; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
; RUN: opt < %s -passes="print<cost-model>" 2>&1 -disable-output -S -mtriple=riscv64 -mattr=+v,+f,+d,+zvfh,+zvfbfmin | FileCheck %s --check-prefixes=CHECK,ZVFH
; RUN: opt < %s -passes="print<cost-model>" 2>&1 -disable-output -S -mtriple=riscv64 -mattr=+v,+f,+d,+zvfhmin,+zvfbfmin | FileCheck %s --check-prefixes=CHECK,ZVFHMIN

define void @floor() {
; CHECK-LABEL: 'floor'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.floor.bf16(bfloat undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.floor.v2bf16(<2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.floor.v4bf16(<4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.floor.v8bf16(<8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.floor.v16bf16(<16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.floor.nxv1bf16(<vscale x 1 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.floor.nxv2bf16(<vscale x 2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.floor.nxv4bf16(<vscale x 4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.floor.nxv8bf16(<vscale x 8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.floor.nxv16bf16(<vscale x 16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.floor.f32(float undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x float> @llvm.floor.v2f32(<2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x float> @llvm.floor.v4f32(<4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x float> @llvm.floor.v8f32(<8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x float> @llvm.floor.v16f32(<16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x float> @llvm.floor.nxv1f32(<vscale x 1 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x float> @llvm.floor.nxv2f32(<vscale x 2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x float> @llvm.floor.nxv4f32(<vscale x 4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x float> @llvm.floor.nxv8f32(<vscale x 8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %20 = call <vscale x 16 x float> @llvm.floor.nxv16f32(<vscale x 16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.floor.f64(double undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %22 = call <2 x double> @llvm.floor.v2f64(<2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %23 = call <4 x double> @llvm.floor.v4f64(<4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %24 = call <8 x double> @llvm.floor.v8f64(<8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %25 = call <16 x double> @llvm.floor.v16f64(<16 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 1 x double> @llvm.floor.nxv1f64(<vscale x 1 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %27 = call <vscale x 2 x double> @llvm.floor.nxv2f64(<vscale x 2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %28 = call <vscale x 4 x double> @llvm.floor.nxv4f64(<vscale x 4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %29 = call <vscale x 8 x double> @llvm.floor.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call bfloat @llvm.floor.bf16(bfloat undef)
  call <2 x bfloat> @llvm.floor.v2bf16(<2 x bfloat> undef)
  call <4 x bfloat> @llvm.floor.v4bf16(<4 x bfloat> undef)
  call <8 x bfloat> @llvm.floor.v8bf16(<8 x bfloat> undef)
  call <16 x bfloat> @llvm.floor.v16bf16(<16 x bfloat> undef)
  call <vscale x 1 x bfloat> @llvm.floor.nxv1bf16(<vscale x 1 x bfloat> undef)
  call <vscale x 2 x bfloat> @llvm.floor.nxv2bf16(<vscale x 2 x bfloat> undef)
  call <vscale x 4 x bfloat> @llvm.floor.nxv4bf16(<vscale x 4 x bfloat> undef)
  call <vscale x 8 x bfloat> @llvm.floor.nxv8bf16(<vscale x 8 x bfloat> undef)
  call <vscale x 16 x bfloat> @llvm.floor.nxv16bf16(<vscale x 16 x bfloat> undef)
  call float @llvm.floor.f32(float undef)
  call <2 x float> @llvm.floor.v2f32(<2 x float> undef)
  call <4 x float> @llvm.floor.v4f32(<4 x float> undef)
  call <8 x float> @llvm.floor.v8f32(<8 x float> undef)
  call <16 x float> @llvm.floor.v16f32(<16 x float> undef)
  call <vscale x 1 x float> @llvm.floor.nxv1f32(<vscale x 1 x float> undef)
  call <vscale x 2 x float> @llvm.floor.nxv2f32(<vscale x 2 x float> undef)
  call <vscale x 4 x float> @llvm.floor.nxv4f32(<vscale x 4 x float> undef)
  call <vscale x 8 x float> @llvm.floor.nxv8f32(<vscale x 8 x float> undef)
  call <vscale x 16 x float> @llvm.floor.nxv16f32(<vscale x 16 x float> undef)
  call double @llvm.floor.f64(double undef)
  call <2 x double> @llvm.floor.v2f64(<2 x double> undef)
  call <4 x double> @llvm.floor.v4f64(<4 x double> undef)
  call <8 x double> @llvm.floor.v8f64(<8 x double> undef)
  call <16 x double> @llvm.floor.v16f64(<16 x double> undef)
  call <vscale x 1 x double> @llvm.floor.nxv1f64(<vscale x 1 x double> undef)
  call <vscale x 2 x double> @llvm.floor.nxv2f64(<vscale x 2 x double> undef)
  call <vscale x 4 x double> @llvm.floor.nxv4f64(<vscale x 4 x double> undef)
  call <vscale x 8 x double> @llvm.floor.nxv8f64(<vscale x 8 x double> undef)
  ret void
}

define void @floor_fp16() {
; ZVFH-LABEL: 'floor_fp16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.floor.f16(half undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <2 x half> @llvm.floor.v2f16(<2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <4 x half> @llvm.floor.v4f16(<4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <8 x half> @llvm.floor.v8f16(<8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x half> @llvm.floor.v16f16(<16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 1 x half> @llvm.floor.nxv1f16(<vscale x 1 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 2 x half> @llvm.floor.nxv2f16(<vscale x 2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 4 x half> @llvm.floor.nxv4f16(<vscale x 4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 8 x half> @llvm.floor.nxv8f16(<vscale x 8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %10 = call <vscale x 16 x half> @llvm.floor.nxv16f16(<vscale x 16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'floor_fp16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call half @llvm.floor.f16(half undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x half> @llvm.floor.v2f16(<2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x half> @llvm.floor.v4f16(<4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x half> @llvm.floor.v8f16(<8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x half> @llvm.floor.v16f16(<16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x half> @llvm.floor.nxv1f16(<vscale x 1 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x half> @llvm.floor.nxv2f16(<vscale x 2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x half> @llvm.floor.nxv4f16(<vscale x 4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x half> @llvm.floor.nxv8f16(<vscale x 8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x half> @llvm.floor.nxv16f16(<vscale x 16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call half @llvm.floor.f16(half undef)
  call <2 x half> @llvm.floor.v2f16(<2 x half> undef)
  call <4 x half> @llvm.floor.v4f16(<4 x half> undef)
  call <8 x half> @llvm.floor.v8f16(<8 x half> undef)
  call <16 x half> @llvm.floor.v16f16(<16 x half> undef)
  call <vscale x 1 x half> @llvm.floor.nxv1f16(<vscale x 1 x half> undef)
  call <vscale x 2 x half> @llvm.floor.nxv2f16(<vscale x 2 x half> undef)
  call <vscale x 4 x half> @llvm.floor.nxv4f16(<vscale x 4 x half> undef)
  call <vscale x 8 x half> @llvm.floor.nxv8f16(<vscale x 8 x half> undef)
  call <vscale x 16 x half> @llvm.floor.nxv16f16(<vscale x 16 x half> undef)
  ret void
}

define void @ceil() {
; CHECK-LABEL: 'ceil'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.ceil.bf16(bfloat undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.ceil.v2bf16(<2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.ceil.v4bf16(<4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.ceil.v8bf16(<8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.ceil.v16bf16(<16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.ceil.nxv1bf16(<vscale x 1 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.ceil.nxv2bf16(<vscale x 2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.ceil.nxv4bf16(<vscale x 4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.ceil.nxv8bf16(<vscale x 8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.ceil.nxv16bf16(<vscale x 16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.ceil.f32(float undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x float> @llvm.ceil.v2f32(<2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x float> @llvm.ceil.v4f32(<4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x float> @llvm.ceil.v8f32(<8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x float> @llvm.ceil.v16f32(<16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x float> @llvm.ceil.nxv1f32(<vscale x 1 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x float> @llvm.ceil.nxv2f32(<vscale x 2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x float> @llvm.ceil.nxv4f32(<vscale x 4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x float> @llvm.ceil.nxv8f32(<vscale x 8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %20 = call <vscale x 16 x float> @llvm.ceil.nxv16f32(<vscale x 16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.ceil.f64(double undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %22 = call <2 x double> @llvm.ceil.v2f64(<2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %23 = call <4 x double> @llvm.ceil.v4f64(<4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %24 = call <8 x double> @llvm.ceil.v8f64(<8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %25 = call <16 x double> @llvm.ceil.v16f64(<16 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 1 x double> @llvm.ceil.nxv1f64(<vscale x 1 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %27 = call <vscale x 2 x double> @llvm.ceil.nxv2f64(<vscale x 2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %28 = call <vscale x 4 x double> @llvm.ceil.nxv4f64(<vscale x 4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %29 = call <vscale x 8 x double> @llvm.ceil.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call bfloat @llvm.ceil.bf16(bfloat undef)
  call <2 x bfloat> @llvm.ceil.v2bf16(<2 x bfloat> undef)
  call <4 x bfloat> @llvm.ceil.v4bf16(<4 x bfloat> undef)
  call <8 x bfloat> @llvm.ceil.v8bf16(<8 x bfloat> undef)
  call <16 x bfloat> @llvm.ceil.v16bf16(<16 x bfloat> undef)
  call <vscale x 1 x bfloat> @llvm.ceil.nxv1bf16(<vscale x 1 x bfloat> undef)
  call <vscale x 2 x bfloat> @llvm.ceil.nxv2bf16(<vscale x 2 x bfloat> undef)
  call <vscale x 4 x bfloat> @llvm.ceil.nxv4bf16(<vscale x 4 x bfloat> undef)
  call <vscale x 8 x bfloat> @llvm.ceil.nxv8bf16(<vscale x 8 x bfloat> undef)
  call <vscale x 16 x bfloat> @llvm.ceil.nxv16bf16(<vscale x 16 x bfloat> undef)
  call float @llvm.ceil.f32(float undef)
  call <2 x float> @llvm.ceil.v2f32(<2 x float> undef)
  call <4 x float> @llvm.ceil.v4f32(<4 x float> undef)
  call <8 x float> @llvm.ceil.v8f32(<8 x float> undef)
  call <16 x float> @llvm.ceil.v16f32(<16 x float> undef)
  call <vscale x 1 x float> @llvm.ceil.nxv1f32(<vscale x 1 x float> undef)
  call <vscale x 2 x float> @llvm.ceil.nxv2f32(<vscale x 2 x float> undef)
  call <vscale x 4 x float> @llvm.ceil.nxv4f32(<vscale x 4 x float> undef)
  call <vscale x 8 x float> @llvm.ceil.nxv8f32(<vscale x 8 x float> undef)
  call <vscale x 16 x float> @llvm.ceil.nxv16f32(<vscale x 16 x float> undef)
  call double @llvm.ceil.f64(double undef)
  call <2 x double> @llvm.ceil.v2f64(<2 x double> undef)
  call <4 x double> @llvm.ceil.v4f64(<4 x double> undef)
  call <8 x double> @llvm.ceil.v8f64(<8 x double> undef)
  call <16 x double> @llvm.ceil.v16f64(<16 x double> undef)
  call <vscale x 1 x double> @llvm.ceil.nxv1f64(<vscale x 1 x double> undef)
  call <vscale x 2 x double> @llvm.ceil.nxv2f64(<vscale x 2 x double> undef)
  call <vscale x 4 x double> @llvm.ceil.nxv4f64(<vscale x 4 x double> undef)
  call <vscale x 8 x double> @llvm.ceil.nxv8f64(<vscale x 8 x double> undef)
  ret void
}

define void @ceil_fp16() {
; ZVFH-LABEL: 'ceil_fp16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.ceil.f16(half undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <2 x half> @llvm.ceil.v2f16(<2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <4 x half> @llvm.ceil.v4f16(<4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <8 x half> @llvm.ceil.v8f16(<8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x half> @llvm.ceil.v16f16(<16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 1 x half> @llvm.ceil.nxv1f16(<vscale x 1 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 2 x half> @llvm.ceil.nxv2f16(<vscale x 2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 4 x half> @llvm.ceil.nxv4f16(<vscale x 4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 8 x half> @llvm.ceil.nxv8f16(<vscale x 8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %10 = call <vscale x 16 x half> @llvm.ceil.nxv16f16(<vscale x 16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'ceil_fp16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call half @llvm.ceil.f16(half undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x half> @llvm.ceil.v2f16(<2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x half> @llvm.ceil.v4f16(<4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x half> @llvm.ceil.v8f16(<8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x half> @llvm.ceil.v16f16(<16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x half> @llvm.ceil.nxv1f16(<vscale x 1 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x half> @llvm.ceil.nxv2f16(<vscale x 2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x half> @llvm.ceil.nxv4f16(<vscale x 4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x half> @llvm.ceil.nxv8f16(<vscale x 8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x half> @llvm.ceil.nxv16f16(<vscale x 16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call half @llvm.ceil.f16(half undef)
  call <2 x half> @llvm.ceil.v2f16(<2 x half> undef)
  call <4 x half> @llvm.ceil.v4f16(<4 x half> undef)
  call <8 x half> @llvm.ceil.v8f16(<8 x half> undef)
  call <16 x half> @llvm.ceil.v16f16(<16 x half> undef)
  call <vscale x 1 x half> @llvm.ceil.nxv1f16(<vscale x 1 x half> undef)
  call <vscale x 2 x half> @llvm.ceil.nxv2f16(<vscale x 2 x half> undef)
  call <vscale x 4 x half> @llvm.ceil.nxv4f16(<vscale x 4 x half> undef)
  call <vscale x 8 x half> @llvm.ceil.nxv8f16(<vscale x 8 x half> undef)
  call <vscale x 16 x half> @llvm.ceil.nxv16f16(<vscale x 16 x half> undef)
  ret void
}

