Newsgroups: | comp.compilers |
From: | moss@cs.umass.edu (Eliot Moss) |
References: | <1990Jun1.194941.5781@esegue.segue.boston.ma.us> <1990Jun4.212226.18389@esegue.segue.boston.ma.us> <1990Jun7.010349.2097@esegue.segue.boston.ma.us> <1990Jun7.190315.4578@esegue.segue.boston.ma.us> |
Date: | Sun, 10 Jun 90 22:05:35 GMT |
Organization: | Dept of Comp and Info Sci, Univ of Mass (Amherst) |
Keywords: | optimize |
In-Reply-To: | tli@%phakt.usc.edu's message of 7 Jun 90 19:03:15 GMT |
I would like to note that a number of languages are standardized such that
order of evaluation of certain items is *not specified*, and it is legal for
the compiler to choose any order it likes, non-associativity, etc.,
notwithstanding. Frequently optimizers "assume associativity" under this
rubric. If the programmer has a program where she cares about the difference,
then it is up to her to force evaluation order somehow. (I'm not saying I
agree or disagree with this; I'm merely stating it as a fact of life ...)
--
J. Eliot B. Moss, Assistant Professor
Department of Computer and Information Science
Lederle Graduate Research Center
University of Massachusetts
Amherst, MA 01003
(413) 545-4206; Moss@cs.umass.edu
[Indeed, such ambiguity is more the rule than the exception. Fortran and C
are both standardized that way. -John]
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.