[AArch64] Emit TBAA metadata for SVE load/store intrinsics

In Clang we can attach TBAA metadata based on the load/store intrinsics
based on the operation's element type.

This also contains changes to InstCombine where the AArch64-specific
intrinsics are transformed into generic LLVM load/store operations,
to ensure that all metadata is transferred to the new instruction.

There will be some further work after this patch to also emit TBAA
metadata for SVE's gather/scatter- and struct load/store intrinsics.

Reviewed By: paulwalker-arm

Differential Revision: https://reviews.llvm.org/D119319

Author: sdesmalen-arm

clang/lib/CodeGen/CGBuiltin.cpp

@@ -9028,7 +9028,10 @@ Value *CodeGenFunction::EmitSVEMaskedLoad(const CallExpr *E,
 
   BasePtr = Builder.CreateBitCast(BasePtr, MemEltTy->getPointerTo());
   Function *F = CGM.getIntrinsic(BuiltinID, MemoryTy);
-  Value *Load = Builder.CreateCall(F, {Predicate, BasePtr});
+  auto *Load =
+      cast<llvm::Instruction>(Builder.CreateCall(F, {Predicate, BasePtr}));
+  auto TBAAInfo = CGM.getTBAAAccessInfo(LangPTy->getPointeeType());
+  CGM.DecorateInstructionWithTBAA(Load, TBAAInfo);
 
   return IsZExtReturn ? Builder.CreateZExt(Load, VectorTy)
                      : Builder.CreateSExt(Load, VectorTy);
@@ -9056,7 +9059,11 @@ Value *CodeGenFunction::EmitSVEMaskedStore(const CallExpr *E,
 
   BasePtr = Builder.CreateBitCast(BasePtr, MemEltTy->getPointerTo());
   Function *F = CGM.getIntrinsic(BuiltinID, MemoryTy);
-  return Builder.CreateCall(F, {Val, Predicate, BasePtr});
+  auto *Store =
+      cast<llvm::Instruction>(Builder.CreateCall(F, {Val, Predicate, BasePtr}));
+  auto TBAAInfo = CGM.getTBAAAccessInfo(LangPTy->getPointeeType());
+  CGM.DecorateInstructionWithTBAA(Store, TBAAInfo);
+  return Store;
 }
 
 // Limit the usage of scalable llvm IR generated by the ACLE by using the