A Closer Look At The GCC 8 Compiler Performance On Intel Skylake
With the Himeno pressure solver was another test where the -march=native tuning hard lowered the performance on GCC 8 compared to GCC 7, but at least its performance was still ahead of the generic -O3.
To no surprise, compiling code generally takes longer on GCC 8.1 than GCC 7.3 due to the growing size of the compiler with more optimizations, etc.
C-Ray at -O3 found some nice performance improvements on this Intel Xeon Skylake system with GCC 8.1.
The AOBench performance meanwhile regressed when using the native microarchitecture tuning.
Those wishing to dig through more of these GCC 7.3 vs. GCC 8.1 compiler benchmarks on this Xeon E3 system can find more data via this OpenBenchmarking.org result file. There is also related reading with the GCC 8.1 tests on five AMD/Intel systems and trying Clear Linux with GCC 8.
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.