define void @trunc() {
; CHECK-LABEL: 'trunc'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.trunc.bf16(bfloat undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.trunc.v2bf16(<2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.trunc.v4bf16(<4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.trunc.v8bf16(<8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.trunc.v16bf16(<16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.trunc.nxv1bf16(<vscale x 1 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.trunc.nxv2bf16(<vscale x 2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.trunc.nxv4bf16(<vscale x 4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.trunc.nxv8bf16(<vscale x 8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.trunc.nxv16bf16(<vscale x 16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.trunc.f32(float undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %12 = call <2 x float> @llvm.trunc.v2f32(<2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %13 = call <4 x float> @llvm.trunc.v4f32(<4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %14 = call <8 x float> @llvm.trunc.v8f32(<8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %15 = call <16 x float> @llvm.trunc.v16f32(<16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 1 x float> @llvm.trunc.nxv1f32(<vscale x 1 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 2 x float> @llvm.trunc.nxv2f32(<vscale x 2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 4 x float> @llvm.trunc.nxv4f32(<vscale x 4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %19 = call <vscale x 8 x float> @llvm.trunc.nxv8f32(<vscale x 8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %20 = call <vscale x 16 x float> @llvm.trunc.nxv16f32(<vscale x 16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.trunc.f64(double undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %22 = call <2 x double> @llvm.trunc.v2f64(<2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %23 = call <4 x double> @llvm.trunc.v4f64(<4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %24 = call <8 x double> @llvm.trunc.v8f64(<8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %25 = call <16 x double> @llvm.trunc.v16f64(<16 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 1 x double> @llvm.trunc.nxv1f64(<vscale x 1 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %27 = call <vscale x 2 x double> @llvm.trunc.nxv2f64(<vscale x 2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %28 = call <vscale x 4 x double> @llvm.trunc.nxv4f64(<vscale x 4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %29 = call <vscale x 8 x double> @llvm.trunc.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call bfloat @llvm.trunc.bf16(bfloat undef)
  call <2 x bfloat> @llvm.trunc.v2bf16(<2 x bfloat> undef)
  call <4 x bfloat> @llvm.trunc.v4bf16(<4 x bfloat> undef)
  call <8 x bfloat> @llvm.trunc.v8bf16(<8 x bfloat> undef)
  call <16 x bfloat> @llvm.trunc.v16bf16(<16 x bfloat> undef)
  call <vscale x 1 x bfloat> @llvm.trunc.nxv1bf16(<vscale x 1 x bfloat> undef)
  call <vscale x 2 x bfloat> @llvm.trunc.nxv2bf16(<vscale x 2 x bfloat> undef)
  call <vscale x 4 x bfloat> @llvm.trunc.nxv4bf16(<vscale x 4 x bfloat> undef)
  call <vscale x 8 x bfloat> @llvm.trunc.nxv8bf16(<vscale x 8 x bfloat> undef)
  call <vscale x 16 x bfloat> @llvm.trunc.nxv16bf16(<vscale x 16 x bfloat> undef)
  call float @llvm.trunc.f32(float undef)
  call <2 x float> @llvm.trunc.v2f32(<2 x float> undef)
  call <4 x float> @llvm.trunc.v4f32(<4 x float> undef)
  call <8 x float> @llvm.trunc.v8f32(<8 x float> undef)
  call <16 x float> @llvm.trunc.v16f32(<16 x float> undef)
  call <vscale x 1 x float> @llvm.trunc.nxv1f32(<vscale x 1 x float> undef)
  call <vscale x 2 x float> @llvm.trunc.nxv2f32(<vscale x 2 x float> undef)
  call <vscale x 4 x float> @llvm.trunc.nxv4f32(<vscale x 4 x float> undef)
  call <vscale x 8 x float> @llvm.trunc.nxv8f32(<vscale x 8 x float> undef)
  call <vscale x 16 x float> @llvm.trunc.nxv16f32(<vscale x 16 x float> undef)
  call double @llvm.trunc.f64(double undef)
  call <2 x double> @llvm.trunc.v2f64(<2 x double> undef)
  call <4 x double> @llvm.trunc.v4f64(<4 x double> undef)
  call <8 x double> @llvm.trunc.v8f64(<8 x double> undef)
  call <16 x double> @llvm.trunc.v16f64(<16 x double> undef)
  call <vscale x 1 x double> @llvm.trunc.nxv1f64(<vscale x 1 x double> undef)
  call <vscale x 2 x double> @llvm.trunc.nxv2f64(<vscale x 2 x double> undef)
  call <vscale x 4 x double> @llvm.trunc.nxv4f64(<vscale x 4 x double> undef)
  call <vscale x 8 x double> @llvm.trunc.nxv8f64(<vscale x 8 x double> undef)
  ret void
}

define void @trunc_fp16() {
; ZVFH-LABEL: 'trunc_fp16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.trunc.f16(half undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <2 x half> @llvm.trunc.v2f16(<2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <4 x half> @llvm.trunc.v4f16(<4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <8 x half> @llvm.trunc.v8f16(<8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x half> @llvm.trunc.v16f16(<16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 1 x half> @llvm.trunc.nxv1f16(<vscale x 1 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 2 x half> @llvm.trunc.nxv2f16(<vscale x 2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 4 x half> @llvm.trunc.nxv4f16(<vscale x 4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 8 x half> @llvm.trunc.nxv8f16(<vscale x 8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %10 = call <vscale x 16 x half> @llvm.trunc.nxv16f16(<vscale x 16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'trunc_fp16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call half @llvm.trunc.f16(half undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x half> @llvm.trunc.v2f16(<2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x half> @llvm.trunc.v4f16(<4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x half> @llvm.trunc.v8f16(<8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x half> @llvm.trunc.v16f16(<16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x half> @llvm.trunc.nxv1f16(<vscale x 1 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x half> @llvm.trunc.nxv2f16(<vscale x 2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x half> @llvm.trunc.nxv4f16(<vscale x 4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x half> @llvm.trunc.nxv8f16(<vscale x 8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x half> @llvm.trunc.nxv16f16(<vscale x 16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call half @llvm.trunc.f16(half undef)
  call <2 x half> @llvm.trunc.v2f16(<2 x half> undef)
  call <4 x half> @llvm.trunc.v4f16(<4 x half> undef)
  call <8 x half> @llvm.trunc.v8f16(<8 x half> undef)
  call <16 x half> @llvm.trunc.v16f16(<16 x half> undef)
  call <vscale x 1 x half> @llvm.trunc.nxv1f16(<vscale x 1 x half> undef)
  call <vscale x 2 x half> @llvm.trunc.nxv2f16(<vscale x 2 x half> undef)
  call <vscale x 4 x half> @llvm.trunc.nxv4f16(<vscale x 4 x half> undef)
  call <vscale x 8 x half> @llvm.trunc.nxv8f16(<vscale x 8 x half> undef)
  call <vscale x 16 x half> @llvm.trunc.nxv16f16(<vscale x 16 x half> undef)
  ret void
}

define void @rint() {
; CHECK-LABEL: 'rint'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.rint.bf16(bfloat undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.rint.v2bf16(<2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.rint.v4bf16(<4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.rint.v8bf16(<8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.rint.v16bf16(<16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.rint.nxv1bf16(<vscale x 1 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.rint.nxv2bf16(<vscale x 2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.rint.nxv4bf16(<vscale x 4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.rint.nxv8bf16(<vscale x 8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.rint.nxv16bf16(<vscale x 16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.rint.f32(float undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %12 = call <2 x float> @llvm.rint.v2f32(<2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %13 = call <4 x float> @llvm.rint.v4f32(<4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %14 = call <8 x float> @llvm.rint.v8f32(<8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %15 = call <16 x float> @llvm.rint.v16f32(<16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 1 x float> @llvm.rint.nxv1f32(<vscale x 1 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 2 x float> @llvm.rint.nxv2f32(<vscale x 2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 4 x float> @llvm.rint.nxv4f32(<vscale x 4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %19 = call <vscale x 8 x float> @llvm.rint.nxv8f32(<vscale x 8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %20 = call <vscale x 16 x float> @llvm.rint.nxv16f32(<vscale x 16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.rint.f64(double undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %22 = call <2 x double> @llvm.rint.v2f64(<2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %23 = call <4 x double> @llvm.rint.v4f64(<4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %24 = call <8 x double> @llvm.rint.v8f64(<8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %25 = call <16 x double> @llvm.rint.v16f64(<16 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 1 x double> @llvm.rint.nxv1f64(<vscale x 1 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %27 = call <vscale x 2 x double> @llvm.rint.nxv2f64(<vscale x 2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %28 = call <vscale x 4 x double> @llvm.rint.nxv4f64(<vscale x 4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %29 = call <vscale x 8 x double> @llvm.rint.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call bfloat @llvm.rint.bf16(bfloat undef)
  call <2 x bfloat> @llvm.rint.v2bf16(<2 x bfloat> undef)
  call <4 x bfloat> @llvm.rint.v4bf16(<4 x bfloat> undef)
  call <8 x bfloat> @llvm.rint.v8bf16(<8 x bfloat> undef)
  call <16 x bfloat> @llvm.rint.v16bf16(<16 x bfloat> undef)
  call <vscale x 1 x bfloat> @llvm.rint.nxv1bf16(<vscale x 1 x bfloat> undef)
  call <vscale x 2 x bfloat> @llvm.rint.nxv2bf16(<vscale x 2 x bfloat> undef)
  call <vscale x 4 x bfloat> @llvm.rint.nxv4bf16(<vscale x 4 x bfloat> undef)
  call <vscale x 8 x bfloat> @llvm.rint.nxv8bf16(<vscale x 8 x bfloat> undef)
  call <vscale x 16 x bfloat> @llvm.rint.nxv16bf16(<vscale x 16 x bfloat> undef)
  call float @llvm.rint.f32(float undef)
  call <2 x float> @llvm.rint.v2f32(<2 x float> undef)
  call <4 x float> @llvm.rint.v4f32(<4 x float> undef)
  call <8 x float> @llvm.rint.v8f32(<8 x float> undef)
  call <16 x float> @llvm.rint.v16f32(<16 x float> undef)
  call <vscale x 1 x float> @llvm.rint.nxv1f32(<vscale x 1 x float> undef)
  call <vscale x 2 x float> @llvm.rint.nxv2f32(<vscale x 2 x float> undef)
  call <vscale x 4 x float> @llvm.rint.nxv4f32(<vscale x 4 x float> undef)
  call <vscale x 8 x float> @llvm.rint.nxv8f32(<vscale x 8 x float> undef)
  call <vscale x 16 x float> @llvm.rint.nxv16f32(<vscale x 16 x float> undef)
  call double @llvm.rint.f64(double undef)
  call <2 x double> @llvm.rint.v2f64(<2 x double> undef)
  call <4 x double> @llvm.rint.v4f64(<4 x double> undef)
  call <8 x double> @llvm.rint.v8f64(<8 x double> undef)
  call <16 x double> @llvm.rint.v16f64(<16 x double> undef)
  call <vscale x 1 x double> @llvm.rint.nxv1f64(<vscale x 1 x double> undef)
  call <vscale x 2 x double> @llvm.rint.nxv2f64(<vscale x 2 x double> undef)
  call <vscale x 4 x double> @llvm.rint.nxv4f64(<vscale x 4 x double> undef)
  call <vscale x 8 x double> @llvm.rint.nxv8f64(<vscale x 8 x double> undef)
  ret void
}

