Announcement

Collapse
No announcement yet.

DragonFlyBSD 3.2.1 Battles Against Linux For Speed

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

  • DragonFlyBSD 3.2.1 Battles Against Linux For Speed

    Phoronix: DragonFlyBSD 3.2.1 Battles Against Linux For Speed

    The much-anticipated release of DragonFlyBSD 3.2 is now available as it enhances its performance to better compete with Linux in multi-core environments...

    http://www.phoronix.com/vr.php?view=MTIyMTM

  • #2
    Originally posted by phoronix View Post
    Phoronix: DragonFlyBSD 3.2.1 Battles Against Linux For Speed

    The much-anticipated release of DragonFlyBSD 3.2 is now available as it enhances its performance to better compete with Linux in multi-core environments...

    http://www.phoronix.com/vr.php?view=MTIyMTM
    a monolithic kernel is always faster than a kernel that passes messages queues. it would be almost impossible to beat linux in performance.

    Comment


    • #3
      Obligatory "Why is everyone working on different things? Why don't we all work together in unison to build the ultimate kernel/distro/package format" post.

      Comment


      • #4
        Originally posted by Ancurio View Post
        Obligatory "Why is everyone working on different things? Why don't we all work together in unison to build the ultimate kernel/distro/package format" post.
        Everyone is working on different things, because everyone is trying to achieve different goals.
        There is no perfect (suitable for every task) kernel/distro/package format.

        Comment


        • #5
          Originally posted by garegin View Post
          a monolithic kernel is always faster than a kernel that passes messages queues. it would be almost impossible to beat linux in performance.
          DragonFlyBSD doesn't use message quene so much.
          http://www.phoronix.com/scan.php?pag...tem&px=MTIwNjA
          Of course 2.6.32 is a bit old.

          Comment


          • #6
            It is monolithic.. with some messaging infrastructure when needed

            Originally posted by garegin View Post
            a monolithic kernel is always faster than a kernel that passes messages queues. it would be almost impossible to beat linux in performance.
            The DragonFly kernel *is* monolithic. There are some various api's in kernel to support different types of multithreaded scenarios
            when those scenarios would be better than some other alternative (e.g. a lock, etc) -
            But it is *not* a mach-style microkernel where everything is message passing, etc.

            disclamer: I am a dragonfly dev

            Comment


            • #7
              Originally posted by LightBit View Post
              Everyone is working on different things, because everyone is trying to achieve different goals.
              There is no perfect (suitable for every task) kernel/distro/package format.
              http://en.wikipedia.org/wiki/Sarcasm

              Comment


              • #8
                Originally posted by dfcat View Post
                The DragonFly kernel *is* monolithic. There are some various api's in kernel to support different types of multithreaded scenarios
                when those scenarios would be better than some other alternative (e.g. a lock, etc) -
                But it is *not* a mach-style microkernel where everything is message passing, etc.

                disclamer: I am a dragonfly dev
                Wikipedia disagrees with you: https://en.wikipedia.org/wiki/DragonFly_BSD#Kernel.

                "DragonFly's kernel is a hybrid, containing features of both monolithic and microkernels, such as the message passing capability of microkernels enabling larger portions of the OS to benefit from protected memory, as well as retaining the speed of monolithic kernels for certain critical tasks. The messaging subsystem being developed is similar to those found in microkernels such as Mach, though it is less complex by design. DragonFly's messaging subsystem has the ability to act in either a synchronous or asynchronous fashion, and attempts to use this capability to achieve the best performance possible in any given situation.[7]"
                There's more to kernels than just a _strict_ monolithic or microkernel approach.

                Originally posted by Ancurio View Post
                https://en.wikipedia.org/wiki/Stupidity

                4 years ago the _microkernel_ Minix3 creator Andrew Tanenbaum was granted 2,5 million € for his "Research on Really Reliable and Secure Systems Software": http://www.minix3.org/news/index.html

                I don't know how well you can reload drivers or whole subsystems after a crash in a monolithic kernel but usually the main focus is very fundamentally on performance and then on everything else: http://www.minix3.org/other/research-projects.html

                Automatic recovery from fatal system errors

                In most operating systems, a fatal error within the operating system, for example, referencing an invalid pointer, leads almost immediately to a system crash. In MINIX 3, such an error leads to the crash of one of the operating system components, but not the entire system. The crash is reported to a system component called the reincarnation server, which takes an appropriate action, typically including logging the event, notifying the system administrator, and restarting the failed component. For stateless components, the recover is very quick and completely transparent to application processes, that is, they do not even notice the failure and recovery. Research on automatic recovery of stateful components is underway. Sample paper on recovery in MINIX 3.
                Last edited by ArchLinux; 11-04-2012, 01:40 PM.

                Comment


                • #9
                  Originally posted by ArchLinux View Post
                  What is your fucking point?

                  Comment


                  • #10
                    Originally posted by Ancurio View Post
                    What is your fucking point?
                    The irony.

                    Comment


                    • #11
                      Originally posted by ArchLinux View Post
                      Wikipedia disagrees with you: https://en.wikipedia.org/wiki/DragonFly_BSD#Kernel.


                      There's more to kernels than just a _strict_ monolithic or microkernel approach.
                      http://www.realworldtech.com/forum/?...urpostid=65915

                      Comment


                      • #12
                        Originally posted by ArchLinux View Post
                        Wikipedia disagrees with you: https://en.wikipedia.org/wiki/DragonFly_BSD#Kernel.

                        There's more to kernels than just a _strict_ monolithic or microkernel approach.
                        When these kernels are called hybrid kernels, they are usually called that because they support loadable modules. They can still be compiled as strictly monolithic kernels and they behave no differently than "true" monolithic kernels once loaded. As far as the developers working on them are concerned, they are monolithic kernels.

                        Comment


                        • #13
                          Originally posted by ryao View Post
                          When these kernels are called hybrid kernels, they are usually called that because they support loadable modules.
                          Uhh.. no?

                          They are called that because some of their services live in user space.

                          Correction. Yeah, I've seen that through Wikiquote but I still use the term. Just makes more sense.
                          Last edited by ArchLinux; 11-05-2012, 01:01 PM.

                          Comment


                          • #14
                            who the fuck runs bsd??

                            their drivers suck even more (if that's possible)

                            the versions of common shit like firefox and chromium are lagging behind.

                            gnome 2? lol wtf

                            Comment


                            • #15
                              Originally posted by garegin View Post
                              a monolithic kernel is always faster than a kernel that passes messages queues.
                              If coded to peak efficiency for all possible H/W configs.

                              I note that monolithic S/W designs are MUCH harder to code to peak efficiency.

                              Comment

                              Working...
                              X