[llvm-branch-commits] [llvm] [LV] Vectorize down-counting floating-point argmin/argmax reductions (PR #209827)

[Florian Hahn via llvm-branch-commits]

https://github.com/fhahn commented:

It looks like the PR contains a few independent features, which would be good to submit separately for review, e.g. starting with support for counting down inductions?

That would also make it easier to make sure all code paths are covered by tests.

For FP support, there's https://github.com/llvm/llvm-project/pull/198300

https://github.com/llvm/llvm-project/pull/209827