; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve.fp -verify-machineinstrs -o - %s | FileCheck %s

define arm_aapcs_vfpcc <16 x i8> @test_vqmovnbq_s16(<16 x i8> %a, <8 x i16> %b) {
; CHECK-LABEL: test_vqmovnbq_s16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovnb.s16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <16 x i8> @llvm.arm.mve.vqmovn.v16i8.v8i16(<16 x i8> %a, <8 x i16> %b, i32 0, i32 0, i32 0)
  ret <16 x i8> %0
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovnbq_s32(<8 x i16> %a, <4 x i32> %b) {
; CHECK-LABEL: test_vqmovnbq_s32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovnb.s32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <8 x i16> @llvm.arm.mve.vqmovn.v8i16.v4i32(<8 x i16> %a, <4 x i32> %b, i32 0, i32 0, i32 0)
  ret <8 x i16> %0
}

define arm_aapcs_vfpcc <16 x i8> @test_vqmovnbq_u16(<16 x i8> %a, <8 x i16> %b) {
; CHECK-LABEL: test_vqmovnbq_u16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovnb.u16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <16 x i8> @llvm.arm.mve.vqmovn.v16i8.v8i16(<16 x i8> %a, <8 x i16> %b, i32 1, i32 1, i32 0)
  ret <16 x i8> %0
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovnbq_u32(<8 x i16> %a, <4 x i32> %b) {
; CHECK-LABEL: test_vqmovnbq_u32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovnb.u32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <8 x i16> @llvm.arm.mve.vqmovn.v8i16.v4i32(<8 x i16> %a, <4 x i32> %b, i32 1, i32 1, i32 0)
  ret <8 x i16> %0
}

define arm_aapcs_vfpcc <16 x i8> @test_vqmovntq_s16(<16 x i8> %a, <8 x i16> %b) {
; CHECK-LABEL: test_vqmovntq_s16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovnt.s16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <16 x i8> @llvm.arm.mve.vqmovn.v16i8.v8i16(<16 x i8> %a, <8 x i16> %b, i32 0, i32 0, i32 1)
  ret <16 x i8> %0
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovntq_s32(<8 x i16> %a, <4 x i32> %b) {
; CHECK-LABEL: test_vqmovntq_s32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovnt.s32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <8 x i16> @llvm.arm.mve.vqmovn.v8i16.v4i32(<8 x i16> %a, <4 x i32> %b, i32 0, i32 0, i32 1)
  ret <8 x i16> %0
}

define arm_aapcs_vfpcc <16 x i8> @test_vqmovntq_u16(<16 x i8> %a, <8 x i16> %b) {
; CHECK-LABEL: test_vqmovntq_u16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovnt.u16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <16 x i8> @llvm.arm.mve.vqmovn.v16i8.v8i16(<16 x i8> %a, <8 x i16> %b, i32 1, i32 1, i32 1)
  ret <16 x i8> %0
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovntq_u32(<8 x i16> %a, <4 x i32> %b) {
; CHECK-LABEL: test_vqmovntq_u32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovnt.u32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <8 x i16> @llvm.arm.mve.vqmovn.v8i16.v4i32(<8 x i16> %a, <4 x i32> %b, i32 1, i32 1, i32 1)
  ret <8 x i16> %0
}

define arm_aapcs_vfpcc <16 x i8> @test_vqmovunbq_s16(<16 x i8> %a, <8 x i16> %b) {
; CHECK-LABEL: test_vqmovunbq_s16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovunb.s16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <16 x i8> @llvm.arm.mve.vqmovn.v16i8.v8i16(<16 x i8> %a, <8 x i16> %b, i32 1, i32 0, i32 0)
  ret <16 x i8> %0
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovunbq_s32(<8 x i16> %a, <4 x i32> %b) {
; CHECK-LABEL: test_vqmovunbq_s32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovunb.s32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <8 x i16> @llvm.arm.mve.vqmovn.v8i16.v4i32(<8 x i16> %a, <4 x i32> %b, i32 1, i32 0, i32 0)
  ret <8 x i16> %0
}

define arm_aapcs_vfpcc <16 x i8> @test_vqmovuntq_s16(<16 x i8> %a, <8 x i16> %b) {
; CHECK-LABEL: test_vqmovuntq_s16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovunt.s16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <16 x i8> @llvm.arm.mve.vqmovn.v16i8.v8i16(<16 x i8> %a, <8 x i16> %b, i32 1, i32 0, i32 1)
  ret <16 x i8> %0
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovuntq_s32(<8 x i16> %a, <4 x i32> %b) {
; CHECK-LABEL: test_vqmovuntq_s32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vqmovunt.s32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = tail call <8 x i16> @llvm.arm.mve.vqmovn.v8i16.v4i32(<8 x i16> %a, <4 x i32> %b, i32 1, i32 0, i32 1)
  ret <8 x i16> %0
}

define arm_aapcs_vfpcc <16 x i8> @test_vqmovnbq_m_s16(<16 x i8> %a, <8 x i16> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovnbq_m_s16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovnbt.s16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
  %2 = tail call <16 x i8> @llvm.arm.mve.vqmovn.predicated.v16i8.v8i16.v8i1(<16 x i8> %a, <8 x i16> %b, i32 0, i32 0, i32 0, <8 x i1> %1)
  ret <16 x i8> %2
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovnbq_m_s32(<8 x i16> %a, <4 x i32> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovnbq_m_s32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovnbt.s32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
  %2 = tail call <8 x i16> @llvm.arm.mve.vqmovn.predicated.v8i16.v4i32.v4i1(<8 x i16> %a, <4 x i32> %b, i32 0, i32 0, i32 0, <4 x i1> %1)
  ret <8 x i16> %2
}

