OpenCL ROCm Linux vs. AMDGPU-PRO Benchmarks

ROCm 1.4 benchmarks on Ubuntu 16.04 compared to AMDGPU-PRO. OpenCL benchmarks by Michael Larabel for a future article on Phoronix.com.

HTML result view exported from: https://openbenchmarking.org/result/1701170-RI-OPENCLROC11&grr.

ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLCompilerFile-SystemScreen ResolutionR9 FuryRX 460RX 480RX 460RX 480R9 Fury ROCm 1.4 ROCm 1.4 ROCm 1.4 AMDGPU-PRO 16.50 AMDGPU-PRO 16.50 AMDGPU-PRO 16.50Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores)MSI C236A WORKSTATION (MS-7998) v1.0Intel Sky Lake16384MB256GB TOSHIBA-RD400Sapphire AMD Radeon R9 FURY / NANO 3968MBRealtek ALC1150Acer B286HKIntel ConnectionUbuntu 16.044.6.0-kfd-compute-rocm-rel-1.4-16 (x86_64)Unity 7.4.0X Server 1.18.3modesetting 1.18.34.1 Mesa 11.2.0 Gallium 0.4OpenCL 2.0 AMD-APP (2300.5)GCC 5.4.0 20160609 + Clang 4.0 + LLVM 4.0.0ext43840x2160LLVMpipe3.3 Mesa 11.2.0 Gallium 0.4AMD Radeon RX 460 2048MB4.4.0-59-generic (x86_64)amdgpu 1.1.994.5.13462OpenCL 2.0 AMD-APP (2236.5)GCC 5.4.0 20160609AMD Radeon RX 480 8192MBSapphire AMD Radeon R9 Fury 4096MBOpenBenchmarking.orgCompiler Details- --build=x86_64-linux-gnu --disable-browser-plugin --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-arch-directory=amd64 --with-default-libstdcxx-abi=new --with-multilib-list=m32,m64,mx32 --with-tune=generic -v Processor Details- Scaling Governor: intel_pstate powersaveGraphics Details- R9 Fury: ROCm 1.4, RX 460: AMDGPU-PRO 16.50, RX 480: AMDGPU-PRO 16.50, R9 Fury: AMDGPU-PRO 16.50: GLAMOREnvironment Details- RX 460: ROCm 1.4, RX 480: ROCm 1.4: LIBGL_ALWAYS_SOFTWARE=1

luxmark: GPU - Luxball HDRluxmark: GPU - Microphoneluxmark: GPU - Hotelmandelgpu: GPUmandelbulbgpu: GPUjuliagpu: GPUdarktable: Server Room - OpenCLdarktable: Masskrug - OpenCLdarktable: Boat - OpenCLrodinia: OpenCL Heartwallrodinia: OpenCL Myocyteparboil: OpenCL TPACFparboil: OpenCL LBMparboil: OpenCL BFSshoc: OpenCL - Texture Read Bandwidthshoc: OpenCL - Bus Speed Readbackshoc: OpenCL - Bus Speed Downloadshoc: OpenCL - Max SP Flopsshoc: OpenCL - FFT SPshoc: OpenCL - TriadR9 FuryRX 460RX 480RX 460RX 480R9 Fury ROCm 1.4 ROCm 1.4 ROCm 1.4 AMDGPU-PRO 16.50 AMDGPU-PRO 16.50 AMDGPU-PRO 16.50119955695120182051996.2744388927.1273072755.801.486.094.986.45362.482.2436.811.43214.5310.8611.325330.67399.7110.59366438128295516.3329562658.9046101692.272.487.059.5713.51169.923.9061.841.4791.145.275.722158.12158.215.21919698759296261.8749050438.6770675082.100.995.935.727.28244.312.2438.491.49193.498.388.375815.52403.227.945547262389735552080.1532208376.9850807022.252.837.209.517.97131.5977.357.146.932066.69245.136.25140666924239981101281.9048517365.8081972594.400.995.764.375.3588.30160.5714.2013.665750.69508.209.401939476812402107202116.4043447360.4075992404.701.796.304.226.38223.2514.2113.697131.18751.864.12OpenBenchmarking.org

