diff --git a/test/CodeGen/ARM/vacgt.ll b/test/CodeGen/ARM/vacgt.ll
index ebf590e..2a9c7ae 100644
--- a/test/CodeGen/ARM/vacgt.ll
+++ b/test/CodeGen/ARM/vacgt.ll
@@ -1,7 +1,8 @@
-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep {vacgt\\.f32} %t | count 2
+; RUN: llvm-as < %s | llc -march=arm -mattr=+neon | FileCheck %s
 
 define <2 x i32> @vacgtf32(<2 x float>* %A, <2 x float>* %B) nounwind {
+;CHECK: vacgtf32:
+;CHECK: vacgt.f32
 	%tmp1 = load <2 x float>* %A
 	%tmp2 = load <2 x float>* %B
 	%tmp3 = call <2 x i32> @llvm.arm.neon.vacgtd(<2 x float> %tmp1, <2 x float> %tmp2)
@@ -9,6 +10,8 @@
 }
 
 define <4 x i32> @vacgtQf32(<4 x float>* %A, <4 x float>* %B) nounwind {
+;CHECK: vacgtQf32:
+;CHECK: vacgt.f32
 	%tmp1 = load <4 x float>* %A
 	%tmp2 = load <4 x float>* %B
 	%tmp3 = call <4 x i32> @llvm.arm.neon.vacgtq(<4 x float> %tmp1, <4 x float> %tmp2)
diff --git a/test/CodeGen/ARM/vadd.ll b/test/CodeGen/ARM/vadd.ll
index b2b0e23..cebf27c 100644
--- a/test/CodeGen/ARM/vadd.ll
+++ b/test/CodeGen/ARM/vadd.ll
@@ -1,11 +1,8 @@
-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep {vadd\\.i8} %t | count 2
-; RUN: grep {vadd\\.i16} %t | count 2
-; RUN: grep {vadd\\.i32} %t | count 2
-; RUN: grep {vadd\\.i64} %t | count 2
-; RUN: grep {vadd\\.f32} %t | count 2
+; RUN: llvm-as < %s | llc -march=arm -mattr=+neon | FileCheck %s
 
 define <8 x i8> @vaddi8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
+;CHECK: vaddi8:
+;CHECK: vadd.i8
 	%tmp1 = load <8 x i8>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = add <8 x i8> %tmp1, %tmp2
@@ -13,6 +10,8 @@
 }
 
 define <4 x i16> @vaddi16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
+;CHECK: vaddi16:
+;CHECK: vadd.i16
 	%tmp1 = load <4 x i16>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = add <4 x i16> %tmp1, %tmp2
@@ -20,6 +19,8 @@
 }
 
 define <2 x i32> @vaddi32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
+;CHECK: vaddi32:
+;CHECK: vadd.i32
 	%tmp1 = load <2 x i32>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = add <2 x i32> %tmp1, %tmp2
@@ -27,6 +28,8 @@
 }
 
 define <1 x i64> @vaddi64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
+;CHECK: vaddi64:
+;CHECK: vadd.i64
 	%tmp1 = load <1 x i64>* %A
 	%tmp2 = load <1 x i64>* %B
 	%tmp3 = add <1 x i64> %tmp1, %tmp2
@@ -34,6 +37,8 @@
 }
 
 define <2 x float> @vaddf32(<2 x float>* %A, <2 x float>* %B) nounwind {
+;CHECK: vaddf32:
+;CHECK: vadd.f32
 	%tmp1 = load <2 x float>* %A
 	%tmp2 = load <2 x float>* %B
 	%tmp3 = add <2 x float> %tmp1, %tmp2
@@ -41,6 +46,8 @@
 }
 
 define <16 x i8> @vaddQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
+;CHECK: vaddQi8:
+;CHECK: vadd.i8
 	%tmp1 = load <16 x i8>* %A
 	%tmp2 = load <16 x i8>* %B
 	%tmp3 = add <16 x i8> %tmp1, %tmp2
@@ -48,6 +55,8 @@
 }
 
 define <8 x i16> @vaddQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
+;CHECK: vaddQi16:
+;CHECK: vadd.i16
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i16>* %B
 	%tmp3 = add <8 x i16> %tmp1, %tmp2
@@ -55,6 +64,8 @@
 }
 
