[llvm] [SystemZ] Use scheduling model for load latency (PR #217617)

[via llvm-commits]

MarkVeerasingam wrote:

@uweigand @JonPsson1 Thanks for the review and benchmarking.

I agree that selectively using the scheduling model for TCK_Latency isn't justified without a concrete optimization that benefits from it. 

Since the current results are essentially neutral, I'll drop this approach for now. I think this PR is in a good state to be closed. If something comes up where we need this, I'm happy to revisit it!

https://github.com/llvm/llvm-project/pull/217617