diff --git a/lib/Target/ARM/ARMISelDAGToDAG.cpp b/lib/Target/ARM/ARMISelDAGToDAG.cpp
index 713cc2e..27b0ed2 100644
--- a/lib/Target/ARM/ARMISelDAGToDAG.cpp
+++ b/lib/Target/ARM/ARMISelDAGToDAG.cpp
@@ -1357,6 +1357,7 @@
         case MVT::v4i16: Opc = ARM::VLD2d16; break;
         case MVT::v2f32:
         case MVT::v2i32: Opc = ARM::VLD2d32; break;
+        case MVT::v1i64: Opc = ARM::VLD2d64; break;
         }
         SDValue Chain = N->getOperand(0);
         const SDValue Ops[] = { MemAddr, MemUpdate, MemOpc, Chain };
diff --git a/lib/Target/ARM/ARMInstrNEON.td b/lib/Target/ARM/ARMInstrNEON.td
index 334f496..11117cc 100644
--- a/lib/Target/ARM/ARMInstrNEON.td
+++ b/lib/Target/ARM/ARMInstrNEON.td
@@ -193,6 +193,9 @@
 def  VLD2d8   : VLD2D<0b0000, "vld2.8">;
 def  VLD2d16  : VLD2D<0b0100, "vld2.16">;
 def  VLD2d32  : VLD2D<0b1000, "vld2.32">;
+def  VLD2d64  : NLdSt<0,0b10,0b1010,0b1100, (outs DPR:$dst1, DPR:$dst2),
+                      (ins addrmode6:$addr), IIC_VLD1,
+                      "vld1.64\t\\{$dst1,$dst2\\}, $addr", "", []>;
 
 def  VLD2q8   : VLD2Q<0b0000, "vld2.8">;
 def  VLD2q16  : VLD2Q<0b0100, "vld2.16">;
diff --git a/lib/Target/ARM/NEONPreAllocPass.cpp b/lib/Target/ARM/NEONPreAllocPass.cpp
index 0c167a1..e795e1a 100644
--- a/lib/Target/ARM/NEONPreAllocPass.cpp
+++ b/lib/Target/ARM/NEONPreAllocPass.cpp
@@ -49,6 +49,7 @@
   case ARM::VLD2d8:
   case ARM::VLD2d16:
   case ARM::VLD2d32:
+  case ARM::VLD2d64:
   case ARM::VLD2LNd8:
   case ARM::VLD2LNd16:
   case ARM::VLD2LNd32:
diff --git a/test/CodeGen/ARM/vld2.ll b/test/CodeGen/ARM/vld2.ll
index 4456f60..23f7d2c 100644
--- a/test/CodeGen/ARM/vld2.ll
+++ b/test/CodeGen/ARM/vld2.ll
@@ -4,6 +4,7 @@
 %struct.__neon_int16x4x2_t = type { <4 x i16>, <4 x i16> }
 %struct.__neon_int32x2x2_t = type { <2 x i32>, <2 x i32> }
 %struct.__neon_float32x2x2_t = type { <2 x float>, <2 x float> }
+%struct.__neon_int64x1x2_t = type { <1 x i64>, <1 x i64> }
 
 %struct.__neon_int8x16x2_t = type { <16 x i8>,  <16 x i8> }
 %struct.__neon_int16x8x2_t = type { <8 x i16>, <8 x i16> }
@@ -50,6 +51,16 @@
 	ret <2 x float> %tmp4
 }
 
+define <1 x i64> @vld2i64(i64* %A) nounwind {
+;CHECK: vld2i64:
+;CHECK: vld1.64
+	%tmp1 = call %struct.__neon_int64x1x2_t @llvm.arm.neon.vld2.v1i64(i64* %A)
+        %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: vld2Qi8:
 ;CHECK: vld2.8
@@ -94,6 +105,7 @@
 declare %struct.__neon_int16x4x2_t @llvm.arm.neon.vld2.v4i16(i8*) nounwind readonly
 declare %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2.v2i32(i8*) nounwind readonly
 declare %struct.__neon_float32x2x2_t @llvm.arm.neon.vld2.v2f32(i8*) nounwind readonly
+declare %struct.__neon_int64x1x2_t @llvm.arm.neon.vld2.v1i64(i8*) nounwind readonly
 
 declare %struct.__neon_int8x16x2_t @llvm.arm.neon.vld2.v16i8(i8*) nounwind readonly
 declare %struct.__neon_int16x8x2_t @llvm.arm.neon.vld2.v8i16(i8*) nounwind readonly