rocksdb/build_tools/build_detect_platform
Radheshyam Balasundaram f0660d5253 Adding NUMA support to db_bench tests
Summary:
Changes:
- Adding numa_aware flag to db_bench.cc
- Using numa.h library to bind memory and cpu of threads to a fixed NUMA node
Result: There seems to be no significant change in the micros/op time with numa_aware enabled. I also tried this with other implementations, including a combination of pthread_setaffinity_np, sched_setaffinity and set_mempolicy methods. It'd be great if someone could point out where I'm going wrong and if we can achieve a better micors/op.

Test Plan:
Ran db_bench tests using following command:
./db_bench --db=/mnt/tmp --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=134217728 --max_bytes_for_level_base=1073741824 --disable_wal=0 --wal_dir=/mnt/tmp --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --duration=300 --benchmarks=readwhilewriting --use_existing_db=1 --num=157286400 --threads=24 --writes_per_second=10240 --numa_aware=[False/True]

The tests were run in private devserver with 24 cores and the db was prepopulated using filluniquerandom test. The tests resulted in 0.145 us/op with numa_aware=False and 0.161 us/op with numa_aware=True.

Reviewers: sdong, yhchiang, ljin, igor

Reviewed By: ljin, igor

Subscribers: igor, leveldb

Differential Revision: https://reviews.facebook.net/D19353
2014-07-07 10:53:31 -07:00

