(void*)' has a different exception specifier size_t malloc_usable_size(void *); ^~~~~~~~~~~~~~~~~~ In file included from /go/src/github.com/cockroachdb/cockroach/c-deps/rocksdb.src/db/db_impl.cc:20:0: /go/native/x86_64-unknown-linux-musl/jemalloc/include/jemalloc/jemalloc.h:78:33: note: from previous declaration 'size_t malloc_usable_size(void*) throw ()' # define je_malloc_usable_size malloc_usable_size ^ /go/native/x86_64-unknown-linux-musl/jemalloc/include/jemalloc/jemalloc.h:239:41: note: in expansion of macro 'je_malloc_usable_size' JEMALLOC_EXPORT size_t JEMALLOC_NOTHROW je_malloc_usable_size( ^~~~~~~~~~~~~~~~~~~~~ CMakeFiles/rocksdb.dir/build.make:350: recipe for target 'CMakeFiles/rocksdb.dir/db/db_impl.cc.o' failed ``` This works around the issue by rearranging the sources such that jemalloc's headers are never in the same scope as the system's malloc header. The jemalloc issue has been reported as well, see: https://github.com/jemalloc/jemalloc/issues/778. cc tschottdorf Closes https://github.com/facebook/rocksdb/pull/2188 Differential Revision: D5163048 Pulled By: siying fbshipit-source-id: c553125458892def175c1be5682b0330d80b2a0d
2017-06-01 07:41:44 +02:00
db/malloc_stats.cc
db/managed_iterator.cc
db/memtable.cc
db/memtable_list.cc
db/merge_helper.cc
db/merge_operator.cc
Compaction Support for Range Deletion Summary: This diff introduces RangeDelAggregator, which takes ownership of iterators provided to it via AddTombstones(). The tombstones are organized in a two-level map (snapshot stripe -> begin key -> tombstone). Tombstone creation avoids data copy by holding Slices returned by the iterator, which remain valid thanks to pinning. For compaction, we create a hierarchical range tombstone iterator with structure matching the iterator over compaction input data. An aggregator based on that iterator is used by CompactionIterator to determine which keys are covered by range tombstones. In case of merge operand, the same aggregator is used by MergeHelper. Upon finishing each file in the compaction, relevant range tombstones are added to the output file's range tombstone metablock and file boundaries are updated accordingly. To check whether a key is covered by range tombstone, RangeDelAggregator::ShouldDelete() considers tombstones in the key's snapshot stripe. When this function is used outside of compaction, it also checks newer stripes, which can contain covering tombstones. Currently the intra-stripe check involves a linear scan; however, in the future we plan to collapse ranges within a stripe such that binary search can be used. RangeDelAggregator::AddToBuilder() adds all range tombstones in the table's key-range to a new table's range tombstone meta-block. Since range tombstones may fall in the gap between files, we may need to extend some files' key-ranges. The strategy is (1) first file extends as far left as possible and other files do not extend left, (2) all files extend right until either the start of the next file or the end of the last range tombstone in the gap, whichever comes first. One other notable change is adding release/move semantics to ScopedArenaIterator such that it can be used to transfer ownership of an arena-allocated iterator, similar to how unique_ptr is used for malloc'd data. Depends on D61473 Test Plan: compaction_iterator_test, mock_table, end-to-end tests in D63927 Reviewers: sdong, IslamAbdelRahman, wanning, yhchiang, lightmark Reviewed By: lightmark Subscribers: andrewkr, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D62205
2016-10-18 21:04:56 +02:00
db/range_del_aggregator.cc
db/repair.cc
db/snapshot_impl.cc
db/table_cache.cc
db/table_properties_collector.cc
db/transaction_log_impl.cc
db/version_builder.cc
db/version_edit.cc
db/version_set.cc
db/wal_manager.cc
db/write_batch.cc
db/write_batch_base.cc
db/write_controller.cc
db/write_thread.cc
env/env.cc
env/env_chroot.cc
Encryption at rest support Summary: This PR adds support for encrypting data stored by RocksDB when written to disk. It adds an `EncryptedEnv` override of the `Env` class with matching overrides for sequential&random access files. The encryption itself is done through a configurable `EncryptionProvider`. This class creates is asked to create `BlockAccessCipherStream` for a file. This is where the actual encryption/decryption is being done. Currently there is a Counter mode implementation of `BlockAccessCipherStream` with a `ROT13` block cipher (NOTE the `ROT13` is for demo purposes only!!). The Counter operation mode uses an initial counter & random initialization vector (IV). Both are created randomly for each file and stored in a 4K (default size) block that is prefixed to that file. The `EncryptedEnv` implementation is such that clients of the `Env` class do not see this prefix (nor data, nor in filesize). The largest part of the prefix block is also encrypted, and there is room left for implementation specific settings/values/keys in there. To test the encryption, the `DBTestBase` class has been extended to consider a new environment variable called `ENCRYPTED_ENV`. If set, the test will setup a encrypted instance of the `Env` class to use for all tests. Typically you would run it like this: ``` ENCRYPTED_ENV=1 make check_some ``` There is also an added test that checks that some data inserted into the database is or is not "visible" on disk. With `ENCRYPTED_ENV` active it must not find plain text strings, with `ENCRYPTED_ENV` unset, it must find the plain text strings. Closes https://github.com/facebook/rocksdb/pull/2424 Differential Revision: D5322178 Pulled By: sdwilsh fbshipit-source-id: 253b0a9c2c498cc98f580df7f2623cbf7678a27f
2017-06-27 01:52:06 +02:00
env/env_encryption.cc
env/env_hdfs.cc
env/mock_env.cc
memtable/alloc_tracker.cc
memtable/hash_cuckoo_rep.cc
memtable/hash_linklist_rep.cc
memtable/hash_skiplist_rep.cc
memtable/skiplistrep.cc
memtable/vectorrep.cc
memtable/write_buffer_manager.cc
monitoring/histogram.cc
monitoring/histogram_windowing.cc
monitoring/instrumented_mutex.cc
monitoring/iostats_context.cc
monitoring/perf_context.cc
monitoring/perf_level.cc
monitoring/statistics.cc
monitoring/thread_status_impl.cc
monitoring/thread_status_updater.cc
monitoring/thread_status_util.cc
monitoring/thread_status_util_debug.cc
options/cf_options.cc
options/db_options.cc
options/options.cc
options/options_helper.cc
options/options_parser.cc
options/options_sanity_check.cc
port/stack_trace.cc
table/adaptive_table_factory.cc
table/block.cc
table/block_based_filter_block.cc
table/block_based_table_builder.cc
table/block_based_table_factory.cc
table/block_based_table_reader.cc
table/block_builder.cc
table/block_fetcher.cc
table/block_prefix_index.cc
table/bloom_block.cc
table/cuckoo_table_builder.cc
table/cuckoo_table_factory.cc
table/cuckoo_table_reader.cc
table/flush_block_policy.cc
table/format.cc
table/full_filter_block.cc
table/get_context.cc
table/index_builder.cc
table/iterator.cc
table/merging_iterator.cc
table/meta_blocks.cc
table/partitioned_filter_block.cc
table/persistent_cache_helper.cc
table/plain_table_builder.cc
table/plain_table_factory.cc
table/plain_table_index.cc
table/plain_table_key_coding.cc
table/plain_table_reader.cc
table/sst_file_writer.cc
table/table_properties.cc
table/two_level_iterator.cc
tools/db_bench_tool.cc
tools/dump/db_dump_tool.cc
tools/ldb_cmd.cc
tools/ldb_tool.cc
tools/sst_dump_tool.cc
util/arena.cc
util/auto_roll_logger.cc
util/bloom.cc
util/coding.cc
util/compaction_job_stats_impl.cc
util/comparator.cc
util/compression_context_cache.cc
support for concurrent adds to memtable Summary: This diff adds support for concurrent adds to the skiplist memtable implementations. Memory allocation is made thread-safe by the addition of a spinlock, with small per-core buffers to avoid contention. Concurrent memtable writes are made via an additional method and don't impose a performance overhead on the non-concurrent case, so parallelism can be selected on a per-batch basis. Write thread synchronization is an increasing bottleneck for higher levels of concurrency, so this diff adds --enable_write_thread_adaptive_yield (default off). This feature causes threads joining a write batch group to spin for a short time (default 100 usec) using sched_yield, rather than going to sleep on a mutex. If the timing of the yield calls indicates that another thread has actually run during the yield then spinning is avoided. This option improves performance for concurrent situations even without parallel adds, although it has the potential to increase CPU usage (and the heuristic adaptation is not yet mature). Parallel writes are not currently compatible with inplace updates, update callbacks, or delete filtering. Enable it with --allow_concurrent_memtable_write (and --enable_write_thread_adaptive_yield). Parallel memtable writes are performance neutral when there is no actual parallelism, and in my experiments (SSD server-class Linux and varying contention and key sizes for fillrandom) they are always a performance win when there is more than one thread. Statistics are updated earlier in the write path, dropping the number of DB mutex acquisitions from 2 to 1 for almost all cases. This diff was motivated and inspired by Yahoo's cLSM work. It is more conservative than cLSM: RocksDB's write batch group leader role is preserved (along with all of the existing flush and write throttling logic) and concurrent