// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
// REQUIRES: aarch64-registered-target
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - -x c++ %s | FileCheck %s -check-prefix=CPP-CHECK
// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -emit-llvm -o - -x c++ %s | FileCheck %s -check-prefix=CPP-CHECK
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -fallow-half-arguments-and-returns -S -O1 -Werror -Wall -o /dev/null %s
// A simple used,unused... macro, long enough to represent any SVE builtin.
// CHECK-LABEL: @test_svclastb_s8(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.clastb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]])
// CHECK-NEXT: ret <vscale x 16 x i8> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z16test_svclastb_s8u10__SVBool_tu10__SVInt8_tu10__SVInt8_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.clastb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret <vscale x 16 x i8> [[TMP0]]
//
svint8_t
// CHECK-LABEL: @test_svclastb_s16(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.clastb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[FALLBACK:%.*]], <vscale x 8 x i16> [[DATA:%.*]])
// CHECK-NEXT: ret <vscale x 8 x i16> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z17test_svclastb_s16u10__SVBool_tu11__SVInt16_tu11__SVInt16_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.clastb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[FALLBACK:%.*]], <vscale x 8 x i16> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret <vscale x 8 x i16> [[TMP1]]
//
svint16_t
// CHECK-LABEL: @test_svclastb_s32(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.clastb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[FALLBACK:%.*]], <vscale x 4 x i32> [[DATA:%.*]])
// CHECK-NEXT: ret <vscale x 4 x i32> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z17test_svclastb_s32u10__SVBool_tu11__SVInt32_tu11__SVInt32_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.clastb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[FALLBACK:%.*]], <vscale x 4 x i32> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret <vscale x 4 x i32> [[TMP1]]
//
svint32_t
// CHECK-LABEL: @test_svclastb_s64(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.clastb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[FALLBACK:%.*]], <vscale x 2 x i64> [[DATA:%.*]])
// CHECK-NEXT: ret <vscale x 2 x i64> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z17test_svclastb_s64u10__SVBool_tu11__SVInt64_tu11__SVInt64_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.clastb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[FALLBACK:%.*]], <vscale x 2 x i64> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret <vscale x 2 x i64> [[TMP1]]
//
svint64_t
// CHECK-LABEL: @test_svclastb_u8(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.clastb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]])
// CHECK-NEXT: ret <vscale x 16 x i8> [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z16test_svclastb_u8u10__SVBool_tu11__SVUint8_tu11__SVUint8_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.clastb.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], <vscale x 16 x i8> [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret <vscale x 16 x i8> [[TMP0]]
//
svuint8_t
// CHECK-LABEL: @test_svclastb_u16(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.clastb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[FALLBACK:%.*]], <vscale x 8 x i16> [[DATA:%.*]])
// CHECK-NEXT: ret <vscale x 8 x i16> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z17test_svclastb_u16u10__SVBool_tu12__SVUint16_tu12__SVUint16_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.clastb.nxv8i16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x i16> [[FALLBACK:%.*]], <vscale x 8 x i16> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret <vscale x 8 x i16> [[TMP1]]
//
svuint16_t
// CHECK-LABEL: @test_svclastb_u32(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.clastb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[FALLBACK:%.*]], <vscale x 4 x i32> [[DATA:%.*]])
