[ARM] Take into account address spaces in interleaved access vectorization

[jketema]

Summary

The vector being loaded might not be in address space 0. In this case the vldn/vstn call creation will trigger an assert in CallInst:Create, because the argument type and the parameter are in different address spaces. Resolve this by adding an appropriate address space cast.

[sbaranga]

lib/Target/ARM/ARMISelLowering.cpp

11708 ↗

(On Diff #35137)

This is interesting.. We ideally wouldn't want to introduce address space casts when not necessary (I can imagine some users using the address space qualifier to help with alias analysis).

We should be able to get a variant of the vldn intrinsic that operates on the correct address space.

[jketema]

lib/Target/ARM/ARMISelLowering.cpp

11708 ↗

(On Diff #35137)

Maybe. I would need some pointers on how to change the code in that case, because this would require changes to IntrinsicsARM.td, which I'm not very familiar with.

I'm also wondering: is alias analysis information still being used during instruction lowering or later?

[sbaranga]

lib/Target/ARM/ARMISelLowering.cpp

11708 ↗

(On Diff #35137)

I've looked into this and changing IntrinisicsARM.td should be fairly simple, all that's needed is to replace in definitions like int_arm_neon_vld2 the llvm_ptr_ty argument with a llvm_anyptr_ty.

However it turns out that this changes the intrinisic name mangling (which might be a problem with backward compatibility) and also requires clang changes. This is of course a bit complicated..

[jketema]

lib/Target/ARM/ARMISelLowering.cpp

11708 ↗

(On Diff #35137)

Thanks for the pointer regarding llvm_anyptr_ty.

Maybe map, say, __builtin_neon_vld3_v to @llvm.arm.neon.vld3.v2i32.p0i8 instead of @llvm.arm.neon.vld3.v2i32 and keep that as the only exposed version? Or are there objections to that solution too?

[sbaranga]

lib/Target/ARM/ARMISelLowering.cpp

11708 ↗

(On Diff #35137)

Yes, we would have to do that and also add auto-upgrade support for the old forms of the intrinsics. Perhaps an email should be sent to llvm-dev as well.

[jketema]

lib/Target/ARM/ARMISelLowering.cpp

11708 ↗

(On Diff #35137)

I'm not sure what you're saying. Sticking with vld3 Do you mean you both want to expose @llvm.arm.neon.vld3.v2i32.p0i8 and @llvm.arm.neon.vld3.v2i32? Or do you want more exposed?

I'm also not quite sure what you mean by auto-upgrade support?

[sbaranga]

lib/Target/ARM/ARMISelLowering.cpp

11708 ↗

(On Diff #35137)

I think we should make the change to IntrinsicARM.td (replace llvm_ptr_ty with llvm_anyptr_ty for the vldN intrinsics). At that point we will only have the new style intrinsics in llvm IR (like @llvm.arm.neon.vld3.v2i32.p0i8) and the old ones ( like @llvm.arm.neon.vld3.v2i32) will not be supported anymore.

We therefore need to add a case in lib/IR/AutoUpgrade.cpp for the old style vldN intrinsics so that we can convert old style intrinsics to the new one and not break backwards compatibility.

Regarding the clang changes (__builtin_neon_vld*..), yes, these should be emitting @llvm.arm.neon.vld3.v2i32.p0i8.

[jketema]

lib/Target/ARM/ARMISelLowering.cpp

11708 ↗

(On Diff #35137)

Got it, thanks for the explanation. I was not aware of lib/IR/AutoUpgrade.cpp.

I'll update this patch to take your comments into account and whip up an associated patch for clang. Once that's done I'll also send a message to llvm-dev to ask for more feedback.

[jketema]

This now upgrades the vld[234] and vst[234] intrinsics to take into account an address space, as discussed.

Open issues as far as I'm concerned:

• Should vld1 and vst1 also be upgraded for uniformity? Similarly for the vld[234]lane and vst[234]lane instructions?
• Should all the tests referring to vld[234] and vst[234] be updated to use the new intrinsic names, or are they allowed to depend on the auto-upgrade code?

[sbaranga]

Thanks, Jeroen!

In D12985#252574, @jketema wrote:

This now upgrades the vld[234] and vst[234] intrinsics to take into account an address space, as discussed.

Open issues as far as I'm concerned:

• Should vld1 and vst1 also be upgraded for uniformity? Similarly for the vld[234]lane and vst[234]lane instructions?

They seem to have the same issues, so I would say yes.

• Should all the tests referring to vld[234] and vst[234] be updated to use the new intrinsic names, or are they allowed to depend on the auto-upgrade code?

I think they should use the new names, and have different tests to check that the old ones get auto-upgraded.

[jketema]

This extends the patch to cover all of vld[1234], vls[234]lane, vst[1234], and vst[234]lane. All tests now use the updated intrinsic names, and new tests have been added to test auto-upgrading.

Should I still email llvm-dev about this? If so, what exaclty should I ask about.

Forgot to say in my previous comment. The related clang patch is http://reviews.llvm.org/D13127

[sbaranga]

In D12985#254767, @jketema wrote:

This extends the patch to cover all of vld[1234], vls[234]lane, vst[1234], and vst[234]lane. All tests now use the updated intrinsic names, and new tests have been added to test auto-upgrading.

Should I still email llvm-dev about this? If so, what exaclty should I ask about.

Forgot to say in my previous comment. The related clang patch is http://reviews.llvm.org/D13127

Both patches look good!

The purpose of the email to llvm-dev would be to announce the interface change and to ask if this would be a problem for anyone (it shouldn't be since you've added the auto-upgrade).

[sbaranga]

LGTM

[jketema]

Committed. Thanks for all the feedback!

[sbaranga]

Thanks for all the work! :)

llvm/trunk/include/llvm/IR/IntrinsicsARM.td

 def int_arm_neon_vrinta : Neon_1Arg_Intrinsic;
 def int_arm_neon_vrintz : Neon_1Arg_Intrinsic;
 def int_arm_neon_vrintm : Neon_1Arg_Intrinsic;
 def int_arm_neon_vrintp : Neon_1Arg_Intrinsic;
 
 // De-interleaving vector loads from N-element structures.
 // Source operands are the address and alignment.
 def int_arm_neon_vld1 : Intrinsic<[llvm_anyvector_ty],
-                                  [llvm_ptr_ty, llvm_i32_ty],
+                                  [llvm_anyptr_ty, llvm_i32_ty],
                                   [IntrReadArgMem]>;
 def int_arm_neon_vld2 : Intrinsic<[llvm_anyvector_ty, LLVMMatchType<0>],
-                                  [llvm_ptr_ty, llvm_i32_ty],
+                                  [llvm_anyptr_ty, llvm_i32_ty],
                                   [IntrReadArgMem]>;
 def int_arm_neon_vld3 : Intrinsic<[llvm_anyvector_ty, LLVMMatchType<0>,
                                    LLVMMatchType<0>],
-                                  [llvm_ptr_ty, llvm_i32_ty],
+                                  [llvm_anyptr_ty, llvm_i32_ty],
                                   [IntrReadArgMem]>;
 def int_arm_neon_vld4 : Intrinsic<[llvm_anyvector_ty, LLVMMatchType<0>,
                                    LLVMMatchType<0>, LLVMMatchType<0>],
-                                  [llvm_ptr_ty, llvm_i32_ty],
+                                  [llvm_anyptr_ty, llvm_i32_ty],
                                   [IntrReadArgMem]>;
 
 // Vector load N-element structure to one lane.
 // Source operands are: the address, the N input vectors (since only one
 // lane is assigned), the lane number, and the alignment.
 def int_arm_neon_vld2lane : Intrinsic<[llvm_anyvector_ty, LLVMMatchType<0>],
-                                      [llvm_ptr_ty, LLVMMatchType<0>,
+                                      [llvm_anyptr_ty, LLVMMatchType<0>,
                                        LLVMMatchType<0>, llvm_i32_ty,
                                        llvm_i32_ty], [IntrReadArgMem]>;
 def int_arm_neon_vld3lane : Intrinsic<[llvm_anyvector_ty, LLVMMatchType<0>,
                                        LLVMMatchType<0>],
-                                      [llvm_ptr_ty, LLVMMatchType<0>,
+                                      [llvm_anyptr_ty, LLVMMatchType<0>,
                                        LLVMMatchType<0>, LLVMMatchType<0>,
                                        llvm_i32_ty, llvm_i32_ty],
                                       [IntrReadArgMem]>;
 def int_arm_neon_vld4lane : Intrinsic<[llvm_anyvector_ty, LLVMMatchType<0>,
                                        LLVMMatchType<0>, LLVMMatchType<0>],
-                                      [llvm_ptr_ty, LLVMMatchType<0>,
+                                      [llvm_anyptr_ty, LLVMMatchType<0>,
                                        LLVMMatchType<0>, LLVMMatchType<0>,
                                        LLVMMatchType<0>, llvm_i32_ty,
                                        llvm_i32_ty], [IntrReadArgMem]>;
 
 // Interleaving vector stores from N-element structures.
 // Source operands are: the address, the N vectors, and the alignment.
 def int_arm_neon_vst1 : Intrinsic<[],
-                                  [llvm_ptr_ty, llvm_anyvector_ty,
+                                  [llvm_anyptr_ty, llvm_anyvector_ty,
                                    llvm_i32_ty], [IntrReadWriteArgMem]>;
 def int_arm_neon_vst2 : Intrinsic<[],
-                                  [llvm_ptr_ty, llvm_anyvector_ty,
-                                   LLVMMatchType<0>, llvm_i32_ty],
+                                  [llvm_anyptr_ty, llvm_anyvector_ty,
+                                   LLVMMatchType<1>, llvm_i32_ty],
                                   [IntrReadWriteArgMem]>;
 def int_arm_neon_vst3 : Intrinsic<[],
-                                  [llvm_ptr_ty, llvm_anyvector_ty,
-                                   LLVMMatchType<0>, LLVMMatchType<0>,
+                                  [llvm_anyptr_ty, llvm_anyvector_ty,
+                                   LLVMMatchType<1>, LLVMMatchType<1>,
                                    llvm_i32_ty], [IntrReadWriteArgMem]>;
 def int_arm_neon_vst4 : Intrinsic<[],
-                                  [llvm_ptr_ty, llvm_anyvector_ty,
-                                   LLVMMatchType<0>, LLVMMatchType<0>,
-                                   LLVMMatchType<0>, llvm_i32_ty],
+                                  [llvm_anyptr_ty, llvm_anyvector_ty,
+                                   LLVMMatchType<1>, LLVMMatchType<1>,
+                                   LLVMMatchType<1>, llvm_i32_ty],
                                   [IntrReadWriteArgMem]>;
 
 // Vector store N-element structure from one lane.
 // Source operands are: the address, the N vectors, the lane number, and
 // the alignment.
 def int_arm_neon_vst2lane : Intrinsic<[],
-                                      [llvm_ptr_ty, llvm_anyvector_ty,
-                                       LLVMMatchType<0>, llvm_i32_ty,
+                                      [llvm_anyptr_ty, llvm_anyvector_ty,
+                                       LLVMMatchType<1>, llvm_i32_ty,
                                        llvm_i32_ty], [IntrReadWriteArgMem]>;
 def int_arm_neon_vst3lane : Intrinsic<[],
-                                      [llvm_ptr_ty, llvm_anyvector_ty,
-                                       LLVMMatchType<0>, LLVMMatchType<0>,
+                                      [llvm_anyptr_ty, llvm_anyvector_ty,
+                                       LLVMMatchType<1>, LLVMMatchType<1>,
                                        llvm_i32_ty, llvm_i32_ty],
                                       [IntrReadWriteArgMem]>;
 def int_arm_neon_vst4lane : Intrinsic<[],
-                                      [llvm_ptr_ty, llvm_anyvector_ty,
-                                       LLVMMatchType<0>, LLVMMatchType<0>,
-                                       LLVMMatchType<0>, llvm_i32_ty,
+                                      [llvm_anyptr_ty, llvm_anyvector_ty,
+                                       LLVMMatchType<1>, LLVMMatchType<1>,
+                                       LLVMMatchType<1>, llvm_i32_ty,
                                        llvm_i32_ty], [IntrReadWriteArgMem]>;
 
 // Vector bitwise select.
 def int_arm_neon_vbsl : Intrinsic<[llvm_anyvector_ty],
                         [LLVMMatchType<0>, LLVMMatchType<0>, LLVMMatchType<0>],
                         [IntrNoMem]>;
 
 

llvm/trunk/lib/IR/AutoUpgrade.cpp

 #include "llvm/IR/DiagnosticInfo.h"
 #include "llvm/IR/Function.h"
 #include "llvm/IR/IRBuilder.h"
 #include "llvm/IR/Instruction.h"
 #include "llvm/IR/IntrinsicInst.h"
 #include "llvm/IR/LLVMContext.h"
 #include "llvm/IR/Module.h"
 #include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/Regex.h"
 #include <cstring>
 using namespace llvm;
 
 // Upgrade the declarations of the SSE4.1 functions whose arguments have
 // changed their type from v4f32 to v2i64.
 static bool UpgradeSSE41Function(Function* F, Intrinsic::ID IID,
                                  Function *&NewFn) {
   // Check whether this is an old version of the function, which received
if (Name.startswith("arm.neon.vclz")) {
                                "llvm.ctlz." + Name.substr(14), F->getParent());
       return true;
     }
     if (Name.startswith("arm.neon.vcnt")) {
       NewFn = Intrinsic::getDeclaration(F->getParent(), Intrinsic::ctpop,
                                         F->arg_begin()->getType());
       return true;
     }
+    Regex vldRegex("^arm\\.neon\\.vld([1234]|[234]lane)\\.v[a-z0-9]*$");
+    if (vldRegex.match(Name)) {
+      auto fArgs = F->getFunctionType()->params();
+      SmallVector<Type *, 4> Tys(fArgs.begin(), fArgs.end());
+      // Can't use Intrinsic::getDeclaration here as the return types might
+      // then only be structurally equal.
+      FunctionType* fType = FunctionType::get(F->getReturnType(), Tys, false);
+      NewFn = Function::Create(fType, F->getLinkage(),
+                               "llvm." + Name + ".p0i8", F->getParent());
+      return true;
+    }
+    Regex vstRegex("^arm\\.neon\\.vst([1234]|[234]lane)\\.v[a-z0-9]*$");
+    if (vstRegex.match(Name)) {
+      static Intrinsic::ID StoreInts[] = {Intrinsic::arm_neon_vst1,
+                                          Intrinsic::arm_neon_vst2,
+                                          Intrinsic::arm_neon_vst3,
+                                          Intrinsic::arm_neon_vst4};
+
+      static Intrinsic::ID StoreLaneInts[] = {Intrinsic::arm_neon_vst2lane,
+                                              Intrinsic::arm_neon_vst3lane,
+                                              Intrinsic::arm_neon_vst4lane};
+
+      auto fArgs = F->getFunctionType()->params();
+      Type *Tys[] = {fArgs[0], fArgs[1]};
+      if (Name.find("lane") == StringRef::npos)
+        NewFn = Intrinsic::getDeclaration(F->getParent(),
+                                          StoreInts[fArgs.size() - 3], Tys);
+      else
+        NewFn = Intrinsic::getDeclaration(F->getParent(),
+                                          StoreLaneInts[fArgs.size() - 5], Tys);
+      return true;
+    }
     break;
   }
+
   case 'c': {
     if (Name.startswith("ctlz.") && F->arg_size() == 1) {
       F->setName(Name + ".old");
       NewFn = Intrinsic::getDeclaration(F->getParent(), Intrinsic::ctlz,
                                         F->arg_begin()->getType());
       return true;
     }
     if (Name.startswith("cttz.") && F->arg_size() == 1) {
void llvm::UpgradeIntrinsicCall(CallInst *CI, Function *NewFn) {
   std::string Name = CI->getName();
   if (!Name.empty())
     CI->setName(Name + ".old");
 
   switch (NewFn->getIntrinsicID()) {
   default:
     llvm_unreachable("Unknown function for CallInst upgrade.");
 
+  case Intrinsic::arm_neon_vld1:
+  case Intrinsic::arm_neon_vld2:
+  case Intrinsic::arm_neon_vld3:
+  case Intrinsic::arm_neon_vld4:
+  case Intrinsic::arm_neon_vld2lane:
+  case Intrinsic::arm_neon_vld3lane:
+  case Intrinsic::arm_neon_vld4lane:
+  case Intrinsic::arm_neon_vst1:
+  case Intrinsic::arm_neon_vst2:
+  case Intrinsic::arm_neon_vst3:
+  case Intrinsic::arm_neon_vst4:
+  case Intrinsic::arm_neon_vst2lane:
+  case Intrinsic::arm_neon_vst3lane:
+  case Intrinsic::arm_neon_vst4lane: {
+    SmallVector<Value *, 4> Args(CI->arg_operands().begin(),
+                                 CI->arg_operands().end());
+    CI->replaceAllUsesWith(Builder.CreateCall(NewFn, Args));
+    CI->eraseFromParent();
+    return;
+  }
+
   case Intrinsic::ctlz:
   case Intrinsic::cttz:
     assert(CI->getNumArgOperands() == 1 &&
            "Mismatch between function args and call args");
     CI->replaceAllUsesWith(Builder.CreateCall(
         NewFn, {CI->getArgOperand(0), Builder.getFalse()}, Name));
     CI->eraseFromParent();
     return;

llvm/trunk/lib/Target/ARM/ARMISelLowering.cpp

bool ARMTargetLowering::lowerInterleavedLoad(
   if (EltTy->isPointerTy())
     VecTy =
         VectorType::get(DL.getIntPtrType(EltTy), VecTy->getVectorNumElements());
 
   static const Intrinsic::ID LoadInts[3] = {Intrinsic::arm_neon_vld2,
                                             Intrinsic::arm_neon_vld3,
                                             Intrinsic::arm_neon_vld4};
 
-  Function *VldnFunc =
-      Intrinsic::getDeclaration(LI->getModule(), LoadInts[Factor - 2], VecTy);
-
   IRBuilder<> Builder(LI);
   SmallVector<Value *, 2> Ops;
 
   Type *Int8Ptr = Builder.getInt8PtrTy(LI->getPointerAddressSpace());
   Ops.push_back(Builder.CreateBitCast(LI->getPointerOperand(), Int8Ptr));
   Ops.push_back(Builder.getInt32(LI->getAlignment()));
 
+  Type *Tys[] = { VecTy, Int8Ptr };
+  Function *VldnFunc =
+      Intrinsic::getDeclaration(LI->getModule(), LoadInts[Factor - 2], Tys);
   CallInst *VldN = Builder.CreateCall(VldnFunc, Ops, "vldN");
 
   // Replace uses of each shufflevector with the corresponding vector loaded
   // by ldN.
   for (unsigned i = 0; i < Shuffles.size(); i++) {
     ShuffleVectorInst *SV = Shuffles[i];
     unsigned Index = Indices[i];
 
if (EltTy->isPointerTy()) {
     Op1 = Builder.CreatePtrToInt(Op1, IntVecTy);
 
     SubVecTy = VectorType::get(IntTy, NumSubElts);
   }
 
   static Intrinsic::ID StoreInts[3] = {Intrinsic::arm_neon_vst2,
                                        Intrinsic::arm_neon_vst3,
                                        Intrinsic::arm_neon_vst4};
-  Function *VstNFunc = Intrinsic::getDeclaration(
-      SI->getModule(), StoreInts[Factor - 2], SubVecTy);
-
   SmallVector<Value *, 6> Ops;
 
   Type *Int8Ptr = Builder.getInt8PtrTy(SI->getPointerAddressSpace());
   Ops.push_back(Builder.CreateBitCast(SI->getPointerOperand(), Int8Ptr));
 
+  Type *Tys[] = { Int8Ptr, SubVecTy };
+  Function *VstNFunc = Intrinsic::getDeclaration(
+      SI->getModule(), StoreInts[Factor - 2], Tys);
+
   // Split the shufflevector operands into sub vectors for the new vstN call.
   for (unsigned i = 0; i < Factor; i++)
     Ops.push_back(Builder.CreateShuffleVector(
         Op0, Op1, getSequentialMask(Builder, NumSubElts * i, NumSubElts)));
 
   Ops.push_back(Builder.getInt32(SI->getAlignment()));
   Builder.CreateCall(VstNFunc, Ops);
   return true;

llvm/trunk/test/Analysis/BasicAA/cs-cs.ll

 ; RUN: opt < %s -basicaa -aa-eval -print-all-alias-modref-info -disable-output 2>&1 | FileCheck %s
 target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:32:32-f32:32:32-f64:32:32-v64:32:64-v128:32:128-a0:0:32-n32"
 target triple = "arm-apple-ios"
 
-declare <8 x i16> @llvm.arm.neon.vld1.v8i16(i8*, i32) nounwind readonly
-declare void @llvm.arm.neon.vst1.v8i16(i8*, <8 x i16>, i32) nounwind
+declare <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8*, i32) nounwind readonly
+declare void @llvm.arm.neon.vst1.p0i8.v8i16(i8*, <8 x i16>, i32) nounwind
 
 declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
 declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
 
 declare void @a_readonly_func(i8 *) noinline nounwind readonly
 
 define <8 x i16> @test1(i8* %p, <8 x i16> %y) {
 entry:
   %q = getelementptr i8, i8* %p, i64 16
-  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) nounwind
-  call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16)
-  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) nounwind
+  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) nounwind
+  call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16)
+  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) nounwind
   %c = add <8 x i16> %a, %b
   ret <8 x i16> %c
 
 ; CHECK-LABEL: Function: test1:
 
 ; CHECK: NoAlias:      i8* %p, i8* %q
-; CHECK: Just Ref:  Ptr: i8* %p        <->  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4
-; CHECK: NoModRef:  Ptr: i8* %q        <->  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4
-; CHECK: NoModRef:  Ptr: i8* %p        <->  call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16)
-; CHECK: Both ModRef:  Ptr: i8* %q     <->  call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16)
-; CHECK: Just Ref:  Ptr: i8* %p        <->  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4
-; CHECK: NoModRef:  Ptr: i8* %q        <->  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4
-; CHECK: NoModRef:   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4 <->   call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16)
-; CHECK: NoModRef:   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4 <->   %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4
-; CHECK: NoModRef:   call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16) <->   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4
-; CHECK: NoModRef:   call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16) <->   %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4
-; CHECK: NoModRef:   %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4 <->   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4
-; CHECK: NoModRef:   %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) #4 <->   call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16)
+; CHECK: Just Ref:  Ptr: i8* %p        <->  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4
+; CHECK: NoModRef:  Ptr: i8* %q        <->  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4
+; CHECK: NoModRef:  Ptr: i8* %p        <->  call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16)
+; CHECK: Both ModRef:  Ptr: i8* %q     <->  call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16)
+; CHECK: Just Ref:  Ptr: i8* %p        <->  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4
+; CHECK: NoModRef:  Ptr: i8* %q        <->  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4
+; CHECK: NoModRef:   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4 <->   call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16)
+; CHECK: NoModRef:   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4 <->   %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4
+; CHECK: NoModRef:   call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16) <->   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4
+; CHECK: NoModRef:   call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16) <->   %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4
+; CHECK: NoModRef:   %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4 <->   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4
+; CHECK: NoModRef:   %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) #4 <->   call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16)
 }
 
 define void @test2(i8* %P, i8* %Q) nounwind ssp {
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
   ret void
 
 ; CHECK-LABEL: Function: test2:

llvm/trunk/test/Analysis/BasicAA/intrinsics.ll

 ; RUN: opt -basicaa -gvn -S < %s | FileCheck %s
 
 target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:32:32-f32:32:32-f64:32:32-v64:32:64-v128:32:128-a0:0:32-n32"
 
 ; BasicAA should prove that these calls don't interfere, since they are
 ; IntrArgReadMem and have noalias pointers.
 
 ; CHECK:      define <8 x i16> @test0(i8* noalias %p, i8* noalias %q, <8 x i16> %y) {
 ; CHECK-NEXT: entry:
-; CHECK-NEXT:   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) [[ATTR:#[0-9]+]]
-; CHECK-NEXT:   call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16)
+; CHECK-NEXT:   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) [[ATTR:#[0-9]+]]
+; CHECK-NEXT:   call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16)
 ; CHECK-NEXT:   %c = add <8 x i16> %a, %a
 define <8 x i16> @test0(i8* noalias %p, i8* noalias %q, <8 x i16> %y) {
 entry:
-  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) nounwind
-  call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16)
-  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) nounwind
+  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) nounwind
+  call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16)
+  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) nounwind
   %c = add <8 x i16> %a, %b
   ret <8 x i16> %c
 }
 
 ; CHECK:      define <8 x i16> @test1(i8* %p, <8 x i16> %y) {
 ; CHECK-NEXT: entry:
 ; CHECK-NEXT:   %q = getelementptr i8, i8* %p, i64 16
-; CHECK-NEXT:   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) [[ATTR]]
-; CHECK-NEXT:   call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16)
+; CHECK-NEXT:   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) [[ATTR]]
+; CHECK-NEXT:   call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16)
 ; CHECK-NEXT:   %c = add <8 x i16> %a, %a
 define <8 x i16> @test1(i8* %p, <8 x i16> %y) {
 entry:
   %q = getelementptr i8, i8* %p, i64 16
-  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) nounwind
-  call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16)
-  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) nounwind
+  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) nounwind
+  call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16)
+  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) nounwind
   %c = add <8 x i16> %a, %b
   ret <8 x i16> %c
 }
 
-declare <8 x i16> @llvm.arm.neon.vld1.v8i16(i8*, i32) nounwind readonly
-declare void @llvm.arm.neon.vst1.v8i16(i8*, <8 x i16>, i32) nounwind
+declare <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8*, i32) nounwind readonly
+declare void @llvm.arm.neon.vst1.p0i8.v8i16(i8*, <8 x i16>, i32) nounwind
 
 ; CHECK: attributes #0 = { nounwind readonly argmemonly }
 ; CHECK: attributes #1 = { nounwind argmemonly }
 ; CHECK: attributes [[ATTR]] = { nounwind }

llvm/trunk/test/Analysis/TypeBasedAliasAnalysis/intrinsics.ll

 ; RUN: opt -tbaa -basicaa -gvn -S < %s | FileCheck %s
 
 target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:32:32-f32:32:32-f64:32:32-v64:32:64-v128:32:128-a0:0:32-n32"
 
 ; TBAA should prove that these calls don't interfere, since they are
 ; IntrArgReadMem and have TBAA metadata.
 
 ; CHECK:      define <8 x i16> @test0(i8* %p, i8* %q, <8 x i16> %y) {
 ; CHECK-NEXT: entry:
-; CHECK-NEXT:   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) [[NUW:#[0-9]+]]
-; CHECK-NEXT:   call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16)
+; CHECK-NEXT:   %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) [[NUW:#[0-9]+]]
+; CHECK-NEXT:   call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16)
 ; CHECK-NEXT:   %c = add <8 x i16> %a, %a
 define <8 x i16> @test0(i8* %p, i8* %q, <8 x i16> %y) {
 entry:
-  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) nounwind, !tbaa !2
-  call void @llvm.arm.neon.vst1.v8i16(i8* %q, <8 x i16> %y, i32 16), !tbaa !1
-  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %p, i32 16) nounwind, !tbaa !2
+  %a = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) nounwind, !tbaa !2
+  call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %q, <8 x i16> %y, i32 16), !tbaa !1
+  %b = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %p, i32 16) nounwind, !tbaa !2
   %c = add <8 x i16> %a, %b
   ret <8 x i16> %c
 }
 
-declare <8 x i16> @llvm.arm.neon.vld1.v8i16(i8*, i32) nounwind readonly
-declare void @llvm.arm.neon.vst1.v8i16(i8*, <8 x i16>, i32) nounwind
+declare <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8*, i32) nounwind readonly
+declare void @llvm.arm.neon.vst1.p0i8.v8i16(i8*, <8 x i16>, i32) nounwind
 
 ; CHECK: attributes #0 = { nounwind readonly argmemonly }
 ; CHECK: attributes #1 = { nounwind argmemonly }
 ; CHECK: attributes [[NUW]] = { nounwind }
 
 !0 = !{!"tbaa root", null}
 !1 = !{!3, !3, i64 0}
 !2 = !{!4, !4, i64 0}
 !3 = !{!"A", !0}
 !4 = !{!"B", !0}

llvm/trunk/test/CodeGen/ARM/2010-05-20-NEONSpillCrash.ll

 ; RUN: llc -mtriple=arm-eabi -mattr=+neon -O0 -optimize-regalloc -regalloc=basic %s -o /dev/null
 
 ; This test would crash the rewriter when trying to handle a spill after one of
-; the @llvm.arm.neon.vld3.v8i8 defined three parts of a register.
+; the @llvm.arm.neon.vld3.v8i8.p0i8 defined three parts of a register.
 
