AMD AOCC 1.3 Compiler Benchmarks vs. GCC vs. LLVM Clang 2 x AMD EPYC 7601 compiler benchmarks on a future article for Phoronix.com by Michael Larabel. GCC 8.2.0: Processor: 2 x AMD EPYC 7601 32-Core @ 3.10GHz (64 Cores / 128 Threads), Motherboard: Dell 02MJ3T (1.2.5 BIOS), Chipset: AMD Family 17h, Memory: 516096MB, Disk: 120GB SSDSCKJB120G7R + 20 x 500GB Samsung SSD 860, Graphics: Matrox Matrox G200eW3, Monitor: VE228, Network: Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA OS: Ubuntu 18.10, Kernel: 4.19.0-041900-generic (x86_64), Compiler: GCC 8.2.0, File-System: ext4, Screen Resolution: 1600x1200 LLVM Clang 7.0: Processor: 2 x AMD EPYC 7601 32-Core @ 3.10GHz (64 Cores / 128 Threads), Motherboard: Dell 02MJ3T (1.2.5 BIOS), Chipset: AMD Family 17h, Memory: 516096MB, Disk: 120GB SSDSCKJB120G7R + 20 x 500GB Samsung SSD 860, Graphics: Matrox Matrox G200eW3, Monitor: VE228, Network: Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA OS: Ubuntu 18.10, Kernel: 4.19.0-041900-generic (x86_64), Compiler: Clang 7.0.0-3 + LLVM 7.0.0, File-System: ext4, Screen Resolution: 1600x1200 AMD AOCC 1.3: Processor: 2 x AMD EPYC 7601 32-Core @ 3.10GHz (64 Cores / 128 Threads), Motherboard: Dell 02MJ3T (1.2.5 BIOS), Chipset: AMD Family 17h, Memory: 516096MB, Disk: 120GB SSDSCKJB120G7R + 20 x 500GB Samsung SSD 860, Graphics: Matrox Matrox G200eW3, Monitor: VE228, Network: Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA OS: Ubuntu 18.10, Kernel: 4.19.0-041900-generic (x86_64), Compiler: Clang 7.0.0, File-System: ext4, Screen Resolution: 1600x1200 Rodinia 2.4 Test: OpenMP Streamcluster Seconds < Lower Is Better GCC 8.2.0 ...... 23.33 |======================================================= LLVM Clang 7.0 . 21.81 |=================================================== AMD AOCC 1.3 ... 22.41 |===================================================== High Performance Conjugate Gradient 3.0 GFLOP/s > Higher Is Better GCC 8.2.0 ...... 0.50 |=============================================== LLVM Clang 7.0 . 0.50 |=============================================== AMD AOCC 1.3 ... 0.60 |======================================================== PolyBench-C 4.2 Test: Correlation Computation Seconds < Lower Is Better GCC 8.2.0 ...... 5.36 |====================================================== LLVM Clang 7.0 . 5.53 |======================================================== AMD AOCC 1.3 ... 5.33 |====================================================== PolyBench-C 4.2 Test: 3 Matrix Multiplications Seconds < Lower Is Better GCC 8.2.0 ...... 3.70 |================================================== LLVM Clang 7.0 . 4.13 |======================================================== AMD AOCC 1.3 ... 3.65 |================================================= FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 2048 Mflops > Higher Is Better GCC 8.2.0 ...... 27361 |======================================================= LLVM Clang 7.0 . 26574 |===================================================== AMD AOCC 1.3 ... 26619 |====================================================== FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 2048 Mflops > Higher Is Better GCC 8.2.0 ...... 14829 |====================================================== LLVM Clang 7.0 . 14580 |===================================================== AMD AOCC 1.3 ... 15056 |======================================================= Timed HMMer Search 2.3.2 Pfam Database Search Seconds < Lower Is Better GCC 8.2.0 ...... 6.52 |======================================================= LLVM Clang 7.0 . 6.12 |=================================================== AMD AOCC 1.3 ... 6.69 |======================================================== Timed MAFFT Alignment 7.392 Multiple Sequence Alignment Seconds < Lower Is Better GCC 8.2.0 ...... 3.66 |======================================================== LLVM Clang 7.0 . 3.40 |==================================================== AMD AOCC 1.3 ... 3.53 |====================================================== Fhourstones 3.1 Complex Connect-4 Solving Kpos / sec > Higher Is Better GCC 8.2.0 ...... 10431 |===================================================== LLVM Clang 7.0 . 10905 |======================================================= AMD AOCC 1.3 ... 10799 |====================================================== SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better GCC 8.2.0 ...... 1796 |==================================================== LLVM Clang 7.0 . 1925 |======================================================== AMD AOCC 1.3 ... 1766 |=================================================== SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better GCC 8.2.0 ...... 556 |========================================================= LLVM Clang 7.0 . 552 |========================================================= AMD AOCC 1.3 ... 