[m-dev.] Bug on 32bit systems.

[Julien Fischer]

Hi,

On Fri, Jan 4, 2013 at 2:20 PM, Paul Bone <paul at bone.id.au> wrote:
> On Fri, Jan 04, 2013 at 02:01:20PM +1100, Julien Fischer wrote:
>> On Fri, Jan 4, 2013 at 1:53 PM, Paul Bone <paul at bone.id.au> wrote:
>>
>> I've compiled versions of G12 with both the broken and working compilers -- this
>> is all in grade hlc.gc.trseg -- and the C files we generate for G12
>> are identical for
>> both compilers.  As are the C files that both Mercury compilers generate for the
>> standard library.
>> Both Mercury compilers are invoking the C compiler identically.
>>
>
> That all seems to 'narrow down' to some environment variable or C macro
> differences.

I could compare the preprocessed C files.

> Alternatively we'd need to compare the contents of the object
> files without comparing the object files' timestamps and other unstable
> information.
>
> Is there any way I could get/reproduce the same test case so I could also
> test this.  Ie, I wonder if it happens on Linux.  (MC do use Windows so I
> need to support that too).

Not without a copy of CPLEX and Opturion's version of G12.
I am trying to find a better test case (i.e. one that doesn't require
either of those things)
but haven't yet been able to come up with one.
(At some point I'm going to try this out on 32-bit Linux in order to
see if the same
thing occurs.)

One thing that should be done is building the compiler as in my broken version,
and then running all the tests using that compiler.  (In at least one
low-level and one
high-level C grade.)

Cheers,
Julien.