Announcement

Collapse
No announcement yet.

GCC 8.1 vs. GCC 7.3 Compiler Benchmarks On Five AMD/Intel Linux Systems

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

  • #11
    thanks for the benchmarks always love it when you focus on GCC benchmarks

    Comment


    • #12
      What's the word on Specter/Meltdown mitigations in GCC 7.3 & 8.1? Does the compiler automatically activate the retpoline and other mitigations when it's built and building for the x86-64 architecture (which would seem prudent so long as there's a knob to disable it) or does it require explicit options passed and if so does -march=native tweak those knobs? It would be relevant to just about any series of benchmarks these days (with/without because not all archs need it, and not all work loads even on x86-64 require it)

      Comment


      • #13
        Originally posted by stormcrow View Post
        What's the word on Specter/Meltdown mitigations in GCC 7.3 & 8.1? Does the compiler automatically activate the retpoline and other mitigations when it's built and building for the x86-64 architecture (which would seem prudent so long as there's a knob to disable it) or does it require explicit options passed and if so does -march=native tweak those knobs? It would be relevant to just about any series of benchmarks these days (with/without because not all archs need it, and not all work loads even on x86-64 require it)
        Retpolines are not on by default. So far all userland is built without them for most distros.

        Comment


        • #14
          I know these benchmarks are not intended for comparing different CPUs against each other, but rather compiler versions on the same CPU, but I am still curious: why are the EPYC results so low for most of the benchmarks? What caused that system to perform so much worse in many benchmarks compared to the others?

          Comment


          • #15
            Originally posted by tajjada View Post
            I know these benchmarks are not intended for comparing different CPUs against each other, but rather compiler versions on the same CPU, but I am still curious: why are the EPYC results so low for most of the benchmarks? What caused that system to perform so much worse in many benchmarks compared to the others?
            cpu clock speed probably factors into it

            Comment


            • #16
              What's up with the Coffee Lake performance? It regressed a lot of quite a few tests.

              Comment

              Working...
              X