GCC 8.0 vs. LLVM Clang 6.0 On AMD EPYC

At the beginning of January I posted some early LLVM Clang 6.0 benchmarks on AMD EPYC while in this article is comparing the tentative Clang 6.0 performance to that of the in-development GCC 8.0. Both compilers are now into their feature freeze and this testing looked at the performance of generated binaries both for generic x86_64 as well as being tuned for AMD's Zen "znver1" microarchitecture.

From the AMD EPYC 7601 32-core / 64-thread system running within the Tyan B8026T70AE24HR, benchmarks were done of the latest in-development GCC 8.0 and LLVM Clang 6.0 compilers. Tests were done when the CFLAGS/CXXFLAGS were set to "-O3 -march=x86-64" and then again at "-O3 -march=znver1" for seeing the performance of generic x86-64 targeting and then again when specifically targeting the compilers at these first-generation Zen CPU cores. Tests were done from Ubuntu 17.10 running on the Linux 4.13 kernel x86_64.

These compiler benchmarks were carried out using the Phoronix Test Suite open-source and automated benchmarking software.