// 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_svhistcnt_s32_z(
// 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.histcnt.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]])
// CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z20test_svhistcnt_s32_zu10__SVBool_tu11__SVInt32_tS0_(
// 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.histcnt.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
//
svuint32_t test_svhistcnt_s32_z(svbool_t pg, svint32_t op1, svint32_t op2)
{
  return SVE_ACLE_FUNC(svhistcnt,_s32,_z,)(pg, op1, op2);
}

// CHECK-LABEL: @test_svhistcnt_s64_z(
// 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.histcnt.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]])
// CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z20test_svhistcnt_s64_zu10__SVBool_tu11__SVInt64_tS0_(
// 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.histcnt.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
//
svuint64_t test_svhistcnt_s64_z(svbool_t pg, svint64_t op1, svint64_t op2)
{
  return SVE_ACLE_FUNC(svhistcnt,_s64,_z,)(pg, op1, op2);
}

// CHECK-LABEL: @test_svhistcnt_u32_z(
// 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.histcnt.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]])
// CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z20test_svhistcnt_u32_zu10__SVBool_tu12__SVUint32_tS0_(
// 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.histcnt.nxv4i32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 4 x i32> [[TMP1]]
//
svuint32_t test_svhistcnt_u32_z(svbool_t pg, svuint32_t op1, svuint32_t op2)
{
  return SVE_ACLE_FUNC(svhistcnt,_u32,_z,)(pg, op1, op2);
}

// CHECK-LABEL: @test_svhistcnt_u64_z(
// 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.histcnt.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]])
// CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
//
// CPP-CHECK-LABEL: @_Z20test_svhistcnt_u64_zu10__SVBool_tu12__SVUint64_tS0_(
// 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.histcnt.nxv2i64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]])
// CPP-CHECK-NEXT:    ret <vscale x 2 x i64> [[TMP1]]
//
svuint64_t test_svhistcnt_u64_z(svbool_t pg, svuint64_t op1, svuint64_t op2)
{
  return SVE_ACLE_FUNC(svhistcnt,_u64,_z,)(pg, op1, op2);
}