 define <4 x i32> @vaddQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
+;CHECK: vaddQi32:
+;CHECK: vadd.i32
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i32>* %B
 	%tmp3 = add <4 x i32> %tmp1, %tmp2
@@ -62,6 +73,8 @@
 }
 
 define <2 x i64> @vaddQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
+;CHECK: vaddQi64:
+;CHECK: vadd.i64
 	%tmp1 = load <2 x i64>* %A
 	%tmp2 = load <2 x i64>* %B
 	%tmp3 = add <2 x i64> %tmp1, %tmp2
@@ -69,6 +82,8 @@
 }
 
 define <4 x float> @vaddQf32(<4 x float>* %A, <4 x float>* %B) nounwind {
+;CHECK: vaddQf32:
+;CHECK: vadd.f32
 	%tmp1 = load <4 x float>* %A
 	%tmp2 = load <4 x float>* %B
 	%tmp3 = add <4 x float> %tmp1, %tmp2
diff --git a/test/CodeGen/ARM/vaddhn.ll b/test/CodeGen/ARM/vaddhn.ll
index b2d1b14..ee36646 100644
--- a/test/CodeGen/ARM/vaddhn.ll
+++ b/test/CodeGen/ARM/vaddhn.ll
@@ -1,9 +1,8 @@
-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep {vaddhn\\.i16} %t | count 1
-; RUN: grep {vaddhn\\.i32} %t | count 1
-; RUN: grep {vaddhn\\.i64} %t | count 1
+; RUN: llvm-as < %s | llc -march=arm -mattr=+neon | FileCheck %s
 
 define <8 x i8> @vaddhni16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
+;CHECK: vaddhni16:
+;CHECK: vaddhn.i16
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i16>* %B
 	%tmp3 = call <8 x i8> @llvm.arm.neon.vaddhn.v8i8(<8 x i16> %tmp1, <8 x i16> %tmp2)
@@ -11,6 +10,8 @@
 }
 
 define <4 x i16> @vaddhni32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
+;CHECK: vaddhni32:
+;CHECK: vaddhn.i32
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i32>* %B
 	%tmp3 = call <4 x i16> @llvm.arm.neon.vaddhn.v4i16(<4 x i32> %tmp1, <4 x i32> %tmp2)
@@ -18,6 +19,8 @@
 }
 
 define <2 x i32> @vaddhni64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
+;CHECK: vaddhni64:
+;CHECK: vaddhn.i64
 	%tmp1 = load <2 x i64>* %A
 	%tmp2 = load <2 x i64>* %B
 	%tmp3 = call <2 x i32> @llvm.arm.neon.vaddhn.v2i32(<2 x i64> %tmp1, <2 x i64> %tmp2)
diff --git a/test/CodeGen/ARM/vaddl.ll b/test/CodeGen/ARM/vaddl.ll
index 26ab17b..21c9600 100644
--- a/test/CodeGen/ARM/vaddl.ll
+++ b/test/CodeGen/ARM/vaddl.ll
@@ -1,12 +1,8 @@
-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep {vaddl\\.s8} %t | count 1
-; RUN: grep {vaddl\\.s16} %t | count 1
-; RUN: grep {vaddl\\.s32} %t | count 1
-; RUN: grep {vaddl\\.u8} %t | count 1
-; RUN: grep {vaddl\\.u16} %t | count 1
-; RUN: grep {vaddl\\.u32} %t | count 1
+; RUN: llvm-as < %s | llc -march=arm -mattr=+neon | FileCheck %s
 
 define <8 x i16> @vaddls8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
+;CHECK: vaddls8:
+;CHECK: vaddl.s8
 	%tmp1 = load <8 x i8>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = call <8 x i16> @llvm.arm.neon.vaddls.v8i16(<8 x i8> %tmp1, <8 x i8> %tmp2)
@@ -14,6 +10,8 @@
 }
 
 define <4 x i32> @vaddls16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
+;CHECK: vaddls16:
+;CHECK: vaddl.s16
 	%tmp1 = load <4 x i16>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = call <4 x i32> @llvm.arm.neon.vaddls.v4i32(<4 x i16> %tmp1, <4 x i16> %tmp2)
@@ -21,6 +19,8 @@
 }
 
