[llvm-dev] [SVE][AArch64] Codegen for a scalable vector splat

[Renato Golin via llvm-dev]

On Thu, 29 Aug 2019 at 14:24, Graham Hunter <Graham.Hunter at arm.com> wrote:
> If we introduce the new node, I certainly expect it to be used for
> fixed width vectors as well (though possibly not right away). Is there
> anything special about predication in this use case, or is a simple
> splat sufficient?

There are many patterns that predication can do to constant splats,
but I was just thinking of implementing the tail loop in the same for
predicated fixed width as we do with SVE. So a simple splat would work
for now.

--renato