-; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
-; RUN: grep vdup.8 %t | count 4
-; RUN: grep vdup.16 %t | count 4
-; RUN: grep vdup.32 %t | count 10
+; RUN: llc < %s -march=arm -mattr=+neon | FileCheck %s
define <8 x i8> @v_dup8(i8 %A) nounwind {
+;CHECK: v_dup8:
+;CHECK: vdup.8
%tmp1 = insertelement <8 x i8> zeroinitializer, i8 %A, i32 0
%tmp2 = insertelement <8 x i8> %tmp1, i8 %A, i32 1
%tmp3 = insertelement <8 x i8> %tmp2, i8 %A, i32 2
}
define <4 x i16> @v_dup16(i16 %A) nounwind {
+;CHECK: v_dup16:
+;CHECK: vdup.16
%tmp1 = insertelement <4 x i16> zeroinitializer, i16 %A, i32 0
%tmp2 = insertelement <4 x i16> %tmp1, i16 %A, i32 1
%tmp3 = insertelement <4 x i16> %tmp2, i16 %A, i32 2
}
define <2 x i32> @v_dup32(i32 %A) nounwind {
+;CHECK: v_dup32:
+;CHECK: vdup.32
%tmp1 = insertelement <2 x i32> zeroinitializer, i32 %A, i32 0
%tmp2 = insertelement <2 x i32> %tmp1, i32 %A, i32 1
ret <2 x i32> %tmp2
}
define <2 x float> @v_dupfloat(float %A) nounwind {
+;CHECK: v_dupfloat:
+;CHECK: vdup.32
%tmp1 = insertelement <2 x float> zeroinitializer, float %A, i32 0
%tmp2 = insertelement <2 x float> %tmp1, float %A, i32 1
ret <2 x float> %tmp2
}
define <16 x i8> @v_dupQ8(i8 %A) nounwind {
+;CHECK: v_dupQ8:
+;CHECK: vdup.8
%tmp1 = insertelement <16 x i8> zeroinitializer, i8 %A, i32 0
%tmp2 = insertelement <16 x i8> %tmp1, i8 %A, i32 1
%tmp3 = insertelement <16 x i8> %tmp2, i8 %A, i32 2
}
define <8 x i16> @v_dupQ16(i16 %A) nounwind {
+;CHECK: v_dupQ16:
+;CHECK: vdup.16
%tmp1 = insertelement <8 x i16> zeroinitializer, i16 %A, i32 0
%tmp2 = insertelement <8 x i16> %tmp1, i16 %A, i32 1
%tmp3 = insertelement <8 x i16> %tmp2, i16 %A, i32 2
}
define <4 x i32> @v_dupQ32(i32 %A) nounwind {
+;CHECK: v_dupQ32:
+;CHECK: vdup.32
%tmp1 = insertelement <4 x i32> zeroinitializer, i32 %A, i32 0
%tmp2 = insertelement <4 x i32> %tmp1, i32 %A, i32 1
%tmp3 = insertelement <4 x i32> %tmp2, i32 %A, i32 2
}
define <4 x float> @v_dupQfloat(float %A) nounwind {
+;CHECK: v_dupQfloat:
+;CHECK: vdup.32
%tmp1 = insertelement <4 x float> zeroinitializer, float %A, i32 0
%tmp2 = insertelement <4 x float> %tmp1, float %A, i32 1
%tmp3 = insertelement <4 x float> %tmp2, float %A, i32 2
; Check to make sure it works with shuffles, too.
