CodeGen peephole: fold redundant phys reg copies

[oota-llvm.git] / test / CodeGen / X86 / vec_insert-5.ll

diff --git a/test/CodeGen/X86/vec_insert-5.ll b/test/CodeGen/X86/vec_insert-5.ll
index b72044aee30b35cca8b814797313849fa73e6a6b..14b57e76dc8faf9fc1bfa469ce796cc992194030 100644 (file)
--- a/test/CodeGen/X86/vec_insert-5.ll
+++ b/test/CodeGen/X86/vec_insert-5.ll
@@ -1,3 +1,4 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -march=x86 -mattr=+sse2,+ssse3 | FileCheck %s
 ; There are no MMX operations in @t1
 
@@ -8,8 +9,8 @@ define void  @t1(i32 %a, x86_mmx* %P) nounwind {
 ; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; CHECK-NEXT:    shll $12, %ecx
 ; CHECK-NEXT:    movd %ecx, %xmm0
-; CHECK-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,0,0,1]
-; CHECK-NEXT:    movlpd %xmm0, (%eax)
+; CHECK-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,0,1,1]
+; CHECK-NEXT:    movq %xmm0, (%eax)
 ; CHECK-NEXT:    retl
  %tmp12 = shl i32 %a, 12
  %tmp21 = insertelement <2 x i32> undef, i32 %tmp12, i32 1
@@ -25,10 +26,10 @@ define <4 x float> @t2(<4 x float>* %P) nounwind {
 ; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; CHECK-NEXT:    movaps (%eax), %xmm1
 ; CHECK-NEXT:    xorps %xmm0, %xmm0
-; CHECK-NEXT:    shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[0,0]
-; CHECK-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,0],xmm1[2,0]
+; CHECK-NEXT:    shufps {{.*#+}} xmm1 = xmm1[0,0],xmm0[2,0]
+; CHECK-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,0]
 ; CHECK-NEXT:    retl
-  %tmp1 = load <4 x float>* %P
+  %tmp1 = load <4 x float>, <4 x float>* %P
   %tmp2 = shufflevector <4 x float> %tmp1, <4 x float> zeroinitializer, <4 x i32> < i32 4, i32 4, i32 4, i32 0 >
   ret <4 x float> %tmp2
 }
@@ -37,11 +38,11 @@ define <4 x float> @t3(<4 x float>* %P) nounwind {
 ; CHECK-LABEL: t3:
 ; CHECK:       # BB#0:
 ; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
-; CHECK-NEXT:    movaps (%eax), %xmm0
-; CHECK-NEXT:    xorps %xmm1, %xmm1
-; CHECK-NEXT:    shufps {{.*#+}} xmm0 = xmm0[2,3],xmm1[0,0]
+; CHECK-NEXT:    movapd (%eax), %xmm0
+; CHECK-NEXT:    xorpd %xmm1, %xmm1
+; CHECK-NEXT:    unpckhpd {{.*#+}} xmm0 = xmm0[1],xmm1[1]
 ; CHECK-NEXT:    retl
-  %tmp1 = load <4 x float>* %P
+  %tmp1 = load <4 x float>, <4 x float>* %P
   %tmp2 = shufflevector <4 x float> %tmp1, <4 x float> zeroinitializer, <4 x i32> < i32 2, i32 3, i32 4, i32 4 >
   ret <4 x float> %tmp2
 }
@@ -52,10 +53,10 @@ define <4 x float> @t4(<4 x float>* %P) nounwind {
 ; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; CHECK-NEXT:    movaps (%eax), %xmm0
 ; CHECK-NEXT:    xorps %xmm1, %xmm1
-; CHECK-NEXT:    shufps {{.*#+}} xmm0 = xmm0[3,0],xmm1[0,0]
-; CHECK-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,2],xmm1[0,0]
+; CHECK-NEXT:    shufps {{.*#+}} xmm0 = xmm0[3,0],xmm1[1,0]
+; CHECK-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,2],xmm1[2,3]
 ; CHECK-NEXT:    retl
-  %tmp1 = load <4 x float>* %P
+  %tmp1 = load <4 x float>, <4 x float>* %P
   %tmp2 = shufflevector <4 x float> zeroinitializer, <4 x float> %tmp1, <4 x i32> < i32 7, i32 0, i32 0, i32 0 >
   ret <4 x float> %tmp2
 }
@@ -63,7 +64,7 @@ define <4 x float> @t4(<4 x float>* %P) nounwind {
 define <16 x i8> @t5(<16 x i8> %x) nounwind {
 ; CHECK-LABEL: t5:
 ; CHECK:       # BB#0:
-; CHECK-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15],zero
+; CHECK-NEXT:    psrlw $8, %xmm0
 ; CHECK-NEXT:    retl
   %s = shufflevector <16 x i8> %x, <16 x i8> zeroinitializer, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 17>
   ret <16 x i8> %s
@@ -72,7 +73,7 @@ define <16 x i8> @t5(<16 x i8> %x) nounwind {
 define <16 x i8> @t6(<16 x i8> %x) nounwind {
 ; CHECK-LABEL: t6:
 ; CHECK:       # BB#0:
-; CHECK-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15],zero
+; CHECK-NEXT:    psrlw $8, %xmm0
 ; CHECK-NEXT:    retl
   %s = shufflevector <16 x i8> %x, <16 x i8> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
   ret <16 x i8> %s
@@ -86,3 +87,21 @@ define <16 x i8> @t7(<16 x i8> %x) nounwind {
   %s = shufflevector <16 x i8> %x, <16 x i8> undef, <16 x i32> <i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 1, i32 2>
   ret <16 x i8> %s
 }
+
+define <16 x i8> @t8(<16 x i8> %x) nounwind {
+; CHECK-LABEL: t8:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15],zero
+; CHECK-NEXT:    retl
+  %s = shufflevector <16 x i8> %x, <16 x i8> zeroinitializer, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 8, i32 9, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 17>
+  ret <16 x i8> %s
+}
+
+define <16 x i8> @t9(<16 x i8> %x) nounwind {
+; CHECK-LABEL: t9:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15],zero
+; CHECK-NEXT:    retl
+  %s = shufflevector <16 x i8> %x, <16 x i8> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 7, i32 8, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 14, i32 undef, i32 undef>
+  ret <16 x i8> %s
+}