writers are blocked until all memtable insertions have completed and the sequence number has been advanced, to preserve linearizability. My test config is "db_bench -benchmarks=fillrandom -threads=$T -batch_size=1 -memtablerep=skip_list -value_size=100 --num=1000000/$T -level0_slowdown_writes_trigger=9999 -level0_stop_writes_trigger=9999 -disable_auto_compactions --max_write_buffer_number=8 -max_background_flushes=8 --disable_wal --write_buffer_size=160000000 --block_size=16384 --allow_concurrent_memtable_write" on a two-socket Xeon E5-2660 @ 2.2Ghz with lots of memory and an SSD hard drive. With 1 thread I get ~440Kops/sec. Peak performance for 1 socket (numactl -N1) is slightly more than 1Mops/sec, at 16 threads. Peak performance across both sockets happens at 30 threads, and is ~900Kops/sec, although with fewer threads there is less performance loss when the system has background work. Test Plan: 1. concurrent stress tests for InlineSkipList and DynamicBloom 2. make clean; make check 3. make clean; DISABLE_JEMALLOC=1 make valgrind_check; valgrind db_bench 4. make clean; COMPILE_WITH_TSAN=1 make all check; db_bench 5. make clean; COMPILE_WITH_ASAN=1 make all check; db_bench 6. make clean; OPT=-DROCKSDB_LITE make check 7. verify no perf regressions when disabled Reviewers: igor, sdong Reviewed By: sdong Subscribers: MarkCallaghan, IslamAbdelRahman, anthony, yhchiang, rven, sdong, guyg8, kradhakrishnan, dhruba Differential Revision: https://reviews.facebook.net/D50589
2015-08-15 01:59:07 +02:00
util/concurrent_arena.cc
util/crc32c.cc
util/delete_scheduler.cc
util/dynamic_bloom.cc
util/event_logger.cc
util/file_reader_writer.cc
util/file_util.cc
util/filename.cc
util/filter_policy.cc
util/hash.cc
util/log_buffer.cc
util/murmurhash.cc
util/random.cc
util/rate_limiter.cc
util/slice.cc
util/sst_file_manager_impl.cc
util/status.cc
util/status_message.cc
util/string_util.cc
util/sync_point.cc
util/sync_point_impl.cc
util/testutil.cc
util/thread_local.cc
util/threadpool_imp.cc
2016-03-15 18:57:33 +01:00
util/transaction_test_util.cc
util/xxhash.cc
utilities/backupable/backupable_db.cc
utilities/blob_db/blob_compaction_filter.cc
utilities/blob_db/blob_db.cc
utilities/blob_db/blob_db_impl.cc
utilities/blob_db/blob_dump_tool.cc
utilities/blob_db/blob_file.cc
utilities/blob_db/blob_log_reader.cc
utilities/blob_db/blob_log_writer.cc
utilities/blob_db/blob_log_format.cc
utilities/blob_db/ttl_extractor.cc
utilities/cassandra/cassandra_compaction_filter.cc
utilities/cassandra/format.cc
utilities/cassandra/merge_operator.cc
utilities/checkpoint/checkpoint_impl.cc
utilities/col_buf_decoder.cc
utilities/col_buf_encoder.cc
utilities/column_aware_encoding_util.cc
2016-01-28 15:44:31 +01:00
utilities/compaction_filters/remove_emptyvalue_compactionfilter.cc
utilities/date_tiered/date_tiered_db_impl.cc
utilities/debug.cc
utilities/document/document_db.cc
utilities/document/json_document.cc
utilities/document/json_document_builder.cc
utilities/env_mirror.cc
utilities/env_timed.cc
utilities/geodb/geodb_impl.cc
utilities/leveldb_options/leveldb_options.cc
utilities/lua/rocks_lua_compaction_filter.cc
utilities/memory/memory_util.cc
utilities/merge_operators/bytesxor.cc
utilities/merge_operators/max.cc
utilities/merge_operators/put.cc
utilities/merge_operators/string_append/stringappend.cc
utilities/merge_operators/string_append/stringappend2.cc
utilities/merge_operators/uint64add.cc
utilities/option_change_migration/option_change_migration.cc
utilities/options/options_util.cc
2016-08-03 02:15:18 +02:00
utilities/persistent_cache/block_cache_tier.cc
utilities/persistent_cache/block_cache_tier_file.cc
utilities/persistent_cache/block_cache_tier_metadata.cc
utilities/persistent_cache/persistent_cache_tier.cc
utilities/persistent_cache/volatile_tier_impl.cc
utilities/redis/redis_lists.cc
utilities/simulator_cache/sim_cache.cc
utilities/spatialdb/spatial_db.cc
utilities/table_properties_collectors/compact_on_deletion_collector.cc
utilities/transactions/optimistic_transaction_db_impl.cc
utilities/transactions/optimistic_transaction.cc
utilities/transactions/pessimistic_transaction.cc
utilities/transactions/pessimistic_transaction_db.cc
utilities/transactions/snapshot_checker.cc
utilities/transactions/transaction_base.cc
utilities/transactions/transaction_db_mutex_impl.cc
utilities/transactions/transaction_lock_mgr.cc
utilities/transactions/transaction_util.cc
utilities/transactions/write_prepared_txn.cc
utilities/transactions/write_prepared_txn_db.cc
utilities/transactions/write_unprepared_txn.cc
utilities/transactions/write_unprepared_txn_db.cc
utilities/ttl/db_ttl_impl.cc
utilities/write_batch_with_index/write_batch_with_index.cc
