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

jan@neuroinformatik.ruhr-uni-bochum.de (Jan Vorbrueggen)
Mon, 21 Aug 1995 15:59:25 GMT

          From comp.compilers

Related articles
[29 earlier articles]
Re: Order of argument evaluation in C++, etc. msb@sq.com (1995-08-18)
Re: Order of argument evaluation in C++, etc. ka@socrates.hr.att.com (1995-08-19)
Re: Order of argument evaluation in C++, etc. hbaker@netcom.com (1995-08-21)
Re: Order of argument evaluation in C++, etc. chase@centerline.com (1995-08-21)
Re: Order of argument evaluation in C++, etc. chase@centerline.com (1995-08-21)
Re: Order of argument evaluation in C++, etc. bobduff@world.std.com (1995-08-21)
Re: Order of argument evaluation in C++, etc. jan@neuroinformatik.ruhr-uni-bochum.de (1995-08-21)
Re: Order of argument evaluation in C++, etc. bill@amber.ssd.hcsc.com (1995-08-21)
Re: Order of argument evaluation in C++, etc. burley@gnu.ai.mit.edu (1995-08-23)
Re: Order of argument evaluation in C++, etc. jthill@netcom.com (1995-08-24)
Re: Order of argument evaluation in C++, etc. hbaker@netcom.com (1995-08-23)
Re: Order of argument evaluation in C++, etc. way@cis.udel.edu (Thomas Way) (1995-08-23)
Re: Order of argument evaluation in C++, etc. jmccarty@spdmail.spd.dsccc.com (1995-08-24)
[7 later articles]
| List of all articles for this month |

Newsgroups: comp.compilers
From: jan@neuroinformatik.ruhr-uni-bochum.de (Jan Vorbrueggen)
Keywords: C++, optimize
Organization: Institut fuer Neuroinformatik, Ruhr-Universitaet Bochum, Germany
References: 95-07-068 95-08-115
Date: Mon, 21 Aug 1995 15:59:25 GMT

bobduff@world.std.com (Robert A Duff) writes:


      Unfortunately, there are still some "stylistic" rules left -- for example,
      it's still illegal for a function to have an 'out' parameter, which is very
      annoying.


Wow. Even occam2, which is "holier-than-thou" for almost anybody, allows
that. (Actually, a function can return any _list_ of scalar values, due to the
multiple assignment possible. Now, people also want to return fixed-size
arrays...)


      I wonder if it's possible to design the language so that the compiler has a
      better idea of exactly which sub-expressions have side-effects, and what
      those side effects are. (Ada and C++ pretty much require the compiler to
      assume the worst, for a call to a separately compiled function.)


Occam2 allows this, as any function _must_ be side-effect free. This is
checked by the compiler (only IN arguments, results as above, no I/O
statements allowed, no procedure calls). Thus, an occam2 compiler can assume
the best.


Jan


--


Post a followup to this message

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