Announcement

Collapse
No announcement yet.

Linux Sees Fix For "Critical" Scheduler Bug Introduced A Few Months Ago

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

  • Linux Sees Fix For "Critical" Scheduler Bug Introduced A Few Months Ago

    Phoronix: Linux Sees Fix For "Critical" Scheduler Bug Introduced A Few Months Ago

    Coming out Friday night was a series to address a "critical" scheduler issue with the Linux kernel...

    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
    Found the thread on lkml also which makes it easier to read and grab the patches.

    Hopefully this makes it in by tomorrow for 5.4-rc7, as it didn't patch cleanly on 5.3.9 for me but went smoothly with 5.4-rc6. Thanks for the heads up Michael.

    Comment


    • #3
      >Not writing your OS in Rust

      Comment


      • #4
        Good fix. But I'm going to continue nagging about this.

        Performance. Regression. Testing.

        Comment


        • #5
          Originally posted by josh_walrath View Post
          >Not writing your OS in Rust
          Yeah, let's rewrite entire kernel with very young, not verified at all in real life scenarios the Rust language. Can you provide some examples of competitive C counterparts written in Rust?

          Comment


          • #6
            Originally posted by Volta View Post

            Yeah, let's rewrite entire kernel with very young, not verified at all in real life scenarios the Rust language. Can you provide some examples of competitive C counterparts written in Rust?
            ripgrep

            Comment


            • #7
              Originally posted by flashmozzg View Post

              ripgrep
              I'm sure ripgrep is great. It's not an Operating System kernel written entirely in Rust though (which I believe was the comparison Volta was asking for).

              Comment


              • #8
                Originally posted by josh_walrath View Post
                >Not writing your OS in Rust
                Instead of whining, fork the kernel and start the conversion process to Rust.

                Isn't that a common retort that we hear in these forums when someone thinks they have a better solution?

                Seriously...

                Let's see how far you get. Let's see how successful you are compared to a Linux kernel in C.

                Who knows, you might attract some corporate interest somewhere that will help fund your efforts.

                Comment


                • #9
                  Originally posted by ermo View Post

                  I'm sure ripgrep is great. It's not an Operating System kernel written entirely in Rust though (which I believe was the comparison Volta was asking for).
                  Well "competitive C counterparts" read as counterpart to any used C program.
                  As for OSes specifically - Redox could serve as an example.

                  Comment


                  • #10
                    Originally posted by flashmozzg View Post

                    Well "competitive C counterparts" read as counterpart to any used C program.
                    As for OSes specifically - Redox could serve as an example.
                    Yes, exactly. I'm curious if there will be promised benefits from using Rust on the kernel level? I mean concurrency, races, safeness aren't those need to be implemented manually?

                    Comment

                    Working...
                    X