Announcement

Collapse
No announcement yet.

RadeonSI Ups Its Compiler Threads To Let Shader-DB Run Faster On Modern Systems

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

  • RadeonSI Ups Its Compiler Threads To Let Shader-DB Run Faster On Modern Systems

    Phoronix: RadeonSI Ups Its Compiler Threads To Let Shader-DB Run Faster On Modern Systems

    The RadeonSI compiler queue can now run across more CPU cores/threads of modern systems though it appears this will primarily just benefit those running the shader-db shader test cases...

    http://www.phoronix.com/scan.php?pag...mpiler-Threads

  • #2
    Should hopefully translate to faster load times on many core CPUs for games as well

    Comment


    • #3
      "threads times three divided by four" is an odd way of saying "threads*0.75"

      Comment


      • #4
        Originally posted by schmidtbag View Post
        "threads times three divided by four" is an odd way of saying "threads*0.75"
        it is a difference between integer and floating point numbers

        Comment


        • #5
          It's just literally what the code does :-)

          Code:
          num_comp_hi_threads = hw_threads * 3 / 4;
          And it's not the same for the computer! Since these are integer operations the result is not the same as "threads*0.75"

          Comment


          • #6
            Moah performance.!!


            Any idea how would this translate to gallium nine performance for Starcraft 2 for instance?

            Comment


            • #7
              Originally posted by droste View Post
              It's just literally what the code does :-)

              Code:
              num_comp_hi_threads = hw_threads * 3 / 4;
              And it's not the same for the computer! Since these are integer operations the result is not the same as "threads*0.75"
              Though in practice, I'm not aware of any computers where it wouldn't be. 12, 16, 24, 32, 36, 48, etc. cores all divide evenly. I guess maybe in a virtual environment there might be an odd # of cores assigned.

              Comment


              • #8
                Originally posted by smitty3268 View Post

                Though in practice, I'm not aware of any computers where it wouldn't be. 12, 16, 24, 32, 36, 48, etc. cores all divide evenly. I guess maybe in a virtual environment there might be an odd # of cores assigned.
                Maybe some multi-CPU setup where the CPU's don't have same amount of cores?

                Comment


                • #9
                  marek what were the effects of this change in shaderdb execution time?

                  Comment


                  • #10
                    Originally posted by schmidtbag View Post
                    "threads times three divided by four" is an odd way of saying "threads*0.75"
                    Or simply three quarters

                    Comment

                    Working...
                    X