[DAGCombiner] Generalize FADD constant combines to work with vectors

Updated the FADD combines to work with vectors as well as scalars.

Differential Revision: http://reviews.llvm.org/D13416

llvm-svn: 249251

2 files changed, 49 insertions, 24 deletions

diff --git a/llvm/test/CodeGen/X86/fadd-combines.ll b/llvm/test/CodeGen/X86/fadd-combines.ll
index 5eb455d14c1..225b6421c81 100644
--- a/llvm/test/CodeGen/X86/fadd-combines.ll
+++ b/llvm/test/CodeGen/X86/fadd-combines.ll
@@ -11,13 +11,12 @@ define float @fadd_zero_f32(float %x) #0 {
 define <4 x float> @fadd_zero_4f32(<4 x float> %x) #0 {
 ; CHECK-LABEL: fadd_zero_4f32:
 ; CHECK:       # BB#0:
-; CHECK-NEXT:    xorps %xmm1, %xmm1
-; CHECK-NEXT:    addps %xmm1, %xmm0
 ; CHECK-NEXT:    retq
   %y = fadd <4 x float> %x, zeroinitializer
   ret <4 x float> %y
 }
 
+; CHECK: float 3
 define float @fadd_2const_f32(float %x) #0 {
 ; CHECK-LABEL: fadd_2const_f32:
 ; CHECK:       # BB#0:
@@ -28,17 +27,21 @@ define float @fadd_2const_f32(float %x) #0 {
   ret float %z
 }
 
+; CHECK: float 5.000000e+00
+; CHECK: float 5.000000e+00
+; CHECK: float 5.000000e+00
+; CHECK: float 5.000000e+00
 define <4 x float> @fadd_2const_4f32(<4 x float> %x) #0 {
 ; CHECK-LABEL: fadd_2const_4f32:
 ; CHECK:       # BB#0:
 ; CHECK-NEXT:    addps {{.*}}(%rip), %xmm0
-; CHECK-NEXT:    addps {{.*}}(%rip), %xmm0
 ; CHECK-NEXT:    retq
   %y = fadd <4 x float> %x, <float 1.0, float 2.0, float 3.0, float 4.0>
   %z = fadd <4 x float> %y, <float 4.0, float 3.0, float 2.0, float 1.0>
   ret <4 x float> %z
 }
 
+; CHECK: float 3
 define float @fadd_x_fmul_x_c_f32(float %x) #0 {
 ; CHECK-LABEL: fadd_x_fmul_x_c_f32:
 ; CHECK:       # BB#0:
@@ -49,18 +52,21 @@ define float @fadd_x_fmul_x_c_f32(float %x) #0 {
   ret float %z
 }
 
+; CHECK: float 2.000000e+00
+; CHECK: float 3.000000e+00
+; CHECK: float 4.000000e+00
+; CHECK: float 5.000000e+00
 define <4 x float> @fadd_x_fmul_x_c_4f32(<4 x float> %x) #0 {
 ; CHECK-LABEL: fadd_x_fmul_x_c_4f32:
 ; CHECK:       # BB#0:
-; CHECK-NEXT:    movaps {{.*#+}} xmm1 = [1.000000e+00,2.000000e+00,3.000000e+00,4.000000e+00]
-; CHECK-NEXT:    mulps %xmm0, %xmm1
-; CHECK-NEXT:    addps %xmm1, %xmm0
+; CHECK-NEXT:    mulps {{.*}}(%rip), %xmm0
 ; CHECK-NEXT:    retq
   %y = fmul <4 x float> %x, <float 1.0, float 2.0, float 3.0, float 4.0>
   %z = fadd <4 x float> %x, %y
   ret <4 x float> %z
 }
 
+; CHECK: float 3
 define float @fadd_fmul_x_c_x_f32(float %x) #0 {
 ; CHECK-LABEL: fadd_fmul_x_c_x_f32:
 ; CHECK:       # BB#0:
@@ -71,18 +77,21 @@ define float @fadd_fmul_x_c_x_f32(float %x) #0 {
   ret float %z
 }
 
