[llvm] [AMDGPU]: Unpack packed instructions overlapped by MFMAs post-RA scheduling (PR #157968)

[Jeffrey Byrnes via llvm-commits]

https://github.com/jrbyrnes approved this pull request.

LGTM --

There are still a few important TODOs -- namely, we should combine the loops, and look into supporting more opcodes. Also, there are several code cleanup / minor optimizations I can nitpick on, but I don't think we should be blocking the PR for those things.

https://github.com/llvm/llvm-project/pull/157968