From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Mon, 21 Nov 2016 15:28:21 +0000 (+0000)
Subject: [X86][SSE] Add SSE reciprocal estimate tests
X-Git-Tag: android-x86-7.1-r4~24197
X-Git-Url: http://git.osdn.net/view?a=commitdiff_plain;h=ab814ecdff8983c9dcfef7aa7de457288a6d7f2e;p=android-x86%2Fexternal-llvm.git

[X86][SSE] Add SSE reciprocal estimate tests

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287543 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/test/CodeGen/X86/recip-fastmath.ll b/test/CodeGen/X86/recip-fastmath.ll
index bd622d0442e..0a99254cd62 100644
--- a/test/CodeGen/X86/recip-fastmath.ll
+++ b/test/CodeGen/X86/recip-fastmath.ll
@@ -1,5 +1,6 @@
-; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx  | FileCheck %s --check-prefix=AVX
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=CHECK --check-prefix=SSE
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx  | FileCheck %s --check-prefix=CHECK --check-prefix=AVX
 
 ; If the target's divss/divps instructions are substantially
 ; slower than rcpss/rcpps with a Newton-Raphson refinement,
@@ -10,17 +11,34 @@
 ; differences of x86 reciprocal estimates.
 
 define float @f32_no_estimate(float %x) #0 {
+; SSE-LABEL: f32_no_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
+; SSE-NEXT:    divss %xmm0, %xmm1
+; SSE-NEXT:    movaps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: f32_no_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
 ; AVX-NEXT:    vdivss %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    retq
-;
   %div = fdiv fast float 1.0, %x
   ret float %div
 }
 
 define float @f32_one_step(float %x) #1 {
+; SSE-LABEL: f32_one_step:
+; SSE:       # BB#0:
+; SSE-NEXT:    rcpss %xmm0, %xmm2
+; SSE-NEXT:    mulss %xmm2, %xmm0
+; SSE-NEXT:    movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
+; SSE-NEXT:    subss %xmm0, %xmm1
+; SSE-NEXT:    mulss %xmm2, %xmm1
+; SSE-NEXT:    addss %xmm2, %xmm1
+; SSE-NEXT:    movaps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: f32_one_step:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vrcpss %xmm0, %xmm0, %xmm1
@@ -30,12 +48,28 @@ define float @f32_one_step(float %x) #1 {
 ; AVX-NEXT:    vmulss %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    vaddss %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    retq
-;
   %div = fdiv fast float 1.0, %x
   ret float %div
 }
 
 define float @f32_two_step(float %x) #2 {
+; SSE-LABEL: f32_two_step:
+; SSE:       # BB#0:
+; SSE-NEXT:    rcpss %xmm0, %xmm2
+; SSE-NEXT:    movaps %xmm0, %xmm3
+; SSE-NEXT:    mulss %xmm2, %xmm3
+; SSE-NEXT:    movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
+; SSE-NEXT:    movaps %xmm1, %xmm4
+; SSE-NEXT:    subss %xmm3, %xmm4
+; SSE-NEXT:    mulss %xmm2, %xmm4
+; SSE-NEXT:    addss %xmm2, %xmm4
+; SSE-NEXT:    mulss %xmm4, %xmm0
+; SSE-NEXT:    subss %xmm0, %xmm1
+; SSE-NEXT:    mulss %xmm4, %xmm1
+; SSE-NEXT:    addss %xmm4, %xmm1
+; SSE-NEXT:    movaps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: f32_two_step:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vrcpss %xmm0, %xmm0, %xmm1
@@ -49,23 +83,39 @@ define float @f32_two_step(float %x) #2 {
 ; AVX-NEXT:    vmulss %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    vaddss %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    retq
-;
   %div = fdiv fast float 1.0, %x
   ret float %div
 }
 