 define <2 x i64> @vaddls32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
+;CHECK: vaddls32:
+;CHECK: vaddl.s32
 	%tmp1 = load <2 x i32>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = call <2 x i64> @llvm.arm.neon.vaddls.v2i64(<2 x i32> %tmp1, <2 x i32> %tmp2)
@@ -28,6 +28,8 @@
 }
 
 define <8 x i16> @vaddlu8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
+;CHECK: vaddlu8:
+;CHECK: vaddl.u8
 	%tmp1 = load <8 x i8>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = call <8 x i16> @llvm.arm.neon.vaddlu.v8i16(<8 x i8> %tmp1, <8 x i8> %tmp2)
@@ -35,6 +37,8 @@
 }
 
 define <4 x i32> @vaddlu16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
+;CHECK: vaddlu16:
+;CHECK: vaddl.u16
 	%tmp1 = load <4 x i16>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = call <4 x i32> @llvm.arm.neon.vaddlu.v4i32(<4 x i16> %tmp1, <4 x i16> %tmp2)
@@ -42,6 +46,8 @@
 }
 
 define <2 x i64> @vaddlu32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
+;CHECK: vaddlu32:
+;CHECK: vaddl.u32
 	%tmp1 = load <2 x i32>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = call <2 x i64> @llvm.arm.neon.vaddlu.v2i64(<2 x i32> %tmp1, <2 x i32> %tmp2)
diff --git a/test/CodeGen/ARM/vaddw.ll b/test/CodeGen/ARM/vaddw.ll
index e06f94a..3e9a8de 100644
--- a/test/CodeGen/ARM/vaddw.ll
+++ b/test/CodeGen/ARM/vaddw.ll
@@ -1,12 +1,8 @@
-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep {vaddw\\.s8} %t | count 1
-; RUN: grep {vaddw\\.s16} %t | count 1
-; RUN: grep {vaddw\\.s32} %t | count 1
-; RUN: grep {vaddw\\.u8} %t | count 1
-; RUN: grep {vaddw\\.u16} %t | count 1
-; RUN: grep {vaddw\\.u32} %t | count 1
+; RUN: llvm-as < %s | llc -march=arm -mattr=+neon | FileCheck %s
 
 define <8 x i16> @vaddws8(<8 x i16>* %A, <8 x i8>* %B) nounwind {
+;CHECK: vaddws8:
+;CHECK: vaddw.s8
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = call <8 x i16> @llvm.arm.neon.vaddws.v8i16(<8 x i16> %tmp1, <8 x i8> %tmp2)
@@ -14,6 +10,8 @@
 }
 
 define <4 x i32> @vaddws16(<4 x i32>* %A, <4 x i16>* %B) nounwind {
+;CHECK: vaddws16:
+;CHECK: vaddw.s16
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = call <4 x i32> @llvm.arm.neon.vaddws.v4i32(<4 x i32> %tmp1, <4 x i16> %tmp2)
@@ -21,6 +19,8 @@
 }
 
 define <2 x i64> @vaddws32(<2 x i64>* %A, <2 x i32>* %B) nounwind {
+;CHECK: vaddws32:
+;CHECK: vaddw.s32
 	%tmp1 = load <2 x i64>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = call <2 x i64> @llvm.arm.neon.vaddws.v2i64(<2 x i64> %tmp1, <2 x i32> %tmp2)
@@ -28,6 +28,8 @@
 }
 
 define <8 x i16> @vaddwu8(<8 x i16>* %A, <8 x i8>* %B) nounwind {
+;CHECK: vaddwu8:
+;CHECK: vaddw.u8
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = call <8 x i16> @llvm.arm.neon.vaddwu.v8i16(<8 x i16> %tmp1, <8 x i8> %tmp2)
@@ -35,6 +37,8 @@
 }
 
 define <4 x i32> @vaddwu16(<4 x i32>* %A, <4 x i16>* %B) nounwind {
+;CHECK: vaddwu16:
+;CHECK: vaddw.u16
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = call <4 x i32> @llvm.arm.neon.vaddwu.v4i32(<4 x i32> %tmp1, <4 x i16> %tmp2)
@@ -42,6 +46,8 @@
 }
 