LuxMark

OpenCL Device: GPU - Scene: Luxball HDR

R9 FuryRX 460RX 480OpenBenchmarking.orgScore, More Is BetterLuxMark 3.0OpenCL Device: GPU - Scene: Luxball HDRROCm 1.4AMDGPU-PRO 16.504K8K12K16K20KSE +/- 17.34, N = 3SE +/- 75.47, N = 3SE +/- 17.00, N = 3SE +/- 9.82, N = 3SE +/- 0.67, N = 3SE +/- 68.10, N = 3119951939436645547919614066

LuxMark

OpenCL Device: GPU - Scene: Microphone

R9 FuryRX 460RX 480OpenBenchmarking.orgScore, More Is BetterLuxMark 3.0OpenCL Device: GPU - Scene: MicrophoneROCm 1.4AMDGPU-PRO 16.5016003200480064008000SE +/- 15.04, N = 3SE +/- 17.84, N = 3SE +/- 6.98, N = 3SE +/- 13.50, N = 35695768126236924

LuxMark

OpenCL Device: GPU - Scene: Hotel

R9 FuryRX 460RX 480OpenBenchmarking.orgScore, More Is BetterLuxMark 3.0OpenCL Device: GPU - Scene: HotelROCm 1.4AMDGPU-PRO 16.505001000150020002500SE +/- 0.00, N = 3SE +/- 11.46, N = 3SE +/- 0.58, N = 3SE +/- 1.00, N = 3SE +/- 2.40, N = 3SE +/- 6.94, N = 3120124023818979872399

MandelGPU

OpenCL Device: GPU

R9 FuryRX 460RX 480OpenBenchmarking.orgSamples/sec, More Is BetterMandelGPU 1.3pts1OpenCL Device: GPUROCm 1.4AMDGPU-PRO 16.5020M40M60M80M100MSE +/- 71744.88, N = 3SE +/- 30521.44, N = 3SE +/- 165178.15, N = 2SE +/- 126265.24, N = 382051996.27107202116.4028295516.3335552080.1559296261.8781101281.901. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

MandelbulbGPU

OpenCL Device: GPU

R9 FuryRX 460RX 480OpenBenchmarking.orgSamples/sec, More Is BetterMandelbulbGPU 1.0pts1OpenCL Device: GPUROCm 1.4AMDGPU-PRO 16.5011M22M33M44M55MSE +/- 2304744.64, N = 6SE +/- 117840.12, N = 3SE +/- 561923.72, N = 4SE +/- 81023.55, N = 344388927.1243447360.4029562658.9032208376.9849050438.6748517365.801. (CC) gcc options: -O3 -lm -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL

JuliaGPU

OpenCL Device: GPU

R9 FuryRX 460RX 480OpenBenchmarking.orgSamples/sec, More Is BetterJuliaGPU 1.2pts1OpenCL Device: GPUROCm 1.4AMDGPU-PRO 16.5020M40M60M80M100MSE +/- 985012.76, N = 3SE +/- 160084.77, N = 3SE +/- 97714.65, N = 2SE +/- 94849.94, N = 3SE +/- 500734.10, N = 273072755.8075992404.7046101692.2750807022.2570675082.1081972594.401. (CC) gcc options: -O3 -march=native -ftree-vectorize -funroll-loops -lglut -lOpenCL -lGL -lm

Darktable

Test: Server Room - Acceleration: OpenCL

R9 FuryRX 460RX 480OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.2.1Test: Server Room - Acceleration: OpenCLROCm 1.4AMDGPU-PRO 16.500.63681.27361.91042.54723.184SE +/- 0.02, N = 3SE +/- 0.04, N = 6SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 4SE +/- 0.01, N = 31.481.792.482.830.990.99

Darktable

Test: Masskrug - Acceleration: OpenCL

R9 FuryRX 460RX 480OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.2.1Test: Masskrug - Acceleration: OpenCLROCm 1.4AMDGPU-PRO 16.50246810SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 36.096.307.057.205.935.76

Darktable

Test: Boat - Acceleration: OpenCL

R9 FuryRX 460RX 480OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.2.1Test: Boat - Acceleration: OpenCLROCm 1.4AMDGPU-PRO 16.503691215SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.77, N = 6SE +/- 0.02, N = 3SE +/- 0.01, N = 34.984.229.579.515.724.37

