diff --git a/test/CodeGen/ARM/vcge.ll b/test/CodeGen/ARM/vcge.ll
index 6dfc6dd..6017d41 100644
--- a/test/CodeGen/ARM/vcge.ll
+++ b/test/CodeGen/ARM/vcge.ll
@@ -1,13 +1,8 @@
-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep {vcge\\.s8} %t | count 2
-; RUN: grep {vcge\\.s16} %t | count 2
-; RUN: grep {vcge\\.s32} %t | count 2
-; RUN: grep {vcge\\.u8} %t | count 2
-; RUN: grep {vcge\\.u16} %t | count 2
-; RUN: grep {vcge\\.u32} %t | count 2
-; RUN: grep {vcge\\.f32} %t | count 2
+; RUN: llvm-as < %s | llc -march=arm -mattr=+neon | FileCheck %s
 
 define <8 x i8> @vcges8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
+;CHECK: vcges8:
+;CHECK: vcge.s8
 	%tmp1 = load <8 x i8>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = icmp sge <8 x i8> %tmp1, %tmp2
@@ -16,6 +11,8 @@
 }
 
 define <4 x i16> @vcges16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
+;CHECK: vcges16:
+;CHECK: vcge.s16
 	%tmp1 = load <4 x i16>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = icmp sge <4 x i16> %tmp1, %tmp2
@@ -24,6 +21,8 @@
 }
 
 define <2 x i32> @vcges32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
+;CHECK: vcges32:
+;CHECK: vcge.s32
 	%tmp1 = load <2 x i32>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = icmp sge <2 x i32> %tmp1, %tmp2
@@ -32,6 +31,8 @@
 }
 
 define <8 x i8> @vcgeu8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
+;CHECK: vcgeu8:
+;CHECK: vcge.u8
 	%tmp1 = load <8 x i8>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = icmp uge <8 x i8> %tmp1, %tmp2
@@ -40,6 +41,8 @@
 }
 
 define <4 x i16> @vcgeu16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
+;CHECK: vcgeu16:
+;CHECK: vcge.u16
 	%tmp1 = load <4 x i16>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = icmp uge <4 x i16> %tmp1, %tmp2
@@ -48,6 +51,8 @@
 }
 
 define <2 x i32> @vcgeu32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
+;CHECK: vcgeu32:
+;CHECK: vcge.u32
 	%tmp1 = load <2 x i32>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = icmp uge <2 x i32> %tmp1, %tmp2
@@ -56,6 +61,8 @@
 }
 
 define <2 x i32> @vcgef32(<2 x float>* %A, <2 x float>* %B) nounwind {
+;CHECK: vcgef32:
+;CHECK: vcge.f32
 	%tmp1 = load <2 x float>* %A
 	%tmp2 = load <2 x float>* %B
 	%tmp3 = fcmp oge <2 x float> %tmp1, %tmp2
@@ -64,6 +71,8 @@
 }
 
 define <16 x i8> @vcgeQs8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
+;CHECK: vcgeQs8:
+;CHECK: vcge.s8
 	%tmp1 = load <16 x i8>* %A
 	%tmp2 = load <16 x i8>* %B
 	%tmp3 = icmp sge <16 x i8> %tmp1, %tmp2
@@ -72,6 +81,8 @@
 }
 
 define <8 x i16> @vcgeQs16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
+;CHECK: vcgeQs16:
+;CHECK: vcge.s16
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i16>* %B
 	%tmp3 = icmp sge <8 x i16> %tmp1, %tmp2
@@ -80,6 +91,8 @@
 }
 
 define <4 x i32> @vcgeQs32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
+;CHECK: vcgeQs32:
+;CHECK: vcge.s32
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i32>* %B
 	%tmp3 = icmp sge <4 x i32> %tmp1, %tmp2
@@ -88,6 +101,8 @@
 }
 
 define <16 x i8> @vcgeQu8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
+;CHECK: vcgeQu8:
+;CHECK: vcge.u8
 	%tmp1 = load <16 x i8>* %A
 	%tmp2 = load <16 x i8>* %B
 	%tmp3 = icmp uge <16 x i8> %tmp1, %tmp2
@@ -96,6 +111,8 @@
 }
 
 define <8 x i16> @vcgeQu16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
+;CHECK: vcgeQu16:
+;CHECK: vcge.u16
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i16>* %B
 	%tmp3 = icmp uge <8 x i16> %tmp1, %tmp2
@@ -104,6 +121,8 @@
 }
 
 define <4 x i32> @vcgeQu32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
+;CHECK: vcgeQu32:
+;CHECK: vcge.u32
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i32>* %B
 	%tmp3 = icmp uge <4 x i32> %tmp1, %tmp2
@@ -112,6 +131,8 @@
 }
 