 define <4 x float> @v4f32_no_estimate(<4 x float> %x) #0 {
+; SSE-LABEL: v4f32_no_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    movaps {{.*#+}} xmm1 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
+; SSE-NEXT:    divps %xmm0, %xmm1
+; SSE-NEXT:    movaps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: v4f32_no_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vmovaps {{.*#+}} xmm1 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
 ; AVX-NEXT:    vdivps %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    retq
-;
   %div = fdiv fast <4 x float> <float 1.0, float 1.0, float 1.0, float 1.0>, %x
   ret <4 x float> %div
 }
 
 define <4 x float> @v4f32_one_step(<4 x float> %x) #1 {
+; SSE-LABEL: v4f32_one_step:
+; SSE:       # BB#0:
+; SSE-NEXT:    rcpps %xmm0, %xmm2
+; SSE-NEXT:    mulps %xmm2, %xmm0
+; SSE-NEXT:    movaps {{.*#+}} xmm1 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
+; SSE-NEXT:    subps %xmm0, %xmm1
+; SSE-NEXT:    mulps %xmm2, %xmm1
+; SSE-NEXT:    addps %xmm2, %xmm1
+; SSE-NEXT:    movaps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: v4f32_one_step:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vrcpps %xmm0, %xmm1
@@ -75,12 +125,28 @@ define <4 x float> @v4f32_one_step(<4 x float> %x) #1 {
 ; AVX-NEXT:    vmulps %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    vaddps %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    retq
-;
   %div = fdiv fast <4 x float> <float 1.0, float 1.0, float 1.0, float 1.0>, %x
   ret <4 x float> %div
 }
 
 define <4 x float> @v4f32_two_step(<4 x float> %x) #2 {
+; SSE-LABEL: v4f32_two_step:
+; SSE:       # BB#0:
+; SSE-NEXT:    rcpps %xmm0, %xmm2
+; SSE-NEXT:    movaps %xmm0, %xmm3
+; SSE-NEXT:    mulps %xmm2, %xmm3
+; SSE-NEXT:    movaps {{.*#+}} xmm1 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
+; SSE-NEXT:    movaps %xmm1, %xmm4
+; SSE-NEXT:    subps %xmm3, %xmm4
+; SSE-NEXT:    mulps %xmm2, %xmm4
+; SSE-NEXT:    addps %xmm2, %xmm4
+; SSE-NEXT:    mulps %xmm4, %xmm0
+; SSE-NEXT:    subps %xmm0, %xmm1
+; SSE-NEXT:    mulps %xmm4, %xmm1
+; SSE-NEXT:    addps %xmm4, %xmm1
+; SSE-NEXT:    movaps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: v4f32_two_step:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vrcpps %xmm0, %xmm1
@@ -94,23 +160,49 @@ define <4 x float> @v4f32_two_step(<4 x float> %x) #2 {
 ; AVX-NEXT:    vmulps %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    vaddps %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    retq
-;
   %div = fdiv fast <4 x float> <float 1.0, float 1.0, float 1.0, float 1.0>, %x
   ret <4 x float> %div
 }
 
 define <8 x float> @v8f32_no_estimate(<8 x float> %x) #0 {
+; SSE-LABEL: v8f32_no_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    movaps {{.*#+}} xmm2 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
+; SSE-NEXT:    movaps %xmm2, %xmm3
+; SSE-NEXT:    divps %xmm0, %xmm3
+; SSE-NEXT:    divps %xmm1, %xmm2
+; SSE-NEXT:    movaps %xmm3, %xmm0
+; SSE-NEXT:    movaps %xmm2, %xmm1
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: v8f32_no_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vmovaps {{.*#+}} ymm1 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
 ; AVX-NEXT:    vdivps %ymm0, %ymm1, %ymm0
 ; AVX-NEXT:    retq
-;
   %div = fdiv fast <8 x float> <float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0>, %x
   ret <8 x float> %div
 }
 
