Announcement

Collapse
No announcement yet.

Pay For Faster Linux Kernel Performance? There's Patches For That

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

  • #31
    Originally posted by alpha_one_x86 View Post
    When the official kernel will support -march=native as CONFIG?
    https://github.com/graysky2/kernel_gcc_patch
    ## VGA ##
    AMD: X1950XTX, HD3870, HD5870
    Intel: GMA45, HD3000 (Core i5 2500K)

    Comment


    • #32
      I have see that's into alternative kernel here:
      http://packages.gentoo.org/category/sys-kernel
      I don't remember what kernel.

      Yes, but harder than test the same program on arm, x86, mips, ...

      And gcc 2 don't support too... Who will compil the last linux kernel with gcc < 4.4?
      Developer of Ultracopier/Supercopier and of the game CatchChallenger

      Comment


      • #33
        Somebody must be really insane to pay for a custom kernel config. The app on github looks like a system optimizer - if i would know something that would enhance speed I would have it by default. Well lots of ppl pay for Windows optimizations, so why not for Linux. It's is just like RAM optimizers, they do nothing useful just let the user "feel" better. If you feel better after paying, everybody is happy.

        Comment


        • #34
          Originally posted by alpha_one_x86 View Post
          I have see that's into alternative kernel here:
          http://packages.gentoo.org/category/sys-kernel
          I don't remember what kernel.

          Yes, but harder than test the same program on arm, x86, mips, ...

          And gcc 2 don't support too... Who will compil the last linux kernel with gcc < 4.4?
          Documentation/Changes

          "Current Minimal Requirements
          [..]
          o Gnu C 3.2 # gcc --version
          "

          It's kernel developers choice - not mine.

          Anyway, the whole discussion about optimizations reminds me
          http://fun.irq.dk/funroll-loops.org/

          Comment


          • #35
            An polish web site did some test of this kernel http://www.dobreprogramy.pl/Cala-pra...ews,59826.html and I know this guy, all changes with he makes in the source code are public if he made some things, but mostly his changes are config changes and setting compiler flags with you don't need to made public according to GPL. But it still is some immoral, I am sure if you apply march=native patch for kernel, adjust config and some sys and proc entries you will get 99% the same result as him.

            Comment


            • #36
              Originally posted by marceel View Post
              Nice work.
              Try to expand your business and optimize other libraries.
              This.

              Comment


              • #37
                Do You know any alternative for AMD-based PC's?
                That's why.

                Comment


                • #38
                  Originally posted by michal View Post
                  When someone will submit a patch, push it through review and prove that it works fine.
                  Or they will say they'll talk to "the compiler guys", then never reply again. Just like they did with this patch: https://lkml.org/lkml/2010/12/20/18 - Does anyone (Bridgman) know why it was never mainlined? I'm using it since years...

                  Comment


                  • #39
                    I'm polish and I visit dobreprogramy.pl pretty often.

                    Those guys who make this kernel are well known among community for spamming forums and comment section with shameless advertisments. In general all of this is kinda suspicious and I honestly would avoid prebuild kernels from unknown source which said group is.

                    I'd say that until some main kernel developer or even Linus himself comment on this - avoid this shit.

                    Comment


                    • #40
                      Originally posted by TAXI View Post
                      Or they will say they'll talk to "the compiler guys", then never reply again. Just like they did with this patch: https://lkml.org/lkml/2010/12/20/18 - Does anyone (Bridgman) know why it was never mainlined? I'm using it since years...
                      I'll just quote

                      "What other linux will let you have a vector optimized wordprocessor?"

                      http://fun.irq.dk/funroll-loops.org/

                      Comment

                      Working...
                      X