// CHECK-NEXT: ret <vscale x 4 x i32> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z17test_svclastb_u32u10__SVBool_tu12__SVUint32_tu12__SVUint32_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.clastb.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[FALLBACK:%.*]], <vscale x 4 x i32> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret <vscale x 4 x i32> [[TMP1]]
//
svuint32_t
// CHECK-LABEL: @test_svclastb_u64(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.clastb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[FALLBACK:%.*]], <vscale x 2 x i64> [[DATA:%.*]])
// CHECK-NEXT: ret <vscale x 2 x i64> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z17test_svclastb_u64u10__SVBool_tu12__SVUint64_tu12__SVUint64_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.clastb.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[FALLBACK:%.*]], <vscale x 2 x i64> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret <vscale x 2 x i64> [[TMP1]]
//
svuint64_t
// CHECK-LABEL: @test_svclastb_f16(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 8 x half> @llvm.aarch64.sve.clastb.nxv8f16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x half> [[FALLBACK:%.*]], <vscale x 8 x half> [[DATA:%.*]])
// CHECK-NEXT: ret <vscale x 8 x half> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z17test_svclastb_f16u10__SVBool_tu13__SVFloat16_tu13__SVFloat16_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 8 x half> @llvm.aarch64.sve.clastb.nxv8f16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x half> [[FALLBACK:%.*]], <vscale x 8 x half> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret <vscale x 8 x half> [[TMP1]]
//
svfloat16_t
// CHECK-LABEL: @test_svclastb_f32(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x float> @llvm.aarch64.sve.clastb.nxv4f32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x float> [[FALLBACK:%.*]], <vscale x 4 x float> [[DATA:%.*]])
// CHECK-NEXT: ret <vscale x 4 x float> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z17test_svclastb_f32u10__SVBool_tu13__SVFloat32_tu13__SVFloat32_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 4 x float> @llvm.aarch64.sve.clastb.nxv4f32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x float> [[FALLBACK:%.*]], <vscale x 4 x float> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret <vscale x 4 x float> [[TMP1]]
//
svfloat32_t
// CHECK-LABEL: @test_svclastb_f64(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 2 x double> @llvm.aarch64.sve.clastb.nxv2f64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x double> [[FALLBACK:%.*]], <vscale x 2 x double> [[DATA:%.*]])
// CHECK-NEXT: ret <vscale x 2 x double> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z17test_svclastb_f64u10__SVBool_tu13__SVFloat64_tu13__SVFloat64_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call <vscale x 2 x double> @llvm.aarch64.sve.clastb.nxv2f64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x double> [[FALLBACK:%.*]], <vscale x 2 x double> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret <vscale x 2 x double> [[TMP1]]
//
svfloat64_t
// CHECK-LABEL: @test_svclastb_n_s8(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call i8 @llvm.aarch64.sve.clastb.n.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], i8 [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]])
// CHECK-NEXT: ret i8 [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z18test_svclastb_n_s8u10__SVBool_tau10__SVInt8_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call i8 @llvm.aarch64.sve.clastb.n.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], i8 [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret i8 [[TMP0]]
//
int8_t
// CHECK-LABEL: @test_svclastb_n_s16(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = bitcast i16 [[FALLBACK:%.*]] to half
// CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 8 x i16> [[DATA:%.*]] to <vscale x 8 x half>
// CHECK-NEXT: [[TMP3:%.*]] = tail call half @llvm.aarch64.sve.clastb.n.nxv8f16(<vscale x 8 x i1> [[TMP0]], half [[TMP1]], <vscale x 8 x half> [[TMP2]])
// CHECK-NEXT: [[TMP4:%.*]] = bitcast half [[TMP3]] to i16
