Re: [petsc-dev] MetSetBlockSize issue
On Wed, Apr 25, 2012 at 4:58 PM, Jed Brown <[email protected]> wrote:
On Wed, Apr 25, 2012 at 16:41, Barry Smith <[email protected]> wrote:
I admit MatSetVirtualColumnBlockSize() is clunky and would be happy to hear alternatives.
What about having a separate row and column block size? The BAIJ and MAIJ formats can continue to store data using bs=gcd(rowbs,colbs), but those row and column block sizes indicate that redundancy exists in the data structure, so it can be utilized by appropriate operations.
I agree. In fact, P's "column block size" (cbs) has exactly that meaning in the GAMG code (e.g., formProl0() in gamg/agg.c) -- it sets matrix values in adjacent multiples of cbs within each row. MatPtAP() could easily impart that block size (cbs = rbs, in this case) to its result. This way the meaning of "column block size" and "row block size" are meaningful for any matrix and not just within some particular context. Dmitry.
participants (1)
-
Dmitry Karpeev