Compiler Optimizations For AMD Steamroller/Kaveri

Compiler benchmarks by Michael Larabel for a future article on Phoronix.com

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1408161-KH-COMPILERO99
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts

Limit displaying results to tests within:

Audio Encoding 2 Tests
Timed Code Compilation 2 Tests
C/C++ Compiler Tests 7 Tests
CPU Massive 7 Tests
Creator Workloads 5 Tests
Encoding 3 Tests
Multi-Core 5 Tests
Programmer / Developer System Benchmarks 2 Tests
Server CPU Tests 3 Tests
Single-Threaded 2 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
k8
August 13 2014
 
k8-sse3
August 13 2014
 
barcelona
August 13 2014
 
bdver1
August 13 2014
 
bdver2
August 13 2014
 
bdver3
August 13 2014
 
bdver3 + flto
August 13 2014
 
Invert Hiding All Results Option
 

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

k8

Compiler Notes: --disable-multilib --enable-checking=release --enable-languages=c,c++
Processor Notes: Scaling Governor: acpi-cpufreq ondemand

k8-sse3

barcelona

bdver1

bdver2

bdver3

bdver3 + flto

Processor: AMD A10-7800 Radeon R7 12 Compute Cores 4C+8G @ 3.50GHz (4 Cores), Motherboard: ASUS A88X-PRO, Chipset: AMD Family 15h, Memory: 7168MB, Disk: 64GB OCZ AGILITY, Graphics: ASUS AMD Radeon R7 200 1024MB, Audio: AMD Kaveri HDMI/DP, Monitor: SyncMaster, Network: Realtek RTL8111/8168/8411

OS: Ubuntu 14.04, Kernel: 3.16.0-031600-generic (x86_64), Desktop: Unity 7.2.2, Display Server: X Server 1.15.1, Display Driver: radeon 7.4.99, OpenGL: 3.3 Mesa 10.3.0-devel (git-b1843a2 2014-08-10 trusty-oibaf-ppa+gallium-nine) Gallium 0.4, Compiler: GCC 4.10.0 20140810, File-System: ext4, Screen Resolution: 2560x1600

Compiler Optimizations For AMD Steamroller/KaveriOpenBenchmarking.orgPhoronix Test SuiteAMD A10-7800 Radeon R7 12 Compute Cores 4C+8G @ 3.50GHz (4 Cores)ASUS A88X-PROAMD Family 15h7168MB64GB OCZ AGILITYASUS AMD Radeon R7 200 1024MBAMD Kaveri HDMI/DPSyncMasterRealtek RTL8111/8168/8411Ubuntu 14.043.16.0-031600-generic (x86_64)Unity 7.2.2X Server 1.15.1radeon 7.4.993.3 Mesa 10.3.0-devel (git-b1843a2 2014-08-10 trusty-oibaf-ppa+gallium-nine) Gallium 0.4GCC 4.10.0 20140810ext42560x1600ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionCompiler Optimizations For AMD Steamroller/Kaveri BenchmarksSystem Logs- --disable-multilib --enable-checking=release --enable-languages=c,c++- Scaling Governor: acpi-cpufreq ondemand

k8k8-sse3barcelonabdver1bdver2bdver3bdver3 + fltoLogarithmic Result OverviewPhoronix Test SuiteTimed PHP CompilationC-RayGraphicsMagickTimed Apache CompilationFLAC Audio EncodingGraphicsMagickGraphicsMagickLAME MP3 EncodingGraphicsMagickGraphicsMagickFFmpegHimeno BenchmarkTime To CompileTotal TimeSharpenTime To CompileWAV To FLACBlurResizingWAV To MP3L.A.THWB Color SpaceH.2.H.T.N.DP.P.S

Compiler Optimizations For AMD Steamroller/Kaveribuild-php: Time To Compilec-ray: Total Timegraphics-magick: Sharpenbuild-apache: Time To Compileencode-flac: WAV To FLACgraphics-magick: Blurgraphics-magick: Resizingencode-mp3: WAV To MP3graphics-magick: Local Adaptive Thresholdinggraphics-magick: HWB Color Spaceffmpeg: H.264 HD To NTSC DVhimeno: Poisson Pressure Solverk8k8-sse3barcelonabdver1bdver2bdver3bdver3 + flto58.0996.894261.066.747710220.857712921.73870.3358.1296.934261.076.749210220.617612921.78869.5358.1565.575861.077.229111922.287713721.77871.1760.5547.436861.295.2810012419.638113721.95883.0660.5747.326761.715.5210012419.608113721.69872.2220.2347.316761.405.5910012419.618113721.60882.11184.1547.416796.695.3710012419.678213721.65880.95OpenBenchmarking.org