define void @rint_fp16() {
; ZVFH-LABEL: 'rint_fp16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.rint.f16(half undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <2 x half> @llvm.rint.v2f16(<2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <4 x half> @llvm.rint.v4f16(<4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <8 x half> @llvm.rint.v8f16(<8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x half> @llvm.rint.v16f16(<16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 1 x half> @llvm.rint.nxv1f16(<vscale x 1 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 2 x half> @llvm.rint.nxv2f16(<vscale x 2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 4 x half> @llvm.rint.nxv4f16(<vscale x 4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 8 x half> @llvm.rint.nxv8f16(<vscale x 8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %10 = call <vscale x 16 x half> @llvm.rint.nxv16f16(<vscale x 16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'rint_fp16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call half @llvm.rint.f16(half undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x half> @llvm.rint.v2f16(<2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x half> @llvm.rint.v4f16(<4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x half> @llvm.rint.v8f16(<8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x half> @llvm.rint.v16f16(<16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x half> @llvm.rint.nxv1f16(<vscale x 1 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x half> @llvm.rint.nxv2f16(<vscale x 2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x half> @llvm.rint.nxv4f16(<vscale x 4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x half> @llvm.rint.nxv8f16(<vscale x 8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x half> @llvm.rint.nxv16f16(<vscale x 16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call half @llvm.rint.f16(half undef)
  call <2 x half> @llvm.rint.v2f16(<2 x half> undef)
  call <4 x half> @llvm.rint.v4f16(<4 x half> undef)
  call <8 x half> @llvm.rint.v8f16(<8 x half> undef)
  call <16 x half> @llvm.rint.v16f16(<16 x half> undef)
  call <vscale x 1 x half> @llvm.rint.nxv1f16(<vscale x 1 x half> undef)
  call <vscale x 2 x half> @llvm.rint.nxv2f16(<vscale x 2 x half> undef)
  call <vscale x 4 x half> @llvm.rint.nxv4f16(<vscale x 4 x half> undef)
  call <vscale x 8 x half> @llvm.rint.nxv8f16(<vscale x 8 x half> undef)
  call <vscale x 16 x half> @llvm.rint.nxv16f16(<vscale x 16 x half> undef)
  ret void
}

define void @lrint() {
; CHECK-LABEL: 'lrint'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call i64 @llvm.lrint.i64.bf16(bfloat undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %2 = call <2 x i64> @llvm.lrint.v2i64.v2bf16(<2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %3 = call <4 x i64> @llvm.lrint.v4i64.v4bf16(<4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %4 = call <8 x i64> @llvm.lrint.v8i64.v8bf16(<8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %5 = call <16 x i64> @llvm.lrint.v16i64.v16bf16(<16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %6 = call <vscale x 1 x i64> @llvm.lrint.nxv1i64.nxv1bf16(<vscale x 1 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %7 = call <vscale x 2 x i64> @llvm.lrint.nxv2i64.nxv2bf16(<vscale x 2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %8 = call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4bf16(<vscale x 4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %9 = call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8bf16(<vscale x 8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %10 = call <vscale x 16 x i64> @llvm.lrint.nxv16i64.nxv16bf16(<vscale x 16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %11 = call i64 @llvm.lrint.i64.f32(float undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x i64> @llvm.lrint.v2i64.v2f32(<2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x i64> @llvm.lrint.v4i64.v4f32(<4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %14 = call <8 x i64> @llvm.lrint.v8i64.v8f32(<8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %15 = call <16 x i64> @llvm.lrint.v16i64.v16f32(<16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x i64> @llvm.lrint.nxv1i64.nxv1f32(<vscale x 1 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %17 = call <vscale x 2 x i64> @llvm.lrint.nxv2i64.nxv2f32(<vscale x 2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %18 = call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4f32(<vscale x 4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %19 = call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8f32(<vscale x 8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %20 = call <vscale x 16 x i64> @llvm.lrint.nxv16i64.nxv16f32(<vscale x 16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %21 = call i64 @llvm.lrint.i64.f64(double undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %22 = call <2 x i64> @llvm.lrint.v2i64.v2f64(<2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %23 = call <4 x i64> @llvm.lrint.v4i64.v4f64(<4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %24 = call <8 x i64> @llvm.lrint.v8i64.v8f64(<8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %25 = call <16 x i64> @llvm.lrint.v16i64.v16f64(<16 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %26 = call <vscale x 1 x i64> @llvm.lrint.nxv1i64.nxv1f64(<vscale x 1 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %27 = call <vscale x 2 x i64> @llvm.lrint.nxv2i64.nxv2f64(<vscale x 2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %28 = call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4f64(<vscale x 4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %29 = call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call i64 @llvm.lrint.i64.bf16(bfloat undef)
  call <2 x i64> @llvm.lrint.v2i64.v2bf16(<2 x bfloat> undef)
  call <4 x i64> @llvm.lrint.v4i64.v4bf16(<4 x bfloat> undef)
  call <8 x i64> @llvm.lrint.v8i64.v8bf16(<8 x bfloat> undef)
  call <16 x i64> @llvm.lrint.v16i64.v16bf16(<16 x bfloat> undef)
  call <vscale x 1 x i64> @llvm.lrint.nxv1i64.nxv1bf16(<vscale x 1 x bfloat> undef)
  call <vscale x 2 x i64> @llvm.lrint.nxv2i64.nxv2bf16(<vscale x 2 x bfloat> undef)
  call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4bf16(<vscale x 4 x bfloat> undef)
  call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8bf16(<vscale x 8 x bfloat> undef)
  call <vscale x 16 x i64> @llvm.lrint.nxv16i64.nxv16bf16(<vscale x 16 x bfloat> undef)
  call i64 @llvm.lrint.i64.f32(float undef)
  call <2 x i64> @llvm.lrint.v2i64.v2f32(<2 x float> undef)
  call <4 x i64> @llvm.lrint.v4i64.v4f32(<4 x float> undef)
  call <8 x i64> @llvm.lrint.v8i64.v8f32(<8 x float> undef)
  call <16 x i64> @llvm.lrint.v16i64.v16f32(<16 x float> undef)
  call <vscale x 1 x i64> @llvm.lrint.nxv1i64.nxv1f32(<vscale x 1 x float> undef)
  call <vscale x 2 x i64> @llvm.lrint.nxv2i64.nxv2f32(<vscale x 2 x float> undef)
  call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4f32(<vscale x 4 x float> undef)
  call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8f32(<vscale x 8 x float> undef)
  call <vscale x 16 x i64> @llvm.lrint.nxv16i64.nxv16f32(<vscale x 16 x float> undef)
  call i64 @llvm.lrint.i64.f64(double undef)
  call <2 x i64> @llvm.lrint.v2i64.v2f64(<2 x double> undef)
  call <4 x i64> @llvm.lrint.v4i64.v4f64(<4 x double> undef)
  call <8 x i64> @llvm.lrint.v8i64.v8f64(<8 x double> undef)
  call <16 x i64> @llvm.lrint.v16i64.v16f64(<16 x double> undef)
  call <vscale x 1 x i64> @llvm.lrint.nxv1i64.nxv1f64(<vscale x 1 x double> undef)
  call <vscale x 2 x i64> @llvm.lrint.nxv2i64.nxv2f64(<vscale x 2 x double> undef)
  call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4f64(<vscale x 4 x double> undef)
  call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8f64(<vscale x 8 x double> undef)
  ret void
}

define void @lrint_fp16() {
; CHECK-LABEL: 'lrint_fp16'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call i64 @llvm.lrint.i64.f16(half undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %2 = call <2 x i64> @llvm.lrint.v2i64.v2f16(<2 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %3 = call <4 x i64> @llvm.lrint.v4i64.v4f16(<4 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %4 = call <8 x i64> @llvm.lrint.v8i64.v8f16(<8 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %5 = call <16 x i64> @llvm.lrint.v16i64.v16f16(<16 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %6 = call <vscale x 1 x i64> @llvm.lrint.nxv1i64.nxv1f16(<vscale x 1 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %7 = call <vscale x 2 x i64> @llvm.lrint.nxv2i64.nxv2f16(<vscale x 2 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %8 = call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4f16(<vscale x 4 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %9 = call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8f16(<vscale x 8 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %10 = call <vscale x 16 x i64> @llvm.lrint.nxv16i64.nxv16f16(<vscale x 16 x half> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call i64 @llvm.lrint.f16(half undef)
  call <2 x i64> @llvm.lrint.v2f16(<2 x half> undef)
  call <4 x i64> @llvm.lrint.v4f16(<4 x half> undef)
  call <8 x i64> @llvm.lrint.v8f16(<8 x half> undef)
  call <16 x i64> @llvm.lrint.v16f16(<16 x half> undef)
  call <vscale x 1 x i64> @llvm.lrint.nxv1f16(<vscale x 1 x half> undef)
  call <vscale x 2 x i64> @llvm.lrint.nxv2f16(<vscale x 2 x half> undef)
  call <vscale x 4 x i64> @llvm.lrint.nxv4f16(<vscale x 4 x half> undef)
  call <vscale x 8 x i64> @llvm.lrint.nxv8f16(<vscale x 8 x half> undef)
  call <vscale x 16 x i64> @llvm.lrint.nxv16f16(<vscale x 16 x half> undef)
  ret void
}

define void @llrint() {
; CHECK-LABEL: 'llrint'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call i64 @llvm.llrint.i64.bf16(bfloat undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %2 = call <2 x i64> @llvm.llrint.v2i64.v2bf16(<2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %3 = call <4 x i64> @llvm.llrint.v4i64.v4bf16(<4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %4 = call <8 x i64> @llvm.llrint.v8i64.v8bf16(<8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %5 = call <16 x i64> @llvm.llrint.v16i64.v16bf16(<16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %6 = call <vscale x 1 x i64> @llvm.llrint.nxv1i64.nxv1bf16(<vscale x 1 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %7 = call <vscale x 2 x i64> @llvm.llrint.nxv2i64.nxv2bf16(<vscale x 2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %8 = call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4bf16(<vscale x 4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %9 = call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8bf16(<vscale x 8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %10 = call <vscale x 16 x i64> @llvm.llrint.nxv16i64.nxv16bf16(<vscale x 16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %11 = call i64 @llvm.llrint.i64.f32(float undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x i64> @llvm.llrint.v2i64.v2f32(<2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x i64> @llvm.llrint.v4i64.v4f32(<4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %14 = call <8 x i64> @llvm.llrint.v8i64.v8f32(<8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %15 = call <16 x i64> @llvm.llrint.v16i64.v16f32(<16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x i64> @llvm.llrint.nxv1i64.nxv1f32(<vscale x 1 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %17 = call <vscale x 2 x i64> @llvm.llrint.nxv2i64.nxv2f32(<vscale x 2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %18 = call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4f32(<vscale x 4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %19 = call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8f32(<vscale x 8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %20 = call <vscale x 16 x i64> @llvm.llrint.nxv16i64.nxv16f32(<vscale x 16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %21 = call i64 @llvm.llrint.i64.f64(double undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %22 = call <2 x i64> @llvm.llrint.v2i64.v2f64(<2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %23 = call <4 x i64> @llvm.llrint.v4i64.v4f64(<4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %24 = call <8 x i64> @llvm.llrint.v8i64.v8f64(<8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %25 = call <16 x i64> @llvm.llrint.v16i64.v16f64(<16 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %26 = call <vscale x 1 x i64> @llvm.llrint.nxv1i64.nxv1f64(<vscale x 1 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %27 = call <vscale x 2 x i64> @llvm.llrint.nxv2i64.nxv2f64(<vscale x 2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %28 = call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4f64(<vscale x 4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %29 = call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call i64 @llvm.llrint.i64.bf16(bfloat undef)
  call <2 x i64> @llvm.llrint.v2i64.v2bf16(<2 x bfloat> undef)
  call <4 x i64> @llvm.llrint.v4i64.v4bf16(<4 x bfloat> undef)
  call <8 x i64> @llvm.llrint.v8i64.v8bf16(<8 x bfloat> undef)
  call <16 x i64> @llvm.llrint.v16i64.v16bf16(<16 x bfloat> undef)
  call <vscale x 1 x i64> @llvm.llrint.nxv1i64.nxv1bf16(<vscale x 1 x bfloat> undef)
  call <vscale x 2 x i64> @llvm.llrint.nxv2i64.nxv2bf16(<vscale x 2 x bfloat> undef)
  call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4bf16(<vscale x 4 x bfloat> undef)
  call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8bf16(<vscale x 8 x bfloat> undef)
  call <vscale x 16 x i64> @llvm.llrint.nxv16i64.nxv16bf16(<vscale x 16 x bfloat> undef)
  call i64 @llvm.llrint.i64.f32(float undef)
  call <2 x i64> @llvm.llrint.v2i64.v2f32(<2 x float> undef)
  call <4 x i64> @llvm.llrint.v4i64.v4f32(<4 x float> undef)
  call <8 x i64> @llvm.llrint.v8i64.v8f32(<8 x float> undef)
  call <16 x i64> @llvm.llrint.v16i64.v16f32(<16 x float> undef)
  call <vscale x 1 x i64> @llvm.llrint.nxv1i64.nxv1f32(<vscale x 1 x float> undef)
  call <vscale x 2 x i64> @llvm.llrint.nxv2i64.nxv2f32(<vscale x 2 x float> undef)
  call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4f32(<vscale x 4 x float> undef)
  call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8f32(<vscale x 8 x float> undef)
  call <vscale x 16 x i64> @llvm.llrint.nxv16i64.nxv16f32(<vscale x 16 x float> undef)
  call i64 @llvm.llrint.i64.f64(double undef)
  call <2 x i64> @llvm.llrint.v2i64.v2f64(<2 x double> undef)
  call <4 x i64> @llvm.llrint.v4i64.v4f64(<4 x double> undef)
  call <8 x i64> @llvm.llrint.v8i64.v8f64(<8 x double> undef)
  call <16 x i64> @llvm.llrint.v16i64.v16f64(<16 x double> undef)
  call <vscale x 1 x i64> @llvm.llrint.nxv1i64.nxv1f64(<vscale x 1 x double> undef)
  call <vscale x 2 x i64> @llvm.llrint.nxv2i64.nxv2f64(<vscale x 2 x double> undef)
  call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4f64(<vscale x 4 x double> undef)
  call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8f64(<vscale x 8 x double> undef)
  ret void
}

