From: Sanjay Patel <spatel@rotateright.com>
Date: Mon, 3 Jul 2017 15:27:19 +0000 (+0000)
Subject: [x86] auto-generate complete checks for tests; NFC
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=3e805b83e39b81a215c3a2d7f1f21483d49a7463;p=llvm

[x86] auto-generate complete checks for tests; NFC

These all used 'CHECK-NOT' which isn't necessary if we have complete checks.
There were also over-specifications in the RUN params such as CPU model.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@307033 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/test/CodeGen/X86/bool-simplify.ll b/test/CodeGen/X86/bool-simplify.ll
index a0a1c364662..7f7f9791d90 100644
--- a/test/CodeGen/X86/bool-simplify.ll
+++ b/test/CodeGen/X86/bool-simplify.ll
@@ -1,45 +1,62 @@
-; RUN: llc < %s -march=x86-64 -mattr=+sse4.1,-avx,+rdrnd,+rdseed | FileCheck %s
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1,-avx,+rdrnd,+rdseed | FileCheck %s
 
 define i32 @foo(<2 x i64> %c, i32 %a, i32 %b) {
+; CHECK-LABEL: foo:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    ptest %xmm0, %xmm0
+; CHECK-NEXT:    cmovnel %esi, %edi
+; CHECK-NEXT:    movl %edi, %eax
+; CHECK-NEXT:    retq
   %t1 = call i32 @llvm.x86.sse41.ptestz(<2 x i64> %c, <2 x i64> %c)
   %t2 = icmp ne i32 %t1, 0
   %t3 = select i1 %t2, i32 %a, i32 %b
   ret i32 %t3
-; CHECK: foo
-; CHECK: ptest
-; CHECK-NOT: testl
-; CHECK: cmov
-; CHECK: ret
 }
 
 define i32 @bar(<2 x i64> %c) {
+; CHECK-LABEL: bar:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    ptest %xmm0, %xmm0
+; CHECK-NEXT:    jne .LBB1_2
+; CHECK-NEXT:  # BB#1: # %if-true-block
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    retq
+; CHECK-NEXT:  .LBB1_2: # %endif-block
+; CHECK-NEXT:    movl $1, %eax
+; CHECK-NEXT:    retq
 entry:
   %0 = call i32 @llvm.x86.sse41.ptestz(<2 x i64> %c, <2 x i64> %c)
   %1 = icmp ne i32 %0, 0
   br i1 %1, label %if-true-block, label %endif-block
-if-true-block:                                    ; preds = %entry
+if-true-block:
   ret i32 0
-endif-block:                                      ; preds = %entry,
+endif-block:
   ret i32 1
-; CHECK: bar
-; CHECK: ptest
-; CHECK-NOT: testl
-; CHECK: jne
-; CHECK: ret
 }
 
 define i32 @bax(<2 x i64> %c) {
+; CHECK-LABEL: bax:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    ptest %xmm0, %xmm0
+; CHECK-NEXT:    sete %al
+; CHECK-NEXT:    retq
   %t1 = call i32 @llvm.x86.sse41.ptestz(<2 x i64> %c, <2 x i64> %c)
   %t2 = icmp eq i32 %t1, 1
   %t3 = zext i1 %t2 to i32
   ret i32 %t3
-; CHECK: bax
-; CHECK: ptest
-; CHECK-NOT: cmpl
-; CHECK: ret
 }
 
-define i16 @rnd16(i16 %arg) nounwind uwtable {
+define i16 @rnd16(i16 %arg) nounwind {
+; CHECK-LABEL: rnd16:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    rdrandw %cx
+; CHECK-NEXT:    cmovbw %di, %ax
+; CHECK-NEXT:    addl %ecx, %eax
+; CHECK-NEXT:    # kill: %AX<def> %AX<kill> %EAX<kill>
+; CHECK-NEXT:    retq
   %1 = tail call { i16, i32 } @llvm.x86.rdrand.16() nounwind
   %2 = extractvalue { i16, i32 } %1, 0
   %3 = extractvalue { i16, i32 } %1, 1
@@ -47,14 +64,16 @@ define i16 @rnd16(i16 %arg) nounwind uwtable {
   %5 = select i1 %4, i16 0, i16 %arg
   %6 = add i16 %5, %2
   ret i16 %6
-; CHECK: rnd16
-; CHECK: rdrand
-; CHECK: cmov
-; CHECK-NOT: cmov
-; CHECK: ret
 }
 
