AOCC 3.1 Compiler Performance Against Clang 12, GCC 11 On AMD EPYC
Following the recent benchmarks seeing how AMD's new AOCC 3.1 compiler has brought some performance improvements over the prior AOCC 3.0 release that introduced initial Zen 3 optimizations, here are some benchmarks looking at how that latest AMD Optimizing C/C++ Compiler performance compares to the upstream LLVM Clang 12 compiler for which it is based as well as against GCC 11 as the latest GNU compiler release that remains common to Linux systems.
In my testing of the AMD Optimizing C/C++ Compiler over the past week, AOCC 3.1 has been a pleasant incremental update over AOCC 3.0 that debuted alongside the EPYC 7003 "Milan" processors back in March.
My AOCC 3.1 testing has been largely with the AMD EPYC 7543 32-core Milan processor with the Tyan S8036GM2NE platform that I have been testing the past number of weeks. For this latest-generation AMD server platform tested with the recent Ubuntu 21.04, AOCC 3.1 has generally been delivering a slight edge up over the latest upstream LLVM Clang 12.0.1 compiler as well as the GCC 11.1 stable release. Throughout the CFLAGS/CXXFLAGS were maintained at "-O3 -march=znver3" for catering to this Zen 3 / Milan server processor.