utilities/write_batch_with_index/write_batch_with_index_internal.cc
$<TARGET_OBJECTS:build_version>)
if(HAVE_SSE42 AND NOT FORCE_SSE42)
if(NOT MSVC)
set_source_files_properties(
util/crc32c.cc
Port 3 way SSE4.2 crc32c implementation from Folly Summary: **# Summary** RocksDB uses SSE crc32 intrinsics to calculate the crc32 values but it does it in single way fashion (not pipelined on single CPU core). Intel's whitepaper () published an algorithm that uses 3-way pipelining for the crc32 intrinsics, then use pclmulqdq intrinsic to combine the values. Because pclmulqdq has overhead on its own, this algorithm will show perf gains on buffers larger than 216 bytes, which makes RocksDB a perfect user, since most of the buffers RocksDB call crc32c on is over 4KB. Initial db_bench show tremendous CPU gain. This change uses the 3-way SSE algorithm by default. The old SSE algorithm is now behind a compiler tag NO_THREEWAY_CRC32C. If user compiles the code with NO_THREEWAY_CRC32C=1 then the old SSE Crc32c algorithm would be used. If the server does not have SSE4.2 at the run time the slow way (Non SSE) will be used. **# Performance Test Results** We ran the FillRandom and ReadRandom benchmarks in db_bench. ReadRandom is the point of interest here since it calculates the CRC32 for the in-mem buffers. We did 3 runs for each algorithm. Before this change the CRC32 value computation takes about 11.5% of total CPU cost, and with the new 3-way algorithm it reduced to around 4.5%. The overall throughput also improved from 25.53MB/s to 27.63MB/s. 1) ReadRandom in db_bench overall metrics PER RUN Algorithm | run | micros/op | ops/sec |Throughput (MB/s) 3-way | 1 | 4.143 | 241387 | 26.7 3-way | 2 | 3.775 | 264872 | 29.3 3-way | 3 | 4.116 | 242929 | 26.9 FastCrc32c|1 | 4.037 | 247727 | 27.4 FastCrc32c|2 | 4.648 | 215166 | 23.8 FastCrc32c|3 | 4.352 | 229799 | 25.4 AVG Algorithm | Average of micros/op | Average of ops/sec | Average of Throughput (MB/s) 3-way | 4.01 | 249,729 | 27.63 FastCrc32c | 4.35 | 230,897 | 25.53 2) Crc32c computation CPU cost (inclusive samples percentage) PER RUN Implementation | run |  TotalSamples | Crc32c percentage 3-way   | 1    |  4,572,250,000 | 4.37% 3-way   | 2    |  3,779,250,000 | 4.62% 3-way   | 3    |  4,129,500,000 | 4.48% FastCrc32c     | 1    |  4,663,500,000 | 11.24% FastCrc32c     | 2    |  4,047,500,000 | 12.34% FastCrc32c     | 3    |  4,366,750,000 | 11.68% **# Test Plan** make -j64 corruption_test && ./corruption_test By default it uses 3-way SSE algorithm NO_THREEWAY_CRC32C=1 make -j64 corruption_test && ./corruption_test make clean && DEBUG_LEVEL=0 make -j64 db_bench make clean && DEBUG_LEVEL=0 NO_THREEWAY_CRC32C=1 make -j64 db_bench Closes https://github.com/facebook/rocksdb/pull/3173 Differential Revision: D6330882 Pulled By: yingsu00 fbshipit-source-id: 8ec3d89719533b63b536a736663ca6f0dd4482e9
2017-12-20 03:20:50 +01:00
PROPERTIES COMPILE_FLAGS "-msse4.2 -mpclmul")
endif()
endif()
if(HAVE_POWER8)
list(APPEND SOURCES
util/crc32c_ppc.c
util/crc32c_ppc_asm.S)
endif(HAVE_POWER8)
if(WIN32)
list(APPEND SOURCES
port/win/io_win.cc
port/win/env_win.cc
port/win/env_default.cc
port/win/port_win.cc
port/win/win_logger.cc
port/win/win_thread.cc)
if(WITH_XPRESS)
list(APPEND SOURCES
port/win/xpress_win.cc)
endif()
if(WITH_JEMALLOC)
list(APPEND SOURCES
port/win/win_jemalloc.cc)
endif()
else()
list(APPEND SOURCES
port/port_posix.cc
env/env_posix.cc
env/io_posix.cc)
endif()
set(ROCKSDB_STATIC_LIB rocksdb${ARTIFACT_SUFFIX})
set(ROCKSDB_SHARED_LIB rocksdb-shared${ARTIFACT_SUFFIX})
set(ROCKSDB_IMPORT_LIB ${ROCKSDB_SHARED_LIB})
option(WITH_LIBRADOS "Build with librados" OFF)
if(WITH_LIBRADOS)
list(APPEND SOURCES
utilities/env_librados.cc)
list(APPEND THIRDPARTY_LIBS rados)
endif()
if(WIN32)
set(SYSTEM_LIBS ${SYSTEM_LIBS} Shlwapi.lib Rpcrt4.lib)
set(LIBS ${ROCKSDB_STATIC_LIB} ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
else()
cross-platform compatibility improvements Summary: We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS. See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build. I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream. Closes https://github.com/facebook/rocksdb/pull/2199 Differential Revision: D5054042 Pulled By: yiwu-arbug fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
2017-05-15 23:42:32 +02:00
set(SYSTEM_LIBS ${CMAKE_THREAD_LIBS_INIT})
set(LIBS ${ROCKSDB_SHARED_LIB} ${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
add_library(${ROCKSDB_SHARED_LIB} SHARED ${SOURCES})
target_link_libraries(${ROCKSDB_SHARED_LIB}
${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
set_target_properties(${ROCKSDB_SHARED_LIB} PROPERTIES
LINKER_LANGUAGE CXX
VERSION ${ROCKSDB_VERSION}
SOVERSION ${ROCKSDB_VERSION_MAJOR}
