// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
// RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64-none-linux-gnu -target-feature +sve2 -S -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64-none-linux-gnu -target-feature +sve2 -S -O1 -Werror -Wall -emit-llvm -o - -x c++ %s | FileCheck %s -check-prefix=CPP-CHECK
// RUN: %clang_cc1 -fclang-abi-compat=latest -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve2 -S -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -fclang-abi-compat=latest -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve2 -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_svtbx_s8(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.tbx.nxv16i8(<vscale x 16 x i8> [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]], <vscale x 16 x i8> [[INDICES:%.*]])
// CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z13test_svtbx_s8u10__SVInt8_tS_u11__SVUint8_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.tbx.nxv16i8(<vscale x 16 x i8> [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]], <vscale x 16 x i8> [[INDICES:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
//
svint8_t test_svtbx_s8(svint8_t fallback, svint8_t data, svuint8_t indices)
{
  return SVE_ACLE_FUNC(svtbx,_s8,,)(fallback, data, indices);
}

// CHECK-LABEL: @test_svtbx_s16(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.tbx.nxv8i16(<vscale x 8 x i16> [[FALLBACK:%.*]], <vscale x 8 x i16> [[DATA:%.*]], <vscale x 8 x i16> [[INDICES:%.*]])
// CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z14test_svtbx_s16u11__SVInt16_tS_u12__SVUint16_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.tbx.nxv8i16(<vscale x 8 x i16> [[FALLBACK:%.*]], <vscale x 8 x i16> [[DATA:%.*]], <vscale x 8 x i16> [[INDICES:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
//
svint16_t test_svtbx_s16(svint16_t fallback, svint16_t data, svuint16_t indices)
{
  return SVE_ACLE_FUNC(svtbx,_s16,,)(fallback, data, indices);
}

// CHECK-LABEL: @test_svtbx_s32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.tbx.nxv4i32(<vscale x 4 x i32> [[FALLBACK:%.*]], <vscale x 4 x i32> [[DATA:%.*]], <vscale x 4 x i32> [[INDICES:%.*]])
// CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z14test_svtbx_s32u11__SVInt32_tS_u12__SVUint32_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.tbx.nxv4i32(<vscale x 4 x i32> [[FALLBACK:%.*]], <vscale x 4 x i32> [[DATA:%.*]], <vscale x 4 x i32> [[INDICES:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
//
svint32_t test_svtbx_s32(svint32_t fallback, svint32_t data, svuint32_t indices)
{
  return SVE_ACLE_FUNC(svtbx,_s32,,)(fallback, data, indices);
}

// CHECK-LABEL: @test_svtbx_s64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.tbx.nxv2i64(<vscale x 2 x i64> [[FALLBACK:%.*]], <vscale x 2 x i64> [[DATA:%.*]], <vscale x 2 x i64> [[INDICES:%.*]])
// CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z14test_svtbx_s64u11__SVInt64_tS_u12__SVUint64_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.tbx.nxv2i64(<vscale x 2 x i64> [[FALLBACK:%.*]], <vscale x 2 x i64> [[DATA:%.*]], <vscale x 2 x i64> [[INDICES:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
//
svint64_t test_svtbx_s64(svint64_t fallback, svint64_t data, svuint64_t indices)
{
  return SVE_ACLE_FUNC(svtbx,_s64,,)(fallback, data, indices);
}

// CHECK-LABEL: @test_svtbx_u8(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.tbx.nxv16i8(<vscale x 16 x i8> [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]], <vscale x 16 x i8> [[INDICES:%.*]])
// CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z13test_svtbx_u8u11__SVUint8_tS_S_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.tbx.nxv16i8(<vscale x 16 x i8> [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]], <vscale x 16 x i8> [[INDICES:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 16 x i8> [[TMP0]]
//
svuint8_t test_svtbx_u8(svuint8_t fallback, svuint8_t data, svuint8_t indices)
{
  return SVE_ACLE_FUNC(svtbx,_u8,,)(fallback, data, indices);
}

// CHECK-LABEL: @test_svtbx_u16(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.tbx.nxv8i16(<vscale x 8 x i16> [[FALLBACK:%.*]], <vscale x 8 x i16> [[DATA:%.*]], <vscale x 8 x i16> [[INDICES:%.*]])
// CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z14test_svtbx_u16u12__SVUint16_tS_S_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.tbx.nxv8i16(<vscale x 8 x i16> [[FALLBACK:%.*]], <vscale x 8 x i16> [[DATA:%.*]], <vscale x 8 x i16> [[INDICES:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
//
svuint16_t test_svtbx_u16(svuint16_t fallback, svuint16_t data, svuint16_t indices)
{
  return SVE_ACLE_FUNC(svtbx,_u16,,)(fallback, data, indices);
}