define void @llrint_fp16() {
; CHECK-LABEL: 'llrint_fp16'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call i64 @llvm.llrint.i64.f16(half undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %2 = call <2 x i64> @llvm.llrint.v2i64.v2f16(<2 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %3 = call <4 x i64> @llvm.llrint.v4i64.v4f16(<4 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %4 = call <8 x i64> @llvm.llrint.v8i64.v8f16(<8 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %5 = call <16 x i64> @llvm.llrint.v16i64.v16f16(<16 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %6 = call <vscale x 1 x i64> @llvm.llrint.nxv1i64.nxv1f16(<vscale x 1 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %7 = call <vscale x 2 x i64> @llvm.llrint.nxv2i64.nxv2f16(<vscale x 2 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %8 = call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4f16(<vscale x 4 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %9 = call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8f16(<vscale x 8 x half> undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %10 = call <vscale x 16 x i64> @llvm.llrint.nxv16i64.nxv16f16(<vscale x 16 x half> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call i64 @llvm.llrint.f16(half undef)
  call <2 x i64> @llvm.llrint.v2f16(<2 x half> undef)
  call <4 x i64> @llvm.llrint.v4f16(<4 x half> undef)
  call <8 x i64> @llvm.llrint.v8f16(<8 x half> undef)
  call <16 x i64> @llvm.llrint.v16f16(<16 x half> undef)
  call <vscale x 1 x i64> @llvm.llrint.nxv1f16(<vscale x 1 x half> undef)
  call <vscale x 2 x i64> @llvm.llrint.nxv2f16(<vscale x 2 x half> undef)
  call <vscale x 4 x i64> @llvm.llrint.nxv4f16(<vscale x 4 x half> undef)
  call <vscale x 8 x i64> @llvm.llrint.nxv8f16(<vscale x 8 x half> undef)
  call <vscale x 16 x i64> @llvm.llrint.nxv16f16(<vscale x 16 x half> undef)
  ret void
}

define void @lround() {
; CHECK-LABEL: 'lround'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call i32 @llvm.lround.i32.bf16(bfloat poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %2 = call <2 x i32> @llvm.lround.v2i32.v2bf16(<2 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %3 = call <4 x i32> @llvm.lround.v4i32.v4bf16(<4 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %4 = call <8 x i32> @llvm.lround.v8i32.v8bf16(<8 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %5 = call <16 x i32> @llvm.lround.v16i32.v16bf16(<16 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %6 = call <vscale x 1 x i32> @llvm.lround.nxv1i32.nxv1bf16(<vscale x 1 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %7 = call <vscale x 2 x i32> @llvm.lround.nxv2i32.nxv2bf16(<vscale x 2 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %8 = call <vscale x 4 x i32> @llvm.lround.nxv4i32.nxv4bf16(<vscale x 4 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %9 = call <vscale x 8 x i32> @llvm.lround.nxv8i32.nxv8bf16(<vscale x 8 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %10 = call <vscale x 16 x i32> @llvm.lround.nxv16i32.nxv16bf16(<vscale x 16 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %11 = call i32 @llvm.lround.i32.f32(float poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %12 = call <2 x i32> @llvm.lround.v2i32.v2f32(<2 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %13 = call <4 x i32> @llvm.lround.v4i32.v4f32(<4 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %14 = call <8 x i32> @llvm.lround.v8i32.v8f32(<8 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %15 = call <16 x i32> @llvm.lround.v16i32.v16f32(<16 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %16 = call <vscale x 1 x i32> @llvm.lround.nxv1i32.nxv1f32(<vscale x 1 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %17 = call <vscale x 2 x i32> @llvm.lround.nxv2i32.nxv2f32(<vscale x 2 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %18 = call <vscale x 4 x i32> @llvm.lround.nxv4i32.nxv4f32(<vscale x 4 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %19 = call <vscale x 8 x i32> @llvm.lround.nxv8i32.nxv8f32(<vscale x 8 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %20 = call <vscale x 16 x i32> @llvm.lround.nxv16i32.nxv16f32(<vscale x 16 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %21 = call i32 @llvm.lround.i32.f64(double poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %22 = call <2 x i32> @llvm.lround.v2i32.v2f64(<2 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %23 = call <4 x i32> @llvm.lround.v4i32.v4f64(<4 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %24 = call <8 x i32> @llvm.lround.v8i32.v8f64(<8 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %25 = call <16 x i32> @llvm.lround.v16i32.v16f64(<16 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %26 = call <vscale x 1 x i32> @llvm.lround.nxv1i32.nxv1f64(<vscale x 1 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %27 = call <vscale x 2 x i32> @llvm.lround.nxv2i32.nxv2f64(<vscale x 2 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %28 = call <vscale x 4 x i32> @llvm.lround.nxv4i32.nxv4f64(<vscale x 4 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %29 = call <vscale x 8 x i32> @llvm.lround.nxv8i32.nxv8f64(<vscale x 8 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %30 = call <vscale x 16 x i32> @llvm.lround.nxv16i32.nxv16f64(<vscale x 16 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %31 = call i64 @llvm.lround.i64.bf16(bfloat poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %32 = call <2 x i64> @llvm.lround.v2i64.v2bf16(<2 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %33 = call <4 x i64> @llvm.lround.v4i64.v4bf16(<4 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %34 = call <8 x i64> @llvm.lround.v8i64.v8bf16(<8 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %35 = call <16 x i64> @llvm.lround.v16i64.v16bf16(<16 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %36 = call <vscale x 1 x i64> @llvm.lround.nxv1i64.nxv1bf16(<vscale x 1 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %37 = call <vscale x 2 x i64> @llvm.lround.nxv2i64.nxv2bf16(<vscale x 2 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %38 = call <vscale x 4 x i64> @llvm.lround.nxv4i64.nxv4bf16(<vscale x 4 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %39 = call <vscale x 8 x i64> @llvm.lround.nxv8i64.nxv8bf16(<vscale x 8 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %40 = call <vscale x 16 x i64> @llvm.lround.nxv16i64.nxv16bf16(<vscale x 16 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %41 = call i64 @llvm.lround.i64.f32(float poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %42 = call <2 x i64> @llvm.lround.v2i64.v2f32(<2 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %43 = call <4 x i64> @llvm.lround.v4i64.v4f32(<4 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %44 = call <8 x i64> @llvm.lround.v8i64.v8f32(<8 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %45 = call <16 x i64> @llvm.lround.v16i64.v16f32(<16 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %46 = call <vscale x 1 x i64> @llvm.lround.nxv1i64.nxv1f32(<vscale x 1 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %47 = call <vscale x 2 x i64> @llvm.lround.nxv2i64.nxv2f32(<vscale x 2 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %48 = call <vscale x 4 x i64> @llvm.lround.nxv4i64.nxv4f32(<vscale x 4 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %49 = call <vscale x 8 x i64> @llvm.lround.nxv8i64.nxv8f32(<vscale x 8 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %50 = call <vscale x 16 x i64> @llvm.lround.nxv16i64.nxv16f32(<vscale x 16 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %51 = call i64 @llvm.lround.i64.f64(double poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %52 = call <2 x i64> @llvm.lround.v2i64.v2f64(<2 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %53 = call <4 x i64> @llvm.lround.v4i64.v4f64(<4 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %54 = call <8 x i64> @llvm.lround.v8i64.v8f64(<8 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %55 = call <16 x i64> @llvm.lround.v16i64.v16f64(<16 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %56 = call <vscale x 1 x i64> @llvm.lround.nxv1i64.nxv1f64(<vscale x 1 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %57 = call <vscale x 2 x i64> @llvm.lround.nxv2i64.nxv2f64(<vscale x 2 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %58 = call <vscale x 4 x i64> @llvm.lround.nxv4i64.nxv4f64(<vscale x 4 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %59 = call <vscale x 8 x i64> @llvm.lround.nxv8i64.nxv8f64(<vscale x 8 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %60 = call <vscale x 16 x i64> @llvm.lround.nxv16i64.nxv16f64(<vscale x 16 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call i32 @llvm.lround.i32.bf16(bfloat poison)
  call <2 x i32> @llvm.lround.v2i32.v2bf16(<2 x bfloat> poison)
  call <4 x i32> @llvm.lround.v4i32.v4bf16(<4 x bfloat> poison)
  call <8 x i32> @llvm.lround.v8i32.v8bf16(<8 x bfloat> poison)
  call <16 x i32> @llvm.lround.v16i32.v16bf16(<16 x bfloat> poison)
  call <vscale x 1 x i32> @llvm.lround.nxv1i32.nxv1bf16(<vscale x 1 x bfloat> poison)
  call <vscale x 2 x i32> @llvm.lround.nxv2i32.nxv2bf16(<vscale x 2 x bfloat> poison)
  call <vscale x 4 x i32> @llvm.lround.nxv4i32.nxv4bf16(<vscale x 4 x bfloat> poison)
  call <vscale x 8 x i32> @llvm.lround.nxv8i32.nxv8bf16(<vscale x 8 x bfloat> poison)
  call <vscale x 16 x i32> @llvm.lround.nxv16i32.nxv16bf16(<vscale x 16 x bfloat> poison)
  call i32 @llvm.lround.i32.f32(float poison)
  call <2 x i32> @llvm.lround.v2i32.v2f32(<2 x float> poison)
  call <4 x i32> @llvm.lround.v4i32.v4f32(<4 x float> poison)
  call <8 x i32> @llvm.lround.v8i32.v8f32(<8 x float> poison)
  call <16 x i32> @llvm.lround.v16i32.v16f32(<16 x float> poison)
  call <vscale x 1 x i32> @llvm.lround.nxv1i32.nxv1f32(<vscale x 1 x float> poison)
  call <vscale x 2 x i32> @llvm.lround.nxv2i32.nxv2f32(<vscale x 2 x float> poison)
  call <vscale x 4 x i32> @llvm.lround.nxv4i32.nxv4f32(<vscale x 4 x float> poison)
  call <vscale x 8 x i32> @llvm.lround.nxv8i32.nxv8f32(<vscale x 8 x float> poison)
  call <vscale x 16 x i32> @llvm.lround.nxv16i32.nxv16f32(<vscale x 16 x float> poison)
  call i32 @llvm.lround.i32.f64(double poison)
  call <2 x i32> @llvm.lround.v2i32.v2f64(<2 x double> poison)
  call <4 x i32> @llvm.lround.v4i32.v4f64(<4 x double> poison)
  call <8 x i32> @llvm.lround.v8i32.v8f64(<8 x double> poison)
  call <16 x i32> @llvm.lround.v16i32.v16f64(<16 x double> poison)
  call <vscale x 1 x i32> @llvm.lround.nxv1i32.nxv1f64(<vscale x 1 x double> poison)
  call <vscale x 2 x i32> @llvm.lround.nxv2i32.nxv2f64(<vscale x 2 x double> poison)
  call <vscale x 4 x i32> @llvm.lround.nxv4i32.nxv4f64(<vscale x 4 x double> poison)
  call <vscale x 8 x i32> @llvm.lround.nxv8i32.nxv8f64(<vscale x 8 x double> poison)
  call <vscale x 16 x i32> @llvm.lround.nxv16i32.nxv16f64(<vscale x 16 x double> poison)
  call i64 @llvm.lround.i64.bf16(bfloat poison)
  call <2 x i64> @llvm.lround.v2i64.v2bf16(<2 x bfloat> poison)
  call <4 x i64> @llvm.lround.v4i64.v4bf16(<4 x bfloat> poison)
  call <8 x i64> @llvm.lround.v8i64.v8bf16(<8 x bfloat> poison)
  call <16 x i64> @llvm.lround.v16i64.v16bf16(<16 x bfloat> poison)
  call <vscale x 1 x i64> @llvm.lround.nxv1i64.nxv1bf16(<vscale x 1 x bfloat> poison)
  call <vscale x 2 x i64> @llvm.lround.nxv2i64.nxv2bf16(<vscale x 2 x bfloat> poison)
  call <vscale x 4 x i64> @llvm.lround.nxv4i64.nxv4bf16(<vscale x 4 x bfloat> poison)
  call <vscale x 8 x i64> @llvm.lround.nxv8i64.nxv8bf16(<vscale x 8 x bfloat> poison)
  call <vscale x 16 x i64> @llvm.lround.nxv16i64.nxv16bf16(<vscale x 16 x bfloat> poison)
  call i64 @llvm.lround.i64.f32(float poison)
  call <2 x i64> @llvm.lround.v2i64.v2f32(<2 x float> poison)
  call <4 x i64> @llvm.lround.v4i64.v4f32(<4 x float> poison)
  call <8 x i64> @llvm.lround.v8i64.v8f32(<8 x float> poison)
  call <16 x i64> @llvm.lround.v16i64.v16f32(<16 x float> poison)
  call <vscale x 1 x i64> @llvm.lround.nxv1i64.nxv1f32(<vscale x 1 x float> poison)
  call <vscale x 2 x i64> @llvm.lround.nxv2i64.nxv2f32(<vscale x 2 x float> poison)
  call <vscale x 4 x i64> @llvm.lround.nxv4i64.nxv4f32(<vscale x 4 x float> poison)
  call <vscale x 8 x i64> @llvm.lround.nxv8i64.nxv8f32(<vscale x 8 x float> poison)
  call <vscale x 16 x i64> @llvm.lround.nxv16i64.nxv16f32(<vscale x 16 x float> poison)
  call i64 @llvm.lround.i64.f64(double poison)
  call <2 x i64> @llvm.lround.v2i64.v2f64(<2 x double> poison)
  call <4 x i64> @llvm.lround.v4i64.v4f64(<4 x double> poison)
  call <8 x i64> @llvm.lround.v8i64.v8f64(<8 x double> poison)
  call <16 x i64> @llvm.lround.v16i64.v16f64(<16 x double> poison)
  call <vscale x 1 x i64> @llvm.lround.nxv1i64.nxv1f64(<vscale x 1 x double> poison)
  call <vscale x 2 x i64> @llvm.lround.nxv2i64.nxv2f64(<vscale x 2 x double> poison)
  call <vscale x 4 x i64> @llvm.lround.nxv4i64.nxv4f64(<vscale x 4 x double> poison)
  call <vscale x 8 x i64> @llvm.lround.nxv8i64.nxv8f64(<vscale x 8 x double> poison)
  call <vscale x 16 x i64> @llvm.lround.nxv16i64.nxv16f64(<vscale x 16 x double> poison)
  ret void
}