define <8 x i8> @v_shuffledup8(i8 %A) nounwind {
+;CHECK: v_shuffledup8:
+;CHECK: vdup.8
%tmp1 = insertelement <8 x i8> undef, i8 %A, i32 0
%tmp2 = shufflevector <8 x i8> %tmp1, <8 x i8> undef, <8 x i32> zeroinitializer
ret <8 x i8> %tmp2
}
define <4 x i16> @v_shuffledup16(i16 %A) nounwind {
+;CHECK: v_shuffledup16:
+;CHECK: vdup.16
%tmp1 = insertelement <4 x i16> undef, i16 %A, i32 0
%tmp2 = shufflevector <4 x i16> %tmp1, <4 x i16> undef, <4 x i32> zeroinitializer
ret <4 x i16> %tmp2
}
define <2 x i32> @v_shuffledup32(i32 %A) nounwind {
+;CHECK: v_shuffledup32:
+;CHECK: vdup.32
%tmp1 = insertelement <2 x i32> undef, i32 %A, i32 0
%tmp2 = shufflevector <2 x i32> %tmp1, <2 x i32> undef, <2 x i32> zeroinitializer
ret <2 x i32> %tmp2
}
define <2 x float> @v_shuffledupfloat(float %A) nounwind {
+;CHECK: v_shuffledupfloat:
+;CHECK: vdup.32
%tmp1 = insertelement <2 x float> undef, float %A, i32 0
%tmp2 = shufflevector <2 x float> %tmp1, <2 x float> undef, <2 x i32> zeroinitializer
ret <2 x float> %tmp2
}
define <16 x i8> @v_shuffledupQ8(i8 %A) nounwind {
+;CHECK: v_shuffledupQ8:
+;CHECK: vdup.8
%tmp1 = insertelement <16 x i8> undef, i8 %A, i32 0
%tmp2 = shufflevector <16 x i8> %tmp1, <16 x i8> undef, <16 x i32> zeroinitializer
ret <16 x i8> %tmp2
}
define <8 x i16> @v_shuffledupQ16(i16 %A) nounwind {
+;CHECK: v_shuffledupQ16:
+;CHECK: vdup.16
%tmp1 = insertelement <8 x i16> undef, i16 %A, i32 0
%tmp2 = shufflevector <8 x i16> %tmp1, <8 x i16> undef, <8 x i32> zeroinitializer
ret <8 x i16> %tmp2
}
define <4 x i32> @v_shuffledupQ32(i32 %A) nounwind {
+;CHECK: v_shuffledupQ32:
+;CHECK: vdup.32
%tmp1 = insertelement <4 x i32> undef, i32 %A, i32 0
%tmp2 = shufflevector <4 x i32> %tmp1, <4 x i32> undef, <4 x i32> zeroinitializer
ret <4 x i32> %tmp2
}
define <4 x float> @v_shuffledupQfloat(float %A) nounwind {
+;CHECK: v_shuffledupQfloat:
+;CHECK: vdup.32
%tmp1 = insertelement <4 x float> undef, float %A, i32 0
%tmp2 = shufflevector <4 x float> %tmp1, <4 x float> undef, <4 x i32> zeroinitializer
ret <4 x float> %tmp2
}
-define <2 x float> @v_shuffledupfloat2(float* %A) nounwind {
- %tmp0 = load float* %A
- %tmp1 = insertelement <2 x float> undef, float %tmp0, i32 0
- %tmp2 = shufflevector <2 x float> %tmp1, <2 x float> undef, <2 x i32> zeroinitializer
- ret <2 x float> %tmp2
+define <8 x i8> @vduplane8(<8 x i8>* %A) nounwind {
+;CHECK: vduplane8:
+;CHECK: vdup.8
+ %tmp1 = load <8 x i8>* %A
+ %tmp2 = shufflevector <8 x i8> %tmp1, <8 x i8> undef, <8 x i32> < i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1 >
+ ret <8 x i8> %tmp2
+}
+
+define <4 x i16> @vduplane16(<4 x i16>* %A) nounwind {
+;CHECK: vduplane16:
+;CHECK: vdup.16
+ %tmp1 = load <4 x i16>* %A
+ %tmp2 = shufflevector <4 x i16> %tmp1, <4 x i16> undef, <4 x i32> < i32 1, i32 1, i32 1, i32 1 >
+ ret <4 x i16> %tmp2
+}
+
+define <2 x i32> @vduplane32(<2 x i32>* %A) nounwind {
+;CHECK: vduplane32:
