Re: Order of argument evaluation in C++, etc.

jthill@netcom.com (Jim Hill) writes:


      Unless I'm very much mistaken, alias identification is an insoluble
      problem, and flagging order dependencies in expressions runs straight into
      that wall.


Well, I guess it's insoluble in general (no my area of expertise), but it is
certainly possible and useful. It's clear that this is really what we're
talking about.


      To digress for a paragraph, I'm all for better tools to tag possible
      problems, but I don't know of any commercial compiler that gets even the
      simplest cases.


Well, then you don't know enough 8-)...have a look at the Inmos occam2
compilers. They have some quite complicated alias checking (helped by language
rules to avoid those nasty interprocedural/global-variable problems) which,
then, is used to assist you in avoiding/diagnosing those nasty multiple-use
problems in a parallel programme. And that in a language that allows you to
specify, at runtime, aliases for parts of your data...


Jan


--

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