Announcement

Collapse
No announcement yet.

Benchmarking Linux With The Retpoline Patches For Spectre

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

  • Benchmarking Linux With The Retpoline Patches For Spectre

    Phoronix: Benchmarking Linux With The Retpoline Patches For Spectre

    While the Kernel Page Table Isolation (KPTI) patches were quick to land in the mainline Linux kernel for addressing the Meltdown CPU vulnerability, the "Retpoline" patches are still being worked on as the leading approach on the Linux side for dealing with the Spectre CPU vulnerability. The Retpoline patches are said to have little impact on performance, but here are our benchmarks of these kernel patches for seeing how they affect a variety of AMD and Intel systems.

    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
    Thanks for running these cross-platform tests.
    Additionally, that Postgres benchmark that finally compares post-KPTI Intel chips to non-KPTI AMD parts including their ultra-high end Epyc 7601 REALLY needs some publicity to put things into context and to fight a bunch of the FUD out there. That goes triple for when AMD gets forced to turn KPTI on later because anybody who thinks that Meltdown was the last race-condition bug that will ever be found and that AMD hardware is immune from all attacks is in for a rude wakeup call.

    Comment


    • #3
      I don't know why Intel needs to pay for a PR department; all they need is you. You're so far up their rear-end that it isn't even funny.

      Comment


      • #4
        F*CK! this was suposed to have "negligible performance impact"

        Comment


        • #5
          For very high values of "negligible" ?

          Comment


          • #6
            Originally posted by chuckula View Post
            Thanks for running these cross-platform tests.
            Additionally, that Postgres benchmark that finally compares post-KPTI Intel chips to non-KPTI AMD parts including their ultra-high end Epyc 7601 REALLY needs some publicity to put things into context and to fight a bunch of the FUD out there. That goes triple for when AMD gets forced to turn KPTI on later because anybody who thinks that Meltdown was the last race-condition bug that will ever be found and that AMD hardware is immune from all attacks is in for a rude wakeup call.
            Interesting way to spread FUD. Pretend to be fighting it with the first sentence. Then preceed to spread FUD with the next... Troll?

            Comment


            • #7
              Keep in mind these tests were all done with KPTI present and enabled, just comparing Retpoline today.
              Yikes.

              Comment


              • #8
                Originally posted by perpetually high View Post
                Keep in mind these tests were all done with KPTI present and enabled, just comparing Retpoline today.
                Yikes.
                Michael Just to clarify: You have KPTI (unnecessarily) enabled for AMD?

                Comment


                • #9
                  Michael, did you run with KPTI enabled on the Ryzen and EPYC tests? If so, why? AMD has stated that KPTI is not needed on their processors, and independent tests have confirmed that Ryzen and EPYC are not susceptible to the Meltdown bug. If you have the time, please re-run the AMD tests without KPTI.

                  Thanks!

                  Comment


                  • #10
                    Please re-run the AMD tests without KPTI.

                    Comment

                    Working...
                    X