CXX_STANDARD 11
OUTPUT_NAME "rocksdb")
endif()
add_library(${ROCKSDB_STATIC_LIB} STATIC ${SOURCES})
target_link_libraries(${ROCKSDB_STATIC_LIB}
${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
if(WIN32)
add_library(${ROCKSDB_IMPORT_LIB} SHARED ${SOURCES})
target_link_libraries(${ROCKSDB_IMPORT_LIB}
${THIRDPARTY_LIBS} ${SYSTEM_LIBS})
set_target_properties(${ROCKSDB_IMPORT_LIB} PROPERTIES
COMPILE_DEFINITIONS "ROCKSDB_DLL;ROCKSDB_LIBRARY_EXPORTS")
if(MSVC)
set_target_properties(${ROCKSDB_STATIC_LIB} PROPERTIES
COMPILE_FLAGS "/Fd${CMAKE_CFG_INTDIR}/${ROCKSDB_STATIC_LIB}.pdb")
set_target_properties(${ROCKSDB_IMPORT_LIB} PROPERTIES
COMPILE_FLAGS "/Fd${CMAKE_CFG_INTDIR}/${ROCKSDB_IMPORT_LIB}.pdb")
endif()
endif()
option(WITH_JNI "build with JNI" OFF)
if(WITH_JNI OR JNI)
message(STATUS "JNI library is enabled")
add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/java)
2016-01-28 15:44:31 +01:00
else()
message(STATUS "JNI library is disabled")
endif()
# Installation and packaging
if(WIN32)
option(ROCKSDB_INSTALL_ON_WINDOWS "Enable install target on Windows" OFF)
endif()
if(NOT WIN32 OR ROCKSDB_INSTALL_ON_WINDOWS)
if(CMAKE_INSTALL_PREFIX_INITIALIZED_TO_DEFAULT)
if(${CMAKE_SYSTEM_NAME} STREQUAL "Linux")
# Change default installation prefix on Linux to /usr
set(CMAKE_INSTALL_PREFIX /usr CACHE PATH "Install path prefix, prepended onto install directories." FORCE)
endif()
endif()
include(GNUInstallDirs)
include(CMakePackageConfigHelpers)
set(package_config_destination ${CMAKE_INSTALL_LIBDIR}/cmake/rocksdb)
configure_package_config_file(
${CMAKE_CURRENT_LIST_DIR}/cmake/RocksDBConfig.cmake.in RocksDBConfig.cmake
INSTALL_DESTINATION ${package_config_destination}
)
write_basic_package_version_file(
RocksDBConfigVersion.cmake
VERSION ${ROCKSDB_VERSION}
COMPATIBILITY SameMajorVersion
)
install(DIRECTORY include/rocksdb COMPONENT devel DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}")
install(
TARGETS ${ROCKSDB_STATIC_LIB}
EXPORT RocksDBTargets
COMPONENT devel
ARCHIVE DESTINATION "${CMAKE_INSTALL_LIBDIR}"
INCLUDES DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}"
)
install(
TARGETS ${ROCKSDB_SHARED_LIB}
EXPORT RocksDBTargets
COMPONENT runtime
RUNTIME DESTINATION "${CMAKE_INSTALL_BINDIR}"
LIBRARY DESTINATION "${CMAKE_INSTALL_LIBDIR}"
INCLUDES DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}"
)
install(
EXPORT RocksDBTargets
COMPONENT devel
DESTINATION ${package_config_destination}
NAMESPACE RocksDB::
)
install(
FILES
${CMAKE_CURRENT_BINARY_DIR}/RocksDBConfig.cmake
${CMAKE_CURRENT_BINARY_DIR}/RocksDBConfigVersion.cmake
COMPONENT devel
DESTINATION ${package_config_destination}
)
endif()
option(WITH_TESTS "build with tests" ON)
if(WITH_TESTS)
set(TESTS
cache/cache_test.cc
cache/lru_cache_test.cc
db/column_family_test.cc
db/compact_files_test.cc
db/compaction_iterator_test.cc
db/compaction_job_stats_test.cc
db/compaction_job_test.cc
db/compaction_picker_test.cc
db/comparator_db_test.cc
db/corruption_test.cc
db/cuckoo_table_db_test.cc
db/db_basic_test.cc
db/db_blob_index_test.cc
db/db_block_cache_test.cc
db/db_bloom_filter_test.cc
db/db_compaction_filter_test.cc
db/db_compaction_test.cc
db/db_dynamic_level_test.cc
db/db_flush_test.cc
db/db_inplace_update_test.cc
db/db_io_failure_test.cc
db/db_iter_test.cc
Change and clarify the relationship between Valid(), status() and Seek*() for all iterators. Also fix some bugs Summary: Before this PR, Iterator/InternalIterator may simultaneously have non-ok status() and Valid() = true. That state means that the last operation failed, but the iterator is nevertheless positioned on some unspecified record. Likely intended uses of that are: * If some sst files are corrupted, a normal iterator can be used to read the data from files that are not corrupted. * When using read_tier = kBlockCacheTier, read the data that's in block cache, skipping over the data that is not. However, this behavior wasn't documented well (and until recently the wiki on github had misleading incorrect information). In the code there's a lot of confusion about the relationship between status() and Valid(), and about whether Seek()/SeekToLast()/etc reset the status or not. There were a number of bugs caused by this confusion, both inside rocksdb and in the code that uses rocksdb (including ours). This PR changes the convention to: * If status() is not ok, Valid() always returns false. * Any seek operation resets status. (Before the PR, it depended on iterator type and on particular error.) This does sacrifice the two use cases listed above, but siying said it's ok. Overview of the changes: * A commit that adds missing status checks in MergingIterator. This fixes a bug that actually affects us, and we need it fixed. `DBIteratorTest.NonBlockingIterationBugRepro` explains the scenario. * Changes to lots of iterator types to make all of them conform to the new convention. Some bug fixes along the way. By far the biggest changes are in DBIter, which is a big messy piece of code; I tried to make it less big and messy but mostly failed. * A stress-test for DBIter, to gain some confidence that I didn't break it. It does a few million random operations on the iterator, while occasionally modifying the underlying data (like ForwardIterator does) and occasionally returning non-ok status from internal iterator. To find the iterator types that needed changes I searched for "public .