 define <4 x i32> @vcgeQf32(<4 x float>* %A, <4 x float>* %B) nounwind {
+;CHECK: vcgeQf32:
+;CHECK: vcge.f32
 	%tmp1 = load <4 x float>* %A
 	%tmp2 = load <4 x float>* %B
 	%tmp3 = fcmp oge <4 x float> %tmp1, %tmp2
diff --git a/test/CodeGen/ARM/vcgt.ll b/test/CodeGen/ARM/vcgt.ll
index f926083..47115ca 100644
--- a/test/CodeGen/ARM/vcgt.ll
+++ b/test/CodeGen/ARM/vcgt.ll
@@ -1,13 +1,8 @@
-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep {vcgt\\.s8} %t | count 2
-; RUN: grep {vcgt\\.s16} %t | count 2
-; RUN: grep {vcgt\\.s32} %t | count 2
-; RUN: grep {vcgt\\.u8} %t | count 2
-; RUN: grep {vcgt\\.u16} %t | count 2
-; RUN: grep {vcgt\\.u32} %t | count 2
-; RUN: grep {vcgt\\.f32} %t | count 2
+; RUN: llvm-as < %s | llc -march=arm -mattr=+neon | FileCheck %s
 
 define <8 x i8> @vcgts8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
+;CHECK: vcgts8:
+;CHECK: vcgt.s8
 	%tmp1 = load <8 x i8>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = icmp sgt <8 x i8> %tmp1, %tmp2
@@ -16,6 +11,8 @@
 }
 
 define <4 x i16> @vcgts16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
+;CHECK: vcgts16:
+;CHECK: vcgt.s16
 	%tmp1 = load <4 x i16>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = icmp sgt <4 x i16> %tmp1, %tmp2
@@ -24,6 +21,8 @@
 }
 
 define <2 x i32> @vcgts32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
+;CHECK: vcgts32:
+;CHECK: vcgt.s32
 	%tmp1 = load <2 x i32>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = icmp sgt <2 x i32> %tmp1, %tmp2
@@ -32,6 +31,8 @@
 }
 
 define <8 x i8> @vcgtu8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
+;CHECK: vcgtu8:
+;CHECK: vcgt.u8
 	%tmp1 = load <8 x i8>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = icmp ugt <8 x i8> %tmp1, %tmp2
@@ -40,6 +41,8 @@
 }
 
 define <4 x i16> @vcgtu16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
+;CHECK: vcgtu16:
+;CHECK: vcgt.u16
 	%tmp1 = load <4 x i16>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = icmp ugt <4 x i16> %tmp1, %tmp2
@@ -48,6 +51,8 @@
 }
 
 define <2 x i32> @vcgtu32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
+;CHECK: vcgtu32:
+;CHECK: vcgt.u32
 	%tmp1 = load <2 x i32>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = icmp ugt <2 x i32> %tmp1, %tmp2
@@ -56,6 +61,8 @@
 }
 
 define <2 x i32> @vcgtf32(<2 x float>* %A, <2 x float>* %B) nounwind {
+;CHECK: vcgtf32:
+;CHECK: vcgt.f32
 	%tmp1 = load <2 x float>* %A
 	%tmp2 = load <2 x float>* %B
 	%tmp3 = fcmp ogt <2 x float> %tmp1, %tmp2
@@ -64,6 +71,8 @@
 }
 
 define <16 x i8> @vcgtQs8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
+;CHECK: vcgtQs8:
+;CHECK: vcgt.s8
 	%tmp1 = load <16 x i8>* %A
 	%tmp2 = load <16 x i8>* %B
 	%tmp3 = icmp sgt <16 x i8> %tmp1, %tmp2
@@ -72,6 +81,8 @@
 }
 
 define <8 x i16> @vcgtQs16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
+;CHECK: vcgtQs16:
+;CHECK: vcgt.s16
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i16>* %B
 	%tmp3 = icmp sgt <8 x i16> %tmp1, %tmp2
@@ -80,6 +91,8 @@
 }
 
 define <4 x i32> @vcgtQs32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
+;CHECK: vcgtQs32:
+;CHECK: vcgt.s32
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i32>* %B
 	%tmp3 = icmp sgt <4 x i32> %tmp1, %tmp2
@@ -88,6 +101,8 @@
 }
 
 define <16 x i8> @vcgtQu8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
+;CHECK: vcgtQu8:
+;CHECK: vcgt.u8
 	%tmp1 = load <16 x i8>* %A
 	%tmp2 = load <16 x i8>* %B
 	%tmp3 = icmp ugt <16 x i8> %tmp1, %tmp2
@@ -96,6 +111,8 @@
 }
 
 define <8 x i16> @vcgtQu16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
+;CHECK: vcgtQu16:
+;CHECK: vcgt.u16
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i16>* %B
 	%tmp3 = icmp ugt <8 x i16> %tmp1, %tmp2
@@ -104,6 +121,8 @@
 }
 
 define <4 x i32> @vcgtQu32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
+;CHECK: vcgtQu32:
+;CHECK: vcgt.u32
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i32>* %B
 	%tmp3 = icmp ugt <4 x i32> %tmp1, %tmp2
@@ -112,6 +131,8 @@
 }
 
 define <4 x i32> @vcgtQf32(<4 x float>* %A, <4 x float>* %B) nounwind {
+;CHECK: vcgtQf32:
+;CHECK: vcgt.f32
 	%tmp1 = load <4 x float>* %A
 	%tmp2 = load <4 x float>* %B
 	%tmp3 = fcmp ogt <4 x float> %tmp1, %tmp2