Re: [AMBER-Developers] MKL in updated configure scripts etc.

From: case <case.biomaps.rutgers.edu>
Date: Mon, 1 Mar 2010 20:10:40 -0500

On Mon, Mar 01, 2010, Scott Brozell wrote:
> >
> > I am trying to work out how MKL works these days with building sander. We
> > used to have -DMKL specified which would control the ifdef around veclib.f
> > but now it looks like we do not do this anymore. Thus from what I can see we
> > have:
>
> Based on browsing the configure* logs, my guess is that the loss of -DMKL
> is an oversight.

I agree. I haven't tracked it down, but I hope everyone will (continue to)
be careful in checking things in. This kind of thing usually happens when
someone copies a file from somewhere else into a CVS working directory. When
such files are checked in, they may easily blow away other commits. If you
keep your working files under CVS control, then CVS will force you to do an
update before committing.

Sometimes it's just mildly annoying and easily fixed, as in update 1.75
to amber11/src/configure. But other cases may go unnoticed for a while
and lead to mysterious errors later on. But -DMKL should be put back into
configure. Linux may allow duplicates, but some other systems will complain.

[This is only a mild admonishment -- I know that I've broken the build several
times recently, by not being careful enough.]

> Note also that we HAVE_MKL which appears to be redundant:

It should be changed. We need to do an ifdef audit soon, and try to minimize
their use.

...thanks....dac


_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Mon Mar 01 2010 - 17:30:02 PST
Custom Search