LLVM Clang 3.3 vs. GCC 4.8 - Intel Core-AVX2 Haswell Intel Core i7-4770K testing with a Intel DH87RL motherboard looking at the GCC 4.7, GCC 4.8, LLVM Clang 3.2, and LLVM Clang 3.3 compiler performance with core-avx2 Haswell optimizations. Intel Core i7 Haswell core-avx2 compiler benchmarks for a future article on Phoronix by Michael Larabel. LLVM Clang 3.3: Processor: Intel Core i7-4770K @ 3.50GHz (8 Cores), Motherboard: Intel DH87RL, Chipset: Intel 4th Gen Core DRAM, Memory: 15360MB, Disk: 240GB OCZ VERTEX3, Graphics: Intel Haswell Desktop, Audio: Intel Haswell HDMI, Monitor: VA2431, Network: Intel Connection I217-V OS: Ubuntu 13.10, Kernel: 3.10.0-999-generic (x86_64), Desktop: Unity 7.0.0, Display Server: X Server 1.13.3, Display Driver: intel 2.21.9, OpenGL: 3.0 Mesa 9.1.3, Compiler: Clang 3.3 + LLVM 3.3, File-System: ext4, Screen Resolution: 1920x1080 GCC 4.8.1: Processor: Intel Core i7-4770K @ 3.50GHz (8 Cores), Motherboard: Intel DH87RL, Chipset: Intel 4th Gen Core DRAM, Memory: 15360MB, Disk: 240GB OCZ VERTEX3, Graphics: Intel Haswell Desktop, Audio: Intel Haswell HDMI, Monitor: VA2431, Network: Intel Connection I217-V OS: Ubuntu 13.10, Kernel: 3.10.0-999-generic (x86_64), Desktop: Unity 7.0.0, Display Server: X Server 1.13.3, Display Driver: intel 2.21.9, OpenGL: 3.0 Mesa 9.1.3, Compiler: GCC 4.8.1, File-System: ext4, Screen Resolution: 1920x1080 LLVM Clang 3.2: Processor: Intel Core i7-4770K @ 3.50GHz (8 Cores), Motherboard: Intel DH87RL, Chipset: Intel 4th Gen Core DRAM, Memory: 15360MB, Disk: 240GB OCZ VERTEX3, Graphics: Intel Haswell Desktop, Audio: Intel Haswell HDMI, Monitor: VA2431, Network: Intel Connection I217-V OS: Ubuntu 13.10, Kernel: 3.10.0-999-generic (x86_64), Desktop: Unity 7.0.0, Display Server: X Server 1.13.3, Display Driver: intel 2.21.9, OpenGL: 3.0 Mesa 9.1.3, Compiler: Clang 3.2 + LLVM 3.2svn, File-System: ext4, Screen Resolution: 1920x1080 GCC 4.7.3: Processor: Intel Core i7-4770K @ 3.50GHz (8 Cores), Motherboard: Intel DH87RL, Chipset: Intel 4th Gen Core DRAM, Memory: 15360MB, Disk: 240GB OCZ VERTEX3, Graphics: Intel Haswell Desktop, Audio: Intel Haswell HDMI, Monitor: VA2431, Network: Intel Connection I217-V OS: Ubuntu 13.10, Kernel: 3.10.0-999-generic (x86_64), Desktop: Unity 7.0.0, Display Server: X Server 1.13.3, Display Driver: intel 2.21.9, OpenGL: 3.0 Mesa 9.1.3, Compiler: GCC 4.7.3, File-System: ext4, Screen Resolution: 1920x1080 Smallpt 1.0 Global Illumination Renderer; 100 Samples Seconds < Lower Is Better LLVM Clang 3.3 . 142 |========================================================= GCC 4.8.1 ...... 25 |========== LLVM Clang 3.2 . 140 |======================================================== GCC 4.7.3 ...... 25 |========== Primesieve 4.2 1e12 Prime Number Generation Seconds < Lower Is Better LLVM Clang 3.3 . 326.85 |====================================================== GCC 4.8.1 ...... 79.24 |============= GCC 4.7.3 ...... 79.36 |============= Timed ImageMagick Compilation 6.8.1-10 Time To Compile Seconds < Lower Is Better LLVM Clang 3.3 . 34.35 |==================== GCC 4.8.1 ...... 78.61 |============================================== LLVM Clang 3.2 . 31.94 |=================== GCC 4.7.3 ...... 93.67 |======================================================= Timed PHP Compilation 5.2.9 Time To Compile Seconds < Lower Is Better LLVM Clang 3.3 . 21.03 |=================================== GCC 4.8.1 ...... 33.30 |======================================================= LLVM Clang 3.2 . 19.59 |================================ GCC 4.7.3 ...... 32.92 |====================================================== C-Ray 1.1 Total Time Seconds < Lower Is Better LLVM Clang 3.3 . 27.03 |====================================================== GCC 4.8.1 ...... 17.02 |================================== LLVM Clang 3.2 . 27.46 |======================================================= GCC 4.7.3 ...... 21.45 |=========================================== SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better LLVM Clang 3.3 . 1666.24 |===================================================== GCC 4.8.1 ...... 1164.63 |===================================== LLVM Clang 3.2 . 1670.77 |===================================================== GCC 4.7.3 ...... 1163.52 |===================================== BLAKE2 20121223 Phoronix Test Suite v4.8.0m0 Cycles Per Byte < Lower Is Better LLVM Clang 3.3 . 7.45 |======================================================= GCC 4.8.1 ...... 5.30 |======================================= LLVM Clang 3.2 . 7.54 |======================================================== GCC 4.7.3 ...... 