OpenCL Radeon Open-Source vs. Catalyst

Radeon R600g and RadeonSI Gallium3D Clover OpenCL benchmarks with clpeak using Linux 4.5 + Mesa 11.2-devel. Compared to Catalyst. Benchmark results by Michael Larabel of Phoronix.com for a future article.

HTML result view exported from: https://openbenchmarking.org/result/1602034-GA-1602026GA07.

ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionOpen-SourceCatalyst HD 6870 HD 7950 R9 285 R9 290 R7 370 HD 6870 HD 7950 R9 285 R9 290 R7 370 R9 290xIntel Core i7-5960X @ 3.50GHz (16 Cores)Gigabyte X99-UD4-CFIntel Xeon E7 v3/Xeon16384MB2 x 120GB SanDisk SDSSDA12 + 120GB Samsung SSD 850Sapphire AMD Radeon HD 6870 1024MBRealtek ALC1150DELL P2415QIntel ConnectionUbuntu 15.104.5.0-rc2amdgpu-pp+ (x86_64) 20160201UnityX Server 1.17.2radeon 7.6.993.3 Mesa 11.2.0-devel (padoka PPA) Gallium 0.4GCC 5.2.1 20151010ext43840x2160XFX AMD Radeon HD 7950/8950 OEM / R9 280 3072MB4.1 Mesa 11.2.0-devel (padoka PPA) Gallium 0.4XFX AMD Radeon R9 285/380 2048MBamdgpu 1.0.99XFX AMD Radeon R9 290 4096MBradeon 7.6.99MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MBSapphire AMD Radeon HD 6800 1024MB4.2.0-27-generic (x86_64)fglrx 15.20.34.5.13399XFX AMD Radeon HD 7900 3072MBSupported device 6939 2048MBSupported device 67B1 4096MBMSI AMD Radeon R7 370 4096MBIntel Core i7-4790K @ 4.50GHz (8 Cores)ASUS SABERTOOTH Z97 MARK 1Intel 4th Gen Core DRAM512GB OCZ VERTEX4 + 256GB CORSAIR FORCE LXSupported device 67B0 4096MB (1030/1250MHz)AMD Hawaii HDMI AudioASUS PB278Realtek RTL8111/8168/8411Mageia 54.1.15-desktop-2.mga5 (x86_64)KDE 4.14.5X Server 1.16.4fglrx 15.30.34.4.13416GCC 4.9.2 + Open64 opencc: option-- d Open Chinese Convert (OpenCC) Command Line Tool 0.4.3Author: BYVoid Bug Report: github /BYVoid/OpenCC/issuesUsage: opencc [Options]Options:-i [file]--input=[file] Read original text from [file].-o [file]--output=[file] Write converted text to [file].-c [file]--config=[file] Load configuration of con from [file].-v-- Print and build information.-h--help Print this help.With no input file reads standard input and writes converted stream to standard output.Default configuration(zhs2zht.ini) will be loaded if not set. + Clang 3.5.2 + LLVM 3.5.22560x1440OpenBenchmarking.orgKernel Details- Open-Source: HD 6870, Open-Source: HD 7950, Open-Source: R9 285, Open-Source: R9 290, Open-Source: R7 370: amdgpu.powerplay=1Compiler Details- Open-Source: HD 6870, Open-Source: HD 7950, Open-Source: R9 285, Open-Source: R9 290, Open-Source: R7 370, Catalyst: HD 6870, Catalyst: HD 7950, Catalyst: R9 285, Catalyst: R9 290, Catalyst: R7 370: --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 -vProcessor Details- Scaling Governor: intel_pstate performanceGraphics Details- Open-Source: HD 6870: EXA- Open-Source: HD 7950: GLAMOR- Open-Source: R9 285: GLAMOR- Open-Source: R9 290: GLAMOR- Open-Source: R7 370: GLAMOREnvironment Details- Catalyst: HD 6870: LIBGL_DRIVERS_PATH=/usr/lib/fglrx/dri:/usr/lib/x86_64-linux-gnu/dri:/usr/lib/dri:/usr/lib32/fglrx/dri:/usr/lib/i386-linux-gnu/dri- Catalyst: HD 7950: LIBGL_DRIVERS_PATH=/usr/lib/fglrx/dri:/usr/lib/x86_64-linux-gnu/dri:/usr/lib/dri:/usr/lib32/fglrx/dri:/usr/lib/i386-linux-gnu/dri- Catalyst: R9 285: LIBGL_DRIVERS_PATH=/usr/lib/fglrx/dri:/usr/lib/x86_64-linux-gnu/dri:/usr/lib/dri:/usr/lib32/fglrx/dri:/usr/lib/i386-linux-gnu/dri- Catalyst: R9 290: LIBGL_DRIVERS_PATH=/usr/lib/fglrx/dri:/usr/lib/x86_64-linux-gnu/dri:/usr/lib/dri:/usr/lib32/fglrx/dri:/usr/lib/i386-linux-gnu/dri- Catalyst: R7 370: LIBGL_DRIVERS_PATH=/usr/lib/fglrx/dri:/usr/lib/x86_64-linux-gnu/dri:/usr/lib/dri:/usr/lib32/fglrx/dri:/usr/lib/i386-linux-gnu/dri- Catalyst: R9 290x: LIBGL_DRIVERS_PATH=/usr/lib/dri:/usr/lib64/dri