// CHECK-NEXT: ret i16 [[TMP4]]
//
// CPP-CHECK-LABEL: @_Z19test_svclastb_n_s16u10__SVBool_tsu11__SVInt16_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = bitcast i16 [[FALLBACK:%.*]] to half
// CPP-CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 8 x i16> [[DATA:%.*]] to <vscale x 8 x half>
// CPP-CHECK-NEXT: [[TMP3:%.*]] = tail call half @llvm.aarch64.sve.clastb.n.nxv8f16(<vscale x 8 x i1> [[TMP0]], half [[TMP1]], <vscale x 8 x half> [[TMP2]])
// CPP-CHECK-NEXT: [[TMP4:%.*]] = bitcast half [[TMP3]] to i16
// CPP-CHECK-NEXT: ret i16 [[TMP4]]
//
int16_t
// CHECK-LABEL: @test_svclastb_n_s32(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = bitcast i32 [[FALLBACK:%.*]] to float
// CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 4 x i32> [[DATA:%.*]] to <vscale x 4 x float>
// CHECK-NEXT: [[TMP3:%.*]] = tail call float @llvm.aarch64.sve.clastb.n.nxv4f32(<vscale x 4 x i1> [[TMP0]], float [[TMP1]], <vscale x 4 x float> [[TMP2]])
// CHECK-NEXT: [[TMP4:%.*]] = bitcast float [[TMP3]] to i32
// CHECK-NEXT: ret i32 [[TMP4]]
//
// CPP-CHECK-LABEL: @_Z19test_svclastb_n_s32u10__SVBool_tiu11__SVInt32_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = bitcast i32 [[FALLBACK:%.*]] to float
// CPP-CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 4 x i32> [[DATA:%.*]] to <vscale x 4 x float>
// CPP-CHECK-NEXT: [[TMP3:%.*]] = tail call float @llvm.aarch64.sve.clastb.n.nxv4f32(<vscale x 4 x i1> [[TMP0]], float [[TMP1]], <vscale x 4 x float> [[TMP2]])
// CPP-CHECK-NEXT: [[TMP4:%.*]] = bitcast float [[TMP3]] to i32
// CPP-CHECK-NEXT: ret i32 [[TMP4]]
//
int32_t
// CHECK-LABEL: @test_svclastb_n_s64(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = bitcast i64 [[FALLBACK:%.*]] to double
// CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 2 x i64> [[DATA:%.*]] to <vscale x 2 x double>
// CHECK-NEXT: [[TMP3:%.*]] = tail call double @llvm.aarch64.sve.clastb.n.nxv2f64(<vscale x 2 x i1> [[TMP0]], double [[TMP1]], <vscale x 2 x double> [[TMP2]])
// CHECK-NEXT: [[TMP4:%.*]] = bitcast double [[TMP3]] to i64
// CHECK-NEXT: ret i64 [[TMP4]]
//
// CPP-CHECK-LABEL: @_Z19test_svclastb_n_s64u10__SVBool_tlu11__SVInt64_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = bitcast i64 [[FALLBACK:%.*]] to double
// CPP-CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 2 x i64> [[DATA:%.*]] to <vscale x 2 x double>
// CPP-CHECK-NEXT: [[TMP3:%.*]] = tail call double @llvm.aarch64.sve.clastb.n.nxv2f64(<vscale x 2 x i1> [[TMP0]], double [[TMP1]], <vscale x 2 x double> [[TMP2]])
// CPP-CHECK-NEXT: [[TMP4:%.*]] = bitcast double [[TMP3]] to i64
// CPP-CHECK-NEXT: ret i64 [[TMP4]]
//
int64_t
// CHECK-LABEL: @test_svclastb_n_u8(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call i8 @llvm.aarch64.sve.clastb.n.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], i8 [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]])
// CHECK-NEXT: ret i8 [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z18test_svclastb_n_u8u10__SVBool_thu11__SVUint8_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call i8 @llvm.aarch64.sve.clastb.n.nxv16i8(<vscale x 16 x i1> [[PG:%.*]], i8 [[FALLBACK:%.*]], <vscale x 16 x i8> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret i8 [[TMP0]]
//
uint8_t
// CHECK-LABEL: @test_svclastb_n_u16(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = bitcast i16 [[FALLBACK:%.*]] to half
// CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 8 x i16> [[DATA:%.*]] to <vscale x 8 x half>
// CHECK-NEXT: [[TMP3:%.*]] = tail call half @llvm.aarch64.sve.clastb.n.nxv8f16(<vscale x 8 x i1> [[TMP0]], half [[TMP1]], <vscale x 8 x half> [[TMP2]])
// CHECK-NEXT: [[TMP4:%.*]] = bitcast half [[TMP3]] to i16
// CHECK-NEXT: ret i16 [[TMP4]]
//
// CPP-CHECK-LABEL: @_Z19test_svclastb_n_u16u10__SVBool_ttu12__SVUint16_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = bitcast i16 [[FALLBACK:%.*]] to half
// CPP-CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 8 x i16> [[DATA:%.*]] to <vscale x 8 x half>
// CPP-CHECK-NEXT: [[TMP3:%.*]] = tail call half @llvm.aarch64.sve.clastb.n.nxv8f16(<vscale x 8 x i1> [[TMP0]], half [[TMP1]], <vscale x 8 x half> [[TMP2]])
