Announcement

Collapse
No announcement yet.

AMD & NVIDIA Acceleration With GCC, LLVM Clang

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • AMD & NVIDIA Acceleration With GCC, LLVM Clang

    Phoronix: AMD & NVIDIA Acceleration With GCC, LLVM Clang

    With yesterday's merging of AMD HSA support into GCC 6, questions have been raised by readers in taking advantage of this acceleration support for AMD APUs and related items...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    Indeed, an updated guide to getting things working would be appreciated, especially if HCC is among the compilers probed.

    Comment


    • #3
      As of LLVM 3.8, the AMDGPU backend supports emitting code (binary ISA, not HSAIL) that can be executed with the HSA runtime. You just need to make sure you are using this runtime branch: https://github.com/tstellarAMD/hsa-runtime/tree/hcc and the HSA 1.6 driver release from here: https://github.com/HSAFoundation/HSA-Drivers-Linux-AMD. There are instructions in the samples directory for the runtime branch for how to make this work.

      Comment


      • #4
        Queston to those that actually got it working/running, what was the speedup times, if any?

        Comment


        • #5
          There has been more work on the HSAIL backend since July, just not on the default branch shown on github

          Comment

          Working...
          X