 %struct.__neon_int8x8x3_t = type { <8 x i8>, <8 x i8>, <8 x i8> }
 
-declare %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8*, i32) nounwind readonly
 
-declare void @llvm.arm.neon.vst3.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst3.p0i8.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind
 
 define <8 x i8> @t3(i8* %A1, i8* %A2, i8* %A3, i8* %A4, i8* %A5, i8* %A6, i8* %A7, i8* %A8, i8* %B) nounwind {
-  %tmp1b = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A2, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
+  %tmp1b = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8* %A2, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
   %tmp2b = extractvalue %struct.__neon_int8x8x3_t %tmp1b, 0 ; <<8 x i8>> [#uses=1]
   %tmp4b = extractvalue %struct.__neon_int8x8x3_t %tmp1b, 1 ; <<8 x i8>> [#uses=1]
-  %tmp1d = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A4, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
+  %tmp1d = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8* %A4, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
   %tmp2d = extractvalue %struct.__neon_int8x8x3_t %tmp1d, 0 ; <<8 x i8>> [#uses=1]
   %tmp4d = extractvalue %struct.__neon_int8x8x3_t %tmp1d, 1 ; <<8 x i8>> [#uses=1]
-  %tmp1e = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A5, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=1]
+  %tmp1e = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8* %A5, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=1]
   %tmp2e = extractvalue %struct.__neon_int8x8x3_t %tmp1e, 0 ; <<8 x i8>> [#uses=1]
-  %tmp1f = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A6, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=1]
+  %tmp1f = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8* %A6, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=1]
   %tmp2f = extractvalue %struct.__neon_int8x8x3_t %tmp1f, 0 ; <<8 x i8>> [#uses=1]
-  %tmp1g = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A7, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
+  %tmp1g = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8* %A7, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
   %tmp2g = extractvalue %struct.__neon_int8x8x3_t %tmp1g, 0 ; <<8 x i8>> [#uses=1]
   %tmp4g = extractvalue %struct.__neon_int8x8x3_t %tmp1g, 1 ; <<8 x i8>> [#uses=1]
-  %tmp1h = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A8, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
+  %tmp1h = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8* %A8, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
   %tmp2h = extractvalue %struct.__neon_int8x8x3_t %tmp1h, 0 ; <<8 x i8>> [#uses=1]
   %tmp3h = extractvalue %struct.__neon_int8x8x3_t %tmp1h, 2 ; <<8 x i8>> [#uses=1]
   %tmp2bd = add <8 x i8> %tmp2b, %tmp2d           ; <<8 x i8>> [#uses=1]
   %tmp4bd = add <8 x i8> %tmp4b, %tmp4d           ; <<8 x i8>> [#uses=1]
   %tmp2abcd = mul <8 x i8> undef, %tmp2bd         ; <<8 x i8>> [#uses=1]
   %tmp4abcd = mul <8 x i8> undef, %tmp4bd         ; <<8 x i8>> [#uses=2]
-  call void @llvm.arm.neon.vst3.v8i8(i8* %A1, <8 x i8> %tmp4abcd, <8 x i8> zeroinitializer, <8 x i8> %tmp2abcd, i32 1)
+  call void @llvm.arm.neon.vst3.p0i8.v8i8(i8* %A1, <8 x i8> %tmp4abcd, <8 x i8> zeroinitializer, <8 x i8> %tmp2abcd, i32 1)
   %tmp2ef = sub <8 x i8> %tmp2e, %tmp2f           ; <<8 x i8>> [#uses=1]
   %tmp2gh = sub <8 x i8> %tmp2g, %tmp2h           ; <<8 x i8>> [#uses=1]
   %tmp3gh = sub <8 x i8> zeroinitializer, %tmp3h  ; <<8 x i8>> [#uses=1]
   %tmp4ef = sub <8 x i8> zeroinitializer, %tmp4g  ; <<8 x i8>> [#uses=1]
   %tmp2efgh = mul <8 x i8> %tmp2ef, %tmp2gh       ; <<8 x i8>> [#uses=1]
   %tmp3efgh = mul <8 x i8> undef, %tmp3gh         ; <<8 x i8>> [#uses=1]
   %tmp4efgh = mul <8 x i8> %tmp4ef, undef         ; <<8 x i8>> [#uses=2]
-  call void @llvm.arm.neon.vst3.v8i8(i8* %A2, <8 x i8> %tmp4efgh, <8 x i8> %tmp3efgh, <8 x i8> %tmp2efgh, i32 1)
+  call void @llvm.arm.neon.vst3.p0i8.v8i8(i8* %A2, <8 x i8> %tmp4efgh, <8 x i8> %tmp3efgh, <8 x i8> %tmp2efgh, i32 1)
   %tmp4 = sub <8 x i8> %tmp4efgh, %tmp4abcd       ; <<8 x i8>> [#uses=1]
-  tail call void @llvm.arm.neon.vst3.v8i8(i8* %B, <8 x i8> zeroinitializer, <8 x i8> undef, <8 x i8> undef, i32 1)
+  tail call void @llvm.arm.neon.vst3.p0i8.v8i8(i8* %B, <8 x i8> zeroinitializer, <8 x i8> undef, <8 x i8> undef, i32 1)
   ret <8 x i8> %tmp4
 }

llvm/trunk/test/CodeGen/ARM/2010-05-21-BuildVector.ll

;CHECK: vldr s
   %16 = ashr i32 %15, 30
   %.sum14 = add i32 %16, 4
   %17 = getelementptr inbounds float, float* %table, i32 %.sum14
 ;CHECK: vldr s
   %18 = load float, float* %17, align 4
   %tmp5 = insertelement <4 x float> %tmp7, float %18, i32 3
   %19 = fmul <4 x float> %tmp5, %2
   %20 = bitcast float* %fltp to i8*
-  tail call void @llvm.arm.neon.vst1.v4f32(i8* %20, <4 x float> %19, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v4f32(i8* %20, <4 x float> %19, i32 1)
   ret void
 }
 
-declare void @llvm.arm.neon.vst1.v4f32(i8*, <4 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v4f32(i8*, <4 x float>, i32) nounwind

llvm/trunk/test/CodeGen/ARM/2010-06-11-vmovdrr-bitcast.ll

 ; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o /dev/null
 ; rdar://8084742
 
 %struct.__int8x8x2_t = type { [2 x <8 x i8>] }
 
 define void @foo(%struct.__int8x8x2_t* nocapture %a, i8* %b) nounwind {
 entry:
  %0 = bitcast %struct.__int8x8x2_t* %a to i128*  ; <i128*> [#uses=1]
  %srcval = load i128, i128* %0, align 8                ; <i128> [#uses=2]
  %tmp6 = trunc i128 %srcval to i64               ; <i64> [#uses=1]
  %tmp8 = lshr i128 %srcval, 64                   ; <i128> [#uses=1]
  %tmp9 = trunc i128 %tmp8 to i64                 ; <i64> [#uses=1]
  %tmp16.i = bitcast i64 %tmp6 to <8 x i8>        ; <<8 x i8>> [#uses=1]
  %tmp20.i = bitcast i64 %tmp9 to <8 x i8>        ; <<8 x i8>> [#uses=1]
- tail call void @llvm.arm.neon.vst2.v8i8(i8* %b, <8 x i8> %tmp16.i, <8 x i8> %tmp20.i, i32 1) nounwind
+ tail call void @llvm.arm.neon.vst2.p0i8.v8i8(i8* %b, <8 x i8> %tmp16.i, <8 x i8> %tmp20.i, i32 1) nounwind
  ret void
 }
 
-declare void @llvm.arm.neon.vst2.v8i8(i8*, <8 x i8>, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst2.p0i8.v8i8(i8*, <8 x i8>, <8 x i8>, i32) nounwind

llvm/trunk/test/CodeGen/ARM/2010-06-29-PartialRedefFastAlloc.ll

 ; redef, it cannot also get %Q0.
 
 ; CHECK: vld1.64 {d16, d17}, [r{{.}}]
 ; CHECK-NOT: vld1.64 {d16, d17}
 ; CHECK: vmov.f64
 
 define i32 @test(i8* %arg) nounwind {
 entry:
- %0 = call <2 x i64> @llvm.arm.neon.vld1.v2i64(i8* %arg, i32 1)
+ %0 = call <2 x i64> @llvm.arm.neon.vld1.v2i64.p0i8(i8* %arg, i32 1)
  %1 = shufflevector <2 x i64> undef, <2 x i64> %0, <2 x i32> <i32 1, i32 2>
  store <2 x i64> %1, <2 x i64>* undef, align 16
  ret i32 undef
 }
 
-declare <2 x i64> @llvm.arm.neon.vld1.v2i64(i8*, i32) nounwind readonly
+declare <2 x i64> @llvm.arm.neon.vld1.v2i64.p0i8(i8*, i32) nounwind readonly

llvm/trunk/test/CodeGen/ARM/2011-08-12-vmovqqqq-pseudo.ll

 ; RUN: llc %s -mtriple=thumbv7-apple-darwin -verify-machineinstrs -mcpu=cortex-a9 -O0 -o -
 ; Make sure that the VMOVQQQQ pseudo instruction is handled properly
 ; by codegen.
 
 define void @test_vmovqqqq_pseudo() nounwind ssp {
 entry:
-  %vld3_lane = call { <8 x i16>, <8 x i16>, <8 x i16> } @llvm.arm.neon.vld3lane.v8i16(i8* undef, <8 x i16> undef, <8 x i16> undef, <8 x i16> zeroinitializer, i32 7, i32 2)
+  %vld3_lane = call { <8 x i16>, <8 x i16>, <8 x i16> } @llvm.arm.neon.vld3lane.v8i16.p0i8(i8* undef, <8 x i16> undef, <8 x i16> undef, <8 x i16> zeroinitializer, i32 7, i32 2)
   store { <8 x i16>, <8 x i16>, <8 x i16> } %vld3_lane, { <8 x i16>, <8 x i16>, <8 x i16> }* undef
   ret void
 }
 
-declare { <8 x i16>, <8 x i16>, <8 x i16> } @llvm.arm.neon.vld3lane.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind readonly
+declare { <8 x i16>, <8 x i16>, <8 x i16> } @llvm.arm.neon.vld3lane.v8i16.p0i8(i8*, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind readonly

llvm/trunk/test/CodeGen/ARM/2012-01-24-RegSequenceLiveRange.ll

cond.end295:                                      ; preds = %entry
   %shuffle.i38.i.i1036 = shufflevector <2 x i64> zeroinitializer, <2 x i64> undef, <1 x i32> zeroinitializer
   %shuffle.i37.i.i1037 = shufflevector <1 x i64> %shuffle.i39.i.i1035, <1 x i64> %shuffle.i38.i.i1036, <2 x i32> <i32 0, i32 1>
   %0 = bitcast <2 x i64> %shuffle.i37.i.i1037 to <4 x float>
   %1 = bitcast <4 x float> undef to <2 x i64>
   %shuffle.i36.i.i = shufflevector <2 x i64> %1, <2 x i64> undef, <1 x i32> zeroinitializer
   %shuffle.i35.i.i = shufflevector <2 x i64> undef, <2 x i64> undef, <1 x i32> zeroinitializer
   %shuffle.i34.i.i = shufflevector <1 x i64> %shuffle.i36.i.i, <1 x i64> %shuffle.i35.i.i, <2 x i32> <i32 0, i32 1>
   %2 = bitcast <2 x i64> %shuffle.i34.i.i to <4 x float>
-  tail call void @llvm.arm.neon.vst1.v4f32(i8* undef, <4 x float> %0, i32 4) nounwind
-  tail call void @llvm.arm.neon.vst1.v4f32(i8* undef, <4 x float> %2, i32 4) nounwind
+  tail call void @llvm.arm.neon.vst1.p0i8.v4f32(i8* undef, <4 x float> %0, i32 4) nounwind
+  tail call void @llvm.arm.neon.vst1.p0i8.v4f32(i8* undef, <4 x float> %2, i32 4) nounwind
   unreachable
 
 for.end:                                          ; preds = %entry
   ret void
 }
 
 ; Check that pseudo-expansion preserves <undef> flags.
 define void @foo3(i8* %p) nounwind ssp {
 entry:
-  tail call void @llvm.arm.neon.vst2.v4f32(i8* %p, <4 x float> undef, <4 x float> undef, i32 4)
+  tail call void @llvm.arm.neon.vst2.p0i8.v4f32(i8* %p, <4 x float> undef, <4 x float> undef, i32 4)
   ret void
 }
 
 declare arm_aapcs_vfpcc void @bar(i8*, float, float, float)
-declare void @llvm.arm.neon.vst1.v4f32(i8*, <4 x float>, i32) nounwind
-declare void @llvm.arm.neon.vst2.v4f32(i8*, <4 x float>, <4 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v4f32(i8*, <4 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst2.p0i8.v4f32(i8*, <4 x float>, <4 x float>, i32) nounwind

llvm/trunk/test/CodeGen/ARM/2012-05-10-PreferVMOVtoVDUP32.ll

 ; RUN: llc -mtriple=arm-eabi -mcpu=swift %s -o - | FileCheck %s
 ; <rdar://problem/10451892>
 
 define void @f(i32 %x, i32* %p) nounwind ssp {
 entry:
 ; CHECK-NOT: vdup.32
   %vecinit.i = insertelement <2 x i32> undef, i32 %x, i32 0
   %vecinit1.i = insertelement <2 x i32> %vecinit.i, i32 %x, i32 1
   %0 = bitcast i32* %p to i8*
-  tail call void @llvm.arm.neon.vst1.v2i32(i8* %0, <2 x i32> %vecinit1.i, i32 4)
+  tail call void @llvm.arm.neon.vst1.p0i8.v2i32(i8* %0, <2 x i32> %vecinit1.i, i32 4)
   ret void
 }
 
-declare void @llvm.arm.neon.vst1.v2i32(i8*, <2 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v2i32(i8*, <2 x i32>, i32) nounwind

llvm/trunk/test/CodeGen/ARM/2012-08-27-CopyPhysRegCrash.ll

 ; RUN: llc < %s -mcpu=cortex-a8 -march=thumb
 ; Test that this doesn't crash.
 ; <rdar://problem/12183003>
 
 target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:32:64-v128:32:128-a0:0:32-n32-S32"
 target triple = "thumbv7-apple-ios5.1.0"
 
-declare { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.arm.neon.vld3.v16i8(i8*, i32) nounwind readonly
+declare { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.arm.neon.vld3.v16i8.p0i8(i8*, i32) nounwind readonly
 
-declare void @llvm.arm.neon.vst1.v16i8(i8*, <16 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v16i8(i8*, <16 x i8>, i32) nounwind
 
 define void @findEdges(i8*) nounwind ssp {
   %2 = icmp sgt i32 undef, 0
   br i1 %2, label %5, label %3
 
 ; <label>:3                                       ; preds = %5, %1
   %4 = phi i8* [ %0, %1 ], [ %19, %5 ]
   ret void
 
 ; <label>:5                                       ; preds = %5, %1
   %6 = phi i8* [ %19, %5 ], [ %0, %1 ]
-  %7 = tail call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.arm.neon.vld3.v16i8(i8* null, i32 1)
+  %7 = tail call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.arm.neon.vld3.v16i8.p0i8(i8* null, i32 1)
   %8 = extractvalue { <16 x i8>, <16 x i8>, <16 x i8> } %7, 0
   %9 = getelementptr inbounds i8, i8* null, i32 3
-  %10 = tail call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.arm.neon.vld3.v16i8(i8* %9, i32 1)
+  %10 = tail call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.arm.neon.vld3.v16i8.p0i8(i8* %9, i32 1)
   %11 = extractvalue { <16 x i8>, <16 x i8>, <16 x i8> } %10, 2
-  %12 = tail call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.arm.neon.vld3.v16i8(i8* %6, i32 1)
+  %12 = tail call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.arm.neon.vld3.v16i8.p0i8(i8* %6, i32 1)
   %13 = extractvalue { <16 x i8>, <16 x i8>, <16 x i8> } %12, 0
   %14 = extractvalue { <16 x i8>, <16 x i8>, <16 x i8> } %12, 1
   %15 = getelementptr inbounds i8, i8* %6, i32 3
-  %16 = tail call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.arm.neon.vld3.v16i8(i8* %15, i32 1)
+  %16 = tail call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.arm.neon.vld3.v16i8.p0i8(i8* %15, i32 1)
   %17 = extractvalue { <16 x i8>, <16 x i8>, <16 x i8> } %16, 1
   %18 = extractvalue { <16 x i8>, <16 x i8>, <16 x i8> } %16, 2
   %19 = getelementptr inbounds i8, i8* %6, i32 48
   %20 = bitcast <16 x i8> %13 to <2 x i64>
   %21 = bitcast <16 x i8> %8 to <2 x i64>
   %22 = bitcast <16 x i8> %14 to <2 x i64>
   %23 = shufflevector <2 x i64> %22, <2 x i64> undef, <1 x i32> zeroinitializer
   %24 = bitcast <1 x i64> %23 to <8 x i8>
; <label>:5                                       ; preds = %5, %1
   %91 = bitcast <4 x i16> %90 to <1 x i64>
   %92 = shufflevector <1 x i64> undef, <1 x i64> %91, <2 x i32> <i32 0, i32 1>
   %93 = bitcast <2 x i64> %92 to <8 x i16>
   %94 = tail call <8 x i8> @llvm.arm.neon.vshiftn.v8i8(<8 x i16> %93, <8 x i16> <i16 -8, i16 -8, i16 -8, i16 -8, i16 -8, i16 -8, i16 -8, i16 -8>)
   %95 = bitcast <8 x i8> %56 to <1 x i64>
   %96 = bitcast <8 x i8> %94 to <1 x i64>
   %97 = shufflevector <1 x i64> %95, <1 x i64> %96, <2 x i32> <i32 0, i32 1>
   %98 = bitcast <2 x i64> %97 to <16 x i8>
-  tail call void @llvm.arm.neon.vst1.v16i8(i8* null, <16 x i8> %98, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v16i8(i8* null, <16 x i8> %98, i32 1)
   %99 = icmp slt i32 undef, undef
   br i1 %99, label %5, label %3
 }
 
 declare <4 x i16> @llvm.arm.neon.vqshiftnu.v4i16(<4 x i32>, <4 x i32>) nounwind readnone
 
 declare <8 x i8> @llvm.arm.neon.vshiftn.v8i8(<8 x i16>, <8 x i16>) nounwind readnone
 
 declare <4 x i16> @llvm.arm.neon.vqrshiftnu.v4i16(<4 x i32>, <4 x i32>) nounwind readnone
 
 declare <4 x i32> @llvm.arm.neon.vmullu.v4i32(<4 x i16>, <4 x i16>) nounwind readnone
 
 declare <8 x i16> @llvm.arm.neon.vmaxu.v8i16(<8 x i16>, <8 x i16>) nounwind readnone
 
 declare <8 x i16> @llvm.arm.neon.vabs.v8i16(<8 x i16>) nounwind readnone

llvm/trunk/test/CodeGen/ARM/2013-10-11-select-stalls.ll

 ; REQUIRES: asserts
 ; RUN: llc < %s -mtriple=thumbv7-apple-ios -disable-ifcvt-diamond -stats 2>&1 | FileCheck %s
 ; Evaluate the two vld1.8 instructions in separate MBB's,
 ; instead of stalling on one and conditionally overwriting its result.
 ;
 ; Update: After if-conversion the two vld1.8 instructions are in the same MBB
 ; again. So we disable this if-conversion to eliminate its influence to this
 ; test.
 
 ; CHECK-NOT: Number of pipeline stalls
 define <16 x i8> @multiselect(i32 %avail, i8* %foo, i8* %bar) {
 entry:
-  %vld1 = call <16 x i8> @llvm.arm.neon.vld1.v16i8(i8* %foo, i32 1)
-  %vld2 = call <16 x i8> @llvm.arm.neon.vld1.v16i8(i8* %bar, i32 1)
+  %vld1 = call <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* %foo, i32 1)
+  %vld2 = call <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* %bar, i32 1)
   %and = and i32 %avail, 3
   %tobool = icmp eq i32 %and, 0
   %retv = select i1 %tobool, <16 x i8> %vld1, <16 x i8> %vld2
   ret <16 x i8> %retv
 }
 
-declare <16 x i8> @llvm.arm.neon.vld1.v16i8(i8* , i32 )
+declare <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* , i32 )

llvm/trunk/test/CodeGen/ARM/2014-01-09-pseudo_expand_implicit_reg.ll

; CHECK: VST1d64Q %R{{[0-9]+}}<kill>, 8, %D{{[0-9]+}}, pred:14, pred:%noreg, %Q{{[0-9]+}}_Q{{[0-9]+}}<imp-use,kill>
   %s3 = bitcast <8 x i8> %t3 to <1 x i64>
 
   %tmp0 = bitcast <1 x i64> %s2 to i64
   %tmp1 = bitcast <1 x i64> %s3 to i64
 
   %n0 = insertelement <2 x i64> undef, i64 %tmp0, i32 0
   %n1 = insertelement <2 x i64> %n0, i64 %tmp1, i32 1
 
-  call void @llvm.arm.neon.vst4.v1i64(i8* %m, <1 x i64> %s0, <1 x i64> %s1, <1 x i64> %s2, <1 x i64> %s3, i32 8)
+  call void @llvm.arm.neon.vst4.p0i8.v1i64(i8* %m, <1 x i64> %s0, <1 x i64> %s1, <1 x i64> %s2, <1 x i64> %s3, i32 8)
 
   call void @bar(<2 x i64> %n1)
 
   ret void
 }
 
 %struct.__neon_int8x8x4_t = type { <8 x i8>,  <8 x i8>,  <8 x i8>, <8 x i8> }
 define <8 x i8> @vtbx4(<8 x i8>* %A, %struct.__neon_int8x8x4_t* %B, <8 x i8>* %C) nounwind {
 ; CHECK: vtbx4:
 ; CHECK: VTBX4 {{.*}}, pred:14, pred:%noreg, %Q{{[0-9]+}}_Q{{[0-9]+}}<imp-use>
   %tmp1 = load <8 x i8>, <8 x i8>* %A
   %tmp2 = load %struct.__neon_int8x8x4_t, %struct.__neon_int8x8x4_t* %B
         %tmp3 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 2
         %tmp6 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 3
   %tmp7 = load <8 x i8>, <8 x i8>* %C
   %tmp8 = call <8 x i8> @llvm.arm.neon.vtbx4(<8 x i8> %tmp1, <8 x i8> %tmp3, <8 x i8> %tmp4, <8 x i8> %tmp5, <8 x i8> %tmp6, <8 x i8> %tmp7)
   call void @bar2(%struct.__neon_int8x8x4_t %tmp2, <8 x i8> %tmp8)
   ret <8 x i8> %tmp8
 }
 
-declare void @llvm.arm.neon.vst4.v1i64(i8*, <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64>, i32)
+declare void @llvm.arm.neon.vst4.p0i8.v1i64(i8*, <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64>, i32)
 declare <8 x i8>  @llvm.arm.neon.vtbx4(<8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>) nounwind readnone
 declare void @bar2(%struct.__neon_int8x8x4_t, <8 x i8>)
 declare void @bar(<2 x i64> %arg)

llvm/trunk/test/CodeGen/ARM/arm-interleaved-accesses.ll

 define void @store_undef_mask_factor4(i32* %ptr, <4 x i32> %v0, <4 x i32> %v1, <4 x i32> %v2, <4 x i32> %v3) {
   %base = bitcast i32* %ptr to <16 x i32>*
   %v0_v1 = shufflevector <4 x i32> %v0, <4 x i32> %v1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
   %v2_v3 = shufflevector <4 x i32> %v2, <4 x i32> %v3, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
   %interleaved.vec = shufflevector <8 x i32> %v0_v1, <8 x i32> %v2_v3, <16 x i32> <i32 0, i32 4, i32 8, i32 undef, i32 undef, i32 5, i32 9, i32 13, i32 2, i32 6, i32 10, i32 14, i32 3, i32 7, i32 11, i32 15>
   store <16 x i32> %interleaved.vec, <16 x i32>* %base, align 4
   ret void
 }
+
+; The following test cases check that address spaces are properly handled
+
+; CHECK-LABEL: load_address_space
+; CHECK: vld3.32
+define void @load_address_space(<4 x i32> addrspace(1)* %A, <2 x i32>* %B) {
+ %tmp = load <4 x i32>, <4 x i32> addrspace(1)* %A
+ %interleaved = shufflevector <4 x i32> %tmp, <4 x i32> undef, <2 x i32> <i32 0, i32 3>
+ store <2 x i32> %interleaved, <2 x i32>* %B
+ ret void
+}
+
+; CHECK-LABEL: store_address_space
+; CHECK: vst2.32
+define void @store_address_space(<2 x i32>* %A, <2 x i32>* %B, <4 x i32> addrspace(1)* %C) {
+ %tmp0 = load <2 x i32>, <2 x i32>* %A
+ %tmp1 = load <2 x i32>, <2 x i32>* %B
+ %interleaved = shufflevector <2 x i32> %tmp0, <2 x i32> %tmp1, <4 x i32> <i32 0, i32 2, i32 1, i32 3>
+ store <4 x i32> %interleaved, <4 x i32> addrspace(1)* %C
+ ret void
+}

llvm/trunk/test/CodeGen/ARM/coalesce-subregs.ll

 ; RUN: llc < %s -mcpu=cortex-a9 -verify-coalescing -verify-machineinstrs | FileCheck %s
 target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:32:64-v128:32:128-a0:0:32-n32-S32"
 target triple = "thumbv7-apple-ios0.0.0"
 
 ; CHECK: f
 ; The vld2 and vst2 are not aligned wrt each other, the second Q loaded is the
 ; first one stored.
 ; The coalescer must find a super-register larger than QQ to eliminate the copy
 ; setting up the vst2 data.
 ; CHECK: vld2
 ; CHECK-NOT: vorr
 ; CHECK-NOT: vmov
 ; CHECK: vst2
 define void @f(float* %p, i32 %c) nounwind ssp {
 entry:
   %0 = bitcast float* %p to i8*
-  %vld2 = tail call { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32(i8* %0, i32 4)
+  %vld2 = tail call { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32.p0i8(i8* %0, i32 4)
   %vld221 = extractvalue { <4 x float>, <4 x float> } %vld2, 1
   %add.ptr = getelementptr inbounds float, float* %p, i32 8
   %1 = bitcast float* %add.ptr to i8*
-  tail call void @llvm.arm.neon.vst2.v4f32(i8* %1, <4 x float> %vld221, <4 x float> undef, i32 4)
+  tail call void @llvm.arm.neon.vst2.p0i8.v4f32(i8* %1, <4 x float> %vld221, <4 x float> undef, i32 4)
   ret void
 }
 
 ; CHECK: f1
 ; FIXME: This function still has copies.
 define void @f1(float* %p, i32 %c) nounwind ssp {
 entry:
   %0 = bitcast float* %p to i8*
-  %vld2 = tail call { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32(i8* %0, i32 4)
+  %vld2 = tail call { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32.p0i8(i8* %0, i32 4)
   %vld221 = extractvalue { <4 x float>, <4 x float> } %vld2, 1
   %add.ptr = getelementptr inbounds float, float* %p, i32 8
   %1 = bitcast float* %add.ptr to i8*
-  %vld22 = tail call { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32(i8* %1, i32 4)
+  %vld22 = tail call { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32.p0i8(i8* %1, i32 4)
   %vld2215 = extractvalue { <4 x float>, <4 x float> } %vld22, 0
-  tail call void @llvm.arm.neon.vst2.v4f32(i8* %1, <4 x float> %vld221, <4 x float> %vld2215, i32 4)
+  tail call void @llvm.arm.neon.vst2.p0i8.v4f32(i8* %1, <4 x float> %vld221, <4 x float> %vld2215, i32 4)
   ret void
 }
 
 ; CHECK: f2
 ; FIXME: This function still has copies.
 define void @f2(float* %p, i32 %c) nounwind ssp {
 entry:
   %0 = bitcast float* %p to i8*
-  %vld2 = tail call { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32(i8* %0, i32 4)
+  %vld2 = tail call { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32.p0i8(i8* %0, i32 4)
   %vld224 = extractvalue { <4 x float>, <4 x float> } %vld2, 1
   br label %do.body
 
 do.body:                                          ; preds = %do.body, %entry
   %qq0.0.1.0 = phi <4 x float> [ %vld224, %entry ], [ %vld2216, %do.body ]
   %c.addr.0 = phi i32 [ %c, %entry ], [ %dec, %do.body ]
   %p.addr.0 = phi float* [ %p, %entry ], [ %add.ptr, %do.body ]
   %add.ptr = getelementptr inbounds float, float* %p.addr.0, i32 8
   %1 = bitcast float* %add.ptr to i8*
-  %vld22 = tail call { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32(i8* %1, i32 4)
+  %vld22 = tail call { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32.p0i8(i8* %1, i32 4)
   %vld2215 = extractvalue { <4 x float>, <4 x float> } %vld22, 0
   %vld2216 = extractvalue { <4 x float>, <4 x float> } %vld22, 1
-  tail call void @llvm.arm.neon.vst2.v4f32(i8* %1, <4 x float> %qq0.0.1.0, <4 x float> %vld2215, i32 4)
+  tail call void @llvm.arm.neon.vst2.p0i8.v4f32(i8* %1, <4 x float> %qq0.0.1.0, <4 x float> %vld2215, i32 4)
   %dec = add nsw i32 %c.addr.0, -1
   %tobool = icmp eq i32 %dec, 0
   br i1 %tobool, label %do.end, label %do.body
 
 do.end:                                           ; preds = %do.body
   ret void
 }
 
-declare { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32(i8*, i32) nounwind readonly
-declare void @llvm.arm.neon.vst2.v4f32(i8*, <4 x float>, <4 x float>, i32) nounwind
+declare { <4 x float>, <4 x float> } @llvm.arm.neon.vld2.v4f32.p0i8(i8*, i32) nounwind readonly
+declare void @llvm.arm.neon.vst2.p0i8.v4f32(i8*, <4 x float>, <4 x float>, i32) nounwind
 
 ; CHECK: f3
 ; This function has lane insertions that span basic blocks.
 ; The trivial REG_SEQUENCE lowering can't handle that, but the coalescer can.
 ;
 ; void f3(float *p, float *q) {
 ;   float32x2_t x;
 ;   x[1] = p[3];
if.else:                                          ; preds = %entry
   %3 = load float, float* %arrayidx4, align 4
   %vecins5 = insertelement <2 x float> %vecins, float %3, i32 0
   br label %if.end
 
 if.end:                                           ; preds = %if.else, %if.then
   %x.0 = phi <2 x float> [ %vecins3, %if.then ], [ %vecins5, %if.else ]
   %add.ptr = getelementptr inbounds float, float* %p, i32 4
   %4 = bitcast float* %add.ptr to i8*
-  tail call void @llvm.arm.neon.vst1.v2f32(i8* %4, <2 x float> %x.0, i32 4)
+  tail call void @llvm.arm.neon.vst1.p0i8.v2f32(i8* %4, <2 x float> %x.0, i32 4)
   ret void
 }
 
-declare void @llvm.arm.neon.vst1.v2f32(i8*, <2 x float>, i32) nounwind
-declare <2 x float> @llvm.arm.neon.vld1.v2f32(i8*, i32) nounwind readonly
+declare void @llvm.arm.neon.vst1.p0i8.v2f32(i8*, <2 x float>, i32) nounwind
+declare <2 x float> @llvm.arm.neon.vld1.v2f32.p0i8(i8*, i32) nounwind readonly
 
 ; CHECK: f4
 ; This function inserts a lane into a fully defined vector.
 ; The destination lane isn't read, so the subregs can coalesce.
 ; CHECK-NOT: vmov
 ; CHECK-NOT: vorr
 define void @f4(float* %p, float* %q) nounwind ssp {
 entry:
   %0 = bitcast float* %p to i8*
-  %vld1 = tail call <2 x float> @llvm.arm.neon.vld1.v2f32(i8* %0, i32 4)
+  %vld1 = tail call <2 x float> @llvm.arm.neon.vld1.v2f32.p0i8(i8* %0, i32 4)
   %tobool = icmp eq float* %q, null
   br i1 %tobool, label %if.end, label %if.then
 
 if.then:                                          ; preds = %entry
   %1 = load float, float* %q, align 4
   %arrayidx1 = getelementptr inbounds float, float* %q, i32 1
   %2 = load float, float* %arrayidx1, align 4
   %add = fadd float %1, %2
   %vecins = insertelement <2 x float> %vld1, float %add, i32 1
   br label %if.end
 
 if.end:                                           ; preds = %entry, %if.then
   %x.0 = phi <2 x float> [ %vecins, %if.then ], [ %vld1, %entry ]
-  tail call void @llvm.arm.neon.vst1.v2f32(i8* %0, <2 x float> %x.0, i32 4)
+  tail call void @llvm.arm.neon.vst1.p0i8.v2f32(i8* %0, <2 x float> %x.0, i32 4)
   ret void
 }
 
 ; CHECK: f5
 ; Coalesce vector lanes through phis.
 ; CHECK: vmov.f32 {{.*}}, #1.0
 ; CHECK-NOT: vmov
 ; CHECK-NOT: vorr
 ; CHECK: bx
 ; We may leave the last insertelement in the if.end block.
 ; It is inserting the %add value into a dead lane, but %add causes interference
 ; in the entry block, and we don't do dead lane checks across basic blocks.
 define void @f5(float* %p, float* %q) nounwind ssp {
 entry:
   %0 = bitcast float* %p to i8*
-  %vld1 = tail call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* %0, i32 4)
+  %vld1 = tail call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* %0, i32 4)
   %vecext = extractelement <4 x float> %vld1, i32 0
   %vecext1 = extractelement <4 x float> %vld1, i32 1
   %vecext2 = extractelement <4 x float> %vld1, i32 2
   %vecext3 = extractelement <4 x float> %vld1, i32 3
   %add = fadd float %vecext3, 1.000000e+00
   %tobool = icmp eq float* %q, null
   br i1 %tobool, label %if.end, label %if.then
 

 if.end:                                           ; preds = %entry, %if.then
   %a.0 = phi float [ %add4, %if.then ], [ %vecext, %entry ]
   %b.0 = phi float [ %add6, %if.then ], [ %vecext1, %entry ]
   %c.0 = phi float [ %add8, %if.then ], [ %vecext2, %entry ]
   %vecinit = insertelement <4 x float> undef, float %a.0, i32 0
   %vecinit9 = insertelement <4 x float> %vecinit, float %b.0, i32 1
   %vecinit10 = insertelement <4 x float> %vecinit9, float %c.0, i32 2
   %vecinit11 = insertelement <4 x float> %vecinit10, float %add, i32 3
-  tail call void @llvm.arm.neon.vst1.v4f32(i8* %0, <4 x float> %vecinit11, i32 4)
+  tail call void @llvm.arm.neon.vst1.p0i8.v4f32(i8* %0, <4 x float> %vecinit11, i32 4)
   ret void
 }
 