Timed PHP Compilation

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 5.2.9Time To Compilek8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona4080120160200SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.86, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 358.1258.09184.1520.2360.5760.5558.15-march=k8-sse3-march=k8-march=bdver3 -flto-march=bdver3-march=bdver2-march=bdver1-march=barcelona1. (CC) gcc options: -O3 -pedantic -ldl -lz -lm

C-Ray

This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Timek8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona20406080100SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.09, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.06, N = 396.9396.8947.4147.3147.3247.4365.57-march=k8-sse3-march=k8-march=bdver3 -flto-march=bdver3-march=bdver2-march=bdver1-march=barcelona1. (CC) gcc options: -lm -lpthread -O3

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests to stress the system's CPU. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: Sharpenk8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona1530456075SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 342426767676858-march=k8-sse3-march=k8-march=bdver3 -flto-march=bdver3-march=bdver2-march=bdver1-march=barcelona1. (CC) gcc options: -std=gnu99 -fopenmp -O3 -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lxml2 -lz -lm -lpthread

Timed Apache Compilation

This test times how long it takes to build the Apache HTTP Server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.7Time To Compilek8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona20406080100SE +/- 0.14, N = 3SE +/- 0.28, N = 3SE +/- 0.10, N = 3SE +/- 0.10, N = 3SE +/- 0.27, N = 3SE +/- 0.04, N = 3SE +/- 0.15, N = 361.0761.0696.6961.4061.7161.2961.07

FLAC Audio Encoding

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.0WAV To FLACk8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona246810SE +/- 0.01, N = 5SE +/- 0.01, N = 5SE +/- 0.01, N = 5SE +/- 0.01, N = 5SE +/- 0.01, N = 5SE +/- 0.01, N = 5SE +/- 0.01, N = 56.746.745.375.595.525.287.22-march=k8-sse3-march=k8-march=bdver3 -flto-march=bdver3-march=bdver2-march=bdver1-march=barcelona1. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests to stress the system's CPU. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: Blurk8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona20406080100SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3927710010010010091-march=k8-sse3-march=k8-march=bdver3 -flto-march=bdver3-march=bdver2-march=bdver1-march=barcelona1. (CC) gcc options: -std=gnu99 -fopenmp -O3 -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lxml2 -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: Resizingk8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona306090120150SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3102102124124124124119-march=k8-sse3-march=k8-march=bdver3 -flto-march=bdver3-march=bdver2-march=bdver1-march=barcelona1. (CC) gcc options: -std=gnu99 -fopenmp -O3 -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lxml2 -lz -lm -lpthread

LAME MP3 Encoding

LAME is an MP3 encoder licensed under the LGPL. This test measures the time required to encode a WAV file to MP3 format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.99.3WAV To MP3k8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona510152025SE +/- 0.02, N = 5SE +/- 0.27, N = 5SE +/- 0.02, N = 5SE +/- 0.02, N = 5SE +/- 0.01, N = 5SE +/- 0.04, N = 5SE +/- 0.02, N = 520.6120.8519.6719.6119.6019.6322.28-march=k8-sse3-march=k8-march=bdver3 -flto-march=bdver3-march=bdver2-march=bdver1-march=barcelona1. (CC) gcc options: -pipe -O3 -lncurses -lm

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests to stress the system's CPU. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: Local Adaptive Thresholdingk8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona20406080100SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 376778281818177-march=k8-sse3-march=k8-march=bdver3 -flto-march=bdver3-march=bdver2-march=bdver1-march=barcelona1. (CC) gcc options: -std=gnu99 -fopenmp -O3 -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lxml2 -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: HWB Color Spacek8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona306090120150SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3129129137137137137137-march=k8-sse3-march=k8-march=bdver3 -flto-march=bdver3-march=bdver2-march=bdver1-march=barcelona1. (CC) gcc options: -std=gnu99 -fopenmp -O3 -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lxml2 -lz -lm -lpthread