*Iterator" in the code. Here's an overview of all 27 iterator types: Iterators that didn't need changes: * status() is always ok(), or Valid() is always false: MemTableIterator, ModelIter, TestIterator, KVIter (2 classes with this name anonymous namespaces), LoggingForwardVectorIterator, VectorIterator, MockTableIterator, EmptyIterator, EmptyInternalIterator. * Thin wrappers that always pass through Valid() and status(): ArenaWrappedDBIter, TtlIterator, InternalIteratorFromIterator. Iterators with changes (see inline comments for details): * DBIter - an overhaul: - It used to silently skip corrupted keys (`FindParseableKey()`), which seems dangerous. This PR makes it just stop immediately after encountering a corrupted key, just like it would for other kinds of corruption. Let me know if there was actually some deeper meaning in this behavior and I should put it back. - It had a few code paths silently discarding subiterator's status. The stress test caught a few. - The backwards iteration code path was expecting the internal iterator's set of keys to be immutable. It's probably always true in practice at the moment, since ForwardIterator doesn't support backwards iteration, but this PR fixes it anyway. See added DBIteratorTest.ReverseToForwardBug for an example. - Some parts of backwards iteration code path even did things like `assert(iter_->Valid())` after a seek, which is never a safe assumption. - It used to not reset status on seek for some types of errors. - Some simplifications and better comments. - Some things got more complicated from the added error handling. I'm open to ideas for how to make it nicer. * MergingIterator - check status after every operation on every subiterator, and in some places assert that valid subiterators have ok status. * ForwardIterator - changed to the new convention, also slightly simplified. * ForwardLevelIterator - fixed some bugs and simplified. * LevelIterator - simplified. * TwoLevelIterator - changed to the new convention. Also fixed a bug that would make SeekForPrev() sometimes silently ignore errors from first_level_iter_. * BlockBasedTableIterator - minor changes. * BlockIter - replaced `SetStatus()` with `Invalidate()` to make sure non-ok BlockIter is always invalid. * PlainTableIterator - some seeks used to not reset status. * CuckooTableIterator - tiny code cleanup. * ManagedIterator - fixed some bugs. * BaseDeltaIterator - changed to the new convention and fixed a bug. * BlobDBIterator - seeks used to not reset status. * KeyConvertingIterator - some small change. Closes https://github.com/facebook/rocksdb/pull/3810 Differential Revision: D7888019 Pulled By: al13n321 fbshipit-source-id: 4aaf6d3421c545d16722a815b2fa2e7912bc851d
2018-05-17 11:44:14 +02:00
db/db_iter_stress_test.cc
db/db_iterator_test.cc
db/db_log_iter_test.cc
db/db_memtable_test.cc
db/db_merge_operator_test.cc
db/db_options_test.cc
db/db_properties_test.cc
db/db_range_del_test.cc
db/db_sst_test.cc
db/db_statistics_test.cc
db/db_table_properties_test.cc
db/db_tailing_iter_test.cc
db/db_test.cc
db/db_test2.cc
db/db_universal_compaction_test.cc
db/db_wal_test.cc
db/db_write_test.cc
db/dbformat_test.cc
db/deletefile_test.cc
Fix race condition causing double deletion of ssts Summary: Possible interleaved execution of background compaction thread calling `FindObsoleteFiles (no full scan) / PurgeObsoleteFiles` and user thread calling `FindObsoleteFiles (full scan) / PurgeObsoleteFiles` can lead to race condition on which RocksDB attempts to delete a file twice. The second attempt will fail and return `IO error`. This may occur to other files, but this PR targets sst. Also add a unit test to verify that this PR fixes the issue. The newly added unit test `obsolete_files_test` has a test case for this scenario, implemented in `ObsoleteFilesTest#RaceForObsoleteFileDeletion`. `TestSyncPoint`s are used to coordinate the interleaving the `user_thread` and background compaction thread. They execute as follows ``` timeline user_thread background_compaction