define void @llround() {
; CHECK-LABEL: 'llround'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call i64 @llvm.llround.i64.bf16(bfloat poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %2 = call <2 x i64> @llvm.llround.v2i64.v2bf16(<2 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %3 = call <4 x i64> @llvm.llround.v4i64.v4bf16(<4 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %4 = call <8 x i64> @llvm.llround.v8i64.v8bf16(<8 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %5 = call <16 x i64> @llvm.llround.v16i64.v16bf16(<16 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %6 = call <vscale x 1 x i64> @llvm.llround.nxv1i64.nxv1bf16(<vscale x 1 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %7 = call <vscale x 2 x i64> @llvm.llround.nxv2i64.nxv2bf16(<vscale x 2 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %8 = call <vscale x 4 x i64> @llvm.llround.nxv4i64.nxv4bf16(<vscale x 4 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %9 = call <vscale x 8 x i64> @llvm.llround.nxv8i64.nxv8bf16(<vscale x 8 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %10 = call <vscale x 16 x i64> @llvm.llround.nxv16i64.nxv16bf16(<vscale x 16 x bfloat> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %11 = call i64 @llvm.llround.i64.f32(float poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %12 = call <2 x i64> @llvm.llround.v2i64.v2f32(<2 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %13 = call <4 x i64> @llvm.llround.v4i64.v4f32(<4 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %14 = call <8 x i64> @llvm.llround.v8i64.v8f32(<8 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %15 = call <16 x i64> @llvm.llround.v16i64.v16f32(<16 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %16 = call <vscale x 1 x i64> @llvm.llround.nxv1i64.nxv1f32(<vscale x 1 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %17 = call <vscale x 2 x i64> @llvm.llround.nxv2i64.nxv2f32(<vscale x 2 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %18 = call <vscale x 4 x i64> @llvm.llround.nxv4i64.nxv4f32(<vscale x 4 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %19 = call <vscale x 8 x i64> @llvm.llround.nxv8i64.nxv8f32(<vscale x 8 x float> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %20 = call <vscale x 16 x i64> @llvm.llround.nxv16i64.nxv16f32(<vscale x 16 x float> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %21 = call i64 @llvm.llround.i64.f64(double poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %22 = call <2 x i64> @llvm.llround.v2i64.v2f64(<2 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %23 = call <4 x i64> @llvm.llround.v4i64.v4f64(<4 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 23 for instruction: %24 = call <8 x i64> @llvm.llround.v8i64.v8f64(<8 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %25 = call <16 x i64> @llvm.llround.v16i64.v16f64(<16 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %26 = call <vscale x 1 x i64> @llvm.llround.nxv1i64.nxv1f64(<vscale x 1 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %27 = call <vscale x 2 x i64> @llvm.llround.nxv2i64.nxv2f64(<vscale x 2 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %28 = call <vscale x 4 x i64> @llvm.llround.nxv4i64.nxv4f64(<vscale x 4 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %29 = call <vscale x 8 x i64> @llvm.llround.nxv8i64.nxv8f64(<vscale x 8 x double> poison)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %30 = call <vscale x 16 x i64> @llvm.llround.nxv16i64.nxv16f64(<vscale x 16 x double> poison)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call i64 @llvm.llround.i64.bf16(bfloat poison)
  call <2 x i64> @llvm.llround.v2i64.v2bf16(<2 x bfloat> poison)
  call <4 x i64> @llvm.llround.v4i64.v4bf16(<4 x bfloat> poison)
  call <8 x i64> @llvm.llround.v8i64.v8bf16(<8 x bfloat> poison)
  call <16 x i64> @llvm.llround.v16i64.v16bf16(<16 x bfloat> poison)
  call <vscale x 1 x i64> @llvm.llround.nxv1i64.nxv1bf16(<vscale x 1 x bfloat> poison)
  call <vscale x 2 x i64> @llvm.llround.nxv2i64.nxv2bf16(<vscale x 2 x bfloat> poison)
  call <vscale x 4 x i64> @llvm.llround.nxv4i64.nxv4bf16(<vscale x 4 x bfloat> poison)
  call <vscale x 8 x i64> @llvm.llround.nxv8i64.nxv8bf16(<vscale x 8 x bfloat> poison)
  call <vscale x 16 x i64> @llvm.llround.nxv16i64.nxv16bf16(<vscale x 16 x bfloat> poison)
  call i64 @llvm.llround.i64.f32(float poison)
  call <2 x i64> @llvm.llround.v2i64.v2f32(<2 x float> poison)
  call <4 x i64> @llvm.llround.v4i64.v4f32(<4 x float> poison)
  call <8 x i64> @llvm.llround.v8i64.v8f32(<8 x float> poison)
  call <16 x i64> @llvm.llround.v16i64.v16f32(<16 x float> poison)
  call <vscale x 1 x i64> @llvm.llround.nxv1i64.nxv1f32(<vscale x 1 x float> poison)
  call <vscale x 2 x i64> @llvm.llround.nxv2i64.nxv2f32(<vscale x 2 x float> poison)
  call <vscale x 4 x i64> @llvm.llround.nxv4i64.nxv4f32(<vscale x 4 x float> poison)
  call <vscale x 8 x i64> @llvm.llround.nxv8i64.nxv8f32(<vscale x 8 x float> poison)
  call <vscale x 16 x i64> @llvm.llround.nxv16i64.nxv16f32(<vscale x 16 x float> poison)
  call i64 @llvm.llround.i64.f64(double poison)
  call <2 x i64> @llvm.llround.v2i64.v2f64(<2 x double> poison)
  call <4 x i64> @llvm.llround.v4i64.v4f64(<4 x double> poison)
  call <8 x i64> @llvm.llround.v8i64.v8f64(<8 x double> poison)
  call <16 x i64> @llvm.llround.v16i64.v16f64(<16 x double> poison)
  call <vscale x 1 x i64> @llvm.llround.nxv1i64.nxv1f64(<vscale x 1 x double> poison)
  call <vscale x 2 x i64> @llvm.llround.nxv2i64.nxv2f64(<vscale x 2 x double> poison)
  call <vscale x 4 x i64> @llvm.llround.nxv4i64.nxv4f64(<vscale x 4 x double> poison)
  call <vscale x 8 x i64> @llvm.llround.nxv8i64.nxv8f64(<vscale x 8 x double> poison)
  call <vscale x 16 x i64> @llvm.llround.nxv16i64.nxv16f64(<vscale x 16 x double> poison)
  ret void
}

define void @nearbyint() {
; CHECK-LABEL: 'nearbyint'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %1 = call bfloat @llvm.nearbyint.bf16(bfloat undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.nearbyint.v2bf16(<2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.nearbyint.v4bf16(<4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.nearbyint.v8bf16(<8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.nearbyint.v16bf16(<16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.nearbyint.nxv1bf16(<vscale x 1 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.nearbyint.nxv2bf16(<vscale x 2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.nearbyint.nxv4bf16(<vscale x 4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.nearbyint.nxv8bf16(<vscale x 8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.nearbyint.nxv16bf16(<vscale x 16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %11 = call float @llvm.nearbyint.f32(float undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x float> @llvm.nearbyint.v2f32(<2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x float> @llvm.nearbyint.v4f32(<4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x float> @llvm.nearbyint.v8f32(<8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x float> @llvm.nearbyint.v16f32(<16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x float> @llvm.nearbyint.nxv1f32(<vscale x 1 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x float> @llvm.nearbyint.nxv2f32(<vscale x 2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x float> @llvm.nearbyint.nxv4f32(<vscale x 4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x float> @llvm.nearbyint.nxv8f32(<vscale x 8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %20 = call <vscale x 16 x float> @llvm.nearbyint.nxv16f32(<vscale x 16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %21 = call double @llvm.nearbyint.f64(double undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %22 = call <2 x double> @llvm.nearbyint.v2f64(<2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %23 = call <4 x double> @llvm.nearbyint.v4f64(<4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %24 = call <8 x double> @llvm.nearbyint.v8f64(<8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %25 = call <16 x double> @llvm.nearbyint.v16f64(<16 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 1 x double> @llvm.nearbyint.nxv1f64(<vscale x 1 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %27 = call <vscale x 2 x double> @llvm.nearbyint.nxv2f64(<vscale x 2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %28 = call <vscale x 4 x double> @llvm.nearbyint.nxv4f64(<vscale x 4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %29 = call <vscale x 8 x double> @llvm.nearbyint.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call bfloat @llvm.nearbyint.bf16(bfloat undef)
  call <2 x bfloat> @llvm.nearbyint.v2bf16(<2 x bfloat> undef)
  call <4 x bfloat> @llvm.nearbyint.v4bf16(<4 x bfloat> undef)
  call <8 x bfloat> @llvm.nearbyint.v8bf16(<8 x bfloat> undef)
  call <16 x bfloat> @llvm.nearbyint.v16bf16(<16 x bfloat> undef)
  call <vscale x 1 x bfloat> @llvm.nearbyint.nxv1bf16(<vscale x 1 x bfloat> undef)
  call <vscale x 2 x bfloat> @llvm.nearbyint.nxv2bf16(<vscale x 2 x bfloat> undef)
  call <vscale x 4 x bfloat> @llvm.nearbyint.nxv4bf16(<vscale x 4 x bfloat> undef)
  call <vscale x 8 x bfloat> @llvm.nearbyint.nxv8bf16(<vscale x 8 x bfloat> undef)
  call <vscale x 16 x bfloat> @llvm.nearbyint.nxv16bf16(<vscale x 16 x bfloat> undef)
  call float @llvm.nearbyint.f32(float undef)
  call <2 x float> @llvm.nearbyint.v2f32(<2 x float> undef)
  call <4 x float> @llvm.nearbyint.v4f32(<4 x float> undef)
  call <8 x float> @llvm.nearbyint.v8f32(<8 x float> undef)
  call <16 x float> @llvm.nearbyint.v16f32(<16 x float> undef)
  call <vscale x 1 x float> @llvm.nearbyint.nxv1f32(<vscale x 1 x float> undef)
  call <vscale x 2 x float> @llvm.nearbyint.nxv2f32(<vscale x 2 x float> undef)
  call <vscale x 4 x float> @llvm.nearbyint.nxv4f32(<vscale x 4 x float> undef)
  call <vscale x 8 x float> @llvm.nearbyint.nxv8f32(<vscale x 8 x float> undef)
  call <vscale x 16 x float> @llvm.nearbyint.nxv16f32(<vscale x 16 x float> undef)
  call double @llvm.nearbyint.f64(double undef)
  call <2 x double> @llvm.nearbyint.v2f64(<2 x double> undef)
  call <4 x double> @llvm.nearbyint.v4f64(<4 x double> undef)
  call <8 x double> @llvm.nearbyint.v8f64(<8 x double> undef)
  call <16 x double> @llvm.nearbyint.v16f64(<16 x double> undef)
  call <vscale x 1 x double> @llvm.nearbyint.nxv1f64(<vscale x 1 x double> undef)
  call <vscale x 2 x double> @llvm.nearbyint.nxv2f64(<vscale x 2 x double> undef)
  call <vscale x 4 x double> @llvm.nearbyint.nxv4f64(<vscale x 4 x double> undef)
  call <vscale x 8 x double> @llvm.nearbyint.nxv8f64(<vscale x 8 x double> undef)
  ret void
}