 define <2 x i64> @vaddwu32(<2 x i64>* %A, <2 x i32>* %B) nounwind {
+;CHECK: vaddwu32:
+;CHECK: vaddw.u32
 	%tmp1 = load <2 x i64>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = call <2 x i64> @llvm.arm.neon.vaddwu.v2i64(<2 x i64> %tmp1, <2 x i32> %tmp2)
diff --git a/test/CodeGen/ARM/vand.ll b/test/CodeGen/ARM/vand.ll
index b3eaf1c..d61e435 100644
--- a/test/CodeGen/ARM/vand.ll
+++ b/test/CodeGen/ARM/vand.ll
@@ -1,8 +1,8 @@
-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep vand %t | count 8
-; Note: function names do not include "vand" to allow simple grep for opcodes
+; RUN: llvm-as < %s | llc -march=arm -mattr=+neon | FileCheck %s
 
 define <8 x i8> @v_andi8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
+;CHECK: v_andi8:
+;CHECK: vand
 	%tmp1 = load <8 x i8>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = and <8 x i8> %tmp1, %tmp2
@@ -10,6 +10,8 @@
 }
 
 define <4 x i16> @v_andi16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
+;CHECK: v_andi16:
+;CHECK: vand
 	%tmp1 = load <4 x i16>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = and <4 x i16> %tmp1, %tmp2
@@ -17,6 +19,8 @@
 }
 
 define <2 x i32> @v_andi32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
+;CHECK: v_andi32:
+;CHECK: vand
 	%tmp1 = load <2 x i32>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = and <2 x i32> %tmp1, %tmp2
@@ -24,6 +28,8 @@
 }
 
 define <1 x i64> @v_andi64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
+;CHECK: v_andi64:
+;CHECK: vand
 	%tmp1 = load <1 x i64>* %A
 	%tmp2 = load <1 x i64>* %B
 	%tmp3 = and <1 x i64> %tmp1, %tmp2
@@ -31,6 +37,8 @@
 }
 
 define <16 x i8> @v_andQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
+;CHECK: v_andQi8:
+;CHECK: vand
 	%tmp1 = load <16 x i8>* %A
 	%tmp2 = load <16 x i8>* %B
 	%tmp3 = and <16 x i8> %tmp1, %tmp2
@@ -38,6 +46,8 @@
 }
 
 define <8 x i16> @v_andQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
+;CHECK: v_andQi16:
+;CHECK: vand
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i16>* %B
 	%tmp3 = and <8 x i16> %tmp1, %tmp2
@@ -45,6 +55,8 @@
 }
 
 define <4 x i32> @v_andQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
+;CHECK: v_andQi32:
+;CHECK: vand
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i32>* %B
 	%tmp3 = and <4 x i32> %tmp1, %tmp2
@@ -52,6 +64,8 @@
 }
 
 define <2 x i64> @v_andQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
+;CHECK: v_andQi64:
+;CHECK: vand
 	%tmp1 = load <2 x i64>* %A
 	%tmp2 = load <2 x i64>* %B
 	%tmp3 = and <2 x i64> %tmp1, %tmp2
diff --git a/test/CodeGen/ARM/vbic.ll b/test/CodeGen/ARM/vbic.ll
index dbc11ea..591f491 100644
--- a/test/CodeGen/ARM/vbic.ll
+++ b/test/CodeGen/ARM/vbic.ll
@@ -1,8 +1,8 @@
-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep vbic %t | count 8
-; Note: function names do not include "vbic" to allow simple grep for opcodes
+; RUN: llvm-as < %s | llc -march=arm -mattr=+neon | FileCheck %s
 
 define <8 x i8> @v_bici8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
+;CHECK: v_bici8:
+;CHECK: vbic
 	%tmp1 = load <8 x i8>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = xor <8 x i8> %tmp2, < i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1 >
@@ -11,6 +11,8 @@
 }
 
 define <4 x i16> @v_bici16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
+;CHECK: v_bici16:
+;CHECK: vbic
 	%tmp1 = load <4 x i16>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = xor <4 x i16> %tmp2, < i16 -1, i16 -1, i16 -1, i16 -1 >
@@ -19,6 +21,8 @@
 }
 
 define <2 x i32> @v_bici32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
+;CHECK: v_bici32:
+;CHECK: vbic
 	%tmp1 = load <2 x i32>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = xor <2 x i32> %tmp2, < i32 -1, i32 -1 >
@@ -27,6 +31,8 @@
 }
 