-declare <4 x float> @llvm.arm.neon.vld1.v4f32(i8*, i32) nounwind readonly
+declare <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8*, i32) nounwind readonly
 
-declare void @llvm.arm.neon.vst1.v4f32(i8*, <4 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v4f32(i8*, <4 x float>, i32) nounwind
 
 ; CHECK: pr13999
 define void @pr13999() nounwind readonly {
 entry:
  br i1 true, label %outer_loop, label %loop.end
 
 outer_loop:
  %d = phi double [ 0.0, %entry ], [ %add, %after_inner_loop ]

llvm/trunk/test/CodeGen/ARM/dagcombine-concatvector.ll

 bb:
   %tmp = extractvalue [4 x i64] %vec.coerce, 0
   %tmp2 = bitcast i64 %tmp to <8 x i8>
   %tmp3 = shufflevector <8 x i8> %tmp2, <8 x i8> undef, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
   %tmp4 = extractvalue [4 x i64] %vec.coerce, 1
   %tmp5 = bitcast i64 %tmp4 to <8 x i8>
   %tmp6 = shufflevector <8 x i8> %tmp5, <8 x i8> undef, <16 x i32> <i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
   %tmp7 = shufflevector <16 x i8> %tmp6, <16 x i8> %tmp3, <16 x i32> <i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
-  tail call void @llvm.arm.neon.vst1.v16i8(i8* %arg, <16 x i8> %tmp7, i32 2)
+  tail call void @llvm.arm.neon.vst1.p0i8.v16i8(i8* %arg, <16 x i8> %tmp7, i32 2)
   ret void
 }
 
-declare void @llvm.arm.neon.vst1.v16i8(i8*, <16 x i8>, i32)
+declare void @llvm.arm.neon.vst1.p0i8.v16i8(i8*, <16 x i8>, i32)

llvm/trunk/test/CodeGen/ARM/neon_spill.ll

 
 declare arm_aapcs_vfpcc %0** @func2()
 
 declare arm_aapcs_vfpcc %2* @func3(%2*, %2*, i32)
 
 declare arm_aapcs_vfpcc %2** @func4()
 
 define arm_aapcs_vfpcc void @foo(%3* nocapture) nounwind align 2 {
-  call void @llvm.arm.neon.vst4.v4i32(i8* undef, <4 x i32> <i32 0, i32 1065353216, i32 1073741824, i32 1077936128>, <4 x i32> <i32 1082130432, i32 1084227584, i32 1086324736, i32 1088421888>, <4 x i32> <i32 1090519040, i32 1091567616, i32 1092616192, i32 1093664768>, <4 x i32> <i32 1094713344, i32 1095761920, i32 1096810496, i32 1097859072>, i32 16) nounwind
+  call void @llvm.arm.neon.vst4.p0i8.v4i32(i8* undef, <4 x i32> <i32 0, i32 1065353216, i32 1073741824, i32 1077936128>, <4 x i32> <i32 1082130432, i32 1084227584, i32 1086324736, i32 1088421888>, <4 x i32> <i32 1090519040, i32 1091567616, i32 1092616192, i32 1093664768>, <4 x i32> <i32 1094713344, i32 1095761920, i32 1096810496, i32 1097859072>, i32 16) nounwind
   %2 = call arm_aapcs_vfpcc  %0** @func2() nounwind
   %3 = load %0*, %0** %2, align 4
   store float 0.000000e+00, float* undef, align 4
   %4 = call arm_aapcs_vfpcc  %2* @func3(%2* undef, %2* undef, i32 2956) nounwind
   call arm_aapcs_vfpcc  void @func1(%0* %3, float* undef, float* undef, %2* undef)
   %5 = call arm_aapcs_vfpcc  %0** @func2() nounwind
   store float 1.000000e+00, float* undef, align 4
   call arm_aapcs_vfpcc  void @func1(%0* undef, float* undef, float* undef, %2* undef)
   store float 1.500000e+01, float* undef, align 4
   %6 = call arm_aapcs_vfpcc  %2** @func4() nounwind
   %7 = call arm_aapcs_vfpcc  %2* @func3(%2* undef, %2* undef, i32 2971) nounwind
   %8 = fadd float undef, -1.000000e+05
   store float %8, float* undef, align 16
   %9 = call arm_aapcs_vfpcc  i32 @rand() nounwind
   %10 = fmul float undef, 2.000000e+05
   %11 = fadd float %10, -1.000000e+05
   store float %11, float* undef, align 4
-  call void @llvm.arm.neon.vst4.v4i32(i8* undef, <4 x i32> <i32 0, i32 1065353216, i32 1073741824, i32 1077936128>, <4 x i32> <i32 1082130432, i32 1084227584, i32 1086324736, i32 1088421888>, <4 x i32> <i32 1090519040, i32 1091567616, i32 1092616192, i32 1093664768>, <4 x i32> <i32 1094713344, i32 1095761920, i32 1096810496, i32 1097859072>, i32 16) nounwind
+  call void @llvm.arm.neon.vst4.p0i8.v4i32(i8* undef, <4 x i32> <i32 0, i32 1065353216, i32 1073741824, i32 1077936128>, <4 x i32> <i32 1082130432, i32 1084227584, i32 1086324736, i32 1088421888>, <4 x i32> <i32 1090519040, i32 1091567616, i32 1092616192, i32 1093664768>, <4 x i32> <i32 1094713344, i32 1095761920, i32 1096810496, i32 1097859072>, i32 16) nounwind
   ret void
 }
 
-declare void @llvm.arm.neon.vst4.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst4.p0i8.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind
 
 declare arm_aapcs_vfpcc i32 @rand()

llvm/trunk/test/CodeGen/ARM/out-of-registers.ll

 ; RUN: llc -O3 %s -o - | FileCheck %s
 ; ModuleID = 'fo.c'
 target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:64:128-a0:0:32-n8:16:32-S64"
 target triple = "thumbv7-none-linux-gnueabi"
 
 ; CHECK: vpush
 ; CHECK: vpop
 
 define void @foo(float* nocapture %A) #0 {
   %1= bitcast float* %A to i8*
-  %2 = tail call { <4 x float>, <4 x float>, <4 x float>, <4 x float> } @llvm.arm.neon.vld4.v4f32(i8* %1, i32 4)
+  %2 = tail call { <4 x float>, <4 x float>, <4 x float>, <4 x float> } @llvm.arm.neon.vld4.v4f32.p0i8(i8* %1, i32 4)
   %3 = extractvalue { <4 x float>, <4 x float>, <4 x float>, <4 x float> } %2, 0
   %divp_vec = fdiv <4 x float> <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>, %3
   %4 = extractvalue { <4 x float>, <4 x float>, <4 x float>, <4 x float> } %2, 1
   %div3p_vec = fdiv <4 x float> <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>, %4
   %5 = extractvalue { <4 x float>, <4 x float>, <4 x float>, <4 x float> } %2, 2
   %div8p_vec = fdiv <4 x float> <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>, %5
   %6 = extractvalue { <4 x float>, <4 x float>, <4 x float>, <4 x float> } %2, 3
   %div13p_vec = fdiv <4 x float> <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>, %6
-  tail call void @llvm.arm.neon.vst4.v4f32(i8* %1, <4 x float> %divp_vec, <4 x float> %div3p_vec, <4 x float> %div8p_vec, <4 x float> %div13p_vec, i32 4)
+  tail call void @llvm.arm.neon.vst4.p0i8.v4f32(i8* %1, <4 x float> %divp_vec, <4 x float> %div3p_vec, <4 x float> %div8p_vec, <4 x float> %div13p_vec, i32 4)
  ret void
 }
 
 ; Function Attrs: nounwind
 declare i32 @llvm.annotation.i32(i32, i8*, i8*, i32) #1
 
 ; Function Attrs: nounwind readonly
 
 ; Function Attrs: nounwind
-declare void @llvm.arm.neon.vst4.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32) #1
-declare { <4 x float>, <4 x float>, <4 x float>, <4 x float> } @llvm.arm.neon.vld4.v4f32(i8*, i32) #2
+declare void @llvm.arm.neon.vst4.p0i8.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32) #1
+declare { <4 x float>, <4 x float>, <4 x float>, <4 x float> } @llvm.arm.neon.vld4.v4f32.p0i8(i8*, i32) #2
 
 ; Function Attrs: nounwind
 
 attributes #0 = { nounwind "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="true" "no-nans-fp-math"="true" "stack-protector-buffer-size"="8" "unsafe-fp-math"="true" "use-soft-float"="false" }
 attributes #1 = { nounwind }
 attributes #2 = { nounwind readonly }
 
 !llvm.ident = !{!0}
 
 !0 = !{!"Snapdragon LLVM ARM Compiler 3.4"}
 !1 = !{!1}

llvm/trunk/test/CodeGen/ARM/reg_sequence.ll

 ; CHECK:        vshrn.i32
 ; CHECK-NOT:    vmov d
 ; CHECK-NEXT:   vst1.16
   %0 = getelementptr inbounds %struct.int32x4_t, %struct.int32x4_t* %vT0ptr, i32 0, i32 0 ; <<4 x i32>*> [#uses=1]
   %1 = load <4 x i32>, <4 x i32>* %0, align 16               ; <<4 x i32>> [#uses=1]
   %2 = getelementptr inbounds %struct.int32x4_t, %struct.int32x4_t* %vT1ptr, i32 0, i32 0 ; <<4 x i32>*> [#uses=1]
   %3 = load <4 x i32>, <4 x i32>* %2, align 16               ; <<4 x i32>> [#uses=1]
   %4 = bitcast i16* %i_ptr to i8*                 ; <i8*> [#uses=1]
-  %5 = tail call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %4, i32 1) ; <<8 x i16>> [#uses=1]
+  %5 = tail call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %4, i32 1) ; <<8 x i16>> [#uses=1]
   %6 = bitcast <8 x i16> %5 to <2 x double>       ; <<2 x double>> [#uses=2]
   %7 = extractelement <2 x double> %6, i32 0      ; <double> [#uses=1]
   %8 = bitcast double %7 to <4 x i16>             ; <<4 x i16>> [#uses=1]
   %9 = sext <4 x i16> %8 to <4 x i32>             ; <<4 x i32>> [#uses=1]
   %10 = extractelement <2 x double> %6, i32 1     ; <double> [#uses=1]
   %11 = bitcast double %10 to <4 x i16>           ; <<4 x i16>> [#uses=1]
   %12 = sext <4 x i16> %11 to <4 x i32>           ; <<4 x i32>> [#uses=1]
   %13 = mul <4 x i32> %1, %9                      ; <<4 x i32>> [#uses=1]
   %14 = mul <4 x i32> %3, %12                     ; <<4 x i32>> [#uses=1]
   %15 = lshr <4 x i32> %13, <i32 12, i32 12, i32 12, i32 12>
   %trunc_15 = trunc <4 x i32> %15 to <4 x i16>
   %16 = lshr <4 x i32> %14, <i32 12, i32 12, i32 12, i32 12>
   %trunc_16 = trunc <4 x i32> %16 to <4 x i16>
   %17 = shufflevector <4 x i16> %trunc_15, <4 x i16> %trunc_16, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7> ; <<8 x i16>> [#uses=1]
   %18 = bitcast i16* %o_ptr to i8*                ; <i8*> [#uses=1]
-  tail call void @llvm.arm.neon.vst1.v8i16(i8* %18, <8 x i16> %17, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %18, <8 x i16> %17, i32 1)
   ret void
 }
 
 define void @t2(i16* %i_ptr, i16* %o_ptr, %struct.int16x8_t* nocapture %vT0ptr, %struct.int16x8_t* nocapture %vT1ptr) nounwind {
 entry:
 ; CHECK-LABEL:        t2:
 ; CHECK:        vld1.16
 ; CHECK-NOT:    vmov
 ; CHECK:        vmul.i16
 ; CHECK:        vld1.16
 ; CHECK:        vmul.i16
 ; CHECK-NOT:    vmov
 ; CHECK:        vst1.16
 ; CHECK:        vst1.16
   %0 = getelementptr inbounds %struct.int16x8_t, %struct.int16x8_t* %vT0ptr, i32 0, i32 0 ; <<8 x i16>*> [#uses=1]
   %1 = load <8 x i16>, <8 x i16>* %0, align 16               ; <<8 x i16>> [#uses=1]
   %2 = getelementptr inbounds %struct.int16x8_t, %struct.int16x8_t* %vT1ptr, i32 0, i32 0 ; <<8 x i16>*> [#uses=1]
   %3 = load <8 x i16>, <8 x i16>* %2, align 16               ; <<8 x i16>> [#uses=1]
   %4 = bitcast i16* %i_ptr to i8*                 ; <i8*> [#uses=1]
-  %5 = tail call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %4, i32 1) ; <<8 x i16>> [#uses=1]
+  %5 = tail call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %4, i32 1) ; <<8 x i16>> [#uses=1]
   %6 = getelementptr inbounds i16, i16* %i_ptr, i32 8  ; <i16*> [#uses=1]
   %7 = bitcast i16* %6 to i8*                     ; <i8*> [#uses=1]
-  %8 = tail call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %7, i32 1) ; <<8 x i16>> [#uses=1]
+  %8 = tail call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %7, i32 1) ; <<8 x i16>> [#uses=1]
   %9 = mul <8 x i16> %1, %5                       ; <<8 x i16>> [#uses=1]
   %10 = mul <8 x i16> %3, %8                      ; <<8 x i16>> [#uses=1]
   %11 = bitcast i16* %o_ptr to i8*                ; <i8*> [#uses=1]
-  tail call void @llvm.arm.neon.vst1.v8i16(i8* %11, <8 x i16> %9, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %11, <8 x i16> %9, i32 1)
   %12 = getelementptr inbounds i16, i16* %o_ptr, i32 8 ; <i16*> [#uses=1]
   %13 = bitcast i16* %12 to i8*                   ; <i8*> [#uses=1]
-  tail call void @llvm.arm.neon.vst1.v8i16(i8* %13, <8 x i16> %10, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %13, <8 x i16> %10, i32 1)
   ret void
 }
 
 define <8 x i8> @t3(i8* %A, i8* %B) nounwind {
 ; CHECK-LABEL:        t3:
 ; CHECK:        vld3.8
 ; CHECK:        vmul.i8
 ; CHECK:        vmov r
 ; CHECK-NOT:    vmov d
 ; CHECK:        vst3.8
-  %tmp1 = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
+  %tmp1 = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8* %A, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=2]
   %tmp2 = extractvalue %struct.__neon_int8x8x3_t %tmp1, 0 ; <<8 x i8>> [#uses=1]
   %tmp3 = extractvalue %struct.__neon_int8x8x3_t %tmp1, 2 ; <<8 x i8>> [#uses=1]
   %tmp4 = extractvalue %struct.__neon_int8x8x3_t %tmp1, 1 ; <<8 x i8>> [#uses=1]
   %tmp5 = sub <8 x i8> %tmp3, %tmp4
   %tmp6 = add <8 x i8> %tmp2, %tmp3               ; <<8 x i8>> [#uses=1]
   %tmp7 = mul <8 x i8> %tmp4, %tmp2
-  tail call void @llvm.arm.neon.vst3.v8i8(i8* %B, <8 x i8> %tmp5, <8 x i8> %tmp6, <8 x i8> %tmp7, i32 1)
+  tail call void @llvm.arm.neon.vst3.p0i8.v8i8(i8* %B, <8 x i8> %tmp5, <8 x i8> %tmp6, <8 x i8> %tmp7, i32 1)
   ret <8 x i8> %tmp4
 }
 
 define void @t4(i32* %in, i32* %out) nounwind {
 entry:
 ; CHECK-LABEL:        t4:
 ; CHECK:        vld2.32
 ; CHECK-NOT:    vmov
 ; CHECK:        vld2.32
 ; CHECK-NOT:    vmov
 ; CHECK:        bne
   %tmp1 = bitcast i32* %in to i8*                 ; <i8*> [#uses=1]
-  %tmp2 = tail call %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32(i8* %tmp1, i32 1) ; <%struct.__neon_int32x4x2_t> [#uses=2]
+  %tmp2 = tail call %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32.p0i8(i8* %tmp1, i32 1) ; <%struct.__neon_int32x4x2_t> [#uses=2]
   %tmp3 = getelementptr inbounds i32, i32* %in, i32 8  ; <i32*> [#uses=1]
   %tmp4 = bitcast i32* %tmp3 to i8*               ; <i8*> [#uses=1]
-  %tmp5 = tail call %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32(i8* %tmp4, i32 1) ; <%struct.__neon_int32x4x2_t> [#uses=2]
+  %tmp5 = tail call %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32.p0i8(i8* %tmp4, i32 1) ; <%struct.__neon_int32x4x2_t> [#uses=2]
   %tmp8 = bitcast i32* %out to i8*                ; <i8*> [#uses=1]
   br i1 undef, label %return1, label %return2
 
 return1:
 ; CHECK:        %return1
 ; CHECK-NOT:    vmov
 ; CHECK-NEXT:   vadd.i32
 ; CHECK-NEXT:   vadd.i32
 ; CHECK-NEXT:   vst2.32
   %tmp52 = extractvalue %struct.__neon_int32x4x2_t %tmp2, 0 ; <<4 x i32>> [#uses=1]
   %tmp57 = extractvalue %struct.__neon_int32x4x2_t %tmp2, 1 ; <<4 x i32>> [#uses=1]
   %tmp = extractvalue %struct.__neon_int32x4x2_t %tmp5, 0 ; <<4 x i32>> [#uses=1]
   %tmp39 = extractvalue %struct.__neon_int32x4x2_t %tmp5, 1 ; <<4 x i32>> [#uses=1]
   %tmp6 = add <4 x i32> %tmp52, %tmp              ; <<4 x i32>> [#uses=1]
   %tmp7 = add <4 x i32> %tmp57, %tmp39            ; <<4 x i32>> [#uses=1]
-  tail call void @llvm.arm.neon.vst2.v4i32(i8* %tmp8, <4 x i32> %tmp6, <4 x i32> %tmp7, i32 1)
+  tail call void @llvm.arm.neon.vst2.p0i8.v4i32(i8* %tmp8, <4 x i32> %tmp6, <4 x i32> %tmp7, i32 1)
   ret void
 
 return2:
 ; CHECK:        %return2
 ; CHECK:        vadd.i32
 ; CHECK-NOT:    vmov
 ; CHECK:        vst2.32 {d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}}
   %tmp100 = extractvalue %struct.__neon_int32x4x2_t %tmp2, 0 ; <<4 x i32>> [#uses=1]
   %tmp101 = extractvalue %struct.__neon_int32x4x2_t %tmp5, 1 ; <<4 x i32>> [#uses=1]
   %tmp102 = add <4 x i32> %tmp100, %tmp101              ; <<4 x i32>> [#uses=1]
-  tail call void @llvm.arm.neon.vst2.v4i32(i8* %tmp8, <4 x i32> %tmp102, <4 x i32> %tmp101, i32 1)
+  tail call void @llvm.arm.neon.vst2.p0i8.v4i32(i8* %tmp8, <4 x i32> %tmp102, <4 x i32> %tmp101, i32 1)
   call void @llvm.trap()
   unreachable
 }
 
 define <8 x i16> @t5(i16* %A, <8 x i16>* %B) nounwind {
 ; CHECK-LABEL:        t5:
 ; CHECK:        vld1.32
 ; How can FileCheck match Q and D registers? We need a lisp interpreter.
 ; CHECK:        vorr {{q[0-9]+}}, {{q[0-9]+}}, {{q[0-9]+}}
 ; CHECK-NOT:    vmov
 ; CHECK:        vld2.16 {d{{[0-9]+}}[1], d{{[0-9]+}}[1]}, [r0]
 ; CHECK-NOT:    vmov
 ; CHECK:        vadd.i16
   %tmp0 = bitcast i16* %A to i8*                  ; <i8*> [#uses=1]
   %tmp1 = load <8 x i16>, <8 x i16>* %B                      ; <<8 x i16>> [#uses=2]
-  %tmp2 = call %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2lane.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1, i32 1) ; <%struct.__neon_int16x8x2_t> [#uses=2]
+  %tmp2 = call %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2lane.v8i16.p0i8(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1, i32 1) ; <%struct.__neon_int16x8x2_t> [#uses=2]
   %tmp3 = extractvalue %struct.__neon_int16x8x2_t %tmp2, 0 ; <<8 x i16>> [#uses=1]
   %tmp4 = extractvalue %struct.__neon_int16x8x2_t %tmp2, 1 ; <<8 x i16>> [#uses=1]
   %tmp5 = add <8 x i16> %tmp3, %tmp4              ; <<8 x i16>> [#uses=1]
   ret <8 x i16> %tmp5
 }
 
 define <8 x i8> @t6(i8* %A, <8 x i8>* %B) nounwind {
 ; CHECK-LABEL:        t6:
 ; CHECK:        vldr
 ; CHECK:        vorr d[[D0:[0-9]+]], d[[D1:[0-9]+]]
 ; CHECK-NEXT:   vld2.8 {d[[D1]][1], d[[D0]][1]}
   %tmp1 = load <8 x i8>, <8 x i8>* %B                       ; <<8 x i8>> [#uses=2]
-  %tmp2 = call %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 1) ; <%struct.__neon_int8x8x2_t> [#uses=2]
+  %tmp2 = call %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8.p0i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 1) ; <%struct.__neon_int8x8x2_t> [#uses=2]
   %tmp3 = extractvalue %struct.__neon_int8x8x2_t %tmp2, 0 ; <<8 x i8>> [#uses=1]
   %tmp4 = extractvalue %struct.__neon_int8x8x2_t %tmp2, 1 ; <<8 x i8>> [#uses=1]
   %tmp5 = add <8 x i8> %tmp3, %tmp4               ; <<8 x i8>> [#uses=1]
   ret <8 x i8> %tmp5
 }
 
 define void @t7(i32* %iptr, i32* %optr) nounwind {
 entry:
 ; CHECK-LABEL:        t7:
 ; CHECK:        vld2.32
 ; CHECK:        vst2.32
 ; CHECK:        vld1.32 {d{{[0-9]+}}, d{{[0-9]+}}},
 ; CHECK:        vorr q[[Q0:[0-9]+]], q[[Q1:[0-9]+]], q[[Q1:[0-9]+]]
 ; CHECK-NOT:    vmov
 ; CHECK:        vuzp.32 q[[Q1]], q[[Q0]]
 ; CHECK:        vst1.32
   %0 = bitcast i32* %iptr to i8*                  ; <i8*> [#uses=2]
-  %1 = tail call %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32(i8* %0, i32 1) ; <%struct.__neon_int32x4x2_t> [#uses=2]
+  %1 = tail call %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32.p0i8(i8* %0, i32 1) ; <%struct.__neon_int32x4x2_t> [#uses=2]
   %tmp57 = extractvalue %struct.__neon_int32x4x2_t %1, 0 ; <<4 x i32>> [#uses=1]
   %tmp60 = extractvalue %struct.__neon_int32x4x2_t %1, 1 ; <<4 x i32>> [#uses=1]
   %2 = bitcast i32* %optr to i8*                  ; <i8*> [#uses=2]
-  tail call void @llvm.arm.neon.vst2.v4i32(i8* %2, <4 x i32> %tmp57, <4 x i32> %tmp60, i32 1)
-  %3 = tail call <4 x i32> @llvm.arm.neon.vld1.v4i32(i8* %0, i32 1) ; <<4 x i32>> [#uses=1]
+  tail call void @llvm.arm.neon.vst2.p0i8.v4i32(i8* %2, <4 x i32> %tmp57, <4 x i32> %tmp60, i32 1)
+  %3 = tail call <4 x i32> @llvm.arm.neon.vld1.v4i32.p0i8(i8* %0, i32 1) ; <<4 x i32>> [#uses=1]
   %4 = shufflevector <4 x i32> %3, <4 x i32> undef, <4 x i32> <i32 0, i32 2, i32 0, i32 2> ; <<4 x i32>> [#uses=1]
-  tail call void @llvm.arm.neon.vst1.v4i32(i8* %2, <4 x i32> %4, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v4i32(i8* %2, <4 x i32> %4, i32 1)
   ret void
 }
 
 ; PR7156
 define arm_aapcs_vfpcc i32 @t8() nounwind {
 ; CHECK-LABEL: t8:
 ; CHECK: vrsqrte.f32 q8, q8
 bb.nph55.bb.nph55.split_crit_edge:
exit:          ; preds = %bb.i19
   unreachable
 
 bb14:                                             ; preds = %bb6
   ret i32 0
 }
 
 ; This test crashes the coalescer because live variables were not updated properly.
 define <8 x i8> @t11(i8* %A1, i8* %A2, i8* %A3, i8* %A4, i8* %A5, i8* %A6, i8* %A7, i8* %A8, i8* %B) nounwind {
-  %tmp1d = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A4, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=1]
+  %tmp1d = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8* %A4, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=1]
   %tmp2d = extractvalue %struct.__neon_int8x8x3_t %tmp1d, 0 ; <<8 x i8>> [#uses=1]
-  %tmp1f = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A6, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=1]
+  %tmp1f = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8* %A6, i32 1) ; <%struct.__neon_int8x8x3_t> [#uses=1]
   %tmp2f = extractvalue %struct.__neon_int8x8x3_t %tmp1f, 0 ; <<8 x i8>> [#uses=1]
   %tmp2bd = add <8 x i8> zeroinitializer, %tmp2d  ; <<8 x i8>> [#uses=1]
   %tmp2abcd = mul <8 x i8> zeroinitializer, %tmp2bd ; <<8 x i8>> [#uses=1]
   %tmp2ef = sub <8 x i8> zeroinitializer, %tmp2f  ; <<8 x i8>> [#uses=1]
   %tmp2efgh = mul <8 x i8> %tmp2ef, undef         ; <<8 x i8>> [#uses=2]
-  call void @llvm.arm.neon.vst3.v8i8(i8* %A2, <8 x i8> undef, <8 x i8> undef, <8 x i8> %tmp2efgh, i32 1)
+  call void @llvm.arm.neon.vst3.p0i8.v8i8(i8* %A2, <8 x i8> undef, <8 x i8> undef, <8 x i8> %tmp2efgh, i32 1)
   %tmp2 = sub <8 x i8> %tmp2efgh, %tmp2abcd       ; <<8 x i8>> [#uses=1]
   %tmp7 = mul <8 x i8> undef, %tmp2               ; <<8 x i8>> [#uses=1]
-  tail call void @llvm.arm.neon.vst3.v8i8(i8* %B, <8 x i8> undef, <8 x i8> undef, <8 x i8> %tmp7, i32 1)
+  tail call void @llvm.arm.neon.vst3.p0i8.v8i8(i8* %B, <8 x i8> undef, <8 x i8> undef, <8 x i8> %tmp7, i32 1)
   ret <8 x i8> undef
 }
 
-declare <4 x i32> @llvm.arm.neon.vld1.v4i32(i8*, i32) nounwind readonly
+declare <4 x i32> @llvm.arm.neon.vld1.v4i32.p0i8(i8*, i32) nounwind readonly
 
-declare <8 x i16> @llvm.arm.neon.vld1.v8i16(i8*, i32) nounwind readonly
+declare <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8*, i32) nounwind readonly
 
 declare <4 x i16> @llvm.arm.neon.vshiftn.v4i16(<4 x i32>, <4 x i32>) nounwind readnone
 
-declare void @llvm.arm.neon.vst1.v4i32(i8*, <4 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v4i32(i8*, <4 x i32>, i32) nounwind
 
-declare void @llvm.arm.neon.vst1.v8i16(i8*, <8 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v8i16(i8*, <8 x i16>, i32) nounwind
 
-declare void @llvm.arm.neon.vst3.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32)
+declare void @llvm.arm.neon.vst3.p0i8.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32)
 nounwind
 
-declare %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8*, i32) nounwind readonly
 
-declare %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32(i8*, i32) nounwind readonly
+declare %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32.p0i8(i8*, i32) nounwind readonly
 
-declare %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8(i8*, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
+declare %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8.p0i8(i8*, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
 
-declare %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2lane.v8i16(i8*, <8 x i16>, <8 x i16>, i32, i32) nounwind readonly
+declare %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2lane.v8i16.p0i8(i8*, <8 x i16>, <8 x i16>, i32, i32) nounwind readonly
 
-declare void @llvm.arm.neon.vst2.v4i32(i8*, <4 x i32>, <4 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst2.p0i8.v4i32(i8*, <4 x i32>, <4 x i32>, i32) nounwind
 
 declare <4 x float> @llvm.arm.neon.vrsqrte.v4f32(<4 x float>) nounwind readnone
 
 declare void @llvm.trap() nounwind

llvm/trunk/test/CodeGen/ARM/spill-q.ll

 ; RUN: llc < %s -mtriple=armv7-elf -mattr=+neon -arm-atomic-cfg-tidy=0 | FileCheck %s
 ; PR4789
 
 %bar = type { float, float, float }
 %baz = type { i32, [16 x %bar], [16 x float], [16 x i32], i8 }
 %foo = type { <4 x float> }
 %quux = type { i32 (...)**, %baz*, i32 }
 %quuz = type { %quux, i32, %bar, [128 x i8], [16 x %foo], %foo, %foo, %foo }
 
-declare <4 x float> @llvm.arm.neon.vld1.v4f32(i8*, i32) nounwind readonly
+declare <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8*, i32) nounwind readonly
 
 define void @aaa(%quuz* %this, i8* %block) {
 ; CHECK-LABEL: aaa:
 ; CHECK: bfc {{.*}}, #0, #4
 ; CHECK: vst1.64 {{.*}}sp:128
 ; CHECK: vld1.64 {{.*}}sp:128
 entry:
   %aligned_vec = alloca <4 x float>, align 16
   %"alloca point" = bitcast i32 0 to i32
   %vecptr = bitcast <4 x float>* %aligned_vec to i8*
-  %0 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* %vecptr, i32 1) nounwind ; <<4 x float>> [#uses=1]
+  %0 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* %vecptr, i32 1) nounwind ; <<4 x float>> [#uses=1]
   store float 6.300000e+01, float* undef, align 4
-  %1 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind ; <<4 x float>> [#uses=1]
+  %1 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind ; <<4 x float>> [#uses=1]
   store float 0.000000e+00, float* undef, align 4
-  %2 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind ; <<4 x float>> [#uses=1]
-  %ld3 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %2 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind ; <<4 x float>> [#uses=1]
+  %ld3 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld4 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld4 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld5 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld5 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld6 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld6 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld7 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld7 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld8 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld8 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld9 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld9 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld10 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld10 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld11 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld11 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld12 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld12 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
   %val173 = load <4 x float>, <4 x float>* undef               ; <<4 x float>> [#uses=1]
   br label %bb4
 
 bb4:                                              ; preds = %bb193, %entry
   %besterror.0.2264 = phi <4 x float> [ undef, %entry ], [ %besterror.0.0, %bb193 ] ; <<4 x float>> [#uses=2]
   %part0.0.0261 = phi <4 x float> [ zeroinitializer, %entry ], [ %23, %bb193 ] ; <<4 x float>> [#uses=2]
   %3 = fmul <4 x float> zeroinitializer, %0       ; <<4 x float>> [#uses=2]

llvm/trunk/test/CodeGen/ARM/vcge.ll

 ;CHECK: vcle.f32
 entry:
   %0 = fcmp ole <4 x float> undef, zeroinitializer
   %1 = sext <4 x i1> %0 to <4 x i16>
   %2 = add <4 x i16> %1, zeroinitializer
   %3 = shufflevector <4 x i16> %2, <4 x i16> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
   %4 = add <8 x i16> %3, <i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1>
   %5 = trunc <8 x i16> %4 to <8 x i8>
-  tail call void @llvm.arm.neon.vst1.v8i8(i8* undef, <8 x i8> %5, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i8(i8* undef, <8 x i8> %5, i32 1)
   unreachable
 }
 