define void @nearbyint_fp16() {
; ZVFH-LABEL: 'nearbyint_fp16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.nearbyint.f16(half undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <2 x half> @llvm.nearbyint.v2f16(<2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <4 x half> @llvm.nearbyint.v4f16(<4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <8 x half> @llvm.nearbyint.v8f16(<8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x half> @llvm.nearbyint.v16f16(<16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 1 x half> @llvm.nearbyint.nxv1f16(<vscale x 1 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 2 x half> @llvm.nearbyint.nxv2f16(<vscale x 2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 4 x half> @llvm.nearbyint.nxv4f16(<vscale x 4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 8 x half> @llvm.nearbyint.nxv8f16(<vscale x 8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %10 = call <vscale x 16 x half> @llvm.nearbyint.nxv16f16(<vscale x 16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'nearbyint_fp16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %1 = call half @llvm.nearbyint.f16(half undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x half> @llvm.nearbyint.v2f16(<2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x half> @llvm.nearbyint.v4f16(<4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x half> @llvm.nearbyint.v8f16(<8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x half> @llvm.nearbyint.v16f16(<16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x half> @llvm.nearbyint.nxv1f16(<vscale x 1 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x half> @llvm.nearbyint.nxv2f16(<vscale x 2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x half> @llvm.nearbyint.nxv4f16(<vscale x 4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x half> @llvm.nearbyint.nxv8f16(<vscale x 8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x half> @llvm.nearbyint.nxv16f16(<vscale x 16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call half @llvm.nearbyint.f16(half undef)
  call <2 x half> @llvm.nearbyint.v2f16(<2 x half> undef)
  call <4 x half> @llvm.nearbyint.v4f16(<4 x half> undef)
  call <8 x half> @llvm.nearbyint.v8f16(<8 x half> undef)
  call <16 x half> @llvm.nearbyint.v16f16(<16 x half> undef)
  call <vscale x 1 x half> @llvm.nearbyint.nxv1f16(<vscale x 1 x half> undef)
  call <vscale x 2 x half> @llvm.nearbyint.nxv2f16(<vscale x 2 x half> undef)
  call <vscale x 4 x half> @llvm.nearbyint.nxv4f16(<vscale x 4 x half> undef)
  call <vscale x 8 x half> @llvm.nearbyint.nxv8f16(<vscale x 8 x half> undef)
  call <vscale x 16 x half> @llvm.nearbyint.nxv16f16(<vscale x 16 x half> undef)
  ret void
}

define void @round() {
; CHECK-LABEL: 'round'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.round.bf16(bfloat undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.round.v2bf16(<2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.round.v4bf16(<4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.round.v8bf16(<8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.round.v16bf16(<16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.round.nxv1bf16(<vscale x 1 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.round.nxv2bf16(<vscale x 2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.round.nxv4bf16(<vscale x 4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.round.nxv8bf16(<vscale x 8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.round.nxv16bf16(<vscale x 16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.round.f32(float undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x float> @llvm.round.v2f32(<2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x float> @llvm.round.v4f32(<4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x float> @llvm.round.v8f32(<8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x float> @llvm.round.v16f32(<16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x float> @llvm.round.nxv1f32(<vscale x 1 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x float> @llvm.round.nxv2f32(<vscale x 2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x float> @llvm.round.nxv4f32(<vscale x 4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x float> @llvm.round.nxv8f32(<vscale x 8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %20 = call <vscale x 16 x float> @llvm.round.nxv16f32(<vscale x 16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.round.f64(double undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %22 = call <2 x double> @llvm.round.v2f64(<2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %23 = call <4 x double> @llvm.round.v4f64(<4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %24 = call <8 x double> @llvm.round.v8f64(<8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %25 = call <16 x double> @llvm.round.v16f64(<16 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 1 x double> @llvm.round.nxv1f64(<vscale x 1 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %27 = call <vscale x 2 x double> @llvm.round.nxv2f64(<vscale x 2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %28 = call <vscale x 4 x double> @llvm.round.nxv4f64(<vscale x 4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %29 = call <vscale x 8 x double> @llvm.round.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call bfloat @llvm.round.bf16(bfloat undef)
  call <2 x bfloat> @llvm.round.v2bf16(<2 x bfloat> undef)
  call <4 x bfloat> @llvm.round.v4bf16(<4 x bfloat> undef)
  call <8 x bfloat> @llvm.round.v8bf16(<8 x bfloat> undef)
  call <16 x bfloat> @llvm.round.v16bf16(<16 x bfloat> undef)
  call <vscale x 1 x bfloat> @llvm.round.nxv1bf16(<vscale x 1 x bfloat> undef)
  call <vscale x 2 x bfloat> @llvm.round.nxv2bf16(<vscale x 2 x bfloat> undef)
  call <vscale x 4 x bfloat> @llvm.round.nxv4bf16(<vscale x 4 x bfloat> undef)
  call <vscale x 8 x bfloat> @llvm.round.nxv8bf16(<vscale x 8 x bfloat> undef)
  call <vscale x 16 x bfloat> @llvm.round.nxv16bf16(<vscale x 16 x bfloat> undef)
  call float @llvm.round.f32(float undef)
  call <2 x float> @llvm.round.v2f32(<2 x float> undef)
  call <4 x float> @llvm.round.v4f32(<4 x float> undef)
  call <8 x float> @llvm.round.v8f32(<8 x float> undef)
  call <16 x float> @llvm.round.v16f32(<16 x float> undef)
  call <vscale x 1 x float> @llvm.round.nxv1f32(<vscale x 1 x float> undef)
  call <vscale x 2 x float> @llvm.round.nxv2f32(<vscale x 2 x float> undef)
  call <vscale x 4 x float> @llvm.round.nxv4f32(<vscale x 4 x float> undef)
  call <vscale x 8 x float> @llvm.round.nxv8f32(<vscale x 8 x float> undef)
  call <vscale x 16 x float> @llvm.round.nxv16f32(<vscale x 16 x float> undef)
  call double @llvm.round.f64(double undef)
  call <2 x double> @llvm.round.v2f64(<2 x double> undef)
  call <4 x double> @llvm.round.v4f64(<4 x double> undef)
  call <8 x double> @llvm.round.v8f64(<8 x double> undef)
  call <16 x double> @llvm.round.v16f64(<16 x double> undef)
  call <vscale x 1 x double> @llvm.round.nxv1f64(<vscale x 1 x double> undef)
  call <vscale x 2 x double> @llvm.round.nxv2f64(<vscale x 2 x double> undef)
  call <vscale x 4 x double> @llvm.round.nxv4f64(<vscale x 4 x double> undef)
  call <vscale x 8 x double> @llvm.round.nxv8f64(<vscale x 8 x double> undef)
  ret void
}

define void @round_fp16() {
; ZVFH-LABEL: 'round_fp16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.round.f16(half undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <2 x half> @llvm.round.v2f16(<2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <4 x half> @llvm.round.v4f16(<4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <8 x half> @llvm.round.v8f16(<8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x half> @llvm.round.v16f16(<16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 1 x half> @llvm.round.nxv1f16(<vscale x 1 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 2 x half> @llvm.round.nxv2f16(<vscale x 2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 4 x half> @llvm.round.nxv4f16(<vscale x 4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 8 x half> @llvm.round.nxv8f16(<vscale x 8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %10 = call <vscale x 16 x half> @llvm.round.nxv16f16(<vscale x 16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'round_fp16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call half @llvm.round.f16(half undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x half> @llvm.round.v2f16(<2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x half> @llvm.round.v4f16(<4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x half> @llvm.round.v8f16(<8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x half> @llvm.round.v16f16(<16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x half> @llvm.round.nxv1f16(<vscale x 1 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x half> @llvm.round.nxv2f16(<vscale x 2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x half> @llvm.round.nxv4f16(<vscale x 4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x half> @llvm.round.nxv8f16(<vscale x 8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x half> @llvm.round.nxv16f16(<vscale x 16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call half @llvm.round.f16(half undef)
  call <2 x half> @llvm.round.v2f16(<2 x half> undef)
  call <4 x half> @llvm.round.v4f16(<4 x half> undef)
  call <8 x half> @llvm.round.v8f16(<8 x half> undef)
  call <16 x half> @llvm.round.v16f16(<16 x half> undef)
  call <vscale x 1 x half> @llvm.round.nxv1f16(<vscale x 1 x half> undef)
  call <vscale x 2 x half> @llvm.round.nxv2f16(<vscale x 2 x half> undef)
  call <vscale x 4 x half> @llvm.round.nxv4f16(<vscale x 4 x half> undef)
  call <vscale x 8 x half> @llvm.round.nxv8f16(<vscale x 8 x half> undef)
  call <vscale x 16 x half> @llvm.round.nxv16f16(<vscale x 16 x half> undef)
  ret void
}

define void @roundeven() {
; CHECK-LABEL: 'roundeven'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.roundeven.bf16(bfloat undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.roundeven.v2bf16(<2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.roundeven.v4bf16(<4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.roundeven.v8bf16(<8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.roundeven.v16bf16(<16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.roundeven.nxv1bf16(<vscale x 1 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.roundeven.nxv2bf16(<vscale x 2 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.roundeven.nxv4bf16(<vscale x 4 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.roundeven.nxv8bf16(<vscale x 8 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.roundeven.nxv16bf16(<vscale x 16 x bfloat> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.roundeven.f32(float undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x float> @llvm.roundeven.v2f32(<2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x float> @llvm.roundeven.v4f32(<4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x float> @llvm.roundeven.v8f32(<8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x float> @llvm.roundeven.v16f32(<16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x float> @llvm.roundeven.nxv1f32(<vscale x 1 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x float> @llvm.roundeven.nxv2f32(<vscale x 2 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x float> @llvm.roundeven.nxv4f32(<vscale x 4 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x float> @llvm.roundeven.nxv8f32(<vscale x 8 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %20 = call <vscale x 16 x float> @llvm.roundeven.nxv16f32(<vscale x 16 x float> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.roundeven.f64(double undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %22 = call <2 x double> @llvm.roundeven.v2f64(<2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %23 = call <4 x double> @llvm.roundeven.v4f64(<4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %24 = call <8 x double> @llvm.roundeven.v8f64(<8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %25 = call <16 x double> @llvm.roundeven.v16f64(<16 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 1 x double> @llvm.roundeven.nxv1f64(<vscale x 1 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %27 = call <vscale x 2 x double> @llvm.roundeven.nxv2f64(<vscale x 2 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %28 = call <vscale x 4 x double> @llvm.roundeven.nxv4f64(<vscale x 4 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %29 = call <vscale x 8 x double> @llvm.roundeven.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call bfloat @llvm.roundeven.bf16(bfloat undef)
  call <2 x bfloat> @llvm.roundeven.v2bf16(<2 x bfloat> undef)
  call <4 x bfloat> @llvm.roundeven.v4bf16(<4 x bfloat> undef)
  call <8 x bfloat> @llvm.roundeven.v8bf16(<8 x bfloat> undef)
  call <16 x bfloat> @llvm.roundeven.v16bf16(<16 x bfloat> undef)
  call <vscale x 1 x bfloat> @llvm.roundeven.nxv1bf16(<vscale x 1 x bfloat> undef)
  call <vscale x 2 x bfloat> @llvm.roundeven.nxv2bf16(<vscale x 2 x bfloat> undef)
  call <vscale x 4 x bfloat> @llvm.roundeven.nxv4bf16(<vscale x 4 x bfloat> undef)
  call <vscale x 8 x bfloat> @llvm.roundeven.nxv8bf16(<vscale x 8 x bfloat> undef)
  call <vscale x 16 x bfloat> @llvm.roundeven.nxv16bf16(<vscale x 16 x bfloat> undef)
  call float @llvm.roundeven.f32(float undef)
  call <2 x float> @llvm.roundeven.v2f32(<2 x float> undef)
  call <4 x float> @llvm.roundeven.v4f32(<4 x float> undef)
  call <8 x float> @llvm.roundeven.v8f32(<8 x float> undef)
  call <16 x float> @llvm.roundeven.v16f32(<16 x float> undef)
  call <vscale x 1 x float> @llvm.roundeven.nxv1f32(<vscale x 1 x float> undef)
  call <vscale x 2 x float> @llvm.roundeven.nxv2f32(<vscale x 2 x float> undef)
  call <vscale x 4 x float> @llvm.roundeven.nxv4f32(<vscale x 4 x float> undef)
  call <vscale x 8 x float> @llvm.roundeven.nxv8f32(<vscale x 8 x float> undef)
  call <vscale x 16 x float> @llvm.roundeven.nxv16f32(<vscale x 16 x float> undef)
  call double @llvm.roundeven.f64(double undef)
  call <2 x double> @llvm.roundeven.v2f64(<2 x double> undef)
  call <4 x double> @llvm.roundeven.v4f64(<4 x double> undef)
  call <8 x double> @llvm.roundeven.v8f64(<8 x double> undef)
  call <16 x double> @llvm.roundeven.v16f64(<16 x double> undef)
  call <vscale x 1 x double> @llvm.roundeven.nxv1f64(<vscale x 1 x double> undef)
  call <vscale x 2 x double> @llvm.roundeven.nxv2f64(<vscale x 2 x double> undef)
  call <vscale x 4 x double> @llvm.roundeven.nxv4f64(<vscale x 4 x double> undef)
  call <vscale x 8 x double> @llvm.roundeven.nxv8f64(<vscale x 8 x double> undef)
  ret void
}

