AMDGPU: Remove unnecessary pointer bitcast

Author: arsenm

llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp

@@ -1033,16 +1033,14 @@ AMDGPUPromoteAllocaImpl::getLocalSizeYZ(IRBuilder<> &Builder) {
   DispatchPtr->addDereferenceableRetAttr(64);
 
   Type *I32Ty = Type::getInt32Ty(Mod->getContext());
-  Value *CastDispatchPtr = Builder.CreateBitCast(
-      DispatchPtr, PointerType::get(I32Ty, AMDGPUAS::CONSTANT_ADDRESS));
 
   // We could do a single 64-bit load here, but it's likely that the basic
   // 32-bit and extract sequence is already present, and it is probably easier
   // to CSE this. The loads should be mergeable later anyway.
-  Value *GEPXY = Builder.CreateConstInBoundsGEP1_64(I32Ty, CastDispatchPtr, 1);
+  Value *GEPXY = Builder.CreateConstInBoundsGEP1_64(I32Ty, DispatchPtr, 1);
   LoadInst *LoadXY = Builder.CreateAlignedLoad(I32Ty, GEPXY, Align(4));
 
-  Value *GEPZU = Builder.CreateConstInBoundsGEP1_64(I32Ty, CastDispatchPtr, 2);
+  Value *GEPZU = Builder.CreateConstInBoundsGEP1_64(I32Ty, DispatchPtr, 2);
   LoadInst *LoadZU = Builder.CreateAlignedLoad(I32Ty, GEPZU, Align(4));
 
   MDNode *MD = MDNode::get(Mod->getContext(), std::nullopt);