Benchmarks by Michael Larabel for a future article.
New Default - Generic Retpolines Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-yTrUTS/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0xa201016Java Notes: OpenJDK Runtime Environment (build 11.0.14+9-Ubuntu-0ubuntu2.20.04)Python Notes: Python 2.7.18 + Python 3.8.10Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
AMD / LFENCE Prior Default Processor: AMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3904 BIOS), Chipset: AMD Starship/Matisse, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 Plus, Graphics: AMD Radeon RX 6800 16GB (2475/1000MHz), Audio: AMD Device ab28, Monitor: ASUS MG28U, Network: Realtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200
OS: Ubuntu 20.04, Kernel: 5.17.0-051700rc7daily20220310-generic (x86_64), Desktop: GNOME Shell 3.36.9, Display Server: X Server 1.20.13, OpenGL: 4.6 Mesa 22.0.0-devel (LLVM 13.0.1 DRM 3.44), Compiler: GCC 9.4.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-yTrUTS/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0xa201016Java Notes: OpenJDK Runtime Environment (build 11.0.14+9-Ubuntu-0ubuntu2.20.04)Python Notes: Python 2.7.18 + Python 3.8.10Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Vulnerable: LFENCE + srbds: Not affected + tsx_async_abort: Not affected
AMD Retpolines Linux 5.17 OpenBenchmarking.org Phoronix Test Suite AMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads) ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3904 BIOS) AMD Starship/Matisse 32GB 1000GB Sabrent Rocket 4.0 Plus AMD Radeon RX 6800 16GB (2475/1000MHz) AMD Device ab28 ASUS MG28U Realtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200 Ubuntu 20.04 5.17.0-051700rc7daily20220310-generic (x86_64) GNOME Shell 3.36.9 X Server 1.20.13 4.6 Mesa 22.0.0-devel (LLVM 13.0.1 DRM 3.44) GCC 9.4.0 ext4 3840x2160 Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Compiler File-System Screen Resolution AMD Retpolines Linux 5.17 Benchmarks System Logs - Transparent Huge Pages: madvise - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-yTrUTS/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - NONE / errors=remount-ro,relatime,rw / Block Size: 4096 - Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0xa201016 - OpenJDK Runtime Environment (build 11.0.14+9-Ubuntu-0ubuntu2.20.04) - Python 2.7.18 + Python 3.8.10 - New Default - Generic Retpolines: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected - AMD / LFENCE Prior Default: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Vulnerable: LFENCE + srbds: Not affected + tsx_async_abort: Not affected
New Default - Generic Retpolines vs. AMD / LFENCE Prior Default Comparison Phoronix Test Suite Baseline +13.5% +13.5% +27% +27% +40.5% +40.5% 54% 6.4% 5.3% 5% 3.7% 3.3% 2.6% 2.1% Context Switching Create Threads Throughput Rand Write - IO_uring - No - Yes - 4KB D.T.P rotate W.i - Google Chrome 3.2% Create Files W.i - Firefox 2.4% Kraken - Firefox 2% Stress-NG OSBench Sockperf Flexible IO Tester PostMark GIMP Selenium OSBench Selenium WireGuard + Linux Networking Stack Stress Test Selenium New Default - Generic Retpolines AMD / LFENCE Prior Default
AMD Retpolines Linux 5.17 build-linux-kernel: allmodconfig selenium: Jetstream 2 - Firefox wireguard: selenium: ARES-6 - Firefox selenium: Jetstream 2 - Google Chrome keydb: stress-ng: Context Switching selenium: StyleBench - Google Chrome rocksdb: Rand Fill Sync rocksdb: Rand Fill rocksdb: Read Rand Write Rand build-linux-kernel: defconfig influxdb: 4 - 10000 - 2,5000,1 - 10000 influxdb: 64 - 10000 - 2,5000,1 - 10000 selenium: StyleBench - Firefox selenium: ARES-6 - Google Chrome sqlite-speedtest: Timed Time - Size 1,000 selenium: Speedometer - Google Chrome sockperf: Throughput