[petsc-dev] Maintaining row alignment in matrices (especially Inode and odd-sized (S)BAIJ)

Jed Brown jed at 59A2.org
Wed Oct 6 20:00:49 CDT 2010

Alright, same model as MATCRL and CSRPERM. Would be nice to eventually avoid
the copy, but there's no wasted code to do that latter if the performance is
clearly worth it.


On Oct 7, 2010 2:50 AM, "Barry Smith" <bsmith at mcs.anl.gov> wrote:

  Make a whole new subclass of SeqAIJ (parallel to the Inode) that does all
this cool stuff and copies into the new aligned data structures (rather than
keeping the data in the same data structure (as the current inode does).
 We'll just have to get the factorization stuff to work eventually once you
show good performance gain for MatMult_SeqAIJ_AlignedInode().


On Oct 6, 2010, at 7:04 PM, Jed Brown wrote:

> Looking at assembly generated from the Inode kernel...
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20101007/ded43cf5/attachment.html>

More information about the petsc-dev mailing list