// 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 +sve -target-feature +bf16 -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 +sve -target-feature +bf16 -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 +sve -target-feature +bf16 -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 +sve -target-feature +bf16 -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_svstnt1_bf16(
// 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:    tail call void @llvm.aarch64.sve.stnt1.nxv8bf16(<vscale x 8 x bfloat> [[DATA:%.*]], <vscale x 8 x i1> [[TMP0]], bfloat* [[BASE:%.*]]), !tbaa [[TBAA6:![0-9]+]]
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z17test_svstnt1_bf16u10__SVBool_tPu6__bf16u14__SVBFloat16_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:    tail call void @llvm.aarch64.sve.stnt1.nxv8bf16(<vscale x 8 x bfloat> [[DATA:%.*]], <vscale x 8 x i1> [[TMP0]], bfloat* [[BASE:%.*]]), !tbaa [[TBAA6:![0-9]+]]
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_bf16(svbool_t pg, bfloat16_t *base, svbfloat16_t data)
{
  return SVE_ACLE_FUNC(svstnt1,_bf16,,)(pg, base, data);
}

// CHECK-LABEL: @test_svstnt1_vnum_bf16(
// 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 bfloat* [[BASE:%.*]] to <vscale x 8 x bfloat>*
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr <vscale x 8 x bfloat>, <vscale x 8 x bfloat>* [[TMP1]], i64 [[VNUM:%.*]], i64 0
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.nxv8bf16(<vscale x 8 x bfloat> [[DATA:%.*]], <vscale x 8 x i1> [[TMP0]], bfloat* [[TMP2]]), !tbaa [[TBAA6]]
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z22test_svstnt1_vnum_bf16u10__SVBool_tPu6__bf16lu14__SVBFloat16_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 bfloat* [[BASE:%.*]] to <vscale x 8 x bfloat>*
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr <vscale x 8 x bfloat>, <vscale x 8 x bfloat>* [[TMP1]], i64 [[VNUM:%.*]], i64 0
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.nxv8bf16(<vscale x 8 x bfloat> [[DATA:%.*]], <vscale x 8 x i1> [[TMP0]], bfloat* [[TMP2]]), !tbaa [[TBAA6]]
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_bf16(svbool_t pg, bfloat16_t *base, int64_t vnum, svbfloat16_t data)
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_bf16,,)(pg, base, vnum, data);
}