[petsc-dev] CUBLAS interface for sequential dense matrices

Matthew Knepley knepley at gmail.com
Sun Nov 6 08:46:49 CST 2011


On Sun, Nov 6, 2011 at 1:15 PM, Stefano Zampini
<stefano.zampini at gmail.com>wrote:

> Hi,
>
> I'm developing a code for a BDDC preconditioner in PETSc.
> Since I'm interested in using sequential dense matrices for GPU runs of my
> code, I wish to know when an interface to CUBLAS will be available for
> petsc-dev version of the software.
>

CUBLAS is already used in other parts of the code, but we had no plans to
use it for dense matrices, because we have little time
and many projects, and not much call for dense things. If you want to
follow the example of aijseqcusp and make desneseqcuda
that would be fine.

  Thanks,

     Matt


> Regards,
> --
> Stefano
>



-- 
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/20111106/dd90c911/attachment.html>


More information about the petsc-dev mailing list