GCC 10 vs. GCC 11 Compiler Performance On AMD Zen 3

After recently looking at the early LLVM Clang 12 compiler performance on the AMD Ryzen 9 5950X, in today's benchmarking is a look at how the GCC 11 compiler performance is looking in its near final state compared to GCC 10 under a variety of build CFLAGS/CXXFLAGS configurations on the AMD Zen 3 desktop.

This round of compiler benchmarking is focused on the GCC 10.2 versus GCC 11.0.0 (20210207 development snapshot) performance with the AMD Ryzen 9 5950X. Both GNU Compiler Collection releases were built in the same release configuration mode. The tested CFLAGS/CXXFLAGS under each compiler included:

-O2 -march=x86-64 for the generic and mid-level optimized experience as commonly use by many Linux distributions.

-O2 -march=native for sticking to -O2 but with catering to use the newer CPU instruction set extensions.

-O3 -march=native for enabling more optimizations and catering to the CPU under test, the configuration many enthusiasts and HPC users prefer.

-O3 -march=native -flto for also looking at the performance with link-time optimizations (LTO) enabled.

The Ryzen 9 5950X system was running Ubuntu 20.10 with the Linux 5.8 kernel.

Via the Phoronix Test Suite a wide variety of C/C++ open-source benchmarks were carried out under these compilers. More GCC (and Clang) benchmarks from a diverse spectrum of hardware will come as the official Clang 12 and GCC 11 releases materialize over the next month or two.