GCC, LLVM/Clang 3.0, AMD Open64

Compiler performance benchmarking of GCC 4.6.2 RC, LLVM 3.0, and latest Open64 from AMD. All phoronix test suite benchmarks were built with CFLAGS/CXXFLAGS of -march=native and -mtune=native under each compiler. Tests by Michael Larabel for a future Phoronix.com article.

HTML result view exported from: https://openbenchmarking.org/result/1110220-AR-OPENLLVM254.

ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen Resolution2 x AMD Opteron 2384Intel Core i7 GCC 4.6.2-RC LLVM/Clang 3.0-RC1 AMD Open64 4.2.5.2 GCC 4.6.2-RC LLVM/Clang 3.0-RC12 x AMD Opteron 2384 @ 2.70GHz (8 Cores)TYAN S2932/S2932-ENVIDIA MCP554096MB165GB Western Digital WD1600YS-01S + 128GB ityATI Device 9456ATI R6xx HDMIDELL S2409WUbuntu 11.103.0.0-12-generic (x86_64)Unity 4.22.0X Server 1.10.4radeon 6.14.992.1 Mesa 7.11 Gallium 0.4GCC 4.6.2ext41920x1080165GB Western Digital WD1600YS-01SClang 3.0 + LLVM 3.0Open64 4.2.5.2Intel Core i7-2630QM @ 2.00GHz (8 Cores)Compal PBL21Intel 2nd Generation Core Family8192MB128GB Super Talent STT_FTM28G225HNVIDIA GeForce GT 540M 2048MB (672/900MHz)Realtek ALC259Realtek RTL8111/8168B + Atheros AR9285 WirelessNVIDIA 280.134.1.0 NVIDIA 280.13GCC 4.6.2Clang 3.0 + LLVM 3.0OpenBenchmarking.orgSystem Details- Compiz was running on this system.

c-ray: Total Timesmallpt: Global Illumination Renderer; 100 Samplescompress-7zip: Compress Speed Testopenssl: RSA 4096-bit Performancejohn-the-ripper: Traditional DESjohn-the-ripper: MD5john-the-ripper: Blowfishmafft: Multiple Sequence Alignment2 x AMD Opteron 2384Intel Core i7 GCC 4.6.2-RC LLVM/Clang 3.0-RC1 AMD Open64 4.2.5.2 GCC 4.6.2-RC LLVM/Clang 3.0-RC163.011141610151.4819923331225380636.0067.0197451.1323626671139966038.2427.89161972354000713967234.7373.151431319167.5034686671109771225.4664.4762867.0837353331032863629.63OpenBenchmarking.org

C-Ray

Total Time

2 x AMD Opteron 2384Intel Core i7OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeGCC 4.6.2-RCLLVM/Clang 3.0-RC1AMD Open64 4.2.5.21632486480SE +/- 0.07, N = 3SE +/- 0.19, N = 3SE +/- 0.01, N = 3SE +/- 1.10, N = 3SE +/- 0.35, N = 363.0167.0127.8973.1564.47

Smallpt

Global Illumination Renderer; 100 Samples

2 x AMD Opteron 2384Intel Core i7OpenBenchmarking.orgSeconds, Fewer Is BetterSmallpt 1.0Global Illumination Renderer; 100 SamplesGCC 4.6.2-RCLLVM/Clang 3.0-RC12004006008001000SE +/- 0.00, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3114974143628

7-Zip Compression

Compress Speed Test

2 x AMD Opteron 2384Intel Core i7OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 9.13Compress Speed TestGCC 4.6.2-RCAMD Open64 4.2.5.23K6K9K12K15KSE +/- 138.92, N = 3SE +/- 41.60, N = 3SE +/- 213.53, N = 6161011619713191

OpenSSL

RSA 4096-bit Performance

2 x AMD Opteron 2384Intel Core i7OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.0.0aRSA 4096-bit PerformanceGCC 4.6.2-RCLLVM/Clang 3.0-RC11530456075SE +/- 0.03, N = 4SE +/- 0.03, N = 4SE +/- 0.00, N = 4SE +/- 0.16, N = 451.4851.1367.5067.08

John The Ripper

Test: Traditional DES

2 x AMD Opteron 2384Intel Core i7OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.7.8Test: Traditional DESGCC 4.6.2-RCLLVM/Clang 3.0-RC1AMD Open64 4.2.5.2800K1600K2400K3200K4000KSE +/- 3382.96, N = 3SE +/- 2333.33, N = 3SE +/- 4509.25, N = 3SE +/- 25562.56, N = 3SE +/- 2962.73, N = 319923332362667235400034686673735333

John The Ripper

Test: MD5

2 x AMD Opteron 2384Intel Core i7OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.7.8Test: MD5GCC 4.6.2-RCLLVM/Clang 3.0-RC1AMD Open64 4.2.5.23K6K9K12K15KSE +/- 3.33, N = 3SE +/- 11.68, N = 3SE +/- 0.33, N = 3SE +/- 6.69, N = 3SE +/- 1.20, N = 3122531139971391109710328

John The Ripper

Test: Blowfish

2 x AMD Opteron 2384Intel Core i7OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.7.8Test: BlowfishGCC 4.6.2-RCLLVM/Clang 3.0-RC1AMD Open64 4.2.5.22004006008001000SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 1.33, N = 3SE +/- 0.33, N = 3806660672712636

Timed MAFFT Alignment

Multiple Sequence Alignment

2 x AMD Opteron 2384Intel Core i7OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 6.706Multiple Sequence AlignmentGCC 4.6.2-RCLLVM/Clang 3.0-RC1AMD Open64 4.2.5.2918273645SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 336.0038.2434.7325.4629.63


Phoronix Test Suite v10.8.4