-declare void @llvm.arm.neon.vst1.v8i8(i8*, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v8i8(i8*, <8 x i8>, i32) nounwind

llvm/trunk/test/CodeGen/ARM/vector-DAGCombine.ll

 ; operands with i16 elements.
 define void @test_i16_constant_fold() nounwind optsize {
 entry:
   %0 = sext <4 x i1> zeroinitializer to <4 x i16>
   %1 = add <4 x i16> %0, zeroinitializer
   %2 = shufflevector <4 x i16> %1, <4 x i16> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
   %3 = add <8 x i16> %2, <i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1>
   %4 = trunc <8 x i16> %3 to <8 x i8>
-  tail call void @llvm.arm.neon.vst1.v8i8(i8* undef, <8 x i8> %4, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i8(i8* undef, <8 x i8> %4, i32 1)
   unreachable
 }
 
-declare void @llvm.arm.neon.vst1.v8i8(i8*, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v8i8(i8*, <8 x i8>, i32) nounwind
 
 ; Test that loads and stores of i64 vector elements are handled as f64 values
 ; so they are not split up into i32 values.  Radar 8755338.
 define void @i64_buildvector(i64* %ptr, <2 x i64>* %vp) nounwind {
 ; CHECK: i64_buildvector
 ; CHECK: vldr
   %t0 = load i64, i64* %ptr, align 4
   %t1 = insertelement <2 x i64> undef, i64 %t0, i32 0

llvm/trunk/test/CodeGen/ARM/vld-vst-upgrade.ll

+; RUN: llc -mtriple=arm-eabi -mattr=+neon < %s | FileCheck %s
+
+%struct.__neon_int32x2x2_t = type { <2 x i32>, <2 x i32> }
+%struct.__neon_int32x2x3_t = type { <2 x i32>, <2 x i32>, <2 x i32> }
+%struct.__neon_int32x2x4_t = type { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> }
+
+; vld[1234] auto-upgrade tests
+
+; CHECK-LABEL: test_vld1_upgrade:
+; CHECK: vld1.32 {d16}, [r0]
+define <2 x i32> @test_vld1_upgrade(i8* %ptr) {
+  %tmp1 = call <2 x i32> @llvm.arm.neon.vld1.v2i32(i8* %ptr, i32 1)
+  ret <2 x i32> %tmp1
+}
+
+declare <2 x i32> @llvm.arm.neon.vld1.v2i32(i8*, i32) nounwind readonly
+
+; CHECK-LABEL: test_vld2_upgrade:
+; CHECK: vld2.32 {d16, d17}, [r0]
+define %struct.__neon_int32x2x2_t @test_vld2_upgrade(i8* %ptr) {
+  %tmp1 = call %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2.v2i32(i8* %ptr, i32 1)
+  ret %struct.__neon_int32x2x2_t %tmp1
+}
+
+declare %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2.v2i32(i8*, i32) nounwind readonly
+
+; CHECK-LABEL: test_vld3_upgrade:
+; CHECK: vld3.32 {d16, d17, d18}, [r1]
+define %struct.__neon_int32x2x3_t @test_vld3_upgrade(i8* %ptr) {
+  %tmp1 = call %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3.v2i32(i8* %ptr, i32 1)
+  ret %struct.__neon_int32x2x3_t %tmp1
+}
+
+declare %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3.v2i32(i8*, i32) nounwind readonly
+
+; CHECK-LABEL: test_vld4_upgrade:
+; CHECK: vld4.32 {d16, d17, d18, d19}, [r1]
+define %struct.__neon_int32x2x4_t @test_vld4_upgrade(i8* %ptr) {
+  %tmp1 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8* %ptr, i32 1)
+  ret %struct.__neon_int32x2x4_t %tmp1
+}
+
+declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8*, i32) nounwind readonly
+
+; vld[234]lane auto-upgrade tests
+
+; CHECK-LABEL: test_vld2lane_upgrade:
+; CHECK: vld2.32 {d16[1], d17[1]}, [r0]
+define %struct.__neon_int32x2x2_t @test_vld2lane_upgrade(i8* %ptr, <2 x i32> %A, <2 x i32> %B) {
+  %tmp1 = call %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2lane.v2i32(i8* %ptr, <2 x i32> %A, <2 x i32> %B, i32 1, i32 1)
+  ret %struct.__neon_int32x2x2_t %tmp1
+}
+
+declare %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2lane.v2i32(i8*, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
+
+; CHECK-LABEL: test_vld3lane_upgrade:
+; CHECK: vld3.32 {d16[1], d17[1], d18[1]}, [r1]
+define %struct.__neon_int32x2x3_t @test_vld3lane_upgrade(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C) {
+  %tmp1 = call %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3lane.v2i32(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C, i32 1, i32 1)
+  ret %struct.__neon_int32x2x3_t %tmp1
+}
+
+declare %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
+
+; CHECK-LABEL: test_vld4lane_upgrade:
+; CHECK: vld4.32 {d16[1], d17[1], d18[1], d19[1]}, [r1]
+define %struct.__neon_int32x2x4_t @test_vld4lane_upgrade(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C, <2 x i32> %D) {
+  %tmp1 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4lane.v2i32(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C, <2 x i32> %D, i32 1, i32 1)
+  ret %struct.__neon_int32x2x4_t %tmp1
+}
+
+declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
+
+; vst[1234] auto-upgrade tests
+
+; CHECK-LABEL: test_vst1_upgrade:
+; CHECK: vst1.32 {d16}, [r0]
+define void @test_vst1_upgrade(i8* %ptr, <2 x i32> %A) {
+  call void @llvm.arm.neon.vst1.v2i32(i8* %ptr, <2 x i32> %A, i32 1)
+  ret void
+}
+
+declare void @llvm.arm.neon.vst1.v2i32(i8*, <2 x i32>, i32) nounwind
+
+; CHECK-LABEL: test_vst2_upgrade:
+; CHECK: vst2.32 {d16, d17}, [r0]
+define void @test_vst2_upgrade(i8* %ptr, <2 x i32> %A, <2 x i32> %B) {
+  call void @llvm.arm.neon.vst2.v2i32(i8* %ptr, <2 x i32> %A, <2 x i32> %B, i32 1)
+  ret void
+}
+
+declare void @llvm.arm.neon.vst2.v2i32(i8*, <2 x i32>, <2 x i32>, i32) nounwind
+
+; CHECK-LABEL: test_vst3_upgrade:
+; CHECK: vst3.32 {d16, d17, d18}, [r0]
+define void @test_vst3_upgrade(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C) {
+  call void @llvm.arm.neon.vst3.v2i32(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C, i32 1)
+  ret void
+}
+
+declare void @llvm.arm.neon.vst3.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind
+
+; CHECK-LABEL: test_vst4_upgrade:
+; CHECK: vst4.32 {d16, d17, d18, d19}, [r0]
+define void @test_vst4_upgrade(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C, <2 x i32> %D) {
+  call void @llvm.arm.neon.vst4.v2i32(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C, <2 x i32> %D, i32 1)
+  ret void
+}
+
+declare void @llvm.arm.neon.vst4.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind
+
+; vst[234]lane auto-upgrade tests
+
+; CHECK-LABEL: test_vst2lane_upgrade:
+; CHECK: vst2.32 {d16[1], d17[1]}, [r0]
+define void @test_vst2lane_upgrade(i8* %ptr, <2 x i32> %A, <2 x i32> %B) {
+  call void @llvm.arm.neon.vst2lane.v2i32(i8* %ptr, <2 x i32> %A, <2 x i32> %B, i32 1, i32 1)
+  ret void
+}
+
+declare void @llvm.arm.neon.vst2lane.v2i32(i8*, <2 x i32>, <2 x i32>, i32, i32) nounwind
+
+; CHECK-LABEL: test_vst3lane_upgrade:
+; CHECK: vst3.32 {d16[1], d17[1], d18[1]}, [r0]
+define void @test_vst3lane_upgrade(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C) {
+  call void @llvm.arm.neon.vst3lane.v2i32(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C, i32 1, i32 1)
+  ret void
+}
+
+declare void @llvm.arm.neon.vst3lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind
+
+; CHECK-LABEL: test_vst4lane_upgrade:
+; CHECK: vst4.32 {d16[1], d17[1], d18[1], d19[1]}, [r0]
+define void @test_vst4lane_upgrade(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C, <2 x i32> %D) {
+  call void @llvm.arm.neon.vst4lane.v2i32(i8* %ptr, <2 x i32> %A, <2 x i32> %B, <2 x i32> %C, <2 x i32> %D, i32 1, i32 1)
+  ret void
+}
+
+declare void @llvm.arm.neon.vst4lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind

llvm/trunk/test/CodeGen/ARM/vld1.ll

 ; RUN: llc -mtriple=arm-eabi -float-abi=soft -mattr=+neon %s -o - | FileCheck %s
 
 ; RUN: llc -mtriple=arm-eabi -float-abi=soft -mattr=+neon -regalloc=basic %s -o - \
 ; RUN:  | FileCheck %s
 
 define <8 x i8> @vld1i8(i8* %A) nounwind {
 ;CHECK-LABEL: vld1i8:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vld1.8 {d16}, [r0:64]
-  %tmp1 = call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %A, i32 16)
+  %tmp1 = call <8 x i8> @llvm.arm.neon.vld1.v8i8.p0i8(i8* %A, i32 16)
   ret <8 x i8> %tmp1
 }
 
 define <4 x i16> @vld1i16(i16* %A) nounwind {
 ;CHECK-LABEL: vld1i16:
 ;CHECK: vld1.16
   %tmp0 = bitcast i16* %A to i8*
-  %tmp1 = call <4 x i16> @llvm.arm.neon.vld1.v4i16(i8* %tmp0, i32 1)
+  %tmp1 = call <4 x i16> @llvm.arm.neon.vld1.v4i16.p0i8(i8* %tmp0, i32 1)
   ret <4 x i16> %tmp1
 }
 
 ;Check for a post-increment updating load. 
 define <4 x i16> @vld1i16_update(i16** %ptr) nounwind {
 ;CHECK-LABEL: vld1i16_update:
 ;CHECK: vld1.16 {d16}, [{{r[0-9]+}}]!
   %A = load i16*, i16** %ptr
   %tmp0 = bitcast i16* %A to i8*
-  %tmp1 = call <4 x i16> @llvm.arm.neon.vld1.v4i16(i8* %tmp0, i32 1)
+  %tmp1 = call <4 x i16> @llvm.arm.neon.vld1.v4i16.p0i8(i8* %tmp0, i32 1)
   %tmp2 = getelementptr i16, i16* %A, i32 4
          store i16* %tmp2, i16** %ptr
   ret <4 x i16> %tmp1
 }
 
 define <2 x i32> @vld1i32(i32* %A) nounwind {
 ;CHECK-LABEL: vld1i32:
 ;CHECK: vld1.32
   %tmp0 = bitcast i32* %A to i8*
-  %tmp1 = call <2 x i32> @llvm.arm.neon.vld1.v2i32(i8* %tmp0, i32 1)
+  %tmp1 = call <2 x i32> @llvm.arm.neon.vld1.v2i32.p0i8(i8* %tmp0, i32 1)
   ret <2 x i32> %tmp1
 }
 
 ;Check for a post-increment updating load with register increment.
 define <2 x i32> @vld1i32_update(i32** %ptr, i32 %inc) nounwind {
 ;CHECK-LABEL: vld1i32_update:
 ;CHECK: vld1.32 {d16}, [{{r[0-9]+}}], {{r[0-9]+}}
   %A = load i32*, i32** %ptr
   %tmp0 = bitcast i32* %A to i8*
-  %tmp1 = call <2 x i32> @llvm.arm.neon.vld1.v2i32(i8* %tmp0, i32 1)
+  %tmp1 = call <2 x i32> @llvm.arm.neon.vld1.v2i32.p0i8(i8* %tmp0, i32 1)
   %tmp2 = getelementptr i32, i32* %A, i32 %inc
   store i32* %tmp2, i32** %ptr
   ret <2 x i32> %tmp1
 }
 
 define <2 x float> @vld1f(float* %A) nounwind {
 ;CHECK-LABEL: vld1f:
 ;CHECK: vld1.32
   %tmp0 = bitcast float* %A to i8*
-  %tmp1 = call <2 x float> @llvm.arm.neon.vld1.v2f32(i8* %tmp0, i32 1)
+  %tmp1 = call <2 x float> @llvm.arm.neon.vld1.v2f32.p0i8(i8* %tmp0, i32 1)
   ret <2 x float> %tmp1
 }
 
 define <1 x i64> @vld1i64(i64* %A) nounwind {
 ;CHECK-LABEL: vld1i64:
 ;CHECK: vld1.64
   %tmp0 = bitcast i64* %A to i8*
-  %tmp1 = call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %tmp0, i32 1)
+  %tmp1 = call <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8* %tmp0, i32 1)
   ret <1 x i64> %tmp1
 }
 
 define <16 x i8> @vld1Qi8(i8* %A) nounwind {
 ;CHECK-LABEL: vld1Qi8:
 ;Check the alignment value.  Max for this instruction is 128 bits:
 ;CHECK: vld1.8 {d16, d17}, [r0:64]
-  %tmp1 = call <16 x i8> @llvm.arm.neon.vld1.v16i8(i8* %A, i32 8)
+  %tmp1 = call <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* %A, i32 8)
   ret <16 x i8> %tmp1
 }
 
 ;Check for a post-increment updating load.
 define <16 x i8> @vld1Qi8_update(i8** %ptr) nounwind {
 ;CHECK-LABEL: vld1Qi8_update:
 ;CHECK: vld1.8 {d16, d17}, [{{r[0-9]+}}:64]!
   %A = load i8*, i8** %ptr
-  %tmp1 = call <16 x i8> @llvm.arm.neon.vld1.v16i8(i8* %A, i32 8)
+  %tmp1 = call <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* %A, i32 8)
   %tmp2 = getelementptr i8, i8* %A, i32 16
   store i8* %tmp2, i8** %ptr
   ret <16 x i8> %tmp1
 }
 
 define <8 x i16> @vld1Qi16(i16* %A) nounwind {
 ;CHECK-LABEL: vld1Qi16:
 ;Check the alignment value.  Max for this instruction is 128 bits:
 ;CHECK: vld1.16 {d16, d17}, [r0:128]
   %tmp0 = bitcast i16* %A to i8*
-  %tmp1 = call <8 x i16> @llvm.arm.neon.vld1.v8i16(i8* %tmp0, i32 32)
+  %tmp1 = call <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8* %tmp0, i32 32)
   ret <8 x i16> %tmp1
 }
 
 define <4 x i32> @vld1Qi32(i32* %A) nounwind {
 ;CHECK-LABEL: vld1Qi32:
 ;CHECK: vld1.32
   %tmp0 = bitcast i32* %A to i8*
-  %tmp1 = call <4 x i32> @llvm.arm.neon.vld1.v4i32(i8* %tmp0, i32 1)
+  %tmp1 = call <4 x i32> @llvm.arm.neon.vld1.v4i32.p0i8(i8* %tmp0, i32 1)
   ret <4 x i32> %tmp1
 }
 
 define <4 x float> @vld1Qf(float* %A) nounwind {
 ;CHECK-LABEL: vld1Qf:
 ;CHECK: vld1.32
   %tmp0 = bitcast float* %A to i8*
-  %tmp1 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* %tmp0, i32 1)
+  %tmp1 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* %tmp0, i32 1)
   ret <4 x float> %tmp1
 }
 
 define <2 x i64> @vld1Qi64(i64* %A) nounwind {
 ;CHECK-LABEL: vld1Qi64:
 ;CHECK: vld1.64
   %tmp0 = bitcast i64* %A to i8*
-  %tmp1 = call <2 x i64> @llvm.arm.neon.vld1.v2i64(i8* %tmp0, i32 1)
+  %tmp1 = call <2 x i64> @llvm.arm.neon.vld1.v2i64.p0i8(i8* %tmp0, i32 1)
   ret <2 x i64> %tmp1
 }
 
 define <2 x double> @vld1Qf64(double* %A) nounwind {
 ;CHECK-LABEL: vld1Qf64:
 ;CHECK: vld1.64
   %tmp0 = bitcast double* %A to i8*
-  %tmp1 = call <2 x double> @llvm.arm.neon.vld1.v2f64(i8* %tmp0, i32 1)
+  %tmp1 = call <2 x double> @llvm.arm.neon.vld1.v2f64.p0i8(i8* %tmp0, i32 1)
   ret <2 x double> %tmp1
 }
 
-declare <8 x i8>  @llvm.arm.neon.vld1.v8i8(i8*, i32) nounwind readonly
-declare <4 x i16> @llvm.arm.neon.vld1.v4i16(i8*, i32) nounwind readonly
-declare <2 x i32> @llvm.arm.neon.vld1.v2i32(i8*, i32) nounwind readonly
-declare <2 x float> @llvm.arm.neon.vld1.v2f32(i8*, i32) nounwind readonly
-declare <1 x i64> @llvm.arm.neon.vld1.v1i64(i8*, i32) nounwind readonly
-
-declare <16 x i8> @llvm.arm.neon.vld1.v16i8(i8*, i32) nounwind readonly
-declare <8 x i16> @llvm.arm.neon.vld1.v8i16(i8*, i32) nounwind readonly
-declare <4 x i32> @llvm.arm.neon.vld1.v4i32(i8*, i32) nounwind readonly
-declare <4 x float> @llvm.arm.neon.vld1.v4f32(i8*, i32) nounwind readonly
-declare <2 x i64> @llvm.arm.neon.vld1.v2i64(i8*, i32) nounwind readonly
-declare <2 x double> @llvm.arm.neon.vld1.v2f64(i8*, i32) nounwind readonly
+declare <8 x i8>  @llvm.arm.neon.vld1.v8i8.p0i8(i8*, i32) nounwind readonly
+declare <4 x i16> @llvm.arm.neon.vld1.v4i16.p0i8(i8*, i32) nounwind readonly
+declare <2 x i32> @llvm.arm.neon.vld1.v2i32.p0i8(i8*, i32) nounwind readonly
+declare <2 x float> @llvm.arm.neon.vld1.v2f32.p0i8(i8*, i32) nounwind readonly
+declare <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8*, i32) nounwind readonly
+
+declare <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8*, i32) nounwind readonly
+declare <8 x i16> @llvm.arm.neon.vld1.v8i16.p0i8(i8*, i32) nounwind readonly
+declare <4 x i32> @llvm.arm.neon.vld1.v4i32.p0i8(i8*, i32) nounwind readonly
+declare <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8*, i32) nounwind readonly
+declare <2 x i64> @llvm.arm.neon.vld1.v2i64.p0i8(i8*, i32) nounwind readonly
+declare <2 x double> @llvm.arm.neon.vld1.v2f64.p0i8(i8*, i32) nounwind readonly
 
 ; Radar 8355607
 ; Do not crash if the vld1 result is not used.
 define void @unused_vld1_result() {
 entry:
-  %0 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) 
+  %0 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1)
   call void @llvm.trap()
   unreachable
 }
 
 declare void @llvm.trap() nounwind

llvm/trunk/test/CodeGen/ARM/vld2.ll

 %struct.__neon_int16x8x2_t = type { <8 x i16>, <8 x i16> }
 %struct.__neon_int32x4x2_t = type { <4 x i32>, <4 x i32> }
 %struct.__neon_float32x4x2_t = type { <4 x float>, <4 x float> }
 
 define <8 x i8> @vld2i8(i8* %A) nounwind {
 ;CHECK-LABEL: vld2i8:
 ;Check the alignment value.  Max for this instruction is 128 bits:
 ;CHECK: vld2.8 {d16, d17}, [r0:64]
-  %tmp1 = call %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2.v8i8(i8* %A, i32 8)
+  %tmp1 = call %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2.v8i8.p0i8(i8* %A, i32 8)
         %tmp2 = extractvalue %struct.__neon_int8x8x2_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int8x8x2_t %tmp1, 1
         %tmp4 = add <8 x i8> %tmp2, %tmp3
   ret <8 x i8> %tmp4
 }
 
 define <4 x i16> @vld2i16(i16* %A) nounwind {
 ;CHECK-LABEL: vld2i16:
 ;Check the alignment value.  Max for this instruction is 128 bits:
 ;CHECK: vld2.16 {d16, d17}, [r0:128]
   %tmp0 = bitcast i16* %A to i8*
-  %tmp1 = call %struct.__neon_int16x4x2_t @llvm.arm.neon.vld2.v4i16(i8* %tmp0, i32 32)
+  %tmp1 = call %struct.__neon_int16x4x2_t @llvm.arm.neon.vld2.v4i16.p0i8(i8* %tmp0, i32 32)
         %tmp2 = extractvalue %struct.__neon_int16x4x2_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int16x4x2_t %tmp1, 1
         %tmp4 = add <4 x i16> %tmp2, %tmp3
   ret <4 x i16> %tmp4
 }
 
 define <2 x i32> @vld2i32(i32* %A) nounwind {
 ;CHECK-LABEL: vld2i32:
 ;CHECK: vld2.32
   %tmp0 = bitcast i32* %A to i8*
-  %tmp1 = call %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2.v2i32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2.v2i32.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_int32x2x2_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int32x2x2_t %tmp1, 1
         %tmp4 = add <2 x i32> %tmp2, %tmp3
   ret <2 x i32> %tmp4
 }
 
 define <2 x float> @vld2f(float* %A) nounwind {
 ;CHECK-LABEL: vld2f:
 ;CHECK: vld2.32
   %tmp0 = bitcast float* %A to i8*
-  %tmp1 = call %struct.__neon_float32x2x2_t @llvm.arm.neon.vld2.v2f32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_float32x2x2_t @llvm.arm.neon.vld2.v2f32.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_float32x2x2_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_float32x2x2_t %tmp1, 1
         %tmp4 = fadd <2 x float> %tmp2, %tmp3
   ret <2 x float> %tmp4
 }
 
 ;Check for a post-increment updating load. 
 define <2 x float> @vld2f_update(float** %ptr) nounwind {
 ;CHECK-LABEL: vld2f_update:
 ;CHECK: vld2.32 {d16, d17}, [r1]!
   %A = load float*, float** %ptr
   %tmp0 = bitcast float* %A to i8*
-  %tmp1 = call %struct.__neon_float32x2x2_t @llvm.arm.neon.vld2.v2f32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_float32x2x2_t @llvm.arm.neon.vld2.v2f32.p0i8(i8* %tmp0, i32 1)
   %tmp2 = extractvalue %struct.__neon_float32x2x2_t %tmp1, 0
   %tmp3 = extractvalue %struct.__neon_float32x2x2_t %tmp1, 1
   %tmp4 = fadd <2 x float> %tmp2, %tmp3
   %tmp5 = getelementptr float, float* %A, i32 4
   store float* %tmp5, float** %ptr
   ret <2 x float> %tmp4
 }
 
 define <1 x i64> @vld2i64(i64* %A) nounwind {
 ;CHECK-LABEL: vld2i64:
 ;Check the alignment value.  Max for this instruction is 128 bits:
 ;CHECK: vld1.64 {d16, d17}, [r0:128]
   %tmp0 = bitcast i64* %A to i8*
-  %tmp1 = call %struct.__neon_int64x1x2_t @llvm.arm.neon.vld2.v1i64(i8* %tmp0, i32 32)
+  %tmp1 = call %struct.__neon_int64x1x2_t @llvm.arm.neon.vld2.v1i64.p0i8(i8* %tmp0, i32 32)
         %tmp2 = extractvalue %struct.__neon_int64x1x2_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int64x1x2_t %tmp1, 1
         %tmp4 = add <1 x i64> %tmp2, %tmp3
   ret <1 x i64> %tmp4
 }
 
 define <16 x i8> @vld2Qi8(i8* %A) nounwind {
 ;CHECK-LABEL: vld2Qi8:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vld2.8 {d16, d17, d18, d19}, [r0:64]
-  %tmp1 = call %struct.__neon_int8x16x2_t @llvm.arm.neon.vld2.v16i8(i8* %A, i32 8)
+  %tmp1 = call %struct.__neon_int8x16x2_t @llvm.arm.neon.vld2.v16i8.p0i8(i8* %A, i32 8)
         %tmp2 = extractvalue %struct.__neon_int8x16x2_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int8x16x2_t %tmp1, 1
         %tmp4 = add <16 x i8> %tmp2, %tmp3
   ret <16 x i8> %tmp4
 }
 
 ;Check for a post-increment updating load with register increment.
 define <16 x i8> @vld2Qi8_update(i8** %ptr, i32 %inc) nounwind {
 ;CHECK-LABEL: vld2Qi8_update:
 ;CHECK: vld2.8 {d16, d17, d18, d19}, [r2:128], r1
   %A = load i8*, i8** %ptr
-  %tmp1 = call %struct.__neon_int8x16x2_t @llvm.arm.neon.vld2.v16i8(i8* %A, i32 16)
+  %tmp1 = call %struct.__neon_int8x16x2_t @llvm.arm.neon.vld2.v16i8.p0i8(i8* %A, i32 16)
         %tmp2 = extractvalue %struct.__neon_int8x16x2_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int8x16x2_t %tmp1, 1
         %tmp4 = add <16 x i8> %tmp2, %tmp3
   %tmp5 = getelementptr i8, i8* %A, i32 %inc
   store i8* %tmp5, i8** %ptr
   ret <16 x i8> %tmp4
 }
 
 define <8 x i16> @vld2Qi16(i16* %A) nounwind {
 ;CHECK-LABEL: vld2Qi16:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vld2.16 {d16, d17, d18, d19}, [r0:128]
   %tmp0 = bitcast i16* %A to i8*
-  %tmp1 = call %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2.v8i16(i8* %tmp0, i32 16)
+  %tmp1 = call %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2.v8i16.p0i8(i8* %tmp0, i32 16)
         %tmp2 = extractvalue %struct.__neon_int16x8x2_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int16x8x2_t %tmp1, 1
         %tmp4 = add <8 x i16> %tmp2, %tmp3
   ret <8 x i16> %tmp4
 }
 
 define <4 x i32> @vld2Qi32(i32* %A) nounwind {
 ;CHECK-LABEL: vld2Qi32:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vld2.32 {d16, d17, d18, d19}, [r0:256]
   %tmp0 = bitcast i32* %A to i8*
-  %tmp1 = call %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32(i8* %tmp0, i32 64)
+  %tmp1 = call %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32.p0i8(i8* %tmp0, i32 64)
         %tmp2 = extractvalue %struct.__neon_int32x4x2_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int32x4x2_t %tmp1, 1
         %tmp4 = add <4 x i32> %tmp2, %tmp3
   ret <4 x i32> %tmp4
 }
 
 define <4 x float> @vld2Qf(float* %A) nounwind {
 ;CHECK-LABEL: vld2Qf:
 ;CHECK: vld2.32
   %tmp0 = bitcast float* %A to i8*
-  %tmp1 = call %struct.__neon_float32x4x2_t @llvm.arm.neon.vld2.v4f32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_float32x4x2_t @llvm.arm.neon.vld2.v4f32.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_float32x4x2_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_float32x4x2_t %tmp1, 1
         %tmp4 = fadd <4 x float> %tmp2, %tmp3
   ret <4 x float> %tmp4
 }
 
-declare %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2.v8i8(i8*, i32) nounwind readonly
-declare %struct.__neon_int16x4x2_t @llvm.arm.neon.vld2.v4i16(i8*, i32) nounwind readonly
-declare %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2.v2i32(i8*, i32) nounwind readonly
-declare %struct.__neon_float32x2x2_t @llvm.arm.neon.vld2.v2f32(i8*, i32) nounwind readonly
-declare %struct.__neon_int64x1x2_t @llvm.arm.neon.vld2.v1i64(i8*, i32) nounwind readonly
-
-declare %struct.__neon_int8x16x2_t @llvm.arm.neon.vld2.v16i8(i8*, i32) nounwind readonly
-declare %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2.v8i16(i8*, i32) nounwind readonly
-declare %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32(i8*, i32) nounwind readonly
-declare %struct.__neon_float32x4x2_t @llvm.arm.neon.vld2.v4f32(i8*, i32) nounwind readonly
+declare %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2.v8i8.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int16x4x2_t @llvm.arm.neon.vld2.v4i16.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2.v2i32.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_float32x2x2_t @llvm.arm.neon.vld2.v2f32.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int64x1x2_t @llvm.arm.neon.vld2.v1i64.p0i8(i8*, i32) nounwind readonly
+
+declare %struct.__neon_int8x16x2_t @llvm.arm.neon.vld2.v16i8.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2.v8i16.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2.v4i32.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_float32x4x2_t @llvm.arm.neon.vld2.v4f32.p0i8(i8*, i32) nounwind readonly

llvm/trunk/test/CodeGen/ARM/vld3.ll

 %struct.__neon_int16x8x3_t = type { <8 x i16>, <8 x i16>, <8 x i16> }
 %struct.__neon_int32x4x3_t = type { <4 x i32>, <4 x i32>, <4 x i32> }
 %struct.__neon_float32x4x3_t = type { <4 x float>, <4 x float>, <4 x float> }
 
 define <8 x i8> @vld3i8(i8* %A) nounwind {
 ;CHECK-LABEL: vld3i8:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vld3.8 {d16, d17, d18}, [r0:64]
-  %tmp1 = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A, i32 32)
+  %tmp1 = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8* %A, i32 32)
         %tmp2 = extractvalue %struct.__neon_int8x8x3_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int8x8x3_t %tmp1, 2
         %tmp4 = add <8 x i8> %tmp2, %tmp3
   ret <8 x i8> %tmp4
 }
 
 define <4 x i16> @vld3i16(i16* %A) nounwind {
 ;CHECK-LABEL: vld3i16:
 ;CHECK: vld3.16
   %tmp0 = bitcast i16* %A to i8*
-  %tmp1 = call %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3.v4i16(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3.v4i16.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_int16x4x3_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int16x4x3_t %tmp1, 2
         %tmp4 = add <4 x i16> %tmp2, %tmp3
   ret <4 x i16> %tmp4
 }
 