-define i32 @rnd32(i32 %arg) nounwind uwtable {
+define i32 @rnd32(i32 %arg) nounwind {
+; CHECK-LABEL: rnd32:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    rdrandl %ecx
+; CHECK-NEXT:    cmovbl %edi, %eax
+; CHECK-NEXT:    addl %ecx, %eax
+; CHECK-NEXT:    retq
   %1 = tail call { i32, i32 } @llvm.x86.rdrand.32() nounwind
   %2 = extractvalue { i32, i32 } %1, 0
   %3 = extractvalue { i32, i32 } %1, 1
@@ -62,14 +81,16 @@ define i32 @rnd32(i32 %arg) nounwind uwtable {
   %5 = select i1 %4, i32 0, i32 %arg
   %6 = add i32 %5, %2
   ret i32 %6
-; CHECK: rnd32
-; CHECK: rdrand
-; CHECK: cmov
-; CHECK-NOT: cmov
-; CHECK: ret
 }
 
-define i64 @rnd64(i64 %arg) nounwind uwtable {
+define i64 @rnd64(i64 %arg) nounwind {
+; CHECK-LABEL: rnd64:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    rdrandq %rcx
+; CHECK-NEXT:    cmovbq %rdi, %rax
+; CHECK-NEXT:    addq %rcx, %rax
+; CHECK-NEXT:    retq
   %1 = tail call { i64, i32 } @llvm.x86.rdrand.64() nounwind
   %2 = extractvalue { i64, i32 } %1, 0
   %3 = extractvalue { i64, i32 } %1, 1
@@ -77,14 +98,17 @@ define i64 @rnd64(i64 %arg) nounwind uwtable {
   %5 = select i1 %4, i64 0, i64 %arg
   %6 = add i64 %5, %2
   ret i64 %6
-; CHECK: rnd64
-; CHECK: rdrand
-; CHECK: cmov
-; CHECK-NOT: cmov
-; CHECK: ret
 }
 
-define i16 @seed16(i16 %arg) nounwind uwtable {
+define i16 @seed16(i16 %arg) nounwind {
+; CHECK-LABEL: seed16:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    rdseedw %cx
+; CHECK-NEXT:    cmovbw %di, %ax
+; CHECK-NEXT:    addl %ecx, %eax
+; CHECK-NEXT:    # kill: %AX<def> %AX<kill> %EAX<kill>
+; CHECK-NEXT:    retq
   %1 = tail call { i16, i32 } @llvm.x86.rdseed.16() nounwind
   %2 = extractvalue { i16, i32 } %1, 0
   %3 = extractvalue { i16, i32 } %1, 1
@@ -92,14 +116,16 @@ define i16 @seed16(i16 %arg) nounwind uwtable {
   %5 = select i1 %4, i16 0, i16 %arg
   %6 = add i16 %5, %2
   ret i16 %6
-; CHECK: seed16
-; CHECK: rdseed
-; CHECK: cmov
-; CHECK-NOT: cmov
-; CHECK: ret
 }
 
-define i32 @seed32(i32 %arg) nounwind uwtable {
+define i32 @seed32(i32 %arg) nounwind {
+; CHECK-LABEL: seed32:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    rdseedl %ecx
+; CHECK-NEXT:    cmovbl %edi, %eax
+; CHECK-NEXT:    addl %ecx, %eax
+; CHECK-NEXT:    retq
   %1 = tail call { i32, i32 } @llvm.x86.rdseed.32() nounwind
   %2 = extractvalue { i32, i32 } %1, 0
   %3 = extractvalue { i32, i32 } %1, 1
@@ -107,14 +133,16 @@ define i32 @seed32(i32 %arg) nounwind uwtable {
   %5 = select i1 %4, i32 0, i32 %arg
   %6 = add i32 %5, %2
   ret i32 %6
-; CHECK: seed32
-; CHECK: rdseed
-; CHECK: cmov
-; CHECK-NOT: cmov
-; CHECK: ret
 }
 
