Loading ...
Sorry, an error occurred while loading the content.

Re: progress on native CMS runtime for GCC

Expand Messages
  • kerravon86
    ... Thanks for that wonderful news about it working fine on z/VM. ... Actually, another option you have, which I would like to see, is to get PDPCLIB working
    Message 1 of 18 , Feb 1, 2009
    • 0 Attachment
      --- In hercules-os380@yahoogroups.com, Dave Jones <dave@...> wrote:
      >
      > Thanks for the information, Bob, I appreciate it. I have
      > installed Paul's latest GCC for CMS product here, and have
      > actually gotten it to compile without errors a fairly
      > significant chuck of code, ~2.4K lines.

      Thanks for that wonderful news about it working
      fine on z/VM.

      > Now I need to see how GCCCMS compares with both
      > the older z/VM C compiler and the brand new z/VM C/C++ one.
      > I'm using them at the moment
      > via their "System Programming" interface, which removes the
      > need for LE, in some cases.

      Actually, another option you have, which I would like
      to see, is to get PDPCLIB working with IBM C. I used
      to have it working on IBM C/370 for MVS, so you will
      see some comments in the assembler mentioning that.
      But it will have stopped working for IBM C for MVS
      with all the changes that have been made, and it was
      never done for VM, so that won't work either.

      If you were able to get that to work, then you can
      use IBM's C compiler, but PDPCLIB for the runtime.
      That means that the source is available, can be
      modified, no licence restrictions, can be used in
      any environment, yet at the same time you get the
      benefits of a commercial C compiler.

      Let me know if you're interested in doing that.
      You could then try it out immediately as the CP
      exit you wanted.

      BTW, I mentioned with the exit that the assembler
      routines could be stubbed out. I think that is
      insufficient and wrong. It is the start.c that
      needs to be modified to bypass all that, call
      main() with special parameters pointing to a
      register save area or something for people to
      make use of the original registers. I'm interested
      in doing that work (setting up a Systems Programming
      Environment for VM) too.

      BFN. Paul.
    Your message has been successfully submitted and would be delivered to recipients shortly.