; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
; RUN: llc -verify-machineinstrs -o - %s -mtriple=aarch64-linux-gnu | FileCheck %s

declare <4 x i16> @llvm.aarch64.neon.vcvtfp2hf(<4 x float>) #2
declare <8 x i8> @llvm.aarch64.neon.tbl1.v8i8(<16 x i8>, <8 x i8>) #2

define <8 x half> @test1(<4 x float> noundef %a) {
; CHECK-LABEL: test1:
; CHECK:       // %bb.0: // %entry
; CHECK-NEXT:    fcvtn v0.4h, v0.4s
; CHECK-NEXT:    ret
entry:
  %vcvt_f16_f321.i = tail call <4 x i16> @llvm.aarch64.neon.vcvtfp2hf(<4 x float> %a)
  %0 = bitcast <4 x i16> %vcvt_f16_f321.i to <4 x half>
  %shuffle.i = shufflevector <4 x half> %0, <4 x half> zeroinitializer, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
  ret <8 x half> %shuffle.i
}

define <8 x i8> @test2(ptr nocapture noundef readonly %in, <8 x i8> noundef %idx) {
; CHECK-LABEL: test2:
; CHECK:       // %bb.0: // %entry
; CHECK-NEXT:    ldr q1, [x0]
; CHECK-NEXT:    shrn v1.8b, v1.8h, #4
; CHECK-NEXT:    tbl v0.8b, { v1.16b }, v0.8b
; CHECK-NEXT:    ret
entry:
  %0 = load <8 x i16>, ptr %in, align 2
  %1 = lshr <8 x i16> %0, <i16 4, i16 4, i16 4, i16 4, i16 4, i16 4, i16 4, i16 4>
  %vshrn_n = trunc <8 x i16> %1 to <8 x i8>
  %vtbl1.i = shufflevector <8 x i8> %vshrn_n, <8 x i8> zeroinitializer, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
  %vtbl11.i = tail call <8 x i8> @llvm.aarch64.neon.tbl1.v8i8(<16 x i8> %vtbl1.i, <8 x i8> %idx)
  ret <8 x i8> %vtbl11.i
}

define <8 x i8> @tbl1v8i8(ptr nocapture noundef readonly %in, <8 x i8> noundef %idx) {
; CHECK-LABEL: tbl1v8i8:
; CHECK:       // %bb.0: // %entry
; CHECK-NEXT:    ldr q1, [x0]
; CHECK-NEXT:    shrn v1.8b, v1.8h, #4
; CHECK-NEXT:    tbl v0.8b, { v1.16b }, v0.8b
; CHECK-NEXT:    ret
entry:
  %0 = load <8 x i16>, ptr %in, align 2
  %1 = lshr <8 x i16> %0, <i16 4, i16 4, i16 4, i16 4, i16 4, i16 4, i16 4, i16 4>
  %vshrn_n = trunc <8 x i16> %1 to <8 x i8>
  %vtbl1.i = shufflevector <8 x i8> %vshrn_n, <8 x i8> zeroinitializer, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
  %vtbl11.i = tail call <8 x i8> @llvm.aarch64.neon.tbl1.v8i8(<16 x i8> %vtbl1.i, <8 x i8> %idx)
  ret <8 x i8> %vtbl11.i
}

define <8 x i16> @addpv4i16(<4 x i16> noundef %a, <4 x i16> noundef %b) {
; CHECK-LABEL: addpv4i16:
; CHECK:       // %bb.0: // %entry
; CHECK-NEXT:    addp v0.4h, v0.4h, v1.4h
; CHECK-NEXT:    ret
entry:
  %vpadd_v2.i = tail call <4 x i16> @llvm.aarch64.neon.addp.v4i16(<4 x i16> %a, <4 x i16> %b)
  %shuffle.i = shufflevector <4 x i16> %vpadd_v2.i, <4 x i16> zeroinitializer, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
  ret <8 x i16> %shuffle.i
}

