// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
// RUN: %clang_cc1 -no-opaque-pointers -triple aarch64-none-linux-gnu -target-feature +sve2 -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -no-opaque-pointers -triple aarch64-none-linux-gnu -target-feature +sve2 -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - -x c++ %s | FileCheck %s -check-prefix=CPP-CHECK
// RUN: %clang_cc1 -no-opaque-pointers -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve2 -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -no-opaque-pointers -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve2 -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - -x c++ %s | FileCheck %s -check-prefix=CPP-CHECK

// REQUIRES: aarch64-registered-target

#include <arm_sve.h>

#ifdef SVE_OVERLOADED_FORMS
// A simple used,unused... macro, long enough to represent any SVE builtin.
#define SVE_ACLE_FUNC(A1,A2_UNUSED,A3,A4_UNUSED) A1##A3
#else
#define SVE_ACLE_FUNC(A1,A2,A3,A4) A1##A2##A3##A4
#endif

// CHECK-LABEL: @test_svwhilerw_s8(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.whilerw.b.nxv16i1.p0i8(i8* [[OP1:%.*]], i8* [[OP2:%.*]])
// CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z17test_svwhilerw_s8PKaS0_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.whilerw.b.nxv16i1.p0i8(i8* [[OP1:%.*]], i8* [[OP2:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP0]]
//
svbool_t test_svwhilerw_s8(const int8_t *op1, const int8_t *op2)
{
  return SVE_ACLE_FUNC(svwhilerw,_s8,,)(op1, op2);
}

// CHECK-LABEL: @test_svwhilerw_s16(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.whilerw.h.nxv8i1.p0i16(i16* [[OP1:%.*]], i16* [[OP2:%.*]])
// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv8i1(<vscale x 8 x i1> [[TMP0]])
// CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z18test_svwhilerw_s16PKsS0_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.whilerw.h.nxv8i1.p0i16(i16* [[OP1:%.*]], i16* [[OP2:%.*]])
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv8i1(<vscale x 8 x i1> [[TMP0]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
svbool_t test_svwhilerw_s16(const int16_t *op1, const int16_t *op2)
{
  return SVE_ACLE_FUNC(svwhilerw,_s16,,)(op1, op2);
}

// CHECK-LABEL: @test_svwhilerw_s32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.whilerw.s.nxv4i1.p0i32(i32* [[OP1:%.*]], i32* [[OP2:%.*]])
// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv4i1(<vscale x 4 x i1> [[TMP0]])
// CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z18test_svwhilerw_s32PKiS0_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.whilerw.s.nxv4i1.p0i32(i32* [[OP1:%.*]], i32* [[OP2:%.*]])
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv4i1(<vscale x 4 x i1> [[TMP0]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
svbool_t test_svwhilerw_s32(const int32_t *op1, const int32_t *op2)
{
  return SVE_ACLE_FUNC(svwhilerw,_s32,,)(op1, op2);
}

// CHECK-LABEL: @test_svwhilerw_s64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.whilerw.d.nxv2i1.p0i64(i64* [[OP1:%.*]], i64* [[OP2:%.*]])
// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv2i1(<vscale x 2 x i1> [[TMP0]])
// CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z18test_svwhilerw_s64PKlS0_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.whilerw.d.nxv2i1.p0i64(i64* [[OP1:%.*]], i64* [[OP2:%.*]])
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv2i1(<vscale x 2 x i1> [[TMP0]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
svbool_t test_svwhilerw_s64(const int64_t *op1, const int64_t *op2)
{
  return SVE_ACLE_FUNC(svwhilerw,_s64,,)(op1, op2);
}

// CHECK-LABEL: @test_svwhilerw_u8(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.whilerw.b.nxv16i1.p0i8(i8* [[OP1:%.*]], i8* [[OP2:%.*]])
// CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z17test_svwhilerw_u8PKhS0_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.whilerw.b.nxv16i1.p0i8(i8* [[OP1:%.*]], i8* [[OP2:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP0]]
//
svbool_t test_svwhilerw_u8(const uint8_t *op1, const uint8_t *op2)
{
  return SVE_ACLE_FUNC(svwhilerw,_u8,,)(op1, op2);
}

