diff --git a/lib/Transforms/InstCombine/InstCombineCalls.cpp b/lib/Transforms/InstCombine/InstCombineCalls.cpp
index ed078ec1c51..d53f3291e71 100644
--- a/lib/Transforms/InstCombine/InstCombineCalls.cpp
+++ b/lib/Transforms/InstCombine/InstCombineCalls.cpp
@@ -702,6 +702,32 @@ Instruction *InstCombiner::visitCallInst(CallInst &CI) {
     }
     break;
 
+  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: {
+    unsigned MemAlign = GetOrEnforceKnownAlignment(II->getArgOperand(0));
+    unsigned AlignArg = II->getNumArgOperands() - 1;
+    ConstantInt *IntrAlign = dyn_cast<ConstantInt>(II->getArgOperand(AlignArg));
+    if (IntrAlign && IntrAlign->getZExtValue() < MemAlign) {
+      II->setArgOperand(AlignArg,
+                        ConstantInt::get(Type::getInt32Ty(II->getContext()),
+                                         MemAlign, false));
+      return II;
+    }
+    break;
+  }
+
   case Intrinsic::stackrestore: {
     // If the save is right next to the restore, remove the restore.  This can
     // happen when variable allocas are DCE'd.
diff --git a/test/Transforms/InstCombine/neon-intrinsics.ll b/test/Transforms/InstCombine/neon-intrinsics.ll
new file mode 100644
index 00000000000..3ad09cc6c69
--- /dev/null
+++ b/test/Transforms/InstCombine/neon-intrinsics.ll
@@ -0,0 +1,25 @@
+; 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)
+
+@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)
+  %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)
+  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