; 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: vpunpcklbw %xmm
+; CHECK-NEXT: vpunpckhbw %xmm
+; CHECK-NEXT: vpshufd $85
; 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: vpunpckhwd %xmm
+; CHECK-NEXT: vpshufd $85
; 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: vmovlhps %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
ret <4 x i64> %vecinit6.i
}
-; CHECK: vinsertf128 $1
-; CHECK-NEXT: vpermilps $0
+; CHECK: vpermilpd $0
+; 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
; shuffle (scalar_to_vector (load (ptr + 4))), undef, <0, 0, 0, 0>
; To:
; shuffle (vload ptr)), undef, <1, 1, 1, 1>
-; CHECK: vmovaps
+; CHECK: vmovdqa
+; CHECK-NEXT: vpshufd $-1
; CHECK-NEXT: vinsertf128 $1
-; CHECK-NEXT: vpermilps $-1
define <8 x float> @funcE() nounwind {
allocas:
%udx495 = alloca [18 x [18 x float]], align 32
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: vpshufd $0
+; 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: vpshufd $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>
}
; CHECK: vextractf128 $1
+; CHECK-NEXT: vpshufd
; 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>