[petsc-dev] Fwd: [petsc-maint #69763] PETSc and OSKI

[Aron Ahmadia]

I would bet a beer that Jed's Intel intrinsics beat the pants off of
anything OSKI has to offer at this point.

A

On Fri, Apr 8, 2011 at 7:50 AM, Barry Smith <bsmith at mcs.anl.gov> wrote:

>
>  Who put this in and does it do anything?
>
>   Barry
>
>
> Begin forwarded message:
>
> > From: Kadir Akbudak <kadir at cs.bilkent.edu.tr>
> > Date: April 8, 2011 6:42:40 AM CDT
> > To: petsc-maint at mcs.anl.gov
> > Subject: [petsc-maint #69763] PETSc and OSKI
> > Reply-To: petsc-maint at mcs.anl.gov, Kadir Akbudak <
> kadir at cs.bilkent.edu.tr>
> >
> > Dear Sir or Madam,
> >
> > There is such a configuration script,
> > petsc-3.1-p8/config/PETSc/packages/oski.py, in the PETSc distribution.
> > Does this mean that PETSc uses OSKI sparse matrix-vector routines? If
> > yes, is it possible to use shared libraries? It is written as:
> > self.liblist   = [['oski/liboski_Tid.a','oski/liboski.a']] This means
> > the libraries are static.
> >
> > In fact I want to use benefits of OSKI together with my sparse matrix
> > ordering tool, in a numerical method which can be performed by PETSc.
> >
> > Yours Faithfully,
> >
> > Kadir
> >
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20110408/6c6ad525/attachment.html>