Announcement

Collapse
No announcement yet.

Clang Lands A Virtual File-System

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

  • Clang Lands A Virtual File-System

    Phoronix: Clang Lands A Virtual File-System

    LLVM's Clang compiler now has a virtual file-system implementation...

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

  • #2
    With great software comes great bloat.

    Comment


    • #3
      Originally posted by carewolf View Post
      With great software comes great bloat.
      Why not tell the developer, he's probably unaware of this, I'm sure he'll listen to your wise conclusion.

      Edit: I noticed it's just a few hundred LOCs, I thought it would take at least a few thousand.

      Comment


      • #4
        Originally posted by mark45 View Post
        Why not tell the developer, he's probably unaware of this, I'm sure he'll listen to your wise conclusion.

        Edit: I noticed it's just a few hundred LOCs, I thought it would take at least a few thousand.
        It is not really a virtual file-system, that would be bloat (hence the joke). It is really just an new index over data already availble (the install path). As such it is very little code, but also not a virtual file-system.

        Comment


        • #5
          Originally posted by carewolf View Post
          With great software comes great bloat.
          So, they're making things more efficient by doing path-mapping tricks in memory instead of expensive disk activity (in the form of unnecessary file copies), and you're calling that bloat?

          Comment


          • #6
            Originally posted by carewolf View Post
            With great software comes great bloat.
            The commit added only 300 lines of code...

            Comment


            • #7
              Originally posted by carewolf View Post
              It is not really a virtual file-system, that would be bloat (hence the joke). It is really just an new index over data already availble (the install path). As such it is very little code, but also not a virtual file-system.
              Quit while you're behind.

              Comment


              • #8
                I think you buried the lead here. They're talking about implementing modules in C++. That's huge.

                Comment


                • #9
                  Originally posted by Skrapion View Post
                  I think you buried the lead here. They're talking about implementing modules in C++. That's huge.
                  Thats not actually not new news unless you missed some previous news stories. Modules, as opposed to tons of header files, are on the white board for the next C++ Spec update

                  Comment


                  • #10
                    Originally posted by Ericg View Post
                    Thats not actually not new news unless you missed some previous news stories. Modules, as opposed to tons of header files, are on the white board for the next C++ Spec update
                    That sentence was a grammatical nightmare... Lets try this again.

                    Unless you've missed a few headlines, that is not actually 'new' news. Modules are supposed to a part of the next C++ Spec update, pdf on them here: http://llvm.org/devmtg/2012-11/Gregor-Modules.pdf

                    Comment


                    • #11
                      Oh, trust me, I've been following it closely. But the first modules paper came out in 2004, and the latest modules paper came out in 2012 (the same year as the slides you posted) and there hasn't really been any news since.

                      The possibility of having an implementation of modules after a decade of talk? Yeah, that's news.

                      Comment

                      Working...
                      X