552 |========================================================= SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better GCC 8.2.0 ...... 231 |========================================================= LLVM Clang 7.0 . 218 |====================================================== AMD AOCC 1.3 ... 226 |======================================================== SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better GCC 8.2.0 ...... 2554 |=================================================== LLVM Clang 7.0 . 2513 |================================================== AMD AOCC 1.3 ... 2827 |======================================================== SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better GCC 8.2.0 ...... 3953 |============================================= LLVM Clang 7.0 . 4911 |======================================================== AMD AOCC 1.3 ... 3795 |=========================================== SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better GCC 8.2.0 ...... 1688 |======================================================== LLVM Clang 7.0 . 1429 |=============================================== AMD AOCC 1.3 ... 1430 |=============================================== TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better GCC 8.2.0 ...... 853137 |=================================================== LLVM Clang 7.0 . 911886 |====================================================== AMD AOCC 1.3 ... 892960 |===================================================== x264 2018-09-25 H.264 Video Encoding Frames Per Second > Higher Is Better GCC 8.2.0 ...... 143 |======================================================= LLVM Clang 7.0 . 147 |========================================================= AMD AOCC 1.3 ... 145 |======================================================== Stockfish 9 Total Time Nodes Per Second > Higher Is Better GCC 8.2.0 ...... 102909858 |================================================= LLVM Clang 7.0 . 107358539 |=================================================== AMD AOCC 1.3 ... 107516593 |=================================================== Swet 1.5.16 Average Operations Per Second > Higher Is Better GCC 8.2.0 ...... 493058080 |=================================================== LLVM Clang 7.0 . 212870591 |====================== AMD AOCC 1.3 ... 230056272 |======================== Timed Apache Compilation 2.4.7 Time To Compile Seconds < Lower Is Better GCC 8.2.0 ...... 25.56 |======================================= LLVM Clang 7.0 . 25.07 |====================================== AMD AOCC 1.3 ... 36.18 |======================================================= Timed LLVM Compilation 6.0.1 Time To Compile Seconds < Lower Is Better GCC 8.2.0 ...... 150 |=========================================== LLVM Clang 7.0 . 145 |========================================== AMD AOCC 1.3 ... 198 |========================================================= Timed PHP Compilation 7.1.9 Time To Compile Seconds < Lower Is Better GCC 8.2.0 ...... 69.33 |===================== LLVM Clang 7.0 . 109.00 |================================= AMD AOCC 1.3 ... 179.00 |====================================================== Primesieve 7.1 1e12 Prime Number Generation Seconds < Lower Is Better GCC 8.2.0 ...... 6.25 |======================================================== LLVM Clang 7.0 . 5.78 |==================================================== AMD AOCC 1.3 ... 5.81 |==================================================== AOBench Size: 2048 x 2048 - Total Time Seconds < Lower Is Better GCC 8.2.0 ...... 44.51 |=================================================== LLVM Clang 7.0 . 48.46 |======================================================= AMD AOCC 1.3 ... 45.19 |=================================================== Bullet Physics Engine 2.81 Test: Raytests Seconds < Lower Is Better GCC 8.2.0 ...... 3.14 |======================================================== LLVM Clang 7.0 . 3.15 |======================================================== AMD AOCC 1.3 ... 3.14 |======================================================== Bullet Physics Engine 2.81 Test: 3000 Fall Seconds < Lower Is Better GCC 8.2.0 ...... 5.07 |====================================================== LLVM Clang 7.0 . 5.23 |======================================================== AMD AOCC 1.3 ... 5.07 |====================================================== Bullet Physics Engine 2.81 Test: 1000 Stack Seconds < Lower Is Better GCC 8.2.0 ...... 5.81 |===================================================== LLVM Clang 7.0 . 6.15 |======================================================== AMD AOCC 1.3 ... 5.87 |===================================================== Bullet Physics Engine 2.81 Test: 1000 Convex Seconds < Lower Is Better GCC 8.2.0 ...... 5.37 |======================================================== LLVM Clang 7.0 . 5.26 |======================================================= AMD AOCC 1.3 ... 