 ;Check for a post-increment updating load with register increment.
 define <4 x i16> @vld3i16_update(i16** %ptr, i32 %inc) nounwind {
 ;CHECK-LABEL: vld3i16_update:
 ;CHECK: vld3.16 {d16, d17, d18}, [{{r[0-9]+}}], {{r[0-9]+}}
   %A = load i16*, i16** %ptr
   %tmp0 = bitcast i16* %A to i8*
-  %tmp1 = call %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3.v4i16(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3.v4i16.p0i8(i8* %tmp0, i32 1)
   %tmp2 = extractvalue %struct.__neon_int16x4x3_t %tmp1, 0
   %tmp3 = extractvalue %struct.__neon_int16x4x3_t %tmp1, 2
   %tmp4 = add <4 x i16> %tmp2, %tmp3
   %tmp5 = getelementptr i16, i16* %A, i32 %inc
   store i16* %tmp5, i16** %ptr
   ret <4 x i16> %tmp4
 }
 
 define <2 x i32> @vld3i32(i32* %A) nounwind {
 ;CHECK-LABEL: vld3i32:
 ;CHECK: vld3.32
   %tmp0 = bitcast i32* %A to i8*
-  %tmp1 = call %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3.v2i32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3.v2i32.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_int32x2x3_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int32x2x3_t %tmp1, 2
         %tmp4 = add <2 x i32> %tmp2, %tmp3
   ret <2 x i32> %tmp4
 }
 
 define <2 x float> @vld3f(float* %A) nounwind {
 ;CHECK-LABEL: vld3f:
 ;CHECK: vld3.32
   %tmp0 = bitcast float* %A to i8*
-  %tmp1 = call %struct.__neon_float32x2x3_t @llvm.arm.neon.vld3.v2f32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_float32x2x3_t @llvm.arm.neon.vld3.v2f32.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_float32x2x3_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_float32x2x3_t %tmp1, 2
         %tmp4 = fadd <2 x float> %tmp2, %tmp3
   ret <2 x float> %tmp4
 }
 
 define <1 x i64> @vld3i64(i64* %A) nounwind {
 ;CHECK-LABEL: vld3i64:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vld1.64 {d16, d17, d18}, [r0:64]
   %tmp0 = bitcast i64* %A to i8*
-  %tmp1 = call %struct.__neon_int64x1x3_t @llvm.arm.neon.vld3.v1i64(i8* %tmp0, i32 16)
+  %tmp1 = call %struct.__neon_int64x1x3_t @llvm.arm.neon.vld3.v1i64.p0i8(i8* %tmp0, i32 16)
         %tmp2 = extractvalue %struct.__neon_int64x1x3_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int64x1x3_t %tmp1, 2
         %tmp4 = add <1 x i64> %tmp2, %tmp3
   ret <1 x i64> %tmp4
 }
 
 define <1 x i64> @vld3i64_update(i64** %ptr, i64* %A) nounwind {
 ;CHECK-LABEL: vld3i64_update:
 ;CHECK: vld1.64 {d16, d17, d18}, [r1:64]!
         %tmp0 = bitcast i64* %A to i8*
-        %tmp1 = call %struct.__neon_int64x1x3_t @llvm.arm.neon.vld3.v1i64(i8* %tmp0, i32 16)
+        %tmp1 = call %struct.__neon_int64x1x3_t @llvm.arm.neon.vld3.v1i64.p0i8(i8* %tmp0, i32 16)
         %tmp5 = getelementptr i64, i64* %A, i32 3
         store i64* %tmp5, i64** %ptr
         %tmp2 = extractvalue %struct.__neon_int64x1x3_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int64x1x3_t %tmp1, 2
         %tmp4 = add <1 x i64> %tmp2, %tmp3
         ret <1 x i64> %tmp4
 }
 
 define <16 x i8> @vld3Qi8(i8* %A) nounwind {
 ;CHECK-LABEL: vld3Qi8:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vld3.8 {d16, d18, d20}, [r0:64]!
 ;CHECK: vld3.8 {d17, d19, d21}, [r0:64]
-  %tmp1 = call %struct.__neon_int8x16x3_t @llvm.arm.neon.vld3.v16i8(i8* %A, i32 32)
+  %tmp1 = call %struct.__neon_int8x16x3_t @llvm.arm.neon.vld3.v16i8.p0i8(i8* %A, i32 32)
         %tmp2 = extractvalue %struct.__neon_int8x16x3_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int8x16x3_t %tmp1, 2
         %tmp4 = add <16 x i8> %tmp2, %tmp3
   ret <16 x i8> %tmp4
 }
 
 define <8 x i16> @vld3Qi16(i16* %A) nounwind {
 ;CHECK-LABEL: vld3Qi16:
 ;CHECK: vld3.16
 ;CHECK: vld3.16
   %tmp0 = bitcast i16* %A to i8*
-  %tmp1 = call %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3.v8i16(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3.v8i16.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_int16x8x3_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int16x8x3_t %tmp1, 2
         %tmp4 = add <8 x i16> %tmp2, %tmp3
   ret <8 x i16> %tmp4
 }
 
 define <4 x i32> @vld3Qi32(i32* %A) nounwind {
 ;CHECK-LABEL: vld3Qi32:
 ;CHECK: vld3.32
 ;CHECK: vld3.32
   %tmp0 = bitcast i32* %A to i8*
-  %tmp1 = call %struct.__neon_int32x4x3_t @llvm.arm.neon.vld3.v4i32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_int32x4x3_t @llvm.arm.neon.vld3.v4i32.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_int32x4x3_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int32x4x3_t %tmp1, 2
         %tmp4 = add <4 x i32> %tmp2, %tmp3
   ret <4 x i32> %tmp4
 }
 
 ;Check for a post-increment updating load. 
 define <4 x i32> @vld3Qi32_update(i32** %ptr) nounwind {
 ;CHECK-LABEL: vld3Qi32_update:
 ;CHECK: vld3.32 {d16, d18, d20}, [r[[R:[0-9]+]]]!
 ;CHECK: vld3.32 {d17, d19, d21}, [r[[R]]]!
   %A = load i32*, i32** %ptr
   %tmp0 = bitcast i32* %A to i8*
-  %tmp1 = call %struct.__neon_int32x4x3_t @llvm.arm.neon.vld3.v4i32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_int32x4x3_t @llvm.arm.neon.vld3.v4i32.p0i8(i8* %tmp0, i32 1)
   %tmp2 = extractvalue %struct.__neon_int32x4x3_t %tmp1, 0
   %tmp3 = extractvalue %struct.__neon_int32x4x3_t %tmp1, 2
   %tmp4 = add <4 x i32> %tmp2, %tmp3
   %tmp5 = getelementptr i32, i32* %A, i32 12
   store i32* %tmp5, i32** %ptr
   ret <4 x i32> %tmp4
 }
 
 define <4 x float> @vld3Qf(float* %A) nounwind {
 ;CHECK-LABEL: vld3Qf:
 ;CHECK: vld3.32
 ;CHECK: vld3.32
   %tmp0 = bitcast float* %A to i8*
-  %tmp1 = call %struct.__neon_float32x4x3_t @llvm.arm.neon.vld3.v4f32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_float32x4x3_t @llvm.arm.neon.vld3.v4f32.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_float32x4x3_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_float32x4x3_t %tmp1, 2
         %tmp4 = fadd <4 x float> %tmp2, %tmp3
   ret <4 x float> %tmp4
 }
 
-declare %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8*, i32) nounwind readonly
-declare %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3.v4i16(i8*, i32) nounwind readonly
-declare %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3.v2i32(i8*, i32) nounwind readonly
-declare %struct.__neon_float32x2x3_t @llvm.arm.neon.vld3.v2f32(i8*, i32) nounwind readonly
-declare %struct.__neon_int64x1x3_t @llvm.arm.neon.vld3.v1i64(i8*, i32) nounwind readonly
-
-declare %struct.__neon_int8x16x3_t @llvm.arm.neon.vld3.v16i8(i8*, i32) nounwind readonly
-declare %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3.v8i16(i8*, i32) nounwind readonly
-declare %struct.__neon_int32x4x3_t @llvm.arm.neon.vld3.v4i32(i8*, i32) nounwind readonly
-declare %struct.__neon_float32x4x3_t @llvm.arm.neon.vld3.v4f32(i8*, i32) nounwind readonly
+declare %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3.v4i16.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3.v2i32.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_float32x2x3_t @llvm.arm.neon.vld3.v2f32.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int64x1x3_t @llvm.arm.neon.vld3.v1i64.p0i8(i8*, i32) nounwind readonly
+
+declare %struct.__neon_int8x16x3_t @llvm.arm.neon.vld3.v16i8.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3.v8i16.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int32x4x3_t @llvm.arm.neon.vld3.v4i32.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_float32x4x3_t @llvm.arm.neon.vld3.v4f32.p0i8(i8*, i32) nounwind readonly

llvm/trunk/test/CodeGen/ARM/vld4.ll

 %struct.__neon_int16x8x4_t = type { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> }
 %struct.__neon_int32x4x4_t = type { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> }
 %struct.__neon_float32x4x4_t = type { <4 x float>, <4 x float>, <4 x float>, <4 x float> }
 
 define <8 x i8> @vld4i8(i8* %A) nounwind {
 ;CHECK-LABEL: vld4i8:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vld4.8 {d16, d17, d18, d19}, [r0:64]
-  %tmp1 = call %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4.v8i8(i8* %A, i32 8)
+  %tmp1 = call %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4.v8i8.p0i8(i8* %A, i32 8)
         %tmp2 = extractvalue %struct.__neon_int8x8x4_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int8x8x4_t %tmp1, 2
         %tmp4 = add <8 x i8> %tmp2, %tmp3
   ret <8 x i8> %tmp4
 }
 
 ;Check for a post-increment updating load with register increment.
 define <8 x i8> @vld4i8_update(i8** %ptr, i32 %inc) nounwind {
 ;CHECK-LABEL: vld4i8_update:
 ;CHECK: vld4.8 {d16, d17, d18, d19}, [r2:128], r1
   %A = load i8*, i8** %ptr
-  %tmp1 = call %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4.v8i8(i8* %A, i32 16)
+  %tmp1 = call %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4.v8i8.p0i8(i8* %A, i32 16)
   %tmp2 = extractvalue %struct.__neon_int8x8x4_t %tmp1, 0
   %tmp3 = extractvalue %struct.__neon_int8x8x4_t %tmp1, 2
   %tmp4 = add <8 x i8> %tmp2, %tmp3
   %tmp5 = getelementptr i8, i8* %A, i32 %inc
   store i8* %tmp5, i8** %ptr
   ret <8 x i8> %tmp4
 }
 
 define <4 x i16> @vld4i16(i16* %A) nounwind {
 ;CHECK-LABEL: vld4i16:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vld4.16 {d16, d17, d18, d19}, [r0:128]
   %tmp0 = bitcast i16* %A to i8*
-  %tmp1 = call %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4.v4i16(i8* %tmp0, i32 16)
+  %tmp1 = call %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4.v4i16.p0i8(i8* %tmp0, i32 16)
         %tmp2 = extractvalue %struct.__neon_int16x4x4_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int16x4x4_t %tmp1, 2
         %tmp4 = add <4 x i16> %tmp2, %tmp3
   ret <4 x i16> %tmp4
 }
 
 define <2 x i32> @vld4i32(i32* %A) nounwind {
 ;CHECK-LABEL: vld4i32:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vld4.32 {d16, d17, d18, d19}, [r0:256]
   %tmp0 = bitcast i32* %A to i8*
-  %tmp1 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8* %tmp0, i32 32)
+  %tmp1 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32.p0i8(i8* %tmp0, i32 32)
         %tmp2 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 2
         %tmp4 = add <2 x i32> %tmp2, %tmp3
   ret <2 x i32> %tmp4
 }
 
 define <2 x float> @vld4f(float* %A) nounwind {
 ;CHECK-LABEL: vld4f:
 ;CHECK: vld4.32
   %tmp0 = bitcast float* %A to i8*
-  %tmp1 = call %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4.v2f32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4.v2f32.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_float32x2x4_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_float32x2x4_t %tmp1, 2
         %tmp4 = fadd <2 x float> %tmp2, %tmp3
   ret <2 x float> %tmp4
 }
 
 define <1 x i64> @vld4i64(i64* %A) nounwind {
 ;CHECK-LABEL: vld4i64:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vld1.64 {d16, d17, d18, d19}, [r0:256]
   %tmp0 = bitcast i64* %A to i8*
-  %tmp1 = call %struct.__neon_int64x1x4_t @llvm.arm.neon.vld4.v1i64(i8* %tmp0, i32 64)
+  %tmp1 = call %struct.__neon_int64x1x4_t @llvm.arm.neon.vld4.v1i64.p0i8(i8* %tmp0, i32 64)
         %tmp2 = extractvalue %struct.__neon_int64x1x4_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int64x1x4_t %tmp1, 2
         %tmp4 = add <1 x i64> %tmp2, %tmp3
   ret <1 x i64> %tmp4
 }
 
 define <1 x i64> @vld4i64_update(i64** %ptr, i64* %A) nounwind {
 ;CHECK-LABEL: vld4i64_update:
 ;CHECK: vld1.64 {d16, d17, d18, d19}, [r1:256]!
         %tmp0 = bitcast i64* %A to i8*
-        %tmp1 = call %struct.__neon_int64x1x4_t @llvm.arm.neon.vld4.v1i64(i8* %tmp0, i32 64)
+        %tmp1 = call %struct.__neon_int64x1x4_t @llvm.arm.neon.vld4.v1i64.p0i8(i8* %tmp0, i32 64)
         %tmp5 = getelementptr i64, i64* %A, i32 4
         store i64* %tmp5, i64** %ptr
         %tmp2 = extractvalue %struct.__neon_int64x1x4_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int64x1x4_t %tmp1, 2
         %tmp4 = add <1 x i64> %tmp2, %tmp3
         ret <1 x i64> %tmp4
 }
 
 define <16 x i8> @vld4Qi8(i8* %A) nounwind {
 ;CHECK-LABEL: vld4Qi8:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vld4.8 {d16, d18, d20, d22}, [r0:256]!
 ;CHECK: vld4.8 {d17, d19, d21, d23}, [r0:256]
-  %tmp1 = call %struct.__neon_int8x16x4_t @llvm.arm.neon.vld4.v16i8(i8* %A, i32 64)
+  %tmp1 = call %struct.__neon_int8x16x4_t @llvm.arm.neon.vld4.v16i8.p0i8(i8* %A, i32 64)
         %tmp2 = extractvalue %struct.__neon_int8x16x4_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int8x16x4_t %tmp1, 2
         %tmp4 = add <16 x i8> %tmp2, %tmp3
   ret <16 x i8> %tmp4
 }
 
 define <8 x i16> @vld4Qi16(i16* %A) nounwind {
 ;CHECK-LABEL: vld4Qi16:
 ;Check for no alignment specifier.
 ;CHECK: vld4.16 {d16, d18, d20, d22}, [r0]!
 ;CHECK: vld4.16 {d17, d19, d21, d23}, [r0]
   %tmp0 = bitcast i16* %A to i8*
-  %tmp1 = call %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4.v8i16(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4.v8i16.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_int16x8x4_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int16x8x4_t %tmp1, 2
         %tmp4 = add <8 x i16> %tmp2, %tmp3
   ret <8 x i16> %tmp4
 }
 
 ;Check for a post-increment updating load. 
 define <8 x i16> @vld4Qi16_update(i16** %ptr) nounwind {
 ;CHECK-LABEL: vld4Qi16_update:
 ;CHECK: vld4.16 {d16, d18, d20, d22}, [r1:64]!
 ;CHECK: vld4.16 {d17, d19, d21, d23}, [r1:64]!
   %A = load i16*, i16** %ptr
   %tmp0 = bitcast i16* %A to i8*
-  %tmp1 = call %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4.v8i16(i8* %tmp0, i32 8)
+  %tmp1 = call %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4.v8i16.p0i8(i8* %tmp0, i32 8)
   %tmp2 = extractvalue %struct.__neon_int16x8x4_t %tmp1, 0
   %tmp3 = extractvalue %struct.__neon_int16x8x4_t %tmp1, 2
   %tmp4 = add <8 x i16> %tmp2, %tmp3
   %tmp5 = getelementptr i16, i16* %A, i32 32
   store i16* %tmp5, i16** %ptr
   ret <8 x i16> %tmp4
 }
 
 define <4 x i32> @vld4Qi32(i32* %A) nounwind {
 ;CHECK-LABEL: vld4Qi32:
 ;CHECK: vld4.32
 ;CHECK: vld4.32
   %tmp0 = bitcast i32* %A to i8*
-  %tmp1 = call %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4.v4i32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4.v4i32.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_int32x4x4_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_int32x4x4_t %tmp1, 2
         %tmp4 = add <4 x i32> %tmp2, %tmp3
   ret <4 x i32> %tmp4
 }
 
 define <4 x float> @vld4Qf(float* %A) nounwind {
 ;CHECK-LABEL: vld4Qf:
 ;CHECK: vld4.32
 ;CHECK: vld4.32
   %tmp0 = bitcast float* %A to i8*
-  %tmp1 = call %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4.v4f32(i8* %tmp0, i32 1)
+  %tmp1 = call %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4.v4f32.p0i8(i8* %tmp0, i32 1)
         %tmp2 = extractvalue %struct.__neon_float32x4x4_t %tmp1, 0
         %tmp3 = extractvalue %struct.__neon_float32x4x4_t %tmp1, 2
         %tmp4 = fadd <4 x float> %tmp2, %tmp3
   ret <4 x float> %tmp4
 }
 
-declare %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4.v8i8(i8*, i32) nounwind readonly
-declare %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4.v4i16(i8*, i32) nounwind readonly
-declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8*, i32) nounwind readonly
-declare %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4.v2f32(i8*, i32) nounwind readonly
-declare %struct.__neon_int64x1x4_t @llvm.arm.neon.vld4.v1i64(i8*, i32) nounwind readonly
-
-declare %struct.__neon_int8x16x4_t @llvm.arm.neon.vld4.v16i8(i8*, i32) nounwind readonly
-declare %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4.v8i16(i8*, i32) nounwind readonly
-declare %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4.v4i32(i8*, i32) nounwind readonly
-declare %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4.v4f32(i8*, i32) nounwind readonly
+declare %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4.v8i8.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4.v4i16.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4.v2f32.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int64x1x4_t @llvm.arm.neon.vld4.v1i64.p0i8(i8*, i32) nounwind readonly
+
+declare %struct.__neon_int8x16x4_t @llvm.arm.neon.vld4.v16i8.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4.v8i16.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4.v4i32.p0i8(i8*, i32) nounwind readonly
+declare %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4.v4f32.p0i8(i8*, i32) nounwind readonly

llvm/trunk/test/CodeGen/ARM/vlddup.ll

 %struct.__neon_int8x8x2_t = type { <8 x i8>, <8 x i8> }
 %struct.__neon_int4x16x2_t = type { <4 x i16>, <4 x i16> }
 %struct.__neon_int2x32x2_t = type { <2 x i32>, <2 x i32> }
 
 define <8 x i8> @vld2dupi8(i8* %A) nounwind {
 ;CHECK-LABEL: vld2dupi8:
 ;Check the (default) alignment value.
 ;CHECK: vld2.8 {d16[], d17[]}, [r0]
-  %tmp0 = tail call %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8(i8* %A, <8 x i8> undef, <8 x i8> undef, i32 0, i32 1)
+  %tmp0 = tail call %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8.p0i8(i8* %A, <8 x i8> undef, <8 x i8> undef, i32 0, i32 1)
   %tmp1 = extractvalue %struct.__neon_int8x8x2_t %tmp0, 0
   %tmp2 = shufflevector <8 x i8> %tmp1, <8 x i8> undef, <8 x i32> zeroinitializer
   %tmp3 = extractvalue %struct.__neon_int8x8x2_t %tmp0, 1
   %tmp4 = shufflevector <8 x i8> %tmp3, <8 x i8> undef, <8 x i32> zeroinitializer
         %tmp5 = add <8 x i8> %tmp2, %tmp4
         ret <8 x i8> %tmp5
 }
 
 define <4 x i16> @vld2dupi16(i8* %A) nounwind {
 ;CHECK-LABEL: vld2dupi16:
 ;Check that a power-of-two alignment smaller than the total size of the memory
 ;being loaded is ignored.
 ;CHECK: vld2.16 {d16[], d17[]}, [r0]
-  %tmp0 = tail call %struct.__neon_int4x16x2_t @llvm.arm.neon.vld2lane.v4i16(i8* %A, <4 x i16> undef, <4 x i16> undef, i32 0, i32 2)
+  %tmp0 = tail call %struct.__neon_int4x16x2_t @llvm.arm.neon.vld2lane.v4i16.p0i8(i8* %A, <4 x i16> undef, <4 x i16> undef, i32 0, i32 2)
   %tmp1 = extractvalue %struct.__neon_int4x16x2_t %tmp0, 0
   %tmp2 = shufflevector <4 x i16> %tmp1, <4 x i16> undef, <4 x i32> zeroinitializer
   %tmp3 = extractvalue %struct.__neon_int4x16x2_t %tmp0, 1
   %tmp4 = shufflevector <4 x i16> %tmp3, <4 x i16> undef, <4 x i32> zeroinitializer
         %tmp5 = add <4 x i16> %tmp2, %tmp4
         ret <4 x i16> %tmp5
 }
 
 ;Check for a post-increment updating load. 
 define <4 x i16> @vld2dupi16_update(i16** %ptr) nounwind {
 ;CHECK-LABEL: vld2dupi16_update:
 ;CHECK: vld2.16 {d16[], d17[]}, [r1]!
   %A = load i16*, i16** %ptr
         %A2 = bitcast i16* %A to i8*
-  %tmp0 = tail call %struct.__neon_int4x16x2_t @llvm.arm.neon.vld2lane.v4i16(i8* %A2, <4 x i16> undef, <4 x i16> undef, i32 0, i32 2)
+  %tmp0 = tail call %struct.__neon_int4x16x2_t @llvm.arm.neon.vld2lane.v4i16.p0i8(i8* %A2, <4 x i16> undef, <4 x i16> undef, i32 0, i32 2)
   %tmp1 = extractvalue %struct.__neon_int4x16x2_t %tmp0, 0
   %tmp2 = shufflevector <4 x i16> %tmp1, <4 x i16> undef, <4 x i32> zeroinitializer
   %tmp3 = extractvalue %struct.__neon_int4x16x2_t %tmp0, 1
   %tmp4 = shufflevector <4 x i16> %tmp3, <4 x i16> undef, <4 x i32> zeroinitializer
   %tmp5 = add <4 x i16> %tmp2, %tmp4
   %tmp6 = getelementptr i16, i16* %A, i32 2
   store i16* %tmp6, i16** %ptr
   ret <4 x i16> %tmp5
 }
 
 define <2 x i32> @vld2dupi32(i8* %A) nounwind {
 ;CHECK-LABEL: vld2dupi32:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vld2.32 {d16[], d17[]}, [r0:64]
-  %tmp0 = tail call %struct.__neon_int2x32x2_t @llvm.arm.neon.vld2lane.v2i32(i8* %A, <2 x i32> undef, <2 x i32> undef, i32 0, i32 16)
+  %tmp0 = tail call %struct.__neon_int2x32x2_t @llvm.arm.neon.vld2lane.v2i32.p0i8(i8* %A, <2 x i32> undef, <2 x i32> undef, i32 0, i32 16)
   %tmp1 = extractvalue %struct.__neon_int2x32x2_t %tmp0, 0
   %tmp2 = shufflevector <2 x i32> %tmp1, <2 x i32> undef, <2 x i32> zeroinitializer
   %tmp3 = extractvalue %struct.__neon_int2x32x2_t %tmp0, 1
   %tmp4 = shufflevector <2 x i32> %tmp3, <2 x i32> undef, <2 x i32> zeroinitializer
         %tmp5 = add <2 x i32> %tmp2, %tmp4
         ret <2 x i32> %tmp5
 }
 
-declare %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8(i8*, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
-declare %struct.__neon_int4x16x2_t @llvm.arm.neon.vld2lane.v4i16(i8*, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
-declare %struct.__neon_int2x32x2_t @llvm.arm.neon.vld2lane.v2i32(i8*, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
+declare %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8.p0i8(i8*, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
+declare %struct.__neon_int4x16x2_t @llvm.arm.neon.vld2lane.v4i16.p0i8(i8*, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
+declare %struct.__neon_int2x32x2_t @llvm.arm.neon.vld2lane.v2i32.p0i8(i8*, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
 
 %struct.__neon_int8x8x3_t = type { <8 x i8>, <8 x i8>, <8 x i8> }
 %struct.__neon_int16x4x3_t = type { <4 x i16>, <4 x i16>, <4 x i16> }
 
 ;Check for a post-increment updating load with register increment.
 define <8 x i8> @vld3dupi8_update(i8** %ptr, i32 %inc) nounwind {
 ;CHECK-LABEL: vld3dupi8_update:
 ;CHECK: vld3.8 {d16[], d17[], d18[]}, [r2], r1
   %A = load i8*, i8** %ptr
-  %tmp0 = tail call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3lane.v8i8(i8* %A, <8 x i8> undef, <8 x i8> undef, <8 x i8> undef, i32 0, i32 8)
+  %tmp0 = tail call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3lane.v8i8.p0i8(i8* %A, <8 x i8> undef, <8 x i8> undef, <8 x i8> undef, i32 0, i32 8)
   %tmp1 = extractvalue %struct.__neon_int8x8x3_t %tmp0, 0
   %tmp2 = shufflevector <8 x i8> %tmp1, <8 x i8> undef, <8 x i32> zeroinitializer
   %tmp3 = extractvalue %struct.__neon_int8x8x3_t %tmp0, 1
   %tmp4 = shufflevector <8 x i8> %tmp3, <8 x i8> undef, <8 x i32> zeroinitializer
   %tmp5 = extractvalue %struct.__neon_int8x8x3_t %tmp0, 2
   %tmp6 = shufflevector <8 x i8> %tmp5, <8 x i8> undef, <8 x i32> zeroinitializer
   %tmp7 = add <8 x i8> %tmp2, %tmp4
   %tmp8 = add <8 x i8> %tmp7, %tmp6
   %tmp9 = getelementptr i8, i8* %A, i32 %inc
   store i8* %tmp9, i8** %ptr
   ret <8 x i8> %tmp8
 }
 
 define <4 x i16> @vld3dupi16(i8* %A) nounwind {
 ;CHECK-LABEL: vld3dupi16:
 ;Check the (default) alignment value. VLD3 does not support alignment.
 ;CHECK: vld3.16 {d16[], d17[], d18[]}, [r0]
-  %tmp0 = tail call %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3lane.v4i16(i8* %A, <4 x i16> undef, <4 x i16> undef, <4 x i16> undef, i32 0, i32 8)
+  %tmp0 = tail call %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3lane.v4i16.p0i8(i8* %A, <4 x i16> undef, <4 x i16> undef, <4 x i16> undef, i32 0, i32 8)
   %tmp1 = extractvalue %struct.__neon_int16x4x3_t %tmp0, 0
   %tmp2 = shufflevector <4 x i16> %tmp1, <4 x i16> undef, <4 x i32> zeroinitializer
   %tmp3 = extractvalue %struct.__neon_int16x4x3_t %tmp0, 1
   %tmp4 = shufflevector <4 x i16> %tmp3, <4 x i16> undef, <4 x i32> zeroinitializer
   %tmp5 = extractvalue %struct.__neon_int16x4x3_t %tmp0, 2
   %tmp6 = shufflevector <4 x i16> %tmp5, <4 x i16> undef, <4 x i32> zeroinitializer
         %tmp7 = add <4 x i16> %tmp2, %tmp4
         %tmp8 = add <4 x i16> %tmp7, %tmp6
         ret <4 x i16> %tmp8
 }
 
-declare %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3lane.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
-declare %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3lane.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
+declare %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3lane.v8i8.p0i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
+declare %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3lane.v4i16.p0i8(i8*, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
 
 %struct.__neon_int16x4x4_t = type { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> }
 %struct.__neon_int32x2x4_t = type { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> }
 
 ;Check for a post-increment updating load.
 define <4 x i16> @vld4dupi16_update(i16** %ptr) nounwind {
 ;CHECK-LABEL: vld4dupi16_update:
 ;CHECK: vld4.16 {d16[], d17[], d18[], d19[]}, [r1]!
   %A = load i16*, i16** %ptr
         %A2 = bitcast i16* %A to i8*
-  %tmp0 = tail call %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4lane.v4i16(i8* %A2, <4 x i16> undef, <4 x i16> undef, <4 x i16> undef, <4 x i16> undef, i32 0, i32 1)
+  %tmp0 = tail call %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4lane.v4i16.p0i8(i8* %A2, <4 x i16> undef, <4 x i16> undef, <4 x i16> undef, <4 x i16> undef, i32 0, i32 1)
   %tmp1 = extractvalue %struct.__neon_int16x4x4_t %tmp0, 0
   %tmp2 = shufflevector <4 x i16> %tmp1, <4 x i16> undef, <4 x i32> zeroinitializer
   %tmp3 = extractvalue %struct.__neon_int16x4x4_t %tmp0, 1
   %tmp4 = shufflevector <4 x i16> %tmp3, <4 x i16> undef, <4 x i32> zeroinitializer
   %tmp5 = extractvalue %struct.__neon_int16x4x4_t %tmp0, 2
   %tmp6 = shufflevector <4 x i16> %tmp5, <4 x i16> undef, <4 x i32> zeroinitializer
   %tmp7 = extractvalue %struct.__neon_int16x4x4_t %tmp0, 3
   %tmp8 = shufflevector <4 x i16> %tmp7, <4 x i16> undef, <4 x i32> zeroinitializer
   %tmp9 = add <4 x i16> %tmp2, %tmp4
   %tmp10 = add <4 x i16> %tmp6, %tmp8
   %tmp11 = add <4 x i16> %tmp9, %tmp10
   %tmp12 = getelementptr i16, i16* %A, i32 4
   store i16* %tmp12, i16** %ptr
   ret <4 x i16> %tmp11
 }
 
 define <2 x i32> @vld4dupi32(i8* %A) nounwind {
 ;CHECK-LABEL: vld4dupi32:
 ;Check the alignment value.  An 8-byte alignment is allowed here even though
 ;it is smaller than the total size of the memory being loaded.
 ;CHECK: vld4.32 {d16[], d17[], d18[], d19[]}, [r0:64]
-  %tmp0 = tail call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4lane.v2i32(i8* %A, <2 x i32> undef, <2 x i32> undef, <2 x i32> undef, <2 x i32> undef, i32 0, i32 8)
+  %tmp0 = tail call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4lane.v2i32.p0i8(i8* %A, <2 x i32> undef, <2 x i32> undef, <2 x i32> undef, <2 x i32> undef, i32 0, i32 8)
   %tmp1 = extractvalue %struct.__neon_int32x2x4_t %tmp0, 0
   %tmp2 = shufflevector <2 x i32> %tmp1, <2 x i32> undef, <2 x i32> zeroinitializer
   %tmp3 = extractvalue %struct.__neon_int32x2x4_t %tmp0, 1
   %tmp4 = shufflevector <2 x i32> %tmp3, <2 x i32> undef, <2 x i32> zeroinitializer
   %tmp5 = extractvalue %struct.__neon_int32x2x4_t %tmp0, 2
   %tmp6 = shufflevector <2 x i32> %tmp5, <2 x i32> undef, <2 x i32> zeroinitializer
   %tmp7 = extractvalue %struct.__neon_int32x2x4_t %tmp0, 3
   %tmp8 = shufflevector <2 x i32> %tmp7, <2 x i32> undef, <2 x i32> zeroinitializer
         %tmp9 = add <2 x i32> %tmp2, %tmp4
         %tmp10 = add <2 x i32> %tmp6, %tmp8
         %tmp11 = add <2 x i32> %tmp9, %tmp10
         ret <2 x i32> %tmp11
 }
 