+; CHECK: float 2.000000e+00
+; CHECK: float 3.000000e+00
+; CHECK: float 4.000000e+00
+; CHECK: float 5.000000e+00
 define <4 x float> @fadd_fmul_x_c_x_4f32(<4 x float> %x) #0 {
 ; CHECK-LABEL: fadd_fmul_x_c_x_4f32:
 ; CHECK:       # BB#0:
-; CHECK-NEXT:    movaps {{.*#+}} xmm1 = [1.000000e+00,2.000000e+00,3.000000e+00,4.000000e+00]
-; CHECK-NEXT:    mulps %xmm0, %xmm1
-; CHECK-NEXT:    addps %xmm1, %xmm0
+; CHECK-NEXT:    mulps {{.*}}(%rip), %xmm0
 ; CHECK-NEXT:    retq
   %y = fmul <4 x float> %x, <float 1.0, float 2.0, float 3.0, float 4.0>
   %z = fadd <4 x float> %y, %x
   ret <4 x float> %z
 }
 
+; CHECK: float 4
 define float @fadd_fadd_x_x_fmul_x_c_f32(float %x) #0 {
 ; CHECK-LABEL: fadd_fadd_x_x_fmul_x_c_f32:
 ; CHECK:       # BB#0:
@@ -94,13 +103,14 @@ define float @fadd_fadd_x_x_fmul_x_c_f32(float %x) #0 {
   ret float %w
 }
 
+; CHECK: float 3.000000e+00
+; CHECK: float 4.000000e+00
+; CHECK: float 5.000000e+00
+; CHECK: float 6.000000e+00
 define <4 x float> @fadd_fadd_x_x_fmul_x_c_4f32(<4 x float> %x) #0 {
 ; CHECK-LABEL: fadd_fadd_x_x_fmul_x_c_4f32:
 ; CHECK:       # BB#0:
-; CHECK-NEXT:    movaps {{.*#+}} xmm1 = [1.000000e+00,2.000000e+00,3.000000e+00,4.000000e+00]
-; CHECK-NEXT:    mulps %xmm0, %xmm1
-; CHECK-NEXT:    addps %xmm0, %xmm1
-; CHECK-NEXT:    addps %xmm1, %xmm0
+; CHECK-NEXT:    mulps {{.*}}(%rip), %xmm0
 ; CHECK-NEXT:    retq
   %y = fadd <4 x float> %x, %x
   %z = fmul <4 x float> %x, <float 1.0, float 2.0, float 3.0, float 4.0>
@@ -108,6 +118,7 @@ define <4 x float> @fadd_fadd_x_x_fmul_x_c_4f32(<4 x float> %x) #0 {
   ret <4 x float> %w
 }
 
+; CHECK: float 4
 define float @fadd_fmul_x_c_fadd_x_x_f32(float %x) #0 {
 ; CHECK-LABEL: fadd_fmul_x_c_fadd_x_x_f32:
 ; CHECK:       # BB#0:
@@ -119,13 +130,14 @@ define float @fadd_fmul_x_c_fadd_x_x_f32(float %x) #0 {
   ret float %w
 }
 
+; CHECK: float 3.000000e+00
+; CHECK: float 4.000000e+00
+; CHECK: float 5.000000e+00
+; CHECK: float 6.000000e+00
 define <4 x float> @fadd_fmul_x_c_fadd_x_x_4f32(<4 x float> %x) #0 {
 ; CHECK-LABEL: fadd_fmul_x_c_fadd_x_x_4f32:
 ; CHECK:       # BB#0:
-; CHECK-NEXT:    movaps {{.*#+}} xmm1 = [1.000000e+00,2.000000e+00,3.000000e+00,4.000000e+00]
-; CHECK-NEXT:    mulps %xmm0, %xmm1
-; CHECK-NEXT:    addps %xmm0, %xmm1
-; CHECK-NEXT:    addps %xmm1, %xmm0
+; CHECK-NEXT:    mulps {{.*}}(%rip), %xmm0
 ; CHECK-NEXT:    retq
   %y = fadd <4 x float> %x, %x
   %z = fmul <4 x float> %x, <float 1.0, float 2.0, float 3.0, float 4.0>
@@ -133,6 +145,7 @@ define <4 x float> @fadd_fmul_x_c_fadd_x_x_4f32(<4 x float> %x) #0 {
   ret <4 x float> %w
 }
 
