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 1 of 5. 14 Comments.

Continuing in our compiler benchmarks this week are some GCC vs. Clang C/C++ compiler performance benchmarks on Intel's new Skylake processor while testing from Ubuntu Linux 64-bit.

This marks our first compiler comparison using Intel's new Skylake architecture. The compilers tested for this benchmarking roundabout included GCC 4.9.3, GCC 5.2.0, GCC 6.0.0 20150830 SVN, and LLVM Clang 3.8.0 SVN. All of the tested compilers were built from source in their release/optimized modes. With the Clang tests there were two runs from the 3.8.0 code: one with the same CFLAGS/CXXFLAGS as the other compiler runs and then one with "-fopenmp=libomp -fno-exceptions" added in, per yesterday's article about the OpenMP support present since LLVM Clang 3.7.0. Per our set CFLAGS/CXXFLAGS set, and illustrated on the relevant test results, "-O3 -march=native" was utilized for trying to maximize the performance out of this Core i5 6600K processor.

All of these GCC and Clang Linux compiler benchmarks were carried out in a fully-automated and reproducible manner using the open-source Phoronix Test Suite benchmarking software.


Related Articles