define void @roundeven_fp16() {
; ZVFH-LABEL: 'roundeven_fp16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.roundeven.f16(half undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <2 x half> @llvm.roundeven.v2f16(<2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <4 x half> @llvm.roundeven.v4f16(<4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <8 x half> @llvm.roundeven.v8f16(<8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x half> @llvm.roundeven.v16f16(<16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 1 x half> @llvm.roundeven.nxv1f16(<vscale x 1 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 2 x half> @llvm.roundeven.nxv2f16(<vscale x 2 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 4 x half> @llvm.roundeven.nxv4f16(<vscale x 4 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 8 x half> @llvm.roundeven.nxv8f16(<vscale x 8 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %10 = call <vscale x 16 x half> @llvm.roundeven.nxv16f16(<vscale x 16 x half> undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'roundeven_fp16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %1 = call half @llvm.roundeven.f16(half undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x half> @llvm.roundeven.v2f16(<2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x half> @llvm.roundeven.v4f16(<4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x half> @llvm.roundeven.v8f16(<8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x half> @llvm.roundeven.v16f16(<16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x half> @llvm.roundeven.nxv1f16(<vscale x 1 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x half> @llvm.roundeven.nxv2f16(<vscale x 2 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x half> @llvm.roundeven.nxv4f16(<vscale x 4 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x half> @llvm.roundeven.nxv8f16(<vscale x 8 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x half> @llvm.roundeven.nxv16f16(<vscale x 16 x half> undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call half @llvm.roundeven.f16(half undef)
  call <2 x half> @llvm.roundeven.v2f16(<2 x half> undef)
  call <4 x half> @llvm.roundeven.v4f16(<4 x half> undef)
  call <8 x half> @llvm.roundeven.v8f16(<8 x half> undef)
  call <16 x half> @llvm.roundeven.v16f16(<16 x half> undef)
  call <vscale x 1 x half> @llvm.roundeven.nxv1f16(<vscale x 1 x half> undef)
  call <vscale x 2 x half> @llvm.roundeven.nxv2f16(<vscale x 2 x half> undef)
  call <vscale x 4 x half> @llvm.roundeven.nxv4f16(<vscale x 4 x half> undef)
  call <vscale x 8 x half> @llvm.roundeven.nxv8f16(<vscale x 8 x half> undef)
  call <vscale x 16 x half> @llvm.roundeven.nxv16f16(<vscale x 16 x half> undef)
  ret void
}

define void @vp_ceil() {
; CHECK-LABEL: 'vp_ceil'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.ceil.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.ceil.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.ceil.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.ceil.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.ceil.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.ceil.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.ceil.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.ceil.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.ceil.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %10 = call <2 x float> @llvm.vp.ceil.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %11 = call <4 x float> @llvm.vp.ceil.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %12 = call <8 x float> @llvm.vp.ceil.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %13 = call <16 x float> @llvm.vp.ceil.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.ceil.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.ceil.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.ceil.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.ceil.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.ceil.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %19 = call <2 x double> @llvm.vp.ceil.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %20 = call <4 x double> @llvm.vp.ceil.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %21 = call <8 x double> @llvm.vp.ceil.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %22 = call <16 x double> @llvm.vp.ceil.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.ceil.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.ceil.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.ceil.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.ceil.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x bfloat> @llvm.vp.ceil.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
  call <4 x bfloat> @llvm.vp.ceil.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
  call <8 x bfloat> @llvm.vp.ceil.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
  call <16 x bfloat> @llvm.vp.ceil.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x bfloat> @llvm.vp.ceil.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x bfloat> @llvm.vp.ceil.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x bfloat> @llvm.vp.ceil.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x bfloat> @llvm.vp.ceil.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x bfloat> @llvm.vp.ceil.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x float> @llvm.vp.ceil.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  call <4 x float> @llvm.vp.ceil.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  call <8 x float> @llvm.vp.ceil.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  call <16 x float> @llvm.vp.ceil.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x float> @llvm.vp.ceil.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x float> @llvm.vp.ceil.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x float> @llvm.vp.ceil.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x float> @llvm.vp.ceil.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x float> @llvm.vp.ceil.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x double> @llvm.vp.ceil.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  call <4 x double> @llvm.vp.ceil.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  call <8 x double> @llvm.vp.ceil.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  call <16 x double> @llvm.vp.ceil.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x double> @llvm.vp.ceil.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x double> @llvm.vp.ceil.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x double> @llvm.vp.ceil.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x double> @llvm.vp.ceil.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  ret void
}

define void @vp_ceil_f16() {
; ZVFH-LABEL: 'vp_ceil_f16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call <2 x half> @llvm.vp.ceil.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <4 x half> @llvm.vp.ceil.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <8 x half> @llvm.vp.ceil.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <16 x half> @llvm.vp.ceil.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.ceil.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.ceil.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.ceil.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.ceil.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.ceil.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'vp_ceil_f16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x half> @llvm.vp.ceil.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x half> @llvm.vp.ceil.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x half> @llvm.vp.ceil.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x half> @llvm.vp.ceil.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.ceil.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.ceil.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.ceil.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.ceil.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.ceil.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x half> @llvm.vp.ceil.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
  call <4 x half> @llvm.vp.ceil.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
  call <8 x half> @llvm.vp.ceil.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
  call <16 x half> @llvm.vp.ceil.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x half> @llvm.vp.ceil.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x half> @llvm.vp.ceil.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x half> @llvm.vp.ceil.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x half> @llvm.vp.ceil.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x half> @llvm.vp.ceil.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
  ret void
}

define void @vp_floor() {
; CHECK-LABEL: 'vp_floor'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.floor.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.floor.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.floor.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.floor.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.floor.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.floor.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.floor.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.floor.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.floor.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %10 = call <2 x float> @llvm.vp.floor.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %11 = call <4 x float> @llvm.vp.floor.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %12 = call <8 x float> @llvm.vp.floor.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %13 = call <16 x float> @llvm.vp.floor.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.floor.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.floor.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.floor.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.floor.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.floor.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %19 = call <2 x double> @llvm.vp.floor.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %20 = call <4 x double> @llvm.vp.floor.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %21 = call <8 x double> @llvm.vp.floor.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %22 = call <16 x double> @llvm.vp.floor.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.floor.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.floor.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.floor.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.floor.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x bfloat> @llvm.vp.floor.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
  call <4 x bfloat> @llvm.vp.floor.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
  call <8 x bfloat> @llvm.vp.floor.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
  call <16 x bfloat> @llvm.vp.floor.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x bfloat> @llvm.vp.floor.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x bfloat> @llvm.vp.floor.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x bfloat> @llvm.vp.floor.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x bfloat> @llvm.vp.floor.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x bfloat> @llvm.vp.floor.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x float> @llvm.vp.floor.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  call <4 x float> @llvm.vp.floor.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  call <8 x float> @llvm.vp.floor.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  call <16 x float> @llvm.vp.floor.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x float> @llvm.vp.floor.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x float> @llvm.vp.floor.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x float> @llvm.vp.floor.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x float> @llvm.vp.floor.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x float> @llvm.vp.floor.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x double> @llvm.vp.floor.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  call <4 x double> @llvm.vp.floor.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  call <8 x double> @llvm.vp.floor.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  call <16 x double> @llvm.vp.floor.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x double> @llvm.vp.floor.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x double> @llvm.vp.floor.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x double> @llvm.vp.floor.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x double> @llvm.vp.floor.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  ret void
}

define void @vp_floor_f16() {
; ZVFH-LABEL: 'vp_floor_f16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call <2 x half> @llvm.vp.floor.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <4 x half> @llvm.vp.floor.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <8 x half> @llvm.vp.floor.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <16 x half> @llvm.vp.floor.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.floor.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.floor.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.floor.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.floor.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.floor.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'vp_floor_f16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x half> @llvm.vp.floor.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x half> @llvm.vp.floor.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x half> @llvm.vp.floor.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x half> @llvm.vp.floor.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.floor.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.floor.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.floor.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.floor.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.floor.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x half> @llvm.vp.floor.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
  call <4 x half> @llvm.vp.floor.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
  call <8 x half> @llvm.vp.floor.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
  call <16 x half> @llvm.vp.floor.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x half> @llvm.vp.floor.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x half> @llvm.vp.floor.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x half> @llvm.vp.floor.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x half> @llvm.vp.floor.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x half> @llvm.vp.floor.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
  ret void
}

define void @vp_round() {
; CHECK-LABEL: 'vp_round'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.round.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.round.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.round.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.round.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.round.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.round.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.round.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.round.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.round.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %10 = call <2 x float> @llvm.vp.round.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %11 = call <4 x float> @llvm.vp.round.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %12 = call <8 x float> @llvm.vp.round.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %13 = call <16 x float> @llvm.vp.round.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.round.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.round.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.round.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.round.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.round.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %19 = call <2 x double> @llvm.vp.round.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %20 = call <4 x double> @llvm.vp.round.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %21 = call <8 x double> @llvm.vp.round.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %22 = call <16 x double> @llvm.vp.round.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.round.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.round.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.round.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.round.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x bfloat> @llvm.vp.round.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
  call <4 x bfloat> @llvm.vp.round.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
  call <8 x bfloat> @llvm.vp.round.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
  call <16 x bfloat> @llvm.vp.round.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x bfloat> @llvm.vp.round.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x bfloat> @llvm.vp.round.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x bfloat> @llvm.vp.round.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x bfloat> @llvm.vp.round.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x bfloat> @llvm.vp.round.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x float> @llvm.vp.round.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  call <4 x float> @llvm.vp.round.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  call <8 x float> @llvm.vp.round.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  call <16 x float> @llvm.vp.round.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x float> @llvm.vp.round.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x float> @llvm.vp.round.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x float> @llvm.vp.round.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x float> @llvm.vp.round.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x float> @llvm.vp.round.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x double> @llvm.vp.round.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  call <4 x double> @llvm.vp.round.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  call <8 x double> @llvm.vp.round.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  call <16 x double> @llvm.vp.round.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x double> @llvm.vp.round.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x double> @llvm.vp.round.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x double> @llvm.vp.round.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x double> @llvm.vp.round.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  ret void
}

define void @vp_round_f16() {
; ZVFH-LABEL: 'vp_round_f16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call <2 x half> @llvm.vp.round.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <4 x half> @llvm.vp.round.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <8 x half> @llvm.vp.round.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <16 x half> @llvm.vp.round.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.round.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.round.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.round.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.round.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.round.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'vp_round_f16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x half> @llvm.vp.round.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x half> @llvm.vp.round.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x half> @llvm.vp.round.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x half> @llvm.vp.round.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.round.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.round.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.round.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.round.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.round.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x half> @llvm.vp.round.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
  call <4 x half> @llvm.vp.round.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
  call <8 x half> @llvm.vp.round.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
  call <16 x half> @llvm.vp.round.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x half> @llvm.vp.round.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x half> @llvm.vp.round.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x half> @llvm.vp.round.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x half> @llvm.vp.round.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x half> @llvm.vp.round.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
  ret void
}