// CHECK-LABEL: @test_svtbx_u32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.tbx.nxv4i32(<vscale x 4 x i32> [[FALLBACK:%.*]], <vscale x 4 x i32> [[DATA:%.*]], <vscale x 4 x i32> [[INDICES:%.*]])
// CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z14test_svtbx_u32u12__SVUint32_tS_S_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.tbx.nxv4i32(<vscale x 4 x i32> [[FALLBACK:%.*]], <vscale x 4 x i32> [[DATA:%.*]], <vscale x 4 x i32> [[INDICES:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP0]]
//
svuint32_t test_svtbx_u32(svuint32_t fallback, svuint32_t data, svuint32_t indices)
{
  return SVE_ACLE_FUNC(svtbx,_u32,,)(fallback, data, indices);
}

// CHECK-LABEL: @test_svtbx_u64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.tbx.nxv2i64(<vscale x 2 x i64> [[FALLBACK:%.*]], <vscale x 2 x i64> [[DATA:%.*]], <vscale x 2 x i64> [[INDICES:%.*]])
// CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z14test_svtbx_u64u12__SVUint64_tS_S_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.tbx.nxv2i64(<vscale x 2 x i64> [[FALLBACK:%.*]], <vscale x 2 x i64> [[DATA:%.*]], <vscale x 2 x i64> [[INDICES:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP0]]
//
svuint64_t test_svtbx_u64(svuint64_t fallback, svuint64_t data, svuint64_t indices)
{
  return SVE_ACLE_FUNC(svtbx,_u64,,)(fallback, data, indices);
}

// CHECK-LABEL: @test_svtbx_f16(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x half> @llvm.aarch64.sve.tbx.nxv8f16(<vscale x 8 x half> [[FALLBACK:%.*]], <vscale x 8 x half> [[DATA:%.*]], <vscale x 8 x i16> [[INDICES:%.*]])
// CHECK-NEXT:    ret <vscale x 8 x half> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z14test_svtbx_f16u13__SVFloat16_tS_u12__SVUint16_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x half> @llvm.aarch64.sve.tbx.nxv8f16(<vscale x 8 x half> [[FALLBACK:%.*]], <vscale x 8 x half> [[DATA:%.*]], <vscale x 8 x i16> [[INDICES:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 8 x half> [[TMP0]]
//
svfloat16_t test_svtbx_f16(svfloat16_t fallback, svfloat16_t data, svuint16_t indices)
{
  return SVE_ACLE_FUNC(svtbx,_f16,,)(fallback, data, indices);
}

// CHECK-LABEL: @test_svtbx_f32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x float> @llvm.aarch64.sve.tbx.nxv4f32(<vscale x 4 x float> [[FALLBACK:%.*]], <vscale x 4 x float> [[DATA:%.*]], <vscale x 4 x i32> [[INDICES:%.*]])
// CHECK-NEXT:    ret <vscale x 4 x float> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z14test_svtbx_f32u13__SVFloat32_tS_u12__SVUint32_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x float> @llvm.aarch64.sve.tbx.nxv4f32(<vscale x 4 x float> [[FALLBACK:%.*]], <vscale x 4 x float> [[DATA:%.*]], <vscale x 4 x i32> [[INDICES:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 4 x float> [[TMP0]]
//
svfloat32_t test_svtbx_f32(svfloat32_t fallback, svfloat32_t data, svuint32_t indices)
{
  return SVE_ACLE_FUNC(svtbx,_f32,,)(fallback, data, indices);
}

// CHECK-LABEL: @test_svtbx_f64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x double> @llvm.aarch64.sve.tbx.nxv2f64(<vscale x 2 x double> [[FALLBACK:%.*]], <vscale x 2 x double> [[DATA:%.*]], <vscale x 2 x i64> [[INDICES:%.*]])
// CHECK-NEXT:    ret <vscale x 2 x double> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z14test_svtbx_f64u13__SVFloat64_tS_u12__SVUint64_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x double> @llvm.aarch64.sve.tbx.nxv2f64(<vscale x 2 x double> [[FALLBACK:%.*]], <vscale x 2 x double> [[DATA:%.*]], <vscale x 2 x i64> [[INDICES:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 2 x double> [[TMP0]]
//
svfloat64_t test_svtbx_f64(svfloat64_t fallback, svfloat64_t data, svuint64_t indices)
{
  return SVE_ACLE_FUNC(svtbx,_f64,,)(fallback, data, indices);
}