Announcement

Collapse
No announcement yet.

Linux 5.10 Slated To Use New Intel SERIALIZE In Fending Off Speculative Execution Bugs

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

  • Linux 5.10 Slated To Use New Intel SERIALIZE In Fending Off Speculative Execution Bugs

    Phoronix: Linux 5.10 Slated To Use New Intel SERIALIZE In Fending Off Speculative Execution Bugs

    Queued now in the "x86/cpu" development branch ahead of the Linux 5.10 kernel later this year is the change to make use of Intel's new "SERIALIZE" instruction within the kernel's "sync_core" code that is used for stopping the speculative execution and prefetching of modified code...

    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
    Typos:

    Originally posted by phoronix View Post
    SERIALIZE is used for ensuring all flags/register/memory modifications are complete and all buffered wrties drained to memory before proceeding to the next instruction. SERIALIZE comes as a result of the speculative execution bugs hitting Intel particularly hard over the past few years.

    Comment


    • #3
      Originally posted by tildearrow View Post
      Typos:
      See, if serialize was already in the kernel then his keyboard would get the letters in the right order!

      Comment


      • #4
        Yikes!

        Oh my god.

        Ensuring all flags/register/memory modifications are complete and all buffered writes drained to memory before proceeding to the next instruction?

        That's absolutely insane.

        Dump Intel, and adopt AMD.

        That's the only real solution, though I know it's not immediately economically feasible for many companies and individuals.

        Comment


        • #5
          Originally posted by muncrief View Post
          Yikes!

          Oh my god.

          Ensuring all flags/register/memory modifications are complete and all buffered writes drained to memory before proceeding to the next instruction?

          That's absolutely insane.

          Dump Intel, and adopt AMD.

          That's the only real solution, though I know it's not immediately economically feasible for many companies and individuals.



          Not only is it not insane, but AMD will invariably copy it since anybody who has ever wanted to deal with deterministic system behavior would love something like this. Not to mention that since AMD is a late-adopter of SMT it is just as vulnerable to SMT-class vulnerabilities as POWER and Intel systems are and when ARM finally gets around to copying SMT it will be vulnerable too. In fact, plenty of existing systems have features like this for RAS so simplifying it into a single instruction is a great idea.

          Comment

          Working...
          X