[llvm] [AMDGPU] Fix canonicalization of truncated values. (PR #83054)

[Matt Arsenault via llvm-commits]

================
@@ -12574,6 +12595,9 @@ bool SITargetLowering::isCanonicalized(SelectionDAG &DAG, SDValue Op,
     return false;
 
   case ISD::BITCAST:
+    // TODO: This is incorrect as it loses track of the operand's type. We may
+    // end up effectively bitcasting from f32 to v2f16 or vice versa, and the
+    // same bits that are canonicalized in one type need not be in the other.
----------------
arsenm wrote:

should give up on different element size bitcasts? (separate patch though) 

https://github.com/llvm/llvm-project/pull/83054