-declare %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4lane.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
-declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
+declare %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4lane.v4i16.p0i8(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
+declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4lane.v2i32.p0i8(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly

llvm/trunk/test/CodeGen/ARM/vldlane.ll

 %struct.__neon_int32x4x2_t = type { <4 x i32>, <4 x i32> }
 %struct.__neon_float32x4x2_t = type { <4 x float>, <4 x float> }
 
 define <8 x i8> @vld2lanei8(i8* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vld2lanei8:
 ;Check the alignment value.  Max for this instruction is 16 bits:
 ;CHECK: vld2.8 {d16[1], d17[1]}, [r0:16]
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  %tmp2 = call %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 4)
+  %tmp2 = call %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8.p0i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 4)
         %tmp3 = extractvalue %struct.__neon_int8x8x2_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int8x8x2_t %tmp2, 1
         %tmp5 = add <8 x i8> %tmp3, %tmp4
   ret <8 x i8> %tmp5
 }
 
 define <4 x i16> @vld2lanei16(i16* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: vld2lanei16:
 ;Check the alignment value.  Max for this instruction is 32 bits:
 ;CHECK: vld2.16 {d16[1], d17[1]}, [r0:32]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  %tmp2 = call %struct.__neon_int16x4x2_t @llvm.arm.neon.vld2lane.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 8)
+  %tmp2 = call %struct.__neon_int16x4x2_t @llvm.arm.neon.vld2lane.v4i16.p0i8(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 8)
         %tmp3 = extractvalue %struct.__neon_int16x4x2_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int16x4x2_t %tmp2, 1
         %tmp5 = add <4 x i16> %tmp3, %tmp4
   ret <4 x i16> %tmp5
 }
 
 define <2 x i32> @vld2lanei32(i32* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vld2lanei32:
 ;CHECK: vld2.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  %tmp2 = call %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2lane.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
+  %tmp2 = call %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2lane.v2i32.p0i8(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
         %tmp3 = extractvalue %struct.__neon_int32x2x2_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int32x2x2_t %tmp2, 1
         %tmp5 = add <2 x i32> %tmp3, %tmp4
   ret <2 x i32> %tmp5
 }
 
 ;Check for a post-increment updating load.
 define <2 x i32> @vld2lanei32_update(i32** %ptr, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vld2lanei32_update:
 ;CHECK: vld2.32 {d16[1], d17[1]}, [{{r[0-9]+}}]!
   %A = load i32*, i32** %ptr
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  %tmp2 = call %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2lane.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
+  %tmp2 = call %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2lane.v2i32.p0i8(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
   %tmp3 = extractvalue %struct.__neon_int32x2x2_t %tmp2, 0
   %tmp4 = extractvalue %struct.__neon_int32x2x2_t %tmp2, 1
   %tmp5 = add <2 x i32> %tmp3, %tmp4
   %tmp6 = getelementptr i32, i32* %A, i32 2
   store i32* %tmp6, i32** %ptr
   ret <2 x i32> %tmp5
 }
 
 define <2 x float> @vld2lanef(float* %A, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: vld2lanef:
 ;CHECK: vld2.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <2 x float>, <2 x float>* %B
-  %tmp2 = call %struct.__neon_float32x2x2_t @llvm.arm.neon.vld2lane.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
+  %tmp2 = call %struct.__neon_float32x2x2_t @llvm.arm.neon.vld2lane.v2f32.p0i8(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
         %tmp3 = extractvalue %struct.__neon_float32x2x2_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_float32x2x2_t %tmp2, 1
         %tmp5 = fadd <2 x float> %tmp3, %tmp4
   ret <2 x float> %tmp5
 }
 
 define <8 x i16> @vld2laneQi16(i16* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: vld2laneQi16:
 ;Check the (default) alignment.
 ;CHECK: vld2.16 {d17[1], d19[1]}, [{{r[0-9]+}}]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  %tmp2 = call %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2lane.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 5, i32 1)
+  %tmp2 = call %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2lane.v8i16.p0i8(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 5, i32 1)
         %tmp3 = extractvalue %struct.__neon_int16x8x2_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int16x8x2_t %tmp2, 1
         %tmp5 = add <8 x i16> %tmp3, %tmp4
   ret <8 x i16> %tmp5
 }
 
 define <4 x i32> @vld2laneQi32(i32* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: vld2laneQi32:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vld2.32 {d17[0], d19[0]}, [{{r[0-9]+}}:64]
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <4 x i32>, <4 x i32>* %B
-  %tmp2 = call %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2lane.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 2, i32 16)
+  %tmp2 = call %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2lane.v4i32.p0i8(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 2, i32 16)
         %tmp3 = extractvalue %struct.__neon_int32x4x2_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int32x4x2_t %tmp2, 1
         %tmp5 = add <4 x i32> %tmp3, %tmp4
   ret <4 x i32> %tmp5
 }
 
 define <4 x float> @vld2laneQf(float* %A, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: vld2laneQf:
 ;CHECK: vld2.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <4 x float>, <4 x float>* %B
-  %tmp2 = call %struct.__neon_float32x4x2_t @llvm.arm.neon.vld2lane.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, i32 1, i32 1)
+  %tmp2 = call %struct.__neon_float32x4x2_t @llvm.arm.neon.vld2lane.v4f32.p0i8(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, i32 1, i32 1)
         %tmp3 = extractvalue %struct.__neon_float32x4x2_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_float32x4x2_t %tmp2, 1
         %tmp5 = fadd <4 x float> %tmp3, %tmp4
   ret <4 x float> %tmp5
 }
 
-declare %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8(i8*, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
-declare %struct.__neon_int16x4x2_t @llvm.arm.neon.vld2lane.v4i16(i8*, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
-declare %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2lane.v2i32(i8*, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
-declare %struct.__neon_float32x2x2_t @llvm.arm.neon.vld2lane.v2f32(i8*, <2 x float>, <2 x float>, i32, i32) nounwind readonly
-
-declare %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2lane.v8i16(i8*, <8 x i16>, <8 x i16>, i32, i32) nounwind readonly
-declare %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2lane.v4i32(i8*, <4 x i32>, <4 x i32>, i32, i32) nounwind readonly
-declare %struct.__neon_float32x4x2_t @llvm.arm.neon.vld2lane.v4f32(i8*, <4 x float>, <4 x float>, i32, i32) nounwind readonly
+declare %struct.__neon_int8x8x2_t @llvm.arm.neon.vld2lane.v8i8.p0i8(i8*, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
+declare %struct.__neon_int16x4x2_t @llvm.arm.neon.vld2lane.v4i16.p0i8(i8*, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
+declare %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2lane.v2i32.p0i8(i8*, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
+declare %struct.__neon_float32x2x2_t @llvm.arm.neon.vld2lane.v2f32.p0i8(i8*, <2 x float>, <2 x float>, i32, i32) nounwind readonly
+
+declare %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2lane.v8i16.p0i8(i8*, <8 x i16>, <8 x i16>, i32, i32) nounwind readonly
+declare %struct.__neon_int32x4x2_t @llvm.arm.neon.vld2lane.v4i32.p0i8(i8*, <4 x i32>, <4 x i32>, i32, i32) nounwind readonly
+declare %struct.__neon_float32x4x2_t @llvm.arm.neon.vld2lane.v4f32.p0i8(i8*, <4 x float>, <4 x float>, i32, i32) nounwind readonly
 
 %struct.__neon_int8x8x3_t = type { <8 x i8>,  <8 x i8>,  <8 x i8> }
 %struct.__neon_int16x4x3_t = type { <4 x i16>, <4 x i16>, <4 x i16> }
 %struct.__neon_int32x2x3_t = type { <2 x i32>, <2 x i32>, <2 x i32> }
 %struct.__neon_float32x2x3_t = type { <2 x float>, <2 x float>, <2 x float> }
 
 %struct.__neon_int16x8x3_t = type { <8 x i16>, <8 x i16>, <8 x i16> }
 %struct.__neon_int32x4x3_t = type { <4 x i32>, <4 x i32>, <4 x i32> }
 %struct.__neon_float32x4x3_t = type { <4 x float>, <4 x float>, <4 x float> }
 
 define <8 x i8> @vld3lanei8(i8* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vld3lanei8:
 ;CHECK: vld3.8
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  %tmp2 = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 1)
+  %tmp2 = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3lane.v8i8.p0i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 1)
         %tmp3 = extractvalue %struct.__neon_int8x8x3_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int8x8x3_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_int8x8x3_t %tmp2, 2
         %tmp6 = add <8 x i8> %tmp3, %tmp4
         %tmp7 = add <8 x i8> %tmp5, %tmp6
   ret <8 x i8> %tmp7
 }
 
 define <4 x i16> @vld3lanei16(i16* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: vld3lanei16:
 ;Check the (default) alignment value.  VLD3 does not support alignment.
 ;CHECK: vld3.16 {d{{.*}}[1], d{{.*}}[1], d{{.*}}[1]}, [{{r[0-9]+}}]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  %tmp2 = call %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3lane.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 8)
+  %tmp2 = call %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3lane.v4i16.p0i8(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 8)
         %tmp3 = extractvalue %struct.__neon_int16x4x3_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int16x4x3_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_int16x4x3_t %tmp2, 2
         %tmp6 = add <4 x i16> %tmp3, %tmp4
         %tmp7 = add <4 x i16> %tmp5, %tmp6
   ret <4 x i16> %tmp7
 }
 
 define <2 x i32> @vld3lanei32(i32* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vld3lanei32:
 ;CHECK: vld3.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  %tmp2 = call %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3lane.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
+  %tmp2 = call %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3lane.v2i32.p0i8(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
         %tmp3 = extractvalue %struct.__neon_int32x2x3_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int32x2x3_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_int32x2x3_t %tmp2, 2
         %tmp6 = add <2 x i32> %tmp3, %tmp4
         %tmp7 = add <2 x i32> %tmp5, %tmp6
   ret <2 x i32> %tmp7
 }
 
 define <2 x float> @vld3lanef(float* %A, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: vld3lanef:
 ;CHECK: vld3.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <2 x float>, <2 x float>* %B
-  %tmp2 = call %struct.__neon_float32x2x3_t @llvm.arm.neon.vld3lane.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
+  %tmp2 = call %struct.__neon_float32x2x3_t @llvm.arm.neon.vld3lane.v2f32.p0i8(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
         %tmp3 = extractvalue %struct.__neon_float32x2x3_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_float32x2x3_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_float32x2x3_t %tmp2, 2
         %tmp6 = fadd <2 x float> %tmp3, %tmp4
         %tmp7 = fadd <2 x float> %tmp5, %tmp6
   ret <2 x float> %tmp7
 }
 
 define <8 x i16> @vld3laneQi16(i16* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: vld3laneQi16:
 ;Check the (default) alignment value.  VLD3 does not support alignment.
 ;CHECK: vld3.16 {d{{.*}}[1], d{{.*}}[1], d{{.*}}[1]}, [{{r[0-9]+}}]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  %tmp2 = call %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3lane.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1, i32 8)
+  %tmp2 = call %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3lane.v8i16.p0i8(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1, i32 8)
         %tmp3 = extractvalue %struct.__neon_int16x8x3_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int16x8x3_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_int16x8x3_t %tmp2, 2
         %tmp6 = add <8 x i16> %tmp3, %tmp4
         %tmp7 = add <8 x i16> %tmp5, %tmp6
   ret <8 x i16> %tmp7
 }
 
 ;Check for a post-increment updating load with register increment.
 define <8 x i16> @vld3laneQi16_update(i16** %ptr, <8 x i16>* %B, i32 %inc) nounwind {
 ;CHECK-LABEL: vld3laneQi16_update:
 ;CHECK: vld3.16 {d{{.*}}[1], d{{.*}}[1], d{{.*}}[1]}, [{{r[0-9]+}}], {{r[0-9]+}}
   %A = load i16*, i16** %ptr
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  %tmp2 = call %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3lane.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1, i32 8)
+  %tmp2 = call %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3lane.v8i16.p0i8(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1, i32 8)
   %tmp3 = extractvalue %struct.__neon_int16x8x3_t %tmp2, 0
   %tmp4 = extractvalue %struct.__neon_int16x8x3_t %tmp2, 1
   %tmp5 = extractvalue %struct.__neon_int16x8x3_t %tmp2, 2
   %tmp6 = add <8 x i16> %tmp3, %tmp4
   %tmp7 = add <8 x i16> %tmp5, %tmp6
   %tmp8 = getelementptr i16, i16* %A, i32 %inc
   store i16* %tmp8, i16** %ptr
   ret <8 x i16> %tmp7
 }
 
 define <4 x i32> @vld3laneQi32(i32* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: vld3laneQi32:
 ;CHECK: vld3.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <4 x i32>, <4 x i32>* %B
-  %tmp2 = call %struct.__neon_int32x4x3_t @llvm.arm.neon.vld3lane.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 3, i32 1)
+  %tmp2 = call %struct.__neon_int32x4x3_t @llvm.arm.neon.vld3lane.v4i32.p0i8(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 3, i32 1)
         %tmp3 = extractvalue %struct.__neon_int32x4x3_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int32x4x3_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_int32x4x3_t %tmp2, 2
         %tmp6 = add <4 x i32> %tmp3, %tmp4
         %tmp7 = add <4 x i32> %tmp5, %tmp6
   ret <4 x i32> %tmp7
 }
 
 define <4 x float> @vld3laneQf(float* %A, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: vld3laneQf:
 ;CHECK: vld3.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <4 x float>, <4 x float>* %B
-  %tmp2 = call %struct.__neon_float32x4x3_t @llvm.arm.neon.vld3lane.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1, i32 1)
+  %tmp2 = call %struct.__neon_float32x4x3_t @llvm.arm.neon.vld3lane.v4f32.p0i8(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1, i32 1)
         %tmp3 = extractvalue %struct.__neon_float32x4x3_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_float32x4x3_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_float32x4x3_t %tmp2, 2
         %tmp6 = fadd <4 x float> %tmp3, %tmp4
         %tmp7 = fadd <4 x float> %tmp5, %tmp6
   ret <4 x float> %tmp7
 }
 
-declare %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3lane.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
-declare %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3lane.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
-declare %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
-declare %struct.__neon_float32x2x3_t @llvm.arm.neon.vld3lane.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, i32, i32) nounwind readonly
-
-declare %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3lane.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind readonly
-declare %struct.__neon_int32x4x3_t @llvm.arm.neon.vld3lane.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32) nounwind readonly
-declare %struct.__neon_float32x4x3_t @llvm.arm.neon.vld3lane.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, i32, i32) nounwind readonly
+declare %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3lane.v8i8.p0i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
+declare %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3lane.v4i16.p0i8(i8*, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
+declare %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3lane.v2i32.p0i8(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
+declare %struct.__neon_float32x2x3_t @llvm.arm.neon.vld3lane.v2f32.p0i8(i8*, <2 x float>, <2 x float>, <2 x float>, i32, i32) nounwind readonly
+
+declare %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3lane.v8i16.p0i8(i8*, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind readonly
+declare %struct.__neon_int32x4x3_t @llvm.arm.neon.vld3lane.v4i32.p0i8(i8*, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32) nounwind readonly
+declare %struct.__neon_float32x4x3_t @llvm.arm.neon.vld3lane.v4f32.p0i8(i8*, <4 x float>, <4 x float>, <4 x float>, i32, i32) nounwind readonly
 
 %struct.__neon_int8x8x4_t = type { <8 x i8>,  <8 x i8>,  <8 x i8>,  <8 x i8> }
 %struct.__neon_int16x4x4_t = type { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> }
 %struct.__neon_int32x2x4_t = type { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> }
 %struct.__neon_float32x2x4_t = type { <2 x float>, <2 x float>, <2 x float>, <2 x float> }
 
 %struct.__neon_int16x8x4_t = type { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> }
 %struct.__neon_int32x4x4_t = type { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> }
 %struct.__neon_float32x4x4_t = type { <4 x float>, <4 x float>, <4 x float>, <4 x float> }
 
 define <8 x i8> @vld4lanei8(i8* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vld4lanei8:
 ;Check the alignment value.  Max for this instruction is 32 bits:
 ;CHECK: vld4.8 {d{{.*}}[1], d{{.*}}[1], d{{.*}}[1], d{{.*}}[1]}, [{{r[0-9]+}}:32]
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  %tmp2 = call %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 8)
+  %tmp2 = call %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4lane.v8i8.p0i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 8)
         %tmp3 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 2
         %tmp6 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 3
         %tmp7 = add <8 x i8> %tmp3, %tmp4
         %tmp8 = add <8 x i8> %tmp5, %tmp6
         %tmp9 = add <8 x i8> %tmp7, %tmp8
   ret <8 x i8> %tmp9
 }
 
 ;Check for a post-increment updating load.
 define <8 x i8> @vld4lanei8_update(i8** %ptr, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vld4lanei8_update:
 ;CHECK: vld4.8 {d16[1], d17[1], d18[1], d19[1]}, [{{r[0-9]+}}:32]!
   %A = load i8*, i8** %ptr
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  %tmp2 = call %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 8)
+  %tmp2 = call %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4lane.v8i8.p0i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 8)
   %tmp3 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 0
   %tmp4 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 1
   %tmp5 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 2
   %tmp6 = extractvalue %struct.__neon_int8x8x4_t %tmp2, 3
   %tmp7 = add <8 x i8> %tmp3, %tmp4
   %tmp8 = add <8 x i8> %tmp5, %tmp6
   %tmp9 = add <8 x i8> %tmp7, %tmp8
   %tmp10 = getelementptr i8, i8* %A, i32 4
   store i8* %tmp10, i8** %ptr
   ret <8 x i8> %tmp9
 }
 
 define <4 x i16> @vld4lanei16(i16* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: vld4lanei16:
 ;Check that a power-of-two alignment smaller than the total size of the memory
 ;being loaded is ignored.
 ;CHECK: vld4.16 {d16[1], d17[1], d18[1], d19[1]}, [{{r[0-9]+}}]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  %tmp2 = call %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4lane.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 4)
+  %tmp2 = call %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4lane.v4i16.p0i8(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 4)
         %tmp3 = extractvalue %struct.__neon_int16x4x4_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int16x4x4_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_int16x4x4_t %tmp2, 2
         %tmp6 = extractvalue %struct.__neon_int16x4x4_t %tmp2, 3
         %tmp7 = add <4 x i16> %tmp3, %tmp4
         %tmp8 = add <4 x i16> %tmp5, %tmp6
         %tmp9 = add <4 x i16> %tmp7, %tmp8
   ret <4 x i16> %tmp9
 }
 
 define <2 x i32> @vld4lanei32(i32* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vld4lanei32:
 ;Check the alignment value.  An 8-byte alignment is allowed here even though
 ;it is smaller than the total size of the memory being loaded.
 ;CHECK: vld4.32 {d16[1], d17[1], d18[1], d19[1]}, [{{r[0-9]+}}:64]
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  %tmp2 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4lane.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 8)
+  %tmp2 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4lane.v2i32.p0i8(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 8)
         %tmp3 = extractvalue %struct.__neon_int32x2x4_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int32x2x4_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_int32x2x4_t %tmp2, 2
         %tmp6 = extractvalue %struct.__neon_int32x2x4_t %tmp2, 3
         %tmp7 = add <2 x i32> %tmp3, %tmp4
         %tmp8 = add <2 x i32> %tmp5, %tmp6
         %tmp9 = add <2 x i32> %tmp7, %tmp8
   ret <2 x i32> %tmp9
 }
 
 define <2 x float> @vld4lanef(float* %A, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: vld4lanef:
 ;CHECK: vld4.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <2 x float>, <2 x float>* %B
-  %tmp2 = call %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4lane.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
+  %tmp2 = call %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4lane.v2f32.p0i8(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
         %tmp3 = extractvalue %struct.__neon_float32x2x4_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_float32x2x4_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_float32x2x4_t %tmp2, 2
         %tmp6 = extractvalue %struct.__neon_float32x2x4_t %tmp2, 3
         %tmp7 = fadd <2 x float> %tmp3, %tmp4
         %tmp8 = fadd <2 x float> %tmp5, %tmp6
         %tmp9 = fadd <2 x float> %tmp7, %tmp8
   ret <2 x float> %tmp9
 }
 
 define <8 x i16> @vld4laneQi16(i16* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: vld4laneQi16:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vld4.16 {d16[1], d18[1], d20[1], d22[1]}, [{{r[0-9]+}}:64]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  %tmp2 = call %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4lane.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1, i32 16)
+  %tmp2 = call %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4lane.v8i16.p0i8(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1, i32 16)
         %tmp3 = extractvalue %struct.__neon_int16x8x4_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int16x8x4_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_int16x8x4_t %tmp2, 2
         %tmp6 = extractvalue %struct.__neon_int16x8x4_t %tmp2, 3
         %tmp7 = add <8 x i16> %tmp3, %tmp4
         %tmp8 = add <8 x i16> %tmp5, %tmp6
         %tmp9 = add <8 x i16> %tmp7, %tmp8
   ret <8 x i16> %tmp9
 }
 
 define <4 x i32> @vld4laneQi32(i32* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: vld4laneQi32:
 ;Check the (default) alignment.
 ;CHECK: vld4.32 {d17[0], d19[0], d21[0], d23[0]}, [{{r[0-9]+}}]
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <4 x i32>, <4 x i32>* %B
-  %tmp2 = call %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4lane.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 2, i32 1)
+  %tmp2 = call %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4lane.v4i32.p0i8(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 2, i32 1)
         %tmp3 = extractvalue %struct.__neon_int32x4x4_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_int32x4x4_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_int32x4x4_t %tmp2, 2
         %tmp6 = extractvalue %struct.__neon_int32x4x4_t %tmp2, 3
         %tmp7 = add <4 x i32> %tmp3, %tmp4
         %tmp8 = add <4 x i32> %tmp5, %tmp6
         %tmp9 = add <4 x i32> %tmp7, %tmp8
   ret <4 x i32> %tmp9
 }
 
 define <4 x float> @vld4laneQf(float* %A, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: vld4laneQf:
 ;CHECK: vld4.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <4 x float>, <4 x float>* %B
-  %tmp2 = call %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4lane.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1, i32 1)
+  %tmp2 = call %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4lane.v4f32.p0i8(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1, i32 1)
         %tmp3 = extractvalue %struct.__neon_float32x4x4_t %tmp2, 0
         %tmp4 = extractvalue %struct.__neon_float32x4x4_t %tmp2, 1
         %tmp5 = extractvalue %struct.__neon_float32x4x4_t %tmp2, 2
         %tmp6 = extractvalue %struct.__neon_float32x4x4_t %tmp2, 3
         %tmp7 = fadd <4 x float> %tmp3, %tmp4
         %tmp8 = fadd <4 x float> %tmp5, %tmp6
         %tmp9 = fadd <4 x float> %tmp7, %tmp8
   ret <4 x float> %tmp9
 }
 
-declare %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4lane.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
-declare %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4lane.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
-declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
-declare %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4lane.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, <2 x float>, i32, i32) nounwind readonly
-
-declare %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4lane.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind readonly
-declare %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4lane.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32) nounwind readonly
-declare %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4lane.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32, i32) nounwind readonly
+declare %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4lane.v8i8.p0i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind readonly
+declare %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4lane.v4i16.p0i8(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind readonly
+declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4lane.v2i32.p0i8(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind readonly
+declare %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4lane.v2f32.p0i8(i8*, <2 x float>, <2 x float>, <2 x float>, <2 x float>, i32, i32) nounwind readonly
+
+declare %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4lane.v8i16.p0i8(i8*, <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind readonly
+declare %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4lane.v4i32.p0i8(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32) nounwind readonly
+declare %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4lane.v4f32.p0i8(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32, i32) nounwind readonly
 
 ; Radar 8776599: If one of the operands to a QQQQ REG_SEQUENCE is a register
 ; in the QPR_VFP2 regclass, it needs to be copied to a QPR regclass because
 ; we don't currently have a QQQQ_VFP2 super-regclass.  (The "0" for the low
 ; part of %ins67 is supposed to be loaded by a VLDRS instruction in this test.)
 define <8 x i16> @test_qqqq_regsequence_subreg([6 x i64] %b) nounwind {
 ;CHECK-LABEL: test_qqqq_regsequence_subreg:
 ;CHECK: vld3.16
   %tmp63 = extractvalue [6 x i64] %b, 5
   %tmp64 = zext i64 %tmp63 to i128
   %tmp65 = shl i128 %tmp64, 64
   %ins67 = or i128 %tmp65, 0
   %tmp78 = bitcast i128 %ins67 to <8 x i16>
-  %vld3_lane = tail call %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3lane.v8i16(i8* undef, <8 x i16> undef, <8 x i16> undef, <8 x i16> %tmp78, i32 1, i32 2)
+  %vld3_lane = tail call %struct.__neon_int16x8x3_t @llvm.arm.neon.vld3lane.v8i16.p0i8(i8* undef, <8 x i16> undef, <8 x i16> undef, <8 x i16> %tmp78, i32 1, i32 2)
   %tmp3 = extractvalue %struct.__neon_int16x8x3_t %vld3_lane, 0
   %tmp4 = extractvalue %struct.__neon_int16x8x3_t %vld3_lane, 1
   %tmp5 = extractvalue %struct.__neon_int16x8x3_t %vld3_lane, 2
   %tmp6 = add <8 x i16> %tmp3, %tmp4
   %tmp7 = add <8 x i16> %tmp5, %tmp6
   ret <8 x i16> %tmp7
 }
 
 declare void @llvm.trap() nounwind

llvm/trunk/test/CodeGen/ARM/vmov.ll

 entry:
 ;CHECK-LABEL: any_extend:
 ;CHECK: vmovl
   %and.i186 = zext <4 x i1> %x to <4 x i32>
   %add.i185 = sub <4 x i32> %and.i186, %y
   %sub.i = sub <4 x i32> %add.i185, zeroinitializer
   %add.i = add <4 x i32> %sub.i, zeroinitializer
   %vmovn.i = trunc <4 x i32> %add.i to <4 x i16>
-  tail call void @llvm.arm.neon.vst1.v4i16(i8* undef, <4 x i16> %vmovn.i, i32 2)
+  tail call void @llvm.arm.neon.vst1.p0i8.v4i16(i8* undef, <4 x i16> %vmovn.i, i32 2)
   unreachable
 }
 
-declare void @llvm.arm.neon.vst1.v4i16(i8*, <4 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v4i16(i8*, <4 x i16>, i32) nounwind

llvm/trunk/test/CodeGen/ARM/vmul.ll

 define void @distribute(i16* %dst, i8* %src, i32 %mul) nounwind {
 entry:
 ; CHECK-LABEL: distribute:
 ; CHECK: vmull.u8 [[REG1:(q[0-9]+)]], d{{.*}}, [[REG2:(d[0-9]+)]]
 ; CHECK: vmlal.u8 [[REG1]], d{{.*}}, [[REG2]]
   %0 = trunc i32 %mul to i8
   %1 = insertelement <8 x i8> undef, i8 %0, i32 0
   %2 = shufflevector <8 x i8> %1, <8 x i8> undef, <8 x i32> zeroinitializer
-  %3 = tail call <16 x i8> @llvm.arm.neon.vld1.v16i8(i8* %src, i32 1)
+  %3 = tail call <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* %src, i32 1)
   %4 = bitcast <16 x i8> %3 to <2 x double>
   %5 = extractelement <2 x double> %4, i32 1
   %6 = bitcast double %5 to <8 x i8>
   %7 = zext <8 x i8> %6 to <8 x i16>
   %8 = zext <8 x i8> %2 to <8 x i16>
   %9 = extractelement <2 x double> %4, i32 0
   %10 = bitcast double %9 to <8 x i8>
   %11 = zext <8 x i8> %10 to <8 x i16>
   %12 = add <8 x i16> %7, %11
   %13 = mul <8 x i16> %12, %8
   %14 = bitcast i16* %dst to i8*
-  tail call void @llvm.arm.neon.vst1.v8i16(i8* %14, <8 x i16> %13, i32 2)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %14, <8 x i16> %13, i32 2)
   ret void
 }
 
-declare <16 x i8> @llvm.arm.neon.vld1.v16i8(i8*, i32) nounwind readonly
+declare <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8*, i32) nounwind readonly
 
-declare void @llvm.arm.neon.vst1.v8i16(i8*, <8 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v8i16(i8*, <8 x i16>, i32) nounwind
 
 ; Take advantage of the Cortex-A8 multiplier accumulator forward.
 
