[petsc-dev] configure issues

Matthew Knepley knepley at gmail.com
Wed Feb 9 03:26:15 CST 2011

On Wed, Feb 9, 2011 at 3:13 AM, Jed Brown <jed at 59a2.org> wrote:

> On Wed, Feb 9, 2011 at 08:04, Matthew Knepley <knepley at gmail.com> wrote:
>> I can't believe you guys want further screwing around with the C
>> preprocessor. I want all that crap completely
>> out of the system. We are very close to that.
> If you think we are very close to removing all functional tests from
> configure (making petscconf.h contain only options set directly by the user,
> not the result of any tests in configure), then you are delusional. Barry
> just wants a clean way to jump from macro name to the test that defines that
> macro and back. I agree that manually following the "link" is sometimes a
> confusing process as evidenced by the fact that I typically use recursive
> grep for it, but I would rather make the macro definition process
> sufficiently well structured than to write potentially fragile paths into
> the variable itself. This is about naming and links, not about "screwing
> around with the C preprocessor".

1) This is about fooling with the preprocessor, or I would not need any of
these goofy names. How can you deny this?

2) For test information as well as control flow, I want to stay in Python.

3) What do you mean by "link"?


What most experimenters take for granted before they begin their experiments
is infinitely more interesting than any results to which their experiments
-- Norbert Wiener
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20110209/d4196be3/attachment.html>

More information about the petsc-dev mailing list