Re: Testing strategy for compiler

George Neuner <gneuner2@comcast.net> wrote:
(snip)


> But John is correct that this won't test semantics ...
(snip)


> I once used a Pascal compiler that happily accepted garbage like


> for n := 1 to 10 by -1
> do ...


> and generated an unreachable loop body. I forget why the optimizer
> didn't just remove the loop ... probably the implementation used a
> hidden loop counter to guarantee FOR iteration. What the compiler
> should have done was reject the invalid step count.


Is that illegal in PASCAL? I haven't thought about PASCAL recently,
but I don't believe that such a loop is illegal in the other languages
with a similar loop construct.


It might be that unreachable statements without a statement number
are illegal in Fortran:


            GOTO 123
            PRINT *,'can''t get here!'


but otherwise it is usually the optimizer that complains.


I do remember (from a long time ago, well, early in my Fortran days)
that the IBM S/360 Fortran G compiler gives a level 0 message
for unreachable statements, where Fortran H gives a level 4.


In the usual terminology, level 4 is warning, level 8 is error,
level 12 is fatal error, such that level 0 isn't even warning!


The Fortran H optimizer needs to keep track of program flow for
optimization reasons, but Fortran G doesn't.


-- glen

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