AVX-512: Implemented cvtsi2ss/d cvtusi2ss/d instructions with round control for KNL.

[oota-llvm.git] / test / CodeGen / X86 / avx-splat.ll

diff --git a/test/CodeGen/X86/avx-splat.ll b/test/CodeGen/X86/avx-splat.ll
index 417d7b09c513ae1cff99452239dd07819c6eec40..3ea7e386c426254843b2b891e732b7f072f106b7 100644 (file)
--- a/test/CodeGen/X86/avx-splat.ll
+++ b/test/CodeGen/X86/avx-splat.ll
@@ -1,31 +1,25 @@
 ; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=corei7-avx -mattr=+avx | FileCheck %s
 
-; FIXME: use avx versions for punpcklbw, punpckhbw and punpckhwd
 
-; CHECK: vextractf128 $0
-; CHECK-NEXT: punpcklbw
-; CHECK-NEXT: punpckhbw
+; CHECK: vpshufb {{.*}} ## xmm0 = xmm0[5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5]
 ; CHECK-NEXT: vinsertf128 $1
-; CHECK-NEXT: vpermilps $85
 define <32 x i8> @funcA(<32 x i8> %a) nounwind uwtable readnone ssp {
 entry:
   %shuffle = shufflevector <32 x i8> %a, <32 x i8> undef, <32 x i32> <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
   ret <32 x i8> %shuffle
 }
 
-; CHECK: vextractf128 $0
-; CHECK-NEXT: punpckhwd
+; CHECK: vpshufb {{.*}} ## xmm0 = xmm0[10,11,10,11,10,11,10,11,10,11,10,11,10,11,10,11]
 ; CHECK-NEXT: vinsertf128 $1
-; CHECK-NEXT: vpermilps $85
 define <16 x i16> @funcB(<16 x i16> %a) nounwind uwtable readnone ssp {
 entry:
   %shuffle = shufflevector <16 x i16> %a, <16 x i16> undef, <16 x i32> <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>
   ret <16 x i16> %shuffle
 }
 
-; CHECK: vmovd
+; CHECK: vmovq
+; CHECK-NEXT: vmovddup %xmm
 ; CHECK-NEXT: vinsertf128 $1
-; CHECK-NEXT: vpermilps $0
 define <4 x i64> @funcC(i64 %q) nounwind uwtable readnone ssp {
 entry:
   %vecinit.i = insertelement <4 x i64> undef, i64 %q, i32 0
@@ -35,8 +29,8 @@ entry:
   ret <4 x i64> %vecinit6.i
 }
 
-; CHECK: vinsertf128 $1
-; CHECK-NEXT: vpermilps $0
+; CHECK: vmovddup %xmm
+; CHECK-NEXT: vinsertf128 $1
 define <4 x double> @funcD(double %q) nounwind uwtable readnone ssp {
 entry:
   %vecinit.i = insertelement <4 x double> undef, double %q, i32 0
@@ -46,13 +40,10 @@ entry:
   ret <4 x double> %vecinit6.i
 }
 
-; Test this simple opt:
+; Test this turns into a broadcast:
 ;   shuffle (scalar_to_vector (load (ptr + 4))), undef, <0, 0, 0, 0>
-; To:
-;   shuffle (vload ptr)), undef, <1, 1, 1, 1>
-; CHECK: vmovaps
-; CHECK-NEXT: vinsertf128  $1
-; CHECK-NEXT: vpermilps $-1
+;
+; CHECK: vbroadcastss
 define <8 x float> @funcE() nounwind {
 allocas:
   %udx495 = alloca [18 x [18 x float]], align 32
@@ -65,9 +56,9 @@ for_exit499:                                      ; preds = %for_test505.prehead
   br i1 undef, label %__load_and_broadcast_32.exit1249, label %load.i1247
 
 load.i1247:                                       ; preds = %for_exit499
-  %ptr1227 = getelementptr [18 x [18 x float]]* %udx495, i64 0, i64 1, i64 1
+  %ptr1227 = getelementptr [18 x [18 x float]], [18 x [18 x float]]* %udx495, i64 0, i64 1, i64 1
   %ptr.i1237 = bitcast float* %ptr1227 to i32*
-  %val.i1238 = load i32* %ptr.i1237, align 4
+  %val.i1238 = load i32, i32* %ptr.i1237, align 4
   %ret6.i1245 = insertelement <8 x i32> undef, i32 %val.i1238, i32 6
   %ret7.i1246 = insertelement <8 x i32> %ret6.i1245, i32 %val.i1238, i32 7
   %phitmp = bitcast <8 x i32> %ret7.i1246 to <8 x float>
@@ -78,18 +69,17 @@ __load_and_broadcast_32.exit1249:                 ; preds = %load.i1247, %for_ex
   ret <8 x float> %load_broadcast12281250
 }
 
-; CHECK: vinsertf128 $1
-; CHECK-NEXT: vpermilps $0
-define <8 x float> @funcF(i32* %ptr) nounwind {
-  %val = load i32* %ptr, align 4
+; CHECK: vpermilps $4
+; CHECK-NEXT: vinsertf128 $1
+define <8 x float> @funcF(i32 %val) nounwind {
   %ret6 = insertelement <8 x i32> undef, i32 %val, i32 6
   %ret7 = insertelement <8 x i32> %ret6, i32 %val, i32 7
   %tmp = bitcast <8 x i32> %ret7 to <8 x float>
   ret <8 x float> %tmp
 }
 
-; CHECK: vinsertf128  $1
-; CHECK-NEXT: vpermilps  $0
+; CHECK: vpermilps $0
+; CHECK-NEXT: vinsertf128  $1
 define <8 x float> @funcG(<8 x float> %a) nounwind uwtable readnone ssp {
 entry:
   %shuffle = shufflevector <8 x float> %a, <8 x float> undef, <8 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
@@ -97,8 +87,8 @@ entry:
 }
 
 ; CHECK: vextractf128  $1
+; CHECK-NEXT: vpermilps $85
 ; CHECK-NEXT: vinsertf128  $1
-; CHECK-NEXT: vpermilps  $85
 define <8 x float> @funcH(<8 x float> %a) nounwind uwtable readnone ssp {
 entry:
   %shuffle = shufflevector <8 x float> %a, <8 x float> undef, <8 x i32> <i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5, i32 5>