clpeak: Kernel Latencyclpeak: Integer Compute INTclpeak: Single-Precision Floatclpeak: Double-Precision Doubleclpeak: Transfer Bandwidth enqueueReadBufferclpeak: Transfer Bandwidth enqueueWriteBufferOpen-SourceCatalyst HD 6870 HD 7950 R9 285 R9 290 R7 370 HD 6870 HD 7950 R9 285 R9 290 R7 370 R9 290x145.85182.44360.923.585.26105.54491.622345.12547.0310.4015.72143.39650.803102.60204.7510.0914.82118.711578.014608.55596.559.8212.32110.24418.742031.91131.4910.4115.7724.90199.38398.298.809.4934.03516.742292.54536.8410.4713.8826.78647.083177.62202.719.0013.2137.771591.744721.18598.8110.6414.8135.42415.672095.91128.7310.6714.9717.961871.325697.10716.1712.2226.42OpenBenchmarking.org

clpeak

OpenCL Test: Kernel Latency

Open-SourceCatalystOpenBenchmarking.orgus, Fewer Is BetterclpeakOpenCL Test: Kernel LatencyHD 6870HD 7950R9 285R9 290R7 370R9 290x306090120150SE +/- 8.80, N = 6SE +/- 1.78, N = 6SE +/- 14.32, N = 6SE +/- 2.08, N = 6SE +/- 3.85, N = 6SE +/- 0.28, N = 3SE +/- 0.68, N = 3SE +/- 0.38, N = 3SE +/- 2.12, N = 6SE +/- 0.52, N = 5SE +/- 0.42, N = 6145.85105.54143.39118.71110.2424.9034.0326.7837.7735.4217.96

clpeak

OpenCL Test: Integer Compute INT

Open-SourceCatalystOpenBenchmarking.orgGIOPS, More Is BetterclpeakOpenCL Test: Integer Compute INTHD 6870HD 7950R9 285R9 290R7 370R9 290x400800120016002000SE +/- 0.16, N = 3SE +/- 7.09, N = 6SE +/- 0.05, N = 3SE +/- 0.49, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 21.03, N = 6SE +/- 1.77, N = 3SE +/- 0.05, N = 3SE +/- 3.65, N = 3SE +/- 1.20, N = 3182.44491.62650.801578.01418.74199.38516.74647.081591.74415.671871.32

clpeak

OpenCL Test: Single-Precision Float

Open-SourceCatalystOpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Single-Precision FloatHD 6870HD 7950R9 285R9 290R7 370R9 290x12002400360048006000SE +/- 0.08, N = 3SE +/- 47.92, N = 6SE +/- 1.52, N = 3SE +/- 2.13, N = 3SE +/- 0.22, N = 3SE +/- 0.00, N = 3SE +/- 44.66, N = 6SE +/- 0.24, N = 3SE +/- 0.95, N = 3SE +/- 2.25, N = 3SE +/- 5.48, N = 3360.922345.123102.604608.552031.91398.292292.543177.624721.182095.915697.10

clpeak

OpenCL Test: Double-Precision Double

Open-SourceCatalystOpenBenchmarking.orgGFLOPS, More Is BetterclpeakOpenCL Test: Double-Precision DoubleHD 7950R9 285R9 290R7 370R9 290x150300450600750SE +/- 16.28, N = 6SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.00, N = 3SE +/- 11.37, N = 6SE +/- 0.22, N = 3SE +/- 0.01, N = 3SE +/- 0.36, N = 3SE +/- 0.37, N = 3547.03204.75596.55131.49536.84202.71598.81128.73716.17

clpeak

OpenCL Test: Transfer Bandwidth enqueueReadBuffer

Open-SourceCatalystOpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueReadBufferHD 6870HD 7950R9 285R9 290R7 370R9 290x3691215SE +/- 0.10, N = 6SE +/- 0.03, N = 3SE +/- 0.30, N = 6SE +/- 0.25, N = 6SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.13, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 33.5810.4010.099.8210.418.8010.479.0010.6410.6712.22

clpeak

OpenCL Test: Transfer Bandwidth enqueueWriteBuffer

Open-SourceCatalystOpenBenchmarking.orgGBPS, More Is BetterclpeakOpenCL Test: Transfer Bandwidth enqueueWriteBufferHD 6870HD 7950R9 285R9 290R7 370R9 290x612182430SE +/- 0.00, N = 3SE +/- 0.05, N = 3SE +/- 0.59, N = 6SE +/- 0.06, N = 3SE +/- 0.00, N = 3SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.19, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 35.2615.7214.8212.3215.779.4913.8813.2114.8114.9726.42


Phoronix Test Suite v10.8.4