-define i64 @seed64(i64 %arg) nounwind uwtable {
+define i64 @seed64(i64 %arg) nounwind {
+; CHECK-LABEL: seed64:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    rdseedq %rcx
+; CHECK-NEXT:    cmovbq %rdi, %rax
+; CHECK-NEXT:    addq %rcx, %rax
+; CHECK-NEXT:    retq
   %1 = tail call { i64, i32 } @llvm.x86.rdseed.64() nounwind
   %2 = extractvalue { i64, i32 } %1, 0
   %3 = extractvalue { i64, i32 } %1, 1
@@ -122,11 +150,6 @@ define i64 @seed64(i64 %arg) nounwind uwtable {
   %5 = select i1 %4, i64 0, i64 %arg
   %6 = add i64 %5, %2
   ret i64 %6
-; CHECK: seed64
-; CHECK: rdseed
-; CHECK: cmov
-; CHECK-NOT: cmov
-; CHECK: ret
 }
 
 declare i32 @llvm.x86.sse41.ptestz(<2 x i64>, <2 x i64>) nounwind readnone
diff --git a/test/CodeGen/X86/combine-avx-intrinsics.ll b/test/CodeGen/X86/combine-avx-intrinsics.ll
index 64e081523c1..811b1f20833 100644
--- a/test/CodeGen/X86/combine-avx-intrinsics.ll
+++ b/test/CodeGen/X86/combine-avx-intrinsics.ll
@@ -1,59 +1,56 @@
-; RUN: llc < %s -march=x86-64 -mcpu=corei7-avx | FileCheck %s
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx | FileCheck %s
 
 
 define <4 x double> @test_x86_avx_blend_pd_256(<4 x double> %a0) {
+; CHECK-LABEL: test_x86_avx_blend_pd_256:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %1 = call <4 x double> @llvm.x86.avx.blend.pd.256(<4 x double> %a0, <4 x double> %a0, i32 7)
   ret <4 x double> %1
 }
-; CHECK-LABEL: test_x86_avx_blend_pd_256
-; CHECK-NOT: vblendpd
-; CHECK: ret
-
 
 define <8 x float> @test_x86_avx_blend_ps_256(<8 x float> %a0) {
+; CHECK-LABEL: test_x86_avx_blend_ps_256:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %1 = call <8 x float> @llvm.x86.avx.blend.ps.256(<8 x float> %a0, <8 x float> %a0, i32 7)
   ret <8 x float> %1
 }
-; CHECK-LABEL: test_x86_avx_blend_ps_256
-; CHECK-NOT: vblendps
-; CHECK: ret
-
 
 define <4 x double> @test2_x86_avx_blend_pd_256(<4 x double> %a0, <4 x double> %a1) {
+; CHECK-LABEL: test2_x86_avx_blend_pd_256:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %1 = call <4 x double> @llvm.x86.avx.blend.pd.256(<4 x double> %a0, <4 x double> %a1, i32 0)
   ret <4 x double> %1
 }
-; CHECK-LABEL: test2_x86_avx_blend_pd_256
-; CHECK-NOT: vblendpd
-; CHECK: ret
-
 
 define <8 x float> @test2_x86_avx_blend_ps_256(<8 x float> %a0, <8 x float> %a1) {
+; CHECK-LABEL: test2_x86_avx_blend_ps_256:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %1 = call <8 x float> @llvm.x86.avx.blend.ps.256(<8 x float> %a0, <8 x float> %a1, i32 0)
   ret <8 x float> %1
 }
-; CHECK-LABEL: test2_x86_avx_blend_ps_256
-; CHECK-NOT: vblendps
-; CHECK: ret
-
 
