LLVM Clang 3.5 Brings Some Compiler Performance Improvements
Compile times appear a bit slower with LLVM Clang 3.5, but they still hold a big advantage over the GCC compile times.
C-Ray, the multi-threaded ray-tracer that always tends to be interesting for compiler tests, is slightly faster with Clang 3.5 on this core-avx-i system.
There might be regressions with FLAC/LAME audio encoding with the new compiler infrastructure.
Well, that's that for our Ivy Bridge LLVM/Clang testing. There are some performance improvements shown by this testing of LLVM Clang 3.5 but in a few workloads the performance was slightly lower. Stay tuned to Phoronix as our open-source compiler benchmarking continues.
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.