[AMDGPU][GlobalISel] Properly handle lane op legalization for larger vector types

Author: vikramRH

llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp

@@ -5565,6 +5565,7 @@ bool AMDGPULegalizerInfo::legalizeLaneOp(LegalizerHelper &Helper,
     return false;
 
   LLT PartialResTy = LLT::scalar(SplitSize);
+  bool NeedsBitcast = false;
   if (Ty.isVector()) {
     LLT EltTy = Ty.getElementType();
     unsigned EltSize = EltTy.getSizeInBits();
@@ -5573,8 +5574,10 @@ bool AMDGPULegalizerInfo::legalizeLaneOp(LegalizerHelper &Helper,
     } else if (EltSize == 16 || EltSize == 32) {
       unsigned NElem = SplitSize / EltSize;
       PartialResTy = Ty.changeElementCount(ElementCount::getFixed(NElem));
+    } else {
+      // Handle all other cases via S32/S64 pieces
+      NeedsBitcast = true;
     }
-    // Handle all other cases via S32/S64 pieces;
   }
 
   SmallVector<Register, 4> PartialRes;
@@ -5600,7 +5603,12 @@ bool AMDGPULegalizerInfo::legalizeLaneOp(LegalizerHelper &Helper,
     PartialRes.push_back(createLaneOp(Src0, Src1, Src2, PartialResTy));
   }
 
-  B.buildMergeLikeInstr(DstReg, PartialRes);
+  if (NeedsBitcast)
+    B.buildBitcast(DstReg, B.buildMergeLikeInstr(
+                               LLT::scalar(Ty.getSizeInBits()), PartialRes));
+  else
+    B.buildMergeLikeInstr(DstReg, PartialRes);
+
   MI.eraseFromParent();
   return true;
 }