Announcement

Collapse
No announcement yet.

PCC: Portable C Compiler Isn't Quick To Advance

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

  • PCC: Portable C Compiler Isn't Quick To Advance

    Phoronix: PCC: Portable C Compiler Isn't Quick To Advance

    The Portable C Compiler 1.0 was released in April of 2011, but since then there hasn't been many updates out of this open-source compiler that was originally spawned in the late 1970's...

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

  • #2
    I think waiting for C++ support is missing the point. PCC is a C compiler.

    Actually I am impressed of its performance considering that its competitors are backed by such a huge body of developers. PCC can already build Open- and FreeBSD so it is definitely capable. If only someone could crack this challenge

    http://bsdfund.org/bundle/

    things could get interesting

    Comment


    • #3
      If memory serves, there were two main reasons for the interest in resurrecting PCC:

      - A general interest in heirloom code, whether because of simple nostalgia or because such code is believed to be simpler and thus more minimalist. As far as I know, PCC is satisfying this interest quite well enough already.

      - The BSD projects' distaste for GPLv3 had them looking for alternatives to GCC. PCC was briefly considered, but eventually lost out to Clang/LLVM because that project was much further along in meeting the needs of most current software.

      Comment


      • #4
        I wrote a patch a month or two ago to enable PCC to build with Clang, but I forgot to upstream it. I will send that upstream today. Thanks for the reminder.

        Comment


        • #5
          Originally posted by staalmannen View Post
          Actually I am impressed of its performance considering that its competitors are backed by such a huge body of developers.
          Yes I was also surprised, particularly seeing it beating Clang/LLVM. However there were no compiler options listed here which means it could be no optimizations set at all, that and this being only one test (and a synthetic benchmark at that) makes it rather impossible to draw any conclusions from this.

          Comment

          Working...
          X