Rodinia

Test: OpenCL Heartwall

R9 FuryRX 460RX 480OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenCL HeartwallROCm 1.4AMDGPU-PRO 16.503691215SE +/- 0.16, N = 6SE +/- 0.07, N = 3SE +/- 0.08, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 36.456.3813.517.977.285.351. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenCL Myocyte

R9 FuryRX 460RX 480OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenCL MyocyteROCm 1.4AMDGPU-PRO 16.5080160240320400SE +/- 0.09, N = 3SE +/- 51.83, N = 6SE +/- 0.16, N = 2SE +/- 1.77, N = 3SE +/- 26.88, N = 6362.48169.92131.59244.3188.301. (CXX) g++ options: -O2 -lOpenCL

Parboil

Test: OpenCL TPACF

R9 FuryRX 460RX 480OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenCL TPACFROCm 1.40.87751.7552.63253.514.3875SE +/- 0.05, N = 6SE +/- 0.00, N = 3SE +/- 0.02, N = 32.243.902.241. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: OpenCL LBM

R9 FuryRX 460RX 480OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenCL LBMROCm 1.41428425670SE +/- 2.40, N = 6SE +/- 0.13, N = 3SE +/- 0.21, N = 336.8161.8438.491. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: OpenCL BFS

R9 FuryRX 460RX 480OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenCL BFSROCm 1.40.33530.67061.00591.34121.6765SE +/- 0.02, N = 5SE +/- 0.01, N = 3SE +/- 0.02, N = 31.431.471.491. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Texture Read Bandwidth

R9 FuryRX 460RX 480OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: Texture Read BandwidthROCm 1.4AMDGPU-PRO 16.5050100150200250SE +/- 4.26, N = 3SE +/- 1.03, N = 3SE +/- 0.16, N = 3SE +/- 0.69, N = 3SE +/- 1.30, N = 3SE +/- 0.37, N = 3214.53223.2591.1477.35193.49160.571. (CXX) g++ options: -O2 -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Bus Speed Readback

R9 FuryRX 460RX 480OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: Bus Speed ReadbackROCm 1.4AMDGPU-PRO 16.5048121620SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 310.8614.215.277.148.3814.201. (CXX) g++ options: -O2 -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Bus Speed Download

R9 FuryRX 460RX 480OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: Bus Speed DownloadROCm 1.4AMDGPU-PRO 16.5048121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 311.3213.695.726.938.3713.661. (CXX) g++ options: -O2 -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Max SP Flops

R9 FuryRX 460RX 480OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: Max SP FlopsROCm 1.4AMDGPU-PRO 16.5015003000450060007500SE +/- 369.63, N = 6SE +/- 0.69, N = 3SE +/- 0.07, N = 3SE +/- 18.41, N = 3SE +/- 4.14, N = 3SE +/- 30.75, N = 35330.677131.182158.122066.695815.525750.691. (CXX) g++ options: -O2 -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: FFT SP

R9 FuryRX 460RX 480OpenBenchmarking.orgGFLOPS, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: FFT SPROCm 1.4AMDGPU-PRO 16.50160320480640800SE +/- 0.44, N = 3SE +/- 14.35, N = 3SE +/- 0.04, N = 3SE +/- 1.23, N = 3SE +/- 6.05, N = 3SE +/- 2.19, N = 3399.71751.86158.21245.13403.22508.201. (CXX) g++ options: -O2 -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt

SHOC Scalable HeterOgeneous Computing

Target: OpenCL - Benchmark: Triad

R9 FuryRX 460RX 480OpenBenchmarking.orgGB/s, More Is BetterSHOC Scalable HeterOgeneous Computing 2015-11-10Target: OpenCL - Benchmark: TriadROCm 1.4AMDGPU-PRO 16.503691215SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.10, N = 3SE +/- 0.01, N = 3SE +/- 0.14, N = 410.594.125.216.257.949.401. (CXX) g++ options: -O2 -lSHOCCommonOpenCL -lSHOCCommon -lOpenCL -lrt

Phoronix Test Suite v10.8.4