dacapobench: H2 fio: Rand Read - IO_uring - No - Yes - 4KB - Default Test Directory selenium: PSPDFKit WASM - Google Chrome fio: Rand Write - IO_uring - No - Yes - 4KB - Default Test Directory stress-ng: Semaphores selenium: PSPDFKit WASM - Firefox postmark: Disk Transaction Performance selenium: Kraken - Firefox rocksdb: Seq Fill selenium: Kraken - Google Chrome darktable: Boat - CPU-only osbench: Create Threads selenium: WASM collisionDetection - Firefox leveldb: Rand Read sockperf: Latency Ping Pong dacapobench: Tradesoap gimp: unsharp-mask selenium: WASM collisionDetection - Google Chrome gimp: auto-levels dacapobench: Tradebeans gimp: rotate selenium: WASM imageConvolute - Google Chrome selenium: WASM imageConvolute - Firefox darktable: Masskrug - CPU-only darktable: Server Room - CPU-only gimp: resize dacapobench: Jython osbench: Create Files leveldb: Fill Sync darktable: Server Rack - CPU-only New Default - Generic Retpolines AMD / LFENCE Prior Default 658.321 135.176 164.136 28.69 232.327 825870.24 3874572.19 55.4 19837 1383010 2923053 53.204 2139058.0 2287292.1 121 11.09 42.671 249 796686 2230 321000 2841 518333 3280517.54 2471 8825 685.2 1526340 504.4 14.578 10.676384 278.8 13.679 2.513 2628 12.247 252.12 9.586 2952 8.791 20.91 20.6 6.475 5.875 5.499 2975 11.717551 1648.569 0.183 658.127 132.905 160.799 28.61 230.278 833156.26 5967964.27 55.1 19940 1385292 2951794 53.087 2135481.9 2288503.6 121 11.19 42.104 249 838652 2253 326000 2838 544333 3281839.70 2465 9149 698.8 1548865 498.9 14.553 10.037104 276.5 13.623 2.504 2625 12.143 250.81 9.509 2971 8.509 21.57 21.1 6.454 5.847 5.485 2988 11.423668 1636.440 0.182 OpenBenchmarking.org
Selenium OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Firefox AMD / LFENCE Prior Default New Default - Generic Retpolines 30 60 90 120 150 SE +/- 1.06, N = 3 SE +/- 0.74, N = 3 132.91 135.18 1. firefox 97.0.2
WireGuard + Linux Networking Stack Stress Test This is a benchmark of the WireGuard secure VPN tunnel and Linux networking stack stress test. The test runs on the local host but does require root permissions to run. The way it works is it creates three namespaces. ns0 has a loopback device. ns1 and ns2 each have wireguard devices. Those two wireguard devices send traffic through the loopback device of ns0. The end result of this is that tests wind up testing encryption and decryption at the same time -- a pretty CPU and scheduler-heavy workflow. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better WireGuard + Linux Networking Stack Stress Test AMD / LFENCE Prior Default New Default - Generic Retpolines 40 80 120 160 200 SE +/- 0.85, N = 3 SE +/- 1.09, N = 3 160.80 164.14
Selenium OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: ARES-6 - Browser: Firefox AMD / LFENCE Prior Default New Default - Generic Retpolines 7 14 21 28 35 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 28.61 28.69 1. firefox 97.0.2
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Google Chrome AMD / LFENCE Prior Default New Default - Generic Retpolines 50 100 150 200 250 SE +/- 2.34, N = 3 SE +/- 1.84, N = 3 230.28 232.33 1. chrome 99.0.4844.51
KeyDB A benchmark of KeyDB as a multi-threaded fork of the Redis server. The KeyDB benchmark is conducted using memtier-benchmark. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Ops/sec, More Is Better KeyDB 6.2.0 AMD / LFENCE Prior Default New Default - Generic Retpolines 200K 400K 600K 800K 1000K SE +/- 5837.82, N = 3 SE +/- 1303.16, N = 3 833156.26 825870.24 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Stress-NG Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Context Switching AMD / LFENCE Prior Default New Default - Generic Retpolines 1.3M 2.6M 3.9M 5.2M 6.5M SE +/- 395944.87, N = 12 SE +/- 27870.51, N = 3 5967964.27 3874572.19 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