333 lines
11 KiB
Bash
Executable File

#!/bin/sh
#
# Detects OS we're compiling on and outputs a file specified by the first
# argument, which in turn gets read while processing Makefile.
#
# The output will set the following variables:
# CC C Compiler path
# CXX C++ Compiler path
# PLATFORM_LDFLAGS Linker flags
# PLATFORM_SHARED_EXT Extension for shared libraries
# PLATFORM_SHARED_LDFLAGS Flags for building shared library
# PLATFORM_SHARED_CFLAGS Flags for compiling objects for shared library
# PLATFORM_CCFLAGS C compiler flags
# PLATFORM_CXXFLAGS C++ compiler flags. Will contain:
# PLATFORM_SHARED_VERSIONED Set to 'true' if platform supports versioned
# shared libraries, empty otherwise.
#
# The PLATFORM_CCFLAGS and PLATFORM_CXXFLAGS might include the following:
#
# -DLEVELDB_PLATFORM_POSIX if cstdatomic is present
# -DLEVELDB_PLATFORM_NOATOMIC if it is not
# -DSNAPPY if the Snappy library is present
# -DLZ4 if the LZ4 library is present
# -DNUMA if the NUMA library is present
#
# Using gflags in rocksdb:
# Our project depends on gflags, which requires users to take some extra steps
# before they can compile the whole repository:
# 1. Install gflags. You may download it from here:
# https://code.google.com/p/gflags/
# 2. Once install, add the include path/lib path for gflags to CPATH and
# LIBRARY_PATH respectively. If installed with default mode, the
# lib and include path will be /usr/local/lib and /usr/local/include
# Mac user can do this by running build_tools/mac-install-gflags.sh
OUTPUT=$1
if test -z "$OUTPUT"; then
echo "usage: $0 <output-filename>" >&2
exit 1
fi
# we depend on C++11
PLATFORM_CXXFLAGS="-std=c++11"
# we currently depend on POSIX platform
COMMON_FLAGS="-DROCKSDB_PLATFORM_POSIX"
# Default to fbcode gcc on internal fb machines
if [ -d /mnt/gvfs/third-party -a -z "$CXX" ]; then
FBCODE_BUILD="true"
if [ -z "$USE_CLANG" ]; then
CENTOS_VERSION=`rpm -q --qf "%{VERSION}" \
$(rpm -q --whatprovides redhat-release)`
if [ "$CENTOS_VERSION" = "6" ]; then
source "$PWD/build_tools/fbcode.gcc481.sh"
else
source "$PWD/build_tools/fbcode.gcc471.sh"
fi
else
source "$PWD/build_tools/fbcode.clang31.sh"
fi
fi
# Delete existing output, if it exists
rm -f "$OUTPUT"
touch "$OUTPUT"
if test -z "$CC"; then
CC=cc
fi
if test -z "$CXX"; then
CXX=g++
fi
# Detect OS
if test -z "$TARGET_OS"; then
TARGET_OS=`uname -s`
fi
COMMON_FLAGS="$COMMON_FLAGS ${CFLAGS}"
CROSS_COMPILE=
PLATFORM_CCFLAGS=
PLATFORM_CXXFLAGS="$PLATFORM_CXXFLAGS ${CXXFLAGS}"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS"
PLATFORM_SHARED_EXT="so"
PLATFORM_SHARED_LDFLAGS="-shared -Wl,-soname -Wl,"
PLATFORM_SHARED_CFLAGS="-fPIC"
PLATFORM_SHARED_VERSIONED=false
# generic port files (working on all platform by #ifdef) go directly in /port
GENERIC_PORT_FILES=`cd "$ROCKSDB_ROOT"; find port -name '*.cc' | tr "\n" " "`
# On GCC, we pick libc's memcmp over GCC's memcmp via -fno-builtin-memcmp
case "$TARGET_OS" in
Darwin)
PLATFORM=OS_MACOSX
COMMON_FLAGS="$COMMON_FLAGS -DOS_MACOSX"
PLATFORM_SHARED_EXT=dylib
PLATFORM_SHARED_LDFLAGS="-dynamiclib -install_name "
# PORT_FILES=port/darwin/darwin_specific.cc
;;
IOS)
PLATFORM=IOS
COMMON_FLAGS="$COMMON_FLAGS -DOS_MACOSX -DIOS_CROSS_COMPILE -DROCKSDB_LITE"
PLATFORM_SHARED_EXT=dylib
PLATFORM_SHARED_LDFLAGS="-dynamiclib -install_name "
CROSS_COMPILE=true
;;
Linux)
PLATFORM=OS_LINUX
COMMON_FLAGS="$COMMON_FLAGS -DOS_LINUX"
if [ -z "$USE_CLANG" ]; then
COMMON_FLAGS="$COMMON_FLAGS -fno-builtin-memcmp"
fi
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lpthread -lrt"
