Announcement

Collapse
No announcement yet.

ORC Unwinder For Linux 4.14, Boosts Kernel Performance By Disabling Frame Pointers

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

  • ORC Unwinder For Linux 4.14, Boosts Kernel Performance By Disabling Frame Pointers

    Phoronix: ORC Unwinder For Linux 4.14, Boosts Kernel Performance By Disabling Frame Pointers

    Ingo Molnar submitted the Linux x86 Assembly updates today for the 4.14 merge window. What's interesting with the x86/asm code changes is the introduction of the ORC Unwinder...

    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
    From the pull request:

    Remove lguest support: its intended role was that of a temporary proof of
    concept for virtualization, plus its removal will enable the reduction
    (removal) of the paravirt API as well, so Rusty agreed to its removal.
    (Juergen Gross)
    So am I reading this right that they are planning to completely remove paravirtualization support? Sure, hw virt support is fairly common nowadays, but usually they are pretty conservative about removing features that someone might still depend on?

    Comment


    • #3
      Originally posted by jabl View Post
      From the pull request:



      So am I reading this right that they are planning to completely remove paravirtualization support? Sure, hw virt support is fairly common nowadays, but usually they are pretty conservative about removing features that someone might still depend on?
      I usually presume these people run a LTS kernel on Red Hat/SuSE.

      Comment


      • #4
        Originally posted by jabl View Post
        From the pull request:



        So am I reading this right that they are planning to completely remove paravirtualization support? Sure, hw virt support is fairly common nowadays, but usually they are pretty conservative about removing features that someone might still depend on?
        the way I read is that the plan is to minimize paravirt to the absolute minimal possible. Paravirt is pretty expensive in terms of how invasive it is to the code (never mind the performance impact of it)

        Comment


        • #5
          My kernel builds ... as always

          CONFIG_FRAME_POINTER is not set

          [NeteXt'73]

          Comment


          • #6
            Originally posted by rubdos View Post
            I usually presume these people run a LTS kernel on Red Hat/SuSE.
            I'm pretty positive that RHEL/SLES are all run on hardware that supports hw virtualization already.

            I mean even potatoes have hw virtualization support nowadays.

            Comment


            • #7
              Does ORC acronym have any meaning besides joking with dwarf?

              Comment


              • #8
                Code:
                $ uname -a
                Linux localhost 4.12.10-1-ck-sandybridge #1 SMP PREEMPT Wed Aug 30 15:49:36 EDT 2017 x86_64 GNU/Linux
                
                $ zgrep CONFIG_FRAME_POINTER /proc/config.gz
                CONFIG_FRAME_POINTER=y
                Ah. True story on Arch.

                Comment


                • #9
                  It is not set by default on Debian Stable.
                  Code:
                  # CONFIG_FRAME_POINTER is not set

                  Comment


                  • #10
                    Originally posted by andrei_me View Post
                    Does ORC acronym have any meaning besides joking with dwarf?
                    I don't know for sure, but I doubt it.

                    In case you aren't aware, the name DWARF is itself a joke based on the ELF binary format. https://en.wikipedia.org/wiki/DWARF

                    Comment

                    Working...
                    X