Selenium OpenBenchmarking.org Runs / Minute, More Is Better Selenium Benchmark: StyleBench - Browser: Google Chrome AMD / LFENCE Prior Default New Default - Generic Retpolines 12 24 36 48 60 SE +/- 0.42, N = 3 SE +/- 0.13, N = 3 55.1 55.4 1. chrome 99.0.4844.51
Facebook RocksDB This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.22.1 Test: Random Fill Sync AMD / LFENCE Prior Default New Default - Generic Retpolines 4K 8K 12K 16K 20K SE +/- 40.83, N = 3 SE +/- 22.87, N = 3 19940 19837 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.22.1 Test: Random Fill AMD / LFENCE Prior Default New Default - Generic Retpolines 300K 600K 900K 1200K 1500K SE +/- 7351.70, N = 3 SE +/- 7169.40, N = 3 1385292 1383010 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.22.1 Test: Read Random Write Random AMD / LFENCE Prior Default New Default - Generic Retpolines 600K 1200K 1800K 2400K 3000K SE +/- 15263.55, N = 3 SE +/- 1897.32, N = 3 2951794 2923053 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
InfluxDB This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 AMD / LFENCE Prior Default New Default - Generic Retpolines 500K 1000K 1500K 2000K 2500K SE +/- 8821.41, N = 3 SE +/- 1911.27, N = 3 2135481.9 2139058.0
OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 AMD / LFENCE Prior Default New Default - Generic Retpolines 500K 1000K 1500K 2000K 2500K SE +/- 1104.63, N = 3 SE +/- 4793.84, N = 3 2288503.6 2287292.1
Selenium OpenBenchmarking.org Runs / Minute, More Is Better Selenium Benchmark: StyleBench - Browser: Firefox AMD / LFENCE Prior Default New Default - Generic Retpolines 30 60 90 120 150 SE +/- 0.67, N = 3 SE +/- 0.67, N = 3 121 121 1. firefox 97.0.2
OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: ARES-6 - Browser: Google Chrome AMD / LFENCE Prior Default New Default - Generic Retpolines 3 6 9 12 15 SE +/- 0.13, N = 4 SE +/- 0.05, N = 3 11.19 11.09 1. chrome 99.0.4844.51
Selenium OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Google Chrome AMD / LFENCE Prior Default New Default - Generic Retpolines 50 100 150 200 250 SE +/- 1.45, N = 3 SE +/- 0.88, N = 3 249 249 1. chrome 99.0.4844.51
Sockperf This is a network socket API performance benchmark developed by Mellanox. This test profile runs both the client and server on the local host for evaluating individual system performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Messages Per Second, More Is Better Sockperf 3.7 Test: Throughput AMD / LFENCE Prior Default New Default - Generic Retpolines 200K 400K 600K 800K 1000K SE +/- 4485.12, N = 25 SE +/- 5785.67, N = 5 838652 796686 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.29 Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory AMD / LFENCE Prior Default New Default - Generic Retpolines 70K 140K 210K 280K 350K SE +/- 1527.53, N = 3 326000 321000 1. (CC) gcc options: -rdynamic -ll -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
Selenium OpenBenchmarking.org Score, Fewer Is Better Selenium Benchmark: PSPDFKit WASM - Browser: Google Chrome AMD / LFENCE Prior Default New Default - Generic Retpolines 600 1200 1800 2400 3000 SE +/- 9.26, N = 3 SE +/- 11.61, N = 3 2838 2841 1. chrome 99.0.4844.51
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.29 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory AMD / LFENCE Prior Default New Default - Generic Retpolines 120K 240K 360K 480K 600K SE +/- 3929.94, N = 3 SE +/- 4255.72, N = 3 544333 518333 1. (CC) gcc options: -rdynamic -ll -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
