[BBVectorize][X86] Regenerate simple tests

llvm-svn: 306578

4 files changed, 240 insertions, 152 deletions

diff --git a/llvm/test/Transforms/BBVectorize/X86/simple-int.ll b/llvm/test/Transforms/BBVectorize/X86/simple-int.ll
index 7842ec85b6c..ee5b5b3e4d0 100644
--- a/llvm/test/Transforms/BBVectorize/X86/simple-int.ll
+++ b/llvm/test/Transforms/BBVectorize/X86/simple-int.ll
@@ -1,6 +1,8 @@
-target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128"
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s
 
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128"
+
 declare double @llvm.fma.f64(double, double, double)
 declare double @llvm.fmuladd.f64(double, double, double)
 declare double @llvm.cos.f64(double)
@@ -8,72 +10,118 @@ declare double @llvm.powi.f64(double, i32)
 
 ; Basic depth-3 chain with fma
 define double @test1(double %A1, double %A2, double %B1, double %B2, double %C1, double %C2) {
-	%X1 = fsub double %A1, %B1
-	%X2 = fsub double %A2, %B2
-	%Y1 = call double @llvm.fma.f64(double %X1, double %A1, double %C1)
-	%Y2 = call double @llvm.fma.f64(double %X2, double %A2, double %C2)
-	%Z1 = fadd double %Y1, %B1
-	%Z2 = fadd double %Y2, %B2
-	%R  = fmul double %Z1, %Z2
-	ret double %R
 ; CHECK-LABEL: @test1(
-; CHECK: ret double %R
+; CHECK-NEXT:    [[X1:%.*]] = fsub double [[A1:%.*]], [[B1:%.*]]
+; CHECK-NEXT:    [[X2:%.*]] = fsub double [[A2:%.*]], [[B2:%.*]]
+; CHECK-NEXT:    [[Y1:%.*]] = call double @llvm.fma.f64(double [[X1]], double [[A1]], double [[C1:%.*]])
+; CHECK-NEXT:    [[Y2:%.*]] = call double @llvm.fma.f64(double [[X2]], double [[A2]], double [[C2:%.*]])
+; CHECK-NEXT:    [[Z1:%.*]] = fadd double [[Y1]], [[B1]]
+; CHECK-NEXT:    [[Z2:%.*]] = fadd double [[Y2]], [[B2]]
+; CHECK-NEXT:    [[R:%.*]] = fmul double [[Z1]], [[Z2]]
+; CHECK-NEXT:    ret double [[R]]
+;
+  %X1 = fsub double %A1, %B1
+  %X2 = fsub double %A2, %B2
+  %Y1 = call double @llvm.fma.f64(double %X1, double %A1, double %C1)
+  %Y2 = call double @llvm.fma.f64(double %X2, double %A2, double %C2)
+  %Z1 = fadd double %Y1, %B1
+  %Z2 = fadd double %Y2, %B2
+  %R  = fmul double %Z1, %Z2
+  ret double %R
 }
 