# PORT_FILES=port/linux/linux_specific.cc
;;
SunOS)
PLATFORM=OS_SOLARIS
COMMON_FLAGS="$COMMON_FLAGS -fno-builtin-memcmp -D_REENTRANT -DOS_SOLARIS"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lpthread -lrt"
# PORT_FILES=port/sunos/sunos_specific.cc
;;
FreeBSD)
PLATFORM=OS_FREEBSD
COMMON_FLAGS="$COMMON_FLAGS -fno-builtin-memcmp -D_REENTRANT -DOS_FREEBSD"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lpthread"
# PORT_FILES=port/freebsd/freebsd_specific.cc
;;
NetBSD)
PLATFORM=OS_NETBSD
COMMON_FLAGS="$COMMON_FLAGS -fno-builtin-memcmp -D_REENTRANT -DOS_NETBSD"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lpthread -lgcc_s"
# PORT_FILES=port/netbsd/netbsd_specific.cc
;;
OpenBSD)
PLATFORM=OS_OPENBSD
COMMON_FLAGS="$COMMON_FLAGS -fno-builtin-memcmp -D_REENTRANT -DOS_OPENBSD"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -pthread"
# PORT_FILES=port/openbsd/openbsd_specific.cc
;;
DragonFly)
PLATFORM=OS_DRAGONFLYBSD
COMMON_FLAGS="$COMMON_FLAGS -fno-builtin-memcmp -D_REENTRANT -DOS_DRAGONFLYBSD"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lpthread"
# PORT_FILES=port/dragonfly/dragonfly_specific.cc
;;
OS_ANDROID_CROSSCOMPILE)
PLATFORM=OS_ANDROID
COMMON_FLAGS="$COMMON_FLAGS -fno-builtin-memcmp -D_REENTRANT -DOS_ANDROID -DLEVELDB_PLATFORM_POSIX"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS " # All pthread features are in the Android C library
# PORT_FILES=port/android/android.cc
CROSS_COMPILE=true
;;
*)
echo "Unknown platform!" >&2
exit 1
esac
if test -z "$DO_NOT_RUN_BUILD_DETECT_VERSION"; then
"$PWD/build_tools/build_detect_version"
fi
# We want to make a list of all cc files within util, db, table, and helpers
# except for the test and benchmark files. By default, find will output a list
# of all files matching either rule, so we need to append -print to make the
# prune take effect.
DIRS="util db table utilities"
set -f # temporarily disable globbing so that our patterns arent expanded
PRUNE_TEST="-name *test*.cc -prune"
PRUNE_BENCH="-name *bench*.cc -prune"
PORTABLE_FILES=`cd "$ROCKSDB_ROOT"; find $DIRS $PRUNE_TEST -o $PRUNE_BENCH -o -name '*.cc' -print | sort | tr "\n" " "`
PORTABLE_CPP=`cd "$ROCKSDB_ROOT"; find $DIRS $PRUNE_TEST -o $PRUNE_BENCH -o -name '*.cpp' -print | sort | tr "\n" " "`
set +f # re-enable globbing
# The sources consist of the portable files, plus the platform-specific port
# file.
echo "SOURCES=$PORTABLE_FILES $GENERIC_PORT_FILES $PORT_FILES" >> "$OUTPUT"
echo "SOURCESCPP=$PORTABLE_CPP" >> "$OUTPUT"
echo "MEMENV_SOURCES=helpers/memenv/memenv.cc" >> "$OUTPUT"
if [ "$CROSS_COMPILE" = "true" -o "$FBCODE_BUILD" = "true" ]; then
# Cross-compiling; do not try any compilation tests.
# Also don't need any compilation tests if compiling on fbcode
true
else
# If -std=c++0x works, use <atomic>. Otherwise use port_posix.h.
$CXX $CFLAGS -std=c++0x -x c++ - -o /dev/null 2>/dev/null <<EOF
#include <atomic>
int main() {}
EOF
if [ "$?" = 0 ]; then
COMMON_FLAGS="$COMMON_FLAGS -DROCKSDB_ATOMIC_PRESENT"
fi
# Test whether fallocate is available
$CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null <<EOF
#include <fcntl.h>
int main() {
int fd = open("/dev/null", 0);
fallocate(fd, 0, 0, 1024);
}
EOF
if [ "$?" = 0 ]; then
COMMON_FLAGS="$COMMON_FLAGS -DROCKSDB_FALLOCATE_PRESENT"
fi
# Test whether Snappy library is installed
# http://code.google.com/p/snappy/
$CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null <<EOF
#include <snappy.h>
int main() {}
EOF
if [ "$?" = 0 ]; then
COMMON_FLAGS="$COMMON_FLAGS -DSNAPPY"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lsnappy"