5.32 |======================================== SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better LLVM Clang 3.3 . 619.77 |====================================================== GCC 4.8.1 ...... 615.32 |====================================================== LLVM Clang 3.2 . 615.04 |====================================================== GCC 4.7.3 ...... 450.21 |======================================= Timed MAFFT Alignment 6.864 Multiple Sequence Alignment Seconds < Lower Is Better LLVM Clang 3.3 . 6.19 |======================================================== GCC 4.8.1 ...... 5.17 |=============================================== LLVM Clang 3.2 . 5.96 |====================================================== GCC 4.7.3 ...... 5.27 |================================================ Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better LLVM Clang 3.3 . 1419.90 |============================================= GCC 4.8.1 ...... 1593.09 |=================================================== LLVM Clang 3.2 . 1532.51 |================================================= GCC 4.7.3 ...... 1663.97 |===================================================== Timed HMMer Search 2.3.2 Pfam Database Search Seconds < Lower Is Better LLVM Clang 3.3 . 10.94 |================================================== GCC 4.8.1 ...... 10.30 |================================================ LLVM Clang 3.2 . 11.92 |======================================================= GCC 4.7.3 ...... 10.41 |================================================ LAME MP3 Encoding 3.99.3 WAV To MP3 Seconds < Lower Is Better LLVM Clang 3.3 . 14.45 |======================================================= LLVM Clang 3.2 . 12.74 |================================================ SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better LLVM Clang 3.3 . 1263.29 |===================================================== GCC 4.8.1 ...... 1123.80 |=============================================== LLVM Clang 3.2 . 1234.19 |==================================================== GCC 4.7.3 ...... 1177.86 |================================================= SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better LLVM Clang 3.3 . 237.86 |=================================================== GCC 4.8.1 ...... 245.00 |===================================================== LLVM Clang 3.2 . 246.79 |===================================================== GCC 4.7.3 ...... 251.67 |====================================================== TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better LLVM Clang 3.3 . 624749 |===================================================== GCC 4.8.1 ...... 599455 |=================================================== LLVM Clang 3.2 . 624323 |===================================================== GCC 4.7.3 ...... 631626 |====================================================== Tachyon 0.98.9 Total Time Seconds < Lower Is Better LLVM Clang 3.3 . 10.98 |======================================================= LLVM Clang 3.2 . 10.44 |==================================================== SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better LLVM Clang 3.3 . 1827.34 |==================================================== GCC 4.8.1 ...... 1773.35 |================================================== LLVM Clang 3.2 . 1774.03 |=================================================== GCC 4.7.3 ...... 1861.55 |===================================================== FFmpeg 1.1 H.264 HD To NTSC DV Seconds < Lower Is Better LLVM Clang 3.3 . 13.18 |======================================================= GCC 4.8.1 ...... 12.81 |===================================================== LLVM Clang 3.2 . 12.77 |===================================================== GCC 4.7.3 ...... 12.86 |====================================================== Apache Benchmark 2.4.3 Static Web Page Serving Requests Per Second > Higher Is Better LLVM Clang 3.3 . 25295.82 |=================================================== GCC 4.8.1 ...... 25786.15 |==================================================== LLVM Clang 3.2 . 25888.95 |==================================================== GCC 4.7.3 ...... 25743.99 |==================================================== x264 2013-06-08 H.264 Video Encoding Frames Per Second > Higher Is Better LLVM Clang 3.3 . 153.15 |===================================================== GCC 4.8.1 ...... 156.34 |====================================================== LLVM Clang 3.2 . 155.35 |====================================================== GCC 4.7.3 ...... 155.33 |======================================================