// CHECK-LABEL: @test_svwhilerw_u16(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.whilerw.h.nxv8i1.p0i16(i16* [[OP1:%.*]], i16* [[OP2:%.*]])
// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv8i1(<vscale x 8 x i1> [[TMP0]])
// CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z18test_svwhilerw_u16PKtS0_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.whilerw.h.nxv8i1.p0i16(i16* [[OP1:%.*]], i16* [[OP2:%.*]])
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv8i1(<vscale x 8 x i1> [[TMP0]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
svbool_t test_svwhilerw_u16(const uint16_t *op1, const uint16_t *op2)
{
  return SVE_ACLE_FUNC(svwhilerw,_u16,,)(op1, op2);
}

// CHECK-LABEL: @test_svwhilerw_u32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.whilerw.s.nxv4i1.p0i32(i32* [[OP1:%.*]], i32* [[OP2:%.*]])
// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv4i1(<vscale x 4 x i1> [[TMP0]])
// CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z18test_svwhilerw_u32PKjS0_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.whilerw.s.nxv4i1.p0i32(i32* [[OP1:%.*]], i32* [[OP2:%.*]])
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv4i1(<vscale x 4 x i1> [[TMP0]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
svbool_t test_svwhilerw_u32(const uint32_t *op1, const uint32_t *op2)
{
  return SVE_ACLE_FUNC(svwhilerw,_u32,,)(op1, op2);
}

// CHECK-LABEL: @test_svwhilerw_u64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.whilerw.d.nxv2i1.p0i64(i64* [[OP1:%.*]], i64* [[OP2:%.*]])
// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv2i1(<vscale x 2 x i1> [[TMP0]])
// CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z18test_svwhilerw_u64PKmS0_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.whilerw.d.nxv2i1.p0i64(i64* [[OP1:%.*]], i64* [[OP2:%.*]])
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv2i1(<vscale x 2 x i1> [[TMP0]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
svbool_t test_svwhilerw_u64(const uint64_t *op1, const uint64_t *op2)
{
  return SVE_ACLE_FUNC(svwhilerw,_u64,,)(op1, op2);
}

// CHECK-LABEL: @test_svwhilerw_f16(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.whilerw.h.nxv8i1.p0f16(half* [[OP1:%.*]], half* [[OP2:%.*]])
// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv8i1(<vscale x 8 x i1> [[TMP0]])
// CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z18test_svwhilerw_f16PKDhS0_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.whilerw.h.nxv8i1.p0f16(half* [[OP1:%.*]], half* [[OP2:%.*]])
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv8i1(<vscale x 8 x i1> [[TMP0]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
svbool_t test_svwhilerw_f16(const float16_t *op1, const float16_t *op2)
{
  return SVE_ACLE_FUNC(svwhilerw,_f16,,)(op1, op2);
}

// CHECK-LABEL: @test_svwhilerw_f32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.whilerw.s.nxv4i1.p0f32(float* [[OP1:%.*]], float* [[OP2:%.*]])
// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv4i1(<vscale x 4 x i1> [[TMP0]])
// CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z18test_svwhilerw_f32PKfS0_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.whilerw.s.nxv4i1.p0f32(float* [[OP1:%.*]], float* [[OP2:%.*]])
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv4i1(<vscale x 4 x i1> [[TMP0]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
svbool_t test_svwhilerw_f32(const float32_t *op1, const float32_t *op2)
{
  return SVE_ACLE_FUNC(svwhilerw,_f32,,)(op1, op2);
}

// CHECK-LABEL: @test_svwhilerw_f64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.whilerw.d.nxv2i1.p0f64(double* [[OP1:%.*]], double* [[OP2:%.*]])
// CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv2i1(<vscale x 2 x i1> [[TMP0]])
// CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z18test_svwhilerw_f64PKdS0_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.whilerw.d.nxv2i1.p0f64(double* [[OP1:%.*]], double* [[OP2:%.*]])
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <vscale x 16 x i1> @llvm.aarch64.sve.convert.to.svbool.nxv2i1(<vscale x 2 x i1> [[TMP0]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i1> [[TMP1]]
//
svbool_t test_svwhilerw_f64(const float64_t *op1, const float64_t *op2)
{
  return SVE_ACLE_FUNC(svwhilerw,_f64,,)(op1, op2);
}