define <8 x i16> @addv4i16(<4 x i16> noundef %a, <4 x i16> noundef %b) {
; CHECK-LABEL: addv4i16:
; CHECK:       // %bb.0: // %entry
; CHECK-NEXT:    add v0.4h, v1.4h, v0.4h
; CHECK-NEXT:    ret
entry:
  %add.i = add <4 x i16> %b, %a
  %shuffle.i = shufflevector <4 x i16> %add.i, <4 x i16> zeroinitializer, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
  ret <8 x i16> %shuffle.i
}

define <16 x i8> @rshrn(<8 x i16> noundef %a, <4 x i16> noundef %b) {
; CHECK-LABEL: rshrn:
; CHECK:       // %bb.0: // %entry
; CHECK-NEXT:    rshrn v0.8b, v0.8h, #3
; CHECK-NEXT:    ret
entry:
  %vrshrn_n1 = tail call <8 x i8> @llvm.aarch64.neon.rshrn.v8i8(<8 x i16> %a, i32 3)
  %shuffle.i = shufflevector <8 x i8> %vrshrn_n1, <8 x i8> zeroinitializer, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
  ret <16 x i8> %shuffle.i
}

define <16 x i8> @tbl1(<16 x i8> %a, <8 x i8> %b) {
; CHECK-LABEL: tbl1:
; CHECK:       // %bb.0: // %entry
; CHECK-NEXT:    tbl v0.8b, { v0.16b }, v1.8b
; CHECK-NEXT:    ret
entry:
  %vtbl11 = tail call <8 x i8> @llvm.aarch64.neon.tbl1.v8i8(<16 x i8> %a, <8 x i8> %b)
  %shuffle.i = shufflevector <8 x i8> %vtbl11, <8 x i8> zeroinitializer, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
  ret <16 x i8> %shuffle.i
}

define <2 x double> @fadd(double noundef %x, double noundef %y) {
; CHECK-LABEL: fadd:
; CHECK:       // %bb.0: // %entry
; CHECK-NEXT:    movi v2.2d, #0000000000000000
; CHECK-NEXT:    fadd d0, d0, d1
; CHECK-NEXT:    mov v2.d[0], v0.d[0]
; CHECK-NEXT:    mov v0.16b, v2.16b
; CHECK-NEXT:    ret
entry:
  %add = fadd double %x, %y
  %vecinit1 = insertelement <2 x double> poison, double %add, i64 0
  %vecinit2 = insertelement <2 x double> %vecinit1, double 0.0, i64 1
  ret <2 x double> %vecinit2
}

define <16 x i8> @bsl(<4 x i16> noundef %a, <4 x i16> noundef %c, <4 x i16> noundef %d, <4 x i16> noundef %b) {
; CHECK-LABEL: bsl:
; CHECK:       // %bb.0: // %entry
; CHECK-NEXT:    bsl v0.8b, v1.8b, v2.8b
; CHECK-NEXT:    ret
entry:
  %vbsl3.i = and <4 x i16> %c, %a
  %0 = xor <4 x i16> %a, <i16 -1, i16 -1, i16 -1, i16 -1>
  %vbsl4.i = and <4 x i16> %0, %d
  %vbsl5.i = or <4 x i16> %vbsl4.i, %vbsl3.i
  %1 = bitcast <4 x i16> %vbsl5.i to <8 x i8>
  %shuffle.i = shufflevector <8 x i8> %1, <8 x i8> zeroinitializer, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
  ret <16 x i8> %shuffle.i
}

define <16 x i8> @load(ptr %a, <8 x i8> %b) {
; CHECK-LABEL: load:
; CHECK:       // %bb.0: // %entry
; CHECK-NEXT:    ldr d0, [x0]
; CHECK-NEXT:    ret
entry:
  %vtbl11 = load <8 x i8>, ptr %a
  %shuffle.i = shufflevector <8 x i8> %vtbl11, <8 x i8> zeroinitializer, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
  ret <16 x i8> %shuffle.i
}


declare <8 x i8> @llvm.aarch64.neon.rshrn.v8i8(<8 x i16>, i32)
declare <4 x i16> @llvm.aarch64.neon.addp.v4i16(<4 x i16>, <4 x i16>)