 define <4 x double> @test3_x86_avx_blend_pd_256(<4 x double> %a0, <4 x double> %a1) {
+; CHECK-LABEL: test3_x86_avx_blend_pd_256:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    vmovaps %ymm1, %ymm0
+; CHECK-NEXT:    retq
   %1 = call <4 x double> @llvm.x86.avx.blend.pd.256(<4 x double> %a0, <4 x double> %a1, i32 -1)
   ret <4 x double> %1
 }
-; CHECK-LABEL: test3_x86_avx_blend_pd_256
-; CHECK-NOT: vblendpd
-; CHECK: ret
-
 
 define <8 x float> @test3_x86_avx_blend_ps_256(<8 x float> %a0, <8 x float> %a1) {
+; CHECK-LABEL: test3_x86_avx_blend_ps_256:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    vmovaps %ymm1, %ymm0
+; CHECK-NEXT:    retq
   %1 = call <8 x float> @llvm.x86.avx.blend.ps.256(<8 x float> %a0, <8 x float> %a1, i32 -1)
   ret <8 x float> %1
 }
-; CHECK-LABEL: test3_x86_avx_blend_ps_256
-; CHECK-NOT: vblendps
-; CHECK: ret
-
 
 declare <4 x double> @llvm.x86.avx.blend.pd.256(<4 x double>, <4 x double>, i32)
 declare <8 x float> @llvm.x86.avx.blend.ps.256(<8 x float>, <8 x float>, i32)
diff --git a/test/CodeGen/X86/combine-avx2-intrinsics.ll b/test/CodeGen/X86/combine-avx2-intrinsics.ll
index 2714b26c914..9a548f6b7f0 100644
--- a/test/CodeGen/X86/combine-avx2-intrinsics.ll
+++ b/test/CodeGen/X86/combine-avx2-intrinsics.ll
@@ -1,88 +1,83 @@
-; RUN: llc < %s -march=x86-64 -mcpu=core-avx2 | FileCheck %s
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx2 | FileCheck %s
 
 ; Verify that the backend correctly combines AVX2 builtin intrinsics.
 
 
 define <16 x i16> @test_x86_avx2_pblendw(<16 x i16> %a0) {
+; CHECK-LABEL: test_x86_avx2_pblendw:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %res = call <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16> %a0, <16 x i16> %a0, i32 7)
   ret <16 x i16> %res
 }
-; CHECK-LABEL: test_x86_avx2_pblendw
-; CHECK-NOT: vpblendw
-; CHECK: ret
-
 
 define <4 x i32> @test_x86_avx2_pblendd_128(<4 x i32> %a0) {
+; CHECK-LABEL: test_x86_avx2_pblendd_128:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %res = call <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32> %a0, <4 x i32> %a0, i32 7)
   ret <4 x i32> %res
 }
-; CHECK-LABEL: test_x86_avx2_pblendd_128
-; CHECK-NOT: vpblendd
-; CHECK: ret
-
 
 define <8 x i32> @test_x86_avx2_pblendd_256(<8 x i32> %a0) {
+; CHECK-LABEL: test_x86_avx2_pblendd_256:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %res = call <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32> %a0, <8 x i32> %a0, i32 7)
   ret <8 x i32> %res
 }
-; CHECK-LABEL: test_x86_avx2_pblendd_256
-; CHECK-NOT: vpblendd
-; CHECK: ret
-
 
 define <16 x i16> @test2_x86_avx2_pblendw(<16 x i16> %a0, <16 x i16> %a1) {
+; CHECK-LABEL: test2_x86_avx2_pblendw:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %res = call <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16> %a0, <16 x i16> %a1, i32 0)
   ret <16 x i16> %res
 }
-; CHECK-LABEL: test2_x86_avx2_pblendw
-; CHECK-NOT: vpblendw
-; CHECK: ret
-
 
 define <4 x i32> @test2_x86_avx2_pblendd_128(<4 x i32> %a0, <4 x i32> %a1) {
+; CHECK-LABEL: test2_x86_avx2_pblendd_128:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %res = call <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32> %a0, <4 x i32> %a1, i32 0)
   ret <4 x i32> %res
 }
-; CHECK-LABEL: test2_x86_avx2_pblendd_128
-; CHECK-NOT: vpblendd
-; CHECK: ret
-
 
