Re: [boost] Accelerating algorithms with SIMD - Segmented iterators and alternatives

Subject: Re: [boost] Accelerating algorithms with SIMD - Segmented iterators and alternatives
From: Sebastian Redl (sebastian.redl_at_[hidden])
Date: 2010-10-19 02:41:58

On 19.10.2010, at 02:15, David Abrahams wrote:

> I don't understand what you're talking about in that section above,
> but I think you're making things too complicated because you're
> fixated on the idea of a heterogeneous sequence.  Here's a simple
> extension to the hierarchical algorithm formulation that I think could
> accomodate SIMD.  The key change is the segment_fill function, which
> gives us an additional customization point with which to capture SIMD
> segments

Oh, that's nice.

Sebastian