define void @vp_roundeven() {
; CHECK-LABEL: 'vp_roundeven'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.roundeven.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.roundeven.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.roundeven.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.roundeven.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.roundeven.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.roundeven.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.roundeven.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.roundeven.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.roundeven.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %10 = call <2 x float> @llvm.vp.roundeven.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %11 = call <4 x float> @llvm.vp.roundeven.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %12 = call <8 x float> @llvm.vp.roundeven.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %13 = call <16 x float> @llvm.vp.roundeven.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.roundeven.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.roundeven.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.roundeven.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.roundeven.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.roundeven.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %19 = call <2 x double> @llvm.vp.roundeven.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %20 = call <4 x double> @llvm.vp.roundeven.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %21 = call <8 x double> @llvm.vp.roundeven.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %22 = call <16 x double> @llvm.vp.roundeven.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.roundeven.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.roundeven.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.roundeven.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.roundeven.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x bfloat> @llvm.vp.roundeven.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
  call <4 x bfloat> @llvm.vp.roundeven.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
  call <8 x bfloat> @llvm.vp.roundeven.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
  call <16 x bfloat> @llvm.vp.roundeven.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x bfloat> @llvm.vp.roundeven.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x bfloat> @llvm.vp.roundeven.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x bfloat> @llvm.vp.roundeven.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x bfloat> @llvm.vp.roundeven.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x bfloat> @llvm.vp.roundeven.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x float> @llvm.vp.roundeven.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  call <4 x float> @llvm.vp.roundeven.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  call <8 x float> @llvm.vp.roundeven.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  call <16 x float> @llvm.vp.roundeven.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x float> @llvm.vp.roundeven.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x float> @llvm.vp.roundeven.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x float> @llvm.vp.roundeven.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x float> @llvm.vp.roundeven.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x float> @llvm.vp.roundeven.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x double> @llvm.vp.roundeven.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  call <4 x double> @llvm.vp.roundeven.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  call <8 x double> @llvm.vp.roundeven.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  call <16 x double> @llvm.vp.roundeven.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x double> @llvm.vp.roundeven.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x double> @llvm.vp.roundeven.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x double> @llvm.vp.roundeven.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x double> @llvm.vp.roundeven.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  ret void
}

define void @vp_roundeven_f16() {
; ZVFH-LABEL: 'vp_roundeven_f16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call <2 x half> @llvm.vp.roundeven.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <4 x half> @llvm.vp.roundeven.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <8 x half> @llvm.vp.roundeven.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <16 x half> @llvm.vp.roundeven.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.roundeven.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.roundeven.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.roundeven.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.roundeven.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.roundeven.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'vp_roundeven_f16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x half> @llvm.vp.roundeven.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x half> @llvm.vp.roundeven.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x half> @llvm.vp.roundeven.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x half> @llvm.vp.roundeven.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.roundeven.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.roundeven.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.roundeven.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.roundeven.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.roundeven.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x half> @llvm.vp.roundeven.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
  call <4 x half> @llvm.vp.roundeven.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
  call <8 x half> @llvm.vp.roundeven.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
  call <16 x half> @llvm.vp.roundeven.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x half> @llvm.vp.roundeven.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x half> @llvm.vp.roundeven.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x half> @llvm.vp.roundeven.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x half> @llvm.vp.roundeven.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x half> @llvm.vp.roundeven.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
  ret void
}

define void @vp_roundtozero() {
; CHECK-LABEL: 'vp_roundtozero'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.roundtozero.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.roundtozero.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.roundtozero.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.roundtozero.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.roundtozero.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.roundtozero.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.roundtozero.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.roundtozero.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.roundtozero.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x float> @llvm.vp.roundtozero.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x float> @llvm.vp.roundtozero.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x float> @llvm.vp.roundtozero.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x float> @llvm.vp.roundtozero.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.roundtozero.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.roundtozero.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.roundtozero.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.roundtozero.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.roundtozero.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %19 = call <2 x double> @llvm.vp.roundtozero.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %20 = call <4 x double> @llvm.vp.roundtozero.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %21 = call <8 x double> @llvm.vp.roundtozero.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %22 = call <16 x double> @llvm.vp.roundtozero.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.roundtozero.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.roundtozero.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.roundtozero.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.roundtozero.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x bfloat> @llvm.vp.roundtozero.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
  call <4 x bfloat> @llvm.vp.roundtozero.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
  call <8 x bfloat> @llvm.vp.roundtozero.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
  call <16 x bfloat> @llvm.vp.roundtozero.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x bfloat> @llvm.vp.roundtozero.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x bfloat> @llvm.vp.roundtozero.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x bfloat> @llvm.vp.roundtozero.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x bfloat> @llvm.vp.roundtozero.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x bfloat> @llvm.vp.roundtozero.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x float> @llvm.vp.roundtozero.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  call <4 x float> @llvm.vp.roundtozero.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  call <8 x float> @llvm.vp.roundtozero.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  call <16 x float> @llvm.vp.roundtozero.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x float> @llvm.vp.roundtozero.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x float> @llvm.vp.roundtozero.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x float> @llvm.vp.roundtozero.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x float> @llvm.vp.roundtozero.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x float> @llvm.vp.roundtozero.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x double> @llvm.vp.roundtozero.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  call <4 x double> @llvm.vp.roundtozero.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  call <8 x double> @llvm.vp.roundtozero.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  call <16 x double> @llvm.vp.roundtozero.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x double> @llvm.vp.roundtozero.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x double> @llvm.vp.roundtozero.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x double> @llvm.vp.roundtozero.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x double> @llvm.vp.roundtozero.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  ret void
}

define void @vp_roundtozero_f16() {
; ZVFH-LABEL: 'vp_roundtozero_f16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call <2 x half> @llvm.vp.roundtozero.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <4 x half> @llvm.vp.roundtozero.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <8 x half> @llvm.vp.roundtozero.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <16 x half> @llvm.vp.roundtozero.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.roundtozero.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.roundtozero.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.roundtozero.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.roundtozero.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.roundtozero.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'vp_roundtozero_f16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x half> @llvm.vp.roundtozero.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x half> @llvm.vp.roundtozero.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x half> @llvm.vp.roundtozero.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x half> @llvm.vp.roundtozero.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.roundtozero.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.roundtozero.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.roundtozero.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.roundtozero.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.roundtozero.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x half> @llvm.vp.roundtozero.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
  call <4 x half> @llvm.vp.roundtozero.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
  call <8 x half> @llvm.vp.roundtozero.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
  call <16 x half> @llvm.vp.roundtozero.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x half> @llvm.vp.roundtozero.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x half> @llvm.vp.roundtozero.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x half> @llvm.vp.roundtozero.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x half> @llvm.vp.roundtozero.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x half> @llvm.vp.roundtozero.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
  ret void
}

define void @vp_rint() {
; CHECK-LABEL: 'vp_rint'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.rint.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.rint.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.rint.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.rint.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.rint.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.rint.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.rint.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.rint.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.rint.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x float> @llvm.vp.rint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x float> @llvm.vp.rint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x float> @llvm.vp.rint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x float> @llvm.vp.rint.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.rint.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.rint.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.rint.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.rint.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.rint.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %19 = call <2 x double> @llvm.vp.rint.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %20 = call <4 x double> @llvm.vp.rint.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %21 = call <8 x double> @llvm.vp.rint.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %22 = call <16 x double> @llvm.vp.rint.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.rint.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.rint.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.rint.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.rint.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x bfloat> @llvm.vp.rint.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
  call <4 x bfloat> @llvm.vp.rint.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
  call <8 x bfloat> @llvm.vp.rint.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
  call <16 x bfloat> @llvm.vp.rint.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x bfloat> @llvm.vp.rint.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x bfloat> @llvm.vp.rint.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x bfloat> @llvm.vp.rint.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x bfloat> @llvm.vp.rint.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x bfloat> @llvm.vp.rint.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x float> @llvm.vp.rint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  call <4 x float> @llvm.vp.rint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  call <8 x float> @llvm.vp.rint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  call <16 x float> @llvm.vp.rint.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x float> @llvm.vp.rint.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x float> @llvm.vp.rint.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x float> @llvm.vp.rint.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x float> @llvm.vp.rint.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x float> @llvm.vp.rint.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x double> @llvm.vp.rint.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  call <4 x double> @llvm.vp.rint.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  call <8 x double> @llvm.vp.rint.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  call <16 x double> @llvm.vp.rint.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x double> @llvm.vp.rint.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x double> @llvm.vp.rint.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x double> @llvm.vp.rint.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x double> @llvm.vp.rint.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  ret void
}

define void @vp_rint_f16() {
; ZVFH-LABEL: 'vp_rint_f16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call <2 x half> @llvm.vp.rint.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <4 x half> @llvm.vp.rint.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <8 x half> @llvm.vp.rint.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <16 x half> @llvm.vp.rint.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.rint.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.rint.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.rint.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.rint.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.rint.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'vp_rint_f16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x half> @llvm.vp.rint.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x half> @llvm.vp.rint.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x half> @llvm.vp.rint.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x half> @llvm.vp.rint.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.rint.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.rint.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.rint.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.rint.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.rint.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x half> @llvm.vp.rint.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
  call <4 x half> @llvm.vp.rint.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
  call <8 x half> @llvm.vp.rint.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
  call <16 x half> @llvm.vp.rint.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x half> @llvm.vp.rint.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x half> @llvm.vp.rint.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x half> @llvm.vp.rint.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x half> @llvm.vp.rint.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x half> @llvm.vp.rint.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
  ret void
}

define void @vp_nearbyint() {
; CHECK-LABEL: 'vp_nearbyint'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.nearbyint.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.nearbyint.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.nearbyint.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.nearbyint.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.nearbyint.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.nearbyint.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.nearbyint.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.nearbyint.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.nearbyint.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %10 = call <2 x float> @llvm.vp.nearbyint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %11 = call <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %12 = call <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %13 = call <16 x float> @llvm.vp.nearbyint.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.nearbyint.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.nearbyint.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.nearbyint.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.nearbyint.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.nearbyint.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %19 = call <2 x double> @llvm.vp.nearbyint.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %20 = call <4 x double> @llvm.vp.nearbyint.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %21 = call <8 x double> @llvm.vp.nearbyint.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %22 = call <16 x double> @llvm.vp.nearbyint.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.nearbyint.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.nearbyint.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.nearbyint.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.nearbyint.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x bfloat> @llvm.vp.nearbyint.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
  call <4 x bfloat> @llvm.vp.nearbyint.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
  call <8 x bfloat> @llvm.vp.nearbyint.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
  call <16 x bfloat> @llvm.vp.nearbyint.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x bfloat> @llvm.vp.nearbyint.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x bfloat> @llvm.vp.nearbyint.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x bfloat> @llvm.vp.nearbyint.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x bfloat> @llvm.vp.nearbyint.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x bfloat> @llvm.vp.nearbyint.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x float> @llvm.vp.nearbyint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  call <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  call <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  call <16 x float> @llvm.vp.nearbyint.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x float> @llvm.vp.nearbyint.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x float> @llvm.vp.nearbyint.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x float> @llvm.vp.nearbyint.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x float> @llvm.vp.nearbyint.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x float> @llvm.vp.nearbyint.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  call <2 x double> @llvm.vp.nearbyint.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  call <4 x double> @llvm.vp.nearbyint.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  call <8 x double> @llvm.vp.nearbyint.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  call <16 x double> @llvm.vp.nearbyint.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x double> @llvm.vp.nearbyint.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x double> @llvm.vp.nearbyint.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x double> @llvm.vp.nearbyint.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x double> @llvm.vp.nearbyint.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  ret void
}

define void @vp_nearbyint_f16() {
; ZVFH-LABEL: 'vp_nearbyint_f16'
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call <2 x half> @llvm.vp.nearbyint.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call <4 x half> @llvm.vp.nearbyint.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <8 x half> @llvm.vp.nearbyint.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <16 x half> @llvm.vp.nearbyint.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.nearbyint.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.nearbyint.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.nearbyint.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.nearbyint.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.nearbyint.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVFHMIN-LABEL: 'vp_nearbyint_f16'
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x half> @llvm.vp.nearbyint.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x half> @llvm.vp.nearbyint.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x half> @llvm.vp.nearbyint.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x half> @llvm.vp.nearbyint.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x half> @llvm.vp.nearbyint.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x half> @llvm.vp.nearbyint.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x half> @llvm.vp.nearbyint.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x half> @llvm.vp.nearbyint.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x half> @llvm.vp.nearbyint.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  call <2 x half> @llvm.vp.nearbyint.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
  call <4 x half> @llvm.vp.nearbyint.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
  call <8 x half> @llvm.vp.nearbyint.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
  call <16 x half> @llvm.vp.nearbyint.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
  call <vscale x 1 x half> @llvm.vp.nearbyint.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
  call <vscale x 2 x half> @llvm.vp.nearbyint.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
  call <vscale x 4 x half> @llvm.vp.nearbyint.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
  call <vscale x 8 x half> @llvm.vp.nearbyint.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
  call <vscale x 16 x half> @llvm.vp.nearbyint.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
  ret void
}