Announcement

Collapse
No announcement yet.

A Run Down Of VT Switching On Linux

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

  • #91
    Originally posted by Ericg View Post
    What I was trying to say was:

    There's not a problem with targetting the 1% but, like here, we run into a problem where the 1% don't like something...but the other 99% have great ideas that they can use with what the 1% doesn't like. Do we limit ourselves because the 1% doesn't like something, or do we just accept that you can't make everyone happy and move forward?
    In this case you move forward, but that isnt the hard part. The hard part is knowing when it's the 1% or when it's the 90%. Most devs can't tell the difference.

    Comment


    • #92
      Originally posted by curaga View Post
      Revoke() is already implemented on BSDs, and I believe on Solaris too. So these people disagree.
      Solaris uses a user space daemon for this, not kernel syscalls.

      This section describes the Solaris (TM) system administration and maintenance utilities and is for system and network administrators.

      Comment


      • #93
        Originally posted by alanc View Post
        Solaris uses a user space daemon for this, not kernel syscalls.

        http://docs.oracle.com/cd/E26502_01/...daemon-1m.html
        Alanc...fucking...BOOM
        All opinions are my own not those of my employer if you know who they are.

        Comment


        • #94
          Originally posted by Vim_User View Post
          What comes next, adding systemd in the mix? There are still tasks out there where init systems are not needed and bring absolutely no advantages (rather disadvantages, having more maintenance work and more parts that can be buggy), why should we change that?
          A bug in the kernel is no different from a bug in the minimal init. The change makes the kernel's bug surface smaller, after all.

          And yea, as Ericg noted, what kind of embedded systems are you working with that don't need an init, anyway? You do need some way of launching bash, at least, don't you?

          Comment


          • #95
            Originally posted by Vim_User View Post
            This is not about having to learn a new way, this is about increasing maintenance work and potential bugs, doubling the count of packages that are to maintain and test.

            Evolution also teaches that more sophisticated systems are more prone to break and that sometimes the simpler systems live while the complicated die. When we are at taking lessons already.
            I was under the impression that these changes would decrease LOC and complexity.

            Comment


            • #96
              Originally posted by GreatEmerald View Post
              And yea, as Ericg noted, what kind of embedded systems are you working with that don't need an init, anyway? You do need some way of launching bash, at least, don't you?
              The kernel can do that fine. Make the test, launch your system with the kernel parameter init=/bin/bash.

              Comment


              • #97
                Originally posted by erendorn View Post
                I was under the impression that these changes would decrease LOC and complexity.
                Maybe LOC, but how is making sure that several new modules interact correctly decreasing complexity?

                Comment


                • #98
                  Originally posted by Vim_User View Post
                  Maybe LOC, but how is making sure that several new modules interact correctly decreasing complexity?
                  Different distinct functionalities won't magically interact better the second you put them in the same module. A part of a given module can do the exact same things that an other part does not expect.

                  Comment


                  • #99
                    Originally posted by alanc View Post
                    Solaris uses a user space daemon for this, not kernel syscalls.

                    http://docs.oracle.com/cd/E26502_01/...daemon-1m.html
                    Thanks for the correction.

                    Originally posted by Ericg
                    Alanc...fucking...BOOM
                    Did you perhaps miss "I believe Solaris also"? I was only sure of BSD. I didn't claim Solaris as a fact, based on a belief, unlike you do, religious preacher.
                    You're no different to people spreading their god. They also preach based on beliefs, not facts.

                    Originally posted by erendorn
                    I was under the impression that these changes would decrease LOC and complexity.
                    Every additional layer by the very definition increases complexity.

                    Comment


                    • Originally posted by Vim_User View Post
                      The kernel can do that fine. Make the test, launch your system with the kernel parameter init=/bin/bash.
                      Aha, so you are using an init after all. Bash itself gets launched, but it doesn't matter. If instead of bash itself, you launch a dispatcher that spawns a terminal and bash for you, then you end up with the exact same thing you have right now. The only difference is that now the terminal is in kernelspace, and after that it will be in userspace.

                      Comment

                      Working...
                      X