 define <1 x i64> @v_bici64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
+;CHECK: v_bici64:
+;CHECK: vbic
 	%tmp1 = load <1 x i64>* %A
 	%tmp2 = load <1 x i64>* %B
 	%tmp3 = xor <1 x i64> %tmp2, < i64 -1 >
@@ -35,6 +41,8 @@
 }
 
 define <16 x i8> @v_bicQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
+;CHECK: v_bicQi8:
+;CHECK: vbic
 	%tmp1 = load <16 x i8>* %A
 	%tmp2 = load <16 x i8>* %B
 	%tmp3 = xor <16 x i8> %tmp2, < i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1 >
@@ -43,6 +51,8 @@
 }
 
 define <8 x i16> @v_bicQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
+;CHECK: v_bicQi16:
+;CHECK: vbic
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i16>* %B
 	%tmp3 = xor <8 x i16> %tmp2, < i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1 >
@@ -51,6 +61,8 @@
 }
 
 define <4 x i32> @v_bicQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
+;CHECK: v_bicQi32:
+;CHECK: vbic
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i32>* %B
 	%tmp3 = xor <4 x i32> %tmp2, < i32 -1, i32 -1, i32 -1, i32 -1 >
@@ -59,6 +71,8 @@
 }
 
 define <2 x i64> @v_bicQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
+;CHECK: v_bicQi64:
+;CHECK: vbic
 	%tmp1 = load <2 x i64>* %A
 	%tmp2 = load <2 x i64>* %B
 	%tmp3 = xor <2 x i64> %tmp2, < i64 -1, i64 -1 >
diff --git a/test/CodeGen/ARM/vbsl.ll b/test/CodeGen/ARM/vbsl.ll
index 37ddf4d..bf81304 100644
--- a/test/CodeGen/ARM/vbsl.ll
+++ b/test/CodeGen/ARM/vbsl.ll
@@ -1,8 +1,8 @@
-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep vbsl %t | count 8
-; Note: function names do not include "vbsl" to allow simple grep for opcodes
+; RUN: llvm-as < %s | llc -march=arm -mattr=+neon | FileCheck %s
 
 define <8 x i8> @v_bsli8(<8 x i8>* %A, <8 x i8>* %B, <8 x i8>* %C) nounwind {
+;CHECK: v_bsli8:
+;CHECK: vbsl
 	%tmp1 = load <8 x i8>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = load <8 x i8>* %C
@@ -14,6 +14,8 @@
 }
 
 define <4 x i16> @v_bsli16(<4 x i16>* %A, <4 x i16>* %B, <4 x i16>* %C) nounwind {
+;CHECK: v_bsli16:
+;CHECK: vbsl
 	%tmp1 = load <4 x i16>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = load <4 x i16>* %C
@@ -25,6 +27,8 @@
 }
 
 define <2 x i32> @v_bsli32(<2 x i32>* %A, <2 x i32>* %B, <2 x i32>* %C) nounwind {
+;CHECK: v_bsli32:
+;CHECK: vbsl
 	%tmp1 = load <2 x i32>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = load <2 x i32>* %C
@@ -36,6 +40,8 @@
 }
 
 define <1 x i64> @v_bsli64(<1 x i64>* %A, <1 x i64>* %B, <1 x i64>* %C) nounwind {
+;CHECK: v_bsli64:
+;CHECK: vbsl
 	%tmp1 = load <1 x i64>* %A
 	%tmp2 = load <1 x i64>* %B
 	%tmp3 = load <1 x i64>* %C
@@ -47,6 +53,8 @@
 }
 
 define <16 x i8> @v_bslQi8(<16 x i8>* %A, <16 x i8>* %B, <16 x i8>* %C) nounwind {
+;CHECK: v_bslQi8:
+;CHECK: vbsl
 	%tmp1 = load <16 x i8>* %A
 	%tmp2 = load <16 x i8>* %B
 	%tmp3 = load <16 x i8>* %C
@@ -58,6 +66,8 @@
 }
 
 define <8 x i16> @v_bslQi16(<8 x i16>* %A, <8 x i16>* %B, <8 x i16>* %C) nounwind {
+;CHECK: v_bslQi16:
+;CHECK: vbsl
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i16>* %B
 	%tmp3 = load <8 x i16>* %C
@@ -69,6 +79,8 @@
 }
 
