[Discuss] Test for positive gives wacko results for underflowed values for the gcc compiler suite (including the gcc compiler, and g77 and gfortran compilers)

[David Bronaugh]

Alan W. Irwin wrote:
> On 2007-06-25 16:14-0700 David Bronaugh wrote:
>
>> Have you searched the GCC bug database? It might be a good idea at 
>> this point.
>
> Hi David:
>
> I just now tried a search for "underflow" at
> http://gcc.gnu.org/bugzilla/query.cgi and got 8 hits, but none of them 
> seem
> relevant.
>
> Normally, I have the confidence to make bug reports, but I must say in 
> this
> case I am somewhat intimidated since they already have a case about
> floating-point representations issues (see 
> http://gcc.gnu.org/bugs.html#nonbugs_general), my gcc compiler is 
> pretty out
> of date (I don't have the latest Ubuntu), and besides that I know very
> little about compilers.
>
> David, if you confirm the result for my latest simple C example on your
> latest/greatest gcc and think it really must be a gcc bug, do you have 
> the
> confidence/knowledge to report the problem yourself?
I'll have to thoroughly examine the code and ask questions first, make 
sure that I know exactly what's going on. After that, yes, I'd be 
willing to make the report.

David