Re: [AMBER-Developers] SSE2 / SSE3 settings no longer used for ifort?

From: Ben Roberts <roberts.qtp.ufl.edu>
Date: Fri, 23 Oct 2009 20:05:53 -0400

On Fri, 2009-10-23 at 16:45 -0700, Ross Walker wrote:
> Hi All,
>
> It seems that the config.h prepared by ./configure is no longer setting the
> SSE2 / SSE3 optimization flags for the Intel compiler. This can reduce
> performance by >30% in some cases. Does anyone know why this was turned off?
>
> Specifically this is things like:
>
> ifort -axWPS
>
> The problem is that different versions of the Intel compilers use different
> sets of flags. Once again an example of the patients running the asylum.
> However, we could set -fast as the default option which would turn on:
>
> -xP -O3 -ipo
>
> The ipo is a problem since it means all the optimization is done at link
> stage and you lose all the benefits of parallel builds and each new make is
> very expensive. We could specify
>
> -axWPS
>
> but a lot of this is now changed and deprecated in ifort >11.0
>
> Hence should we have an intel9 and intel10 target which uses -O3 -ip -axWPS
> and an Intel11 target which uses -O3 -ip -axSSE4.2,SSE4.1,SSSE3,SSE3,SSE2
>
> Comments?

Hi Ross,

My immediate question (perhaps influenced by my recent development
efforts on the Mac) is: How well supported are these new flags between
platforms?

Cheers,
Ben


_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Fri Oct 23 2009 - 17:30:02 PDT
Custom Search