 %struct.uint8x8_t = type { <8 x i8> }
 
 define void @distribute2(%struct.uint8x8_t* nocapture %dst, i8* %src, i32 %mul) nounwind {
 entry:
 ; CHECK: distribute2
 ; CHECK-NOT: vadd.i8
 ; CHECK: vmul.i8
 ; CHECK: vmla.i8
   %0 = trunc i32 %mul to i8
   %1 = insertelement <8 x i8> undef, i8 %0, i32 0
   %2 = shufflevector <8 x i8> %1, <8 x i8> undef, <8 x i32> zeroinitializer
-  %3 = tail call <16 x i8> @llvm.arm.neon.vld1.v16i8(i8* %src, i32 1)
+  %3 = tail call <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* %src, i32 1)
   %4 = bitcast <16 x i8> %3 to <2 x double>
   %5 = extractelement <2 x double> %4, i32 1
   %6 = bitcast double %5 to <8 x i8>
   %7 = extractelement <2 x double> %4, i32 0
   %8 = bitcast double %7 to <8 x i8>
   %9 = add <8 x i8> %6, %8
   %10 = mul <8 x i8> %9, %2
   %11 = getelementptr inbounds %struct.uint8x8_t, %struct.uint8x8_t* %dst, i32 0, i32 0
   store <8 x i8> %10, <8 x i8>* %11, align 8
   ret void
 }
 
 define void @distribute2_commutative(%struct.uint8x8_t* nocapture %dst, i8* %src, i32 %mul) nounwind {
 entry:
 ; CHECK: distribute2_commutative
 ; CHECK-NOT: vadd.i8
 ; CHECK: vmul.i8
 ; CHECK: vmla.i8
   %0 = trunc i32 %mul to i8
   %1 = insertelement <8 x i8> undef, i8 %0, i32 0
   %2 = shufflevector <8 x i8> %1, <8 x i8> undef, <8 x i32> zeroinitializer
-  %3 = tail call <16 x i8> @llvm.arm.neon.vld1.v16i8(i8* %src, i32 1)
+  %3 = tail call <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* %src, i32 1)
   %4 = bitcast <16 x i8> %3 to <2 x double>
   %5 = extractelement <2 x double> %4, i32 1
   %6 = bitcast double %5 to <8 x i8>
   %7 = extractelement <2 x double> %4, i32 0
   %8 = bitcast double %7 to <8 x i8>
   %9 = add <8 x i8> %6, %8
   %10 = mul <8 x i8> %2, %9
   %11 = getelementptr inbounds %struct.uint8x8_t, %struct.uint8x8_t* %dst, i32 0, i32 0
for.body:                                         ; preds = %for.cond.loopexit, %for.body.lr.ph
   br i1 undef, label %for.cond.loopexit, label %for.body33.lr.ph
 
 for.body33.lr.ph:                                 ; preds = %for.body
   %.sub = select i1 undef, i32 0, i32 undef
   br label %for.body33
 
 for.body33:                                       ; preds = %for.body33, %for.body33.lr.ph
   %add45 = add i32 undef, undef
-  %vld155 = tail call <16 x i8> @llvm.arm.neon.vld1.v16i8(i8* undef, i32 1)
+  %vld155 = tail call <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* undef, i32 1)
   %0 = load i32*, i32** undef, align 4
   %shuffle.i250 = shufflevector <2 x i64> undef, <2 x i64> undef, <1 x i32> zeroinitializer
   %1 = bitcast <1 x i64> %shuffle.i250 to <8 x i8>
   %vmovl.i249 = zext <8 x i8> %1 to <8 x i16>
   %shuffle.i246 = shufflevector <2 x i64> undef, <2 x i64> undef, <1 x i32> zeroinitializer
   %shuffle.i240 = shufflevector <2 x i64> undef, <2 x i64> undef, <1 x i32> <i32 1>
   %2 = bitcast <1 x i64> %shuffle.i240 to <8 x i8>
   %3 = bitcast <16 x i8> undef to <2 x i64>

llvm/trunk/test/CodeGen/ARM/vst1.ll

 ; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s
 
 define void @vst1i8(i8* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst1i8:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vst1.8 {d16}, [r0:64]
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  call void @llvm.arm.neon.vst1.v8i8(i8* %A, <8 x i8> %tmp1, i32 16)
+  call void @llvm.arm.neon.vst1.p0i8.v8i8(i8* %A, <8 x i8> %tmp1, i32 16)
   ret void
 }
 
 define void @vst1i16(i16* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst1i16:
 ;CHECK: vst1.16
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  call void @llvm.arm.neon.vst1.v4i16(i8* %tmp0, <4 x i16> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst1.p0i8.v4i16(i8* %tmp0, <4 x i16> %tmp1, i32 1)
   ret void
 }
 
 define void @vst1i32(i32* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst1i32:
 ;CHECK: vst1.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  call void @llvm.arm.neon.vst1.v2i32(i8* %tmp0, <2 x i32> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst1.p0i8.v2i32(i8* %tmp0, <2 x i32> %tmp1, i32 1)
   ret void
 }
 
 define void @vst1f(float* %A, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: vst1f:
 ;CHECK: vst1.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <2 x float>, <2 x float>* %B
-  call void @llvm.arm.neon.vst1.v2f32(i8* %tmp0, <2 x float> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst1.p0i8.v2f32(i8* %tmp0, <2 x float> %tmp1, i32 1)
   ret void
 }
 
 ;Check for a post-increment updating store.
 define void @vst1f_update(float** %ptr, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: vst1f_update:
 ;CHECK: vst1.32 {d16}, [r1]!
   %A = load float*, float** %ptr
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <2 x float>, <2 x float>* %B
-  call void @llvm.arm.neon.vst1.v2f32(i8* %tmp0, <2 x float> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst1.p0i8.v2f32(i8* %tmp0, <2 x float> %tmp1, i32 1)
   %tmp2 = getelementptr float, float* %A, i32 2
   store float* %tmp2, float** %ptr
   ret void
 }
 
 define void @vst1i64(i64* %A, <1 x i64>* %B) nounwind {
 ;CHECK-LABEL: vst1i64:
 ;CHECK: vst1.64
   %tmp0 = bitcast i64* %A to i8*
   %tmp1 = load <1 x i64>, <1 x i64>* %B
-  call void @llvm.arm.neon.vst1.v1i64(i8* %tmp0, <1 x i64> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst1.p0i8.v1i64(i8* %tmp0, <1 x i64> %tmp1, i32 1)
   ret void
 }
 
 define void @vst1Qi8(i8* %A, <16 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst1Qi8:
 ;Check the alignment value.  Max for this instruction is 128 bits:
 ;CHECK: vst1.8 {d16, d17}, [r0:64]
   %tmp1 = load <16 x i8>, <16 x i8>* %B
-  call void @llvm.arm.neon.vst1.v16i8(i8* %A, <16 x i8> %tmp1, i32 8)
+  call void @llvm.arm.neon.vst1.p0i8.v16i8(i8* %A, <16 x i8> %tmp1, i32 8)
   ret void
 }
 
 define void @vst1Qi16(i16* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst1Qi16:
 ;Check the alignment value.  Max for this instruction is 128 bits:
 ;CHECK: vst1.16 {d16, d17}, [r0:128]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  call void @llvm.arm.neon.vst1.v8i16(i8* %tmp0, <8 x i16> %tmp1, i32 32)
+  call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %tmp0, <8 x i16> %tmp1, i32 32)
   ret void
 }
 
 ;Check for a post-increment updating store with register increment.
 define void @vst1Qi16_update(i16** %ptr, <8 x i16>* %B, i32 %inc) nounwind {
 ;CHECK-LABEL: vst1Qi16_update:
 ;CHECK: vst1.16 {d16, d17}, [r1:64], r2
   %A = load i16*, i16** %ptr
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  call void @llvm.arm.neon.vst1.v8i16(i8* %tmp0, <8 x i16> %tmp1, i32 8)
+  call void @llvm.arm.neon.vst1.p0i8.v8i16(i8* %tmp0, <8 x i16> %tmp1, i32 8)
   %tmp2 = getelementptr i16, i16* %A, i32 %inc
   store i16* %tmp2, i16** %ptr
   ret void
 }
 
 define void @vst1Qi32(i32* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst1Qi32:
 ;CHECK: vst1.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <4 x i32>, <4 x i32>* %B
-  call void @llvm.arm.neon.vst1.v4i32(i8* %tmp0, <4 x i32> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst1.p0i8.v4i32(i8* %tmp0, <4 x i32> %tmp1, i32 1)
   ret void
 }
 
 define void @vst1Qf(float* %A, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: vst1Qf:
 ;CHECK: vst1.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <4 x float>, <4 x float>* %B
-  call void @llvm.arm.neon.vst1.v4f32(i8* %tmp0, <4 x float> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst1.p0i8.v4f32(i8* %tmp0, <4 x float> %tmp1, i32 1)
   ret void
 }
 
 define void @vst1Qi64(i64* %A, <2 x i64>* %B) nounwind {
 ;CHECK-LABEL: vst1Qi64:
 ;CHECK: vst1.64
   %tmp0 = bitcast i64* %A to i8*
   %tmp1 = load <2 x i64>, <2 x i64>* %B
-  call void @llvm.arm.neon.vst1.v2i64(i8* %tmp0, <2 x i64> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst1.p0i8.v2i64(i8* %tmp0, <2 x i64> %tmp1, i32 1)
   ret void
 }
 
 define void @vst1Qf64(double* %A, <2 x double>* %B) nounwind {
 ;CHECK-LABEL: vst1Qf64:
 ;CHECK: vst1.64
   %tmp0 = bitcast double* %A to i8*
   %tmp1 = load <2 x double>, <2 x double>* %B
-  call void @llvm.arm.neon.vst1.v2f64(i8* %tmp0, <2 x double> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst1.p0i8.v2f64(i8* %tmp0, <2 x double> %tmp1, i32 1)
   ret void
 }
 
-declare void @llvm.arm.neon.vst1.v8i8(i8*, <8 x i8>, i32) nounwind
-declare void @llvm.arm.neon.vst1.v4i16(i8*, <4 x i16>, i32) nounwind
-declare void @llvm.arm.neon.vst1.v2i32(i8*, <2 x i32>, i32) nounwind
-declare void @llvm.arm.neon.vst1.v2f32(i8*, <2 x float>, i32) nounwind
-declare void @llvm.arm.neon.vst1.v1i64(i8*, <1 x i64>, i32) nounwind
-
-declare void @llvm.arm.neon.vst1.v16i8(i8*, <16 x i8>, i32) nounwind
-declare void @llvm.arm.neon.vst1.v8i16(i8*, <8 x i16>, i32) nounwind
-declare void @llvm.arm.neon.vst1.v4i32(i8*, <4 x i32>, i32) nounwind
-declare void @llvm.arm.neon.vst1.v4f32(i8*, <4 x float>, i32) nounwind
-declare void @llvm.arm.neon.vst1.v2i64(i8*, <2 x i64>, i32) nounwind
-declare void @llvm.arm.neon.vst1.v2f64(i8*, <2 x double>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v8i8(i8*, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v4i16(i8*, <4 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v2i32(i8*, <2 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v2f32(i8*, <2 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v1i64(i8*, <1 x i64>, i32) nounwind
+
+declare void @llvm.arm.neon.vst1.p0i8.v16i8(i8*, <16 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v8i16(i8*, <8 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v4i32(i8*, <4 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v4f32(i8*, <4 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v2i64(i8*, <2 x i64>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v2f64(i8*, <2 x double>, i32) nounwind

llvm/trunk/test/CodeGen/ARM/vst2.ll

 ; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s
 
 define void @vst2i8(i8* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst2i8:
 ;Check the alignment value.  Max for this instruction is 128 bits:
 ;CHECK: vst2.8 {d16, d17}, [r0:64]
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  call void @llvm.arm.neon.vst2.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 8)
+  call void @llvm.arm.neon.vst2.p0i8.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 8)
   ret void
 }
 
 ;Check for a post-increment updating store with register increment.
 define void @vst2i8_update(i8** %ptr, <8 x i8>* %B, i32 %inc) nounwind {
 ;CHECK-LABEL: vst2i8_update:
 ;CHECK: vst2.8 {d16, d17}, [r1], r2
   %A = load i8*, i8** %ptr
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  call void @llvm.arm.neon.vst2.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 4)
+  call void @llvm.arm.neon.vst2.p0i8.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 4)
   %tmp2 = getelementptr i8, i8* %A, i32 %inc
   store i8* %tmp2, i8** %ptr
   ret void
 }
 
 define void @vst2i16(i16* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst2i16:
 ;Check the alignment value.  Max for this instruction is 128 bits:
 ;CHECK: vst2.16 {d16, d17}, [r0:128]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  call void @llvm.arm.neon.vst2.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 32)
+  call void @llvm.arm.neon.vst2.p0i8.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 32)
   ret void
 }
 
 define void @vst2i32(i32* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst2i32:
 ;CHECK: vst2.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  call void @llvm.arm.neon.vst2.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst2.p0i8.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1)
   ret void
 }
 
 define void @vst2f(float* %A, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: vst2f:
 ;CHECK: vst2.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <2 x float>, <2 x float>* %B
-  call void @llvm.arm.neon.vst2.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst2.p0i8.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, i32 1)
   ret void
 }
 
 define void @vst2i64(i64* %A, <1 x i64>* %B) nounwind {
 ;CHECK-LABEL: vst2i64:
 ;Check the alignment value.  Max for this instruction is 128 bits:
 ;CHECK: vst1.64 {d16, d17}, [r0:128]
   %tmp0 = bitcast i64* %A to i8*
   %tmp1 = load <1 x i64>, <1 x i64>* %B
-  call void @llvm.arm.neon.vst2.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 32)
+  call void @llvm.arm.neon.vst2.p0i8.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 32)
   ret void
 }
 
 ;Check for a post-increment updating store.
 define void @vst2i64_update(i64** %ptr, <1 x i64>* %B) nounwind {
 ;CHECK-LABEL: vst2i64_update:
 ;CHECK: vst1.64 {d16, d17}, [r1:64]!
   %A = load i64*, i64** %ptr
   %tmp0 = bitcast i64* %A to i8*
   %tmp1 = load <1 x i64>, <1 x i64>* %B
-  call void @llvm.arm.neon.vst2.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 8)
+  call void @llvm.arm.neon.vst2.p0i8.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 8)
   %tmp2 = getelementptr i64, i64* %A, i32 2
   store i64* %tmp2, i64** %ptr
   ret void
 }
 
 define void @vst2Qi8(i8* %A, <16 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst2Qi8:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vst2.8 {d16, d17, d18, d19}, [r0:64]
   %tmp1 = load <16 x i8>, <16 x i8>* %B
-  call void @llvm.arm.neon.vst2.v16i8(i8* %A, <16 x i8> %tmp1, <16 x i8> %tmp1, i32 8)
+  call void @llvm.arm.neon.vst2.p0i8.v16i8(i8* %A, <16 x i8> %tmp1, <16 x i8> %tmp1, i32 8)
   ret void
 }
 
 define void @vst2Qi16(i16* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst2Qi16:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vst2.16 {d16, d17, d18, d19}, [r0:128]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  call void @llvm.arm.neon.vst2.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 16)
+  call void @llvm.arm.neon.vst2.p0i8.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 16)
   ret void
 }
 
 define void @vst2Qi32(i32* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst2Qi32:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vst2.32 {d16, d17, d18, d19}, [r0:256]
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <4 x i32>, <4 x i32>* %B
-  call void @llvm.arm.neon.vst2.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 64)
+  call void @llvm.arm.neon.vst2.p0i8.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 64)
   ret void
 }
 
 define void @vst2Qf(float* %A, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: vst2Qf:
 ;CHECK: vst2.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <4 x float>, <4 x float>* %B
-  call void @llvm.arm.neon.vst2.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst2.p0i8.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
   ret void
 }
 
 define i8* @vst2update(i8* %out, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst2update:
 ;CHECK: vst2.16 {d16, d17}, [r0]!
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  tail call void @llvm.arm.neon.vst2.v4i16(i8* %out, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 2)
+  tail call void @llvm.arm.neon.vst2.p0i8.v4i16(i8* %out, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 2)
   %t5 = getelementptr inbounds i8, i8* %out, i32 16
   ret i8* %t5
 }
 
 define i8* @vst2update2(i8 * %out, <4 x float> * %this) nounwind optsize ssp align 2 {
 ;CHECK-LABEL: vst2update2:
 ;CHECK: vst2.32 {d16, d17, d18, d19}, [r0]!
   %tmp1 = load <4 x float>, <4 x float>* %this
-  call void @llvm.arm.neon.vst2.v4f32(i8* %out, <4 x float> %tmp1, <4 x float> %tmp1, i32 4) nounwind
+  call void @llvm.arm.neon.vst2.p0i8.v4f32(i8* %out, <4 x float> %tmp1, <4 x float> %tmp1, i32 4) nounwind
   %tmp2 = getelementptr inbounds i8, i8* %out, i32  32
   ret i8* %tmp2
 }
 
-declare void @llvm.arm.neon.vst2.v8i8(i8*, <8 x i8>, <8 x i8>, i32) nounwind
-declare void @llvm.arm.neon.vst2.v4i16(i8*, <4 x i16>, <4 x i16>, i32) nounwind
-declare void @llvm.arm.neon.vst2.v2i32(i8*, <2 x i32>, <2 x i32>, i32) nounwind
-declare void @llvm.arm.neon.vst2.v2f32(i8*, <2 x float>, <2 x float>, i32) nounwind
-declare void @llvm.arm.neon.vst2.v1i64(i8*, <1 x i64>, <1 x i64>, i32) nounwind
-
-declare void @llvm.arm.neon.vst2.v16i8(i8*, <16 x i8>, <16 x i8>, i32) nounwind
-declare void @llvm.arm.neon.vst2.v8i16(i8*, <8 x i16>, <8 x i16>, i32) nounwind
-declare void @llvm.arm.neon.vst2.v4i32(i8*, <4 x i32>, <4 x i32>, i32) nounwind
-declare void @llvm.arm.neon.vst2.v4f32(i8*, <4 x float>, <4 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst2.p0i8.v8i8(i8*, <8 x i8>, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst2.p0i8.v4i16(i8*, <4 x i16>, <4 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst2.p0i8.v2i32(i8*, <2 x i32>, <2 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst2.p0i8.v2f32(i8*, <2 x float>, <2 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst2.p0i8.v1i64(i8*, <1 x i64>, <1 x i64>, i32) nounwind
+
+declare void @llvm.arm.neon.vst2.p0i8.v16i8(i8*, <16 x i8>, <16 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst2.p0i8.v8i16(i8*, <8 x i16>, <8 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst2.p0i8.v4i32(i8*, <4 x i32>, <4 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst2.p0i8.v4f32(i8*, <4 x float>, <4 x float>, i32) nounwind

llvm/trunk/test/CodeGen/ARM/vst3.ll

 ; RUN: llc -mtriple=arm-eabi -mattr=+neon -fast-isel=0 -O0 %s -o - | FileCheck %s
 
 define void @vst3i8(i8* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst3i8:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;This test runs at -O0 so do not check for specific register numbers.
 ;CHECK: vst3.8 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}:64]
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  call void @llvm.arm.neon.vst3.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 32)
+  call void @llvm.arm.neon.vst3.p0i8.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 32)
   ret void
 }
 
 define void @vst3i16(i16* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst3i16:
 ;CHECK: vst3.16
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  call void @llvm.arm.neon.vst3.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst3.p0i8.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1)
   ret void
 }
 
 define void @vst3i32(i32* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst3i32:
 ;CHECK: vst3.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  call void @llvm.arm.neon.vst3.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst3.p0i8.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1)
   ret void
 }
 
 ;Check for a post-increment updating store.
 define void @vst3i32_update(i32** %ptr, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst3i32_update:
 ;CHECK: vst3.32 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}]!
   %A = load i32*, i32** %ptr
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  call void @llvm.arm.neon.vst3.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst3.p0i8.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1)
   %tmp2 = getelementptr i32, i32* %A, i32 6
   store i32* %tmp2, i32** %ptr
   ret void
 }
 
 define void @vst3f(float* %A, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: vst3f:
 ;CHECK: vst3.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <2 x float>, <2 x float>* %B
-  call void @llvm.arm.neon.vst3.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst3.p0i8.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1)
   ret void
 }
 
 define void @vst3i64(i64* %A, <1 x i64>* %B) nounwind {
 ;CHECK-LABEL: vst3i64:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;This test runs at -O0 so do not check for specific register numbers.
 ;CHECK: vst1.64 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}:64]
   %tmp0 = bitcast i64* %A to i8*
   %tmp1 = load <1 x i64>, <1 x i64>* %B
-  call void @llvm.arm.neon.vst3.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 16)
+  call void @llvm.arm.neon.vst3.p0i8.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 16)
   ret void
 }
 
 define void @vst3i64_update(i64** %ptr, <1 x i64>* %B) nounwind {
 ;CHECK-LABEL: vst3i64_update
 ;CHECK: vst1.64 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}]!
         %A = load i64*, i64** %ptr
         %tmp0 = bitcast i64* %A to i8*
         %tmp1 = load <1 x i64>, <1 x i64>* %B
-        call void @llvm.arm.neon.vst3.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 1)
+        call void @llvm.arm.neon.vst3.p0i8.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 1)
         %tmp2 = getelementptr i64, i64* %A, i32 3
         store i64* %tmp2, i64** %ptr
         ret void
 }
 
 define void @vst3Qi8(i8* %A, <16 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst3Qi8:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;This test runs at -O0 so do not check for specific register numbers.
 ;CHECK: vst3.8 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}:64]!
 ;CHECK: vst3.8 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}:64]
   %tmp1 = load <16 x i8>, <16 x i8>* %B
-  call void @llvm.arm.neon.vst3.v16i8(i8* %A, <16 x i8> %tmp1, <16 x i8> %tmp1, <16 x i8> %tmp1, i32 32)
+  call void @llvm.arm.neon.vst3.p0i8.v16i8(i8* %A, <16 x i8> %tmp1, <16 x i8> %tmp1, <16 x i8> %tmp1, i32 32)
   ret void
 }
 
 define void @vst3Qi16(i16* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst3Qi16:
 ;CHECK: vst3.16
 ;CHECK: vst3.16
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  call void @llvm.arm.neon.vst3.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst3.p0i8.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1)
   ret void
 }
 
 ;Check for a post-increment updating store.
 define void @vst3Qi16_update(i16** %ptr, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst3Qi16_update:
 ;CHECK: vst3.16 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}]!
 ;CHECK: vst3.16 {d{{.*}}, d{{.*}}, d{{.*}}}, [r{{.*}}]!
   %A = load i16*, i16** %ptr
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  call void @llvm.arm.neon.vst3.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst3.p0i8.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1)
   %tmp2 = getelementptr i16, i16* %A, i32 24
   store i16* %tmp2, i16** %ptr
   ret void
 }
 
 define void @vst3Qi32(i32* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst3Qi32:
 ;CHECK: vst3.32
 ;CHECK: vst3.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <4 x i32>, <4 x i32>* %B
-  call void @llvm.arm.neon.vst3.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst3.p0i8.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 1)
   ret void
 }
 
 define void @vst3Qf(float* %A, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: vst3Qf:
 ;CHECK: vst3.32
 ;CHECK: vst3.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <4 x float>, <4 x float>* %B
-  call void @llvm.arm.neon.vst3.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst3.p0i8.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
   ret void
 }
 
-declare void @llvm.arm.neon.vst3.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind
-declare void @llvm.arm.neon.vst3.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, i32) nounwind
-declare void @llvm.arm.neon.vst3.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind
-declare void @llvm.arm.neon.vst3.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, i32) nounwind
-declare void @llvm.arm.neon.vst3.v1i64(i8*, <1 x i64>, <1 x i64>, <1 x i64>, i32) nounwind
+declare void @llvm.arm.neon.vst3.p0i8.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst3.p0i8.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst3.p0i8.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst3.p0i8.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst3.p0i8.v1i64(i8*, <1 x i64>, <1 x i64>, <1 x i64>, i32) nounwind
 
-declare void @llvm.arm.neon.vst3.v16i8(i8*, <16 x i8>, <16 x i8>, <16 x i8>, i32) nounwind
-declare void @llvm.arm.neon.vst3.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, i32) nounwind
-declare void @llvm.arm.neon.vst3.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind
-declare void @llvm.arm.neon.vst3.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst3.p0i8.v16i8(i8*, <16 x i8>, <16 x i8>, <16 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst3.p0i8.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst3.p0i8.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst3.p0i8.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, i32) nounwind

llvm/trunk/test/CodeGen/ARM/vst4.ll

 ; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s
 
 define void @vst4i8(i8* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst4i8:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vst4.8 {d16, d17, d18, d19}, [r0:64]
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  call void @llvm.arm.neon.vst4.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 8)
+  call void @llvm.arm.neon.vst4.p0i8.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 8)
   ret void
 }
 
 ;Check for a post-increment updating store with register increment.
 define void @vst4i8_update(i8** %ptr, <8 x i8>* %B, i32 %inc) nounwind {
 ;CHECK-LABEL: vst4i8_update:
 ;CHECK: vst4.8 {d16, d17, d18, d19}, [r1:128], r2
   %A = load i8*, i8** %ptr
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  call void @llvm.arm.neon.vst4.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 16)
+  call void @llvm.arm.neon.vst4.p0i8.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 16)
   %tmp2 = getelementptr i8, i8* %A, i32 %inc
   store i8* %tmp2, i8** %ptr
   ret void
 }
 
 define void @vst4i16(i16* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst4i16:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vst4.16 {d16, d17, d18, d19}, [r0:128]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  call void @llvm.arm.neon.vst4.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 16)
+  call void @llvm.arm.neon.vst4.p0i8.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 16)
   ret void
 }
 
 define void @vst4i32(i32* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst4i32:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vst4.32 {d16, d17, d18, d19}, [r0:256]
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  call void @llvm.arm.neon.vst4.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 32)
+  call void @llvm.arm.neon.vst4.p0i8.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 32)
   ret void
 }
 
 define void @vst4f(float* %A, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: vst4f:
 ;CHECK: vst4.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <2 x float>, <2 x float>* %B
-  call void @llvm.arm.neon.vst4.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst4.p0i8.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1)
   ret void
 }
 
 define void @vst4i64(i64* %A, <1 x i64>* %B) nounwind {
 ;CHECK-LABEL: vst4i64:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vst1.64 {d16, d17, d18, d19}, [r0:256]
   %tmp0 = bitcast i64* %A to i8*
   %tmp1 = load <1 x i64>, <1 x i64>* %B
-  call void @llvm.arm.neon.vst4.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 64)
+  call void @llvm.arm.neon.vst4.p0i8.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 64)
   ret void
 }
 
 define void @vst4i64_update(i64** %ptr, <1 x i64>* %B) nounwind {
 ;CHECK-LABEL: vst4i64_update:
 ;CHECK: vst1.64 {d16, d17, d18, d19}, [r1]!
         %A = load i64*, i64** %ptr
         %tmp0 = bitcast i64* %A to i8*
         %tmp1 = load <1 x i64>, <1 x i64>* %B
-        call void @llvm.arm.neon.vst4.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 1)
+        call void @llvm.arm.neon.vst4.p0i8.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 1)
         %tmp2 = getelementptr i64, i64* %A, i32 4
         store i64* %tmp2, i64** %ptr
         ret void
 }
 
 define void @vst4Qi8(i8* %A, <16 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst4Qi8:
 ;Check the alignment value.  Max for this instruction is 256 bits:
 ;CHECK: vst4.8 {d16, d18, d20, d22}, [r0:256]!
 ;CHECK: vst4.8 {d17, d19, d21, d23}, [r0:256]
   %tmp1 = load <16 x i8>, <16 x i8>* %B
-  call void @llvm.arm.neon.vst4.v16i8(i8* %A, <16 x i8> %tmp1, <16 x i8> %tmp1, <16 x i8> %tmp1, <16 x i8> %tmp1, i32 64)
+  call void @llvm.arm.neon.vst4.p0i8.v16i8(i8* %A, <16 x i8> %tmp1, <16 x i8> %tmp1, <16 x i8> %tmp1, <16 x i8> %tmp1, i32 64)
   ret void
 }
 
 define void @vst4Qi16(i16* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst4Qi16:
 ;Check for no alignment specifier.
 ;CHECK: vst4.16 {d16, d18, d20, d22}, [r0]!
 ;CHECK: vst4.16 {d17, d19, d21, d23}, [r0]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  call void @llvm.arm.neon.vst4.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst4.p0i8.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1)
   ret void
 }
 
 define void @vst4Qi32(i32* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst4Qi32:
 ;CHECK: vst4.32
 ;CHECK: vst4.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <4 x i32>, <4 x i32>* %B
-  call void @llvm.arm.neon.vst4.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst4.p0i8.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 1)
   ret void
 }
 
 define void @vst4Qf(float* %A, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: vst4Qf:
 ;CHECK: vst4.32
 ;CHECK: vst4.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <4 x float>, <4 x float>* %B
-  call void @llvm.arm.neon.vst4.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst4.p0i8.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
   ret void
 }
 
 ;Check for a post-increment updating store.
 define void @vst4Qf_update(float** %ptr, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: vst4Qf_update:
 ;CHECK: vst4.32 {d16, d18, d20, d22}, [r1]!
 ;CHECK: vst4.32 {d17, d19, d21, d23}, [r1]!
   %A = load float*, float** %ptr
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <4 x float>, <4 x float>* %B
-  call void @llvm.arm.neon.vst4.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
+  call void @llvm.arm.neon.vst4.p0i8.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
   %tmp2 = getelementptr float, float* %A, i32 16
   store float* %tmp2, float** %ptr
   ret void
 }
 
-declare void @llvm.arm.neon.vst4.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind
-declare void @llvm.arm.neon.vst4.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32) nounwind
-declare void @llvm.arm.neon.vst4.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind
-declare void @llvm.arm.neon.vst4.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, <2 x float>, i32) nounwind
-declare void @llvm.arm.neon.vst4.v1i64(i8*, <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64>, i32) nounwind
-
-declare void @llvm.arm.neon.vst4.v16i8(i8*, <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8>, i32) nounwind
-declare void @llvm.arm.neon.vst4.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16>, i32) nounwind
-declare void @llvm.arm.neon.vst4.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind
-declare void @llvm.arm.neon.vst4.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst4.p0i8.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst4.p0i8.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst4.p0i8.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst4.p0i8.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, <2 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst4.p0i8.v1i64(i8*, <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64>, i32) nounwind
+
+declare void @llvm.arm.neon.vst4.p0i8.v16i8(i8*, <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst4.p0i8.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16>, i32) nounwind
+declare void @llvm.arm.neon.vst4.p0i8.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst4.p0i8.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32) nounwind

llvm/trunk/test/CodeGen/ARM/vstlane.ll

; // CHE-CK: vst1.32 {d17[1]}, [r0]
   ret void
 }
 
 define void @vst2lanei8(i8* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst2lanei8:
 ;Check the alignment value.  Max for this instruction is 16 bits:
 ;CHECK: vst2.8 {d16[1], d17[1]}, [r0:16]
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  call void @llvm.arm.neon.vst2lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 4)
+  call void @llvm.arm.neon.vst2lane.p0i8.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 4)
   ret void
 }
 
 define void @vst2lanei16(i16* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst2lanei16:
 ;Check the alignment value.  Max for this instruction is 32 bits:
 ;CHECK: vst2.16 {d16[1], d17[1]}, [r0:32]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  call void @llvm.arm.neon.vst2lane.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 8)
+  call void @llvm.arm.neon.vst2lane.p0i8.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 8)
   ret void
 }
 
 ;Check for a post-increment updating store with register increment.
 define void @vst2lanei16_update(i16** %ptr, <4 x i16>* %B, i32 %inc) nounwind {
 ;CHECK-LABEL: vst2lanei16_update:
 ;CHECK: vst2.16 {d16[1], d17[1]}, [r1], r2
   %A = load i16*, i16** %ptr
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  call void @llvm.arm.neon.vst2lane.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 2)
+  call void @llvm.arm.neon.vst2lane.p0i8.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 2)
   %tmp2 = getelementptr i16, i16* %A, i32 %inc
   store i16* %tmp2, i16** %ptr
   ret void
 }
 
 define void @vst2lanei32(i32* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst2lanei32:
 ;CHECK: vst2.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  call void @llvm.arm.neon.vst2lane.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
+  call void @llvm.arm.neon.vst2lane.p0i8.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
   ret void
 }
 
 define void @vst2lanef(float* %A, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: vst2lanef:
 ;CHECK: vst2.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <2 x float>, <2 x float>* %B
-  call void @llvm.arm.neon.vst2lane.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
+  call void @llvm.arm.neon.vst2lane.p0i8.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
   ret void
 }
 
 define void @vst2laneQi16(i16* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst2laneQi16:
 ;Check the (default) alignment.
 ;CHECK: vst2.16 {d17[1], d19[1]}, [r0]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  call void @llvm.arm.neon.vst2lane.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 5, i32 1)
+  call void @llvm.arm.neon.vst2lane.p0i8.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 5, i32 1)
   ret void
 }
 
 define void @vst2laneQi32(i32* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst2laneQi32:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vst2.32 {d17[0], d19[0]}, [r0:64]
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <4 x i32>, <4 x i32>* %B
-  call void @llvm.arm.neon.vst2lane.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 2, i32 16)
+  call void @llvm.arm.neon.vst2lane.p0i8.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 2, i32 16)
   ret void
 }
 
 define void @vst2laneQf(float* %A, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: vst2laneQf:
 ;CHECK: vst2.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <4 x float>, <4 x float>* %B
-  call void @llvm.arm.neon.vst2lane.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, i32 3, i32 1)
+  call void @llvm.arm.neon.vst2lane.p0i8.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, i32 3, i32 1)
   ret void
 }
 
-declare void @llvm.arm.neon.vst2lane.v8i8(i8*, <8 x i8>, <8 x i8>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst2lane.v4i16(i8*, <4 x i16>, <4 x i16>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst2lane.v2i32(i8*, <2 x i32>, <2 x i32>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst2lane.v2f32(i8*, <2 x float>, <2 x float>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.p0i8.v8i8(i8*, <8 x i8>, <8 x i8>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.p0i8.v4i16(i8*, <4 x i16>, <4 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.p0i8.v2i32(i8*, <2 x i32>, <2 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.p0i8.v2f32(i8*, <2 x float>, <2 x float>, i32, i32) nounwind
 
