[llvm] [RISCV] Vectorize phi for loop carried @llvm.vp.reduce.* (PR #131974)

via llvm-commits llvm-commits at lists.llvm.org
Tue Mar 18 22:56:19 PDT 2025


NexMing wrote:

> The original patch only supported ordered fadd. Why does this to support more than just vp.reduce.fadd?



> > The original patch only supported ordered fadd. Why does this to support more than just vp.reduce.fadd?
> 
> Oh I guess its because of the scalar input that the regular reduce intrinsics don't have?

That's right.  I think this optimization will not change the order of operations.

https://github.com/llvm/llvm-project/pull/131974


More information about the llvm-commits mailing list