 ; Basic depth-3 chain with fmuladd
 define double @test1a(double %A1, double %A2, double %B1, double %B2, double %C1, double %C2) {
-	%X1 = fsub double %A1, %B1
-	%X2 = fsub double %A2, %B2
-	%Y1 = call double @llvm.fmuladd.f64(double %X1, double %A1, double %C1)
-	%Y2 = call double @llvm.fmuladd.f64(double %X2, double %A2, double %C2)
-	%Z1 = fadd double %Y1, %B1
-	%Z2 = fadd double %Y2, %B2
-	%R  = fmul double %Z1, %Z2
-	ret double %R
 ; CHECK-LABEL: @test1a(
-; CHECK: ret double %R
+; CHECK-NEXT:    [[X1_V_I1_1:%.*]] = insertelement <2 x double> undef, double [[B1:%.*]], i32 0
+; CHECK-NEXT:    [[X1_V_I1_2:%.*]] = insertelement <2 x double> [[X1_V_I1_1]], double [[B2:%.*]], i32 1
+; CHECK-NEXT:    [[X1_V_I0_1:%.*]] = insertelement <2 x double> undef, double [[A1:%.*]], i32 0
+; CHECK-NEXT:    [[X1_V_I0_2:%.*]] = insertelement <2 x double> [[X1_V_I0_1]], double [[A2:%.*]], i32 1
+; CHECK-NEXT:    [[X1:%.*]] = fsub <2 x double> [[X1_V_I0_2]], [[X1_V_I1_2]]
+; CHECK-NEXT:    [[Y1_V_I2_1:%.*]] = insertelement <2 x double> undef, double [[C1:%.*]], i32 0
+; CHECK-NEXT:    [[Y1_V_I2_2:%.*]] = insertelement <2 x double> [[Y1_V_I2_1]], double [[C2:%.*]], i32 1
+; CHECK-NEXT:    [[Y1:%.*]] = call <2 x double> @llvm.fmuladd.v2f64(<2 x double> [[X1]], <2 x double> [[X1_V_I0_2]], <2 x double> [[Y1_V_I2_2]])
+; CHECK-NEXT:    [[Z1:%.*]] = fadd <2 x double> [[Y1]], [[X1_V_I1_2]]
+; CHECK-NEXT:    [[Z1_V_R1:%.*]] = extractelement <2 x double> [[Z1]], i32 0
+; CHECK-NEXT:    [[Z1_V_R2:%.*]] = extractelement <2 x double> [[Z1]], i32 1
+; CHECK-NEXT:    [[R:%.*]] = fmul double [[Z1_V_R1]], [[Z1_V_R2]]
+; CHECK-NEXT:    ret double [[R]]
+;
+  %X1 = fsub double %A1, %B1
+  %X2 = fsub double %A2, %B2
+  %Y1 = call double @llvm.fmuladd.f64(double %X1, double %A1, double %C1)
+  %Y2 = call double @llvm.fmuladd.f64(double %X2, double %A2, double %C2)
+  %Z1 = fadd double %Y1, %B1
+  %Z2 = fadd double %Y2, %B2
+  %R  = fmul double %Z1, %Z2
+  ret double %R
 }
 
 ; Basic depth-3 chain with cos
 define double @test2(double %A1, double %A2, double %B1, double %B2) {
-	%X1 = fsub double %A1, %B1
-	%X2 = fsub double %A2, %B2
-	%Y1 = call double @llvm.cos.f64(double %X1)
-	%Y2 = call double @llvm.cos.f64(double %X2)
-	%Z1 = fadd double %Y1, %B1
-	%Z2 = fadd double %Y2, %B2
-	%R  = fmul double %Z1, %Z2
-	ret double %R
 ; CHECK-LABEL: @test2(
-; CHECK: ret double %R
+; CHECK-NEXT:    [[X1:%.*]] = fsub double [[A1:%.*]], [[B1:%.*]]
+; CHECK-NEXT:    [[X2:%.*]] = fsub double [[A2:%.*]], [[B2:%.*]]
+; CHECK-NEXT:    [[Y1:%.*]] = call double @llvm.cos.f64(double [[X1]])
+; CHECK-NEXT:    [[Y2:%.*]] = call double @llvm.cos.f64(double [[X2]])
+; CHECK-NEXT:    [[Z1:%.*]] = fadd double [[Y1]], [[B1]]
+; CHECK-NEXT:    [[Z2:%.*]] = fadd double [[Y2]], [[B2]]
+; CHECK-NEXT:    [[R:%.*]] = fmul double [[Z1]], [[Z2]]
+; CHECK-NEXT:    ret double [[R]]
+;
+  %X1 = fsub double %A1, %B1
+  %X2 = fsub double %A2, %B2
+  %Y1 = call double @llvm.cos.f64(double %X1)
+  %Y2 = call double @llvm.cos.f64(double %X2)
+  %Z1 = fadd double %Y1, %B1
+  %Z2 = fadd double %Y2, %B2
+  %R  = fmul double %Z1, %Z2
+  ret double %R
 }
 
