[llvm] [AMDGPU] select v_sat_pk from two i16 or v2i16 (PR #121124)

[Matt Arsenault via llvm-commits]

https://github.com/arsenm commented:

I think the approach of using these patterns should be revisited. These patterns are unwieldy and duplicate generic combiner logic.

We should be using the generic TRUNCATE_*SAT* nodes. The only issue is that we don't want to make v2i8 legal, but we do not have to. We can custom lower these nodes on the illegal v2i8 type, use a target specific node and bitcast from the packed-as-i16 form of the instruction to the v2i8 



https://github.com/llvm/llvm-project/pull/121124