[llvm] [AMDGPU] Support TFE D16 format buffer loads on gfx8/10/11/12 (PR #211465)

[Arseniy Obolenskiy via llvm-commits]

================
@@ -6898,7 +6908,58 @@ bool AMDGPULegalizerInfo::legalizeBufferLoad(MachineInstr &MI,
     }
   }
 
-  if (IsTFE) {
+  if (IsTFE && IsD16 && Ty.isVector()) {
----------------
aobolensk wrote:

Yes, renamed the PR

https://github.com/llvm/llvm-project/pull/211465