 ; Basic depth-3 chain with powi
 define double @test3(double %A1, double %A2, double %B1, double %B2, i32 %P) {
-	%X1 = fsub double %A1, %B1
-	%X2 = fsub double %A2, %B2
-	%Y1 = call double @llvm.powi.f64(double %X1, i32 %P)
-	%Y2 = call double @llvm.powi.f64(double %X2, i32 %P)
-	%Z1 = fadd double %Y1, %B1
-	%Z2 = fadd double %Y2, %B2
-	%R  = fmul double %Z1, %Z2
-	ret double %R
 ; CHECK-LABEL: @test3(
-; CHECK: ret double %R
+; CHECK-NEXT:    [[X1:%.*]] = fsub double [[A1:%.*]], [[B1:%.*]]
+; CHECK-NEXT:    [[X2:%.*]] = fsub double [[A2:%.*]], [[B2:%.*]]
+; CHECK-NEXT:    [[Y1:%.*]] = call double @llvm.powi.f64(double [[X1]], i32 [[P:%.*]])
+; CHECK-NEXT:    [[Y2:%.*]] = call double @llvm.powi.f64(double [[X2]], i32 [[P]])
+; CHECK-NEXT:    [[Z1:%.*]] = fadd double [[Y1]], [[B1]]
+; CHECK-NEXT:    [[Z2:%.*]] = fadd double [[Y2]], [[B2]]
+; CHECK-NEXT:    [[R:%.*]] = fmul double [[Z1]], [[Z2]]
+; CHECK-NEXT:    ret double [[R]]
+;
+  %X1 = fsub double %A1, %B1
+  %X2 = fsub double %A2, %B2
+  %Y1 = call double @llvm.powi.f64(double %X1, i32 %P)
+  %Y2 = call double @llvm.powi.f64(double %X2, i32 %P)
+  %Z1 = fadd double %Y1, %B1
+  %Z2 = fadd double %Y2, %B2
+  %R  = fmul double %Z1, %Z2
+  ret double %R
 }
 
 ; Basic depth-3 chain with powi (different powers: should not vectorize)
 define double @test4(double %A1, double %A2, double %B1, double %B2, i32 %P) {
-	%X1 = fsub double %A1, %B1
-	%X2 = fsub double %A2, %B2
-        %P2 = add i32 %P, 1
-	%Y1 = call double @llvm.powi.f64(double %X1, i32 %P)
-	%Y2 = call double @llvm.powi.f64(double %X2, i32 %P2)
-	%Z1 = fadd double %Y1, %B1
-	%Z2 = fadd double %Y2, %B2
-	%R  = fmul double %Z1, %Z2
-	ret double %R
 ; CHECK-LABEL: @test4(
-; CHECK: ret double %R
+; CHECK-NEXT:    [[X1:%.*]] = fsub double [[A1:%.*]], [[B1:%.*]]
+; CHECK-NEXT:    [[X2:%.*]] = fsub double [[A2:%.*]], [[B2:%.*]]
+; CHECK-NEXT:    [[P2:%.*]] = add i32 [[P:%.*]], 1
+; CHECK-NEXT:    [[Y1:%.*]] = call double @llvm.powi.f64(double [[X1]], i32 [[P]])
+; CHECK-NEXT:    [[Y2:%.*]] = call double @llvm.powi.f64(double [[X2]], i32 [[P2]])
+; CHECK-NEXT:    [[Z1:%.*]] = fadd double [[Y1]], [[B1]]
+; CHECK-NEXT:    [[Z2:%.*]] = fadd double [[Y2]], [[B2]]
+; CHECK-NEXT:    [[R:%.*]] = fmul double [[Z1]], [[Z2]]
+; CHECK-NEXT:    ret double [[R]]
+;
+  %X1 = fsub double %A1, %B1
+  %X2 = fsub double %A2, %B2
+  %P2 = add i32 %P, 1
+  %Y1 = call double @llvm.powi.f64(double %X1, i32 %P)
+  %Y2 = call double @llvm.powi.f64(double %X2, i32 %P2)
+  %Z1 = fadd double %Y1, %B1
+  %Z2 = fadd double %Y2, %B2
+  %R  = fmul double %Z1, %Z2
+  ret double %R
 }
 
diff --git a/llvm/test/Transforms/BBVectorize/X86/simple-ldstr.ll b/llvm/test/Transforms/BBVectorize/X86/simple-ldstr.ll
index 2c05f30d081..a81d9638f5e 100644
--- a/llvm/test/Transforms/BBVectorize/X86/simple-ldstr.ll
+++ b/llvm/test/Transforms/BBVectorize/X86/simple-ldstr.ll
@@ -1,8 +1,21 @@
-target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s
 
