Server Processors

[Anonymous]

We are pleased to announce the GA release of AMD Optimizing C/C++ Compiler (AOCC) v1.2 for our public developer community, is now officially available on https://developer.amd.com/amd-aocc/.

The AOCC compiler system is a high performance, production-quality code generation tool. The AOCC environment provides various options to developers when building and optimizing C, C++, and Fortran applications targeting 32-bit and 64-bit Linux platforms. The AOCC compiler system offers a high level of advanced optimizations, multi-threading and processor support that includes global optimization, vectorization, inter-procedural analyses, loop transformations, and code generation. AMD also provides highly optimized libraries, that can optimize performance for each x86 processor core when utilized. The AOCC Compiler Suite simplifies and accelerates development and tuning for x86 applications.

What's New in Release v1.2

• AOCC v1.1 migrated to LLVM 6.0 release (llvm.org, 8th Mar)
• Includes the support for retpolines (from LLVM 6.0) to mitigate "branch target injection" (variant #2) of the "Spectre" speculative side channel attacks described by Google Project Zero and the Spectre paper​
• Introducing Flang, as an experimental Fortran frontend in addition to existing stable and supported DragonEgg Fortran frontend. Refer to Flang (alpha) - The Fortran compiler documentation for more details

[vchuravy]

The download links on https://developer.amd.com/amd-aocc/ are broken.

[Anonymous]

Thank you for letting me know vchuravy. I have updated the links, please try again and let me know if you are still experiencing issues.

[bero]

Is the source for this compiler/patchset available anywhere? Would be great to get it into Linux distros so Zen boxes would perform better out of the box...

In OpenMandriva, we use clang as our primary system compiler, so the patchset would really help making the entire OS faster.