 define <8 x float> @v8f32_one_step(<8 x float> %x) #1 {
+; SSE-LABEL: v8f32_one_step:
+; SSE:       # BB#0:
+; SSE-NEXT:    rcpps %xmm0, %xmm4
+; SSE-NEXT:    mulps %xmm4, %xmm0
+; SSE-NEXT:    movaps {{.*#+}} xmm2 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
+; SSE-NEXT:    movaps %xmm2, %xmm3
+; SSE-NEXT:    subps %xmm0, %xmm3
+; SSE-NEXT:    mulps %xmm4, %xmm3
+; SSE-NEXT:    addps %xmm4, %xmm3
+; SSE-NEXT:    rcpps %xmm1, %xmm0
+; SSE-NEXT:    mulps %xmm0, %xmm1
+; SSE-NEXT:    subps %xmm1, %xmm2
+; SSE-NEXT:    mulps %xmm0, %xmm2
+; SSE-NEXT:    addps %xmm0, %xmm2
+; SSE-NEXT:    movaps %xmm3, %xmm0
+; SSE-NEXT:    movaps %xmm2, %xmm1
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: v8f32_one_step:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vrcpps %ymm0, %ymm1
@@ -120,12 +212,41 @@ define <8 x float> @v8f32_one_step(<8 x float> %x) #1 {
 ; AVX-NEXT:    vmulps %ymm0, %ymm1, %ymm0
 ; AVX-NEXT:    vaddps %ymm0, %ymm1, %ymm0
 ; AVX-NEXT:    retq
-;
   %div = fdiv fast <8 x float> <float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0>, %x
   ret <8 x float> %div
 }
 
 define <8 x float> @v8f32_two_step(<8 x float> %x) #2 {
+; SSE-LABEL: v8f32_two_step:
+; SSE:       # BB#0:
+; SSE-NEXT:    movaps %xmm1, %xmm2
+; SSE-NEXT:    rcpps %xmm0, %xmm3
+; SSE-NEXT:    movaps %xmm0, %xmm4
+; SSE-NEXT:    mulps %xmm3, %xmm4
+; SSE-NEXT:    movaps {{.*#+}} xmm1 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
+; SSE-NEXT:    movaps %xmm1, %xmm5
+; SSE-NEXT:    subps %xmm4, %xmm5
+; SSE-NEXT:    mulps %xmm3, %xmm5
+; SSE-NEXT:    addps %xmm3, %xmm5
+; SSE-NEXT:    mulps %xmm5, %xmm0
+; SSE-NEXT:    movaps %xmm1, %xmm3
+; SSE-NEXT:    subps %xmm0, %xmm3
+; SSE-NEXT:    mulps %xmm5, %xmm3
+; SSE-NEXT:    addps %xmm5, %xmm3
+; SSE-NEXT:    rcpps %xmm2, %xmm0
+; SSE-NEXT:    movaps %xmm2, %xmm4
+; SSE-NEXT:    mulps %xmm0, %xmm4
+; SSE-NEXT:    movaps %xmm1, %xmm5
+; SSE-NEXT:    subps %xmm4, %xmm5
+; SSE-NEXT:    mulps %xmm0, %xmm5
+; SSE-NEXT:    addps %xmm0, %xmm5
+; SSE-NEXT:    mulps %xmm5, %xmm2
+; SSE-NEXT:    subps %xmm2, %xmm1
+; SSE-NEXT:    mulps %xmm5, %xmm1
+; SSE-NEXT:    addps %xmm5, %xmm1
+; SSE-NEXT:    movaps %xmm3, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: v8f32_two_step:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vrcpps %ymm0, %ymm1
@@ -139,7 +260,6 @@ define <8 x float> @v8f32_two_step(<8 x float> %x) #2 {
 ; AVX-NEXT:    vmulps %ymm0, %ymm1, %ymm0
 ; AVX-NEXT:    vaddps %ymm0, %ymm1, %ymm0
 ; AVX-NEXT:    retq
-;
   %div = fdiv fast <8 x float> <float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0>, %x
   ret <8 x float> %div
 }
diff --git a/test/CodeGen/X86/sqrt-fastmath.ll b/test/CodeGen/X86/sqrt-fastmath.ll
index 68424c60aa6..af2dcc495f5 100644
--- a/test/CodeGen/X86/sqrt-fastmath.ll
+++ b/test/CodeGen/X86/sqrt-fastmath.ll
@@ -1,5 +1,6 @@
-; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx | FileCheck %s --check-prefix=AVX
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=CHECK --check-prefix=SSE
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx  | FileCheck %s --check-prefix=CHECK --check-prefix=AVX
 
