[llvm] [AMDGPU] Optimize fcanonicalize/fneg/fsub with packed bf16 math ops (PR #197318)

[Matt Arsenault via llvm-commits]

https://github.com/arsenm requested changes to this pull request.

Patch content is still going beyond what the title says. You should separate out the min and max changes 

https://github.com/llvm/llvm-project/pull/197318