[Cplex] Short SIMD vector types

Nelson, Clark clark.nelson at intel.com
Tue Jun 4 02:03:23 CEST 2013


> > A big issue with that approach though is that it is naturally limiting
> > and verbose when it comes to varying vector widths.  Is it good to
> have
> > int2,int4,int8,int16... ad nauseum? Perhaps it is, but it may be worth
> > considering defining a type with more generic semantics as well to
> allow
> > the writing of code to "native SIMD width" rather than specific
> widths,
> > especially given the proliferation of available widths in newer
> > processors/accelerators.
> 
>       +1.  The fixed-width approach requires writing nonportable code.

Portability is somewhat in the eye of the beholder -- especially considering that we're contemplating doing something in the standard arena. But OpenCL code, for example, is inherently non-scalable, unlike array sections, as from Cilk Plus.

Clark


More information about the Cplex mailing list