[llvm] [RISCV] Lower unmasked zero-stride vp.stride to a splat of one scalar load. (PR #97394)

[Pengcheng Wang via llvm-commits]

wangpc-pp wrote:

> I guess we could handle the case for constant EVLs. Are they common in practice?

I think we can't know it because we don't have any **practices** since almost no VP intrinsic will be generated currently.

https://github.com/llvm/llvm-project/pull/97394