Re: non trivial constant folding

Anton Ertl <anton@mips.complang.tuwien.ac.at> writes
> "Mihai Christodorescu" <mihai@cs.wisc.edu> writes:
>> There is also another problem: the optimized expression can behave
>>differently than the original expression, due to operations being executed
>>in different order/with different operands.
>
>And even if the original code had undefined behaviour, a particular
>compiler should implement a consistent behaviour for that, if possible
>(for practical reasons, like finding the bug).


I have to disagree with this. One of the points of specifying
'undefined' behaviour for a construct is to give latitude to the
implementors to assume that such a case can be ignored when
considering how to optimize.


Hence, for the following (undefined) C code, I would certainly assume
that an optimised and non-optimised version could have different
behaviours:


                i = i++; /* undefined behaviour */


--
Conor O'Neill, Bristol, UK

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