jeff f4e6149d82 [AMDGPU] Use V_PERM to match buildvectors when inputs are not canonicalized (i.e. can't use V_PACK) ...

If we can not prove that f16 operands of a buildvector are canonicalized, then we can not lower into a V_PACK. In this scenario, we would previously lower into some combination of and(sdwa), shr, or. This patch allows for matching into V_PERM instead.

Change-Id: Ifa4a74fdb81ef44f22ba490c7fdf81ec8aebc945

2022-10-03 12:58:29 -07:00

8.4 KiB

Raw Blame History

View Raw