 define <8 x i32> @test2_x86_avx2_pblendd_256(<8 x i32> %a0, <8 x i32> %a1) {
+; CHECK-LABEL: test2_x86_avx2_pblendd_256:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %res = call <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32> %a0, <8 x i32> %a1, i32 0)
   ret <8 x i32> %res
 }
-; CHECK-LABEL: test2_x86_avx2_pblendd_256
-; CHECK-NOT: vpblendd
-; CHECK: ret
-
 
 define <16 x i16> @test3_x86_avx2_pblendw(<16 x i16> %a0, <16 x i16> %a1) {
+; CHECK-LABEL: test3_x86_avx2_pblendw:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    vmovaps %ymm1, %ymm0
+; CHECK-NEXT:    retq
   %res = call <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16> %a0, <16 x i16> %a1, i32 -1)
   ret <16 x i16> %res
 }
-; CHECK-LABEL: test3_x86_avx2_pblendw
-; CHECK-NOT: vpblendw
-; CHECK: ret
-
 
 define <4 x i32> @test3_x86_avx2_pblendd_128(<4 x i32> %a0, <4 x i32> %a1) {
+; CHECK-LABEL: test3_x86_avx2_pblendd_128:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    vmovaps %xmm1, %xmm0
+; CHECK-NEXT:    retq
   %res = call <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32> %a0, <4 x i32> %a1, i32 -1)
   ret <4 x i32> %res
 }
-; CHECK-LABEL: test3_x86_avx2_pblendd_128
-; CHECK-NOT: vpblendd
-; CHECK: ret
-
 
 define <8 x i32> @test3_x86_avx2_pblendd_256(<8 x i32> %a0, <8 x i32> %a1) {
+; CHECK-LABEL: test3_x86_avx2_pblendd_256:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    vmovaps %ymm1, %ymm0
+; CHECK-NEXT:    retq
   %res = call <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32> %a0, <8 x i32> %a1, i32 -1)
   ret <8 x i32> %res
 }
-; CHECK-LABEL: test3_x86_avx2_pblendd_256
-; CHECK-NOT: vpblendd
-; CHECK: ret
-
 
 declare <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16>, <16 x i16>, i32)
 declare <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32>, <4 x i32>, i32)
diff --git a/test/CodeGen/X86/combine-sse41-intrinsics.ll b/test/CodeGen/X86/combine-sse41-intrinsics.ll
index 1916883c201..0c8e7b317ec 100644
--- a/test/CodeGen/X86/combine-sse41-intrinsics.ll
+++ b/test/CodeGen/X86/combine-sse41-intrinsics.ll
@@ -1,89 +1,81 @@
-; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=corei7 | FileCheck %s
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=sse4.1 | FileCheck %s
 
 
 define <2 x double> @test_x86_sse41_blend_pd(<2 x double> %a0, <2 x double> %a1) {
+; CHECK-LABEL: test_x86_sse41_blend_pd:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %1 = call <2 x double> @llvm.x86.sse41.blendpd(<2 x double> %a0, <2 x double> %a1, i32 0)
   ret <2 x double> %1
 }
-; CHECK-LABEL: test_x86_sse41_blend_pd
-; CHECK-NOT: blendpd
-; CHECK: ret
-
 
 define <4 x float> @test_x86_sse41_blend_ps(<4 x float> %a0, <4 x float> %a1) {
+; CHECK-LABEL: test_x86_sse41_blend_ps:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %1 = call <4 x float> @llvm.x86.sse41.blendps(<4 x float> %a0, <4 x float> %a1, i32 0)
   ret <4 x float> %1
 }
-; CHECK-LABEL: test_x86_sse41_blend_ps
-; CHECK-NOT: blendps
-; CHECK: ret
-
 