+; CHECK: float 3
 define float @fadd_x_fadd_x_x_f32(float %x) #0 {
 ; CHECK-LABEL: fadd_x_fadd_x_x_f32:
 ; CHECK:       # BB#0:
@@ -143,6 +156,10 @@ define float @fadd_x_fadd_x_x_f32(float %x) #0 {
   ret float %z
 }
 
+; CHECK: float 3.000000e+00
+; CHECK: float 3.000000e+00
+; CHECK: float 3.000000e+00
+; CHECK: float 3.000000e+00
 define <4 x float> @fadd_x_fadd_x_x_4f32(<4 x float> %x) #0 {
 ; CHECK-LABEL: fadd_x_fadd_x_x_4f32:
 ; CHECK:       # BB#0:
@@ -153,6 +170,7 @@ define <4 x float> @fadd_x_fadd_x_x_4f32(<4 x float> %x) #0 {
   ret <4 x float> %z
 }
 
+; CHECK: float 3
 define float @fadd_fadd_x_x_x_f32(float %x) #0 {
 ; CHECK-LABEL: fadd_fadd_x_x_x_f32:
 ; CHECK:       # BB#0:
@@ -163,6 +181,10 @@ define float @fadd_fadd_x_x_x_f32(float %x) #0 {
   ret float %z
 }
 
+; CHECK: float 3.000000e+00
+; CHECK: float 3.000000e+00
+; CHECK: float 3.000000e+00
+; CHECK: float 3.000000e+00
 define <4 x float> @fadd_fadd_x_x_x_4f32(<4 x float> %x) #0 {
 ; CHECK-LABEL: fadd_fadd_x_x_x_4f32:
 ; CHECK:       # BB#0:
@@ -173,6 +195,7 @@ define <4 x float> @fadd_fadd_x_x_x_4f32(<4 x float> %x) #0 {
   ret <4 x float> %z
 }
 
+; CHECK: float 4
 define float @fadd_fadd_x_x_fadd_x_x_f32(float %x) #0 {
 ; CHECK-LABEL: fadd_fadd_x_x_fadd_x_x_f32:
 ; CHECK:       # BB#0:
@@ -183,6 +206,10 @@ define float @fadd_fadd_x_x_fadd_x_x_f32(float %x) #0 {
   ret float %z
 }
 
+; CHECK: float 4.000000e+00
+; CHECK: float 4.000000e+00
+; CHECK: float 4.000000e+00
+; CHECK: float 4.000000e+00
 define <4 x float> @fadd_fadd_x_x_fadd_x_x_4f32(<4 x float> %x) #0 {
 ; CHECK-LABEL: fadd_fadd_x_x_fadd_x_x_4f32:
 ; CHECK:       # BB#0:
diff --git a/llvm/test/CodeGen/X86/fmul-combines.ll b/llvm/test/CodeGen/X86/fmul-combines.ll
index 7d75611e133..42e538646ff 100644
--- a/llvm/test/CodeGen/X86/fmul-combines.ll
+++ b/llvm/test/CodeGen/X86/fmul-combines.ll
@@ -86,15 +86,13 @@ define <4 x float> @fmul_v4f32_two_consts_no_splat_non_canonical(<4 x float> %x)
 }
 
 ; More than one use of a constant multiply should not inhibit the optimization.
-; Instead of a chain of 2 dependent mults, this test will have 2 independent mults. 
-; CHECK: float 5.000000e+00
-; CHECK: float 1.200000e+01
-; CHECK: float 2.100000e+01
-; CHECK: float 3.200000e+01
+; Instead of a chain of 2 dependent mults, this test will have 2 independent mults.
+; CHECK: float 6.000000e+00
+; CHECK: float 1.400000e+01
+; CHECK: float 2.400000e+01
+; CHECK: float 3.600000e+01
 ; CHECK-LABEL: fmul_v4f32_two_consts_no_splat_multiple_use:
 ; CHECK: mulps
-; CHECK: mulps
-; CHECK: addps
 ; CHECK: ret
 define <4 x float> @fmul_v4f32_two_consts_no_splat_multiple_use(<4 x float> %x) #0 {
   %y = fmul <4 x float> %x, <float 1.0, float 2.0, float 3.0, float 4.0>