[petsc-dev] Why does configure ignore compiler output?

[Matthew Knepley]

On Mon, Mar 18, 2013 at 9:22 PM, Jed Brown <jedbrown at mcs.anl.gov> wrote:

>
> On Mon, Mar 18, 2013 at 8:10 PM, Matthew Knepley <knepley at gmail.com>wrote:
>
>> It was before that commit, as you can see that I just use the 1 that was
>> already there.
>>
>
> Yeah, I see you just declared the default differently there.
>
> It's also amusing that your name is SpikeGPU according to bitbucket. ;-)
>
>
>> I think it was that compilers too often spewed a bunch of meaningless
>> crap.
>>
>
> Why have the feature in the first place if it's never turned on? We could
> delete a bunch of code if we just said "compiler output is always ignored",
> which is effectively what we have here. None of the nightlies even flip
> that default.
>
> And, more importantly right now, how am I supposed to check for compiler
> support in a way that interprets the following as non-support?
>
> "/tmp/petsc-NZgbtU/PETSc.Configure/conftest.c", line 3: warning:
>  attribute "unused" is unknown, ignored
>

You can always turn that switch off for your test. What we were finding is
that some crappy compilers
were killing everything by putting in meaningless output for every compile.
We don't rip it out because
maybe those compilers die and we flip the switch in the new release.

   Matt

-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20130318/a2d74bcc/attachment.html>