 define <8 x i16> @test_x86_sse41_pblend_w(<8 x i16> %a0, <8 x i16> %a1) {
+; CHECK-LABEL: test_x86_sse41_pblend_w:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %1 = call <8 x i16> @llvm.x86.sse41.pblendw(<8 x i16> %a0, <8 x i16> %a1, i32 0)
   ret <8 x i16> %1
 }
-; CHECK-LABEL: test_x86_sse41_pblend_w
-; CHECK-NOT: pblendw
-; CHECK: ret
-
 
 define <2 x double> @test2_x86_sse41_blend_pd(<2 x double> %a0, <2 x double> %a1) {
+; CHECK-LABEL: test2_x86_sse41_blend_pd:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    movaps %xmm1, %xmm0
+; CHECK-NEXT:    retq
   %1 = call <2 x double> @llvm.x86.sse41.blendpd(<2 x double> %a0, <2 x double> %a1, i32 -1)
   ret <2 x double> %1
 }
-; CHECK-LABEL: test2_x86_sse41_blend_pd
-; CHECK-NOT: blendpd
-; CHECK: movaps %xmm1, %xmm0
-; CHECK-NEXT: ret
-
 
 define <4 x float> @test2_x86_sse41_blend_ps(<4 x float> %a0, <4 x float> %a1) {
+; CHECK-LABEL: test2_x86_sse41_blend_ps:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    movaps %xmm1, %xmm0
+; CHECK-NEXT:    retq
   %1 = call <4 x float> @llvm.x86.sse41.blendps(<4 x float> %a0, <4 x float> %a1, i32 -1)
   ret <4 x float> %1
 }
-; CHECK-LABEL: test2_x86_sse41_blend_ps
-; CHECK-NOT: blendps
-; CHECK: movaps %xmm1, %xmm0
-; CHECK-NEXT: ret
-
 
 define <8 x i16> @test2_x86_sse41_pblend_w(<8 x i16> %a0, <8 x i16> %a1) {
+; CHECK-LABEL: test2_x86_sse41_pblend_w:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    movaps %xmm1, %xmm0
+; CHECK-NEXT:    retq
   %1 = call <8 x i16> @llvm.x86.sse41.pblendw(<8 x i16> %a0, <8 x i16> %a1, i32 -1)
   ret <8 x i16> %1
 }
-; CHECK-LABEL: test2_x86_sse41_pblend_w
-; CHECK-NOT: pblendw
-; CHECK: movaps %xmm1, %xmm0
-; CHECK-NEXT: ret
-
 
 define <2 x double> @test3_x86_sse41_blend_pd(<2 x double> %a0) {
+; CHECK-LABEL: test3_x86_sse41_blend_pd:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %1 = call <2 x double> @llvm.x86.sse41.blendpd(<2 x double> %a0, <2 x double> %a0, i32 7)
   ret <2 x double> %1
 }
-; CHECK-LABEL: test3_x86_sse41_blend_pd
-; CHECK-NOT: blendpd
-; CHECK: ret
-
 
 define <4 x float> @test3_x86_sse41_blend_ps(<4 x float> %a0) {
+; CHECK-LABEL: test3_x86_sse41_blend_ps:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %1 = call <4 x float> @llvm.x86.sse41.blendps(<4 x float> %a0, <4 x float> %a0, i32 7)
   ret <4 x float> %1
 }
-; CHECK-LABEL: test3_x86_sse41_blend_ps
-; CHECK-NOT: blendps
-; CHECK: ret
-
 
 define <8 x i16> @test3_x86_sse41_pblend_w(<8 x i16> %a0) {
+; CHECK-LABEL: test3_x86_sse41_pblend_w:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    retq
   %1 = call <8 x i16> @llvm.x86.sse41.pblendw(<8 x i16> %a0, <8 x i16> %a0, i32 7)
   ret <8 x i16> %1
 }
-; CHECK-LABEL: test3_x86_sse41_pblend_w
-; CHECK-NOT: pblendw
-; CHECK: ret
-
 
 declare <2 x double> @llvm.x86.sse41.blendpd(<2 x double>, <2 x double>, i32)
 declare <4 x float> @llvm.x86.sse41.blendps(<4 x float>, <4 x float>, i32)