Stress-NG Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Semaphores AMD / LFENCE Prior Default New Default - Generic Retpolines 700K 1400K 2100K 2800K 3500K SE +/- 806.72, N = 3 SE +/- 1006.44, N = 3 3281839.70 3280517.54 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
Selenium OpenBenchmarking.org Score, Fewer Is Better Selenium Benchmark: PSPDFKit WASM - Browser: Firefox AMD / LFENCE Prior Default New Default - Generic Retpolines 500 1000 1500 2000 2500 SE +/- 4.06, N = 3 2465 2471 1. firefox 97.0.2
PostMark This is a test of NetApp's PostMark benchmark designed to simulate small-file testing similar to the tasks endured by web and mail servers. This test profile will set PostMark to perform 25,000 transactions with 500 files simultaneously with the file sizes ranging between 5 and 512 kilobytes. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org TPS, More Is Better PostMark 1.51 Disk Transaction Performance AMD / LFENCE Prior Default New Default - Generic Retpolines 2K 4K 6K 8K 10K SE +/- 110.33, N = 3 SE +/- 102.67, N = 3 9149 8825 1. (CC) gcc options: -O3
Selenium OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: Kraken - Browser: Firefox AMD / LFENCE Prior Default New Default - Generic Retpolines 150 300 450 600 750 SE +/- 3.22, N = 3 SE +/- 5.51, N = 3 698.8 685.2 1. firefox 97.0.2
Facebook RocksDB This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.22.1 Test: Sequential Fill AMD / LFENCE Prior Default New Default - Generic Retpolines 300K 600K 900K 1200K 1500K SE +/- 18783.82, N = 3 SE +/- 1865.93, N = 3 1548865 1526340 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Selenium OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: Kraken - Browser: Google Chrome AMD / LFENCE Prior Default New Default - Generic Retpolines 110 220 330 440 550 SE +/- 2.96, N = 3 SE +/- 1.44, N = 3 498.9 504.4 1. chrome 99.0.4844.51
Darktable Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Darktable 3.0.1 Test: Boat - Acceleration: CPU-only AMD / LFENCE Prior Default New Default - Generic Retpolines 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 14.55 14.58
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Threads AMD / LFENCE Prior Default New Default - Generic Retpolines 3 6 9 12 15 SE +/- 0.10, N = 15 SE +/- 0.05, N = 3 10.04 10.68 1. (CC) gcc options: -lm
Selenium OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: WASM collisionDetection - Browser: Firefox AMD / LFENCE Prior Default New Default - Generic Retpolines 60 120 180 240 300 SE +/- 3.40, N = 3 SE +/- 0.83, N = 3 276.5 278.8 1. firefox 97.0.2
LevelDB LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Random Read AMD / LFENCE Prior Default New Default - Generic Retpolines 4 8 12 16 20 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 13.62 13.68 1. (CXX) g++ options: -O3 -lsnappy -lpthread
Sockperf This is a network socket API performance benchmark developed by Mellanox. This test profile runs both the client and server on the local host for evaluating individual system performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.7 Test: Latency Ping Pong AMD / LFENCE Prior Default New Default - Generic Retpolines 0.5654 1.1308 1.6962 2.2616 2.827 SE +/- 0.025, N = 5 SE +/- 0.021, N = 5 2.504 2.513 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
GIMP GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: unsharp-mask AMD / LFENCE Prior Default New Default - Generic Retpolines 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 12.14 12.25
Selenium OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: WASM collisionDetection - Browser: Google Chrome AMD / LFENCE Prior Default New Default - Generic Retpolines 60 120 180 240 300 SE +/- 0.84, N = 3 SE +/- 1.95, N = 3 250.81 252.12 1. chrome 99.0.4844.51
GIMP GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: auto-levels AMD / LFENCE Prior Default New Default - Generic Retpolines 3 6 9 12 15 SE +/- 0.010, N = 3 SE +/- 0.009, N = 3 9.509 9.586
