[PATCH] D75138: [WIP][AMDGPU] Eliminate the ScratchWaveOffset register from the calling convention

[Scott Linder via Phabricator via llvm-commits]

scott.linder added a comment.

I'm going to go ahead with trying to eliminate the need for an FP completely in entry functions and then update the review with a more complete set of test updates. I'm sure the issue I was having with defining and using `hasFP` consistently between ISel and PEI could be worked around, but putting that effort into eliminating the FP entirely in entry functions seems more productive.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D75138/new/

https://reviews.llvm.org/D75138