dolfin team mailing list archive

[Johan Jansson <johanjan@xxxxxxxxxxxxxxxx>]

On Tue, Oct 04, 2005 at 06:56:36PM -0500, Anders Logg wrote:
> I have verified that the new FFC BLAS mode works for the stabilized
> convection-diffusion module in DOLFIN (which is sufficiently complex
> to catch any obvious bugs).
> 

Great job. I'll try it out with the elasticity forms as soon as I have
time.

> I still need to run some more extensive benchmarks, but the first
> indications seem to be that the new BLAS mode is actually slower and
> the generated code actually takes longer to compile. Possible
> explanations:
> 
> 1. For the convection-diffusion form, most of the work is computing
> the geometry tensor and that is performed in the same way in both
> cases. Maybe we could generate some loop-based code for computing the
> geometry tensor also.
> 
> 2. Assigning to array entries (G[i] = ... in BLAS mode) seems to take
> longer to compile than assigning to variables (double Gi = ... in
> default mode).
> 
> 3. It actually matters that the default mode of FFC removes any
> products with zeros in the tensor product. BLAS does not know about
> zeros.

This seems to be a severe limitation. For PDE systems with Lagrange
elements (the typical case I guess), there will be lots of zeros. From
previous discussions about FFC/Ferari, the conclusion was that
skipping the zero elements was the dominant optimization for computing
the element matrix.

  Johan