GIMP GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: rotate AMD / LFENCE Prior Default New Default - Generic Retpolines 2 4 6 8 10 SE +/- 0.030, N = 3 SE +/- 0.005, N = 3 8.509 8.791
Selenium OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: WASM imageConvolute - Browser: Google Chrome AMD / LFENCE Prior Default New Default - Generic Retpolines 5 10 15 20 25 SE +/- 0.23, N = 4 SE +/- 0.15, N = 3 21.57 20.91 1. chrome 99.0.4844.51
OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: WASM imageConvolute - Browser: Firefox AMD / LFENCE Prior Default New Default - Generic Retpolines 5 10 15 20 25 SE +/- 0.03, N = 3 SE +/- 0.12, N = 3 21.1 20.6 1. firefox 97.0.2
Darktable Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Darktable 3.0.1 Test: Masskrug - Acceleration: CPU-only AMD / LFENCE Prior Default New Default - Generic Retpolines 2 4 6 8 10 SE +/- 0.015, N = 3 SE +/- 0.012, N = 3 6.454 6.475
OpenBenchmarking.org Seconds, Fewer Is Better Darktable 3.0.1 Test: Server Room - Acceleration: CPU-only AMD / LFENCE Prior Default New Default - Generic Retpolines 1.3219 2.6438 3.9657 5.2876 6.6095 SE +/- 0.011, N = 3 SE +/- 0.009, N = 3 5.847 5.875
GIMP GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.18 Test: resize AMD / LFENCE Prior Default New Default - Generic Retpolines 1.2373 2.4746 3.7119 4.9492 6.1865 SE +/- 0.031, N = 3 SE +/- 0.025, N = 3 5.485 5.499
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Files AMD / LFENCE Prior Default New Default - Generic Retpolines 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 11.42 11.72 1. (CC) gcc options: -lm
LevelDB LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Fill Sync AMD / LFENCE Prior Default New Default - Generic Retpolines 400 800 1200 1600 2000 SE +/- 4.66, N = 7 SE +/- 10.74, N = 4 1636.44 1648.57 1. (CXX) g++ options: -O3 -lsnappy -lpthread
Darktable Darktable is an open-source photography / workflow application this will use any system-installed Darktable program or on Windows will automatically download the pre-built binary from the project. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Darktable 3.0.1 Test: Server Rack - Acceleration: CPU-only AMD / LFENCE Prior Default New Default - Generic Retpolines 0.0412 0.0824 0.1236 0.1648 0.206 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 0.182 0.183
New Default - Generic Retpolines Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-yTrUTS/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0xa201016Java Notes: OpenJDK Runtime Environment (build 11.0.14+9-Ubuntu-0ubuntu2.20.04)Python Notes: Python 2.7.18 + Python 3.8.10Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 10 March 2022 08:31 by user phoronix.
AMD / LFENCE Prior Default Processor: AMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3904 BIOS), Chipset: AMD Starship/Matisse, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 Plus, Graphics: AMD Radeon RX 6800 16GB (2475/1000MHz), Audio: AMD Device ab28, Monitor: ASUS MG28U, Network: Realtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200
OS: Ubuntu 20.04, Kernel: 5.17.0-051700rc7daily20220310-generic (x86_64), Desktop: GNOME Shell 3.36.9, Display Server: X Server 1.20.13, OpenGL: 4.6 Mesa 22.0.0-devel (LLVM 13.0.1 DRM 3.44), Compiler: GCC 9.4.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-yTrUTS/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0xa201016Java Notes: OpenJDK Runtime Environment (build 11.0.14+9-Ubuntu-0ubuntu2.20.04)Python Notes: Python 2.7.18 + Python 3.8.10Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Vulnerable: LFENCE + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 10 March 2022 13:10 by user phoronix.