-declare void @llvm.arm.neon.vst2lane.v8i16(i8*, <8 x i16>, <8 x i16>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst2lane.v4i32(i8*, <4 x i32>, <4 x i32>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst2lane.v4f32(i8*, <4 x float>, <4 x float>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.p0i8.v8i16(i8*, <8 x i16>, <8 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.p0i8.v4i32(i8*, <4 x i32>, <4 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst2lane.p0i8.v4f32(i8*, <4 x float>, <4 x float>, i32, i32) nounwind
 
 define void @vst3lanei8(i8* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst3lanei8:
 ;CHECK: vst3.8
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  call void @llvm.arm.neon.vst3lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 1)
+  call void @llvm.arm.neon.vst3lane.p0i8.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 1)
   ret void
 }
 
 define void @vst3lanei16(i16* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst3lanei16:
 ;Check the (default) alignment value.  VST3 does not support alignment.
 ;CHECK: vst3.16 {d16[1], d17[1], d18[1]}, [r0]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  call void @llvm.arm.neon.vst3lane.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 8)
+  call void @llvm.arm.neon.vst3lane.p0i8.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 8)
   ret void
 }
 
 define void @vst3lanei32(i32* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst3lanei32:
 ;CHECK: vst3.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  call void @llvm.arm.neon.vst3lane.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
+  call void @llvm.arm.neon.vst3lane.p0i8.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 1)
   ret void
 }
 
 define void @vst3lanef(float* %A, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: vst3lanef:
 ;CHECK: vst3.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <2 x float>, <2 x float>* %B
-  call void @llvm.arm.neon.vst3lane.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
+  call void @llvm.arm.neon.vst3lane.p0i8.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
   ret void
 }
 
 define void @vst3laneQi16(i16* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst3laneQi16:
 ;Check the (default) alignment value.  VST3 does not support alignment.
 ;CHECK: vst3.16 {d17[2], d19[2], d21[2]}, [r0]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  call void @llvm.arm.neon.vst3lane.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 6, i32 8)
+  call void @llvm.arm.neon.vst3lane.p0i8.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 6, i32 8)
   ret void
 }
 
 define void @vst3laneQi32(i32* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst3laneQi32:
 ;CHECK: vst3.32
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <4 x i32>, <4 x i32>* %B
-  call void @llvm.arm.neon.vst3lane.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 0, i32 1)
+  call void @llvm.arm.neon.vst3lane.p0i8.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 0, i32 1)
   ret void
 }
 
 ;Check for a post-increment updating store.
 define void @vst3laneQi32_update(i32** %ptr, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst3laneQi32_update:
 ;CHECK: vst3.32 {d16[0], d18[0], d20[0]}, [r1]!
   %A = load i32*, i32** %ptr
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <4 x i32>, <4 x i32>* %B
-  call void @llvm.arm.neon.vst3lane.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 0, i32 1)
+  call void @llvm.arm.neon.vst3lane.p0i8.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 0, i32 1)
   %tmp2 = getelementptr i32, i32* %A, i32 3
   store i32* %tmp2, i32** %ptr
   ret void
 }
 
 define void @vst3laneQf(float* %A, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: vst3laneQf:
 ;CHECK: vst3.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <4 x float>, <4 x float>* %B
-  call void @llvm.arm.neon.vst3lane.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1, i32 1)
+  call void @llvm.arm.neon.vst3lane.p0i8.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1, i32 1)
   ret void
 }
 
-declare void @llvm.arm.neon.vst3lane.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst3lane.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst3lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst3lane.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.p0i8.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.p0i8.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.p0i8.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.p0i8.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, i32, i32) nounwind
 
-declare void @llvm.arm.neon.vst3lane.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst3lane.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst3lane.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.p0i8.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.p0i8.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst3lane.p0i8.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, i32, i32) nounwind
 
 
 define void @vst4lanei8(i8* %A, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst4lanei8:
 ;Check the alignment value.  Max for this instruction is 32 bits:
 ;CHECK: vst4.8 {d16[1], d17[1], d18[1], d19[1]}, [r0:32]
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  call void @llvm.arm.neon.vst4lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 8)
+  call void @llvm.arm.neon.vst4lane.p0i8.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 8)
   ret void
 }
 
 ;Check for a post-increment updating store.
 define void @vst4lanei8_update(i8** %ptr, <8 x i8>* %B) nounwind {
 ;CHECK-LABEL: vst4lanei8_update:
 ;CHECK: vst4.8 {d16[1], d17[1], d18[1], d19[1]}, [r1:32]!
   %A = load i8*, i8** %ptr
   %tmp1 = load <8 x i8>, <8 x i8>* %B
-  call void @llvm.arm.neon.vst4lane.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 8)
+  call void @llvm.arm.neon.vst4lane.p0i8.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 1, i32 8)
   %tmp2 = getelementptr i8, i8* %A, i32 4
   store i8* %tmp2, i8** %ptr
   ret void
 }
 
 define void @vst4lanei16(i16* %A, <4 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst4lanei16:
 ;CHECK: vst4.16
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <4 x i16>, <4 x i16>* %B
-  call void @llvm.arm.neon.vst4lane.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 1)
+  call void @llvm.arm.neon.vst4lane.p0i8.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 1, i32 1)
   ret void
 }
 
 define void @vst4lanei32(i32* %A, <2 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst4lanei32:
 ;Check the alignment value.  Max for this instruction is 128 bits:
 ;CHECK: vst4.32 {d16[1], d17[1], d18[1], d19[1]}, [r0:128]
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <2 x i32>, <2 x i32>* %B
-  call void @llvm.arm.neon.vst4lane.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 16)
+  call void @llvm.arm.neon.vst4lane.p0i8.v2i32(i8* %tmp0, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, <2 x i32> %tmp1, i32 1, i32 16)
   ret void
 }
 
 define void @vst4lanef(float* %A, <2 x float>* %B) nounwind {
 ;CHECK-LABEL: vst4lanef:
 ;CHECK: vst4.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <2 x float>, <2 x float>* %B
-  call void @llvm.arm.neon.vst4lane.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
+  call void @llvm.arm.neon.vst4lane.p0i8.v2f32(i8* %tmp0, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, <2 x float> %tmp1, i32 1, i32 1)
   ret void
 }
 
 define void @vst4laneQi16(i16* %A, <8 x i16>* %B) nounwind {
 ;CHECK-LABEL: vst4laneQi16:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vst4.16 {d17[3], d19[3], d21[3], d23[3]}, [r0:64]
   %tmp0 = bitcast i16* %A to i8*
   %tmp1 = load <8 x i16>, <8 x i16>* %B
-  call void @llvm.arm.neon.vst4lane.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 7, i32 16)
+  call void @llvm.arm.neon.vst4lane.p0i8.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 7, i32 16)
   ret void
 }
 
 define void @vst4laneQi32(i32* %A, <4 x i32>* %B) nounwind {
 ;CHECK-LABEL: vst4laneQi32:
 ;Check the (default) alignment.
 ;CHECK: vst4.32 {d17[0], d19[0], d21[0], d23[0]}, [r0]
   %tmp0 = bitcast i32* %A to i8*
   %tmp1 = load <4 x i32>, <4 x i32>* %B
-  call void @llvm.arm.neon.vst4lane.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 2, i32 1)
+  call void @llvm.arm.neon.vst4lane.p0i8.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 2, i32 1)
   ret void
 }
 
 define void @vst4laneQf(float* %A, <4 x float>* %B) nounwind {
 ;CHECK-LABEL: vst4laneQf:
 ;CHECK: vst4.32
   %tmp0 = bitcast float* %A to i8*
   %tmp1 = load <4 x float>, <4 x float>* %B
-  call void @llvm.arm.neon.vst4lane.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1, i32 1)
+  call void @llvm.arm.neon.vst4lane.p0i8.v4f32(i8* %tmp0, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1, i32 1)
   ret void
 }
 
 ; Make sure this doesn't crash; PR10258
 define <8 x i16> @variable_insertelement(<8 x i16> %a, i16 %b, i32 %c) nounwind readnone {
 ;CHECK-LABEL: variable_insertelement:
     %r = insertelement <8 x i16> %a, i16 %b, i32 %c
     ret <8 x i16> %r
 }
 
-declare void @llvm.arm.neon.vst4lane.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst4lane.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst4lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst4lane.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, <2 x float>, i32, i32) nounwind
-
-declare void @llvm.arm.neon.vst4lane.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst4lane.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32) nounwind
-declare void @llvm.arm.neon.vst4lane.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.p0i8.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.p0i8.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.p0i8.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.p0i8.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, <2 x float>, i32, i32) nounwind
+
+declare void @llvm.arm.neon.vst4lane.p0i8.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.p0i8.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32, i32) nounwind
+declare void @llvm.arm.neon.vst4lane.p0i8.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32, i32) nounwind

llvm/trunk/test/CodeGen/Thumb2/crash.ll

entry:
   %1 = load <4 x i32>, <4 x i32>* %0, align 16               ; <<4 x i32>> [#uses=1]
   %2 = bitcast i32* %sp1 to <4 x i32>*            ; <<4 x i32>*> [#uses=1]
   %3 = load <4 x i32>, <4 x i32>* %2, align 16               ; <<4 x i32>> [#uses=1]
   %4 = bitcast i32* %sp2 to <4 x i32>*            ; <<4 x i32>*> [#uses=1]
   %5 = load <4 x i32>, <4 x i32>* %4, align 16               ; <<4 x i32>> [#uses=1]
   %6 = bitcast i32* %sp3 to <4 x i32>*            ; <<4 x i32>*> [#uses=1]
   %7 = load <4 x i32>, <4 x i32>* %6, align 16               ; <<4 x i32>> [#uses=1]
   %8 = bitcast i32* %dp to i8*                    ; <i8*> [#uses=1]
-  tail call void @llvm.arm.neon.vst4.v4i32(i8* %8, <4 x i32> %1, <4 x i32> %3, <4 x i32> %5, <4 x i32> %7, i32 1)
+  tail call void @llvm.arm.neon.vst4.p0i8.v4i32(i8* %8, <4 x i32> %1, <4 x i32> %3, <4 x i32> %5, <4 x i32> %7, i32 1)
   ret void
 }
 
-declare void @llvm.arm.neon.vst4.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind
+declare void @llvm.arm.neon.vst4.p0i8.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind
 
 @sbuf = common global [16 x i32] zeroinitializer, align 16 ; <[16 x i32]*> [#uses=5]
 @dbuf = common global [16 x i32] zeroinitializer  ; <[16 x i32]*> [#uses=2]
 
 ; This function creates 4 chained INSERT_SUBREGS and then invokes the register scavenger.
 ; The first INSERT_SUBREG needs an <undef> use operand for that to work.
 define arm_apcscc i32 @main() nounwind {
 bb.nph:
bb:                                               ; preds = %bb, %bb.nph
   %exitcond = icmp eq i32 %1, 16                  ; <i1> [#uses=1]
   br i1 %exitcond, label %bb2, label %bb
 
 bb2:                                              ; preds = %bb
   %2 = load <4 x i32>, <4 x i32>* bitcast ([16 x i32]* @sbuf to <4 x i32>*), align 16 ; <<4 x i32>> [#uses=1]
   %3 = load <4 x i32>, <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @sbuf, i32 0, i32 4) to <4 x i32>*), align 16 ; <<4 x i32>> [#uses=1]
   %4 = load <4 x i32>, <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @sbuf, i32 0, i32 8) to <4 x i32>*), align 16 ; <<4 x i32>> [#uses=1]
   %5 = load <4 x i32>, <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @sbuf, i32 0, i32 12) to <4 x i32>*), align 16 ; <<4 x i32>> [#uses=1]
-  tail call void @llvm.arm.neon.vst4.v4i32(i8* bitcast ([16 x i32]* @dbuf to i8*), <4 x i32> %2, <4 x i32> %3, <4 x i32> %4, <4 x i32> %5, i32 1) nounwind
+  tail call void @llvm.arm.neon.vst4.p0i8.v4i32(i8* bitcast ([16 x i32]* @dbuf to i8*), <4 x i32> %2, <4 x i32> %3, <4 x i32> %4, <4 x i32> %5, i32 1) nounwind
   ret i32 0
 }
 
 ; PR12389
 ; Make sure the DPair register class can spill.
 define void @pr12389(i8* %p) nounwind ssp {
 entry:
-  %vld1 = tail call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* %p, i32 1)
+  %vld1 = tail call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* %p, i32 1)
   tail call void asm sideeffect "", "~{q0},~{q1},~{q2},~{q3},~{q4},~{q5},~{q6},~{q7},~{q8},~{q9},~{q10},~{q11},~{q12},~{q13},~{q14},~{q15}"() nounwind
-  tail call void @llvm.arm.neon.vst1.v4f32(i8* %p, <4 x float> %vld1, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v4f32(i8* %p, <4 x float> %vld1, i32 1)
   ret void
 }
 
-declare <4 x float> @llvm.arm.neon.vld1.v4f32(i8*, i32) nounwind readonly
+declare <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8*, i32) nounwind readonly
 
-declare void @llvm.arm.neon.vst1.v4f32(i8*, <4 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v4f32(i8*, <4 x float>, i32) nounwind
 
 ; <rdar://problem/11101911>
 ; When an strd is expanded into two str instructions, make sure the first str
 ; doesn't kill the base register. This can happen if the base register is the
 ; same as the data register.
 %class = type { i8*, %class*, i32 }
 define void @f11101911(%class* %this, i32 %num) ssp align 2 {
 entry:

llvm/trunk/test/CodeGen/Thumb2/machine-licm.ll

 ; CHECK: vmov.f32 q{{.*}}, #1.000000e+00
   br i1 undef, label %bb1, label %bb2
 
 bb1:
 ; CHECK: %bb1
   %indvar = phi i32 [ %indvar.next, %bb1 ], [ 0, %entry ]
   %tmp1 = shl i32 %indvar, 2
   %gep1 = getelementptr i8, i8* %ptr1, i32 %tmp1
-  %tmp2 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* %gep1, i32 1)
+  %tmp2 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* %gep1, i32 1)
   %tmp3 = call <4 x float> @llvm.arm.neon.vmaxs.v4f32(<4 x float> <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>, <4 x float> %tmp2)
   %gep2 = getelementptr i8, i8* %ptr2, i32 %tmp1
-  call void @llvm.arm.neon.vst1.v4f32(i8* %gep2, <4 x float> %tmp3, i32 1)
+  call void @llvm.arm.neon.vst1.p0i8.v4f32(i8* %gep2, <4 x float> %tmp3, i32 1)
   %indvar.next = add i32 %indvar, 1
   %cond = icmp eq i32 %indvar.next, 10
   br i1 %cond, label %bb2, label %bb1
 
 bb2:
   ret void
 }
 
 ; CHECK-NOT: LCPI1_0:
 
-declare <4 x float> @llvm.arm.neon.vld1.v4f32(i8*, i32) nounwind readonly
+declare <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8*, i32) nounwind readonly
 
-declare void @llvm.arm.neon.vst1.v4f32(i8*, <4 x float>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v4f32(i8*, <4 x float>, i32) nounwind
 
 declare <4 x float> @llvm.arm.neon.vmaxs.v4f32(<4 x float>, <4 x float>) nounwind readnone
 
 ; rdar://8241368
 ; isel should not fold immediate into eor's which would have prevented LICM.
 define zeroext i16 @t3(i8 zeroext %data, i16 zeroext %crc) nounwind readnone {
 ; CHECK-LABEL: t3:
 bb.nph:

llvm/trunk/test/CodeGen/Thumb2/thumb2-spill-q.ll

 ; RUN: llc < %s -mtriple=thumbv7-elf -mattr=+neon -arm-atomic-cfg-tidy=0 | FileCheck %s
 ; PR4789
 
 %bar = type { float, float, float }
 %baz = type { i32, [16 x %bar], [16 x float], [16 x i32], i8 }
 %foo = type { <4 x float> }
 %quux = type { i32 (...)**, %baz*, i32 }
 %quuz = type { %quux, i32, %bar, [128 x i8], [16 x %foo], %foo, %foo, %foo }
 
-declare <4 x float> @llvm.arm.neon.vld1.v4f32(i8*, i32) nounwind readonly
+declare <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8*, i32) nounwind readonly
 
 define void @aaa(%quuz* %this, i8* %block) {
 ; CHECK-LABEL: aaa:
 ; CHECK: bfc r4, #0, #4
 ; CHECK: vst1.64 {{.*}}[{{.*}}:128]
 ; CHECK: vld1.64 {{.*}}[{{.*}}:128]
 entry:
   %aligned_vec = alloca <4 x float>, align 16
   %"alloca point" = bitcast i32 0 to i32
   %vecptr = bitcast <4 x float>* %aligned_vec to i8*
-  %0 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* %vecptr, i32 1) nounwind 
+  %0 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* %vecptr, i32 1) nounwind 
   store float 6.300000e+01, float* undef, align 4
-  %1 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind ; <<4 x float>> [#uses=1]
+  %1 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind ; <<4 x float>> [#uses=1]
   store float 0.000000e+00, float* undef, align 4
-  %2 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind ; <<4 x float>> [#uses=1]
-  %ld3 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %2 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind ; <<4 x float>> [#uses=1]
+  %ld3 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld4 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld4 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld5 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld5 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld6 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld6 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld7 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld7 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld8 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld8 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld9 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld9 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld10 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld10 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld11 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld11 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
-  %ld12 = call <4 x float> @llvm.arm.neon.vld1.v4f32(i8* undef, i32 1) nounwind
+  %ld12 = call <4 x float> @llvm.arm.neon.vld1.v4f32.p0i8(i8* undef, i32 1) nounwind
   store float 0.000000e+00, float* undef, align 4
   %val173 = load <4 x float>, <4 x float>* undef               ; <<4 x float>> [#uses=1]
   br label %bb4
 
 bb4:                                              ; preds = %bb193, %entry
   %besterror.0.2264 = phi <4 x float> [ undef, %entry ], [ %besterror.0.0, %bb193 ] ; <<4 x float>> [#uses=2]
   %part0.0.0261 = phi <4 x float> [ zeroinitializer, %entry ], [ %23, %bb193 ] ; <<4 x float>> [#uses=2]
   %3 = fmul <4 x float> zeroinitializer, %0       ; <<4 x float>> [#uses=2]

llvm/trunk/test/CodeGen/Thumb2/v8_IT_1.ll

 ; RUN: llc < %s -mtriple=thumbv8 -mattr=+neon | FileCheck %s
 ; RUN: llc < %s -mtriple=thumbv7 -mattr=+neon -arm-restrict-it | FileCheck %s
 
 ;CHECK-LABEL: select_s_v_v:
 ;CHECK-NOT: it
 ;CHECK: bx
 define <16 x i8> @select_s_v_v(i32 %avail, i8* %bar) {
 entry:
-  %vld1 = call <16 x i8> @llvm.arm.neon.vld1.v16i8(i8* %bar, i32 1)
+  %vld1 = call <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* %bar, i32 1)
   %and = and i32 %avail, 1
   %tobool = icmp eq i32 %and, 0
   %vld1. = select i1 %tobool, <16 x i8> %vld1, <16 x i8> zeroinitializer
   ret <16 x i8> %vld1.
 }
 
-declare <16 x i8> @llvm.arm.neon.vld1.v16i8(i8* , i32 )
+declare <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* , i32 )
 

llvm/trunk/test/Transforms/InstCombine/neon-intrinsics.ll

 ; RUN: opt < %s -instcombine -S | FileCheck %s
 
 ; The alignment arguments for NEON load/store intrinsics can be increased
 ; by instcombine.  Check for this.
 
-; CHECK: vld4.v2i32({{.*}}, i32 32)
-; CHECK: vst4.v2i32({{.*}}, i32 16)
+; CHECK: vld4.v2i32.p0i8({{.*}}, i32 32)
+; CHECK: vst4.p0i8.v2i32({{.*}}, i32 16)
 
 @x = common global [8 x i32] zeroinitializer, align 32
 @y = common global [8 x i32] zeroinitializer, align 16
 
 %struct.__neon_int32x2x4_t = type { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> }
 
 define void @test() nounwind ssp {
-  %tmp1 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8* bitcast ([8 x i32]* @x to i8*), i32 1)
+  %tmp1 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32.p0i8(i8* bitcast ([8 x i32]* @x to i8*), i32 1)
   %tmp2 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 0
   %tmp3 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 1
   %tmp4 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 2
   %tmp5 = extractvalue %struct.__neon_int32x2x4_t %tmp1, 3
-  call void @llvm.arm.neon.vst4.v2i32(i8* bitcast ([8 x i32]* @y to i8*), <2 x i32> %tmp2, <2 x i32> %tmp3, <2 x i32> %tmp4, <2 x i32> %tmp5, i32 1)
+  call void @llvm.arm.neon.vst4.p0i8.v2i32(i8* bitcast ([8 x i32]* @y to i8*), <2 x i32> %tmp2, <2 x i32> %tmp3, <2 x i32> %tmp4, <2 x i32> %tmp5, i32 1)
   ret void
 }
 
-declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8*, i32) nounwind readonly
-declare void @llvm.arm.neon.vst4.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind
+declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32.p0i8(i8*, i32) nounwind readonly
+declare void @llvm.arm.neon.vst4.p0i8.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind

llvm/trunk/test/Transforms/LoopStrengthReduce/ARM/ivchain-ARM.ll

.lr.ph:                                           ; preds = %0
   %10 = mul i32 %limit, -64
   br label %11
 
 ; <label>:11                                      ; preds = %11, %.lr.ph
   %.05 = phi i8* [ %ref_data, %.lr.ph ], [ %42, %11 ]
   %counter.04 = phi i32 [ 0, %.lr.ph ], [ %44, %11 ]
   %result.03 = phi <16 x i8> [ zeroinitializer, %.lr.ph ], [ %41, %11 ]
   %.012 = phi <16 x i8>* [ %data, %.lr.ph ], [ %43, %11 ]
-  %12 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %.05, i32 1) nounwind
+  %12 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8* %.05, i32 1) nounwind
   %13 = getelementptr inbounds i8, i8* %.05, i32 %ref_stride
-  %14 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %13, i32 1) nounwind
+  %14 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8* %13, i32 1) nounwind
   %15 = shufflevector <1 x i64> %12, <1 x i64> %14, <2 x i32> <i32 0, i32 1>
   %16 = bitcast <2 x i64> %15 to <16 x i8>
   %17 = getelementptr inbounds <16 x i8>, <16 x i8>* %.012, i32 1
   store <16 x i8> %16, <16 x i8>* %.012, align 4
   %18 = getelementptr inbounds i8, i8* %.05, i32 %2
-  %19 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %18, i32 1) nounwind
+  %19 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8* %18, i32 1) nounwind
   %20 = getelementptr inbounds i8, i8* %.05, i32 %3
-  %21 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %20, i32 1) nounwind
+  %21 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8* %20, i32 1) nounwind
   %22 = shufflevector <1 x i64> %19, <1 x i64> %21, <2 x i32> <i32 0, i32 1>
   %23 = bitcast <2 x i64> %22 to <16 x i8>
   %24 = getelementptr inbounds <16 x i8>, <16 x i8>* %.012, i32 2
   store <16 x i8> %23, <16 x i8>* %17, align 4
   %25 = getelementptr inbounds i8, i8* %.05, i32 %4
-  %26 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %25, i32 1) nounwind
+  %26 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8* %25, i32 1) nounwind
   %27 = getelementptr inbounds i8, i8* %.05, i32 %5
-  %28 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %27, i32 1) nounwind
+  %28 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8* %27, i32 1) nounwind
   %29 = shufflevector <1 x i64> %26, <1 x i64> %28, <2 x i32> <i32 0, i32 1>
   %30 = bitcast <2 x i64> %29 to <16 x i8>
   %31 = getelementptr inbounds <16 x i8>, <16 x i8>* %.012, i32 3
   store <16 x i8> %30, <16 x i8>* %24, align 4
   %32 = getelementptr inbounds i8, i8* %.05, i32 %6
-  %33 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %32, i32 1) nounwind
+  %33 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8* %32, i32 1) nounwind
   %34 = getelementptr inbounds i8, i8* %.05, i32 %7
-  %35 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %34, i32 1) nounwind
+  %35 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8* %34, i32 1) nounwind
   %36 = shufflevector <1 x i64> %33, <1 x i64> %35, <2 x i32> <i32 0, i32 1>
   %37 = bitcast <2 x i64> %36 to <16 x i8>
   store <16 x i8> %37, <16 x i8>* %31, align 4
   %38 = add <16 x i8> %16, %23
   %39 = add <16 x i8> %38, %30
   %40 = add <16 x i8> %39, %37
   %41 = add <16 x i8> %result.03, %40
   %42 = getelementptr i8, i8* %.05, i32 %9
   %43 = getelementptr inbounds <16 x i8>, <16 x i8>* %.012, i32 -64
   %44 = add nsw i32 %counter.04, 1
   %exitcond = icmp eq i32 %44, %limit
   br i1 %exitcond, label %._crit_edge, label %11
 
 ._crit_edge:                                      ; preds = %11
   %scevgep = getelementptr <16 x i8>, <16 x i8>* %data, i32 %10
   br label %45
 
 ; <label>:45                                      ; preds = %._crit_edge, %0
   %result.0.lcssa = phi <16 x i8> [ %41, %._crit_edge ], [ zeroinitializer, %0 ]
   %.01.lcssa = phi <16 x i8>* [ %scevgep, %._crit_edge ], [ %data, %0 ]
   store <16 x i8> %result.0.lcssa, <16 x i8>* %.01.lcssa, align 4
   ret void
 }
 
-declare <1 x i64> @llvm.arm.neon.vld1.v1i64(i8*, i32) nounwind readonly
+declare <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8*, i32) nounwind readonly
 
 ; Handle chains in which the same offset is used for both loads and
 ; stores to the same array.
 ; rdar://11410078.
 ;
 ; A9: @testReuse
 ; A9: %for.body
 ; A9: vld1.8 {d{{[0-9]+}}}, [[BASE:[r[0-9]+]]], [[INC:r[0-9]]]
entry:
   %idx.neg6 = sub i32 0, %mul5
   %idx.neg10 = sub i32 0, %stride
   br label %for.body
 
 for.body:                                         ; preds = %for.body, %entry
   %i.0110 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
   %src.addr = phi i8* [ %src, %entry ], [ %add.ptr45, %for.body ]
   %add.ptr = getelementptr inbounds i8, i8* %src.addr, i32 %idx.neg
-  %vld1 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr, i32 1)
+  %vld1 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8.p0i8(i8* %add.ptr, i32 1)
   %add.ptr3 = getelementptr inbounds i8, i8* %src.addr, i32 %idx.neg2
-  %vld2 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr3, i32 1)
+  %vld2 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8.p0i8(i8* %add.ptr3, i32 1)
   %add.ptr7 = getelementptr inbounds i8, i8* %src.addr, i32 %idx.neg6
-  %vld3 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr7, i32 1)
+  %vld3 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8.p0i8(i8* %add.ptr7, i32 1)
   %add.ptr11 = getelementptr inbounds i8, i8* %src.addr, i32 %idx.neg10
-  %vld4 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr11, i32 1)
-  %vld5 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %src.addr, i32 1)
+  %vld4 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8.p0i8(i8* %add.ptr11, i32 1)
+  %vld5 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8.p0i8(i8* %src.addr, i32 1)
   %add.ptr17 = getelementptr inbounds i8, i8* %src.addr, i32 %stride
-  %vld6 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr17, i32 1)
+  %vld6 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8.p0i8(i8* %add.ptr17, i32 1)
   %add.ptr20 = getelementptr inbounds i8, i8* %src.addr, i32 %mul5
-  %vld7 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr20, i32 1)
+  %vld7 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8.p0i8(i8* %add.ptr20, i32 1)
   %add.ptr23 = getelementptr inbounds i8, i8* %src.addr, i32 %mul1
-  %vld8 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr23, i32 1)
+  %vld8 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8.p0i8(i8* %add.ptr23, i32 1)
   %vadd1 = tail call <8 x i8> @llvm.arm.neon.vhaddu.v8i8(<8 x i8> %vld1, <8 x i8> %vld2) nounwind
   %vadd2 = tail call <8 x i8> @llvm.arm.neon.vhaddu.v8i8(<8 x i8> %vld2, <8 x i8> %vld3) nounwind
   %vadd3 = tail call <8 x i8> @llvm.arm.neon.vhaddu.v8i8(<8 x i8> %vld3, <8 x i8> %vld4) nounwind
   %vadd4 = tail call <8 x i8> @llvm.arm.neon.vhaddu.v8i8(<8 x i8> %vld4, <8 x i8> %vld5) nounwind
   %vadd5 = tail call <8 x i8> @llvm.arm.neon.vhaddu.v8i8(<8 x i8> %vld5, <8 x i8> %vld6) nounwind
   %vadd6 = tail call <8 x i8> @llvm.arm.neon.vhaddu.v8i8(<8 x i8> %vld6, <8 x i8> %vld7) nounwind
-  tail call void @llvm.arm.neon.vst1.v8i8(i8* %add.ptr3, <8 x i8> %vadd1, i32 1)
-  tail call void @llvm.arm.neon.vst1.v8i8(i8* %add.ptr7, <8 x i8> %vadd2, i32 1)
-  tail call void @llvm.arm.neon.vst1.v8i8(i8* %add.ptr11, <8 x i8> %vadd3, i32 1)
-  tail call void @llvm.arm.neon.vst1.v8i8(i8* %src.addr, <8 x i8> %vadd4, i32 1)
-  tail call void @llvm.arm.neon.vst1.v8i8(i8* %add.ptr17, <8 x i8> %vadd5, i32 1)
-  tail call void @llvm.arm.neon.vst1.v8i8(i8* %add.ptr20, <8 x i8> %vadd6, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i8(i8* %add.ptr3, <8 x i8> %vadd1, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i8(i8* %add.ptr7, <8 x i8> %vadd2, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i8(i8* %add.ptr11, <8 x i8> %vadd3, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i8(i8* %src.addr, <8 x i8> %vadd4, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i8(i8* %add.ptr17, <8 x i8> %vadd5, i32 1)
+  tail call void @llvm.arm.neon.vst1.p0i8.v8i8(i8* %add.ptr20, <8 x i8> %vadd6, i32 1)
   %inc = add nsw i32 %i.0110, 1
   %add.ptr45 = getelementptr inbounds i8, i8* %src.addr, i32 8
   %exitcond = icmp eq i32 %inc, 4
   br i1 %exitcond, label %for.end, label %for.body
 
 for.end:                                          ; preds = %for.body
   ret void
 }
 
-declare <8 x i8> @llvm.arm.neon.vld1.v8i8(i8*, i32) nounwind readonly
+declare <8 x i8> @llvm.arm.neon.vld1.v8i8.p0i8(i8*, i32) nounwind readonly
 
-declare void @llvm.arm.neon.vst1.v8i8(i8*, <8 x i8>, i32) nounwind
+declare void @llvm.arm.neon.vst1.p0i8.v8i8(i8*, <8 x i8>, i32) nounwind
 
 declare <8 x i8> @llvm.arm.neon.vhaddu.v8i8(<8 x i8>, <8 x i8>) nounwind readnone