 declare double @__sqrt_finite(double)
 declare float @__sqrtf_finite(float)
@@ -10,11 +11,15 @@ declare <8 x float> @llvm.sqrt.v8f32(<8 x float>)
 
 
 define double @finite_f64_no_estimate(double %d) #0 {
+; SSE-LABEL: finite_f64_no_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    sqrtsd %xmm0, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: finite_f64_no_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vsqrtsd %xmm0, %xmm0, %xmm0
 ; AVX-NEXT:    retq
-;
   %call = tail call double @__sqrt_finite(double %d) #2
   ret double %call
 }
@@ -22,26 +27,49 @@ define double @finite_f64_no_estimate(double %d) #0 {
 ; No estimates for doubles.
 
 define double @finite_f64_estimate(double %d) #1 {
+; SSE-LABEL: finite_f64_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    sqrtsd %xmm0, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: finite_f64_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vsqrtsd %xmm0, %xmm0, %xmm0
 ; AVX-NEXT:    retq
-;
   %call = tail call double @__sqrt_finite(double %d) #2
   ret double %call
 }
 
 define float @finite_f32_no_estimate(float %f) #0 {
+; SSE-LABEL: finite_f32_no_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    sqrtss %xmm0, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: finite_f32_no_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vsqrtss %xmm0, %xmm0, %xmm0
 ; AVX-NEXT:    retq
-;
   %call = tail call float @__sqrtf_finite(float %f) #2
   ret float %call
 }
 
 define float @finite_f32_estimate(float %f) #1 {
+; SSE-LABEL: finite_f32_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    rsqrtss %xmm0, %xmm1
+; SSE-NEXT:    movaps %xmm0, %xmm2
+; SSE-NEXT:    mulss %xmm1, %xmm2
+; SSE-NEXT:    movss {{.*#+}} xmm3 = mem[0],zero,zero,zero
+; SSE-NEXT:    mulss %xmm2, %xmm3
+; SSE-NEXT:    mulss %xmm1, %xmm2
+; SSE-NEXT:    addss {{.*}}(%rip), %xmm2
+; SSE-NEXT:    mulss %xmm3, %xmm2
+; SSE-NEXT:    xorps %xmm1, %xmm1
+; SSE-NEXT:    cmpeqss %xmm1, %xmm0
+; SSE-NEXT:    andnps %xmm2, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: finite_f32_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vrsqrtss %xmm0, %xmm0, %xmm1
@@ -54,18 +82,16 @@ define float @finite_f32_estimate(float %f) #1 {
 ; AVX-NEXT:    vcmpeqss %xmm2, %xmm0, %xmm0
 ; AVX-NEXT:    vandnps %xmm1, %xmm0, %xmm0
 ; AVX-NEXT:    retq
-;
   %call = tail call float @__sqrtf_finite(float %f) #2
   ret float %call
 }
 
 define x86_fp80 @finite_f80_no_estimate(x86_fp80 %ld) #0 {
-; AVX-LABEL: finite_f80_no_estimate:
-; AVX:       # BB#0:
-; AVX-NEXT:    fldt {{[0-9]+}}(%rsp)
-; AVX-NEXT:    fsqrt
-; AVX-NEXT:    retq
-;
+; CHECK-LABEL: finite_f80_no_estimate:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    fldt {{[0-9]+}}(%rsp)
+; CHECK-NEXT:    fsqrt
+; CHECK-NEXT:    retq
   %call = tail call x86_fp80 @__sqrtl_finite(x86_fp80 %ld) #2
   ret x86_fp80 %call
 }
@@ -73,30 +99,47 @@ define x86_fp80 @finite_f80_no_estimate(x86_fp80 %ld) #0 {
 ; Don't die on the impossible.
 