// CPP-CHECK-NEXT: [[TMP4:%.*]] = bitcast half [[TMP3]] to i16
// CPP-CHECK-NEXT: ret i16 [[TMP4]]
//
uint16_t
// CHECK-LABEL: @test_svclastb_n_u32(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = bitcast i32 [[FALLBACK:%.*]] to float
// CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 4 x i32> [[DATA:%.*]] to <vscale x 4 x float>
// CHECK-NEXT: [[TMP3:%.*]] = tail call float @llvm.aarch64.sve.clastb.n.nxv4f32(<vscale x 4 x i1> [[TMP0]], float [[TMP1]], <vscale x 4 x float> [[TMP2]])
// CHECK-NEXT: [[TMP4:%.*]] = bitcast float [[TMP3]] to i32
// CHECK-NEXT: ret i32 [[TMP4]]
//
// CPP-CHECK-LABEL: @_Z19test_svclastb_n_u32u10__SVBool_tju12__SVUint32_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = bitcast i32 [[FALLBACK:%.*]] to float
// CPP-CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 4 x i32> [[DATA:%.*]] to <vscale x 4 x float>
// CPP-CHECK-NEXT: [[TMP3:%.*]] = tail call float @llvm.aarch64.sve.clastb.n.nxv4f32(<vscale x 4 x i1> [[TMP0]], float [[TMP1]], <vscale x 4 x float> [[TMP2]])
// CPP-CHECK-NEXT: [[TMP4:%.*]] = bitcast float [[TMP3]] to i32
// CPP-CHECK-NEXT: ret i32 [[TMP4]]
//
uint32_t
// CHECK-LABEL: @test_svclastb_n_u64(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = bitcast i64 [[FALLBACK:%.*]] to double
// CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 2 x i64> [[DATA:%.*]] to <vscale x 2 x double>
// CHECK-NEXT: [[TMP3:%.*]] = tail call double @llvm.aarch64.sve.clastb.n.nxv2f64(<vscale x 2 x i1> [[TMP0]], double [[TMP1]], <vscale x 2 x double> [[TMP2]])
// CHECK-NEXT: [[TMP4:%.*]] = bitcast double [[TMP3]] to i64
// CHECK-NEXT: ret i64 [[TMP4]]
//
// CPP-CHECK-LABEL: @_Z19test_svclastb_n_u64u10__SVBool_tmu12__SVUint64_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = bitcast i64 [[FALLBACK:%.*]] to double
// CPP-CHECK-NEXT: [[TMP2:%.*]] = bitcast <vscale x 2 x i64> [[DATA:%.*]] to <vscale x 2 x double>
// CPP-CHECK-NEXT: [[TMP3:%.*]] = tail call double @llvm.aarch64.sve.clastb.n.nxv2f64(<vscale x 2 x i1> [[TMP0]], double [[TMP1]], <vscale x 2 x double> [[TMP2]])
// CPP-CHECK-NEXT: [[TMP4:%.*]] = bitcast double [[TMP3]] to i64
// CPP-CHECK-NEXT: ret i64 [[TMP4]]
//
uint64_t
// CHECK-LABEL: @test_svclastb_n_f16(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call half @llvm.aarch64.sve.clastb.n.nxv8f16(<vscale x 8 x i1> [[TMP0]], half [[FALLBACK:%.*]], <vscale x 8 x half> [[DATA:%.*]])
// CHECK-NEXT: ret half [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z19test_svclastb_n_f16u10__SVBool_tDhu13__SVFloat16_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call half @llvm.aarch64.sve.clastb.n.nxv8f16(<vscale x 8 x i1> [[TMP0]], half [[FALLBACK:%.*]], <vscale x 8 x half> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret half [[TMP1]]
//
float16_t
// CHECK-LABEL: @test_svclastb_n_f32(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call float @llvm.aarch64.sve.clastb.n.nxv4f32(<vscale x 4 x i1> [[TMP0]], float [[FALLBACK:%.*]], <vscale x 4 x float> [[DATA:%.*]])
// CHECK-NEXT: ret float [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z19test_svclastb_n_f32u10__SVBool_tfu13__SVFloat32_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call float @llvm.aarch64.sve.clastb.n.nxv4f32(<vscale x 4 x i1> [[TMP0]], float [[FALLBACK:%.*]], <vscale x 4 x float> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret float [[TMP1]]
//
float32_t
// CHECK-LABEL: @test_svclastb_n_f64(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CHECK-NEXT: [[TMP1:%.*]] = tail call double @llvm.aarch64.sve.clastb.n.nxv2f64(<vscale x 2 x i1> [[TMP0]], double [[FALLBACK:%.*]], <vscale x 2 x double> [[DATA:%.*]])
// CHECK-NEXT: ret double [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z19test_svclastb_n_f64u10__SVBool_tdu13__SVFloat64_t(
// CPP-CHECK-NEXT: entry:
// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
// CPP-CHECK-NEXT: [[TMP1:%.*]] = tail call double @llvm.aarch64.sve.clastb.n.nxv2f64(<vscale x 2 x i1> [[TMP0]], double [[FALLBACK:%.*]], <vscale x 2 x double> [[DATA:%.*]])
// CPP-CHECK-NEXT: ret double [[TMP1]]
//
float64_t