Re: Re: C++ virtual function calls

  David L Moore wrote:
  > (I threw in y so you can work on dynamic typing algorithms while
  > pretending to do optimization - if you only want to support single
  > dispatching you need to prove y is an x or else not pull that inline)
  >
  > I have not seen any discussion of this in the literature. Indeed, I suspect
  > optimization of C++ is still a largely untapped field.


    It is perhaps true for C++, but it certainly doesn't hold for more
  advanced (should I say real :-) ?) OO languages.


    You should have a look at the research done on Self and Cecil:
  (and perhaps on Dylan at CMU in the following years)


  http://www.cs.ucsb.edu/oocsb/papers/tf-vs-ti.html
  http://www.cs.ucsb.edu/oocsb/papers/type-feedback.html


  http://www.cs.washington.edu/research/projects/cecil/www/papers.html


    Very interesting sites for high-level optimizations techniques
  (as opposed to assembly level optimizations).


    Eric Gouriou
    egouriou@cs.ucla.edu








--

Return to the comp.compilers page.
Search the comp.compilers archives again.