 define x86_fp80 @finite_f80_estimate_but_no(x86_fp80 %ld) #1 {
-; AVX-LABEL: finite_f80_estimate_but_no:
-; AVX:       # BB#0:
-; AVX-NEXT:    fldt {{[0-9]+}}(%rsp)
-; AVX-NEXT:    fsqrt
-; AVX-NEXT:    retq
-;
+; CHECK-LABEL: finite_f80_estimate_but_no:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    fldt {{[0-9]+}}(%rsp)
+; CHECK-NEXT:    fsqrt
+; CHECK-NEXT:    retq
   %call = tail call x86_fp80 @__sqrtl_finite(x86_fp80 %ld) #2
   ret x86_fp80 %call
 }
 
 define float @f32_no_estimate(float %x) #0 {
+; SSE-LABEL: f32_no_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    sqrtss %xmm0, %xmm1
+; SSE-NEXT:    movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; SSE-NEXT:    divss %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: f32_no_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vsqrtss %xmm0, %xmm0, %xmm0
 ; AVX-NEXT:    vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
 ; AVX-NEXT:    vdivss %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    retq
-;
   %sqrt = tail call float @llvm.sqrt.f32(float %x)
   %div = fdiv fast float 1.0, %sqrt
   ret float %div
 }
 
 define float @f32_estimate(float %x) #1 {
+; SSE-LABEL: f32_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    rsqrtss %xmm0, %xmm1
+; SSE-NEXT:    movaps %xmm1, %xmm2
+; SSE-NEXT:    mulss %xmm2, %xmm2
+; SSE-NEXT:    mulss %xmm0, %xmm2
+; SSE-NEXT:    addss {{.*}}(%rip), %xmm2
+; SSE-NEXT:    mulss {{.*}}(%rip), %xmm1
+; SSE-NEXT:    mulss %xmm2, %xmm1
+; SSE-NEXT:    movaps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: f32_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vrsqrtss %xmm0, %xmm0, %xmm1
@@ -106,26 +149,43 @@ define float @f32_estimate(float %x) #1 {
 ; AVX-NEXT:    vmulss {{.*}}(%rip), %xmm1, %xmm1
 ; AVX-NEXT:    vmulss %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    retq
-;
   %sqrt = tail call float @llvm.sqrt.f32(float %x)
   %div = fdiv fast float 1.0, %sqrt
   ret float %div
 }
 
 define <4 x float> @v4f32_no_estimate(<4 x float> %x) #0 {
+; SSE-LABEL: v4f32_no_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    sqrtps %xmm0, %xmm1
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
+; SSE-NEXT:    divps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: v4f32_no_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vsqrtps %xmm0, %xmm0
 ; AVX-NEXT:    vmovaps {{.*#+}} xmm1 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
 ; AVX-NEXT:    vdivps %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    retq
-;
   %sqrt = tail call <4 x float> @llvm.sqrt.v4f32(<4 x float> %x)
   %div = fdiv fast <4 x float> <float 1.0, float 1.0, float 1.0, float 1.0>, %sqrt
   ret <4 x float> %div
 }
 
 define <4 x float> @v4f32_estimate(<4 x float> %x) #1 {
+; SSE-LABEL: v4f32_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    rsqrtps %xmm0, %xmm1
+; SSE-NEXT:    movaps %xmm1, %xmm2
+; SSE-NEXT:    mulps %xmm2, %xmm2
+; SSE-NEXT:    mulps %xmm0, %xmm2
+; SSE-NEXT:    addps {{.*}}(%rip), %xmm2
+; SSE-NEXT:    mulps {{.*}}(%rip), %xmm1
+; SSE-NEXT:    mulps %xmm2, %xmm1
+; SSE-NEXT:    movaps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: v4f32_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vrsqrtps %xmm0, %xmm1
@@ -135,26 +195,56 @@ define <4 x float> @v4f32_estimate(<4 x float> %x) #1 {
 ; AVX-NEXT:    vmulps {{.*}}(%rip), %xmm1, %xmm1
 ; AVX-NEXT:    vmulps %xmm0, %xmm1, %xmm0
 ; AVX-NEXT:    retq
-;
   %sqrt = tail call <4 x float> @llvm.sqrt.v4f32(<4 x float> %x)
   %div = fdiv fast <4 x float> <float 1.0, float 1.0, float 1.0, float 1.0>, %sqrt
   ret <4 x float> %div
 }
 
