GCC 4.8 Compiler On AMD's Eight-Core Piledriver

Written by Michael Larabel in Software on 30 October 2012 at 10:30 AM EDT. Page 1 of 4. 2 Comments.

This month from CPUs based upon AMD's new Piledriver micro-architecture I have delivered results of compiler tuning on AMD's Open64 compiler as well as GCC bdver2 tuning. That initial testing from an AMD FX-8350 Eight-Core processor didn't show any big boost out of the "bdver2" target with the new BMI/TBM/F16C/FMA3 instruction set extensions. Testing in this article from the AMD FX-8350 are GCC compiler benchmarks of the 4.6.3, 4.7.2, and 4.8.0 development snapshots to look for performance improvements on this new high-end AMD processor when using the very latest GCC compiler code.

Before getting too excited though, the GCC 4.8 benchmarks of the AMD FX-8350 Vishera/Piledriver processor aren't too exciting. Even for tests exploiting "-march=native" for the bdver2 optimizations there still isn't much to see out of the new instruction sets found on this second-generation Bulldozer CPU. In some test profiles, there are slight changes in performance between GCC 4.6 and the yet-to-be-released GCC 4.8, but nothing too major.

GCC 4.8 Compiler On AMD Piledriver bdver2

While GCC 4.8 may not be fully exploiting the Piledriver-based AMD CPUs, at least there are a number of other interesting features. GCC 4.8 is nearing its end of stage one development and should be officially released in early 2013.

On the following pages is the GNU Compiler Collection comparison on the three tested GCC versions from the AMD FX-8350. If you missed my launch-day review of the AMD FX-8350, the Linux review is here.


Related Articles