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+
 ; Simple 3-pair chain with loads and stores
 define void @test1(double* %a, double* %b, double* %c) nounwind uwtable readonly {
+; CHECK-LABEL: @test1(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[I0_V_I0:%.*]] = bitcast double* [[A:%.*]] to <2 x double>*
+; CHECK-NEXT:    [[I1_V_I0:%.*]] = bitcast double* [[B:%.*]] to <2 x double>*
+; CHECK-NEXT:    [[I0:%.*]] = load <2 x double>, <2 x double>* [[I0_V_I0]], align 8
+; CHECK-NEXT:    [[I1:%.*]] = load <2 x double>, <2 x double>* [[I1_V_I0]], align 8
+; CHECK-NEXT:    [[MUL:%.*]] = fmul <2 x double> [[I0]], [[I1]]
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast double* [[C:%.*]] to <2 x double>*
+; CHECK-NEXT:    store <2 x double> [[MUL]], <2 x double>* [[TMP0]], align 8
+; CHECK-NEXT:    ret void
+;
 entry:
   %i0 = load double, double* %a, align 8
   %i1 = load double, double* %b, align 8
@@ -16,14 +29,5 @@ entry:
   %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   ret void
-; CHECK-LABEL: @test1(
-; CHECK: %i0.v.i0 = bitcast double* %a to <2 x double>*
-; CHECK: %i1.v.i0 = bitcast double* %b to <2 x double>*
-; CHECK: %i0 = load <2 x double>, <2 x double>* %i0.v.i0, align 8
-; CHECK: %i1 = load <2 x double>, <2 x double>* %i1.v.i0, align 8
-; CHECK: %mul = fmul <2 x double> %i0, %i1
-; CHECK: %0 = bitcast double* %c to <2 x double>*
-; CHECK: store <2 x double> %mul, <2 x double>* %0, align 8
-; CHECK: ret void
 }
 
diff --git a/llvm/test/Transforms/BBVectorize/X86/simple.ll b/llvm/test/Transforms/BBVectorize/X86/simple.ll
index a11e3090f20..0f7ddffbd19 100644
--- a/llvm/test/Transforms/BBVectorize/X86/simple.ll
+++ b/llvm/test/Transforms/BBVectorize/X86/simple.ll
@@ -1,120 +1,149 @@
-target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128"
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s
 
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128"
+
 ; Basic depth-3 chain
 define double @test1(double %A1, double %A2, double %B1, double %B2) {
-	%X1 = fsub double %A1, %B1
-	%X2 = fsub double %A2, %B2
-	%Y1 = fmul double %X1, %A1
-	%Y2 = fmul double %X2, %A2
-	%Z1 = fadd double %Y1, %B1
-	%Z2 = fadd double %Y2, %B2
-	%R  = fmul double %Z1, %Z2
-	ret double %R
 ; CHECK-LABEL: @test1(
-; CHECK: fsub <2 x double>
-; CHECK: fmul <2 x double>
-; CHECK: fadd <2 x double>
-; CHECK: extract
-; CHECK: extract
-; CHECK: ret double %R
+; CHECK-NEXT:    [[X1_V_I1_1:%.*]] = insertelement <2 x double> undef, double [[B1:%.*]], i32 0
+; CHECK-NEXT:    [[X1_V_I1_2:%.*]] = insertelement <2 x double> [[X1_V_I1_1]], double [[B2:%.*]], i32 1
+; CHECK-NEXT:    [[X1_V_I0_1:%.*]] = insertelement <2 x double> undef, double [[A1:%.*]], i32 0
+; CHECK-NEXT:    [[X1_V_I0_2:%.*]] = insertelement <2 x double> [[X1_V_I0_1]], double [[A2:%.*]], i32 1
+; CHECK-NEXT:    [[X1:%.*]] = fsub <2 x double> [[X1_V_I0_2]], [[X1_V_I1_2]]
+; CHECK-NEXT:    [[Y1:%.*]] = fmul <2 x double> [[X1]], [[X1_V_I0_2]]
+; CHECK-NEXT:    [[Z1:%.*]] = fadd <2 x double> [[Y1]], [[X1_V_I1_2]]
+; CHECK-NEXT:    [[Z1_V_R1:%.*]] = extractelement <2 x double> [[Z1]], i32 0
+; CHECK-NEXT:    [[Z1_V_R2:%.*]] = extractelement <2 x double> [[Z1]], i32 1
+; CHECK-NEXT:    [[R:%.*]] = fmul double [[Z1_V_R1]], [[Z1_V_R2]]
+; CHECK-NEXT:    ret double [[R]]
+;
+  %X1 = fsub double %A1, %B1
+  %X2 = fsub double %A2, %B2
+  %Y1 = fmul double %X1, %A1
+  %Y2 = fmul double %X2, %A2
+  %Z1 = fadd double %Y1, %B1
+  %Z2 = fadd double %Y2, %B2
+  %R  = fmul double %Z1, %Z2
+  ret double %R
 }
 