fi
# Test whether gflags library is installed
# http://code.google.com/p/gflags/
# check if the namespace is gflags
$CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null <<EOF
#include <gflags/gflags.h>
using namespace gflags;
int main() {}
EOF
if [ "$?" = 0 ]; then
COMMON_FLAGS="$COMMON_FLAGS -DGFLAGS=gflags"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lgflags"
fi
# check if namespace is google
$CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null <<EOF
#include <gflags/gflags.h>
using namespace google;
int main() {}
EOF
if [ "$?" = 0 ]; then
COMMON_FLAGS="$COMMON_FLAGS -DGFLAGS=google"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lgflags"
fi
# Test whether zlib library is installed
$CXX $CFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null <<EOF
#include <zlib.h>
int main() {}
EOF
if [ "$?" = 0 ]; then
COMMON_FLAGS="$COMMON_FLAGS -DZLIB"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lz"
fi
# Test whether bzip library is installed
$CXX $CFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null <<EOF
#include <bzlib.h>
int main() {}
EOF
if [ "$?" = 0 ]; then
COMMON_FLAGS="$COMMON_FLAGS -DBZIP2"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lbz2"
fi
# Test whether lz4 library is installed
$CXX $CFLAGS $COMMON_FLAGS -x c++ - -o /dev/null 2>/dev/null <<EOF
#include <lz4.h>
#include <lz4hc.h>
int main() {}
EOF
if [ "$?" = 0 ]; then
COMMON_FLAGS="$COMMON_FLAGS -DLZ4"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -llz4"
fi
# Test whether numa is available
$CXX $CFLAGS -x c++ - -o /dev/null -lnuma 2>/dev/null <<EOF
#include <numa.h>
#inlcude <numaif.h>
int main() {}
EOF
if [ "$?" = 0 ]; then
COMMON_FLAGS="$COMMON_FLAGS -DNUMA"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -lnuma"
fi
# Test whether tcmalloc is available
$CXX $CFLAGS -x c++ - -o /dev/null -ltcmalloc 2>/dev/null <<EOF
int main() {}
EOF
if [ "$?" = 0 ]; then
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS -ltcmalloc"
fi
fi
# shall we use HDFS?
if test "$USE_HDFS"; then
if test -z "$JAVA_HOME"; then
echo "JAVA_HOME has to be set for HDFS usage."
exit 1
fi
HDFS_CCFLAGS="$HDFS_CCFLAGS -I$JAVA_HOME/include -I$JAVA_HOME/include/linux -DUSE_HDFS"
HDFS_LDFLAGS="$HDFS_LDFLAGS -Wl,--no-whole-archive -lhdfs -L$JAVA_HOME/jre/lib/amd64"
HDFS_LDFLAGS="$HDFS_LDFLAGS -L$JAVA_HOME/jre/lib/amd64/server -L$GLIBC_RUNTIME_PATH/lib"
HDFS_LDFLAGS="$HDFS_LDFLAGS -ldl -lverify -ljava -ljvm"
COMMON_FLAGS="$COMMON_FLAGS $HDFS_CCFLAGS"
PLATFORM_LDFLAGS="$PLATFORM_LDFLAGS $HDFS_LDFLAGS"
fi
# if Intel SSE instruction set is supported, set USE_SSE=" -msse -msse4.2 "
COMMON_FLAGS="$COMMON_FLAGS $USE_SSE"
PLATFORM_CCFLAGS="$PLATFORM_CCFLAGS $COMMON_FLAGS"
PLATFORM_CXXFLAGS="$PLATFORM_CXXFLAGS $COMMON_FLAGS"
VALGRIND_VER="$VALGRIND_VER"
echo "CC=$CC" >> "$OUTPUT"
echo "CXX=$CXX" >> "$OUTPUT"
echo "PLATFORM=$PLATFORM" >> "$OUTPUT"
echo "PLATFORM_LDFLAGS=$PLATFORM_LDFLAGS" >> "$OUTPUT"
echo "VALGRIND_VER=$VALGRIND_VER" >> "$OUTPUT"
echo "PLATFORM_CCFLAGS=$PLATFORM_CCFLAGS" >> "$OUTPUT"
echo "PLATFORM_CXXFLAGS=$PLATFORM_CXXFLAGS" >> "$OUTPUT"
echo "PLATFORM_SHARED_CFLAGS=$PLATFORM_SHARED_CFLAGS" >> "$OUTPUT"
echo "PLATFORM_SHARED_EXT=$PLATFORM_SHARED_EXT" >> "$OUTPUT"
echo "PLATFORM_SHARED_LDFLAGS=$PLATFORM_SHARED_LDFLAGS" >> "$OUTPUT"
echo "PLATFORM_SHARED_VERSIONED=$PLATFORM_SHARED_VERSIONED" >> "$OUTPUT"
echo "EXEC_LDFLAGS=$EXEC_LDFLAGS" >> "$OUTPUT"
echo "JEMALLOC_INCLUDE=$JEMALLOC_INCLUDE" >> "$OUTPUT"
echo "JEMALLOC_LIB=$JEMALLOC_LIB" >> "$OUTPUT"