thread t1 | FindObsoleteFiles(full_scan=false) t2 | FindObsoleteFiles(full_scan=true) t3 | PurgeObsoleteFiles t4 | PurgeObsoleteFiles V ``` When `user_thread` invokes `FindObsoleteFiles` with full scan, it collects ALL files in RocksDB directory, including the ones that background compaction thread have collected in its job context. Then `user_thread` will see an IO error when trying to delete these files in `PurgeObsoleteFiles` because background compaction thread has already deleted the file in `PurgeObsoleteFiles`. To fix this, we make RocksDB remember which (SST) files have been found by threads after calling `FindObsoleteFiles` (see `DBImpl#files_grabbed_for_purge_`). Therefore, when another thread calls `FindObsoleteFiles` with full scan, it will not collect such files. ajkr could you take a look and comment? Thanks! Closes https://github.com/facebook/rocksdb/pull/3638 Differential Revision: D7384372 Pulled By: riversand963 fbshipit-source-id: 01489516d60012e722ee65a80e1449e589ce26d3
2018-03-28 19:23:31 +02:00
db/obsolete_files_test.cc
db/external_sst_file_basic_test.cc
db/external_sst_file_test.cc
db/fault_injection_test.cc
db/file_indexer_test.cc
db/filename_test.cc
db/flush_job_test.cc
db/listener_test.cc
db/log_test.cc
db/manual_compaction_test.cc
db/memtable_list_test.cc
db/merge_helper_test.cc
db/merge_test.cc
db/options_file_test.cc
db/perf_context_test.cc
db/plain_table_db_test.cc
db/prefix_test.cc
db/repair_test.cc
db/table_properties_collector_test.cc
db/version_builder_test.cc
db/version_edit_test.cc
db/version_set_test.cc
db/wal_manager_test.cc
db/write_batch_test.cc
db/write_callback_test.cc
db/write_controller_test.cc
env/env_basic_test.cc
env/env_test.cc
env/mock_env_test.cc
memtable/inlineskiplist_test.cc
memtable/skiplist_test.cc
memtable/write_buffer_manager_test.cc
monitoring/histogram_test.cc
monitoring/iostats_context_test.cc
monitoring/statistics_test.cc
options/options_settable_test.cc
options/options_test.cc
table/block_based_filter_block_test.cc
table/block_test.cc
table/cleanable_test.cc
table/cuckoo_table_builder_test.cc
table/cuckoo_table_reader_test.cc
table/full_filter_block_test.cc
table/merger_test.cc
table/table_test.cc
tools/ldb_cmd_test.cc
tools/reduce_levels_test.cc
tools/sst_dump_test.cc
util/arena_test.cc
util/auto_roll_logger_test.cc
util/autovector_test.cc
util/bloom_test.cc
util/coding_test.cc
util/crc32c_test.cc
util/delete_scheduler_test.cc
util/dynamic_bloom_test.cc
util/event_logger_test.cc
util/file_reader_writer_test.cc
util/filelock_test.cc
util/hash_test.cc
util/heap_test.cc
util/rate_limiter_test.cc
util/slice_transform_test.cc
util/timer_queue_test.cc
util/thread_list_test.cc
util/thread_local_test.cc
utilities/backupable/backupable_db_test.cc
utilities/blob_db/blob_db_test.cc
utilities/cassandra/cassandra_functional_test.cc
utilities/cassandra/cassandra_format_test.cc
utilities/cassandra/cassandra_row_merge_test.cc
utilities/cassandra/cassandra_serialize_test.cc
utilities/checkpoint/checkpoint_test.cc
utilities/column_aware_encoding_test.cc
utilities/date_tiered/date_tiered_test.cc
utilities/document/document_db_test.cc
utilities/document/json_document_test.cc
utilities/geodb/geodb_test.cc
utilities/lua/rocks_lua_test.cc
utilities/memory/memory_test.cc
utilities/merge_operators/string_append/stringappend_test.cc
utilities/object_registry_test.cc
utilities/option_change_migration/option_change_migration_test.cc
2015-11-12 23:53:19 +01:00
utilities/options/options_util_test.cc
utilities/persistent_cache/hash_table_test.cc
utilities/persistent_cache/persistent_cache_test.cc
utilities/redis/redis_lists_test.cc
utilities/spatialdb/spatial_db_test.cc
utilities/simulator_cache/sim_cache_test.cc
utilities/table_properties_collectors/compact_on_deletion_collector_test.cc
utilities/transactions/optimistic_transaction_test.cc
utilities/transactions/transaction_test.cc
utilities/transactions/write_prepared_transaction_test.cc
utilities/transactions/write_unprepared_transaction_test.cc
utilities/ttl/ttl_test.cc
utilities/write_batch_with_index/write_batch_with_index_test.cc
)
if(WITH_LIBRADOS)
list(APPEND TESTS utilities/env_librados_test.cc)
endif()
set(BENCHMARKS
cache/cache_bench.cc
memtable/memtablerep_bench.cc
tools/db_bench.cc
table/table_reader_bench.cc
utilities/column_aware_encoding_exp.cc
utilities/persistent_cache/hash_table_bench.cc)
add_library(testharness OBJECT util/testharness.cc)
foreach(sourcefile ${BENCHMARKS})
get_filename_component(exename ${sourcefile} NAME_WE)
add_executable(${exename}${ARTIFACT_SUFFIX} ${sourcefile}
$<TARGET_OBJECTS:testharness>)
target_link_libraries(${exename}${ARTIFACT_SUFFIX} gtest ${LIBS})
endforeach(sourcefile ${BENCHMARKS})