 ; Basic chain
 define double @test1a(double %A1, double %A2, double %B1, double %B2) {
-	%X1 = fsub double %A1, %B1
-	%X2 = fsub double %A2, %B2
-	%Y1 = fmul double %X1, %A1
-	%Y2 = fmul double %X2, %A2
-	%Z1 = fadd double %Y1, %B1
-	%Z2 = fadd double %Y2, %B2
-	%W1 = fadd double %Y1, %Z1
-	%W2 = fadd double %Y2, %Z2
-	%V1 = fadd double %W1, %Z1
-	%V2 = fadd double %W2, %Z2
-	%Q1 = fadd double %W1, %V1
-	%Q2 = fadd double %W2, %V2
-	%S1 = fadd double %W1, %Q1
-	%S2 = fadd double %W2, %Q2
-	%R  = fmul double %S1, %S2
-	ret double %R
 ; CHECK-LABEL: @test1a(
-; CHECK: %X1.v.i1.1 = insertelement <2 x double> undef, double %B1, i32 0
-; CHECK: %X1.v.i1.2 = insertelement <2 x double> %X1.v.i1.1, double %B2, i32 1
-; CHECK: %X1.v.i0.1 = insertelement <2 x double> undef, double %A1, i32 0
-; CHECK: %X1.v.i0.2 = insertelement <2 x double> %X1.v.i0.1, double %A2, i32 1
-; CHECK: %X1 = fsub <2 x double> %X1.v.i0.2, %X1.v.i1.2
-; CHECK: %Y1 = fmul <2 x double> %X1, %X1.v.i0.2
-; CHECK: %Z1 = fadd <2 x double> %Y1, %X1.v.i1.2
-; CHECK: %W1 = fadd <2 x double> %Y1, %Z1
-; CHECK: %V1 = fadd <2 x double> %W1, %Z1
-; CHECK: %Q1 = fadd <2 x double> %W1, %V1
-; CHECK: %S1 = fadd <2 x double> %W1, %Q1
-; CHECK: %S1.v.r1 = extractelement <2 x double> %S1, i32 0
-; CHECK: %S1.v.r2 = extractelement <2 x double> %S1, i32 1
-; CHECK: %R = fmul double %S1.v.r1, %S1.v.r2
-; CHECK: ret double %R
+; CHECK-NEXT:    [[X1_V_I1_1:%.*]] = insertelement <2 x double> undef, double [[B1:%.*]], i32 0
+; CHECK-NEXT:    [[X1_V_I1_2:%.*]] = insertelement <2 x double> [[X1_V_I1_1]], double [[B2:%.*]], i32 1
+; CHECK-NEXT:    [[X1_V_I0_1:%.*]] = insertelement <2 x double> undef, double [[A1:%.*]], i32 0
+; CHECK-NEXT:    [[X1_V_I0_2:%.*]] = insertelement <2 x double> [[X1_V_I0_1]], double [[A2:%.*]], i32 1
+; CHECK-NEXT:    [[X1:%.*]] = fsub <2 x double> [[X1_V_I0_2]], [[X1_V_I1_2]]
+; CHECK-NEXT:    [[Y1:%.*]] = fmul <2 x double> [[X1]], [[X1_V_I0_2]]
+; CHECK-NEXT:    [[Z1:%.*]] = fadd <2 x double> [[Y1]], [[X1_V_I1_2]]
+; CHECK-NEXT:    [[W1:%.*]] = fadd <2 x double> [[Y1]], [[Z1]]
+; CHECK-NEXT:    [[V1:%.*]] = fadd <2 x double> [[W1]], [[Z1]]
+; CHECK-NEXT:    [[Q1:%.*]] = fadd <2 x double> [[W1]], [[V1]]
+; CHECK-NEXT:    [[S1:%.*]] = fadd <2 x double> [[W1]], [[Q1]]
+; CHECK-NEXT:    [[S1_V_R1:%.*]] = extractelement <2 x double> [[S1]], i32 0
+; CHECK-NEXT:    [[S1_V_R2:%.*]] = extractelement <2 x double> [[S1]], i32 1
+; CHECK-NEXT:    [[R:%.*]] = fmul double [[S1_V_R1]], [[S1_V_R2]]
+; CHECK-NEXT:    ret double [[R]]
+;
+  %X1 = fsub double %A1, %B1
+  %X2 = fsub double %A2, %B2
+  %Y1 = fmul double %X1, %A1
+  %Y2 = fmul double %X2, %A2
+  %Z1 = fadd double %Y1, %B1
+  %Z2 = fadd double %Y2, %B2
+  %W1 = fadd double %Y1, %Z1
+  %W2 = fadd double %Y2, %Z2
+  %V1 = fadd double %W1, %Z1
+  %V2 = fadd double %W2, %Z2
+  %Q1 = fadd double %W1, %V1
+  %Q2 = fadd double %W2, %V2
+  %S1 = fadd double %W1, %Q1
+  %S2 = fadd double %W2, %Q2
+  %R  = fmul double %S1, %S2
+  ret double %R
 }
 