5.23 |======================================================= Bullet Physics Engine 2.81 Test: 136 Ragdolls Seconds < Lower Is Better GCC 8.2.0 ...... 3.11 |===================================================== LLVM Clang 7.0 . 3.29 |======================================================== AMD AOCC 1.3 ... 3.16 |====================================================== Bullet Physics Engine 2.81 Test: Prim Trimesh Seconds < Lower Is Better GCC 8.2.0 ...... 1.14 |======================================================== LLVM Clang 7.0 . 1.13 |======================================================== AMD AOCC 1.3 ... 1.10 |====================================================== Bullet Physics Engine 2.81 Test: Convex Trimesh Seconds < Lower Is Better GCC 8.2.0 ...... 1.35 |======================================================== LLVM Clang 7.0 . 1.35 |======================================================== AMD AOCC 1.3 ... 1.33 |======================================================= Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 Seconds < Lower Is Better GCC 8.2.0 ...... 13.48 |==================================================== LLVM Clang 7.0 . 14.29 |======================================================= AMD AOCC 1.3 ... 13.47 |==================================================== FLAC Audio Encoding 1.3.2 WAV To FLAC Seconds < Lower Is Better GCC 8.2.0 ...... 13.32 |======================================================= LLVM Clang 7.0 . 12.23 |================================================== AMD AOCC 1.3 ... 12.84 |===================================================== Mencoder 1.3.0 AVI To LAVC Seconds < Lower Is Better GCC 8.2.0 ...... 22.29 |======================================================= LLVM Clang 7.0 . 22.33 |======================================================= AMD AOCC 1.3 ... 22.20 |======================================================= Minion 1.8 Benchmark: Graceful Seconds < Lower Is Better GCC 8.2.0 ...... 56.41 |======================================================= LLVM Clang 7.0 . 55.67 |====================================================== AMD AOCC 1.3 ... 55.19 |====================================================== Minion 1.8 Benchmark: Solitaire Seconds < Lower Is Better GCC 8.2.0 ...... 86.81 |======================================================= LLVM Clang 7.0 . 83.51 |===================================================== AMD AOCC 1.3 ... 86.13 |======================================================= Minion 1.8 Benchmark: Quasigroup Seconds < Lower Is Better GCC 8.2.0 ...... 147 |========================================================= LLVM Clang 7.0 . 142 |======================================================= AMD AOCC 1.3 ... 142 |======================================================= Aircrack-ng 1.3 k/s > Higher Is Better GCC 8.2.0 ...... 81752 |====================================================== LLVM Clang 7.0 . 83710 |======================================================= AMD AOCC 1.3 ... 83435 |======================================================= libjpeg-turbo tjbench 1.5.3 Test: Decompression Throughput Megapixels/sec > Higher Is Better GCC 8.2.0 ...... 147 |========================================================= LLVM Clang 7.0 . 141 |======================================================= AMD AOCC 1.3 ... 145 |======================================================== Redis 4.0.8 Test: GET Requests Per Second > Higher Is Better GCC 8.2.0 ...... 2235376 |===================================================== LLVM Clang 7.0 . 2150558 |=================================================== AMD AOCC 1.3 ... 2150203 |=================================================== Redis 4.0.8 Test: SET Requests Per Second > Higher Is Better GCC 8.2.0 ...... 1419694 |===================================================== LLVM Clang 7.0 . 1364289 |=================================================== AMD AOCC 1.3 ... 1405199 |==================================================== Memcached mcperf 1.5.10 Method: Get Operations Per Second > Higher Is Better GCC 8.2.0 ...... 63050 |=================================================== LLVM Clang 7.0 . 58078 |=============================================== AMD AOCC 1.3 ... 67604 |======================================================= Memcached mcperf 1.5.10 Method: Set Operations Per Second > Higher Is Better GCC 8.2.0 ...... 62454 |======================================================= LLVM Clang 7.0 . 57082 |================================================== AMD AOCC 1.3 ... 43395 |====================================== Hierarchical INTegration 1.0 Test: DOUBLE QUIPs > Higher Is Better GCC 8.2.0 ...... 615241438 |================================================== LLVM Clang 7.0 . 618373040 |================================================== AMD AOCC 1.3 ... 624671701 |===================================================