 define <4 x i32> @v_bslQi32(<4 x i32>* %A, <4 x i32>* %B, <4 x i32>* %C) nounwind {
+;CHECK: v_bslQi32:
+;CHECK: vbsl
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i32>* %B
 	%tmp3 = load <4 x i32>* %C
@@ -80,6 +92,8 @@
 }
 
 define <2 x i64> @v_bslQi64(<2 x i64>* %A, <2 x i64>* %B, <2 x i64>* %C) nounwind {
+;CHECK: v_bslQi64:
+;CHECK: vbsl
 	%tmp1 = load <2 x i64>* %A
 	%tmp2 = load <2 x i64>* %B
 	%tmp3 = load <2 x i64>* %C
diff --git a/test/CodeGen/ARM/vceq.ll b/test/CodeGen/ARM/vceq.ll
index 1af67e8..6e545b7 100644
--- a/test/CodeGen/ARM/vceq.ll
+++ b/test/CodeGen/ARM/vceq.ll
@@ -1,10 +1,8 @@
-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep {vceq\\.i8} %t | count 2
-; RUN: grep {vceq\\.i16} %t | count 2
-; RUN: grep {vceq\\.i32} %t | count 2
-; RUN: grep {vceq\\.f32} %t | count 2
+; RUN: llvm-as < %s | llc -march=arm -mattr=+neon | FileCheck %s
 
 define <8 x i8> @vceqi8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
+;CHECK: vceqi8:
+;CHECK: vceq.i8
 	%tmp1 = load <8 x i8>* %A
 	%tmp2 = load <8 x i8>* %B
 	%tmp3 = icmp eq <8 x i8> %tmp1, %tmp2
@@ -13,6 +11,8 @@
 }
 
 define <4 x i16> @vceqi16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
+;CHECK: vceqi16:
+;CHECK: vceq.i16
 	%tmp1 = load <4 x i16>* %A
 	%tmp2 = load <4 x i16>* %B
 	%tmp3 = icmp eq <4 x i16> %tmp1, %tmp2
@@ -21,6 +21,8 @@
 }
 
 define <2 x i32> @vceqi32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
+;CHECK: vceqi32:
+;CHECK: vceq.i32
 	%tmp1 = load <2 x i32>* %A
 	%tmp2 = load <2 x i32>* %B
 	%tmp3 = icmp eq <2 x i32> %tmp1, %tmp2
@@ -29,6 +31,8 @@
 }
 
 define <2 x i32> @vceqf32(<2 x float>* %A, <2 x float>* %B) nounwind {
+;CHECK: vceqf32:
+;CHECK: vceq.f32
 	%tmp1 = load <2 x float>* %A
 	%tmp2 = load <2 x float>* %B
 	%tmp3 = fcmp oeq <2 x float> %tmp1, %tmp2
@@ -37,6 +41,8 @@
 }
 
 define <16 x i8> @vceqQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
+;CHECK: vceqQi8:
+;CHECK: vceq.i8
 	%tmp1 = load <16 x i8>* %A
 	%tmp2 = load <16 x i8>* %B
 	%tmp3 = icmp eq <16 x i8> %tmp1, %tmp2
@@ -45,6 +51,8 @@
 }
 
 define <8 x i16> @vceqQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
+;CHECK: vceqQi16:
+;CHECK: vceq.i16
 	%tmp1 = load <8 x i16>* %A
 	%tmp2 = load <8 x i16>* %B
 	%tmp3 = icmp eq <8 x i16> %tmp1, %tmp2
@@ -53,6 +61,8 @@
 }
 
 define <4 x i32> @vceqQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
+;CHECK: vceqQi32:
+;CHECK: vceq.i32
 	%tmp1 = load <4 x i32>* %A
 	%tmp2 = load <4 x i32>* %B
 	%tmp3 = icmp eq <4 x i32> %tmp1, %tmp2
@@ -61,6 +71,8 @@
 }
 
 define <4 x i32> @vceqQf32(<4 x float>* %A, <4 x float>* %B) nounwind {
+;CHECK: vceqQf32:
+;CHECK: vceq.f32
 	%tmp1 = load <4 x float>* %A
 	%tmp2 = load <4 x float>* %B
 	%tmp3 = fcmp oeq <4 x float> %tmp1, %tmp2