 ; Basic depth-3 chain (last pair permuted)
 define double @test2(double %A1, double %A2, double %B1, double %B2) {
-	%X1 = fsub double %A1, %B1
-	%X2 = fsub double %A2, %B2
-	%Y1 = fmul double %X1, %A1
-	%Y2 = fmul double %X2, %A2
-	%Z1 = fadd double %Y2, %B1
-	%Z2 = fadd double %Y1, %B2
-	%R  = fmul double %Z1, %Z2
-	ret double %R
 ; CHECK-LABEL: @test2(
-; CHECK: insertelement
-; CHECK: insertelement
-; CHECK: insertelement
-; CHECK: insertelement
-; CHECK: fsub <2 x double>
-; CHECK: fmul <2 x double>
-; CHECK: ret double %R
+; CHECK-NEXT:    [[X1_V_I1_1:%.*]] = insertelement <2 x double> undef, double [[B1:%.*]], i32 0
+; CHECK-NEXT:    [[X1_V_I1_2:%.*]] = insertelement <2 x double> [[X1_V_I1_1]], double [[B2:%.*]], i32 1
+; CHECK-NEXT:    [[X1_V_I0_1:%.*]] = insertelement <2 x double> undef, double [[A1:%.*]], i32 0
+; CHECK-NEXT:    [[X1_V_I0_2:%.*]] = insertelement <2 x double> [[X1_V_I0_1]], double [[A2:%.*]], i32 1
+; CHECK-NEXT:    [[X1:%.*]] = fsub <2 x double> [[X1_V_I0_2]], [[X1_V_I1_2]]
+; CHECK-NEXT:    [[Y1:%.*]] = fmul <2 x double> [[X1]], [[X1_V_I0_2]]
+; CHECK-NEXT:    [[Z1_V_I1_1:%.*]] = insertelement <2 x double> undef, double [[B2]], i32 0
+; CHECK-NEXT:    [[Z1_V_I1_2:%.*]] = insertelement <2 x double> [[Z1_V_I1_1]], double [[B1]], i32 1
+; CHECK-NEXT:    [[Z2:%.*]] = fadd <2 x double> [[Y1]], [[Z1_V_I1_2]]
+; CHECK-NEXT:    [[Z2_V_R1:%.*]] = extractelement <2 x double> [[Z2]], i32 0
+; CHECK-NEXT:    [[Z2_V_R2:%.*]] = extractelement <2 x double> [[Z2]], i32 1
+; CHECK-NEXT:    [[R:%.*]] = fmul double [[Z2_V_R2]], [[Z2_V_R1]]
+; CHECK-NEXT:    ret double [[R]]
+;
+  %X1 = fsub double %A1, %B1
+  %X2 = fsub double %A2, %B2
+  %Y1 = fmul double %X1, %A1
+  %Y2 = fmul double %X2, %A2
+  %Z1 = fadd double %Y2, %B1
+  %Z2 = fadd double %Y1, %B2
+  %R  = fmul double %Z1, %Z2
+  ret double %R
 }
 
