GCC 9 vs. Clang 8 C/C++ Compiler Performance On AMD Threadripper, Intel Core i9
Clang 8 yielded similar performance to GCC 9 for the Redis benchmarks.
The mcperf performance was also similar between compilers.
In total 71 tests were run across the two systems and three tested compilers. All of the data can be found via this OpenBenchmarking.org result file.
When looking at the geometric mean for all of the test results that ran successfully on all of the tested compilers, it was quite interesting. On the Intel Core i9 7980XE system, GCC 9.1 came out slightly ahead of GCC 8.3 but GCC9 measured up to be 7% faster than LLVM Clang 8. On the AMD Ryzen Threadripper 2990WX system, however, Clang 8 and GCC 9 tied for offering the same performance overall. For at least AMD znver1, Clang seems to have a upper-hand to make it more competitive to GCC than we've seen with Intel systems, either due to better Znver1 tuning or other characteristics, but we have follow-up tests running to try to provide an answer on that front. In a number of the benchmarks, Clang 8 yielded faster binaries than GCC 9 and was not something seen during the Skylake-X testing. Anyhow, overall Clang is certainly offering friendly competition to GCC and stay tuned for more of these compiler benchmarks now that GCC 9 has shipped.
If you enjoyed this article consider joining Phoronix Premium to view this site ad-free, multi-page articles on a single page, and other benefits. PayPal or Stripe tips are also graciously accepted. Thanks for your support.