[PATCH] normal_distribution<double> performance improvement with SSE

[Ulrich Drepper]

On Wed, Sep 26, 2012 at 7:32 AM, Jakub Jelinek <jakub@redhat.com> wrote:
> Have you considered also an __AVX__ version handling 4 elements at a time?
> Without __AVX2__ one would need to cast __m256i to __m256d for and/or, as
> AVX1 doesn't have _mm256_and_si256 or _mm256_or_si256, but _mm256_and_pd
> or _mm256_or_pd could be used instead.

One step to do first.  Currently the random number engine interface is
inefficient since it returns a single number.  What we need is an
additional interface to return vectors.  I'd love to use the gcc
vector extensions.  For engines like sfmt this is natural.  There are
a few issues with C++ support for the vector extensions.  Operations
available in C are not supported in C++ yet.