FFmpeg

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 2.1.1H.264 HD To NTSC DVk8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona510152025SE +/- 0.12, N = 3SE +/- 0.05, N = 3SE +/- 0.15, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 321.7821.7321.6521.6021.6921.9521.77-march=k8-sse3-march=k8-march=bdver3 -flto-march=native-march=bdver2-march=bdver1-march=barcelona1. (CC) gcc options: -lavdevice -lavfilter -lavformat -lavcodec -lswresample -lswscale -lavutil -ldl -lXv -lX11 -lXext -lasound -lSDL -lm -pthread -O3 -std=c99 -fomit-frame-pointer -fno-math-errno -fno-signed-zeros -fno-tree-vectorize -MMD -MF -MT

Himeno Benchmark

The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure Solverk8-sse3k8bdver3 + fltobdver3bdver2bdver1barcelona2004006008001000SE +/- 0.82, N = 3SE +/- 0.27, N = 3SE +/- 0.76, N = 3SE +/- 2.02, N = 3SE +/- 3.58, N = 3SE +/- 1.93, N = 3SE +/- 1.09, N = 3869.53870.33880.95882.11872.22883.06871.17-march=k8-sse3-march=k8-march=bdver3 -flto-march=bdver3-march=bdver2-march=bdver1-march=barcelona1. (CC) gcc options: -O3

k8

Compiler Notes: --disable-multilib --enable-checking=release --enable-languages=c,c++
Processor Notes: Scaling Governor: acpi-cpufreq ondemand

Testing initiated at 13 August 2014 18:49 by user phoronix.

k8-sse3

Compiler Notes: --disable-multilib --enable-checking=release --enable-languages=c,c++
Processor Notes: Scaling Governor: acpi-cpufreq ondemand

Testing initiated at 13 August 2014 18:03 by user phoronix.

barcelona

Compiler Notes: --disable-multilib --enable-checking=release --enable-languages=c,c++
Processor Notes: Scaling Governor: acpi-cpufreq ondemand

Testing initiated at 13 August 2014 17:15 by user phoronix.

bdver1

Compiler Notes: --disable-multilib --enable-checking=release --enable-languages=c,c++
Processor Notes: Scaling Governor: acpi-cpufreq ondemand

Testing initiated at 13 August 2014 16:29 by user phoronix.

bdver2

Compiler Notes: --disable-multilib --enable-checking=release --enable-languages=c,c++
Processor Notes: Scaling Governor: acpi-cpufreq ondemand

Testing initiated at 13 August 2014 15:36 by user phoronix.

bdver3

Compiler Notes: --disable-multilib --enable-checking=release --enable-languages=c,c++
Processor Notes: Scaling Governor: acpi-cpufreq ondemand

Testing initiated at 13 August 2014 14:44 by user phoronix.

bdver3 + flto

Processor: AMD A10-7800 Radeon R7 12 Compute Cores 4C+8G @ 3.50GHz (4 Cores), Motherboard: ASUS A88X-PRO, Chipset: AMD Family 15h, Memory: 7168MB, Disk: 64GB OCZ AGILITY, Graphics: ASUS AMD Radeon R7 200 1024MB, Audio: AMD Kaveri HDMI/DP, Monitor: SyncMaster, Network: Realtek RTL8111/8168/8411

OS: Ubuntu 14.04, Kernel: 3.16.0-031600-generic (x86_64), Desktop: Unity 7.2.2, Display Server: X Server 1.15.1, Display Driver: radeon 7.4.99, OpenGL: 3.3 Mesa 10.3.0-devel (git-b1843a2 2014-08-10 trusty-oibaf-ppa+gallium-nine) Gallium 0.4, Compiler: GCC 4.10.0 20140810, File-System: ext4, Screen Resolution: 2560x1600

Compiler Notes: --disable-multilib --enable-checking=release --enable-languages=c,c++
Processor Notes: Scaling Governor: acpi-cpufreq ondemand

Testing initiated at 13 August 2014 19:40 by user phoronix.