GCC 6.1 vs. LLVM Clang 3.9 Compiler Performance

With the Hmmer Pfam database search test, LLVM Clang 3.9 was slightly faster than GCC 6.1 on this Haswell E5 Xeon system.

With the SciMark2 micro-benchmarks, LLVM Clang 3.9 came out ahead by a long shot when looking at the composite metric.

With the Monte Carlo test, GCC 6.1 though is doing much better than Clang.

For a number of the SciMark test runs the results were very close between these competing compilers...

Where Clang came out significantly ahead of GCC to boost its composite performance figure was in the Dense LU Matrix Factorization test.

Clang also won by a noticeable amount in the Jacobi test too.