+;CHECK: vdup.32
+ %tmp1 = load <2 x i32>* %A
+ %tmp2 = shufflevector <2 x i32> %tmp1, <2 x i32> undef, <2 x i32> < i32 1, i32 1 >
+ ret <2 x i32> %tmp2
+}
+
+define <2 x float> @vduplanefloat(<2 x float>* %A) nounwind {
+;CHECK: vduplanefloat:
+;CHECK: vdup.32
+ %tmp1 = load <2 x float>* %A
+ %tmp2 = shufflevector <2 x float> %tmp1, <2 x float> undef, <2 x i32> < i32 1, i32 1 >
+ ret <2 x float> %tmp2
+}
+
+define <16 x i8> @vduplaneQ8(<8 x i8>* %A) nounwind {
+;CHECK: vduplaneQ8:
+;CHECK: vdup.8
+ %tmp1 = load <8 x i8>* %A
+ %tmp2 = shufflevector <8 x i8> %tmp1, <8 x i8> undef, <16 x i32> < i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1 >
+ ret <16 x i8> %tmp2
+}
+
+define <8 x i16> @vduplaneQ16(<4 x i16>* %A) nounwind {
+;CHECK: vduplaneQ16:
+;CHECK: vdup.16
+ %tmp1 = load <4 x i16>* %A
+ %tmp2 = shufflevector <4 x i16> %tmp1, <4 x i16> undef, <8 x i32> < i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1 >
+ ret <8 x i16> %tmp2
+}
+
+define <4 x i32> @vduplaneQ32(<2 x i32>* %A) nounwind {
+;CHECK: vduplaneQ32:
+;CHECK: vdup.32
+ %tmp1 = load <2 x i32>* %A
+ %tmp2 = shufflevector <2 x i32> %tmp1, <2 x i32> undef, <4 x i32> < i32 1, i32 1, i32 1, i32 1 >
+ ret <4 x i32> %tmp2
+}
+
+define <4 x float> @vduplaneQfloat(<2 x float>* %A) nounwind {
+;CHECK: vduplaneQfloat:
+;CHECK: vdup.32
+ %tmp1 = load <2 x float>* %A
+ %tmp2 = shufflevector <2 x float> %tmp1, <2 x float> undef, <4 x i32> < i32 1, i32 1, i32 1, i32 1 >
+ ret <4 x float> %tmp2
+}
+
+define <2 x i64> @foo(<2 x i64> %arg0_int64x1_t) nounwind readnone {
+entry:
+ %0 = shufflevector <2 x i64> %arg0_int64x1_t, <2 x i64> undef, <2 x i32> <i32 1, i32 1>
+ ret <2 x i64> %0
+}
+
+define <2 x i64> @bar(<2 x i64> %arg0_int64x1_t) nounwind readnone {
+entry:
+ %0 = shufflevector <2 x i64> %arg0_int64x1_t, <2 x i64> undef, <2 x i32> <i32 0, i32 0>
+ ret <2 x i64> %0
+}
+
+define <2 x double> @baz(<2 x double> %arg0_int64x1_t) nounwind readnone {
+entry:
+ %0 = shufflevector <2 x double> %arg0_int64x1_t, <2 x double> undef, <2 x i32> <i32 1, i32 1>
+ ret <2 x double> %0
+}
+
+define <2 x double> @qux(<2 x double> %arg0_int64x1_t) nounwind readnone {
+entry:
+ %0 = shufflevector <2 x double> %arg0_int64x1_t, <2 x double> undef, <2 x i32> <i32 0, i32 0>
+ ret <2 x double> %0
}
-define <4 x float> @v_shuffledupQfloat2(float* %A) nounwind {
- %tmp0 = load float* %A
- %tmp1 = insertelement <4 x float> undef, float %tmp0, i32 0
- %tmp2 = shufflevector <4 x float> %tmp1, <4 x float> undef, <4 x i32> zeroinitializer
- ret <4 x float> %tmp2
+; Radar 7373643
+;CHECK: redundantVdup:
+;CHECK: vmov.i8
+;CHECK-NOT: vdup.8
+;CHECK: vstr.64
+define void @redundantVdup(<8 x i8>* %ptr) nounwind {
+ %1 = insertelement <8 x i8> undef, i8 -128, i32 0
+ %2 = shufflevector <8 x i8> %1, <8 x i8> undef, <8 x i32> zeroinitializer
+ store <8 x i8> %2, <8 x i8>* %ptr, align 8
+ ret void
}
projects / oota-llvm.git / blobdiff