define arm_aapcs_vfpcc <16 x i8> @test_vqmovnbq_m_u16(<16 x i8> %a, <8 x i16> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovnbq_m_u16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovnbt.u16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
  %2 = tail call <16 x i8> @llvm.arm.mve.vqmovn.predicated.v16i8.v8i16.v8i1(<16 x i8> %a, <8 x i16> %b, i32 1, i32 1, i32 0, <8 x i1> %1)
  ret <16 x i8> %2
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovnbq_m_u32(<8 x i16> %a, <4 x i32> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovnbq_m_u32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovnbt.u32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
  %2 = tail call <8 x i16> @llvm.arm.mve.vqmovn.predicated.v8i16.v4i32.v4i1(<8 x i16> %a, <4 x i32> %b, i32 1, i32 1, i32 0, <4 x i1> %1)
  ret <8 x i16> %2
}

define arm_aapcs_vfpcc <16 x i8> @test_vqmovntq_m_s16(<16 x i8> %a, <8 x i16> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovntq_m_s16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovntt.s16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
  %2 = tail call <16 x i8> @llvm.arm.mve.vqmovn.predicated.v16i8.v8i16.v8i1(<16 x i8> %a, <8 x i16> %b, i32 0, i32 0, i32 1, <8 x i1> %1)
  ret <16 x i8> %2
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovntq_m_s32(<8 x i16> %a, <4 x i32> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovntq_m_s32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovntt.s32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
  %2 = tail call <8 x i16> @llvm.arm.mve.vqmovn.predicated.v8i16.v4i32.v4i1(<8 x i16> %a, <4 x i32> %b, i32 0, i32 0, i32 1, <4 x i1> %1)
  ret <8 x i16> %2
}

define arm_aapcs_vfpcc <16 x i8> @test_vqmovntq_m_u16(<16 x i8> %a, <8 x i16> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovntq_m_u16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovntt.u16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
  %2 = tail call <16 x i8> @llvm.arm.mve.vqmovn.predicated.v16i8.v8i16.v8i1(<16 x i8> %a, <8 x i16> %b, i32 1, i32 1, i32 1, <8 x i1> %1)
  ret <16 x i8> %2
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovntq_m_u32(<8 x i16> %a, <4 x i32> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovntq_m_u32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovntt.u32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
  %2 = tail call <8 x i16> @llvm.arm.mve.vqmovn.predicated.v8i16.v4i32.v4i1(<8 x i16> %a, <4 x i32> %b, i32 1, i32 1, i32 1, <4 x i1> %1)
  ret <8 x i16> %2
}

define arm_aapcs_vfpcc <16 x i8> @test_vqmovunbq_m_s16(<16 x i8> %a, <8 x i16> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovunbq_m_s16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovunbt.s16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
  %2 = tail call <16 x i8> @llvm.arm.mve.vqmovn.predicated.v16i8.v8i16.v8i1(<16 x i8> %a, <8 x i16> %b, i32 1, i32 0, i32 0, <8 x i1> %1)
  ret <16 x i8> %2
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovunbq_m_s32(<8 x i16> %a, <4 x i32> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovunbq_m_s32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovunbt.s32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
  %2 = tail call <8 x i16> @llvm.arm.mve.vqmovn.predicated.v8i16.v4i32.v4i1(<8 x i16> %a, <4 x i32> %b, i32 1, i32 0, i32 0, <4 x i1> %1)
  ret <8 x i16> %2
}

define arm_aapcs_vfpcc <16 x i8> @test_vqmovuntq_m_s16(<16 x i8> %a, <8 x i16> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovuntq_m_s16:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovuntt.s16 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 %0)
  %2 = tail call <16 x i8> @llvm.arm.mve.vqmovn.predicated.v16i8.v8i16.v8i1(<16 x i8> %a, <8 x i16> %b, i32 1, i32 0, i32 1, <8 x i1> %1)
  ret <16 x i8> %2
}

define arm_aapcs_vfpcc <8 x i16> @test_vqmovuntq_m_s32(<8 x i16> %a, <4 x i32> %b, i16 zeroext %p) {
; CHECK-LABEL: test_vqmovuntq_m_s32:
; CHECK:       @ %bb.0: @ %entry
; CHECK-NEXT:    vmsr p0, r0
; CHECK-NEXT:    vpst
; CHECK-NEXT:    vqmovuntt.s32 q0, q1
; CHECK-NEXT:    bx lr
entry:
  %0 = zext i16 %p to i32
  %1 = tail call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 %0)
  %2 = tail call <8 x i16> @llvm.arm.mve.vqmovn.predicated.v8i16.v4i32.v4i1(<8 x i16> %a, <4 x i32> %b, i32 1, i32 0, i32 1, <4 x i1> %1)
  ret <8 x i16> %2
}

declare <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32)
declare <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32)

declare <16 x i8> @llvm.arm.mve.vqmovn.v16i8.v8i16(<16 x i8>, <8 x i16>, i32, i32, i32)
declare <8 x i16> @llvm.arm.mve.vqmovn.v8i16.v4i32(<8 x i16>, <4 x i32>, i32, i32, i32)

declare <16 x i8> @llvm.arm.mve.vqmovn.predicated.v16i8.v8i16.v8i1(<16 x i8>, <8 x i16>, i32, i32, i32, <8 x i1>)
declare <8 x i16> @llvm.arm.mve.vqmovn.predicated.v8i16.v4i32.v4i1(<8 x i16>, <4 x i32>, i32, i32, i32, <4 x i1>)