 ; Basic depth-4 chain (internal permutation)
 define double @test4(double %A1, double %A2, double %B1, double %B2) {
-	%X1 = fsub double %A1, %B1
-	%X2 = fsub double %A2, %B2
-	%Y1 = fmul double %X1, %A1
-	%Y2 = fmul double %X2, %A2
-	%Z1 = fadd double %Y2, %B1
-	%Z2 = fadd double %Y1, %B2
-	%W1 = fadd double %Y2, %Z1
-	%W2 = fadd double %Y1, %Z2
-	%R  = fmul double %Z1, %Z2
-	ret double %R
 ; CHECK-LABEL: @test4(
-; CHECK: insertelement
-; CHECK: insertelement
-; CHECK: insertelement
-; CHECK: insertelement
-; CHECK: fsub <2 x double>
-; CHECK: fmul <2 x double>
-; CHECK: insertelement
-; CHECK: insertelement
-; CHECK: fadd <2 x double>
-; CHECK: ret double %R
+; CHECK-NEXT:    [[X1_V_I1_1:%.*]] = insertelement <2 x double> undef, double [[B1:%.*]], i32 0
+; CHECK-NEXT:    [[X1_V_I1_2:%.*]] = insertelement <2 x double> [[X1_V_I1_1]], double [[B2:%.*]], i32 1
+; CHECK-NEXT:    [[X1_V_I0_1:%.*]] = insertelement <2 x double> undef, double [[A1:%.*]], i32 0
+; CHECK-NEXT:    [[X1_V_I0_2:%.*]] = insertelement <2 x double> [[X1_V_I0_1]], double [[A2:%.*]], i32 1
+; CHECK-NEXT:    [[X1:%.*]] = fsub <2 x double> [[X1_V_I0_2]], [[X1_V_I1_2]]
+; CHECK-NEXT:    [[Y1:%.*]] = fmul <2 x double> [[X1]], [[X1_V_I0_2]]
+; CHECK-NEXT:    [[Z1_V_I1_1:%.*]] = insertelement <2 x double> undef, double [[B2]], i32 0
+; CHECK-NEXT:    [[Z1_V_I1_2:%.*]] = insertelement <2 x double> [[Z1_V_I1_1]], double [[B1]], i32 1
+; CHECK-NEXT:    [[Z2:%.*]] = fadd <2 x double> [[Y1]], [[Z1_V_I1_2]]
+; CHECK-NEXT:    [[Z2_V_R1:%.*]] = extractelement <2 x double> [[Z2]], i32 0
+; CHECK-NEXT:    [[Z2_V_R2:%.*]] = extractelement <2 x double> [[Z2]], i32 1
+; CHECK-NEXT:    [[R:%.*]] = fmul double [[Z2_V_R2]], [[Z2_V_R1]]
+; CHECK-NEXT:    ret double [[R]]
+;
+  %X1 = fsub double %A1, %B1
+  %X2 = fsub double %A2, %B2
+  %Y1 = fmul double %X1, %A1
+  %Y2 = fmul double %X2, %A2
+  %Z1 = fadd double %Y2, %B1
+  %Z2 = fadd double %Y1, %B2
+  %W1 = fadd double %Y2, %Z1
+  %W2 = fadd double %Y1, %Z2
+  %R  = fmul double %Z1, %Z2
+  ret double %R
 }
 