 define <8 x float> @v8f32_no_estimate(<8 x float> %x) #0 {
+; SSE-LABEL: v8f32_no_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    sqrtps %xmm1, %xmm2
+; SSE-NEXT:    sqrtps %xmm0, %xmm3
+; SSE-NEXT:    movaps {{.*#+}} xmm1 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
+; SSE-NEXT:    movaps %xmm1, %xmm0
+; SSE-NEXT:    divps %xmm3, %xmm0
+; SSE-NEXT:    divps %xmm2, %xmm1
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: v8f32_no_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vsqrtps %ymm0, %ymm0
 ; AVX-NEXT:    vmovaps {{.*#+}} ymm1 = [1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00,1.000000e+00]
 ; AVX-NEXT:    vdivps %ymm0, %ymm1, %ymm0
 ; AVX-NEXT:    retq
-;
   %sqrt = tail call <8 x float> @llvm.sqrt.v8f32(<8 x float> %x)
   %div = fdiv fast <8 x float> <float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0>, %sqrt
   ret <8 x float> %div
 }
 
 define <8 x float> @v8f32_estimate(<8 x float> %x) #1 {
+; SSE-LABEL: v8f32_estimate:
+; SSE:       # BB#0:
+; SSE-NEXT:    rsqrtps %xmm0, %xmm3
+; SSE-NEXT:    movaps {{.*#+}} xmm4 = [-5.000000e-01,-5.000000e-01,-5.000000e-01,-5.000000e-01]
+; SSE-NEXT:    movaps %xmm3, %xmm2
+; SSE-NEXT:    mulps %xmm2, %xmm2
+; SSE-NEXT:    mulps %xmm0, %xmm2
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [-3.000000e+00,-3.000000e+00,-3.000000e+00,-3.000000e+00]
+; SSE-NEXT:    addps %xmm0, %xmm2
+; SSE-NEXT:    mulps %xmm4, %xmm2
+; SSE-NEXT:    mulps %xmm3, %xmm2
+; SSE-NEXT:    rsqrtps %xmm1, %xmm5
+; SSE-NEXT:    movaps %xmm5, %xmm3
+; SSE-NEXT:    mulps %xmm3, %xmm3
+; SSE-NEXT:    mulps %xmm1, %xmm3
+; SSE-NEXT:    addps %xmm0, %xmm3
+; SSE-NEXT:    mulps %xmm4, %xmm3
+; SSE-NEXT:    mulps %xmm5, %xmm3
+; SSE-NEXT:    movaps %xmm2, %xmm0
+; SSE-NEXT:    movaps %xmm3, %xmm1
+; SSE-NEXT:    retq
+;
 ; AVX-LABEL: v8f32_estimate:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    vrsqrtps %ymm0, %ymm1
@@ -164,7 +254,6 @@ define <8 x float> @v8f32_estimate(<8 x float> %x) #1 {
 ; AVX-NEXT:    vmulps {{.*}}(%rip), %ymm1, %ymm1
 ; AVX-NEXT:    vmulps %ymm0, %ymm1, %ymm0
 ; AVX-NEXT:    retq
-;
   %sqrt = tail call <8 x float> @llvm.sqrt.v8f32(<8 x float> %x)
   %div = fdiv fast <8 x float> <float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0, float 1.0>, %sqrt
   ret <8 x float> %div