Re: Bug or feature?

["Randy Brukardt" <randy@rrsoftware.com>]

"Simon Wright" <simon@pushface.org> wrote in message 
news:ly8uq3mown.fsf@pushface.org...
> Robert A Duff <bobduff@shell01.TheWorld.com> writes:
>
>> Of those, only "-gnatE", "-gnato", and "-bargs -T0" are needed for
>> standards conformance.  And as I said, "-gnato" is no longer, or soon
>> will no longer be needed.
>
> -gnato isn't the default for GNAT GPL 2014 or FSF GCC 4.9.0.
>
> Add -fstack-check ?

That's probably a good idea for real programs. The ACATS doesn't need it as 
the tests that used to attempt to exhaust memory have been reigned in. (They 
had nasty effects on targets supporting virtual memory. I remember the first 
time we ran the ACATS on Windows NT, one of those tests allocated an insane 
amount of swap space and then essentially ran from the disk drive. It would 
have taken months to complete. I had to artifically bound the heap size on 
NT in order to eliminate that problem; that's the sort of counterproductive 
thing that one would hope the ACATS is not requiring.) The tests now 
allocate a few megabytes and then give up - thus on virtual memory hosts 
they don't try to test for Storage_Error.

                                           Randy.