 ; Basic chain with shuffles
 define <8 x i8> @test6(<8 x i8> %A1, <8 x i8> %A2, <8 x i8> %B1, <8 x i8> %B2) {
-	%X1 = sub <8 x i8> %A1, %B1
-	%X2 = sub <8 x i8> %A2, %B2
-	%Y1 = mul <8 x i8> %X1, %A1
-	%Y2 = mul <8 x i8> %X2, %A2
-	%Z1 = add <8 x i8> %Y1, %B1
-	%Z2 = add <8 x i8> %Y2, %B2
-        %Q1 = shufflevector <8 x i8> %Z1, <8 x i8> %Z2, <8 x i32> <i32 15, i32 8, i32 6, i32 1, i32 13, i32 10, i32 4, i32 3>
-        %Q2 = shufflevector <8 x i8> %Z2, <8 x i8> %Z2, <8 x i32> <i32 6, i32 7, i32 0, i32 1, i32 2, i32 4, i32 4, i32 1>
-	%R  = mul <8 x i8> %Q1, %Q2
-	ret <8 x i8> %R
 ; CHECK-LABEL: @test6(
-; CHECK-NOT: sub <16 x i8>
-; CHECK: ret <8 x i8>
+; CHECK-NEXT:    [[X1:%.*]] = sub <8 x i8> [[A1:%.*]], [[B1:%.*]]
+; CHECK-NEXT:    [[X2:%.*]] = sub <8 x i8> [[A2:%.*]], [[B2:%.*]]
+; CHECK-NEXT:    [[Y1:%.*]] = mul <8 x i8> [[X1]], [[A1]]
+; CHECK-NEXT:    [[Y2:%.*]] = mul <8 x i8> [[X2]], [[A2]]
+; CHECK-NEXT:    [[Z1:%.*]] = add <8 x i8> [[Y1]], [[B1]]
+; CHECK-NEXT:    [[Z2:%.*]] = add <8 x i8> [[Y2]], [[B2]]
+; CHECK-NEXT:    [[Q1:%.*]] = shufflevector <8 x i8> [[Z1]], <8 x i8> [[Z2]], <8 x i32> <i32 15, i32 8, i32 6, i32 1, i32 13, i32 10, i32 4, i32 3>
+; CHECK-NEXT:    [[Q2:%.*]] = shufflevector <8 x i8> [[Z2]], <8 x i8> undef, <8 x i32> <i32 6, i32 7, i32 0, i32 1, i32 2, i32 4, i32 4, i32 1>
+; CHECK-NEXT:    [[R:%.*]] = mul <8 x i8> [[Q1]], [[Q2]]
+; CHECK-NEXT:    ret <8 x i8> [[R]]
+;
+  %X1 = sub <8 x i8> %A1, %B1
+  %X2 = sub <8 x i8> %A2, %B2
+  %Y1 = mul <8 x i8> %X1, %A1
+  %Y2 = mul <8 x i8> %X2, %A2
+  %Z1 = add <8 x i8> %Y1, %B1
+  %Z2 = add <8 x i8> %Y2, %B2
+  %Q1 = shufflevector <8 x i8> %Z1, <8 x i8> %Z2, <8 x i32> <i32 15, i32 8, i32 6, i32 1, i32 13, i32 10, i32 4, i32 3>
+  %Q2 = shufflevector <8 x i8> %Z2, <8 x i8> %Z2, <8 x i32> <i32 6, i32 7, i32 0, i32 1, i32 2, i32 4, i32 4, i32 1>
+  %R  = mul <8 x i8> %Q1, %Q2
+  ret <8 x i8> %R
 }
 
diff --git a/llvm/test/Transforms/BBVectorize/X86/vs-cast.ll b/llvm/test/Transforms/BBVectorize/X86/vs-cast.ll
index 0c666b11976..297f2d5a7b3 100644
--- a/llvm/test/Transforms/BBVectorize/X86/vs-cast.ll
+++ b/llvm/test/Transforms/BBVectorize/X86/vs-cast.ll
@@ -1,12 +1,19 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -bb-vectorize -S | FileCheck %s
+
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
 target triple = "x86_64-unknown-linux-gnu"
-; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -bb-vectorize -S | FileCheck %s
 
 define void @main() nounwind uwtable {
+; CHECK-LABEL: @main(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <2 x i64> undef to i128
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> undef to i128
+; CHECK-NEXT:    ret void
+;
 entry:
   %0 = bitcast <2 x i64> undef to i128
   %1 = bitcast <2 x i64> undef to i128
   ret void
-; CHECK-LABEL: @main(
 }