[llvm] [VPlan] Optimize more IV increment exit users by using a map (PR #147016)

Shih-Po Hung via llvm-commits llvm-commits at lists.llvm.org
Mon Jul 7 04:41:30 PDT 2025


arcbbb wrote:

> Could we detect them differently?

I see that plan0 now keeps track of the widen-phi and its backedge. Maybe there's a way to use that information to improve how we handle the IV increment at exit.

https://github.com/llvm/llvm-project/pull/147016


More information about the llvm-commits mailing list