GCC vs. Clang Compiler Benchmarking On Intel's Skylake CPU

Written by Michael Larabel in Software on 4 September 2015 at 11:50 AM EDT. Page 2 of 5. 14 Comments.

Even though there was the additional LLVM Clang run with the proper OpenMP arguments set, it didn't seem to make a difference for the Rodinia scientific benchmark that heavily utilizes parallel processing -- thus these OpenMP tests were a clear win for GCC. However, between GCC 4.9 and the current GCC 6 development code for release next year there was no significant performance differences.

With the HMMer test that doesn't make use of OpenMP, the performance between GCC and LLVM Clang were neck-and-neck for this latest-generation Intel desktop processor.

SciMark is a set of micro-tests in C that tend to do very well with LLVM Clang.

With the Monte Carlo sub-test, Clang 3.8 came up shy of GCC on the Skylake Linux system.


Related Articles