# For test util library that is build only in DEBUG mode
# and linked to tests. Add test only code that is not #ifdefed for Release here.
set(TESTUTIL_SOURCE
db/db_test_util.cc
monitoring/thread_status_updater_debug.cc
table/mock_table.cc
util/fault_injection_test_env.cc
utilities/cassandra/test_utils.cc
)
# test utilities are only build in debug
enable_testing()
add_custom_target(check COMMAND ${CMAKE_CTEST_COMMAND})
set(TESTUTILLIB testutillib${ARTIFACT_SUFFIX})
add_library(${TESTUTILLIB} STATIC ${TESTUTIL_SOURCE})
if(MSVC)
set_target_properties(${TESTUTILLIB} PROPERTIES COMPILE_FLAGS "/Fd${CMAKE_CFG_INTDIR}/testutillib${ARTIFACT_SUFFIX}.pdb")
endif()
set_target_properties(${TESTUTILLIB}
PROPERTIES EXCLUDE_FROM_DEFAULT_BUILD_RELEASE 1
EXCLUDE_FROM_DEFAULT_BUILD_MINRELEASE 1
EXCLUDE_FROM_DEFAULT_BUILD_RELWITHDEBINFO 1
)
# Tests are excluded from Release builds
set(TEST_EXES ${TESTS})
foreach(sourcefile ${TEST_EXES})
get_filename_component(exename ${sourcefile} NAME_WE)
add_executable(${CMAKE_PROJECT_NAME}_${exename}${ARTIFACT_SUFFIX} ${sourcefile}
$<TARGET_OBJECTS:testharness>)
set_target_properties(${CMAKE_PROJECT_NAME}_${exename}${ARTIFACT_SUFFIX}
PROPERTIES EXCLUDE_FROM_DEFAULT_BUILD_RELEASE 1
EXCLUDE_FROM_DEFAULT_BUILD_MINRELEASE 1
EXCLUDE_FROM_DEFAULT_BUILD_RELWITHDEBINFO 1
OUTPUT_NAME ${exename}${ARTIFACT_SUFFIX}
)
target_link_libraries(${CMAKE_PROJECT_NAME}_${exename}${ARTIFACT_SUFFIX} testutillib${ARTIFACT_SUFFIX} gtest ${LIBS})
if(NOT "${exename}" MATCHES "db_sanity_test")
add_test(NAME ${exename} COMMAND ${exename}${ARTIFACT_SUFFIX})
add_dependencies(check ${CMAKE_PROJECT_NAME}_${exename}${ARTIFACT_SUFFIX})
endif()
endforeach(sourcefile ${TEST_EXES})
# C executables must link to a shared object
set(C_TESTS db/c_test.c)
set(C_TEST_EXES ${C_TESTS})
foreach(sourcefile ${C_TEST_EXES})
string(REPLACE ".c" "" exename ${sourcefile})
string(REGEX REPLACE "^((.+)/)+" "" exename ${exename})
add_executable(${exename}${ARTIFACT_SUFFIX} ${sourcefile})
set_target_properties(${exename}${ARTIFACT_SUFFIX}
PROPERTIES EXCLUDE_FROM_DEFAULT_BUILD_RELEASE 1
EXCLUDE_FROM_DEFAULT_BUILD_MINRELEASE 1
EXCLUDE_FROM_DEFAULT_BUILD_RELWITHDEBINFO 1
)
target_link_libraries(${exename}${ARTIFACT_SUFFIX} ${ROCKSDB_IMPORT_LIB} testutillib${ARTIFACT_SUFFIX})
add_test(NAME ${exename} COMMAND ${exename}${ARTIFACT_SUFFIX})
add_dependencies(check ${exename}${ARTIFACT_SUFFIX})
endforeach(sourcefile ${C_TEST_EXES})
endif()
option(WITH_TOOLS "build with tools" ON)
if(WITH_TOOLS)
add_subdirectory(tools)
endif()